oars ie ten Me ch eK Lee : eee yen ace HT Ais : a‘ chi! tte. fot iz Rie Sa tinea fe rel ues eg CM PR LE AN aie he ey oe vaya teary Bi Gol et Rete MeL erm et er a ree? eh Wiehe fe Gr ay ante RG SG Ty tam Ble Boke WON Bae NS Gee My Mth tet teat Dae Big te Ry eee ee wale Pree NS yee oe ee eee 4 Higa ee ee et Vi neal: tate ae Via we ee en e's ST TT te ee Kore ase ths Fa! ser s i PROCEEDINGS OF THE LINNEAN SOCIETY NEW SOUTH WALES VOLUME 106 (Nos 465-468; for 1981-82) Sydney The Linnean Society of New South Wales 1983 © Linnean Society of New South Wales Contents of Proceedings Volume 106 NUMBER 1 (No. 465) (Issued 27th October, 1982) JONSELL, B. Linnaeus and his two circumnavigating apostles .............. FOSTER, B. A. Two new intertidal balanoid barnacles from eastern Australia... MCALPINE, D. K., and KENT, D. S. Systematics of Tapezgaster (Diptera: Heleomyzidae) with notes on biology and larval morphology ........... SCHNEIDER, M. A. A comparative morphological study of the reproductive systems of some species of Tapeigaster Macquart (Diptera: Heleomyzidae) . . SHAW, D. E., CANTRELL, B. K., and HOUSTON, K. J. Neurochaeta inversa McAlpine (Diptera: Neurochaetidae) and seed set in Alocasta macrorrhiza (L.) G. Don (Araceae) in southeast Queensland..................... HUTCHINGS, P. A., and RECHER, H. F. The fauna of Australian mangroves. . . Notes and Discussion FORD, R. J. A. W. H. Humphrey, His Majesty’s mineralogist in New South Wales (803-12 —— Avcomment. 24. 25.2 vie eae ee oe ee se wits NUMBER 2 (No. 466) (Issued 27th October, 1982) POWELL, C.McA., NEEF, G., CRANE, D., JELL, P. A., and PERCIVAL, I. G. Significance of late Cambrian (Idamean) fossils in the Cupala Creek Formation, northwestern New South Wales...................20056- ANDERSON, D. T. Origins and relationships among the animal phyla ........ JENKINS, C. J. Late Pridolian graptolites from the Elmside Formation near Yass, Ne WAS GUILE WialeSeum tian ante psec Har eee aos Gi ore elect tous ay wake Gees JENKINS, C. J. Darriwilian (middle Ordovician) graptolites from the Monaro Trough sequence east of Braidwood, New South Wales ............... HUTCHINGS, P. A. The fauna of Australian seagrass beds................. Annexure to Numbers 1 & 2. The Linnean Society of New South Wales. Record of the Annual General Meeting 1981. Reports and balance sheets ............. 123 NUMBER 3 (No. 467) (Issued 10th May, 1983) LARKUM, A. W. D., and WEST, R. J. Stability, depletion and restoration of seagrass! beds oii. ned as Mates ds) ie age ins ed) aie Wc sane a ro WEST, R. J., and LARKUM, A. W. D. Seagrass primary production — a review. Kuo, J. Notes on the biology of Australian seagrasses ................-... GRAY, M. R. The taxonomy of the semi-communal spiders commonly referred to the species Ixeuticus candidus (L. Koch) with notes on the genera Phryganoporus, Ixeuticus and Badumna (Araneae, Amaurobioidea) ......... NUMBER 4 (No. 468) (Issued 10th May, 1983) WEBB, L. J. Sir William Macleay Memorial Lecture 1982. Ecological values of the tropical naimtonestasesOUnCe nen .s uate. cues aeons ira tet eer GRAY, M. R. A new genus of spiders of the subfamily Metaltellinae (Araneae, Amaurobioidea) from southeastern Australia....................... CARR, P. F. A reappraisal of the stratigraphy of the upper Shoalhaven Group and lower Illawarra Coal Measures, southern Sydney Basin, New South Whales ak ak GSP. cee ot ie ti attra) atid at ANcs oe PSR ON eR OO GE ne Ree ae HOLMES, W. B. K., HOLMES, F. M., and MARTIN, H. A. Fossil Eucalyptus remains from the middle Miocene Chalk Mountain Formation, Warrumbungle Mountains, New South Wales...................... WATSON, I. A., and SOUSA, C. N. A. de. Long distance transport of spores of Puccinia graminis tritici in the southern hemisphere................-...- SHAW, D. E., and CANTRELL, B. K. A study of pollination of Alocasia macrorrhiza (L.) G. Don (Araceae) in southeast Queensland..................... SELKIRK, P. M., COSTIN, A. B., SEPPELT, R. D., and SCOTT, J. J. Rabbits, vegetation and erosion on Macquarie Island........................ TINE gilt 21-505, Sto Oe Ro eet Scag oop toate ie Be 263 ZU 287 299 311 323 - = Annexure to Proceedings Vol. 106 THE LINNEAN SOCIETY OF NEW SOUTH WALES RECORD OF THE ANNUAL GENERAL MEETING, 1981 The one hundred and sixth Annual General Meeting was held in the Activities Room, Australian Museum, 6-8 College Street, Sydney on Wednesday, 25th March 198] at 7.30 p.m. The President, Dr F. W. E. Rowe, occupied the Chair. He read the minutes of the one hundred and fifth Annual General Meeting and they were adopted by the meeting. REPORT ON THE AFFAIRS OF THE SOCIETY FOR THE YEAR 1980-81 Publications The Society’s Proceedings were published on the following dates : Vol. 104, Parts 1 & 2, July 1980 Vol. 104, Numbers 3 & 4, January 1981 We record our gratitude to Professor T. G. Vallance for his continuing service as Honorary Editor. Membership During the year, 9 new members were elected to full membership of the Society and 18 joined as associate members. There were 4 resignations and 3 deaths; 5 names of unfinancial members were removed from our membership list and 2 ordinary members became life members after 40 years membership of the Society, making a total of 300 on the 20th March, 1981. Council noted with regret the death of life member, Dr I. M. Mackerras who joined the Society in 1922 and was Linnean Macleay Fellow from 1925 until 1927. We also note with regret the death of Mr E. T. Smith who had been a member since 1953 and of Mr S. L. W. Allman, also a life member, who joined in 1940. DrJ. M. Beattie and Mr K. L. Taylor became life members this year and we record our appreciation of their long service. The Council is pleased to note the growth in the new category of associate membership. Meetings Activities during the year included 4 ordinary general meetings, 1 field excursion and 1 full day symposium. The first meeting was held on 11th June_at the School of Botany, University of New South Wales. Dr J. Bowler of the Research School of Pacific Studies, A.N.U. addressed us on the topic ‘Reconstructing climates of the past 50,000 years: evidence from salt lakes in Northern and South Australia’. On 9th July, in the Australian Museum, Dr D. A. Adamson of Macquarie University showed slides and discussed his research in the Antarctic during the summer of 1979-80. On 24th September, in the Australian Museum, Dr P. Selkirk of Macquarie University spoke to us about ‘Summer on Macquarie Island through a Biologist’s Eye’. The biennial Macleay Memorial Lecture was delivered on 10th September by Professor H. B. Whittington of the Sedgwick Museum in Cambridge, at the Science Centre Auditorium. His subject was ‘Evolutionary problems presented by the appearance of multi-celled animals in the Cambrian’. On Sunday 28th September, members were escorted on a tour of the Royal Botanic Gardens by the Director, Dr Johnson and staff members, Mr Rodd and Mr Dangerfield. A review symposium on ‘The Sydney Basin’ attracted an audience of close to one hundred people to the Australian Museum on Saturday 25th October to hear a panel of speakers deal with the area from various scientific and environmental aspects. Our last meeting for the year was held on 19th November at the Australian Museum when Dr D. A. Adamson addressed us on ‘The biology and geomorphology of a Blue Mountain valley, the Grand Canyon, Blackheath’. Newsletter The LINN Soc News, edited by Dr Helene Martin, continues to be printed quarterly. It consists of details of our programme, reports of resolutions of the Council and other items of interest to members. Library The Council has for some time been concerned about the Society's problems in meeting the increasing cost of maintaining the library. Detailed discussions have been held with officers of the State Library, the National Herbarium, the Australian Museum and the three universities in Sydney. These discussions established that none of these institutions has the capacity or the finance te house and service the Society’s library as a whole, but all are interested in receiving selected parts of it. After extensive discussion of available options, Council decided in December to inform the State Government, the Australian Academy of Science and the interested institutions that unless the costs of running the library could be entirely and permanently undertaken by an outside body within six months, the Council proposed to dispose of the collection in the following manner: 1. As public institutions, the Australian Museum and the National Herbarium should be given first refusal of parts of the library in which they are interested. 2. Consideration should be given to donating to the Macleay Museum any Macleayana, including bound pamphlets. 3. Remaining parts should be offered to the universities of Sydney, New South Wales and Macquarie. 4. The remainder to be dispersed by a decision of Council. Linnean Macleay Fellowship Barbara Porter has completed her second and final year as Linnean Macleay Fellow at the University of New South Wales. Her studies on the structure and function of macropodid salivary glands have disclosed the remarkable secretory capacity of these glands, and the importance of salivary secretions in foregut homeostasis and in electrotype balance in the kangaroos. In many respects, the histology of the glands and the volume and composition of their secretions, resembles ruminant counterparts; the study therefore provides further evidence for the evolutionary convergence of macropods and ruminants with respect to digestive physiology. Miss Porter submitted her Ph. D. thesis in March 1981. Linnean Macleay Lectureship in Microbiology The work this year was concerned mainly with the physiology and cultivation of a new type of oyster mushroom, Pleuwrotus sajor-caju. The mushroom has many advantages over Agaricus bisporus, the only mushroom presently grown in Australia. Apart from the high yield, the oyster mushroom can utilize a wide range of agricultural wastes and grow at a wide range of temperatures. The work is supported by a grant awarded by the Rural Credit Development. Dr K. Y. Cho continues to be the Linnean Macleay Lecturer. Sczence Centre Report At the end of 1980, Mr E. J. Selby resigned as one of the Society’s representatives on the board of Science House Pty Ltd. Being experienced in business affairs, Mr Selby has given invaluable service and the Society is indebted to him. His place has been filled by Mr Harry Wallace, a director of F. T. Wimble with a background of successful business experience. The Society is currently represented on the board of Science House by Mr Harry Wallace, Dr Lyn Moffat, Dr Alex Ritchie and Dr Helene Martin. The Science Centre is now fully let so income from rents has reached a plateau. Hirings of the rooms are about 42% of capacity. The secretarial service operates well and is near the limit for the present staff. The activities in conference organizing have shown a spectacular increase — conferences are organized for the Science Centre and elsewhere, e.g. Wollongong, Broken Hill and overseas. As a trading operation, all of these avenues work well and the income over expenditure is quite satisfactory. However, it in no way services the mortgage of $2.6 million and there is no hope of Science House Pty Ltd ever trading itself out of this debt. A foundation is being created for tax-deductible fund-raising purposes with Mr Harry Wallace as chairman of the fund-raising committee. FINANCIAL REPORT Report of the Honorary Treasurer (Dr A. Ritchie), presented to the Annual General Meeting with the audited balance sheets. Despite increased costs for goods and services, the Society closes the year 1980 with a small surplus. The amount ($2678) now transferred to accumulated funds is less than the surplus for 1979 ($3648) but compares favourably with the $2224 in 1978. Invested funds in the General Account stand at $91434 (down $1782 on 1979, but $2000 redeemed from the MSW&DB awaits reinvestment) and yielded interest of $10952 in 1980, $634 more than in 1979. The remainder of our investments (totalling almost $420,000) is tied in loans to Science House Pty Ltd - which are unlikely to yield a return in the foreseeable future. Total expenditure for the year 1980 was $30,497, substantially more than in 1979 ($26,392). Among items contributing to the increase were: higher audit fees, a grant to the Linnean Macleay Fellow for assistance with research equipment, higher postal costs and greater outlay on Library (a number of subscriptions had been held over or delayed from 1979). Again, following a necessary review by Council, the emoluments of the Secretary and Librarian were increased in acknowledgement of present inflation. The cost of printing the Proceedzngs has also risen but members surely will appreciate the improved quality of the journal and be aware that all editorial work is done by Prof. Vallance at no cost to the Society. Economy in the use of the secretarial services of Science House Pty Ltd resulted in our paying only $7903 in 1980, as against $8097 in the previous year. The total income of $33,175 for 1980 derived chiefly from subscriptions, interest on General Account investments, sales of back issues of the Proceedings and, importantly, from the transferred surplus from the Fellowships Account (see below). In the Fellowships Account, total investments of $128,196 yielded interest of $13,414 in 1980 ($11,620 in 1979). The maximum amount ($3200) permitted by the Equity Court was paid to our Fellow during 1980 and the surplus ($10214) transferred to the General Account. The Bacteriology Account, with investments totalling $36,921 earned interest of $3639 in 1980. The sum of $4000 paid during the year to the University of Sydney as contribution to the salary of the Linnean Macleay Lecturer in Microbiology left a deficit of $361 but this is covered by a previous surplus. Total investments in the Scientific Research Fund (to be known henceforth in honour of its principal benefactor, the Joyce Vickery Scientific Research Fund) by the end of 1980 stood at $49,890, a considerable increase on the 1979 figure and due largely to further releases from the late Dr Vickery’s estate. Interest received during 1980 amounted to $4216; increased income in future is anticipated as the investments start to yield over a full year. Accordingly, Council has resolved that from now on a certain proportion of the annual income will be made available as grants for appropriate scientific research. It plans soon to announce conditions of eligibility for the grants and to invite applications for them during 1981 by a date to be fixed, and regularly thereafter. Council, however, will reserve the right and capacity to receive and grant applications for emergency assistance at any time, One such grant (of $300) from the Joyce Vickery Scientific Research Fund has already been made — to enable Dr S. K. Roy of India to complete a research project at the Royal Botanic Gardens in Sydney. The Council hopes others will emulate Dr Vickery’s generosity by donating to the Fund; by so doing they will enhance the capacity of the Linnean Society of New South Wales to foster the cause of science. Donations to the Fund are tax-deductible. Following discussion of the Honorary Treasurer’s report, a motion that the audited balance sheets for 1980 and the financial report be adopted was carried unanimously by the members present. PRESIDENTIAL ADDRESS Dr Rowe delivered the Presidential Address entitled ‘A Brief Review of Australian Echinoderms since H. L. Clark (1946)’. This address will be published in full in the Proceedings. DECLARATION OF ELECTIONS As there were five nominations to fill six vacancies on Council and no other nominations for President or Councillors were received, no election was necessary. The President declared the following members elected: Dr L. A. S. Johnson, Dr Lynette A. Moffat, Mr G. R. Phipps, Dr F. W. E. Rowe, Professor T. G. Vallance. Our Auditors continue to be W. Sinclair & Co. Dr Rowe introduced Dr Helene A. Martin as the President for 1981-82 and invited her to take the Chair. ill 8L'PS6 965% SI 8hE S&P Ol PSF 16 6h SS1 3 00'T $3960 I 96 961 I OF 046 OL 91 €86 0F 000% 09 7006 LG°686 9LF 00°0S2‘21 00°000°L 00°008°1 00°000 02 00°000 1 00°00€'6 00°00L 9¢ 91 ¥88% 01 966 €€ 120% G9 PPS 16049 I $ doooumop DD aoOD sjassy JUSIIND [B10] coODtOOD HONG nOoOHoO ROD OCU G0 yueg 3@ ysep 6-40 O°0 quno.Dy sdiysmo][oq Agape ueouUulT siojqaq Arpuns GO00000500000000 qunossy juaWIaseuR, DO OR Sea a ae al JUNODIY 913UID V29U9IINS —paiuiry 41g asnopy 2duaIdg — suvoT —sjassy JUarIND prog aBeuleig 2 eeIaMag ‘19jeM uejodonay See: Siren nab Re aaonnh Boe meee Ric-o pantry epursy Pieced eae tradelemnees eS Tre pauwry ‘or @°9 3soD 1W — seinjueqeq pc vedeecmeas a json IW — sueOT YI[eamuoUIUIOZ ++ yurg jusuidojaaaq saoimosay ueleusny »--diop juauidojaaag sadmosay uelpensny ae Aja190¢ Surp[ing yusueUTIeg “M'S'N —sjtsodaq — SJUdUI}SIAUT GOOUGCoOOD DODD OOO sjassy pexty [239.L aooo0e peiury 4g asnopy adua19g — aieys — s0D 1VY — JoUsaAUT palace Boron a JjO U9}311M UONeIDaI1d aq ssaT brats oaromtont oie eb cs 1s0D IV — sauTydeW 2170 Sa Tog “°° + yo u9}1IM uoTeDaI1daq ssaT “* 5) 4soni1y — ssuiy 2 soingxiy ‘oinjruing —sjassy poxly LES boat 969 86r LE8°9 602 9g¢°T S006 686 9LP 916 86 OSL SI 0006 008 'T 00002 000'T 0066 00498 999% $ 6L61 81° FE6 965$ 00°06 99° ZrI I GL Z9L G69 6L°LL9 €E 80 63S $ play suoneuog —peiuwiry 4g asnozy 29ueI9S O06 KDOSD0CGOFF OOD HOAOGKFODbobODO0oGU OG SuIpulqyoog — dA1asayy Panton aiosctoseenecanea Boo 0861 ‘Jaquisszaq asT¢ ‘aouRleg GONKD00DD OD GDOODDnAooO0R reak oy IO} snjding PPK O86I ‘Arenuef ys] ‘oouerleg —spuny paiejnumsoy LES vO5$ OL SPI I ¥80 $6 6L61 O86I ‘t2qQuI900q ISTE 3¥ SB IB9YS sdUL[ eg LNNODOV IVYANAD SATVM HLOOS MAN FO ALAIOOS NVANNIT iv ‘T86T Yo1eIN ‘Jomnseaty, “UO “AIHOLIY V “1861 ‘yore Jo hep parya Ayuam y siya Aaupdg 1e Ca.LVG ‘popuoure se ‘Gh6 (10W UOneNsIdoY sJURJUNODDY dI[qng ey) JopuN paiajsay sjUeJUNOIDY poioeYyy “OD 8 YIVIONIS “M *K32190G ay) Jo syoog ay) Aq pasojasip se pue sn 0) uaAl8 suoT}eULdxa ay) 0} BuIPIOdIe ORG ‘19qQUIEd9q ISTE Ie SB sIIeFJe [e1INueUTy aYy3 JO UONISOd ay3 YI10} sjos A[ID91109 jUNOD9¥ sINjIpuedxa wy suIODUT BuUIAULdUIODDe pu jaIoYS BdURT[eq aAOge ay) ey) UOTUIdO ZY) Jo a1e pUR YG ‘1aquis00q IS[G Papua syIUOUL 9A[IaM} JY} IOJ SaTeM YINOS Man] Jo AyaI90G uesuUUTT ay) JO sp10d91 puke syxOOd 9Y) paiIpne sary 9M LYOdaY SYOLIGNV €3 GLI'Ss$ 098 '62$ 6S GLI‘Se$ 098 '62$ GLUES Ce > OP spuny payefnumMsoy 0} parsioysuel y, 89'S : 1eaX 107 snjding PE LOPIOG)) 2 0 eae es he ainjtpusdxg [210], 66 9% OE ene oc ae ee sasuadxq Surpoaery = Ll FSI qGonocdamovD oOo mOOnoU noo OoON OOo On suoydajay, ILI 9L°€06°L S9DTAIIG [BIIBIVINI9G 1.608 08986 ttre judy asnopy 2ouaI9g 1.86 669609 ttt terse eee sauue[es Oge'G OOO0L ttre au] WOOY 0g Pre Wn el ee Stosseritiensoto.b nadctoratpetcr DHORO-D-cC Onno 160: 0s0,u sjurday 042 €%Z GLI'ES Foo eFO OD OMOOOCO Ooo OMHnOOn OOOO OM OC oO DS auoouy [BIOL 098 ‘62 OF'LS JURUDIUIe] pue sireday Pall 06° LOFT atrafty Meike) sitvlialiesetalsaleltesallalloieMalis@ieSlsiiclivialvallciioisliuliclisiiclelisiinliviteli=¥i=ltalisllsaisalelelis ica) sates 6621 09°Z86 onoooOogHUoDOoONOODOODO Arau0neI1¢ pue Ssunutig 8h9 Gy (‘“‘<“‘éta ODOT OOD DODSODSEGOOGSGOGDODODGDuODE OD sayes yuriday GIZ‘1 MAI || oo abe covaaos $1809 Sunutig pa sSurpaas01g SPL‘ L 00°09 fob oCoOnOCOOODDO Ob ODD OOK OOO OOOO Do Ono e oO Peara09y suoneuog L 16'SI8‘I OdDoOOKOGOO0DHDKGoGoODoO Ooo b Oooo oOo sa3v\sog LEG‘ GORCIIZ0 Eee ee ee ee 0861 ‘1equiasaq 1s, ¢ APUG

NEW SOUTH WALES VOLUME 106 NUMBERS 1 &2 NATURAL HISTORY IN ALL ITS BRANCHES THE LINNEAN SOCIETY OF NEW SOUTH WALES Founded 1874. Incorporated 1884. The Society exists to promote ‘the Cultivation and Study of the Science of Natural History in all its Branches’. It holds meetings and field excursions, offers annually a Linnean Macleay Fellowship for research, contributes to the stipend of the Linnean Macleay Lecturer in Micro- biology at the University of Sydney, and publishes the Proceedings. Meetings include that for the Sir William Macleay Memorial Lecture, delivered biennially by a person eminent in some branch of Natural Science. The Society’s extensive library is housed at the Science Centre in Sydney. Membership enquiries should be addressed in the first instance to the Secretary. Candidates for election to the Society must be recommended by two members. The present annual subscription is $20.00. The current rate of subscription to the Proceedings for non-members is set at $35.00 per volume. Back issues of all but a few volumes and parts of the Proceedings are available for purchase. A price list will be supplied on application to the Secretary. OFFICERS AND COUNCIL 1981-82 President: HELENE A. MARTIN Vice-Presidents: LYNETTE A. MOFFAT", A. RITCHIE, F. W. E. ROWE, J. T. WATERHOUSE Honorary Treasurer: A. RITCHIE Secretary: BARBARA STODDARD . Council: D. A. ADAMSON, M. ARCHER, L. W. C. FILEWOOD, L. A. S. JOHNSON, HELENE A. MARTIN, P. M. MARTIN, LYNETTE A. MOFFAT", P. MYERSCOUGH, G. PHIPPS, A. RITCHIE, A. N. RODD, F. W. E. ROWE, C. N. SMITHERS, T. G. VALLANCE, J. T. WATERHOUSE, B. D. WEBBY, A.J. T. WRIGHT Honorary Editor: T. G. VALLANCE — Department of Geology & Geophysics, University of Sydney, Australia, 2006. Librarian: PAULINE G. MILLS Linnean Macleay Fellow: Linnean Macleay Lecturer in Microbiology: K.-Y. CHO Auditors: W. SINCLAIR & Co. eh The office of the Society is in the Science Centre, 35-43 Clarence Street, Sydney, N.S.W., Australia, 2000. Telephone (02) 290 1612. ©Linnean Society of New South Wales . ' resigned 20 May 1981 Cover motif: Forewing of Permorapisma bisertalis (Neuroptera: Permithonidae), from the Permian of Belmont, f New South Wales 4 Adapted by Len Hay from Proc. Linn. Soc. N.S.W. 51, 1926, p. 278 (Fig. 15) PROCEEDINGS of the LINNEAN SOCIETY NEW SOUTH WALES VOLUME 106 NUMBER 1 } ’ ea bite Linnaeus and his two Circumnavigating Apostles BENGT JONSELL Uppsala Universitet Institutionen for Systematisk Botantk Box 541 75121 UPPSALA Sweden A Lecture delivered under the joint auspices of The Linnean Society of London, The Linnean Society of New South Wales and the Svenska Linnésallskapet in the Macleay Lecture Theatre, University of Sydney, on 26 August 1981, during the 13th International Botanical Congress If you visit Linnaeus’s house in Uppsala and ascend to the floor where his study remains with inost of the furniture intact, you will see on a high glass cupboard a globe of the world made in Akerman’s famous workshop in Uppsala in the middle of the 18th century (Fig. 1). The globe shows with surprising accuracy the continents in broad outline but we can imagine that Linnaeus looked with curiosity and fascination at the arbitrary, dotted contour that runs along the eastern side of Nova Hollandia up to Nova Guinea (Fig. 2), across the strait said to have been kept secret by the Spaniards. Long before Linnaeus knew of the plans for Captain Cook’s first voyage to the Pacific he may well have wondered about this area on his globe and what such a remote region might yield in the way of living forms. As it turned out, two of his pupils — Daniel Solander and Anders Sparrman — were to sail with Cook and by so doing helped unravel some of nature’s secrets on the other side of the world. At the time of Cook’s voyages, as we approach Linnaeus to consider his relations with and expectations of these widest-ranging of his pupils the great teacher is in his fifties, already prematurely aged (Fig. 3). His great botanical contributions — those beginning with the explosion of fundamental works during the years 1735-38 in Holland and culminating in the Speczes Plantarum of-1753 — are finished. But although his principal new discoveries had been published Linnaeus still maintained industrious botanical work. He received innumerable additions to his garden and herbarium and wrote new and essentially augmented editions of Systema Naturae, Genera Plantarum and Species Plantarum, undertakings that later in his life were taken over by compilers abroad. Linnaeus’s garden in Uppsala (Fig. 4) continued to flourish in the early 1760s, twenty years after its energetic restoration following his appointment in 1741 as one of the two professors of medicine at Uppsala University. After its first glory in the days of Olof Rudbeck the elder, the grand old man of Swedish 17th century science, the garden had long been neglected when Linnaeus took it in hand. Soon after 1760, however, Linnaeus turned his attention increasingly to his new estate Hammarby in the countryside southeast of Uppsala. The collections were housed there in a special small museum building and gardening was more successful than in the flat, moist plots in the town. His family by this time was fairly grown up. The four daughters were still unmarried and the only surviving son Carl was now in charge as demonstrator in the Uppsala garden, an arrangement that had drawn much criticism. Linnaeus himself did not leave Uppsala any more, except for meetings of the Proc. Linn. Soc. N.S.W., 106 (1), (1981) 1982 2 LINNAEUS AND HIS CIRCUMNAVIGATING APOSTLES Fig. 1. The globe in Linnaeus’s study (from a colour photo by O. Lindman). Fig. 2. Detail of the globe (Fig. 1) showing Nova Hollandia and surroundings (from a colour photo by O. Lindman). Academy of Science in Stockholm and visits to the Royal family and castles near the capital. That, very briefly, is the pattern of those years. By their journeys his pupils, the apostles, have begun to cover the world and gather its ‘spoils’ (Fig. 5). Linnaeus’s knowledge of the world and of its botany in particular increased substantially as a result of the records and specimens, dried and Proc. Linn. Soc. N.S.W., 106 (1), (1981) 1982 B. JONSELL 3 © Fig. 3. Linnaeus at the age of 40. From a pencil drawing by J. E. Rehn, probably made 1747. HoR TI! UPSALIENSIS PROSPECTUS | AC HRES “ 4 iis pee - pede so ian i 1h : ; ph 4 PS = é = : oa £ ‘ ae = Gk ee a eas . ff QQUrc ere 2 € Ay Fig. 4. The Botanical Garden in Uppsala in the 1740s. From the dissertation Hortus upsaliensis (1745). Proc. Linn. Soc. N.S.W. 106 (1), (1981) 1982 4 LINNAEUS AND HIS CIRCUMNAVIGATING APOSTLES ¥ GRONLAND CS? Fikes BP iO RR TI ae ts. 4 e A 2 | ey _Norra polcirkeln Kin NEA Aw Shes enEh oN ae ‘ @ ‘ 4 = 5) , Xx Tow ; = pe EUROPA > 2S o> BOR KASPISKA > ; b 4 KS, THlasselqul y i 3 Teanten/y ages wd) Si} Hasselquist, 4. ss : ada dpformosa — Keifrans vandkrets Ke einusA ree Aes x Y ‘ & 8g Hawatr-varna Se) @ yy be) Sah eA THe AL Ve Ee Tr: Ales e ____-Ekyators qa} 4 zy Rea Caen a & seein a eae poe er Sa Neue o A Marquesss-Garna MERIKA Timor Nya Hebriderna p, 5. Tahiti ‘idji-Ourna = Ka Vrq ——, — "es “AUSTRALIEN \ pales Tonga-Sarna S “ , Pa ye c= f Sy AR f= sande\ (NW Se ALT RA T EN] Comore INDISKAS OCEAN EN = bas roe * man ‘ Kerguelen / Spat OS Lo Q-meridianen ©; Nya j I serge Shs Zeelent fb \c <> 4 =m, oman 2 (ab c wo Spuman ~ : X, i SOD) Re A 1S aH LAW V, ET Qo i Fig. 5. The voyages of Linnaeus’s apostles. Map compiled by R. E. Fries (Fries, 1951). living, brought back by the travellers. Yet along with the new-found facts an enthusiastic credulity persisted. Linnaeus could still accept that the tropical and southern regions of the world were populated by strange ‘human cousins’. There was, for instance, the Homo caudatus — the tail man — from the Nicobar islands in the Gulf of Bengal, and the Homo troglodytes the white nocturnal species, blind in daylight, wild and thievish, reported from ‘Ethiopia’ (i.e. tropical Africa), Java and other places (Fig. 6) ; both species appear in Systema Naturae of 1758. Linnaeus’s view of the southern lands was based largely on knowledge of southernmost Africa, the rest south of the tropic of Capricorn being virtually unknown to him. In the thesis Flora Capenszs of 1759 details on the flora are mixed with remarks on a strange assemblage of beings between ape and man. ‘Semper aliquid novi ex Africa was a much used proverb among the old Romans, and this is still true in our days’ he says in the introduction to the thesis and goes on to list troglodytes, sathyres, sylvanes, sphinxes, dianae, hamadryades and cynomolges as well as the hottentot — the most awkward of humans and the most remote of all to be included in the species Homo sapiens. ‘Africa monstrifera’ he calls this place in Plantae Rarzores Africanae (Linnaeus, 1760), the end of the world where all miracles of nature seem to have been concentrated. One might even discern here his disappointment at not having seen this world. Sparrman tells us there was nothing Linnaeus regretted more in his life than declining the voyage to South Africa offered him during his years in Holland. What I have said makes it easy to understand Linnaeus’s great expectations when he heard from London of the plans for a scientific voyage to the South Seas, to those very areas still empty on the globe in his study. The great news arrived in a letter of 19 August 1768 from John Ellis, the London merchant and outstanding student of corallines and other fields of natural history and Linnaeus’s most faithful and intimate correspondent in Britain. ‘I must inform you’, Ellis begins, ‘that Joseph Banks Esquire, Proc. Linn. Soc. N.S.W., 106 (1), (1981) 1982 B. JONSELL 5 AN THROPOMORPH Fig. 6. The ‘human cousins’, as illustrated in the dissertation Anthropomorpha (1760). a gentleman of £6000 per annum estate, has prevailed on your pupil Dr Solander, to accompany him in the ship that carries the English astronomers to the newly discovered country in the South sea, Lat. about 20° South, and Long. between 130° and 150° West from London. .. . They are to proceed . . . on further discoveries of the great Southern continent. . . . No people ever went to sea better fitted out for the purpose of Natural History, nor more elegantly. . . . In short, Solander assured me this expedition would cost Mr Banks 10 000 pounds’. And as an act of devotion to Linnaeus Ellis added: ‘All this thing is owing to you and your writings’. Ellis had taken leave of Banks and Solander three days before writing the letter. Linnaeus, always sensible, became enthusiastic on receiving the news. The participation of Solander, one of his most intimate pupils, seemed a guarantee that Linnaeus would have a share of the collections. Solander wrote from Rio de Janeiro, where the expedition had landed three months after leaving Britain, explaining that if ‘Mr Banks will be in the same spirit as now to complete studies in Natural History, we will together make the voyage to Sweden to ask prof. Linnaeus to order our recruits’. The last was an expression borrowed from Linnaeus himself. This makes it appropriate to consider how relations between Linnaeus and Solander had evolved so far. Solander (Figs 7, 8), whose origin lay in the north of Sweden, arrived in Uppsala in 1750. Like many before him, he went with quite other academic intentions but became fascinated by Linnaeus’s teaching in medicine and natural history. Within a few years he was a favourite of Linnaeus. “The wittiest pupil I ever had’, Linnaeus wrote many years later and, on another occasion, ‘I have housed him under my roof, just like a son’. Indeed, Solander in those days was practically one of the family. He accompanied Linnaeus to the Royal castles and collections and joined the family on visits to Dalecarlia, where Linnaeus’s father-in-law lived. He was even considered seriously as husband for Linnaeus’s eldest daughter. Through his close association with Linnaeus Solander’s botanical expertise became impressive. He also gained experience in the field travelling, as the young Linnaeus had done more than twenty years earlier, Proc. Linn. Soc. N.S.W., 106 (1), (1981) 1982 6 LINNAEUS AND HIS CIRCUMNAVIGATING APOSTLES wee Hie C. Hite Fig. 7. Daniel Solander (1733-1782). After the painting by J. Zoffany, now in the possession of the Linnean Society of London. in the mountains of Lapland, probably during two summers. Several of Solander’s collections from those journeys are preserved in Stockholm (Fig. 9). Linnaean pupils began to be sought after in the learned world. The British wanted a man to teach them Linnaean method and so when Linnaeus thought Solander mature enough for a trip abroad to enlarge his education — the sort of thing almost obligatory for learned young Swedes in those days — he proposed for him a year or two in Britain. This was surely a sign of great trust in the 26-year-old Solander, whom Linnaeus now apparently hoped to groom as his own successor. Solander was urged to return after not too long a sojourn. As we know Solander never returned to Sweden. He was immediately and cordially received by Ellis and by Peter Collinson, another merchant and student of natural history, and other Londoners. Solander made friends everywhere — such was his amiable nature — and soon became a well-known figure in society and scientific circles (Fig. 8). “Throw him where you will — he swims’ were Boswell’s words about Solander. His combination of modesty and wit, his skill in conversation made a general impression, as so many London reports of those days acknowledge. Linnaeus long persisted with his own plans for Solander, counting on his return. When he transmitted to Solander an offer of the chair in natural history at the Academy of Science in St Petersburg Linnaeus may well have regarded some years’ Proc. Linn. Soc. N.S.W., 106 (1), (1981) 1982 B. JONSELL 7 Zhe SIMPLING MACARONI. Lhe Slant 4 ‘cove oom 2 froxen LoVe. J Bea : On thellou- Parkes grows Lf SO ia generat 4 Pal accor lo BL by Marly Strered aby (35 772 Fug. 8. Caricature of Solander, published in London after the return of the Endeavour. After the etching dated 1772. service in the Russian capital as a way of securing Solander’s succession to the chair at Uppsala. But Solander’s English friends were appalled at the prospect. Collinson warned that tumults and riots, perhaps even a revolution, might occur in Russia, expressing also more rational arguments about the scientific and commercial isolation of a place where Solander would be buried in obscurity. ‘No doubt you . . . know persons less eminent but every way qualified in botanic science to teach Russian bears’ Collinson protested to Linnaeus in November 1762. The English had no wish to lose the young man who had won their favour. Eventually, but only after repeated pleas from Linnaeus, Solander replied that he would have accepted the offer with uplifted hands, but on certain conditions. Thereafter his direct correspondence with Linnaeus seems to have ceased until the messages from the Endeavour. In 1763 Solander obtained a post at the British Museum. The Swedish scholar Arvid Uggla has concluded that a breach occurred between master and pupil, and for two reasons: first, following a particular grace of the Swedish parliament Linnaeus had his own son nominated as his successor, and second, his eldest daughter had married an officer, Captain Bergencrantz — soon to be a most unhappy marriage. Solander could no longer feel like a son in Linnaeus’s household and was more than content to remain in England. Solander probably made his first acquaintance with finds from the South Seas in Proc. Linn. Soc. N.S.W., 106 (1), (1981) 1982 LINNAEUS AND HIS CIRCUMNAVIGATING APOSTLES ete IV Wit Ait Rees ey nes Gentianella aurea (1) J. M. Gullerr JM. Giller Ge #} Le cl? 2. of f ee ee, ae NOBEL. ee ay hee at ye LOR Ca. Foto parte Fig. 9. Gentiana aurea L., collected by Solander at the coast of northern Norway in 1753. Specimen preserved at the Museum of Natural History, Stockholm. 1765, in a collection of ‘artifacts’ given to the British Museum by Commodore Byron after his circumnavigation on the Dolphin. About a year earlier Solander had met Joseph Banks and they had already planned.to make a study trip to Linnaeus in Uppsala when Banks’s decision to join the Endeavour expedition altered everything. Banks made Solander enthusiastic for the voyage with a result that the latter soon Proc. Linn. Soc. N.S.W., 106 (1), (1981) 1982 B. JONSELL 9 asked: “Would you like a fellow-traveller?’. ‘Someone like you would give me untold treasures and rewards’ was Banks’s reply, and so the business was settled. I shall not here describe the course of this voyage, many times told and doubtless well-known in this part of the world. Let us only recall the days 28 April-5 May 1770 when, as William Stearn has said, rarely indeed can so many new and remarkable plants have been collected in so short a time — enthusiastic collecting that made Cook find Botany Bay the most appropriate name for the place. May I also remind you of Solander’s curious comparison of the termite mounds of the Endeavour River area with the rune stones of the Uppsala plain, perhaps an understandably nostalgic reaction for a person who for ten years had not seen his native country and now found himself at the antipodes. Let us now turn back to Linnaeus in Uppsala and his eagerness for news from the expedition. Immediately after the return of the Endeavour the faithful Ellis informed Linnaeus that it came laden with the greatest treasures of natural history ever brought into any country at one time by two people. Linnaeus replied within the hour that he had never received a more welcome letter, adding: ‘If I were not bound here by 64 years of age, and a worn out body, I would this very day set out for London to see this great hero in botany’. He drew a characteristically bold parallel: ‘Moses was not permitted into Palestine, but only to view it from a distance: so I conceive an idea in my mind of the acquisitions and treasures of those who have visited every part of the globe’. But Ellis also wrote in the same letter to Linnaeus, four days after Endeavour’s return: ‘as to their Natural History I fear I shall not live to see it. They have sufficient for one thousand folio plates’. Unfortunately these were prophetic words. Besides Ellis, Linnaeus also obtained information about the expedition and of Solander’s activities from two of his pupils, Anders Berlin and Henric Gahn, then in London, but heard nothing from Solander himself. Berlin, soon to meet his destiny in Guinea, estimated that 1200 new plant species and 100 new genera had been taken home. Gahn, later a well-known physician in Stockholm (Gahnza (Cyperaceae) was named by Forster in his honour), arrived in London just in time to see the collections and pass on to Linnaeus glimpses, for instance, on the bread fruit tree and about rubber that could be used to erase pencil writing. He presumed Linnaeus would get part of the results but already feared for delays in publication. Ten months after his return Banks signed a letter, drafted by Solander, for Linnaeus, very polite with many excuses and explanations why the specimens reserved for him had not been despatched. But already Linnaeus had written in alarm to Ellis after learning from Berlin that a new voyage to the South Seas was being planned with Banks and Solander of the company. The report almost deprived Linnaeus of sleep. He foresaw the collections of the first expedition being put aside untouched, thrust in some corner to become perhaps the prey of insects and risk destruction. His attitude to life, so often dark in those years, is revealed in the words: ‘I shall be only more and more confirmed in my opinion that the Fates are ever adverse to the greatest undertakings of mankind’. He mentions bitterly that Solander in his letter from Rio de Janeiro had promised to visit Uppsala and it is touching to read of his disappointed hopes for such an occasion: ‘If he had brought some specimens with him I could at once have told him what were new; and we might have turned over books together, and he might have been informed or satisfied upon subjects, which after my death will not be so easily explained’. As we know Banks and Solander never joined Cook’s second expedition — they and their equipment could not be accommodated as Banks wished. The ships left Britain with Johann Reinhold and Georg Forster, father and son, as naturalists but at Proc. Linn. Soc. N.S.W., 106 (1), (1981) 1982 10 LINNAEUS AND HIS CIRCUMNAVIGATING APOSTLES Fig. 10. Anders Sparrman (1749-1820). Engraving after a drawing by M. Mollard. Cape Town they were joined by Anders Sparrman (Fig. 10). This pupil of Linnaeus, 16 years younger than Solander but already a far-ranging, observant traveller, is not as generally known as for instance Solander or Thunberg. Sparrman was brave and full of confidence. When only 17-years-old he had travelled to Canton as physician on a Swedish East India ship on which his services, luckily, are said not to have been much needed. His connection with the famous Swedish East India Company captain Carl Gustaf Ekeberg, who owned an estate in Sparrman’s home parish, helped in this and the following voyage. Five years later he got the opportunity to go to South Africa as tutor to a Dutch family at False Bay in the Cape Province. ‘There after seven months he met the Forsters and within a few days he was on board the Resolutzon for a 28- month trip in the South Seas (Fig. 11) Sparrman was a true Linnaean in his curiosity and devotion to his master whose words were law for him and eagerly defended. But he was an unusual Linnaean in his disregard for systematic method and description, at least in published work. His journeys in virgin South Africa, before and after his voyage with Cook, are vividly related with a highly-developed sense for nature and people, in marked contrast to Thunberg’s contemporary, exact and thorough narrative. Sparrman must have been an amiable and sociable person. He expresses delight in plant hunting with a fellow Linnaean (Thunberg) on Table Mountain, a venture about which Thunberg has not Proc. Linn. Soc. N.S.W., 106 (1), (1981) 1982 B. JONSELL 11 XRTA she SODR 6 med de emyaste er * Vi ceetras: Utgitven at ef | htt Fig. 11. Map of the South Pacific, from Sparrman’s Voyage, part 2. a word to tell. Sparrman has only good to say of the Forsters who often came into conflict with other people, not least with Banks and Solander after Cook’s return to Britain. There is a touch of credulity and naivety in Sparrman, revealed, for instance, in his diary by a certain parading of his cleverness. An amusing example comes from the turning of Resolutzon at its southernmost position among the icebergs (Fig. 12). On that occasion Sparrman hurried to the stern to claim he had been further south than anyone else in the world. That was at Lat. 71° 10’, a record not broken until 1823 when James Weddell reached a more southerly point. Sparrman’s diary is of outstanding value, both readable and informative, but the section from the South Seas is generally regarded as somewhat inferior to that relating his years in South Africa. The South Seas part was completed only decades after Sparrman’s return to Sweden when impressions must have faded and Georg Forster’s Voyage round the World had aptly told the story. As a scientist Sparrman was principally a zoologist. His observations in Africa show that but in the South Seas he is first of all an ethnographer, describing with accuracy and ingenuity the inhabitants of one island after another. He collected quivers and war clubs, pillows and diadems, altogether about 60 specimens that were presented to the Academy of Science in Stockholm where they can now be seen at the Ethnographical Museum. He was also a Proc. Linn. Soc. N.S.W., 106 (1), (1981) 1982 12 LINNAEUS AND HIS CIRCUMNAVIGATING APOSTLES Sie Pee annbadl CE eR = Sy jf 2 C Mita Sy: Berg, Avarvid ba apitain ¢ oe co Skgpd J os ef: Matin » ined frrfhe vatten . JSrd Luakted OF ok Ofler om Oe. -dallopps< Uda #28 ; Fig. 12. The icebergs as illustrated by W. Hodges in Sparrman’s Voyage. competent, interested botanist who collected quite a lot, now mostly in Stockholm. We note only one example of interest: a gentian from Dusky Bay (Fig. 13), the first place reached by Sparrman in New Zealand, where the virgin temperate forest covered the mountains from top to sea — gentian from the antipodes of that gathered by Solander on the coast of Norway. The botanical achievements of Cook’s second voyage have generally been attributed to the Forsters, to such an extent that in an authoritative work like Merrill’s The Botany of Cook’s Voyages Sparrman’s contribution passes unnoticed. His role may never be completely assessed but from diary notes it seems clear the three naturalists formed a team. Johann Reinhold Forster, the father, bears witness to Sparrman’s contribution in the preface to his Characteres Generum Plantarum: ‘Sparrman described the plants and my son depicted them. I devoted my whole time to zoological descriptions. But when Sparrman had more carefully examined the plants, my son and I were often summoned for advice and we discussed together. After that my son compiled the described plants in another volume. I revised it all before the descriptions were again transferred to another volume according to the Linnaean system. And while Sparrman with my son were so occupied, I gathered again new plants and other wealths of Nature so that we should not leave any place with empty hands’. It is significant that Sparrman stands in the background, he never achieved a central position like that gained by both Solander and Thunberg. Eventually at home he was for many years in charge of the collections of the Academy of Science, where his own ethnographica seem to have been the most spectacular items; the naturalza are said to have been in more or less disorder. Towards the end of the century Sparrman was forced to leave this post and lived his last twenty years as a physician among the lower classes in Stockholm. The appearance, only two years before his death in 1820, Proc. Linn. Soc. N.S.W., 106 (1), (1981) 1982 B. JONSELL 13 (OOK One OL Fig. 13. Gentiana saxosa Forst., collected by Sparrman at Dusky Bay, southern New Zealand, in 1773. Specimen preserved at the Museum of Natural History, Stockholm. of the final part of Sparrman’s narrative from the South Seas marks the last time a voice from the Linnaean era was heard. With the return to Britain of the Resolution in 1775 a new important load of plants from the South Seas was added to that brought back earlier by Banks and Solander. In London Solander now worked as a ‘registrator of the world’s botany’, in a way like Linnaeus himself had done 30 years before. But still he made no communication with his old teacher who repeatedly complained that ‘Solander, who may fill hundreds of letters with novelties’ kept silent. “The ungrateful Solander does not send one herb or insect of all that he collected in Insulis australibus novis’ Linnaeus wrote in one of his five autobiographies. Solander certainly was hard at work and that ‘sedentary and luxurious life’ which has been ascribed to him was more a misinterpretation of his seemingly-unhurried manner. He prepared in manuscript many island floras — for Madeira, Tierra de Fuego, Tahiti and New Zealand — with numerous descriptions arranged for the printers (Fig. 14). Others, among them the Australian flora, did not reach that stage. A substantial number of engravings were completed from sketches and paintings (Fig. 15). In a letter to the younger Linnaeus in 1778, the father having died in January of that year, Banks told that 550 plates were then engraved but it would not be possible to include them in a work to be published in the course of the year. Having received nothing from Banks and Solander, Linnaeus the elder promptly got in touch with the Forsters as soon as he learned of their participation in the second Proc. Linn. Soc. N.S.W., 106 (1), (1981) 1982 14 LINNAEUS AND HIS CIRCUMNAVIGATING APOSTLES InsuleOcPacifi onect (AYERS Perea, ES Awe FUE LE Cay rer on UTS ole. LY, Laws cathe Bes Se ela ys Poca wla: Pee hi wer ca Cf PFE A epee. cerrak'~ eee Parr caepticcer fathece coma hirer) a ea CEPA S fee Valeamunien ee bane nl ee Pen as Sepxtceins wd p aaa, operatic Led. We 1 Linn Cp. ph NAL2Z A Vy art eke Finn vcs HEXANDRIA pontap hy le Op ee ann Se pL NACD 1. Prrvcr » 462 oe Kala Ce Chesed . Frrnl. 63 7.2 CLheecrre queen gee fob ; NO oe Che - blr eden i beed unt ts 3 Pals. bor gr Lawler e ‘ rr, cmerrre ad - Bubb. aecMnes 2 ‘ Lice 2 ner tgarren ah vebherrmapher dhe vefosler foie prec Leore co a bok. Mia, ovata, comcard, an havin att vd; PA FED Pes van 7 fra allerrra alicces oa wees pla cute : : . : = bv ere Caparica, pela Lan Qrtherce rena por- Fog. 14. A page of the manuscript of Solander’s unpublished Flora of Tahiti, in the British Museum (Natural History), London. Cook voyage. Indeed, Linnaeus was to encourage them to publish their results before Banks and Solander. And, as Linnaeus probably calculated, some collections from the Forsters reached him through the agency of his most intimate friend, the Stockholm physician Abraham Back. These were the only plants from the South Seas to come under Linnaeus’s inspection. As a result the Supplementum Plantarum, the Proc. Linn. Soc. N.S.W., 106 (1), (1981) 1982 B. JONSELL 15 % Fig. 15. Banksia integrifolia L.f., collected at Botany Bay, painted by Sydney Parkinson on the Endeavour and engraved by Miller in the 1770s. From Britten (1905). manuscript of which was partly elaborated by the elder Linnaeus and edited and extended by his son, includes a number of new Pacific plant genera and species. The work provoked considerable tension between the Banks-Solander circle and Linnaeus filtus, exacerbated by disagreement over the botanical application of Banksza, a name the older Linnaeus in his first enthusiasm had proposed for the whole continent, Terra Australis. Solander, who knew he might have been in the younger Linnaeus’s chair in Uppsala, may have had cause for resentment. But his generous nature took over when Linnaeus ftlzus visited London; Solander cordially introduced him into the Banksian circle. During this time, in May 1782, Solander suffered a sudden stroke at breakfast at Banks’s house, a hard blow to Linnaeus as well as to the absent host. He died a few days later and what had long been feared now became all too clear. Without Solander the professional botanist, publication of the botany from the voyage could hardly be imagined, the more so since Banks was now deeply involved with official duties. Solander is unique among Linnaeus’s pupils in his influence on botanical science. Yet his published works are remarkably few. Besides being a gifted systematist, as is evident from his flora manuscripts, he was a keen observer of morphology. One example of the latter skill is his observation of the Eucalyptus bud with its debated, Proc. Linn. Soc. N.S.W., 106 (1), (1981) 1982 16 LINNAEUS AND HIS CIRCUMNAVIGATING APOSTLES o Fig. 16. Sparrmannia africana L.f. From Curtis's Botanzcal Magazine, plate 516 (1801). M1874. WA Al 3 ee S\N Fig. 17. Solandra grandiflora Sw. From Curtis’s Botanical Magazine, plate 1874 (1817). deciduous cap, for which he suggested an interpretation. In particular he was influential in spreading Linnaean thought and method in Britain, amply realizing the hopes of those who had once begged Linnaeus to send them a pupil. To this day only a fraction of Solander’s work has been published, the main parts Proc. Linn. Soc. N.S.W., 106 (1), (1981) 1982 B. JONSELL 17 printed being the more than 700 copper engravings of Australian plants with descriptions edited by Britten at the beginning of this century (Fig. 15) and now followed by the magnificent editions of Captain Cook’s and Banks’s Florilegza. But the work was never shut away. The generous access granted by Banks to Solander’s material after his death enabled many scientists to profit by and even incorporate many of Solander’s results in their own works. Some, like the Forsters, did so without giving fair credit to the source while Solander was still alive, action that led to their being ostracized by the London scientific community. Most, however, paid full credit to the fundamental importance Solander’s work had for their own, as Gaertner in his De Fructibus et Seminibus Plantarum or Robert Brown in Prodromus Florae Novae Hollandiae and other works that marked the break with Linnaean classification. So the diversity and peculiarity of the material collected and described by the influential Linnaean Solander contributed essentially to the adoption of a non-Linnaean system. Solander’s influence can thus be traced at various levels which ought to be further analysed. I will stop here with only a short retrospective epilogue. Linnaeus, Solander and Sparrman were all brought up in provincial Swedish rectories — Linnaeus in Smaland in the south, Solander in northern Vestrobothnia and Sparrman in the central province of Uppland. Their fathers were clergymen, their grand- or great grandfathers, on one side or the other, had been peasants. Their links with rural life and knowledge were vital, and particularly obvious for Linnaeus. These three Fig. 18. Linnaea borealis L. From a copper plate engraved for J. W. Palmstruch: Svensk Botanzk (1802- 1843) and used in Lindman: Bilder ur Nordens Flora (Stockholm, 1922). Proc. Linn. Soc. N.S.W., 106 (1), (1981) 1982 18 LINNAEUS AND HIS CIRCUMNAVIGATING APOSTLES distinguished men possessed a social and educational background typical of many of those men of learning who made the 18th century in Sweden a brilliant period for science. Their names live by their achievements and also through the well-known plants named in their honour: Sparrmannia L.f. (Fig. 16), the South African tiliaceous genus of many greenhouses, Solandra Sw. (Fig. 17), the spectacular member of the Solanaceae from Central and South America, now widely cultivated, and Linnaea Gron. (Fig. 18), which Gronovius in Holland dedicated to Linnaeus and to which the latter claimed to find a likeness in himself. Linnaeus’s personality may not seem to us to have much in common with this shy, secluded plant of the northern coniferous forests but it has nevertheless become a sort of symbol for Linnaeus and his era. ACKNOWLEDGEMENTS My thanks are due to staff members of the University Library, the Institutes of Art Science and of Systematic Botany at Uppsala University, and of the Department of Botany, Museum of Natural History, Stockholm, for multifarious assistance and photographic work. The Hon. Editor of the Proceedings, Prof. Vallance, is thanked for revising the English text. References BEAGLEHOLE, J. C. (ed.), 1962. — The Endeavour Journal of Joseph Banks 1768-1771. 2 vols. Sydney: Public Library of N.S.W. BLunT, W., and STEARN, W. T., 1973. — Captain Cook’s Florilegium. London: Lion & Unicorn Press. BRITTEN, J., 1901-05. — Illustratzons of Australzan Plants collected in 1770 during Captain Cook’s voyage. London: British Museum. BroserG, G., 1975. — Homo sapiens L. Studzer ¢ Carl von Linnés naturuppfattning och manniskolara. Uppsala/Stockholm: Almavist & Wiksell. Brown, R., 1810. — Prodromus florae Novae Hollandiae et Insulae van-Diemen. Vol. I London: [The author ]. ERIKSSON, G., 1969. — Botanzkens historia 7 Sverige intill adr 1800. Stockholm: Almgqvist & Wiksell. Forster, G., 1777. — A Voyage round the World in His Britannic Majesty's sloop Resolution. 2 vols. London: B. White, J. Robson, P. Elmsly and G. Robinson. ForSTER, J. R., and Forster, G., 1776. — Characteres generum plantarum. London: B. White, T. Cadell and P. Elmsly. 5 Fries, R., 1940. — Daniel Solander. K. Svenska Vet. Akad. Arsb. 38: 279-301. ———, 1951. — De linneanska apostlarnas resor. Kommentarer till en karta. Svenska Linnésallsk. Arsskr. 33-34: 31-40. GAERTNER, J., 1788-91. — De fructibus et seminibus plantarum. Stuttgart/Tubingen: Typis Academiae Carolinae/G. H. Schramm. Granit, R., 1978. — Banks och Solander — tva vanner i 1700-talets London: Jn Granit, R., ed., Utur stubbotan rot (pp. 41-59). Stockholm: Norstedts. ‘ JuEL, O., 1924. — Notes on the-herbarium of Abraham Back. Svenska Linnésallsk. Arsskr. 7: 68-82. LinDMAN, C., 1907, 1909. — A Linnean herbarium in the Natural History Museum in Stockholm, I-II. Arkiv f. Botantk 7(3) ; 9(6). LINDROTH, S., 1967. — Kungl. Svenska Vetenskapsakademiens historia 1739-1818. Vols 1, 2. Stockholm: Almgvist & Wiksell. LINNAEUS, C., 1758-59. — Systema naturae. Ed. 10. Holmiae: Impensis L. Salvii. ———, 1759. — Flora capensis (resp. C. H. Wannman). Diss. Upsaliae. ———, 1760. — Plantae africanae rariores (resp. J. Printz). Diss. Upsaliae. ———, 1760. — Anthropomorpha (resp. C. E. Hoppius) . Diss. Upsaliae. ——, 1909-12. — Bref och skrifvelser af och till Carl von Linné (Frirs, T. M., ed.), I, part 3-6. Stockholm: Ljus. ———, 1957. — Vita Caroli Linnaez (MALMESTROM, E., and UcGLa, A., eds). Stockholm: Almgvist & Wiksell. Linn—E, C. von, fil., 1781. Supplementum plantarum. Braunschweig (Brunsvigae): Impensis Orphanotrophei. Proc. Linn. Soc. N.S.W., 106 (1), (1981) 1982 B. JONSELL 19 MERRILL, E. D., 1954. — The botany of Cook’s voyages. Chronica Botanica 14 (5/6). RAUSCHENBERG, R. A., 1968. — Daniel Carl Solander, naturalist on the ‘Endeavour’. Trans. Amer. phil. Soc. n.s. 58 (8). . SELLING, O., 1962. — Daniel Solanders naturaliekabinett och dess 6den. Svenska Linnesallsk. Arsskr. 45: 128-137. SMITH, B., 1960. — European Vision and the South Pacific. Oxford: Oxford Univ. Press. SmiTH, J. E., 1821. — A Selection of the Correspondence of Linnaeus and other Naturalists. 2 vols. London: Longman, Hurst, Rees. SGDERSTROM, J., 1939. — A. Sparrman’s ethnographical collection from James Cook’s 2nd expedition (1772-1775). Ethnogr. Mus. Sweden, Stockholm, n.s. Publ. 6. SPARRMAN, A., 1783, 1802, 1818. — Resa till Goda Hopps-Udden, Sodra Pol-Kretsen och Omkring Jordklotet samt till Hottentott- och Caffer-landen Gren 1772-76. 2 vols. Stockholm: A. J. Nordstr6ém/C. Delén. STEARN, W. T., 1969. — A Royal Society appointment with Venus in 1769: The voyage of Cook and Banks in the Endeavour in 1768-1771] and its botanical results. Notes Rec. Roy. Soc. London 24: 64-90. Uccia, A. H., 1955. — Daniel Solander och Linné. Svenska Linnésallsk. Arsskr. 37-38: 23-64. Proc. Linn. Soc. N.S.W., 106 (1), (1981) 1982 Co (a Re eer : SOF tale Bae ‘ang Py Re et eee) ae Two new intertidal balanoid Barnacles from eastern Australia BRIAN A. FOSTER (Communicated by D. T. ANDERSON) Foster, B. A. Two new intertidal balanoid barnacles from eastern Australia. Proc. Linn. Soc. N.S.W. 106 (1), (1981) 1982: 21-32. Two species of balanomorph barnacles are described from harbours and estuaries of New South Wales where they are locally very abundant intertidally. Elminius covertus sp. nov., is closely related to E. modestus from New Zealand, and less so to Elminzus kingw from South America, and has long been mistaken for Elmanius modestus which is now known reliably only in Australia from some ports. Hexaminius popeiana gen. nov. and sp. nov. is superficially similar to Elmznzus spp. but has 6 parietal plates. Both species are assigned to the Elmininae, a new subfamily in the Archaeobalanidae. Brian A. Foster, Department of Zoology, University of Auckland, Private Bag, Auckland, New Zealand; manuscript received 21 October 1980, accepted for publication in revised form 19 August 1981. INTRODUCTION Elminius modestus Darwin has .become a well-known barnacle since its introduction and spread in Europe (see full literature citation in Newman and Ross, 1976). It has been assumed that the species was carried to Britain on ships in the 1930- 40 period from Australia and New Zealand (Bishop, 1947). In New Zealand E. modestus is very common in harbours and estuaries where it occurs on a wide variety of substrata in the midlittoral and shallow sublittoral zones (Moore, 1944; Morton and Miller, 1968; Foster, 1978). It is the commonest fouling barnacle in New Zealand harbours. For Australia, the literature on E. modestus is perplexing to one used to the species on New Zealand (and European) shores. The early records of Darwin (1854), and the later account of Pope (1945) for Sydney shores, seem to be the basis for the Australian presence and identity of the species. For Port Jackson, Pope (1945) reported that E. modestus occurred ‘only towards the upper limit of the tidal range which prevents it from becoming much of a nuisance on ships and small craft’. This anomaly in the intertidal range, plus the lack of data on the distribution of the species away from Sydney, prompted a closer look at the sheltered-shore barnacles of New South Wales. It was concluded (Foster, 1980) that the account of Pope (1945) refers to a different species of Elmznzus which occurs along with E. modestus and an undescribed 6-plated balanid on harbour shores in the vicinity of Sydney. This present paper describes both new species and considers their relationships within the Balanomorpha as recently revised by Newman and Ross (1976). The barnacles described in this paper are abundant where they occur. Samples collected and studied contain at least 10, up to 50, specimens, at various stages of development and erosion. The two species of Elminzus are difficult to identify when the shell is extremely eroded, and close examination is required to distinguish all 3 species when they are mud encrusted. Proc. Linn. Soc. N.S.W., 106 (1), (1981) 1982 22 NEW BALANOID BARNACLES SYSTEMATICS Suborder BALANOMORPHA Pilsbry, 1916 Superfamily BALANOIDEA Leach, 1817 Family ARCHAEOBALANIDAE Newman and Ross, 1976 Balanomorph barnacles with labrum thin with pronounced medial cleft; shell wall either solid or with pores of the non-balanine type, inter-laminate figures lacking or simple. Remarks: The Archaeobalanidae contain 10 extant and 4 extinct genera. The solid- walled genera fall into two groups, those that associate with various sponge or coelenterate habitats (Armatobalanus, Membranobalanus, Acasta and Conopea), and those that are free-living (Chzrona, Solidobalanus, Elminius and Notobalanus) . The former have hooks and teeth on cirri III and IV, the latter do not; the latter are considered to be the more primitive. The various forms of tubiferous parietes as seen in Semzbalanus and the extinct Archaeobalanus and Actinobalanus are considered to be independent developments from the much more prevalent parietal pore condition of the Balanidae (Newman and Ross, 1976). I have argued before (Foster, 1978, p. 130), that the calcareous base of balanomorph barnacles has been variously evolved in different groups, and that the primitive condition is a membranous base. Three of the archaeobalanid genera have membranous bases: Semzbalanus, Membranobalanus and Elminius. Elminius is the only 4-plated genus in the archaeobalanids, but its simple cirri, its membranous base, and the non-interlocking radii indicate a basic origin for Elmznzus. The new 6-plated form described below also has simple cirri and a membranous base, and this combination of characters denies its inclusion in any of the aforementioned genera. It also has non-interlocking radii and seems closely allied to Elmznzus. Rather than amend the diagnoses of Elmznzus or of Chzrona (which has a calcareous base) a new genus is proposed for this undescribed species. Furthermore, this species, and those of Elminius, seem to constitute a southern hemisphere subfamily as well defined as the A — 1mm B —_1mm Fig. 1. A: Elminius modestus Darwin, drawn from a New Zealand specimen; B: Elminius covertus n. sp., drawn from a specimen from Swansea, Lake Macquarie. Proc. Linn. Soc. N.S.W., 106 (1), (1981) 1982 B. A. FOSTER 23 northern hemisphere Semibalaninae. I propose the name Elmininae for this subfamily. Genus Elminius Leach, 1825 Balanoid barnacles with a medially cleft labrum; 4 thin, solid parietes, radii simple, base membranous; cirri III and IV without hooks or teeth; basi-dorsal point of penis absent. Type species Elmanzus kingaz Gray, 1831 (subsequent designation, Pilsbry 1916). Remarks: Elminius includes 3 extant species: E. kengz from southern South America, E. modestus from New Zealand, Australia and Europe, and the new species described below. The identity of E. crzstallinus Gruvel from the Azores needs confirmation; it might be a case of introduced E. modestus (Newman and Ross, 1976) or the young 4- plated stage of some other species. The subspecies E. modestus molluscorum Kolosvary from Auckland, and E. modestus laevis Nilsson-Cantell from Melbourne are just two of the forms assumed by this species when growing on various intertidal substrata (Moore, 1944). Elminius modestus Darwin Figs 1A, 2 Synonymy: Elmznzus modestus Darwin, 1854:350; Guiler, 1952:20; Pope, 1966:181; Foster, 1978:95 (see for New Zealand references) ; Newman and Ross, 1976:52 (in part, viz. the European and New Zealand references) ; Foster, 1980: figs 2, 3. Not Pope, 1945:368; Underwood, 1977:23, 217. Elminius modestus molluscorum Kolosvary, 1942 :147 Elminius modestus laevts Nilsson-Cantell, 1925 :49 Australian material examined. Personal collection: Port Jackson (The Spit), 1978; Melbourne (coll. J. S. Buckeridge, 1976) ; Adelaide (coll. A. Gackle, 1979). Could not be relocated in Port Jackson in 1980. Australian Museum: P21284 (Newport Power House, Victoria, 1950), P21287 (Derwent River, Tasmania, 1950), P21288 (Sandy Bay, Hobart, Tasmania, 1951), P21291 (Port Lincoln, South Australia, 1951), P23569 (Port Phillip, Melbourne, boat fouling 1950), P23571 (Port Jackson, Sydney, ship fouling, 1954). There are numerous lots of specimens that have been classified under E. modestus but are referable to E. covertus (below). Description as given by Moore (1944) and Foster (1978). Designation of lectotype: British Museum (Natural History), one of several specimens attached to an intertidal limpet Cellana ornata (Dillwyn), Registered No. B.M. 47.1.15.39, as arrowed in Fig. 2 and now held separately, Reg. No. B.M. 1981.274. The limpet shell is one of four shells on a wooden slab bearing on reverse a label in Darwin’s handwriting. These shells are endemic to New Zealand, and despite the lack of collecting data it is assumed that the Elmznius specimens are part of Darwin’s syntype material and originated from New Zealand. Specimens of the barnacle Chamaesipho columna (Spengler) are included on two of the limpet shells. This association of barnacles and limpets is common on New Zealand harbour shores. Remarks: Darwin (1854, p. 348) stated ‘at Sydney I found E. modestus adhering to oysters in a muddy lagoon, almost separated from the sea, and apparently very unfavourable for cirripedes’. The question arises whether Darwin found specimens of the other species of Elminzus described below, and whether he confounded the two species in his description of E. modestus. Specimens now in the British Museum, some with labels identifying them as E. modestus in Darwin’s handwriting, are indeed E. modestus as known in New Zealand and as has been introduced into Europe. It is Proc. Linn. Soc. N.S.W., 106 (1), (1981) 1982 24 NEW BALANOID BARNACLES A B C. —2 d Lt tr. fue ee, ff Bt oe o hac oe Lk. e Fig. 2. Elminius modestus Darwin. A: lectotype arrowed, on shell of Cellana ornata (Dillwyn). B: lectotype and syntypes on New Zealand gastropod shells — Cellana radians (Gmelin), left, Turbo smaragda Gmelin, bottom, and Cellana ornata (Dillwyn), top and right. C: Darwin’s own label as applied to reverse of slab pictured in B. Depository: British Museum (Natural History) , London. suspected not all Darwin’s material is now in the British Museum, and there is doubt (Foster, 1980) about the labelling and the localities of some specimens he had to study. Nor are his Sydney specimens identifiable, if indeed he collected them. Darwin did not designate types and it seems purely academic to seek out all remnants of Darwin’s barnacles in the possibility that some may not be E. modestus. Insofar as it goes, Darwin’s description of E. modestus does describe E. modestus as it is known in New Zealand and Europe. The barnacles Darwin ‘found’ at Sydney were on oysters, and on Sydney shores oysters are ‘zoned’ below the confounding species. Therefore, it is possible that Darwin did collect E. modestus at Sydney in 1836. More discerning collecting and identification of Elmznzus is required to ascertain the details of the geographic distribution of both species in Australia. If it is found to be restricted to major shipping ports, and absent from the intervening estuaries on the coast, then it is possible that ship-aided dispersal to Australia has occurred, some at least pre-1836. ; Without examination of specimens, it is not possible to determine the true identity of the barnacles listed by Guiler (1952) and Pope (1966), and figured by Underwood (1977). Elminius covertus n.sp. Figs 1B, 3, 4A-C Synonymy: Elminius modestus Pope, 1945:368; Underwood, 1977:23, ?17; Guiler, 1952:20; Pope, 1966:181 Elminius sp. Foster, 1980:614 (figs 1, 3, 4) Material examined. Personal collection: in 1975, Pittwater and Port Jackson; in 1978, Coffs Harbour, Nambucca River (Macksville), Hastings River (Settlement Point, Port Macquarie), Camden Haven, Wallis Lake (Tuncurry), Port Stephens (Tea Gardens, Nelson Bay), Port Hunter (Stockton), Lake Macquarie (Swansea), Broken Bay (Gosford, Bobbin Head, Pittwater), Port Jackson (Vaucluse, The Spit, Bantry Proc. Linn. Soc. N.S.W., 106 (1), (1981) 1982 B. A. FOSTER G H J Fig. 3. Elminius covertus. A: inner view of shell; B: scutum; C: tergum; D: labrum; E: mandible; F: maxilulle; G-I: cirri I-III; J: seta from 5th segment, posterior ramus, cirrus III; K: middle segment, posterior ramus, cirrus VI. All drawn from a 9 mm rostrocarinal diameter specimen. Bay, Roseville Bridge), Botany Bay (Captain Cook Bridge), Port Hacking (Cronulla) ; in 1979, Port Jackson (Rose Bay, coll. M. F. Barker) ; in 1980, Pittwater and Bennelong Point. Australian Museum: Queensland localities, P21306, P21313 (both Stradbrooke Island, 1961); New South Wales localities, P21296 (Hawkesbury, 1967), P21314 Proc. Linn. Soc. N.S.W., 106 (1), (1981) 1982 26 NEW BALANOID BARNACLES (Pittwater, 1962), P21301-(Port Jackson, 1953), P21303 (Port Hacking, 1952) ; Victorian localities, P11643 (Kangaroo Island, Bass Strait, 1945), P21283 (Fisher Island, Bass Strait, 1930), P21297 (Mallacoota Inlet, 1957), P21309 (Westernport Bay, 1962); Tasmanian localities, P21286 (Recherche Bay, 1953), P21288 (Sandy Bay, Hobart (195) 221289 s(Port Anchunme 195i) 220 Ons Gear are oy) P21295 (Port Arthur, 1951); South Australian localities, P21294 (Port Adelaide, 1950) ; Western Australian localities, P21311 (Bunbury Jetty, 1959), P21316 (Oyster Bay, Albany, 1959). ; Deposition of types: Type specimens from Lake Macquarie (Swansea), coll. 8 March 1978. Holotype; Australian Museum Cat. No. P30976. Paratypes; Australian Museum, Sydney (Cat. No. P30977), National Museum, Wellington (Cat. No. Cr2238), British Museum (Natural History), London (Cat. No. 1980. 305-314), U.S. National Museum, Washington (Cat. No. 181718). Description: Shell (Figs 1B, 3A) : flat, with 4 parietal plates, up to 15 mm across and 4 mm high. Parietes thin, solid; basal margin sinuous from within. In uneroded specimens, parietes have pale narrow ribs alternating with pale wine-red coloured spaces between: 4 or 5 ribs per plate in young specimens but up to 9 in larger specimens. In eroded specimens, the shell is grey and granulate apically, but usually with some evidence peripherally of dark reddish-purple laminae and ribbing. Radii narrow, not completely covering underlying alae in uneroded specimens, dark wine-red in colour, not interlocking with adjacent parietes. On erosion, the sutures between parietes remain obvious but the radii and alae are indistinguishable. Base membranous. Orifice pentagonal in outline, elongate in the rostrocarinal axis with a very short rostral side. Opercula (Figs 3B, C) : Scutum about as long or longer than high, the articular ridge occupying half articular margin and curving into the articular furrow. Internally there is no adductor ridge, nor crests for depressor muscles, but a faint adductor muscle scar. Tergum elongated with a very wide articular furrow; articular ridge standing at right angles to the pitted innerface of the plate, curving directly to a spur that merges with the basiscutal angle; basal margin concave, with 4 or 5 crests depending at the carinal end. Externally in live, uneroded specimens, the scuta are dark red with a conspicuous widening white band along the tergal margin of each scutum; opercular membranes are white with 6 pairs of dark spots as shown in Fig. 1B. Mouthparts (Figs 3D-F) : Labrum with 3 teeth on each side of central cleft. Mandible with 5 teeth, the 2nd about % along the cutting edge from the upper tooth, the 5th merging with a pectinate lower angle which bears a prominent curved spine at the bottom point. Maxillule with a wide but shallow notch below the upper pair of spines, the main cutting edge not protuberant bearing 3 or 4 major spines and a clump of short ones at the lower angle. Cirri (Figs 3G-K) : Number of segments in the rami of 5 specimens as follows, anterior ramus first : Shell length (mm) I II Ill IV V VI 8.5 EG 7,6 10,9 16,17 16,18 20,19 5.2 (eroded) 12,6 @) 7 11,10 15,17 23,21 25,23 12.0 14,6 9,7 11,10 17,20 24,19 21,19 12.6 16,6 9.8 U2, 18,19 22) 24 25,26 9.2 (eroded) 16,7 9.8 12,10 D3.) 26,27 BI 27 Cirrus I with anterior ramus 4% as long again as posterior ramus, distal segments elongate with long setae; posterior ramus with segments slightly protuberant Proc. Linn. Soc. N.S.W., 106 (1), (1981) 1982 B. A. FOSTER 27 Fig. 4. A: Elminius covertus, group of specimens from Pittwater, photographed alive, X 0.9; B: E. covertus left and centre, Hexaminius popeiana right, from Swansea, Lake Macquarie, photographed alive, 2.5: C-F: medial face of middle segments of posterior ramus of cirrus III of E. covertus (X 220) C, H. popetana (X 125) D, Elminzus modestus (x 60) E, Elminius king@i (X 220) F. Proc. Linn. Soc. N.S.W., 106 (1), (1981) 1982 28 NEW BALANOID BARNACLES anteriorly. Cirrus II with anterior ramus slightly longer than posterior ramus, segments of both rami setose and slightly protuberant anteriorly. Cirrus III with subequal rami, all but basal 1 or 2 segments on medial face of posterior ramus with pectinate setae (Fig. 4C). These pectinate setae are much stouter than the finely feathered setae of the posterior rami. Cirri IV to VI with square segments with 4, rarely 5, pairs of setae on the anterior edge. Penis 1 x length of cirrus VI, no basal dorsal point or setae. Etymology: with reference to the species being so long unnoticed on harbour shores. Remarks: As both E. modestus and E. covertus may occur together it is desirable to emphasize the distinguishing characters. In young uneroded shells the reddish to buff colour of E. covertus is in marked contrast to the vivid white shell of E. modestus. Also, the narrow contrastingly-coloured ribs of E. covertus differ from the uniformly white and broadly-folded parietes of E. modestus. Eroded shells are harder to identify; close inspection may reveal a stellate basal margin where ‘the continuation of the ribs stick out round the margin in a series of points’ (Pope, 1945:369). As with all intertidal barnacles, there is some variation in the dimensions of the scutum, ranging from almost isosceles (with articular and basal margins equal) to the basal margin being 114 times the length of the articular margin, perhaps a feature of flat conic species growing on different substratum contours. The terga also vary in shape mostly due to the degree of erosion of the outer face. The terga of E. covertus are more elongate than those of E. modestus (‘hatchet- shaped’ of Pope, 1945), with the articular ridge overhanging the spur with which it merges (forming the ‘handle of the hatchet’) and extending as a thickened rim around the apico-carinal region. The triangular region between the depressor muscle crests and the spur is notably thinner, often pitted. In E. modestus the inner surface slopes to the apex and articular ridge, and is roughened in larger specimens. The major point of difference in the appendages is the presence of pectinate setae on the posterior ramus of cirrus III in E. covertus (Fig. 4C), absent in E. modestus (Fig. 4E). In the gaping animals, the tergoscutal flaps reveal the most marked distinguishing character. In E. covertus (see Fig. 4A, B; Fig. 1B) they are white with 6 pairs of discrete black spots, one at the groove, 3 to the carinal side and 2 to the rostral side. On the basis of the opercula, Elmznzus covertus and E. modestus are more closely related to each other than to E. kengz of South America; in E. kengz the tergal spur is separate from the basiscutal angle and aligned with the articular margin as it is in Solidobalanus spp. However, the posterior ramus of cirrus III of E. kengz bears pectinate setae (Fig. 4F) like those in E. covertus (Fig. 4D), lacking in E. modestus (Fig. 4E). Genus Hexaminius nov. Balanoid barnacles with a medially cleft labrum; 6 thin, solid parietes; rostrum not elongated; radii simple; base membranous; cirri III and IV without hooks or teeth; basi-dorsal point of penis vestigial. Remarks: This is a monospecific genus proposed to accommodate the species described below, hereby designated the type species. Hexaminius popezana n.sp. Figs 4B, 4D, 5, 6 Synonymy: Solzdobalanus sp. Foster, 1980: p, 614, fig. 3 Proc. LINN. Soc. N.S.W., 106 (1), (1981) 1982 B. A. FOSTER 29 Material examined. Personal collection: New South Wales localities: In 1975, Careel Bay; in 1978, Hastings River (Settlement Point, Port Macquarie), Wallis Lake (Tuncurry), Port Stephens (Tea Gardens, Karuah, Nelson Bay) ; Lake Macquarie (Swansea) ; Broken Bay (Gosford, Careel Bay, Bobbin Head); Port Jackson (The Spit, Roseville Bridge, Vaucluse) ; Botany Bay (Captain Cook Bridge) ; Port Hacking (Cronulla). Further specimens from Rose Bay (Port Jackson) in 1979, (coll. M. F. Barker) ; in 1980, Pittwater (Broken Bay) and Bennelong Point, Port Jackson. Deposition of types: Holotype and paratypes — Australian Museum, Sydney, Cat. No. P30974 (holotype) Cat. No. P30975 (paratypes) from Lake Macquarie, 8 March 1978. Paratypes — National Museum, Wellington (Cat. No. Cr2239), from Cowan Creek, Hawkesbury, coll. 9 March 1978; British Museum (Natural History) , London (Cat. No. 1980.315-320) , from Lake Macquarie, coll. 8 March 1978; U.S. National Museum (Cat. No. 181719), Cowan Creek, Hawkesbury, coll. 9 March 1978. Description: Shell (Figs 5, 6A): conic, up to 10 mm across and 4 mm high. Rostrum wide; carinolatera about 1/3 width of latera. Compartments solid throughout, thin, sometimes broadly sinuous, without ribs internally, with basal edge truncated. Radii narrow, with simple edges, not interlocking with adjacent parietes, exposing much of the underlying ala. In young and uneroded specimens, shell compartments with broad pale bands alternating with bands of purplish to wine red colour, the latter with white flecks which underly short projecting spines that line the growth ridges. In eroded specimens, colour generally white with narrow darker stripes confined to the depressions between the broad ‘ribs’. Base membranous. Orifice pentagonal in outline, with a very short rostral side. Opercula (Figs 6B, C): Scutum longer than high, articular ridge occupying 2/3 articular margin, with deep articular furrow; internally with very faint muscle attachment scars, and a weak adductor ridge. Tergum waisted by having a concave basal margin, but straight articular margin; deep articular furrow; spur close to but separate from the basiscutal angle, forming an acute angle with the short part of the —_ Imm Fig. 5. Hexaminius popecana n.gen. and n. sp., drawn from a specimen from Rose Bay, Port Jackson. Proc. Linn. Soc. N.S.W., 106 (1), (1981) 1982 30 NEW BALANOID BARNACLES Fig. 6. Hexaminius popezana. A: inner view of shell; B: scutum; C: tergum; D: labrum; E: mandible; F: maxilulle; G-I: cirri I-III; J: seta from 7th segment, posterior ramus, cirrus III; K: middle segment, posterior ramus, cirrus VI. All drawn from a 9 mm rostrocarinal diameter specimen. basal margin on that side, spur curves gradually to the concave basal margin, which on the carinal side is prominent with 4 or 5 crests for the carinal depressor muscles. Externally in live specimens, the scuta are dark purple red with an inconspicuous pale flange along the tergal articular edge of each scutum; opercular membranes off-white with paired markings as shown in Fig. 6C. Proc. Linn. Soc. N.S.W., 106 (1), (1981) 1982 B. A. FOSTER 31 Mouthparts (Figs 6D-F): Labrum cleft, with 3 teeth on either side. Mandible with 5 teeth, the uppermost separated from the 2nd by half the length of the cutting edge; the 3rd, 4th and 5th merging into a molariform lower angle which may bear a few short spines. Maxillule with a slight notch below the upper pair of spines, the edge below not protuberant, with 5 or 6 spines, the lowest 2 of which equal in length the 2 above the notch. Cirri (Figs 6G-K) : The numbers of segments in the rami of cirri of 3 specimens as follows, anterior ramus first : Shell length I II Ill IV V VI 4.8mm 9.6 8,7 9.9 16,17 20,20 21,20 9.6mm 14,7 8,8 10,10 20,21 24,24 25,25 9.6mm 7 8,9 11,10 21,20 2424 26,26 Cirrus I with anterior ramus only slightly longer than posterior ramus. Cirrus IT with subequal rami. Cirrus III with subequal rami, setae on distal 3 or 4 segments of both rami serrated more on one side than the other (Fig. 4D and Fig. 6J). Basal segment of cirri III very broad. Cirri IV to VI with segments slightly longer than wide, each with 5 pairs of setae on the anterior edge. Penis longer than cirrus VI, with a small basal dorsal point bearing 2 setae, and with a prominent pedicle. Etymology: named in honour of Miss Elizabeth Pope, student of Australian barnacles, whose collections in the Australian Museum contain numerous lots of this species. It is a species she has clearly pondered on. Remarks: The poreless shell structure and unmodified 3rd and 4th cirri indicate an affinity with Notobalanus and Chirona; the position of the tergal spur, set aligned to the articular margin and very near to the basiscutal angle, resembles Solzdobalanus and Elminius kingz. The absence of a calcareous base precludes its inclusion in any of these genera. Nor can it be included in the membranous-based genera Semzbalanus and Membranobalanus, because of the poreless shell and simple cirri respectively. The similarity of H. popezana with Elmznzus is striking, particularly with respect to the primitive features of non-toothed cirri, basiscutal position of tergal spur, the membranous base, and the non-interlocking radii. Indeed, Hexamznzus satisfies the requirements of an Elmznzus ancestor as discussed by Foster (1978, p. 97). All species of Elmznzus and Hexaminius are characteristic of shallow sea habitats and estuaries. Their weakly-constructed shells are inappropriate to surf habitats. These species, the Elmininae, may represent the survivors of an early stage in balanid radiation before the development of stronger radially-interlocked shells like those of Solidobalanus. They perhaps indicate a southern hemisphere development to parallel the Semibalaninae (e.g. Semzbalanus balanozdes) of the northern hemisphere. The fouling proclivity of E. modestus has enabled it to overcome oceanic barriers and become sympatric with related species in Australia and Europe. ACKNOWLEDGEMENTS I am grateful to the Director of the Australian Museum for allowing me to study specimens therein, and acknowledge the work of Miss Elizabeth Pope whose personal collecting significantly contributed to the barnacles collection in the Museum. For other specimens I am grateful to Miss A. Gackle, Dr John Buckeridge and Dr Michael Barker. I thank Prof. D. T. Anderson for reading the manuscript, Mr Charles McKenzie and Mr Gary Batt for help with the illustrations, and the Auckland University Research Committee for funding travel to Australia. References BisHop, M. W. H., 1947. — Establishment of an immigrant barnacle in British coastal waters. Nature 159:501. Proc. Linn. Soc. N.S.W., 106 (1), (1981) 1982 32 NEW BALANOID BARNACLES Darwin, C., 1854. — A monograph of the subclass Cirripedia with figures of all the species. The Balanidae, the Verrucidae etc. London: Ray Society. Foster, B. A., 1978. — The marine fauna of New Zealand: Barnacles (Cirripedia Thoracica) . Mem. N.Z. Oceanogr. Inst. 69:1-160. ——, 1980. — Biogeographic implications of re-examination of some common shore barnacles of Australia and New Zealand. Procs. Int. Symp. Mar. Biogeography & Evolution in the Southern Hemisphere, 613-623. N.Z. DSIR Information Ser. 137. GuILER, E. R., 1952. — A list of the Crustacea of Tasmania. Rec. Queen Victoria Mus. 3:15-44. Ko.osvary, G., 1942. — Studien un Cirripedien. Zool. Anz. 137:138-150. Moore, L. B., 1944. — Some intertidal sessile barnacles of New Zealand. Trans. Proc. R. Soc. New Zealand 73 :315-334. Morton, J. E., and MILER, M. C., 1968. — The New Zealand Sea Shore. London, Auckland: Collins. Newman, W. A., and Ross, A., 1976. — Revision of the balanomorph barnacles; including a catalog of the species. Mem. San Diego Soc. Nat. Hist. 9:1-108. NILSsON-CANTELL, C. A. 1925. — Neue und wenig bekannte Cirripeden aus den Museen zu Stockholm und zu Upsala. Ark. Zool. 18A:1-46. Pope, E. C., 1945. — A simplified key to the sessile barnacles found on the rocks, boats, wharf piles and other installations in Port Jackson and adjacent waters. Rec. Australian Museum 21:351-372. ——, 1966. — Sessile barnacles (Thoracica, Cirripedia). Port Phillip Survey, 1957-1963. Mem. Natl Mus. Melbourne 27: 179-182. SOUTHWARD, A. J., and Crisp, D. J., 1963. — Barnacles of European waters. In: Catalogue of main marine fouling organisms, Vol. 1, Barnacles, pp. 1-46. Paris: Organisation for Economic Cooperation and Development. UNDERWOOD, A. J., 1977. — Barnacles. Science Field Guides. Sydney: Reed Education. Proc. LINN. Soc. N.S.W., 106 (1), (1981) 1982 Systematics of Tapezgaster (Diptera: Heleomyzidae) with notes on biology and larval morphology DAVID K. McALPINE and DEBORAH S. KENT McALpineE, D. K., & KENT, D. S. Systematics of Tapezgaster (Diptera: Heleomyzidae) with notes on biology and larval morphology. Proc. Linn. Soc. N.S.W. 106 (1), (1981) 1982: 33-58. The genus Tapezgaster, which inhabits temperate Australia, is reviewed systematically and placed in the monogeneric tribe Tapeigastrini of the family Heleomyzidae. Four species are described as new. New synonymy is recorded for five species. Lectotypes are designated for four nominal species. A neotype is designated for Dryomyza cingulipes Walker. The third instar larva and puparium are described for two species. Brief notes on the biology of the insects are given, with a list of recorded host fungi. David K. McAlpine and Deborah S. Kent, The Australian Museum, Box A285, Sydney South, Australia 2000; manuscript received 22 April 1981, accepted for publication 19 August 1981. INTRODUCTION Although the genus Tapezgaster was reviewed taxonomically by Paramonov (1955) this is considered an appropriate time to present a revision of the genus for several reasons. Type material of all nominal species has been examined (except for one which appears no longer to exist) as well as a considerable amount of new material. As a result it becomes necessary to change the names in use for several species, and to describe four new species. As no additional new species has come to the attention of the senior author in more than twenty years it appears that the register of species is complete or almost so. In listing material the names of the following collectors are abbreviated to the initials: D. H. Colless, I. F. Common, G. Daniels, B. J. Day, G. A. Holloway, Z. R. Liepa, D. K. McAlpine, K. R. Norris, M. S. Upton. Institutions housing material are abbreviated thus: AM, Australian Museum, Sydney; ANIC, Division of Entomology, Commonwealth Scientific and Industrial Research Organization, Canberra; BCRI, Biological and Chemical Research Institute, (N.S.W. Department of Agriculture) , Rydalmere, Sydney; BM, British Museum (Natural History), London; CNC, Canadian National Insect Collection, Agriculture Canada, Ottawa; PM, Muséum National d’Histoire Naturelle, Paris; SPHTM, Commonwealth Institute of Health (formerly School of Public Health and Tropical Medicine), University of Sydney; UQ, Department of Entomology, University of Queensland; USNM, National Museum of Natural History, Smithsonian Institution, Washington, D.C.; WAM, Western Australian Museum, Perth; WM, Naturhistorisches Museum, Vienna. Tribe TAPEIGASTRINI The tribal characters are those of the only genus, but the main distinguishing characters may be summarized as follows: Fronto-orbital plates short and parallel with eye margin; face concave, uniformly Proc. Linn. Soc. N.S.W.,106 (1), (1981) 1982 34 SYSTEMATICS OF TAPEIGASTER sclerotized; prelabrum reduced. Antenna decumbent, with arista inserted sub-basally on segment 3. Thorax with one or 2 dorsocentral bristles and no mesopleural bristle. All femora with ventral spines. Costal spines absent; subcosta complete, diverging from vein 1 distally; vein 6 long, sometimes attaining wing margin. We place the genus Tapezgaster alone in the tribe because it is only distantly related to any other heleomyzid genus, and its nearest ally is unknown. This tribe is considered to have similar status to other groups in the Heleomyzidae, which are probably best regarded as tribes e.g. Suilliini, Allophylopsini, Cnemospathini, Rhinotorini, Trixoscelidini. The species of Tapezgaster are apparently restricted to temperate parts of Australia. Probably, then, the group has evolved in isolation on the Australian continent for much of the Tertiary; or, alternatively, Tapezgaster may be a relict group, whose nearest allies in other regions are extinct. Griffiths (1972) has associated Tapezgaster with the Rhinotorini, to which he gives family rank. We find the resemblances between Tapezgaster and the other genera referred by Griffiths to the Rhinotoridae unconvincing as indicators of close phylogenetic affinity, because most of them are widely distributed among heleomyzoid groups or seem likely to be readily duplicated by convergence. On the other hand, Tapeigaster shows some fundamental differences from the rhinotorine genera. In Tapeigaster the antenna is decumbent, with the arista inserted near the base of segment 3; the prelabrum is reduced; the subcosta forms a distinct tubular vein right to its junction with costa, and diverges from vein 1 distally; vein 6 is gradually reduced distally and generally long (most reduced in T. pulverea) ; vein 7 is distinct; the anal crossvein is transverse and forms with vein 6 a definite posterodistal angle to anal cell. In these characters Tapezgaster contrasts with all true rhinotorine genera (including Apophoneura and Anastomyza). The larva of Tapezgaster is very different from that of the only known rhinotorine larva (Cazrnsimyza robusta (Walker), see McAlpine, 1968). The pupa of Cazrnsimyza lacks the prothoracic respiratory horn which perforates the puparium in Tapezgaster. Griffiths gives as a ground plan condition of Rhinotoridae: ‘Cerci of males small, fused below anus’. In Tapezgaster the cerci are neither significantly reduced in size nor fused. McAlpine (1968) gives reasons for considering the Rhinotorini to be an integral part of the family Heleomyzidae. As Griffiths’s fragmentation of the Heleomyzidae depends largely on the erroneous assumption that characters of the male postabdomen are appropriate for higher classification within the Schizophora, we continue to regard Rhinotorini and equivalent groups as tribes. Because structural divergence between closely related species is usually greater in male postabdominal organs than in any other part of the insect, these characters must be regarded as the least stable above the species level and of low reliability as indicators of broad relationships. Genus TAPEIGASTER Tapeigaster Macquart, 1847: 86-87 (in reprint). Type species T. annulzpes Macquart. Scromyzoptera Hendel, 1917: 46. Type species S. annulata Hendel. A genus of medium-sized to rather large flies of medium to robust build, having most of the characters of the family Heleomyzidae s.l. as defined by Colless and McAlpine (1970). Head rounded, with convex occiput; fronto-orbital plate not sharply differentiated, short, parallel with eye margin, bearing one or 2 fronto-orbital bristles which are sometimes very fine and hair like; postvertical bristles strong, convergent or Proc. Linn. Soc. N.S.W., 106 (1), (1981) 1982 D.K. McALPINE AND D.S. KENT 35 crossed; uppermost postocular bristle well developed; face uniformly sclerotized, concave in profile, without distinct median carina but often with median line slightly raised, with epistomal margin more or less prominent. Antenna deflexed from articulation between segments 1 and 2; segment 1 short, setulose at least on dorsal margin; segment 2 setulose, with one longer dorsal bristle; segment 3 rounded oval, compressed ; arista inserted before middle of dorsal margin of segment 3; segments 4 and 5 small, not notably swollen; segment 6 long, not much swollen basally, with numerous minute hairs throughout. Prelabrum not prominent, only slightly projecting from general level of peribuccal area; proboscis somewhat elongate, with short labella. Thorax with the following bristles: humeral, 1 + 1 notopleurals, presutural (sometimes absent), supra-alar, postalar, intra-alar, one or 2 dorsocentrals, prescutellar acrostichal (sometimes absent), 2 pairs of long scutellars, inserted apically and dorsolaterally respectively, one sternopleural; propleural and mesopleural bristles absent; scutellum somewhat rounded, with short hairs and pubescence-pruinescence. All femora with strong ventral spines, mostly aligned in the anteroventral and posteroventral series; mid femur with one or more posterior preapical bristles, often in a short, oblique series; tarsi with distal segments depressed (least so in T. swbglabra) ; in males fore coxa, ventral surface of femora and tibiae and ventral part of sternopleuron often with long, fine, dense hairs. Wing without dark markings; costa much weakened and notched but not deeply incised at end of subcosta, not broken or weakened just beyond humeral crossvein, with usually 2 to 4 bristles on section bordering first costal cell, otherwise without differentiated bristles, with a series of short stout spinules dorsally bordering subcostal and marginal cells; subcosta well developed throughout, diverging from vein | distally; anal crossvein with sigmoid curvature, almost transverse; vein 6 long, usually gradually reduced distally to an unpigmented fold which reaches wing margin or is discontinued a short distance from it, sometimes pigmented to margin (T. subglabra) ; vein 7 represented by a usually well defined, dorsally convex crease in membrane which is often pigmented. Abdomen with preabdominal spiracles in pleural membrane; d postabdomen with abbreviated tergite 6; sternites 6 and 7 well developed, displaced to left side, connected to the dorsal sternite 8; cerci distinct, separate. 9 post-abdomen with no particularly elongate segments; segments 6 and 7 with separate tergites and sternites; cerci free. Key to Species of Tapezgaster 1. Prosternum without long hairs, with minute pubescence only ........ 2 Prostscnwiaen wun Momer eu > $3 Ss sc ae oe assess asus Se aes ea 3 2. Vein 6 not reaching wing-margin; presutural bristle present; vibrissa arising from brownish spot, subtended by a second fairly strong bristles (uilbrissaenotiso: in amy Other Species) vr. 2 oe a ee pulverea Vein 6 reaching margin though weakened apically; presutural bristle absent (present in all other species) ; vibrissa not arising from a pigmented spot, subtended only by fine hairs .............. subglabra 3. Propleuron with numerous long, fine hairs near centre; mid and liitaderenitoraanotsmotablysswolllemters a) ses Mele oeesolun ye Meee ne so 64 - 4 Propleuron not haired near centre; other characters variable ....... 6 Proc. Linn. Soc. N.S.W., 106 (1), (1981) 1982 36 SYSTEMATICS OF TA PEIGASTER 4. Mesoscutum entirely pruinescent, not shining, with three broad, well- defined, longitudinal grey bands; fore and mid tibiae with only the apical black band distinct; surstylus unequally bilobed, the lobes strongly: procurved seahiich. aii AS cen SH eae ee tener area paramonovt Mesoscutum without longitudinal bands, shining and sparsely _ pruinescent; all tibiae with a black band at apex and another above middle esunstylasio titer wal Sesh © tern Cle epee een eee ee 5 5. Fore femur browned on no more than distal quarter; thoracic pleura in part sparsely pruinescent, parts of mesopleuron and sternopleuron somewhat shining; male terminalia with surstylus inserted close to cercus, undivided, broad basally, with apical part narrow ..... annulata Fore femur with brown anterodorsal stripe from well before middle to apex; thoracic pleura entirely densely pruinescent, not shining; male terminalia with surstylus inserted a short distance in front of cercus, forked, the posterior lobe broadly obtuse ....... Se Aeriiah digitata 6. Prescutellar acrostichal bristle absent or reduced; mesoscutum tawny with three regular, grey-pruinescent, longitudinal bands; abdominal tergite 9 of male with pair of incurved horn-like processes .... annulipes Prescutellar acrostichal bristle well developed; mesoscutum not marked as above; abdominal tergite 9 of male with at most a sibbositysonveaehside i) fs gil icslekpie dt aaron. aaah des ae ele ike aac a as 7 7. Vein 6 discontinued before reaching wing margin; wing membrane with narrow bare zone in front of vein 7; mid femur stouter than hindtone (viewedifirom above)) #4441) a ieee ama Lee PemTies Vein 6 reaching wing margin though faint distally; wing membrane entirely microtrichose near vein 7; hind femur usually stouter than TMMUGIVONME 4 ee Ena Mya Her coe 2eobl sarc ais: cole Re ys ek tO eh eA ee aces ee Eee 8 8. Mid tibia darkened only at apex; hind tibia with grey pruinescence on most of surface; posterior surface of fore femur with neither depression nor transverse grey-pruinescent band; wing not yellowish basally; postfrons with complete silvery-pruinescent orbital stripe, but without silvery orbital spots; abdominal tergite 9 (epandrium) of maleswith eb bOsibyaonkeachysic cy sie sa lenient an aeeee nigricornis Mid tibia darkened sub-basally as well as apically; hind tibia shining, without obvious pruinescence; posterior surface of fore femur with either depression or transverse grey-pruinescent band just beyond middle; wing yellowish basally; postfrons with silvery-pruinescent orbital spots; abdominal tergite 9 of male without paired PUD DOSTEVES H1.0 75. Stele joviolke Pane se eRe eee sac oe meat cies eRe ee ea 9 9. Mesoscutum pale brownish with a large round blackish spot before and behind transverse suture on each side; fore femur with whitish-pruinescent transverse stripe on posterior surface beyond FagvKO (CG | (eas Wiens Neat MR MeeRecind Sear cael Sea Ue ge STROM. Rare ee cinctipes Mesoscutum differently coloured; fore femur without such stripe .... 10 10. Mesoscutum tawny-orange, with distinct white-pruinescent markings on margins only; fore coxa entirely fulvous-orange, largely pruinescent ON OUCER/SUTPACE 2 Silo Daa se a eee ee argyrospila Mesoscutum deep brown with whitish-pruinescent markings on central part; fore coxa with glossy brown zone on outer surface ... brunnezfrons Proc. Linn. Soc. N.S.W., 106 (1), (1981) 1982 D. K. McALPINE AND D. S. KENT 37 Tapergaster nigricornis (Macquart) n. comb. (Fig. 3) Sctomyza nigricornis Macquart, 1851: 277-278, pl. 25, figs. lla, 11b. Tapergaster marginifrons Bezzi, 1923: 74-75; McKeown, 1942: 228, fig.; Paramonov, 1955: 459-461. N. syn. Bezzi has given a detailed description which is quoted by Paramonov. The species is not very closely related to any other species, and is easily recognized from the characters given in the key. Distribution: Queensland — south-east districts as far north as Bundaberg; New South Wales — very widely distributed; Victoria; Tasmania; South Australia — mainly southern districts; Western Australia — southern districts north to Perth and east to southern Nullarbor. Type material of S. nigricornis: “Tasmanie’ (lectotype d, here designated, PM no. 4/8/47); ‘Akarea, Nouvelle-Zélande’ (probably Akaroa, Banks Peninsula), (paralectotype 9, PM no. 4/46). The paralectotype is not conspecific with the lectotype but is a lauxaniid of the genus Sapromyza in the sense in which that genus is employed in New Zealand. It may well be referable to one of the species included by Harrison (1959) in his revision of this group. After careful consideration and discussion we have decided to designate the Tapezgaster specimen as lectotype. Although this means altering the established name Tapezgaster margznifrons, it removes a potential problem for lauxaniid workers. Type material of T. marginifrons: Blue Mountains, New South Wales, 22.iv.1922 (lectotype d, here designated, right wing missing, AM), anon.; Sydney, New South Wales, 2.vi.1922 (paralectotype 2, much damaged, head glued to label, AM), anon. Other material examined (localities only given). Qld.: Woowonga Range, SW of Bundaberg (ANIC) ; Tibrogaren Creek, via Beerburrum (UQ) ; Dayboro (ANIC) ; Camp Mountain (UQ); Capalaba, near Brisbane (UQ); Rosewood (ANIC) ; Lamington National Park (UQ); Inglewood (UQ); Braeside (UQ); Stanthorpe (UQ). N.S.W.: Boonoo Boonoo, near Tenterfield (ANIC) ; Deepwater, near Glen Innes (AM); Graman (BCRI); Mount Gibraltar (AM, ANIC); Bourke (BCRI) ; near Rylstone (ANIC); Kandos (AM); Mount Coricudgy (AM) ; Derriwong, near Condobolin (BCRI) ; Parkes (ANIC) ; Winburndale Nature Reserve, near Bathurst (ANIC) ; 37 km (23 miles) N of St Albans (BCRI) ; Mount Wilson, Mount Tomah, and Mount York, Blue Mountains (AM); Kurrajong (AM); Ooma Creek, NW of Grenfell (ANIC); Katoomba (AM); Blackheath and Leura (BCRI); Wentworth Falls (AM); Maroota, Gordon, Lane Cove, Bronte, Como and Sutherland, Sydney district (AM) ; Manly Reservoir, near Sydney (ANIC) ; Port Hacking (AM) ; Royal National Park (AM); Jenolan Caves (AM); Kanangra Plateau (AM); Mittagong (BCRI); Moss Vale (BCRI); Wombeyan Caves (ANIC); Hilltop (UQ); Abercrombie River, Bummatoo Forest (AM) ; Marulan (ANIC); Robertson (AM) ; Gerroa (AM); Yanco (AM); Narrandera (BCRI); Leeton (BCRI); Bungendore (AM); Royalla (ANIC) ; Billabong Creek, Wanganella (ANIC) ; Billabong Creek, Conargo (ANIC); near Deniliquin (ANIC); Pilot Hill, Bago Forest, near Batlow (ANIC); Talbingo (ANIC); Clyde Mountain (ANIC); Monga, near Clyde Mountain (ANIC); Broulee (ANIC); Gerogery (ANIC); near Adaminaby (ANIC) ; Nimmitabel (ANIC) ; Brown Mountain, near Nimmitabel (AM, ANIC) ; Pambula (BCRI); Wilson’s Valley, Snowy Mountains (AM); The Creel, Snowy Mountains (ANIC) ; Alpine Creek, Snowy Mountains Highway (ANIC). Vzc: Mount Buffalo (ANIC); Seaford (ANIC); Mill Park, Melbourne (AM); near Woodend (AM); Lorne (ANIC); Mount Buangor, near Beaufort (ANIC); Wedderburn (ANIC) ; Mount William, Grampians (AM); Wannon River, Grampians (AM) ; Proc. Linn. Soc. N.S.W., 106 (1), (1981) 1982 38 SYSTEMATICS OF TA PEIGASTER Figs 1-2. Tapezgaster lutetpennis. 1-2. db genitalia. 1. lateral view. 2. posterior view. Warrnambool (UQ); Dimboola (ANIC). Tas.: Derwent Bridge (UQ); Tyenna River, near Mount Field National Park (AM); Mount Wellington (ANIC). S.4.: Moorlands (ANIC); Mount Lofty Range (ANIC); Prospect (UQ); Second Valley Road, Cape Jervis (ANIC) ; Seal Bay, and Bale’s Bay, Kangaroo Island (AM) ; Cape Borda, Kangaroo Island (ANIC) ; Sleaford Bay (ANIC) ; 8 km SSE of Mount Hope (ANIC); N of Nullarbor Homestead (AM). W.A.: Cocklebiddy (ANIC) ; Junana Rock (ANIC) ; Mount Ragged (ANIC) ; NW of Mount Arid (ANIC) ; Thomas River estuary, Esperance district (ANIC); Gibson (ANIC); E of Ravensthorpe (ANIC) ; Mount Magog, Stirling Range (AM) ; Porongurup National Park (ANIC) ; E of Jewel Cave, Augusta (ANIC) ; Darlington (AM). Tapeigaster cinctipes (Walker) n. comb. (Figs 4-5) Heteromyza cinctipes Walker, 1853: 404. Tapergaster bella Paramonov, 1955: 454-455, N. syn. The description of Paramonov, together with the characters indicated in our key, is adequate for identification. The species is polymorphic in abdominal coloration. A majority of specimens from eastern states, including the holotypes of Walker and Paramonov, have the abdominal tergites fulvous with much of tergite 2 and all of tergites 3 and 4 blackish. Specimens from Bendora (allotype 9 of T. bella), Mount Ainslie (paratype d of T. bella), and Monga (1 G) have the abdominal tergites entirely fulvous. The three specimens from Western Australia (both sexes) are intermediate between the two eastern forms, having the dark coloration reduced to a brown patch on each side of tergites 2 to 4, smallest on tergite 2. No difference in the male genitalia can be discerned between the two eastern forms and the difference in shape of the surstylus in the one western male examined is so slight that a significant difference between the populations may not be indicated. Proc. Linn. Soc. N.S.W., 106 (1), (1981) 1982 D.K. McALPINE AND D. S. KENT 39 SO postabdomen. Surstylus broadly ovate, broadly gibbous on posterior margin from a little beyond base, most strongly haired on anterior margin, with shallow, oblique apical cleft producing a blunt tooth on inner surface; sides of hypandrium without series of hairs. Distribution: southern New South Wales; Australian Capital Territory; Victoria; Tasmania; southern Western Australia. Holotype 9 of H. cinctipes: “Van Diemen’s Land’ (Tasmania) (BM), anon. Type materzal of T. bella: Blundell’s clearing, near Canberra, 6.iv.1947 (holotype d, paratype 0, ANIC), S. J. Paramonov; Bendoora, A.C.T., 15.iv.1953 (allotype 2, ANIC), S. J. Paramonov; Mount Ainslie, near Canberra, i1i.1950 (paratype 6, ANIC), A. Floyd; National Park (i.e. Mount Field National Park), Tasmania, xii.1922 (paratype 0d, ANIC), A. L. Tonnoir; Beverley, W.A., 1913 (paratypes 1d, 12, ANIC), ‘D.B.’ Other material examined (localities only given). N.S.W.: Murrumbateman (ANIC); Monga, near Braidwood (ANIC); Broulee (ANIC, AM). A.C.T.: Black Mountain, Canberra (ANIC). Vic.: Mount Macedon (AM). W.A.: Nedlands (ANIC). Tapeigaster argyrospila Bezzi (Fig. 6) Tapeigaster argyrospila Bezzi, 1923: 77-78; Paramonov, 1955: 458-459. The species has been described in some detail by Bezzi and is correctly interpreted by Paramonov. The surstylus is, as usual, quite distinctive. Abdomen of G. Sternite 5 not divided into 2 plates, but rather weakly sclerotized medially; surstylus three-lobed; central lobe rather elongate, tapering; anterior lobe broadly rounded, with short hairs towards margin, not deeply divided off from central lobe; posterior lobe more slender and deeply set off than anterior lobe, much shorter than central lobe. Distribution: Queensland — southern districts only; New South Wales — coast to western plains; Victoria; South Australia. Holotype 3 : Blue Mountains, 11.iii.1922 (AM, ex Health Dept. collection) , anon. Other material examined (localities only given). Qld.: Mount Mowbullan, Bunya Mountains (AM) ; Emuvale (UQ) ; Stanthorpe (UQ). N.S.W.: 11 km (7 miles) E of Mendooran (AM); Brummagem Creek, near Dubbo (AM); Bogan River (AM) ; Comark, near Oberon (BCRI); Leura Falls, Blue Mountains (ANIC) ; Thornleigh, near Sydney (BCRI) ; Upper Middle Harbour, Sydney (AM); Mittagong (BCRI) ; Leeton (AM) ; Billabong Creek, near Wanganella (ANIC). Vzc. : Upper Main Creek, Mornington Peninsula (AM); Hobson’s Bay, Melbourne (AM); near Woodend (AM) ; Cobrum (ANIC). S.A.: Sleaford Bay, Port Lincoln (ANIC). Tapeigaster brunnezfrons Malloch (Fig. 7) Tapeigaster brunnezfrons Malloch, 1927: 16. Tapeigaster vernalis Paramonov, 1955: 457. N. syn. Descriptions have been given by Malloch and Paramonov. These are brief, but as the species resembles T. argyrospila very closely, except in male genitalia, a detailed redescription is not necessary. Although the type of T. brunnezfrons was undoubtedly darker-coloured when Malloch described it the year after the collection date, its coloration is now quite in Proc. Linn. Soc. N.S.W., 106 (1), (1981) 1982 40 SYSTEMATICS OF T4 PEIGASTER Figs 3-10. 3-8. 3 surstyli. 3. Tapecgaster nigricornis. 4. T. cinctipes (Mt. Ainslie, A.C.T.).5. T. cinctipes (Beverley, W.A.). 6. T. argyrospila. 7. T. brunnezfrons. 8. T. annulipes. 9-10. T. paramonow. 9. d genitalia. 