Abstracts No.48 Palaeobiogeography of Australasian Faunas & Floras, 1997, Wollongong by GSAustralia

Geological Society of Australia

ABSTRACTS Number

Palaeobiogeography of Australasian Faunas and Floras University of Wollongong December 8-11,1997 • Association of Australasian Palaeontologists • Geological Society of Australia • International Palaeontological Association • IGCP410 • IGCP 421

PALAEOBIOGEOGRAPHY OF AUSTRALASIAN FAUNAS AND FLORAS DECEMBER 8-11,1997

ABSTRACTS AND PROGRAMME ORGANISED UNDER THE AUSPICES OF SCHOOL OF GEOSCIENCES UNIVERSITY OF WOLLONGONG ASSOCIATION OF AUSTRALASIAN PALAEONTOLOGISTS INTERNATIONAL PALAEONTOLOGICAL ASSOCIATION & INTERNATIONAL GEOLOGICAL CORRELATION PROGRAMMES 410 AND 421

GEOLOGICAL SOCIETY OF AUSTRALIA ABSTRACTS NUMBER 48 PUBLICATION DATE, DECEMBER 8-11,1997 PRINTED BY THE UNIVERSITY OF WOLLONGONG COMPILED AND EDITED BY A.J. WRIGHT

A CONFERENCE HELD UNDER THE SPONSORSHIP OF ASSOCIATION OF AUSTRALASIAN PALAEONTOLOGISTS

INTERNATIONAL PALAEONTOLOGICAL ASSOCIATION

GEOLOGICAL SOCIETY OF AUSTRALIA AND THE SCHOOL OF GEOSCIENCES, UNIVERSITY OF WOLLONGONG

AS WELL AS INTERNATIONAL GEOLOGICAL CORRELATION PROGRAMMES 410 AND 421

CONVENORS: PROFESSOR JOHN TALENT DR GAVIN YOUNG DR TONY WRIGHT

Ill

ISSN 0729 Oil X

Copies of this publication can be obtained from the Geological Society of Australia Incorporated, 706 Wynyard House, 301 George Street, Sydney NSW AUSTRALIA 2000.

A sample citation for this volume is: Antoshkina, A., 1997. Specific reef biotas of the Urals and their importance for the mid-Palaeozoic palaeogeography. Geological Society of Australia Abstracts 44,1-2.

Material presented here may be freely copied for library, educational or research purposes, but replication of any part requires written permission of the Geological Society of Australia, as well as appropriate acknowledgment.

TAXONOMIC/NOMENCLATURAL DISCLAIMER This publication is not deemed valid for taxonomic or nomenclatural purposes (see Article 8b in the ICZN, edition).

TABLE OF CONTENTS Page iv-vi

General conference information

vii-xii

List of contributors

xiii

List of delegates

xiv-xxi

Programme of oral presentations

1-117.

Abstracts

GENERAL CONFERENCE INFORMATION VENUE. The conference will be held at the University of Wollongong, in Building 20, lecture theatre 20.1 on Monday and Tuesday; as we go to press, the Thursday venue is not finalised! Your registration satchel contains, among other things, maps of the campus and city of Wollongong. A campus map is also printed on the front inside cover of this volume.

REGISTRATION. Registration will be held on the afternoon of Sunday December from 2 pm until about 6 pm, in the first floor, southern foyer of Building 41. On Monday S^^, registration will be in the foyer of Building 20 from 8-9 am. ACCOMMODATION.

Most accommodation is at "Kooloobong", a student residence located on Northfields Avenue, at the western end of the campus. In clement weather this residence is a pleasant walk from the conference venue.

PUBLIC LECTURE.

On Monday evening at 7.30 pm, Dr Malcolm Walter will present a public lecture in the large theatre 20.1, again in the "Pentagon'' building, on "Search for Life on Mars". This is intended to attract a wide audience, and to publicize both palaeontology and our conference.

WORKSHOPS and MEETINGS. There will a meeting of the Association of Australasian Palaeontologists in theatre 20.1 on Monday night after the public lecture. It is expected to commence at about 8.30. On Tuesday evening, commencing at about 7.15 pm, following a general introduction in our conference theatre to the International Geological Correlation Programme, there will be separate workshops on the following IGCP themes: IGCP 410: IGCP 421:

The great Ordovician biodiversity event. North Gondwana mid-Palaeozoic biogeography/bioevent patterns in relation to crustal dynamics.

Following these meetings there will be a meeting of the Pander Society.

ORAL PRESENTATIONS. Keynote papers will be allocated 25 minutes plus 5 minutes question time, and other papers will have 15 minutes plus 5 minutes question time. We apologise for this regrettable departure from the advertised times, but it is necessitated by the very full programme and our desire to avoid concurrent sessions. Speakers must contact the chair of their session well before their presentation, and give any loaded slide carousels, with any special instructions, to the projectionist before the beginning of their session. All speakers are responsible for loading their carousel in the correct order and orientation. This loading can be done in the large Geology laboratory. Room 41/153 (NB .. Geosciences building!), where several projectors will be available for final checking. Please write your name on the label stuck to the top of the carousel. Carousels can be obtained from the projectionist, and will be returned to speakers at the end of the session for unloading, after which they should be returned to the projectionist or the conference desk.

VISUAL AIDS. The lecture theatre will have 2 x 35 mm slide projectors operated by remote control from the speakers' podium, and laser pointers will be available. An overhead projector will also be set up for use in the theatre. POSTERS.

Posters should be set up on Sunday afternoon or early Monday morning. Delegates presenting posters will have a scheduled time (see programme) for presentations of a "mini-lecture'' and discussions, at the times indicated in the programme.

REFRESHMENTS. Every attempt will be made to maintain a supply of the necessary goodies for morning and afternoon tea and coffee in the foyer of building 20 outside the lecture theatre. LUNCH. Lunch can be purchased from one of the many campus outlets, especially in the Uni Centre. EVENING MEALS. Meals are available at very few outlets on campus at this time of the year. The Uni Tavern offers liquid refreshments and a light menu including a barbecue and salad menu; we have been assured that this outlet will have no difficulty coping with conference numbers, but all delegates should wear their PAFF name tag to ensure high quality service.

CONFERENCE DINNER. The dinner will be held in the UniCentre, in the upstairs Function Rooms. This will be held on Wednesday at 7 pm for 7.30 pm. Tickets must be purchased by 5 pm on Monday delegates who have already paid will find their ticket in their conference satchel. It should be possible to park cars in the parking station (reached via the western gate on Northfields Avenue), and walk through to the Function Centre. PARKING. At this time of the year, parking areas near buildings 35 and 41 should be reasonably accommodating; free entry to the campus has been granted. Delegates residing at Kooloobong who have a vehicle will find ample safe (but lock that car!) street parking. CONFERENCE PUBLICATIONS. As delegates will know, there will be 2 avenues of publication. One will be a volume in the Oxford University Press Biogeography series; Ms Jill Lane of the Sydney OUP office has been invited to attend the conference on Monday, and is expected to be available for discussions. We understand that a special pre-publication purchase price for this OUP volume will be available to conference delegates. The other publication venue will be a special issue of Historical Biology \ intending contributors should submit their manuscript to one of the Convenors at the conference.

PUBLICATIONS FOR SALE. Some publications will be offered for sale by Misha Frankel, of the Geological Society of Australia; these will include the journal Alcheringa and Memoirs of the Association of Australasian Palaeontologists. The second edition of the MUCEP "Blue Book'' .. directory of Australasian palaeontologists will be on sale at the conference.

Vll

LIST OF CONTRIBUTORS ^indicates the speaker in the case of a multi-author paper

J.C. AITCHISON. Australian Devonian radiolarians. (poster) A. ANTOSHKINA. Specific reef biotas of the Urals and their importance for the mid- Palaeozoic palaeogeography. N.W. ARCHBOLD. The palaeobiogeography of the Australian Permian, (keynote address) N.W. ARCHBOLD. Late Palaeozoic Insecta and Arachnida of Gondwana: a review. M. ARCHER. Messing with mammalian meanderings. (keynote address) J.-S. BAO & J.B. JAGO*. Late Late Cambrian trilobites from the Birch Inlet area, southwestern Tasmania. M. BAMFORD. Australasian floral elements in southern Africa: Tertiary woods of the Myrtales. E. BARONE-NUGENT, A. DRINNAN & S. McLOUGHLIN. Late Triassic flora of the Leigh Creek Coal Measures, South Australia, (poster) C. BENTLEY* & J.B. JAGO. Some comments on agnostoid taxonomy and classification. C. BENTLEY & J.B. JAGO. Revision of the family Catillicephalidae Raymond 1938 (Trilobita). (poster) C.J. BURROW. A poracanthodid acanthodian from the Silverband Formation, the Grampians, Victoria, and its palaeobiogeographical significance, (poster) D.J. CANTRILL. The geographic distribution of the extant fern Lophosoria quadripinnata: implications for Cretaceous climates of the southern high latitudes.

Z.Q. CHEN* & G.R. SHI. Carboniferous brachiopods of China: a review on their distributions in space and time. P. COCKLE. Time, space and environmental relationships in the Silurian succession of Boree Creek, NSW. (poster) A.G. COOK. Australian Devonian gastropod biogeography. A.G. COOK. Devonian gastropods from Brogans Creek, NSW. (poster) R.A. COOPER. Ordovician graptolite species longevity, faunal turnover and biochronology. (keynote address) R.S. CRAIG. The Cenozoic brachiopods of Western Australia. J.G. DOUGLAS* & J.D. TIMS. Biogeography of Early-Middle Palaeozoic floras of southeastern Australia. A. DRINNAN*, T. C. CHAMBERS & S. McLOUGHLIN. Morphology of the recently discovered Wollemi pine {Wollemia nobilis: Araucariaceae) and comparisons to Cretaceous and Tertiary fossils. M.K. EAGLE. A Late Ordovician biostrome fauna from Hailes Knob, New Zealand. M.C. EBACH. Phylogenetic and area cladistic analyses of a Devonian trilobite fauna from western New South Wales. J.A. FAGERSTROM* & O. WEIDLICH. Building the Upper Capitan Reef framework (Permian), Guadalupe Mountains, New Mexico-Texas, U.S.A: inter-relations between paleoecology and abiotic processes. R.A. FORTEY. Ordovician palaeogeography of the Shan-Thai terrane. (keynote address) C. B. FOSTER. Caheniasaccites Bose & Kar, 966, Costatascyclus Felix & Burbidge, 1967 and allied genera: portrait of a migrating gymnosperm from North America, Gondwana, and Russia. L. FRAKES. Phanerozoic GCM results and climate data for Australia, (keynote address) A. GARCIA & A.R. CHIVAS. Biogeography of Quaternary and extant Lamprothamnium Groves (Charophyta) in Australia, (poster) M. GHAVIDEL-SYOOKI. Upper Palaeozoic strata in the Zagros Basin, Southern Iran, (poster)

M. GHAVIDEL-SYOOKI. Biostratigraphy and palaeobiogeography of the Late Ordovician-Early Silurian chitinozoans from the Zagros Basin, Southern Iran. J.A. GRANT-MACKIE* & HJ. CAMPBELL. Australasian Triassic and Jurassic biogeography. (keynote address) E. HAKANSSON. Australasian lunulite bryozoans - the European connection? A. HALLAM. An overview of Australasian palaeobiogeography in its global context, (keynote address) T. HAMLEY. Biogeography of the procolophonids. S. HAND. New world order in bat biogeography. R.A. HENDERSON*, J. CRAMPTON, M. DETTMANN, D. HAIG, R. MOLNAR, S. SHAFIK, J. STILWELL & T. THULBORN. Biogeographical observations on the Cretaceous biota of Australasia. W.B.K. HOLMES. From go to woe - the Triassic vegetation of eastern Australia. J.-J. JAEGER* & Y. CHAIMANEE. The evolution of the Rattus group (Mammalia, Rodentia) during the Plio-Pleistocene in Thailand. P. J. JONES*, B. A. ENGEL, L METCALFE, G. PLAYFORD, J. RIGBY, J. ROBERTS, S. TURNER & G.E. WEBB. Carboniferous biogeography of Australasia. A.S. KASSAB. Variation, biostratigraphy and palaeobiogeography of the Upper Cretaceous coilopoceratid ammonites from Egypt. A. KELMAN. Surface microstriation on Lower Ordovician conodont elements, (poster) A.P. KERSHAW. The palaeobiology of the Quaternary - a partial overview, (keynote address) J.G. LLEWELLYN. Some early studies in biogeography and palaeobiogeography of Australasian faunas and floras. J.A. LONG. Palaeozoic fish faunas of Australia, (keynote address) R. MAWSON & J.A. TALENT. Eastern Molong Platform and adjacent Hill End Trough, New South Wales: Mid-Palaeozoic conodont data.

age-inferences, and implications regarding platform exposure and carbonate sedimentation offshore, (poster) B. McGOWRAN* & Q. LI. Australasian Cainozoic biogeography: environmental framework for a southern-temperate province on a cooling planet, (keynote address) B. McGOWRAN*, M. ARCHER, T. DARRAGH, Q. LI, P. MAXWELL, K. McNAMARA, M. McPHAIL, A. PARTRIDGE, J. RICHARDSON, S. SHAFIK, E. TRUSWELL & M. WARNE. Australasian Cainozoic biogeography S. McLOUGHLIN*, A.N. DRINNAN & S. LINDSTROM. Permian-Triassic floras of East Antarctica: palaeoenvironmental setting and biogeographic affinities. 1. METCALFE. Australian-Southeast Asian breakup in relation to biogeographic patterns in the region, (keynote address) R. MORGAN. Crawney Limestone - a determination of conodont biostratigraphy using polygnathids. (poster) A. MUSSER. Examining relationships of the monotremes in a biogeographical context. T. MYERS. Comparative palaeoecology of Cainozoic Riversleigh faunas, (poster) N. NAGALINGUM, A. DRINNAN & S. McLOUGHLIN. New and revised fossil plants from the Lower Cretaceous Boola Boola flora, Gippsland Basin, Victoria, (poster) R. PARKES & A. SIMPSON. A Late Silurian acanthodian from coeval carbonates in Victoria and New South Wales, (poster) I.G. PERCIVAL*, B.D. WEBBY & Y.Y. ZHEN. Revised Ordovician biostratigraphy of New South Wales. J.W. PICKETT. Silurian corals of Australasian aspect. C.McA. POWELL* & Z.X. LI. Palaeogeographic evolution of the Australasian region since the early Neoproterozoic. (keynote address) J. RIGBY. The Permian flora of the Mt Mulligan Coal Measures, Queensland, Australia, (poster) R.B. RICKARDS. Eastern Australian Silurian graptolites in time and

space, (keynote address) A. RIX & A. SIMPSON. The fauna and flora of the Eocene Redbank Plains Formation: summary of recent progress, (poster) J.D. SCANLON*, M. ARCHER, S. HAND & H. GODTHELP. Preliminary biostratigraphic analysis of snakes from the Tertiary of Riversleigh, Queensland. S. SHAFIK. The Cretaceous of Australia: the biostratigraphic system, (poster) S. SHEN*, N.W. ARCHBOLD & G.R. SHI. Changhsingian (Late Permian) brachiopod palaeobiogeography. J. SHERGOLD*, R. FEIST & D. VIZCAINO. Middle-Late Cambrian transition trilobites of Australo-Sinian aspects in southern France. G.R. SHI. Terrane rafting enhanced by contemporaneous climatic amelioration as a mechanism of biogeographical vicariance: Permian marine biogeography of SE Asia. A. SIMPSON. Late Ordovician conodonts from Gray Creek and the head-waters of Stockyard Creek, north Queensland, (poster) A. SIMPSON. Early Silurian conodonts from the Broken River region, north Queensland, (poster) G.D. STANLEY, Jr. Palaeogeography of Upper Triassic Tethyan reef faunas from the Northwestern Australian Shelf. J.D. STILWELL. Tectonic and palaeobiogeographic significance of the Chatham Islands, South Pacific, Late Cretaceous fauna. J.A. TALENT, J.C. AITCHISON, M. BRADSHAW, G. DARGAN, R. MAWSON, T. WINCHESTER-SEETO, G. YOUNG & Y.-Y. ZHEN. Palaeobiogeography of Australasian Devonian biota. A.-M. TOSOLINF, N. NAGALINGUM, S. McLOUGHLIN & A. DRINNAN. An Early Cretaceous flora from the Lower Strzelecki Group, Gippsland Basin, Victoria. A.J. VADALA* & A.N. DRINNAN. A Late Paleocene leaf flora from Cambalong Creek, Southern Highlands, New South Wales. M.R. WALTER*, J. GEHLING & K. GREY. Australian Neoproterozoic stratigraphy, geography and biogeography. (keynote address)

xu A. WARREN*, R. DAMIANI, C. NORTHWOOD & A. YATES. Palaeobiogeography of Australian fossil amphibians. G.E. WEBB. Problems for palaeobiogeographical analysis of Palaeozoic corals as illustrated by Lower Carboniferous corals of eastern Australia. B.D. WEBBY*, I.G. PERCIVAL, A.H.M. VANDENBERG, J.W. PICKETT, J. POJETA Jr., R. SCHALLREUTER, T. WINCHESTER-SEETO, G.D. EDGECOMBE, R.A. COOPER & Y.Y. ZHEN. Ordovician biogeography of Australasia. G.D. WEBSTER*, P.A. JELL & A.N. DEREWETZKY. Paleobiogeography of Permian echinoderms of Australia. T.C. WINCHESTER-SEETO. Chitinozoans and foraminiferal linings from the Ordovician of Australia: an overview. S. WROE. A new marsupial from the Early Eocene Tingamarra local fauna of Murgon, Southeastern Queensland and its significance in the understanding of Australian marsupial origins, (poster) Weiping YANG. The Early Permian Australasian microflora in West Yunnan and its palaeobiogeographic and tectonic significance. M. YAZDI. Late Devonian bioevents and biogeographic patterns in relation to sea level changes in the Shotori ranges, East Iran. M. YAZDI. Ironstone facies throughout the Frasnian-Famennian in Central Iran: a Gondwana key in East Iran, (poster) G.C. YOUNG. Lower vertebrates of East Gondwana - biogeographic patterns, (keynote address) Y.Y. ZHEN, A.J. WRIGHT & J.S. JELL. Rugosan diversifications and migrations in the Devonian of Australia, (poster)

Xlll

LIST OF DELEGATES ANTOSHKINA, Anna ARCHBOLD, Neil ARCHER, Mike BAMFORD, Marion BARONE-NUGENT, Eroia BASDEN, Alison BEATTIE, Joan BELL, Ken BENTLEY, Chris BORIG, Natalie BRAMMALL, Jenni BROCK, Glenn BURROW, Carole CAMPBELL, Hamish CANTRILL, David CHAIMANEE, Yaowalak CHEN, Xiu-qin CHEN, Zhongqian COCKLE, Peter COOK, Alex COOPER, Roger COTTER, Kaye CRAIG, Rob DARGAN, Gary DARRAGH, Tom DETTMANN, Mary DOUGLAS, John DRINNAN, Andrew EAGLE, Michael EBACH, Malte EDGECOMBE, Greg ENGELBRETSEN, Michael FAGERSTROM, A1 FARRELL, John FERGUSSON, Chris FORTEY, Richard FOSTER, Clinton FRAKES, Larry FREW, Linda GARCIA, Adriana GHAVIDEL-SYOOKI, Mohammad GILLARD, Tom GRANT-MACKIE, Jack GROVER, Monty HAKANSSON, Eckart HALLY, Lee Ann HALLAM, Tony HAMLEY, Tim HAND, Sue HENDERSON, Bob HOLLOWAY, Gary HOLMES, Keith IRWIN, Philip JAGO, Jim JELL, John JELL, Peter JIN, Yu-gan JOHANSON, Zerina JONES, Peter KASSAB, Ahmed KEAR, Ben

KELMAN, Andrew KERSHAW, Peter LAURIE, John LEE, Mike LEONOVA, Tatyana LI, Man-ying LLEWELLYN, Jen LONG, John MAWSON, Ruth McGOWRAN, Brian McHENRY, Ben McLOUGHLIN, Stephen METCALFE, Ian MORGAN, Bob MUNSON, TIM MURRAY, Suzanne MUSSER, Anne MYERS, Troy NAGALINGUM, Nathalie NEIL, John NICOLL, Bob PARKES, Ross PERCIVAL, Ian PICKETT, John POWELL, Chris RICKARDS, Barrie RIGBY, John ROBERTS, John SANDFORD, Andrew SHARK, Samir SHAH, S.C. SHEN, Shuzhong SHERGOLD, John SHI, Guang-r SIMPSON, Andrew SLOAN, Terry SONE, Masatoshi STANLEY, George STILWELL, Jeffrey STRUSZ, Des TALENT, John TIMS, Jackie TONG, Jinnan TOSOLINI, Anne-Marie TRINAJSTIC, Katherine TURNER, Sue WALTER, Malcolm WARREN, Anne WEBB, Greg WEBBY, Barry WEBSTER, Gary WILLIS, Paul WINCHESTER-SEETO, Theresa WRIGHT, Tony WROE, Steve YAN, Jiaxin YANG, Weiping YAZDI, Mehdi YOUNG, Gavin ZHEN,Yong-yi

XIV

PROGRAMME FOR ORAL AND POSTER PRESENTATIONS MONDAY DECEMBER

1997

SESSION 1 (9.00-10.30) Chairman: Tony Wright, Wollongong 9.00-9.10. Opening remarks by Professor Gerard Sutton, Vice Chancellor, University of Wollongong 9.10-9.20. Welcoming remarks by Professor Allan Chivas, Head, School of Geosciences, University of Wollongong. 9.20-9.30. Introductory remarks by John Talent, President, International Palaeontological Association; joint convenor of IGCP 421; and member of Organising Committee, PAFF. 9.30-9.40. John Jell, President, AAP. 9.40-10.00. General remarks about conference arrangements by Tony Wright, member of PAFF Organising Committee. 10.00-10.30. Tony Hallam: An overview of Australasian palaeobiogeography in its global context, (keynote address)

MORNING TEA: 10.30-11-00 POSTERS Eroia BARONE-NUGENT, Andrew DRINNAN & Stephen McLOUGHLIN. Late Triassic flora of the Leigh Creek Coal Measures, South Australia. Peter COCKLE. Time, space and environmental relationships in the Silurian succession of Boree Creek, NSW. Alex COOK. Devonian gastropods from Brogans Creek, NSW. Andrew KELMAN. Surface microstriation on Lower Ordovician conodont elements. Mehdi YAZDL Ironstone facies throughout the Frasnian-Famennian in Central Iran: a Gondwana key in East Iran.

SESSION 2 (11.00-12.50) Chairman: Ian Metcalfe, Armidale 11.00-11.20. Jen Llewellyn: Some early studies in biogeography and palaeobiogeography of Australasian faunas and floras. 11.20-11.50. Chris PowelP and Z. Li. Palaeogeographic evolution of the Australasian region since the Early Neoproterozoic. (keynote address) 11.50-12.20. Larry Frakes: Phanerozoic GCM results and climate data for Australia, (keynote address) 12.20-12.50. Malcolm Walter*, Jim Gehling and Kath Grey: Australian Neoproterozoic stratigraphy, geography and biogeography. (keynote address)

LUNCH: 12,504.50 SESSION 3 (1.50-3.30) Chairman: Richard Fortey, London 1.50-2.10. Jin-song Bao and Jim Jago*: Late Late Cambrian trilobites from the Birch Inlet area, southwestern Tasmania. 2.10-2.30. John Shergold*, Raimund Feist and Daniel Vincaino: Middle-Late Cambrian transition trilobites of Australo-Sinian aspects in southern France. 2.30-2.50. Chris Bentley* and classification.

Jim Jago:

Some comments on agnostoid taxonomy and

2.50-3.10. Glenn Brock*, Michael Engelbretsen, Jim Jago, John Laurie and James Sorauf: Palaeobiogeographic trends of Cambrian faunas from Australia. 3.10-3.30. Michael Eagle: A Late Ordovician biostrome fauna from Hailes Knob, New Zealand.

AFTERNOON TEA: 3.30-3.50 POSTERS Carole BURROW. A poracanthodid acanthodian from the Silverband Formation, the Grampians, Victoria, and its palaeobiogeographical significance. Adriana GARCIA & Allan CHIVAS. Biogeography of Quaternary and extant Lamprothamnium Groves (Charophyta) in Australia. Ross PARKES & Andrew SIMPSON. A Late Silurian acanthodian from coeval carbonates in Victoria and New South Wales. Samir SHAFIK. The Cretaceous of Australia: the biostratigraphic system. Yongyi ZHEN, Tony WRIGHT & John JELL. Rugosan diversifications and migrations in the Devonian of Australia.

SESSION 4 (3.50-6.10) Chairman: Barrie Richards, Cambridge 3.50-4.20. Roger Cooper: Ordovician graptolite species longevity, faunal turnover and biochronology. (keynote address) 4.20-4.40. Ian Percival*, Barry Webby and Yongyi Zhen: Revised Ordovician biostratigraphy of New South Wales. 4.40-5.00. Theresa Winchester-Seeto: Chitinozoans and foraminiferal linings from the Ordovician of Australia: an overview. 5.00-5.30. Richard Fortey: Ordovician palaeogeography of the Shan-Thai terrane. (keynote address) 5.30-5.50. Barry Webby*, Ian Percival, Fens VandenBerg, John Pickett, John Pojeta Jr, Roger Schallreuter, Theresa Winchester-Seeto, Greg Edgecombe, Roger Cooper and Yongyi Zhen: Ordovician biogeography of Australasia. 5.50-6.10. Mohammad Ghavidel-Syooki: Biostratigraphy and palaeo-biogeography of the Late Ordovician-Early Silurian chitinozoans from the Zagros Basin, Southern Iran.

XVI

EVENING MEAL 7.30-8.30 pm ... MALCOLM WALTER. Search for life on Mars.. Theatre 20.1 8.30-9.30 pm AAP Meeting. Theatre 20.1

XVll

TUESDAY

December, 1997

SESSION 1 (8.30-11.00) Chairman: Al Fagerstrom, Corvallis 8.30-9.00. Barrie Rickards: (keynote address)

Eastern Australian Silurian graptolites in time and space,

9.00-9.20. John Pickett: Silurian corals of Australasian aspect 9.20-9.40. Jack Douglas"^ and Jackie Tims: Biogeography of Early-Middle Palaeozoic floras of southeastern Australia. 9.40-10.10. John Long: Palaeozoic fish faunas of Australia, (keynote address) 10.10-10.40. Gavin Young: (keynote address)

Lower vertebrates of East Gondwana - biogeographic patterns,

10.40-UOO. Malte Ebach. Phylogenetic and area cladistic analyses of a Devonian trilobite fauna from western New South Wales

MORNING TEA: 11.00-11.20 POSTERS Chris BENTLEY & Jim JAGO. Revision of the family Catillicephalidae Raymond 1938 (Trilobita). Nathalie NAGALCSIGUM, Andrew DRINNAN & Stephen McLOUGHLEV. New and revised fossil plants from the Lower Cretaceous Boola Boola flora, Gippsland Basin, Victoria. Troy MYERS. Comparative palaeoecology of Cainozoic Riversleigh faunas. John RIGBY. The Permian flora of the Mt Mulligan Coal Measures, Queensland, Australia. Andrew SIMPSON. Late Ordovician conodonts from Gray Creek and the head-waters of Stockyard Creek, north Queensland.

SESSION 2 (11.20-1.00) Chairman: Gary Webster, Pullman 11.20-11.40. Alex Cook: Australian Devonian gastropod biogeography. 11.40-12.00. Mehdi Yazdi: Late Devonian bioevents and biogeographic patterns in relation to sea level changes in the Shotori Ranges, East Iran. 12.00-12.20. John Talent, Jonathon Atchison, Margaret Bradshaw, Gary Dargan, Ruth Mawson, Theresa Winchester-Seeto, Gavin Young & Yong-Yi Zhen. Palaeobiogeography of Australasian Devonian biota. 12.20-12.40. Greg Webb: Problems for palaeobiogeographical analysis of Palaeozoic corals as illustrated by Lower Carboniferous corals of eastern Australia.

XVlll

12.40-1.00. Zhongqian Chen* and Guang Shi: Carboniferous brachiopods of China: a review of their distributions in space and time.

LUNCH: LOO-1-45 SESSION 3 (1.45-3.15) Chairman: Marion Bamford, Witwatersrand 1.45-1.55. Peter Jones*, Brian Engel, Ian Metcalfe, Geoff Playford, John Rigby, John Roberts, Sue Turner and Greg Webb: Carboniferous biogeography of Australasia. 1.55-2.15. A1 Fagerstrom* and Otto Weidlich: Building the Upper Capitan Reef framework (Permian), Guadalupe Mountains, New Mexico-Texas, U.S.A: inter-relations between paleoecology and abiotic processes. 2.15-2.35. Clinton Foster: Caheniasaccites Bose & Kar, 1966, Costatascyclus Felix & Burbidge, 1967 and allied genera: portrait of a migrating gymnosperm from North America, Gondwana, and Russia. 2.35-2.55. Neil Archbold: Late Palaeozoiclnsecta and Arachnida of Gondwana: a review. 2.55-3.15. Guang Shi: Terrane rafting enhanced by contemporaneous climatic amelioration as a mechanism of biogeographical vicariance: Permian marine biogeography of SE Asia.

AFTERNOON TEA: 3.15-3-40 POSTERS Ruth MAWSON & John TALENT. Eastern Molong Platform and adjacent Hill End Trough, New South Wales: Mid-Palaeozoic conodont data, age-inferences, and implications regarding platform exposure and carbonate sedimentation offshore. Andrew SIMPSON. Early Silurian conodonts from the Broken River region, north Queensland. YANG Weiping. The Early Permian Australasian microflora in West Yunnan and its palaeobiogeographic and tectonic significance.

SESSION 4 (3.40-6.00) Chairman: Eckart HMcansson, Copenhagen

3.40-4.00. Gary Webster^\ Peter Jell and Adam Derewetzky: Paleobiogeography of Permian echinoderms of Australia. 4.00-4.20. Shuzhong Shen^\ Neil Archbold and Guang Shi: brachiopod palaeobiogeography.

Changhsingian (Late Permian)

4.20-4.50. Neil Archbold: The palaeobiogeography of the Australasian Permian, address) 4.50-5.10. Anne Warren*, Ross Damiani, Caroline Northwood and Adam Yates: Palaeobiogeography of Australian fossil amphibians.

(keynote

XIX

5.10-5.30. Tim Hamley: Biogeography of the procolophonids. 5.30-6.00. Ian Metcalfe: Australian-Southeast Asian breakup in relation to biogeographic patterns in the region, (keynote address)

EVENING MEAL

EVENING ACTIVITIES Chaired by: Barry Webby (MUCEP, Sydney) and John Talent (MUCEP, Sydney) 7-00- 8.00 Discussions of IGCP programmes. Theatre 20.1, then Geology labs in building 41. 8.00-? Meeting of Pander Society meeting (Wollongong branch)

WEDNESDAY lO'^ December, 1997 9.00-10.00. This time has been set aside for discussions to be convened by the coordinators of the various chapters as required. Meetings will be held in the general vicinity of the large first floor foyer of building 41. Chairs can be borrowed (and replaced, please) from laboratory 41/153. Field trip to Minnamurra rainforest and Black point. Buses will depart from the large parking area near building 35. Lunch supplied at Minnamurra Park centre. Return to Northfields Avenue by 5 pm. Conference Dinner 7 pm for 7.30, Uni Centre Function Rooms.

THURSDAY December

1997

SESSION 1 (8.30-10.40) Chairman: Roger Cooper, Wellington 8.30-8.50. Stephen McLoughlin*, Andrew Drinnan and Sofie Lindstrom: Permian-Triassic floras of East Antarctica: palaeoenvironmental setting and biogeographic affinities. 8.50-9.10. Keith Holmes: From go to woe - the Triassic vegetation of eastern Australia. 9.10-9.30. George Stanley: Palaeogeography of Upper Triassic Tethyan reef faunas from the Northwestern Australian Shelf. 9.30-10.00. Jack Grant-Mackie* and Hamish Campbell: Australasian Triassic and Jurassic biogeography. (keynote address) 10.00-10.20. David Cantrill: The geographic distribution of the extant fern Lophosoria quadripinnata: implications for Cretaceous climates of the southern high latitudes. 10.20-10.40. Anne-Marie Tosohni*, Nathalie Nagalingum, Stephen McLoughlin and Andrew Drinnan: An Early Cretaceous flora from the Lower Strzelecki Group, Gippsland Basin, Victoria.

MORNING TEA: 10.40-11.00 POSTERS Mohammad GHAVroEL-SYOOKI. Upper Palaeozoic strata in the Zagros Basin, Southern Iran. Bob MORGAN. Crawney Limestone - a determination of conodont biostratigraphy using polygnathids. Alan RIX & Andrew SIMPSON. The fauna and flora of the Eocene Redbank Plains Formation: summary of recent progress.

SESSION 2 (11.00-12.50) Chairman: Jack Grant-Mackie, Auckland 11.00-11.20. Eckart Hackansson: Australasian lunulite bryozoans - the European connection? 11.20-11.40. Jeffrey Stilwell: Tectonic and palaeobiogeographic significance of the Chatham Islands, South Pacific, Late Cretaceous fauna. 11.40-12.00. A.S. Kassab: Variation, biostratigraphy and palaeobiogeography of the Upper Cretaceous coilopoceratid ammonites from Egypt 12.00-12.20. Bob Henderson*, James Crampton, Mary Dettmann, David Haig, Ralph Molnar, Samir Shafik, Jeff Stilwell and Tony Thulborn: Biogeographical observations on the Cretaceous biota of Australasia12.20-12.50. Brian McGowran* and Qianyu Li: Australasian Cainozoic biogeography: environmental framework for a southern-temperate province on a cooling planet, (keynote address)

XXI

LUNCH: 12.50-1.50 SESSION 3 (1.50-3.40) Chairman: Jean-Jacques Jaeger, Montpellier 1.50-2.20. Mike Archer: Messing with mammalian meanderings. (keynote address) 2.20-2.40. Brian McGowran*, Mike Archer, Tom Darragh, Qianyu Li, Phillip Maxwell, Ken McNamara, Mike McPhail, Alan Partridge, Joyce Richardson, Samir Shafik, Elizabeth Truswell and Mark Warne. Australasian Cainozoic Biogeography 2.40-3.00. Andrew Drinnan*, T. Carrick Chambers and Stephen McLoughlin: Morphology of the recently discovered Wollemi pine (Wollemia nobilis: Araucariaceae) and comparisons to Cretaceous and Tertiary fossils. 3.00-3.20. Sue Hand: New world order in bat biogeography. 3.20-3.40. Anthony Vadala* and Andrew Drinnan: A Late Paleocene leaf flora from Cambalong Creek, Southern Highlands, New South Wales.

