Abstracts No.27: 7th International Conference on Geochron, Cosmochron & Isotope Geology, 1990, ACT by GSAustralia

Geological Society of Australia

ABSTRACTS Number 27

SEVENTH INTERNATIONAL CONFERENCE ON

GEOCHRONOLOGY COSMOCHRONOLOGY AND ISOTOPE GEOLOGY Monday 24-Saturday 29 September, 1990 Canberra—Australia

CANBERRA 1 990

SEVENTH INTERNATIONAL CONFERENCE ON

GEOCHRONOLOGY COSMOCHRONOLOGY AND ISOTOPE GEOLOGY Monday 24 — Saturday 29 September, 1990 Canberra - Australia

Abstract Volume

Published by the Geological Society of Australia Incorporated 606 ANA House, 301 George Street, Sydney, NSW 2000 1990 ISSN 0729 01IX

Printed by Union Offset Co. Pty Ltd, Canberra

Geological Society of Australia

ABSTRACTS Number 27

Letter from organizers Organizing Committees

Page v vi

Sponsors

vii

Abstracts (in alphabetical order, by first author) Author Index Advertisements

1 117 127

SEVENTH INTERNATIONAL CONFERENCE ON GEOCHRONOLOGY, COSMOCHRONOLOGY AND ISOTOPE GEOLOGY Monday 24 - Saturday 29 September, 1990 Canberra - Australia

We and our sponsors bid you a warm welcome to Canberra and the Seventh International Conference on Geochronology, Cosmochronology and Isotope Geology (ICOG 7), Many would know that this conference had its beginnings at a meeting on the geochronology of Precambrian rocks, held at Edmonton, Alberta, in 1967. Since then, the scope of the conference and the numbers of delegates have grown enormously, despite the fact that ICOG has no formal international affiliations. We have been delighted by the response to our invitation to attend ICOG 7 in Canberra, At the time of this printing, there are 560 delegates attending from 25 countries, and 464 submitted abstracts in this volume. In the spirit oflCOGs past, ICOG 7 is intended for the discussion of recent advances in a wide range of isotope geoscience disciplines through both poster and oral presentations. The scientific program is divided into two sections: topical Symposia, and General Sessions. Symposia consist of a limited number of presentations selected by conveners to provide an overview of afield, or to point to new directions in a developing area of isotope geoscience. The General Sessions (both oral and poster) are divided according to a number of themes that are evident in the detailed program. The format of this meeting, especially the emphasis on poster presentations, will be new to some. We have planned the scientific and social programs in the hope that you will not only find ICOG 7 professionally stimulating, but also that you will have plenty of opportunity to meet with other delegates outside the formal sessions. We are confident that delegates will enjoy and gain from the degree of informality that this program structure brings. Our aim is to make your stay in Canberra pleasant and profitable. If you have any questions or problems, please feel free to ask any of the committee members, technical assistants, or people at the Registration desk for help. We thank you for your response and hope that your visit to Australia will be rewarding.

ICOG 7 Organising Committee

Organising Committee ICOG 7 has been organised by a Canberra-based committee from, and with the support of, the Research School of Earth Sciences (RSES), the Australian National University (ANU), and the Bureau of Mineral Resources (BMR), in association with a number of individuals from other Australian institutions. Members of the Organising Committee are: W. Compston (Chairman) L.P. Black A.R. Chivas M.T. McCulloch I. McDougall R.W. Page J.R. Richards P. Wellman I.S. Williams AustraUan Convention and Travel Services Pty Ltd (ACTS) The Organising Committee has had considerable assistance from the six Symposium conveners, from Mr P. Radley, Mr T. Davies, Mr R. Wylde-Browne, Mrs Jennifer Delhaize and Mrs Pam De Winter (all RSES), and from the people on the Scientific Program Subcommittee: I. McDougall (Chairman) A.R. Chivas W. Compston M.T. Esat R.I. Hill P.D. Kinny W.F. McDonough R.W. Page R.L. Rudnick J.O.H.Stone

Sponsors The Organising Committee is grateful for financial and other support from the following organisations: • Research School of Earth Sciences, ANU •

Bureau of Mineral Resources, Geology & Geophysics

•

Geological Society of Australia Inc. Australian Academy of Science 25th IGC Fund

• • •

International Association of Geochemistry and Cosmochemistry International Geological Correlation Program 280 International Union of Geological Sciences, Subcommission on Geochronology International Science and Technology Advisory Committee (Department of Industry, Technology and Commerce) • International Seminars Support Scheme (Australian Intemational Development Assistance Bureau, AIDAB) • Convention Assistance Scheme, Australian Tourist Commission and the following companies: Aberfoyle Resources Limited Exploration Division Ansett Airlines BHP-Utah Minerals Intemational Exploration Division Billiton Australia Canon Copiers Finnigan MAT Pasminco Mining, Broken Hill Qantas Airways Ltd RGC Exploration Pty Ltd Stockdale Prospecting Limited VG Instruments Western Mining Corporation Exploration Division

ABSTRACTS

ICOG 7

APATITE FISSION TRACK DATING AND THE UPLIFT OF THE APUAN ALPS AND SURROUNDING REGIONS (NORTHERN APENNINES, ITALY) ABBATE, Ernesto; BALESTRIERI, Maria Laura, Dip. Scien ze della Terra, University of Florence, Italy, BIGAZZI Giulio; NORELLI, Pio; QUERCIOLI, Carlo, 1st. Geocrono logia e Geochimica Isotopica, CNR, Pisa, Italy Fifteen apatite concentrates have been dated by fission tracks (FT) using the population method. The samples come from three superimposed tectonic units variously affected by postorogenic uplift and erosion and presently outcrop ping at different elevations. From the Apuan metamorphic sequence, which is the lowest unit. Paleozoic phyllites and metavolcanoclastics give ages of approximately 4 Ma, while the Oligocene metagraywackes of 2 Ma. In the upper unit (Tuscan Nappe)outcrop ping near La Spezia the Oligocene Macigno arenaceous turbi dites and associated conglomerates have been sampled. Whi le the FT ages of the sandstones span from 8 to 5 Ma, an Her cynian granite boulder included in the conglomerates and at the same level of the sandstones gives 50 Ma. Since the apatite crystals of the granite differ in color and shape from those of the sandstones, the contrasting behaviour in track retention might be related to chemical composition differences. In the arenaceous matrix which envelopes the granite boulder the two apatite families have been found together and give a mixed age of 17 Ma. The track length distribution indicates that the grani te apatites still retain an age which is not far from a track reset around the Cretaceous/Tertiary boundary. A prolonged residence in the fading zone seems to be likely both for the sandstones and for the granite boulder. Assuming that in an active mountain belt such as the A pennines FT ages record cooling due to uplift and erosion, and estimating a geothermal gradient of 30 C/km, uplift rates ranging from 0.7 to 1.7mm/a are derived for a region which extends from the Tyrrhenian sea to the Apuan Alps. Relatively young FT ages and rapid uplift are in agreement with the sedimentary history of adjacent N^o gene to Recent Basins.

2.1 GA MAFIC MAGMATISM IN WEST AFRICA: AN EARLY STAGE OF CRUSTAL ACCRETION. ABOUCHAMI. Wafa, BOHER, Muriel, ALBAREDE, Francis, BARBEY, Pierre and MICHARD, Annie, C.R.P.G and University

de Nancy I, BP 20, 54501 Vandoeuvre cedex, France.

Birrimian terranes from West Africa comprise two major units: a dominantly mafic bimodal volcanic unit and a volcano-detrital unit with mostly felsic to intermediate protolith. Current work suggest that they both formed in a short time interval around 2.1 Ga ago. Widespread basaltic magmas from the bimodal unit have been analyzed for REE distributions and Sr-Nd isotopes. Sm-Nd isochrons on tholeiitic lavas were obtained at 2.126 ± 0.024 Ga and initial eNd = 2.9 ± 0.7 for Burkina Faso, at 2.063 ± 0.041 Ga and 3.1 ± 1.0 for Eastern Senegal, data which compare with the age of 2.11 ± 0.09 Ga and initial eNd = 2.1 ± 1.8 obtained in Guyana by Gruau et al. [1985]. Samples from other localities give generally similar results. Although the variations of Sm/Nd ratios and the scatter of eNd(T) values from +0.7 to +4.3 preclude a single origin for those magmas, initial isotopic heterogeneities are unlikely to bias significantly the ages given by the isochrons which are in good a /eement with U-Pb zircon ages [Boher et al., 1989]. Presence of lavas with frequent pillow structures and sediments free of older recycled components suggests that Birrimian terranes formed in ocean basins far from continental influence. The isotopic heterogeneities are not consistent with a MORB-like mantle source. Most lavas are slightly depleted in LREE and inversion of the data through a melting model suggests 5-15 percent melting of a slightly depleted Iherzolite. Oceanic flood basalts similar to those which form oceanic plateaus (e.g. in the Nauru basin) and later accreted to continents as allochtonous terranes represent the most acceptable modern analogue of Birrimian basalts. Deep plumes piercing young lithosphere can generate huge amounts of tholeiites in a short time. Although the 2.1 Ga old magmatic event in West Africa has gone virtually unnoticed in the literature, its spatial extension compares with the distribution of mantle-derived magmatic activity in other major orogenic provinces. If such large crustal segments were overlooked, a spurious pattern of episodic activity of the mantle could arise.

R E V I E W O F K - A R A N D R B - S R D A T I N G STUDII2S O F MESOZOLC L O W - G R A D E M E T A S E D M E N T S IN N E W Z E A L A N D : A G E LIMITS O N B U R U L , U P L I F T A N D TERRANE ASSEMBLY

ADAMS. C.J. and GRAHAM, I.J., Institute of Nuclear Sciences, DSIR, Lower Hutt, New Zealand

K-Ar and Rb-Sr ages of suites of low-grade metasedimentary and metavolcanic rocks from Torlesse and Caples-Waipapa terranes of New Zealand and equivalent Otago Schists indicate widespread burial metamorphism in late Triassic-late Jurassic times. Isotopic resetting and re-equilibration of rocks of various protolithologies was fully complete during this phase. Younger, mid-early Cretaceous ages are related to more local structural effects i.e. obduction of the geosynclinal sequence, and/or terrane assembly and are associated with gold/scheelite mineralisation. Initial Sr isotope ratios confirm the acid-continental (high values) or arcvolcanic derivation (low values) of these suites and allow characterisation of the terranes and their extension into the highergrade Otago Schists.

RADIOCARBON AGES OF THE HOLOCENE LACUSTRINE CARBONATE SEDIMENTS FROM THE COORONG REGION OF SOUTH AUSTRALIA: IMPLICATIONS FOR DETERMINATION OF THE GREAT AUSTRALL\N ARID PERIOD AND SEDIMENTATION RATE IN THE LAKES National University, GPO Box 4, Canberra, ACT 2601, Australia, and John HEAD, Radiocarbon Dating Research, Research School of Pacific Studies, The Australian National University, GPO Box 4, Canberra, ACT 2601, Australia A chain of ephemeral lakes occurs in the coastal Coorong Region (Lat. 36°00' - 36°45' S, Long. 139^30' - 139°55' E) of South Australia. A 5 to 10 cm thick weakly lithified calcrete layer exists within the Holocene carbonate sequence of these lakes. Physicochemical evidence suggest that this calcrete layer was formed as a result of subaereal/vadose zone diagenesis of the carbonate sediments at the lake surface while they remained dry for a certain period of time, thereby acting as an evidence of a great Australian arid period (GAAP) in South Australia. Although the previous workers, based on evidences elsewhere in Australia, proposed an age of about 4000 years BP for this GAAP, determination of the ^"^C ages of the samples of the calcrete layer from two of these lakes (The Kingston and the EMU lakes) gives an age rage of 2250 ± 180 years to 3000 ± 170 years BP. These dates suggest that the GAAP probably lasted at least in South Australia until about 2250 ± 180 years BP. Adopting an average of 2600 years BP for the calcrete layer in the Coorong lake chain the post-GAAP carbonate sedimentation rate was calculated for some of the lakes. This gives significantly varying sedimentation rates for the lakes, such as - 0.03 mm / yr for the Kingstone Lake, 0.09 mm / yr for the EMU Lake, 0.12 mm / yr for the South Stromatolite Lake, 0.38 mm / yr for the Pellet Lake, and 0.12 mm / yr for the Milnes Lake. Radiocarbon dating of a carbonate sediment sample from the bottom part of the Holocene sequence in the Bird Lake indicates the beginning of its evolution at about 6000 years BP. Sample from the top 5 cm of the dolomitie enriched surface sediments of the EMU Lake gives an age of 590 ± 220 years BP probably suggesting that the dolomite is currently in the process of formation in the lake as a result of primary inorganic precipitation from the lake brine.

ICOG 7

ABSTRACTS

IN SITU LOCATION OF INTERSTELLAR SiC in TWO CM METEORITES ALEXANDER, C.M.O'D, SWAN, P., and WALKER. R. M.. McDonnell Center for Space Sciences and Physics Department, Washington University, St. Louis, MO 63130 USA Isotopically anomalous grains of SiC have been found in acid residues of both CM and ordinary chondrites 1»2. We report here the first measurements of such SiC grains, in situ. The grains were found in polished sections of Cold Bokkeveld (CB) and Murchison (MURCH) using a low voltage X-ray mapping technique3. The method uses differences in Si X-ray production rates in conjunction with image processing to find SiC grains even when they are imbedded in a silicate matrix. The W.U. ion probe was used to measure isotopes directly on the sections using negative secondary ions. The work has three goals: 1) to determine if SiC is associated with other isotopically anomalous materials, 2) to compare the properties of the in situ grains with those found in acid residues and 3) to search for other Si-rich grains (e.g., pure Si). A total of 16 SiC grains were found in the CB section, of which 9 were large enough (> l^im) to permit isotopic measurements. Of these, 5 were found to have anomalous carbon (913c = + 350 to + 2400 %o) proving that they were indigenous to the meteorite. The remaining 4 grains had normal isotopic compositions and may or may not be due to terrestrial contamination, although we lean to the view that they are not. Four grains of SiC found in the MURCH section were all anomalous (31^3c = + 150 to + 700 %o). Silicon anomalies were also found. All of the SiC occurred as isolated matrix grains. Several had fissures suggesting that they would have broken into smaller fragments during disaggregation. However, there is no qualitative difference in the size distribution of in situ grains compared to those in acid residues. No grains of other Si-rich material (e.g.. Si, SiN) were found. Additional mapping work on other meteorites will be presented at the meeting. References: [1] Bematowicz et al. (1987) Nature 330, 728-730. [2] Alexander et al. (1989) Meteoritics 24, 247. [3] Swan et al. (1989) LPSC XX, 1093.

REE AND ND ISOTOPE DATA IN BIF FROM HAMERSLEY, WESTERN AUSTRALIA: IMPLICATIONS FOR THE COMPOSITION OF EARLY PROTEROZOIC SEAWATER. ALIBERT. Chantal, and McCULLOCH, Malcolm T., Research School of Earth Sciences, ANU, Canberra 2601, Australia Nd isotopic compositions and REE have been measured in composite drill core iron formations (IF), individual magnetite and chert bands and interlayered shaley macrobands from the Jeerinah, Marra Mamba, Dales Gorge and Joffre formations, deposited in the Hamersley Basin in the intervall 2.7-2.5 Ga. The Dales Gorge and Joffre IF give = 0.4 ± 0.8 at 2.5 Ga. The shaley lithologies (Sbands, Trendall & Blockley, 1970) are an important contributor to the Nd budget of the BIF, owing to their high Fe and Nd contents, and proportion (up to 1/5 of Dales Gorge). Their average eNj = -0.6 ± 0.9 is distinct from that of associated IF. A few S-bands show a clear volcanic component and stand out by their higher eN^ (+1 - +2). By contrast, the black shales from Jeerinah , the IF from Marra Mamba and the shaley bands in the overlying Wittenoom Dolomite, are not distinguishable in terms of their Nd isotopic signature: e^d = 0.5 ± 0.9 at 2.6 Ga. REE patterns of IF show the typical heavy REE enrichment of seawater but lack any clear Ce or Eu anomaly, which is consistent with the expected lower p02 compared to modem marine environment. The low Nd content (0.4 to 4 ppm) suggests a rapid removal from seawater and sedimentation rate. The REE in the S-bands show an enhanced enrichment (-20 ppm Nd) and patterns intermediate between those of IF and those of typical shales such as the McRae Shale. These features, as well as Nd isotopic compositions overlaping with those of Marra Mamba IF, suggest that these deposits could represent hydrothermal precipitates formed under reducing conditions, in accord with the ubiquitous presence of sulfides, siderite and carbon. Such conditions could be found in deep waters of a partially closed basin whose Nd isotopic composition is likely to be controled by the basement formed by the Fortescue volcanics. Conversely, the more positive en^ (0 to +1) and more oxidized character of the Dales Gorge and Joffre IF could be related to an evolution towards an open ocean environment. Finally, a metamorphic age of 2,130 ± 30 Ma is defined by the riebeckite-rich IF. Highly positive e^d at 2.5 Ga in such samples previously led to overrate the hydrothermal contribution to the REE budget of the early Proterozoic seawater, at Hamersley.

INTERSTELLAR SIC IN DIFFERENT METEORITE TYPES: RESULTS AND IMPLICATIONS ALEXANDER. C.M.O^D . WALKER, R.M. AND ZINNER, E., McDonnell Center for the Space Sciences and Physics Department, Washington University, St. Louis, MO 63130 USA

DUPAL-TYPE Pb-Sr-Nd ISOTOPIC ANOMALY IN DOVIREN LAYERED INTRUSION, EASTERN SIBERIA AMELIN. Ju.V., NEYMARK, L.A., NEMCHIN, A.A., RITSK, E.Ju., Institute of Precambrian Geology & Geochronology, USSR Academy of Sciences, Leningrad 199034 USSR

Isotopically anomalous SiC of probable interstellar origin is common in primitive meteorites, having been measured in acid, residues of CMl, CV2, CP and UOC's^ However, concentrations are highly variable ranging from -5 ppm to <.02 ppm in Allende(CV)5.Grains with normal Si and C are also found in the same meteorites in variable proportions relative to the anomalous ones. These normal grains are either indigenous to the meteorites or are terrestrial contaminants. If the former is true (we hope to settle the issue shortly), it suggests that some SiC can be formed in the solar nebula. However that would not change the basic conclusion that the highly anomalous grains are probably pre-solar. Interestingly our data indicate that the SiC may not represent a simple mixture between a nebular component and a single interstellar one- different meteorites apparently have different populations of isotopically anomalous grains. For example, measurements in Chainpur2 give d^^C values > 1000 for 12 of 36 single grains while the corresponding numbers for the more extensively studied Murchison meteorite are 10 out of 89 (excluding normal grains ); Si isotopic groupings are also found in significantly different proportions in different meteorites, even in those of the same type. Although all measurements have been made on grains l^im to lO^m in size, it is possible that size sorting, or other selection effects, either in the nebula or in the processing used to produce the acid residues, are the cause of the differences. Alternatively, the differences could be due to basic heterogeneities in the distribution of dust from different stellar sources in the precursor interstellar dust cloud and their preservation in the ensuing collapse and accretion processes. References: [1] Zinner et al. (1989) G.C.A. 53, 1235-1244. [2] Alexander et al. (1990) Submitted to Met. Soc. [3] Stone et al. (1990) LPSC XXI, 1212-1213. [4] Alexander et al. (1990) LPSC XXI, 9-10. [5] Swan et al. (1989) LPSC XX, 1093-1094.

The Doviren layered intrusion of average picritic composition in the Northern Baikal area intrudes metasediments of the upper Proterozoic age. Sm-Nd and Rb-Sr mineral isochrons on a gabbronorite sample yield ages of 703 ± 40 Ma (2a) and 714 ± 46 Ma respectively. Initial ^'Sr/^^Sr (jgj.^ 70Q Ma) vary from 0.7100 to 0.7135 and E^. (700 Ma) from -13.1 to -16.1. Sr-Nd isotopic data for the surrounding sediments Isr (700 Ma) from 0.708 to 0.716 and Ejs^jj (700 Ma) from -6.8 to -11.0 demonstrate that low ejvj^j in Doviren rocks can not result only from wallrock assimilation. Plagioclase Pb-isotope model parameters correspond to mean crustal ILI = 9.7 and enhanced K = 4.3 (Stacey-Kramers model). These Pb isotopic characteristics can not be identified with any of the global reservoirs in the Plumbotectonics model of Zartman & Doe (1981) and resemble EMI component of Zindler & Hart (1986). The Sr-Nd-Pb isotopic patterns for the Doviren intnision are an extreme variant of the features that define the DUPAL anomaly in some oceanic basalts (Hart, 1984). Such isotopic characteristics might result from meldng of the ancient subcontinental mantle and/or lower crust at early stages of rifting processes.

ABSTRACTS

IC0G7

SOURCES OF A-TYPES GRANITES : Sr-Nd ISOTOPIC EVIDENCE FROM UWEINAT RING-COMPLEXES (S.E. Libya). ANDRE Luc, Musee royal de TAfrique centrale 1980 TERVUREN - BELGIUM

Sr AND Nd ISOTOPIC CHARACTERISTICS OF THE CORNUBIAN BATHOLITH, SOUTH-WEST ENGLAND ANTHONY, Elizabeth Y., Dept. of Geol. Sci., Univ. of Texas, El Paso, TX 79968, USA

Two groups of samples are considered : 1) the coarsegrained rocks (alkali granites, syenites, foid syenites) that form the major part of the ring-complexes (43 Ma); the fine-grained rocks from the downfaulted central part of the complexes (phonolites, trachytes) and the cogenetic dyke suites (basalts, phonolites, trachytes) that cut the complexes and the surrounding gneisses between 41 Ma and 27 Ma. Four groups of rocks are identified isotopically : 1) the foids syenites and the alkali basalts that fit a small domain of the "Lo Nd" array : e... = +3.3 to + 1.3; = - 21.6 to-13.3; ^^ 2) the syfenites and the alkali granites from theUweinat peralkaline and peraluminous series that fall on the "mantle" array : e... = + 1.8 to+1.6; ec = - 5.1 to - 4.0; ^^ ^^ 3) the Archenu alkali granites scattered around the Sr radiogenic side of the "mantle" array : e.,. = + 2.4 to -1.4; = + 9.1 to + 11.9; ^^ 4) the dyke cutting the surrounding Archean rocks that are characterized by a "crustal" isotopic signature : down to - 2.8; e^^ up to.+ 100). These data reveal that : 1) the source region of these A-types granites is to be found in mantle sources comparable to those of oceanic and continental alkali basalts; 2) the coexistence of oversaturated and undersaturated rocks could result from:a) the existence of two geochemically distinct components in the source region, a low ®^Sr/®®Sr (HIMU ?) component and a slightly depleted component from de "mantle" array; b) interactions of the low ®^Sr/®®Sr magmas with lithospheric (crustal or mantle) components; 3) the peraluminous and peralkaline magmas evolved from cogenetic magmas.

The Cornubian Batholith is a strongly peraluminous, tin-mineralized granite of Hercynian age. The main batholith, which constitutes the spine of Cornwall, is intruded by highly-evolved, topaz-bearing granite. This study reports on Sr and Nd isotopic compositions of both mineral separates and wholerock samples and was undertaken to constrain hypotheses concerning the lithology of source materials for the parent magma. All samples, including the topaz granite, and their constituent minerals have an almost constant of -6 (la - 0.35, n = 13). The limited range in values implies a high degree of isotopic homogeneity for the magma and a close genetic relationship between the main phase of plutonism and the later topaz granites. e^a values of about -6 are typical of other Hercynian-aged granites of France and Germany and imply a uniformity in source materials and processes of magma genesis across the belt. Sr isotopic compositions, on the other hand, are variable: minerals with high Rb/Sr values (such as the micas) are discordant with respect to isochrons calculated from whole-rock samples and minerals with low Rb/Sr values. This behavior extends to whole-rock samples with high Rb/Sr, such as the topaz granites, and to megacrystic phases of the granite. The majority of the discordant points lie to the right-hand side of the isochrons, causing them to have low ®'Sr/®®Sr,. Samples with minimal disturbance show little variability and cluster at a "Sr/®®Sr, of 0.711 to 0.712. These values accord with the Nd isotopes in implying a mature crustal lithology for the source materials of the magmas.

OXYGEN ISOTOPE EVIDENCE FOR THE ORIGIN OF PGE M I N E R A L I Z A T I O N IN A L A S K A N - T Y P E I N T R U S I V E S AT FIFIELD, EASTERN AUSTRALIA. A N D R E W , A . S . , C S I R O , D i v i s i o n of Exploration Geoscience, North Ryde, NSW, 2113, Australia, HENSEN, B.J., and DUNLOP, A.C., U n i v e r s i t y of N S W , K e n s i n g t o n , N S W , 2033, Australia.

ISOTOPIC EVIDENCE FOR THE SUPERGENE ORIGIN OF ALUNITES IN SEDIMENT-HOSTED MICRON GOLD DEPOSITS, WESTERN USA AREHART, Greg B., Dept. of Geol. Sciences, University of Michigan, Ann Arbor, MI 48109-1063, U.S.A.; and FOLAND, K.A., Dept. of Geol. & Mineral., Ohio State University, Columbus, OH, 43210, U.S.A.

P G E m i n e r a l i z a t i o n is h o s t e d b y pyroxenites and peridotites within zoned Alaskan-type intrusive complexes at Fifield, NSW. The Owendale and adjacent Tout Intrusive Complexes comprise p e r i d o t i t e , d u n i t e , clinopyroxenite, hornblendite, monzogabbro, monzodiorite a n d m o n z o n i t e . M a g m a t i c l a y e r i n g is variably developed in m o s t r o c k t y p e s e x c e p t dunite and peridotite. Pd-poor and S-poor mineralization is p r e d o m i n a n t l y in cross-cutting clinopyroxene-rich veins and pipes (P-units). Oxygen isotope fractionations between clinopyroxene, plagioclase and magnetite indicate high temperature crystallization. S^^O values of clinopyroxene and magnetite from least altered pyroxenites and P-units are 5.5 and 4.1 permil V - S M O W , respectively, suggesting magmatic crystallization temperatures as high as 1300'C from a mantled e r i v e d m a g m a w i t h S ^ ^ O of 5 . 7 . A l l other values plot within the triangular fields defined by these primary values on S-6 plots, demonstrating closed-system subsolidus exchange t o <670°C a n d n o i n g r e s s b y externally-derived f l u i d s . In p y r o x e n i t e s a n d Pu n i t s , e x c h a n g e of has enriched clinopyroxene and biotite and depleted magnetite. In t h e m o r e f e l s i c r o c k s , f e l d s p a r h a s been enriched and clinopyroxene and magnetite have b e c o m e d e p l e t e d in PGE mineralization h o s t e d w i t h i n P - u n i t s is m a g m a t i c in o r i g i n .

Alunite is a common mineral associated with micron gold deposits along the "Carlin Trend" in western North America. The Trend is a major gold-producing feature containing reported reserves in excess of 60 million ounces.Textures and mineral associations strongly suggest a supergene origin for these alunites. They occur as fine-grained, earthy, nearly pure fracture fillings associated with jarosite and other iron oxides, and other supergene minerals such as pharmacosiderite, crandallite, symplesite, and stibiconite. Alunites from several deposits have been investigated by isotopic methods. Dating of primary (hypogene) mineralization at Post is constrained to prior to 40 Ma, and most other deposits are thought to be of a similar age. In contrast, K/Ar and ^krP^Ax dates on alunites from these deposits cluster near 10 Ma. Such dates are in concordance with major weathering cycles associated with uplift and development of the Basin and Range. Sulfur isotope values from primary pyrite range from +3 to +15 permil; alunite values overlap that range, strongly suggesting that alunite formed from weathering of pyrite. Barite from the same veins as alunite is a coarsely crystalline, earlier phase; barite mineralization extends to great depth (>500 m) whereas alunite is restricted to near-surface (<150 m). 5^S values of barite from these veins are higher (+25-35 permil), indicating that the barite is of different origin from alunite. 5D values of alunite closely approximate local meteoric water (-110 permil); when combined with data, alunites fall very close to the "supergene alunite line."

ICOG 7

ION MICROPROBE STUDIES ON ZIRCONS FROM THE BARBERTON MOUNTAIN LAND: IMPLICATIONS FOR ARCHAEAN GREENSTONE BELT STRATIGRAPHY. ARMSTRONG. R.A., and DE WIT, MJ.*, Departments of Geochemistry and * Geology, University of Cape Town, Rondebosch, 7700, S. Africa, COMPSTON, W., and WILLIAMS, LS., RSES, ANU, GPO Box 4, Canberra, 2601, A.CT, Australia. The Barberton Greenstone Belt, with some of the best preserved, old rock sequences, contains a stratigraphic succession that has long been used as a model for early Earth processes. Recent field and geochemical studies, however, suggest a need to test the stratigraphy as it is presently adopted. In order to establish the original order of deposition and to chronicle the evolution of the greenstone belt, a high resolution time frame has been established by ion microprobe studies on zircons from critical lithologies. The new zircon data indicate that major stratigraphic and structural breaks as well as tectonic repetitions exist in the type area of the Onverwacht Group. Specifically (1) an age of 3538±6 Ma for a tectonic wedge of tona itic gneiss within the Theespruit Formation confirms the presence of a sialic basement and deformed unconformity below that unit. This is the oldest unit yet identified in the greenstone belt. (2) Volcaniclastic sediments of the Theespruit Formation could be younger than the (structurally) overlying mafic-ultramafic volcanics of the Komati Formation. (3) The intermediate igneous rocks of the Hooggenoeg and Theespruit Formations, as well as the trondjhemitic Theespruit pluton are equal in age at c. 3440 Ma. A second major episode of intermediate igneous activity is documented at c. 3250 Ma. Both episodes are associated with major episodes of regional thrusting, evidence that the greenstone belt evolved, at least in its later stages, near or within an active convergent tectonic environment. The results clearly support the suggestion that the rock sequences of the Barberton Mountain Land are part of a litho-tectonic complex. The relative disposition of the original units is not known, but cannot be assumed to be as simple as originally described.

THE PERSISTENT MYTH OF CRUSTAL GROWTH

ARMSTRONG. Richard Lee, Geological Sciences, University of British Columbia, Vancouver, B. C. Canada, V6T 2B4 From the extraterrestrial telescopic, space probe, meteorite and returned sample studies of planetary evolution and terrestrial evidence for early differentiation of core and fluid spheres I feel the conclusion is inescapable that large terrestrial planets of our solar system underwent essentially immediate differentiation into relatively constant core, depleted mantle, enriched crus^ andfluidreservoirs. Differentiation was an early event, carried rapidly to completion. It is a false premise to havetiieformation of sialic crust on Earth dragged out over billions of years after hot accretion. The uniqueness of the Earth arises from its size,retentionof water, and dynamic surface-renewal processes, which have effectively erased all vestiges of the first several hundred million years of its crustal history. A large volume of depleted mantle always existed and an isotopically nearly homogeneous character for mande and crust was only sustained by rapid convective stirring of the silicate Earth. The sigmoidal crust stabilization history curve that is recorded in whole rock Nd and zircon U-Pb dates is a predictable consequence of the highly exponential decline in stirring rates. Positive e Nd values for Archean mande-derived magmas are quantitatively predicted witii this view. Current rates of crustal accretion and loss of crust by sediment subduction and tectonic erosion are in balance, indicating negligible crustal growth at present. Claims that current growth is approximately 1 km3/a are based on incorrect and selectively cited data and have survived and thrived byrepeatedcitation. The idea that the earths crust has grown is a myth, dating from the 19th Century, and established as geochemical dogma in the 1950's and 1960's. It has survived by inertia and endless self-citation. The alternate no-growth view has been sometimes ignored, frequendy questioned and downplayed, often cited only as an end member hypothesis (on the presumption that the truth must lie between extremes), and sometimes acclaimed and supported. Evidence in its favor has been accumulating. The growth myth has survived, however, as the consensus, and been defended with biased models and selected,.and even incorrect,.data, and incomplete theoretical analyses. In science conventional wisdom is difficult to discard. After nearly 25 years the implications of plate tectonics have yet to be fully appreciated by many isotope geochemists.

ABSTRACTS

AGE AND GROWTH RATE OF CONTINENTAL CRUST: A PROJECT OF THE INTERNATIONAL LITHOSPHERE PROGRAM ARNDT, NT. Max-Planck-lnstitut fur Chemie, Saarstr 23. 6500 Mainz. FRG The International Lithosphere Progranfi has started a project intended to infiprove our understanding of the mechan sm and timing of continent growth by making a global compilation of Nd isotope data and U-Pb zircon ages. The project aims to answer the following questions: (1) When did most continental crust form: pre-4 Ga. at the end of the Archean. or throughout geological history? (2) Was continental growth continuous, episodic on a global scale, or episodic on a continental scale? Do individual continents or continental regions have unique growth histories? (3) What was the growth mechanism: accretion of island arcs, reworking of basaltic underplates, plumerelated orogenesis, a uniquely Archean mechanism? (4) How did evolution of continental crust influence the composition and physical state of the mantle? The procedure involves: (1) preparation of maps based on geological, geochronological and geophysical information that define lithotectonic terranes within the continents; (2) compilation of Nd isotopic data and U-Pb zircon ages from all continents; (3) use of Nd isotopic data to define crust formation ages, i.e. the average age of the components of each portion of the crust; and U-Pb zircon ages to establish the timing of orogenic events. A combination of the two types of data enables crustal growth to be distinguished from crustal reworking, and can be used to establish a crustal growth history for each continental region; (4) this information is then used to constmct growth histories for individual continents, which provide a basis for constructing a global crustal-growth curve. Representatives of the task group are currently gathering data from literature and unpublished sources. A preliminary compilation and map will be completed within two years, and final compilation and interpretation is planned for 1995.

534s IN HAMERSLEY BIF, W. AUSTRALIA: THE ROLE OF BACTERIAL ACTIVITY AND THIOSULFATES AS AN ALTERNATIVE TO THE HYDROTHERMAL MODEL ARNOLD. Michel, ALIBERT, Chantal, and JACQUIER, Bertrand, C.R.P.G., B.P. 20, 54501 Vandoeuvre Cedex, France Various lithologies from the Jeerinah, Marra Mamba, Dales Gorge and Joffre formations (black shales, carbonate-cherts and iron oxides facies), deposited in the Hamersley Basin between 2.7 and 2.5 Ga, have been analyzed for sulfur isotopes. Sulfides consist of pyrite and pyrrhotite either disseminated or in mm to cm thick layers. The Jeerinah and Marra Mamba samples give 534S from +3.5 to +9%o (± 0.2%o). Conversely, preliminary data from Dales Gorge-Joffre give 534S values close to 0%o. The former highly positive values seem restricted to Jeerinah-Marra Mamba, characterized by a more reducing character (higher C and S contents, abundant siderite) compared to the overlying Dales Gorge-Joffre formations. Therefore, a hydrothermal model for the origin of S is not supported. Moreover, available SH2 is expected to be easily oxidized into SO by iron hydroxides particulates near vents and on the other side, in modem ocean-ridge systems, 534S of sulfides rarely exceeds 7%o, Alternatively, both bacterial reduction of S04~" (or SO3-) or a thiosulfate decomposition process can explain the measured 534S range. If a small ASO4-XS fractionation is involved (low sulfate content and/or higher seawater T, compared to present-day T), then positive 534S values can be obtained through the biogenic process and the 9%o value represents the minimum composition for marine aqueous sulfates. The evolution from a primitive Archean ocean dominated by SO3"" and SH2 to a modem sulfate-rich ocean, requires intermediate species such a S406"' and S2O3, well known components in modem sub-oxic waters. In a similar way, S04"" and FeS can be produced by reaction between S 2 O 3 - and dissolved iron (Fe++) in a purely inorganic process. Starting from a 534S « 9%o, experimental fractionation data (Agarwala & Thode, 1969) suggest 534S values from -9 to +9 for the produced sulfides, which is actually observed in Dales Gorge-Joffre.

ABSTRACTS

ICOG 7

THE NUCLEOSYNTHESIS CONTRIBUTIONS TO THE SOLAR NEBULA ARNOULD, Marcel, Institut d'Astronomie, d'Astrophysique et de Geophysique, University Libre de Bruxelles, B-1050 Bruxelles, Belgium The discovery of isotopic anomalies in meteorites suggests that the Solar System is made of material from compositionally different and imperfectly mixed reservoirs. One of them comprises the bulk Solar System material, and is considered to be made of the well-homogenized ashes of many nucleosynthesis events. As a consequence, models for the chemical evolution of the Galaxy are called for in order to study its composition. The major nucleosynthetic agents responsible for that evolution are reviewed. The remaining reservoir(s) contain(s) the isotopically anomalous matter, which represents only a very minute fraction of the total Solar System material. It is stressed that such anomalies provide new clues to many important astrophysical problems, and in particular to the early history of the Sun and of Solar System solid bodies, as well as to the nucleosynthetic yields of various types of stars. In fact, the isotopic anomalies offer the very exciting perspective of testing nucleosynthesis models for a very limited number of events, even possibly a single one. This situation is in marked contrast with the one encountered when trying to understand the bulk Solar System composition. Some selected examples are given to illustrate the rich diversity of potential nucleosynthetic mechanisms that could have produced the isotopically anomalous material. In particular, a review is presented of the possible contributions of Red Giant stars, massive mass loosing objects (of the Wolf-Rayet type), novae or supernovae to the observed anomalies, and in particular to those ones due to the in situ decay of radionuclides with half-lives in excess of about 105 yr. Some recent nucleosynthesis predictions based on new nuclear physics data are presented.

GEOCHEMICAL AND ISOTOPIC VARIATIONS ALONG THE ROZA DIKE SYSTEM, COLUMBIA RIVER BASALT GROUP ATKINSON. Suzan and LAMBERT, R. St. J., Department of Geology, University of Alberta, Edmonton, Alta., T6G 2E3, Canada The Roza Member of the Columbia River Basalt Group erupted along a narrow NNW trending zone of fissures and vents in eastern Washington and northeastern Oregon around 15 Ma ago. This line of eruptive centers cuts across the suture between a craton margin and accreted Mesozoic oceanic terranes. The Roza flow has been divided by Martin into six chemical subtypes, also recognizable along the feeder system: thus the nature of the magma source, feeder emplacement and crustal contamination can be deduced. Each subtype erupted along overlapping segments of varying lengths. No systematic migration of activity with time can be seen. The initial five subtypes are characterized by a gradual increase in Ca and Cr with time, and a steady decrease in Ti, P, Zr, and Nb. High precision Sr isotope analysis, using a quintuple collector array, and internally consistent to +/- 0.00003, shows that 87/86 Sr varies randomly from 0.7051 to 0.7053. In contrast, subtype IV has 87/86 Sr = 0.7054 and high incompatible element abundances. 87/86 Sr of plagioclase phenocrysts from all subtypes equal 0.7051. 143/144 Nd varies from 0.51253 to 0.51267 +/- 0.00001: Pb isotopic variation is within error of analysis. These minute variations in isotopic properties place severe constraints on the nature and location of contaminants. All subtypes were derived from a common fractionating source. Any contaminant is unlikely to have been ancient criist, as Nd and Pb in the nearest known crust to the north are distinct to that in the Roza Member.

BONE PHOSPHATE OXYGEN-ISOTOPES OF MODERN AND FOSSIL ELEPHANTS L.K. AYLIFFE and A.R. CHIVAS, Research School of Earth Sciences, Australian National University, A.C.T., 2601, Australia An attempt at calibrating the bone phosphate oxygen-isotope (6^^0p)v5 environmental water oxygen-isotope (S^Ôw) relationship for modern elephants has been made. With this information extrapolations into past environmental water oxygenisotopes (d^Ôw) regimes, and hencc cUmatcs, using fossil elephant bone material may be possible. Two species of modem elephants Elphas maximus (Indian elephant) and Loxodonta africana (African elephant) from a wide range of S^Ôw - climate regimes are observed to have S^Ôp that vary linearly with changes in local 5l80w. Despite slight differences in the diet and physiology of these two species of modern elephant, little difference exists in their vs relationships. Given that Elpfias maximus is anatomically closer to many of the now extinct mammoths than it is to present day Loxodonta africana, it is also quite likely that many fossil elephant species have similar S^Ôp - S^Ôw relationships. S^Ôp data for mammoth from known glacial and interglacial deposits in Great Britain and Europe will be presented, highlighting the potential of this technique for palaeoclimatic reconstructions.

OXYGEN-ISOTOPE (PO4) CONSTRAINTS ON THE ORIGIN OF ISLAND PHOSPHATE DEPOSITS L.K. AYLIFFE, H.H.VEEH* and A.R.CHIVAS, Research School of Earth Sciences, Australian National University, A.C.T., 2601, Australia. *School of Earth Sciences, Flinders University of South Australia, Bedford Park, 5042, Australia. In an attempt to test the hypothesis that insular phosphorites on high islands (e.g. Nauru) formed by chemical reations between seabird guano and carbonate, the oxygen-isotopic composition of phosphate (Sl^Op) in modem guano and in the apatite component of phosphorites of various ages and locations have been determined. As indicated in Figure 1, a distinction between recent apatites which are closely associated with modern guano, and ancient apatites is apparent. If of apatite is unaltered during diagenesis, the results indicate that ancient apatites could not have formed by similar processes without invoking unreasonably high paleotemperatures and/or unrealistic values for seawater. Either the major phosphorite deposits on high islands are unrelated to seabird guano, or the in apatite is reset during diagenesis and thus merely reflects the Sl^Ow of meteoric waters at therespectivesites.

6X). {%o)

Fig. 1. Plot of 6l80p vs 5l80w at different temperatures under equilibrium conditions (t=111.4-4.3(5p-5w)) and Op of various islands phosphates.

A N O M A L O U S N E O N A N D X E N O N I S O T O P E S IN A N ARCHAEAN ANORTHOSITE FROM GREENLAND AZUMA, S., Geophysical Institute, Univ.of Tokyo, Tokyo 113, JAPAN, OZIMA, M. and HIYAGON, H., Geophysical Institute, Univ.of Tokyo, Tokyo 113, JAPAN. Jeffery(^) first reported large anomalies on ^-^Ye, ^îê and ^^^Xe relative to the atmospheric Xe in an Archaean anorthosite(2.9Ga) from the Fiskenaesset Complex in West Greenland. Later, Zadnik and Jeffery^-^ observed anomalies on ''^Ne and '^Ne on the anorthosite. Since the isotopic anomalies recorded in the rock may be the reflection of some specific Archaean geological conditions, we decided to reexamine the problem. We studied an anorthosite from the same locality. We separated hornblende and anorthite from the anorthosite, and made isotopic analyses on the mineral separates. We observed that Xe is essentially atmospheric except for slight ^^Û fission Xe in hornblende in the higher temperature fraction. The result did not confirm the observation by Jeffery. However, we also observed considerable anomalies in both ^^Ne and -îVe in accordance with the results by Zadnik and Jeffery. The former anomaly is most reasonably attributable to and the latter to ^^F. However, we cannot rule out the possibility that the excess in Ne was produced by the cosiTiic ray irradiation, though this would require more than 600Ma surface residence of the anorthosite (less than a few meter from the surface). References (1) Jeffery,P.M.,1971./soiopic composition of xenon from a Greenland anorthosite. Nature (London), 233:260-261. (2) Zadnik,M.G. and Jeffery,PM.,1985. Radiogenic neon in an Archaean anorthosite. Chemi. GeoL, 19-125. GEOCHRONOLOGY OF THE OLONDO GREENSTONE BELT BAADSGAARD. H.. Dept. of Geology, University of Alberta, Edmonton, Alberta T6G2E3, Canada; NUTMAN, A. P., Research School of Eanh Sciences, Australian National University, Canberra ACT 2601, Australia and SAMSONOV, A. v.. Dept. of Petrology, IGEM, USSR Academy of Sciences, Staromonetny 35, Moscow 109017, U.S.S.R. The Olondo Greenstone Belt (0GB) in the central portion of the Olekma Granite-Greenstone Terrain of the Aldan Shield in East Siberia has a narrow synformal structure, shows complex deformation, has steep limbs and splits into two branches in its northern portion. The maximum width of the 0GB is 10 km, but because of folding the true stratigraphic thickness must be less. The 0GB consists of interlayered volcanics and sediments, and is dominated by mafic volcanics, ranging in composition from komatiitic to tholeiitic. Felsic volcanics are also present and zircons from them were analyzed by ion probe (SHRIMP). The westernmost volcanic unit in the 0GB yields a crystallization age of 2986 ± 6 Ma, central and central-western samples 2998 ± 9 Ma and easternmost samples 3003 ± 11 Ma and 3005 ± 5 Ma. The maximum difference in the ages is 17 million years, but the oldest and youngest ages just overlap at the two sigma uncertainty. Thus the somewhat less than 10 km stratigraphic sequence of the OGB accumulated over a maximum time of 17 ± 17 million years. Zircons from orthogneisses adjacent to the OGB have also been dated by SHRIMP. A tonalitic gneiss gave an age of 2862 ± 7 Ma, significantly younger that the OGB. On the other hand, a dioritic gneiss yielded an age of 3018 ± 10 Ma, within uncertainty the same age as that for the OGB felsic volcanic rocks.

ICOG 7

ABSTRACTS

FLUID FLOW HISTORY OF E. TRIASSIC NARRABEEN GROUP SANDSTONES OF THE SOUTHERN SYDNEY BASIN BALG. & KEENE,J., Department of Geolosy, Sydney University, NSW 2006, Australia, and EADINGTON, P. J. & HAMILTON,PJ., CSIRO Div. Exploration Geosci.,POB 136, NRyde, NSW 2113, Australia Petrographic observations and stable isotope and fluid inclusion measurements on diagenetic phases of Narrabeen Sandstones have been obtained in order to document the thermal and chemical evolution of pore fluids during burial. Extensive carbonate cementation is common with precipitation of grain coating morphologies (calcite and siderite) followed by pore filling occurrences of calcite, siderite and ankerite. Subsequently stacks of pseudohexagonal kaolin platelets were formed and later overgrown by diagenetic quartz. The latest cements to form were calcite, ankerite and filamentous pore bridging illite. Oxygen isotope compositions of various cements may be used together with the fluid inclusion data and observations on the relative timing of diagenetic events to constrain the evolution and source(s) of pore waters during burial. Such data indicate progressive enrichment in ^Ô of pore water during burial from depositional 51^0 of approx to approx. -6%o during quartz overgrowth formation. If the illites formed at higher temperatures than did the quartz overgrowth, then pore water S^Ô must have been even more ^Ô - enriched at d^Ô ~ -A7oo. Alternatively illite formation may have occurred at lower temperature as a result of meteoric water flushing at the time of uplift. In this case, the illite 5D values of about -115%o suggest that such water was pre-Tertiary in age. Homogenization temperatures of fluid inclusions in diagenetic quartz lie in the range of 95-120' C which is 40-80* C higher than current formation temperatures at the sampled depths of c a 0.5km. Modelling of vitrinite reflectance values and these fluid inclusion data suggest that the higher temperatures resulted from a combination of a maximum heat flow of 2.1 HFU, somewat higher than the present value of 1.7-1.9HFU, together with loss of c a 1.7km of section consequent to Tertiary uplift and erosion.

THERMOCHRONOLOGY OF A YOUNG OROGENIC BELT, D'EKTRECASTEAUX ISLANDS, PAPUA NEW GUINEA BALDWIN. S.C. Research School of Earth Sciences, The Australian National University, GPO Box 4, Canberra ACT 2601, Australia The D'Entrecasteaux Islands (DI) located off the northeastern coast of the P a p u ^ Peninsula, Papua New Guinea (PNG) occupy a tectonic setting in which the Woodlark Basin spreading centre is propagating westwards into continental crust of the Papuan Peninsula. Domes of high grade metamoiphic rocks (i.e., amphibolites, eclogites, granulites, and migmatites), intruded by large granodiorite bodies, form the basement complex on the DI. A therniochpnologic study of the DI has provided insight into the cooling histor>' of the metamorphic rocks and the role of igneous processes in the formation of the domes. Gneisses contain a primary assemblage of plagioclase+quanzi:gamet+nitile±p>Toxene ± amphibole. Retrograde assemblages include biotite and hornblende coronas around gamet, amphibole replacing pyroxene and minor sericitization of plagioclase. The gneisses cooled rapidly (>100oC/Ma) as indicated by hornblende and biotite ^Âift'Âi apparent ages of -2.5-3.0 Ma and-1.5-1.7 Ma, respectively. Plagioclases yielded saddle-shaped spectra interpreted to be the result of incorporation of vaning degrees of excess Ar. Plagioclase minimum apparent ages of 4.2 and 2.3 Ma are interpreted as the maximum age(s) of thermal event(s) which caused outgassing of plagioclase. Granodiorite which intrudes the basement contains xenoliths of retrograded eclogite. Âif^Âi apparent ages for hornblende,biotite, muscovite, and K-feldspar from the rind of one ^ÂrZ-^Âr mineral ages for retrogressed gneisses, schists, and granodiorites are 2.0-3.0 Ma (amphiboles), 1.5-1.7 Ma (muscovites), and 1.4-1.8 Ma (biotites) indicating emplacement of granodiorites into high levels of the crust closely coincided with retrogression of the metamorphic basement. The granodiorite bodies associated with tiie DI domes represent syn-kinematically emplaced granitoids intruded into an area of active continental extension.

ABSTRACTS

ICOG 7

ORIGIN OF 2.0 TO 2.4 Ga Nd MODEL AGE CRUST IN THE CHEYENNE BELT, SE WYOMING, USA BALL,Thecxiore,T and FARMER,G.Lang, Dept. of Geological Sciences and CIRES Campus Box 216, University of Colorado, Boulder, CO, 80309-0216, USA Nd isotopic studies of Precambrian continental crust in North America have revealed that 2.0-2.4 Ga Nd model age crust commonly separates Archean cratons from younger (1.7-1.8Ga) accreted island arc terranes. The origin of the 2.0-2.4 Ga model age crust is enigmatic and could represent new crustal material formed at 2.0-2.4 Ga, or Proterozoic crust that contains a component of intermixed Archean crustal material. In order to further examine the origin of 2.0-2.4 Ga Nd model age crust a detailed Nd isotopic study was undertaken of Precambrian crustal rocks in the Cheyenne Belt (CB), a Proterozoic suture zone separating the Archean Wyoming craton for the Proterozoic Colorado Province (CP). North of the CB, Proterozoic miogeoclinal rocks have Archean model ages (TDM=2.6-2.8 Ga) and require a sediment provenance solely within the Archean craton. South of the CB, CP crust has T ^ ^ =1-8-2.1 Ga and record no incorporation of Archean crustal material. However within the CB itself paragneisses have T^^ = 2.1-2.4 Ga, intermediate between the values determined for crust north and south of the CB, values that are similar to other 2.0-2.4 Ga Nd model age crust in North America. The location of the paragneisses between the Archean and Proterozoic crust and their probable sedimentary origin suggest that these rocks could be oceanic trench or foredeep basin sediments that were deposited as the Archean and Proterozoic crust converged. The intermediate model ages are most likely the result of mechanical mking of sediments derived from the Archean craton and the CP. The overall lack of 2.0-2.4 Ga Nd model age crust in the area of the CB, compared to the Penokean orogen in the northcentral U.S., can now be explained within the existing model for the formation of the CB. This model postulates overthrusting of the forward portion of the CP onto the Wyoming craton. Subsequent uplift and erosion has now removed the foredeep sediments north of the CB.

RARE GAS TRACING OF CRUSTAL FLUIDS AND HYDROCARBON ACCUMULATION IN EXTENSIONAL BASINS C J BALLENTINE. R K O'NIONS ( Dept. E ^ h Sciences, University of Cambridge, England ) The contribution from mantle, cmstal and atmospheric sources to the noble gas inventory of any crustal fluid ( eg. groundwaters and hydrocarbons ) may be cdculated from the distinctive isotopic stmcture of the noble gases and their relative abundances .In active extensional basins, hydrocarbon maturation and accumulation may be accompanied by Ae emplacement of mantle derived melts and strongly time dependent themial regimes in the crust. These aspects of basin evolution are examined in relation to crustal fluids in two Neogene hydrocarbon bearing basins. Natural gas reservoirs in the Pannonian Basin, Hungary and the Vienna Basin, Austria, provides samples of different hyi-ocarbon maturity down to a depth of 5.5Km. Both basin systems were formed by extension in the Middle Miocene. Consideration of the mass balance of atmospherically derived ^^Ne and ^Âr gives the minimum volume of air equilibrated water required to emplace them, and enables the extent of sample interaction with the groundwater system to be identified and quantified. Methane samples measured from the Hajduszoboslo field, H u n g ^ have CU^/'^Âx ratios equivalent to methane saturated air equilibrated water under reservoir conditions. This is consistent with the dominant mode of methane transport being in solution, with subsequent degassing at the topological high of the reservoir. In all samples, between 2 and 5% of the helium is mantle derived and the radiogenic contribution to these samples increases with depth ( 2lNe/22Ne up to 0.0526, ^ÂT/^ÂI up to 2459 ). The radiogenic mass balance in both systems cannot be accounted for by steady state crustal production since basin formation. The deepest samples retain both the atmospheric and radiogenic end member ratios, while fractionation of both end member components in the shallower samples is in the sense predicted by solubility models.

MIXING PROCESSES BENEATH "DRY" SALT LAKES B A R N E S . Chris, C S I R O Division of Water Resources, PO Box 1 6 6 6 , Canberra City, ACT, 2 6 0 1 , AUSTRALIA, and WOODING, Robin, Centre for Environmental Mechanics, CSIRO, GPO Box 8 2 1 , Canberra. ACT, 2 6 0 1 , AUSTRALIA. Evaporation processes from salt lakes lead to the development of a relatively high density, high salt concentration layer of water near the surface. With sufficiently permeable lake sediments (characterized by a certain Rayleigh number) this leads to density instabilities which advectively mix the isotopically enriched waters from the evaporation zone with the underlying water. In addition the horizontal salinity gradient, produced by regional waters flowing towards the lake to satisfy the evaporative demand, results in the establishment of a large scale circulation beneath the lake, fed by the surface mixing. A mathematical model describing both these processes is used to predict the development of stable Isotope (^H, profiles beneath these systems, and to examine their dependence on system parameters such as evaporation rate, permeability and aquifer thickness. Published stable isotope and other data from a number of Australian salt lakes (Lakes Tyrrell, Frome and Amadeus) are re-examined in the light of this model, and salt and water movements beneath the lakes are inferred.

U-PB AGES OF METAMORPHIC MONAZTTES AND THE TIMING OF ACADIAN METAMORPHIC EVENTS IN THE NORTHERN APPALACHIANS, NEW ENGLAND, USA BARREIRO. Barbara!, BARNETT, Daniel, EUSDEN, J. Dykstra^, and SMITH, Harold, Dartmouth College, Hanover, NH, 03755 USA (now at ^NIGL, Keyworth, NG12 5GG, United Kingdom, and ^Bates College, Lewiston, ME, 04240, USA) Silurian-Devonian sediments deposited in the Connecticut! ValleyGaspe Trough (CVGT) and the Central Maine Terrane (CMT) were deformed, metamorphosed and intruded by Devonian and Carboniferous plutons during the Acadian Orogeny. Many of these pelitic schists contain monazitc. A detailed study of the Perry Mountain-Rangeley Formations in western Maine (Smith and Barreiro, in press) has shown that detrital monazites, if present, break down in biotite grade rocks to allanite and xenotime. Newly formed metamorphic monazites be^n to grow at staurolite grade, and are common to abundant in sillimanite and higher grade rocks. Since these monazites grow below their closure temperature (-700OC), thek concordant or slightly discordant U-Pb ages record their growth during prograde metamorphism. Two distinct groups of ages are revealed in the CMT. An older event (405-390 Ma) occurs in westem Maine and along the westem margin of the CMT in New Hampshire. The age range coincides with the syntectonic S-type plutons of the New Hampshire Plutonic Series, with dated hydrothermal activity at the Bristol graphite vein (Zeitler at al., in press) and with estimates for major nappe formation. Younger ages (375-353 Ma) occur in Maine and Massachusetts in the CMT, and in kyanite-staurolite schists in the CVGS in Vermont, but are for the most pan missing in New Hampshire. The younger ages coincide with ages for undeformed two-mica Concord granites. Discordant ages in southeastern New Hampshire between 390 and 360 Ma may record monazite growth during both events. Amphibolite facies conditions were attained at least twice during "Acadian" metamorphism.The heat transfer mechanism may have been different for each event Peak conditions often accompanied the intrusion of contemporaneous plutons.

ICOG 7 U^Pb ISOTOPE AND TRACE ELEMENT COMPOSITIONS OF BLACK AND KIMBERLEY KEEPS PYRITE, SOUTH AFRICA: IMPLICATIONS FOR AGE AND ORIGIN BARTON. E.S., Bernard Price Institute of Geophysical Research, Univ. of the Witwatersrand, P O Wits 2050, South Africa, and HALLBAUER, D.K., Dept of Geology, Univ of Stellenbosch, Stellenbosch 7700, South Africa In the Black Reef Quartzite Formation of the Transvaal Supergroup (East Rand Gold Field, South Africa) three types of pyrite occur in two gold-bearing conglomerate facies: 1) compact, rounded detrital, 2) rounded, porous and oolitic synsedimentary, and 3) recrystallised authigenic grains. Pb-isotope data for Type 1 pyrite from both facies suggest that all Type 1 pyrites crystallised about 3000 Ma ago from a common source that had a bulk earth composition. Pb-isotope and trace element data for Types 2 and 3 pyrite show very different patterns indicative of their sedimentary/authigenic origin. The Pb data for these pyrites define a steep linear trend on a Pb-Pb plot. If this trend represents an "anomalous Pb" line the maximum age of the source of the radiogenic contaminant is 3800 Ma and the time of formation of the radiogenic Pb component is 2500 Ma. The latter is a max i mum age for Black Reef deposition. Such steep trends, common for many secondary minerals and some rock units in the Witwatersrand depository, may reflect a widespread hydrothermal event. The Pb-isotope compositions, U, Pb and trace element contents of Kimberley Reef detrital pyrites are distinct from that of all the Black Reef pyrites. While the 3000 Ma age for the Black Reef detrital pyrites does not exclude a Witwatersrand source, these data* preclude the Kimberley Reef as a source for the pyrite.

A K-Ar PROFILE THRDUaHi TOE JOINVILLE ' MASSIF AND IHE DOM FELICIANO BELT,SOUIHEE^ BRASIL - TECTONIC IMPLICATIONS.

M.A.S.BASEI, O.SIGA JR. and K.KAWASHITA. IGc.Universidade de Sao Paulo, PO.BOX.20899, CEP 01498.Sao Paulo, Brasil. A K-Ar cooling age profile using biotite, amphibole, plagioclase, K-feldspar and whole rock data from metamorphic rocks of the Ribeira Belt, Joinville Massif and Dom Fcliciano Belt i s presented. The results are f a i r l y coherent v/ith crustal modeling previously obtained from geological and gravimetric data. Three main geochronological d<Dmains are distinguishable, froni m to SE: 1)Ribeira Belt (RB) and the northern part of the Joinville Massif (NJr4); 2)southern part of the Joinville Massif (SJTvl); and 3)Dom Feliciano Belt(DFB). All domains are separated by major thrust faults probably related to lithospheric discontinuities.Geochronological domains 1 and 3 yielded K-Ar values of 750-500 Ma and 600-500 Ma, respectively, v.tiich show a clear influence of the Brasiliano Cycle. The granulite terrain comprising the SJM exhibits ages betv;een 2200 and 1800 Ma, indicating that this area was cool during the Late Proterozoic Brasiliano Cycle. More detailed analysis of the profile reveals thermal differences at the tectonic contacts, interpreted here as a result of the geometrical characteristics of the nappes related to these boundaries. The SJM and NJM limit is marked by a rapid transition from Early Proterozoic ages (2000-1800 Ma) in the SJM, to Late PrxDterozoic ages (600 Ma) in banded gneisses of the NJM. This boundary i s also delimited by a discontinuous belt of mafic bodies and an increas^j in regional metamorphism near the southern limit of the NJM. By contrast, the contact between the SJIvI and DBF i s interpreted as a thrust at a high, relatively cool crustal level.

ABSTRACTS

GEOCHRONOLOGY AND GEOCHEMISTRY OF GROUP A ECLOGITES FROM THE MOLDANUBIAN ZONE OF THE BOHEMIAN MASSIF, CZECHOSLOVAKIA BEARa Brian L., MEDARIS, L. Gordon, Jr., JOHNSON, Clark M., Dept of Geol. and Geophys. University of Wisconsin, Madison, WI, 53706, USA, and BRUECKNER, Hannes K., Queens College, Palisades, NY 11367, USA Eclogite and garnet pyroxenite that occur as lenses within garnet peridotite have been investigated from five localities in the Moldanubian zone of the Bohemian Massif, Czechoslovakia, to establish the age and equilibration conditions of eclogite formation and to evaluate the relative contributions of mande and crust to their petrogenesis. Garnet and omphacite are compositionally zoned with homogenous cores and retrograde rims. Depending on locality, cores of minerals yield minimum temperatures and pressures of 850-1050'C and 15-25 kb, based on Fe-Mg exchange between garnet and clinopyroxene and the jadeite content in clinopyroxene. Eclogite samples range from ne- to hy- normative, are subalkaline, and define a tholeiitic AFM trend. REE patterns for eclogite are relatively flat to slightly LREE depleted. Elevated MgO and Cr contents indicate that these eclogites are not primary melts, and scatter in element variation diagrams suggests that they are cumulate in origin with up to 30% trapped melt-. Sm-Nd ages for gamet-clinopyroxene pairs are 374-327 Ma,* reflecting events related to the Variscan Orogeny, E^^ values and initial ratios are anticorrelated and vary from -3 to -1-6 and 0.7025 to 0.7067, respectively. Field, compositional, and isotopic data indicate that the eclogites were derived from depleted to enriched mantle and were transponed into the crust by enclosing mande-derived peridotites during Variscan Orogenesis.

Late Neogene ^'^Srl^^Sr isotopic record in the western Arabian Sea; ODP Leg 117.

C.J Beets, GeOmarine Center,Vrije Universiteit, De Boelelaan 1085, 1081 HV Amsterdam, The Netherlands. This poster demonstrates the potential of strontium isotopes as a chronostratigraphical tool for late Neogene marine sediments. The sediments which were recovered during ODP Leg 117 in the western Arabian Sea provide a continuous early Miocene to Present record. This interval can be dated with strontium isotopes with a resolution varying from 0.2-1.5 My due to the "stepwise" increase of the 8 ' 7 s r / 8 6 s r ratio in seawater. A comparison with earlier strontium evolution curves of DePaolo from the Tasman Sea and of McKenzie from the Mediterranean confirms the use of these curves as good chronostratigraphical tools for the late Neogene.

ABSTRACTS

ICOG 7

Measurement of cosmogenic 26ai using the ISOLAB 120 and determination of geological cosmic ray exposure histories

N. S. Belshaw. R. K. O'Nions, D. J. Martel, K, Burton, A. S. Cohen

(Dept. of Earth Sciences, Univ. of Cambridge, Downing Street, Cambridge. CB2 3EQ, U. K.)

The Cambridge ISOLAB-120 has an abundance sensitivity of <10-ii at 26 a.m.u. A1 may be ionised efficiently both thermally and by sputtering with high transmission at a mass resolving power of 6000 required to resolve 26^1 from '^^Mg, Thus, quantitative measurement of cosmogenic 26A1 in geological materials becomes possible. Techniques have been developed for the separation of A1 from meteorites, quartz and olivine with minimal mass spectrometric interference of ^^Mg on 26Al. The ion counting background on the ISOLAB120 is currently -0.1 c.p.s., and 26A1 may be quantified at 1 c.p.s or less. Chondrite meteorites, such as Lost City, with 2 wt% AI2O3, and 3 x lOio atoms(26Al).g-i, have 26A1/27A1 of -1.5 x lO-io which may be measured quantitatively using these techniques. Sample sizes of 1 mg may be measured, a factor of 10^ to 10^ lower than those used for counting experiments. 26a1/27a1 ratios are even more favourable in some low-Al targets, e. g. quartz and olivine. Results from such samples are presented and compared with cosmogenic ^He exposure age estimates.

FORMATION OF DEPLETED RESERVOIRS IN THE EARLY ARCHEAN EARTH: CONSTRAINTS FROM COMBINED Nd AND Os ISOTOPIC MODELLING BENNETT. Victoria C., and McCULLOCH, Malcolm T., Research School of Earth Sciences, The Australian National University, Canberra, ACT 2601, Australia It is established, primarily on the basis of initial Nd isotopic compositions of Archean rocks, that a depleted reservoir(s) has existed since the early Archean. The extent and timing of the depletion events however, are not well-documented. In order to further investigate the formation of depleted reservoirs in the early Earth, we are determining the initial Nd isotopic compositions of early Archean samples with well-defined field relationships, and with crystallization ages determined by U-Pb systematics of zircon using the SHRIMP. The most ancient tonalitic gneisses from Mt. Narryer, Western Australia (3.73 Ga, Nutman et al, 1990) have a narrowly defined range of initial from +1.7 to +1.9 . A tonalite from the Amitsoq gneisses, Greenland (3.87 Ga, Nutman, this volume) has a value of +2, whereas two slightly younger gneisses have of +3.2 and +3.1. These samples further document the existence of a widespread depleted reservoir in the Archean earth, although the values we measure are not as radiogenic as previously suggested (cf. Shirey et al, 1986). Osmium isotopes have the potential to test models for formation of depleted reservoirs owing to the differing behavior of the Re-Os and Sm-Nd isotopic systems. For example, the Nd isotopic depletions reflected in the early Archean crust can be produced by long-term storage of mafic-ultramafic material (e.g. Chase and Patchett, 1988X however the complementary Os isotopic compositions, calculated based on reasonable Re-Os partition coefficients, should reflect a retarded evolution of the 1870s/l860s ratio in the depleted source. These effects are not observed, either in published Os isotopic data (e.g. Luck and Allegre, 1980) or in our measurements of Archean osmiridiums from the Pilbara Block of Western Australia. Based on these considerations, we favor formation of a depleted reservoir by progressive extraction of continental crust, with the the relative masses of the depleted mande and crust remaining approximately constant. Further work, using combined Nd and Os isotopic measurements is in progress.

EAST GREENLAND APATITE FISSION TRACK AND MAGMATIC ROCK RECORD OF NORTH ATLANTIC SPREADING AND HOTSPOT EVENTS BERGMAN, S.C.. HENK, Fil., ARCO OU and Gas Co. and ARCO International OU and Gas Co., Piano, TX 75075 USA, and HANSEN, K. Institut for Petrologi, 0ster Voldgade, Copenhagen, Denmark Eocene alkali and tholeiitic basalt dikes and sills are important components of the Jameson Land basin. New K-Ar age (n=6: 50-55 Ma) and petrologic data (n=30) demonstrate the intrusive rocks are part of the regiond Eocene basaltic magmatic suite that fomied during the major Paleogene rifting episode in the North Atlantic at the time the Iceland hotspot moved past East Greenland. The thermal histories of 65 surface Pemiian to Jurassic sedimentary rocks in the Jameson Land basin are constrained by apatite fission track analyses. Mean or pooled apatite fission track ages range from 12 to 230 Ma; many samples possess multiple populations with variably reset grains. Mean track lengths are mosdy 9-12 jum with only a few samples having mean track lengths of 13.3-14.7;am. Two distinct groups are recognized: totally reset samples with mean ages <80 Ma and slightly or non reset samples with ages of 111-230 M^ Rocks are variably reset by Eocene basalt intrusions. A 1000 km^ area at Weper Halvo is uniformly reset (apatite ages of 12-25 Ma). The apatite fission track data record two post-Eocene uplift/erosion events, each correlating with regional plate tectonic events. During the first phase from 55 to 22 Ma, the section slowly cooled tfirough 120-80 OC (at 1+/-0.2 ^C/km) and was slowly uplifted 1 km (at 30+/5 m/m.y.) through 3-4 km depths. The onset of this event correlates with the opening of the North Atlantic at magnetic anomaly 24 time. The more recent phase began at the time the Kolbeinsey Ridge was initiated at 22 Ma (anomaly 6 time). During the last 22 m.y., rocks now exposed at the surface rapidly cooled through 80-0 °C (at 4+/-1 oc/m.y.) and were rapidly uplifted 2-3 km at rates of 130+/-20 m/m.y. Anomaly 6 oceanic cmst occurs east of Jameson Land. The age and composition of magmatic rocks and apatite fission track analyses on sedimentary rocks document Eocene to Miocene plate tectonic events associated with spreading in the North Atlantic. Initiation of spreading in the Norwegian-Greenland Sea was coincident with a widespread mantle melting event associated with the Iceland hotspot.

EVIDENCE OF RAPID TEMPERATURE CHANGE ACROSS K/T BOUNDARY : fi^^O SIGNATURE IN MANGYSHLAK SEDIMENTS. S.K.BHATTACHARYA and A.SARKAR, Physical Research Laboratory, Ahmedabad 380009, India and D.P.NAIDIN, Geological Faculty. Moscow State Unsiversity, Moscow, USSR. An iridium enrichment (a factor of/-100) has been found in K/T boundary clay from a sedimentary sequence from Koshak, Mangyshlak peninsula of the Caspian sea, USSR. The sediments are composed of white chalky material (mostly calcareous nannofossil assemblages) and the K/T boundary can be traced over tens of kilometers. Paleontologic and sedimentologic criteria as well as of benthic foraminifera suggest that these were deposited in a warm shallow sea of about 200 m water depth. We have analysed oxygen and carbon isotopes in bulk sediments from a new locality spanning several meters above and below the boundary.High resolution( mm level) studies show a sudden increase of-^l°/oo in 6 ^^0 followed by similar decrease just across the boundary within an interval of->10 cm whereas 513c values do not show any appreciable change. signal indicates a rapid cooling (of--'4°C) of the surface water and subsequent warming wi t h i n 1 0 0 0 yrs . Absence of any change in e^^C implies no significant change in productivity at this location.

10 ION MICROPROBE U-Th-Pb DATING AND ELECTRON MICRO PROBE EXAMINATION OF ZIRCONS FROM THE EARLY ARCHAEAN NOVOPAVLOVSK COMPLEX, UKRAINIAN SHIELD BIBIKOVA. E.Y., Vemadsky Institute of Geochemistry, Moscow, 117334, USSR WILLL^MS, LS., Research School of Earth Sciences, Australian National University, Canberra, 2601, Australia The oldest rocks known in the Ukrainian shield are orthogneisses, tonalitic in composition, which contain enclaves of ultramafic rocks belonging to the Novopavlovsk complex. They first underwent granulite facies metamorphism, then were retrogressed to amphibolite grade. All rock types contain two generations of zircons, the second being of metamorphic origin. Ion microprobe U-Th-Pb dating of zircons has shown that the age of zircon I in two pyroxenite enclaves is 3633 ± 16 Ma and 3640 ± 11 Ma respectively. Zircon I in the host tonalite is >3609 ± 5 Ma old. Zircon II in one pyroxenite is 2901± 34 Ma, and overgrowths on zircon I are -2600 Ma old. Electron microprobe examination of the zircons has shown a relatively homogeneous distribution of U, Pb, Y, P, and Hf, with Zr/Hf = 30 in zircon I from the tonalites, which is consistent with the magmatic origin of these grains. Zircon I from the pyroxenites has very high levels, and a heterogeneous distribution, of trace elements. The inner part of most grains represents the admixture of rockforming minerals with high concentrations of Ni, Co, Cr and even Ir and Os. Hf in the zircon is up to 3%, mean Zr/Hf = 20. These features indicate a metasomatic origin for the zircons in the pyroxenites. The metamorphic generation of zircon in both rock t ^ e s is very low in trace elements, Zr/Hf = 40, reflecting the higher mobility of Zr in granulitic (CO:) fluid. Thus the age of 3620 Ma corresponds to tonalite emplacement and metasomatic reworking of pre-existing ultramafic rocks.

U-PB ISOTOPE DILUTION ANALYSIS OF ZIRCON FRAGMENTS FOR PETROGENETIC STUDIES: I. AIM AND PREPARATION TECHNIQUES BICKEL.Ruth A . and STEIGER,Rudolf H., Isotope geochemistry at ETH,CH-8092 Zurich, Switzerland The polyorogenic nature of granitoids is often reflected in their zircon population.Even single grains may record a major part of the host rock evolution. This is usually the case when single zircon crystals contain one or more generations of encapsulated cores. Ideally such sample should be dated by use of an ion microprobe which allows excellent geometrical resolution. Higher analytical precision of the isotope dilution method has induced us to look for an alternative method by refining the techniques of mechanical fragmentation. Whereas Scharer & Allegre (1982) used a SOO^g zircon for their original work, our aim was to obtain well localized fragments from grains of 5-10 Mg weight as normally found in the coarse fraction of zircon concentrates. We select grains containing visible cores. Prior to fragmentation exterior and interior morphological features reflecting the growth phases as well as inclusions, coloration and cracks of the single grains are carefully recorded for all crystal faces. Initially the grains are cracked between glass slide and cover glass under the combistereo microscope using a miniature crushing stage that allows observation at high magnification. To prevent uncontrollable bursting the grain is placed in Canada balsam and cooled to slow fragmentation and thus enable video recording of the event. The cover glass is removed from the crushed zircon grain after freezing the balsam by liquid air. This allows to preserve the constellation of fragments as last recorded by video. By playing back the videotape in reverse direction, the original position of each fragment can be reconstructed. The fragments are then removed, washed in xylene and stored in a drop of glycerine. If necessary, further cleaving is done in balsam by use of eye surgeon's tools on micromanipulators.

ICOG 7

ABSTRACTS

ADVECTIVE-DIFFUSIVE TRANSPORT OF OXYGEN, CARBON A N D STRONTIUM ISOTOPIC FRONTS: AN EXAMPLE FROM N A X O S , GREECE B I C K L E . M i c h a e l , J . , B A K E R , J u d y , and CHAPMAN, Hazel, J , Dept. Earth Sciences, U n i v e r s i t y of C a m b r i d g e , CB2 3 E Q , U . K . M o d i f i c a t i o n of step-function isotopic p r o f i l e s across lithological contacts may p r o v i d e information on t i m e - i n t e g r a t e d fluid fluxes as w e l l as on t h e durati^gn of fluid flow events and p o r o s i t i e s . 6 O profiles across m a r b l e - s c h i s t c o n t a c t s in the M i o c e n e m e t a m o r p h i c rocks on N a x o s , G r e e c e , show p r o g r e s s i v e m o d i f i c a t i o n by fluid-flow during p r o g r a d e m e t a m o r p h i s m . T h e asymmetry of t|je alteration with the more extensive <S O reductions in t h e m e t r e - w i d e lower boundary layers of t h e m a r b l e u n i t s is c o n s i s t e n t with u p w a r d s flow t o w a r d s lower g r a d e . The b o u n d a r y layer p r o f i l e s m a y be modelled by combined advective-diffusive t r a n s p o r t of an initial step-function with rapid g r a i n - s c a l e fluid-solid equilibrium. N u m e r i c a l fits g i v e tin^e-integrated fluidfluxes of 0.21 +/- 0.06 m V m and 0.99 / - 0.22 m/m . These fluxes and the diffusive b r o a d e n i n g of t h e b o u n d a r y layer profiles imply p o r o s i t i e s of a b o u t 10 and total durations of the fluid flow event of c a . 1 Ma for an assumed permeability-porosity relationship. ^^ Comparison of 6 O and 6 C t r a n s p o r t in the b o u n d a r y layers implies t h e fluids w e r e waterrich with XCO2 less than 0 . 3 . In c o n t r a s t to the metre-wide oxygen isotopic boundary l a y e r s , strontium isotopic alteration is restricted to c a . 1 cm o r less in width and strontium c o n c e n t r a t i o n s in the low salinity fluids m u s t have b e e n less than c a . 3 p p m .

AN UNUSUAL Rb-Sr ISOCHRON: THE LODGE BAY GRANITE, SASKATCHEWAN, CANADA BIKERMAN. Michael, Department of Geology and Planetary Science, University of Pittsburgh, Pittsburgh, PA 15260, USA The Lodge Bay granite of the Beaverlodge area, northwest Saskatchewan, is a uniform massive leuco-granite intruding Precambrian meta-sedimentary rocks. These country rocks range from granulite facies in the east to amphibolite or lower facies in the west, around the Lodge Bay granite. They cover 600 km^ and include quartzites, pelites, conglomerates, carbonates, and pillow basalts. Rb-Sr and isotopic Sr analyses on 13 samples from the granite spread out well over an isochron with a date of 2839 ± 67 m.y., (MSWD of 0.73, and an initial ^^St/^^St ratio of 0.69276 ± 0.00239). Eliminating one point which fell off the line improved the MSWD to 0.15 and produced a date of 2896 ± 73 and an initial ratio of 0.69134 ±0.00253. A relatively imprecise U-Pb zircon date (Van Schmus et al. 1986) of 3072 ± 41 m.y. suggests that the emplacement age of the Lodge Bay cranite is close to 3 b.y. and supports the very low initial 87Sr/%r ratio. This date and the resultant initial Sr ratio leads to a perplexing paradox: either meteorite ^^Sr/^^Sr values (B ABl, 0.69898; Allende, 0.69877) which have been used as primordial values for primitive mantle may actually represent somewhat more evolved matter, or possibly some unrecognized geological event displaced the isochron so that the intercept was lowered but the slope was left unchanged.

ABSTRACTS

ICOG 7

ULTRASENSITIVE RADIOISOTOPE DETECTION BY ACCELERATOR MASS SPECTROMETRY J.R.Bird, ANSTO, Private Mailbag, Menai, NSW 2234, Australia Accelerator Mass Spectrometry (AMS) is now well established for the determination of long-lived radioisotopes such as ^ORe (1.6 My), ^'C (5.7 ky), ^A1 (705 ky), ^^Cl (301 ky), ^^Ca (100 ky) and (i6 My). There is also growing use for the determination of shoner half-life isotopes such as ''Be (53 d), ^^Si (140 y), etc. The accelerator used as the key component of an AMS system may vary in size from a few MV tandem to a large linear accelerator producing beam energies of 1 GeV or more. Each type offers some advantages in either cost or performance (or both). These parameters are also being improved by recent developments in auxiliary equipment such as ion sources, etc. Applications of AMS results for dating on various time scales of materials such as meteorites, exposed rocks, sediments, ice cores, groundwaters, etc. depend on die retention of the appropriate radioisotope within a closed system. The results are also dependent on natural or anthropogenic production rates and AMS can be used to study these topics. It has been particularly important for the study of variations in and ^'^C production and distribution. For open systems, AMS results provide information on transport and mixing processes - especially when used in conjuncdon with other parameters. AMS can also be used for stable isotope determination, free of molecular interference, with geoscience applications so far being concentrated on the study of platinum group elements.

IN THE WEATHERING ENVIRONMENT M.I. BIRD. F.J. LONGSTAFFE*, A.R. CHIVAS AND W.S. FYFE* Research School of Eanh Sciences, Australian National University, A.C.T. 2601, Australia. *Geology Department, University of Western Ontario, London, Ont N6A 5B7, Canada. The composition of authigenic minerals in the weathering environment is controlled by temperature of formation and the of the water from which the mineral formed. These factors in turn can be related to gross climatic parameters such as temperature and precipitation-evaporation regime. However, this palaeoclimate relationship has, as yet, found only limited quantitative application in the weathering environment. This is due primarily to (i) the complex fine-gained mineralogies commonly encountered, (ii) uncertainties in mineral-water isotope fractionation factors at low temperature, (iii) poor age constraints for weathering (iv) the complex nature of processes which can effect the of meteoric waters prior to mineral formation. Currently, mineral-water fractionation factors (of varying certainty) are available at surficial temperatures for quartz, carbonates, phosphates, a variety of clay minerals (e.g kaolinite, smectite), the Fe/Al oxides and oxy-hydroxides, and Ae Ti oxides. A variety of physical and chemical separation techniques can be values for a range applied to regolith samples to obtain precise of minerals in a single sample. Such techniques have been applied to the simple multi-component weathering system represented by the Holocene karst bauxites of Haiti. values for a total of ten different minerals indicate a close approach to isotopic equilibrium with modem meteoric waters in the region. In contrast, the thick and ancient regolith of the eastern Amazon in Brazil records a complex history of formation. Some generations of minerals are in isotopic equilibrium with modem waters, however others, notably hematite from the weathered BIF deposits of the Caraj^s region and gibbsite from bauxite in the Paragominas region formed from waters strongly depleted in ^^O. This suggests formation of these deposits in a monsoonal climatic regime. The continued refinement of preparation techniques for regolith samples should enable a wide range of palaeoclimatic and genetic problems to be addressed in a quantitative manner.

11 PROBLEMS OF DATING MAHC DYKES: AN EXAMPLE FROM THE VESTFOLD HILLS REGION OF EAST ANTARCTICA L.P. BLACK. J.W. SHERATON, Bureau of Mineral Resources, Canberra, A.C.T. 2601, Australia, P.D. KINNY, R.S.E.S., Australian National University, Canberra, A.C.T. 2601, Australia, C.P. DELOR and L.B. HARRIS, Geology Department, University of Western Australia, Nedlands, W.A. 6009, Australia. Mafic dyke swarms are commonly used for the establishment of relative chronology in complexly deformed terranes. The major problem in defining an absolute chronology is the derivation of primary crystallisation ages for such dykes. At least five discrete generations of dykes are recognised, on the basis of petrography, geochemistry, field relations and isotopic characteristics, within the late Archaean Vestfold Hills region of East Antarctica. The most abundant suites are of tholeiitic composition. One of these, which forms the subject of this study, previously yielded a Rb-Sr wholerock isochron age of 1374±125 Ma (initial ratio = 0.7028±5), which was thought to date emplacement, even though the isotopic analyses are markedly scattered (MSWD = 323). U-Pb zircon ion-microprobe analyses have been made on a typical tholeiitic dolerite and a wholly enclosed 1-2 cm thick quartz diorite vein which strikes perpendicular to the dyke - country rock contacts. All available evidence, including its sharp but not obviously crosscutting contacts, is consistent with this vein being a late-stage magmatic segregation from the tholeiitic magma. Most of the (generally euhedral) zircons in both rock types are of xenocrystic origin, having an age (2483±8 Ma) which is indistinguishable from that of the surrounding felsic gneisses. The 25% of remaining ^ains from the tholeiite crystallised at 1034±24 Ma; 10% of the zircon in the felsic vein crystallised 1248±4 Ma ago. The latter zircons would have been expected to yield a minimum temporal estimate or, more probably, the actual crystallisation age of the mafic dyke. Both ages are distinctiy younger than the Rb-Sr estimate, presumably due to crustal contamination of the mafic magma (note the presence of xenocrystic zircon). Although the combinai data do not unequivocally indicate which of the two younger ages temporally defines dyke crystallisation, the balance of evidence supports the 1248 Ma value. Whether or not this is the correct interpretation, this study shows that the dating of mafic dykes must be done with extreme care.

CRUSTAL GROWTH IN WEST AFRICA AT 2. L GA. BOHER. Muriel, ABOUCHAMI, Wafa, ALBAREDE, Francis, BARBEY, Pierre, MICHARD, Annie, C.R.P.G. and University de Nancy I, BP 20, 54501 Vandoeuvre cedex, France, and MILESI,

Jean-Pierre, B.R.G.M., Oleans, France Early Proterozoic terranes (Birrimian) extend over large areas in West Africa. Four types of Birrimian formations have been distinguished: (1) Highly deformed and metamorphosed formations (amphibolite facies), (2) Volcanosedimentary formations with tholeiitic (basalts) then calcalkaline (andesites to rhyodacites) series and also polymetallic nodules and black shales, (3) Important volumes of clastic turbiditic sediments (schistes and greywackes). (4) Successive generations of calcalkaline to alkaline granites often related to transcurrent tectonic (Ledru et al., 1989; Dupuis et al., 1989). Thermobarometric estimations in granites and in surrounding rocks give similar T and P respectively of 500-550°C and 5-6 kbar. Rb-S^r, Sm-Nd and U-Pb ages of the highly metamorphic terranes (2.20-2.17 Ga) are slighly older than Sm-Nd ages of basalts, zircons U-Pb ages of clastic sediments (2.15 to 2.08 Ga) and granites (ca 2.10 Ga). The duration of the entire orogenic process is less than 100 Ma. Sm-Nd isotopic results show same average eNd(2.10 Ga) and Tdm values respectively of -I-2 and 2.25 Ga independant of the t>T)e of formations. Only granitic or detrital Birrimian formations from the Archean area have eNd(2.10 Ga) and TDM values respectively of -3 and 2.60 Ga intermediate between Birrimian (2, 2.25 Ga) and Archean (-10, 3.00 Ga) values. The short residence age of the crustal protolith (ca 100 Ma) suggest, that important volumes of juvenile crust were created in West Africa at ca 2.1 Ga. Oceanic characteristics of volcanosedimentary formations, lack of Archean participation in sediments and granites, thermobarometric and structural studies suggest that crustal growth occured in an oceanic environment in a 15-20 km minimum thick protolith and is followed by accretion of this new crust against the Archean craton with local thrusting (Feybesse et al., 1989). The Birrimian terranes bridge a major gap in the crustal evolution and mantle activity for a period, 2.10 Ga, assumed to be almost quiescent in other continents.

PRELIMINARY EVIDENCE FOR SEAWATER INTERACTION WITH SILICIC PERALKALINE RHYOLITES IN THE EASTERN PACinC OCEAN: SOCORRO ISLAND, MEXICO BOHRSON. Wendy A, and REID, Mary R, Department of Earth and Space Sciences, UCLA, Los Angeles, CA, 90024, USA. Socorro Island, Mexico is located on a segment of the East Pacific Rise which failed -3.15 Ma.The island is unique because it is the only predominantly silicic peralkaline volcano in the Pacific Ocean. High ^"^Sr^^Sr reported here for peralkaline rhyolites from Socorro provide a unique opportunity to describe the mechanisms capable of producing elevated ^^Sr/^^Sr because magmatism is geographically removed from continental crustal influence. Alkalic basalt, trachybasalt and trachyte samples from Socorro have €Nd values of +5.27 to +6.44 and ^'^Sr/^^Sr between 0.70319 and 0.70365; these ranges are similar to those in basaltic rocks from ocean islands such as the Azores and Hawaii. Although silicic peralkaline rocks have similar £Nd values, S'^Sr/^^Sr are higher, and whole rock-feldspar pairs show isotopic heterogeneity. ^^Sr/^^Sr values for acid leached samples are as follows: whole rock feldspar comendite 0.70729 0.70491 obsidian 0.70458 0.70322 Feldspar-whole rock disequilibria implies that peralkaline magmas may be affected by open-system behavior, and traceelement modeling supports this interpretation. Elevated ^'^Sr/^^Sr for peralkaline rhyolites relative to less evolved rocks suggest that the rhyolites have interacted with a seawater derived component, and their low Sr concentrations (<40 ppm) facilitate recognition of this interaction. Preliminary modeling demonstrates that seawaterrock ratios of only - 1 are required to account for the elevated comendite feldspar ^'^Sr/^^Sr, and ratios of only - 5 are required to explain the comendite whole rock. We suggest that seawater is an important contributor during open-system behavior, and our present work is focusing on defining the extent, mechanisms and timing of seawater interaction with peralkaline rhyolites fi"om Socorro.

ISOTOPIC IMAGING OF DEEP CONTINENTAL LITHOSPHERE: SYSTEMATICS AND APPLICATIONS DOWNUNDER BORG. Scott G., and DEPAOLO, Donald J., Berkeley Center for Isotope Geochemistry, University of California, Berkeley, California, 94720, USA Tectonic studies in Antarctica are greatly facilitated by isotopic mapping of crystalline rocks due to limitations placed on traditional field methods by difficult access. We are assembling an isotopic-" geochemical database that delineates major crustal boundaries along a 2000 km stretch of the Transantarctic Mountains from Victoria Land to the Horlick Mountains. This region records the late Proterozoic to middle Paleozoic development of a large portion of the Panthalassa margin of Gondwana and is unaffected by later compressive tectonism. Isotopic studies of Paleozoic granitic rocks and xenoliths from Late Tertiary volcanic rocks are used to identify and delineate lower crustal provinces and to infer the tectonic environment of pluton emplacement. Isotopic compositions of metasedimentary rocks indicate provenance and constrain tectonodepositional models. The early Paleozoic batholiths contain a predominance of cnistally derived granitic rocks, as evidenced by low initial (at 500 Ma) ENd values (-2 to -12), high s^Sr/s^Sr (0.706 to 0.740), and high (>+9). Although there is evidence for Archean basement, all of the granites were melted from Proterozoic basement (2.0 -1.4 Ga); there is a pronounced paucity of mantle-derived material. Inferred basement ages are 2.7, 1.9±0.1, 1.710.2, and l.ltO.l, the last being a terrane in which there is evidence of substantial mantlederived magma (eNd = +0.5 to +2, < +8). The former two ages represent autochthonous terranes, the latter two are allochthonous and lie outboard of the former. Nd isotopes of metasedimentary units in the allochthonous terranes indicate provenance in a different continent. Middle Paleozoic granites in northem Victoria Land define another Proterozoic (2.0 Ga) province allochthonous to pre-Permian Gondwana. In addition to inferences about crustal structure at the time of pluton emplacement, the isotopic boundaries provide a baseline for evaluating Mesozoic and Tertiary plate movement

ICOG 7

ABSTRACTS

TIMING AND SEQUENCE OF PYRITIZATIGN IN MODERN (FOAM) SEDIMENTS ^ S.H. BOTTOELL^. D. CANFIELD^ and R. RAISWELL^ •^Department of Earth Sciences, University of Leeds, Leeds LS2 9JT, UK; ^ NASA Ames Research Center, Moffett Field, California 94035, USA. Pyritization of modem sediment proceeds by reaction of bacterially-produced H j S with iron derived from a variety of mineral sources. The different mineral sources show a range of reactivity toward H2S, and therefore pyrite formed by reaction with different mineral species should record any progression of H2S S-isotope composition during diagenesis. The most reactive iron minerals will react with, and record, the eariiest H9S isotopic compositions and this composition will evolve to heavier values as pore water sulphate is depleted during diagenesis. We have isolated pyrite associated with different sediment fractions at a number of depth intervals in core from the FOAM site. Analysis has been carried out for pyrite Sisotopic compositions and also residual pore water H j S composition at each depth. Near surface pyrite associated with magnetite has a similar isotope composition to pyrite associated with other iron oxides, but witn increasing depth the less reactive magnetite continues to react with isotopically heavier H2S. The pyrite associated with magnetite thus attains progressively heavier compositions while pyrite associated with the other iron oxide fractions preserve their early, light S compositions. The magnetite-associated pyrite is generally only 1 - 3 9 ^ heavier than the porewater H2S at any depth, but up to 20%Q heavier than the early formed pyrite. Thus pyrite associated with magnetite continues to record the latest stages of bacterial H2S production in these sediments.

Pb ISOTOPIC RESULTS FOR THE RONDA ULTRAMAFIC COMPLEX B. BOURDON, A. Z I N D L E R AND L. R E I S B E R G , Lamont

Doherty Geological Observatory of Columbia University, Palisades, N.Y. 10964 U.S.A., and E. J A G O U T Z , Max Planck Institut fur Chemie, Postfach 3060, D6500 Mainz, F.R.G. The Sr and Nd isotopic ratios of the Ronda Ultramafic Complex have been shown to span nearly the entire range of values obtained from oceanic basalts. The third phase of our continuing investigation of Ronda entails characterization of Pb isotopic systematics. Pb concentrations in hand-picked, acid-washed cpx separated from plagioclase and spinel fades peridotites range from 20 to 63 ppb and have mean and median values of about 40 ppb. Pb isotopic results for Ronda cover most of the known field for mantle compositions, extending from near the depleted end of the MORB field to the most radiogenic of mantle leads from Mangaia, near the HIMU component of Zindler and Hart (1986). A similar range has been observed for Beni Bousera by Hamelin and Allegre (1988). Samples scatter about the Northem Hemisphere Reference Line (NHRL) of Hart (1984) on a plot of 206pb/204p5 versus 207p5/204p5^ The data define a relatively similar array on the 208pb/204pb versus 206p5/204pb diagram, with the more radiogenic samples extending to higher values of 208pb/204p5 than have previously been observed in mande-derived basalts. The Ronda peridotites with HIMU characteristics occur in the plagioclase facies where depletion in basaltic constituents is most severe (Suen and Frey, 1987). In contrast to the major element depletion, this re^on of the massif is marked by an "enriched" Sr and Nd isotopic signature, with relatively high ^Srl^^Sr ratios and very low i^^Nd/w^d ratios (Reisberg and Zindler, 1986/7). In Nd-Pb and Sr-Pb diagrams, the data extend from the MORB field through to HIMU. If the Ronda trend were due to chemical interaction with fluid or meh derived from crustal material at depth, we would expect to see a positive correlation of i43Nd/i44Nd with 206p5/204p5^ the opposite of that which is observed. This negative correlation constitutes the most convincing proof to date that the isotopic effects observed at Ronda at not related to crustal emplacement.

ABSTRACTS

ICOG 7

OXYGEN ISOTOPE EXCHANGE ALONG FLUID F=LOW PATHS: IMPUCATIONS FOR THE ISOTOPIC EVOLUTION OF HYDROTHERMAL FLUIDS AND ESTIMATES OF WATER/ROCK RATIOS BOWMAN, John R. and COOK, Stephen J., Dept. of Geology and Geophysics, University of Utah, Salt U k e City, UT 84112 A necessary consequence of the flow of fluid through the earth's crust is the progressive modification of the oxygen isotopic composition of both fluid and rock within the flow system. Numerical modeling of fluid flow and isotopic exchange indicates that narrow transition zones or isotope infiltration fronts characterize equilibrium exchange flow and are developed in both the fluid and rock columns along the flow paths. As kinetic effects become more important(increaslng infiltration rate, larger grain size, decreasing temperature, or decreasing rate constants) this sharp isotope infiltration front is progressively destroyed. Model calculations indicate that significant kinetic effects are induced in quartzo-feldspathic rocks at T< 400®C for infiltration rates of 0.1 m/yr or greater; in contrast, carbonate rocks can sustain much higher infiltration rates (1 .Om/yr) to even lower temperatures MOO°C) without developing significant kinetic modifications of the ^0-distance profiles. Sharp infiltration fronts are diagnostic of equilibrium exchange during fluid infiltration and should be observed in the high temperature metamorphism of carbonate rocks. Such sharp fronts are observed in the contact metamorphosed dolostones in the Alta Stock aureole, Utah (Cook and Bowman, 1988). Progressive oxygen isotopic exchange In flow systems has several consequences to volumetric W/R ratios cateulated with conventional isotope techniques. Isotopically-calculated W/R ratios will: 1 )progressively underestimate true W/R ratio as a function of increasing pathlength of fluid flow and the extent of prior isotope exchange in the rock column; 2) be a function additionally of kinetic parameters such as T, fluid infiltration rate, time and kinetic rate constants of minerals. Therefore isotopically defined W/R ratios will be much lower than actual W/R ratios for natural flow systems which have evolved along long flow lines, at temperatures below 400°C, or at high fluid Infiltration rates. In such cases, the conventional W/R ratio will reflect primarily the characteristics of the flow path, not the local conditions of fluid-rock exchange.

THE 3.96 Ga ACASTA GNEISSES: REMNANT O F EARTH'S EARLY CRUST BOWRING. S.A., HOUSH, T., ISACHSEN, C.E., PODOSEK, F.A., Department of Earth and Planetary Sciences, Washington University, St. Louis, Missouri, 65130-4899, USA WILLIAMS, I.S. and COMPSTON, W., Research School of Earth Sciences, Australian National University, Canberra, ACT, 2601, Australia The Acasta gneisses of the Slave craton, NWT, Canada, are the Earth's oldest known rocks. They consist of a complex assemblage of tonalitic to granitic gneisses and amphibolites with subordinate ultramafic and metasedimentary rocks. SHRIMP U-Pb analyses of zircons from both tonalitic and granitic gneisses indicate that the rocks formed 3.96 Ga ago. On concordia plots the data define complex arrays interpreted to reflect early Pb loss followed by new zircon growth ca. 3.6 Ga ago. Many zircons are characterized by unzoned cores indistin^shable in age from their zoned 3.96 Ga overgrowths. Nd isotopic data r ^ u i r e that crust older than 3.96 Ga was involved in the generation of these rocks. The Nd chondritic model ages for the p e i s s e s are as old as 4.1 Ga. 207pb/204p5 in feldspar in the gneisses is characteristically high, suggesting either an older highsource region or mixing with radiogenic Pb. The ultramafic rocks have low R E E abundances and Nd isotopic compositions consistent with equilibration with enriched cmstal fluids at ca. 3.6 Ga. The Acasta gneisses are the first example of rocks derived from an early Archean reservoir which was enriched relative to the depleted mantle. The global Nd and Pb data-base for early Archean rocks shows a considerable range in initial isotopic compositions. This range has often been attributed to mantle heterogeneity; however, some of it may represent the involvement of small but variable amounts of older enriched continental crust. An important question for models of crustal evolution is how much pre-3.8 Ga crust is preserved. The recent discovery of the Acasta gneisses emphasises low much is still unknown about the distribution of ages in the Archean cratons of the world.

Sm-Nd, U-Pb, AND K-Ar DATING OF THE Z E R M A T T - S A A S F E E OPfflOLITE, W E S T E R N ALPS. B O W T E L L . Sophie Ann, Dept. of Earth Sciences, University of Leeds, L S 2 9 J T , Great Britain. The Zermatt-Saas Fee ophiolite lies between granitic massifs o f European and Apulian origin. It is thought to represent the ocean that once separated Europe from Italy and the rest of the Apulian plate. The ophiolite and surrounding massifs have undergone subduction to a depth of around 60km. In the adjacent Monte Rosa massif, the high pressure event has been dated at 1 lOMa {^^AI-^^AT, phengite). The aim of this study is to constrain the history of the ophiolite by dating successive metamorphic assemblages displayed by the metabasalts. T h e subduction pressure peak ( 1 7 - 2 0 k b , 5 5 0 - 6 0 0 ' C ) is represented by fresh eclogites and blueschists exposed in a small area of Taschalp, near Zermatt. These high pressure rocks are surrounded by green schist overprinted rocks; the amount o f retrogression ranging from small chlorite rims on garnet to complete greenschist facies recrystallization. Sm-Nd mineral isochron work has proven problematic. Garnets in these rocks have unusually low SmMd ratios, leading to a small range of Sm/Nd ratios and large errors. Results, though imprecise, suggest that closure o f the Sm-Nd system occurred at less than lOOMa. Garnet is thought to become closed to Sm diffusion at 500700°C, thus ages should be close to the age of crystallization. The greenschist overprint has been dated at 37-40Ma (39Ar-40Ar, phengite) in the Monte Rosa and Gran Paradiso massifs. K-Ar dates for paragonites from high pressure metabasalts with and without greenschist overprint fall into the range 40-50Ma. This suggests that greenschist conditions in the ophiolite (8kb, 4 5 0 ' C ) developed prior to the closure of paragonite to Ar (at about 350'C, c f muscovite ?). Although this work is still in its early stages, one interpretation of the results is that the Zermatt-Saas zone has a time-temperature history distinct from that of the surrounding continental massifs. The problem of the age of the greenschist event may be resolvable by UPb isochron methods. Sphene, common in the greenschist assemblage, has a favourable U/Pb ratio and a closure temperattire of at least 500°C. It should be possible to date the sphene crystallization in this way.

COMPARISON OF SR ISOTOPIC COMPOSITIONS OF OREFORMING F L U I D S O F T W O MISSISSIPPI V A L L E Y - T Y P E (MVT) O R E DEPOSITS: VIBURNUM TREND, S E MO AND UPPER MISSISSIPPI V A L L E Y DISTRICT, SW WI, USA B R A N N O N . J . and P O D O S E K , F., Dept. of Earth and Planetary Sciences, Washington University, St. Louis, MO 63130, USA, and VIETS, J., LEACH, D., GOLDHABER, M., R O W A N , L . , U S G S , Denver, C O 8 0 2 2 5 , U S A , and M c L I M A N S , R., E.L duPont de Nemours & Co., Inc., Wilmington, D E 19898 USA The isotopic composition of Sr is a useful tracer for fluid origins and flow-paths of M V T ore deposits, since they are hosted by carbonate sediments of Paleozoic age (with corresponding marine ^^Sr/^^Sr compositions, 0.707-0.709) that are at various stratigraphic levels above the Precambrian basement or basement derived clastics (with typically high Rb/Sr and ^^Sr/^^Sr > 0.709). Measured ^'^Sr/^^Sr ratios in sphalerites from the Middle Ordovician Upper Mississippi Valley District are more radiogenic than fluid inclusions in the same samples, but solid and inclusion Sr compositions agree (^^Sr/^^Sr « 0 . 7 0 9 0 ) when age corrections are applied for decay of ^^Rb for 300 Ma, consistent with a postulated Pennsylvanian-Permian age. The initial is elevated above ^'^Sr/^^Sr of seawater of the same age ( « 0 . 7 0 8 0 ) , indicating that fluids interacted with underlying Cambrian clastics or Precambrian basement rocks. In pronounced contrast, age-corrected (300 my) Sr isotopic compositions of fluid inclusions and their host minerals in the Viburnum Trend do not usually yield the same initials. Initial Sr values of fluid inclusions fall within or slightly above the ranges found for the Upper Cambrian dolostone host ( « 0.7090). Initial Sr values for minerals v a ^ as a function of paragenetic sequence. Early pyrite, chalcopyrite and sphalerite are more radiogenic (0.7408-0.7108), values for most galenas are less radiogenic, extending to those of the host (0.7155-0.7086), and late stage chalcopyrites and quartz again are more radiogenic (0.71560.7097), indicating a much more complicated history of fluid interaction with closely underlying Cambrian or Precambrian rocks.

SOME THERMAL CONSEQUENCES OF REGIONAL EROSION AND THE IMPLICATIONS FOR INTERPRETING APATITE nSSION TRACK CRUSTAL PROFILES BROWN. Roderick and GALLAGHER, Kerry, Department of Geology, La Trobe University, Melbourne, 3083 Australia The hypothesis that regional denudation is the dominant control on cooling in the upper few kilometers of the crust is proposed. This is investigated using a simple one dimensional numerical model for the thermal evolution of the lithosphere. The model is used to assess the competing thermal effects of increased heat flow, caused by simple shear extension of the lithosphere, and regional denudation (erosion) of the crust. The model results are used to predict the variation with depth of apatite fission track ages and confined track length distributions within the crust. The theoretical fission track crustal profiles provide a means for interpreting apatitefissiontrack analysis resultsfi-omthe South Adantic margin of southern Africa. Apatite fission track ages and confined track length distributions obtained from basement samples from the western continental margin of southern Africa record an episode of accelerated cooling which is broadly synchronous with continental rifting between South America and Africa. The apatite ages indicate that cooling began at -140 Ma in the south near Cape Town and occurred progressively later to the north, where thefinalcooling of the margin was initiated at -70 Ma in the Etendeka region of northem Namibia. This cooling of the upper crust is interpreted as being the consequence of regional erosional denudation, coeval with the formation of the new continental margin. This period of accelerated erosion correlates with the period of maximum sedimentation in the offshore basin, providing independent evidence for increased rates of erosion during the Cretaceous. It is proposed that suitable apatitefissiontrack profiles can be used to quantify the magnitude and distribution of erosional episodes. The eroded section can then be reconstructed or "backstacked" and the isostatic unloading and uplift effects of erosion calculated using a technique analogous to backstripping basin stratigraphy. After removal of the isostatic loading and unloading effects of sedimentation and erosion, respectively, comparable estimates of the tectonically induced subsidence and uplift can be made.

CONVERGING Sm-Nd MINERAL AGES IN ECLOGUES AND GARNET PERIDOTITES FROM THE CALEDONIDES OF NORWAY AND THE VARISCIDES OF POLAND AND CZECHOSLOVAIOA H.K. BRUECKNER*. N. BAKUN-CZUBAROW, J.L. RUBENSTONE, and A. ZiNDLER, Lamont-Doherty Geological Observatory of

Columbia University, Palisades, NY 10964 U.S.A. (*also at Queens College of CUNY, Flushing, NY 11367 U.S.A.) Gamet-bearing mafic and ultramafic lithologies record the history of high-pressure metamorphism within the Caledonian and Variscan orogens of nonhem and central Europe. Eclogites that occur directiy within gneisses (country rock eclogites) give Sm-Nd gametwhole-rock-cpx a^es that closely approximate the peak of eciogitefacies metamoiphism (400-447 Ma tor the Caledonides of southern Norway; 327-353 Ma for the easternmost Variscides) during continent-continent collision. Spatially associated jamet peridotites, garnet p>Toxenites, etc. that occur within ultramafic rocks generally give significantly older ages (998-1703 Ma for the Caledonides; 374 to 405 Ma for the Variscides). The differences are attributable to the fact that country rock eclogites formed from low P protoliths (igneous and low-grade metamorphic rocks) that acquired their eclogite facies mineralogies during prograde metamorpnism in the crust, whereas the gametiferous ultramafic rocks originated as high P-T mande assemblages that re-equilibrated at progressively lower P and T as they were inserted into the crust. The retrograde nistory of the garnet peridotites resulted in the recrystallization of a secondary gamet-bearing assemblage of younger a^e (Jamtveit et al., 1990), and zonation in Sm/Nd and i^s^d/i^^Nd in first generation garnets. Zoned first-generation arnets in garnet Iherzolites from the Almklovdalen peridotite (,(Norway) and the Gory Sowie peridotite (Poland) show changes in Sm/Nd, i43Nd/i44^d and decreasing apparent ages (1460 to 1230 and 405 to 385 Ma, respectively) from core to rim. The core and rim ages are considered, respectively, minimum ages of the initiation, and maximum ages of the cessation, of first-generation garnet growth. The garnet ages in Norwegian peridotites are significantiy pre-Caledonian and imply a long period of residence in a relatively cold (<1000°C) mantie region for much of the Proterozoic prior to insertion into the Caledonian orogen. In contrast, garnets in the Polish peridotite define ages that are barely pre-Variscan, with a short (20 Ma) age difference between core and rim, suggesting a source mantle region that was above mineral homogenization temperatures for REE (>1000°C) until shortly before inculcation within the Variscan orogen.

ICOG 7

ABSTRACTS

ISOTOPIC E V O L U T I O N OF THE RHINE RIVER SYSTEM P' BUHL, R.D. N E U S E R , D.K. R I C H T E R , D. RIEDEL, B. R O B E R T S , H. S T R A U S S & J. V E I Z E R Institut f u r G e o l o g i e , R u h r - U n i v e r s i t a t , D-4630 B c c h u m , FRG The Rhine basin has been studied at 27 locat i o n s b e t w e e n the inflow of the upper Rhine Into Lake C o n s t a n c e and E m m e r i c h at the Dutch b o r d e r . S a m p l e d w e r e river w a t e r s and t h r e e layered a r a g o n i t i c s h e l l s of the m i g r a t o r y D r e i s s e n a p o l y m o r p h a . C h a r a c t e r i s t i c chemical f e a t u r e s are the progressively increasing c o n d u c t i v i t y , o r t h o p h o s p h a t e , K and Na cont e n t s . Isotopic t r e n d s in the shells are a d o w n s t r e a m i n c r e a s e in r a d i o g e n i c Sr (0.70850 . 7 0 9 0 ) , h e a v i e r o x y g e n (-11 to -9%o PDB) and lighter c a r b o n (-7 to -10"^ PDB). The 5 of the Rhine w a t e r c h a n g e s from -12 to -10%o. The Alpine melt water is c l e a r l y the feed f o r the upper Rhine with lowland t r i b u t a r i e s - r e f l e c t i n g local precipitation around - slowly contributing to the overall w a t e r b u d g e t . Based on balance c a l c u l a t i o n s , the t r i b u t a r i e s a c c o u n t for 5060% of the lower Rhine d i s c h a r g e . The d o w n r i ver oxygen isotope t r e n d for w a t e r is m i r r o red by the 5 l^O of the s h e l l s . The S ^^c com p o s i t i o n of the Rhine TDC is strongly influenced by soil d e r i v e d C O 2 d e l i v e r e d by the t r i b u t a r i e s . The Rhine system serves as a net source of C O 2 for the a t m o s p h e r e . A 8 7 s r / 8 6 s r of 0 . 7 0 8 5 in the upper p a r t s u g g e s t s t h a t the m a i n s o u r c e of Sr are the M i o c e n e ( b i o ) c h e m i c a l s e d i m e n t s of the Alpine r e g i o n . An i n t e n s i v e c o n t r i b u t i o n from w e a t h e r i n g of P a l e o z o i c s i l i c a t e rocks d e l i v e r e d by t r i b u t a r i e s e n h a n c e s the r a d i o g e n i c ^^Sr component downstream (0.7090).

BETTER UNDERSTANDING OF 39Ar. 40Ar APPARENT AGES BY MONTE- CARLO SIMULATION. A.BURGHELE , Max-Planck-Institut fiir Chemie, D-6500 Mainz, F.R.G. The 39Ar- "Âr degassing patterns, not showing a constant 39Ar / Âr ratio or a monotonic increase to a so-called "plateau", reflecting in the initial stages of gas release the effect of recent argon loss are not well understood. This is especially the case of saddle-shaped release patterns or decreasing ages with hiêr temperature steps. The attempts to explain these effects by diffusion measure ments are very tedious, because assumptions have to be made with respect to the mineral size,shape and composition. In the crystal the argon atoms are confined to definite sites in the lattice. A definite energy called "activation energy" is required to dislodge an argon atom from a site. This energ>' is derived from the thermal motion of the lattice. Considering the gas release as a statistical randomizing process,the motion of the argon atoms is a random walk in three dimensions. By means of Monte- Carlo simulation it is possible to explain the above mentioned gas release patterns,avoiding definite assumptions about dimensions and difiusion charaaeristics of the mineral components in the sample. The movement of an argon atom along a line, in a plane or in a three dimension space can be described by the equation: {r) = I Jn where ris the distance from the initial location, n the number of displacements and / the unit length of the displacement. If we assume that for a given temperature the length / will be constant, the number of displacements will be proportional to the time-interval the heating temperature is applied to the sample. The equation may be written : (r) = iJTiJ where T is the time and ni the number of displacements per unit time for a given temperature.The theoretical approach is verified by the measurements which were performed by repeated degassing at the same temperature.

ABSTRACTS

ICOG 7

THE THERMAL EVOLUTION OF CONTINENTAL CRUST RECORDED BY CRYSTAL GROWTH K.W. BURTON and R.K. O'NIONS, Dept. of Earth Sciences, University of Cambridge, Downing Street, Cambridge. CB2 3EQ, U.K. Understanding the thermal evolution of the continental lithosphere requires a knowledge of the pressure (P) temperature (T) and time (t) paths that rocks experience during metamorphism and crustal deformation. High precision P-T-t information is recorded by differential mineral reaction within an interlayered metasediment from Sulitjelma, North Norway, that underwent metamorphism during the Caledonian orogeny. Garnets in graphitic layers grew isotherm ally at 460+18°C and 5.3+0.5 kbar, yielding concordant Sm-Nd, U-Pb and Rb-Sr ages of 434.1+1.2 Ma, 434.2+1.1 Ma and 435.2±12.0 Ma, respectively. In contrast, garnet growth in adjacent graphite-free layers ceased at 575+32°C and 9.3+1 kbar, and garnet rims preserve concordant Sm-Nd, U-Pb and Rb-Sr ages of 424.6+1.3 Ma, 423.5+1.9 Ma, and 422.4+4.6 Ma. Utilizing the Sm-Nd data these results indicate heating and burial rates of 12.1+Vj^°C Ma-1 and 420±^f^ bars Ma-l, respectively, that relate directly to mineral reactions accompanying prograde metamorphism and crustal deformation. Rb and Sr exchange between coexisting muscovite and biotite continued to 420+4.0 Ma and 386+1.0 Ma, respectively, which constrains cooling during uplift to 6.6±^-j°CMa-l.

IMPLICATIONS OF MANTLE PLUME STRUCTURE FOR THE GEOCHEMISTRY OF THE OIB-SOURCE CAMPBELL. I.H., GRIFHTHS, R.W., Research School of Earth Sciences, The Australian National University, Canberra ACT 2601,AustraHa Experimental studies in viscous fluids show that a newly initiated plume consists of a large bulbous head followed by a narrower feeder conduit. If tiie ascent of the plume is driven by thermal buoyancy, the head entrains the overlying mantle as it rises so that it becomes a mixture of hot mantie from the plume's boundary layer source and cooler overlying mantle. The conduit, connecting the head to the plume's mantle source, does not entrain as it rises provided it continues to rise vertically. Its potential temperature remains high, comparable to that in the boundary layer from which the plume originates and it melts to produce picrites. Continuing plumes, of the type that produce ocean islands chains, also entrain the surrounding mantle as they rise if the conduit becomes titied. The amount of entrainment increases with the degree of tilt and with decreasing mass flux through the plume. Strong plumes, such as Hawaii and Pitcaim are therefore predicted to produce picrites. Since picrites, from both flood basalt provinces and ocean island chains, come from plumes, or the parts, of plumes that are least contaminated by mixing, their geochemistry gives the best sample of the isotopic and incompatible trace element ratios of the mantie in the plume source. The higher the MgO content of a picrite, the lower its contamination and the better the sample.

CONTINENTAL FLOOD BASALTS: SAMPLES OF THE LITHOSPHERE OR DEEP MANTLE? CARLSON. Richard W., Carnegie Institution of Washington, DTM, 5241 Broad Branch Road, Washington. D.C, 20015, U.S.A. Continental flood basalt provinces are accumulations of 10^ to greater than W km^ of basalt erupted over time spans of less than a few million years. The larger of these events reflect melt production rates that are at least an order of magnitude greater than observed at ocean ridges or major hot-spots. Over the last 15 years, geochemical studies have shown that flood basalts range in composition from picrites through basaltic andesites, and from depleted through strongly enriched LILelement contents. Generalizing the large compositional range observed is a risky undertaking because each flood basalt province has its own distinct characteristics. At least two mantle components appear to be involved in most occurrences, however; 1) relatively depleted mantie most like T- or E-MORB sources, and 2) ancient LDL-enriched mantie with low i«Nd/i44Nd, 206pb/204pb and chemical characteristics most often associated with convergent margin volcanism. Both of these components can be found in the lithosphere or shallow asthenosphere. Consequentiy, geochemical data do not require a plume component in flood basalts. If the source of flood basalts primarily is in the shallow mantie, a mechanism must be found to produce large-scale melting over short time periods. Extensive melting within an uprising plume is more easily envisioned, but basalts with chemical and isotopic signatures similar to those of associated oceanic hot-spots are not always the most abundant in a given flood basalt province. Further study of the temporal and geographic variation in the chemical and isotopic characteristics of flood basalts may provide additional insights into die ultimate sources of these magmas and to the manner with which plumes interact with continental lithosphere.