10. db 5th sternite. agreement with material of the species subsequently named vernalis by Paramonov. Though, from the descriptions, the small size of the fronto-orbital bristles and the single pair of dorsocentral bristles in T. brunnezfrons might seem to separate it from T. vernalis, these characters are variable in the available series and asymmetrical combinations of the bristle characters occur. Proc. Linn. Soc. N.S.W.,106 (1), (1981) 1982 D. K. McALPINE AND D. S. KENT 4] Abdomen of 6. Tergite 5 undivided, but with median part of posterior margin weakly sclerotized ; surstylus not lobed as in T. argyrospila, subtriangular, with subacute apex very slightly inclined forwards and short posterobasal projection, on inner surface a strongly developed blade running from posterobasal projection towards, but not reaching, apex. Distribution: New South Wales — coast, tablelands and south-western plains; Australian Capital Territory; Victoria; Tasmania. Holotype 2 of T. brunnetfrons: Newcastle, N.S.W. 7.iv.1926 (USNM), anon. Type material of T. vernalis. Lectotype d, here designated: Wahroonga, near Sydney, 31.x.1926 (SPHTM), anon.; same data (paralectotype 9, SPHTM). These specimens were marked “Typus 6’ and ‘Typus 9’ by Paramonov, and there is no indication of a holotype in his description. Other material examined (localities only given). N.S.W.: near Spring Creek, Ebor- Point Lookout Road (AM); Mooney Mooney Creek, near Gosford (AM); Mount Wilson, Blue Mountains (AM); Mount Tomah, Blue Mountains (ANIC); Katoomba (AM); Wentworth Falls (AM) ; Glenorie (BCRI) ; Gordon, near Sydney (AM); Royal National Park (AM); Manar, near Braidwood (AM); Monga, near Braidwood (ANIC); Batehaven, near Bateman’s Bay (ANIC); Billabong Creek, near Wanganella (ANIC). A.C.T.: Black Mountain, Canberra (ANIC). Tas.: Hellyer Gorge (ANIC) ; Huon-Picton Junction (ANIC). Comparative notes. T. brunnezfrons is very closely related to T. argyrospila, there being few external characters separating them, though the former species is darker. The surstylus of the male shows a quite remarkable degree of difference as described above. Tapeigaster lutecpennis Bezzi (Figs 1-2) Tapeigaster lutetpennis Bezzi, 1923: 76-77. Tapeigaster taylort Malloch, 1935: 94-95. N. syn. A detailed description has been given by Bezzi, but the male postabdominal characters have not been recorded. Abdomen of G : tergite 9 produced posteriorly into two short rounded protuberances ; cerci concealed in a roughly oval membranous cavity beneath and between the protuberances; surstylus very complex (Figs 1-2), with tubercle on basal region and two divergent lobes; hypandrium with numerous rather long hairs medially. Distributzon: Queensland — south-east; New South Wales — coast and warmer eastern margins of tablelands. This species shows a greater degree of restriction to warmer areas than any other in the genus. The record from Wentworth Falls and probably that from Tallong indicate localities in nearby valleys at much lower elevations than the towns named. Holotype 3 of T. luteipennis: Eccleston, near Dungog, N.S.W., 1.i1i.1921 (not 1922 as given by Bezzi) (AM, ex Health Dept. collection) , anon. Holotype 6 of T. taylort: Tallong, near Marulan, N.S.W., no date (SPHTM), F. H. Taylor. Other material examined. Qlid.: Bundaberg (UQ); Montville (UQ), Bunya Mountains (UQ); Brisbane (UQ, ANIC); Tamborine Mountain (AM) ; Cunningham’s Gap, near Maryvale (UQ); Lamington National Park (UQ). N.S.W.: Legume (UQ); Tooloom (UQ); Ulong, East Dorrigo district (AM) ; Dorrigo (ANIC) ; Wentworth Falls, Blue Mountains (AM) ; Heathcote (AM) ; Royal National Park (AM) ; Otford (AM). Proc. Linn. Soc. N.S.W., 106 (1), (1981) 1982 42 SYSTEMATICS OF TAPEIGASTER Notes. It is clear from examination of types and additional material that T. lutecpennis and T. taylor? are conspecific. There is a little variation in the amount of white pruinescence on the mesoscutum and in the brown markings on the abdomen. The only other species with paired gibbosities or tubercles on tergite 9 of the male are T. annulipes and T. nigricornis. In T. annulzpes the tubercles are more elongate and horn-like. In T. nzgricornzs the paired gibbosities are less prominent than in T. lutecpennis and the cerci are much more prominent. Tapezgaster annulipes Macquart (Fig. 8) Tapezgaster annulipes Macquart, 1847: 87, pl. 6, figs. 1, la-le; Bezzi, 1923: 75-76; Colless and McAlpine, 1970, fig. 34.30A. Dryomyza cingulipes Walker, 1857: 220, synonymized by Bezzi, 1923: 75. A useful description is given by Bezzi. It may be added that the propleuron is not haired just below the humeral callus, and vein 6 does not quite reach the wing-margin though longer than in T. dzgztata and allied species. Distribution: Queensland — southern border districts; New South Wales — mainly coast and tableland districts; Victoria; Tasmania; South Australia — Lofty Ranges and Kangaroo Island. The species is very common and widely distributed, being found in forested areas and sometimes inner urban areas. It is apparently absent from some of the warmer, drier districts inhabited by T. nzgricornzs and T. argyrospila. Type material of T. annulipes: no locality on label but Macquart gives ‘Nouvelle- Hollande’, ‘Tapeigaster annulipes. d @. n. g. n. sp. Macq’ (in Macquart’s handwriting) and on separate disc “TYP’ (lectotype G here designated, OX). Macquart probably had 2 specimens in his type series as he gives a single measurement for each sex. The lectotype is the only specimen which we can identify with certainty as belonging to the type series, and its designation as such is essential to preserve current usage of the name T. annulzpes. Also above the cabinet label T. annulzpes in the Bigot collection at Oxford, there are two other flies at least one of which should be a paralectotype of T. annulipes though neither is conspecific with the lectotype. The first is a female pyrgotid with a “IYP’ disc on the pin which could indicate type material if its authenticity could be established. The second specimen is a male Tapeigaster, probably of T. argyrospila but very mouldy. Some credence is given to its possible type status, because Macquart’s pl. 6, fig. 1 (whole insect) , looks more like T. argyrospila than T. annulipes in some respects. On the other hand Macquart’s illustrations are so exceedingly inaccurate that they scarcely provide grounds for safe conclusions. Neotype of D. cingulipes: We designate the lectotype of Tapezgaster annulipes Macquart as the neotype of Dryomyza cingulipes Walker. In so doing we stabilize the otherwise doubtful synonymy first given by Bezzi and repeated by Paramonov (1955). Fixing the identity of D. cingulipes in this way will prevent the use of the epithet for other species of Tapezgaster here referred to by names of later date. In accordance with Article 75(C) of International Code of Zoological Nomenclature we make the following comments. (1) The species is as described by Bezzi (1923) and is distinguished as given above in our key to species of Tapezgaster. (2) The label data are those on the lectotype of T. annulipes given above. (3) Walker’s types of Australian Diptera are, so far as we know, housed in the British Museum (Natural History) ; the Hope Department of Entomology, University Museum, Oxford; and the National Museum of Victoria, Melbourne. McAlpine has been unable to find type material of D. congulipes at the British Museum or Oxford and B. H. Cogan confirms Proc. Linn. Soc. N.S.W., 106 (1), (1981) 1982 D. K. McALPINE AND D.S. KENT 43 its apparent absence from BM. A. Neboiss informs us that there is no type material of D. cingulipes at NMV. We therefore regard this original type material as probably lost. (4) The original description of D. cingulipes provides the only available information on the type material of that species. It almost certainly refers to a species of Tapezgaster and agrees fairly well with T. annulzpes though it also resembles other species, particularly the much less common T. pulverea. (5) The type locality for T. annulzpes is given as ‘Nouvelle Hollande’, that for D. cingulipes as ‘New South Wales’. As the term New South Wales at that time applied to most of the eastern half of mainland Australia (or New Holland), there is no evidence that the two type localities were remote from one another. (6) The neotype belongs to the University Museum, Oxford, England. Other material examined (localities only given). Qld.: Lamington National Park (UQ) ; Stanthorpe (UQ). N.S.W.: 32 km (20 miles) , Glen Innes to Grafton highway (ANIC) ; Mt. Kaputar, near Narrabri (AM, ANIC); New England National Park (ANIC) ; near Spring Creek, Ebor— Pt. Lookout road (AM); Point Lookout, near Ebor (AM) ; Wright’s Lookout, New England National Park (AM) ; near Mt Dagola, Warrumbungles (AM); Nundle State Forest (ANIC); Tomalla (AM) ; Coachwood Gully, 21.7 km SE Threeways (AM); Mt Irvine (AM) ; Mt Wilson, Blue Mts (AM) ; Mt Tomah, Blue Mts (AM, ANIC) ; 8 km NW of Kurrajong-Bell road (ANIC) ; near Mt Banks, Blue Mts (AM); Mt Boyce, Blue Mts (AM); Leura Falls (ANIC) ; Govett’s Leap, Blue Mts (AM); Katoomba (AM); Wentworth Falls, Blue Mts (AM) ; Sassafras Gully, Springwood (AM); Jenolan (AM); 6.4 km (4 miles) N of Jenolan Caves (AM); Mt Queen Pin, Kanangra road, Boyd Plateau (AM); Kanangra (AM); Boyd River Crossing, Kanangra road (AM); Budthingaroo Ck, Kanangra-Boyd National Park (AM) ; Cowan (AM); Milson Island (AM) ; Sydney (AM, UQ); Gordon (AM); Mosman (ANIC); Bexley (AM); Northwood (AM) ; Royal National Park (AM, ANIC); Otford (AM); Cataract Ck, Bulli (ANIC); Mt Keira, near Wollongong (AM) ; Colo Vale (ANIC) ; Mt Gibraltar, Bowral (ANIC) ; 6 km SE of Robertson (AM) ; Fitzroy Falls (AM); Beaumont, 6 km SE of Kangaroo Valley (AM); Seven Mile Beach State Park, Gerroa (AM); Jervis Bay (ANIC) ; Clyde Mountain, near Braidwood (ANIC); Mongarlowe River, Clyde Mountain (ANIC) ; 9 km SE of Bateman’s Bay (ANIC) ; Broulee (ANIC) ; Rutherford Creek, Brown Mtn (AM, ANIC); Brown Mtn, Bega district (ANIC) ; 6.4 km (4 miles) E of Nimmitabel (ANIC); Tumut (ANIC); Pilot Hill, Bago Forest, Batlow (ANIC) ; Kiandra, Alpine Ck (ANIC); Eucumbene-Lookout, Snowy Mts (ANIC); 19.2 km (12 miles) NW of Adaminaby (ANIC); The Creel, Kosciusko (AM, ANIC) ; Wilson’s Valley, Snowy Mts (AM) ; Sawpit Creek, Snowy Mts (AM, ANIC); 3 km S Wragge’s Creek, Kosciusko National Park (ANIC). A.C.T.: Black Mountain (ANIC); Mt Coree (ANIC); Bull’s Head (AM); Uriarra State Forest (ANIC) ; Paddy’s River (ANIC); Lee’s Creek, Brindabella Range (AM); Condor Creek, Brindabella (AM); Brindabella Range (AM). Vzc.: Dynamite Creek, Bonang Highway (ANIC); 21 km (13 miles) W of Matlock (AM); Cement Creek, near Warburton (AM); Warburton (ANIC); Sherbrooke Forest, near Kallista (AM) ; Belgrave (AM); Hobson’s Bay, Melbourne (AM); Mt Macedon (AM); Daylesford (ANIC); Mt Buangor, NE of Beaufort (ANIC). Tas.: 5 km (3 miles) S of Oonah, Waratah Highway (AM); Hellyer Gorge (AM, ANIC); 19.2 km (12 miles) S of Wilmot (AM); Mt Barrow, via Launceston (UQ); 5 km (3 miles) E of Waratah (ANIC); Needles, near Deloraine (AM); Marakoopa Caves, near Mole Creek (AM); 25.6 km (16 miles) NE of Cradle Mt (AM); Breona (UQ); Lake St Clair (AM); Derwent Bridge (ANIC); Franklin River Crossing, Lyell Highway (AM) ; National Park (UQ); Hobart (UQ); Mount Wellington (ANIC). S.4.: Mt Burr Proc. Linn. Soc. N.S.W., 106 (1), (1981) 1982 44 SYSTEMATICS OF TAPEIGASTER (ANIC); Glen Osmond (AM); Waterfall Gully, Burnside (AM); Engelbrook Reserve, near Bridgewater (AM); National Park, Mt Lofty Range (ANIC) ; Upper Ravine des Casoars, Kangaroo Island (AM). Tapeigaster paramonovwi n. sp. (Figs 9-10) d Q. Closely related to T. dégztata and agreeing with description given for that species except as indicated below. Coloration. Postfrons orange-tawny, often deeper tawny-brown laterally, sometimes darkened anteriorly, with whitish pruinescence along orbital margins and to lesser extent on ocellar triangle. Mesoscutum tawny-brown with pale grey, yellowish-edged median stripe from anterior extremity almost to scutellar suture where it is narrowed to a point, and with broad yellowish grey lateral margins; scutellum tawny-brown on central and anterior part of dorsal surface, broadly pale tawny-buff on lateral and posterior margins; pleura with deeper tawny ground colour than in T. digztata, and with even thicker covering of whitish pruinescence. Fore femur with brown anterodorsal longitudinal stripe as in T. dzgitata but broader and more diffuse; fore tibia with distal brown zone occupying nearly half length of tibia, without sub-basal dark band; mid tibia browned apically, with sub-basal dark band at most indistinct; hind tibia coloured approximately as in T. dzgztata. Head with one or 2 fronto-orbital bristles, in the latter case anterior bristle very short. Thorax. Femora of 6 with very numerous, rather short, fine ventral hairs. Abdomen. Sternite 5 of db with each of paired lobes very rounded posteromedially. d postabdomen: surstylus with 2 divergent lobes, both procurved and obtuse, the anterior one larger, also with anterior basal gibbosity which bears much longer hairs than those on lobes; hypandrium with a series of long hairs on each side, the longest ones about as long as surstylus; cercus with anterior subapical tubercle. Dimensions: total length, d 5.2-6.8 mm, 2 4.9-6.5 mm; length of thorax, 5 2.5-3.3 mm, 2 2.7-3.1 mm; length of wing, d 5.6-6.9 mm, 9 6.2-7.5 mm. Distributzon: New South Wales (tablelands) ; Victoria; South Australia; Western Australia (south-west) . Holotype 3: Leather Barrel Creek, Snowy Mountains, 11.11.1979 (AM), B.J.D. and D.K.M. Paratypes. N.S.W.: Mount Kaputar, near Narrabri, xi.1964 (4d, AM), D.K.M.; Mount Wilson, Blue Mountains, xii.1956 (1 9, AM), D.K.M.; 8 km (5 miles) S of Mount Wilson, iv.1971 (1d, AM), D.K.M.; Mount Boyce, Blue Mountains, iii.1963 (1 d, AM), D.K.M.; Mount York, Blue Mountains, x.1960 (1 9, AM), D.K.M.; Falls Creek, near Nowra, x1i.1926 (1 Q, AM), B. Bertram; Clyde Mountain (west slope), near Braidwood, v.1965 (1 9, ANIC), D.H.C.; The Creel, near Mount Kosciusko, xi.1961 (5 d, 29, ANIC), D.H.C.; Wilson’s Valley, Snowy Mountains, 11.1979 (1 d, AM), B.J.D. and D.K.M.; 19.2 km (12 miles) NW of Adaminaby, xi.1961 (1d, ANIC), D.H.C.; Geehi River, xi.1961 (1d, ANIC), D.H.C. 4.C.T.: Black Mountain, Canberra, viii.1968 (1 0, ANIC), I.F.C.; Mount Coree, iv.1968 (2 d, ANIC, 1d, 1 9, CNC), D.H.C., J. W. Boyes; Bendora, ii.1950 (1 2, ANIC), K.R.N.; Cotter River, x.1956 (1 d, ANIC), Z.R.L.; Lee’s Springs, iv.1958 (1 9, ANIC), Z.R.L.; Condor Creek, Brindabella Range, iv.1972 (1 9, AM), D.K.M. Other material examined. Vic.: Frenchman’s Gap, near Wood's Point, iv.1963 (2 ©), AM), D.K.M.; Daylesford, viii.1968 (1 9, ANIC), N. Dobrotworsky; Wannon River, near Jimmy’s Creek, Grampians, xii.1977 (1 9, AM), M. A. Schneider and D.K.M. Tas.: Cradle Mountain, i.1923 (19, ANIC), A. L. Tonnoir. S.A. : Second Proc. Linn. Soc. N.S.W., 106 (1), (1981) 1982 D. K. McALPINE AND D. S. KENT 45 Valley Road, Cape Jervis, x.1975 (1d, ANIC), Z.R.L. W.A.: Darlington, ix-xii.1964 (1d, AM), G. L. Bush; 17.6 km (11 miles) SW of Collie, x.1964 (2d, 292, AM), G. L. Bush; 24 km (15 miles) S of Mumballup, x.1964 (1 d, AM), G. L. Bush; Mount Toolbrunup, Stirling Ranges (10, AM, 1d, WAM), G.A.H.; Warren River, 9.6 km (6 miles) SE of Pemberton, 1.1971, (16, AM, 1d6, WAM), G.A.H. and H. Hughes; Pemberton, xii.1936 (1d, ANIC), K.R.N.; Channybearup, near Pemberton, x.1970 (12, ANIC), D.H.C.; Pimelia, near Pemberton, x.1970 (12, ANIC), D.H.C.; 14.4 km (9 miles) W of Pemberton, x.1970 (1 Q, ANIC), D.H.C.; Porongurup National Park, x.1970 (20, ANIC), D.H.C.; Mount Chudalup, S of Northcliffe, x.1970 (19, ANIC), D.H.C.; Rest Point, Walpole, x.1970 (1 9, ANIC), D.H.C.; Nornalup National Park, x.1970 (2d, 19, ANIC), D.H.C. Comparative notes. T. paramonovwt is related to T. annulata and especially T. digztata but differs from these in having a median grey stripe on the mesoscutum much as in T. annulipes. It differs from T. annulzpes in the presence of hairs on the propleuron and absence of paired horn-like tubercles on the epandrium. The shape of the surstylus is distinct from all related species. Paramonov (1955) referred specimens of this species doubtfully to T. fulva. Tapeigaster digitata n. sp. (Figs 13-17) 6 2. Coloration. Head dull fulvous; cheek and face paler; ocellar spot black. Antenna black to blackish brown. Palpus pale fulvous; labella brown. Mesoscutum brownish fulvous to tawny brown, entirely thinly greyish- to brownish-pruinescent, somewhat shining, with pair of ill-defined whitish-pruinescent marks on anterior margin; pleura pale fulvous to light brown, entirely densely whitish-pruinescent and nowhere shining. Legs fulvous; femora blackish apically; fore femur with rather broad brown to black longitudinal anterodorsal stripe from about basal quarter to apex; tibiae each with brown to blackish ring at apex and before middle; tarsi variably browned distally. Wing hyaline, tinged with greyish brown; haltere dull yellowish. Abdomen brownish, often paler at base and apex, but variable, more densely pruinescent in d than in 9, largely shining in latter sex. Head with usually one distinct fronto-orbital ; vibrissa not duplicated. Thorax. Propleuron with fine rather long hairs below humeral callus; presutural and 2 pairs of dorsocentral bristles present; prescutellar acrostichal bristle absent. Fore coxae usually with pale hairs only, sometimes with one or 2 incipient bristles; fore femur moderately stout, mid and hind femora relatively slender for genus and of almost equal thickness; ventral spines of femora restricted to anteroventral and posteroventral series but usually only the former present on hind femur; femora and tibiae without long, dense ventral hairs in either sex. Wing with veins 3 and 4 apically only very slightly convergent; vein 6 discontinued at a short distance from margin. Abdomen. Sternite 5 of d divided medially into 2 plates, each of which is rather rounded posteriorly with only slight indication of posteromedian angle. d postabdomen with left spiracle 6 situated within lateral extremity of tergite; left spiracle 7 situated higher up behind tergite 6; right spiracles 6 and 7 close together and horizontally aligned in membrane; sternite 6 situated on lower part of left side, with a narrow strip extending round ventral surface, connected to sternite 7 by a short, narrow isthmus; sternite 7 more dorsally placed on left side, broadly continuous with the large, dorsal sternite 8 at its dorsal extremity; epandrium without tubercles or gibbosities, with well-developed free cerci and one pair of basally articulated surstyli a short distance in front of them, surstylus forked into 2 subequal lobes, anterior lobe Proc. Linn. Soc. N.S.W., 106 (1), (1981) 1982 46 SYSTEMATICS OF TAPEIGASTER 12 Figs 11-16. 11-12. Tapecgaster annulata. 11. 3 genitalia. 12. d 5th sternite. 13-16. T. digitata. 13. db genitalia. 14. d 5th sternite. 15. hypandrium. 16. wing of holotype. obtuse, straight, posterior lobe more broadly obtuse slightly posteriorly curved; hypandrium broad, with a series of very short hairs on each side, with pair of slender anterior extensions which join together on median line to enclose a subcircular membranous area, with a broad horizontal plate posteriorly on each side of aedeagus, without distinct gonites; a pair of sclerites in the form of well-developed plates Proc. Linn. Soc. N.S.W., 106 (1), (1981) 1982 D.K. McALPINE AND D.S. KENT 47 between bases of surstyli and posterior to hypandrium, embracing bases of cerci and connected to each side of the rather small median transverse posthypandrial plate ; aedeagus short, stout, curved forwards, with posterior surface membranous, with a sclerotized skeletal strip on each side, and complex apical part. Dimensions: total length, 5 4.8-6.4 mm, 2 4.9-6.1 mm; length of thorax, d 2.4-3.1 mm; length of wing, d 6.0-8.0 mm, 2 7.0-7.8 mm. Distribution: New South Wales (principally tablelands); Australian Capital Territory; Victoria; Tasmania; South Australia (Lofty Ranges and Kangaroo Island). Holotype 6 : below Govett’s Leap, Blue Mountains, 7.x.1956 (AM), D.K.M. Paratypes. N.S.W.: Mount Kaputar, near Narrabri, xi.1964 (1 d, AM), D.K.M.; Wright’s Lookout, New England National Park, iii-iv.1961 (6¢,592, AM, 3d, 39, BM), D.K.M.; Point Lookout, New England National Park, x.1962 (2 3, ANIC), D.H.C.; Barrington Tops via Salisbury, xii.1965 (3 d, 1 9, UQ), B. Cantrell; Tubrabucca, near Barrington Tops, x.1956 (1 3,192, AM),D.K.M.; Mount Wilson, Blue Mountains, 1.iii.iv.vi.x.xi.xii.1956-1977 (12d, 149, AM, 2d, 39, CNC,24, 3 Q, USNM), G.D., M. A. Schneider, D.K.M.; below Govett’s Leap, ix.1957 (2 6, AM), D.K.M.; Katoomba, xii.1956 (1 d, AM), D.K.M.; Leura Falls, Blue Mountains, 1.1973 (1 6, ANIC), D.H.C.; Wentworth Falls, Blue Mountains, x1.xii.1956-59 (4 d, AM), D.K.M.; Woodford, Blue Mountains, xi.1925 (1 4, ANIC), I. M. Mackerras; Royal National Park, x.1956-65 (20, AM), D.K.M.; Bola Creek, Royal National Park, ix.1961 (1 9, ANIC), D.H.C.; Otford, Mlawarra District, x.1957 (1d, 19, AM). D.K.M.; Mount Gibraltar, near Bowral 111.1975 (3 Q, ANIC), Z.R.L.; Lake George, xii.1950 (2 0, ANIC), K.R.N.; Clyde Mountain, near Braidwood, ii.v.x.1960-65 (1d, AM, 5 d, ANIC), S. J. Paramonov, D.H.C., D.K.M.; 5-8 km (3-5 miles) S of Monga, near Braidwood, v.1968 (3d, 32, ANIC), D.H.C. and Z.R.L.; Mount Jagungal, Snowy Mountains, 11.1951 (1 3d, ANIC), L. Pryor; The Creel, Snowy Mountains, xi.1961 (2d, 22, ANIC), D.H.C.; Wilson’s Valley, Snowy Mountains, 11.1963-1979 (1 d, 1 2, AM), D.K.M. and Bye: Charlotte Pass, Snowy Mountains, 11.1963 (19, AM), D.K.M.; Leather Barrel Creek, Snowy Mountains, ii.1979 (8d, 82, AM), B.J.D. and D.K.M.; Rutherford Creek, Brown Mountain, near Nimmitabel, xi.1974 (6d, 39, AM), G.D. A.C.T.: Black Mountain, Canberra, iv.x.1979-80 (1 d, 1 9, ANIC), D.H.C.; Blundell’s, near Canberra, x.1930 (1d, ANIC), A. L. Tonnoir; Uriarra State Forest, iv.xi.1960-61 (1 6, 29, ANIC), D.H.C.; Bendora, 11.1950 (1 d, ANIC), K.R.N.; Condor Creek, Brindabella Range, iv.1972 (1 d, AM), D.K.M.; ‘Piccadilly Circus’, Brindabella Range, iv.1972 (69, AM), D.K.M.; Bull’s Head, Brindabella Range, iv.xi.1968 (3 3,12, ANIC), D.H.C. Other material examined. Vic.: Mount Beauty, x.1961 (3d, 12, ANIC), D.H.C.; Dynamite Creek, Bonang Highway, x.1961 (3d, ANIC), D.H.C.; Bonang Highway, Bendoc Road Junction, xi.1976 (1 6, ANIC), D.H.C. and Z.R.L.; Nowa Nowa, x.1961 (1 d, ANIC), D.H.C.; Noojee, xi.1964 (1 2, ANIC), D. E. Havenstein; Mount Baw Baw, near Tanjilbren, 111.1964 (1 @, AM), G. L. Bush; Frenchman’s Gap near Woods Point, iv.1963 (32, AM), D.K.M.; 20.8 km (13 miles) W of Matlock, iv.1963 (1 2, AM), D.K.M.; Mount Dom Dom (Black Spur), near Healesville, x.1961 (3d, 39%, ANIC), D.H.C.; 8 km SW of Narbethong, xii.1979 (1 3, ANIC), K.R.N.; Cement Creek, near Warburton, iv.1963 (1 29, AM), D.K.M.; Warburton, iv.1963 (2 3, 22, AM), D.K.M.; Sherbrook Forest, i.1966 (2d, AM), D.K.M.; Belgrave, 1.1966 (1 9, AM), D.K.M.; near Melbourne, 11.1931 (1 d, AM), A. Musgrave; Hobson’s Bay, Melbourne, xii.1922 (1 6, AM), A.F.B. Hull; Mount Macedon, near Woodend, xii.1931 (5 d, 3 2, AM), A. Musgrave; Beech Forest, Proc. Linn. Soc. N.S.W., 106 (1), (1981) 1982 48 SYSTEMATICS OF TAPEIGASTER i.1967 (12, ANIC), Z.R.L.; Otway Range, 11 km W of Apollo Bay, xii.1978 (1d, AM), G.D.; Mount William, Grampians, x.1977 (20,39, AM), D.K.M. and M. A. Schneider. Tas.: Mount Barrow, near Launceston, 1.1960 (3 d, AM), D.K.M.; Western Tiers, Lake Highway, 1.1960 (26, AM), D.K.M.; Marakoopa Caves, near Mole Creek, 1.1960 (30,39, AM), D.K.M.; 19.2 km (12 miles) S of Wilmot, i.1960 (3d, AM), D.K.M.; 3.2km (2 miles) S of Oonah, Waratah Highway, 1.1960 (3d, 7 Q, AM), D.K.M.; Hellyer Gorge, Waratah Highway, i.1960 (4¢0,49, AM, 24,192, ANIC), D.K.M., E. F. Riek; Meredith River, 19.2 km (12 miles) from Corinna, 1.1954 (12, ANIC), T. G. Campbell; 11.2 km (7 miles) W of Rosebery, ii.1963 (1 3d, 22, ANIC), I.F.C. and M.S.U.; Lake Saint Clair, 1.1960 (4 d, 3 2, AM), D.K.M.; Franklin River crossing, Lyell Highway, 1.1960 (13d, 5 Q, AM), D.K.M.; near Russell Falls, Mount Field National Park, 1.11.1960-63 (1 2, AM, 2d, 1 @, ANIC), D.K.M., D.H.C.; Mount Wellington, ii.1963 (2 5d, 6 9, ANIC), D.H.C.; Cambridge, x.1965 (10, ANIC), K. L. Taylor; Myrtle Gully, iii.1935 (12, ANIC), W. Rafferty. S.4.: Engelbrook Reserve, near Bridgewater, iv.1967 (1d, 192, AM), D.K.M.; Ravine des Casoars, Kangaroo Island, xi.xii.1977 (5 d, AM), M. A. Schneider and D.K.M. Comparative notes. T. digttata forms with T. annulata and T. paramonow a group of three closely related species characterized by the haired propleuron, absence of strong thickening of the mid and hind femora, and incomplete vein 6. T. digitata differs from T. annulata in its more densely pruinescent thorax, which generally has a less definite orange hue, in having the dark distal zone of the fore femur extended basad as an anterodorsal stripe, in having the lobes of sternite 5 much blunter posteriorly, the lateral hairs of the hypandrium much shorter, and the surstylus not distinctly forked. For comparison with T. paramonow see under that species. Tapergaster annulata (Hendel), n. comb. (Figs 11-12) Scromyzoptera annulata Hendel, 1917: 47. Tapeigaster fulva Malloch, 1926: 553. N. syn. To Malloch’s description we add the following: Coloration generally of a brighter orange tone than in T. dzgztata. Thoracic pleura a little less thickly whitish-pruinescent than in that species with mesopleuron somewhat shining. Fore femur fulvous, dark brown to black only on apical quarter or less. Thorax. Propleuron with fine hairs below humeral callus; presutural bristle present ; prescutellar acrostichal bristle absent. S postabdomen as described for T. digztata with the following notable differences: lobes of sternite 5 more narrowly produced and subacute; surstylus inserted very close to cercus, not forked, very broad basally, contracted into a narrow, straight, posteriorly directed distal part with attenuated apex; hypandrium with a series of conspicuously long hairs, almost as long as surstylus, on each side. Distribution: Queensland — extreme south-east; New South Wales — coast and wetter parts of tablelands districts; recorded in error by Paramonov (1955) from Australian Capital Territory, Victoria, and Tasmania. Type material examined : ‘N. Holl. 878° (i.e. Australia) no other data (holotype d of Sciomyzoptera annulata Hendel, WM); Botany Bay, N.S.W., no date (holotype 6, paratype G, latter damaged, of T. fulva Malloch, USNM), H. Petersen. Other material examined (localities only given). Q/d.: Binna Burra, Lamington National Park (AM). N.S.W.: Mount Gibraltar National Park, 102.4 km (64 miles) W of Grafton (AM); Ulong, East Dorrigo district (AM); Dorrigo (ANIC); New Proc. LINN. Soc. N.S.W., 106 (1), (1981) 1982 D.K. McALPINE AND D.S. KENT 49 Figs 17-21. 17. Tapeigaster digitata, head of holotype. 18-21. T. swbglabra. 18. head of holotype. 19. wing of holotype. 20. db genitalia. 21. hypandrium. England National Park (AM, ANIC, UQ); Upper Allyn, near Gresford (ANIC) ; Wallaroo State Forest, near Karuah (ANIC); 46.7 km (29 miles) NW of Putty (AM); Mount Wilson, Wentworth Falls, and Springwood, Blue Mountains (AM) ; Mooney Mooney Creek, near Gosford (AM); Middle Creek, near Narrabeen (ANIC); Bronte, near Sydney (AM); Royal National Park (AM, ANIC); Otford, Proc. Linn. Soc. N.S.W., 106 (1), (1981) 1982 50 SYSTEMATICS OF TAPEIGASTER Illawarra district (AM, ANIC) ; Cataract Creek, Bulli (ANIC) ; Minnamurra Falls, near Kiama (AM); Kangaroo Valley (ANIC); Clyde Mountain, near Braidwood (ANIC). Paramonov (1955) misidentified T. fulva, most of his specimens being referable to T. digitata and one to T. subglabra. Bezzi placed S. annulata as a synonym of T. annulipes but examination of the holotype shows this to be incorrect.. T. annulata is most closely related to T. dzgztata, q.v. for comparison. Tapergaster pulverea n. sp. (Figs 22-23) 3 2. Coloration. Head fulvous to tawny; cheek and face much paler; ocelli surrounded by a rather large dark brown spot; orbital margins of postfrons with creamy-white pruinescence; vertex with V-shaped creamy-white pruinescent mark pointing posteriorly visible from some angles, its arms passing to each side of postvertical pair of bristles; upper occiput with brownish patch of varying intensity on each side; extremity of vibrissal angle brown. Antenna dark brown; base of segment 3 reddish brown. Palpus pale fulvous; labella brown to tawny. Thorax with tawny ground colour; mesoscutum with pair of whitish-pruinescent paramedian longitudinal stripes which are very marked anteriorly but gradually become narrower and indistinct posteriorly; whitish-pruinescent spots also present sublaterally, one before and one behind transverse suture, one between supra-alar and intra-alar bristles, one on scutellar bridge; a whitish suffusion on humeral callus and on notopleuron; scutellum with whitish spot anteriorly on each side; pleura with brown spot on upper posterior part of mesopleuron and sometimes with additional brownish suffusion, rather extensively whitish-pruinescent. Fore coxa pale yellowish; other coxae yellowish to tawny; femora fulvous, brown apically and less intensely so near middle; tibiae fulvous, each browned at apex, near middle, and, less extensively so, at base; tarsi fulvous, becoming brownish apically. Wing membrane with faint, almost uniform yellowish brown tinge. Haltere pale yellowish. Abdomen tawny, sometimes with variable brown suffusion. Head with usually one strong upper fronto-orbital bristle, occasionally another weaker one in front of it; one or 2 setulae of peristomial series just below vibrissa stronger than others and usually at least half as long as vibrissae. Thorax. Propleuron pruinescent, without hairs; presutural and 2 pairs of dorsocentral bristles present; prescutellar acrostichal bristle absent. Fore coxa with bristles all rather weak and hair-hke, but quite long in 6: fore femur much stouter than other femora, with ventral spines numerous and not entirely seriate; other femora with ventral spines weaker and less numerous; in 6 femora and tibiae with numerous long, fine ventral hairs. Wing with veins 3 and 4 very slightly converging apically; vein 6 shorter than in other species, discontinued at about 0.8 of distance from anal crossvein to margin (measured in direction of vein 6); anal crossvein sloping a little more towards base at posterior end than in other species, posterodistal angle of anal cell thus a little more obtuse. Abdomen. Sternite 5 of d not divided medially. d postabdomen somewhat resembling that of T. digztata; right spiracle 7 not found, possibly absent; surstylus slender beyond the thickened base, falcate, tapering to the narrow obtuse apex, with rather long hairs anteriorly near base; hypandrium with hairs of moderate length in a series on each side; posthypandrium plate much larger than in T. dzgztata and related species, touching hypandrium anteriorly on each side and connected to basal plates of surstyli posteriorly. Proc. Linn. Soc. N.S.W., 106 (1), (1981) 1982 D. K. McALPINE AND D.S. KENT j mh ls Figs 22-26. 22-23. Tapeigaster pulverea. 22. head of holotype. 23. d genitalia. 24. T. nigricornis, larval cephalopharyngeal skeleton. dc = dorsal cornu. es = epistomal sclerite. hs = hypostomal sclerite. ls = ligulate sclerite. mh = mouth-hook. pb = parastomal bar. ps = pharyngeal sclerite. ve = ventral cornu. 25-26. T. annulipes. 25. puparium. 26. pupal respiratory horn. is = inner spiracles. lc = larval cuticle (puparium wall). pe = pupal cuticle. Dimensions: total length, d 4.2-5.9 mm, 2 4.7-5.0 mm; length of thorax, d 2.2-3.0 mm; 2 2.5-2.9 mm; length of wing, d 5.1-6.5 mm, 9 5.5-6.3 mm. Proc. Linn. Soc. N.S.W., 106 (1), (1981) 1982 5] 52 SYSTEMATICS OF TA PEIGASTER Distribution: Queensland — southern border districts; New South Wales — coast and tablelands; Victoria. Holotype 3 : Royal National Park, near Sydney, 14.x.1956 (AM), D.K.M. Paratypes (all from N.S.W.) : same data as holotype (1 Q, AM); Mount Wilson, Blue Mountains, i.i1i.1958-1976 (1d, ANIC, 2d, AM), E. F. Riek, J. J. T. Evans, M. A. Schneider and D.K.M.; Wentworth Falls, Blue Mountains, xi.xil.1956-1958 (2 d, AM), D.K.M.; Sassafras Gully, Springwood, x1.1956 (1 Q, AM), D.K.M.; Seven Mile Beach State Park, near Gerroa,, v.1976 (29, AM) G.A.H.; 24 km (15 miles) SSE of Braidwood, ix.1956 (1d, ANIC), I.F.C. Other material examined. Qld. : Mitchell Gully, 3.2 km (2 miles) E of Cunningham’s Gap, vi.1966 (1 3, ANIC), Z.R.L.; Lamington National Park, x.1934 (1 Q, UQ), F. A. Perkins; Binna Burra, Lamington National Park, v.1964 (1 9, UQ), B. Genn. Vic. : Powelltown, 60 km E of Melbourne, x.1961 (12, ANIC), D.H.C. Comparative notes. T. pulverea is not very close to any other species of the genus, differing from all in having a rather strong bristle just below the vibrissa. In the hairless prosternum it resembles T. subglabra, but it differs from that species in the thicker fore femur, shortened vein 6, more extensively haired mesoscutum, and presence of the presutural bristle. Tapezgaster subglabra n.sp. (Figs 18-21) 3 2 Coloration. Head tawny; cheek and face paler; postfrons often with two longitudinal brown stripes inside orbital margins (varying in intensity); ocelli surrounded by a dark brown spot; anterior region of orbital margins of postfrons with white pruinescence. Antenna dark brown. Palpus and labella tawny to fulvous. Thorax tawny; mesoscutum with faint median longitudinal brown stripe, fading posteriorly to suture; lateral margins of mesoscutum dark brown, extending to wing bases (seen in lateral view as a stripe tapering posteriorly) ; scutellum tawny; pleura tawny, with a faint band of brown tapering anteriorly, running above coxae; prosternum pale fulvous. Fore coxa pale tawny, other coxae tawny; femora tawny, browned apically; fore femur also with longitudinal dorsolateral brown stripe, fading basally, on inner surface; tibiae tawny, each browned at apex and slightly at basal joint; tarsi tawny, becoming brownish apically. Wing membrane with uniform faint yellowish brown tint. Haltere with brown capitellum, otherwise fulvous. Abdomen tawny to brown. Head with two strong upper fronto-orbital bristles, without strong bristle behind vibrissa. Thorax. Presutural bristle absent; two pairs of dorsocentral bristles present; prescutellar acrostichal bristles absent. Propleuron without hairs below humeral callus; prosternum hairless; fine hairs on mesoscutum reduced to pair of acrostichal series, pair of dorsocentral series, and some irregularly placed lateral ones. Fore coxae with a row of bristles along the inner basal edge; mid and hind coxae with a row of bristles along the anterior basal edge; femora only slightly thickened; fore and mid femora slightly stouter than hind femur; ventral spines of femora restricted to anteroventral and posteroventral series, with a reduction in spines from fore to hind femur; femora and tibiae with numerous fine hairs. Wing with vein 6 pigmented all the way to the wing margin; veins 3 and 4 not diverging distally, almost parallel. Abdomen. Sternite 5 of 5 not divided medially. 6 postabdomen: cercus short and small; surstylus large in comparison, broadly L-shaped, its basal section with numerous long hairs along ventral margin, apical section obtusely rounded. Proc. Linn. Soc. N.S.W., 106 (1), (1981) 1982 D. K. McALPINE AND D. S. KENT 53 Hypandrium broad anteriorly, with row of long incurved hairs along inner anterior margin, with a broad plate posteriorly on each side of aedeagus; anterior margin of plate extended into a finger-like projection bearing two incurved hairs. Dimensions: total length, 3 4.6-6.5 mm, 2 5.0-7.0 mm; length of thorax, dS 2.2-3.0 mm, 9 2.2-2.8mm; length of wing, d 5.7-7.3 mm, 2 6.0-7.4mm. Distribution: New South Wales — principally tableland districts; Australian Capital Territory; Victoria. Holotype 3 : Mount Wilson, Blue Mountains, 5.x.1957 (AM), D.K.M. Paratypes. N.S.W.: Wright’s Lookout, New England National Park, iv.1961 (1 6, AM), D.K.M.; Coachwood Gully, 21.7 km SE of Three Ways, xii.1977 (1 Q, AM), G.D.; Tubrabucca, near Barrington Tops, x.1956 (1 d, AM), D.K.M.; Mount Wilson, Blue Mountains, i.iii-iv.v.vi.viii.ix.x.xi.1957-1980 (27d, 289 AM, 24,192, AINE, 2G, BQ, TIME, er, We ING, Bos 4. USINIMD), IDI (Ce, GzlDg, Gosia, Se 2. Simba) he Vien View ACs Schncidens AnG i. lees Springs. ive l958) (a 3, ANIC) Z.R.L.; Mount Coree, x.1961 (1 d, ANIC), D.H.C.; Bull’s Head, Brindabella Range, iv.1958 (2d, ANIC), D.H.C. Other Material Examined. Vic.: Bonang Highway, Bendoc Rd. Junction, 10 km S of Bonang, xi.1976 (20,29, ANIC), D.H.C. and Z.R.L.; Mount Beauty, x.1961 (19, ANIC), D.H.C.; Mount Donna Buang, iv.1963 (1 Q, AM), D.K.M.; Warburton, iv.1963 (30,392, AM), D.K.M.; Belgrave, 1.1966 (1d, AM), D.K.M. Comparative notes. T. subglabra is not very close to any other species of the genus, differing from all in the absence of presutural bristles and reduction of hairs on the mesoscutum. In the hairless prosternum it resembles T. pulverea, but differs from that species in other characters, as indicated in the description of T. pulverea. MORPHOLOGY OF EARLY STAGES The morphology of the early stages of Heleomyzidae s.1. is too poorly known to enable any broad taxonomic inferences to be drawn from the new data. Two points may be made here. The larva of Tapeigaster differs greatly from that of Cazrnsizmyza (McAlpine, 1968) although the two genera have been considered to be related (Griffiths, 1972). We are unable to find any points of resemblance which are not shared with larvae of numerous other schizophoran families. The differences in the facial organs and form of the terminal spiracle-bearing segment are particularly marked. The pupa has a prothoracic respiratory horn which penetrates the puparium and is thus visible externally. Such a process is frequently present in muscoid (calyptrate) pupae, but is apparently known only in certain north-temperate heleomyzids among the acalyptrates. According to Hennigian reasoning the presence of such a plesiomorphic character cannot be employed as evidence for close phylogenetic relationship between the tribes Heleomyzini and Tapeigastrini. Nevertheless it may be statistically significant that the only known retention of this kind of respiratory horn among non-muscoid Schizophora is in groups which, from other characters, seem referable to the family Heleomyzidae. Tapeigaster annultipes Larval material of T. annulpes was obtained from an adult female collected at the Australian Museum, 24 April 1980. The fly laid eggs on a cultivated mushroom (Agaricus sp.). Of the subsequent larvae raised, some were preserved in alcohol for later dissection and the rest left to pupate. Adults failed to emerge so the remaining puparia were either dry mounted or placed in alcohol. It was from this latter material that the detail of the pupal respiratory horn was obtained. Proc. Linn. Soc. N.S.W., 106 (1), (1981) 1982 54 SYSTEMATICS OF TA PEIGASTER Figs 27-32. Tapeigaster annulipes, larva. 27. entire larva in lateral view. 28. facial region. 29. anterior spiracle. 30. posterior view. 31. posterior spiracle. 32. cephalopharyngeal skeleton, with ventral view of epistomal sclerite and dorsal view of hypostomal sclerite and ligulate sclerites. dec = dorsal cornu. ds = dentary sclerite. es = epistomal sclerite. hs = hypostomal sclerite. Is = ligulate sclerite. mh = mouth- hook. pb = parastomal bar. ps = pharyngeal sclerite. ve = ventral cornu. Last instar larva (Figs 27-32) creamy-white, somewhat elongate, circular in cross- section; anterior end tapered; posterior three quarters almost uniform in width; Proc. Linn. Soc. N.S.W., 106 (1), (1981) 1982 D.K. McALPINE AND D.S. KENT 55 posterior end truncate and sloping (Fig. 27). Segment 1 (Fig. 28) divided by a median groove; each lobe with a short, pale yellowish-brown, 2-segmented antennal papilla dorsoapically and a circular sensory plate ventroapically; a smaller third pair of papillae located on each side of atrial opening. Facial area with numerous parallel and branching, minutely serrated, sclerotized ridges posteriorly, and with numerous cuticular teeth anteriorly and on each side of atrial opening, the latter of 4 different types: the foremost ones (type 1) consisting of numerous small subglobular serrated plates occupying a narrow transverse band from each side of anterior extremity of atrial opening to lateral extremity of atrial opening to lateral extremity of facial region; teeth of type 2 lying across anterior extremity of atrial opening and extending into it, more elongate than those of type 1 with fewer, longer cusps, the inner ones simply conical or unicuspid ; teeth of type 3 large and few, lying behind third pair of papillae, each with 6-10 cusps; a triangular sheath-like structure located at posterior margin of the atrial opening projecting into the opening. Segment 2 bearing paired, light yellowish-brown, fan-shaped anterior spiracles on short processes postero- laterally, each bearing 10-14 short apical papillae; a yellowish-brown sclerotized ring encircling base of each process (Fig. 29). Segments 3-12 bearing narrow encircling bands of unpigmented spicules anteriorly. Segments 5-12 bearing ventral creeping pads on anterior margins, each pad bearing unpigmented spicules arranged in short parallel rows, directed both anteriorly and posteriorly. Segment 12 (Fig. 30) bearing a pair of spiracular tubercles and five pairs of papillae. Posterior spiracles set in a concave depression, with 4 pairs of papillae encircling depression on outer rim, and a fifth and largest pair ventrally directed and associated with the anal pore. Posterior spiracular plates (Fig. 31) yellowish brown, located on separate, short tubercles; each plate with 3 narrowly oval apertures at about 65° to each other, containing lamellae just within, and with 4 colourless, much branched interspiracular hydrophobe hairs; a yellowish brown sclerotized subcuticular ring below _ spiracular plate. Cephalopharyngeal skeleton (Fig. 32) black to dark brown. Mouth-hooks well developed, paired, separate, without accessory teeth below the curved apical section, with small window in posteroventral corner and distinctive hook-like nodules on the posterodorsal margin of the basal sections. Dentary sclerites paired, separate, near posteroventral margin of mouth-hooks. Epistomal sclerite not fused to parastomal bars, largely dark-pigmented posteriorly, lighter on anterior median process and margin, with pair of narrow rod-like sclerites extending posteriorly to fuse with inner posterodorsal margin of hypostomal sclerite. Parastomal bars narrow, dilated anteriorly, darkly pigmented, separate from upper margin of hypostomal sclerite and fused posteriorly with pharyngeal sclerites. Hypostomal sclerite H-shaped, not fused posteriorly to pharyngeal sclerites; anterior rami shorter and broader than posterior rami; hypostomal bridge darkly pigmented on all but narrow posterior band. Ligulate sclerites curved, paired and separate; situated anterior to hypostomal sclerite and between posterior margins of mouth-hooks. Pharyngeal sclerites with anterodorsal bridge joining anterior ends of dorsal cornua, bridge with several windows medially; with antero-ventral projections lying below posterior rami of hypostomal sclerite; dorsal cornua narrower and slightly longer than heavily pigmented part of ventral cornua, which are lightly pigmented posteroventrally. Pharyngeal ridges between ventral cornua well developed but lacking pigmentation. Length 10-11 mm. Puparium (Fig. 25) light reddish brown, darker at each end, elongate-ovoid, slightly more curved dorsally than ventrally; posterior end rounded, with posterior spiracular tubercles heavily pigmented and surrounded by concentric ridges; anal plate surrounded by a narrow band of ridges; anterior end tapered in lateral view; anterior Proc. Linn. Soc. N.S.W., 106 (1), (1981) 1982 56 SYSTEMATICS OF TAPEIGASTER spiracles heavily pigmented and compressed, but otherwise as in larva; thoracic segments 1 and 2 with prominent encircling ridges; thoracic segment 3 with slightly less distinct ridges restricted to anterior margin; abdominal segment 1 (segment 4) bearing on its dorsal surface the paired, brown outer pupal respiratory horns, which project through the puparium wall (as in Musca). Outer respiratory horn (Fig. 26) with heavily sclerotized wall and obliquely aligned pores, arising from a short outgrowth of the dorso-lateral edge of the precursor prothorax of the pupa; this outgrowth also bearing inner bilobed spiracles on posterior surface; spiracles slightly domed, with oval pores, divided medially by a bulbous process, opening into cavity between puparium wall and pupal cuticle, each subtended by a crescent-shaped sclerite. TABLE 1. Host Records of Tapezgaster Species Host Records Adults Reared From: Adults Collected On: T. nagricornis ‘Pleurotus lampas’ = Pleurotus nidiformis Berk. ‘Pleurotus lampas’ Amanita grisea Mass. & Rodw. Amanita ochrophylla (Cke. & Mass.) Clel. Boletus luridus Schaeff. ex Fr. ‘Boletus granulatus’ = Suzllus granulatus (L. ex. Fr.) O. Kuntze ‘Boletus portentosus’ = Phaeogyroporus portentosus (Berk. & Br.) McNabb. Cortinarius brunneus (Pers. ex. Fr.) Fr. Agaricus sp. Boletus sp. “Puff ball fungus’ T. annulipes ‘Psalliota campestris’ = Agaricus campestris L. ‘Pleurotus lampas’ ex Fr. T. digitata ‘fungi’ T. cinctipes ‘Pleurotus lampas’ T. paramonow ‘mushrooms’ T. argyrospila ‘Puff ball fungus’ T. brunneifrons ‘fungi’ Tapergaster nigricornis A similarly detailed study of T. négr¢cornis larvae was not possible as the available material consisted of only two larvae (possibly last instar), collected ex fungus from Broulee, N.S.W., 6 May 1980 by Z.R.L. (ANIC). Only one specimen was dissected, the other was left entire. Larvae of T. nigricornis differ little from those of T. annulipes. The minor differences evident are as follows: (a) in facial area, the patterns of cuticular teeth differed slightly in number and arrangement, i.e. teeth of type 2 are all unicuspid and extend into the atrial opening in only a single row; (b) the anterior spiracle, is more broadly fan-shaped, bearing only 10 papillae; each papilla has a single pore opening into a roughly circular cavity lined by tear-shaped protuberances. The major difference between the larvae is in the cephalopharyngeal skeleton; the mouth-hook Proc. Linn. Soc. N.S.W., 106 (1), (1981) 1982 D. K. McALPINE AND D. S. KENT 57 has an elongate, acute, inwardly directed process on the posterodorsal surface; the mouth-hook has a large window in the basal section and a lightly pigmented anterior region. The parastomal bar is parallel-sided along its entire length with slight upward curve anteriorly. The dentary sclerite is absent in both specimens. The ligulate sclerite is large, with incurving dorsal bulge; this sclerite is visible in lateral as well as ventral view (Fig. 24). A comparison of puparia and pupae showed that those of T. negricornis differed little from those of T. annulzpes. NOTES ON BIOLOGY Adults of Tapezgaster annulitpes are often observed on fruiting bodies of agaric and other fungi, and other species of the genus have either been observed by the authors on fungi or recorded on labels as inhabiting fungi. These species are T. nigricornis, T. brunnetfrons, T. cinctipes, T. annulipes, T. digitata, T. paramonovi and T. argyrospila. In the case of T. annulipes males frequently take a position on top of the pileus and defend the position from rival males. Females approaching and flying beneath the pileus (perhaps for oviposition) have been seen to be followed by the male and mounted, sometimes after a struggle, but no time was given to courtship in such instances. Adults of Tapezgaster species are often swept from ferns and low foliage during insect-collecting activities. They are rarely found at night, perhaps because they are mainly diurnal, and they do not habitually settle on tree trunks. Larvae of at least five species of Tapezgaster live in fruiting bodies of fungi. T. annulipes and perhaps all or most other species can be reared in the laboratory on commercial mushrooms, but they are not considered to be pests of this crop. The large number of mushroom spores in facial grooves of preserved T. annulipes larvae suggest that they had been feeding particularly on the gills. Presumably the fungi on which the adults are found are normally those that the larvae infest. Table 1 gives the host records available to us. ACKNOWLEDGEMENTS We are indebted to Dr D. H. Colless, C.S.I.R.O. Division of Entomology, Canberra, and Miss M. A. Schneider, Department of Entomology, University of Queensland, St Lucia, for loans of material. Mr B. H. Cogan, British Museum (Natural History), London, Dr R. Contreras-Lichtenberg, Natural History Museum, Vienna, Dr W. Mathis, National Museum of Natural History, Washington, D.C., and Dr A. Neboiss, National Museum of Victoria, Melbourne, supplied information on types. References Bezzi, M., 1923. — Note on the Australian genus Tapezgaster Macq. (Diptera) with descriptions of new species. Aust. Zool. 3: 72-78. Co..ess, D. H., and McAtpine, D. K., 1970. — Chapter 34. Diptera. In The znsects of Australia: 656-740. Melbourne: Melbourne University Press. GrirFiTHs, G. C. D., 1972. — The phylogenetic classification of Diptera Cyclorrhapha with special reference to the structure of the male postabdomen. The Hague: W. Junk. HENDEL, F., 1917. — Beitrage zur Kenntnis der acalypraten Musciden. Dé. ent. Zettschr., 1917: 33-47. HEeNNic, W., 1958. — Die Familien der Diptera Schizophora und ihre phylogenetischen Verwandtschaftsbeziehungen. Beztr. Ent. 8: 505-688. Macquarr, P. J. M., 1847. — Diptéres exotiques, supplément 2: 104 pp. 6 pl. ——, 1851. — Diptéres exotiques, supplément 4 (pt. 2) : 161-364, pl. 15-28. Proc. Linn. Soc. N.S.W., 106 (1), (1981) 1982 58 SYSTEMATICS OF TA PEIGASTER MALLocH, J. R., 1926. — Notes on Australian Diptera. No. ix. Proc. Linn. Soc. N.S.W. 51: 545-554. ——, 1927. — Notes on Australian Diptera. No. x. Proc. Linn. Soc. N.S.W. 52: 1-16. , 1935. — Notes on and descriptions of new species of Australian Diptera. Aust. Zool. 8: 87-95. McALPINnE, D. K., 1968. — The genus Cazrnsemyza Malloch (Diptera, Heleomyzidae, Rhinotorini). Rec. Aust. Mus. 27: 263-283. PARAMONOY, S. J., 1955. — Notes on Australian Diptera (xx). A Review of the genus Tapezgaster Macq. (Neottiophilidae, Acalyptrata) , Diptera. Ann. Mag. nat. Hist. (12) 8: 453-464. WALKER, F., 1853. — Insecta Saundersiana. Diptera 4: 253-414. ——, 1858. — Characters of undescribed Diptera in the collection of W. W. Saunders, Esq., F.R.S. & c (Continued). Trans. ent. Soc. Lond. (2) 4: 190-235. Proc. Linn. Soc. N.S.W., 106 (1), (1981) 1982 A Comparative Morphological Study of the Reproductive Systems of some Species of Tapeigaster Macquart (Diptera: Heleomyzidae) MARGARET A. SCHNEIDER (Communicated by D. K. McALPINE) SCHNEIDER, M. A. A comparative morphological study of the reproductive systems of some species of Tapeigaster Macquart (Diptera: Heleomyzidae). Proc. Linn. Soc. N.S.W. 106 (1), (1981) 1982: 59-65. The female reproductive systems of Tapezgaster annulipes Macquart, T. annulata (Hendel), T. swbglabra McAlpine and Kent, T. dzgztata McAlpine and Kent, T. pulverea McAlpine and Kent, and T. lutezpennzs Bezzi, and the male re- productive systems of T. annulata, T. lutecpennis, T. pulverea, T. annulipes and T. subglabra are discussed and illustrated either fully or in part to show specific differences. Margaret A. Schneider, Department of Entomology, University of Queensland, St Lucia, Australia 4067; manuscript received 22 July 1981, accepted for publication 23 September 1981. INTRODUCTION Alcohol-preserved specimens of six species of Tapeigaster Macquart (Heleomyzidae) were dissected to investigate whether the male and female reproductive systems would show differences at the species level and similarities which could be considered generic characters. There has been little published on either the male or female reproductive systems of other genera of Heleomyzidae for comparison. As in many families of the Schizophora, Heleomyzidae appear to have a groundplan number of three spermathecae, with the occasional increase to four or decrease to two being apomorphic (Hennig, 1958). In his investigations Sturtevant (1926) found three spermathecae in all the genera except one, in which there were four. The females of all six species of Tapezgaster which were dissected in the present study have only two spermathecae and Hennig found only two in T. margznzfrons Bezzi (= T, nigricornis (Macquart) ). Further comparative studies are needed to determine how often this reduction in the number of spermathecae occurs in Australian genera of Heleomyzidae. In two other endemic genera so far examined, only two spermathecae were found in each of the two species of Austrolerxa McAlpine dissected while three were found in each of the three species of the closely related genus, Diplogeomyza Hendel. It is interesting to note, however, that in Diplogeomyza the two spermathecae on the same side share a common duct throughout its length and are very closely associated. The shape and overall appearance of the spermathecae of Diplogeomyza spp. and Austroleria spp. are somewhat similar and very different from the simple, spherical form on a long duct as found in Tapezgaster spp. The females of Tapezgaster show only slight specific differences such as the relative positions of the spermathecal and accessory gland duct openings into the vagina, the relative size of the structures concerned and the presence or absence of a sclerite in the wall of the vagina in the region of the spermathecal and accessory gland duct openings. Differences between species are more obviously shown in the male reproductive systems, with shape of testes and length and form of accessory glands and the ejaculatory duct being useful characters. Male specimens of T. dzgztata McAlpine and Kent were not available for dissection. Proc. Linn. Soc. N.S.W., 106 (1), (1981) 1982 60 REPRODUCTIVE SYSTEMS OF SOME TA PEIGASTER Proc. Linn. Soc. N.S.W., 106 (1), (1981) 1982 M. A. SCHNEIDER 61 FEMALE REPRODUCTIVE SYSTEM Tapergaster annulipes Macquart (Fig. 1) The ovaries, which occupy the greater part of the abdominal cavity in a gravid female, end in short lateral oviducts which meet to form a muscular, median oviduct. There are two spherical, brown-black spermathecae, one on each side; each spermatheca is surrounded by a white fat mass which, in the undisturbed state, is embedded in the tissue surrounding the ovary on the same side. The spermathecal ducts are long, thin, relatively stout-walled and darkly sclerotized at the distal ends; they enter the vagina at the midline. Each of the two accessory glands is an elongate, pale, thin-walled, semi-transparent structure embedded in tissue adjacent to the spermatheca on the same side. The distal end of each accessory gland duct, adjacent to the gland, is somewhat expanded and muscular, presumably with a sphincter function; the accessory gland ducts enter the vagina at the middle immediately posterior to the spermathecal ducts. The vagina, or posterior part of the median oviduct, is slightly expanded, with thick muscular walls in the region of these ducts, and a small dark sclerite lies in the ventral wall. No morula gland or ventral receptacle is present. Strong muscle bands originating from the posterior end of the vagina and inserted in the wall of the vagina just anterior to the spermathecal ducts had to be cut to expand the vagina to its full length. Tapergaster lutetpennis Bezzi (Fig. 2) Differs from T. annulipes as follows. Median oviduct shorter, about three-fifths length of that in T. annulzpes. Spermathecal ducts pale yellow along most of length, colour slightly darker at proximal ends where they enter the vagina a little lateral to midline on each side. In this region, vagina bulbous with greatly thickened muscular walls and internal yellow, thickened supporting bands associated with openings of the ducts. Accessory glands long, as in T. annulzpes, but distinctly bulbous distally (Fig. 2). Accessory glands ducts more expanded distally, entering vagina immediately posterior to spermathecal duct on each side. Tapezgaster annulata (Hendel) Fig. 3 Differs from T. annulipes as follows. Median oviduct short, about half length of that in T. annulipes. Accessory glands and muscular distal ends of accessory gland ducts much shorter but otherwise accessory gland ducts a little longer than in T. annulipes. Spermathecal and accessory gland ducts meeting vagina lateral to midline as shown in Fig. 3. Nosclerite in ventral vaginal wall. Tapergaster pulverea McAlpine and Kent (Fig. 4) Resembles most closely T. annulata but differs as follows. Vagina in region of spermathecal duct openings more bulbous with thicker muscular walls. Accessory Figs 1-5. Female Tapezgaster spp.: (1) T. annulipes, reproductive system; (2) T. lutezpennis, accessory gland; (3) T. annulata, reproductive system, ovaries not shown; (4, 5,) proximal ends of spermathecal and accessory gland ducts: (4) T. pulverea; (5) T. digitata. ac = accessory gland, acd = accessory gland duct, c = cercus, ft = fat mass, lo = lateral oviduct, m = cut muscle band, mo = median oviduct, ov = ovary, sp = spermatheca, spd = spermathecal duct, T9 — tergite 9, v — vagina. Proc. Linn. Soc. N.S.W., 106 (1), (1981) 1982 62 REPRODUCTIVE SYSTEMS OF SOME TA PEIGASTER ~aC vd— 05mm Proc. Linn. Soc. N.S.W., 106 (1), (1981) 1982 M. A. SCHNEIDER 63 glands and ducts smaller and shorter than in 7. annulata, entering vagina distinctly lateral to spermathecal ducts (Fig. 4). Tapergaster subglabra McAlpine and Kent Only specimen of this species available for dissection proved to be very immature with ovaries not developed and accessory glands small and very thin-walled. Sclerite in ventral wall of vagina not present. Sclerotization of distal ends of spermathecal ducts absent, these ducts meeting vagina as in T. annulipes but accessory gland duct appearing to meet vagina lateral to midline. Tapeigaster digitata McAlpine and Kent (Fig. 5) Although generally agreeing with description given for T. annultpes, differs as follows. Spermathecae mid-brown and spermathecal ducts meeting vagina lateral to midline (Fig. 5). Accessory glands about three-fourths length of those of T. annulipes, pale yellowish rather than white and not as thin-walled. MALE REPRODUCTIVE SYSTEM Tapezgaster annulata (Fig. 6) The testes are golden brown, apically tapered, elongate sacs. The stout, golden- brown vasa deferentia narrow proximally where they enter the apex of the pale- coloured ejaculatory duct. This duct is at first relatively stout and thick-walled but abruptly narrows about half way along its length. Posteriorly it passes into the ejaculatory bulb which has articulated with it, the muscle-covered ejaculatory apodeme. A narrow, strong duct connects this with the aedeagus to which is articulated the large, flat, muscle-covered aedeagal apodeme. The accessory glands which enter the apex of the ejaculatory duct ventral to the vasa deferentia are long, pale yellow and only slightly expanded apically. Tapeigaster annultpes (Fig. 7) Generally similar to T. annulata but differs as follows. Testes (Fig.7) shorter and more expanded basally. Vasa deferentia a little longer. Accessory glands about half as long as those of T. annulata and almost as thick as distal portion of ejaculatory duct, which is wider and longer than in T. annulata. Tapeigaster luterpennis Specimens of this species available for dissection were immature so it 1s difficult to make accurate comparisons with other species. Testes small but of similar shape to those of T. annulipes except not so markedly expanded basally and more rounded apically. Ejaculatory duct wide distally as in T. annulipes but accessory glands in specimens examined long and narrow, about half width of ejaculatory duct, as in T. annulata. Fig. 6-7. Male Tapezgaster spp.: (6) T. annulata, reproductive system; (7) T. annulipes, testis and vas deferens. aa = muscle-covered aedeagal apodeme, ac = accessory gland, ae = aedeagus, ea = muscle covered ejaculatory apodeme, ej = ejaculatory duct, t = testis, vd = vas deferens. Proc. Linn. Soc. N.S.W., 106 (1), (1981) 1982 64 REPRODUCTIVE SYSTEMS OF SOME TA PEIGASTER Tapeigaster pulverea Appears most similar to T. annulata although once again available specimens seemed immature judging from small size of testes, especially in one specimen in which there was also very little pigmentation of testes and vasa deferentia. Testes relatively long and narrow but not as apically tapered as in T. annulata. Accessory glands narrow but about half length of those in T. annulata. Tapeigaster subglabra (Fig. 8) General appearance of male reproductive system of this species quite different from that of any other species dissected as shown in Fig. 8. Testes golden-brown with 8 > €a 05mm Fig. 8. Tapeigaster subglabra, male reproductive system. ac = accessory gland, ae = aedeagus, ea = muscle covered ejaculatory apodeme, ej = ejaculatory duct, t = testis, vd = vas deferens. Proc. Linn. Soc. N.S.W., 106 (1), (1981) 1982 M. A. SCHNEIDER 65 rounded lobe distally. Pale brown vasa deferentia short, entering median ejaculatory duct at its slightly expanded apex. Accessory glands much shorter than those of T. annulata, distinctly expanded at distal ends and entering distal end of ejaculatory duct posterolateral to vasa deferentia. Aedeagal apodeme present although not shown in Fig. 8. DISCUSSION All female specimens dissected had two spherical, dark spermathecae and males had large, golden-brown testes. These could perhaps be regarded as generic characters but further comparative studies need to be made to confirm this. The dissection also showed that even in the females there are specific differences in the morphology of the reproductive systems but it would be difficult to make accurate interpretations concerning closeness of relationships between species from the differences observed. However, in conjunction with the information gained from the dissections of males it would seem that T. pulverea and T. annulata are quite closely related. T. lutezpennis is perhaps closest of T. annulipes but with some similarities to T. annulata. T. subglabra is not very close to any of the other species dissected. ACKNOWLEDGEMENTS I wish to thank Dr D. K. McAlpine of the Australian Museum, Sydney, for his assistance and for supplying specimens for dissection. References Hennic, W., 1958 — Die Familien der Diptera Schizophora und ihre phylo-genetischen Verwandt- schaftsbeziehungen. Beztr. Ent. 8: 505-688. STURTEVANT, A. H., 1926 — The seminal receptacles and accessory glands of the Diptera, with special reference to the Acalypterae. J. New York ent. Soc. 34: 1-21, pls 1-3. Proc. Linn. Soc. N.S.W., 106 (1), (1981) 1982 Neurochaeta inversa McAlpine (Diptera: Neurochaetidae) and Seed Set in Alocasta macrorrhiza (L.) G. Don (Araceae) in southeast Queensland DOROTHY E. SHAW, B. K. CANTRELL, and K. J. HOUSTON SHAw. D. E., CANnTRELL, B. K., & Houston, K. J. Neurochaeta inversa McAlpine (Diptera: Neurochaetidae) and seed set in Alocasta macrorrhiza (L.) G. Don (Araceae) in southeast Queensland. Proc. Linn. Soc. N.S.W. 106 (1), (1981) 1982: 67-82. Forty one intact infructescences of Alocasta macrorrhiza (L.) G. Don were examined for presence or absence of Neurochaeta inversa McAlpine and for numbers of seed set. N. znversa was recorded in all the spathal chambers of 20 infructescences in two rainforest localities, in only two chambers of nine infructescences from two natural stands in cleared areas, but not at all in chambers of 12 infructescences from cultivated garden plants. Numbers of seed were not correlated with the numbers of N. mmversa in the chambers, and seed was still set at sites where no N. inversa was recorded. Pollination, therefore, is probably independent of ovipositing females of N. mversa. Dorothy E. Shaw, Plant Pathology, and B. K. Cantrell and K. J. Houston, Entomology, Department of Primary Industries, Meters Road, Indooroopilly, Australia 4068; manuscript received 2 June 1981, accepted for publication 23 September 1981. INTRODUCTION Alocasta macrorrhiza (L.) G. Don isa large aroid commonly known as ‘cunjevoi’ in Australia. Its inflorescence consists of a spathe and spadix (Fig. 1, A). The spathe is divided into a terminal expanded portion (the spathal blade) and a convoluted persistent base (the spathal tube) which forms the spathal chamber. The spadix consists of four sections (Fig. 1, B) which, from the tip downwards, are: the sterile terminal appendage; the staminate portion with synandria bearing the pollen; the median sterile (m.s.) florets of the constricted portion and the pistillate flowers or ovaries (with ovules) at the base. The stigmas are receptive prior to anthesis. After fertilization, the tip of the spathal tube tightly clasps the constricted portion of the spadix, forming the ‘sealed’ spathal chamber. The unfertilized ovaries in the sealed chamber persist on the maturing infructescence and remain about the same size as at anthesis, but fertilized ovaries (berries) enlarge to accommodate the developing fertilized ovules (seed). The berries, green at first, tuyn yellow and then red at maturity. The spathal tube also enlarges to accommodate the developing seeds. At maturity, the walls of the spathal tube turn from green to yellow, and split longitudinally into four or five segments which unfurl by rolling outwards revealing the ripe berries: McAlpine (1978) described Neurochaeta inversa, noting that it is only known to occur in association with A. macrorrhiza in its original habitat in or at the edge of Proc. Linn. Soc. N.S.W., 106 (1), (1981) 1982 NEUROCHAETA INVERSA McALPINE AND SEED SET 68 0 8Il 119 Lele, I 0¢€ § GOL ws, 0's € 30l Pry yoorg seddy 02 PLL SANOZ ALVIGAWNUALNI 8E 08 *6 6S GG 9 9 al § FI ae G 6 SSI LG v6 v 8I (Gat 6°48 GOL g § 6L 6 YS IZ 6 Pv Sl 9°68 H GP 86 8° rg 89 [Z P LLG 6 $6 L&I 9 9L PGE +V 16 161 G GP G06 9°OL IST v GE 991 +9 68 06 g OP 0066 W LG 1¥G 1 46 Lit L 67 061 6 LY 08 9 0°86 G 081 (wu LG 9 x6 V 9 4 PS 8 *8°G 8 g 96 81 «6 81 LI 6 0°66 L01 (Du GV O8T §°S6 18 g Or CsI 0°96 061 (4 0' rr 0 §81 A 8°0¢ L Sst Snollo[) IW LI L81 § LS FII g 06 99 6 LE 1g SG Ig 181 ian Ill I L°66 L vtl WO neaie]d L'3% L' PPL uo UIE T 9°62 L'L&Sl LSAUOANIVY “ON “AV “ON “AV “ON “AV “ON “ON “AV “ON (AV “ON (AV “ON % “ON WeIqeH Ayryeso'T ais WwIiIqey Ayipeso7: aig peay ied uoneUsIsap Ayryesoy (sa1i19q) peoy ‘LV.LIGVH peay sad vs1anue “NY peey sed paas SaTIRAO pazI[NIIq ‘QUS 1DI2QDY puD KjxDI0) ‘aj2s “‘pvaYy Lad eSIZAUL ‘N pun paas fo suaquinu pup (sars1aq) sarsvno pazyrsaf fo advjuao1ag puv saqunyy Ll aTadVL Proc. Linn. Soc. N.S.W., 106 (1), (1981) 1982 69 D. E. SHAW, B. K. CANTRELL AND K. J. HOUSTON ‘JOU IO pazI[N1ej L9yYIOYM Surysnsunsip Jo Ayyprqissodurt yo asnedaq ‘popnyaut jou (G a[qe J 99S) JOI 19M YIIM SaLIvAG,, 0 09T 8°§9 SII v 0 69 € $¢ vg g 0 19 8° bP 9g (4 0 641 6 16 FOL I 0°0 § 601 UN 0°0 € 601 eon] 3g 0 II Gg Or G 0 SST 1&8 ISI 7 0 96 8°6 61 g 0°0 §°€9 OW 0 G86 186 9S1 L 0 091 6 69 16 9 0°0 G°666G Poicl 0 Ill VY 6h 18 G 0 OF 8°06 Ig v 0°0 GGL UV 0°0 § GIT eYyI-1009 IW 0 v6l 8°64 66 I 00 . 0 P61 S 00 0 FI tH dey 0°0 € 6Il Naddv 0 GV OLI 66 § 0 GE 1 01 96 6 0 9 Gg G I 0°0 HUG YW 00 j LLG quUOWYII0g G 0ST § 08 OLl 9 0 Srl LOL 66 G 0 0S GL6 9¢ a4 SI LS LST \t § 0 Chat § 19 89 6 “ON “AV “ON AV “ON “AV “ON “ON (AV “ON AV “ON “AV ‘ON % “ON WeIGeyY Aqpeso'y EXC WIqey Aypeoo7y ais peay sad uoleUsIsap Aqtpeo0] (satii9q) peoy ‘LV.LIGVH peoy tod vs1aau Ny peoy sed paas SOIIVAO PdZI[N 134 ‘SIS jojIQoYy pup Kjz27VI0] ‘azws ‘poay Lad eSI2AUl ‘NY PUD paas fo siaquinu puD (sarL1aq) sarsoao pazyiysa Jo asDjuaoLag pup LaQqQUiNnny PINUyU0I— | ATAVL Proc. Linn. Soc. N.S.W., 106 (1), (1981) 1982 70 NEUROCHAETA INVERSA McALPINE AND SEED SET rainforest and not with cultivated examples. Adults live on both sides of the leaf blade, on the outer surface of the petiole, within the petiolar cleft and in the spathe of the inflorescence at anthesis. Eggs (which were not observed) are apparently laid on the female part of the inflorescence and larvae live between the developing fruitlets where they are in contact with a ‘watery liquid’ found within the spathe which forms a sealed chamber during the whole of the larval development. Pupariation, and often emergence of adults, takes place while the fruitlets are still unripe and the spathe sealed. The adults, therefore, cannot escape until the spathe splits at maturity. He stated that ovipositing females of N. znversa may act as pollinators of 4. macrorrhiza, but observed that, as many insects visit the flowers, they are probably not the sole pollinating agents. In studies carried out in southeast Queensland in 1980 and 1981, numbers and percentages of fertilized ovaries (berries), numbers of fertilized ovules (seed) and insects found in the sealed chambers of infructescences were recorded in order to determine whether any correlation exists between the numbers of seed set and numbers and types of insects recorded in the chambers. SITES AND SAMPLES Samples were from plants in three general habitats, viz., rainforest, intermediate zones of remnant natural stands in mainly cleared areas, and in gardens. The localities and brief descriptions of the sites are as follows : 1. Rainforest Samples were from clumps in small private holdings, adjacent to large or relatively large reserved areas of rainforest, as below: Lamington Site O’R, on creek in mixed rainforest /sclerophyll area. Plateau: Mt Tamborine: Site BC, on creek in open area near rainforest and sclerophyll forest. Mt Glorious: Sites F, R(L), R(R), H, M, in rainforest, some near dirt road, some near small clearings or thinned rainforest. 2. Intermediate zones Upper Brookfield: Site Sm, isolated natural stand on dry creek in cleared area of secondary growth and weeds; general area under crop and artificial grassland; nearest rainforest remnant about 0.3 km distant. Beechmont: Site K, isolated natural stand on dry creek surrounded by artificial grassland and weeds; nearest rainforest remnants about 0.5 km distant. 3. Garden Chapel Hill: Site S, one cultivated plant in suburban garden; area with housing and gardens, and remnant sclerophyll and rainforest. Mt Coot-tha: Sites AR, ER and MC, scattered planted clumps, some near creek, in suburban Mt Coot-tha Botanic Gardens; area originally sclerophyll forest, planted to simulated rainforest during the last three years. St Lucia: Site NR, large planted clump near creek in small simulated rainforest surrounded by parklands in the University of Queensland. Each site was sampled once except Mt Coot-tha (twice) and Mt Glorious (three times) . Proc. Linn. Soc. N.S.W., 106 (1), (1981) 1982 D. E. SHAW, B. K. CANTRELL AND K. J. HOUSTON 71 The samples consisted of 54 infructescences or heads which were examined for insects. Forty-four heads had intact and sealed spathal chambers; of these, 41 were examined for seed set and three especially for physiological maturity. Ten other heads were too young for seed or had damaged spathal tubes. The samples were at various stages of development from anthesis to near-maturity. Counts for insects and other organisms were made by removing each spathal tube in four or five longitudinal strips and examining the inside of each strip immediately under the stereomicroscope for larvae, pupae, etc., any detected being removed to vials for later study. Any liquid in the spathal chamber was also examined. The outer surfaces of the ovaries were examined microscopically in situ and any organisms removed to vials. Each ovary was then removed with forceps while still under microscopic observation, this being especially necessary in order to detect colourless larvae around the ovary bases and on the white spadix core. SEED AND INSECT RECORDS Numbers of fertelized ovaries (berries) per head The number and percentage of fertilized ovaries (berries) per head for 41 heads with intact, sealed chambers, are given in Table 1. There was considerable variation in the number of berries per head (six to 191), and in the percentage of berries per head, which ranged from 3.5% to 98.1%. The three heads from Beechmont (Site K) all had a low number of fertilized ovaries, viz., 3.5%, 10.1% and 17.0%. The numbers were also low for the three heads from Site R(L), Mt Glorious, with 18.3%, 5.8% and 4.2%. Numbers of seed per head The numbers of seed per head are given in Table 1, ranging from six seeds in six berries (an average of 1.0 seeds per berry) to 354 seeds in 191 berries (an average of 1.9 seeds per berry). The lowest averages of seed per head per site were for Site R(L), Mt Glorious, and Site K, Beechmont, with 10.7 and 27.7 seeds per head per site respectively, while the highest were at Site M, Mt Glorious, and Site ER, Mt Coot-tha, with 222.0 and 222.5 seeds per head per site respectively. Statistical analysis showed that differences between some sites are significant at either the 5% or 1% level (Table 2). TABLE 2 Statistical analysis of average numbers of seeds per head per site HABITAT, Site* Seeds Sites (numbered as in third column) locality Design- Num- per head significantly less than designated site ation ber# per site Av. No. 5% level 1% level RAINFOREST Lamington OR 144.7 Sites 3, 8 Plateau 1 Mt Glorious F 2 183.0 Sites 8, 11 Site 3 R(L) 3 10.7 R(R) 4 180.5 Site 8 Site 3 M 5 220.0 Sites 9, 12 Sitesi3i16) 7,8) 1 H 6 89.6 Proc. Linn. Soc. N.S.W., 106 (1), (1981) 1982 72 NEUROCHAETA INVERSA McALPINE AND SEED SET TABLE 2—continued HABITAT, Site* Seeds Sites (numbered as in third column) locality Design- Num- per head significantly less than designated-site ation ber# per site Av. No. 5% level 1% level INTERMEDIATE ZONES U. Brookfield Sm 7 102.3 Site 3 Beechmont K 8 Dl GARDEN Mt Coot-tha AR 9 75.5 ER 10 222.5 Sites 6, 7, 9, 12 Sites 3, 8, 11 MC 11 63.3 St Lucia NR 12 109.3 Site 3 Garden site 1 at Chapel Hill omitted because only one sample. # Sites numbered consecutively for statistical analysis. Less variation was shown in the average numbers of seed per head per locality, which ranged from 27.7 for Beechmont to 144.7 for Lamington and even less variation in the average numbers of seed per head per habitat, being 77.4 for the intermediate zones, 112.3 for garden areas and 137.1 for rainforest (Table 1). Insects The only insects found within undamaged, still-sealed heads were 1) N. znversa, 2) Nitidulidae and 3) a few miscellaneous organisms, as below: 1. N. znversa The eggs were not located. The numbers of larvae, puparia (with pupae or empty) and adults recorded in the spathal chambers of the intact heads with seed are shown in Table 3, and the totals are also inserted in Table 1. The numbers of adults are given in brackets in Table 3, and are not included in the totals, as those individuals are presumably already included in the numbers of empty puparia. TABLE 3 Occurrence of N. inversa and Nitidulidae in 41 heads of A. macrorrhiza HABITAT, Head N. inversa Nitidulidae locality, design- Larvae Puparia Adults Total Larvae Adults site ation with empty pupa No. No. No. No. No. No. No. RAINFOREST Lamington Plateau O’R 1 31 0 0 0 31 3 0 2 2 18 0 0 20 0 3 0 6 11 0* 17 1 0 Mt Glorious F 2 0 0 46 0* 46 n.s.* n.s.* 3 0 0 42 0* 42 n.s.* n.s.* R(L) 2 0 16 10 0* 26 2 0 3 4 25 5 0* 34 2 0 4 8 19 0 0 27 1 0 Proc. Linn. Soc. N.S.W.,106 (1), (1981) 1982 D. E. SHAW, B. K. CANTRELL AND K. J. HOUSTON TABLE 3—continued Occurrence of N. inversa and Nitidulidae in 41 heads of A. macrorrhiza HaBIiTAT, Head N. inversa locality, design- Larvae Puparia site ation with empty pupa No No. No. R(R) 6 2 29 18 7 21 5 ] M 3 35 0 0 4 32 44 1 5 64 11 1 6 0 1 3 7 0 44 1 H 2 ] 2 0 3 16 2 0 4 0 0 2 5 0 4 2 6 0 6 32 INTERMEDIATE ZONES Upper Brookfield Sm ] 0 0 0 2 0 0 0 3 0 3 10 4 0 0 0 6 0 0 5 Beechmont B 1 0 0 0 D) 0 0 0 3 0 0 0 GARDEN Chapel Hill S 1 0 0 0 Mt Coot-tha AR 4 0 0 0 5 0 0 0 ER 6 0 0 0 7 0 0 0 MC 3 0 0 0 4 0 0 0 5 0 0 0 St Lucia NR 1 0 0 0 2 0 0 0 3 0 0 0 4 0 0 0 Adults Total Soe qocose ooo °& Nitidulidae Larvae Adults No. No. 2 0 2 2 3 0 0 0 1 0 2 0 0 0 1 0 9 0 5 0 0 0 ] 0 11 0 0 0 2 0 0 1 0 0 7 0 2 0 0 0 n.s.* n.s.* 0 0 2 0 0 0 0 0 0 0 0 0 0 0 2 0 0 0 0 0 0 0 * Some adults apparently escaped without detection during the opening of the chamber n.s. = not sought 73 # Numbers of adults given in brackets not included in the total, because the individuals already counted in the numbers of empty puparia. All heads sampled from rainforest sites had one or more stages of N. znversa, while in heads from intermediate zones, only two out of nine heads sampled had any stage of the fly, and then only in low numbers. No N. znversa were recorded in heads from garden areas. Because of the number of zero values in the intermediate and garden sites, the Proc. Linn. Soc. N.S.W., 106 (1), (1981) 1982 74 NEUROCHAETA INVERSA McALPINE AND SEED SET data were transformed (x = \/x + 4) and the transformed means show that both the intermediate and garden sites differ significantly at the 5% or 1% levels from the rainforest sites in numbers of N. znversa present (Table 4). The numbers of N. znversa individuals per head, together with average numbers per site, locality and habitat, are shown in Table 1. N. znversa was recorded in all heads at all the rainforest sites, as reported previously, and seed set per head for these ranged from 4.2% to 96.0%. In the intermediate zones, only two heads at the Upper Brookfield site had a small number of N. znversa with 15.7% and 80.3% seed set respectively, while the four heads without N. znversa at this site had 27.5% to 76.6% seed set. None of the three heads from Beechmont had N. znversa, but seed set was 3.5%, 10.1% and 17.0% respectively. In the garden sites, where N. inversa was absent, seed set ranged from 5.5% to 98.1%. This latter figure, in fact, was the highest percentage of seed recorded in this study. TABLE 4 Statzstzcal analysis of average numbers of N. inversa per head per site HABITAT, Site* N. inversa Sites (numbered as in third column locality Design- Num- per head significantly less than designated site ation. ber# per site Av. No.* 5% level 1% level RAINFOREST Lamington O’R 1 22.3 Sites 7, 8, 9, 10, 11, 12 Plateau Mt Glorious F 2 44.0 Sites 6, 7, 8, 9, 10, 11, 12 R(L) 3 28.9 Site 6 Sites 7, 8, 9, 10, 11, 12 R(R) 4 37.2 Site 6 Sites 7, 8, 9, 10, 11, 12 M D) 36.3 Sites, 6, 7, 8, 9, 10, 11, 12 H 6 10.4 Sites 7, 8, 9, 10, 11 Site 12 INTERMEDIATE ZONES U. Brookfield Sm 7 1.7 Beechmont K 8 0.0 GARDEN Mt Coot-tha AR 9 0.0 ER 10 0.0 MC 11 0.0 St Lucia NR 12 0.0 x Garden Site 1 at Chapel Hill omitted because only one sample. Sites numbered consecutively for statistical analysis. Transformed means. A co-variance analysis was carried out on seed number per site using the numbers of N. znversa as a co-variate, and the regression is not significant. 2. Nitedulidae Larvae of Nitidulidae (Coleoptera) and more rarely, adults, were found in low numbers in some heads at most sites (Table 3), particularly at rainforest and Beechmont sites. The larvae mainly occurred in the ms. florets undergoing deliquescence, on the inner face of the spathal tube and only rarely on the berries. 3. Other insects etc. Other insects and organisms found in undamaged, sealed spathal chambers are recorded in Table 5. They were mainly larvae of unknown Diptera, Ceratopogonidae (biting midges) , oligochaete worms and crustaceans (Syncarida: ?Bathynellacea) . Proc. Linn. Soc. N.S.W.,106 (1), (1981) 1982 D. E. SHAW, B. K. CANTRELL AND K. J. HOUSTON 75 TABLE 5 Occurrence of other insects etc. and damaged berries in 41 heads of A. macrorrhiza HABITAT, Head Insects etc. Damaged berries with locality, design - Prick Hole with Wet rot# into site ation lesion* some rot spadix core No. Ge, No. % RAINFOREST Mt Glorious R(L) 2 = 18 19.4 4 7 28 19.7 R(R) 6 He ] 0.6 7 Few Oligochaeta (worms) at 6 4.9" M 3 ie D 1.8 4 a 37 N78 5 3 2 1.0 6 oe I 0. 7 z 2 1.6 H 5 y 6 6 Ceratopogonidae larvae z 4 3.9 29 22 3 Unknown Diptera larvae 2 Diptera puparia 1 dead flying ant GARDEN Mt Coot-tha MC 3 3 dead adult Thripidae > St Lucia NR 1 16 Ceratopogonidae larvae i 4 3.5 1 Aphididae (1 apterous, 1 immature) 1 ? Hymenoptera (prepupa) 2 3 Oligochaeta (worms) = 3 4 Oligochaeta (worms) z 2 le 4 10 (approx). ? Bathynellacea 5 4 (Syncarida) Crustacea Fusarium sp. (fungus) (sp.1) (fine web and spores over the berries) * * = ‘prick’ lesions present; ** = abundant; ~ = absent. # Impossible to determine whether rotted structures were berries or unfertilized ovaries. Rot extending into spadix core. INSECTS IN YOUNG OR DAMAGED HEADS Insects and other organisms in ten other heads which were either too young to have seed, or where the spathal tube was damaged or gaping, are reported in Table 6. N. inversa larvae, puparia and adults were recorded inside the spathal chambers of most of these rainforest heads. Nitidulid larvae and a few adults were found either outside or inside the spathal chambers, in the latter case mainly near or in the decaying m.s. florets at the tip of the chamber. Eggs of Chloropidae (Diptera) were found mainly inside the chambers of heads picked while the spathal limb was still extended and unwilted, while the larvae were found in decaying tissue outside and inside the spathal chambers. Two species (Cadrema sp. and Hippelates sp.) were reared from larvae in the laboratory, and one specimen of each species was parasitized by a cynipid wasp (Hymenoptera: Cynipidae: Eucoilinae). The large (0.6 x 0.24 mm) opaque white eggs on Mt Glorious heads have not been identified. Only one Proc. Linn. Soc. N.S.W., 106 (1), (1981) 1982 76 NEUROCHAETA INVERSA McALPINE AND SEED SET ceratopogonoid larva was recorded on a Mt Glorious head. Brachypterous female Phoridae (Diptera) were found in the decaying tissue of two damaged heads from Mt Coot-tha. TABLE 6 Insects recorded in ten young or damaged heads of A. macrorrhiza, not reported in previous tables HABITAT, Insects, etc. locality Head maturity and position of N. inversa Nitid- Chlor- Miscell- site, head insects ulidae opidae aneous designation No. * No. * No.* No. * RAINFOREST Mt Glorious R(L) Head 5 Very young, spathal limb still present; no difference in ovary sizes In plastic bag after transport 1L 1 Diptera L 4A 1 Staphy- lididae L In debris at tip on outside of spathal tube 11L 20L Inside spathal chamber 33L 1L 98#E 1 Large white €ss Few flatworms? Head 8 Prior to anthesis, held in lab. 18 days; no enlargement of ovaries but constriction closed Inside chamber near tip 1L 20L 28#E 1 Ceratopo- gonidae L 3P 15L 4P 2A R(R) Head 9 Very young, spathal limb still present; held in lab. 4 days; no perceptible difference in ovary sizes In decaying florets outside spathal chamber 1L 3L Inside chamber 2L aan E 1 large white ess 7L Head 10 Very young, spathal limb still present; held in lab. 5 days; hardly perceptible difference in Ovary sizes Inside chamber 40L 97#E 8 large white eggs 24L M Head 1 Head net bagged before anthesis; 1 ovary slightly enlarged F Inside spathal flap 4E Inside spathal chamber 47L (Some ‘prick’ lesions present) 5P Head 8 Head ripe with partially opened spathal tube Inside chamber 8P H Head 7 Head with slight rot of spathal tube tip Proc. Linn. Soc. N.S.W.,106 (1), (1981) 1982 D. E. SHAW, B. K. CANTRELL AND K. J. HOUSTON 77 TABLE 6 — continued Insects recorded in ten young or damaged heads of A. macrorrhiza, not reported in previous tables HABITAT, Insects, etc. locality Head maturity and position of N. inversa Nitid- Chlor- Miscell- site, head insects ulidae opidae aneous designation No. * No. * No. t No. * Inside chamber 2P 4L 2A Head 8 Head with slight rot of spathal tube tip Inside chamber 15P 2P GARDEN Mt Coot-tha MC Head 1 Staminate portion deliberately severed before anthesis; later rest of spadix rotting In rotting florets 3 Phoridae A Head 2 Small hole with some rot into spathal chamber In rotting spadix 19L 15 Phoridae A 1A *E = egg; L = larva; P = puparium; A = adult. # Probably more eggs present but uncounted. + Eggs uncounted. BIOLOGY OF NEUROCHAETA INVERSA The development of N. znversa from larva to adult within the sealed spathal tube of A. macrorrhiza seems fairly precisely adjusted to the maturation of the infructescence. Details of the maturation of the head will be given separately by D.E.S., but the part of the process relevant to the spathal chamber and the chamber liquid may be summarized as follows: The portion of the spadix (consisting of the terminal sterile appendage, the staminate section and the upper part of the m:s. florets) distal to the constriction withers and falls off. The spathal blade also withers and falls off, usually in its entirety. The tip of the spathal tube constricts to form the sealed spathal chamber (Fig. 1, C) which encloses the lower part of the ms. florets and the pistillate flowers or ovaries. The enclosed m.s. florets then undergo deliquescence, leaving the spadix core bare and contributing to a broth of liquid and flower-part remains. This broth bathes the surface and sides of the ovaries in the sealed chamber and the broth debris often accumulates near or at the base of the chamber. In some young heads, the enclosed liquid, which may not all have been derived from the dissolution of the m.s. florets, measured up to 8 cc. Larvae of N. znversa were usually found on the inner face of the spathal tube, in the broth from the spathal chamber (Fig. 1, D) or on the outer faces of the exposed berries and unfertilized ovaries. These latter larvae, however, moved rapidly between the ovaries when the spathal tube was removed, and were usually only retrieved by the removal of the berries and unfertilized ovaries. The development of the larvae coincides with the deliquescence of the m.s. florets, as described above, so that when the broth resulting from the dissolution is abundant, the larvae are present. Not all the larvae in any one head pupariate at exactly the same time, so that in some cases both larvae and puparia occur together in the same chamber (Table 3), often with the remains of the m.s. florets still discernible. However, by the time all the larvae have pupariated the remains of the m.s. florets have virtually disappeared, together with most of the liquid. There is some latitude in this sequence, however, regarding Proc. Linn. Soc. N.S.W., 106 (1), (1981) 1982 78 NEUROCHAETA INVERSA McALPINE AND SEED SET Proc. Linn. Soc. N.S.W., 106 (1), (1981) 1982 D. E. SHAW, B. K. CANTRELL AND K. J. HOUSTON 79 TABLE 7 Infructescence maturity, amount of liquid in spathal chamber and stages of N. inversain heads from Mt Tamborine Site, head Maturity of head Amount of Stages of N. znversa designation (All heads with spathal liquid in Larvae Puparia Adults tubes still green) spathal with empty chamber pupae No. No. No. No. BC Head 2 Berries enlarged, green trace 0 2 11 Da Head 4 Berries larger, green about 0 16 30 295 0.5 cc Head 3 Berries largest, turning about 1 4 7 i yellow at base (therefore lec ripest) * Some adults apparently escaped without detection during the opening of the chamber. maturity of the head, amount of liquid remaining in the chamber, and stage of development of N. znversa. This is shown, for example, in three heads from Mt Tamborine (Table 7), not included in the other tables. Head 3 from Mt Tamborine was the only one, out of over 50 heads studied, where all three stages (larva, puparia and adults) occurred at the same time. Although its berries were more mature than those on Heads 2 and 4, its chamber contained more liquid than the other two chambers, and this may have contributed to the presence of the larva. The puparia were oriented with the posterior end downwards between the berries, and with the flattened or slightly depressed facet at the anterior end usually parallel to the surface of the spathal tube (Fig. 1, E). In a few cases puparia were found between the base of the ovaries, probably dislodged during transport of the head from the field. McAlpine (1978) reported that the adults which emerge from the puparia cannot escape from the spathal chamber until the fruits are ripe and the spathal tube splits. In this study, adults were present in Heads 5 and 6, Site H, and in Head 7, Site M, at Mt Glorious (Table 3) and also in all three heads from Mt Tamborine (Table 7), and in all cases the spathal tubes were still green, without any yellowing which indicates approaching maturity and which precedes splitting of the tube. The adults were quick to escape from any artificial opening made in the tube, and perhaps escape naturally as soon as the slighest relaxation occurs in the tube, even possibly prior to splitting. Information is still required as to whether the adults feed after emergence while still within the chamber, and whether they remain quiescent between the berry shoulders, or make their way to the tip of the chamber and congregate around the spadix core bared by the dissolution of the m.s. florets, waiting for the first opportunity to escape. N. inversa larvae were never noted causing damage to the ovaries or to the spadix core. No parasites were recorded in any of the specimens. Fig. 1. A, Spathe (left, detached) with spathal blade (sb) and spathal tube (st) and spadix (right) of Alocasia macrorrhiza. B, Later stage of the spadix showing sterile terminal appendage (sta) ; staminate flowers (s); median sterile flowers (msf) and pistillate flowers (p). C, ‘Sealed’ spathal tube enclosing berries. D, Diptera larvae and ‘broth’ from spathal chamber. E, Puparia of Neurochaeta inversa and berries of A. macrorrhiza. F, Tip of spathal tube with rot and sporulating fungus (Fusarzum sp.). Proc. Linn. Soc. N.S.W., 106 (1), (1981) 1982 80 NEUROCHAETA INVERSA McALPINE AND SEED SET DAMAGE TO INFRUCTESCENCE PARTS Damage to berries in sealed, intact heads was of three types, viz., 1) ‘prick’ lesions, 2) hole in the berry wall, and 3) wet rot of the berry. Rot of the spathal tube tip (point 4. below), resulting in gaping, occurred in three other heads. 1. ‘Prick’ lesions In some heads (Table 5) at sites at Mt Glorious, minute red ‘prick’ lesions were noted on some berry walls. In a few cases a narrow border of slightly more translucent tissue surrounded the ‘prick’, but even then the whole diameter of the lesion was still only about 0.5 mm. In no case did any rot of the berry wall or the underlying seeds appear to be associated with these ‘prick’ lesions. The lesions may have been caused by stylet penetration of the pistillate flowers before anthesis by an unknown insect. 2. Hole in berry A hole in the berry wall, sometimes with damage to the enclosed seed, occurred in 14 berries from Mt Glorious and 10 from St Lucia (Table 5). The hole was usually near the base of the berry, and such damage tended to cause a premature change in the colour of the berry wall from green to yellow and even to red. In some cases the damage extended into the spadix core, causing a reddish wet rot at the site of berry attachment. Although larvae of Ceratopogonidae were occasionally found feeding within the holes in the berries, this may merely indicate opportunism and not necessarily a causal relationship. 3. Wet rot of ovaries Wet rot of 116 ovaries (or berries, as it was impossible to determine whether fertilization had taken place or not) was noted in six heads from Mt Glorious (Table 5), one case involving 22.7% of the ovaries on the head. If adjacent ovaries were affected, counts were made on stigmas rather than on the ovaries themselves, as it was difficult to distinguish the individual rotting bodies. In each case the rot extended into the spadix core, again with a wet reddish discolouration of the tissue. It is not known whether the wet rot is an extension of the ‘hole’ condition described above, or whether it is quite distinct. The latter may be the case, as no wet rot was recorded in the St Lucia heads, whereas holes did occur in 10 berries on three heads from this site. 4. Rot of tip of spathal tube Three heads, including Heads 7 and 8 from Site H, Mt Glorious (Table 6), and an unlisted pathology specimen head from Site R(R) at Mt Glorious, had a rot of the spathal tube beginning at the tip, and proceeding downwards, with white sporulating fungal clumps on the rotted tissue (Fig. 1, F). The fungus, which was obtained in axenic culture, was Fusarium sp. 2 (see Addendum), different from Fusarzum sp. 1 recorded as a fine sporulating web on Head 4, St. Lucia (Table 5). Further investigation 1s required to determine whether Fusarzum sp. is a primary pathogen, or, as would seem more likely, a secondary invader following damage by an unknown agent. GENERAL DISCUSSION McAlpine (1978) reported that larvae of Cadrema sp. (Diptera: Chloropidae) were found in the moist decaying upper part of the spadix of A. macrorrhiza, and has confirmed it (pers. comm.) to be the same species as that found in the present study. We also found larvae of Hippelates sp. (Chloropidae) and Nitidulidae occupying that niche provided by the dissolution of the m.s. florets at the tip of the spathal chamber and in the decaying spadix parts outside the spathal chamber. It is not known if they also occur on the decaying distal portion of the spadix which usually withers and falls off the infructescence. If so, the life cycle of these two groups may be completed in the decaying tissue on the ground. Proc. Linn. Soc. N.S.W., 106 (1), (1981) 1982 D. E. SHAW, B. K. CANTRELL AND K. J. HOUSTON 81 Some of the miscellaneous organisms found in the sealed chambers, such as the Ceratopogonidae (biting midges), the oligochaete worms and the ?Bathynellacea (Crustacea: Syncarida) may be chance visitors caught within the chambers before sealing. Oligochaete worms and midges were only recorded in sealed intact heads from St Lucia and Mt Glorious. The worms, however, are difficult to see, and low numbers may have been missed in heads from other sites. The presence of the worms seemed to be associated with thin rings of debris on the inner surface of the spathal tube, although it was not possible to confirm the connection. One of the interesting aspects revealed by this study was the considerable variation in seed set on different heads at the one site (Table 1). Whether this was the result of differences in numbers of the pollinating agent(s), or in the amount of available pollen, or in microclimatic factors at the time of stigma receptivity of each head, is not known. McAlpine (1978) stated that ovipositing N. znversa females may well be polli- nators of A. macrorrhiza, but observed that, as many insects visit the flowers, they are probably not the sole pollinating agents. As shown in Table 1, the numbers of seed set were not related to the numbers of N. znversa recorded in the chambers at the sites where the fly was found. McAlpine (1978) further observed that N. znversa had not been found on cultivated examples of ‘cunjevoi’ and this was confirmed in the present study, as no N. znversa were recorded at these sites, although seed was still set as shown in Table 1. We conclude, therefore, that pollination of A. macrorrhiza is probably independent of ovipositing females of N. znversa in the inflorescence. Carson and Okada (1980) previously noted that there appeared to be no difference in seed set of flowers of Colocasza esculenta (Araceae) from which adult Drosophiella pzstilicola were excluded by bagging, and unbagged controls. They concluded, therefore, that the presence of. this fly was not necessary for full pollination of the flowers. Their work, like that of the present authors, was concerned specifically with an insect associated with the sealed spathal chambers, and not with the elucidation of the roles which may be played by many other insects known to visit the exposed flower parts. ACKNOWLEDGEMENTS The owners of the sites sampled and other individuals who assisted in obtaining the collections, including Mr and Mrs E. J. Frazer, Ms J. Grimshaw, Mrs J. Henry, Mr A. Hiller, Mrs J. Hope, Ms K. Howdesell, Mrs B. Kennedy, Mr and Mrs L. J. Manning, Mr M. Olsen, Mr P. O'Reilly, Mr and Mrs D. Sands, Mr and Mrs W. H. Smith, and Mr H. Caulfield and Mr J. Donnelly (Mt Coot-tha Botanic Gardens) and Dr A. B. Cribb, Dr D. Priest and Mr A. R. Steginga (University of Queensland) are thanked for their co-operation in this study. We are also grateful to Dr P. J. F. Davie, Mr J. F. Donaldson, Dr I. D. Galloway, Mrs M. M. Harris and Dr D. K. McAlpine for identifications, to Ms J. Alder for the statistical analyses, and to the Director, Plant Pathology Branch, Department of Primary Industries, Indooroopilly, for facilities to D. E. S. during the study. References Carson, H. L., and Oxapa, T., 1980. — Drosophilidae associated with flowers in Papua New Guinea. 1. Colocasia esculenta. Kontyu, Tokyo 47: 15-29. McALPINE, D. K., 1978. — Description and biology of a new genus of flies related to Anthoclusea and representing a new family (Diptera, Schizophora, Neurochaetidae). Ann. Natal. Mus. 23: 273-295. Proc. Linn. Soc. N.S.W., 106 (1), (1981) 1982 82 NEUROCHAETA INVERSA McALPINE AND SEED SET ADDENDUM Identifications of Fusarzum spp. 1 and 2, mentioned in the section ‘Damage to Infructescence Parts’, now received from the Commonwealth Mycological Institute, United Kingdom, are Fusartwm solani (Mart.) Sacc. and F. oxysporum Schlecht., respectively. Proc. Linn. Soc. N.S.W., 106 (1), (1981) 1982 1/06 C/a) ¢ The Fauna of Australian Mangroves P. A. HUTCHINGS and H. F. RECHER Hurcuincs, P. A., & RECHER, H. F. The fauna of Australian mangroves. Proc. Linn. Soc. N.S.W. \@@€®, (1981) 1982: 83-121. Mangrove forests are among: the world’s most productive ecosystems. In Australia, mangroves have an extensive distribution and are probably important for the maintenance of estuarine fisheries. Despite these values, there has been relatively little research on the ecology of Australian mangrove communities. Botanical studies have been most extensive, work on mangrove fauna has been largely restricted to a few commercially-important or pest species. Many of the data available are non- quantitative, anecdotal or unpublished. This lack of data on mangrove fauna prevents the development of detailed plans of management and therefore poses long-term conservation problems. In order to identify areas where research on Australian mangrove fauna is needed and, it is hoped, to stimulate such work we summarize information on the Australian mangrove ecosystem and discuss the structure and evolution of these communities. P. A. Hutchings, Department of Marine Invertebrates, and H. F. Recher, Department of Terrestrial Ecology, Australian Museum, P.O. Box A285, Sydney South, Australia 2000; manuscript received 19 May 1981, accepted in revised form 21 October 1981. INTRODUCTION Mangroves are among the most productive of the world’s forests (Westlake, 1963; Lugo and Snedaker, 1974) and contribute importantly to the productivity of tropical and sub-tropical estuaries. Working in southern Florida, Odum and Heald (1975) demonstrated that the organic matter produced by mangroves formed the base of a detritus food chain that culminated in the rich fisheries of Florida Bay. Mangroves occur around the coast of Australia (Lear and Turner, 1977) (Fig. 1). In the estuaries of northern New South Wales, Queensland, the Northern Territory and parts of Western Australia, mangroves form extensive forests which, with seagrasses and salt marshes produce the organic matter that is the base of the fisheries of northern Australia. Many of Australia’s commercial fish species feed on mangrove fauna or use the mangroves as a nursery (Newell and Barber, 1975; Pollard, 1976, 1981). Crab, prawn and oyster are other fisheries intimately associated with mangroves (Ruello, 1973; Staples, 1980 a,b). Despite their ecological and economic importance, research on Australian mangroves has been largely botanical, describing vegetation (MacNae, 1966, 1967; Chapman, 1975; Saenger et al., 1977; Bunt and Williams, 1980, 1981; Williams and Bunt, 1980), floral biology (Clarke and Hannon, 1970, 1971; Duke and Bunt, 1979; Tomlinson et al., 1978, 1979), geomorphology in relation to distribution of mangroves (Thom e¢ al., 1975) and mangrove physiology (Clough and Andrews, 1982). Faunal studies have mostly been restricted to surveys of species present in mangroves (e.g. Hutchings and Recher, 1974; Saenger et al., 1977). Studies on ecosystem dynamics have been initiated (Bunt e¢ al., 1979; Clough and Attiwill, 1982, a, b; Goulter and Allaway, 1979), but there is little information on mangrove productivity or the pathways by which energy flows from mangroves through Australian estuaries. The information that is available tends to support the findings of Odum and Heald (1975). Proc. Linn. Soc. N.S.W., 106 (1), (1981) 1982 84 FAUNA OF AUSTRALIAN MANGROVES 12 7] 3 NO. OF MANGROVE ee, _TROPIC OF CAPRICORN _ SPECIES (ene alae SS ae 38 eee 24 ONE SPECIES 1 20-29 30 ~A 10- 20 SE IONS oe ee ae oe : ONE SPECIES 36 ONE SPECIES Fig. 1. The number of mangrove species increases from temperate areas to the tropics. Only one species, Avicennia marina, occurs around the coast from South Australia to Cape York. With increasing species number, there is a concomitant increase in community complexity. Odum and Heald (1975) examined the food webs in a simple estuarine mangrove community, Rhzzophora mangle, in southern Florida. The community was characterized by a large vascular plant production with little input from plankton or algae. Most of the leaves and stems produced by the mangroves (Duke et al., 1982; Williams et al., 1982) were not eaten by herbivores, but became part of the decaying matter (‘litter’) on the forest floor. Mangrove detritus formed the basic component in the diet of primary marine consumers (‘detritivores’). Among the primary consumers studied by Odum and Heald, detritus of vascular plant origin formed 20 per cent of the diet. The authors concluded that mangrove leaf detritus was the most important element in the food web. However, Redfield (1982) has presented a modified version of their model with increased emphasis on the consumption of mangrove leaves by insects (see Onuf et al., 1977) and mud dwelling herbivores (see Malley, 1978) and decreased emphasis on the importance of bacteria and fungi in the breakdown of mangrove detritus. As in all communities, animals are an integral part of the mangrove ecosystem. Herbivores and detritivores living in the mangrove forest have been shown to be important in the transfer of energy from the mangroves to the estuary (Goulter and Allaway, 1979; Odum and Heald, 1972; Malley,1978). In the absence of detailed information about the fauna of Australian mangroves, models of local mangrove ecosystems must be considered incomplete. In turn, the lack of data on the mangrove ecosystem, its fauna and the relation of these to the estuary seriously hampers the conservation of mangrove forests and the management of estuarine fisheries. As we have previously pointed out (Hutchings and Recher, 1977; Recher and Hutchings, 1980), mangroves are a threatened habitat throughout eastern and southern Australia. The development of a national conservation strategy for mangroves and Proc. LINN. Soc. N.S.W., 106 (1), (1981) 1982 P. A. HUTCHINGS AND H. F. RECHER 85 their management is of critical importance for the sustained production of estuarine and near-shore fisheries (Hutchings and Recher, 1977). In this paper we review what is known of the mangrove fauna in Australia. Data are often sketchy or anecdotal and much of the information presented here is as yet unpublished. We do not apologize for these deficiencies; our intention is to focus future research work on areas of greatest need. MANGROVE FAUNA The mangrove fauna is composed of animals from terrestrial, marine and freshwater environments. It is a diverse fauna with an abundance of vertebrate and invertebrate forms. Some are restricted to mangroves; others occur more widely. For many species the full range of habitats used is unknown and species currently classed as Mangrove endemics may occur elsewhere. Some animals are only temporary residents. Birds may use mangroves for roosting, fish move into and out of the mangroves with the tides, others come in seasonally like flying foxes (Pteropus spp.) to feed on nectar. In this review we have ignored records of animals which do not occur regularly in mangroves. VERTEBRATES Few vertebrates are restricted to mangroves. The majority are casual visitors or accidental occurrences. Saenger et al. (1977), for example, record over 200 species of birds from Australian mangrove habitats, but fewer than half occur regularly (Schodde et al., 1982). Mammals No mammal is endemic to mangroves in Australia, nor is there any species for which mangroves are the principal habitat. The rat, Xeromys myozdes, has been recorded living among mangroves on the Tomkinson River, Arnhem Land, and on Andranang Creek, Melville Island in the Northern Territory (Magnusson et al., 1976). It has also been recorded as frequenting mangroves near Mackay in central Queensland and on North Stradbroke Island in southern Queensland (Van Dyck et al., 1979). In these situations it builds nests of leaves and mud among the mangrove roots of Bruguzera; the nests are flooded on the highest tides (Magnusson et al., 1976). Xeromys feeds on crabs and is at least partially arboreal, but also occurs in a wide range of habitats other than mangroves (R. Strahan, pers. comm.) . Many other terrestrial animals enter mangroves and may obtain part of their food there. Frith (1973) records a number of rodents (Rattus colletti, Mus musculus, Melomys spp., Mesembriomys spp. and Conilurus spp.), bandicoots (Perameles, Isoodon) and the northern brush-tailed possum Tr¢chosurus arnhemensis as occurring in mangroves. The water rat Hydomys chrysogaster forages regularly in mangroves (Frith, 1973; Recher, pers. obs.) . Two species of flying foxes, Pteropus poliocephalus and P. alecto commonly come into the mangroves to feed on nectar or to camp. P. poliocephalus occurs along the entire east Australian coast whereas P. alecto is restricted to tropical mangroves. Another species, P. conspicillatus, sometimes camps in mangroves. Some species of tree-roosting bats such as Tadarzda planiceps, the flat- headed mastiff-bat and T. loriae, the little northern mastiff-bat, are occasionally found in mangroves. The northern blossom bat, Macroglossus lagochilus occurs in stands of Sonneratza alba in Western Australia. Mangroves do not appear to be used by marine mammals. Heinsohn and Wake (1976) stress the importance of sheltered waters for dugong Dugong dugon and the association of dugong with seagrass beds but make no mention of dugong using Proc. Linn. Soc. N.S.W., 106 (1), (1981) 1982 86 FAUNA OF AUSTRALIAN MANGROVES mangroves. Recently dugongs and dolphins have been seen in the mangrove channels on Hinchinbrook Island (Bunt, pers. comm.). Birds Birds are a conspicuous component of all mangrove forests, although they are not abundant. Schodde et al. (1982) suggest that the comparative uniformity of structure within the canopy of mangrove forests, which provides little variety of foraging surfaces, accounts for the small number of individuals. In many regions, mangroves have a fragmented distribution which Schodde et al. suggest may also be important in limiting numbers. Feeding time for ground foragers is restricted by tidal flooding. Nest sites may be limited. These reasons may also explain why many birds occurring in mangroves, also occur in other habitats. Data are not available on whether these species occupy similar niches in all habitats or if they broaden their niche in the mangrove forest. Over 200 bird species have been recorded from Australian mangroves (Saenger et al., 1977). Of these, 14 species are virtually restricted to mangroves; 12 species use mangroves as a primary habitat in part of their range, and 60 species use mangroves regularly throughout the year or in particular seasons (Schodde et al., 1982). This is a rich mangrove bird fauna compared with other parts of the world (MacNae, 1968). The number of species of birds using mangroves and the number of mangrove specialists increases from south to north with increasing floristic richness (Fig. 2). Thus near Sydney (34°S) only one species, the mangrove heron Butorides striatus, is restricted to mangroves. The other birds in temperate mangroves are widely distributed in a variety of aquatic or forest habitats (e.g. rufous whistler Pachycephala rufiventris and white-face heron Ardea novaehollandiae). In contrast, on the Cape York Peninsula (19°S) nineteen species of birds are considered mangrove specialists and seven of these are endemic to mangrove habitats (Schodde et al., 1982). Birds which are wholly or partially dependent on mangroves include two herons, a kite, a rail, a pigeon, a cuckoo, three kingfishers, five pachycephaline flycatchers, four myiagrine flycatchers, three acanthizine warblers, three honeyeaters, a silvereye and a butcher-bird. None shows major morphological adaptations to the mangrove environment. However there is a tendency for bills to be longer than in non-mangrove congeners (Schodde et al., 1982) . The white-breasted whistler Pachycephala laniozdes has a comparatively heavy and hooked bill for handling marine invertebrates. Birds which regularly use mangroves, but which cannot be considered mangrove specialists, include aquatic species and a wide range of passerines. Aquatic and wading birds (e.g. herons, ibis, sandpipers) forage among mangroves and many use mangroves as high tide roosts. Some species (e.g. pied cormorant Phalacrocorax varius, straw-necked ibis Threskzornis aethiopicus) nest in mangroves (White, 1917; Seton, 1971; Schodde et al., 1982). Among the passerines, honeyeaters commonly visit mangroves for nectar when blossom is abundant. The sulphur crested cockatoo Cacatua galerita has been seen feeding on the fruits of Lumnitzera littorea in north Queensland by Bunt and his co-workers (pers. comm.). Other birds (e.g. rufous fantail Rhipidura rufifrons and grey fantail R. fuliginosa) frequent mangroves during migration. Reptiles Although they may be important feeding grounds, for the majority of terrestrial reptiles using mangroves, mangroves are peripheral habitats. In some areas, mangroves are the only habitable forest and may, therefore, be important corridors for the movement of individuals. Reptiles are common in tropical mangroves, but are rarely seen in temperate Proc. Linn. Soc. N.S.W., 106 (1), (1981) 1982 P. A. HUTCHINGS AND H. F. RECHER 87 35 (10 ) tae 45 (8) Faunal areas based on terrestrial birds ( after Kikkawa & Pearse, 1969 ) No. of bird species ( no. of species confined to or most abundant in mangroves ) Total for Australia is ~ 58 (14) species of terrestrial birds Fig. 2. Kikkawa and Pearse (1969) divided Australia into regions based on the similarity of the bird fauna. These regions are convenient to illustrate the number of species of birds which occur regularly in mangroves. The number of bird species endemic to mangroves in each region are shown in parentheses. Species number and the number of endemics increase from south to north. This latitudinal gradient in bird species diversity is correlated with an increasing area of mangroves and the richer floristic and structural complexity of tropical mangrove forests. ones. The salt water crocodile Crocodylus porosus is probably the best known mangrove resident; a notoriety derived from its occasional consumption of humans. Salt water crocodiles occur in river systems from Broome in Western Australia, around the Northern Territory and south to Maryborough in Queensland (Cogger, 1979). The animal is most abundant in the tropics and comes into mangroves to feed. Juvenile crocodiles take crabs (especially sesarmids) , prawns, mud skippers, and small fish. Larger animals take mud crabs Scylla, birds, and mammals including Xeromys and flying foxes (Taylor, 1979). Crocodiles do not nest in mangroves (Webb et al., 1977). The fresh water crocodile C. johnstond occurs rarely in mangroves and is probably accidental at the landward margins of the mangroves and in less saline areas. Cogger (1979) has recorded the pitted-shelled turtle Carettochelys insculpta in the Daly, Victoria and Alligator River systems in the Northern Territory where it may enter mangrove areas. A small number of snakes and lizards occur regularly in mangroves. Mangroves Proc. Linn. Soc. N.S.W., 106 (1), (1981) 1982 88 FAUNA OF AUSTRALIAN MANGROVES are the primary habitat for the mangrove monitor Varanus ¢ndicus which occurs in the Northern Territory, Torres Strait and Cape York (Cogger, 1979). It feeds on crabs and mud skippers. Four other goannas, V. semiremex, V. prasinus, V. ttmorensis and V. tristts are common in tropical mangroves, but range widely in other habitats (Cogger, 1979). The littoral skink Emoza atrocostata is common in mangrove areas of the Torres Strait and northern Cape York Peninsula (Cogger, 1979). The Northern Water Dragon (Lophognathus temporalis) feeds on insects in the mangroves of northwest Australia and the Northern Territory (Swanson, 1976). Several species of pythons (Lzaszs fuscus, L. oltvaceus) occasionally use mangroves and are attracted by large camps of flying foxes on which they feed. The carpet snake (Morelia spilotes) and Liaszs amethystinus and two common colubrid snakes, the green tree snake Dendrelaphis punctulatus and brown tree snake Bozga rregularts also occur in mangrove habitats. The file snake Acrochordus granulatus is virtually restricted to mangroves in northeastern and northern Australia. It lives on the mud flats in front of the mangroves where it feeds on crabs and fish (Cogger, 1979). Mangroves are also an important habitat for the bockadam Cerberus rhynchops, the white-bellied mangrove snake Fordonza leucobalia, and the mangrove snake Myron richardsoni which occur in northern Australia (Cogger, 1979.). Some sea snakes are estuarine and occur commonly in mangroves (Cogger, 1979.). Ephalophis mertoni, E. greyt and Hydrelaps darwiniensis are found in mangroves in the Northern Territory and Hydrophis elegans and Aipysurus eydouxw are found in central Queensland. Amphibians Two frogs have been reported from mangrove habitats in the Northern Territory. The northern dwarf tree frog Lztorza bicolor has been recorded on the Wildman River, in an area greatly affected by freshwater, and the marbled frog Lymnodynastes convexiusculus occurs in the Cerzops tagal zone of the South Alligator River (Hegerl et Glo, UDI) Fishes Fishes are a conspicuous component of the mangrove ecosystem with large numbers invading the mangrove forest at high tide and retreating to deeper waters as the tide falls. Either as juveniles or adults, virtually all the common fishes of the estuaries can be found in mangroves (Ellway and Hegerl, 1972; Shine et al., 1973; Saenger et al., 1977; Beumer, 1978; Blaber, 1980). Few species are restricted to mangroves. The major group of fishes adapted to living in mangroves belong to the gobiid sub-family Oxcidercinae. This includes mud skippers which are represented in Australia by five species in two genera, Perrophthalmus and Perzophthalmodon. Mud skippers are common throughout the tropics but are absent from temperate mangroves. They occur only as far south as Hervey Bay in Queensland (25°S). Other representatives of this family are the genera Boleophthalmus and Scartelaos which burrow in the mud and occur in the Northern Territory and Queensland. Other gobies are common in mangroves, but are also found elsewhere in the estuary. Included are the genera Mugilogobius, Taenzoides and Arenigobzus which occur throughout Australia. The oyster blenny Omobranchus occurs in frontal mangroves where residual water is trapped. All these fish can stand some exposure to air. Most live in burrows where some water is present even during low tide. A diverse fish fauna occurs in creeks or lagoons in the mangroves where permanent water occurs. These will be estuarine species, characteristic of the geographical region (Saenger et al., 1977; Ellway and Hegerl, 1972 and Shine et al., Proc. Linn. Soc. N.S.W., 106 (1), (1981) 1982 P. A. HUTCHINGS AND H. F. RECHER 89 1973). Among the more common estuarine species which frequent mangroves are toados Torquiger hamiltoni, mullet Myxus elongatus, and fortesques Centropogon australis. Blaber (1980) has designated feeding categories of the fish of Trinity Inlet, Cairns. Beumer (1978) has studied the feeding ecology of black bream (Acanthopagrus berda), bony bream (Anodontostoma chacunda), hair fin goby (Ctenogobius crineger) and milk-spotted toad fish (Chelonodon patoca) in a mangrove creek in the Townsville region. INVERTEBRATES There are two components to the invertebrate fauna of mangrove communities, spiders and insects associated with the forest canopy and aquatic animals inhabiting the intertidal. This fauna is richer than the vertebrate community. Our approach in reviewing the information available on mangrove invertebrates has been to use a classification based on taxonomic and ecological attributes. We first consider insects and spiders, but then review the aquatic fauna according to where they occur in the mangroves. Insects Although neither the insects nor spiders of mangroves have been extensively studied in Australia there is a great variety of species. These show a variable degree of dependence on mangroves. Whether the diversity in mangroves is similar to that in other forest habitats at the same latitude or whether mangrove forests have an array of mangrove specialists is unknown. C. Smithers (pers. comm.) suggests that each species of mangrove probably will have an associated unique suite of insect species. In general, insects in mangroves appear to have the same range of adaptations that they show in other forest environments. For many species, mangroves are an important habitat and insects may be favoured by the often large, almost mono- specific stands of trees within the forest. A comprehensive list of insects from a temperate mangrove system is provided in the Australian Littoral Society Report of Towra Point, Botany Bay (1977). The only detailed study of the insect and spider communities of mangroves has been carried out by Simberloff and Wilson (1970) and Wilson and Simberloff (1969a, b) in the Florida Keys and more recently in north Queensland*. Simberloff and Wilson fumigated small mangrove islands and observed the recolonization of the islands with time. Most work in Australia on mangrove insects has been on species of potential commercial importance (termites) or which may be involved in the transmission of disease (mosquitoes, biting midges) . Termites are an important component of tropical mangroves. Three species of Kalotermitidae occur in the Darwin region: Neotermes insularis, Cryptotermes secundus and C. domesticus. C. domesticus is widely distributed in South-East Asia and the Western Pacific and occurs in both natural and man-made habitats. Whether the mangrove populations in Australia are indigenous, or were introduced is not known. Cryptotermes secundus has also been recorded from Rhzzophora at Kalumburu in Western Australia, and from Avecennza at Groote Eylandt. At Cooloola in southern Queensland Cryptotermes primus occurs in species of Avecennza and *The extensive insect collections made by Simberloff on Hinchinbrook Island have been deposited with Dr R. Taylor, C.S.I.R.O. Proc. Linn. Soc. N.S.W., 106 (1), (1981) 1982 90 FAUNA OF AUSTRALIAN MANGROVES Rhizophora. At Weipa, mangroves support large colonies of Inczsttermes barrettz. Another species, Nasutztermes graveolus, is common on the landward fringe. This termite builds arboreal nests externally on the trunk or branches. Some species may range below high tide level. Mastotermes darwiniensts has been recorded in this situation on Cerzops in the Darwin region, but this is unusual (Miller and Watson, pers. comm.). The location of this termite at high tide is unknown. It may nest in the forest canopy. The same