AFTERNOON TEA: 3-40-400 SESSION 4 (4.00-6.00) Chairman: George Stanley, Missoula 4.00-4.20. Anne Musser: Examining relationships of the monotremes in a biogeographical context. 4.20-4.40. John Scanlon*, Mike Archer, Sue Hand and Hank Godthelp: Preliminary biostratigraphic analysis of snakes from the Tertiary of Riversleigh, Queensland. 4.40-5.00. Jean-Jacques Jaeger* and Yaowalak Chaimanee: The evolution of the Rattus group (Mammalia, Rodentia) during the Plio-Pleistocene in Thailand. 5.00-5.20. Marion Bamford: Australasian floral elements in southern Africa: Tertiary woods of the Myrtales. 5.20-5.40. Rob Craig: The Cenozoic brachiopods of Western Australia. 5.40-6.00. Peter Kershaw: The palaeobiology of the Quaternary - a partial overview. CONCLUDING COMMENTS AND REFRESHMENTS

SPECIFIC REEF BIOTAS OF THE URALS AND THEIR IMPORTANCE FOR THE MID-PALEOZOIC PALEOGEOGRAPHY Anna ANTOSHKINA Institute of Geology, Komi Science Centre, Uralian Division, Russian Academy of Science, 167610 Syktyvkar, RUSSIA Upper Silurian-Lower Devonian (Ludlow-Emsian) reefs are exposed along the western and eastern slopes of the Ural Mountains from Pay-Khoy to Mugodzhary more than 2000 km. They formed barrier belts both on a continental shelf margin (Western Urals) and on a shelf margin of a volcanic island arc (Eastern Urals). The largest Ludlow and Pragian barrier reefs (up to 500 m thick and up to 1000 m thick respectively) were formed in the Pechora Urals (Shuysky 1973; Antoshkina 1994). The Ludlow-Emsian reefs were developed in subtidal environments on the shelf margin and stromatolite-like microbial-metazoan associations were the common costructors of these reefs biofabrics. The Ludlow frame builders comprise stromatoporoids and corals encrusted by GirvanellaAikQ cyanobacteria and very specific Ikella-Fistulella (microbialproblematic hydroid) and stromatolite-aphrosalpingid associations. Aphosalpingid sponges occur as the nucleus for growth of concentric stromatolitic mats formed by Renalcis, Girvanella, Epiphyton, Ludlovia, Hecetaphyton and Sphaerina. These spongelike organisms are not an essenial part of reefs but they played the role of bafflers and binders in the consolidation of reefal sediments. The Pragian-Emsian reef framework is mainly composed of the Ikella-Fistulella association and diverse green algae; especially lanciculids and dasyclads are abundant. Diverse microbial assemblages, except for EpiphytonAikt forms, consist of Renalcis, Ikella, Girvanella, Rothpletzella, Garwoodia and Hedstroemia. Similar aphrosalpingid-bearing microbial-metazoan reef builders occur in the Ludlow buildups of the Salair (Zhuravleva & Miagkova 1981), in the Alexander terrane of southeastern Alaska (Soja 1991) and in the Pridoli ones of the southwestern and westcentral Alaska (Clough & Blodgett 1988). The microbial association of the Northern Urals Ludlow reef has an extraordinarily high degree of taxonomic similarity with one of coeval reefs in the Alexander terrane (Soja & Antoshkina 1997). The problematic hydroid Fistulella, characteristic of the Silurian-Lower Devonian reefs in the Urals and the Lower Devonian reefs in southwestern Siberia (Krasnow et al. 1986), has also been recognised in the framework of the reef mound complexes of southwestern Alaska (Clough & Blodgett 1988). Thus the reef microbial-metazoan (microbial-aphrosalpingid and microbidX-Fistulella) associations, being important contributors to reef growth, were limited in their geographic distribution in the Mid-Paleozoic. Because they occur only in the Urals, Salair, southwestern Siberia and southwest-central Alaska, a definitive paleobiogeographic link existed between these regions during Late Silurian-Lower Devonian. The reef bioconstructions in these regions are not identical in taxonomic composition but share common conspecific, reef-dwelling metazoans and congeneric microbiotas indicating that they evolved separately but in geographically contiguous regions. Migration of biotas.

provided by the Uralian seaway, was perhaps hmited by the low latitudes of the Northern Hemisphere equatorial zone. Antoshkina, A.L, 1994. Rifie v Paleozoe Pechorskogo Urala. Rossiyskaya. Akademiya Nauk, StPetersburg, 154 pp. Clough, J.G. & Blodgett, R.B., 1988. Silurian-Devonian algal reef complex of southwest Alaska. In Geldsetzer, H.H.J., James, N.P. & Tebbutt, G.E. (eds), Reefs Canada and adjacent area. Canadian Society of Petroleum Geologists Memoir 13, 404-407. Krasnov, V.L, Stepanov, S.A. & Ratanov, L.S., 1986. Rifovie systemy v srednem Paleozoe Sibiri. In Kalio, D.L. and Klaamann, B.P., (eds), Teoria i praktika ekostratigrafii. Valgus, Taallin, 237244. Shuysky, V.P., 1973. Nizhnedevonskie rifostroyashie vodorosli Urala. Akademiya Nauk, Moskow, 155 pp. Soja, C.M., 1991, Origin of Silurian reefs in the Alexander terrane of southeastern Alaska. Palaios 6, 111126. Soja, C.M. & Antoshkina, A.L, 1997, Coeval development of Silurian stromatohte reefs in Alaska and the Ural Mountains: Implications for paleogeography of the Alexander terrane. Geology 25, 539-542. Zhuravleva, I.T. & Myagkova, E.I., 1981, Materialy k izucheniyu Archaeta. Trudy Instituta Geologii i GeophisikiSOANSSSR^U, 1-223.

THE PALAEOBIOGEOGRAPHY OF THE AUSTRALASIAN PERMIAN N.W. ARCHBOLD School of Aquatic Science and Natural Resources Management, Deakin University, Rusden Campus, Clayton, VIC 3168 During the Permian, the present Australian continent was a component of the northeastern portion of Gondwana. The cold to warm-temperate climate change during the Permian and the combination of cold eastern marine surface currents and cool to warm western surface currents collectively explain many of the palaeobiogeographical features of Australian Permian faunas and floras. Marine faunas invariably show an increase in diversity from south (Tasmania) to north (Bowen Basin) and from east, with endemic genera, to the west with faunas and floras demonstrating stronger links with Tethyan, peripheral Gondwanan and Indian fossil sequences. Eastern Australian marine faunas, particularly those of Queensland show strong links with the scattered faunas of New Zealand (excluding the fusulinid bearing terranes). Permian marine faunas and terrestrial floras of Western Australia are well summarized in Skwarko (1993) and ongoing work continues on foraminiferans, molluscs (particularly bivalves and gastropods), brachiopods, echinoderms, conodonts and microfloras. References to eastern Australian Permian faunas and floras are numerous and diverse and the broad nature of their relationships are well understood although considerable systematic work remains to be undertaken. Brachiopod faunas of the Australian Permian (Archbold 1996) are reasonably typical of all fossil groups in terms of the palaeobiogeographical relationships that they demonstrate. Some fossil groups are totally or mostly restricted to western basins (eg. conodonts, most ammonoids and the elasmobranch Helicoprion) with connections to the Tethyan and Uralian seas. Some Eastem Australian fossil groups developed strong endemism, e.g. the brachiopod family Ingellarellidae, in adapting to cold water conditions. Isolated faunas and floras from the Permian of Irian Jaya demonstrate links to the Australian fossil record and to that of southeast Asia (notably the Shan Thai terrane, a continental fragment also with significant Permian faunal links to Western Australia, see Archbold & Shi 1995).

Archbold, N.W., 1996. Paleobiogeography of Australian Permian brachiopod faunas. In Copper, P. & Jin, J.-S. (eds), Brachiopods, Proceedings of the Third International Brachiopod Congress, Sudbury, Canada. 19-23. A.A. Balkema, Rotterdam. Archbold, N.W. & Shi, G.R., 1995. Permian brachiopod faunas of Western Australia: Gondwanan-Asian relationships and Permian climate. Journal of Southeast Asian Earth Sciences 11, 207-215. Skwarko, S.K., (ed.), 1993. Palaeontology of the Permian of Western Australia Bulletin of the Geological Survey of Western Australia 136, 1-417, 102 plates.

LATE PALAEOZOIC INSECTA AND ARACHNIDA OF GONDWANA: A REVIEW N.W. ARCHBOLD School of Aquatic Science and Natural Resources Management, Deakin University, Rusden Campus, Clayton, VIC 3168 Late Palaeozoic Insecta and Arachnida are potentially powerful biostratigraphical tools for improving correlations between terrestrial coal measure sequences across Gondwana, judging from northern hemisphere experience (e.g. Rodendorf et al 1961; Durden 1969). Nevertheless these groups have received relatively minor attention within the Gondwanan continents. The oldest Gondwanan insects are from the Namurian of South America (Argentina) and include well-preserved members of the Protodonata and Megasecoptera. Late Carboniferous (Stephanian) members of the Palaeodictyoptera, Protorthoptera and the Paraplecoptera as well as the Blattodea (cockroaches) are also known from Argentina. Blattodea and Protorthoptera are also documented from Sao Paulo, Brazil. The only known Late Palaeozoic arachnids from Gondwana (Gondwanarachne and the remarkable giant Megarachne) are also from the Late Carboniferous (Stephanian) of Argentina. Permian insects are known from a scatter of horizons across the Gondwanan continents. African records include an earliest Permian protorthopterid from the Dwyka Series of Zimbabwe and poorly dated early Permian? (associated with Glossopteris) Protorthoptera and Mecoptera from Madagascar. Collembola, Blattodea, Protorthoptera, Plecoptera (stoneflies), Hemiptera (bugs) and Coleoptera (beetles) have been described from the Middle Ecca near Pretoria. Late Permian records include Protorthoptera (Democratic Republic of the Congo) and Megasecoptera, Protorthoptera, Blattodea, Orthoptera (grasshoppers etc.), Miomoptera, Protolytroptera, Hemiptera, Mecoptera (scorpionflies) and Neuroptera (lacewings) from South Africa. The Permian record from Antarctica is meagre with only Hemiptera and nymphs of possible Plecoptera or Protorthoptera. The South American Permian includes Mecoptera, Neuroptera, Hemiptera, Blattodea, Coleoptera and Odonata (dragonflies) with Odonata also known from the Falkland Islands. The Early Permian (Sakmarian) of India has yielded Blattodea and Protodonata while the later Early Permian has yielded Blattodea, Hemiptera, Mecoptera and Coleoptera. The Australian Permian includes an earliest Permian palaeodictyopteran from the Wynyard Tillite, Tasmania. A moderately diverse Djhulfian insect fauna including representatives of the Odonata, Plecoptera, Megaloptera, Psocoptera (psocids), Protolytroptera, Hemiptera, Mecoptera, Neuroptera, Glossolytrodea, Trichoptera (caddisflies), Coleoptera, and ancestral Diptera (2-winged flies) is known from the Awaba Tuff Member of the Eleebana Formation, northern Sydney Basin. Permian insect faunas of Gondwana possess a mixture of representatives of archaic orders, many of which became extinct at the end of the Permian, the only major extinction event affecting insects (Labandeira & Sepkoski 1993), and orders of insects still extant. The potential of insect fossils for yielding palaeoecological data is significant (Shear & Kukalova-Peck 1990).

Durden, CJ., 1969. Pennsylvanian correlation using blattoid insects. Canadian Journal of Earth Sciences 6, 1159-1177. Labandeira, C.C. & Sepkoski, J.J. f]., 1993. Insect diversity in the fossil record. Science 261, 310-315. Rodendorf, B.B., Bekker-Migdisova, E.E., Martynova, O.M. & Sharov, A.G., 1961. Paleozoiskie nasekomye Kuznetskogo Basseina. Trudy Paleontologicheskogo Instituta 85, 1-707, 40 pis. Shear, W.A. & Kukalova-Peck, J., 1990. The ecology of Paleozoic terrestrial arthropods: the fossil evidence. Canadian Journal of Zoology 68,1807-1834.

MESSING WITH MAMMALIAN MEANDERINGS M. ARCHER School of Biological Science, University of New South Wales

Prior to the 1980s, simplistic 'just-so stories' developed to explain modem distribution patterns of Australia's mammals were possible because of an almost total lack of information about pre-Quatemary distributions. Recent discoveries in Australian Cretaceous and Eocene, Oligocene, Miocene and Pliocene deposits have left very few of the older concepts on their feet. Monotreme and microbiotheriid distribution patterns have, for example, gone from continental to intercontinental, and marsupials from intercontinental to global, both examples involving epic dispersals followed by vicariance events. The question of early terrestrial placentals in Australia is blowing up into a fine controversy that becomes ever more complicated. Presumptions about the areas of origin and consequent dispersals of many groups, such as pseudocheirids, phalangerids and petaurids, have been tipped on their heads. Several mammalian groups thought never to have set toe upon this southern land, such as mystacinids and dendromurines, have suddenly checked in with important palaeobiogeographic consequences. Palaeoecological origins for other groups, such as notoryctids and koalas, have proved equally unexpected and in some cases led to predictions about potential unrecognised components to distribution patterns for living animals. While most groups of Australian mammals have begun to give up some of their palaeobiogeographic secrets, pteropodids, numbats and tarsipedids remain a total mystery, with no Tertiary fossil records of any kind. Analysis of Tertiary communities as a whole, which in some cases have 44% more family-level diversity than living descendant communities, enables recognition of probably serendipitous (inappropriate) gaps in modem distribution pattems. Documenting wider geographic and palaeoecological pre-Pleistocene distribution pattems for modem mammals encourages conservationists to consider the value of artificially dispersing threatened taxa into extralimital but formerly occupied habitats, particularly as modem ranges undergo massive collapse during the coming century. Palaeobiodiversity and palaeobiogeography provide new insights that should be added to modem understanding to reconsider current conservation priorities among threatened living mammals. While in many cases understanding about palaeobiogeography supports current wisdom, in others it most certainly does not. The most recent intemational controversy about the palaeobiogeography of Australian mammals involves a claimed early Cretaceous placental, Ausktribosphenos nyctos, from Victoria (Rich et al, 1997), discovery of which would indeed, as its original describers claim, up-end most current views about tribosphenid evolution. As will be shown, reexamination of this very interesting specimen indicates that, to the contrary, it is not a placental mammal but rather a structural, temporal and geographic intermediate between the world's most derived peramurids (e.g., European Jurassic Peramus and South American early Cretaceous Vincelestes) and the most plesiomorphic monotremes (e.g., the early Cretaceous Steropodon). As such, it is the oldest monotreme known and very informative about the phylogenetic and geographic origins of this distinctive group of Gondwanan mammals.

LATE LATE CAMBRIAN TRILOBITES FROM THE BIRCH INLET AREA, SOUTHWESTERN TASMANIA Jin-song BAG & Jim JAGG Department of Applied Geology, School of Engineering, University of South Australia, The Levels, SA 5095. Richly fossiliferous late Late Cambrian rocks occur within a fault slice about 8.5km west of Birch Inlet, southwestern Tasmania (Jago 1972, Jago & Brown 1989, Bao 1995, Laurie et al 1995). About 760m of Cambrian sediments outcrop in this area. The bottom 425m is poorly known, but consists of siltstones, minor tuffs and greywacke conglomerate overlain by purple and grey-green siltstone and fine breccia. This is overlain by 30m of fine green sandstone and siltstone. The top 305m of the succession comprises fine sandstone and siltstone which is micaceous in places. Much of this succession is fossiliferous, with the abundance and variety of fossils increasing up section. The top 60m of this succession is mainly unfossiliferous siltstone; within the top 10-12m there are some beds of quartz sandstone and conglomerate up to 60m thick. This succession is overlain by the Ordovician Gordon Group. The Birch Inlet fauna contains the following trilobites: Lotagnostus trisectus (Salter), a new genus of agnostoids, Micragnostus cf. serus (Fortey), Micragnostus sp., Neoagnostus sp., other agnostoids, Charchaqia halli Jago, C curvata Troedsson, Parabolinella triarthra (Callaway), Proteuloma huochengensis (Zhang), Macropyge cf. carinata Lu, Onchonotellus sp., Cemuolimbus sp., Macropyge sp., plus new species of the following genera, Skljarella, Niobella, Proceratopyge, Olenus, and Hedinaspis, a new genus of the Richardsonellidae and an indeterminate member of the Remopleuridacea. The most abundant taxa are C. halli, Skljarella sp.nov., and the new agnostoid genus. This fauna is of late Late Cambrian age (Payntonian to earliest Datsonian) on the north Australian biostratigraphic scale. It is an offshore fauna which is similar to those of the Southeast China Faunal Province. This is in contrast to the late Late Cambrian faunas from Black Mountain in Queensland (the standard section for north Australia) which are dominated by shallow water faunas similar to those of the North China Faunal Province. Correlation from the Birch Inlet faunas to the late Late Cambrian shallow water faunas is made via the Shenjiawan fauna of Hunan, which contains a mixture of deep and shallow water forms.

Bao, Jin-song, 1995. Some Middle and Late Cambrian trilobites from western and northwestern Tasmania. M.Sc. thesis. University of South Australia (unpublished). Jago, J.B., 1972. The youngest recorded Tasmanian Cambrian trilobites. Search 3, 173-174. Jago, J.B. & Brown, A.V., 1989. Middle to Upper Cambrian Fossiliferous sedimentary rocks. Geological Society of Australia, Special Publication 15, 74-83. Laurie, J.R., Jago, J.B. & Bao Jin-song, 1995. Review of Tasmanian Cambrian biostratigraphy. AGSO Record, 1995/69, 1-31.

AUSTRALASIAN FLORAL ELEMENTS IN SOUTHERN AFRICA: TERTIARY WOODS OF T H E MYRTALES MARION K. BAMFORD Bernard Price Institute for Palaeontology, University of the Witwatersrand, Private Bag 3, P.O. WITS 2050, SOUTH AFRICA In the last few years fossil woods have been collected from Tertiary deposits in southern Africa. One site, Bogenfels, in Namibia, has a lot of Myrtaceous woods, with characteristics of Callistemonoxylon and Syzygioxylon. They are probably Eocene in age. Today Syzygium occurs in southern Africa, but not Callistemon. Other fossil woods are those of the Combretaceae, both of the Terminalia and Comb return types. These genera are well represented throughout Africa today but some are very climate specific. The fossil and modem taxa and distribution are compared for the African representatives, and also briefly compared with the Australasian taxa. There are large gaps in the Tertiary wood record for southem Africa at this stage but an attempt is made to reconstruct the distribution of the Myrtales.

SOME COMMENTS ON AGNOSTOID TAXONOMY AND CLASSIFICATION Christopher BENTLEY & Jim JAGG Department of Applied Geology, School of Engineering, University of South Australia, The Levels, SA 5095.

Agnostoids were small, isopygous arthropods which flourished from the late Early Cambrian to the end of the Late Cambrian and were extinct by the end of the Ordovician. They are extremely useful zone fossils for most of the Middle and Late Cambrian; for example, if the time scale of Young & Laurie (1996) is accepted, the average time span for trilobite zones, mainly agnostoid-based, in the Australian Middle and Late Cambrian is about 520,000 years. In agnostoid taxonomy and classification much emphasis has been placed on small differences which are more or less visible depending on the preservation of the specimens, e.g. Opik (1967) noted that "some 130 single characters have been used in describing the many species of Pseudagnostus'\ The authors support the suggestion of Pratt (1992) that the deuterolobe, highly regarded by previous workers as a taxonomic tool in Pseudagnostus and related genera, is in fact a tme axial expansion and that the axial furrows are homologous in all agnostoids. The first modern comprehensive classification of agnostoids was that of Opik (1967) who used seven criteria, several of which are unworkable. After assessing 179 generic and subgeneric taxa, Shergold et al (1990) reduced this, in a detailed discussion, to 119 valid genera and subgenera. It is suggested that even 119 is an overestimate with intrageneric and intraspecific variation now being taken more into account (e.g. by Westrop et al 1996). The classification of agnostoids should be based on axial features because they are directly related to the function performed by the part of the anatomy they reflect and are likely to be more conservative and expressive of relationships than marginal features such as deliquiate or nondeliquiate border furrows and marginal spines. Such features are more useful at lower taxonomic levels. Many agnostoid taxa show a large degree of intraspecific and interspecific variation, including the presence, absence and relative strength of preglabellar median furrows, degree of effacement of furrows, presence or absence of scrobiculation and the strengths of posterolateral spines. Such characters should not be used in higher level taxonomy if they are variable at the specific and generic levels. Axial features as used here means the characters of the glabella and of the pygidial axis. An analysis has been carried out using 39 agnostoid taxa. These included the nominate genera of the subfamilies recognized by Shergold et al. (1990) plus others chosen from the same groups, and newer taxa, such as Nahannagnostus Pratt, 1992. The purposes of the analysis were to test the higher level classification of Shergold et al (1990) and also to attempt to find characters which are important in higher level agnostoid taxonomy. Using a combination of 26 taxonomic characters and two "time" characters, the 37 agnostoid genera under consideration were divided into two major groups, with five families in one group and four in another, a total of nine families rather than six (Shergold et al 1990) and three (Opik 1967) for the same genera.

Opik, A.A., 1967. The Mindyallan Fauna of North-Western Queensland. Bulletin of the Bureau of Mineral Resources, Geology and Geophysics, Australia 74, (1) 404 pp., (2) 167 pp. + 67 pi. Pratt, B.R., 1992. Trilobites of the Marjuman and Steptoean stages (Upper Cambrian), Rabbitkettle Formation, southern Mackenzie Mountains, northwest Canada. Palaeontographica Canadiana 9, 179 pp. Shergold, J.H., Laurie, J.R. & Sun X.-W., 1990. Classification and review of the trilobite order Agnostida Salter, 1864: an Australian perspective. Report of the Bureau of Mineral Resources, Geology and Geophysics, Australia 296, 1-93. Westrop, S.R., Ludvigsen, R. & Kindle, C.H., 1996. Marjuman (Cambrian) agnostoid trilobites of the Cow Head Group, western Newfoundland. Journal of Paleontology 70, 804-829. Young, G.C. & Laurie, J.R., (eds), 1996. An Australian Phanerozoic Timescale. Oxford University Press, Melbourne, 279 pp.

11 PALAEOBIOGEOGRAPHIC TRENDS OF CAMBRIAN FAUNAS FROM AUSTRALIA

Glenn A. BR0CK\ Michael J, ENGELBRETSEN\ Jim JAGO^ John LAURIE^ & James E. SORAUF"^ ^Centre for Ecostratigraphy and Palaeobiology, School of Earth Sciences, Macquarie University, NSW 2109 ^Department of Applied Geology, School of Engineering, University of South Australia, The Levels, SA 5095 ^Australian Geological Survey Organisation, G.P.O- Box 378, Canberra ACT 2601 ^State University of New York at Binghamton, Binghamton, New York 13902-6000, U.S.A. The lack of formal stage subdivisions for the Early Cambrian of Australia continues to thwart accurate global biostratigraphic correlation, and has thus seriously impeded attempts to undertake meaningful palaeobiogeographic analysis of Early Cambrian faunas. Despite this, it is still possible to gain some insight into palaeobiogeographic trends of Early Cambrian fossils from Australia by comparing taxa of broadly similar age using local or regional biostratigraphic schemes. Early Cambrian taxa from Australia were almost entirely confined to the tropical, shallow water, north-south trending epicontinental seaway responsible for deposition of thick carbonate sequences within the Adelaide Geosyncline, Officer Basin and the Gnalta Shelf, western N.S.W. Atdabanian polymeroids include dolerolenids, protolenids, ichangiids, emuellids and redlichiids that have close affinities to faunas in Antarctica, Morocco, Spain, France, China and Siberia. Eodiscid genera recorded from Australia (except Serrodiscus and Opsidiscus) are also found in Laurentia, Siberia, Southem Europe, South China and Antarctica. This distributional pattern indicates that eodiscids were pelagic or had a teleplanic larval stage, and were less constrained by oceanic barriers than benthic trilobites. Water temperature may also have affected the distribution of some genera. The composition and distribution of the Australasian archaeocyathan fauna reflects Australia's position adjacent to, and united with, Antarctica within Cambrian East Gondwana. The archaeocyaths that flourished during the mid-late Early Cambrian are now preserved in the Amadeus and Georgina Basins (Northern Territory), Arrowie, Stansbury and Officer Basins (South Australia), Gnalta Shelf (New South Wales) and Ross Fold Belt and outliers in Antarctica. Of the 240 valid archaeocyath species (now placed in 96 genera) from Australia and Antarctica, 26 are common to both continents, supporting the concept of a unified Australian-Antarctic province. Altogether, the Australian-Antarctic province includes archaeocyathan faunas equivalent in age to the Atdabanian, Botomian and Toyonian stages of the Siberian Early Cambrian, as well as unique Antarctic examples from the Middle and Late Cambrian. Additionally, six species (in five genera) of radiocyaths have been named, together spanning the same age interval. Tommotian-equivalent strata of the province are mainly siliciclastic rocks deposited in environments which were unconducive to archaeocyathan or radiocyathan development. Early Cambrian Ungulate brachiopod genera from Australia include Askepasma, "Lingulella'\ Acrothele, Paterina, Botsfordia, Dictyonina, Neobolus, Kyrshabaktella,

Karathele, Eoobolus, Vandalotreta and Edreja, Askepasma is restricted to Australia, with Dictyonina and Edreja common to both Australia and Kazakhstan, Neobolus and Eoobolus are common to Australia and Pakistan. Vandalotreta, Eoobolus and Karathele provide strong links with Antarctica. The distribution of Early and Middle Cambrian corals, including the Tabulaconida {Mooripora, Arrowipora and the skeletal sponge Flindersipora) and the Cothonida (Cothonion and Lipopora), is restricted to the Adelaide Geosyncline. The taxonomic instability amongst Early Cambrian small shelly fossil faunas hinders accurate biogeographic comparisons, but the widespread (in some cases restricted) distribution of coeloscleritophorans (chancelloriids and halkeriids), cambroclaves, tommotiids (e.g. Lapworthella, Dailyatia), hyolithelminthes, hyoliths, palaeoscolecidans and molluscs across much of Gondwana suggests these groups have the potential to provide useful palaeobiogeographic information. Early Middle Cambrian, Ordian/Early Templetonian, polymeroid faunas include redlichiids, xystridurids and oryctocephalids. Late Templetonian/Floran faunas are more diverse, with the presence of xystridurids, ptychoparids, dolichometopids and oryctocephalids displaying affinities with Antarctica, Siberia and North China. The presence of dolichometopids, such as Fuchouia and Amphoton, damesellids, mapanids and nepeids in the upper part of the Middle Cambrian (Undillan and Boomerangian) indicates affinities with eastern Asia and Antarctica, although some connection with North America is evident. Middle Cambrian agnostid genera whose members are used in the various biostratigraphic schemes are cosmopolitan. These include Ptychagnostus, Acidusus, Lejopyge, Triplagnostus (Triplagnostus), Pentagnostiis, Hypagnostus and Doryagnostus, Others, including Goniagnostus (Goniagnostus), G. (Allobodochus), Tomagnostus, Diplagnostus and Diplorrhina, have a similar distribution, except for their absence from western and northwestern North America, which may be a result of oceanic circulation patterns. Analysis of the Middle-Late Cambrian distribution of some key articulate brachiopods (e.g. Nisusia, Arctohedra, Yorkia, Wimanella, Diaphora, Cambrotrophia and a range of newly erected genera), molluscs (e.g. Latouchella, Yochelcionella, Protowenella and Anarbarella) Ungulates (e.g. Treptotreta, Amictocracens, Stilpnotreta, Anobalotreta, Neotreta, Micromitra and Kyrshabaktella), conodonts, and small shelly fossils (e.g. Tretocylichne) re-inforces the strong faunal links with New Zealand, Antarctica, eastern Laurentia, Baltica, Siberia and Kazakhstan and indicates relative proximity of the carbonate platforms from these regions. There is a major faunal reorganisation at the start of the Idamean, both in Australia and elsewhere. Idamean polymeroid faunas become more diverse with time. Throughout the Idamean there is a mixture of cosmopolitan genera and other genera with strong affinities to one or more of Antarctica, Siberia, East Asia, Kazakhstan and North America. The Iverian shallow water faunas include a mixture of cosmopolitan genera plus others with affinities to East Asia, Kazakhstan and Antarctica, and some connection with North America, Spain and Turkey. The base of the Payntonian is marked by the incoming of a fauna with almost totally East Asian affinities with species of tsinaniids, saukiids, shumardiids and leiostegiids prominent. Throughout the Middle and Late Cambrian, deeper water Tasmanian faunas show a more cosmopolitan aspect than do the shallow water faunas of central and northern Australia. The relatively large proportion of cosmopolitan genera indicates that agnostids were pelagic, an interpretation supported by

their enrolled life habit. The restriction of many genera to the tropics in both the Middle and Late Cambrian indicates that water temperature may also have affected their distribution. Late Cambrian brachiopod faunas from cratonic Australia and accreted terranes include cosmopolitan taxa such as Billingsella, Eoorthis, Micromitra, ''Lingulella" and Acrothele, as well as numerous taxa with more restricted distributions such as Kozhchinella, known only from Queensland and the Kuznetsk-Alatau of Russia. Late Cambrian lingulate taxa from Australia include Angulotreta, Physotreta, Eurytreta, PalaeoboluSy Dactylotreta, Treptotreta, Anabolotreta, Orbithele, Neotreta, Quadrisonia, Picnotreta, Stilpnotreta and Schizambon. The lingulates show strongest affinities with the United States, Antarctica and Kazakhstan, each having seven genera in common with Australia. Angulotreta, Schizambon, Quadrisonia, Orbithele, Anabolotreta, Dactylotreta and Physotreta are also known from the United States with Dactylotreta, Quadrisonia, Schizambon, Angulotreta, Treptotreta, Picnotreta and Stilpnotreta present in Antarctica. Kazakhstan has Dactylotreta, Quadrisonia, Angulotreta, Treptotreta, Neotreta, Eurytreta and Anabolotreta in common with Australia. The Siberian Platform has very strong links with the Australian faunas at this time as the two regions have six genera, Angulotreta, Anabolotreta, Dactylotreta, Neotreta, Schizambon and Stilpnotreta, in common.

THE GEOGRAPHIC DISTRIBUTION OF THE EXTANT FERN Lophosoria quadripinnata: IMPLICATIONS FOR CRETACEOUS CLIMATES OF THE SOUTHERN HIGH LATITUDES. David J. CANTRILL British Antarctic Survey, Natural Environment Research Council, Madingley Road, High Cross, Cambridge CB3 OET, U.K. Fertile fern foliage from Lower Cretaceous (Aptian) strata on President Head, Snow Island (Antarctic Peninsula) contains the distinctive spore Cyatheacidites annulatus. The foliage is morphologically similar to the Lower Cretaceous form-genera Gleichenites and Microphyllopteris, whose time and space distribution in Gondwana matches that of the dispersed spore Cyatheacidites. Some of the Cretaceous Southern Hemisphere material assigned to Gleichenites and Microphyllopteris is probably allied to the Lophosoriaceae rather than the Gleicheniaceae as had been previously supposed. The nearest living relative, Lophosoria quadripinnata, is today confined to South and Central America. It grows within a mean annual temperature range of 8-22° C, and a mean annual precipitation range of 195-1977 mm. The presence of Lophosoria cupulatus at palaeolatitudes of 55-65° S implies that during the Aptian the southern high latitudes were a minimum of 12° C warmer than the present day.

CARBONIFEROUS BRACHIOPODS OF CHINA: A REVIEW OF THEIR DISTRIBUTIONS IN SPACE AND TIME Z. Q. CHEN & G. R. SHI School of Aquatic Science and Natural Resources Management, DeakinUniversity, Rusden Campus, 662 Blackburn Rd., Clayton, VIC 3168 The spatio-temporal distributions of the Carboniferous brachiopod faunas in China are reviewed, resulting in the recognition of 3 provinces (Xizang -West Yunnan, Tethyan and Tianshan-Hinggan) and 8 subprovinces. Stratigraphical assemblages of brachiopods are outlined and tabulated for each subprovince, with brief discussions provided wherever appropriate on the comparison and correlations of these assemblages. Brachiopod successions across the Devonian/Carboniferous, ToumasianA^isean, Lower/Upper Carboniferous, and the Carboniferous/Permian boundaries are discussed with a view to identifying some key brachiopod taxa for defining the respective boundaries and Carboniferous stages.

AUSTRALIAN DEVONIAN GASTROPOD BIOGEOGRAPHY ALEX COOK Queensland Museum, P.O. Box 3300, South Brisbane, QLD 4101 Although other elements of the Australian Devonian faunas have been studied in detail, gastropods have received comparatively little attention. Early Devonian faunas in New South Wales, Victoria and Queensland are partly known, and recent work on Middle Devonian faunas in north Queensland has filled a large gap in the taxonomic dataset. Australian Late Devonian gastropods have not yet been described. Examination of the biogeography is limited by the paucity of work on Asian faunas. Australian Early and Middle Devonian gastropod faunas are sufficiently endemic to constitute a distinct subprovince within the Old World Realm. Early Devonian platyceratid-crinoid interactions, so common in Old World and North Americas realm faunas, are limited or undeveloped in Australian faunas. Givetian faunas are dominated by a plexus of nodose, lirate gyronematines which, whilst clearly related to Old World Realm forms, are characteristic of a significant radiation in the eastern Gondwanaland subprovince. Platyceratids s.s. are a very minor faunal element during this time. Frasnian faunas, known cursorily from the Canning and Bonaparte Basins, are cosmopolitan, showing relationships to faunas from the Iberger Kalk, Germany, Naples faunas. New York and the Voronje Beds of the Urals.

ORDOVICIAN GRAPTOLITE SPECIES LONGEVITY, FAUNAL TURNOVER AND BIOCHRONOLOGY R.A. COOPER Institute of Geological and Nuclear Sciences, P.O. Box 30368, Lower Hutt, NEW ZEALAND The Ordovician graptolite sequence of Australasia, based on the combined sequences of Victoria and New Zealand, is one of the richest and most finely zoned in the world. Thirty-two zones and subzones are recognised. The Australasian zonal succession has proved widely applicable; it can be directly applied in Newfoundland, and the western USA and Canada, and is closely similar to schemes used in south China, cordilleran South America and Siberia. The sequence can therefore be taken as representative of a large portion of the Ordovician globe. The recent census and review of the zonal distribution of species and subspecies therefore provides a high resolution data set for analysis of macroevolutionary rates at the species level in a planktic group. Two hundred and ninety species level taxa are used in the present study. The duration of the zones has been obtained by using the graptolite composite standard sequence in pelagic shales (for the Early Ordovician) and the conodont composite standard sequence (for the Late Ordovician); these were calibrated and adjusted to fit zircon-dated levels (12 horizons). The resulting time scale has been used to measure macroevolutionary rates, including species diversity, originations, extinctions, faunal turnover, and species longevity. Faunal turnover is uneven in rate but is high across most zone boundaries, with peaks at the La3/Bel boundary, where there is a major increase in diversity, and the Da3/Da4 boundary, where there is a sharp decline in diversity. The changes in rate of faunal turnover and diversity are not obviously related to sea level change and their cause is uncertain. Species longevity averages 3.917 Ma and ranges from 8.25 Ma in the Dicranograptidae, to 1.53 Ma in the Sigmagraptidae. Interestingly, mean species duration for species which occupy the shallow water (epipelagic) depth zone (43 species) is 5.37 Ma, whereas species which occupy the deep water depth zone (57 species) average less than half of this at 2.19 Ma. The deep water zone (150m - ? 1000m) was a high-risk high-reward one, nitrogen rich and oxygen poor with abundant food resources but dangerously unstable. The shallow water zone (the surface turbulent waters of the ocean) was more stable over geological time. The high diversity and short life span of species in the deep water biotope may represent a 'get rich quick but risk early death' gamble by the Ordovician plankton.

THE CENOZOIC BRACHIOPODS OF WESTERN AUSTRALIA Robert Samuel CRAIG School of Applied Geology, Curtin University of Technology, G.RO. Box U 1987, Perth, WA 6001. To date only five Cenozoic brachiopods have been described from Western Australia. These include one rhynchonellid, Tegulorhynchia boongeroodensis McNamara, 1983, two craniaceans, Westralicrania allani Cockbain, 1967, and W. zenobiae Craig, 1997, and tv^o terebratulids, recta Richardson, 1991 diud Adnatida gnangarensis Richardson, 1991. In recent years collecting has yielded Cenozoic brachiopods from four sedimentary basins: the Carnarvon Basin, the Perth Basin, the Bremer Basin and the Eucla Basin. In the Carnarvon Basin brachiopods have been found in the Cape Range and the Giralia Range. Here, brachiopods occur in the Cardabia Formation comprising the Mid to Late Paleocene Boongerooda Greensand, the Late Paleocene Wadera Calcarenite and Pirie Calcarenite, the Late Paleocene to Early Eocene Cashin Calcarenite and the Middle Eocene Jubilee Calcarenite. Above this occurs the Middle to Late Eocene Giralia Calcarenite. Brachiopods found here include W. allani, T, boongeroodensis and several terebratulids, including several species of Rhyzothyris, otherwise known from the Late Eocene to Pliocene of New Zealand and Antarctic islands. The Late Oligocene to Early Miocene Cape Range group has yielded two species of terebratulid brachiopod. The Late Eocene Werrilup Formation in the Bremer Basin contains a new species of Craniacea, Westralicrania zenobiae Craig, 1997 and 14 species of terebratulid, including Gryphus labiatus, Stethothyris pectoralis, Aldingia furculifera, Victorithyris divaricata and a Terebratulina species. In the southern Eucla Basin brachiopods occur in the Late Eocene Toolinna Limestone, the Middle-Late Eocene Wilson Bluff Limestone, the Late Oligocene to Middle Miocene Abrakurrie Limestone and the Late Pliocene Roe Calcarenite. Species from the three older formations include the craniacean W. zenobiae and more than twenty species of terebratulids including: Aliquantula insolita, Diedrothyris johnstoniana, Magadina lunata, Terebratula cyclica and Victorithyris garabaldiana. These, like those found in the Werrilup Formation, are consistent with species found in the Tertiary of south-eastern Australia, although the westem Australian specimens tend to show an earlier appearance. Only one brachiopod is known from the Late Pliocene Roe Calcarenite, a new species of Neothyris, This genus had been considered, until now, to have been endemic to New Zealand. Brachiopods from the Perth Basin are of Late Pliocene to Early Pleistocene age. Two of the Late Pliocene brachiopods, Anakinetica recta and Adnatida gnangarensis, have been described by Richardson. Another 11 species remain to be described.

In total, some 50 species of Tertiary brachiopod have been identified from Western Australia. These are now under investigation to determine their taxonomic and evolutionary position, as well as their palaeogeographic and palaeoclimatic significance.