MANTLE AND CRUSTAL LEAD ISOTOPE SIGNATURES OF Au MINERALIZATION IN THE WESTERN TECTONIC DOMAINS OF THE LACHLAN FOLD BELT, NSW. CARR. G.R. and DEAN, J.A. CSIRO Division of Exploration Geoscience, North Ryde, NSW, 2113, Australia. Gold and base metal mineralization of porphyry, epithermal and volcanogenic associations in the western tectonic domains of the Lachlan Fold Belt of NSW have complex Pb isotope initial compositions which reflect the origin of Pb from the crust and from two mantle sources. Initial Pb isotope ratios of deposits containing crustal Pb vary slightly within and between domains, plot on or close to orogene growth curves, and have model ages consistent with known geological ages. These deposits are associated with Ordovician to Devonian felsic volcanic rocks or deep water turbidites (e.g. Mineral Hill, Elura). Initial Pb isotope ratios of deposits containing mantle derived Pb show significant depletion in 207pt, relative to crustal Pb and have highly variable ^^^Ph/^^yh ratios. These deposits are associated with Ordovician to Devonian andesites which may have a shoshonitic affiliation. They occur as porphyry deposits associated with monzonitic high-level intrusions and as probable epithermal deposits. The wide variation in isotopic composition is interpreted to have resulted from mixing of two mantle sources; one with MORB like U, Th and Pb contents and one relatively depleted in Pb. The major ore deposits (e.g. Goonumbla, Gidginbung) have a distinctive isotopic signature dominated by the depleted Pb source.

I30T0PIC EVIDENCE FOR CONVERGENT DEVONIAN AND CARBONIFEROUS MAGMATIC ARCS IN THE EASTERN LACHLAN FOLD BELT, NSW CARR, Paul F., and JONES, Brian G., Department of Geology, University of Wollongong, Wollongong, NSW, 2500, AustraUa. The Marulan Batholith and comagmatic Bindook Volcanic Complex crop out as a near meridional belt in the easternmost exposures of the Lachlan Fold Belt. The batholith is subvolcanic, only partly unroofed, and consists of 14 mappable plutons which comprise only about 16% of the total outcrop area of 1400 sq. km. K-Ar ages for biotite separates, together with Rb-Sr biotite-whole rock data, indicate emplacement between 406 and 388 Ma and an initial Sr isotopic ratio of approximately 0.7060. These isotopic data accord with biostratigraphic evidence which indicates a minimum age of Late Devonian for both the volcanic rocks and plutons, and a maximum age of Late Silurian to earliest Devonian for the volcanic episode. The Marulan Batholith, therefore, has the same general Late Silurian to Early Devonian age as most batholiths in the southeastern Lachlan Fold Belt and constitutes part of the world-wide magmatic event at about 400 Ma. The Marulan Batholith and Bindook Volcanic Complex are spatially associated with several contact-aureole plutons which have Rb-Sr biotite-whole rock ages about 325 Ma and an initial Sr isotopic ratio of approximately 0.7049. This close association of granites of different ages is unusual for the Lachlan Fold Belt and marks the intersection of a Devonian magmatic arc, to the south, with a Carboniferous magmatic arc extending along the western margin of the New England Fold Belt. Isotopic differences between the Devonian and geographically associated Carboniferous granites constrain the timing of emplacement of the source rocks for the younger granites to be contemporaneous with, or subsequent to, the emplacement of the Marulan Batholith. A NOVEL TECHNIQUE FOR DETERMINING SEDIMENTATION RATES FOR VARIABLE ^lOpB ACTIVITY VERSUS DEPTH PROFILES IN MARINE SEDIMENTS CARROLLJoLynn. Marine Science Program, LERCHE, Ian, and LUI, Jianchang, Geological Sciences Department, University of South Carolina, Columbia,SC 29208, USA 210pb geochronolo^ is a powerful method for determining sedimentation rates in nearshore marine environments. Marine sediments adsorb ^^^Pb from seawater. After burial, ^lOpb activities decrease with time at a rate commensurate with the decay coefficient X = 3 A x 10-2 atoms/yr. If 2l0pb adsorption to sediments is constant, if sedimentation rate is constant, and if sediment compaction is ignorable then 210pb versus depth profiles can easily be used to determine the sedimentation rate from the equation of radioactive decay, 210pb(z)=210pb(z=0)exp(- X z/v) where z is depth and v=sedimentation rate. Often in rapidly depositing environments, neither source nor sedimentation is constant, producing 2l0pb activities which vary nonexponentially with depth in sediment cores and even increasing with depth over restricted depth ranges. Although the causes of such variations are known, techniques for separating sedimentation from source variations have achieved only limited success to date. We have achieved success determining variations of source and sedimentation in sediment cores by modeling ^lOpb profiles as 2lOpb(z)=210pb(z=0)exp[(- X z/v)+Ian/n7i(sin(n7iz/zmax))+ Zbn/n7c( 1-cos(n7rz/zmax))] where summations are for n=l,2,3..., the terms involving an(bn) refer to sedimentation (source) and zmax is the greatest depth of measurement. A nonlinear method detemiines the best parameters for 2lOpb(z=0),v, and the coefficients an & bp which satisfy the data by least square error analysis. The non-linear procedure guarantees to keep the time to depth conversion positive at all depths. The method will be presented using 2iOp5 ^ata for Amazon Shelf and Black Sea sediments. The procedure is of general applicability for all radioactive isotopes.

ICOG 7

ABSTRACTS

LOW ^ 1 8 0 PROTEROZOIC THOLEIITE MAGMAS FROM THE SCOURIE DYKE SWARM, LEWISIAN COMPLEX, NW SCOTLAND. CARTWRIGHT, Ian. Dept. Earth Sciences, Monash Univ., Clayton, Victoria, 3168 Australia and VALLEY, John W . Dept. Geology and Geophysics, Univ. Wisconsin, Madison, WI 53706, USA. The Scourie dyke swarm is dominated by quartz tholeiites (-80% of dykes) with minor volumes of picrites, olivine gabbros and norites. The dykes were intruded at depths of >15kra in two episodes at 2.0 and 2.4Ga. Unaltered tholeiites have a primary mineralogy of Plg-J-Cpx+Mt+Qtz+Opx+Gnt+Hbl and chilled margins. The tholeiites were variably altered to Hbl +Plg assemblages during the 1.7Ga Laxfordian event. Seven samples of unretrogressed tholeiite dykes from near Scourie yield whole rock (WR) values of 2 . 0 4 + 0 . 1 4 ^ , lighter than typical mafic igneous rocks. Constituent minerals are isotopically concordant (Pig = 3.1 to 4 . 1 ^ , Cpx = 1.4 to 1 . 9 ^ , Mt = -0.3 to 0.65Jo) and preserve high-temperature (>500oC) ^ i s o fractionations. The dykes intrude isotopically heterogeneous granulite f a d e s gneisses fî8o(WR) values: tonalitic = 9 . 3 - 9 . 8 ^ , n=4; ultramafic = 5.2-5.7ô, n=4; basic = 6 . 3 - 6 . 9 ^ , n=5; metasediments ^'i®0(WR) = 8.5-9.9?^, n=3). Retrogressed tholeiitic dykes have S (WR) = 5.6-6.5ô due to the influx of fluids that were in equilibrium with the surrounding gneisses. Interaction of the dykes with surface-derived fluids is unlikely because of: a) the high temperature 6 fractionations; b) presence of primary minerals; c) the depth of intrusion; d) the heterogeneous isotopic values of the country rocks. It is proposed that the Scourie dykes were intruded as low <^1^0 magmas, possibly derived from a mantle region that contained subducted relics of hydrothermally altered oceanic lithosphere. If the ^^Ô values are typical, several thousand km^ of lowc^^Ô magma may have been produced during the Proterozoic.

THE ARGHEAN CAR30NATITE-BEARING ALKALINE INTRUSION AT KAMINAK LAKE, N.W.T., CANADA - AN lONMICROPROBEAGE. CAVELL.P.A. and BAADSGAARD H., Department of Geology, University of Alberta, Edmonton, Alberta, Canada, T6G 2E3, WIJBRANS J.R., ZWO Laboratorium voor Isotopen Geologie, 1081 HV Amsterdam, The Netherlands and COMPSTON W., Research School of Earth Sciences, Australian National University, G.P.O. Box 4, Canberra, ACT 2601, Australia. The Kaminak Lake alkaline complex intrudes both volcanic and younger gabbroic to adamellite units of the Archean Rankin-Ennadai granitoid-greenstone belt in the Churchill Province of the North American craton. The complex consists of silica-undersaturated feldspathic syenites (±nepheline) and an ijolitic suite (± biotite and melanite), both cut by late carbonatite dykes. Regional metamorphic grade of host and alkaline intrusive rocks is lower greenschist. U-Pb ion microprobe (SHRIMP) analysis of zircons from syenites and ijolites yielded an intrusion age of 2659 ± 5 Ma. Zircons show complex zoning with a wide range of U and Th contents and a Th/U range from 0.6 to 9. There is no evidence for remnant older cores nor for post-crystallisation resetting of the U-Pb system. Adjacent felsic volcanic rocks, near the top of the stratigraphic sequence, were previously dated at 2692 ± 1 Ma. The time span for emplacement of the host plutonic suite is not yet well constrained. Argon-laser hornblende ages for gabbros and tonalites range between 2600 and 2730 Ma. Kaminak Lake is one of the oldest carbonatite-bearing intrusions yet reported. Its presence and age suggest an extensional tectonic regime in a young granitoid-greenstone belt in the late Archean.

ABSTRACTS

ICOG 7

SURFACE TRACK LENGTH DISTRIBUTIONS FOR COMPLEX THERMAL HISTORIES A. CHAMBATJDFT. M. MARS, M. REBETEZ and M. GRIVET Laboratoire de Microanalyses Nucleaires U.F.R. Sciences et Techniques La Bouloie -16 route de Gray 25030 Besangon Cedex - France Fission track length analysis is of fundamental importance for the fission track method since it provides information about the thermal history of minerals. Depending on the kinds of analyzed traclK, i.e., confined tracks or surface tracks, the trend of length distributions is different, each technique showing advantages and drawbacks. Whereas confined track analysis is often used, the study of surface track length distributions must be further developed, especially for complex geological histories. With this goal in mind, we first carried out a laboratory simulation of short-time thermal events in a Durango apatite crystal followed by neutron irradiation. Two populations of tracks were registered in the crystal, one of which was partially annealed, and the track projected lengths were then studied. In these samples, the resulting length distributions emphasize the presence of two track populations. A linear regression method leads to graphic dissociation of the two track populations and gives fundamental information about each population. Based on this work, several complex thermal histories were considered and the corresponding length distributions were analyzed. Finally, an application of this study was tested on an apatite sample with a complex geological history.

GRANITES FROM THE HODGKINSON PROVINCE, NORTH QL^ENSLAND; ORIGIN AND IMPLICATIONS FOR CRUSTAL EVOLUTION: ISOTOPIC AND GEOCHEMICAL EVIDENCE CHAMPION. D.C., CHAPPELL, B. W., Geology Dept., ANU, Canberra, ACT, 2601, Australia, and McCULLOCH, M.T., RSES, ANU, Canberra, ACT, 2601, Australia. The Hodgkinson Province (HP) is an Ordovician to Devonian, flysch-dominated, supracrustal sequence that forms the northern part of the Tasman Orogen. Both I- and S-type granites of Carboniferous to Permian age are found in the province. The S-tjpe granites cover about 2500 km^ and are very felsic (Si02 = 68-77%; average = 74%). Geochemistry of the granites (higher CaO, AI2O3, Ba, REE, Y) suggest that exposed sediments of the HP were not the source. Additionally the initial 8751/8651 ratios of the granites (0.710-0.725) are much lower than that of the HP sediments (at 280250 Ma) indicating a less fractionated (and more immature) source for the S-type granites as suggested by the geochemistry. \Miether this source is part of a deeper level of the HP or an underlying older unit is not known. Preliminary initial CNd values for S-type granites are similar (-5.9 and -6.1, depleted mantle model ages of 1.7 and 2.04 Ga) and suggest that the protolith consists of dominantly old (reworked) upper crustal material. Permian I-type ^nites outcrop along the very eastemmost part of the HP. They exhibit a wide compositional range (SiOi = 63-76%) and are characterised by high Cr, V and Ni, suggesting some mafic input. Nd-Sr isotope systematics of the l-iype ^anites suggest an involvement of older crust (eNd -4.4 to -3.9, depleted mantle model ages of 1.3-1.48 Ga, and initial 87sr/86sr of 0.7072-0.7158) and can be interpreted as either a mixtur^f younger mafic and older felsic components or from a relatively mafic old source, both consistent with geochemistry. Both geochemistry and initial S^Sr/S^Sr, especially for the more mafic rocks, suggest that the ENd values are not due to interaction with sediments of the HP, or for that matter with other sediments. This involvement of older infracrustal material suggests the existence of old, Proterozoic (?) crust beneath the eastern Hodgkinson Province. If this is indeed the case then it suggests that lateral accretion has not played as an important process in formation of the north Queensland crust as previously thought.

QUANTIFIED FLUID CIRCULATION DURING HERCYNIAN METAMORPHISM SHOWN BY DIFFERING STRONTIUM AND OXYGEN ISOTOPIC MEASUREMENTS. CHAPMAN HJ and BICKLE MJ Department of Earth Science, University of Cambridge, Downing Street, Cambridge CB2 3EQ U.K. Strontium isotopic diffusion distances of 0.4m in carbonate and ~7m in pelite-psammite of the Hercynian Trois Seigneurs Massif in the French Pyrenees have been estimated from detailed ®^Sr/®®Sr profiling. A depositional/diagenetic age for these sediments of 457Ma V. 20Ma with initial ®^Sr/®®Sr ratio of 0.7085 V . 0.0004 were given from 10m length scale averaging of the numerous thin slice and hand-specimen sized whole-rock analyses. The model pre-Hercynian ®^Sr/®®Sr profile across a 15m wide metacarbonate horizon based on the 450Ma depositional age and 0.7086 initial ®®S^®®Sr ratio is compared to the measured post-Hercynian ®^Sr/®^Sr prdfile. Strontium isotopic transport distances inferred from this comparison were smaller than predicted from the pervasive homogenisation of oxygen isotopic values. The systematic shifts in oxygen isotopic compositions in the higher-grade parts of this high temperature - low pressure metamorphic sequence were explained as a consequence of pervasive influx of surface derived or connate water during prograde Hercynian metamorphism (Wickham SM and Taylor HP (1985) Contrib mineral petrol 91 122-137). Strontium values do not support this interpretation and either flow was perfectly layer parallel or more probably the regional scale alteration of oxygen took place by fluid circulation in the brittie regime prior to or during the early stage of metamorphism. Flow along cracks with incomplete diffusion exchange between fluid and wall rock would allow greater differences in oxygen and strontium isotopic transport than a pervasive advective transport with localised fluid - solid in equilibrium. POSSIBLE MIXED SOURCE ROCKS IN THE BEGA BATHOLITH: CONSTRAINTS PROVIDED BY COMBINED CHEMICAL AND ISOTOPIC STUDIES B.W. CHAPPELL. Department of Geology, Australian National University, and M.T. MCCULLOCH, Research School of Earth Sciences, Australian National University The Bega Batholith is a large (8940 km^) C:aledonian-age (420-380 Ma) granite complex occurring at the eastern edge of the Lachlan Fold Belt in south-eastern Australia. It is dominantly (>99%) of I-type character. It extends for 300 km parallel to the continental margin and it shows very striking lateral chemical and isotopic asymmetry, through a distance of 80 km. Changes in chemical composition across this batholith are dominated by decreasing Na and Sr contents towards the older continent to the west. In the same direction, initial 87Sr/86Sr ratios increase from around 0.704 to 0.709 and e^d values decrease from +4 to -9. 5180 values increase slightly, and less systematically, from near 8%o to 9-9.5%o. All of these changes are apperently consistent with an increasing proportion of a sedimentary component in the ^nites on the western side of the batholith. We have examined the isotopic features of the more westerly granites in detail, since these are the most isotopically evolved and provide the opportunity to look at the relationship between chemical and isotopic composition. The Glenbog Suite comprises 12 separate plutons with a total area of 1500 km^ extending for 265 km close to the western edge of the batholith. All rocks in this suite share distinctive petrographic and chemical features and have relatively evolved isotopic compositions. The latter are similar throughout the suite, but not uniform, with initial 87Sr/86Sr ratios ranging from 0.7083 to 0.7108 and e^d values from -7.2 to -5.5. In the samples that have been studied isotopically, total FeO ranges from 1.7 to 4.7%. There is no correlation between the chemical and isotopic compositions, so that most of the variation in isotopic values is found throughout the range of chemical composition. This observation can be used to argue that any mixing pf isotopically different components must have occurred prior to partial melting of the source rocks and this places strong constraints on the development of those sources in the bathoHth.

AN ION MICROPROBE STUDY OF SECULAR BORON ISOTOPE VARIATIONS IN TOURMALINES. CHAUSSIDON Marc. CRPG-CNRS, BP20, 54501 Vandoeuvreles-Nancy, France and ALBAREDE Francis, ENSG-CRPG, BP20, 54501 Vandoeuvre-les-Nancy, France. Recent measurements of HB/I^B ratios in tourmalines from massive sulphide deposits and tourmalinites by thermal ionization mass spectrometry of cesium metaborate have shown large variations («-23%c<31lB<«+22%o) which are interpreted as resulting mainly from the relative contribution of marine and nonmarine evaporites in the source of these rocks. In the present study we have calibrated an IMS3f ion probe for the determination of in situ 31 iB values in tourmalines. The analyses are obtained with a 2-5|im in diameter O" primary beam, secondary ions being analyzed at mass resolution of 2000 so that ^^BH interference on the ^^B peak is resolved. For given analytical conditions, instrumental mass discrimination varies linearly with the product Mass/Charge ratio by ion emissivity of the octahedral cations (Fe+Mg+Mn+Ti+Li) from values of «-65%o for Li-rich tourmalines up to values of «-60%o for Mg-Fe rich tourmalines. Reproducibility of the d^^B measurements on tourmaline standards is currently better than ±l%o. Results obtained on single crystals in various magmatic and sedimentary rocks (from 3.8 Gy to present, e. g. Isua WestGreenland, Swaziland South Africa, Tanco Canada, Birrimien rocks from Africa, Himalayan leucogranites) range between -30%o and +\7%o. No d^^B variation in excess 2-3%o has been found in chemically zoned tourmalines suggesting low magnitude (<5%o) isotope fractionations between tourmaline and hydrothermal fluids. There is no simple correlation between the age and the d^^B values though the most negative d^^B values are found in some «3.1 Gy old rocks from Swaziland. On the contrary the large B^^B variations found between rocks of different origins seem to be related in part to the bulk chemistry of the tourmaline grains, which suggests either a large-scale isotope fractionation and/or mixing processes. 1 Palmer and Slack (1989) Contrib. Mineral. Petrol. 103,434-451. 2 Swihart and Moore (1989) G.C.A. 53, 911-916.

HIMU-EM: THE FRENCH POLYNESL\N CONNECTION CHAUVEL, Catherine and HOFMANN, Albrecht W., MaxPlanck Institut fiir Chemie, Postfach 3060, D-6500 Mainz, FRG H M U , EM 1 and EM 2 are three of the main mantle reservoirs that give rise to oceanic island basalts. They represent the end members That produce the extreme isotopic compositions measured on intraplate volcanics. HIMU is characterized by very radiogenic Pb isotopic compositions, EM 1 by'relatively unradiogenic Sr, Nd and Pb isotopic compositions, and EM 2 by Nd isotopes similar to EM 1 but with much more radiogenic Sr and Pb. In French Polynesia, all three mantle components are represented in volcanic rocks. Characteristic HIMU signature is found in Tubuai, Mangaia and Rurutu, EM 1 is present in the source of Rarotonga volcanics and EM 2 dominates the composition of most Society Islands. Intermediate values between the 3 end members are found in most islands. We suggest that the three reservoirs are not independent but are physically related in the mande. The HDv4U reservoir is thought to be recycled oceanic crust that lost part of its lead through hydrothermal processes prior to and during subduction. EM 1 and EM 2 represent a mixture of the same recycled oceanic crust and sediments that were subducted together with the crust. The trace element pattern and the isotopic composition of HIMU lavas can be quantitatively modelled using a mixture of -25% old recycled MORB crust and 15% mantle peridotite. The extreme Pb isotopic composition is reached assuming that the Pb loss from oceanic crust occurred 2 Ga ago through hydrothermal alteration and preferential dissolution of sulfides in the subduction zone. Pb isotopic compositions similar to those of EM 1 and EM 2 can also be modelled assuming that sediments with average crustal Pb isotopic compositions were subducted and recycled into the mantle together with the underlying MORB oceanic crust. Pelagic sediments (|i~5 and K-6) account for the Pb isotopic composition of EM 1 whereas terrigenous sediments (|i~10 and K~4.5) evolve towards the EM 2 end member. A few percent of sediments in the recycled crustsediment mbcture will destroy the characteristic Pb isotopic signature of the HIMU component. This, together with the low probability of isolating oceanic crust in the mantle for >2 Ga, explains why the HIMU component is sampled so rarely by OIB.

ICOG 7

ABSTRACTS

an-Nd SYSTEMRTICS IN CJQRSTRL RANGE SEDIMEin'S, TAIWAN: CASE STUDY ON A M3DE3^ ARC-OONTINENT CX)LLISICN CHEN Chen-Honq and LEE Typhoon, Inst. E^rth S c i , Academia Sinica, Taipai, Taiwan, ROC Twenty-four fine-grained sediments of late Paleozoic to Miocene fonreitions, sixteen shales and six sand-mud pairs of turbidites from arc-related Coastal Range, as well as bottom sediments from eight major rivers in Taiwan were studied isotopically to (1) explore the relationship between the eroded sediments and their source region lithologies; (2) check if mud and sand in Coastal Range turbidites were from different sources; (3) see whether or not the shaly sediments in Coastal Range can be a good indicator of the ongoing arc-continent collision at Taiwan. Our results show that the Nd isotopic ratios of Taiwan river sediments ( 6 R d = - 1 1 . 4 ± 1 . 0 ) , are similar to, but more homogeneous than, those for sedimentary formations (-11.6±3.2) in their source areas. The dcndnance of continentderived ccnponents in fine-grained sediments of Coastal Range (-7 to -14) precludes dranatic increase of end, thus preventing the use of Nd isotopes to identify the fresh mantle input from the ongoing collision. This is contrary to the observation that major orogenic events left strong signals i n End of large basins (Michard et al., 1985). In contrast, large amounts of volcanic materials in quartz-poor sands makes these sediments (-1.3 to -5.2) a good indicator of this collision event. Therefore, the lack of end increase in fine-grained sediments does not necessarily mean that there occurred no orogeny. Isotopic study on sand-mud pairs in turbidites of the Coastal Range shows that these sediments can be explained by the mixing between the continental crust represented by average Taiwan river sediments and the northern Luzon arc volcanics. Unmixing of differ^t provenance coirponents may have taken place during sedimentary sorting in the turbidity current depositional environment.

THE PRESENCE OF IN PROTOPLANETS CHEN, J. H. and G. J. Wasserburg, Lunatic Asylum, Div. of Geological & Planetary Sciences, Caltech, Pasadena, CA 91125 We report the Pd-Ag data in metal, sulfide and phosphide phases in a wide variety of meteorites. The results provide a basis for evaluating the relationship between and in meteorite parent bodies. The excess C^Âg) is inferred to be due to the in situ decay of (ti/2=6.5xl0®y), and the difference in between meteorites appears to provide a chronometer for early planet formation and differentiation. There is a good correlation between and for different pieces of metal from group IVA irons. The slope ratio of 2.4 x 10"^ and of this correlation line corresponds to a the initial ^Âg/^Âg is close to normal silver. These results support the common origin of tiiese meteorites. By contrast, group IVB and some anomalous irons also show and plot far below the IVA reference line and require a more complex history. We have recently extended our search for ^Âg' to the metal phases in pallasites and mesosiderites. The results on Eagle Station show the same ^ÂgPÂg ratio as in normal silver and indicate 6.3x10"^. The results on two other pallasites, Brenham and Glorietta Mountain show a of '-5%c. This implies a of -1x10'^. Horse Creek, an anomalous mesosiderite or hexahedrite, shows a value of 5.5%o and a ^"'AgV^^'Pd of 6.4x10-^. Morristown, a 3A mesosiderite, yields a of 2.6±0.6%o, and The common presence of in many iron meteorites, some pallasites and possibly in mesosiderites constitutes strong evidence in support of the presence of live during the segregation of Fe-Ni. If these meteorites were produced from a reservoir which had the same initial ^"'Pd/'^'Pd ratio, and the anomaly is from in situ decay of the differences in ratios indicate time scales of major initial chemical fractionation of Pd/Ag among meteorite parent bodies. The results indicate that the formation time intervals between meteorites were very short (<40my). The relationship of this nominal relative time scale to an absolute time scale is not yet established. (#703)

ABSTRACTS

ICOG 7

STUDIES ON THE CHARACTERISTICS OF Sr, O, Pb ISOTOPES IN MARINE SEDIMENTS AND PRIMARY AGE DATING OF ZIRCONS FROM SOUTH CHINA SEA CHEN Yuwei, GUI Xuntang, Institute of Geochemistry Academia Sinica, Guangzhou Division, China, CHEN Chaomuo, South China Sea Institute of Oceanology, Academia Sinica, China Sixteen surface sediments and 3 submarine cores collected from the South China Sea from the area of 108-118°E, 3 - i r N were analysed for Sr, O, Pb isotopes. Sr isotope ratios in the carbonate fraction of sediments were homogeneous with the average of ^'^ST/^^ST 0.70928 which may represent the Sr isotopic composition of ocean water at that time. In the silicate fraction the pair of Sr and Pb isotope ratios have a negative relationship and have regional distribution features on the north part of S'N 87sr/86sr 0.709080.71246, 5I8O 16.01-21.86%o, on the southwestern part ^'^Sifi^Sr 0.719, 15%o. Pb isotope ratios in carbonate fraction of sediments are 206pb/207p5 1.2015-1.2740 in the deep water, 206pb/207p5 1.1763-1.999 in shallow water. Pb isotope ratios in silicate fractions of southeastern part correspond to present ocean lead, on the northwestern part Pb isotope ratios due to the Pb isotope of mixing type being derived from old rocks. Vertical distribution profile of Sr and Pb isotopes in three cores indicated the different sedimentary environment. The isotope results could be presumed at least 4 sedimentation regions in this area. In addition, some different characteristics between Sr and Pb isotopes during marine sedimentation process were obser\^ed. Microscopically morphology observations and electron probe analyses of zircons isolated from sediments of this area have been done and yield ages 1640 ± 42 Ma, 236 ± 77 Ma by 207pb/206pb method. This paper is one of the projects of South China Sea Scientific Expedition.

Sr-Nd ISOTOPE SYSTEMATICS IN CHARNOCKITES FROM SOUTHERN DsTDIA CHOUDHARY. A.K., HARRIS, N.B.W., HAWKESWORTH, C.J., and VAN CALSTEREN, P.W.C., Department of Eanh Sciences, The Open University, Milton Keynes, MK7 6 A A, U.K. Sr and Nd isotope analyses of incipient chamockites from Kerala indicate that Sr isotopes equilibrated between whole rock, biotite and feldspar at 485 ± 4 Ma, whereas whole-rock-gamet Nd ages lie in the range 1790 ± 80 to 560 ± 20 Ma. These data suggest that whereas Sr isotopes equilibrated during uplift at 485 Ma, Nd isotopes record at least two phases of mobility: the first during regional metamorphism prior to 1790 Ma, and the second at or after 5 ^ Ma during chamockite formation. Analyses of a gneiss-chamockite pair indicate that there is a -10% decrease in Sm/Nd and a similar increase in Rb/Sr during chamockite formation. However, model Nd ages (2730-2440 Ma) are only slightly older than model Sr ages (2440-2090 Ma) which precludes large scale Rb/Sr or Sm/Nd fractionation during the crustal history of these rocks. Chamockite formation, therefore, occurred between 485 and 560 Ma in Kerala, consistent with that in Sri Lanka and with regional thermal events in both Africa and Antarctica. However, chamockites to the north of the Palghat-Cauvery shear zone formed prior to 2500 Ma and biotite-whole rock ages record uplift at 2400 Ma. South Indian chamockites were thus formed during at least two distinct periods of crustal evolution. For the younger event, models of chamockite formation invoking large scale magmatic or fluid movement are ruled out by the small Rb/Sr and Sm/Nd fractionation between gneiss and chamockite.

SINGLE MINERAL Sm-Nd DATING OF ORE DEPOSITION CHESLEY, John T., HALLTDAY. A.N., Dept of Geologic Sciences, U. of Michigan, Ann Arbor, MI, 48109 USA, SHEPHERD. T.J., British Geologic Survey, Keyworth, Nottingham NG12 5GG, U.K., and SCRIVENER. R.C. British Geologic Survey, Exeter, EX4 6BX, U.K. Current understanding of the timing and duration of hydrothermal mineralization is limited by our inability to date directly specific stages of mineral deposition. The Comubian ore deposits in southwest England are associated with Hercynian two-mica granites that were emplaced as discrete plutons into Upper Paleozoic sediments at -280290 Ma . Metalliferous hydrothermal mineralization in Cornwall can be subdivided into four stages: 1) Early sheeted Sn-W vein complexes; 2) Sn-bearing tourmaline veins 3) East-west polymetallic sulfide veins; and 4) Late-stage north-south polymetallic fluorite veins, locally termed "crosscourses". Sm-Nd isochron ages based solely on fluorites from E-W stage 3 veins were obtained at both the South Crofty (SO and Wheal Jane (WJ) mines. Hydrothermal mineralization at the SC and W7 mines is related to the Carnmenellis Pluton (-285 Ma). Mineralization at WJ is hosted by Paleozoic sediments, whereas mineralization at SC is hosted by the granite. The Sm-Nd isochron age of fluorites from SC is 259 ± 7 Ma (2a) with an initial eNd = -5.7 and that of fluorite from WJ is 260 ± 18.0 Ma (2a) with an initial eNd = -7.1. ratios vary from - 0.1 to 0.5 at both mines and can vary by as much as 0.1 to 0.4 within a single hand sample. The similar isochron ages preclude fluid mixing or deposition from genetically different fluids as the cause of variation in Sm-Nd ratios. The initial eNd values suggests that the REE's at both deposits are dominantly derived from the granite. At the WJ mine, however, eNdi was probably lowered as a result of the mineralizing fluid interacting with the sediments or mixing with meteoric fluids. The ages also indicate that hydrothermal mineralization post-dates granite emplacement by 15 to 25 Myrs. The ability to establish the age of mineralization directly using a single mineral has important implications in defining the overall duration of granite-related hydrothermal activity as well as evaluating current models of hydrothermal ore deposition. In addition, direct dating of fluorite may be possible with other tx-pes of hydrothermal mineralization (e.g. MVT deposits).

SHRIMP ZIRCON DATING OF TIMESCALE SAMPLES — CONSTRAINTS AND ADVAxNTAGES

CLAQUE-LONG, J., RSES, ANU, Canberra, Australia, and MAXWELL. S., Applied Geology, UNSW, Australia. The ability of the SHRIMP ion microprobe to resolve complex zircon populations is being applied to Palaeozoic timescale problems. The most useful information comes from ^^^Pb/^^'U ratios, which have good measurement precision and are insensitive to initial Pb correction. Accuracy is determined by calibration to the known Pb/U of a standard zircon via the well established SHRIMP calibration relationship (see Figure), and analytical accuracy better than ±4Ma (2a) is proving routinely achievable in samples of Carboniferous age. The Martins Creek Andesite at the boundary between the Rhipodomella fortimuscula and Delipinia aspinosa Zones is an important Carboniferous marker. Its zircon population includes a magmatic component; significant Pb loss causing Permian apparent ages in several grains; and optically invisible Proterozoic cores in some crj'Stals. SHRIMP'S ability to resolve the magmatic age from this complex population has resulted in a geologically accurate age where other, equally precise, but less selective analytical methods may have yielded a meaningless mixed ase. 0.4-^= ^ Standard zircon SL13 (572 Ma)

0.2

Carboniferous zircons

measured UO/U 0.1 7.2 7.6 8.0 8.4 SHRIMP calibration diagram comparing data for a timescale sample with analyses of the standard zircon. The calibration relationship between PbAJ and UGAJ is drawn through the data.

ICOG 7

SHRIMP DETERMINATION OF THE AGE OF THE DEVONIAN-CARBONIFEROUS BOUNDARY

CLAQUE-LONG. J., RSES, ANU, Canberra, Australia, JONES, P., BMR, Canberra, Australia, ROBERTS, J., & MAXWELL, S., Applied Geology, University of NSW, Kensington, Australia. The SHRIMP ion microprobe has been used to date zircons from a 1cm thick bentonite located at the Devonian-Carboniferous boundary in the Hasselbachtal stratotype section in W. Germany, and a tuff at the same biostratigraphic level in Australia. Multiple replicate analyses have yielded indistinguishable ages, giving 353.2 ± 4.0 Ma (2a) for the age of the boundary (see Concordia diagram). Obtaining an age in this stratotype boundary section depended on the ability of SHRIMP to date a very small sample and to target small areas of structurally complex zircons. The bentonite is within centimetres of both the conodont and ammonoid definitions of the boundary; future refinements of the age of the boundary may depend on reanalysis of this uniquely placed volcanic. Demonstration that this thin horizon is amenable to accurate dating highlights the value of defming boundaries in the vicinity of known dateable volcanics.

0.065 -

0.045 -

Hasselbachtal Bed 79 SHRIMP zircon data

353.2 ± 4.0 Ma (2a)

0.025

0.2

207,

0.4

Pb/235u

OJ6

DATING METAMORPHISM USING URANIUM-RICH METASOMATIC TITANITE IN CALCSILICATES. R.A.CLIFF, Dept. of Earth Sciences, University of Leeds, Leeds LS2 9JT,UK Detailed studies of thermal history require determination of the age of peak metamorphism as well as stages in. subsequent cooling and uplift. Titanite (sphene) is one of the few minerals that have sufficiently high closur'e temperature, for U - P b , to act as closed systems immediately following crystallisation in the amphibolite facies. In the Alps titanites have survived regional metamorphic re-heating to 600-650°C with little or no resetting of their U-Pb system. Calc-silicates, typically containing titanite, often result from metasomatic interactions between carbonate rocks and pelites. In the Dalradian of Connemara, Ireland, a metasomatic diopside-rock was formed at the peak of upper amphibolite facies metamorphism by the influx of fluids from dehydration of pelites, into a dolomite marble, which acted as a major aquifer during metamorphism. Titanite from this rock is unusually U-rich (c.400ppm), resulting in highly radiogenic lead which defines a precise 478 ± 3 Ma concordant age. Metamorphic conditions are close to or below estimated titanite closure temperature so that the age is interpreted as a new and more direct determination of the age of peak metamorphism in Connemara. Similar metasomatic rocks are commonly found in other belts and further analyses are in progress to establish whether the unusual uranium contents are typical of titanite from such rocks and thus provide a means to date metamorphic crystallisation generally.

ABSTRACTS

CARBONATE CONCRETIONS QUANTIFY DIAGENETIC MASS TRANSPORT OF SILICA AND ALUMINA INTO SANDSTONES MAX COLEMAN. BP Research Centre, Sunburyon-Thames, Middx. TW16 7LN, U.K. & JON GLUYAS BP Norway Exploration, Forusbeen 35, PO Box 197, 4033 Forus, Stavanger, Norway. Deeply-buried, Jurassic marine sandstones form North Sea oilfield reservoirs. Sandstones of the Brae, Gam and Ula formations, presently at depths between 3.5 and 4.7 km, contain carbonate concretions, formed early in the diagenetic history of the sediment. Evidence for early growth is: pre-compaction grain texture, detrital quartz grains without overgrowths, preservation of unaltered feldspar grains and 32 - 40% intergranular cement (which, if it is not displacive, re-presents an equivalent porosity, indicative of only very shallow burial). 513CpDB of the carbonate cements, varying between 0 ®/oo and -10.5 7oo, and presence of associated minor pyrite imply formation from degradation of organic matter by microbial sulphate reduction, with a contribution from seawater-bicarbonate, and confirm an early origin. However, both the poikilotopic texture and 5180pdb» -3.6 7oo to -13.5 °/oo, imply recrystallisation of the original cement. Despite this, the other evidence shows no disturbance of the original, detrital material within the concretion. Depositional mineralogy and facies of sediment within concretions are identical to those outside and it is not clear why the concretions formed where they did. They are not radially zoned. Their compositions have been used to estimate net change in chemical composition of a whole sedimentary unit between early diagenesis and deep, present burial depths, by comparison with sediment outside the concretions. Multiple samples provided statistically-valid mean values, which when corrected for compactional differences showed that, despite feldspar dissolution, the amount of silica added during diagenesis varied between 295 ± 150 and 355 ± 155 kg.m-3 (present-day volume). Only in one case was a statistically significant amount of alumina added, 4.2 ± 0.4 kg.m-^. These data provide a quantitative constraint on models of subsurface fluid flow. GEOLOGIC AND ISOTOPIC CONSTRAINTS ON EVOLUTION OF THE PRE-3.5 BILLION YEAR CRUST-MANTLE SYSTEM KENNETH D. COLLERSON, Earth Sciences, University of California, Santa Cniz, CA, 95064, USA Models of early chemical differentiation of the Earth are limited by uncertainties in primitive crust-mantle compositions and knowledge of the extent and time-scales of mantle fractionation. The record of pre-3.5 Byr crust, preserved fortuitously in many cratonic blocks affords direct information about crust-mantle differentiation during the first billion years of Earth history. Geological relations in these ancient terrains are commonly obscurred by the effects of superimposed thermotectonism and isotope systems commonly yield equivocal information regarding age and source chemistry. Developments in analytical instrumentation (e.g., SHRIMP) and techniques have greatly assisted in identification of ancient crust and have allowed inferences to be made concerning crust formation. Despite these advances, a major difficulty in directiy addressing the problem of early crust formation relates to the extremely limited data-base of early crustal compositions. This may reflect the fact that early crust and mantle lithosphere was dynamically unstable and was efficientiy recycled back into the mantie as a result of intense meteorite bombardment and tiiermally driven turbulent convection Extensive development of tonalite-dominated continental crust was probably strongly limited by rate of formation of the oceans, and extent of hydration of mafic crust. Generally positive eNd values (+1 to +5) of Archean rocks require source regions with Sm/Nd ratios greater than the chondritic value for a significant period of time prior to melt formation. Geochemical fractionation in the early mantie, reflects extraction of light rare earth element (LREE) enriched material. However, the nature of this enriched reservoir is equivocal; possibilities include earlier continental crust (such as Acasta gneiss crust), enriched mantle, or subducted mafic/ultramafic crust. The principal requirement istiiatthe enriched component must have been isolated from the homogenising effect of convection in the upper mantie for a long time period (0.3-0.4 Byr). eNd(O) values ranging from +18 to +56 for ca. 3.8 Byr komatiites from Labrador show tiiat depleted, extremely heterogeneous mantie formed before 4.1- 4.2 Byr. This fractionation may have occurred during consolidation of a primordial magmasphere.

ABSTRACTS

ICOG 7

ZIRCON U-PB AGES RELEVANT TO THE CAMBRIAN NUMERICAL TIME-SCALE COMPSTON. W., WILLIAMS, I.S., RSES, ANU., Canberra, ACT 2601, Australia, KIRSCHVINK, J., Div. of Geological & Planetary Sciences, Caltech, California, 91125, U.S.A., and ZHANG, Zichao, Chinese Academy of Geological Sciences, Yichang, China. New zircon U-Pb ages from Early Cambrian volcanics in South China and Morocco strongly favour younger estimates for the Cambrian time-scale as proposed by Odin (1982). Zircons were extracted from a tuff in the Serie Lie de Vin, Tiout, Anti-Atlas Mountains, Morocco, -350 m below the first appearance of early Cambrian trilobites and archaeocyathids, close to the Tommotion-Atdabanian boundary. Nearly all zircons, as measured by ion microprobe, show equal 206p5/238u ages to within 2% (a) and average 521 ± 4 Ma (ta). Exceptions are a few areas that have lost Pb. The mean 207pi5/206p5 ages of the zircons agree at 515 ± 21 Ma (ta), with precision limited by counting statistics. The maximum age of the Tiout tuff from these data is the older limit for the mean 207pb/206p5 age, 536 Ma. Accepting this would imply that nearly all the zircons lost closely the same sm^fractionof their radiogenic Pb {ca. 3%), which seems improbable. A more realistic limit to the tuff age is 525 Ma, based on their maximum mean 206pb/238u age. The Tommotian-Atdabanian numerical boundary therefore is unlikely to be older than this value. The comparitively simple Pb isotope systematics of the Tiout tuff zircons warrant conventional isotope dilution analysis on single-grains to improve precision in 207pb/206pb and accuracy in 206pb/238u. Less conclusive results have been obtained for zircons from a bentonite recently found within the Sinian-Cambrian sequence at Meishucun, close to the base of the Cambrian. An age is indicated that is significantly younger than the ca. 570-580 Ma shale isochrons from overlying Atdabanian strata. Further ion probe study on these samples is in progress. If the younger age is sustained, the older shale results imply the presence of substantial amounts of inherited ilHte.

ION MICROPROBE U-Pb ZIRCON DATING WITHIN THE LOWER CAMBRLW OF SOUTH AUSTRALA J.A.COOPER & R.J.F. JENKINS, Geology and Geophysics, The University of Adelaide, Adelaide, 5000, South Australia. W.COMPSTON & I.S.WILLL^MS, Research School of Earth Sciences, Australian National University, Canberra, A.C.T, 2600, Australia. The problem of dating the early Cambrian has been explored in the continuous Adelaidian-Cambrian sequence of South Australia. An isolated and narrow (3.5-4.5 cm) felsic tuff occurs high in the exposed section of the Heatherdale Shale at Sellicks Hill, South of Adelaide, near the west coast of Fleurieu Peninsula. Occasional specimens of Concoryphid trilobites have been reported from immediately above and below the tuff bed, which itself buries trilobite trackways. This, with other regional trilobite evidence, and Archaeocyathids at a slightly lower level in a layer-cake stratigraphy, have together suggested a late Atdabanian - early Botomian age. Zircons extracted from the tuff have been dated by the U-ThPb method using the SHRIMP ion-microprobe. Rare inherited grains and slightly discordant analysed areas have been easily identified and their data systematically excluded. Slightly high recordings of mass 207, significant at this age level, biased some earlier 207Pb/235U and 207Pb/206Pb age calculations upwards. Subsequent efforts to suppress suspected hydride interference considerably reduced this interference but these two sytems remain relatively insensitive to the method in thistimerange. The ion-microprobe 238U/206Pb age calculation suffers no such interference and is considerably more sensitive. It indicates that the Atdabanian - Botomian boundary is younger than currently assumed, and consequently suggests that the age of the base of the Cambrian is younger than the 570 Ma that is favoured in most modem time scales.

THE CRUSTAL EVOLUTION OF SOUTH AMERICA U.G.Ctordani; W.Teixeira; C.C.G.Tassinari; K.Kawashita (Instituto de Geociencias Uhiversidade de Si) Paulo - Brasil) Integrated interpretation of Sr, Pb and Nd isotopes has led to the estimation of the continental growth of South America. The continent has been divided into crustal domains with internally coherent structure and geochronological patterns. Assuming an uniform crustal thickness for,the Brazilian Shield, and considering the special case of the Andes, the crustal volumes of each of the domains have been calculated. Then, from the available isotopic constraints, the estimated amounts of continental crust accreted in each period of time have been established. The main periods of episodic crustal growth for South America are related to the Late Archean and Early to Mid-Proterozoic. During Late Proterozoic and Phanerozoic,some accretion occurred, but crustal reworking and crustal contamination of mantle derived material were the predominant processes. Considering the geological constraints of each domain, their age and isotopic geochemical parameters,the following proportions of continental growth through geological time were estimated: Early and Late Archean 33% Early Proterozoic 40% Middle Proterozoic 15% Late proterozoic 5% Phanerozoic 7%

Hf ISOTOPE RELATIONSHIPS AND U-Pb AGES FOR THE LATE-ARCHEAN ABITIBI GREENSTONE BELT, CANADA. F. CORFU and S. NOBLE, Dept of Geology, Royal Ontario Museum, Toronto, Ontario, Canada, M5S 2C6. Single filament techniques have been developed for the isotopic analysis of Hf and Lu. Hf is run using Ir-Mo-C as activators yielding results that are comparable with those reported for the Patchett triple filament technique. Lu is run as a metal on C; this method eliminates the need to chemically separate the REE since Yb is easily ionized prior to Lu. The initial Hf ratios in zircons from all representative igneous components of the southern Abitibi belt, including ultramafic, calc-alkalic and alkalic units, define e-values that range from +9 to -h4.5. This is consistent with the existing e-Nd data for the belt that span about -1-3 to +1 and indicate derivation from depleted mantle. Sialic crust was not involved in the formation of the belt, an interpretation also consistent with the paucity of pre-2750 Ma inherited and detrital zircons in igneous and sedimentary rocks. Zircon geochronology indicates that the bulk of the greenstone belt evolved through episodes of komatiitic, tholeiitic and calcalkalic volcanism between mainly 2730 and 2700 Ma, followed by regional folding, thrusting and calc-alkalic plutonism between 2700 and 2690 Ma. A subsequent magmatic event produced alkalic volcanic and plutonic rocks and was associated with nonmarine sedimentation at 2685-2675 Ma. These events were concluded by thrusting, local folding and faulting. As shown by other workers, gold-mineralization, hosted by extensive fault zones, formed 50-100 Ma after build-up of the greenstone belt

ICOG 7

ABSTRACTS

THE ORIGIN OF RHYOLITE IN THE TAUPO VOLCANIC ZONE, NEW ZEALAND; A STRONTIOM, NEODYMIUM, LEAD AND OXYGEN ISOTOPE STUDY. CORLETT. Graeme and CHAPPELL, B.W. Department of Geology, McCULLOCH, M.T., and WOODHEAD, J. D., Research School of Earth Sciences, Australian National University, Canberra, A.C.T. 2601, Australia. Rhyolite rocks, their minerals and basalts from the Cainozoic Maroa Volcanic Centre in the Taupo Volcanic Zone have been analysed for strontium, neodymium, lead and oxygen isotopes. Mesozoic grey wackes which outcrop to the west and east of the volcanic zone have also been analysed for these isotopes to test theories invoking their contribution to rhyolite magma genesis. Sr and Nd rhyolite isotope data (87Sr/86Sr = 0.70499 to 0.70579; e^d = 2.0 to 0.1) suggest simple mixing of primitive basalt (87Sr/86Sr = 0.70382; e^^ = 5.3) with either Western greywacke, which compositionally reflects an island-arc provenance or Eastern greywacke, composed mainly of continentally-derived sediments. However Pb and isotopic compositions of rhyolitic plagioclase separates (206pb/204p5 = 18J65 ^o 18.832; 207pb/204pb = 15.6I8 to 15.632; 208pb/204pb = 38.636 to 38.724; 5I80 = 6.1 to 7.9, average = 6.7) indicate minimal involvement of Western greywacke (206pb/204pb = 18.696; 207pb/204pb = 15.619; 208pb/204pb = 38.571; 5I80 = 8.7) which has been suggested by earlier workers (e.g. Ewart and Stipp, 1968) as a likely source of the rhyolite magma. Models integrating 6 i s o t o p e data with Sr-Nd-Pb isotope systems, using Maroa basalt (2o6pb/204pb = 18.650; 207pb/204'pb = 15.580; 208pb/204pb = 38.447; S^^O = 5.7) and EasternCTeywacke(206pb/204pb = 18.828 ; 207pb/204pb = 15.642; 208pb/2d^pb = 38.729; 5180 = 12.6) and its hydrothermally altered equivalent as end member components, also cannot satisfactorily explain the rhyolite compositions by either simple mixing, or combined assimilation-crystal fractionation processes. The preferred model for rhyolite genesis is partial melting of a relatively homogeneous intermediate igneous source that has been contaminated prior to melting.

Sr ISOTOPE DISCORDANCE PATTERNS IN HYDROTHERMALLY ALTERED PROTEROZOIC GRANITES. CREASER, Robert, La Trobe University, Department of Geology Bundoora VIC 3083 AUSTRALL\. Now at: California Institute of Technology, Division of Geological and Planetary Sciences, Pasadena CA. 91125 USA. Proterozoic granites from South Australia are precisely dated at 1585-1600 Ma by U-Pb zircon and titanite methods. However, RbSr total-rock (TR) systems record ages that are typically 100 Ma younger than the U-Pb ages on a regional scale. Study of Sr-isotope TR system for regional granites and mineral systems for individual panites has been undertaken to investigate this age discordance. 0isotope analyses of selected minerals are also presented. TR Rb-Sr analyses from regional samples show a coherent pattern of Sr isotope discordance. With increasing total-rock Rb/Sr ratios, Sr "model ages" or calculated initial ^^Sr/^^Sr ratio (Ri) at the U-Pb age decrease systematically. Felsic g^ites have very young model ages or impossibly low apparent Ri. Only granites with high Sr content have remained closed systems since crystallisation. Study of Sr-isotopes in minerals reveals a similar pattern of discordance for all granites. Biotite ages range from the U-Pb age (-1600 Ma) to 1530 Ma and are always older than feldspar ages. However, the biotite ages are not interpreted to have simple chronological significance (such as cooling ages) as 0-isotope data suggests significant fluid exchange after crystallisation. Feldspar pairs and isochrons mostly record ages of -1500 Ma, similar to the age determined for hydrothermal sericite from a nearby orebody. All feldspars show O-isotope evidence for fluid interaction. The young Rb-Sr ages are interpreted to result primarily by loss of Sr* from alkali feldspar, facilitated by fluid penetration along perthite lamellae and cleavage traces. Titanite, homblende and augite have accepted some of this Sr*. Apatite appears remarkably resistant to Sr isotopic disturbance. Six apatites from rocks interpreted to be related on geochemical and Nd isotopic grounds record similar Ri (0.7087 ± 3). The loss of Sr* from alkali feldspar, the resistance of apatite to Sr isotope disturbance and the partially reset biotite ages can all be explained by an episode of hydrothermal alteration at a temperature < 250°C using available Sr diffusion data.

NEW CONSTRAINTS ON THE THERxMAL EVOLUTION OF THE PERIADRIATIC LINEAMENT FROM APATITE HSSION TRACK ANALYSIS COYLE, David A., Department of Geology, La Trobe University, Melboume, 3083 Australia, RATSCHBACHER, Lothar, and KNAUS, Annette, Institut fiir Geologie der Universitat Tubingen, D-7400 Tubingen, West Germany Fifteen samples were collected from Sudtirol, northern Italy, and Kamten, Austria, and subjected to apatite fission-track analysis (AFTA). Ages range from 10.8 Ma on the Periadriatic Lineamant south of the Tauem window, to 70 Ma in Austridic basement rocks in Kmten. Confined mean lengths range from 10 to 14 microns, which together with their distributions aid in determining the specific thermal histories through which each sample has evolved. In Kamten, the low temperature thermal history is related to the periadriatic intrusive event at ca. 28 Ma, with no evidence of regional heating related to the Alpine orogeny. The thermal effects of the intrusions are narrow, several hundreds of metres for an 120° C isotherm, which indicates high-level intrusion into cool rocks. Thermal modelling based upon the confined length distributions suggests that the intrusions cooled linearly from 250° C at ca. 28 Ma toca.60°Cby 10 Ma. This is also the case in Sudtirol south of the Periadriatic Lineament, where the pattem of AFTA ages and confined lengths is similar to that in Kamten. North of the Periadriatic Lineament in Sudtirol there was Miocene regional heating above 120°C. In the region between the DefereggerAntholzer-Vals (DAV) line and the Periadriatic Lineament, the temperatures were not high enough to reset the Paleozoic Rb/Sr biotite system, and so the AFTA data places important constraints on the late Alpine evolution of the three blocks delineated by these faults. Offsets in the AFTA ages across the DAV line and the Periadriatic Lineament in Sudtirol indicate up to several hundreds of metres of late Miocene vertical displacement across these lineaments, with the northen blocks moving up relative to the southern blocks.

LASER MICROPROBE INVESTIGATION OF SULFUR ISOTOPE VARIATION IN A SEAFLOOR HYDROTHERMAL SPIRE, AXIAL SEAMOUNT, JUAN DE FUCA RIDGE CROWE, Douglas, and VALLEY, John, Dept. of Geology, Univ. of Wisconsin, Madison, WI, 53706, USA, and SCOTT, Steve, Dept. of Geology, Univ. of Toronto, Toronto, M5S lAl, Canada, and HANNINGTON, Mark, Geological Survey of Canada, Ottawa, Ontario, KIA 0E8, Canada The Axial seamount site is located on the Juan de Fuca Ridge at 45* 57'N, 130' 02'W. Low temperature (<35* C) hydrothermal fluids currendy emanate from numerous vents, cracks and fissures within the caldera, and several large (to 12m) silica + sulfide + sulfate hydrothennal spires reflect previous higher temperature (to 300' C) hydrothermal activity. The sulfide assemblage in these inactive spires is comprised mainly of sphalerite + wurtzite + marcasite. Laser microprobe traverses across five temporally distinct fluid conduits within one of the inactive higher temperature spires reveal significant temporal variation in d^^S, both within individual conduits and between conduits. The maximum intraconduit 5-^S variation is 5.0 %o and the maximum inter-conduit variation is 7.4 %o. Inter-conduit variation occurs primarily due to variable amounts of seawater mixing with hydrothermal fluids witiiin the spire; generally decreases through time due to lesser amounts of seawater that invade more mature, heavily mineralized spires. Intra-conduit variation has not been documented at this scale (5 %o within 1 mm), and the potential mechanisms responsible for this variation include 1) within-spire seawater-hydrothermal fluid mixing, which also produces the inter-conduit variations, 2) overprinting of earlier formed sulfide phases by later hydrothermal pulses or 3) more deep-seated convection and redox processes in the seafloor subsurface that alter the hydrothermal fluid S'^S. These data confirm the results of Crowe et al. (1990, GCA), and show the imponance of the 100 |im spot size of the new laser microprobe technique for deciphering complex hydrothermal systems.

ABSTRACTS

ICOG 7

GEOCHRONOLOGY OF THE RES Ni-Cu DEPOSIT, FLIN FLON, MANITOBA, CANADA: THERMAL HISTORY OF A METAMORPHIC TERRANT. GUMMING. G.L.. Depanment of Physics, University of Albena, Edmonton, Alberta, Canada, T6G 2J1 and KRSTIC,D., Dept. of Physics, University of Alberta. Intermediate composition gneisses which surround the ultramafic host rock of the Res Ni-Cu deposit in northern Manitoba began to coolfromelevated temperatures of approximately 700 °C at 1860 Ma when new zircons formed. Subsequently, the temperature decreased very slowly, with a relaxation time of 160 Ma, cooling through the blocking temperature for hornblende at 1799 Ma and that of biotite at 1735 Ma. Using published values for blocking temperatures, we find that the temperature-time data are consistent with a simple exponential decrease in temperature and do not require discrete pulses of thermal activity during the first 150 Ma of cooling. There is no indication of temperatures as high as the blocking temperature for biotite subsequent to 1735 Ma. Data from sphene yield an age close to 1820 Ma, intermediate between that of zircon and hornblende, and consistent with a blocking temperature of about 525 °C, while apatite gives an age of 1810 Ma and therefore a blocking temperature between that of sphene and hornblende, i.e. about 510 °C. The good precision on all the ages obtained, provides rather precise relative blocking temperamres for zircon, sphene, apatite, hornblende and biotite. Pb/Pb ages on the gneisses are in excellent agreement with the zircons ages, indicating a rather high blocking temperature for whole rock and mineral Pb/Pb systems. Zircons in the ultramafic host of the ore body are slightly younger than the gneisses, close to 1847 Ma.

40Ar-39Ar DATING OF FLUID INCLUSIONS IN QUARTZ DAI Tongmo,QIU Huaning, LI Zhenghua, PU Zhipin, Institute of Geochemistry (Guangzhou Branch) Academia Sinica. Guangzhou Wushan 510640 In order to investigate the possibility of dating the minerogenetic age, argon isotopes in fluid inclusions in quartz associated with granite-hosted tungsten mineralization in Lushui, Western Yunnan, have been determined by the '^0Ar-39Ar technique. Two samples of coexisting muscovite and quartz in the piece of tungsten ore were selected. The muscovite was analysed by stepped heating, and the quartz by stepped heating and by crushing in a ultra-vacuum device instead of the crucible for heating. Grains were crushed by repeatedly lifting and dropping the pesde, using an electromagnet. The age spectrum or the muscovite by stepped heating is essendally flat. Its plateau age is 38.8 ± 0.6 Ma and the isochron age is 38.6 ± 0.8 Ma. The two ages are concordant and indicate the time of the tungsten mineralization. The saddleshaped age spectrum of the quartz by stepped heating shows the existence of excess "^^Ar, in which the low plateau age is 31.4 Ma. The quartz by crushing forms a smoothly falling age spectrum marked by high ages at low value of ^^Ar release, and the minimum age is higher than its real age. The isochron diagram by crushing in vacuo yields sensible information. The line corresponds to an age of 38.6 ±1.9 Ma, which shows excellent coincidence with the results obtained from the muscovite and agrees within experimental error with the low plateau aae of the quartz by stepped heating. Another analysis of quartz in the miarolite from Kuichi, Fujian province, also yields sensible information.

LATE MID-PROTEROZOIC CRUSTAL GROWTH IN THE NORTH ATLANTIC REGION J.S. DALY/ Department of Geology, University College Dublin, Belfield, Dublin 4, Ireland In the Adirondack sector of the Grenville orogen, the oldest (c. 1.35 Ga) plutons are magmatic arc tonalites with Sm-Nd model ages (tDM) within 70 Ma of their crystallization ages indicating minimal involvement of significantly older crust in their genesis. 1150 Ma old granitoids associated with massif anorthosites (AMCG suite) have an average tDM age of 1410 Ma, similar to the tonalites. Excluding one alaskite with a high Sm/Nd ratio suggestive of intracrustal rare earth element fractionation, the later c. 1100-1150 Ma old granitoids have tDMs of c. 1330 Ma. All of the younger granitoids have positive initial £Nd values and their chemistry is consistent with intracrustal melting of the arc tonalites. The only indication of significantly older material is provided by pelitic metasediment with a tDM age of 2075 Ma. Contrary to the traditional view of the Grenville as an ensialic orogen, these results dononstrate that the Adirondack sector of the belt comprises a juvenile late Middle Proterozoic terrane. Crust of this age (1.3-1.4 Ga) has also been identified in Proterozoic inliers in the Appalachian-Caledonian belt such as the Chester Dome, Vermont and the Erris Complex of NW Ireland. Potentially greater volumes occur as detritus in the North Atlantic region e.g. in Proterozoic belts such as the Kalak terrane of Arctic Norway, the Dalradian of Ireland and in Caledonian and younger sequences.

ND-SRISOTOPIC CHARACTERISTICS OF MARC DYKES FROM THE SOUTHERN ADELAIDE FOLDBELT, SOUTH AUSTRALIA DALY, J. Stephen, Dept of Geology, University College, Dublin, Belfield, Dublin 4, Ireland, UU, Son^a and GRAY, Chris M., Dept of Geology, La Trobe University, Bundoora, Victorian 3083, Australia Tholeiitic dykes (now metadolerites and amphibolites) intruded the metasedimentary rocks (mainly the Cambrian Kanmantoo Group) in the southern Adelaide Foldbelt, South Australia at various stages of the Delamerian Orogeny at about 500 Ma ago. Although detailed mapping demonstrates that some dykes intruded pre- to syn-Dl, post-Dl and pre- to syn-D2 while others are post-D2 in age, the age range is likely to be less than 30 Ma. Geochemically the dykes resemble oceanic tholeiites especially in their low abundances of incompatible elements, flat REE patterns and in the Nd-Sr isotopic signamres of the best preserved or least contaminated samples. At 500 Ma, the latter have initial ^'^Sr/^^Sr ratios of 0.7030 and initial ENd values of +5.8 to at least +8 and plot in the MORB field in the eNd - ^'^ST/^ST as defined by Ito et al. (1987). One amphiboUte has a high initial 87Sr/86Sr ratio of 0.7104 and a high positive ENd value of +6.7. However some samples of amphibolite and metadolerite have low £n(J values between -1 and +3.6 probably as a result of contamination with a cnistal component such as that seen in a granitic dyke with an initial ENd value of -10 and a depleted mantie model age of about 1800 M a, or in the enclosing Kanmantoo Group metasediments which have similar Nd isotopic features. Considering other available geological and geophysical evidence, the isotopic and geochemical data for the mafic dykes are best interpreted in light of a tectonic model in which the Kanmantoo Group was deposited in a well-advanced extensional sedimentary basin which was most probably a marginal basin/back-arc basin near the present soutiiwest boundary of the Tasman Fold Beh in southeast Australia.

IC0G7

RB-SR AND SM-ND CONSTRAINTS ON THE AGE OF MINERALISATION AND THE ORIGIN OF HYDROTHERMAL FLUIDS IN SW ENGLAND. DARBYSHIRE. D.P.F. NERC Isotope Geosciences Laboratory, Keyworth, UK and SHEPHERD, TJ. British Geological Survey, Keyworth, UK. To address the complex genetic relationship between different styles of polymetallic mineralization and granite magmatism in the SW England orefield, use has been made of whole rock, mineral and fluid inclusion Rb-Sr and Sm-Nd isotope systematics. A fluid inclusion Rb-Sr isochron for quartz from an E-W trending, early tin-tungsten vein at the South Crofty Mine has shown that the Sn mineralization is 20 Ma younger than the host Cammenellis granite (290 ± 2 Ma). Application of this technique to later, N-S trending, base metal-fluorite-barite veins in the Tamar Valley indicates an early Triassic age (c. 235Ma), thus suggesting multiple stages of mineralization over an extended period after granite emplacement Data will be presented to introduce a new approach to dating mineral deposits based on fluorite Sm-Nd isochrons. Fluorite is common to both vein systems and is enriched in E REE (up to 200ppm) and exhibits a wide range in Sm/Nd ratios. Fluorites from an E-W vein at the Pendarves Mine yield an Sm-Nd isochron age for mineralization which is consistent with the established RbSr timeframe. eNd values suggest either precipitation of fluorite from magmatic fluids or isotopic equilibration of extemally derived fluids with the subsurface granites. eSr values show a bimodal distribution indicating more than one source of strontium. This variation may reflect the mixing of magmatic and non-magmatic fluids during hydrothermal mineralization as demonstrated by the fluid inclusion and stable isotope data for the orefield.

H E L I U M A N D C A R B O N I S O T O P E R A T I O S IN C R U S T A L F L U I D S T H E E A S T E R N R I F T S Y S T E M IN K E N Y A

DARLING> W G , British Geological Survey, Wallingford, 0X10 8BB, UK, and GRIESSHABER, E, Department of Earth Sciences, University of Cambridge, Cambridge CB2 3EQ, UK Isotopic measurements of hydrothermal fluids in the Kenya rift reveal an upper mantle helium component reaching a maximum near the northern end. Peak ^He/^He values range from 5.5 to 8.0 R^, though nearsurface effects tend to lower values away from the immediate vicinity of hydrothermal plumes. The peak values indicate that local faulting above magma chambers allows a greater flux of mantle He to the surface than the larger rift-margin faults. Although the Kenya rift is believed to be structurally unconnected with the Ethiopian rift, implying a separate underlying mantle plume, there is no sign of a hot spot similar to the Ethiopian one where ^He/^He values reach ~15 R a . Possibly lateral compressive forces have playecT a proportionately greater role in the updoming associated with the Kenya rift. Isotopic values for CO2 lie in a restricted range of -3.5 ± \%o S^^C for most of the hydrothermal sources, probably indicating a largely crustal origin. However, evidence from the most active hydrothermal sites that becomes more negative with increasing ^He/^He suggests an increasing mantle contribution. Carbon isotopes in CH4 also occupy a relatively small range of -23 ± 5%. S ^ h , This precludes a biogenic origin, and probably also a mantle contribution. High-temperature synthesis of CH4 from CO2 and Ho cannot be discounted on isotopic grounds, but hydrotnermal systems of widely differing inferred maximum temperature give comparable <5^^C(CH4) results, suggesting that thermogenic reactions at relatively low temperature are responsible.

ABSTRACTS

THE USE OF STABLE ISOTOPES TO DETERMINE THE TIMING AND ORIGIN OF A BIF-HOSTED CU-AU DEPOSIT, NORTHWEST AUSTRALIA. GARRY DAVIDSON. Centre for Ore Deposit & Exploration Studies, Hobart, Tasmania, Australia, 7005. Starra and Trough Tank are stratabound Cu-Au deposits hosted by stratiform magnetite-hematite iron formation in the Mount Isa Eastern Succession, Australia. They are part of the rift-related Middle Proterozoic Mary Kathleen Group, which was deposited in a shallow to evaporitic setting, and subsequently deformed and metamorphosed. The location of Starra (consisting of 4 geographically separate orebodies), on the margin of a major greenschist facies shear zone (ore metamorphism estimated at 500 550° C, 3.5 - 4.5 kbar) has led to a more complex structural history for the ores than for the surrounding rocks. Both structural/metamorphic and syngenetic ore genesis models have been proposed for the ores. Trough Tank is far less strained, but was metamorphosed to upper amphibolite grade (680 - 720° C, -A kbar). It is in many respects similar to Starra. Carbon isotopes were obtained from carbonate in mineralised ironstone, hangingwall metasediment matrix, and syn to postD2(LP) veins and shears in the metasediments. Calcite-hosted fluid inclusions in the latter homogenised between means of 114° and 200° C for individual samples, with salinities of 4.0 to 23.1 wt. % equiv. NaCl. Sediment matrix varied between -5.3 and +0.0%o, overlapping with the vein range of -5.1 to 1.4%c.Ironstone carbonate (recrystallised) varied between Sl^C = -15.5 and -3.8%c, requiring a source component of oxidised organic carbon. Fluid-rock reactions are interpreted to have altered the sediment matrix S^^C value towards the vein-fluid composition (reflecting mixing between deep metamorphic and shallow meteoric waters within the Starra Shear). In contrast, in ironstone carbonates at Starra and Trough Tank do not converge with the vein-fluid compositions, and must predate the veining/shearing deformation. Carbonates from the two deposits form a single unusual trend of increasing with decreasing S^^c, which suggests a S^^O-enriched connate water contribution to the the ore fluid. This trend can be regarded as a "target signature" in the region for ores of the Starra type. ARC MAGMATISM AND CRUSTAL GROWTH: CONSTRAINTS FROM THE CENTRAL ANDES DAVIDSON. J.P.. Dept Earth & Space Sci., UCLA. Los Angeles, USA, WORNER, G.W., Inst. Geowissenschaften, Univ. Mainz, FRG, MOORBATH, S., Dept Eanh Sci., Univ. Oxford, UK, and HARMON, R.S., NERC Isotope Geoscience Lab, Keyworth, UK An assessment of the role of arc magmatism in formation of the continental crust is critically dependent on evaluation of the relative roles of mantle-derived flux versus intra-crustal recycling. In island arcs the mantle-crust flux is dominantly basaltic and therefore cannot represent one stage fractionation of tonalitic crust from the mantle. In the Central Andes, volcanic rocks more closely approximate andesite/tonalite, and may accordingly be interpreted in terms of addition of new continental crust, if derived directly from the mantle. The Neogene-Recent volcanic front of the Central Andes (22°17°S) forms a single arc segment underlain by unusually thick (6070km) continental crust. Stratocones from the front have erupted a range in compositions from high-K calc-alkaline basalt (51.4% Si02) to rhyolite (74.2 %Si02), although intermediate compositions predominate. Along strike trends in isotopic composition are observed among the most mafic lavas erupted: ^'^Sr/^^Sr increases northward from c. 0.7058 at 22°S to c. 0.706 at IQ^S, and remains essentially constant to 17°S. 206pt)/2a4pb ratios decrease sharply from 18.7 at 22oS to 17.8 north of 20"S, probably reflecting the change in Pb isotopic composition of the underlying basement which varies in age from Paleozoic in the south to Precambrian in the north. This may in turn indicate the position of the southwestern margin of the Brazilian shield, pointing to conn-ol on magma isotope compositions by magma crust interaction. Primary melts in the Central Andes are derived from an asthenospheric mantle wedge, and are similar in chemical character to primitive magmatic arc rocks woridwide. The parental magmas in the central Andes are distinguished by relatively high ^''Sr/^^Sr, and low i43Nd/i44isTd^ Ba/Ce and Ce/Ta compared with primitive arc rocks. They are therefore the product of extensive interaction with the continental crust, which has thickened considerably over the past 15Ma, and do not represent addition of tonalitic crust directly from the mantle.

ABSTRACTS

IC0G7

THE CRYSTALLIZATION AGES OF MINERALS FROM A SIUCIC MAGMA CHAMBER DAVIES. G.R . ^ L L I D A Y , AJST., Department of Geological Sciences,University of Michigan, Ann Arbor, MI 48109-1063, USA and MAHOOD, G A . , Department of Geology, Stanford University, Stanford, CA 94305-2115, USA. High-silica rhyolite lavas of Glass Mountain, California, are characterised by among the most extreme Rb/Sr ratios measured to date (up to 4000). Despite their essentially aphyric nature the glasses record evidence of extensive feldspar fractionation with Sr contents between 2 and 0.06 ppm. Biotite and inclusion bearing quartz have Rb/Sr ratios up to 7 times higher than their host glasses. Coupled with precise ratios we have the potential to resolve age differences on the order of 500yr. Glass- whole rocks define a series of geographically controlled isochrons (e.g. 2.08± 0.04 Ma and 1.91 ± 0.02 Ma ) that are significantly older than the eruption ages, which implies that the rhyolite magmas had a substantial magma chamber residence prior to eruption, up to 0.7 My. The sparse mineral assemblages display significant Sr isotope disequilibrium but with no detectable Pb or Nd disequilibrium apparently ruling out a xenocryst origin for the minerals. Rb-Sr Kfeldspar-glass (G), plagioclase^ and quartz-G ages are close to their respective whole rock isochron ages, i.e. circa 2.0 Ma. In contrast biotite-G and biotite-feldspar ages are up to 0.3 My younger. The biotite ages are still substantially older than the eruption ages. There are three possible interpretations of these data; 1) the mineral-glass ages represent crystallization ages, 2) the biotite is in fact recording the eruption age and the Ar ages on the glasses are incorrect (still impling that the feldspars record crystallization ages), 3) the biotite crystallized with the quartz and feldspar but subsequently underwent partial Sr equilibration with the host magma. All three interpretaions imply an extensive residence time for the minerals in a magma without undergoing Sr equilibration and suggest that Sr diffusion in highly silicic magmas is significantly slower than predicted from experimental studies. ^^ Ar/ '^^Ar studies of biotite and K-feldspar are currently underway to determine the rhyolite eruption ages and consequently unequivically establish the time over which the minerals have remained in disequilibrium with their host magma.

MANTLE PLUMES, MANTLE STIRRING AND HOTSPOT CHEMISTRY DAVIES, Geoffrey F., Research School of Earth Sciences, Australian National University, Canberra, ACT 2601, Australia Recent advances in understanding plume dynamics allow the role of mantle plumes, the presumed cause of volcanic hotspots and source of the hotpsots* oceanic island basalts (OIBs), to be specified in more d e t a i l than h i t h e r t o . Their role in sampling a mantle with single-layer convection and an increase in viscosity with depth by 2 or 3 orders of magnitude i s considered here. Plumes would sample thin tabular regions at the bottom of the mantle. This region would be expected in the assumed mantle model to have properties appropriate to OIB sources, including greater heterogeneity, greater age and less depletion of incompatible- elements than the shallow mantle sampled by ocean ridges. All proposed sources of recycled material, including oceanic crust, oceanic sediments and continental lithosphere, can be accommodated in t h i s model. However i t i s noted that clear evidence for a source with primitive r e f r a c t o r y element r a t i o s i s s t i l l lacking, as i s undisputed evidence for differences between MORB source and plume source '•®Ar/'®Ar and r a t i o s . "^^Ar/'^Ar r a t i o s do not exclude nearly complete early degassing of the mantle. High 'He/'*He r a t i o s in some plumes might come from some remaining l e s s degassed mantle or'from the core.

25 EVAPORATION OF ROCKS AND MINERALS AND IMPLICATIONS FOR EARLY SOLAR SYSTEM PROCESSES DAVIS. Andrew M.*. HASfflMOTO, Akihikot, CLAYTON, Robert N.*^ and MAYEDA, Toshiko K.*, * Enrico Fermi Institute, ^Department of the Geophysical Sciences, ^Department of Chemistry, University of Chicago, Chicago, IL 60637 USA, ^HarvardSmithsonian Center for Astrophysics, Cambridge, MA 02138 USA Calcium-aluminum-rich refractory inclusions (CAI's) in carbonaceous chondrites can contain O, Mg, Si, Ca, Ti and Cr whose isotopic compositions are significantly mass fractionated, up to a few percent amu"^ from normal (=terrestrial) values. Isotopic fractionation between gaseous and condensed phases is likely to be the source of these effects. Because of the large massfractionationeffects observed, it has been proposed that they are kinetic isotope effects produced by nonequilibrium evaporation and/or condensation processes that occurred in the primordial solar nebula. To verify this idea, we have partially evaporated synthetic forsterite and silica in vacuum for various durations and different temperatures. Evaporation of molten forsterite led to substantial enrichments in the heavy isotopes of O, Mg and Si, whereas evaporation of solid forsterite and molten silica led to neglible massfractionationeffects in the riesidue. In order for large mass fractionation effects to be produced in the residue by Rayleigh fractionation, the heavy isotopes which concentrate on the surface layer of the residue as a result of preferred evaporation of light isotopes must migrate promptly to the interior. Molten forsterite has high diffusion rates and low viscosity, satisfying this condition; diffusion rates are lower and viscosity is higher in solid forsterite and molten silica, accounting for the small fractionations seen in these residues. These results imply that CAFs that show large heavy isotope enrichments must have been molten during the evaporation episode. Evaporation experiments are being extended to the other elements for which mass fractionation effects have been observed: Ca, Ti and Cr.

DISTINGUISHING PLUTONIC AND VOLCANOGENIC MINERALIZATION IN THE LACHLAN FOLD BELT OF EASTERN AUSTRALIA USING Pb ISOTOPES DEAN, J . A . and CARR, G.R. CSIRO D i v i s i o n of E x p l o r a t i o n G e o s c i e n c e , N o r t h Ryde, NSW, 2113, A u s t r a l i a . I n t h e e a s t e r n t e c t o n i c domains of the L a c h l a n Fold B e l t (LFB) of NSW and i n the C h a r t e r s Towers P r o v i n c e of n o r t h eastern Q u e e n s l a n d , Au and b a s e m e t a l mineralization may be a s s o c i a t e d w i t h O r d o v i c i a n t o S i l u r i a n v o l c a n o g e n i c e v e n t s o r t o Devonian g r a n i t o i d s which i n t r u d e t h e s e q u e n c e s . Thus granite d e r i v e d v e i n m i n e r a l i z a t i o n may be e n c o u n t e r e d w i t h i n v o l c a n i c s e q u e n c e s and i s not r e a d i l y d i s t i n g u i s h a b l e from p o s s i b l e stockwork v o l c a n o g e n i c m i n e r a l i z a t i o n . The Pb i s o t o p e i n i t i a l r a t i o s of volcanog e n i c m i n e r a l i z a t i o n t h r o u g h o u t t h e LFB show remarkable homogeneity d e s p i t e t h e wide g e o g r a p h i c d i s t r i b u t i o n of t h e d e p o s i t s and a r a n g e i n a g e s of a t l e a s t 50 Ma. The Pb i s o t o p e r a t i o s of unambiguously g r a n i t e - d e r i v e d v e i n and s k a r n m i n e r a l i z a t i o n show l i m i t e d v a r i a t i o n , and i n g e n e r a l ^ a r e more r a d i o g e n i c t h a n t h e v o l c a n o g e n i c signat u r e . Where s u c h m i n e r a l i z a t i o n h a s n o t b e e n f o u n d , t h e e x p e c t e d i n i t i a l r a t i o s of minera l i z a t i o n associated with p a r t i c u l a r granite p l u t o n s can b e e s t i m a t e d by t h e a n a l y s i s of K - f e l d s p a r s w i t h low U/Pb r a t i o s . To a c h i e v e t h e b e s t e s t i m a t e w i t h o u t t h e need t o c o r r e c t for radiogenic addition, multiple sequential d i s s o l u t i o n s h a v e been u n d e r t a k e n . The r e sults are similar to those for the Pb-rich v e i n and s k a r n m i n e r a l i z a t i o n . Thus i n most a r e a s i t is possible to a s s e s s t h e m e t a l l o g e n i c a s s o c i a t i o n of e x p l o r a t i o n s a m p l e s b a s e d on Pb i s o t o p e s .

ICOG 7

ABSTRACTS

REVISED CHRONOSTRATIGRAPHY OF THE NGORORA FORMATION AND CALIBRATION OF THE PALEOMAGNETIC REVERSAL TIME SCALE FOR THE MIDDLE MIOCENE DEINO, A., and Drake, R., Geochronology Center of the Institute of Human Origins, 2453 Ridge Rd., Berkeley CA 94709 USA; TAUXE, L., Scripps Institute of Oceanography, La Jolla, CA 92093 USA; MONAGHAN, M., Department of Geophysical Sciences, University of Chicago, Chicago, IL 60637 USA Single-crystal laser-fusion ^^Arft^Ai dating results have yielded precise eruption ages for tephra from volcaniclastic rocks of the middle Miocene Ngorora Formation, Kenya Rift Valley. The type section near Kabarsero ranges in age from -13.1 Ma at the base to 10.5 Ma near the top. Extrapolation of the sedimentation rate inferred from the dated units suggests that the youngest Ngorora strata at Kabarsero may be ~8-9 Ma. These upper strata would then be contemporaneous with distal outcrops attributed to the Ngorora Formation at Ngeringerowa that range in age from approximately 8.9 to 9.5 Ma. We have also obtained chrono-, litho-, and paleomagnetic stratigraphies for Ngorora Formation sediments at Bartabwa that correspond in age to the lower half of the type section. A composite paleomagnetic stratigraphy for the period 10.513 Ma has been derived from these sections. Interpolated ages for magnetozone boundaries are about 1.4% older than corresponding boundaries in the geomagnetic reversal time scale of Berggren et al. (1985). Much of the discrepancy can be eliminated if an inferred change in sea-floor spreading rate occurred at 13 Ma or earlier, rather than at 10.42 Ma as they suggested Sedimentation rates within the Kabarsero and Bartabwa sections are highest toward the base of the sections where they overlie the voluminous, basin-filling Tiim Phonolites. These initial rapid sedimentation rates are recorded in sediments only 0.1 to 0.3 Ma older than the underlying lava flows. These relationships suggest that the basin in which the Ngorora Formation was deposited was not predominandy a rift valley created by extensional tectonics, but instead was controlled by subsidence feauire resulting from the removal of large volumes of magma from the lithosphere.

D/H ANALYSIS OF AMPHIBOLES FROM MANTLE LHERZOLITE XENOLITHS BY ION PROBE DELOULE. Etienne, ALBAREDE, Francis & SHEPPARD, Simon M.F., Centre de Recherches Petrographiques et Geochimiques (C.R.P.G.-C.N.R.S.) & Ecole Nationale Superieure de Geologic, BP 20, 54501 Vandoeuvre Cedex, France. An analytical procedure has been established to measure in situ D/H ratios in hydroxylated minerals on a few tens jim scale by ion probe, with a precision of ±10%o (Deloule et al, 1989, EOS,70, 489) .This technique has been applied to determine the D/H ratio of amphiboles occurring as rare disseminated minerals in peridotites. These amphiboles include pargasite from the Lherz peridotite, Pyrenees (France), and kaersutite and pargasite in ultramafic xenoliths and megacrysts from the Massif Central (France), Nunivak Island (Alaska) and Salt Lake Crater (Hawaii). For the Massif Central and Salt Lake Crater, 9Dsmow variations up to 50%o were measured both on the scale of single crystals (^mm) and among crystals of the same sample. Chemical variations are minor and not correlated with dD. Combining the diffusion data for hydrogen in amphiboles with these results implies that (1) the dD of the infiltrating fluid was heterogeneous, and (2) the time interval between exchange and volcanism was very short (<few months). The 3D measured in Lherz (-92%o and -65%o) and Nunivak Island (-93%c to -69%o) are in the accepted range for uncontaminated upper mantie minerals. The higher 3D (-59%o to -28%o) from the Massif Central xenoliths may involve contamination of the subcontinental lithosphere by subduction of seawater altered oceanic crust. The very light 3D (-120%o and -100%o) observed for hawaiian xenoliths could reflect the influence of deep chemical plume. The 3D heterogeneity observed for the latter two localities probably results from different mixing processes, rather than from a fractionation process.

POST-ARCHAEAN TECTONIC EVOLUTION OF THE WEST MARGIN OF THE YILGARN CRATON

THE ANORTHOSITE-BEARING ANOROGENIC RING COMPLEXES OF AIR (NIGER) : SR, ND, PB ISOTOPE GEOCHEMISTRY DEMAIFFE D , , Lab. Geol.-Petrol.-Geochr. , Univ. Libre Bruxelles, 50, av. Roosevelt, 1050 BrusselsBelgium ; MOREAU C . , Dept.Geol, Univ. ClermontFerrand, France ; BROWN W.L., CRPG, Nancy, France ; WEIS D. (same as DEMAIFFE D . )

J. R. DELAETER, Curtin University of Technology, Kent Street, Bentley, W. A., 6012, AustraUa, and W. G. LIBBY. Geological Survey of W. A., 100 Plain Street, Perth, W. A., 6000, Australia Biotite Rb-Sr dating indicates that most of the 1000 Km extent of the western margin of the Yilgam Craton of Western Australia, dated by a variety of methods as Archaean, has been tectonically active during post-Archaean time. In contrast, most biotite dates in the main body of the craton are Archaean, averaging about 2500 Ma. Wrapping around the northwestern comer of the Yilgam Craton is a broad 400 Km arc yielding biotite dates at about 1600 Ma. These dates correlate with major 1600 Ma tectonic activity within the adjacent Gascoyne Province, implying that the 1600 Ma Gascoyne mobile belt extended southwestward from the Gascoyne Province along the west side of the Yilgam Craton. To the south, a triangular area extending for 400 Km along the west margin of the Yilgam Craton and continuing into the adjacent Proterozoic Albany-Fraser Province has been biotitedated at 450 to 500 Ma, suggesting 5-10 Km of early Paleozoic uplift and providing an upper limit for emplacement of the Albany-Fraser Province.

Plagioclasic cumulates (anorthosite, leucogabbro, troctolite, .. .) are common in the anorogenic alkaline ring complexes of Air (Niger) ; they are associated with granites (sometimes peralkaline) and syenites occurring as ring-dykes or central plugs. Several types of intrusion;? have been recognized in the field : small (< 2 km) funnel shaped types (Taguei, Abontorok), a large massif-type body (Ofoud), a stratiform intrusion (Bous) and a large (65 km) ring-dyke (Meugueur-Meugueur). They intrude a 30 km thick gneissic basement of Pan-African age. Geochemically the mafic and felsic rocks appear to have complementary REE patterns. Plagioclasic cumulates have low REE contents (Laj^ = 2 . 5 - 1 0 ) , high (La/Yb)j^ ratios (3.8-6) and large positive Eu anomalies (Eu/Eu* = 1 . 8 - 3 . 1 ) while acid rocks have high REE (La., = 90-400) with negative Eu anomalies (Eu/Eu* = 0.7-^6.21). g^ g^ The plagioclase-rich rocks have low ( Sr/ Sr). ratios (0.7030 to 0.7065) with slightly positive (+0.5 to +2.6) or negative (down to -2) GJ,, values. Each intrusion has a distinct isotopic signature implying either a distinct source (for example a slightly depleted OIB-type source for Bous : g , +2.6 ; 0.7030) or different discrete degrees of contamination of a common magma by continental materials. The granites and syenites are not comagmatic with the basic rocks : they have typical crustal isotopic signature (®^Sr/®^Sr)i > 0.708 ; e Nd = -7 to - 1 2 . 5 ) .

ABSTRACTS

IC0G7

Isotopic changes of the Total Dissolved Inorganic Carbon, as tracers of liquid-solid interactions in the unsaturated zone Dever L. & Pontes J.Ch. Laboratoirc d'Hydrologie et de Geochimie Isotopique,Universite Paris-Sud, bat. 504, 91405 ORSAY France The water percolates througth the unsaturated zone of the chalk (20 m depth) with a rate close to 0.75 according to a piston flow model. The hydrochemical facies is calcium bicarbonate. From 0,5 m depth the water is saturated with respect to the calcite. The carbon-13 of the Total Dissolved Inorganic Carbon (TDIC) increases downwards from 2m (-12.2 %o) to 10m (-8.2 %o) depth. In the same range of depth the radiocarbon contents of the TDIC decreases from 117 to 85 p.m.c. Carbon isotope mass balance indicates a progressive mixing with depth between the carbon of the liquid phase and that of the solid phase. The mixing results from calcite dissolution-precipitation processes which occur even at saturation and deals with 500 nm of the exchange layer at the external part of the coccolith according to the stable isotopic mass balance.Such exchange involves from 3 % (2 m) to 40 % (11 m) of the TDIC under closed system condition. A comparison of the calculated and measured dissolved radiocarbon profiles indicate that the subsurface of the solid phase is not a radiocarbon sink. The apparent dissolution rate which is close to 10'^' mmole takes into account the precipitation of "modem carbon" and the dissolution of "dead" aM "modern carbon" previously precipitated under non-steady state conditions. This is confirmed by the radiocarbon contents (from 0.5% to 2.5%) which are measured on the total solid phases. This non-steady state at the liquid-solid interface is the result of modifications of the carbon dioxide profil generated in the unsaturated zone through recent forestation.

Tectonic and eustatic effects in the Gulf of Corinth investigated through high precision ^ ^ ^ ages and ^^Sr/^^Sr ratios of corals. A.Dia. A.S. Cohen and R.K. O'Nions Department of Earth Sciences, University of Cambridge, Downing Street, Cambridge CB2 3EQ (U.K.) Mass spectrometric ^^^U-^^'^U-^^^Th ages of coral aragonite combined with ^^Sr/^^Sr ratios for the same materials provide the basis for an investigation of tectonic uplift rates in relationship to glacio-eustatic changes and seawater evolution over the last 350ka. Such techniques have been applied to corals from raised beaches on the south side of the gulf of Corinth, Greece. The Corinth basin is a continental fault-bound basin formed by crustal extension, northeast of subduction along the Hellenic Trench, and proximal to the Aegean volcanic arc. Coral samples from terraces on the uplifted south side of the Gulf of Corinth, confirm estimates for an uplift rate over the last 300 ka of ^ 0.3 m ka Sr isotope compositions of coral aragonite have been determined to very high precision (< lOppm in ^^Sr/^^Sr at 95% confidence level), and resolve an increase of lOOppm over a period of 125 ka. This rate of increase compares with an average increase of 220ppm per 100 ka in marine Sr through much of the Cenozoic. Variations in ^^Sr/^^Sr are considered in the light of climate (glacio-eustatic), and tectonic controls on weathering and continental run-off.

27 EVIDENCE FOR AN EARLY ARCHEAN CRATONIC NUCLEUS IN THE SUPERIOR PROVINCE OF ONTARIO Dickin, A.P., McNutt, R.H., Mueller, E., Department of Geology, McMaster University, Hamilton, Ontario L8S 4M1 & Beakhouse, G., Ontario Geological Survey, 77 Grenville Street, Toronto, Ontario M7A 1W4. Tonalitic gneisses from the length of the Winnipeg River belt were analysed by the Sm/Nd method. Results below show epsilon Nd values as low as -4.3 at 3.17 Ga, a U/Pb zircon age determined from this area by Corfu (1988) and interpreted as the time of formation of the Cedar Lake tonalite. However, the range of epsilon Nd at 3.17 Ga is evidence for a pre3.2 Ga crustal history for the precursors of the Winnipeg River gneisses.

Pb/Pb ages were determined on zircons from some samples by the direct Pb evaporation technique (Kober, 1986). A wide range of 207/206 ages up to a maximum of 3.38 Ga was measured, but clusters of ages are found from 2.7 to 3.2 Ga. Taken together, the evidence suggests the presence of an Early Archean cratonic nucleus in the Winnipeg River belt of the Superior Province that was severely reworked during the Middle and Late Archean.

A STUDY ON SILICON ISOTOPE COMPOSITIONS OF SOME ROCKS AND MINERALS D I N G Tiping, W A N Defang, JIANG Shaoyong and LIU Zijian, Institute of Mineral Deposits, CAGS, Beijing 100037, P.R.China.

More than 250 samples of rocks and minerals were analysed for their Si isotope compositions. The results were reported as d^Si values relative to NBS-28, and the precision obtained was ± 0.1(%o). The total range of <5^°Si variations observed is 5.3(%o), ranging from - 1 . 9 to 3-4 (%o). The <5^°Si values of meteorites, tiktites and igneous rocks are similar to those reported by Douthitt(1982). However, some significant observations have been made on variation among different siliceous rocks, quartz veins and clay minerals and on isotope fractionation between coexisting minerals in igneous and metamorphic rocks. The value of kaolinite may be used in distinguishing their origin. Our S^^Si results of weathering kaolinite vary from - 1 . 0 to - 1 . 9 ( % o ) . In contrast, the S^^Si values of hydro thermal kaolinite are from - 0 . 1 to 0 . 1 ( % o ) . The sedimentary kaolinite have ^^Si values of -0.1

1.2(%o).

The values of metamorphosed siliceous rocks can also be used in determing their'original rocks. As we know, the values of quartz sand and sandstone are all close to 0. In contrast, a sample of diatom earth shows a negative d^^Si value of-l.l(%o). The deep sea radiolarian rocks have S^^S'i values o f - 0 . 3 0.6(%o), but the shallow sea radiolarian rocks have values of 0.2—0.8(%o). The siliceous dolomite and siliceous stromatolite have d^^Si values of 2.4 A preliminary survey on silicon isotope compositions of more than 10 ore deposits shows that their values of vein quartz may indicate their origin of silicon. The S^^Si values of coexisting minerals in igneous and metamorphic rocks show regular variation, i.e. the values decrease from quartz to feldspar, muscovite, biotite and garnet. This trend may provide a possibility of using Si isotope fractionation to study the forming temperature of rocks.

GEOCHEMISTRY AND GEOCPIRONOLOGY OF MAHC ROCKS OF THE OLONDO GREENSTONE BELT, ALDAN SHIELD, SIBERIA, U.S.S.R. N. DOBRETSOV. Institute of Geology and Geophysics, Siberian Branch, U.S.S.R. Academy and Sciences, Novosibirsk, U.S.S.R. The -3000 Ma Olondo greenstone belt is located the Olekma granite-greenstone block, which forms the western part of the Aldan Shield, Siberia, The Olondo greenstone belt contains volcanic and volcano-sedimentaiy rocks with komatiitic, tholeiitic, andesitic and rhyolitic compositions. Intrusive rocks are represented by abundant ultrabasic (dunites and peridotites) and basic (differentiated gabbro) bodies. The gabbros are compositionally similar to the volcanic rocks of tholeiitic composition. All of these rocks are deformed and metamorphosed, and commonly have amphibolite facies assemblages. However, pillow structures are locally preserved in mafic rocks. Cross-cutting layered bodies and dykes of picritic affinity have yielded an age of-2100 Ma. The mafic volcanic rocks show chondrite normalised REE pattems ranging from LREE depleted to moderately LREE enriched relative to HREE. The picrites are compositionally distinct from the older mafic suites of the belt The geochemical signatures of these rocks are used to constrain late Archaean and early Proterozoic upper mantle evolution in the Aldan region.

KINETICS AND THERMOCHRONOKETRY MARTIN H. DODSON, Department of Earth Sciences, University of Leeds, Leeds LS2 9JT, England. Kinetic approaches to thermal history depend upon a constantly shifting interplay of theory and experiment with natural observations; theory supplieJs the conceptual foundation, and predicts some simple idealised relationships amongst closure (or opening) temperatures, kinetic parameters, and rates of temperature change. Laboratory data on diffusion is much harder to acquire than on fission track annealing; the shortage of such data remains a serious problem for kinetic thermometry. Natural 'experiments' in contact zones help to fill the gap, but rely heavily on theoretical recpnstrucions of thermal histories. Regional patterns of ages (e.g. the Alps) provide opportunities to test interpretations against independent geological constraints. The increasing availability of high spatial resolution isotopic analyses within mineral grains, and the need to understand Ar-^^Ar step-heating curves, lends point to theoretical studies of concentration profiles. Work of this kind should clarify the vexed question of diffusion geometry what exactly are the size and shape of the regions within mineral grains to which one wishes to apply diffusion theory? Gaps in the theory include diffusional exchange with a finite reservoir, the effects of extreme anisotropy, and systematic methods of integrating thermal modelling with thermochronometry.

ICOG 7

ABSTRACTS

Pb-ISOTOPE COMPOSITION OF GALENA FROM MALAYSIA DOE, Bruce R., U.S. Geological Survey, 923 National Center, Reston, VA 22092, USA Seven new samples of galena from Malaysia have Pb isotope ratios of 18.609 to 19.057 for X or Pb(206)/Pb(204), 15.704 to 15.765 for Y or Pb(207)/Pb(204), and 38.950 to 39.385 for Z or Pb(208)/Pb(204). Both the East Coast (ECP) and West Coast (WCP) igneous provinces of Bignell and Snelling (1977) are represented. The range in isotope ratios is surprisingly small in that the WCP plutons are Late Triassic in age, are exclusively S-type, and have a concealed basement age of 1500-1700 Ma, whereas the ECP plutons are Permo-Triassic, are both I- and Stype, and have a concealed basement age of 1100-1400 Ma (Liew and McCulloch, 1985). Although all their values of (Sr-87/Sr-86) ^^ are at least somewhat high for mantle ratios (>0.705), the WCP has some of the highest values known (up to 0.7512). Probably the most distinguishing feature of the Pb isotopes is the high value of Y, which has a distinct continental upper-crust nature. Because the Stacey-Kramers model Pb ages range from 260 Ma (close to the age of the deposits) to -450 Ma, there is no indication of cratonized crust, (values of U/Pb and" Th/Pb reduced by high-grade metamorphism in the Precambrian). Similar Pb-isotope ratios are known from Belitung Island, of Indonesia (Jones et al., 1977), which is included in an extension of the Malaysian tin belt to the SE by Hosking (1970), but not from adjacent Sumatra. Pb isotope ratios similar to Malaysia are also known in Java. The other major tin province, Bolivia, also has many high values of Y (Tilton et al., 1981).

ISOTOPE RATIO MONITORING MASS SPECTROMETRY: A POSSIBLE APPROACH TO A STT^LE ISOTOPE MICROPROBE C.B. Douthitt, Finnigan MAT, San Jose, CA USA High precision measurement of stable isotope ratios with microprobe resolution is bein'g pursued by combining lasers with magnetic sector MS. Most approaches make use of the McKinney (1950) dual viscous flow inlet system which was designed to circumvent problems which at that time precluded making absolute measurements with the .l%oprecision required to resolve variations in natural abundcmces. The MS has been improved to the point where, for samples ^ 1 bar-ywl, the precision is almost entirely limited by counting statistics. However, the requirements of the inlet system for viscous flow and sample/standard comparison require samples two to three orders of magnitude larger than are required from counting statistics for precisions at the .l%r level. An entirely new inlet system has been developed in order to isotopically characterize samples as they elute from a GC; the technique is described as isotope ratio monitoring MS. Samples are entrained in a stream of He, passed through a GC into a combustion chamber, and the CO2 pulses introduced directly into the MS ion source. Ncinomole samples of CO2 can be analyzed in seconds to precision of .1%«, approaching the limits imposed by counting statistics. The lower limit on sample size is dictated by the signal-tobackground ratio; femtogram quantities of the metabolites of heavily ^^c-iabelled compounds have been detected. Isotope ratio monitoring MS allows the analysis of samples of a size and with a speed that are commensurate with proposed laser ablation techniques, and with a precision appropriate for expected variations; with suitable modifications, it appears well-suited to laser ablation/fluorination of silicates, oxides and sulfides, and to stepped combustion of meteoritic carbon.

ABSTRACTS

IC0G7

RADIOISOTOPIC AGES OF THE NORTH PATAGONIAN BATHOLITH AND THEIR RELATION TO THE LIQUINEOFQUI FAULT ZONE, CHILE (40°-42® S. LAT.) Drake, R., Geochronology Center, Institute of Human Origins, 2453 Ridge Rd., Berkeley CA 94709 USA; Herve, F. and Munizaga, F., Depto. de Geologia y Geofisica, Univ. de Chile, Santiago, CHILE, Pankhurst, R.J., NERC Isotope Geosciences Lab., Keyworth, Nottingham, UNITED KINGDOM. The W. margin of the N. Patagonian Batholith consists of several different ages of plutons (gabbros<granodiorite>granites)based on K-Ar, single-crystal laser-fusion ^^Aifl^Ai, and Rb-Sr WR dating. The Liquine-Ofqui Fault Zone (LOFZ) cuts N-S through this batholith which is exposed by extensive Pleistocene glaciation.. Plutons W of the LOFZ exhibit Miocene ages: 12-10 Ma (K-Ar) and 16+/-lMa (Rb-Sr), Lat.40°30' to 41^30' S and 20 Ma (K-Ar) near Chaiten, 43° S. In contrast, plutons E of the LOFZ yield midCretaceous ages: 100-115 Ma (Ar/Ar) along Estero Reloncavi, 41.5° S. and 110-115 Ma (Rb-Sr) at Futaleufu and Palena, 43° 20' to 43°40' S. Mio-Pliocene ages: 10-3 Ma (Ar/Ar) also occur on Isla Llancahue, 42° 10' S.. These data imply that the locus of subduction related arc magmatism at these latitudes has remained nearly stationary since the Jurassic in contrast to the 150-2(X)km. eastward migration seen further north along the W margin of SA. during the same time period. The diversity and distribution of ages on each side of the LOFZ needs to be better defined but may reflect a late T e r t i ^ tectonic activity of displacement and/or block rotations accompanied by continued intrusion of plutons.

A THERMAL HISTORY FRAMEWORK FOR THE EROMANGA BASIN: AN INTEGRATED APPROACH USING AFTA (Apatite Fission Track Analysis) AND VITRINITE REFLECTANCE IAN R. BUDDY, Geotrack International, PO Box 4210, Melbourne University, Victoria, 3052, Australia The thermal histories of a number of hydrocarbon exploration wells from the Mesozoic Eromanga Basin of South Australia have been assessed using Apatite Fission Track Analysis (AFTA). Modelling of apatite fission track parameters using a laboratory based description of the annealing kinetics (Laslett et al, 1987: Chem. GeoL, Isot. Geosci, 65, 1-13) indicates that high presentday geothermal gradients (up to ~7()°C/km) in parts of the basin have developed within the last 5 Ma. This suppons interpretations derived from coalification studies (Kanstler et al., 1978: APEA J. 18,143-156) that temperatures have increased recently. AFTA data shows that much of the sediment in the Eromanga Basin sequence was derived from contemporaneous volcanogenic sources. The Jurassic Birkhead Formation in South Australia, like its lateral equivalent to the east, the Walloon Coal Measures, was derived largely from volcanogenic sources. The presence of contemporaneous volcanogenic detritus provides an explanation for the nature of the boundaries between the quartzose, basementderived reservoirs like the Hutton Fm and Namur mbr and the associated "dirty" (volcanogenic) reservoirs of the Birkhead Fm and Murta mbr. The section above the E. Cretaceous Murta mbr, including the Winton Fm, is dominated by volcanogenic detritus. The effect of chlorine in apatite on the annealing response, first proposed from work in the Otway Basin, has been observed in volcanogenic samples from Eromanga Basin sequence, confirming the generality of the concept. Further, it is suggested that the CI content of detrital apatite provides good information on the nature of a sediment's provenance. Sediments derived from volcanic terrains contain apatites with a wide range of CI contents, while "granitic" basement terrains typically provide a population of apatite with a much more restricted, F-rich, compositional range. Integration of AFTA with vitrinite reflectance data has allowed the construction of detailed thermal and hydrocarbon maturation histories for the Eromanga Basin, as well as providing important information on the regional controls on reservoir quality.

THE GEOCHRONOLOGY OF THE AKIA TERRANE, SOUTHERN WEST GREENLAND DUKE. M.J.M. and BAADSGAARD H., Department of Geology, University of Alberta, Edmonton, Alberta, Canada, T6G 2E3, NUTMAN, A.P. and COMPSTON W., Research School of Earth Sciences, Australian National University, Canberra, ACT 2601, Australia, McGREGOR, V.R., Atanmiik, 3912 Sukkertoppen, Greenland, Denmark. The Archaean high-grade gneiss complex of the Nuuk region of southern West Greenland was divided into four tectono-stratigraphic terranes by Friend et al. (1988). One of these terranes, the Akia terrane, is composed largely of amphibolite grade tonalitic-dioritic type-Nuk gneisses (3070-2940 Ma) and, across Godthabsfjord, granulite facies quartzofeldspathic gneisses of the Nordlandet peninsula. Although the relationship of the type-Nuk gneisses and granulites of Nordlandet is uncertain, the granulites have been considered approximately the same age as the type-Nuk gneisses. Ion microprobe U-Pb ages using the SHRIMP at ANU have been determined for single zircons separated from the type-Nuk gneisses and Nordlandet granulites. Ages such as 3019±4 Ma (2a) have been measured for type-Nuk gneisses, while cores and rims of zircons from the granulites give 3192±6 Ma (2a), and approximately 3020 Ma, respectively. Obviously, the Nordlandet granulites can no longer be considered temporal equivalents of the type-Nuk gneisses and some refinement of the Akia "terrane" is in order. Contrary to the distinct tectono-stratigraphic nature of a terrane there are areas within the late Archaean Akia terrane mapped as early Archaean Amitsoq gneisses. However, in all cases these rocks do not display the unradiogenic Pb isotopic signature characteristic of the type-Amitsoq gneisses. This is evidence for their misidentification in the field. Further chemical and isotopic studies are in progress and will be used to present a comprehensive picture of the history of the Akia terrane. Friend, C.R.L. et al. (1988) Nature 335, 535.

DECCAN FLOOD BASALT VOLCANISM AND THE KyT BOUNDARY: LINKED TO MANTLE PLUME INITIATION R.A. DUNCAN, College of Oceanography, Oregon State University, Corvallis, OR 97331 U.S.A. The hotspot now producing magma beneath the volcanic island of Reunion (westem Indian Ocean) burst into activity with massive volumes of rapidly erupted basalts over much of westem India at about 66 Ma, coincident with the Cretaceous/Tertiary boundary. High-precision "^^Ar-^^Ar plateau and isochron ages, coupled with magneto-strati graphic data indicate that the entire 1.5 x lO^km^ volume of basalts was erupted in 0.5 million years or less. The large volume and rapid eruption rates are best explained as decompression melting from the head of a mantle plume on arrival at the base of the lithosphere. DiapMc rise of the warm plume material from a deep, gravitationally unstable layer (coremantle boundary) established the conduit through which later hotspot melting was maintained by a significantly lower flux of mantle upwelling. Similar plume initiation events produced other flood basalt provinces on continents and in the ocean basins. These events have been correlated with other mass extinctions but appear to be non-periodic. Subsequent products of the Reunion hotspot were the volcanic foundations of the Laccadives and Maldives Ridges, the Chagos Bank, and much of the Mascarene Plateau, and the young volcanic islands of Mauritius and Rodrigues. Precise radiometric ages are now available for 10 sites along this volcanic trail. The linear geometry and north to south age progression are compatible with motion of first the Indian and more recendy the African plates over a stationary hotspot beneath Reunion. This focused, upper mande melting anomaly has remained fixed with respect to other prominent hotspots in the Atlantic and Indian Ocean basins, and provides a ready frame of reference for reconstructions of plates surrounding the westem Indian Ocean.

ICOG 7

OROGENY AND COOLING IN THE SCOTTISH DALRADIAN Peter DYMOKE: The Department of Geology and Geophysics, Adelaide University, G.P.O.Box 498, S.A. 5001, Australia: There are many examples of the survival of old mineral cooling ages in rocks which show petrological evidence of heating to temperatures higher than the closure temperatures of the isotopic systems concerned. However, there are few, if any, examples of the opposing case where ages are reset to younger values in assemblages which show no petrological evidence of post - peak metamorphic heating above the mineral geochronometer's closure temperature. An example of this latter case is presented from the South West Highlands (S WH) of the Scottish Dalradian. In the Dalradian, the earliest (D1.2) deformation phases occurred before 590 Ma. High pressure syn-Di.2 greenschist to epidoteamphibolite facies assemblages are regionally dominant in the SWH. In strong contrast, lower P - higher T assemblages were dominant elsewhere in the Dah^dian and are currently believed to have formed during a subsequent (D3) deformation phase ca. 490 - 510 Ma ago. There was therefore a hiatus up to 80 - 100 Ma long during which deformation and metamorphism were absent. Rb-Sr phen^te ages from the SWH vary between 460 - 500 Ma, while Rb-Sr biotite and K-Ar phengite and biotite ages are between 475 and 440 Ma old. Syn-Di.2 metamorphic temperatures were similar to or less tlian the normally assumed closure temperature for Sr in phengite. The relatively young Rb-Sr ages from syn D1.2 phengites are therefore apparently inconsistent with the survival of these older assemblages. This contradiction can be resolved if a process can be invoked which maintains these temperatures at these depths for the duration of the hiatus. Such a process must also account for the absence of deformation in that hiatus. A geodynamic model which accounts for the thermal consequences of partitioned strain between the crust and mantle lithosphere suggests that excessive heating of mid to deep crustal levels can be prevented even if thick lithosphere is maintained for 80-100 Ma. Accounting for lithospheric gravitational stability in the models predicts a maximum 30 - 50 Ma time for this hiatus, which may still be long enough to explain the resetting to younger cooling ages obser\'ed in the SWH.

STABLE ISOTOPE RATIOS OF CHLORINE IN DIFFUSION CONTROLLED SYSTEMS EGGENKAMP. Hans and KREULEN, Rob, Department of Geochemistry, University of Utrecht, P.O. box 80.021, 3508 TA Utrecht, The Netherlands ''C1/"C1 ratios were measured on samples from two diffusion controlled systems. The Dutch Lake IJssel contained fresh water before 1600, after which it became saline. In 1932 the lake was closed from the sea by a dam and became fresh again. The pore-water in the sediments reflect this history. Chloride from the saline pore-water layer diffused into both fresh water layers. The different diffusion rates of "CI and "CI produced isotope separation. Diffusion started from saline water with a normal <S"C1 value of C/^o. From the saline to the underlying fresh water a diffusion controlled gradient of 5"C1 is found from +0.5 to -0.4Voo. The starting 5"Cl-value of the saline water before diffusion into the upper freshwater layer was the inherited, positive value +0.5Voo- In this part a diffusion gradient of 6"C1 is found from +0.4 to +0.8Voo- In spite of the large difference in diffusion time, the diffusion coefficients calculated for both fresh water layers are the same (Djsd = 2.95*10"'° D3,CI = 2.89*10-'° m's"') . This implies that the isotope distribution is adequately described by a simple diffusion model and that fluid flow was negligible. Formation waters from the Dutch Westland show lineair 5"C1/C1" relations in samples from a sandstone (slope 0.28) and a sandstone-shale (slope 1.37). The steeper slope in the sandstone-shale is probably caused by a more effective isotope separation compared to the sandstone.

ABSTRACTS

ISOTOPIC AGES FROM TfiE OBAN MASSIF AND SOUTH EAST LOKOJA: IMPLICATIONS FOR THE EVOLUTION OF THE BASEMENT COMPLEX OF NIGERIA EKWUEME, Barth N., Department of Geology, University of Calabar, P.M.B. 1115 Calabar, Nigeria, and CAEN-VACHETTE, M., Laboratoire de Geochronologie, Universite de Clermont II, 5 rue Kessler, Clermont Ferrand, France The evolution of the basement complex rocks of the Oban massif and Southeast Lokoja in Southeastern and Central Nigeria respectively involved two orogenies : Kibaran and Pan - African. The main phase of the Kibaran deformation and metamorphism in these two areas occurred at the same time C. 1313±37 Ma to 1515±72 Ma ago. In the Oban massif this Kibaran event was accom.panied by plutonic activity which emplaced rocks of charnockitic affinity about 1289±153 Ma ago. These ages thus confirm the active participation of the Kibaran orogeny in the evolution of the basement complex as well as the occurrence of Kibaran charnockites in Nigeria. The Kibaran rocks were subjected to Pan African tectogenesis, the main phase of which occurred at the same time in both areas at C. 676±26 Ma to 687±13 Ma ago. A post- tectonic Pan - African event which was marked by diaphthoresis occurred in the Oban massif 527±16 Ma and is believed to be coeval with a 546±24" Ma event which reworked the charnockites in the area. The peak thermal event in the Oban massif occurred 510±11 Ma^and caused widespread migmatitization of schists and gneisses.

MANTLE CONTAMINATION AS RECORDED IN SULFURBEARING DL\MOND INCLUSIONS ELDRIDGE. C.S.. GeoL Dept. and RSES, ANU, Canberra, ACT 2601, Australia, COMPSTON, W., RSES, ANU, WILLL\MS, LS., RSES, ANU, BRISTOW, J., Anglo American Corp. Ltd, Research , PO Box 106, Crown Mines, 2025, South Africa and HARRIS, J.W., Dept. of GeoL, U of Glasgow, G128QQ, U.K. The mantle environment maintains sulfur in the reduced state so that pristine mantle sulfur is characterized by meteoritic sulfur isotopic compositions (0 %o). The sulfur isotopic compositions of thirty selected diamond inclusions have been determined using the ion microprobe SHRIMP. Sulfide minerals studied include po, mss, pnt and ccp from six kimberlite pipes in southern Africa. The sulfides range in isotopic compositionsfrom-10 to +14 %c demonstrating that crustal contamination or other sulfur isotopic fractionating processes must have occurred for at least some diamond sources. Sulfides obtained from the Premier, Roberts Victor and Koffiefontein pipes exhibit isotopic homogeneity near 0 %o and might be derived, therefore, from a pristine mantle source, while those from the Jagersfontein and Orapa pipes range from low values (-10 %o) to those greater than 0 %c. Sulfides from the Sierra Leone pipe are confined to the high end of the range (>8 %c). The data suggest that not only has isotopically heavy sulfur, perhaps from altered seafloor basalts, been add^ to the subcontinental mande during subduction, but that isotopically light sulfur, possibly from biogenic sulfide in sediments, was introduced as well. Moreover, isotopic heterogeneity in individual kimberlite pipes may indicate that diamond formation spans a considerable interval of time and/or depth, reflecting introduction of surface-altered basaltic or sedimentary material to kimberiite source regions during diamond growth. Ion probe measurements of Pb in the sulfide inclusions varyfroma few tenths to a few hundred ppm, and many give satisfactory Pb isotopic compositions. Model Pb ages rangefrom- equal to the pipe age to much greater (2.5 Ga). Some sulfides contain highly radiogenic Pb unsupported by in situ U and Th, possibly added from an external source. Pb isotopic ratios vary widely between some inclusions from the same pipe and parallel their dispersion in 5 ^"^S values, suggesting a variable source Pb composition.

ABSTRACTS

ICOG 7

DEVELOPMENT OF A HIGH-ABUNDANCE SENSITIVITY MASS SPECTROMETER FOR TIMS, SIMS, AND S I R I M S J. ENGLAND, A. ZINDLER, L. REISBERG, S. WEAVER, J.L. RUBENSTONE, B. BOURDON, F. MARCANTONIO, AND V .

SALTERS, Lamont-Doherty Geological Observatory of Columbia University, Palisades, N.Y. lOW U.S.A. The Lamont VG Isolab 54 is a prototype double focussing isotope ratio mass spectrometer that allows the measurement of atoms thermally ionized on a hot filament (TIMS), secondary ions sputtered using an ion beam (SIMS), and sputtered neutrals resonantly ionized using a laser (sputter-induced resonance ionization mass spectrometry—SIRIM^S). Its forward geometry (electrostatic analyzer followed by magnetic analyzer) allows the simultaneous collection of different isotopes of the same element. The Lamont instrument is the first to have an ion-counting multicollector based on a multi-channel plate as well as the traditional Faraday cups. A second electrostatic analyzer after magnet and multicollector is equipped with an ion-counting Daly system, and yields an abundance sensitivity of --0.1 ppm/amu at mass 230. The Duoplasmatron ion source, operating at up to 20kv, produces a primary Ar+ ion beam for SIMS and SIRIMS. The p r i m ^ beam is velocity-filtered in the Duoplasmatron column, yielding primary beam currents of pamp at the sample. An excimer laser, firing at up to 80 Hz, pumps a tunable dye laser whose output is fr^uency-doubled to produce UV photons in the range 270-295 nm. Tiiis range includes several resonance lines for Os and Re, the principal elements of interest for SIRIMS. Present applications of the Lamont Isolab 54 include precise isotopic analysis of U and Th by TIMS, Th and Hf by SIMS, and Os by SIRIMS. A discussion of the problems and relative merits of each technique as applied to geological samples will be given during the presentation together with a more detailed description of the instrument, the operating characteristics of its components and the design changes and improvements that have been incorporated during the development period at Lamont.