species has been recorded from Corio Bay, Queensland, at the southern limit of distribution of Mastotermes on the Queensland coast (Ellway, 1974). The grasshopper Valanga irregularis has been seen feeding in the mangrove canopy at Corio Bay (Ellway, 1974). The pygmy grasshopper Coptotettzx mastricatus has been recorded on the mud of mangroves. In the New World mangroves, many species of crickets and a number of tettigoniids occur (Rentz, pers. comm.). The beetle fauna of mangroves is diverse and largely undescribed. It probably includes species which feed on dead wood as larvae, others whose larvae feed on mangrove seed pods, while the adults feed on foliage. Some weevils have been recorded from mangrove seed capsules (Zimmerman, pers. comm.). MacNae (1968) records coccids covering the leaves and young twigs of Rhizophoracea (species not given) in north Queensland. Diptera occur in mangroves as both free living and parasitic forms and are a diverse component of the mangrove fauna. Fourteen species were collected by Hutchings and Recher (1974) at Careel Bay, near Sydney. Some Diptera, such as Melanagromyza avicenniae which is restricted to mangroves, appear to feed on new shoots. Diptera are also associated with rotting wood. The adults and larvae of Copidita nigronotata are secondary timber borers. Other dipterans specialize on the fruits and propagules of mangroves (see Hutchings and Recher, 1974, and Spencer, WIT) « According to Common and Waterhouse (1972) several species of butterflies are commonly associated with mangroves. All belong to the Lycaenidae (blue butterflies). Hypochrysops epicurus which occurs from Port Macquarie to Brisbane, appears to be restricted to stands of Avicennza marina. Hypochrysops apelles and H. narcissus are characteristic elements of the mangrove fauna, although their larvae also feed on non-mangrove species. Hypochrysops apelles which occurs along the east Australian coast from north of Yeppoon, feeds on Rhizophora stylosa, Bruguzera gymnorhiza, Ceriops tagal and Avicennia marina. Ogyris amaryllis hewttsonz which occurs from Maryborough to Cairns, is common around mangroves, as its larvae feed on the mistletoe A myema mackayense which grows on mangroves. A less common butterfly Nacaduba kurava occurs from the Richmond River (New South Wales) to Cape York. It feeds on Aegiceras corniculatum but also uses other plants. The larvae and pupae of Acrodtpsas illidgez have been found in the nests of the ant Crematogaster laeviceps near Brisbane. The ant nests in hollow branches of mangroves. Three to four larvae or pupae of Acrodipsas were present in each colony; it seems likely that the larvae of A. zlledgez feed on the immature stages of the ant (Common, pers. comm.). Similarly, Hypochrysops appollo is associated with ants in mangroves, as it breeds in ant house plants (Saenger et al., 1977). The larvae of the tortricid moth Procalyptis parooptera feed on the leaves of Ceriops tagal at Yeppoon, firmly joining adjacent leaves with silk to form a shelter in which the larvae live. Pupation occurs in this shelter. This empty shelter is then used by the larvae of the butterfly Hypochrysops apollo as a daytime retreat where it is attended by ants and from which they emerge at night to feed on the surrounding foliage (Common and Waterhouse, 1972). The moth Macrocyttara expressa Proc. LINN. Soc. N.S.W., 106 (1), (1981) 1982 P. A. HUTCHINGS AND H. F. RECHER 9) (Cossidae) is restricted to mangroves. The larvae of this species tunnel gregariously in the trunks of Excoecaria agallocha. The larvae of Cenoloba_ obliteralis (Oxychirotidae) develop in the cotyledons of fallen seeds of Avicennia marina (Common, 1970). All the Lepidoptera mentioned occur in eastern Australia but other species closely associated with mangroves probably occur in other regions. Two unidentified species have been recorded from South Australian mangroves (Butler et al., 1977). The ant fauna of Australian mangroves has not been studied in detail but many species occur. No species has been found which is totally restricted to mangroves. For ants, colonization of mangroves means coping with the tidal cycle. Thus most mangrove ants are arboreal and several species nest in hollow twigs. Dead mangroves provide shelter for three species of ants (not identified) in Corio Bay, Queensland (Ellway, 1974). Some species nest in the canopy while other ants nest on the ground where they are inundated by the tide (Taylor, pers. comm.). Overseas, the ant fauna in the Florida Keys has been well studied by Wilson (1964). MacNae (1968) records the weaver ant Oecophylla smaragdina in Bruguiera, Cerzops and Sonneratia forests in north Queensland. This species has a similar life history to species of Oecophylla occurring in coconut palm forests in Zanzibar which have been described by Vanderplank (1960). MacNae also suggests that species of ants which occur below high tide level may be able to trap air in their burrows with plugs of mud. Whether ants living in inland areas where freshwater flooding regularly occurs, adopt a similar strategy is unknown. If they do not, then the ants found in mangrove muds have developed a unique adaptation enabling them to invade the intertidal environment. An unidentified species occurs at low tide in the Hinchinbrook region (Hutchings, pers. obs.) and presumably adopts this strategy. Another unidentified species at Hinchinbrook has been seen living in tunnels made by Teredo in the prop roots of Rhizophora. Two species of ant house plants Myrmecodia antoinim and Hydnophytum formicarum, occur in northeastern Queensland. Their tuberous stems are hollow and often used by ants of several species. The ant Phezdole myrmecodizae is restricted to this habitat, where it obtains protection from desiccation and predation. The advantages to the plants are less clear, but the ants may protect them from herbivores. In contrast to other groups of insects, there is considerable information about the mosquito (Lee et al., 1980; Griffiths, in ms.) and biting midge fauna of mangroves. Mosquitoes are typically found in the ephemeral pools at the back of the mangroves, rather than within the mangroves where there is regular tidal exchange. In disturbed mangroves, stagnant pools may be created as a result of dredge and fill operations. These pools are ideal mosquito habitat. The three most common mosquito species occurring in Australian mangroves are Aedes vigilax, A. alternans and Culex sttiens. The distribution of A. wgzlax is largely coastal, with a special relationship to the low- lying land of mangroves and mangrove zones (Sinclair, 1976) . It also occurs inland. Typically, A. wgilax breeds in temporary waters and is intermittent in occurrence. The water must be stagnant and exposed to the sun. Sunlight is necessary for the growth of the algal plankton on which the larvae feed. Reproduction occurs largely during the summer wet season. Aedes can complete larval development within 10 days. Low temperatures inhibit breeding and greatly prolong the duration of the larval stages (Iyengar, 1965; Griffiths, in ms. and pers. comm.). Aedes alternans occurs in all mainland States, and is particularly associated with estuarine areas. It also occurs inland. A. alternans breeds in temporary stagnant pools, both fresh and brackish, exposed to the sun. The eggs are laid singly (probably on mud at the edge of drying pools) and can withstand drying. The eggs hatch when Proc. Linn. Soc. N.S.W., 106 (1), (1981) 1982 92 FAUNA OF AUSTRALIAN MANGROVES the depression fills with water (Marks, 1967, 1971). The eggs take seven days to hatch under favourable conditions but below 21°C hatching is delayed. In coastal situations the larvae of A. alternans are often found with those of A. wgzlax. The larvae of A. alternans are predacious from the second instar onwards on other mosquito larvae. The adults are strong fliers, and are known to migrate with A. wgzlax, but only in the summer (Hamlyn-Harris, 1933). Culex sittens occurs from New South Wales northwards, and westwards to Western Australia. In the Northern Territory, Hill (1917) found C. sztzens abundant near the coast and inland as far as 55 km south of Darwin, breeding in pools, hollow stumps, tins, crab holes, mangrove swamps, weedy lagoons, inland waterholes and shallow wet-season accumulations of water on grassland. Marks (1953) also found it in a variety of habitats, in sunlit, muddy, brackish pools and in deep shade in mangrove swamps, both amongst roots and in more open water, and in footprints on mud flats. Three species of Anopheles also occur in mangroves. Anopheles amictus hilli breeds most abundantly in brackish water, and occurs in north Australia from Western Australia to the Queensland/New South Wales border. A. annulipes occasionally breeds in brackish water and occurs throughout Australia. A. farautz is found in the north of Australia from Western Australia to the east coast of Queensland. It sometimes breeds in brackish water. Perry (1946) reports it breeding in extensive brackish water pools in the Solomon Islands. These were high in organic debris and subject to tidal fluctuations. Similar habitats in Australian mangroves are likely to be suitable. Biting midges belong to the family Ceratopogonidae and are erroneously called sandflies. More than twenty species are associated with mangroves; many are still undescribed. For most, knowledge is restricted to the adult form. This information is largely based on emergence trapping which may or may not indicate larval habitat. A considerable amount of work has been carried out on the biology of coastal biting midges, with a view to developing control techniques. Debenham (1979, and pers. comm.) has reviewed the biology of coastal species of biting midges. Of the twenty-odd species, seven are common in mangroves. Culzcozdes henry occurs in coastal southern Queensland, coastal and subcoastal New South Wales; it appears to prefer muddy sand as a larval habitat, possibly with tree cover. The pupae float on the water surface with their long axes parallel to the surface, and are unable to submerge. The major source of blood is unknown, but people are attacked. Reye (1972a) ranks it as the seventh most important pest midge of coastal Queensland, while noting that it reaches pest proportions only in restricted localities. C. hzstr¢o occurs in Australia from coastal northeastern Australia south to the Sydney region. Immature stages have been collected from a Juncus pool at Careel Bay (New South Wales). At Townsville, C. historto occurs on mangrove flats well inside the mouth of the Ross River. Populations are maximal during the summer and negligible in winter (Kay and Fanning, 1974). It has been only recorded feeding on birds. C. magneszanus occurs on the coast and offshore islands of the Northern Territory and Queensland. It has been collected biting humans (Lee and Reye, 1955), but it is probably an avian feeder. Activity is nocturnal; peak activity in the Townsville area occurs between 2100 and 9300 hr (Lee and Reye, 1961). C. marmoratus occurs in Australia from north Queensland to southern New South Wales. Larvae occur in low-lying estuarine zones, often in association with Salicornia, Sporobolus wirginicus or Sueda maritima. The eggs cannot withstand desiccation and breeding occurs after spring tides which flood the salt marsh. This species is known to occur outside its breeding area. It has been reported to bite humans and can be a pest in coastal regions. Lee et al. (1963) suggest it is an opportunistic feeder, with wallabies its primary native host. Biting activity is primarily Proc. LINN. Soc. N.S.W., 106 (1), (1981) 1982 P. A. HUTCHINGS AND H. F. RECHER 93 crepuscular. C. molestus occurs throughout coastal eastern Australia. Reye (1972a) has characterized the larval habitat as clean sand in the open or among trees disturbed by slight to moderate wave or current action. A wide range of salinities is tolerated. The larvae occur with the top 7.5 cm of sand, and there is evidence that the pupae will drift in on a rising tide, be stranded and emerge there. This species has invaded the sandy banks of canal (housing) estates in southern Queensland (Reye, 1971). C. molestus has been recorded attacking people, but except in canal estates, it is not considered a problem (Kettle et al., 1975). C. subtmmaculatus occurs throughout coastal eastern Australia. Immature stages are found on sandy estuarine foreshores, and in the Salzcornza zone. They have also been found in the vicinity of crab holes (often associated with Avwcennza), and Reye (1969a,b) suggests that the presence of soldier crabs Mictyrzs platycheles is essential for the breeding of this species. Kettle (1977) has recorded the larvae feeding on polychaete worms and desmids. Mass emergences appear to be correlated with neap tides. The adults are opportunistic feeders. Activity is largely crepuscular but can be diurnal if humidity is high and there is little wind. The last common species is C. ornatus which occurs in the Torres Strait islands and coastal northern Australia. It does not occur south of Tin Can Bay in Queensland but this species has been often confused taxonomically giving it an erroneous distribution. The larval habitat is the mean neap tide zone of estuarine areas, mud substrate, completely sheltered from wave action, and with a dense tree cover, often Rhizophora stylosa. Reye (1972a) has suggested that there is an obligatory association with Aegiceras corniculatum. Adults are abundant in mangrove swamps, and Reye (1972a) regards this species as the most important Queensland pest species. It also feeds on flying foxes and birds. Spiders The spider fauna of Australian mangroves is undescribed but from casual observations appears to be rich in species and individuals. Hutchings and Recher (1974) recorded 18 species in temperate mangroves at Careel Bay, New South Wales. At Towra Point, Botany Bay (New South Wales), 35 species were found (Australian Littoral Society, 1977). Forty two species were collected from Trinity Bay, Cairns (Hegerl and Davie, 1977), and 56 species in Corio Bay, near Rockhampton (Ellway, 1974). Far fewer species (6) have been recorded from South Australia (Butler et al., 1977). These few observations do not allow for many generalizations. The diversity of spiders associated with mangroves increases rapidly from temperate to tropical zones, but all the spiders recorded from mangroves also occur in other terrestrial habitats. McCormick (1978) suggests that spiders in mangroves may be highly seasonal in occurrence and in areas subjected to regular flooding the fauna may be severely restricted. From the information available, it appears that the spider fauna of mangroves may be recruited from adjacent habitats. Of the spiders found in mangroves the orb web building spiders are the most evident. This group includes both the slant orb web weavers of the genus Tetragnatha (the large jawed spiders) which elsewhere are associated with stream or lake-side habitats and the vertical orb web weavers of the genus Eriophora. Both are adapted for catching flying prey. Foliage dwelling spiders seem to be rather uncommon. Other groups of spiders occur on the ground or on the lowest vegetation. Most evident in these situations are the wolf spiders (Geolycosa sp,) and allied hunting spiders of the family Pisauridae (Dolomedes sp.). Presumably these are opportunistic hunters, foraging on the exposed mud at low tide. Dolomedes, however, is often Proc. Linn. Soc. N.S.W., 106 (1), (1981) 1982 94 FAUNA OF AUSTRALIAN MANGROVES associated with aquatic habitats and has been observed to move underwater to avoid predators or take prey (Gray, pers. comm.). Scorpions and mites have been reported from Hinchinbrook Island (Bunt, pers. comm.). Aquatic Invertebrates This review is restricted to macro-invertebrates. Although microfauna such as nematodes and protozoa is abundant in mangroves, its presence has been poorly documented in Australia. Only a single study has been carried out on the nematode fauna. Dacraemes and Coomans (1978) collected 25 species of nematodes from a single sample in the mangroves at Lizard Island, Great Barrier Reef, suggesting that a rich fauna is awaiting study. Overseas, more attention has been directed to the microfauna (see Newell, 1974, for review). The aquatic invertebrates of mangroves are primarily marine in origin with some fresh water animals occurring at the freshwater, marine interface in the upper reaches of the estuaries. In addition there is a group of marine pulmonate molluscs which is probably related to terrestrial and freshwater pulmonate gastropods. However they have had a long history of independent evolution (Zilch, 1959). Within the intertidal zone, the mangrove fauna is dominated by polychaetes, crustaceans and molluscs (Appendices 1,2,3). In addition to the intertidal fauna, tide pools and channels with permanent water have a characteristic estuarine subtidal fauna. Again this is dominated by polychaetes, crustaceans and molluscs, but echinoderms, ascidians, coelenterates and sponges may be present (Saenger et al., 1980) . Sponges and ascidians are also associated with sea grasses. This subtidal fauna is arguably part of the mangrove system, but for the purposes of this review, we consider only the intertidal fauna. According to where they live, there are five distinct groupings of marine invertebrates within the intertidal of mangrove forests: encrusting epifauna, mangrove epifauna, substrate epifauna, substrate infauna and wood-boring infauna. The composition of the fauna within each of these categories is determined by the physical environment and there are distinct patterns of zonation (Hutchings and Recher, 1974). Zones are determined by sediment structure, tidal and salinity regimes and period of inundation. The distribution of mobile animals will also vary according to the time of day and night. Thus the intertidal fauna changes through time and is patterned on both a horizontal and vertical scale in space. Little of this complexity has been quantified and the summary which follows must be accepted as a simplification of an exceedingly complex situation. Studies which provide information on relative abundances of marine invertebrates within mangrove habitats are limited to Butler et al. (1974, 1975) (South Australia), Hutchings and Recher (1974), Hutchings et al. (1977), Weate (1975), McCormick (1978) (New South Wales), Shine et al. (1973) (Queensland) and Hegerl et al. (1979) (Northern Territory) . Epifauna The epifauna consists of sessile and mobile animals which live on the surface of the forest floor, on mangrove trunks and on mangrove roots and pneumatophores. Encrusting fauna Throughout most of Australia, the encrusting fauna is dominated by the oyster Saccostrea commerczalts and barnacles. The dense mass of these animals provides a sheltered environment for a rich and mobile fauna of errant polychaetes, crustaceans and gastropods. Encrusting animals occur throughout the mangroves, but require inundation on each tide and are best developed at the edges of the mangrove forest Proc. Linn. Soc. N.S.W., 106 (1), (1981) 1982 P. A. HUTCHINGS AND H. F. RECHER 95 where the flow of water is high and prolonged inundation occurs. As in other intertidal environments (e.g. rocky foreshores), barnacles tend to dominate in the upper reaches and oysters or mussels at lower tide levels. In New South Wales, encrusting fauna has a similar pattern of zonation irrespective of locality (McCormick, 1978). The greatest concentration of animals occurs in the frontal areas. McCormick (1978) suggests that frequency of tidal inundation and shade are the most critical factors affecting the abundance of this fauna. In areas of low light intensity, epiphytic algae (Bostrychia sp., Caloglossa sp., Catenella sp.) grow over the pneumatophores of Avicennia and around the base of tree trunks. These algae provide a microhabitat for small gastropods and amphipods. The algae, together with oysters may in some places retain enough moisture for spionid and nereid polychaete worms to occur. Broadly similar patterns and communities of encrusting organisms have been reported from South Australia, Victoria and Queensland. The barnacle Elmznzus modestus and the serpulid Galeolaria caespitosa occur on the pneumatophores and lower trunks of Avicennia marina in the upper part of Spencer and St Vincent Gulf and at Ceduna in South Australia (Womersley and Edmonds, 1958). In Victoria, the fauna on Avcennza is characterized by the snail Bembiccum melanostomum, the mussel Mytzlus edulis planulatus, and the barnacle Chamaesipho columna (Smith et al., 1975). The Sydney rock oyster Saccostrea commercialis and the barnacle Balanus amphitrite dominate the encrusting epifauna on Avicennza in New South Wales (Dakin et al., 1952; Hutchings and Recher, 1974; Hutchings et al., 1977). The snail Lzttortna scabra and the isopod Ligza australienszs are also typical of this community. Weate (1975) working on the lower Myall River on the central coast of New South Wales (32°S) recorded ten species of molluscs and four crabs living on mangroves. Some, such as the gastropods Salenator solida and Meloszdula zonata, grazed both on the trunks of Avicennza and the surface of the mud substrate. Other molluscs, including Patellocda mimula, Xenostrobus securis, Tatea ruftlabris and Lasaea australis occurred only on the encrusting oyster Saccostrea commerczalis. On the central coast of New South Wales, Hutchings and Recher (1974) recorded 26 species of molluscs and crabs living on or in the mangroves at Careel Bay (Appendices 2 and 3). The mangrove epifauna at Brooklyn, also on the Hawkesbury River estuary but upstream from Careel Bay, has a less diverse fauna, although the total number of individuals appears similar (Hutchings et al., 1977; pers. obs.). Species such as Balanus amphitrite, Paragrapsus laevis, Austrocochlea constricta and Melostdula zonata are absent at Brooklyn, but common at Careel Bay. The difference is probably the result of lower salinities and possibly of higher silt loads at Brooklyn. One of the most conspicuous animals in the mangrove epifauna is the gastropod Littorina scabra. L. scabra occurs throughout the mangrove forest and can be extremely abundant. In Queensland and northern Australia, L. scabra occurs with a second species of Littorzna, n. sp., (Ponder, pers. comm.). Other molluscs on mangrove trunks include Cerzthedea obtusa and Nerita lineata. Oncis sp., a pulmonate slug, is also present. Several studies in Queensland mangroves have not recorded the precise habitat where species were collected (e.g. Hegerl and Davie, 1977; Shine et al., 1973). However, the mangrove epifauna seems to be more diverse with decreasing latitude. Saenger et al. (1979) studied subtidal fouling organisms in the Calliope River, Queensland. Several of the species which colonized their settlement plates also occur on pneumatophores and tree trunks in the frontal zone of mangroves. These include such species as the barnacle Balanus amphitrite, the serpulid polychaete Fecopomatus Proc. Linn. Soc. N.S.W., 106 (1), (1981) 1982 96 FAUNA OF AUSTRALIAN MANGROVES uschakow and the oyster Saccostrea commercizalis. Yhey found that Balanus, Ficopomatus and the bryozoan Electra sp. characterize a pioneer community which is replaced by a community dominated by Saccostrea, Xenostrobus, Balanus and Ficopomatus. A mosaic of pioneer and climax phases is the common condition found on naturally occurring substrates such as mangrove pneumatophores and tree trunks in the study area. Two species of oysters (Saccostrea commerczalis and S. echinata) have been recorded from mangroves adjacent to the coral reef at Low Isles. The gastropod Morula was a common associate (Stephenson et al., 1958). A recent survey of the fauna of mangroves in the Kakadu National Park, Northern Territory, by Hegerl et al. (1979) does not indicate specific habitats for the molluscs collected, although the encrusting oysters Saccostrea echinata and S. commerczalis were collected. Barnacles were common on the trunks of Sonneratia, Camptostemon and Rhizophora, but were not identified. Substrate Epifauna The animals which live on the surface of the forest floor are mainly molluscs. Crabs are also common, but the majority live in burrows, coming out only to forage and are discussed under infauna. The molluscs are dominated by gastropods which are zoned in relation to tidal inundation. McCormick (1978) found that the frontal zone of the mangroves is the most diverse, although the diversity varied at the six sites he studied along the New South Wales coast. He found no trend towards increasing diversity from southern to northern New South Wales. However the sites varied greatly in their salinity regimes and mangrove structure which may explain these findings. Except for McCormick’s (1978) study seasonal fluctuations in substrate epifauna have been almost completely neglected. He found maximum densities during the winter and spring. Five families of molluscs dominate the epifauna, the Neritidae, Littorinidae, Potamididae, Cerithiidae and Ellobiidae (MacNae, 1968). Scattered information exists on the preferred habitats within the mangroves for some species. The two species of Terebralia, T. palustris and T. sulcata, prefer muddy substrates whereas species of Cassidula occur among decaying vegetation. The pulmonate slugs Onchzdium sp. are cryptic, burying themselves in mud or hiding under debris. Stephenson et al. (1958) found Onchidium, Quoyia and Bembicium concentrated among shingle under Bruguzera bushes at Low Isles. Many gastropods occur throughout the mangroves, on the surface of mud, on pneumatophores and in association with logs. The number of molluscs in the surface epifauna increases from the temperate zone to the tropics. Although almost certainly underestimates, 16 species have been recorded in South Australia, 24 in Victoria, 33 in New South Wales and 57 in northern Queensland (Appendix 3). There are distinct temperate and tropical elements to this fauna. Conuber melanostoma and C. sordida (Naticidae) are restricted to temperate mangroves and are absent from central Queensland and further north. In contrast Cassidula augulifera (Ellobiidae) is restricted to tropical mangroves although C. augulzfera does extend as far south as southern Queensland. In the family Neritidae, 6 species are restricted to tropical or sub-tropical mangroves and only N. atramentosa occurs in temperate mangroves and southern Queensland. As more information on the molluscs becomes available the distribution patterns will become clearer and will probably continue to substantiate the pattern of greater diversity of molluscs in tropical mangroves than in temperate systems (Fig. 3). MacNae (1968) hypothesized that the molluscs occurring in temperate mangroves such as Westernport Bay and the gulfs of South Australia are typical Proc. Linn. Soc. N.S.W., 106 (1), (1981) 1982 P. A. HUTCHINGS AND H. F. RECHER 97 My 146° 140. 12° ~~ we 6° «—_(54, 75) 18° pe £26__ Tropic of Capricorn 30° \] (22, ° 6) { Molluscs, Crustaceans Fg. 3. Several groups of marine invertebrates appear to be more abundant and richer in species in the tropics than in the temperate zones. Data on molluscs and crustaceans are most complete and are used here to illustrate this point for Australian mangrove forests. estuarine species and that specialized mangrove species are restricted to tropical areas. MacNae (1966) also suggested that the fauna associated with Sydney Harbour mangroves is a depauperate tropical one with all the species derived from Queensland. However he based this statement on incomplete species lists provided by Dakin e¢ al. (1952). The mangrove mollusc fauna near Sydney is in fact quite rich and is a mixture of tropical and temperate forms. There is relatively little information on the abundance of any of these animals. Hutchings et al. (1977) recorded densities of 0.4 individuals per m? for Littorzna scabra and 55.5 individuals per m? for a species of Tatea in the mangroves at Brooklyn on the Hawkesbury River. Numbers of Tatea can exceed 10,000 individuals per m? in salt marsh at the edge of mangroves (Ponder, pers. comm.). Butler et al. (1975) Proc. Linn. Soc. N.S.W., 106 (1), (1981) 1982 98 FAUNA OF AUSTRALIAN MANGROVES provide a list of macrobenthic animals found in mangrove areas at different locations in South Australia and rank these by relative abundance. Except for McCormick’s (1978) work in New South Wales, there are no measures of the seasonal change of the substrate epifauna. McCormick (1978) found that maximum densities occur in winter and spring at Towra Point on Botany Bay and Patonga Creek on Broken Bay, Hawkesbury River (Fig. 4). Substrate Infauna This fauna includes animals that live in the sediment for all or part of their lives. It is restricted to the upper layers of mud and consists of sedentary and mobile animals. The mobile species such as crabs feed on the surface of the mud. The majority of the fauna live in semi-permanent or permanent burrows. Burrows penetrate through the oxygenated layers of mud into the anaerobic layers, where they are restricted in depth by the fibrous matted root systems of the mangroves. Some of the terrestrial fauna such as ants may trap air in their burrows during high tide and marine animals such as crabs trap water during low tide. The infauna has been neglected in mangrove faunal studies and well-planned, long term collecting throughout Australian mangroves is needed. The sedentary animals are represented by sipunculans, echiuroids, polychaetes and nemerteans. There is a single species of sipunculan, Phascolosoma arcuatum. It occurs in tropical £12 — < = 10 (2) =) = 8 iva 3 a 6 €q = 4 o Pd MHHW i MSL = MLLW ! y 1 A : 100 1 200 1 AREA 9! AREA 1 ! AREA 2 1 AREA 3 ' SALT MARSH DISTANCE FROM MUD FLAT (m) PATONGA CREEK; 33° 32'S, 151° 16 'E, (JANUARY) AREA O AREA 1 EPIFAUNA INFAUNA TREE FAUNA Fig. 4. McCormick (1978) studied the horizontal distribution of animals in several areas of mangroves in New South Wales. These data on species number for the mangrove stands at Patonga on the Hawkesbury River are presented in this figure. Both epifauna and infauna are richest in species at the lower tide levels where there is regular inundation by the tides. The terrestrial invertebrate fauna is most abundant in the middle zones where refuges during high tide occur and absent from frontal areas (area 3) which are covered by high tides. Proc. LINN. Soc. N.S.W., 106 (1), (1981) 1982 P. A. HUTCHINGS AND H. F. RECHER 99 mangroves in Australia and extends into South East Asia (Edmonds, 1980). Phascolosoma occurs throughout the mangroves, constructs extensive galleries in the substrate, and may occur in large numbers. The echiuroid Ochetostoma australiense occurs on the mud flats in front of the mangroves from northern New South Wales to southern Queensland. This species has a commensal crab, bivalve and gastropod associated with it in Queensland (Ponder, pers. comm.). Polychaetes probably occur in all mangrove areas in Australia although they have rarely been collected systematically. The species which have so far been recorded are listed in Appendix I but many more species probably occur. Polychaetes are restricted to the wetter seaward margins of the mangroves or near regions of permanent water (Hutchings, pers. obs.). They occur in the less consolidated sediments where burrowing is easy. Burrows do not extend into the dense root systems of the mangroves. Several families are represented but nereids, spionids and capitellids are the most abundant groups. From the data available it appears that the polychaete fauna is more diverse in tropical regions. Most of the species occurring in temperate mangrove systems also occur in other estuarine habitats. Ai) exception is the terebellid Hadrachaeta aspeta which occurs only in mangroves or on mud flats immediately adjacent to the mangroves (Hutchings, pers. obs.). Whether more mangrove specialists occur in the tropics remains to be investigated. Some of the species such as the nereids and spionids construct semi-permanent mucous-lined tubes. Many of the species occurring in the substrate also occur in rotten logs or in pockets of water trapped among the encrusting epifauna on pneumatophores, prop roots and tree trunks. There are limited data on the abundance of polychaetes in mangroves. Hutchings et al. (1977) and McCormick (1978) have some data for New South Wales. In general densities of polychaetes appear low with less than one individual per m7’. However, McCormick (1978) found that numbers changed seasonally and more extensive sampling could reveal higher population sizes. Numbers of polychaetes in Queensland mangroves appear to be higher than in New South Wales. Nemerteans occur in frequently-inundated areas of mangroves where sediments are relatively unconsolidated. A tropical species, Pantenonemertes winsort, occurs beneath bark and in cavities in rotten wood of Avicennia marina and Ceriops tagal (Moore and Gibson, 1981). At Careel Bay, Tubulanus polymorphus occurs in sea grass beds (Zostera sp.) adjacent to the mangroves and probably extends into the frontal zone of the mangrove forest (Hutchings and Recher, 1974). Turbellarians also occur in mangroves throughout New South Wales (Hutchings and Recher, pers. obs.), but have not been identified. Molluscs are represented in the substrate of mangrove habitats by large numbers of bivalves. McCormick (1978) reported up to four species from his sites in New South Wales. Glauconome plankta, Arthritica helmsi sp., Laternula sp., Tellina deltoidalis and Venerupis crenata were the most abundant species. Species diversity is higher in tropical mangroves and a number of species may be restricted to mangrove habitats (Fig. 3). MacNae (1968) suggested that some mangrove bivalves literally form a ‘cocoon’ among the matted roots of the mangroves. Bivalves are most abundant along the seaward fringe of the mangrove forest and decrease in abundance and diversity with decreasing tidal inundation (McCormick, 1978). Crustaceans are a conspicuous component of the substrate fauna (Appendix 2). Crabs are the most obvious. The snapping shrimp Alpheus and the burrowing prawn or nipper Callianassa as well as isopods and amphipods are less visible, but are probably equally abundant. Certainly one of the most characteristic sounds of Proc. Linn. Soc. N.S.W., 106 (1), (1981) 1982 100 FAUNA OF AUSTRALIAN MANGROVES mangrove environments at low tide is the ‘snapping’ of alpheids in their burrows. Crabs are represented by large numbers of species in several families (Appendix 2) and are among the best known of the mangrove fauna (Fig. 3). In tropical mangroves the burrows of the mud lobster Thalasstna anomola and the mud crab Scylla serrata are conspicuous. Juvenile Scylla are particularly common in mangroves although they also occur among sea grasses and the retention of extensive mangrove forests may be critical for the management of this commercially- important species. Once the juveniles reach a carapace width of 60-80 mm they tend to move into subtidal areas, but continue to forage among the mangroves on the rising tide (Hill, 1979, 1980, and pers. comm.). The use of mangroves is accentuated in places where the density of sub-tidal benthic animals is low, perhaps as a result of freshwater conditions during the wet season, such as in far north Queensland. There is a distinct zonation of crab species within the mangrove forest. In Western Australia, George and Jones (in press) correlated the distribution of fiddler crabs Uca with sediment grain size. Elsewhere zonation appears to be determined largely by period of tidal inundation. Yates (1978) studied the ecology of crabs in mangroves at Patonga Creek on Broken Bay, New South Wales (33°S). Two ocypodids Heloeczus cordiformis and Australoplax tridentata, are most abundant on the seaward fringe of the mangroves with numbers decreasing towards land. Heteropanope serratifrons and Ilyograpsus paludicola occur only on the seaward margins. By contrast Sesarma erythrodactyla is most abundant in the highest zones with numbers decreasing towards the seaward edge. Similarly, Helograpsus haswellianus occurs only on the landward margins. Paragrapsus laevis is restricted to the middle regions of the mangroves. Helice leachi, which occurs only in small numbers, is restricted to the higher levels of the shore. At Patonga there are only two species of mangrove, Avicenna marina and Aegiceras corniculatum. The distribution and abundance of crab species cannot be correlated with either species (Yates, 1978). In the tropics some species of crab are restricted to particular zones and may be dependent on the species of mangrove. According to MacNae (1968), Uca lactea is associated with Avicennia, especially when it is growing in sand. If the sand is very fine U. bellator occurs and is also common in areas shaded by outermost bushes of Cerzops. However, George and Jones (in press) disagree with MacNae’s identifications and the precise nature of the distribution of his Uca species in Australian mangroves must now be viewed with caution. The larger sesarmids, being mainly herbivores, occur in areas of seedlings of Avicenna, Bruguzera and Ceriops. Whereas Macrophthalmus depressus, an omnivore, is restricted to sandy substrates in drainage channels (MacNae, 1968). Yates (1978) recorded seasonal changes in the abundance of crabs. Maximum densities of most species occur during summer. Paragrapsus laevts is most abundant in winter. Some species, such as Sesarma erythrodactyla and Paragrapsus laevis, move to lower zones in the mangroves to spawn. Crabs also show daily activity patterns. During the day, at low tide, crabs tend to remain in their burrows. Hutchings and Recher (1974) recorded maximum concentrations of crabs at night on a rising tide. Many of the crabs were climbing pneumatophores and trees to forage. Wood Infauna The fauna associated with dead and living wood is not restricted to mangroves, but because logs tend to be trapped and accumulate in mangrove habitats, mangroves have a rich wood-boring fauna. In Australia, the wood-boring fauna is dominated by the teredinid molluscs of which 31 species in eight genera have been described (Turner et al., 1972; Turner and McKay, 1979). Holdich and Harrison (1980) have Proc. Linn. Soc. N.S.W.,106 (1), (1981)1982 P. A. HUTCHINGS AND H. F. RECHER 101 recently described several species of isopods Gnathia from dead wood found in Queensland mangroves. Animals such as the ship worms (Teredinidae) and gnathid isopods are important in the breakdown of wood in marine habitats. However, there is some evidence that the wood must first be invaded by fungi before it can be colonized by wood boring animals (Cragg and Swift, 1980). Leightley (1980) has found marine fungi in the wood of the five species of Australian mangrove so far examined. The succession of marine fungi in wood has been extensively studied in Florida by Newell (1974) and similar patterns can be expected in Australia. In Papua New Guinea, Cragg and Swift (1980) have drawn analogies between terrestrial decomposers such as termites and beetles and the marine timber borers. The wood-boring fauna is not restricted to dead wood. In Florida, an isopod Sphaeroma terebrans attacks the live prop roots of Rhizophora mangle. Simberloff et al. (1978) suggest that the damage by isopods coupled with insect attack on aerial roots may stimulate root branching. For every root produced by the tree, 1.4 roots reach the substrate thereby providing greater support for the plant. There is no information for Australia on the wood-boring fauna of live mangroves. Associated with these wood-boring organisms is a range of animals seeking protection from predation, desiccation, etc., such as barnacles, limpets, crabs, amphipods, nemerteans and polychaetes. This fauna is most diverse in logs trapped in the frontal margins of the mangroves which are frequently inundated. Under logs, small pockets of water are often trapped which provide suitable microhabitats for many of these animals. In temperate mangroves the crab Sesarma erythrodactyla commonly occurs under logs (Yates, 1978). Graham et al. (1975) found 34 species of molluscs associated with logs in Trinity Bay, Cairns. These molluscs often use the tunnels and cavities created by the wood-boring teredinids. Also, some species such as Ellobtum aurtsjudea seek refuge in the log during the day, and feed on the muddy substrates in the Rhzzophora and Bruguzera zones at night or on cloudy, rainy days. One species of nemertean Pantznonemertes winsori has been recorded from tropical mangroves (Moore and Gibson, 1981). It occurs beneath bark or in cavities in rotten fallen timber of Awicennia marina and Ceriops tagal in the upper tide levels. At Gladstone, central Queensland (23°S), many species of molluscs including Melampus castaneus, M. striatus, Ellobtum aurisjudea, Onchidium sp., Bactranophorus sp., Nerita lineata, Isognomon ephippium have been collected from logs (Saenger, pers. comm.). Microgastropods (e.g. Assemznae spp. and Iravadza sp.) are also found on logs in tropical mangroves (Ponder, pers. comm.). Ponder has evidence that differences in the mollusc fauna can be attributed to differences in the moisture content of the log and the frequency of inundation. In the Northern Territory several species of crab utilize rotting logs in the mangroves of Kakadu National Park (Hegerl et al., 1979). Hegerl et al. (1979) found that Metopograpsus quadridentatus, Metopograpsus sp., and Clzstocoeloma merquzensis were most abundant in logs in the upper part of the tidal range. Nannosesarma batavicum and Sesarma sp. occur in logs in the Rhizophora zone. S. krausst borneensis is restricted to sites which are frequently inundated and Epzxanthus dentatus occurs in Rhizophora, Sonneratia/Camptostemon zones. S. darwinensis commonly occurs under decaying logs. Saenger (pers. comm.) collected the crabs Clistocoeloma merguiense, Sesarma molluccensis and S. elongata from logs among mangroves at Gladstone. PATTERNS It helps to understand the ecology of mangrove ecosystems if it is realized that mangroves are essentially forests with a muddy intertidal substrate. The abundance of Proc. Linn. Soc. N.S.W., 106 (1), (1981) 1982 102 FAUNA OF AUSTRALIAN MANGROVES animals, their distribution within the mangrove forest, and their adaptations to the physical and biological environment can then be explained in the same ways that one would discuss a terrestrial forest or estuarine mudflat. For much of the fauna, mangroves are no more than an extension of their ‘normal’ habitat. Relatively few ” animals are restricted to mangroves or show specific adaptations to the mangrove environment. Mangrove endemics have probably evolved in response to a limited array of unique features. Mangroves are flooded regularly by the tide, for varying periods depending on the tidal cycle, and at varying times during the day. For terrestrial organisms, this restricts movement and limits the amount of time available for foraging on the forest floor. McCormick (1978) has suggested that the canopy fauna in mangroves which is subject to partial or complete flooding of the canopy is depauperate for this reason. High tannin content and concentrations of salt on or within the leaves of mangroves cause special problems for herbivores. The resistance of mangrove leaves to breakdown may also limit the ability of marine detritivores to use them. The mangrove community reflects the merging of the marine and terrestrial systems. In temperate areas, at their most complex, mangroves form fairly open forests with a simple canopy; in the tropics, mangroves may be open or closed and are often layered with distinct zonation. The terrestrial fauna is largely restricted to the forest canopy, and the marine fauna to the forest floor and lower levels of trees. Thus there is a horizontal partitioning of these two parts of the mangrove community. Within the two faunal groups, subdivisions occur. For terrestrial animals, these divisions may be related to the horizontal zonation of the mangroves. For example, stands of Rhizophora may provide vertical complexity, nesting sites or shelter for a particular species of bird, whereas stands of Avicennia may not offer these advantages. As mangroves are often a continuation of nearby forest, the partitioning within these forests should be compared with that occurring in mangroves. Such comparisons may explain, for example, why some birds which occur in adjacent habitats are absent from mangroves. For the marine fauna, mangroves provide additional habitat to that found on muddy shores. The dense canopy of the forest provides protection against desiccation and may offer cover against predators. Mangrove conditions are less rigorous in terms of salinity and temperature change than those on an exposed mudflat. The less rigorous environment and the firm substrates provided by roots, stems and logs enable some animals to survive which might otherwise be absent from the intertidal environment of estuaries. As a generalization, most of the marine fauna occurring in temperate mangroves also occurs on adjacent muddy or rocky shores. Mangrove specialists or endemics are by and large restricted to the tropics (e.g. Figs 2 and 3). Either the tropical mangrove forest offers habitats not available in tropical estuaries (e.g. solid substrates) or the cooler conditions, high humidity and shelter from the sun within’ the mangrove forest are critical for survival. For the marine fauna, zonation horizontally across the substrate from low to high water mark and vertically up roots and trunks and within the substrate, occurs in response to the periodicity of tidal inundation (Fig. 4). This may be modified by drainage patterns and tidal creeks within the mangroves. Behavioural, physiological and morphological adaptations to water loss will largely determine the tide levels at which particular species can live. The distribution of animals may vary during a tidal cycle or during their life cycle; some species for example migrate seawards to spawn. Like the terrestrial fauna, the distribution of the marine fauna may be determined by Proc. Linn. Soc. N.S.W., 106 (1), (1981) 1982 P. A. HUTCHINGS AND H. F. RECHER 103 the zonation of the mangroves themselves. For example, the prop roots of Rhzzophora and the buttress roots of Bruguzera may provide different microhabitats for marine animals in comparison to the pneumatophores of Avicennia. The density of pneumatophores, the amount of debris on the floor, and fallen logs are important for cryptic species and wood fauna, in providing refuges and protection from desiccation. Zonation of sediment within the mangroves is important to infaunal species. Also the density of seedlings and the canopy cover which provide shade, affect animals living on the surface of the mud. Salinity gradients along a river will determine how far each species penetrates upstream. Superimposed upon all these factors are latitudinal gradients along the coast. The number of mangrove species increases from south to north with the greatest diversity occurring in northeastern Queensland and northwestern Western Australia (Lear and Turner, 1977) (Fig. 1). Coincident with the increasing number of species, the extent of mangrove forests and their structural complexity also increase. Although data are lacking for most tropical mangroves and for Western Australia in particular, there appears to be a precipitous increase in the diversity of the mangrove fauna which parallels that of the forest (Figs 2 and 3). As in the case of birds (Schodde et al., 1982), the increased diversity of species is accompanied by an increased number of mangrove specialists and endemics (Fig. 3). The three major groups of marine animals, polychaetes, crustaceans and molluscs (Fig. 3), all increase in species number from south to north (Appendices 1, 2, 3) and similar patterns can be expected for other groups. Pianka (1978) and others have reviewed the many theories advanced to explain the increased diversity of tropical animal communities relative to those in similar habitats within the temperate zones. It is not possible to separate completely the different factors which may have led to the greater diversity of tropical systems. In the case of mangroves, the greater structural complexity of the tropical forests, the increased number of plant species and larger areas are all related to increased faunal diversity. The area of mangroves and the distance between stands, in particular, may explain the much greater numbers of birds which use mangroves as a primary habitat in the tropics when compared to temperate mangrove forests. This is simply a logical extension of the theories of island biogeography (see MacArthur and Wilson, 1967). The increased structural complexity and greater number of plant species would also permit a greater number of bird species to co-exist (Recher, 1971). Schodde et al. (1982) provide examples of resource apportionment among closely related birds co- existing in northern mangrove forests. The white-breasted whistler Pachycephala laniotdes apparently forages extensively on the muddy forest floor where it takes various marine invertebrates while the co-occurring mangrove golden whistler P. melanura forages for insects in the canopy. Two co-habiting kingfishers Alcedo azurea and A. pusilla differ in size (28 to 35 grams versus 10 to 13 grams) and would be expected to take different sized prey. Other examples given by Schodde e¢ al. (1982) include flycatchers Myagra, fantails Rhzpidura, and warblers Gerygone. Similar patterns are evident among marine invertebrates with various crabs separated within the mangroves by size of sediment particles (Uca spp.), type of food or capacity to withstand desiccation. As data become available, it is likely that significant regional differences in community composition and diversity will be evident between tropical mangrove forests in northern Australia. Stands of mangroves in northwestern Australia are apparently much more uniform in species composition than those in northeastern Australia. Historical events related to recent glacial periods and the separation of northeastern and northwestern marine environments by the Torres Strait land bridge Proc. Linn. Soc. N.S.W., 106 (1), (1981) 1982 104 FAUNA OF AUSTRALIAN MANGROVES have probably had a significant effect on the evolution of distinctive mangrove communities in these regions. During the time the land bridge existed there were enhanced opportunities for plants and animals to move between mangrove forests in eastern Australia and eastern New Guinea. These were isolated from mangroves occurring along the western fringes of the continent. Schodde et al. (1982) discuss the origins of the mangrove avifauna in the context of the Torres Strait land bridge. They suggest that northwestern Australia was linked to New Guinea by a mangrove-rich land connection, Arafura Land. This, they argue, was the centre of evolution for mangrove bird specialists. In addition, the mangroves in this area are geologically older and have been stable over geological time. Those in northern Queensland have been affected by fluctuations in sea levels which periodically eliminated mangrove areas (Coventry et al., 1980). Cape York was connected to the Gulf of Papua which had relatively few mangrove forests and hence fewer mangrove birds. Whether similar patterns exist for other groups of mangrove animals is unknown. As is evident from the material presented in this review, information on mangrove fauna is sketchy. There is a particular need for quantitative studies of fauna and for work in tropical mangrove forests. Until such information is available it will not be possible to evaluate fairly the theories of Schodde and his colleagues or those of MacNae (1968) regarding the evolution and dispersion of mangrove animals in Australia. Nor, we submit, will it be possible to develop sound management and conservation strategies for what must be regarded as one of Australia’s richest ecosystems. ACKNOWLEDGEMENTS This review would have been impossible without the help of many people, and we should like to thank especially H. Cogger, I. Common, P. Davie, M. Debenham, R. George, M. Griffiths, D. Hoese, I. Loch, P. Marks, L. Miller, W. Ponder, D. Rentz, E. Reye, C. Smithers, J. Taylor, T. Taylor, T. Watson, F. Wells and E. Zimmerman, who provided unpublished information. We also thank J. Davie, M. Gill and T. Kingston for reviewing the manuscript and A. Murray for artwork and K. Handley for literature searching. However, we take full responsibility for any inaccuracies in reporting the distribution of the fauna. References AUSTRALIAN LITTORAL Society, 1977. — An investigation of management options for Towra Point, Botany Bay. Report prepared for the Australian National Parks and Wildlife Service. BARNES, R. S. K., 1967. — The Macrophthalminae of Australasia; with a review of the evolution and morphological diversity of the type genus Macrophthalmus (Crustacea: Brachyura). Trans. Zool. Soc. Lond.31: 195-262. ———, 1971. — Biological results of the Snellius Expedition XXIII. The genus Macrophthalmus (Crustacea: Brachyura). Zool. Verh., Leiden 115: 1-40. BEUMER, J. P., 1978. — Feeding ecology of fishes from a mangrove creek in north Queensland, Australia. /. Fish Biol. 12: 475-490. BLABER, S.J. M., 1980. — Fish of the Trinity Inlet system of north Queensland with notes on the ecology of fish faunas of tropical Indo-Pacific estuaries. Aust. J. mar. Freshwater Res. 31: 137-146. BuNT, J., Boro, K. G., and Boro, G., 1979. — A survey method for estimating potential levels of mangrove forest primary production. Mar. Biol. 52: 123-129. Proc. Linn. Soc. N.S.W., 106 (1), (1981) 1982 P. A. HUTCHINGS AND H. F. RECHER 105 ———, and WiLLiams, W. T., 1980. — Studies in the analysis of data from Australian tidal forests (Mangroves’). 1. Vegetational sequences and their graphic representation Aust. J. Ecol. 5: 383-390. ———, 1981. — Vegetational relationships in the mangroves of tropical Australia. Marine Ecology: Progress Series 4: 349-359. Burter, A. J., 1974. — An account of further ecological researches conducted by the third year class in Ecology and Marine Biology at Garden Island, Port Adelaide, March-April, 1974. Unpublished reports, Zoology Dept., University of Adelaide. ———, Depers, A. J., McKitiup, S. C., and THomas, D. P., 1975. — The conservation of mangrove swamps in South Australia. A Report to the Nature Conservation Society of S.A. (Zoology Dept., University of Adelaide). ———, 1977. — A survey of mangrove forests in South Australia. South Aust. Nat. 51: 34-49. CampBELL, B. M., 1967. — The Australian Sesarminae (Crustacea: Brachyura) : Five species of Sesarma (Chiromantes). Mem. Qd Mus. 15 (1): 1-19. ———, and GrirFIN, D. J. G., 1966. — The Australian Sesarminae (Crustacea: Brachyura): Genera Helice, Helograpsus nov., Cyclograpsus, and Paragrapsus. Mem. Qd Mus. 14 (5) : 127-174. CuapmaN, V.J., 1975. — Mangrove vegetation. Leutershausen, Germany: J. Cramer. CiarkE, L., and Hannon, N., 1970. — The mangrove swamp and salt marsh communities of the Sydney district. III. Plant growth in relation to salinity and waterlogging. J. Ecol. 58: 351-370. ———, 1971. — The mangrove swamp and salt marsh communities of the Sydney district. IV. The significance of species interaction. J. Ecol. 59: 535-554. CLoucu, B. F., and AnpreEws, T. J., 1982. — Physiological processes in mangroves. In: B. F. CLouGH, (ed.), Structure functzon and management of mangrove ecosystems in Australia. Canberra: ANU Press. ———, and ATTIWILL, P., 1982 a. — Primary processes in mangroves. In: B. F. CLoucu, (ed.), Structure function and management in mangrove ecosystems in Australia. Canberra: ANU Press. ———, 1982 b. — Primary production in mangroves. In: B. F. CLouGu, (ed.), Structure function and management of mangrove ecosystems in Australza. Canberra: ANU Press. CoccER, H. G., 1979. — Reptiles and amphibians of Australia. 2d edition. Sydney, Wellington, London: A.H.and A. W. Reed. Common, I. W. B., 1970. — Lepidoptera. In: I. M. Mackerras, (ed.), The Insects of Australia (pp. 765- 866). Melbourne: Melbourne Univ. Press. ———, and WATERHOUSE, D. F., 1972. — Butterflies of Australza. Sydney: Angus and Robertson. Coventry, R. J., Hoptey, D., CAMPBELL, J. B., Doucuas, I., Harvey, N., KERsHAw, A. P., OLIVER, J., Puipps, C. V. G., and Pyr, K., 1980. — The Quaternary of northeastern Australia. In: J. STEVEN, (ed.), Development of N.E. Australia (pp. 375-417). Brisbane: Geological Society of Australia, Qld Division. Cracc, S. M., and Swirr, M. J., 1980. — The contribution of fungi and marine borers to wood decay of some mangrove communities of Papua New Guinea. Second. Int. Sym. Biol. Management of Mangroves and Tropical Shallow Water Communites. (Abstract only) p. 22. CRANE, J., 1975. — The Fiddler Crabs of the World. Ocypodidae: Genus Uca. Princeton, New Jersey: Princeton Univ. Press. Dakin, W. J., BENNETT, I., and Pope, E., 1952. — Australian Seashores. Sydney: Angus and Robertson. Davie, P. J., — A preliminary checklist of Brachyura (Crustacea: Decapoda) associated with Australian mangrove forests (unpublished ms.) . DEBENHAM, M. L., 1979. — An annotated checklist and bibliography of Australasian region Ceratopogonidae, (Diptera, Nematocera). Entomology Monograph No. 1. Canberra: Aust. Government Publishing Service. DECRAEMER, W., and Coomans, A., 1978. — Scientific report on the Belgian Expedition to the Great Barrier Reef in 1967. Nematodes XII. Ecological notes on the nematode fauna in and around mangroves on Lizard Island. Aust. J]. mar. Freshwater Res. 29: 497-508. Duke, N. C., and Bunt, J. S., 1979. — The genus Rhzzophora (Rhizophoraceae) in north eastern Australia. Aust. J. Bot. 27; 657-678. , and WILLIAMs, W. T., (in press) — Mangrove litter fall in north eastern Australia I. Annual totals by component in selected species. Aust. J. Bot. 29(5). Epmonps, S. J., 1980. — A revision of the systematics of Australian Sipunculans (Sipuncula). Rec. S. Aust. Mus. 18(1) : 1-74. Exiway, C. P., 1974. — An ecological study of Corio Bay, Central Queensland. Report commissioned by the Capricorn Coast Protection Council. , and HEGERL, E. J., 1972. — Fishes of the Tweed River estuary. Operculum 2: 16-23. Fritu, H.J., 1973. — Wildlife Conservation. Sydney: Angus and Robertson. Gay, F. J., and Watson, J. A. L., (in press). — The genus Cryptotermes in Australia (Isoptera: Kalotermitidae). Aust. J. Zool. Proc. Linn. Soc. N.S.W., 106 (1), (1981) 1982 106 FAUNA OF AUSTRALIAN MANGROVES GeorGE, R. W., and Knorr, M. E., 1965. — The ocypode ghost crabs of Western Australia (Crustacea: Brachyura). J. Roy. Soc. West. Aust. 48(1) : 15-21. ———, and Jones, D. S., (in press). — West Australian and Northern Territory fiddler crabs (subfamily Ocypodinae). Rec. West. Aust. Mus. Goutter, P. F. E. and ALLAway, W. G., 1979. — Litter fall and decomposition in a mangrove stand Avicennia marina (Forsk.) Vierh., in Middle Harbour, Sydney. Aust. J. mar. Freshwater Res. 30: 541-546. GraHaM, M., GRrIMSHAW, J., HEGERL, E., McNALty, J., and Timmins, R., 1975. — Cairns wetlands. A preliminary report. Operculum 4 (3-4) : 117-148. GrirFiTHs, M. — Review of the Culicidae from the Australasian region (in ms.). Hate, H. M., 1927. — The Crustaceans of South Australia. Handbooks of the flora and fauna of South Australia. Adelaide: Govt. Printer. HaMiyn-Harris, R., 1933. — Some ecological factors involved in the dispersal of mosquitoes in Queensland. Bull. ent. Res. 24: 229-232. HEGERL, E. J., and Davig, J. D., 1977. — The mangrove forests of Cairns, Northern Australia. Mar. Res. Indonesza 18: 23-59. ———, Davir, P. J. F., Craripce, G. F., and Eruiotr, A. G., 1979. — The Kakadu National Park mangrove forests and tzdal marshes. Volume 1. A review of the literature and results of a freld reconnaissance. Report prepared for the Australian National Parks and Wildlife Service. Brisbane: Australian Littoral Society. ——, and Tarte, D. M., 1974. — A reconnaissance of the Capricorn Coast tidal wetlands. Operculum 4 (2) : 50-62. ———, and Timmins, R. D., 1973. — The Noosa River tidal swamps: a preliminary report on the flora and fauna. Operculum 3 (4) : 38-43. HEINSOHN, G. E., and Wake, J. A., 1976. — The importance of the Fraser Island region to dugongs. Operculum 5: 15-18. HIu, B. J., 1979. — Aspects of the feeding strategy of the predatory crab Scylla serrata. Mar. Bzol. 55: 209- 214. ———, 1980. — Effects of temperature on feeding activity in the crab Scylla serrata. Mar. Biol. 59: 189- 192. HILL, G. F., 1917. — Report on some Culicidae of the Northern Territory. Bull. Nth. Terr. Aust. 17: 1-8. Ho.picH, D. M., and Harrison, K., 1980. — The crustacean isopod genus Gnathza Leach from Queensland waters with descriptions of nine new species. Aust. J. mar. Freshwater Res. 31: 215-240. HuTcHincs, P. A., PICKARD, J., RECHER, H. F., and Weate, P. B., 1977. — A survey of mangroves at Brooklyn, Hawkesbury River, New South Wales. Operculum Jan. 1977: 105-112. ———, and Recuer, H. F., 1974. — The fauna of Careel Bay with comments on the ecology of mangroves and sea-grass communities. Aust. Zool. 18: 99-128. ———, 1977. — The management of mangroves in an urban situation. Mar. Res. Indonesza. 18: 1-13. IyENGAR, M. O. T., 1965. — Epidemiology of filariasis in the South Pacific. Techn. Pap. S. Pacif. Comm. 148: 1-183. Kay, B. H., and FANNING, I. D., 1974. — Brunswick Heads midge study. Rep. Qld. Inst. med. Res. 29: 11- 12. KETTLE, D.S., 1977. — Biology and bionomics of blood sucking ceratopogonids. A. Rev. Ent. 22: 33-51. ———, ReyE, E. J., and Epwarps, P. B., 1979. — Distribution of Culicozdes molestus (Skuse) (Diptera: Ceratopogonidae) in man-made canals in south-eastern Queensland. Aust. J. mar. Freshwater Res. 30: 653 — 660. KIkKAWA, J., and PEARSE, K., 1969. — Geographical distribution of land birds in Australia — a numerical analysis. Aust. J. Zool. 17: 821-40. Lear, R., and Turner, T., 1977. — Mangroves of Australia. Brisbane: Univ. of Queensland Press. Ler, D. J., Hicks, M. M., GrirritHs, M., RussELt, R. C., and Marks, E. N., 1980. — The Culiczdae of the Australasian region. Commonwealth Inst. of Health Monograph Series, Entomology Monograph No. 2. Vol. 1. Canberra: Aust. Government Publishing Service. ———, and ReyeE, E. J., 1955. — Australian Ceratopogonidae (Diptera, Nematocera) Part VII. Notes on the genera Alluaudomyia, Ceratopogon, Culicovdes and Lastohelea. Proc. Linn. Soc. N.S.W. 79: 233-246. ———, 1961. — An investigation of the possible role of biting midges (Diptera, Ceratopogonidae) , in the transmission of arthropod-borne virus diseases at Townsville. Proc. Linn. Soc. N.S.W. 86: 230-236. ———, Reve, E. J., and Dyce, A. L., 1963. — ‘Sandflies’ as possible vectors of disease in domesticated animals in Australia. Proc. Linn. Soc. N.S.W. 87: 364-376. LEIGHTLEY, L. E., 1980. — Biodeterioration of mangrove wood by marine fungi. Second Int. Symp. Bzol. Management of Mangroves and Tropical Shallow Water Communities. (Abstract only) p. 36. Lucas, J. S., 1980. — Spider crabs of the family Hymenosomatidae (Crustacea: Brachyura) with particular Proc. Linn. Soc. N.S.W., 106 (1), (1981) 1982 P. A. HUTCHINGS AND H. F. RECHER 107 reference to Australian species: Systematics and biology. Rec. Aust. Mus. 33 (4) : 148-247. Luco, A. E., and SNEDAKER, S. C., 1974. — The ecology of mangroves. Ann. Rev. Ecol. Syst. 5: 39-65. MacArruur, R. H., and Witson, E. O., 1967. — The Theory of Island Biogeography. Princeton: Princeton Univ. Press. MacNae, W., 1966. — Mangroves in eastern and southern Australia. Aust. J. Bot., 14(1) : 67-104. ———, 1967. — Zonation within mangroves associated with estuaries in North Queensland. In: G. H. CRONIN, (ed.), Estwarzes. Amer. Assoc. Adv. Sct. Publ. 83: 432-441. ——— 1968. — A general account of the fauna and flora of mangrove swamps and forests in the Indo-West Pacific region. Advances in Marine Biology, 6: 73-270, F. S. RussELt (ed.). London: Academic Press. Macnusson, W. E., Wess, G. J. W., and Tay or, J. A., 1976. — Two new locality records, a new habitat and a nest description for Xeromys myozdes Thomas (Rodentia: Muridae). Aust. Wild. Res. 3: 153-157. Matty, D. F., 1978. — Degradation of mangrove leaf litter by the tropical sesarmid crab Chzromanthes onychophorum. Mar. Biol. 49: 377-386. Marks, E. N., 1953. — Report of the National Mosquito Control Committee. Appendix A. Ann. Rep. Hith. med. Serv. Qd 1952-53: 104-108. ———, 1967. — An atlas of common Queensland mosquitoes; with a guide to common Queensland biting midges by E. J. Reye. Brisbane, Queensland. (Mimeo., 91 pp.). ———, 1971. — Key to mosquitoes: In: Freld manual for tzdal swamps. Subtropical eastern Australia. Brisbane: Queensland Littoral Society. (Mimeo, pp. 6.0-6.9). McCormick, W. A., 1978. — The ecology of benthic macrofauna in New South Wales mangrove swamps. Kensington: Univ. of New South Wales, M.Sc. thesis, unpubl. McNEILL, F. A., 1926. — Studies in Australian carcinology No. 2. (A revision of the family Mictyridae and the status and synonymy of the genus Megametope and its contained species.) Rec. Aust. Mus. 15: 100-131. Moore, J., and Gipson, R., 1981. — The Geonemertes problem. J. Zool. Lond. 194: 175-201. NEWELL, B. S., and BARBER, W. E., 1975. — Estuaries important to Australian Fisheries. Aust. Fash. 34: 17-22. NEWELL, S. Y., 1974. — Mangrove fungi; the succession in the mycoflora of red mangrove Rhzzophora mangel L. seedling. Fungz: 51-88. Opum, H. T., and HEatp, E. J., 1975. — Mangrove forests and aquatic productivity. In: A. D. HasLeEr, (ed.), Coupling of land and water systems ; ecological studies 10. New York: Springer-Verlag. Onur, C. P., TEAL, J. M., and VaAuiELa, I., 1977. — Interactions of nutrients, plant growth and herbivory in a mangrove ecosystem. Ecology 58: 514-526. Perry, W. J., 1946. — Observations on the bionomics of the principal malaria vector in the New Hebrides, Solomon Islands. J. natn. Malar. Soc. 5: 127-139. PIANKA, E. R., 1978. — Evolutzonary ecology. 2nd Edition. New York: Harper and Row. POLLARD, D. A., 1976. — Estuaries must be protected. Aust. Fésh. 35: 1-5. ———, 1981. — Estuaries are valuable contributors to fisheries production. Aust. Fesh. 40(1) : 7-9. RecHER, H. F., 1971. — Bird species diversity: a review of the relation between species number and environment. Proc. Ecol. Soc. Aust. 6: 135-152. , and Hurcuincs, P. A., 1980. — The waterlogged forest. Aust. Nat. Hist. 20: 87-96. REDFIELD, J. A., 1982. — Trophic relationships in mangrove communities. In: B. F. CLoucu, (ed.), Structure function and management of mangrove ecosystems in Australia. Canberra: ANU Press. REYE, E. J., 1969a. — Mapping the habitat of Culicozdes subtmmaculatus. Univ. Qld. Dep. Ent., roneoed, 1 pp. ——, 1969b. — Larval survey of C. swhimmaculatus habitats. Univ. Qld. Dep. Ent., roneoed, 2 pp. ——, 1971. — [Untitled. ] Ceratopogonidae. Inf. Exch. 7: 3-4. ——, 1972a. — Pest biting midges. Some observations, December 1972. Univ. Qld. Dep. Ent., roneoed, 8 RuELLO, N. V., 1973. — The influence of rainfall on the distribution and abundance of the school prawn Metapenaeus macleay: in the Hunter River, Australia. Mar. Bzol. 23: 221-228. SAENGER, P., SpecHT, M. M., SpecHT, R. L., and CHApMaN, V. J., 1977. — Mangrove and coastal salt- marsh communities in Australasia. In (pp. 293-345) : Ecosystems of the world. Vol. 1. Wet coastal formations. Amsterdam: Elsevier Publ. Co. ———, STEPHENSON, W. and Moverty, J., 1979. — The subtidal fouling organisms of the Calliope River and Auckland Creek, central Queensland. Mem. Qd Mus. 19: 399-412. ——, 1980. — The estuarine macrobenthos of the Calliope River and Auckland Creek, Queensland. Mem. Qd Mus. 20: 143-161. ScHoppE, R. Mason, I. J., and Gitt, H. B., 1982. — The avifauna of the Australian mangroves — A brief review of composition, structure and origin. In: B. F. CLoucn, (ed.), Structure, function and Proc. Linn. Soc. N.S.W., 106 (1), (1981) 1982 108 FAUNA OF AUSTRALIAN MANGROVES management of mangrove ecosystems in Australia. Canberra: ANU Press. SETON, D. H. C., 1971. — Mangrove rookery near Ayr, Queensland. Aust. Berdwatcher 4: 96-97. SHANCO, P., and Timmins, R. D., 1975. — Reconnaissance of southern Bustard Bay tidal wetlands. Operculum 4 (3-4) : 149-154. SHINE, R., Ettway, C. P., and HEGERL, E. J., 1973. — A biological survey of the Tallebudgera Creek estuary. Operculum 3 (5-6) : 59-83. i SIMBERLOFF, D. S., BROwN, B. J., and Lowrie, S., 1978. — Isopod and insect root borers may benefit Florida. Sczence 201: 630-632. ——, and Witson, E. O., 1970. — Experimental zoogeography of islands. A two-year record of colonization. Ecology 51: 934-937. SINcLaIR, P., 1976. — Notes on the biology of the salt marsh mosquito Aedes vigzlax (Skuse) in south eastern Queensland. Qd Nat. 21: 134-139. SmiTH, B., CoLEMAN, N., and Watson, J. E., 1975. — The invertebrate fauna of Westernport Bay. Proc. R. Soc. Vict. 87: 149-155. SPENCER, K. A., 1977. — A revision of the Australian Agromyzidae (Diptera). West Aust. Mus. Special Publ. 8: 1-255. STAPLES, D. J., 1980a. — Ecology of juvenile and adolescent banana prawns Penaeus merquiensis de Man, in a mangrove estuary and adjacent offshore area of the Gulf of Carpentaria. I. Immigration and settlement of postlarvae. Aust. J. mar. Freshwater Res. 31: 635-652. ——, 1980b. — Ecology of juvenile and adolescent banana prawns Penaeus merquiensis, in a mangrove estuary and adjacent offshore area of the Gulf of Carpentaria. II. Emigration, population structure and growth of juveniles. Aust. J. mar. Freshwater Res. 31: 653-665. STEPHENSON, W., ENDEAN, R., and BENNETT, I., 1958. — An ecological survey of the marine fauna of Low Isles, Queensland. Aust. J. mar. Freshwater Res. 9: 261-317. SWANSON, S., 1976. — Lizards of Australia. Sydney: Angus and Robertson. Taytor, J. A., 1979. — The foods and feeding habit of subadult Crocodylus porosus Schneider in northern Australia. Aust. Wildl. Res. 6: 347-359. Tuom, B. G., WricuT, L. D., and Coteman, J. M., 1975. — Mangrove ecology and deltaic estuarine geomorphology, Cambridge Gulf-Ord River, Western Australia. J. Ecol. 63: 203-232. TOMLINSON, P. B., Bunt, J. S., PRimack, R. B., and Duke, N. C., 1978. — Lummnitzera rosea (Combretaceae) — its status and floral morphology. J. Arnold Arboretum 59 (4) : 342-351. ——, Primack, R. B., and Bunt, J. S., 1979. — Preliminary observations on floral biology in mangrove Rhizophoraceae. Biotropica 11 (3) : 256-277. TURNER, R. D., et al., 1972. — Survey of marine borers. The family Teredinidae in Australian waters. CSIRO Project P. 5-11. Univ. NSW Project 12-045-15. 1970-1972. 158 pp. 34 maps. TuRNER, R., and McKay, J., 1979. — Banka nezialia n.sp. Mollusca: Bivalvia: Teredinidae from Australia and New Zealand and its relationships. J. Roy. Soc. New Zealand 9:453-473. VANDERPLANK, F. L., 1960. — The bionomics and ecology of the red tree ant Oecophylla sp., and its relationship to the coconut bug Pseudotheraptus way: Brown (Coreidae). J. Anim. Ecol 29: 15-33. VaN Dyck, S., BAKER, W. W., and GILLETTE, D. D., 1979. — The false water rat, Xeromys myozdes on Stradbroke Island, a new locality in southeastern Queensland. Proc. Roy. Soc. Qd 90: 84. Warp, M., 1933. — New genera and species of marine Decapoda Brachyura from the coasts of New South Wales and Queensland. Aust. Zool. 7(5) : 377-394. WEATE, P. B., 1973. — A study of the fauna of littoral and sublittoral zones at Towra Point, Botany Bay. (Unpublished manuscript, 17 pp.). ——, 1975. — A study of the wetlands of the Myall River. Operculum 4: 105-113. Wess, G. J. H., MEssett, H., and MAGNusson, W., 1977. — The nesting of Crocodylus porosus in Arnhem Land, Northern Australia. Copeza 2: 238-249. WELLS, F. E., and Stack-SmiTH, S. M., 1981. — Zonation of molluscs in a mangrove swamp in the Kimberley, Western Australia. In. Bzological Survey of Mitchell Plateau and Admiralty Gulf, Kimberly, Western Australia. Part 9: 265-274. Perth: Western Australian Museum. WEscoTT, G., 1976. — The intertidal crabs of Victoria. An introduction, checklist and key to adults. Vac. Nat. 93 (6) : 237-245. WESTLAKE, D. F., 1963. — Comparisons of plant productivity. Bzo. Rev. 38: 385-425. White, H. L., 1917. — Northern Australian birds. Emu 16: 117-168. WILLIAMS, W. T., and Bunt, J. S., 1980. — Studies in the analysis of data from Australian tidal forests (‘Mangroves’). II. The use of an asymmetric monothetic divisive classificatory program. Aust. J. Ecol. 5: 391-396. ——, Bunt, J.S., and Duke, N. C., 1981. — Mangrove litter fall in north eastern Australia. II. Periodicity. Aust. J. Bot. 29(5) : 555-563. WILSON, E. O., 1964. — The ants of the Florida keys. Breviora 210: 14 pp. ——, and Simsertorr, D. S., 1969a. — Experimental zoogeography of islands. Defaunation and Proc. Linn. Soc. N.S.W., 106 (1), (1981) 1982 P. A. HUTCHINGS AND H. F. RECHER 109 monitoring techniques. Ecology 50: 267-278. —— 1969b. Experimental zoogeography of islands. The colonisation of empty islands. Ecology 50: 278- 296. WonmersLey, H. B. S., and Epmonps, S. J., 1958. — A general account of the intertidal ecology of South Australian coasts. Aust. J. mar. Freshwater Res. 9: 217-260. YaTEs, R., 1978. — Aspects of the ecology and reproductive biology of crabs in a mangrove swamp at Patonga Creek, NSW. Sydney: University of Sydney, M.Sc. thesis, unpubl. ZitcH, A., 1959. — Gastropoda von Wilhelm Wenz, Teil 2. Euthyneura fortgesetzt von Adolf Zilch 1959. Handbuch der Paldaozoologie, Band 6. Berlin: Gebriider Borntraeger. Proc. Linn. Soc. N.S.W., 106 (1), (1981) 1982 110 FAUNA OF AUSTRALIAN MANGROVES APPENDIX 1 POLYCHAETES RECORDED FROM AUSTRALIAN MANGROVES NT Qld NSW Vict. SA WA Nth Cent. Sth POLYCHAETA Polynoidae Lepidonotus sp. xX Sigalionidae xX Amphinomidae Eurythoe complanata (Pallas) ».4 Phyllodocidae Phyllodoce duplex McIntosh xX P. malmgreni Gravier Xx x P. novaehollandiae Kinberg x Phyllodoce sp. x Nereididae Australonerezs ehlers: (Augener) Ceratonerezs erythraeensis Fauvel* xX C. mirabilis Kinberg xx x es Namalycastis abiuma Mueller Namalycastis sp. x Namanereis quadraticeps (Blanchard) x Neanthes cricognatha (Ehlers) x N. vaalu Kinberg x Nerezs uncinula (Russell) : Nerets sp. X Perinerevs vallata (Grube) x Platynereis cf. dumerili antipoda Hartman x x x Pseudonerets rottnestiana (Augener) Xx Nephtyidae Nephtys australiensis Fauchald xX Xx x Glyceridae Glycera americana Leidy Glycera sp. Gonzada sp. x x Eunicidae Eunice antennata (Savigny) xX Marphysa sanguinea Montagu Xx x Marphysa sp. x Lumbrineridae Lumbrineris sp. x Onuphidae Diopatra dentata Savigny x x Orbiniidae Leztoscoloplos sp. Xx Scoloplos (S.) simplex (Hutchings) xX Scoloplos sp. Xx x Spionidae Boccardia sp. Xx Polydora sp. xX Prionospio sp. Xx Scolecolepis indica Fauvel Xx Spionidae spp. Xx Magelonidae x Chaetopteridae Chaetopterus sp. x DS Xx Proc. Linn. Soc. N.S.W.,106 (1), (1981) 1982 P. A. HUTCHINGS AND H. F. RECHER 111 APPENDIX 1 POLYCHAETES RECORDED FROM AUSTRALIAN MANGROVES NT Qld NSW Vict. SA WA Nth Cent. Sth Mesochaetopterus minutus Potts x Cirratulidae Cirriformia tentaculata Montagu Xx Opheliidae Armandtia intermedia Fauvel x Capitellidae Barantolla lepte Hutchings x x Capitella capitata (Fabricius) x Notomastus torquatus (Hutchings & Rainer) x x Notomastus sp. Xx Xx Maldanidae Euclymene sp. x Oweniidae Owenza fuszformis Delle Chiaje xX X Terebellidae Hadrachaeta aspeta Hutchings Xx Lysilla pacifica Hessle x Pista sp. x Rhinothelepus sp. xX Thelepus setosus (Quatrefages) x Amphitrite rubra (Risso) Xx Terebella ehrenbergz Xx Sabellidae : x Laonome sp. xX Serpulidae Ficopomatus uschakow (Pillai) Xx Galeolaria caespitosa Savigny x Salmacina sp. Xx *This species has recently been split into several species, by Hutchings and Turvey (manuscript) and Hutchings and Glasby (in press) . Ceratonerezs erythraeensis probably does not occur in Australia. Sources of references : Butler et al., 1977; Hutchings and Recher, 1974; Hutchings et al., 1977; McCormick, 1978; Queensland Museum (unpublished report), 1974; Saenger et al., 1977; also unpublished data of Hutchings. Proc. Linn. Soc. N.S.W., 106 (1), (1981) 1982 112 FAUNA OF AUSTRALIAN MANGROVES APPENDIX 2 CRUSTACEANS RECORDED FROM AUSTRALIAN MANGROVES NT Qld NSW Vict. SA WA Nth Cent. Sth CIRRIPEDIA Iblidae Ibla cumingz Darwin ».« Balanidae Balanus amphitrite Darwin X x X xX x xX X Elminius modestus Darwin xX Tetraclitidae Tetraclita coerulescens (Splenger) T. squamosa (Bruguiére) T. vitiata Darwin Chthamalidae Chamaesipho columna (Spengler) xX Chthamalus caudatus Pilsbury Xx xX C. malayenszs Pilsbury xX Xx x x Tanaidacea Paratanaidae Paratanais sp. Xx Isopoda Anthuridae Cyathura sp. xX Haliophasma sp. x Cirolanidae Cirolana sp. x Limnoriidae Limnoria lignorum (Rathke) Limnorza sp. x Sphaeromatidae Amphrovdella sp. Eubranchiata sp. Exosphaeroma ali Baker E. alata Baker xX Exosphaeroma sp. xX Pseudosphaeroma sp. Sphaeroma quoyanum Milne-Edwards Xx Xx Xx S. terebrans Baker Xx S. walkerz Stebbing Sphaeroma sp. Xx x Chitonopsts sp. Ligiidae Ligza australiensts (Dana) Ligia sp. Xx Amphipoda Cheluridae Chelura terebrans Philippi xX Corophiidae Corophium sp. xX xX xX Ericthonzus sp. x Gammaridae Victorzopisa australiensts (Chilton) Xx Talitridae Orchestza sp. x Talorchestza sp. xX mK KM KK x -K KM * * Proc. Linn. Soc. N.S.W., 106 (1), (1981) 1982 P. A. HUTCHINGS AND H. F. RECHER 113 APPENDIX 2 CRUSTACEANS RECORDED FROM AUSTRALIAN MANGROVES NT Qld NSW Vict. SA WA Nth Cent. Sth Hyalellidae Parhyallela sp. x Indet sp. x DECAPODA Penaeidae Metapenaeus macleay: (Haswell) xX Penaeus latisulcatus Kishinouye x P. plebejus Hess Xx Alpheidae Alpheus edwardsi (Andouin) x x Alpheus sp. xX Atyidae Paratya sp.: Xx Palaemonidae Leander intermedius Stimpson x L. littoreus McCulloch x Macrobrachium intermedzum Stimpson x Palaemon serenus Heller Xx Callianassidae Callianassa australiensis Dana x Xx 4 Laomediidae Laomediza healyi Yaldwyn & Wear Xx Grapsidae Eriocheir spinosa Milne-Edwards x Xanthidae Pilumnopaeus serratifrons Kinahan x Ocypodidae Scopimerinae Ilyoplax dentata Ward Ilyoplax sp. x Scopimera inflata Milne-Edwards Scopimera sp. Tmethypocoelis sp. X Heloecius cordiformzs (Milne-Edwards) x Xx x Ocypode ceratophthalma (Pallas) x x Xx ».4 O. convexa Quoy & Gaimard O. cordimanus Desmarest x ».« x Xx O. fabricit Milne-Edwards Uca capricornis Crane Uca sp. A Ucasp. B Uca sp. C Uca dussumierz (Milne-Edwards) Xx Xx U. australiae Crane U. coarctata Milne-Edwards Xx U. flammula Crane X U. signata (Hess) Xx U. longzdigita (Kingsley) U. seismella Crane x xX U. polzta Crane xX X U. tetragonon (Herbst) X x KX ~ xX xX ~ x XM mx xX KK KM xX ~ * ~*~ Pata va Km KM mM mM OM ea KK MK xx mK MK Proc. Linn. Soc. N.S.W., 106 (1), (1981) 1982 114 FAUNA OF AUSTRALIAN MANGROVES APPENDIX 2 CRUSTACEANS RECORDED FROM AUSTRALIAN MANGROVES NT Qld NSW Vict. SA Nth Cent. Sth U. dampieri Crane x U. vomeris McNeill x x 4 x Xx U. triangularts (Milne-Edwards) x x U. mjobergi Rathbun x x Xx Xx U. perplexa (Milne-Edwards) Xx x X Macrophthalminae Australoplax trindentata (Milne-Edwards) Xx x XxX ».4 Cletstostoma wardi Rathbun xX Xx x Letpocten sordidulum Kemp xX Xx xX Macrophthalmus abercrombiez Barnes x M. boscw Audouin & Savigny xX xX x M. convexus Stimpson Xx Xx Xx M. crassipes Milne-Edwards XxX D4 x x x M. darwinensis Barnes D4 M. japonicus (De Haan) M. latifrons Haswell x M. latrezllec (Desmarest) Xx Xx Xx ».« M. pacificus Dana Xx x x x M. punctulatus Miers xX xX Xx xX M. setosus Milne-Edwards x x Xx M. telescopicus (Owen) xX xX x xX Paracleistostoma mcneilli (Ward) ».4 x x Mictyridae Mictyris liungstonez McNeill ».« Xx XxX Xx M. longicarpus Latreille xX xX xX xX xX Xx M. platycheles Milne-Edwards Xx Xx Xx Grapsidae Grapsinae Metopograpsus frontalzs Miers xX x Xx Xx M. latzfrons (White) xX x M. quadridentatus Stimpson x Metopograpsus sp. Xx x x x Ilyograpsus paludicola (Rathbun) xX Xx ?Ilyograpsus sp. Xx Varuninae Varuna litterata (Fabricius) xX xX x x Sesarminae Clistocoeloma merguiense De Man Xx Helice leachzi Hess wm eM x x xX xx XK ~ Pad Helograpsus haswellianus (Whitelegge) Nannosesarma sp. A Nannosesarma sp. B Xx Xx x Xx Nannosesarma sp. C Xx x x Paragrapsus gaimardz Milne-Edwards P. laevis (Dana) x Xx P. quadridentalus Milne-Edwards x x xX Sarmatium crassum Dana x x x Sesarma (Chiromantes) brevicristatum x Campbell S. (Chzromantes) darwiniensis x Campbell ’ Proc. Linn. Soc. N.S.W., 106 (1), (1981) 1982 P. A. HUTCHINGS AND H. F. RECHER MOLLUSCS RECORDED FROM AUSTRALIAN MANGROVES S. (Chiromantes) liidum Milne-Edwards S. (Chiromantes) messa Campbell S. (Chiromantes) semperz longicristatum Campbell S. (Chiromantes) semperz 2semperz Burger S. (Holometopus) sp. 1 S. (Holometopus) sp. 2 S. (Holometopus) sp. 3 S S . (Parasesarma) erythrodactyla Hess . (Parasesarma) leptosoma Hilgendorf (Parasesarma) ?moluccensis De Man (Parasesarma) sp. 1 . (Parasesarma) sp. 2 . (Bresedium) sp. (aff. brevepes) (Bresedium) sp. (Neoepzsesarma) sp. (aff. brockzz) (Neoepisesarma) sp. (2Neoepisesarma) sp. . (Neosarmatium) fourmanozri Séréne ) ANNHAHAHHAHY . (Neosarmatium) meznertz De Man . (Neosarmatium) sp. . (Neosesarma) sp. . (Sesarmozdes) kraussz borneenszs Tweedie S. (Tiomanium) indica Milne- Edwards Xanthidae Epixanthus dentatus (White) E. frontalis Milne-Edwards Heteropanope glabra Stimpson H. serratifrons (Kinahan) AnHnAN ?Heteropilumnus sp. Myomenippe fornasinz (Bianconi) Unident. sp. Goneplacidae ?Rhizopa sp. Speocarcinus sp. Portunidae — Scylla serrata (Forskal) Hymenpsomatidae Elamenopsis aspinifera Lucas E. lineata Milne-Edwards E. octagonalis (Kemp) E. torrensica Lucas Elamenopsis sp. 3 Elamenopsis sp. 4 Halicarcinus bedford: Montgomery APPENDIX 2 NT Qld Nth Cent. xX x x x xX x x x x X x x xX xX xX x xX xX x x xX xX xX xX Pr xX x xX x xX xX x xX xX xX xX xX xX xX xX x xX xX X xX xX xX xX MK xX xX xX x XK xX xm x Xx NSW Vict. x Nth x xX xX x xX SA 115 WA Xx Proc. Linn. Soc. N.S.W., 106 (1), (1981) 1982 116 FAUNA OF AUSTRALIAN MANGROVES APPENDIX 2 (continued) Amarinus laevis (Targioni Tozzetti) Xx x A. paralacustris (Lucas) Xx Xx xX Xx A. latinasus Lucas x Hymenosoma hodgkint Lucas x x Xx x Sources of References: Aust. Litt. Soc., 1977; Barnes, 1967, 1971; Butler, 1973, 1974; Butler et al., 1975, 1977; Campbell, 1967; Campbell and Griffin, 1966; Crane, 1975; Davie (unpub.); Ellway, 1974; George and Jones (in press) ; George and Knott, 1965; Graham e¢ al., 1975; Hale, 1927; Hegerl and Tarte, 1974; Hegerl et al., 1979; Holdich and Harrison, 1980; Hutchings and Recher, 1974; Hutchings et al., 1977; Lucas, 1980; McCormick, 1978; McNeill, 1926; Shanco and Timmons, 1975; Shine et al., 1973; Saenger, 1977; Ward, 1933; Weate, 1973 (unpub.), 1977; Wescott, 1976. Davie in his unpublished manuscript has verified many of the decapod identifications but the other crustaceans are based entirely on the literature. So some of the identifications and their distributions may be inaccurate. Proc. Linn. Soc. N.S.W., 106 (1), (1981) 1982 P. A. HUTCHINGS AND H. F. RECHER 117 APPENDIX 3 MOLLUSCS RECORDED FROM AUSTRALIAN MANGROVES NT Qld NSW Vict. SA WA Nth Cent. Sth POLYPLACOPHORA Acanthozostera gemmata Iredale & Hull xX x x GASTROPODA Fissurellidae Montfortula rugosa (Quoy & Gaimard) x Xx Patellidae Cellana tramoserica (Sowerby) Xx Xx x x Acmaeidae Patelloida mimula (Iredale) x x x x xX Trochidae Astrocochlea adelaidae (Philippi) A. concamerata (Wood) XxX A. constricta (Lamarck) x Xx Xx Xx Xx A. torr¢ Cotton & Godfrey Euchelus atratus (Gmelin) Montodonta labio (L.) Phastanotrochus sp. Prothalotia comtessz Iredale Neritidae Nerzta atramentosa Reeve N. albicilla L. N. chamaeleon L. N. lineata Gmelin N. planospira Anton N. plicata L. N. ualenensis Lesson Nerztena crepidularza Lamarck Littorinidae Bembicium auratum (Quoy & Gaimard) x B. melanostomum (Gmelin) Littorina acutispira Smith L. irrorata Say L. scabra (L.) L. undulata Gray Peasvella tantilla (Gould) Hydrobiidae Hydrobia buccinoides (Quoy & Gaimard) Posticobia braziert (Smith) x Tatea kestevenz Iredale T. rufilabris (Adams) Stenothyridae Stenothyra sp. x X x Iravadiidae Travadza sp. ».« xX x Assimineidae Assiminea relata Cotton 4 Xx A. tasmanica (1. Woods) Xx Assiminea sp. Xx x Turritellidae Turritella terebra (L.) Xx x Xx xX m~ X mm wm * ra * x xX “KM mK K mm KK rad ms x x MX ~*~ xX me Mm Mh Mm OM mr xX x Kx ~ x KM mm ra ~ mM mM Planaxidae Planaxis sulcatus (Born) Quoyza decollata (Quoy & Gaimard) Rissoidea Rissozna sp. x xX Vitrinellidae ~ x Proc. Linn. Soc. N.S.W., 106 (1), (1981) 1982 118 FAUNA OF AUSTRALIAN MANGROVES APPENDIX 3 (continued) Pseudoliota sp. Potamididae Cerithidea largilliertz (Philippi) C. obtusa (Lamarck) C. cingulata (Gmelin) Pyrazus ebeninus (Bruguiére) Telescopium telescopium (L.) Terebralia palustris (L.) T. sulcata (Born) Velacumantus australis (Quoy & Gaimard) Cerithiidae Clypeomorus carbonarius (Sowerby) Strombidae Lambis lambis (L.) Strombus luhuanus (L.) Naticidae Polinices conicus (Lamarck) P. melanostoma (Swainson) P. sordida (Swainson) Muricidae Bedeva hanley: (Angas) Cronia aurantiaca (Hombron & Jacquinot) Homalocantha secunda (Lamarck) Lepszella vinosa (Quoy & Gaimard) Morula marginalba (Blainville) Naguetia capucina (Lamarck) Buccinidae Cominella eburnea (Reeve) C. lineolata (Lamarck) Melongenidae Volema cochlidium (L.) Nassariidae Nassartus burchard: (Dunker zn Philippi) N. cf. dorsatus (Roding) N. oltvaceus (Bruguiére) N. melaniozdes Reeve N. pauperatus (Lamarck) N. jonasiz (Dunker) N. pullus (Linné) Collumbellidae Columbella duclostana Pyrene scripta (Lamarck) Zafra sp. Mitridae Mitra retusa Lamarck Bullinidae Bullina lineata (Gray) Haminoeidae Atys naucum (L.) Haminoea sp. Ellobiidae Cassidula angulifera (Petit) C. nucleus (Gmelin) C. rugata Menke C. sowerbyana (Pfeioffer) Melampus striatus (Pease) Proc. Linn. Soc. N.S.W., 106 (1), (1981) 1982 NT ta em KM KM axe KM Qld Nth Cent. x x x x Xx xX x x XxX x ».4 >.< xX xX x xX x x x XxX x x x x x x x x Xx X x x X x x X x x x NSW Vict. SA WA Sth Xx x x Xx x x x Xx x Xx x x x x x x X x x x x x x x x x x x xX x x x x x x x x x x x x x x x x x x x x x x X x x x x x x P. A. HUTCHINGS AND H. F. RECHER 119 APPENDIX 3 (continued) NT Qld NSW Vict. SA WA Nth Cent. Sth M. castaneus (Muehlfeldt) Ellobium aurisjudae (L.) Xx Melosidula granulosa Iredale M. zonata (Adams & Adams) Ophicardelus ornatus (Férussac) O. quoy: (Adams & Adams) O. sulcatus (Adams & Adams) xX Pythia scarabaeus (L.) x Amphibolidae Salinator burmana (Blanford) S. fragilis (Lamarck) S. solida von Martens Siphonariidae Stphonaria bifurcatus Reeve S. dentzculata Quoy & Gaimard S. marza (Iredale) Onchidiidae Onchidina australis Semper O. beutschlw Stantschinsky O. daemelu Semper O. verruculatum Cuvier xX xX Oncis chameleon (Brazier) Oncis sp. x Aplysiidae A plysza sp. xX Bursatella leachi (Blainville) BIVALVIA Solemyidae Solemya teraereginae (Iredale) x Arcidae Anadara trapezia (Deshayes) ».4 x D4 Arca sp. x Barbatia sp. x x Mytilidae Amygdalum beddomez Iredale x Modiolus inconstans (Dunker) Xx Xx Musculus variocosus (Gould) Mytilus edulis L. Trichomya hirsuta (Lamarck) Xenostrobus pulex (Lamarck) x X. securis (Lamarck) Isognomonidae Isognomon isognomon (L.) Tsognomon’ nucleus (Lamarck) Xx x Isognomon cf. wtrea (Reeve) X Isognomon sp. Xx Pinnidae Pinna bicolor (Menke) x Xx x Pteriidae Pinctada margaritifera (L.) xX x Ostreidae Ostrea nomades Iredale xX Saccostrea amasa (Iredale) Xx x xX S. cuccullata commerczalis (Iredale & Roughley) xX x x Xx Xx S. echinata (Quoy & Gaimard) x Anomiidae Xx maxx KM KM ~*~ mm mM MK mx mM mx Xs xx KK mx xx KK mw ~ em ~*~ x mx x xxx KM ~*~ x * xx x x KM eM OM Km x KM Proc. Linn. Soc. N.S.W.,106 (1), (1981) 1982 120 FAUNA OF AUSTRALIAN MANGROVES APPENDIX 3 (continued) Enigmonia aenigmatica (L.) Patro australis (Gray) Lucinidae Arthritica helmsz (Hedley) Cavatidens omissa Iredale Leptonidae Lasaea australzs (Lamarck) Geloinidae Batzssa violacea (Deshayes) Geloina coaxans (Gmelin) Veneridae Circe tumidum (Bolten) Paphia hiantina (Lamarck) Tapes wathngi Iredale Venerupis crenata (Lamarck) V. crebrelamellata Tate Tellinidae Tellina deltordalis Lamarck T. australis Deshayes T. capsoides Lamarck Mesodesmatidae Mesodesma altenai de Rooij Schuiling Mactridae Notospisula trigonella (Lamarck) Glauconomidae Glauconome cf. cumingz (Prime) G. virens (L.) G. plankta (Iredale) Teredinidae Bankia australis (Calman) . campanelata Moll & Roch . carinata (Grey) . bipalmulata (Lamarck) . graciles Moll . nord Moll . rochi Moll Bactranophorus thoracitzes (Gould) Bactranophorus sp. Dicyathifer manni (Wright) Lyrodus massa (Lamy) L. medilobata (Edmonson) L. pedicellatus (Quatrefages) L. bipartita (Jeffreys) Nausitora dunlope: Wright N. globosa Sivickis Nototeredo edax (Hedley) Teredo bartschi Clapp T. johnsoni Clapp T. navalis (L.) T. furcifera Von Martens T. clappi Bartsch T. poculzfer Iredale T. princessae (Sivickis) T. trranularis Edmonson Hiatellidae Fluviolanatus armarus Laseron Pholadidae Bowswwns xx KK * x KKK Proc. LINN. Soc. N.S.W., 106 (1), (1981) 1982 me ~ ~*~ Xs ead wm mK AK KKK KM x Kx AK mK mM KM KK OM we ax x x ~~ xX mK mK ~ x x xX nm x ~ axe KKM KM KX * wx x ~*~ x P. A. HUTCHINGS AND H. F. RECHER 12] APPENDIX 3 (continued ) Martesza striata L. x Laternulidae Laternula creccina Reeve axe x L. tasmanica (Reeve) x L. vagina (Reeve) x x Sources of references: Aust. Litt. Soc., 1977; Butler, 1973, 1974, et al. 1975, 1977; Ellway, 1974; Graham et al., 1975; Hegerl and Davie, 1977, et al. 1979; Hutchings and Recher, 1974; Hutchings et al., 1977; McCormick, 1978; Shine et al., 1973; Saenger et al., 1976; Turner et al., 1972; Weate, 1973 (unpub.), 1975; Wells and Slack-Smith, 1981; also unpublished data of Hutchings and Australian Littoral Society. The species in this appendix are based entirely on the literature and the material has not been verified so some of the identifications and their distributions may be inaccurate. Proc. Linn. Soc. N.S.W., 106 (1), (1981) 1982 Notes and Discussion A. W.H. HUMPHREY, HIS MAJESTY’S MINERALOGIST IN NEW SOUTH WALES, 1803-12 — ACOMMENT R. J. FORD Department of Geology, University of Tasmania, Box 252C, G.P.O., Hobart, 7001. In his paper concerning the activities of the mineralogist, A. W. H. Humphrey, particularly with respect to his mineral collecting in Van Diemen’s Land, Vallance (1981) was puzzled by the lack of any subsequent knowledge of the occurrence of the ‘green garnets’ that were collected by Humphrey and Robert Brown during an expedition to the Huon River district in 1804 and may have been like those offered for sale in 1812 by Heuland. The source of the ‘garnets’ and also of the ‘pitchstone’ mentioned by Humphrey as associated with them has until recently been obscure. Observations now reported indicate a distinctive provenance and, furthermore, show that Humphrey’s ‘green garnet’ is an epidote with unusual habit. A short summary of the geology of Port Cygnet is in order here, at least with respect to the Cretaceous alkaline igneous rocks for which the district is noted. There are three distinct groups of such rocks in the area: (a)a group referred to as syenite porphyries that contain as main minerals plagioclase as well as potash feldspar, (b) rocks usually bearing phenocrysts of sanidine, and termed sanidine porphyries, in which the feldspar is almost exclusively potash feldspar, and (c)hybrid rocks, produced by reaction between potassium-rich magma (from which the sanidine porphyries have formed) and pre-existing Jurassic dolerite. The alkaline rocks occur at Port Cygnet as small dykes, sills and irregular intrusions into Permian and Jurassic country rocks. The rock of major significance here is the garnet trachyte described by Macleod and White (1900). This is related to the sanidine porphyries, but lacks the porphyritic texture due to absence of sanidine phenocrysts. It is most unusual in bearing phenocrysts of light-brown spessartine garnet (Ford, 1967), represented by sporadic euhedral crystals but more usually rounded though with some faces well developed, and also less abundant phenocrysts of rounded, dark-green epidote grains. The usual habit of crystallized epidote is columnar and striated with the crystal elongated parallel to the ‘b’ crystallographic axis which is in contrast to that observed for this rock (Fig. 1) where crystal faces are not developed. Massive and granular epidote is more typical of skarns and contact metamorphic zones. Epidote in this rock from Port Cygnet was not reported by Macleod and White, who were concerned with the garnet, and, indeed, is not present in their specimen (No. X 1435), now in the collection of the Tasmanian Museum. Both sets of phenocrysts range in size from less than 1 mm to about 10 mm in diameter and commonly bear white rims of potash feldspar which has Proc. Linn. Soc. N.S.W., 106 (1), (1981) 1982 124 NOTES AND DISCUSSION Fig. 1. Garnet trachyte specimen from Port Cygnet showing rounded phenocrysts of epidote with white rims of potash feldspar. One cent coin (17.5 mm diam.) added for scale. apparently crystallized onto the phenocrysts (Fig. 1). The groundmass of the rock is dense, dark grey in colour. It is made up of closely-packed fine-grained potash feldspar laths showing a trachytic flow structure and also some patches of iron chlorite and occasional biotite. This rock, unique to the area, occurs as a near-vertical dyke, approximately 0.6 m thick and exposed for about 20 m on the shoreline of Port Cygnet north of Langdon’s Point at the locality having co-ordinates 5217 750N, 506 800E on the D’Entrecasteaux sheet of the Tasmanian 1:100 000 topographic series. The origin of the epidote in the garnet trachyte is enigmatic. It shows no reaction relationship to its environment and it is difficult to determine whether the phase is a primary crystallate or a relict from older rocks. Bulk chemistry, particularly trace elements and an MnO content of about 2%, suggests the trachyte formed from a parent potassic magma which had been modified by assimilation of other material. This is certainly the source of Humphrey’s ‘green garnet’. One would like to think that he correctly identified the brown garnet by crystal form and then concluded that the epidote was a green variety since these two minerals are texturally very similar in the host rock. The fine, dark-grey groundmass could also constitute the pitchstone matrix referred to by Humphrey (Vallance, 1981: 139). The locality of the outcrop is marked on the accompanying map (Fig. 2). This would confirm Vallance’s deduction of their having encountered the Port Cygnet estuary on their way back to Hobart. It is quite possible that Humphrey and Brown came to Port Cygnet at the point where the garnet trachyte outcrops, because within about 100 metres north and south of the above, there are prominent outcrops of sanidine porphyry dykes with spectacular euhedral phenocrysts of (010) tablets of yellow-white sanidine crystals up to 40 mm maximum dimension, and having a swirl structure in a green aegirine-bearing matrix. In contrast the outcrop of the garnet trachyte is much more subdued. For someone like Humphrey who was looking for minerals it would seem to be incredible that he would fail to comment on the sanidine Proc. Linn. Soc. N.S.W., 106 (1), (1981) 1982 NOTES AND DISCUSSION MAP OF HUON DISTRICT Basalt centres Garnet trachyte RANELAGH HUONVILLE i <= w = & w Q GRID) NORTH DrENTRECASTEAUL Fig. 2. Map of the Huon district showing localities referred to in the text dykes had he seen them. Even today it is difficult to walk along the western shore of Port Cygnet unless the tide is very low. Hence it is most likely that Brown and Humphrey happened to come to the Port Cygnet shore at the garnet trachyte outcrop and skirted the estuary by walking along the tops of the shoreline banks. If Brown and Humphrey reached the Huon River by crossing the summit of Mt Wellington and following the Mountain River as indicated by Vallance (1981: 115), the only rock type they could have seen apart from sediments and dolerite would have been basalt. On this route there are two small outcrops of basalt — one being at the Mountain River bridge on the Ranelagh to Hobart road, the other near the summit of Mt Wellington close to the Australian Broadcasting Commission TV transmission tower. If a separate source for pitchstone is required these may qualify although Humphrey’s letter states that the ‘garnet’ has an intimate association with its pitchstone matrix. The Mountain River locality does not contain olivine nodules but these do occur in some abundance at the Mt Wellington locality. If Humphrey had collected from here because of its particular location, I am sure that he would have made special note of it. These latter points have been added for sake of completeness as the author believes the evidence supports the Port Cygnet locality as the source of A. W.H. Humphrey’s ‘green garnets’. ACKNOWLEDGEMENTS Following earlier correspondence with Professor Vallance I would express my Proc. Linn. Soc. N.S.W., 106 (1), (1981) 1982 125 126 NOTES AND DISCUSSION appreciation to the Society for inviting me to present these comments in the Proceedings and to Professor T. Vallance and Dr M. R. Banks for suggesting improvements to the manuscript. References Forp, R. J., 1967. — A re-appraisal of johnstonotite. Pap. Proc. Roy Soc. Tasm. 101: 11-12. Mac eop, W. A., and WuitE, O. E., 1900. — On the occurrence of a new species of garnet at Port Cygnet. Pap. Proc. Roy. Soc. Tasm. for the years 1898-1899: 74-76. VALLANCE, T. G., 1981. — The start of government science in Australia: A. W. H. Humphrey, His Majesty’s Mineralogist in New South Wales, 1803-12. Proc. Linn. Soc. N.S.W. 105 (2): 107-146. Communicated by T. G. Vallance ; accepted for publication 16 December 1981. Proc. Linn. Soc. N.S.W., 106 (1), (1981) 1982 PROCEEDINGS of the LINNEAN SOCIETY NEW SOUTH WALES VOLUME 106 _ NUMBER 2 Significance of Late Cambrian (Idamean) Fossils in the Cupala Creek Formation, northwestern New South Wales €. MCA. POWELL, G. NEEF, D. CRANE, P. A. JELL, and I. G. PERCIVAL PowELL, C. McA., Neer, G., CRANE, D., JELL, P. A., & PeRcivat, I. G. Significance of Late Cambrian (Idamean) fossils in the Cupala Creek Formation, northwestern New South Wales. Proc. Linn. Soc. N.S.W. 106 (2), (1981) 1982: 127-150. Late Cambrian (Idamean) fossils occur about 22 km southeast of Nuntherungie station, northwestern New South Wales, in the upper part of a terrigenous, transgressive succession (Cupala Creek Formation — new name) lacking carbonates. The succession, about 1 km thick, consists of a basal conglomeratic unit overlain by red and grey sandstone, the latter containing brachiopods, which in turn is overlain by olive-grey siltstone containing trilobites and brachiopods. The basal part of the formation represents one of the first pulses of clastics derived from the rising Delamerian mountains to the south. Sedimentation was probably rapid from streams intermediate between braided and meandering sinuosity. The middle and upper parts of the formation are paralic and neritic, respectively. The Cupala Creek Formation unconformably overlies the ?Vendian or Early Cambrian Copper Mine Range Beds, and is unconformably overlain by the ?Late Devonian Mulga Downs Group. The nearby Kandie Tank Limestone has the same stratigraphic relations as the Cupala Creek Formation, but is probably younger (post- Idamean — pre-Payntonian). Six trilobite species, including one new genus, Notoaphelaspis Jell, gen. nov., and one new species, N. orthocephalis Jell, sp. nov., two brachiopods and a mollusc are described. The discovery of fossils in the Cupala Creek Formation demonstrates that the Copper Mine Range Beds were folded before the Late Cambrian. The Cupala Creek Formation was subsequently folded about southeasterly-trending axes between Late Cambrian and Late Devonian, and the whole succession was again folded, probably in the Carboniferous. C. McA. Powell and D. Crane, School of Earth Sciences, Macquarie University, North Ryde, Australia 2113; G. Neef, W. S. and L. B. Robinson University College, University of New South Wales, P. O. Box 334, Broken Hill, Australia 2880; P. A. Jell, National Museum of Victoria, 285-321 Russell Street, Melbourne, Australia 3000, and I. G. Percival, Department of Geology and Geophysics, University of Sydney, Australia 2006; manuscript received 7 July 1981, accepted for publication in revised form 18 November 1981. INTRODUCTION The White Cliffs 1:250,000 geological map (Rose et al., 1964) shows a north- northwesterly-trending syncline on the northeastern side of, and truncated by, the more westerly-trending structure of Copper Mine Range. Sediments in the north- northwesterly trending syncline are marked as ?Proterozoic on the map, but suggested to be Lower Devonian in the explanatory notes (Rose, 1974) on the basis of cor- relation with rocks about 20 km south near Mt Daubney and Old Gnalta homestead where fragmentary plant remains had been found (Brown zn Freeman, 1965; Neef et al., in prep.). The overlying, more shallowly-dipping and westerly-trending quartzose sandstone and siltstone in Copper Mine Range were correlated by Rose (1974) with Proc. Linn. Soc. N.S.W., 106 (2), (1981) 1982 128 SIGNIFICANCE OF LATE CAMBRIAN FOSSILS the Middle to Upper Devonian Mulga Downs Group further east. The structure thus indicated is a mid-Devonian unconformity (Webby, 1972; Evans, 1977) , which two of us (Crane and Powell) set out to investigate as part of a more regional study of mid- Devonian tectonism. Upper Silurian to Lower Devonian xX eae Cretaceous to Recent Alluvium -” _[5" White, mature quartz MULGA DOWNS" sandstone and siltstone GROUP pe Angular Unconformity ={ Green/Grey siltstones and =| fine sandstones CUPALA Maroon to grey lithic sandstones SR are Lithic conglomerate Angular Unconformity Micaceous siltstones and shale of flyschoid aspect Bedding Bedding, younging determined —— Anticline. Syncline (with plunge) RR Minor folds — 2_ Mapped ——_ Probable Lees (with thro} Inferred indicated COPPERMINE RANGE BEDS --~—~sB Airphoto trend “Measured section A~ ©1 Marine fossil locality == Access track S| i A B Section line Thickness of conglomeratic unit in metres Fig. 1. Geological map and cross section of the Cupala Creek Formation. Proc. Linn. Soc. N.S.W., 106 (2), (1981) 1982 400 300 200 100 sea level C. McA. POWELL, G. NEEF, D. CRANE, P. A. JELL ANDI. G. PERCIVAL 129 Late Cambrian (Idamean) trilobites and brachiopods were found by Crane and Neef thus ruling out the correlation with the plant-bearing part of the Mt Daubney Beds (also known as the Cootawundy Beds, Pogson and Scheibner, 1971), which lie west of the Koonenberry Fault. Previous correlation of the two successions was tentative, on the basis of photogeological similarity (Rose et al., 1964; Wilson, 1967). Three of us (Crane, Neef and Powell) mapped the area, examining the stratigraphy, sedimentology and structure, and searching for fossils. Jell described the trilobites and mollusc, and Percival the brachiopods. The main purpose of this paper is to document the occurrence of Idamean fossils and to point out their significance in determining the timing of regional deformation in northwestern New South Wales. CUPALA CREEK FORMATION The strata, herein named Cupala Creek Formation, are well exposed over an area of 4.5 km? in the dry tributaries at the headwaters of Cupala Creek, and there are also exposures in the low hills between the tributaries (Fig. 1). The structure (discussed in detail below) is a faulted syncline that plunges gently south-southeast and is erosionally truncated by mature quartzarenites of the overlying Mulga Downs Group (Fig. 2). Approximately 1 km of sediment is present in the southern part of the area, where the type section was measured (Fig. 3, A-B on Figs 1 and 2). The stratigraphic succession consists of a basal conglomeratic unit overlain by a sandstone unit in which a pale-grey sandstone is sandwiched between a lower and upper red sandstone. The top of the formation is an olive-grey siltstone interbedded with very fine sandstone. In the northern part of the area, only the conglomeratic and sandstone units are present. Fig. 2. Oblique aerial photograph, viewing north, shows steeply dipping Cupala Creek Formation overlain by gently dipping Mulga Downs Group. Type section is A-B, and Cupala Creek is in the centre background. The most fossiliferous locality containing numerous trilobites (1 in Fig. 1) is at C. Proc. Linn. Soc. N.S.W., 106 (2), (1981) 1982 130 SIGNIFICANCE OF LATE CAMBRIAN FOSSILS Fossil Localities MULGA DOWNS 9! and 92 GROUP Metres silt very fine fine Ly) Trilobites 9) Outcrop of Cambrian © Station Homestead ~——= Major Fault trends 0 10 20 30 ———$- km Fig. 8. Distribution of Cambrian strata in northwestern New South Wales. CAMBRIAN DEPOSITS IN NORTHWESTERN NEW SOUTH WALES Cambrian strata are known from at least seven other areas in northwestern New South Wales (Rose and Brunker, 1969; Warris, 1967; Wopfner, 1966; and, more recently, Neef, unpub. data) (Fig. 8). A precise age to zone or stage level is known for two of these areas, and the age of the others is inferred by lithological correlation. Fig. 9 summarizes the present stage of knowledge. Cambrian deposits are most extensive near Mt Wright where Early to earliest Middle Cambrian limestone, shale and volcanics are separated from latest Cambrian strata by an unconformity (Opik, 1975; Kruse, 1978). Early Ordovician strata of the Mt Wright district can be traced southward to the Scopes Range (near Bilpa station) so that the underlying conformable strata are probably Late Cambrian (Rose and Brunker, 1969). The stratigraphic relationship of Early or Middle Cambrian Proc. LINN. Soc. N.S.W., 106 (2), (1981) 1982 ORDOVICIAN CAMBRIAN LATE BOOMERANGIAN Leiopyge Loevigato<—I7 uu I (=) a = > = oO <= uJ VENDIAN C. McA. POWELL, G. NEEF, D. CRANE, P. A. JELL ANDI. G. PERCIVAL 137 STAGES COMARTO MT. WRIGHT CUPALA CK KANDIE TANK KAYRUNNERA AIRSTRIP KOONENBERRY MT. ARROW- SMITH PAYNTONIAN PRE - PAYNTONIAN POST - IDAMEAN IDAMEAN s Erixanium sentum MINDYALLA UNDILLAN 4 FLORA TEMPLETONIA ORDIAN OLENELLIAN Stage ? 2 Undesignated : { Irvingella tropica Stigmatoa diloma Proceratopyge cryptica Glyptagnostus reticulatus Glyptagnostus stolidotus N4 Cylagnostus quasivespa Erediaspis eretes ‘A Goniagnostus nathorsti Ptychagnostus punctuosus Sf Evagnostus opimus Ptychagnostus atavus N Ptychagnostus gibbus N k <"\y Zo-nowm wm 4n-—Frvuc/\\> <= “ihe o> —> < < oo eS < a: ee Known Turbidite dips eee precisely Shale +siltstone E==] Limestone | inferred Sandstone Volcanics Angular uncon- 3] Conglomerate formity with Metamorphic maximum Breccia complex time range Fig. 9. Cambrian biostratigraphy of northwestern New South Wales, from Henderson (1976), Shergold (1971), and Opik (1968). Age of Cambrian strata from Opik (1968), Shergold (1971), Warris (1967), Wopfner (1966) and Jell (unpub. data). Circled numbers refer to the following formations. 1, Mount Wright Volcanics; 2, Cymbric Vale Formation; 3, Coonigan Formation; 4, Nootumbulla Sandstone; 5, Wonaminta Complex; 6, Copper Mine Range Beds; 7, Cupala Creek Formation; 8, Kandie Tank Limestone. Middle Cambrian is used in the sense of Opik (1968) , and Late Cambrian in the sense of Daily and Jago (1975). Proc. Linn. Soc. N.S.W., 106 (2), (1981) 1982 138 SIGNIFICANCE OF LATE CAMBRIAN FOSSILS fossiliferous sediments near Mt Arrowsmith to the underlying ?Early Cambrian volcanics is unknown (Wopfner, 1966; Warris, 1967). The Yandaminta Quartzite (Warris, 1967), which underlies the fossiliferous Early Ordovician Tabita Formation (Shergold, 1971), is separated from the Early or Middle Cambrian strata by an unconformity at Mt Arrowsmith. Between Mt Wright and Mt Arrowsmith, in the Nuntherungie-Kayrunnera area, there are three isolated areas of Cambrian strata (including the Cupala Creek Formation described herein). About 13 km southeast of Nuntherungie, the Kandie Tank Limestone (Pogson and Scheibner, 1971) contains the latest Cambrian or Early Ordovician brachiopod, Lznnarssonella sp., and distacodont conodonts (Warris, 1967). The Cupala Creek Formation lies about 9 km southeast of the exposed Kandie Tank Limestone. Trilobites collected by Neef from near Kayrunnera airstrip are of Mindyallan age (Jell, unpub. work) and are probably from the same outcrops from which Opik (1966) reported Mindyallan trilobites. Rose (1974) reported Lzngula sp. in the lower beds of a small outcrop of possible Cambrian rock at “The Bunkers’ (GR557194, 1:250,000 White Cliffs geological map); however, these lower beds are apparently continuous and conformable with quartzose sandstone and siltstone of the Mulga Downs Group, so that their age remains uncertain. Cambrian strata are present in a narrow graben formed by the bifurcation of the Koonenberry Fault north of Wonaminta station at Koonenberry Gap. Here a ?Late Cambrian conglomerate and a ?Late Cambrian hyolithid-bearing limestone are known (Warris, 1967). The flyschoid Copper Mine Range Beds, which unconformably underlie Cupaia Creek Formation and Kandie Tank Limestone, are considered by Pogson and Scheibner (1971) to be either Adelaidean or Early Cambrian. Edwards (1978) suggested that the mafic rocks of the Wonaminta Block are Early Cambrian, and reasoned that the Copper Mine Range Beds are also Early Cambrian in age — i.e. deeper-water equivalent of Early Cambrian shelf deposits at Mt Wright — an interpretation consistent with that of Scheibner (1973, fig. 7). EARLY PALAEOZOIC HISTORY OF NORTHWESTERN NEW SOUTH WALES The Wonaminta Complex (Wonaminta Block, 1971 Tectonic Map of Australia) is a regionally metamorphosed succession (up to biotite grade) of sediments and some volcanics considered to be Precambrian (Rose and Brunker, 1969). The mafic rocks intruding the Wonaminta Complex have been shown on the tectonic map of New South Wales as Early Palaeozoic island-arc intrusives, and a similar interpretation was suggested by Edwards (1978) as one of the possibilities. In this interpretation (Scheibner, 1973) a volcanic arc in the west (Mt Wright volcanic arc) is fringed with shelf sediments (Gnalta Shelf) and a fore-arc terrace (White Cliffs Deeper Terrace) to the east, and here the Copper Mine Range Beds would have accumulated. An alternative hypothesis is that the mafic igneous rocks represent basaltic outpourings associated with the Vendian or earliest Cambrian rifting of the Palaeo- Pacific margin (Veevers, 1976). The alkaline nature of the basalts on the western margin of the Wonaminta Block (Edwards, 1978) is consistent with this hypothesis, in which case the Copper Mine Range Beds would be turbidites on a passive, divergent continental margin. No matter which model of the Early Cambrian palaeogeography is correct, the Copper Mine Range Beds (clasts of which occur in the basal Cupala Creek Formation) were highly folded and eroded before the Idamean. The age of this folding cannot be constrained more tightly than latest Precambrian to earliest Late Proc. Linn. Soc. N.S.W., 106 (2), (1981) 1982 C. McA. POWELL, G. NEEF, D. CRANE, P. A. JELL ANDI. G. PERCIVAL 139 Cambrian from the Cupala Creek area, because of the uncertain age of the Copper Mine Range Beds. Other stratigraphic successions in northwestern New South Wales (Fig. 9) show a paucity of Mindyallan and Idamean deposits. Admittedly, the lacunae in each of the stratigraphic successions are of differing magnitudes, but if, for argument’s sake, we assume that the uplift associated was broadly synchronous for this region (50 km X 200 km), the uplift occurred either in the Undillan to Mindyallan stages (the Mootwingee Movement of Webby, 1978) or in the latest Precambrian. The importance of the Cupala Creek Formation in this interpretation is that it marks the upper limit of the age of the uplift. Scheibner (1976, p.141; 1978) concluded that deformation in northwestern New South Wales (Delamerian Orogeny) started in late Early Cambrian time. Further afield, in southeastern South Australia, deformation and uplift began in the Late Cambrian and continued with increasing intensity into the Early Ordovician (Thomson, 1969). In Tasmania, uplift with deformation (Tyennan Orogeny) began in the Middle Cambrian and continued into the Early Ordovician (Campana zn Banks, 1962; Williams, 1978). The possible late Middle to earliest Late Cambrian deformation in northwestern New South Wales would correlate with the early movements in the Delainerian Fold Belt of Rutland (1976). In the Mt Wright area, the contact between the latest Early to earliest Middle Cambrian Coonigan Formation and the conglomeratic Late Cambrian Nootumbulla Sandstone, is paraconformable north of the Lawrence Fault (Fig. 8). South of the fault, however, the conglomeratic base of the Nootumbulla Sandstone rests disconformably on the Early Cambrian Cymbric Vale Formation, the Coonigan Formation having been removed by mid-Cambrian erosion (Leu, 1980). Abundant grey-green sandstone clasts (possibly derived from the Early Cambrian Cymbric Vale Formation of the Mt Wright area) in the, basal conglomeratic unit of Cupala Creek Formation, suggest that the Cymbric Vale Formation may have been deeply eroded in the Idamean. Clasts of Cymbric Vale-type sediment, and limestone clasts bearing brachiopods identical to those of the overlying Coonigan Formation, also occur at Koonenberry Gap (Warris, 1967). The Kandie Tank Limestone appears to be Early Ordovician because it contains distacodont conodonts. However, the fossil evidence is not strong, and in view of the terrigenous nature of Early Ordovician strata at Mt Arrowsmith and Mt Wright in contrast with the limestone facies inferred for the Late Cambrian in the Comarto Hills and Koonenberry areas, Kandie Tank Limestone may be Late Cambrian. The timing of the southeast-trending folding of the Cupala Creek Formation cannot be constrained from the Copper Mine Range area any more accurately than post-Idamean and pre-Mulga Downs Group (assumed to be Middle or Late Devonian). Webby (1978) suggested that, during the latest Cambrian and Early Ordovician, clastics were deposited from northward- and eastward-flowing rivers in a delta complex. The Cupala Creek Formation represents an early pulse of these clastics, derived from the Delamerian Mountains rising to the south and west. According to Webby (1978) these clastics were probably deformed in the Middle Ordovician Dulingari Movement. The last folding occurred after the Mulga Downs Group was deposited. These open folds, commonly with nearly vertical to slightly overturned dips on the western limb of the syncline adjacent to the Koonenberry Fault, and gentle (15° to 50°) dips on the eastern limb, probably occurred during the Carboniferous when the rest of the Lachlan Fold Belt was similarly folded (Powell et al., 1980). Tilting, rotation and minor folding associated with the steep faults trending 100° occurred in the Cupala Creek Formation at this time. Proc. Linn. Soc. N.S.W., 106 (2), (1981) 1982 140 SIGNIFICANCE OF LATE CAMBRIAN FOSSILS SYSTEMATIC PALAEONTOLOGY The trilobites and mollusc were collected from locality 1, and brachiopods from localities 2-5 (Fig. 1). Trilobite debris is also present at locality 3. All described material is housed in the Palaeontological Collection of the National Museum of Victoria (numbers prefixed P). ARTHROPODA (P.A.J.) Class TRILOBITA Walsh, 1771 Order MIOMERA Jaekel, 1909 Superfamily AGNOSTOIDEA McCoy, 1849 Family DIPLAGNOSTIDAE Whitehouse, 1936 Subfamily PSEUDAGNOSTINAE Whitehouse, 1936 Genus PSEUDAGNOSTUS Jaekel, 1909 Pseudagnostus sp. aff. cdalis Opik, 1967 Fig. 10, 1,2 Material: One cranidium P61493 and one pygidium P61494. Discussion: This species is related to P. zdalzs in the squarish glabellar anterior and weak median preglabellar furrow, in the shape of the basal glabellar lobes and in the wide border furrow. In the pygidium the shape of the posterior of the deuterolobe is particularly reminiscent of P. zdalzs and the extremely wide border furrow and narrow pleural area are also similarities. The preservation of this material is too inferior for positive specific identification but enough morphology is revealed to indicate a resemblance to P. zdalzs. Order POLYMERA Jaekel, 1909 Superfamily PTYCHOPARIOIDEA Matthew, 1887 Family EULOMATIDAE Kobayashi, 1955 Genus STIGMATOA Opik, 1963 Stigmatoa tyson? Opik, 1963 Fig. 10, 3-6 1963 St?gmatoa tysoni Opik, p.92, pl. 4, fig. 3 1976 Stigmatoa tysoni Henderson, p.354, pl. 51, figs 8, 9 Materzal: One well-preserved (P61496) and three damaged (P61495, 61497 and 61498) cranidia. Dimensions: Cranidial lengths range from 4 to 7 mm. Diagnosis: As given by Henderson (1976) and Opik (1963). Discussion: There can be no doubt that this material belongs to S. tysonz; glabellar furrows, occipital ring and palpebral areas are all distinctive. Fig. 10. 