20 BIOGEOGRAPHY OF EARLY-MIDDLE PALAEOZOIC FLORAS OF SOUTHEASTERN AUSTRALIA

J.G. DOUGLAS^ & J.D. TIMS^ ^Grieve St., Warrnambool, VIC 3281 ^Beaumaris, VIC 3193 Poorly preserved and scattered structures possibly representing algal remains have been recorded from beds of Cambrian and Ordovician age in Victoria. Putative algal remains including Buthrotrephis sp. have been described from beds of Late Silurian and Early Devonian age in the Melbourne Trough. However, no fossils which could be considered land plant remains have been found in beds older than Late Silurian. The appearance of the highly organized Baragwanathia longifolia, other described species and associated (but as yet unpublished) plants in beds of Ludlow (Late Silurian) age therefore represents a remarkable episode in the history of plant development. The plant assemblage is one of the oldest vascularized land floras and certainly the most complex and diverse Silurian flora known. Probably 30 plant species are present in the Late Silurian and Early Devonian beds and with other co-fossilized plant and animal remains, help elucidate the palaeogeography of the time and more specific concepts such as shoreline position. Fourteen species have been described. Other remains which are attributable to Subdivisions of plants but are too incomplete to ascribe to species, indicate at least as many species again. One of us (JDT) believes the flora grew in a fresh water marsh environment associated with a major river system in which periodic flooding carried vast amounts of plant debris into the Melbourne Trough. Co-fossilized with the flora are graptolites, cephalopods and other marine animals which suggests to the other (JGD), onshore strandline accumulation. The distribution of Baragwanathia is examined in an attempt at reconciliation with shoreline reconstructions based on sedimentary facies and other criteria. Geological aspects of the key Melbourne Trough localities are outlined and compared with those of other successions in southeastern Australia, notably the Mullamuddy Formation of New South Wales and the Mathinna Beds of Tasmania. The effect of the "newly introduced" vegetative cover on the hitherto uninterupted movement of water on the earth's surface is presented as having momentous impact on terrestrial sedimentation. Lastly a review of the range in time of Baragwanathia and associated plants is aimed at providing an insight into the nature and palaeogeographic potential of this pioneer colonizer of the earth.

MORPHOLOGY OF THE RECENTLY DISCOVERED WOLLEMI PINE {Wollemia nobilis: ARAUCARIACEAE) AND COMPARISONS TO CRETACEOUS AND TERTIARY FOSSILS Andrew N. DRINNANt, T. Carrick CHAMBERS* & Stephen McLOUGHLINt t School of Botany, University of Melbourne, Parkville, VIC 3052. * Royal Botanic Gardens, Mrs Macquaries Road, Sydney, NSW 2000.

The extant araucariacean conifer Wollemia nobilis (Wollemi Pine) was discovered in 1994 and occurs as two very small populations growing in isolated sandstone gorges around 200 km northwest of Sydney. Macro- and micro-morphological details of Wollemia nobilis are described and illustrated and compared to the extant genera Agathis and Araucaria and with selected araucarian fossils from the Mesozoic of Australia. Adult and juvenile shoots of Wollemia differ in leaf arrangement, leaf shape, and cuticular features; in these features they are most similar to Araucaria, The cone scales have a long, distal spine reminiscent of Araucaria section Eutacta, but the winged seeds that are ontogenetically free from, and shed independently from, the cone scale are similar to Agathis. Shoots with variable leaf types, Araucaria-\\kt cone scales, and Agathis-Wkt winged seeds, are found in several plant fossil assemblages from the Cretaceous of Australia; these fossil conifers, which had been recognized as araucarian, can now be favourably compared with Wollemia. Pollen of Wollemia is indistinguishable from the fossil palynogenus Dilwynites, which has a fossil record extending back to the Late Cretaceous in Australia and New Zealand. Re-examination of the Mesozoic and Tertiary palaeofloras will most probably reveal an important contribution of Wollemia to the fossil record of Araucariaceae.

A LATE ORDOVICIAN BIOSTROME FAUNA FROM HAILES KNOB, NEW ZEALAND MICHAEL K. EAGLE Auckland Museum, Private Bag 92018, Auckland, NEW ZEALAND A late Ordovician biostrome faunal assemblage is described from Arthur Marble, Hailes Knob, Upper Takaka, north-west Nelson, South Island, New Zealand. The fossil assemblage is located within part of a single Gondwanan Early Palaeozoic deposit of a continuous belt of eastern sediments analogous to the Central Victorian, Melbourne Trough attributable to the Benambran Orogen. This assemblage includes the first record of an Ordovician bivalve from New Zealand, Ambonychia hailensis, the only Ambonychia known from Australasia. It also records for the first time the pelmatazoan camerate crinoid genera Reteocrinus, Aetheocrinus and Archaeocrinus and the only inadunate pelmatazoan crinoid genus known from New Zealand, Deocrinus, Additionally, the faunal assemblage includes unidentified sponge and brachiopod impressions, tabulate corals including Proheliolites goldfussU Plasmoporella cf. inflata, Paleopora inordinata, the rugose coral genus Streptelasma, and conodonts including Belodina, Evidence for paleocurrents, presumably detritus and plankton-laden, suggests an environmental niche suitable to suspension feeding. This is confirmed by the shallow water, sedentary, suspension feeding character of the invertebrate assemblage. It is suggested that the tabulate and rugose corals acted as key species in establishing a reef framework at this time. The byssate bivalve A. hailensis and the sessile pelmatozoan crinoids were ideally suited to such a reef environment, able to exploit nutrientrich currents and the maximum area of hard substrate available for colonisation. Isolation probably encouraged divergent evolution and species recorded from the Hailes Knob locality are most likely indicative of a local Gondwanan shelf marine province that restricted entry to some faunas and denied others. Because the fauna exhibits affinities with North American and European species, it is proposed that the occurrence of these Palaeozoic invertebrates in a southern Gondwanan offshore shallow marine facies at this time is a result of oceanic migration.

PHYLOGENETIC AND AREA CLADISTIC ANALYSES OF A DEVONIAN TRILOBITE FAUNA FROM WESTERN NEW SOUTH WALES Malte C. EBACH Department of Geology and Geophysics, University of Sydney, Sydney, NSW Eleven trilobite species have been discovered in the Lower Devonian (Pragian) of the Biddabirra Formation, Cobar, New South Wales. Six new species of trilobites belonging to five genera have been run through a phylogenetic analysis program PAUP (Swofford 1984), to produce a series of cladograms showing their relation within their respective genus and family. The genera included in the analysis are: Kainops, Paciphacops, Edgecombeaspis, Kettneraspis, Leonaspis, Acanthopyge and Cordania. Analyses using new and previous data (Ramskold & Chatterton 1988) have shown different results once new data and longer heuristic searches have been run. Within the phacopids, all Australian Paciphacops species show a closer affinity with Kainops, than to Paciphacops, and have therefore been assigned to Kainops. A new genus of odontopleurid from western New South Wales shows a close affinity with both Edgecombeaspis and Leonasipis. This close affinity with Edgecombeaspis would place the new genus and Kettneraspis, but not Dicanthaspis as outgroups in Adrains' (1997) analysis. The four consensus trees produced by the analyses, have been run through COMPONENT (Page 1993), a component analysis program. The analysis produced a series of area cladograms, and a consensus tree using assumption 2 (Nelson & Platnick 1981), as a result of producing the most resolved cladograms. The area cladogram shows the biogeographic distribution of the genera, a distribution that is similar for all Australian species. Adrain, J.M. & Ramskold, L., 1997. Silurian Odontopleurinae (Trilobita) from the Cape Phillips Formation, Arctic Canada. Journal of Paleontology 71, 237-260. Page, R.D.M., 1993. COMPONENT, Version 2.0. Natural History Museum, London. Ramskold, L. & Chatterton, B.D.E., 1988. Revision and subdivision of the polyphyletic 'Leonaspis' (Trilobitaj. Transactions of the Royal Society of Edinburgh: Earth Sciences 82, 333-381. Nelson, G. & Platnick, N., 1981. Systematics and Biogeographic/cladistics and vicariance. Columbia University Press, New York. Swofford, D.L., 1984. PAUP: phylogenetic analysis using parsimony. Version 2.4. Illinois Natural History Survey, Champaign, Illinois.

BUILDING T H E UPPER CAPITAN REEF FRAMEWORK (PERMIAN), GUADALUPE MOUNTAINS, NEW MEXICO-TEXAS, U.S.A: INTERRELATIONS BETWEEN PALEOECOLOGY AND ABIOTIC PROCESSES J.A. FAGERSTROM^ & O. WEIDLICtf ^Department of Geosciences, Oregon State University, Corvallis, OR 97331, U.S.A. ^Technical University of Berlin, Institut of Applied Geology, II, Sekr. EB 10, ErnstReuter-Platz 1, D-10587 Berlin, GERMANY Analyses of large acetate sheet tracings, close-up photos and numerous sub-horizontal quadrat surfaces at five localities near the base of the Guadalupe Mountains Escarpment indicate that the framework of the upper Capitan reef (Fagerstrom and Weidlich, in press) was built by about 34 species: one codiacean, 17 calcisponges, 9 Bryozoa, one richthofenid brachiopod, some crinoids and 4 Problematica. This widespread fossil community included members of the Constructor, Baffler and Binder Guilds; open surface phylloid algal and cryptic sponge-bryozoan dominated sub-communities were locally important. Upper Capitan reefs differ from more typical Phanerozoic reefs in their low diversity and packing density of the frame-building biota, low micrite content and abundant micro-frameworks with inter-grown syndepositional cements (botryoidal and isopachous, fibrous calcite: Weidlich & Fagerstrom 1997). Solitary, sub-cylindrical and multi-branched, clonal, erect sponges (and frondosefenestrate Bryozoa?) were the initial frame-builders in both open and cryptic habitats. Selective larval recruitment of erect sponges (and Bryozoa?) to hard substrates produced continuous upward accretion of the initial reef framework. On open surfaces and in pores formed by tabular sponges and fenestrate Bryozoa, erect and pendant sponges were supported in their hydrodynamically unstable growth positions by encrusters of uncertain taxonomic affinities, chiefly Archaeolithoporella hidensis but also Tubiphytes obscurus and an unnamed tubular organism. Rapid upward growth of A. hidensis, microbial micrite sheaths and syndepositional cements on outer sponge walls impeded ambient water circulation to the sponges, led to "creeping sponge death by suffocation" and eventual complete encrustation. Subsequent pore-filling of the initial framework by intemal sediment (packstone-grainstone; derived from the framework and the back-reef shelf/platform) and voluminous syndepositional marine-phreatic cements completed the frame-building process. Fagerstrom, J.A. & Weidlich, O. (in press). Origin of the upper Capitan-Massive Limestone (Permian), Guadalupe Mountains, New Mexico-Texas: Is it a reef? Geological Society of America, Bulletin. Weidlich, O. & Fagerstrom, J.A. (1997). Evolution of the upper Capitan-Massive Limestone (Permian), Guadalupe Mountains, New Mexico. Brigham Young University, Studies in Geology.

ORDOVICIAN PALAEOGEOGRAPHY OF THE SHAN-THAI TERRANE R.A. FORTEY Department of Palaeontology, The Natural History Museum, Cromwell Rd, London SW7 5BD, U.K. As the Ordovician faunas of the Shan-Thai Terrane become better known their palaeogeography becomes more interesting. The Terrane is a fault-bounded block including much of Burma and Thailand, which was assuredly peripheral to the main Gondwana continent in the Ordovician - but exactly where? The main groups used for biogeographic purposes have been trilobites, molluscs and brachiopods. In the earlier part of the Ordovician thick limestone successions have yielded a number of taxa which have been claimed as indicating close palaebiogeographic ties with the North China Platform and cratonic successions in Australia, where similarly shallow-water deposits certainly occur. Data published by the author and L.R.M. Cocks in the last year on southern Thai faunas has added new data on the later Middle and Upper Ordovician. Trilobite faunas of Caradoc age and younger are identical at species level to faunas from (among others) the Pagoda Limestone of the South China platform; while some of the taxa may be widespread, others, so far, are largely endemic to the Pa Kae Formation and the Pagoda, and thus support a 'switch' from North to South China affinities through the Ordovician. Recent description of the Himantia fauna brachiopods from the Ashgill Wang Tong Formation in S. Thailand also suggests (1) that the faunas described from the Mayanmar, Shan States by F.R.C. Reed are the same (2) that these may be senior synonyms of some species described from the South China platform. They are accompanied by the widespread trilobite Mucronaspis, the spread of which is related to the onset of the late Ordovician glaciation. The Ordovician biogeography of the Shan Thai faunas provides a good example of the how palaeontological evidence can contribute to reconstruction of terrane history, and how ambiguities can lead to new questions.

Caheniasaccites BOSE & KAR, 1966, Costatascyclus FELIX & BURBRIDGE, 1967 AND ALLIED GENERA: PORTRAIT OF A MIGRATING GYMNOSPERM FROM NORTH AMERICA, GONDWANA, AND RUSSIA C. B. FOSTER Australian Geological Survey Organisation, P.O. Box 378, Canberra, ACT 2601 Costatascyclus Felix & Burbridge, 1967 accommodates morphologically distinctive monosaccate pollen that appear disaccate due to bilateral constriction of the saccus, and possess, in the constricted areas, radiating folds of exoexine. The genus was originally described from shale units of the Springer Formation, southern Okalahoma. Conodonts from the correlative Imo Formation, belonging to the Adetognathuis unicornis zone, indicate that the Springer Formation is of Late Carboniferous, upper Namurian A (Amsbergian) age. A Namurian to Westphalian B range for the type species C. crenatus is confirmed by the 1986 study by Ravn of Iowa coals. In Australia, as yet undescribed monosaccate pollen, showing radial exoexinal folds and bilateral symmetry, first appear in the Spelaeotriletes (al. Anabaculites) ybertii Assemblage in the northern Australian Bonaparte Basin. They have been assigned, informally by the author, to Caheniasaccites Bose & Kar, 1966, first described from Early Permian glacial and periglacial deposits form Zaire. Elsewhere in Gondwana, Caheniasaccites spp. are common in Early Permian (?AsselianSakmarian) assemblages of the Granulatisporites confluens Zone of Foster in Foster & Waterhouse (1988) and are rarely seen in younger assemblages. In the Bonaparte Basin the age of the S.ybertii assemblage is poorly constrained: brachiopods belonging to the Echinoconchus gradatus fauna, occurring 10s of metres below the palynozone indicate that the base of the S. ybertii is not older than Namurian. If the first occurrence of monosaccate pollen is used as an isochronous marker, then the base of the zone also equates with the European Namurian, but independent evidence is required to avoid circular reasoning. SHRIMP dating of equivalent palynofloras in eastern Australia, belonging to either the S. ybertii or the overlying Diatomozonotriletes birkheadensis Zone are of Namurian age (see Roberts et al. 1995). Pollen with this distinctive morphology was described by as Florinites? plicatus f. magnus Inossova sp. et f. nov. (Westphalian to Stephanian) and Potonieisporites radiosus Shwartsman (Stephanian to Asselian) in the 1976 Atlas of microspores and pollen from the Upper Carboniferous and Lower Permian of the Donets Basin by Inossova et al. The equivalent European age ranges for these taxa are shown in parentheses. The distinctive morphology, relatively narrow age range, and the initial widespread geographic occurrence, in the Namurian, of the Caheniasaccites-Costatascyclus complex, suggests a common gymnosperm origin, at least at family level. No in situ pollen records, and therefore plant associations, are yet known. In Gondwana, the parent plants were established, or perhaps originated, as part of the pre-Permian flora, and re-emerged after the Early Permian glaciation. It is tempting to suggest that, prior to the Permian glacial period, the parent plants migrated northwards, only to return when climatic conditions ameliorated. Roberts, J., Claoue-Long, J., Jones, P.J. & Foster, C.B., 1995. SHRIMP zircon age control of Gondwanan sequences in Late Carboniferous and Early Permian Australia. Geological Society of London Special

Publication 89, 145-174.

PHANEROZOIC GCM RESULTS AND CLIMATE DATA FOR AUSTRALIA L. A.FRAKES University of Adelaide Quantitative data bearing on Phanerozoic climates of Australia are summarized and comparisons are made with the predictions of atmospheric General Circulation Models over eleven time slices from the Silurian to the Miocene. Summaries of the model-derived mean temperature and precipitation conditions are given individually for each of four regions (Southeast, Northeast, Central and West). Model results provide the only means of estimating climate conditions in many cases but there is fair agreement between the two approaches in most cases where data are available. The Southeast experienced winter freezing conditions from the Permian through until the middle Cretaceous and was consistently the coolest of the four regions, in part because of its proximity to the pole but also because it was the site of elevated terrain throughout most of the Phanerozoic. It also tended to be among the wettest parts of Australia over this time and was subject to frequent winter storms. The Northeast, in contrast, often was the warmest part of the continent, although strong seasonality and freezing winter temperatures were common during the Mesozoic and Palaeozoic. Strong seasonality (>30 degrees C contrast between winter and summer) also apparently characterized most of the record for the Central region, except for the Tertiary, most of the Jurassic, and the early Palaeozoic. Together with the Southeast, this region experienced the most intense effects of winter, especially in the Triassic. Triassic seasonality also affected the West region, and, interestingly, the highest summer temperatures are predicted for the West for most of the time slices studied. The models used today are imperfect but nonetheless offer a means of prediction and perhaps will stimulate the gathering of quantitative data in a continent where information is hard to come by and for which not much has yet been presented.

BIOSTRATIGRAPHY AND PALAEO-BIOGEOGRAPHY OF THE LATE ORDOVICIAN-EARLY SILURIAN CHITINOZOANS FROM THE ZAGROS BASIN, SOUTHERN IRAN Mohammad GHAVIDEL-SYOOKI Exploration Division, National Iranian Oil Company, P.O. Box 1065, Tehran, IRAN A thick lower Palaeozoic succession is well-developed in Tang-e-Zakeen at Kuh-eFaraghan, approximately 80 km north of Bandar Abbas city. The succession is 807 m thick and mainly consists of clastic sediments and subordinate limestone. This succession has been divided into the Seyahou (741 m) and Sarchahan (66 m) Formations, respectively belonging to the late Ordovician and Early Silurian. The Syahou Formation consists of variegated shales, siltstones and subordinate limestones. Sedimentary structures such as cross-bedding, symmetrical ripple-marks, trace fossils (trails, tracks and burrows) and large concretions are common in the Seyahou Formation, whereas they are absent from the Sarchahan Formation. The Seyahou Formation contains orthocerids, trilobites, graptolites, brachiopods and crinoids, whereas only graptolites are present in the Sarchahan Formation. A total of 200 samples were selected and treated for palynological study. Fortunately, all samples contain well-preserved and abundant chitinozoans and acritarchs. The Seyahou Formation is characterized by Cyathochitina companuliformis, Rhabdochitina usilata, Calpichitina lenticularis, Jenkinochitina tanvillensis, Cyathochitina fistulosa, Belonechitina micracantha, Acanthochitina barbata, Desmochitina bulla, Desmochitina minor, Armoricochitina nigerica and Ancyrochitina merga. The Upper Ordovician chitinozoan species of the Seyahou Formation were compared with those from elsewhere. This comparison revealed that there are very strong affinities with those of the same age from Saudi Arabia, Libya, Algeria, Morocco and South America. Likewise, the Sarchahan Formation (Early Silurian) contains well-preserved and abundant diagnostic chitinozoan species, including.Siphonochitina fragilis, Pterochitina perivelata and Ancyrochitina ancyrea. This comparison revealed that the Early Silurian chitinozoan species of the Zagros Basin are quite similar to those of Saudi Arabia, North Africa and Southern Europe. Therefore, based upon the chitinozoan fauna, the Zagros Basin was part of the gondwanan landmass during the lower Palaeozoic.

AUSTRALASIAN TRIASSIC AND JURASSIC BIOGEOGRAPHY J.A. GRANT-MACKIE^ & H J . CAMPBELL^ ^ Department of Geology, University of Auckland, Auckland, NEW ZEALAND ^Institute of Geological and Nuclear Sciences, Lower Hutt, NEW ZEALAND The Australasian part of the Gondwana land surface in the Triassic supported a flora differentiated into 2 zones: cool, including East Australia and New Zealand (Ipswich Flora; 40"" E-70E'' S) and warmer in Western Australia - the northwest shelf - Papua New Guinea (Onslow Flora; 25"" E-35E° S). This latitudinal zonation continued into the Early Jurassic before general cooling produced an almost uniform Gondwana flora through to the Early Cretaceous. The few terrestrial vertebrates do not contradict this pattern for the TriassicJurassic. Fresh-water fish and amphibians show an eastem Gondwana - east Laurasia link, probably via Asian island terranes situated then in east Tethys/westem Pacific. Marine patterns are more complex. However, most simply stated, there was a warmer water province along the southern margin of Tethys and a cooler zone on the East Gondwana - SW Panthalasssa margin, with some members of the latter confined to the shallower waters of that region, others able to cross the deeper central Panthalassa to Central America, and still others linked along the Gondwana coast to West Antarctica and Andean South America. The relative strength of these links varied through time with a combination of global climatic changes and local plate tectonic activity. The New Zealand - New Caledonia marine Triassic (Murihiku terrane), with its very high level of endemism amongst benthic genera, has traditionally constituted the small Maorian Province which lasted through to the late Early Jurassic. Reconsideration of ammonoid faunas indicates that this province may have been initiated in the Middle, not Early Triassic, as previously thought. Recently discovered ammonoids in the Maitai terrane prove to be of Early Triassic age and Tethyan and cosmopolitan affinity, with little hint of a separate austral province at that time. This probable extension of Tethyan links in the Early Triassic is also likely to have encompassed the Maryborough basin of southeast Queensland. The Rakaia terrane of eastem New Zealand also lay in a region of Tethyan influence but with links also to the Murihiku Triassic, whereas the Waipapa complex lay in lower latitudes and shows no Murihiku relations. Around the Early/Middle Jurassic boundary changes to the north greatly improved marine communication between the eastem Tethys and southwest Panthalassa. The Maorian Province disappeared and Murihiku, Waipapa and Torlesse areas shared an influx of east Tethyan forms. The Pahau part of the Torlesse lay far enough away from the others to show a stronger Tethyan influence. Communication also extended to West Antarctica and southem South America, but ammonite faunas show strong trans-Panthalassa links in addition with Mexico.

AUSTRALASIAN LUNULITE BRYOZOANS - THE EUROPEAN CONNECTION? Eckart HAKANSSON Geological Institute, University of Copenhagen, Oster Voldgade 10, DK-1350 Copenhagen, DENMARK Lunulite bryozoans constitute one of the most conspicuous groups in cheilostome history. In addition to their highly organized, very characteristic colonies, they share a very complex biogeographic history, with a pattern combining dramatically changing global diversity through time with very distinct shifts in geographic distribution from their first appearance in the Late Cretaceous of NW Europe. In the Australasian biogeographic realm their record has been traced back to the late Paleocene, but here significant proliferation did not take off until the Miocene. This development shows no sign of abatement up to the present day, where lunulite bryozoans constitute a prolific, conspicuous endemic element in the shelf seas of Australia and New Zealand, a situation which is in strong contrast to the preceeding biogeographic history of the group. Thus, in the Eocene, lunulites had an equally prominent stronghold in North America, while a fair amount of contemporaneous species have nevertheless been recorded from most of the rest of the World. In contrast, in the latest Cretaceous they flourished in NW Europe, reaching their highest diversity ever, while the rest of the World has yielded just a handful of species of this age. The complexity of this biogeographic pattern naturally raises the question as to what extent lunulite history involves migration as opposed to independent, local evolution of free-living clades of lunulite body plan. However, recent investigations already indicate that very few lunulite species may have survived the Cretaceous-Tertiary boundary into the Danian, and possibly none of the survivors lived through the Paleocene. Hence the Eocene peak in lunulite history was possibly derived entirely from lunulite clades evolved independently during the early part of the Paleogene. In view of this possibility the predominantly Neogene lunulite history of Australasia should also be reconsidered. The talk presents an updated overview on lunulite biogeography, and reconsiders the origin of the Australasian lunulites based on preliminary results from a study of the oldest Australasian lunulite fauna (Thanetian, Carnarvon Basin, Western Australia).

AN OVERVIEW OF AUSTRALASIAN PALAEOBIOGEOGRAPHY IN ITS GLOBAL CONTEXT A. HALLAM School of Earth Sciences, University of Birmingham, Birmingham B15 2TT, U.K. Alfred Wegener can with some justification be regarded as the father of modem Australasian palaeobiogeography. With characteristic prescience, he observed that, faunally speaking, Australia is "like a foreign body from another world", and attributed its distinctiveness to a long history of geographic isolation followed by Cenozoic collision with South-East Asia, as indicated by the line of sharp faunal discontinuity known as Wallace's Line. Modern tectonic reconstructions suggest that the major collision took place about 25 million years ago. There is a well established generalised track, to use Croizat's term, of living biota including marsupials, ratite and galliform birds, chelid turtles, chironomid midges, Nothofagus and proteacean plants to suggest former land links between South America, Antarctica and Australia, with good evidence that Nothofagus in particular cannot disperse effectively across marine barriers. A primitive platypus has been found in the Cretaceous of White Cliffs, NSW; both comparative morphology and serology suggests that all living Australian marsupials are more closely related to each other than to any New World family, suggesting derivation from a common stock. According to Tedford marsupial immigrants dispersed in either the Late Cretaceous or Palaeocene and subsequently became isolated, with the radiation probably being initiated by insectivorous didelphines. There is abundant Nothofagus pollen in the Maastrichtian of Australia and New Zealand and Proteacea in the Late Cretaceous and Palaeogene of Australia. A continuous land link must have existed in the late Mesozoic between South America and Australia. A rift zone between Australia and Antarctica became established in the Late Jurassic; the record of magnetic anomalies indicates that deep sea extended as far as the longitude of western Victoria by 80 Ma. No marine connection was established from the Indian to the Pacific Ocean until Eocene times. If such a connection were established earlier, it would be difficult to account for the dissimilarity of Late Cretaceous-Palaeogene shallow-water molluscan faunas between southwest Australia and southeast AustraliaNew Zealand. Geological and palaeontological evidence suggests a land bridge along the South Tasman Rise between North Victoria Land and south east Australia. Final separation took place around 38 Ma, so land dispersal ceased then. In the latest Cretaceous New Zealand separated completely from Antarctica and by the Late Palaeocene had migrated a long way north, parallel to Australia. The absence of any fossil or endemic extant mammals in New Zealand may indicate that Greater Antarctica was not populated by mammals prior to the separation of New Zealand. The close proximity of New Zealand and the similar ages of tectonic episodes suggests one orogenic belt extending to New Zealand. By the Late Eocene-Early Oligocene dispersal would have been very improbable, especially to and from Australia. Faunal continuity was very limited because of the sharply differentiated marsupials of Australia and South America. Their absence from New Zealand may mean that marsupial dispersals between Greater Antarctica and Australasia were limited to the latest Cretaceous.

The Early Cenozoic climates of Australia were characterised by greater humidity than today, with rain forest vegetation dominating the landscape across the southern margin and locally spreading inland to central Australia; the high fem diversity, types of tree and epiphytic fungi point overwhelmingly to high humidity. There are many endemic Palaeocene marine molluscs in New Zealand. A number of taxa which survived well into the mid-Cenozoic in Antarctica disappeared from Australasia after the Palaeocene. This disappearance of palaeaustral species is thought to be due to the northward migration of Australia and New Zealand. By mid-Eocene times warm water Indo-Pacific elements such as conid, harpid and cypraeid gastropods made their first appearance in New Zealand. Before the mid-Oligocene Australia-New Guinea-New Zealand was part of the Indo-Australia plate, well separated from the Java-Borneo part of the Asian Plate and Solomon-New Hebrides-Fiji part of the Pacific Plate. By the mid- to Late Oligocene there was an intermingling of larger foraminifera such as Lepidocyclina and Miogypsina, another biogeographic response to the northward migration of Australasia. As a consequence of this, both biogeographic and oxygen isotope data suggest a temperature rise from the Early to Middle Cenozoic followed by a fall towards the present, which is anomalous compared with the rest of the world, which reveals an overall temperature decline after the Early Eocene. The conventional view of the end-Cretaceous mass extinction of terrestrial vertebrates relies heavily on negative evidence, largely from the fossil record of the Northern Hemisphere. In contrast, molecular data provide hard evidence for a Cretaceous diversification of modern mammals and birds, with mass survival across the K-T boundary. Whereas in former times Laurasia was held to be a centre for terrestrial vertebrate evolution, Gondwana, of which Australasia is a key component, is now seen as a likely major source, as has long been acknowledged for the angiosperms. There is evidence of a distinctive southern high-latitude fauna of Cretaceous terrestrial vertebrates. Thus the prolific Early Cretaceous dinosaur faunas of south east Australia, with a high proportion of endemic relict taxa, are dominated by hypsilophodontids. This indicates among other things that at least some dinosaurs could cope with high latitudes. Hypsilophodonts are also known from the Upper Cretaceous of both New Zealand and Antarctica, and Crame believes that Antarctica could have been a major source of evolutionary novelties. For most of the Mesozoic Australasia formed part of the Gondwana supercontinent, with the initiation of separation between East and West Gondwana taking place early in the Late Jurassic; with the opening of the Indian Ocean, Australia-Antarctica began to separate from India in the Valanginian. A consensus has developed that much of South East Asia consists of terranes rifted from the margin of East Gondwana. Kannemeyerid dicynodont reptiles have been found in the Lower Triassic of Australia, South Africa and South America, while in the Upper Triassic prosauropod dinosaurs occur in Australia and are widespread elsewhere. The cosmopolitan distribution of such reptiles points to the absence of climatic and physical barriers to dispersal. Unlike elsewhere, labyrinthodont amphibians persist from the Triassic into the Jurassic of Australia. The Dicroidium flora ranges up to the Camian within Gondwana, including Australia. Doubt has been thrown on the distinctiveness of a Maorian Province in the South West Pacific during the Late Triassic and Early Jurassic, because some of the diagnostic bivalves and brachiopods have now been found also in Chile. The story of New Zealand in the Mesozoic is complicated

by the presence of displaced terranes as elsewhere on the Pacific margins. The influx of Tethyan faunas into New Zealand in the mid-Jurassic is more likely to be due to high sealevel stand than warmer climate, as formerly believed. The Cretaceous marks the earliest time in the Mesozoic when distinctive austral biotas became well established. Much of this centred on Australasia, as illustrated for example by the distribution of the dimitobelid belemnites. With regard to the Late Palaeozoic, Young distinguishes several fish provinces for the Early Devonian. Based on thelodonts and other groups the endemism of the East Gondwana region, based mainly on Australian evidence, persisted into the Late Devonian. Emsian marine placoderms of eastern Australia show some affinities with South China. In the Late Devonian provinciality declined. Thus, whereas in the Early Devonian there had been strong isolation between South China and Australia, by the Late Devonian the sinolepids (antiarch placoderms) had reached eastern Australia. Phyllolepid placoderms may have originated in East Gondwana, and their appearance in Euramerica near the Frasnian-Famennian boundary signifies the establishment for the first time of a persistent terrestrial connection, because this is a non-marine group. The Carboniferous tetrapod record, hitherto confined to Euramerica, has now been extended to Australia. The Permian Glosssopteris flora of Australia is interpreted by Ziegler et al as part of a South Temperate or Gondwana biome. Permian marine brachiopod distributions indicate according to Waterhouse and Bonham Carter both cold temperate (New Zealand) and polar (eastern Australia) provinces. For the Early Palaeozoic, the Lower Ordovician trilobite and brachiopod cratonic faunas are biogeographically the most instructive. Gondwana regions include both Australia and South China, which fall into the same trilobite province. In contrast Silurian marine invertebrates are relatively cosmopolitan, with low-diversity temperate faunas in south east Australia. Much uncertainty persists about the Cambrian, because of limited or ambiguous data. Attention is drawn to the radical palaeogeographic reconstructions of Dalziel and Dobson, respectively utilising in part the distribution of Australian trilobites and graptolites. Cooper, A. & Penny, D., 1997. Mass survival of birds across the Cretaceous-Tertiary boundary: molecular evidence. Science 215, 1109-1113. Crame, J.A., 1997. An evolutionary framework for the polar regions. Journal of Biogeography 24, 1-9. Dobson, J.E., 1996. A palaeogeographic link between Australia and eastern North America : a New England connection? Journal of Biogeography 23, 609-617. Hall, R., 1996. Reconstructing Cenozoic SE Asia. In: Hall, R. & Blundell, D. (eds.). Tectonic Evolution of Southeast Asia. Geological Society of London Special Publication 106, 153-184. Hallam, A., 1994. An Outline of Phanerozoic Biogeography. Oxford University Press, Oxford.

BIOGEOGRAPHY OF THE PROCOLOPHONIDS Tim HAMLEY Department of Zoology, University of Queensland Procolophonids were small, superficially lizard-like reptiles that entered the fossil record in the Late Permian and attained an almost world-wide distribution before their apparent demise toward the end of the Triassic. Amongst the taxa generally accepted as procolophonids, four evolutionarily sequential groups can be recognised: firstly, the relatively primitive Late Permian forms such as Owenetta from South Africa; second, the slightly more derived Permo-Triassic forms such as Tichvinskia from Russia and Pentaedrusaurus from China; third, the Early Triassic Procolophon type taxa from South Africa and Antarctica and Eumetabolodon from China; and finally, the more derived, Upper Triassic forms such as Hypsognathus from North America. New procolophonids from Australia appear to be most like the Tichvinskia-Pentaedrusaurus types but are slightly less derived, despite being nominally more recent according to current stratigraphical correlations. These findings are discussed in terms of their palaeo-biogeographical and stratigraphical implications.

35 NEW WORLD ORDER IN BAT BIOGEOGRAPHY

SJ. HAND School of Biological Science, University of New South Wales, Sydney, NSW 2052. Startling fossil discoveries, on-going taxonomic revisions and new phylogenetic understanding have combined to turn Australasian bat biogeography upside down. No longer can Australia be viewed merely as a receptacle for an influx of Asian bats in the middle Tertiary as the Australian Plate neared southeast Asia. A long and complex historical biogeography is indicated for Australasian bats. The world's extant bat radiation is now believed to have occurred in the Southern Hemisphere (modem bat groups evolving there from isolated immigrant archaic bat groups in the Paleocene or early Eocene), with Australia standing centre stage. Today, bats comprise nearly one third of Australian non-marine mammals; taxonomic revisions and new field survey techniques (involving the routine use of increasingly sophisticated electronic detectors of speciesspecific sonar calls) have pushed the tally of extant Australian bats to 87 species. Concurrently, new discoveries of Tertiary bat fossils at Murgon (SE Qld), Riversleigh (NW Qld), Bullock Creek (NT), and Chinchilla (SE Qld) have dramatically increased the depth and breadth of the Australian fossil bat record. Recent findings with major implications for historical biogeography include the following: (1) The history of bats in Australia is much longer than previously thought (at least 55 Ma) and hence among the oldest for bats anywhere in the world; (2) Many bat families had already radiated in Australia by 25 million years ago (e.g. Hipposideridae, Megadermatidae, Molossidae, Mystacinidae); (3) Endemicity in the Australian bat fauna (extant and extinct (+)) is much higher than once appreciated; endemic genera that can now be added to Macroderma and Rhinonicteris include Vespadelus, Falsistrellus, Scotorepens, Scoteanax, Australonycteris (+), Riversleigha (+), Xenorhinos (+) and Miophyllorhina (+). Indeed, a large component of the extant Australian fauna, i.e. members of the Vespertilionini tribe (34 spp.), possibly represents a monophyletic endemic group; (4) Although many Australian genera demonstrably have northem (Asian) affinities (e.g. Marina, Pipistrellus, Megaderma, etc.) and/or broad Old World distributions (e.g. Taphozous, Pteropus, Miniopterus, etc.), it is not entirely clear in which direction dispersals occurred; (5) Southern ('Gondwanan') affinities of the Australian bat fauna are indicated by: (a) the discovery of plesiomorphic members of the New Zealand bat family Mystacinidae in a number of Australian Tertiary deposits, indicating an Australian Oligocene origin for that family; and (b) discovery of a new Australian Miocene bat family, allied to both the Mystacinidae and Noctilionidae, suggesting an Australian origin for basal members of the most diverse of all bat groups, the South American superfamily Noctilionoidea (=Phyllostomoidea).