SOME PRACTICAL ASPECTS OF Pb ISOTOPE STUDY BY SIMS E.U. ENGSTROM*. S. CLAESSON** and A-LODDING'-^^ -'Dept.of Physics, Chalmers Univ. of Technology, S-A1296 Gothenburg, Sweden "•"Swedish Museum of Natural History, Box 50007, S-10A05 Stockholm, Sweden Pb isotopic measurement by Secondary Ion Mass Spectrometry (SIMS) is a powerful tool in isotope geology. Reproducible measurements can be made on very small selected areas on a specimen. Exact SIMS determination of lead isotope ratios requires a thorough grasp of numerous possibly involved experimental artifacts. The present contribution discusses such factors and how to cope with them. A subject to be particularly addressed is to what extent a commercially produced magnetic-sector multi-purpose SIMS (such as Cameca IMS 3F/4F or VG 1X70) can be used for the tasks for which bigger instruments (SHRIMP Mk. II, Cameca 1270, VG Isolab 54 5e 120) are presently being designed, and what kind of hint the performance of the first mentioned type of ion probe might contribute for the forthcoming instrumental development. Recent results of Pb dating of zircons, performed with an IMS 3F equipment, are to be presented, as well as a discussion of the limitations and potentials of the technique.

31 Rb-Sr AGES OF DL^GENESIS OF Mg-RICH CLAYS, PERML^N EVAPORTTE SEQUENCE, PALO DURO BASIN, TEXAS, USA ERWIN, M. E., LONG. L. E., Dept. of Geological Sciences, Univ. of Texas at Austin, Austin, TX 78713, USA, and FISHER, R. S., Bureau of Economic Geology, Univ. of Texas Mid to late-Permian sediments in the shallow (3(XX) m), intracratonic Palo Duro Basin, Texas, consist of multiple cycles of sabkha carbonate-evaporite-redbed assemblages. Chemical and X-ray diffraction analyses indicate that Al-rich detrital clays (illite, smectite, mixed-layer illite/smectite) have been altered to Mg-rich authigenic clays (chlorite/smectite, other mixed-layer chloritic clays), probably in open brine pools or in the presence of pore-fluid brine following shallow burial. We performed Rb-Sr analyses of diagenetic clay from three core samples from each of the Tubb (oldest). Lower San Andres, Upper Seven Rivers, and Salado-Tansil (youngest) evaporitic formations to examine the timing of clay diagenesis. After an ultra-fine clay fraction (<0.05 micrometer) consisting of the diagenetic component was separated by centrifugation, it was divided into 3 or 4 aliquots for treatment with different cation exchange reagents to remove loosely bound Rb and Sr, then analyzed isotopically. Data from treated and untreated aliquots plot as linear arrays on a Rb-Sr isochron diagram, giving "leachochrons" whose ages and initial ratios we regard as meaningful. Initial 87Sr/86Sr (0.7070 to 0.7076) equals or slightly exceeds that of mid-Permian seawater (0.7070 to 0.7075), and the isochron ages are close to the presumed age of deposition. Combined data from the Tubb Formation yield t = 261 ± 11 Ma (0.7076, 9 data points). The Lower San Andres yields t = 244 ± 34 Ma (0.7103,13 points). Excluding an aberrant sample, a 6-point isochron yields t = 276 ± 7 Ma and initial ratio = 0.7070. The Upper Seven Rivers isochron provides t = 268 ± 14 Ma (0.7076, 10 points). The Salado-Tansil isochron corresponds to t = 250 ± 18 Ma (0.7078,12 joints). Ages and initial ratios indicate that diagenesis shortly folowed deposition, mediated by fluid whose Sr isotope composition resembled that of contemporary seawater. Apparent ages of coarser clay (0.2-0.5um) are 320-340 Ma with large uncertainties, suggesting the presence of a detrital component. With the ion-exchange leaching technique, good-quality data were obtained from smectitic clay that contains as little as 0.5 to 2 ppm of Rb. Mg ISOTOPE FRACTIONATION IN LUNAR SOILS Tezer M. ESAT. Research School of Earth Sciences, The Australian National University, Canberra 2601, Australia. Previous work has established the presence of large mass dependent enrichments in I80/I60 and 30si/28si in bulk lunar soils for the initial O2 and Si extracted by partial fluorination.The anomalies are believed to reside on the surface coatings of soil grains and appear to be correlated with concentrations of solar-wind H2and have been generally attributed to solar-wind sputtering and gravitational segregation in the lunar gravity field. In contrast to the large effects in O (2.5% /amu) and Si (1.3% /amu) attempts to extract Ca from lunar soils have failed to show any large effects. We have previously extracted Mg from 4 different bulk lunar soil samples (14163, 15021, 15271, 66081) by water leaches. The leached Mg was isotopically indistinguishable from terrestrial. We have extended these experiments to individual pyroxene grains, hand-picked from each sample. Large single grains up to 500 ^im or several smaller grains of similar total mass were immersed in HF+HNO3 for about 30 seconds. The leached Mg which ranged from about 20 ng to 60 ng constituted less than 0.2% of the total mass and corresponds to the removal of a surface layer to a depth of approximately 0.2 ^m. About 15 ng is sufficient to obtain a 24Mg beam of TxlO'^^ \ foj- several hours. Two batches from each soil sample were analysed. The data show no resolvable effects beyond the ±1.5%o limit for instrumental mass fractionation. The absence of large mass dependent fractionation effects in Mg and Ca in lunar soils is in sharp contrast to per cent type anomalies in O and Si. For Mg we have <iemonstrated that water leaches or HF fluorination of bulk soil samples or surface leaching of individual pyroxene crystals produces sufficient quantities of Mg for isotopic analysis. The isotopic composition of this Mg is indistinguishable from terrestrial Mg. We cannot offer any plausible explanation for the difference in results for O and Si as opposed to Mg and Ca except that: 1) The present leaching procedures do not remove significant amounts of isotopically fractionated Mg. 2) The leached Mg is somehow contaminated with Mg from easily soluble bulk phases. However, the most plausible explanation may be that the effects reported from the original O and Si experiments are artifacts of the fluorination extraction process.

OS ISOTOPES IN MARINE AND TERRIGENOUS SEDIMENTS ESSER. B. K..and K. K. Turckian, Dept. of Geology and Geophysics, Yale Unitverstiy, New Haven CT 06511, USA Osmium concentrations and isotopic compositions have been measured by SIMS in a variety of marine and continental sediments to determine the sources of Os to the marine system. Crustal Os: The Os isotopic composition of the continental crust was determined by analyzing North American loess and Amazon, Changjiang and Mississippi River shelf sediment Measured crustal 187o^l86os for all sediments falls between 10 and 11; age-corrected i87os/l86os for the total upper continental crust is 13.9. Sediment Os concentrations are 15-90 pg/g; the mean upper crustal Os concentration.is 50 pg/g. Pollutant Os: In Long Island Sound, Os is enriched in surface sediment relative to deep sediment. In adjacent New Haven Harbor, excess Os is correlated with excess organic carbon derived from sewage sludge. Os concentrations are 40-770 pg/g; 1870s/l860s is 1.6-8.5 and is correlated with 1/Os. If this correlation is due to mixing, the extrapolated anthropogenic endmember l87os/i86os is ~1 and is indistinguishable from analyzed New Haven Harbor sewage sludge. Cosmic Os: Two Equatorial Pacific pelagic clays with known sedimentation rates (DOMES sites A and C) have bulk Os = 137-163 pg/g and l87os/l86os = 5.9-6.6. Oxalic acid leaches of sites A and C manganese nodules and of site C pelagic clay have 1870s/1860s = 8.0-8.5. If leaches extract hydrogenous Os, nonhydrogenous 1870s/1860s in site C clay is 5.7 by mass-balance and much less radiogenic than continental crust. As this pelagic clay is free of metalliferous or volcaniclastic phases, the non-radiogenic phase must be particulate meteoritic material with 1870s/^^^s = 1. Since continental detritus has 1 8 7 0 s / l ^ s = 10-11 we can calculate the cosmic component by mass balance. The calculated meteoritic flux is 4-6 X 10^ tons/y and is comparable to previous estimates using 53Mn and Ir. Seawater Os: Most abyssal manganese nodules and all reduced marine sediments have I870s/l860s = 7-9 and are less radiogenic than continental crust. Mass-balance suggests that 60-90% of seawater Os is derived from weathering of continental crust The source of non-radiogenic Os to seawater must be oceanic lithosphere or cosmic dust

ICOG 7

ABSTRACTS

LASER ABLATION STUDIES OF CARBON AND OXYGEN ISOTOPES IN CARBONATES EXLEY R.A. and FREEDMAN, P.A., VG ISOTECH Ltd, Middlewich, Cheshire CWIO OHT, U.K. Carbon and oxygen isotopic compositions of carbonates have been measured using a continuous wave Nd-YAG laser to liberate CO^. Spatial resolution of a few lO's of microns is achieved, with liberated volumes of CO2 of a few micro!itres. Individual spots or rastered areas may be analysed on slabs or thick sections. The evolved CO is analysed immediately via an on-line connection to mass spectrometer. The effects of laser power and operating mode on reproducibility and accuracy were determined by analysis of well characterized standards and natural samples. Optimum reproducibility is obtained with precise control of input laser power and is better than 0.2 per mil for both delta ^^C and delta ^^o. No significant mass fractionation effects are seen for carbon isotopes, and I80 shows only small and reproducible deviations from expected compositions ( 0.5 per mil).

THE ISOTOPIC RECOGNITION OF SINTERS IN EPITHERMAL GOLD DEPOSITS G.R. EWERS. Bureau of Mineral Resources, GPO Box 378, Canberra, ACT, 2601, Australia

ADVANCES IN THE PRACTICE AND APPLICATION OF LASER PROBES TO S^^'S ANALYSIS OF GEOLOGICAL SAMPLES FALLiCK, A. E., BURGESS, R., BOYCE, A. J. and KELLEY, S. P., Isotope Geology Unit, S.U.R.R.C., East Kilbride, Glasgow, G75 OQU, Scotland

Surficial deposits of siliceous sinter are a common feature of present-day hot springs in areas of active volcanism, but they are rarely recognised in older, eroded volcanic terranes because of their low preservation potential. Recently, ancient sinters have been widely reported from gold prospects in the DevonianCarboniferous Drummond Basin in northeast Queensland: their recognition provides tangible evidence of a palaeosurface associated with an extinct geothermal system and indicates the potential for a significant mineralised epithermal gold system beneath and possibly adjacent to the sinter. Although sinter recognition is clearly important in a prospect area, the characterisic features of modem sinters may be absent or simply not preserved in ancient sinters. In the absence of diagnostic features, a rock incorrectly interpreted as a sinter could lead to false assumptions about the potential of an epithermal prospect, and possibly to a misdirected exploration program Oxygen isotope data for sinters from epithemial gold prospects in the northern Drummond Basin indicate the potential to characterise them on the basis of their isotopic signature. For example, at the Conway prospect, sinters are consistently and significantly more

Laser microprobe extraction systems linked to conventional stable isotope ratio mass spectrometers offer several attractive features compared to conventional techniques. Principal amongst these is the potential for high precision isotope ratio measurement, coupled to in situ analysis with moderately good spatial resolution. S^^S assay of sulphides is possible by using a laser beam as a localised heat source to combust the sample in an oxygen atmosphere. In such experimental systems, combustion of sulphides to SO^ can introduce sulphur isotopic fractionation. Detailed experiments have been conducted, with both argon ion and Nd YAG continuous wave lasers, to demonstrate that this fractionation (A S) is mineral-dependent. A^^S varies with the bond strength of the sulphide but appears to be independent of the sample ^^S/'^^S ratio. The precision routinely attainable is similar to that achieved with orthodox combustion of drilled samples (around ± 0 . 2 7 o o , la). Detailed traverses have been performed on a variety of geological samples from different environments of formation. Conventional gas purification procedures are employed to produce clean prior to mass spectrometry. Galena vein S^^S profiles from Lower Carboniferous 70M tonne Pb-Zn sedex deposit at Navan, Eire have been determined by both conventional drilling and by laser probe techniques. The in situ measurements demonstrate broader range in S^^S and the rate of change of is more accurately established. Models which envisage mixing of two sulphur sources are thus better constrained.

^^^SMOW enriched (+11.5±0.7 per mil) than nearby vein quartz (+6.3±1.9 per mil). These sinters are consistent with deposition at 90-100OC from a fluid having a value of about -9 to -8 per mil. Such marked ^^O-enrichment relative to higher-temperature vein quartz can be explained by the rapid increase in the O isotope fractionation factor for the quartz-water pair with decreasing temperature, particularly below lOQOC. Recr>^stallization of an originally amorphous sinter to cryptocrystalline quartz does not appear to have affected the isotopic signature of the sinter.

ABSTRACTS

ICOG 7

ISOTOPIC RECORD ON PALEOZOIC CRUST EVOLUTION WITHIN HERCYNIAN RIFT IN HAINAN ISLAND, CHINA FANG,Zhong, TAO,Xiancong, XIA,Bangdong, and YANG,Jiedong, Department of Earth Sciences, Nanjing University, Nanjing 210008, P.R.China

ISOTOPIC AND CHEMICAL SYSTEMATICS OF MURCfflSON REFRACTORY INCLUSIONS: IMPLICATIONS FOR LRTE FRACTIONATION IN THE SOLAR NEBULA FEGLEY. Bruce, Max-Planck-Institutfur Chemie, Mainz 0-6500, Federal Republic of Germany, and IRELAND, Trevor, Research School of Earth Sciences, ANU, Canberra ACT 2601, Australia.

HainanÎsland is a.continental crust island ( and 108.5^-110Ê). Authors have found that there was a Hercynian rift in the main part of it. The basement rocks in the rift are amphibolite facies, Rb-Sr WR isochron 601, 554 and 470Ma, an altered mineral isochron 257Ka; a hornblend K-Ar age 311^1a, The gneiss Pb-Pb V/R isochron 1700Ma shov;s the age of ancient crust materials. In the nucleus of it there are greenstones with bimodal volcanics and Fe-Co-Cu deposits, which are the typical sedimentary-metamorphosed formations in rifts. Their Rb-Sr V/R isochron 330Ka, dolomite and himatj } Pb model ages 200-400Ma, besides some r. ided zircons Pb-Pb isochron 1288Ma and U-Pb disconcordia 1352Ma, vîch were sampled from the schists and ferrolite. These dates show various sources of sediments in pre-rifting and the rifting periods. Voluminous prophyroblastic granites are widely located in it, their Rb-Sr TO isochron 320Ma and mineral isochron 31^Ka. Based on the Sr and 0 isotope evidence they were anatectic crust products in the rifting, v/hich were composed of 53% felsic and basic crust mass. Conclusion: Some materials of mantle (basalts) entered into upper crust in the rift and the crustal rocks were metamorphosed or remeIted m d e r Hercynian rifting conditions which were the crust stritched, deep seated pressure decrease and h i ^ e r geothermal gradient.

The abundances of the lithophile refractory trace elements (LRTE) are highly enriched in refractory inclusions from carbonaceous chondrites relative to average solar abundances. In particular, the Allende (CVS) Group II REE pattern with its depletions in the most refractory as well as most volatile LRTE requires that condensation was a primary process in the fractionation of these elements. Refractory inclusions from the Murchison (CM2) chondrite differ from those of Allende in that isotopic and chemical characteristics are largely correlated in different mineralogical types. Independent of the mineralogical groups there are two features that hold for all Murchison inclusions: (1) No inclusion has both large ^^Ti anomaly and excess ^^Mg* (from the in situ decay of ^^Al), and (2) those inclusions that have ultrarefractory-depleted LRTE abundances (c.f. Group II) all have excess ^^Mg* provided they do not violate (1). These systematics suggest that solid precursors played a large roll in the preservation of isotopic anomalies. However, chemical fractionations may have also been preserved in the precursors and the LRTE patterns are not necessarily associated with condensation in the solar nebula. The formation of the Allende Group II pattern requires multistage condensation processes with isolation of the condensate from the cooling gas. So too the LRTE abundances of the Murchison refractory inclusions require complex cooling histories if the observed elemental fractionations are to be produced. Not only are the ultrarefractory-depleted condensate patterns preserved in Murchison inclusions, but also the ultrarefractory-enriched residue complement. However, the two patterns are not perfectly complementary and interelement fractionations other than by gas-solid condensation have played a role. Crystal-liquid partitioning appears to have produced a smooth variation in abundances with ionic radius. Further work involving a reevaluation of activity coefficients based on experimental and natural systems could also have implications for the production of these patterns.

IDTIMS SINGLE GRAIN ANALYSES OF lON-MICROPROBED ZIRCONS; A COMPARISON OF U-Pb TECHNIQUES THROUGH GEOLOGIC TIME. FANNING. C.M., WILLIAMS, I.S., McCULLOCH, M.T. and COMPSTON, W., Research School of Earth Sciences, A.N.U., Canberra, ACT 2601, Australia.

BEHAVIOUR OF RADIOGENIC ISOTOPES IN CRUSTAL ANATEXIS (THE TUSCAN MAGMATIC PROVINCE: A CASE HISTORY) FERRARA G. & TONARINI S., Istituto Geocronologia e Geochimica Isotopica, CNR, Pisa (Italy): INNOCENT! F. , Dipartimento Scienze della Terra, Pisa (Italy).

The application of the ion-microprobe has enabled the direct determination of Pb isotopic compositions and U/Pb ratios in discrete (~30[im) portions of individual zircon grains. In many cases, these studies have shown that apparently simple single zircon crystals contain discrete zones and/or portions that crystallised at different times. Isotope dilution analysis using thermal ionisation mass spectrometry (IDTIMS) remains the most accurate and precise method to determine U/Pb ratios and has the potential advantage relative to SHRIMP in that the larger sample size enables determination of more precise and accurate Pb isotopic ratios. An obvious disadvantage of IDTIMS is that it cannot determine ctystallisation ages for even single crystals that contain temporally discrete zones or portions, nor crystals which have suffered heterogeneous Pb loss. We present here the first IDTIMS analyses of the very same zircons that have been analysed by SHRIMP. Samples were chosen with U-Pb zircon ages of 320,400, 800,1150,2450, 3800,3960 and 4100 Ma. The zircons had been mounted in epoxy, sectioned and analysed using SHRIMP. The temporally and structurally simplest "half' grains were removed from the polished mount, dissolved by HF vapour transfer together with a mixed 205pb.235u tracer and analysed using a MAT261 multicollector mass spectrometer. The 206p5/204p|5 ratios were determined using Faraday cups in combination with the electron multiplier. The U-Pb data enable direct comparisons to be made between IDTIMS and SHRIMP. The results show excellent agreement between the two methods. IDTIMS 207p5/206p5 ratios are more precise than those determined by SHRIMP and for concordant, simple grains or homogeneous sub-grain fragments give more tightly constrained ages. In general, IDTIMS yields more precise "average" U-Pb analyses whereas SHRIMP yields more accurate crystallisation ages for the 30 |im spot analysed.

The behaviour of different isotopic systematics during the process of production of cnistal anatectic melts is discussed. The Tuscan Magmatic Province, in Italy, is considered to be dominantly and, in some cases, exclusively of crustal origin; it is mainly composed by intermediate-acid intrusive and volcanic rocks outcropping in south Tuscany and north Latium. Isotopic data show that the products of some volcanic centres are the result of a process of interaction of anatectic melts and subcrustal m a ^ a s . For this study we report only data on samples which are considered only of cmstal origin. New detailed isotopic studies of young silicic volcanic rocks of Roccastrada (Rb/Sr age 2.45 Ma), Tolfa-Ceriti Volcanic Complex (K/Ar age 3.5 Ma) and of granitic body of Giglio island (Rh/Sr age 5.0 Ma) together with the already pubblished data from the S. Vincenzo volcanites (K/Ar age 4.7 Ma) and the granitic intrusion of Mt. Capanne, Elba island (Rb/Sr age 6.8 Ma) allow a more detailed comparison of the geochemical and isotopic characteristics of the intrusive and effusive rocks. The Nd isotopic ratios are low (.51210-.51222) and relatively homogeneous for all the studied samples as already seen for the Pb isotopes. In contrast initial Sr isotopic ratios are high either for the granitic bodies (.7113-.7176) and for volcanic rocks (.7131-.7250) and represent an important tracer for the characterization of the source materials. In the rapid-cooled volcanites the crystal-matrix isotopic relationships allow to investigate the melting process; evidence of strong isotopic disequilibrium in S.Vincenzo centre suggests for this silicic magma, an origin by a mixture of partial melts and restite crystals; a different situation is depicted by Roccastrada rhyolites that exhibit a complete equilibrium between mineral-phenocryst phases and matrix supporting the hypothesis of a complete separation of the melts from the source. Sr isotopic compositions of minerals separated from granitic rocks show only small degree of disequilibrium, probably due to the longer cooling time.

ISOTOPIC EVIDENCES FOR PHENOCRYST-MATRIX DISEQUILIBRIUM IN THE INTERACTION BETWEEN CRUSTAL AND SUBCRUSTAL MAGMAS FERRARA G. & TONARINI S., Istituto Geocronologia e Geochimica Isotopica, CNR, Pisa (Italy); TAYLOR H.P., Caltech, Pasadena (USA); TURIB.. Dipartimento di Scienze della Terra, Roma (Italy) Partial melting of crustal rocks can generate granitic melts with different characteristics; isotopic systematics can be used to decipher the origin of these magmas. Time-scale for melting and crystallizzation in basalt-induced crustal melting are very short and isotopic disequilibrium between phenocrysts and matrix in silicic magmas and in mafic inclusions can be maintained. An interesting example is represented by the isotopic disequilibrium between different minerals, separated from mafic inclusions and host rocks of the Tolfa-Ceriti Volcanic Complex, Tuscan Anatectic Province (Italy): the products of this complex are mainly represented by trachytic and rhyolitic lava domes 3.5 Ma old. Magmatic enclaves of trachybasaltic and rhyolitic compositions are often present. Sr-Nd isotopic compositions of the Tolfa-Ceriti trachytes range between .71407-.71314 and .51209-51213 respectively. Pla^oclases, sanidines and glasses from the same rocks are nearly in isotopicic equilibrium (Pl= .71387.71405; Kf= .71386-.71399, Gl= .71278-.71388); pyroxenes and biotites from the same rocks exhibit lower ®7Si/86Sr ratios, i.e. .71209-.71355; and .71213-.71375 respectively. The^^So^^Sr ratios of the Tolfa-Ceriti enclaves (w.r.) range between .70778 and .71311; i43Nd/i44Nd ratios are between .51242-.51221. Xenocrysts separated from the enclaves show a strong isotopic disequilibrium with their matrix.87Siy«6Sr range from .70821-.71441 and from .71384-.71417 in PI and Kf respectively; the overall groundmass variation is from .70775 to .71097 . The high isotopic composition of PI and Kf rules out the possibility that most of them have crystallized from the pristine subcrustal magmas and the values, close to those determined on PI and Kf separated from the host rocks, suggest a more acid magma (like the one which generated trachytic lavas). Clinopyroxenes (from the enclaves) are characterized by the lowest ^^Sr/^^Sr ratios (.70718-.70843) probably close to the isotopic composition of the original subcrustal materials. The Tolfa-Ceriti trachytes are probably the melting product of a relatively homogeneous pelitic source («7Sr/86Sr - .7140 and ^^^Nd/^'^^l^d " .51212) only slightly contamined with hybrid magmas, represented by the mafic enclaves.

ICOG 7

ABSTRACTS

THE APPLICATION OF WIDE-DISPERSION MASS SPECTROMETERS TO K-Ca GEOCHRONOLOGY FLETCHER. Ian R. , ROSMAN, K.J.R., School of Physical Sciences, Curtin University, GPO Box U1987, WA 6001, and McNAUGHTON, N.J., Dept. of Geology, University of WA, Nedlands, WA 6009, Australia. A K-Ca facility is being developed, based on the anticipated installation of a MULTI II upgrade on a VG 354 mass spectrometer which will provide a beam dispersion greater than 4 amu at Ca. This enables pseudosimultaneous collection of all isotopes from to ^^Ca by using ^^Ca/'^'^Cai as an internal normalising pair. Using this mass span and a double spike it is possible to determine Ca concentration, natural Ca isotopic fractionation, and radiogenic ^"Ca in a single mass spectrometric analysis. K concentrations are measured without /ff^s fractionation uncertainties, using a spike. The initial focus of this project is on comparisons between K-Ca and established chronometers. Preliminary results on a suite of metamorphically reset biotites from the Yilgarn Craton indicate substantially higher blocking temperatures for K-Ca than for Rb-Sr, though the actual blocking temperatures are not yet determined.

APATITE nSSION TRACK CONSTRAINTS ON THE EXTENSIONAL EVOLUTION OF THE GOLD BUTTE CRUSTAL SECTION, SOUTH VIRGIN MOUNTAINS, NEVADA. FITZGERALD. P.G., Geology Dept., Arizona State University, Tempe, AZ 85287; FRYXELL, J.E., Geology Dept., California State Univ., San Bernardino, CA 92407 and WERNICKE, B.P., Dept. of Earth and Planetary Sciences, Harvard Univ., Cambridge, MA 02138

PROTEROZOIC LITHOSPHERIC ENRICHMENT AND ITS PROFOUND INFLUENCE ON THE ISOTOPIC COMPOSITION OF PHANEROZOIC MAGMAS. FODEN, J., and TURNER, S., Dept. of Geology and Geophysics, University of Adelaide, MICHARD, A. C.R.P.G., CNRS, Vandoeuvre les Nancy, France.

The Gold Butte block (GBB) appears to be a cross-section through the upper 10-15km of Colorado Plateau crust, tilted eastward and tectonically denuded via top-to-the-west Tertiaiy normal faulting. Thus the E side of the block exposes steeply dipping cratonic Paleozoic section (just 5-10km W of the western boundary of the modem Plateau), ~2-3km thick with Proterozoic basement to the W continuously exposed for an apparent thickness of - 15km structurally below the basal Cambrian unconformity (u/c). Western portions of the GBB contain a zone of mylonitic gneisses and chlorite breccias, similar to those observed in Cordilleran core complexes, with W trending lineations and top-to-the-W shear sense. Garnet gneiss, a major lithology within the block, shows progressively more retrogression approaching the zone of mylonite and chlorite breccia. Apatite fission track (AFT) dating and confined track length distributions have been used to determine the time of unroofing of the GBB and hence the timing of extension in this part of the Basin and Range Province, as well as test the hypothesis that it is arelativelyintact crustal section. AFT ages are given with respect to structural depth beneath the Cambrian unconformity. Ten samples were collected in an E-W transect across the block, from just beneath the u/c (0.1 km) to just east of the chlorite breccia (~14km). From 0.1 to 1.2km ages decreased from 50 to 34 Ma, and had length distributions of 12.4 - 13.2 ^lm with standard deviations (o) of ~2 ^m. These indicate significant time spent within the partial annealing zone (PAZ) following Laramide uplift (l-2km of erosion of Mesozoic strata) but prior to extension and later uplift. From 1.6 to -lOkm ages are identical within error, with a weighted mean of 15 ± 0.5 Ma. Mean lengths are -14.5 ^m and o = -1.3 p-m. This reflects rapid cooling of these samples below 10®C due to tilting and uplift of the GBB at ~15 Ma, effectively constraining the time of extension to mid Miocene time, and synchronous with the onset of rapid synrift sedimentation in the region. The base of the PAZ (-110°C) was therefore at a depth of ~ 1.5km below the u/c (or at 4-6km depth prior to block tilting). The most western sample (~14km) from completely retrograded garnet gneiss close to the chlorite breccia does not, however, fit the straightforward pattern above. This sample had an age of 77 ± 4 Ma, and a mean length of 12.7 ^lm with a = 2.1 [iin, indicating a substantial degree of track annealing.

The Nd and Sr isotopic compositions of continental rocks demand much greater degrees of Rb/Sr and Nd/Sm fractionation than those predicted during substantial degrees of fusion of the mantle. As granites have Nd/Sm ratios like their lithospheric sources, such secondary melting cannot generate observed isotopic characteristics either. The initiation of Palaeozoic continental growth on the south eastern margin of Gondwanaland is recorded by Cambrian rocks in south eastern Australia and in previously contiguous parts of Antarctica (Victoria Land and the Transantarctic Mountains). The South Australian Delamerian Orogeny , the Antarctic Ross Orogeny and tectonic events in Tasmania all take place during the time 520 - 480 Ma. The Nd and Sr-isotopic compositions of South Australian Cambro-Ordovician granites reflect the strong influence of a 1.1 ± 0.2 Ga. (depleted mantie) model age component and are little influenced by the regional middle and upper crust which has 2 - 2.4 Ga. Nd model ages. The isotopic characteristics demanded of this source (and its geochemical composition) are the same as those required by the extensive Tasmania and Antarctic Ferrar Jurassic dolerites. Deformed Iherzolite xenoliths from Quaternary volcanics in S.E. South Australia, with very low two pyroxene equilibration temperatures have the same isotopic characteristics These factors indicate the existence of a widespread upper mantie lithospheric source with an important component whose average age is Late Proterozoic and which has influenced the compositions of magmas over a protracted period. Worldwide Phanerozoic magmatism from many seperate localities within and at the the margins of the fragments of Gondwanaland have Sr- and Nd- isotopic compositions which implicate a common component. Examples include Hercynian and Tertiary ^anites in Europe and potassic Cainozoic volcanic rocks in the Mediterranean and Indonesia. These observations reflect sub-continental lithospheric metasomatic enrichment, favoured by the assemblage of the Late Proterozoic supercontinent.

ABSTRACTS

ICOG 7

FORMATION OF OOGENETIC QUARTZ AND NEPHELINE SYENITES DUE TO CRUSTAL INTERACTION: A TALE OF TWO COMPLEXES FOLAND. K. A., LANDOLL, J. D. , Dept. of Geology & Mineralogy, Ohio State Univ., Columbus, OH, 43210, USA, HENDERSON, C.M.B., Dept. of Geology, The University, Manchester M13 9PL, UK, and CHEN, JF., Univ. Science & Technology, Hefei, Anhui, PRC Apparently oogenetic quartz and nepheline syenites occur at many locations and explaining the formation of these silica over- and undersaturated lithologies has been problematic because of the thermal divide in petrogeny's residua system (PRS). Such associations are under investigation in epizonal alkaline complexes. Detailed ^"^Sr/^^Sr and i^^Nd/^^^Nd studies along with chemical work have been conducted at two complexes: Marangudzi (178 Ma) in southeast Zimbabwe; and, Mt. Brome (122 Ma) in southern Quebec, Canada. These contain an early gabbro mass and units of quartz syenite along the periphery and nephelinebearing rocks in the center. Isotopic, chemical, and age relations suggest that the units of each are oogenetic, formed by open-system processes from mantle-derived p a r e n t s . Syenites form coherent groups with projected compositions falling on both sides of the divide in PRS. Their ^'^STC/^^STC and ^^^Nd/i^^Nd ratios which show large variations are inversely correlated and are also correlated with bulk chemical parameters (e.g., Mg/Fe, Si02). The relationships indicate input of crustal material by open-system processes with greater amounts for the most quartz-rich syenites. The silica over- and undersaturated associations may be explained by incorporation of Si02-rich crust as solid or melt by an initially u n d e r s a t u r a t e d magma. The observed relationships can be modeled as AFC-type processes which are consistent with the phase relations of the PRS. Such processes may account for silica oversaturated rocks from initially undersaturated magmas more generally and similar associations more specifically in continental settings.

ISOTOPE PALAEOHYDROLOGY PONTES. J.Ch., University dc Paris-Sud, Laboraioirc d'Hydrologie, et de G6ochimic Isotopique Specific difriculdes encountered in applying stable isotopes to continental sites are caused by the ambiguity relating to the relationship between isotope ratios (S^^q or S^^C) and temperature which is also dependent on the isotopic ratio of the environmental fluid (either water or dissolved inorganic carbon). Nevertheless, the following situauons can be considered in continental systems: (1) under the influence of a temperate climate, the 5^80 of authigenic carbonates in lakes with high turnover rates may be related to the S^^O of rainfall which in turn can be linked to the air temperature in the catchment area. (2) under the influence of a dry climate, either cold or warm, but especially in tropical lakes, the S^^O of authigenic carbonates mainly reflects the evaporation/precipitation ratio. Lakes fed by shaJJow ijuifers, like those in sand dunes, permit detailed reconstructions of the local palaeohydrology. Lakes which are affected by the discharge of confined aquifers are also of interest because they promptly reflect fluctuations in recharge conditions, viz. the difference between precipitation and evapotranspiration. Variations in ^^c contents of lacustrine carbonates are mainly due to the various contributions of several potential sources of dissolved inorganic carbon (DIG): atmospheric CO2, photosynthetic CO2, bicarbonate from the aquiferj, CO2 cogenetic with methane, and emanations of deep-seated CO2. Variations in the ^^C contents of fossil lacustrine carbonates can give qualitative information on the primary productivity in the lake, its turn-over rate, the conditions of mixing, and the redox conditions in bottom sediments. Of special importance is the case where equilibrium conditions with atmospheric CO2 can be inferred from PDB values close to zero or slightly positive. This phenomenon indicates that the carbonate provides suitable material for ^^C dating. In this case, especially, inorganic calcimicrite precipitates close to the surface of the water. Case studies showing predictable concordance or discordance in age are discussed. Carbon-13 studies of carbonates may be complemented with l^c profiles on organic matter from lacustrine core material. Organic matter derived from aquatic vegetation (frequently C4 plants) or vegetation debris supplied from the catchment area during periods of high run-off may be identified. Palaeohydrological information provided by stable-isotope typology of lacustrine deposits can be compared to reconstructions given by bioindicators—such as diatoms and ostracods. Ostracods are especially important in that respect since they have the potential to provide information on the following: bioenvironment, stable-isotopes, dating O^C and ThAJ disequilibrium) and trace elements (e.g. Sr2+, Mg2+, Ba2+). From the promising results obtained so far it is expected that substantial development of isotope palaeohydrology will occur in the near future, especially as detailed reconstructions cannot be easily drawn from marine deposits which have low sedimentation rates.

40Ar/39Ar THERMOCHRONOLOGY OF THE COLORADO RIVER EXTENSIONAL CORRIDOR, CALIFORNIA, USA, WITH IMPLICATIONS FOR LOW-ANGLE DETACHMENT FAULTS FOSTER, David A., Department of Geology, La Trobe University, Bundoora, Melbourne, 3083 Australia Lithospheric extension centered in the metamorphic core complexes of the Colorado River extensional corridor, SE California, USA, resulted in the juxtaposition of middle and upper crust during midTertiary time. ^^Aift^Ai analyses from the footwalls of the core complexes reveal rapid Miocene cooling and have implications for the dips of regional scale detachment faults (DF). Four hornblendes from mylonitized granodiorite beneath the Sacramento DF give ages between 22 and 24 Ma, which are interpreted to record the time of mylonitization caused by the onset of extension. Coexisting biotites and K-feldspars yield plateau ages of 17-18 Ma and age gradients from 15-20 Ma, respectively. Diffusion parameters from these Kfeldspars suggest that their age gradients correspond to cooling between -330° and -200'C. Seven hornblendes from Late Cretaceous and Proterozoic rocks from the footwall to the Chemehuevi DF give ages of 80-65 Ma. Ten biotites from this mountain range decrease in age from 65-17 Ma from SW to NE, in the direction of fault movement. Plateaux ages of ten K-feldspars which record cooling below 300-350°C also diminish from 52 to 19 Ma in the direction of deeper level exposures. Minimum ages of these K-feldspars with closure temperatures of 200-260'C decrease from 25 to 15 Ma towards the NE. From this down dip reduction in age we infer that when extension started in the Chemehuevi Mountains, rocks exposed in the SW and NE parts of the footwall were at -180' and -350°C, respectively. Assuming a geothermal gradient of 30 ± lO'C/km, suggests that the exposed part of the Chemehuevi DF initiated at a dip of 5'-30'. This estimate is supported by homblende barometry of Tertiary(?) plutons that yield emplacement pressures of -3.2 kbar. Preliminary results from the Whipple DF suggest a similar dip. Average post-Cretaceous cooling ratesfromthese three metamorphic core complexes were 5-10'C/Ma during the Early Tertiary and as high as 50'C/Ma during Miocene extension.

ISOTOPIC CHEMISTRY AND SEDIMENTOLOGY OF DISTAL BENGAL FAN SEDIMENTS FRANCE-LANORD. Christian, MTCHARD. Annie, C.R.P.G.,BP20, 54501-Vandceuvre, France, BOUQUILLON, Anne, U.S.T.L.F.A., 59655-Villeneuve d'Ascq, France, TIERCELIN, Jean-Jacques, U.B.O. 29287 Brest, France. Coarse and fine sediments from the 17 m.y. record of the Leg 116 Holes drilled in the distal part of the Bengal fan have been investigated for their mineralogy and Nd, Sr and O isotope compositions. Both the mineralogy and the combined isotopic compositions are compatible with an Himalayan origin for the major part oftiieinput. This conclusion is particularly well supported because the three isotopic systems are independent of each other. Except during the 6 to 1 m.y. period, the mineralogical assemblages are poor in alteration minerals. The very high sedimentation rates, the low S^^O («-12%o) and high ^'^Sifi^Sv values (« 0.740), unusual for detrital material, all suggest that the processes of erosion, transport and deposition were very intense and rapid. These characteristics are of course related to the dramatic development of the Himalaya where uplift rates reached 1000 m/m.y. The long history of sedimentation also impliestiiatthe Himalayan relief has existed at least since the middle Miocene. Whereas the mineralogical assemblages and the rate of sedimentation are variable, especially during the 6 to 1 m.y. period,tiieNd-isotoj^ signature, which is the most reliable witness of the origin, is remarkably similar throughout the studied 17 m.y. (eNdo« -15). This stability implies that the major source of eroded material remained the same during the 17 m.y. and therefore that the High Himalaya Crystalline or similar formations have always been tiie most exposed to erosion.

36 ISOTOPIC EFFECTS OF OPEN-SYSTEM METAMORPHISM, MORTON PASS, WYOMING FROST, Carol D. and FROST, B.R., Dept. of Geology & Geophysics, Univ. of Wyoming, Laramie WY 82071 USA In the contact aureole of the 1.44 Ga Laramie Anorthosite Complex (LAC) an interbedded sequence of Archean p e l i t e s , marbles and metabasites has been subjected to metamorphism at temperatures as high as 900C. At one l o c a l i t y 150 m from the contact with the LAC, hb-opx-cpx metabasites are strongly dehydrated where they abut marbles. Samples from a second l o c a l i t y of the same metabasite horizon 300 m from the LAC contact are at upper amphibolite f a c i e s . Comparison of the major and trace element geochemistry o f the two l o c a l i t i e s reveals considerable element mobility at granulite f a c i e s . The breakdown reactions o f hb to opx and cpx require open system f l u x of HzO-poor f l u i d outward from the marble and concurrent sodium l o s s . Whole rock Sm-Nd isochrons from both l o c a l i t i e s have been reset to the time of metamorphism, but during open-system dehydration at the higher grade l o c a l i t y Sm and Nd concentrations, Sm/Nd r a t i o s and i n i t i a l Nd i s o t o p i c r a t i o s are a f f e c t e d by p r e f e r e n t i a l l o s s of ^^^Nd and Nd over Sm. Rb-Sr i s o t o p i c compositions at the Ipwer grade l o c a l i t y are variably disturbed depending upon the extent of hornblende breakdown. Against the marble, the Rb-Sr isochron i s r e s e t to a date 300 Ma too young. These data can be explained by p r e f e r e n t i a l l o s s of ®^Sr and Rb from hornblende which results in an isochron with an i n i t i a l l y negative slope. At Morton Pass, knowledge of the time of metamorphism and the mineral reactions involved allows the i d e n t i f i c a t i o n of open-system e f f e c t s on i s o t o p i c systems. Both Rb-Sr and Sm-Nd whole rock systems are a f f e c t e d at temperatures equivalent to upper amphibolite and granulite f a c i e s . I f the i s o t o p i c data were interpreted according to closed-system models, i n c o r r e c t c r y s t a l l i z a t i o n and metamorphic ages, i n i t i a l Sr and Nd isotope r a t i o s and Nd model ages would be obtained.

Sub^ MTING aid REE GBDCHQflSIRY of the CSik BRAVA MAFIC-ULTEWMAFIC OGMEUX, GENHtAL BRAZIL. FUGI,M.Y.,(*) Kobe University, Kobe JP, NAKflMORA^., Kobe University, Kobe, JP, YAMflMiro.K., Kobe University, Kobe, JP, and NCHDA,S., Kyoto Sangyo University, Kyoto, JP. (*) Present Address: Bishimetal Exploration Co.,Tolqô, JP The Cana Brava conplex is the northmost among three metamoriîosed mafic-ultramafic complexes which make up a 300Kn long gabbroperidotite belt crossing the central region of Brazil in NNE direction. Understanding the origin of these ccn?3lexes is a must to decipher the evolution of the Precambrian basenent of this region, ^^ch is believed to include some of the oldest terrains outcropping in the Brasilian territory. In this wDrk, 20 sanples ranging from gabbros to pyroxenites and amphibolites of the Cana Brava ccraplex \jere analysed to major and trace (including REE) elenents. Among these, six samples were analysed for Nd and Sm isotopic compositions. An integrated analysis of major and trace elements suggests that all units of this conplex are oogenetic and fractional crystallization of a basaltic magma is a plausible model for the origin of this complex. MDreover, iôle-rock REE features ware found to be reproducible by closed-system fractional crystallization model calculations, Oiile REE features of mineral separates suggest that their parental magma was LREE-enriched. Four among six samples analysed for Sm and Nd isotopic canposifine an isochron of 19701 69 m.a., with initial d/"^Nd=0.510015t0.000073 and it \jas interpreted as the crystallization age of this complex. The other t w sanqiLes that do not lie on the isochron were interpreted in terms of isotopic heterogeneities of the parental magma. By assuming that these sanples are contemporary to the other four vMch define the isochron, the obtained initial Nd ratio were 0.509726 (E =-7.00) and 0.509601.(-9.55). These results put forvgard that, if CftLs complex was actually formed by crystallization of a magma inside a single chamber, then more than one injection are necessary to explain the isotopic as well as some petrologic features of this conplex. The negative values of ^ ^ were considered to be a result of contamination of the rising magma with old, acidic crustal rocks.

ICOG 7

ABSTRACTS

ACROSS BACK-ARC BASIN VARIATION IN THE Sr AND Nd ISOTOPIC COMPOSITIONS OF THE CENOZOIC ALKALI BASALTS FROM SW JAPAN AND EASTERN ASIAN CONTINENT FUJIBAYASHI. Norie. Dept. of Geol. and Mineral.. Hokkaido Univ.. Sapporo. 060, Japan. NAGAO. Takashi. Dept. of Mineral. Sci. and Geol., Yamaguchi Univ., Yamaguchi. 753. Japan, and KAGAMI. Hiroo, Inst, for Study of the Earth's Interior. Okayama Univ.. Misasa, 682-02, Japan Cenozoic alkali basalts are widely distributed in the eastern Asian area between the Japan Sea side of Japan and the eastern China. It is a controversial topic wether the alkali basalt magmatism correlates to subduction of the Pacific plate or to formation of the back-arc basin. Japan Sea. Sr and Nd isotopic data measured on samples from SW Japan provides an important constraint on this topic. The Sr for samples from the Japan Sea side is more radiogenic and the Nd is less radiogenic than for those from the Pacific side. This sapatial variation in Sr and Nd isotopic compositions makes a marked contrast with that shown by the NE Japan Quaternary arc volcanics. When the isotopic data are compiled for the alkali basalts from eastern Asia, the whole data set shows that most 'enriched' signature locates near the Japan Sea. The rocks become isotopically more 'depleted' with distance from the Japan Sea toward both the Pacific and continental sides. The alkali basalts from these areas mostly lie within the 'mantle array' on a plot of the ®'^Sr/®^Sr vs. ^'^^Nd/^'^^Nd. and may not be affected by crustal components. The isotopic variation of the Cenozoic alkali basalts discussed here is likely to be correlated to the mantle upwelling relating to rifting of the Japan Sea.

NEW K-AR AGES FROM METAMORPHIC ROCKS OF THE NEW ENGLAND FOLD BELT, EASTERN AUSTRALIA AND TECTONICS OF THE BELT FUKUI. Shiro and ITAYA, Tetsumaru, Hiruzen Research Institute, Okayama Univ. of Science, Okayama 700, Japan, WATANABE, Teruo, Department of Geology and Mineralogy, Hokkaido Univ., Sapporo 060, Japan and LEITCH, Evan C., Department of Applied Geology, Univ. of Technology, Sydney, NSW 2007, Australia New K-Ar mica ages from low and midium grade rocks provide evidence for three metamorphic events in the southern New England Fold Belt, eastern Australia. Glaucophane schist found within serpentine melange in the Pigna Barney region yields an age of 482Ma. the oldest radiometric age so far from New England. It is part of a 250km long belt of serpentine melange which includes blocks of Ordovician-Silurian limestone, and metabasalt/metagabbro, amphibolite and eclogite of unknown age. Some of the mafic rocks have REE patterns and Sm-Nd and Rb-Sr ratios that suggest primitive arc affinity whereas others have compositions similar to oceanic tholeiite. The belt probably comprises parts of an Early Paleozoic arc-trench system. Late Carboniferous high P/T metamorphism is indicated by ages of about 300Ma from the southern part of the Yarrowitch block. The protolith of these metamorphic rocks includes chert and mafic volcanics and they comprise the more deeply buried part of a subduction complex that formed at a well-established Devonian and Carboniferous convergent plate boundary. The age is that of the final accretion of the terranes that comprise southern New England. Ages of about 260Ma have been derived from the Tia Wongwibinda complexes and the Nambucca slate belt. The rocks are of low or intermediate P/T facies series and metamorphism accompanied inversion of a deep Early Permian rift basin and widespread terrane dispersion. It is possible that the increased heat flow indicated a result of the crustal extension that accompanied development of the rift basin.

ABSTRACTS

ICOG 7

NOBLE GAS ION IMPLANTATION INTO MINERALS T. Futagami and M. Ozima, Geophysical Institute, University of Tokyo, Bunkyoku, Tokyo 113, Japan. S. Nagai and Y. Aoki, Department of Research, Japan Atomic Energy Research Institute, Takasaki-shi, Gunma-ken 370-12, Japan. To understand better the implantation mechanism of solar energetic particles on lunar soils, meteorites, and other extraterrestrial materials, we carried out laboratory experiments, where slices of ilmenite and olivine (forsterite) were irradiated with ion beams of helium, neon, and argon at energies of 20, 40, and 200KeV. Implanted noble gases were extracted by stepwise heating in high vacuum, and their amount was measured with a mass spectrometer. The experimental results are the following: Noble gas degassing patterns were independent of crystal orientation, nor of roughness of the target surface. They were dependent of irradiation energy and of target material. Noble gases implanted at higher energy were retained up to higher temperatures. Noble gas degassing occurred over the wider temperature range for ilmenite than for olivine. Theoretical estimate on the penetration depth of noble gases into minerals shows that there is not much difference in the penetration depth between ilmenite and olivine. However, there was considerable difference in degassing patterns between them. We have at present no explanations for the discrepancy between the theory and observation.

INTERRELATIONSHIPS BETWEEN CONTINENTAL FREEBOARD, TECTONICS AND MANTLE TEMPERATURE. GALER, SJ.G., Max-Planck-Institut fiir Chemie, Postf. 3060, 6500 Mainz, FR Germany. The plate model of the oceanic lithosphere [Parsons, 1982] has been combined with parameterizations of melting by isentropic upwelling beneath ocean ridges [McKenzie & Bickle, 1988]. The volume of the ocean basins and thickness of the oceanic crust soderived permits the continental freeboard to be calculated by isostatic balance. Such calculations were performed numerically for a range of proposed plate tectonic models and masses of continental crust in the geological past. Ancient freeboard has been estimated by many authors using sedimentary facies and conditions at the peak of metamorphism, for example. From these studies it has been widely accepted that freeboard has remained nearly constant (conservatively ±0.5 km) from the middle Archean (-3.5 Ga) to the present. Such a balance depends upon the areal extent of the continents, the shallow mantle potential temperature (Tp), the oceanic lithosphere thickness and its rate of creation; the present continental mass decreases freeboard by -1.5 km. To some degree the parameters may be traded-off to maintain invariant freeboard. This is very sensitive to Tp, through the oceanic crustal thickness, but is insensitive to the oceanic plate creation rate. It is demonstrated that no reasonable freeboard balance exists if Tp in the middle Archean exeeded today's value by more than ~150°C. If a comparable amount of continental crust existed then, Tp could not have been more than ~50°C hotter. These results argue, in a plate tectonic regime, for a 'cool' mantie and <25% of the present continental mass existing at 3.5 Ga. Extensive recycling to the mantie of a once large Archean continental crust is compatible with Nd isotopic data [Armstrong, 1981], but cannot plausibly support constant freeboard. With regard to the Earth's thermal evolution, high post-accretional temperatures appear to have substantially decreased by the middle Archean. This suggests that heat transport and internal temperature (i.e., Nu-Ra) are strongly coupled during mantie convection. In addition, the high eruption temperatures of komatiites are probably unrepresentative of the shallow Archean mande.

THERMAL MODELING OF NORMAL FAULT BLOCKS WITH A CASE STUDY FROM THE BASIN AND RANGE PROVINCE GANS, PhiL, MILLER. Elizabeth, Geology Dept., Stanford U. CA 94305, BROWN. Rod, Geology Dept., LaTrobe U., Bundoora, Victoria 3083, HOUSEMAN. Greg, and LISTER. Gordon, Geology Dept., Monash U., Clayton, Victoria 3168 Extension of the upper crust is commonly accomplished by sets of parallel planar normal faults that rotate during progessive slip. We have modeled the thermal evolution of such fault blocks and investigated the effect of varying (a)initial fault dip, (b) fault spacing, (c) strain rate, (d) initial geothermal gradient, (e) rate of erosion, and (f) strain rates at deeper levels in the lithosphere. For realistic geometric parameters and strain rates, significant extension (6 >1.5) occurs on time scales that are slow with respect to the thermal time constant of the fault blocks (<100 Ka) but fast on the time scale of the lithosphere (50 Ma). Thus, fault blocks rotate through isotherms that remain subhorizontal but collapse by a factor of 6. P-T-t paths within the rotating fault blocks are strongly dependent on botfi the strain rate and the position within the fault block. Three general domains exist: Rocks between the midline of the fault block and the upper bounding fault undergo rapid uplift and cooling, points along the midline undergo isothermal decompression, and points between the midline and the lower bounding fault undergo slow heating and uplift We have developed an interactive program that illustrates, for a specified geometry and strain rate, P-T-t paths for any point in a rotating fault block. Spatial arrays of thermochronologic data from real fault blocks can thus be inverse modeled to constrain such parameters as the initial fault dip, magnitude of rotation, paleoslip rates, and paleogeothermal gradients. The Kem Mts.-Deep Creek Range is a 13 km-thick fault block that has been tilted 35-45° westward during Cenozoic extension. Detailed E-W and vertical dating transects in progess utilizing ^^Ar/^^Ar (muse, bio, kspar) and fission track (ap, zr) suggest onset of rapid extension and tilting at -35 Ma and continuing, perhaps episodically, until <16 Ma. Differential uplift and rotation of the fault block is best illustrated by mica and kspar ages which young in the downdip direction. Apatite age contours are nearly horizontal in the deeper levels of the fault block, demonstrating that rotation was nearly completed by the time these levels cooled through 100® C

LEAD ISOTOPE STUDY ON QINGCHENGZI AND ITS PERIPHERAL PB-ZN DEPOSITS, NE CHINA GAO. Z., Department of Geology, La Trobe University, Bundoora, Victoria 3083, LAMBERT, I.B., Elisian Resources , 2 Bonwick Place, Canberra 2605, SUN, S. S., Bureau of Mineral Resources, Canberra 2601, and GULSON, B.L., CSIRO, North Ryde. The Qingchengzi Pb-Zn de^sit, located in NE China, is hosted by the Lower Proterozoic Liaohe Group. The carbonates and argillaceous rocks which comprise the Group have been metamorphosed to amphiboUte facies. Proterozoic and Mesozoic intrusive activity is widespread in the region. Pb-Zn mineralization in the region occurs as conformable and vein type ore bodies. Lead isotope analyses were carried out on galena samples from the Qingchengzi and other smaller deposits in the region. Three types of lead isotope signatures were identifed: (a) Conformable lead: This type of lead with 206pb/204p5 ratios from 15.33 to 15.69 and a model age of about 1600 Ma, is confined to small conformable deposits marginal to the Qingchengzi deposit In tiie standard 207pb^04p5 vs 206pb/204p5 diagram, the ratios plot close to the growth curve. (b) Lead from mixed sources: This type of lead is confined to the Qingchengzi dejwsit. The 206pb/204p5 f^j. ^j^e conformable and vein-type ores range from 17.677 to 17.861 and 17.554 to 17.848 respectively. The model age (500 Ma) is much younger than tiie host rock. (c) Radiogenic lead: This type of lead is found at Zhaijiagou, a small vein type deposit north of Qingchengzi. The 206p5y204p5 ratios range from 22.52 to 23.47, yielding a future model age (Lin et al., 1982). The isotope data suggest that the lead in the Qingchengzi deposit was derived from dual sources, namely, lead from the conformable ores, and radiogenic lead from a source similar to the Zhaijiagou deposit. The lead in the region was most probably remobilized during the Hercynian magmatic event

CRUST AND MANTLE EVOLUTION OF THE EUROPEAN HERCYNIDES GEBAUER. Dieter, Laboratory for Isotope Geochemistry and Mass Spectrometry, Swiss Federal Institute of Technology, 8092 Zurich, Switzerland and WILLL\MS, Ian S., Research School of Earth Sciences, The Australian National University, Canberra, A.C.T. 2601, Australia The distribution of primary magmatic ages of detrital zircons from various Cambrian and Ordovician (meta)sediments of the European Hercynides (EH) is used to derive semi-quantitative estimates on crustal growth rates. After correction for rec>[cling, the existence of 4 Megacycles of crustal growth until the beginning of the Cambrian can be established by the SHRIMP data: 2.5 - 2.7 Ga, 2.2 -1.8 Ga, 1.2 - 0.9 Ga and 0.7 - 0.55 Ga. Using the same method, especially the two latter megacycles have also been detected in Ordovician fl)^sch sediments of SE Australia. Additionally, a very low activity of zircon-forming events around 1.5 Ga is typical for both the EH and SE Australia. Thus, an episodic character of crustal growth of Gondwana, at least for the scale of the sampled provenance, is indicated. Taking the few literature data for episodic crustal growth of the other supercontinent, Laurasia, crustal growth tends to be increasingly continuous the larger the scale and thus might be truly continuous planetarywise. For the Gondwana-derived crust of the EH the amount of pre-Panafrican crust is considerably higher when compared to estimates based on Rb-Sr data or Sm-Nd data. In agreement with the youngest detrital zircon grains, Rb-Sr model ages on (meta)sediments from the EH can be used as maximum depositional ages. Sm-Nd model ages, yielding ca. 1.8 Ga for the average age of the continental crust of the EH, probably are also too young as previous work did not take into account the considerably higher proportion of mafic Archean crustal rocks. For the same reason, a correction is necess^ for the zircon data assuming negligible amounts of zircons in mafic rocks when compared to felsic rocks. Applying such a correction to the SHRIMP data for the EH, ca. 50 % ot continental crust existed by the end of the Archean at 2.5 Ga, whereas only about 10% accumulated between 3.8 Ga and 2.7 Ga, in obvious contrast to the situation in Laurasia. Absolute crustal growth rates for the EH vary from virtually no growth to 16 km^ a-i and average ca. 3 km^ a-i for the Archean and ca. 2 km^ a'^ for the Proterozoic. Probable inherited mantle zircons formed 1(X) Ma to 3(X) Ma before major crust forming events, possibly in an upwelling mantle during oceanization when relatively littie continental crust was formed. AND Pb ISOTOPE STUDY OF THE STRINGER ZONE BENEATH THE HELLYER SULPHIDE DEPOSIT, TASMANIA GEMMELL. J.B. and LARGE, R.R., Centre for Ore Deposit and Exploration Studies, University of Tasmania, Hobart, Tas. and BARREIRO, B., Dept of Earth Sciences, Dartmouth College, Hanover, NTi, USA Underlying the Hellyer volcanogenic massive sulphide deposit is an extensive, well developed, and well preserved stringer zone. The deposit is hosted in the Middle Cambrian mafic Que-Hellyer volcanics. Fluids responsible for the precipitation of the massive sulphide flowed through two stages of syn-mineralisation veining that cross cut the pervasive footwall alteration. A detailed sulphur isotope investigation has shown distinct variations between vein stages in the stringer zone. Stage A quartz-pyritecarbonate veins have values that are isotopically heavier (5.7-11.1 %o, av. 9.9 %c) than the Stage B base metal sulphide-rich veins (3.8-8.9 %o, av. 6.9 %c). Stage A veins near the outer margins of the stringer zone are isotopically very heavy (S^'^S 11.9-40.7 %o, av. 25.0 %c). Average lead isotope values (^^^Pb/^^Pb, ^^^Pb/ô^Pb, ô^Pb/^^Pb) for Stage A and B veins are 18.425, 15.660, 38.367 and 18.394, 15.640,38.275, respectively. Lead isotope data for Stage B veins lie on the same trend as the sulphide ore, while lead isotope data for Stage A veins lie on a distinctly different trend. In a model explaining the sulphur and lead isotîc characteristics of the Heliyer stringer zone, Cambrian seawater (5 S -30 %o) entered the convection system and became progressively reduced. In the early stages of the hydrothermal system the proportion of reduced seawater sulphate was high (high water/rock ratios) giving rise to fluids with heavier b^^S values, as observed in the Stage A veins. Lead in these veins was derived from the immediate footwall sequence andesite volcanics. As the convection system intensified and scavenged deeper, rock sulphur began to dominate the hydrothermal fluid (low water/rock ratios), as seen in the Stage B veins. Lead in these veins was derived from clastic sediments and felsic Central Volcanic Sequence rocks, and possibly Precambrian basement, underlying the Que-Hellyer volcanics. This fluid was dominant in the formation of the Hellyer volcanogenic massive sulphide deposit.

ICOG 7

ABSTRACTS

POTASSIUM-ARGON DATING MORS : EXAMPLE FROM JUAN DE FUCA RIDGE (EAST PACIFIC). P-Y.Gilloti,andJ.Karsten2

1 Centre dcs Faibles RadioacUvitds (CEA-CNRS), F 91191 Gif sur Yvette 2 Hawaii Institute of Geophysic, Univ. of Hawaii, Honolulu HI 96822 USA

The feasability of potassium-argon measurements for the dating of young MORE has been explored using the Cassignol techniques (Cassignol and Gillot, 1982; Gillot and Comette, 1986) especially developed for the dating of young rocks. The problem that arises in the dating of subaquatic lavas is the possible inheritance of radiogenic 40Ar from the magma due to the pressure of the superimposed water column (Dabymple and Moore, 1968); the disturbing effect of that inheritance is accentuated by the very low K content of the MORE (0.02 to 0.2 %). Nevertheless, this excess of 40Ar is recognized essentially in the glassy crust of the pillows (Dahymple and Moore, 1968). 40Ar inheritance is also linked to the early crystallised phenocrysts. Ten samples from the Juan de Fuca ridge (east Pacific) were chosen as a function of their distance from the ridge axis to test this application of potassium-argon dating. Their depth of recovery is comprised between 2,000 and 3,000 m. The groundmass was selected (i.e. the phenocryst phase was eliminated by means of magnetic and heavy liquid separations); the samples were extracted from the core of the pillows, that is the part of the lava that slowly cooled and gently microcrystallized. The results, ranging from 0 to 3.5 Ma show a satisfactory correlation with distance from the ridge axis at the order of magnitude of 1 Ma. Within the last million year, it remains satisfacto^ for 5 (per 8) of the samples, within a reasonable range; i.e. it is within the range of the analytical accuracy which is typically ± 35,000 years (2s). The results which fall out may be caused by excess of argon; very tiny in the absolute (<10^ ^ at.g'^ exces 40Ar per g)it is proportionally important compared to the very low K content of the samples (<0.05%). These spurious results might also be related to complexities of the local tectonics. These results encourage us to pursue a systematic dating study on lavas sampled in the different sectors of the ridge in order to fix the conditions of this application of the K-Ar dating. APATITE FISSION TRACK 'STRATIGRAPHY' - PROFILES OF FISSION TRACK AGE AND TRACK LENGTH IN THE CRUST GLEADOW. Andrew and BROWN, Roderick, Victorian Institute of Earth and Planetary Sciences, Department of Geology, La Trobe University, Bundoora Vic 3083 Australia The limited stability of fission tracks in apatite under the temperature conditions found in the upper 3-4 km of the crust leads to progressive changes in apparent fission track age with depth. Temperatures in this region cause a progressive annealling of the tracks by shortening their lengths and consequently reducing the apparent age. On a geological time-scale this annealing occurs mostly within a temperature range of about 60°-110°C, although some annealing occurs even down to ambient surface temperatures. Quantitative modelling of the thermal annealing of tracks in apatite shows that distinctive profiles of apparent fission track age and length can be related to a number of simple thermal history styles. These may be compared to actual profiles observed in deep bore-holes or over a range of elevations in areas of topographic relief. The main types of profile obser\'ed tend to be either a linear decrease of apparent age with increasing depth, which relates to continuous uplift and denudation, or a concave-upwards profile produced by partial annealing in environments of tectonic stability or sedimentary accumulation. More complex thermal histories produce compound profiles which are essentially combinations of these two elements with the different segments often being separated by a pronounced inflexion point. Such an inflexion point is characteristic of a discrete episode of cooling. A distinctive 'boomerang'-shaped trend of track length with apparent age is characteristic of the progressive annexing of tracks through a partial annealing zone. Apparent fission track ages in apatites can thus define an invisible 'stratigraphy' in otherwise homogeneous rock masses that can provide quantitative constraints on the thermal and tectonic history. Examples of the major profile t>'pes have been identified in various geological environments with particularly important applications in tectonic studies, detecting low-grade thermal events and sedimentary basin analysis.

ABSTRACTS

ICOG 7

TEMPORAL VARIATIONS IN THE SOURCES OF ORE FLUIDS FOR MESOTHERMAL GOLD DEPOSITS S,D. GOLDING, Dept Geology & Mineralogy, Univ. of Queensland, QLD 4072, N.J. McNAUGHTON, & D.I. GROVES, Key Centre for Strategic Mineral Deposits, Univ. of Western Australia, WA 6009, R.J. BULTITUDE, Queensland Depart of Resource Industries, GPO Box 194, Brisbane, QLD 4001, S.G. PETERS, Central Norseman Gold Corporation, Box 56, Norseman, WA 6443 Mesothermal gold deposits occur in Archaean granitoidgreenstone terranes and in younger fold belts. In general, gold was deposited at 300 + 50 C and 1 to 2 Kb in quartz vein systems and shear zones in metamorphic rocks. Ore fluids were low salinity, near-neutral and relatively reduced, with variable CO^ contents. Despite the consistency in the physicochemical parameters of mesothermal gold ore fluids, there is significant debate regarding the sources of the auriferous fluids with models that favour crustal + mantle outgassing, mantle degassing-granulitization, magmatic devolatilization and deep circulation of meteoric waters. Calculated oxygen and hydrogen isotope compositions of Archaean gold ore fluids appear to preclude a contribution of meteoric water; they indicate the ore fluids were mainly metamorphic or juvenile in origin. The provinciality evident in the carbon and lead isotope data for hydrothermal minerals reflects source heterogeneities for these elements and appears to implicate a lower crustal source for some of the lead. Stable and radiogenic isotope data, together with geologic and fluid inclusion studies suggest that many Proterozoic mesothermal gold deposits are granitoidrelated, with evidence for a connate and/or metamorphic fluid component in some mineralizing systems. Ore fluids in younger mesothermal gold deposits have similarly enriched oxygen isotope compositions to the Precambrian gold deposits, but their fluid 6D values are variably depleted and other isotope data suggest that upper crustal sources predominated.

GEOCHRONOLOGY AND PETROGENESIS OF MORB USING 238u_230Th DISEQUILIBRIUM GOLDSTEIN, S.J., MURRELL. M,T. and JANECKY, D.R., Los Alamos National Laboratorv MS-J514, Los Alamos, NM, 87545, USA. DELANEY, J.R., University of Washington, Seattle, 98195, USA. GLAGUE, D.A., US Geological Survey, Menlo Park, GA, 94025, USA. A highly precise mass spectrometric method of analysis was used to determine 238u_234u_230xh in basalt glasses from the Juan de Fuca (JDF) and Gorda ridges. Direct dating of MORB was accomplished by measuring the decrease in excess with time in off-axis samples. In addition, initial "0Th/232Th ratios were used to examine the recent Th/U evolution of the mantle sources beneath these ridges. Initial ^^^Th excesses in the axial samples generally range from 12 to 40%. The influence of sea water alteration on U-Th disequilibrium was evaluated using and appears to be small (6234u=0); hence the ^^^Th excesses are believed to be magmatic in origin. 238u_230'pj^ ^^^^ increase systematically with distance from axis, and ages of 50 to 200 ka were obtained with uncertainties of 10 to 20 ka. For the Endeavour sepient of JDF, the full spreading rate is found to be 6 cm/yr with asymmetry in spreading between the two halves (4 cm/yr West and 2 cm/yr East). For North Gorda, a half spreading rate of 3 cm/yr was obtained. These spreading rates are in good agreement with those based on magnetic anomalies and topographic constraints. For both ridges, samples located near the rift axis are typically yoimger than predicted by these spreading rates which may indicate recent eruptive activity over a broad (2 to 3 km wide) area. Our initial ^^^Th/^^^Th results point to major differences in partial melting and Th/U in the sources of certain ridge segments.

HOW YOUNG IS THE MORB MANTLE? GOLDSTEIN, Steven L. and CHAUVEL, Catherine, MaxPlanck-Institut fur Chemie, Postfach 3060, 6500 Mainz, FRG. Pb isotopes in oceanic basalts generally fall to the right of the geochron, requiring an increase in U/Pb ratio (^i) sometime in the past. In MORB they form a linear array which has been interpreted either as an isochron (--1.7 Ga), or as a mixing line between HIMU and depleted mantle (DMM) reservoirs. Both inteipretations explain present day Pb isotopic compositions as resulting from ancient differentiation, and both are unconvincing. As an isochron, the Pb isotopes in MORB must reflect mantie-wide differentiation events at ' ^ I J Ga. Differentiation events between 1.5 and 2 Ga are also required to make a HIMU reservoir. If MORB is a mixture of HIMU and DMM, HIMU must be a very large mantle reservoir, yet it is only found in few oceanic islands. We suggest that MORB and HIMU sources have been affected by the same processes, and one of the effects is to increase U/Pb ratios. HIMU sources are the surviving remnants of ancient fractionation events, whereas Pb isotopes in MORB mainly reflect mixing of more recently made high U/Pb sources with mantle having lower 206pb/2(Wpb ratios. The MORB Pb isotope array can be generated by the same increase in [L as is necessary to make HIMU compositions, but with the increase occurring more recendy (0-1.2 Ga to account for the whole range of Pb isotopes in MORB). Thus the MORB Pb isotopic array need not be considered as an isochron or a HIMU-DMM mixing array. Rather, it reflects continuous differentiation and convective mixing in the upper mantle. The differentiation age of MORB sources (in this case the average timing of U/Pb increase) is indicated by the representative range of Pb isotopes in MORB rather than the slope of the MORB array, and is on average much younger than 1 Ga. This young differentiation age is consistent with geophysical modelling of the upper mantle which suggests fast average mixing times of much less than a billion years (e.g. Christensen, 1989).

ND ISOTOPES IN THE OCEANS: THE CIRCUM-ANTARCTIC CONNECTION. GOLDSTEIN, S. L., Max-Planck-Institut fiir Chemie, Postfach 3060, 6500 Mainz, FRG; ALBAREDE, F. and DAUTEL, D., CRPG-ENSG, Vandoeuvre, France. The earliest Nd isotopic studies of Mn nodules (O'Nions et al. 1978, Piepgras et al. 1979, Goldstein and O'Nions 1981) showed that Nd isotopic compositions within individual ocean basins are fairly uniform, but that they vary between different oceans basins (e.g. eNd =-2±2 (Pacific), -7.5±1.5 (Indian) and -12±2 (Atlantic)). The causes of these variations have not yet been adequately explained. It has been suggested that they reflect different continental inputs into different oceans, or alternatively, addition of Nd to seawater from weathering of basalt from the North Atiantic to the Pacific. The intermediary between all of the oceans is the Circum-antarctic current, which moves west to east around the earth. In order to examine the gross, relationships between water movement and inter-ocean Nd isotopic variations, we have analyzed nearly sixty Mn nodules from around the Circum-antarctic Ocean. The data fall into two groups, averaging e^Td =-8.5±l from South America eastward to Australia, and -6±1 south of the Pacific. The transition zones are sharp, with huge amounts of exchange-more than 50% of Circum-antarctic Nd is replaced by Pacific Nd south of the Tasman Sea, and a similar amount by Atiantic Nd east of the Drake Passage. These transitions occur where there is large deep water input. The variations in end between oceans do not reflect gradual changes between the Nonh Atlantic to the Pacific. The basinal characteristics form independentiy in the Atiantic and tiie Pacific, and reflect average continental input into the Atiantic and an extra, young, volcanic-derived input into the Pacific. Unlike the other oceans, the Circum-antarctic has no significant input of continental paniculates from rivers and winds, and appears to have a primary role in mixing Nd in the oceans. Its average value of eNd =-7.5 reflects the global oceanic average. The intermediate ejsM of the Indian Ocean compared to the Pacific and Atlantic reflects derivation of the Indian Ocean Nd primarily from the Circum-antarctic ocean, rather than direct mixing of Pacific and Atiantic waters.

ISOTOPIC CONSTRAINTS ON THE ORIGIN OF THE TOM STRATIFORM ZN-PB-BARITE DEPOSIT, MACMILLAN PASS, YUKON CANADA GOODFELLOW, Wayne, Geological Survey of Canada, 601 Booth Street, Ottawa, Canada K1A 0E8. The Tom deposit is hosted in Late Devonian Ijasinal carbonaceous shales and cherts near the eastern margin of the r'rft-controlled Selwyn Basin. The deposit consists of the West, East and Southeast zones which together comprise 15 million tonnes of ore averaging 7% Zn, 4.6% Pb and 49 g/tonne Ag. More locally, the Tom deposit occurs in sub-basins formed by subsidence along major synsedimentrary faults. Both the West and East zones are underlain by fault scarp breccias and are truncated to the south by a horst structure. Near the southern margins of the deposits, interiaminated barite, sphalerite, galena, pyrite and chert of the sedimentary-hydrothermal fades have been brecciated, veined and replaced by ferroan carbonate, sulphides and sulphosalt minerals. Sulphur, strontium and carbon isotopes were determined for hydrothermal minerals from the different ore fades and the endosing rocks. Values of S^^S in barite are near Middle Devonian seawater values. Sulphur isotopes in laminated pyrite define an evdutionary trend of increasing values which range between 6 and 8°/oo through the ore zone, and reach a masimum value of 30 °/oo in the overlying Frasnian chert. Values for galena in the sedimentary-hydrothermal fades are comparable to pyrite, but incarease to values of 12 to 14°/oo in coarse grained galena from the replacement vent complex. Strontium isotope ratios in barite and carbonate are more radiogenic than Late Devonian seawater, and decrease systematically upward within the ore zone and away from the hydrothermal vent complex. Variations in ®^Sr/®Sr ratios are controlled by the mixing ratio of hydrothermal fluids with ambient seawater. Sulphur isotopes in the Tom deposit indicate two major sulphur sources. Most of the sulphur in bedded sulphides originated from an ambient, H2Srich water column whereas sulphur in the vent complex was derived from barite sulphate inorganically reduced to sulphide during reaction with lowhydrothermal fluids. This is supported by replacement textures and isotopically negative values in ferroan carbonate from the vent complex which suggest that most of the carbonate was derived from organic compounds that were oxidized to carbonate during sulphate reduction.

U-PB STUDY OF CHONDRITIC PHOSPHATES C.Gopel, G. Manhes,CJ. Allegre ( Lab. Geochimie & Cosmochimie, 4, Place Jussieu, 75252 PARIS CEDEX 05) The first planetary bodies formed at 4.55 JE and underwent thermal processes shortly after their formation: - fusion and differenciation (achondrite and iron meteorite parent bodies) or smaller heating (up to 11(X)°K) in the case of the parent body of ordinary chondrites. The study of this thermal evolution allows to precise the condition of the formation of these first objects and to characterize the effective heating source(s). Radiometric data indicate that this thermal evolution lasted 10^ years and that the size of the parent bodies ranges from 100 to 300 km. The data are also in agreement with an onion shell structure of these parent bodies, the most metamorphosed meteorites formed in the core, surrounded by sucessive layers of less metamorphosed type 5 through type 3 material. We started an U-Pb investigation of phosphates from H-, L- and LL chondrites. Phosphates are an acessory mineral phase in ordinary chondrites and formed during metamorphism. Enriched in U compared to the bulk meteorite, they exhibit today very radiogenic Pb isotopic compositions, allowing very precise age determinations (± IxlO^y)- Three types of information can be obtained: 1) - within a class of chondrites (particulary H- chondrites) different closure time of the phosphates of different petrologic t>'pes allows to estimate the time intervall of metamorphism, the cooling rate and by that the size of the parent body, 2) - comparaison between the phosphates of type 6 chondrites from different classes (H, L, LL) indicates if the pic of metamorphism happened at the same time and 3)- age determinations of type 6 phosphates may allow to distinguish between accretion metamorphism and metamorphism by internal iieating, the first one characterized by older ages compared to younger ages measured of material originating at grater depth. We analysed the U-Pb systematics of phosphates which were separatedfromtype 5 and 6 meteorites. H5 phosphates are 20-50 xlO^ y older than H6 phosphates, this is in agreement with an onion-shell model. A great age difference between phosphates of H6 and LL6 meteorites was found, H6 meteorites being younger, this is difficult to reconcile with a model where the LL- parent body is much larger than the H- chondrite parent body.

ICOG 7

ABSTRACTS

Rb-Sr STUDY OP MULTI-STAGE ILLITE EVOLUTION IN THE LOWER CAMBRIAN CLAYSTONES I,M,GOROKHOY. Inst.Precambr.Geol.Geochron. 15^054 W i n ^ d , USSR, N^CLAUER, Centre Geochim.Surface, 67084 Strasbourg, France, T.LoTURCHENKO, N^N^MELNIKOY, E.P.KUTYAVTN, A.Y.BASKAKOV, Inst.Precambr.Geol.Geochron. 199034 Leningrad, USSR, C.BONNOT-COURTOIS, Labor.Geomorpli., Dinard, France, E.PIRRUS, Inst.Geol,, 200101 Tallinn, USSR <2/km fractions from the lower Cambrian "blue clays" of Estonia and province of Leningrad contain three illite generations. Detrital 2M illite particles have sharp borders sometimes rolled up and are most abundant in 0.6-2^111 fraction. This illite I has low values of the crystallinity index usual for epi- and anchizones. Its minimum age is of 722 Ma. The 0#1-0,2Am fraction includes authigenio 1M illite II formed under closedsystem conditions during burial of the sediments 533 Ma ago. High values of the crystallinity index typical of a zone of diagenesis and lath-like particles are specific to illite II. Intermediate fractions involve mixtures of illites I and II. IM(IMd) illite III observed in the finest <0p1/,m fraction is also authigenic. Its mineralogical and morphological properties are similar to those of illite II. The foimation of illite III was induced by the non-marine subsurface water influx due to faulting or regional emergence whicKtook place 430-480 Ua ago under open-system conditions. The three illite generations were crystallized in the environments with different initial Sr isotopic ratios which are of 0.7100, 0.7137 and 0.7109 for illites I, II and III respectively.

MAGNESIUM ISOTOPE SYSTEMAHCS IN REFRACTORY INCLUSIONS FROM THE CVS OONDRTIE EFREMOVKA GOSVVAMI J.N. and SRINIVASAN, G., Physical Research Laboratory, Ahmedabad 380009, India, and ULYANOV, A.A., Vernadsky Institute of Geochemistry and Analytical Chemistry, Moscow 117975, USSR. The calcium-aluminium-rich inclusions (CAI) in the CVS chondrite Efremovka are characterized by the absence of secondary alteration products, prevalent in the AUende CAIs, and may represent more pristine solar system material available for laboratory analysis. Magnesium isotopic compositions of different mineral phases in a set of petrographically documented Efremovka CAIs have been determined by the ion-probe technique. Measurement precision of<2%o (2cr^) could be achieved during analysis of terrestrial standards and isotopically spiked anorthositiLc glasses. Studies of non-linear effect in Mg-26 as weU as intrinsic massfractionation (defined as the difference in A (25) between sample and standard) have been completed for a set of coarse grained refractory inclusions. Meimte-rich (Type Bl) CAIs from Efremovka show weU behaved Mg-Al isochron with initial (Al-26/Al-27)o close to the canonical value of A/5 X 10"^. The magnesium isotopic compositions in spinels from different inclusions suggest differences in the magnesium isotopic composition (Mg-26/Mg-24) of the source material with positive excursions of upto 2%o ( (^m = 0.4%o). Two coarse-grained inclusions with similar Mg-Al isochron (within experimental uncertainties) showed a nearly reverse trend in their magnesium isotopic mass fractionation as one moved from the core region to the boundary of the inclusions. Complex evaporationdistillatLon processes need to be invoked to explain the observed magnesium isotopic systematics in the refractory inclusions from Efremovka.

ABSTRACTS

IC0G7

HELIUM ISOTOPE GEOCHEMISTRY OF THE AUSTRALIAN-ANTARCTIC DISCORDANCE GRAHAM. David, LUPTON, John, Dept. of Geological Sciences, University of California, Santa Barbara, CA 93106 KLEIN, Emily, Duke University, Durham, NC 27708 CHRISTIE, David, PYLE, Doug, OSU, Corvallis, OR 97331 The Australian-Antarctic Discordance is a tectonically complex region of the ocean basin south of Australia, which includes a deep portion of the Southeast Indian Ridge and numerous fracture zones. Pb, Sr, and Nd isotopes in basalts from a 1700 km-long segment of the ridge centered on the AAD previously revealed the presence of two distinct isotopic groupings. Basalts to the west of 124®E longitude have Indian Ocean affinities, with more radiogenic Sr and less radiogenic 206p5/204p5^ ^j^jig ^Q jh^ ^^st of 127°E have Pacific isotope characteristics. We have analyzed the same samples for ^He/'^He, He and Ne concentrations, plus an additional suite of closely-spaced samples from within the AAD. He and Ne were extracted by in vacuo crushing. Analytical precision for ^He/^He is ±0.5% (2a). ^He/'^He ratios show a change in isotope characteristics at the location of the previously identified Pb, Sr and Nd isotope boundary. "Indian" samples have lower and more variable ^He/^He ratios, between 6.3-7.3 Ra (Ra is the atmospheric ratio of 1.39x10-6). In contrast, "Pacific" samples have higher ^He/^He ratios of 7.3-7.6 Ra and, like Sr and Nd, show a narrow range of values. All ^He/^He ratios are at the lower limit of the MORB range. The "dynamic range" in He isotope ratios (i.e., observed variability relative to the analytical precision) is about 35, comparable to that for 87Sr/86Sr (0.7025-0.7035) and about onehalf that for 206pb/204p5^ in detail, ^He/^He is inversely coirelated with ^^Sr/^^Sr in the Indian samples. The inverse correlation is remarkably linear (r= -0.99, n=10), with Pacific samples defining the most depleted compositions. Extension of the array to radiogenic He and Sr isotope ratios suggests that an enriched mantie component, such as that found at the islands of Kerguelen, Tristan da Cunha and Gough, contributes to the regional composition of the upper mande.

THE THERMAL SENSITIVITY OF FISSION TRACKS IN APATITE, ZIRCON AND SPHENE - A REVIEW PAUL F. GREEN. Geotrack International. P.O. Box 4120, Melbourne University, Victoria 3052, AUSTRALIA Because of the sensitivity of fission tracks in apatite to temperamres of ~100°C or less over geological timescales, the thermal resetting of apatite fission track ages can be direcdy studied in deep well sections which attain such temperatures. This, combined with the ease with which the processes causing the thermal resetting can be studied in the laboratory, provides arigorous,quantitative understanding of the thermal sensitivity of fission tracks in apatite, from which the magnitude andtimingof maximum paleotemperatures in the range 20-130°C can be determined with a precision of ~10°C. Extrapolation of a laboratory-based kinetic description of fission track annealing in an apatite of average composition can readily explain annealing effects in apatites of similar composition from geological situations in which thermal histories are well constrained. Observations of differential annealing in samples containing apatites with a spread of compositions, and laborator}' experiments on apatites of different composition have clearly shown that tiie F/Cl ratio exerts the dominant control, with F-apatites more sensitive than Cl-apatites. Experiments are in progress to quantify this effect. In distinction to most methods of geochronology, open-system behaviour of the AFTA system is common, and provides the basis of application of the technique. Apatite fission track ages can rarely be interpreted in terms of "closure temperamres", and such an approach may lose significant information on the thermal histoty of a sample. Fission tracks in zircon and sphene are sensitive to higher temperatures, around 200-300°C. However the thermal sensitivities of these systems are less well understood in detail, because of the difficulties in studying these systems in the laboratory, and the lack of data from deep wells that reach present temperatures of -'200°C or more. The best available constraint on the stability of tracks in zircon comes from continually cooled sequences where zircon data can be set in the context of other systems, while few data are available for sphene. Compositional influences appear to affect zircon, although controls have yet to berigorouslyidentified. Thus in anyrigoroussense, the thermial stabilities of the zircon and sphene fission track systems must be regarded as uncertain, at present.

CORRELATED SR, ND AND PB ISOTOPIC SYSTEMS AND CHEMICAL COMPOSITION IN BASALTS OF THE NEWER VOLCANICS, VICTORIA, AUSTRALIA GRAY. C.M., La Trobe University, Melbourne 3083, Australia, TATSUMOTO, M., and LUDWIG, K.R., USGS, Denver, Colorado, U.S.A. and PRICE, R.C., La Trobe University. The Newer Volcanic Province of Victoria (15000km2) is dominated by Pliocene to Holocene plains lava flows of a basaltic icelanditetholeiite-transitional hawaiite-hawaiite association that has become progressively more alkalic with time. 87Sr/86Sr for -450 samples outlines eastern (radiogenic) and westem (less radiogenic) geographic sectors with distinctiy different maximum and minimum isotopic ratios. The sectors are separated by a Palaeozoic basement tectonic boundary indicating a mantle lithospheric influence on basalt composition. Overall, 87Sr/86Sr declines withtimein both sectors, has a broad positive correlation with Si02%, and a distinctive variation with Sr and other incompatible element contents. A 143Nd/l^Nd - 87Sr/86Sr diagnmi has separate eastern and westem linear tholeiite trends which lie within the 'mantle array' and converge to a common modem hawaiite point. The coherence of the trends argues for a mixing origin. The majority of Iherzolite inclusion data from the province fall in the triangular shape between the trends. On a 207pb/204pb. 206pb/204pb diagram data are resolved into adjoining eastem and westem groups both of which increase in 206pb/204pi3fromolder tholeiitic to younger alkalic rocks. All three isotopic systems have similar temporal and compositional behaviour and Sr and Nd isotopic compositions are directly correlated. In addition, the Pb system can be inte^ted graphically in a simple fashion by recognition of the identical geometry of data in 87Sr/86Sr - Sr ppm and 207pb/204pb. 206pb/204pb diagrams. Hence, isotopic systems and chemical compositions of all plains basalts are inter-related. The petrogenesis of the basalts is described by the interaction of at least six components. Only one of these components, which has hawaiitic character, is independent of lithospheric structures and is considered to be of asthenospheric origin. The remainder of the component are confined to one sector and are thus of mantie lithospheric origin.

TIME SCALES FOR BOUNDARY EVENT CHANGES IN THE 1«0/1«O OF SEAWATER GREGORY, R.T. Department of Geological Sciences, Southern Methodist University, Dallas, Texas 75275 Oxygen isotopic data from hydrothermally-altered oceanic pillow lavas suggest that the value of seawater has not changed very much over geologic time. However, several studies on sedimentary roc& attribute 3-6 per mil changes in the values of carbonate rocks to changes in the «0 of seawater. These changes appear to occur over very short time intervals (<20 Ma), and therefore, it is important to examine the characteristic times associated with primary processes that may affect the isotopic composition of seawater. Material balance equations describing the secular change in the »0 of seawater resulting from isotopic exchange between the lithosphere and the hydrosphere have solutions that give the steady-state value of seawater as: E ki(6f - Ai)/S k -,, where the A| are the bulk fractionation factors between the various rock reservoirs (i) and water, the 6-; are the initial»O values of the various rock reservoirs, the k values are the rate constants in units of 1/time, and E kj the total rate. For a global ridge system with a spreading rate of 3 km^ /yr and a 4 km depth of seawater penetration, the k value for exchange between the oceanic crustal reservoir and seawater « 0.008/Myr with a mean lifetime (1/k) of 125 Ma. Similarly, a weathering rate of 2.5 km^ /yr yields a k value for isotopic exchange between seawater and the continents due to weathering of 0.002/Myr with a mean lifetime of 420 Ma. Even for high estimates of continental weathering rates («7 km® /yr), the mean life for the weathering process is still 150 Ma. By these mechanisms, a 6 per mil change in the »0 of seawater requires 150 km^/w of continental weathering which would occur over » 15 Ma years at the expense of processing about a third of the contmental crust! Clearly, such changes m the composition of the oceans should be reflected in dramatic changes in the tectonic regime of the Earth. From this analysis of the time constants involved in the global cycles, apparent boundary event changes in the values of carbonates must be explained by other mechanisms.

42 FIRST EVIDENCE FOR ARCHAEAN ROCKS AND EARLY PROTEROZOIC HIGH-GRADE METAMORPHISM WITHIN THE WEST EUROPEAN HERCYNIAN BELT : THE BAY OF BISCAY GRANULITES GUERROT, C., CAESS, Institut de Geologie, University de Rennes, France and Isotope Geochemistry, ETH, Zurich, Switzerland, and PEUCAT, J.J., CAESS, Universite de Rennes, France. Dredge hauls and diving surveys have shown the existence of granulite facies rocks on both sides of the Bay of Biscay (Western France and Northern Spain). Sub-concordant U-Pb results on zircons and monazite data divide these rocks into two groups of distinct age: Archaean (2.7 Ga) and early Proterozoic (1.9 Ga). The early Proterozoic age is interpreted to refer to a magmatic event essentially contemporaneous with high-grade metamorphism. Similarly, the Archaean age is associated with crustal growth probably also linked to a period of high-grade metamorphism. Whereas the zircons from the Archaean samples do not show any disturbance related to early Proterozoic metamorphism, Rb-Sr and Sm-Nd isotopic results on minerals (clinopyroxene - garnet - plagioclase - ilmenite) from one Archaean sample prove their involvement in the 1.9 Ga high-grade event. Isotopic studies (Sr-Nd) on whole rocks indicate that the early Proterozoic magmas have been variably contaminated by an old component, which could be represented by the Archaean samples. It is important to point out that for the first time Archaean rocks have been identified in the Western European Hercynian belt and that an early Proterozoic high-grade metamorphic event has been proven to exist; all other granulitic rocks known so far in Western Europe have a Palaeozoic age for metamorphism. The ages reported by this study are similar to those of inherited zircons from all over Europe and allow for comparison with the West African craton.

G E N E R A T I O N S O F G O L D M I N E R A L I Z A T I O N IN T H E R E D LAKE AREA, NORTHERN ONTARIO, CANADA G U L S O N , B . L . and M I Z O N , K . J . , C S I R O D i v i s i o n of E x p l o r a t i o n G e o s c i e n c e , N o r t h Ryde NSW 2113, Australia, ATKINSON, B., M i n i s t r y of N o r t h e r n D e v e l o p m e n t , R e d L a k e Ontario, Canada. T h e R e d L a k e area is a m a j o r p r o d u c e r f r o m Archean greenstone lode gold deposits. It was t h e m o s t s u i t a b l e s i t e t o e v a l u a t e t h e u s e of P b i s o t o p e s in g e n e t i c ( s o u r c e , s u l f i d a t i o n m e c h a n i s m ) and e x p l o r a t i o n a s p e c t s of A r c h e a n g r e e n s t o n e l o d e g o l d d e p o s i t s b e c a u s e o f : (i) good stratigraphic and structural control w i t h h i g h - p r e c i s i o n z i r c o n U - P b a g e s ; (ii) t h e - 3 0 0 Ma f o r d e v e l o p m e n t of t h e b e l t ; and (iii) o c c u r r e n c e of b o t h m a j o r a n d m i n o r d e p o s i t s in r o c k s of v a r y i n g a g e s a n d c o m p o s i t i o n a n d in d i f f e r e n t d e f o r m a t i o n z o n e s . F r o m o v e r 350 a n a l y s e s of r o c k s , s u l f i d e s and gold there appear to be at least two g e n e r a t i o n s of g o l d f o r m a t i o n . T h e f i r s t , an o l d e r o n e f o u n d in t h e m a j o r C a m p b e l l / D i c k e n s o n ( A . W . W h i t e ) m i n e s is b a s e d o n t h e low 6/4 v a l u e s (13.1) f o u n d in s u l f i d e s and coe x i s t i n g g o l d . M i n e r a l i z a t i o n d a t a lie o n t h e s a m e a r r a y s for t h e - 3 0 0 0 M a old m a f i c a n d felsic volcanics and the simplest interpretat i o n w o u l d b e for a c o m m o n v o l c a n i c / m i n e r a l i z a t i o n s o u r c e . T h e s e c o n d is a y o u n g e r e v e n t a t - 2 7 0 0 M a b a s e d o n a s e r i e s of P b - P b a r r a y s for s u l f i d e s , and m a g n e t i t e from U - P b dated 2720 M a s t o c k s (Dome and M c K e n z i e in t h e e a s t ; R e d C r e s t in t h e w e s t ) a n d , U - P b a n d Pb-Pb ages for highly radiogenic arsenopyrite from Balmer Lake (F. C o r f u , pers. c o m m u n . , 1 9 8 7 ) , a n d m o d e l "ages" f o r g a l e n a a n d Pbr i c h s u l f i d e s . C u r r e n t i d e a s on g o l d f o r m ation place it at 2700-2720 M a .

ICOG 7

ABSTRACTS

I S O T O P I C V A R I A T I O N S IN G A B B R O I C INTRUSIONS PERFORATING THE PROTEROZOIC CONTINENTAL C R U S T , BAMBLE, SOUTH NORWAY: IMPLICATIONS FOR THE E V O L U T I O N OF THE U N D E R L Y I N G M A N T L E . HAAS. G.J.L.M. de, Institute of Earth S c i e n c e s , U n i v e r s i t y of U t r e c h t , P . O . B o x 8 0 . 0 2 1 , 3508 T A U t r e c h t , T h e N e t h e r l a n d s ; VERSCHURE, R.H., Isotope Geology Laboratory, Free University, de Boelelaan 1085, 1081 HV Amsterdam, The Netherlands. R b / S r a n d S m / N d s y s t e m a t i c s of s e v e r a l g a b b r o i c intrusions from the Bamble amphiboliteg r a n u l i t e t r a n s i t i o n z o n e , w h i c h is t h o u g h t t o r e p r e s e n t a p a r t of t h e P r o t e r o z o i c lower continental crust, have been studied. The area suffered metamorphism during the Kongsbergian (1600-1500 M a ) a s w e l l as t h e S v e c o n o r w e g i a n (1200-950 M a ) E v e n t . Sm-Nd age determinations yielded preK o n g s b e r g i a n a g e s for s o m e g a b b r o s a s w e l l a s S v e c o n o r w e g i a n a g e s for o t h e r s . T h e i n i t i a l e^^ values seriously deviate from the postulated Nd e v o l u t i o n c u r v e for t h i s p a r t of t h e B a l t i c mantle, which suggests that the upper mantle w a s e s s e n t i a l l y u n d e p l e t e d u p t o » 1.7 G a . T w o o l d (al.7 G a ) as w e l l as t w o , a s s u m e d , y o u n g (al.l G a ) g a b b r o s h a v e i n i t i a l e^^ of + 5 . 5 . A y o u n g g a b b r o (dated a t 1.1 G a ) , t h e m o s t p r i m i t i v e of a l l g a b b r o s u n d e r consideration, h o w e v e r h a s an i n i t i a l e^^ of + 1 . 8 , p l o t t i n g close to the postulated c u r v e . These results point towards varied depletion of t h e s u b - B a m b l e m a n t l e of w h i c h t h e o n s e t t o o k p l a c e b e f o r e 1.7 G a . R b / S r d a t a of t h e i n t r u s i o n s w i l l a l s o b e reported.

THE ISOTOPIC EVOLUTION OF FOSSIL PLUMES HALLIDAY, A L E X N., Dept of Geological Sciences, U. of Michigan, Ann Arbor, MI, 48109 USA, and FITTON, J.GODFREY, Grant Institute of Geology, University of Edinburgh, Edinburgh, EH93 JW, UK. The recent (<10Ma) basaltic volcanism along the Clameroon line displays an increase in maximum ^o^pb/^wpb and 208pb/204pb to t>'pical HIMU values with proximity to the continent / ocean bound^. However there is no systematic variation in 207pb/204p5 with distance. The data form a series of sub-parallel Pb-Pb arrays or "isochrons" each of which is represented by the data for an individual volcanic centre. The offsets in 206pb/204pb and between each volcanic centre are therefore explicable in terms of relatively young fractionations in U/Pb and TTi/Pb ratios in the upper mantle. A possible explanation fot this is that these variations in Pb isotopic composition refelect variations in parent/daughter ratio produced in the upper mantle at c 125 Ma as a result of migration of very small melt fractions. We propose that the relationship with the continent/ocean boundary reflects the effect of a plume or thermal which was instrumental in the formation of the continental margin at this location at 125 Ma. As the plume head cooled from the margins, low degree melt fractions were trapped progressively in toward the centre. This process resulted in a zoning of mantle enrichment with the most enriched compositions at the centre. Supporting evidence for trace elementfractionationby small melt fractions in the sources of Cameroon line basalts can be found in the general level of alkali enrichment and in the fact that there is a positive correlation between Rb/Ba ratio and Rb concentration, which is contrary to that found in MORE for example. The results of this study suggest that enrichment via melts can account for at least some of the characteristics of HDv4U basalts including displacement below the Nd-Sr "mantle array" without resorting to recycling models or advocating CO2 metasomatism.

ABSTRACTS

IC0G7

THERMAL HISTORY, ILLITE GROWTH AND K-Ar AGES FOR JURASSIC SANDSTONES, VIKING GRABEN, NORTH SEA HAMILT0N:P.J.. CSIRO Div of Exploration Geoscience,POB 136, N Ryde,NSW 2113, Australia, GILES,M., Shell UK, Shell Mex House, Strand, London WC2R ODX, England and AINSWORTH.P., SURRC, Glasgow G75 OQU, Scotland. K-Ar dating is used extensively to assess timing of illite diagenesis relative to the time of hydrocarbon charge to petroleum reservoirs. This application has been assessed for Brent sands from the E. Shetland Basin, Viking Graben, N. Sea by integration of age data with burial history curves, quartz overgrowth fluid inclusion temperatures and petrographic observations . In summary:1) Youngest illite ages from "clean" sandstones are less than those from nearby "dirty" sandsones. The latter reflect either contamination with older detrital illite or bias towards earlier illite diagenesis as a result of easier nucleation on pre-existing clay. 2)0il zones in "clean" sandstones have illite ages for finest size fractions that generally match calculated tinges of oil migration and approximate the cessation of illite formation 3)Neoformed illite is observed to first appear at ca. 2300m depth, equivalent to a temperature of about 80' C.The corresponding time is lens Ma greater than the youngest illite ages — a time span significantly greater than the measured age difference between coarsest and finest fractions of an illite separate. 4)Illite, and by inference quartz, diagenesis can persist at reduced rates in water wet portions of oil zones after oil accumulation. High fluid inclusion temperatures that have been suggested to record the action of abnormally hot fluids prior to oil migration are more appropiately interpreted as reflecting maximum burial conditions of the reservoir. A model for illite growth is proposed that explains some of the above observations together with the fact that none of the K-Ar data provides evidence of recent diagenesis (<15 Ma). An understanding of the importance of fluid flow rates in competition with illite forming reaction rates suggests that the aggregate nature of size fractions is always biased towards sampling of grains formed at the maximum rate of growth.

TEMPERATURE HISTORY IN DANISH SEDIMENTARY BASINS AS REVEALED BY THE HSSIGN TRACK METHOD.

DYNAMIC EVOLUTION OF THE ICELAND HOT MANTLE PLUME: A PB ISOTOPE PERSPECTIVE B. B. HANAN. San Diego State University, San Diego, CA 92182, U.S.A., and J-G. SCHILLING, University of Rhode Island, Kingston, RI 02881, U.S.A.

O-ISOTOPE CHARACTER OF THE LOWER CRUST: NORTH EASTERN AUSTRALIA AND WESTERN EUROPE HARMON. R.S. NERC Isotope Geosciences Laboratory, Keyworth, UK and FOWLER M.B., Department of Geology, Oxford Polytechnic, Oxford , UK The granulite-facies xenoliths occasionally carried to the earth's surface in mafic lavas provide important information about the composition, age, formation, and evolution of the lower crust. Oisotope ratios have been determined for four well-studied xenolith suites from NE Australia (Chudleigh & McBride) and W. Europe (Eifel & Massif Central). The Chudleigh (C) and Eifel OE) xenoliths are mafic compositions considered to represent crystallized mantlederived magmas and/or their derivative crystal cumulates. The McBride (MB) and Massif Central (MC) xenoliths are more compositionally heterogeneous, ranging from mafic to felsic lithologies considered to have been derived from a variety of protoliths including mafic melts and cumulates, mafic restites, felsic melts, and metasediments. Measured (%o SMOW) ranges are: C = -h6.2 to +7.0, E = +5.9 to +8.7, MB +8.0 to +13.2, and MC = +6.2 to +11.0. The three most salient features of the xenolith Oisotope data are: (i) the large (>7 %o) overall variaiton in whole-rock 180/160 ratios, (ii) the distinct difference in character between igneous and metasedimentary lithologies, and (iii) the generally high-^^0 nature of the mafic igneous granulites relative to typicd peridotite upper mantle. Given the similarity of this xenolith ^Ô data to that for other localities worldwide, these 0-isotope characteristics are likely widespread and representative features of the present-day lower crust. The variable ^ ratios indicate that the 0-isotope character of the deepest continental crust has not been modified extensively by pervasive interaction or equilibration with fluids of either external or internal derivation that would have tended to homoginize ^Ô/'Ô ratios on a large scale. For the four xenolith suites studied, enrichments and 0-isotope variability are correlated with radiogenic (Sr, Nd, & Pb) isotope variations. Higher ^Ô/^Ô ratios are observed for xenolith with higher 87Sr/86Sr and 206pb/204pb ratios and lower i43Nd/i44Nd ratios suggesting extensive interaction of mantle-derived magmas with more ^Ô-rich and isotopically evolved old crust during episodes of magmatism that produce new, mafic lower crust.

Pb Isotope data for Iceland Tertiary basalts from two distinct paleorift zones spanning 14.1-2.2 Ma and from the Reykjanes Ridge and Neovolcanic zones of Iceland (0.7-0 Ma) require 3-component mantle source mixing in relative Pb triangular space. The components are the 206pb nch plume, the 207p5 nch asthenosphere, and 208p5richcontinental derived lithosphere material. Fixed mass proportions between the 207p5 and 208pb components are observed at 4 distinct periods of time (14.1-13.7 Ma, 13.7-7.5 Ma, 7.5-2.2 Ma and 0.7-0 Ma) resulting in 4 distinct, but converging, mixing arrays that help in constraining the Pb isotopic composition of the 206pb rich plume component. Apparently the Iceland plume has supplied material of relatively constant Pb isotopic composition throughout Iceland's 15 my history. The Iceland plume component shows maximum influence at 7.5 Ma, coincident with a maximum in volcanic production rate on Iceland and along the Reykjanes Ridge. The 208p5 rich component is kindred to that of the Hebridean lithospheric keel (Menzies et al., 1987). The ^OSpb rich component has a maximum influence at 13.7 Ma and decreases thereafter. High ^He/^He recorded in modern Iceland basalts suggests a plume derivation from the lower mantle. The observed cyclic Pb isotope variations obser\'ed for the Tertiary Iceland basalts suggest a model where a chain of blobs rich in 206p5 rise through and interact with a ^O'^Pb rich asthenosphere which may have been contaminated by some delaminated continental lithosphere during the opening of the N-Atlantic. Alternatively, entrainment of the 208pb rich component could have taken place during ascent of the plume through the 670Km discontinuity where recycled continental material may have accumulated, or for that matter at the core mantle boundary.

10, DK-1350 Kobenhavn K, Denmark,' JENSEN, Peter Klint, Ris0 National Laboratory, DK-4000 Roskilde, Denmark, and KUNZENDORF, Helmar, Riso National Laboratoiy, DK-4000 Roskilde, Denmark. Fission track analyses have been performed for wells from the Danish part of the Central Graben and the Danish Sub-basin. The aim of the study is to reveal differences in temperature history and vertical movements for the basins and compare the results to results from basin modelling. Results from the wells A2 and Q1 reveal gradually declining age ranges and mean track lengths downhole, starting below ca. 1 km depth and probably in equilibrium with the present geothermal gradient. Crossing the basin eastwards and proceeding into the Danish basin major differences in temperature history are revealed by the fission track method. Results from unannealed samples (at ca. 1 km depth) reveal age ranges and apatite mean track lengths comparable to ranges found for the Guldfaks area, Norway. The length distribution patterns for the unannealed samples show skewed broad patterns and age distributions which are not comparable with published (and unpublished, Kirsten Hansen) distributions for Scandinavia. Thus the vertical movements and temperature history of the source areas represented in the sediments may reflect earlier stages in the Scandinavian history, or the material may originate from sources outside Scandinavia.

92Nb A N D THE INTERPRETATION OF SHORT-LIVED NUCLEDIC ABUNDANCES IN THE EARLY SOLAR SYSTEM HARPER. Charles L., NASA JSC, Houston, TX 77058 U.S.A. While demonstrations of the existence of live 26Al, 53Mn, I07pd, 1291, l46Sm, and 244Pu in the early solar system constitute some of the most exciting, far-reaching discoveries in isotopic studies, the timings and sitings of their respective episodes of nucleosynthesis remain essentially obscure. This lacuna in understanding represents a major challenge in isotopic research. Two types of models [cf., 1] have been proposed: (i) thermonuclear nucleosynthesis, and (ii) spallogenic 'local' production. With the exception of actinides, inferences from nuclidic abundances to cosmochronology are contingent on spallative production limits. As the early sun was more active, the question is not was there production, but how much? Consider T1/2 = 103 My l46Sm [(146Sm/144Sm -0.007]: contemporary particle flux on the lunar surface is estimated to generate a l46Sm/l44Sm saturation ratio of -10-6 via 147Sm(n,2/i), l48Sm(«,5/i), l44Sm(p,2/i), and l47Sm(p,p/i )[2]. If an upward scaling of 103 for bulk matter proton irradiation (to -102t)cm-2, >10MeV) in the solar accretion era is conceivable (it is), then (i46Sm/l44Sm)5PXLL as high as -0.1% would be expected. l46Sm/l44Sm is particularly interesting because it is a p -only ratio and comparable (l46Sm/Sm = 2 x 10-4) to analagous non p -only ratios in early solar system differentiates: 26A1/A1 (-10-5), 53Mn/Mn ( - 1 0 - 6 ) . i07Pd/Pd (-10-5), and I29i/i (~10-4). Explosive thermonuclear production ratio estimates are: ^1461^144 -^-7 and - 0 . 0 2 for p -, and y - models. The former is consistent with a continuous nucleosynthesis model. The latter requires a (too) large late-spike [c/., 3]. 92Nb (T1/2 = 36 My) is an odd-odd p,r -shielded, s -bypassed nuclide, produced 'locally' by 92Zr(p,/i) and 93Nb(/i,2n)[4], and sparingly in the explosive p - and / -processes [5]: thus widely differing (92Nb/93Nb)/(l46Sm/l44Sm) ratios are predicted for the explosive and 'local' scenarios. We have identified Nb-rutile in acid residue of the Toluca iron meteorite. Nb/Zr is 103 X chondrites, allowing detection of (92Nb/93Nb)^j5A£ > 2 x 10-6 with our current resolution on e92Zr ( - ± l e ) , for 500ng loadings. Analytical results will be presented. Refs: [1] Wasscrburg and Amould, 1987. Uct. Not. Phys., v. 287, p. 262; [2] Le Roullcy et al, 1978. Lunar Planet. ScL, DC, 640; [3] Prinzhofcr et al., 1989. Ap. J., 344: L81; [4] Clayton et al., 1977. Ap. J., 214: 300; [5] Woosley and Howard, 1978. Ap. J.. 36: 285.

ICOG 7

ABSTRACTS

*0Ar/3 9Ar THERMOCHRONOLOGY FOR K-FELDSPARS HAVING A DISTRIBUTION OF DIFFUSION DOMAIN SIZES 90024, USA, Oscar M. UDVERA, Div. G&ol. ^ e t . Sd.,^Caltech, Pasadena, CA 91125, USA, Frank M. RICHTER, Dept. Geophys. Sd., Univ. Chicago, Chicago, IL 60637, USA Many and perhaps most *°Ar/3®Ar age spectra for slowly cooled alkali feldspars are significantly different from model age spectra calculated assuming a single diffusion domain size. In addition, Arrhenius plots calculated from the measured loss of ®®Ar during the step heating experiment show departures from linearity that are inconsistent with diffusion from domains of equal size. The Dodson closure model has been extended to apply to minerals with a discrete distribution of diffusion domam sizes and we find it offers an mtemally consistent explanation for commonly observed features of alkali feldspar age spectra and their associated Arrhenius plots. Because these domains contain a range of closure temperatures, single Kfeldspar samples may reveal a broad segment of a cooling history, rather than the single datum usually expected. We have tested and confirmed the predictions of the domain distribution model through non-conventional argon extraction experiments which also reveal that the domains are an intrinsic property of small, individual K-feldspar cr)'stals. Because the form of Arrhenius plots varies with laboratory heating schedule for samples with a domam size distribution, we propose a plot (the log(r/ro) plot) in which the log of the deviation from the diffusion law for the smallest apparent domain (ro) is plotted against ® Ar released yielding domain size data independent of laboratory heating schedule. We now recognize small but significant differences in activation energy among diffusion domains in K-feldspars. A consequence of even relatively small variations in activation energy among domains is that the shape of the age spectrum may change with varying laboratory heating schedules resulting in, for example, age spectra havmg a characteristic hump. Further, variations in activation energy of -5 kcal/mol can result in differences in calculated dosiu-e temperature up to 30®C from that obtained assuming equal activation energies. Arrhenius and log(r/rQ) plots have the potential to reveal even small differences in E between domains. Exceedingly high estimates of apparent activation energy resulting from reversed heatmg experiments appear to reflect annealmg of subgrain features which define the smallest diffusion domain size. Use of the diffusion compensation relationship allows assessment of the relative distribution of domain sizes in samples containing multiple activation energies.

GEOCHEMISTRY OF THE WOLF-DARWIN LINEAMENT AND RIDGE-PLUME INTERACTION IN THE GALAPAGOS HARPP, Karen S., WHITE, WiIliam.M., Dept. of Geol. Sci., Cornell Univ., Ithaca, NY 14853, USA and DUNCAN, R.A. Coll. of Oceanography, Oregon State Univ., Corvallis, OR USA

C O N T I N E N T A L BASALT SOURCE REGIONS A N D I N TERACTIONS - A COMPARATIVE INVESTIGATION HART. William K . , Geology D e p t . , Miami U n i v . , Oxford, OH 45056, USA

The Galapagos Islands exhibit geochemical patterns different from those of more familiar hotspots. Basalts have increasingly MORBlike signatures toward the center of the archipelago, with greater enrichment around the margins, reflecting entrainment of asthenosphere into the center of the hotspot plume. A curious feature of the Galapagos is the Wolf-Darwin Lineament, located between the Islands and the Galapagos Spreading Center (GSC). Pb isotopes from the Lineament are somewhat distinct from the main Galapagos array, with slightly higher 208pb and 207pb for a ^ven 206pb. Our seabeam surveys and dregding reveal that it consists of numerous small volcanic centers and two larger volcanos. Wolf and Darwin Is. 87sr/86sr ratios at the nonhemmost dredge on the lineament are , slightly lower than the maxVyv.^ ; imum values that occur along the adjacent segment of the GSC. 87sr/86sr and La/Sm ratios pass through a slight maximum near Wolf and Darwin and then decrease to MORB-like values to the southeast This isotopic pattem may be a separate phenomenon from the asthenosphere entrainment in the main body of the archipelago. Higher isotope ratios in the GSC compared to average MORB are presumably caused by entrainment of plume material into the ridge. The pattern along the Wolf-Darwin Lineament may be a result of deeper mixing of plume and ridge material.

Geochemical information derived from uncontaminated continental basaltic rocks and from mantle xenoliths entrained in kimberlites and alkalic magmas corroborates the oceanicbased four principle component mantle ''reservoir" model ( D M M - H I M U - E M I - E M I I ) . This paper compares and contrasts the formation and evolution of distinct mantle source regions beneath two important Cenozoic volcano/tectonic provinces; the Rio Grande Rift region (USAXRGR) and the Ethiopian R i f t / A f a r ( E R A ) . Three c o m m o n threads link these regions; gross similarity in present-day tectonic processes , evidence of HIMU-EMI components in the subcontinental l i t h o s p h e r e , and e v i d e n c e of a m a n t l e source or set of processes that produces basalts with ^^Sr/^^Sr = 0 . 7 0 3 5 , ^^^Nd/^^^Nd - 0 . 5 1 3 0 and - ^ W ^ ^ ^ P b = 1 8 . 5 ( P R E M A characteristics). The latter features are particularly interesting given the distinct pre-Cenozoic crustal evolution histories of these regions, and hence the possible range of chronologies and chemistries of mantle enrichments and depletions. For the R G R , attention is focused on an off-axis tholeiitic to alkaline basalt f i e l d (7 to <1 Ma). Trace e l e m e n t and isotopic data for u n c o n t a m i n a t e d basalts and e n t r a i n e d nodules from this area define arrays bounded by mixtures of DMM(PREMA)-HIMU-EMI-type components. In the E R A , trac« element and isotopic data for uncontaminated tholeiitic to alkaline basalts (30 to <1 Ma) define more complex mantle interactions involving spatially and tempK)rally correlative input from sources with geochemical traits of LoNd (HIMUEMI), EMU, PREMA, and DMM(MORB). The nature and timing of processes yielding characteristics of the above d e f i n e d "reservoirs" is discussed. The c o m parative nature of this discussion allows inferences to be drawn regarding the development of global suboceanic and subcontinental mantle heterogeneities.

ABSTRACTS

ICOG 7

NITROGEN ISOTOPE ANOMALY IN ORDINARY CHONDRITES Ko HASHIZUME and Naoji SUGIURA, Geophysical institute, University of Tokyo, Tokyo, 113, JAPAN. Petrographic studies of ordinary chondrites showed that they have been metamorphosed to various degrees. The amounts of volatile trace elements in ordinary chondrites decrease with the increase in the metamorphic grade of chondrites. There are two possible explanations for these observations. 1) Volatile trace elements have been lost from the parent body during the metamorphism of a intially homogeneous parent body. 2) Central part of a parent body must have accreted earlier at higher temperatures than the outer part of the parent body. Nitrogen isotope may give a good support on resolving the problem. In case 1, nitrogen isotope should show mass dependent isotope fractionation, though in case 2, nitrogen isotopic ratio is expected to be nearly constant. The results for the nitrogen isotopic ratio are shown in the figure. The contribution of cosmogenic nitrogen has been corrected for the values in the figure. The variation of the isotopic ratio is difficult to account by the case 2. The basic tendency of the results seems to favor the case 1, since the nitrogen isotopic ratio of type 3 are tend to be lower than those of higher petrologic types.

Pefrologic type

Sr- AND Nd-ISOTOPE COMPOSITIONS OF ARCHAEAN CUNOPYROXENES: DERIVATION OF EXTREMELY ENRICHED MAGMAS FROM A DEPLETED MANTLE SOURCE. HATTORI. Keiko, Ottawa-Carleton Geoscience Centre, University of Ottawa, Ontario, K1N 6N5, Canada, and HART, Stanley R., Woods Hole Oceanographic Institution, Woods Hole, Massachusetts, 02543,USA. To better understand isotopic evolution of Archaean mantle, Sr- and Nd-isotopic compositions of pure primary clinopyroxenes from 7 Archaean alkaline igneous rocks (syenite and minette) were determined. The samples, covering a "'50km area in Kirkland Lake, are typical alkaline rocks found in the Superior Province. They overlie an already tilted supracrustal sequence and represent the latest magmatism in the Archaean shield. They are chemically very similar to shoshonites in modern arc settings, with low Nb, Ta, and Ti and high LREE and LILE. The degree of the enrichment is extreme in Archaean examples. High contents of these elements in pyroxene measured by ID and ion probe indicate that the source magmas were indeed enriched. Pyroxenes contain Ba (7 to 180 ppm), Sr (21 to 1100 ppm), Sm (4 to 30 ppm), Nd (16 to 90 ppm), Eu (1 to 10 ppm). ^Present-day shoshonites are characterized by higher 87sr/86sr and lower values than the associated calc-alkaline rocks. Examples include those in the Aeolian arc (Luais, 1988), Italy (Hawkesworth and Vollmer, 1979), Wyoming and Navajo Provinces (Salters, 1989), and Highland of PNG (Page and Johnson. 1974j. But the Arnhaean nvrnyenfis have low initial 87c>r/86.qr ^0.70112 tO and o'Sr/o^Sr show a weak positive correlation. Contiguous, older (~2700Ma) mafic and komatiitic volcanic rocks of the greenstone belt were derived from a once depleted mantle source (Machado et al., 1986). The isotope data of the alkaline rocks indicate that the extremely enriched magmas were also supplied from a similar depleted mantle source.