1, 2, Pseudagnostus sp. aff. cdalis Opik, 1967. 1, internal mould of cephalon, P61493, x7.2; 2, damaged internal mould of pygidium, P61494, x7.2. 3-6. Stigmatoa tysoni Opik, 1963. 3, badly damaged and deformed cranidium P64195, x7.2; 4a, b, dorsal and right anterolateral oblique views of internal cranidial mould, P61496, x5.4.; 5, internal mould of damaged cranidium, P61497, x3.6; 6, internal mould of damaged cranidium, P61498, x5.4 7, 8. ? Aphelaspis sp. aff. A. australis Henderson, 1976. Internal moulds of damaged cranidia, P61499 and 61500, x6.3 and x7.2 respectively. : 9-12. Prismenaspis? sp. nov. 9, latex cast of damaged and tectonically shortened cranidium P61501, x2.7; 10-12, internal moulds of damaged and distorted cranidia, P61502, 61503 and 61504, x2.7. 13, 14. Genus and species indeterminate. Internal moulds of damaged distorted cranidia P61505 and 61506, x3.6. 15-17. Proplina sp. Right and left anterolateral and dorsal views of internal mould, P61507, x1.8. Proc. Linn. Soc. N.S.W., 106 (2), (1981) 1982 14] C. McA. POWELL, G. NEEF, D. CRANE, P. A. JELL ANDI. G. PERCIVAL Proc. Linn. Soc. N.S.W., 106 (2), (1981) 1982 142 SIGNIFICANCE OF LATE CAMBRIAN FOSSILS Family PTEROCEPHALIIDAE Kobayashi, 1935 Subfamily APHELASPIDINAE Palmer, 1960 Genus 4 PHELASPIS Resser, 1935 Type species: A. walcotti Resser 1938, p.59, pl. 13, fig. 14. Diagnosis: As given by Palmer (1965, p.58) with the addition made by Henderson (1976, p.345 in remarks on Eugonocare) that the interocular width approximately equals cranidial length. Discusston: In Australia this genus is represented by A. australis Henderson (1976, p.342, pl. 49, figs. 5-7) and by a species from limestone in eastern Victoria (Thomas and Singleton, 1956). A case could be made to exclude australzs from A phelaspis on the basis of its very much shorter frontal area, its well impressed glabellar furrows, its longer palpebral lobes, and its more rounded glabellar anterior. I consider that it should remain in A phelaspis but that it is a distant relative of the North American species described to date including A. brachyphasts Palmer, 1962 which Henderson considered the ‘nearest match’. Future detailed work on Australian faunas of this age may well show that australis belongs to a lineage quite separate from the North American Aphelaspis. For the present it is placed in Aphelaspis as is the material compared with it herein. 2A phelaspis sp. aff. A. australis Henderson, 1976 Fig. 10, 7,8 Material: Two poorly preserved, damaged internal moulds of cranidia, P61499 and 61500. Description: Cranidium approximately as long as interocular width which is slightly compressed. Glabella slightly tapering, anteriorly truncated, with almost straight sides. Axial furrow well impressed but slightly shallower anteriorly. Frontal area short (approximately 40% of glabellar length). Border furrow shallow. Border short and convex, of uniform length except where the facial sutures cut across it. Interocular cheeks narrow. Palpebral lobes elongate, slightly raised and opposite middle of glabella. Discusston: These specimens resemble A phelaspis australis in their short frontal area measuring 40% of glabellar length and in their elongate palpebral lobes. They also resemble 24 phelaspis sp. B. of Opik, 1963 in the type of border, glabellar shape, short frontal area and narrow interocular cheeks. However, their poor preservation and ignorance of the rest of the exoskeleton make closer identification impossible. Family PTEROCEPHALIIDAE Kobayashi, 1935 Genus NOTOAPHELASPIS gen. nov. Etymology: Notios Gr. southern, with North American genus name A phelaspis (from apheles Gr. smooth). Fig. 11. 1-10. Notoaphelaspis orthocephalis gen. et. sp. nov. la, latex of cranidium, Ib internal mould of same cranidium P61508, x3.2 (approx.); 2a, right anterolateral oblique view and, 2b, dorsal view of internal mould of holotype cranidium P61509, x2.2; 3, internal mould of cranidium with fixed cheeks only slightly displaced and three thoracic segments P61510, x1.4; 4, internal moulds of damaged thorax and pygidium, P61511, x1.8; 5, 6, internal moulds of damaged cranidia P61512 and 61513, x1.8 and x1.4 respectively; 7a internal mould and 7b, latex cast from external mould of damaged cranidium and partial thorax P61514, x1.8; 8, 9, internal moulds of damaged cranidia P61515, 61516, x2.7; 10, internal mould of free cheek broken down genal spine to reveal the external mould of the ventral surface P61517, x3.2. 11. Notoaphelaspis sp. cf. N. orthocephalis gen. et sp. nov. internal mould of incomplete cranidium P61518, x2.7. Proc. LINN. Soc. N.S.W.,106 (2), (1981) 1982 C. McA. POWELL, G. NEEF, D. CRANE, P. A. JELL ANDI. G. PERCIVAL 143 tee Sight ie < é bee Proc. Linn. Soc. N.S.W., 106 (2), (1981) 1982 144 SIGNIFICANCE OF LATE CAMBRIAN FOSSILS Type species: N. orthocephalis gen. et sp. nov. Diagnosis: Aphelaspid with rectangular glabella, wide interocular cheeks, transverse eye ridges, anteriorly placed palpebral lobes, long downsloping preglabellar field, well impressed anterior border furrow, medially elongate anterior border, non-spinose pleural tips, and transverse, non-spinose pygidium. Discussion: Although this species could, with some difficulty, be placed in A uae the combination outlined above is quite unlike any known species of that genus and the eye lines, glabellar shape and deep border furrow are particularly distinctive. It has no close relatives in North America. Eugonocare Whitehouse, 1939 is distinguished by its broader glabella with sloping lateral furrows, anterior border, more sloping eye ridges and longer pygidium. Its glabella and the furrows thereon are extremely similar to those of Erzxanium which are themselves very distinctive among trilobites. The major differences between N. orthocephalis and E. sentum are the glabella shape, cephalic border, pleural tips and the pygidium. In view of the marked changes in pleural tip and pygidial morphology along an aphelaspid lineage (Palmer, 1965) the possibility that N. orthocephalis is an earlier member of the Erixanzum lineage (and therefore a member of the Erixaniidae) should not be overlooked. Such placement would require considerable expansion of the Family concept given by Opik (1963). Notoaphelaspis orthocephalis sp. nov. Fig. 11, 1-10 Etymology: Orthos Gr. right (angled) ; kephale Gr. head; for the angular glabella. Holotype: P61509. Other material: Several cranidia, two thoraxes, one free cheek and two pygidia including the paratypes P61508 and P61510-61517. Diagnosis: Up to 8 cm in cephalic length. Cranidial length and interocular width approximately equal. Glabella rectangular, two-thirds cephalic length, with angular anterolateral corners and truncated anterior, with occipital furrow poorly impressed axially but not reaching axial furrow, with three pairs of very weakly impressed glabellar furrows as circular depressions on either side of axis, well separated from axial furrow. Preglabellar field relatively long, down sloping and slightly swollen in front of glabella. Anterior border furrow well impressed. Anterior border convex and expanded medially. Each interocular cheek as wide as glabella at midlength of relatively long arcuate slightly upturned palpebral lobes. Eye ridges transverse; very slightly posteriorly directed laterally in a few specimens. Facial sutures diverging anteriorly from palpebral lobe at about 45° posteriorly running laterally at a low angle to the transverse line so that the posterior part of the fixed cheek is short and wide. Posterior border furrow distinct, of almost constant length throughout, defining a short convex border also of constant length. Free cheek with convex genal field, well impressed border furrow, and long slender slightly incurved genal spine that continues the curve of the lateral margin. Thorax of at least 9 segments (no complete thorax available). Axis highly arched and relatively narrow. Axial furrow simply a change of slope. Articulating line about mid-width of each pleural area. Articulating facet very wide, steep and flat. Pleural furrows prominent occupying most of the pleural area. Anterior and posterior pleural margins parallel. Pygidium transverse, twice as wide as long, moderately convex with axis standing above pleural areas. Axis of three rings and a posteriorly rounded terminus that interrupts the border furrow. Pleural areas convex, crossed by three pleural and one interpleural (including anterior border furrow) furrows. Border furrow weakly impressed but not impressed at all behind the Proc. Linn. Soc. N.S.W., 106 (2), (1981) 1982 C. McA. POWELL, G. NEEF, D. CRANE, P. A. JELL ANDI. G. PERCIVAL 145 axis. Border widening anteriorly, not present behind the axis. The margin forms a smooth arc, except behind the axis where it is excavated to the extent that it comes very Close to the axis. Discussion: The material figured here is in a coarse matrix, is tectonically distorted to some degree in each specimen and is mainly of damaged internal moulds but the morphology available is so distinctive as outlined under discussion of the genus as to demand a separate taxonomic name. Comparison with any existing species is superfluous. The sample is too small to determine any intraspecific variation and no morphogenetic variation is visible over the size range available. As discussed below the small cranidium (Fig. 11, 7) could possibly be a juvenile of this species although this is a remote chance. Notoaphelaspis sp. cf. N. orthocephalis sp. nov. Higa tet Material: One fragmentary cephalon P61518. Discusston: This specimen differs from N. orthocephalis in that its anterior border does not taper laterally, its preglabellar field is shorter and its interocular cheeks are relatively narrower. Family UNCERTAIN Genus PRISMENASPIS Henderson, 1976 Prismenaspis? sp. nov. Fig. 10, 9-12 Materval: Six cranidia in various degrees.of distortion due to deformation, including P61501-61504. Description: Moderately large cranidium (up to 1.5 cm long) approximately as wide as long with broad squat anteriorly tapering glabella just over half cranidial length. Glabellar furrows not well impressed but 1P connected adaxially to 2P in at least one specimen (Fig. 10, 11). Occipital ring tapering laterally and with an almost imperceptible median node. Axial furrow well impressed with a marked pair of fossulae in front of the glabella adaxial to the rounded anterolateral corners. Preglabellar field slightly inflated medially, short and not distinguished from an almost indiscernible, long, shallow, border furrow which in turn is not separated from a long, flat, medially inflated border that tapers laterally and bears a low but distinct transverse ridge that may or may not correspond to the edge of the doublure. Eye ridges well defined and running posterolaterally to short wide, kidney-shaped palpebral lobes. Facial sutures diverging slightly forward and sloping posteriorly behind the palpebral lobe. Posterior portion of fixed cheek triangular, crossed by long border furrow that runs forward laterally to leave a lengthening and flattening posterior border. Discussion: The state of preservation and small number of specimens make identification of this with any existing species impossible but it does not appear to be similar to any existing taxon. Even at the generic level the assignment to Przsmenaspis is very tentative and requires expansion of that genus to include apparently unornamented species. This is in conflict with Henderson’s concept of the genus as he stressed the pustulose ornament on both internal moulds and external surfaces. However, related genera such as Dunderburgza encompass smooth and ornamented species. Moreover, Henderson described four pairs of lateral glabellar furrows in Prismenaspts but material of the New South Wales species is too poorly preserved to Proc. Linn. Soc. N.S.W., 106 (2), (1981) 1982 146 SIGNIFICANCE OF LATE CAMBRIAN FOSSILS see any but the 1P glabellar furrows. Glabellar shape, well-impressed axial furrow with prominent fossulae, narrow interocular cheeks, wide posterior portion of the fixed cheeks, position and length of palpebral lobes, and most importantly, the flat frontal area relate this species to Prismenaspis. Prismenaspis propinquum is the most closely related species but differs in ornament, glabellar shape and lack of transverse ridge on anterior border. As with Prismenaspis there are several younger Siberian genera described by Rozova (1968, 1977) that appear similar, including Ketyna, Monosulcatina, Mansiella, and Maduzya but the comparisons are obscured by the states of preservation and incomplete information. Genus and species indeterminate Fig. 10, 13, 14 Materzal: Two damaged cranidia P61505 and 61506. Description: Small cranidium longer than wide with subrectangular glabella having only very slightly tapering sides, angular anterolateral corners, and no furrows impressed. Axial furrow well impressed, without fossulae. Preglabellar furrow transverse. Preglabellar field short, shorter than border. Border furrow shallow throughout and indistinct medially where the border is expanded into a poorly- defined plectrum. Border relatively long, tapering gradually laterally. Interocular cheeks narrow. Palpebral lobes elongate, weakly curved and defined by shallow palpebral furrows. Posterior part of fixed cheek rather long and narrow with a prominent laterally lengthening border furrow running across it. Discussion: These two specimens are distinctive in their clearly truncated glabellar anterior, and broadly convex anterior border. They may belong to a genus that has not yet been named but a new name has not been assigned because of the small number of poorly-preserved internal moulds available and because of the lack of knowledge on the remainder of the exoskeleton. Perhaps the most distinctive aspect of this material is the long, convex anterior border, that is unique among this type of trilobite of this age. As with Prismenaspis the closest morphology is to be found among younger trilobites from the northwestern Siberian Platform. Monosulcatina laeve Rozova, 1963, particularly the cranidium, figured by Rozova in 1968 (pl. 9, figs 11, 12) is very close in morphology to the N.S.W. specimen. MOLLUSCA (PeASTs) Class MONOPLACOPHORA Wenz in Knight, 1952 Order TRYBLIDIIDA Lemche, 1957 Family TRYBLIDIIDAE Pilsbry in Zittel-Eastman, 1899 Genus PROPLINA Kobayashi, 1933 Proplina sp. Fig. 10, 15-17 Materzal: One shell P61507. Description: A large shell 14 mm high, 21 mm wide and 30 mm long with the shape of an inverted asymmetrical cone. Apex overhangs the margin of the oval aperture. Internal and external surfaces of shell exhibit fine closely-spaced growth lines parallel to the apertural margin. Dorsal surface of internal mould with a variety of curving impressions but none of these are symmetrical nor could they be interpreted as muscle scars. Discussion: Ignorance of the muscle pattern and of the apex prevent specific assignment but the exterior surface, position of the apex and apertural shape make generic placement clear. The apex is presumably pointed but the infilling has broken Proc. Linn. Soc. N.S.W., 106 (2), (1981) 1982 C. McA. POWELL, G. NEEF, D. CRANE, P. A. JELL ANDI. G. PERCIVAL 147 on the surface shown in Fig. 10, 16. As there is only one specimen available and the infilling does not easily separate from the external mould it has not been fully prepared out. BRACHIOPODA (iGaPs) Class ARTICULATA Huxley, 1869 Order ORTHIDA Schuchert & Cooper, 1932 Suborder CLITAMBONITIDINA Opik, 1934 Superfamily BILLINGSELLACEA Schuchert, 1893 Family BILLINGSELLIDAE Schuchert, 1893 Genus BILLINGSELLA Hall & Clarke, 1892 Billingsella sp. Fig. 12, 1-5 Material: Eleven specimens (P61939-P61949) from locality 2; five specimens (P61950-P61954) from locality 4; three specimens (P61955-P61958) from locality 5; two specimens (P61959 and P61960) from locality 3. Description: Exterior of valves. Outline subquadrate; maximum width at about midlength of pedicle valve, and occurring in posterior half of brachial valve; hingeline equal to, or slightly less than, maximum width; cardinal extremities obtuse to right- angled. Profile biconvex, with deeper brachial valve (depth about one-third valve length) bearing shallow posteromedian sulcus. Ornament poorly preserved on available specimens, but apparently finely multicostellate with prominent widely- spaced concentric growth lamellae. Ventral interarea long, equal to between 0.2 and Fig. 12. 1-5. Billingsella sp. All x2.9. 1-2. Pedicle valve internal mould and corresponding latex cast, P61939. 3. Pedicle valve internal mould, P61940. 4. Brachial valve internal mould, P61941. 5. Brachial valve external (latex cast), P61942. 6. obolid indet., P61938, x3.5. Proc. Linn. Soc. N.S.W., 106 (2), (1981) 1982 148 SIGNIFICANCE OF LATE CAMBRIAN FOSSILS 0.3 valve length, flat to weakly curved, orthocline to apsacline; delthyrium covered for about one-third its length by strongly arched pseudodeltidium. Dorsal interarea much lower, less than one-third length of ventral interarea; planar, anacline. Pedicle valve intertor. Teeth stout, extending from hingeline to floor of valve; dental plates lacking. Muscle field large, extending to between 0.4 and 0.5 valve length, and occupying approximately one-quarter valve width; elliptical diductors divided posteriorly by short, thin median ridge extending forward from apical callus; central adductor impression expands rapidly in front of median ridge. Mantle canal impressions prominent, of saccate pattern with thick vascula media weakly divergent medially, strongly arcuate anterolaterally. Broad shallow groove-like depression extends from hingeline adjacent to valve margin. Brachial valve interior. Cardinalia simple, comprising fine ridge-like cardinal processes, and short widely divergent rod-like brachiophores, flanking cup-shaped sockets. Low, broad, median ridge extends from notothyrial platform to about one- third valve length, bisecting faintly quadripartite muscle field. Mantle canal impressions not discernible. Measurements (in mm) Figured specemens. P61939 (pedicle valve): length 13.1, width 12.4, length of muscle field 6.5; P61940 (pedicle valve): length 9.5, width 10.8, length of muscle field 4.1; P61941 (brachial valve): length 8.8, width 11.7; P61942 (brachial valve) : length 7.2, width 9.2. Pedicle valves, 8 specomens. Range of lengths: 9.5-13.1, range of widths: 10.8-14.0, range of length/width ratios: 0.88-1.08, average length/width ratio: 0.99. Brachial valves, 9 specimens. Range of lengths: 6.3-10.4, range of widths: 7.7-14.0, range of length/width ratios: 0.73-0.82, average length/width ratio: 0.78. Discussion: Billingsella is restricted to strata of Middle Cambrian to Early Ordovician age, being particularly abundant in the Late Cambrian. The species found in the Cupala Creek Formation is most similar to (and may prove to be conspecific with) Bullingsella sp. from the Middle to Late Cambrian Mariner Formation (Cooper et al., 1976) of Northern Victoria Land, Antarctica, described and illustrated by MacKinnon (zm Shergold et al., 1976). Relatively minor differences between these forms include a proportionately shorter pseudodeltidium and slightly less pronounced dorsal median sulcus on New South Wales specimens. In other details such as outline, profile, dimensions and internal structures, species from these two areas are virtually indistinguishable. In common with the Antarctic species, the New South Wales form resembles several North American species, such as B. perfecta Ulrich and Cooper and B. texana Bell, typical of the Late Cambrian. Trilobites associated with the Antarctic Billingsella indicate its age as late Idamean, Erzxaniwm sentum Zone (Shergold et al., WOTOE Class INARTICULATA Huxley, 1869 Order LINGULIDA Waagen, 1885 Superfamily LINGULACEA Menke, 1828 Family OBOLIDAE King, 1846 Obolid indet. Fig. 12,6 Discussion: A single poorly-preserved valve (P61938) was found at locality 3. The available material does not permit precise identification or age determination. Measurements: Length 9.2 mm, width 7.0mm. Proc. Linn. Soc. N.S.W., 106 (2), (1981) 1982 C. McA. POWELL, G. NEEF, D. CRANE, P. A. JELL ANDI. G. PERCIVAL 149 ACKNOWLEDGEMENTS Powell and Crane were supported by the Australian Research Grants Com- mittee. David Larsen and Jeff Vaughan (W. S. & L. B. Robinson University College) helped in the field and Elaine Prendergast (Macquarie University) point-counted the sandstone samples. Mr Bill Gall (Nuntherungie station) is thanked for hospitality and he kindly piloted the light aircraft from which Fig. 2 was taken. Ian Percival acknowledges financial support from a Rotary Foundation Graduate Fellowship, and use of facilities at the Museum of Invertebrate Paleontology, Department of Geology, University of Kansas. Erwin Scheibner, John Shergold, John Talent and John Veevers. offered helpful comments on the manuscript. References BANKS, M., 1962. — Ordovician System. J. geol. Soc. Aust. 9:147-176. BRUNKER, R. L., OFFENBERG, A., and RosE, G., 1967. — 1:500,000 Koonenberry geological sheet. Geol. Surv. N.S.W. Cooper, R. A., JAco, J. B., MACKINNON, D. I., Stes, J. E., and BRappock, P. E., 1976. —Cambrian fossils from the Bowers Group, Northern Victoria Land, Antarctica (preliminary note). N.Z. J. Geol. Geophys. 19: 283-288. Dairy, B., and Jaco, J. B., 1975. — The trilobite Lejopyge Hawle and Corda and the Middle-Upper Cambrian boundary. Palaeontology 18:527-550. Epwarpbs, A. C., 1978. — Tectonic implications of the immobile trace-element geochemistry of mafic rocks bounding the Wonaminta Block. J. geol. Soc. Aust. 25:459-465. FoLk, R. L., 1974. — Petrology of sedimentary rocks. Austin, Texas: Hemphill’s. Evans, P. R., 1977. — Petroleum geology of western New South Wales. Aust. Petrol. Explor. Assn. J. 17: 42-49). FREEMAN, R. N., 1965. — Geological appraisal of productive potential of P.E.L. 114, New South Wales. Rep. Planet Explor. Co. Pty. Ltd. (Geol. Surv. N.S.W. Petrol. File, Geol. 1965/1) (unpubl.) HENDERSON, R. A., 1976. — Upper Cambrian (Idamean) trilobites from western Queensland, Australia. Palaeontology 19: 325-364. Kruse, P. D., 1978. — New Archaeocyatha from the Early Cambrian of the Mt. Wright area, New South Wales. Alcheringa 2: 27-47. Leu, M. R., 1980. — Silicified early Middle Cambrian trilobites from the Gnalta District, N.S.W., and their stratigraphic context. North Ryde, N.S.W.: Macquarie University, B.A.(Hons) thesis, unpubl. eek Moore, P. S., 1979. — Stratigraphy and depositional environments of the Billy Creek Formation (Cambrian), central and northern Flinders Ranges, South Australia. Trans. R. Soc. South Aust. 103: 197-211. Neer, G., Epwarps, A. C., HoLZBERGER, I., Harty, J., KELLY, R., and VAUGHAN, J. — The geology of the Mt Daubeny/Wertago area (in prep.). Opik, A. A., 1963. — Early Upper Cambrian fossils from Queensland. Bull. Bur. Miner. Resour. Geol. Geophys. Aust. 64: 1-133. ———, 1966. — The early Upper Cambrian crisis and its correlation. J. Proc. R. Soc. N.S.W. 100: 9-14. ———, 1967. — The Mindyallan fauna of north-western Queensland. Bull. Bur. Miner. Resour. Geol. Geophys. Aust. 74: 1-404 and 1-67. ———, 1968. — The Ordian Stage of the Cambrian and the Australian Metadoxididae. Bull. Bur. Miner. Resour. Geol. Geophys. Aust. 92: 133-170. ———, 1975. — Cymbric Vale fauna of New South Wales and Early Cambrian biostratigraphy. Bull. Bur. Miner. Resour. Geol. Geophys. Aust. 159: 1-78. ———, 1979. — Middle Cambrian agnostids: Systematics and biostratigraphy. Bull. Bur. Miner. Resour. Geol. Geophys. Aust. 172: 1-188. PackHaM, G. H., (ed.), 1969. — The geology of New South Wales. J. geol. Soc. Aust. 16: 1-654. Parmer, A. R., 1965. — Trilobites of the Pterocephaliid Biomere in the Great Basin, United States. Prof. Pap. U.S. geol. Surv. 493: 1-105. Pogson, D. J., and ScHEIBNER, E., 1971. — Pre-Upper Cambrian sediments east of the Copper Mine Range, New South Wales. Q. Notes Geol. Surv. N.S.W. 4: 3-8. Proc. Linn. Soc. N.S.W., 106 (2), (1981) 1982 150 SIGNIFICANCE OF LATE CAMBRIAN FOSSILS PowELL, C. McA., GILFILLAN, M. A., and Henry, N. M., 1978. — Early east-southeast trending folds in the Sofala Volcanics, N.S.W. J. Proc. R. Soc. N.S.W. 111: 121-128. ———, Fercusson, C. L., and WI..iams, A. J., 1980. — Structural relationships across the Lambian Unconformity in the Hervey Range-Parkes area, N.S.W. Proc. Linn. Soc. N.S.W. 104: 195-204. Rircuie, A., 1973. — Wuttagoonaspis gen. nov., an unusual arthrodire from the Devonian of Western New South Wales, Australia. Palaeontographica Abt. A. 143: 58-72. Ross, G., 1974. — Explanatory notes on the White Cliffs 1: 250,000 Geological Sheet. Geol. Surv. N.S. W. Sydney, 48 p. ———, LoupEN, A. G., and O’CoNNELL, P., 1964. — 1:250,000 Geological Sheet SH 54-12, White Cliffs. Geol. Surv. N.S. W. ———, and BRUNKER, R. L., 1969. — The Upper Proterozoic and Phanerozoic geology of north-western New South Wales. Proc. Australas. Inst. Min. Metall. 229: 105-120. Rozova, A. V., 1968. — Biostratigraphiya i trilobity Verkhnego Kembriya 1 Nizhnego Ordovika severo- zapada Sibirskoi platformy. Trudy Inst. Geol. Geofiz. stb. Otd., Moscow 36: 1-196. ———, 1977. — Nekotorye verkhnekembriyskie i nizhneordovikskie trilobity bassinov rek Rubnoy, Khantayki, Kureyki i Letney. Jn Biostratigrafiya i fauna verkhnego kembriya 1 pogranichnykh s nim sloev. Trudy Inst. Geol. Geofiz. sib. Otd., Moscow 313: 54-84. RUTLAND, R. W. R., 1976. — Orogenic evolution of Australia. Earth. Sct. Rev. 12: 161-196. SCHEIBNER, E., 1973. — A plate tectonic model of the Palaeozoic tectonic history of New South Wales. /. geol. Soc. Aust. 20: 405-426. ———, 1976. — Explanatory notes on the tectonic map of New South Wales. Geol. Surv. N.S.W., Sydney. ———, 1978. — Tasman Fold Belt system in New South Wales — general description. Tectonophysics 48: 297-316. SHERGOLD, J. H., 1971. — Resume of data on the base of the Ordovician in northern and central Austria [sec=Australia]. Mém. Bur. Rech. géol. min. 73: 391-402. ———, Cooper, R. A., MacKinnon, D. I., and YOcHELSON, E. L., 1976. — Late Cambrian Brachiopoda, Mollusca, and Trilobita from Northern Victoria Land, Antarctica. Palaeontology 19: 247-291. Tuomas, D. E., and SINGLETON, O. P., 1956. — The Cambrian stratigraphy of Victoria. 20th Int. Geol. Congr. Mexico, 1956. El Sistema Cambrico 2: 149-163. THomson, B. P., 1969. — The Kanmantoo Group. In PaRkIN, L. W., (ed.), Handbook of South Australian Geology, Geol. Surv. South Aust., 102-108. VEEVERS, J. J., 1976. — Early Phanerozoic events on and alongside the Australasian-Antarctic platform. /. geol. Soc. Aust. 23: 183-206. Warkis, B. J., 1967. — The Palaeozoic stratigraphy and palaeontology of north-western New South Wales. Sydney: University of Sydney, Ph.D. thesis, unpubl. Wessy, B. D., 1972. — Devonian geological history of the Lachlan Geosyncline. J. geol. Soc. Aust. 19: 99- 123. ———, 1978. — History of the Ordovician continental platform and shelf margin of Australia. J. geol. Soc. Aust. 25: 41-63. WILLIAMS, E., 1978. — Tasman Fold Belt system in Tasmania. Tectonophysics 48: 159-205. WILsoN, R. B. 1967. — Geological appraisal of the Mootwingee area, New South Wales. Aust. Petrol. Explor. Assn. J. 7: 103-114. WopPrNeER, H., 1966. — Cambro-Ordovician sediments from northeastern margin of the Frome Embayment (Mt Arrowsmith, N.S.W.). J. Proc. R. Soc. N.S.W. 100: 163-177. Proc. Linn. Soc. N.S.W., 106 (2), (1981) 1982 Origins and Relationships among the Animal Phyla D. T. ANDERSON ANDERSON, D. T. Origins and relationships among the animal phyla. Proc. Linn. Soc. N.S.W. 106 (2), (1981) 1982: 151-166. Conceptual and factual advances in the study of animal phylogeny have emerged from recent work in palaeontology, functional morphology and comparative embryology. Many earlier proposals on the origin and relationships of the animal phyla can now be seen to be erroneous. The fossil record of the Precambrian and early Cambrian has not revealed links between phyla, but it places the time of origin and radiation of the metazoan phyla at earlier than 700 million years ago. Functional morphological studies have raised many possibilities of the convergent or parallel evolution of phyla, especially among those of simpler body construction. In particular, these studies have called into question the validity of hypothetical ancestors and of a monophyletic origin of the Metazoa. Comparative embryology remains a major source of positive information about the relationships between the animal phyla. Two major assemblages, the spiral cleavage assemblage and the deuterostome assemblage, can be identified by this approach. More information is needed on a number of phyla. A synthesis of all three lines of investigation emphasizes the likelihood that metazoans arose from protozoans on several occasions. D. T. Anderson, School of Biological Sciences, The University of Sydney, Australia, 2006; manuscript recetved 15 October 1981, accepted for publication 18 November 1981. INTRODUCTION A student once wrote in an essay for me, ‘Phylogenetic relationships is a very doubtful area of zoology’. This a sentence of delightful ambiguity. I think all zoologists would agree with it, but for different reasons. To some, the word doubtful would mean dubious, questionable, verging on the disreputable. To others, quite a strong group, the doubt would be as to whether phylogenetic relations is a part of scientific zoology at all. A third group, however, would take it that there are still matters of doubt and unresolved questions concerning animal phylogeny, and would regard this as an important challenge for zoologists to struggle with. I subscribe to the last of these views, but I am well aware that there are plenty of dubious phylogenies as well as some spectacularly unscientific ones in the recent literature. I shall mention some of these as we proceed. I am also aware that a lack of suitable factual knowledge leaves many questions still unanswered concerning phylogenetic relationships, but I think we have now reached the stage where some firmly-based statements can be made. Many attempts have been made over the years to evaluate the origins of the metazoan phyla and their phylogenetic relationships. Until quite recently there was a strong tendency among animal phylogeneticists to vie with one another in the erection of monophyletic trees for the Metazoa, branching in various ways. Figs 1 and 2 show two examples of this. Preconception was the basic philosophical stance in these attempts, with the known facts being used as hooks from which to suspend the imagined tree. This approach has not lost its vogue. Papers are still being published which contain trees of this type, replete with nonfunctional hypothetical ancestors. A good example of this, first published in 1976, is shown in Fig. 3. Such works are of the same conceptual status as land bridges in biogeography. In the last twenty years, however, and especially in the last decade, very substantial conceptual and factual advances have been made by many workers Proc. Linn. Soc. N.S.W., 106 (2), (1981) 1982 152 ORIGINS AND RELATIONSHIPS AMONG ANIMAL PHYLA Arthr. s Annelids Platyhelminths Nemerteans Brachiopods Se eos Hemichordates lophophorates Coelomate Animals Radiate Animals Scyphozoans Hydrozoans Anthozoans Sporozoans Protozoans Sarcodinians Flagellates Fig. 1. The phylogenetic tree of Marcus (after Barnes, 1968). interested in the problem of metazoan phylogenetic relationships. It may seem perhaps surprising that this problem, which was at the forefront of zoological thinking in the latter part of the nineteenth century, should receive a revival of interest in the latter part of the twentieth century; but I think this can be explained by the importance of phylogeny to zoology as a whole. Zoology, the science of animals, is couched in terms of generalizations based properly and entirely on an acceptance of the view that evolution has occurred. The better we understand this evolution in terms of origins and lineages, the better will be our generalizations about its products. CONCEPTUAL ADVANCES What then are the conceptual advances which are assisting in this endeavour? There are three. The first is a greater emphasis on the notion that facts take precedence over preconceptions. Paraphrasing two famous pragmatists, Sidnie Manton and Henry Ford, hypothetical animals are bunk! The hypothetical urcrustacean shown in Fig. 4 and the hypothetical ancestral mollusc well known amongst zoologists (Yonge and Thompson, 1976), while of value in comprehending the phyla to which they are ascribed, have no place in thinking about the relationships between phyla. Gaps in the evidence must be accepted for what they are and not filled with inventions to satisfy the urge for a monophyletic oversimplification. Proc. Linn. Soc. N.S.W., 106 (2), (1981) 1982 D. T. ANDERSON 153 Molluscs Arthropods Chordates ee ee Hemichordates lophophorates _—_ Brachiopod seudocoelomat¢s Echinoderms Flatworms Coelenterates Acoeloid Ancestor Sporozoans Ciliophorans Sponge s \ Rhizoflagellat Sarcodinians Flagellates Fig. 2. The phylogenetic tree of Hanson (after Barnes, 1968). Secondly, deriving from advances in many aspects of zoology, it is now realized that animals exist as functionally integrated complexes of dynamically interacting components, in which there are intrinsic constraints on change as well as the possibilities of change over evolutionary time. Putting this more simply, evolution must be functional and function must be viewed holistically. Any phylogenetic proposal which ignores the requirements of functional gradualism (Anderson, 1967; Dullemeijer, 1980) must be wrong. This is not to say that any proposal based on holistic functional thinking must be right. There are other ways in which error can potentially creep in. Functional thinking may, for example, be employed spuriously to give credence to imaginary evolutionary sequences. Gutmann (1981) treats the deuterostomes in this way, expounding the evolution of all deuterostomes from a segmented chordate stem, itself evolved by an entirely hypothetical ‘functional’ route. Proc. Linn. Soc. N.S.W., 106 (2), (1981) 1982 154 ORIGINS AND RELATIONSHIPS AMONG ANIMAL PHYLA Arthropoda \ Annelida a Chordata Articulata Mollusca : d es rer Heihordela — Ae Pogonophora Spiralian Archicoélomate Cnidaria Ctenophora Porifera Blastaea Fig. 3. The phylogenetic relationships of the Metazoa according to Siewing (1976), redrawn and simplified. Nowadays, however, most zoologists and many palaeontologists have begun to include functional thinking in their phylogenetic interpretations (e.g. Bock, 1981; Szalay, 1981), though some palaeontologists still proceed in blissful unawareness of this need. Proc. Linn. Soc. N.S.W., 106 (2), (1981) 1982 D. T. ANDERSON 155 antennule antenna mandible maxillule maxilla first trunk limb Fug. 4. The hypothetical urcrustacean of Hessler and Newman (redrawn after Hessler ang Newman, 1973). The speculations of Jefferies (Jefferies and Lewis, 1978; Jefferies, 1979, and earlier) on the Calcichordata as ancestral chordates exemplify this well, as Philip (1979) has remarked. Curiously enough, this change of approach has led in turn to a third conceptual advance with a somewhat opposite effect. This is the realization that for the simpler kinds of animal organization, with few interacting components, it is very difficult to exclude the possibility of parallel or convergent evolution. Simple metazoans which may seem alike are not necessarily related. I shall return to this later. FACTUAL ADVANCES We are thinking about animal phylogeny with a new kind of honesty and understanding as a result of these conceptual changes, but this new approach derives basically from the availability of new facts about animals and their history. During the last twenty years, factual knowledge has advanced on three fronts pertinent to the problem of phylogenetic relationships. The early part of the fossil record of animals in the later Precambrian and Cambrian has become much better known (e.g. Brasier, 1979). The functional morphology of the modern phyla has become much better understood; and the embryonic development of many of the modern phyla has been more deeply analysed in functional terms. The interplay of fact and concept has now reached a stage at which the evidence of all three of these lines of investigation can be brought together in a more comprehensive way. I shall briefly review each area and then attempt to establish a synthesis from them. Proc. Linn. Soc. N.S.W., 106 (2), (1981) 1982 156 ORIGINS AND RELATIONSHIPS AMONG ANIMAL PHYLA 1 Jellyfish 2 Charniodiscus 3 Tribrachidium 4 Dickinsonia 5 Spriggina Fig. 5. Some representatives of the Ediacara fauna (after Glaessner) . The fossil evidence For the metazoan phyla, the fossil record of the later Precambrian is sparse but revealing (Durham, 1979; Paul, 1980; Runnegar, in press). It is well exemplified by the famous Ediacara fauna of South Australia and elsewhere, shown in Figure 5 (Glaessner, 1969, 1971, 1972, 1976, 1979, 1980; Glaessner and Wade, 1966; Wade, 1970, 1972; Ford, 1980; Birket-Smith, 1981). Beginning about 680 million years ago, cnidarians, polychaetes, echiuroids, arthropod-like animals and various other metazoans were already present. Evolution had thus reached the stage of complex, soft-bodied coelomate animals by the time metazoan fossils began to appear. This is further borne out by the appearance of pogonophorans before the onset of the Cambrian (Brasier, 1979) and a variety of complex skeletonized forms such as molluscs, echinoderms, brachiopods and trilobites (Figs 6, 7) in the early Cambrian (Runnegar, 1980a, and in press) . This evidence tells us that the important separations of the metazoan phyla took place more than 700 million years ago, before any fossil metazoans were laid down. It also tells us that the fossils that are present all belong to discrete phyla, some modern, some extinct, and include no animals that can be interpreted as missing links. No new phyla are directly known to have evolved since the onset of the Cambrian, although some groups that might be accorded the status of phyla (Archaeocyatha, Hyolitha, Trilobita and others; see Fig. 25) became extinct in the Palaeozoic. Fig. 7. The Cambrian evolution of the trilobites (after Bergstrom, 1979). Proc. Linn. Soc. N.S.W., 106 (2), (1981) 1982 D. T. ANDERSON 157 Confirmation of these conclusions arises from the recent studies of the Middle Cambrian Burgess Shale fauna (Conway Morris, 1979a; Whittington, 1979, 1980). In this work we find extraordinary evidence of the early presence of a variety of soft bodied animals belonging to modern phyla. The Burgess Shale fauna includes priapulids, polychaetes, marine uniramians and even chordates (Figs 8-11). The phylum Crustacea is also represented in the Burgess Shale by Canadaspis, and in the Upper Cambrian by ostracods (Fig. 16) and cephalocarids (Briggs, 1977; Miller, OOM LO Sis): Ji (G Ze 12 Figs 8-12. 8, A polychaete of the middle Cambrian Burgess Shale (after Conway Morris, 1979b). 28 mm. 9, A priapulid of the middle Cambrian Burgess Shale (after Conway Morris, 1979a). 180 mm. 10, The middle Cambrian marine uniramian Aysheaza (redrawn from Whittington, 1980). 60 mm. 11, The middle Cambrian chordate Prkaza (based on Conway Morris and Whittington, 1979). 40 mm. 12, The middle Cambrian Opabznza, phylum unknown (after Whittington, 1980). 50mm. In addition, the Burgess Shale fauna contains many representatives of groups of animals that have not survived and do not belong within the modern phyla. These include such forms as Opabinia (Fig. 12), Hallucigenia (Fig. 13) and strange arthropods like Yohora (Fig. 14), Marrella (Fig. 15) and Odorata (Conway Morris, 1979a; Whittington, 1979, 1980, 198la; Briggs, 1981). It seems that there were several (perhaps many) distinct groups of arthropods in the Cambrian fauna in Proc. Linn. Soc. N.S.W., 106 (2), (1981) 1982 158 ORIGINS AND RELATIONSHIPS AMONG ANIMAL PHYLA addition to the Trilobita, Crustacea, Chelicerata and Uniramia (Whittington, 1979, 1981b; Manton and Anderson, 1979). Whether Opabinia, Hallucigenza, the strange arthropods and the other unique animals of the Burgess Shale should be interpreted as members of discrete extinct phyla is debatable, but they are certainly indicative of the existence of Cambrian groups of animals that do not fit the definitions of any of the modern phyla. Like the Precambrian and early Cambrian metazoans, none of the Burgess Shale animals is intermediate between or directly ancestral to any of the recognized phyla. Their structural complexity, elucidated now in impressive detail, gives further evidence of a long prior history of metazoan evolution. AS NS IES = (Soaat erie ae 14 UA XQ Figs 13-16. 13, The middle Cambrian Halluczgenca, a fossil worthy of a Bestiary (after Whittington, 1980). 18mm. 14, The middle Cambrian arthropod Yohoza, phylum unknown (after Whittington, 1980). 23 mm. 15, The middle Cambrian arthropod Marrella, phylum unknown (after Whittington, 1980). 20 mm. 16, The upper Cambrian ostracod Vestrogothia (after Muller, 1979). 0.5 mm. Proc. Linn. Soc. N.S.W., 106 (2), (1981) 1982 D. T. ANDERSON 159 From the point of view of the origins and relationships of the animal phyla, then, the fossil evidence provides nothing directly, except a confirmation of the timing of events. This puts the onus directly on functional morphology and embryology to provide some answers within this context. Functional morphology Curiously enough, what this discipline has done for phylogenetic relationships among the animal phyla is mostly negative. The recognition that animals are highly functionally integrated and adapted in relation to habit and habitat has led to the further realization that they are often more different than they seem. Common ancestries then become much less plausible. The separation of the arthropods into three modern phyla and probably several extinct phyla (Anderson, 1973; Manton, 1973, 1977; Schram, 1978; Manton and Anderson, 1979; Whittington, 1979, 1981b) is perhaps the most spectacular example of how functional morphology has demolished old beliefs, but there are others. The fact that structure is functionally related to habit and habitat is well illustrated in a broad sense by the case of the coelomic worms. Functional studies (Clark, 1979) have demonstrated a clear relationship between: 1. an unsegmented coelom and burrowing slowly in compact substrata, 2. asegmented coelom and burrowing actively in looser substrata, 3. a trimeric coelom and tubicolous life. Each of these conditions could have evolved independently more than once and none needs any other as a functional prerequisite. For example, there is no reason to suppose on functional morphological ground that the Sipunculida and the Priapulida, both unsegmented coelomate burrowers, are related to one another. The Phoronida and the Pogonophora both show a trimeric tubicolous condition, but the REPART pre annular TRU postannular A FOREPART TRUNK OPISTHOSOMA Fig. 17. The adult and larva of a pogonophoran (after Southward, 1980). Proc. Linn. Soc. N.S.W., 106 (2), (1981) 1982 160 ORIGINS AND RELATIONSHIPS AMONG ANIMAL PHYLA Pogonophora (Fig. 17) are now thought to be specialized descendants of the annelids (Southward, 1980; Jones, 1981), while phoronids are clearly lophophorates (Zimmer, 1973; Hermann, 1980). The trimeric coelomate condition, therefore, has evolved more than once and of itself is not evidence for phylogenetic affinity, only of adaptation to tubicolous life. Even tubicolous polychaetes such as serpulids tend to a trisomic condition, while retaining metameric segmentation. This case epitomizes the general effect of functional morphology on phylogenetic thinking. It has made us more wary of overemphasizing general resemblances and more honest about the gaps in the story. It has also made us realize the likelihood of convergent evolution among complex phyla and even more, the likelihood of con- vergent evolution among the animals of the simpler phyla. On the other hand, functional morphology has not assisted in the positive identification of interphylum relationships, except perhaps in relating ectoprocts to endoprocts (Nielsen, 1977) and pogonophorans to annelids (Southward, 1980). Fig. 18. Typical stages in polychaete spiral cleavage, (a) Arenzcola (b) Amphitrite (after Anderson, 1973). apical tuft stomach protonephridium Fug. 19. A polychaete trochophore larva. Proc. Linn. Soc. N.S.W., 106 (2), (1981) 1982 D. T. ANDERSON 161 Fig. 20. Stages in the spiral cleavage of the egg of the pogonophoran Nerezlinum murmanicum (after Gureeva, 1979). Comparative embryology Where, then, are we to turn for evidence of phylogenetic relationships between phyla? Fortunately, for a number of phyla, an answer still lies in embryology. Detailed functional analyses of development have shown that developmental patterns can remain astonishingly conservative despite wide divergences in adult form and complexity. It is necessary, of course, to make allowance for functional specializations of the developmental process in relation to its own requirements such as planktotrophic adaptations and/or the presence of yolk in the embryo. Taking these into account, we can discern certain well-established cases where embryological studies strongly indicate phylogenetic affinities. Spiral cleavage development is one such case. The pattern of spiral cleavage development which includes 4d mesoderm and a subsequent larva of the trochophore type (Figs 18-20) links many phyla, including molluscs, annelids, pogonophorans and uniramians, with the platyhelminths (Anderson, 1973, 1979; Gureeva, 1979; Bakke, 1980). Mostly these phyla cannot be positively linked on other evidence. Other kinds of spiral cleavage development occur in other phyla, linking them more remotely to the 4d group. These include (Figs 21, 22) rotifers, gastrotrichs, nematodes and crustaceans (Anderson, 1973, 1979, in press ; Joffe, 1979). Another major grouping on embryological grounds is recognizable from the radial cleavage, deuterostome, trimeric pattern of development which links the hemichordates, chordates, echinoderms and chaetognaths (Philip, 1979; Hermann, 1980, Pross, 1980). This kind of evidence works well for complex animals where enough information is available, though a number of groups of complex animals have still not been investigated sufficiently to allow phylogenetic inferences to be drawn about them in embryological terms. de aren . caaee if 1A ay 2B 2D Fog. 21. Stages in the spiral cleavage of the egg Fig. 22. Stages in the cleavage of the egg of a of a rotifer, Neogossea (after Joffe, 1979). nematode, Prionchulus (after Joffe, 1979). Proc. Linn. Soc. N.S.W., 106 (2), (1981) 1982 162 ORIGINS AND RELATIONSHIPS AMONG ANIMAL PHYLA Embryological evidence is not satisfactory, on the other hand, for the simpler groups, where development offers less information. Sponges, ctenophores and cnidarians cannot be related to each other or to other phyla on embryological grounds. The need for an integrated functional interpretation of the sequence of development must also be emphasized. Evidence of this kind, when incomplete, is easily misused. Attempts to relate the Pycnogonida to the Chelicerata and the Pentastomida to the branchiuran Crustacea (Riley et al., 1978; Schram, 1978) on the basis of inadequate description of a few embryo stages (Fig. 23) must remain totally unconvincing, though there are good functional morphological reasons why the pycnogonids should be included in the phylum Chelicerata (Manton, 1978; Bergstrom et al., 1980). dorsal organ ganglion somite Fig. 23. Two stages in the embryonic development of the pentastomid Rezghardza sternae (after Riley et al., 1978). PORIFERA * CNIDARIA 9O4/ 4 X CTENOPHORA xf iy Ui Can stage Blastuloid Choanoflagellate colony Fig. 24. The phylogeny of the Radialia, as interpreted by Salvini-Plawen (1978) Proc. Linn. Soc. N.S.W., 106 (2), (1981) 1982 D. T. ANDERSON 163 o jes) ~ | < (@) a Oo fo) fo) xn > oD ; : | ate Pe ale S 2y < Zz | = i) ?4— —Archaeocyatha Zz > CHOANOFLAGELLATES -|— — — > ?-—} ——Porifera a mneee ZOOFLAGELLATES — —|— —@»> ?-—-—} —-Cnidaria ? — —-Ctenophora pe _ Platyhelminthes [ + 7 ~ _-Entoprocta - a 7 ao GAG) > SK ~>Ectoprocta | / SPIRAL \_ ee | CLEAVAGE | ~ —Nemertea { \ WORMS AP ~ SL U---\NSV IX. > Sipunculida | 4 \\ Se / \ JA. > Echiurida / yw / SS i \ \ \ Annelida / \\ Ppogonophora / NNN / * ‘“Uniramia / NGA | Ba ee \ Tardigrada ao aS \ CILIATA / _ SPIRAL . Mollusca Ei is So CLEAVAGE y= anes E \. WORMS L& wl nmi cmUstaced > - ROSES i aa Se pei Chelicerata | Ss XS ~ ~ x SRS ~~ Rotifera a ~Nematoda ~ Nematomorpha Gastrotricha — — Kinorhyncha G TRIMERIC i ~( COELOMATE Pee Et ean ea WORMS _/ Sieh ~Brachiopoda pease S ( DEUTERONT \\ STOMES x ) Hemichordata Se See Se _— Chaetognatha ~ Chordata Ny ‘Echinodermata ?-|— —-Priapulida — —-Acanthocephala lies === = —— alas — — Pentastomida ies + 7 yo ——Trilobita hee ——_ 2+ — —Hyolitha ee 27 —— OTHER | a Fal ge Q4+—— — —) ) _ BURGESS —s igeser Pam SHIAUE Conglomerate y <. This paper & This fauna Elmside Formation Philip (1971) : ; nN SR Qe! pb} ¥ = as Silt one tities Cowridge cw Ue Mudstoneqy~ JU Siltstone i : ah See Las Rormation Link & Druce 2 (1972) Graptolite faunas of Jaeger (1967) Rosebank Shale Black Bog Shale Formation Silverdale Formation Laidlaw Formation Cliftonwood Limestone O’Briens Creek Sandstone Douro Volcanics Fig. 1. Stratigraphic position of fauna, in relation to other graptolites and conodonts. Column modified after Link and Druce (1972). Cowridge Siltstone 4 Formation /:: “SPRING MOUNT’ XY Fig. 2. Location of fauna. Map modified from Link and Druce (1972). Proc. Linn. Soc. N.S.W., 106 (2), (1981) 1982 C.J. JENKINS 169 For the greater part, the section is structurally uncomplicated and dips on the bedding are at 35° towards 130°. Only at the northern end do small faults and gentle folds appear. The preservation of the graptolites is good. They are flattened but not tectonically deformed, and the carbonized periderm is usually still present. A few sparsely-distributed trilobites, brachiopods, conularids and _ scolecodonts are associated with the graptolites. Some beds contain abundant broken plant material. THE GRAPTOLITES Linograptus posthumus Richter 1875. This form is exceedingly abundant in all collections through about 20 m stratigraphic thickness. Several specimens with sicular cladia have been recovered (Figs 3A,B). As in typical forms of this species in other regions the thecal spacings are about 9-10 th/cm. The stipes widen imperceptibly from 0.3-0.4 mm at th 1 to 0.7-0.8 mm distally. Minimum widths between the distal thecal apertures are 0.3 mm. Broken, separated sections representing just the metasiculae are 1-1.2 mm long and rather tubular in shape (0.2 mm wide). Distal thecal apertures occasionally show evidence of cortical thickening and low, rounded lateral lappets (Fig. 3D). The longest observed stipe portion is 3 cm (not including the long extended virgula). L. posthumus is a long-ranging species from the Ludlovian into the Lochkovian (Devonian) and is of little use for dating the fauna precisely. Monograptus cf. angustidens Pfibyl 1940. The graptolites represented in Figs 3 F, I-L, O-Q appear to constitute a single, morphologically variable species. This is indicated by the slight continuous variation that affects virtually all dimensions. In each case the variation coefficients are about 15%. The variation is also seen in their general appearances. Thus, while some specimens are like Monograptus boucekz Pribyl in having a straight dorsal margin to the thecal series at the proximal end, others have a tapered stipe proximally, which is more like M. angustidens. The collections are thus apparently intermediate between these two species. Jaeger (1967) has already described M. boucekz from the stratigraphically lower Cowridge Siltstone of the Yass syncline. These older specimens, however, show no tendency to adopt the characteristics of M. angustzdens. The following dimensions are representative of the collections. The rhabdosomes may reach 40 mm in length, but are usually less than 20 mm. The siculae average 1.2 mm in length (range: 0.9-1.8 mm) and their apices reach to levels between the th 1 aperture and just above the th 2 aperture. In specimens with a straight dorsal margin to the thecae, the 0.25 mm wide sicula produces a strong dorsal deflection of the rhabdosome. In others that are related more closely to M. angustzdens this deflection only corrects for the strong curvature between th 1 and th 10. A virgellar spine usually projects down and ventrally at 20-35" from the axis of the rhabdosome. The thecae are strongly hooked, though in a few individuals this form may weaken very slightly for the thecae beyond th 10. Stipe widths are 0.6-0.9 mm at th 1, 1.0-1.3 mm at th5, 1.25-1.5 mm at th 10 increasing steadily to (at most) 1.7 mm distally. Stipe widths between the thecal apertures grow from 0.4 mm just above th 1, to 0.9 mm distally. Thecal spacings are 13-14/cm proximally and 10-11/cm beyond th 10. The distal interthecal septa measure about 0.8 mm in length and are inclined at about 20° to the dorsal margin. Some of the dimensions differ consistently from the values found in overseas specimens of M. angustidens and M. bouceki, perhaps reflecting geographic Proc. Linn. Soc. N.S.W., 106 (2), (1981) 1982 LATE PRIDOLIAN GRAPTOLITES Proc. Linn. Soc. N.S.W., 106 (2), (1981) 1982 C. J. JENKINS 171 subspeciation. The siculae are generally shorter and do not extend up as far as th 3. The thecae are less closely spaced proximally and the stipes are wider at full development. Monograptus cf. formosus Boutek 1931. A single incomplete specimen (Fig. 3E) represents a 9 mm long portion of strongly dorsally curved stipe that is only 0.25 mm wide between the thecae and 0.7- 0.8 mm wide at the hooked, partly-isolated thecae. The interthecal septa are about 0.5 mm long. The fragment appears to represent a middle portion of the graptolite, where the thecae are spaced at 10/cm. These dimensions compare well with those of the M. formosus from the Rosebank Shale in the Yass syncline, (Jaeger, 1967, pl, 14, figs. b, c) and from overseas successions. The form is not Monograptus paraformosus Jackson and Lenz (1969) as the thecae are only moderately isolated. Monograptus transgrediens Perner 1899. This graptolite is present, but never common in the beds. It attains 30 mm in length and 1.8 mm in width. Except for th 1 which appears to be slightly hooked or to have lappets, all thecae are simple. Distal thecae are spaced at 9th/cm, proximal thecae at about 12th/cm. The sicula is about 1 mm long and has a ventral virgella. The rhabdosome is strongly tapered at the proximal end, causing a pronounced ventral curvature of the dorsal margin. Monograptus transgrediens cf. praecipuus Pribyl 1940. Only two specimens are available (Figs 3 G-H). They possess the distinctive ventral curvature of the proximal parts in the stipes, the rather narrow stripe width (1.2-1.7 mm) in fully-developed overseas examples), and the mainly pristiograptid thecae, except for th 1 and 2 which Lenz and Jackson (1971) describe as having a ‘beak-like profile’. This may be due to the presence of substantial lappets at the apertures. The Yass specimens reach up to 23 mm in length and have stipe widths of 0.5-0.7 mm at th 1, 0.7-0.8 mm at th 5, and 0.8-1.0 mm at th 10. One possible distal fragment (not figured) has a width of 1.8 mm. One sicula is 1.3 mm long, with its apex lying level with the aperture of th 1. The primary difference between this and Overseas representatives is the lesser stipe width at th 10. FAUNAL RELATIONSHIPS Late Silurian and Early Devonian graptoloid faunas show no detectable geographic provincialism and can be used for very accurate worldwide biostratigraphic correlation (Jaeger, 1978). The fauna in the lower Elmside Formation is clearly late Pridolian in age. It is a natural continuation of the graptolite faunas occurring lower in the Yass sequence and it contains several species in common with the older faunas. M. angustidens (as M. unzformis angustidens) is reported as a short-ranged species characteristic of the very base of the Lochkovian (Jaeger, 1977) . However, it is also recorded from the very uppermost Pridolian, apparently below the first appearance of Monograptus uniformis uniforms Pfibyl — as for instance in Canada, Fig. 3. Graptolites from the Elmside Formation, near Yass, NSW. A-D. Linograptus posthumus (Richter) . A, 35971. B, 35907. C, 35965. D, 35967. E, Monograptus formosus Boucek, 35910. F, I-L, O-Q. Monograptus cf. angustidens Pribyl. F, 35920. I, 35920. J, 35895. K, L, 35964. O, 35896. P, 35899. Q, 35908. G-H. Monograptus transgrediens cf. praecipuus Pribyl. G, 35911 & 35916. H, 35915. M-N, 35925. All specimens held in Australian National University Palaeontology Collection. All magnifications are X 5.0 except C, D, F, K and L which are X 10. Proc. Linn. Soc. N.S.W., 106 (2), (1981) 1982 Nei LATE PRIDOLIAN GRAPTOLITES Poland and probably Czechoslovakia (Lenz and Jackson, 1971; Tomcezyk et al., 1977; Teller, 1964; Jaeger, 1967). Representatives in the lower Elmside Formation, appear to represent an early intermediate of this form with the older species M. boucekz, which is usually regarded as typical of graptolite horizons from lower to middle Pridolian. M. unzformzs — the primary species which indicates the base of the Devonian — has not appeared in any of the graptolite collections. A further indication that the faunas are not Devonian is the presence of M. transgredzens (Jaeger, 1977). The Elmside Formation graptolite assemblages reinforce the conodont-based biostratigraphy of Link and Druce (1972) which dated limestones bearing Icriodus ‘woschmzdti in the upper Elmside Formation as basal Lochkovian. However, they have made it possible to locate the boundary of the Devonian more accurately in the succession, and show that it lies in the middle part of the Elmside Formation. ACKNOWLEDGEMENTS The first discovery of the fauna was made during a Geological Society of Australia Sunday picnic (November 1980). I am indebted to Dr K. A. W. Crook for drawing attention to the occurrence and checking a draft of the manuscript. References Jackson, D. E., and Lenz, A. C., 1969. — Latest Silurian graptolites from Porcupine River, Yukon Territory. Bull. geol. Surv. Can. 182: 17-29. JAEGER, M., 1967. — Preliminary stratigraphical results from graptolite studies in the Upper Silurian and Lower Devonian of southeastern Australia. J. geol. Soc. Aust. 14: 281-286. ——, 1977. — Graptolites, pp. 337-345 zn MartTINssoN, A., (ed.), The Silurcan-Devonian boundary. (International Union of Geological Sciences, Series A, number 5.). Stuttgart: Schweizerbart’sche Verlagsbuchhandlung. ———, 1978. — Late graptoloid faunas and the problem of graptoloid extinction. Acta pal. Pol. 23: 497- 521. LENz, A. G., and JAcksOoN, D. E., 1971. — Latest Silurian (Pridolian) and Early Devonian Monograptus of northwestern Canada. Bull. geol. Surv. Can. 192: 1-24. Link, A. G., 1970. — Age and correlations of the Siluro-Devonian strata in the Yass Basin, New South Wales, J. geol. Soc. Aust. 16: 711-22. ———, and DruceE, E. C., 1972. — Ludlovian and Gedinnian conodont stratigraphy of the Yass Basin, New South Wales, Bull. Bur. Miner. Resour. Aust. 134: 1-136. Puitip, G. M., 1971. — The Silurian-Devonian boundary in Australia. Geol. Newsletter (1971) 1: 12-15. Strusz, D. L., 1980. — The Encrinuridae and related trilobite families, with a description of Silurian species from southeastern Australia. Palaeontographica (A) 168: 1-148. TELLER, L., 1964. — Graptolite fauna and stratigraphy of the Ludlovian deposits of the Chelm borehole, eastern Poland. Stud. geol. Pol. 13: 1-88. Tomczyk, H., PajcHLowA, M., and Tomczykowa4, E., 1977. — Poland, pp. 65-83 7m MARTINSSON, A., (ed.), The Stlurtan-Devonian boundary. (International Union of Geological Sciences, Series A, number 5). Stuttgart: Schweizerbart’sche Verlagsbuchhandlung. Proc. Linn. Soc. N.S.W., 106 (2), (1981) 1982 Darriwilian (Middle Ordovician) Graptolites from the Monaro Trough sequence east of Braidwood, New South Wales C. J. JENKINS JENKINS, C. J. Darriwilian (Middle Ordovician) graptolites from the Monaro Trough sequence east of Braidwood, New South Wales. Proc. Linn. Soc. N.S.W. 106 (2), (1981) 1982: 173-179. Darriwilian graptolites have been found in hemipelagic mudstones interbedded with quartz-rich arenites of turbidite origin, in the Monga area, 13 km ESE of Braidwood, N.S.W. This is the oldest reliably-dated sediment of the Monaro Trough sequence, in the easternmost part of the Lachlan Fold Belt of New South Wales. The fauna is diverse and is correlated with assemblages close to the boundaries of the Da 2 and Da 3 Zones in Victoria and the Paraglossograptus tentaculatus and Diplograptus? decoratus Zones of New Zealand. C. J. Jenkins, Department of Geology, Australian National University, P.O. Box 4, Canberra City, Australia 2600; manuscript received 10 November 1981, accepted for publication 16 December 1981. INTRODUCTION This short article describes a find of Middle Ordovician graptolites from the very easternmost part of the Lachlan Fold Belt in southeastern New South Wales. The locality lies 13 km directly ESE of Braidwood, in a region of geology that has been termed the Monaro Trough by Crook and Powell (1976) and others. The discovery is a valuable one because potentially it indicates the environments of deposition at a very early stage in the development of the Monaro Trough, in the eastern part of the Lachlan Fold Belt. Furthermore, it marks a considerable extension of the extent of Middle Ordovician sedimentation into that broad region of the Lachlan Fold Belt which lies east of Canberra. Previously the nearest proven occurrence of Middle Ordovician strata was at Canberra, 75 km to the west (Nicholl, 1980). The existence of beds of this age in the region has been suspected (Webby, 1976), but only on very speculative stratigraphic grounds. The Wagonga Beds, for example, have been thought to range through the Middle Ordovician, but are now believed to be entirely Upper Ordovician in age (Jenkins et al., in prep.). The locality lies well to the east of the strike of the Middle Ordovician graptolite localities in Victoria — for instance those at Tabberabbera (Beavis et al., 1976, p. 34). Other Ordovician graptolite localities lie scattered around Braidwood (Sherrard, 1954, 1962) but none of these faunas is older than the Eastonian division of the Upper Ordovician. GEOLOGICAL SETTING The locality occurs in the southern bank of a road cutting on the highway, 650 m northwest of the crossing of the Mongarlowe River (Fig. 1). It lies 15.5 km by road east of Braidwood, near Currajugg homestead and at grid reference 655673 on the Monga 1: 100,000 sheet. At the present time there is no formal lithostratigraphic name for the sediments. The graptolites occur in two of the four isolated lumps of black mudstone that are Proc. Linn. Soc. N.S.W., 106 (2), (1981) 1982 174 DARRIWILIAN GRAPTOLITES GOULBURN e YASS © COOMA Fig. 1. Location of fauna. exposed in the road cut. All lie within 20 m of the western end of the outcrop. Dips are generally at 35° to the east but open folding, faulting and the intrusion of a number of 1-3 m thick (?)dolerite dykes causes variations. The sediments are all slightly hornfelsed. The mudstones appear to be of hemipelagic origin and are surrounded by well- bedded grey- and purple-toned, coarse- to fine-textured greywackes, siltstones and mudstones which are interpreted as turbidites. Most beds are very continuous laterally, but some of the turbidite-siltstones and all of the hemipelagic black mudstones rapidly pinch out laterally, in one case through 150 mm thickness in only 1-3 m distance. This may have been due to ‘boudinage’ effects in the soft-sediments (McCrossan, 1958) or to penecontemporaneous and small-scale slumping of the deposits — a common process on continental slopes (Nardin et al., 1979). Four separate occurrences of the black mudstone have been discovered, each less than 1 m in extent and 150 mm thick. The westernmost has disrupted bedding and disoriented fragmented graptolites, both suggesting a degree of post-depositional fluidization or remobilization. The lenses clearly do not represent clastic blocks within the greywackes. One of the greywackes has been examined in thin section (ANU slide 13110). The estimated proportions of the minerals are: 30% quartz (angular, unstrained grains) , 20% altered coarse detrital micas (?originally biotite, now sericite) and 50% groundmass (probably mainly of sericite). Grains of reworked black mudstone and particles of green detrital chlorite and green euhedral tourmaline were also present, but altogether constituted far less than 1% of the rock. THE GRAPTOLITES The following species have been identified in the mudstone bearing the most prolific fauna, about 12 m from the western limit of the outcrop and about 2 m above the roadside gutter level: Didymograptus cognatus Harris & Thomas, Tetragraptus sp., ?Pseudobryograptus sp., Isograptus sp., ?Apiograptus crudus (Harris & Thomas), Glossograptus acanthus Elles & Wood, Paraglossograptus cf. tentaculatus (J. Hall), Cryptograptus cf. schaefert Lapworth, Glyptograptus intersitus Harris & Proc. LINN. Soc. N.S.W., 106 (2), (1981) 1982 C. J. JENKINS 175 Fog. 2. A-D, Didymograptus cognatus Harris & Thomas; A, B, proximal ends, specimens 35869, 35887, C, D, distal stipe portions, 35837, 35837; E-G, Tetragraptus sp., 35858, 35877, 35878; H-J, Isograptus sp., 35880; 35883; 35882; K-N, Glossograptus acanthus Elles & Wood, 35876, 35855, 35847, 35886; O, scalariform Glossograptus indet. 35874; P, ?Apiograptus crudus Harris & Thomas, 35888; Q-T, Paraglossograptus cf. tentaculatus (J. Hall), 35862, 35846, 35856, 35851; U-Y, Cryptograptus inutzles Hall, 35871, 35885, 35863, 35867, 35868. All at x5 magnification. All specimens in the Australian National University Geology Department Palaeontology Collection. Proc. Linn. Soc. N.S.W., 106 (2), (1981) 1982 176 DARRIWILIAN GRAPTOLITES Thomas, Diplograptus? decoratus (Harris & Thomas). A brachiopod — perhaps Palaeoglossa — has also been collected. The other black mudstones are apparently barren except for one about 2 m from the western end of the cutting, which has yielded one small specimen of each of D. cognatus and ?Glossograptus sp. The state of preservation of the graptolites is poor and tectonic strain has produced the thin and flattened specimens which are now replaced by phyllosilicates. The bars drawn on Figs 2 and 3 represent the direction of maximum relative elongation as seen in the bedding. In some cases taxonomic problems and the poor preservation have led to some difficulties with the identifications. The remainder of this section discusses the details of the identifications. The specimens of D. cognatus (Fig. 2 A-D) are fragmentary. They are close in form and in their dimensions to several described species including Didymograptus Fig. 3. A-F, 2G, J. Glyptograptus intersitus Harris & Thomas, specimens 35844, 35849, 35875, 35842, 35839, 35873, 35879, 35865; ?H, I. Pseudoclimacograptus differtus (Harris & Thomas) 35854, 35848; 2K, L. Diplograptus? decoratus (Harris & Thomas) 35838, 35864. All at x5 magnification. All specimens in the Australian National University Geology Department Palaeontology Collection. Proc. Linn. Soc. N.S.W., 106 (2), (1981) 1982 C. J. JENKINS 177 compressus Harris & Thomas and Didymograptus nicholsoni Lapworth so the specific identification is not entirely certain. Small Tetragraptus species similar to that of Fig. 2 E-G are common in Middle Ordovician strata worldwide, but few have been formally named or distinguished from Lower Ordovician forms. The name ‘T. clarkfielde’ of Thomas (1960) which probably represents conspecific Darriwilian material from Victoria is unfortunately a nomen nudum. The identification of 2Pseudobryograptus sp. is especially tentative, being based only on a few stipe fragments. The small Isograptus sp. (Fig. 2 H-J) is a common form, close to the Darriwilian Tsograptus sp.’ of Cooper (1973, p. 89). It may represent a population of small-sized I. forczpiformzs Ruedemann or I. wictoriae divergens (Harris) which are both common large species at this level (Cooper, 1973). ?Apiograptus crudus is identified on the basis of only one specimen (Fig. 2 P) which is clearly not Glossograptus as the thecae are simple and dichograptid. Furthermore, the long spines and produced ventral apertural processes of that genus are clearly not present and the rhabdosome is bluntly rounded proximally. In spite of the poor preservation and the tectonic distortions, an identification as Phyllograptus or Pseudotrigonograptus can be discounted for the distal end is untapered, a nema is present, and there is no evidence of the scalariform 3rd and 4th stipes. Although ?4. crudus is generally regarded as restricted in range to the Yapeenian (Cooper, 1979, p. 80; Cooper and McLaurin, 1974) it appears to be present in the Darriwilian in both N.S.W. and Victoria. Harris and Thomas (1935, fig. 1, no. 13) record a specimen in association with Darriwilian faunas east of Bendigo, Victoria and this specimen was included in the synonymy of A. crudus in Cooper and McLaurin (1974). The identifications of the glossograptids and cryptograptids are fairly free of problems and follow those of Cooper (1979). G. acanthus (Fig. 2 K-N) is an especially common form. Some small specimens (Fig. 2 U-Y) probably represent Cryptograptus inutzlis, a species that is commonly recorded at this horizon. The laciniae on the specimens identified as P. cf. tentaculatus (Fig. 2 Q-T) are often difficult to see, but a few instances have been arrowed on the figures. The rhabdosomal shape and the dimensions of the specimens agree with those of P. tentaculatus. Another specimen in the fauna (Fig. 20) probably represents one of the above glossograptids in scalariform view. The details of thecal form on the diplograptids have been badly affected by poor preservation and the tectonic strains, particularly where the specimens have undergone a strong degree of shortening (e.g. Fig. 3 B, F). Nevertheless P. dzffertus, G. intersitus and D.? decoratus can be recognized from a number of the better- preserved specimens (Fig. 3). None of the specimens of D.? decoratus has the distal vane structure which is so often present in the species. FAUNAL RELATIONSHIPS AND CORRELATIONS The fauna is obviously very similar to those of the Darriwilian in Victoria and New Zealand. These similarities are the basis of proposed stratigraphic correlations. The zonation of the Victorian faunas is based on a succession of diplograptid species and the presence together of G. zntersttus and D.? decoratus in the Braidwood fauna indicates that it represents a level close to the boundary between the Da 2 and Da 3 zones of Victoria (Thomas, 1960). However, the greater abundance of G. zntersitus suggests that it lies within the Da 2 zone. The remainder of the species are consistent with this conclusion, except perhaps ?A. crudus which is generally regarded as significantly older, and also P. dzffertus which Thomas (1960) depicts as a Da 4 zone species. However as noted already D. crudus is known from the Darriwilian in Victoria. Furthermore, P. differtus was recorded from the “D 1 zone” by Harris and Proc. Linn. Soc. N.S.W., 106 (2), (1981) 1982 178 DARRIWILIAN GRAPTOLITES Thomas (1935) and in Harris (1935); (their now superseded ‘D 1 zone’ equates with the present Da 2 and Da 3 zones; Webby 1976, p. 422). The zonation of the Middle Ordovician graptolites from the South Island of New Zealand recently described by Cooper (1979), is based on the first appearances of a few diverse forms of graptolites. Thus, the occurrence of D.? decoratus in the Braidwood assemblage indicates the D.? decoratus Zone. However its association with P. cf. tentaculatus, G. acanthus and G. intersztus is suggestive of horizons rather low in the zone. The general species composition confirms this as it matches that of the underlying P. tentaculatus Zone more closely than the Zone of D.? decoratus. The only species inconsistent with such a level is, again, ?A4. crudus. Middle Ordovician faunas of similar aspect are known also from Texas (Zone 9 of ‘Hallograptus etheridge: (= P. tentaculatus) ; Berry, 1960) and Peru (Lemon and Cranswick, 1956). Comments on the relations between the Braidwood fauna and the two other Middle Ordovician graptolite assemblages known in N.S.W. from Canberra (Opik, 1958) and near Narrandera (Sherrard, 1954) must await re-examinations of these collections, though it is clear from the listed faunas (Packham, 1969, pp. 89, 102) that levels in the Darriwilian are again represented. CONCLUSIONS The locality extends the known extent of Middle Ordovician deposition in south eastern Australia. It is a significant one, being the lowest reliably-dated horizon in the Lachlan Fold Belt east of the Canberra-Molong High, in the area termed the Monaro Trough. The sediments are not significantly different in character from the younger flysch, being quartz-rich turbidites and black shales, indicating that similar, reasonably deep ocean conditions prevailed at this early stage in the region’s development. ACKNOWLEDGEMENTS Dr M. J. Rickard (ANU) first discovered the locality. Peter Ward and Dr K. A. W. Crook (ANU) have helped with subsequent collecting. Both Dr Rickard and Dr Crook kindly checked an early draft of the article. References BEAvIs, F. C., FERGUSON, J. A., TALENT, J. A., TATTAM, C. M., and Tuomas, D. E., 1976. — Ordovician. In Douctas, J. G., and Fercuson, J. A., (eds), The geology of Victoria. Spec. Publ. Geol. Soc. Aust. 5: 24-44. Berry, W. B.N., 1960. — Graptolite faunas of the Marathon Region, West Texas. Publ. Bur. Econ. geol. Univ. Texas 6005: 1-179. Cooper, R. A., 1973. — Taxonomy and evolution of Isograptus Moberg in Australasia. Palaeontology 16: 45-115. ———, 1979. — Ordovician geology and graptolite faunas of the Aorangi Mine area, north-west Nelson, New Zealand. Palaeont. Bull. geol. Surv. N.Z. 47: 1-127. ———, and McLaurin, A. N., 1974. — Apiograptus gen. nov. and the origin of the biserial graptoloid rhabdosome. Spec. Pap. Palaeont. 13: 75-85. Crook, K. A. W., and PowELL, C. Mc. A., 1976. — The evolution of the southeastern part of the Tasman geosyncline. 25th Int. Geol. Congr. Field Guide 17A. Harris, W. J., 1935. — The graptolite succession of Bendigo East, with suggested zoning. Proc. R. Soc. Vict. 47: 314-37. ———, and Tuomas, D. E., 1935. — Victorian graptolites (New Series) Part III. Proc. R. Soc. Vict. 47: 288-313. JENKINS, C. J., Kipp, P. R., and MILts, K. J., 2m prep. — Eastonian age graptolites from the Wagonga Beds, near Batemans Bay, N.S.W. Proc. LINN. Soc. N.S.W., 106 (2), (1981) 1982 C. J. JENKINS 179 Lemon, R. R. H., and Cranswick, J. S., 1956. — Graptolites from Huacar, Peru. Publ. Mus. Hist. nat., Lima, C. (Geol.) 5: 1-31. McCrossan, R. G., 1958. — Sedimentary ‘Boudinage’ structures in the Upper Devonian Ireton Formation of Alberta. J. sedim. Petrol. 28: 316-20. Narpin, IT. R., Epwarps, B. D., and Fors.in, D. S., 1979. — Santa Cruz Basin, California Borderland: Dominance of slope processes in basin sedimentation. Spec. Publ. Soc. Econ. Paleont. Min. 27: 209- 21. Nico, R. S., 1980. — Middle Ordovician conodonts from the Pittman Formation, Canberra, A.C.T. B.M.R. J. Aust. Geol. Geophys. 5: 150-153. Opik, A. A., 1958. — The Geology of the Canberra City district. Bull. Bur. Miner. Resourc. 32, 1-99. PAckHAM, G. H. (ed.), 1969. — The geology of New South Wales. J. geol. Soc. Aust. 16: 1-654. SHERRARD, K. M., 1954. — The assemblages of graptolites in New South Wales. J. Proc. R. Soc. N.S.W. 87: 73-101. ———, 1962. — Further notes on assemblages of graptolites in New South Wales. J. Proc. R. Soc. N.S.W. 95: 167-78. Tuomas, D. E., 1960. — The zonal distribution of Australian graptolites. J. Proc. R. Soc. N.S.W. 94: 1-58. Wespy, B. D., 1976. — The Ordovician System in South-eastern Australia, pp. 417-446 in BAssETT, M. G. (ed.). The Ordovician System: proceedings of a Palaeontological Association symposium, Birmingham, September 1974. Cardiff: Univ. Wales Press and National Museum of Wales. Proc. Linn. Soc. N.S.W., 106 (2), (1981) 1982 H aah an ray PV ha : i ‘Na : Ayal scalars ; ’ ; Wd Ty Da Ne) Pi ie iy me wed MT GL bey ‘s 106 Ci/2): The Fauna of Australian Seagrass Beds PAT HUTCHINGS Hutcnines, P. A. The fauna of Australian seagrass beds. Proc. Linn. Soc. N.S.W. 486-65, (1981) 1982: 181-200. Australia has a rich seagrass flora (26 species), widely distributed in estuaries and shallow coastal embayments. The fauna of the seagrass beds is diverse and consists of infaunal, epifaunal, epiphytic and epibenthic species. This community is composed of temporary and permanent residents exhibiting tidal, diurnal and seasonal fluctuations. The literature dealing with the fauna of Australian seagrass beds is reviewed and possible food-web relationships and energy-flows within seagrass communities are postulated. P. A. Hutchings, Curator of Marine Invertebrates, Australian Museum, P.O. Box A285, Sydney South, Australia 2000; manuscript recetved 19th May 1981, accepted in revised form 16 December 1981. INTRODUCTION Seagrass beds occur in estuaries and shallow coastal embayments throughout mainland Australia (Larkum, 1976a). In many estuaries they form extensive beds. In eastern temperate Australia, seagrass beds often occur with mangrove and salt marsh communities which together constitute .coastal wetlands (Hutchings and Recher, 1974; AMSA, 1977). In tropical regions this relationship rarely occurs. Instead, seagrass beds may be associated with coral reefs. Recently in the Caribbean, Ogden and Zieman (1977) have shown the close relationship between coral reefs and seagrasses. Herbivorous coral reef fish migrate from the reef to nearby beds of Thalassza testudinum (Banks ex Konig) to feed. Carnivorous fish and grazing sea urchins move at night from the reef to the seagrass beds to feed. Such migrations have not been documented in Australia. The role of seagrass beds in estuaries and shallow coastal embayments has become apparent during the past decade in Australia. Seagrass beds provide important habitats for juvenile stages of commercial fish species (Bell et al., 1978a, b) and prawns (Young, 1978), They stabilize and trap sediments improving water clarity (Orth, 1977). A rich fauna is associated with the beds. There are few herbivores of Australian seagrasses and seagrass beds also produce large amounts of organic matter which is then broken down by the fauna of seagrass beds to detritus on which estuarine food chains are dependent (Kirkman and Reid, 1979). Overseas, considerable information on the fauna of seagrass beds exists. An overview of seagrass faunas throughout the world is provided by Kikuchi and Péres (1976). They describe the fauna of each region, and then attempt to assess the importance of seagrasses in providing a structural habitat, shelter and a food resource. Food web relationships and energy flows in the seagrass ecosystems are discussed. Other studies have concentrated on regional seagrass communities (Orth, 1973, 1977; Heck, 1977, 1979). Heck studied in detail the epibenthic invertebrates of Thalassza beds in Panama. Considerable seasonal fluctuations in both species number and abundance occurred. Sites varied in species composition and abundance and Heck attributed this to the proximity of surrounding habitats especially coral reefs which Proc. LINN. Soc. N.S.W., 106 (2), (1981) 1982 182 FAUNA OF AUSTRALIAN SEAGRASS BEDS contain a number of species that utilize the seagrass meadows. Other workers like Nelson (1979) have studied in detail one faunal component, such as the amphipods. Other workers have experimentally manipulated seagrass beds to assess the value of habitat complexity in determining species richness and abundance (Heck and Wetstone, 1977). Young and Young (1977, 1978) have studied the effects of predation at various times of the year by caging, adding dense populations of suspension feeders and providing organic enrichment. Breakdown of seagrass into detritus has been investigated by Robertson and Mann (1980). Isopods and amphipods are important in shredding the intact dead seagrass leaves. The decomposition of organic detritus has been extensively reviewed by Fenchel (1977). Utilization of this detritus and export of this productivity out of the seagrass beds has been considered by Ogden and Zieman (1977; Ogden, 1976) who have studied fish feeding in Caribbean seagrass beds. OBJECTIVES OF THE REVIEW To collate and synthesize all the available information on Australian seagrass faunas. This will reveal the dominant species whose life histories and trophic status merit detailed investigations. Such information is essential for estimates of secondary productivity to be made for although it is widely recognized that Australian seagrass beds are important, this is largely based on inference from overseas data. Data on secondary productivity of Australian seagrass beds are non-existent. From overseas data and limited Australian data there are sound financial and ecological reasons for maintaining seagrass beds (Larkum, 1981). Maintenance requires management but we lack the necessary information for implementation. Instead, as the use of estuaries increases (Burnley, 1974), seagrass beds are being degraded. Larkum (1976b) has shown that the seagrass beds on the northern shores of Botany Bay, New South Wales, have disappeared because of pollution from Cooks River. Seagrass beds on the southern shores of Botany Bay are affected by increased wave action and decreased water clarity caused by dredging associated with the port development. Similar port and industrial developments in Cockburn Sound, Western Australia, have also caused widescale disappearance of seagrass beds during the 1970s and this is well documented by Cambridge (1979). As estuarine resources are limited in Australia it may be necessary to sacrifice some seagrass beds to keep others. This necessitates the ranking of beds in terms of ecological importance. This can only be done if detailed information exists on a variety of seagrass communities in differing environments. Finally, it is hoped that this review will reveal the major gaps in our knowledge and stimulate further research. FLORISTICS OF SEAGRASSES Seagrasses are angiosperms which are able to live permanently in seawater. Not only are they able to fulfil their normal metabolic activities when fully submerged in seawater but can also complete their generative cycle. All seagrasses have in common a well developed anchoring system. In addition, they can compete successfully with primarily marine organisms such as algae. This combination of properties is so rare among angiosperms that only fifty two species of seagrass have been described (den Hartog, 1970, 1979). Other species are currently being described (Larkum, pers. comm.). Twenty six of these species occur in Australia (Larkum, 1976a). Since 1976 three new species have been described Pos¢donza senuosa, Cambridge and Kuo, P. Proc. Linn. Soc. N.S.W., 106 (2), (1981) 1982 P. HUTCHINGS 183 augustifolia, Cambridge and Kuo, and Halophila tricestata, Greenway. Larkum suggests that latitude 25°S on the west coast and 30°S on the east coast of Australia form a boundary between temperate and tropical species. South of these latitudes the most successful community is formed by Poszdonza australis Hook f. This species forms extensive monospecific beds in sheltered conditions. In New South Wales and Western Australia the beds rarely extend below -10m (depth below mean low water mark), but in the Great Australian Bight, they occur in more exposed conditions to depths of -40m. Poszdonia is not a colonizing species and only appears after the substrate has been stabilized by species such as Amphibolis antarctica (Labill) in Western and South Australia and Zostera capricornt Aschers in New South Wales. The development of P. australis is accompanied by the formation of a tough, wave resistant rhizome ‘matte’ (cf. Molinier and Picard, 1952). Larkum (1976a) has summarized on a state-wide basis the distribution of seagrasses. Since Larkum’s review, Shepherd and Womersley (1976) have described the seagrass communities around St. Francis Island in the northern Great Australian Bight. New South Wales State Fisheries are currently documenting seagrass distributions along the coast (Evans and Gibbs, in press) . Considerable gaps in distribution data occur, especially in tropical regions where many species occur. Den Hartog (1970) provides the only information on tropical seagrass communities, none of which reach the density or areal size of Poszdonza australis in temperate waters. a map detailing the distribution and extent of Australian seagrasses is given by McComb et al., 1981. Associated with the seagrasses may be rhizophytic companion species such as Caulerpa and Halimeda. Such species occur mainly in sub-tropical or tropical communities. A film of microalgae, especially diatoms, occurs on the muddy or sandy substratum in between the seagrasses. Algae are often trapped by the seagrass and may continue to grow in the shelter of the beds (den Hartog, 1979). This associated flora has been neglected in Australia. Primary Productivity Studies McRoy and McMillan (1977) have suggested that such common plants as seagrasses with such a high primary production must be of immense ecological importance. Consequently, ‘seagrass ecosystems’ have become a popular focus for research throughout the world, including Australia. Recently, considerable advances have been made in measuring the rates of primary production of some temperate species here in Australia. ; The preferred method of measuring primary productivity amongst seagrass workers seems to be measuring the production of organic matter directly, i.e. by measuring the increase in blade length. This method has the distinct advantage that it can give an integrated measure of primary productivity over a number of days. The method has drawbacks, in that it measures only the productivity of part of a plant, the frond, and it does not take into account any losses of dissolved organic carbon compounds (Larkum, 1981), which may be significant (Kirkman and Reid, 1979). Rates of leaf blade production of Posidonza australis range from 0.7-5.5 g dry wt. m~ day, depending on site and season. It is significantly higher in clear waters, such as Jervis Bay, than in disturbed areas like Botany Bay, New South Wales. Maximum rates occur in the middle of the geographical range of this species, which is Spencer Gulf, South Australia (West and Larkum, 1979). Leaf production is markedly seasonal with a maximum production between November and February, and a minimum in July to August. However, if flowering Proc. Linn. Soc. N.S.W., 106 (2), (1981) 1982 184 FAUNA OF AUSTRALIAN SEAGRASS BEDS occurs in July, then leaf blade production is severely curtailed. In Botany Bay three to four crops of leaves are produced annually. In areas of frequent storms, biomass figures may be relatively constant (West and Larkum, 1979). In some areas such as Westernport Bay, Victoria, nitrogen limits the growth of Heterozostera tasmanica during spring and early summer (Bulthuis and Woelkerling, 1981). This species and some other seagrasses, including Zostera marina, can absorb nutrients from the sediments and translocate them to the leaves (Twilley e¢ al., 1977). Zostera capricorn shows a similar growth pattern (Larkum et al., in press) which seems to be more closely related to water temperature than to solar radiation. New leaf production continues fairly uniformly throughout the year with a slight increase in the warmer months. Turnover times for leaves vary from about 33 days in the summer months to 67 days for the winter period (Kirkman et al., 1982). In Port Hacking, New South Wales, Zostera grows 3-4 times faster than Poszdonia australis. As these two seagrasses occur in varying densities, comparisons of production per unit area of seagrass beds are more meaningful. Zostera production varies from 0.56-1.82 g m7? day" while P. australis ranges from 0.57-1.44 g m? day! — throughout the year (Kirkman and Reid, 1979; Kirkman et al., 1982). SEAGRASS FAUNAS STRUCTURE The animal communities of seagrasses are considered here in four major categories : a. infauna — animals living in the sediment associated with the rhizomes of the seagrasses, b. epifauna — animals living on the surface of the mud at the base of the seagrass, c. sessile communities on the blades of the seagrass and, d. epibenthic animals, swimming in the water column between the seagrass. SAMPLING METHODS When designing sampling programmes the three-dimensional nature of the habitat must be considered, for each of the dimensions has an associated fauna. Seagrass communities which are exposed to air during low tide, lose one of the dimensions as the blades collapse eliminating the space between them. This dimension is restored on a rising tide. As discussed later, the fauna shows distinct diurnal, tidal and seasonal variations in composition. The fauna can also be divided into permanent and temporary residents. Permanent residents spend their entire lives within the seagrass beds. Temporary residents spend only part of their life within the community. This element is rather heterogeneous and consists of : 1. fishes and crustaceans which spend part of their juvenile life in the beds, 2. tidal migrants, which move into the beds on a rising tide and leave on an ebb tide and terrestrial species which feed in the beds during low tide, 3. vertebrate grazers, such as ducks, turtles and dugongs. Thus a seagrass community is in a constant state of flux as the relationship of the community and its environment change. The community can be considered as having a central core of residents whose numbers fluctuate seasonally, and migrants whose composition varies according to the time of day, tide and season. To devise one sampling programme to study seagrass faunas is impossible. Each group of fauna necessitates a specific sampling programme, with various periodicities and intensities. Proc. Linn. Soc. N.S.W., 106 (2), (1981) 1982 P. HUTCHINGS 185 a. Infauna: The dense rhizome ‘matte’ of seagrass beds which is well developed in Poszdonza is the major problem to overcome in sampling infauna. Collett et al. (in MS) used a hand-operated corer with a good cutting edge to penetrate the ‘matte;. Poiner (1980) successfully used a 0.1 m? Smith — MacIntyre grab in beds of Halophila and Hutchings et al. (1978) used a similar grab in Zostera and patchy Poszdonia. This grab is not efficient in dense Poszdonza. Collett et al. (in MS) found that 15-20 samples (0.03 m*) in Poszdonza beds in New South Wales were needed to sample 70% of the smaller infaunal species. The samples were collected from an area of visually uniformly dense beds. Hutchings e¢ al. (1978) sampling Zostera and patchy Poszdonza found that the number of samples required to reach the upper asymptote of the respective species-volume curve varied from 4-10 samples. All infaunal studies should ascertain the minimum number of samples necessary to collect 70% or more of the species present. This has not been standard practice. The most commonly used mesh size is 1 mm (Collett ef al., in MS; Poiner, 1980; Rainer, 1981; Rainer and Fitzhardinge, 1981). To facilitate the sorting of animals from in amongst the large amount of plant matter, add Rose Bengal to the 7% formalin solution, until a deep pink colour develops. All living, animal material becomes dark pink in colour, considerably facilitating sorting. Micro-crustaceans can be collected from the sieved fractions by running a steady stream of water over the samples. The micro-crustaceans float off the samples which are gently agitated and are retained in a sieve. Poiner (1980) determined the density of seagrasses in the area of his infaunal samples by counting the number of nodes per 0.25 m?. He also analysed the sediment using standard methods of granulometric analysis (Buchanan, 1971). Such associated data are invaluable in interpreting the benthic fauna and should become standard practice. Lewis and Stoner (1981) have recently compared macrofaunal sampling methods in seagrass beds. Three different-sized core samples were tested for their relative efficiencies. Core samples were sieved with both 0.5 and 1.0 mm mesh screens. The smaller core was the most efficient and this was attributed to the problems of penetrating the dense root ‘matte’. The 0.5 mm sieve retained approximately 50% more individuals than the 1.0 mm sieve. For sampling larger infauna such as bivalves. Anadara and Trapezza, 0.25 m®* quadrats are laid and the sediment searched within each quadrat by a diver. Hutchings and Recher (1974) used 40 such quadrats. b. Epzfauna: Techniques that have been used include visual observations by divers along transects (Rainer, 1981; Shepherd, 1974) and small seines pulled through the seagrass (Hutchings and Recher, 1974). Both methods are qualitative. No quantitative study of this community has been undertaken in Australia. c. Epiphytes : No quantitative survey has been carried out on this fauna. Hutchings and Recher (1974), while measuring density of seagrasses, scraped off the associated fauna. The polychaetes are described by Hutchings and Rainer (1979). d. Epzbenthic: Young (1978) sampled with a roller beam trawl between fixed markers Proc. LINN. Soc. N.S.W., 106 (2), (1981) 1982 186 FAUNA OF AUSTRALIAN SEAGRASS BEDS fortnightly over a 12 month period. All stations were sampled within 2 days of the new and full moon. All samples were collected within 2 hours of the night high tide. The trawls were made towing into the current. Various mesh sizes were tested and optimum size for crustaceans was 2 mm for the body of the net and a cod end of 1 mm. A larger mesh size of 2.5 mm for the body of the net was used in a subsequent study (Young, 1981). Robertson and Howard (1978) sampled the zooplankton in Zostera and Heterozostera beds, by using plankton tows of 10 minutes duration. The mesh size was 254 um and the net had a 30 cm mouth diameter. The upper zone of the water column was sampled by maintaining the top of the net at 10 cm. The volume of water sampled was measured by a digital flow meter. For fish, Scott (1981) used a mesh of 1.5 mm and trawls of 1-2 minutes duration covering a distance of 30-50 m. Sampling was considered sufficient when no additional species were found. Robertson and Howard (1978) used a 30 m seine net (1 cm mesh bunt and 1.25 cm mesh wings) to catch large, fast swimming*and mid-water living fish. Smaller benthic fish were obtained by using a hand laid 10 m (1 mm mesh) seine net. Samples were taken on the rising and falling tide, by day and by night. Robertson measured the bias introduced into density estimates through fish avoiding the seine net in the study area. The results indicate no apparently significant difference in the efficiency of the net in capturing the major pelagic species, but efficiency values were much lower for benthic species. Although the efficiency of capture differed, the nettings accurately reflected the relative abundance of species within each group. DESCRIPTION OF THE FAUNA a. INFAUNA Infauna has been the best studied faunal component of seagrass beds probably because of the relative ease of collecting quantitative data. Even so, quantitative data are restricted to temperate beds of Posedonza, Ruppia, Halophila and Zostera. A recent survey of Poszdonza beds along the New South Wales coast by Collett et al. (in MS) has shown that a large number of benthic invertebrates occur (363 species) dominated by polychaetes, crustaceans and molluscs. A large number of species (211 or 54.5%) are restricted to one site and of these 102 (26.9%) are represented by single individuals. Only two species occur at all sites. No characteristic Posidonia fauna occurs. Rather, the fauna is determined by the hydrological and sediment characteristics of a particular bed. Latitudinal gradients seem unimportant. Analysis of the fauna shows that marine-dominated sites are more similar than estuarine sites. The number of species and individuals occurring at sites varied greatly as did community indices. Many of the species occurring in Poszdonza beds also occur in Zostera beds or in seagrass-free sediments, in the same area. Other seagrass communities studied quantitatively include those of Tuggerah Lakes, New South Wales (Powis and Robinson, 1980). In the lakes, Zostera capricornt, Ruppia spiralis and Halophila ovalis occur typically as discrete monospecific stands. At only one site, does a mixed community of Zostera and Ruppia occur. Powis and Robinson found that the sites could be split on the basis of sediment and associated vegetation, in terms of number of species, and individuals, diversity and characteristic species. Seagrass communities are generally more diverse than mud communities. The Zostera sites share some species in common, although the numerical dominance varies between sites. Ruppia sites have lower numbers of species than Zostera sites. Mixed communities of Zostera and Ruppza have the highest Proc. Linn. Soc. N.S.W., 106 (2), (1981) 1982 P. HUTCHINGS 187 number of species and diversity. Halophila beds are intermediate between Ruppia and Zostera in terms of number of species and diversity. Beds of Halophila ovalis and H. spinulosa have been extensively studied in Moreton Bay by Poiner (1980). He found 137 species, of which 51% are polychaetes, 33% crustaceans (including 16% amphipods and 8% decapods) and 11% molluscs (9% of which are bivalves). The majority of these studies have shown that a richer infauna is associated with seagrasses than with bare substrate. In Moreton Bay, one of the beds of Halophila ovalis, which Poiner (1980) studied, disappeared during the winter, as a result of storms and strong wave action, and reappeared during the calm, summer months. Poiner found with an increase in plant growth during the summer, diversity of the infaunal community increases significantly. The rise is due to an increase in the number of individuals rather than additional species. Several reasons have been advanced as to why seagrass communities have richer benthic faunas than bare sediment. Seagrasses dampen the sediment water interface by stabilizing sediments, baffling currents and damping wave action (Zieman, 1972; Orth, 1977). This stabilizing effect of seagrasses could allow increased settlement of larvae, prevent adults or juveniles from resuspension and reduce disruptive disturbances by, for instance, nektonic predators, surface scavengers and wave action (Poiner, 1980). Orth (1977) reported the impact of the blue crab (Callinectes sapidus) on the infauna of seagrasses. Powis and Robinson (1980) suggest that seagrass beds may provide a habitat of high structural complexity which, according to MacArthur (1965), would have a high species diversity (Heck and Wetstone, 1977). However, the abundance of infauna seems to vary according to the type of seagrass. Powis and Robinson (1980), as already mentioned, found that Zostera capricornz has a richer fauna than either Halophila or Ruppia. Hutchings and Recher (1974) sampled the infauna of Posedonza australis and Z. capricorni and found that Zostera supports a more diverse community, both in terms of number of individuals and species than Poszdonza. Powis and Robinson suggest that the well-developed ‘matte’ of Poszdonza restricts the development of an infaunal community. By contrast, some of the Posidonza beds sampled by Collett et al. (in MS) yielded large numbers of individuals and species (e.g., Towra Point yielded 148 species with 4,480 individuals per m’). Powis and Robinson (1980) further suggest that the differences in infaunal diversity of different seagrasses may be related to the efficiency of individual species in binding sediment. They suggest that Zostera is more efficient than Ruppza. Orth (1977) also stresses the stabilizing properties of Zostera marina. Stable substrates according to Sanders (1969) can support communities of high diversity. Ruppza, with its poor root system and small blades, is not as efficient as Zostera in either of these roles and hence has lower diversity communities. Powis and Robinson do not speculate on Halophila beds which have intermediate values of diversity. This species does not form dense beds in Tuggerah Lakes. There is some contradiction in the literature, regarding the endemicity of seagrass faunas. Kikuchi and Péres (1976) developed the concept of parallelism which suggests that seagrass beds share many species in common or a ‘species’ is replaced by a closely related equivalent in a different geographical locality. They give several examples from the epifauna of Zostera beds in Japanese waters and imply it would be valid for infauna. The data from Pos¢donza beds along the New South Wales coast do not support this concept. For Collett e¢ al. (in MS) were not able to define a characteristic fauna. Each site has a unique assemblage of species. Powis and Robinson (1980), similarly, did not record any species totally restricted to a particular species of seagrass in Tuggerah Lakes. Also, in Careel Bay many species occurring in Proc. Linn. Soc. N.S.W., 106 (2), (1981) 1982 188 FAUNA OF AUSTRALIAN SEAGRASS BEDS TABLE 1 Species of Polychaetes recorded from nine Zostera capricorni beds in N.S.W. Harmothoe praeclara Parahalosydna chrysostichtus Paralepidonotus ampulliferus Sigalion ovigerum Sthenelazs boa Eumida sanguinea Gyptis sp. Phyllodoce novaehollandiae Sphaerosyllis sublaevis S. nr. semzverrucosa Typosyllis varvegata Australonerets ehlersi Ceratonereis erythraeensis* C. mirabilis Platynerets dumerilu antipoda Nephtys australienszs Glycera americana Marphysa sanguinea Lumbrinerts latrezlli Leztoscoloplos bifurcatus L. normalis Natneris grubez australis Phylo felix Polydora soczalts Prionospio malmgrenz Pseudopolydora kempi Magelona dakini Chaetopterus variopedatus Cirriformia filigera C. tentaculata Hyboscolex longiseta Armandia intermedia Polyophthalmus pictus Barantolla lepte Mediomastus californzensts Notomastus torquatus Scyphoproctus djiboutiensis Arenicola bombayensis Owenia fusiformis Amphictets dalmatica Lysilla apheles L. pacifica Rhinothelepus lobatus Proc. Linn. Soc. N.S.W., 106 (2), (1981) 1982 Merimbula Towra Point x x Weeney Bay mx x x KK Bonnet Bay Sites Oo BY) sj eee ah DO) ae O xX >< xX x xX xX > XG gt XE xX XG RX xX XK XxX xX xX OX xX XxX xX xX xX x XxX xX xX XOX XxX xX XxX >.