BIOGEOGRAPHICAL OBSERVATIONS ON THE CRETACEOUS BIOTA OF AUSTRALASIA Bob HENDERSON (compiler)\ James CRAMPTON^ Mary DETTMANN^ David HAIG^ Ralph MOLNAR^ Samir SHAFIK^ Jeff STILWELL^ & Tony THULBORN^ 1 School of Earth Sciences, James Cook University, Douglas, QLD 4811 2 Institute of Geological and Nuclear Sciences, PC Box 30368, Lower Hutt, NEW ZEALAND 3 Department of Botany, University of Queensland, St Lucia QLD 4072 4 Department of Geology and Geophysics, University of Western Australia, Nedlands WA 6009 5 Queensland Museum, PC Box 300, South Brisbane QLD 4101 6 Australian Geological Survey Organisation, GPO Box 378, Canberra ACT 2601 7 Department of Zoology, University of Queensland, St Lucia QLD 4072 The Cretaceous System is well represented in Australasia. The generation of new seafloor to the west, south and east of continental Australia, associated with the Cretaceous fragmentation of Gondwana, led to the development of extensive passive margin sedimentary systems. Substantial tracts of these systems are displayed on the western and southern continental borderlands of Australia. The separation of greater New Zealand from Marie Byrd Land in the mid-Cretaceous, combined with opening of the Tasman Sea, induced passive downflexing of of the newly separated microcontinent and the development of major transgressive sediment systems in its eastern parts. In addition, late Early Cretaceous epeirogenesis in eastern Australia combined with the a high stand of global sealevel lead to the development of a vast epicontinental sedimentary platform. All of these sediment systems accumulated in high palaeolatitudes. They are predominantly of shallow marine or terrestrial facies and they contain a rich and varied palaeontological record of Aptian-Maastrichtian age. The Early Cretaceous (Berriasian-Barremian) record for Australasia, however, is poor. Mollusca comprise the dominant macrofossil grouping. The biogeographical affinities of Australasian ammonite faunas shows marked changes through time. Early and midCretaceous faunas of the continental borderlands are distinctively cosmopolitan in character, sharing most genera and several species with western Europe. A like flavour applies to Aptian ammonites of the east Australian epeiric sea but Albian assemblages of this tract are provincial in aspect and show the first clear evidence of an Austral (or South Pacific) faunal realm. Late Cretaceous (Campanian - Maastrichtian) assemblages show a marked latitudinal control on the aspect of faunas. For New Zealand, which lay beyond 80^ S, kossmaticeratid ammonites of Austral affinities predominated whereas for western Australia, biogeographic affinities with the northern hemisphere persisted. Australasian belemnite assemblages are comprised exclusively of the Dimetobelidae which have a distinctive Austral distribution. Aptian-Albian belemnites from Papua New Guinea contrast in that they include elements with European affinities via a Tethyan distribution. Bivalves, gastropods and scaphopods display a patchy but variably rich stratigraphic record with an overall moderate level of diversity recognised in Australia and Papua New Guinea (late Aptian-Albian, Cenomanian, Santonian, late Maastrichtian), New Zealand (Aptian-late Maastrichtian) and the Chatham Islands (late Albian-Santonian, late

Campanian-late Maastrictian). Species- and genus-level diversity increased throughout the period to an acme in the Maastrichtian. Bivalves are generally more diverse than gastropods, and scaphopods are depauperate. The final fragmentation of Gondwana during the Late Cretaceous in concert with ensuing changes in oceanic circulation, lowering sea surface temperatures and rising latitudinal temperature gradients, enhanced the distinctiveness of the fauna as reflected by its increasingly endemic character at species level. This faunal provincialism resulted from evolutionary divergence of Australasian Late Cretaceous Mollusca, reflecting range retractions from a formerly cosmopolitan Mesozoic world. Benthic foraminifera are known from every Cretaceous stage but are best documented from the Aptian and Albian where they are widely distributed. They belong to four main faunal associations: (1) the Ammobaculites association characteristic of the silled interior seas, characterised by siliceous agglutinated foraminifera and a distinctive hyaline assemblage apparently adapted to cool water, low oxygen and slightly brackish conditions; (2) the Marsonella association found in biogenic-carbonate muds and sands and in calcareous siliciclastic muds of the open continental shelf and upper slope, characterised by calcareous agglutinated foraminifera and a diverse hyaline assemblage apparently adapted to normal marine neritic-bathyal conditions; (3) a deep-water association of siliceous agglutinated species found below the carbonate compensation depth in clays; and (4) a "larger" orbitoid association found in the biogenic carbonate sand characteristic of warm, clear shallow seas. The Ammobaculites association has an established range of upper Hauterivian to Turonian and has been widely recognised in the epicintinental platform, the Papuan Basin and basins of northwestern Australia. The Marssonella association is known from successions proximal to continental margins of Albian-Maastrictian age throughout Australasia and from Papua New Guinea. The deep water association is known from the Berriasian-Aptian of the Argo Abyssal Plain and the Late Cretaceous of Papua New Guinea and New Zealand. The "larger" orbitoid association is known from shallow marine limestones structurally emplaced in the New Guinea Orogenic Belt east and north of the Papuan Basin. Very few of the genera represented in these associations are endemic to the Australasian region and many of the species present are cosmopolitan within the limits of of their preferred faunal associations. Elsewhere, the Ammobaculites association is best known from the Western Interior Basin of North America and the Western Siberian Lowlands; the Marsonella association is recognised on most continents; a similar deep water association is known from the North Atlantic Ocean and in the Carpathians; and the "larger" orbitoid association is known widely in basins marginal to Tethys. Planktic foraminifera usually accompany the Marsonella association and also occur in the deep water association. The oldest assemblages occur in Aptian sediments from the Cuvier Abyssal Plain. In the epeiric and marginal basins, planktic foraminifera first appeared within the Early Albian. They migrated into the interior basins during maximium flooding episodes sometime after radiolaria had become established in the marine plankton of what were shallow seas. In general the planktic fauna is diverse and transitional between assemblages of warm water affiliation characterised by keeled Planomalina - Rotalipora - Dicarinella - Marginotruncana - Globotruncana and an assemblage of cool water affilitaion dominated by Hedbergella-WhiteinellaRugoglobigerina and biserial species. The north-south boundary of these assemblages fluctuated during time. The Australasian planktic foraminifera belong to cosmopolitan

taxa. The distribution of calcareous nannofossils was greatly influenced by the temperature regime. Valanginian assemblages in NW Australia comprise mixtures of cool-water and warm-water species, suggesting connections with Tethys. Similarly mixed Albian assemblages covered a wide, mid-latitude province extending in an arc from the Naturaliste Plateau and Carnarvon Basin to the Papuan Basin of Papua New Guinea. Coeval Albian assemblages in the Eromanga Basin contained more abundant cool water species and lacked those of warm water affiliation, reflecting a location close to the pole. Later in the period, during intervals of depressed surface water temperature and increasing differentiation of water masses (e.g. during the Campanian-Maastrichtian), assemblages broadly reflected their latitudinal position. For warm intervals, such as the Santonian, assemblages were less clearly differentiated. These trends are particularly evident along the western margin af Australia. Continental Cretaceous vertebrates from Australasia are poorly known and largely resticted to the Early Cretaceous record. For most of the period, the record is inadequate for the discrimination of regionality. Aptian - Albian body fossils are suggestive of regionalisation with large herbivorous dinosaurs feeding both at ground level and in canopy vegetation at heights of up to 6 m in Queensland whereas for New South Wales and Victoria only small forms feeding within 2 m of the ground are known. Footprints, however, indicate the presence of large forms in New South Wales at least, and the presence of small forms in Queensland. The depositional regime in Victorian Otway Group, which hosts the known record, may have induced a taphonomic bias towards the preservation of small forms only. A contrast is also apparent in aquatic assemblages with late-surviving temnospondyls represented in Victoria where crocodilians are very rare whereas temnospondyls are unknown for Queensland and New South Wales but crocodilians are not uncommon. The Broome ichnofauna (Neocomian) provides one of the most diverse ichnofossil records of dinosaurs known from any region of the world and indicates that the known body fossil record is largely incomplete. Consequently, the Australasian Cretaceous vertebrate record must be interpreted with great caution. The Cretaceous vegetation of Australia was floristically heterogeneous. Much of the region was forested but woodlands, heathlands and aquatic communities were also represented. The high latitude forests had an open structure throughout the Cretaceous and podocarps and araucarians were important canopy components. Earliest Cretaceous understory communities included pteridosperms, cycadeoids, cycads, and crypogram associations and these were modified during Aptian-Cenomanian times by shrubby angiosperms. Angiosperms entered the canopy of southern Australian forests by the Santonian and included diverse Proteaceae, which, by the Campanian were associated with increasingly diverse angiosperms including Nothofagus, Winteraceae, Trimeniaceae and Epacridaceae.

FROM GO TO WOE - THE TRIASSIC VEGETATION OF EASTERN AUSTRALIA W.B. Keith HOLMES ^'Noonee Nyrang'^ Wellington, NSW 2820. The boundaries of the Triassic Period are marked by extinction episodes. The catastrophic extinction event at the Permian/Triassic boundary (251 Ma) saw the demise of 95% of marine species world-wide. In Eastern Australia, the highly productive Glossopteris Flora - source of some of the world's greatest coal reserves - was obliterated at this time. In the Ulawarra Coal Measures of the Southern Sydney Basin, a radical change of microflora occurs within the uppermost coal seam - the Bulli Seam. Macrofossils of a new impoverished plant assemblage are found in the sediments directly overlying this seam and also above the coal measures in the Cox Gap area in the Hunter Coalfield. This earliest Triassic plant assemblage was characterised by the presence of ''Thinnfeldia" callipteroides -a pteridosperm (seed-fern) with several-times forked fronds, IsoetesA\kt lycopods, Schizoneura horsetails, conifers, and ferns known only from sterile fronds. This short-lived flora was soon succeeded on the vast Sydney Basin floodplain by a littleknown flora dominated by forests of Voltziopsis conifers. By Late Scythian time the conifers were replaced by a distinctive broadleaf flora containing Dicroidium zuberi (a bipinnate seed-fern with forked leaves), other seed-ferns, taeniopterids with simple leaves, horsetails and lycopods. Fossilised leaves and fertile organs from plants of the Dicroidium zuberi Flora are preserved in the seacliffs north of Sydney and from localised sites in the Blue Mountains and in grey shales of the Lome Basin on the NSW north coast. Monodominant thickets of the lycopod Pleuromeia (=Cydomeia) which grew in areas of high water table are also preserved as complete plants or detached leaves and cones. The Dicroidium zuberi Flora flourished until the Late Anisian when it was enriched by the arrival of Dicroidium odontopteroides and many other diverse elements. The Dicroidium odontopteroides Flora, which was widespread throughout the southern continent of Gondwana, is preserved in the upper part of the Wianamatta Group in the Sydney Basin. Rich assemblages also occur in the Ladinian Esk Trough of Southern Queensland and the Nymboida Sub-Basin of northern NSW This flora of high diversity contained representatives of all modem plant groups (with the exception of the angiosperms), as well as many forms that were to become extinct. A contemporaneous flora with more sclerophyllous forms is preserved in the lowermost sediments of the Great Australian Basin in the Dubbo area of western NSW The Dicroidium odontopteroides Flora continued into the Camian when rich and diverse plant communities flourished in the Ipswich and Clarence-Moreton Basins of southern Queensland and in the Newtown and Brady Formations in Tasmania. Studies by Walkom, Jones & de Jersey, Retallack and Webb of the taxonomy and vegetation types during the Early to Late Middle Triassic from the Sydney, Ipswich and Esk areas, indicate the richness of land vegetation during a time of moist temperate cHmatic conditions even though Australia was attached to eastern

Antarctica and lay at high latitudes - 50^ - 60^ south of the equator. By the Late Triassic the climate was tending warmer and drier. Scanty evidence from the few known sedimentary deposits of this time - the upper Leigh Creek deposits in South Australia and Abedare and Raceview Formations in Queensland shows that the Dicroidium odontopteroides Flora was in sharp decline with only a limited range of taxa still present. The Triassic/Jurassic boundary was marked by severe vertebrate extinctions. The palynofloral turnover at the boundary was sharp and substantial. Australian Jurassic floras had a more cosmopolitan aspect and were dominated by conifers, ferns, cycads and cycadeoids and with minor occurrences of seed-ferns, Equisetum horsetails and ginkgos. In this presentation I will focus on the rich and diverse Ladinian Nymboida Flora which flourished ca 237 million years ago. A taxonomic study of this flora based on >2,000 catalogued slabs is in progress. The flora, comprising ca 60 taxa, includes bryophytes, sphenophytes, lycopods, ferns, pteridosperms, ginkgos, cycads, conifers and some enigmatic remains. The fossil material forms a comprehensive sample of the associations and successions of vegetation which occupied a space of less than 2 hectares through a limited span of geological time. The ecological setting was a mosaic of plant communities characteristic of an alluvial floodplain that was periodically overwhelmed by deposits of mud, silt or sand from overbank flood episodes.

THE EVOLUTION OF THE RATTUS GROUP (MAMMALIA, RODENTIA) DURING THE PLIO-PLEISTOCENE IN THAILAND Jean-Jacques JAEGER* & Yaowalak CHAIMANEE** * Institut des Sciences de PEvolution, CNRS-Univ. Montpellier II, CC64, Place Bataillon, 34095-Montpellier Cedex 5, FRANCE. ** Paleontology Section, Department of Mineral Resources, Rama VI Road, Bangkok-10400, THAILAND. The Rattus group represents one of the most spectacular exemple of adaptive radiation within the evolution of mammals. We distinguish two levels, one concerning the diversification of the genus Rattus relatively to its sistergroups {Maxomys, Niviventer, Leopoldamys, Berylmys), the other consisting of numerous species within the genus Rattus which are largely distributed over Southeast Asia and Australasia. The understanding of the origin of Rattus, the age of its diversifications and the causes of these successive radiations represent an important challenge for evolutionary biology. Recent molecular data suggest a recent origin of the genus Rattus (8-6 Ma) followed much later by an exceptionally fast diversification within Rattus (since 2.5 Ma). According to these results, Maxomys would represent the oldest group, followed by the NiviventerLeopoldamys group with Sundamys-Berylmys-Bandicota and Rattus as sister-group. We constructed the phylogenetic relationships of the Thai representative of these taxa using only molar characters, which represent the only preserved characters in the fossil record. Our cladogram, based on 65 molar characters, is largely congruent with the molecular data. Some differences concern the more early position of Rattus relative to its sister group consisting of Bandicota- Berylmys and of Niviventer-Leopoldamys. Early position of Maxomys, close relationships between Niviventer and Leopoldamys appear similarily to the molecular data. Paleontological data relative to 20 karstic or cave localities largely distributed over all Thailand indicate that, during the Pliocene, Rattus was represented by a few number of species, confirming previous observations made in Southern China. The Pliocene community of Murid rodents was significatively different from the living one, with extinct genera or species and only one species of Rattus, The number of Rattus species seems to increase progressively during the Pleistocene. The analysis of all available evidences indicate that during the latest Tertiary, there was a significative amount of grasslands by comparison to the present day, indicating stronger seasonality. During the Pleistocene, climate seem to have become wetter with less seasonality and more widespread evergreen forests over the country. During the late middle Pleistocene, the climate was wet and cooler than today, and the rodent composition indicate a downward shift of at least 1000 meters in vegetational zones. We relate this development of evergreen forests during the Pleistocene in South East Asia to the radiation of the genus Rattus which seems to have spread from a southern refuge to the north. This climatic history seems in good agreement with some climatic global models which correlate cooling and increasing humidity through the Plio-Pleistocene in South East Asia to the uplift of Tibet plateau.

CARBONIFEROUS BIOGEOGRAPHY OF AUSTRALASIA P. J. JONES\ B. A. ENGEL^ L METCALFE^ G. PLAYFORD^ J. RIGBY^ J. ROBERTS^ S. TURNER^ & G. E. WEBB^ ^ Australian Geological Survey Organisation, PO Box 378, Canberra City, ACT 2601 ^ Department of Geology, University of Newcastle, Callaghan, NSW 2308 ^ Division of Earth Sciences, School of Physical Sciences & Engineering, University of New England, Armidale, NSW 2351 Department of Earth Sciences, University of Queensland, Brisbane, QLD 4072 ^ School of Natural Resource Sciences, Queensland University of Technology, GPO Box 2434, Brisbane, QLD 4001 ^ Department of Applied Geology, University of New South Wales, Sydney, NSW 2052 ^ Queensland Museum, PO Box 3300, South Brisbane, QLD 4101 Biogeographic evidence from the major Australasian biotas are reviewed for the Early Carboniferous (Dinantian) and Late Carboniferous Silesian) subperiods. They are recorded mainly from a range of biotopes on the land, shelf margins, island arcs and basins in eastem Australia (New England Orogen), and in shallow eperic seas in intracratonic basins of the western part of the continent (Bonaparte, Canning, Carnarvon Basins). The groups studied here include ammonoids, brachiopods, bryozoans, conodonts, corals, foraminiferids, ostracods, and trilobites within the marine environment, and vertebrates, plants, spores and pollen, which spanned both continental and marine realms. Their biogeographic relationships are deduced at the biotope level (ecostratigraphy), by comparing the geographic distribution of similar communities. Limitations include gaps in the data base, resulting in monographic bias. Both faunal and floral evidence suggest a low palaeolatitudal position of the Australian block during the Early Carboniferous. Dinantian foraminiferal faunas from the Bonaparte Basin belong to the Tethyan Realm, with close connections to Eurasia, particularly Laos and Vietnam. Tournaisian conodont faunas of the near-shore {Clydagnathus) biofacies from the Western Australian Basins show links with western Europe, Russia, and the Kuznetsk Basin. In eastem Australia the Tournaisian basinal (Siphonodella) biofacies is characteristic of the same facies in South China, Europe, and the US midcontinent. In the Visean, conodont cosmopolitanism is more marked in Western Australia than in the east. Other cosmopolitan groups include the brachiopods from eastem and Western Australia, and ammonoids from eastem Australia, both with Toumaisian faunas having links with North America Europe, Russia and Kazakhstan. In the early Visean a global flooding event introduced new faunas into Australia from the same provenances. By the late Visean, during a period of diminishing diversity of brachiopod species, minor links were established with northwestern Argentina. Marine benthic ostracod faunas from Western Australia, mainly from the Bonaparte Basin, consist of cosmopolitan genera, and support the major links indicated by the brachiopods and ammonoids; some provincialism is indicated, which unite northwestem Australia, South China and parts of northem Gondwana. Early Carboniferous marine fish scales and teeth remains from Western Australia and the New England Orogen indicate connections with the US midcontinent, Europe, and Russia. Comparisons and definition of trilobite provinces within the Early Carboniferous are still conjectural issues.

Other Early Carboniferous biota show distinctive provinciahsm. The coral fauna of the Bonaparte Basin has Chinese affinities (Kueichouphyllum-faun^i), and that of eastern Australia is highly endemic. The endemic taxa were mostly confined to shallow reefal settings, whereas the cosmopolitan taxa, which are longer-ranging, is part of a deeperwater fauna. Despite the paucity of Carboniferous bryozoan faunas in Western Australia, a marked difference in the character of the Permian faunas found in both eastern and Western Australia, suggests regional differences that may have originated during the Carboniferous. The spore-pollen flora of the Australian Early Carboniferous {Granulatisporites frustulentus Microflora) has an unambiguous regional character, but contains some cosmopolitan taxa enabling correlation with Europe. The lower vascular plants {Lepidodendropsis Flora) are independent of the Eurasian floras. Non-marine fish scales and teeth remains from the Toumaisian of Victoria and Queensland (Drummond Basin) have similar biotopes in North America and Europe. Mid-Visean tetrapods from the Drummond Basin pre-date their occurrences in North America and Scotland. A rapidly diminishing faunal and floral diversity occurred in Australia in the latest Visean and earliest Namurian, heralding a major global cooling change Pennsylvanian Ice Age), marked by a major turnover of biota in the earliest Namurian. The ammonoids, 94% of brachiopod species in the latest Visean of NSW, the conodont and fish faunas, and virtually all the corals became extinct, and were replaced by new brachiopod, molluscan, bryozoan and trilobite faunas {Levipustula levis Zone). Gondwana connections with South America are indicated by the faunal similarities which existed between Australia and Argentina. A weaker, but interesting, brachiopod similarity is with the Baikal region, Siberia. The palynofloras support Australia-Argentina links, with some similarities evident from Oman on the northern Gondwana fringe.

44 VARIATION, BIOSTRATIGRAPHY AND PALAEOBIOGEOGRAPHY OF THE UPPER CRETACEOUS COILOPOCERATID AMMONITES FROM EGYPT

A.S- KASSAB Geology Department, Faculty of Science, Assiut University, Assiut 71516, EGYPT Turonian successions are widely distributed and well exposed at several localities in Egypt, including beds extremely rich in ammonites. Coilopoceratids are the most common among the ammonite builders of the Turonian carbonates. The aim of the present paper is to clarify the variation, biostratigraphic importance and palaeobiogeographic affinity and distribution of the Turonian coilopoceratid ammonites. Material for the present study was collected from Sinai and the Eastern Desert, Egypt. Based on biometrical and variational analyses, a full account of ontogenetic and intraspecific variation in populations of the ammonite Coilopoceras Hyatt is given. It is concluded that two coilopoceratid species can be easily differentiated in the material studied, namely Coilopocersas newelli Benavides-Caseres and Coilopoceras requinianum (D'Orbigny). Several species which were previously identified in the literature as separate are considered to be synonyms of the two species. Thus the study sequence is classified into two ammonite zones of Late Turonian age, from base to top: the Coilopocersas newelli interval zone and the Coilopoceras requinianum total range zone. Analysis of the palaeogeographic distribution of the studied coilopoceratid ammonites indicates that there was a sort of marine connection between Egypt and other Tethyan countries during the Upper Turonian.

THE PALAEOBIOLOGY OF THE QUATERNARY - A PARTIAL OVERVIEW A. Peter KERSHAW Centre for Palynology and Palaeoecology, Department of Environmental Science, Monash University, Clayton, VIC 3168

Geography

and

The geological position of the Quaternary period has provided data which is both advantageous to, and problematic for, the development a palaeobiological overview. On one hand, continuity with the present has promoted detailed environmental and ecological interpretation of essentially modem biotas supported by a variety of high resolution dating techniques, for the latter part of the period. On the other hand, the geological youth of the Quaternary has resulted in the neglect of conventional stratigraphic study of most of the period. The situation is also complicated, for terrestrial environments at least, by extreme climatic variability which has resulted in discontinuous fossiliferous sequences, and by a lack of consensus on a definition of the Quaternary. Formally the Plio-Pleistocene boundary is placed at around 1.8 my but it is becoming increasingly clear that this does not satisfy suggested criteria that the boundary should be clearly marked biostratigraphically and indicate the initiation of climatic deterioration which characterises the Quaternary. It is almost certain that this boundary will be lowered to approximately 2.6 Ma which coincides with the end of a major global cooling between 3 and 2.6 Ma, the Gauss/Matuyama palaeomagnetic reversal, a change in dominance from orbital precession to orbital obliquity cyclicity, the first major incursion of ice-rafted debris into the North Atlantic Ocean and with a number of biostratigraphic markers (Partridge 1997). The proposed Pliocene/Pleistocene boundary at 2.6 Ma also provides a valuable definition for the base of the Quaternary, particularly in southern Australasia. Around this time there was a sharp reduction in the extent of rainforest and a number of regional rainforest extinctions. These may have been due to a change from a summer to winter-dominated rainfall pattern as well as increasing climatic variability. The basis for the development of present sclerophyll vegetation types was also established. However, the lack of record continuity through the Early Quaternary, which falls between the temporal interests and resources of stratigraphic and evolutionary, and ecological palaeobiologists, inhibits significant insights into subsequent developments until the Mid Pleistocene. There is little doubt that major changes within the Quaternary occurred during latter part of the period. Superimposed on the extreme eccentricity-dominated cyclicity in climate were directional changes in environmental conditions which resulted in major regional and continental floral and faunal extinctions. Contributing factors may have been increased aridity, increased temperatures particularly in northern Australia, and higher levels of biomass burning possibly associated with time(s) of human arrival. Increased numbers of long, continuous palaeoecological records and improved dating of them is allowing the construction of realistic scenarios to explain late Quaternary extinctions.

SOME EARLY STUDIES IN BIOGEOGRAPHY AND PALAEOBIOGEOGRAPHY OF AUSTRALASIAN FAUNAS AND FLORAS J. G. LLEWELLYN MS 544, Clifton, QLD 4361 Early naturalists in Australasia were interested in both the unusual and the familiar types of extant and fossil flora and fauna. Robert Brown, an English botanist, spent four years in Australia and had described about 4200 species by 1810. He was, like others at that time, a creationist and, when he noticed the differences between the flora of western and eastern Australia and the similarities of the Proteacea between Australia and the Cape of Good Hope, assumed they were all different creations - biogeography did not come into it. By the 1830s evolution following on from a creation was generally accepted and it was generally believed that there had been one creation and that each species had devolved from one pair and developed, changed, migrated and on occasions become extinct, because of environmental changes. This meant that centres of creation, migrations and changes in species in both space and time had to be considered. Modern biogeography was born. Throughout the nineteenth century huge amounts of data were amassed and theories were constantly being modified. Creation, followed by evolution as designed by God, followed by evolution by competition; the acceptance of unique species followed by blurring of the demarkation lines; geological catastrophes followed by uniformitarianism; a shrinking and cooling earth followed by a growing earth from space debris; a Biblical earth a few thousand years old - then a geological earth hundreds of millions of years old - then a thermodynamic earth earth less than 100 million years old - then, in the early 20th Century, a radioactive earth thousands of millions of years old. Through all changes the theories trailed behind the facts and the facts were bent to fit each theory until the theory was abandoned or modified. One theory which did not change during the nineteenth and early twentieth centuries was that the continents and oceans were geographically fixed; and so all biogeographic and palaeobiogeographic work was done within this framework. The only movement was vertical as the continents and oceans rose and sank to preserve isostatic balance. To explain biogeographic links between continents, the idea of old sunken continents became popular. Those relating to Australasia included an Indian Oceanic continent, "Lemuria", linking Africa, India and Australia in the Palaeozoic and again in the Mesozoic to early Tertiary; a southern Pacific continent, "The Melanesian Plateau" existing in the Mesozoic linking Australia, New Zealand and the western Pacific islands; and "Gondwana", a continent linking India, Australasia, South America, Africa and Antarctica in the Palaeozoic and later in the Mesozoic. These continents provided some solutions but also posed quite a few problems and by the 1890s they had "shrunk" to land bridges and narrow peninsulas, or archipelagos, linking the land masses. In an effort to understand Australian biology and palaeontology most people working in the field were, at one time or another, biogeographers. Between the 1870s and the early 1900s there were several people active in the study of biogeography in Australia and New

Zealand. Ralph Tate and Baldwin Spencer both set out a pattern for Australian biogeography, Charles Hedley and others studied marine biogeography, Henry Dean studied flora, and Hutton studied patterns in New Zealand. All of them considered the biological and palaeontological discoveries from New Guinea, the Pacific islands, Africa, India, Malaya, Antarctica and South America. They were also aware of Northern Hemisphere efforts because, besides the personal links Australasian scientists had overseas, the various societies, museums and geological surveys around the world exchanged journals with each other - in some cases in Australia, over 1000 a year were received by individual institutions. In the early twentieth century, data continued to be collected and sorted, but existing theoretical explanations were becoming inadequate. The main interest of the time was the exploration and correlation of data from New Guinea and Antarctica. More material was also beginning to come in from South America to fit into southern biogeographic patterns.

PALAEOZOIC FISH FAUNAS OF AUSTRALIA John LONG Western Australian Museum, Francis Street, Perth, WA 6000. The oldest known accepted vertebrate fossils come from the Lower-Middle Ordovician Amadeus Basin sequence of central Australia, although possible Late Cambrian vertebrate fragments have recently been reported from the Georgina Basin (Young et al 1996). The Ordovician faunas contain the agnathan fishes Arandaspis and Porophoraspis and an assemblage of microvertebrate remains including scales of Areyongia, a possible early chondrichthyan, and the agnathan Apedolepis (Young 1997). Silurian fishes are poorly known in Australia, apart from one articulated body of an acanthodian from near Yea, Victoria (Burrow & Young 1997) and some microvertebrate assemblages from Queensland (Turner 1993). Australia has a diverse and well-preserved record of Devonian fishes. Three-dimensional preservation of skulls, and partially articulated fishes from the Early Devonian (PragianEmsian) Limestones near Taemas-Wee Jasper, Cooma (NSW) and at Waratah Bay and Buchan (Vic.) have produced a great diversity of mainly endemic taxa of placoderms, lungfishes, and isolated microvertebrate remains of sharks, thelodonts, onychodontiforms, porolepiforms and palaeoniscoids. Early-Middle Devonian boundary faunas are known from the Cobar region (NSW) and from the Toko Syncline in Queensland, dominated by the Wuttagoonaspis assemblage of endemic placoderms with rare osteichthyan and acanthodian remains. The Eifelian Hatchery Creek Conglomerate, near Canberra, has another important fauna containing the endemic antiarchs Monarolepis and Sherbonaspis (Young & Gorter 1981; Young 1988) along with our oldest osteolepiform fish, and thelodonts. The Middle Devonian - Late Devonian boundary faunas include Mt Howitt, Victoria, with over 17 taxa of endemic species of placoderms, lungfishes, acanthodians, osteolepiforms, and a primitive coelacanth (Long 1991, 1995, 1997). The famous Gogo fish sites in the northwest of W.A. have a diverse, well-preserved fauna of over 40 taxa, mostly placoderms and lungfishes, but also palaeoniscoids, onychodontiforms and rare osteolepiforms (Long 1995). Many well-preserved Late Devonian fish sites occur in the Lachlan Fold Belt of eastern Australia; most notable is the mass kill site at Canowindra (Famennian) which has yielded placoderms, the endemic osteolepiform Canowindra, well-preserved rhizodont and tristichopterid osteolepiforms (Johanson & Ahlberg 1997), and a ?rhynchodipterid lungfish. Carboniferous fishes comprising palaeoniscoids, acanthodians, rhizodonts and lungfish are well-preserved from Mansfield in Victoria (Long 1991) and from a few sites in central Queensland (Drummond Basin). Several Lower Carboniferous microvertebrate sites are known throughout Australia which also contain diverse shark faunas (Turner 1993). Permian fishes are rare, but exquisitely preserved palaeoniscoids are known from the Blackwater Shale of SW Queensland (Campbell & Phuoc 1983). The Permian of the Carnarvon Basin, W.A. has produced some Helicoprion tooth whorls and a few fish scales, teeth and spines (Long 1991).

Burrow, C J . & Young, G.C., 1997. An articulated acanthodian tail from the Late Silurian of Victoria, Australia, 16. Abstracts, Conference on Australasian Vertebrate Evolution, Palaeontology and Systematics, July 7th-llth, 1997. Western Australian Museum, Perth. Campbell, K.S.W. & Le Duy Phuoc, 1983. A Late Permian actinopterygian fish from Australia. Palaeontology 26, 33-70. Johanson, Z. & Ahlberg, P.E., 1997. New tristichopterid Osteolepiformes: Sarcopterygii) from the Mandagery Sandstone (Famennian) near Canowindra, NSW, Australia. Transactions of the Royal Society of Edinburgh: Earth Sciences Long, J.A., 1991a. The long history of fossil fish, 336-428. In P. Vickers-Rich, J.N. Monaghan, R.F. Baird & T.H. Rich (eds) Vertebrate Palaeontology in Australasia. Pioneer Design Studios with Monash University Publications Committee, Melbourne. Long, J.A., 1995. The Rise of Fishes. 500 million years of Evolution. University of New South Wales Press, Sydney, 228 pp. Long, J.A., 1997. A primitive new coelacanth (Osteichthyes: Actinistia) from the late Middle Devonian of Mt Howitt, Victoria, 41-42. Abstracts, Conference on Australasian Vertebrate Evolution, Palaeontology and Systematics, July 7th-llth, 1997. Western Australian Museum, Perth. Turner, S. 1993a. Palaeozoic microvertebrate biostratigraphy of eastern Gondwana, 174-207. In J.A. Long (ed.). Palaeozoic Vertebrate Biostratigraphy and Biogeography. Belhaven Press, London. Young G.C. & Gorter, J.D., 1981. A new fish fauna of Middle Devonian age from the TaemasAVee Jasper region of New South Wales. Bulletin of the Bureau of Mineral Resources, Geology and Geophysics (Australia) 209, 83-147. Young, G.C., 1988a. Antiarchs (placoderm fishes) from the Devonian Aztec Siltstone, southern Victoria Land, Antarctica. Palaeonto graphical^! A, 1-125. Young, G.C., 1996. Ordovician microvertebrate remains from the Amadeus Basin, Central Australia. Journal of Vertebrate Paleontology 17, 1-25. Young, G.C., Karatajute-Talimaa, V.N. & Smith, M.M., 1996. A possible Late Cambrian vertebrate from Australia. A^^mre 383, 810-812.

50 AUSTRALASIAN CAINOZOIC BIOGEOGRAPHY

Brian McGOWRAN (on behalf of: M. Archer, T. Darragh, Q. Li, P. Maxwell, B, McGowran, K. McNamara, M. McPhail, A. Partridge, J. Richardson, S. Shafik, E. Truswell & M. Warne) Department of Geology & Geophysics, The University of Adelaide, SA 5005 The environmental scenario comprises: (i) Tasman and Coral Sea spreading and accelerated Australia-Antarctica divergence in the Palaeogene; "modem" topographic relief is of late Neogene age; (ii) Global cooling, increased gradients, increased contrasts from the Subtropical Convergence to the tropics, and increased aridification of Australia all began in the late Palaeogene; (iii) Isolation of the terrestrial biotas is variously suggested to have occurred as early as Paleocene, as late as late Eocene; (iv) The southern-neritic and adjoining realms have been controlled by the interplay of warm currents with cool waters south of the STC; and (v) The Australian continent is unique in the preservation of its old and deep regolith, the long-term, strong influence of ENSO, and the poor exposure of its strata-all in contrast to New Zealand. Neritic realm, Molluscan faunas around the fragments of eastem Gondwana were provincial (with a strong trans-Tasman contrast in the late Paleocene) before a homogenizing and modernizing in the later Eocene; new provinces developed in the Oligo-Miocene before a second homogenizing in the late Pliocene. Brachiopod assemblages include cosmopolitan members (most taxa) and the terebratellids, exclusively southern and responsible for the abundant articulates in southern Australasia. The terebratellids show a particularly clear Gondwanan centre of origin. The rich bryozoan assemblages of southern Australia have close similarities with New Zealand faunas, but there are few common elements between the modem Antarctic shelf faunas and the Australian Cainozoic-Recent. The tropical Indo-West Pacific modem fauna shows several similarities with Australian groups, but generalizations on dispersal await systematic studies. Three major patterns in ostracod biogeography include (i) the evolution of new SW Pacific taxonomic clades in shallow marine environments; (ii) deep-ocean colonization when Late Cretaceous, Gondwanan, neritic clades invaded deep Pacific environments; (iii) faunas of the modem neritic IndoPacific realm came to dominate Australasian neritic environments during late Neogene cooling. Echinoids followed global trends in peaking in the later Eocene then bottlenecking across the Eocene/Oligocene boundary. After the highly endemic late Oligocene faunas, fluctuations through the Neogene are now seen more in terms of nutrient supply than water temperature (as previously), as well as immigrations from the IndoPacific region. Neritic benthic foraminifera display four phases of concurrent immigration and endemism in shallow epifaunas: later Eocene, Oligocene, early-middle Miocene and Pliocene. Brief incursions by phytosymbiotic larger species mark both warmings and an invigorated Leeuwin Current. Pelagic realm. Calcareous nannofossils with known bipolar distribution occur abundantly on the Australian southern margin but are virtually absent in NW Australia.

The abundance of warm-water species along the southern margin decreases in an easterly direction, tracing the intermittent flow of the Leeuwin Current back to the mid Eocene. In the southern Palaeogene, dinoflagellate microfloras comprise mixed cosmopolitan and cool-water taxa of the "Transantarctic Flora" with episodic incursions of the warmer-water Wetzelielloideae. A subtropical province of chorate dinoflagellate cysts is recognised in western and northeastern carbonates. Planktonic foraminifera have figured in the tracing of the equatorwards-shift by climatic belts, punctuated by warming reversals. This is seen in both species incursions and ratios of cool/eutrophic to warm/oligotrophic faunal components. The evolution and geography of marine mammals, especially the cetaceans, is driven by nutrient levels, ultimately by steepening physical gradients. Terrestrial realm. Terrestrial floras trace continental drying out in episodic and reversed changes, with overall shrinking from late Paleocene to early middle Miocene (and return in the Pliocene). Australia has moved equatorwards but floras and coals reflect more the global trajectories of environmental change. Terrestrial vertebrate faunas have long been celebrated for isolation and convergence, at first in dispersal from the north, later in vicariance and dispersal from the south. With almost no Palaeogene fossil record, indirect cladistic and molecular arguments have been invoked to address the time of isolation from Antarctica against the history of the South Tasman Rise. Evolution and geography are broadly consistent with scenarios of environmental change.

