Barta oh eae FPR HE teh ele ue ON BA whe WE aire Mead net pedir ada Ny tenia Bata he Cas Fam ge WET Pe Sarge a9 Me Sera Jemwite fafa ite eb Ya eee ME Hate Lt eh Oe aren OO ata 8 emer Saar mT Pb errr Re ISS HE TE Tip te WAI Cate ty on SAP a ee vets EM me PP alte! Cphaste tena! Medd yond cedar riteg hy Vie meh Waar, TOL Beene iste an pyrene, Ym Rae g aia tne ths SETAE) we sthU} Poaurkes 4 Tee owe » at Peletyeeyinanie Fae aNS ve NERESES aN Pinan thew gmer tyes Onset Piura. CABIN Wi ete LA Sederece ni tra, eonitee H a) SO ae eesa tH PY teeta Pesitte a an TAS Sterne orene renter HEE ys tehrese ly Y ite Set WAIN rere rere poanes teres Dieta oer er ee SHR OTEN cyan aw onan Bese a ruys Babess Para vees CNV TENE med dt abl tdhe afar antsy cae Prevenieeaind HON ety eae PM MAG Ie rh ee ybist Pees ye 2h vegee : Aise Pe hain Nera EEA x SON ys fam 3 bevel as the ty SEV ANN ee EE AN tye AUPE Dette ge aoe aye): NN ee SIR IN ot Bons “ ° ort inst Meher e weve: g AN shack YVekaee eewge reat ae y a VatOhena ped ctan wie FEY EL ed Saas. cra rene ane tehere et rere Abeer boas BV96§ Ay Ek wenty out hey ‘yy dete SPN AS eb rat RR WNT SS es Wers etye nan Saath swe MY Ge : rar Recnitestred Are NES Ace Baten ernst Seen ows) eran gh, Fons erat gral Pay io ea Saat Ou vam ata ee td bier fete PROCEEDINGS of the LINNEAN SOCIETY NEW SOUTH WALES | VOLUME 123 pS WSS Se \ | <— >> WN "A NATURAL HISTORY IN ALL ITS BRANCHES THE LINNEAN SOCIETY OF NEW SOUTH WALES ISSN 0370-047X 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 Microbiology 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. 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 $53.00 The current rate of subscription to the Proceedings for non- members is set at $80.00 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 Secratary. OFFICERS AND COUNCIL 2000-2001 President: J.P. BARKAS Vice-presidents: R.J. KING, K.L. WILSON, A. RITCHIE, R.A.L. OSBORNE Honorary Treasurer: 1.G. PERCIVAL Secretary: C.G. FORD Council: A.E.J. ANDREWS, M.L. AUGEE, J. BARKAS, M.R. GRAY, G.J. HARDEN, M.S. MOULDS, D.R. MURRAY, P.J. MYERSCOUGH, R.A.L. OSBORNE, I.G. PERCIVAL, A. RITCHIE, S. ROSE, K.L. WILSON Honorary Editor: M.L. AUGEE Linnean Macleay Lecturer in Microbiology: P.R. REEVES Auditors: Phil Williams Carbonara The postal address of the Society is P.O. Box 137, Matraville, N.S.W. 2036, Australia. Telephone and Fax +61 2 9662 6196. Mobile 0408 69 39 74 E-mail: linnsoc @acay.com.au Home page: http://www.acay.com.au/~linnsoc/welcome.html ©) Linnean Society of New South Wales Cover motif: Pilliga mouse drawn by David De Nardi from photo by Barry Fox (see Jefferys and Fox, this volume) PROCEEDINGS of the LINNEAN SOCIETY NEW SOUTH WALES JAN 25 2002 For information about the Linnean Society of New South Wales, its publications and activities, see the Society’s homepage http://www.acay.com.au/~linnsoc/welcome.html VOLUME 123 December 2001 Proc. Linn. Soc. N.S.W., 123. 2001 7 et ou ss EDITORIAL The Linnean Society of NSW is pleased to announce that our journal has finally achieved coverage in SciSearch (previously known as Science Citation Index-Expanded) and Current Contents. The Proceedings of the Linnean Society of New South Wales has of course long been included in Biological Abstracts and Zoological Records. Jean-Claude Herremans, a councillor of the Linnean Society of NSW, has completed a computer based index of the entire 122 volumes of the journal. Part of this index is on our web page, but it has become so large that a search using most commercial servers is very slow and cumbersome. Therefore the index has been put onto 31/2 inch floppy discs. A set of two, one disc in alphabetical order by authors’ surnames and the other broken down into large subject areas, can be obtained from the