< >< x OX xX xX X peas xx x x Lower Myall River Wallis Lake xx x Wollomba River Number of sttes COO OU mT DDT OO i OT Om OR OOOO OO ODT OO OO 8 P. HUTCHINGS 189 TABLE 1 (continued) Species of Polychaetes recorded from nine Zostera capricorni beds in N.S.W. Sites n g sl 2 28 oes ew Stein eas a 5 Sali Oe Wea aging 9 omen cata eS GO Bs ty, Ce ice Oy ells ea Bee) cee Gur steel a etc mene mais Se academe at ES aes OP\ ie) Seles Streblosoma acymatum x 1 Pista typha Xx xX 2 Amphiglena pacifica Xx 1 Janua (D.) foraminosa x 1 J. (D.) brasiliensis x 1 Total number of species 1 1 WG as 1s7y 7 9 We 4 Sources of Data: Careel Bay — Hutchings and Recher (1974) and Hutchings and Rainer (1979) Towra Point and Weeney Bay — ALS (1978) Wallis Lake — Hutchings et al. (1978) Lower Myall River — Weate (1975) ; Atkinson et al. (1981) Wollomba River — Day (1975) Bonnet Bay — Branagan et al. (1976) Tuggerah Lakes — Powis and Robinson (1980) Merimbula — Hutchings and Murray (in prep.) *See footnote p. 190. Proc. Linn. Soc. N.S.W., 106 (2), (1981) 1982 190 FAUNA OF AUSTRALIAN SEAGRASS BEDS the Zostera beds also occur in Postdonza (Hutchings and Recher, 1974). No characteristic species are associated with the Zostera mueller? and Heterozostera tasmanica communities in Westernport Bay, Victoria (Littlejohn et al., 1974). Similarly, the Zostera communities in Gippsland Lakes (Poore et al., 1977) lack characteristic species. However, Poiner (1980) found 35 species restricted to beds of Halophila ovalts and H. spinulosa but does not list them. Even if the concept of Kikuchi and Pérés (1976) is rejected, there is evidence that similar trophic levels operate within seagrass communities. Collett et al. (in MS) found that the very different faunal communities in the Poszdonza beds are all dominated by detritus feeders, either feeding on detritus within the sediment (deposit feeders) or in suspension (filter feeders). Similarly, the infauna of Zostera muelleri and Heterozostera tasmanica in Westernport Bay, Victoria (Littlejohn et al., 1974) are dominated by detritus feeders. Of the 44 species present, 14 are selective deposit feeders, 10 are non-selective deposit feeders and 3 are filter feeders. Polychaetes represented 50% of the biovolume. As polychaetes are often the dominant group of animals in the infauna, I have extracted from the literature species occurring in Zostera in New South Wales. Many of these studies were carried out by myself or else I confirmed the identifications so that consistent names have been given to the species (for several of these species pose taxonomic problems). The sites in Table 1 have been arranged latitudinally. Zostera capricorn flourishes in a variety of salinities, from almost fully marine, as in Careel Bay, to upper estuarine areas such as Wollumba River which flows into Wallis Lake or Bonnet Bay on the Woronora River, where considerable fluctuations in salinity occur. Fully marine sites such as Careel Bay have the richest polychaete fauna (37 species) and upper estuarine areas such as Bonnet Bay and Wollumba River have relatively few species (3 and 4, respectively) which may be abundant, however. Such data reinforce the concept that although seagrasses provide some benefits in stabilizing sediment, factors such as salinity regimes have an overriding effect on the faunal community. The polychaete fauna varies considerably between sites, and 30 species (62.5%) occur at one site only; no species is shared by all 9 sites. However, certain families of polychaetes are commonly present in infaunal communities, for example, all sites possess at least one species of nereid (Australonerezs ehlersi, Ceratonerezs erythraeensis*, C. mirabilis or Platynerets-dumerilu antipoda) and all sites except Bonnet Bay have at least one species of capitellid (Barantolla lepte, Medzomastus californiensis, Notomastus torquatus, Scyphoproctus djiboutiensis). These two families are well represented in other seagrass beds (Collett e¢ a/., in MS) and Hutchings and Turvey (in MS) have collected 10 species of nereids from seagrass beds in South Australia. Capitellids are non-selective deposit feeders. Nereids are often assumed to be carnivores, as they have well developed jaws but are probably scavengers or omnivores. Fluctuations in seagrass infaunal communities i. Seasonality Seagrass beds exhibit strong seasonality of growth, maximum growth occurs during the spring and summer. In the autumn and winter, much of this growth dies and is broken off and may be washed out of the seagrass beds. Severe storms may cause loss of seagrass leaves, and even rip out the seagrass ‘matte’. *Ceratonereis erythraeensis is a complex of species currently being described by Hutchings and Turvey (in MS). Ceratonerezs erythraeensis Fauvel probably does not occur in Australia. Proc. Linn. Soc. N.S.W., 106 (2), (1981) 1982 P. HUTCHINGS 191 Hutchings and Recher (1974) recorded strong seasonality of infauna in both Posidonia and Zostera beds. Both showed fluctuations in the number of species, with maximum numbers occurring in June. Numbers of individuals also fluctuated with maximum numbers in November in Zostera beds and in June in Poszdonza. Mass settlement of some species occurs. Two polychaetes Scoloplos (S.) simplex (originally referred to as Haploscoloplos n. sp.) and Ceratonerezs erythraeensts (originally referred to as Nerezs diversicolor but see footnote here) settle in large numbers between June and November and, similarly, the mollusc Macoma deltozdalis. Many species occur sporadically in the seagrass beds (Hutchings and Recher, 1974). Poiner (1980), working on Halophila communities, measured the community flux at 2-monthly intervals. Community flux is a measure of the degree to which species contributions within a particular site (measured as weighted change in abundances of individual species) are exchanged or given up altogether between sampling periods. The community flux is low for permanent beds of Halophila, whereas the flux in beds showing marked seasonal changes, fluctuate widely. In these beds the number of species remains relatively constant, but an increase in the number of individuals is strongly correlated with an increase in the density of Halophila. i. Flooding Severe floods occurred in Tuggerah Lakes while Powis and Robinson (1980) were sampling the infauna. Halophila beds were severely reduced. Associated with the rain was a severe storm with strong winds. This resulted in the devastation of some beds of Zostera and Ruppra. After the floods limited re-establishment of fauna in muddy areas occurred; this did not happen in the seagrass areas. Powis and Robinson suggest that the loss of Halophila roots, resulted in a decrease in habitat availability. In the lower Myall River and Myall Lakes, New South Wales, widespread destruction of Zostera occurs periodically, probably as a result of salinity reductions. The effect on the associated benthic faunas has not been substantiated (Atkinson et ail, Ns), b. EPIFAUNA No quantitative data are available on this community. Qualitative studies have been carried out in Cockburn Sound, Western Australia. Echinoderms (Marsh and Devaney, 1978), and molluscs (Wells, 1978, and Wilson et al., 1978) were identified. With the demise of seagrass beds (Cambridge, 1979) one of the echinoids Temnopleurus michaelsenz is also declining in numbers. This species was abundant in the early 70s. Similar studies have been carried out in the upper Spencer Gulf of South Australia by Shepherd (1974). He found an inverse correlation between the density of Posidonia and the abundance of filter-feeding molluscs, Penna, Malleus and Trichomya, the sea urchin Helzoczdaris and the ascidian Polycarpa. Water movement under the seagrass canopy is about 30% of that above the canopy. This may explain the relatively small numbers of filter-feeding organisms occurring in dense seagrass beds. The bivalve Anadara trapezia recruits successfully only periodically so that within a seagrass bed one age or size class dominates (Hutchings and Recher, 1974; Dixon, pers. comm.). Since 1974, there has not been a successful recruitment to Careel Bay, New South Wales, and widespread mortality (perhaps naturally as the population ages) is currently occurring (Recher, pers. comm.) . Cc. SESSILE COMMUNITIES ON THE BLADES OF THE SEAGRASSES Proc. Linn. Soc. N.S.W., 106 (2), (1981) 1982 192 FAUNA OF AUSTRALIAN SEAGRASS BEDS The epiphytic fauna of Australian seagrass beds has not been studied as a separate component. Anink (1980) reports that the number of epifaunal species so far observed is far fewer than the number reported settling on artificial substrates. d. EPIBENTHIC COMMUNITIES Young (1978) has extensively studied the distribution of one component of epibenthic communities, namely prawns in Moreton Bay, Queensland. Four species of penaeid prawn were studied in detail. They utilize all available littoral areas in the bay, only some of which contain seagrasses. The distribution of prawns differs but is related to the prevailing salinity and temperature regime and the seagrasses. The maximum abundances of post-larvae of all four species occur within two seagrass communities (Zostera capricornt and Halophila ovalis or Z. capricorni, Halodule uninervis and H. ovalzs). Three of the four species show increase in abundances in summer which coincides with the increased abundance of Z. capricornz and H. ovalzs. However, the other species has reduced numbers at this time suggesting no simple relationships. Young was unable to isolate or attribute the relative importance of these individual factors. A similar complex pattern exists among the entire epibenthic community (Young and Wadley, 1979). The type of sediment and the presence of seagrasses are again important but the major factor is salinity. Salinity regimes are major determinants in seagrass distribution, but depth is also important (Young and Kirkman, 1975). Young and Wadley (1979) were unable to quantify the importance of specific seagrass beds in determining their associated epibenthic fauna. In a subsequent study, Young (1981) attempted to eliminate some of these variables. He selected two adjacent beds of Postdonza australis and Zostera capricornt, in similar depths of water in Port Hacking, New South Wales. Significant differences in the species composition and abundances of the epibenthic fauna occur, with Posidonia being the richer community. Young suggests that the differences in the fauna of the two seagrasses are controlled by external events leading to variable recruitment success, although he also suggests that the resources offered by the seagrasses present a habitat which many species recognize and utilize. Wadley (1981), working in the same area, showed that temporal heterogeneity of the epibenthos, although significant, is small compared with spatial heterogeneity. Differences in the fauna associated with the month and season of sampling show strong interaction with the type of habitat. The absence of change in the major components of the fauna during flood and the overall temporal homogeneity in samples indicate relatively stable communities despite the unpredictable environment. Young (1981) further suggests that the more diverse community of Poszdonza australis is not significantly more stable than the Zostera community. This greater diversity of epibenthic fauna is the reverse of infaunal communities, where the Zostera is more diverse (Hutchings and Recher, 1974). Fish Relatively few studies have been carried out on the fish communities of seagrass beds. Scott (1981) sampled the fish fauna of Posedonia beds in Geographe Bay, Western Australia. Nineteen species occur, and the most abundant species is the odacid Neoodax radiatus (weed whiting) . Proc. Linn. Soc. N.S.W., 106 (2), (1981) 1982 P. HUTCHINGS 193 Hoese (1978, pers. comm.) has studied the fish communities occurring in seagrass beds in coastal lagoons in New South Wales. These lagoons are periodically closed to the sea. The communities fluctuate both in terms of number of individuals and species. Recruitment appears to be solely from the sea, when the lagoons are open, rather than from other parts of the lagoon. This is in contrast to Scott’s (1981) study where the fish complete their life cycle within the seagrass. Other fish studies by Bell et al. (1978a, b), Conacher et al. (1979) and Littlejohn et al. (1974) have been related to investigating the feeding patterns of the fish. This will be discussed in the energy flow section. Zooplankton Holoplanktonic calanoid copepods and meroplanktonic decapod larvae are numerically dominant in the water column during the day in Zostera beds in Westernport Bay, Victoria (Robertson and Howard, 1978). However, the number of such animals is significantly lower than that collected at night. This pattern of low catches near the surface during the day and high numbers at night has been recorded on a number of previous occasions (Howard, unpublished data). These obligate planktons have also been taken in planktonic trawls made close to the substrate during the day. Thus estimates of zooplankton samples taken near the surface underestimate the true density of these organisms throughout the whole water column during a 24- hour period. Amphipods are benthic during the day, at night they form a conspicuous component of the plankton. Ostracods, which are infaunal by day also enter the water column at night but at lower densities than amphipods. ENERGY FLOWS WITHIN THE SEAGRASS COMMUNITIES Kirkman and Reid (1979) have constructed a carbon budget for Poszdonza australis in Port Hacking, near Sydney. They measured biomass, calculated grazing rates by herbivores, estimated the amount of floating seagrass detached from the beds and collected the sinking Poszdonza detritus below the seagrass beds. Small seasonal variations in biomass occur. Three species of Monacanthidae (leatherjackets), Monacanthus chinensis, Meuchenza freycineti and M. trachylepszs consume large amounts of Poszdonza. Bell et al. (1978b) calculated that adult M. chinensis have a diet consisting of 41.7% P. australis by estimated volume. However, looking at the entire Poszdonia beds, herbivores do not consume an appreciable amount of attached Poszdonza (Conacher et al., 1979). The amount of floating Poszdonza is related to winds and, unlike Zostera, is not washed up onto beaches in New South Wales. However, in South Australia large beds accumulate on beaches adjacent to seagrass beds. Pos¢donza occurs typically in greater depths than Zostera. Leaves that are added to the organic detrital cover of sediments make up the majority of particulate organic C lost from the seagrass beds. Only heavily epiphytized leaves sink. Between 27-71% of the biomass collected in the traps is epibiota. Dissolved organic carbon accounted for the greatest loss of organic C from the system. A less extensive study of the Zostera beds in Port Hacking by Kirkman e¢ al. (1982) has shown that the growth of Zostera is 3-4 times that of Poszdonza in the same Proc. Linn. Soc. N.S.W., 106 (2), (1981) 1982 194 FAUNA OF AUSTRALIAN SEAGRASS BEDS environment. The production figures are for Zostera 0.56-1.82 g m™ day while P. australzs shows a production rate of 0.57-1.44 g m7” day" through the year. It seems likely that a similar breakdown of the C budget occurs in Zostera beds as Kirkman and Reid (1979) have described for Poszdonza in the same environment. Larkum (1981) has estimated the percentage contribution of each of the primary producers in the estuarine ecosystem of Botany Bay. Although seagrass beds do not occur as extensive beds, they contribute 25% to the total budget of the bay. This percentage includes the contribution made by the epiphytic algae which occur on the seagrass blades. Assuming that Kirkman and Reid’s (1979) data from Port Hacking are of general applicability, the important contribution of seagrasses to the estuarine ecosystem is the continual supply of leaves to the organic detrital cover of the sediments. Attiwill and Clough (1974) have also stressed the importance of this production of organic matter by the seagrass beds in Westernport Bay, Victoria. Little work has been done on the breakdown or shredding of this continual supply of leaves. Brand (1977) has experimentally investigated the production of detritus from the breakdown of Zostera by the amphipods Parhyalella spp. This was to quantify part of the model of the C budget proposed by Brand et al. (1974) for estuarine ecosystems. The amphipods ingest the epiphytes on the blades of the Zostera. Brand does not provide any evidence that the amphipods actually obtain nutrients from the plant matter but implies that nutrition is obtained mainly from the attached biota. The amphipods produce nitrogen-enriched faecal pellets which facilitate further breakdown of the Zostera fragments. Seagrass leaf tissue is extremely refractile (Godshalk and Wetzel, 1978) and the decay of the structural carbohydrates may be the rate limiting step in the transfer of energy through seagrass ecosystems (Harrison and Mann, 1975). Animals which can utilize these fragments appear to be mainly amphipods (Robertson and Mann, 1980), although some herbivorous polychaetes may be involved. Overseas, attempts have been made to trace the origin of detritus in the diets of estuarine animals by measuring the ratio of '?C:'*C. These ratios differ sufficiently in seagrasses and phytoplankton to allow the source of the detritus to be traced (Fry and Parker, 1979; McConnaughey and McRoy, 1979; Thayer et al., 1978). As mentioned earlier, few fish of seagrass beds in Australia are herbivores. The leatherjackets do obtain some nutrients from ingested P. australis fragments (Conacher et al., 1979). Most of the fish occurring in seagrass are carnivores (Littlejohn et al., 1974; Scott, 1981) feeding on the detritus-feeding crabs, molluscs and polychaetes (Robertson, 1980). Although Robertson and Howard (1978) have documented the importance of zooplankton for several fish species, fish diets may shift during the year as prey species fluctuate (Littlejohn et al., 1974). The importance of detritus feeders (Tenore, 1977) which dominate the benthos is related to the further breakdown of the plant matter by mechanical abrasion as it passes through the guts of the detritus feeders (Fenchel, 1970, 1977). This increases the surface area to volume ratio, and hence the bacterial and algal populations on the particles (Newell, 1965). Many of these detritus feeders are probably obtaining nutrients from the associated flora and fauna rather than from the seagrasses. During the production of detritus some organic compounds are produced which can be taken up directly by the seagrasses themselves (Bulthuis and Woelkerling, 1981) so that a close relationship between the infauna and seagrasses occurs. Where they occur, vertebrate grazers of seagrass beds may have considerable local impact on the beds. Dugong dugon, a marine mammal, occurs in large herds in coastal waters of northern Australia. The dugong feeds exclusively on seagrasses (Heinsohn and Birch, 1972). In the Townsville region they feed selectively on Proc. Linn. Soc. N.S.W., 106 (2), (1981) 1982 P. HUTCHINGS 195 Diplanthera and Cymodoce. Two other common seagrass species in the region, Enhalus acorovdes (L.f.) Royle and Syringodzum isoetifolium, are not eaten. Dugongs eat the entire plant (leaves, rhizomes and roots) , and ingest some animals and algae. They graze most heavily in beds of low density (biomass 10-30 g dry matter/m7’) , producing trails of a depth of 3-5 cm. They remove on average 63% of the seagrass from the trails with a maximum of 86% (Heinsohn et al., 1977). Two captive dugongs ate 50-55 kg of seagrass per day (equivalent to 2 tonnes dry weight/annum) (Jones, 1967). During feeding some seagrass is damaged although not ingested. Assuming a grazing efficiency of 63% and a seagrass community biomass of 94 kg dry matter/m* (Heinsohn, unpublished data from Cleveland Bay, Townsville) 3.5 ha is required to support a dugong for a year. The area over which dugongs graze is not known but should be if the impact of their grazing on seagrass beds is to be assessed. Turtles also feed on seagrasses, but not exclusively. The green turtle Chelonza mydas feeds on marine algae and seagrasses, including species of Zostera, Enhalus, Thallasstna, Postdonia and Halodule. Loggerhead turtles Caretta caretta, also eat seagrasses, but are primarily carnivores. Similarly, Hawksbill turtles Eretmochelys zmobricata, eat seagrasses but mainly as juveniles (Rebel, 1974). Wading birds are important temporary residents during low tide. Water fowl such as black swans, Cygnus atratus, are estimated to eat 230 tonnes of Ruppia and the freshwater Potamogeton a year in the Blackwood River estuary in Western Australia (Hodgkin, 1978). Delroy (1974) calculated that 90% of food of black ducks (Anas supercil), musk duck (Bzzzura lobata) and grey teal (Anas gibberifrons) consisted of tubers and seeds of Ruppia and Lamprothaminium in the Coorong, South Australia. They feed by pulling plants to the surface, thus damaging more than they eat. They feed only during the summer on seagrases in this estuary. Ferguson Wood (1959) suggested that black swans caused great damage to Zostera capricorni beds. Seagrass beds are also used by a variety of water birds including pelicans (Pelecanus conspicillatus), cormorants, etc. Pelicans and cormorants feed on schooling fish within the seagrass beds. For birds such as herons (Ardeidae) which stalk their prey, the presence or absence of seagrass beds probably determines the number of individuals in an area. In dense seagrass beds, water clarity is improved, more sedentary fish are present, giving herons ample opportunity to feed during low tide. In areas of sparse or no beds, the water is more turbid, fewer sedentary species of fish are present and herons must expend far more energy in catching fish. Such areas will have fewer herons than areas with dense seagrass beds. Recher (pers. comm.) points out that although herons and egrets are predominantly fish eaters, in marine habitats crustaceans are an important part of their diet. Migrant waders, are less dependent on seagrass beds. They are limited to low tide and will often feed in non-grass situations. They feed on molluscs, annelids and crustaceans, although a particular species may be highly selective. For example, the knot feeds on surface-living molluscs whereas the bartailed godwit Limosa lapponica feeds preferentially on polychaetes and the eastern curlew Numenius madagascariensis on Callzanassa sp. (Recher, pers. comm.). DISCUSSION Concluding this review, I would like to consider the direction of future studies of seagrass communities. Recent trends in ecological studies of these communities overseas have been to manipulate the system experimentally. Orth (1977) has been able to quantify the importance of seagrass beds in Proc. Linn. Soc. N.S.W., 106 (2), (1981) 1982 196 FAUNA OF AUSTRALIAN SEAGRASS BEDS stabilizing sediments and the production of organic material by artificially removing Zostera. He could then compare intact Zostera with clipped beds. Such methods could be extended to quantify the relative importance of different species of seagrasses in Australia. Other experiments have tested the hypothesis that seagrass provides shelter from predation, by caging experiments (Heck and Thoman, 1981). Such information is essential if the much-quoted statement that seagrasses act as nursery grounds and provide shelter and protection is to be substantiated. How do different seagrasses compare? Are some species more efficient than others? Young and Young (1978) have experimentally investigated the factors important in regulating species densities and diversities in seagrass associated benthos. Such studies urgently need to be conducted here in Australia: for this experimental approach seems to be a productive way of determining the factors important in maintenance of seagrass communities. Unless we know these factors, the management of seagrass communities cannot be soundly based. We also need more integrated studies like those of Kirkman and Reid (1979) but these should be extended to include the faunal components. Rates of turn-over of the dominant animals must be determined. The data from Australian seagrass communities clearly show the importance of seagrasses. Some of the primary productivity studies indicate the variations between sites (Larkum, 1981). Are similar fluctuations occurring in the animal communities? Larkum (1981) has suggested that these variations are sometimes caused by urban stress. Unless we understand the entire community we will not recognize the early symptoms of a community under stress, and thus be able to take remedial action. Loss of seagrasses will have serious impacts on the fishing, oyster and prawning industries (AMSA, 1977; Young, 1978). Oil spills and the use of oil dispersants and other types of industrial pollution severely affect the growth of seagrass beds (Larkum, 1976b) and hence affect the associated fauna (Jacobs, 1980). The number of oil spills and incidents of industrial pollution are increasing and unless checked will have major effects on the seagrass ecosystems and subsequently on estuarine and shallow coastal ecosystems. ACKNOWLEDGEMENTS I should like to thank Drs Arnold, Collett, Hodgkin, Rainer and Young and Mss Marsh and Wadley for providing unpublished material and reports, or papers in press. Drs Larkum, Rainer and Recher criticized an initial draft for which I am most grateful. The paper also benefited from the discussions which occurred after the presentation of this paper at the Seagrass Workshop held at Sydney University, February, 1980. Mr Glasby and Ms Handley helped in the compilation of the bibliography. References ANINK, P. J., 1980, — Epiphytic fauna of seagrass beds. National Seagrass Workshop, University of Sydney, 1980. Abstracts pp. 6-7. ATKINSON, G. E., HutcHincs, P. A., JOHNSON, M., JOHNSON, W. D., and Me vi.te, M. D., 1981. — An ecological investigation of the Myall Lakes region. Aust. J. Ecol. 6: 299-329. ATTIWILL, P. M., and CLoucu, B. F., 1974. — The role of mangrove and seagrass communities in nutrient cycling in Westernport Bay. Report to Westernport Bay Environmental Study, Ministry for Conservation, Victoria, Australia. 35 pp. AUSTRALIAN Litrorat SociETy, (ALS), 1978. — An investigation of management options for Towra Proc. Linn. Soc. N.S.W., 106 (2), (1981) 1982 P. HUTCHINGS 197 Point, Botany Bay. Report prepared for the Australian National Parks and Wildlife. 396 pp. AUSTRALIAN MARINE ScIENCES AssocIATION, (AMSA), 1977. — Guidelines for the protection and management of estuaries and estuarine wetlands. 12 pp. By Cotierr, L. and Hurcuincs, P. A. BELL, J. D., BuRcHMoRE, J. J., and PoLLarp, D. A., 1978a. — Feeding ecology of a scorpaenid fish, the fortescue Centropogon australis, from a Posidonza seagrass habitat in New South Wales. Aust. J. mar. Freshwater Res. 29: 175-85. ——,, 1978b. — Feeding ecology of three sympatric species of leatherjacket (Pisces: Monacanthidae) from a Postdonia seagrass habitat in New South Wales. Aust. J. mar. Freshwater Res. 29: 631-43. BRANAGAN, D. F., Hurcuines, P. A., RecuHer, H., Back, D., and Co..is, P., 1976. — Report on the environmental impact of a proposed residential development at Bonnet Bay. Prepared for Sutherland Shire Council by the Bonnet Bay Ratepayers Assc. 33 pp. Branp, G. W., 1977. — Epifaunal activity in seagrass breakdown. Report to Westernport Bay Enuironmental Study, Ministry for Conservation, Victoria, Australia. 17 pp. ——, Buttuuis, D., and Curr, W., 1974. — Energy flow and its regulation in the Westernport Bay ecosystem. Report to the Westernport Bay Environmental Study, 1973-74. Project Report No. 4.3.11. Ministry for Conservation, Victoria, Melbourne, December. 12 pp. & tables. BUCHANAN, J. B., 1971. — Measurements of the physical and chemical environments: Sediments. pp. 30-52 in I.B.P. Handbook No. 16. Methods for the Study of Marine Benthos, Houmes, N. A., and McInryreE, A. D., (eds.). Oxford: Blackwell. BuLTuuis, D. A., and WoELKERLING, W. J., 1981. — Effects of insitu nitrogen and phosphorus enrichment of the sediments on the seagrass Heterozostera tasmanica (Martens ex Aschers.) den Hartog in Westernport Bay, Victoria, Australia. J. exp. mar. Biol. Ecol. 53: 193-207. BurnLey, I. H., 1974. — Elements of population growth in Australia’s coastal zones: 1947-1971. pp. 3-20 in Report of Symposium on the Impact of Human Activities on Coastal Zones, Australian UNESCO Committee for Man and the Biosphere, Publ. No. 1. Canberra: Australian Government Publishing Service. CAMBRIDGE, M. L., 1979. — Cockburn Sound Environmental Study — Technical Report on Seagrass. West Australian Department of Conservation and Environment, Report No. 7. 100 pp. CouiueTr, L. C., Hurcuines, P. A., Gipss, P. J., and Co.iins, A. J., (im MS). — The macro-benthic fauna of Postdonia australis meadows in New South Wales, Australia. Submitted to Aquat. Bot. ConacHER, M. J., Lanzinc, W. J. R., and Larkum, A. W. D., 1979. — Ecology of Botany Bay. II. Aspects of the feeding ecology of the fanbellied leatherjacket, Monacanthus chinensis (Pisces: Monacanthidae) in Poszdonza australis seagrass beds in Quibray Bay, Botany Bay, New South Wales. Aust. J. mar. Freshwater Res. 30: 387-400. Day, R. W., 1975. — A biological survey of Tuncurry Creek, New South Wales. Operculum 4: 96-105. Deroy, L. B., 1974. — The food of waterfowl (Anatidae) in the southern Coorong saltwater habitat of South Australia. S$. Aust. Ornzthol. 26; 157-163. DEN HartToe, C., 1970. — The Seagrasses of the World. Amsterdam: North-Holland. ——, 1979, — Seagrasses and seagrass ecosystems, an appraisal of the research approach. Aquat. Bot. 7: 105-117. Evans, P., and Gisss, P. J., (in press). — Distribution and seagrasses in five New South Wales coastal lagoons. N.S.W. State Fisheries, Technical Report. 30 pp. FENCHEL, T., 1970. — Studies on the decomposition of organic detritus derived from the turtle grass Thalassza testudinum. Limnol. Oceanogr. 15: 14-20. ——, 1977. — Aspects of the decomposition of seagrasses. pp. 123-145 in McRoy, C. P., and HELFFERICH, C. P. (eds.). Seagrass Ecosystems, a Scientific Perspective. New York: Marcel Decker Inc. FERGUSON Woop, E. J., 1959. — Some East Australian seagrass communities. Proc. Linn. Soc. N.S.W. 84: 218-226. Fry, B., and Parker, P. L., 1979. — Animal diet in Texas seagrass meadows:d'*C evidence for the importance of benthic plants. Est. Coast. Mar. Scz. 8: 499-509. GODSHALK, G. L., and WETZEL, R. G., 1978. — Decomposition of aquatic angiosperms. III. Zostera marina L. and a conceptual model of decomposition. Aquat. Bot. 5: 301-328. Harrison, P. G., and Mann, K. H., 1975. — Detritus formation from eelgrass (Zostera marina): the relative effects of fragmentation, leaching and decay. Lzmnol. Oceanogr. 20: 924-934. Heck, K. L., 1977. — Comparative species richness, composition, and abundance of invertebrates in Caribbean seagrass (Thalassia testudinum) meadows (Panama). Mar. Bzol. 41: 335-348. ——, 1979. — Some determinants of the composition and abundance of mobile macroinvertebrate species in tropical and temperate turtlegrass (Thalassza testudinum) meadows. J. Biogeog. 6: 183-200. ——, and THoman, T. A., 1981. — Experiments on predator — prey interactions in vegetated aquatic habitats. J. exp. mar. Biol. Ecol., 53: 125-134. , and WETSTONE, G. S., 1977. — Habitat complexity and invertebrate species richness and abundance in tropical seagrass meadows. J. Bzogeog. 4: 135-142. Proc. Linn. Soc. N.S.W., 106 (2), (1981) 1982 198 FAUNA OF AUSTRALIAN SEAGRASS BEDS HEINSOHN, G. E., and BircH, W. R., 1972. — Foods and feeding habits of the dugong, Dugong dugon (Erxleben) , in northern Queensland, Australia. Mammalia 36 (3) : 414-422. ——, WakE, J., Marsu, H., and Spain, A. V., 1977. — The dugong Dugong dugon in the seagrass system. Aquaculture 12 (3) : 235-248. Hopexin, E. P., 1978. — An environmental study of the Blackwood River estuary, Western Australia, 1974-75. A Report to the Estuarine and Marine Adusory Committee of the Environmental Protection Authority, Rep. No. 1, Department of Conservation and Environment, 78 pp. Hoese, D., 1978. — Fishes in seagrass communities. Aust. Nat. Hist. 19(5) : 170-173. Hurcuines, P. A., and RECHER, H. F., 1974. — The fauna of Careel Bay with comments on the ecology of mangrove and seagrass communities. Aust. Zool. 18: 99-128. ——, Nicou, P. I., and O’Gower, A. K., 1978. — The marine macrobenthic communities of Wallis and Smiths Lakes, New South Wales. Aust. J. Ecol. 3: 79-90. ——, and Rainer, S. F., 1979. — The polychaete fauna of Careel Bay, Pittwater, Australia. J. Nat. Hist. (Lond.) 13: 745-796. —, and Turvey, S. P. (in MS). — The Nereididae of South Australia. Submitted to Trans Roy. Soc. S. Aust. Jacoss, R. P., 1980. — Effects of the Amoco Cadiz oil spill on the seagrass community at Roscoff with species references to the benthic infauna. Mar. Ecol. Prog. Ser. 2: 207-213. Jones, S., 1967. — The dugong — its present status in the seas around India with observations on its behaviour in captivity. Int. Zoo. Yearb. 7: 215-220. Kikucui, T., and Péres, J. M., 1976. — Consumer ecology of seagrass beds. pp. 147-193 in McRoy, C. P. and HELFFERICH, C., (eds.). Seagrass Ecosystems: a scientific perspective. New York: Marcel Dekker Inc. KirkKMAN, H., and Rerp, D. D., 1979. — A study of the role of the seagrass Poszdonza australis in the carbon budget of an estuary. Aquat. Bot. 7: 173-183. —, Cook, I. H., and Reip, D. D., (1982). — Biomass and growth of Zostera capricornz Aschers in Port Hacking, N.S.W., Australia. Aquat. Bot. 12: 57-67. Larkum, A. W. D., 1976a. — Recent research on seagrass communities in Australia. pp. 247-262 in McRoy, C. P. and HELFFERICH, C., (eds.). Seagrass Ecosystems: a scientific perspective. New York: Marcel Dekker Inc. ——, 1976b. — Ecology of Botany Bay. I. Growth of Poszdonza australis (Brown) Hook F. in Botany Bay and other bays in the Sydney Basin. Aust. J. mar. Freshwater Res. 27: 117-27. ——, 1981. — Marine primary productivity. pp. 370-385 in CLayTon, M. N., and Kine, R. J., (eds.). Marine Botany an Australasian Perspective. Melbourne: Longman Cheshire Press. ——, Co.etT, L. C., and WILLIAMs, R. J., (in press). — Ecology of Botany Bay. III. The standing stocks, growth and shoot production of Zostera capricorni in Botany Bay, New South Wales. Aust. J. Mar. Freshwater Res. Lewis, F. G., and Sroner, A. W., 1981. — An examination of methods for sampling macrobenthos in seagrass meadows. Bull. Mar. Scz. 31: 116-124. LITTLEJOHN, M. J., Watson, G. F., and RoBErTSON, A. I., 1974. — The ecological role of macrofauna in eelgrass communities. pp. 1-16 in SHApIRO, M. A. (ed.). A preliminary report on the Westernport Bay environmental study. Victoria: Ministry for Conservation. MacArTuHur, R. H., 1965. — Patterns of species diversity. Bzol. Rev. 40: 510-533. McComs, A. J., CAMBRIDGE, M. L., KIRKMAN, H., and Kuo, J., 1981. — The biology of Australian seagrasses. Chapter 9, pp. 259-293 in Pate, J. S., and McComs, A. J., (eds.). Biology of Australian Plants. Nedlands: University of Western Australia Press. McConnaucuey, T., and McRoy, C. P., 1979. — !°C label identifies eelgrass (Zostera marina) carbon in an Alaskan estuarine food web. Mar. Bzol. 53: 263-269. McRoy, C. P., and McMILLAN, C., 1977. — Production ecology and physiology of seagrasses. pp. 53-87 in McRoy, C. P., and HELFFERICH, C., (eds.). Seagrass Ecosystems: a Scientific Perspective. New York, Basel: Marcel Dekker Inc. Marsu, L. M., and Devaney, D. M., 1978. — The benthic fauna of Cockburn Sound, Western Australia. Part II: Coelenterata, 8 pp. Part III: Echinodermata: Ophiuroidea, 8 pp. Part IV: Echinodermata, Crinoidea, Asteroidea, Echinoidea and Holothuroidea. 23 pp. Western Australian Museum, Perth. Unpublished report submitted to the West Australian Department of Conservation and Environment. Mo.inier, R., and PicarD, J., 1952. — Recherches sur les herbiers de phanerogames marines du littoral Méditerranéen Francais. Inst. Oceanog. Annales. 27: 157-234. NELSON, W. G., 1979. — An analysis of structural pattern in an eelgrass (Zostera marina L.) amphipod community. J. Exp. Mar. Biol. Ecol. 39: 231-264. NEWELL, R. C., 1965. — The role of detritus in the nutrition of two marine deposit feeders, the prosobranch Hydrobia ulvae and the bivalve Macoma balthica. Proc. zool. Soc. Lond. 114: 25-45. Proc. Linn. Soc. N.S.W., 106 (2), (1981) 1982 P. HUTCHINGS 199 OcpEN, J. C., 1976. — Some aspects of herbivore plant relationships on Caribbean reefs and seagrass beds. Aquat. Bot. 2: 103-116. OcpEN, J. G., and Zieman, J. C., 1977. — Ecological aspects of coral reef-seagrass bed contacts in the Caribbean. Proc. Third Int. Coral Reef Symp., Miami: 377-383. OrTH, R. J., 1973. — Benthic infauna of eelgrass, Zostera marina, beds. Chesapeake Sci. 14: 258-269. ——, 1977. — The importance of sediment stability in seagrass communities. pp. 281-300. CouLt, B. C., (ed.). in Ecology of Marine Benthos. Vol. VI, Belle W. Baruch Library in Marine Science. Columbia S.C.: University of South Carolina Press. Poors, G. C. B., (in MS). — Benthic communities of the Gippsland Lakes, Victoria. ——, CotEeMAn, N., and KupEnov, J. D., 1977. — Development of a benthic research programme for the Gippsland Lakes. Paper No. 143 in the Environmental Studies Series, Ministry for Conservation, Victoria. 33 pp. Poiner, I. R., 1980. — A comparison between species diversity and community flux rates in the macrobenthos of an infaunal sand community and a seagrass community of Moreton Bay, Queensland. Proc. Roy. Soc. Qd 91: 21-36. Powis, B. J., and Rosinson, K. I. M., 1980. — Benthic macrofaunal communities in the Tuggerah Lakes, N.S.W. Aust. J. mar. Freshwater Res. 31: 803-815. RAINER, S. F., 1981. — Temporal patterns in the structure of macrobenthic communities of an Australian estuary. Est. Coast. Shelf Science 13 (6) : 597-620. ——, and FITZHARDINGE, R., 1981. — Benthic communities in an estuary with periodic deoxygenation. Aust. J]. mar. Freshwater Res. 32 (2) : 227-43. Reset, T. P., 1974. — Sea turtles — and the turtle industry of the West Indies, Florida and the Gulf of Mexico. Miami: University of Miami Press. Revised edition. ROBERTSON, A. I., 1980. — The structure and organisation of an eelgrass fish fauna. Oecologia 47: 76-82. ——, and Howarp, R. K., 1978. — Diel trophic interactions between vertically-migrating zooplankton and their fish predators in an eelgrass community. Mar. Biol. 48: 207-213. ——, and Mann, K. H., 1980. — The role of isopods and amphipods in the initial fragmentation of eelgrass detritus in Nova Scotia, Canada. Mar. Biol. 59: 63-69. SANDERS, H. L., 1969. — Benthic marine diversity and the stability — time hypothesis. Bookhaven Symp. Biol. 22: 71-81. ScorrT, J. K., 1981. — The seagrass fish fauna of Geographe Bay, Western Australia. J. Roy. Soc. W. Aust., 63: 97-102. Se SHEPHERD, S. A., 1974. — An underwater survey near Crag Point in Upper Spencer Gulf. South Australian Dept. of Fisheries, Tech. Rep. No. 1, 29 pp. , and WomersLEY, H. B. S., 1976. — The subtidal algae and seagrass ecology of St. Francis Island, South Australia. Trans. Roy. Soc. S. Aust. 100: 177-191. TENORE, K. R., 1977. — Food chains pathways in detrital feeding benthic communities: A review with new observations on sediment resuspension and detrital recycling. pp. 37-55 in Coutt, B. C., (ed). Ecology of Marine Benthos, Belle W. Baruch Institute for Marine Biology. Columbia S.C.: University of South Carolina Press. THAYER, G. W., Parker, P. L., LaCrorx, M. W., and Fry, B., 1978. — The stable carbon isotope ratio of some components of an eelgrass Zostera marina bed. Oecologia 35: 1-12. TwiLLey, R. R., BRINSON, M. M., and Davis, G. J., 1977. — Phosphorus’absorption, trabslocation and secretion in Nuphar luteum. Limnol. Oceanogr. 22: 1022-1032. Wap ey, V. A., 1981. — Spatial and temporal heterogeneity in the epibenthic fauna of estuarine sand and seagrass beds. Aust. J. Ecol. 6: 217. WEATE, P. B., 1975. — A study of the wetlands of the Myall River. Operculum 4: 105-113. WELLS, F. E., 1978. — A quantitative examination of the benthic molluscs of Cockburn Sound, Western Australia. Western Australian Museum, Perth. Unpublished report submitted to the West Australian Department of Conservation and Environment. 118 pp. West, R. J., and Larxum, A. W. D., 1979. — Leaf productivity of the seagrass Poszdonza australis in eastern Australian waters. Aquat. Bot. 7: 57-65. Witson, R. B., Kenprick, G. W., and BREARLEY, A., 1978. — The benthic fauna of Cockburn Sound, Western Australia Part 1: Prosobranch, gastropod and bivalve molluscs, Western Australian Museum, Perth. Unpublished report submitted to the West Australian Department of Conservation and Environment Vol. J and IT. Younc, D. K., and Youne, M. W., 1977. — Community structure of the macrobenthos associated with seagrasses of the Indian River Estuary, Florida. pp. 359-381 in CouLL, B. C., (ed.), Ecology of Marine Benthos, Belle W. Baruch Institute for Marine Biology. Columbia, S.C.: University of South Carolina Press. ——, 1978, — Regulation of species densities of seagrass — associated macrobenthos: Evidence from field experiments in the Indian River estuary, Florida. J. Mar. Res. 36: 569-593. Proc. Linn. Soc. N.S.W.,106 (2), (1981) 1982 200 FAUNA OF AUSTRALIAN SEAGRASS BEDS YounG, P. C., 1978. — Moreton Bay, Queensland: A nursery area for juvenile penaeid prawns. Aust. /. mar. Freshwat. Res. 29: 55-75. ——, 1981. — Temporal changes in the vagile epibenthic fauna of two seagrass meadows (Zostera capricorni, Postdonia australis). Mar. Ecol. Progress Series 5: 91-102 , and KirkMAN, H., 1975. — The seagrass communities of Moreton Bay, Queensland. Aquat. Bot. 191-202. , and WapLEy, V. A., 1979. — Distribution of shallow-water epibenthic macrofauna in Moreton Bay, Queensland, Australia. Mar. Bzol. 53: 83-97. ZiEMAN, J. C., 1972. — Origin of circular beds of Thalassia (Spermatophyta: Hydrocharitaceae) in South Biscayne Bay, Florida, and their relationship to mangrove hammocks. Bull. Mar. Sct. 22: 559-574. Proc. Linn. Soc. N.S.W., 106 (2), (1981) 1982 ADVICE TO AUTHORS The Linnean Society of New South Wales publishes in its Proceedings original papers and review articles dealing with biological and earth science. Papers of general significance are preferred but the Proceedings also provides a medium for the dissemination of useful works of more limited scope. Manuscripts will be received for assessment from non-members as well as members of the Society though non-members must communicate their works through a member. Subject to acceptance, a member’s paper may be given priority in publication over that of a non-member. Manuscripts (originals and two copies of text and illustrations) should be forwarded to the Secretary, Linnean Society of New South Wales, 35-43 Clarence Street, Sydney, Australia, 2000. Authors who are members of the Society are supplied with 25 free offprints of their papers after publication. Additional copies may be purchased, provided an order is placed when the corrected proofs are returned to the Honorary Editor. Donations towards the cost of publishing papers are always most welcome. In the case of lengthy papers or those with many illustrations or tables, contributions from authors, and especially non-member authors, may be requested at the discretion of Council. On publication a paper and the copyright thereof become the property of the Society. Requests to use copyright material should be directed to the Secretary. PREPARATION OF MANUSCRIPTS Copy must be typewritten, double-spaced throughout, on one side only of good quality A4 (210 X 297 mm) paper. Margins not less than 25 mm wide all round are required. All pages should be numbered serially and securely fastened together. The desired positions for all figures and tables should be indicated in the left-hand margin of the text. For taxonomic papers the Botanical or Zoological Codes of Nomenclature, as appropriate, must be followed. Generic and specific names should be clearly marked by underlining. Papers should be written in clear, concise English. The Style Manual for Authors and Printers of Australian Government Publications (Second Edition, 1972) is a useful guide. Spelling should conform to that preferred by the Oxford English Dictionary. The general design of a paper should follow the scheme: (1) Title and author’s name — all in capitals. (2) A concise Abstract, complete in not more than 200 words, indicating the scope of the paper. Authors should adopt the lay-out used in this issue of the Proceedings, including details of postal address but leaving spaces for editorial insertions. (3) Main text. Footnotes should be avoided.,The text may be divided into sections introduced by short headings set out as in this issue. (4) Acknowledgements, if any. (5) References. These should be cited in the text by author’s name and date, e.g., Bullough (1939) or (Bullough, 1939) according to the context and listed alphabetically by authors under References thus: BuL.Loucu, W. S., 1939. — A study of the reproductive cycle of the minnow in relation to the environment. Proc. zool. Soc. Lond., Ser. A, 109: 79-108. Titles of periodicals should be abbreviated as in the World List of Scientific Periodicals. If more than one work by the same author published in the same year is cited, use a, b, etc., after the year in both text and list of references. Titles of books should be quoted in full together with the place of publication, the name of the publisher and the edition if other than the first. Illustrations: Authors should note that illustrative matter (both photographs for half-tone reproduction and line drawings) is now printed in the text, not as separate plates. All illustrations must therefore be marked as figures. A number of small photographs may be arranged to form one figure. The individual parts of such a composite illustration should be clearly marked (preferably by capital letters) and identified in the caption. All captions must be typed on a separate sheet or sheets. The maximum printed dimensions for figures will normally be 125 X 200 mm; larger formats will be considered only in exceptional circumstances. Figures must therefore be designed to yield clear images within the limits of a single page. Close attention should be paid to the matter of scale on figures. Where possible add a linear scale (with the dimension clearly marked) to the figure rather than trust that a statement of scale in the caption will be correct after the plate-maker and printer have finished their jobs. All line drawings should be in India ink on a suitable surface, such as Bristol board, tracing linen or plastic film. In general, however, the platemaker prefers to work from good quality, glossy photographs rather than originals of various sizes. Authors are urged to supply such photographic reproductions which, if made to a scale appropriate to the size of a printed page, will show whether ornament and lettering can be read in the final print. All photographs, whether for half-tone or line illustrations, should be high-contrast, glossy prints. Each illustration should be identified (author’s name, Fig. no. and orientation) in pencil on the back. Tables should be submitted in a clear format on separate sheets. Like illustrations, they should be designed to fit a single page of the journal. PROCEEDINGS of LINNEAN SOCIETY OF NEW SOUTH WALES VOLUME 106 Issued October 27, 1982 CONTENTS: | NUMBER 1 1 By JONSELE Linnaeus and his circumnavigating Apostles 21. B.A.FOSTER Two new intertidal balanoid Barnacles from eastern Australia 33. D.K.McALPINE and D.S. KENT Systematics of Tapeigaster (Diptera: Heleomyzidae) with Notes on Biology and larval Morphology 59 M.A.SCHNEIDER A comparative morphological Study of the reproductive Systems of some poses of Tapeigaster Macquart (Diptera: Heleomyzidae) 67 D.E.SHAW, B.K. CANTRELL and K. J. HOUSTON Neurochaeta inversa McAlpine (Diptera: Neurochaetidae) and Seed Set in Alocasia macrorrhiza (L.) G. Don (Araceae) in southeast Queensland 83. P.A. HUTCHINGS and H. F. RECHER The Fauna of Australian Mangroves 123 Notes and Discussion ~R.J. FORD A. W. H. Humphrey, His Majesty’s Mineralogist in New South Wales. 1803-12 = A Comment NUMBER 2 127 C.McA. POWELL, G. NEEF, D. CRANE, P. A. JELL and |. G. PERCIVAL Significance of Late Cambrian (Idamean) Fossils:in the Cupala Creek Formation, northwestern New South Wales 151. D.T. ANDERSON Origins and Relationships among the Animal Phyla 167 C.J.JENKINS Late Pridolian Graptolites from the Elmside Formation near Yass, New South Wales 173. C.J. JENKINS Darriwilian (Middle Ordovician) Graptolites from the Monaro Trough Sequence east of Braidwood, New South Wales 181 P. A. HUTCHINGS The Fauna of Australian Seagrass Beds LINNEAN SOCIETY OF NEW SOUTH WALES SHS — Record of the ANNUAL GENERAL MEETING 1981, Reports and Balance heets.°- Senne ee eee a een enna ten USN Tm Printed by Southwood Press Pty Limited, 80-92 Chapel Street, Marrickville 2204 PROCEEDINGS | of the | LINNEAN — SOCIETY NEW SOUTH WALES VOLUME 106 NUMBERS 3 &4 = NATURAL HISTORY IN ALL ITS BRANCHES THE LINNEAN SOCIETY OF NEW SOUTH WALES Founded 1874. Incorporated 1884. The Society exists to promote ‘the Cultivation and Study of the Science of Natural History in all its Branches’. It holds meetings and field excursions, offers annually a Linnean Macleay Fellowship for research, contributes to the stipend of the Linnean Macleay Lecturer in Micro- biology at the University of Sydney, and publishes the Proceedings. Meetings include that for the Sir William Macleay Memorial Lecture, delivered biennially by a person eminent in some branch of Natural Science. The _ Society’s extensive library is housed at the Science Centre . n Sydney. Membership enquiries should be addressed in the first instance to the Secretary. Candidates for election to the Society must be recommended by two members. The present annual subscription is $20.00. The current rate of subscription to the Proceedings for non-members is set at $35.00 per volume. Back issues of all but a few volumes and parts of the Proceedings are available for purchase. A price list will be supplied on epplication to the aia pes s a & ag Bead — Nee a 4. Se apes OFFICERS AND COUNCIL 1981-82 _ President: HELENE A. MARTIN Vice-Presidents: LYNETTE A. MOFFAT!, A. RITCHIE, F. W. E. ROWE, J. T. WATERHOUSE Honorary Treasurer: A. RITCHIE Secretary: BARBARA STODDARD Council: D. A.~ ADAMSON, M. ARCHER, L. W. CC. FILEWOOD, cy Auxs: JOHNSON, HELENE A. MARTIN, P. M. MARTIN, LYNETTE A. MOFFAT! P. MYERSCOUGH, G. PHIPPS, A. RITCHIE, A. N. RODD, F. W. E. ROWE, C. N. SMITHERS, T. G. VALLANCE, J. T. WATERHOUSE, B. D. WEBBY, A.J. T. WRIGHT Honorary Editor: T. G. VALLANCE — Department of Geology & Geophysics, University of Sydney, Australia, 2006. Librarian: PAULINE G. MILLS aE Linnean Macleay Fellow: Linnean Macleay Lecturer in Microbiology: K.-Y. CHO Auditors: W. SINCLAIR & Co. The office of the Society is in the Science Centre, 35-43 Clarence Street, Sydney, N.S.W., Australia, 2000. Telephone (02) 290 1612. © Linnean Society of New South Wales ' resigned 20 May 1981 Cover motif: Forewing of Permorapisma biserialis (Neuroptera: Permithonidae), from the Permian of Belmont, New South Wales Adapted by Len Hay from Proc. Linn. Soc. N.S.W. 51, 1926, p. 278 (Fig. 15) un en PROCEEDINGS of the LINNEAN SOCIETY NEW SOUTH WALES e : PRR pn ee Marine Biological Laboratory LIBRARY AUG 15 1983 Woods Mole, Mass. VOLUME 106 NUMBER 3 Stability, Depletion and Restoration of Seagrass Beds* ’ A. W. D. LARKUM and R. J. WEST LARKUM, A. W. D., & WEST, R. J. Stability, depletion and restoration of seagrass beds. Proc. Linn. Soc. N.S.W. 106 (3), (1982) 1983: 201-212. The hypothesis that seagrasses are well adapted to the environment in which they live is questioned. Evidence for instability in seagrass beds is documented. Good evidence for depletion of seagrass beds from natural causes exists. In addition it is clear that man-induced depletion has been very severe in many populated coastal areas. The causes of instability in seagrass beds need to be understood both as regards the prevention of further depletion and the restoration of seagrasses in depleted areas. Several techniques have been used successfully for the restoration of seagrasses in small scale experiments but these are expensive and labour-intensive. A. W. D. Larkum, School of Biological Sciences, University of Sydney, Sydney, Australia 2006, and R. J. West, N.S.W. State Fisheries, 211 Kent Street, Sydney, Australia 2000 (formerly School of Biological Sciences, University of Sydney); manuscript received 19 May 1981, accepted for publication in revised form 16 December 1981. INTRODUCTION Seagrasses are angiosperms with a rooted rhizome system which limits their distribution generally to unconsolidated sedimentary substrates in relatively calm waters (Thalassodendron pychyrizum which grows on rocky reefs in Western Australia (Cambridge, 1975) is an exception to this rule). This distribution, in turn, has a special significance for both depletion and restoration, for it means that the favoured sites for seagrasses — estuaries and sheltered embayments — are also the favoured sites for human settlement and activity. As a result there has in the past been much depletion of seagrass beds (Phillips, 1978). However it should be pointed out that it is only recently that man has had any real impact on seagrasses and it is really only the current life styles of world civilizations that are inimical to co-existence at such sites of seagrasses and people. With the growing awareness of the biology of seagrasses has also come an awareness of the economic importance of seagrasses in a variety of ways (see e.g. Thayer et al., 1976; Phillips, 1978). And out of this have come a number of attempts to restore seagrasses to depleted areas (see Table 1). However, perhaps the best place to start is on the subject of the stability of seagrass ecosystems in their natural state, since our ideas on the ability of seagrass communities to withstand man-induced impacts depend on this crucial point. STABILITY OF SEAGRASS BEDS Introduction Seagrasses it has been claimed are well adapted to the environment in which they live (e.g. den Hartog, 1979). The supporting evidence for this statement is not com- pletely convincing. Resolution of this matter is very important since on it rests the degree of stability which may be expected from ‘natural’ seagrass beds and on this in turn rest any assumptions that may be made as to the cause of depletions of beds subjected to man-induced impacts. * A paper presented at the Australian Seagrass Workshop, University of Sydney, 1981. Proc. LINN. Soc. N.S.W. 106 (3), (1982) 1983 202 STABILITY, DEPLETION AND RESTORATION OF SEAGRASS BEDS Seagrasses are certainly adapted to their environment otherwise they would not be there. The real question to ask is ‘are they well adapted?’ This is a difficult question to answer. It could be argued that the lack of speciation within seagrasses is a sign of poor fitness. Seagrasses probably evolved early in the evolution of angiosperms; but whereas there are approximately 300,000 species of flowering plants there are only 50 odd species of seagrasses. The counter-argument would be that seagrasses are so well adapted to their environment that further adaptation and speciation has been un- necessary. However this sort of argument is unsatisfactory in the light of the vast speciation in terrestrial grasses, freshwater macrophytes and reef fish (Ehrlich, 1975) to take but a few examples. Furthermore the recent documentation of speciation in Posidonia (Cambridge and Kuo, 1979) in south western Australia indicates that speciation has taken place in an area which appears to be ideal for the growth of seagrasses. A number of inherent weaknesses may exist in the strategy of a submersed marine angiosperm. These weaknesses include: — 1) sexual propagation 2) vegetative propagation 3) occlusion of leaves by epiphytes 4) internal aeration 5) salinity control and turgor regulation 6) susceptibility to microbial attack 7) requirement for moderately high light environments. At present there are few hard facts about any of these in relation to the ‘fitness’ of seagrasses. Johnston (1979) has made the interesting suggestion that the high leaf productivity of seagrasses in comparison with other plants is a result of the need to provide new photosynthetic tissue as the older parts of the leaves become overgrown by epiphytes and epizoans. Despite the fact that seagrass leaf productivity is not ex- ceptional (terrestrial tropical grasses such as sugar cane and napier grass have a productivity several fold higher — see Boardman and Larkum, 1974), the suggestion has some merit and if it is true leads to the constraint that seagrasses have less energy to devote to sexual and vegetative propagative mechanisms. As regards salinity control, Tyerman (1979) has shown that the leaf meristem tissue is hypotonic to seawater and further experiments (unpublished) have shown that exposure of the meristem to seawater leads to its death. Thus mechanical damage of the sheath due to wave action, disease, or animal attack (and of course human activities such as bottom trawls or army ‘ducks’) can have serious consequences for seagrasses. Internal aeration does not apparently pose any problems for seagrasses; freshwater macrophytes which in terms of speciation are much more successful than seagrasses seem to cope with this con- straint. Nevertheless the necessity of having rhizomes and roots in an anaerobic en- vironment must place constraints on the growth and productivity of such a plant and must make it more susceptible to microbial attack. Presumably the high concentration of phenolics and other aromatic compounds in the roots and rhizomes of seagrasses (Cariello and Zanetti, 1979) is an antibacterial response. Evidence for instability in seagrass beds The best example of instability in seagrass beds is the ‘wasting disease’ of Zostera marina which caused a near catastrophic decline in stocks of this seagrass in the early 1930s. This example is dealt with further in the next section. However in summary it appears that a slight climatic rise in temperature may have set off the demise (Rasmussen, 1977). Apart from this example there is a great deal of evidence to support recent cyclical changes in stocks of Zostera marina (Orth, 1976; den Hartog Proc. LINN. Soc. N.S.W. 106 (3), (1982) 1983 A.W.D. LARKUMANDR. J. WEST 203 1979). In fact McRoy and Bridges (1974) suggested that the ‘wasting-disease’ of the 1930s was an extreme cycle of a rhythmic oscillation. Evidence for cyclical changes in other seagrasses is not available but this probably reflects the absence of investigations rather than the absence of the phenomenon. Further evidence for instability can be found in a few studies of various transient phenomena, usually induced by wave-action. Patriquin (1975) has described the ‘migration’ of blowouts in beds of Thalassia testudinum in the West Indies. Storm- generated, crescentic holes in the beds were observed to move shorewards at the rate of ca 0.5 cm/day. It was estimated that any one point would be ‘recurrently eroded and restabilized at intervals of the order of 5-15 years’. Syringodium filiforme was involved in the recolonization of the blowouts on the leeward slope. Similar blowouts have been described in Western Australia (Cambridge, 1975) in P. australis beds (there Amphibolis antarctica is involved in the recolonization). In both examples the bed is in a dynamic equilibrium i.e. the seagrass system is able to cope with the factor causing the in- stability. However it does indicate a rather delicate balance in the system: in both examples the dominant seagrass is absent or very much depleted in more exposed sites. Thus as pointed out by den Hartog (1979) we generally see today the end result of past Fig. 1. Rings in Zostera capricorni bed in Botany Bay. Scale bar represents 20 m. Proc. LINN. Soc. N.S.W. 106 (3), (1982) 1983 204 STABILITY, DEPLETION AND RESTORATION OF SEAGRASS BEDS evolutionary and successional events and assume that the systems we see are stable. This does not mean however that such events are not going on but they will un- doubtedly be more difficult to detect. Recently we have detected what appears to be a cyclical event in the growth of Zostera capricornt (Larkum et al., in prep.). Rings of Z. capricornt were observed by means of high-resolution aerial photographs of the northern shore of Botany Bay (Fig. 1). The rings occurred in a shallow region which previously had been an area of largely bare, shifting sand. However after the establishment of the armoured wall for Port Botany, the area became stabilized and wave action was almost eliminated. Under these conditions Z. capricorni recolonized most of the area within a few years. It was under these conditions that the rings developed. As can be seen the middle of the rings was bare with only loose sand. Inspection of the rings revealed no apparent cause for the phenomenon. A possible cause for the die-back in the centres could be the depletion of an essential mineral. However this seems unlikely. Another explanation could be that apical meristems, which are critical for seagrass production (Tomlinson, 1974), grow initially in an outward radial direction; eventually the oldest shoots on the rhizomes die and cannot be replaced unless seedlings establish in the centre region or unless one ring grows into another. If this explanation is correct then it will be a further example of the ‘apical meristem dependence’ of seagrasses and illustrates another possible weakness in seagrass systems. It may well be that the inability of seagrasses to generate secondary apical meristems results in the slow regrowth of eroded areas (West, 1980). The causes of natural depletion of Seagrass Beds Research into the natural depletion of seagrass beds has not been extensive in the past and the cause(s) of depletion in any one instance has never been fully established. A number of possible causes have been put forward including disease (Rasmussen, 1977), climatic changes (Rasmussen, 1977; Orth, 1976), natural cycles (McRoy and Bridges, 1974) sediment movement (Kirkman, 1978), salinity changes (Orth, 1976), sea level changes (den Hartog, 1977) and faunal influences (Orth, 1976; Ferguson Wood, 1959). The most famous example of natural depletion of seagrasses is the ‘wasting disease’ of Zostera marina which resulted in a nearly catastrophic decline in the stocks of this seagrass on both sides of the Atlantic in the early 1930s. Although it is generally agreed that dead plants were attacked by a fungus (Ophzobolus halimus) and a slime mold (Labyrinthula macrocystis) Rasmussen (1977) has put forward good arguments to suggest that the real cause of the ‘disease’ was a rise in sea level temperatures in the early 1930s and that the weakened plants were then susceptible to attack by the fungal and slime mold pathogens. Rasmussen (1977) also suggested that seed germination was inhibited at the higher temperatures and that plants growing in brackish water were immune to the ‘disease’. Orth (1976) documented a 70% decline in Zostera marina in Chesapeake Bay (U.S.A.) from 1972 to 1974 and a virtual disappearance in 1975. This period was accompanied by a rise in average temperatures and a marked rise in water tem- peratures (Orth, 1976, fig. 7). If the arguments of Rasmussen are accepted then the growth of Z. marina must be seen as a finely balanced process in which temperature, salinity and susceptibility to pathogen attack are finely balanced. There are two examples of apparently natural depletion of seagrasses in Australia. One is at Corner Inlet, Victoria, which is the only site in continental Victoria where Postdonia australis occurs. Large beds have existed there in the past, as evidenced by fibre remains, but in recent years the extent of the beds has decreased (Poore, 1978). Proc. LINN. Soc. N.S.W. 106 (3), (1982) 1983 A.W. D.LARKUM ANDR. J. WEST 205 Sedimentation in the region has caused a general rise of the seabed in the region and this in turn may have altered the salinity regime of the area (P. australis is a strictly stenoholine species). However the exact cause of the decline is not known. One might more profitably enquire as to why P. australis is absent from the remainder of the coast of Victoria but occurs at Eden in N.S.W. and after Port MacDonnel in South Australia. The second example is the depletion of Zostera capricorni in northern areas of Moreton Bay (Kirkman, 1978). This depletion seems to have occurred as a result of sand inundation at a rate faster than the upward growth of the plants. At the same time adjacent healthy beds were subject to increased grazing which contributed to the overall depletion. TABLE 1 Man-induced impacts affecting seagrass communities Impact Seagrass community Place Effect Reference Turbidity Posidonia oceanica S. France Large-scale Meinesz & associated depletion Laurent, 1976 with at lower eutroplication limit Posidonza australis Cockburn Large-scale McComb ef al. , Sound, W.A. depletion 1981 Posidonia australis Botany Bay Loss of Larkum, 1976b. deeper beds Thalassia testudinum Florida Large-scale Taylor et al. , depletion 1973 Thalassia testudinum Virgin Local Dong et al. , Islands depletions 1972 Nichols e¢ al. , 1972 Zostera capricorni/ Tuggerah Loss of Higginson, 1971 Ruppia spiralis Lakes, NSW deeper beds Turbidity Thalassia testuainum Virgin Depletions Van Eepoel et al. , associated Islands, 1971 with West Indies dredging Thalassia testudinum St. Thomas, Depletions Grigg, 1970 West Indies Hydraulic Thalassia testudinum/ Tampa and Extensive Godcharles, clam dredge Halodule beaudette Tarpen losses 1971 Syringodium filiforme Springs, Florida Dredging and Thalassia testudinum Boca Ciega Local Taylor & filling Bay, Florida depletions Saloman, 1968 Thalassia testudinum Redfish Bay, Local Odum, 1963 Texas depletions Sewage Thalassia testudinum Biscayne Bay, Local McNulty, 1970 effluent Florida depletions Thalassia testudinum Florida Local Hammer, 1972 depletions Posidonia oceanica S. France Large-scale Peres & depletions Pickard, 1974 Posidonta australis Adelaide Severe local Shepherd, 1970 S.A. depletion Proc. LINN. Soc. N.S.W. 106 (3), (1982) 1983 206 STABILITY, DEPLETION AND RESTORATION OF SEAGRASS BEDS Hot water Thalassia testudinum Florida Large Zieman, 1976 effluents depletion Florida Large Thorhaug et al. , depletion 1978 Salinity Thalassia testudinum Florida Loss of McMillan & changes Halodule beaudetter leaves and Moseley, 1967 Syringodium filtforme some depletion Thalassia testudinum Florida Die-back Zieman, 1975 but recovery Overgrazing Thalassia testudinum Florida Local Camp et al. , by Lytechinus depletion 1972 variegatus Oil pollution Thalassia testudinum Puerto Local Diaz-Piferrer, Rico losses 1962 Phyllospadix torreyt Santa Local Foster et al., Barbara depletion 1971 Physical Thalassia testudinum Florida Small-scale Zieman, 1976 disturbance losses but by boats spreading Industrial Posidonia australis Cockburn Local Cambridge, 1975 effluents Sound losses MAN-INDUCED DEPLETION OF SEAGRASS BEDS There are a great number of examples of man-induced depletions of seagrass beds (Table 1) and there are a great number of possible or known causes amongst which are the following: — 1) turbidity increases associated with dredging, 2) turbidity increases associated with industrial or urban influences, 3) turbidity increases associated with eutrophication, 4) toxic chemicals, 5) hot water effluents, 6) oil spills, 7) activities of commercial fishermen using bottom trawls, 8) changes in salinity, 9) sewage. Much of the evidence on these factors comes from overseas studies as can be seen from Table 1, and much of it is less than fully convincing. It relies on field observations and circumstantial evidence. Many of the sites where depletion occurred were subject to multiple impacts and it may not be justified to single out one factor. Clearly careful experiments with proper controls are needed. In Australia there are a number of published reports on man-induced depletion of seagrass beds (Shepherd, 1970; Higginson, 1971; Cambridge, 1975; Larkum, 1976 a, b; Shepherd and Van der Borch, 1977). In no case, so far has the cause been established categorically except for the mechanical removal of P. australis beds in the Spencer Gulf, South Australia. In fact in all of the other examples multiple impacts existed. The most clear-cut case would seem to be the disappearance of P. australis near sewage outfalls off the coast of Adelaide (Shepherd, 1970) yet even this has been disputed (Steffensen, unpublished). It would seem that transplant experiments with Proc. LINN. Soc. N.S.W. 106 (3), (1982) 1983 A.W.D.LARKUM ANDR. J. WEST 207 proper controls should be able to provide a definitive answer. In Cockburn Sound (McComb etal., 1981), in Botany Bay (Larkum, 1976a, 1976b and unpublished) and in the Tuggerah Lakes increased turbidity would seem to be the most important factor. In Botany Bay wave-action is also a clear-cut factor at some sites (Larkum, 1976b). However at all these places a number of impacts are present. In the Tuggerah Lakes some depletion is almost certainly due to hot-water effluents from electric power stations (Weiner, personal communication). Poorly documented evidence exists also for the following: 1) depletion due to oil and dispersants in Botany Bay and, ii) depletion due to hot water effluent in Botany Bay (Bunnerong Power Station). Depletion of seagrasses has almost certainly occurred in Port Phillip, Sydney Method Plugs Plugs Plugs Plug Plugs Plugs Plugs & Turfs Turfs Turfs Turfs Turions Turions Turions Seeds Seeds Seeds Worker and Publication Date Kelly et al. (1971) Breedveld (1976) Breedveld (1975) Kirkman (1976) Phillips (1974) Phillips (1974) Phillips (1980) Ranwell et al. (1974) Ranwell et al. (1974) Larkum (1976) Kelly et al. (1971) Eleuterius (1975) Phillips (1974) Phillips (1974) Thorhaug (1974) Thorhaug & Hixon (1975) TABLE 2 Transplantation of Seagrasses by Various Workers using Various Techniques (Adapted from Thorhaug and Austin, 1976) NAA = Naphthylacetic acid Place Anchoring Method Tampa Bay, Florida Tampa, Florida Tampa, Florida Moreton Bay, Australia Tampa Bay, Florida Tampa Bay, Florida Redfish Bay, Texas Norfolk, England Norfolk, England Botany Bay, Australia Tampa Bay, Florida Biloxi, Mississippi Whidbey Isle, Washington Friday Harbor, Washington South Biscayne Bay, Florida North Biscayne Bay, Florida Tin can or Bag 30 cm substrate into hole 30 cm substrate into hole Dug holes Buried with soil Buried with soil Buried with soil Spaded into hole Spaded into hole Dug holes 1. Construction rods 2. Brick pipes Wire-mesh anchor and Construction rod Iron pipes and trenches Iron rods and trenches Plastic Plastic peat- pots Chemical Dimensions of Success Additive Transplant NAA 20 cm? Control 40% Exper. 15% 5% NAA 30cm 0-100% 5% Root Deep post-hole dip digger 5% NAA 30cm 100% 5% Root Deep post-hole dip digger None None 10 cm? Some None 10cm? None None 60 cm” 80-100% None ca22 x 15 Varied x 10cm deep None ca22 x 10 100% year 1 x 15cm deep 35% year 2 None 50 x 50cm 50-100 % 10% NAA Short shoot 18% in some and rhizome None 45 x 45cm 3% 5% None None Short shoot Dependent and rhizome on depth portion 100% to none None Single seeds None 10% NAA Single seeds 80% 10% NAA Single 15%-55% Species Thalassia testudinum Thalassia testudinum Syringodium filiforme Zostera capricornt Halodule wright Thalassia testudinum T. testudinum A. wright - Zostera noltir Zostera nolti Z. marina Zostera capricornt & Posidonia australis Thalassia testudinum T. testudinum H. wright Cymodocea manatorum Zostera marina Zostera marina Thalassia testudinum Thalassia testudinum Proc. LINN. Soc. N.S.W. 106 (3), (1982) 1983 208 STABILITY, DEPLETION AND RESTORATION OF SEAGRASS BEDS Harbour (and the Parramatta River) and the Brisbane River, but good documentation of these events and their causes is lacking. It is worth noting in passing that there has been no published work anywhere in the world on the effect of toxic wastes such as heavy metals and chlorinated hydrocarbons on seagrasses or seagrass communities. Considering the economic importance attached to seagrass beds (see e.g. Thayer et al., 1976; and Phillips, 1978) it is surprising that more public money is not being spent on investigating the causes of the various depletions of seagrasses in Australia. The lack of funding can partly be seen as a result of ignorance and the great expanse of completely untouched seagrass beds in southern and western Australia. RESTORATION OF SEAGRASS BEDS Introduction Restoration of seagrass beds does occur naturally at times, as for instance after the ‘wasting disease’ of the 1930s. However artificial restoration is an important development that has come about in recent years mainly as a result of man-induced depletions of seagrass beds. For instance in Florida (U.S.A.) real estate development involving dredging of Thalassia testudinum meadows resulted in large scale depletion of this seagrass in the 1960s and had the Florida State Board of Conservation setting up a programme to restore this seagrass to several areas (Phillips, 1974). Also in Florida, hot water effluents led to damage of Thalassia beds and restoration by transplanting was effected (Thorhaug and Roessler, 1976), In fact the history of transplants goes all the way back to the ‘wasting disease’ period of the 1930s when rather ineffectual attempts were made to revegetate areas with Zostera marina (see Phillips, 1974). Table 2 lists all published attempts to date to transplant seagrasses. Techniques a) Vegetative Two basic techniques exist. They are: — al) transplantation of seagrass plants without accompanying sediment. The plants (including rhizomes and several shoots) are replanted in sediment of the new site (Phillips, 1974; Larkum, 1976a), a2) transplantation of seagrass turfs, i.e. plants are removed zn situ in un- disturbed sediment placed into trays or bags and relocated into excavated holes at the new site (Ranwell et al., 1974; Fuss and Kelly, 1969; Breed- veld, 1975). Anchoring may be a problem. Dislodgement may occur as a result of storms or through the activities of fishermen or prawners. A number of devices have been tried. Kelly e al. (1971) mention cans, sacking polyethylene, pipes, construction rods and breeze bricks. To this list can be added steel mesh for reinforcing concrete (Larkum, 1976a) and plastic netting (Ranwell et al., 1974). b) Seedlings. This method has been pioneered with success by Thorhaug (1974, 1976, 1979). Results and Discussion Transplanting of seagrasses by the vegetative method has in most cases met with reasonably good initial success. A great deal depends on the species of seagrass. Pioneer and rapidly growing species may quickly spread out and cover a large area at the new site, e.g. Zostera nolti: (Ranwell et al., 1974), Zostera marina (Phillips, 1974). In this way an area may be ‘seeded’ with the minimum of work and money. On ihe other hand with ‘climax’ species such as Thalassia testudinum and Posidonia spp. growth rate Proc. LINN. Soc. N.S.W. 106 (3), (1982) 1983 A.W.D. LARKUM ANDR. J. WEST 209 may be much slower and replanting might mean the ‘sod by sod’ coverage described by Thorhaug (1979). In fact Kelly et a/. (1971) note that regression may occur at trans- plant sites because of death of older shoots and lack of replacement. This results from the ‘meristem dependence’ of seagrass populations (Tomlinson, 1974). Few seagrasses have dormant meristems and horizontal vegetative spreading depends on the presence of an apical meristem. In an established seagrass bed there may be many shoots but few apical meristems. Thus in vegetative transplants no apical meristem may be present and expansion growth cannot then occur. Apical meristems can generate from older (short) shoots but this is not common. Seedlings generate an apical meristem directly and are thus ideal for transplant work providing that large quantities of seeds can be obtained and germinated and provided that these can be sown effectively. With Thalassia testudinum this can be done easily and economically (Thorhaug and Austin, 1976). This species also has the ad- vantage that seedlings grow rapidly producing a ‘long’ shoot and several ‘short’ shoots within six months. With Poszdonia australis seedling growth is much slower and production of an apical meristem (within this species the second shoot) does not occur for at least a year (McComb et al., 1981; West, 1980). Austin in Thorhaug and Austin (1976) has costed the operation of seagrass transplants and for Thalassia seedling transplants this works out to be about $U.S. 13,000 per ha. Ranwell et a/. (1974) costed their vegetative process at £1,000 per ha. (1973 prices). It would therefore appear that restoration of at least some depleted sites is possible and not outrageously expensive. However in all the successful experiments the sites have not been subject to stresses or impacts (for example the hot water discharge involved in the area of depletion used by Thorhaug (1974) had been discontinued before the transplant experiments were conducted). In areas where man- induced impacts occur and where the most depletions have resulted restoration of seagrass beds may be much more difficult, at least until the major causal factor has been identified and controlled. Nevertheless such situations do provide the opportunity to study the effects of a variety of impacts on seagrasses. CONCLUSION Australia is richly endowed with seagrasses both in terms of species and in terms of the extent of seagrass beds. This affords many opportunities for the study of stability/instability, evolutionary changes, successional changes and natural depletions, all of which are in great need of much further research. Seagrass com- munities may be less stable than they were previously held to be. The overall wealth of Australian seagrasses may suggest that local depletion of seagrass beds near to several state capitals is unimportant. This may be so but until we know much more about the stability of natural seagrass systems, natural depletions, man-induced depletions and the role of seagrass beds in the natural economy of coastal waters there needs to be a note of caution in any acceptance of local depradations and a positive attitude to the possibility of restoration of seagrass beds. In any event there will probably always exist a need for the restoration of seagrass beds as a result of natural depletions and activities such as dredging, harbour development and offshore mining. References Appy, C. E., 1947 — Eelgrass planting guide. Maryland Conserv. 24: 16-17. BEER, S., and WAISEL, Y., 1979 — Some photosynthetic carbon fixation properties of seagrasses. Aquatic Botany 7: 129-138. BOARDMAN, K., and LAaRKuM, A. W. D., 1974 — Biological conversion of solar energy. In MESSEL, H., and BUTLER, S. T., (eds) Solar Energy 125-181. Sydney: Shakespeare Head Press. Proc. LINN. Soc. N.S.W. 106 (3), (1982) 1983 210 STABILITY, DEPLETION AND RESTORATION OF SEAGRASS BEDS BREEDVELD, J. F. VAN., 1975 — Transplanting of seagrasses with emphasis on the importance of substrate. Florida Marine Research Publications 17: 1-22. Ilustr. CAMBRIDGE, M. L., 1975 — Seagrasses of south-western Australia with special reference to the ecology of Posidonia australis Hook F ‘n a polluted environment. Aquatic Botany 1: 149-161. , and Kuo, J., — Two new species of seagrasses from Australia, Poszdonia sinuosa and P. angustifolia (Posidoniaceae). Submitted to Aquatic Botany. Camp, D. K., Coss, S. P., and VAN BREEDVELD, J. F., 1973 — Overgrazing of seagrasses by a regular urchin Lytechinus variegatus. Bioscience 23: 37-38. CARIELLO, L., and ZANETTI, L., 1979 — Distribution of chicoric acid during leaf development of Poszdonza oceanica. Botanica Marina X XII: 359-360. DIAZ-PIFERRER, M., 1962 — The effects of oil on the shore of Guanica, Puerto Rico. Assoc. Island. Mar. Lab. 4th Meeting Curacao. pp 12-13. Abstract Deep-Sea Res. 11: 855-856. DonG, M., ROSENFELD, J., REDMANN, G., ELLIOT, M., BALAZy, J., POOLE, B., RONNHOLM, K., KENISBERG, D., NOVAL, P., CUNNINGHAM, C., and KARNOw, C., 1972 — The role of man-induced stresses in the ecology of Long Reef and Christiansted Harbour, St. Croix, U.S. Virgin Islands. Spec. Publ. W. Indies Laboratory. St. Croix: Fairleigh Dickinson University. Drew, E. A., 1979 — Physiological aspects of primary production in seagrasses. Aquatic Botany 7: 139-150. DRYSDALE, F. R., and BARBOUR, M. G., 1975 — Responses of the marine angiosperm Phyllospadix torreyi to certain environmental variables: A preliminary study. Aquatic Botany 1: 211-247. EHRLICH, P. R., 1975 — The population biology of coral reef fishes. Ann. Rev. Ecol. Syst. 6: 211-247. ELEUTERIUS, L. N., 1975 — Submergent vegetation for bottom stabilization. Estuarine Research, Vol. II: 439-456. New York: Academic Press. : FELLER, R. J., TAGHON, G. L., GALLAGHER, E. D., KENNY, G. E., and JuMARS, P. A., 1979 — Im- munological methods for food web analysis in a soft-bottom benthic community. Marine Biology 54: 61-74. FERGUSON WOOD, E. J., 1959 — Some east Australian seagrass communities. Proc. Linn. Soc. N.S. W. 84: 218-226. FOSTER, M., NEUSHUL, N., and ZINGMARK, R. — The Santa Barbara Oil Spill. Part 2. Initial effects on intertidal and kelp bed organisms. Environ. Pollut. 2: 115-134. Fuss, G. M. (Jr.), and KELLY, J. A. (Jr.)., 1969 — Survival and growth of seagrasses transplanted under artificial conditions. Bull. Mar. Sc1. 19: 351-365. GODCHARLES, M. F., 1971 — A study of the effects of a commercial hydraulic clam dredge on benthic communities in estuarine areas. Florida Dep. Nat. Res., Mar. Res. Lab., Tech. Ser. No. 64: 1-55. GRIGG, D. I., 1970 — Some effects of dredging on water quality and coral reef ecology. Caribb. Cons. Assoc. Newsl. 1: 22-27. HAMMER, L., 1972 — Anaerobiosis in marine algae and marine phanerograms. Proc. 7th Int. Seaweed Symp. Sapporo: 414-419. DEN HarToa, C., 1979 — Seagrasses and seagrass ecosystems: an appraisal of the research approach. Aquatic Botany 7: 105-117. HIGGINSON, F. R., 1971 — Ecological effects of pollution in Tuggerah Lakes. Proc. Ecol. Soc. Aust. 5: 143- 152. HouGu, R. A., and WETZEL, R. G., 1977 — Photosynthetic pathways of some aquatic plants. Aquatic Botany 3: 297-313. JOHNSTONE, I. M., 1979 — Papua New Guinea seagrasses and aspects of the biology and growth of Enhalus acoroides (L.f.) Royle. Aquatic Botany 7: 197-208. KELLY, J. A. (Jr.)., Fuss, C. M. (Jr.), and HALL, J. R., 1971 — The transplanting and survival of Turtle Grass Thalassia testudinum, in Boca Ciega Bay, Florida. Fishery Bull. 69: 273-280. KIRKMAN, H., 1977 — Growing Zostera capricornt Aschers. in tanks. Aquatic Botany 4: 367-372. , 1978 — Decline of seagrass in northern areas of Moreton Bay, Queensland. Aquatic Botany 5: 63-76. LARKUM, A. W. D., 1976a — The ecology of Botany Bay. I. The growth of Posidonza australis (Brona) Hook f. in Botany Bay and other bays of the Sydney Basin. Aust. J. Mar. Freshw. Res. 27: 117-127. ——, 1976b — Botany Bay: National asset or national disaster. Operculum 5: 67-75. ——,, COLLETT, L. C., and WILLIAMS, R. J. (Jr.)— Ecology of Botany Bay. III: The Standing Stocks, Growth, and Shoot pproduction of Zostera capricorn in Botany Bay, N.S.W. In preparation. McComgs, A. J., CAMBRIDGE, M. C., KIRKMAN, H., and Kuo, J., 1981 — The biology of Australian seagrasses. In (pp. 258-293) PATE, J. S., and MCComgB, A. J., (eds), The Biology of Australian Plants. Nedlands: Univ. of Western Australia Press. MCMILLAN, C,, and MOSELEY, F. N., 1967 — Salinity tolerances of five marine spermatophytes in Redfish Bay, Texas. Ecology 48: 503-506. , and PHILLIPS, R. C., 1979. Differentiation in habitat response among populations of New World seagrasses. Aquatic Botany 7: 185-196. MCNULTY, J. K., 1970 — Effects of abatement of domestic sewage pollution on the benthos volumes of PROC. LINN. Soc. N.S.W. 106 (3), (1982) 1983 A.W.D. LARKUM ANDR. J. WEST 211 zooplankton and the fouling organisms of Biscayne Bay, Florida. Stud. Trop. Oceanogr. 9: 107. McRoy, C. P., and BRIDGES, K., 1974 — Dynamics of seagrass ecosystems. Proc. Ist Int. Congr. Ecol. The Hague, Sept 1974: 374-375. MEINESZ, A., and LAURENT, R., 1978 — Cartographie et etat de la limite inferieure de |’herbier de Postdonia oceanica dans les Alpes-maritimes (France). Botanica Marina. X XI: 513-526. Mukal, H., Alot, K., Koike, I., Izumi, H., OHTsu, N., and Hairort, A., 1979 — Growth and organic production of eelgrass (Zostera marina L.) in temperate waters of the Pacific coast of Japan. In Growth analysis in spring-summer. Aquatic Botany 7: 47-56. NICHOLS, M. M., VAN EEPOEL, R., GRIGG, D., BRODY, R., SALLENGER, A., OLLMON, J., and CRENA, R., 1972 — Environment, water and sediments of Christiansted Harbour, St. Croix. Division of En- vironmental Health, Government of the Virgin Islands. Rept. No. 16. OpuM, W. E., ZIEMAN, J. C., and HEALD, E. J., 1972 — The importance of vascular plant detritus to estuaries. Proc. of the Coastal Marsh and Estuary Management Symposium, Baton Rouge, Louisiana. ORTH, R., 1976 —The demise and recovery of eelgrass Zostera marina, in the Chesapeake Bay, Virginia. Aquatic Botany 2: 141-159. PATRIQUIN, D., 1973 — Estimation of growth rate, production and age of the marine angiosperm Thalassia testudinum Konig. Carb. J. Sci. 13: 111-123. ——, 1975 — ‘Migration’ of blowouts in seagrass beds at Barbados and Carriacon, West Indies, and its ecological and geological implications. Aquatic Botany 1: 163-189. PERES, J.M., and PICKARD, J., 1975 — Causes de la rarefaction et de la disparition des herbiers de Posidonia oceanica sur les ctes Frangaises de la Mediterranée. Aquatic Botany 1: 133-139. PHILLIPS, R. C., 1978 — Seagrasses and the coastal marine environment. Oceanus Summer, 1978: 30-40. , 1960 — Observations on the ecology and distribution of the Florida seagrasses. Prof. Pap. Serv. Florida Board Conserv. 2: 1-72. ——, 1974 — Transplantation of seagrasses, with special emphasis on eelgrass, Zostera marina L. Aquaculture 1: 161-176. ——., 1980 — Response of transplanted and indigenous Thalassia testudinum. Banks ex Konig and Halodule wrighti: Aschers to sediment loading and cold stress. Contrib. Mar. Sci. 23: 79-87. ——., VINCENT, M. K., and HUFFMAN, R. T., 1978 — Habitat development field investigations, Port St. Joe, Seagrass Demonstration site Port St. Joe, Florida. Summary Report, D-78-33, 1978. Depart- ment of the Army (U.S.A.), Waterways Expt. St. Corps of Engineers Vicksburg, Mississippi. POoORE, R., 1978 — Report on depletion of Posidonia australis. Marine Studies Group, Dept. of Con- servation. RANWELL, D. S., WYER, D. W., BOORMAN, L. A., PIZZEY, J. M., and WATERS, R. J., 1974 — Zostera transplants in Norfolk and Suffolk, Great Britain. Aquaculture 4: 185-198. RASMUSSEN, E., 1977 — The wasting disease of eelgrass (Zostera marina) and its effects on environmental factors and fauna. In Seagrass ecosystems — a scientific perspective (ed. C. P. MCROy): 1-51. New York: Dekker. SHEPHERD, S. A., 1970 — Preliminary report upon degradation of seagrass beds at North Glenelg. Repts S. Austr. Dept of Fisheries 33: 16-19. , and VON DER BoORCH, C. C., 1977 — Environmental effects of mining marine fibre. SAFIC (Magazine of the Fisheries Branch, S.A.,and S. Aust. Fishing Ind. Council), May 1977, no. 12: 13- 15. TAYLOR, J. L., and SALOMAN, C. H., 1968 — Some effects of hydraulic dredging and coastal development in Boca Ciega Bay, Florida. Fish. Bull. 67: 213-241. , and PREST, K. W. (Jr.), 1973 — Harvest and regrowth of turtle grass (Thalassva testudinum) in Tampa Bay, Florida. Fish. Bull. 71: 145-148. THAYER, G. W., ADAMS, S. M., and LACRorx, M. W., 1975 — Structural and functional aspects of a recently established Zostera marina community. In CRONIN, L. E., (ed.), Estuarine Research I: 518-540. New York: Academic Press. ——, ENGEL, D. W., and LACRoIx, M. W., 1977 — Seasonal distribution and changes in the nutritive quality of living, dead and detrital franctions of Zostera marina L. J. exp. mar. Biol. Ecol. 30: 109-127. ——, WOLFE, D. A., and WILLIAMS, R. B., 1976 — The impact of Man on seagrass systems. American Scientist 63: 288-296. THORHAUG, A., 1974 — Transplantation of the seagrass Thalassia testudinum KOnig. Aquaculture 4: 177-183. , 1979 — The flowering and fruiting of restored Thalassia beds. Aquatic Botany 6: 189-192. , and AusTIN, C. B., 1976 — Restoration of seagrasses with economic analysis. Environmental Con- servation 3: 259-267. ——., BLAKE, N., and SCHROEDER, P. B., 1978 — The effect of heat effluents from power plants on seagrass (Thalassia) communities quantitatively comparing estuaries in the subtropical to the tropics. Marine Pollution Bulletin 9: 181-187. , and HIxon, R., 1975 — Revegetation of Thalassia testudinum in a multiply stressed estuary, North Proc. LINN. Soc. N.S.W. 106 (3), (1982) 1983 212 STABILITY, DEPLETION AND RESTORATION OF SEAGRASS BEDS Biscayne Bay, Florida. pp 12-17 in Second Ann. Congr. on Restoration of Coastal Veg. in Fla. Ed. R. R. LEWIS. Tampa Fla: Hillsborough College Press. 214 pp. , SEGAR, D., and ROESSLER, M. A., 1974 — Impact of a power plant on a subtropical estuarine en- vironment. Mar. Pollut. Bull. 4: 166-169. TOMLINSON, P. B., 1974 — Vegetative morphology and meristem dependence, the foundation of productivity in seagrasses. Aquaculture 4: 107-130. TYERMAN, S., 1979 —Turgor regulation and the development of water potential gradients in Posidonia. In (pp. 465-468) Plant Membrane Transport: Current Conceptual Issues, R. M. SPANSWICK,, W. J. LUCAS and J. Dainty, (eds). Amsterdam: Elsevier/North Holland. VAN EEPOEL, R. R., GriGG, D. I., BRODy, R. W., and RAYMOND, W., 1971 — Water quality and sediments of Lindbergh Bay, St. Thomas. Carrib. Res. Inst., Water Pollut. Rept no. 11: 1-33. YOUNG, C., and KIRKMAN, H., 1975 — The seagrass communities of Moreton Bay, Queensland. Aquatic Botany 1: 191-202. WEST, R.J., 1980 — A study of growth and primary production of the seagrass Poszdonia australis Hook. f. Sydney: University of Sydney, M.Sc. thesis, unpubl. ZIEMAN, J. C., 1975 — Seasonal variation of turtle grass Thalassia testudinum Konig with references to temperature and salinity effects. Aquatic Botany 1: 107-124. ——., 1976 — The ecological effects of physical damage from motor boats on turtle grass beds in southern Florida. Aquatic Botany 2: 127-139. Proc. LINN. Soc. N.S.W. 106 (3), (1982) 1983 Seagrass Primary Production — a Review* R. J. WEST and A. W. D. LARKUM WEST, R. J., & LARKuM, A. W. D. Seagrass primary production — a review. Proc. Linn. Soc. N.S.W. 106 (3), (1982) 1983:213-223. Seagrasses dominate many inshore benthic communities in Australia and contribute a high proportion of the total fixed carbon of the inshore ecosystem. The methods in use to measure primary production of seagrass communities with special reference to Australia are reviewed, and some of the comparable results listed. Seagrass productivity is a reflection of climate, local environmental conditions, and individual growth strategies of particular species, but is high in comparison with most other plant communities. Light, temperature, salinity, nutrients, sediment and human influence are reviewed as possible controls on seagrass production. R. J. West, N.S. W. State Fisheries, P.O. Box N211, Grosvenor Street, Sydney, Australia 2001, and A. W. D. Larkum, School of Biological Sciences, University of Sydney, Sydney, Australia 2006; manuscript received 19 May 1981, accepted for publication in revised form 16 December 1981. INTRODUCTION Production refers to the capacity of a biological system to form organic matter, and the rate of this formation is called productivity. Primary production is the capture of radiant solar energy by plants and subsequent conversion of carbon and water to organic matter during photosynthesis. The measurement of primary productivity is of importance as it can give an in- dication of the capability of a biological system to support a food web of secondary producers and leads to a better understanding of the growth strategies of plant species. Productivity measurements have the units dry matter produced per unit area per unit time, usually expressed as grams dry weight per square metre per day, or tonnes dry weight per hectare per year. When the relationship between carbon content and dry weight of the plant material is known, then another useful unit is grams carbon per square metre per hour. Although dry weight has been used extensively in this review, one should bear in mind that the variability in ash content (interspecific, and seasonal) may make comparisons of productivity in terms of dry weight misleading. Wherever necessary, the values of McRoy and McMillan (1977) have been adopted for the sake of comparison. These authors report that ‘available data on eelgrass indicate an organic weight of about 80% of the dry weight, 47% of which is carbon’. There are two broad methods often used to measure primary productivity (Larkum, 1981). One is based on the calculation of net photosynthetic rates, while the other is a measure of growth. The former is a short-term method, the period of ex- perimentation being from a matter of minutes, to perhaps as long as twenty-four hours, and involves the measurement of radioactive carbon fixation or of net oxygen production. The latter is a long term method, based on the increases in the dry matter over a period of weeks, months or years. In most seagrasses, at least three processes, photorespiration (Larkum, 1981), the loss of dissolved organic carbon (Brylinsky, 1977) and formation and translocation of carbon reserves (Dawes and Lawrence, 1979), will ensure that net photosynthesis and growth are not equivalent over the short term. Despite this, direct comparisons of the methods have been attempted (Bittaker and Iverson, 1976). * A paper presented at the Australian Seagrass Workshop, University of Sydney, 1981. Proc. LINN. Soc. N.S.W. 106 (3), (1982) 1983 214 SEAGRASS PRIMARY PRODUCTION METHODS OF MEASUREMENT The short-term methods of estimation of primary production include measurement of oxygen production, or of radioactive carbon uptake (Vollenweider, 1971). Both of these methods may be affected by the recycling of gases in the extensive lacunal system possessed by most seagrasses (Zieman, 1968; McRoy and McMillan, 1977). Methods employed to measure oxygen release from seagrass communities may involve the use of closed vessels, such as glass bottles or large acrylic chambers, or they may involve measurement in open waters (Vollenweider, 1971). The principal method of oxygen determination is the Winkler chemical method (Strickland and Parsons, 1972), although oxygen electrode techniques are also widely in use (Pijanowski, 1975). Diurnal oxygen concentrations in a flowing water mass provided one of the earliest short term measures of productivity of aquatic plants (Odum, 1956, 1957). This method made no allowances for planktonic metabolism, and relied heavily on a crude estimate of water volume and speed. Loss of oxygen through bubbling, and changes in concentration due to reaeration were further problems. Several later studies relied on this method (Odum, Burkholder and Rivero, 1959; Quasim and Bhattathiri, 1971). Improvements have been made to this technique by the measurement of the planktonic contribution (through the use of light and dark bottles), and of the oxygen reaeration (Nixon and Oviatt, 1972). A further improvement is in the isolation of a known volume of water (Weiner and Kirkman, 1980). A variation of the above technique is the collection of whole plants (or of leaf sections) which are subsequently incubated in glass jars either zm sztu or under con- trolled laboratory conditions, and assayed for oxygen production (Jones, 1968; Buesa, 1974; Drew, 1978). The most sophisticated method employed in the measurement of oxygen release from seagrass communities is the simultaneous recording of oxygen, temperature, light and salinity within a completely enclosed chamber, with circulating seawater (Weiner and Kirkman, 1980; Clough and Attiwill, 1980). Clough and Attiwill (1980) have used this method to predict the primary productivity of Zostera mueller: in Western Port Bay, Victoria. These authors have calculated a photosynthetic response curve from several diurnal incubations. This treatment could be substantially improved by increasing the number of diurnal incubations to exclude seasonal variations in photosynthesis, and by the continuous measurement of light at the plant-canopy level. However the problem created by the existence of internal gas lacunae remains unresolved. Wetzel (1964) and Drew and Larkum (1967) pioneered the in situ use of radioactive isotopes to measure primary productivity of aquatic plants. These workers used bell jars to incubate single plants which were later assayed for carbon-14. McRoy (1974) adopted a similar method with the seagrass Zostera marina in Alaskan waters, while Drew and Jupp (1976) measured uptake by Poszdonia oceanica in Mediterranean waters. Penhale (1977) measured primary productivity of Thalassia testudinum and its epiphytes by a carbon-14 method, this time improved by the reduction of unstirred layers through the circulation of the seawater surrounding the plants. The use of radio- isotopes in the field is somewhat difficult and does not overcome the problem of the possible recycling of gases in the lacunae (McRoy and McMillan, 1977). Long term methods of measurement of primary productivity were first used in relation to seagrasses in Denmark in the early part of this century. Petersen (1913) estimated the primary productivity of Zostera marina by harvesting this seagrass at several stages of its growth cycle, and concluded that these plants made an important contribution to production in the estuarine system. He did not however take into account the rapid turnover of leaf material that is often characteristic of seagrasses, for Proc. LINN. Soc. N.S.W. 106 (3), (1982) 1983 R. J. WEST ANDA. W. D. LARKUM 215 example, of many Zostera species (Sand-Jensen, 1975). Petersen’s estimates were therefore very much lower than the true production rates. Such biomass figures are still often used as an estimate of productivity of plants (Westlake, 1963), although turnover of biomass is the more important concept. This can be defined as the number of times mean biomass is replaced throughout the annual growing cycle: Tusenowrae ite Annual increase in biomass (crops per year) mean biomass Zieman (1968, 1974, 1975) developed a marking technique to measure leaf production in meadows of Thalassia testudinum in waters off Florida. This method was described at the International Seagrass Workshop in the Netherlands as the only reliable method to measure directly leaf production of seagrasses (McMillan et al. , 1973). It has been used in many studies involving Thalassia testudinum (Zieman, 1975; Patriquin, 1973; Greenway, 1977), Zostera marina (Sand-Jensen, 1975) and Posidonia australis (West and Larkum, 1979; Kirkman and Reid, 1979). Few workers have attempted to measure rhizome growth in seagrasses. Patriquin (1973) introduced a method based on the time interval between new leaf production, the dry weight of rhizome sections, and the number of leaf scars on the rhizome. This method appears to work adequately for Thalassia testudinum and Posidonia australis (West, 1980). Sand-Jensen (1975) has used the increase in shoot numbers in fixed quadrats to estimate rhizome production for Zostera marina. COMPARISON OF PRODUCTIVITY MEASUREMENTS The results of many productivity studies have been summarized elsewhere (McRoy and McMillan, 1977; Zieman and Wetzel, 1980). These authors have made little or no reference to the many different methods adopted in the individual productivity studies. However, in this review, the recommendation of the In- ternational Seagrass Workshop held in the Netherlands (McMillan e¢ al., 1973) is adopted, and only results based on leaf-making techniques are included. Some representative values for leaf-production of several seagrass species are shown in Table 1, and some available data concerning rhizome production shown in Table 2. There are often large differences in biomass between stands of the same seagrass species, depending on localized environmental conditions, and this leads to variations in the productivity estimates between such stands. The differences in biomass may be due to different shoot densities, or variability in leaf length and width. West and Larkum (1980) compared five study sites and found large variations in biomass, densities and leaf characteristics between stands of Posidonia australis, however leaf turnover was relatively constant (2.8-4.5 crops of leaves per year). This consistency in mean leaf turnover is also apparent from a number of studies carried out on various seagrass species (Table 1), although seasonal variations often mask the observation. Thus it would seem that for many seagrass species, the number of leaves produced each year by individual shoots may be quite consistent, although biomass, density and leaf characteristics vary, depending on local conditions, such as depth, sediment, age of bed, degree of wave exposure and exposure to salinity changes. Thus care must be taken when assessing the productivity of a large area. It is not often possible to make an estimate of the area of seagrass beds, as say by aerial photography and then to multiply this area by the productivity found for plants in an experimental area. Plant density is not always linearly related to plant cover (Larkum et al., in press) and plant cover is what is usually estimated from aerial photography. Proc. LINN. Soc. N.S.W. 106 (3), (1982) 1983 SEAGRASS PRIMARY PRODUCTION 216 G6] ‘uesue{-pues 1861 ‘smyyng (qndun) umyieT T861 72 72 QUOD IN 6261 ‘e8pruquieo 6261 ‘e5pluquied 6L6] ‘PIOY B UeUIyAITY 6L6T “WNyIe'T 29 SOM e26r unbineg 916] ‘Aemusai5y GL6] ‘UeWIaIZ IOUIIIJFOY (ueaut) ¢-9 (ueaut) 0’ / (uvour) 042 (ueowl) ¢°¢ (ueour) ¢°z CT8C 8 rI-cT (ueaut) 7°98 €1-6 6 JIZAOUIN T, 00S-00€ (uvour) 0'9 (uvout) 6%Z 0'01-2'% (uvaut) (gz Aep/,w/'mp's zu/mp's AyAnONpoig sseuloIg uoyonpoLg-f{oaT sspssvay fo saps T dldVv LT “W19}S 19919 pue Jeo] sopnpout — dors Sutpueys , 97e19d UI9}-pjoo 9}e19dUI9}-pjoo o7e10du19)-wW1eM 9)e19dUI9}-WIeEM o1eroduie}-wiieMm a1eroduia}-wiIeM 9)e19du19)-ulIeEM 93e19du19}-ulIeEM jeordon eo1don jeotdon aqeunl{ yrewusg DULLDUL DLB]SOF BIIOII A 4» DIUDUSD} DlaySOZOLIOET 1us0911g09 v1aso7z psonuis DiUuopisoy SYDLSND DIUOpISOT epnuliog eoreure [” eplloyy wnuipnysa] DISsD{DY J. Ayrpeso'T sseIseag Proc. LINN. Soc. N.S.W. 106 (3), (1982) 1983 217 R.J. WEST AND A. W. D. LARKUM G16] ‘uesue{-pueg O86T “S9M €L6, ‘umbineg (ueour) ¢'T (uvout) ¢T'( dUdIIJOY ‘IR /sdoin JQ9AOUIN T, Aep/.w/ Mp’3 AWATIONpOIg uoIMpoLg auloziyy ssosdvag fo sajvULS Ty Z ATV L DULLDUL DLIISOZ S1]DLSND DIUOPISOT i eae ha ae fee ee See ee areunty) Proc. LINN. Soc. N.S.W. 106 (3), (1982) 1983 218 SEAGRASS PRIMARY PRODUCTION Larkum ¢ al. (in press) found that a detailed analysis of the beds of Zostera capricorni in Botany Bay gave a productivity value which was 5 fold lower than the estimate ob- tained by the simple product of total area and a productivity value obtained in a dense, shallow bed. From the data available (Table 2) it would appear that rhizome production is extremely variable between species. Rhizome production is obviously a reflection of the growth habit of individual species, and the stability of a particular seagrass stand. For example, West (1980) has shown that turnover of rhizome material is slow, but depends markedly on site, in stands of Poszdonia australis in Botany Bay. Some erect shoots may be more than 20 years in age (cf Kuo and Cambridge, 1978). Other seagrasses must exhibit a high turnover of rhizomes, as these species appear to replace rhizome annually (personal observation). In general, it may be said that the production of organic matter in seagrass beds occurs at a high rate when compared with most other plant communities (Westlake, 1963), and is a reflection of climate, local environmental conditions and individual growth strategies of particular species. CONTROLS ON PRODUCTIVITY Although turnover of organic matter may be consistent in established stands of seagrasses, there are many factors which can influence the biomass that can be achieved or which influence the seasonality in growth at particular sites, and some of these factors will now be discussed. Light and temperature Light is the most critical ecological factor for plant growth as it directly affects the rate of photosynthesis. The lower depth limit to the establishment of seagrasses is generally considered to be due to the minimum light intensity required by a particular seagrass (Phillips, 1978), although few conclusive experiments have been performed. Ostenfeld (1905) was one of the first workers to attempt to correlate distribution patterns of a seagrass (Zostera marina) to underwater irradiance. The short term methods of measuring productivity (see earlier) generally yield information concerning the effect of light on photosynthetic rate, however the variability in methods and light units measured makes comparison of this work dif- ficult. For example, Jones (1968) and Buesa (1974, 1975) have used leaf sections isolated in bottles, Drew (1978) used leaf slices in small jars, McRoy (1974) used whole plants in jars, Penhale (1977) used whole plants in circulating seawater, and Clough and Attiwill (1979) used whole plants (zn situ) in benthic chambers. Most of these workers used different measures of light intensity. Drew (1979) has compared the photosynthetic response of six seagrass species, and found that, in all cases, a classic P-I curve for plants intermediate between sun and shade adaptation could be demonstrated and that thermal damage occurred above 30- 35°C. Further this author has found that at least one species, Halophila stipulacea, is capable of a chloroplast re-orientation that may be a protective mechanism in high light situations. Trocine, Rice and Wells (1981) have also noted chloroplast clumping in a Halophila species and suggest that this results in an increased photosynthetic tolerance to ultraviolet-B radiation. McRoy and McWilliam (1977) suggest that production can be expressed in terms of light quantity. They report three separate studies giving estimates of between 5.3- 117.0 microgram carbon per langley per gram biomass, and propose that the similarity in result may lead to a useful relationship from which productivity could be estimated from light intensity. PROC. LINN. Soc. N.S.W. 106 (3), (1982) 1983 R. J. WEST ANDA. W. D. LARKUM 219 Backman and Barilotti (1976) have shown that turion density of Zostera marina is reduced when irradiance is reduced. Thus these authors present perhaps the best evidence to show that a coastal development that affects turbidity will also affect seagrass growth and production. Temperature is also known to affect photosynthetic rates of seagrasses (Biebl and McRoy, 1971), and the effect of high temperature on the growth of some species is well documented (Zieman, 1975; Thorhaug, Blake and Schroeder, 1978). Zieman (1975) found that leaf production in Thalassia testudinum was near zero at temperatures above 35°C or below 25°C. The interaction of light and temperature in field situations makes the individual effects of these environmental variables difficult to study, and has led to some disagreement and confusion as to whether light, or temperature, is responsible for seasonality of growth rates of seagrasses. The variation in method for the measurement of seagrass production (i.e. leaf growth or photosynthetic rate) also adds to the con- fusion. From our own studies with Poszdonza australis it would appear that temperature controls individual leaf growth rates provided that light intensity has remained suf- ficiently high to allow the plants to accumulate reserves in the rhizome (West, 1980; West and Larkum, in preparation). Salinity There are few references in the literature to the effect of salinity changes on seagrass productivity. Biebl and McRoy (1971) report a reduction in photosynthesis, when salinity is reduced from seawater, for Zostera marina, while Ogata and Matsui (1965) found for Zostera nana that a reduction in photosynthesis with increasing salinity above seawater was due principally to carbon limitation. Other laboratory studies have involved the transplantation of seagrasses to aquaria at various salinities (McMillan and Mosely, 1967; McMahon, 1968; Mc- Millan, 1974). McRoy and McMillan (1977) summarize these results: “Halodule shows that broadest salinity tolerance . . . Halophila the narrowest ... and Thalassia and Syringodium show intermediate tolerance’ . Zieman (1975) found that productivity decreased above and below the optimum salinity of 30 parts per thousand for Thalassia testudinum. West (1980) found that large short term changes in salinity could cause mortality of Posidonia australis, but there was no significant effect on leaf growth of surviving plants. The effect of salinity on establishment of seagrass species requires further studies so that an attempt to explain distribution patterns can be made. Sediment and nutrients Little is known concerning the sediment requirements of most seagrasses although some general observations have been listed by Den Hartog (1970). For example, Posidonia australis generally occurs on sandy substratum, whereas Zostera capricorni is not restricted to sand, and grows on a wide range of substrates (Harris, King and Ellis, 1979). One advantage that seagrasses have over most algae is that they are not dependent , on the nutrient availability in the water column. Both anatomical studies and physiological experiments indicate that seagrasses can take advantage of the often rich nutrient supply offered by the sediment and interstitial water. Kuo and Cambridge (1978) found a reduced vascular system in Poszdonta australis, but all the elements required for nutrient uptake and transfer were present. Further, McComb et al. (1981) suggest that the wide occurrence of bacterial colonies, fungi and Proc. LINN. Soc. N.S.W. 106 (3), (1982) 1983 220 SEAGRASS PRIMARY PRODUCTION other micro-organisms found in the rhizosphere and peripheral root tissue of Postdonia australis may be involved in nutrient uptake and nitrogen fixation in seagrass roots. Further evidence exists for the occurrence of nitrogen fixation in either the phyllosphere or rhizosphere of other seagrasses (Patriquin and Knowles, 1972; Goering and Parker, 1972). Capone et al. (1979) have found a relationship between productivity (biomass) and nitrogen fixation in the Thalassia testudinum community. McRoy, Blackburn and Klug (1981) have suggested that the level of nitrogen fixation and sulphur reduction in sediments associated with seagrasses is related to the successional stage of the seagrass community. Phosphate absorption has been demonstrated for the roots and rhizomes of eelgrass (McRoy and Barsdate, 1970; Penhale and Thayer, 1980), and for Poszdonia australis (Larkum, unpublished data). Human influence The effect of man on seagrass productivity is a result of industrial and urban development along shallow marine embayments and along coastal fringes. Thayer, Wolfe and Williams (1976) have summarized many of the impacts of man on seagrass systems. Light, salinity, water turbulence and nutrient levels are all affected by the degree of foreshore development, and toxic contamination, dredging and ‘filling-in’ should be added as further possible impacts. Evidence exists for a decline in Poszdonia australis beds near urban areas in New South Wales, South Australia and Western Australia (Larkum, 1977). There has been a continued reduction in the area colonized by Poszdonia australis in Botany Bay (N.S.W.) through both increased turbidity and water turbulence (Larkum, 1976; West, 1980). Aerial photographic evidence confirms the loss of 20% in the area covered by seagrasses, in Botany Bay, in the last fifteen to twenty years. A decline of Posidonia australis in Cockburn Sound, Western Australia, has also been described (Cambridge, 1975, 1979). Similar situations have been documented for the meadows of Posidonia oceanica off the French coastline (Meinesz and Laurent, 1978). The recovery of seagrass communities in degraded areas may be extremely slow, even if the environmental stress is relieved. This results from the fact that many seagrasses appear to rely heavily on vegetative reproduction, and rhizome growth is often very slow. Artificial revegetation is difficult and costly and thus careful management of existing seagrass areas 1s imperative. SECONDARY PRODUCTION As stated, seagrasses produce large quantities of organic matter. Seagrass meadows also offer a diverse habitat for secondary producers (Kikuchi and Peres, OD): Although leaf material may be grazed quite heavily in tropical areas by fish (Ogden, 1976), turtles (Hirth, Klikoff and Harper, 1973) and dugongs (Heinsohn, Wake, Marsh and Spain, 1977) there are few direct grazers in temperate regions (Kikuchi and Peres, 1977). Conacher et al. (1979) have shown that the major fish grazer of Postdonia australis, Monacanthus chinensis removes less than 1% of the productive capacity. Nevertheless these fish do obtain some nutrients from ingested leaf fragments. Kirkman and Reid (1979) have studied the role of a seagrass (P. australis) in the carbon-budget of an estuary. Only 3% of the total seagrass biomass (in terms of carbon) was consumed by herbivores, whereas the remainder was lost as exuded dissolved organic carbon and detached leaves in approximately equal proportions. PROC. LINN. SOc. N.S.W. 106 (3), (1982) 1983 R. J. WEST ANDA. W. D. LARKUM 221 Thus, secondary production in temperate regions occurs primarily through the detrital food chain (Fenchel, 1977):— Leaching of dissolved organic and inorganic materials, Mechanical breakdown of leaves (waves or ingestion), Bacterial and fungal decomposition, Attack by bacteriovorous microfauna and detritivores, These small fauna are prey for large carnivores. Oe OO Ne GENERAL CONCLUSIONS Care is required in the selection of a method to measure productivity and Zieman’s leaf marking technique (Zieman, 1974) is still preferred. Primary production of many seagrass species is high and related to growth strategy of individual species and local environmental conditions. The ability of seagrasses to inhabit sandy substrates, to use sediment nutrients, and to accumulate carbon reserves are important factors which ensure a relatively stable food supply and habitat, and this is reflected in a diverse and important faunal community. References BACKMAN, T. W., and BARILOTTI, D. C., 1976 — Irradiance reduction: effects on standing crop of eelgrass Zostera marina in a coastal lagoon. Marine Biology 34:33-40. BIEBL, R., and MCRoy, C. P., 1971 — Plasmatic resistance and rate of respiration and photosynthesis of Zostera marina at different salinities and temperatures. Marine Biology 8: 48-56. BITTAKER, H. T., and IVERSON, R. L., 1976 — Thalassia testudinum productivity: a field comparison of measurement methods. Marine Biology 37: 39-46. BUESA, R. J., 1974 — Population and biological data on turtle grass (Thalassia testudinum Konig, 1805) on the northwestern Cuban shelf. Aquaculture 4: 207-227. ——, 1975 — Population biomass and metabolic rates of marine angiosperms on the northwestern Cuban shelf. Aquatic Botany 1: 11-23. BULTHUIS, D. A., 1981 — Seasonal variation in standing crop, density and leaf growth rate of Heterozostera tasmanica in Westernport and Port Philip, Victoria. (Abstract). In, Australian Seagrass Workshop, 1981. Sydney: University of Sydney. CAMBRIDGE, M. L., 1975 — Seagrasses of southwestern Australia with special reference to the ecology of Posidonia australis Hook f. in a polluted environment. Aquatic Botany 1: 149-161. ——.,, 1979 — Cockburn Sound Technical Report on Seagrass. Report No. 7. Department of Conservation and Environment, Western Australia. CAPONE, D. G., PENHALE, P. A., OREMLAND, P. S., and TAYLOR, B. F., 1979 — Relationship between productivity and Ny (C)H,) fixation in a Thalassia testudinum community. Limnol. Oceanogr. 24: 117- 125. CLOUGH, B. F., and ATTIWILL, P. M., 1980 — Primary productivity of Zostera mueller: Irmisch ex Aschers. in southern Australia. Aquatic Botany 9: 1-13. CONACHER, M. J., LANZING, W. J. R., and LARKuM, A. W. D., 1979 — The ecology of Botany Bay II. Aspects of the feeding ecology of the fanbellied leatherjacket, Monacanthus chinesis (Pisces: Monacanthidae) in Posidonia australis seagrass beds in Quibray Bay, Botany Bay, New South Wales. Australia: J. mar. Freshw. Res. 30: 387-400. Dawes, C. J., and LAWRENCE, J. M., 1979 — Effects of blade removal on the proximate composition of the rhizome of the seagrass Thalassia testudinum Banks ex K6énig. Aquatic Botany 7: 255-266. DEN HARTOG, C., 1970 — The seagrasses of the world. Amsterdam: North-Holland. Drew, E. A., 1978. Factors affecting photosynthesis and its seasonal variation in the seagrasses Cymodocea nodosa (Ucria) Aschers, and Poszdonia oceanica (L) Delile, in the Mediterranean. J. Exp. mar. Biol. Ecol. 31: 173-194. ——., 1979 — Physiological aspects of primary production in seagrasses. Aquatic Botany 7: 139-150. , and Jupp, B. P., 1976 — Some aspects of the growth of Poszdonia oceanica in Malta. pp. 357-367 in, Drew, E. A., LYTHGOE, J. N., and Woops, J. D., (eds), Underwater Research. London: Academic Press. , and LARKUM, A. W. D., 1967 — Photosynthesis and growth in Udotea petiolata, a green algae from deep water. Underwater Ass. Rep. 1966-67. Proc. LINN. Soc. N.S.W. 106 (3), (1982) 1983 222 SEAGRASS PRIMARY PRODUCTION FENCHEL, T., 1977 — Aspects of the decomposition of seagrasses. pp. 123-145 in, McRoy, C. P., and HELFFERICH, C., (eds), Seagrass Ecosystems. New York: Marcel Dekker. GOERING, J. J., and PARKER, P. L., 1972 — Nitrogen fixation by epiphytes on seagrasses. Limmol. Oceanogr. 17: 320-323. GREENWAY, M., 1974 — The effects of cropping on the growth of Thalassia testudinum (K6nig) in Jamaica. Aquaculture 4: 199-206. Harris, M. McD., KING, R. J., and ELLs, J., 1979 (1980) — The eelgrass Zostera capricorni in Illawarra Lake, New South Wales. Proc. Linn. Soc. N.S. W. 104 (1): 23-33. HEINSON, G. E., WAKE, J., MARSH, H., and SpaIN, A. J., 1977 — The dugong (Dugong dugon Miller) in the seagrass system. Aquaculture 12: 235-248. HirTH, H. F., Kiikorr, L. G., and HARPER, K. T., 1973 — Seagrasses at Khar Umaira, Peoples Democratic Republic of Yemen, with special reference in their role in the diet of the green turtle Chelonia mydas. Fish. Bull. 71: 1093-1097. JONES, J. A., 1968 — Primary productivity by the ttropical marine turtle grass Thalassia testudinum Konig, and its epiphytes. University of Miami: Ph.D. thesis. unpubl. KIKUCHI, T., and PERES, J. M., 1977 — Consumer ecology of seagrass beds. pp. 147-193 in, MCRoy, C. P., and HELFFERICH, C.., (eds), Seagrass Ecosystems. New York: Marcel Dekker. KIRKMAN, H., and REID, D. D., 1979 — A study of the role of the seagrass Poszdonza australis in the carbon budget of an estuary. Aquatic Botany 7: 173-183. Kuo, J., and CAMBRIDGE, M. L., 1978 — Morphology, anatomy and histochemistry of the Australian seagrasses, genus Posidonia KGnig (Posidoniaceae). II. Rhizome and root of Posidonza australis Hook f. Aquatic Botany 5: 191-206. LarkuM, A. W. D., 1976 — The ecology of Botany Bay. I. Growth of Posidonia australis (Brown) Hook. f. in Botany Bay and other bays of the Sydney Basin. Aust. J. mar. Freshw. Res. 27: 117-127. ——, 1977 — Seagrass research in Australia. In Seagrass Ecosystems: A Scientific Perspective, (ed.) C. P. MCcRoy and C. HELFERRICH, pp. 247-262. New York: Marcel Dekker. ——,, 1981 — Marine primary productivity. In Marine Botany. An Australasian Perspective, CLAYTON, M. C., and KING, R. J., (eds), pp. 369-385. Melbourne: Longman-Cheshire. —., COLLETT, L. C., and WILLIAMS, R. W., in press. The ecology of Botany Bay. III. The growth and productivity at Zostera marina. Aust. J. mar. Freshw. Res. McComs, A. J. J., CAMBRIDGE, M. L., KIRKMAN, H., and Kuo, J., 1981 — The biology of Australian seagrasses. In PATE, J. S., and McComs, A. J., The Biology of Australian Plants, pp. 258-293. Nedlands: University of Western Australia Press. McManan, C. A., 1968. Biomass and salinity tolerance of shoalgrass and manateegrass in Lower Lachlan Madre. Texas J. Wildlife Manag. 32: 501-506. MCMILLAN, C., 1974 — Salt tolerance of mangroves and submerged aquatic plants. In REINOLD, R. J., and QUEEN, W. H.., (eds), Ecology of halophytes, pp. 379-390. New York: Academic Press. —., McRoy, C. P., BARSDATE, R. J., DONZE, M., LUTHER, H., THORHAUG, A., and TOMLINSON, P. B., 1973 — I. Productivity/physiology working group. Seagrass Ecosystems — Proc. Intern — Seagrass Workshop, pp. 5-12. The Netherlands: Leiden. , and MosgELy, F. N., 1967 — Salinity tolerance of five marine spermatophytes of Redfish Bay, Texas. Ecology 48: 503-506. McRoy, C. P., 1974 — Seagrass productivity — carbon uptake experiments in the eelgrass, Zostera marina. Aquaculture 4: 131-137. ——., and BARSDATE, R. J., 1970 — Phosphate absorption in eelgrass. Limnol. Oceanogr. 15: 6-13. , and MCMILLAN, C., 1977. Production ecology and physiology of seagrasses. In MCROY, C. P., and HELFFERICH, C., Seagrass Ecosystems, pp. 53-87. New York: Marcel Dekker. ——., BLACKBURN, H., and KLUuG, M. J., 1981 — Nitrogen and sulphur turnover in sediments associated with tropical Australian seagrasses. Australian Seagrass Workshop, 1981. Sydney: University of Sydney. MEINESZ, A., and LAURENT, R., 1978 — Cartographie et etat de la limite inferieuse de l’herbier de Poszdonza oceanica dans les Alpes-maritimes (France). Botanica marina XX1: 513-526. NIXON, 8S. W., and OviaTT, C. A., 1972 — Preliminary measurements of midsummer metabolism in beds of eelgrass, Zostera marina. Ecology 53: 150-153. OGDEN, J. C., 1976 — Some aspects of plant-herbovose relationships on Caribbean reef fishes and seagrass beds. Aquatic Botany 2: 103-116. OpuM, H. T., 1956 — Primary production in flowing waters. Limnol. Oceanogr. 1: 103-117. , 1957 — Trophic structure and productivity of Silver Springs, Florida. Ecological Monographs 27: 55- 112. ——.,, BURKHOLDER, P. R., and RIVERO, J., 1959 — Measurements of productivity of turtle-grass flats, reefs and the bahia fosforesante of Southern Puerto Rico. Publ. Inst. Mar. Scr. Texas 6: 159-170. Proc. LINN. SOc. N.S.W. 106 (3), (1982) 1983 R.J. WEST AND A. W. D. LARKUM 223 OcatTa, E., and Matsut, T., 1965 — Photosynthesis in several marine plants of Japan as affected by salinity, drying and pH with attention to their growth habits. Botanica marina VIII: 199-217. OSTENFELD, C. H., 1905 — Preliminary remarks on the distribution and biology of the Zostera in Danish seas. Bot. Tidsskr. 