AUSTRALASIAN CAINOZOIC BIOGEOGRAPHY: ENVIRONMENTAL FRAMEWORK FOR A SOUTHERN-TEMPERATE PROVINCE ON A COOLING PLANET Brian McGOWRAN & Qianyu LI Department of Geology & Geophysics, The University of Adelaide, SA 5005 Historical background. In the period bracketed by organic evolution and continental drift, Australia was at the end of one of the lines out of the northern centre of evolution. No matter how cogent the Permo-Triassic patterns supported continental dispersal, the dominating paradigms of Cainozoic biogeography firmly supported biotic dispersal. Biotic patterns informed tectonic patterns (landbridges, successfully; continental drift, less happily). In the next period, post-drift, pre-palaeoceanography (i.e. pre-DSDP and ODP oceanic drilling), terrestrial dispersal was reassessed; marine patterns (e.g. marsupiate echinoids, penguins, planktonic foraminifera) suggested inherited coolwater biotas from the south and subsequent tropical inputs when contact was made to the north. From the early 1970s the new breakup and spreading scenarios began to inform biogeography. Since the rise of palaeoceanography there have appeared rather precise models of global cooling in steps with reversals, watermass shifts, possible eustatic sea level curves, all well-constrained geochronologically. Just now we are entering an exciting phase of ''evolutionary palaeoecology" which will draw on better and better physical scenarios through time, better and better notions of communities in stasis or at least stability through geological time scales, and a place for biogeography in all this. There is no longer a clear distinction between Hutchinson's "the environmental theatre" and "the evolutionary play". Close to the Recent and long past the breakup of Pangaea, we are rather less concerned with matching biotas to drifting and accreting continental fragments than are our pre-Pangaea colleagues. Natural divisions of the Cainozoic record. The table emphasizes the "second-order" punctuation of the Cainozoic record by coolings ("chills") giving four natural divisions. Cenozoic global transformation nine abrupt major steps including four chills high-frequency climatic oscillations at increasing amplitude IX Middle Pliocene Chill IV 2.4Ma Early Pliocene warming reversal; (?) partial icecap meltdown y i l l Latent Miocene Messinian drawdown Antarctic icecaps; lowest sea level since Gondwana ice age VII Middle Miocene C M / / / / 13.8Ma Early to middle Miocene Monterey carbon excursion punctuated warming to Miocene climatic optimum Miocene oscillation begins in late Oligocene. VI Middle OUgocene lowered sealevel event Psychrospheric ocean: glacial cycles early Oligocene V Chill II Terminal Eocene Event s.s. 33.7Ma major transformation in evolutionary palaeoecology Auversian facies shift ''The icehouse cometh." IV late middle Eocene Khirthar restoration Terminatioix of early Eoceae warm period; shortlived icecaps? ni ChiU I at early/middle Eocene boundary -49Ma Early Eocene peaks, warmth & sea level, lowest global gradients II Paleocene/Eocene boundary isotopic spike recovery and radiation in planktonic communities I Terminal Maastrichtian mass extinction etc.

southern Australia continental drying resumed Pliocene reversal-Leewin Current-warm biotas minimal strmigraphic records, all facies Bairnsdalian regression: continental drying Balcombian-Batesfordian warm & wet peak: pelagic, neritic, & terrestrial realms Oligo-Miocene oscillation begins minimal strmigraphic records, all facies Aldingan transgression on southern margin, neritic bottlenecking; land floras unresolved mesothermal rainforests first strong Leeuwin Current:warm biotas Wilson Bluff transgression. no strata dated in Australia last terrestrial contact with AntarcticaAmerica megathermal (angiosperm) rainforests brief marine transgressions in marginal facies wet-conifer forests

The Australian succession clearly parallels the global succession at this level. The changes that forced the planet into its - 5 0 myr cooling trend boil down to the closure of Tethys and other lowlatitude gateways and the opening of the circum-Antarctic oceanway and other high-latitude gateways. The interval labelled as the "Khirthar Restoration" marks the most profound division of the Cainozoic environmental and biotic successions, the natural split between the old prepsychrospheric world and the new psychrospheric, icecap-prone world. The reconstruction of various geochemical trends have revealed a wholesale reorganization of oceanic geochemistry at the same time, christened the "Auversian facies shift". Punctuation not diachrony. It is intuitively appealing to conceive of a cooling planet forcing climatic/watermass belts towards the equator, thus generating a strongly diachronous pattern. This notion has informed the character of Australian neritic faunas warming through time and the seemingly non-isochronous surfaces of bioevents (speciations and extinctions) used in biostratigraphy. It has influenced the perceptions of palaeobotanists and carbonate sedimentologists. Australia moved towards and into the tropics through -30° of latitude whilst the planet cooled, thus balancing two opposing trends to generate a remarkable steady-state environment on this continent through the Cainozoic era. Not so: diachrony is a hangover from Lyellian gradualism. History of all kinds is strongly punctuated. Ocean-floor spreading and continental geodynamics are episodic. Global climate changes from one state to the next by rapid shifts. The record of broad changes in global sealevel, most recently from the Cretaceous high to the Neogene low, is punctuated by numerous, globally synchronous depositional sequences. The Phanerozoic fossil record is punctuated by mass extinction; other theories include "turnover pulses" (clustered changes punctuating evolutionary stasis and triggered by abrupt climatic change) and "coordinated stasis", a pervasive temporal pattern of blocks of community stability. Boucot has called this alternation of prolonged stability and rapid collapse "reconciling d'Orbigny with Darwin". Whilst some of this is still controversial (is "coordinated stasis" merely "coincident relative stability"?) there is no comfort there for diachrony. A better model is stepped environmental shifts forcing isotherms to advance and retreat through tens of degrees latitude rather quickly (see figure). Within each block of Cainozoic time there will be lower-order reversals which, in southern Australia, are the outcome of a waltz beween the warm marginal currents (east and west where the Leeuwin is defiantly counter-gyral) advancing and retreating, and the Subtropical Convergence retreating and advancing. This waltz began during the Khirthar restoration ("the icehouse cometh") and became more animated with each step towards the late Neogene ice ages. We believe that biogeographic pattems under that control will be essentially isochronous and found in all realms-neritic, oceanic and terrestrial. Southern Australasia neither is the passive recipient of terrestrial emigrations from the northern hemisphere, "where all the action is" in the old perceptions, nor merely the accumulator of spinoffs from the IndoPacific marine province. This region has been embroiled in the most profound of the forces driving global environmental transformation during the Cainozoic-AustraliaAntarctica divergence and the growth of the West Wind Drift and Circum-Antarctic Current. McGowran, B., Li, Q. & Moss, G.D., 1997. The neritic carbonate record in southern Australia: the biogeohistorical framework. In Coolwater carbonates in spece and time, N.P. James & Clarke, J.D (eds). Society of Economic Paleontologists and Mineralogists, Special Volume 56, 185-203. McGowran, B., Li, Q., Padley, D., McKirdy, D.M., Cann, J. & Shafik, S. (in press). The Leeuwin Current at geological time scales: biogeographic impact on southern Australia. Palaeogeography, Palaeoclimatology, Palaeoecology. McGowran, B. & Li, Q. (in press). Cainozoic climatic change and its implications for understanding the Australian regolith. Proceedings of Regolith 96. Austraian Journal of Earth Sciences, Special Volume.

MIOCENE 0

OLIGOC 20

EOCENE 40

PALEOC 60

70 Ma

20°C

70°-

(Upper) Mcdem oceanic pattern. During late Neogene glacials the Conveigence moves into the Bi^t and the Leeuwin Current shuts down. (Lower) Australia's migration on a time-latitude grid. Three oceanic grossplot isotherms indicate three wamiings at this timescale: Paleocene to eaiiy Eocene, later Eocene, eariy -middle Miocene. Dotted: trajectory of western Bight

54 PERMIAN-TRIASSIC FLORAS OF EAST ANTARCTICA: PALAEOENVIRONMENTAL SETTING AND BIOGEOGRAPHIC AFFINITIES

Stephen McLOUGHLIN^ Andrew N. DRINNAN^ & Sofie LINDSTROM^ of Botany, The University of Melbourne, Parkville, VIC 3052. 2-JSchool Division of Historical Geology and Palaeontology, Department of Geology, Lund University, Solvegatan 13, S-223 62 Lund, SWEDEN. Mid-Permian to Upper Triassic sedimentary rocks of the Amery Group represent the only Palaeozoic-Mesozoic sedimentary sequence exposed in East Antarctica outside the Transantarctic Mountains. The Amery Group sediments in the Prince Charles Mountains were centrally located within the Gondwanan supercontinent. The fossil biota of the Amery Group is thus important for correlation of strata between the now dispersed Gondwanan landmasses, for interpretation of interior Gondwanan palaeoclimates and environments, and for understanding the nature of high-latitude terrestrial biotic turnovers at the Permian-Triassic boundary. The Permian fossil assemblages of the Radok Conglomerate and Bainmedart Coal Measures (lower and middle Amery Group) are dominated by low-diversity, glossopteriddominated, coal-forming mire floras. Cordaitaleans, ferns, herbaceous lycophytes and sphenophytes represent minor components of the flora. Faunal remains are limited to very scarce exoskeleton fragments of arthropods, insect larval borings in wood, and simple sinuous and irregular sand-filled trails and burrows in floodbasin sediments. Analysis of growth rings in glosspterid woods suggest that the climate was strongly seasonal and that there were significant intraseasonal perturbations to growth conditions evidenced by numerous false rings. The biogeographic affinities of the fossil macrobiota are indistinct, but the palynoflora shows some similarities with other southern Gondwanan assemblages, notably from east Africa, India, the Transantarctic Mountains, and Western Australia. The Permian-Triassic boundary is marked by the termination of coal sedimentation, extinction of glossopterid and cordaitalean gymnosperms, and the appearance of peltaspermalean gymnosperms (Lepidopteris) and a range of herbaceous lycophytes. Drier climates developed in Middle Triassic times as evidenced by calcrete crusts and desiccation cracks in well-developed red-bed sequences. Upper Triassic macro- and palynofloras from the Flagstone Bench Formation (upper Amery Group) show an increase in floristic diversity. These floras incorporate a range of conifers, corystosperms, ferns, lycophytes, sphenophytes, and algae. Late Triassic conifer woods show well-defined growth rings reflecting seasonal conditions, but no statistical analysis of the growth rings has yet been carried out for comparison to the Late Permian woods. The Late Triassic (Norian) palynofloras contain Tethyan elements and show closest similarities to assemblages from India and northwestern Australia.

AUSTRALIAN-SOUTHEAST ASIAN BREAKUP IN RELATION TO BIOGEOGRAPHIC PATTERNS IN THE REGION

Ian METCALFE Division of Earth Sciences, University of New England, Armidale NSW 2351. East and Southeast Asia comprises a complex assembly of allochthonous continental terranes, the boundaries of which are marked by suture zones that represent the remnants or sites of former ocean basins that once separated the now juxtaposed pieces of unsubductable continental lithosphere of the terranes of the region. Biogeographical and other data suggest that all of the principal East and Southeast Asian continental terranes were located on the Himalayan-Australian margin of Gondwanaland in the Early Palaeozoic and that they formed a "Greater Gondwanaland". Close faunal affinities between the east and Southeast Asian terranes and with Australia occur in the Cambrian, Ordovician and Silurian, and distinctive Asian-Australian faunal provinces can be recognised during this time interval. Multidisciplinary data (eg. biogeographic, biostratigraphic, palaeomagnetic, structural/tectonic) suggest that the Asian terranes rifted and separated from NW Australian Gondwanaland as three continental slivers in the Devonian, late Early Permian, and Late Triassic-Late Jurassic. During the separation of these continental slivers, three ocean basins, the Palaeo-Tethys, Meso-Tethys and CenoTethys, are interpreted to have opened between each successive sliver and Australian Gondwanaland. The first sliver, interpreted to have separated in the Devonian, included North China, South China, Tarim and Indochina. Changing biogeographic patterns in the Devonian on these separating terranes can be explained by changing continent-ocean and land-sea configurations resulting from the rifting processes and separation from Gondwanaland. By Carboniferous times, faunas and floras of terranes that formed the first sliver to separate had no Gondwanaland connections, and palaeomagnetic data indicate that these terranes were separated from the parent craton. The clockwise rotation of Gondwanaland in the Late Carboniferous sent Australia into high southern latitudes and the Gondwanaland glaciation affected much of Australia during the Late Carboniferous and Early Permian. During this time, glacial ice reached the marine environment of the northeast Gondwanaland margin and glacial-marine sediments were deposited on the Gondwanaland margin including the Sibumasu, Qiangtang and Lhasa terranes now located in East and Southeast Asia. Typical Gondwanaland fossils are also found on these terranes in the Lower Permian and cold-water faunas and cold water oxygen isotopic signatures are associated with the Lower Permian glacial-marine sediments. Early Permian rifting led to the separation of the Sibumasu and Qiangtang terranes from the Himalaya-NW Australian margin of Gondwanaland at the end of the Early Permian. These terranes drifted rapidly northwards during the Permian and significant changes in both brachiopod and fusulinid biogeographic provinces on these terranes during the Permian are interpreted as being primarily due to the separation and northwards drift of the Cimmerian continental terranes, coupled with climatic amelioration following retreat of the Gondwanaland glaciation. Rifting and separation of a third continental sliver from Gondwanaland began in the Late Triassic with the separation of the Lhasa block and progressed eastwards with the Late Jurassic separation of West Burma and the small Sikuleh and Natal terranes of SW Sumatra and other small microcontinental fragments now located in Borneo and West Sulawesi. Following the final breakup of Gondwanaland, both India and Australia travelled northwards to collide with Eurasia.

India is here interpreted, based on biogeographic and palaeomagnetic data, to have had its initial contact with Eurasia at around 60 Ma. Initial collision of Australia with the Philippine Sea plate of Southeast Asia occurred at about 25 Ma. A series of palaeogeographic reconstructions depicting the breakup of Australian Gondwanaland during the Phanerozoic and the relationships of these events to changing biogeographic patterns is presented.

EXAMINING RELATIONSHIPS OF THE MONOTREMES IN A BIOGEOGRAPHICAL CONTEXT A.M, MUSSER School of Biological Science, University of New South Wales

Monotremes are the egg-laying mammals - the platypus and echidnas - that today are found only in Australia and New Guinea. However, this ancient subclass was previously much more diverse, and so far as is known at least the platypus family (Omithorhynchidae) occupied a much more extensive range than does the living representative, Omithorhynchus. Monotremata is comprised of four families: the semiaquatic, carnivorous Steropodontidae and Omithorhynchidae; the possibly semi-aquatic shellfish specialist Kollikodontidae; and terrestrial ant-, termite- and worm-eating echidnas, or Tachyglossidae. The monotypic Steropodontidae and Kollikodontidae are known only from early Cretaceous (110 Ma) opal-bearing deposits at Lightning Ridge in New South Wales, Australia. The oldest known omithorhynchid, Monotrematum sudamericanum, was recovered from early Paleocene (61-63 Ma) deposits from Patagonia, Argentina, a find that confirms the Gondwanan nature of the monotremes and indicates that omithorhynchids occupied Antarctica in addition to Australia and South America. In Australia omithorhynchids are known from the Oligocene to the present, with Omithorhynchus restricted to the green eastem fringe of the continent. Tachyglossids are known from the Miocene to the present, and are found in both Australia {Tachyglossus) and New Guinea {Tachyglossus and Zaglossus), The historical distribution of the monotremes is plotted and potential dispersal routes, vicariance events and the effects of drift on monotreme populations are examined. The contraction in range for omithorhynchids, driven by aridification of the Australian continent and the isolation and subsequent chilling of the Antarctic continent, is put into a contemporary perspective urging conservation measures to protect the living species from further habitat loss. Palaeoecology of fossil sites yielding monotreme material illustrates tremendous change throughout the extraordinarily long history of this enigmatic mammalian group. Theories on higher-level monotreme relationships are reviewed and distributions of proposed sister taxa - therian mammals, multituberculates and marsupials - are compared to that of the monotremes.

REVISED ORDOVICIAN BIOSTRATIGRAPHY OF NEW SOUTH WALES Ian G. PERCIVAL\ Barry WEBBY^ & Yongyi ZHEN^ ^Specialist Services Section, Geological Survey of New South Wales, P.O. Box 76, Lidcombe NSW 2141 ^Centre for Ecostratigraphy and Palaeobiology, School of Earth Sciences, Macquarie University NSW 2109 The oldest Ordovician rocks occur in the far western region of the state, where Early Ordovician strata conformably succeed latest Cambrian rocks containing Cordylodus proavus in the Koonenberry Belt (=Gnalta Shelf) NE of Broken Hill. Complete details of the stratigraphy of this area remain to be elucidated, but elements of the basal Ordovician C. lindstromi conodont zone have been identified in the Kandie Tank Limestone south of Kayrunnera (P. Kruse, pers. comm.). The youngest Ordovician horizons presently known in the Koonenberry Belt are found in the Tabita Formation at Mt Arrowsmith; descriptions and illustrations of the conodonts have never been formally published, but it seems that these faunas are only late Bendigonian or slightly younger (possibly early-middle Arenig) in age. Thus there is a slight overlap with the oldest Ordovician strata in the Lachlan Fold Belt of central NSW, where early Bendigonian graptolites occur in both the Yarrimbah Chert of the Parkes Platform, and the Hensleigh Siltstone on the northern Molong High. A significant conodont fauna dominated by Bergstroemognathus extensus and Juanognathus variabilis, which is newly recognised from beds immediately underlying the graptolites in the Hensleigh Siltstone, indicates contemporaneous deposition with the Koonenberry Belt, and can be more widely correlated with the early Bendigonian Emanuel Formation (Canning Basin) conodont assemblages B to D, equivalent to the Prioniodus elegans Baltoscandian zone, the upper part of the North American Acodus deltatus - Oneotodus costatus zone, and conodont zone D of the Utah succession. A substantial hiatus spanning the Castlemainian, Yapeenian and early Darriwilian then ensued, during which time apparently no shallow water deposition took place across the state (or if it did no trace now remains). All known Middle Ordovician shelly faunas in this region of the Lachlan Fold Belt are Da3 or younger, as is also the case for the New England Fold Belt. The almost continuous biostratigraphic succession above this point is first developed in limestones in the lower Goonumbla Volcanics on the Parkes Platform, from where John Pickett has reported Pygodus anserinus and P. serra, indicative of Fauna 5-6 in the North American zonation. Equivalent or very slightly younger strata now being studied on the northern Molong High yield conodont faunas, including the distinctive Appalachignathus, of latest Darriwilian to early Gisbomian age. These faunas occur both in very shallow water environments, and in allochthonous limestones within deep water facies equivalents. A late Gisbomian conodont fauna, obtained from a limestone member higher in the Fairbridge Volcanics, fills in the zonation beneath the early to late Eastonian shelly faunas which have been well-documented over the past 30 years on the Molong High and Parkes Platform. Extensive areas of Eastonian strata are now also known from the Capertee Volcanic Rise, based on the occurrence of species of Taoquopognathus.

^ IGP publishes with permission of the Director General, NSW Department of Mineral Resources. ^ A contribution of IGCP 410, "The Great Ordovician Biodiversification Event"

SILURIAN CORALS OF AUSTRALASIAN ASPECT John PICKETT Geological Survey of New South Wales, P.O. Box 536, St Leonards NSW 2065 In the Silurian of Australia, the Tabulata are represented by perhaps twenty-eight genera, and the Rugosa by forty-five. All of the tabulates are widely distributed, except for the endemic Hattonia which occurs in the the Late Silurian and Early Devonian in NSW and Queensland. Three genera of rugosans are monotypic, endemic, and their occurrences are restricted to the type localities: Angullophyllum, Coronoruga and Mictocystis, Of the remaining rugosans, ten show promise as indicators of regional character: Nipponophyllum, Zenophila, Mucophyllum, Toquimaphyllum, Labechiellata, Idiophyllum, Nanshanophyllum, Stylopleura, Yassia and Zelolasma. The pattern of coral distribution favours the configuration of crustal blocks adapted by Rong et al (1995) after Scotese & McKerrow (1990), particularly in reference to Siberia, since the 90"^ rotation shifts coral faunas in the Inner Mongolian area away from almost 60°N to a more rational 35^N. During most of the Silurian, Australia lay, as now, on the western margin of a protoPacific ocean, with no known land masses of appreciable size immediately to the east. In the west, the crustal blocks in closest proximity were, according to the reconstruction of Rong et al.. South China, the Borneo-Malayan block, Tibet and Indo-China. Significantly, at least seven, and possibly eight of the selected ten genera have their first occurrences on either the Tibet, South China or Australian blocks. This pattern correlates well with that of the Retziella brachiopod fauna of Rong et al (1995), and suggests that the area may have been a centre of origin for coral genera through most of the Silurian and into the Devonian. However, the more distal occurrences of the coral genera lie in areas in which other brachiopods faunas have been recognised, notably the subrianids. If west-flowing North and South Equatorial currents were operative during the Silurian, this would have hindered eastward migration. Kazakhstan and the extreme northern part of the north American plate were the next in a westerly direction, so it is possible that the equatorial currents account for the migration of Australasian elements to the Canadian Arctic, California and the Great Basin. The revised positions of Siberia and Kazakhstan also better fit the coral distribution. Rong J.-y., Boucot, AJ., Su Y.-z. & Strusz, D.L., 1995. Biogeographical analysis of Late Silurian brachiopod faunas, chiefly from Asia and Australia. Lethaia 28, 39-60. Scotese, C.R. & McKerrow, W.S., 1990. Palaeozoic world maps and symposium introduction. In: Palaeozoic Palaeogeography and Biogeography. Geological Society of London Memoir 12, 1-21. ^ Published with permission of the Director General, NSW Department of Mineral Resources.

PALAEOGEOGRAPHIC EVOLUTION OF THE AUSTRALASIAN REGION SINCE THE EARLY NEOPROTEROZOIC C.McA. Powell & Z.X. Li Tectonics Special Research Centre, Department of Geology & Geophysics, The University of Western Australia, Nedlands, Western Australia 6907 Australia's palaeogeographic evolution in the last billion years has been influenced by its position in two supercontinents, Rodinia and Gondwanaland, and its palaeolatitude. The oldest supercontinent, Rodinia, assembled during the late Mesoproterozoic along sutures represented by fold belts extending from the Grenville Province in eastern Laurentia through the Eastern Ghats of India into the Albany-Fraser-Musgrave fold belts in Australia. There may also have been a Rodinian suture between northern Australia and the Western Australian craton along the northern margin of the Pilbara block. In Rodinia, Australia lay adjacent to Laurentia along the Tasman Line, possibly with South China in between. Australia was in mid-northern latitudes for most of the Neoproterozoic, though by the late Neoproterozoic, the equator lay close to what is now the southern margin of Australia. Australia's Neoproterozoic palaeogeography is dominated by the shallow intracratonic Centralian superbasin, which may have extended into western Laurentia, and, in the latter half of the era, by low-latitude glaciation. The breakup of Australia's eastern Rodinian margin, around the end of the Sturtian glaciation, gave rise to the Palaeo-Pacific Ocean. Closure in the late Neoproterozoic of the Mozambique and other oceans between Australia-East Antarctica-India and the Brazilian craton gave rise to the fleeting supercontinent Pannotia. The second supercontinent to influence Australia's palaeogeography was Gondwanaland, which formed at the beginning of the Phanerozoic, possibly as a result of the separation of Laurentia from Pannotia to create the early lapetus Ocean. In the Early Cambrian, the eastern margin of Australia lay along the Tasman Line facing the Palaeo-Pacific Ocean, with all other parts of Australia lying in the interior of the supercontinent. Breakup in northwestern Australia, with possibly the Tarim block separating from the Kimberley, was related to the global continental breakup around the Precambrian/Cambrian boundary, and gave rise to the Antrim Plateau basalts and the Bonaparte basin. Westward subduction of the Palaeo-Pacific Ocean along the eastern margin of Australia-Antarctica commenced during the Early Cambrian in northern Victoria Land and in the Middle Cambrian in South Australia, and led to the Cambro-Ordovician Ross-Delamerian Mountains. A new palaeogeographic cycle began in the Ordovician, with retreat of the magmatic arc from Australia's eastern continental margin, as a marginal sea and offshore island arc developed. A shallow seaway across Australia in the Late Cambrian persisted into the Ordovician, but gradually the sea regressed, and desert-like conditions formed in Central Australia and the adjacent Canning basin. Silurian to mid-Devonian was an interval of rapidly changing palaeogeography in eastern Australia as carbonate platforms flanking narrow, deep volcanogenic troughs formed in a dextral transtensional tectonic setting. On the Australian craton to the west, terrestrial desert conditions prevailed, with some shallowmarine areas in the Bonaparte and Canning basins.

Widespread deformation in Tasman orogenic zone in the Middle Devonian to Early Carboniferous, was accompanied by the development of an Andean-style magmatic arc along the Pacific continental margin of Australia, and big foreland and epicontinental basins in Central Australia. The biggest Phanerozoic mountain-building stage in Central Australia occurred in the mid-Carboniferous, as part of a world-wide Variscan orogenic episode associated with the collision of Gondwanaland with Laurussia. The Pangean supercontinent was completed in the earliest Permian by the collision of Siberia with Laurussia along the Urals. During the Cambrian to Early Carboniferous, Australia lay mainly in tropical latitudes, drifting slightly southward. In the late Visean, a rapid polar shift occurred, and by the Westphalian the South Pole lay less than 20"" south of southeastern Australia. The Late Carboniferous and earliest Permian palaeogeography of Australasia was dominated by glacial conditions. Large ice sheets covered much of interior Australia, though by the Asselian, these were diminishing and a brief shallow-marine transgression occurred in many marginal basins. Transtensional basins associated with dextral oroclinal shear along the Panthalassan eastern margin of Australia developed in the Late Carboniferous and persisted until the Late Permian, when an Andean-style magmatic arc was reestablished. Large foreland basins inboard of the Late Permian to Early Triassic magmatic arc accumulated major coal deposits during Late Permian volcanic phases, but drastic climatic changes at the end of the Permian, possibly caused by global greenhouse conditions, led to red-bed deposition in the Early Triassic. Pangea began to rift in the mid-Triassic, and by the Late Triassic, the Cimmerian blocks, which lay off northwestern Australia throughout the Palaeozoic, had left the northern margin of Gondwanaland. A new Andean-style continental magmatic arc became established along the Pacific margin of Australia, and continental sedimentation occurred in interior Australia. The palaeolatitude of eastern Australia remained high. The breakup of Gondwanaland began between East (Australia-Antarctica-India-Madagascar) and West (South America-Africa) Gondwanaland in the Callovian, with breakup between AustraliaAntarctica and the northern part of Greater India commencing in the late Tithonian. Continental extension between Australia and Antarctica began in the Late Jurassic and continued until mid-Cretaceous. The Andean-style magmatic arc along the Pacific margin of Australia also persisted until this time, although it may have been under extension in the Cretaceous. An important change in palaeogeography occurred in the mid-Cretaceous. For most of the preceding 400 million years, the eastern margin of Australia had been occupied by a magmatic arc either on the continental margin or inferred to be nearby off-shore separated by a marginal sea from the Gondwanan landmass. At 96 Ma, seafloor spreading began between Australia and Antarctica, and the continental magmatic arc rotated eastward from Australia, initially as part of the Lord Howe Rise (LHR), and later, when the LHR split to form an interarc basin that grew into a marginal sea, as an island arc now recognised as Tonga-Kermadec. On continental Australia, andesitic volcanism ceased, and the great foreland basin that had spread across Australia's interior dried out. Australia's palaeolatitude was high throughout the Cretaceous, but, owing to global greenhouse conditions, no ice sheets were developed.

At the beginning of the Palaeogene, Australia commenced its northward drift. Seafloor spreading between Australia and Antarctica was at first slow, with a narrow Red Sea style of opening, and continental connections between Australia and Antarctica, throughout most of the Palaeogene. Around 45 Ma, the rate of spreading between Australia and Antarctica increased sharply, possibly driven by the slow-down in India's northward movement related to its collision with Asia. By the mid-Oligocene, the Southern Ocean connected freely with the Pacific Ocean south of Tasmania, and the circum-Antarctic current became established, thereby triggering glaciation in the Antarctica. Northern Australia reached the tropics by the beginning of the Miocene, and Australia has progressively moved northwards at 7 to 8 cm each year since. Palaeoclimatic and palaeoenvironmental conditions in Australia in the Neogene can be related partly to this northern drift, and also to global climatic changes associated with the Pleistocene Ice Age.

EASTERN AUSTRALIAN SILURIAN GRAPTOLITES IN TIME AND SPACE R.B, RICKARDS Department of Earth Sciences, University of Cambridge, Downing St., Cambridge CB2 3EQ, ENGLAND Evidence is adduced to demonstrate that the long-held view that planktic graptolites were open ocean plankton, and evolved in that niche, is flawed. It is considered that most evolution of graptoloids occurred on shelf seas, not so far removed from palaeoenvironments occupied by benthic graptolites. A series of environments are envisaged: near shore, off shore open shelf, off shore relatively deep troughs, shelf edge and upper shelf slope. The nutrient supply necessary to support such planktonic niches is considered to be in accord with present day nutrient supply systems. Examples will be chosen from parts of the European Silurian, the Argentinian Ordovician of San Juan Province, but primarily from the Melbourne Trough, the Lachlan Fold Belt east of the Wagga Metamorphic Belt and the Graveyard Creek Subprovince. Some attempt will be made to equate the latest hydrodynamic research on graptolites, using the Laser Doppler Anemometry, to the hydrodynamic ecosystems envisaged.

PRELIMINARY BIOSTRATIGRAPHIC ANALYSIS OF SNAKES FROM THE TERTIARY OF RIVERSLEIGH, QUEENSLAND J.D. SCANLON*, M. ARCHER, S. HAND & H. GODTHELP School of Biological Science, University of New South Wales Representatives of four families of snakes are known from the Riversleigh deposits: Typhlopidae (presumed Ramphotyphlops spp.); Madtsoiidae (at least six species, genera Yurlunggur, Wonambi, Nanowana and an undescribed new genus); Pythoninae {Booidea\ at least three species, comparable to members of Liasis, Morelia and Python); and Elapidae (at least three species, comparable to terrestrial Hydropheinae but not referable to extant genera). Snake remains are known from at least 37 sites in the Riversleigh area, ranging in age from late Oligocene (System A of Archer et al. 1989) to Pleistocene, according to current provisional estimates of stratigraphic sequence and age. Typhlopids and madtsoiids occur in Systems A, B and C; pythons in A, B, C and the Pliocene Rackham's Roost deposit; elapids mainly in System C, with one occurrence in System B (RSO Site) and a Pleistocene record from Atomic Fissure LF. While madtsoiids are known from the Eocene of Murgon, southeastern Queensland (Scanlon 1993), Riversleigh has the oldest known Australian records of the other families. Identifications below the family level are not currently possible for most occurrences of Typhlopidae, Elapidae and Pythoninae, because of conservative vertebral morphology and the rarity of diagnostic cranial material. Within Madtsoiidae, four groups (genera) can be readily distinguished on the basis of vertebrae. Yurlunggur and Wonambi occur in Systems A, B and C, while Nanowana is known in B and C only, and an undescribed new genus in B only. Sympatry of congeneric species can be shown for Yurlunggur and Nanowana, but most sites have not produced material diagnostic at the species level. Within Wonambi, material from White Hunter Site (System A) and most System B sites is relatively uniform, while Ringtail Site (System C) has relatively large vertebrae which may represent a distinct species. All Riversleigh Wonambi, however, are much smaller than the late Pleistocene W, naracoortensis. The undescribed new genus is apparently rare, with identifiable remains known so far from Greaser's Ramparts Site (thought to be equivalent to System B), and Upper Site (System B). Analysis at the family level does not demonstrate any major changes in taxonomic composition of the fauna through time at Riversleigh, but some turnover at lower taxonomic levels is indicated. The fossil record here and overseas is consistent with a model based on mid-Tertiary dispersal from Asia to the Australian region in the case of pythons, elapids and typhlopids.

CHANGHSINGIAN (LATE PERMIAN) BRACHIOPOD PALAEOBIOGEOGRAPHY Shuzhong SHEN, N.W, ARCHBOLD & G.R- SHI School of Aquatic Science and Natural Resources Management, Deakin University, Rusden Campus, 662 Blackburn Road, Clayton, VIC 3168 Based on an analysis of the occurrences and composition of brachiopod faunas, the authors recognize that Changhsingian brachiopods occur in northern and southern Japan, South China, northern, southern, central and eastern Xizang (Tibet), northwestern China, Transcaucasia, Iran, the Salt Range of Pakistan, Kashmir, northwest Nepal, north and northwestern Thailand, northern Vietnam, Central Slovenia, Hungary, Italy, Greece and New Zealand. There is no boreal-type brachiopod faunas of the Changhsingian known at present. The Gondwana-type brachiopod faunas of the Changhsingian are only reported from New Zealand. The Changhsingian brachiopods in the Austrazean Province mainly consist of species of the genera Aperispirifer, Spiriferella, Tomiopsis and Notospirifer. The Tethyan Province can be further divided into three subprovinces based on the compositions of brachiopod faunas during the Changhsingian. The northern Tethyan Subprovince may only include the Kitakami massif of northern Japan, which is characterised by the coexistence of Megousia, Tschemyschewia, Orthothrix, Paramarginifera and Eolyttonia. The Cathaysian Subprovince includes Yangtze Block, Qiangtang Terrane, Qaidam Basin, Maizuru Belt of southern Japan, Caucasian Arc, Iranian Block, Indochinese Arcs and Chisos Arc. The occurrences of Peltichia, Araxathyris, Prelissorhynchia and Janiceps characterize this subprovince. It seems that there is no clear distinction in brachiopod composition to separate the western Tethyan Subprovince and eastern Tethyan Subprovince during the Changhsingian. The southern Tethyan Subprovince mainly covers the Sibumasu Arc and the Peri-Gondwana regions including the Salt Range of Pakistan, Kashmir, southem Xizang and northwest Nepal. The Changhsingian brachiopod faunas in this subprovince are usually yielded just below the Triassic Otoceras Bed and characterised by the coexistence of Spiriferella, Neospirifer, Waagenoconcha with some small elements such as Waagenites, Hustedia, Crurithyris and Martinia etc.

MIDDLE-LATE CAMBRIAN TRANSITION TRILOBITES OF AUSTRALOSINIAN ASPECTS IN SOUTHERN FRANCE John SHERGOLD, Raimund FEIST & Daniel VIZCAINO Institut des Sciences de I'Evolution, University of Montpellier II, 34095 Montpellier, FRANCE A diverse trilobite association has been recovered from a limestone coquina in the basal Val d'Homs Formation at Ferrals-les-Montagne, Montagne Noire, southern France. Taxa provisionally assigned to Ammagnostus, Prochuangia, Procemtopyge, Paraacidaspis, Shengia, Abharella, ''Stigmatoa"' and "Olentella'' occur with the previously described Bergeronites latifalcatus Feist & Courtessole and Prochuangia gallica Feist & Courtessole at this locality. Whereas no North European or American elements are represented, the assemblage is related to faunas from South-Central China, Central Australia, Iran, Kazakhstan and southern Siberia. In comparing the variously interpreted range charts of related forms it is difficult to assess the precise age of the French fauna. It certainly supersedes levels with the last Middle Cambrian Paradoxides that are already associated with Asian invaders such as Chelidonocephalus and Dorypyge. Although Bergeronites and Proceratopyge are reported from both latest Middle and Late Cambrian, the other taxa so far are known from early Late Cambrian strata. Some, such as Prochuangia, Proceratopyge and Stigmatoa range quite high in the Late Cambrian. As index agnostoids such as Lejopyge laevigata, Glyptagnostus stolidotus and G. reticulatus are lacking, a precise age will only become apparent when the systematic palaeontology and exact ranges of the taxa are fully explored region by region. Currently, we think the assemblage may be correlated with that of the Liostracina-Chatiania Assemblage-Zone of South-Central China and the Glyptagnostus stolidotus Zone of Australia, on the basis of the occurrence of Ammagnostus, but it could range as young as the G. reticulatus Zone. At first sight, the Ferrals fauna appears to represent a mixture of Middle and Late Cambrian trilobites from more than one biostratigraphical zone. Generic associations described elsewhere, such as the occurrence of Bergeronites and Prochuangia on the same pieces of rock, lack of intraclasts and authigenic minerals, suggest an absence of reworking and an autochthonous deposit. The fauna belongs to the subtropical, shallow carbonate platform, agnostoid-poor biofacies that was also developed in Iran (Mila Formation) and South-Central China (Laochatian Formation of western Hunan and eastern Guizhou). According to these data, the Montagne Noire must have probably remained within the warm-water latitudinal belt at the Middle-Late Cambrian transition, with the implication that formerly developed models of the high latitudinal position of this area should perhaps be revised.