secretary of the society for $420.00. The discs are PC formatted and in text files. The Linnean Society of NSW plans a Symposium on Monotreme Biology in conjunction with the Australian Mammal Society’s AGM in Sydney in 2003. This will be the third in a series, begun in 1978 when the first was hosted by the Royal Zoological Society of NSW. The second was held in 1992. We anticipate that volume 125 of the Proceedings will include papers from this symposium. Anyone wishing to receive details of the 2002 symposium as they become available should contact the secretary of the Linnean Society by post or e-mail (addresses are on the inside front cover). IMPORTANT NOTICE - CHANGE IN JOURNAL SIZE From volume 124 this journal will be published in A4 size. This will allow considerable cost savings as the editorial team will be able to prepare each volume almost to the point of “print ready”. However the journal is prepared by unpaid volunteers and this change will require authors to cooperate in preparing manuscripts in a form that can be readily incorporated into our system. Details will be set out in revised “Instructions to Authors” which will be available through our web page and on request by post. However the main requirements will be that text is in Word for Windows (PC format) and that illustrations are presented as hardcopy or TIF or PDF files. M.L. Augee PhD FRZS Editor Proc. Linn. Soc. N.S.W., 123. 2001 2 at to eee te Topic o AeNog oil Sree Aste. vena YOU i LIMA sorron THATsONE es iat med & a2 ‘seittesiay gan a % at ey, re ce Tho a ues eo “Ss jesitper he bie Sean aed fa Cpe OF} awobii att (a et ok 704 ; oo a a nk 1 ve i vibe HOE RE ORAL 208 vem aT Historic Overview of Algal Blooms in Marine and Estuarine Waters of New South Wales, Australia PENELOPE AJANI, GUSTAAF HALLEGRAEFF 7 AND TIM PRITCHARD! !Water Science Section, Environment Protection Authority, New South Wales, PO Box A290 Sydney South, NSW, 1232, Australia 2School of Plant Science, University of Tasmania, GPO Box 252-55, Hobart, Tasmania, 7001, Australia Ajani, P., Hallegraeff, G. and Pritchard, T. (2001). Historic overview of algal blooms in marine and estuarine waters of New South Wales, Australia. Proceedings of the Linnean Society of New South Wales 123, 1-22. A compendium of algal bloom reports for marine and estuarine waters of New South Wales, Australia, for the period 1890-1999 is presented. The majority of blooms have been harmless water discolourations predominantly caused by the large heterotrophic dinoflagellate Noctiluca scintillans, or the filamentous cyanobacterium Trichodesmium erythraeum. Other harmless species that have bloomed include the dinoflagellates Gymnodinium sanguineum (= Akashiwo sanguinea) and Prorocentrum minimum, the surf diatom Anaulus australis, the coccolithophorid Gephyrocapsa oceanica and the ciliate Mesodinium rubrum. Species that have produced blooms and are potentially harmful to marine organisms include the silicoflagellate Dictyocha octonaria, the dinoflagellates Gonyaulax polygramma and Scrippsiella trochoidea and the diatoms Thalassiosira spp. and Chaetoceros spp. The toxic raphidophyte species Heterosigma akashiwo, Chattonella cf globosa and Haramonas sp., the dinoflagellates Gymnodinium galatheanum (= Karlodinium micrum), Dinophysis acuminata and Alexandrium catenella, and the diatoms Pseudo-nitzschia multiseries and P. australis have also been identified as bloom-forming species in these waters. Reports of algal blooms have apparently increased considerably since 1990 but the data may be biased because of the ad hoc nature of these reports. For this reason it is difficult to identify the cause(s) of this apparent increase in bloom frequency. Contributing factors may include the expansion in coastal