27: 123-125. PATRIQUIN, D. G., 1973 — Estimation of growth rate, production and age of the marine angiosperm Thalassia testudinum K6nig. Carrib. J. Sci. 13: 111-121. PENHALE, P. A., 1977 — Macrophyte — epiphyte biomass and productivity in an eelgrass (Zostera marina L.) community. J. exp. mar. Biol. Ecol. 26: 211-214. , and THAYER, G. W., 1980 — Uptake and transfer of carbon and phosphorus by eelgrass (Zostera marina L.) and its epiphytes. J. exp. mar. Biol. Ecol. 42: 113-123. PETERSEN, C. G. J., 1913 — Om baedeltangens (Zostera marina) aarsproduktion i de Danske Farvande. In, JUNGERSEN, H. F. E., and WARMING, E., (eds), Mindeskr. Japetusteenstrups. Fodsel. 1 (9): 1-20. Copenhagen: Bianco Lunos Bogtrykkeri. PHILLIPS, R. C., 1978 — Seagrasses and the coastal marine environment. Oceanus 21: 30-40. PIANOWSKI, B. S., 1975 — Dissolved oxygen sensors — theory of operation, testing and calibration technique. Symposium on Chemistry and Physics of Aqueous Solutions, Toronto, pp. 373-381. QuasIM, S. Z., and BHATTATHIRI, P. M. A., 1971 — Primary production of seagrass bed on Kavaratt Atoll (Laccadives). Hydrobiologia 38: 29-38. SAND-JENSEN, K., 1975 — Biomass, net production and growth dynamics in an eelgrass (Zostera marina L.) population in Vellerup Vig, Denmark. Ophelia 14: 185-201. STRICKLAND, J. D. H., and PARSONS, T. R., 1972. A practical handbook of seawater analysis. 2nd. ed. Canadian Fish. Res. Brd. Bull. 167. THAYER, G. W., WOLFE, D. A., and WILLIAMS, R. B., 1975 — The impact of men on seagrass systems. American Sct. 63: 288-296. THORHAUG, A., BLAKE, B., and SCHROEDER, P. B., 1978 — The effect of heated effluents from power plants on seagrass (Thalassia) communities, quantitatively comparing estuaries in the sub-tropics to the tropics. Marine Pollution Bulletin 9: 181-187. TROCINE, R. P., RICE, J. D., and WELLS, G. N., 1981 — Inhibition of seagrass photosynthesis by ultraviolet-B radiation. Plant Physiol. 68: 74-81. VOLLENWEIDER, R. A., 1971. A manual on methods for measuring primary production in aquatic en- vironments. J.B. P. Handbook. No. 12. Oxford: Blackwell. WEINER, P. S., and KIRKMAN, H., 1980 — Continual recording technique used to measure release of oxygen from a seagrass community with an acrylic insulation chamber. C.S./.R.O. Aust. Div. Fish. Oceangr. Rep. 96. WEST, R. J., 1980 — A study of growth and primary production of the seagrass, Postdonia australis. Hook. f. Sydney: University of Sydney, M.Sc. thesis, unpubl. , and LARKUM, A. W. D., 1979 — Leaf productivity of the seagrass Posidonia austraiis in eastern Australian waters. Aquatic Botany 7: 57-65. WESTLAKE, D. F., 1963 — Comparisons of plant productivity. Biol. Rev. 28: 385-425. WETZEL, R. G., 1964 — A comparative study of primary productivity of higher aquatic plants, periphyton and phytoplankton in a shallow lake. Int. Rev. ges. Hydrobiol., 49: 1-61. ZIEMAN, J. C., 1968 — A study of growth and decomposition of the seagrass Thalassia testudinum. Florida: University of Miami, M.Sc. thesis. ——., 1974 — Methods for the study of the growth and production of turtle grass Thalassia testudinum Konig. Aquaculture, 4: 139-143. ——, 1975 — Quantative and dynamic aspects of the ecology of turtle grass, Thalassza testudinum. In CRONIN, E. L., (ed), Estuarine research, Vol. 1. New York: Academic Press. , and WETZEL, R. G., 1980 — Productivity in seagrasses: methods and rates. In PHILLIPS, R. C., and McRoy, C. P., (eds), Handbook of Seagrass Biology: an ecosystem perspective, pp. 87-116. New York: Garland STPM Press. Proc. LINN. Soc. N.S.W. 106 (3), (1982) 1983 Ge ah aie Notes on the Biology of Australian Seagrasses* J. Kuo (Communicated by A. W. D. LARKUM) Kuo, J. Notes on the biology of Australian seagrasses. Proc. Linn. Soc. N.S.W. 106 (3), (1982) 1983: 225-245. This contribution focuses on the anatomy, histochemistry and ultrastructure of vegetative organs of Australian seagrasses in relation to their possible functions. Detailed anatomical and ultrastructural structures vary with species. Some structures are similar to those in terrestrial plants and others are unique in seagrasses for adaptation to a marine environment. Some details of the taxonomy, distribution, habitat and life history of Australian seagrasses are also included. The structure of the phyllosphere and rhizosphere in seagrasses is discussed as well as the possible significance of the role of epiphytes and epifauna. J. Kuo, Electron Microscopy Centre, University of Western Australia, Nedlands, Australia 6009, manuscript recewwed 19 May 1981, accepted for publication in revised form 17 November 1982. INTRODUCTION Seagrasses are aquatic angiosperms which are completely adapted to life in the marine environment. According to Arber (1920) and den Hartog (1970) they require at least the following adaptations to be able to colonize the marine environment suc- cessfully: (1) they live fully submerged in seawater, (2) they have an anchoring system, (3) they have a hydrophilous pollination, (4) they are able to cope with a high salinity. Seagrasses are abundant in Australian waters and play a significant role in marine food chains (Larkum, 1977; King, 1981; McComb et al., 1981). The ecology of seagrasses including Australian species has been intensively reviewed (McRoy and Helfferich, 1977; Larkum, 1977; Phillips and McRoy,, 1980; McComb et a/., 1981) and will not be dealt with here. Knowledge on other biological aspects of these marine vascular plants is still fragmented. This contribution, which emphasizes anatomy, cellular structure and function, is concerned with these adaptations and other biological aspects of marine angiosperms in Australian waters. A comparative leaf morphology and gross anatomy of seagrasses has been described by Tomlinson (1980). TAXONOMY, DISTRIBUTION AND HABITATS Taxonomy Den Hartog (1970) contributed an excellent taxonomic study on seagrasses of the world, listing 49 species belonging to twelve genera and four families of angiosperms regarded as seagrasses. A brief taxonomic study of eastern Australian Zostera has been made by Jacobs and Williams (1980). A comprehensive taxonomy of temperate Australian seagrasses will be available soon (Womersley and Robinson, in preparation). Australia is well endowed with seagrasses; there are 25 described species belonging to all genera except the genus Phyllospadix represented in Australian waters (Table 1). Three of these species have only recently been described (Greenway, 1979; Cambridge and Kuo, 1979), and four new Posidonia species are to be described (Kuo and Cambridge, in preparation). Dehydrated specimens, which were collected as drift on the beach and deposited in herbaria, were used by den Hartog (1970). However, herbaria have recently received an increasing number of seagrass specimens taken from their natural habitats. Unfortunately, most herbaria have stored seagrasses in a dehydrated state and the author found it difficult to study the morphology of these dried seagrasses. * A paper presented at the Australian Seagrass Workshop, University of Sydney, 1981. Proc. LINN. Soc. N.S.W. 106 (3), (1982) 1983 226 BIOLOGY OF AUSTRALIAN SEAGRASSES TABLE 1 The Australian Seagrasses * 9) Endemic to~ Genus! Species Australia Distribution? Zosteraceae (Potamogetonaceae) Zostera capricornt Aschers. C Qld., NSW (12) mucronata den Hartog E S SA,WA muellert Irmisch & Aschers. E SSA, Tas, Vic. Heterozostera tasmanica (Martens ex Aschers.) E S SA, WA, Vic., (1) den Hartog NSW, Tas. Posidoniaceae (Potamogetonaceae) Posidonia angustifolia Cambridge & Kuo E S WA,SA (5) australis Hook F. E S NSW, Vic., Tas., SA, WA ostenfeldi den Hartog E S WA,SA sinuosa Cambridge & Kuo E S WA,SA Cymodoceaceae (Zannichelliaceae; Potamogetonaceae) Halodule pinifolia (Miki) den Hartog N Qld. (6) uninervis (Forsk.) Aschers. N WA, Qld. in Boissier Cymodocea angustata Ostenfeld E N WA (4) rotundata Ehrenb. & Hempr. N Qld. ex Aschers. serrulata (R.Br.) Aschers. & Magnus N Qld.,WA Syringodium isoetifolium (Aschers.) Dandy C WA, Qld. (2) Thalassodendron ciliatum (Forsk.) den Hartog N Qld. (2) pachyrhizum den Hartog E S WA Amphibolis antarctica (Labill.) Sonder et E S WA,SA, (2) Aschers. Vic., Tas. griffithu (J. M. Black) den Hartog E S WA,SA Hydrocharitaceae Enhalus acoroides (LL.F.) Royle N Qld. (1) Thalassia hemprichu (Ehrenb.) Aschers. N Qld. (2) Halophila decipiens Ostenfeld N Qld., NSW (9) ovalis (R.Br.) Hook F. C WA, Vic., NSW, Qld., NT ovata Gaud. in Freycin. N WA, Qld. spinulosa (R.Br.) Aschers. N WA, Qld. tricostata Greenway 1B: N Qld. 1. Number in (_) represents total world species in the genus. Only the genus Phyllospadix (5) is not represented in Australia. 2. E: species endemic to Australia. 3. N, occurring only in northern, tropical waters; S, occurring in southern, temperate waters; C, cosmopolitan, occurring in tropical and temperate waters. Qld., Queensland; NSW, New South Wales; Vic., Victoria; SA, South Australia; Tas., Tasmania; WA, Western Australia; NT, Northern Territory. Distribution A brief biogeography of Australian seagrasses has been discussed by Specht (1981). Australian seagrasses show a high degree of endemism as well as having a strong affinity with those of other continents and island groups. Half of the Australian species, and the genera Amphibolis and Heterozostera, are endemic to Australia, while some species, including Enhalus acoroides (L.f.) Royle, Thalassodendron ciliatum (Forsk.) Proc. LINN. Soc. N.S.W. 106 (3), (1982) 1983 J.KUO 227 Fig. 1. A typical temperate Australian seagrass Posidonia australis Hook F. Bd: Leaf blade; Sh: Leaf sheath; Fr: Fibres; Rt: Roots; Ep: Epiphytes. Note the rhizome is covered by the fibres. PRoc. LINN. Soc. N.S.W. 106 (3), (1982) 1983 228 BIOLOGY OF AUSTRALIAN SEAGRASSES den Hartog, Syringodium tsoetifolium (Aschers.) Dandy are also found elsewhere in the Indo-Pacific region. Halophila ovalis (R.Br.) Hook.f. has a much wider distribution. Within Australian waters, some genera (e.g. Enhalus, Cymodocea, Halodule, Thalassia and most species of Halophila) are found only in tropical areas, others (e.g. Poszdonza, Amphibolis, Heterozostera and most species of Zostera) occur only in temperate waters. On the other hand, Syringodium tsoetifolium and Halophila ovalis are found in both temperate and tropical waters. The latitudes 30°S on the east coast and 25°S on the west coast of Australia appear to be transition points for a great number of tropical and temperate seagrass communities (Larkum, 1977). No information is available at present on seagrasses of nothern Australia, from Port Hedland in Western Australia to the Gulf of Carpentaria in Queensland (McComb et al. , 1981). Habitat Different seagrasses grow on different substrates (den Hartog, 1970). Halophila is found in all kinds of habitats from open sea to estuaries and inlets. Cymodocea, Thalassia and Zostera can also be found on various substrates and the depth limit of most of the species of these genera is 10 to 12 metres. Syringodium is restricted usually to the upper sublittoral zone where it sometimes appears rarely to make up monospecific meadows. Enhalus, Amphibolis, Thalassodendron and Posidonia are also restricted to the upper sublittoral zone with some exceptions. In Western Australia, Cambridge (1975) found that Thalassodendron pachyrhizum den Hartog occurs only on limestone and granitic reefs and never on sands, and could be found to a depth of at least 40 metres. She also noticed that Amphibolis usually inhabits sand and sand-covered reefs as either dense meadows or small clumps, while Poszdonia ostenfeldi den Hartog is found in sandy substrates on unsheathed sand banks. Posidonia australis Hook. f. forms vast meadows in embayments enclosed by reefs, shoals and islands. P. australis appears to be quite tolerant of salinity fluctuations, as it has been reported in Shark Bay where salinity can be as high as oe (Cambridge, 1979, personal communication). The largest, densest seagrass beds are found in St Vincent and Spencer Gulfs of South Australia and in Shark Bay in Western Australia. The area of the latter embayment has been estimated at more than 10,000 km? (Davis, 1970). There is no doubt that the most extensive seagrass community in these beds is that formed by Posidonza spp. VEGETATIVE STRUCTURE AND FUNCTION Seagrasses show a high degree of uniformity in their morpological appearance; most genera have well-developed rhizomes and linear or strap-shaped leaves, Halophila being the only exception. However, the anatomical structure of the seagrasses exhibits a considerable degree of diversity. Some seagrass genera are monomorphic (i.e. with shoots bearing only one kind of leaf foliage leaves, Postdonia, Zostera, Heterozostera, Cymodocea, Halodule and Enhalus) and others are dimorphic (i.e. with two types of shoot- bearing scale leaves and foliage leaves respectively, Amphibolis, Thalassodendron, Syringodium, Halophila and Thalassia). Species of Halophila have shoots of variable length. A typical seagrass, P. australis (Fig. 1), has a creeping rhizome which at each node produces two branching roots and a unit of three foliage leaves protected by a sheath at the base. 1. LEAF BLADE (Figs 2-4). A transverse section of leaf blade (Fig. 2, A-G) reveals a thin cuticle covering a single cell layer of epidermis which lies above the parenchyma (mesophyll) or fibre cells (mechanical tissue). Numerous vascular bundles and air lacunae are embedded among the parenchyma cells. Proc. LINN. Soc. N.S.W. 106 (3), (1982) 1983 J. Kuo 229 Fig. 2. The leaf blade anatomy of seagrasses and the leaf epidermal morphology of some Australian Posidonia. A-G. Transverse sections of the leaf blade of A, Syringodium isoetifolium; B, Zostera muelleri; C, P. australis; D, Z. muellerti (SEM); E, Heterozostera tasmanica (SEM); F, P. ostenfeldu (SEM); G, Enhalus acoroides (SEM). A: Air-lacunae; E: Epidermis; V: Vascular bundles. H-J. The surface view of the Posidonia leaf epidermis (all same magnification). H, P. australis; I, P. angustifolia; J, P. sinuosa. Proc. LINN. Soc. N.S.W. 106 (3), (1982) 1983 230 BIOLOGY OF AUSTRALIAN SEAGRASSES Cuticle: There are no stomata in seagrasses (Doohan and Newcomb, 1976; Kuo, 1978) (Fig. 2, H-J). The appearance of the cuticle under electron microscopy differs among species. With the exception of P. sznuosa (Cambridge and Kuo, 1982) the genus Postdonia (Kuo, 1978; Fig. 3, D) has an unusual porous texture. The inner face of the cuticle has oval cavities in Zostera muellert Irmisch & Aschers and Heterozostera tasmanica (Martens ex Aschers.) den Hartog (Fig. 3, F). A similar structure has been observed in Z. capensis and it has been suggested that it might have an association with iron ab- sorption, secretion and storage (Barnabas et al. , 1977) or might even have an excretory function (Barnabas ét al. , 1980). The cuticle may appear as a thin, electron-transparent layer in P. sinuosa, Thalassia hemprichit (Ehrenb.) Aschers., Cymodocea serrulata (R.Br.) Aschers. & Magnus, Cymodocea rotundata Ehrenb. & Hempr. ex Aschers. and Syringodium isoetifolium (Fig. 3, E; Doohan and Newcomb, 1976). The cuticle has been reported as absent in the northern hemisphere Thalassia testudinum Banks ex Konig (Benedict and Scott, 1976). It is thought that the cuticle of the submerged leaves offers little resistance to carbon diffusion from surrounding water in the absence of a func- tional stomatal system. Epidermal Cells: The shape of the epidermal cells is sometimes a useful taxonomic character (Cambridge and Kuo, 1979; Fig. 2, H-J). The thickened walls of epidermal cells are rich in pectin and celluloses but are not lignified (Kuo, 1978). Epidermal transfer cells have been observed in Thalassia, Cymodocea, Halophila, Halodule, Zostera (Fig. 3, G) and Heterozostera (Jagel, 1973, 1982; Birch, 1974; Doohan and Newcomb, 1976; Barnabas et al., 1977; Kuo, unpublished) but are absent in Poszdonia (Fig. 3, H), Amphibolis, Syringodium and Thalassodendron (Kuo, 1978; unpublished). An unusual structure has been observed in Halophila epidermis (Birch, 1974). Epidermal cells contain most of the leaf chloroplasts as well as numerous mitochondria, Golgi bodies and microbodies (Doohan and Newcomb, 1976; Kuo, 1978; Fig. 3, G-H). Phenolic materials may also occur in the epidermal cells of certain species. The epidermal cells appear to have active photosynthetic and mitochondrial metabolisms associated with osmoregulation (Jagel, 1973; 1982). These cells may be the primary site of carbon dioxide fixation and carbon dioxide uptake. The products of photosynthesis in the epidermal cells may be transported to the mesophyll cells and to the vascular systems via plasmodesmata (Kuo, 1978). The !°C/!*C ratio, measured as 6 !°C values in the range - 3.0 to - 19.0%, have been recorded in seagrass leaves (Smith and Epstein, 1971; Doohan and Newcomb, 1976; Benedict and Scott, 1976; Andrews and Abel, 1979; McMillan e¢ al., 1980). These values are within the range usually associated with C, terrestrial plants that fix carbon dioxide during photosynthesis with phosphoenolpyruvate carboxylase (Smith and Epstein, 1971). The high variability in 6 '3C values might be associated with a variable photosynthesis metabolism in seagrasses (McMillan et al., 1980) and it has been thought that the C, pathway might operate in seagrass photosynthesis (Benedict and Scott, 1976). However, recent biochemical studies of four Australian tropical seagrasses by Andrews and Abel (1979) indicate that photosynthesis occurs via the C3 rather than the C, pathway. Studies of leaf anatomy, which do not show the ‘Kranz’ anatomy of terrestrial C, plants (Fig. 1, A-C) (Jagel, 1973; Doohan and Newcomb, 1976; Barnabas et al., 1977; Kuo, 1978; Tomlinson, 1980) and of photorespiration (Hough, 1976) suggested that seagrasses may not be typical C3 or Cy plants. At the pH of seawater, there is more bicarbonate and less carbon dioxide available in seawater than there is in the air. Thus, it is still uncertain whether seagrasses take up carbon in carbon dioxide or bicarbonate form from surrounding water for photosynthesis (Beer et al., 1977; Sand-Jensen, 1977). PROC. LINN. SOc. N.S.W. 106 (3), (1982) 1983 231 Fig. 3. The structure of the leaf epidermis and the cuticle of some Australian seagrasses. A. The portion of the leaf blade in P. ostenfeldi. Note that the epidermal cells (E) contain tannins and fibre cells (F) associated with epidermal and other hypodermal cells. B and C. A pronounced cuticle (arrows) covering the epidermal cells (E) of P. australis, sudan black and toluidine blue stained respectively. D-F. Electron micrographs of seagrass cuticle. D, P. australis cuticle (C) has a porous appearance. E, an electron-transparent cuticle (arrow) in Springodium isoetifollum. F, cup-shaped invaginations (arrows) occur in the outer epidermal wall (W) of Z. muellert. G. Electron micrograph of the epidermal cell of Z. muellert. Note that the cell has many chloroplasts; more cell wall ingrowths (asterisks) occur in the basal than the distal tangential walls. Some plasmadesma (arrow) are present between the epidermal cells. H. Electron micrograph of the epidermal cell of P. australis. Note that the cell also has many chloroplasts (CH) and a large nucleus (N) but is lacking cell wall ingrowths. Some plasmodesmata (arrow) also occur on the walls between cells. Proc. LINN. Soc. N.S.W. 106 (3), (1982) 1983 232 BIOLOGY OF AUSTRALIAN SEAGRASSES It is believed that the leaves of seagrasses have the capacity to take up nutrients from surrounding water (Arber, 1920; Sculthorpe, 1967; Raven, 1981). Using radioactive isotopes the uptake of carbon (Harrison, 1978; Barbour and Radosevich, 1979), phosphorus (McRoy and Barsdate, 1970), cadmium (Faraday and Churchill, 1979; Brinkhuis e¢ a/., 1980) and manganese (Brinkhuis et al., 1980) by the leaf has been demonstrated for several seagrasses. However, there are no similar studies on Australian seagrasses. Mesophyll Cells and Air-lacunae: Mesophyll cells of seagrasses are thin-walled but highly vacuolated. The thin peripheral cytoplasm contains few chloroplasts with small starch granules. The mesophyll cells surround air lacunae of varying size (Fig. 2, A-G) and in some genera (e.g. Zostera) the surface area of air lacunae may occupy half of a leaf surface area in cross section (Fig. 2, D). It has been shown that the air lacunae contain oxygen and carbon dioxide (Zieman, 1974). The storage of these gases in the air lacunae could swell the leaf blades of Thalassia testudinum up to 200-250% during the day (Zieman, 1974). The presence of air lacunae has been considered important in seagrass photosynthesis, particularly in re-fixation of carbon dioxide. Oxygen produced in photosynthesis could build up in these lacunae, and respiratory and photorespiratory carbon dioxide could diffuse into them as well. Zelitch (1971) estimated that up to 50-67% of photorespired carbon dioxide in T. testudinum leaves was recycled. Z. muellert grows in both intertidal and subtidal mudflats in Westernport Bay, Victoria. The surface area of air-lacunae in the leaves of the intertidal form is only about one-third that in the leaves of the subtidal form (Kuo, in preparation). This finding correlates well with the photosynthesis rate in this species. Clough and Attiwill (1980) found that the photosynthesis rate of the subtidal form is about three times that of the intertidal form. There are many septa interrupting the air-lacunae along the leaf. Each septum consists of a group of small parenchyma cells with minute pores at their intercellular spaces. Numerous external wall ingrowths projecting into these pores are present in the septa of Zostera muellert and Z. capricorni but are absent from those of more than twenty seagrass species belonging to twelve genera examined so far (Kuo, in preparation). Vascular System: In general, each vascular bundle is surrounded by a layer of sheath cells (Fig. 4, B, E), though Benedict and Scott (1976) could not distinguish sheath cells in Thalassia testudinum. The sheath cells of Posidonia have a thin but lignified wall (Kuo, 1978), those of Syringodium have suberin lamellae in the wall, those of Thalassodendron and Amphibolis have a thick but not lignified wall and those of Zostera and Heterozostera have cell wall ingrowths (Kuo, unpublished). The vascular sheath cells of seagrasses therefore appear to have the potential to regulate solute movement between the mesophyll and the vascular tissues, but this has never been demonstrated ex- perimentally. The number and size of xylem elements in the vascular bundles of seagrasses, as well as in other aquatic plants, are much reduced in comparison with terrestrial plants (Sculthorpe, 1967). Xylem walls in seagrasses have little lignification and secondary wall thickening (Kuo, 1978; Fig. 4, D, G). The reduced xylem has led some resear- chers to suggest that there is little xylem transport in seagrasses (e.g. Tomlinson, 1972), but precise experimental work on this point appears to be lacking. The structure of the phloem tissues in seagrasses is similar to that of terrestrial plants. Sieve tubes of Posidonia contain a plasmalemma, endoplasmic reticulum, mitochondria and plastids but lack nuclei (Kuo, 1978; Fig. 4, G). Companion cells are rich in cytoplasm. Sieve Proc. LINN. Soc. N.S.W. 106 (3), (1982) 1983 233 ; & ie Le 7Ou ’ : Fig. 4. The leaf vascular structure in some Australian seagrasses. A. A polarizing micrograph of P. australis leaf blade. Notice that the cell walls of the epidermal and fibre cells are intensively birefringent but that of vascular bundle (V) is not so. B. The sheath cells of vascular bundle (V) in P. sznuosa is not so intensively stained with PAS reaction. Fibre cells (arrows) occur between the epidermal (E) and the hypodermal cells. C.(SEM) and D. The vascular bundle (V) of Z. mueller’. Note that a single xylem lacunae (X) separates from the phloem tissue (P). Fibre cells associated with vascular tissue but not with the epidermal cells (E). E. Both the epidermal cell contents and the sheath cells of the vascular bundle (V) in P. australis leaf blade are strongly autofluorescent. F. Nacreous wall sieve elements (arrows) are found in the vascular bundles of H. tasmanica leaf sheath. G. Electron micrograph of the vascular bundle in P. australis leaf blade. Note that the bundle is surrounded by a layer of lignified sheath cells (B). The walls of xylem elements (X) are poorly secondary thickness and weakly lignified and that of sieve tubes (S) and of parenchyma cells are normal. H. Nacreous wall sieve tubes (N) and cell wall ingrowths in parenchyma cells (P) occur in the vascular bundle of Z. mueller: leaf blade. Proc. LINN. Soc. N.S.W. 106 (3), (1982) 1983 234 BIOLOGY OF AUSTRALIAN SEAGRASSES tubes of the family Zosteraceae and the sole genus Halodule of the family Cymodoceacae have nacreous or thick walls with reduced lumens and the companion cells of these species have cell wall ingrowths (Kuo, 1983; Fig. 4, F, H). The reduction of lumen area in nacreous sieve elements relates to the translocation rate and has received little attention in terrestrial plants (Esau and Cheadle, 1958). Mechanical Tissues: The prominent fibre cells (Fig. 3, A, C; Fig. 4, B) of the leaves are of particular interest. Mechanical tissues occur only in the leaves of certain genera, they may be associated with vascular bundles (e.g. Zostera) or distributed among parenchyma tissue (e.g. Postdonia). They have thickened walls, which consist of pectin and cellulose but do not contain lignin (Kuo, 1978). Thus they could provide tensile strength but retain a high degree of flexibility, allowing the leaves to withstand vigorous ocean wave action. The distribution of these fibre groups in the leaves of Postdonia can be used as a taxonomic character (Sauvageau, 1890; Cambridge and Kuo, 1979). 2. LEAF SHEATH (Fig. 5). The leaf sheath encloses the basal portion of the leaves and usually lies beneath the sediment surface (Fig. 5, A). The anatomy and cellular structures of the sheath of Posidonia differ markedly from the blade (Kuo, 1978; Fig. 5, B, C). The cuticle is electron-transparent and is not porous. A thin lignified or suberized layer is present on the outer wall of abaxial epidermal cells. This layer or the cuticle may act as a physical barrier between the inner developing leaf tissue and surrounding water. Tyerman (1979, personal communication) reported that the sheath may function to maintain a relatively low osmotic pressure for the leaves meristem of P. australis. In contrast to the blade the sheath epidermal cells of Poszdonza lack chloroplasts and are highly vacuolated with a thin peripheral cytoplasm. The structure of vascular bundles is similar to those in the blade. Groups of fibre bundles are widely distributed among the parenchyma tissues of the sheath (Kuo, 1978; Fig. 5, B) but in contrast to the fibre bundles of the blade, they are lignified and because of this they persist on the rhizomes long after the other tissues of the leaf sheath have rotted away (Fig. 5, A). The sheath fibres of Posidonia can be deposited both beneath living seagrass meadows and on ocean floors bordering the coast (Kuo and Cambridge, 1978). Deposits of these marine fibres in Spencer and St Vincent Gulfs in South Australia are present in such quantities that the fibres were harvested over a limited period (1905 to 1915) for grain bags, paper making and insulation material (Winterbottom, 1917; Reid and Smith, 1919). 3. STEM AND RHIZOME (Fig. 5). In transverse sections, both erect stem and rhizomes of Amphibolis, Zostera and Thalassodendron exhibit similar morphological features (Ducker et al., 1977; Fig. 5, D, E, G). The central stele has a central xylem surrounded by phloem bundles (Fig. 5, E). The stele is surrounded by the cortex in Poszdonia or by aerenchyma in Amphibolss, Thalassodendron and Zostera. Two distinct cortical zones can be recognized in Amphibolis and Zostera based upon morphological and histochemical characters. Starch grains are prominent in the cortical cells of Postdonia rhizomes, and groups of lignified fibre cells are scattered through the cortex of Postdonia rhizomes (Kuo. and Cambridge, 1978; Fig. 5, G). A few peripheral vascular strands are distributed among the cortical tissues (Fig. 5, E). The distribution and number of vascular strands appear to be a species characteristic (den Hartog, 1970; Ducker e¢ al., 1977). A distinct cuticle covers both stem and rhizome. Cell walls of the epidermal and hypodermal cells are thickened and lignified and sometimes these cells contain polyphenolic materials (Fig. 5, F). A few Proc. LINN. Soc. N.S.W. 106 (3), (1982) 1983 235 4 28s Fig. 5. The anatomy and the leaf sheaths and rhizomes in seagrasses. A. Old leaf sheaths of P. australis contain numerous fibres. B and C. Transverse sections of the leaf sheaths of P. australis and Z. muelleri respectively. Many air-lacunae (A) occur in both species but P. australis has more fibre bundles among the parenchyma tissue. V: vascular bundles. D. A scanning electron micrograph of a P. australis rhizome in a transverse view. E. A transverse section of Syringodium tsoetifolium rhizome shows numerous vascular bundles (V), air-lacunae (A) and large tannins-containing cells (T) but there are no fibre bundles in the rhizome. F. The cell walls of both the epidermal (E) and the hypodermal cells (H) of P. australis rhizome are autofluorescent but those of the cortical cells (C) are not. G. Transverse section of P. australis rhizome showing the central stele has a xylem element (X) surrounded by numerous phloem bundles (P), many large fibre bundles scatter among the cortical tissue which contains numerous minute starch granules. H. A higher magnification of a fibre bundle in P. australis rhizome showing the middle lamellae of the fibre cells are autofluorescent. Proc. LINN. Soc. N.S.W. 106 (3), (1982) 1983 236 BIOLOGY OF AUSTRALIAN SEAGRASSES bacteria and diatoms are often present on the surface of rhizomes and numerous epiphytic algae occur on the stem. 4. ROOTS (Fig. 6). Both rhizomes and roots serve as anchor systems for seagrasses. The external morphology of roots may vary between species and appears to be associated with their ecological habitat and mineral nutrient environment. For example, the roots of Halophila and Zostera are creamy in colour and soft, with many root hairs, and appear to be suitable for survival on soft substrates. Posdonia roots are thick but relatively soft with many branches; the root hairs are sparse, and the plants live on the sandy ocean floor (Figs 1, 6, A). Hard, wavy and branched roots are usually found in Amphibolis, and appear suitable to anchoring on harder substrates such as rocks or marine fibres. The roots of Thalassodendron are thick and strong, with shiny black surface roots, and have few branches. These roots apparently penetrate the interstices of the rocky substrates and anchor the plants against the turbulence of ocean waves. The anatomy of the roots, however, appears to be very similar among the species. All roots have a distinct root cap (Fig. 6, B). The central stele (Fig. 6, C-E) consists mainly of well developed phloem and a central weakly lignified xylem, and is surrounded by an endodermis which has a Casparian strip in the radial cell wall (Kuo and Cambridge, 1978; Fig. 6, H). Cortical cells are thin-walled and may have many large air-lacunae. Cell walls of the epidermal cells and three to four layers of hypodermal cells are thickened and lignified in harder roots, such as those of Thalassodendron, Amphibolis and Posidonia (Fig. 6, F). The thickened hypodermal walls may have suberin lamellae (Kuo and Cambridge, 1978; Fig. 6, 1). Nutrients are taken up by roots and rhizomes of seagrasses from the sediment and are transported to all parts of the plant. Directions and rates of nutrient transport are likely to be dependent upon the morphology, age and physical condition of the individual plants as well as the species. This translocation probably occurs in the xylem and phloem of the vascular system. However, McRoy and Barsdate (1970) showed that the leaves and stems of Zostera marina L. absorbed more phosphorus than did the roots and rhizomes. Wetzel and Penhale (1979) found the rates of carbon uptake and transport from roots and rhizomes to the leaves of Halodule wrghtii Aschers. are much higher than those for Zostera marina and Thalassia testudinum. Thus it appears that both leaves and roots of seagrasses may have the capacity to take up nutrients, although the quantitative role in nutrient uptake in a natural environment is still unknown. Data on nutrient uptake by Australian seagrasses are still not available. LIFE HISTORY Some seagrasses appear to show infrequent flowering. Cymodocea serrulata flowers have been recorded once, and these were male (Kirkman, 1975) while Tomlinson (1969) observed an average arrangement of male:female shoot ratio of 4:1 for Thalassza testudinum in Florida. Other species show regular flowering and some have the male and female flowers on the same shoot — e.g. Poszdonia australis whose flowers are initiated in autumn when day lengths are shortening and temperatures are falling, and the flowers open in spring (Cambridge, 1975). Aspects of floral morphology and development in seagrasses have been in- vestigated only for some members of the family Cymodoceaceae (Isaac, 1969; Kay, 1971; Ducker et al., 1978; Tomlinson and Posluszny, 1978). But this information is lacking for other families of seagrasses, probably due to the difficulty of obtaining material. Posidonia flowers (Fig. 7, A) are bisexual with a perianth. Mature pollen grains of most seagrass species are thread-like with a pair of fine curved hooks at the Proc. LINN. Soc. N.S.W. 106 (3), (1982) 1983 237 Fig. 6. The anatomy of P. australis roots. A. A scanning electron micrograph of a mature root. B. A median longitudinal section of a mature root tip with a pronounced root cap (RC). C. A transverse section of a mature root showing the stele is surrounded by a series of air-lacunae (A), many thin-walled cortical cells (C) and a layer of the thin-walled epidermal cells (E). D. A scanning electron micrograph of a mature root in a transverse view. E. A high magnification of a stele with a distinct layer of endodermis (EF). F. A high magnification of the root periphery showing that the thin-walled epidermal cells (E) with thicker- walled hypodermal cells (H). G. Numerous starch granules (arrows) are located in the basal portion of the root cap cells. H. A distinct Casparian strip (CS) is present in the radial walls of two adjacent endodermal cells. I. Electron dense suberin lamellae (arrows) are present in thickened walls (W) of the adjacent hypodermal cells. Proc. LINN. Soc. N.S.W. 106 (3), (1982) 1983 238 BIOLOGY OF AUSTRALIAN SEAGRASSES end (Pettitt and Jermy, 1975; Pettitt, 1976, 1980; Ducker et al., 1978; Fig. 7, B), probably for attachment to the stigma surface. In contrast to terrestrial flowering plants, the pollen of seagrasses lack an exine or it is poorly developed (Fig. 7, C, D). The pollen is released under water and drifts about submerged. The mechanism of seagrass pollination has been discussed elsewhere (Ducker and Knox, 1976; Ducker et al., 1978; Pettitt, 1980). Enzymatic properties of the pollen wall and stigma pellicle in seagrasses are comparable with those found in terrestrial flowering plants. Pettitt (1980) suggested a similar mechanism of cuticle erosion in both pollen grain and stigma might well follow compatible pollination both on land and in the sea. The fruits of Postdonta mature about three months after anthesis (Fig. 7, E, F). Cambridge (1975) estimated that P. australis in Cockburn Sound, Western Australia, may produce 500 fruits per square metre of meadow. Each seed is enclosed by a thin membrane which resembles the seed coat of terrestrial plants, and then by a pericarp which has a well-developed aerenchyma system. The bulk of the seed consists of en- dospermic tissue which serves as nutrient storage. Starch and lipid are the main nutrient reserves and storage protein is lacking. Nitrogen, phosphorus and other macro-and micro-elements occur in concentrations which are in general comparable to those of terrestrial plants (Hocking e¢ al. , 1980). Mature fruits with their pericarps are released in summer and float to the water surface. After a day or two the pericarp ruptures and the seeds sink to the floor of the sediment (Fig. 7, G). Germination may begin even before the seeds are released from the pericarp. The young seedlings (Fig. 7, H) may depend upon the nutrient storage of the seed up to nine months after germination (Hocking et al. , 1981). In Amphibolis and Thalassodendron the germination of the seed is viviparous. The flowers of these seagrasses occur on separate plants in summer. The fruits develop at the apices of the lower branches of female plants. Mature embryos burst through the apices of their pericarps and appear as little seedlings, which continue their growth on the parent plant for some time (Fig. 7, I). The seedlings, with cup-like pericarp structures, are released from the parent plants in winter and float in the ocean. Finally, the seedlings become anchored in the sediments or to the bases of plants in meadows by the comb-like lobes of the upper pericarp. Then the stems begin to grow rapidly and later shoots produce rhizomes with roots. Cambridge and Kuo (in press) describe an unusual form of vegetative propagation in Heterozostera tasmanica. Small propagules develop in the non-fertile erect stems of the parent plant during spring and summer. Eventually, erect stems carrying propagules break off from the parent plants and drift away. These propagules may then find a suitable location for further development. Although the rate of leaf production in an established seagrass meadow has been estimated (West and Larkum, 1979; Cambridge, 1981), little is known about either the rate of underground tissue production or the growth rate from a seedling to its maturity in any seagrass. Old leaves of Postdonia plants usually break off at the junction between the blade and sheath in autumn and leaf litter from seagrass may accumulate as high as two metres along some parts of the south-west Australian coastline during the winter. It is washed back into the ocean by late spring. There is no doubt that seagrasses are playing an important role in marine food chains. RHIZOME BRANCHING Despite the large numbers of seeds produced, seagrass seedlings have been in- frequently observed in the field and vegetative propagation by growth of rhizomes is probably of greater importance in the maintenance and spread of seagrasses than is seed production (Tomlinson, 1974). PROC. LINN. SOc. N.S.W. 106 (3), (1982) 1983 J.KUO 239 je SUH IU neers I ga fe eee i HOA) ey a) j 2 U2 6 US 16 17 VR Ya 2ID 2 22a 2 Fig. 7. Flowers, pollen grains, fruits and germination in seagrasses. A. Flowering in P. australis. B. Thread-like pollen grains of P. australis. C and D. Electron micrographs of pollen grains of P. australis showing that pollen has both sperm (SN) and vegetative (VN) nucleus but lacks exine on the walls. E. Fruits of P. australis. F. Seeds of Poszdonza (left: P. ostenfeldi1; centre: P. sinuosa; right: P. australis). G. Three germinating stages in P. australis. Note that there are no roots in the seedlings. H. Six-months old seedling of P. ostenfeldi. Note it bears many leaves and roots, however, the seed (arrow) appear still intact. I. A young (viviparous) seedling of Amphibolis antarctica showing a cup-shaped modified pericarp (arrow) attached to the stem of the parent plant. Proc. LINN. Soc. N.S.W. 106 (3), (1982) 1983 240 BIOLOGY OF AUSTRALIAN SEAGRASSES An excellent study on rhizome branching systems in seagrasses was made by Tomlinson (1974). He found rhizome branching in some genera (e.g. Amphibolis, Thalassodendron, Heterozostera) is sympodial and in others (e.g. Posidonia, Zostera, Cymodocea, etc.) is monopodial. The branching sequence may be (a) continuous (i.e. branch at every node, Syringodium, Halophila); (b) diffuse (a.e. branches produced at irregular intervals, Postdonta, Zostera, etc.) or (c) periodic (i.e. branches produced at regular intervals (Thalassza). In addition to relative uniformity of the marine en- vironment, the success of vegetative growth rather than propagation from seed in seagrasses could result in a low frequency of sexual recombination and, consequently, in a slow rate of evolutionary change. It is possibly for these reasons that Poszdonia oceanica (L.) Delile from the Mediterranean is similar to P. australis of Australia, and Thalassia testudinum of the Caribbean 1s closely related to 7. hemprichi of tropical Indo- Pacific regions. The species in both cases appear to have diverged relatively little since they were originally separated from one another in the Miocene (den Hartog, 1970). PHYLLOSPHERE AND RHIZOSPHERE Phyllosphere. Seagrass leaves bear varying numbers of epiphytes, depending upon age, season and environmental conditions. The epiphytes include bacteria, diatoms, algae (Fig. 8, A-E), and animals or ‘zoo-epiphytes’ such as hydroids (Fig. 8, F), sponges, etc. In general the older portions of the leaf support more epiphytes, and sometimes epiphytes may occlude half of the leaf area of a seagrass standing crop (Fig. 8, I). Even on the surface of newly emerged leaf blades there are numerous bacteria and diatoms (Fig. 8, A, B). As far as the abundance of epiphytic algae in Australian seagrasses is concerned, more than twenty species have been found on Poszdonia australis from Kangaroo Island, South Australia (Womersley, 1956); more than fifty species were found on Zostera capricornt Aschers., Heterozostera tasmanica (Martens ex Aschers.) den Hartog, and P. australis in Botany Bay and Jervis Bay, New South Wales (May et al. , 1978); Ducker et al. (1977) have recorded more than one hundred species of epiphytic algae on Amphibolis around Australia. The level of salinity may affect the growth of epiphytes. Den Hartog (1970) noted that P. australis from the high salinity areas such as the sea inlets in the Sydney area and Shark Bay in Western Australia have considerably fewer epiphytes than P. australis from an open sea. The attachment of epiphytes to the host appears to be limited to the cuticle of the host (Fig. 8, G) though Ducker and Knox (1978) have observed the red alga Heteroderma cymodoceae to erode locally the cuticle and outer wall of the epidermal cells of Amphibolis antarctica stems. Epiphytes could reduce seagrass photosynthetic rates by acting both as a barrier to carbon dioxide uptake and by reducing light intensity (Sand-Jensen, 1977; Cam- bridge, 1979). On the other hand, seagrasses may be able to reduce their epiphyte stands by producing new photosynthetic tissues, or perhaps by excreting toxic sub- stances (Sand-Jensen, 1977). The transfer of nutrients between epiphytic algae and Thalassia testudinum has been demonstrated (Goering and Parker, 1972; Harlin, 1973; McRoy and Goering, 1974). Rhizosphere. The roots of seagrasses could also provide an environment for micro- organisms. The micro-organisms in the rhizosphere may play an important role in aiding nutrient uptake, affecting pathogen invasion and in nitrogen fixation. Numerous bacteria and other micro-organisms have been found in the rhizospheres of several seagrass species (Kuo et al., 1981; Fig. 8, H), and appear to penetrate only the periphery of the host tissue. However, it is still not known whether those micro- organisms are able to fix nitrogen. Patriquin (1972) hypothesized that bacterial PROC. LINN. SOc. N.S.W. 106 (3), (1982) 1983 J. KUO 241 eS ba ‘oe Fig. 8. Epiphytes and rhizosphere in seagrasses. A-D. Scanning electron micrographs show various epiphytes associated with the leaf blade of P. ostenfeldit. A, numerous bacteria; B, bacteria and diatoms; C, bacteria, diatoms and calcareous red algae; D, bacteria, diatoms and unknown organisms had damaged the epidermal surface. E. A fungal hyphae-like structure on the leaf blade of P. australis. G. A transverse section of P. sinuosa leaf blade showing numerous epiphytes are present on the leaf surface (V: vascular bundles; E: epidermal cells). H. Electron micrograph of P. australis rhizosphere showing some bacteria located on the outer surface of root epidermal cells (E). I. Numerous algae epiphytes (Ep) in association with the leaf blades of a P. australis plant. Proc. LINN. Soc. N.S.W. 106 (3), (1982) 1983 242 BIOLOGY OF AUSTRALIAN SEAGRASSES nitrogen fixation probably operated in seagrass rhizospheres, and Patriquin and Knowles (1972) provided experimental support. However, McRoy et al. (1973) could barely detect C.H, reduction associated with sediments, roots, rhizomes or leaves of Thalassia testudinum. Capone et al. (1979) found that dead leaves of T. testudinum have only one third of the nitrogen concentration of green leaves, and they estimated bacterial nitrogen fixation in the sediments supplied up to a quarter to half of the nitrogen required for leaf production. CONCLUDING REMARKS Seagrasses are abundant in Australian waters and many of them are endemic. Investigation of the biology of Australian seagrasses is still in an early stage. Natural habitats of seagrasses should be more carefully studied. Seagrasses have yet to be collected from the north-west of this continent to map Australian species completely. A critical but comprehensive taxonomic study of Australian seagrasses, particularly in the genera Halophila, Posidonia and Zostera is urgently needed. The storage of seagrass specimens in liquid as well as dehydrated states in herbaria is highly desired. Some structures of seagrasses are similar to those in terrestrial plants, while others are specialized adaptations in a marine environment. The development of various vegetative and reproductive structures as well as germination in seagrasses is not well understood. A combined biochemical, physiological and structural study on seagrasses in relation to nutrient uptake, photosynthesis, translocation and transportation is required. The relationship between epiphytes, zoo-epiphytes, rhizosphere and seagrasses also requires more attention. ACKNOWLEDGEMENTS I acknowledge with appreciation the continuing fruitful discussions with Dr M. L. Cambridge and many encouragements and interest in the study by Dr A. W. D. Larkum. Dr L. K. Abbott, Dr A. W. D. Larkum and Professor A. J. McComb made helpful comments on the manuscript. Special thanks go to Drs K. Abel, J. W. An- drews, M. L. Cambridge, H. Kirkman, S. V. Lewis, E. Nazaro and J. Stewart for their supply of seagrass specimens. Many thanks to Mr S. Blackburn and Mr G. Lane for their photographic assistance and to Mrs M. Sedlak for her patience and skill typing many drafts of this manuscript. This study was supported by funding from the Australian Research Grants Committee and the Australian Marine Sciences and Technology Advisory Committee. References ANpDREWS, T. J., and ABEL, K. M., 1979 — Photosynthetic carbon metabolism in seagrasses, !*C labelling evidence for the C3 pathway. Plant physiol. 63: 650-656. ARBER, A., 1920 — Water Plants. A study of Aquatic Angiosperms. Cambridge: Cambridge Univ. Press. BARBOUR, M. G., and RADOSLEVICK, S. R., 1979 —!4C uptake by the marine angiosperm Phyllospadix scoulert. Amer. J. Bot. 66: 301-306. BARNABAS, A. D., BUTLER, V., and STEINKE, T. D., 1977 — Zostera capensis Setchell. I. Observations on the fine structure of the leaf epidermis. Z. Pflanzenphysiol. 85: 417-427. , BUTLER, V., and STEINKE, T., 1980 — Zostera capensis Setchell. Il. Fine structure of the cavities in the wall of leaf blade epidermal cells. Z. Pflanzenphysiol. 99: 95-103. BEER, S., ESCHEL, A., and WAISEL,, Y, 1977 — Carbon. metabolism of seagrasses. I. The utilization of exogenous inorganic carbon species in photosynthesis. J. Exp. Bot. 28: 1180-1187. BENEDICT, C. R., and SCOTT, J. R., 1976 — Photosynthetic carbon metabolism of a marine grass. Plant Phystol. 57: 876-880. BIRCH, W. R., 1974 — The unusual epidermis of the marine angiosperm Halophila. Thou. Flora (Jena) 163: 410-414. Proc. LINN. Soc. N.S.W. 106 (3), (1982) 1983 J.KUO 243 ——,, 1981 — Morphology of germinating seeds of the seagrass Halophila spinulosa (R.Br.) Aschers. (Hydrocharitaceae). Aquatic Botany. 17: 79-90. BRINKHUIS, B. H., PENELLO, W. F., and CHURCHILL, A. C., 1980 — Cadmium and manganese flux in eelgrass Zostera marina. 11. Metal uptake by leaf and root-rhizome tissue. Mar. Biol. 58: 187-196. CAMBRIDGE, M. L., 1975 — Seagrasses of southwestern Australia with special reference to the ecology of Posidonia australia Hook F . in a polluted environment. Aquatic Bot. 1: 149-161. ——, 1979 — Cockburn Sound Environmental Study. Technical Report on Seagrass. Dept. Conservation & Environment, Western Australia. Report No. 7. ——, 1981 — Ecological studies on seagrasses of South Western Australia with particular reference to Cockburn Sound. Nedlands: University of Western Australia, Ph.D. thesis, unpubl. , and Kuo, J., 1979 — Two new species of seagrasses from Australia, Posidonia sinuosa and P. angustifolia (Posidoniaceae). Aquatic Bot. 6: 307-328. , and , (in press) — An unusual method of vegetative propagation in Australian Zosteraceae. Aquatic Bot. , and , (1982) — Morphology, anatomy and histochemistry of the Australian seagrasses of the genus Posidonia Konig (Posidoniaceae). III. Poszdonta sinuosa Cambridge & Kuo. Aquatic Bot. 14: 1-14. CLOUGH, B. F., and ATTIWILL, P. M., 1980 — Primary productivity of Zostera muellerit Irmisch ex Aschers. in Westernport Bay, (Victoria, Australia). Aquatic Bot. 9: 1-13. Davies, G. R., 1970 — Carbonate bank sedimentation, Eastern Shark Bay; Western Australia. Memoir 13, American Association of Petroleum Geologists: 85-168. Doonan, M. E., and Newco, E. H., 1976 — Leaf ultrastructure and 6!°C values of three seagrasses from the Great Barrier Reef. Aust. J. Plant Physiol. 3: 9-23. DUCKER, S. C., and KNOx, R. B., 1976 — Submarine pollination in seagrass. Nature (Lond.) 263: 705-706. , and , 1978 — Allelloparasitism between a seagrass and algae. Naturwiss. 65: 391-392. —, FoorD, N. J., and KNOx, R. B., 1977 — Biology of Australian seagrasses: The genus Amphibolis C. Agardh (Cymodoceaceae). Aust. J. Bot. 25: 67-95. ——, PETTITT, J. M., and Knox, R. B., 1978 — Biology of Australian seagrasses: pollen development and submarine pollination in Amphibolis antarctica and Thalassodendron ciliatum (Cymodoceaceae). Aust. J. Bot. 26: 265-285. ESAU, K., and CHEADLE, V. I., 1958 — Wall thickening in sieve elements. Proc. Acad. Sci. U.S.A. 44: 546- 553. FARADAY, W. E., and CHURCHILL, A. C., 1979 — Uptake of cadmium by the eelgrass Zostera marina. Mar. Biol. 53: 293-298. GOERING, J. J., and PARKER, P. L., 1972 — Nitrogen fixation by epiphytes on seagrasses. Limnol. Oceanogr. 17: 320-323. GREENWAY, M., 1979 — Halophila tricostata (Hydrocharitaceae), a new species of seagrass from the Great Barrier Reef region. Aquatic Bot. 7: 67-70. HARLIN, M. M., 1973 — Transfer of products between epiphytic marine algae and host plants. J. Phytol. 9: 243-248. HARRISON, P. G., 1978 — Patterns of uptake and translocation of '*C by Zostera americana den Hartog in the laboratory. Aquatic Bot. 5: 193-197. DEN HARTOG, C., 1970 — The Seagrasses of the World. Amsterdam: North Holland. DEN HARTOG, C., 1977 — Structure, function and classification in seagrass communities. In MCRoy, C. P., and HELFFERICH, C., (eds). Seagrass ecosystems, pp. 89-121. New York: Dekker. HOCKING, P. J., CAMBRIDGE, M. L., and McComs, A. J., 1980 — Nutrient accumulation in the fruits of two species of seagrass, Postdonia australis and P. sinuosa. Ann. Bot. 45: 149-161. ===, , and , 1981 — The nitrogen and phosphorus nutrition of developing plants of two seagrasses. Posidonia australis and Posidonia sinuosa. Aquatic Bot. 11: 245-261. Houcu, R. A., 1976 — Light and dark respiration and release of organic carbon in marine macrophytes of the Great Barrier Reef region. Aust. J. Plant Physiol. 3: 63-68. Isaacs, F. M., 1969 — Floral structure and germination in Cymodocea ciliata. Phytomorphology 19: 44-51. Jacoss, S. W. L., and WILLIAMS, A., 1980 — Notes on the genus Zostera s. lat. in New South Wales. Telopea 1: 451-455. JAGEL, R. H., 1973 — Studies of a marine grass Thalassia testudinum. I. Ultrastructure of osmoregulatory leaf cells. Amer. J. Bot. 60: 1003-1009. ——, 1982 — Further evidence for osmoregulation in epidermal leaf cells of seagrasses. Amer. J. Bot. (in press). Kay, Q. ae ., 1971 — Floral structure in the marine angiosperms Cymodocea serrulata and Thalassodendron ciliatum (Cymodocea ciliata). Bot. J. Linnean Soc. 64: 423-429. KING, R. J., 1981 — Marine angiosperms: seagrasses. In, Marine Botany: an Australian Perspective, CLAYTON, M.N., and KING, R. J., (eds), pp. 200-210. Melbourne: Longman Cheshire. Proc. LINN. Soc. N.S.W. 106 (3), (1982) 1983 244 BIOLOGY OF AUSTRALIAN SEAGRASSES KIRKMAN, H., 1975 — Male floral structure in the marine angiosperm Cymodocea serrulata (R.Br.). Aschers. and Mangus. Bot. J. Linn. Soc. 70: 267-268. Kuo, J., 1978 — Morphology, anatomy and histochemistry of the Australian seagrasses of the genus Posidonia Konig (Posidoniaceae). I. Leaf blade and leaf sheath of Poszdonza australis Hook F. Aquatic Bot. 5: 171-190. ——, 1983 — The nacreous walls of sieve elements in seagrasses. Amer. J. Bot. 70: 159-164. , and CAMBRIDGE, M. L., 1978 — Morphology, anatomy and histochemistry of the Australian seagrasses of the genus Posidonia Konig (Posidoniaceae). II. Rhizome and root of Poszdonia australis Hook f. Aquatic Bot. 5: 191-206. —, McComs, A. J., and CAMBRIDGE, M. L., 1981 — Ultrastructure of the seagrass rhizosphere. New Phytol. 89: 139-143. LARKUM, A. W. D., 1977 — Recent research on seagrass communities in Australia. In MCROy, C. P., and HELFFERICH, C.., (eds). Seagrass Ecosystems. pp. 247-262. New York: Dekker. May, V., COLLINS, A. J., and COLLETT, L. C., 1978 — A comparative study of epiphytic algal com- munities on two common genera of sea-grasses in eastern Australia. Aust. J. Ecol. 3: 91-104. McComs, A. J., CAMBRIDGE, M. L., KIRKMAN, H., and Kuo, J., 1981 — Biology of Australian seagrasses. In, PATE, J. S., and McComs, A. J.: The Biology of Australian Plants. pp. 258-293. Nedlands: Univ. of Western Australia Press. McMILLAN, C., PARKER, P. L., and Fry, B., 1980 — !°C/'?C ratios in seagrasses. Aquatic Bot. 9: 237-249. McRoy, C., P., and BARSDATE, R. J., 1970 — Phosphate absorption in eelgrass. Limnol. Oceanogr. 15: 6-13. , and GOERING, J. J., 1974 — Nutrient transfer between seagrass Zostera marina and its epiphytes. Nature 248: 173-174. : , and CHANEY, B., 1973 — Nitrogen fixation associated with seagrasses. Limnol. & Oceanogr. 18: 998-1002. , and HELFFERICH, C., 1977 — Seagrass Ecosystems. New York: Dekker. PATRIQUIN, D., 1972 — The origin of nitrogen and phosphorus for growth of the marine angiosperm Thalassia testudinum. Mar. Biol. 15: 35-46. , and KNOWLES, R. G., 1972 — Nitrogen fixation in the rhizosphere of marine angiosperms. Mar. Biol. 16: 48-58. PETTITT, J. M., 1976 — Pollen wall and stigma surface in the marine angiosperms Thalassia and Thalassodendron. Micron 7: 21-32. ——, 1980 — Reproduction in seagrasses: Nature of the pollen and receptive surface of the stigma in the Hydrocharitaceae. Ann. Bot. 45: 257-271. , and JERMY, A. C., 1975 — Pollen in hydrophilous angiosperms. Micron 5: 377-405. PHILLIPS, R. C., and McRoy, C. P., 1980 — A Handbook of Seagrass Biology. New York: Garland STAM Press. RAVEN, J. A., 1981 — Nutrition strategies of submerged benthic plants: the acquisition of C, N. and P by rhizophytes and halophytes. New Phytol. 88: 1-30. REID, J., and SMITH, H. G., 1919 — An investigation of the ‘marine fibre’ of Poszdonza australis. Comm. Aust. Inst. Sci. Ind. Bull. 14. Melbourne: Government Printer: 60 pp. SAND-JENSEN, K., 1977 — Effect of epiphytes on eelgrass photosynthesis. Aquatic Bot. 3: 55-63. SAUVAGEAU, C., 1980 — Observations sur la structure des feuilles des plantes aquatiques. J. Bot. 4: 221- 229, 237-245. SCULTHORPE, C. D., 1967 — The Biology of Aquatic Vascular Plants. London: Edward Arnold. SMITH, B. N., and EpsTeIN, S., 1971 — The categories of !°C/!2C for high plants. Plant Physiol. 47: 380-384. SPECHT, R. L., 1981 — Biogeography of halophytic angiosperms (salt-marsh, mangrove and sea-grass). In, KEAST, R., (ed), Ecological Biogeography of Australia, pp. 577-589. The Hague, Boston, London: W. Junk. Sarit TOMLINSON, P. B., 1969 — On the morphology and anatomy of turtle grass. Thalassia testudinum (Hydrocharitaceae). II. Anatomy and development of the root in relation to function. Bull. Mar. Scz. 19: 57-71. ——, 1972 — On the morphology and anatomy of turtle grass Thalassia testudinum (Hydrocharitaceae). IV. Leaf anatomy and development. Bull. Mar. Sct. 22: 75-92. ——., 1974 — Vegetative morphology and meristem dependence. The foundation of productivity in seagrasses. Aquaculture 4: 107-130. ——., 1980 — Leaf morphology and anatomy in seagrasses. In, PHILLIPS, R. C., and McRoy, C. P., (eds), A Handbook of Seagrass Biology, pp. 7-28. New York: Garland STAM Press. , and PosLuszNy, U., 1978 — Aspects of floral morphology and development in the seagrass Syringodium filiforme (Cymodoceaceae). Bot. Gaz. 139: 333-345. WEST, R. J., and LARKUM, A. W. D., 1979 — Leaf productivity of the seagrass Poszdonia australis, in eastern Australian waters. Aquatic Bot. 7: 57-65. PROC. LINN. SOc. N.S.W. 106 (3), (1982) 1983 J.KUO 245 WETZEL, R. G., and PENHALE, P. A., 1979 — Transport of carbon and excretion of dissolved organic carbon by leaves and root/rhizomes in seagrasses and their epiphytes. Aquatic Bot. 6: 149-158. WINTERBOTTOM, D. C., 1917 — Marine fibre. Dept. Chemistry Bull. No. 4, South Australia, 36 pp. WOMERSLEY, H. B. S., 1956 — The marine algae of Kangaroo Island. IV. The algal ecology of American River inlet. Aust. J. Mar. Freshwater Res. 7: 64-87. ZELITCH, I., 1971 — Photosyntheses and Photorespiration and Plant Productivity. New York: Academic Press. ZIEMAN, J. C., 1974 — Methods for the study of the growth and production of turtle grass, Thalassia testudinum Konig. Aquaculture 4: 139-143. Proc. LINN. Soc. N.S.W. 106 (3), (1982) 1983 The ‘Taxonomy of the semi-communal Spiders commonly referred to the Species Jxeutzcus candidus (1. Koch) with Notes on the Genera Phryganoporus, Ixeuticus and Badumna (Araneae, Amaurobioidea) M. R. GRAY Gray, M. R. The taxonomy of the semi-communal spiders commonly referred to the species Ixeuticus candidus (LL. Koch) with notes on the genera Phryganoporus, Ixeuticus and Badumna (Araneae, Amaurobioidea). Proc. Linn. Soc. N.S.W. 106 (3), (1982) 1983: 247-261. Phryganoporus Simon 1908 is synonymized with Badumna Thorell 1890. The synonymy of Jxeutzcus Dalmas 1917 with Badumna is supported. Spiders previously placed in the genus Phryganoporus or the species J. candidus (L. Koch) are referred to either B. candida (L. Koch), B. gausapata (Simon) n. comb. or B. vandiemeni n. sp.; these three species form the candida species group of the genus Badumna. Biological notes are given. M. R. Gray, Department of Arachnology, The Australian Museum, P.O. Box A285, Sydney South, Australia 2001; manuscript recewed I September 1981, accepted for publication in revised form 17 February 1982. INTRODUCTION Several authors have considered the taxonomic status of the semi-social amaurobioid spiders described by L. Koch (1872) as Amaurobius candidus from Queensland and Simon (1908) as the genus Phryganoporus from Victoria and Western _ Australia. However, at present these spiders are usually referred to the genus Jxeuticus as a single species, J. candidus (L. Koch). They are widely distributed in open forest and woodland habitats where their communal webs are usually built in low tree or shrub foliage. Occasionally, heavy infestations occur in orchards or shrubland pastures; the extensive webbing can severely inhibit foliage growth and prevent grazing. TAXONOMIC BACKGROUND Three species were originally recognized within Phryganoporus Simon 1908: P. gausapatus (Simon 1906) from Victoria (originally described in Amaurobius C. Koch) with a subspecies, P. g. occidentalis Simon 1908 from Western Australia; P. nigrinus Simon 1908 and P. tubicola Simon 1908 from Western Australia. Amaurobius candidus L. Koch 1872 had previously been described from Queensland and was later transferred to Ixeuticus Dalmas 1917 by Roewer (1954). This placement has been followed by most subsequent authors (McKeown, 1963; Dondale, 1966; Hickman, 1967; Main, 1971, 1976). Dondale (1966) redescribed J. candidus but the description and figures given were of a related species J. martius (Simon 1899). As a result of this confusion Leech (1971, 1972) suggested that J. martius may be a junior synonym of J. candidus. In fact, as previously noted by Lehtinen (1967), J. martius is a junior synonym of Amaurobius longinquus (L. Koch 1872) also referred to Ixeuticus by Dalmas (1917). Lehtinen (1967) retained the genus Phryganoporus to which he transferred J. can- didus to create the new combination P. candidus (L. Koch). Lehtinen also synonymized Proc. LINN. Soc. N.S.W. 106 (3), (1982) 1983 248 SEMI-COMMUNAL SPIDERS two of Simon’s species, P. gausapatus and P. nigrinus, with P. candidus but retained P. tubicola Simon. Subsequently, Main (1971) synonymized all of the species formerly described in Phryganoporus into a single widely distributed species, J. candidus. THE GENERIC PROBLEM Lehtinen (1967) synonymized Ixeuticus Dalmas 1917 (type species A. martius Simon 1899 described from New Zealand but probably introduced from eastern Australia) with Badumna Thorell 1890 (type species B. Azrsuta Thorell 1890 from Java, Indonesia). Subsequent authors have not followed Lehtinen (Forster, 1970; Main, 1971, 1976; Leech, 1971, 1972). Only Leech (1972) supported his rejection of the synonymy, arguing first that J. martius lacks functional tarsal claws on the female palps whereas B. hirsuta has definite, pectinate palpal claws as noted by Thorell (1890) and second that the anterior median eyes of his J. martius were largest whereas Thorell indicated that the anterior and posterior median eyes of B. hirsuta were subequal. However, all females of J. longinquus (syn. I. martius) examined by me possess well- developed palpal tarsal claws with seven to ten pectinations; and while the A.M.E. were usually the largest, this was not the case in all of the specimens examined or in other species of Jxeuticus. Kulczynski (1908) provides an excellent description and figures of B. hirsuta. These indicate clearly that the genitalic characteristics (male palp and female epigynum) of this species are extremely similar to those of J. longinquus. Consequently, Lehtinen’s synonymy of Jxeuticus with Badumna seems entirely valid. One result of this has been a change in the name of a very common and widely distributed species, the black house spider. Previously Jxeutscus robustus (L. Koch), Lehtinen (1967) synonymized it with Amaurobius insignis L. Koch 1872, so creating the new combination Badumna insignis (L. Koch). Main (1971) noted the inadequacy of the characters used by Simon (1908) as a basis for maintaining the separation of Phryganoporus and Ixeuticus. By placing all species of Phryganoporus into I. candidus Main effectively synonymized the two genera but retained the junior synonym, Jxeutzcus. ‘The position of Badumna was not considered here. In contrast, Lehtinen (1967) retained Phryganoporus while recognizing the synonymy of Jxeuticus and Badumna. His criteria for the retention of Phryganoporus relate mainly to spination and the structure of the male palp. However, specimens examined by me do not support the spination differences cited; representatives of both Phryganoporus and Badumna (sensu Lehtinen) have a rather constant pattern of ventral metatarsal spination of 221, whereas ventral tibial spination varies widely in Phryganoporus (010 to 222) and is not an adequate generic character. The male palpal tibia of Phryganoporus is stated by Lehtinen to possess a single basodorsal ( = retrodorsal basal) process. However, there is also a retroventral process which is equivalent to that present in Badumna species. Two or three retrolateral to retrodorsal palpal tibial processes are commonly present in Badumna. These may be placed apically as in B. longinquus or basally as in B. inornata Simon. Lehtinen also cites the basally protruding tegulum of the Phryganoporus male palp as a generic character; however, this feature simply represents the accentuation of a character trend already apparent in Badumna species. Unlike Badumna, Phryganoporus may possess a patellar process but this is not significantly developed in all species. Similarities in male palpal morphology between Phryganoporus and Badumna are readily apparent. Both possess an S-shaped, spiniform embolus, a distally tapering, marginally folded conductor and a spoon-shaped, membranous median apophysis. The female genitalia also share a common pattern consisting of an anterior fossa Proc. LINN. Soc. N.S.W. 106 (3), (1982) 1983 M.R.GRAY 249 bounded posteriorly by a prominent, transverse ridge with lateral teeth present. The internal genitalia of Phryganoporus species show a consistent pattern that is very similar to that of B. longinquus (Fig. 20). Both genera share a similar tarsal organ and trichobothrial plate morphology (Figs 23-28). Both also possess a complex, strongly branched tracheal system confined to the abdomen. This presents a marked contrast with another related, but distinct genus, Forsterina Lehtinen, which possesses a simple (unbranched) tracheal system. The latter finding is of wider interest because it suggests that for the Australian fauna this character may be usable at generic level only, whereas it has been used at the super- family level by Forster (1970) and Forster and Wilton (1973) to separate their Dic- tynioidea and Amaurobioidea. One anomaly evident from this is their placement of the related genera Jxeuticus and Reinga Forster, a close relative of Forsterina, into different superfamilies. In summary, the features used by Lehtinen to maintain Phryganoporus seem inadequate to justify its separation from the large and rather variable genus Badumna. The only unequivocal character at present available to Phryganoporus is its possession of a single retrodorsal tibial process on the male palp compared to two or three in Badumna. However, the tibial processes seem to be rather labile characters, both in shape and number, in these spiders. Behavioural traits may be valid generic characters but the social behaviour shown by members of Phryganoporus does not seem of special significance — social tendencies are apparent in members of Badumna also e.g. B. socialis (Rainbow). Consequently, I think ‘it justifiable to consider Phryganoporus Simon, like Ixeuticus, to be a junior synonym of Badumna Thorell. A generic revision of Badumna would certainly require the delineation of either subgeneric or species group categories. Here, the spiders formerly placed within Jxeutecus candidus (L. Koch) or Phryganoporus Simon are regarded as forming the candida species group within the genus Badumna. Badumna Thorell Badumna Thorell 1890: 322. Type species: Badumna hirsuta Thorell 1890. Phryganoporus Simon 1908. N. syn. Type species: Amaurobius gausapatus Simon 1906 Ixeuticus Dalmas 1917. Type species: Amaurobius martius Simon 1899 Badumna candida species group Three species are recognized here: B. candida (L. Koch), a widespread, variable species found in Queensland, New South Wales and South and Western Australia. B. gausapata (Simon) from southeastern Australia; and B. vandiement n. sp. from Tasmania. Diagnosis Medium sized (carapace length 2.4 to 4.1 mm) cribellate spiders which live in both communal and solitary webs. Carapace silvery brown in colour, white hairs abundant. Abdomen light brown with a dark brown mid-dorsal stripe followed by several light to dark brown chevron markings with white hair tufts laterally. Legs banded brown and grey. Anterior median eyes or anterior lateral eyes largest. Cheliceral teeth, retrolateral 2-4, prolateral 3-5. Cymbium large, broad. Embolus of male palp sinuously curved (S-shaped), proximal part of tegulum strongly protuberant basally. Male palpal tibia with a retrodorsal basal and a retrolateral ventral process; Proc. LINN. Soc. N.S.W. 106 (3), (1982) 1983 250 SEMI-COMMUNAL SPIDERS dorsal patellar process well developed or rudimentary to absent. Epigynum with a prominent, subdistal, transverse ridge posterior to an unpaired fossa; lateral teeth distal to subdistal. Tracheal system complex, confined to abdomen. Repositories: Australian Museum (A.M.); Queensland Museum (Q.M.); Tasmanian Museum and Art Gallery (T.M.); Zoologische Museum, Hamburg (Z.M.H.); Museum National d’Histoire Naturelle, Paris (M.N.H.N.); Australian National Insect Collection (A.N.I.C.). Badumna candida (1. Koch), new comb. Figs 1-11, 23, 35-37 Amaurobius candidus L. Koch 1872 Phryganoporus gausapatus occidentalis Simon 1908 Phryganoporus nigrinus Simon 1908 Phryganoporus tubicola Simon 1908; Lehtinen 1967 Ixeuticus candidus Roewer 1954; Main 1971 Phryganoporus candidus Lehtinen 1967 Diagnosis Patellar process on male palp rudimentary to absent; median apophysis directed apico-laterally. Lateral teeth of epigynum distal; fossa widest in central to posterior half. Cheliceral teeth, retrolateral 2-4 and prolateral 3-5. Ventral tibial spination, first leg 010-222. Metatarsal trichobothria; first leg 4-5. MALE (S 144, O.M.) Measurements (mm): Body length 5.68. Carapace length 2.58, width 1.90. Abdomen length 3.10, width 2.11. Colour: Carapace silvery brownish-grey with numerous white hairs; brown patches lateral and posterior to the fovea. Chelicerae and sternum dark brown, the sternum with dark brown hairs only. Legs with silvery grey and brown bands; ventral surfaces of coxae with white hairs. Abdomen light brown with a broad, dark brown mid-dorsal stripe, paler centrally and less than half as long as abdomen, bordered by lateral patches of white hairs. Immediately posterior to this stripe is a light brown patch partly delimited anterolaterally and posterolaterally by four dark brown spots. Behind this is a row of five to six dark brown chevron markings, paler brown centrally, which are separated from each other by thin lines of white hairs which form white hair tufts laterally. Dark brown flecks are present on the lateral abdomen particularly lateroventrally where they form a more or less distinct longitudinal dark line. This is separated on each side by a moderately broad line of white pigment from a broad, dark brown midventral stripe running between the epigastric fold and the spinnerets. Carapace: Longer than wide in ratio 1:0.74. Clypeus height about 1.5 times the diameter of an A.M.E. Cephalic area well developed, fovea a narrow slit. Eyes: A.L.E. > P.L.E. > A.M.E. > P.M.E. in ratio 1:0.86:0.73:0.71. Interdistance ratios, A.M.E. - A.M.E. 0.50: A.M.E. - A.L.E. 0.69: A.L.E. - P.L.E. 0.27: P.L.E. Figs 1-11. Badumna candida. 1-3 male palp; 1, retrolateral; 2, ventral; 3, prolateral. 4-7, female genitalia: 4, epigynum, 9, internal genitalia (Girraween N.P., Qld); 6, epigynum, 7, internal genitalia (Brookton, W.A.). 8, female, dorsal. 9, male abdomen, dorsal (dark colour morph). 10-11, male palp (TYPE of P. tubicola): 10, ventral: 11, tibia and patella, retrolateral. Scale lines 0.2 mm. PROG. LINN. Soc. N.S.W. 106 (3), (1982) 1983 Proc. LINN. Soc. N.S.W. 106 (3), (1982) 1983 252 SEMI-COMMUNAL SPIDERS - P.M.E. 1:P.M.E. - P.M.E. 0.92. M.O.Q. length, anterior width, posterior width ratio 0.97: 0.81:1. Lateral eyes slightly protuberant. From above, anterior eye row slightly recurved, width 0.88 mm; posterior eye row slightly procurved, width 0.96 mm. The A.L.E., P.L.E. and P.M.E. all have broad, band-like tapeta, diffuse in A.M.E. Chelicerae: Boss present. Fang groove with 2-3 teeth on retromargin, 3-4 on promargin. Mazxillae: Subparallel, slightly convergent, twice as long as wide. Labium: Wider than long in ratio 1:0.79. Widest subbasally, shallowly notched apically and basolaterally. Sternum: Cordate, shortly pointed posteriorly; longer than wide in ratio 1:0.87. Male palp: Cymbium short and broad, bulb large. Tegulum and proximal embolus protrude strongly basally on the prolateral side. Embolus a large, sinuous, S-shaped spine supported by a similarly sinuous, folded membranous conductor, both ending retrolateral-ventral to the apex of the cymbium. Median apophysis membranous, broad, spoon-shaped and, in ventral view, directed apicolaterally. Tibia with a blunt retrolateral-ventral process directed ventrally and a pointed retrodorsal basal process directed retro-dorsally. Patellar process indistinct to absent. Legs: 1243. Spination: Leg 1, femur p 011, d 112, tibia p 11, r 11, v 122 or 022, metatarsus p 11 or 12, r 11, d 02 or 12, v 221; leg 2, femur p 011, d 112, tibia p 11, r 11, d 012, metatarsus p 11, r11,d 12, v 221; leg 3, femurd 113, tibiap 11, r11,v 012, metatarsus d 11, r11, d 112, v 221; leg 4, femur d 112 or 113, tibia p 11, r 11, v 112, metatarsus p 011, r 001, d 222, v221. Calamistrum weak. Tarsal claws: superior with 9-11 pectinations; inferior with 2-3 pectinations. Hairs ciliate. Trichobothria: single row on tarsus and metatarsus; tarsus of first and second legs with 5, others with 3, placed in central half to third; metatarsus of third leg with 5, remainder with 4, placed in distal three quarters to half, or distal quarter on fourth leg. Bothria collariform with fine, longitudinal striae on proximal plate. Tracheal system: Complex, consisting of four strongly branched tubes confined to the abdomen; spiracle of moderate width, as wide as cribellum. Crbellum: Bipartite, spinning area reduced, strongly sclerotized posteriorly. Spinnerets: Six, short. Anterior lateral pair broad, conical, basally approximated with a very short distal segment: posterior lateral pair thinner and slightly longer, the distal segment one-third of the total length. FEMALE (S 144, O.M.) Similar to male except as indicated below. Measurements (mm): Body length 7.00. Carapace length 2.84, width 2.04. Abdomen length 4.42, width 3.43. Eyes: A.L.E. > A.M.E. = P.L.E. > P.M.E. in ratio 1:0.92:0.92:0.87. Interdistance ratios, A.M.E. - A.M.E. 0.40: A.M.E. - A.L.E. 0.61: A.L.E. - P.L.E. 0.16: P.L.E. Proc. LINN. Soc. N.S.W. 106 (3), (1982) 1983 M.R.GRAY 253 - P.M.E. 1: P.M.E. - P.M.E. 0.89. M.O.Q. length, anterior width, posterior width ratio 0.99:0.78:1. Chelicerae: Fang groove with 4 teeth on retromargin, 5 on promargin. Labium: Wider than long in ratio 1:0.84. Sternum: Longer than wide in ratio 1:0.81. Palp: Tarsal claw with 8-9 pectinations. Legs: 1243. Spination: leg 1, femur p 011, d 112, tibia p 11, r 11, v 122, metatarsus p 011, r011,d 012, v 221; leg 2, femur p 011, d 112, tibiap 11, r11, v 112, metatarsus p 011, r011,d 112, v 220 or 221; leg 3, femur d 133, tibia p 11, r 11, v 012, metatarsus p 011, r011,d 212, v 221; leg 4, femur d 112, tibiap 11, r11, v 112, metatarsus p 011, r 001, d 122, v 121 or 221. Calamistrum well developed and occupying the proximal to central half of the metatarsus. Tarsal claws: superior with 9-11 pectinations; inferior with 2-3 pectinations. Trichobothria (legs 1 to 4): tarsus 5, 4, 4, 4; metatarsus 5, 3, 4, 4. Bothria collariform with several poorly defined ridges curving medially from the lateral margins of the proximal plate and converging upon its base. Tarsal organ an oval opening situation at the distal side of a low, mound ornamented with a few in- distinct semi-circular folds; tarsal organ mound poorly delimited and about three times longer than opening. Cribellum: Bipartite, sclerotized posteriorly and at median partition, spinning areas well developed, spigots strobilate. Genitalia: Epigynal fossa a rounded depression about as long as wide, widest centrally to posteriorly. Fossa bounded posteriorly by a broad, transverse, chitinous ridge; anterior margin of ridge indented. Lateral teeth distal to ridge. Internal genitalia with broad, singly coiled seminal ducts; receptacula small and adjacent near mid line; a broad, curved fertilization duct extends posteriorly. Holotype female: Amaurobius candidus L. Koch 1872 from Bowen, Queensland, Australia. Zoologische Museum, Hamburg. Araneae type cat. no. 11. Museum Godeffroyi cat. no. 7852. Material examined HOLOTYPE female, Bowen, Qld (Mus. Godeff. 7852, Z.M.H.). Male and female (S 144, O.M.), Southwood, 30km west of Moonie, Old, R. Raven; 24.8.1973; from communal web. 2 males and 4 females (Q.M.), Girraween National Park, nr. Stanthorpe, Qld. Male (KS 6941, A.M.), ‘Burnside’, near Margaret River, W.A., M. Gray, 26.1.1979; taken as juvenile in solitary web. Female (KS 6938, A.M.), Torbay, W.A., B.Y. Main 10.10.1977. Female (KS 6939, A.M.), Brookton, W.A., B.Y. Main, 5.5.1977. Male (Ar 811, M.N.H.N.), TYPE of Phryganoporus tubtcola Simon, Denham, W.A. Male (KS 6937, A.M.), Wanaaring, N.S.W., 28.3.1977; taken as juvenile from communal web. Female (KS 6940, A.M.), 21km east of Parkes, N.S.W., M. Gray, 8.4.1972; from solitary web. 6 males, 10 females (KS 5089, A.M.), 6km east of Dubbo, N.S.W., M. Gray, 21.8.1980; from communal web. 7 females, 1 male (KS 8663, A.M.), Kimba, Eyre Peninsula, S.A. Proc. LINN. Soc. N.S.W. 106 (3), (1982) 1983 254 SEMI-COMMUNAL SPIDERS Variation Measurements (mm): Males: carapace length 2.44-2.88, width 1.65-2.10. Females: carapace length 2.84-3.80, width 2.04-2.54. Colour: Sternum with brown or brown and white hairs. In addition to normally pigmented spiders a colour form with increased melanic pigmentation occurs in southwest Australia. As adults these spiders have a dark brown carapace and the light brown patch normally placed immediately behind the mid dorsal abdominal stripe is replaced by a dark brown chevron marking (Fig. 9). The lateral abdominal areas are silvery grey in colour. These spiders correspond well with Simon’s P. nigrinus from Boyanup, W.A., here synonymized with B. candida. Sub- specific status may prove appropriate for this distinctive colour morph. Chelicerae: Retrolateral teeth 2-4, prolateral teeth 3-5. Spination: Leg 1, ventral tibia 010-222. Genitalia: Epigynal fossa as long as wide or wider than long; lateral margins evenly or unevenly curved. Badumna gausapata (Simon 1906), new comb. Figs 12-18, 26 Amaurobius gausapatus Simon 1906 Phryganoporus gausapatus Simon 1908 Phryganoporus candidus Lehtinen 1967 Txeuticus candidus Main 1971 Similar to B. candida and agreeing with the description given for that species except as indicated below. Diagnosis Definite patellar process on male palp; median apophysis directed apico- laterally. Lateral teeth of epigynum subdistal; fossa widest in central to posterior half. Cheliceral teeth, retrolateral 2-3 and prolateral 5-6. Ventral tibial spination, first leg 112-222. Metatarsal trichobothria, first leg 5-6. MALE (KS 6942, A.M.) Measurements (mm): Body length 7.05. Carapace length 3.45, width 2.61. Abdomen length 3.60, width 2.31. Colour: As for B. candida. Sternum with brown and white hairs. Carapace: Longer than wide in ratio of 1:0.76. Clypeus height equals 1.25 diameters of an A.M.E. Eyes: A.M.E,. > P.L.E. > A.L.E. > P.M.E. in ratio 1:0.98: 0.96: 0.83. In- terdistance ratios, A.M.E. - A.M.E. 0.38: A.M.E. - A.L.E. 0.53: A.L.E. - P.L.E. Figs 12-18. Badumna gausapata. 12-14, male palp: 12, retrolateral; 13, ventral; 14, prolateral. 15-17, female genitalia (SYNTYPES): 15-16, epigyna; 17, internal genitalia. 18, female abdomen, dorsal. Figs 19-22. Badumna longinquus. 19-20, female genitalia: 19, epigynum; 20, internal genitalia. 21-22, male palp: 21, ventral; 22, tibia and patella, retrolateral. Scale lines 0.2 mm. Proc. LINN. Soc. N.S.W. 106 (3), (1982) 1983 M.R.GRAY 255 Proc. LINN. Soc. N.S.W. 106 (3), (1982) 1983 256 SEMI-COMMUNAL SPIDERS 0.23: P.L.E. - P.M.E. 0.89: P.M.E. - P.M.E. 1. M.O.Q. length, anterior width, posterior width ratio 0.98: 0.80:1. From above anterior eye row recurved, width 0.97 mm, posterior eye row slightly procurved, width 1.09 mm. Chelicerae: Fang groove with 2 retromarginal teeth, 4 promarginal teeth. Maxillae: Parallel, twice as long as wide. Labium: Wider than long in ratio 1:0.80; surface convex. Sternum: Longer than wide in ratio 1:0.79. Male palp: Median apophysis membranous, moderately narrow and spoon-shaped; in ventral view directed apico-laterally. Patellar process a short, bluntly pointed, finger- like projection, placed dorsally. Legs: 1243. Spination: Leg 1, femur p 011, d 112, tibia p 101, r 101, v 122, metatarsus p 011, r101,d012, v 221: leg 2, femur p 001, d 213, tibia p 11, r 11, v 112, metatarsus p 101, r101,d 012, v 221: leg 3, femur d 113, tibia p 11, r 11, v 012, metatarsus p 011, r011,d 212, v 221: leg 4, femur d 113, tibia p 11, r 11 or 111, v 112 or 122, metatarsus p 011, r 001, d 222, v 221. Calamistrum very weak. Trichobothria (legs 1 to 4): tarsus 5, 4, 4, 3: metatarsus 5, 4, 3, 3. FEMALE (KS 6086, A.M.) Similar to male except as indicated below. Measurements (mm): Body length 7.60. Carapace length 3.38, width 2.21. Abdomen length 4.45, width 2.95. Colour: Sternum with brown hairs, white hairs absent. Eyes: A.L.E. > P.L.E. > A.M.E. > P.M.E. in ratio 1:0.95: 0.86:0.85. Interdistance ratios, A.M.E. - A.M.E. 0.36: A.M.E. - A.L.E. 0.48: A.L.E. - P.L.E. 0.22: P.L.E. - P.M.E. 1:P.M.E. - P.M.E. 0.76. M.O.Q. length, anterior width, posterior width ratio 1:0.81:1. Chelicerae: Fang groove with 3 or 4 retromarginal teeth, 4 or 5 promarginal teeth. Labium: Wider than long in ratio 1:0.75. Sternum: Longer than wide in ratio 1:0.81. Palp: Tarsal claw with 7-8 pectinations. Legs: 1423. Spination: leg 1, femur p 011, d 122, tibia p 11, r 101, v 222, metatarsus p 011, r011,d 112, v 221; leg 2, femur p 0111, d 122, tibia p 11, r11, v 122, metatarsus pii,ri11,d112, v 221; leg 3, femur d 133, tibia p 11, r 11, d 212, v 221; leg 4, d 112, tibia p 11, r11, v 112, metatarsus p 111, r 001, d 1012, v 221. Trichobothria (legs 1 to 4): tarsus 6, 4, 4, 4; metatarsus 6, 5, 4, 4. Bothria collariform, surface of proximal plate with narrow, semi-longitudinal ridges. Tarsal organ opening oval, placed near the distal margin of a poorly defined, elongate mound approximately four times as long as opening. Proc. LINN. Soc. N.S.W. 106 (3), (1982) 1983 M.R. GRAY 257 Figs 23-25. Tarsal organs, leg 1: 23, B. candida (similar in B. gausapata); 24, B. vandiemeni; 25, B. longinquus. Figs 26-28. Trichobothrial bases, tarsus, leg 1: 26, B. gausapata; 27, B. vandiement; (both types present in B. candida) 28, B. longinquus. Scale lines 5 p. Genitalia: Epigynal fossa wider than long, widest in central to posterior area, margins smoothly curved. Lateral teeth subdistal. Anterior margin of transverse ridge not indented. Material examined SYNTYPE females (AR 810, Paris), Victoria, 1903. 1 male (KS 6942, A.M.), Canberra, A.C.T., M.S. Upton, 24.5.1965. 5 females (KS 6086, A.M.), Black Mountain, Canberra, A.C.T., M. R. Gray, 1.10.1980. 4 males, 3 females (KS 6090, A.M.), Black Mountain, Canberra, A.C.T., 3.6.1965. 3 males, 4 females (KS 6088 A.M.), Wee Jasper, N.S.W., M. Gray 30.10.1980. 4 males (A.N.I.C.) Black Mountain, Canberra, A.C.T., I. F. B. Common, 4.6.1965 (from light trap). Variation Measurements (mm): Males; carapace length 3.05-3.48, width 2.19-2.61. Females: carapace length 3.38-2.44, width 2.21-1.22. Chelicerae: Retrolateral teeth 2-3, prolateral teeth 4-5. Spination: Leg 1, ventral tibia 122-222. Badumna vandiemeni n. sp. Figs 24, 27, 29-34 Ixeuticus candidus Hickman 1967 Similar to B. candida and agreeing with the description given for that species except as indicated below. Diagnosis Definite patellar process on male palp; median apophysis directed laterally. Lateral teeth of epigynum subdistal; fossa widest in anterior half. Mid-dorsal ab- dominal colour pattern light brown posteriorly. Cheliceral teeth, retrolateral 2 and Proc. LINN. Soc. N.S.W. 106 (3), (1982) 1983 258 SEMI-COMMUNAL SPIDERS Figs 29-34. Badumna vandiemeni. 29, male abdomen, dorsal. 30-32, male palp: 30, retrolateral; 31, ventral; 32, prolateral. 33-34, female genitalia: 33, epigynum; 34, internal genitalia. Scale lines 0.2 mm. prolateral 4-5. Ventral tibial spination, first leg 121-222. Metatarsal trichobothria, first leg 7-9. MALE (KS 6976, A.M.), Holotype Measurements (mm): Body length 8.55. Carapace length 4.06, width 2.80. Abdomen length 4.58, width 3.70. Colour: Dorsal abdominal stripe fairly long, about half as long as abdomen. Posterior chevrons indistinct, light brown except for small, lateral, dark brown patches of Proc. LINN. Soc. N.S.W. 106 (3), (1982) 1983 M.R.GRAY 259 pigment. Sternum with dark brown and white hairs. Carapace: Longer than wide in ratio of 1:0.69. Clypeus height 1.25 diameters of an A.M.E. yee eNO. > ACL. > PE. > P.M_E. in‘ ratio of 1:0°95:0.92:0.84. In- terdistance ratios, A.M.E. - A.M.E. 0.39: A.M.E. - A.L.E. 0.42: A.L.E. - P.L.E. 0.20: P.L.E. - P.M.E. 1.00: P.M.E. - P.M.E. 0.90. M.O.Q. length, anterior width, posterior width ratio 1:0.82:0.96. From above anterior eye row slightly recurved, width 1.20 mm; posterior eye row slightly procurved, width 1.34 mm. Chelicerae: Retrolateral teeth 2; prolateral teeth, 4-5. Mazxillae: Subparallel, slightly convergent, twice as long as wide. Labium: Wider than long in ratio 1:0.80. Sternum: Longer than wide in ratio 1:0.81. Male palp: Median apophysis moderately narrow and spoon-shaped: directed laterally, almost horizontal in ventral view. Patellar process dorsaily placed, short, bluntly pointed and adorned basally with white, spatulate hairs. Legs: 1243. Spination: leg 1, femur p 011, d 112, tibia p 11, r 11, v 222, metatarsus p OMe OlpecdhOO2 ve22 leos?* femur pO0ll or 0111 d t2) tibiap ir 11 122, metatarsus p 101, r 101, d 002, v 222: leg 3, femur p 0111, d 113, tibiap 11, r11v 022: metatarsus p 0101, r 0101, d 112, v 221: leg 4, femur d 113, tibia p 11, r 11, v 112, metatarsus p 111, r 001, d 112, v 221. Calamistrum weak. Trichobothria (legs 1 to 4): tarsus 6, 5, 5, 5: metatarsus 7, 6, 6, 5. FEMALE (KS 6977, A.M.), Paratype Similar to male except as indicated below. Measurements (mm): Body length 8.60. Carapace length 3.69, width 2.48. Abdomen length 4.95, width 3.50. Colour: Dorsal abdominal stripe less than half as long as abdomen. Carapace: Longer than wide in ratio 1:0.67. Eyes: A.M.E. > A.L.E. > P.L.E. = P.M.E. in ratio 1:0.95:0.85:0.85. Interdistance ratios, A.M. — A.M_E. 0.42; A.M.LE,-— A.1L.E. 0.71: A.L.E. - P.L.E. 0.23: P.L.E. - P.M.E. 0.86. P.M.E. - P.M.E. 1.00 M.O.Q. length, anterior width, posterior width ratio 1:0.73:0.99. Eye row width, anterior 1.20 mm; posterior 1.34 mm. Chelicerae: Retrolateral teeth, 2; prolateral teeth, 4. Labium: Wider than long in ratio 1:0.84. Sternum: Longer than wide in ratio 1:0.85. Legs: 1243. Spination: leg 1, femur p 0011, d 112, tibia p 11, r 11, v 121; metatarsus p 011, r 011, d 002, v 221; leg 2, femur p 0011, d 112, tibia p 11, r 11, v 121, metatarsus p 011, r011, d 002, v 221; leg 3, femur d 113, tibia p 11, r11, v 012, metatarsus p 011, r 011, d 002, v 221; leg 4, femur d 112, tibia p 11, r 11, v 012, metatarsus p 011, r 0011, d 002, v 121. Trichobothria (legs 1 to 4): tarsus 5, 5, 4, 5; metatarsus 9, 5, 7, 5. Bothria collariform with several, well defined, semi-circular ridges on proximal plate. Tarsal organ a small, oval opening, acutely pointed proximally, placed near the distal margin of a well delimited, oval mound six times as long as opening; surface of mound ornamented by fine striae. Genitalia: Epigynal fossa wider than long, widest anteriorly, lateral margins sloping inwards to subdistal lateral teeth. Transverse ridge not indented anteriorly. Proc. LINN. Soc. N.S.W. 106 (3), (1982) 1983 SEMI-COMMUNAL SPIDERS 37 Figs 35-37. Badumna candida, webs. 35, communal web, Dubbo, N.S.W.; 36, solitary web retreat, Goonoo S.F., N.S.W.; 37, solitary web retreat of juvenile male, dark colour form, Margaret River, W.A. Types Holotype Male — KS 6976 (A.M.). Eaglehawk Neck, Tas., M. R. Gray, 3.7.1980; from solitary web on shrub (Acacia uriczfolta). Paratypes — Female, KS 6977 (A.M.), same data as holotype. Female, KS 6978 (A.M.), Eaglehawk Neck, Tas., V. V. Hickman, 6.3.1960. 3 females, J.763 (T.M.), Lauderdale, Tas., April 1971. Variation Measurements (mm): Females: carapace length 2.95-3.69; carapace width 1.97-2.40. Spination: Leg 1, ventral tibia 121-222. BIOLOGICAL NOTES Spiders of the B. candida species group make both communal and solitary webs. As noted by Main (1971) communal web populations are made up mainly of juveniles, but adults are also often present; some, at least, may complete their life cycles within the communal web. However, many leave to take up a solitary existence soon after they mature. Late instar juveniles as well as adults are involved in such dispersal as both subadult and mature spiders can be found in solitary webs. The webs (Figs 35-37) are built among the foliage of various low, sclerophyllous trees and shrubs. Solitary webs are small, the retreat usually being fastened along a stem while the irregular cribellate snare extends a short way into the surrounding foliage. Communal webs vary greatly in size and may encompass much of the foliage of a shrub or branch. Up to 95 spiders have been recorded from a single large web. Counts of penultimate juveniles in communal webs often showed marked disproportions in sex ratios; whether this simply represents differential dispersal or involves some other factors is not clear. Though their retreat tubes are independent of each other, juvenile spiders will hunt and feed together on the same prey animal in the shared catching part of the web. Main (1976, fig. 40e) noted that the structure of the solitary web in south western populations of B. candida consisted of a short, bag-like tube of silk fastened onto a branch, the small sheet web radiating out from it. Similar solitary nests are built by members of the candida species group in eastern Australia (Fig. 36), though sometimes Proc. LINN. Soc. N.S.W. 106 (3), (1982) 1983 M.R.GRAY 261 their retreat tubes are more elongated. In southwestern Australia long, horn-like retreats containing juvenile males of the dark colour form of B. candida (reared to maturity in the laboratory) have been collected (Fig. 37). One to three egg sacs can be found embedded in the silk-plant-food detritus wall matrix of the female retreats. The sacs are circular to oval, flattened spheres with a definite circumferential seam varying in diameter from 4 to 7 mm. The outer silk is flocculent and attaches the sac closely to the retreat wall: inside this is a more finely woven, thin layer of silk. The eggs are non-glutinous and vary from 0.6 to 0.8 mm in diameter. Each sac contains from 13 to 49 eggs. Sacs were found in retreats from both solitary and communal webs. ACKNOWLEDGEMENTS My thanks go to Dr M. Hubert, Dr B. Y. Main, Prof V. V. Hickman, Dr V. Davies and Mr M. Upton for the loan of material. Inking and shading of drawings was done by Ms P. Greer. References DALMAS, C. de, 1917. — Araignées de Nouvelle-Zélande. Ann. Soc. Ent. France 86: 317-430. DONDALE, C. D., 1966. — The spider fauna (Araneida) of deciduous orchards in the Australian Capital Territory. Aust. J. Zool. 14: 1157-92. FORSTER, R. R., 1970. — The spiders of New Zealand, Pt. III. Otago Mus. Bull. 3: 1-184. , and WILTON, C. L., 1973. — The spiders of New Zealand, Pt. 1V. Otago Mus. Bull. 4: 1-309. HICKMAN, V. V., 1967. — Some common spiders of Tasmania. Hobart: Tasmanian Museum and Art Gallery. KULCZYNSKI, V., 1908. — Symbola and faunam aranearum Javae at Sumatrae cognoscendam. I. Mygalomorphae et Cribellatae. Bull. Acad. Sci. Cracov. 1908: 527-581. Koch, L., 1872. — Die Arachniden Australiens. Nuremburg: Ludwig Korn. LEECH, R., 1971. — The introduced Amaurobiidae-of North America and Callobius hokkatdo n.sp. from Japan (Arachnida:Araneida), Canad. Ent. 103: 23-32. , 1972. — A revision of the Nearctic Amaurobiidae. Mem. Ent. Soc. Can. 84: 1-182. LEHTINEN, P. T., 1967. — Classification of the Cribellate Spiders and some allied families with notes on the evolution of the suborder Araneomorphae. Ann. Zool. Fenn. 4: 199-468. MAIN, B. Y., 1964, 1967. — Spiders of Australia, (2 editions). Brisbane: Jacaranda Press. 1971. — The common ‘colonial’ spider Ixeuticus candidus (Koch) and its synonyms (Dic- tynidae: Araneae). J. Roy. Soc. W.A. 54 (4): 119-120. , 1976. — Spiders. Sydney: Collins. Australian Naturalist Library. McKeown, K. C., 1963. — Australian spiders. Sydney: Angus and Robertson. Roewer, C. F., 1942-54. — Katalog der Araneae, 2 vois. Bremen: Wissenschaften Paul Budy. SIMON, E., 1899. — Ergebnisse einer Reise nach dem Pacific. Arachnoideen. Zool Jahrb., Syst. 12: 411-437. SIMON, E., 1906. — Etude sur les Araignées de la section des Cribellates. Ann. Soc. Ent. Belg. 50: 284-308. , 1908. — In Die Fauna Siidwest Australiens 1 (12): 359-446 (W. MICHAELSEN AND R. HARTMEYER (eds) ). Jena: Gustav Fischer. THORELL, T., 1890. — Studi sui Ragni malesie papuani. [V. Ann. Mus. Civ. St. Nat. Genova 28: 1-419. ? Proc. LINN. Soc. N.S.W. 106 (3), (1982) 1983 PROCEEDINGS of the LINNEAN SOCIETY NEW SOUTH WALES VOLUME 106 NUMBER 4 THE SIR WILLIAM MACLEAY MEMORIAL LECTURE 1982 Ecological Values of the Tropical Rainforest Resource L. J. WEBB Honorary Fellow, School of Australian Environmental Studies, Griffith University, Nathan, Brisbane, Australia 4111* [Delivered 8 September 1982] ‘Tf we knew enough no choice would be trivial, and it is our duty to acquire the knowledge which will enable us to moralize our everyday actions, both by the study of available statistics and by encouraging statistical inquiry elsewhere’ (Haldane, 1928). In medieval times Wisdom was separated into Metaphysics, Morals, and Natural Philosophy which was equivalent to natural science. In the seventeenth century, Newton along with Galileo and Descartes established the prestige and paramountcy of mathematics and physics, and the domination of scientific thinking was well under way. The term ‘natural philosophy’ became restricted to the exact physical sciences, which were joined by chemistry at the beginning of the 19th century. It remained for Playfair in 1812 to identify Natural History simply as a descriptive and non- experimental science that dealt with vegetable, animal, and mineral objects (Pantin, 1968). Much later, natural history became known as the biological and earth sciences, parts of which became increasingly quantitative, and involved with finer grades of structures and molecular events. In this hierarchical re-arrangement of the sciences, which emphasized precision and experimentation, the so-called descriptive biological sciences stayed at the bottom. Thus within a century or two of western scientific progress, biology and the study of organism-environment relationships that was to become known as ecology lost their early links with morals and metaphysics, became ‘poor relations’ of the exact sciences, and were obliged to renounce intrinsic human values in the promotion of objective scientific knowledge. It is relevant that William Macleay, that great Australian naturalist whose memory we celebrate tonight, identified the Linnean Society of New South Wales, when it was founded just over a century ago, as a ‘society of natural history’ that he hoped would succeed. It was founded not only ‘to promote knowledge’, but also ‘for the progress of this community, and for the welfare of humanity’. Even then, it seems, the extraordinarily broad interests of early naturalists such as Macleay enabled them to foreshadow the necessary integration of the social and natural sciences — an in- tegration that only now, in international projects such as the Man and the Biosphere programme, is being hesitantly and not too effectively approached. While admiring what Sir Otto Frankel (1970) called Sir William Macleay’s ‘remarkable prescience’, we may wonder what Sir William would have thought about * Formerly, Rainforest Ecology Unit, Division of Plant Industry, CSIRO Laboratories, Indooroopilly, Australia 4068. Proc. LINN. Soc. N.S.W. 106 (4), (1982) 1983 264 ECOLOGICAL VALUES OF THE TROPICAL RAINFOREST RESOURCE the future losses to human welfare when biological and physical sciences became so rigidly separated, when science became so thoroughly insulated from ethics, and when there was nothing left but utter disdain for metaphysics. It is more helpful for the biologist today, who retains something of his natural history heritage, to understand the fundamental contrast and separation as not be- tween biological and physical sciences but, as Pantin (1968) put it, ‘between unrestricted and restricted sciences’. Thus the ecologist who is obliged to follow the analysis of problems into many other sciences, and even non-sciences, generally must rely on ‘natural experiments’; and to gain a broad and integrative view must sacrifice exactness. Yet ecological observation is not mere description, because perception depends on experience, training, and current concepts. It is hard to accept that observations can be strictly neutral when research directions are selected beforehand, priorities are set by economic standards, and the presentation and interpretation of data are coloured by the requirements of a particular development. In other words, it may not always be possible to separate questions of fact from questions of value. If we are honest about it, I suggest that we all can find some favourite examples. It has become clear to me that decisions about the use and management of natural resources are unavoidably influenced by value judgements. Questions of fact and questions of value become inseparable in the broad issues of ‘ecological problems’, as distinct from narrowly-circumscribed technical ‘problems in ecology’ (Passmore, 1974; Ashby, 1978). The ecologist as an unrestricted scientist is therefore continually im- plicated with the ‘is’ of fact, and the ‘ought’ of valuation that is not intrinsic to his study and rests on an ethical system of some kind. I wonder what Macleay would have done, had he, with his broad humanist and what we may call his wide ecological interests, been confronted as we are now by this thorny path to applied ecology, to the resolution of choices impelled by knowledge of ecological consequences, and to dawning insights about how certain resources have more values than utilitarian. In this lecture I shall try to identify some ecological values from a knowledge of rainforests. It seems that understanding how valuable a resource really is must await the accumulation of enough facts and experience: then crystallization occurs to reveal unsuspected facets. First of all I give an international perspective to ecological problems encountered in Australian tropical and subtropical rainforests. Then I briefly examine the Australian forests: their history and interactions with other vegetation and with man, which in turn provide a basis of comparison with rainforests of our northern neighbours. Finally I return to the question of values — ecological and beyond — that the rainforests project. The colligation of facts and intrinsic values prompts the identification of fresh moral choices, and suggests new responsibilities for ecology that has now become ‘the genesis of a science of man and nature’ (di Castri, 1981). About the time that Macleay and T. H. Huxley met in Sydney in the middle of last century, the world’s population had reached its first billion. Nowadays it takes only 14 years to add another billion. Of the people who will be added to the world’s population by the end of the century, about 90 per cent (about 1.5 billion) will be in the tropics, i.e. in the less developed, poor, third world countries of Asia, Africa, and Latin America. Already, in the third world, about 2 billion of the rural poor, in their struggle to survive, exhaust the soils, deplete the forests, destroy wildlife, and pollute the waters. As the direct and indirect result of technological developments and the ‘pullulating millions’, the world’s tropical and subtropical moist forests are disappearing at an accelerating rate. The rates of conversion vary from country to country, and among Proc. LINN. Soc. N.S.W. 106 (4), (1982) 1983 L. J. WEBB 265 different land systems within a particular zone, so that average rates may be quite misleading. They have been heavily dramatized, with estimates varying from 20 to 40 hectares per minute (Myers, 1980). Unfortunately, most of the forests that have so far been cleared are the moister complex types with the greatest array of genetic resources at lower altitudes. Jacobs (1978a) illustrates this for Malaysia by noting that there are 9 different types of dipterocarp forest below 300 m, as opposed to 6 types between 300 and 700 m, and only 3 types between 700 m and 1200 m altitude. In the tropical humid region of North Queensland, this is more pronounced, as there are about 25 main structural types of rainforest below 800 m altitude but only about 8 main structural types in the 800-1600 m altitudinal zone. These facts indicate that proportionately more diversity is lost in converting lowland rainforest types. Of the estimated 3 million species of plants and animals in the tropics, only about half-a-million have scientific names. It has been predicted that, at the present rate of destruction of the tropical and subtropical moist forests throughout the world, one-half to one million species of plants and animals will become extinct. Thus to retain as much natural or near-natural tropical forests as possible usually means ‘conserving the unknown’ (MAB Ecology in Action Exhibit, Paris, 1981). In stressing that less than half of the world’s predicted threatened plants are known, we must realize that rare plants are no longer ‘the playchild of a botanical elite’. In countries such as India, where there is so much reliance on the wild plant resource, plant conservation is an essential part of economic development (Synge, 1982). Most people, especially biologists, would agree that species extinctions are a bad thing, although people’s reasons vary. This is, however, a moral judgment. Since morals are the practice of ethics, and ethics is the science of morals, any opinion and action about species extinction must have ethical roots. Such judgment, which is freely made by biological scientists and some others, is therefore worth examining in a little more detail. The MAB Ecology in Action Exhibit in Paris (1981) claims that con- serving both the known and the unknown is not just a question of altruism but of human survival, and proposes three justifications: ‘to preserve the genetic information lost through species’ extinctions; to keep open the options for facing our future requirements; and to conserve a natural diverse environment which will satisfy us morally and aesthetically’. This seems a fair proposition, and far from extremist. While acknowledging expediency and the necessities of practical life, it recognizes that less tangible values are of importance, and second to none. There is thus a spectrum of ecological values for the conservation of nature, with pragmatism and self-interest at one end and the rights of nature apart from man at the other. Myers (1978), while claiming that ‘the problem of disappearing species could eventually be seen as one of the greater ‘sleeper issues’ of the late 20th century’, neatly summarized the utilitarian reasons for conserving animal and plant genetic resources in tropical rainforests: for many particular uses in modern agriculture, in medicine and pharmaceutics, and in industrial processes. At the other end of the continuum there are quasi-religious reasons for the con- servation of so-called ‘non-resources’ (Ehrenfeld, 1976). Non-resources are species and biotic communities that do not have an economic value, or demonstrated potential value for use by man here and now. Some people try to justify non-resources to make them competitive with economic resources. For example, that recreational and aesthetic values are essential for mental health; species diversity and polycultures are less risky than monocultures; natural communities may provide clues to organization that ensures survival; certain species may be sensitive environmental impact in- dicators; natural ecosystems could serve as models for reconstructing degraded ones; Proc. LINN. Soc. N.S.W. 106 (4), (1982) 1983 266 ECOLOGICAL VALUES OF THE TROPICAL RAINFOREST RESOURCE intact ecosystems are needed for teaching purposes. All these examples suffer from being too remote and nebulous to be convincing as political arguments. The ultimate justification is rarely articulated: a deep-seated fear that irreversible changes which reach a significant scale ‘may carry a hidden and unknowable risk of serious damage to humans and their civilizations’ (Ehrenfeld, 1976). Homocentric as this fear is, it is clearly impossible to assign resource values and priorities to non-resources that can rank with, and excel, the economic values of resources as treasured by a totally ex- ploitative society. That neither development nor preservation can be total, however, gives grounds for believing that fresh value systems will continue to emerge. Elton (1958) understood the spectrum of human attitudes towards co-existence with nature as a series of ‘three absolute questions . . . waiting patiently to be an- swered’ (which a quarter of a century later is still the case). He identified the questions as religious, aesthetic and intellectual, and practical. This is how he put the least scientific of these questions: ‘The first, which is not usually put first, is really religious. There are some millions of people in the world who think that animals have a right to exist and to be left alone, or at any rate that they should not be persecuted or made extinct as a species. Some people will believe this even when it is quite dangerous to themselves’. Examples of this belief are multiplying everywhere in the world. We witness them locally as campaigns and demonstrations in various rainforest areas in eastern Australia from north Queensland to south-west Tasmania. In recent times, in October 1978, the tropical botanist Jacobs (1978b) taunted the 8th World Forestry Congress at Jakarta, the theme of which was ‘Forests for People’, by a paper quoting the contrary title: “People for Forests’. Jacobs sensed a growing feeling in industrialized Europe that power over nature is no longer what it used to be, and predicted a change in forest policy towards mixed natural forests as well as, or in places instead of, monocultures. Jacobs argued that ‘People for Forests’ would have carried ‘a more dignified appeal’, and implied ‘a notion of respect, and respect befits people’. He cited his colleague P. van Heynsbergen, an Amsterdam legal scholar, who went further, and published in 1977 a ‘Declaration of the Rights of Animal and Plant Life’ with eight paragraphs, the first of which reads that, in principle, ‘Each living creature on earth has the right to exist, independent of its usefulness to humans’. This idea is echoed by the current Charter of Nature to be considered by the United Nations (Kamanda wa Kamanda, 1980). These sentiments are no longer the eccentricities of a cultivated few in the so- called First World, nor are they restricted to philosophers, poets, oriental sects, and primitive Aboriginal society. It is worth considering Ashby’s (1978) claim that the most effective defenders of nature today are not humanists, but scientists — mainly natural scientists aided by a few social scientists. To be effective — in a wider campaign — does require a certain scientific objectivity, because this is a basic prerequisite for communication by scientific workers who are not gifted like poets in this art, and because objective data can be tested by any other scientist. Scientific concern that nature be valued for its intrinsic worth reflects Elton’s (1958) passion for conservation of the greatest possible ecological variety, as well as contemporary ecological intuitions about what interdependence in ecosystems and other holistic principles mean in the evolution of the human condition. Somehow, despite the ingenuity of systems ecologists, and the synecologists’ vision of ‘emergent properties’ at different levels of organization, holism as a working principle used by western scientists has done little to help solve the urgent problems of conservation and development in the developing countries of the tropics. Moreover, even in the most developed country of all, Odum (1977) noted that controversy had PRoc. LINN. Soc. N.S.W. 106 (4), (1982) 1983 L. J. WEBB 267 clouded the ‘holistic strategy for ecosystem development’, and suggested that the disappointing performance of the United States in the International Biological Programme could be attributed to lack of ‘unifying themes or concepts’. Yet a holistic frame of reference — the concept of environment in its totality, the habit of ‘looking around corners’, the prediction of inevitable effects in space and time of present actions — was being established slowly and firmly in widening sections of society by the ideas, whatever their media, of ‘planet earth’, ‘the global village’, and so on. Similarly but in a quite different context, spokesmen in the developing countries (e.g. Salim, 1982; Soemarwoto, 1982) were beginning to urge the use of traditional knowledge, and of holistic rather than analytical methods as a basis of ecological guidelines for development through conservation. Earlier ecological guidelines produced by the International Union for the Conservation of Nature and Natural Resources (Poore, 1976) had generally been applied from ‘top to bottom’ and had had little impact among the people. Given this international and partly philosophical background, we can examine the values of the tropical rainforest resource to the Australian community. We should note that the definition of ‘resource’ is surprisingly wide, and includes computable wealth in materials; something in reserve; something to which one has recourse in difficulty; provision of relief or recovery and a means of spending leisure time. Passmore (1974) went much further with the definition of ‘nature’, teasing out its meanings in which man ranges from despot to primitivist, and noting its tangents with metaphysics and sentiment. Throughout the four decades of my studies in CSIRO I have been exposed to an expanding spectrum of uses and ‘non-uses’ of rainforests for people. The facts of science — science that traditionally makes disinterested inventories of ‘nature out there’ — were only part of the spectrum. .The spectrum was always being coloured, rainbow-like, through the prism of different values. Always there were the covert and unacknowledged moral choices. But even more than this seems at stake. What of those people who believe, who somehow intuitively know, that ancient and prolonged existence in nature of Australian rainforest relicts has value? That it confers a right for these remnants to continue to exist as objects in nature? That this is now a new kind of ‘justification’ for not exterminating or damaging any more of them? Let us examine some examples of the evolution of the less tangible values of rainforests that are not readily separated as scientific facts i.e. as evidence free from inference, opinion, and advocacy, and that are therefore bound to involve moral choices. This examination will require brief reference to some recent developments in tropical rainforest botany and ecology in Australia. At the outset it should be acknowledged that any ecologist studying physical and biological processes in rainforests of the tropics and subtropics of the world cannot exclude human ecological processes from his chosen ecosystem. Often he cannot see the trees for the people; or in places like Peninsular Malaysia, for the extensive food and industrial monocultures into which nearly all the lowland rainforests have been con- verted. Food crops are needed for subsistence, if they can keep pace with exploding population. Industrial crops provide necessary income, although this may end up in foreign hands. Wholesale conversions of rainforest such as wood-chipping, e.g. in the Gogol Valley, Papua New Guinea, ignore ecological safeguards (Webb, 1977). Economic development becomes the manipulation of things, and any increases in human welfare ‘are seen as being spin-offs from a rapidly growing economic system in which such things as income and consumption inequalities and natural resource destruction are considered normal’ (De’ Ath, 1982). Proc. LINN. Soc. N.S.W. 106 (4), (1982) 1983 268 ECOLOGICAL VALUES OF THE TROPICAL RAINFOREST RESOURCE What are our historical perspectives as a nation about rainforests? Australia is the only developed country largely in the tropics. The area of rainforests throughout the whole of the continent when Europeans arrived was relatively small, about 1 per cent; plus another half per cent or so if coastal sclerophyll-rainforest mixtures, and scattered fragments in drier subcoastal north-eastern areas are added. The rainforest area when Aboriginals arrived over 40,000 years ago was somewhat larger than when the first Europeans settled 200 years ago. The area was much larger during earlier moist warm interglacial periods. And long before man arrived, beyond the onset of arid conditions in the Plio-Pleistocene, and stretching back in time through the Tertiary to the Cretaceous and the origin of the flowering plants, rainforests (or moist closed forests) covered most if not all of the soils of the land mass that became Australia. Current interpretations of the evolution of the rainforest and sclerophyll forest floras have reversed traditional botanical doctrine and the ‘invasion theory’ (Keast, 1981; Barlow, 1981). Previously it was thought that the sclerophylls were the most ancient, the only truly Australian (autochthonous) elements; whereas the rainforests were ‘alien and invasive’ immigrants. However, the sharing of many taxa by the tropical and subtropical rainforests in Australia and Indo-Malesia-Oceania is now construed as indicating common origins of palaeotropical stocks in Gondwanic times, as well as various migrations since (Webb and Tracey, 1981a; Webb, Tracey and Jessup, 1982). Since European settlement, about two-thirds to three-quarters of the area of coastal and subcoastal rainforests and monsoon