PB ISOTOPES FROM RONDA: GEOCHEMICAL AND FLUID DYNAMIC MODEL OF A MANTLE DL^PIR HAURI, Erik H. and HART, Stanley R., MIT/Woods Hole Joint Program, Woods Hole, MA 02139 USA U, Th, and Pb concentrations, Pb, Sr and Nd isotopic compositions and trace elements are reported for pure cpx and garnet separates from peridotites and mafic layers from the Ronda orogenic peridotite, Spain. Pb isotopic compositions of peridotites span a large range from very depleted to enriched values (206pb/204pb 17.05-19.91), while mafic layers are more restricted (206pb/204pb 18.20-18.95). The isotope systematics of the peridotites are best explained as the result of mixing between three components which are distinct in Pb, Sr and Nd concentrations and isotopic compositions. The data are consistent with an origin of garnet peridotites as mixtures of garnet pyroxenite and depleted peridotite, while plagioclase peridotites have been subject to a metasomatic interaction with radiogenic Pb. Samples of plagioclase peridotite have interacted chemically with surrounding mafic layers. Cpx/garnet partition coefficients for U, Th and Pb average 7, 20 and 72 respectively, indicating that garnet will influence U-Th-Pb fractionations at small melt fractions (<1%). Ce/Pb ratios of spinel and garnet facies clinopyroxenes are similar to oceanic basalts, suggesting similar apparent partition coefficients for Ce and Pb during mantle melting. A comprehensive fluid dynamic model is presented to explain mineralogical and geochemical features of Ronda as a mantle diapir. CHEMICAL BUDGETS AT SUBDUCTION ZONES HAWKESWORTH, Chris, ELLAM, Rob, and McDERMOTT, Frank, Department of Earth Sciences, The Open University, Milton Keynes, MK7 6AA, U.K. The published results on subduction related basalts have been compiled and a general model developed from the average data on basalts from different destructive plate margin suites. The average data for highly incompatible elements vary by factors of 20-5, and inversion of the data arrays indicates that much of the observed variation is consistent with simple partial melting models. In contrast, subduction-related trace element signatures typically suggest a higher slab-derived flux than do the associated isotope variations. Moreover, published models conclude that the LlL element contents of island arc tholeiites are dominated by the contribution from the subducted slab, whereas those of continental margin rocks are dominated by that from the mantle wedge. To reconcile such observations it has been argued that the slab-derived flux may be best evaluated in areas where extreme HPS element depletion indicates that the mande wedge was unusually depleted in incompatible elements. This in turn suggests that the slab contributes -25 % of the Sr, and <15 % of the Kb, Ba, K and Th in average new crust, and that the remainder is scavenged from the mantle wedge. The effects of this, and of other published slab-derived fluxes, on the minor and trace element contents of the residual slab have been explored for a selection of slab compositions. For the lower estimates of the slab-derived flux, there is little detectable effect on the trace element distribution in a range of slab compositions; for example, if the slab contribute^ust 25% of the Sr in new crust, that may represent as little as 3% of the Sr in the subducted crust. Higher slab-derived fluxes are more difficult to reconcile with the available radiogenic isotope data, but if the high LII7HFS element ratios of the continental crust are to be attributed to the slab contribution, the slab must contain a recycled continental contribution (Kay, 1980). Several authors have suggested that the residual slab will have some of the features of ocean island basalts, but none of the models evaluated here readily generate a positive Nb and Ta anomaly in the residual slab. Thus, mat feature of OIB is more likely to reflect small degrees of melting of a depleted mantle source. Unless there is an unexpectedly large'LIL element flux from the slab to the mantle wedge, in addition to that sampled by subduction-related magmatism, it would appear that slab differentiation during subduction is not a major factor in the development of distinctive trace element signatures in the upper mantle.

ICOG 7

ABSTRACTS

TIMING AND ORIGIN OF MIDCONTINENT RIFT ALKALINE MAGMATISM, NORTH AMERICA HEAMAN, L., Dept. of Geology, Royal Ontario Museum, Toronto, Ontario, M5S 2C6, Canada and MACHADO, N., Dept. of Geology, Universite du Quebec a Montreal, Montreal, Quebec, H3C 3P8, Canada.

Nd-0 ISOTOPIC CORRELATION IN ICELANDIC THOLEIITES: EVIDENCE FOR I 8 0 / I 6 0 HETEROGENEITIES IN THE MANTLE BENEATH ICELAND? HEMOND. Christophe. HOFMANN, A.W. and ARNDT, N. Max-Planck-Institut fur Chemie, Postfach 3060, D-6500 Mainz, F.R.Germany)

The Midcontinent Rift System (MRS) in North America is one of the largest known Precambrian intracontinental rifts and provides an excellent opportunity to examine temporal and spatial relationships between tholeiitic and alkaline magmatism in an ancient rift setting. Previous studies on MRS volcanic rocks have shown that rift volcanism was predominantly tholeiitic, large volume (>1,000,000 cubic km), and short duration (22 m.y. between 1108 and 1086 Ma). In contrast, the associated alkaline magmatism is preserved as a number of discrete, off-axis plutonic complexes that intrude Archean crust north of the rift.

Nd isotopic compositions were measured on 19 samples from all active rift zones on Iceland. The suite included some picrites and quartz tholeiites but the majority were olivine tholeiites. In a previously published model, low 2'30xh/232xh, low S^^O, and high s'^Sr/S^Sr in these rocks had been explained by contamination of prima^ tholeiitic magmas with old, hydrothermally-altered Icelandic basaltic crust. However, a compilation of our results and literature data shows positive correlations between ^^^Nd/^^^Nd and 1^0/^60: eNd varies from +7.1 to +10.8 as b^^O changes from +3.9 to +6.25. To produce such a variation in eNd in Icelandic volcanics = 0.12-0.28), requires -200 Ma, which greatly exceeds the maximum age of Icelandic crust. Mantle reservoirs with different Nd isotopic compositions are indicated. On the other hand, the low 5180 values cannot be produced by partial melting or other mantle processes, nor are they found in hydrothermally-altered oceanic crust: they imply interaction between l8o-depleted meteoric waters and basaltic rocks, as has been documented on Iceland. We propose two alternative explanations for the correlations, (a) common Iceland tholeiites are mixtures of melts from plume and MORB sources. Many interact with altered crust during emplacement and acquire low The trend from low to high eNd and 6^80 corresponds to a transition from contaminated basalts from a mixed source, to uncontaminated picrites from the pure MORB source, (b) the basalts with low eNd and 6^80 are derived from Iceland-like crust that was hydrothermally altered by l8o-depleted meteoric water before it was recycled through the mantle. This means that part of the low I80 signature is a source feature, while another part, seen particularly in the rhyolites, is produced by recent processes.

The Coldwell Complex (CC) represents the largest (28 km diameter) MRS alkaline intrusion and contains a plethora of rock types. A detailed U-Pb zircon/ baddeleyite study of five samples indicate that the majority of phases were emplaced during the earliest stages of rift development at 1108 +/- 1 Ma; documenting contemporaneous tholeiitic and alkaline magmatism at the onset of rifting. The Sr-Nd-Pb isotopic composition of selected minerals from the earliest intrusive phases of the CC combined with previously published isotopic data for the most primitive MRS olivine tholeiites and the nearby Prairie Lake carbonatite provide evidence for the presence of at least two isotopically distinct subcontinental mantle reservoirs beneath the MRS; 1) a nearly chondritic source (lithosphere?) documented by the vast majority of MRS magmatism (e^d =+0.5; esr=+2.5) and 2) a "depleted" source (asthenosphere?) documented by the carbonatites ( € N d = + 3 ; = - 1 0 ) .

DATING LATE PLEISTOCENE PEAT DEPOSITS BY URANIUM/THORIUM DISEQUILIBRIUM HEUNIS, H. and VAN DER PLIGHT, J., Centre for Isotope Research, University of Groningen, Westersingel 34, 9718 CM Groningen, the Netherlands Systematic studies showed that U / T h ages obtained from peat showed good agreement with the ^^C ages (Vogel and Kronfeld, 1980 and Van der Wijk, 1987). A new study is concentrated on open system behaviour of peat layers and correction models -for environmental thorium ( Th^^^ t 0). The top and bottom parts of peat layers often behave as~open systems due to the fact that there is an exchange of hexavalent and tetravalent uranium. Depending on the over- and underlying sediments the peat-sediment boundary is often the boundary between oxidizing and reducing environment. Examples from Interglacial- (Eemian) peat deposits will be discussed (Heijnis et al., to be published). Series of samples were taken from several Interglacial-peat layers, the deduced radiometric data can be understood in terms of open-system behaviour of top and bottom part. Environmental thorium corrections were performed using the so-called "leachate only" model of Schwarcz (1989). Having elucidated these experimental problems, future research will be focused on dating peat layers in N.W. Europe which are believed terrestrial equivalents of deep-sea stage 5. In a lot of cases the biostratigraphy does not solve the question whether a peat layer is Interglacial or Interstadial, so complementary dating will be carried out. Some peat samples taken from long-terrestrial records in Australia, containing a climatic history of the last 200.000-150.000 years were dated as well. This provides a chronological framework beyond the limit of ^^C (± 40 ka BP). 1. Schwarcz, H.P. et al., 1989. Dirty calcites, chemical geology. Isotope Geoscience 80, 35-43. 2. Van der Wijk, A., 1987. Radiometric dating by alpha spectrometry on uranium series nuclides. PhD Thesis, University of Groningen. 3. Vogel, J.C. and Kronfeld, J., 1980. A new method for dating peat. J. Afr. J. Sci., 76, 557-558.

COMPOSITIONAL VARIATION IN THE LAU BASIN: IMPLICATIONS FOR THE FATE OF RECYCLED MATERIAL DURING ARC MAGMATISM. HERGT. Janet, HAWICESWORTH, Chris, Department of Earth Sciences, The Open University, Milton Keynes, MKT 6 A A, U.K., and PEARCE, Julian, Department of Geolosy, University of Durham, Durham, DHl 4HN, U.K. New Sr and Nd isotopic results for dredge material from the Lau basin between 18 and 20° S are similar to published results from the same vicinity, and display significant offsets from typical Pacific MORB values towards higher ^'^Sr/^^Si at similar i^3Nd/i44Nd. 87Sr/86Sr varies from 0.703186 to 0.703383, and i43Nd/i44Nd from 0.513003 to 0.513137. Elsewhere in the basin, data for some samples trend down the mantle array (^^Sr/^^Sr 0.7032-0.7036 and i43Nd/i44Nd 0.51295-0.51310) towards the field for Tonga. The new data confirm these results indicating that the magma source in this back-arc environment is not strictly analogous to that of Pacific MORB. In the northern part of the Lau Basin, similar basalts bear strong affinities with Samoan magmatism so that Sr, Nd and Pb compositions are displaced from Pacific MORB values (ie. 87Sr/86Sr 0.7036-0.7047, i43Nd/i44Nd 0.51275 - 0.51290, A7/4, 8-15 and A8/4, ~57 ) towards the field for Samoan basalts on various isotope diagrams. In contrast, the smaller offset from the MORB field observed in the new data from the central part of the basin do not show this Samoan influence; the ^'^ST/^^ST is lower and i43Nd/i^Nd higher (ref. above), and in particular, the displacement above the NHRL on a plot of 208pb/2(54p5 vs 206pb/204>b (A8/4 -43) is not matched by similar displacement on a 207pb/204pt) vs 206p5/204p5 diagram (where the data plot within the Pacific MORB field). The contrast in the behaviour of the Pb isotope signatures is more consistant with intramantle fractionation of the source U/Th ratios than the incorporation of a sedimentary component.

ABSTRACTS

ICOG 7

ISOTOPIC ANALYSES OF FISSIOGENIC NUCLIDES FROM OKLO NATURAL NUCLEAR REACTOR Hiroshl HIDAKA and Akimasa MASUDA Department of Chemistry, The University of Tokyo, Tokyo 113, Japan Oklo uranium ore, known as fossil nuclear reactor, includes much isotopically anomalous elements because of nuclear fission. Especially rare-earth elements (REE) among any other elements are much produced by fission of uranium-235, and are also sensitive to some nuclear reactions in the reactor. Analyses of REE fissiogenic nuclides provide us with useful information, i.e. the scale of nuclear reaction and geochemical behavior of radioactive nuclides. The isotopic compositions of elements deeply associated with nuclear reaction (REE, Ba, Sr, U etc.) were determined by thermal ionization mass spectrometer in four samples from different reactor zones at Oklo mine (Zones 2, 7-9, and 12). Geochemical behavior of radioactive nuclides could be easily characterized from fission product yield curve (plots of each fraction of fissiogenic nuclides vs. its mass). According to these data, REE, Ru and Pd had been fairly retainable in the reactor, while Ba, Sr, Rb had been largely lost out of reactor for about 2.0 b. y. Some nuclides among REE have extremely large neutron capture cross sections, and such nuclides enable us to estimate nuclear parameters. A new calculation method employing U and Gd isotopic composition has been developed to evaluate parameters (integrated neutron flux, average neutron temperature, conversion factor of uranium). These results are in good agreement with those obtained by conventional method.

PLUME TECTONICS: AN IMPORTANT MECHANISM FOR THE REWORKING AND INTERNAL DIFFERENTLATION OF CONTINENTAL CRUST HILL, Robert 1. and Campbell, Ian H., Research School of Earth Sciences, Australian National University, PO Box 4, Canberra, ACT 2600, Australia Ion probe U/Pb dating of zircons from the Late Archean granitegreenstone terrains of the southeastern Yilgam Block of Western Australia shows that basalt eruption began -2705 Ma ago, and was succeeded by two major episodes of felsic magmatism 2685-2690 and 2660-2665 Ma ago. Inherited zircons from granitic and felsic volcanic rocks indicate that they were derived through anatexis of much older, complex cnist, parts of which are up to 3400 Ma old. The time lag of 15-20 Ma between initiation of basalt eruption and the start of voluminous felsic magmatism has been interpreted as the time taken for heat to be conducted from the top of a mantle starting plume into the crustal anatectic zone. The voluminous basalts were derived by melting within the main volume of the head, whereas the komatiites are the products of melting within a high-T central jet. Felsic volcanic and granitic rocks were derived through anatexis of older crust with an average andesitic composition. The fmal product of Late Archean thermal reworking was internally differentiated crust with an upper granitic layer containing greenstone inserts. The model for plume-initiated crustal reworking developed for the Yilgam {plume tectonics) has now been generalised by incorporating inferences from geophysical fluid dynamics studies witii geological observations from younger terrains. New mantie plumes consist of a large diameter head and a potentially long-lived conduit or tail. Plume head provinces are up to 2500 km across, commonly contain evidence for early uplift followed by extension, extensive basalts, and a time gap of 10-20 Ma between first basalts and emplacement of voluminous anatectic melts. Plume tail provinces are characterised by lines of time-progressive magmatic features, which can range from lines of isolated volcanic centres through elongate volcanic filled troughs. The estimated repeat times are 700-800 Ma for plume head and 1500-3000 Ma for plume tail provinces, respectively. The repeat time for plume head reworking is similar to the time-scale for orogenic episodicity inferred for most continents.

NOBLE GASES IN SUBMARINE GLASSES FROM MOR'S AND LOIHI SEA MOUNT: CONSTRAINTS ON EARLY HISTORY OF THE EARTH HIYAGON. ILL, M. OZIMA, S. ZASHU, Geophysical Institute, University of Tokyo, Tokyo 113, Japan, and H. SAKAI, Ocean Resarch Institute, University of Tokyo, Minamidai, Nakano, Tokyo 164, Japan. We have measured isotopic composition of all stable noble gases in submarine glasses from Mid Oceanic Ridges (MAR, EPR and Galapagos Rise) and from Loihi Sea Mounts Our results are as follows: (1) High '^^Ne/'^^Ne ratios up to about 13 and high ^^Ne/'^'^Ne ratios up to 0.057 are found in samples from all the Ridges (EPR, MAR and Galapagos Rise) and a high ^"Ne/^^Ne ratio of 11.8 is also found in a Loihi sample. (2) These high ^^lie/^^ê ratios are always accompanied by high ^Âr/'^Âr ratios (>2000), which suggest little air contamination in the samples. (3) Excess ^^^Xe and excess ^-^^"^-^^Xe are found in some samples from MOR*s and both excesses are correlated, i.e., the former is always accompanied by the latter, and vice versa. Isotope systematics of He and Ne suggests that the relatively high ^"Ne/^'^Ne ratio in a Loihi sample represents the original signature of the source material of Loihi (possibly lower mantle) rather than contamination of MORB-like material. A possible scenario to explain - high '^"Ne/'^^Ne ratios in the mantle would be fractionation of atmospheric Ne from solartype ratio (^"Ne/'^'^Ne =13.6) to the atmospheric ratio (9.8) due to extensive loss of the atmosphere at some time in early history of the Earth.

RECYCLED IDEAS ON THE ORIGIN OF OIB HOFMANN, A.W., Max-Planck-Institut fur Chemie, Postfach 3060, 6500 Mainz, F.R. Germany. OIBs fall into distinct isotopic subtypes, HIMU, EMI, EM2, and (perhaps) PREMA (Zindler & Hart, 1986). None of these has the combined chemical and isotopic characteristics of primitive mantie; instead, they have been differentiated from a rather uniform reservoir, which is chemically complementary to the continental crust (Hofmann et al. 1986). Creation and subduction of oceanic crust and lithosphere together constitute a veiy large-scale recycling process and are by far the largest differentiation process still occurring on Earth. It is extremely unlikely that the chemical heterogeneities displayed by OIBs should not reflect this recycling process. This inference is reinforced by the evidence (NbAJ and Ce/Pb correlations witii s^Sr/s^Sr) that one type of OIB (EM2) contains recycled continental crust, probably in tiie form of subducted pelagic sediment. The combined MORB-OIB reservoir yields younger isotopic model ages than the mean continental crust, because it is cominuopusly remixed and redifferentiated by convection and sea floor spreading. The available geochemical data are most consistent with tiie interpretation that all types of OIBs represent recycled material. Thus, HIMU represents recycled N-type oceanic crust which has lost Pb through hydrotiiermal ridge processes. PREMA and mainarray-type OIBs (such as Hawaii and Iceland) represent recycled Etype oceanic crust and/or former oceanic plateaus and OIBs, and EM2 sources are recycled oceanic crust with minor additions of sediment. EMI is chemically the least well characterized and therefore least understood type. Its isotopic composition (with relatively low 87Sr/86Sr, i^s^d/i^^Nd, and 206p5/204p5 ratios) resembles that of some lower continental crustal rocks. Unresolved potential difficulties with this general recycling model include the high 3He in some OIBs and tiie apparent absence of highly altered oceanic crust from all OIB sources (Hart & Staudigel, 1988). However, these difficulties appear minor when compared with those encountered by non-recycling models.

TWO COMPONENT MIXING IN LOW MAGMAS RELATED TO CONTINENTAL RIFTING: A GEOCHEMICAL AND Sr, Nd, Pb AND O ISOTOPE STUDY OF TERTIARY SYENITIC ROCKS AT KANGERDLUGSSUAQ, EAST GREENLAND HOLM, Paul Martin, Geological Institute, University of Copenhagen, 0ster Voldgade 10, DK-1350 Copenhagen, Denmark The Kaerven syenite complex is the earliest recorded magmatic body related to the continental break-up and formation of the North Atlantic ocean in the Kangerdlugssuaq area where the present day Icelandic hot spot probably was active in the Paleocene. The range and compositions of radiogenic isotopes at Kasrven are extreme compared to other magmatic rocks sofar analysed from the area: ®^Sr/®^r= 0.708-0.720, 0.5127-0.5120,15.0-16.5,'^Pb/^^Pb=14.84-15.10 with a limited variation of 208pb/2^Pb= 36.7-37.2. Oxygen is very light with mostly 5^®0= 3.7-5.2 o/oo. Pairs of feldspar and quartz have A =1.0-2.5 o / o o indicating equilibration a t . magmatic temperatures. A good correlation between all isotopic ratios together with elemental information indicate two-component mixing of the magmas or in the sources. AFC processes were not the major contributor to the isotopic character of the magmas. One end-member has relatively light oxygen, radiogenic Nd and Pb and undradiogenic Sr and is closely related to the rocks of the adjacent and later Kangerdlugssuaq intrusion. The local Archaean high grade crust is isotopically distinct from this. The light oxygen of present day Icelandic basalts, closely related to East Greenland CFBs, has been modelled to be inherited from hydrothermally altered basaltic crust. The light oxygen of the Icelandic and Kangerdlugssuaq rocks either do not have similar origin or a deeper source is indicated.

NEON ISOTOPE ANOMALIES OBSERVED IN BASALT GLASSES FROM LOIHI AND KILAUEA, HAWAII HONDA. M., PATTERSON, D., DOULGERIS, A., McDOUGALL, L, Research School of Earth Sciences, The Australian National University, Canberra, ACT 2601, Australia, and CLAGUE, D.A., U.S. Geological Survey, Menlo Park, CA 94025, U.S.A. Non-atmospheric Ne isotopic ratios have been reported by earlier workers from materials derived from the mantle, such as MORBs and diamonds. The previous Ne results lie on a clear correlaticJn line in ÔNe/^^Ne-îNe/^^Ne space with one end member being atmospheric Ne, the other end member enriched in îNe and ÔNe relative to ^^Ne. In order to see if this Ne trend is general for mantle derived materials, and in particular those from the so-called undepleted mantle source, we have measured Ne isotopic ratios in fresh basaltic glasses dredged from Loihi Seamount and the East Rift Zone of Kilauea (hereafter referred to as the L-K samples), which are regarded as a manifestation of the Hawaiian mantle plume or hotspot. Neon isotopic ratios measured on the L-K samples lie on a clear correlation line, which has a slope intermediate between those of the MORB line and the mass fractionation line, all lines passing through atmospheric Ne. This indicates that the correlation observed for the L-K samples cannot be explained by mass fractionation processes operating on atmospheric Ne, nor by mixing with MORB Ne. This requires that the source of the nonatmospheric Ne in the L-K samples must have different characteristics compared with the MORB source. The MORB and L-K correlation lines could be interpreted as mixing between present-day atmospheric Ne, and at least two distinctive non-atmospheric Ne components in the sources. Possibilities include sources with both nucleogenic and solar components in order to explain the enrichment in ^^Ne and isotopes. If this is the case, the steeper slope of the L-K line compared with the MORB line in 20Ne/22Ne-2iNe/22Ne space implies that solar Ne is more enriched relative to nucleogenic Ne in the L-K source than in the MORB source.

ICOG 7

ABSTRACTS

AN ISOTOPIC EVIDENCE ON PALAEOENVIRONMENTAL EVOLUTION SINCE THE LATE PLEISTOCEI^E IN THE TAIWAN STRAIT HONG ASKI Third Institute of Oceanography, State Oceanic Administration , Xiamen(Amoy), P.R. China, P.C. 361005 In order to find the evidence on palaeoenvironmental evolution since the Late Pleistocene in the Taiwan Strait, the writer chose foraminifera in the core MD756 coliJinn sample for 0-isotope analysis, and combining with some other results such as the '''C dating , has worked out the 0-isotope variation cu^ T h e core MD756124''52.o'N, 119''15.5'e) locates at the outside of Qianzhou Bay, where the depth of water is 34m. The length of the column sample, which being clayey silt throughout, is about 557 centimeters. The whole column sample is divided into 11 sections. Vi/e pick out the benthic foraminifera, clean up the organic pollutants on the siirface, and then, conduct the isotope analysis on the MAT-251 mass spectrometer. The 0-isotope curve and the '^C dating results(some of the '^C dating data are based on the calculation from the average sedimentary rate) show that, at 16.7ka and 12.2ka, the S^^O values deviate negatively twice, indicating the rise of the palaeotemperatxire of the water; at 13o6ka to 15.7ka and 10.2ka it deviate positively twice, indicating the drop of palaeotemperature of water, the range of deviation of S'^0 values of the sample reaches All these results tell that the Taiwan Strait region at least \mdergone times of climate variation cycle since the Late Pleistocene.

MAJOR AND TRACE ELEMENT CHEMISTRY AND D/H, I80/I60, 87sr/86SR AND 143ND/144MD RATIOS IN ROCKS FROM THE OKINAWA TROUGH: MAGMA GENESIS AT THE SPREADING CENTER OF THE MARGINAL BACK-ARC BASIN Hiroji HONMA. Institute for Study of the Earth's Interior, Okayama University,Misasa,Tottori-ken 682-02, Japan* Shigeru IIZUMI, Department of Geology, Shimane University,Matsue,690,Japan, Minoru KUSAKABE*, Hiroo KAGAMI*, Hitoshi SAKAI, Ocean Research Institute, University of Tokyo, 1-15-1, Minamidai, Nakanoku, Tokyo 164, Japan. The Okinawa Trough is an active marginal back arc basin behind the Ryukyu arc-trench system, now opening along the south-eastern margin of the Asian continent. One dacite and four basalt samples from knolls and ridges of the central graben of the Okinawa Trough have been analyzed for major elements, trace elements (Ni, Cr, V, Zn, Rb, Sr, Nd, Sm, Y and Zr) and isotope ratios. Sr and Nd isotope ratios of rocks of the Ryukyu volcanic chain, the volcanic front of the Ryukyu arc, also have been analyzed and compared with those of the Okinawa Trough. In contents of AI2O3, total iron, MgO, CaO, Na20, Ni, Cr,and V, the basalts are quite similar to N-MORB's, whereas they have much higher KoC^ Rb, Sr, H2O contents and D/H, I S 0 / I 6 0 and ^'^Sr/^^Sr ratios and lower 143Nd/144Nd ratios than do N-MORB's, suggesting the contribution of subduction-related, mainly sedimentary components to the source of magmas of the Okinawa Trough. Okinawa Trough basalts are roughly similar to intra-oceanic back-arc basalts, but they have much lower 143Md/144N<j ratios and higher K, Rb, Sr contents and I 8 0 / I 6 0 and ^'^ST/^^ST ratios. This may be due largely to difference in the nature of subduction-related materials. Volcanic rocks of the Ryukyu arc are similar to those of the Okinawa Trough in Sr-Nd isotopic ^ s t e m atics, but many of them have distinctly higher ^'^Sr/ S6sr ratios and lower 143isid/144Nd ratios, which may have resulted from contamination by crustal materials.

ABSTRACTS

IC0G7

STRONTIUM AND OXYGEN ISOTOPE SYSTEMATICS OF GRANITOIDS IN YANGBAJAIN-QUXU, TIBET, CHINA HU Ruizhong, YU Jinsheng, and GUI Xuntang, Institute of Geochemistry, Academia Sinica, Guiyang, Guizhou Province 550002, China The area studied, lying just to the north of the Indus-Yarlung Tsangpo suture zone, is a part of Transhimalaya or Kangdese plutonic belt. There are successively Quxu batholith, Lhasa and Yangbajain plutons from south to north in the area. They were intruded into metasediments and metavolcanics of Palaeozoic and Mesozoic ages about 50 ± 5 Ma ago. The plutonic rocks are composed dominantly of granodiorites for the Quxu, and granites for the Lhasa and Yangbajain. Based on the data of Sr and O isotopic compositions of some 60 samples, this paper discusses the genesis of the intrusions. The (87Sr/86Sr)i of the Quxu ranges from 0.7036 to 0.7083, and 5I8O from +6.88 to +9.40%o. A plot of b^^O versus (87Sr/86Sr)i has a clear positive slope. The lower end of the trend line is close to the "mantle representative point". These data can be modelled by fractional crystalization and simultaneous assimilation of crustal rocks by mantle-derived magma. The (87Sr/86Sr)i of the Lhasa and Yangbajain, varying from 0.7042 to 0.7075, is similar to those of the Quxu, but their b^^O values are much lower (+2.79 to 6.69%c). It is considered that the low 5I80 values are the result of present day or recent near-surface geothermal activity. The genesis of these two intrusions is probably the same as the Ouxu, because no correlations exist between the 5I80 and (87Sr/8t>Sr)i values for the Yangbajain and the Lhasa plutons, but analysis of quartz in each intrusion places them on the Quxu 518O trend line.

FACTORS AFFECTING HELIUM ISOTOPE RATIOS IN THE TAUPO VOLCANIC ZONE, NEW ZEALAND HULSTON. John R., Institute of Nuclear Sciences, DSIR, PO Box 31-312, Lower Hutt, New Zealand, and LUPTON, John E., University of California, Santa Barbara, California 93106. USA Geothermal gases from many of the geothermal fields in the central section of the NE-SW trending Taupo Volcanic Zone have ^He/^He ratios in the range 5.9-7.3 times the atmospheric ratio (5.9-7.3 RA), which is typical for mantle helium from subduction zone systems. In the Wairakei field helium-3 varies only a few per cent from 7.2 R^, despite evidence (eg from under and over saturation of other noble gases relative to air saturated water values) of considerable boiling at depth. This is in sharp contrast to published measurements from Yellowstone National Park, USA, where boiling at depth results in considerably lower helium-3 in the depleted v/aters. This difference is possibly due, in part, to Wairakei being an "older area" in terms of water throughput (as indicated by the low oxygen-18 shift) from which the radiogenic components have largely been flushed. The Ohaaki-Broadlands area, which is closer to the eastern edge of the volcanic zone has a larger oxygen-18 shift and helium isotope ratios which vary from 6 R^ in the NW section to ~ 3 R^ in the SB section, resulting in a factor of two variation across the whole fleld. Since the Cl/B ratio also varies across this field the change in helium-3 is probably due to variable interaction at depth with rocks containing both boron and radiogenic helium. Present indications are that the helium isotope ratios are affected both by the flux of mantle helium released associated in the subduction process and also by the presence of sedimentary rocks beneath the geothermal field and at the edges of the volcanic zone.

49 LOW-TEMPERATURE THERMAL HISTORY OF PROSPECTIVE SITES FOR RADIOACTIVE WASTE DISPOSAL: AN APATITE FISSION TRACK STUDY IN THE ALPINE FORELAND BASIN OF SWITZERLAND. Anthony J. Hurford, Fission Track Research Group, Dept. of Geological Sciences, University and Birkbeck Colleges, Gower Street, London WCIE 6BT. U.K. An intensive drilling and investigative progran^e has been carried out in Switzerland since 1980, aimed at locating sites suitable for storage of high, medium and low-activity radioactive waste produced by the nuclear industry. Investigation has centred upon Hercynian crystalline massifs of the Alpine chain, and crystalline basement underlying Tertiary molasse, Permo-Triassic and Carboniferous sediments in the northern foreland basin. An essential factor in evaluating the suitability of a prospective storage site is the predictive modelling of both the future stability of an area and the secure preservation of stored radionuclides. Understanding the thermal history of a tectonic unit provides a baseline upon which such forward nxKielling can be grounded. Fission track analysis of apatite from sediments and underlying basement in the foreland basin has provided a means of determining low-temperature history. In each of the three investigated wells similar patterns of lowered fission track age and reduced track length patterns are indicative of past elevated temperatures. Alternative hypotheses suppose additional kilometre-scale burial by Jurassic and Cretaceous sediments, with subsequent uplift and removal of section in earliest Tertiary times, and/or enhanced Terti^ heatflow with the passage of hot-fluids. In either case, the previously unrecognised late Mesozoic - early Cenozoic heating of the crystalline basement provides an important constraint for predictive models for storage of waste.

MIGRATION OF EXCESS ARGON RECORDED IN BIOTITE AROUND A DIKE CONTACT HYODO, H., Hiruzen Research Institute,Okayama Univ. of Science, Okayama, 700, Japan, and YORK, D., Dep. of Physics, Univ. of Toronto, Toronto, M5S 1A7, Canada ^^Ar/^^Ar age analyses using laser step heating were carried out on single-grain biotites (0.5 mm in size) from the contact zones between a Grenville diabase dike (570 + 3 Ma) and the country rock (tonalitic gneiss, 1000 Ma) near Mattawa, Ontario, Canada. The single grain analysis by laser step heating allowed us to examine the few unaltered grains available. At the contact, the biotite ^^/^r/^^f^r integrated ages in the country rock were almost totally reset to the age of the dike emplacement. They gradually rise with increasing distance from the contact, and exceed the reference age of the country rock, creating an asymmetric wave-like pattern of ages across the dike with a maximum of about 1400 Ma. Since simple thermal conduction creates only a symmetric pattern of isotherms around a dike, the asymmetry suggests that convection of fluid in the rock may have played an important role. A qualitative model for the process of forming the excess peak will be discussed, considering the effect of argon partial pressure and the argon blocking process in biotite during course of its cooling history. The asymmetric partial pressure distribution was presumably due to fluid movement initiated by the dike which was emplaced in a tilted attitude. On a totally different tectonic scale, a similar pattern of excess K-Ar ages near the Grenville Front (fault zone) was reported by Wanless et al.. (1970). These phenomena suggest that the analysis of excess argon may provide a new tool to analyze temperature and the movement of fluids as a function of time in plutonic and tectonic environments.

ICOG 7

SR AND ND ISOTOPE STUDIES ON MAFIC INCLUSIONS IN I-TYPE GRANITOIDS Shigeru IIZUML Department of Geology, Shimane University, Matsue, Shimane, 690, Japan, and Hiroji HONMA, I n s t i t u t e for Study of the Earth^s Interior, Okayama University, Misasa, T o t t o r i , 682-02, Japan. Based on petrographic, petrochemical and Sr-Nd isotope data, we examine the genesis of mafic inclusions in I type g r a n i t o i d s . Samples were collected from two Permian plutons (the Moonbi and Valcha Road Adamellites) in the New England Fold Belt, eastern Australia, and four Cretaceous plutons (the Neu, Ningyo-toge and Chi2uKoshihata Granites and Okutsu Granodiorite) in southwest Japan. I n i t i a l Sr and Nd isotope r a t i o s of most mafic inclusions in the Moonbi and Walcha Road Adamellites and the Neu granite are very close to those of their host granitoids. This means that the host granitoids and their mafic inclusions were derived from the same parents or were i s o t o p i c a l l y equilibrated with each other at the magmatic stage of the host g r a n i t o i d s . On the other hand, almost a l l mafic inclusions in the Okutsu Granodiorite have s i g n i f i c a n t l y lower Sr and higher Nd isotope ratios than the host rocks. Sr and Nd isotope modification of the magma during i t s evolution in the crust could be suggested for the Okutsu Granodiorite. In contrast with t h i s , some mafic inclusions in the Ningyo-toge and ChizuKoshihata Granites have s i g n i f i c a n t l y higher Sr and lower Nd isotope r a t i o s than their host rocks. Contribution of crustal materials (mostly sediments) to the magmas could be suggested for genesis of these mafic inclusions.

ESR DATING OF CORALS AND PRE-ANNEALING EFFECT ON ESR SENSITIVITIES

IKEDA. Sumiko. KASUYA. Masao and IKEYA, Motoji ; Department of Physics, Faculty of Science, Osaka University, Toyonaka 560, Japan

Electron spin resonance (ESR) dating using natural radiation damage has become a useful method for age determination of ara—a. 'gonitic corals. Four ESR signals. A, B, C, and D are commonly found in corals (Fig.). 2015 2010 2005 2.000g-value'.995 Signal C, which increases with gammairradiation, is practically used for dating. We investigated pre-annealing effects on ESR sensitivities using Recent coral samples. Pre-annealing at 60-200*C for one hour significantly changed gamma-ray sensitivities of ESR signals; increase for signal A and decrease for signals B and C. The results imply that changes in ESR sensitivity during geologic times may have to be considered in ESR dating. Another implication is a possibility of using ESR sensitivities as indicators of paleotemper^ ature.

ABSTRACTS

SINGLE ZIRCON AGES FROM A GREYWACKE OF THE TORLESSE TERRANE, NEW ZEALAND IRELAND.Trevor. Research School of Earth Sciences, ANU, Canberra ACT2601, Australia. The sediments of the Torlesse Terrane and their Haast Schist metamorphic equivalents constitute a major fraction of the land area of South Island, New Zealand. The sediments consist predominantly of sparsely fossiliferous greywackes and mudstones with deposition ages that range from the Carboniferous through to the Cretaceous. Their quartzofeldspathic composition has suggested an origin from a continental source and paleomagnetic reconstructions have indicated Marie Byrd Land, Antarctica as a possible provenance area. As a constraint on the provenance of the sediments, single zircons were separated from a sample of greywacke and analyzed with the SHRIMP ion microprobe for their U-Pb ages. The greywacke sample was collected from Lake Aviemore near a middle Triassic fossil locality (Corbies Creek). Two morphological types of zircon were recognised: euhedral crystals were probably released from lithic fragments during sample preparation whereas rounded pitted grains have clearly experienced sedimentary processing. The euhedral zircons dominate the population being 65 of 79 analyzed. They have U concentrations of 200-700 ppm and ThAJ close to unity, and yield a mean 206pb»y238u age of 260 ± 4 Ma (2am). This is substantially older than the deposition age range of 209-233 Ma, however a rim on one of the 260 Ma grains has high U, Th, and common lead and has a 206pb'/238u age of 211 ± 10 Ma. The rounded zircons have ages of 340 Ma (1 zircon), 500-550 Ma (7), 680 Ma (2), 1090 Ma (3), and 1200 Ma (1). These ages indicate that the greywackes have sampled material that is as old as middle Proterozoic, although this material is )robably volumetrically minor as indicated by the preponderance of ithic 260 Ma material. The provenance age of the zircons must have changed over the age range of the Torlesse sediments and only a complete stratigraphic sampling might finally constrain the provenance area. The zircon ages contrast with the model Nd age of Torlesse sediments which is approximately 1200 Ma. Possible explanations for this difference include: zircon is absent from the older material, zircon formedfromrecycled older crust, this sediment is not representative of the Torlesse and has a younger age, or the Nd systematics do not in fact reflect the average age of these sediments. GAS GEOCHEMISTRY OF THE HYDROTHERMAL FLUIDS IN THE MIDDLE OKLNAWA TROUGH BACK ARC BASIN ISHIBASHLJun-ichiro, SANO,Yuji, WAKITA,Hiroshi, Laboratory for Earthquake Chemistry, Univ. Tokyo, Tokyo 113, Japan, and GAMO,Toshitaka, TSUTSUMI,Makoto, SAKAI, Hitoshi, Ocean Res. Inst., Univ. Tokyo, Tokyo 164, Japan The Okinawa Trough, located southwest to Japan, is an active back-circ basin formed by extension within the continental lithosphere. At the JADE hydrothermal field 127»05'^) in the middle Okinawa Trough, high temperature (> 300®C) fluid accompan}ing mineral deposits was discovered in 1988. This is the first observation of the hydrothermal circulation within intra-continental back-arc bcisins. The JADE hydrothermal fluid shows significant enrichment in gas components. Dissolved CO2 concentration of the endmember fluid is more than one order of magnitude higher than that of the mid-ocean ridge hydrothermal sj'stems. Another interesting phenomenon is venting of liquid CO2 at two sites about 300-500 m apart from the center of the hydrothermal activity. The ^He/^He ratios of 6.2 - 6.8 R/Ratm and of CO2 of -4.8 - -5.8 per mill of the JADE fluid indicate that these gases are of mantle origin. However, the He isotopic ratios are lower than those of the mid-ocean ridge hydrothermal systems ( 7.8 R/Ratm ), and fall within the range for volcanic gases along island arcs. The C02/^He ratios of (13 - 34) x 10^ are also similar to the island arc volcanic gases. The gas geochemistry of the JADE hydrothermal fluid is typical of island arc tj^e magmatism.

ABSTRACTS

ICOG 7

K-AR AGE STUDY OF SUBDUCTION-RELATED MATERIALS IN SOUTHWEST JAPAN ITAYA, Tetsumaru, Hiruzen Research Institute, Okayama University of Science, Okayama 700, Japan There are three major lithologic belts along the pacific side of southwest Japan, the Sanbagawa, the Chichibu and the Shimanto which are composed of various type of rocks of the Triassic to Oligocene subductionaccretion complex and have suffered high P/T or subduction-related metamorphism. The author has dated the K-Ar age of white mica from the metamorphic rocks collected systematically from these belts with my colleagues to reveal their tectonic evolution. The results are reviewed here. White mica is a common constitutent mineral in low to middle grade metamorphic rocks, and its K-Ar age is generally the same as the Ar-Ar age and is useful in investigating of metamorphic events of the host rocks. The newly developed mass spectrometer with high sensitivity and precision was applied to the dating. This age dating system requirs less than 50mg of material in each run. The results show that the Sanbagawa experienced peak metamorphism in the lower Cretaceous(n0-120Ma), and the high grade rocks had the average cooling and uplift rates of 9-120C/Ma and 0.4-0.5mm/year. The Chichibu is subdived into the three belts, the northern, central and southern. The northern belt belons to a lower grade part of the Sanbagawa, being 110-130Ma in age, and the central belt, called the Kurosegawa belt, was composed of low grade metamorphics(l90-230Ma) of a subduction complex and episodic igneous, metamorphic and sedimentary rocks(Silurian to Permian). The Shimanto belt consists of upper Cretaceous to Oligocene subductioncomplex materials. These K-Ar mica ages were concordant with the fossil ages of their protolith. This chronological approach made it possible to reconstruct the tectonic evolution of Triassic to Oligocene subduction-accretion complex in southwest Japan.

Sr-Nd-Pb ISOTOPIC STUDY OF ULTRA HIGH PT ROCKS: IMPLICATIONS FOR MANTLE EVOLUTION. E. Jagoutz^ V.S. Shatsky^ N.V. Sobolev^- ^ Max-Planck Institut fiir Chemie, D6500 Mainz, F.R.G. ^Inst. of Geology and Geophysics, Siberian Division U.S.S.R Academy of Sciences 630090 Novosibirsk 90, U.S.S.R. The Kokchetav massif is known as the first occurrence of diamonds in metamorphic rocks. The diamonds occur as inclusions in relict garnets from retrograded garnet and biotite gneisses. Other high pressure indicators (as K-feldspar exsolution lamellaes in clinopyroxene) demonstrate that the diamonds crystallized under static conditions at pressures > 40 kb. A chemical heterogeneous protolith was formed 2 Ga ago. As indicated by the very radiogenic Pb isotopes units with a very high U (U/Pb > 200) were contained in this protolith. Most of this U was lost by the high PT metamorphism. The pro-grade metamorphism was dated by the loss of U, while the end of the metamorphism was dated using the Sm-Nd mineral age of the high PT mineral relicts separated from this rock. The peak of metamorphism occurred 533 ± 10 Ma ago, while the duration of this metamorphism might be as short as 14 Ma from the loss of U to the cooling age of high pressure garnet and clinop}Toxen. It appears that these newly discovered geological processes might have occurred frequently in the past. During subduction U and Rb might be lost by the breakdown of hydrous phases producing high potassic volcanics. However Pb will be lost by the eclogitisation and dominate the mantle Pb, because of the low Pb concentration of the mande rocks. The contamination is less noticeable in Sr and especially Nd isotopes in the mantle rocks. The Pb isotopes found in mantle rocks might represent the fossilized traces of such events getting slowly homogenized by the mantle convection. This model explaines the observed variations in Pb and Sr isotopes in the mantle, which suggest strong and non-systematic U/Rb fractionations, while the U/K ratio remains constant.

DIRECT DATING OF SEDIMENTARY SEQUENCES: Pb-Pb GEOCHRONOLOGY OF CARBONATE ROCKS JAHN, Bor-ming, and CUVELLIER, Henri, University de Rennes, Institut de Geologie, CAESS-CNRS, 35042 Rennes C6dex, France. Carbonate rocks occur in all major sedimentary sequences and represent about a fifth in volume of the earth's sedimentary rocks. Thus, any methods that can be used to date them should have significant impact on stratigraphic studies. Recent U-Pb isotopic works (1-6) on carbonates of Archean to Mesozoic ages have shown such a promise. Carbonate rocks often, though not always, possess very radiogenic Pb isotopic components, occasionally reaching a level of zircon Pb. Very credible and geologically significant isochron ages have been obtained either for little to non-metamorphosed limestones and dolomites (dating sedimentary ages, 2,4,6) or for marbles (dating metamorphic ages, 3,5). A study (6) on carbonate rocks from the Schmidtsdrif Formation (equivalent to the Transvaal Dolomite) of the Transvaal Supergroup, South Africa, revealed that the stromatolitic limestones are not Proterozoic in age, but they were deposited in the late Archean (T = 2557 ± 49 Ma, 2sigma). This new age suggests that a division of the Transvaal Supergroup is probably required because it comprises both Archean and Proterozoic formations. A systematic study of the entire sedimentary column of the classical Jixian Section in China is being undertaken. Preliminary results show that the basal Changcheng (Great Wall) Group was formed not earlier than 1750 Ma as opposed to the formerly proposed boundary at 1950 Ma. The Tuanshanzi Formation, which is in the middle part of the Changcheng Group, is precisely dated at 1556 ± 10 Ma ratios from ca. 16 to 180). Some critical problems will be evaluated.

- Global catastrophes and volcanic events : a comparison of volcanic and industrial outgassings under normal and catastrophic conditions. M.JAVOY&G.MICHARD (Laboratoire de gdochimie des isotopes stables, Laboratoire de geochimie des eaux, IPGP et University de Paris VII, 4, place jussieu, 75251 Paris cedex 05). The Cretaceous Tertiary boundary is currently explained by two main scenarios: extraterrestrial or volcanic events. The last one is supported by the more and more exact synchronism between KTB and the Deccan Traps and the very short duration of this volcanic event (10^ to 4 10^ ans). Independently of the fact that the Deccan Traps are or are not the primary cause of the mass extinction, their synchronism imposes that very large quantities of CO2 have been released in a very short period of time by discontinuous events and directly into the atmosphere, contrarily to the normal, midoceanic ridge mode. It is shown that whereas this normal activity corresponds to CO2 fluxes comparable lo that of the early industrial period, the Deccan pulses correspond to the present industrial fluxes, with comparable duration (5 to 100 years) for a single event. The nocivity of these emissions is enhanced relative to industrial ones mainly because of much higher SO2/CO2 ratios, leading to acidification of the oceans. It is shown that the Deccan volcanism has probably rapidly distroyed the planctonic fauna, leading to the diss^jpearance of the so called Williams Riley pump. In this work we calculate the chemical and isotopic effects of CO2 outgassing from a total volume of 1-2 10^ km^ of Deccan traps with a probable carbon concentration around 3000 ppm. We suppose that the Deccan events were distributed as fifty major events in a timespan of 10^ to 10^ ans. The total amount of CO2 outgassed is 1.6 10^ moles, that is about half the present ocean content We discuss the behaviour of the ocean as a function of the rate and mode of outgassing. We show that, in any case, this corresponds, in a short time span, to the dissolution of an equivalent quantity of carbonate sediments. Thereprecipitationof this excess carbonate corresponds grossly to the size of the S^^c anomaly observed in KTB sediments, with the right isotopic composition due to progressive neutralisation by continental alteration. Thereturnto normal S^^C is due to resuming of the biological activity.

52 ISOTOPIC AND GEOCHEMICAL CONSTRAINTS ON MIXING PROCESSES IN SUPRA-SUBDUCTION ZONE ENVIRONMENTS: EVIDENCE FROM ORDOVICIAN VOLCANICS AND OPHIOLITES. JENNER, G., Earth Sciences, Memorial Univ., St John's, NF A l B 3X5, CANADA Recent arc volcanics show limited overlap in cNd with MORE. This is consistent with arc sources containing a minor amount of subducted, old, continent-derived sediment. Other geochemical features of arc volcanism are also consistent with this suggestion. Nonetheless, debate about the role of subducted sediments versus ocean island mantle sources in determining arc isotopic compositions continues. Integrated geochemical and isotopic studies of Ordovician oceanic volcanics and ophiolites shows that these rocks record the evolution of all stages of arc and back-arc development. A spectrum of mantle-derived compositions and geochemical signatures are found. Of particular importance is the recognition of a transitional (TRANS) and refractory (RARC) series of arc volcanics, not documented in modem arcs. TRANS are intermediate between arc tholeiites (DARC) and DMORB. eNd in DARC, TRANS and DMORB range from -1-5.5-7.3, -H7.2-8.4 and -1-6.5-8.6, respectively. R A R C are characterized by extreme REE and HFSE depletion and cNd -1 to +4. Models required to explain R A R C suggest that arc metasomatism and/or melt extraction can be extremely complex. Tlie complete transition between arc and non-arc trace element and Nd isotopic signatures in these samples is in complete accord with mixing models involving a subducted sediment component. There is no evidence that the arc volcanics acquire their chemical signatures from an ocean island component. Modern arc volcanics recording the complete nature of the interaction between depleted mantle and subducted components presumably exist, but are not easily recovered.

BASIN MODELLING AND ANNEALING OF FISSION TRACKS IN THE DANISH A R E A JENSEN. Peter Klint, Ris0 National Laboratory, DK4000 Roskilde, Denmark, HANSEN, Kirsten, Institute for Petrology, Copenhagen University, 0ster Voldgade 10, DK1350 Copenhagen K, Denmark, and KUNZENDORF, Helmar, Ris0 National Laboratory, DK-4000 Roskilde Denmark. A mathematical model has been developed by which the thermal history of rocks may be determined from fission track data. Basic input data to the model is the fission track length distribution, the surface track density, and the 238U concentration. The number of tracks within a certain track length interval represents the time necessary to create them. The track length distribution frequency may therefore be converted to a number of age intervals. For each age determination (and for each column of the histogram) the fanning plot annealing model is used to determine the paleotemperature. Biases of the measured track length distribution of etched tracks are discussed in relation to the track length distribution of unetched tracks. The likelyhood that a track cuts a path to the surface, and is etched, is proportional to its length. The probability that a track is accepted as beeing horizontal during observation is inversely proportional to its length. The mathematical model is applied to reveal the thermal history of sedimentary rocks from the Danish area. The results are discussed in relation to basin model studies. Fission tracks, as a thermal indicator, has a promissing potential compared to other thermal indicators such as vitrinite reflectance and/or biomarkers.

ICOG 7

ABSTRACTS

CHEMICAL AND ISOTOPIC COMPOSITION OF THE MANTLE BENEATH THE MENDOCINO TRIPLE JUNCTION, WESTERN USA JOHNSON. Clark, and MEDARIS, L. Gordon, Jr., Dept. of Geology, University of Wisconsin, Madison, WI, 53706, USA, and COLEMAN, Robert, Dept. of Geology, Stanford University, Stanford, CA. 94305, USA Basaltic lavas erupted in the San Andreas fault zone at 10 to 1 Ma followed northward migration of the Mendocino triple junction (MTJ) and provide a unique view of the subcontinental mantle at its margin. The mildly alkalic lavas have [Ce/Yb]N 2 to 10, initial ^'^Sr/^^Sr^ 0.7025 to 0.7030, ENd= +6 to +11. 206pb/204pb= 18.7 to 19.2; the isotopic compositions overlap those of MORB, although the lavas have elevated 207pb/204pb and 208pb/204pb ratios (15.60 to 15.64 and 38.3 to 38.5, respectively) relative to MORB. These data support models for asthenospheric upwelling during migration of the MTJ. Clinopyroxene and orthopyroxene in Cr-diopside spinel Iherzolite (SPL) nodules that occur in some of the lavas have chemical compositions that are similar to those of SPL nodules from the interior of W USA: AI2O3, 5.5 to 6.3 wt% and 3.9 to 4.8 vit%; Cr203, 0.8 to 1.2 wt% and 0.4 to 0.6 wt%, respectively. These data suggest that the mantle beneath the MTJ is still fertile. Pyroxenite nodules and xenocrysts in the host lavas represent basaltic intrusions near the MOHO that were emplaced during magmatic thickening of the lithosphere associated with passage of the MTJ. In contrast to the MTJ lavas, young mafic lavas from the arc-related Mt. Lassen volcanic center, 150 km NE. generally have initial 87sr/86sr= 0.7035 to 0.7043 and +4 to +6, although Pb isotope ratios overlap those of the MTJ lavas (BuDen and Clynne, 1990, JGR). Assuming that the continental-margin MTJ lavas were derived from mantle that Is devoid of subducted material, the higher Sr and lower Nd isotope ratios of the Mt. Lassen lavas may reflect the magnitude of the subduction component in these arc-related volcanic rocks.

RB-SR AND SM-ND AGES OF METAMORPIC ROCKS FROM SOUTITvVESTERN AREA OF SRI LANKA H.KAGAMI,Institute for Study of the Earth's Interior, Okaya^ University,Misasa, Itottori-ken 682-02, Japan,M.a'JADA,Faculty of Science,Yamguchi University, Yamaguchi 753,Japan, Y.OSANAI,Fukuoka University of Education,Mun^ta 811-41,Japan, Y.HIROI, Faculty of Science,Chiba University,Chiba260,Japan Sri Lanka is geologically divided into three unites ; Vijayan Coiplex, Highland Series-South\^^stem Group and Wanni (West Vijayan) Corplex, fron southeast to northwest. High-grade metamorphic rocks are mainly distributed in the Highland Series-Southvvestem Group. Newton and Hansen (1986) stated a possibility that the high grade terrain in Sri Lanka is deep-crust section. Ages of metamorphic rocks from the Highland Series and Wanni Corplex have been measured using Rb-Sr and Sm-Nd systems. Granulites collected frcm southern area (Ganpola-Nuwara Eliya) of the Highland Series have various ages as follows ; 2600-2900Ma (Nd model age), 2500±130Ma (Rb-Sr I^fi: isochron age, initial 87Sr/86Sr ratio=0.718),2330±30Ma(Sm-Nd Wr isochron age, initial eNd=-0.4) and early Paleozoic (ca. 500Ma). The first age indicates an initial separation of crustal materials frcm mantle source. The second age of 2330Ma (Sm-Nd) indicates large-scale extraction of granitic melt frcm the crustal materials. The last age shows events such as small-scale formation of granitic melt and granulite facies metamorphism. The granulites fron Ganpola-Nuwara Eliya have long and corplicated history. Gneisses and granulites collected fron Wanni Corplex and central area (Kandy) of the Highland Series have ages of ca.lOOOria and early Paleozoic. Initial separation age of these rocks fron the mantle source is probably close to the first age. The second age, v M c h is also reported frcm Ganpola-Nuwara Eliya, implies regional metamorphism in Sri Lanka.

ABSTRACTS

IC0G7

CERIUM AND NEODYMIUM ISOTOPIC COMPOSITIONS IN SEA WATER INFERRED FROM LIMESTONES K A M I O K A , H i k a r i . G e o l o g i c a l Survey of J a p a n , Tsukuba, Ibaraki 305, Japan, and MASUDA,Akimasa, D e p a r t m e n t of C h e m i s t r y , University of Tokyo, Bunkyo-ku, Tokyo 113, Japan The similarity of REE abundance patterns b e t w e e n limestones and sea water suggests the usefulness of limestones as the "fossil of sea water" for REE. Ce and Nd isotopic compositions and REE a b u n dances of several Japanese limestonesCCarboniferous-^ Triassic) are measured. Isotopic data are corrected back to the formation age. Initial values are such that £ C e = - 0 . 7 - - 4 . 6 and £ N d = - 1 . 3 — 5 . 8 . On £ Ce vs. £ Nd diagram, these results are plotted in the third quadrantC £ Ce<0, £Nd<0), which are consistent with the study on Pacific Fe-Mn nodules. The possible sources of REE in sea water are continental c r u s t ( £ Ce>0, £ N d < 0 ) and depleted mantle ( £ C e < 0 , £ Nd>0). So Ce-Nd i s o t o p i c c o m p o s i t i o n s should basically be explained as the mixing of these two end members. When average values for MORB( £ Ce= - 2 , £ Nd=+ll,Ce/Nd=l.l) and crustal materials( £ Ce= +1.3, £ Nd=-ll) are chosen as two end members, the mixing line to fit the data of limestones requires Ce/Nd ratio for continental end member to be less than 0.1. This value is extremely low compared to the available data for crustal materials such as shales ( ~ 2 ) and river waters(0.5'-2.5). This indicates that isotopic compositions of sea water cannnot be explained as the r e s u l t of the simple mixing of these two end members. The effect of depleted mantle is greater in Ce than in Nd for the sea water in which Japanese limestones were formed. As another process, preferential removal of Ce in sea water and resulting mixing of Ce-depleted component and oceanic mantle must be taken into consideration to make any interpretation about Ce-Nd systematics in ocean.

U - P B GEOCHRONOLOGY O F A R C H E A N PLUTONISM IN T H E BARBERTON REGION, S. AFRICA: 800 Ma O F CRUSTAL EVOLUTION KAMO, S.L, DAVIS, D.W., Jack Satterly Geochronology Laboratory, Royal Ontario Museum, 100 Queen's Park, Toronto, Ontario, M5S 2C6, Canada, and DeWIT, M.J., Department of Geology, University of Cape Town, Rondebosch, 7700, Cape Town, South Africa The oldest rocks associated with the Barberton Greenstone Belt (BGB) are highly deformed tonalite-trondhjemite gneisses at the base of the Onverwacht Group (Steynsdorp Pluton, 3509 +6/-5 Ma; tectonically interleaved gneiss sliver, 3537 -I-6/-6 Ma). Volcanism in the Onverwacht Group is associated in time with emplacement of early plutons such as the Doornhoek Pluton (3448 -}-4/-3 Ma) and the Theespruit Pluton (3443 +4/-3 Ma). A felsic dyke (3467 +19/-7 Ma) intrudes the Komati Fm. Thrusting occurred at ca. 3450 in the southern part of the belt. A later thrusting event in the northern part of the B G B may be associated with Fig Tree and Moodies Group sedimentation and emplacement of the Kaap Valley tonalite at 3227 + / - 1 Ma. A sjTideformational ignimbrite found between the Fig Tree and Moodies Groups yields a preliminary age of 3227 + 8 / - 3 Ma, and a predeformational porphyry is 3229 + / - 3 Ma. Large sheet-like potassic plutons such as the Nelspruit and Mpuluzi plutons (3106 -I-2A3 Ma and 3107 + 4 / - 3 Ma, respectively) to the north and south of the BGB are coeval with syenitic magmatism (Boesmanskop Pluton, 3107 +1-2 Ma) within the belt. This episode may also be associated with deformation since it is coeval with the Stentor Pluton (3106 -1-6/7 Ma) which is affected by thrusting. A late phase of magmatism involved the emplacement of small granitic intrusions such as the Mpageni Pluton at ca. 2740 Ma.

53 THERMAL AND TECTONIC fflSTORY OF TARANAKI BASIN WELLS, NEW ZEALAND, ASSESSED B Y APATITE FISSION TRACK ANALYSIS KAMP, PETER J. J., Department of Earth Sciences, University of Waikato, Hamilton 2001, New Zealand, and GREEN, Paul F., Geotrack International, P.O. Box 4120, Melbourne University, Victoria 3052, Australia Apatite Fission Track Analysis (AFTA) has been applied to samples from four hydrocarbon well sections in the southern part of Taranaki basin (New Zealand). Samples from three of the wells (Fresne-1, North Tasman-1, Surville-l) show that the successions were exposed to higher temperatures in the past, through deeper burial, whereas in the fourth well (Kupe-1), formations are currently at their maximum temperatures since deposition. In Fresne-1, apatite fission track ages decrease rapidly with depth from - 8 3 Ma near the surface to Ma at 2.5 km in the well (present temperature of 79°C). Single grain fission track ages and track length data suggest that maximum paleotemperature, 80°C higher than present values, were attained in the mid-Miocene. Cooling, via uplift and erosion of 3.0 ± 0.3 km of section, started at - 1 2 Ma, and probably continued until ~8 Ma; within this interval uplift occurred at a rate of 0.75 mm/y. Apatite fission track ages also decrease with depth to values approaching stratagraphic age in Surville-l and North Tasman-1. Single grain ages and track length data suggest mid-Miocene maximum temperatures -40® and 25°C, respectively, higher than at present, corresponding to <2.0 ± 0.5 and 1.0 ± 0.3 km, respectively, of late Miocene-Pliocene uplift and erosion. In Kupe-1, apatite ages consistently increase with depth from - 4 0 Ma near the surface to - 1 3 6 Ma at the bottom. The single grain ages and track length data show no evidence of significant annealing as a result of heating due to burial. Rather, the AFTA data are dominated by apatite provenance, specifically, the thermo-tectonic development of the source areas. The fission track data for Kupe-1 highlight the role of the time scale of heating on the kinetics of annealing in apatite: rapid (<5 Ma) increases in temperature of about 55°C, bringing the temperature at the bottom of the well to 95°C, through burial, have not produced significant annealing in any except the deepest sample.

EVOLUTION OF THE LITHOSPHERE AND ITS INTERACTION WITH THE UNDERLYING MANTLE AS INFERRED FROM NOBLE GAS ISOTOPE DATA KANEOKA,Ichiro, Earthquake Research Institute, University of Tokyo, Bunkyo-ku, Tokyo 113, Japan, and TAKAOKAjNobuo, Faculty of Science, Yamagata University, Yamagata 990, Japan The continental lithosphere is considered to be relatively old and thick, but the oceanic lithosphere is not, which would reflect the difference in their evolutional processes. Ultramafic nodules from the continental and oceanic regions give us significant information concerning the natures of these lithospheres. Based on noble gas isotope data for such ultramafic nodules together with relevant host rocks, the evolution of the lithosphere and their relevance to the underlying mantle were examined for some typical areas including the African continent, Hawaii and some other areas. The noble gas data such as Ar and Xe isotopes for ultramafic nodules in South African kimberlites suggest that the evolution of the African continental lithosphere might have experienced a non-uniform evolution, implying the addition of the components from different sources in the mantle. Ultramafic nodules from oceanic hot spot areas such as Hawaii and Reunion Islands indicate signatures similar to those of MORB. This implies the addition of MORB source components during the growth of the oceanic lithosphere, which is definitely different from the circumstances of the evolution of the continental lithosphere. To characterize such components in ultramafic nodules. He isotopes are not always so effective to elucidate them because of sensitiveness to the secondary effects such as addition of radiogenic components, loss of He and/or equilibration of isotopic ratios through relatively long time. Apparently, the continental lithosphere seems to have experienced more complicated processes than the oceanic lithosphere through their evolution.

ICOG 7

ABSTRACTS

^"Ar/^Âr LASER-PROBE: N W OPPORTUNITIES M.I. Karpenko and V.V. Ivanenko, IGEM of the USSR Acad. Sci., Staromonetny 35, Moscow 109017, USSR The LMA-1 pulsed laser is being used in ^ÂiPÂi studies in our laboratory. Our very clean system, together with the high spatial resolution achievable and the momentary nature of the laser pulse, has made it possible to sample on the lO'^g level. The accuracy of the laser analysis system has been tested using standard samples. For example, a ÂiPÂi apparent age map using a 200 )im sampling grid has been made of a 0.3 cm^ piece of biotite with an artificially disturbed K-Ar system (Karpenko and Ivanenko, 1986). However, against the advantages of the pulsed laser method for local melting, is the disadvantage that single analyses cannot give any information on the disturbance of the K-Ar system of the sample. The attempts to avoid this drawback (Miiller et al., 1977; York et al., 1981) have not solved this problem. However a new method for the detection of disturbance of the K-Ar system has been worked out in our laboratory (Ivanenko and Karpenko, 1987). The method uses one inherent property of the laser beam, the predominant absorbtion of the beam energy on cr>'stal defects, which are mainly responsible for a heterogeneous distribution of the argon and for disturbance of the K-Ar isotopic system. The suggested method consists of a series of repeated laser shots to the same point of the sample. The resulting pattem of the apparent age spectrum is similar to the one found in the stepwise heating method.

ION MICROPROBE ANALYSIS OF " S r « r IN CARBONATES A. K. KENNEDY, I. D. Hutcheon, P. J. Wyllie and G. J. Wasserburg. Geol. & Planet. Sci., Caltech, Pasadena, CA 91125. Calcite and dolomite are ubiquitous, occurring in sedimentary, metamorphic and igneous rocks, mantle peridotites and alteration materials. Given the potential of high spatial resolution (10/im) Sr isotopic analysis of carbonates by ion microprobe, and the limited applications to date, we have compared low and high mass resolution techniques by analyzing carbonate standards with known ^St/^St and carbonates from a kimberlitic dyke. Ion microprobe measurements of "^Sr/^Sr are complicated by isobaric interferences (Caj, CaMgO, FeOj, CaNaO, CaPO, CaAlO, CaSiO, MnO^, CaSiO, CaK, CaCO^ and Rb) between masses 86 to 88. Previous studies [1,2] employed low mass resolving power (MRP) and peak stripping to correct for CaMgO, Caj and CaNaO interferences. Only Caj, Rb and CaK, cannot be resolved from Sr at MRP of 4500; however, MRPs between 6000 and 34000 are required to separate the interferences from each other. Thus, peak stripping is feasible only when interferences other than Ca2, CaMgO and CaNaO are negligible. Potential difficulties with high MRP include '"Rb and decreased sensitivity (2-5 x below low MRP). High and low MRP techniques give comparable "Sr/^Sr values for carbonates with high Sr (>2000ppm) or low concentrations of Na, Al, Si, P, K, Fe and Mn. For low Sr, only high MRP analysis gave correct results. The micaceous kimberlite dyke at Dixonville, PA contains carbonates with extremely heavy C and O isotopic compositions [3]. We analyzed groundmass carbonates and carbonate inclusions from a narrow (0.2m) section of the dyke which is zoned from the margin to the center in both C (6^^C=-4 to 24%c) and O (6"0=16 to 25%o). Our data show (1) low MRP analysis gives incorrect "St/^St Qow MRP 0.7147+0.0020, high MRP 0.7040 +0.0010, errors are 2a on the mean), (2) a wide range of " S t K t ratios (0.7040 ±0.0010-0.7319±0.0020) across the dyke and (3) the heavy C and O component is associated with the higher "Sr/®®Sr and reflects crustal contamination. [1] R.A.Exley & A.P.Jones (1983) CMP 83, 288-292. [2] D.E.ScatenaWachel (1986) PhD thesis, U. Chicago, 327p. [3] P.'Deines (1968) GCA 32, 613-625.

RADIOACTIVE DISEQUILIBRIUM DATING OF CORALS BY NUCLEAR TRACK DETECTION KASUYA. Masao, IKED A, Sumiko, Physics Department, Faculty of Science, Osaka University, Toyonaka 560, Japan Age determinations of fossil aragonitic corals are essential to the studies of Quaternary climatic and tectonic history, because formation of coral reefs is closely related to sea-level changes. We present here a simple method of radioactive disequilibrium dating of corals using nuclear track detectors. The natural alpha activity in a coral sample was monitored by an alpha-track plastic detector (CR-39), and the uranium content of sample was measured by the induced fission-track density in a muscovite detector after thermal neutron irradiation. Alpha activity of a sample increases witii increasing age, and becomes saturated with the transition from radioactive disequilibrium state to equilibrium state. The age of a sample can be estimated from the alpha-track productivity per uranium content Experimental results for 10 corals of various estimated ages (0 to 500,000 years) showed a good agreement with a theoretical fit and supported the validity of the method. We hope that tiiis low-cost technique will contribute to Quaternary geochronology.

CONTRADICTIONS IN THE ISOTOPIC AND ELEMENTAL SIGNATURES OF GRANITOID SUITES THAT INTRUDE MAJOR CRUSTAL BOUNDARIES IN EASTERN CANADA. KERR, Andrew, Geological Survey Branch, P.O. Box 8700, St. John's, NF, AlB 4J6, Canada; FRYER, Brian, and JENNER, George, Department of Earth Sciences, Memorial University, St. John's, NF, AlB 3X5, Canada. In Labrador, EN(i(t)CHUR values of Early Proterozoic (1.8 to 1.72 Ga), post-orogenic, granitoid rocks change from negative (-13 to -3) to positive (+2 to -1-6) across a line interpreted as the boundaiy between an Archean craton and an accreted "juvenile" Proterozoic crustal province. A belt of distinctive ''A-type" granites display a shift in € Nd ^om -6 to +6 across this boundary, but have constant elemental geochemical patterns on either side of it. In Eastern Newfoundland, Paleozoic plutons that intrude the platformal Avalon Zone have positive G ^ d values that probably reflect the short crustal residence period of their Pan-African basement relative to older basement of unknown affinity beneath the adjacent Gander Zone, where plutons commonly have negative G^d- ^ stitching pluton of "I-type" granite that intrudes the Gander-Avalon boundary displays an analogous internal change in Gjsjd from -1-2 to -5, but is generally homogeneous in elemental terms. These Nd isotopic patterns clearly indicate juxtaposition of distinct basement provinces, but trans-boundary contrasts are apparentiy invisible in the major and trace element patterns of these suites. This raises doubts about the use of elemental patterns in granites as "images" of postulated lower crustal source regions. Decoupling of elemental and isotopic patterns may indicate gross compositional similarity amongst lower crustal regions of contrasting ages, or a dominance of new, mantie-derived material over recycled crust in the petrogenesis of some granites. Significant isotopic contrasts within a unit interpreted as a single pluton, as observed at the GanderAvalon boundary, indicate that the isotopic heritage of local source regions can be retained within a single large magma chamber that is grossly homogeneous in elemental terms. This interesting feature suggests that single-sample isotopic data, as commonly used in regional studies of granitoid rocks, should be interpreted with caution. It also raises a number of interesting questions concerning petrogenetic models for granitoid rocks.

ABSTRACTS

ICOG 7

KLYUCHEVSKOY VOLCANO, KAMCHATKA USSR: GEOCHEMICAL EVIDENCE FOR CRUSTAL CONTAMINATION? KERSTING A B (Dept. of Geol. Sci., Univ. of Michigan, Ann Arbor. Michigan, USA) ARCULUS R J (Dept. of Geol. & Geophy., Univ. of New England, Armidale, Australia) KYSER T K (Dept. of Geosci., Univ. of Saskatchewan, Saskatoon, Canada) OKRUGIN V M, KHRENOV A P, and FEDOTOV S A (Inst. Volcan., Petropavlovsk, Kamchatsky, USSR) Klyuchevskoy volcano, located on the Kamchatka peninsula at the intersection of the Aleutian and Kamchatka trenches, is presently one of the most active arc volcanoes with a mean eruption rate of 0.7 m^s-i. The volcanic suite is dominated by basaltic lava flows and pyroclastic deposits that range in composition from high-MgO (10 wt%)-low-Al203 (13.5 wt %) (HMLA) to lowMgO (4 wt%)-high-Al203 (18 wt %) (LMHA). HMLA basalts contain olivine (Fo 80-90 cores), clino-and orthopyroxene phenocrysts all normally zoned in a groundmass of plagioclase, pigeonite and spinel. LMHA basalts contain plagioclase (An 82-55 cores), olivine (Fo 80-65 cores), clino-and orthopyroxene phenocrysts preserving both normal and reverse zoning in a groundmass of plagioclase, olivine and spinel. Two pyroxene thermometry indicates high temperatures of equilibration ranging from 1100 to 1190oC. Major and trace element variations are most consistenUy explained by an active 0'Hara-st>'le erupting-recharging-fractionating magma chamber in which fractionation of parental HMLA melts produce LMHA basalts. Radiogenic isotope ratios for Sr (87Sr/86Sr = 0.70353-0.70370) and Nd (i43Nd/i44Nd = 0.51302-0.51311) of the basalts define a narrow range that plots within the MORE field. The Sr isotope ratios are slightly more radiogenic than most ratios from the Kurile and Aleutian basalts. Thegi^o value of whole rocks (+7.2 to +7.8%o), plagioclase phenocrysts (+8.3 to +8.5%o) and olivine phenocrysts (+6.9 to +7.4 %o) are higher than the range for typical mantle values. Together, Sr and O isotope ratios indicate that crustal contamination is involved in the evolution of the Klyuchevskoy basalts. Aqueous fluid inclusions hosted in plagioclase phenocrysts in the LMHA basalts have gD values from -100 to -150 %o. These data coupled with low homogenization temperatures (80 - 180oC) suggest that some flows have been hydrothermally altered by meteoric water. Hydrothermal alteration by meteoric fluid is limited in extent as stable isotope ratios for the phenocr>'Sts remain undisturbed.

THERMOCHRONOLOGY OF THE CENTRAL APPALACfflAN PIEDMONT: CONSTRAINTS FROM HSSIGN-TRACK DATING KOHN. Barry, P. Dept. of Geology, Ben Gurion University, Beer Sheva 84105, Israel, LUTZ, T.M., WAGNER, M.E. and OMAR, G.I. Dept. of Geology, University of Pennsylvania, PA 19104, USA. A total of 121 sphene, zircon and apatite fission track ages (FTA) have been determined on Grenville and Paleozoic crystalline rocks, and the Mesozoic Newark Series of the central Appalachian Piedmont. Most sphene and zircon FTA are almost concordant within analytical errors at -190 ± 20 Ma and indicate rapid cooling from sphene to zircon closure temperatures (-280° to 200°C) between times indicated by the ages. Near concordance of sphene and zircon FTA over the study area (-16,000 sq km) is indicative of cooling following total annealing during a regional thermal event. The Mesozoic thermal event may have resulted from localised Early Jurassic tholeiitic magmatism at -201 Ma recorded in the Newark Basin or a hydrothermal event postulated in Newark Series sediments at -175-178 Ma . Although the two events are separated by - 2 5 Ma the precision of the FT data cannot resolve between the two possibilities. However, the near concordance of sphene and zircon FT ages over a large tract of crystalline basement enclosing the Newark Basin and largely devoid of Early Jurassic magmatism strongly suggests that magmatism and hydrothermal activity alone were not responsible for extensive resetting of the FT clocks. We view the anomalous Early Jurassic thermal regime as a consequence of lithospheric thinning and Late Triassic-Early Jurassic rifting along the eastern North American margin. Consideration of sphene and zircon closure temperatures, Rb-Sr data from coexisting micas, Newark series stratigraphy and crustal extension models indicate that Mesozoic rifting in the study area produced minimum paleothermal gradients of-50°C/km. Apatites from basement and Newark Series rocks mostly yield Late Jurassic-Early Cretaceous FT ages. Apatite FTA and confined track length data have been used for modelling thermal histories. The models, using a Monte Carlo routine, indicate a discrete phase of Late Jurassic-Early Cretaceous denudation (-3 km). A similar phase of cooling indicated by recent apatite FT studies in the Northern Appalachian basin suggests a regional extent for this event.

AGE OF THE CENOMANIAN-TURONIAN BOUNDARY IN THE WESTERN INTERIOR OF THE UNITED STATES Kowallis, B.J., Christiansen, E.H., Department of Geology, Brigham Young University, Provo, UT 84602, USA, and Deino, A.L., Geochronology Center, Institute of Human Origins, Berkeley, CA 94709, USA New high precision *°Ar/"Ar laser-microprobe ages of individual sanidines from three bentonites spanning the Cenomanian-Turonian boundary in the Western Interior of the United States suggest that the boundary may be older than generally reported. The lowermost bentonite comes from the Late Cenomanian Sciponoceras gracile biozone and gives an average age of 93.50 ± 0.18 Ma (2a, standard error of the mean) for 14 grains. The middle bentonite comes from the Late Cenomanian Neocardioceras juddii biozone and gives an average age of 93.33 ± 0.12 Ma for 29 grains. The upper bentonite comes from near the base of the Turonian immediately above the first occurrence of the basal Turonian bivalve species Mytiloides and gives an average age of 93.46 ± 0.36 Ma for 12 grains. These ages are in good agreement with K-Ar biotite ages (92.8± 1.8 Ma, 93.7± 1.8 Ma, and 96.6±4.8 Ma) obtained by Obradovich & Cobban (1976, recalculated with new constants and reported by Odin & Obradovich, 1982) from Late Cenomanian bentonites of the Western Interior; they may possibly be the same bentonites we dated. These workers also report a 91.0±1.8 Ma age from an Early Turonian bentonite from the Inoceramus labiatus biozone from the Western Interior. These biotite ages have not been considered reliable, however, because the biotites contained less than 5% potassium. Our ages are also compatible with Early Turonian, K-Ar biotite ages (ranging from 91.511.8 to 93.6±2.4 Ma) obtained by Lanphere (1982) from the Seabee Formation in Alaska. These Alaskan biotites had K contents of 5.5% to 7.2%. Based upon the high precision age data, we propose that the age of the Cenomanian-Turonian boundary in North America is about 93 to 93.5 Ma. The currently accepted age for this boundary is reported as 91 Ma in all the major published time scales. The 91 Ma age for the boundary is derived primarily from the large number of glaucony ages from European strata.

CARBON AND NITROGEN ISOTOPES OF INDONESIAN VOLCANIC GASES KREULEN. Rob, MEESTERBURRIE, Jan, POORTER, Rene, Department of Geochemistry, University of Utrecht, P.O.Box 80021, 3508 TA Utrecht, The Netherlands, and SRIWANA, Terry, Volcanological Survey of Indonesia, Jalan Diponegoro 57, Bandung, Indonesia. The Indonesian Sunda-Banda arc volcanoes are related to subduction of oceanic crust in the western part of the arc and subduction of Australian continental crust in the eastern part. Volcanic gases were sampled from volcanoes in the transition zone between oceanic subduction and continental subduction. Stable isotopes of the gases show relationships that seem to correlate with the ^He/'*He data recently reported by Hilton and Craig (1989) for the same sampling sites. Egon volcano with low ^He/''He (mostly crustal or sedimentary helium) has a relatively high 5"C of -l.SVoo and positive (5"N values up to +3Voo* Lewotolo volcano with high ^He/'^He (more mantle helium) has an average i5"C of -3.6Voo and negative <S"N values down to -5Voo« Different fumaroles on Lewotolo volcano, covering a range in vent temperatures from lOOX to more than 400®C, show a remarkably narrow range in Nitrogen isotope ratios vary from sample to sample, reflecting different proportions of air contamination. At the time of writing this abstract, the analytical work was not completed. Sulfur and chlorine isotopes will probably be available at the meeting.

ICOG 7

56 CONVENTIONAL U-PB GEOCHRONOLOGY SINGLE ZIRCON GRAINS. T.KROGH Dept. of Geology, Royal Ontario Museum, Toronto, Ontario, Canada, M5S 2C6.

Precise U-Pb analysis of single grains and parts of grains from Archean mafic granulites containing as little as 8 ppm U indicate a sequence of core formation during dehydration, overgrowth formation during rehydration and cross cutting pegmatite intrusion after brittle fracturing. The same sequence is found at deeper levels but all stages are younger, implying a form of accretionary underplating by supracrustal rocks in the interval 2650 to 2585 Ma. Pebbles as young as 2680 Ma prove that rocks formed at least 20 Ma after the cessation of volcanism are present. Gold deposits in the world's most productive greenstone belt nearby are coeval with the underplating. Abrasion to remove secondary lead loss allows precise two stage discordia to be defined. One Grenville quartzite yields a 1687-996 line within error for a ± 2 Ma *Tb/^Pb age uncertainty, for 2 single and 2 multigrain fractions from 2 samples. Data for 7 of 8 fractions from 2 paragneiss samples are colinear between 1744 and 1062 Ma. Melt pods in the Killarney area yield metamorphic zircon and titanite that are colinear between 1453 and 982 Ma (some Grenville!) and titanite and zircon yield a 21 point line between 1651 and 988 Ma in the Front Zone in Labrador. Melts were produced below thrust sheets in the 1651 batholith. Data for six single core free grains from the melts indicate a 2 sigma uncertainty for a single analysis of ± 3.5 Ma with a mean age of 995 ± 1.6 Ma. Four of the six contained between 1.5 and 2.5 pg total common Pb, a fifth contained 4 pg while the sixth had 29 pg (but fell on the bench during loading). Measured ratios of 206 to 204 lead in four of the six samples are between 20,000 and 52,000.

SINGLE ZIRCON EVAPORATION AS A TOOL IN UNDERSTANDING EARLY ARCHAEAN CRUSTAL EVOLUTION: THE BARBERTON-SWAZILAND TERRAIN KRONER, Alfred, Institut fiir Geowissenschaften, Universitat Mainz, Postfach 3980, 6500 Mainz, FRG Ambiguities in age relations between different lithological units in the Barberton greenstone belt (BGB) of South Africa, surrounding TTG plutons and rocks of the Ancient Gneiss Complex of Swaziland could be resolved by high-precision zircon dating using the single grain evaporation method. Dacitic flows and pyroclastic rocks in die upper Onverwacht Group yield 207pb/206pb ages between 3445±6 and 3416±10 Ma (2a errors) and contain older crustal components reflected by a 3504±8 Ma old zircon xenocrystThese data document ~30 Ma of felsic volcanic activity which is coeval with emplacement of some of the TTG plutons bordering the BGB and dated at 3440±4-3445±3 Ma. The small Steynsdorp Pluton was emplaced at 3490±4 Ma and thus predates the Onverwacht volcanic suite. It contains tonalitic xenoliths and zircon xenocrysts as old as 3531±3 Ma and suggests that such plutons are unlikely to be derived from short-lived mafic precursors but from older tonalitic crust that may have existed up to 200 Ma prior to greenstone formation. Zircons from dacitic tuff and agglomerate in the basal Fig Tree Group have ages between 325915 and 3241±6 Ma, and one sample contains an extraordinary complex population of zircon xenocrysts with ages up to 3522+4 Ma, again suggesting involvement of preOnverwacht sialic crust in the generation of the felsic melts. Zircons from upper Fig Tree felsic rocks have ciystallization ages of 322513 and 322714 Ma and also contain rare xenocrysts as old as 348414 Ma. The Kaapvalley tonalite pluton (322615 Ma) is coeval with upper Fig Tree felsic volcanism. The Fig Tree felsic cycle also lasted for - 3 0 Ma and documents a significant hiatus of -150 Ma between the Onverwacht and Fig Tree Groups. The zircon ages unambiguously prove that felsic units in the BGB stratigraphy are not due to tectonic repetition. The new zircon ages provide a maximum age of about 3450 Ma for the beginning of BGB deposition and also show tiiat the Onverwacht and Dwalile greenstones (Swaziland) are coeval. The latter are interbedded with continent-derived sediments, and the frequent occurrence of zircon xenocrysts in Onverwacht felsic flows also argues for a pre-greenstone basement below the BGB depository.

ABSTRACTS

2.0 GA OLD CARBONATITE COMPLEX IN PRECAMBRIAN CHARNOCKITES, HOGENAKAL, SOUTH INDIA. ANIL KUMAR and GOPALAN^K, National Geophysical Research Institute,Hyderabad; RAO,B.B, and NATARAJAN,M, Geology Indian Institute of Technology,Bombay. Hogenakal carbonatites associated with pyroxenite dykes are emplaced along NE-SW trending major crustal faults through late Archean charnockites. We report preliminary Rb-Sr evidence that these rocks are one of the rare relicts of > 2 Ga old carbonatites in the world with its implication to the understanding of sub-Dharwar mantle at that time and the antiquity of the above deep seated lineaments. The complex consists of syenite, pyroxenite, silicate-sovite and sovite with their mutual relations indicating this sequence of their formation. Similarity of their REE patterns and trend of LREE enrichment suggest their genetic association as possibly products of fractionation of a primary alkaline ultramafic magma. A good Rb-Sr internal isochron based on apatite, calcite, pyroxene and biotite separated from a silicate-sovite sample gives an age of 2008 + 26 Ma and Sr^ = 0.7017 +1. If the biotite Ts secondary, the primary age of the complex may be even older. Sr C^) ^^ -5.5 +0.5 implies a source with a substantial prior history of LIL depletion. If the emplacement was along pre-existing crustal faults, the latter must predate 2 Ga. In the light of the present result, the much younger mineral ages of -the nearby Sevattur carbonatite complex are being reexamined.

OXYGEN ISOTOPE EXCURSIONS IN SOME ARCHAEAN AND PROTEROZOIC CARBONATE SUCCESSIONS OP INDIA KUMAR. B.. MATHUR, R., National Geophysical Research Institute^ Hyderabad 500 007, India, and TEWARI, V.C., Wadia Institute of Himalayan Geology, Dehradun 248 001, India The oxygen isotope data of ArchaeanProterozoic carbonate successions from parts of South India and Lesser Himalaya show a positive (5l9o trend from +16 to +23 %o (SMOW) with decreasing age (middle Archaean to middle Proterozoic) followed by sharp d^^O excursions (+17 to +29 %o vs. SMOW) at the end of Proterozoic. The excursions of middle-late Proterozoic age relate to marked changes in evolutionary stages of life and carbon budget. The lighter ISO/ISQ ratios of marine sedimentary carbonates with increasing age have been attributed to : i) Post depositional equilibrium exchange with isotopically light meteoric waters, or (ii) deficient nature of the earlier ocean waters. Evaluation of our data add another possible interpretation to these trends. We suggest that positive trend with decreasing age may reflect progressive increase in atmospheric oxygen concentration and sharp fi^^O excursions during late Proterozoic may point to large changes in atmospheric oxygen level from Pasteur level to Present Atmospheric Level (P.A.L.) or higher than P.A.L.

ABSTRACTS

IC0G7

'•OAr/aaAr AGE SPECTRA OF SHOCKED MICROCLINE FROM THE CENTRAL UPLIFT OF THE MANSON IMPACT STRUCTURE SUPPORT A LINK WITH THE K-T BOUNDARY KUNK, Michael J., U.S. Geological Survey, Reston, VA USA The discovery of an iridium anomaly within K-T boundary sections was the first suggestion of a bolide impact(s) related to that boundary. This iridium anomaly has since been found in K-T boundary sections worldwide. The impact hypothesis also has been strengthened by the discovery of shocked mineral and lithic grains within the K-T boundry layer, however, the location of any crater(s) that can be linked to the boundary event has been elusive. 40Ar/39Ar age spectra results on shocked microcline from a core taken from within the crystalline central uplift of the Manson Impact Structure (MIS), located in Iowa, suggest that it is a likely candidate for a K-T boundary event. The age spectra compare favorably with theoretical diffusion models and suggest that the microcline was partially reset at 65.7±1.0 Ma with 98% argon loss at the time of impact. The mechanism of argon loss is thought to be shock induced heating. Other lines of evidence that also suggest that the MIS is a likely candidate include (1) the worldwide size and abundance of shocked grains within the K-T boundary layer which are greatest in the Western Interior of North America and suggest a nearby impact, and (2) compositional similarities between the mineralogy of the shocked grains found within the K-T boundary layer and that of the target rocks at the MIS. Although the MIS is somewhat smaller (35 km) than what would be expected for a K-T bolide impact, existing evidence warrents its further investigation.

THE GEOCHEMISTRY OF TAIWAN GRANITOIDS A N D ITS TECTONIC IMPLICATION FOR SE C H I N A L A N . C h i n q - Y i n q ^ , L E E , Typhooni and J A H N , Bor-Ming2; ^ I n s t . Earth Sci,,Academia Sinica,Taipei ROC; ^Inst, Geologie, Univ. Rennes, France To better understand the crustal evolution of southeastern C h i n a , we have studied the geochemistry of the granitic gneisses in the basement of T a i w a n . The chemical compositions of the gneisses show that they are peraluminus in character with normative corundum. The Sr-Nd isotopic compositions of the gneisses are highly v a r i a b l e . Initial end (at 90 Ma B P ) for gneisses range from 0 to -12 and initial ^"^Sr/^^Sr cluster around 0.7051 to 0 . 7 1 0 8 . These values suggest that the source material of the gneisses are heterogeneous and are resulted from the mixing between the mantle material such as the adjacent oceanic metabasites (end 4 to 11) and the old continental crust such as the nearby metapelites (-7 to - 1 1 ) . The amount of the continental crust component inferred varies from 10 to 80 % . Nd crustal resident ages w . r . t . the depleted m a n t l e reservoir ( T D K ) for all gneisses range from 0.9 to 1.5 Ga and are in general agreement with the U - P b zircon upper intercept age of 1.0 to 1.7 Ga (Jahn et a l . , 1 9 8 6 ) . These old model ages imply that since late Proterozoic the southeastern margin of the Asian continent has witnessed primarily the recycling of old continental crustal m a t e r i a l . The m u l t i p l e magmatic and metamorphic events caused by the repeated collisions with and subduction of oceanic plates (paleo-Pacific) added only minor amounts of fresh input from the m a n t l e .

HIGH-RESOLUTION ^^At/^^AT GEOCHRONOLOGY OF TERTIARY VOLCANIC ROCKS, WESTERN U.S.A. LANPHERE, M.A., U.S. Geological Survey, Menlo Park, CA 94025, U.S.A., SAWYER, D.A., U.S. Geological Survey, Denver, CO 80225, U.S.A., FLECK, R.J., U.S. Geological Survey, Menlo Park, CA 94025, U.S.A. ^°Ar/®®Ar measurements on biotite and sanidine from silicic volcanic rocks of the central San Juan caldera cluster (CSJ) and the Southwestern Nevada Volcanic Field (SWNVF) indicate that ages can be measured with analytical precisions, as given by weighted standard deviations, of less than 0.4%. In the CSJ eight major ash-flow tuffs were erupted from seven calderas during a period of 2 m.y. about 26-28 Ma. The pooled estimate of the standard deviation for weightedmean age-spectrum plateau ages is about 0.5%. The age of the Fish Canyon Tuff, the most voluminous ash-flow tuff in the CSJ, based on age-spectrum plateau ages of biotite and sanidine is 27^55 + 0.08 Ma. This agrees with conventional K-Ar ages of 27.55 + 0.17 on biotite and 27.53 ± 0.23 Ma on hornblende from the Fish Canyon Tuff. Ages of coexisting minerals from a single sample or single minerals from two samples of the same tuff indicate that the age of an ash-flow tuff can be determined with an analytical precision of 0.3-0.4% (Is). The volcanic units of the SWNVF comprise 10 major assemblages, each having distinct petrologic characteristics. Analyses of sanidine using a continuous laser dating system yielded total-fusion ages with analytical precisions of 0.6-0.7%. Averages of 4-6 sanidine analyses yielded ages with weighted standard deviations of less than 0.3% for tuffs 7.5-16 Ma. The ^°Ar/39Ar data suggest that large-volume silicic caldera volcanism is episodic within the multiple-caldera fields of the CSJ and SWNVF. Major caldera-forming ash-flow tuff eruptions were separated by 0.2-0.3 Ma in CSJ, while in SWNVF repose intervals ranged from 0.3-1 Ma. Most related pre- and post-caldera lava eruptions took place within 0.1 Ma of caldera formation.

HIGH RESOLUTION CHRONOSTRATIGRAPHY ROCCAMONFINA VOLCANO (ITALY)

A. Ballini^ F. Barberi^, F. Mezzctti^, M. Oddone^, l.M. Villa^ 1- 1st. Geocronologia CNR, Pisa; 2-Dip. Scienze d. Terra, Pisa; 3- Dip. Chlmica Genende, Pavia.

Roccamonfina Volcano, between Rome and Naples, is famous for being the type locality of numerous geochemical and isotopic studies (see Taylor et al, EPSL4i (1979) 81, & references therein). It is a strato-volcano and lies in the southern part of the Roman Volcanic Province (RVP). From a geothermal well we obtained a new series of samples extending all the way through the volcanic deposits including 2 lavas with lower K ( a K S " , also called •^KS"), underlying HKS pyroclastites, whose biotites yield between 650 and 600 ka bp. In our approach to volcanological reconstruction, the first stage is establishing a correlation map of stratigraphic columns of the outcrops and the recognition of key marker beds. ^^Ai/^^Ar stepwise heating on high'potassium minerals (sanidine and leucite, mostly from pumices) allows a fairly good time-resolution, and thus a tight constraint on the time-scale of volcanic evolution. The data now available indicate that the volcano is probably younger than previously thought. Among the first products there are fissural eccentric lavas with high-K affinity, one of which yields 600±3 ka. A HKS pyroclastite recovered from the drillcore yielded 586±2 ka. The age around 590 ka is coeval with the onset of activity in the northern RVP. The volcanic edifice was built up by a series of lavas and pyroclastites. 7 samples from two thick pyroclastic columns, representing a great part of the stratocone, indicate that the activity was almost continuous from about 400 to 335 ka. A huge ignimbrite, Brown Leucitic Tuff (335±3 ka) closed the HKS activity, shortly followed by the White Trachytic Tuff. This KS ignimbrite is now shown to be subdivided into at least three main events (318±2, 301±2, <283 ka). The final activity consists of intracalderic lava dome extrusions. We were able to show: 1. a substantial contemporaneity in the onset of volcanic activity along a 300 km-long line, from Vulsini to Roccamonfina; 2. a flawless agreement between stratigraphy and chronology; 3. the scarcity of excess Ar in most, but not all, spectra; 4. the close .proximity of HKS and KS eruptions: the switch takes a few ka only.

ON THE ARGON RELEASE MECHANISMS OF HYDROUS MINERALS IN VACUO LEE. James K.W., LO, Ching-Hua, and ONSTOTT, Tullis C., Department of Geological and Geophysical Sciences, Princeton University, Princeton, New Jersey, 08544, U.S.A. The release mechanisms of argon from hydrous minerals have been examined in a series of in-vacuo heating experiments utilizing 40Ar/39Ar isothermal and step-heating runs and in-situ high temperature X-ray diffractometry. The results demonstrate that at temperatures above the range of mineral stability, argon release is dominantly controlled by the structural breakdown of the minerals. At lower temperatures, however, argon release appears to follow known diffusion mechanisms. Calculated D/a^ values at these temperatures are approximately 2-4 orders of magnitude higher than those obtained from previous hydrothermal experiments. From the low-temperature isothermal-heating experiments, the D/a2 values are very consistent throughout most of the experiment, except in the first few steps. Although other factors that could potentially affect this variation of D/a^ values were evaluated, such as grain size distributions within the sample and the use of approximation equations for volume diffusion, they cannot explain this phenomenon. Rather, this may be a result of a small but finite lag time in sample heating and the effect of crystal dislocations on the release of argon. We suggest that the low-temperature j^rtion of any apparent-age spectrum reflects an argon distribution controlled by either structural defects/dislocations in the crystal and/or volume diffusion. Moreover, because the decomposition of the hydrous minerals at higher temperatures (> 600' C for biotite and > 900' C for hornblende) strongly influences the release of argon, ^ÂiPÂx ratios would tend to be homogenized for the gas released in these temperature steps. In order to obtain more accurate information about the argon isotopic distribution through ^Âr/^Âr step-heating, temperature schedules should include more steps below the decomposition temperatures of the respective hydrous minerals. In general, any interpretation of ^^krf^Âi apparent age spectra must carefully consider the stability range in vacuo and at elevated temperatures of the materials to be dated.

ICOG 7

ABSTRACTS

COMPOSITIONAL VARIATIONS ACCOMPANYING METAMORPHISM OF SUBDUCTED OCEANIC LITHOSPHERE: IMPLICATIONS FOR GENESIS OF ARC MAGMAS AND MANTLE REPLENISHMENT LEEMAN. W.P.. MORAN, A.E., and SISSON, V.B., Keith-Wiess Geological Laboratory, Rice University, Houston, TX 77251 Recycling of subducted oceanic crust and sediments is widely invoked to account for compositional distinctions between volcanic arc magmas and those from other tectonic settings. Correlations between B and short-lived cosmogenic ^^Be (Morris, Leeman, and Tera, 1990) in magmas from five arcs strongly support derivation of B and ^^Be from a common source - most plausibly subducted sediment. However, our geochemical studies of numerous metamoiphic suites suggest that many elements that are enriched in arc magmas (e.g., Cs, Rb, B, As, Sb, Pb, W, U) are progressively depleted in igneous and sedimentary materials as they follow a prograde subduction geotherm. In fact, LREE normalized ratios (e.g., Pb/Ce) of the above elements are commonly similar to mantle values in high grade equivalents of these materids. This evidence precludes simple addition of bulk sediment or altered oceanic crust to the mande as a means of creating arc magma sources. Aqueous fluids can efficiently extract these elements from subducted materials, effectively resulting in their one-way fractionation into the overlying accretionary prism and mantle wedge and limiting return transport to the deep mantle. Addition of small amounts ( « 1 wt. percent) of such fluids to 'primitive' mantle compositions can easily produce bulk compositions that upon partial melting are capable of generating many arc basalts. There remain numerous problems in explaining how mass transport physically occurs. Most fundamental is how these elements are retained to depths (ca. 100 km) below volcanic arcs given their proclivity to enter dehydration fluids at shallower depths. Detailed explanations must await more information on actual thermal gradients and element solubilities as functions of P, T, and fluid composition. The effiKtiveness of subduction recycling in buffering mantle composition is uncertain but possibly overexaggerated, at least for those elements mobilized in subduction zone fluids.

SEDIMENTATION RATE BASED ON i^Be MIXING MODEL LEE. Typhooni>2, y o u , Chen-Fengi and LIU, Tsung-Kuei2; ^Inst. Earth Sci,,Academia Sinica,Taipei ROC; ^Inst, Geology, Nat'l Taiwan Univ., Taipei, ROC

KOMATIITES FROM KAMBALDA: WHY SOME ARE CONTAMINATED AND OTHERS ARE NOT. LESHER. C.M. Department Geology, Univ Alabama, "I uscaloosa, AL 35487, USA, and ARNDT. NT. MaxPlanck-lnstitut fur Chemie, Saarstr 23, 6500 Mainz, FRG

We have applied the i^Be mixing models of Brown et al. (1985) developed for the sea off southern California to the data of You et al (1989) to estimate the sedimentation rate for the continental shelf around Taiwan. Our model-dependent rates for 31 sites increase from 14 cm/kyr at 30 km offshore to 370cm/kyr near shore and compare favorably with those directly determined from i^C. This new approach does not rely directly on radioactive decay but depends on modeling the lORe in the shelf sediments as mixtures of continental runoff and pelagic inflow. It does not require special sample material and is insensitive to turbation effects. The larger the contrast between the continental runoff and the pelagic inflow is, the better the resolution of the technique is. Thus it is particularly useful in tectonically active areas where the high erosion rates leads to low residence time on the surface hence low Be concentration in the runoff. Our estimated sedimentation rates represent 102-103 years averages in modern basins west of Taiwan. The paleo-basins now onland Taiwan have sedimentation rates increasing from 10 cm/kyr in the Miocene to 400 cm/kyr in the Pleistocene. We speculate that this increase is the result of a westerly movement of the shoreline towards the deposition location. Such a migration of shoreline was in turn caused by the crustal shortening and uplifting owing to the arc-continent collision which started 5 Ma BP in eastern Taiwan.

Komatiites at Kambalda have morphological and chemical features that clarify the manner in which they interacted with continental crust during ascent and eruption. The lower komatiite member comprises differentiated flows with thick, largely cumulate central channels, and thinner flanking segments, interlayered with sediments. The upper member contains thinner differentiated spinifex flows and massive porphyritic units. Major and trace elements and Nd isotopic compositions of the komatiites vary widely but change systematically with stratigraphic position. Spinifex-textured lava in central channels is highly magnesian (16-31% MgO) with depleted LREE ([La/Sm]N = 0.44-0.58) and high eNd (1.7 to 4.5). Flanking segments are more evolved (MgO=1221%), [La/Sm]N=0.49-0.87, eNd =2.2), as are lavas in the upper member (MgO=16-25%), [La/Sm]N=0.57-0.82, eNd =2.7). These chemical and isotopic variations are attributed to differences in the extent of contamination, augmented by olivine fractionation and accumulation. During the main eruptive phase, flow rates were high, flowage was turbulent, and wall rocks encountered by the komatiite were actively eroded and assimilated. The flanking segments of komatiite flows in the lower member represent contaminated lava that overflowed the main channels during this phase of eruption. Later, as eruption rates declined, the flow regime changed from turbulent to laminar, and thermal erosion ceased. Virtually uncontaminated lava then filled the central channels, purging" them of contaminated lava. Komatiites in the upper member may owe their contaminated character to interaction with crust that had been heated by the passage of earlier komatiitic magmas and was therefore more readily assimilated, or to volcanological influences that prevented the emplacement of late-stage uncontaminated lava .

ABSTRACTS

ICOG 7

HOT-SPOT TRACKING AROUND THE U.K? Cherry L.E. LEWTS. Anthony J. HURFORD Fission Track Research Group, University College London, Gower St, London WCIE 6BT, England and Paul F. GREEN, Geotrack International, Department of Geology, University of Melbourne, Parkville, Victoria 3052, Australia. Apatite fission track analysis has been used to evaluate the thermal and tectonic history of the Irish Sea, Cheshire Basin and surrounding onshore regions of the British Isles. Single grain ages and track length data suggest that maximum paleotemperatures were reached at the end of the Cretaceous and that this was followed by significant uplift during the Early Tertia^. These results support previous work which also identified uplift at this time both in the Lake District and the East Midlands thereby confirming a picture of regional doming during the Tertiary rather than localised episodes of inversion. The timing of this uplift event coincides with the opening of die North Atiantic and fits well with models that predict significant doming over low-density mantie hot-spots prior to a major rifting event. However, old apatite fission track ages (>300 Ma) found right across Scotiand from the Shetiand Isles to the Western Isles do not substantiate this episode of uplift despite their close proximity to well known sites of Tertiary igneous activity where elevation might be expected to be at its peak. Further anomalous results are seen in granites which record significantiy younger ages than tiie surrounding sediments. Since pdaeotemperatures derived from the sediments indicate that normal geothermal gradients existed during the Tertiary, it appears tiiat the granites are local 'hot-spots' acting either as 'chimneys' for the increased mantie heat flow predicted attiiattime,or as pathways for hot fluids channelled up deep-seated ft-actures which have been reactivated during a change in the regional stress regime.

Sm-Nd ISOTOPIC CHRONOLOGY OF THE MAJOR TECTONIC EVENTS FOR THE QINLING-DABIE OROGENIC BELT LI Shuguang. LIU Deliang, CHEN Yizhi, GE Ningjie, University of Science and Technology of China, Hefei, Anhui 230026, PRC, HART, S.R., Massachusetts Institute of Technology, Cambridge, MA 02139, USA, ZHANG Guowei, Northeastem University, Xian 710069, PRC, and ZHANG Zongqing, Chinese Academy of Geological Sciences, Beijing 100037, PRC Thetimeof ophiolite obduction events and a collision event between the North and South China blocks have been determined by the SmNd mineral isochron method for the Qinling-Dabie orogenic belt, China. An age of 983 ± 140 Ma for a garnet amphibolite aureole of the Suangshugou Alpine peridotite and an age of 403 ±17 Ma obtained from the Lajimiao gabbro-norite indicate the existence of two ophiolite obduction events in the late Proterozoic and Palaeozoic in the North Qinling belt. The collision melange developed in the Dabie mountains is represented by C-type eclogite blocks and ultramafic massifs which are emplaced in solid state into the Dabie Group — a Proterozoic metamorphic complex. Three identical ages, 244 ± 11 Ma for an eclogite in the Raobazhai ultramafic massif, and 224 ± 20 Ma and 221 ± 5 Ma for C-type eclogites from the North and South Dabie Mountains respectively, document that the collision melange was formed in the early Triassic. Three major orogenic stages for the Qinling-Dabie orogenic belt are suggested by the above results: (1) in the later Proterozoic and earlier Palaeozoic, the North Qinling belt was like the trench-arcbasin system of the western Pacific ocean, (2) in the later Palaeozoic, the rear arc basin was closed andtiieNorth Qinling belt was like the Andes belt on the eastern Pacific coast, (3) in tiie early Triassic, the North and South China blocks were united and the Qinling-Dabie orogenic belt was like the Alpine belt — a continentalcontinental collision orogenic belt.

H-0 ISOTOPE GEOCHEMISTRY OF AN A-TYPE GRANITE STOCK, MANZISHAN, NE CHINA LI Pei-Zhong, and YU Jin-Sheng, Guangzhou Branch, Institute of Geochemistry, Academia Sinica, Wushan, Guangzhou, Guangdong, China. The Nianzishan A-type granite stock is located in the Heilongjiang province, north east China. From a H-0 isotope study, two important discoveries were made. (1) The 51^0 values may be contoured with the highest values in the centre and decreasing toward the margins of the stock. This suggeststiiatmeteoric ground water was supplied from the outside and heated by tiie stock itself. It can be modelled by computer calculation. That means it would be a common phenomenon for a magma rock pluton or stock. (2) There is a vertical relationship between the oD values and water content for the whole-rock samples of die stock and the 5D values of the recent meteoric water of the local area are so high that the 5D of the stock cannot be decreased to -145%o. This means that the Ddepletion of the stock can be accounted for by an open-system magma degassing process.

Sr-Nd-Pb-0 ISOTOPIC CONSTRAINTS ON THE ORIGIN OF SILURIAN GRANITOID ROCKS IN ZHUGUANGSHAN BATHOLITH, SOUTHEAST CHINA : IMPLICATIONS FOR THE CRUST DEVELOPMENT LI Xianhua, Institute of Geochemistry, Academia Sinica, Guangzhou 510640, China, TATSUMOTO, M., U.S.Geological Survey, M.S. 963, Box 25046, Denver, Co. 80225, USA. Whole-rock Sr, Nd, O and K-feldspar Pb isotopes have been determined for the Silurian granitoid rocks in Zhuguangshan Batholith, southeast China. Sr-Nd, Pb-Pb and Sr-0 isotopic systematics, topology in six-dimensional isotopic multispace and mixing model calculations indicated that the granitoid rocks were derived from the Proterozoic basement rocks without significant mantle-derived magma and granulite facies lower continental materials. The source materials were fairly heterogeneous, and mainly composed of continent-derived sedimentary rocks, geosynclinal sediments and altered volcanic rocks. Based on Nd model ages relative to DM and relict zircon U-Pb ages, the primary crust probably formed 2.5 Ga ago, which consisted of the core of the Archaean continental crust A large-scale differentiation of mantie/crust and an increase oftiiecontinent took place at about 1.8 Ga, consistent withtiieworldwide crust formation event of 1.7-1.9 Ga. There was another important regional crustal increase during magmatism-tectonic movements of 1.3-1.4 Ga, in which the previous crust was reworked and/or remelted intensively.

ICOG 7

ABSTRACTS

Th-U-Pb EVOLUTION OF THE MORE MANTLE AND UPPER CONTINENTAL CRUST T.C. LIEW and A.W. Hofmann, Max. Planck Institut fiir Chemie, Postfach 3060,6500 Mainz, FRG Pb isotopic data for early Archean galenas indicate the Earth's primitive mantie had K(23^Th/204pb)=4.3 and ^i(238u/204pb).8.5. Relative to these parameters, N-MORBs and the PREMA tovalent mantle) composition of Zindler and Hart (1986) have 20^/204p5 and 208p5/2(Mp5 values that indicate derivation from sources with time-integrated high-^, low K and low Th/Pb. MORBs display a 'U paradox', rather than a 'Pb paradox', in the sense that the Th/Pb decrease is consistent with the melt-residue incompatibility order of both elements as well as the depletions recorded by the Sm-Nd, RbSr and Lu-Hf systems but the U/Pb increase is not. The anomalous behaviour of the U-Pb system is attributed to enrichment of U in MORB sources, relative to Th and Pb, rather than to loss of Pb. Unlike the 238u-206pb system, the 235u.207pb system does not show overall time-integrated U enrichment. This suggests U enrichment is a post-Archean phenomenon that produced little increase in 207p5/204p5 because of the small quantities of shortlived 235u that remained. Time-integrated Th/U ratios (calculated using 208pb*/206pb* data) for galenas from massive sulphide deposits decrease from 4.3 in the early Archean to -3.9 by -1.5 Ga, but then appears to have remained roughly constant for the last ~L5 Ga. If the isotopic evolution of such galenas is assumed to reflect that of the upper continental crust, then the time-integrated ThAJ systematics actually requires an increase in measured Th/U or, more likely, a loss of U from the upper crust during the last 1-2 Ga. The isotopic record suggests that the timing and inferred transfer directions of U for the depleted mantle and upper crust are complementary, and are consistent with the predictions of the model of a U return flux to the upper mantle via subduction of hydrothermally altered oceanic crust as proposed by Hofmann and White (1982).

GEOCHRQNOLOGY AND GEOCHEI^'ISTRY OF THE QUATEEU^ARY VOLCAGES THE Kl^^LUN MOU!^TAI!^S,NORTH TIBET PLATEAU IIU Jiaqi, Institute of Geology,Chinese Academy of Sciences, P.O.Box 654,Beijing loopll, China On the top of towering Kunlun Mountain in north Tibet Plateau there are more than ten volcanoes with evident truncate scape and tens of volcanelloes like hillocks distributed in Ashikule Basin. They,together with 200 km^ of Isva flow, formed a volcanic group located on 4650 m a . s . l . . T h i s i s among the highest volcanic group in the World. Other lava flow widely covered northern h i l l s i d e of Kunlun Mountain along the Keliya River. According to volcanic geological occurrence,geomorphic and dating by K-Ar method, most volcanoes occurred in the Quaternary from l,65I''5a to the present, the l a s t volcanic eruption i s in 1951. There volcanic rocks look basalt,but not true. They rich in s i l i c a and a l k a l i (K20+Ka20)^6, K20>Na20) which increase with the lapse of time. So most volcanic rock are called trachyandesites rich in potash named "kunlunite". These volcanic rocks also rich in rare earth elements (total REE >470 ppm,max 729 ppm), especially light rare earth elements (La/Yb)n 59-111.9 with strong fluctuation. Isotope composition of Rb,Sr,Sm,Nd were measured by VQ554 massspectrometer.The results are Rb 108.68-175.64 ppm, Sr 915.68-1250.23 ppm, Sr-87/Sr-86 0.708820-0.710540, Sm 15.599-21.559 Ppm, Nd IO4.86154.95 ppm, Nd-145/Nd-144 0.512221-0.512565, Nd - 5 . 0 -7.7. I t suggested that magma originated from upper mantle and to be contaminated throughout crust of the earth. Volcanism closely related with upwarping of Tibet Plateau,

ISOTOPIC/GEOCHEMICAL STRATIGRAPHY AND PALEOTECTONIC ANALYSIS: MESOZOIC FOREARC SANDSTONES OF CALIFORNIA LINN. A.M., DEPAOLO, D.J., Department of Geology and Geophysics, University of California, Berkeley, CA, 94720, U.S.A.. and INGERSOLL, R.V., Department of Earth and Space Sciences, University of California, Los Angeles, CA, 90024, U.S.A.

THE ISOTOPE GEOCHEMISTRY AND ORIGIN OF STRATIFORM COPPER(IRON-SULFUR-GOLD)DEPOSITS IN THE TONGLING AREA, ANHUI PROVINCE, CHINA Yuqing Liu Institute of Mineral Deposits, Chinese Academy of Geological Sciences, Baiwanzhuang Road 26, Beijing 100037, China

The large isotopic gradients in continental margin volcanic/ Plutonic arcs make it possible to obtain high resolution provenance information in arc-derived sediments. Sediments of the Great Valley forearc basin in California record the magmatic and tectonic evolution of the Sierra Nevada arc over nearly 100 Ma. Arc rocks vary in e^d from +6 to -7 and in 87Sr/86Sr ratios from 0.703 to 0.708 normal to the continental margin. Sandstones derived from the arc (148 to 73 Ma) have a correlative systematic geochemical stratigraphy in which eNd changes from +7 to -5 and 87Sr/86Sr ratios change from 0.705 to 0.707. The decrease in eNd with time correlates with an increase in K2O, Rb, Pb, Ba, U, Th and La, a decrease in MgO, and with the previously documented decrease in plagioclaseto-k-feldspar ratio. The eNd-esr relationship and correlation with major and trace elements indicates that volcanic and Plutonic arc rocks were the primary source of sediments. Pre-arc metasedimentary rocks contributed 5 to 25% of the detritus based on sandstone petrography. The isotopic composition of the igneous component pinpoints the position of the volcanic front, which from geomorphological considerations is typically located approximately 90% of the distance from the river mouth to the headwaters. The ENd and ®7Sr/86Sr ratios, which are a function only of transverse position in the arc, indicate that the volcanic front was located in a westerly position from 148 to 100 Ma and then migrated eastward during the Late Cretaceous at a rate of 2 km/Myr.

The Tongling area is one of the important copper-producing areas in China. In the area there are more than ten large and intermediate-sized copper, iron, sulfur and gold deposits, such as the Dongguashan and Tongguanshan deposits. Since the 1950's, they were considered to be typical skarn-type deposits. In recent years, by means of the geochemical studies of S,0,C,H,Sr,Pb isotopes of minerals and rocks and H,0,C istpoes in mineral inclusions, the author has found that these deposits have distinct stratabound characteristics. These source beds consist of the calcareous-magnesian ore-bearing carbonate formations of the Middle to Upper Carboniferous System and Upper Permian System and Lower Triassic System in some littoral—subneritc hollows. This horizon corresponds to the polymetallic ore bed in the open Huanglong sea area of South China. The formation of ore shoots is related to the Yanshanian tectono-magmatic hydrothermal activity and the activity of subsurface meteoric water in the heated circulation. As the source bed was subjected to those activities, facies changes, remobilization of ore substances and their reprecipization occurred, thus forming stratiform and irregular ore shoots. They show the characteristics of syngenetic sedimentary—superimposed and remoulded mineralization of the stratabound deposit. In the same ore field, different types of deposit are related to the same intrusion and assume a "five-storey" temporal-spatial distribution patlcrn, thus forming a distinctive minerogenetic models.

ABSTRACTS

ICOG 7

ATYPICAL STABLE ISOTOPE SIGNATURES FROM EPITHERMAL MINERALIZATION, NEVADOS DEL FAMATINA DISTRICT, LA RIOJA PROVINCE, ARGENTINA. LOSADA-CALDERON. A., Earth Sc Dept, Monash Univ, Clayton, Vic, 3168, Australia; Gregory, R.T., Dept of Geol Sc, Southern Methodist Univ, Dallas, Tex, 75275, USA, and; Bloom, M.S., Earth Sc Dept, Monash Univ, Clayton, Vic, 3168, Australia. Tertiary (» 5.5 Ma.) porphyry dacite intrudes Ordovician metapelites in the Nevados del Famatina (NF) mining district. Mo-Cu-Au porphyry style mineralization is known to occur at NF (4000 to 4800 m elevation) in three open annular patterns encompassing the intrusive rocks. Peripheral to the intrusive centre is the La Mejicana (LM) high sulfidation epithermal famatinite-gold deposit. Whole-rock S^^O and 8D from NF porphyry dacite with Ksilicate alteration are 8.9%o and -75.3%o, respectively. Quartz in veinlets from NF with quartz-sericite alteration suggest fluids with calculated values between 5.4 to 6.3%c and 5D from -65.7 to -73.5%o. from fluids in equilibrium with sericite range from 8.87 to 9.1%o. LM 51^0 calculated from quartz veins with alunitekaolinite alteration vary from 1.7 to 2.4%o, while 5D ranges from -26.4 to '49.9%o. The fluid inclusion population most closely related to goldtelluride-sulphosalt mineralization and associated with kaolinitealunite alteration has salinities between 20 to 32 eq.wt.% NaCl at temperatures between 270°C and 340°C, which are atypical of epithermal deposits. These occur in quartz veins having crustiform and open filling textures characteristic of epithermal mineralization. The and 5D, strongly enriched relative to meteoric waters having S^^O <-15%o and 5D <-110%o anticipated at the elevation where mineralization occurs, are unexpected for an epithermal deposit. These isotopic signatures may be due to meteoric water strongly shifted by processes in the epithermal environment and mixing with magmatic-hydrothermal fluids.