settlements, an increase in awareness of environmental issues such as water quality, possible changes in anthropogenic nutrient input and/or the effects of large-scale oceanographic phenomena and/or climate change. Manuscript received 22 August 2001, accepted for publication 24 October 2001. KEYWORDS: algal bloom, ciliate, coccolithophorid, cyanobacterium, diatom, dinoflagellate, estuaries, raphidophyte, silicoflagellate. INTRODUCTION Fluctuations in the nutrient status of coastal waters, either of natural origin or associated with anthropogenic disturbances, can lead to changes in species composition and abundance of marine and estuarine microalgae. This may result in algal blooms that threaten fish resources, human health and ecosystem function and the recreational amenity of beaches and embayments. Other algal blooms are simply harmless, transient pulses in response to episodic nutrient enrichment such as from coastal upwelling events (Smayda 1997). Whatever factors affect their formation, the focus on algal blooms is increasing. Proc. Linn. Soc. N.S.W., 123. 2001 2 ALGAL BLOOMS IN N.S. W. Global evidence is emerging of an apparent increase in the distribution and frequency of algal blooms (Hallegraeff 1993). Eutrophication can lead to enhanced phytoplankton growth (including nuisance and/or toxic algal blooms) or a change in the species composition of phytoplankton and other organisms (Oviatt et al. 1989; Graneli and Moreira 1990; Riegman et al. 1992; Pan and Rao 1997). Hallegraeff reviewed harmful algal blooms in the Australian region in 1992 and suggested that from the 1970s to the 1990s there had been an escalation of harmful blooms in Australian marine and estuarine waters. Consequently, algal blooms have been targeted as a key environmental indicator for long term monitoring in Australian waters (Ward et al. 1998). There is, however, no systematic reporting of algal blooms in New South Wales (NSW) marine and estuarine waters. The NSW coastline is 1,900 km in length and ranges from warm subtropical in the north to cool temperate in the south ( Fig. 1). It has approximately 700 beaches; with large sandy beaches in the north and smaller pocket beaches bounded by rocky headlands in the south. It is naturally divided into three regions based on broad oceanographic characteristics and the geological structure of the coastline (EPA 1995). The northern region is under the dominant influence of the warm East Australian Current, while the cooler waters and currents from Bass Strait influence the southern region. The central region is a mixed zone. The majority of the human population (6.3 million) is concentrated in the central zone in three major cities, Sydney, Newcastle and Wollongong, and all of these cities are associated with major estuaries. Increased pressures on the entire NSW coastal zone (population increase and coastal development, tourism, water quality and sewage disposal) mean that the potential for algal blooms is increasing. The NSW aquaculture industry, presently worth $42-45 million, is projected to increase to $250 million by 2010. A three-year ocean nutrient and phytoplankton study carried out by the New South Wales Environment Protection Authority (NSW EPA) has been completed recently (1995-1998). That study characterised ambient nutrient concentrations and more specifically, identified the relative significance of various sources of nutrients in NSW offshore coastal waters. The focus of the investigation was on the extent of slope water intrusions and their associated nutrient concentrations, comparing this to anthropogenic nutrient inputs, including the three major deepwater ocean outfalls and estuarine discharges. Phytoplankton blooms at various locations along coastal NSW were related