forests in the tropics and subtropics have been cleared, leaving about 10,000 km?. Rainforest conversion has varied from virtually 100 per cent on the lowlands and tablelands of the tropical and subtropical coast, to virtually nil on high mountains. We now view the last fragments of Australian rainforests through the eyes of fresh scientific understanding that is most conveniently dated from last year, which marked the publication of Keast (1981), and the first International Botanical Congress to be held in Australia. The rare and scattered distribution of the rainforests is the legacy of million-year-old climatic sifting, thousand-year-old Aboriginal burning, and century- old impacts of Europeans. Their rarity has now become valuable. The network of refugia for distinct community-types, and the abundance of disjunct and different kinds of endemic plant species, are highly relevant to the location of reserves for nature conservation, especially for scientific purposes. This is one of the aims of the State National Parks and Wildlife Services, and a central aim of the UNESCO Biosphere Reserves and the World Conservation Strategy. For Australian rainforests, ‘small’ may also be ‘beautiful’ (with an efficient buffer zone), so that a scattering of carefully selected smaller reserves — as well as larger ones — is essential to preserve not only within-species variation, but also relictual species and species-assemblages. There are virtually no data for minimum breeding population sizes for different categories of plants in Australian rainforests. Many tree species are restricted to very few sites (Tracey, 1981). These rare species are common where they occur. Are their population sizes stable? If the refugia are ancient, and the. species are of low vagility, are the rare species tending to extinction in the absence of immigration? Australian rainforest distributions provide ready-made and natural experiments in island biogeography. Elsewhere, as in Amazonia, scattered remnants have to be preserved artificially for this purpose. The age, mutability, and diversity of refugia, and the endemic and primitive species in them, should all be considered in judgments about size, shape, and distribution of nature reserves (Webb and Tracey, 1981a; Kikkawa etal. , 1981). Proc. LINN. Soc. N.S.W. 106 (4), (1982) 1983 L. J. WEBB 269 The transitions and interspersions with eucalypts and other sclerophylls of the mostly closed, broad-leaved Australian rainforest and monsoon forest types that, unlike the sclerophylls, have many genera in common with Indo-Malesia, have ecological properties that are unique in the world. Boundaries in the tropics and subtropics are often abrupt, as the result of a combination of factors dominated by surface fire. Boundaries are generally gradual and extensive in temperate and montane regions where crown fires are a dominant factor, and in wet tropical and subtropical regions on poorer soils. The ‘tall dark wall’ of rainforest rising at the edge of the sunlit eucalypt grassy woodland — a metaphor beloved by some early authors — 1s actually only one of many types of transition. At least five types of transition of interspersion of the rainforest and sclerophyll forest floras can be recognized (Webb and Tracey, 1981b; Webb, Tracey and Williams, 1983). The evolutionary significance of the interspersed taxa, both phylogenetically and syngenetically, has not been studied in detail. The transitions and mixtures are nevertheless highly relevant to our understanding of community development and stability involving the rainforest and sclerophyll floras. It is certainly not helpful for this scientific purpose to discard or confuse ecological facts and hypotheses by pragmatic forestry typology based on commercial timber volumes. Commercial hardwoods such as Brush Box (T7istania conferta) reach relatively huge volumes in these transitional forests or ‘mixed rainforests’ at advanced stages of succession. Large Brush Box trees up to approximately 1200 years old were the subject of confrontations in the forest at Terania, and in court at Sydney, New South Wales, between forestry and the timber industry, and conservation groups (Prineas and Elenius, 1980). The conflicting values placed on these trees, either as logs or as ancient objects in nature, provide a classical example of the politics of ecology, and of the power of popular support for ‘non-wood’ values of Australian moist forests. Ulterior factors have also entered the botany of Australian rainforests at the highest levels of vegetation classification. A comprehensive floristic analysis of Australian rainforests, either by tree species or tree genera, separates the subtropical formation-class at the third and fourth division respectively in the hierarchy (Webb and Tracey, 1981a, 1981b; Webb, Tracey and Williams, 1983). The subtropical rain- forests, often characterized by Araucarian emergents that are of great antiquity and of high commercial value, should not, therefore, be regarded simply as impoverished versions of the tropical types farther north. The separate ecological identity of the subtropical rainforests has to be taken into account when judging the adequacy of existing rainforest reserves and national parks, especially when compared with those of the tropical and the montane types. Species richness and the maintenance of diversity in natural and relatively un- disturbed communities provides another example of the exercise of contrasting value judgements by ecologists and others. Given the fact that in complex rainforest com- munities, many species are represented by relatively few individuals in any one area, and remembering our ignorance of the numbers of individuals that constitute a sur- vival threshold for a particular species, what changes in species can be predicted in a given area as the result of artificial disturbance, such as logging? Patchiness or clustering of organisms at different scales in space and time — so-called intrinsic and extrinsic diversity — is difficult to explain, and both stochastic and determinate processes seem to be involved (Webb et al. , 1972; Letouzey, 1978; Connell et al. , 1982). Regeneration and succession have become ecological problems that have sprouted ideological roots. Several major types of succession have been described by different models, beginning with Egler (1954), and extending to Slatyer (1977) and Noble and Slatyer (1980). Alternative pathways and possible endpoints in reconstructive secondary succession of tropical and subtropical rainforest in north-eastern Australia Proc. LINN. Soc. N.S.W. 106 (4), (1982) 1983 270 ECOLOGICAL VALUES OF THE TROPICAL RAINFOREST RESOURCE were clearly explained by Hopkins (1981). These and other ecological studies show that regeneration and canopy development by secondary species in one rainforest type (e.g. simple notophyll vine forest characterized by Coachwood), are not comparable with another type (e.g. complex notophyll vine forest characterized by a mixture of species including Booyong). Data from long-term monitoring of species over representative areas are not available as a basis for scientific judgment, which has wider terms of reference than forestry decisions about adequacy of canopy closure by secondary species, and where the restoration of an ecosystem similar to the original one controls the time scale since logging. Scientifically, the long-term advantages of conserving diversity in natural communities have been emphasized by many biologists with interests ranging from population dynamics (e.g. Elton, 1958; Connell, Tracey and Webb, 1982) to evolutionary genetics (Frankel, 1970). So much for examples of ecological research which have harboured tacit value judgments and have involved intuitive reference to ecological precepts. We have also undertaken a project, with somewhat tenuous scientific links, to try to explicate, and thus to a limited extent communicate, our impressions as experienced ecologists of the rainforests. The attempt also has a practical basis: to determine criteria for selecting and managing national parks and similar conservation areas in northern Australia. Commercial values of material commodities, and protective values for soil and water are already accepted as of demonstrable economic worth. Attention is focused rather towards the other end — the cultural values — of the spectrum. The aim is to try to identify and objectify the values of tropical rainforest and associated forest and land- scape for scientific, educational, recreational, and aesthetic purposes (Kikkawa, Webb and Tracey, 1974). A co-operative project with K. J. Polakowski, Landscape Ar- chitecture, University of Michigan, to determine aesthetic values is of special interest. It partly uses techniques for identifying relevant attributes at the macro-level of land- scapes, by applying them to the micro-level of forestscapes. It describes perceptions of the rainforest environment in terms of ecological site factors, about which we have a great deal of quantitative information. The perceptions are then translated into a limited number of perceptual effects and emotional states, using various physical descriptors (vegetation structure, flora, fauna, sound, colour, light, etc.). The aesthetic experience of an (interested) observer is an interaction between his cultural and behavioural world and the intricacy and unfolding variety of the rainforest en- vironment (Polakowski, Webb and Kikkawa, 1982). This experience is, from a synecological point of view, a kind of ‘emergent property’ (cf. Odum, 1977). The complexity and variability of natural ecosystems including humans cannot be captured even by the most sophisticated statistical methods. It can be argued that the sense of aesthetics in such situations is not a luxury, but a tool for survival. Aesthetic appeal may play a central role as a human judgment of the viability of ecosystems, and thus be somehow advantageous (Grossmann, 1982). Although it is appreciated that the identification of very broad categories of aesthetic experience must remain sketchy, the conceptual framework does bring together, in a very satisfying way, complex scientific and other values that were otherwise isolated. The framework should provide a stimulating approach to forest environments for children in schools, and for interested adults in the community. It could also be used as a guide to Aboriginal environmental perceptions that are part of the heritage of natural history. Of wider ramifications geographically is our current attempt to match rainforest habitats in northern Australia and Indo-Malesian countries. This attempt is of theoretical and practical interest, and has social values. The aim of the project, partly funded by the UNESCO Man and the Biosphere (MAB) Project, is to use structural Proc. LINN. Soc. N.S.W. 106 (4), (1982) 1983 L. J. WEBB Zajfi| typology, supplemented by elementary environmental information and floristic data for common tree genera, to match forest types and habitats on a site-to-site basis. Given the ancient floristic links already noted between Australia and other south-east Asian countries, well-matched habitats with a high proportion of shared genera would suggest parallel evolution (homology) through common descent. Intuitive comparisons of forest types in the field during reciprocal visits of botanists to Thailand and Australia indicate a remarkable extent of shared genera in certain deciduous monsoonal types e.g. 66-90 per cent of Australian tree genera are shared with homologous habitats in Thailand. Evergreen lower montane forest types share fewer genera e.g. 33-55 per cent. Numerical analysis of the structural data is now being attempted, and will hopefully yield a more objective measure of comparison. Northern Australian rain- forest and monsoon forest patches, although small and scattered, provide the only vegetation for structural and floristic comparison with other countries. This means that the sclerophylls, that have mostly displaced the rainforests and monsoon forests, and that have superior adaptations to fire, drought, and low soil fertility, provide a unique reservoir for plant introduction elsewhere, once comparable habitat-types are established by the systematic use of structural data. Besides enriching ecologic biogeographical understanding, the comparisons could contribute to agricultural and forestry practice. More than this, habitat-matching could help develop a ‘common ecological language’ based on the biotic communities of the region, despite their many man-imposed variations. Perhaps we should also say that tracing this ancient lineage of the Australian flora contributes, in the best tradition of natural history, to the enlargement of Australian cultural identity. Although a common ecological language for the tropical region has yet to be written, there is no doubt in my mind that it will contain many cultural values and ethical overtones. I select only one recent scientific example, from the Man and the Biosphere (MAB) Workshop in Kuala Lumpur, January 1982. The speaker is the Director of the Forest Research Institute at Kepong, Peninsular Malaysia; and he is urging multi-species planting and the retention of adequately-sized blocks of natural forests in forestry areas: *.. . It is acknowledged that any of the above practices suggested could result in reduction of profits to the plantation owner. It is also true that the ‘pest’ population to the plantations could increase. However, it is our contention that sacrifice is necessary for the sake of our environment . . . To limit future generations to an environment of monotony is to limit their ultimate self-development. There is a moral obligation of society to provide the necessary environment to maintain the myriads of floral and fauna species of the world. Any loss and ultimate extinction of plant and animal life is a loss of human society’ (Salleh and Hashim, 1982). And so we return to where we began: to the meaning of values and their place in the ecological language that we all share in some way. Conventional ethics provide religious morality, and social morality and justice. These deal with man-man relationships. Is ecological morality and ecological justice now to be seen as a third- level development for man-nature relationships? Thus the question about the intrinsic values of nature remains. I believe there are signs that this question may be considered safely at the level of science. Environmental values will continue to be ‘good’ or to be ‘bad’. There seems no possibility of con- sensus, no definitive formulae, to decide how the protection of nature and the needs of modern society can be reconciled. But it is becoming increasingly clear that we must learn to live honestly with questions that will always remain open (Ashby, 1976). It seems wise to reject the concept of human domination, and to change the old ecological view from ‘man the outsider’ to ‘man the insider’ in ecosystems that involve nature (di Castri, 1981). Proc. LINN. Soc. N.S.W. 106 (4), (1982) 1983 272 ECOLOGICAL VALUES OF THE TROPICAL RAINFOREST RESOURCE A re-assertion of ecological values cannot avoid involving feelings and sympathies (Wright, 1973). Ecological value systems by themselves are not enough. Ecological values are the results of valuing processes in relation to ecological principles (such as interdependence, diversity, continuity and change, the long-term versus the short- term, and so on). The separation of scientific research and problem-solving from valuing, all of which are in turn integrated in decision-making, is becoming in- creasingly questionable. It seems that novel and complex ecological situations must be matched by revised ethics if man and his culture, and if nature (which he needs, but which does not need him) are to survive. We must find appropriate language, and appropriate values as part of it, for the description of complex ecological situations to be resolved by choice. These are the situations with tangled and seemingly incompatible strands: scientific values guided by the demands of objectivity, and ecological values grown indistinguishable from the responsibilities of care and preservation. International forestry authorities are beginning to accept ethical responsibilities in - forest management: for example, the recent statement by the President of the In- ternational Union of Forestry Research Organizations: “The post-war generation of foresters has witnessed the most savage and fast destruction of forests in the entire history of man’s work in the world. Quite often foresters are taking part in such destruc- tion . . . Research has become a part of those economic concepts which aim towards exploitation . . . More than any other economic branch, forestry has been sat in the dock of the world, accused of destroying nature, for which it is only partly to be blamed . . . Research must be based on ethical principles of responsibility . . . Forestry and its research have to return to nature, which so far has not been done. Tomorrow will take man back to nature, because there is no other way’ (Mlinsek, 1982). What this forester, as a scientist, is asserting is that moral choices are inescapable in research as well as its applications. We must learn to recognize and acknowledge these choices. Such acknowledgment should not weaken our pursuit of objectivity, but make us better scientists as we become aware of more and more of the components of the ecological equation. There are no simple solutions, but it seems clear that ethics and value judgments may no longer remain private affairs for individuals or bureaucracies that make decisions. Thus our values will evolve through the choices made. In other words, we will discover that the last vestiges of Australian rainforests are not so good to cut as they are ‘good to think’ (cf. Dwyer, 1982, p. 186). ACKNOWLEDGEMENTS I wish to thank Professor Jiro Kikkawa for several helpful suggestions after reading the draft of this lecture. Helpful comments on the final manuscript were also made by Mrs Judith Wright McKinney and Sir Rutherford Robertson. References ASHBY, E., 1976. — Towards an environmental ethic. Nature 262: 84-85. , 1978. — Reconciling Man With The Environment. Stanford: Stanford University Press. BARLOW, B. A., 1981. — The Australian flora: Its origin and evolution. In Flora of Australia, Vol. I, pp. 25- 75. Canberra: Australian Government Publishing Service. CONNELL, J. H., TRACEY, J. G., and WEBB, L. J., 1982. — The role of compensatory recruitment, growth and mortality in maintaining rainforest tree diversity. Ecological Monographs (in press). De’ATH, C., 1982. — Forest conservation practices in Papua New Guinea. In Traditional Conservation in Papua New Guinea: Implications for Today (Ed. L. MORAUTA, J. PERNETTA and W. HEANEY), pp. 203- 215. Monograph 16, Institute of Applied Social and Economic Research, Boroko, PNG. Proc. LINN. Soc. N.S.W. 106 (4), (1982) 1983 L. J. WEBB 273 DI CASTRI, F., 1981. — Ecology — the genesis of a science of man and nature. UNESCO Courier, April 1981, pp. 6-11. Dwyer, P. D., 1982. — Wildlife conservation and tradition in the highlands of Papua New Guinea: In Traditional Conservation in Papua New Guinea: Implications for Today. (Ed. L. MORAUTA, J. PERNETTA and W. HEANEY), pp. 173-189. Monograph 16, Institute of Applied Social and Economic Research, Boroko, PNG. EGLER, F. E., 1954. — Vegetation science concepts. I. Initial floristic composition — a factor in old field vegetation development. Vegetatio 4: 412-417. EHRENFELD, D. W., 1976. — The conservation of non-resources. American Scientist, 64: 648-656. ELTON, C.S., 1958. — The Ecology of Invasions by Animals and Plants. London: Methuen. FRANKEL, O. H., 1970. — Variation — The essence of life. Proc. Linn. Soc. N.S. W. 95: 158-169. GROSSMANN, W. D., 1982. — Viable systems: more complex, variable, even chaotic and more reliable than is often recognized. Computing Centre, University of Hamburg. MS. HALDANE, J. B. S., 1928. — Science and Ethics. Conway Memorial Lecture. London: Watts & Co. Hopkins, M. S., 1981. — Disturbance and change in rainforests and the resulting problems of functional classification. In Vegetation Classification in Australia. Ed. A. N. GILLISON and D. J. ANDERSON, pp. 42-52. Canberra: CSIRO and ANU Press. Jacoss, M., 1978a. — Botanical philosophy on the selection of rain forest reserves in Indonesia. Paper delivered to 8th World Forestry Congress, Jakarta, October 1978. ——,, 1978b. — Significance of the tropical rain forests on 12 points. Paper delivered to 8th World Forestry Congress, Jakarta, October 1978. KAMANDA wa KAMANDA, 1980. — Report for the inclusion of an item in the provisional Agenda of the Thirty Fifth Session. United Nations General Assembly. ‘Draft World Charter for Nature.’ A/35/141, 11 June 1980. KEAST, A., (Ed.), 1981. — Ecological Biogeography of Australia. 3 vols. The Hague: W. Junk. KIKKAWA, J., WEBB, L. J., and TRACEY, J. G., 1974. — A multidisciplinary project to establish criteria for selection and management of national parks in North Queensland. In Vol. 1, Working Papers, Seventh Tnennial Conference, Institute of Foresters of Australia, Caloundra, Queensland. , DALE, M. B., MONTEITH, G. B., TRACEY, J. G., and WILLIAMS, W. T., 1981. — Gradients and boundaries of monsoon forests in Australia. Proc. Ecol. Soc. Aust. 11: 39-52. LETOUZEY, R., 1978. — Floristic composition and typology. In Tropical Forest Ecosystems: A State-of- Knowledge Report. pp. 91-111. Paris: UNESCO. MLINSEK, D., 1982. — Preface to IUFRO News No. 37, International Union of Forestry Research Organisations, Vienna. Myers, N., 1978. — Conservation of forest animal and plant genetic resources in tropical rainforests. Paper delivered to 8th World Forestry Congress, Jakarta, October 1978. , 1980. — Conversion of Tropical Moist Forests. Washington: National Academy of Sciences. NOBLE, I. R., and SLATYER, R. O., 1980. — The use of vital attributes to predict successional changes in plant communities subject to recurrent disturbances. Vegetatio 43: 5-21. OpuM, E. P., 1977. — The emergence of ecology as a new integrative discipline. Sczence 195: 1289-1293. PANTIN, C. F. A., 1968. — The Relations Between The Sciences. Cambridge: Cambridge University Press. PASSMORE, J., 1974. — Man’s Responsibility For Nature. London: Duckworth. POLAKOWSKI, K., WEBB, L., and KIKKAWA, J., 1982. — Methods to derive perceptual descriptions, per- ceptual effects and aesthetic values for tropical rainforests. MS. : PoorE, D., 1976. — Ecological Guidelines for Development in Tropical Rain Forests. Morges (Switzerland): IUCN. PRINEAS, P., and ELENIUS, E., 1980. — Why log Terania Creek? National Parks Assn. of New South Wales, 399 Pitt St., Sydney. SALIM, E., 1982. — The second World Conservation Lecture, delivered at the Royal Institution on 18 March 1982. World Wildlife Fund, Switzerland. SALLEH MOHD Nor and HASHIM MouD Nor, 1982. — Plantations — are we creating a sterile system? Paper presented at the Workshop on Ecological Basis for Rational Utilization in the Humid Tropics of South East Asia. U.P.M. Serdang, Malaysia, 18-22 January 1982. SLATYER, R. O., 1977. — Dynamic changes in terrestrial ecosystems: patterns of change, techniques for study, and applications to management. Paris: UNESCO. SOEMARWOTO, O., 1982. — Proceedings of Man and the Biosphere Workshop on rational utilization of resources in the humid tropics, Universiti Pertanian Malaysia, Serdang, January 1982 (in press). SyNGE, H., 1982. — Endangered plant conservation: Looking ahead. JUCN Bulletin (n.s.) 13: 33. Tracey, J. G., 1981. — Australia’s rainforests: where are the rare plants and how do we keep them? In The Biological Aspects of Rare Plant Conservation, Ed. H. SYNGE, pp. 165-178. London: John Wiley. Wess, L. J., 1977. — Ecological considerations and safeguards in the modern use of tropical lowland Proc. LINN. Soc. N.S.W. 106 (4), (1982) 1983 274 ECOLOGICAL VALUES OF THE TROPICAL RAINFOREST RESOURCE rainforests as a source of pulpwood: example, the Madang area, Papua New Guinea. Office of Environment and Conservation, Waigani, PNG. , and TRACEY, J. G., 1981a. — Australian rainforests: Patterns and change. In Ecological Biogeography of Australia, Ed. A. KEAST, pp. 607-694. The Hague: W. Junk. , and , 1981b. — The rainforests of northern Australia. In Australian Vegetation, Ed. R. H. GROVES, an. 67-101. Cambridge: Cambridge University Press. ; , and Jessup, L. W., 1982. — Recent evidence for autochthony of Meiswalinn tropical and subtropical rainforest fois elements. Telopea (in press). —, , and WILLIAMS, W. T., 1972. — Regeneration and pattern in the subtropical rainforest. /. Ecol. 60: 675-95. —-, , and , 1976. — The value of structural features in tropical forest typology. Aust. J. Ecol. 1: 3-28. ——., , and , 1983. — A floristic framework of Australian rainforests. MS. —, 5 , and LANCE, G. N., 1970. — Studies in the numerical analysis of complex rainforest communities. V. A comparison of the properties of floristic and physiognomic-structural data. /. Ecol. 58: 203-232. WRIGHT, J., 1973. — The individual in a new environmental age. Austral. J. Social Issues 8: 3-9. Proc. LINN. Soc. N.S.W. 106 (4), (1982) 1983 A new Genus of Spiders of the Subfamily Metaltellinae (Araneae, Amaurobioidea) from southeastern Australia M. R. GRAY Gray, M. R. A new genus of spiders of the subfamily Metaltellinae (Araneae, Amaurobioidea) from southeastern Australia. Proc. Linn. Soc. N.S.W. 106 (4), (1982) 1983: 275-285. The metaltelline genus, Austmusza gen. nov., from southeastern Australia and three species, A. wilsoni sp. nov., A. kioloa sp. nov., and A. lindi sp. nov., are described. This represents the first record of the subfamily Metaltellinae outside the neotropical region. The presence of a simplified tracheal system in Austmusza is noted. M. R. Gray, The Australian Museum, P.O. Box A285, Sydney South, Australia, 2000; manuscript recewed 11 May 1982, accepted for publication in revised form 20 October 1982. INTRODUCTION The subfamily Metaltellinae has previously been recorded only from the southern and central neotropical region (synanthropic in the southern nearctic (Leech, 1971) ). Lehtinen (1967) recognized four genera in the subfamily: Metaltella Mello-Leitao 1931, Ciniflella Mello-Leitao 1921, Calacadia Exline 1940 and Exlinea Lehtinen 1967. Leech (1971) synonymized Exlinea with Metaltella and noted (Leech, 1972) that Ciznzflella should be regarded as a nomen dubium (type material lost, description inadequate). Metaltella is a cribellate genus while both Calacadia and the genus described below, Austmusia gen. nov. from eastern Australia, are ecribellate. These genera are ap- parently derived from a palaeoaustral fauna which occupied the southern parts of Gondwanaland. METALTELLINAE Lehtinen 1967 Diagnosis: Ecribellate or cribellate; cribellum bipartite. Chelicerae with 2-6 retrolateral and 3-7 prolateral teeth. Labium longer than wide. Male palp with reduced primary conductor. Secondary conductor a large, falciform, anteriorly-directed process en- closing a long, curved, slender emboius. Tegulum more or less bulbous basally. Median apophysis absent. Fixed anterior tegular apophysis present or absent. Epigynum with a large, flat, well-sclerotized plate, lateral teeth present or absent. Internal genitalia usually with convoluted ducts, sometimes simple, with associated diverticula. KEY TO GENERA 1. Ecribellate. Cheliceral teeth 2 (retro.) and 2-3 (pro.) Epigynum with Sinomfalerbercl PeetmOIaMOMe: 25 s:5) 4 4.5. ec vik- ahs costae 4 ue ops Saaeeee nae eee eae 2 — Cribellate. Cheliceral teeth 4-6 and 5-7. Epigynum with elongate lateral OSU Maal ohk Bee aie hae eS aE De PI ho SPO ac onic eto Metaltella 2. Male palp with large patellar apophysis. Epigynal plate unmodified .. Calacadia — Male palp lacking patellar process. Epigynal plate with a pair of depressions roofed anteriorly by thinlaminae .................. Austmusia Austmusia gen. nov. Diagnosis: Ecribellate. Cheliceral groove with 2 retrolateral, 3 prolateral teeth. Male Proc. LINN. Soc. N.S.W. 106 (4), (1982) 1983 276 A NEW GENUS OF SPIDERS Figs 1-4. Body pattern, females. 1 and 2, Austmusia wilsoni sp. nov. 1, lateral; 2, dorsal. 3, A. kzoloa sp. nov., dorsal abdomen. 4, A. lindi sp. nov., dorsal abdomen. Scale lines 0.45 mm. Proc. LINN. Soc. N.S.W. 106 (4), (1982) 1983 M.R.GRAY 277 palp with a vestigial primary conductor. Basal tegulum bulbous; tegular process hook- like. Patellar process absent. Epigynum with two shallow depressions roofed anteriorly by thin cuticular laminae. Two tracheal tubes confined to abdomen. Description: Large spiders, 10 to 14 mm body length. Carapace dark brown, darkest anteriorly; central cephalic and marginal thoracic areas lighter orange brown. Ab- domen dark brownish-grey dorsally with well-defined fawn markings (Figs 1-4). Ventral abdomen light brown with two lateral, dark grey, longitudinal stripes. Chelicerae black with an orange boss. Legs banded dark and light brown. Carapace (Figs 1, 2) almost glabrous, cephalic area prominent, arched longitudinally, raised at fovea. Clypeus vertical, height three to four times the width of an A.M.E. Anterior median eyes smallest. From above both eye rows straight to slightly recurved; from in front both rows procurved (Fig. 8). Cheliceral teeth, Fig. 7; maxillae, labium, sternum, Fig. 5. Sternum with three prominent groups of lyriform organs near each lateral margin. Legs 4123 (females) or 1423 (males). Trochanters weakly notched. Hairs plumose, feathery hairs absent. Ventral spination, legs 1 and 2, tibia 222, metatarsus 221. Tarsal organ tear-shaped, widest distally (Figs 18, 19). Tiichobothria in single row on tarsi and metatarsi, double row on tibiae. Trichobothrial bases collariform; hood striated, margin entire (Figs 16, 17). Superior tarsal claws with eight to thirteen teeth; inferior claw with four to eight teeth. Toothed bristles present ventrally on distal tarsus. Abdominal spiracle narrow and situated just in front of colulus. Colulus a large, rectangular, pilose plate, slightly wider than long, distal margin slightly notched (Fig. 6). Spinnerets six, short. Epiandrous glands absent. Tracheal system consists of two tubes only, oimined to abdomen (Fig. 31). Palpal organ (Figs 9-11) with a vestigial primary conductor. Basal tegulum bulbous, Figs 5-8. Austmusia wilsoni sp. nov. 5, maxillae, labium, sternum. 6, spinnerets, colulus, spiracle. 7, cheliceral teeth. 8, eyes, anterior view. Scale lines 0.5 mm. Proc. LINN. Soc. N.S.W. 106 (4), (1982) 1983 278 A NEW GENUS OF SPIDERS Figs 9-15. Austmusia wilsoni sp. nov. 9-11, male palp. 9, prolateral; 10, ventral; 11, retrolateral (e = em- bolus, ta = tegular apophysis, 1° = primary conductor, 2° = secondary conductor). 12-13, male palpal tibia and patella; 12, retrolateral; 13, ventral. 14-15, female genitalia. 14, epigynum; 15, internal genitalia, dorsal, left hand side. Scale lines 0.2 mm. PROC. LINN. SOc. N.S.W. 106 (4), (1982) 1983 M.R.GRAY 279 separated from apical part by a deep transverse furrow. Apical tegulum with a fixed hook-like process arising subapically on retrolateral side. Ejaculatory duct visible on lower tegulum. Male palpal tibia (Figs 12-13) with a prominent, retrolateral basal process and a retrolateral ventral lamina which ends apically in a small blunt hook. Patellar processes absent. Epigynum (Fig. 14) broad and strongly sclerotized with a pair of large, shallow depressions, open posteriorly but roofed anteriorly by thin cuticular laminae. Lateral teeth absent. External foveae placed in the posterior ends of two deep, curved grooves near each lateral epigynal margin. Internal genitalia (Fig. 15) with convoluted sper- mathecal ducts (partly visible externally) into which a pair of diverticula open. Spermathecae adjacent near midline, elongate, narrowest centrally. Fertilization ducts short, broad. Type Species: Austmusia wilsoni sp. nov. Etymology: Exline (1960) named the Chilean metaltelline genus Calacadia after the Californian Academy of Sciences. This provides a precedent for similarly naming the closely related Australian genus, Austmusza, after its repository institution, The Australian Museum. Type Repository: The Australian Museum, Sydney (A.M.). Austmusia wilsoni sp. nov. (Figs 1-2, 9-15, 16, 18, 31) Diagnosis: Male palp with apex of primary conductor level with highest part of tegular process (Fig. 11). On the basal tegulum a broad segment of the ejaculatory duct loops apically between two slender diverging segments of the duct (Figs 9, 10). Female genitalia with long spermathecae, the anterior ends bent 45° laterally (Fig. 15). Female holotype (A.M., KS 5736) Measurements (mm) — Carapace length 5.57 (5.10-6.40), width 3.51 (3.19-3.98), height 2.21 (2.02-2.50). Abdomen length 5.50 (5.25-6.10), width 4.25 (3.80-4.65). Colour pattern — as for genus. Posterior paired abdominal markings often fused medially. Carapace — longer than wide in ratio 1: 0.63. Clypeus height 4 diameters of an A.M.E. Eyes — ALE > PLE > PME > AME in ratio of 1: 0.87: 0.74: 0.52. Interdistances (mm): AME-AME 0.14, AME-ALE 0.15, ALE-PLE 0.13, PLE-PME 0.38, PME-PME 0.26. M.O.Q. (mm): length 0.58, anterior width 0.44, posterior width 0.64. Eye row width (mm): anterior 1.25; posterior 1.50. From above A.E.R. straight, P.E.R. recurved; from in front A.E.R. gently procurved, P.E.R. strongly procurved. Labium — longer than wide in ratio 1: 0.63 Sternum — longer than wide in ratio 1: 0.83 Legs: 4123. Lengths, legs 1-4 (mm): 18.28, 16.38, 14.94, 18.55. Spination: Leg 1, femur p 01, d 213; tibia p 11, r 11, v 222; metatarsus p 012, r 012, v 221. Leg 2, femur p 0111, d 01212; tibia p 011, r 001, v 222; metatarsus p 112, r 012, v 221. Leg 3, femur p 0111, d 1202; tibia p 011, r011, d 010, v 222; metatarsus p 112, r 112, d 010, v 221. Leg 4, femur p 101, d 1102; tibia p 101, r 101, d 01, v 212; metatarsus p 112 or 111,r 112 or 111, d 010 or 011, v 221. Tarsal organ a narrow, tear-shaped opening, widest distally, not on obvious mound. Trichobothria, legs 1-4: tarsus 10, 9, 9, 9; metatarsus 9,9, 9, 9. Tarsal claws; superior 11-13 teeth; inferior 4-8 teeth. Genitalia — Figs 14, 15. Spermathecae visible externally, elongate with the apical third to half bent 45° laterally. Spermathecal length approximately half the width of the epigynum (measured between the lateral foveal margins). Diverticulum duct narrow Proc. LINN. Soc. N.S.W. 106 (4), (1982) 1983 280 A NEW GENUS OF SPIDERS 2 Bae Figs 16-19. Sensilla, first tarsus. 16-17, trichobothrial base. 16, Austmusza wilsoni; 17, A. kioloa. 18-19, tarsal organ. 18, A. wilsonz; 19, A. kioloa. Scale lines: 16-17, 0.006 mm; 18-19, 0.002 mm. with a small, rounded, distal part which lies well lateral to and below the level of the anterior end of the spermatheca. Fertilization ducts short and broad. Male paratype (A.M., KS 7795) Similar to female except as follows— Measurements (mm) — Carapace length 5.49 (4.59-6.41), width 3.79 (3.20-4.21), height 2.16 (1.98-2.45). Abdomen length 5.30 (4.85-5.75), width 3.50 (3.00-3.80). Carapace — longer than wide in ratio 1: 0.67. Clypeus height 3 diameters of an AME. Eyes PLE > ALE > PME > AME in ratio 1: 0.95: 0.81: 0.72. Interdistances (mm): AME-AME 0.08, AME-ALE 0.14, ALE-PLE 0.06, PLE-PME 0.26, PME-PME 0.18. M.O.Q. (mm): length 0.58, anterior width 0.42, posterior width 0.58. Eye row width (mm); anterior 1.10; posterior 1.29. Chelicerae — cheliceral teeth: retrolateral 2; prolateral 3 with an additional microtooth. Labium — longer than wide in ratio 1: 0.79. Sternum — longer than wide in ratio 1: 0.78. Legs — 1423. Length, legs 1-4 (mm): 19.04, 16.76, 14.12, 18.84. Spination: Leg 1, femur p 0011, d 122; tibia p 112, r 011, v 222; metatarsus p 11, r 010, v 221. Leg 2, p 1011, d 122; tibia p 11, r 11, v 222; metatarsus p 112, r012, v 221. Leg 3 femur p 101, d 122; tibia p 11, r 11, d (1)01, v 22; metatarsus p 112, r 112, d 010, v 221. Leg 4, femur p 101, d 112; tibia p 11, r 011, d(1)01, v 212; metatarsus p 112, r 112, d 010, v 221. Trichobothria, legs 1-4: tarsus 9, 8, 8, 9; metatarsus 10, 9, 8, 8. Tarsal claws: superior 11-12; inferior 4-8. Palp — Figs 9-13. Apex of reduced primary conductor level with highest part of hooked tegular process. Course of ejaculatory duct on basal tegulum — Figs 12, 13. Proc. LINN. Soc. N.S.W. 106 (4), (1982) 1983 M.R.GRAY 281 Types: Holotype female — KS 5736 (A.M.), Cathedral of Ferns, Mt. Wilson, N.S.W., M. Gray, 17.4.1974; under log, depauperate closed forest. Paratypes — Male, KS 7795 (A.M.), Mt. Wilson, N.S.W., M. Gray and C. Hor- seman, May 1978; in pitfall trap, tall open forest. Female, KS 2046 (A.M.), Mt. Wilson, N.S.W., M. Gray and C. Horseman, October/November 1978; in pitfall trap, tall open forest. Male, KS 3877 (A.M.), Waterfall Trail, Mt. Wilson, C. Horseman, August 1979; in pitfall trap, closed forest. Male, KS 7796 (A.M.), Mt. Victoria, 14.5.1972, G. S. Hunt; under log in valley, tall open forest. Male, KS 7797 (A.M.), Mt. Edwards, Boyd Plateau, N.S.W., M. Gray, 15.5.1971; under log, tall open forest. Female, KS 7798 (A.M.), Blood Filly Creek, Boyd Plateau, N.S.W., M. Gray 26.3.1976; in rotting log, tall open forest. Austmusia kioloa sp. nov. (Figs. 3, 17, 19, 20-26) Similar to A. wilsoni except as follows: Diagnosis: Male palp with apex of primary conductor clearly lower than highest part of tegular process (Fig. 22). Ventrally, two slender segments of ejaculatory duct on the basal tegulum first converge anteriorly, then diverge, the prolateral duct curving laterally away (Figs 20, 21). Spermathecae long, apical third enlarged and laterally protuberant (Fig. 26). Female holotype (A.M., KS 4551). Measurements (mm) — Carapace length 4.87 (4.48-5.25), width 3.15 (2.88-3.43), height 1.89 (1.60-2.18). Abdomen length 5.70 (5.45-5.95), width 4.70 (4.50-4.85). Carapace — longer than wide in ratio 1: 0.65. Clypeus height 3.5 diameters of an AME. Eyes — ALE > PLE > PME > AME in ratio 1; 0.95: 0.82: 0.68. Interdistances (mm): AME-AME 0.11, AME-ALE 0.16, ALE-PLE 0.08, PLE-PME 0.30, PME-PME 0.23. M.O.Q. (mm): length 0.53, anterior width 0.39, posterior width 0.61. Eye row width (mm), anterior 1.09; posterior 1.37. Labium — longer than wide in ratio 1: 0.80. Sternum — longer than wide in ratio 1: 0.81. Legs — 4123. Lengths, legs 1-4 (mm): 14.68, 12.79, 11.37, 15.36. Spination: Leg 1, femur p 002, d 111; tibia p 101, r 101, v 222; metatarsus p 111, r 012, d 010, v 221. Leg 2, femur p 101, d 1112; tibia p 101, r 101, v 222; metatarsus p 112, r012,d 010, v 221. Leg 3, femur p 101, d 1112; tibiap 11,r101,d 11, v 212; metatarsus p 112, r 112, d 010, v 221. Leg 4, femur p 001, d 112; tibia p 101, r 101, d 01, v 212; metatarsus p 111,r111,d 012, v 221. Tarsal organ a tear-shaped opening, widest distally, placed on alow, poorly delimited mound. Trichobothria, legs 1-4: tarsus 8, 8, 7, 8; metatarsus 8, 9, 7, 8. Tarsal claws: superior 8-11 teeth; inferior 4-7 teeth. Genitalia — Figs 25, 26. Basal parts of spermathecae visible externally. Spermathecae elongate, apical third laterally bulbous and rounded. Spermathecal length greater than half epigynal width. Diverticulum duct narrow, distal part elongate, lying over or adjacent to apical part of spermathecae. Fertilization ducts longer than in A. wzlsonz. Male paratype (A.M., KS 1651) Similar to female except as follows: Measurements (mm) — Carapace length 4.67 (3.90-5.15), width 3.37 (2.93-3.60), height 1.70 (1.45-2.03). Abdomen length 4.20 (3.80-4.30), width 3.10 (2.85-3.25). Carapace — longer than wide in ratio 1: 0.73. Clypeus height 3 diameters of an A.M.E. PROC. LINN. Soc. N.S.W. 106 (4), (1982) 1983 282 ANEW GENUS OF SPIDERS Figs 20-26. Austmusia kioloa sp. nov. 20-22, male palp. 20, prolateral; 21, ventral; 22, retrolateral. 23-24, male palpal tibia and patella. 23, ventral; 24, retrolateral. 25-26, female genitalia. 25, epigynum; 26, internal genitalia, dorsal, left hand side. Scale lines 0.2 mm. Proc. LINN. Soc. N.S.W. 106 (4), (1982) 1983 M.R. GRAY 983 Eyes — ALE = PLE > PME > AME in ratio 1: 1: 0.95: 0.85. Interdistances (mm): AME-AME 0.10, AME-ALE 0.14, ALE-PLE 0.08, PLE-PME 0.25, PME-PME 0.20. M.O.Q. (mm.): length 0.51, anterior width 0.44, posterior width 0.55. Eye row width (mm): anterior 1.03; posterior 1.30. Labium — longer than wide in ratio 1: 0.72. Sternum — longer than wide in ratio 1: 0.83. Legs — 1423. Lengths, legs 1-4 (mm): 17.42, 15.42, 13.53, 16.83. Spination: Leg 1, femur p 002, d 1111; tibia p 101, r 101, v 222; metatarsus p 112, r 012, v 221. Leg 2, femur p 101, d 1112; tibia p 101, r 101, v 222; metatarsus p 111, r 011, d 012, v 221. Leg 3, femur p 101, d 1112; tibia p 11, r 101, d 01, v 212; metatarsus p 111, r111,d 012, v 221. Leg 4, femur p 101, d 112; tibia p 101, r 101, d 01, v 212; metatarsus p 111,r111,d 012, v 221. Trichobothria, legs 1-4: tarsus 10, 9, 8, 9; metatarsus 8, 8, 9, 9. Tarsal claws: superior 8-10 teeth; inferior 4-7 teeth. Palp — Figs 20-24. Apex of reduced primary conductor well below the highest part of the hooked tegular process. Course of ejaculatory duct on basal tegulum — Figs 20, 21. Types Holotype female — KS 4551 (A.M.), Kioloa Forest Drive, Kioloa S.F., N.S.W., M. Gray and C. Horseman, May/June 1979; in pitfall trap, tall open forest. Paratypes — Male, KS 1651 (A.M.), Kioloa S.F. (rest area on Princes Highway), N.S.W., M. Gray, 14.8.1978; under litter, tall open forest. Male, KS 3115 (A.M.), Kioloa Forest Drive, Kioloa S.F., N.S.W., C. Horseman, May/June 1979; in pitfall trap, tall open forest. Female, KS 2810 (A.M.), same data as KS 3115, March 1979. Male, KS 1730 (A.M.), same data as KS 3115, August 1978. Austmusia lind: sp. nov. (Figs 4, 27, 28) Similar to A. wilsoni except as follows: Diagnosis: Female genitalia with short spermathecae, visible externally, apical end swollen; spermathecal length less than half the width of the epigynum (between lateral foveal margins). Female holotype (A.M., KS 1777) Figs 27-28. Austmusia lindi sp. nov., female genitalia. 27, epigynum. 28, internal genitalia, dorsal, left hand side. Scale lines 0.2 mm. PROC. LINN. Soc. N.S.W. 106 (4), (1982) 1983 284 A NEW GENUS OF SPIDERS 3] 30 29 Figs 29-31. Abdominal tracheal system. 29, Badumna insignis (L.K.). 30, Metaltella stmoni (Keysl.). 31, Austmusia wilsont sp. nov. Scale lines 0.4 mm. Measurements (mm) — Carapace length 6.42, width 4.07, height 2.50. Abdomen length 6.7, width 4.2. Colour pattern — dorsal, paired abdominal markings not fused posteriorly. Carapace — longer than wide in ratio 1: 0.63. Clypeus height 3.5 diameters of an AME. Eyes — ALE = PLE > PME > AME in ratio 1: 1: 0.83: 0.78. Interdistances (mm): AME-AME 0.18, AME-ALE 0.19, ALE-PLE 0.12, PLE-PME 0.38, PME-PME 0.30. M.O.Q. (mm): length 0.64, anterior width 0.52, posterior width 0.70. Eye row width (mm): anterior 1.35, posterior 1.73. Labium — longer than wide in ratio 1: 0.77. Sternum — longer than wide in ratio 1: 0.70. Legs — 4123. Lengths, legs 1-4 (mm): 17.53, 15.72, 13.71, 18.13. Spination: Leg 1, femur p 002, d 11(2); tibia p 101, r 001, v 222; metatarsus p 111, r 011, d 002, v 221. Leg 2, femur p 101, d 122; tibia p 101, r 111, v 222; metatarsus p 111, r011,d012,v 221. Leg 3, femur p 101, d 122; tibia p 11, r 11, d(1)1, v 212; metatarsus p 1101, r 111, d 012, v 221. Leg 4, femur p 001, d 112; tibia p 101, r 101, d (1) 01, v 212; metatarsus p 111, r 111, d 012, v 221. Trichobothria, legs 1-4; tarsus 8, 8, 7, 7; metatarsus 8, 8, 7, 7. Tarsal claws; superior 9-13 teeth; inferior 4-8 teeth. Genitalia — Figs 27, 28. Spermathecae relatively short, less than half the width of the epigynum (between the lateral foveal margins); anterior third swollen. Diverticula large, irregularly ovoid, apex lying just lateral to and above the anterior end of the spermathecae; duct broad. Fertilization ducts longer than in A. walsonz. Types Holotype female — KS 1777 (A.M.), Dingo Creek, Lind National Park, Victoria, M. Gray and C. Horseman, 24.3.1978; in rotting log, mixed closed/tall open forest. NOTES ON TRACHEAL SYSTEMS The tracheal respiratory system of Austmusia gen. nov., is unique within the PROG. LINN. SOc. N.S.W. 106 (4), (1982) 1983 M.R.GRAY 285 Amaurobioidea (sensu Lehtinen, 1967) so far investigated. Within the Dic- tynioidea/Amaurobioidea (sensu Forster, 1970; Forster and Wilton, 1973) two basic tracheal patterns have been recognized: A ‘complex’ pattern consisting of four finely branching tracheal tubes, often arising from a wide spiracle, either confined to the abdomen (Fig. 29) or extending through into the cephalothorax; and a simple pattern consisting of four unbranched tracheal tubes, usually arising from a narrow spiracle, and confined to the abdomen (Fig. 30). In Austmusia, the pattern is further simplified (Fig. 31), only the two medial tracheal tubes being present; they are unbranched, confined to the abdomen and arise from a narrow spiracle placed immediately in front of the spinnerets (Fig. 6). However, this pattern is not characteristic of all metaltelline spiders. Metaltella smoni (Keyserling, 1877) has the simple pattern of four unbranched abdominal tracheal tubes (Fig. 30). Unfortunately no specimens of Calacadia spp. were available for dissection. ACKNOWLEDGEMENTS Type material of Calacadia Exline was made available by Dr F. Arnaud, California Academy of Sciences. A male of Metaltella simoni (Keys.) was lent by Dr H. W. Levi, Museum of Comparative Zoology, Harvard. References EXLINE, H., 1960. — Rhoicinine spiders (Pisauridae) from western South America. Proc. Calif. Acad. Sct. 4th Ser. 29 (17): 577-620. FORSTER, R. R., 1970. — The spiders of New Zealand, Part III. Otago Mus. Bull. 3: 1-184. , and WILTON, C. L., 1973. — The spiders of New Zealand, Part IV. Otago Mus. Bull. 4: 1-309. LEECH, R., 1971. — The introduced Amaurobiidae of North America and Callobius hokkaido n.sp. from Japan (Arachnida: Araneidae). Canad. Ent. 103: 23-32. ——.,, 1972. — A revision of the Nearctic Amaurobiidae (Arachnida: Araneida). Mem. Ent. Soc. Canad. No. 84: 1-128. LEHTINEN, P. T., 1967. — Classification of the cribellate spiders and some allied families, with notes on the evolution of the suborder Araneomorpha. Ann. Zool. Fenn. 4: 199-468. A Reappraisal of the Stratigraphy of the upper Shoalhaven Group and lower Illawarra Coal Measures, southern Sydney Basin, New South Wales PAUL F. CARR CarR, P. F. A reappraisal of the stratigraphy of the upper Shoalhaven Group and lower Illawarra Coal Measures, southern Sydney Basin, New South Wales. Proc Linn. Soc. N.S. W. 106 (4), (1982) 1983: 287-297. Recent mapping and reappraisal of previous work have clarified the stratigraphy and nomenclature of the upper Shoalhaven Group and lower Illawarra Coal Measures of the southern Sydney Basin. The interval between the Berry Siitstone and Illawarra Coal Measures is termed the Broughton Formation which can be subdivided in the Gerringong-Wollongong region where distinctive latite flows occur. In stratigraphic sequence the members of the Broughton Formation are Westley Park Sandstone (lowermost unit), Blow Hole Latite, Kiama Sandstone, Bumbo Latite, Jamberoo Sandstone, Saddleback Latite, Dapto Latite and Cambewarra Latite (uppermost unit). The Saddleback and Dapto Latite Members are considered to be separate flows which occur at similar stratigraphic positions below the Cambewarra Latite. The Pheasants Nest Formation of the Illawarra Coal Measures overlies the Broughton Formation and contains four latite members: Five Islands, Calderwood (new unit), Minumurra and Berkley. P. F. Carr, Department of Geology, University of Wollongong, P.O. Box 1144, Wollongong, Australia 2500; manuscript received 5 August 1981, accepted for publication in revised form 18 August 1982. INTRODUCTION The southern Sydney Basin contains Permian and Triassic marine and non- marine sedimentary rocks which were subjected to four phases of igneous activity between the Late Permian and the Late Oligocene (Carr and Facer, 1980). Permian rocks are subdivided into two groups, the lower of which comprises the marine strata of the Shoalhaven Group, the other consists of the overlying non-marine strata of the Illawarra Coal Measures. Diversity of opinion concerning the mode of emplacement, and the number, of Late Permian igneous rock-units which crop out in the Wollongong-Gerringong region (Fig. 1) has been a major factor in the lack of agreement on the stratigraphic status and relationships of many Late Permian units in the southern Sydney Basin. Remapping by the author has clarified the stratigraphic nomenclature of the upper Shoalhaven Group and lowermost Illawarra Coal Measures. Rationalization of this confused terminology prompts the present author to propose one new name and a number of changes in status. Important previously published stratigraphic schemes and the scheme proposed herein are shown in Fig. 2. STRATIGRAPHIC SETTING The Permian sequence in the southern Sydney Basin unconformably overlies earlier Palaeozoic strata. In the southern extremity of the basin minor coal measure sedimentation (Clyde Coal Measures) occurred as a lateral western equivalent to the shallow marine Conjola Sub-group. This latter unit is conformably overlain by the Wandrawandian Siltstone which in turn underlies the Nowra Sandstone. Boundaries between all the units in the lower Shoalhaven Group represent facies changes which are slightly diachronous. All of the units so distinguished appear to grade westwards into Proc. LINN. Soc. N.S.W. 106 (4), (1982) 1983 288 STRATIGRAPHY OF THE SOUTHERN SYDNEY BASIN OSI [aa] sedimentary and SSS >: WOLLONGONG PERMIAN igneous rocks =i f:: ILLAWARRA 3 = COAL MEASURES —— a FLINDERS PERMIAN is KEMBLA * ISLET SHOALHAVEN e a GROUP fo) fo) fe} _STOCKYARD 2 MOUNTAIN , 2 -~ ce) KANGAROO [- o VALLEY 5 : GERRINGONG CAMBEWARRA ° MOUNTAIN Fig. 1. Locality map and simplified geology of the Wollongong-Gerringong region. the Megalong Conglomerate (McElroy e¢ al., 1969). The Berry Siltstone conformably overlies both the Nowra Sandstone and the Megalong Conglomerate. Joplin e al. (1952) grouped the marine and igneous rocks between the Berry Siltstone and Illawarra Coal Measures into the Gerringong Volcanics and named the unit above the Berry Siltstone the Broughton Tuff. They considered the Broughton Tuff to be divisible in the Gerringong-Wollongong region because of the presence of distinctive extrusive units. Rose (1966) included the Gerringong Volcanics as a Sub- group of the Shoalhaven Group. He recognized the Budgong Sandstone Member (of the Berry Formation) which he considered was stratigraphically partially equivalent to the Broughton Tuff. The distinctive igneous units near the top of the Shoalhaven Proc. LINN. Soc. N.S.W. 106 (4), (1982) 1983 PAUL F. CARR 289 Group had formation status (Harper, 1915; Joplin et al., 1952; Rose, 1966) but were termed members of the “‘Gerringong volcanic facies’? by Bowman (1970, 1974). Sandstones between these igneous rocks have been described as formations (Harper, 1915; Rose, 1966) and as members (Joplin e a/., 1952; Bowman, 1970, 1974). West of the volcanic flows the interval between the Berry Siltstone and the Illawarra Coal Measures has also had a varied nomenclature including Broughton Tuff (Joplin et al., 1952; Rose, 1966) and Budgong Sandstone (Bowman, 1970, 1974). Recent discussions on the ages of the marine faunas of the Shoalhaven Group may be found in Runnegar and McClung (1975) and McClung (1978). The Shoalhaven Group is conformably overlain by the non-marine deposits of the Pheasants Nest Formation which is the lowermost unit of the Illawarra Coal Measures. Harper (1915), Joplin e al. (1952) and Bowman (1970, 1974) have recognized two distinctive igneous rock-units within the Pheasants Nest Formation. BROUGHTON FORMATION In the present study the interval between the Berry Siltstone and the top of the Shoalhaven Group is termed the Broughton Formation which is subdivided into three sandstone members and five latite members in the Wollongong-Gerringong area. The basal unit of the formation is the Westley Park Sandstone Member which conformably and gradationally overlies the Berry Siltstone at Mount Coolangatta (Fig. 1). The nature of this boundary was recognized by Card and Jaquet (1903) and McElroy et al. (1969). Subdivision of the formation is entirely based on and, indeed, is only made possible by the presence of interbedded latites. Problems in stratigraphic terminology have arisen due to restriction, in large part, of some volcanic flows to the coastal strip. The top of the Cambewarra Latite Member or equivalent boulder horizon marks the top of the Shoalhaven Group (Harper, 1915, p. 223) and thus the top of the Broughton Formation. The name Broughton is chosen by the author in preference to Gerringong (equal priority) as the former proposal appears to reflect most closely the original intention of the nomenclature proposed by Jaquet et al. (1905, plate VII). In addition, usage of the name Broughton Formation allows the term Gerringong volcanics to be used as an informal name for the succession characterized by volcanic flows in the upper Shoalhaven Group and lower Illawarra Coal Measures. The term Budgong is rejected since it was introduced by Rose (1966) long after both Broughton and Gerringong had been initially defined (Joplin et al., 1952). Westley Park Sandstone Member. Jaquet et al. (1905) defined the Westley Park Tuffs which were renamed the Westley Park Tuff Member, Westley Park Sandstone and Westley Park Sandstone Member by Joplin ef al. (1952), Raam (1968) and Bowman (1970, 1974) respectively. The unit gradationally and conformably overlies the Berry Siltstone at Mount Coolangatta and is conformably overlain by the Blow Hole Latite Member (Fig. 2). The unit has been described by Raam (1968, 1969) as a shallow marine arenite with minor siltstone and conglomerate, and attains a maximum thickness of 45 m. Blow Hole Latite Member. Jaquet et al. (1905) named the Blow Hole Flow which was renamed the Blow Hole Latite by Joplin et al. (1952). The member is stratigraphically the lowest of the Permian flows in the Kiama-Wollongong region with present outcrops confined to a narrow zone from Kiama southwards for 22 km to Mount Coolangatta (Fig. 1). A maximum exposed thickness of 50 m occurs on the eastern side of Sad- dleback Mountain (Fig. 1). Bowman (1974) has reported that the unit consists of three flows. Proc. LINN. Soc. N.S.W. 106 (4), (1982) 1983 STRATIGRAPHY OF THE SOUTHERN SYDNEY BASIN 290 eu0}syis Aueg uoNBws04 peey dsea uojjBWws04 yoveg Aeqqed uo}}BWwI04 julog seddeus QUu0ISHIS UB/PUBMBIPUBM j|QUOJSpUBS BIMON SQUIOW GUO}SpUBS yueg Aeysem JOQUEW ONIe7 e810H MOlg sequew euojspues ewely JOQUOW 8}}87) oqung — euojspues oosoquer sOQWweW 931387 yoBqeippes JOQUEW 931387 o}d8q JOQUEW 231}87 BiB MeqWBD uo Bwi04 uojyBnoig dNOw¥d-ENs V1OfNOS dNOw¥S N3SAVHIVOHS ‘ulaiay posodoid ainje[suoUIOU puke sautoyos os1ydeis8ije1)s poysttqnd Ajsnotaasid juej10duly 7 “a2 QUOISHIS UBIPUBMEIPUB A @UO}SpUBS BIMON euosyIg Aseg dnOusd-8ns ViI0rNoo uojyBwso4 Bjofuo5 QU0ISHIS UBIPUBMBIPUB MA @U0}SpUBS BIMON sequen euoyspues| BuoBpng sOQWOEW SUO}SpUuBS ysBdg AeysoM JOQUOW 93187 810H MOG JOQuoW euojspuBs Bwely JOQUeW 931387 s0QuoW euoj\spues ooseque JOQUEW 231)87 WOIBQSIPPBS soQquew 231387 BiuBMEQqUBD JOQUOW 0})87 ojdeqg JOQUeW eUuoJspuBS UO}YBNOIg euoyspues BuobBpng dNOYD N3AVH 1IVOHS alee uojjewsio4 Asseg 01 ysbd AgyseM pejye}juese4)/pUuN 8Uu0}8IIS UB/pUuBMBIPUB A seuojspues pue seuoj}spnw snoeoeyjny BNpBIN s0}B8J8WO/BUu0D BlofuoD euoj}spues BJMON dNOuS N3SAVHIVOHS 8315 BIMON e1Bys Asseg SEIBYS [BP}OUlID JOQUOW 4)NL ybdg AgseM 64387 810H MOG syn] Bwely dNOYSD N3SAVHIVOHS JOQUeW JN e6uey o1y oye] @10H Mog SHIN Yue Kosem MOl[4 8j/OH MOIG soquow =m sun] BweIy Syjn] Bwey HNL uoyyBnoig 0487 oquing jjNy Oosequer e487 Moeqe|ppes-oydeg eyAyoRsy, BIIBMOQWBD SOINVOIOA DNOONIEESD #NL uojyBnoig Moly oquing sOQue~Z WN, oosequer sjjn_ Oosequer oye) yOBQeIPpPeS oye] BluBMEQqUeD JOQUOW 03/387 spuejs} eAl4 sJOQueW 07387 poomieple9 JOQWOW 071/387 BuunWNUI;W sJEQUEW 9387 Ao}410g ETE LAY AGNLS LN3AS3ud uolNjJBwso4 sjuBsBeyd NOILVWYOS dNOYS-ENsS GNV1E3SsNND S3YNSVSW 1VOO VEYEVMVTI dnowd-sns JOQUeW 23387 Buunweuuip sequew 631387) Aojexs0g sejoB} sojUBDIOA Bucbuj11eH uolnBwso4 sjuBsBeyd NOILVNYOS (¥Z6L) NVWMOG dNOWS-ENs GNVTdAasWNNd S3uUNSVSW 1VOO VeeVMV TTI Ya8sW3n 6487 Ao|y19g | Oye] BuUINWeULIW NOLLVWYOS (9961) 3SO¥ S3YNSVWAW 1VOO VYEVMVTII dNOus-sns) dnouwd (ZS6L) oye) Ao|y19g SOINVOIOA SNOONINYSD MOl4 SpuBls] BA4 MOl4 y9BQeIppes-01deq AOj4 BUBMeQWUBD mod Aopeg — == a] oye) BaunWNUW NOILVNYOS "ye 38 NITdGOF S3YNSVAW 1VOO VEYVMV I dnoud mols BUNUWNUT NOILVAYOS S3IW3S ANIYWN Y3ddN SSYNSVAW 1VOO YaddN (SL6L) YAdYVH Proc. LINN. Soc. N.S.W. 106 (4), (1982) 1983 PAUL F.CARR 291 The unit contains pillows and breccias consisting of latite in a sedimentary matrix showing soft-sediment deformation structures. Both pillows and breccias have been described in detail by Raam (1964, 1969) who considered that these features indicate that the Blow Hole Latite Member was contemporaneous with the associated sedimentary rocks. Since the sandstone units both above and below the Blow Hole Latite Member contain sedimentary structures and an abundant fauna indicative of a near-shore marine environment (Raam, 1968), the Blow Hole Latite Member is in- terpreted as a near-shore, submarine flow on to wet, unconsolidated sediments. Locally the flow intruded into the sediments to show restricted intrusive contacts. Kiama Sandstone Member. This unit conformably overlies the Blow Hole Latite Member and was defined as the Kiama Tuff by Jaquet et al. (1905). Joplin et al. (1952) renamed the unit the Kiama Tuff Member and Raam (1968, 1969) referred to this arenite as the Kiama Sandstone. A maximum thickness of 53 m is developed at Jamberoo (Fig. 1). The faunal assemblages indicate deposition in a shallow marine environment (Raam, 1968). Bumbo Latite Member. The Bumbo Flow which overlies the Kiama Sandstone Member was originally named by Jaquet et a/. (1905) and was renamed the Bumbo Latite by Joplin e al. (1952). This unit is one of the most extensive and most voluminous Per- mian flows of the southern Sydney Basin and outcrops over a total area of ap- proximately 240 km?. A maximum thickness of 150 m is developed at Saddleback Mountain (Fig. 1). Bowman (1974) has reported that the unit consists of three flows in some areas. The base of the Bumbo Latite Member is well exposed at Bombo Point where the contact is either sharp or is marked by a basal breccia up to 3 m thick. Sub-vertical to vertical pipe-like masses of breccia, 1 to 10 m in diameter, occur at many localities. Contacts between the breccia pipes and the surrounding latite are normally sharp, with a decrease in the size and abundance of vesicles in the latite away from the core of the breccia zone. Both the basal breccia and pipe-like masses of breccia consist of clasts of latite in a matrix of sedimentary material and cement of zeolite and hematite. The sedimentary material is indistinguishable from the underlying Kiama Sandstone Member and in places shows contorted bedding which has been interpreted by Raam (1964) as due to soft-sediment deformation. The breccias appear to have formed as a result of the interaction of magma with the underlying water-saturated sediment. Steam produced at this contact fragmented the base of the lava flow and, in some areas, dragged up wet sediment into the overlying volcanic rock (Raam, 1964). Jamberoo Sandstone Member. The Jamberoo Sandstone Member conformably overlies the Bumbo Latite Member and is conformably overlain by the Saddleback Latite Member. Jaquet ef a/. (1905) defined the unit as the Jamberoo Tuffs but Joplin et al. (1952) changed the name to the Jamberoo Tuff Member. Subsequently Raam (1968) renamed the unit the Jamberoo Sandstone. A maximum thickness of 155 m is developed at Jamberoo (Jaquet et al., 1905). Saddleback Latite Member. Nomenclature of the Saddleback Latite Member has been varied and problematical. The extrusion was defined as the Saddleback Flow by Jaquet et al. (1905) and renamed the Dapto-Saddleback Flow by Harper (1915) who con- sidered that the two unconnected outcrops in the vicinity of Port Kembla in the north and Saddleback Mountain in the south resulted from the same flow. Joplin e¢ al. (1952) accepted Harper’s opinion and redefined the two outcrops as one unit — the Sad- dleback Latite. Bowman (1970, 1974) considered that the Port Kembla and Saddleback Mountain outcrops resulted from two flows at different stratigraphic levels. Thus Bowman (1970, 1974) recognized the Saddleback Latite Member in the south, and the Proc. LINN. Soc. N.S.W. 106 (4), (1982) 1983 292 STRATIGRAPHY OF THE SOUTHERN SYDNEY BASIN Dapto Latite Member in the north and he considered the latter unit to overlie the Cambewarra Latite Member. In the present investigation the Saddleback and Dapto Latite Members are considered to be separate flows which occupy similar stratigraphic positions below the Cambewarra Latite Member. Stratigraphic relations between these units are discussed in a later section of this paper. Outcrops of the Saddleback Latite Member extend around the Illawarra escarp- ment from the southern side of Saddleback Mountain to Woodhill and the eastern limit of Kangaroo Valley. The unit does not occur on the northern slopes of Saddleback Mountain but crops out further to the north along the scarp on the southern flanks of Stockyard Mountain. The maximum thickness of 35 m is developed on the southern side of Saddleback Mountain. Harper (1915) interpreted the Saddleback Latite Member as a submarine ex- trusion. This view was supported by Lowder (1964) who found probable pillow structures at several localities. Also, the presence of vesicles near the top of the unit is suggestive of emplacement at, or near, the earth’s surface. Thus the Saddleback Latite is interpreted as being a submarine lava flow. Dapto Latite Member. All outcrops of the Dapto Latite Member occur in the vicinity of Port Kembla within an area of approximately 100 km?. Flagstaff Hill (Fig. 1) and its vicinity provide the most extensive and thickest (85 m) outcrop and the unit thins to the north, west and southwest away from this area. Various features present in the Dapto Latite Member and surrounding sedimentary rocks have been interpreted as indicating that the unit is a flat-topped laccolith intruded into unconsolidated Permian marine sedimentary rocks (Wilshire and Hobbs, 1962). However, since the time of publication of their article, drilling and extensive road-making activities by the New South Wales Department of Main Roads have provided access to much more information regarding the mode of emplacement of the Dapto Latite Member. As the drill-core is not available (lost or destroyed), the borehole logs of Cook (1966) were used to provide sub-surface data on the Dapto Latite Member. Integration of the data from outcrops and the boreholes provides the following interpretation: (1) The top of the Dapto Latite Member is uneven and is overlain by approximately 3 m of sandstone containing abundant feldspar crystals and fragments of the underlying latite. The fragments of igneous rock are usually rounded, are concentrated into particular horizons, and are isolated from the main mass of igneous rock. The clasts of the Dapto Latite Member in the sedimentary rocks are undoubtedly detrital in origin. (11) Although fine-scale individual horizons are difficult to follow in outcrop, near horizontal bedding (dip 3° to north-northwest) is apparent in the sedimentary rocks above the Dapto Latite Member. Apart from localized draping of bedding over the detrital latite clasts there is no evidence of bedding distortion that would be expected to accompany the intrusion of a large mass of igneous rock into unconsolidated sediments. (i) The Dapto Latite Member consists of two flows which, over much of the region, are separated by a breccia. The borehole logs of Cook (1966) describe the breccia as a complex succession of sandstone, siltstone and blocks of latite ranging from 5 cm to 3 m in vertical extent. The sedimentary rocks contain clasts of feldspar and latite and, apart from steeply inclined contacts with the latite blocks, are horizontally bedded with some cross-bedding. Cambewarra Latite Member. The Cambewarra Latite Member is the uppermost unit of the marine Shoalhaven Group and is the most extensive of the Permian extrusions in Proc. LINN. Soc. N.S.W. 106 (4), (1982) 1983 PAUL F. CARR 29 Le) the southern Sydney Basin. A maximum thickness of 70 m occurs at Woodhill. The flow has a pronounced physiographic expression and forms steep slopes or cliffs ex- tending from Browns Mountain to the northern side of Stockyard Mountain (Fig. 1). Between Stockyard Mountain and Wollongong, the Cambewarra Latite Member is represented by a boulder horizon (Harper, 1915). Numerous amygdules and vesicles near the top of the Cambewarra Latite Member indicate that crystallization occurred at or near the earth’s surface. A subaqueous, extrusive mode of emplacement is indicated by the presence of pillow structures with an inter-pillow matrix of shale and sandstone (Lowder, 1964). Marine fossils both below and within the equivalent unit suggest deposition in a marine en- vironment (Harper, 1915, p. 223). A boulder horizon containing numerous vesicular and amygdaloidal clasts of the Cambewarra Latite Member in a volcarenite matrix extends beyond the northern limit of the flow to Wollongong. Harper (1915) considered that the boulder horizon originated by contemporaneous erosion and deposition of clasts of Cambewarra Latite Member by a strong marine current flowing towards the north. The direction of the current was deduced from the absence of the boulder horizon south of Saddleback Mountain and the decrease in size of the clasts in the northerly direction whereas marine deposition is indicated by the presence of marine fossils in the horizon. A southerly source is consistent with recent work on the sedimentology of the upper part of the Shoalhaven Group and lower part of the Illawarra Coal Measures (Bowman, 1974; B. G. Jones, pers. comm. 1980). Harper (1915) traced the boulder horizon as far north as the Wollongong Water Supply Service Reservoir which was situated 5 km north-northeast of Flagstaff Hill (Fig. 1). Excavation of a 30 m deep cutting through the western flank of Flagstaff Hill by the New South Wales Department of Main Roads has exposed a conglomerate 3 to 5 m stratigraphically above the Dapto Latite Member. The conglomerate, which contains rounded clasts of amygdaloidal igneous rock in an arenite matrix, is considered by the author to represent the boulder horizon in this region. PHEASANTS NEST FORMATION The Broughton Formation is conformably overlain by non-marine deposits of the Pheasants Nest Formation at the base of the Cumberland Sub-group of the Illawarra Coal Measures (Fig. 2). Where the Cambewarra Latite Member or the equivalent boulder horizon is lacking the contact between the Shoalhaven Group and Illawarra Coal Measures is transitional (Wilson, 1969, p. 372). Sedimentary strata of the Pheasants Nest Formation consist of interbedded sandstone, siltstone and shale which lack marine fossils but contain fossil tree stumps, logs and plants. Only the four latite members interbedded with these sedimentary strata are described here. Five Islands Latite Member. The Five Islands Flow was named by Harper (1915) but for consistency of nomenclature is renamed herein the Five Islands Latite Member. The unit is very restricted in outcrop and occurs only on Flinders Islet which is the nor- thernmost of the Five Islands (Fig. 1). Pillow structures up to 1 m in diameter indicate a subaqueous, extrusive mode of emplacement. As the Five Islands Latite Member is the only unit which crops out on Flinders Islet, the stratigraphic position of the flow is problematical. The sedimentary strata of the Broughton Formation on the mainland nearest to Flinders Islet are almost horizontal with a slight dip (2° to 3°) towards the northwest. Evidence for major faulting is lacking. Also, the islands south and southeast of Flinders Islet are composed of Dapto Latite Member. Assuming no major faulting occurs between Flinders Islet Proc. LINN. Soc. N.S.W. 106 (4), (1982) 1983 294 STRATIGRAPHY OF THE SOUTHERN SYDNEY BASIN and the mainland or the other islands, the Five Islands Latite Member must be stratigraphically above the Dapto Latite Member. A lower limit on the stratigraphic separation between the Dapto Latite Member and Five Islands Latite Member cannot be inferred due to the lack of outcrops in proximity to Flinders Islet. However, an upper limit on the stratigraphic separation may be inferred from the regional dip of 3° to the northwest and the minimum distance between outcrops of the Five Islands and Dapto Latite Members. The nearest outcrop of the Dapto Latite Member is Bass Islet which is 1700 m southeast of Flinders Islet (i.e. down dip from the Five Islands Latite Member) and implies a maximum stratigraphic separation of 90 m. Thus the Five Islands Latite Member is taken to occur in the lower part of the Illawarra Coal Measures. The Calderwood, Minumurra and Berkley Latite Members occupy similar stratigraphic positions to that of the Five Islands Latite Member but the exact relationship between the former three units and the Five Islands Latite Member is unknown. On the basis of similarity of petrography to some samples from Kiama and Jamberoo, Chalmers (1941) mapped Flinders Islet as a remnant of the Bumbo Latite Member. However, the Five Islands Latite Member is geochemically distinct from the other latites (Carr, unpublished data) and on the basis of the inferred stratigraphic relationships outlined previously, Flinders Islet is considered to be composed of the Five Islands Latite Member and not the Bumbo Latite Member. Calderwood Latite Member. The Calderwood Latite Member has not been recognized previously. The name is here proposed for the unit which crops out around Calder- wood (Fig. 1) and is 3 m stratigraphically above the Cambewarra Latite Member (i.e. within, and near the base of, the Illawarra Coal Measures). The Calderwood Latite Member is restricted in outcrop area and extends discontinuously from the north- eastern side of Stockyard Mountain towards the north-northwest for a distance of 9 km. A maximum thickness of 38 m occurs near Calderwood (GR 908728 Robertson 1:25,000 Topographic Sheet 9028-IV-N First Edition) but the unit thins rapidly towards the north, south and west. A probably extrusive mode of emplacement is indicated by the high proportion of volcaniclastic material in the sedimentary rocks both above and below the unit. The volcaniclastic material above and below the Calderwood Latite Member is possibly equivalent to Bowman’s (1974) Tappitallee Mountain Tuff Member. The outcrop on the northeastern side of Stockyard Mountain (GR 958685 Albion Park 1:25,000 Topographic Sheet 9028-I-N First Edition) corresponds to the outcrop which Jaquet et al. (1905) and Lowder (1964) mapped as Saddleback Latite Member occurring at the top of the Cambewarra Latite Member. These authors were aware of the anomalous stratigraphic position but considered that the Saddleback Latite Member at this locality formed a topographic high at the time of eruption of the Cambewarra Latite Member. The present investigation, however, indicates that the outcrop on the northeastern side of Stockyard Mountain is separated from the un- derlying Cambewarra Latite Member by 3 m of sandstone and shale of the Illawarra Coal Measures. Bowman (1970, 1974) did not record either the Saddleback Latite Member un- derlying the Cambewarra Latite Member between Saddleback and Stockyard Mountain, or the Calderwood Latite Member on the northeastern side of Stockyard Mountain. However, the other outcrops of the latter unit were mapped as Dapto Latite Member (Bowman, 1970, 1974). These outcrops occur above the Cambewarra Latite Member and thus he concluded that the Dapto Latite Member is stratigraphically Proc. LINN. Soc. N.S.W. 106 (4), (1982) 1983 PAUL F. CARR 295 above the Cambewarra Latite Member. In spite of Harper’s (1915) recognition of the Cambewarra boulder bed as far north as Wollongong, Bowman did not mention the possibility of another flow which was stratigraphically above the Cambewarra Latite Member, and which cropped out in the Calderwood region. Minumurra Latite Member. Jaquet et al. (1905) defined the Minumurra Flow which was subsequently renamed the Minumurra Latite (Joplin e¢ a/., 1952). The incorrect name Minnamurra Latite has also been used extensively (e.g. Lowder, 1964; Raam 1969; Bowman, 1970, 1974). The unit occurs approximately 40 m above the base of the Illawarra Coal Measures and extends as a discontinuous sheet from the northern side of Saddleback Mountain to a point 15 km west of Port Kembla. A maximum thickness of 35 m occurs at Minnamurra River. The base of the Minumurra Latite Member is even but the top of the unit un- dulates by up to 0.5 m in a horizontal distance of 3 m. Horizontally-bedded sand- stones, shales and siltstones fill the depressions in the upper surface and individual sedimentary units are thicker in the depressions than ove: the accompanying ridges (Lowder, 1964) (GR 904647 Kangaroo Valley 1:25,000 Topographic Sheet 9028-IV-S First Edition). An extrusive mode of emplacement is supported by the occurrence in joints near the top of the unit of sedimentary material lithologically different from the sedimentary rocks overlying the Minumurra Latite Member (Lowder, 1964). No evidence in support of an intrusive mode of emplacement has been found previously or in the present study. Since the Minumurra Latite Member occurs within the Illawarra Coal Measures, the unit is considered to be an extrusion which erupted on to the Permian land surface or into a shallow subaqueous environment. Berkley Latite Member. Harper (1915) defined the Berkley Flow which was renamed the Berkeley Latite (McElroy, 1952), and the Berkley Latite (Joplin e a/., 1952). The incorrect spelling Berkeley has been used in many publications (e.g. McElroy, 1952; Raam, 1969; Bowman, 1970, 1974). The base of the unit ranges between 15 and 40 m above the top of the Shoalhaven Group and thus occurs at approximately the same stratigraphic level as the Minumurra Latite Member. This variation in stratigraphic position possibly reflects the palaeotopography. The Berkley Latite Member has been considered to be intrusive (McElroy, 1952) and extrusive (Harper, 1915; Rose, 1966; Raam, 1969; Bowman, 1970, 1974). The latter conclusion is supported by the present study for the following reasons: (i) clasts of Berkley Latite Member occur in the sandstone immediately above the unit at Mount Kembla (GR 005876 Wollongong 1:25,000 Topographic Sheet 9029-II-S First Edition); (ii) apart from localized draping of the bedding over the latite clasts, the sandstone above the Berkley Latite Member shows no evidence of distortion of bedding as would be expected to accompany intrusion of a large mass of igneous rock into unconsolidated sediments (GR 012869 Wollongong 1:25,000 Topographic Sheet 9029-II-S First Edition); and (iii) in the Figtree region, the Berkley Latite Member consists of two flows separated by approximately 3 m of sedimentary rocks of the Illawarra Coal Measures (GR 026875 Wollongong 1:25,000 Topographic Sheet 9029-II-S First Edition). The features described and cited as evidence for emplacement as a sill (McElroy, 1952) possibly represent localized intrusive contacts produced by restricted intrusion of the flow into unconsolidated sediments. The flow has a maximum thickness of ap- proximately 35 m in the Figtree region and thins to the north and southwest. Proc. LINN. Soc. N.S.W. 106 (4), (1982) 1983 296 STRATIGRAPHY OF THE SOUTHERN SYDNEY BASIN STRATIGRAPHIC RELATIONSHIPS BETWEEN THE DAPTO AND SADDLEBACK LATITE MEMBERS Harper (1915) considered the Dapto Latite Member to be a northern extension of the Saddleback Latite Member. Since there is no evidence of continuity between the outcrops at Saddleback Mountain and Port Kembla, and since the thickness of igneous rock decreases away from a maximum for each area, the units are considered to have been separate flows. The top of the Cambewarra Latite Member or the equivalent boulder horizon marks the top of the marine Shoalhaven Group (Harper, 1915). The contact between the Saddleback Latite Member and the overlying Cambewarra Latite Member is poorly exposed but no evidence has been found for the existence of sedimentary material between the two latites. By comparison, the Dapto Latite Member is separated from the Cambewarra boulder horizon by at least 3 m of sedimentary strata. Thus both the Dapto and Saddleback Latite Members occur within the Shoalhaven Group and occupy similar stratigraphic positions below the Cambewarra Latite Member or the equivalent boulder horizon. Assuming that deposition of the boulder horizon is essentially synchronous with eruption of the Cambewarra Latite Member, then either (i) the Saddleback and Dapto Latite Members are of the same age; or (il) the Dapto Latite Member is slightly older than the Saddleback Latite Member. The implications of these equally plausible alternatives are discussed below. Contemporaneous eruption of the Saddleback and Dapto Latite Members implies that subsequent to the extrusion of these flows and prior to the emplacement of the Cambewarra Latite Member and boulder horizon, sediment accumulated in the Port Kembla region whereas no deposition occurred in the vicinity of Saddleback Moun- tain. Alternatively, if the Dapto Latite Member is slightly older than the Saddleback Latite Member, sediment accumulated on top of the Dapto Latite Member prior to, and contemporaneously with, eruption of the Saddleback Latite Member. Subsequent to the eruption of the latter flow, the Cambewarra Latite Member and boulder bed were emplaced. SUMMARY AND CONCLUSIONS Strata of the upper Shoalhaven Group and lower Illawarra Coal Measures of the southern Sydney Basin have had a varied nomenclature. Reappraisal of previous work and recent mapping by the author have led to the recognition of nine distinct extrusive igneous rock-units in the Wollongong-Gerringong region. These flows provide a convenient basis for the subdivision of the strata of the upper Shoalhaven Group and lower Illawarra Coal Measures. The uppermost formation of the Shoalhaven Group is termed the Broughton Formation which contains five Latite Members — Blow Hole (lowermost unit), Bumbo, Saddleback, Dapto and Cambewarra (uppermost unit). The sedimentary strata between the Blow Hole and Bumbo Latite Members, and between the Bumbo and Saddleback Latite Members are known as the Kiama Sandstone Member and Jamberoo Sandstone Member respectively. Four Latite Members (Five Islands, Calderwood, Minumurra and Berkley) are recognized in the Pheasants Nest Formation of the Cumberland Sub-group of the Illawarra Coal Measures. ACKNOWLEDGEMENTS I am grateful to A. C. Cook, B. G. Jones and A. J. Wright for their helpful advice and constructive criticism of the manuscript. Proc. LINN. Soc. N.S.W. 106 (4), (1982) 1983 PAUL F. CARR 297 References Bowman, H. N., 1970. — Palaeoenvironment and revised nomenclature of the upper Shoalhaven Group and Illawarra Coal Measures in the Wollongong-Kiama area, New South Wales. Rec. geol. Surv. N.S. W., 12: 163-182. —, 1974. — Geology of the Wollongong, Kiama and Robertson 1:50,000 sheets. Geol. Surv. N.S.W., Sydney. CARD, G. W., and JAQUET, J. B., 1903. — The geology of the Cambewarra Range, New South Wales, with especial reference to the volcanic rocks. Rec. geol. Surv. N.S.W., 7: 103-140. CarR, P. F., and FACER, R. A., 1980. — Radiometric ages of some igneous rocks from the southern and southwestern coalfields of New South Wales. Search, 11: 382-383. CHALMERS, R. O., 1941. — The petrology of the Five Islands, Port Kembla, New South Wales. Rec. Aust. Mus. , 21: 27-42. Cook, A. C., 1966. — Report on the geology of Flagstaff Hill, Parish of Wollongong, New South Wales. (Unpublished). HARPER, L. F., 1915. — Geology and mineral resources of the Southern Coalfield. Part I. The south coastal portion. Mem. geol. Surv. N.S. W., 7. JAQUET, J. B., CARD, G. W., and HarpEr, L. F., 1905. — The geology of the Kiama-Jamberoo district. Rec. geol. Surv. N.S. W., 8: 1-66. Jopuin, G. A., HANLON, F. N., and Noakes, L. C., 1952. — Wollongong — 4 mile Geological Series. Explan. Notes, Bur. Miner. Resour. Geol. Geophys. Aust. LOWDER, G. G., 1964. — The geology of the Minnamurra Falls area. Sydney: University of Sydney, B.Sc. (Hons) thesis, unpubl. McCLunG, G., 1978. — Morphology, palaeoecology and biostratigraphy of Ingelarella (Brachiopoda: Spiriferida) in the Bowen and Sydney Basins of eastern Australia. Pubs. geol. Surv. Qd, 365: 11-87. McELRoy, C. T., 1952. — Evidence of the intrusive nature of the Berkeley Latite, Wollongong district, N.S.W. Aust. J. Scv., 15: 100. ——,, BRANAGAN, D. F., RAAM, A., and CAMPBELL, K. S. W., 1969. — Shoalhaven Group. Jn PACKHAM, G. H. (ed.), The geology of New South Wales. J. geol. Soc. Aust. , 16: 357-366. RaAAM, A., 1964. — Geology of the Minnamurra-Gerroa area. Sydney: University of Sydney, B.Sc. (Hons) thesis, unpubl. ——,, 1968. — Petrology and diagenesis of Broughton Sandstone (Permian), Kiama district, New South Wales. J. sedim. Petrol. , 38: 319-331. : ——, 1969. — Gerringong Volcanics. Jn PACKHAM, G. H. (ed.), The geology of New South Wales. /. geol. Soc. Aust. , 16: 366-368. ROSE, G., 1966. — Wollongong 1:250,000 Geological Series Sheets SI 56-9. Geol. Surv. N.S. W., Sydney. RUNNEGAR, B., and MCCLUNG, G., 1975. — A Permian time scale for Gondwanaland. Jn CAMPBELL, K. S. W. (ed.), Gondwana geology, 425-441. Canberra: Australian National University Press. WILSHIRE, H. G., and Hosss, B. E., 1962. — Structure, sedimentary inclusions, and hydrothermal alteration of a latite intrusion. J. Geol. , 70: 328-341. WILSON, R. G., 1969. — Illawarra Coal Measures. A. Southern Coalfield. Jn PACKHAM, G. H. (ed.), The geology of New South Wales. /. geol. Soc. Aust. , 16: 370-379. Proc. LINN. Soc. N.S.W. 106 (4), (1982) 1983 Fossil Eucalyptus Remains from the Middle Mauocene Chalk Mountain Formation, Warrumbungle Mountains, New South Wales W.B. K. HOLMES, F. M. HOLMES and H. A. MARTIN HOLMES, W. B. K., HOLMES, F. M., & MARTIN, H. A. Fossil Eucalyptus remains from the Middle Miocene Chalk Mountain Formation, Warrumbungle Mountains, New South Wales. Proc. Linn. Soc. N.S. W. 106 (4), (1982) 1983: 299-310. Eucalyptus fruits and leaves are preserved in the diatomite of the Middle Miocene Chalk Mountain Formation (new name), Warrumbungle Mountains, New South Wales. The fossil eucalypts show some features of advanced states in fruit and venation and are compared with some extant species. Based on present knowledge of Eucalyptus sensu lato, the fossils provide evidence of two separate phylogenetic lines being present in Middle Miocene time. The microflora recovered from the diatomite and associated lignite includes pollen grains of Myrtaceae (Myrtaceidites spp) of which a few are referable to Eucalyptus. W. B. K. Holmes, Hon. Research Fellow, Department of Geology, Unwwersity of New England, Armidale, Australia 2351; F. M. Holmes, ‘Noonee Nyrang’, Gulgong Road, Wellington, Australia 2820 and H. A. Martin, School of Botany, University of New South Wales, Ken- sington, Australia 2033; manuscript received 8 June 1982, accepted for publication 18 August 1982. INTRODUCTION Despite the present day prominence of Eucalyptus and the great diversity of its species, little is known of its evolutionary history from the fossil record (Martin, 1978; Briggs and Johnson, 1979). For phytogeographical reasons alone, there can be little doubt that the genus is of ancient Australian origin (Barlow, 1981); however, the ‘eucalypt-type’ pollen, Myrtaceidites eucalyptoides Cookson and Pike, which is probably bloodwood-Angophora first appears in the Oligocene (Martin, 1981). Fossil leaves which resemble eucalypts in gross form but which were inadequately described and figured, have been recorded from many poorly-documented Tertiary localities in New South Wales, Victoria, Tasmania, South Australia and Central Australia. References to these occurrences are given in Duigan (1951). Lange (1978) reported a diverse assemblage of Lepidospermae fruit, including a number of forms closely similar to Eucalyptus, preserved as impressions on silicified slabs of unknown age from the arid zone of South Australia. Ambrose et al. (1979) have also collected Eucalyptus fruit similarly preserved to those of Lange. The age of these fossils from Stuart Creek, northwest of Lake Torrens in South Australia is thought to be either Eocene-Oligocene or Miocene. Deane (1902) described several forms of eucalypt-like leaves from the Berwick Quarry in Victoria. A re-investigation of this flora by Nelson and Webb (in prep.) has shown that the cuticle of one of the leaves is similar to the cuticle of some extant Eucalyptus. Impressions of a eucalypt-like fruit and of rainforest leaves are also present but no Eucalyptus type pollen has been observed in the microflora which is dominated by Nothofagus pollen (Nelson and Webb, pers. comm.). Eucalyptus fruit and leaves and the microflora from the Middle Miocene diatomite deposit at Bugaldie near Coonabarabran in northwestern New South Wales are described below. Associated with the eucalypts are remains of leaves and flowers of plants (Holmes, in prep.) which have affinities with extant plants growing in situations with a much higher rainfall than the present Bugaldie average annual rainfall of 650 mm. Also in the diatomite are fossils of fish (Hills, 1946), a bird described as an owlet- Proc. LINN. Soc. N.S.W. 106 (4), (1982) 1983 300 MIDDLE MIOCENE FOSSIL EUCALYPTUS nightjar (Rich and McEvey, 1977), diatoms (Thomas and Gould, 1981 a and b), insects and freshwater bivalves. GEOLOGY AND AGE OF THE DIATOMITE Chalk Mountain, which lies 6 km south of Bugaldie village, is a remnant of an extensive dissected plateau of late Permian, Triassic and Jurassic freshwater sediments capped by basalt associated with the Warrumbungle Volcano Complex to the south. Lacustrine sediments which are interbedded between basalt flows on the Jurassic sediments and a 25 metre thick overlying basalt flow, outcrop around the escarpment of Chalk Mountain. Diatomite in these sediments was commercially exploited from 1919 until 1980. Griffin (1961) gave a detailed description of the geological and commercial aspects of the deposit but he did not name the formation. Reports by Herbert (1968a and b) deal with the diatomite. Chalk Mountain Formation (Holmes and Holmes, new name). The Chalk Mountain Formation is a lake deposit over 15 metres in thickness and is composed of almost horizontal layers of mudstone, clay, diatomite, three interbedded horizons of tuf- faceous clay and a thin band of lignite. The type locality is Quarry A on the western face of Chalk Mountain immediately below the basalt flow capping the mountain (grid reference 199148, Gilgandra 1: 250 000 Geological Series Sheet SH 55-16). The base of the formation is obscured at this locality by the basaltic talus slope and by quarry waste. On the roadway up the northern slope of the mountain the basal beds of mudstone and clay rest on basalt. The section exposed in Quarry A is shown in Table I, The area of lake sediments that have disappeared due to erosion around Chalk Mountain is unknown, but it is likely that the original lake was very much larger than the 38 hectares of the existing deposit. Plant macro-fossils have been collected only from the diatomite. Plant micro-fossils occur in both the diatomite and the thin band of lignite near the top of the formation. For a horizontal distance of about thirty metres in from the outcropping face, the diatomite has weathered to a pure white lightweight material much of which has been removed by open-cut mining and some tunnelling. Beyond the weathered zone the diatomite is of a grey to black colour due to the presence of organic matter and is known as ‘black earth’. The ‘black earth’ was only sparingly exploited due to the necessity to process it to remove the organic matter. In the weathered zone of the diatomite the plant macro-fossils are preserved as colourless or limonite-stained impressions and in the unweathered zone as colourless impressions TABLE 1 Type Section of Chalk Mountain Formation exposed in Quarry A on Chalk Mountain, Bugaldie (after Griffin, 1961) Basalt Clay with lignite band 0.91 m Tuff, clay and grit 4.05 m Clay and diatomite 0.13 m Tuff, clay, diatomite and gritty sandstone 3.04 m Diatomite high grade 0.08 m Clay and tuff Diatomite high grade Mudstone, tuff and clay Bottom Basalt, trachyte and andesite overlying Jurassic Pilliga Sandstone Proc. LINN. Soc. N.S.W. 106 (4), (1982) 1983 W.B.K. HOLMES, F. M. HOLMES AND H. A. MARTIN 301 or carbonaceous compressions. The diatomite exhibits seasonal varves each 0.33 mm to 0.5 mm thick (Thomas and Gould, 1981b). A deposition time of between 14 000 and 21 000 years would have been needed to accumulate the total thickness of 7 metres of diatomite, assuming the varves to be annual. The bands of tuff within the deposit indicate contemporaneous eruptive volcanic activity during the period of lacustrine sedimentation. K/Ar dating of igneous rocks from localities around the Warrumbungle complex (Dulhunty and McDougall, 1966; McDougall and Wilkinson, 1967; Wellman and McDougall, 1974) show that the volcanic activity took place during the period 17 million years to 14 million years before the present. The nearest sites to Chalk Mountain with K/Ar dates are Looking Glass Mountain (16.6 + 0.6 m.y.), 15 km to the west-northwest, and Woorut Mountain (15.5 + 0.4m.y.), 20 km to the south (Wellman and McDougall, 1974). As no study has been made of the geology of the northern portion of the Warrumbungle Volcano, the stratigraphic relationships of these localities to Chalk Mountain are not known. Therefore, with present information, we can only date the diatomite as being within the age limits of the active Warrumbungle Volcano — 17 million years to 14 million years, Middle Miocene. THE FOSSIL EUCALYPTS Family MYRTACEAE Genus Eucalyptus L’ Herit. Eucalyptus bugaldiensis Holmes and Holmes sp. nov. Fig. 1 A, Fig. 2 A-D Diagnosis: Fossil infructescence of a group of three umbellasters of seven or fewer small, ES? SS Z Fig. 1. A Eucalyptus bugaldiensis sp. nov. AMF61713 holotype x 2. B Eucalyptus sp. Leaf form B AMF61721 x 2.C Eucalyptus sp. Leaf form A AMF61724 x 2. Proc. LINN. Soc. N.S.W. 106 (4), (1982) 1983 302 MIDDLE MIOCENE FOSSIL EUCALYPTUS hemispherical, ribbed, woody fruit attached by tapering pedicels to peduncles. Rim of fruit broad; exerted valves broad and low. Description: Infructescence of three umbellasters on peduncles 5 mm - 10 mm in length, and 0.8 mm in width attached at a common point to a stem. Umbellasters composed of seven or fewer fruit attached to the peduncle apex by pedicels (anthopodia) 5 mm - 10 mm in length widening upwards to merge into the base of the fruit. Fruit hemispherical, 4 mm - 5 mm in diameter. External surface ornamented by 7 - 10 longitudinal ribs which occasionally fork near the rim. Rim flat ca 1 mm in width. Exerted valves forming a broad low triangular projection 0.5 mm above the rim. Type material: Holotype AMF61713 and counterpart AMF61714. Paratypes AMF61715-20. The Australian Museum, Sydney, N.S.W., Australia. Type locality: Diatomite beds in Quarry A on Chalk Mountain, 6 km south of Bugaldie village near Coonabarabran, New South Wales. Grid reference 199148, Gilgandra 1: 250 000 Geological Series Sheet SH 55-16. Chalk Mountain Formation, Middle Miocene. Discussion: The fossil fruits of Eucalyptus bugaldiensis are preserved as impressions in the soft diatomite. The hollow moulds contain no organic material. The surface of the diatomite was hardened with Bedacryl 122X and latex casts were then taken of the holotype and its counterpart. When whitened with ammonium chloride, the latex casts revealed some of the finer details not readily seen in the original specimens. The distal ends of the peduncles show scars where additional buds or fruits have been aborted or lost during development or preservation. The umbellaster on AMF61715 has five fruit attached by pedicels to the peduncle. In addition to the still-attached fruit, the distal end of the peduncle shows the abscission scars of another two pedicels. ‘This suggests that the total flower number per umbellaster was seven, a number which may be ex- pected in dichasial inflorescences in a family with primarily opposite and decussate phyllotaxy. Extant species in subgenus Symphyomyrtus very often have seven flowered umbellasters. The arrangement on the holotype of three umbellasters about a common point, is found in the eucalypt conflorescence sub-types of T,; and 83 of Johnson (1972, fig. 3). The T, sub-type characterizes the small tropical subgenus Telocalyptus (section Equatoria) and the S3 sub-type occurs mainly in the subgenus Eudesmia, but the fossil fruits are quite different from any in Eudesmia. However, three umbellasters can be so attached in some variants of sub-type S, also (L. A. S. Johnson, pers. comm.) when the phyllotaxy in the inflorescence region is opposite rather than disjunct-opposite (Briggs and Johnson, 1979), as for instance in some inflorescences of Eucalyptus microtheca of subgenus Symphyomyrtus section Adnataris. The infructescence and the individual fruits of E. bugaldiensis are very close in form to some extant species of Eucalyptus. However, due to the type of preservation and to the lack of supporting organs, this new species cannot at present be placed with certainty in any particular subgenus. The fruits of E. microtheca F. Muell. (Blake, 1953, pl. 30, figs 20-36) are closely similar in size and form to those of E. bugaldiensis. E. microtheca is very widely distributed throughout the continent (Blake, 1953, map 24) on ground that is subject to seasonal flooding. E. raveretiana F. Muell. from areas near the coast in central and northern Queensland has similar but rather smaller fruit. None of the diverse forms of fossil Eucalyptus fruits from South Australia (Lange, 1978; Ambrose et al., 1979) appear to be similar to E. bugaldiensis but their different aspect of preservation makes comparisons difficult. The cluster of fruits from Berwick, Victoria (Nelson and Webb, in prep.) is superficially similar to £. bugaldiensis but lacks evidence of valves. Proc. LINN. Soc. N.S.W. 106 (4), (1982) 1983 W. B. K. HOLMES. F. M. HOLMES AND H. A. MARTIN 303 S aN Fig. 2. A-D Eucalyptus bugaldiensis Holmes & Holmes sp. nov. Holotype. A, B AMF61713,A x 1;B x 4.6, D AMF61714 counterpart of AMF61713,C x 1;D x 4. Proc. LINN. Soc. N.S.W. 106 (4), (1982) 1983 MIDDLE MIOCENE FOSSIL EUCAL YPTUS C, D Eucalyptus sp. Leaf form B N > 4 —Q x < st N ™~ Xo) fy = < < one FOnou ee, os so =e 6 Sree 22 ac SA ST R44 Ry FA + = 2) an oy By we Ee Proc. LINN. Soc. N.S.W. 106 (4), (1982) 1983 W.B.K. HOLMES, F.M. HOLMES AND H. A. MARTIN 305 Eucalyptus sp. Leaf-form A Fig. 1 C, Fig. 3 A-B Description: Leaf lanceolate to narrow lanceolate, falcate, asymmetrical about lamina base; margin scarious, entire, sometimes insect-damaged. Length 8 cm-14 cm; width 1.5 cm-2 cm, widest at one quarter of length from the base and tapering gradually to an obtuse apex. Petiole curved, to 2 cm in length. Midvein prominent, tapering from the petiole to the apex. Secondary veins parallel, ca 1 mm apart, sometimes with a minor vein in between; attached to the midvein at 50°-60° and running straight to the intramarginal vein which runs close and parallel to the leaf margin. Tertiary veins form an irregular reticulum of about four rows of areoles between the secondary veins. Material: AMF61721-3. Discussion: The venation pattern of leaf-form A is of the Transversae class (Cambage, 1913; Maiden, 1922) (not to be equated with the section Transversaria of Pryor and Johnson, 1971) and is present in the subgenera Corymbia and Blakella of Eucalyptus and in Angophora which Pryor and Johnson (1971) have suggested is also of subgeneric rank in Eucalyptus s.1. Extant species with venation similar to leaf-form A are the widespread Angophora floribunda (Sm.) Sweet and Eucalyptus trachyphloia F. Muell., a bloodwood tree, which grows on the drier sandy soils of the Upper Hunter River and Pilliga areas of New South Wales and into Queensland. Eucalyptus sp. Leaf-form B Fig. 1 B, Fig. 3C-D Description: Leaf lanceolate, slightly falcate; margin entire; apex obtuse; lamina base asymmetrical. Width 16 mm-23 mm; length 8 cm-12 cm. Secondary veins sub- parallel, spaced irregularly, 5-6 per cm, running a slightly undulate course to the intramarginal vein at an angle of 55°-60° to the midvein. Secondary veins occasionally fork and anastomose to form a very elongate reticulum. Petiole curved, to 2 cm in length. Material: AMF61724-5. MMF15284. Discussion: MMF15284 (Fig. 3 C) shows portions of five leaves, one of which is at- tached to a slender stem which has sub-opposite projections from where the other leaves may have become detached. Leaf-form B differs from leaf-form A by the wider spaced and irregular course of the secondary veins. The venation pattern is in- termediate between the Transversae and Obliquae classes (Cambage 1913; Maiden 1922). This general pattern of leaf venation is present in extant species within the subgenus Symphyomyrtus section Adnataria series Oliganthinae or series Pruinosae and in the subgenus Telocalyptus series Degluptae. Leaves of Eucalyptus raveretiana F. Muell. (a member of Telocalyptus) bear a close resemblance to this fossil form. PALYNOLOGY OF THE CHALK MOUNTAIN FORMATION (H.A.M.) Table 2 lists the taxa found in the ‘black earth’ diatomite and in the lignite from near the top of the formation. Preservation is rather poor which limits identifications, and this may be the result of initial poor conditions at the time of deposition or of subsequent weathering on exposure. Nevertheless, the assemblages are workable and invaluable because they can be dated by independent evidence. The two assemblages are essentially the same, although there is a quantitative difference between the diatomite and the lignite. This difference could result from the different conditions of deposition. Diatomite is formed in lakes and so would ac- cumulate more wind-blown pollen. Because lignite forms in swamps most of the pollen Proc. LINN. Soc. N.S.W. 106 (4), (1982) 1983 306 MIDDLE MIOCENE FOSSIL EUCALYPTUS TABLE 2 The Pollen Frequencies Fossil Name ALGAE Botryococcus braunu Kitzing FERN SPORES Cyathea paleospora Martin Deltoidospora inconspicua Martin Laevigatosporites ovatus Wilson & Webster Polypoditdites sp. Reticuloidosporites minisports Martin Unknown spore GYMNOSPERMS Araucariacites australis Cookson Cupressaceae sp. indet Dacrydium florinu (Cookson & Pike) Cookson Phyllocladus palaeogenicus Cookson & Pike Podocarpus australiensis (Cookson & Pike) Martin Podocarpus elliptica (Cookson) Martin ANGIOSPERMS Cupanierdites orthoteichus Cookson & Pike Encipites sp. Graminidites media Cookson Haloragacidites harrisit (Couper) Harris Loranthaceae Malvacipollis diversis Harris Milfordia sp. Monosulcites cf Palmae Myrtaceidites eucalyptordes Cookson & Pike Myrtaceidites cf M. eucalyptordes Cookson & Pike M. parvus Cookson & Pike Myrtaceae not identified further Nothofagus aspera Cookson N. emarcida Cookson Polyporina chenopodiaceoides Martin Proteacidites sp. Quintinia psilatispora Martin Tnicolporites scabratus Harris Tricolporites cf Cunoniaceae Tricolporites cf Goodeniaceae Tricolporopollenites wvanhoensis Martin Tniontes minisculus McIntyre T. orbiculatus McIntyre Tnorites cf Ulmaceae Unknown tricolpate/tricolporates Number of kinds of unknowns Botanical Affinity Botryococcus brauni Cyathea spp. Total Fern Spores Agathis and Araucaria Cupressaceae Dacrydium Sec. B Phyllocladus spp. Podocarpus Sect. Dacrycarpus Podocarpus, most other sections Total Gymnosperms Tribe Cupaneae (Sapindaceae) Epacridaceae Gramineae Casuarina Loranthaceae Austrobuxus— Dissiliaria (Euphorbiaceae) Restionaceae Like Palmae Bloodwood—Angophora, possibly other genera Possibly other eucaly pts Possibly T71stanta—Backhousta—Baeckia and other genera Myrtaceae Total Myrtaceae Nothofagus, the menztesu type (e.g. N. moorez, N. cunninghamii) Nothofagus, the brassi type Total Nothofagus Chenopodiaceae and some Amaranthaceae Proteaceae Quintinia like Cunoniaceae like Goodeniaceae probably Rutaceae like Apananthe— Celtis Notes: (1) Subtotals may not add up exactly because individual percentages have been rounded off. (2) + indicates a presence but not counted. (3) The use of ‘like’ indicates a tentative identification. Proc. LINN. Soc. N.S.W. 106 (4), (1982) 1983 Diato- Lig- mite nite %o % + it 58) 29 i129) 2.4 1.5 5.6 0.5 0.2 paren Mae: 46 17.2 10.7. 18.1 0.8 DP 4.6 1.0 0.5 5.3 5.4 14) WO.) 29.0 34.8 + 1.2 0.2 13.0 24.0 0.2 a + 0.7 0.5 0.8 6.8 0.2 S75 5.2 25.2 3.2 0.8 0.5 APIO oe 3 0.5 0.8 + 2.0 3.0 0.7 0.2 0.2 0.2 1.0 0.2 22.1 13.2 No. No. 7 6 W. B. K. HOLMES, F.M. HOLMES AND H. A. MARTIN 307 would come from the plants growing in and around the edges of the swamps. Fern spores are more abundant and diverse in the lignite. The swamp environment would be more suitable for fern growth. Gymnosperms are well represented in both assemblages but are slightly more abundant in the lignites and of this group, Araucariaceae are the most common. Casuarina is almost twice as abundant in the lignites. Some species of Casuarina grow on the swamp edges today. The Myrtaceae are more abundant in the diatomite than in the lignite. Un- fortunately, the identification of pollen to genera within this family is not particularly reliable. The bloodwood-Angophora type is only found in the lignite in a very low concentration. Judging from surface samples, Rose (1981) found that this pollen type does not disperse aerially far from the source of production. However, water transport cannot be ruled out; but swamp vegetation would act as a filter and prevent the pollen from travelling far (Birks and Birks, 1980). The other ‘possibly eucalypt’ type is smaller than the bloodwood-Angophora type, and Rose found that it is more widely distributed. This second type has been found only in the diatomite. Myrtaceous pollen which is not like that of eucalypts is also present (see Table 2). The majority of the pollen cannot be identified beyond the family level because of poor preservation or crumpling which obscures diagnostic features. Nothofagus is present but in very low concentration, indicating transport from a distant source. The small amount of the brassi type may have come from plants growing a long way from the catchment of this lake. Little can be said of the taxa present in low frequencies, for these are usually low pollen producers. The in- determinate tricolpate/tricolporate group is quite abundant. These pollen assemblages must represent forests, judging from the low content of herbaceous elements such as Gramineae and the absence of Compositae. The high content of Myrtaceae in the diatomite together with the low content of Nothofagus in both samples fits the Myrtaceae phase of the Pliocene (Martin, 1973). Malvacipollis diversus and Triporopollenites ivanhoensis are usually found only in the lower part of the older phase (Martin, unpubl.). The Pliocene age was based only on the general geology of the Riverina region (Martin, 1973), so it may not be reliable. None of the diagnostic species of the mid-late Miocene Trporopollenites bellus Zone, described from the Gippsland Basin of south eastern Victoria (Stover and Partridge, 1973), has been found at Chalk Mountain and the general description of this Zone is a poor fit for the present assemblages. The base of this Zone contains abundant Nothofagus which decreases upwards through it. Its upper limit possibly extends into the Pliocene but the relationship to the overlying Myrtaceae phase (older) has not been documented. The problem is one of geographic variation during the Tertiary. In time-equivalent assemblages, Nothofagus is more abundant in southeastern Victoria than in New South Wales. Conversely, Myrtaceae are more abundant in New South Wales, particularly in the most westerly sites (Martin, 1983). Since the change from an assemblage with high Nothofagus to one with high Myrtaceae depends on the climate becoming drier, this change should have occurred earlier in the drier, inland regions than on the wetter coastal southeastern part of Australia. The Chalk Mountain microfloral assemblages show that the change had occurred in this region by the mid Miocene, some 14-17 million years ago. THE MIDDLE MIOCENE CHALK MOUNTAIN PALAEOENVIRONMENT The Tertiary pollen record is largely that of rainforest assemblages (Martin, 1978, p. 200) and most deposits of plant macro-fossils show a preponderance of mesophyllous leaves which constituted the ‘Cinnamomum’ flora (Sussmilch, 1937). The preservation Proc. LINN. Soc. N.S.W. 106 (4), (1982) 1983 308 MIDDLE MIOCENE FOSSIL EUCAL YPTUS of pollen and other plant remains as fossils usually depends on the presence of lakes, swamps or bogs. Therefore, a strong bias exists against the fossilization of plants adapted to growing in drier and harsher environments. Beadle (1981) has suggested that scleromorphic plants began to differentiate at the margins of rainforests and developed along declining soil-fertility gradients. Johnson and Briggs (1981) discussed the possibility of differentiation of scleromorphic taxa during the Early Tertiary in nutrient-deficient forest sites. The large tracts of low-fertility soils derived from the underlying Jurassic Pilliga Sandstone surrounding the Warrumbungles would have provided such an environment. At the present time eucalypts characteristically occur in open forest and woodland associations and only a few species have been successful in rainforest margins (Barlow, 1981). In addition to the eucalypt remains, the Chalk Mountain fossil flora includes fern fragments, podocarp leaves, Ceratopetalum sp. flowers and myrtaceous, lauraceous and other leaves that show similarities in gross form with present-day rainforest species in high rainfall areas of northeastern New South Wales and eastern Queensland. This mixed fossil flora should not be regarded as evidence that eucalypts in the Middle Miocene were adapted to rainforest conditions. Two present-day environmental situations provide a similar mixture of leaves. In moist sclerophyll forests on the North Coast of New South Wales, rainforest will develop, in the absence of fire, beneath tall Eucalyptus trees such as E. grandis W. Hill ex Maiden, E. saligna Sm. and occasionally E. microcorys F. Muell. Along the coastal strip of New South Wales and Queensland, rainforest species sometimes form a fringing forest along the banks of watercourses. Sclerophyll forest abuts and often overhangs this rainforest. Leaf-litter in the stream bed is a mixture of eucalypt and rainforest plant remains. At Chalk Mountain in Middle Miocene time, the rich soils derived from the basalt, the moist conditions adjacent to the lake, and the probable higher rainfall (Kemp, 1978) would have favoured rainforest growth. Sandstone hills around the lake and areas away from the soil-improving influence of the basalt would have supported a sclerophyll forest or woodland to provide the eucalypt, herb and grass remains that would have been carried by wind or water to mingle with the rainforest remains in the lake sediments. The high percentage of gymnosperm and Casuarina pollen probably indicates the proximity of these plants to the site of deposition. CONCLUSION On the basis of leaf venation, the Chalk Mountain fossils show that by the Middle Miocene the Eucalyptus genus s.1. had differentiated into at least two of the phylogenetic groups as proposed by Johnson (1972, fig. 1). The fossil eucalypts show some features of advanced states in fruit and venation (L. A. S. Johnson, pers. comm.). This suggests that a suitable environment for their development must have existed very much earlier in the Tertiary. The paucity of eucalypt remains in the fossil record probably reflects the lack of opportunity for scleromorphic plants to become fossilized. The lack of similarity between the Miocene pollen assemblages from the Gipp- sland Basin and Chalk Mountain is probably due to geographic and climatic factors rather than a great difference in age of deposition. K/Ar dating of the basalts above and below the Chalk Mountain Formation is desirable to determine the exact age of the locality. ACKNOWLEDGEMENTS The authors wish to thank Mr V. Mills and Mr and Mrs G. Hawker of Bugaldie for assistance and information on the Chalk Mountain area; Dr J. W. Pickett and Dr Proc. LINN. Soc. N.S.W. 106 (4), (1982) 1983 W.B.K. HOLMES, F.M. HOLMES AND H. A. MARTIN 309 A. Ritchie for the loan of specimens from the Mining and Geological Museum and the Australian Museum respectively; Mr R. S. Jones for preparing the latex casts; Mr P. J. Althofer for making available his private collection; Professor N. C. W. Beadle, Mr J.B. Williams, Dr A. G. Floyd, Dr L. A. S. Johnson and Mr D. F. Blaxell for helpful discussion on the relationship of the fossils to extant species; Professor J. F. G. Wilkinson for information on the Warrumbungle Volcano; Mr D. Nelson and Dr J. A. Webb for useful comments and for information on their unpublished investigations of the Berwick fossil eucalypts. W. B. K. Holmes gratefully acknowledges support from the Science and Industry Endowment Fund. References AMBROSE, G. J., CALLEN, R. A., FLINT, R. B., and LANGE, R. T., 1979. — Eucalyptus fruits in stratigraphic context in Australia. Nature 280: 387-389. BARLOW, B. A., 1981. — The Australian Flora: Its origin and evolution. In Flora of Australia, Vol. 1. Canberra: Australian Govt Printer. BEADLE, N. C. W., 1981. — Origins of the Australian angiosperm flora. In KEAST, A., Ecological Biogeography of Australia. The Hague: W. Junk. BirKS, H. J. B., and BirKs, H. H., 1980. — Quaternary Palaeoecology. London: Edward Arnold. BLAKE, S. T., 1953. — Botanical contributions of the Northern Australia regional survey. 1. Studies on Northern Australian species of Eucalyptus. Aust. J. Bot. 1 (2): 185-352. BRIGGS, B. G., and JOHNSON, L. A. S., 1979. — Evolution in the Myrtaceae — evidence from inflorescence structure. Proc. Linn. Soc. N.S. W. 102 (4): 157-256. CAMBAGE, R. H., 1913. — Development and distribution of the genus Eucalyptus. J. Proc. R. Soc. N.S. W. 47: 18-58. DEANE, H., 1902. — Preliminary report on the fossil flora of Pitfield, Mornington, Sentinel Rock (Otway Coast), Berwick and Wonwron. Rec. Geol. Surv. Vict. 1: 13-32. DuiGan, S. L., 1951. — A catalogue of the Australian Tertiary flora. Proc. R. Soc. Vict. 63: 41-56. DULHUNTY, J. A., and MCDOUGALL, I., 1966. — Potassium-argon dating of basalts in the Coonabarabran- Gunnedah district, New South Wales. Aust. J. Scz. 28: 393-394. GRIFFIN, R. J., 1961. — The Bugaldi (Chalk Mountain) diatomaceous earth deposit, N.S.W. Tech. Rept Depi Mines N.S. W. 7: 19-36. HERBERT, C., 1968a. — The mineral industry of New South Wales. 14. Diatomite. Sydney: Dept of Mines, N.S.W. , 1968b. — Diatomite deposits in the Warrumbungles. Rept. Geol. Surv. N.S. W., 1968/195 unpubl. HILs, E. S., 1946. — Fossil Murray Cod (Maccullochella macquartensis) from diatomaceous earths in New South Wales. Rec. Aust. Mus. 21 (7): 386-382. Jounson, L. A. S., 1972. — Evolution and classification in Eucalyptus. Proc. Linn. Soc. N.S.W. 97(1): 11-29. , and Briccs, B. G., 1981. — Three old southern families — Myrtaceae, Proteaceae and Restionaceae. In KEAST, A., Ecological Biogeography of Australia. The Hague: W. Junk. Kemp, E. M., 1978. — Tertiary climatic evolution and vegetation history in the south-east Indian Ocean region. Palaeogeogr. Palaeoclimatol. Palaeoecol. 