LITHOLOGY AND AGE OF THE BASEMENT AT CAPE DUBOUZET, ANTARCTIC PENINSULA W.P. LOSKE, Institute for General and Applied Geology, 8000 Munchen, Luisenstr. 37, FRG, and H. MILLER, A. MILNE, F. HERVE Low-grade metasedimentary rocks dominate the preMiddle Jurassic geology of the northern Antarctic Peninsula. The discovery of metamorphic xenoliths in a Mesozoic tonalitic intrusive allowed a study to be made of the previously unknown basement lithologies. U-Pb analysis of euhedral zircons (223 +42/-64 Ma) and Sm-Nd analysis of garnet whole rock pairs (183 +/-64 Ma) yielded a unexplicable imprecise early Mesozoic intrusion age. Zircons from a metatonalite xenolith are mostly colourless-yellowish and contain abundant translucent and opaque inclusions. Fractions <100 m are euhedral prismatic with one or two pyramids, whereas larger fractions have a rounded shape. Simple correlation of grain-size, uranium-content and degree of discordance indicate a oogenetic evolution. Seven grain-size fractions (<42->12S m ) define a reference line (MSWD 1.5) which intersects the concordia at 291 +14/-26 Ma and 1698 +88/-82 Ma. The typology of zircons from a hornblende-gneiss xenolith is similar, except for their shape, which is subhedral even for the smaller fractions. Their disturbed pattern of grain-size, uranium content and degree of discordance may be due to a complex newgrowth/inherited nuclei/Pb-loss ratio. Eight grainsize fractions define a regression line (MSWD .94) intersecting the concordia at 307 +12/-14 Ma and 1635 +308/-288 Ma. Both upper intercept dates indicate a formation of the zircons during the Lower Proterozoic and a Pbloss/new-growth event during the Carboniferous. This supports further evidence for a Hercynian metamorphic event affecting the continental basement of the northern Antarctic Peninsula.

61 HIGH-PRECISION MASS-SPECTROMETRIC 230xh.23AU.238U DATING ACROSS A CALCITE VEIN WITH CONTINUOUS GROWTH FROM 570 KA TO 60 KA, DEVIL'S HOLE, NEVADA. U.S.A. Ludwig, K.R., Simmons, K.R., Szabo, B.J., and Riggs, A.C., U.S. Geol. Surv., Denver CO 80225 USA, & Hoffman, R . J . , U.S. Geol. Surv., Carson City, 89701, USA Devil's Hole, a fault-controlled cavern in Paleozoic limestone, i s localized on a 1 to 5-m-wide open f r a c t u r e that extends for >115 m below the land surface. Below the water table, c a l c i t e coats the walls of the f r a c t u r e to a thickness of -36 cm, and contains an average of 450 ppb U. Replicate mass-spectrometrie (MS) ^^Th-^V-^^^ analyses of 20 300-mg samples across the vein shows that c a l c i t e was continuously deposited from 570 ka u n t i l 60 ka, when deposition abruptly ceased. I n i t i a l 234u/238u was roughly constant a t -2.7 (activity r a t i o ) , with a resolvable v a r i a t i o n of -±.08. MS runs for 234u/238u were automated, with run-to-run precisions of 2-3 permil (2-sigma) yielding analytical 234u/238u age-errors of ±13 ky over the whole age-range (though t o t a l 23Au/238u age-errors are limited to +17 by the variation in i n i t i a l 234u/238u) , ^S runs for used graphite-doped Re filaments that were each checked for mass-spectral purity by automated MS scans, and yielded typical 2sigma uncertainties of 2-4 permil, resulting in 230Th/234u age-uncertainties of 0.4-1.0 ky for 60-160 ka samples, 10-15 ky for 400-440 ka samples, and 30-50 ky for 460-570 ka samples. Replicate analyses agreed within assigned errors plus the 1.7 ky uncertainty a r i s i n g from the - 1 mm sample-location error. Deviations of more than -3 permil in ^"^Th/^U from closed-system r a t i o s by the oldest (>400 ka) samples, or in the atomic 230Th/234u/238u assumed for secular-equilibrium, would have yielded s i g n i f i c a n t 230Th/234u-234u/238u age-discordances. Therefore the analytical concordance of the oldest Devil's Hole ^^^Th/^H - 234u/238u ^ges confirms the assigned errors, the assumption of closed-system behavior, and the v a l i d i t y of the "Harwell" uraninite standard used to define atomic 230Th/234u/238u r a t i o s .

ISOTOPIC ANOMALIES IN THE IRON GROUP ELEMENTS LUGMAIR G. W., University of California San Diego, Scripps Institution of Oceanography, La Jolla, CA 92093, USA The discovery in Ca-Al-rich Inclusions (CAD from the Allende meteorite of isotopic anomalies in oxygen, and later in Ca, Sr, Ba, Sm, and Nd in two of the still very rare FUN inclusions, was followed by a decade of intense and systematic investigation of isotopic variat i o n s in t h e I r o n - P e a k (IP) elements. U b i q u i t o u s anomalies were found not only in the highly anomalous FUN inclusions but in the normal non-FUN CAIs for almost all elements with more than two stable isotopes from Ca through Zn. The anomalies in normal CAIs, up to permil in size relative to terrestrial isotopic composition, are characterized by correlated overabundances of the neutron rich isotopes of the IP elements which include ^^Ca, ^oxi, ^ c r , and ^Zn. These excesses are mostly accompanied by small deficits in the lighter isotopes such as ^^ca and ^"'Ti. With rare except i o n s t h e i s o t o p i c a n o m a l i e s a p p e a r to be homogeneously distributed within an inclusion. The exceptions include large isotopic variations found either in ^®Ca and in single r e f r a c t o r y mineral grains, measured by ion microprobe techniques, or in ^^Cr by l e a c h i n g p r o c e d u r e s on bulk samples from c a r bonaceous meteorites. Various nucleosynthetic p r o d u c t i o n models have been developed to account for the excesses found in the neutron rich isotopes of IP elements. It appears t h a t some form of the n e u t r o n - r i c h e-process with mixing of products from stellar zones with different neutron excesses can b e s t explain the experimental observations. In p a r t i c u l a r , the recently discovered small excesses in non-FUN inclusions conform in a surprisingly quantitative manner with these model predictions while the much lower than exjjected effects may result from dilution with isotopically normal Zn prior to or during CAI formation.

ICOG 7

CALCIUM AND SULFUR ISOTOPES AND RARE EARTH ELEMENT CONCENTRATIONS IN THE OLDHAMTTE OF UNEQUILIBRATED ENSTATITE CHONDRITES LUNDBERG. L., CROZAZ, G., and ZINNER, E., McDonnell Center for the Space Sciences, Washington University, St. Louis, MO 63130 USA E chondrites, the most reduced meteorites, contain an accessory mineral, oldhamite (CaS), which may be the refractory equivalent of hibonite and perovskite in the more oxidized carbonaceous chondrites and, therefore, a carrier of isotopic anomalies. This work is a correlated study of Ca isotopic compositions, REE abundances and mineral associations of CaS in the least equilibrated E chondrites, Qingzhen (EH3), Indarch (EH4), Yamato 691 (EH3) and ALHA77156 (EH3-4). S isotopes were also measured in CaS, niningerite (MgS) and troilite (FeS). In the Antarctic meteorites, only weathered CaS is observed whose Ca isotopic composition could not be analyzed. Using CaS in Norton County (E achondrite) as a standard, a power fractionation law was found to be more appropriate than the exponential law used earlier[l]. Previous data are reevaluated and additional measurements reported. In Indarch, no Ca anomalies are present in 10 CaS grains. However, 4 out of 15 grains in Qingzhen show depletions in 48ca (-7.1 ± 2.5, -4.6 ± 2.3, -5.0 ± 2.4, -3.4 ± 2.0%c, 2a errors). Small negative anomalies (-0.26 ± 0.07%o) were reported in whole rock samples of 7 enstatites [2]. However, the S isotopic composition of 4 CaS, 2 MgS and 5 FeS grains in Qingzhen are normal within errors of 2-4%©. Four distinct REE patterns, designated A, B, C, and D[l], are observed in the CaS of Qingzhen while only types A and B are represented in the CaS of Indarch. In the weathered areas of ALHA77156, REE patterns A, B, and C are observed and in Yamato 691 only types B and D are found. REE abundances are available for two of the grains with anomalous Ca. Their patterns are type B and C. The mineral occurrences of the 4 anomalous grains are also different. Thus, there is no obvious trend between S'^oCa anomalies and either REE abundances or CaS occurrences. The Ca isotopic anomalies show that oldhamites have a primitive origin. [1] Lundberg et al. (1989) Meteoritics 24,296-297. [2] Gao X. and Thiemens M. (1990) LPSC XXI, 401-402.

HELIUM ISOTOPES IN SUBMARINE THOLEIITIC BASALTS FROM THE ISLAND OF HAWAII LUPTON. John. GRAHAM, David, (Dept. of Geological Sciences and Marine Science Institute, University of California, Santa Barbara, CA 93106) and GARCIA, Michael (Hawaii Institute of Geophysics, University of Hawaii, Honolulu, HI 96822) We report analyses of ^He/'^He, and He and Ne contents for a suite of basalts dredged from the submarine rift zones of Kilauea, Mauna Loa, Mauna Kea, and Kohala. Gases were released by melting basalt glass or by in vacuo crushing of either glass or pristine olivine phenocrysts separated from the submarine tholeiites. These samples have also been analyzed for major elements, volatiles, and stable isotopes (Garcia et al., JGR 94,525,1989). Magmatic signatures are clearly present in all the samples, but not in all of the phases. Helium isotope ratios in Kilauea and Mauna Loa glasses have ranges of R / R a = -17.2 and 14.3 -15.8, respectively, with typical He contents of lO'^ cc/g (R = 3He/4He and R a = Rair = 1-39 x 10-6). Mauna Kea submarine glasses had much lower ratios of 1.5 - 9.1 Ra and much lower He contents of "5 x 10-^ cc/g. However, ^He/'^He ratios in olivine phenocrysts analyzed thus far from Mauna Kea are between 11.6 - 14.3 Ra, and the He contents of the olivines are higher than that found in any of the Mauna Kea glasses. This indicates that while the glasses have apparently suffered outgassing or alteration, the magmatic signature is preserved in the Mauna Kea olivines. The Kohala olivines (no suitable glass was recovered from the Kohala dredges) ranged from of 7.9 - 13.2 Ra, which is consistent with temporal variability in ^He/'^He reported at a single Hawaiian volcano such as Mauna Loa. Our results for Mauna Kea and Kohala olivines are the first observations of ^He/^He ratios higher than the typical MORE ratio for these volcanoes, supporting the idea that a relatively undegassed primitive mantle component contributed to the tholeiitic phase of all volcanoes on the island of Hawaii.

ABSTRACTS

GEOCHEMICAL CHARACTERISTICS OF 4100-4200 M.Y. CRUST: THE REE RECORD IN DETRITAL ZIRCONS FROM THE NARRYER GNEISS COMPLEX, WESTERN AUSTRALIA MAAS, Roland, Div. Engineering & Sci., Curtin University of Technology, Perth, WA 6001, Australia, KINNY, Peter D. and FROUDE, Derek O., WILLIAMS, Ian S., COMPSTON, William, Research School of Earth Sciences, Australian National University, Canberra, ACT 2601, Australia. Detrital zircons with U-Pb ages > 4 Ga in ca. 3 Ga old quartz-rich metasediments of the Narryer Gneiss Complex (NGC) represent the oldest terrestrial material found to date. They probably originate in differentiated crust but whether this crust formed part of an extensive sialic continental block or whether it was associated with mafic/ultramafic oceanic crust (an Iceland-type setting) is unclear. Here we present ionprobe REE data for detrital zircons from Mount Narryer and Jack Hills in an attempt to characterize their parent rocks and place some constraints on the nature of this ancient crust The measured pattems for 3.9 -4.25 Ga zircons show the strong HREE enrichment which is characteristic of zircon, as well as Eu depletion and low LREE levels with pronounced positive Ce anomalies. The overall pattems are similar in both the Mt. Narryer and Jack Hills zircons, and in younger (<3.6 Ga) zircons of comparable morphology and U-Th levels from the same detrital populations. The significance of these patterns is uncertain but circumstantial evidence suggests they are of primary igneous origin. If this is correct, recent data on REE zircon/silicate melt partitioning obtained on the ionprobe would indicate parent melts characterized by high Lan/Smn, Eu depletion and unfractionated HI^E; only one grain appears to have formed from a HREE-fractionated melt. Collectively, these model melt features, coupled with the presence of qz-Kf-rich polymineralic and monazite inclusions in some of the old NGC zircons and U-Pb geochronological evidence are most consistent with an intrusive, multi - component source of overall evolved, granitic composition. The inferred Eu depletion and flat HREE of the parent melt indicate an intracrustal origm by remelting of earlier, more primitive felsic crustal material. This and volumetric considerations tend to rule out an Iceland-type setting for the 4.1-4.3 Ga crust but favour the existence of an extensive, repeatedly reworked sialic block.

THERMAL HISTORIES IN ANCIENT HIGH GRADE METAMORPmC TERRANES: THE MUSGRAVE RANGES, CENTRAL AUSTRALIA. MABOKO. M.A.H., McDOUGALL, I. and ZEITLER, P.K. R.S.E.S, Australian National Univ, Canberra, Australia. The reconstruction of thermal histories in ancient high grade metamorphic terranes is often complicated by polymetamorphism, multiple deformation and deep erosion, effects which are rarely encountered in younger orogens. By combining various thermochronological systems with petrological thermobarometry, however, it is possible to see through some of these complexities and reconstruct relatively precise segments of the post-metamoiphic t h e c a l history, which can then be used to constrain tectonic evolution. As an example of such an approach, we report a retrograde PTt path for a Proterozoic granulite terrane from the Musâve Ranges, Central Australia. An anatectic chamockitic granitoid emplaced in granulites at -12 kb and 860 ^C yields a zircon U-Pb age of 1186 ± 10 Ma. An adamellitic granitoid emplaced at - 6 kb and 670 ^C yields a similar age (1193 ± 8 Ma), suggesting rapid post-metamorphic decompression. The rapid decompression was followed by slow cooling as evidenced by the fact that hornblende and biotite from the granulites did not close to Ar diffusion until after -930 Ma and -690 Ma repectively. The slow cooling continued until early Palaeozoic time when the granulites were exhumed along the Woodroffe Thrust and emplaced on top of an older amphibolite fades gneissic terrane. Flat Âr-^Âr age spectra obtained from syn-tectonic muscovite growing in shear zones genetically related to the thrust date the exhumation at 530 550 Ma. Lower intercept ages of -400 Ma from steadily rising K-feldspar spectra date fmal cooling of the granulites to temperatures of -200 ®C. The proctrated thermal history suggests a tectonic history involving rapid (> -Imm/year), possibly tectonically-driven uplift of the lower part of a tiiickened crustal segment (-40 km depth) to a depth of 15-20 km. The uplift probably re-established isostatic balance such that prolonged slow cooling with very little uplift ensued until -530- 550 Ma when renewed tectonic activity exhumed the granulites along the Woodroffe Thrust

ABSTRACTS

IC0G7

COUNTING CRISTALLOGRAPHIC SECTION FOR FISSION TRACK DATING OF DURANGO APATITE M.MARS. M.REBETEZ, A.CHAMBAUDET and M.GRIVET Laboratoire de Microanalyses Nucl6aires U.F.R. Sciences et Techniques La Bouloie -16 route de Gray 25030 Besan9on Cedex - France This paper considers the cristallographic section influence on fission track analysis in Durango apatite. From measurements carried out on two basal and prismatic sections of this apatite, we have studied the behaviour of different parameters used in the fission track method. A large number of projected lengtiis were first measured in order to obtain the corresponding statistical distributions. These histograms confirmed the general triangular trend of projected length distributions and gave identical mean projected lengths for the two cristallographic sections. The track densities were then calculated for the two sections. The results also showed no significant differences in the mean value for a set of measurements although there were sometimes notable divergences between two successive countings. Next, the liUAn geometry factor for the two slices with two different extemal detectors was studied. In all cases, the results showed an indiscernible value of 0.5 taking into account measurements errors. Finally, fission track ages calculated with the population method and the extemal detector method for each cristallographic section are given. The mean ages are 31.3 ± 1.2 m.y. for the prismatic section and 32.3 ± 0.4 m.y. for tiie basal section. The results seem to indicate that where there is no partial annealing, such as in Durango apatite, the observed section has no influence on fission track dating.

XENOLOGY OF CHONDRITIC METAL MARTI, K.1), KIM, J. D, FELLAS, P.2), PERRON, C.2), DChem. Dept., B-017, Univ. of Calif., San Diego, La Jolla, CA. 92093, 2)Museum National d'Histoire, Naturelle, Mineralogie 61 Rue de Buffon , 75005 Paris, FRANCE. We identified a new xenon component (FVM-Xe) in metal of the H4 chondrite Forest Vale [ 1 ]. Subsequent work revealed that FVM-Xe is observed in H4 chondrites but not in an H6. The new component was identified by its isotopic signature which displays a 3.4% shift in the 134Xe/136Xe ratio, when compared to chondritic trapped Xe(FVC), a component which is also present in the metal. In order to further constrain the distribution and origin of this new component, we are investigating grain-size fractions of metal separates of H3 chondrite Dhajala. FVM-Xe is enriched in the smaller metal grains and, in stepwise pyrolysis following lowtemperature combustion, shows a bimodal release peaking around 600 and 1200C. The release properties of radiogenic 129Xer, presumably from inclusions in the metal, are quite distinct. Our tentative identification of FVM-Xe, as a surface-sited component, appears to be incomplete. At present, it has only been identified in the metal. We are exploring three possible origins: (1) a fission origin due to neutron-induced fission reactions on 235U, prior to meteorite accretion (possibly in C-rich aggregates [2]), (2) implantation of recoils due to fission of transuranic nuclides and (3) a signature indicating a presolar relict. Since the cosmic-ray exposure age of Dhajala is only 6Ma, the spallation component is not expected to seriously impede an evaluation of the isotopic systematics of fission-shielded isotopes (124Xe to 130Xe). It appears that FVM-Xe, taken as unmixed reservoir, is isotopically inconsistent with trapped Xe components observed in other solar system reservoirs. References: [1] Marti, K . ^ 3I. (1989), Z. NaUirforsch. 963-967. [2] Breariey, A. (1990), Geochim. Cosmochim. A c t a ^ 831-850.

OSMIUM I S O T O P I C CONTRAINTS ON T H E FORMATION OF MANTLE RESERVOIRS MARTIN. Candace E.1.2, and TUREKIAN, Karl K.l, ^Department of Geology and Geophysics, Yale University, P.O. Box 6666, New Haven, CT 06511, USA; ^Department of Terrestrial Magnetism, Carnegie Institution of Washington, 5241 Broad Branch Road NW, Washington, D.C. 20015, USA The measured osmium isotopic compositions (reported as ^^'Ôs/l^Ôs) of uncontaminated oceanic peridotites and basalts range between 1.0032 ± 0.0072 and 1.111 ± 0.011 (2G). This range reflects chemical fractionation of the incompatible parent element Re from compatible Os in the mantle during crust production and subsequent isolation of the Re-depleted reôns of the mantie for time periods in excess of 1000 Ma. We consider that ocean island basalts which have high ^He/^He ratios give the best indication of undepleted mantle Os isotopic composition; this corresponds to a present-day l87os/i86os of 1.10 and a 187Re/l86os of 4.12 for the bulk silicate earth. Oceanic peridotites which are depleted in terms of mineralogy and chemical composition have 1870s/^86os and 187Re/186os which are low relative to the bulk earth but above the previous estimate of bulk earth values (A116gre and Luck, Namrc 302:130-132,1983). A material balance for mantle reservoirs based on Re-Os and assuming that average depleted mantie has present-day of 1.05 yields results which are different from those obtained from incompatible element calculations. The amount of depleted mantie must be small (<5%) in order to satisfy the crust-depleted mantieundepleted mantle system, as opposed to >15% from other isotopes. There are several possible reasons for the discrepancy: (1) Os, a compatible element, may be more resistant to isotopic redistribution and re-equilibration within the mantie mediated by fluids or melts than are tiie incompatible elements; (2) incompatible elements are removed from the mantie by processes which do not affect Re-Os; and/or (3) the mass balance considers only the continental crust, whereas the formation of oceanic crust over geologic time and particularly early in earth history will have a major effect on the amount of Re-depleted mantie. Ancient subducted oceanic crust present within the mantie could be a major reservoir of radiogenic Os which is not stored in the upper continental crust.

U P P E R C R E T A C E O U S S T R A T I G R A P H Y W I T H Sr I S O T O P E S J.M. M^-ARTHUR. Dept. of Geol. Sciences, University College London, London WCIE 6BT, UK, M. F. THIRLWALL, Dept. of Geology, Royal Holloway and Bedford New College, Surrey, TW20 OEX, UK, W.J. KENNEDY, Dept. of Earth Science, Oxford, 0X1 3PR, UK, and A. S. GALE, School of Earth Sciences, Thames Polytechnic, London, El 2NG, UK. In order to provide a quantitative template with which to correlate world-wide, and against which to integrate diverse stratigraphic schemes for the Upper Cretaceous, we present a high-precision curve of ®^Sr/®^Sr for marine Sr during the Cenomanian-lowermost Maastrictian. It is closely integrated with macrofossil and microfossil stratigraphy in order to emphasise the importance of full biostratigraphic control, and its documentation, in compiling such curves and we demonstrate its use for correlation. Data derive mostly from a stratigraphic borehole through the English Chalk at Trunch in Norfolk. Stratigraphic resolution using ®^Sr/®®Sr varies with the rate at which ®^Sr/®^Sr changes and is typically < ± 1 Ma for the Santonian and Campanian stages. The ®^Sr/®^Sr decreases upwards through the condensed Cenomanian section and the lower Turonian, is stable at 0.70738 from the middle Turonian through to the upper Coniacian. It then increases sharply to 0.70745 at the Coniacian-Santonian boundary, thereafter increasing at a lesser rate, in a linear manner, to a value of 0.70772 at the Campanian-Maastrichtian boundary. Fortuitously, where the biostratigraphy is poorest the ®^Sr/®^Sr stratigraphy is best ie. the rate of change in "^'^'Sr/'^'^Sr is greatest. Isotopic correlation over a distance of 350 km within the Chalk of the UK is, at some stratigraphic levels, as good as the best available biostratigraphy^ When testing isotope correlation outside the UK using sections with well-documented biostratigraphy, similar excellent agreement is found, examples of which will be given.

ICOG 7

ABSTRACTS

Sr ISOTOPIC COMPOSITION OF EARLY ARCHEAN BARTTE AND LIMITS ON THE EARTH'S INITIAL ^^Sv/^ST RATIO McCULLOCH M.T. Research School of Earth Sciences, Australian National University, Canberra, 2601, Australia. The initial Sr isotopic composition of the Earth is a poorly defined but important parameter and in principle can be used to provide constraints on the origin and evolution of the Earth-Moon system. Planetismals have relatively high Rb/Sr ratios which are up to a factor of xlO greater than differentiated planetary bodies, hence 87Sr/86Sr ratios evolve rapidly part in 10^ per million years) immediately prior to and during planetary formation. Uncertainties in the Earth's primitive Rb/Sr ratio, a volatile/involatile ratio, enable only a lower limit to the Bulk Earth Best Initial (BEBI) to be inferred from initial 87Sr/86Sr ratios of achondrites (BABI, ADOR) or refractory condensates (ALL) (Wasserburg et al., 1977). In an attempt to directiy constrain BEBI, 87Sr/^Sr ratios have been determined in barite (BslSO^)fromthe early Archean (-3450 Ma) Pilbara Block of Western Australia. The barite has a complex origin, being formed by fluids derived from seawater as well as surrounding basalts, but its low 87Rb/86sr ratio (<0.0001) allows preservation of primitive 87sr/86sr ratios. From the directly measured ^'^Sifi^Sr ratio of 0.700513±9, similar to that previously reported for mineral separates from Onverwacht Group basalts, a relatively high ratio of 0.70050±2 is implied for the early Archean mantle. The higher than anticipated Sr comj^sition of the early Archean mantie has several important ramifications. Firstly it suggests that the 87Rb/86Sr ratio characteristic of the present-day depleted mantie (-0.042) was established in the early Archean. Secondly it gives an estimated upper limit of BEBI at ~4500 Ma of 0.69940±2. This ratio is significantiy higher than that determined for the Moon (0.69903±2) and with a 87Rb/86Sr ratio of -0.4 (-1/2 of average chondrites) for the terrestrial precursors implies that the Earth was formed within an interval of -80 Ma after ADOR and the Moon within -20 Ma. This chronology is compatible with an early (>4520 Ma) giant impact origin for the Moon and requires that the final accretion of the Earth was completed relatively late (>4460 Ma). These timescales are consistent with those required for cooling of the lunar magma ocean and implies a relatively cool,finalstage for the accretion of the Earth.

SOURCES OF INTRAPLATE BASALTS: DEEP MANTLE, ASTHENOSPHERE, LITHOSPHERE McDONOUGH. W. F.. RSES, Australian National University, Canberra, 2601, Australia & SUN, S.-s., Bureau Mineral Resources, Canberra, 2601, Australia. Many intraplate basalts (OIB & Continental) result from upwelling mantle plumes. Although distinct from MORB, the correlated chemical (e.g. Ce/Pb, Nb/U) and isotopic systematics of intraplate basalts & MORB demonstrate a common evolution from an initially depleted mantie reservoir. Many of their chemical and isotopic characteristics can be related primarily to the recycling of oceanic lithosphere. Distinctive types of intraplate basalts (HIMU, EM, DUPAL) are, however, the result of inputsfromseconda^ sources (e.g., crust, asthenosphere, lithospheric mantie). Identifying and understanding the origin of these types can provide important insights into mantie dynamics. A myriad of sources are invoked to explain the composition of intraplate basalts. Fundamental to any model, however, is that the compositions of intraplate basalts are the integrated product of the plume source history (deep mantle) and shallow components derived from plume-mantie interactions during ascent (i.e., asthenospheric and lithospheric sources). In unraveling this complex picture several features stand out: (1) Alkalic intraplate basalts have a constant La/Ce (-0.48) which is not controlled by melting processes and indicates a uniform source enrichment process. (2) Whereas carbonatites and Group I kimberlites can have HIMU characteristics, large volume basalts (ocean plateaus, cont. flood basalts) do not appear to. This feature, along with other chemical and isotopic characteristics of these basalts, suggests that their chemistry is controlled by shallow litiiospheric (crust or mantie) components. (3) The influence of a sediment-added component versus a continental lithospheric mantie (CLM) component on both oceanic and continental intraplate basalts is not uniquely constrained by Sr-NdPb isotope systematics. However, recent data onfloodbasalts and Group I & II kimberlites showtiiatOs-Hf isotope systematics may provide such critical data. Specifically, a CLM. component is characterized by a long-term depleted Os-Hf isotope signature. (4) Combined chemical and isotopic data do not support hypothesized primitive mantle sources for intraplate basalts.

MASS-SPECTROMETRIC MEASUREMENTS OF 230Th/232Th IN 45 Ka VOLCANIC ROCKS F MCDERMOTT TE ELLIOTT, DW WRIGHT P VAN CALSTEREN & O HAWKESWORTH Department of Earth Sciences, Open University, Milton Keynes

HORNBLENDE AS A GEOCHRONOMETER McDOUGALL. I., MABOKO, M.A.H., BALDWIN, S.L., RICHARDS, J.P., FTTZ GERALD, J.D., Research School of Earth Sciences, The Australian National University, Canberra, ACT 2601, Australia Hornblende, as a relatively common mineral, with a moderately high closure temperature for argon (-500"C), is widely utilized for K/Ar and ^^Aift^Ar dating to obtain age estimates for emplacement and cooling of igneous rocks and for thermochronological studies in metamorphic terranes. Hornblende 40Ar/39Ar age spectra range from essentially flat to forms interpreted as reflecting argon loss or gain, and to forms that are difficult to interpret. Although hornblendes from igneous rocks commonly yield simple,flat,age spectra, providing good estimates in many cases for thetimesince crystallization and cooling, others show evidence for incorporation of excess argon through saddle-shaped spectra (e.g. Porgera, PNG) especially in volatile-rich environments. Age spectra measured on hornblendes from metamorphic rocks show great diversity, including a proportion tiiat are complex and irregular, often not readily interpretable. Many simple age spectra obtained from thermally disturbed or recrystallized hornblendes can be interpreted in a straightforward manner, and yield reliable and useful chronological information. However, examples from Central Australia from amphibolite to granulite facies rocks, indicate that some relatively simple hornblende age spectra, including virtually flat spectra, cannot always be interpreted at face value. Data from adjacent samples, with a common P,T history, indicate that excess argon is present in the hornblendes, but not obviously manifest in the shape of the age spectra. Excess argon apparently is incorporated as the mineral cools through its closure temperature, owing to a significant and locally quite variable partial pressure of argon in the environment. To evaluate the significance of age spectra on amphiboles, especially in metamorphic rocks, it is essential to measure a number of samples, preferablyfrx)mlocalities in close proximity to one another, to test for consistency; measurements on coexisting minerals also are desirable. Further laboratory experiments are urgently required on well characterized hornblendes, inducing both argon loss and argon gain, to assist in understanding the processes involved, and to guide the interpretations of age spectra measured on natural samples.

Thorium isotope ratios of230Th/232Th=.00000580 have been measured with a conventional MAT261 multicollector massspectrometer with 230xh detected on the Secondary Electron Multiplier equipped witii pulse counting. For this technique it is essential to determine the magnitude and shape of the background on the 230xh caused by the Ippm abundance sensitivity of the instrument and the large ^^Th. Compared with alpha counting our precision of 2.3% (2a) is three tofivetimesbetter, we require 200 ng of Th compared to l^ig and an analysis takes hours rather than weeks.This is achieved mainly because of the more favourable counting statistics even on the minor isotope, compared with the low decay constant of 232xh.We have evaluated our methods by analysing various international rock standards and when plotted on tiie conventional (230Th/232Th) (238u/232Th) they faU on the equiline.We have applied our methods to lavasfromthe Pantelleria volcano and have showntiiansome were formed at around 45 Ka by partial melting and mixing, not longer than 5 Ka earlier, of rocks with a large range in Th/U, establish^ for at least 1 Ga.

ABSTRACTS

ICOG

USE OF 'tOAr/SÂr AGE SPECTRA AND PALE0MA6NETISM TO UNRAVEL THE STRATIGRAPHY OF A LARGE SILICIC VOLCANIC FIELD, SOUTHWESTERN NEW MEXICO McINTOSH, W.C., CHAPIN, C.E., NMBMMR, Socorro. NM 87801, SUTTER, J.F., and KUNK, M.J., USGS, Reston, VA Z Z O W I M t+0Ar/39Ar age spectra and paleomagnetic data have greatly aided in correlating >25 voluminous (1001200 km3), 36 to 24 Ma regional ignimbrites in the 40,000 km2 Mogollon-Datil volcanic field of southwestern New Mexico. Precision of the ^Âr/sÂr age spectra is similar to that reported from single-crystal laser fusion i+0Ar/39Ar dating studies. Age spectra from sanidine separates (97 spectra, 85 samples) yield plateau ages showing la precisions of +0.25-0.4% (+0.07-0.14 Ma). Plateau ages from multiple samples (n=3-8) of individual ignimbrites show la within-unit precision of ±0.1-0.4% (±0.04-0.13 Ma). Unit-mean ages agree with independently established stratigraphic order. Ignimbrite outflow sheets show laterally and vertically uniform paleomagnetic remanence directions. About 40% of the units show distinctive unit-mean directions that differ by >15% from timeaveraged field directions. ^0Ar/39Ar and paleomagnetic correlation criteria are complementary: ^Âr/^Âr data allow the discrimination of paleomagnetically similar units, and paleomagnetism helps distinguish between units that are altered, lack sanidine, or are too close in age to be resolved by the '+OAr/^Âr method. Used in concert with lithologic and stratigraphic data, '+0Ar/39Ar spectrum ages and paleomagnetic data have allowed the development of a precise stratigraphic framework for the Mogollon-Datil volcanic field. The precisely dated geomagnetic polarity record also provides calibration points for the Eocene-Oligocene geomagnetic polarity time scale.

THE MICROSTRUCTURE OF ZIRCON AND ITS INFLUENCE ON THE Pb/U AGE MEASURED BY THE ION MICROPROBE A . C . M C L A R E N . J.D. H T Z G E R A L D , a n d I.S. W I L L I A M S ,

Research School of Earth Sciences, Australian National University, Canberra, A.C.T., Australia. TEM observations of zircon suggest that certain microstructural features may be associated with discordant PbAJ ages measured with the ion microprobe. Some of the growth bands in the rim of a large Sri Lankan crystal are optically isotropic and one such band (ca. 100 ^im wide; designated I) was compared with adjacent anistropic regions (designated A). The I-band has a significantly higher U and Pb content, and its electron diffraction pattern consists of very diffuse rings suggestive of baddele>dte. TEM images reveal a granularity on the unit cell scale. A-regions consist of oriented zircon crystallites (ca. 10 nm in size) in a matrix of material with the same characteristics as the I-band. Annealing in air at T < 900'C has little effect, but at that T the diffraction rings become relatively sharp and can be indexed unequivocally as baddeleyite which is present as randomly oriented crystallites (ca. 10 nm in size), presumably in association with silica, in both the I-band and A-regions. However, there is no significant growth of the zircon crystallites in the Aregions. On annealing at 1250'C the baddeleyite ^stallites in the Iband grow to ca. 100 nm and the associated silica glass matrix is now easily identified. However, the A-regions recrystallize to a zircon single crystal with precipitates of baddeleyite (ca. 40 nm in size) and, presumably, some silica glass. All the measured PbAJ ages are reverse discordant (corresponding to an apparent Pb gain), apart from two which are concordant (age ca. 570 Ma). The discordance (measured age/concordant age) is consistentiy greater for the I-band, suggesting the discordance may be some matrix effect. However, the large discordances measured in the specimen annealed at 1250'C (340% in the I-band; 118% in the A-region) suggest an additional effect: during annealing Pb may concentrate in the silica glass phase which sputters preferentially leading to an apparent Pb gain. These observations may be relevant to the dating of zircons in rocks which have experienced high temperatures.

SR ISOTOPES AND GROUNDWATERS R.H. MCNUTT, Dept. of Geology, McMaster University, Hamilton, Ontario, Canada, L8S 4M1 In studies of groundwater systems of the Precambrian Shield of Canada, it is observed that in a given hydrologic system, of the 87Sr/86Sr ratio the deeper, generally more saline groundwaters in igneous or meta-igneous rocks is the result of dissolution of plagioclase feldspar and its alteration products. Shallow, more dilute waters from the same area tend to have more radiogenic values. Equilibrium dissolution and kinetic calculations based on the the model of Made and Fritz (1989), but modified to include Sr isotopic data, show that in the early stages of reaction between a very dilute water and a granitic host rock, the ratio increases rapidly with time due to dissolution of Rb-rich (and therefore 87Sr-rich) potassium feldspar. With time, as reaction continues, the ratio in the water approaches that of the plagioclase (Fritz and McNutt, in prep). The calculations suggest that the time required for the groundwater to take on the plagioclase signature is very short in the geological sense. A study of aquifers in neighbouring limstone and sandstone units demonstrate both the utility and limitations of the 87Sr/86Sr ratio as a tracer of aquifers (Plummer and McNutt, in prep). The two water systems have very different isotopic ratios so are sensitive to any mixing between them. But the carbonate-hosted groundwater has much higher Sr contents, so the use of the 87Sr/86Sr ratio is most useful for the study of chemical modifications to the sandstone aquifer. Strontium isotopic studies of groundwater systems are best done in conjunction with stable isotopes (eg.,0 and H) and with other radiogenic isotopic systems (e.g., U-series). Some examples will be given.

NATURAL ISOTOPIC TRACER TECHNIQUES FOR MAPPING HYDRODYNAMIC SYSTEMS IN HYDROCARBON RESERVOIRS Euan M E A R N S . A m e R A H E I M , Harald J O H A N S E N a n d

Helge STRAY, Institutt For Energiteknikk, P.O. Box 40, N2007 Kjeller, Norway. Understanding the degree and pattern of pore-volume communication, both vertically and laterally between wells, is of major benefit in optimising recovery from hydrocarbon reservoirs. Of particular importance is the recognition of permeability barriers at an early stage of field development The methods most commonly used for this purpose are pressure testing during early stages of field development and artificial tracer injection tests which are only applicable at later stages of field development A third approach is to examine the isotope chemistry of the fluids in the reservoir, ie. the formation water and the hydrocarbons, as well as the isotopic compositions of authigenic minerals formed in equilibrium with these fluids. The reasoning here is that diffusion and mechanical mixing processes should lead to uniform isotopic signatures in well connected bodies of fluid while variations in isotopic signature may be preserved in or evolve between isolated bodies of fluid. The stable isotope signatures of the organic fluids are mainly linked to the source rocks and isotopic variations observed in the reservoir are related to the reservoirs filling history. The Sr isotopic signature of formation waters are the result of interaction between the initial formation water and the minerals that comprise the formation, and isotopic variations may reflect both insitu diagenetic reactions or movements of fluid within the reservoir. In both the organic and inorganic cases pore volume architecture and communication control the preservation of isotopic variations. One problem with this type of approach is obtaining sufficientiy numerous, uncontaminated samples of formation fluids. We have developed techniques which allows us to look at variations in 87Sr/86Sr in fonnation waters at the meter scale which essentially means looking at the composition of waters in individual beds. Our results show complex variations in this parameter with depth. At the present time we do not fully understand the implications of these variations which seem to reflect the compound interplay of vertical permeability, diagenetic reaction rates, formation mineralogy, degree of hydrocarbon saturation and fluid movements within the reservoir.

66 Pb DISTRIBUTION IN ULTRAMAFIC NODULES FROM THE UPPER MANTLE BENEATH THE WESTERN UNITED STATES MEIJER, A., and ROGERS, P. S. Z., Isotope and Nuclear Chemistry, Los Alamos National Laboratory, Los Alamos, New Mexico, 87545, United States of America The distribution of Pb has been studied in ultramafic nodules collected from several localities in the western U. S. Highly purified clino- and orthopyroxene separates from several Type lA (i.e., depleted) nodules contain 10-40 and 5-20 ppb Pb, respectively. Importantly, the U/Pb ratios of the clinopyroxenes are high (50-120) while the Sm/Nd ratios are greater than the chondritic ratio. Part of the Pb in the nodules is contained within sulfide phases. PIXE probe analyses of the Pb content of various sulfide grains yield an average Pb concentration of 17 ppm. If the nodules contain on the order of 200 ppm S and if this sulfur is present entirely in the form of sulfide phases, these phases would contribute on the order of 10 ppb to the whole-rock Pb content. This is a significant (i.e., >30%) fraction of the Pb anticipated in depleted mantle materials given certain assumptions about the initial Pb concentration of primitive mantle material and degrees of melting required to produce the depleted character. On the basis of these and other observations, we suggest the U/Pb ratio of depleted mantle material is determined primarily by the clinopyroxene content and the clinopyroxene to sulfide ratio of the residual mantle remaining after partial melting.

NO INFLUENCE OF ICELAND HOTSPOT ON KOLBEINSEY MID-OCEAN RIDGE BASALTS

MERTZl, D.F.,DEVEY1, C. W., TQDT^, W., STOFFERSl, P., HOFMANN^ A.W. iGeol. Inst Uni. Kiel, Olshausenstr. 40, D-2300 Kiel, FRG ^Max-Planck-Inst. fur Chemie, Saarstr. 23, D-65(X) Mainz, FRG Published geochemical data from the Reykjanes Ridge south of Iceland show convincing evidence for mixing between MORB and the Iceland hotspot. To investigate whether the sample type of mixing occurs to the north of Iceland, we collected samples from the Kolbeinsey Ridge between Iceland's north shore and the Spar fracture zone at ca. 69° N. The 15 Kolbeinsey Ridge samples analysed for their isotope ratios show limited isotopic variations (^^Sr/^^Sr = 0.70280 to 0.70298, l43Nd/l44Nd = 0.51307 to 0.51323, 206pb/204pb = 17.912 to 18.053, 207pb/204pb = 15.365 to 15.440 and 208pb/204p5 = 37.459 to 37.657). In a global context, the Kolbeinsey basalts show the least radiogenic Pb isotopic ratios yet found on the MOR system, lying beneath the fields of OIB and MORB on a 206pb/204pb vs. 207pb/204pb diagram. Sr and Nd isotope ratios show systematic variations with latitude, becoming more enriched (^'^ST/^^ST increases, i43Nd/l44Nd decreases) towards Iceland. The Pb isotopes show no such relationship. These geographical variations imply &at the Kolbeinsey source had higher time-integrated Rb/Sr and Nd/Sm in tiie north than in the south, but that the Th/Pb and U/Pb ratios were constant (and low) over the whole region. Mixing calculations involving Pb vs. Sr or Nd isotopes allow us to exclude the Iceland source as a possible endmember in the genesis of the Kolbeinsey Ridge basalts. Furthermore, although there is good evidence for mixing between hotspot and MORB along the Reykjanes Ridge, and although the tectonic settings of the Reykjanes and Kolbeinsey Ridges are similar, the isotopic variations show that hotspot-MORB mixing along the Kolbeinsey Ridge does not occur. These data help to place limits on the extent of interaction between the MORB and OIB reservoirs.

ICOG 7

ABSTRACTS

IMS 1270 : A NEW SIMS INSTRUMENT DEDICATED TO ISOTOPIC RATIO MEASUREMENTS. H,N. MIGEON> E. DE CHAMBOST, B. RASSER, C. CONTY. CAMECA.92400 COURBEVOIE.FRANCE. In order to achieve optimum performance for isotopic ratio measurements at high mass resolution, Cameca has designed a new magnetic sector SIMS instrument. Based on the same design as the IMS 4F, the most utilized instrument with over 180 units in operation/ the IMS 1270 works in both the ion microscope and microprobe modes. The instriament is equipped with a primary ion column delivering a beam diameter adjustable down to 50 nm for the Cesium beam and 120 nm for Oxygen bombardment. The electron gun permits analysis of insulating specimens looking at positive or negative secondary ions. In the negative mode autocompensation of the electrical charges occurs thanks to a cloud of almost zero energy electrons created a few microns above the sample. The secondary ion optical system consist of an immersion-transfer system, a double focusing spectrometer with a magnetic sector having a radius of curvature of 585 mm and a multicollection system allowing simultaneous measurement of up to 5 isotopes. The use of both primary ion sources together with the charge compensation system and a high performance spectrometer-counting system will allow ultimate results for lead dating in zircons, measuring thorium isotopes or simultaneously collecting the oxygen isotopes in the negative mode. Preliminary results on instrumental performance and isotopic measurements will be shown.

EARLY PROTEROZOIC DETRITAL GARNET IN MESOZOIC ACCRETIONARY COMPLEX ROCKS, WEST ANTARCTICA. MILLAR, I. L., MDLNE, A. J., and PANKHURST R. J., British Antarctic Survey, Natural Environment Research Council, High Cross, Madingley Road, Cambridge CB3 OET, U.K. Accretionary complex rocks of the South Orkney Islands were metamorphosed at lower amphibolite facies at around 200 Ma. During an Sm-Nd isotopic study of garnet from these rocks, distinctive orange garnet shards were found in quartz-gamet-mica schists from Signy Island. Sm-Nd analysis of this garnet reveals that it is .significantly too radiogenic to have grown within the metasediments during Mesozoic times, and cannot possibly be genetically related to the whole-rock from which it was separated. Modelling of the isotopic data from the orange garnets suggests a minimum age for garnet growth of around 2.4 Ga. The orange garnet is thought to represent a Proterozoic detrital component. However, on recent reconstructions of Gondwana, the nearest sources of early Proterozoic detritus lie over 1000km from the South Orkney Islands, implying a long history of transport and deposition for the garnet. The detrital garnet has preserved its neodymium isotopic characteristics despite Mesozoic amphibolite facies metamorphism. In northern Antarctic Peninsula, Permian-Triassic sedimentary rocks of the Trinity Peninsula Formation (TPF) contain two populations of detrital garnet, one of which is orange in colour, and similar in appearance to the Proterozoic detrital component from the South Orkney Islands. However, this orange garnet falls on an isochron defined by garnets from basement rocks at Target Hill, northern Antarctic Peninsula, dating Carboniferous metamorphism. The presence of a possible source on the Antarctic Peninsula for the Carboniferous garnet component in the TPF suggests that the source region of Carboniferous detrital zircons within the formation may also be found on the Peninsula.

ABSTRACTS

ICOG 7

GEOCHRONOLOGIC STUDIES BEARING ON THE ORIGIN AND EVOLUTION OF MYLONITES BENEATH NORMAL FAULT SYSTEMS, BASIN AND RANGE PROVINCE MILLER, EX., GANS. P.B.. LEE. J.. Geology, Stanford University, CA. 94305 U.S.A. and BROWN. R.. Geology, La Trobe University, Bundoora, Victoria 3083 The origin and nature of low-angle detachment faults and the mechanisms by which they translate mid-crustal mylonites to the surface remain topics of much debate. An extensive 40Ar/39Ar and fission track data base in the Snake Range metamorphic core complex. Basin and Range province, U.S.A., provides new insight into these questions. The data indicate uplift is a protracted process accomplished by movement on a series of steeper faults rather than a single gently-dipping fault and that the low-angle of metamorphic core complex detachment faults is an artifact of successive superimposed strains and rotations. Combined 40Ar/39Ar geochronology and geologic constraints indicate onset of extension and mylonite development at about 7-8 km depth, 36 Ma continuing to 24 Ma, younging in the slip direction. Elevated T gradients existed during extension; if isotherms were subhorizontal, mylonites may have been generated at steeper angles than their present gentle dips. Apatite fission track ages from samples collected vertically beneath the detachment fault and parallel to its down-dip movement direction (26 samples) yield ages between 22±5 and 15±3 that are indistinguishable regardless of their east-west location, present elevation or higher T cooling history. Possibly somewhat younger ages 12±-14±2 were obtained from the structurally deepest exposures in canyon bottoms. Horizontal track length distributions have mean lengths of 14.6-15.1 um (S.D.=l.l-1.8um) and indicate relatively rapid cooling (15°C/km) through the apatite annealing zone (120-60°C) followed by residence at low, near-surface temperatures. Movement on a younger, steeper system of mapped faults was likely responsible for the observed rapid uplift of the range through 120°C between 20-15 Ma. Apatite FT ages of conglomerate clasts in the hanging wall of this younger fault system support this interpretation. Normal and reverse drag of the detachment and underlying ductile fabrics by movement on these younger faults produced the characteristic domical geometry of the complex we see today.

DIRECT EVIDENCES OF IMPACT FROM SHOCKED QUARTZ GRAINS IN KyT BOUNDARY MIURA, Yasunori, Faculty of Science, Yamaguchi University, Yamaguchi 753, Japan The Cretaceous-Tertiary (K/T) boundary problem can be solved by geochemical and mineralogical data. Contents of Pt-group elements (Ir etc.) and shocked quartz with multiple lamellar texture are considered to be the most confirmative data in determination of the K/T boundary. There are three kinds of the K/T boundary: 1) Abrupt IT anomaly and type I shocked quartz: this is observed in sea-sedimental layers of the Italian, Danish, Austrian, Tunisian K/T boundary samples. 2) Type I shocked quanz (type I) without In this is observed in Japanese K/T boundary sample from Hokkaido. 3) Abrupt IT anomaly and type II shocked quartz with multiple lamellae: this is observed in K/T b o u n d ^ sample from Colorado, U.S.A. Therefore, it is found that the above type II shocked data are surely the direct evidence of meteoritic impact process which could not be explained by normal terrestrial geologic (i.e. volcanic and tectonic) processes. References: [1] Alvarez L.W. et al. (1980): Science, 208, 10951107. [2] Bohor B.F. et al. (1986): Science, 224, ^ - m . [3] Miura Y. et al. (1988): Global Catastrophes in Earth History, 124125. [4] Miura Y. (1989): Meteoritics, 24-4, 305. [5] Izett, G. (1985): EOS, 66, 1149-1150. [6] Miura Y. (1990): LPSC XXI (NASA-LPI), 793-794.

CARBON AND NITROGEN STABLE ISOTOPE SYSTEMATICS OF PALEOHYDROTHERMAL FLUIDS ASSOCIATED WITH'S' TYPE GRANITES, S.W. ENGLAND Martin F. MILLER. British Geological Survey (NERC), Keyworth, Nottingham, NG12 5GG, UK; Stuart R. BOYD and Colin T. PILLINGER, Planetary Sciences Unit, The Open University, Walton Hall, Milton Keynes, MK7 6AA, U.K.

SULFUR DIFFUSION IN NATURAL SPHALERITE Z, MIZUTA. Institute of Mining Geology, Akita University, Akita, 010 Japan, B. J. Giletti, Department of Geological Sciences, Brown University, Pro\ddence, RI, 02912 U.S.A., and S. D. SCOTT, Department of Geology, University of Toronto, Toronto, Ontario, M5S 3B1 CANADA

The isotopic characterisation of trace quantities of carbon dioxide, methane, nitrogen and argon in ancient hydrothermal fluids provides a powerful investigative tool to constrain models relating to the provenance of the fluid. Although the isotopic composition of carbon-bearing species in crustal (and mantle) fluids has been the focus of attention in a variety of recent studies, little regard has been paid to nitrogen isotopes, although nitrogen is acknowledged to be a common fluid component in the continental crust. We report here the application of a low-blank, stepped thermal extraction procedure, devised to prepare sub-nanomole quantities of nitrogen gas for stable isotopic analysis by high-sensitivity static vacuum mass spectrometry, to investigate traces of fluid inclusion nitrogen in paleohydrothermal fluids associated with granite cupolas in the east of the Comubian batholith, S.W. England. The batholith is an St>'pe ilmenite series, two-mica tourmaline-bearing ânite, hosting extensive post-emplacement hydrothermal activity which also extends into the local low-grade, regionally-metamorphosed Devonian-Carboniferous volcanic-sedimentary envelope. -Âr-^Âr reports in the literature have documented a trace paleoatmospheric component in fluids in this region, whereas the present fluid inclusion b^^C (CO2) data suggest the predominant involvement of a lower crustal component, if ± e coexisting fluid inclusion methane traces derived from a separate carbon reservoir. Fluid inclusion nitrogen S^^N values in the region of +8%o were obtained, indicative of an organic precursor. The source of N2is probably the oxidation of NH4+, which is an abundant species in both the country rocks and locally within the batholith. Equilibrium thermodynamic modelling has been used to explore the reaction of NH4+ with C® and C^^ species, as an alternative to a solely diagenetic origin for the observed methane; the implications of this, in terms of carbon source reservoirs, will be discussed.

Diffusion profiles of ^^S tracer in natural sphalerite were determined experimentally as a function of temperature (580 - 7(XfQ using secondary mass spectrometry. Airhenius relations for diffusion coefficients of ^ S in iron-poor natural sphalerite (2 mole% FeS, Arasawa, Japan) and in iron-rich sphalerite (27 mole% FeS, Chihuahua, Mexico) with the ciystallographic direction normal to (110) are shown in Figure 1. Diffusion rates of ^ S in Fe-rich sphalerite arc 2 to 6 times higher than the rates in Fe-poor sphalerite. The faster diffusion of ^ S in Fe-rich sphalerite suggests the strong correlations to lattice constant increase with increasing iron contents in sphalerite. ^ Z n and ^"^Fc depth profiling shows that faster zinc-iron interdiffu1.0 1.1 sion simultaneously took 1000/T (K) place with ^^S diffusion. Fig. 1 Arrhenius relations for diffusion coefficients of ^ S in natural sphalerite.

Retrograde Meteoric-Hydrothermal Alteration of the Boehls Butte Anorthosite, Northern Idaho, USA MORA, Claudia I., Department of Geological Sciences, University of Tennessee, Knoxville, TN, 37996, U.S.A. The Boehls Butte anorthosite, exposed within the broad bor der zone of the northern ID batholith, remains one of the most enigmatic Proterozoic anorthosite bodies. A stable iso topic study was initiated to evaluate episodic fluid-rock inter action affecting the anorthosite, its unique plag mineralogy (and+lab+an) and complex contact with aluminous schists. Stable 0-H isotopic analyses of plag (undifferentiated), hbl and w.r. indicate that the Boehls Butte anorthosite variably interacted with heated meteoric-hydrothermal fluids after the peak of regional metamorphism. Values of 5^®0(SMOW) range from +7.9 to -5.8 o/oo (n=26). Plag and hbl are not in isotopic equilibrium, with A(plag-hbl)=-4.0 to -7.4. Values of 6D(hbl, biot) range from -51 to -136 o/oo (n=7), indicating fluid flow in several samples affected <5D, but was insufficient to shift Hydrothermal fluid flow was predominantly channelized, with gradients in 5"0(plag) as large as 11.9o/oo over 250 m. The anorthosite displays very fine (<0.5 mm) brittle or diffuse fractures, sometimes only visible using hot-cathode cathodoluminescence, and high degrees of fracturing correlate with low values of Values of and (5D(hbl) in gar-amphib dikes within the anorthosite appear unshifted and these rocks may have been impermeable to retrograde fluid flow. Evaluation of A(plag-H20) and A(hbl.H20) over a range of T (150-600 °C) suggests that the anorthosite did not equilibrate with the heated meteoric-hydrothermal fluids or meteoric-hydrothermal fluid compositions were substantially shifted. Hydrothermal alteration may be related to early Tertiary meteoric-hydrothermal convection developed around the Idaho batholith upon intrusion and rapid uplift of the terrane. THE SUBDUCTION COMPONENT IN ARC VOLCANISM: B, CS, RB, AND BA SYSTEMATICS MORRIS, Julie, TERA, Fouad, Dept. Terrestrial Magnetism, Carnegie Inst. Washington, 5241 Broad Branch Rd, Washington, D.C. 20015, U.S.A., LEEMAN, W.P., Rice University, Houston TX, 77251, and HARMON, R.S., NERC Isotope Geosciences Laboratory, Keyworth NG12 5GG, U.K. La^'as from the Aleutian, Bismarck, Central America, and S. Chile (33-42°S) arcs form well-correlated straight lines (r^ > 0.S6), each of unique slope, on a plot of i°Be/Be vs B/Be (Morris et al, Nature, 1990). Similzir simple relationships are beginning to emerge on plots of ^°Be/Be vs Rb/Be, Ba/Be and Ba/La for la\^ of the Bismarck and Central America arcs, and for ^°Be/Be vs Cs/Be for basalts of S. Chile. The apparent two component mixing trends imply that the subducted component (? fluids) in each arc is homogeneous in composition, and that the mass balance of ^°Be, B, Cs, Rb, and Ba is dominated by the subducted component, with all types of mantle and/or crust acting as indistinguishable diluents of the subduction signature. Thus, the subduction component may be isolated, chEiracterized and quantified. Siliceous volcanics of S. Chile depart from the basalt trend (subduction zone mixing) on a plot of ^°Be/Be vs Cs/Be and form a three component field indicating upper crustal contamination, allowing the two processes to be distinguished, even where both occur. Where good two component mixing systematics exist and crustal contamination can be ruled out, the mixing line can be extrapolated to specify the average composition of the sub-arc mantle prior to present-day subduction modification, as monitored by (Ti/2=1.5Ma). This approach suggests that slab-derived B, Rb, and Cs are not retained in the sub-arc mantle, while the residence time for Ba is long enough to permit its accumulation in the wedge through time.

ICOG 7

ABSTRACTS

FLUID MOVEMENTS RELATED TO DETACHMENT FAULTING IN THE WHIPPLE MT. METAMORPHIC CORE COMPLEX, USA: STABLE ISOTOPIC AND PETROLOGIC EVIDENCE MORRISON. Jean, ANDERSON, J.L. and DAVIS, G.A., Dept. Geological Sci., University of Soutliern California, Los Angeles, CA 90089-0740 USA Disseminated and vein calcite in mylonitic quartzofeldspathic gneisses and plutonic rocks from the Whipple Mt. metamorpliic core complex was precipitated by low temperature detachment fault-related fluids. The calcite has been analyzed for and to determine 1) the extent to which detachment faultrelated fluids have affected lower-plate rocks, and 2) the source of the fluids. The Whipple Mt. core complex documents widespread crustal extension in the SW Cordillera of N America. Principle structural features of the complex include 1) a mylonitic sliear zone in lower-plate rocks which manifests early ductile deformation, and 2) a low-angle, late-stage brittle detachment fault that separates upper- from lower-plate rocks, along which mid-crustal rocks have been brought to tlie surface. Aphanitic cataclasites and microbreccias underlie the detachment fault surface and range in thickness from a few cm up to 2 meters. Extensive hydrothermal alteration is associated witli detachment faulting and is manifest by chlorite ± epidote ± calcite. Effects of the hydrothermal fluids are most pronounced at tlie detachment fault, but veins of chlorite ± epidote ± calcite crosscut all lower-plate lithologies at di{?tances as much as 250-300m below the fault. The quartzofeldspathic gneisses contain between 1.41 and 0.01 weight percent calcite. Values of S^^OsjioK range from 8.9 to 33.2 per rail and ^^Cp^o values range from -8.4 to 2.9 per mil. The data support the petrologic evidence for low-T fluid circulation and suggest that the fluids were largely meteoric in origin.

ARCHEAN CRUST-MANTLE E V O L U T I O N OF THE WYOMING CRATON: EVIDENCE FOR A N C I E N T E N R I C H E D MANTLE M u e l l e r . P . A . , U n i v e r s i t y of F l o r i d a , G a i n e s v i l l e , FL 3 2 6 1 1 U S A , and W O O D E N , J . L . , U . S . Geological S u r v e y , M e n l o P a r k , CA 94025 USA T h e A r c h e a n Wyoming A g e P r o v i n c e can b e subdivided into at l e a s t t h r e e d i s t i n c t subp r o v i n c e s or terranes: t h e M o n t a n a M e t a sedimentary terrane (MMT), t h e B e a r t o o t h - B i g Horn m a g m a t i c terrane (BBMT), and t h e Wyoming G r e e n s t o n e terrane (WGT). E a c h area is characterized by d i s t i n c t i v e l i t h o l o g i c a s s o c i a t i o n s , a g e s , t e c t o n i c s t y l e s , and m e t a m o r p h i s m . T h e M M T and B B M T yield the oldest ages in the p r o v i n c e (>3.6 Ga) and two s i g n i f i c a n t c r u s t - f o r m i n g c y c l e s h a v e been identified at 2.75 Ga in t h e B B M T and at 3.25 Ga in t h e BBMT and M M T . C o m p r e h e n s i v e chemical (major, t r a c e , R E E ) , i s o t o p i c (Pb, N d , S r ) , and g e o c h r o n o l o g i c (U-Pb zircon) d a t a sets for rocks from these areas s u g g e s t t h a t sialic crust h a s been extracted from r e l a t i v e l y enriched m a n t l e during t h e i n t e r v a l 3.2-2.7 G a . The overall character of t h i s m a n t l e h a s been one t h a t exhibits e n r i c h e d i s o t o p i c r a t i o s . These include eNd's of - 1 to - 2 , ®^Sr/®^Sr of .7022, and secondary Pb i s o c h r o n s t h a t intersect S t a c e y - K r a m e r s Pb e v o l u t i o n curves at y o u n g e r a g e s , or lie a b o v e t h i s c u r v e . There is, h o w e v e r , no strong e v i d e n c e for comparably enriched elemental r a t i o s (e.g., low S m / N d ) . This a p p a r e n t c o n t r a d i c t i o n can b e resolved if recycling of old crust into t h e m a n t l e leads to partial isotopic e q u i l i b r a t i o n w i t h the mantle while t h e bulk mantle is d e p l e t e d in LIL elements as predicted b y t h e i r bulk D ' s .

ABSTRACTS

ICOG 7

Sr AND S ISOTOPIC CONSTRAINTS ON THE GENESIS OF STRATA BOUND (MANTO TYPE) COPPER DEPOSITS OF CHILE. MUNIZAGA, F., Dept. Geol., U. Chile P.O. Box 13518, Stgo-Chile. TASSINARI, C. & K. KAWASHITA, Inst, of Geosc. USP, Cx. P. 20899, Brasil. ZENTILLI, M., Dept. Geol., Dalh. Univ., Halifax, B3H.3J5 Canada. Volcanic-hosted, strata-bound (iranto type) copper depo sits constitute, in terms of production, the second most significant group in Chile. The source of metals, mineralizing fluids, timing and concentration processes for these deposits remains controversial, and models for their genesis range from volcanic syngenetic, to epigenetic with fluids of iragmatic, burial metamorphic, oceanic or meteoric origin. In an attenpt to constrain the genetic hypotheses and establish a ba seline for comparison among deposits, we have analyzed the isotopic canpositions of Sr in carbonates and S in sulphides from (A) Carolina de Michilla and Mantos Blancos, hosted by Triassic-Jurassic rocks, and (B) Lo Aguirre, El Soldado ard Punta del Cobre, hosted by Mid^etaceous rocks. 8'sr/S^Sr ratios for calcite within (A) range from 0.7055 to 0.7068, v^ereas those from (B) are lower at 0.7040-0.7056. tonineralized marine limestones of Jurassic and Cretaceous age within the host sequences ha ve corpositions of 0.7066-0.7093 and 0.7069-0.7071, respectively. 87sr/86sr in the calcites is indistingui shable from that of the host volcanic sequences and of essentially coeval intrusives. 6 values for ore sulphides in (B) range from -5.2 to +1.5 per mil and cluster around 0, suggesting a minor amount of fractionation, vtereas those of (A) range from -2.4 to -6.5 per mil. The data preclude the participation of large volumes of seawater in the genesis of most of the deposits and are compatible with the low water/rock ratio deduced for them by other authors. These deposits may not all have a common genesis, despite similarities in mineralogy and setting.

Rb-Sr DATING OF SPHALERITES FROM MISSISSIPPI VALLEY TYPE (MVT) DEPOSITS NAKAL S.. HALLIDAY, A. N., KESLER, S. E., AND JONES, H. D., Department of Geological Sciences, The University of Michigan, Ann Arbor, MI 48109-1063, U. S. A. The direct dating of sulfide minerals is one of the most difficult problems in geochronology. This problem hinders the understanding of the genesis of MVT deposits, which mainly consists of sphalerites, galena and gangue minerals. No direct dating of these type of the deposits has been tried with a few exceptions. We have tackled this problem by applying the Rb-Sr dating method to sphalerites. Sphalerites were crushed in deionized water and leached to remove their inclusion fluids. Although Sr concentrations in the sphalerite crystals were low (less than 1 ppm), they had high and variable ^'Rb/^^Sr ratios which allowed dating. Rb-Sr data for sphalerites from Coy mine. East Tennessee, define an isochron which gives an age of 377+27 Ma, with scatter which results from isotopic inhomogeneity of the ore fluids. These results indicate that the mineralization is older than mid-Paleozoic and are also inconsistent with the hypothesis that late Paleozoic fluid migration during the Alleghanian orogeny triggered the ore formation. Our preliminary results indicate that sphalerite is a suitable mineral for Rb-Sr dating, although at this stage we can not say where the Rb and Sr are located in the sphalerites. The application of this dating method will shed light on the genesis of MVT deposits.

69 THERMOLUMINESCENCE DATING OF QUATERNARY FLUVIAL AND AEOLIAN DEPOSITS IN AUSTRALIA. N A N S O N . Gerald C., PRICE, David M., Department of Geography, University of Wollongong, P.O. Box 1144, Wollongong, N.W.S. 2500, Australia, and SHORT, Stephen, A., Environmental Radiochemistry Laboratory, Australian Nuclear Science and Technology Organisation, P.M.B. 1, Menai, N.S.W. 2234, Australia. Study of terrestrial environmental change during the Quaternary has been seriously hampered by the lack of a reliable dating technique that extends the range of radiocarbon. Here we report thermoluminescence (TL) age determinations on alluvial sediments over the last 250,000 years in the Carpentaria and Eyre basins of northern and central Australia. These results are consistent with U/Th age determinations of pedogenic/diagenic minerals (gypcretes, calcretes and ferricretes) precipitated within these alluvia. Extensive alluvial sand sheets beneath 2-3m of overlying mud along the the channels of Cooper Creek and the Diamantina River yield two distinct groupings of TL ages. These sands were laid down by large meandering channels that in age broadly correspond with but slightly postdate the penultimate and ultimate interglacial maxima. They represent two episodes of distinct fluvial and probably pluvial activity in central Australia. A minor fluvial episode also occurred -40,000-50,000 years ago. Interestingly, the Holocene "interglacial" as yet shows no sign of pronounced fluvial activity. The overlying alluvial muds date at less than 85,000 years ago and represent a mostly arid phase; 17 TL dates from aeolian deposits on interfluves in the Eyre Basin show the continental dune fields to have been active over this period, culminating in a period of intense aridity and dune building during the last glacial maximum. Aeolian TL dates of -170,000, and -275,000 years suggest that arid phases separated the last two interglacials and preceded the older one but apparently dune migration has destroyed most of the TL evidence for these.

Sm-Nd ISOTOPE SYSTEMATICS IN LATE ARCHAEAN SUPRACRUSTAL SEQUENCES FROM THE PILBARA AND KAAPVAAL CRATONS D.R. NELSON. A.F. TREND ALL, Geol. Survey of Western AustraUa, Perth, 6000, J.R. de LAETER, I.R. FLETCHER, School of Physics and Geosciences, Curtin Univ. of Tech., Perth, 6001, and N.J. GROBLER, Dept. of Geology, Univ. of the Orange Free State, Bloemfontein 9300, South Africa. Sm-Nd isotopic data were obtained on the ca. 2.7 Ga igneous rocks of the Fortescue Group of the Pilbara Craton, Western Australia and the Ventersdorp Supergroup of the Kaapvaal Craton, southern Africa. Fortescue Group samples have Sm/Nd ratios indicating moderate LREE-enrichment and e^d values between -1.5 and -4.4. No relationship exists between the extent of differentiation and Sm/Nd or Although these characteristics are consistent with extensive crustal contamination (ca. 50% of the original mass), geochemical considerations and die absence of zircon xenocrysts in the Fortescue Group felsic lavas argue against crustal contamination and in favour of derivation from LREE-enriched, low e^d mantie sources. Ventersdorp Supergroup samples are moderately LREEenriched and have negative values, between 0 and -3.4, which are positively correlated with Sm/Nd. The presence of zircon xenocrysts in some Ventersdorp samples provides compelling evidence of some crustal contamination, but their comparatively low abundance is difficult to reconcile with the extensive amounts of crustal assimilation required to explain the observed LREEenrichment and negative e^d values if this were the only cause, and an enriched mantle origin for the Ventersdorp igneous rocks is probable. Sm-Nd isochron dates of 3078 ±107 Ma and 3308 ±138 Ma were obtained for Fortescue Group samples from the west Pilbara and for Ventersdorp Supergroup samples respectively. These dates are significanUy older tiian the eruption ages of these sequences indicated by U-Pb zircon data, and suggest that Sm-Nd fractionation occurred within the subcontinental lithospheric mantle sources of tiiese rocks within ca. 500 Ma of formation of the Kaapvaal and Pilbara granite-greenstone terranes. Similarities in the field geology, geochemistry and isotopic characteristics "of the Fortescue and Ventersdorp sequences and of Phanerozoic examples of continental flood basalt volcanism suggest a common mode of origin, possibly involving derivation from subcontinental lithospheric mantle sources which have been modified by subduction processes.

ICOG 7

ABSTRACTS

EVOLUTION OF THE ALDAN SHIELD, EASTERN SIBERIA A.A.Nemchin, Institute of Precambrian Geology and Geochronology, Leningrad, USSR. The south eastern part of the Archaean Aldan Shield of Eastern Siberia consists of the Olekma Granite-Greenstone Terrain, the Aldan Granulite-Gneiss Terrain and the Stanovik mobile belt The Olekma Granite-Greenstone Terrain contains two generations of greenstone belts. The Tungurcha greenstone belt is the oldest with volcanism dated at 3.2-3.3 Ga. Volcanism in the younger Olondo and Tas-Miele greenstone belts took place 3.0 b.y ago. The belts are surrounded by tonalite-trondhjemite bodies which were emplaced synchronous with 3.0 Ga volcanism. Negative eNd (about 3-4 units) for some trondhjemites suggests an earlier sialic crust. The last tectono-thermal event took place 2.6-2.7 b.y. ago and was accompanied by the intrusions of K20-rich granites. Local shearing occurred 2.0 Ga ago. Ion microprobe analysis on zircon from trondhjemite gneisses of the Aldan Granulite-Gneiss Terrain give an age of 3.35 Ga (Nutman, personal communication, 1989). Two younger granite emplacement events at 3.0 Ga and 1.8-2.0 Ga occur in the Aldan Granulite-Gneiss Terrain. 2.0 Ga granites were emplaced simultaneously with granulite facies metamorphism .The Sm-Nd data suggests these granites were generated by remobilisation of the gneisses. Similar events took place in the Stanovik mobile belt. The last event was a high grade metamorphism dated at 1.82.0 Ga. This reached granulite facies north of the Stanovik Fault and upper amphibolite facies south of the fault. Cratonization of the Aldan Granite-Gneiss Terrain occurred about 3.3 b.y. ago. Granitoids emplaced at about 3.0 Ga and 2.0 Ga were generated by the reworking of older gneisses without the addition of new crustal material. Granitoids in the Olekma Granite-Greenstone Terrain and in the Stanovik mobile belt represent the addition of new sialic material at about 3.0 Ga.

LEAD ISOTOPES, ^ ^ ^ ^ PARAMETER AND THE CRUSTAL PRE-HISTORY OF ROCKS Nevmark. Leonid A., Institute of Precambrian Geology & Geochronology, USSR Academy of Sciences, Leningrad, 199034, U.S.S.R. The delta-type diagram (Stacey, 1986) is modified using 207pb/204pb versus 206pb/204pb fractional deviations from the model mantle values of corresponding age. Parameters used in the model for Pb-isotope evolution in depleted mantle are: Tp = 4.57 Ga, |ii = 3.6, T2 = 4.5 Ga, \X2 = 8.27. These were chosen to give modern values for lead isotopic ratios equal to those of the intersection between NHRL and the 4.5 Ga geochron. The data falling into the NE- and SW-quadrants of the above plot enable the age of a source-region of the rocks under investigation to be determined using the equation: 20rpb/206pb = (1/137.88) (exp X^^Hjû' exp X^T) (exp Y^ToM-exp XiD (T - age of rock, p^T^m - model age of the source-region). p^Tdj^ value is analogous to Nd model age (^^Tî^), but is independent of the mixing of mantle and crustal materials during petrogenesis. ^^T^û has the additional advantage of allowing model age determinations for rocks with Sm/Nd close to that of DM. Possible intracrustal U-Pb fractionation (metamorphic or sedimentary processes) is a significant limitation of this approach. Marked differences between the ^^Tdj^ model ages for modem oceanic volcanics source-regions (^1.8 Ga) and Phanerozoic continental igneous rocks and modem pelagic sediments U2.9 Ga) have been revealed and are interpreted to reflect the upper limit of the age of the oldest recycled oceanic crust and of the separation from depleted mantle of continental type crust correspondingly. Curves for relating both 238u/204pb and 232Th/238u (k) with time are obtained, which show increasing ji and decreasing K in DM through geological time. The reason for this might be the recycling of oceanic crustal components. The geochronological patterns of ^^T^j^ for some igneous and metamorphic rocks from the USSR and Mongolia are treated and compared with istopic ages and ^^T^j^ values. A good agreement between ^^T^j^ and ^^Tj^î parameters exists for crustsal derived granitoids with negative EncI- Granitoids of a mixed crust-mantle origin have ^^Tpj^ values lower than ^Tj^^-

HYDROTHERMAL PROCESSES DURING FORMATION OF THE CONTINENTAL CRUST NEWSOM. Horton. and SIMS, Kenneth, Dept. of Geology, University of New Mexico, Albuquerque, NM 87131, USA

I-XE STUDIES OF ALLENDE AND TIESCHITZ CHONDRULES AND RIMS NICHOLS , R.H. JR., HOHENBERG C.M. and OLINGER C.T., McDonnell Center for the Space Sciences, Physics Department, Washington University, St. Louis, MO 63130 USA, RUBIN A.E., Instiuite of Geophysics and Planetary Physics, University of California, Los Angeles, CA 90024 USA

During oceanic mantle-derived volcanism, Newsom et al. [1] and Sims et al. [2] showed that Mo, As, Sb, and Pb are incompatible elements, and correlate with the light REE, such as Ce or Pr. Tungsten is a highly incompatible element and correlates with U, Th or Ba. We have obtained new data for Mo, W, As and Sb in continental crustal material, including loesses. For the elements As, Sb and W, we have discovered evidence of an extra enrichment in the upper crust, compared to the enrichment expected from their incompatible behavior during oceanic volcanism. A similar enrichment of Pb was discovered by Newsom et al. [1], where it was shown from isotopic data that the expected ratio for the primitive Earth's mantle fell between the mantle and crustal ratio, consistent with an extraction of Pb from the mantle. In contrast, the ratios for Mo (e.g. Mo/Ce) are essentially constant for samples of mantle and crustal materials of all geological ages. The elements affected by the extra enrichment, As, Sb, W, and Pb are consistent with a hydrothermal enrichment process. The possibility that the enrichment is due to an upper-crust lowercrust fractionation is currently being investigated. In the oceanic samples, the constant ratios of As, Sb and W with lithophile elements of similar incompatibility, are consistent with a two stage model, proposed by Hofmann [3] based on data for Pb, involving early extraction of the crust from the mantle followed by homogenization of the mantle. References: [1] Newsom et al. (1986) Earth Planet. Sci. Lett. 80. 299. [2] Sims et al. (1990) in Origin of the Earth. Newsom and Jones eds., Oxford Press, in press. [3] Hofmann (1988) Earth Planet. Lett. 90, 297.

I-Xe studies were performed on four chondrules and four coarse-grained chondrule rims from the Allende meteorite; similar analysis is in progress for Tieschitz. Noble gas data collected here help constrain the conditions of chondrule/rim formation and subsequent metamorphism. Ages are calculated relative to a Bjurbole standard. I-Xe closure occurred within 2 Ma of Bjurbole for all Allende samples. The I-Xe ages of the rims are generally older than the chondrules. One out of four chondrules formed before Bjurbole whole rock, but three out of four rims pre-date it. In the one chondrule-rim pair, the rim is clearly older than the chondrule. Because the I-Xe ages are based on Xe released at higher temperatures than the temperature at which rim sintering occurs (~ 1000 °C), the I-Xe structure may not have been ifected by rim formation. Thus, the I-Xe ages probably refer not to the rims themselves, but to their fine-grained matrix-matend precursors. Increasing initial 129i/r27j increasing extraction temperature indicates that xenon closure occurred first in the higher temperature mineral phases, suggesting cooling or postformational alteration of the I-Xe system. Estimated "cooling" rates (10-300 °C/Ma), coupled with the near-equivalent trends observed for the chondrules and rims, initially suggest that post-formational processes in the regolith, rather than the nebula, are responsible for the I-Xe fine structure. However, we conclude that I-Xe ages are probably dating nebular events rather than parent body processes with the regular structure observed occurring either in the parent body or in the nebula. The near antiquity (from the short half-life of ^^^I) and nearisochronism suggested by the overall I-Xe patterns in all objects indicate that the observed processing must have occurred quite early.

ABSTRACTS

ICOG 7

Re-Os SYSTEMATICS OF IRON METEORITES NIEMEYER, S., Gerlach, D., Russ, G. P., Nuclear Chem. Div., Lawrence Livermore National Laboratory, Livermore, CA 94550, U.S.A. Our Re-Os chronological study of meteorites has initially focused upon the lAB iron meteorites because of their chondritic affinities and their apparent early formation, as inferred from 4.5 Ga ages for silicate inclusions. However, recent laboratory measurement of the half-life coupled with previous Re-Os analyses of iron meteorites indicates an age of only 4.15 Ga for irons of all classes. To reduce uncertainty, we have calibrated our Os isotopic spike against the Os standard used in the half-life experiment. We have used a refined technique to analyze four lAB iron meteorites and have overcome earlier difficulties in achieving isotopic equilibration between Os in spikes and samples. Three samples lie within error of the 4.55 Ga "isochron" defined by Walker and Morgan's (1989) fit to a subset of their carbonaceous chondrites. Model ages calculated using the corresponding initial i87os/1880s of 0.0965 average 4.58 Ga. Using the initial Os isotopic ratio of 0.0977 defined by iron meteorites, the average model age for these three irons is 4.42 Ga. Thus, Re-Os data for these irons are consistent with a closed system evolution since the formation of chondrites. In contrast, a fourth lAB iron, Pitts, lies below the 4.55 Ga isochron with a model age of 4.0 Ga. E i t h e r it has a lower initial than the other lABs, or it was disturbed at this later time. More lAB samples will be analyzed to more precisely define an isochron for this class, as well as explore the extent to which some lAB irons have been disturbed. Then we will continue our previous work on IIIAB irons which indicated ages significantly less than 4 Ga. Finally, we will analyze three samples of the highly unusual HE irons which have silicate ages ranging from 3.8 to 4.6 Ga. This work was performed under the auspices of the U.S. Dept. of Energy by LLNL under contract No. W-7405-Eng-48.

CRUSTAL EVOLUTION IN T H E SUPERIOR PROVINCE: U-Pb AND Nd ISOTOPIC CONSTRAINTS FROM T H E T R O U T LAKE CONFEDERATION LAKE GRANITEGREENSTONE TERRANE. NOBLE, S.R., Royal Ontario Museum, Toronto / NIGL, Keyworth, NG12 5GG U.K., EVENSEN, N.M., Dept. of Geology, University of Toronto, M5S l A l and KROGH, T.E, Royal Ontario Museum, Toronto, M5S 2C6, Canada.

MEASUREMENTS OF EXPOSURE TIME AND EROSION RATES OF TERRESTRIAL SURFACES USING IN SITU PRODUCED lOBe, 26A1, AND 2lNe IN QUARTZ NISHIIZUMI. K., KOHL, C. P., ARNOLD, J. R., GRAF, T., MARTI, K., Dept. of Chemistry, Univ. of Calif., San Diego, La Jolla, CA 92093, LAL, D., Scripps Inst. Oceanography, Univ. of Calif., San Diego, La JoUa, CA 92093 Measurements of in situ produced cosmogenic nuclides in terrestrial surface material, made possible by the development of accelerator mass spectrometry and high sensitivity noble gas mass spectrometry, permit absolute determination of exposure time and erosion rates for short and long range geomorphologic processes covering the time period 10^ to 10'^ years. We have developed and applied the in situ 10Be-26Al dating method in collaboration with the University of Pennsylvania Accelerator group. This method is seen to have several advantages over other proposed datmg methods. The production rates of l^Be and ^Âl were determined to be 6.0 and 36.8 atom/g Si02 year at sea level (geomagnetic latitude >50°); the ratio ^âI/IObc at production is 6.L The method has been applied in several areas of study. (1) Antarctic mountain ranges : Extremely low erosion rates (lO'^ cm/y or less) and long surface exposure times (over 10^ to 10^ years) were found for most of the locations. (2) The dating method was successfully applied to determine the time of impact for Meteor Crater, Arizona, and is being applied to date Australian meteorite craters. The pace of erosion has been found to be slow for the Australian continent. (3) A variety of samples (sands, sediments, loess, etc.) are being investigated as an initial sur\'ey prior to a world wide erosion study. Measurements of in situ produced îNe in aliquots of the same quartz samples can give integral cosmic ray exposure times allowing a more realistic modehng of exposure history. .5-10x107 atoms/g of cosmogenic îNe were found for two Antarctic rocks. This is the range expected from the calculated 10Be-26Al exposure in these two cases. A third sample has a large îNe excess, indicating a complex exposure history. ^He concentrations in at least one sample is low, indicating incomplete retention in quartz. l^C measurements in aliquots of the quartz are underway.

PROTEROZOIC SUPERCONTINENT AND OCEANIC AN0XL\: EVIDENCE FROM SULFUR ISOTOPES AND TRACE ELEMENTS IN BLACK SHALES FROM THE McARTHUR BASIN, NORTHERN TERRITORY, AUSTRALIA NORMAN, M.D., Research School of Earth Sciences, Australian National University, Canberra, A.C.T. 2601 Australia, DONNELLY, T.H., CSIRO Division of Water Resources, Canberra, A.C.T. 2601, JACKSON, M.J., and CRICK, I.H., Bureau of Mineral Resources, Canberra, A.C.T. 2601 Middle Proterozoic sediments in the McAnhur Basin comprise a thick (>6 km) sequence of shales, sandstones, and carbonates deposited 1800 to 1400 Ma. Some shales of the Roper Group are rich in organic matter (up to 8% TOC), low in thermal mauirity, and probably sourced the live oil discovered during drilling in the basin. Evidence for the depositional environment of these black shales is contradictory. Sedimentological indicators (e.g., glauconite) favour an open marine setting, but pyrite sulfur isotopic compositions are quite heavy to 4-46) and more consistent with a lacustrine or lagoonal setting with a restricted sulfate supply. Trace element data demonstrate large enrichments of V (to 400 ppm). Mo (to 130 ppm) and U (to 15 ppm) in the organic-rich units relative to average shale. Similar enrichments are common in marine black shales, but are virtually unknown in lacustrine shales. We therefore favour a marine setting for these shales, although the unusually heavy pyrite sulfur isotopic compositions require conditions atypical of modern marine environments. The low thermal maturity, large range of S^'^S, and lack of evidence for hydrothermal mineralization in Roper Group shales suggests that vigorous bacterial sulfate reduction was responsible for pyrite formation. If so, such heavy pyrite sulfur isotopic compositions may be a consequence of several factors, including locally restricted conditions within a marginal marine rift basin, rapid rates of sulfate reduction vs. seawater diffusion into the sediments, or seawater with unusually low sulfate concentrations or heavy sulfur isotopic compositions. Several mid-Proterozoic occurrences of heavy pyrite sulfur are known, suggesting global development of a common set of conditions. The formation and eventual break-up of a supercontinent, and associated oceanic anoxic events during the Proterozoic may account for these occurrences.

The Uchi-Confederation greenstone belt (UCB) and the adjacent Trout Lake batholith (TLB) in the western Uchi Subprovince have a longer history of volcanism and plutonism and a greater range in e Nd than most other Superior Province granitegreenstone terranes. Episodic UCB volcanism occurred at approximately 2.96 Ga, 2.84 Ga and 2.74 Ga. The bulk of the volcanism was part of a regional 2.76-2.735 Ga arc-building event. Two stages of intrusive activity at 2.838-2.806 Ga and 2.699 Ga in the TLB were asynchronous with major volcanism in the region, e Nd's, mainly from felsic volcanics in the UCB and TLB tonalites, range from -.7 to .4 at 2.96 Ga for the UCB, -1.5 to .5 at 2.84-2.806 Ga for the UCB and TLB, -2.4 to 1.7 at 2.741-2.735 Ga for the UCB. Nd data and geochemical constraints suggest that the TLB and many UCB volcanics were derived partly from crustal and/or LREE-enriched mantle sources ( > 3 Ga old), and depleted mantle (DM) rather than derivation solely by mixing of crustal and DM-derived material. By 2.8 Ga regional variations in isotopic reservoirs had developed; western Uchi Subprovince crustal and enriched mantle reservoirs had greater LREEenrichment and/or an earlier separation from DM than in the eastern Uchi Subprovince. The LREE-enriched reservoirs remained an important influence throughout the development of the UCB-TLB terrane. Geological and isotopic data favor the interpretation of the UCB as a tectonic stack and the UCB-TLB terrane to be the product of episodic collision and tonalitic intrusion of three volcanic arcs in a convergent margin setting.

GEOCHEMISTRY OF QUATERNARY VOLCANIC ROCKS FROM CONTINENTAL COLLISION ZONE IN EASTERN TURKEY Kenji NOTSU, Laboratory for Earthquake Chemistry, University of Tokyo, Tokyo 113, Japan, Tadahide UI, Jun-ichi MATSUDA, Department of Earth Sciences, Kobe University, Kobe 657, Japan, and Tuncay ERGAN, MTA, Geology Department, Ankara, Turkey Turkey forms a part of the Alpine-Himalayan convergent system. In the eastern Turkey, the Arabian plate collides with the Eurasian plate and volcanism related to the continental collision, giving rise to calc-alkarine and alkarine magmas, occurred during Neogene and Quaternary times. The volcanic centers are aligned across the plate boundary of the Eurasian and Arabian plates; from NE to SW, Ararat, Tendurek, Suphan and Nemrut volcanoes on the Eurasian plate and Karacadag volcano on the Arabian plate. We determined major, trace element, Sr isotopic compositions and K-Ar ages of Quaternary volcanic rocks from these 5 volcanoes. Sr-87/Sr-86 ratios of 4 calc-alkaline andesites from Ararat and Suphan volcanoes are in the range 0.70398 0.70467, and are similar to those from many island arcs. However, trace element abundance patterns are different between Turkish and island arc calc-alkaline rocks, implying the lower-crust melting model for the genesis of Turkish calc-alkaline magmas. Sr-87/Sr-86 ratios of 10 alkaline rocks from Nemrut volcano scatter widely, from 0.70352 to 0.70823. The lowest value, 0.70352, is obtained in the primitive basalt, and is within the range of the ratios of Karacadag basalts (0.70334-0.70389 for 11 samples). Trace element abundance patterns of Nemrut and Karacadag basalts are quite similar. These features suggest that the source mantle of alkaline basalts is chemically and isotopically similar under the Eurasian and Arabian plates, although crustal structures of the both plates are completely different due to the collision of the Arabian plate to the Eurasian plate.

SPATIAL, TEMPORAL AND ARTIFICIALLY INDUCED VARIATIONS IN ISOTOPIC COMPOSITION OF STEAM IN THE GEYSERS GEOTHERMAL FIELD Nuti Sergfo, Fancelli Ruggero and D'Amore Franco, International Institute for Geothermal Research. 56126 Pisa. Italy Strong spatial zonations have been observed in the isotopic compositions of fluids produced in many vapor-dominated geothermal fields. Studies of the behaviour of the isotope compositions of steam delivered by geothermal wells during production have led to" development of models of the origin and circulation of fluids in vapor-dominated systems. The temporal evolution of the isotopic composition of the fluid from some wells in the southeastern part of The Geysers geothermal field led to develop models of steam origin. Here we present a model that interprets all observed isotopic compositions in this part of The Geysers, their spatial distribution (horizontal and vertical) and their variations with time induced by exploitation and reinjection. This zone of the field is located close to the recharge area. Meteoric water infiltrated into the underground (delta oxygen-18 « -8 and delta deuterium = -54) and isotopic exchange between water and the rocks (a maturation process) determined an oxygen-shift. The resulting vapor will have an unchanged hydrogen isotopic composition, whereas the concentration of oxygen-18 will increase from the southeast edge of the field toward the centre; partial condensation originated an overlying condensate layer more concentrated in heavy isotopes whith formation of a gas cap at shallower levels of steam of a lighter composition. In the study area reinjection started at the same time as production in 1980. Oxygen and hydrogen isotopic compositions of injected water are much less negative than in the deep steam due the concentration through fractionation of heavy isotopes in the condensate from the cooling towers. Contribution to production from this source will start first in the wells that have the best structural connection with the plume around the injection well. With time the number of wells affected by this source will increase as a consequence of increase in size of the liquid plume and the local pressure gradients.

ICOG 7

ABSTRACTS

NEW OLD RCX:KS FROM GREENLAND A.P. NUTMAN, Research School of Earth Sciences, Australian National University, Canberra, A.C.T. 2601, Australia The AkuUeq terrane of the Godthibs^ord region, southern West Greenland is bounded by Archaean faults and contains rocks of very varied age and origin, which were assembled in the late Archaean. Two major bodies of early Archaean rocks occur in the Akulleq terrane; the Faeringehavn sub-terrane in the south and the Isukasia sub-terrane in the north. New SHRIMP zircon dating has been carried out in order to resolve more detail in the history of these early Archaean rocks. In the Fasringehavn sub-terrane, the pol)yhase Amitsoq gneisses contain enclaves of supracmstal rocks, which are collectively grouped together as the AkQia association. An Amitsoq tonalitic gneiss which intrudes the type Akilia association unit (comprising amphibolites, ultramafic rocks and quartzites) on Akilia island yields a weighted mean 207pb/206p5 ^g^ ^f 3877±10 Ma (2a), making this the oldest known rock in Greenland. From field evidence, the Akilia association unit at this locality must be even older, which at >3880 Ma makes it the oldest-known body of supracmstal rocks in the world. However, it need not follow that all bodies of supracmstal rocks that have been grouped together as the Akilia association are this ancient. In the Isukasia sub-terrane, further SHRIMP U-Pb zircon studies of the Isua supracmstal belt have been carried out. A layered felsic rock from the belt yields a weighted-mean ^^"^PhP-^^Pb age of 3808±4 Ma (2a). This agrees well with the SHRIMP zircon age of 3807±2 Ma for a previously analysed felsic rock from the belt (Compston et al., 1986). These ages must give the maximum age of the Isua supracmstal belt and may be close to its tme age. The possibility that >3800 Ma cmst contributed to the sedimentary rocks of the belt (Jacobsen & Dymek, 1988) will be discussed in the light of SHRIMP dating of detrital zircons in Isua supracmstal quartzites and pelites. Compston., W., et al., 1986: Earth Planet. Sci. Lett. 80,71-81. Jacobsen, S. & Dymek, R., 1988: J. Geophys. Res. 93, 338-354.

DETAILED GE0CHR0N0L(X3Y OF THE ARCHAEAN AULIAN GNEISS COMPLEX (AGC), UKRAINIAN SHIELD, USSR, ACCORDING TO SHRIMP lON-MICROPROBE STUDY ON INT>IVIDUAL GENERATIONS OF ZIRCONS A.P. NUTMAN, W. COMPSTON, Research School of Earth Sciences, ANU, Canberra ACT 2601, Australia, and I.V. CHERN^SHEV. A.V. SAMSONOV, IGEM of the USSR Acad. Sci., Staromonetny 35, Moscow 109017, USSR Zircon populations of AGC rocks have been divided into several t>'pes on the basis of grain morphology. The U-Pb age of these t ^ e s was determined by SHRMP and was used to resolve the details of the evolution of the AGC. Most of the 206p5/238u and 207p|3/206p|5 g^gg concordant or subconcordant; the weighted mean 207p5/205pt, values for the each zircon type are used as the best age estimates. In an older migmatitic group of gabbros, tonalitic and trondhjemitic gneisses, magmatic zircon forms -80, 50 and 5% of the total zircon population respectively, and yield ages of 3178±9, 3196±13 and 3122±2 Ma. Metamorphic zircons in these gneisses yield indistinguishable ages; 3079±9 Ma (tonalitic gneisses) and 3081±6 Ma (trondhjemitic gneisses). Zircon in amphibolite (metagabbro?) is represented by two types interpreted to be metamorphic in origin. The ages of these two types are indistinguishable and together yield an age of 307 8±3 Ma. In a younger group of non-migmatised tonalite and trondhjemite, the ages of magmatic (80% abundance) and metamorphic zircons are indistinguishable, and yield ages of 3073±13 Ma (tonalite) and 3058±7 Ma (trondhjemite). Thus the AGC appears to have formed over 100-120 million years. Gabbro and tonalite protoliths were formed in the period 3180-3190 Ma, followed by high grade metamorphism and further tonalite and trondhjemite injection between 3060 and 3080 Ma. The zircon results provides no evidence that any of these rocks are -3900 Ma, postulated from Sm-Nd whole rock isotopic studies of the gabbros and amphibolites.

ABSTRACTS

IC0G7

Melts and otherfluidsin the development of continental crust. R.K. O'Nions Department of Earth Sciences, University of Cambridge, Downing Street, Cambridge CB2 3EQ (U.K.) Generation of continental crust must involve more than one stage of melting. One of these involves melt fractions that are sufficiently small to produce thefractionatedtrace element budget characteristic of all continental crust. These melts are predominandy hydrous. A key feature of the small melts is their tendency to freeze into cool lithosphere, but they are easily remobilised subsequently. Examples of these processes and their role in continental evolution are discussed. The volatile budget of the small melts in the past and at present may be quantified once the melt fractions as a function of depth are determined from lithophile trace element abundances. In areas of continental crustal extension where the amount of melt produced during extension is determined in this way, the associated volatiles are traced readily particularly using rare gases, and their residence time in continental crust thereby assessed. The principal method of volatile loss from the continents is essentially through fluid circulation. The distribution of heat production in the continental crust is assessed in temis of melting and hydrous fluid transport.

MEGACRYST-BASALT RELATIONSHIPS: ISOTOPIC CONSTRAINTS SUZANNE. Y. O'REILLY. School of Earth Sciences, Macquarie University, Sydney 2109, Australia and W.L. GRIFFIN, CSIRO Div. of Exploration Geoscience, Box 136, N. Ryde, NSW 2113, Australia Megacrysts of anorthoclase, clinopyroxene and amphibole are common in primitive Tertiary basalts of New South Wales and Victoria, Australia. Their origin and relationship to their host basalts has been controversial. Analysis of Sr and Nd isotopes in 25 megacrysts from eight localities has shown that: (1) Individual megacryst species at a single locality show ranges in Sr and Nd isotopic composition that exceed analytical error by factors of 5-50. (2) Different megacryst species from a single locality are not isotopically identical with one another, nor with the host basalt (3) No meaningful isochron relationships exist among the megacrysts from any locality. (4) The general range in Sr and Nd isotopic composition of the megacrysts is roughly equivalent to the range in the basalts within the same province. These data indicate that the megacrysts are not deepseated phenocrysts of dieir host basalts; they represent the crystallization products of earlier magmas, intruded into the lower crust and/or upper mande. The ranges in isotopic composition from each locality suggest the repeated intrusion of generally similar, but isotopically distinct, batches of magma. This intrusion may be related to the Tertiary magmatism, or may be older. Since megacrysts are not genetically related to their host magmas, megacryst/basalt partition coefficients should be used with great caution.

GEOCHEMICAL TERRANES OF THE UPPER MANTLE BENEATH EASTERN AUSTRALIA O'REILLY S.Y.I, CHEN. Y.D.I and W. L. GRIFnN^ 1. School of Earth Sciences, Macquarie Uni., Sydney, NSW, 2109, Australia; 2. Div. Explor. Geosci., CSIRO, N. Ryde, NSW, 2113, Australia Isotopic studies on peridotitic xenoliths entrained in alkalic basalts from Qld, NSW and western Victoria have revealed distinct geochemical terranes in the upper mantle beneath eastern Australia which may correlate with different tectonic histories. The xenolith suite from Victoria shows the greatest isotopic range, forming an elongate field spreading into the "enriched mantle" region. eNd ranges from +6.4 to -8.3 and initial ^^Sr/^^Sr from 0.70335 to 0.71053. The xenolith suite from southern New South Wales defines a more restricted field but still shows a spread in 87sr/86sr (0.70287 to 0.70837) with eNd from +8 to -2.0. In contrast, the Queensland xenolith suite is more radiogenic in eNd (+11.8 to -1.6) but has an even more restricted range in 87sr/86sr (0.70232 to 0.70565). These isotopic differences also reflect variations in trace element abundance and mode. For example, LREE and Sr range to higher values in the Victorian xenolith suite and metasomatic phases (eg amphibole, mica apatite) are more common. The major metasomatic effects evident in the Victorian and NSW xenoliths, while geochemically distinctive, are consistent with different degrees of metasomatism of a mantle wedge by fluids and melts derived from subduction processes. Fragments of such subduction-affected basaltic melts may be represented by granulite/ pyroxenite mantle xenoliths from southern and central NSW. These xenoliths define trends in eNd and eSr consistent with mixing between maride wall-rock and an end member which is eidier similar to ocean-island or sea-floor basalts altered by sea water, or to old upper crust material reflecting long-term enrichment in Rb/Sr and LREE. The Qld xenoliths do not show strong geochemical evidence of subduction effects. The large population of Victorian xenoliths analysed isotopically also reveals the imprint of at least three distinct metasomatic episodes at times ranging from about 300-500 Ma to recent. As Phanerozoic eastern Australia has evolved by successive accretion of marginal arcs, it is reasonable to find components from subducted material and Paleozoic subduction-related volcanism in the upper mantle beneath parts of Eastern Australia. APATITE FISSION TRACK EVIDENCE FOR TERTL\RY UPLIFT IN THE NORTHEASTERN BROOKS RANGE, ALASKA O'SULLIVAN. Paul B., Dept. of Geology, La Trobe University, Bundoora Victoria 3083, Australia; DECKER, John, ARCO Alaska Inc., Box 100360, Anchorage, AK 99510; BERGMAN, Steven C., ARCO Oil and Gas Co., 2300 W. Piano Pkwy, Piano, TX 75075; WALLACE, Wesley, K., Dept. of Geology and Geophysics, University of Alaska, Fairbanks, AK 99775 New Apatite Fission Track Analysis (AFTA) data from Devonian igneous rocks and Mesozoic and Tertiary sedimentary rocks exposed in the northeastern Brooks Range (NEBR) of northern Alaska help constrain the regional tectonic and thermal history of the NEBR. Data are available from four new locations; (A) the Okpilak Batholith, (B) the Jago River section east of the Sadlerochit Mountains, (C) the Marmot Syncline and, (D) the Sagwon Bluffs region. AITA data on rocks from a vertical section through the Devonian-aged Okpilak Batholith (n=17) indicate two distinct cooling events. The first resulted in rapid (<3-5 m.y.) cooling through 125°-60°C at -40-45 Ma. The presence of the top of a fossil annealing zone near the top of the vertical section indicates a maximum age of the second cooling at 30 Ma. AFTA data on sedimentary rocks from the Paleocene to Eocene Jago River Formation (n=5) indicate rapid cooling (<3-5 m.y.) at approximately 20 Ma. AFTA data on Albian-aged sedimentary rocks exposed at Marmot Syncline (n=9) indicate cooling at -4550 Ma. Data from Early Tertiary sedimentary rocks (n=4) exposed at Sagwon Bluffs along the Sadlerochit River show the region has not been heated above 60°C since deposition and individual grain ages reflect the thermal histories of the provenance terrains. These new data, combined with previous results showing major cooling due to uplift and erosion at 60 Ma at Bathtub Ridge and at 45 Ma along the Canning River west of the Sadlerochit Mountains, now defme three major phases of cooling due to uplift and erosion in northeastern Alaska at --60 Ma, -40-50 Ma, and -20 Ma.

74 HIGH-RESOLUTION 238u-206pb ZIRCON DATING OF TERTIARY BENTONITES AND FISH CANYON TUFF: A TEST FOR AGE "CONCORDANCE" BY SINGLE-CRYSTAL ANALYSIS OBERLI, F., FISCHER, H . and MEIER, M . , Isotope Geochemistry, ETH, CH-8092 Zurich Zircon populations of pyroclastic rocks frequently contain inherited components, e.g., old nuclei derived from assimilated crust or exotic grains admixed by detrital or explosive processes. Conventional U-Pb isotope measurements on young zircons typically yield linear or semi-linear, reversely discordant data patterns in 207pb/235u-206pb/238u space, depending on the population's complexity and smoothening of sample heterogeneities by integrating multi-grain analysis. Precision and accuracy of linear-fit derived discordia/ concordia intercept ages are adversely affected by low-angle intercept relationships, especially, if the inherited components have Paleozoic or younger ages. The ages then become rather sensitive to deviations from the ideal (2-component) model, to analytical errors and to radioactive disequilibrium. Grain-bygrain analysis, however, in combination with the high time resolution provided by the 238u-206pb system of zircon, permits to select "concordant" crystals devoid of inheritance and to reject data influenced by old components. This is demonstrated by results from late Eocene to early Oligocene Tethyan bentonites (N Italy) and from Miocene Molasse bentonites (N Switzerland). Application of this technique to zircons from the Fish Canyon Tuff (Colorado) yields a disequilibriumcorrected mean age of 28.41±0.05 Ma (95 % c.l. of precision; estimated accuracy 0.1 Ma), discrepant from published, equally precise and reproducible sanidine and biotite 40Ar-j9Ar ages of 27.8 Ma. Apart from the possibility that this age difference may be related to geological phenomena (e.g., closure of zircon prior to eruption), the result also rises questions with respect to comparability of U-Pb and 40Ar-39Ar ages and associated calibration problems.

PHANEROZOIC UPLIFT HISTORY OF THE WESTERN MARGIN OF THE GULF OF SUEZ RIFT, EGYPT O M A R , G.I, L U T Z , T.M., G e o l o g y D e p t . U n i v . of Penn., Phila., PA 1 9 1 0 4 , U.S.A., STECKLER, M.S., L-DGO of Columbia Univ., P a l i s a d e s , N.Y. 1 0 9 6 4 , U.S.A., K O H N , B.P., Geology Dept. Ben Gurion Univ., Beer Sheva 84105, Israel. Fifty eight apatite fission-track (FT) ages and corresponding Horizontal Confined Track Length Distributions (HCTLDs)reveal that fault-bounded basement blocks, constituting the rift shoulder of the Gulf of Suez, were affected by several uplift events during the Phanerozoic. Three groups of apatite FT ages have been identified on apatite age/mean l e n g t h p l o t and by the u s e of f o r w a r d m o d e l l i n g . In every group HCTLDs change progressively from unimodal and positively skewed to broad b i m o d a l to unimodal narrow and negatively skewed. This age-HCTLDs pattern in each group is consistent with samples collected from a rock column that r e s i d e d for a long p e r i o d of t i m e at temperatures between 50 and 110°C and that was subsequently exposed at the surface during the Red Sea major rift-flank uplift at 21-23 Ma ago. Correcting the oldest age in the three groups for annealing experienced at temperatures of >20°C yields ages of 500, 370 and 260 M a . These ages are interpreted as the times when different sets of blocks w e r e uplifted from temperatures of about 110 to about 50°C during a Paleozoic uplift event. This event was followed by an Early C r e t a c e o u s , Late Cretaceous and Early Miocene-Present uplift events.

ICOG 7

ABSTRACTS

IMPLICATIONS OF THE LASER MICROPROBE FOR MODELS OF ARGON TRANSPORT IN MINERALS ONSTOTT.T.C. and LEE, J.K.W., Dept. of Geological and Geophysical Sciences, Princeton University, Princeton, NJ 08544, U.S.A., and PHILLIPS, D., AngloAmerican Research Laboratories, Johannesburg, Rep. of South Africa In recent years the laser microprobe has revealed the existence of apparent age zonations within single grains of hornblende, biotite, and muscovite. These results are surprising considering that the size of the ablation pits made by lasers, which ^ i c a l l y vary from lO's of microns up to 100 microns in diameter, approach the dimensions of diffusion radii for these minerals as inferred from hydrothermal bomb experiments. In some cases, a CI zonation, as measured by ^^Ar, also occurs with the apparent age zonation. This CI zonation has been confirmed by electron microprobe measurements and does not appear to be an artifact of the laser microprobe technique. In those cases where the thermal history is constrained and the samples have been characterized by electron microprobe and TEM observations, the age and CI zonation is consistent with a diffusion process. The irregularity and non-Fickian shape of the profiles and the existence of submicron to micron size dislocation structures in many instances suggest to us that a significant component of pipe diffusion occurs along these structures. Depending upon the dislocation density, the mineral-liquid partition coefficient for Ar, and the magnitude of the pipe diffusion coefficient relative to the volume diffusion component, volume diffusion will tend to be the dominant mechanism of exchange for small grain sizes. With increasing grain size, pipe diffusion will be the dominant mechanism for low fractional exchange. This could explain the grain size variations in bulk diffusivity of Ar observed in the hydrothermal bomb experiments on biotite and hornblende. TTieoretical closure temperatures will increase with grain size up to a maximum radius where it levels off. In slowly cooled terrains, therefore, the cooling ages should vary with the grain size for grain sizes below this maximum radius.

ISOTOPE ABUNDANCE ANOMALIES IN HEAVY ELEMENTS OTT, Ulrich, Max-Planck-Institut fiir Chemie, D-6500 Mainz, F.R.G. The first isotopic anomaly detected in meteorites was in the rare gas xenon. Then it was found that not only was the isotopic composition of Xe in carbonaceous meteorites different from that of terrestrial atmospheric xenon but that there also exists a component enriched in the light and the heavy isotopes now referred to as HL-Xe and carried by tiny 0.003 /xm) diamonds. The light and heavy branches of HL-Xe resemble Xe producedln the p- and r-processes of nucleosynthesis, resp., but their exact production mechanisms remain to be identified in detail. We have, however, a clear understanding of the other type of isotopic anomaly first discovered in Xe also: s-process Xe thought to be sited in rare (few ppm) SiC. Associated effects have been detected in Kr, Ba, Sr and (possibly) Rb. These results allow us to refine our ideas about the slow neutron capture process of nucleosynthesis. They also show that the s-process that produced these trace elements was different from the average s-process that contributed the sprocess nuclides to the solar system abundances. The composition of s-process Ba indicates that the seed nuclei were exposed to a total neutron dose lower than inferred from solar system abundances, while s-Kr is not of a uniform composition and indicates that production occurred under variable conditions of neutron density. To produce the observed s-Sr-, s-Rb, and average s-Kr compositions, the average neutron density must have been higher and the temperature lower than inferred from solar system abundances. Apart from the anomalies that are most prominent in the noble gases there are others. These include Ag in iron meteorites, Sr and Ba in FUN Ca-Al-rich inclusions in Allende and Hg in a variety of meteorites.

ABSTRACTS

ICOG 7

MANTLE DEGASSING: NOBLE GAS CONSTRAINTS OZIMA Minoru, Geophysical Institute, University of Tokyo, Tokyo 113, Japan. Recent precise isotopic measurements of noble gases in diamonds have enabled us to infer the noble gas state in the ancient mantle, which together with the noble gas data in MORBs representing the modern mantle state imposes crucial constraints on mantle-atmosphere evolution models. The major constraints thus obtained are: (1) Primordial He nnd Ne were solar-like, but not atmospheric nor meteoritic. (2) There are significant excesses in ^^^Xe and in ^^i-ise^g jj^ modern and ancient mantle. (3) Apart from the excesses, Xe isotopic compositions are similar between the mantle and the atmosphere. The implications from these constraints are: (a) The present atmospheric Ne was formed by fractionation of solar-like primordial Ne in the atmosphere, (b) The excess Xe must have been established very early in the Earth's history. This suggests as the progenitor for the excess ^^^Xt as generally accepted, but favour ^^^Pu over ^^^U for the origin of the excesses in Recent experimental results on noble gas distribution between crystal and melt indicate that contrary to the commonly held view magma transport processes cannot be responsible for the major mantle degassing. Present noble gas data favour impact degassing as the major mantle degassing process.

CRUSTAL ORIGIN OF CARBONADO (DIAMOND): NOBLE GAS EVIDENCE OZIMA,M. ZASHU,S., Geophysical Institute, Univ.ol Tokyo, Tokyo, 113, and TOMURA,K., Nuclear Energy Institute, Rikkyo University, Yokohama 240-01, Japan Carbonados are aggregates of raicrocrystals of diamonds, and show various features typical of the crust such as lighter carbon isotopes, inclusion of crustal minerals, and non- association with kimberlites. However,the origin of carbonado is still controversial. We investigated noble gas isotopic compositions in a carbonado from Poxoreu, Brazil. The carbonado we studied is about 1 ca., very porous and dark. Xe, Kr and Ne isotopic compositions are totally different from the atmospheric ones. Xe and Kr are essentially derived from spontaneous fission of Ne contains large fraction of nucleogenic components from and He and AT are highly radiogenie. U content measured on the graphite residue after the noble gas extraction gives 2.3 ppm. If this U content represents the carbonado, the fission Xe and Kr are essentially parentless, or they must have been imported. Since the range for fission Xe and Kr is about 20/i , the carbonado must have formed in direct contact with uranium minerals to accomodate the observed amount of fission Xe and Kr. Therefore, the existence of parentless fission Xe and Kr appears to favour the speculation by Kaminskii(l) that crustal carbon was converted to diamond (carbonado) under the influence of radiation from U in uranium minerals. (1) Kaminskii,F.V.(1987) Doklady Academy Nayk.USSR, Tom294, 439

THE BROKEN HILL OREBODY, AUSTRALIA: DEPOSITIONAL AGE AND TIMING OF HIGH-GRADE METAMORPHISM PAGE. R.W.. Bureau of Mineral Resources, Canberra, A.C.T., Australia, LAING,W.P., James Cook University of North Queensland, Townsville, Queensland, and STEVENS, B.P.J., Geological Survey of NSW, Broken Hill, NSW, Australia. Despite numerous geochronological studies over the past 3 decades, the primary age and correlation of Broken Hill Group supracrustals enclosing the 300 Mt Ag-Pb-Zn orebody are not known, and the depositional origin of the orebody's immediate host rocks remains a matter of much contention. This is largely because past studies have concentrated on high-grade rocks near to Broken Hill, where intense deformation and granulite-facies recrystallisation mask field relations as well as textural features, and primary isotopic signatures are generally modified or obliterated. This study focuses primarily on rocks of lower metamorphic grade, two felsic pyroclastic units, which are stratigraphically equivalent to the ore at Ae top and middle of the Broken Hill Group. Conventional, multi-grain U-Pb analyses of euhedral-subhedral slender zircons are discordant and non-linear on the concordia diagram, precluding precise age interpretation. Ion microprobe analyses on SHRIMP enable resolution of this complex pattern. We identify a major magmatic population at 16901:5 Ma, minor xenocryst groups at 1780 Ma and 1860 Ma, and fewer older xenocrysts in the range 2200 to 2700 Ma. The 1690±5 Ma age for igneous cr>'stallisation determines the depositional age of this part of the Broken Hill Group. This provides a firm chronological link between the early Proterozoic sequences enclosing major stratiform Ag-Pb-Zn orebodies at Broken Hill, Mount Isa (1670±18 Ma) and McArthur River (1690±27 Ma). The inherited xenocryst components, commonly mantled by 1690 Ma zircon overgrowths, reflect complex source(s) consistent with Nd model ages of 2200-2300 Ma. Stratigraphically ^uivalent gneisses in the granulite-facies terrane nearer to Broken Hill, have U-Pb zircon age patterns similar to the lower grade rocks, but complicated by new zircon growth and recrystallisation during the high-grade metamorphism at 1600 Ma. Consistent U-Pb monazite ages are also close to 1600 Ma, suggesting that their blocking temperature is close to that of zircon, or that cooling to < 600°C was achieved relatively rapidly.

THE V6 SECTOR 54 WITH 30CM ELECTROSTATIC FILTER FOR HIGH ABUNDANCE SENSITIVITY MEASUREMENTS OF VERY LOW ABUNDANCE ISOTOPES PALACZ, Z., WALDER, A., FREEDMAN P.A., VG ISOTECH Ltd, Middlewich, Cheshire CWIO OHT, U.K. A 30cm radius electrostatic energy filter (ESF) has been used to improve the abundance sensitivity of a standard 54cm radius magnetic sector instrument. The filter removes low energy ions produced by inelastic collisions in the vacuum analyser by passing the 8KV ion beam between two plates biased at +/- 400V. Transmission through the filter is virtually 100%, flat topped peak shapes are retained, and the stability of the gain between a post filter ion counting system and pre-filter Faraday cup is about 0.1% over 1 hour. Abundance sensitivity-the contribution of the major isotope on the baseline beneath the minor is improved from 2ppm without the filter to O.lppm at 237 with respect to mass 238. By biasing the voltages on the filter it is possible to further imprgrve this figure to 0.05 ppm, by increasing the resolution to 1200. At 0.1 ppm abundance sensitivity there is no evidence of a 232-^^ tail beneath the peak, Thorium isotope ratio data on known pC counting standards are presented.

ICOG 7

ABSTRACTS

OXYGEN ISOTOPE CONSTRAINTS ON ADDITION OF MANTLE-DERIVED COMPONENTS TO CONTINENTAL CRUST IN THE PHANEROZOIC PALIN, J. Michael, Dept. of Geology and Geophysics, Yale Univ., P.O. Box 6666, New Haven, CT 06511, USA Convergent margin complexes (CMCs) represent large geochemical reactors in which mantle-derived components are mixed with exogenous components consisting of biogenic, authigenic, weathered, and eroded materials (10<5l8o<35). These components are initially far from ISQ/^^q equilibrium and undergo oxygen isotope exchange during metamorphism. The length and time scales of exchange depend on the physical and chemical character of the materials, the presence and movement of fluids, and the stress-temperature history. Mantle-derived components become l^O-enriched and exogenous components become l80-depleted by this process. Metamorphic cores of ancient CMCs indicate that most oxygen isotope exchange occurs under low-grade, fluidrich conditions prior to the attainment of peak metamorphic conditions. Apparent decreases in bulk with increasing metamorphic grade in many of these terranes indicates either (1) exchange with surfacederived fluids at high temperatures or (2) systematic tectonostratigraphic variations in the ratio of mantlederived to exogenous components. Relations among O, Sr, Nd, and Pb isotopes within individual terranes (and in associated granitoids) are consistent with the latter hypothesis. The relatively constant S^^q of Phanerozoic clastic sediments and seawater indicates that no net addition of mantle-derived components to the continental crust has occurred over the last 600 Ma.