to this information. As part of this study the Commonwealth Scientific and Industrial Research Organisation (CSIRO) long-term 100m coastal station off Port Hacking was revisited in 1997-98 to investigate phytoplankton patterns and their hydrological environment and to compare these patterns to previous investigations at the same location (Ajani et al. 2001). It is useful to discuss recorded algal blooms in light of these data, as well as the links found between the environmental variables (factors which may promote algal blooms) and the phytoplankton community composition. Three types of algal blooms can be distinguished - those that are harmless water discolourations, potentially harmful to marine organisms (although non-toxic) or potentially toxic to humans. Human illness associated with harmful algae is due to the naturally occurring toxins which are transferred to humans through the consumption of seafood products. The most significant public health problems caused by harmful algae are Amnesic Shellfish Poisoning (ASP), Ciguatera Fish Poisoning (CFP), Diarrhetic Shellfish Poisoning (DSP), Neurotoxic Shellfish Poisoning (NSP) and Paralytic Shellfish Poisoning (PSP). Each of these syndromes is the result of different causative organisms that produce a range of toxins and risks to humans. These have been discussed by various authors (Taylor 1990; Hallegraeff 1991, 1992, 1993; Smayda 1997). Except for ASP (which is caused by diatoms), all other syndromes are caused by biotoxins synthesised by dinoflagellates. Proc. Linn. Soc. N.S.W., 123. 2001 P. AJANI, G. HALLEGRAEFF AND T. PRITCHARD 3 Figure 1. Map of New South Wales coastal zone Potentially toxic phytoplankton are not always toxic in every situation and it is anticipated that in the future other phytoplankton species may prove to be toxic under certain conditions (UNESCO 1995). In addition, the factors that promote algal populations to significantly increase (deviate from their normal cycle of biomass, 1.e. to ‘bloom’) are varied. A range of physical, chemical and biological variables are involved and there may be quite different bloom determinants depending on the type and location of the water body (eg estuarine versus offshore waters). Methods and Data Assessment This paper discusses visible bloom events that have been recorded in NSW marine (coastal and offshore) and estuarine waters. Various NSW government agencies, local councils, water authorities, universities and the public have contributed to these data. Published references to these blooms have been used wherever possible. Other data have been collated from NSW Environment Protection Authority [formally the State Pollution Control Commission (SPCC)] unpublished file reports and similar reports from Australian Water Technologies (AWT). Suitably qualified scientific officers (including the authors) identified the causative organisms in these bloom incidents. Blooms listed as “unidentified” were not examined by scientific officers for various reasons such as no sample was collected or because of sample deterioration prior to examination. Other potentially harmful species have also been identified in these waters but have not yet reached bloom proportions. These are also referenced and discussed. Proc. Linn. Soc. N.S.W., 123. 