TERRANE RAFTING ENHANCED BY CONTEMPORANEOUS CLIMATIC AMELIORATION AS A MECHANISM OF BIOGEOGRAPHICAL VICARIANCE: PERMIAN MARINE BIOGEOGRAPHY OF SE ASIA G.R. SHI School of Aquatic Science and Natural Resources Management, Deakin University, Rusden Campus, 662 Blackburn Road, Clayton, VIC 3168 Permian marine sequences and invertebrate faunas are widely distributed in all mainland terranes of SE Asia. A review of the spatial and temporal distributions of all major Permian marine invertebrate groups in this region, reinforced by the results of recent Permian stage-by-stage statistical analyses of western Pacific brachiopods, reveals that three biotic provinces are present in SE Asia during the Permian. The Cathaysian Province occupied the Simao, Indo-China and East Malaya blocks throughout the Permian. The Sibumasu Province of the Shan-Thai terrane (s.s.), Tengchong and Baoshan blocks developed in Late Sakmarian and continued to exist until, probably, the end of Midian when the same blocks joined the Cathaysian Province. Throughout this period, the Sibumasu Province carried a transitional fauna, which showed a progressively stronger affinity to the Cathaysian Province in the north and increasingly weaker affinity to the Gondwanan provinces (Westralian and Austrazean) in the south. From Asselian to Early Sakmarian, the Shan-Thai terrane, Tengchong and Baoshan blocks belonged to the shortlived Indoralian Province, which then also included Australia, India, the Himalayan and Lhasa terranes. The marked change of marine provinciality of the Shan-Thai terrane (s.s.), Tengchong and Baoshan blocks cannot be explained by the tectonic vicariance (rift-drift) model alone, nor can it be accounted for solely by migration of climatic zones. An interplay of both of these factors during the Permian is considered to be the most likely cause responsible for this marked change of marine provinciality of these blocks.

PALAEOGEOGRAPHY OF UPPER TRIASSIC TETHYAN REEF FAUNAS FROM THE NORTHWESTERN AUSTRALIAN SHELF George D. STANLEY, Jr. Department of Geology, University of Montana, Missoula, MT 59812, USA One of the big surprises in oceanographic investigations came this decade as a result of ODP Leg 122 drilling, AGSO dredging and seismic exploration on the northwestern Australian shelf. Discoveries from the northern Exmouth Plateau (Wombat Plateau) and the nearby Rowley Terrace (offshore Canning Basin) provided the first examples of reef development and tropical marine faunas yet known from Triassic rocks of the Australian region. Well-dated Upper Triassic (Norian-Rhaetian) carbonates from the ODP site 764 core provided details of a new reef complex and associated shallow-water carbonate rocks which developed along a rifted continental margin of northeastern Gondwana. Seafloor spreading and rifting began in Late Triassic and continued into Jurassic and Cretaceous times. Carbonate-dominated Upper Triassic rocks from the Rowley Terrace and the Exmouth Plateau reach more than 800 m in thickness. One patch reef complex, over 200 m thick, developed above thicker and older Upper Triassic rocks of marine and fluviodeltaic origin. Patch reefs propagated during Late Triassic sea-level rise and were terminated at the end of the Triassic (Rhaetian), but overlying Lower Jurassic rocks record continued carbonate deposition with transition to siliciclastic rocks in Toarcian to Oxfordian time. Calcisponges and corals constructing the reef were supplemented by disjectoporoidhydrozoan communities. Paleoecologic reef stages are related to biotic interactions and sea-level changes on tropical shallow-water carbonate shelves of this region. Species of foraminifers, sponges, disjectoporoids, corals, spongiomorphs, brachiopods, bivalves, ostracodes and echinoderms represent a Tethyan assemblage similar in taxonomic composition to well known Upper Triassic reef complexes of central Europe. Corals were significant, high-growing elements of the reef biota. One of the principal reefbuilding corals, Retiophyllia sellae, is known from the Polish Tatra Mountains and the Northern and Southern Alps. Other species such as Retiophyllia paraclathrata, Pamiroseris rectilamellosa, Margarosmilia charlyana, Astraeomorpha crassisepta, and Astraeomorpha confusa are common reef-building and reef-dwelling corals previously recognized from the Norian and Rhaetian stages of the Northern Calcareous Alps of Austria and Germany and Lombardy, Italy. Reef faunas of offshore Australia are among the easternmost examples yet recognized from the ancient coast of northern Gondwana. Palaeogeographically they belong to a broadly distributed tropical Tethyan faunal province, once extending from central Europe to the Himalayas, Timor, Papua New Guinea and Japan, but they compare most closely with faunas from central Europe. Triassic occurrences on the northwestern Australian shelf add much to our understanding of this deeply buried margin of Pangea and its subsequent breakup. The reefs are remnants of more extensive, carbonate platforms and reef complexes now largely removed by tectonic rifting.

70 TECTONIC AND PALAEOBIOGEOGRAPHIC SIGNIFICANCE OF THE CHATHAM ISLANDS, SOUTH PACIFIC, LATE CRETACEOUS FAUNA

Jeffrey D. STILWELL Department of Earth Sciences, James Cook University of North Queensland, Townsville, QLD 4811 The Cretaceous fauna of the Chatham Islands, South Pacific, is essentially a tectonically controlled facies fauna, with origins relating to divergent plate motions and concomitant deposition of volcaniclastic, transgressive sediments in half-grabens in the Chatham Rise region during the Late Cretaceous. At least 60 macroinvertebrate (predominantly Mollusca) taxa and one vertebrate have been recorded from the Kahuitara Tuff (Campanian-Lower Maastrichtian) of Pitt Island. The separation of the New Zealand subcontinent, including the Chatham Rise and Tasman Sea region, from the Marie Byrd Land sector of the Gondwana margin, along with changing oceanic circulation, lowering sea-surface temperatures and believed increasing temperature gradients, enhanced the distinctiveness of the fauna, as reflected in the strong species-level endemicity of the fossil record. Shoaling from volcanic activity in the Chatham Islands region created substrates suitable for colonization of a characteristic hardground community dominated by epifaunal suspensions feeders (ca. 41%), followed by lower percentages of infaunal suspension feeders (ca. 30%), epifaunal browsers (ca. 14%), deposit feeders (ca. 8%), and carnivores (ca. 5%). The Kahuitara Tuff faunule is divided into four biogeographic groupings at genus- and subgenus-level: Indo-Pacific/Tethyan (ca. 37%), cosmopolitan (ca. 34%), palaeoaustral (ca. 28.5%), and endemic (ca. 8.5%). Palaeoaustral taxa are inclusive of endemic groups, in accordance with C. A. Fleming's original ideas. These percentages suggest an overall relatively warm-water, semi-global biogeographic flavour at this taxonomic level. Some 43% of taxa from the Kahuitara Tuff are found in mainland New Zealand coeval faunas and about 41% are endemic to the Chatham Islands, differences being attributable largely to facies and to a much lesser degree to geographic isolation. At species-level, endemic taxa of cosmopolitan or wide-ranging Indo-Pacific/Tethyan and palaeoaustral genera/subgenera (83%) dominate, with negligible representation of widespread or cosmopolitan species and endemic species of endemic genera/subgenera. It is suggested that the Kahuitara Tuff faunule represents evolutionary divergence, reflecting range retractions of a former cosmopolitan, early to late Mesozoic world. Nearly all Kahuitara Tuff taxa are endemic to either the Chatham Islands or mainland New Zealand. Similarities of the fauna with other coeval faunas around the rim of the southern circum-Pacific are moderate to weak at genusand subgenus-level, indicating a degree of provincialism and isolation, especially at specieslevel. The Chatham Islands fauna probably belonged to the short-lived, Campanian to Maastrichtian, Weddellian Biotic Province of W. J. Zinsmeister. The Kahuitara Tuff fauna evolved from a mixture of Austral and Boreal elements during the Late Cretaceous with evidence of approximately 40% of taxa having ancestors in the Chatham Islands-New Zealand region of the Gondwana supercontinent. Changes in composition across the K-T boundary in the Chatham Islands were dramatic with very few genus- and subgenus-level taxa in the Kahuitara Tuff present in the Upper Palaeocene to Lower Eocene Red Bluff Tuff.

71 PALAEOBIOGEOGRAPHY OF AUSTRALASIAN DEVONIAN BIOTA

JOHN A. TALENT\ JONATHON C. AITCHISON^ MARGARET BRADSHAW^ GARY DARGAN^ RUTH MAWS0N\ THERESA WINCHESTER-SEETO\ GAVIN YOUNG^ & YONG-YIZHEN^ ^ MUCEP, Earth Sciences, Macquarie University 2109 ^ Department of Earth Sciences, University of Hong Kong, CHINA ^PlaNet NZ, Christchurch, NEW ZEALAND Specialist Services Section, Geological Survey of Nev^^ South Wales, Lidcombe 2141 Documenting movement of continental plates through time lies in the hands of the palaeomagnetists and palaeobiogeographers: the fomier can postdict latitudinal positions for land masses and the latter, using similarity indices of the biotas, can give the best estimation of longitude and distances apart of former blocks. A high degree of continental fragmentation tends to produce higher endemism, and, conversely, at times when continental pieces move together and/or dock to form larger land masses, biotas become increasingly cosmopolitan. It is well known that global biogeographic history has an element of cyclicity about it. Through time, provinciality has waxed and waned repeatedly from high levels of provinciality, such as in the Early Devonian, and Late Carboniferous-Permian, to high levels of cosmopolitanism, as in the Late Devonian-Early Carboniferous. Such fluctuations have been ascribed to changing orientations and patterns of aggregation and dispersal of continental blocks, attendant changes in oceanic circulation patterns and, for terrestrial biota, resultant changes in climate. The Devonian typifies a cycle of such a pattern with endemism increasing to a maximum in early Emsian time, then decreasing through the Middle Devonian to remarkable cosmopolitanism in the Late Devonian. Precision in palaeobiogeographic studies relies on exact stratigraphic alignments of the faunas or floras being compared. Traditionally, most stratigraphic alignments have been made on the basis of macrofauna (especially brachiopods) and, though obviously more sensitive to ecology and more differentiated biogeographically, macrofaunas retain considerable value for regional correlation, especially in the absence of carbonates. In recent years, deliberations of the Subcommission on Devonian Stratigraphy have produced stability as regards an international standard scale, based on conodonts, for the Devonian. We have, therefore, attempted to relate our stratigraphic alignments to this zonal scheme. In some cases the intervals represented by individual conodont zones correspond broadly with regional stratigraphic and macrofaunal intervals, but in many cases, especially in the Late Devonian, they are finer. Early Devonian biota, indicating a high degree of provinciality, have been delineated as belonging to 3 distinct realms (the Eastern North American, the cold water Malvinokaffric, and the wide-spread Old World Realm that have been further subdivided into provinces: the Tasman Province or East Gondwana Province, the New Zealand Province, the Rhenish-Bohemian Province or Eumerica Province (including Laurentia and Baltica), the Uralian Province or Siberian and Tuva provinces, and the distinctive Malvinokaffric Province. Globally, benthic faunas, including brachiopods, trilobites and molluscs increase in provinciality through Early Devonian time to a maximum in the early

Emsian, then decline to a minimum in the Late Devonian (Famennian). At the acme of provincialism, the degree of dissimilarity of New Zealand bivalve and brachiopod faunas from the Reefton area and those from Australia became so pronounced that a discrete New Zealand Province came into existence. Links with both the Malvinokaffric and Tasman provinces indicate the Reefton region lay close to both these regions. That the Malvinokaffric Province included Antarctica is clear from the occurrence of brachiopods such as Pleurothyrella and Australocoelia. Even pelagic biota such as conodonts, fish, and chitinozoans were highly provincial during the Early Devonian. The paucity in Australia of conodont genera such as Icriodus and Pelekysgnathus is thought to be related to juxtaposition of the continental land masses at the time rather than biofacies related. Eastern Australian conodont funas have strong affinities with those from southern central Asia (Uzbekistan and Tajikistan). Analysis of both fish and chitinozoan faunas has also indicated a high degree of provincialism for the Early Devonian. Although some new species of radiolarians have been reported from the Early Devonian of eastem Australia, insufficient data are to hand to decipher patterns of provinciality. Endemism persisted into the Middle Devonian although provinciality became less clearly defined. Fish faunas indicate completely different cephalaspid/heterostracan faunas in the Uralian Province to that of East Gondwana whereas faunas from Tarim, South and North China and Indochina shared a distinctive and highly endemic galeaspid-yunnanolepid assemblage. Similarity of fish faunas from South Africa, South America and southern Victoria Land, Antarctica, indicate continuation of the Malvinokaffric Province. Provinciality in corals, brachiopods, trilobites and molluscs declined globally through the Middle Devonian into the Frasnian. By the late Famennian, all groups had become highly cosmopolitan. The occurrence of sinolepid antiarchs in eastem Australia (the only record of an endemic Asian fish group outside of Asia) suggests a continental connection with previously isolated Asian terranes. Famennian conodont faunas, notably from north Queensland, include species in common with faunas from central Kazakhstan suggesting an oceanic link between the two areas. Biogeographic differentiation of plants also declined in the Late Devonian, argued to be caused by a marine regression and generation of a land connection between Gondwana and Euramerica. Although there was a single floral realm by the Late Devonian (with some ecological segregation), Australian floras showed closest affinity with those from Kazakhstan, China, Japan and Siberia.

AN EARLY CRETACEOUS FLORA FROM THE LOWER STRZELECKI GROUP, GIPPSLAND BASIN, VICTORIA Anne-Marie TOSOLINI, Nathalie NAGALINGUM, Stephen McLOUGHLIN & Andrew DRINNAN School of Botany, The University of Melbourne, Parkville, VIC 3052 Plant macrofossil assemblages from exposures of the lower Strzelecki Group in the Boola Boola Forest, north of Traralgon, Gippsland Basin, are dominated by a suite of conifers including undescribed species of Pagiophyllum, Elatocladus and Brachyphyllum, Subdominant gymnospermous elements of the floras include Bennettitaleans (Ptilophyllum boolensis and P. fasciatum), Pentoxylaleans {Taeniopteris daintreei), and a range of pteridosperms of uncertain alliance {Pachypteris austropapillosa, Thinnfeldia sp. and a new leaf type previously assigned to Reinitsia). Most of these groups can be identified from dispersed cuticle assemblages based on characteristic stomatal arrangements. However some taxa, like Taeniopteris and "Reinitsia'\ have delicate cuticles that are readily destroyed during treatment with acid to remove mesophyll tissues. These groups may be under-represented in dispersed cuticle assemblages. Ferns are abundant within discrete horizons and were probably the chief understorey components of the vegetation. Equisetaleans are poorly represented in the macrofloras and lycophytes appear to be absent. However, this absence may be a preservational artefact as a diverse range of lycophytic megaspores are present in the palynoflora. Both conifer and seed-fern remains are characterized by the small size of their leaves compared to other Mesozoic macrofloral assemblages. This feature may have been a response to growth at high latitudes (>70''S) in cold climates where in modem floras large leaves are prone to frost damage. Abundant epiphyllous fungi preserved on a range of leaf taxa are suggestive of a relatively humid climate. Few other climatic indicators are available in the Neocomian floras. Both entire-margined and compound leaves are represented in the floras. However, application of angiosperm-based morpho-climatic models based on ratios of entire to compound leaf forms are probably inappropriate as many of the Mesozoic gymnosperms belong to now extinct groups. Analogies with angiosperm foliage may not take into account phylogenetic and developmental constraints on the form of Mesozoic gymnosperm leaves. A Neocomian age is proposed for the Boola Boola assemblages based on the association of Murospora florida, Cooksonites variabilis, Cyclosporites hughesi, and Dictyosporites speciosus, and the absence of Crybelosporites stylosus from the palynofloras. The lower Strzelecki Group is dominated by coarse sandstones and conglomerates and was deposited predominantly within braided fluvial environments. Processing of specific lithofacies has, in many cases, yielded distinctive dispersed cuticle and spore/pollen assemblages that provide additional data for the interpretation of local depositional environments and the distribution of plant groups in ancient ecosystems. The Victorian Neocomian floras share a range of genera and species with Early Cretaceous floras from South Australia, Westem Australia, Queensland and the Northern Territory. More remote affinities are evident with Early Cretaceous assemblages from the

Antarctic Peninsula and India, and relatively few taxa are shared with western Gondwanan (South African and South American) floras.

A LATE PALEOCENE LEAF FLORA FROM CAMBALONG CREEK, SOUTHERN HIGHLANDS, NEW SOUTH WALES Anthony J. VADALA & Andrew N. DRINNAN School of Botany, The University of Melbourne, Parkville, VIC 3052. A detailed study of mummified leaf remains from Cambalong Creek in the Southern Highlands of New South Wales has provided information regarding a floral community growing in the area approximately 58-60 million years ago (Thanetian stage). Information from these leaves elucidates the Tertiary history and modem biogeography of significant elements of the modem flora of Australia, particularly Lauraceae, Nothofagaceae and Proteaceae. Predominance of Lauraceae (as the fossil genus Laurophyllum, and the extant genera Litsea and Neolitsed) reiterates the prominence of this family in southeastem Australian floras throughout the Tertiary. The Proteaceae component of the flora includes taxa from tribes Banksieae, Macadamieae and Oriteae, and reinforces the diversity of subfamily Grevilleoideae by the Thanetian. Other components of the flora include taxa from Elaeocarpaceae (Elaeocarpus), Eucryphiaceae (Eucryphia), Nothofagaceae (Nothofagus), and Sapindaceae {Heterodendrum). The conifer flora is composed of taxa from Araucariaceae (Araucarioides), and Podocarpaceae {Dacrycarpus and Phyllocladus). The extant analogues of most components of the flora are in the mesotherm response category, which suggests a warmer climate than at present in the higher latitudes of the Southem Highlands during the Thanetian. The importance and relevance of cuticular micromorphology to the systematics and taxonomy of both extinct and modem taxa of plants is also emphasised.

AUSTRALIAN NEOPROTEROZOIC STRATIGRAPHY, GEOGRAPHY AND BIOGEOGRAPHY M.R. WALTER\ J. GEHLING^ & K. GREY^ ^ School of Earth Sciences, Macquarie University, NSW 2109 ^ Department of Geology, Flinders University, Adelaide SA 5001 ^ Geological Survey of Western Australia, Perth, WA 6004 There is an emerging precision in Neoproterozoic biostratigraphy resulting mostly from new discoveries of acritarchs, particularly the spectacularly rich and diverse microfloras of the Amadeus and Officer Basins. Metazoan body and trace fossils are making a contribution, and stromatolites continue to be useful. Biostratigraphy is complemented by isotope chemostratigraphy using carbon, sulfur and strontium. When added to litho- and sequence stratigraphy these results provide sufficient time resolution to permit attempts at palaeogeographic interpretations, especially for the "terminal" Proterozoic (Ediacarian). Meaningful biogeography is still largely beyond our grasp, but some preliminary observations are possible. Most of the Neoproterozoic sedimentary rocks of Australia are located in the Adelaide Rift Complex and the Centralian Superbasin. The Rift Complex is a intracratonic rift at a high angle to the associated east-west elongated epicratonic sag of the Superbasin. In the earliest Cambrian the Flinders zone of the Adelaide Rift Complex was transformed to a failed arm or aulacogen by continental breakup along its southem part. According to many interpretations, Australia was joined in the Neoproterozoic with India, Antarctica, and Laurentia, such that the Tasman Line faced the Canadian-Wyoming cordilleran line. The configuration of the north-south trending Adelaide Rift Complex and the east-west trending Centralian Superbasin was mirrored by the basins in Laurentia to form a T, which split at about the end of the Neoproterozoic by growth of a precursor of the Pacific Ocean. The Neoproterozoic stratigraphy of Australia can be analysed in terms of four supersequences, with finer subdivision possible in the Ediacarian and possibly also earlier. Supersequence 1, about 800Ma old, appears to have two successive assemblages of both acritarchs and stromatolites. Supersequence 2 starts with the Sturtian glacial deposits. The overlying shales, siltstones and carbonates contain characteristic stromatolites, a huge and unprecedented sulfur isotope anomaly and a major carbon isotope anomaly. Supersequence 3, starting with the Marinoan glacials, has a succession of at least three acritarch assemblages defined by a total of about 50 taxa. A single taxon of metazoan trace fossil is known from the Georgina Basin, and there are distinctive stromatolites. Supersequence four encompasses the Ediacara biota, which at least in Australia cannot be subdivided further. As far as can be determined from the limited data available, many Neoproterozoic stromatolites appear endemic to Australia. Several forms are widespread in Australia and useful for correlation, but few have global distributions. There is a rich record of microbes and protists in the form of acritarchs in siliciclastic rocks and microfossils in chert. This record changes profoundly from the mid-Neoproterozoic to the Ediacarian. During both time intervals shallow-water, near-shore communities consisted almost exclusively of prokaryotic coccoid and filamentous microorganisms derived from benthic

microbial communities. Many taxa are common to both ages. Protistan phytoplankters are present, but numbers are generally small, species diversity is low and taxa are often simple in structure. Assemblages from shelfal facies in both pre-Ediacarian and basal Ediacarian are also similar. They contain large numbers of phytoplankters, but species diversity is generally lov^, and morphologies tend to be simple. By contrast, assemblages in the middle Ediacarian are very diverse and contain highly complex, exceptionally large species. Taxa appear to be short-ranging. This diversity and the rapid changes make it possible to subdivide the middle Ediacarian into three zones. It may be more than coincidence that the marked increase in complexity occurs immediately after the Acraman impact event, although it might reflect initiation of grazing by planktonic metazoans. The Ediacara biota is cosmopolitan. Most of the organisms lived on the sea floor just below fairweather wavebase, but those of NW Canada and the Avalon terrane of Newfoundland and England were deeper shelf communities. At least 20 genera have been identified in two or more localities around the globe. Distinctive frondose forms such as Chamia and Pteridinium are known from four continents including Australia. The discoidal form Beltanelliformis is even more widely distributed. The biotas of the Flinders Ranges and the White Sea are remarkably similar, suggesting proximity, contrary to most continental reconstructions.

PALAEOBIOGEOGRAPHY OF AUSTRALIAN FOSSIL AMPHIBIANS Anne WARREN, Ross DAMIANI, Caroline NORTHWOOD & Adam YATES School of Zoology, La Trobe University, Bundoora, Melbourne, VIC 3083. Australia has representatives of the earliest known tetrapod vertebrates from the Late Devonian and evidence that their amphibian descendants remained here at least until the Visean. We have no record of tetrapods from later in the Carboniferous or the earlier parts of the Permian and scanty remains only, from the Late Permian. A single Permian record from the Sydney Basin, if correctly identified, indicates the presence of advanced temnospondyls (stereospondyls) much earlier in the fossil record than formerly predicted. We propose a Gondwanan refuge for late Permian temnospondyls which became extinct in the latest Permian elsewhere. In contrast, our Early Triassic amphibian fauna is diverse, geographically widespread, and contains stem stereospondyls indicative of a Gondwanan origin of the entire radiation of Mesozoic stereospondyls. Australia was also a refuge for post-Triassic temnospondyl amphibians, with two Early Jurassic representatives and the last known survivor of that group in the Early Cretaceous.

PROBLEMS FOR PALAEOBIOGEOGRAPHICAL ANALYSIS OF PALAEOZOIC CORALS AS ILLUSTRATED BY LOWER CARBONIFEROUS CORALS OF EASTERN AUSTRALIA Gregory E. WEBB Department of Earth Sciences, University of Queensland, Brisbane, QLD 4072 Problems encountered in analyzing the palaeogeography of eastern Australian Lower Carboniferous (EALC) corals can be grouped into three major classes, which relate to: 1) taxonomic practice; 2) techniques used to compare regional faunas; and 3) sample bias. Correct taxonomic assignment is critical to any palaeobiogeographic analysis, but the level of confidence assigned to taxonomic determinations varies considerably. The true affinities of many EALC corals remain unknown owing to poorly preserved, rare, or inadequately described material. However, confidence is no higher for other taxa that are adequately known and described, because the type species for comparable foreign genera are poorly known or inadequately described. Species described as Amplexocarinia from Queensland (Webb 1990) are adequately known, but the juvenile morphology of the type species of the genus is unknown. Hence, it is impossible to be confident that Australian and other global occurrences of Amplexocarinia-likc corals are congeneric. This problem is particularly severe, because homeomorphy is rampant in Palaeozoic corals. Lithostrotionoid EALC corals, long considered congeneric with European forms, have been shown to represent a largely endemic clade (Webb 1990, 1994). Hence, great care must be taken in Palaeobiogeographic analysis to note the degree of taxonomic confidence placed in faunas from all compared regions. Techniques used to compare faunas impart related, but different, problems. Whether the emphasis is placed on endemic or shared taxa, and assuming taxonomic consistency, cooccurrence of a genus in two regions may have very different implications. Two populations may be freely interbreeding, implying close biogeographic connection, or they may represent completely isolated, relict populations inherited independently from a cosmopolitan precursor. Alternately, the populations both may be derived from the same contemporaneous source population, but have no direct genetic interchange between them. Hence, presence/absence data on the occurrence of genera is less important than an understanding of the history of the clades that contain the genera. Much evidence suggests that corals of the EALC and contemporaneous Japanese terranes were derived from the same stock, but there is much less evidence for direct genetic interchange. Sampling bias has been shown to be an important control of local coral diversity (Webb et al 1997). Poor sampling may bias against rare endemic taxa, but also biases diversity gradients used for the analysis of latitude. Of 45 analyzed EALC coral genera, 40.0 % are monospecific and 35.6 % are known from a single horizon/locality. Only 42.2 % of genera occur in both Queensland and New South Wales. These indices suggest that coral sampling is very incomplete in the EALC. Hence, the eastern Australian coral latitudinal gradient reported by Pickett & Wu (1990) may result more from the increased distribution of limestone (which itself may be latitudinally controlled) and greater sampling intensity in Queensland, than from a biological latitudinal diversity gradient alone.

Pickett, J. W. & Wu, W.-S., 1990. The succession of Early Carboniferous coral faunas in eastern Australia and China. Alcheringa 14, 89-108. Webb, Gregory E., 1990. Lower Carboniferous coral fauna of the Rockhampton Group, eastern-central Queensland. Memoir of the Association of Australasian Palaeontologists 10,1-167. Webb, Gregory E., 1994. Parallelism, non-biotic data and phylogeny reconstruction in paleobiology. Lethaia 27, 185-192. Webb, Gregory E., Sando, W. J., & Raymond, A. (in press). Mississippian coral latitudinal diversity gradients (western interior United States): testing the limits of high resolution diversity data. Journal of Paleontology 70.

ORDOVICIAN BIOGEOGRAPHY OF AUSTRALASIA B.D. WEBBY\ LG. PERCIVAL^ A.H.M. VANDENBERG^ J.W. PICKETT^ J. POJETA Jr.\ R. SCHALLREUTER^ T. WINCHESTER-SEETO\ G.D. EDGECOMBE^ R.A. COOPER^ & Y.Y. ZHEN^ ^Centre for Ecostratigraphy and Paleobiology, School of Earth Sciences, Macquarie University, NSW 2109 ^Specialist Services Section, Geological Survey of New South Wales, P.O. Box 76, Lidcombe NSW 2141 ^Geological Survey of Victoria, Department of Agriculture, Energy and Minerals, P.O. Box 2145, Fitzroy, VIC 3065 "^U.S. Geological Survey, Museum of Natural History, Smithsonian Institution, Washington D.C., 20560, U.S.A. ^Geologisch-Palaontologisches Institut, Universitat Hamburg, D-20146 Hamburg, GERMANY ^Palaeontology Section, Australian Museum, P.O. Box A285, Sydney South, NSW 2000 ^Institute of Geological and Nuclear Sciences, P.O Box 30368, Lower Hutt, NEW ZEALAND Biogeographic relationships of more important Australasian biotas are assessed for the Early Ordovician (Warendian-Chewtonian), Middle Ordovician (CastlemainianDarriwilian) and Late Ordovician (Gisbomian-Bolindian) epochs. They are characterised as low-latitude faunas that either lived in the shallow seas encroaching part of eastern Gondwana, or occupied a range of habitats in the shelf margins, island arcs and basins that rimmed the vast Panthalassic Ocean. The groups studied include: corals,stromatoporoids, sponges, molluscs, brachiopods, trilobites, ostracodes, conodonts, graptolites and chitinozoans. Only a small component of the Early-Middle Ordovician biotas from the platform cover, now confined to intracratonic basins, provide diagnostic biogeographic data. A few nautiloids, gastropods, and brachiopods from the Canning and Georgina basins exhibit close provincial ties to counterparts in Thailand and Malaysia (Shan-Thai terrane); while others have less close connections to Tasmania, China, South America, Western Europe or North America. The chitinozoans from these same basins exhibit broader relationships, being linked with assemblages of low-latitude, "pelagic" aspect in North America and Spitsbergen. The graptolite faunas belong to the circum-equatorial, "warm-water". Pacific Province, and include both the varied assemblages of deeper-water aspect in Eastern Australia and New Zealand, and relatively shallow, "epipelagic" zone associations of the Canning Basin. Conodonts of the cratonic basins and the shelf margins (e.g., Tasmanian Shelf, and Gnalta Shelf in western New South Wales) are of warm-water aspect (representatives of the North American Midcontinent Province), and those from basinal successions of Victoria and central & southern NSW are cooler (?deeper) water faunas (members of the North Atlantic Province). Late Ordovician "island arc" faunas of central NSW additionally exhibit a component of North Chinese provincialism. The significance of the wide distribution of ostracode subfamily, the Pillinae, in the Ordovician successions is not yet clear; it is only known elsewhere from South America, possibly China and Siberia.

Other Late Ordovician faunas show distinctive patterns of provincialism. The Tasmanian nautiloids have a high proportion of endemic genera, and a component of nektobenthonic forms, in contrast to the NSW "island-arc" nautiloids that have lower diversities and an exclusively nektonic mode-of-life suggesting they alone were able to cross the marginal sea (Wagga Trough) that separated the island arc from the Tasmanian Shelf. The NSW gastropod and brachiopod faunas show remarkably close links with North America, especially the Klamath Mts of California and Alaska, less close relationships with Kazakhstan, SE China and Siberia. The East Australian trilobite faunas have been characterized as belonging to the Eokosovopeltis-Pliomerina province of the circumequatorial Remopleuridid Realm, with close links to SE Asia, China, Korea and Kazakhstan, but New Zealand with its occurrences of the South American genus Incaia should perhaps be grouped in a separate province. The Late Ordovician coral and stromatoporoid faunas belong to the American-Siberian Realm. At the provincial level, central NSW stromatoporoids have closest links with SE Asian, N., NW & SE Chinese, Korean and Russian Altai-Sayan assemblages. Though the Tasmanian stromatoporoid assemblages are not closely similar to their counterparts in central NSW, they still exhibit reasonably close links to Asia (e.g. NW China), as well as closer relationships to North American faunas. The corals include both North American and Asian (Russian Altai-Sayan) links. The isolated late Eastonian assemblages of north-east NSW have closest relationships with the North American "Red River" faunas. Occurrences of early Bolindian agetolitids in central NSW and North Queensland suggest close links with China, Central Asia, Kazakhstan and Alaska. IGP & JWP publish with permission of the Director General, NSW Department of Mineral Resources. This abstract is a contribution of IGCP 410 (The Great Ordovician Biodiversification Event).

PALEOBIOGEOGRAPHY OF PERMIAN ECHINODERMS OF AUSTRALIA Gary D. WEBSTER\ Peter A. JELL^ & Aram N. DEREWETZKY^ ^Department of Geology, Washington State University, Pullman, WA, USA, 991642812 ^Queensland Museum, P. O. Box 3300, South Brisbane, QLD 4101 Permian echinoderms of Australia are represented by crinoids, blastoids, echinoids and asteroids. They are known from marginal basin sediments of early Sakmarian into early Wuchiapingian age. Artinskian faunas are the most diverse and abundant, and known from eastern and Western Australia. Roadian and Wordian faunas are known only from eastern Australia. The Wuchiapingian fauna is known from Western Australia and is one of the two youngest Permian faunas known from cups and crowns worldwide. Paleogeographic position of the Permian faunas of Western Australia is from 39. r S lat., 97.5^ E long., to 55.9^ S lat., 90.6^ E long., whereas faunas of eastern Australia were living between Al.T S lat., 130.9^ E long., and 6 9 . r s lat., 131.1^ E. long. Permian echinoderms of Australia may be grouped into three general categories, Australian endemics, Tethyan endemics, and non-endemics. The endemics, both Australian and Tethyan, are advanced forms, whereas the non-endemics are conservative forms. The conservative forms are mostly cosmopolitan taxa initially reported from the Devonian or more commonly the Early Carboniferous and their occurrence in the Permian of Australia often represents a range extension. Echinoderms are a significant part of the cooler water Permian faunas of Australia. They were adapted to living in a clastic rich enviroment, atypical of Paleozoic echinoderms. Crinoids are the dominant echinoderm and represented by camerates, disparids, cyathocrinitids, and poteriocrinitids. Poteriocrinitids dominate the crinoid faunas. Blastoids are dominated by neoschismatids.

CHITINOZOANS AND FORAMINIFERAL LININGS FROM THE ORDOVICIAN OF AUSTRALIA: AN OVERVIEW Theresa WINCHESTER-SEETO Centre for Ecostratigraphy and Palaeobiology, School of Earth Sciences, Macquarie University, NSW 2109. Elsewhere in the world, Ordovician chitinozoans have been used extensively for biostratigraphic correlation; however, in Australia, they have been poorly documented. Studies to date have centred primarily on the Lower and Middle Ordovician Nambeet, Thangoo and Goldwyer formations of the Canning Basin, (Combaz & Peniguel 1972; Achab & Millipied 1980), with additional work in progress on material from the Goldwyer & Nita formations (unpub. data, C. Foster & T. Winchester-Seeto). This is, in part, overlapped by work from the Georgina Basin on the Coolibah and Nora formations (Playford & Miller 1988). The studies span the period from the late Lancefieldian (Zone La3) through to the late Darriwilian, possibly Da3. (early Arenig to Llanvim). Within this time interval, seventeen genera are known to exist, but only six have so far been observed in Australian strata. In general, the chitinozoan genera from Australia have a very simple morphology {Conochitina, Lagenochitina, Desmochitina, Calpichitinal), with Cyathochitina the only genus with a carina and Belonechitina the only genus with an ornamented vesicle. At the species level, Australian faunas have very few species in common with northern Gondwana or Baltica, but are similar in composition to those from Laurentia (especially Quebec and western Newfoundland) and Spitsbergen. Conochitina symmetrica Taugourdeau & Jekhowsky, found at the base of the Arenig (correlated with the Tetragraptus approximatus Zone), is the exception and has been recovered from Australia (Nambeet Formation), the Sahara, Bohemia and Quebec and has been used in zonal schemes set up by Paris (1990) for northern Gondwana and by Achab (1989) for western Newfoundland and Quebec; however, beyond this there is little similarity between the two schemes. Achab (1989) identified six chitinozoan biozones spanning the interval from the beginning of the Arenig to early Llanvim; four of these biozones can be identified in Australian faunas. The pattems of distribution of chitinozoan species suggests a palaeolatitudinal influence operating in the the Early and Middle Ordovician. Linings of agglutinated foraminiferans have been recovered from Palaeozoic strata, ranging from the Early Cambrian to Late Devonian. A review of Ordovician foraminiferal linings observed from Australian strata is given.

Achab, A., 1989. Ordovician chitinozoan zonation of Quebec and western Newfoundland. Journal of Paleontology 63, 14-24. Achab, A. & Millipeid, P. 1980. Conochitina symmetrica Taugourdeau & Jekhowsky, a guide fossil to the lower Ordovician. Fifth International palynological Conference, Cambridge, p. 3. Combaz T.H. & Peniguel, G., 1972. Etude palynostratigraphique de I'Ordovicien dans quelques sondages du basin de Canning (Australie occidentalj. Bulletin du Centre de Recherches de Pau, SNPA 6, 121167. Paris, F., 1990. The Ordovician biozones of the northern Gondwana Domain. Reviews of Palaeobotany & Palynology 66, 181-209. Playford G. & Miller, M.A., 1988. Chitinozoa from Lower Ordovician strata of the Georgina Basin, Queensland (Australia). Geobios 21, 17-39.