Rb-Sr, Pb-Pb and Sm-Nd GEOCHRONOLOGY OF THE HIGH GRADE GRANULITE GNEISS TERRAIN OF RAMACHANDRAPURAM AREA, DISTRICT KAMARA3, TAMILNADU, SOUTH INDIA. B.K. PANDEY, K. RAMESH KUMAR, VEENA KRISHNA, R. DHANA RA3U AND J.N. GUPTA. Atomic Minerals Dvision, Department of Atomic Energy, Hyderabad 500 016 (INDIA). The bio-mu-ga granites, charnockites and hbl-bio-ga quartzo-felspathic gneisses from Ramachandrapuram area (9°18'-9°32'N,77°29'-77° 42'E) yield Rb-Sr isochron ages of 1137 ± 17 Ma (IR = 0.7056 ± 3), 1127 ± 23 Ma (IR=0.7052 ± 2) and 1031 ± 23 Ma (IR=0.7085 ± 2) respectively. Their Pb-Pb data define linear arrays corresponding to ages of 3099 ± 137 Ma, 2825 ± 282 Ma and 1731 ± 2^3 Ma respectively (modelMi= 8.^-8.5). Whereas the granites yi-eld Sm-Nd isochron age of 2063 ± 138 Ma (^Jfcp 6.3 ± 2.0), the charnockites and gneisses show scatter and generally fall above the isochron for the granites. The Rb-Sr isochron ages of 1.1 Ga are interpreted as the time of high grade granulite facies metamorphic event. The Sm-Nd date of 2.1 Ga records the most probable time of the generation of the precursor crystal segment from a depleted mantle reservoir. The scatter of Sm-Nd data for the charnockites and the gneisses could be due to disturbances (mostly decrease) in the Sm/Nd ratios of these rocks during the 1.1 Ga event. The linear arrays defined by the Pb-Pb data, which yield ages older than even the Sm-Nd date, appear to be the spurious 'transposed palaeoisochrons' possibly caused by the incomplete homogenisation of Pb isotopes accompanied by severe loss of U during the 1.1 Ga metamorphism. The presence of 1.1 Ga high grade metamorphic event which is broadly correlatable to the high grade metamorphic event in Eastern Ghat Belt (INDIA) and Sri Lanka, has been recorded for the first time in Southern Indian granulite gneiss terrain.

CONTROLS OVER SEAWATER Sr ISOTOPE RATIOS LN THE CENOZOIC PALMER, M.R., Department of Geology, Bristol University, Bristol, U.K., and EDMOND, J.M., Department of Earth, Atmospheric and Planetary Sciences, Massachusetts Institute of Technology, Cambridge, Massachusetts, U.S.A. Recent advances in our understanding of the controls over the Sr isotope geochemistry of submarine hydrothermal fluids and river waters required a revision of the values used for these fluxes of Sr isotopes to the oceans (Palmer & Edmond, 1989). This, in turn, required that we take a revised look at the Sr isotope evolution of the oceans. Models have been developed that incorporate; recycling of marine Sr through carbonate-rich sediment porewaters and uplifted limestones, decreases in hydrothermal fluxes resulting from a decrease in mid-ocean ridge crustal production rates, and the influence of the Ganges-Brahmaputra river system that developed following the collision of the Indian subcontinent with Asia. These models are based on documented geologic events and indicate that the ^'^Sr/^^Sr ratio of the oceans since 65 Ma is characterised by gradual increase due to a fall in oceanic crustal production rates, upon which is superimposed a more rapid increase since 40 Ma related to the uplift of the Himalayas and the development of the Ganges-Brahmaputra river system. This river system has a high dissolved load because of the rapid uplift rate of the Himalayas and the high rainfall in the area. Its Sr isotope composition is high because of large exposures of radiogenic granites within its catchment The observation that individual river systems can have a major impact on the Sr isotope evolution of the oceans as the result of regional climatic, geologic and tectonic factors places severe limits on our ability to interpret the Sr isotope evolution of the oceans in the more distant past. M.R. Palmer & J.M. Edmond (1989) EPSL 92, 11-26.

PROTEROZOIC CRUSTAL GROWTH AND ITS MODERN ANALOG IN THE CANADIAN CORDILLERA PATCHETT. P.J.. Dept. of Geosciences, University of Arizona, Tucson, AZ 85721, USA, and SAMSON, S.D., Dept. of Geology, Syracuse University, Syracuse, NY 13244, USA In Proterozoic time (2.5-0.6 Ga), the crust of several studied regions grew rapidly. Examples are the 1.9-1.7 Ga terranes of Europe, Greenland and North America, the 1.6(?)-1.0 Ga Grenville orogenic system, the Australian Proterozoic and the 0.9-0.7 Ga crust of Egypt-Saudi Arabia. Persistent features of these Proterozoic crustal additions are: (1) high abundance of "arc-like" volcanic/ plutonic complexes of calc-alkaline affinity; (2) end values in igneous rocks between -2 and +8 at time of activity, suggesting a predominantly juvenile mantle origin; (3) generation of large amounts of juvenile continental crust in restrict^ regions and time periods. The crust of 1.9-1.7 Ga age in Nonh America, Greenland and Europe, for example, would represent 110% of today's plausible island-arc generation rate, and over 70% of any plausible Proterozoic rate. It seems to be required that at several times in the Proterozoic, the plate-tectonic regime was capable of accreting large proportions of the globally-produced arc complexes into small regions. Our, and other, studies in the Canadian Cordillera show that this region is a near-perfect modem analog for Proterozoic crustal growth. The Canadian Cordillera is unusual among Phanerozoic orogens in showing (1) pro^essive Mesozoic and Cenozoic accretion of discrete tectonostratigraphic terranes, with internal histories extending back to early Paleozoic; (2) evidence that large volumes of terranes represent crust of broadly arc-like character produced from the mantle during Phanerozoic time. An additional much-debated feature of the Cordillera is northward translation of terranes along the North American margin, due to the nonhward component of Pacific Plate motion. Translation of accreted arc complexes towards a single zone may also have been a feature of major Proterozoic crustal accretion events. It is clear that an eventual continental collision in the Canadian Cordillera, with attendant metamorphism and melting, will result in a belt of rocks lithologically, geochemically and isotopically very similar to juvenile Proterozoic belts.

ABSTRACTS

ICOG 7

CONTAMINATION OF OCEANIC BASALTS WITH ATMOSPHERE DERIVED NOBLE GASES PATTERSON. D., HONDA, M. and McDOUGALL, L, Research School of Earth Sciences, The Australian National University, Canberra, ACT 2601, Australia. The incorporation of a component of atmospheric noble gases in mantle derived samples, such as dredged glassy basalts from mid-oceanic ridges (MORBs) and intraplate oceanic islands (OIBsX is of critical importance to noble gas studies of mantle geochemistry. In extreme cases atmospheric contamination may completely obscure any mantie noble gas signal that may have been present. Models for the geochemical structure and evolution of the Earth's mantle-crust-atmosphere system may be constrained by noble gas data only after identifying and correcting for secondary atmospheric components within the samples. Noble gas data from volcanic glasses dredged from Loihi Seamount, Hawaii, provide an excellent example of contamination of basaltic samples with atmosphere derived noble gases. The most likely source for the atmospheric component is interaction of the Loihi magma with seawater carr>'ing dissolved atmosphere-derived noble gases. Simple models of endmember mixing between present day atmosphere and a nonatmospheric mantle derived component can account for the observed variations in noble gas isotopic ratios in the Loihi sample suite. The elemental noble gas ratios ÔNe/^Âr, ^KifiÂi and ^^^XtfiÂi show more complex variations, which can be modelled as the result of contamination coupled with solubility controlled elemental fractionation between melt and volatile phases. Such solubility controlled fractionation may occur during equilibration of the melt with ambient volatiles, or during bubble formation and gas loss processes. The noble gases are very sensitive geochemical tracers for detecting interaction of oceanic basalts and seawater. Massive contamination effects can be observed in the noble gases without causing detectable effects in other geochemical tracers such as Sr isotopic and Th activity ratios.

5I80 IN DISSOLVED PHOSPHATE PAYTAN, Adina, LUZ, Boaz, and KOLODNY. Yehoshua, Department of Geology, The Hebrew University of Jerusalem, Jerusalem, Israel. Isotopic exchange between oxygen of dissolved phosphate and water is extremely slow in inorganic reactions and rapid in the presence of enzymes, i.e. when phosphate is biologically processed. Isotopic analysis of oxygen in dissolved phosphate may hence serve as tracer of the efficiency of biological cycling of the phosphate. Dissolved phosphate was extracted by passing large volumes of water through columns of chitosan-ammonium-molybdate (CMA). CMA was prepared by coating freshly formed chitosan beads with ammonium molybdate. Recovery yields of phosphate from seawater average 70%. No isotopic fractionation of oxygen has been observed in the extraction process. Isotopic exchange in the system dissolved phosphate-waterbiomass was traced by using ISQ enriched water (30%c) and a phosphate fertilizer of known isotopic composition (6%o) of its oxygen (6p). When an isotopically "light" (6%o) phosphate fertilizer is added to growth ponds, the apparent 6p of soft tissues approaches rapidly a value determined solely by B^^O of the pond water and is not influenced by 5p of the fertilizer. The biomass is about 19 to 23%o enriched in 18O with respect to the pond water. Growth and decomposition experiments of algae at different temperatures (between 0 and 35oC) show that 5p in algal organic matter and in P04-^ released to sea water upon its decomposition, is not a function of temperature, in sharp differencefrom5p of skeletal apatite in fish. 5p of dissolved phosphate in the ocean varies between 19.7 and 20A%o (Longinelli et al., 1976, our results). These values suggest that it is determined by cycling through the oceanic biomass. Such cycling is apparentiy too rapid to permit tracing of its efficiency. In Lake Kinnereth (N. Israel) phosphate is supplied into the lake by the Jordan from two sources: a^ basaltic apatite with 5p « 5-10%o, and b. sedimentary apatite with 5p « 2Q%o. The changing 5p in the lake sediments is a reflection both of mixing of these sources and increasing equilibration fromtiieJordan entrance southwards.

A STUDY OF FISSION TRACKS IN APATITE USING HIGHRESOLUTION TRANSMISSION ELECTRON MICROSCOPY:

WHAT CAN WE LEARN FROM VARIOUS CHRONOLOGIES ON SIZES, INTERNAL STRUCTURES, AND METAMORPHISM OF CHONDRITES ? PELLAS,P., Museum,61 rue Buffon,75005 Paris,France

A METHOD FOR THE IN-SITU ANNEALING.

OBSERVATION

PAUL, Tracy A., Department of Geology, Arizona State University, Tempe, Arizona, USA 85287-1404. Electron bombardment, such as that which occurs in a transmission electron microscope (TEM), can alter or destroy crystalline materials and microstructures, including fission tracks. The addition of cold-stage and anticontamination instruments to a JEOL JEM2000FX analytical TEM have facilitated obser\'ation of beamsensitive fission tracks in apatite. By adjusting the heating current to the cold stage, the sample temperature is varied from -170 to 4-10°C and thereby track annealing induced for in-situ observation. Spontaneous and induced fission tracks in apatite cr>'stals from Durango, Mexico have been examined using the TEM. Observations of induced fission tracks in laboratory-annealed samples indicate that near the atomic level, fission tracks are not perfectiy linear features. Many tracks are not continuous along their length, but are comprised of subparallel segments. Although track orientation is random, width appears to be crystallographically controlled, with tracks parallel to [0001] displaying the greatest widths (-40 A). Tracks perpendicular to [0001] display widths of -25 A. The time over which fission tracks in Durango apatite are stable in the microscope is dependent on the current density on the specimen and the temperature of the specimen. Current densities > 90 pA/cm^ result in radiolytic degradation of the apatite structure. Damage is accelerated at temperatures above LN2 (-178°C). Fission tracks appear to be more resistent to radiolytic damage than the host apatite. However, destruction of the host material negates imaging of the tracks. At LN2 temperature and -15 pA/cm^ current density, tracks can be viewed for 20 to 70 minutes. At this current density, track annealing commences at ~-60°C. Presently, the restoration process cannot be viewed when the sample temperature is above +10°C due to electron-beam damage of the host. The rate at which individual tracks anneal is variable and may reflect differences in track orientation, grain composition, and degree of partial annealing.

Search for coherences between age data obtained on meteorites through absolute and relative chronologies has been a major challenge during the past thirtyyears with very little success yet. Kith a larger number of selected samples analyzed, time has come to describe the landscape that is emerging. 1) Among the very short-lived radionuclides it does not appear that 1129, A126, Mn53 or Pd 107 correlate in any satisfactory way with absolute chronologies. On the other hand, recent results based on SmlA6 (Tl/2 : 103 Ma) seem to have an exciting prospect to describe early evolution of fractionated meteorite (1). New data from Pu244 fission xenon and fission tracks in phosphates and their adjacent silicate phases fairly well describe the length of metamorphism in the various equilibrated layers (petrologic types 4,5,6) of H and LL OC asteroids. The results indicate onion-like internal structures for both asteroids, with a larger size for the LL-body (250-300kra in diameter) than for the H-body (<150km). 2) Recent Pb-Pb/U-Pb ages of phosphates from several chondrites (2), along with many Ar-Ar ages obtained on bulk samples of the same objects, have made it possible to estimate absolute lengths of metamorphism as defined by the interval between the closures of UPb and K-Ar systems. These metamorphism intervals are in good agreement with those defined by Pu-Xe/fission tracks intervals. It remains to verify that initial Sr87/Sr86 values of phosphates from those chondrites do also allow to define metamorphism intervals in agreement with those defined above. This depends, however. on the Rb/Sr ratio of phosphates environments that have to be ascertained. References: 1) Prinzhofer et al. (1990) LPS 21, 981. 2) Gopel et al. (1989) Meteoritics 24, 270.

STRONTIUM, OXYGEN AND BORON ISOTOPES AS TRACERS OF MAGMATIC EVOLUTION AND CRUSTAL ASSDvlILATION AT MID-OCEAN RIDGES PERFIT. M R Pepart ment of Geology, University of Florida, Gainesville FL 32611; JONASSON, IR and GREGOIRE, D C (Geological Survey of Canada, Ottawa, KIA 0E8, Canada) Comprehensive investigations of B and Sr concentrations and B, Sr and 0 isotopes have been completed on a cogenetic suite of mid-ocean ridge glasses and altered lavas from the eastern Galapagos Rift. In fresh glasses, B concentrations range from --0.5 ppm in MORB to 9.1 ppm in andesite and show systematic increases with other highly incompatible elements (Zr, Y, Ce); except in some andesites with B values around 5 ppm. In contrast, B in altered samples range from 5 ppm to 30 ppm and reach 124 ppm in silica-oxide hydrothermal deposits. Poor correlations exist between B and volatile (e.g. CI) concentrations, and there is a general decrease in B in the most highly altered samples; those with the lowest Sr and highest ^^Sr/^^Sr. Boron dB=-48'U.7) isotope ratios O^BI^^B) range from 3.85 to 4.06 in glasses and from 3.97 to 4.10 in altered lavas. The lowest values agree with those previously reported in meteorites and peridotites. Increases in ^^B/^^B in more fractionated glasses correlate with increased Sr and B contents and slight, but significant, shifts in ^^Sr/^^Sr (.70251-.70272) and ^^SO (6.2-5.5) Slightly to moderately altered rocks have ^^B/^^B values similar to the more evolved glasses but higher B contents and elevated ^'^Srl^^Sr and ^^B/^^B values that approach seawater (NASS = 4.18). Calculations indicate the isotopes and elemental variations obser\'ed in the glasses are a result of assimilationfractional crystallization (AFC) and minor mixing in shallow level magma chambers. Fractional crystallization of parental MORB may have exceeded 80%, and the mass of assimilated altered crust is limited to <15%. Mixing between primitive unaltered magmas and evolved contaminated magmas may explain some of the overenrichments in incompatible elements. Boron appears to be a very sensitive indicator of contamination in subaxial magma chambers and the hydrothermal alteration of oceanic crust.

40Ar/39Ar AND U-Pb DATING OF THE GOONUMBLA PORPHYRY COPPER-GOLD DEPOSITS, NSW, AUSTRALIA CAROLINE PERKINS, Geology Department and Research School of Earth Sciences, The Australian National University, G.P.O. Box 4, Canberra, ACT 2601, Australia, IAN McDOUGALL and JON CLAOUE-LONG, Research School of Earth Sciences, The Australian National University, G.P.O. Box 4, Canberra, ACT 2601, Australia High precision dating of alteration minerals and host rocks to mineral deposits provides information important for the understanding of ore genesis. The ^ÂrftÂi technique of K-Ar dating has considerable potential in determining the age of hydrothermal phases associated with mineralization, and U-Pb dating of magmatic zircons provides a means by which the emplacement time of propylitically altered host magmatic rocks to deposits may be established. The geochronology of the relatively undeformed Goonumbla porphyry copper-gold deposits near Parkes in the Lachlan Fold Belt, NSW, Australia, has been investigated using ^ÂiflÂi and U-Pb dating. The Goonumbla deposits are hosted in quartz monzonite pipes which intrude a trachyte-trachyandesite volcanic succession. The ^Âr/^Âr age spectra from Goonumbla hydrothermal sericite indicates that vein and groundmass (potassic and phyllic) alteration were broadly contemporaneous at 439.2 ± 1.2 Ma (la). Sericite has undergone partial argon loss as the result of emplacement of nearby granitiods between about 370-420 Ma. The U-Pb dating of magmatic zircons by the SHRIMP ion microprobe yields an age of 438.5 ± 3.6 Ma (lo) for the Nelungaloo Volcanics which underlie the mineralized Goonumbla sequence, and suggests that the two successions are largely coeval. The ^0Ar/39Ar age spectra obtained in this investigation from fine grained sericite demonstrates that there is considerable potential for application of this dating method to Phanerozoic metallogenic deposits. The close agreement in the ages determined by the 40Ar/39Ar and U-Pb methods gives confidence in application of both dating techniques to altered Palaeozoic rocks.

ICOG 7

ABSTRACTS

SPINEL-PERIDOTITE XENOLITHS FROM EASTERN PARAGUAY: Sr-Nd ISOTOPIC COMPOSITIONS B E A R I N G ON S U B - C O N T I N E N T A L MANTLE E V O L U T I O N PETRINI, R., Istituto di Geocronologia e Geochimica Isotopica, CNR, Pisa, Italy; CIVETTA, L., Dipariimento di Geofisica e Vulcanologia, Napoli, Italy; COMIN-CHIARAMONTI, P., Istituto di Mineralogia, Petrografia e Geochimica, Palermo, Italy; DEMARCHI, G. and PICCIRILLO, E.M., Istituto di Mineralogia e Petrografia, Trieste, Italy.

Protogranular spinel-lherzolites to dunites occur as xenoliths in nephelinites at Nemby Hill (eastern Paraguay). Relict to well-preserved clinopyroxenes may show spinel exsolutions a n d / o r reveal spongy texture including glassy rods. Orthopyroxenes show spinel exsolutions, while spongy texture is rare and confined to crystal boundar}'. All the xenoliths are enriched in incompatible elements, particularly those characterized by "blebs", probably related to original hydrous phase(s). Sr-Nd isotope ratios have been measured on whole rock, clinopyroxene and orthopyroxene. Present-day SrNd isotopic compositions on whole rocks are in the ranges 0.7040-0.7048 and 0.5125-0.5133 respectively. In general, the j-^tio decreases as increasing the refractoiy character of the xenoliths. This cannot be simply ascribed to different degrees of melting, but suggests the intervention of metasomatic processes. The Nd isotopic compositions in clinopyroxene range between 0.51267-0.51348, and are not linearly correlated with the corresponding I'^'^Sm/^'^Nd ratios. On the contrary, most whole rocks and orthopyroxene define a linear correlation which if interpreted as an isochron would yield an age of c. 500 Ma (Brazilian Cycle). This age possibly dates metasomatic processes which affected isotopically depleted mantle source(s) relative to the Bulk Earth. Clinopyroxene may show, even in the same xenolith, quite variable SIMS Sm/Nd ratios (max. 0.0570.605) believed related to variable degrees of incompatible-element enrichment during metasomatism. U-Pb AGES OF ZIRCONS FROM CONGLOMERATE CLASTS IN THE JACK HILLS METASEDIMENTARY BELT, YILGARN CRATON, WESTERN AUSTRALIA R.T.PIDGEON. S.A.WILDE, School of Physics and Geosciences, Curtin Uni. of Tech.,Perth, W.A.,6102, and W.COMPSTON, Research School of Earth Sciences, Australian National University, Canberra, A.C.T.,2601. Twelve percent of detrital zircons from a sample of metaconglomerate from the Jack Hills, located in the Archaean Narryer Gneiss Complex of the Yilgarn Craton of Western Australia, have U-Pb ages in excess of 4 Ga and one grain has a minimum age of 4,276±6 My (Compston and Pidgeon, Nature, 321, 766-9, 1986). The metaconglomerate consists of a sandy matrix supporting deformed quartzitic pebbles up to 4 cm in diameter. The analysed zircons were from a composite sample of pebbles and matrix and it was not clear whether the zircons were contained in the pebbles, the matrix or both. Further mapping of the Jack Hills Metasedimentary Belt has identified a persistant boulder bed which appeares to be an ancient river channel. Clasts, although flattened, are commonly in the order of 12 cm in length.and consist of recrystallised sandstone and some may be chert : no boulders of igneous rock were found. The purpose of this paper is to describe the U-Pb systems of zircons extracted from a number of individual sandstone clasts . The spectrum of zircons ages from individual boulders is very similar to that found in the original sample of pebble conglomerate. Within the limitations of the number of zircons measured it appears that about 10% of the zircon pains are >4Ga. No individual clast has been found which contain a population of zircons with an unusually high proportion of >4Ga grains. This suggests that the zircons were mixed before being incorporated in sandstone which was later eroded and fragmented to form boulders in river channels. Except for zircons with ages >4Ga , the ages of zircon from the clasts reflect those found in gneisses of the Nanyer Gneiss Complex suggesting that the gneissic rocks were the major source of the clastic deuitus deposited ca. 3.1 Ga ago.

ABSTRACTS

ICOG 7

Sr AND Nd ISOTOPIC EVIDENCE FOR OCCLUSION OF TWO MANTLE COMPONENTS IN CUBIC DIAMONDS FROM ZAIRE PIER. J., PODOSEK, F. and BRANNON, J., Dept. of Earth and Planetary Sciences, Washington University, St. Louis, MO 63130, USA, and OZIMA, M. Geophysical Institute, University of Tokyo, Tokyo 113, Japan. Sr and Nd isotopic compositions and Rb, Sr, Sm, Nd, Ca, and K abundances have been measured for leaches of inclusions in two cubic diamonds (DZ89 and DZ90) from Zaire. The diamonds were cleaned in HF, crushed, and successively leached in H2O (DZ90 only), HCl, and HF. The elemental data suggest that the diamonds did not grow in identical environments. Although both have similar, low Sm/Nd ratios (0.13-0.15), DZ90 has substantially lower Rb/Sr and K/Ca ratios than DZ89. The isotopic data indicate that at least two distinct components were occluded during the growth of each diamond; however, it is unclear whether these components were fluids or partially crystalline at the time of trapping. Component 1, which is soluble in HCl (DZ89) and H2O (DZ90), has Sr and Nd isotopic signatures close to those of Bulk Earth, whereas component 2, which is soluble in HF and is thus expected to be a silicate phase, has slightly negative Epsilon Nd and unsupported radiogenic Sr (values range from 0.71465 to 0.71950). Relative to Ca, component 1 has lower abundances of K, Rb, Sr, Sm, and Nd than component 2. The HCl leach of DZ90 displays intermediate values suggesting that it represents a mixture of the two end-members. Within each diamond, the physical distribution of inclusions concentrating these two components is presently unknown. They may be systematically arranged, reflecting periods of growth in different reservoirs, or randomly distributed resulting from variable trapping of two immiscible or poorly mixed fluids. Rapid growth, as indicated by the cubic habit of the diamonds, may have been triggered by interaction between the two components. The time elapsed between growth of the diamonds and their transport to the surface, which is thought to have occurred during the late Cretaceous, has been a subject of debate. Relative to the evolution of Epsilon Nd in the depleted mantle reservoir, the maximum age for both diamonds appears to be 400 m.y. and may be appreciably younger. Growth immediately prior to eruption cannot be excluded.

Pb ISOTOPIC EVIDENCE OF GENETIC RELATIONSHIPS BETWEEN MAFIC AND ACIDIC HERCYNL\N MAGMATISM IN THE WESTERN SOUTHERN ALPS (NW ITALY) PINARELLI Laura. DELMORO Aldo, PARDINI Giancario, Istituto di Geocronologia e Geochimica Isotopica, Pisa, Italy, BORIANI Attilio, Dip.to di Scienze della Terra, Milano, Italy &VANNUCCI Riccardo, Dip.to di Scienze della Terra, Pavia, Italy The Ivrea Verbano Zone (IV) and the Serie dei Laghi (SL) are two sections of the lower and intermediate, respectively, continental cnist in the Southem Alps, tectonically juxtaposed. They were intruded, in Hercynian time, by mafic (IV) and acidic to intermediate-basic (SL) magmatites. Till now, the two magmatic series have been separately considered; however, recent geochemical and Sr isotopic studies suggest a genetic link between the two rock series. We present here new lead isotopic data on the late-Hercynian intnisives from the SL. The results can be summarized as follows: Granitic vlutons - (Montorfano, Baveno-Mottarone, Pella, Alzo-Roccapietra) - They contain typical crustal lead, with Mu values varying between 9.82 and 9.99, and show a Pb isotopic etherogeneity both among and within different bodies. Basic-intermediate intnisives - (Brissago, M.Zuccaro, Valsesia) They exhibit variable Mu values, ranging between 9.83 and 10.25. Their 206p|j/204pb and 207pb/204pb ratios are lower and higher, respectively, than those of the granites. In the 207pb/204pb vs. 206pb/204pb plot, samples coming from different outcrops arrange in two groups (Brissago + Valsesia and M.Zuccaro) showing a positive correlation between the two parameters, but with two different slopes. The crustal Pb isotopic signature of the examined plutons and dykes contrasts with their Sr isotopic ratios (^"^sf/sesr = 0.7051-0.7110), acsribed to the origin from an anomalous mantle source. On the other hand, the Mafic Complex from the IV contains crustal leads too, and the possibility that the primary magma for this complex derived from an enriched mantle source (high Mu and W values) has been considered. We suggest that the parental magma of the Hercynian intnisives from SL was generated in a similar way, and had the same Pb isotopic compositipn of the Mafic Complex from the Ivrea Zone.

THE FIRST 106-108 YEARS OF SOLAR SYSTEM HISTORY PODOSEK. F.A., McDonnell Center for the Space Sciences, Washington University, St Louis, MO 63130, USA The best estimate for the absolute age of the solar system is about 4.59 Ga; the absolute age is established with sufflcient precision and reliability to satisfy any probable need. There is also considerable interest in resolving a history of events in the solar nebula and in the early evolution of planetary bodies, on timescales ranging from less than 10^ yr to around 10^ yr. Our understanding of this early phase of solar system, however, is less satisfactory than the amount of available pertinent data might suggest, and some important questions and apparent contradictions remain unresolved. This presentation will review isotopic data relevant to early solar system chronology, including high precision absolute ages (especially U-Pb-Pb ages) and relative chronologies based on short-Hved radionucHdes (26Al, 53Mn, lO^pd, I29i) and on the evolution of initial ^^sr/sasr^ Overall, straightforward interpretation of isotopic chronometers leads to a picture of nebular history spanning a few Ma to about 10*7 yr, and to chondrite parent body thermal histories ranging to about 10^ yr. There are some cases, notably among igneous achondrites, for which different chronometers applied to the same objects lead to consistent results. Agreement of different chronometers is the exception rather than the rule, however. In some cases, different results from different chronometers can be understood readily in terms of their recording different isotopic closure events, but such an interpretation is not always easy. A problem of particular interest is that isotopic results suggest a duration of the solar nebula considerably longer than estimates based on non-isotopic (mechanical and hydrodynamic) considerations. This presentation will also review suggestions which have been advanced to (possibly) reconcile such discrepancies, including the hypothesis that short-lived radionuclides were very heterogeneously distributed in the solar nebula and reinterpretation of putative nebular events in terms of parent body processes.

Double Beta Decay of

A Geochemical Approach

H. Poths, University of New Mexico, Albuquerque, NM 87131, USA R. Rundberg, Los Alamos National Laboratory, Los Alamos, NM, 87545, USA Double beta decay is a second order weak process which converts two neutrons in a nucleus into two protons. The long lifetimes of nuclei undergoing this decay make direct observations of this effect very difficult. Different commonly considered double beta decay diagrams (e.g., emission of two neutrinos or emission of no neutrino at all) result in different calculated lifetimes of the nuclei and have implications for the nature of the neutrino. We decided to determine the amount of resulting from the decay of in a 2-billion year old molybdenite sample from Greenland. In a preliminary measurement we found that the extracted Ru was isotopically normal within 1 percent. With the measured amount of indigenous Ru in the molybdenite plus blank Ru from the extraction chemistry, we could estimate the lifetime of to be >1x10^' years. We hope that reduction of the blank and improvements in the determination of the isotopic composition of Ru can increase the limit for the lifetime of further.

80 NOBLE GASES IN SEDIMENTS ASSOCIATED WITH NATURAL GASES J. Poths*. B.M. Kennedy*, E.A. Bryant', J.H. Reynolds* Los Alamos National Laboratory, Los Alamos, NM, 87545, USA; *Dept. of Physics, Univ. of California, Berkeley, CA 94720, USA Recent investigations have found elevated ^He/He ratios in natural gases compared to that expected for such tectonically quiet sedimentary basins (1,2). Several, as yet speculative, possibilities exist for creating ^He "excesses" in this setting: input of mantle helium, input of cosmic dust, nuclear reactions on locally concentrated Li, or input of surface-exposed materials bearing cosmogenic noble gases. In an attempt to distinguish among the above possibilities, we have analyzed bulk sediment samples and the associated natural gas samples from 5 formations in 2 gas wells in the Hugoton/Panhandle gas field of Oklahoma and Texas. Noble gases contained in the natural gas samples exhibit isotopic effects comparable to samples from other gas fields ( 1 , 2 , 3 ) i ^ H e « O.2R3, ^^Ne excesses of 25-40% relative to atmospheric Ne, of -1200, and about 1 % excess at the fissiogenic isotopes of Xe. The bulk core samples also show these effects, but with factors of two spread in the isotopic ratios for samples from various formations within a given well. However, the concentrations of the noble gases in the solid samples appear to be orders of magnitude lower than needed to account for the large concentrations of noble gases associated with the natural gas. The sediments may have already released most of their noble gases, or the noble gases may have been scavenged from a much larger volume of rock, possibly during the migration of the natural gas to its trap. Refs:l)E.R. Oxburgh et al.. Nature 324, 632, 1986; 2) Hiyagon et al., Terra Cognita 6,104,1986; 3) Kennedy et al.. Earth Planet Sci. Lett, in press

ICOG 7

ABSTRACTS

STABLE ISOTOPES FROM D E S E R T SOIL CARBONATES IN PALEOECOLOGiC RECONSTRUCTON QUADE, Jay, CERLING, Thure E., and BOWMAN. John R.. Department of Geology and Geophysics, University of Utah, Salt Lake City, Utah 8 4 1 1 2 The ( P D B ) and E'^SO ( P D B ) content of pedogenic carbonate In Holocene age soils decrease with elevation increase in mountain ranges in the southern Great Basin, USA. At high elevation, the b ^ ^ C ( P D B ) of coexisting carbonates (> 3 0 cm soil depth) and soil organic matter differ by 1 4 to 1 5 permil, consistent with attainment of isotopic equilibrium between all oxidized carbon species derived from irreversible oxidation of organic matter prior to carbonate formation. Organic matter, and therefore coexisting soil carbonate, are depleted in ""^c, reflecting the dominance of C3 plants in the local biomass. At lower elevation, soil carbonates become progressively enriched in "'^c, largely because of an increasing proportion of C4 and CAM plants in local cover. Decreasing soil respiration rates, allowing penetration of atmospheric C O 2 deep into the soil, also contribute to this enrichment. Two important observations emerge: (1) soil caribonate underlying each of the main vegetation zones along the elevation transects is isotopically distinct, and (2) parent material has no apparent Influence on the carbon isotopic composition of soil carbonates. The S ' ^ ^ Q ( P D B ) of soil carbonate and rainfall both decrease with elevation. However, the average b ' ^ ^ O ( P D B ) value of deep ( > 5 0 cm) pedogenic cartx)nate is enriched about three permil relative to values we predict should be in equilibrium with the d ^ ^ O ( P D B ) of average local meteoric water. We used these isotopic patterns to resolve the origin of near surface carbonates that encrust fault fractures in Trench 14 near Yucca Mountain, Nevada, site of the proposed high level nuclear waste repository. It has been suggested that these carbonates were deposited by springs sometime In the Quaternary. However, actual spring carbonates that precipitated from the regional aquifer elsewhere are 3 to 7 permil more depleted in "•^O than the Trench 1 4 occurrences. The results from Trench 14 cart^onates are highly consistent with the b^^C (PDB) and 5''®0 (PDB) content of modern soil carbonates now found about 7 5 0 m above the site. Therefore, the Trench 1 4 carbonates we analysed are pedogenic in origin, but they probably formed in the presence of vegetation and rainfall found at the site during glacial maxima.

40Ar/39Ar DATING OF QUATERNARY PLAGIOCLASE: EXAMPLES FROM THE TAUPO VOLCANIC ZONE, N.Z. PRINGLE, Malcolm S.. U.S. Geoi. Survey, Menlo Park, CA, 94025, U.S.A.; Bruce HOUGHTON, New Zealand Geol. Survey, Rotorua, New Zealand; and Phil B. GANS, Dept. of Geophysics, Stanford Univ., Stanford, CA, 94305, U.S.A.

APPLICATION OF THE LASER MICROPROBE TO THE ANALYSIS OF O AND C ISOTOPIC COMPOSITIONS OF CALCITE CEMENTS IN RESERVOIR SANDSTONES Ame RAHEIM. Harald JOHANSEN, Derk S T U F H O O R N and Euan MEARNS, Institutt For Energiteknikk, P.O. Box 40, N-2007 Kjeller, Norway.

We have developed ^^Ar/^^Ar continuous laser and resistance furnace techniques which allow us to date 20 mg of Quaternarx plagioclase to better than 20,000 yr (1 sigma). Two advances have led to this development: (1) the new generation of ultraclean rare gas mass spectrometers, with typical blanks of 4*10-1® mol m/z«40, 2*10-17 mol m/z=39, and 1*10-17 mol m/z=36, and (2) the use of a continuous laser or resistance furnace to clean feldspar of most of its atmospheric contamination. With the continuous laser system, we defocus an argon laser to about 5 mm wide and degass up to 20 mg of feldspar at 4 watts. The degassed crystals are then fused with a focused laser beam in groups of about 5 mg. The major advantage of the laser extraction system is its negligible contribution to the total system blank. The major disadvantages are uneven heating of individual crystals during degassing and sample size limitations during fusion. With the resistance furnace system, we heat 20 mg of feldspar to 800-900°C to degass the crystals, and then perform a step heating experiment on the cleaned sample. The major advantage of the furnace system is the ability to evenly heat the crystals during the degassing step. The major disadvantage is the higher blank—about 2 times the laser system below 1000°C, while above 1000°C and ®®Ar increase exponentially in a near atmospheric ratio. Even so, the blank at 1400°C is still only 4*10'l^ mol. We illustrate the application of these techniques on plagioclase separated from ash flow tuffs of the Taupo Volcanic Zone (TVZ), New Zealand. Despite detailed geological mapping of this silicic caldera system which has erupted 10^ km^ over the last 2 m.y., our understanding of this exceedingly active history has been hampered by a lack of precise age data. Our preliminary age results have extended the age of TVZ volcanism back to 2.1 Ma, with major peaks of silicic volcanism at 1.2, 0.7, and 0.4 Ma. The frequency of major caldera-forming events is 55,000 years, exceptionally high for rhyolitic systems.

The traditional method for measuring the b^^O and b^^C of calcite cements normally involves bulk acid digestion of samples. Where two or more generations of calcite or other carbonates are present this method has the disadvantage of dissolving several components and the results obtained therefore reflect the weighted means of these components and may be essentially meaningless. The laser microprobe has potential to ablate calcite on a scale of 50-l(X)|im and thereby allow measurement of and values in individual calcite pains or within zones of complex grains thus providing the possibility to unravel complex diagenetic histories. This relatively new technique has yielded promising results, however, standardisation of procedures are required before general application can be accepted. Preliminary results for coarse-grained poikilotopic calcite fix)m a North Sea reservoir are to be reported. In one 7-8mm crystal varied from -10.2 to -12.7 (all units are per mill PDB) with a mean value of -11.7 (n=13). This compares favorably with the bulk acid digestion result of -11.7 for the same sample. The corresponding b^^C values rangefrom-10.4 to -12.2 with a mean of -11.4 which compares with a bulk acid digestion value of -12.2. The laser results show that this sample is relatively homogeneous and in this case bulk sampling also provides an accurate result. In another sample, 51^0 in ferroan calcite rangedfrom-1.0 to -17.2 with a mean value of -10.2 (n=9). This compares with a bulk acid digestion result of 11.5. The corresponding b^'^C values range from -10.1 to -14.8 with a mean of -12.3 compared to a bulk value of -12.9. The mean laser and bulk results are again comparable, however, in this case the bulk result masks the.complex history revealed by the laser. These results show that the laser microprobe produces and values which are comparable with traditional techniques so long as the laser system is properly calibrated.

ABSTRACTS

ICOG 7

PROTOLITH VARIATION IN T H E P R O T E R O Z O I C RAPAKIVI GRANITES O F EASTERN FENNOSCANDIA: Nd AND Pb ISOTOPES RAMO, O.T., University of Helsinki, Department of Geology, P.O. Box 115, SF-00171, Helsinki, Finland; HUHMA. H. and VAASJOKI. M., Geological Survey of Finland, SF-02150, Espoo, Finland The eastern Fennoscandian shield (southern Finland and Soviet Karelia) was the site of extensive anorogenic magmatism 1.65 to l i 5 Ga ago. The juvenile Svecofennian 1.9 Ga crust was reworked and a considerable volume of silicic magmas forming rapakivi granite complexes was emplaced at upper crustal levels. The ca. 1.64 Ga old rapakivi granites of southeastern Finland (e.g., the Wiborg batholith) show CNdOO values from -2.9 to -02 and the ca. 1.58 Ga old rapakivi'granites of southwestern Finland have cNdOO values between -3.1 to -1.4. Tdm model ages are nearly constant with an average value of 2.06 ±0.03 Ga (1 S.D.). Pb isotopic data on K-feldspars show /i2 (S&K) values from 957 to 9.98, averaging 9.76±0.11(1 S.D.). These isotopic characteristics are very close to those measured for the juvenile 1.9 Ga crust As the anorogenic granites probably represent average lower crust composition the conclusion is that the juvenile crust was not internally differentiated in terms of Nd and Pb isotopes at the time of emplacement of the granites. The rapakivi granites of Soviet Karelia show different isotopic compositioa Their CNd(L54 Ga) values are more negative (-8.1 to -5.7) and fi2 values calculated for K-feldspar separates are distinctly lower (8.73 and 8.75). This suggests two sources for these granites: a juvenile one probably equal to that tapped by the Finnish rapakivi granites and an old Archean one with retarded Nd and Pb isotopic evolutioa Mass balance calculations show that the latter component contributed approximately one third to these granites. On the basis of Pb isotopic data, a H2 value of about 6 is calculated for the Archean component This is in contrast to the high U/Pb character of the Archean upper crust and indicates major intracrustal differentiation of Pb isotopes in the Archean CTatoa

Multicollection SIMS measurements of oxygen-isotope ratios on extra terrestrial samples P.M.Read and D Smith, VG MicroTrace Ltd. Ion Path, Road Three, Winsford, Cheshire. CW7 3BX. and I C Lyon and G Turner, Department of Geology, University of Manchester, Manchester. Ml3 9PL. England. In situ secondary ion mass spectrometry (SIMS) measurements of oxygen isotope ratios have been made using the VG Isolab 54 multicollector ion probe. Analysis is performed on 10 um diameter areas at points on polished sections of extra-terrestrial samples. A primary Cs ion beam is utilised to provide high sensitivity and the secondary o" beam is analysed. Sample charging effects are found to be significantly reduced relative to conventional ion microscope analysis due to the much lower extraction field r ^ u i r e d in the ion microprobe. This yields stable 0~ signals without the use of electron beams to neutralise residual charging. Investigations of the utility of electron beam charge compensation are reported. The multicollector comprises of four off axis microchannel plate (MCP) ion counting channels, a selectable axial Faraday detector or electron multiplier, and four off axis Faraday detectors. In the present studies 0~ was detected on an off axis Faraday at 500 mass resolving power (RP), on the axial electron multiplier at 8000 RP and 0~ on an MCP channel at 1000 RP. High mass resolution is required on the axial channel to remove the 0H~ interference which can be several orders of magnitude greater than the "^'o signal. Multicollection has two major advantages relative'to peak switching for oxygen isotopic analysis. Firstly errors due to primary Cs beam instabilities, which are at the level of 0.2% over a 30 minute measurement period, are eliminated. Secondly, the usual requirement in peak switching to limit the ^ô" signal to -5 x 10^ counts per second (cps) to minimise dead time correction errors is removed. In the present studies the ^ 0~ signal is -lo" A (-5 x 10 cps) and a precision on the ^Ô/^Ô ratio of 0.1% is achieved in 30 minutes. Reproducibility of the ratio at the same point on the sample is -0.1%.

ETCHED FISSION TRACKS IN MINERALS: ABOUT THE SHAPE OF LENGTH DISTRIBUTIONS M. REBETEZ. A. CHAMBAUDET, M. MARS and M. GRIVET Laboratoire de Microanlyses Nucleaires U.F.R. Sciences et Techniques La Bouloie - 1 6 route de Gray 25030 Besan9on Cedex - France

Surface track and confined track measurements are both used to obtain length distributions used in the fission track dating method. Each of these measurements has its advantages and drawbacks, but neither has been corroborated by an extensive theory until now. The few theoretical track length distribution theories proposed so far have not taken all the experimental aspects into account, especially with respect to selective chemical etching and total etching time. A model of surface track length distributions was first constructed. Projected lengths and real lengths were calculated and the corresponding distributions versus etching parameters were plotted. This algorithm was then extented to confined track length distributions, track in track (TINT) and track in cleavage (TINC) and the results were ^ p h e d in the same way. The obtained distributions were analyzed and compared to the experimental results from other authors. The track under-etching and overetching conditions have considerable influence on several minerals. These results indicate that strict experimental conditions are required to study track length distributions. In such conditions, significant chronothermometric information can be obtained to interpret complex thermal histories.

SIMULATIONS OF IN-SITU COSMOGENIC-NUCLIDE PRODUCTION ON EARTH* c, Robert C., Space Science k Technology Div., Los Alamos National Laboratory, Los Alamos, NM 87545 USA The interactions of galactic-cosmic-ray particles in the Earth's atmosphere produce a cascade of particles. Some neutrons and muons in this cascade reach the Earth's surface and produce nuclides. Long-Hved radionuclides, Hke 5730-year ^^C, 0.3-Ma 36C1, 0.7-Ma Al, and 1.5-Ma i^Be, and stable nuclides, like ^He and ^^Ne, made in situ in certain materials can be used to study their recent exposure history. To determine ages, or erosion rates in rocks from which minerals are sampled, production rates for these cosmogenic nuclides are needed. A series of irradiations at the Los Alamos Meson Physics Facility (LAMPF) have simulated the production long-lived radionuclides in surface rocks. Some short-lived radionuclides were coimted in these and monitor targets, and long-lived radionuclides were measured by accelerator mass spectrometry (AMS) by a large number of collaborators. Most of the measurements are complete, and we are analyzing and interpreting the data. The results will assist in determining better rates for the in situ production of these nuclides. Neutrons are the dominant producer of nuclides in the top meter. Quartz and silicon were exposed to neutrons with an energy spectnmi similar to that in the Earth's surface. Using AMS. i^C, and ^^Al were measured. The i^Be/^^Al ratio in the quartz is different than that observed in proton irradiations but is consistent with that observed in situ in young rocks. The measurements are being unfolded to get better cross sections for the neutron reactions producing these nuclides.^ Starting near a meter's depth, muons become the major source of cosmogenic nuclides at depth. Several targets were exposed to stopped negative muons (/x") and a quartz target was irradiated with w 70 MeV /x" at LAMPF, and many short- and long-lived radionuclides have been measured. The ^°Be/^®Al ratio in the quartz exposed to stopped muons is similar to that in the neutron-irradiated quartz, which will made data interpretation less sensitive to a sample's depth history. • Work done under the auspices of the US DOE.

82 RADIOGENIC AND STABLE ISOTOPE PATTERNS IN BARITE FROM ORE DEPOSITS: BRIDGING THE GAP BETWEEN SULPHIDES AND HOST LITHOLOGIES REID D L, Geochemistry, U Cape Town, S Africa, BARR J M, WELKE H J, and SMITH C B, Bernard Price Institute of Geophysics, U Witwatersrand, Johannesburg, S Africa Barite is a mineral that offers unique opportunities for isotope tracer studies of certain ore deposits as it bridges the gap between metal sulphides and associated lithologies. Barite is commonly associated with a wide variety of sulphide ore deposits and is usually a product of the same depositional process. Practically all the isotope systems investigated routinely can be accessed with this mineral, thereby allowing, for example, an internal check on lead and sulphur isotope relationships, as well as providing additional constraints in the form of strontium and oxygen isotope patterns. Application of barite isotope characterisation is particularly useful in the study of Proterozoic sediment-hosted polymetallic sulphide deposits, since isotope patterns in host lithologies can be influenced by additional factors such as inherited detrital components and metamorphism. In a study of the giant sedex metal sulphide deposits in central Bushmanland of South Africa, barite has been subjected to sulphur, oxygen, strontium and lead isotope analysis to enable further characterisation. Lead isotope patterns for barite and sulphides overlap, confirming a close genetic relationship. Strontium isotope ratios in the barite are very high, compared with estimated Proterozoic sea water. All the barites are enriched in ^S and and variation in ^^S correlates positively with pointing to primary control by the mineralising fluids, rather than later metamorphic effects. The above isotopic patterns taken together with trace element abundances in the barites will be discussed in terms of placing constraints on the nature, source and transport of the exhalative fluids responsible for the Bushmanland ore deposits.

TESTING Ar- A t POST-2 Ga AGES FOR PSEUDOTACHYLITE FROM THE VREDEFORT STRUCTURE (SOUTH AFRICA) REIMOLD, W , U . , Econ. Geol. Res. Unit, Univ. Witwatersrand, P . O . WITS 2050, Johannesburg, RSA, STEPHAN, T. and JESSBERGER, E . K . , MPI f . Kemphysik, P.O. Box 103980, D-6900 Heidelberg, FRG. 40

39 Ar- Ar stepheating results from Vredefort pseudotachylite suggested that these breccias were formed over the period from 2 . 2 - 1 . 1 Ga ago (Reimold et al, 1990). Rb-Sr isotope data, unpublished ^ A r - Ar results by the late Hugh Allsopp and geological evidence were cited in support of these data, which are in sharp contrast with previous theory linking all deformation phenomena at Vredefort to a single catastrophic event at ^^2 Ga ago. In order to test Jhe validity of these results and the suggestion that Ar- Ar dating of pseudot achy lite matrix may be a useful chronological tool for the dating of tectonic events, mineral separates from the host rocks to two of the "younger" pseudotachylite samples were analysed. For charnockitic granite NW-28, host rock to pseudotachylite 31A dated at 1.44 ± 0.08 Ga, a biotite fraction yielded a well-defined plateau age of 2 . 0 7 ± 0.01 Ga, while a hornblende separate gave a plateau age of 3.03 ± 0.03 Ga, but shows evidence for later disturbance (resetting to 2.2 Ga for low T gas release). No significant post-2 Ga thermal effect is indicated for these samples. For Outer Granite Gneiss sample USA-30A, host rock to pseudotachylite 30A dated at 1.39 ± 0.08 Ga, biotite, hornblende, plagioclase and microcline (splits of separates that yielded a 2 . 0 1 ± 0 . 0 5 Ga Rb-Sr isochron age) gave complex gas release spectra. Ages of 1.76 ± 0.04, 1.66 ± 0.05, 1.57 ± 0 . 0 4 and 1.39 ± 0.01 Ga were calculated, respectively. The microcline age is in excellent agreement with the pseudotachylite age. Both biotite and hornblende are partially chloritised, which may be the reason for Ar loss. (Ref.: Reimold, W . U . , Jessberger, E.K. and Stephan, T. (1990) Tectonophysics, 171, 139-152).

ICOG 7

ABSTRACTS

MEASUREMENT OF OSMIUM ISOTOPIC COMPOSITIONS BY LASER RESONANCE IONIZATION OF SPUTTERED ATOMS

L. REISBERG. P. MARCANTONIO, J. ENGLAND. A. ZINDLER, S. WEAVER AND J. RUBENSTONE, Lament Doheity Geological Observatory

of Columbia University, Palisades, N.Y. 10964 U.S.A.

The strong fractionation of Re and Os that occurs during partial melting makes the Re-Os isotopic system a potentially very useful tool for understanding the evolution of the earth's crust and mantle. However, the exploitation of this system has been limited by analytical difficulties resulting from the extremely low concentrations of Os in crustal rocks (generally a few parts per trillion). This problem is compounded by the poor ionization efficiency of Os during thermal emission. Two techniques that have been used successfully to measure the isotopic ratios of nanogram quantities of Os are sputter-induced ionization [Luck (1980)] and laser resonance ionization of thermally ablated neutral atoms [Walker and Fasset (1986)]. At Lamont, we are using a pulsed excimerpumped tunable dye laser to resonantly ionize Os atoms sputtered by a primary beam of Ar+ ions produced by a duoplasmatron. The primary beam is pulsed synchronously with the laser, and delay generators are used to adjust the timing of the plume of sputtered material so that it matches precisely the arrival of the laser pulse. Thus particles are sputtered from the sample only when photons are available to ionize them. The Os isotopic composition is then measured in the Lamont VG Isolab 54 mass spectrometer [England et al., this volume]. We have found the efficiency of resonance ionization to be approximately 1000 times greater than the ionization produced by sputtering alone, with the exact extent of resonance ionization strongly dependent on the geometry of overlap between the plume of sputtered atoms and the laser beam. Isobaric molecular interferences, produced either by sputtering or by non-resonant ionization, are quite small relative to the size of the resonant signal, and can be completely corrected for by subtracting off-resonance background measurements. We believe that this technique will allow us to measure quantities of Os that are substantially smaller than can be measured currently by the sputtering technique alone.

APPLICATION OF SINGLE-CRYSTAL LASER FUSION 40Ar/39Ar DATING TO THE DETERMINATION OF SEDIMENTARY PROVENANCE RENNE. P.. and BECKER, T., Geochronology Center of the Institute of Human Origins, 2453 Ridge Rd., Berkeley CA 94709 USA 40Ar/39Ar laser probe systems are capable of dating individual cr>'stals of sand-sized potassic mineral âins (e.g. feldspars, micas, hornblende) as young as Quaternary in age with high precision. Step-heating with the laser allows assessment of possible argon loss or excess argon, as with conventional bulk ^Âr/^Âr analysis. Thus argon retention ages of detrital minerals can be used as criteria for inferring the specific provenance of sandstones. In addition, single ciystal argon retention ages can provide a wealth of information about both the pre-depositional and post-depositional histories of detrital grains, including clarification of source region tectonics, sediment transport and diagenetic processes. An example is given from the Middle Eocene arkosic Montgomery Creek Formation of northern California. Laser probe dating of more than 100 grains of biotite and muscovite indicate derivation primarily from the Idaho Batholith region. Rapid unroofing, erosion and sedimentation of muscovite grains dated as young as 47 Ma require minimum uplift rates of ca. 2 mm/yr for the Idaho Batholith. The age distribution of Montgomery Creek muscovite grains is consistent with a single fluviatile system transporting these sediments. However, preliminary results from total-fusion analyses of a few hornblende grains and step-heating analyses of several biotite grains suggest that the Klamath Mountains (or another source area containing both hornblende and biotite with apparent ages greater than 150 Ma) may have had some input into the early drainage system and perhaps also in the latter period of basin filling. If this is the case, identifying periods of unique source terrain input will provide clues to the timing of regional tectonic events. Many biotite grains are variably vermiculitized, and the age of this post-depositional alteration inferred from analysis of composite grains is between 4 and 10 Ma.

ABSTRACTS

Cooling histories and their tectonic implication for the Indian plate crystalline thrust stack of northern Pakistan. Paki David C. R e x ^ Peter J. Treloar^, and Phillip G. Guise^ 1. 2.

ICOG 7

Department of Earth Sciences, Leeds University, LS2 9 J T,U.K. ,U.F D ^ aartmei r t o e n t of Geology, Imperial College, London, ^2BP, U.K.

In northern Pakistan, the crystalline internal zones of the Indian plate have a post metamorphic cooling history which is documented here by amphibole and mica Ar-Ar, K-Ar and Rb-Sr geochronology. Cooling ages for hornblende (c. 500°C) muscovite (c. 350°C) and biotite (c. 300°C) are 34 ±4, 24 to 30, 20 to 29 My respectively. These ages, together with those derived from fission track data (Zeitler 1985) demonstrate a rate of cooling which peaked between 30 and 20 My, culminating with cooling through the apatite blocking temperature at 19 My. This cooling was initiated during the post metamorphic evolution of the Indian plate south verging imbricate thrust stack. The bulk of the cooling, which is equivalent to an overburden stripping of some 8 ± 2 km, reflects exhumation of this thrust stack through a combination of erosion (recorded in the Miocene molasse sediments of the forland basins) and crustal extension within the upper levels of the Indian plate and structurally overlying the M.M.T. zone. Both erosion and extension were the direct consequence of the evolution of the thrust stack and record the destruction of the resultant topographic high.

Pb ISOTOPES AND THE TERRAINS OF EASTERN AUSTRALIA J.R.RICHARDS, Australian National University, Canberra, ACT, 2601, Australia, and G.L.CUMMING, University of Alberta, Edmonton, T6G 2J1, Canada. Information on galenas in eastern Australia, principally from the Lachlan Fold Belt, has now about doubled in amount since the last Meeting (Richards, 1986). The contrasts in isotopic character persist. This display is an attempt to express these contrasts in a semiquantitative manner, and to seek possible correlation with the terrains outlined in recent geochemical and geophysical studies (Chappell et al., 1988; Wellman, 1987 § 1988). The chosen isotopic parameter is the 23 of the source rock as it would be if sampled today (yp). The growth-curve ("crustal") value is 10.75±0.03 (2.SD), whereas samples plotting close to the Basalt Reference Line of Hart (1984) yield yp values at near 10.40. The space in between has been subdivided into four (coloured) ranges based on comfortable multiples of the standard deviation, and these have been plotted on the map showing fold-belt outlines. Broad correlations are seen, between the "crustal" values which predominate in the Lachlan and the Thomson/Hodgkinson Fold Belts of North Queensland. In the New England Fold Belt and its northerly correlative along the southern Queensland coast, the y-value is lower, hinting at a source rock somewhat depleted in U. The very low-y localities are now 5; four occur within the north-central sector of the NSW Lachlan. All have "basalt"-type y-values. One is related to the Gilmore Suture (G, Carr, pers.comm.) but for the others connection with identifiable structural features remains obscure. Chappell B.W.et al., Aust.J.Earth Sci., 505, 1988. Hart S.R., Nature, 753, 1984. Richards J.R., Terra Cognita, 6, 229, 1986. Wellman P., BMR Res,News Lett,,6, 12, 1987. Wellman P., BMR Res.News Lett., 8, 14, 1988,

Nd AND Sr ISOTOPE VARIATIONS IN THE CENTRAL SAN JUAN CALDERA CLUSTER, COLORADO, USA RICIPUTI, Lee, and JOHNSON, Clark, Dept. of Geology, University of Wisconsin, Madison, Wisconsin, 53706, USA, and SAWYER, David, and LIPMAN, Peter, U.S. Geological Survey, Denver, Colorado, 80225, USA (Ig^) ratios for 9 large-volume ash-flow e Nd and initial tuffs (50-3000 km^) erupted from 6 calderas of the central San Juan cluster (Oligocene) lie between compositions inferred for regional Proterozoic crust (<-12, >0.708, respectively) and mande (-4 to 0, <0.7045). Both Nd and Sr isotope ratios of the tuffs become more mantle-like with time for two discrete caldera cycles, but the isotopic variations between the different tuffs do not correlate with elemental compositions, e ^^ values for the 1st cycle tuffs (28.3-26.9 Ma) increase from —8.0 for the oldest Masonic Park and Fish Canyon Tuffs to ~-6.0 for the youngest Snowshoe Mm Tuff, while Is^ ratios decline from 0.7067 to 0.7058. The multiple tuffs from the youngest and northwestemmost San Luis caldera complex form a separate cycle, although the e^d-time trends are similar: e^^ values increase from —7.8 for the 26.45 Ma Rat Creek Tuff to —6.8 for the 26.15 Ma Nelson Mtn Tuff. Trends in Ig^ ratios for this cycle are less clear due to zonations within individual tuffs. The Nd and Sr isotope trends can be modelled by AFC processes in which mantle-derived magmas assimilate and progressively hybridize Proterozoic crust. Calculations suggest that a 4-unit increase in e^^ value of assimilated crust is needed to generate the 2 unit increase between the Fish Canyon and Snowshoe Mtn Tuffs. Assuming a conservative 1:4 ratio of extrusive:intrusive volumes, input of >160,(XX) km^ of mantle-derived magma into the crust is required to produce the volcanic and associated plutonic rocks of the central caldera cluster. The modelling implies a 1:2 to 2:1 mixture of recent mantle-derived material:Proterozoic crust in the tuff-forming magmas.

EARTH'S OXYGEN ISOTOPIC COMPOSITION Francois ROBERT, Agnes REJOU-MICHEL. Laboratoire de Gôchimie des Isotopes Stables; IPGP and University Paris VII, 75251 PARIS CEDEX 05, FRANCE. Experimental: We have recently improved our precision on the determination of Si'Ô/a^Ô ratio: 0.52288 ± 0.00086 for the MORB's. Assuming that the and relationship can be still approximated by linear equations within such sensitive limits, the equations of the lines defined by different types of rocks have been determined : 1) Onverwacht cherts (3.5 B.Y. old) 2) modem cherts (>100 M.Y.) 3) lower and upper mantie derived lavas 4) sea and rain waters and 5) lunar samples. Results : The mantie lavas values between +4.0 and +7.0%c) are systematically below the line defined by modem and precambrian cherts (by 0.1 %o); their mean S^'Ô value for = 0 is: + 0.24 ± 0.02%o. Precambrian and modem cherts have a mean S^'Ô values (for b^Ô = 0) of 0.16 ± 0.03%o and 0.05 ± 0.30%o, respectively. SMOW and rain water have a mean S^'Ô value for = 0 almost indistinguishable from 0%o: = 0.033 ± 0.008%o. The two lunar samples lie exactly on the correlation defined by the terrestrial mantie basalts. Interpretation: As noted previously [1,2], tiie data reveal no oxygen isotopic heterogeneity within the moon-earth system (within a precision better tiian ± 0.05%o on 616), illustrating the common origin of these planets. It is important to note tiiat tiie isotopic compositions of the oldest precambrian cherts show no departure from the present day average chert fractionation line (0.51510.003 and 0.521±0.008, respectively) and have a similar 0 intercept. This indicates that, if the water on Earth was inherited from an extraterrestrial source (different in ^ ^ from the bulk mantle), the oceans were already in isotopic equilibrium with mantle rocks 3.5 B.Y. ago. The difference between the intercept at = -K).24%o for die mantie basalts and the mean oceanic value (~0%o), is still not well understood. References: [1] Mayeda T.K. & Clayton R.N. Proc. Lunar Planet Sci. Conf. IV^ (1980) p. 1145-1151. [2] Javoy M., Pineau F. & Delorme H. Chem. Geology (1986) 57, 41-62.

Isotopic compositions of minerals condensed in plasma; application to cosmochemistry. Fran9ois ROBERr, Agnes REJOU-MICHEL^ Jean-Louis BIRCK" and Fran9ois FLORY**.

(•) Lab. de G6ochimie et Cosmochimie, IPG, 2 Place Jussieu 75251 Paris CEDEX 05; France. (»») Lab. Optique. E. N. S. Domaine Universitaire Saint-J6r6me. 13 397 MarseiUe CEDEX 13; France.

We recently reported [1] precise measurements of the oxygen isotopic fractionation of minerals formed by condensation in plasma, evaporation and condensation at temperatures ÔOOK, and evaporation by electronic impact. This study investigated if isotopic anomalies similar to those observed in meteoritic minerals or during the synthesis of ozone [2] could be obtained in minerals of cosmochemical interest. Experiments. TiO was vaporized at a pressure of 10'^ atm. using a 8 keV electron gun and simultaneously irradiated by 250 eV oxygen atoms for 30 min. Amorphous Ti02 was deposited on an aluminium foil. A similar experiment was repeated with TÔj to compare the oxygen isotopes of the two minerals containing different number of oxygen atoms. The isotopic composition of the oxygen beam was also determined by irradiating a silicon foil. Results. The slope of the correlation bewteen and S^'Ô in the 3 isotope diagram of oxygen was always close to 0.52 although the isotopic fractionation is as high as +100%o in The isotopic compositions (and yields) vary with the thickness of the minerals. This suggests a temporal evolution in the proportion of oxygen atoms supplied by the ^Ô-rich beam as opi^sed to the TiO target which react to form Ti02. The isotopic composition of the oxygen atoms ori^nated from the ion source, is normal. No marked isotopic difference exists between Ti02 and Ta20r. A consequence for the formation of the solar system can be derived: during its formation, chemical reactions induced by the interaction between the primordial intense solar wind and the solar nebula, did not produce isotopic anomalies. References. [1] Robert F.(1989) Lunar and Planetary Sci. Conf. XX, 908-909. [2] Thiemens M. H. and Heidenreich III (1983) Science 219, 1073-1074.

SULPHUR ISOTOPE STUT)IES OF EPITHERMAL AU-AGPB-ZN-CU DEPOSITS IN THE HAU-RAKI GOLDFIELD NEW ZEALAND ROBINSON. Brian W., Institute of Nuclear Sciences, DSIR, P.O. Box 31 312, Lower Hutt, New Zealand, and CHRISTIE, Anthony B., New Zealand Geological Survey, DSIR, P.O. Box 30 368, Lower Hutt, New Zealand Numerous separate gold-silver/base metal epithermal deposits are located in the Coromandel peninsula of New Zealand's North Island. This is a region of Tertiary volcanic rocks overlying Mesozoic greywacke basement and the Miocene-Pliocene mineralisation occurs as fissure-filling quartz/carbonate lodes and occasional sinters. Average values for the commonly found sulphide minerals are: Pyrite +2.1 ± 1.5 (n=114), Chalcopyrite +2.1 + 0.8 (n=77). Sphalerite + 2 . 0 ± 1 . 1 (n=105), Galena +0.3 ± 1.0 (n=76). Mineral pairs indicate temperatures of 250350°C for base metal mineralisation; precious metals formed at lower temperatures. Sulphate minerals have values from + 5 to +25 %o representing some movement to isotopic equilibrium with the HjS source of +2%o. In comparison, HjS in the present day geothermal systems to the south have average values of +4.8 ± 1.2%o (n=83). A large difference in C/S ratios between the recent volcanic fumaroles and geothermal systems suggests that sulphur is fixed as sulphide minerals beneath the latter (compare the Tertiary epithermal deposits). As the fluids move closer to the surface small amounts of sulphate form and some degree of isotopic equilibrium may take place. The small isotopic difference between the sulphur source in the Coromandel (+2%o) and the geothermal areas (+5%o) presumably represents the amount of sedimentary material which has entered the parent magmas. Sulphur isotopic compositions of greywacke/argillite basement rocks are very variable: average +3.7 ± 9.7%o (n=73).

ICOG 7

ABSTRACTS

PRECISE CALIBRATION OF TRACER COMPOSITIONS FOR SM-ND ISOTOPIC STUDIES . J.C RODDICK, R.W. SULLIVAN AND F.6.L. DUDAS, Geological Survey of Canada, Ottawa, Ontario, KIA 0E8 Sm-Nd studies require accurate tracer compositions used for determination of sample '^'Sm/^^Nd and ^'^Nd/'^^Nd. The accurate determination of tracer isotopic compositions is limited by mass fractionation and lack of an internal reference normalization ratio. The figure presents a calibration technique (modified after Hofmann, EPSL10,397,1971) which corrects for mass fractionation. For appropriate isotopic ratios, a mixing line between normal Nd and tracer Nd defines a locus of compositions with no mass fractionation. Superimposed on the mixing lines are mass fractionation paths. At the critical mixture composition mixing and fractionation lines coincide. The critical mixture and normal Nd composition define the mixing line. Intersection of this mixing hne with the tracer fractionation line determines the tracer composition relative to reference normal Nd. The GSC tracer is ORNL ""^Nd tracer modified by the addition of a small amount of ^^^Nd. A larger a n g l e of intersection of 100.0 mudng and tracer fractionation lines compared to the 80.0 ORNL tracer permits the GSC 60.0 composition to be accurately 40.0 determined to within 0.02% per 20.0 mass unit. An a n a l o g o u s procedure applies 0.3 NORMAL Nd to a Sm tracer.

PROVENANCE AGES OF THE TORRIDONIAN SANDSTONE OF NW SCOTLAND USING SINGLE GRAIN U-PB ZIRCON ANALYSIS ROGERS, G., KROGH, T. E., BLUCK, B. J. and KWOK, Y. Y. (Scottish Universities Research and Reactor Centre, East Kilbride, Glasgow, G75 OQU, U.K.) The Torridonian sandstone of NW Scotland represents an essentially unmetamorphosed clastic succession of late Proterozoic age. Sedimentological studies show that the upper part of the Torridonian, the Applecross Formation, was deposited mainly as channel bars within a large river system. Palaeocurrent directions indicate derivation from the northwest. This suggests drainage areas outside NW Scotland, further onto the Laurentian plate. In order to assess the age of the source of the Applecross Formation detrital zircons were separated from a channel bar deposit, handpicked into a variety of populations based on morphology and colour, and analysed individually. The majority of the 22 analyses are <1% discordant with ages falling into three distinct groups: 1088-1193, 16251662 and 2628-2857 Ma. Grains of identical appearance give ages up to 1.3 Ga apart. The spectrum of ages suggests a major provenance in crust of Grenville, mid-Proterozoic and Archaean age, which appears inconsistent with eastern Greenland, the postulated source of previous models. Whilst the Archaean grains may have had a local provenance from the Lewisian, the younger grains do not represent such a source. A provenance in Labrador seems to best satisfy the available data. It is unclear, however, if NW Scotland lay adjacent to Labrador during Torridonian deposition or whether sediment could have been dispersed into older basins which were then re-eroded into the Torridonian. Such factors influence models of the postulated disposition of crustal blocks during the late Proterozoic.

ABSTRACTS

IC0G7

CORALS AS PALAEOENVIRONMENTAL INDICATORS ROKSANDIC ZARKO, Bureau of Mineral Resources, Canberra, ACT, 2601, Australia Hermatypic (hard) coral skeletons from GBR, are the product of symbiotic biological activity between photosynthetic algae zooxanthellae and coral polyps. The carbon isotope ratio of the skeleton reflects this activity but also the environmental signature derived from the seawater bicarbonate. By determining the biologically controlled fractionation it may be possible to reconstruct isotopic signature related to palaeoclimate. The magnitude of the metabolically derived carbon in the skeletal carbonate of Acropora grandis has now been determined for the extreme morphological habitats. For maximum photosynthetic rates at 10 m depths, there is 25 % contribution to the skeletal carbonate. When photosynthesis completely ceases, as simulated by placing an opaque dome over the colony for 8 weeks to indicate depths greater than 50 m and where the light becomes the limiting factor, there is less than 5 % of metabolic carbon in the skeleton. The remainder is due to the bicarbonate that directly reflects the seawater conditions. For palaeoclimate predictions the major remaining obstacle is to reliably measure the isotopic composition of organic matter remaining in the fossil coral.

PROVENANCE INSIGHTS FROM U-PB AND SM-ND ANALYSES OF DETRITAL MONAZITE ROSS, G.M., Geological Survey of Canada, 3303 33rd St NW, Calgaiy, Alberta T2L 2A7 and R.R. PARRISH, F.O. DUDAS, Geological Survey of Canada, 601 Booth St., Ottawa, Ontario KIA 0E8 Monazite commonly is overlooked as a detrital phase in isotopic studies of sediment provenance yet, by virtue of its high concentration of U and Nd, it provides a means of characterizing source areas in terms of age and Nd isotopic composition. Single detrital monazite grains from the Middle Proterozoic Bonner Formation (Belt Supergroup, southwestern Montana) have been analysed using U-Pb and Sm-Nd isotopes, along with single grain analyses of detrital zircon with the following results: U-Pb ages from 8 single zircon grains range from 1701 - 1860 Ma with two ca. 2.6 Ga grains. Concordant UPb ages from 5 single monazite grains range from 1643 - 1786 Ma. Sm-Nd analyses of the same monazite grains yield C^^ at the time of crystallization of +0.7 to -14.5. These data suggest that the source area for the Bonner Formation was composed of Early Proterozoic crust that had interacted with variable amounts of, in part, evolved Archean crust. Sm-Nd whole rock model ages of 1.9 - 2.1 Ga for Belt Supergroup sediments thus reflect the crustal inheritance of the Proterozoic magmatic rocks in the source area and do not correspond to an actual crust or source area age. These results place constraints on the age and isotopic characteristics of the source area and demonstrate the utility of detrital monazite in provenance studies.

INCOMPATIBLE TRACE ELEMENTS IN FLUIDS T R A P P E D IN MANTLE NODULES J.M. R O S E N B A U M * . J.L. RUBENSTONE, AND A. ZINDLER,

DYNAMICS AND EVOLUTION OF VOLCANIC S Y S T E M S USING T H AND U ISOTOPES MEASURED BY S I M S J.L. RUBENSTONE. A. ZINDLER, B. BOURDON, J. ENGLAND, L. REISBERG, B. HAMELIN, V. SALTERS, S. WEAVER, W.-X.

Lamont-Doherty Geological Observatory of Columbia University, Palisades, NY 10964 U.S.A. (*now at Enrico Fermi Institute, Univ. of Chicago, Chicago, IL 60637 U.S.A.) Fluid inclusions in minerals from spinel Iherzolite nodules are rare samples of C02-rich fluid or silicate melt present within the mantle. We nave detennined the trace element distributions in Iherzolite nodules by comparison of IDMS measurements of clean, leached inclusion-free and inclusion-bearing pyroxene separates in nodules from the southwestern USA (San Carlos, Kilbourne Hole, and Potrillo Maar) and Nunivak Island, Alaska. Absolute fluid concentrations are obtained by combining data from sample aliquots with different amounts of trapped fluid. Although volumetrically trivial within most mantle minerals, even small amounts of fluid (<0.003 wt.%) will dominate the nodule budget for Rb, Cs, Ba, Th, and Pb, elements most incompatible in the minerals. In contrast, Sm, Nd, and Sr are concentrated in clinopyroxene compared to the fluid phase. Two nodules with different inclusion compositions were smdied in detail. Inclusions in a Iherzolite from Nunivak Island are primarily CO2 with a trace of glass, while a San Carlos Iherzolite contains silicate glass inclusions with minor amounts of CO2. The abundances of Cs, Rb, Sr, Pb, U, and Th in the inclusions are the same within uncertainty for both nodules (Cs ppm, Rb -15 ppm, Sr -1000 ppm, Pb -35 ppm, U -8 ppm, Th -25 ppm), but differ for K and Ba (K -430()0 and 12100 ppm, Ba -1200 and -3000 ppm, in Nunivak and San Carlos, respectively). The trace element abundance patterns in these fluids are similar to those of carbonatites or kimberhtes. If the fluids and host minerals are in chemical equilibrium, then K is relatively more compatible in the the C02-rich Nunivak fluid, and U, Th, and Ba are relatively more compatible in the solute-rich, C02-bearing fluidfromthe San Carlos nodule. Whereas simple batch melting of a fluid-free fertile Iherzolite to varying degrees can match the the abundances of the mineral-hosted trace elements (Sm, Nd, Sr) in average basalt types, the calculated concentrations for the fluid-hosted elements are much too low in the model melts. A mixture of these melts and the observed fluids still fails to reproduce the trace element signatures of typical mafic magmas, suggesting that the fluid phase now preserved in the nodules cannot be directly involved in basalt production without first undergoing further chemical fractionation.

LI, AND F. MARCANTONIO, Lamont-Doherty Geological Observatory of Columbia University, Palisades, NY 10964 U.S.A.; J.A. WOLFF, Department, of Geology, University of Texas, Arlington, TX 76()19 U.S.A. The timing of mass transfer and chemical fractionation in magmas is critical to understanding processes beneath active volcanoes. Temporal data currently lacking include simple eruption frequency at inaccessible volcanic centers (e.g., mid-ocean ridges), rates of magma transit from zones of anatexis, and duration of stora|e and fractionation at intermediate to low pressures. Given sufficient measurement precision, systematics of U and Th isotopes in the 238U decay senes record magma evolution over the past several 100 Ka of volcanic activity. Alpha-spectrometry methods for U-series isotopes are limited for longer-lived nuclides in volcanic rocks by low precision, long counting times, and large sample requirements. To overcome these limitations, we have developed techniques for measuring 230Th/232Th using SIMS on the Lamont Isolab 54 (Zindler et al., this volume) that permit isotopic measurements of reasonable quantitites of Th- and U-poor materials (such as MORB glass, phenocrysts or ^oundmass from low-Th lavas, and minerals from cognate xenoliths) and minute amounts of Th-bearing accessory phases or high-Th pumice. Our initial work has been with MORB and with evolved compositions from oceanic islands. Dredged glasses from the axial valley of the Mid-Atiantic Rid^e and the crest of the East Pacific Rise were hand-picked and leacned to eliminate palagonite and secondary Mn-oxide, which can contain extreme enrichments in 230Th. Density separates (produced using heavy liquids in an ultracentrifuge) of fine-grained, holocrystalline basah from individual off-axis samples from the EPR have a range of 238u/232Th suitable for determining eruption age with an intemal isochron. Late Q u a t e m ^ pyroclastic eruptions from within the Las Canadas caldera on Tenerife (Canary Islands) include both basaltic and phonolitic pumice, as well as at least three distinct phenocr}'st assemblages. U-Th isotope s)[stematics of crystals and glasses help constrain timing of crystallization and magma mixing. Results of both studies will be discussed at the meeting.

IC0G7

ABSTRACTS

Pb ISOTOPES IN THE LOWER CONTINENTAL CRUST: NO SOLUTION TO THE Pb PARADOX RUDMCK, R. L. RSES, Aust. Nat. Univ., Canberra, 2601, Australia and GOLDSTEIN, S.L., M.P.I, fur Chemie, Saarstr. 23, D-65(X) Mainz, W. Germany The depleted mantle shows long term depletion in Rb/Sr and Nd/Sm which complements the continents. However, both the upper continental crust and depleted mantle show enrichment in U/Pb, causing their Pb isotopic compositions to be more radiogenic than bulk earth. This has been termed the Pb paradox. Granulites are generally U-depleted, thus the lower crust has been suggested to be the unradiogenic Pb reservoir needed to resolve the Pb paradox. A common feature of lower crustal xenoliths erupted through tectonically young crust is that they have radiogenic Pb isotopic compositions (filing in and above the MORB field on a 20fepb/204pb vs. 2O7pb/2(0pb diagram), have low U/Pb ratios and have old Nd model ages (TDNOImplications of these observations are: (1) the protoliths have lost U only recentiy, (2) cumulate rocks (where low U/Pb is primaty) having radiogenic Pb are young, and (3) the old TDM ages in these young rocks are a product of mixing between underplated mafic magma and pre-existing continental crust The mixing trends show that this pre-existing lower crust had low eNd and radiogenic Pb. The few xenoliths tiiat have unradiogenic Pb isotopes (plotting to the left of the geochron), are erupted in areas that have remained tectonically stable over long time periods (e.g., southern Africa, northern Tanzania). The Pb isotopic compositions of the lower crust thus appear to be related to the tectonothermal age of the crust: unradiogenic Pb can only develop in old, stable regions. During orogeny, the Pb in the lower crust is mixed with radiogenic upper crustal Pb and mantie-derived Pb to "rejuvenate" the Pb in the lower crust (similar to the orogene of Plumbotectonics (Zartman and Doe, 1981)). These observations allow compositions of avarge lower crustal Pb to be calculated using crustal tectonothermal age and upper crustal Pb estimates (e.g., Stacey and Kramers, 1977). The resulting compositions lie to the left of tiie 4.57 Ga geochron but are too radiogenic to balance the upper crust and upper mantie Pb reservoirs so to place the bulk earth on the geochron. Thus the lower crust, on its own, does not appear to resolve the Pb paradox.

OSMIUM ISOTOPE SYSTEMATICS IN THE BUSHVELD COMPLEX RUIZ, Joaquin, and McCANDLESS. Tom, Department of Geosciences, University of Arizona, Tucson, Arizona 85721, USA The origin and evolution of the melts that formed the Bushveld Complex and associated ore deposits is still not clearly understood. Strontium and Pb isotope evidence has been used to argue for assimilation of crustal material by the Bushveld magmas. However, these isotope data do not properly constrain the origin of the magmas. Osmium isotopes can help constrain the source of the melts that formed the Bushveld because of the large Re-Os fractionation between crust and mantie. Previous Os isotope work on the Bushveld has been confined to minerals from one horizon the Merensky Reef. Laurite-RuSi, erlichmanite-0sS2, which are the main osmium bearing phases of the reef yield 1870s/1860s ratios of 1.40, which is much greater than the ratio of the mantie at the time and requires a component of crustal Os. Here we report the first Os data on sulfides from the UG2-chromitite layer, which is approximately 3(X) meters below the Merensky Reef and is a major host of platinum-group elements (PGE) from the Bushveld. The Os isotope measurements were made on an inductively coupled plasma mass spectrometer (ICP-MS). The Os was distilled from the sample and concentrated as a hexachloride. The resulting solution was directiy analyzed by ICP-MS. Multiple analyses of samples and standards indicate a better than 1% precision for the I8'70s/186os ratio, which is adequate for this study. The analyzed samples contain the Os mostly in laurite although some erlichmanite is also present. All samples from the UG2 analyzed for this study have 18/ ratios of about 1.40. These values are identical to those of the Merensky Reef and indicate that the Bushveld complex had very homogenous Os isotopic ratios laterally (samples from different parts of the same horizon) and also in different horizons (UG2 horizon vs Merensky Reef). These isotopic characteristics are difficult to reconcile with hydrothermal alteration and argue strongly for assimilation of crustal materials by the Bushveld magmas. The homogeneity in the Os isotope data suggest that crustal assimilation had to be early in the evolution of the Bushveld magmas. If multiple injections of magma produced the layered complex, as suggested by some authors, then the magmas had to be tapping a reservoir that had homogenized in Os isotopes.

IMPLICATIONS FOR CRUSTAL UNDERPLATING AND ISOTOPIC RESERVOIRS FROM Nd-Sr-Pb DATA FROM LOWER CRUSTAL XENOLITHS FROM THE WESTERN U^ AND MEXICO. RUIZ, Joaquin, Department of Geosciences, University of Arizona, Tucson, Arizona 85721, U.S A. The lower crust has long been thought to consist of predominantly mafic and intermediate granulites. The question remains, however, whether the granulites are mainly cumulates, restites or mafic magmas that underplated the lower crust. A better understanding of the nature of the lower crust would severely constrain models for crustal evolution. Recent seismic and petrologic studies of areas undergoing extensional tectonism have concluded that mantiederived mafic magmas are added to the lower crust; in these models underplating becomes a significant process in the petrogenesis of the lower crust. However, there are important constraints that argue against this interpretation for the petrogenesis of the lower crust. First, the amount of underplating possible in areas of passive extension can be calculated by thermal modelling, which suggests that the vast amounts of underplating are not possible simply by thinning of the crust. For example, an extreme case of 400% passive extension would only produce 5 km of basalt with melting at depths of 100 km. The thermal models indicate that if extensive underplating by mantle-derived basalts occurred, the partial melting of the mantle was not caused by tl^inm'ng of the crust alone, and heating of the mantle by other sources is necessary. Second, xenolith data indicate that the mafic and intermediate lower crustal xenoliths from Arizona, New Mexico and Mexico generally consist of cumulates, as well as lesser amounts of mantle-derived basalt. In many cases, garnet-rich metasedimentary granulites with REE patterns similar to NASC are also abundant. Geobarometry of the metapelites indicates that the metasedimentary rocks equilibrated at pressures consistent with those of the lowermost crust. The mafic and intermediate orthogneiss xenoliths have Nd-Sr isotopic compositions that are close to CHUR indicating that large amounts of mantle with CHUR Nd-Sr isotopic values is being tapped or that mantle-derived material effectively homogenizes with crustal material to produce an isotopic reservoir with CHUR isotopic characteristics. Regarding Pb, some xenoliths plot to the left of the geochron, but many xenoliths plot to the right of the geochron in the MORB field. It seems from these data that the lower crust cannot be a common Pb reservoir but that it may be considered a region with CHUR Nd-Sr isotopic characteristics. The xenolith data require that the lower crust have a complex history that includes magmatic processes such as intracrustal differentiation and underplating and the tectonic incorporation of sediments that were once at the Earth's surface.

B, Li, AND lO^Be SYSTEMATICS OF OKMOK AND RECHESHNOI VOLCANOES, UMNAK ISLAND, ALEUTIANS RYAN, Jeffrey, Dept Terrestrial Magnetism, Washington, D.C. 20015,USA; MILLER, Daniel, L-DGO, Palisades, NY, USA; MORRIS, Julie, Dept Terr. Magnetism; LANGMUIR, Charles, L-DGO Okmok and Recheshnoi, the adjacent tholeiitic and calc-alkaline volcanic centers on Umnak island, show major differences in their B, Li and Be systematics. The recent lavas from Okmok are among the most boron enriched in the Aleutian arc, ranging from 15-55 ppm B with B/Be ratios of from 25-50 that show little systematic change in samples ranging from basalt to dacite. 10/9Be ratios in these samples range from 5 to 9 x lO-^l and the l^Be/B ratio is essentially constant in the basaltic samples. All the Okmok basaltic samples fall within error of the ^O/^Be-B/Be correlation for the Aleutian arc of Moms et al (1990; Nature). Li also shows enrichments at Okmok, with Li/Yb ratios of 4-5 in primitive basalts that decline to 2 in rhyolites. By contrast, primitive lavas from Recheshnoi are a factor of 3-4 lower in both boron contents (~5 ppm) and B/Be ratios (9), but higher in their LiA^b ratios (7). Both LiA^ and B/Be ratios increase substantially with increasing differentiation at Recheshnoi. Although B/K,Li/REE, and K/Li systematics of the Recheshnoi rocks imply the presence of a fractionating phase with affinities for REE and K (amphibole?), most of the changes in B/Be and Li/Yb are due to increases in B (5 to 95 ppm) and Li (9 to 47 ppm) too large to be explained by fractional crystallization. , The high B/Be, lO/^Be, and Li/Yb in the primitive Okmok rocks suggest the substantial influence of a fluid bome, sediment derived subducted component during their petrogenesis. Low B/Be ratios in primitive Recheshnoi rocks imply less of a subducted component was available despite its proximity to Okmok (45 km), but large B and Li increases during differentiation point to later inputs of a B and Li enriched component to the magma chamber.

ABSTRACTS

ICOG 7

THE HF-ISOTOPE CONNECTION. VJ.M. SALTERS, J. ENGLAND, and A. ZINDLER, Lamont-

Doherty Geological Observatory of Columbia University, Palisades, N.Y. 10964 U.S.A.

A Nd and Hf isotopic study of oceanic volcanics has been undertaken to further examine the correlation between the two isotopic systems and to determine the I'^^Hf/^'^^Hf characteristics of mantle components as described by Zindler and Hart (1986). Ocean Island Basalts (OIB) form an array on a i^^Nd/i^Nd versus diagram, whereby the mantle endmembers as defined in Sr-Nd-Pb isotopic space can be extended to the Hfsystem. For MORE, however, Hf and Nd isotopes are decoupled and substantial I'^^Hf/^'^'^Hf variation in the mantle source is required. Melting models in which Lu/Hf and Sm/Nd are fractionated during the formation of the MORE reservoir can explain the complete range in Nd and Hf isotopes in MORE, given tiiat variable amounts of both diopside and garnet are present during melting. A model in which MORE-type mantle is mixed with (or veined by) low i76Hf/i77Hf-high i43Nd/i44Nd material is also consistent with the isotope data. The low l76Hf/i77Hf component may be generated in the deep mantie by a low Lu/Hf ratio resulting from die presence of perovskite, which will effectively increase the bulk partition coefficients for Hf and Zr (and Ti). Tests performed on the Lamont Isolab indicate that a hot SIMS technique similar to that developed for Th (Zindler et al., this volume) increases the ionization efficiency of Hf by two orders of magnitude over standard thermal techniques. This permits measurement of i76Hf/i77Hf on 5-lOng of total Hf, and, for the first time, Hf isotopic ratios can be measured directly in mantle peridotites. Most peridotites are depleted in High Field Strength Elements (HFSE) compared to the REE. The technique will be used to measure i'76Hf/i'77Hf ratios in HFSE depleted mantie peridotites to determine age of the depletion and to asses to relationship between peridotites and volcanics.

Sm-Nd, U-Pb ZIRCON AND REE DATA FOR AN ARCHAEAN ACID VOLCAxNO-PLUTONIC SUITE, UKRAINIAN SHIELD, USSR A.V. SAMSONOV. D.Z. ZHURAVLEV and E.V. BIBIKOVA, IGEM of the USSR Acad. Sci., Staromonetny 35. Moscow 109017, USSR Isotope-geochemical studies of acid volcanics in greenstone belts and adjacent tonalite-trondhjemite plutons within the Middle Cisdnieprovia granite-green stone terrain have been carried out to look for genetic relationships and to recognise the source of the acid melts. Acid volcanic and plutonic rocks of the Verchovtsevo greenstone belt form a unified calc-alkaline-like trend from dacite/tonalite to plagiogranite/trondhjemite. Conventional U-Pb dating of zircons has >delded discordia upper intercept aees of Ma for the tonalites and 3103±23 Ma for the plagiogranites; that is, they are indistinguishable in age. Sm-Nd whole rock data for the acid volcanic and plutonic rocks approximate to a single isochron of 3117±204 Ma (eT^d(3^7)+1.14±0.8). Dacites and tonalites have the same (chonarite normalised) REE patterns, with moderately fractionated light and heavy REE and small negative Eu anomalies. The REE pattems in the trondhjemites and plagiogranites are less fractionated and have larger negative Eu anomalies than the dacites and tonalites. The obtained data suggest a temporal and genetic relationship between the acid volcanics of the Greenstone sequences and the adjacent plutonic rocks, and they may be regarded as both belonging to a single suite. The REE data suggest that plagioclase fractionation was involved in their genesis, and the positive value suggests that their source wa^mafic material rather than older sialic crust.

PHANEROZOIC CRUSTAL GROWTH IN THE ALASKAN AND CANADIAN CORDILLERA SAMSON. S.D.. Dept. of Geology, Syracuse University, Syracuse, NY 13244-1070, USA, and PATCHETT, P.J., Dept. of Geosciences, University of Arizona, Tucson, AZ 85721, USA Because the North American Cordillera is a mosaic of suspect terranes the isotopic characterization of this major Phanerozoic orogenic belt can only be accomplished by the characterization of the individual terranes that comprise the orogen. Nd and Sr isotopic ratios have been measured on more than 100 samples of the largest terranes including the Alexander, Wrangellia, and Stikine terranes, the Coast Mountains batholith (CMB), and the Gravina belt. Initial cNd and ratios range from -1 to +9.6 and 0.70276-0.70663 for the terrane samples, and range from -3.5 to -0.2 and 0.70494-0.70607 for the Upper Cretaceous-Eocene CMB. Tdm ages range from 340 to 1100 Ma for the sedimentary rocks of the four terranes and range from 620 to 1070 for the CMB. These data demonstrate the isotopically juvenile nature of the terranes and, to a lesser degree, the CMB. Estimates of the mean Nd concentrations and isotopic compositions of the region studied, and of the depleted-mantie, indicate that the terranes are composed of greater than 90% mantie-derived material. The CMB is composed of approximately 60-70% new, mantie-derived crust and 30-40% recycled Proterozoic crust The high END and low TDM ages of the Cordilleran samples are similar to other Pacific margin terranes, such as those in New Zealand, but are in marked contrast to other Phanerozoic orogens such the Himalayas, the Hercynides, and Caledonides. An estimate of the growth rate of the part of the Cordillera studied is about 0.20.25 km^ a-i, or about 20-25% of tiie Cenozoic world-wide crustal growth rate. The formation, amalgamation and accretion of isotopically juvenile terranes has obviously been a very important mechanism of crustal formation tiiroughout Phanerozoic time and may also have been a fundamentally important mode of crustal growth during the Precambrian.

CORRECTIONS TO FISSION TRACK AGES OF NATURAL CRYSTALLINE MINERALS SANDHU. A.S., RAMOLA, R.C., SINGH, S., VIRK, H.S. Department of Physics, Guru Nanak Dev University, Amritsar143005, India. The cr>'stalline mineral track detectors are generally used as a tool for dating rocks and for revealing geothermal history of an area. It is now well established that the fission tracks in terrestrial materials get shortened partially or completely owing to geothermal annealing. This results in a younger apparent age of the sample concerned. Thus, in order to obtain correct age results, an annealing correction must be applied. Currently, tiie two age correction techniques viz: the track size and the age plateau are generally employed. In the present work we propose a new method to correct the apparent fission track ages of crystalline minerals.

HELIUM AND CARBON FLUXES IN LAKE NYOS, CAMEROON: CONSTRAINT ON NEXT GAS BURST SANO. Yuji, Lab. for Earthq. Chem., Univ. of Tokyo, Bunkyo-Ku, Tokyo, Japan, KUSAKABE, Minom, Inst, for Study of Earth's Interior, Okayama Univ., Misasa, Tottori, Japan, HIRABAYASHI, Jun-ichi, Kusatsu-Shirane Volcano Obs., Kusatsu, Gunma, Japan, NOJIRI, Yukihiro, National Inst, for Environ. Studies, Tsukuba, Ibaraki, Japan, SHINOHARA, Hiroshi, Dept. Chem. Tokyo Inst, of Tech., Meguro-Ku, Tokyo, Japan, NJINE, Thomas, Lab. of General Biol., Univ. of Yaunde, Yaunde, Cameroon, TANYILEKE, Greg, IRGM/CRH, Yaunde, Cameroon On 21 August, 1986, a lethal gas burst issued from Lake Nyos in Cameroon, westem Africa, killing at least 1700 people. Altfiou^ a consensus has that the victims died of asphyxiation by CO2 of magmatic origin, it is still unknown how frequent such catastrophic degassing occurs at the lake. The CO2 flux at the bottom of Lake Nyos is estimated based on the measurement of ^He/^He and 4jjg/20Ne ratios and chemical compositions in gases exsolved from the lake water. A calculated CO2 flux at the bottom of Lake Nyos is (4.4±2.3) x 10^3 cm^STP/year. Combined with the estimated release in the August 1986 event of (8±2) x lO^^ cm^STP, the CO2 flux suggests that the gas burst may happen about once in 18±10 years, although the significant uncertainty should be taken into account for the frequency resulting from assumptions such as steady-state fluxes in the lake and small fractionation of the OHe ratio during the degassing event. Several yeariy measurements of hypolimneiic fluxes and seasonal measurements of epilimnetic fluxes are needed to betterfixthe recurrence interval. In addition, regional ^He/^He survey of carbonated mineral springs in northwestem Cameroon is reported in the view of geotectonics.

WEATHERING IN HIMALAYAS : INFERENCES FROM CHEMICAL AND ISOTOPIC ANALYSIS OF HIGH ALTITUDE STREAMS M . M . SARIN, J.R.TRIVEDI, S .KRISHNASWAMI, Physical Research Laboratory, Ahmedabad 380009, India, and K.K.SHARMA, Wadia Institute of Himalayan Geology, Dehra Dun 248001, India. We have made systematic measurements of Sr, Ra, U isotopes and major elements in a number of Himayalan streams : BhagiraM, Bhilangna, Alaknanda and their tributaries (Ganga source waters) draining at altitudes ranging from 600 to 6750 m. The total dissolved salts (TDS) vary in the range 40 to 245 mg/1. The high TDS content (80 mg/1) in the Bhagirathi at Gangotri (near its source), is a reflection of intense chemical weathering in these regions. Calcium and magnesium are ttie major cations (-85% of TZ"^) and bicarbonate accounts for nearly 2/3 of the anions, m most of the streams, sulphate is quite significant in the anion balance and substantially higher than chloride. In the Bhagirathi and Kedarganga at Gangotri, sulphate accounts for as much as 80% of the anions and Ca:S04 ratio is 0.8. These observations suggest that major ion chemistry of ttiese streams is dominated by weattiering of carbonates, the high conc. of SO4 results from dissolution of gypsum and/or oxidation of pyrites associated with carbonates. The high conc. of silica (100 to 400ymole/l), Na excess over a and (Na+K):TZ+ ratio of 0.15 suggest that at least 15% of tiie cations in these waters are derived from silicate weathering. This is also reflected in high radiogenic ^'^Sr/^^Sr ratios, 0.7330 - 0.7984, some of the highest observed for surface waters. Although the Sr/Ca ratio in the Bhagirathi main stream is dominated by carbonates, the high radiogenic Sr requires weathering of Rb rich (low Sr) silicates. The 238u conc. in most of these streams centres around 2yg/l and 234u/238u activity ratio is close to unity, consistant with the intense chemical weathering of sedimentary terrains.

ICOG 7

ABSTRACTS

CORRELATED ENRICHMENT OF URANIUM AND ORGANIC CARBON IN AN EASTERN ARABIAN SEA CORE DURING THE LAST GLACIAL MAXIMUM (LGM) A . S A R K A R , S.K.BHATTACHARYA, and M . M . S A R I N Physical Research Laboratory, Ahmedabad 380009, India. A d e t a i l e d a n a l y s i s of in planktonic foraminifera. CaC03(%), total organic carbon and i t s S ^ ^ C a l o n g w i t h uranium content h a s been done in a core from the e a s t e r n A r a b i a n sea ( 1 0 ° N . 72°E\ water d e p t h : 2 5 2 3 m ) . The S ^ ^ O s t r a t i g r a p h y and ^^C ages establish an u n d i s t u r b e d s e d i m e n t a t i o n ( 2 . 2 c m / k y r ) w i t h LGM located at 32 cm ( S a r k a r et a l , N a t u r e , 343, p.549, 1990). The total productivity during glacial times was lower ( i n d i c a t e d by l e s s CaC03) due to a weaker summer monsoon. This also e x p l a i n s h i g h e r (by 2 ° / o o ) v a l u e s of g l a c i a l ( o r g a n i c ) due to l e s s e r input of terrigenous organic matter r e s u l t i n g from reduced river discharge. The uranium ( U ) content i s not constant throughtout the c o r e ; two depth zones have c o n c e n t r a t i o n s h i g h e r by s e v e r a l f a c t o r s (at /«22kyr and ^ 5 0 k y r ) . These l e v e l s are a l s o c h a r a c t e r i s e d by h i g h o r g a n i c c a r b o n , C ( o r g ) , content. The U-C(org) c o r r e l a t i o n i s probably indicative of r e l a t i v e l y a n o x i c condit i o n s during those p e r i o d s l e a d i n g to b e t t e r p r e s e r v a t i o n of organic matter and a s s o c i a t e d uranium. Post d e p o s i t i o n a l d i a g e n e s i s does not seem to be responsible for the high U-C(org) values based on nearly uniform c o n c e n t r a t i o n of Mn w i t h d e p t h .

NOBLE GASES IN C A R B O N A T I T E S A N D THEIR IMPLICATIONS ON THE S U B C O N T I N E N T A L UPPER M A N T L E SASADA, T ^ and H. H I Y A G A O N , Geophysical Institute, U n i v e r s i t y of Tokyo, Bunkyoku, Tokyo 113, Japan. Most of geochemical information about mantle available at present is r e s t r i c t e d to oceanic areas, i.e., data of MORB and OIB. To obtain geochemical information of subcontinental upper mantle, we have measured isotopic composition of noble gases in carbonatites. Carbonatites w o u l d be suitable samples for geochemical study of subcontinental upper mantle because (i) they are of mantle origin (juding from Nd, Sr and Pb data{l)), (ii) they are mostly r e s t r i c t e d to continental areas, and (iii) wide v a r i a t i o n in their ages (from SOOOMa to the present) could provide useful information about geochemical evolution of the s u b c o n t i n e n t a l mantle. First, we analyzed five carbonatite samples, two from Africa and three from Canada. P r e l i m i n a r y results show they are enriched in radiogenic c o m p o n e n t s such as ^He, ^Âr and ^-^Û-f issiogenic Xe, which, unfortunately, cover up the original signatures of the mantle. Secondly, we separated pyroxene, which has low c o n c e n t r a t i o n s of K, U, etc., and analyzed noble gases in it. A preliminary analysis of pyroxene shows relatively low ^Âr/^Âr ratio of about 320, suggesting radiogenic components were effectively removed. In considering high retentivity of n o b l e gases in pyroxene, this low ratio may represent the original signature of the s u b c o n t i n e n t a l mantle. If this is the case, the subcontinental mantle is isotopically different from MORB-type upper mantle. (1) K. Bell and J. Blenkinsop (1987), Geochim. C o s m o c h i m . Acta, 291-298.

ABSTRACTS

ICOG 7

ISOTOPIC AND CHEMICAL COMPOSITIONS OF PRECIPITATION COLLECTED EVERY 12 HOURS THROUGHOUT ONE YEAR IN TOYAMA, JAPAN SATAKE, Hiroshi and SAKAMOTO, Noriko, Department of Earth Sciences, Toyama University, Toyama 930, Japan Toyama is located in the northern part of central Honshu, Japan, facing the Sea of Japan. Precipitation samples were collected every 12 hours during December 1985 to April 1987. 6D and 6^®0 values of these samples range from -139 to 31 % o and -19.4 to 3.9 %o respectively. They do not show any seasonal variation. The d-parameter (d=6D-86^®0) is an important index for characterizing the vapor origin. The d-parameter observed in Toyama ranges from 0 to 40 % o . This parameter is constant (0 to 10 %o ) in April to August. It increases monotonously to 40 %© from September to December and decreases from January to March. Thus the d-parameter shows seasonal variation. In the Sea of Japan, the temperature of the sea water is higher than the atmospheric temperature in September to March. The seasonal variation of the dparameter shows good correlation to that of the temperature difference between the sea water and the atmosphere. The d-parameter changes from 0 to 40 %o with the increase of the temperature difference up to 10 This correlation may be explained as follows. When the atmospheric temperature is lower than the temperature of the sea water, the air on the sea will be warmed and decrease its relative humidity. The degree of decrease in the relative humidity will be enlarged with the larger temperature difference. The evaporation of the sea water to the air having lower humidity makes the larg'er d-parameter of the vapor (Merlivat and Jouzel, 1979). Thus d-parameter may correlate with the temperature difference. The similar variations of the d-parameter were observed i n the offshore region of the Far East (IAEA, 1983; Liu, 1984). This seasonal variation may be the essential characteristic of the precipitation in this region.

Rb-Sr, Sm-Nd AND K-Ar DATA ON GRANITOIDS FROM THE PACIFIC COAST OF MEXICO BETWEEN 21* AND 17®N. SCHAAF. P.. KdHLER. H.. MOLLER-SOHNIUS. D. & v.DRACH. v.. Mineralogisch-Petrogrophisches Institut der Universitat, Thersienstr. 41, 8 0 0 0 MUNCHEN 2, Germany. Following the coast from NW to SE. the four bathoUths of Punta de Mita (Pta.M.), Puerto Vallarta (PV). Manzonillo (Mzo) and Jilotlan (Jin) were investigated. The granitoids are considered to be the roots of a magmatic island arc system. The oldest rocks known in this area are jurassic sediments. Results are given in the following table: NW Mzo. PV Jin. Pta. M. KOMPLEX 88±2 68±4 62±7 INTRUSION (MQ) 92I9 ®^Sr/®®Sr|n .70511i8 .70471113 .7034914 .7037614 85 59-50 Hbl.{K-Ar) (MQ) 58-73 Bi (K-Ar) (MQ) 55-55 83 64-57 60 Bi (Rb-Sr) (MQ) 48-55 75-84 59-59 55-59 C-Nd In -2.3to 0 -45to+1.8 ^3to+5 +2.7to-^3.7 T-NdDM (Mq) 1170-500 1500-500 360-250 510-310 Cooling rate CC/MQ) - (110) - 70 - 130 - 160 (600-300-C) DM pQrQmeters Qre deduced from ultrQmQfic rocks from the Trans MexicQn VolcQnic Beit.

Slightly negative e-Ndjn-values and Nd-model ages from Pta.M.- and PV-complexes indicate an old unknown basement. HoA'ever, Nd data from Mzo - and Jln.-plutons suggest younger basement rocks and less crustal contamination. A similar trend can be seen from NW to SE in decreasing intrusion ages and ®^Sr/®®Srjn- values. Locally. Pta.M. granitoids display gneissic texture. The differences between intrusion- and related mineral ages give evidence of metamorphism ( s 5 0 Ma), probably due to strike- slip displacements. The smaller plutons of Mzo. and Jin. show high cooling rates caused by a rapid regional uplift.

AGE-SIGNIFICANCE OF U-PB SYSTEMATICS IN IMPACT-SHOCKED ROCKS SCHARER Urs, Laboratoire de Gdochronologie, University Paris Vn, 2 place Jussieu, 75251 Paris Cedex 05, France, and DEUTSCH Alex, Institut fiir Planetologie, Universitat Munster, Wilhelm-Klemm Str. 10, D-4400 Munster, W-Germany. In order to evaluate the physical conditions required to reset the UPb chronometer in impact shocked rocks, naturally and experimentally shocked minerals were investigated in combination with measurements on unshocked counterparts. For both studies, minerals rich in U and radiogenic Pb were selected, representing phases highly susceptible to reveal shock-induced U-Pb fractionation. For the experiments, Proterozoic zircon, titanite and monazite were shocked to 35, 47.5, and 59 GPa, whereas naturally shocked zircon and monazite were extracted from (50-60 GPa) shocked crystalline fragments of the Haughton crater, Canada. In experimentally and naturally shocked samples post-shock temperatures are similar, lying above 1200° C. For all three types of minerals, shock-waves produce drastic changes in the crystal lattices causing planar structures, mosaicism, strong lowering of birefringences, turbidization, and decolorization of the grains. Moreover, X-ray patterns indicate transition of the crystals into polycr>'Stalline aggregates of < 10 um block-size. These shock-phenomenons are more pronounced in experimentally shocked grains. Detailed U-Pb analyses show that neither experimental or natural shock-waves do not fractionate Pb-isotopes within the analytical precision of ± 0.1 and for chemical fractionation, no evidence in favour of shock-induced U-Pb fractionation is observed for monazite. Small degrees (< 2 %) of discordancy in shocked zircon and titanite are due to time-integrated Pb-loss in the rock, and not to the shock event. In consequence, the data substantiate that shock-wave metamorphism up to 60 GPa/>1200° C, followed by rapid cooling of the shocked phases does not measurably effect the U-Pb chronometer. Primary ages and pre-shock discordancy patterns are well preserved. This observation questions the view that shockwave metamorphism in rapidly cooling systems produces significant U/Pb fractionation i.e., that lower intercept-ages of discordant U-Pb patterns (concordia diagram) date shock events in moderately to highly shocked systems. ND-SR-PB ISOTOPIC VARIATIONS ALONG THE GULF OF ADEN: EVIDENCE FOR MANLE PLUME-CONTINENTAL LITHOSPHERE INTERACTION. J-G. SCHILLING and B. B. HANAN, Graduate School of Oceanography, University of Rhode Island, Kingston, RI, USA We report on REE and Nd-Sr-Pb isotopic composition of basalts along a 1700-km long profile in the Gulf of Aden-Gulf of Tadjoura-Ardoukoba Rift in East Afar. A 800-km long isotopic and La/Sm gradient, sinusoidally modulated, is apparent from the Afar eastward. The first enrichment peak occurs in the Gulf of Tadjoura where diffuse extension of the Danakil-Ashia continental lithospheric block and westward rift propagation is progressing. The second enrichment peak at 46°E is associated with a mantle buoyancy anomaly and related constructional volcanism. East of 48°E, the MORBs are typically light REE depleted, whereas 206pb/204pb and 87sr/86sr slightly increase, suggesting recent decoupling. In Nd-Sr-Pb isotope ratio space, 3 distinct mixing vectors are observed which point towards mantle source endmembers interpreted as continental lithosphere along the the Gulf of Tadjoura (a hybrid EM1-EM2), a mantle plume (HiMU-like) which dominates the 46°E anomaly, and the depleted asthenosphere east of 48°E (DUPAL-like). Combined data for the Gulf of Aden-Red Sea-Afar-Ethiopian rifted zones suggest a radial pattern of geochemical and isotopic variation about the Afar. A dynamicalthermal model is presented. It invokes both passive rifting/seafloor spreading in the Red Sea-Gulf of Aden, and the flattening and interaction of the head of a torus-like thermal mantle plume with the Pan-African continental lithosphere which is slowly moving northward. The thinning of the lithosphere by thermal and advective erosion does not go to completion. A ternary mixing model indicates that within the 800-1000 km radius of influence of the Afar plume, melting of the lithosphere and the asthenosphere dominates initially. Only along the 3 rifting zones intersecting the ring composed of original HiMU-like plume material, 450 ±150 km in radius, does the plume play a more dominant role. We further suggest that the stem-feeder of the plume has now been tilted and deflected towards the Afar as a result of the northward migration of the Gulf of Aden/Red Sea spreading centers which act as sinks, and the likelihood that the plume is encountering an Afiican lithosphere increasing in age, thickness and rigidity.

90 Sm-Nd AND U-Th-Pb ISOTOPIC EVIDENCE FOR THE SOUTHWARD EXTENT OF EARLY ARCHAEAN CONTINENTAL CRUST IN NORTHERN LABRADOR SCHI0TTE, L., SHIREY, S.B., CARLSON, R.W., Carnegie Institution of Washington (DTM), Washington, D.C. 20015, USA, HANSEN, B.T., Institut fiir Mineralogie, D-4400 Munster, West Germany, NOBLE, S., Royal Ontario Museum, Toronto, Ontario M5S 2C6, Canada, and BRIDGWATER, D., Geologisk Museum, K0benhavn 1350 K, Denmark The Labrador coast north of Nain is dominated by gneisses affected by 2.7 to 2.8 Ga granulite facies metamorphism. Although the whole rock Rb-Sr, Sm-Nd and U-Th-Pb isotopic systems were severely dismrbed during this event, it has been possible to trace early Archaean continental crust down to the Okak area, about 100 km south of the type area of the 3.73 Ga Uivak gneiss. Samples from a selected outcrop of gneiss that was partially remelted during the late Archaean granulite facies metamorphism yield a late Archaean Sm-Nd isochron with an initial €am of about -6. This un-radiogenic Nd, together with high amounts of unsupported radiogenic Sr and a large spread in 207Pb/204Pb compared to 206Pb/204Pb shows that the continental crust here had a pre-metamorphic history of several hundred million years. On close inspection, the Pb isotopic signature is different from that of the Uivak gneisses further north. The pre-granulite facies U/Pb ratio varies on outcrop scale from very low to very high when compared with mantle or bulk earth Pb evolution. The Pb also developed in a low ThAJ environment (ThAJ ca. 2) before the granulite facies event, possibly an indication of a higher proportion of metasediment in the parent rock. This Pb signature could serve as a tracer for the extent of this rock complex. Abundant mafic orthogneisses in the Okak area have more juvenile Nd and Pb isotopic compositions than the early Archaean gneisses. Single zircon U-Pb age dating of these mafic gneisses shows that they are the southward extension of the ca. 3240 Ma Lister gneisses. They were emplaced as granulites and were partly retrogressed during emplacement of leucocratic granitoid sheets at ca. 2990 Ma. Thus their granulite facies mineralogy is without direct relation to the regional metamorphism that took place between 2.8 and 2.7 Ga.

GEOCHRONOLOGIC AND ISOTOPIC STUDIES (RB/SR, U/PB, SM/ND) ON POLYMETAMORPHIC HP-ROCKS FROM THE BLACK FOREST. FRG. SCHLEICHER, Helmut, Mineralogisch-petrograph.lnst., Albertstr.23b, D-78 Freiburg, Fed.Rep.Germany Rb/Sr, U/Pb and Sm/Nd data are reported for eclogitic amphibolites from the pre-Hercynian basement of the Black Forest (FRG). Subsequent to their igneous formation these rocks suffered a polyphase metamorphic history starting with a HP-HT metamorphism (eclogitic stage) and ranging down via a MP-HT (granulitic) stage and a LP-HT metamorphism to a LP-LT retrograde evolution connected with the uplift history of the basement (Stenger et al.1989). The most dominating LP-HT metamorphic stage has overprinted the preceeding HP and MP events and can be correlated to the climax of the Variscan orogeny as proven by concordant U/Pb monazite ages {325-335 Ma, Werchau et al.1989) from various metamorphites. Sm/Nd Tdm ages (1.2-2.4 Ga) as well as some U/Pb zircon ages indicate a Proterozoic formation of the protoliths. An Ordovician age for the eclogite facies HP event is supported by upper intercept ages of 460-490 Ma for zircons from a number of eclogite occurrences. Sm/Nd isochron ages, however, point to a scatter in age between 327-390 Ma for the closure of the HP mineral paragenesis (V^erchau 1989). This feature is postulated to indicate time differences in tectonic uplift for the individual eclogite bodies during Variscan thrust tectonics. This is consistent with Rb/Sr results and secondary U/Pb zircon ages. A very detailed study including U/Pb, Rb/Sr and Sm/Nd dating as well as a Sr isotopic profile was done on a special occurrence (Silberberg). The results give an insight into a very complex polymetamorphic history with major events at about 2 Ga, 490 Ma, 390 Ma, and 340 Ma. The detailed petrological and geochemical investigations give evidence for a complete resetting of the Rb/Sr whole rock system in the course of Variscan shear tectonics. Stenger et al.(1989): Tectonophys. 157,117. Werchau et al. (1989): Beih.Europ.J.Min. 1,No.1. Werchau (1989) pers.comm.

ICOG 7

ABSTRACTS

ISOTOPIC COMPOSITION OF ORGANIC AND INORGANIC NITROGEN DURING SEDIMENTATION AND DIAGENESIS SCHOLTEN/ S v e n a n d K R E U L E N , R o b , D e p a r t m e n t of G e o c h e m i s t r y , U n i v e r s i t y of U t r e c h t , P . O . B o x 8 0 . 0 2 1 , 3508 T A U t r e c h t , t h e N e t h e r l a n d s Three different nitrogen fractions were studied on samples from c l a y , organic-rich sediments a n d s h a l e s . T h e s e f r a c t i o n s are f i x e d (Nî^), e x c h a n g e a b l e (N^J a n d o r g a n i c n i t r o g e n (N^rg). T h e e n v i r o n m e n t of d e p o s i t i o n s e e m s to h a v e a p r o n o u n c e d e f f e c t o n f i x e d n i t r o g e n : on a v e r a g e , (S^Nî^ r a n g e s f r o m <3Voo in N W E u r o p e a n f r e s h w a t e r c l a y s t o >5Voo Atlantic deep-sea c l a y s . T h e o b s e r v e d s y s t e m a t i c i n c r e a s e of with the depth at which sedimentation t o o k p l a c e m u s t r e f l e c t t h e c h a n g e s in a v e r a g e i s o t o p i c c o m p o s i t i o n of t h e d i f f e r e n t n i t r o g e n f r a c t i o n s in t h e w a t e r c o l u m n . In t w o h o m o g e n o u s , o r g a n i c - r i c h t u r b i d i t e s f r o m t h e M a d e i r a A b y s s a l P l a i n , w h i c h are partly altered by oxidation, organic matter has b e e n d e c r e a s e d a b o v e t h e o x i d a t i o n f r o n t s by 7 5 % . It appears that oxidation has a strong e f f e c t o n <!>"N«rg and H o w e v e r , 5"N,i^-values show only a small change from 5.7Voo to 5.2Voo upon oxidation. T h e e f f e c t of m a t u r a t i o n on t h e n i t r o g e n f r a c t i o n s w a s s t u d i e d o n s a m p l e s from t h e Kimmeridge and Posidonien shales. With i n c r e a s i n g t e m p e r a t u r e an i n c r e a s e of b o t h N^^^c o n t e n t and (S^Nî, (up t o 3Voo) h a s been o b s e r v e d . T h e a m m o n i u m i n v o l v e d is p r o b a b l y d e r i v e d from d e c a y i n g o r g a n i c m a t t e r , t h o u g h no c l e a r t r e n d of 5"N„rg w i t h m a t u r a t i o n w a s f o u n d .

COSMIC RAY EXPOSURE AGES OF H3- AND H4-CH0NDRITES SCHULTZ, L., WEBER H. W., and BEGEMANN F. Max-Planck-Institut fiir Chemie, D-6500 Mainz, FRG Many cosmic ray exposure ages of chondrites are based on noble gas analyses which were carried out more than 20 years ago. Since that time experimental techniques have been considerably improved and production rate adjustments (e.g. shielding corrections) are better known. To obtain a more reliable distribution function of exposure ages we have measured the concentration and isotopic composition of He, Ne and Ar in 48 H3- and H4-chondrites. Only meteorites which were observed to fall were analysed to avoid secondary effects like weathering. Together with 15 analyses taken from the literature this investigation covers all known H3 and H4 falls. Taken together, the H-Chondrite Falls exposure age distribution (see Fig.) shows the well g known cluster around 8 J Ma. This peak is inter- -g • H3 prated as a major colli- ^^ • H4 sional event which, at ^ that time, broke up a considerable part of the Hgroup parent body. Note, Exposure age (Ma) however, that the peak is entirely due to the more evolved H4 group; about 40% of the H4 falls are found in this peak but only one out of 13 H3's has an exposure age of about 8 Ma. This observation speaks against the derivation of both H3 and H4 chondrites from a single parent body with onion-shell structure wh.re the least metamorphosed H3-group resides at the surface.