2001 4 ALGAL BLOOMS IN N.S.W. Table 1. Harmless algal blooms recorded in New South Wales’s marine (M) and estuarine (E) waters Date Location Bloom Taxa July- Aug 30-32 Sydney Harbour (E) Gymnodinium sanguineum! Oct-72 Taree and Coffs Harbour (M) Trichodesmium sp* Dec-72 Palm Beach to Cronulla (M) Trichodesmium sp* Oct-80 Alexandra Canal (Cooks River, Sydney) (E) Gymnodinium sanguineum? Oct-81 Lane Cove River (Sydney) (E) Gymnodinium sanguineum? Aug-82 Lake Macquarie (Central Coast) (E) Noctiluca scintillans? Dec-83 Newcastle, Narrabeen, Foster, Bondi Beach (M) Trichodesmium sp* Apr-84 Lane Cove River (Sydney) (E) Mesodinium rubrum? Noy-84 Lane Cove River (Sydney) (E) Mesodinium rubrum? Dec-84 Sydney to Wollongong (M) Trichodesmium sp? Feb-86 Lane Cove River (Sydney) (E) Mesodinium rubrum? Jan-89 Sydney coastal waters and Jervis Bay toUlladulla (M) Trichodesmium sp* Aug-92 Lake Macquarie (Central Coast) (E) Noctiluca scintillans? Sep-92 Berowra Creek (Hawkesbury River) (E) N. scintillans/G. sanguineum> Dec-92 Jervis Bay (M) Gephyrocapsa oceanica* Jan-93 Sydney beaches and Port Kembla (M) Noctiluca scintillans? Jun-93 Lake Illawarra (E) Noctiluca scintillans* 1994 Berowra Creek (Hawkesbury River) (E) Gymnodinium sanguineum* Feb-94 Sydney northern beaches (M) Noctiluca scintillans* Apr-94 Ham & Chicken Bay (E) Noctiluca scintillans? Jul-94 Berowra Creek (Hawkesbury River) (E) Pseudonitzschia pungens* Nov-94 Sydney northern beaches (M) Noctiluca scintillans* Noy-94 Offshore from Port Hacking and Wollongong (M) Noctiluca scintillans* Noy-94 Narrabeen Lakes (Sydney) (E) Noctiluca scintillans* Dec-94 Newcastle Beach (M) Noctiluca scintillans* Dec-94 Jervis Bay (M) Noctiluca scintillans* Mar-95 Pearl Beach (Sydney) (M) Noctiluca scintillans* Mar-95 Berowra Creek (Hawkesbury River) (E) Prorocentrum minimum* Aug-95 North Harbour (Sydney) (M) Gymnodinium sanguineum* Sep-95 Offshore from Sydney Heads (M) Noctiluca scintillans* Oct-95 Offshore from Boat Harbour and Anna Bay (M) Trichodesmium erythraeum* Feb-96 Stanwell Park to Astinmer (M) Noctiluca scintillans* Feb-96 Wamberal Beach (Central Coast) (M) Noctiluca scintillans4 Mar-96 Sussex Inlet (E) Noctiluca scintillans* Mar-96 Paradise Beach (St Georges Basin) (E) Noctiluca scintillans* Mar-96 Hymans Beach (Jervis Bay) Noctiluca scintillans* Mar-96 Sussex Inlet (E) Noctiluca scintillans* Aug-96 Shelley Beach (Manly) (M) Mesodinium rubrum* Aug-96 Greenwich Baths (Parramatta River) (E) Mesodinium rubrum* Oct-96 Offshore from Boat Harbour and Anna Bay (M) Trichodesmium erythraeum* Oct-96 Port Stephens to Broughton Island (M) Trichodesmium erythraeum* Oct-96 Manly (Sydney) (M) Noctiluca scintillans* Jan-97 Sydney Beaches (Warriewood to Manly) (M) Noctiluca scintillans* Jan-97 Evans Head and Coffs Harbour (M) Trichodesmium erythraeum* Jan-97 St Georges Basin and Jervis Bay (M) Noctiluca scintillans* Jan-97 Toowoon Bay and south to Bateau Bay (M) Noctiluca scintillans* Jan-97 Port Stephens (M) Noctiluca scintillans* Feb-97 Central Coast (M) Noctiluca scintillans* Feb-97 Jervis Bay (M) Noctiluca scintillans‘ Feb-97 Sydney Beaches (Collaroy to Coogee) (M) Noctiluca scintillans* Feb-97 Coledale to Astinmer (M) Noctiluca scintillans* Feb-97 Manly (Sydney) (M) Noctiluca scintillans* Proc. Linn. Soc. N.S.W., 123. 2001 P. AJANI, G. HALLEGRAEFF AND T. PRITCHARD 5 Feb-97 Warriewood (Sydney) (M) Noctiluca scintillans* Feb-97 Vincentia (Jervis Bay) (M) Noctiluca scintillans* Feb-97 Berowra Creek (Hawkesbury River) (E) Pseudonitzschia pungens* Apr-97 Frenchmans Bay (Botany Bay) (E) Noctiluca scintillans* Api-97 Lake Macquarie (E) Noctiluca scintillans* Jul-97 Stockton Bight to Birubi Point (M) Anaulus australis* Aug-97 Bass Point (Sydney) (M) Noctiluca scintillans* Sep-97 St. George’s Basin Beach (E) Noctiluca scintillans* Sep-97 Otford and Werrong (Royal National Park) (M) Noctiluca scintillans* Oct-97 Newcastle (M) Noctiluca scintillans* Dec-97 Illawarra Region (M) Noctiluca scintillans* Dec-97 Sydney northern beaches and Port Stephens (M) Noctiluca scintillans* Jan-98 Sydney northern beaches and Central Coast (M) Noctiluca scintillans* Jan-98 Boat Harbour to Boulder Bay (M) Trichodesmium erythraeum* Jan-98 Stanwell Park to Kiama (M) Noctiluca