24: 169-208. LANGE, R. T., 1978. — Carpological evidence for fossil Eucalyptus and other Leptospermeae (sub-family Leptospermoideae of Myrtaceae) from a Tertiary deposit in the South Australian arid zone. Aust. /. Bot. 26: 221-233. MAIDEN, J. H., 1922. — A critical revision of the genus Eucalyptus. Vol. 6. Sydney: Government Printer. MartTIN, H. A., 1973. — Upper Tertiary palynology in southern New South Wales. Geol. Soc. Aust. Spec. Publ. 4: 35-54. ——.,, 1978. — Evolution of the Australian flora and vegetation through the Tertiary: evidence from pollen. Alcheringa 2: 181-202. ——., 1981. — The Tertiary flora. In KEAST, A., Ecological Biogeography of Australia. The Hague: W. Junk. ——, 1983. — Changing Cenozoic barriers and the Australian paleobotanical record. Annals Missourt Bot. Gdns 69: 625-667. McDouGA tt, I., and WILKINSON, J. F. G., 1967. — Potassium-argon dates on some Cainozoic rocks from north-eastern New South Wales. /. geol. Soc. Aust. 14: 225-234. NELSON, D., and WEb3, J. A., in prep. — A Tertiary flora from Berwick, Victoria. Pryor, L. D., and JOHNSON, L. A. S., 1971. —A classification of the Eucalypts. Canberra: A.N.U. Press. Proc. LINN. Soc. N.S.W. 106 (4), (1982) 1983 310 MIDDLE MIOCENE FOSSIL EUCAL YPTUS RICH, P. V., and McEvey, A., 1977. — A new owlet nightjar from the Early to Mid-Miocene of Eastern New South Wales. Mem. Nat. Mus. Vict. 38: 247-253. ROSE, S., 1981. — Palynology and history of the Holocene at Dry Lake, Thirlmere, N.S.W. Kensington: Univ. N.S.W., B.Sc. (Hons) thesis, unpubl. SToveR, L. E., and PARTRIDGE, A. D., 1973. — Tertiary and late Cretaceous spores and pollen from the Gippsland Basin, southeastern Australia. Proc. R. Soc. Vict. 85: 237-286. SUSSMILCH, C. A., 1937. — The geological history of the Cainozoic Era in New South Wales. Proc. Linn. Soc. N.S. W. 62: vili-xxxiil. THomas, D. P., and GOULD, R. E., 1981 a. — Tertiary non-marine diatoms from eastern Australia: descriptions of taxa. Proc. Linn. Soc. N.S.W. 105 (1): 23-52. , 1981 b. — Tertiary non-marine diatoms from eastern Australia: palaeoecological in- terpretation and biostratigraphy. Proc. Linn. Soc. N.S.W. 105(1): 53-63. WELLMAN, P., and MCDOUGALL, I., 1974. — Potassium-argon ages on the Cainozoic volcanic rocks of New South Wales. J. geol. Soc. Aust. 21(3): 247-271. Proc. LINN. Soc. N.S.W. 106 (4), (1982) 1983 Long distance ‘Transport of Spores of Puccinia graminis tritici in the southern Hemisphere I. A. WATSON and C. N. A. DE SOUSA WaTSsON, I. A., and DE Sousa, C. N. A. Long distance transport of spores of Puccinia graminis tritici in the southern hemsiphere. Proc. Linn. Soc. N.S.W. 106 (4), (1982) 1983: 311-321. Evidence is presented that viable uredospores of the wheat stem rust fungus Puccinia graminis tritict reached Australia from Africa in 1968-69. Two strains 326- 1,2,3,5,6 and 194-1,2,3,5,6 both very dissimilar from any previously found in Australia were collected in 1969 from Clinton, S. Australia and Tichborne, N.S.W. respectively. These strains or derivatives from them were found to be indistinguishable from their African counterparts. The weather systems responsible for the transport of the spores across the Indian Ocean are discussed. I. A. Watson, 49 Old Castle Hill Road, Castle Hill, Australia 2154, and C. N. A. de Sousa, EMBRAPA, Passo Fundo RS 99100, Brazil (both formerly of the Department of Agricultural Botany, The University of Sydney, Sydney); manuscript received 21st April 1982, accepted for publication 21 July 1982. INTRODUCTION Rust pathogens of cereals are widely distributed throughout the world and survive on native or introduced host plants. Regardless of the geographical area, these pathogens exist as a multiplicity of strains differentiated by their pathogenicity, morphology, physiological and other characters. The origin of this variability can be attributed to a number of different causes. Both sexual and asexual processes may proceed side by side in some countries but where the alternate host is insignificant, mutation can explain most of the variation. It is generally recognized that where geographical areas are isolated the rust flora of those areas becomes characteristic of them. The extent of the differences between areas will depend on the amount of recombination occurring within them due probably to sexual reproduction in the fungus and on the survival of mutations. One of the most outstanding examples of this geographical isolation is seen in the linseed rust fungus Melampsora lini (Pers.) Lév. When comparisons are made between isolates from North America and Australia, Bison is found to be susceptible in the former but highly resistant to a group of strains in Australia. Strains avirulent on Bison in Australia commonly survive on the native flax Linum marginale Cunn. which has not yet been shown to have specific genes for resistance. It has already been pointed out (Watson, 1977) that on a global basis the total variation in Puccinia graminis f.sp. tritict Eriks. and Henn. that can be demonstrated in any one area is related not only to sexual reproduction and mutation but to the extent of wheat breeding activities especially where programs have made use of gene specific resistance. Reports appearing in the literature however do not always give the true picture of the range of variability, since the latter is revealed only when suitable single gene differentials are used in the survey work. In Puccinia graminis tritict maximum variability occurs in North America where the alternate host is common and a large effort is expended on breeding resistant cultivars using the well documented Sr genes. There is thus extensive variation around the genes of the standard set of differentials and also around the Sr genes of the parents used by the breeders. In Australia by contrast, there are few standard races, but a multiplicity of components have evolved within each following mutation and the cultivation of resistant wheats of diverse genotypes. Once these variants of the population have been Proc. LINN. Soc. N.S.W. 106 (4), (1982) 1983 312 LONG DISTANCE TRANSPORT OF SPORES catalogued and described any spectacular deviations from them can be easily recognized and will indicate that something other than normal evolution has occurred. ‘TRANSPORTATION OF SPORES (1) Over Land areas. Pathogen gene flow between regions will tend to have a unifying effect on their rust flora. The effectiveness of wind transport of spores in contributing to this uniformity will depend on the proximity of the regions and any impediments offered by natural barriers. It is suggested that eastern and western Europe may be different regions (Hogg et al. , 1969) but the rapid spread of race 77 of P. recondita Rob. ex Desm. following the release of wheats with the IB/IR translocation indicate that there is a ready exchange of spore material over most of this land area. The distances covered by any one spore movement are probably relatively short and seldom exceed 500-700 km (M. Boskovic, personal communication). It has been known for a long time that spores of P. graminis tritici are transported on the North American continent along the Puccinia pathway, but the extent of the distances covered in any one step is not clear from the reports. As in Europe it would appear to be of the order of several hundred km (Hogg et al. , 1969). On the Australian continent the movement of spores of P. graminis tritici over much longer distances and in different directions has been well documented over the last 25 years (Hogg et al., 1969). The results from earlier work were confirmed in 1973. In that year practically no stem rust was found in Western Australia but there was an epidemic in the eastern states. The composition of the flora in both the east and west was well known. Sometime after this epidemic 12 strains of eastern origin were isolated in Western Australia for the first time (Luig, 1977). Presumably the spores were caught in the anticyclones which may extend for 6,000 km and which, in the November-April period, feature easterly winds over the continent. Spores must have been transported about 3,500 km from east to west over the harsh inland because there are no congenial hosts between the areas which were the source of the spores and the points of new infections. (ii) Over Seas. Adjoining countries seldom present any barriers to the movement of rust spores but as they become increasingly separated by seas such as the Mediterranean, the Baltic and the Tasman, the possibility of spore transport from one to the other is reduced. From the summary presented by Hogg ¢t al. (1969) it is clear that movement of rusts across water is readily occurring in Europe and spores in transit have been trapped in the atmosphere. While there is no reason to doubt this movement of P. graminis tritici the designation of the strains involved has not been very precise as only the broad categories of standard race numbers were known. Hirst and Hurst (1967) believe that to establish proof of aerial dissemination of spores there needs to be ‘incontrovertible association between the source and new colonies’ or ‘interception of migrant spores’. Work in the Australia-New Zealand area has adhered rigidly to the first of these requirements and where spore movement of rusts is postulated, the extensive genetic markers in the material at the source and at the new site have been identical as far as is known. There are several reasons why, despite the extent of the land and water in this geographical area, the rust organisms can be so accurately characterized. First, the annual race survey has been in progress for 60 years and for much of that time the classification of the strains has been very precise because the genes involved in dif- ferentiation have included those of the standard series as well as those in the Australian supplementals. Consequently it has been possible to relate to some extent the local flora with that found overseas. Second, and very important, Australia is an isolated con- Proc. LINN. Soc. N.S.W. 106 (4), (1982) 1983 I. A. WATSON ANDC.N.A.DESOUSA 313 tinent, politically uniform so that collections from a vast area can be made and examined at will. Since wind currents transport spores around the country the national wheat crop in years of favourable climatic conditions must serve as a huge biological trap to snare not only local inoculum but an occasional viable spore that may reach the continent from elsewhere. Because of this isolation it has been possible to construct an evolutionary pathway for the rusts of the Australia-New Zealand area in which the important events have been underlined (Watson, 1981; Burdon et al. , 1982). A third factor in the build-up of knowledge on rust epidemiology is the continuous documentation of the population structure of the New Zealand rusts. Arrival of a new rust in Australia is followed after a short interval of some months by its isolation in New Zealand (Hogg é¢ al., 1969; Luig, 1977). There is now irrefutable evidence that cereal rust spores are transported across the Tasman Sea at frequent intervals, a distance of at least 2,000 km. (iii) Over Oceans. There are as yet no reports of wind-borne inoculum of P. graminis tritict having crossed any of the world’s oceans. With the discovery of the coffee rust organism (Hemuileia vastatrix Berk. & Br.) in Brazil in 1970 it was suggested (Bowden et al., 1971) that since trade winds blow frequently and for long periods across the Atlantic Ocean spores of this organism had been transported from the African con- tinent. Since the strains observed on the two continents showed similar characters this appeared likely but Ingold (1978) believes there is some doubt as to whether this organism is normally dispersed by wind. Australian workers have speculated on the possible overseas origin of several strains of P. graminis tritict that have successfully colonized this country. Luig (1977) discusses this in some detail but gives no data to pin point the origin of any strain common to the area. The original six standard races found 60 years ago were characteristic of Australia but no origin ‘for them has been suggested. Luig (1977) described how these and other rusts were replaced with the passage of time by more aggressive strains. Three periods 1925, 1948-54 and 1968-69 are important in this replacement. Entry of new rusts to the country at those times probably involved wind transport of spores across the great distances of the Pacific or Indian Oceans. The possibility of accidental or deliberate introduction by man must be largely discounted since com- munication between the critical areas of Australia and Africa would be minimal. In the period 1948-54 standard race 21 was isolated as a type completely new to Australia. This race was at the time among the most common in the world but there were no comparative data from overseas to suggest from where it came. Almost 20 years later in 1968-69 further dramatic changes occurred in the population of P. graminis tritici in Australia. Intensive investigations were commenced to uncover the possible origin of the strains involved in this latest event. ISOLATION OF STRAINS FROM AFRICA AND A COMPARISON WITH THEIR AUSTRALIAN COUNTERPARTS (i) Early suggestions of Similarities. Luig (1977) has already outlined some of the details concerning studies of two Australian strains of P. graminis tritict which differed markedly from any previously isolated in this area. Both collections had been made in 1969 by the senior author, one 326-1,2,3,5,6 from Clinton, S. Australia, the other 194-1,2,3,5,6 from Tichborne, N.S.W. The material on which each was collected had few large isolated pustules and carried the gene S76. Although these collections were part of the annual race survey conducted by Luig and Watson the identifications revealed something unique. The first clue that overseas Proc. LINN. Soc. N.S.W. 106 (4), (1982) 1983 314 LONG DISTANCE TRANSPORT OF SPORES inoculum may have been involved in the origin of these two rusts came from data sent by Dr J. C. Santiago who was then at Elvas, Portugal. These data showed that stan- dard races 194 and 222 were present in Mozambique and there was evidence that pathogenicity on some lines of the Australian supplementals was the same as that of the two new local rusts. In order to make detailed comparisons of rusts from the two continents inoculum was requested from Angola, Rhodesia (now Zimbabwe) and Mozambique. (ii) Detailed studies of Pathogenicity of African rusts. Two collections were received from Angola, one from Nova Lisboa the other from Caconda. A single collection was sent from Salisbury, Rhodesia. Because these collections were from overseas, strict quarantine precautions were taken during the studies with them. All tests were made in glasshouses at the University of Sydney, Glebe, thus minimizing the possibility of spores escaping and lodging on a congenial host. At the conclusion of each test the plant material and the rust were collected and destroyed. The strain components of the three collections were identified using the dif- ferentials of Stakman et al. (1962), supplemented by the eleven local differentials adopted in Australia by Luig and Watson (1977). After the initial multiplication of spores, inoculum was placed onto certain key varieties vzz. Marquis (S77b), Reliance (S75), Acme (S79g), Einkorn (S721), Vernal Emmer (S79e), McMurachy ($76), Yalta ($711) and C.1.12632 (SrTt,). Many isolations were made and finally from single pustules five strains were identified from Nova Lisboa and one from Caconda. ‘Two strains were found in the Salisbury material. According to the classification of Luig and Watson these were designated as in Table 1 and the seedling infection types on certain wheats are given in Table 2. The two Rhodesian strains were differentiated on Acme and Barleta Benvenuto and no differences could be found between Rhodesia 2 and Angola 4. Although all differentials were tested, complete separation of the strains was possible using only those five of Table 2. (111) Studzes on the Australian counterparts. As indicated above large isolated pustules were found in 1969 at Clinton, S.A. Laboratory tests showed this culture 69822 to be of strain 326-1,2,3,5,6. As reported by Luig (1977) this combination 1(S76), 2(S711), 3(S79b), 5(S717), 6(S78) of virulences had not been previously reported in Australia. Moreover and very spectacularly, this strain was avirulent on plants with S7r7b a host- parasite interaction that had not been seen in Australia for about 30 years. The strain was Clearly new to the population. TABLE 1 Strains of P. graminis tritici zsolated from Angola and Rhodesia Culture Number Location Australian Designation Angola 1 Nova Lisboa 194 — 1,2,3,5,6,7 ee 2 ny 194 — 1,2,3,5,6 oe 3 ae 343 — 1,2,3,5,6 ie 4 me 222 — 1,2,3,5,6,11 re, 5 gi 194 — 1,2,3,5,6,* 2 6 Caconda 222 — 1,2,3,5,6 Rhodesia 1 Salisbury 34 — 1,2,3,5,6 4 2 He 222 — 1,2,3,5,6,11** * Acme mesothetic ** As Angola 4 PRoc. LINN. Soc. N.S.W. 106 (4), (1982) 1983 I.A. WATSON ANDC.N.A.DESOUSA 315 TABLE 2 Seedling infection types on key differentials selected from the standard and local sets when inoculated with seven strains of P. graminis tritici from Angola and Rhodesia Differential Rust Cultures from Standard Marquis S77b Reliance S75 Acme S79g Norka $715 Barleta Benvenuto $7BB mm MPP Shortly after the collection at Clinton was made culture 691042 was found about 1,000 km east at Tichborne, N.S.W. It was classified as 194-1,2,3,5,6. Hence within a short time two rusts vastly different from any previously seen in this country were recovered. They differed only in the infection types found on plants with S77b. Luig however, in unpublished work, has shown that although avirulence on plants with S715 had been found before in Australia these two strains produced a distinctly lower in- fection type on plants with S715 than all other Australian strains with P15. At the time of commencement of these studies in 1972 we had access to both these strains, but on comparing the African rusts it was found that only one, Angola 2 (194-1,2,3,5,6), had an Australian counterpart even though six of them had the numbers 1,2,3,5,6. Both 326 and 194 are avirulent on plants with $75 and a careful search was made for variants of them with virulence. Strain 326-1,2,3,5,6 was soon predominant in southern Australia and since Summit (575) was widely grown in Victoria the appropriate mutation could be anticipated. This did not appear, however, until the rust epidemic of 1973 when at Kerang, Victoria, one of us (I.A.W.) collected rust (culture 73879) on a roadside plant of Hordeum leporinum Link. It was identified as 343-1,2,3,5,6 and represented the second counterpart of the African rusts, in this case Angola 3. The third counterpart was of Angola 6 (222-1,2,3,5,6) and this was collected at Willowie, S. Australia in 1974 (culture 74220). Luig and Watson later identified a fourth counterpart (culture 74654) 194-1,2,3,5,6,7 (Angola 1) at Yacup in W. Australia in 1974-75 and it was used in some of the tests. (iv) Comparison of the Pathogenicity of the Angola strains and their Australian counterparts. (a) On the standard and local differentials. When the components of the African collections had been purified and when from the survey of Luig and Watson it was found that in the Australian wheatfields counterparts of at least three of them were present, detailed comparisons were commenced. The Angola numbers were 2(194-1,2,3,5,6), 3(343-1,2,3,5,6) and 6(222-1,2,3,5,6). The corresponding three Australian cultures were 691042, 73879 and 74220. The six cultures were used in pairs to inoculate seedlings of the standard and supplemental differentials. The infection types of the Angola cultures are presented in Table 3 and were identical with those of their respective counterparts. From these readings it was apparent that the only genes important in the differentiation of the six cultures were S75 and Sr7b. We concluded that this set of genetic material could not be used to distinguish between the members of each pair of strains. (b) On entries in the International Virulence Gene Survey. At the time of analysing the African collections one of us (I.A.W.) was involved in testing the pathogenicity of Proc. LINN. Soc. N.S.W. 106 (4), (1982) 1983 316 LONG DISTANCE TRANSPORT OF SPORES TABLE 3 Seedling infection types at 17-22°C of 11 standard differentials and 11 local supplementals when tested with the three strains of P. graminis tritici from Angola Standard Differential Little Club Marquis S77b Reliance S75 Kota Arnautka Mindum Spelmar Kubanka Acme S79g Einkorn Vernal Australian Angola Culture and Strain Supplemental Angola Culture and Strain 2 3 6 2 3 6 194-1,2, 343-1,2, 222-1,2, 194-1,2, 343-1,2, 222-1,2, 3,5,6 3,5,6 3,5,6 3,5,6 3,5,6 3,5,6 4 4 4 McMurachy $76 4 4 4 4 2 4 Yalta S711 4 4 4 0 4 4 W2402 S77b Sr9b 4 2 4 4 4 4 C.1. 12632 SrTt, sil = ole ile 4 4 4 Renown Sr7b Sr17 4 2 4 4 4 4 Mentana S78 4 4 4 4 4 4 Norka S715 xX x x oD = oD = oD) = Festiquay 5730 aD, a) a7) i ;2= ;2= Agropyron intermedium : : : ;1- ;1- ;1- Entrelargo 2+ 2+ 2+ de Montijo ;1- ;1- ;1- Barleta Benvenuto S7BB x x xX several Australian strains on seedlings of a collection of 117 wheats. These had been selected as part of a survey to catalogue on a world basis the geographic location of virulence genes and combination of them. The group had been divided into six sections according to the alleles they carried. Group 1 Group 2 Group 3 Group 4 Group 5 Group 6 Alleles with world wide usefulness as differentials. S7 genes such as 5,6,8,9b,11 and Tt, were included. Alleles useful as differentials only in certain areas of the world. S7 genes 7a,/b,9a,10,13,14,15,17,23 and several others as yet uncatalogued were included here. Combinations of known alleles. These had two or more Sr genes per line. Combinations of unknown alleles which were found to react differentially from one region of the world to another. They were complex genotypes and included such cultivars as Thatcher, Chris, Selkirk, Timgalen and Gamut. Lines of unknown genetic constitution which were highly resistant to all strains in many locations. The group included lines of Triticum turgidum L. as well as many lines of T. aestivum L. from the International Spring Wheat Rust Nursery. Lines commonly used as susceptible parents in crosses and comprised Line E W3498, W 2691 a Little Club derivative, Marquis and Chinese Spring. All 117 lines of this collection had been tested with the available Australian strains that closely resembled those from Africa. As well they were tested with Angolan rusts numbers 1,3,4, and 6 and Rhodesia number 1. From the tests it was confirmed that the only genes effective in differentiation were S75, S77b, S715 and S7BB. Angola 2 was not compared with culture 691042 in these tests but again no separation between members of the pairs was possible despite the extent of the genetic variability in the host material. (v) Comparisons of other Characters in the African and Australian Rusts. Proc. LINN. Soc. N.S.W. 106 (4), (1982) 1983 I. A. WATSON ANDC.N.A.DESOUSA 317 (a) Shape and Size of Uredospores. The length of the uredospores of Angola 3,4 and 6 were compared with the Australian cultures 73879 and 74220. On the basis of measure- ments on 950 spores of each there were no significant differences between Angola 3(33.48 + 1.75) and 73879 (34.56 + 2.18) and between Angola 6(34.32 + 2.99) and 74220 (34.08 + 2.91) at the P = 0.01 level although there was a marginal difference between the first two at P = 0.05. Angola 4(30.08+ 2.06) was distinctly rounder than any other culture but it had no Australian counterpart. In width of spore all cultures examined fell within the range 17.644 1.84 — 18.684 1.25p. (b) Uredospore Colour. There was no difference between Angola 2,3 and 6 and their Australian counterparts. Angola 4 and Rhodesia 2 with the same pathogenicity both differed from the other African rusts in having yellowish rather than brown uredospores. (c) Growth in Axenic Culture. Uredospores of Angola 3 and 6 along with their Australian counterparts were sprayed separately onto Williams’s (1971) agar medium in Petri dishes. The material was observed under the microscope 24 hours later and the structures showing were identical in all four cultures. This comparison as with those for pathogenicity, spore morphology and colour was of no value in demonstrating dif- ferences between corresponding rusts from the two continents. DISCUSSION AND CONCLUSIONS (1) Pathogenicity and other Characters of the Cultures. It is apparent from the results obtained that there is considerable diversity among the strains of P. gramuinis tritici in that part of the African continent represented by Angola and Zimbabwe. It was not the purpose of the study to examine in detail the extent of this variability but to compare African rusts with those from Australia where there were known counterparts. Four Angola cultures 1(194-1,2,3,5,6,7), 2(194-1,2,3,5,6), 3(343-1,2,3,5,6) and 6(222-1,2,3,5,6) each had counterparts in Australia which were respectively cultures 74654, 691042, 73879 and 74220. When the overseas cultures were examined in paired comparisons with the corresponding local strains no differences in the members of each pair could be detected, even though the host material was very diverse. The uredospores were of the same size within reasonable limits of variation and although one Angola culture was clearly different from the others in spore shape, no counterpart for it was found in Australia. Growth in axenic culture which is sometimes useful in distinguishing between strains revealed no differences between local and introduced material. Although four counterparts were examined we are postulating that strains 326-1,2,3,5,6 and 194-1,2,3,5,6 arrived initially and the others 343-1,2,3,5,6, 194-1,2,3,5,6,7 and 222-1 ,2,3,5,6 evolved from them here by mutation. (ii) Significance of certain genes for Avirulence in the Australian environment. Luig has already pointed out (1977) that when found late in 1969 the two cultures 69822 and 691042 carried the formula 1,2,3,5,6 from their infection on the local supplementals. This was a new combination for Australia. Virulence on plants with S76 (1) gave the first im- portant hint in the field of this new material. In the previous four seasons 1965, 1966, 1967 and 1968 only seven isolates with this ability had been found east of the Nullarbor Plain among 1684 examined during the period (Luig and Watson, 1970). In addition both these new strains had another rare character, virulence on plants with 578. In the five years prior to 1969 only 12 isolates from a total of 2603 from throughout Australia were virulent on plants with this gene and not one of these could attack the Sr6 S78 combination. Thus to find these two cultures each with such a rare genotype £6 8 was unexpected enough, but all the more so when one of them in addition had avirulence on plants with S77b. Proc. LINN. Soc. N.S.W. 106 (4), (1982) 1983 318 LONG DISTANCE TRANSPORT OF SPORES It has previously been pointed out that during the 15 years prior to 1970 new patterns of pathogenicity arose in the Australian rust population due to the progressive introgression of new alleles into it (Watson and Luig, 1966). About 1950 the fungal alleles p5, £8, p15 and pBB predominated locally in P. graminis tritict. With the arrival of 21-0 all this changed since this strain brought with it P5, P8, P15 and PBB. It is clear from our studies that of the eight alleles at these four loci, seven of them were found in the African collections. P8 was missing. This all points to Africa also as the source of standard race 21 in the 1948-54 period and we strongly suggest this as a real possibility. (iii) The origin of strain 343-1, 2,3,5,6. It may be argued that our failure to recover this strain from the African material suggests some alternative origin for it. We pointed out earlier however that only three collections were available and these may not have been adequate to sample the area. If spores were windblown from Africa it would be ex- pected that they would be of the most common strains. We know from Dr Santiago that components of race 194 had been present in Central Africa and in 1970 Fonseca (1974) found races 194 and 222 in Mozambique. To the best of our knowledge avirulence on plants with S77b had not been identified in the area. However the ease with which its presence was established in our collections demonstrates that it was part of the population at the time of these introductions. During the interval between the movement to Australia (1968-69) and the isolation in our tests (1972-73) the population may have shifted from 326-1,2,3,5,6 to 343-1,2,3,5,6 by simple mutation at the P5 locus. The demonstration of contrasting pathogenicity on plants with S77b requires good control over environment especially temperature and without it the four rusts of this family can easily be misclassified as shown in Fig. 1. P5 to ps 320 ee P7b |to p7b P7b to| p7b P5 tops 194 S$ 222 Fig. 1. Changes in Standard race number with single gene mutations at the locus indicated. We believe that 326-1,2,3,5,6 entered Australia about 1968 multiplied on susceptible cultivates in southern states especially on Halberd (S76), mutated to virulence at the P5 locus and under selection pressure resulting from the cultivation of wheats such as Summit (S75) those mutants became apparent as 343-1,2,3,5,6 (73879). (iv) Other similarities between African and Australian rusts. All the evidence we have presented and the evidence of Luig and Watson (1970) and Luig (unpublished) show a strong resemblance between the rust cultures we have compared. Unless gene flow has been involved, the similarity is much greater than would be expected in cultures originating in countries separated by the Indian Ocean. While this conclusion has been reached mainly from comparisons of pathogenicity, Burdon e¢ al. (1982) have reached the same conclusion using isozymes as genetic markers. This new approach can also be used to trace evolution in rust populations. While these workers did not use material identical with ours their results completely confirm the close similarity between the strains of the two continents. No isozyme differences were found between the isolates of standard races 21, 194, 222, 326 and 343. Moreover two isolates from Angola viz 2 and PROC. LINN. Soc. N.S.W. 106 (4), (1982) 1983 I. A. WATSON ANDC.N.A.DESOUSA 319 Fig. 2. Initial trajectories of 3 identical constant level balloons released almost simultaneously from Men- doza, Argentina on 20.9.71 (Partially redrawn from Morel and Bandeen, 1973). C = Clinton, S. Aust., T = Tichborne, N.S.W., N = Nova Lisboa, Angola, P = Pretoria, S. Africa, R = Rio de Janeiro, Brazil, B.A. = Buenos Aires, M = Mendoza, Argentina, L = Lima, Peru, SPR = South Polar Region. Stippling indicates the probable source of spores. 4 were found to have the same phenotype as the following Australian rusts: 194-1,2,3,5,6 (S837 from Queensland), 326-1,2,3,5,6 (69822) and 343-1,2,3,5,6 (S1205 from Queensland). These workers have also proposed an African origin for strain 21-0 and this supports the conclusion we have also reached. (v) Weather conditions making spore transport possible. The shortest distance between the wheat fields of Africa and Australia is approximately 12,000 km. To postulate natural movement of viable spores across this distance of water and the establishment of colonies from them in Australia we must assume ‘a delicate synchrony of biological and physical processes’ (Hirst and Hurst, 1967). It has been demonstrated that the rust strains from the two areas are indistinguishable, and it is known that physical processes are available to elevate spores of them trom the crop to considerable heights. According to Ingold (1978) rust uredospores are set free around midday, and under relatively dry conditions vigorous thermals could carry the spores high into the air. Proc. LINN. Soc. N.S.W. 106 (4), (1982) 1983 320 LONG DISTANCE TRANSPORT OF SPORES Schematic air paths shown by Ludlam (1967) suggest how they may be carried by rain clouds into the high troposphere, and possibly to the jet stream. Once into the atmosphere there are a number of meteorological systems which could transport these spores to Australia. The first would be the anticyclones moving at a height of up to 3 km and covering the distance in about six days. While there will be a tendency for the spores to fall or to be washed out, circulation will have the effect of lengthening the period when they are airborne. The fall speeds of the spores would be of the order of 1 cms"! so that having reached an initial height of 5 km over the ocean, drift could carry them easterly for another 10,000 km. If they are in winds of 10 ms"! and falling at 1 cm s! they would cross the Indian Ocean at the rate of about 1000 km each day (Pedgley, 1980). As indicated above, some spores may reach the jet stream at a height of about 7-12 km and travel at much greater speed covering the distance in about two days. It has become apparent from the work of Lalley et al. (1966) and of Morel and Bandeen (1973) that jet streams in the southern hemisphere are much more continuous and rapid than originally estimated. Monitored balloons released in New Zealand and Argentina completed the global circuit in 10-12 days. Some of these trajectories are partially redrawn in Fig. 2, and show that the balloons may pass over the critical areas of both Africa and Australia. Presumably if spores reached the jet streams they could do the same, and eventually subside following air current originating in local thun- derstorms. Once at ground level the spores would need to reach a congenial host on which to colonize. Apparently in 1968-69 the favourable climatic conditions which enabled Australia to produce a record wheat crop were also those allowing migrant rust spores to become established. Intercontinental travel for spores of P. graminis tritict is hazardous and of rare occurrence. It is for this reason that we believe Australia has been involved in this migration only three times in the last 60 years. ACKNOWLEDGEMENTS The authors would like to thank Mr J. Colquhoun, Senior Research Meteorologist of the Department of Science, Sydney, and Dr I. Watson of the School of Earth Sciences at Macquarie University for help in interpreting the weather data. The work is based on a thesis submitted to the University of Sydney by the junior author for the degree of M.Agr. while holding a fellowship sponsored by the Food and Agriculture Organisation. We are grateful to Mr L. N. Balaam, Dean of the Faculty of Agriculture and Mrs Helen Keys for the typing of the manuscript. References BOWDEN, J., GREGORY, P. H., and JOHNSON, C. G., 1971. — Possible wind transport of coffee rust across the Atlantic Ocean. Nature 229: 500-501. BURDON, J. J., MARSHALL, D. R., Luic, N. H., and Gow, D. J. S., 1982. — Isozyme studies on the origin and evolution of Puccinza graminis f. sp. tritici in Australia. Aust. J. Biol. Sct. 35: 231-238. FONSECA, N., 1974. — Racas fisiologicas de Puccinia graminis f. sp. tritici (Eriks. and E. Henn.) em Mocambique 11. — Resultados da prospecca feita em 1971. Inst. de Investigacao Agronomica de Mocambique. Informacao Tecnica No. 77. Hocec, W. H., Hounam, C. E., MALLIK, A. K., and ZADOKS, J. C., 1969. — Meteorological factors af- fecting the epidemiology of wheat rusts. World Meteorologic Organisation Tech. Note No. 99. Hirst, J. M., and Hurst, G. W., 1967. — Long-distance spore transport. In GREGORY, P. H., and MONTEITH, J. L., (Eds.), Azrborne Microbes. Cambridge: Cambridge Univ. Press. INGOLD, C. T., 1978. — Dispersal of micro-organisms. In SCOTT, P. R., and BAINBRIDGE, A., (Eds), Plant disease Epidemiology. Oxford: Blackwell Scientific Publications. Proc. LINN. Soc. N.S.W. 106 (4), (1982) 1983 I. A. WATSON ANDC.N.A. DESOUSA 321 LALLEY, V. E., LICHFIELD, E. W., and SOLOT, S. B., 1966. — Global Horizontal Sounding Technique. World Meteorologic Organisation Bull. 15: 124-128. LupDLaM, F. H., 1967. — The circulation of air, water and particles in the troposphere. In GREGORY, P. H., and MONTEITH, J. L., (Eds), Azrborne microbes. Cambridge: Cambridge Univ. Press. Luic, N. H., 1977. — The establishment and success of exotic strains of Puccinia graminis tritici in Australia. Proc. Ecol. Soc. Aust. 10: 89-96. , and WATSON, I. A., 1970. — The effect of complex genetic resistance in wheat on the variability of Puccinia graminis f. sp. tritici. Proc. Linn. Soc. N.S.W. 95: 22-45. , and , 1977. — The role of barley, rye and grasses in the 1973-74 wheat stem rust epiphytotic in southern and eastern Australia. Proc. Linn. Soc. N.S. W., 101: 65-76. MoREL, P., and BANDEEN, W., 1973. — The Eole experiment. Early results and current objectives. Bull. Amer. Meteorolog. Soc. 54: 298-306. PEDGLEY, D. E., 1980. — Weather and airborne organisms. World Meteorologic Organisation Tech. Note No. 173. STAKMAN, E. C., STEWART, D. M., and LOEGERING, W. Q., 1962. — Identification of physiologic races of Puccinia graminis var. tritict. Agr. Res. Service E-617 (Rev. ). 53 p. WaTSON, I. A., 1977. — The national wheat rust control programme in Australia. Vestnik sel skochozja —‘ jstvennot Nauki No. 6, 123-133. (Russian). ——, 1981. — Wheat and its rust parasites in Australia. In EvANS, L. ‘T., and PEACOCK, W. J., (Eds), Wheat science — Today and tomorrow. Cambridge: Cambridge Univ. Press. , and Luic, N. H., 1966. — S715 — a new gene for use in the classification of Puccinia graminis var. tritict. Euphytica 15: 239-250. WILLIAMS, P. G., 1971. — A new perspective of the axenic culture of Puccinia graminis f. sp. tritict from uredospores. Phytopathology 61: 994-1002. Proc. LINN. Soc. N.S.W. 106 (4), (1982) 1983 A Study of the Pollination of Alocasia macrorrhiza (L.) G. Don (Araceae) in southeast Queensland DOROTHY E. SHAW and B. K. CANTRELL SHAW, D. E., AND CANTRELL, B. K. A study of the pollination of Alocasia macrorrhiza (Araceae) in southeast Queensland. Proc. Linn. Soc. N.S.W. 106 (4), (1982) 1983: 323-335. Flowering of Alocasia macrorrhiza (L.) G. Don in southeast Queensland was in- vestigated and appears to be protogynous. Insects in the open spathal chambers, on the spadix after closure of the constricting area, and in the sealed chamber of the nearly mature infructescences were recorded to determine possible pollinators. The number of seed set in an experiment involving various combinations of emasculation and bagging of 143 inflorescences including controls appear to indicate cross pollination by insects. Dorothy E. Shaw, Plant Pathology, and B. K. Cantrell, Entomology, Department of Primary Industries, Mewers Road, Indooroopilly, Australia 4068; manuscript received 26 June 1982, accepted for publication 20 October 1982. INTRODUCTION Alocasia macrorrhiza (LL.) G. Don (Araceae) occurs in wet tropical Asia, eastern Australia and the Pacific Islands; to some of the latter at least it was probably in- troduced, as well as to other parts of the wet tropics and subtropics. In Australia it occurs in the high rainfall coastal areas and adjacent plateau regions in Queensland and New South Wales, from about 12° to 34°S latitude. Hamilton (1898) reported the beetle, Brachypeplus murray: Macleay (Coleoptera: Nitidulidae) as the pollinator of A. macrorrhiza, and that a species of Agaristidae (Lepidoptera) also aids in fertilization. McAlpine (1978) described Neurochaeta inversa (Diptera: Neurochaetidae) (whose life cycle is intimately connected with the flowers of this plant growing in its natural habitat in eastern Australia) which he thought may act as the pollinator. He observed, however, that as many insects visit the flowers, it is probably not the sole agent. Shaw et al. (1982) reported that A. macrorrhiza still set seed in gardens where no N. inversa were recorded, and in remnant natural stands now surrounded by mainly non- forested areas, where occurrence of N. inversa was sparse. It was also shown that seed set in plants from rainforest areas was not related to the numbers of N. inversa recorded in chambers of developing and matured infructescences. They concluded, therefore, that pollination is probably independent of ovipositing females of N. znversa. The present study records the stages of flowering and the insects associated with the inflorescences or flowering heads (hereafter called ‘heads’) and in the nearly mature infructescences. Also reported are the results of an experiment involving emasculation and bagging of heads before anthesis, with controls, to provide information on the possibilities of selfing or crossing, and of wind or insect pollination, as measured by seed set. SITES AND METHODS Sites A. macrorrhiza has been under study since 1980, but the work reported below was mainly carried out in the flowering period October 1981 to March 1982 in southeast Proc. LINN. Soc. N.S.W. 106 (4), (1982) 1983 324 POLLINATION OF ALOCASIA MACRORRHIZA Queensland, at sites in rainforest areas along the coastal range west and southwest of Brisbane, and in simulated rainforest areas (gardens) in Brisbane suburbs. Details of the sites are as follows: Rainforest: Three sites in small private holdings adjacent to large or relatively large reserved areas of rainforest at Mt Nebo, Mt Glorious and Mt Tamborine, at about 520, 650 and 500 m above sea level (a.s.1.) respectively; plants at the first and third sites were growing in and at the edge of streams; at each site plants occurred in scat- tered clumps. Garden: Three sites in the Mt Coot-tha Botanic Gardens, developed as simulated rainforest in natural sclerophyll forest during the last four years, and one site at the University of Queensland, St Lucia, surrounded by parklands; all sites about 50 m a.s.l.; plants at one site at Mt Coot-tha and most plants at St Lucia, at the edge of streams; at each site plants occurred in scattered clumps. Methods At each site inflorescences were examined throughout the entire flowering period and observations made on the various stages of flowering, both in the field and on excised heads removed to the laboratory. In some in situ (non-treatment) and other excised heads, small hinged ‘windows’ about 1 cm square were cut into the wall of the spathal chamber for stigma viewing, the flap being resealed with adhesive tape. All sites were visited as frequently as possible after the treatments below were applied. Any rotting heads were harvested when the rotting became evident, but heads maturing were allowed to remain in situ until near maturity but before actual splitting of the wall of the spathal chamber to ensure no seeds were lost. Each head was examined at harvest, and if any fertilization had occured, the numbers of unfertilized ovaries, fertilized ovaries (berries) and fertilized ovules (seed) were determined. a. Treatments Emasculation and bagging were carried out only on ‘buds’ when each spadix was still enclosed by the furled spathe, prior to anthesis. The treatments were as follows: 1. Emasculated, mesh-bagged. This involved the excision of the distal portion of the spathe (consisting of the furled spathal limb, the sterile terminal appendage, the portion bearing the staminate flowers and the upper portion of the median sterile florets) at the constricted zone. The excised part of the spadix was removed from the excised spathal limb and checked to ensure that all the staminate flowers had been removed. The portion of the spadix remaining on the plant (consisting of the lower part of the median sterile florets and the pistillate flowers) was enclosed by a nylon mesh bag fine enough to keep out insects larger than thrips — no thrips, however, although they could physically enter, were found inside the bags. Emasculated, unbagged. As above, but unbagged. 3. Non-emasculated, whole head paper-bagged. A white paper bag normally used in maize pollination studies was employed. 4. Non-emasculated, whole head plastic-bagged. A commercially available clear plastic bag was found to be suitable. 5. Non-emasculated, whole head mesh-bagged. Similar bag as used in Treatment 1. In some cases the bags were distended by the insertion of a framework composed of two metal rings maintained in position on a rigid metal upright attached to the peduncle. 6. Non-emasculated, unbagged, labelled. Heads untouched, labelled as controls while still in bud. During each visit to the field, at least one set of the six above treatments were prepared, heads being labelled individually. Occasionally additional heads of some of NO Proc. LINN. SOc. N.S.W. 106 (4), (1982) 1983 DOROTHY E. SHAW ANDB. K. CANTRELL 325 the treatments were included if extra heads were available. All the paper, plastic and mesh bags were tied at the base with nylon net tape. 7. Non-emasculated, unbagged, unlabelled. Early maturing heads at each site, previously unused in our experiments, were also harvested as further controls for comparison. b. Insects Records were made of all insects observed within the open spathal chamber when the pistillate flowers were exposed and those later found around the staminate flowers and on the accumulated pollen of the spathal ‘catchment area’ above the constricted portion at the base of the spathal limb. An aspirator was used to collect representative samples of all species of insects encountered (from non-treatment heads). Any insects in the sealed spathal chambers of treatment heads at harvest, whether immature or adult, were also recorded and retained. RESULTS Flowering After the green spathal limb unfurls, the constricting portion at the base of the limb gapes widely, with a gap of 3-4 mm around the spadix, so that the pistillate flowers can be partially viewed from above. At this stage the subsessile stigmas appear dull, white and rough, and the 2-3-(4) short lobes are seen microscopically to consist of a cushion of short hyaline ‘hairs’ ca 138 to 227 um long and ca 13 to 23 wm wide (mean 157.1 by 16.2 um), to which the sticky pollen grains adhere. The chamber is open for about one day in the field, after which the constricting portion of the spathe closes tightly around the spadix in the region of the median sterile florets, sealing off the chamber. Pollen is then extruded in white cirri from the synandria of the staminate flowers for about two days, and fallen pollen accumulates in the pollen catchment area formed at the base of the spathal limb abave the constriction. At this stage, stigmas in the enclosed chamber are glistening, white and glutinous. (The only exceptions noted to the above were a few heads from 650 m a.s.1., excised at the beginning of anthesis, and then held in the laboratory at about 50 m a.s.1. Some pollen in each head began to extrude, mainly from the lower staminate flowers, before the final closure of the constricting portion. This may have been because the heads had not been visited by insects, and the ‘unfertilized’ stigmas may have been receptive longer, perhaps delaying the closure. Also, as Stanley and Linskens (1974) pointed out, altitude and temperature affect pollen dehiscence, so that the warmer temperature in the laboratory at the lower altitude may have caused extrusion of the pollen earlier than normal. This, however, was an unnatural situation.) A fragrance, present at the beginning of anthesis, was still present after pollen extrusion, though less pronounced. Details of pollen studies will be published separately, but are summarized here as follows: When extruded in the cirri, the pollen is sticky and will adhere to a scalpel blade, but the stickiness decreases with age, so that accumulated pollen, although still mainly in clumps, could be launched into the air by deliberately jarring the head. In the cirri, the pollen appears white, but when shaken out of the head is deep ivory, about plate 4A2 in the Methuen Handbook of Colour (Kornerup and Wanscher, 1961). The grains are microscopically colourless, spherical, 32.3-47.54m long by 30.3-47.5 um wide (mean 41.2 x 39.3 mm) excluding ornamentation of surface spinules ca 2 um long. Twenty-two samples of pollen shaken from 10 heads, examined microscopically, revealed no pollen other than that of A. macrorrhiza. In germination tests on modified Brewbaker and Kwack (1973) medium, pollen shaken from heads in the laboratory at the beginning of each day was mainly viable during the first and Proc. LINN. Soc. N.S.W. 106 (4), (1982) 1983 326 POLLINATION OF ALOCASIA MACRORRHIZA TABLE 1 Presence ( +) or absence (—) of insects recorded in the open chambers (O) and around the spadix after sealing of the chambers (AS) of non-treatment heads HABITAT, Insects locality, N. inversa T. Brachypeplus Staphyl- Taento- Other* site carbonaria sp. inidae thrips O AS O AS O AS O AS O AS O AS RAINFOREST Mt Nebo + + + + + + + + Sto 5 Mt Glorious + + + + + + + + - + 27 Mt Tamborine + + - - + + + + - + BaOaae) GARDEN Mt Coot-tha Site 1 +# - + + + + + + ey 6 Site 2 - = ft + fs fi + + See 1 Site 3 +# - + + + + + + - + 8 St Lucia - - - = 4 f + ns SE % Other = 1 A. mellifera # 2 Pteromalidae # 3 Curculionidae (5 specimens: 4 open chamber; 1 around spadix) 4 P. fimetarius # 5 Cyphon sp. (2 specimens) 6 T. ? minor # 7 Hippelates sp. (2 specimens) 8 Dolichopodidae # # only single individuals second days of production, with much lower and slower germination on the third day and only rarely on the fourth day. Some sterile (non-staining) inviable pollen was present in samples from all sites. Insects 1. Within the open chamber Four species of insects were commonly encountered within the open chamber (Table 1). These were N. inversa, Trigona carbonara Smith (Hymenoptera: Apidae), Brachypeplus sp. (Coleoptera: Nitidulidae) and an unidentified alaocharine staphylinid (Coleoptera: Staphylinidae). The two beetles were present at all localities while N. inversa was limited to rainforest localities except for two specimens (only one captured) at Mt Coot-tha. Although common at most localities, T. carbonaria was not recorded during any visit to Mt Tamborine or St Lucia. Several other insects were occasionally recorded within open spathal chambers, but usually only as single individuals. These were Taenzothrips (Isochaetothrips) sp. (Thysanoptera: Thripidae), Cyphon sp. (Coleoptera: Scirtidae), Triphyllus ?minor Lea (Coleoptera: Mycetophagidae), an unidentified derelomine weevil (Coleoptera: Curculionidae), Hippelates sp. (Diptera: Chloropidae), an unidentified dolichopodid (Diptera: Dolichopodidae) and an _ unidentified pteromalid (Hymenoptera: Pteromalidae) (Table 1). 2. On the spadix after closure of the constricting portion Four species of insects were commonly encountered on the spadix (or in the ac- cumulated pollen) after closure of the spathal constriction. These were 7. carbonara, Brachypeplus sp., the staphylinid and Taeniothrips (Isochaetothrips) sp. All were present at each locality except for 7. carbonara which was absent as noted previously (Table 1). Proc. LINN. Soc. N.S.W. 106 (4), (1982) 1983 DOROTHY E. SHAW ANDB. K. CANTRELL 327 TABLE 2 Insects other than Neurochaeta inversa recorded in sealed chambers of matured heads * Insects” HABITAT, Head Brachypeplus Chloro- Diptera Fermi- Collem- locality designation sp. pidae Muscidae unidentified cidae bola No. No. No. No. No. No. RAINFOREST Mt Nebo 7A.4 2L 2P Mt Glorious 6A.1 1 L# X.1 iL X.2 1L Mt Tamborine 1A.1 1L 1A.4 2L 9A.1 1L 9A.2 1L X.1 1A X.2 6P WIL X.4 1A 1P DI 2L GARDEN Mt Coot-tha Site 2 1A.1 iL Chambers in 131 observed heads (including unlabelled controls) were without insects, other than presence of NV. inversa in some. Life forms: L = larva; P = pupaor puparium; A = adult. # Larvae in one of two damaged ovaries. = Dead. TABLE 3 Numbers and fate of heads in seven treatments Heads in the following classes HABITAT, Emasculated Non-emasculated locality, = site Mesh- Unbagged Paper- Plastic- Mesh- Unbagged, Unbagged, bagged bagged bagged bagged labelled unlabelled M R M R M R M R ME TR EIR M R No. No. No. No. No. No. No. No. No. No. No. No. No. No. RAINFOREST Mt Nebo 0 1 0 1 4 0 Se Mt Glorious 0 5 1 2 0 2 0 3 0 7 2 3 3 Mt Tamborine 0 6 60 6 0 2 4 0 5 1 i} GARDEN Mt Coot-tha Site 1 0 3. (0 3 0 2 3 1 2 3 1 Dh ok Site 2 0 3 2 1 0 32> 2 0 Sie 33 1 1 ex Site 3 0 1 0 2 0 Se) 3 0 1 3 1 Sk St Lucia 0 3 0 1 0 3 4 0 2px Total OP A1s eS SS as QO ts OQ ag LU} 28} 7 Bil sx M = Matured; all heads with seed (details Table 4) R = Rotted; all heads without seed except four with few seeds (details Table 4) x = Some rotted heads occurring naturally, but uncounted. Proc. LINN. Soc. N.S.W. 106 (4), (1982) 1983 328 POLLINATION OF ALOCASIA MACRORRHIZA TABLE 4 Numbers of seed set in treated heads Heads in the following classes HABITAT, Emasculated Non-emasculated locality, Ri Sane hy en ES Pe EE TL a Ee ee ee site Mesh- Unbagged Paper- Plastic- Mesh- Unbagged, Unbagged, bagged bagged bagged bagged labelled unlabelled No. No. No. No. No. No. No. RAINFOREST Mt Nebo 6* 0 66 310 229 381 191 260 Mt Glorious 0 37 0 0 0 0 131 0 0 0 0 0 0 140 0 0 0 0 2 103 0 0 148 0 0 138 0 0 Mt Tamborine 0 0 0 De a 33 223 0 0 0 0 0 69 183 0 0 0 0 152 434 0 0 0 0 41 180 0 0 0 0 147 0 0 5 64 101 GARDEN Mt Coot-tha Site 1 0 0 0 0 13 174 205 0 0) 0 0 0 0 106 0 0 0 0 130 173 Site 2 0 173 0 0 0 223 137 0 12 0 0 0 0 0 0 0 0 6 44 Site 3 0 0 0 0 0 173 109 0 0 0 155 18 0 0 0 98 0 14 0 St Lucia 0 0 0 101 2 0 0 137 40 0 0 288 44 Total heads, No. 18 21 15 19 19 30 21 Total seeds, No. 0 228 0 2 18 2649 3117 Seed, Mean 0) 10.9 0 0.1 0.9 88.3 148.4 * Heads rotted, although a few seeds were set. Proc. Linn. Soc. N.S.W. 106 (4), (1982) 1983 DOROTHY E. SHAW ANDB. K. CANTRELL 329 Occasional encounters were Apis mellifera L. (Hymenoptera: Apidae), N. inversa, Phalacrus fimetarius (F'.) (Coleoptera: Phalacridae) and the unidentified weevil (Table 1). 3. Within the sealed spathal chamber near maturity The main species found within the sealed spathal chambers near maturity was N. inversa, as recorded in Table 5. It occurred, however, only at the rainforest sites, not in the gardens. A few occasional insects were also recorded, as listed in Table 2. Inflorescence treatments In all, 139 heads were involved in Treatments 1 to 6. Of these, 10 were lost through accident, stealth, etc., and seven heads in Treatment 3 were discarded because of perforation of the paper bags due to heavy rainfall and possibly pressure from the growing heads, giving an actual total of 122 heads. In addition, 21 heads in Treatment 7 were included in the results for comparison, making a total of 143 heads. The results are reported in Tables 3 to 6. Table 3 shows the number of heads maturing or rotting for each treatment. No difference in behaviour occurred between any site, locality or habitat, nor was any difference noted between heads bagged with paper bags compared with those with plastic bags. Some rotted heads occurred naturally, as shown in the labelled and unlabelled controls. Only those heads matured where seed had set except for four, which between them had only 12 berries with 15 seeds. Conversely, the only heads which rotted were those where no seed had set, apart from the four exceptions mentioned previously. The details of seed set are given in Table 4. The only heads which set seed were those of the controls, except for four unbagged, emasculated heads (one of which rotted) and one plastic-bagged and two mesh-bagged non-emasculated heads (two of which rotted). Table 5 shows the numbers of fertilized ovaries (berries) and fertilized ovules (seed) for all matured heads and the numbers (or absence) of N. imversa (mainly puparia) recorded in the sealed chambers of these heads. Seed set varied considerably at most sites, ranging from 1.5% (St Lucia) to 99.3% (Mt Nebo). The numbers of N. inversa per head ranged from three (Mt Tamborine) to 77 (Mt Glorious) with averages of 24.2, 26.5 and 17.5 respectively for the three rainforest localities (Mt Nebo, Mt Glorious and Mt Tamborine). There was no relation between numbers of N. znversa and the numbers of seed set per head, locality or habitat. No N. inversa were recorded in heads from the garden areas, but up to 88.3% of seed was still set. DISCUSSION Pollination ecology In A. macrorrhiza the pistils (megasporophylls) and stamens (microsporophylls) are present on different parts of the flowering head of the same plant. The exposure of the stigmas by gaping of the constricting areas at the unfurling of the spathal limb, presumably at the time of receptivity of the stigmatic surface, and the closing of the constricting portion of the spathe (therefore isolating the pistillate flowers) before the extrusion of the pollen, indicate protogyny. The precise period and peak of stigma receptivity, measured by the number and maximum of seeds set, need to be deter- mined by controlled crossing at different times during anthesis. The possibility of pollination by wind (anemophily) as against insect pollination (entomophily) is examined below. The pollen when first extruded in cirri above the surface of the staminate flowers (and in the best position for pick-up by wind currents) is at its stickiest and so less likely to be wind dispersed. Also, the extended, upright and slightly incurved spathal limb would act as a protector and deflector of wind from the rear of the head. After the Proc. LINN. Soc. N.S.W. 106 (4), (1982) 1983 330 POLLINATION OF ALOCASIA MACRORRHIZA TABLE 5 Numbers and percentages of fertilized ovaries (berries) and numbers of seed and N. inversa per head, locality and habitat for all matured heads HABITAT, Fertilized Seed per head N. inversa per head locality, ovaries head (berries) designation per head Locality Habitat Locality Habitat No. % No. Mean Mean No. Mean Mean RAINFOREST 156.9 21.4 Mt Nebo 239.5 24.2 6A.1 55 49.1 66 21 6A.2 131 89.1 229 26 7A.1 124229 191 46 7A.2 107 76.4 260 11 X.1 150 995.5 310 16 X.2 14) 99.3) 381 25 Mt Glorious 116.2 26.5 6A.1 87 = 558.8 148 11 6A.2 88 59.9 138 8 X.1 128 975.7 131 22 X.2 140) 9772 140 11 X.3 134 61.9 103 77 8B 262A 37 30 Mt Tamborine 136.0 17.5 1A.1 29 22.3 33 3 1A.4 46 =28.9 69 29 9A.1 117. 92.1 152 7 9A.2 S22 3ED 41 18 X.1 148) 728) 223 12 X.2 117 = 46.3 183 19 X.3 210 =98.1 434 31 X.4 85 61.6 180 6 X.5 96 49.0 147 29 X.6 5 28) 5 9 X.7 47 = 44.8 64 20 X.8 Ouro 101 27 GARDEN 107.3 0 Mt Coot-tha Site 1 133.5 0 1A 132 60.6 174 0 5A.1 80 63.5 130 0 5A.2 145 87.9 173 0 X.1 136 = 88.3 205 0 X.2 Uy 43.9) 106 0 1D 10 6.5 13 0 Site 2 99.2 0 1A 143 77.3 223 0 4A.1 6 5.3 6 0 4A.2 33 56.9 44 0 X.1 117 += 83.0 137 0 1B 155 74.9 173 0 4B 10 5.8 12 0 Site 3 94.5 0 1A.1 92220 173 0 1A.2 119 58.3 155 0 Proc. LINN. Soc. N.S.W. 106 (4), (1982) 1983 DOROTHY E. SHAW ANDB. K. CANTRELL 331 5A 10 7.6 14 0 ll 62 30.7 109 0 X.2 14 8.6 18 0 X.3 75 42.9 98 0 St Lucia 102.0 0 3A.1 7A 538.0 101 0 3A.2 IQ FO 137 0 3A.3 Nile 7: 288 0 3A.4 94 19.5 44 0 Yen 2 185 2 0 xe? Sela 40 0 TABLE 6 Theoretical expectancies of selfing vs crossing and wind vs insect pollination and the actual results Heads in the following classes Emasculated Non-emasculated Mechanism Paper- and Mesh- plastic- Mesh- bagged Unbagged bagged bagged Unbagged SELFED! no seed no seed seed seed seed CROSSED Wind Pollen reaching chamber of emasculated heads” seed seed no seed seed seed Pollen not reaching chamber ? of emasculated heads” no seed no seed no seed seed seed Insects Attractant still present in emasculated heads* no seed seed no seed no seed seed Attractant excised in emasculated heads* no seed no seed no seed no seed seed Tf insect crossed, attractant excised, no. of heads with: (Theoretical) Seed 0 0 0 0 30 No seed 18 21 34 19 0 (Actual) Seed 0 4 1 2 24 (2 seeds) (few seeds) No seed 18 17 33 17 6 ' Normally protogynous, but selfing is perhaps possible in absence of cross pollen or pollinator(s) if some residual stigma receptivity is still present at time of pollen extrusion, unless self-incompatible. 2 Cross pollen would need to find the small entrance to the spathal chamber. 3 Attractant (possibly fragrance) probably situated in the appendage/staminate portion of the spadix removed with emasculation. pollen has fallen, when it is less sticky, the grains are fairly well protected from wind pick-up, being within the overlap at the base of the spathal limb in front (in the pollen catchment area) and the spadix itself at the back. However, neither occurrence (that is, stickiness when extruded and protection within the base of the overlapping spathal limb) need necessarily completely exclude the possibility of dispersal of some pollen Proc. LINN. Soc. N.S.W. 106 (4), (1982) 1983 332 POLLINATION OF ALOCASIA MACRORRHIZA grains by wind. This may occur, as previously reported, by deliberately jarring the head. Such jarring could occur in nature by falling branches, accidental encounters by birds or ground animals and possibly by high winds. Faegri and Pijl (1979) have stated that an obvious condition for effective wind pollination is the presence of wind, as found in open, sparse vegetation or in the top layer of closed, multi-layered vegetation types, and that in dense forest vegetation wind is so slight and infrequent that anemophily is contra-indicated. At all the sites except Site 3 at Mt Coot-tha the plants under study were part of the understorey, and at most sites were well protected by topography (being in or on the banks of streams) and by lush gully vegetation. Provided pollen grains are fortuitously distributed, the total output of pollen produced, N, gives a number of ‘effective’ grains, i.e. grains carrying out pollination, n, which is a function of the areas of stigmatic surfaces, a, and the total area of the surroundings, A: n _ a (Faegri and Pijl, 1979). The production of inflorescences on Nig A any plant of A. macrorrhiza is sequential, so that when a head is in anthesis, the previous head is already in ‘pod’ (with the spathal chamber sealed) and the subsequent head is still in bud, with the spathe still furled. Therefore pollen, if wind-borne, would need to encounter heads on other plants at the right stage of stigma receptivity. Such heads could occur on plants in neighbouring clumps, though these are often scattered and sparse. Wind pollination is considered much less precise than biotic pollination, and to achieve its objective it presumes a very high incidence of pollen grains near their source, the dilution factor going by a power of 3 of the distance (Faegri and Pil, 1979). While considerable pollen of A. macrorrhiza is produced in any one head, the overall quantity in any one area would not be great, and also it is either sticky when at source or when older and less sticky is protected from easy wind pick-up in the pollen catch- ment area, as discussed previously. Even if wind-borne pollen were to encounter a recently unfurled head with receptive stigmas, the pollen has to reach those receptive surfaces. The chance of wind-borne pollen falling under gravity directly into an open chamber through the gaping constricting area would seem rather remote. Alter- natively, wind-borne grains could strike the inner surface of the extended upright spathal limb, if the limb happened to be at an appropriate angle to the line of flight, but again the grains would need to fall directly, or down the side of the spathal limb, onto a receptive surface. Even if a successful landing on a receptive stigma (pollination) was achieved, successful fertilization (germination on the stigma and fertilization of the ovules) may not necessarily follow, because the less sticky the pollen and therefore more readily available for wind dispersal, the older and less viable it is, and therefore less capable of achieving germination and fertilization. Faegri and Pil (1979) pointed out that one of the factors of major importance in anemophiles is the buoyancy of the pollen grains, which increases as the size of the grain decreases, and considered that pollen of anemophiles belongs to the smaller size classes with a ‘typical’ diameter of 20-30-(60) um, even if equally small or even smaller pollen grains are found in many entomophiles. Contributing to the effectivity of pollen transport in anemophiles and to the buoyancy of the pollen is the fact that grains do not adhere to each other, but are smooth and dry and are spread separately or in very small groups. Entomophilous pollen, on the other hand, is ornamented and sticky and the grains stick together (Faegri and Pijl, 1979). Thus the size of A. macrorrhiza pollen (mean 41.2 x 39.3 wm exclusive of ornamentation) measured in this study is in the larger part of the range for an anemophile, and their ornamentation of spinules and their stickiness suggest an entomophile. In A. macrorrhiza therefore, the extrusion of sticky pollen in cirri, the continued Proc. LINN. Soc. N.S.W. 106 (4), (1982) 1983 DOROTHY E. SHAW ANDB. K. CANTRELL 333 though decreased stickiness of the pollen with age, coupled with decrease in viability, the occurrence of spinules on the pollen grains and the short stigmatic hairs all suggest pollination by insects (entomophily) rather than wind pollination (anemophily). In addition, the occurrence of the fragrance at anthesis may also indicate entomophily. Insects N. inversa was present in all open chambers (Table 1) examined from the rainforest localities, usually moving on the walls of the chamber, seldom near the stigmas. Flies were seldom encountered on the spadix, either before or after closure of the spathal constriction, and were much more often seen running over the spathal limb. On three occasions copulating pairs were noted, once within an open chamber and twice on the inside face of the spathal limb. Brachypeplus sp. and the aleocharine staphylinid were usually moving over the ovaries below the stigmas in open chambers. No beetles, when examined microscopically, had pollen adhering, but this could have been dislodged during capture. T. carbonaria collected from the open chambers were from the stigmas, often with heavy pollen loads in their corbiculae. Macroscopically the corbicular pellets appeared homogeneous in colour and texture. Microscopical examination of samples from six pellets chosen at random showed that they were entirely composed of A. macrorrhiza pollen, except for four foreign grains from one sample. 7. carbonaria was therefore mainly ‘A. macrorrhiza constant’ and must be considered to be a possible pollinating agent since they were also often observed collecting pollen on heads with sealed chambers. Of the insects occasionally recorded within the open chambers (Table 1) the thrips were present in large numbers on the spadix of such inflorescences at all localities and the presence of some in the chamber would be expected. The records of the remaining species cannot be easily explained but is probably due to chance occurrence. In any case, their rarity discounts any possibility of their being effective pollinators. Considering the insects on the exposed spadix after closure of the spathal con- striction (Table 1), 7. carbonaria were observed visiting and collecting pollen at most sites, and their absence at others is probably due to lack of colonies in the area. Brachypeplus sp. and the staphylinid were always present, often immersed in the ac- cumulated pollen of the spathal catchment area above the spathal constriction. However, no beetles were ever observed arriving or leaving and experiments on their vagility are necessary to determine their ability as pollinators. The small size of the thrips makes them unlikely pollinators despite their common occurrence, and like the beetles, their vagility is unknown. Normally ubiquitous, the honeybee was only observed visiting on a single oc- casion, despite their presence on other flowers at all localities. The only other insects encountered apart from N. inversa (already mentioned) (Table 1) were the phalacrid and the unidentified weevil, both single individuals. The numbers of N. inversa recorded in the sealed chambers (Table 5) confirms the report of McAlpine (1978) and Shaw et al. (1982) that they occurred only in rainforest. (It should be noted, however, that two adult N. inversa were found in two open chambers at two of the Mt Coot-tha sites (Table 1). This may indicate that the fly is now finding its way to this area, which has only been established for four years.) The few occasional insects recorded in the sealed chambers (Table 2) show a similar basic composition, though less extensive, to those recorded by Shaw et al. (1982, table 5). In his studies on the pollination of A. macrorrhiza in the Illawarra District near Sydney, Hamilton (1898) recorded the beetle, Brachypeplus murray: in the open chambers. He also made some observations not substantiated by us, viz., beetles feeding on liquid exuded by the pistils and beetles, becoming confined in sealed spathal Proc. LINN. Soc. N.S.W. 106 (4), (1982) 1983 334 POLLINATION OF ALOCASIA MACRORRHIZA chambers, later burrowing their way out. We noted no such feeding habit, nor any perforated chambers. Hamilton (1898) also observed an unidentified species of agaristid (Lepidoptera: Agaristidae) in open chambers and claimed that they aided in fertilization. Larvae of the agaristid Crurta donovani (Boisduval) were noted at all rainforest localities during our study, where they were often seen inside buds and open chambers feeding on stigmas and ovaries. However, as a single head contains sufficient tissue to allow a larva to reach maturity they are unlikely to be effective pollinators. If indeed A. macrorrhiza is insect pollinated as suggested by this study, then Brachypeplus sp. and the staphylinid would appear to be the most likely candidates since they occur at all localities examined. While N. inversa and T. carbonaria might appear to be more likely to be effective in this role, their absence from some localities where good seed set occurred does not support this, or at least suggests that a number of pollinating agents are involved, and that a resolution requires further investigations. Emasculation/bagging experiment The results of the emasculation/bagging experiment given in Table 3 show that all heads rotted if no seed was set (except four with a few seeds only) and conversely, the only heads which matured were those where seed had set (other than the four ex- ceptions). The results in Tables 3-6 show that, as seed occurred (other than in seven heads) only in unbagged, non-emasculated heads (Treatments 6 and 7), A. macrorrhiza is not apomictic, because if it were, seed could have occurred in all heads, unless pseudogamic. Mesh-bagging allowed seed set (18 seeds) in only two heads out of 36 (Table 4) and therefore wind pollination is probably unlikely as a general occurrence. The 18 seeds (set in 12 berries) may have resulted from chance wind-borne or self pollen. Emasculation resulted in seed set (228 seeds) in four out of 39 heads, compared with 2669 seeds in 27 out of 83 non-emasculated heads including labelled controls (Table 4). The four emasculated heads which had some seed set were unbagged and could have been fertilized by wind- or insect-borne pollen. In the non-emasculated heads, only two seeds were set in one out of 34 heads in any imperviously-bagged head (paper or plastic) and only 18 seeds in two heads out of 19 in the mesh-bagged heads (Table 4); the two seeds may have been the result of self pollination while the 18 seeds may have resulted from self or chance wind pollination. Twenty-four of the unbagged, non-emasculated (labelled) controls (Treatment 6) set 2649 seeds against six heads which did not (Table 4); the non-seed set in these heads may have been due to lack of pollen at the time of stigma receptivity and/or lack of pollinating agent(s). The results shown in Table 5 confirm data previously reported by Shaw et al. (1982) viz., that there is great variation in the numbers of seed set per head (in the present study from 1.5% to 99.3% in the 48 heads, all controls except two); that there is no relation between numbers of seed set and numbers of N. inversa recorded in the sealed chambers (which it is presumed reflects the numbers of ovipositing females in the open chambers) at the rainforest sites; that no N. znversa were recorded in the sealed chambers at the garden sites but despite this seed up to 88.3% was still set in garden site heads. Table 6 shows the theoretically expected seed set figures deduced from the treatments applied in our experiment. The actual results, also shown in Table 6, most nearly approximate those of crossing with insect pollination with the attractant excised in the emasculated heads. Small differences from the theoretical results occurred in the emasculated unbagged heads, where four heads set 228 seeds (possibly insect or chance wind-borne pollination); in the imperviously-bagged non-emasculated heads, where Proc. LINN. Soc. N.S.W. 106 (4), (1982) 1983 DOROTHY E. SHAW ANDB. K. CANTRELL 335 two seed set in one head (probably self pollen); and in the mesh-bagged non- emasculated heads, where 18 seeds set in two heads (possibly self or wind-borne pollen), whereas no seed was expected in any of these classes. Also, all control (labelled) heads were expected to set seed but six heads out of 24 did not, a situation also found in the unlabelled controls. The results of the experiment therefore appear to indicate cross pollination by insects, with little attraction for the insects if the appendage/staminate portion of the spadix is removed. ACKNOWLEDGEMENTS The owners of the sites sampled and other individuals who assisted in obtaining the collections, including Mrs J. and Miss L. Cantrell, Mr A. Hiller, Mrs J. Hope, Ms K. Howdesell, Mr and Mrs L. J. Manning, Mr A. L. Marstella, Mr and Mrs W. Scattini and also Mr H. Caulfield (Mt Coot-tha Botanic Gardens) and Dr A. B. Cribb and Mr A. R. Steginga (University of Queensland) are thanked for their co-operation in this study. We are also grateful to Mr T. A. Weir and Dr E. C. Zimmerman, both of the Commonwealth Scientific and Industrial Research Organization, Canberra, and Dr I. D. Galloway and Mr K. J. Houston, both Department of Primary Industries, Brisbane, for identification of insects, and to the Director, Plant Pathology Branch, of the same Department, Indooroopilly, for facilities to D.E.S. during the study. References BREWBAKER, J. L., and KWAck, B. H., 1963. — The essential role of calcium ion in pollen germination and pollen tube growth. Amer. J. Bot. 50: 859-865. FAEGRI, K., and PYL, L. van der, 1979. — The principles of pollination ecology. Oxford: Pergamon Press. HAMILTON, A. G., 1898. — On the methods of fertilization of some Australian plants. Report of the Seventh Meeting of the Australasian Association for the Advancement of Science, held at Sydney, 4898. Published by the Association. (pp. 557-565). KORNERUP, A., and WANSCHER, J. H., 1961. — Methuen handbook of colour. London: Methuen & Co. Ltd. MCALPINE, D. K., 1978. — Description and biology of a new genus of flies related to Anthoclusza and representing a new family (Diptera, Schizophora, Neurochaetidae). Ann. Natal Mus. 32: 273-295. SHAW, D. E., CANTRELL, B. K., and Houston, K. J., 1982. — Neurochaeta inversa McAlpine (Diptera: Neurochaetidae) and seed set in Alocasia macrorrhiza (L.) G. Don (Araceae) in southeast Queensland. Proc. Linn. Soc. N.S.W. 106: 67-82. STANLEY, R. G., and LINSKENS, H. F., 1974. — Pollen: biology, biochemistry, management. Berlin, New York: Springer-Verlag. Proc. LINN. Soc. N.S.W. 106 (4), (1982) 1983 Rabbits, Vegetation and Erosion on Macquarie Island P. M. SELKIRK, A. B. COSTIN, R. D. SEPPELT and J. J. SCOTT SELKIRK, P. M., Costin, A. B., SEPPELT, R. D., & ScoTT, J. J. Rabbits, vegetation and erosion on Macquarie Island. Proc. Linn. Soc. N.S.W. 106 (4), (1982) 1983: 337-346. Rabbits were introduced to subantarctic Macquarie Island in about 1880. Selective grazing by rabbits has been important in changing the floristic composition and structure of grassland and herbfield vegetation, and the effects of grazing and burrowing activities on erosion have concerned scientific visitors to the island for some years. Re-examination in 1980 of sites first documented in 1958 shows increases in erosion at some sites, and revegetation of some areas. Photographic documentation is provided for both sets of observations, and changes noted in the 1980 photographs are discussed. The role of rabbits in erosion of grassland sites is difficult to assess, and may in the past have been overestimated. P. M. Selkirk, School of Biological Sciences, Macquarie University, North Ryde, Australia 2113; A. B. Costin, Centre for Resource and Environmental Studies, Australian National University, Canberra, Australia 2600; R. D. Seppelt, Antarctic Division, Department of Science and Technology, Channel Highway, Kingston, Australia 7150, and J. J. Scott, Geography Depart- ment, Monash University, Clayton, Australa 3168; manuscript recewed 6 July 1982, accepted for publication in revised form 20 October 1982. INTRODUCTION Macquarie Island (54°30'S, 158°57'E) is a small subantarctic island with a moist, cool, windy, oceanic climate. The vegetation includes tussock grassland, herbfield, fen and bog communities, and feldmark (Taylor, 1955). Since their first introduction in 1879 or 1880 (Cumpston, 1968), rabbits have spread over most of the island and their numbers have increased dramatically. Their population was estimated at 50 000 in 1956, 150 000 in 1965-6 (Sobey e¢ al., 1973), 50 000 in 1974 (Skira, 1979) and up to 150 000 in 1977-78 (Copson e¢ al., 1981). The introduction of rabbits appears to have affected a number of components of the en- vironment. They form the major item of food for feral cats on the island (Jones, 1977). Taylor (1979) described the role of rabbits in the demise of the Macquarie Island parakeet. The effect of rabbit grazing on the floristic composition and structure of tussock grassland and herbfield was described by Taylor (1955), and further discussed by Costin and Moore (1960), Johnston (1966) and Jenkin ef a/. (1981). The effects of rabbit grazing and burrowing on erosion have concerned scientific visitors to the island for some years, and have been the subject of several reports. Taylor (1955) and Costin and Moore (1960) concluded that selective grazing by rabbits of the stabilizing species in grassland, aggravated by burrowing, has substantially accelerated erosion of steep slopes. Johnston (1966) and Griffin (1980) concluded that rabbit-induced erosion is of minor importance in the overall erosion picture on Macquarie Island. Griffin (1980) considered that soil creep is the major erosion mechanism on the island, and that rabbit grazing has had little influence on this. COMPARISON OF PHOTOGRAPHS Eight sites, initially photographed in connection with a study of the role of rabbits in erosion in December 1958 (Costin and Moore, 1960) were revisited and rephotographed during January and February 1980. The sites show some changes over Proc. LINN. Soc. N.S.W. 106 (4), (1982) 1983 338 RABBITS, VEGETATION AND EROSION ON MACQUARIE ISLAND the 21-year period between the two sets of photographs. There have been some in- creases in erosion (deepening of gullies, increased stone exposure in bare peat areas) and revegetation of some areas. Comparison of the photographs and a discussion of changes in the sites follows. Vascular plant nomenclature follows that of Greene and Walton (1975) except for Luzula crinita which follows Edgar (1966). Bryophyte nomenclature follows Seppelt (1980). Figure 1: These steep slopes on which Poa foliosa tussock occurs appear unchanged in 1980 compared with 1958. Immediately above the limit of the photograph, however, rabbit grazing is evident at the top of the steep grassland slopes having spread downwards from the gentler herbfield slopes above. Figure 2: The area of bare peat in right foreground has enlarged by 1980, compared with 1958. The Poa foliosa tussocks on the knoll are stunted. The area of white (dead) P. foltosa in the 1958 photo is, in 1980, occupied by a Luzula crinita — Festuca contracta short tussock vegetation amongst which shoots and small clumps of P. foliosa are growing quite prolifically. Present vegetation on the knoll includes Acaena minor, Azorella selago, Colobanthus muscoides, Epilobium linnaeoides, Festuca contracta, Luzula crinita, Poa foliosa, Ranunculus biternatus, Breutelia pendula, Diucranoloma billardiert, Ditrichum strictum, Rhacomitrium crispulum. Figure 3: The dark area of stable tussock (Poa foliosa) in the 1958 photo appears un- changed in 1980. The white areas are now grassland with Agrostis magellanica — Luzula crinitta dominated vegetation forming more or less continuous cover. Both the fresh ‘peat flow’ landslip area (p) and the tumbled area (t) below it in the 1958 photograph have, in 1980, a complete plant cover. The present vegetation in both ‘peat flow’ and ‘tumbled’ areas is similar (unpublished quadrat data, J. J. Scott), including: Acaena minor, Agrostis magellanica, Epilobium linnaeoides, Luzula crinita, Poa annua, Ranunculus biternatus, Breutelia pendula, Drepanocladus sp., Marchantia berteroana, Thuidium furfurosum. To the west of the peat flow area are a number of old tussock bases, (b) presumably rabbit grazed, some now overgrown with Luzula crinita and Acaena minor. There is abundant evidence of rabbits in the area of photograph 3: rabbits, burrows, squats and grazed tussock are all to be seen. In the lower left-hand corner of the 1980 photo is an area of landslip which has occurred since 1958. The ridge (r) between the old and the new slip areas is covered with ‘terracettes’ the origin of which is ascribed by Fig. 1. From a few metres southeast of Hurd Point hut, looking to west. In ail figures, photographs (A) taken in 1958 by A. B. Costin, previously published in Costin and Moore (1960), photographs (B) taken in 1980 by P. M. Selkirk. Letters on photographs are discussed in text. Fig. 2. From about 150 m altitude on west side of scree-slope track above Hurd Point hut, looking down onto knoll to west of scree slope. Fig. 3. North of Windsor Bay, south of Hurd Point-Caroline Cove track, taken from base of rocky outcrop to north of track. Fig. 4. Above Hurd Point, west of scree-slope track, on slope above hill in Fig. 2. Fig. 5. Above Hurd Point, east of scree-slope track, on plateau edge. Photograph B taken from closer to background ridge than photograph A. Fig. 6. On Hurd Point-Caroline Cove track, below rocky knoll from which Figs 3 and 7 were taken. Fig. 7.* East of Petrel Peak, north of Windsor Bay, south of Fig. 6, immediately to east of Fig. 3. Fig. 8.* In Jessie Niccol Creek valley, from point at which Overland Track crosses south arm of Jessie Niccol Creek, looking northwest toward Mt. Blake. * Note that in Costin and Moore (1960) captions for photographs 7 and 8 were reversed. Proc. LINN. Soc. N.S.W. 106 (4), (1982) 1983 339 P.M. SELKIRK, A. B. COSTIN, R. D. SEPPELT AND J. J. SCOTT Proc. LINN. Soc. N.S.W. 106 (4), (1982) 1983 RABBITS, VEGETATION AND EROSION ON MACQUARIE ISLAND 340 Proc. LINN. SOc. N.S.W. 106 (4), (1982) 1983 P.M. SELKIRK, A. B. COSTIN, R. D. SEPPELT AND J. J. SCOTT Proc. LINN. Soc. N.S.W. 106 (4), (1982) 1983 RABBITS, VEGETATION AND EROSION ON MACQUARIE ISLAND 342 PROC. LINN. SOc. N.S.W. 106 (4), (1982) 1983 P.M. SELKIRK, A. B. COSTIN, R. D. SEPPELT AND J. J. SCOTT 343 Griffin (1980) to creep, solifluction and/or small scale slumping processes. We believe the terracettes are initiated by small-scale slumping, then may be aggravated by rabbits using them for squats and territorial digs, and the whole ultimately overgrown by vegetation (J. J. Scott, unpublished data). Figure 4: The tongue of vegetation on the far side of the bare peat area in the 1958 photograph is, in 1980, an elongated island. The bare peat area has enlarged slightly, with a greater area of bare stones, and appears to be forming into a stony erosion pavement which should offer some protection from frost-heave, wind- and water- erosion to the remaining peat (Wimbush and Costin, 1979; Totterdell and Nebauer, 1973) allowing its colonization by mosses and angiosperms. Figure 5: The areas bare of vegetation on the background ridge are similar in area in the two photographs although perhaps more stones are exposed in 1980. The 1958 photograph caption describes rabbit burrowing undercutting dead tussocks above the soil slip in the left background, but by 1980 there is no evidence of tussock grass having been there very recently. Rabbits were still active in the area in 1980. On the background slope (25°), present vegetation includes: Acaena magellanica, Agrostis magellanica, Azorella selago, Cardamine corymbosa, Epilobium nerterioides, Luzula crinita, Ranunculus biternatus, Stellaria decipiens, Breutelia pendula, Drepanocladus sp., Psilopilum australe, Rhacomitrium crispulum. The present vegetation on the gentle slope (5°-15°) of the foreground ridge, much less prone to erosion than the steeper ridge behind, includes: Acaena minor, Cardamine corymbosa, Epilobium nertertoides, Luzula crinita, Ranunculus biternatus. Figure 6: The total area of bare sites has not altered substantially between 1958 and 1980, but some changes are visible. In the centre of the 1980 photograph a larger area of stony pavement is visible than in 1958, exposed presumably by wind- and water- erosion of surrounding peat particles. There is in 1980 a more distinct drainage line (d) connecting the two non-vegetated areas. Figure 7: In the bare areas on the upper slopes, gullies have deepened and some ‘islands’ of vegetation (i) have disappeared since 1958, presumably by the action of frost, wind and water. Lower down the slope, gullies have deepened, exposing a greater area of bare soil along the gully sides. Other areas, bare in 1958, show some revegetation (v) in 1980. Figure 8: The creek gully in 1980 is somewhat infilled at its junction with the small tributary from the northwest, compared with the 1958 photograph. The infill material has presumably been brought down by the tributary creek. Since its deposition it has been densely covered with bryophytes. Upstream of the junction with the tributary, however, the creek gully still has quite steep sides as in 1958. GENERAL DISCUSSION In trying to assess the contribution of rabbits to soil erosion on Macquarie Island, a number of factors must be considered: effect of rabbit activities (burrowing, squats, territorial digs, grazing), size of rabbit population, extent of rabbit activities, and other factors contributing to surface instability. Undercutting of tussocks by rabbit burrowing was described by Costin and Moore Proc. LINN. Soc. N.S.W. 106 (4), (1982) 1983 344 RABBITS, VEGETATION AND EROSION ON MACQUARIE ISLAND (1960). The suggested role of rabbit squat and territorial digging activities in ac- centuating terracettes has been mentioned above. It is by their selective grazing that rabbits exert most effect on the Macquarie Island environment. Areas may be com- pletely denuded of their vegetation when this was composed entirely of palatable species. These areas may be subject to erosion until colonized by less palatable and/or grazing-adapted species. An example is a steep slope above the beach just south of Nuggets Point beach, previously clothed with Stilbocarpa polaris and a little Poa foliosa on which rabbit grazing was severe during 1978-9 (G. Copson, pers. comm.). The area was bared when the Stlbocarpa was grazed down to its rhizomes. In 1980, rapid colonization of the soil surface was being achieved principally by Marchantia berteroana (via its abundantly-produced gemmae), but also by Poa annua, Epilobium linnaeoides, E. nertertordes and Cardamine corymbosa. Marchantia berteroana seemed only rarely to be touched by the rabbits which were still present in the area. Erosion of the surface appeared minimal. The recolonization of areas bared by rabbit grazing appears similar in general terms to the recolonization of areas bared of vegetation by landslips, but with some important differences ( J. J. Scott, unpublished data). These arise because landslips commonly remove all vegetation plus a peat layer, while selective grazing by rabbits usually leaves some (possibly less palatable) vegetation on the site. For example, grazing on the coastal slopes usually removes the dominant species Poa foliosa and Stilbocarpa polaris, allowing remaining subsidiary species such as Ranunculus biternatus, Acaena minor and Agrostis magellanica to achieve local dominance, in similar fashion to the infilling of gaps between senescing Poa foliosa tussocks (Ashton, 1965). A landslip removes both dominant and subsidiary species, as well as surface litter, seeds and underground rhizomes in a peat layer of varying thickness. Revegetation proceeds much more slowly than on areas bared by grazing. In the absence of further grazing once an area has been bared, both landslip surface and areas grazed by rabbits can eventually be revegetated by their former dominant species. However, if regeneration of Poa foliosa and Stilbocarpa polaris is suppressed by continued grazing of landslip- or rabbit-bared surfaces, a short tussock grassland characterized by Agrostis magellanica, Luzula crinita and Acaena spp. develops. In areas whose dominant vegetation includes a mixture of palatable and un- palatable species, selective grazing by rabbits can have a very marked effect on species composition. For example, the rabbit population of the Flat Creek valley, near Bauer Bay, was dramatically reduced by myxomatosis during winter 1979. The quantities of Pleurophyllum hookeri, Stellaria media and Cerastium fontanum, all preferentially eaten by rabbits, were observed to increase noticeably in this valley between the beginning and the end of the 1979-80 summer. The floristic composition of the short tussock grasslands and herbfields has been particularly affected by rabbit grazing. Pleurophyllum hookert now seems less widespread in the herbfields than Taylor’s (1955) photographs and descriptions would suggest. The highly palatable Pleurophyllum is badly damaged when rabbits graze the leaves right down to the stem stump. Species which are now more abundant in the herbfield than Taylor’s (1955) descriptions suggest may be less palatable to rabbits (e.g. Luzula crinita) or able to withstand and thrive despite grazing, by virtue of abundant lateral shoot growth (e.g. Acaena spp.), tillering (Poa annua, Agrostis magellanica) or seed set (Acaena spp., Poa annua, Agrostis magellanica, Uncinia compacta). Copson et al. (1981) note that the majority of rabbits live in the herbfield vegetation, including the short tussock grasslands mentioned above. Few rabbits live in the tussock grassland, bog, fen and feldmark formations as described by Taylor (1955), although these vegetation types are grazed by rabbits from nearby herbfield areas (Copson et al., 1981). Proc. LINN. Soc. N.S.W. 106 (4), (1982) 1983 P.M. SELKIRK, A. B. COSTIN, R. D. SEPPELT AND J. J. SCOTT 345 Early descriptions (Cumpston, 1968) and recent estimates (Sobey et al., 1973; Skira, 1979; Copson et al., 1981) of rabbit numbers on the island are consistent with the suggestion that, in common with introduced species elsewhere, the rabbit population on Macquarie Island was low for a long period after its introduction, went through a period of rapid expansion to a maximum, perhaps beyond the capacity of the habitat to support the numbers, then fell again to a lower level, in balance with the capacity of the habitat to support it. Within this general picture, numbers appear to have fluctuated: for example, a peak between 1880 and 1906 is discussed by Taylor (1979) using data from Cumpston (1968), and a peak occurred in the 1977-78 summer due to several successful breeding seasons preceding it (Copson et al., 1981). It is suggested by Copson é al. that annual population fluctuations are a result of seasonal changes in weather, with extended drier periods leading to higher survival rate of kittens. Dense rabbit populations have occurred in different places on the island at dif- ferent times (maps, Sobey é al., 1973). Detailed rabbit counts have been made in recent years by officers of the Tasmanian National Parks and Wildlife Service but the size of the population at any time is not easy to estimate. The times of peak populations in the past may never be known, but it may be assumed that maximal effect by rabbits on an environment would correspond with peak populations in the area. Such an area would be especially vulnerable to accelerated erosion, if there were a lag between the initial damage to the taller-growing dominant plants (Poa foliosa, Stilbocarpa polaris) and the subsequent succession of lower-growing herbs and mosses, as in Figs 2, 3 and 5. The observations of Costin and Moore (1960) may have been made soon after a period of maximal rabbit population in the Hurd Point area, and consequently maximal deleterious effect on the local environment. Although rabbit activities may have provided a trigger for more than usual peat instability, it is clear that the predicted extension of damaged areas (Costin and Moore, 1960) is not occurring, except locally. Griffin (1980) discussed a number of erosive forces active on Macquarie Island, namely soil creep, landslips triggered both by unusually heavy rain periods and by earthquakes, wind erosion and rabbit activities. In addition, royal penguin, gentoo penguin and giant petrel activity on some of the lower coastal slopes contributes locally to the erosion of peaty soil between Poa foliosa tussocks, thereby affecting the balance and stability of the tussocks. After prolonged rain, slipping of these unstable situations may occur. In the presence of these other erosive forces on Macquarie Island, the role of rabbits in initiating and contributing to continued erosion of grassland areas is difficult to assess, and may in the past have been overestimated. Through their selective grazing, however, rabbits have had major effects in changing the floristic composition and structure of much of the original grassland and herbfield vegetation. ACKNOWLEDGEMENTS Permission from Tasmanian National Parks and Wildlife Service, and logistic support from Australian Antarctic Division for visits to Macquarie Island by the authors are gratefully acknowledged. Thanks are due to Mr G. Copson for help in location of sites, to Dr J. Jenkin for valuable comments on this paper, to Dr N. Wace and Dr B. Fennessey for helpful discussions, and to the editors of the Journal of Ecology for permission to reproduce photographs originally published in that journal. Proc. LINN. Soc. N.S.W. 106 (4), (1982) 1983 346 RABBITS, VEGETATION AND EROSION ON MACQUARIE ISLAND References ASHTON, D. H., 1965 — Regeneration pattern of Poa foliosa Hook F. on Macquarie Island. Proc. R. Soc. Vict. 79:215-233. Corson, G. R., BROTHERS,, N. P., and SkiIRA, I. J., 1981 — Distribution and abundance of the rabbit, Oryctolagus cuniculus (L.), at subantarctic Macquarie Island. Aust. Wildl. Res. 8:597-611. CosTIN, A. B., and Moore, D. M., 1960 — The effects of rabbit grazing on the grasslands of Macquarie Island. J. Ecol. 48: 729-732. CUuMPSTON, J. S., 1968 — Macquarie Island. ANARE Scientific Reports, Series A(1), Narrative. Antarctic Division, Melbourne: 1-380. EpGaR, E., 1966 — Luzula in New Zealand. N.Z. J. Bot. 4: 159-184. GREENE, S. W., and WALTON, D. W. H., 1975 — An annotated checklist of the Subantarctic vascular flora. Polar Record 17: 473-484. GRIFFIN, B. J., 1980 — Erosion and rabbits on Macquarie Island: some comments. Pap. Proc. R. Soc. Tasm. 114: 81-83. JENKIN, J. F., JOHNSTONE, G. W., and Copson, G. R., 1981 — Introduced animal and plant species on Macquarie Island. Colloque sur les Ecosystémes Subantarctiques, Paimpont, C.N.F.R.A. 51: 301-313. JOHNSTON, G. C., 1966 — Macquarie Island and its rabbits. Tasm. J. Agric. 37: 277-280. JONES, E., 1977 — Ecology of the feral cats Felis catus (L.), (Carnivora: Felidae) on Macquarie Island. Aust. Wildl. Res. 4:249-262. SEPPELT, R. D., 1980 — A synoptic moss flora of Macquarie Island. Antarctic Division Technical Memorandum Number 93, 8 pp. SKIRA, I. J., 1979 — Studies of the rabbit population on Macquarie Island. Hobart: University of Tasmania, M.Sc. thesis, unpubl. SOBEY, W. R., ADAMS, K. M., JOHNSTON, G. C., GOULD, L. R., Simpson, K. N. G., and KEITH, K., 1973 — Macquarie Island: the introduction of the European rabbit flea Spzlopsyllus cuniculi (Dale) as a possible vector for myxomatosis. J. Hyg., Camb. 71: 299-308. TAYLOR, B. W., 1955 — The flora, vegetation and soils of Macquarie Island. ANARE Reports, Series B, Vol. II, pp. 190-192. TAYLOR, R. H., 1979 — How the Macquarie Island parakeet became extinct. N.Z. J. Ecol. 2: 42-45. TOTTERDELL, C. J., and NEBAUER, N. R., 1973 — Colour aerial photography in the reappraisal of alpine soil erosion. J. Soil Conserv. Serv. N.S.W. 29: 130-158. WIMBUSH, D. J., and CosTIN, A. B., 1979 — Trends in vegetation at Kosciusko. III. Alpine range transects 1959-1978. Aust. J. Bot. 27: 833-871. Proc. LINN. Soc. N.S.W. 106 (4), (1982) 1983 Index VOLUME 106 Page Alocasia macrorrhiza, Crustaceans from Australian mangroves... . . 112 pollamationlot; fs Suk ha de sls tka ee 323. Cupala Creek Formation (Cambrian)....... 129 SECCbse tpl peusra tachi: oaAlens Mh. stot ule 67 Amaurobioidea, (Araneae) ........... 247, 275 Dapto Latite Member, of Broughton For- Anderson, D. T., Origins and relationships ration (HEAVEN). o5cceces¢eeoccone. 292 among the animal phyla .............. 151 Darriwilian (middle Ordovician) graptolites AUDIOS TOS co. 028 ROS ao. Hi Fegeo aenueat hE ER ee ME 142 from the Monaro Trough sequence east of ?Aphelaspis sp. aff. Braidwood, New South Wales, by C. J. Al, GUSTS Ew fonts oe ae ee a ee 142 Henkin sie dihesye ts alee etre Ria eae 173 Araceae (Alocasia macrorrhiza) ........... ORS 2 3) eee Diptera Neunochachica cli lei aeer 67 Araneae, Amaurobioidea............. UE Pea Oipterawilele omilyzidac ein ie ee 33, 59 Archaeobalanidae, (Balanoidea)........... 22 AOMTISED GSDa WOonc 6 cae ea en dea ae Hoe 275 Ecological values of the tropical rainforest Austmusia kiolbasp.nov. ............-.-.-. 281 resource, Macleay Memorial Lecture, by Austmusia lindi sp.nov. ...........+.----. 283 IE Wich breast aes Sere atic eee 263 Austmusia wilsonisp.nov. ................ OE Eiliminvusicovertusts DOV Ae eee 24 A. W. H. Humphrey, His Majesty’s Hilminausynodes tus apie ee eee 23 mineralogist in New South Wales, 1803-12 Elmside Formation (Siluro-Devonian) ...... 167 —Acomment,byR.J.Ford.......... 123 Eucalyptus bugaldiensis sp. nov.............. 301 Eucalyptus spp. (middle Miocene) .......... 305 PRC LUUITNT Oleaeg Ne opp ea ae Bie ay Ric tea visas apse sah 249 Fauna of Australian mangroves, by P. A. Badumna candidacomb. nov. .............. 250 Hutchings & H.F. Recher ............ 83 Badumna gausapatacomb. nov.............. 254 Fauna of Australian seagrass beds, by P. A. Badumna vandiemeni sp. nov................ 257 Flutchinigss seh hee oer oes cence 181 Barnaclesmioalanoidyes se so 4 oe sae center 21 Five Islands Latite Member, of Pheasants Nest Berkley Latite Member, of Pheasants Nest - Kormation| (Permian) se eee eee 293 Formation (Permian) ................ 295. Ford, R. J., A. W. H. Humphrey, His [ETI RECALLS Ovaen cet eR Gt Siri. paren Ses er oa Ac 147 Majesty’s mineralogist in New South Blow Hole Latite Member, of Broughton Wales, 1803-12 — Acomment ......... 123 Formation (Permian) ................ 289s Fossil Eucalyptus remains from the middle Brachiopods, Cambrian................. 147 Miocene Chalk Mountain Formation, Broughton Formation, of Shoalhaven Groap Warrumbungle Mountains, New South (Renmian)) beaters. cen ns, getters. 28. 289 Wales, by W. B. K. Holmes, F. M. Bugaldie, N.S.W., fossil Eucalyptus from..... 299 Folmme sya lel ay Ate Van; Gan 299 Bumbo Latite Member, of Broughton For- Foster, B. A., Two new intertidal balanoid mation((Permian)),.3.°.>.-..-...- 2 291 barnacles from eastern Australia........ 21 Calderwood Latite Member (new name), of Graptolites 25) as28 ee ee e 167, 173 Pheasants Nest Formation (Permian) .... 294 Gray, M. R., A new genus of spiders of the Cambewarra Latite Member, of Broughton subfamily Metaltellinae (Araneae, Amauro- Formation (Permian) ................ 292 bioidea) from southeastern Australia. .... 275 Cambrian (Idamean) rocks and fossils, north- Gray, M. R., The taxonomy of the semi- WWE SLOTIMVINS AV cee arues cog eo tas ai tsiad apal Geset ce oes = 136. communal spiders commonly referred to Cantrell, B. K., see Shaw, D. E. the species Jxeuticus candidus (L. Koch) with Carr, P. F., A reappraisal of the stratigraphy notes on the genera Phryganoporus, Ixeuticus of the upper Shoalhaven Group and lower and Badumna (Araneae, Amaurobioidea).. 247 Illawarra Coal Measures, southern Sydney Basin, New South Wales.............. 23 tleleomyzidae.y (Diptera) ys. eee 33599 Chalk Mountain Formation (middle Miocene) IE CERTTE TOS (GEN INO 20 oo gy 0002 2o58 5555 28 TENNPINEN IDSA o (oh aes Ubou alla suse eos ence aleenes eet OND 300 Hexaminius popeianasp.nov. ...........-.- 28 Comparative morphological study of the Holmes, F. M., see Holmes, W. B. K. reproductive systems of some species of Holmes, W. B. K., Holmes, F. M., & Martin, Tapeigaster Macquart (Diptera: Heleo- H. A., Fossil Eucalyptus remains from the myzidae), by M. A. Schneider.......... 59 middle Miocene Chalk Mountain For- Costin, A. B., see Selkirk, P. M. mation, Warrumbungle Mountains, New Crane, D., see Powell, C. McA. SouthWralesiii sinus 6.0 eres Ase 299 Proc. LINN. Soc. N.S.W. 106 (4), (1982) 1983 348 Houston, K. J., see Shaw, D. E. Jelnnanyoores7, AN. Wo lal, oc sccooncccocunde: Hutchings, P. A., The fauna of Australian seagrass beds Hutchings, P. A., & Recher, H. F., The fauna of Australian mangroves.............. Idamean (late Cambrian) fossils........... Illawarra Coal Measures (lower), Permian, of southern Sydney Basin Txeuticus candidus syn... ........+005++005. Jamberoo Sandstone Member, of Broughton Formation (Permian) Jell, P. A., see Powell, C. McA. Jenkins, C. J., Darriwilian (middle- Ordovician) graptolites from the Monaro Trough sequence east of Braidwood, New South Wales 0s 55 2s ois aie ciate: Jenkins, C. J., Late Pridolian graptolites from the Elmside Formation near Yass, New SouthiWaalesi ira. sins cola suroei estes Jonsell, B., Linnaeus and his two circum- navigating apostles.................. Kent, D.S., see McAlpine, D. K. Kiama Sandstone Member, of Broughton Formation (Permian) Kuo, J., Notes on the biology of Australian SEAGT ASSES eNetreee aun sets her Npussaencons © Larkum, A. W. D., & West, R. J., Stability, depletion and restoration of seagrass beds . Larkum, A. W. D., see also West, R. J. Late Pridolian graptolites from the Elmside Formation near Yass, New South Wales, byi@is Jienkinsiew see ice ee wacree aie a Linnaeus and his two circumnavigating apostles, by B. Jonsell................ Linnean Society of New South Wales. Record of the annual general meeting 1981. Reports and balance sheets..... annexure Linograptus posthumus...................-. Long distance transport of spores of Puccinia graminis tritici in the southern hemisphere, by I. A. Watson & C.N.A.deSousa .... McAlpine, D. K., & Kent, D. S., Systematics of Tapeigaster (Diptera: Heleomyzidae) with notes on biology and larval morphology. . . Macleay Memorial Lecture, 1982.......... Macquarie Island, rabbits, vegetation and ELOSIONN jest ees teers Guages ae oe eee Mangroves, faunaofAustralian........... Martin, H. A., see Holmes, W. B. K. Metaltellinae, Amaurobioidea............ Minumurra Latite Member, of Pheasants Nest Formation (Permian) Miocene Eucalyptus remains Mollusc, Cambrian Molluscs from Australian mangroves Proc. LINN. Soc. N.S.W. 106 (4), (1982) 1983 INDEX 123 181 83 127 287 247 291 173 311 Monaro Trough sequence east of Braidwood, graptolites Monograptus cf. angustidens ................ Monograptus cf. formosus.................. Monograptus transgrediens.................. Monograptus transgrediens cf. praecipuus........ Neef, G., see Powell, C. McA. Neurochaeta inversa McAlpine (Diptera: Neurochaetidae) and seed set in Alocasia macrorrhiza (L.) G. Don (Araceae) in southeast Queensland, by D. E. Shaw, B. K. Cantrell & K. J. Houston........... New genus of spiders of the subfamily Metaltellinae (Araneae, Amaurobioidea) from southeastern Australia, by M. R. Gray Notes on the biology of Australian seagrasses, bya)e Kuo cbsceet ne oes Stal om) caucenee ae Notoaphelaspis gen. nov..................-- Notoaphelaspis orthocephalis sp.nov........... Notoaphelaspis sp. cf. N. orthocephalis sp. nov... . Obolid, (Obolidae), indet................ Ordovician (middle), graptolites........... Origins and relationships among the animal phyla, by D.T.Anderson............. Palynology, Chalk Mountain Formation. ... . Percival, I. G., see Powell, C. McA. Permian stratigraphy, south Sydney Basin .. . Pheasants Nest Formation, lower Illawarra Coal Measures (Permian)............. Phryganoporus syn.............220+-204-- Pollination of Alocasia macrorrhiza........... Polychaetes from Australian mangroves Polychaetes from seagrass beds............ Powell, C. McA., Neef, G., Crane, D., Jell, P. A., & Percival, I. G., Significance of late Cambrian (Idamean) fossils in the Cupala Creek Formation, northwestern New South Pridolian, (Siluro-Devonian), graptolites .... Prismenaspis?'sp. MOV. . =... 5-252 25: PrOplina sp satel is, Sashes cies pita den curmaee te terere Pseudagnostus sp. aff. idalis Puccinia graminis tritici, transport of spores of. . . Rabbits, vegetation and erosion on Macquarie Island, by P. M. Selkirk, A. B. Costin, R. D. Seppelt & J. J. Scott............... Rainforest, tropical, ecological values....... Reappraisal of the stratigraphy of the upper Shoalhaven Group and lower Illawarra Coal Measures, southern Sydney Basin, New South Wales, by P. F.Carr........ Recher, H.F., see Hutchings, P. A. Reproductive systems of Tapemgaster......... Saddleback Latite Member, of Broughton Formation (Permian) Schneider, M. A., A comparative mer- 173 169 171 171 171 67 275 225 142 144 145 148 173 151 305 287 293 247 323 110 188 127 167 145 146 140 311 337 263 287 59 291 phological study of the reproductive systems of some species of Tapeigaster Macquart (Diptera: Heleomyzidae)...... 59 Seagrass and seagrass beds... .. 181, 201, 213, 225 Seagrass primary production — a review, by R. J. West & A.W. D.Larkum......... 213 Selkirk, P. M., Costin, A. B., Seppelt, R. D., & Scott, J. J., Rabbits, vegetation and erosion on Macquarie Island........... 337 Seppelt, R. D., see Selkirk, P. M. Scott, J. J., see Selkirk, P. M. Shaw, D. E., & Cantrell, B. K., A study of pollination of Alocasia macrorrhiza (L.) G. Don (Araceae) in southeast Queensland... 323 Shaw, D. E., Cantrell, B. K., & Houston, K. J., Neurochaeta inversa McAlpine (Diptera: Neurochaetidae) and seed set in Alocasia macrorrhza (L.) G. Don (Araceae) in southeast Queensland................ 67 Shoalhaven Group (upper), Permian, of southern Sydney Basin............... 287 Significance of late Cambrian (Idamean) fossils ‘in the Cupala Creek Formation, north- western New South Wales, by C. McA. Powell, G. Neef, D. Crane, P. A. Jell & I. (Gaeercivale yan Seay a. cierr Gs suacot oss) eet 127 Solandeny Daniel’. tcc. so) sa: cls a ae 1 Sousa, C. N. A. de, see Watson, I. A. SpauumiankyAnders)-e1420-2 2) oe ons 1 Stability, depletion and restoration of seagrass beds, by A. W. D. Larkum & R.J.West.. 201 Stigmatoa tysont) . 6. ee es 140. Study of the pollination of Alocasta macrorrhiza (L.) G. Don (Araceae) in southeast Queensland, by D. E. Shaw & B. K. (Clanntinellls cacceaseede pe peteens icc eens eer 323 INDEX Sydney Basin, southern, stratigraphy of ..... Systematics of Tapeigaster (Diptera: Heleo- myzidae) with notes on biology and larval morphology, by D. K. McAlpine & D. S. INE n bites sr gusty crpie ean Reese Ge apo Sts cee ee 33 TG PeUgas lenin Sedo t, 5 hye cn eneAS tg POPS LENORE 34, 59 Tapeigaster annulatacomb.nov. ....... . 48, 61, 63 Tapeigaster annulipes.............+.+... 42, 61, 63 Tapeigaster argyrospila..............+.+4+. 39 Tapeigaster brunneifrons..........-..+400-- 39 Tapeigaster cinctipes comb. nov.............. 38 Tapeigaster digitatasp.nov................ 45, 63 Tapeigaster luteipennis ................ 41, 61, 63 Tapeigaster nigricornis comb. nov. ...........- 37 Tapeigaster paramonovi sp.nov.............. 44 Tapeigaster pulvereasp.nov. ...........- 50, 61, 64 Tapeigaster subglabrasp.nov. .......... 52, 63, 64 Mrilobitess sje ree nes ee Se ates eee Se cee 140 Two new intertidal balanoid barnacles from eastern Australia, by B.A. Foster....... 21 Watson, I. A., & Sousa, C. N. A. de, Long distance transport of spores of Puccinia graminis tritici in the southern hemisphere. . Webb, L. J., Ecological values of the tropical rainforest resource (Macleay Memorial Lecture, 1982) West, R. J.. & Larkum, A. W. D., Seagrass 311 primary production — areview......... 213 West, R. J., see also Larkum, A. W. D. Westley Park Sandstone Member, of Broughton Formation (Permian)........ 289 Yass, N.S.W., graptolitesfrom............ 167 Proc. LINN. Soc. N.S.W. 106 (4), (1982) 1983 ADVICE TO AUTHORS The Linnean Society of New South Wales publishes in its Proceedings original papers and review articles dealing with biological and earth science. Papers of general significance are preferred but the Proceedtngs also provides a medium for the dissemination of useful works of more limited scope. Manuscripts will be received for assessment from non-members as well as members of the Society though non-members must communicate their works through a member. Subject to acceptance, a member’s paper may be given priority in publication over that of a non-member. Manuscripts (originals and two copies of text and illustrations) should be forwarded to the Secretary, Linnean Society of New South Wales, 35-43 Clarence Street, Sydney, Australia, 2000. Authors who are members of the Society are supplied with 25 free offprints of their papers after publication. Additional copies may be purchased, provided an order is placed when the corrected proofs are returned to the Honorary Editor. Donations towards the cost of publishing papers are always most welcome. In the case of lengthy papers or those with many illustrations or tables, contributions from authors, and especially non-member authors, may be requested at the discretion of Council. On publication a paper and the copyright thereof become the property of the Society. Requests to use copyright material should be directed to the Secretary. PREPARATION OF MANUSCRIPTS Copy must be typewritten, double-spaced throughout, on one side only of good quality A4 (210 X 297 mm) paper. Margins not less than 25 mm wide all round are required. All pages should be numbered serially and securely fastened together. The desired positions for all figures and tables should be indicated in the left-hand margin of the text. For taxonomic papers the Botanical or Zoological Codes of Nomenclature, as appropriate, must be followed. Generic and specific names should be clearly marked by underlining. Papers should be written in clear, concise English. The Style Manual for Authors and Printers of Australian Government Publications (Second Edition, 1972) is a useful guide. Spelling should conform to that preferred by the Oxford English Dictzonary. The general design of a paper should follow the scheme: (1) Title and author’s name — all in capitals. (2) A concise Abstract, complete in not more than 200 words, indicating the scope of the paper. Authors should adopt the lay-out used in this issue of the Proceedzngs, including details of postal address but leaving spaces for editorial insertions. (3) Main text. Footnotes should be avoided. The text may be divided into sections introduced by short headings set out asinthisissue. (4) Acknowledgements, if any. (5) References. These should be cited in the text by author’s name and date, e.g., Bullough (1939) or (Bullough, 1939) according to the context and listed alphabetically by authors under References thus: BuLLoucH, W. S., 1939.— A study of the reproductive cycle of the minnow in relation to the environment. Proc. zool. Soc. Lond., Ser. A, 109: 79-108. Titles of periodicals should be abbreviated as in the World List of Scientific Periodicals. If more than one work by the same author peblished in the same year is cited, use a, b, etc., after the year in both text and list of references. Titles of books should be quoted in full together with the place of publication, the name of the publisher and the edition if other than the first. Illustrations: Authors should note that illustrative matter (both photographs for half-tone reproduction and line drawings) is now printed in the text, not as separate plates. All illustrations must therefore be marked as figures. A number of small photographs may be arranged to form one figure. The individual parts of such a composite illustration should be clearly marked (preferably by capital letters) and identified in the caption. All captions must be typed on a separate sheet or sheets. The maximum printed dimensions for figures will normally be 125 x 200 mm; larger formats will be considered only in exceptional circumstances. Figures must therefore be designed to yield clear images within the limits of a single page. Close attention should be paid to the matter of scale on figures. Where possible add a linear scale (with the dimension clearly marked) to the figure rather than trust that a statement of scale in the caption will be correct after the plate-maker and printer have finished their jobs. All line drawings should be in India ink on a suitable surface, such as Bristol board, tracing linen or plastic film. In general, however, the platemaker prefers to work from good quality, glossy photographs rather than originals of various sizes. Authors are urged to supply such photographic reproductions which, if made to a scale appropriate to the size of a printed page, will show whether ornament and lettering can be read in the final print. All photographs, whether for half-tone or line illustrations, should be high-contrast, glossy prints. Each illustration should be identified (author’s name, Fig. no. and orientation) in pencil on the back. Tables should be submitted in a clear format on separate sheets. Like illustrations, they should be designed to fit a single page of the journal. PROCEEDINGS of LINNEAN SOCIETY OF NEW SOUTH WALES VOLUME 106 Issued May 10, 1983 CONTENTS: NUMBER 3 201 A.W.D.LARKUM and R. J. WEST Stability, Depletion and Restoration of Seagrass Beds 213 R.J.WEST andA. W. D. LARKUM Seagrass primary Production — aReview 225 ..J. KUO : Notes on the Biology of Australian Seagrasses 247 M.R.GRAY The Taxonomy of the semi-communal Spiders commonly referred to the Species Ixeuticus candidus (L. Koch) with Notes on the Genera Phryganoporus, Ixeuticus and Badumna (Araneae, Amaurobioidea) NUMBER 4 263 L.J.WEBB Sir William Macleay Memorial Lecture 1982 Ecological Values of the tropical Rainforest RESOUICE 275 M.R.GRAY - A new Genus of Spiders of the een Metaltellinae (Araneae, Amaurobioidea) from southeastern Australia’ 287 P.F.CARR A Reappraisal of the Stratigraphy of the upper Shoalhaven Group and lower Illawarra Coal Measures, southern Sydney Basin, New South Wales 299 W.B.K.HOLMES, F.M. HOLMES and H. A. MARTIN Fossil Eucalyptus Remains from the middle Miocene Chalk Mountain Formation, Warrumbungle Mountains, New South Wales ' 311 |. A. WATSON andC.N. A.deSOusA Long distance Transport of Spores of Puccinia graminis tritici in the southern Hemisphere 323 D.E.SHAW and B. K. CANTRELL A Study of Pollination of Alocasia macrorrhiza (L.) G. Don (Araceae) in southeast Queensland . 337. +P.M.SELKIRK, A. B. COSTIN, R.D. SEPPELT and J. J. SCOTT Rabbits, Vegetation and Erosion on Macquarie Island 347 INDEX to Volume 106 : Printed by Southwood Press Pty Limited, 80-92 Chapel Street, Marrickville 2204 ACME, a BOOKAIN’ NG CO., HH. FEB29 984 100 CAMBRIDGE tT nT WH LAEU I Saas ae dG AL et “eh ht nay AE HAE sy sect ba Tate's Teh aBh hay he tae : : Sey femay ges ath 33 a - ; Waeiien et MRA SW hA ahteaeebe 5 ‘ sa hone tae Peary errand ae 3 k pare eae Pome acne Baer ght Sot htt i heen et Fe heck ot at th ostinato teeth $ 4 fo thd oe fhe bee WALGER AM ae oy ie toa Nar Sarina’ eae ue ereree ear tee eileen tie MAAS RS Viasat ante Dice sae. SCR SS nite rye Peer hoary hy yard Cn eT ae ape ab yas) we wen ut uM ast Renee) (en Oe at) He uy ge te a gy Bh: ney baad ‘oS " aay we ne send et ‘ eeu a brane oe eal 7 »’ wevtae ernie wand, whitey Ae Bais , Bey caked Oars vb San!