ABSTRACTS

ICOG 7

PALEOATMOSPHERIC ARGON IN OLIVINE FROM AN ARCHEAN KOMATUTE AND IN MAGNETITE FROM A PROTEROZOIC BANDED IRON FORMATION SCHWARTZMAN. David W., Dept. Geology & Geography, Howard University, Washington, D.C. 20059, U.S.A., SUTTER, John F., U.S. Geological Survey, MS-908, Reston, VA 22092 U.S.A., and NISBET, Euan G., Dept. Geological Sciences, University of Saskatchewan, Saskatoon, Saskatchewan S7N OWO, Canada Measurements of the 40Ar/36Ar ratio (R) of the atmosphere in the geologic past would provide a powerful constraint on the degassing history of the Earth, since virtually all of the Âr in the atmosphere and solid Earth is a produa of the decay of ^ K while the 36Ar is primordial. Olivinefroma Zimbabwean komatiite (age = 2.7 Ga) and magnetitefromthe Biwabik Banded Iron Formation (BIF), U.S.A. (age =1.9 Ga) were analyzed using the 40Ar/39Ar incremental heating method, with data interpretation on an isotope correlation (isochron) diagram (36Ar/40Ar versus 39Ar/40Ar). Three high temperature steps (950,1050, and 1 lOOoC) in the olivine analysis yield an R of 148 ± 13 (2a) with an isochron age of 1.76 Ga. Although the komatiite is 2.7 biUion years old, numerous dikes of mid-Proterozoic age intrude the region, possibly causing gain of K and/or gain or loss of 'Âr in the olivine at this time. Since some K may have been present in the olivine at the time of crystallization 2.7 billion years ago, and any argon gained in mid-Proterozoic time would have had a higher R than atmospheric argon at 2.7 Ga, the R of 148 obtainedfromthe olivine is a maximum value for atmospheric argon dissolved in the komatiite upon eruption 2.7 billion years ago. An apparent isochron formed by the 1050,1075, and llOOoC steps from the magnetite analysis yields an R of 96 ± 19 (2a) and an age of 3.35 Ga, presumably a record of an inherited detrital phase in this midProterozoic BIF. The inferred paleoatmospheric ÂiftÂi ratios obtained in this study are consistent with catastrophic degassing of 36Ar early in Earth history with significant recycling of sialic cmst since the Archean.

Rb-Sr AGE OF THE PRECAMBRIAN-CAMBRIAN BOUNDARY IN GARHKAL HIMALAYA, INDIA. SHARMA,K.K, RAMESHWAR RAO,D, AZMI,R.J, Wadia Institute of Himalayan Geology, Dehra Dun; GOPALAN,K, and PANTULU,G.V. C, National Geophysical Research Institute, Hyderabad, India. A 3m thick phosphorite band in the Tal Formation, Garhwal Himalaya is believed to mark the PC-C boundary following the discovery within it of oldest small shelly faunal association highly correlatable with that of Zone I of Meishucunian Stage in China, the pre-Tommotian Manykaian Stage in Siberia and the Yukon Territory of Canada (Azmi, 28th IGC Abstracts 1, 1989). We report a strict upper limit to the age of this boundary based on the Rb-Sr dating of carbonaceous black shales directly underlying the phosphorite horizon. 9 whole-rock shale samples from a 15-25cm thick band from the Sukhet Block of Maldeota phosphorite mine near Dehra Dun contain mainly illite with very little detritus. Rb and Sr analysis of these sam.ples gave a well-defined isochron corresponding to an age of 624 + 10 (2'ar)Ma and Sr i of 0.7093 +1 (2 ~ ) . This age most probably marks the time of early diagenesis. As the black shale directly underlies the phosphorite boundary layer, the above age is either very close to the age of the PC-C boundary or a strict upper limit to the same. The marginally younger age of 610 + 10 Ma for the PC-C boundary in the Yangtze and Yunnan regions in China is based on extrapolation of younger Rb-Sr ages of black shales lying about 20m above the boundary (Xing et al., Geol. Mag. 121, 155, 1989) .

91 UNSCRAMBLING SHALES: THE OXYGEN ISOTOPE MINERALOGICAL APPROACH SHEPPARD. Simon M.F. and TAIEB, Raouf, Centre de Recherches Petrographiques et Geochimiques, BP 20,54500 Vandoeuvre-les-Nancy, France Shales and other clay minerals rich rocks, constituting more than 60 % of the world's sediments, are actually represented by 9 whole rock O-isotope values 1 This situation reflects in part analytical problems related to the presence of organic matter, carbonates (calcite, dolomite, siderite), etc., often in more than trace quantities. Whole rock b^K) values are needed for mass balance calculations. Thirty whole rock non-metamorphosed Mesozoic shales (Paris basin, Tunisia, Columbia) have b^Ô SMOW values of 13 to 24 derived from the analysis of the calcite (b^Ô = 23 to 30 %o, b^^C = - 8 to + 3 %c PDB) and calcite - free organic matter - free shale (b^Ô = 13 to 22 %o). Calcite is systematically iÔ-enriched compared to the bulk silicate fraction. For a series of samples from a given sedimentary formation, bulk silicate values combined with X-ray and/or chemical analyses can be applied to (1) determine the scale of homogenization, and (2) if the individual samples behaved as closed systems, estimate the S^Ô values of the two principal mineral constituents - clays and quartz (not necessarily a single population). For the Jurassic Paris and Cretaceous Tunisian basins b^Ô bulk silicate values tend to decrease with increase in quartz content. This implies that (1) S^^G detrital quartz was lower than b^Ô detrital clay minerals, and (2) isotopic homogenization has not been closely approached on the scale of our sampling (metric to hectometric) for temperatures of diagenesis up to 180°C. For low degrees of diagenesis (T < 80-100°C), isotopic exchange among coexisting silicates is generally minor. From the b^Ô bulk silicate-% quartz relations, the b^Ô values of the detrital quartz and clay minerals can be estimated and used to characterize the provenance of these minerals.

EVOLUTION OF ARCHEAN CRUST: EVIDENCE FROM Ce AND Nd ISOTOPE RATIOS AND REE ABUNDANCES Hiroshi SHIMIZU and Akimasa MASUDA Dept. of Chemistry. Fac. of Science, The Univ. of Tokyo, Kongo, Tokyo 113, Japan La-Ce and Sm-Nd isotopic data and REE abundances are reported for the Archean Amitsoq gneisses of West Greenland, Archean ironformations in the Isua and Malene supracrustals of the same area and Archean cherts in the Pilbara Block of Western Australia. These data will provide valuable information on the crustal evolution at the early stage of the Archean in light of REE geochemistry. Among the Amitsoq gneisses, the grey gneiss (3.82 Ga) in the Godthaab area is considered to have been derived from a chondritic source with (La/Ce)j^=l.l and (Nd/Sm)j^= 0.99, while it is concluded that the grey gneisses and the banded gneiss (3.71 Ga) in the Isukasia area were formed from a LREE-depleted source with (La/Ce)N=0.6 and (Nd/Sm)N=0.97. Initial ratios at 2.8 Ga for iron-formations in the Malene supracrustals give (La/Ce).90 and (Nd/Sm)i^=0.99 for the sources between 4.56 and 2.8 Ga; i.e., a La-depleted pattern or a pattern with a small positive Ce anomaly. Similar REE features are indicated for the sources of the Isua iron-formations. The isotopic data for the Amitsoq gneisses and the iron-formations from West Greenland suggest that sources for these rocks had a chiefly chondritic or rather mantle-like depleted nature. These results may imply poor development of continental crust enriched in LREE at the Archean West Greenland. Ce and Nd isotopic data for the Archean cherts from the Pilbara Block suggest that sources had a REE pattern with (La/Ce)p^ ratio larger than unity.

CONSTRAINTS ON CONTINENTAL EVOLUTION FROM Pb and Nd ISOTOPIC STUDIES OF ARCHEAN CRUST SHIREY. S.B., Carnegie Institution of Washington, DTM, 5241 Broad Branch Road N.W., Washington DC 20015. Important goals of Pb and Nd isotopic studies of Archean crust have been to constrain early Earth evolution, identify the mantle sources of continental crust and to facilitate comparisons of Archean continental additions with those in the Proterozoic to Phanerozoic. Nearly all mantle-derived volcanic rocks and the majority of granitoids that make up the Archean cratons have positive initial e^d with a near constant offset relative to CHUR of +2 to +4 from 3.8 Ga to 2.7 Ga. A similar constancy is evident in the Pb-Pb data which show an average model \ii of 8.3 and average model KI of 4.2 with variability increasing for both parameters in the late Archean. The Nd isotopic data require that the mande source for the crust had undergone some form of pre-3.8 Ga depletion. This depletion most likely did not occur in a magma ocean because the positive e^d signature is the opposite of that expected for perovskite fractionation. Separation of pre-3.8 Ga crust of either oceanic or continental composition could have depleted the upper mantle. Oceanic crust is favored for several reasons: 1) it was subductable, 2) it was abundant, 3) it had a longer mantle residence time and 4) it would have a smaller effect on the Pb data. The relatively constant eNd of the depleted mantie from 3.8 to 2.7 Ga requires partial mixing of the depleted mantle with some enriched reservoir. This could have been crustal level contamination of mantle melts during emplacement or deeper level mantle mixing with recycled oceanic crust. Both are difficult to resolve but the latter matches the observed Archean crustal Pb isotopic heterogeneity. Crustal addition via volcanic arc subduction-recycling occurred at some craton margins and often was rapid, involved juvenile crust and left little imprint on the Nd and Pb isotopic systems. The Proterozoic marks a rise in the z^d of the depleted mantle, reflecting permanent removal of the Archean cratons from the upper mantle. It also marks a decrease in the initial ENd of sediments and granitoids, reflecting the recycling of older crustal components that marks Phanerozoic crustal evolution.

ICOG 7

ABSTRACTS

URANIUM-THORIUM DATING OF CALCRETES, GYPCRETES AND FERRICRETES IN ALLUVIUM: INDICATORS OF PALAEOCLIMATIC CHANGE. SHORT. SA.. Environmental Radiochemistry Laboratory, Australian Nuclear Science and Technology Organisation, P.M.B. 1, Menai, N.S.W. 2234, Australia, and NANSON, G.C., Department of Geography, University of Wollongong, P. O. Box 1144, Wollongong, N.S.W. 2500, Australia. We have made chronostratigraphic studies of some major alluvial systems in Eastern, Central and Northem Australia. At the same time we compared thermoluminei^nce dates from coarse quartz grains in clean alluvium with (^^^h-ingrowth) uranium-thorium dates for indurated calcretes, gypcretes and ferricretes developed diagenetically within the alluvia. Calcretes and gypcretes chosen for dating were in the form of indurated rhizomoiphs, nodules or fused horizons (hard pans). Ferricretes were comprised of indurated pisoliths, hard pans or ferruginised fossil wood ^ e r e necessary a correction for detrital, allogenic, uranium and ^-^^h was employed. Such accumulations usually form quite rapidly in alluvium: at least 1 mm per 100 years. Similar ages at some locations suggest the climatic and hydrologic conditions which favour calcrete formation will also favour, where local chemistry is favourable, gypcrete and/or ferricrete formation. These minerals did not accumulate during the last two glacial maxima at -18 ka and -135 ka BP. Major accumulation phases were usually associated with, but actually followed, the periods of greatest fluvial, and therefore probably pluvial, activity. More arid conditions are apparently a general prerequisite for maximum accumulation and induration. The only exceptions are small calcrete, gypcrete and ferricrete nodules which formed after the last glacifi maximum. Thermoluminescence dating indicates this period has not yet been one of major fluvial activity. These accumulations are probably destined for chemical reworking during the next major fluvial period. We suggest that U/Th dating of indurated calcretes, gypcretes and ferricretes in low relief alluvial plains can give useful chronological indication of major palaeoenvironmental changes from wet to more arid conditions in the Late Quaternary.

A MULTI-DETECTOR RESONANCE IONIZATION MASS SPECTROMETER FOR Re-Os ISOTOPIC STUDIES OF CRUST-MANTLE EVOLUTION S H I R E Y . S.B., CARLSON, R.W., BROWN, L. and LAMBERT, D.D.,Carnegie Institution of Washington, DTM,5241 Broad Branch Rd., NW, Washington, DC 20015

Xes-XeN AGE OF URANIUM VEINS IN ERZGEBIRGE DEPOSIT, GERMANY SHUKOLYUKOV, Yu.A., MESHICK. A.P., SHARYFZADE, H.B., Vemadsky Institute of Geochemistry and Anal>tical Chemistry, USSR Academy of Sciences, Kosygin str. 19 Moscow, USSR

A resonance ionization mass spectrometer (RIMS) consisting of a 20Hz Nd YAG-tunable dye laser, a nine-inch radius magnetic sector, a 30 kV electrostatic magnifying stage, and a detector array of three separate continuous-dynode electron multipliers has been constructed for Re-Os isotopic analysis. This RIMS instrument employs atomization by either a thermally-pulsed, miniaturized filament or a pulsed Cs+ sputter gun. Ionization of Re and Os is achieved by a two-photon process at 297.6 nm and 290.9 nm respectively. The multipliers have a gain >10^ and use A1 conversion dynodes fixed at the local 30 kV potential. By magnifying the image (3.5 X) with the electrostatic lens to a lateral separation of 5 mm at the second focus, the multipliers can be positioned close enough together to allow for the simultaneous detection of adjacent Os isotopes. The high-voltage lens has other advanages: 1) supplying higher energy ions and 2) making the detectors less susceptible to stray electrons and ions. Data are taken in a three-step peak hopping mode to eliminate multiplier gain differences while collecting ions from three isotopes of Os per step. The t}T)ical Os efficiency of 5 X 10-"'is limited primarily by the low laser duty cycle. The multiple detector array serves to minimize this inefficiency by allowing collection of data from ^^^Os for 85% of the analysis time. This RIMS instrument has demonstrated an accuracy of better than 1% for i^^Os/iseQs on two solution standards at low abundance levels and seven terrestrial osmiridiums in interlaboratory and inter-technique comparisons. Studies are underway on layered intrusions (Stillwater, Skaergaard, Bushveld, Great Dyke) to determine the role of crust vs mantie sources for parental magmas, Archean to Proterozoic komatiites to define the Os isotopic evolution of the mantle and kimberlites and lamproites to understand the history of melt removal from the subcontinental mantle.

The studied uranium veins are located in the exocontact of a granitic intrusion which cut the complicated volcano-sedimentary sequence metamorphosed in greenschist facies. Generally, veins of the Erzgebirge deposit consist of two U-bearing formations. The first one which is more abundant can be divided into quartz - coffinite pitchblende and calcite-pitchblende mineral assemblages (uqk). The second U-bearing formation is composed of dolomite-pitchblende and pitchblende - selenite mineral associations (eb). U-Pb isochron ages of uqk and eb are 260-270 Ma and 140-160 Ma respectively. Forty five pitchblendes (mainly from uqk veins) were sampled even homogeneously from all explored parts of the deposit (2000m deep, 800m wide). Xes-Xen ages obtained vary from 120 to 270 Ma, with maximum of the distribution corresponding to 160 Ma. Several pitchblendes were dated by the U-Pb isochron method. The U-Pb and Xes-Xen ages are concordant. The relationship between Xe^-Xen ages and occurrence of U-veins and space correlation with 5^'^S values in pyrites confirm the Xes-Xen ages are true. The isotopic 3ata obtained suggest that in the ore of Erzgebirge deposit prolonged intravein metasomatism may have occurred which may contribute to the dissolution and redeposition of uranium minerals.

ABSTRACTS

ICOG 7

INFRA RED SPECTROSCOPY AS A ROUTINE M E T H O D FOR T H E S E M I - Q U A N T I T A T I V E DETERMINATION OF T H E CHEMICAL COMPOSITION OF APATITES FOR FISSION TRACK ANALYSIS. R. SIDDALL and M. J. MENDELSSOHN. Dept. of Geological Sciences, University College London, Gower Street, London. U. K., WCIE 6BT. A quick method of determining the chemical composition of apatites witii respect to fluorine and chlorine has been devised using infra-red (IR) reflected light microspectroscopy. Apatite has the chemical formula Ca5(P04)3(0H,F,Cl). It has been shown tiiat there are significant differences in the annealing kinetics between the fluor-apatite and the chlor-apatite end members of that solid solution series (the OH content has been shown to have no effect on the annealing) (Crowley and Cameron, 1988, in press). Chlorine-rich apatites are more resistant to annealing than are fluorine-rich apatites. Naturally, this phenomena has a pronounced effect upon the age produced by dating methods. A chlorapatite gives an older age and a different thermal history to that of a fluorapatite having the same geological history. The IR technique offers a routine method of increasing the validity of the data produced by apatite fission track analysis. This is because the actual samples which were dated and on which the track lengths were measured, can be analysed in situ on the original mount, thus giving a far more practical, and faster, method tiian using an electron microprobe for chemical determinations. To quantify the F:C1 content the crystal orientation dependency of the spectra and the effects of absorption and crystal substrates on intensity measurements must be understood. These problems can be overcome by systematic measurements on apatites already characterised by microprobe analyses. An increase in chlorine content is shown by a decrease in wavenumber of the peaks in the spectrum. As a result of this, plotting chlorine content versus wavenumber can give calibration curves from which the chemical composition of the crystal can be indicated. The annealing properties of fission tracks in apatite are controlled not only by the intensity and duration of thermal events, but also by the chemical composition of the crystal in which they occur. Nd ISOTOPIC EVIDENCE FOR ULTRA-DEPLETED MANTLE FROM THE EARLY PROTEROZOIC ARUNTA INLIER, CENTRAL AUSTRALL\ W.J.SIVELL. J.X. ZHAO, G.E. MORTIMER and M.T. McCULLOCH, Research School of Earth Sciences, ANU, Canberra, ACT 2601, Australia. Initial 143^(1/144^^ ratios, ranging from e^d = +8.2 to -4.4, for 1.7 to 1.8 Ga mafic volcanics and intrusives from the Harts Range meta-igneous complex (HRMC), and mafic dykes from the Whistleduck Dyke Swarm, Arunta Inlier, central Australia, include the highest values yet reported for Proterozoic igneous rocks. The wide range of together with the exceptionally high initial ratios implies not only a long-term depleted mantle source for the HRMC rocks, but also constrains models of continental underplating involving heterogeneous upper mantle. Lower HRMC metatholeiites are LREE-depleted, have Zr/Nb>45 and uniformly high E^d values of +6.3 to +8.2 that are significantly more positive than the highest values (+3.3 to +6.5) reported for 1.7 Ga to 1.9 Ga volcanics from SW North America, Greenland, Finland and Canada. These values imply an even more LREE depleted composition for their source reservoirs than that proposed for depleted mantle evolution (e.g. DePaolo, 1981; McCulloch and Compston, 1981) and are similar to modem intra-oceanic island arcs. Upper HRMC metatholeiites are LREE-enriched, Zr/Nb<45, with variable but lower e^d = +6.3 to -1.8. Anorthosites intruding the HRMC have E^^ = -6.6 to -4.4. The possible existence of an -2600 Ma (minimum age) mantle source with £Nd(T)s = +3 and fSm/Nd = o.22 would be compatible with the production of an early Proterozoic residual depleted mantle and obviates the need to generate HRMC basalts from a highly refractory source. This source may represent the garnet-bearing residuum, resulting from the production of late Archean tonalites. An alternate model involves melting of a layered, predominantly cumulate ultramafic-gabbroic >2200 Ma underplate with the potential to generate basaltic (and anorthositic) magmas having a range of fSm/Nd. Melts were derived from mafic and felsic underplate layers with a range of Sm/Nd ratios and E^d at the time of partial melting. Fractionated LREE enriched portions of this t>'pe of underplate could have been the source for the widespread early Proterozoic magmatism.

A RIFT-RELATED ORIGIN FOR THE ARCHAEAN-PROTEROZOIC BOUNDARY IN NORTHERN SWEDEN SKIOLD, Torbjorn, Laboratory for Isotope Geology, P.O.Box 50 007, S-104 05 Stockholm, Sweden, and OHLANDER, Bjorn, Department of Economic Geology, Lulea University, S-951 87 Lulea, Sweden A common opinion expressed earlier was that the 1.9 Ga old Skellefte sulphide ore district, a remnant of a volcanic arc, developed along the edge of the reworked Archaean craton in northern Sweden. However, we have recently shown, by relatively sparsely distributed Sm-Nd isotopic analyses, that the Archaean paleoboundary most probably is situated somewhere to the north of the Skellefte district. Now we can present a larger amount of Sm-Nd isotopic data which together with geochemical and U-Pb zircon data allow a more detailed outline of the boundary. The initial epsilon Nd-values of 1.9-1.8 Ga igneous rocks, which in this case reflect variable contamination of assimilated continental crust, rapidly shift from -4 to 0 in a narrow zone sub-parallel to the Skellefte district but about 100 km closer to the core of the old craton. The zone represents a direct continuation of the Raahe-Ladoga zone in Finland and seems to coinside with major tectonic lineaments as interpreted from aeromagnetic and gravimetric measurements. We speculate that this zone is the result of a continental break-up with subsequent transgression of the early Proterozoic volcanic belts more than 1.9 Ga ago. Later, spreading shifted to convergence which caused subduction and orogenesis close to the remaining part of the old continent and consequently presenting a complete Wilson cycle.

YOUNG MONAZITE U-PB AGES FROM SCHISTS: DATING METAMORPHISM IN THE NANGA PARBAT SYNTAXIS, HIMALAYAS, NORTHWESTERN PAKISTAN SMITH. Harold A.. CHAMBERLAIN, C. Page, Dartmouth College, Hanover, NH 03755, USA, and ZEITLER, Peter K., Lehigh University, Bethlehem, PA 18015, USA Monazite is a dependable geochronometer for moderate (staurolite grade) metamorphism of pelitic schists (Smith and Barrreiro, Contrib. Min. Pet., in press). Under the appropriate pressuretemperature and original whole-rock chemistry conditions, monazite forms as a metamorphic mineral. The U-Pb system in monazite then records the age of the metamorphism, even under the influence of subsequent metamorphisms up to and including sillimanite grade. In this study, we have applied monazite U-Pb dating techniques to schists from a region of the Himalayas in Pakistan that has experienced recent, rapid uplift (Zeitler, Tectonics, 4:127151, 1985). Three traverses from the Raikhot Fault (Main Mantle Thrust) into the Nanga Parbat syntaxis have yeilded several schists containing monazite. Individual monazite fractions show slight reverse discordance when corrected for Th disequilibrium. Multiple size/magnetic fractions from each sample have ages that agree within the limits of analysis. Thus each schist exhibits a single age, interpreted as the age of metamorphism. Ages range from -^l 1 Ma to as young as ~4 Ma. In no instance have we observed evidence for inherited ages. These dates raise several interesting questions. (1) Is the monazite indeed coeval with the mineral assemblages? If so, does the arrival of metamorphic monazite at the surface corroborate conclusions of recent, extreme uplift rates or do they suggest more moderate, but still high average rates? (2) Can our understanding of the pressure and temperature necessary to form monazite in these pelites offer constraints on the local geotherm? (3) Do these young monazites offer new insight into formation of this important geochronometer?

A RAPE), INEXPENSIVE TECHNIQUE FOR ANALYSIS OF I8O/I60 IN WATER USING PRE-EVACUATED CONTAINERS SOCKL Richard A., C23 Lockheed ESCO, 2400 NASA Rd. 1, Houston, TX, 77058, KARLSSON, H. R., and GIBSON, E. K., SN2, NASA-JSC, Houston, TX 77058. As a part of NASA's human exploration initiative we are developing a new rapid, low-cost method for measuring the oxygen isotopic ratios of sm^ water samples (< 1.5 ml). Although our application of this technique is for determining energy requirements of humans during long duration space flights (doubly labeled water method), it may also be used routinely as a rapid, inexpensive means of measuring S^^O in waters. Our technique is based on the traditional CO2-H2O equilibration method of Epstein and Mayeda (1953), and makes use of inexpensive $0.15 ea.) off-the-shelf, pre-evacuated, rubberstoppered glass vials, eliminating the need for costly, custom-built glassware. Analysis of a suite of laboratory standards and 3 international standards (V-SMOW, SLAP, and GISP) yields a precision of ± 0.04%o (la). Glass vials (7cc) are injected with a syringe (size #20) which is connected to a small vacuum extraction line consisting of both a CO2 reservoir and a sample tube port. The syringe is evacuated by embedding the tip into the rubber stopper on top of the vial, then pumping. After back-filling with -30 ^imoles of CO2, the vial is removedfromthe line and injected with -1.4 ml H2O using a 5cc gas-tight s^nge. CO2-H2O charged vials are allowed to equilibrate while shaking in a 25'C bath. After 3 hours, when equilibration is complete, the vials are placed in liquid nitrogen-filled dewars for 3 min. tofreezethe CO2-H2O mixmre. Thefrozenmixture is placed back on the line where the syringe isreinserted,the liquid nitrogen dewar is replaced with dry ice-alcohol, and the CO2 is collected into a sample tube for mass spectrometric analysis. While this method presently requires the use of a small extraction line, modifications utilizing commercially available C02-charged vials may eliminate the need for any vacuum apparatus, allowing sample extraction to occur on the inlet system of the mass spectrometer. Epstein, S. and Mayeda, T. K. (1953) GCA, 4, 213-224.

Rb-Sr CHRONOLOGY OF STROMOTOLITIC CHERTS FROM THE DHARWAR SUPER GROUP, SOUTH INDIA. SRINIVASAN,R, SIVARAMAN,T.V,PANTULU,G.V.C, GOPALAN,K, and NAQVI.S.M, National Geophysical Research Institute, Hyderabad 500007, India. Morphologically diverse, actively and passively branched columnar stromotolites have recently been discovered in the Vanivilas^ Forination, Chitradurga Group of Dharwar Supracrustal sequence with a depositional age of > 2.6 Ga. Since such evolved stromotolites are very rare in pre-Riphean times, it is important to establish a minimum age for the Dharwar stromotolites more directly. Following the successful Rb-Sr dating of cherts from the Onverwacht Group, South Africa by Weis and Wasserburg (1987), we have analyzed stromotolitic cherts occurring in the type area of Vanivilas Formation. The cherts consist of stratiform and domical stromotolitic layers intercalated with dolomite. They are compact with dark grey laminations composed of meta-argillites with minor graphite. Such micaeous cherts give a satisfactory isochron with an age of 2.56 + 0.08 Ga and Sr^ of 0.7126 + 3. This age is consistent with the 2.6 Ga age of the intrusive Chitradurga granite and associated volcanics. The initial Sr ratio of cherts is however higher than that the associated volcanics and granites. We interpret the age of the cherts as corresponding to the Sr equilibration and closure of their Rb/Sr systems when major tectonic and metamorphic activity ceased in the area and as setting a minimum age for the evolved stromotolites in the Dharwar Supergroup.

ICOG 7

ABSTRACTS

RADIOMETRIC DATING METHOD FOR PETRIFIED WOOD AND ITS APPLICATION TO FOSSILS FROM DECCAN INTERTRAPPEAN BEDS SRIVASTAVA, A.P. and RAJAGOPALAN, G. Department of Radiometric Dating, Birbal Sahni Institute of Palaeobotany, 53, University Road, P.B. No. 106, Lucknow - 226 007, India. Different mineral grains, formed during the petrification process, are seen in the thin sections of petrified wood. The Apatite crystals, with clear grain boundaries, are observed in the vessel part of the transverse thin sections of the specimens. F-T dating technique has been applied to these in-situ Apatite grains to deduce the age of the petrified wood. The sedimentary beds, exposed in between the successive lava flows in the South Central India, are known as Deccan Intertrappean beds. These beds contain abundant plant f o s s i l s . In order to ascertain the age of these plant fossils for reconstructing the vegetational history and palaeoenvironment of the region, petrified wood samples have been dated using F-T dating method. This report deals with the dating of the petrified wood samples collected from the Intertrappean beds exposed at the localities in Madhya Pradesh; Shahpura, Deorikhurd, Ghughua, Umaria, Silther, Mehdwani, Parapani in Mandla district and Keria, Mohgaonkalan in Chhindwara district. The specimen from Deorikhurd, Umaria and Ghughua are dicotyledonous wood while the rest are palm woods, described under the generic name Palmoxylon. The obtained F-T ages range from 45 Ma to 55 Ma.

SR ISOTOPES IN WATER-ROCK INTERACTION: TIMNA AREA, DEAD SEA RIFT, ISRAEL Geology, The Hebrew University of Jerusalem, and BEYTH, Michael, The Geological Survey of Israel, Jerusalem, Israel Saline springs are found in the Timna region, about 40km north of Elat in southern Israel. TDS of those springs reaches 16gr/l and their temperature is up to 32°C. The waters are of the Ca chloride type (Ca > SO4 + HCO3), strongly Mg depleted (Ca/Mg « 100), and relatively low ^"^ST/^ST ratios (« 0.7062). Such values can be explained by high temperature, subsurface reactions between rift brines and rocks of basaltic composition. Reaction temperature is estimated as about 250°C from silica - temperature relationships. The alteration process of a basalt by saline water may be balanced with respect to Ca - Mg exchange: Ca(Sr) plagioclase + Mgbrine —> (chlorite + epidote) + C:a(Sr)brine Since Sr content of both water and rock is controlled by the same process as Ca, Sr can be used for the reconstruction of reaction progress, and source rock identification. The 87Sr/86Sr and the Sr/Ca ratios in the water are determined by: a. the strontium distribution coefficient (XSr) between the formed minerals and the solution; b. 87Sr/86Sr and Sr/Ca ratios in the initial brines (subscript Li); and ^'^Sr/^^Sr and Sr/Ca ratios in the altered rock (subscript R) according to : .86sr RlCalRl^ Cau

Ca u Ca J Ca, \Ca, CaJ Solution of the equation leads to the following conclusions: a. The subsurface rock which was involved in the reactions is an alkali basalt which is the basalt type abundant in the rift valley; It could not be MORB. b. The Ca/Mg ratio of the initial solution which reacted in the subsurface was about 2. Such a value is typical for oil field brines in theriftvalley. l86Sr L

-2 Ca/R

ABSTRACTS

IC0G7

BULK OCEAN CRUST ALTERATION AT DSDP SITE 417 AxND 418: O, Sr, Nd ISOTOPES, K, Rb, CO2 Hubert Staudigel. Scripps Institution of Oceanography UCSD A-015, LaJolla, CA 92093, USA Stan Hart, Woods Hole Oceamographic Inst. Woods Hole MA 02543 Brian M. Smith, Earth Sci. Div. LBL UCB, Berkeley, CA 94720 H.-U. Schmincke, GEOMAR, 2300 Kiel 13, W. Germany We have approximated the bulk compositions of >100m3 sized compositional domains the oceanic crust through analysis of composite samples from DSDP Sites 417 and 418. Composites were mixed to represent distinctly altering lithological types, volcaniclastics (VCL), submarine extrusives (FLO) and the down-section variations in "average" oceanic crust with fixed proportions of volcaniclastics (6%), massive flows (24%) and pillow lavas (70%), at one extremely altered section (417A) and two moderately altered sites (417D and 418A). Extreme alteration was found for VCL and the 417A composites with ^^ST/^^STO > 0.7052, 5180SMOW>15; K20>1.8, H20>4.5, C02> 3.5 (all wt%), Rb>35ppm, while l^SNd/l^^Nd remained essentially unchanged >0.5130. Addition of more than 15 mole% K2O, H2O and CO2 to these materials provides a potential source for metasomatic fluids and/or reduces the solidus of subducted oceanic crust. Such compositional domains will play a preferred role during island arc magma generation processes and mantle metasomatism. 87Sr/86Sro, 87Sr/86Srmeas, sI^OSMOW, K2O, H2O, and CO2 typically display positive inter-element correlations, and down-hole trends with the most altered samples in the top 150m and the least altered at the bottom of 418A, at 550m. The uppermost 50 m display the highest CO2, but somewhat lower ^'^SI/^^STQ, and H2O, than samples 100m below the surface of the oceanic crust. In particular the similarly altered Sites 417D and 418A display a high degree of coherence, providing a consistency check for composite samples. The average compositions of typical uppermost, average upper 500m and 400-500m oceanic crust are: ^'^ST/^^STO'. 0.7040, 0.7042, 0.7039; 87Sr/86Srmeas: 0.7045, 0.7046, 0.7039; SI^Osmow: 11.2, 10.0, 8.5; K2O: .72, 0.56, 0.15 wt%; Rb: 16, 13, 1 ppm; H2O: 2.2, 2.7, 1.0 wt%, and CO2: 6.3, 2.9, 1.0 wt%, respectively.

PRECISE AGES OF THE LAST INTERGLACIAL EVENT AT HUON PENINSULA STEIN, M.^ Wasserburg, GJ,\ Aharon, P.^ Chen, J.H.\ Zhu, Z.R.^ Bloom, A.L.-* & Chappell, J.^ 'Geol. & Planet. Sci., Caltech, Pasadena, CA 91125; ^Dept. Geol., LSU, Baton Rouge, LA 70803; ^Dept. Biol. &Geom., Canberra, ACT 2601; 'Dept. Geol. Sci., Cornell U., Ithaca, NY 14813 Emergent coral reef terraces are a geological record of sea level changes and tectonic movements of coastal areas during the late Quaternary. The extensive flight of uplifted reef terraces which occurs along the northern flank of the Huon Peninsula in Papua New Guinea contains a particularly good record of the last 250Ky. The Huon terraces were the target of an international expedition which took place in July 1988. In particular, we searched for suitable samples for U-series dating in a reef complex designated as VII, which is correlated with the last interglacial episode and high sea level stand. The mineralogy, petrography, 0 and C isotopic compositions and Mg,Sr concentrations of 11 corals (some in duplicates) were determined along with analysis of U-Th compositions. ^^^Th-Û ages of corals which contain >99% aragonite, retain their primary aragonitic textures, U/Sr ratios around 4x 10"* and initial 5Û(T) values close to the present seawater value of 150, are classified as "most satisfactory (MS) ages." The ages of corals which show initial 6Û(T) higher than 170 are considered "questionable." The "MS ages"fromterrace VII cluster around 119 and 135Ky. The simplest sea level model for terrace VII implies a continuous rise from 135 to 120Ky, which would be analogous with the Holocene record observed at this area (Chappell et al., in prep.). The lack of samples in the interval time between 135 to 120Ky could result from erosion. An erosional unconformity in terrace VII appears to support this. Alternatively, this may have been the result of two periods of rapid sea level rise. The older ages reported here precede the Milankovitch peak for stage 5e by at least 7Ky. We do not have necessary and sufficient conditions to determine that no diagenesis processes have affected the ages. This remains the critical uncertainty. If the present high precision data are accepted as correct, then the validity of the Milankovitch theory of climate change must be reassessed. (#704)

U-PB ISOTOPE DILUTION ANALYSIS OF ZIRCON FRAGMENTS FOR PETROGENETIC STUDIES: I I . ANALYTICAL RESULTS STEI6ER.Rudolf H. and MEIER,Martin, Isotope geochemistry at ETH,CH-8092 Zurich, Switzerland Resolution of zircon U-Pb age determination has been considerably improved with the introduction of single grain analysis and particularly the advent o f , t h e ion microprobe SHRIMP at ANU. Our approach by isotope dilution analysis of selected zircon fragments also reveals a surprisingly large spread of the U/Pb ratios within a single grain. Thanks to acid-washing of the fragments prior to analysis and to good isotopic blank control (llpg of total Pb), precision of U/Pb ratios and Pb/Pb ratios for 1 ^g fragments with total radiogenic Pb content of 60-300pg is sufficiently high as to permit meaningful interpretation of data. This is of special importance, if the data points plot close to concordia seemingly yielding a sequence of concordant ages, whereas they actually determine a chord. First applications of the method were made to polyorogenic rocks from E Greenland, where earlier zircon U-Pb dating of a Caledonian granite by classical methods had resulted in an implausible post-Caledonian lower intersection age. The rim fragments of two single grains each gave a concordant age at about 400 Ma in accordance with the Rb-Sr mineral isochron. Together with the respective core fragments they yielded discordia lines intersecting at 3991504 and 403-1508 Ma. Interestingly, the relative large volume of euhedral internal zonation surrounding the core is also of Caledonian age. In another case, a discordia line based on 8 extreme size and magnetic fractions had suggested a Grenville age granite whose zircons had suffered Pb loss during the Caledonian. On the basis of only three fragments from a single zircon grain, the resulting discordia line intersecting at 422+-15 and 977+-40 (95% C.L.) with a concordant rim fragment at 422 Ma required reinterpretation as Caledonian granite generated from reactivated Grenville crust.

ISOTOPIC CONSTRAINTS ON THE INTERACTION OF BASALTIC MAGMA AND CONTINENTAL CRUST STEWART. Brian W., Dept. of Earth & Space Sci., UCLA, Los Angeles, CA 90024, U.S.A. and DEPAOLO, Donald J., Berkeley Center for Isotope Geochemistry, Dept. of Geology & Geophysics, Univ. of Calif., Berkeley, CA 94720, U.S.A. Geochemical variations observed in continental flood basalts are variously attributed to interaction with continental crust or to heterogeneities in the mantle from which they are derived. In order to understand the mechanisms and extent of basaltic magmacontinental crust interaction, we have undertaken a study of isotopic variations in layered mafic intrusions. Mafic magma chambers are conmionly thought to be the most likely setting for extensive assimilation of crustal material, both because the accumulated crystallizing basaltic magma represents a substantial heat source for melting of crust, and because the long residence time of magma in magma chambers (lO^-lO^ yr) should promote interaction. Nd and Sr isotopes were measured in samples spanning the exposed stratigraphy of the Skaergaard intrusion. East Greenland, the Muskox intrusion, Canada and the Kiglapait intrusion, Labrador. All of these are relatively shallow-level intrusions emplaced at least partially into low-melting, granitic composition walli^k with very different Nd and Sr isotopic ratios; thus these isotopes are sensitive tracers for basalt-wallrock interaction. Calculations using combined assimilation-fractional crystallization models for each magma chamber show that the maximum assimilation rate achieved was 6% of the cr>^stallization rate. This relatively low rate is attributed to the density and viscosity contrast between mafic magma and fused granitic composition wallrock; the lighter silicic melt remains at the roof of the magma chamber over the lifetime of the system, and exchange between the two liquids is very inefficient. Variations within the Muskox intrusion suggest that some interaction may occur during influxes of new magma into the chamber, but that very little mixing takes place during normal periods of cr>'Stallization. At the calculated assimilation rates, the amount of cr>'stallization needed to produce the variations observed in continental flood basalts would result in lavas with unacceptably high degrees of differentiation.

THE FUNDAMENTAL ROLE OF MANTLE DERIVED M A G M A IN THE PRODUCTION OF A LARGE V O L U M E FELSIC VOLCANIC PROVINCE,GAWLER R A N G E S , SOUTH AUSTRALIA. STEWART,K., and FODEN, J., Department of Geology and Geophysics, University of Adelaide, GPO Box 498, Adelaide, South Australia 5001. The role.of mafic magma in the genesis of large volume acid volcanics has often been inferred but direct evidence is usually lacking. The 1592 Ma. Gawler Range Volcanic (GRV) province however, represents a massive outpouring of felsic magma (>25,000km2) which is directly attributable to intrusion of mafic magma into the continental crust. Much of the GRVprovince is underlain by a large positive gravity anomaly interpreted from geophysics to be a major mafic intrusive whose upper surface is at approximately 5 kilometres depth and whose volume may be of equivalent magnitude to that of the Bushveld Complex. Gabbros, approximately contemporaneous to the GRV, have been intersected on the edge of this mafic body. Other evidence for the involvment of mafic magma in the GRV province include the eruption of basalts early in the eruption history,and the occurrence of mafic inclusions in some of the large volume ash flow tuffs whose form and texture indicate they were molten at the time of eruption. The early erupted volcanics exhibit marked compositional diversity with dacites to high silica rhyolites interlayered with basalts.The later erupted units are dominated by a number of large volume, highly welded, dacitic to rhyodacitic ash flow tuffs each of which is minerallogically and chemically homogeneous. Mafic mineral assemblages in the volcanics are mainly pigeonite+augite, and calculated temperatures range from approximately 1300 ^C for MgO rich basalts to >900 ^C for rhyolitic magmas. The chemistry of the entire province is high-K and calc-alkaline All samples,from basalt to high silica rhyolite, are enriched in Ba,Th,Zr,Y, K,Rb and LREE. Two types of basalt exist; at a similar MgO content one type has higher P2O5, Ti02, LREE, Zr, Y and Ba. The volcanics plot with initial Sr ranging from 0.703 to 0.707 and epsilon Nd values from +1.2 to -8 indicating the combined roles of fractional crystallization, lithospheric fusion and magma mixing.

EXOTIC SILICON CARBIDE IN CARBONACEOUS CHONDRITES STONE J., HUTCHEON I.D.*, EPSTEIN S. & WASSERBURG GJ.*. Division of Geological & Planetary Sciences, 170-25, Caltech, Pasadena, CA, 91125, U.S.A. (*Also Lunatic Asylum, Geol. Planet. Sci., Caltech). Various fractions of the SiC in carbonaceous and enstatite chondrites host distinct isotopic anomalies in Si, C, N and noble gases. Clarifying relations amongst anomalies, and their nucleosynthetic origins, requires separation of complex SiC populations into subsets of related grains, whose isotopic make-up can be studied in isolation. One such subset, common to CI, CM2 & C3V chondrites is marked by coarse grainsize and distinctive 'platy* surface texture. A second distinct subset, of coarse smooth-surfaced grains occurs in Indarch (E4). These subsets show distinct Si isotope systematics (determined by measurements of individual micron-sized ^ains using the PANURGE ion microprobe). Indarch grains are essentially normal in Si isotopic composition, compatible with a solar-system origin. In contrast, 'platy' grains in Cchondrites consistently carry anomalous Si, with correlated variations in 529si and S^Osi exceeding I00%o (Fig.l). The slope of the Fig.l array is 1.4, and it bypasses the normal Si composition. Together these constraints indicate isotopic variation in the grains due to mixing of two exotic Si components, neither related to solar Si, and suggest an extra-solar origin. Isotopic variability amongst the 'platy' grains implies heterogeneous Si at their site of formation. If they Murchison are red giant condensates (as suggested by ^^c, 14n and sOrgueil process isotope enrichments reported for other grains), a source of Si isotopic variability (e.g. core Heburning) must be found. Alternatively, condensation of these grains may have followed incomplete mixing in a stellar envelope, or involved mixing between Si components produced in separate shells of an evolved star. For example, condensation of incompletely mixed C- and Nebuming (2930si.rich) and O-buming 40 90 140 (28Si-rich) supernova ejecta could yield grains with Si isotopic 630si ( ^ ) compositions like those observed.

ICOG 7

ABSTRACTS

PROTEROZOIC ORGANIC CARBON - ITS PRESERVATION AND ISOTOPIC COMPOSITION STRAUSS, H., Ruhr-Universitat, 4630 Bochum, FRG; DES MARAIS, D.J., NASA Ames Research Center, Moffett Field, CA 94035, USA; HAYES, J.M., Indiana University, Bloomington, IN 47405, USA; SUMMONS, R.E., Bureau of Mineral Resources, Canberra, A.C.T. 2601, Australia A comprehensive study has been made of abundance, preservation and isotopic composition of Proterozoic organic carbon (kerogen), allowing the evaluation of variations within the global carbon cycle. The assessment of kerogen preservation through determination of its elemental composition (C,H,N) reveals a generally advanced degree of alteration. Thermal alteration, however, not only affects kerogen preservation (H/C decreases) but also its carbon isotopic composition. Thus, useful information from carbon isotope data regarding evolutionary trends and/or depositional environments can only be obtained from reasonably well preserved kerogens. Carbon isotopic analyses of minimally altered kerogens (H/C > 0.2) indicate a steady increase of 13C from -34 to -31 Ic through the early and middle Proterozoic while the inorganic carbon record seems to lie near 0 The late Proterozoic shows significant secular variations of both records with a difference of 28 lo. Explanations for the trends of & C values for kerogens invoke global changes: changes in the relative burial rates of organic vs. inorganic carbon and/or changes in the isotopic discrimination associated with synthesis, alteration, and preservation of organic matter. Furthermore, the steady decrease ofaS, the isotopic difference between organic and carbonate carbon, is likely to reflect at least in part the proposed long-term decline in atmospheric CO2 levels.

FISSION-TRACK EVIDENCE FROM THE MT. GRIFFrTH-BSSION WALL PROHLE FOR LATE CRETACEOUS UPLIFT IN THE SCOTT GLACIER AREA OF THE TRANS ANTARCTIC MOUNTAINS STUMP. Edmund and FITZGERALD, Paul G., Department of Geology, Arizona State University, Tempe, Arizona 85287, U.S.A. The Scott Glacier area of the Transantarctic Mountains (TAM) is underlain primarily by an extensive batholitic complex of Cambro-Ordovician age. Border phases of the batholith include a Late Proterozoic sequence of graywacke and argillite (La Gorce Formation) and a Late Cambrian (?) silicic porphyry (Wyatt Formation) of volcanic and hypabyssal origin. Collecting for apatite fission-track analysis was done throughout the area toward determining the uplift/cooling history of that portion of the TAM. A vertical profile was collected at approximately 100 m intervals on the Mt. Griffith-Fission Wall massif, from the summit (3,190 m) to the base (1,183 m), from massive Wyatt Formation at the top through a board transition of increasingly gneissic character to coarse-grained granite in the lower portion. Apatite fission-track ages range from 119 Ma to 75 Ma, with an apparent break in slope on an age vs. elevation plot at about 80 Ma and 2500m. Mean track lengths for all samples ranges between 12.8 and 13.8 \Lm with standard deviations from 1.6 to 2.3 p.m. There is no significant variation on either side of the break in slope. Our interpretation is that the upper portion of the Mt. Griffith-Fission Wall massif lay within the partial annealing zone (PAZ) following Jurassic magmatism that reset ages to zero, until uplift approximately 80 Ma, at which time it moved into the upper portion of the PAZ. Samples beneath the break in slope had ages of zero prior to -80 Ma at which time they were uplifted 1-1.5 km into the PAZ. Subsequent uplift, probably in the eariy Cenozoic, and erosion has placed the massif at its present elevation. In both northern and southern Victoria Land, 1,000-2,000 km NV/ of Scott Glacier, initiation of uplift of the TAM is indicated at about 55-60 Ma. By comparison, uplift in the Scott Glacier area appears to have commenced 20-25 m.y. earlier, during the late Cretaceous.

ABSTRACTS

ICOG 7

ANOMALOUS NITROGEN IN Y74191 (L3) CHONDRITE Naoji SUGIURA and Ko HASHIZUME, Geophysical Institute, University of Tokyo, Tokyo, 113, JAPAN. Presolar grains such as diamonds and SiC have been identified in primitive chondrites. Both diamonds and SiC are chemically stable, so they are enriched in HF/HCl residues and relatively easy to identify. Other presolar grains which are chemically less stable may exist in primitive meteorites. Isotope anomaly of nitrogen may be used as an indicator of the presence of presolar grains, because it tends to show huge isotope anomaly. We have measured nitrogen isotope ratio of about 20 chondrites, and found a striking isotope anomaly 750°/oo-) in a L3 chondrite (Y74191). Neon and Argon are also measured to make sure that the anomaly is not due to spallogenic nitrogen. It is noted that the release pattern of the anomalous nitrogen is similar to that of primordial Ar. The anomalous nitrogen is not particularly enriched in the magnetic fraction, the non-magnetic fraction nor in the HF/HCl residue. Compared with anomalous nitrogen in other meteorites (Bencubbin, Polymict Ureilites, ALH85085), the following similarity is noted. 1) Two or more components of anomalous nitrogen seem to exist in each meteorite. 2) The component released at higher temperature by combustion experiments heis higher and is dissolved by HF/HCl. The presence of such common features for this anomalous nitrogen in different meteorites suggests that this component was ubiquitous in the primitive solar nebula.

CARBON ISOTOPIC COMPOSITION OF INDIVIDUAL SEDIMENTARY ORGANIC COMPOUNDS: AN ADVANCED TECHNIQUE FOR ESTABLISHING THEIR ORIGINS SUMMONS Roger E., BOREHAM Christopher J. and ROKSANT)IC Zarko, Bureau of Mineral Resources, GPO Box 378, Canberra City, ACT, 2601, Australia Total organic carbon from living microbial mats of Hamelin Pool, Shark Bay, Western Australia is isotopically heavy (b'^^C = -8 to -14 %o PDB) compared to organic matter from open marine environments. As expected, bulk lipidic components are lighter by only about 3%o. Individual hydrocarbons from a deep (7m) algal mat with -12%o show a wide dispersion of signatures from -8 to -28%c PDB and attesting to a variety of sources. Hydrocarbons from algal source organisms have b^^C -12%o while cyanobacterial markers are at about -20%o. Waxy n-alkanes have signatures as light as 28%c and have probably not been generated in-situ. A more likely origin is from vascular plants from terrestrial environs. ^ Individual hydrocarbons from fossil organic matter show similar wide dispersions in carbon isotopic composition. In some cases, the measured values can be linked to biochemical peculiarities of the source organisms. In other cases the signatures can indicate a primary origin or one from bacterial re-working. Isotopic signatures, in combination with accurate chemical structure information, is a powerful means of studying the origins and fate of sedimentary organic matter. The technique has direct application to petroleum exploration and studies of palaeoenvironment and carbon cycling.

CHEMICAL AND ISOTOPIC CHARACTERISTICS OF ULTRAPOTASSIC ROCKS Sun, S.-s., Bureau Mineral Resources, Canberra, ACT, Australia Ultrapotassic magmas show a spectrum of comjwsitions and occur in modem subduction environments, post orogenic belts, continental rifts and stable cratons. Integrated chemical, isotopic and experimental petrological studies provide insights into the various processes (e.g., subduction vs intraplate, H20-rich vs C02-rich fluid metasomatism) and sources (e.g., lithosphere, asthenosphere) involved in their formation. Conceivably, many ultrapotassic magmas are integrated melts derived from different sources that have evolved through multistage processes. Despite such complexities in magma generation, systematic chemical and isotopic variations are, however, found in ultrapotassic provinces. Post-collisional ultrapotassic lavas, like their counterparts in arc settings, have strong sediment and subduction zone signatures, i.e., depletion of HFSE, Pb isotopic compositions reflecting modem segments, negative Eu anomdies, and high Ba/La, Ba/Nb, Pb/Ce, Cs/Rb. Additionally, ultrapotassic rocks from some arcs (e.g., Sunda) with EM2(Dupal)-type characteristics may be mixtures of an OIB-type component and a subduction zone component Lamproites with high Ti and Nb/La ~ 1 from older intraplate environments (e.g., Kimberley, W. Australia) show evidence of progressive metasomatic overprinting of their mantle source. Although sediment subduction is not required for the K-enrichment and phlogopite formation in the lithospheric mantle sources of these ultrapotassic rocks, Pb and C isotope data support the existence of a sediment component. Pb isotope evolution models require a high |i environment (upper crust) early in the source history followed by a low |i environment (due to high Pb in sediments). The composition of other intraplate ultrapotassic rocks commonly show chemical and isotopic affinity to Dupal OIB, but in more exaggerated form, suggesting a possible genetic connection between them. Lithospheric sources suggested for Cenozoic ultrapotassic rocks of the East African Rift and in SE Australia may have been modified by Dupal OIB melts. It is plausible that delaminated ancient lithospheric mande with exaggerated Dupal characteristics is further modified through intraplate metasomatism and mixing in the convecting mantle, eventually becoming a Dupal plume source. ION-PROBE Pb/U ISOTOPIC AGES OF A YOUNG SPECTRUM OF GEM ZIRCONS FROM THE EAST AUSTRALIAN BASALTIC BELT SUTHERLAND F.L.. Mineral Section, The Australian Museum, Sydney, NSW, 2000, and KINNY P.D., Research School of Earth Sciences, Australian National University, Canberra, ACT, 2601. Gem zircons in the East Australian Cainozoic-Mesozoic basaltic belt are scattered over 3000 kilometres. They vary in colour, crystal forms, and uranium contents. Potential sources are (1) felsic plutons related to Cainozoic-Mesozoic volcanism, (2) pegmatoids in Mesozoic-Palaeozoic granitoids, (3) metamorphic basements of Palaeozoic-Precambrian age. Fission-track ages (1 to 190 Ma) suggest ages reset by volcanism. Over sixty zkcons from forty sites were dated by SHRIMP U-Pb isotopic analysis to resolve their origins. Most of the analysed zircons (>80%) are 'young' (0-100±0-22 Ma), their ages corresponding to volcanism after opening of tiie Tasman Sea. The measured ^^s mostiy non-radiogenic so only 206pb/238u ages were determined. Ion-probe PbAJ ages close to fission-track ages link many zircons to subvolcanic fractionation. Others are clearly older than the age of eruption and thus represent earlier magmatism. Other zircons have Mesozoic ion-probe Pb/U ages (140-290 ± 10-25 Ma). Some of these match fission-track ages and local basalt ages. Others have much younger fission-track ages (40-60 Ma), suggesting prolonged residence in plutonic chambers before expulsion. A few low-U zircon suites have late Mesozoic ion-probe PbAJ ages (67-147±8-56 Ma), close to fission-track ages. These resemble kimberlitic zircons and may relate to a lull between Mesozoic and post-Tasman volcanism. Similar low-U zircons with young Pb/U isotopic ages (0-10 Ma) come from highly under-saturated centres. Ion-probe Pb/U ages on zircon inclusions in New England sapphires (33-36±3 Ma) reinforce their volcanic association. No ages have been found which correspond to local Palaeozoic granitoids or older basement rocks.

98 ANALYTICAL STUDIES ON APPLICATION OF RE-OS GEOCHRONOMETER TO MOLYBDENITE SUZUKI. H . SHIMIZU, and A . MASUDA Department of C h e m i s t r y . Faculty of Science, The University of T o k y o . T o k y o . 113. Japan Re-Os geochronometer is the promising method especially for application to sulfide ore deposits, because Re and Os are sidero- or chalco-phile. The difficulty in Os isotope measurement has prevented the practical use of this m e t h o d . A l t h o u g h L u c k and Allegre first succeeded in dating m o l y b d e n i t e , there remain some problems in p r a c t i c a l use of this m e t h o d . In our study, Re and Os abundances in molj'bdenite from S e i k y u , J a p a n , were measured using ICP-MS b y isotope dilution m e t h o d . We found out that it w a s not easy for us to attain the isotopic equilibrium for Os in acid digestion, but this problem can be eliminated by the following m e t h o d . Molybdenite is decomposed by HNO3 in a glass ampoule. The precipitate of poly-molybdenum oxide formed is dissolved by N H 3 and added to the sample solution. Then Os is separated bv distillation. It was shown that radiogenic -^^'Os increases gradually with time during distillation. Table 1 shows the time-integrated ratios, Os contents and ^ Effect of distillation time on ages calculated i87os/i92os ratio. Os abundance and age. from these ratios. By taking Os/ppb Age/Ma account of the time/hrchange of the5 I87Qg/192Qg^ 35.6 44.3 0.299 0--1 44.9 55.4 hours are enough 0 - 2 0.362 55.8 for distillation 0--3 68.9 0.438 time. T h u s , the 59.2 73.1 0.462 age of the sam60.2 74.4 0.469 O'-S ple has been determined to •time-integrated. be 60.2Ma.

T H E ROLE OF CRUSTAL CONTAMINATION IN THE PETROGENESIS O F TASMANIAN DOLERITES R.J.SWEENEY AND D.H GREEN, Geology Department, University of Tasmania, Hobart 7001, Australia Tasmanian dolerites represent a compositionally uniform and distinctive (e.g. 54-56% SiOz, ^'^Srl^^Sr initial ratio .709-.714) eruption of basaltic magma in the Mesozoic volcanics of Gondwanaland. The geochemical variation observed within the suite is attributed to the effects of crystal fractionation and alteration by meteoric groundwaters (Hergt et al., J.Petrol., v30, 1989). However, the elevated Si02, K2O and Rb abundances and 87Sr/86Sr initial ratios relative to other Mesozoic tholeiites (e.g. Karoo) and other bas;Utic volcanics in the sub-region (e.g. Victorian Jurassic alkaline basalts, Tasmanian Cainozoic basalts) is highly suggestive of some contamination. Clearly this contamination must have occured prior to the intrusion of the dykes and the generation of geochemical variations within the group. Possible sites for this contamination are (1) the mantle source or (2) the granitic continental crust. A model for the derivation of the dolerites by a high degree of melting of a mantle source 'contaminated' with 3% subducted sediment (Post-Archaean Terrestrial Shale) has recently been proposed (Hergt et al., 1989). Here, a model of contamination of a parental Tasmanian dolerite composition with typical Tasmanian granitic crust is proposed, assuming a coupled assimilationfractional crystallization (AFC) mechanism. In this model a primary, mantle derived picritic parent (Mg#72) characteristically evolves in deep crustal magma chambers. It is suggested that much of the distinctive geochemistry (including variations in Sr-, Nd- and Pb-isotope ratios) may be ascribed to crustal contamination prior to the intrusion of the observed suite of dolerites as an alternative to mantle source contamination proposed by Hergt et al. (1989) to generate a high ^"^Sr/^^Sr mantle reservoir. The proposed parental (primary?) Tasmanian dolerite (Mg#72) composition is more enriched in the LILE (K, Ba and probably Rb) relative to the HFSE (Ti, P, Zr, Nd and Sm) when compared to t>'pical MORB and this reflects a mantle reservoir geochemically distinct from a MORB source, including a major element signature suggesting derivation from a residual mantle.

ICOG 7

ABSTRACTS

CALIBRATION OF THE CENOZOIC TIME SCALE AND THE INTERCALIBRATION OF POTASSIUM-ARGON STANDARDS SWISHER. C.C., Geochronology Center of the Institute of Human Origins, 2453 Ridge Rd., Berkeley CA 94709 USA High precision 40Ar/39Ar dates are rapidly modifying the calibration of the Cenozoic Time Scale. Despite the improved precision obtainable with 40Ar/39Ar dating, the accuracy of these dates still depend upon the ages of conventional K-Ar laboratory standards. Recently, Fish Canyon Sanidine FC-1 (27.8Ma) and Minnesota hornblende MMhb-I (520.4Ma) have been proposed as interlaboratory 40Ar/39Ar standards. To check the accuracy of the reported ages of MMhb-I and FC-1, sample splits were irradiated for 12 hours at the Los Alamos reactor along with K-Ar standard biotites GHC-305, ANU GA-1550 and USGS SB-3. The three biotite standards have been used widely for the calibration of dates that currently serve as calibration points for the Cenozoic Time Scale. Ten 40Ar/39Ar single crystal dates on each sample indicate a grain to grain reproducibility (la) of less than 0.2% for FC-1 and approximately 1.0% for MMhb-I, while the reproducibility of the biotite standards ranged from 0.4% to 0.9%. Mean 40Ar/39Ar ages for MMhb-I and Fish Canyon were calculated using each the three K-Ar standards as monitor minerals. Using the published ages on the K-Ar standards for the calculation of J, resulted in dates for MMhb-I and FC-1 that averaged approximately 1.0% younger than their published K-Ar ages of 520.4Ma and 27.8Ma. The difference between the age of these standards and their K-Ar ages has profound affects on the calibration of the Cenozoic Time Scale. For example, if the younger ages are correct then recent the 40Ar/39Ar dates of 66.2 Ma for the iridium bearing K/T boundary and 34.0 Ma the E / 0 boundary in North America would yield younger ages of 65.5Ma and 33.7 Ma respectively.

T H E R M A L AxNNEALING C H A R A C T E R I S T I C S O F FISSION TRACKS IN ZIRCON TakahiroTAGAML Hisatoshi ITO, Noriko HASEBE and Susumu NISHIMURA, Dept. of Geology and Mineralogy, Faculty of Science, Kyoto Univ., Kyoto 606, Japan. In spite of its great potentiality as a tool for thermal history analysis, zircon fission-track thermochronometry has not been extensively used because of ambiguous thermal annealing behaviors of tracks and, accordingly, of an imprecisely known closure temperature for track retension. The ambiguity has been derived from the change of anisotropic etching nature and thus, possibly, of etching efficiency of tracks during annealing, reflecting loss of alpha-radiation damage accumulated in zircons. Hence it is followed that observed reduction of track density in annealed zircons cannot directly be regarded as true reduction. To assess the effect of change of etching anisotropy on observed track density, we carried out one-hour isochronal annealing on zircon cr>'stals KT06, separated from a rhyolitic welded tuff of the Koto Rhyolites, southwest Japan, and measured the angular distribution of etched tracks along with their density and horizontal confined track length. Several important findings are provided; drastic changes of time and anisotropy of track etching, anisotropy of track annealing, and extra-Poissonian variation in grain-age distribution. It is also demonstrated that the apparent reduction of etching efficiency in annealed zircons is not significant, so long as the adopted criteria for etching tracks are strictly maintained. Consequently, the relationship between track density and length reduction in zircon is found to be approximately linear with an apparent intercept of -0.2 on the length axis, which is considerably different from that in apatite previously reported. The combined influences of possible biases in track length measurement can reasonably explain the obser\'ed forms of density-length relationship in the two minerals. Further additional investigations to quantitatively analyze the biases will be presented.

ABSTRACTS

ICOG 7

DATING OF METAMORPHIC AND IGNEOUS ROCKS FROM S0R RONDANE MTS, ANTARCTICA AND THE MOVEMENT OF THE ANTARCTICA YUTAICA TAKIGAMI, Kanto Gakuen Univ., Ohtashi, Gunma, 373, Japan and MINORU FUNAKI, National Institute of Polar Research, Itabashi-ku, Tokyo, 173, Japan The S0r Rondane Mountain areas, situated in East Antarctica, are composed of various types of metamorphic rocks reprensented by gneiss, amphibolite, migmatite and schist, granitic rocks and the associated dikes. From 1984, this area has been investigated by the Japanese Antarctic Research Expedition and a lot of rock samples ( e.g. gneiss, granite, syenite and dolerite dike ) were collected for geologic and paleomagnetic studies. To reveal the thermal history of this area and the movement of Antarctica, "^^Ar-^^Ar and K-Ar datings were performed for rock samples which were also examined by paleomagnetic studies. It has been revealed that the thermal metamorphism and the intrusive activities took place during the period between - 440 and -^ 550 Ma based on obtained ages*. The movement of the Antarctica is discussed by using the paleomagnetic data such as the position of the VGP (Virtual Geomagnetic Pole).

MIDDLE PALAEOZOIC EXTINCTION EVENTS: ISOTOPIC SIGNATURES TALENT. J.A.. andMAWSON, R., Macquarie University, North Ryde, NSW, 2109, Australia and ANDREW, A.S., HAMILTON, P.J., and WHITFORD, D.J., CSIRO Division of Exploration Geoscience, North Ryde, NSW, 2113, Australia Four sharp cut-backs in generic diversity (c. 40-50%) have been recently recognized during Silurian and Devonian times: very late in the Ludlow (Pentamerid Event = PE), in the late Emsian (Emsian Event = EME), in the late Eifelian (Eifelian Event = EFE) and in the late Givetian (Stringocephalid Event = SE). The first corresponds to an abrupt decimation with extinction of most pentamerid brachiopods, the last to a dramatic wipe-out of large stringocephalid brachiopods, globally important as rock-builders immediately prior to the extinction event. The last three events were of such importance that, despite the large number of new genera in the late Emsian and Middle to Late Devonian, the rate of increase was not sufficient to offset the consequences of recurrent decimation. In a pilot study using S^^C values on whole rocks and brachipods across the EFE, a sudden increase of «6 permil correlates with a 46% extinction of late Eifelian brachipod genera. Global extinction events at the K/T and P/T boundaries are associated with• decreases in S^^C of similar magnitudes, consistent with a reduction in organic productivity. The increase in S^^C values at the EME, is similar to that recognized-by others for the "Kellwasser Event" (KE) at the Frasnian-Famennian boundary (Late Devonian), and may be related to blooming of phytoplankton concomitant with shelly fauna extinction.

99 PRECISE DETERMINATION OF Os-184 ABUNDANCE TSUYOSHI TANAKA, Geological Survey of Japan Higashi 1-1-3, Tsukuba, Ibaraki 305, Japan is produced only in the p-process, whereas other osmium isotopes are produced in s- and/or r- processes. Small amount of osmium isotopes are' difficult to measure by the conventional mass spectrometry. 184os/15Qos ratios in meteorites therefore have been examined by 7-ray spectrometry on neutron irradiated samples. From the activity ratios of their activation products and those of a terrestrial standard, the ^^^Os abundance was used to be calculated as a relative value to the terrestrial reagent. However, the 184os abundance in the terrestrial reagent has not been determined well. IMA, ICP-MS, RIMS and AMS have recently become powerful tools for Os-isotope works. These mass spectrometers, however, have not yet given precise abundance of 184os. Here the 184Os abundance is determined to be 0.0172 ±0.0002 %. The measuring technique is 7-ray spectrometry on a neutron-irradiated samples. 184os enriched reagent of Oak Ridge National Laboratory (184os and I^Oqs abundances are 5.45 ±0.06 % and 23.7 ±0.1%, respectively), together with natural samples, is irradiated. Natural 184os abundance (0.0172 ±0.0002 %) is determined as a relative value to the Oak Ridge 184os enriched reagent. The value obtained here is very close to, but smaller than, the values reported before. Uncertainties owing to the counting statistics and to reproducibility of repeated measurenients are less than 0.5 %. Most of the uncertainties on the above value comes from the uncertainty on 184os abundance in 184os enriched reagent.

Sr ISOTOPIC COMPOSITIONS IN FLUID-ROCK SYSTEMS IN EL TATIO AND PUCHULDIZA GEOTHERMAL FIELDS: NORTHERN CHILE COLOMBO C.G.TASSINARI(l),ALFREDO LAHSEN(2), FRANCISCO MUNIZAGA(2), KEI SATO(l) AND PAULO E.MORI : 1- INSTITUTO DE GEOCIENCIAS UNIVERSI TY OF SAO PAULO PO.BOX 20899, BRAZIL: 2-DEPARTAMENTO DE GEOLOGIA, UNIVERSITY OF CHI LE, PO. BOX 13518 - correo 21 ft 7

Rb and Sr concentrations and Sr/ Sr com positions have been measured in neo-formed minerals, host-rocks and in thermal mineral waters from Puchuldiza and El Tatio geothermal fields in northern Chile. The investigated areas are located in the Andean Chain along the Altiplano Block wich is made up by the Pliocene-Quatemaiy volcanic belt (12 to 4 ma.). This unit is corrposed by large igniiribrite ^eet intercalated v;ith clastic sediments and dacitic-andesitic flows. The neo-formed minerals are: halite, polyalite,santite, thenardite, sulfur, cinnabre, jarosite, kaolin^ te, smectite, illite and amorphous silica. The isotopic data are: EL Tatio: 87Sr/86Sr = 0.7089 - 0.7109 for neoformed minerals and 0.7073 - 0.7096 for host-rocks. Puchuldiza: 87Sr/86Sr = 0.7061 - 0.7070 for neofonned minerals; 0.7061 - 0.7077 for host-rocks and about 0.7065 for thennal water. From the above isotopic data it is clear the Sr isotopic homogeneization during the fluidhost-rocks interaction process. Thus the measured Sr initial ratio for epithermal deposits reflect the mean Sr isotopic composition of the host rocks.

100 FURTHER EVIDENCE FROM Be-B SYSTEMATICS FOR THE HOMOGENEITY OF THE SUBDUCTED COMPONENT IN ARC MAGMATISM: CASE OF THE KURILEKAMCHATKA ARC TERA. Fouad, MORRIS, Julie, Dept. Terrestrial Magnetism, Carnegie Inst. Washington, 5241 Broad Branch Rd, Washington, D.C. 20015, U.S.A., LEEMAN, W.R, Rice University, Houston, TX 77251, U.S.A. and TSVETKOV, A.A., Inst, of Geology and Geochemistry, Academy of the Sciences, Moscow, USSR. The isotopic ratio ^°Be/^Be eind the elemental ratio B/Be were determined in 25 historic (1857-1983) lavas from the KurileKamchatka arc. The isotopic ratios are distinct in the two segments of the arc; being 1 to 4 x 10"^^ in Kamchatka and 8 to 42 X 10"^^ in the Kuriles. The difference is primarily due to contrasting abundances which, in 10® atoms/g, range from 0.5 to 1.2 in Kamchatka and from 2.2 to 7.9 in the Kuriles. is a radioactive isotope (ti/2=1.5Ma) which because of its atmospheric origin and the ease of being hydrolized, is highly enriched in oceanic sediments (5000 X 10® a/g) in contrast to < 10® a/g in MORB and OIB. Boron is similar to in that it is strongly enriched in oceanic sediments (50-150 ppm) compared to MORB and OIB ( ^ 1 to 2 ppm respectively), and its levels in arc are primarily controlled by a subduction component. In good accord with our results on four other axes (1) Be isotopic ratios are correlated with the B/Be ratios in the KurileKamchatka samples. The two segments of the axe provide two distinct linear trends converging on the origin. Each trend is interpreted as a mixing line of two well defined end members; a circumstance that requires that the subducted component involved in magmatism be effectively homogenized to a chaxacteristic composition along the length of each segment. Ref. (1) Morris, J. et al., Nature, 1990.

K-Ar GEOCHRONOLOGY OF IGNEOUS ROCKS FROM MIDDLE BAJA CALIFORNIA PENINSULA, MEXICO TERRELL. David J., Subdireccion de Exploracion, Instituto Mexicano del Petroleo, Eje Central Lazaro Cardenas #152, D.F. 07730, Mexico, MITCHELL, John G., School of Physics, University of Newcastle upon Tyne, N^l 7RU, U.K., and SAUNDERS, Andrew D., Department of Geology, Leicester University, LEI 7RH, U.K. Dating igneous events in Baja California is carried out in hope of understanding the geochemical evolution of the peninsula and its formation related to the opening of the Sea of Cortes. Geophysical evidence suggests that subduction of oceanic lithosphere under the western margin of the North American plate occurred since the Cretaceous and ended during the Miocene at approximately 12.5 Ma. Present spreading rates estimated around 3 cm/yr and geological evidence suggest that the Gulf of Califomia started to open at about 4 Ma ago. Thus leaving 8 Ma between cessation of subduction and the opening of the Gulf of Califomia where the igneous activity can provide information about the fate of the spreading centre. We present K-Ar dates of La Purisima, La Giganta and Tres Virgenes, where a major episode around 8 Ma ago is apparentiy related to the existence of the spreading centre under the peninsula, suggesting that the spreading centre was subducted with the oceanic plate under the peninsula and later appeared in the opening of the Gulf.

ICOG 7

ABSTRACTS

PB-SR-ND ISOTOPE GEOCIIEMISTRY OF BASALTS FROM FUERTEVENTURA, CANARY ISLANDS, SPAIN: THE DEVELOPMENT OF HMU MANTLE THIRLWALL. M.F.. Dept of Geology, Royal HoUoway and Bedford New College, Egham, Surrey TW20 OEX, UK. Miocene (27-14 Ma) shield tholeiites and alkali basalts from the eastern Canary Island of Fuerteventura are followed by posterosional lavas (<4 Ma) ranging from tholeiite to basanite. Shield lavas show fractionation trends from ankaramite to hawaiite, while posterosional lavas appear close to primary liquid compositions. The 206pb/204pb . 207pb/204pb correlation Cage' 1.5 ± 0.2 Ga, MS WD 3.8) suggests either simple two-stage U-Pb evolution or mixing between DMM and HIMU mantle sources, with posterosionals having greater DMM component (206pb/204p5 = 18.8-19.3) than the shield basalts (19.3-20.0). Correlations between incompatible element ratios (e.g. Ba/La, La/Nb) and Pb isotopes are not consistent with DMM-HIMU mixing, as samples with the strongest relative Nb enrichment have the least radiogenic Pb. Further, Nd and Sr isotopes are decoupled from both Pb and chemical parameters, and the Nd-Sr correlation for posterosional lavas is displaced towards HIMU relative to that for the shield lavas (contrasting to the HIMU Pb in the shield).. In order to explain the data by mixing, at least four mande source components would be required, none of which bears any strong similarity to DMM. Little evidence exists for pelagic sediment involvement. The decoupling of the Pb isotope systematics from the Sr-Nd systematics could be explained if the HIMU isotopic signature were a transient feature in the mantle related to Phanerozoic chemical fractionations. This is suggested by 238u 206p5 pseudoisochrons of Phanerozoic apparent age: the high observed \i values result in 206pb/204pt)4oo close to estimated MORB values at that time, with slightly elevated 207pb/204pb. High source is indicated by Ce/Pb ratios of 30-70, higher than most oceanic basalts, normal KAJ ratios, and the lack of a - 7i/Nb (i.e. %melt?) correlation in the shield lavas. TDM ages for the lavas are also Phanerozoic for both Sr and Nd, but substantial Sm/Nd fractionation probably took place on melting, and it is thought that the Nd-Sr covariations date from earlier events in the mantle.

MANTLE & SLAB-DERIVED SIGNATURES IN ARC MAGMAS THOM. A.. RSES, ANU, Canberra ACT 2601, Australia, and ARNDT, N.T., Max-Planck-Institut fur Chemie, Mainz, FRG An important difference between magmas generated in subduction zones, and magmas produced in other tectonic environments lies in the decoupling of alkali and alkaline-earth elements (LILE), REE, and HFSE (Nb, Hf, Zr, Ti). Arc magmas have elevated abundances of the LILE relative to REE. LREE patterns range from depleted to enriched, while HREE patterns are relatively constant and flat. Whereas LREE-depleted samples have chondritic Zr/Sm and no obvious HFSE anomalies, LREE enriched samples display apparent HFSE-depletions. Compared to MORB, arc lavas have high Sr and low Nd isotopic compositions. Current models for arc magma genesis explain high LILE/REE and distinct isotopic compositions by an admixture of highly LILE enriched slab-derived fluids and the overlying mantie wedge. The origin of apparent Nb-, Hf-, Zr-, and Ti- depletions and the LREE enrichments in subduction-related magmas remains controversial. Based on literature data from the Kurile, Mariana, and New Hebrides arcs, we propose an alternative model involving fluid metasomatism, partial melting in the mantie wedge, and fractional crystallization. The relative concentrations of Nb, Hf, Zr, and Ti are attributed to the mantle wedge, while enrichments of LILE and REE relative to HFSE are produced by fluid metasomatism of the wedge by dehydration of the subducted slab. These enrichments are best explained by the higher solubility of these elements in aqueous fluids and their enrichment factors are proportional to the amount of fluid added to the mantie wedge. Compositions of the dehydrating fluids change with depth in the subduction zone. Fluids released at shallow levels are in equilibrium with amphibole and plagioclase, are moderately LREE enriched, and metasomatize the source of tholeiitic arc magmas. Fluids released at greater depths are in equilibrium with garnet and clinopyroxene, are strongly LREE enriched and HREE depleted, and lead to the formation of calcalkaline magmas. The presence or absence of different HFSE enriched phase(s) in the slab may cause a decoupling of individual HFSE anomalies. Volumes of slab-derived fluids, as determined by trace element models, are consistent with constraints from Sr-Nd-Pb isotopic systematics.

ABSTRACTS

101

ICOG 7

PROVENANCE OF THE PRE-DEVONIAN SEDIMENTS OF ENGLAND AND WALES: Sm-Nd ISOTOPIC EVIDENCE. EJ.THOROGOOD, Dept. of Earth Sciences, Parks Road, Oxford, U.K. eNd(t) values of the latest Precambrian Longmyndian sediments of England progressively fall from +1.7 to -1.8, indicating a provenance dominated by the underlying but penecontemporaneous Uriconian Volcanic Complex (8^^(600) -1.1 to +4.2), but with an increasing detrital contribution from preUriconian metamorphic sources (eNd(667) -8 to -7) up-section. By contrast Upper Cambrian and Tremadoc strata of England (deposited over c. 35Ma) have uniform Sm-Nd isotopic characteristics -8.5 to -7.5, = 0.103-0.116), typical of sediments sampling large areas of long established continental crust. Middle to late Silurian strata of central England and the Welsh Borderland (deposited over c. 10 Ma) also have uniform Sm-Nd isotopic characteristics (endiX) -9 to -8, ''^^Sm/^^Nd = 0.110-0.118). This Sm-Nd isotopic data suggests England and Wales received sediment from an adjacent large continent in the Upper Cambrian and Tremadoc and in the Silurian. These sources can be identified from faunal and palaeomagnetic evidence which places England and Wales close to Gondwana in the Cambrian and Lower Ordovician and close to Baltica (and also Laurentia) in the Silurian. In the Lower Ordovician, England and Wales (as part of the Avalonian microplate) rifted from Gondwana and moved northwards through the Ordovician as the lapetus and Tomquist Oceans closed. Ordovician and Silurian bursts of volcanicity in Wales supplied juvenile detritus but this contribution to contemporaneous sediments was diluted during and rapidly masked shortly after active volcanism by extensive reworking of older (lower £^^(0) sediment.

ARE ARCHEAN LEAD EVOLUTION MODELS FOR SUPERIOR PROVINCE UNIVERSALLY OR REGIONALLY APPLICABLE ? THORPE. R. I., Geological Survey of Canada, 601 Booth Street, Ottawa, Ontario, K1A 0E8, Canada If a series of leads evolve, single-stage, in a unifonn sourcefromTq until, at very different times, they form galenas, then the two-stage equation gives slopes and intercepts for lines from each lead to the initial source composition (Aq, Bq). The recast equation shows the slopes and intercepts defme a line yielding Aq and Bq as its negative slope and intercept. Pb isotope and zircon U-Pb data for W. Superior (WS) and Abitibi (Ab) volcanogenic massive sulphide deposits (VMSD) defme distinct paleoisochrons. If the Pb evolved singlestage, the linears indicate respective source ages of 4560 ± 250 Ma and 4415 ± 110 Ma. On this basis, and using data for syngenetic deposits with older known ages, realistic Archean models can be obtained. For the WS model, unpublished ages of 3238 ± 3 Ma for a Barberton deposit and 3465 ± 3 Ma at Yandicoogina, Pilbara (used for Big Stubby Pb), provide precise older control points. Parameters for the WS (To = 4560, Ao= 9.05, Bo = 9.8646, Co = 29.343) and Ab (To = 4415, Ao= 9.4094, Bo = 10.6306, Cq = 29.624) models are not obviously related to meteorite compositions. A calculated source ThAJ ratio of 4.108 ± 0.087 agrees with previous results. Slave Province VMSD, mosdy 2660 to 2700 Ma in age, fit the WS model very well, although the Slave linear is less steep (~ 0.88) than the WS and Ab lines. WS model ages for remobilized and epigenetic Pb at Isua, Greenland, Lumby Lake and Wawa, Ontario, are, as required, younger than respective host ages. Published ages for Pilbara and Yilgam VMSD are imprecise, but WS model ages for Whim Creek (2945 Ma) and for galena in Kylena Basalt (2724 Ma) agree with known constraints. WS model ages for Teutonic Bore, Kambalda, Golden Grove and Prospect X in Australia, seem too great, but Ab model ages, except for Duketon (2812 Ma) and Prospect X (2987 Ma), are no better. An intermediate model appears suitable for Teutonic Bore, Kambalda and Duketon. To establish if the WS and Ab models are globally applicable, or if regional discrepancies exist, will require precise ages for other Archean VMSD in S. Africa, Kenya, India, Finland, USSR and Australia. It is anticipated that mixing models will in future replace these approximate, although useful, single-stage models.

GEOCHEMICAL

EUJri VOLCANO,

EVOLUTION

JTAPAN.

TOGASHI, Shigeko and YAMAZAKI, Haruo, Geol.Surv.Japan, Tsukuba, 305, Japan Mt.Fuji is the highest and most representative active stratovolcano in the east Japan arc. The purpose of this paper is to present major and trace element concentrations and Nd-Sr isotope ratios and to discuss the source mantle of magma and the differentiation in magma chambers. To cover the lack of data, we analyze a large number of boulders in mud flow deposits and scoria in fall deposits of the older ejecta, of which age can be estimated by tephrochronology. Fuji Volcano has frequently erupted basalts of medium-K series and high-alumina basalt rock suite through the past 80,000 years. The Sr and Nd isotope ratios are constant and fall in the range of the most depleted island arc basalts (Fig.l). The concentrations of incompatible elements of the older stage are lower than those of the younger stages a t the same SiOa or FeO«/MgO level. The variation of depletion in incompatible elements with larger ionic radii is greater than those with smaller ones(Fig.2). Ni, Cr and Sr concentrations of each stage plot on trends associated with olivinespinel-plagioclsLse (-pyroxenes) fractionation. The magma of Fuji Volcano was supplied from a slightly heterogeneous depleted source mantle by steady state partial melting, changed with restricted fractional crystallization in short-lived shallow magma chambers, and erupted frequently. Fig.l Fig.2 100 l^Nd

EARLY EARTH: KASS-FLUXES, KAGKATISM AND DEGASSING I.N.TOLSTIKHIN, Geological Institute, Kola Sci. Center, Apatite, 184200, USSR Recent studies of noble gas partitioning among solid, melt and gas enable to express the rate of terrestrial uplifting degassing as a function of mantle melting, and partial degassing of melts and subduction of cooled material.A direct geochemical problem, describing these processes in terms of simple transport equations, was solved by means of numerical modelling. Ve infer from the model computation: the rates of mass-flux from the deep earth to its surface; i.e. to the reservoir subjected to degassing (and the rates of reflux), and the degassing rates which are different for different gases. Thus, the mass-flux amounted to g/year during initial interval 0.01 Ga of the earth evolution; about 70% of -'He, 97% of -'^-kr and 89% of ^ --^'Xe had been released from the mantle into the atmosphere at that time; the average flux was about 4*10^® g/year during the following 0.01-0.1 Ga and about 76% of ='He, 99% of --^ Ar and 91% of ^--^'Xe had been released by the end of this interval, the mean mass ratio between melt and solid F = 0. 1 being used. The above time scale clearly shows that the most intensive degassing of the earth occurred directly in the course of accretion. The degassing was accompanied by "U"fractionation of noble gases in the mantle: light gases (due to high solubility and preservation in partially degassed melts) and heavy ones (due to high partition coefficients and preservation in solids) were released from the mantle much more slowly than argon. Bulk concentrations of K=0.048%, U=0.038 ppm and the following contemporary fluxes (at/cm^sec) were computed: f(3He)=4.3; f (4He)=2*10'^- and f (40Ar) =0. 7*10'=-• The residence times for K and Ar in the upper nantle are 1.1 and 4.1 Ga, which gives a high K-Ar age 4.4 Ga, in contrast to the known estimates of about 2 Ga.

ICOG 7

102

JUVENILE HELIUM IH AJCIEITT ROCKS I.N.TOLSTIKHIir, I. L. KAMENSKY, V, S. DOKUCHAEVA & V.R.VETRIIT Geological Institute, Kola Scientific Center of the USSR Academy of Sciences, Apatite 184200 USSR, ®He/^He ratio is widely used as a geochemical tracer of recent geodynamic processes, it decreases in rocks and fluids of the continental crust from mantle values ^10*10-^- to crustal radiogenic ones 1*10-® during Ga. It has been shown, however, that in some cases rocks (minerals) well preserve He, trapped in the course of crystallization; correspondingly, the isotope composition of HeTRAR sheds light on the nature of ancient terrestrial fluids, A low ratio of HeRAo/HerRAR in rocks is provided by: a high conservation of HexRAR as compared with HeRAD and/or low ratio of (U+Th)/HerRAR. High concentrations of HCTRAR are observed in crustal charnockite rocks from Vezha Tundra Complex (the Kola Peninsula); their age, 2,8 Ga, was obtained by the U-Pb zircon method and a similar value by K-Ar one. Amphibole is shown to be a principle concentrator of He: ['^He] attains 670*10-^- cm^^Vg; the ratio of -=He/^He = 0.2*10"® is considerably higher than the radiogenic value; [-"He] is about 25 times higher than those which could be produced in situ by nuclear processes. In the course of crystallization there was an increase of ('^He/^^'Ar)TRAR ratio from ^2,5 in Hyp to >> 10 in Amph. The second case is typical of ultramafic rocks, ores and minerals from the Konche Pluton (the Kola Peninsula), characterized by CU] 0.001-0.01 ppm and C^He] up to 480»10-® cm®/g, which gives C^He]/[U] up to 520 (atomic) and U-He "age" 10 Ga. The ratios of -He/^He vary within the limits (0.16 - 0.74)*10-'='J they exceed the radiogenic value but are much lower than the mantle one due to contamination of the mantle melt by crustal fluids, carriers of crustal HeRAo. Certain relationships are found between -"He/^He and Mg/(Fe+Kg) values, showing that some olivine crystalIs were formed under higher temperatures as compared with those originated in a magma chamber.

SYSTEMATICS IN PRIMITIVE METEORITES TORIGOYE, Noriko, Geophysical Institute, University of Tokyo, Kongo, Tokyo, 113, Japan, and SHIMA, Masako, National Science Museum, Taito-ku, Tokyo, 110, Japan. In this work, Rb-Sr internal relationships of Qingzhen and Yamato-6901 (E3), Yilimia (E5), and Khairpur (E6) were analysed. This is the first report on both the least and the most metamorphosed enstatite chondrites. The present results are summarized in Table I. includ-^ ing other data obtained from previous investigations. Only silicate fractions in E3 show the ages much younger than 4.5 billion years, while data from sulfide phases are highly disturbed. Most of previous internal Rb-Sr works on type 3 chondrites show disturbed ®^Rb-®^Sr relations (mineral separates from Tieschitz H3 and Sharps LL3), or isochron with ages much younger than A.5 billion years (chondrules from Allende C3). The origin of these young ages together with elemental distributions will be discussed with comparing primitive chondrites to equilibrated chondrites. Table I. Rb-Sr relations in enstatite chondrites. Types Rb-Sr data from mineral separates 3 silicate phases - 2.1 Ga (this work) sulfide phases is highly disturbed 4

4.3 to 4.5 Ga [1-3] two of them show scattered isochron

scattered isochron arround 4.3 Ga [3]

silicate phases - 4.48±0.04 Ga (this work) sulfide phases is slightly scattered

References:[1] M . Shima and M . Honda, Earth Planet. Sci.Lett., 2, 337, 1967. [2] K . Gopalan and G . W . Wetherill, J . Geophys. Res., 75, 3457, 1970. [-3] J-F. Minster, L.-P. Ricard, and C . J . Allegre, Earth Planet. Sci. Lett., 420, 1979.

ABSTRACTS

TIME-SCALE CALIBRATION BY HIGH-PRECISION UPb ZIRCON DATING OF THE ERCALL GRANOPHYRE (Pe) AND VOLCANIC ASHES IN THE ORDOVICIAN AND LOWER SILURIAN STRATOTYPES OF BRITAIN TUCKER.R.D.. Ross,R.J.Jr., WILLL\MS,S.H. & PHARAOH,T.C. Geochronology Laboratory, Royal Ontario Museum, 100 Queen's Park, Toronto, Ontario CANADA M5S 2C6 Employing improved techniques to the conventional method of U-Pb dating, we have analyzed small zircon fractions (9-120 fig) from biostratigraphically controlled volcanic ashes and an intrusive igneous rock, and dated key Paleozoic time-markers with a precision better than 1% (±2 Ma). Four volcanic ashes from Britain for which fission-track ages were previously reported have yielded improved ages of: 438.7 ±2.0 Ma for the lower Silurian zone of Coronograptus cyphus at Dob's Linn, southern Scotland: 457.5 ±2.2 Ma for a Caradoc (Longvillian) ash near Bala, north Wales, and; 465.7 ±2.1 and 464.6 ±1.8 Ma for the Didymograptus artus Zone and the type Didymograptus murchisoni Zone, respectively, of the Llanvirn Series at Arenig Fawr and Abereiddi Bay, Wales. Another volcanic ash from the zone of Dicellograptus anceps (P. pacificus Subzone) of the Ashgill Series at Dob's Linn has been dated at 445.7 ±2.4 Ma, suggesting placement of the OrdovicianSilurian time boundary at approximately 441 Ma. A sixth volcanic ash from Caradocian strata of the central U.S. (Spechts Ferry Shale, Missouri) is 453.7 ±1.8 Ma old, indicating that the Rocklandian Stage of the Mohawkian Series is only slightly younger than the Longvillian Stage of the Caradoc Series in Britain. Finally, the Ercall Granophyre which rests unconformably beneath the Wrekin Quartzite Formation of Early Cambrian age (late Tommotian - early Atdabanian) is dated at at 560.2 ±2.0 Ma, thereby providing a reliable maximum age for the base of the Cambrian System in Britain.

GEOCHRONOLOGY OF A GREENSTONE BELT AND ITS ASSOCIATED GRANITOIDS. A . TUREK, Department of Geology, University of Windsor, Windsor, Ontario, N9B 3P4, Canada The Archean Michipicoten greenstone belt of the Superior province in Ontario evolved over ca. 220 Ma. U-Pb zircon ages indicate the belt formed between 2888 and 2668 Ma, as a result of repeated volcanism and plutonism. There were at least four volcanic cycles at 2888, 2743, 2717, and 2696 Ma, and five Plutonic events, four of which coincide with the volcanism, the fifth is at 2668 Ma and marks the Kenoran orogeny in this region. The volcanic rocks range from magnesium and iron rich tholeiitic basalts to calc-alkalic rhyolites. There is no relation between composition and age. The granitoids occur as plutons that are either internal or external to the supracrustal rocks. In terms of ages there is no difference between the internal or external plutons. The composition of the plutonic rocks is variable and includes trondjhemite, granite, syenite and gabbro, again there is no connection between composition and age. The external granitic terrains are of batholithic dimensions and formed by repeated plutonism over 220 Ma. The ages obtained for the Michipicoten belt are comparable to adjacent belts, however, they are not identical and there are discernable offsets in time between similar events in the different belts.

ABSTRACTS

103

ICOG 7

NOBLE GAS ISOTOPES AND HALOGENS IN FLUID INCLUSIONS G TURNER, F M STUART, M BANNON and J M SAXTON, Geology Dept, Manchester U n i v e r s i t y , Manchester M13 9PL, U K . There is great potential for using noble gas isotopes to study palaeofluids in fluid ingjusig^s. Opportunities are widened if the Ar~ Ar technique is used, since abundances of K , C a , C I , Se, B r , T e , I , Ba and U can be obtained from neutron induced A r , Kr and Xe isotopes. The combination of noble gases and halogens is particularly useful. Several areas are being studied to establish characteristics which can be used to define fluid sources and processes Fluids in vein quartz associated with the SW England granites correspond to a model of a rapid hydrothermal (meteoric) circulation system, with primary (Sn, W ) fluids being enriched in parentless ^^Ar and the later (polymetallic sulphides) fluids lying on a meteoric dilution trend. Fluids in quartz and fluorite from the (MVT) North P|gnines ore field are superficially similar but with Ar concentrations up to 10 times A S W , indicating a major unidentified fractionation m e c h a n i s m . Recoil effects in Ar in fluorite provide a measure of the size distribution of inclusions. A quartz vein in folded Carboniferous sediments has a composition typical of 1-200 Ma evolution of average crust and may be an example of conngte w a t e r . Several Phanerozoic cherts show low CI/ Ar ratios and negligible parentless a r g o n , characteristic of near surface fresh water hot spring deposits. In contrast all Proterozoic and Archaean g^erts analysed show high concentrations of parentless Ar relative to C I , suggesting involvement of deep slowlg circulating fluids rich in radiogenic gases. High Ar/ Ar ratios were probably typical of Proterozoic crustal^gluids because of the higher production rate of Ar and greater flux associated with crust formation. Cl-Ar correlations in H2O/CO2 rich micro inclusions in diamond imply concentrations of these fluids in the upper mantle less than 200ppm.

EVIDENCE FOR INVOLVEMENT OF AN ENRICHED LITHOSPHERIC MANTLE SOURCE IN LATEOROGENIC BIMODAL MAGMATISM, SOUTH AUSTRALIA TURNER. S., FODEN, J. and SANDIFORD, M., Department of Geology and Geophysics, University of Adelaide, GPO Box 498 Adelaide, South Australia 5001. Theoretical modelling indicates that thinning of the subcontinental lithospheric mantle is likely to occur towards the end of continental orogenesis, thus providing a geodynamic explanation for the late (post-deformation) magmatlsm common to many orogenic foldbelts. Because the subcontinental lithospheric mantle is likely to be enriched in incompatible elements relative to the convecting asthenosphere it is possible to test whether it constitutes a significant component in these late orogenic magnias. Intrusion of a bimodal magmatic suite consisting of gabbros and basaltic dykes, A-type granites and rhyolites coincided with the end of convergent deformation in the Cambro-Ordovician southern Adelaide fbldbelt. Intrusion occurred at high crustal levels within sillimanite-bearing metasediments locally containing kyanite, implying some 10km of denudation in the <10Ma allowed between the end of convergent deformation and magma intrusion, and suggesting that deformation was terminated by an increase in orogenic potential energy associated with mantle lithospheric thinning. The geochemistry of the magmatic suite shows enrichment of incompatible elements such as LREE, Kb, Zr, Y, Rb and K. However isotopic data permits only a minimal proterozoic crustal component (<5%), even in the silicic A-t}npe granites and rhyolites. eNd varies from +5 to -5 whilst initial Sr ranges between 0.703 and 0.707 placing the suite near the mantle array. The samples show a trend towards the low Nd, low Sr isotopic values typical of enriched mantle xenoliths rather than a trend towards evolved crustal material. The incompatible element enrichment therefore suggests a significant component from an enriched lithospheric''mantle. Nd model ages are around 0.8-1.4 Ga apparently predating the period of lithospheric growth associated with thermal sag following the extension during which deposition in the Adelaide Geosyncline was initiated (1.0-0.8 Ga).

THE GENESIS OF THE ORIKABE ZONED PLUTON, NORTHERN JAPAN: ISOTOPIC AND GEOCHEMICAL APPROACH UJIIE, Masumi, MATSUHISA, Yukihiro, and TOGASHI, Shigeko, Geological Survey of Japan, Higashi 1-1-3, Tsukuba, 305, Japan The Orikabe zoned pluton is one of the I-type intrusions of Cretaceous age in the Kitakami mountains. The rocks have a high-K calc-alkalic nature. The marginal facies is composed of a variety of gabbro to monzogranite, while the central facies, which intruded slightly later, is composed of granodiorite. Major element compositions change continuously with respect to SiOa content, except for K2O. The central facies rocks lie on a trend which is lower in K2O than that of the marginal facies rocks with the same SiOa content. The genesis and evolution of the magmas were studied in terms of chemical composition, and oxygen and strontium isotopes. Both the marginal and central facies rocks have values of +6.5 to +8.3%. which lie on a single, smooth trend, increasing with SiOa content. This 5'®0 range is. however, about higher than the mantle value. Initial Sr ratios were estimated to be about 0.7039 for the marginal facies. and 0.7042 for the central facies. From major element and oxygen isotopic data, a fractional crystallization model can explain the variation within the marginal facies. The central facies rocks can also be derived from the basic magma of the marginal facies by the same model with a different mode of crystal settling. The crystallization of hydrous minerals may have produced the difference in K2O variation trends. Strontium isotopic data suggest that there was a slight inhomogeneity of the source rocks, or assimilation of material with higher Sr ratios occurred for the magma of the central facies. Judging from isotopic data, the source of the magmas was crustal material which probably had basaltic compositions.

MASS FRACTIONATIONS OF MAGNESIUM AND SILICON DURING GASCONDENSATION PROCESS Chiaki Uyeda, Akira Tsuchiyama Takaiuitu Yamanaka and Jun Okano, Institute of Geological Science, College of General Education, Osaka University, Toyonaka, Osaka 560, Japan The mass fractionations in gas-condensation process was studied by measuring Mg and Si isotopes of laboratory condensed silicates, taking various conditions of the condensation process into account. Esat et al.(1986) measured previously the Mg isotope abundance in laboratory distillated samples using terrestrial pyroxene as the starting material and the mass fractionations observed in condensates and residues were comparable to those found in the CAI inclusions of Allende meteorites. Meanwhile, TsuchiyaiDa(1988) has carried cut a gas-condensation experiment in the Mg-Si-O-H system using a newly constructed furnace. The starting material (powders of synthetic forsterite crystals) was evaporated at about 1630°C. The evaporated gas firstly condensed as forsterite on the cold finger surface below llOO^C, and successively as mixtures of forsterite and Si at medium temperature, and as Mg-rich amorphous material below 500°C. In the present study, isotope abundances of Mg and Si were measured for each condensate phase, and the dependence wiih respect to the condensation temperature Tc was studied. The condensates with high condensation temperature were fractionated toward the heavy isotopes with respect to the starting material, for both Mg and Si. However as Tc decreases, the condensates became enriched with the light isotopes. Both ^^Mg/^-^Mg and ^°Si/2®Si ratios showed reduction of about 40tos Tc decreased from 1100°C to 400°C. The residue was enriched with the heavy isotopes. The mechanism of this large mass fractionation in condensates can be analysed by assuming the Rayleigh fractionation process between the oversaturated gas phase and condensed silicate phases. The Tc dependence of mass fractionation will be calculated from the fraction of residual gas, which are obtained from the measured amount of Mg and Si in the condensates, and compared with the experimental isotope values. REFEREtx^CES l)Esat,M.T.et al.(1986):Nature 319,576-578. 2)Tsuchiya ma,A. (1988):Abst.13th symp.antarct.meteorites,123-125. 3)Uyeda,C. et al.(1989):Secondary Ion Mass Spectrmetry,SIMS VII in print.

104

ICOG 7

CRUSTAL DEVELOPMENT OF THE FENNOSCANDIAN (BALTIC) SHIELD IN FINLAND Matti Vaasjoki and Hannu Huhma Geological Survey of Finland U - P b zircon and Sm-Nd whole rock data elucidate the development of the Precambrian crust within the Finnish part of the Fennoscandian Shield. After the formation of the Archaean craton between 3.1 and 2.6 Ga, little material was added to the continental crust over a period of 0.5 Ga, and the Nd model ages of about 2.3 Ga from turbidites have been shown to be a result of the mixing of Archaean and 1.9—2.1 Ga old Proterozoic sources. Layered gabbros, about 2.44 Ga in age, and 2.2-2.1 Ga old diabases are results of early Proterozoic rifting of the Archaean continent. They, and the 1.97 Ga old ophiolites, have f(Nd) values from - 2 to +4. The data on the 1.89-1.78 Ga granitoids show that the continental crust of the Svecofennian mobile zone was generated largely of mantle material when these rocks were emplaced. In contrast, coeval granitoids at the Karelian continental margin and transgressing the Archaean craton consist largely of remobilized Archaean continental crust. The anorogenic rapakivi granites, 1.67-1.54 Ga in age, consist of remobilized Svecofennian crust. Thus isotope data demonstrate two major periods of crustal growth in Finnish Precambrian: the first one in late Archaean times and the second during the culmination of the early Prroterozoic Svecokarelian orogeny.

PETROGENESIS OF HIGH SILICA GRANITES AND VOLCANITES IN WESTERN BERGSLAGEN, C.SWEDEN AS SEEN THROUGH NEODYMIUM ISOTOPE SYSTEMATICS: EVIDENCE FOR A 2.0-2.4 GA CRUSTAL PRECURSOR VALBRACHT. Peter J . , Laboratory for Isotope Geology, VU, De Boelelaan 1085, 1081 HV Amsterdam, The Netherlands The Early Proterozoic Bergslagen Older Granite suite and the associated comagmatic, >12 km thick, sequence of felsic metavolcanites in Central Sweden are 1.88 ± 0.02 Ga old according to published U-Pb zircon data. These rocks form part of a bimodal magmatic series, which includes about 8.5 vol.% tholeiitic basalts. The granites are fine- to medium-grained, high silica, biotite granites, which were emplaced at high levels in their co-magmatic extrusives, mainly pyroclastica, with rhyolitic plugs in higher parts of the succession. Both felsic and mafic rocks are affected by syn- to post-magmatic- hydrothermal subseafloor processes. The geochemistry of least-altered Older granites ranges from early I-type VAG (LREE enriched, Eu/Eu* > 0.71) to late S- and A-type WPG (flat REE, Eu/Eu' < 0.71), with Si02 contents between 72-79 wt.%. The felsic volcanites have characteristics similar to the I-type granites. Some volcanites show LREE depleted patterns with positive Ce anomalies. Seventeen granite samples show initial e^ variations ranging from +2 to -2. T^^ model ages vary between 2.0-2.2 Ga, with Sm/Nd ratios of 0.17-0.20. Nine felsic volcanite samples show initial e^. variations between +2 and -2. Tp^ model ages are 2.02.4 Ga, with Sm/Nd ratios of 0.20-0.23. Three felsic volcanite samples have initial e^^ of -4, -11 and -13, low LREE contents (Nd concentrations less than 3 ppm), positive Ce anomalies and Sm/Nd ratios of 0.38, 0.4 3 and 0.32, respectively. Except for these three samples there is no apparent correlation of initial c^j for the felsic rocks with either whole rock chemistry or geographic distribution. The anomalously low initial c^ of three felsic volcanites has been related to shallow seawater-rock interaction. The variations in initial of the other felsic rocks can be explained by either 1.88 Ga mixing of mantle derived material with older (Archaean?) crust or by partial melting of an earlier, 2.0-2.4 Ga Early proterozoic crustal precursor. The latter possibility is favoured because of the clear bimodality of the 1.88 Ga magmatism, the large volume (>20.000 km') of high silica magma, the lack of correlation between initial and whole rock chemistry and published reports of 2.2 Ga detrital zircons in Early Proterozoic metasediments to the south of Bergslagen. If the VAG characteristic of the early Older granites is taken as a memory effect, the precursor crust may have been formed in an about 2.2 Ga magmatic arc. A continuously southward receding subduction zone (1.90-1.86 Ga) could then lead to back-arc lithospheric extension in the Bergslagen region, with generation of magma through adiabatic mantle upwelling. The resulting mafic magmachamber could, next to being the lower crustal equivalent of the tholeiitic basalts, have acted as the heat source to generate the large volume of felsic magma and explain todays large crustal thickness (>«42 km) in western Bergslagen.

ABSTRACTS

LASER MICROPROBE AND CONVENTIONAL BrFs 5IS0 ANALYSES OF SIUAN GRANITE : A COMPARISON VALLEY. JW. ELSENHEIMER, D, KOMOR, S, BAKER, K, and CROWE, D, Dept. of Geology, University of Wisconsin, Madison, WI53706, USA Over 50 individual laser microprobe determinations of S^Ô have been made on a 2x2xlcm chip of granite from the Siljan Ring, Sweden. Targeted minerals were ablated using an 11 watt (CW) NdYAG laser, creating pits with an average size of 500x500p.m. Opaque minerals absorb the laser beam and react readily in 100-200 Torr of BrF5. Pit size depends on the duration of reaction, beam diameter, and reactivity of the mineral. The precision of the measurements is influenced by the amount of mineral inclusions or degree of alteration at each target site. Quartz, normally transparent to Nd-YAG radiation, was andyzable due to magnetite inclusions. Consistent technique (reaction time, laser power, beam diameter, cryogenic cleaning procedures) is essential to optimize precision. Multiple laser microprobe analyses of plagioclase megacrysts (conventional 5^^0=9.4) from an Adirondack anorthosite yields an average 5180=8.8±0.4 (n=17), suggesting that laser analyses are accurate within precision, or slightly b^Ô depleted. The Siljan sample is pan of a unique vertical section drilled 6.8km deep in mid-Proterozoic granites. Conventional oxygen isotope analysis of these granites shows that in quartz increases with depth from near 8 at the surface to 9.5 at 5760m, while feldspar decreases from near 10 to 7.1 (Komor & Valley, 1990, CMP). Samples from above a 60m thick diabase sill at 1500m depth are not equilibrated; reversed fractionations (AQt-Kfs=0.2 to -1.7) correlate with brick red color in granite and high S^Ô in calcite crack cements (to 22.3) indicating prolonged, low temperature alteration, conditions conducive to forming isotopic zonation. Laser microprobe determinations of in the red granite (1006m depth) are: Qtz 8.5±0.9 (n=23), Kfs 10.5±2.2 (n=20), Plag 7.3±1.1 (n=8). Conventional isotopic analysis of these minerals yields: Qtz?=8.6, and Flds=9.2. These data suggest that the range in S^Ô for the Siljan granite represents real sample heterogeneity. Analysis of S^Ô zonation will cause revision of previous estimates of the temperature and duration of hydrothermal activity. PETROLOGICAL AND GEOCHEMICAL EVIDENCE FOR OPENSYSTEM EVOLUTION OF MAGMAS IN THE BANDA ARC, EAST INDONESIA MJ. VAN BERGEN. P.Z. VROON and M.A. ELBURG, Institute of Earth Sciences, University of Utrecht, P.O. Box 80021, 3508TA Utrecht, the Netherlands Petrological and geochemical studies of active volcanoes of the Banda Arc indicate that shallow-level evolution processes involving fractionation, mixing and assimilation modified primary magmatic compositions to variable extents. The importance of these processes directly bears upon the mterpretation of isotopic signatures that are extremely 'continental' for an intra-oceanic island arc. The evidence is supported by bulk chemical and isotopic contrasts between mafic mclusions and host lavas, textural and compositional disequilibria of mafic phenocrystal and groundmass phases, presence of (?meta-) sedimentary xenoliths and of a (?resorbed) silica mineral in the matrix of some andesites, and the rare occurrence of zircon xenocrysts. Assimilation models were tested for selected LK, CA and HKCA volcanoes usmg reasonable 'continental' compositions and local sediments as endmembers. For the LK series of Banda crystal fractionation plus minor assimilation could account for small isotomc differences between basalts/andesites and dacites (®SrrSr = 0.7045-48), whereas combined fractionation and mixmg is consistent with the uniform isotopic compositions of lavas (Si02 = 5869 wt%) from the presently active cone. Petrological evidence for open-system behaviour is most pronounced m the volcanoes with 'continental' affinities. Models combining fractionation, mbdng and assimilation were successfully tested for Nila (HKCA) and are able to explain its isotopic ranges (®'Sr/®^Sr=0.7065-78). The highly phyric CA andesites of Serua, showing the most extreme 'continental' SrNd signatures, call for less conventional (but still acceptable) evolution mechanisms, partly because of restricted raises of trace elements concentrations in contrast to large isotopic variations (®'Sr/®^Sr = 0.7075-95). Mineral-chemical and textural evidence suggests that many of the andesitic lavas m the BA originate from mixing with more SiOj-rich magmas, possibly melted crust or sediments. Acidic lavas were not found at the active CA and HKCA volcanoes but rhyodacites are present in a bimodal association on Romang (''SrrSr = 0.7085-93), situated in an adjacent inactive segment where the arc has collided with the Australian continent. Although the effects of subducted continental contaminants cannot be dismissed, and along-arc variations in K-group and other incompatible elements are probably related to source heterogeneity, our data suggest that high-level interaction with the arc crust, sediments or partial melts has played an important additional role.

ABSTRACTS

ICOG 7

MORPHOLOGY OF ZIRCON AND BADDELEYITE AS A KEY TO THE INTERPRETATION OF U-Pb AGES OTTO VAN BREEMEN, Geological Survey of Canada, Ottawa, Ontario, KIA 0E8, Canada In general, the number of cr^'stal faces exhibited by zircon is proportional to the viscosity of the surrounding medium in which it grows. Thus zircons grown in peraluminous magma have more facets than those of peralkaline magmas. Metamorphic zircons, grown in a solid medium, have numerous facets. Cr}'stals are of round to ovoid shape with unequally developed facets resulting in a lack of s>Tmnetry. Gneisses of plutonic origin commonly contain multifaceted metamorphic rims on prismatic cores of igneous origin. In coronitic metagabbros, baddeleyite of igneous origin is often overgrown by aggregates of fine zircon. In some rocks these rims have been documented to be overgrown by metamorphic minerals belonging to the rock's main metamorphic assemblage, thus texturally linking zircon growth to metamorphism. In granulite facies terranes, the preservation of primary ages for cores of igneous zircon or baddeleyite testifies to the high closure temperature of the UPb system in both these minerals. Ovoid zircon grains with asymmetrical facets, frequently forming synneusis twins, may also be found in granite emplaced in shear zones. Etched sections of such zircons show regularly zoned cores surrounded by discordant and irregularly zoned mantles, both the same age. In these examples, magma with zircon growing on the liquidus is inferred to have been introduced into a partially cr}'Stalline mush and the irregular discordant zoning is attributed to impingement with surrounding grains moving in a dynamic environment. Similar internal zircon textures have been found in granites where cores and rims are separated by a significant age interval. These cases have been interpreted in terms of granite remelted during younger tectonism. Concordant isotopic chronometry of mineral growth during regional metamorphism: contstraints on thermal histories. D. Vance and R.K. O'Nions (Dept. Earth Sciences, University of Cambridge, United Kingdom) Understanding the thermal histories of rocks undergoing regional metamorphism requires not only information on the pressure and temperature histories of individual mineral grains, but also chronometric data of sufficient spatial resolution such that the P-T path preserved, for example , by a zoned garnet can be given a framework in time. Vance and O'Nions (1990) utilised recent advances in the techniques of isotope geochemistry to obtain precise information on the peak of the Ciedonian metamorphism in Newfoundland, as well as to set some limits on the timing of the prograde and cooling history prior and subsequent to this peak. An extension of those techniques to other well-studied regional metamorphic terrains is reported here and the generality of the earlier results is tested. Samples have been studied from both Vermont and the Swiss Alps. The rims of gamets from the Gassets Schist, Vermont yield concordant Sm-Nd and U-Pb isochron ages of 423.1Ma and 425.8Ma respectively. These are interpreted as dating the peak of metamorphism. In addition, detailed P-T modeling of the gamets has shown the metamorphic peak to have ocurred at 633 ± 26°C and 9.4 ± 3.0kbar and that the rocks were buried 16km and heated by 55®C during garnet growth. These results, when combined with RbSr cooling ages on micas and published ^Âr-^Âr mineral ages, yield cooling rates of 2-10°C Ma"l subsequent to the peak of metamorphism. Estimates for the timing of the peak of the Alpine metamorphism in the Swiss Alps are derived from U-Pb ages for gamets which, taken together with Rb-Sr ages on micas and published ^Âr-^Âr mineral ages, yield post-peak cooling rates of upto25XMa-l.

105 ISOTOPIC EVOLUTION OF SEAWATER: STATE OF THE ART AND LIMITATIONS VEIZER, Jan, Institut fiir Geologie, Ruhr-Universitat Bochum, 4 630 Bochum 1 , German Federal Republic The technical advances in mass-spectrometry over the recent decades enabled elucidation of isotopic composition of past seawater for sulfur, oxygen, carbon, strontium and neodymium. In all instances, geochronological limitations control the ultimate resolution of the curves. For strontium isotopes, an extension of the "high resolution isotope stratigraphy" to some Paleozoic intervals is feasible, but only to the limits of conodont zonation. For carbon isotopes, these limitations are compounded also by incorporation of ^^C depleted, biogenically generated, CO2 into the shells. In addition, recent data indicate an evolving A^^C (oceanic TDC/ organic carbon), with implications for carbon cycle, terrestrial greenhouse and the development of the biosphere. Still more complicated, and at this stage most controversial, is the nature of the oxygen isotope curve. "Hard rock" geochemists favour a near present-day for past seawater, whereas the "sedimentary" ones advocate an depleted ancient sediments, either because the ancient oceans were hot or because the magnitude of terrestrial rock-water interaction has been severely underestimated. Consideration of the known internal structures of isotopic curves for seawater indicates that they reflect non-linear systems. That is, they are two dimensional projections of multidimensional attractors. In future, the modelling should utilize therefore the approaches of "chaos" dynamics as opposed to the existing linear models.

BORON-ISOTOPE COMPOSITION OF MODERN BIOGENIC CARBONATES: A POTENTIAL NEW TRACER FOR PALAEOENVIRONMENTS VENGOSH, Avner (Research School of Earth Sciences, The Australian National University, Canberra, Australia), KOLODNY. Y., STARINSKY, A. (Department of Geology, the Hebrew University, Jersusalem, Israel), CHTVAS. A.R. and McCULLOCH, M.T. (Research School of Earth Sciences. The Australian National University, Canberra, Australia). The abundances and isotopic composition of boron have been examined by negative thermal-ionization mass spectrometry in modem, biogenic calcareous skeletons from the open ocean; Gulf of Elat, Israel; and the Great Barrier Reef, Australia. The 6 l l B values of the selected species (Foraminifera, Pteropoda, corals. Gastropoda, and Pelecypoda) range from 14.2%o to 32.2%o and overlap with 6l 1b values of modem deep-sea carbonate sediments (5l lB= 8.9%o to 26.2%o). The variations of 5 l l B may be controlled by isotopic exchange of boron species in which 10b is preferentially partitioned into the tetrahedral species, and coprecipitation of different proportions of trigonal and teu-ahedral species in the calcium carbonates. Low 5llB values (-•15%o) may indicate preferential incorporation of tetrahedral species, whereas the higher 51 1b values (~30%o) may indicate: 1) uptake of both boron species assuming equilibrium with seawater; 2) preferential incorporation of B(OH)4- from insitu high pH internal fluids; or/and 3) isolation of the internal reservoir from seawater. The B content and 5 l l B values of deep-sea sediments and Foraminifera shells are used to estimate the global oceanic sink of elemental boron by calcium carbonate and the consequent oceanic 1 iB-enrichment, as 6.4±0.9xl0l0 g/yr and 1.2±0.3xl0l2 permil-g/yr, respectively. The boron-isotope composition of authigenic, well preserved carbonate skeletons may provide a useful tool to record secular boron-isotope variations in seawater at various times in the geological record. In addition, boron isotopes may record variations of the alkalinity of seawater. The potential use of boron-isotope geochemistry in skeletons as a tracer for palaeoenvironments is demonstrated in Ostracoda and Foraminifera from the Gulf of Carpentaria, Ausu-alia. The 6l Ifi values of glacial-age, buried skeletons (4.0%o and 4.9%o) are lower than that of their modem equivalents (13.3%o and 17.6%o). This may reflect a "terrestrial" boron-isotope signature of the water in the Gulf during the Late Quaternary when it was isolated from the ocean.