LATE DEVONIAN BIOEVENTS AND BIOGEOGRAPHIC PATTERNS IN RELATION TO SEA LEVEL CHANGES IN THE SHOTORI RANGES, EAST IRAN M. YAZDI Department of Geology, University of Isfahan, IRAN Fluctuation of the relative water level in the early Frasnian Bahram Formation from terrestrial conditions to shallow water can be documented by reworking and gaps within the sedimentation in the Shotori Range and Central Iran. A new study has been made of two Frasnian biostromal horizons in the Shishtu Formation: at the base of Howz-e-Dorah (Shotori Range), Chahriseh (Esfahan), Damghan and around Central Iran (older than jamieae Zone). The depauperate fauna recovered from these biostromal horizons is in harmony with shallow water to shoaling conditions that must have occurred at the base of the jamieae Zone. The increasing number of faunal elements together with other palaeoenvironmental features indicate a change to an outer shelf and open marine environment. This is also indicated by the occurrence of goniatites in intervals dated as late jamieae Zone and basal linguiformis Zone (Yazdi, 1996). The low number of faunal elements in the Shotori Range in horizons of an age close to the Frasnian/Famennian boundary has been reported from other localities globally (Sandberg et al 1988). The base of the Famennian stage of the Shishtu Formation (in the Shotori Range), can be defined by shallow water and ironstone facies. The lack of palmatolepid species in the early Famennian indicates a shallow to nearshore position. Terrestrial to back-reef environment is suggested for the early Famennian {triangularis Zone to end of crepida Zone)in both the Shotori Range and in Central Iran. The high number of goniatites and increasing number of palmatolepids within the rhomboidea Zone to postera Zone indicates open marine or deeper water for both these localities in Iran. A Gondwanan link is suggested by the Famennian goniatite species in the Shotori Range and Central Iran. From the postera Zone to the Devonian/Carboniferous boundary, the palaeoenvironmental indicators show shallowing occurring in both these localities as well as in other parts of the world (Girard 1994). Girard, C., 1994. Conodont Biofacies and Event stratigraphy across the D/C boundary in the stratotype area (Montagne Noire, France). Courier Forschungsinstitut Senckenberg 168, 399-309. Sandberg, C.A., Ziegler, W., Dreesen, R. & Butler, J.L., 1988. Late Frasnian mass extinction: conodont event stratigraphy, global change & possible causes. Courier Forschungsinstitut Senckenberg 106, 263-307. Yazdi, M, 1996. Late Devonian-Carboniferous conodont biostratigraphy of the Tabas area, Eastern Iran. Unpublished PhD thesis, Macquarie University, Sydney, Australia, 221 pp.

LOWER VERTEBRATES OF EAST GONDWANA - BIOGEOGRAPHIC PATTERNS GAVIN C. YOUNG Geology Department, Faculty of Science, Australian National University, Canberra ACT 0200 Distribution patterns of Palaeozoic vertebrates vary markedly through time (Long 1993). The Ordovician Arandaspis fauna of the Amadeus Basin (the oldest significant assemblage from East Gondwana, and the World) has close affinity with the Sacabambaspis fauna of Bolivia (Young 1997). The different Astraspis fauna in the North American Harding Sandstone establishes a Gondwana pattern at the beginning of the vertebrate fossil record. Diverse Siluro-Devonian chondrichthyan-acanthodian assemblages (Malvinokaffric Province) distinguish West from East Gondwana, the latter with its Early-Middle Devonian endemic Wuttagoonaspis-Pituriaspis fauna (Young 1981, 1987, 1991). At that time Laurentia and Baltica had a completely different cephalaspid/heterostracan fauna, whilst Asian terranes (Tarim, South and North China, Indochina) shared a distinctive and highly endemic galeaspid-yunnanolepid assemblage (Young & Janvier, in press). The known distribution pattern of Siluro-Devonian vertebrates in eastern Asia is more consistent with modem geography than with a Palaeozoic geography of widely separated Asian terranes. The wuttagoonaspidphyllolepid lineage in East Gondwana indicates a new continental connection to Euramerica (via West Gondwana, not Asia) near the Frasnian-Famennian boundary. This also explains the known distribution of early tetrapods (represented in East Gondwana by Early-Late Devonian trackways, and Famennian-Visean skeletal remains; Thulbom et aL 1996). The latest Devonian occurrence of sinolepid antiarchs in the Lachlan Foldbelt suggests a connection at that time with previously isolated Asian terranes (Ritchie et al 1992; Rich & Young 1996). These patterns contradict some other ideas on the palaeogeographic setting of East Gondwana (e.g. connection with Asian terranes in the Early Devonian; wide Middle-Late Devonian ocean between Gondwana and Euramerica), based largely on palaeomagnetic data. The claimed superiority of such quantitative data is rejected. Cladistic methods (Young 1986, 1990, 1995a,b) can be applied to palaeogeographic analysis to accommodate various qualitative data sets relevant to the history of continental or terrane dispersion and accretion, permitting true data integration, and parsimony analysis of competing hypotheses. Long, J.A. (Editor), 1993. Palaeozoic Vertebrate Biostratigraphy and Biogeography. Belhaven Press, London, 369 pp. Rich, T.H. & Young, G.C., 1996. Vertebrate biogeographic evidence for connections of the east and southeast Asian terranes with Gondwana. Australian Journal of Earth Sciences 43, 625-634. Ritchie, A., Wang, S.T., Young, G.C. & Zhang, G.R., 1992. The Sinolepidae, a faniily of antiarchs (placoderm fishes) from the Devonian of South China and eastern Australia. Records of the Australian Museum 44, 319-370. Thulborn, T., Warren, A., Turner, S. & Hamley, T., 1996. Early Carboniferous tetrapods in Australia. Nature 381, 777-780. Young, G.C., 1981. Biogeography of Devonian vertebrates. Alcheringa 5,225-243. Young, G.C., 1986. Cladistic methods in Paleozoic continental reconstruction. Journal of Geology 94, 523-537.

Young, G.C., 1987. Devonian palaeontological data and the Armorica problem. Palaeogeography, Palaeoclimatology, Palaeoecology 60, 283-304. Young, G.C., 1990. Devonian vertebrate distribution patterns, and cladistic analysis of palaeogeographic hypotheses. In Palaeozoic biogeography and palaeogeography (ed W.S. McKerrow & C.R. Scotese). Geological Society of London Memoir 12, 243-255. Young, G.C., 1991a. The first armoured agnathan vertebrates from the Devonian of Australia, 67-85. In Chang, M.M., Liu, Y.H., & Zhang, G.R. (editors). Early Vertebrates and Related Problems of Evolutionary Biology. Science Press, Beijing, China, 514 pp. Young, G.C., 1995a. Application of cladistics to terrane history - parsimony analysis of qualitative geological data. Journal of Southeast Asian Earth Sciences 11, 167-176. Young, G.C., 1995b. Early vertebrates and palaeogeographic models. Geobios, Manoire Speciale 19, 129134. Young, G.C., 1997. Ordovician microvertebrate remains from the Amadeus Basin, central Australia. Journal of Vertebrate Paleontology 17, 1-25. Young, G.C. & Janvier, P. (in press). Early-Middle Palaeozoic vertebrate faunas in relation to Gondwana dispersion and Asian accretion. Final results volume, IGCP 321 'Gondwana dispersion and Asian accretion'. Balkema, Amsterdam.

AUSTRALIAN DEVONIAN RADIOLARIANS JONATHAN C. AITCHISON Department of Earth Sciences, University of Hong Kong, Pokfulam Road, HONG KONG, CHINA Western Australia: The richest and best preserved Devonian radiolarians occurrences in Australia are the diverse and remarkably well-preserved Frasnian faunas which have been described from carbonate concretions of the Gogo Formation, Canning Basin, Western Australia. (Nazarov et al 1982; Nazarov & Ormiston 1983; Aitchison 1993b). The Gogo Formation, is a basinal succession which interdigitates with marginal slope facies and consists mainly of shale, siltstone and thin lenticular beds of limestone (Nazarov et al, 1982). Calcareous nodules which weather out of the shales are composed of sandy micritic limestone and commonly contain well preserved fossils. The fauna is exceptionally well preserved and this is quite atypical compared to the kind of preservation conditions experienced by radiolarian workers who deal with deeper water cherty facies. New England Orogen of Eastern Australia: Well-preserved Lower to Middle evonian radiolarians have been known from the Tamworth district in eastern Australia since the end of last century (David & Howchin 1896). Hinde (1899) described specimens from thin sections cut from samples collected by David & Pittman (1899). For almost 100 years very little further attention was paid to radiolarians from this area and the Hinde fauna was only recently reassessed by Aitchison & Stratford (1997). The past decade has, however, witnessed a renaissance of radiolarian research in the Devonian of eastern Australia. New generation radiolarian work was initiated primarily because these microfossils were recognised as having potential to provide the age control upon which to base studies into the tectonic assembly of the New England orogen (NEO). Utilising HF acid-leaching techniques to extra radiolarians from their surrounding matrix studies of the three dimensional form of these microfossils made their identification and later zonation more achievable. The most abundant source of Devonian radiolarians in the NEO is from the volcaniclastic sediments of the Gamilaroi terrane (Flood & Aitchison 1988, 1992), the westernmost terrane in the NEO. This terrane comprises a complex association of volcaniclastic sediments, tuffs, volcanic rocks and minor carbonates formed in an intraoceanic island-arc setting (Aitchison & Flood 1994; Stratford & Aitchison 1996) and accreted to the eastern margin of Gondwana sometime during the Late Devonian (Flood & Aitchison 1992). Recent work has confirmed the presence of abundant radiolarians in tuffaceous sediments of the Gamilaroi terrane (Aitchison 1988b, c, 1993a; Aitchison & Flood 1992; Aitchison et al 1992; Dongal 1995; Ishiga 1989; Ishiga & Leitch 1988; Ishiga et al 1987, 1988; Spiller 1992, 1993; Stratford, 1995; Stratford & Aitchison 1997; Metcalfe et al 1997). Although abundant, radiolarians are commonly not well preserved and they are often not accompanied by other fossils making correlation with other biozonations problematic. Aitchison & Stratford (1997) revised and reassessed the original material described by Hinde and that a workable biostratigraphy could be established within this terrane with further detail of occurrences throughout the terrane documented by Aitchison et al (in

press). It is likely that correlative of the Gamilaroi terrane present elsewhere in the NEO may also contain abundant radiolarians. Seven distinctive radiolarian assemblages have been defined from the Gamilaroi terrane (Aitchison et al, in press) and they form the basis of an uppermost Lower to Middle Devonian radiolarian biozonation. In ascending order these partial range assemblages are: Stigmosphaerostyla horrida, Helenifore laticlavium, Circulaforma admissarius, Helenifore pilosidiscus, Protoholoeciscus hindea, Ceratoikiscum regalinodus and Trilonche minax assemblages. Gamilaroi terrane sedimentation occurred during the interval Early (Pragian) to Late (Frasnian) Devonian in a volcanic island arc environment in which radiolarians were abundant but sedimentary facies were locally variable. Although assemblages are dominated by spumellarians they also include more distinctive ceratoikiscid forms. Radiolarian assemblages can be used for terrane-wide correlation and recent work from Japan indicates that these assemblages may be cosmopolitan. Devonian radiolarians of the NEO are not restricted to the Gamilaroi terrane, rather they found in many of the terranes in the orogen (Ishiga et al. 1987; Aitchison 1988, 1993a) and their study is on-going. Data are presently insufficient to permit any clear distinction of provinciality amongst the relatively few occurrences of Devonian radiolarians known globally. Described faunas show close similarity with the most significant differences between faunal assemblages apparently being related to preservational factors. Faunas described from basinal successions such as those from the Gogo Formation of Western Australia and the Domanik of Russia (Afanasieva 1997) include a remarkable diversity of well-preserved forms which are commonly delicate and/or ornate. Faunas from sediments which most likely accumulated in deeper water abyssal environments, such as those from the NEO or many Russian faunas (Nazarov 1988), are of lower diversity and are dominated by robust forms which might be more likely to have survived partial dissolution or diagenesis. Such faunal differences are probably not the result of provinciality. Afanasieva, M.S., 1997. Biostratigraphic significance of some Early Frasnian radiolarians, 3. Eighth Meeting of the International Association of Radiolarian Paleontologists abstracts: Paris/Beirville. Aitchison, J.C., 1988a, Late Paleozoic radiolarian ages from the Gwydir terrane. New England orogen, eastern Australia. Geology 793-795. Aitchison, J.C., 1988b. Middle-Late Devonian Radiolaria from the Yarrimie Formation, Tamworth Group, northeastern New South Wales, Australia, 4. In Schmidt-Effmg, R. & Braun, A. (eds) First International Conference on Radiolaria (EURORAD V). Geologica et Palaeontologica. Aitchison, J. C., 1988c. Radiolaria from the southem part of the New England Orogen, 49-60. In Kleeman J. D. (ed.) New England Orogen Tectonics and Metallogenesis. University of New England, Armidale. Aitchison, J. C., 1993a. Albaillellaria from the New England orogen, Eastern NSW, Australia. Marine Micropaleontology 15, 353-368. Aitchison, J. C., 1993b. Late Devonian (Frasnian) Radiolaria of the Canning Basin, Western Australia. Palaeontographica 228A, 105-128. Aitchison, J. C., and Flood, P. G., 1992. Implications of radiolarian research for analysis of subduction complex terranes in the New England Orogen, NSW, Australia. Palaeogeography, Palaeoclimatology, Palaeoecology 96, 89-102. Aitchison, J. C. & Flood, P. G., 1994. Gamilaroi terrane: a Devonian rifted intra-oceanic island arc assemblage, NSW, Australia. In Smellie, J. L. (ed.), Volcanism associated with extension at consuming plate margins. Geological Society, London Special Publication 81,155-168. Aitchison, J. C. & Stratford, J. M. C., 1997. Middle Devonian (Givetian) Radiolaria from Eastern New South Wales, Australia: a reassessment of the Hinde (1899) fauna. Neues Jahrbuch fiir Geologic und Palaontologie Abhandlungen 203, 369-390 Aitchison, J. C. Flood, P. G. & Spiller, F. C. P., 1992. Tectonic setting and paleoenvironment of terranes in the southem New England orogen as constrained by radiolarian biostratigraphy. Palaeo geography, Palaeoclimatology, Palaeoecology 94, 31-54. Aitchison, J. C., Stratford, J. M. C. & Spiller, F. C. P. (in press). A Lower and Middle Devonian radiolarian biozonation from the Gamilaroi terrane, New England Orogen, eastem Australia. Micropaleontology.

David, T. W. E. & Howchin, W. 1896. The occurrence of Radiolaria in Palaeozoic rocks in N. S. Wales. Proceedings of the Linnean Society of New South Wales 4, 553-570. David, T. W. E. & Pittman, E. F. 1899. On the Palaeozoic radiolarian rocks of New South Wales. Geological Society of London Quarterly Journal 55,16-37. Dongal, G. M. S., 1995. Early Devonian (Pragian and early Emsian) fauna from the eastern Tamworth terrane. New South Wales. Memoirs of the Association of Australasian Palaeontologists 18,131-142. Flood, P. G. & Aitchison, J. C., 1988. Tectonostratigraphic terranes of the southern part of the New England Orogen, 711. In Kleeman, J. D. (ed.). New England Orogen Tectonics and Metallogenesis. University of New England, Armidale. Flood, P. G. & Aitchison, J. C., 1992. Late Devonian accretion of the Gamilaroi Terrane to eastern Gondwana: provenance linkage suggested by the first appearance of Lachlan Fold Belt-derived quartzarenite. Australian Journal of Earth Sciences 9, 539-544. Hinde, G. J. 1899. On the Radiolaria in the Devonian Rocks of New South Wales. Quarterly Journal of the Geological Society of London 55, 38-64. Ishiga, H., 1988. Paleontological study of radiolarians from the southern New England Fold Belt, Eastern Australia, 7793. In Iwasaki, M., lizumi, S., Watanabe, T. & Ishiga, H. (eds). Preliminary Report on the Geology of the New England Fold Belt, Australia, 1. Co-operative Research Group of Japan and Australia, Matsue, Japan. Ishiga, H. & Leitch, E. C., 1988. Stratigraphy of the western part of the Hastings Block, New England Fold Belt, Eastern Australia, 33-45. In Iwasaki, M., lizumi, S., Watanabe, T. & and Ishiga, H. (eds). Preliminary Report on the Geology of the New England Fold Belt, Australia, 1. Co-operative Research Group of Japan and Australia, Matsue, Japan. Ishiga, H., Leitch, E. C., Watanabe, T. & Iwasaki, M., 1988. Radiolarian and Conodont biostratigraphy of siliceous rocks from the New England Fold Belt. Australian Journal of Earth Sciences 35, 73-80. Ishiga, H., Leitch, E. C., Naka, T., Watanabe, T. & Iwasaki, M., 1987. Late Devonian Paleoscenidiidae from the Hastings block. New England Fold Belt, Australia. Earth Science (Chikyu Kagaku) 41, 297-302. Metcalfe, I., Aitchison, J. C. & Stratford, J. M. C., 1997. Lower Devonian (Emsian) microfauna from the Gamilaroi Terrane at Glenrock in the southern New England orogen. New South Wales. Proceedings of the Linnean Society of New South Wales 118, 123-130. Nazarov, B.B., 1988. Radiolyarii Paleozoya. Prakticheskoe Rukovodstvo po Mikrofaune SSSR. Leningrad, Nedra, 232 p. Nazarov, B. B. & Ormiston, A. R., 1983. Upper Devonian (Frasnian) radiolarian fauna from the Gogo Formation, Western Australia. Micropaleontology 29, 454-466. Nazarov, B. B., Cockbain, A. E. & Playford, P. E., 1982. Late Devonian Radiolaria from the Gogo Formation, Canning Basin,Westem Australia. Alcheringa6, 161-173. Spiller, F. C. P., 1992. Radiolarian biostratigraphy of the western Hastings Block, N. S. W. MSc. (Prelim.) thesis. University of New England, Australia, (unpubl). Spiller, F. C. P., 1993. Radiolarian ages of sediments in the western Hastings Block, NSW, 243-246. In Flood P. G. & Aitchison J. C. (eds.) New England Orogen, eastern Australia. Publications of the Department of Geology and Geophysics, University of New England, Armidale. Stratford J. M. C., 1995. The formation of an intra-oceanic island arc rift: A case study based on the SiluroDevonian Gamilaroi terrane at Glenrock Station, southern New England orogen. New South Wales. Ph.D thesis, University of Sydney, (unpubl.). Stratford, J. M. C. & Aitchison, J. C., 1996. Devonian intra-oceanic arc rift sedimentation - facies development in the Gamilaroi terrane. New England orogen, eastern Australia. Sedimentary Geology 101,173-193. Stratford, J. M. C. & Aitchison, J. C., 1997. Lower to Middle Devonian radiolarian assemblages from the Gamilaroi terrane, Glenrock Station, NSW, Australia. Marine Micropaleontology 30, 225-250.

LATE TRIASSIC FLORA OF THE LEIGH CREEK COAL MEASURES, SOUTH AUSTRALIA Eroia BARONE-NUGENT, Andrew DRINNAN & Stephen McLOUGHLIN School of Botany, University of Melbourne, Parkville, VIC 3052 The Leigh Creek Coalfield is located in central South Australia about 550 km north of Adelaide. The coalfield is comprised of five discrete fault-bounded basins (Lobes A-E) in which sedimentation initiated at slightly different times during the Late Triassic. Lobes A and E remain covered by Quatemary alluvium and past mining operations have completely removed the coal-bearing sequences from Lobes C and D. Mining activity is now restricted to Lobe B (also known as the Telford Basin). The Telford Basin contains approximately 1000 m of non-marine sediments. Three coal-rich intervals occur within this basin, namely the lower, middle and upper coal series. These intervals are separated by thick siltstone- and sandstone-dominated sequences with abundant siderite concretions. This present study focuses on analysis of the fossil floras from sediments within and immediately above each of the coal-bearing intervals within this basin. The Leigh Creek Coal Measures contain macrofossil flora of comparatively low diversity and which is overwhelmingly dominated by pteridosperms of the family Corystospermaceae. Minor constituents of the macroflora include ginkgophytes, conifers, equisetalean sphenophytes, and osmundaceous ferns. Fossilised freshwater molluscs {Unio spp.) and fish {Leighiscus hillsi) have been found in the lower part of the Telford Basin succession, indicating that oxygenated aquatic conditions were present. A collection of macrofossil foliage from the Leigh Creek Coal Measures was first described at the end of the last century by Etheridge (1895), and this collection was revised by Chapman & Cookson (1926). Subsequent investigations have included a palynological study to determine the age of the coal deposit, and a taxonomic analysis of some of the spore-pollen species (Playford & Dettman 1965). However, no previous study has fully documented the plant macro- or mesofossil assemblages from the coal measures. Although this study is at an early stage, initial results indicate that floristic changes occur through the coal measure succession and that these changes may be related to shifting sedimentary facies and differing successional states within the Triassic coal-forming environments. Most notable is the decline in animal remains and shifts in the representation of corystosperm (Dicroidium) species between the lower and middle series coals. The extended aim of this study is to thoroughly document the macro- and mesofossil components of the palaeoflora of the Telford Basin and interpret the floristic and depositional changes within the succession. Chapman, F. & Cookson, LC., 1926. A revision of the 'Sweet' collection of Triassic plant remains from Leigh's Creek, South Australia Transactions of the Royal Society of South Australia 50, 163-178. Etheridge, R. Jr, 1895. Additional plant remains from the Leigh Creek Coalfield Central Australia. Transactions of the Royal Society of South Australia 19, 138-145. Playford, G. & Dettman, M.E., 1965. Rhaeto-Liassic plant microfossils from the Leigh Creek Coal Measures, South Australia. Senckenbergiana lethaea 46, 127-181

REVISION OF THE FAMILY CATILLICEPHALIDAE Raymond, 1938 (TRILOBITA) Christopher BENTLEY & Jim JAGG Department of Applied Geology, School of Engineering, University of South Australia, The Levels, SA 5095. The Catillicephalidae has been a dumping ground for Middle and Upper Cambrian trilobites of small size and having 'plump', expanded glabellae. In this paper we examine the relationships between the various taxa which have been included in the family. The Catillicephalidae as presently constituted contains the only known example of trilobites with a rostral plate or a ventral median suture being included in the same family. As the ventral median suture is believed to have arisen only once (Fortey & Chatterton, 1988), it appears unlikely that it could appear separately in the Catillicephalidae and the Asaphacea ( a monophyletic suborder of trilobites). It is much more likely that those trilobites that have this structure do not belong in the Catillicephalidae. The family as presently viewed is therefore polyphyletic, or at least, diphyletic. It is a difficult group to analyse, there being such a degree of homeomorphy. This is bome out by the results of cladistic analysis. Thirty one taxa were included, with two outgroups - Syspacephalus (a ptychopariid) and Anomocare (an anomocarid Asaphina). In an analysis of all taxa, two branches showed strongly. These were separated at Anomocare and run again, to give better resolution. The trilobites in the first group resolved into two groups, containing taxa referred to the Onchonotopsidae Shaw, 1966, and the Catillicephalidae. The former contained some trilobites which have been referred to the Lonchocephalidae Hupe, 1953. Lonchocephalus itself appears to be closer to the outgroup and is retained in the Lonchocephalidae. The Catillicephalidae contains Catillicephala, Buttsia, Pemphigaspis, Distazeris and Onchonotellus. These taxa have all been traditionally referred to the family. Welleraspis plots near the base of the tree and could be considered close to the ancestral form of the two groups, as could the form illustrated by Robison (1964) as genus and species indet. 3. Pratt (1992) suggested this form as possibly ancestral to the catillicephalids. Neither of these forms is referred to either family. These two groups (the Catillicephalidae and Onchonotopsidae) probably represent diverging evolutionary paths from closely related ancestors of a generalised ptychoparioid stock (represented in this analysis by Syspacephalus, suggested by Robison (pers. comm., 1997). The two groups could be united as subfamilies in a single family, as has been suggested by Pratt (1992), but given the evident divergence between the two (glabella expanding forwards in the Catillicephalidae, rearwards in the Onchonotopsidae, for example) it is probably best to separate them. Both groups appear in the late Middle Cambrian. The Onchonotopsidae were extinct by the end of the Dresbachian. Of the Catillicephalidae, Onchonotellus survived beyond the Franconian, into the Lower Ordovician. The second tree resolved to the extent that all of the taxa analysed are placed in an expanded Isocolidae Angelin, 1854, with the exception of Triarthropsis, which plotted closer to Anomocare (the outgroup) and is therefore not included in the Isocolidae. Its placement is uncertain, although it would be within the Asaphina, as is the placement of the Isocolidae.

This group probably represents a second expansion by trilobites into the niches previously inhabited by the Catillicephalidae, involving again small size and the forward expansion and inflation of the glabella. There is no analogue of the Onchonotopsidae in the late Upper Cambrian. This group first appeared in the Trempealeauian and was extinct by the end of the Ordovician,

Fortey, R.A. & Chatterton, B.D.E., 1988. Classification of the trilobite suborder Asaphina. Palaeontology 31, 165-222. Pratt, B.R., 1992. Trilobites of the Marjuman and Steptoean stages (Upper Cambrian), Rabbitkettle Formation, southern Mackenzie Mountains, northwest Canada. Palaeontographica Canadiana 9, 179 pp. Robison, R.A., 1964. Late Middle Cambrian faunas from western Utah. Journal of Paleontology 38, 510566.

A PORACANTHODID ACANTHODIAN FROM THE SILVERBAND FORMATION, THE GRAMPIANS, VICTORIA, AND ITS PALAEOBIOGEOGRAPHICAL SIGNIFICANCE Carole J. BURROW Vertebrate Palaeontology Laboratory, Department of Zoology, University of Queensland, QLD 4072 Rare gnathostome scales co-occur with scales of the agnathan thelodont Turinia fuscina in the sandstones of locality 3, unit 3 {sensu Talent & Spencer-Jones 1963) in the Silverband Formation of the Grampians. The scales derive from an ischnacanthid acanthodian Poracanthodes sp. cf. P. qujingensis, Tumer (1986) revised the age of the Silverband fauna to Late Silurian/Early Devonian, based on similar associations of thelodontid agnathans, climatiid and ischnacanthid acanthodians, lingulids, and ostracods in northern hemisphere deposits of this age. She tentatively assigned the few gnathostome scales identified at that time to Gomphonchus or Poracanthodes, while mentioning the possibility that they could be from a shark such as Antarctilamna, Thirteen scales have now been identified, mostly preserved in natural section on exposed surfaces. They differ from scales of Antarctilamna in having a convex base, canals in the posterior crown radiating from the centre of the scale, and pore rows on the crown surface; they also differ in lacking the centripetal arrangement of crown growth zones. Although the crown ornament is preserved on only a few scale fragments, the scales are closely comparable in all features to those of Poracanthodes qujingensis from the Late Silurian (Ludlow, crispa zone) of the Miaogao Formation, Yunnan, south China (Wang & Dong 1989), supporting the Late Silurian age for the Silverband fauna suggested by Turner (1986). Similar scales have also been reported from the Late Silurian (Ludlow, crispa-eosteinhomensis zones) in the Cumnock area of New South Wales (Parkes in Badsen et al, in press). Poracanthodids are useful in microvertebrate zonations, as they occur in both terrigenous and carbonate deposits. P. punctatus is a Pridoli index fossil in the Standard Silurian Microvertebrate Column (Marss et al. 1995), and several other species are quite widespread geographically. The occurrences of P. qujingensis and P, sp. cf. P. qujingensis support a close biogeographical relationship between Australia and south China during the Late Silurian/Early Devonian. Badsen, A., Burrow, CJ., Hocking, M., Parkes, R. & Young, G.C. (in press). Siluro-Devonian microvertebrates from southeastern Australia. In: A. Blieck & S. Turner (eds) IGCP 328 Final Report. Courier Forschungsinstitut Senckenberg. Marss, T., Fredholm, D., Karatajute-Talimaa, V., Turner, S., Jeppsson, L. & Nowland, G., 1995. Silurian Vertebrate Biozonal scheme. Geobios, Manoire Speciale 19, 369-372. Talent, J.A. & Spencer-Jones, D., 1963. The Devono-Carboniferous fauna of the Silverband Formation, Victoria. Proceedings of the Royal Society of Victoria 6, 1-11. Turner, S. 1986. Vertebrate fauna of the Silverband Formation, Grampians, western Victoria. Proceedings of the Royal Society of Victoria 98, 53-62. Wang N-z. & Dong Z-z. 1989. Discovery of Late Silurian microfossils of agnatha and fishes from Yunnan, China. Acta Palaeontologica Sinica 28, 192-206.

LOWER DEVONIAN BRACHIOPODS FROM DONG WUJIMQIN, INNER MONGOLIA, AND THE EXTENT OF THE BALKHASH-OKHOTSK PALEOBIOGEOGRAPHIC PROVINCE CHEN Xiuqin Nanjing Institute of Geology and Palaeotology, Academia Sinica^Nanjing The richly fossihferous Lower Devonian sequence at Dong Wujimqin, Inner Mongolia, is dominated by brachiopods originally studied by Su (1976,1980). A vast amount of new material has enabled significant improvement in knowledge of the Dong Wujimqin fauna. Serial sections are presented for many of the genera including Reeftonia, Tridensilis, Eatonia, Borealirhynchia, Wilsoniella, Spinatrypa, Pacificocoelia, Merista and Brachyspirifer. The generic position of all species, especially those formerly assigned to Reeftonia, Mesoleptostrophia, Eoschuchertella, Latonotoechia and Paraspirifer have been reconsidered. Also reconsidered, on the basis of larger populations, have been all cases where several species of a genus have been reported from a specific horizon at a single locality. The process of synonymisation has reduced the number of accepted taxa to 33 genera and 36 species. Many workers have discussed biogeographic patterns in the Lower Devonian biogeography of China, primarily on the basis of brachiopod faunas, but faunas previously documented, often on the basis of less than adequate material, have resulted in less than compelling, qualitative conclusions. New material has enabled a statistically underpinned biogeographic analysis of the region covering northem Xingjiang, Gansu, west and northeast Inner Mongolia, the Lesser and the Greater Khingan mountains, as well as adjoining regions and countries, to as far afield as central Kazakhstan. The formerly accepted Balkhash-Junggar and Mongolia-Okhotsk provinces are concluded to be parts of a single province, here referred to by the composite name Balkhash-Okhotsk province, the latter forming part of the Old World Realm.

TIME, SPACE AND ENVIRONMENTAL RELATIONSHIPS IN THE SILURIAN SUCCESSION OF BOREE CREEK, NSW. PETER COCKLE MUCEP, School of Earth Sciences, Macquarie University, NSW. The succession in the Boree Creek - Cheesemans Creek area appears to span most of the Silurian and possibly even extends into the earliest Devonian (Lochkovian). Apart from a small portion of the Borenore Limestone, chronologic relationships between the various lithographic units have yet to be determined. A study of conodonts from the Boree and Borenore Limestones was carried out to shed light on these relationships, especially between the Wallace Shale and the Borenore Limestone. Ten sections (144 sampled intervals) through the Boree Creek and Borenore Limestones; produced approximately 1000 conodont elements representing Apsidognathus, Kockelella, Oulodus, Distomodus, Pseudooneotodus, Ozarkodina, Walliserodus and Panderodus, Four species of Kockelella have been identified: Kockelella variabilis Walliser, 1957, K ranuliformis Walliser, 1964, K latidentata Bischoff, 1986 and K n. sp. A Klapper & Murphy, 1974. Together the conodonts suggest an age-range from late Llandovery in the Boree Limestone, to the Ludlow in the Borenore Limestone. Bedded limestone horizons from within the Wallace Shale yielded conodonts indicative of the variabilis Zone. The fauna and flora from the acid insoluble residues is very diverse. It includes several species of foraminifers, bryozoans, sponges, holothurians, the green alga Lancicula, brachiopods and an abundance of pelmatozoan ossicles.

BIOGEOGRAPHY OF QUATERNARY AND EXTANT LAMPROTHAMNIUM GROVES (CHAROPHYTA) IN AUSTRALIA. Adriana GARCIA* & Allan CHIVAS School of Geosciences, University of Wollongong, NSW 2522, Wollongong (*also Visiting Fellow from CONICET, ARGENTINA). The distribution within Australia of Lamprothamnium Groves (Charophyta) and the morphology of its gyrogonites have been studied from lakes from westem Victoria and several areas of South Australia and New South Wales. A statistical comparison has been made of modem, Holocene and Late Quaternary populations as a function of geographical distribution, ecology and climate. Lamprothamnium is found throughout the world being characterized by its paucity (only 4-6 species are known) and by its adaptation to environments with changing salinity, living in salinities from fresh to hypersaline. Two living species have been recognized in Australia, Lamprothamnium macropogon (A. Braun) Ophel and L. succinctum (A. Braun in Ascherson) R.D. Wood. L macropogon is widespread in Australia and has been found also in New Zealand, South Africa and China (Groves and Allen 1935; Wood & Mason 1977; van Raam 1995). Lamprothamnium succinctum has a wider distribution in tropical areas from Africa, South America and Asia and has been mentioned for the first time in Australia by Daily (1969) as the variety australiensis. Although the first record of this genus is from the Lower Cretaceous (Soulie-Marsche 1989), fossil Lamprothamnium has been studied only from two localities in Australia: Pleistocene sediments from Lake Eyre (Garcia and Chivas 1996) and Holocene sediments from Tom Thumb Lagoon, in Wollongong (Garcia 1996, ms). The mentioned references, and this work are the first studies of fossil Australian charophytes although it has been demonstrated that charophytes, known since the Upper Silurian, are very useful in chronology and stratigraphical correlations. Daily, F. K. 1969. A Ljmprothamnium succinctum (Characeae) with imperfect cortex. Bulletin of the Torrey Botanical Club 96, 656-660. Garcia, A., 1996. Charophytes from Tom Thumb Lagoon, Australia: an indicator of palaeoceanographic changes in the Holocene. Annual Meeting of IGCP 367, Sydney, p. 22-23. Garcia, A., (ms). Charophytes as indicators of changes in sea-level: mid-late Holocene sequence from Tom Thumb Lagoon, Australia Australian Journal of Botany. Garcia, A. & Chivas, A., 1996. Pleistocene charophytes from Madigan Gulf, Lake Eyre, South Australia. Second International Symposium on Extant and Fossil Charophytes, Madison, USA, p. 17. Groves, J. & Allen, G.O., 1935. A review of the Queensland Charophyta. Proceedings of the Royal Society of Queensland XLVI, 34-59. Soulie-Marche, L, 1989. Etude comparee de gyrogonites de charophytes actuelles et fossiles et phylogenie des genres actuels. Ed. Rev. PhD Thesis, 237 pp. Wood, R. D. & Mason, R., 1977. Characeae of New Zealand. New Zealand Journal of Botany 15, 87-180. Van Raam, J. C., 1995. The Characeae of Tasmania. Nova Hedwigia 110, 1-80.

99 PALYNOBIOSTRATIGRAPHY AND PALAEOBIOGEOGRAPHY OF THE UPPER PALAEOZOIC STRATA IN THE ZAGROS BASIN, SOUTHERN IRAN.

Mohammad GHAVIDEL-SYOOKI Exploration Division, National Iranian Oil Company, P.O. Box 1065, Tehran, IRAN During the past decade, the writer has carried out a detailed palyno-biostratigraphic study on the Palaeozoic strata in the Zagros Basin. This basin encompasses the southern part of Iran and most of the adjacent Persian Gulf, forming a SE-NW trending linear belt, approximately 1400 km long and 250 km wide, which is separated from the rest of the Iranian Platform by the 5-10 km wide Main Zagros Thrust or Crush Zone. Among the Palaeozoic rock units of this basin, the Faraghan Formation has received minimal interest since it lacks a marine fauna. Therefore, based on stratigraphic position, or some plant remains, it has been assigned a Lower Permian or Permo-Carboniferous age. The writer's palynological investigations have, however, revealed that part of the Faraghan Formation was deposited during the Devonian Period and another part in the Lower Permian time. As a result, the use of Faraghan Formation was dropped with the approval of the National Iranian Stratigraphic Committee and replaced by the Zakeen Formation and Chal-i-Sheh Formation. This paper aims to demonstrate some index palynomorph taxa from these rock units, in order to resolve some aspects of palaeobiogeographic relationships of the Zagros Basin during the upper Palaeozoic time. A total of 500 surface and subsurface samples from the Zakeen and Chal-i-Sheh Formations were treated for palynomorph entities. The 150 morphotype species that were encountered permit the recognition of five local stratigraphic assemblage zones. Zones I through IV appear in the Zakeen Formation, suggesting a Devonian age. Zone V occurs in the Chal-i-Sheh Formation representing the Lower Permian time. Therefore, based on palynological data, there is a hiatus between the Zakeen Formation (Devonian) and Chali-Sheh Formation (Lower Permian), which encompasses the Famennian and the whole Carboniferous Period. Some of the Devonian acritarch species of the Zakeen Formation have been recorded from Europe and North America, including Chomotriletes vedugensis, Chomotriletes bistchoense, Melikeriopola venulosa, Cymatiosphaera perimembrana and Polyedryxium decorum. However, 14 species of the Zakeen Formation have only been recorded from the Gneudna Formation (Frasnian) of Westem Australia, consisting of Deltotosma intonsum, Papulogobata annulata, Navifusa exilis, Lophosphaeridium segregum, Dictyotidium granulatum, D. confragum, Evittia geometrica, Elektoriskos tenuis^ Histopalla capillosa, Helosphaeridium guttatum, H. microclavatum, Gorgonisphaeridium discissum, G. condensum and Somphophragma micellum. The occurrence of the abovementioned acritarch taxa in the Devonian sediments of southern Iran, suggests that the Zagros Basin and westem Australia were at about the same palaeo-latitude along the southern shore of the Palaeo-Tethys ocean during the Devonian Period. Moreover, the lower Permian miospore species, derived from the Chal-i-Sheh Formation are: Corisaccites alutas, Boutakojfites elongatus, Mabuitasaccites ovatus, Striomonosaccites brevis, Striomonosaccites triangularis, Caheniasaccites ellipticus, C.