106 Ar-Ar STUDIES OF MALANI COMPUEX IN RAJASTHAN, INDIA : EVUBNCES FOR DFFERENT MAGMATIC EPISODES. VENKATESAN, T.R., RATHORE, S.S., Physical Research Laboratory, Ahmedabad 380009, India and SRIVASTAVA, R.K., Geology Dept.. Univ. of Rajasthan, Udaipur. India. The Malani volcanic province, predominantly rhyoUtes, covering 50,000 sq.km. in Rajasliian is believed to represent a single magmatic event around 745 ± 10 Ma. Besides rhyoUtes, the province has granitic bodies, minor intrusives and extrusives of basic, intermediate and alkaline rocks. Detailed geochemical studies enabled identification of various rock types of Malani which will classify into four associations unrelated in time, space and origin. We have carried out Ai>Ar dating (Pande et al., 1988 for precedures) of various rock types from following three associations: 1) Basalt-andesite-daciterhyolLte association of Diri and Gurapratap; 2) intermediate alkaline rocks of Tavidar, Samau and Pokaran and 3) alkaline rocks of Mundwara, Sarnau DardaU. Ar-Ar plateau ages of 68.2 ± 0.5 Ma and 66.3 ± 0.5 Ma of a trachyte and an andesite respectively of Tavidar association are same within 2 ^ errors, but are slightly older than the rhyolites which have a weighted mean age (three samples) of 64.9 ± 0.2 Ma. This suggests that Tavidar association represents a much younger magmatic event. An essexite from Mundwara alkaline association yielded a plateau age of 76.0 ± 0.2 Ma which is different from that of Tavidar association. Age spectrum of Diri basalt does not yield a plateau but shows a cooling history with high temperature fractions (4 steps and 9.3% 39-Ar) yielding an age of 693 ± 5 Ma representing the Malani magmatic event. These results clearly indicate that the Malani complex preserves records of atleast three magmatic events. Reference: Pande et al., (1988) Geol. Soc. of India Memoir 10, pp 145-150. RADIOGENIC NOBLE GAS RATIOS AND DEGASSING OF MAGMATIC MELTS A.B. Verchovsky and Yu.A. Shukolyukov, Vernadsky Institute of Geochemistry and Analytical Chemistry, USSR Academy of Sciences, Kosygin str.l9, Moscow 117975, USSR Investigations of radiogenic and nucleogenic noble gas ratios such as "^He/^^Ne, ^He/^^^Ar, provide important information about noble gas fractionation processes.One of them is magmatic fractionation which is caused by differencesin the noble gas solubility in the melt. Measurements of the radiogenic noble gas ratios (mostly '^He/'^^Ar) in MORB glasses show that usually there is no solution equilibrium between gases in the glass and those in the vesicles.There are several conceivable reasons for it: (1) preferential loss of He from the glass mainly under natural but in some cases also under laboratory conditions, (2) in situ formation of radiogenic gases in the glass after eruption of the basalts, (3) occurence of two different types of vesicles in the glass, with gases which correspond to different levels of the degassing melt. One of the degassing steps probably takes place in the magmatic chamber and another during eruption of the basaltic melt. A simple numerical degassing model of the basaltic melt shows that the degassing is a multi-stage process. The model can explain the abundance of the light noble gases in MORB glasses but not that of the heavy ones. Most of the Xe is probably not degassed but stays in the melt. * Present adress: Max-Planck-Institut fiir Chemie, SaarstraBe 23, Postfach 3060 D-6500 Mainz, F.R.G.

ICOG 7

ABSTRACTS

SEAWATER ALTERATION EFFECTS ON SR - ND - PB ISOTOPE SYSTEMATICS OF MID - X E A N RIDGE BASALTS VERMA, Surendra P., Max-Planck-InstItut f. Chemle, Abt. Geochemie, Mainz, F. R. Germany and (present address) Depto. de Geoterinia, Institute de Invest Igadones Electricas, Apdo Postal 475, Cuernavaca, Mor. 62000, Mexico.

Using hot ("140 C) conc. HCI, a detailed Isotoplc and trace element acld-leaching study of "less-altered" and "more-altered" splits of a crystalline r i 2 Ma) and a glassy C I S Ma) MORB from NE Pacific has been carried out by collecting two successive acid-leached fractions (LI and L2) and residues (R) left after leaching, and analyzing ail of them along with several blanks for Sr, Nd and Pb isotopes and Rb-Sr, Sm-Nd and U-Th-Pb concentrations. During oxidation of the crystalline basalt, the "less-oxidized" (LO) and the "more-oxidized" (MO) splits have high 87/86Sr, similar 143/144Nd and significantly different Pb-isotope ratios. The inferred whole-rock (WR) isotope ratios for LO and MO show upon oxidation only a smaiI increase for Sr (0.70316 for LO to 0.70324 for MO), a very small and probably insignificant decrease for Nd (0.51320 to 0.51321), but a very large increase for Pb (18.16 to 19.91, 15.48 to 15.73 and 37.66 to 39.09). During palagonitizatlon of the basaltic glass, the "fresh" (FrG) and the "palagonitized" (PaG) splits show a very large increase in 87/86Sr, a very slight but statistically significant decrease in 143/144Nd and a small but significant increase in Pb-lsotope ratios. The inferred HR isotope ratios for FrG and PaG show upon palagonitizatlon a very large Increase for Sr (0.70277 to 0.70432), a very small decrease for Nd (0.51308 to 0.51310), and a small Increase for Pb (18.41 to 18.44, 15.51 to 15.52 and 37.92 to 38.02). During oxidation of the crystalline MDRB, the m Rb/Sr and U/Pb increased by -755; and -500X, whereas Sm/Nd, Th/Pb and ThAJ decreased by ' A X , -30* and '90% respectively. On the other hand during palagonltization of the basaltic glass, the WR Rb/Sr, Sm/Nd and U/Pb Increased by "2753:, "23; and "33; respectively whereas Th/Pb and Th/U both decreased by "603;. These data cannot be explained by: possible contamination during drilling or sampling, the measured blank-levels, simple two-component mixing with average Pacific seawater, or in-situ radiogenic growth after emplacement. A viable mechanism may be an elemental and isotoplc exchange (under seawater-dominated conditions) between basalt and seawater that had interacted earlier with overlying sediments.

Pb AND Nd ISOTOPIC STUDY OF THE FOOTWALL ROCKS OF THE NORANDA Cu-Zn SULFIDE DEPOSITS, QUEBEC VERVOORT, Jeffrey D. and WHITE, WiUiam M., Dept. of Geol. Sci., Cornell Univ., Ithaca, NY 14853, USA. Mafic to felsic volcanics of the Blake River Group in the southcentral Abitibi greenstone belt of Canada are host to the Noranda volcanogenic massive sulfide (VMS) deposits. Whole rock Pb-Pb data from two rhyolitic units (Nonhwest Rhyolite-NWR and Amulet Rhyolite-AR), an andesitic unit (Rusty Ridge Andesite-RRA) of the footwall volcanics, and an intermediate plutonic unit (Flavrian Granodiorite-FG) that intrudes it fall on a line with an age of 2.694 ± .014 Ga., within error of zircon ages of units within this sequence (2.698-2.701 Ga). This line also passes through a tighdy defined field of the Noranda sulfides, suggesnng that the Pb in the sulfides and the footwall lithogies share a common origin and that the formation of the sulfides and the volcanic pile were contemporaneous at approximately 2.7 Ga. The large spread and colinearity in whole rock Pb-Pb ratios indicate that U and Pb were highly mobile during hydrothermal alterarion at or shonly after formation of the volcanic pile but have not been significantly disturbed since that time. We have not yet been able to resolve, on the basis of this preliminary data, the relationships between mafic and felsic lithologies or their relative contribution to the formation of the sulfides. Initial £K(1 values rangefrom+2 to -r3, indicating the footwall rocks were mantlederived, or at least contained a substantial recent mantle component. Based on this and the Pb-Pb co-lineark NWR ity of sulfides and • FG footwall rocks, it 0 RRA seems likely the Pb ,20 i AR • sulfide in the sulfides was wholly or largely of recent mantle origin at the time the sul16 fides formed. Thus the VMS sulfide 14 data may be useful 10 20 50 40 50 60 in defining the mantle Pb isotope 205 Pb/^^Pb evolution curve.

ABSTRACTS

107

ICOG 7

AFAR MAGMATISM: NATURE AND EVOLUTION O F MANTLE SOURCES VTDAL Ph..Univ. Blaise Pascal C l e r m o n t F e r r a n d , France, VELLUTINI P.J.,PIGUET P.,ISERST, Republique Djibouti, VINCENT J.,COULON Ch., Univ. Marseille , France. Afar triangle is located at the junction between three rifts : two oceanic (Red Sea and Aden) and one continental (Ethiopia). It is a very favourable area for understanding the interactions between lithosphere and asthenosphere in the course of the opening of a continental rift by progressive stretching and thinning of continental crust, leading to oceanisation. The present work focuses on the territory of Djibuti Republic where volcanism has occurred almost continuously since 25 Ma. old rhyolites to the presently active Asal rift which has been described as an emerged segment of a midoceanic ridge. With the exception of the initial rhyolitic episode which corresponds to crustal melting, the isotopic (Sr, Nd, Pb) data plot in all diagrams in the fields of intraplate oceanic volcanism. In addition, as the oceanisation process goes on, the sources are evolving from an enriched mantle source t o w a r d s the MORE field. However, a t h i r d component (HIMU?) must be taken into account since the Sr-Pb isotopic trend intersects the MORB field at its enriched end. In addition, basalts from the present active axial zones (Asal and Manda Inakir) are unlike typical depleted MORB. In conclusion, our data allow to show that the oceanisation process is not yet c o m p l e t e d , .contrary to what has been previously suggested.

THERMOCHRONOLOGY ? LM. Villa (1st. Geocronologia CNR, Pisa, Italy) The detennination of the cooling rate of rocks relies on measuring cooling ages on minerals of known closure temperature. Since minerals show an amazing number of parameters governing the loss of radiogenic daughters, the optimum would seem in principle to determine age and individual closure temperature in one single experiment. In the case of Ar, this is impossible for hydrated minerals, leaving Kfeldspar as the only common K-rich mineral. However, K-feldspar is mineralogically very complex (see e.g. Parsons & Brown 1984) and a number of theoretical shortcuts have been used to make some sort of first-order mathematical treatment. There is supposed to be only one process operating from 100 to 1000°C, which has a single activation energy E. Deviations from simple relationships are hocussed away by a suitably large number of fudge factors involving ad hoc definitions of "effective gram sizes**. The energy required for removal of neutral Ar from the sanidine lattice, 50 kcal/mole, by far exceeds its van der Waals binding (2-4 kcal/mole) and resembles the diffusive energ>' of lattice-forming ions such as Mg, Fe and O in olivine (42-72 kcal/moleJurewicz & Watson, 1988). Thus, At can only diffuse by pushing past the lattice atoms and exploiting extended and point defects. Manning (1974) derives diffusion equations for each type of defect which generalize Pick's Law. Lasajga (1981) argues that different transport r e ^ e s are operating at increasing temperatures because of the different kinetics of the defect types involved. Experimental results (Fechtig et al, 1963; Huneke et al, 1967) had shown that isothermal heating protocols can overcome some of the artefacts; the latter showed that He and Ne diffusion was indeed a sum of processes with increasing activation energies. The present experimental results on a dozen aliquots of a granitic microcline show that: (1) Arrhenius trajectories are a function of the heating schedule, as is intuitive; (2) Arrhenius alignments are formed in most experimental conditions, but they have variable slop^; (3) etching after irradiation provides a direct link between leachability (i.e. enhanced density of extended defects: Lasaga, 1981) and Ar loss; (4) analysis of two sized splits (75 |i; 75-300 jli) of a synthetic glass obtained by melting & quenching (= disordering) the microcline uphold Parson et al's (1988) suggestion that perthite boundaries are no Ar pathways. So where does all this leave thermochronology, in its published (and renounced) or unpublished varieties? It is hard to see how it ever could make the transition to a "-metry", i.e. a science which can correctly quantify parameters (e.g. geobarometry, geothermometry); it will remain associated to astrology and taurocoprology.

SOURCE CONTAMINATION VERSUS HIGH-LEVEL ASSIMILATION IN THE BANDA ARC VOLCANICS: SR^D,PBISOTOPIC EVIDENCE P.Z. VROON. M J. VAN BERGEN, Institute of Earth Sciences, Univ. of Utrecht, P.O. Box 80021, 3508TA Utrecht, the Netherlands, and W.M. WHITE, Dept. of Geological Sciences, ComeU Univ., Ithaca, NY 14853. U.SA. The Banda Arc (East Indonesia) is reputed as an intra-oceanic island-arc where magmagenesis is influenced by subducted terrigenous material. Anomalously high Sr- and low Nd-isotopic ratios of the volcanics compared to other island arcs support this model (e.g., Whitford and Jezek, 1979, Whitford et al., 1981). We discuss the relative importance of subducted source contaminants and high-level assimilation in the arc crust, based on new Sr-Nd-Pbisotopic and chemical data of six active volcanoes along a 500km stretch of the BA, as well as of sediments spatially distributed parallel to these on the adjacent Australian shelf. Isotopic ranges of the volcanics are extremely large with Banda (LK) and Manuk (CA) in the NE having 'oceanic arc like-' (87/86 = 0.7045-55, 143/144 = 0.51291-73, 206/204=18.6-18.7), and ISTila, Teon and Damar (all HKCA) in the SW showing 'continental' affinities (87/86 = 0.7065-83, 143/144=0.51269-52, 206/204 = 193-19.4). Serua (CA) in the central part of the arc is the most extreme in Sr and Nd ratios (87/86 < 0.7095, 143/144 >0.51240), though not in Pb (206/204 <19.10). Except for Banda and Manu^ vwthin-suite 87/86 and 143/144 variations are large (but always limited for Pb), and are consistent with hi^-level assimilation processes. Correlations with differentiation in^ces, petrographic evidence and presence of (meta)sedimentary xenoliths corroborate such models. However, even the most mafic and least contaminated rocks in the SW BA (and the easternmost Sunda Arc) still display relatively high 87/86 (-0.7065) and low 143/144 («0.51260). These areas are close to the locus of collision with the Australian continental margin, and coincide with a rerional anomaly of low ^e/^He (Hilton and Craig, 1989). Averaged ^Nd/^'^Nd of the sediments change from 0.51220 in the NE to 0.51200 in the SW, consistent with the along arc variation in the volcanics. Preliminary Pb-isotope results show a similar shift but also tend to suggest that Pb in the sediments is not sufficiently radiogenic to explain the signatures of the volcanoes in the SW. Our results call for multiple components superimposed on variable(?) mantle signatures to explain the isotopic and chemical ranges found. These probably include subducted continental material, but the extreme isotopic anomalies are caused by contributionsfi-omhigh-level assimilants.

ISOTOPIC EVIDENCES FOR METAMORPHIC FLUID MIGRATE THROUGH GRAIN BOUNDARIES IN MARBLE DURING METAMORPHISM H.VADA,Institute of Geosciences, Shizuoka Univ.Shizuoka, 422,Japan, Y.Arita, Department of Earth Sciences, Nagoya Univ. Nagoya, 464,Japan and K.Iwaki, Institute of Geosciences, Shizuoka Univ.422, Japan. During metamorphism, the metamorphic fluid is an important clue to transportation of chemical elements and heat. The migration mechanis of the fluid is a question under debate vhether such fluid migrates pervasively in the whole rock or channelizedly through vein or fracture systems. Marbles are believed to have low permeability for iBetamorphic fluid and to be relatively impermeable property during metamorphism. On the other hand, metamorphic fluid reacts with impure limestone and sometimes releases a large amount of carbon dioxide. Thus the interaction between metamorphic fluid and the limestone is important for metamorphic processes. We intended to clarify the metamorphic fluid migration pass through the limestones in various types of metamorphics. Ourspecial attention was paid for isotopic heterogeneity or isotopic zonation of individual calcite and graphite cyrstals. For the contact metamorphism in the Hida metamorphic belt, central Japan, isotopic homogenization of metamorphic fluid was mostly attained within a channelized zone near the granite contact where decarbonation reactions occurred. But the other part of marble, calcite grain have an isotopic zoning. The cores of individual calcite and graphite crystals have an isotopic compositions which are same to the values before the contact metamorphism. Only the grain boundary, the lowering of isotopic ratios specially in oxygen can be observed. This is caused by the reaction with metamorphic fluid migrated into the marble. Such oxygen isotopic trough around the grain boundaries within a few hundred micrometers is strongly related to the distance from the granite contact. The isotopic trough diminished at the locality about 1.3km from the contact. The interaction between metamorphic fluid and minerals that is not known by the mineral paragenesis gives an inportant knowledge to qualitative transportation of metamorphic fluid in the crust Ve will discuss the migration mechanism of metamorphic fluid in the other localities of the Hida sietamorphic belt and a different types of metamorphic region of the Adirondack Mountains, New York.

108

PROJECTED FISSION-TRACK LENGTH INFORMATION IN THE THERMAL HISTORY RECONSTRUCTION OF ROCKS WAGNER, Gunther, A., HFJL, Ewald, Max-Planck-Institut fur Kemphysik, D-6900 Heidelberg, Germany and VAN DEN HAUTE, Peier, Laboratorium voor Aardkunde, Rijks Universitet, B-9000 Gent, Belgium In order to reconstruct the low temperature tail of the geological T-t evolution, the use of the fission-track record in apatite is now widely accepted. Numerous successful applications to various rock-types and geological settings have been reported. Apart from the regional and vertical variation of the fission-track ages it is the length distribution of spontaneous fission-tracks in apati te that gives information on the thermal history of rocks. As a diagnostic tool for thermal history analysis, commonly the lengths of "confined" tracks are used. Since confined tracks are etched within the cr>'stal volume they are different from the tracks ("semi-tracks" or "projected" tracks) etched on polished surfaces for age determination. While working on apatite samples from the German Deep Continental Drilling Project KTB a system for T-t-path reconstruction has been developed that is based on projected track lengths. For monotonously cooling rocks this system allows the determination of the 60'C-cooling age in addition to the lOO'Ccooling age conventionally derived from apatite fission-track ages. In particular, the core-samples from various depths and actual temperatures within the KTB pilot drill-hole are used to test the validity of the proposed system under geological in-situ conditions. These results will be presented. Various examples of reconstructing the thermal history of the co'Stalliiie basement and the sedimentary basins in the KTB surroundings will be reported.

Ar- Ar LASER PROBE DATING OF AUTHIGENIC FELDSPARS AS A TOOL FOR INTERPRETING FLUID INCLUSION DATA F.WALGENWITZ, Elf Aquitaine, 64018 PAU Cedex, France M.PAGEL,CREGU,BP23,54501 VANDOEUVRE LES NANCY,France H. MALUSKI, USTL, 34060 MONTPELLIER Cedex, France Fluids trapped in inclusions at a P-T combination above the liquid/vapor curve, homogenize at a temperature (Th) lower than the trapping temperature (Tt). Hence a temperature correction (5T), whose the magnitude depends on the trapping pressure, must be added to Th to get Tt (Tt - Th + 6T). In sedimentary basins, pressure may be estimated from burial curves if the time of fluid inclusion trapping in authigenic minerals is known. Most commonly the age of trapping remains unknown because mineral composition is unsuitable for radiometric dating (quartz, carbonates...). Authigenic K-Feldspars from hydrocarbon bearing reservoirs on the Angola margin, offered a unique chance in dating the trapping of primary oil and aqueous inclusions, and tjiys i^^estimating the pressure from burial curves. The Ar- Ar laser probe method has been used on polished rock slides for this purpose, avoiding Argon contamination from detrital minerals when using the conventional K-Ar method on isolated mineral fractions. A lower Miocene age was found (25.6 +3.2 Ma), indicating that pressure at trapping was close to the present reservoir pressure. The temperature correction, as a result of pressure, leads to Tt in fairly good agreement with Tt determined by intersecting isochores of oil and aqueous fluids.

ICOG 7

ABSTRACTS

RB-SR OPEN SYSTEM BEHAVIOUR AND MINERALISATION IN BUSHVELD GRANITES, SOUTH AFRICA WALRAVEN, F., STRYDOM. J.H., and STRYDOM. N., Geological Sur\^ey, P/Bag X112, Pretoria 0001, South Africa. The variation of Sr-isotopic compositions has been studied in and surrounding Sn pipes occurring in the Bobbejaankop and Lease Granites in the northern lobe of the Bushveld Complex, South Africa. These pipes formed conduits for late- magmatic to hydrothermal fluid circulation and are frequently richly Snmineralised. Calculated model ages and the ^^Sr/^^Sr ratios at the time of granite emplacement are anomalous and are lower in the immediate vicinity of the Sn pipes than in the granites elsewhere. They are at a minimum in the Sn pipes themselves. This pattern complements that observed in the Bushveld granites on a larger scale where low Rb-Sr ages are obtained which correlate with the degrees of alteration and differentiation of the granites, their stratigraphic position and their proximity to mineralisation. Loss of radiogenic ^^Sr is considered to be the primary cause of the low ages and low ratios and is attributed to latestage, low-temperature hydrothermal fluids interacting with the granites. These fluids evolved from the early, high-temperature fluids responsible for the Sn mineralisation and form part of an extremely long-lived hydrothermal system. The distribution of ^^ST^ST ratios in the vicinity of the Sn pipes suggests that the degree of ^^Sr loss directly reflects the intensity of fluid-rock interaction and that this was greater where fluid flow was at a maximum. This in turn suggests that the degree of ^^Sr loss can be used to delineate zones of greater fluid-rock interaction which therefore constitute areas of potential Sn mineralisation.

GEOCHEMICAL AND TEMPORAL CONSTRAINTS ON MAGMA GENESIS OF THE NGORONGORO VOLCANIC HIGHLAND, TANZANIA WALTER, Robert C., Geochronology Center, 2435 Ridge Road, Berkeley, CA 94709, MANEGA, Paul C. and FARMER, G. Lang, Geology, Univ. of Colorado, Boulder, CO 80309, HART, William K., Geology, Miami Univ., Oxford, OH 45056 The Ngorongoro Highland consists of seven major volcanic centers, including three large caldera structures. The oldest rocks are early Pliocene carbonatites and alkaline silicate rocks associated with Sadiman (Hay, 1987). K-Ar, 40Ar/39Ar and petrographic data, suggest a general temporal and compositional progression from southwest to northeast across the complex. The Ngorongoro caldera, northeast of Sadiman, erupted quartz-trachytes from 2.5 to 1.8 Ma; the Olmoti caldera erupted trachytic ignimbrites and lavas from 1.75 to 1.1 Ma; and the Embegai caldera erupted nephelinetrachyte tephras and lavas 0.6 Ma to 0.5 Ma. Two units adjacent to the highland, a phonolite from Engelosin and a trachyte from Molambo, are dated at 2.68 Ma and 1.04 Ma, respectively. Preliminary isotopic data demonstrate that Engelosin has 87Sr/86Sr and eNd values of 0.706771 ± 10 and -15.0 ± 0.3, respectively, whereas Molambo (Olmoti?) and Embegai have 87Sr/86Sr and eNd values of 0.703953 ± 7 and -0.43 ±0.1, and 0.704085 ±6 and 1.1 ± 0.1. The relatively low 87/86 and high eNd values for Molambo and Embegai suggest that these lavas were derived in pan from an OIB-type asthenospheric (?) mande, perhaps with a component from a higher 87/86 and lower eNd source involved as well. The Engelosin stock, west of the main boundary fault, has substantially higher 87/86 and lower eNd values and possesses a higher alkali content than Molambo or Embegai. These data indicate that Engelosin may have derived largely from ancient (Pre-Cambrian) lithospheric mantle, which suggests that at 2.7 Ma, such mande was still intact, at least at the margins of the developing rift. We anticipate that a detailed study of the Ngorongoro Highland volcanic rocks will yield an understanding of the response of the deep-lithosphere to rifting, based on spadal and temporal variations in elemental, and Sr and Nd isotopic data.

ABSTRACTS

ICOG 7

109

OXYGEN AND CARBON ISOTOPIC COMPOSITIONS OF MOLLUSKS FROM THE LATE PLEISTOCENE SZEKOU FORMATION, SOUTHERN TAIWAN WANG. Chuna-Ho and PENG, Tsung-Ren Institute of Earth Sciences, Academia Sinica, P.O. Box 23-59, Taipei, Taiwan 10098, ROC Oxygen and carbon isotopic compositions have been determined in fossil molluscan shells from the late Pleistocene Szekou Formation of southern Taiwan. These fossil mollusks have been well preserved based on evidences of X-ray diffraction, stable isotopic data and their occurrence in the field. The following isotopic ranges were found (relative to PDB standard): b^^O, -2.36%o to 1.62%o;

3 c , -3.26%o to 3.08%o.

The stable isotopic data of fossil shells indicate that they lived in environments beginning with a pure marine one and ending with a freshwater influence. Heavier oxygen and lighter carbon isotopic ratios of these fossils relative to modern counterparts around Taiwan indicate that these fossil mollusks grew in a glacial period (oxygen isotope

Sm-Nd, Rb-Sr AND U-Pb SYSTEMS IN GRANITE-PORPHYRY AND SKARN ORE DEPOSITS IN THE SUZHOU REGION, EASTERN CHINA WANG Yinxi. YANG Jiedong, Modem Analytical Center, Nanjing University, Nanjing 210008, China, and WANG Jianming, YANG Nianqiang, Institute of Geology, Mineral Resources, Test and Application of Jiangsu Province, Nanjing 210008, China The Suzhou granite-poiphyry body is located on the active Mesozoic continental margin of eastern China. The skam ore deposits (Pb, Zn, Ag and Cu) are usually found in the nearby contact thermal zones between the granite-porphyry body and the wall rock. No dimetasomatism was found in the contact zones. The U-Pb isochron of 5 zircon samples from the granite-porphyry body gives T = 243 ± 17 Ma. The geochemical characteristics from the body are: = 10.5, (87Sr/86Sr)i = 0.7080, d^^O = +3%o and 5Eu = 0.49-0.65. This body is an I-type granite. Its genesis possibly is as an intrusion of upper mantie-derived magma along a WE trending deep fault, successively incorporating upper crustal materials. The ore body of the skam has been dated by the Sm-Nd and Rb-Sr isochron methods at 232 ± 7 Ma and 201 ± 11 Ma respectively. The geochemical characteristics from the ore body are: e (J) = -10.2, (87Sr/86Sr)i = 0.7098 and 5Eu = 0.55. Sm-Nd age reflects the age when the skam ore deposit was formed. According to the information, the granite-porphyry has been thought to be a parent of rock in which the skam ore deposit was formed. The source material of the ore deposit was mainly derived from the lower crust and the upper mantie-derived magma.

stage 3), as supported by the "^^C dating results. The calculated mean uplift rate for the Szekou Formation shows a comparable value ( - 3 mm/yr) to the value of Holocene mean uplift rate in this area, suggesting a constant uplifting since the last 45 Ka.

ABIOGENETIC METHANE FROM HOT SPRING IN VOLCANIC ROCK AREA AT PLATE COLLISION BOUNDARY WANG Xianbin, XU Sheng, VTN Qibin, Laboratory of Biogeochemistry and Gasgeochemistry, Lanzhou Institute of Geology, Academia Sinica, Lanzhou 730000, China and XUE Xiaofeng, WANG Wenyi, Yunnan Province Institute of Geoscience, Kunming, China Whether or not there exists large amount of original abiogenetic methane in the internal earth has aroused interests of many researchers all over the world. Chemical and isotopic compotion of hot spring gases in volcano area of Tengchong (Yunnan, China) show their characteristics of abiogenetic deep source. The concentration of He is in the range of 31.5-129.5 ppm, that of Ha is 75.5-801.8 ppm, and CH^ 0.34-1.03%; Carbon isotope composition of methane is 6^^C=-15 — -20.8%o; And the R/Ra ratio of He is 2.9-4.3. Tengchong volcano rock area is in the eastern parts of the collision boundary between Indian Plate and Eurasian Plate. From Neogene to Holocene, a series of volcanic eruption of different lithology accured due to alternate actions of tenssion and compression. Tectonic activity along the plate collision boundary advantaged the original abiogenetic gases in the internal earth to migrate up toward the surface of it.

This project is supported by the NSFC

PETROLOGY AND ISOTOPE GEOCHEMISTRY OF SPINEL ± GARNET LHERZOLITES FROM PALEI AIRE, S. CHILE WARE, A.J., The Open University, Milton Keynes, UK, KEMPTON, P.P., HARMON, R.S., NERC Isotope Geosciences Lab, Keyworth, UK, STOSCH, H.-G., Universitat Koln, Koln, FDR, and LOPEZ-ESCOBAR, L., Universidad de Chile, Santiago, Chile Metasomatic processes and the origin of trace-element enrichment in the subcontinental mantle are considered in the context of major and trace element mineral chemistry, and Sr-, Nd-, and 0-isotope data from spinel (sp) ± garnet (gar) Iherzolites from the Palei Aike (PA) volcanic field, S. Chile. PA peridotites have Nd- and Sr-isotope values that overlap the field of MORE (0.51309-0.51292; 0.7024-0.7038). Most cpx and v^^hole rocks have LREE-enriched, concave-downward patterns unlike either Type la (LREE-depleted) or Type lb (LREE-enriched) sp Iherzolites worldwide. HREE contents are low, similar to cpx from gar Iherzolites. Ion microprobe data confirm this pattern in both core and rim, but rims may be 2X enriched in all REE. AI2O3 contents of most PA cpxs (2.6-6.6wt%) are lower than Type la sp Iherzolites worldwide, and overlap the fields for Type lb sp and gar Iherzolites. values (%oSMOW) for xenolith minerals are: ol=-f5.1 to -1-6.0, cpx=-f-5.7 to +5.9, and opx=+5.8 to -1-6.0; Ax-y plots involving olivine cut across the line A=0 with a steep positive slope. 0-isotope disequilibrium and REE enrichment of cpx rims relative to cores are interpreted as the result of fluid-rock interactions during metasomatism. However, other aspects of the major and trace element data are poorly explained by.conventional models of mantle metasomatism and better understoodan terms of magmatic/metamorphic processes operating in the mantle.

110

ICOG 7

ABSTRACTS

Mechanisms of argon release from Himalayan metamorphic hornblendes J. WARTHO. D.C. REX, P.G. GUISE, R.A. CLIFF, & RJ. KNIPE, Department of Earth Sciences, University of Leeds, Leeds, West Yorkshire, LS2 9JT, U.K. Hornblende breakdown and the systematics of argon release during routine "^Âr-^Âr stepwise dating are open to speculation. Mechanisms of argon release have been investigated in a series of vacuum-heating experiments on hornblende samples so that better constraints can be placed on the interpretation of discordant 'U'shaped age spectra. Splits of unirradiated hornblende samples have been heated to known temperatures and analyzed using a variety of techniques including electron microprobe, scanning electron microscope, transmission electron microscope, and powder camera X-ray diffraction. Photoemission electron microscopy has been used to make 'real-time' observations on vacuum-heated geological specimens up to temperatures of lOOO'C. Detailed studies of the breakdown of heated hornblendes reveal a sequence of decomposition events correlated to argon release: 1) HomblendeÔxyhomblende (initial release at 750- SOO'C), 2) Oxyhomblende-^"Amphiboloid" + Fine-grained products (major release at 990-1030'C), 3) "Amphiboloid" + Fine-grained products^Mg-rich glasses (secondary major release >1080'C). The studies so far indicate that the release of argon during "^Âr^Âr stepwise heating is completely controlled by the breakdown sequence of the hornblende structure.

COMPLETE PbAJ-ANALYSIS OF UNSPIKED SAMPLES BY MEASURING Pb-ISOTOPES ONLY WENDT. I., and CARL, C., BGR, Hannover, HABFAST, K., and TUTTAS, D., Finnigan MAT, Bremen, and WENDT, J.I., MPI, Mainz In certain cases it is desirable to analyse small samples without chemical separation and spiking of U and Pb (single grains). However, a 207pb/206pb-aee can be determined by Pb isotopes only, assuming concordant 207pb/235u and 206pb/238u-ages. Due to completely different evaporation and ionisation conditions for Pb and U no PbAJ ratios can be obtained. However, in case of radioactive equilibrium the radioactive Pb isotope 2l0pb and the isotope 238u have a fixed ratio: 210pb/238u = 238x./2i0;^ = 5.(X) • 10"9, i.e. a (concordant) age of 300 Ma which corresponds to 206pb,/238u-ratio of 4.76 . 10-2 would have a 210pb/206pbr-ratio of 1.05 • 10"7. We have used the multicollector system of the Finnigan 261 spectrometer where masses 204,206,207, 208 are measured in Faraday cups and mass 210 in an electron multipler detector and ion counting. The system was calibrated by artificial mixtures of 2l0pb and 206pb (together with 207pb, 208pb and 204pb) with 210pb/206pb-ratios of 10-3, io-4, io-5.10-6 and lO'^. It could be shown that for ratios < 10-5 the ^OSpb-abundance sensitivity to mass 210 of 1.7 • 10-7 has to be considered. The results of some ciystals of different torbemite and zircon samples measured by means of the conventional isotope dilution methods as well as by the 2l0pb/206pb.method were in good agreement.

DETECTION OF CONFINED 238u FISSION-TRACKS IN MINERALS BY HEAVY ION IRRADIATION

DECIPHERING THE ISOTOPIC EVOLUTION OF DEPLETED MANTLE WILLIAM M. WHITE, Dept. of Geological Sciences, Cornell University, Ithaca, NY 14853. USA There are two possible approaches to establishing the isotopic evolution of the depleted mantle: study of ancient melting products and inferences bas^ on present-day isotope and parent/daughter ratios. Both approaches are beset with difficulties. The rarity of preservation of ancient MORS, and the difficulty of its identification is the principal problem with the first approach. Nevertheless some inferences are possible. Positive eNd and enf values in many Archean mantle-derived rocks suggests rapid early depletion of the upper mantle and more or less linear subsequent growth of isotope ratios, implying no further net depletion. The second approach, inferring evolution from present day isotope systematics, requires a knowledge of parent/daughter ratios in the depleted mantle, which are not re^iily deducedfromparent/daughter ratios in MORB because of the effects of magmatic processing. Nevertheless a few inferences can be made. Average Rb/Sr in MORB, an upper limit for depleted mantle, is so low that d(87Sr/86Sr)/dt < 0.00040/Ga, hence present growth of ^7Sr/86Sr due to Rb decay is insignificant A significant breakthrough camefromTh isotope ratio measurements which allow inference of K (ThAJ) in the depleted mantle. The present KDM IS about 2.5, but Pb isotopes require a time-integrated K of about 3.7. From values of K and Rb/Sr, inferences can dso be made about |I (238u/204p5) in the depleted mantle: the present day ratio is lower than bulk earth |i, which is in turn lower than the p. necessary to produce observed Pb isotope ratios. All these observations are consistent in suggesting that the isotope evolution of the depleted mantle controlled not by in situ radioactive decay but byfluxesfrom other reservoirs. The source of this flux is difficult to establish: neither primitive mantle, sediment, nor continental crust seems to have the requisite chemical and isotopic properties. Mantle plumes are a significant influence on the composition of MORB on regional scales (100-1000 km), and may partly account for differences in isotopic composition between ocean basins. Fluid dynamic models also indicate significant mixing and entrainment of asthenosphere by plumes. The plume flux to the upper mantle may therefore have a significant influence on the upper mantle chemistry.

K. WATANABE and E. IZAWA, Dept. of Mining, Kyushu U n i v e r s i t y , Fukuoka 812, Japan and T. HONDA, Atomic Energy Research Laboratory, Musashi I n s t i t u t e of Technology, Kawasaki 215, Japan Length measurement of f i s s i o n - t r a c k s in common a c c e s s o r y m i n e r a l s ( e . g . a p a t i t e & z i r c o n ) i s an^ e x c e l l e n t t e c h n i q u e s t o e l u c i d a t e t h e thermochronological h i s t o r y of geological bodies. In order t o measure the f u l l track l e n g t h , t h e s e c o n f i n e d t r a c k s which i n t e r s e c t with s u r f a c e spontaneous tracks or cleavage planes may be used. However, in minerals with lower track d e n s i t i e s , confined tracks may be d i f f i c u l t to d e t e c t . To apply the technique to thermal h i s t o r y analysis of minerals from younger i g n e o u s r o c k s ( i . e . P l i o - P l e i s t o c e n e z i r c o n or Miocene a p a t i t e ) , a new technique involving heavy ion i r r a d i a t i o n has been investigated. Z i r c o n and a p a t i t e w i t h s p o n t a n e o u s t r a c k d e n s i t i e s r a n g i n g from 10^ t o 10^ cm"^ w e r e irradiated with 58^110+ ions (80.2MeV) perpendicular to the c r y s t a l surface using a Tandem accelerator (lOMV) and a target of gold f o i l . The fluence of the ion beam on the sample was measured u s i n g a semi-conductor detector, and controlled by varying the scattering angle (9) between the sample and the incident beam. After chemical etching of the sample, p a r a l l e l heavy ion t r a c k s (10 - 20 Um) became v i s i b l e under the microscope. Confined 238u f i s s i o n tracks i n t e r s e c t i n g these heavy ion tracks could then be e a s i l y detected and their lengths measured.

ABSTRACTS

IC0G7

RIDGE PROPAGATION INTO CONTINENTAL CRUST IN THE WOODLARK BASIN: A Sr-Pb-Nd-0 ISOTOPIC STUDY OF RECENT VOLCANIC ROCKS WHITFORD. D.J., DEAN, J.A., ANDREW, A.S., BINNS, R.A., and WHELLER, G.E., CSIRO D i v i s i o n of Exploration G e o s c i e n c e , North Ryde, NSW, 2113, Australia A t t h e w e s t e r n m a r g i n of t h e W o o d l a r k Basin, between Papua New Guinea and the Solomon I s l a n d s , a seafloor spreading ridge is p r o p a gating westward into continental crust. Over a distance of 90 k m , v o l c a n i c rocks range from basalt to rhyolite. In the eastern sector, MORB-like tholeiites, more evolved tholeiites and "andesites" occur in rifts and submarine volcanoes. Further west, peralkaline rhyolites and K- and Mg-rich andesites were dredged from a submarine volcano near the d'Entrecasteaux Islands. Sr, Pb and Nd isotopic trends reflect variations in m a n t l e s o u r c e s . For example, 8 7 s r / 8 6 s r r a t i o s r a n g e f r o m 0 . 7 0 2 8 in M O R B t o 0 . 7 0 3 0 in t r a n s i t i o n a l t h o l e i i t e s t o 0.70350 . 7 0 4 0 in a n d e s i t i c r o c k s w h e r e a s 2 0 6 p b / 2 0 4 p b r a t i o s v a r y f r o m 18.47 in t h o l e i i t e s t o 1 8 . 6 4 in r h y o l i t e s . Sr in t h e v e s i c u l a r submarine rhyolites appears to have re-equilibrated with seawater, although Nd and Pb indicate affinities with the spatially related Mg-rich andesites. Oxygen isotope values from glassy rocks show little variation from MORBs to rhyolites. Samples with values less than 5 permil (V-SMOW) suggest some interaction with seawater. Whether the inferred source modification reflects the transition to sub-continental m a n t l e or an earlier subduction e v e n t r e m a i n s uncertain. T h e r e is little e v i d e n c e for any direct continental crustal input.

THE COMMON LEAD COMPOSITION OF FELDSPARS MEASURED WITH THE SHRIMP ION MICROPROBE WaEDENBECK. Michael, Research School of Earth Sciences, the Australian National University, Canberra, A.C.T., 2601, Australia In order to more accurately characterize the common lead composition in zircons, a procedure using the SHRIMP ion microprobe has been developed for measuring the lead spectrum in oogenetic feldspars. With a 40}im spatial resolution, the ion microprobe is capable of detecting in situ variations in the lead composition of feldspars. Analyses are conducted on polished thin sections cast in epoxy disks. A small plug containing a lead rich (2.0 wt.% Pb02) potassium feldspar standard is included for monitoring machine sensitivity and isotopic fractionation. Measurements are conducted with a 5 n.Amp primary 02" beam at M/AM = 6500. T>'pical sensitivities for lead are 20 counts / sec / ppm. Included in the single collector peak hop sequence are mass stations for both ThO and UO, which are used to detect the presence of thorium and uranium in the volume being analyzed. No evidence has been found for lead hydride interferences on either the ^o^pb or masses. Machine induced mass fractionation favoring the lighter isotopes is <3%o. Based on counting statistics, during a 25 minute analysis the 204pb/206pb^ 207pb/2l56pb and 208pb/206pb ratios in a sample containing 20 ppm total Pb can be determined to a l a precision of 7%o, 5%c and 5%c, respectively. Initial results from Archean feldspars from Western Australian granitoids show that different lead isotopic compositions may exist between optically similar areas. Such variations are not correlated with the presence of Th or U, hence it is unlikely these differences reflect the in situ decay of radioactive elements. In one case the 204pb/206pb of a single grain varied from exterior to interior from 0.0760 to 0.0542, respectively. Such large inhomogeneities must reflect an open system behavior, allowing lead to enter the feldspar from its surroundings. Where heterogeneous feldspar populations have been measured they show good correlation between their lead isotopic ratios, implying a mixing between the original magmatic component and one single contaminant. The non-radiogenic lead in zircons is best characterized as the least radiogenic component found in the feldspars. Continued work will better describe the geologic process affecting feldspar common lead.

111 T h e T e r t i a r y m e t a m o r p h i c e v o l u t i o n of t h e b l u e schist terrane of Sifnos, G r e e c e : a study of s i n g l e mica grains by a r g o n laser probe. WIJBRANS. J.R.. Isotope Geology Group, Department of Earth Sciences, Free University, De Boelelaan 1085, 1081 HV Amsterdam, The Netherlands, SGHLIESTEDT, M., Inititut fiir Mineralogie, Universitat Hannover, D3000 Hannover, F.R. Germany, YORK, D., Geophysics Group, Department of Physics, University of Toronto, Toronto, Canada M5S 1A7 Using the laser heating technique it is possible to obtain detailed age information from young metamorphic micas either by stepwise heating with a broad defocussed laser beam, or by spot fusion of small parts of the crystal with a tightly focussed beam. Comparing the two techniques on different grains from a single micaceous vein shows that the integrated results over the spot fusion grid and the age spectrum are in very good agreement. The range in ages found in the spot fusion experiment is somewhat larger than that found for the step heating experiment which suggests that during stepwise heating some smoothing of the 40Ar* gradients may have occurred. If this is the case, apparent plateaus in mildly disturbed age spectra must be interpreted as younger estimates for the age of the mineral. Ages of micas from the overprinted rocks of central Sifnos cluster tightly around 30 Ma, with some evidence for local overprinting as young as 18.9 ±0.3 Ma. The oldest age found in the blueschist eclogite zone is 45.6 ± 0.6 Ma for a mica from an eclogite from the schists in the far north of the island. The centre of the blueschist zone yielded ages of 41 Ma. Ages as young as 36.1 ± 0.3 Ma occur at the southern margin of the blueschist eclogite zone, and ages in the range 30 to 35 Ma were measured on micas from schist horizons within the adjacent main marble unit. Micas from micaceous layers within the main marble unit yielded ages of 40 Ma. Thus micas from the schists surrounding the main marble unit have remained open to loss of radiogenic argon considerably longer than those from within the main marble unit. Alternatively, the schist micas have lost argon during an event not felt by the micas within the main marble unit.

TECTONIC AND THERMAL HISTORY OF SIFNOS. CYCLADES. GREECE: ARGON AGES AND NUMERICAL MODELS WIJBRANS. J.R.. VAN WEES. J.D.. STEPHENSON. R.A .. a n d C L O E T I N G H . S . A . P . L . . Institute of E a r t h Sciences, Vrije Universiteit. De Boelelaan IO85. 1081 HV Amsterdam, Netherlands The southern margin of the Eurasian Plate was formed during Late Cretaceous to early Tertiary convergence of oceanic lithosphere and microcontinental blocks. T h e C y c l a d i c b l u e s c h i s t s on Sifnos c o n s i s t of continental platform deposits that were metamorphosed at e x t r e m e p r e s s u r e s and moderate temperatures: conditions in the ranges 13-1? kbar and 450-530*C. U p l i f t f o l l o w i n g this m e t a m o r p h i s m w a s r a p i d , stepwise, and took place at least in part in an extensional environment. Two limiting models for uplift are differential uplift of the entire section as a unit and tectonic juxtaposition of a number of slices with similar but distinct thermal and tectonic histories. The geochronological database obtained by laser step heating of selected mica grains from schist horizons showed gradually decreasing ages from 45.6 Ma in the north to 30 Ma in central Sifnos. Greenschist overprinting in central Sifnos is ca 19 My old. but m i n o r , l o c a l , g r e e n s c h i s t overprinting in north Sifnos predates M a . Micas from the geographically intermediate main marble unit yielded 40 Ma ages, distinctly older than ages found for micas in the surrounding schists. In combination, these findings point to the possible existence of two separate blocks with resetting of schist micas in the base of the upper block. Numerical modelling is used to explore the thermal consequences of and to provide constraints on various possible tectonic histories.

112

ICOG 7

ABSTRACTS

CARBON ISOTOPE VARIATION IN A ZONED BULTFONTEIN DIAMOND DETERMINED BY SIMS M. WILDING and B. HARTE. Department of Geology and Geophysics, University of Edinburgh, Edinburgh EH9 3JW, UK, and J.W. HARRIS, Department of Geology, University of Glasgow, Glasgow, G12 8QQ, UK

BERYLLIUM-10; A METHOD FOR TRACING AND DATING ANTARCTIC ICE SHEET FLUCTUATION IN THE PLIOCENE ERA. G.S.WILSON. Antarctic Research Centre,Victoria University of Wellington, P.O.Box 600, Wellington, New Zealand.

The study was carried out upon a polished plate cut parallel to the {001} direction of a diamond from the Bultfontein mine. South Africa. Cathodoluminescence (CL) showed a pattern of cyclic growth involving discontinuities between zones, and with several blue zones of cuboid growth overlain by octahedral zones of green luminescence. The i^C/^^C ratio of the zones was determined on the NERC/Edinburgh University Cameca IMS 4f, ion microprobe using a Cs"*" ion source. The isotope ratio at each point on the diamond was determined by 100 alternate measurements of each isotope peak, and between each point similar measurements were made on a homogenous synthetic diamond of known S^^C. Variation in the analysis at any point was limited to a standard deviation of 0.5®/®® and the results are believed to be accurate to T/®®. The carbon isotopic values of the diamond range from S^^C of -5.8 to -10.9®/®® vs PDB, and show a consistent correlation with particular zones in the CL pattern as follows:

Shallow marine strata of the Antarctic continental margin record, from the character of the sediment deposited, changes in the extent of polar ice. Alternations of till and marine mudstone facies mark glacial advances and retreat phases. This stratigraphic record has been drilled in the McMurdo Sound area in valleys (DVDP-10,11) and fjords (CIROS-2), revealing a varied history of glacial advance and retreat over the last several million years.

CL zone No. of points analysed range A 5 -6.3 to -7.3 B 6 -7.1 to 9.5 C 5 -10.4 to -10.9 D 2 -9.0 to -10.2 E 1 -5.8 F 3 -8.2 to - 8.9 G 1 -6.5 The data indicate changes in diamond precipitation conditions, with either variation in the ^^^C of the source material precipitating diamond, or varying fractionation of the isotopes at the time of precipitation. The fractionation could occur with deposition from a vapour of initial bulk close to the expected mantle value of -5®/®®, with deposition being caused by variation in oxygen fugacity (P. Deines, 1980, Geochim. et Cosmochim. Acta, 44, p.943-961).

UNTANGLING THE SOURCES OF GRANITES BY DATING mHERTTED ZIRCONS WILLIAMS. I.S., COMPSTON, W. Research School of Earth Sciences, Australian National University, Canberra, ACT, 2601 CHAPPELL, B.W., CROOK, K.A.W., Geology Department, Australian National University, Canberra, ACT, 2601 CHEN, Y.D., Macquarie University, Sydney, NSW, 2109 The major and trace element compositions of the mid-Palaeozoic< granites of SE Australia show that they were derived from a variety of sources, some consisting mainly of recycled Supracrustal rocks and others mainly of unweathered Infracmstal rocks. Nd (mean) model ages of those sources are mid to late Proterozoic. However, Nd cannot be used to date individual source components, or to explain why their mean age may show systematic geographic changes. All the granites contain inherited zircon as small, mostly indistinguishable, nucleii within some of the melt-derived zircons. SHRIMP has been used to survey the U-Th-Pb isotopic composition of more than 1000 zircons from 20 granites and their enclaves, and so identify and directly date these remnants of the granites' protoliths. The various granites are 440 to 380 Ma old. The inherited zircon ages group: 650-450 Ma, 1075-800 Ma, and a continuum from 3350 to 1175 Ma. Each age group is represented in most granites, but their absolute and relative abundances differ. The most primitive granites, on the E side of the I-type Bega Batholith, contain rare inherited zircons mostly 650-450 Ma old. The abundance and mean age of the inheritance increases as the granites' chemical and isotopic compositions become more evolved westwards across the batholith. The enclaves contain considerably less inheritance. In the S-type granites further W every zircon core is inherited—melt-derived zircon forms only thin mantles. The S-types have the same inheritance ages as the I-types, except for an extra Archaean age group. Zircons from the c o u n ^ rocks (Ordovician flysch) show identical age groups to the inheritance, but the 650-450 Ma group is relatively less abundant. These data show that major and previously unrecognized igneous activity in SE Australia at 1075-800 and 650-450 Ma generated the sources of the I-type granites. It also contributed most of the detritus to the S-type granites' sources and the early Palaeozoic flysch. It can be demonstrated that the older inheritance in the I-type granites came from minor sediment in their sources.

The cored strata have been dated using various approaches, including foraminifer and diatom biostratigraphy, magnetostratigraphy and K-Ar dating but resolution is poor and there are some serious conflicts. It is proposed to use beryllium-10, which has a half life 1.5 Ma, to provide an independent estimate of age for the cored strata. The cored strata, which are 165 m thick in CIROS-2 (Ferrar Fjord) and 318 m thick in DVDP 11 (adjacent Taylor Valley), comprise alternations of till and marine mudstone. The latter is similar in composition and clay mineralogy to modem sea floor mud being deposited in nearby Granite Harbour. A comparison of l^Be abundances in the mudstone facies of the cores with that of modem sea floor mud should allow the plotting of a decay curve with depth and an age to be estimated for each sample. A comparison of ages inferred in this way from the CIROS-2 and DVDP-11 cores, which are 20 km apart but in separate valleys, will provide a check on the consistency of the results.

TEMPORAL EVOLUTION IN OCEANIC INTRAPLATE VOLCANOES. WOODHEAD. Jon.D, Research School of Earth Sciences, Australian National University, G.P.O. Box 4, Canberra, A.C.T. 2601, Australia. Considerable research effort in recent years has concentrated on the geochemical characterisarion of oceanic intraplate volcanic provinces. However, it is now becoming apparent that significant structure is present within the geochemical data for individual volcanic provinces. For example, in the Hawaiian islands, the combination of geochemical with geochronological and stratigraphic studies has revealed at least three distinct phases of volcanic activity, each with a unique trace element and isotopic signature. Similar behaviour has also been noted from Samoa, Kerguelen, the Marquesas and, most recently, from Pitcaim island. A detailed geochemical investigation of the Marquesan chain in French Polynesia is presented, with both new analyses and a large compilation of literature data. Particular emphasis is placed on the marked isotopic and trace element shifts noted between shield-building and post-shield activity: these are seen to occur systematically at each island in the group. Available data strongly suggest that shield-building lavas are sourced predominantly by melting of the sub-oceanic lithosphere. Comparisons with other island chains reveal that the nature of plume-lithosphere interaction varies considerably between intraplate volcanic provinces. The initial stages of plume-related volcanism are rarely sampled once voluminous shield-building activity has commenced: only in those rare situations where active seamounts have been located at the ends of 'hotspot' trends can this phenomenon be investigated. In Oct-Dec 1989, detailed SEABEAM mapping by the German research vessel 'Sonne' revealed an area of active seamounts in southeast Polynesia, represenring the current expression of the Pitcaim-Gambier mantle plume. Fresh, glassy samples, recovered from these seamounts, provide an opportunity to evaluate the geochemistry of these initial stages of volcanism in the context of a detailed geochemical stratigraphy already established for Pitcaim island (Woodhead & McCulloch, 1989). Preliminary trace element and isotopic data for these samples will be presented.

ABSTRACTS

ICOG 7

113

THE ORIGIN OF GRANITES IN THE NORTHERN GREAT BASIN USA, REVISITED: NEW CONSTRAINTS FROM Sr,Nd AND Pb ISOTOPIC DATA WRIGHT. J. E.. Dept. of Geology, Rice University, Houston, TX, 77251, USA, and WOODEN, J.L., U.S.G.S., 345 Middlefield Rd., Menlo Park, CA, 94305, USA.

Rb AND Sr C O N T E N T S , A N D STRONTIUM ISOTOPIC COMPOSITIONS OF ARC VOLCANICS: MAGMATIC D I F F E R E N T I A T I O N A N D A S S I M I L A T I O N IN R E F I L L E D MAGMA C H A M B E R T.YANAGI, Department of Earth and Planetary Sciences, Kyushu University, Fukuoka 812, Japan

Previous isotopic investigations of granitic plutons in the northern Great Basin (NGB) have not considered that observed isotopic systematics might vary as a function of the tectonic setting and age of the plutonic rocks. Magmatism in the NGB occurred discontinuously over an interval of approximately 130 Ma and can be divided into three distinct tectonic environments: 1) A Middle Jurassic-Early Cretaceous magmatic arc, 2) A Late Cretaceous magmatic arc coeval with crustal thickening during development of the Sevier thrust belt, 3) Cenozoic magmatism synchronous with crustal extension. Another critical point that has been overlooked is that all plutons, at least in the eastem NGB, older that - 7 5 Ma are allocthonous due to shortening associated with the Sevier thrust belt. This implies that the construction and interpretation of isotopic boundaries (i.e. Sr 0.706 and eNd -7 isopleths) without consideration of the age, tectonic setting, and degree of allochthoneity of the plutonic rocks may be in error. Indeed, cmnew Sr, Nd, and Pb isotopic data reveal that isotopic systematics of plutonic rocks in the eastem NGB are strongly correlated with age and tectonic setting. Jurassic-Early Cretaceous plutons have a limited range of Sr, and Nd isotopic values that clutster near bulk earth. All but one of these plutons have eNd values less negative than -7 despite their location well to the east of the ENd -7 line. These plutons appear to have been derived from subcontinental mantle lithosphere with little if any crustal involvement. The chemical and isotopic composition of Late Cretaceous plutons on the other hand indicate that they represent crustal melts produced during crustal thickening associated with the Sevier thrust belt. Cenozoic plutons appear to be mixtures of subcontinental mantle lithosphere and crustal reservoirs. The isotopic systematics of Late Cretaceous and Cenozoic plutons also delineate a previously unrecognized east-west trending ancient crustal boundary between relatively depleted Archean crust to the north, and relatively undepleted Ptoterozoic crust to the south.

In order to d e f i n e t h e m a g m a t i c differe n t i a t i o n , and the c o n t a m i n a t i o n process, major chemical compositions, Rb and Sr c o n t e n t s and s t r o n t i u m isotopic c o m p o s i t i o n s w e r e d e t e r m i n e d on s a m p l e s from lava flows of Quaternary Kinpo volcano in Kyushu, Southwest J a p a n . The r o c k s are all suba l k a l i c . A l l but a few t h o l e i i t i c b a s a l t s are c a l c - a l k a l i c . Sr c o n t e n t ranges from 387 to l l O S p p m , a n d j^^tio from 0.70377 to 0 . 7 0 5 3 7 . M a r k e d linear r e l a t i o n s h i p s are found between ®^Sr/®®Sr ^^^ 1/Sr of the r o c k s . T h e s e linear t r e n d s m o v e s t o w a r d the low ® ^ S r / ® ^ S r r a t i o s i d e w i t h t i m e . Repetitive occurence of basalts and basaltic andesites in the stratigraphic succession and lines of petrographic evidence for magma mixing suggest the o p e r a t i o n s of a p e r i o d i c a l l y r e f i l l e d m a g m a chamber under Kinpo v o l c a n o . The linear relationships b e t w e e n ® ^ S r / ® ® S r and 1/Sr, and also b e t w e e n some m a j o r o x i d e c o m p o n e n t s are interpreted as the m a n i f e s t a t i o n of e r u p t i o n of l a v a s d u r i n g the r e p e a t e d m a g m a mixing. The h i g h ®'^sr/®^Sr r a t i o s are p o s t u r a t e d to b e d u e to the c o n t a m i n a t i o n b y melts derived from the granulitic basic lower crust d u r i n g the u p w a r d m i g r a t i o n of the c h a m b e r to a t t a i n the heat balance. Based on a s s e m b l e d facts from M y o k o , S a k u r a jima and K i n p o v o l c a n o e s , the c o n c e p t of r e f i l l e d m a g m a c h a m b e r is d e v e l o p e d and t h e e v o l u t i o n of it is p r e s e n t e d .

HELIUM ISOTOPE GEOCHEMISTRY IN NATURAL GAS XU Yongchang, Laboratory of Biogeochemistry and Gasgeochemistry; Lanzhou Institute of Geology, Acaderaia Sinica, Lanzhou 730000, China We have systematically studied He in natural gas from oil/gas-bearing basins in China. The results show that ^He/ He in natural gas has a apparent relativity with "^Âr/^Âr. Because 'He/'^He is higher than 1.4x 10" , '*°Ar/2®Ar presents a relative high ratio, which may be explained that Ar and He are related to mantle origin. In more cases, however, Ar and He present negetive relationship. This may imply that the origin of Ar and He is related to the crust. Like ''Âr, "He in natural gas also has a similar chronological accumulation effect, which is coincident with the linear relation between 'He/'^He and the storage age of natural gas. A equation, ^Ci=-35.4-0.916('He/'^He)x 10"'®(r=0.7), has been obtained by linear regression analyses. It shows that 'He/'^He in natural gas has a certain relativity with maturity of CHu. A series of COa and He gas pools, with C02>90% and He from 0.25% to 1.34% respectively, are distributed along the Tanlu large zone which is located in the east of China. in gas pools is in the range of -3.4%o-6.1%o, 'He/'^He=4.9 — 7.2x10"^ On the basis of the geological background, it is reasonable to imagine that CO2 and He originate from mantle degassing, then migrate along the Tanlu Fault zone and reserve in the sedimentary starta.

GE0CHR0N0L06Y OF STANNIFEROUS PLACERS IN NORTHEASTERN T A S h ^ I A , AUSTRALIA YIM. UrrSS U.-S. . Department o-f Geography & Geology, University o-f Hong Kong, Hong Kong (address until end o-f December - Centre -for Australian Regol i th Studies, Australian National University, 6P0 Box 4, Canberra, ACT 2601, Australia) The geochr onol ogy o-f stann i-f erous placers in northeastern Tasmania was studied using a range o-f dating methods. Out o-f these, -fission track and electron spin resonance dating o-f alluvial zircons deriving -from granitic and basaltic source rocks gave results consistent those o-f Rb-Sr and K-Ar dating, Sedimentologic, pedologic, palaeo-oceanographic and palaeo-botanic evidence also examined to provide strat igraph i c control are -found to be in agreement with the conclusions drawn. Placers are -found to be the product o-f 'long' periods o-f geologic time by episodic recycling dating back at least to the Permain. Deep leads were stratigraphical ly con-fined between the Middle Eocene <ca. 47Ma) and Middle Miocene <ca. 16Ma) episodes o-f basaltic volcanic activity. Their burial have been caused by the Oligocene low sea-level stand when erosion and sedimentation rates increased over the highlands and piedmont zones respectively. In addition to deep leads, comparatively shallow leads also occur either -filling bedrock channels or burying erosional sur-faces cut across older alluvial sediments. These leads are lower in economic value because o-f their smaller dimensions, and, they may have been continuously reworked to the present day. Quaternary placers -formed by the thawing o-f upland snow-fields during glacial periods are present as -fluvial terrace deposits along parts o-f the Ringarooma valley. The importance o-f geochronology and strat i graph i c control in providing understanding on the genesis o-f stann i-ferous placers is emphasised.

ICOG 7

114 STABLE ISOTOPIC STUDIES OF THE EAST TAIWAN OPHIOLITE Tzen-Fu YUI, Institute of Earth Sciences, Academia Sinica, Taipei, Taiwan, R.O.C. The East Taiwan Ophiolite (ETO) has been suggested to have been formed at an active ridge-ridge transform fault in a marginal basin (Suppe et al., 1981). Its peculiar tectonic setting merits detailed geochemical studies in order to enhance our knowledge about the geochemical changes which occurred during a hydrothermal alteration of the oceanic crust. Stable isotopic studies of the ETO show a slightly decreasing trend in oxygen-isotope compositions from basalt through plagiogranite/gabbro to ultramafic rocks. This is a common variation trend as documented in many other studies of ophiolites and oceanic crusts. The amphiboles from the ETO gabbros have 6D values ranging from -9 to -19%^ Carbonisotope compositions of calcite veins from the ETO gabbros and basalts range from -3.1 to values as low as -54.5%.. These hydrogen- and the lower end of the carbon-isotope compositions are the highest and the lowest values, respectively, ever reported for samples with similar occurrences. These extreme values could only be accounted for by the ETO*s peculiar tectonic setting. The hydrogen-isotope compositions of lizardites/ chrysotiles in the ETO serpentinites range from -57 to -91%- These values are not compatible with the suggestion that, on the basis of the available isotopic data as well as the field occurrence, most of the lizardites/chrysotiles should have formed in oceanic environments. Post-formation hydrogenisotope exhange between lizardites/chrysotile and the local meteoric waters during/after the on-land emplacement of serpentinites is postulated as a probable explanation. The antiogorites, however, seem to have retained their indigenous 6D values.

A COMPARISON OF Pb-Sr-Nd ISOTOPES IN OLD AND YOUNG CONTINENTAL LITHOSPHERIC MANTLE. ZARTMAN. Robert E., and FUTA, Kiyoto, U.S. Geological Survey, Mail stop 963, Denver Federal Center, Denver, Colorado 80225 A keel of lithospheric mantle underlies most of the continental crust and extends downward to a transition zone into the asthenosphere. The chemical and isotopic properties of this reservoir are known mainly from the basalts and xenoliths that sample it. The distinctive trace element and isotope composition of the lithosphere provides important clues to the Earth's differentiation history, including evidence of interaction with the continental crust, oceanic crust, and asthenosphere. Old lithospheric mantle exhibits Pb, Sr, and Nd isotopic patterns, which confirm the existence of longlived, chemically complex systems, but tend to mask its initial character. The mantle under newly-created crust, on the other hand, is typified by less isotopic variability, which emphasizes the heterogeneity in the materials from which the lithosphere is assembled. A comparison of the Pb, Sr, and Nd isotopes in alkaline to sub-alkaline basalt derived from Late Archean (northeastern China) and Phanerozoic (Patagonian plateau, Argentina) sub-crustal mantle is useful in identifying "end-member" components of the lithosphere. One component, having an isotopic composition between PREMA and HIMU, either continues to evolve virtually unchanged after incorporation into the lithosphere, or is, itself, a relatively new addition to even old lithosphere. Another component, beginning with a BSE (EMI?) isotopic composition, undergoes significant reduction in U/Pb and Sm/Nd upon incorporation into the lithosphere, and, with time, shows increasingly retarded evolution of tt and negative G ^ d - v a l u e s . Somewhat elevated initial values of P, 7 , and Gs,, and reduced value of in both PREMA-HIMU and BSE components compared to MORB, a trait shared with OIB, may identify an additional recycled continental crust component.

ABSTRACTS

5I8O AND 513c ISOTOPIC VARIATION OF HYDROTHERMAL CARBONATES FROM THE ROSEBERY, HERCULES AND SOUTH HERCULES DEPOSITS, WESTERN TASMANIA: INTERPLAY OF CAMBRIAN VS. DEVONIAN SYSTEMS KPITNZAW and R. R. LARGE, Centre for Ore Deposit and Exploration Studies, University of Tasmania, Hobart, Tasmania. The Rosebery, Hercules and the South Hercules deposits are volcanogenic, Cambrian style massive sulphide mineralisation in western Tasmania which have been affected by the Devonian Tabberabberan Orogeny and granite intrusion. Hydrothermal carbonates are found with both the Cambrian exhalative mineralisation and the Devonian granite-related tin mineralisation (e.g. Cleveland deposit, Collins, 1981; Renison Bell deposit, Patterson et al. 1981). The carbonates display wide textural variation including bladed, dissemination, spheroids and recrystallised massive carbonates which are sometimes cut by late fracture filling quartz-carbonate±fIuorite veins. The following S^^O and S^^C values of the carbonates have been recorded for the major ore deposits in the region;

Cambrian Rosebery Hercules South Hercules Pgypnian Cleveland Renison Bell

5I8O (SMOW)

513c (PDB)

+9.9 - +13.9 %o +10.8 - +14.3 %o +9.8 - +16.7 %o

-5.0 - -0.7 %o (n=21) -3.3 --3.3%o(n=7) -3.5 - +0.6 %c (n=10)

+ 1 1 . 2 - +16.71

-7.0- -1.8%c(n=8) (n=3) - 6 . 0 - -4A%c

+12.7 • +15.31

Although 5180 isotopic values give comparable ranges for the Cambrian and the Devonian hydrothermal carbonates, the Devonian hydrothermal carbonates have an apparent lighter 51 ^C isotopic composition than those of the Cambrian hydrothermal carbonates. The 513c isotopic ratios of the Cambrian hydrothermal carbonates do not appear to have been significantly altered by exchange with later Devonian fluids. On this basis, the carbon isotopic ratios of the carbonates can be used as a guide to distinguish Cambrian vs. Devonian st>'les of mineralisation in western Tasmania.

U-Pb DATING OF VERY Y O U N G LEUCOGRANITES ( N W HIMALAYAS) USING A N I O N MICROPROBE ZEITLER. Peter K., Dept. of Geological Sciences, Lehigh University, Bethlehem, PA, 1 8 0 1 5 , USA, and CHAMBERLAIN, C. Page, Dept. of Earth Sciences, Dartmouth College, Hanover, NH 0 3 7 5 5 , USA In the Himalaya of northwestern Pakistan, small discordant bodies of leucogranite intrude basement gneisses of the N a n g a Parbat-Haramosh Massif (NPHM). As the first step in assessing the tectonic significance of these granites, v^e hove used the SHRIMP ion microprobe to obtain U-Pb dotes on zircon. Zircons of igneous appearance are scarce in the NPHM leucogronites, and with one exception ore associated with xenocrvstic components, which occur as discrete grains and as cores rimmed by young zircon. Zircons of igneous appearance are high in U (5,000 to 4 0 , 0 0 0 ppm , low in T h / U (0.01 to 0.05), and distinctly blue in co or. The xenocrystic components give ages of ~ 1 8 5 0 Mo, (known to be the age of the Iskere Gneiss, the main basement unit of the NPHM). In sharp contrast, zircons of igneous appearance from three different samples yield a concordant spread of very young ages: 2.2 to 4.3 Ma; 5 to 8 M o ; and 7 to 12 Ma. W e feel that our data may record a true range in crystallization ages because the observed spread in ages is not readily attributable to typical phenomena such OS inheritance or Pb loss. Further, the spatial and temporal coincidence of the NPHM leucogronites with on episode of rapid unroofing in the NPHM leads us to suggest that rapid decompression should be considered as on alternative mechanism for leucogranite genesis.

ABSTRACTS

ICOG 7

MESOZOIG TECTONIC ISOTOPE GEOLOGY IN EASTERN ZHANG Ligang,

Yichang Institute of Geology

115 GHm* and

Mineral Resources,CAGS,Yichang,Hubei,445003,China ^iesozoic granitoids and felsic volcanics,many with associated ore deposits,are widespread in eastern China.On the basis of Pb isotope data for 215 feldspars from 85 Mesozoic granitoid bodies and lots of ore minerals,and 1250 Sr,0,Kd analyses for 98 Mesozoic granitic intrusive bodies and volcanic rocks from eastern China,in which the ranges of isotope ratio values for these measiirements are as follows: 206/204=15.709 to 19.025,207/204=14.395 to 15.851, 208/204=55.801 to 59.252; 187/86=0.7042 to 0.7850; 1143/144=0.51165 to 0.51226, and S 0 values range from 6 . 5 to 1 4 . 6 % . .An important feature found here is that the values of isotope ratios,especially P b , are subject to tectonic backgrounds, and have very little relations to the age of rocks.As so,we think that three types of crustal basement rocks exist under eastern China, and we can further subdivide the area of these three types of rocks into ten tectoic isotopic provinces.Each isotopic province has distinguishable isotope compositions and isotopic interrelations.Such as in north China tectonic isotopic province have the most nonradiogenic Pb and the lowest 187/86, S O and ^ N d ( T ) v a l u e s , but in Nanling tectonic isotopic province have the most radiogenic Pb and the highest 187/86, i 0 values and a rather low ^ N d ( T ) value among the ten tectonic isotoDic province.Pb,Sr,0,Nd isotope compositions contrain both the developing history and the metallogenic regularities of tectonic terranes. The isogram of Pb isotope ratios on which the boundaries and forms of different isotopic province are sketched out were plotted for the Mesozoic igneous rocks in eastern China on the basis of data described above. *This project is supporting b y the National Natural Science Foundation of China

The Permian-Triassic Boundary Age In China Zhang Zichao Yichang Inst, of Geology and Mineral Resources, G A G S , P.O.Box 502, Yichang, 445003f China Defining the boundary age between Permian and Triassic is an important problem in chronostratigraphy and time scale study. Harland(1982), Odin(l982) and Snelling(l985) assigned 248 and 245 Ma as this boundary age respectively. Lee Zichang(I988) proposed 235+5 Ma as the boundary age between Permian and Triassic in China. In order to confirm this result the author recently worked out three U-Pb age data on zircons from bentonite beds at or very close to the boundary level in different locations. Zircons from an Ophiceras sp.-bearing bentonite bed at the base of Triassic Daye Group in west Hubei gives a lower intersection U-Pb age of 250+4 M a . The fossils can be correlated with those of Griesbachian Stage in Europe. The second zircon sample from a bentonite bed at the top of Permian Dalong F m . in the same area fits a lower intersection age of 237 +8 M a . The third one was taken from a bentonite bed at the top of Late Permian in east Shichuan. The 207pb-206pi3 age measured by direct evaporation on the filament of thermal ion source in mass spectrometer is 239+9 M a . The.se new data support the conclusion drawn by Lee Zichang in *The Geological Time Scale of China'. Although the lower intersection age on the concordia curve by zircons may be a bit younger owing to difusion Pb loss, all these ages are in good agreement with the Rb-Sr isochron and K-Ar ages on volcanic rocks obtained by Lee Zichang and indicate that the P-T boundary age would not be as old as 245 M a . The difference between Chinese data and those from Europe and Australia may be owing to the stratigraphical correlation between different continents and uncertainties in measured data. Anyway more studies are necessary to further confirm the conclusion has drawn.

PROVENANCE OF SEDIMENTS FROM THE AMADEUS BASIN, CENTRAL AUSTRALIA: Sm-Nd CONSTRAINTS J.X. ZHAO & M.T. McCULLOCH, Research School of Earth Sciences, Australian National University, Canberra, Australia The Amadeus Basin is a late Proterozoic to Phanerozoic ensialic depositional basin, located between two Proterozoic basement terrains of different crustal formation ages, the older Arunta Block (2.0 - 2.2 Ga) to the north and the younger Musgrave Block (1.7 -1.9 Ga) to the south. Sediments analysed have a wide range of fSm/Nd from -0.31 to -0.61, irrespective of grainsize and rock types. Their Zfi^iO) values range from -21.2 to -14.2 and their eisid(deposition) values, mostly from -8.3 to -11.3. The calculated T^j^ values for samplesfromthe same stratigraphic unit are well correlated with fSm/Nd^ samples of different stratigraphic units forming sub-parallel arrays. These phenomena are inte^reted in terms of REEfractionationand preferential sorting of REE-rich phases during sedimentary recycling. This process leads to either an increase or a decrease in fSm/Nd Qf ^ sediment relative to its provenance. Therefore, a method is developed, which can be used for estimating average provenance age and average REEfractionationage of a stratigraphic unit, assuming the average fSm/Fid value of the source is known or can be estimated. The average provenance age for each stratigraphic unit of the Amadeus Basin has been calculated using this method. The Nd isotopic data indicate the Amadeus Basin sediments are dominated by a two-source mixing provenance, which is probably related to the older Arunta ( fSm/Nd ^ .0.42 and £Nd(0) - -20.4) and younger Musgrave Blocks ( fSm/Nd ^ .0.40 and eNd(O) - -14.7) and/or their equivalents. Almost aU the analyses fall into the region defined by the two basement block trajectories. Stratigraphically younger sediments plot closer to evolutionary trajectory of the Musgrave Block. This result, together with the average provenance ages estimated above, indicates the proportion of materialfromthe Musgrave Block increases as sediments become stratigraphically younger. The Nd isotopic data preclude a substantial involvement of Archaean sources. Nd isotopic features of the Amadeus Basin sediments suggest that: (1). REEfractionationduring sedimentary recycling can in some cases be an important factor and ne^s to be considered; (2). the provenance of Amadeus Basin sediments is controlled by its local source regions, which mitigates against the use of Nd constraints from platformal clastic sediments for generalised models of cmstal evolution.

SULFUR ISOTOPE FRACTIONATION BETWEEN SULFATE AND SULFIDE IN HYDROTHERMAL ORE DEPOSITS: DIAEQUILIBRIUM VS. EQUILIBRIUM PROCESS ZHENG,Yong-Fei, Geochemisches Institut, Universitat Gottingen, Goldschmidtstr. 1, D-3400 Gottingen, FRG 34 Correlation arrays of sulfur isotope data on 8 S vs. A S diagrams are presented for coexisting sulfate and sulfide pairs from some hydrothermal^ore deposits. Taking into account dissolved H^S and SO, in hydrothermal fluids during precipitation of both sulfate and sulfide, a 4-phase closed system is suggested for describing the conservation of mass among all sulfurbearing phases on the S - A diagrams. The covariation in the S S values of both sulfate and sulfide is ascribed to isotopic exchange between oxidized and reduced sulfur species during precipitation of the minerals. The isotopic exchange could be a thermodynamic process due to simple cooling of high temperature fluids, which leads to an equilibrium fractionation, or a kinetic process due to mixing of two solutions, ^^ich results in a disequilibrium fractionation. The S S value of total sulfur in the fluids could change due to the precipitation of«the minerals, or due to the escape of H^S and/or SO,. Sulfur isotope data for anhydrite and pyrite pairs from the Luohe iron deposit in the Yangtze River Valley and for barite and sulfide pairs from the Bad Grund Pb-Zn deposit in the Harz not only show good correlation arrays on the S - A d i a g r a m s but also yield unreasonably high isotope temperatures which indicate a disequilibrium fractionation between coexisting sulfate and sulfide pairs. A mixing of two solutions is suggested as the cause for the mineral precipitation. The oxidation of aqueous H2S during the hydrothermal mixing is proposed responsible for the decrms^ in,the S S values of sulfate and the increase in the S S values of sulfide.

116

ICOG 7

ABSTRACTS

CONTINENTAL GROWTH AND SEDIMENT EVOLUTION: ND ISOTOPIC PROFILE OF THE SINIAN FORMATION, NORTHERN CHINA. ZHOU, X., Institute of Geology, Academia Sinica, Beijing, PRC, and GOLDSTEIN, S. L., Max-Planck-Institut fur Chemie, Postfach 3060, 6500 Mainz, FRG. Nd isotopic compositions of clastic sediments have yielded much information on the growth history of the continental crust, because they reflect average crustal residence ages (Tcr). In most studies, samples have been taken from localities over large areas, allowing only the gross features of continental growth history to be inferred; whether the Nd isotopic evolution of the continents has been nearly gradual through time, or whether there have been episodic, is hidden by the scatter of the global data. In order to address these problems we have analyzed samples from the Sinian Formation, a well-documented, nearly undeformed sedimentary section in northern China recording deposition on a passive margin between ~l.85-0.55 Ga BP, with a thickness of >10 km. It lies adjacent to 2.5-2.6 Ga old craton regions which experienced extensive tectonic/magmatic events at 1.8 Ga. The basal sediments have end (L8 Ga)= -4 and Tcr ages of 2.6 Ga, consistent with derivation from the regional Archean basement. Major input of juvenile material occurs in 1.7 Ga sediments, which is clearly indicated by higher eNd(i) of -0. The data do not show gradual changes with time, rather eNd(i) values oscillate wildly. At 1.8, 1.5, and 0.9 Ga BP the sediments are dominated by Archean craton, at other times the sediments appear to be variable mixtures of 2.6 and 1.8 Ga old material. There is a 300 Ma hiatus prior to the Cambrian; the Cambrian sample is the only one with Nd isotopic compositions typical of Phanerozoic sediments and having a model age less then 2 Ga (Tcr = 1.8 Ga). Post-Archean crustal growth in Sinian source areas occurred primarily in a single major event at -1.8 Ga. Other ^owth events do not stand out-a small amount of juvenile addition is compatible with the data at -1.2 Ga, but it cannot be distinguished from an extra large contribution of 1.8 Ga old material. Thus it appears that crustal residence ages of the sediments are governed by addition of new continent during a few major crust building episodes and the homogenizing effects of sedimentaiy recycling.

THE ORIGIN, EVOLUTION, AND INTERACTION OF CHEMICALLY DISTINCT MANTLE COMPONENTS A. ZINDLER AND K.-H. PARK, Lamont-Doherty Geological Observatory of Columbia University. Palisades, N.Y. 10964 U.S.A. Isotopic studies of mantle-derived volcanic materials document the existence of several distinct chemical clans in the mantle: DMM, the depleted material from which MORBs are produced; HIMU, a component enriched in U relative to Pb with a mean age of ~2 Ga; and two types of enriched material (EM I and EM II) which ha e affinities with various types of continental crust and lithosphere. The systc'Tiatic geographic distributions and associations of HIMU and the enriched materials, together with their chemical characteristics, suggest origins via various types of recycling, including subduction of oceanic and continental crust, and delamination of subcontinental lithosphere (e.g., McKenzie and O'Nions, 1983). The aging of these materials, necessary to produce their isotopic characteristics, may occur within the continental crust or lithosphere, or within a mid- or lower mantle thermal boundary layers. South Pacific hotspot emanations are dominated by mixtures of the HIMU and EM II components, and have trace element characteristics which are complementary to island arc basalts. These isotopic and chemical characteristics may result from the subduction of oceanic crust and a portion of the associated sediment cover, which have been modified by processing in the sub-arc environment. In contrast. South Atlantic and Indian Ocean hotspots often display EM I characteristics which can result from detachment of the subcontinental lithosphere during continental breakup. Individual hotspots and groups of hotspots are often characterized by distinctive signatures, resulting from mixtures amongst the various enriched components, which can persist for more than 10® years (as evidenced by isotopic similarities between ancient seamounts of the western Pacific and modem volcanism in the south Pacific; Staudigel et al., 1990). Do the source regions themselves represent the particular mixtures, or are the processes of di^irism and volcanism at individual localities capable of generating reproducible mixtures of physically distinct components? Evidence for the latter hypothesis derives from the observation that the chemical and isotopic character and degree of variability of an oceanic volcanic center depends strongly on the age of the plate Uirough which it is erupted. This suggests that the thermal structure of the asthenosphere and lithosphere, which evolves with the aging of the overiying plate, may control the degree to which plume-derived components mix with ambient upper mantle material Gargely DMM) during diapiric ascent

Sm-Nd, U-Pb ZIRCON (SHRIMP) AND Rb-Sr INVESTIGATIONS OF THE AULIAN GNEISS COMPLEX, AND A PROBLEM OF THE BASEMENT TO THE MIDDLE CISDNTEPROVIA GNIEISS-GREENSTON^ TERRAEvI, UKRAINIAN SHIELD, USSR D.Z. ZHURAVLEV. A.V. SAMSONOV, I.S. PUCHTEL, A.K. SIMON and I.V. CHERNYSHEV, IGEM of the USS Acad. Sci., Staromonetny 35, Moscow 109017, USSR New geochronological data shows that there are three stages in the geological history^of the Aulian Gneiss Complex (AGC). 1 (?) Early Archaean stage - formation of tholeiitic-komatiitic volcanic protoliths now represented by the mafic bodies in the AGC, based on a Sm-Nd whole rock isochron age of 3920±14 Ma obtained on these mafic rocks. 2 Late Archaean magmatic stage (3200-3120 Ma), dated by SHRIMP U-Pb zircon and Sm-Nd whole rock methods. This stage involved the emplacement of the igneous protoliths of tonalitic and trondhjemitic gneisses, and tholeiitic metagabbro. 3 Late Archaean metamorphic stage (3090-3050 Ma), dated by U-Pb zircon and Sm-Nd, Rb-Sr whole rock methods, and included migmatisation and emplacement of trondhjemites. Geochronological data shows that the lower metamorphic grade volcanic-sedimentary sequence and associated tonalite-trondhjemite plutons of the Middle Cisdnieprovia granite greenstone complex (GGC) formed between 3200 and 3000 Ma. On the basis of the obtained isotopic data, only the metavolcanic mafic rocks of the AGC (stage 1, above), may be considered as possible basement for the late Archaean GGC greenstone belts; other AGC lithologies (stages 2 and 3) are the same age as the rocks of the GGC. On this basis, the lower metamorphic grade GGC and the higher ^ade AGC may be considered as supra- and infracomplexes of a single tectonic strucuire. Ininal Nd and Sr isotope values of felsic rocks from the GGC and AGC suggest mantle or crust-basalt sources for their parent melts.

SIMS MEASUREMENTS OF 230xh/232Th RATIOS IN ROCKS A. ZINDLER. J.L. RUBENSTONE, J. ENGLAND, B. BOURDON, L. REISBERG, B. R\MELIN, V. SALTERS, S. WEAVER, W.-X. LI, AND F. MARCANTONIO, Lamont-Doherty Geological Observatory of Columbia University, Palisades, N.Y. 10964 U.S.A.^ Investigations of 238u.230'ph systematics in rocks and minerals have been limited by the large sample requirements (~lîg Th) and modest levels of precision (5-8% 2a) obtainable via a-spectroscopy. The recent application of TIMS techniques to 230x^/232'^^ analysis have resulted in a reduction in sample requirements to ~250ng and an improvement in precision to 0.5-1% (Goldstein et al., 1990). However, the extraction of 250ng of Th from MORB, for example, still requires on the order of 1 g of glass. Even where preparation of such large samples is possible, contamination (e.g., by manganese coatings on basalt glasses) is a constant worry, and the hand-picking of wains an arduous task. As many of the most important applications of 238u.230Yh systematics in hardrock geochemistry necessitate the analysis of U- and Th-depleted materials, we have concentrated our efforts on the development of a procedure that significantly reduces the samplerequirementsof current a-spectroscopv and TIMS techniques. We have developed a method for analyzing ^^Q^ U, ~i0ng of total Th by using the Lamont Isolab 54 in a SIMS mode. A sample of Th, chemically purified using the usual techniques, is loaded into a graphite matrix on a zone-refined Re filament The sample is then sputtered using a 5kV, -lâmp Ar+ primary ion beam. Molecular contaminants near mass 230 are reduced to acceptable levels by a combination of heating the sample and energy filtering at the +25V level. The ratio is then measured by simultaneous collection of 230xh on a Daly detector and ^^^Th in a Faraday cup. Looking at a 60V energy window, the abundance sensitivity of the machine is approximately 1x10"'' at mass 230, with respect to 232. Results for a series of 9 standards (Lamont ThI; aliquots ranging from 5ng to lOOng total Th) analyzed by this method yield an average ^^h/^^^-j^ ^^QJ^ ratio of (6.435±0.040)xl0-6 (uncertainty is at the 2a level). This compares to (6.463±0.034)xl0-^ for TIMS analyses of 250ng aliquots (made using the Isolab), and to (6.380±0.160)xl0-^ measured by a-spectroscopy (9-day count of 20pig of Th). As an independent confirmation of the ^^^/^^^Th measured by this new technique, we have analyzed a non-hydrated rhyolite glass that is thought to be completely free of secondary alteration, and is old enough for the U-series isotopes to have reached secular equilibrium. Three analyses of -10 ng aliquots of Th from the rhyolite yielded an average ^^^/^^^Th atom ratio of (7.680±0.050)xl0-^. Together with the U and Th concentration determinations, these results lie on the "equiline" in a plot of (^^^Th/^^^Th) vs. (^^Û/^^^Th), and confirm the accuracy of the Th isotopic determination. Further results for rock samples will be discussed at the meeting.

ABSTRACTS

117

ICOG 7

Abbate E. Abouchami W. Adams CJ. Aharon P. Ahmad R. Ainsworth P. Albarede F. Alexander C.M.O'D. Alibert C. Allfegre CJ. Amelin J.V. Anderson J.L. Andre L.G.R. Andrew A.S. Anthony E.Y. AokiY. Arculus RJ. Arehart G. AritaY. Armstrong R.A. Armstrong R.L. Amdt N.T. Arnold J.R. Arnold M.G. Amould M. Atkinson B. Atkinson S.S. Ayliffe L.K. Azmi RJ. Azuma S.-L Baadsgaard H. BaiG. Baker J. Baker K. Bakun-Czubarow N. Baldwin S.L. Balestrieri M.L. Ball T.T. Ballentine C.J. Ballini A. Bannon M. Barberi F. Barbey P. Barnes C. Bamett D. Barr J.M. Barreiro B. Barton E. Basel M.A.S. Baskakov A.V. Beakhouse G. Beard B.L. Becker T.A. Beets C.J.

AUTHOR INDEX

1 1, 11 1 95 1 43 1, 11, 18, 26, 39 2.2 2,4 40 2 68 3 3, 99, 111 3 37 55 3 107 4 4 4, 46, 58, 100 71 4 5 42 5 5,5 91 6 6, 16, 29 6 10 104 14 6, 64 1 7 7 57 103 57 1, 11 7 7 82 7, 38 8 8 40 27 8 82 8

Begemann F. Belshaw N. Bennett V.C. Bergman S.C. BeythM. Bhattacharya S.K. Bibikova E. V. Bickel R.A. Bickle M.J. Bigazzi G. Bikerman M. Binns R.A. Birck J.-L. Bird J.R. Bird M.I. Black L.P. Bloom A.L. Bloom M.S. Bluck B.J. Boher M. Bohrson W.A. Bonnot-Courtois C. Boreham C.J. Borg S.G. Boriani A. Bottrell S.H. Bouquillon A. Bourdon B. Bowman J.R. Bowring S.A. Bowtell S.A. Boyce A.J. Boyd S.R. Brannon J. Bridgwater D. Bristow J. Brown L. Brown R.W. Brown W.L. Brueckner H. Bryant E.A. Buhl D. Bultitude R.J. Burgess R. Burghele A. Burton K. Burton K.W. Caen-Vachette M. Campbell I.H. Canfield D. CarlC. Carlson R.W. Carr G.R. Carr P.F.

90 9 9 9, 73 94 9,88 10, 87 10 10, 17 1 10 111 84 11 11 11 95 61 84 1,11 12 40 97 12 79 12 35 12, 31, 85, 116 13, 80 13 13 32 67 13, 79 90 30 92 14, 37, 38, 67 26 8, 14 80 14 39 32 14 9 15 30 15, 47 12 110 15, 90, 92 15, 25 16

118

16 Carroll J. Cartwright I. 16 Cavell P. A. 16 Cerling T.E. 80 17, 63, 81 Chambaudet A. Chamberlain C.P. 93,114 17 Champion D.C. Chapin C.E. 65 Chapman H. 10, 17 17,22, 112 Chappell B.W. Chappell J. 95 Chaussidon M.F.R. 18 Chauvel C. 18, 39 Chen C. 19 Chen C.-H. 18 Chen J-F. 35 Chen J.H. 18, 95 Chen Y. 59 Chen Y.D. 73, 112 Chen Y.-W. 19 72, 116 Chemyshev I. Chesley J.T. 19 5, 5, 11, 105 Chivas A.R. Choudhary A.K. 19 Christiansen E.H. 55 84 Christie A.B. 41 Christie D. 78 Civetta L. Claesson S. 31 Clague D.A. 48 Claou6-Long J.C. 19, 20, 78 Clauer N. 40 Clayton R.N. 25 20, 110 Cliff R.A. Cloetingh S.A.P.L. 111 Cohen A.S. 9,27 Coleman M.L. 20 Coleman R. 52 Collerson K.D. 20 Comin-Chiaramonti P. 78 Compston W. 4, 13, 16, 21, 21, 29, 30, 33,72, 78, 112 Conty C. 66 Cook S.J. 13 Cooper J. A. 21 Cordani U.G. 21 Corfu F. 21 22 Corlett G. Coulon Ch. 107 22 Coyle D.A. Creaser R.A. 22 Crick I.H. 71 Crook K.A.W. 112 Crowe D.E. 22, 104 Crozaz G. 62 Cumming G.L. 23, 83 Cuvellier H. 51

ICOG 7

D'Amore F. DaiT. Daly J.S. Darbyshire D.P.F. Darling W.G. Dautel D. Davidson G.J. Davidson J. Davies G.F. Davies G.R. Davis A.M. Davis D.W. Davis G.A. De Chambost E. De Laeter J.R. De Wit M.J. Dean J. A. Decker J. Deino A.L. Del Moro A. Delor C.P. Deloule E. Demaiffe D. Demarchi G. DePaolo D.J. Des Marais D.J. Deutsch A. Dever L. Devey C.W. Dia A.N.M. Dickin A.P. Ding T. Dobretsov N. Dodson M.H, Doe B.R. Dokuchaeva V.S. Donnelly T.H. Doulgeris A. Douthitt C.B. Drach D.V. Drake R.E. Dudas F.6.L. Duddy I.R. Duke M.J.M. Duncan R.A. Dunlop A.C. Dymoke P.L. Eadington P.J. Edmond J.M. Eggenkamp H.G.M. Ekwueme B.N. Elburg M.A. Eldridge C.S. Elenheimer D. EUam R. Elliott T.E. England J.

ABSTRACTS

72 23 23, 23 24 24 39 24 24 25 25 25 53 68 66 26, 69 4, 53 15, 25, 111 73 26, 55 79 11 26 26 78 12, 60, 95 96 89 27 66 27 27 27 28 28 28 102 71 48 28 89 26, 29 84, 85 29 29 29, 44 3 30 6 76 30 30 104 30 104 45 64 31, 82, 85, 87, 116

ABSTRACTS

Engstrom E.U. Epstein S. ErcanT. Erwin M.E. Esat T.M. Esser B.K. Eusden J.D. Evensen N.M. Ewers G.R. Exley R.A. FaUick A.E. Fancelli R. Fang Z. Fanning M. Farmer G.L. Fedotov S.A. Fegley B. Ferrara G. Fischer H. Fisher R.S. Fitton J.G. Fitz Gerald J.D. Fitzgerald P.G. Fleck R.J. Fletcher I.R. Flory F. Foden J.D. Foland K.A. Fontes J.Ch. Foster D.A. Fowler M.E. France-Lanord C. Freedman P.A. Frost B.R. Frost C.D. Froude D.O. Fryer B. Fryxell J.E. Fugi M.Y. Fujibayashi N. Fukui S. Funaki M. Futa K. Futagami T. Fyfe W.S. Gale A.S. Galer S.J.G. Gallagher K. Gamo T. Gans P.B. Gao Z. Garcia M. Ge N. Gebauer D. Gemmell J.B. Gerlach D. Gibson E.K.

ICOG 7

119

31 96 72 31 31 32 7 71 32 32 32 72 33 33 7, 108 55 33 33, 34 74 31 42 64, 65 34, 96 57 34, 69 84 34, 96, 103 3, 35 27, 35 35 43 35 32, 75 36 36 62 54 34 36 36 36 99 114 37 11 63 37 14 50 37, 67, 80 37 62 59 38 38 71 94

Giles M. Giletti B.J. Gillot P.-Y. Gleadow A.J.W. Gluyas J. GoldhaberM. Golding S. Goldstein S.J. Goldstein S.L. Goodfellow W.D. Gopalan K. Gopel C. Gorokhov I.M. Goswami J.N. GrafT. Graham D. Graham I.J. Gray C.M. Green D.H. Green P.F. Gregoire D.C. Gregory R.T. Griesshaber E. Griffin W.L. Griffiths R.W. Grivet M. Grobler N.J. Groves D.I. Guerrot C. Gui X. Guise P.G. Gulson B.L. Gupta J.N. Haas de G.J.L.M. Habfast K. Hallbauer D.K. Halliday A.N. Hamelin B. Hamilton P.J. Hanan B.B. Hannington M. Hansen B.T. Hansen K. Harmon R.S. Harper C.L. Harpp K.S. Harris J.W. Harris L.B. Harris N.B.W. Harrison T.M. Hart S.R. Hart W.K. Harte B. Hasebe N. Hashimoto A. Hashizume K. Hattori K.

43 67 38 38 20 13 39 39 39, 39, 86, 116 40 56, 91, 94 40 40 40 71 41, 62 1 23, 41 98 41, 53, 59 78 41, 61 24 73, 73 15 17, 63, 81 69 39 42 19, 49 83, 110 37, 42 76 42 110 8 19, 25, 42, 69 85,116 6, 43, 99 43, 89 22 90 9, 43, 52 24, 43, 68, 109 44 44 30, 112 11 19 44 45, 45, 59, 95 44, 108 112 98 25 45, 97 45

120

Hauri E.H. Hawkesworth CJ. Hayes J.M. Head J. Heaman L.M. Heijnis H. Hemond C. Henderson C.M.B. Henk F.H. Hensen BJ. Hergt J.M. Herv€ F. Hfjl E. Hidaka H. Hill R.L Hirabayashi J.i. HiroiY. Hiyagon H. Hoffman RJ. Hofmann A.W. Hohenberg C.M. Holm P.M. Honda M. Honda T. Hong A. Honma H. Houghton B. Houseman G. Housh T. Hu R. Huhma H. Hulston J.R. Hurford A.J. Hutcheon I.D. Hyodo H. Dzumi S. IkedaS. Ikeya M. Ingersoll R.V. Innocenti F. Ireland T.R. Isachsen C.E. Ishibashi J.-i. ItayaT. ItoH. Ivanenko V.V. Iwaki K. Izawa E. Jackson M.J. Jacquier B. Jagoutz E. Jahn B.-M. Janecky D.R. Javoy M. Jenkins R.J.F. Jenner G. Jensen P.K.

ICOG 7

45 19, 45, 46, 64 96 1 46 46 46 35 9 3 46 29, 61 108 47 47 88 52 6, 47, 88 61 18, 46, 47, 60, 66 70 48 48, 77 110 48 48, 50 80 37 13 49 81, 104 49 49, 59 54, 96 49 48, 50 50, 54 50 60 33 33, 50 13 50 36, 51 98 54 107 110 71 4 12, 51 51, 57 39 51 21 52, 54 43, 52

Jessberger E.K. Jiang S. Johansen H. Johnson C.M. Jonasson I.R. Jones B.G. Jones H,D. Jones P. Kagami H. Kamensky I.L. Kamioka H. Kamo S.L. Kamp P.J.J. Kaneoka I. Karlsson H.R. Karpenko M.I. Karsten J. Kasuya M. Kawashita K. Keene E.J. Kelley S.P. Kempton P.D. Kennedy A.K. Kennedy B.M. Kennedy W.J. Kerr A. Kersting A. Kesler S.E. Khrenov A.P. Kim J. Kinny P.D. Kirschvink J. Klein E. Knaus A. Knipe R.J. Kohl C.P. Kohler H. Kohn B.P. Kolodny Y. Komor S. Kowallis B.J. Kreulen R. Krishna V. Krishnaswami S. Krogh T.E. Kroner A. Krstic D. Kumar A. Kumar B. Kumar K.R. Kunk M.J. Kunzendorf H. Kusakabe M. Kutyavin E.P. Kwok Y.Y. Kyser T.K. Lahsen A.

82 27 65, 80 8, 52, 83 78 16 69 19,20 36, 48, 52 102 53 53 53 53 94 54 38 50, 54 8, 21, 69 6 32 109 54 80 63 54 55 69 55 63 11, 62, 97 21 41 22 110 71 89 55, 74 77, 105 104 55 30, 55, 90 76 88 56, 71, 84 56 23 56 56 76 57, 65 43, 52 48, 88 40 84 55 99

ABSTRACTS

ICOG 7

Laing W.P. LalD. Lambert D.D. Lambert LB. Lambert R.S.J. Lan C.-Y. Landoll J.D. Langmuir C. Lanphere M. Large R.R. Laurenzi M.A. LazarB. Leach D. Lee J. Lee J.K.W. LeeT. Leeman W.P. Leitch E.G. Lerche L Lesher C.M. Lewis C.L.E. Li P.-z. Lis. Li W-X. LiX. LiZ. Libby W.G. Liew T.C. Linn A. Lipman P. Lister G. Liu D. Liu J. Liu S. Liu T.-K. Liu Y. LiuZ. Lo C.H. Lodding A.R. Long L.E. Longstaffe F.J. Lopez-Escobar L. Losada-Calderon A. Loske W.P. Lovera O.M. Ludwig K.R. Ludwig K.R. Lugmair G.W. Lui J. Lundberg L.L. Lupton J. Lutz T.M. LuzB. Lyon LC. Maas R. Maboko M.A.H. Machado N.

75 71 92 37 5 57 35 86 57 38, 114 57 94 13 67 58, 74 18, 57, 58 58, 68, 100 36 16 58 59 59 59 85,116 59 23 26 60 60 83 37 59 60 23 58 60 27 58 31 31 11 109 61 61 44 41 61 61 16 62 41, 49, 62 55, 74 77 81 62 62, 64 46

121

Mahood G.A. Maluski H. Manega P.C. Manh^s G. Marantonio F. MarsM. Martel D.J. Marti K. Martin C.E. Masuda A. Mathur R. Matsuda J.-L Matsuhisa Y. Mawson R. Maxwell S. Mayeda T.K. McArthur J. McCandless T. McCulloch M.T. McDermott F. McDonough W.F. McDougall I. McGregor V.R. Mcintosh W.C. McLaren A.C. McLimans R. McNaughton N.J. McNutt R.H. Meams E.W. Medaris L.G. Jr Meesterburrie J. Meier M. Meijer A. Melnikov N.N. Mendelssohn M.J. Mertz D.F. Meshick A.P. Mezzetti F. Michard A. Migeon H.N. Milesi J.-P Millar I.L. Miller D. Miller E.L. Miller H. Miller M.F. Milne A.J. Mitchell J.G. Miura Y. Mizon K.J. Mizuta T. Monaghan M. Moorbath S. Mora C.I. Moran A.E. Moreau C.

25 108 108 40 31, 82, 85, 116 17, 63, 81 9 63, 71 63 47, 53, 91, 98 56 72 103 99 19, 20 25 63 86 2, 9, 17, 22, 33, 64, 93, 105, 115 45, 64 64 48, 62, 64, 77, 78 29 65 65 13 34, 39 27, 65 65, 80 8, 52 55 74, 95 66 40 93 66 92 57 1, 11, 34, 35, 51 66 11 66 86 37, 67 61 67 61, 66 100 67 42 67 26 24 68 58 26

122

Mori P.E. Morris J. Morrison J. Mortimer G.E. MueUerE. Mueller P. A. MuUer-Sohnius D. Munizaga F. MurreUMT. Nagai S. Nagao T. Naidin D.P. Nakai S. Nakamura N. Nanson G.C. Naqvi S.M. Natarajan M. Nelson D.R. Nemchin A. Neuser R.D. Newsom H.E. Neymark L. Nichols R.H. Jr. Niemeyer S. Nisbet E.G. Nishiizumi K. Nishimura S. Njine T. Noble S.R. Nohda S. Nojiri Y. Norelli P. Norman M.D. Notsu K. Nuti S. Nutman A.P. O'Nions R.K. O'Reilly S.Y. O'Sullivan P.B. Oberli F. Oddone M. Ohlander B. Okano J. Okrugin V.M. dinger C.T. Omar G.I. Onstott T.C. Osanai Y. OttU. Owada M. Ozima M. Page R.W. Pagel M. Palacz Z.A. Palin J.M. Palmer M.R. Pandey B.K.

ICOG 7

99 68, 86, 100 68 93 27 68 89 29, 69, 99 39 37 36 9 69 36 69, 92 94 56 69 2, 70 14 70 2, 70 70 71 91 71 98 88 21, 71, 90 36 88 1 71 72 72 6, 29, 72, 72 7, 9, 15, 27, 73, 105 73, 73 73 74 57 93 103 55 70 55, 74 58, 74 52 74 52 6, 37, 47, 75, 75, 79 75 108 75 76 76 76

Pankhurst R.J. Pantulu G.V.C. Pardini G. Park K.-H. Parrish R.R. Patchett P. J. Patterson D. Paul T.A. Paytan A. Pearce J. Pellas P. Peng T.-R. Perfit M.R. Perkins C. Perron C. Peters S.G. Petrini R. Peucat J.J. Pharaoh T.C. Phillips D. Piccirillo E.M. Pidgeon R.T. Pier J. Piguet P. Pillinger C.T. Pinarelli L. Pirrus E. Podosek F.A. Poorter R. Poths H. Poths J. Price D.M. Price R.C. Pringle M.S. Pu Z. Puchtel I.S. Pyle D. Qiu H. Quade J. Quercioli C. R&heim A. Raiswell R. Rajagopalan G. Raju R.D. Ramo O.T. Ramola R.C. Rao B.B. Rao D. R. Rasser B. Rathore S.S. Ratschbacher L. Read P. Rebetez M. Reedy R.C. Reid D. Reid M.R. Reimold W.U.

ABSTRACTS

29, 66 91,94 79 116 85 76, 87 48, 77 77 77 46 63, 77 109 78 78 63 39 78 42 102 74 78 78 79 107 67 79 40 13, 13, 79, 79 55 79 80 69 41 80 23 116 41 23 80 1 65, 80 12 94 76 81 87 56 91 66 106 22 81 17, 63, 81 81 82 12 82

ABSTRACTS

Reisberg L.C. Rejou-Michel A. Renne P. Rex D.C. Reynolds J.H. Richards J.P. Richards J.R. Richter D.K. Richter P.M. Riciputi L. Riedel D. Riggs A.C. Ritsk EJ. Robert F. Roberts B. Roberts J. Robinson B.W. Roddick J. C. Rogers G. Rogers P.S.Z. Roksandic Z, Rosenbaum J.M. Rosman K.J.R. Ross G. Ross RJ. Jr Rowan L. Rubenstone J.L. Rubin A.E. Rudnick R.L. Ruiz J. Rundberg R. Russ G.P. Ryan J.G. Sakai H. Sakamoto N. Salters V.J.M. Samson S.D. Samsonov A.V. Sandhu A. Sandiford M. Sano Y. Sarin M. Sarkar A. Sasada T. Satake H. Sato K. Saunders A.D. Sawyer D.A. Saxton J.M. Schaaf P. Scharer U. Schilling J.-G. Schi0tte L. Schleicher H. Schliestedt M. Schmincke H.-U. Scholten S.O.

ICOG 7

123

12, 31, 82, 85, 116 83, 84 82 83, 110 80 64 83 14 44 83 14 61 2 83, 84 14 19,20 84 84 84 66 85, 97 85 34 85 102 13 14,31, 82, 85, 85, 116 70 86 86, 86 79 71 86 47, 48, 50 89 31, 85, 87, 116 76, 87 6, 72, 87, 116 87 103 50, 88 88, 88 9, 88 88 89 99 100 57, 83 103 89 89 43, 89 90 90 111 95 90

Schulz L. Schwartzman D.W. Scott S.D. Scrivener R.C. Sharma K.K. Sharyf-Zade H.B. Shatsky V.S. Shepherd T.J. Sheppard S.M.F. Sheraton J.W. Shima M. Shimizu H. Shinohara H. Shirey S.B. Short S.A. Shukolyukov Yu. A. Siddall R. Siga 0. Jr Simmons K.R. Simon A.K. Sims K. Singh S. Sisson V.B. Sivaraman T. V. Sivell W.J. Skiold T. Smith B.M. Smith C.B. Smith D. Smith H. Smith H.A. Sobolev N.V. Socki R.A. Srinivasan G. Srinivasan R. Srivastava A.P. Srivastava R.K. Sriwana T. Starinsky A. Staudigel H. Steckler M.S. Steiger R.H. Stein M. Stephan T. Stephenson R.A. Stevens B.P.J. Stewart B.W. Stewart K.P. Stijfhoom D. Stoffers P. Stone J. Stosch H.G. Strauss H. Stray H. Strydom J.H. Strydom N. Stuart P.M.

90 91 22, 67 19 91 92 51 19,24 26, 91 11 102 91, 98 88 90, 92, 92 69, 92 92, 106 93 8 61 116 70 87 58 94 93 93 95 82 81 7 93 51 94 40 94 94 106 55 94, 105 95 74 10, 95 95 82 111 75 95 96 80 66 96 109 14, 96 65 108 108 103

124 Stump E. Sugiura N. Sullivan R.W. Summons R.E. Sun S.-s. Sutherland F.L. Sutter J.F. Suzuki K. Swan P. Sweeney RJ. Swisher C.C. Szabo BJ. Tagami T. Taieb R. Takaoka N. Takigami Y. Talent J. A. Tanaka T. Tanyileke G. TaoX. Tassinari C.C.G. Tatsumoto M. Tauxe L. Taylor H.P. TeixeiraW. Tera F. Terrell DJ. Tewari V.C. Thirlwall M.F. Thorn A. Thorogood EJ. Thorpe R.I. Tiercelin J.-J. Todt W. Togashi S. Tolstikhin IN. TomuraK. Tonarini S. Torigoye N. Treloar P. J. Trendall A.F. Trivedi J.R. Tsuchiyama A. Tsutsumi M. Tsvetkov A.A. Tucker R.D. Turchenko T.L. Turek A. Turekian K.K. Turi B. Turner G. Turner S.P. Tuttas D. Ui T. Ujiie M. Ulyanov A.A. Uyeda C.

ICOG 7

96 45, 97 84 96, 97 37, 64, 97 97 65, 91 98 2 98 98 61 98 91 53 99 99 99 88 33 21, 69, 99 41, 59 26 34 21 68, 100 100 56 63, 100 100 101 101 35 66 101, 103 101,102 75 33, 34 102 83 69 88 103 50 100 102 40 102 32, 63 34 81, 103 34, 103 110 72 103 40 103

Vaasjoki M. Valbracht P.J. Valley J. Van Bergen M.J. Van Calsteren P. Van den Haute P. Van der Plicht J. Van Wees J.D. van Breemen 0. Vance D. Vannucci R. Veeh H. Veizer J. Vellutini P.J. Vengosh A. Venkatesan T.R. Verchovsky A. Verma S.P. Verschure R.H. Vervoort J.D. Vetrin V.R. VidalP. Viets J. Villa I.M. Vincent J. Virk H.S. Vroon P.Z. Wada H. Wagner G.A. Wagner M.E. WakitaH. WalderA. Walgenwitz F. Walker R.M. Wallace W.K. Walraven F. Walter R.C. WanD. Wang C.-H. Wang J. WangW. Wang X. Wang Y. Ware A.J. Wartho J.-A. Wasserburg G.J. Watanabe K. Watanabe T. Weaver S. Weber H.W. Weis D. Welke H.J. Wen Q. Wendt I. Wendt J.I. Wernicke B.P. Wheeler G.E.

ABSTRACTS

81, 104 104 16, 22, 104 104, 107 19, 64 108 46 111 105 105 79 5 14, 105 107 105 106 106 106 42 106 102 107 13 57, 107 107 87 104, 107 107 108 55 50 75 108 2,2 73 108 108 27 109 109 109 109 109 109 110 18, 54, 95, < 110 36 31, 82, 85, : 90 26 82 109 110 110 34 111

ABSTRACTS

ICOG 7

White W.M. Whitford DJ. Wiedenbeck M.G. Wijbrans J.R. Wilde S.A. WUdingM. Williams I.S. Williams S.H. Wilson G.S. Wooden J.L. Woodhead J.D. Wooding R. Womer G.W. Wright D.W. Wright J.E. Wyllie PJ. XiaB. Xu S. Xu Y. Xue X. Yamamoto K. Yamanaka T. Yamazaki H. Yanagi T. Yang J. Yang N. Yim W.W.-S. York D. You C.-F. Yu J.-S Yui T.-Fu Zartman R.E. Zashu S. ZawK. Zeitler P.K. Zentilli M. Zhang G. Zhang L. Zhang Z. Zhang Z. Zhao J.-X. Zheng Y.-F. Zhou X. Zhu Z.R. Zhuravlev D.Z. Zindler A. Zinner E.

125

44, 106, 107, 110 99, 111 111 16, 111, 111 78 112 4, 10, 13, 21, 21, 30, 33 38, 65, 112 102 112 68, 113 22, 112 7 24 64 113 54 33 109 113 109 36 103 101 113 33, 109 109 113 49,111 58 49, 59 114 114 47, 75 114 62, 93, 114 69 59 115 59 21, 115 93, 115 115 116 95 87, 116 12, 14, 31, 82, 85, 85, 87, 116, 116 2, 62

The VG Prism Series II Stable Isotope Ratio Mass Spectrometer

The VG PlasmaOuad P02 ICP-MS

A total commitment from VG Instruments The VG Sector 54 Thermal lonisation Mass Spectrometer

Outstanding performance with innovative ion optics for routine high precision measurements and the added flexibility to undertake the most demanding, novel experiments. High abundance sensitivity The 30cm electrostatic filter (ESF) for use with the VG Sector 54 provides

very high abundance sensitivity suitable for U/Th disequilibrium and similar studies. The ESF is the only proven, reliable method which gives high abundance sensitivity, high transmission and long-term stability.

The VG Prism Series II Stable Isotope Ratio Mass Spectrometer Essential analytical power for the Geochemist, Geologist and Oceanographist - proven ion optics, a wide range of sample preparation systems and comprehensive computer control for precision analyst of very small gas samples. Laser ablation The PRISM Series II is also available with a direct laser ablation system. Laser ablation allows the measurement of and '^Of'O ratios in carbonate overgrowths and the measurement of sulphur isotopes

The V G Sector 54 Thermal lonisation Mass Spectrometer

across microveins at a resolution of 10 microns.

The VG PlasmaOuad P02 ICP-MS The world's leading instrument for rapid direct measurement of elemental concentrations from percent to ppt levels. Full elemental and isotopic coverage with easy to use software and a complete range of sample introduction techniques, including Laser ablation and Electrothermal vaporisation.

VG INSTRUMENTS VG Instruments Inorganic Group Aston Way, Middlewich, Cheshire CW10 OHT, Englana Telephone: (44) 0606 846725 ^ t ^ Q f ^ S Fax: (44) 0606 846487 Instruments F ' ^ Instrun

Isolab analysis chamber and rapid sample entry lock

VG ISOLAB A new generation ion microDrobe Several key features establish the VG ISOLAB as leading the field in multiple collection ion microprobe technology. The ultimate multiple collector ion microprobe • High spatial resolution imaging of negative and positive ions • High sensitivity at high mass resolution for separation of isobaric interferences • Multiple collection for enhanced sensitivity and precision of isotope ratio measurements • Analysis of insulators • High dynamic range depth profiling • Ultra high vacuum for ultimate detection limits of H, C, N, and O The ultimate isotope laboratory • Analysis of smaU samples through selected use of SIMS, TIMS, Laser ionisation or combination thereof • In situ isotopic analysis of insulators • Optical, secondary electron and ion imaging for ease of small sample location

Secondary ion mass spectrometry (SIMS) is a well established analytical technique for elemental mapping, isotopic analysis and depth profiling IQ the material, earth and biological sciences. Where high lateral resolution is required SIMS exhibits superior sensitivity and dynamic range relative to other surface techniques. SIMS instrumentation routinely provides detection limits at the ppm level and in favourable cases can operate at the ppb level. The VG ISOLAB represents a dramatic improvement in sensitivity relative to existing SIMS instrumentation and a substantial increase in precision for isotope ratio measurement. This is achieved through increased analyser transmission and the use of multiple collection. VG ISOLAB also provides the option of laser ionisation to address the problem of poor sensitivity for elements that are difficult to ionise by SIMS. A thermal ionisation source is available for TIMS and combined SIMS/TIMS.

• Ultra high abundance sensitivity allows the measurement of low abundance isotopes that would normally be obscured by their high abundance neighbours • Advanced data system provides fully automated data collection and flexibility to design experiment without software changes • Modular design allows specific configuration to individual needs and provides for future enhancements

VG INSTRUMENTS VG Instruments Inorganic Group Aston Way, Middlewich, Cheshire CWIO OHT, England Telephone: (44) 0606 846725

--CONS

Fax: (44) 0606 846487

PI^instruments

ISOTOPE RATIOS GEOLOGICAL SAMPLES? For any analytical challenge contact Finnlgan MAT

Finnigon miiT

U.S.A., Finnigan MAT, 355 River Oaks Parkway, San Jose, CA 95134-1991 • (408) 433-4800 Federal Republic of Germany, Finnigan MAT GmbH, Postfach 14 40 62, D-2800 Bremen 14 • (04 21) 54 93-0 France, Finnigan MAT S.a.r.l., Pare Club Orsay Universite, 2, Rue J. Monod, F-91893 Orsay Cedex • (01) 69 41 98 00 Italy, Finnigan MAT S.rl., Via Valadier 37B, 1-00193 Rome • (06) 6878057 Japan, Finnigan MAT Instruments, Inc., Shiozaki Building, 2-7-1 Hirakawa-cho, Chiyoda-ku, Tokyo 102, Japan • (03) 221-1001 The Netherlands, Finnigan MAT Benelux, Landjuweel 7, NL-3905 PE Veenendaal • (08385) 27266 People's Republic of China, Finnigan MAT China, Inc., No. B7, Baishiquiao Road, Beijing 100081 • (861) 831 6579 Sweden, Finnigan MAT AB, Vastertorpsvagen 135, S-126 59 Hagersten, • (08) 970150 United Kingdom, Finnigan MAT Ltd., Paradise, Heme! Hempstead, Herts HP2 4TG • (0442) 233555

ANUTECH Pty Ltd's instrument division AUSTRALIAN SCIENTIFIC INSTRUMENTS

Invites AllICOG attendees and interested parties to visit the SHRIMP display at Melville Hall, in the Australian National University during the conference week.

•PIONEERING THE TECHNIQUE'

SHRiMP

'PERFECTING THE TECHNOLOGY'

This display details the science and engineering involved in the construction of the Sensitive High Resolution ion Microprobe [SHRIMP II], the worlds most powerful instrument for in situ geological dating and micro chemical analysis. Highly Informative visual display

Tours of the renowned SHRIMP facility at the RSES arranged

Customer enquiries personally addressed

Refreshments available between 5.30 and 6.30 pm

AUSTRALIAN SCIENTIFIC INSTRUMENTS A DIVISION OF ANUTECH PTY LTD