100

indicus, Marsupipollenites stratus, Hogiasaccites transitus, Walikalesaccites ellipticus, Fusacolpites ovatus, F. fusus, Tiwariasporis gondwanensis, T. flavatus, Distriamonocolpites ovalis, Decussatisporites circularis, Plicatipollenites indicus, F.densus, Bascanisporites undosus, Weylandites magnus, Microbaculispora tentula, Horriditriletes ramosus, Verrucosisporites andersonii and Punctatisporites gretensis. The above-mentioned miospore species from the Chal-i-Sheh Formation have also been reported from the Lower Permian of Africa, India and Australia. Therefore, based on palynological data, it would be reasonable to consider the Zagros Basin part of the gondwanan supercontinent.

101

SURFACE MICROSTRIATION ON LOWER ORDOVICIAN CONODONT ELEMENTS ANDREW KELMAN Australian Geological Survey Organisation, P.O. Box 378, Canberra ACT 2601 Abundant, well-preserved collections of discrete conodont elements have been recovered from the Ordovician (early Arenig) Emanuel Formation of the Canning Basin, Western Australia. On close examination the surfaces of many of these elements show longitudinal microstriations. These features vary in size, morphology and distribution. Elements with similar microstriation may belong to the apparatus of the same biologic species. Species with striate elements may be part of a lineage that represents a major division of the Euconodonta.

102

E A S T E R N MOLONG PLATFORM AND ADJACENT HILL END TROUGH, NEW SOUTH WALES: MID-PALAEOZOIC CONODONT DATA, AGE-INFERENCES, AND IMPLICATIONS R E G A R D I N G PLATFORM E X P O S U R E AND C A R B O N A T E SEDIMENTATION OFFSHORE RUTH MAWSON & JOHN A. TALENT Macquarie University Centre for Ecostratigraphy and Palaeobiology, School of Earth Sciences, Macquarie University, NSW 2109 Conodont data are presented from a dozen stratigraphic sections through mid-Palaeozoic carbonate-bearing units, principally from the Dripstone-Euchareena area on the west flank of the Hill End Trough, and from the Jesse Limestone at Limekilns in east-central New South Wales. The principal units sampled are the Early Devonian Cunningham Formation - including its Tolga, Red Hill and Nubrigyn members formerly construed as discrete formations - and the Jesse Limestone. Conodonts from the Tolga Member, an apron-like platform-margin sequence of interbedded shales, siltstones and allodapic carbonate grainstones draped over the Cuga Burga Volcanics, are typically sparse (a conodont per kg or less), except for the first 2-3 m of section, decrease in productivity up-sequence, and have a conspicuously high proportion of broken material testifying to substantial transport. The Tolga Member is suggested to represent carbonate debris swept intermittently eastwards from the adjacent Molong carbonate platform to interfinger with basinal shales. The age indicated by conodonts from this interval is late Lochkovian. A mudstone sequence with intermittent beds of fine-grained allodapic or ?hemipelagic limestones within the Cunningham Formation sensu stricto, stratigraphically intermediate between the Tolga and Red Hill members of produced well preserved early Pragian conodonts displaying a conspicuously lower level of breakage than conodonts from the Tolga Member. The Red Hill Limestone Member produced low but consistent yields of pireneae Zone (late Pragian) conodonts with a breakage ratio intermediate between the Tolga Member and the Cunningham Formation sensu stricto. A scatter of small lenses at about the same horizon towards Mumbil, interpreted as splays from the main Red Hill carbonate fan, also produced pireneae Zone conodonts, as did isolated clasts and runs of carbonate clasts in the Cunningham Formation sensu stricto below the Red Hill Limestone Member. The latter, lithified in the source area prior to emplacement, herald development of the carbonate channel deposits within the major Red Hill submarine fan. Conodonts from the Nubrigyn Member are mostly from limestone boulders in debrisflows, from allodapic limestones, and from isolated limestone megaclasts. Though mostly not compelling as to age, some of these have produced chronologically constraining conodont data indicative of late Lochkovian\Pragian carbonate sedimentation in the source areas, presumably the Garra Limestone of the adjacent Molong carbonate platform. Some of the clasts and\or debris-flow carbonates from high in the Nubrigyn sequence were derived from carbonate bodies post-dating the carbonate record preserved on the Molong

103

Platform. They are evidence for persistence of a former tract or tracts of shallow water carbonate sedimentation on the platform through much or all of Emsian time. We now have evidence from autochthonous sections in the Garra Limestone on the Molong High for continuous carbonate sedimentation through at least 6 conodont zones, delta to dehiscens, and infer from the ages of allochthonous carbonates in the Hill End Trough to the east, that carbonate sedimentation persisted on the Molong Platform for at least 3 more conodont zones, from perhonus Zone to early in the serotinus Zone. Also presented are conodont data from a ?deeper context within the Hill End Trough, from a late Emsian carbonate fan (?emplaced by grain-flow), the Jesse Limestone, at Limekilns c. 60 km east of Euchareena. It proved to be serotinus Zone though with early dehiscens Zone micrite clasts low in the succession testifying to a former but now "lost" deeper water carbonate biofacies of that age formerly existing somewhere upslope - in which direction from Limekilns is problematic. Silurian carbonate units encountered while mapping the western flank of the Nubrigyn outcrop-tract were sampled for conodonts, mostly without encouraging results; conodont data are presented for one of these, a tongue of Narragal Limestone, or a carbonate channel deposit in the Barnby Hills\Wallace Shale consisting of apparently Narragalderived limestone, allochthonous at least in its basal beds. Silurian carbonate clasts from within the Nubrigyn Member, though mostly not compelling as to precise age, include a large megaclast of Wenlock or Ludlow age high in the Nubrigyn sequence north-west of Euchareena. Viewed regionally, the sampling of carbonate sequences offshore in the Hill End Trough has produced results consistent with a pattem of reciprocal sedimentation: development of carbonate fans within the Hill End Trough correlative 'time-wise' with regressive events on the adjacent carbonate platform - where they tended to be expressed as units of massive or poorly bedded limestones, often with fenestral fabrics indicative of sabka-type sedimentation. It is also suggested that the Nubrigyn Member sedimentation may have been primarily tectonically driven, its rhythmic character reflecting intermittent movement on an advancing deformation front to the west. *This is a contribution to IGCP 421 Mid-Palaeozoic biogeographyXbioevent patterns in relation to crustal dynamics.

104

CRAWNEY LIMESTONE - A DETERMINATION OF CONODONT BIOSTRATIGRAPHY USING POLYGNATHIDS BOB MORGAN MUCEP, School of Earth Sciences, Macquarie University, NSW Conodont data obtained from acid-leaching of limestone samples from a stratigraphic section has shown the Crawney Limestone (Yarrimie Formation, Tamworth Group) to be of a late Eifelian age. This is in accordance with the age given by Pedder (1968). On the basis of coral faunas Pedder suggested correlation of the Crawney Limestone with the Moore Creek Limestone and the Timor Limestone. Polygnathid conodonts indicate the Crawney Limestone was deposited during kockelianus-ensensis time. The presence of Tortodus kockelianus kockelianus, a zonal conodont, throughout the sampled section supports a kockelianus age. While the occurrence of Polygnathus eiflius, P. trigonicus, P, angusticostatus and P, pseudofoliatus supports a late kockelianus to early ensensis age, there has been no occurrence of any polygnathid species to mark the beginning of the ensensis zone. A comparison of the conodont faunas of the Crawney Limestone and Moore Creek Limestone, at Moore Creek (Philip 1966) and Wyaralong and Attunga (Mawson & Talent 1994), has shown that the main elements of the faunas are polygnathids and belodellids. Even though, there is a greater abundance of Tortodus kockelianus kockelianus within the Crawney Limestone, the similarity of the overall composition of the faunas suggests that the Crawney Limestone represents at least the lower portion of the Moore Creek Limestone. Mawson R. & Talent, J.A., 1994. The Tamworth Group (mid-Devonian) at Attunga, New south Wales: conodont data and inferred ages. Courier Forschungsinstitut Senckenberg 168, 37-59. Pedder, A.E.H., 1968. The Devonian System of New England, New South Wales, Australia. In Oswald, D.H. (ed.) International Symposium on the Devonian System, 135-142. Philip, G.M. (1966) Middle Devonian conodonts from the Moore Creek Limestone, northern New South Wales. Journal and Proceedings of the Royal Society of New South Wales 100, 151-161.

105

COMPARATIVE PALAEOECOLOGY OF CAINOZOIC RIVERSLEIGH FAUNAS Troy MYERS School of Biological Science, University of New South Wales, Kensington, NSW 2052 Presently over 300 fossil sites, spanning the late Oligocene to late Pleistocene, are recognised in the Riversleigh World Heritage fossil deposits. To date few studies have examined the spatial and temporal synecology of Riversleigh fossil faunas. Species lists have been compiled for four local palaeofaunas, and are in progress for a further six. The selection of local palaeofaunas to be investigated was influenced by the uniformity of lithology and degree of horizontal and vertical confinement of each site, as well as by the sampling effort already undertaken. An attempt was also made to include a suite of sites that covered the temporal range of the Riversleigh deposits. After cataloguing approximately 2000 cranio-dental elements, NISP (the number of identifiable specimens per taxon) values were determined for all known Family, Genus and Species level taxa at Wayne's Wok (WW), Neville's Garden (NG), Cleft of Ages (COA) and Keith's Chocky Block (KCB) sites. Chi-square analysis of the NISP-based relative abundance distributions of these local palaeofaunas, suggests they are not homogeneous as regards the abundances of their component taxa. However, two-sided Kolmogorov-Smimov tests and ANOVA suggest that there is some underlying similarity in structure between WW and NG sites, and between KCB and COA sites. Ecodiversity indicators, incorporating taxonomy, diet and body-size, were used in conjunction with relative abundance analyses to compare palaeocommunity structure between sites and to potential extant analogues. In addition, cenogram analysis (Legendre 1986) was employed for comparing palaeocommunities with one another and with extant communities, and for the determination of palaeoenvironments. Preliminary results from ecodiversity indicators and cenogram analysis suggest a similarity in palaeocommunity structure between KCB & COA, and between NG & WW local palaeocommunities, thereby supporting the similarity observed in relative abundance distributions for these sites. At this stage it is only possible to say that these site combinations are at the very least indicative of separate palaeocommunity types. Separating differences resulting from taphonomic factors and real palaeoecological factors may allow for further classification of these site combinations into palaeocommunities. Legendre, S. 1986. Analysis of mammalian communities from the late Eocene and Oligocene of southern France. Palaeovertebrata 16, 191-212.

106

NEW AND REVISED FOSSIL PLANTS FROM THE LOWER CRETACEOUS BOOLA BOOLA FLORA, GIPPSLAND BASIN, VICTORIA Nathalie NAGALINGUM, Andrew DRINNAN & Stephen McLOUGHLIN School of Botany, The University of Melbourne, Parkville, VIC 3052. Early Cretaceous (Neocomian-Barremian) plant macrofossil assemblages from Boola Boola, Gippsland Basin, contain gymnosperms, lycophytes and ferns. Conifers are the dominant elements in the flora and incorporate Brachyphyllum gippslandicum McCoy, Pagiophyllum victoriensis sp. nov., and a new species of Elatocladus. Pteridosperms are also important elements of the flora but in most cases the suprageneric affinities of the taxa are uncertain due to a lack of attached reproductive organs. A new genus is erected for pteridosperms of unknown affinity and includes Reinitsia variabilis Douglas and Thinnfeldia pinnata Walkom. Pteridosperm foliage previously attributed to Pachypteris austropapillosa Douglas is reassigned to Pachydermophyllum based on a reassessment of the frond morphology and stomatal organization. Taeniopteris daintreex (a pentoxylalean), Thinnfeldia sp. cf. T. chunakalensis Sah & Dev (of uncertain affinities), and Palissya elegans Parris, Drinnan & Cantrill (a coniferous or pteridospermous strobilus) are sub-dominant elements of the flora. The palynoflora from the Boola Boola sediments comprises a diverse array of fern, bryophyte, lycophyte, pteridophyte, Cycadales/Bennettitales/Pentoxylales and conifer palynomorphs. Not all of the palynofloral groups are represented or are distinguishable in the macrofloras. Both the macroflora and palynoflora from Boola Boola have limitations when used separately to estimate past floristic diversity. A combination of the macrofloristic and palynofloristic data provides a more complete picture of the flora. Macrofloral assemblages underrepresent lycophytes while palynoassemblages underestimate the diversity of pteridosperms. The Boola Boola assemblage has generic, and some specific, similarities to the gymnosperm-dominated Early Jurassic Talbragar flora, particularly in the representation of Palissya, Taeniopteris, Elatocladus, and various pteridosperm and fern taxa. Comparison of these floras indicates that little floristic change occurred between the Early Jurassic and Early Cretaceous, or that the age of the Talbragar flora may require reassessment. Notable differences between the Boola Boola flora and the BarremianAptian Koonwarra assemblage, such as the greater diversity of pteridosperms in the former and the presence of ginkgophytes and angiosperms in the latter, implies a more rapid turnover of species during the Early Cretaceous. Alternatively, some of these differences may be a consequence of fluvio-lacustrine facies controls on the composition of macro-plant assemblages.

107 A LATE SILURIAN ACANTHODIAN FROM COEVAL CARBONATES IN VICTORIA AND NEW SOUTH WALES *

Ross PARKES^ & Andrew SIMPSON^ ^ Centre for Ecostratigraphy and Palaeobiology, School of Earth Sciences, Macquarie University NSW 2109 ^ Department of Earth Sciences, the University of Queensland, St Lucia QLD 4072 Microvertebrate remains of the Late Silurian acanthodian Poracanthodes cf. P, qujingensis Wang N-z & Dong Z-z have been recovered from the well-known ocurrence of Cowombat Formation limestone at Native Dog Plain in eastern Victoria, and from unnamed late Silurian limestones near the Bell River west of The Gap, between Molong and Wellington, NSW. The Native Dog Plain occurrence is one of many, often substantial lenticular Silurian (Llandovery-Pridoli) limestones outcropping in the headwaters of the Indi and Buchan Rivers. Conodont data indicate a Ludlow to Pridoli age for this particular unit (Simpson & Talent 1995). The fish remains were obtained from low in the unit where conodont faunas are dominated by long-ranging taxa such as Ozarkodina excavata excavata and Panderodus unicostatus. Some 100 m stratigraphically higher than the fish occurrence, a small number of the latest Ludlow to Pridoli index form Ozarkodina remscheidensis eosteinhomensis has been documentedi (Simpson & Talent 1995). More significantly, a single occurrence of the Ludlow Coryssognathus dubius is recorded from above the' fish horizon. The unnamed limestone near the Bell River east of The Gap occurs within a predominantly clastic sequence broadly equated with the Bamby Hills Shale and approximately coeval with the Native Dog Plain occurrence. Conodont faunas from this unit are presently under study (J.R. Farrell, Macquarie University). Poracanthodes qujingensis was previously reported from the Late Silurian crispus Zone Miaogao Formation of eastem Yunnan. Its presence in south-eastern Australia is consistent with other instances of Late Silurian Sino-Australian palaeogeographic similarities. Simpson, A.J. & Talent, J.A., 1995. Silurian conodonts from the headwaters of the Indi (upper Murray) and Buchan rivers, southeastern Australia, and their implications. Courier Forschungsinstitut Senckenberg 182, 79-215. This is a contribution to IGCP 421.

108

THE PERMIAN FLORA OF THE MOUNT MULLIGAN COAL MEASURES, QUEENSLAND, AUSTRALIA John RIGBY School of Natural Resource Sciences, Queensland University of Technology, GPO Box 2434, Brisbane QLD 4001. Mid to Late Permian floras from the Mount Mulligan Coal Measures are described. The assemblages include species of Glossopteris restricted to the coal measures, namely G. jenseni sp. nov, G. mulligani sp. nov, G. sundesmites sp. nov, with G. ballii sp. nov. also known from the Bowen Basin, central Queensland. The presence of Trizygia speciosa and Cyclodendron leslii indicate groth as a tropical or subtropical flora in a lake depositional system. C. leslii grew in a mangrove-like environment. The otherm, noxi-Glossopteris species, are widely distributed both in area and age, in many parts of Queensland.

109

THE FAUNA AND FLORA OF THE EOCENE REDBANK PLAINS FORMATION: SUMMARY OF RECENT PROGRESS Alan RIX^ & Andrew SIMPSON^ ^ Faculty of Arts, University of Queensland ^ Geology Museum, Department of Earth Sciences, University of Queensland Over the last few years, the fine-grained ferruginous mudstones of the Redbank Plains Formation, west of Brisbane, have been the subject of a rekindling of palaeontologic interest largely driven by the fossil finds of amateur collectors. The mudstones are thought to have been deposited in a fresh-water lake. The Formation is overlain by basalts dated at 46.8 million years by K/Ar methods and a middle Eocene age is inferred. Fish discovered from the site include Percalates antiquus, Phareoides queenslandicus (originally described by Hills, 1935) and the lungfish Neoceratodus gregoryi. Recent discoveries also include the plastron, pelvic bones, costal, peripherals and carapace of a number of chelid turtles. This extends the antiquity of the Chelidae in Australia considerably, the oldest previous finds were from Miocene strata. This material is currently being prepared for publication. Other vertebrate material includes impressions of the pedal digits of a large dromorthid (Vickers-Rich & Molnar, 1996), the specimen represents one of the oldest skeletal fossils of a bird yet recorded in Australia. A rich invertebrate fauna has also been collected from the site. Insects include Aneural apicalis, Austropanorpa australis, Centrocnemisl imperfecta, Chorista sobrina, Euporismites balli, Protasmanina nana, Protolimnophila superba and Scolipopites bryani (Rix, 1991). The insects are the subject of further study at present. A number of ostracods and the bivalve Unio sp. have also been found. Recently recovered plant specimens are also the subject of on-going study. These include Grevillia sp.. Eucalyptus sp. a serrated-edged "Banksia"' type, palm fronds and unidentified Proteaceans. It is anticipated that the Redbank Plains fossil site will become a special scientific reserve under the trusteeship of the Ipswich City Council with the agreement of the current landowners, thus enabling its full scientific potential to be developed. Hills, E.S., 1935. Tertiary fresh water fishes from southern Queensland. Memoirs of the Queensland Mw^^wm 10, 157-174. Rix, A., 1991. A re-examination of the Redbank Plains Tertiary fossil site, Queensland. The Fossil Collector 32/33(1), 31-36, Vickers-Rich, P. & Molnar, R., 1996. The foot of a bird from the Eocene Redbank Plains Formation of Queensland, Australia. Alcheringa 20, 21-29.

110

THE CRETACEOUS OF AUSTRALIA: THE BIOSTRATIGRAPHIC SYSTEM Samir SHAFIK Australian Geological Survey Organisation

Biostratigraphic dating of sedimentary sequences presumes that, initially, the biozonation being applied is without serious problems, represents a time continuum, and has been firmly linked (at least partly) to other biozonations and timescales; this is certainly applied to the Cretaceous sequences of Australia. The pressing need for numeric ages in other disciplines of geology, which use time lines to identify sedimentary packages (e.g. seismic stratigraphy, geological modelling, etc.), has resulted recently in numeric ages being arbitrarily assigned to many local Cretaceous zonal boundaries. The usual problems of precision in zonal correlation, such as the concept of the acme zone, or boundaries being defined by two or more mutually exclusive bioevents etc., and biostratigraphic problems caused by inadequate palaeogeographic data, seem to have magically disappeared, as have possible problems related to the use of informal local zones whose definitions have not been published. A concerted effort by all local biostratigraphers is needed to address this serious deficiency in Australian Cretaceous biostratigraphy. Attention should probably be given initially to identifying age anchors for local biozones (strong ties with other fossil group zonations, both local and global, and with timescales) and weaknesses in links between the various local zonations (foraminifera, calcareous nannofossil, spores and pollen, dinoflagellate, ammonite etc.) and their global counterparts. However, the formal definition, by publication, of local zones and their reference sections (e.g. the nannofossil KCN and most foraminiferal C zones) is a prerequisite for any serious attempt at evaluating the level of maturity of the biostratigraphic system in Australia. Precise determination of the geographic distribution of biomarkers, their stratigraphic ranges, the stratigraphic resolution of individual biozonation, and its applicability in various basinal settings, etc. could follow. The ages of many local dinoflagellate zones, particularly in NW Australia, have been invariably linked to the perceived ages of the local calcareous planktic zones. However, a cursory examination of the stratigraphic ranges of several key foraminiferal species, allegedly used to determine the foraminiferal C zones in two continuously cored ODP sites in NW Australia (762 and 761), throws some doubt on the validity of these zones; notwithstanding that the stratigraphic ranges of some calcareous planktic species in NW Australia differ from those generally accepted elsewhere. The coincidence of formational boundaries in the Cretaceous Sherbrook Group of the Otway Basin with the spore and pollen and dinoflagellate zonal boundaries is another example of the potential problems in Australian Cretaceous biostratigraphy.

Ill

LATE ORDOVICIAN CONODONTS FROM GRAY CREEK AND THE HEADWATERS OF STOCKYARD CREEK, NORTH QUEENSLAND * Andrew SIMPSON Department of Earth Sciences, University of Queensland, St Lucia, QLD 4072 The Carriers Well Formation is the westernmost unit of the Camel Creek Subprovince lying adjacent to the Gray Creek Fault Zone in the Broken River region of north Queensland (Withnall 1993). It is a heterogeneous unit with a wide variety of subordinate, sporadically outcropping carbonate lithologies. Some limestones have a macrofauna of crinoids, brachiopods, nautiloids and corals (Withnall 1993). Reconnaissance sampling by the Geological Survey of Queensland has recovered a small Late Ordovician conodont fauna from outcrops in Gray Creek and the head-waters of Stockyard Creek. Conodonts include Amorphognathus ordovicicus, Plectodina Iflorida, Taoqupognathus tumidus, Aphelognathus Ifloweri, Belodina confluens, Pseudobelodina sp., Panderodus gracilis, Panderodus panderi and Dapsilodus sp. Many of these taxa are cosmopolitan Late Ordovician forms; the presence of A. ordovicicus indicating an Ashgill age. The occurrence of Taoqupognathus tumidus, however, is restricted to China and Australia. Taoqupognathus ranges from middle Caradoc to early Ashgill strata. Zhen & Webby (1995) indicated that this genus is of considerable value for high precision correlation of Australian Late Ordovician sequences and noted that T. tumidus is a late form. The Carriers Well Formation is closest in age (early Ashgill) to the Fork Lagoon Beds of the Anakie Inlier where A. ordovicicus and T tumidus have also been recovered (Palmieri 1978). Palmieri, V., 1978. Late Ordovician conodonts from the Fork Lagoon Beds, Emerald area, central Queensland. Publication, Queensland Geological Survey 369, 1-31. Withnall, LW., 1993. Stratigraphy of the Camel Creek Subprovince, In Withnall, LW. & Lang, S.C. (eds). Geology of the Broken River Province, north Queensland. Queensland Geology 4, 15-34. Zhen, Y.Y. & Webby, B.D., 1995. Ordovician conodonts of the Cliefden Caves Limestone Group (Caradoc), central New South Wales. Courier Forschungsinstitut Senckenberg 182, 265-305. •

This is a contribution to IGCP 410 and IGCP 421.

112

EARLY SILURIAN CONODONTS FROM THE BROKEN RIVER REGION, NORTH QUEENSLAND * Andrew SIMPSON Department of Earth Sciences, the University of Queensland, St Lucia QLD 4072 A number of small, isolated, allochthonous carbonates outcrop within the aerially extensive, predominantly pelitic, Early Silurian Quinton Formation in the Broken River region of north Queensland. In the northern areas, the Quinton Formation is a thick, largely turbiditic sequence with carbonates preserved in the bed of Gray Creek at "Top Hut" and at Tomcat Creek 3.8 km ESE of "Top Hut". Both these limestones have an Early Silurian conodont fauna. In the vicinity of the Broken River Crossing, 26 km to the south, the lateral equivalent of the Quinton Formation is thinner and less turbiditic (Sloan et al, 1995). One isolated, presumably allochthonous, carbonate within pelitic lithologies at the Crossing has also yielded Early Silurian conodonts. The conodonts Pseudolonchodina expansa, Distomodus staurognathoides, Aulacognathus bullatus and Panderodus n. sp. A have been recovered from the Top Hut limestone. Whilst some of these taxa are relatively long ranging Early Silurian forms, the presence of A. bullatus constrains the age of this unit, with reasonable precision, to the late Llandovery celloni Zone; a marginally older age is, however, a remote possibility. The 60 m thick limestone sequence at Tomcat Creek has yielded the conodonts Pseudolonchodina fluegeli, Distomodus staurognathoides, Walliserodus curvatus, Ozarkodina cf. O. hadra and Panderodus n. sp. A. Although this fauna permits less chronological precision than at Top Hut, a generalised age extending from the late Llandovery celloni Zone to the late Llandovery to early Wenlock amorphognathoides Zone can be inferred. The small limestone at the Broken River Crossing has yielded Pseudolonchodina expansa, Distomodus staurognathoides, Oulodus jeannae and Astropentagnathus irregularis. The occurrence of A. irregularis allows an age diagnosis of late Llandovery celloni Zone or slightly older. These three allochthonous units therefore align chronologically, indicating probable derivation from the same general source. Jell et al (1993) summarised the known graptolite data from the same stratigraphic interval indicating late Llandovery (Telychian) ages ranging from the turriculatus to greistoniensis graptolite zones for numerous localities in the south. A slightly younger, but still late Llandovery age was suggested for graptolites from Top Hut. The lack of any major chronological disparity between the conodonts from the allochthonous limestones and the graptolites of the enclosing clastics indicates the penecontemporaneous nature of erosion and redeposition of these carbonates. Jell, J.S., Simpson, AJ., Mawson, R. & Talent, J.A., 1993. Biostratigraphic summary. In WITHNALL, I.W. & LANG, S.C. (eds). Geology of the Broken River Province, north Queensland. Queensland Geology 4, 239-245.

113

Sloan, T.R., Talent, J.A., Mawson, R., Simpson, AJ., Brock, G.A., Engelbretsen, MJ., Jell, J.S., Aung, A.K., Pfaffenritter, C., Ttrotter, J. & Withnall, LW. 1995. Conodont data from Silurian-Middle Devonian carbonate fans, debris flows, allochthonous blocks and adjacent autochthonous platform margins: Broken River and Camel Creek areas, north Queensland, Australia. Courier Forschungsinstitut Senckenberg 182, 1-77. * This is a contribution to IGCP 421.

114

A NEW MARSUPIAL FROM THE EARLY EOCENE TINGAMARRA LOCAL FAUNA OF MURGON, SOUTHEASTERN QUEENSLAND AND ITS SIGNIFICANCE IN THE UNDERSTANDING OF AUSTRALIAN MARSUPIAL ORIGINS STEPHEN WROE Vertebrate Palaeontology Laboratory, School of Biological Sciences, University of New South Wales 2052 Djarthia murgonensis, from the early Eocene Tingamarra Local Fauna of Murgon in southeastern Queensland, shows a combination of derived and plesiomorphic features that permit phylogenetic placement within either Didelphidae or Australidelphia. Detailed character analysis suggests that many features, previously treated as synapomorphies for both didelphids and australidelphians, are poorly supported. Because unassigned tarsal material attributable to both taxa is known from the Murgon deposits, D, murgonensis can not be placed in either clade with confidence. If this taxon is australidelphian it suggests that the ancestor of Australian marsupials and microbiotheriids can not, at present, be distinguished from basal ameridelphian taxa on the basis of dental synapomorphies. Alternatively, if the ameridelphian tarsal material is associated with D, murgonensis, then its placement within Didelphidae is could be argued. Either way, the presence of both meridelphian and australidelphian tarsal elements in the Eocene deposits of Murgon disproves recent claims that marsupial faunas of South America and Australia are manifestly distinct, excepting for the australidelphian affinity of South American microbiotheres.

115

THE EARLY PERMIAN AUSTRALASIAN MICROFLORA IN WEST YUNNAN AND ITS PALAEOBIOGEOGRAPHIC AND TECTONIC SIGNIFICANCE Weiping YANG Nanjing Institute of Geology and Paleontology, Academia Sinica, 210008, Nanjing, PEOPLE'S REPUBLIC OF CHINA The early Permian microflora study in West Yunnan, S.W.China shows that the spores and pollen content from the Tengchong Massif and the Baoshan Massif in West Yunnan were of Gondwana orgin. In particular, the palynological assemblages from West Yunnan can be correlated with those in Australia as well as the rest of Gondwanaland. In the Tengchong Massif, the TP miospore zone is indicated by Microbaculispora tentula - Jayantisporites pseudozonatus and included the following Australasian species: Microbaculispora tentula, Jayantisporites pseudozonatus,, Punctatisporites greatness, Verrucosisporites subsaccata, Vittatina fasciolata, Horriditriletes tereteangulatus, Protohaploxypinus sp., Propinguispora praetholus, Procoronaspora spinosa, Brevitriletes sp.,Spelaeotriletes sp. and Retusotriletes sp. This TP zone has been dated as early Permian (Asselian) and correlated with stage 2 and probably part of the Pseudoreticulatispora confluens Oppel Zone in Australia (Yang et al. 1996). The Pseudoreticulatispora confluens Oppel Zone comparable assemblage has been found in the Baoshan Massif. The palynomorphs extracted from the Dingjiazhai Formation are included as follows: Microbaculispora tentula, Pseudoreticulatispora confluens, Jayantisporites variabilis, Horriditriletes tereteangulatus, Indotriradites niger, Vittatina fasciolata, Weylandites lucifer, W. magmus, Protohaploxypinus limpidus, P. amplus, P. rugatus, Striatopodocarpites cancellatus, Striatoabieites multistriatus, Retusotriletes diversiformis, R, sp., Verrucosisporites cf. andersonii, Calamospora microrugosa, Limitisporites cf. rectus, Plicatipollenites spp., Auroraspora sp., Dictyotriletes sp., Interradispora sp., Sahnites spp. Due to the coexistence of Microbaculispora tentula, Horriditriletes tereteangulatus, Atriatoabieites multistriatus and Plicatipollenites spp., this palynological assemblage in Baoshan could be called Pseudoreticulatispora confluens Oppel-Zone in terms of Foster's (1988) definition. Although the palynological assemblages found in Tengchong and Baoshan are not so rich or abundant, it seems to be to correlate these assemblages with the stage 2 microflora and the P, confluens assemblage in Australia and even the whole of Gondwanaland based on some key taxa. Therefore, together with the sedimentological study (Wopfner & Jin 1994), the presence of the Gondwana realm of phytogeography in West Yunnan could be recognised. Further, the extent of Gondwana glaciation into West Yunnan could also be claimed. Finally, the different derivation from Gondwanaland for the Tengchong Massif and the Baoshan Massif is proposed. The Tengchong one is of the same origin as northwestern Australia, while the Baoshan one is inferred to be within the Gondwana geotectonic domain showing a much close relationship with southern Tibet and India.

116

IRONSTONE FACIES THROUGHOUT THE FRASNIAN-FAMENNIAN IN CENTRAL IRAN: A GONDWANA KEY IN EAST IRAN

M. YAZDI Department of Geology, University of Isfahan, Isfahan, IRAN The red sandy oolitic Cephalopod beds of the Shishtu Formation in the Shotori Range and in number of sections in Central Iran were systematically sampled for conodonts in order to establish an improved chronological framework. The age of these beds, including ironstone horizons, at Kale Sardar and Howz-e-Dorah was determined from the conodonts recovered (Yazdi, 1996). At Howz-e-Dorah, a basal 3-6 m interval is dated as early Famennian while 40 m higher, a horizon rich in cephalopods, is dated as expansa Zone (Yazdi, 1996). At Kale Sadar, the lower ironstone horizon is not represented as the section commences within the Cepalopod beds. 15 m above the base of the section at Kale Sadar, however, a fossiliferous brown to red oolitic limestone horizon is dated as expansa Zone. These distinctive horizons can be identified in other parts of Central Iran. Their location elsewhere should assist in the reconstruction of Gondwanan palaeoenvironments in the area. " Important palaeobiogeographic links between Iran, north-west Australia and north Africa in the Frasnian and Famennian are indicated by the goniatite faunas which include Beloceras tenuistriatum and Platyclymenia platyclymenia. These found in sandy oolitic limestone beds, the host lithology for goniatite groups in the Shotori Range, Africa and Australia (Becker & House 1994). Becker, T. & House, M.R., 1994. Kellwasser Events and goniatite succession in the Devonian of Montagne Noire, with comments on possible causation. Courier Forschungsinstitut Senckenberg 169, 45-77. Yazdi, M, 1996. Late Devonian-Carboniferous conodont biostratigraphy of the Tabas area, Eastern Iran. Unpublished PhD thesis, Macquarie University, Sydney, Australia, 221 pp.

117

RUGOSAN DIVERSIFICATIONS AND MIGRATIONS IN THE DEVONIAN OF AUSTRALIA Yong-yi ZHEN\ A.J. WRIGHT^ & J.S. JELL^ 1 Centre for Ecostratigraphy & Palaeobiology, School of Earth Sciences, Macquarie University, NSW 2109 2 School of Geosciences, University of Wollongong, Wollongong, NSW 2522 3 Department of Earth Sciences, University of Queensland, QLD 4072 Some 96 rugose genera have been recorded from various Devonian carbonate successions in Australia, mainly from Pragian to Frasnian marine shelf settings. They are dominated by disphyllids (21), ptenophyllids (12), spongophyllids (12), phillipsastreids (10), endophyllids (6), cystiphyllids (6) and stringophyllids (5), making up 75 percent of the total number of the genera. The biggest turnovers were in the Pragian and Givetian, and profound extinction events are marked in the late Emsian, late Givetian and late Frasnian; this is basically the same pattern as shown by the turnover rate curves of the Old World Realm and Eastern Americas Realm. The evolutionary innovation and diversification of the Early Devonian rugose corals in eastern Australia are characterized by a high turnover rate in the late Lochkovian Pragian and strong dynamism of radiation from late Pragian to mid Emsian, implying vigorous dispersal to south China, central Asia and Europe. After a high intensity of origination in the Pragian, the maximum diversity was reached in the Emsian. Phillipsastreids, and Devonian endophyllids like Endophyllum and Tabulophyllum, made their first appearance in the Pragian and became common in the Emsian. The earliest elements of the stringophyllids occurred in Emsian strata. As these elements are recorded mainly from the Middle, or even Upper, Devonian in other provinces of the Old World Realm, they may have originated in eastern Australia during the Early Devonian. Following a severe drop of generic diversity in the Eifelian, faunal diversity reached another high peak in the early-mid Givetian as a result of the influx of elements probably originated from south China, central Asia, Europe, northwestern Canada and other provinces of the Old World Realm. The Frasnian fauna is generally cosmopolitan and low diversity. It seems that diversifications of Australian Devonian rugose corals and the migrations out and in the region were strongly influenced by global events, specifically the eustatic sea level changes, as well as regional tectonic events. The latter, especially in the Tasman Fold Belt, resulted in severe drop in diversity in the Eifelian due to the regional regression and the significantly increasing of erosion which carries large quantities of terrestrial clastic sediments into shelf or near shore regions.

University of Wollongong

Key to Buildings 1 materials engineering 2 engineering 3 informatics 4 engineering 5 eastern classrooms 6 engineering laboratories 7 thermodynamics 8 mechanical engineering 9 engineering 10 kids uni (childcare centre) 11 Wollongong UniCentre-retail centre & other services 12 computer science 13 recreation & acquatic centre 14 central lecture theatre 15 Austin Keane building, Informatics, geology, ITS 16 library 18 chemistry/physics 19 arts 20 communications centre 21 education laboratories 22 journalism, education resource centre 23 education 24 music centre 25 creative arts & Long gallery 27 movement laboratory 28 gymnasium 29 centre for research policy 30 Aboriginal education centre 31 buildings and grounds 32 creative arts, printery, central store 35 biological sciences, informatics 36 administration 37 Kooloobong 38 health sciences 39 lllawarra Technology Corporation (ITC) 39 Wollongong International College I 40 commerce, Hope Theatre 41 the new science building 42 the science annexe 43 computer lab 44 computer science annexe 2 45-47 classrooms 48 computer lab 49 temporary building 50 Islamic Education Centre 51-54 classrooms 63 visitors lodge 67 McKinnon Building