scintillans* Jan-98 Cronulla and Illawarra Region (M) Noctiluca scintillans* Jan-98 Batemans Bay (M) Noctiluca scintillans* Feb-98 Cronulla (M) Noctiluca scintillans* Feb-98 Clifton (Illawarra region) (M) Noctiluca scintillans* Feb-98 Whale Beach (Sydney) (M) Noctiluca scintillans* Feb-98 Collaroy Beach (Sydney) (M) Noctiluca scintillans* Mar-98 Sydney beaches and Bonnie Hill (Coffs Harbour) (M) Noctiluca scintillans* Mar-98 Coffs Harbour (M) Trichodesmium erythraeum* Mar-98 Evans Head (M) Trichodesmium erythraeum* Apr-98 Wollongong and Batemans Bay (M) Trichodesmium erythraeum* Apr-98 Richmond River (Ballina) (M) Trichodesmium erythraeum* May-98 Wallaga Lake (E) Noctiluca scintillans* Aug-98 St Georges Basin (E) Noctiluca scintillans* Sep-98 Lake Macquarie (E) Noctiluca scintillans* Sep-Oct-98 Gudgen Creek (Kingscliff, Grafton) (E) Trichodesmium erythraeum* Oct-98 Berowra Creek (Hawkesbury River) (E) Pseudo-nitzschiat* pseudodelicatissima* Dec-98 Illawarra Region (Thirroul and Coalcliff) (M) Noctiluca scintillans* Dec-98 North Coast NSW (various reports) (M) Trichodesmium erythraeum* Dec-98 Lake Illawarra (E) Noctiluca scintillans* Jan-99 North Coast NSW (various reports) (M) Trichodesmium erythraeum* Mar-99 Offshore from Jervis Bay Trichodesmium erythraeum* Oct-99 Richmond River E Trichodesmium erythraeum* Nov-99 Lake Illawarra (E) Noctiluca scintillans* Nov/Dec-99 Seven Mile Beach (Gerringong) (M) Anaulus australis* Dec-99 Sydney northern beaches (M) Noctiluca scintillans* ' Dakin and Colefax 1933, ? Blackburn and Cresswell 1993, > Hallegraeff 1995, * EPA unpublished RESULTS Harmless Algal Blooms The majority of visible algal blooms in NSW coastal waters have been harmless water discolourations (Table 1). Most of these blooms were of the large (200-800um cell diameter) heterotrophic dinoflagellate Noctiluca scintillans ‘Red tides’ of this species have been reported in offshore, coastal and estuarine waters including coastal lagoons of NSW. The majority of blooms occurred in spring and summer and especially after heavy rainfall. Also the resultant high ammonia content released after cell lysis may irritate fish that will generally avoid the bloom area (Okaichi and Nishio 1976). Proc. Linn. Soc. N.S.W., 123. 2001 ALGAL BLOOMS IN N.S.W. Figure 2a. LM. Filamentous cyanobacterium Trichodesmium erythraeum producing raft-like bundles, up to 750 um long; b. LM. Balloon-shaped, colourless dinoflagellate Noctiluca scintillans, 200-500 um diameter; c. SEM. Dinoflagellate Gonyaulax polygramma, showing ornamented cellulose plates with longitudinal ridges, cells 29-66 um long; d. LM. Large, 50-80 Um long, unarmoured dinoflagellate Gymnodinium sanguineum (“sanguineum’=blood); e. SEM. Calcareous nanoplankton Gephyrocapsa oceanica, 15 4m diameter; f. SEM. Triangular, armoured dinoflagellate cells, 10-15 um wide, of Prorocentrum minimum, covered with minute spinules; g. TEM. weakly silicified cell of the centric diatom Thalassiosira partheneia, 10 um diameter; h. TEM. Pennate diatom bloom of Pseudo-nitzschia pseudodelicatissima. Cells 57 to 150 um long. POT Fee th Proc. Linn. Soc. N.S.W., 123. 2001 P. AJANI, G. HALLEGRAEFF AND T. PRITCHARD 7 N. scintillans ( Fig. 2b) has long been present in NSW waters with its earliest record being related to its bioluminescent properties which were observed in Sydney Harbour by Bennett in 1860. In 1933 Dakin and Colefax referred to this species as a relatively minor component of the phytoplankton community in NSW coastal waters. Hallegraeff (1995) collated an historic overview of visible algal blooms in Australian waters with the first visible bloom of N. scintillans recorded in NSW waters in August 1982 (Lake Macquarie, central NSW). Since then, N. scintillans blooms appear to have increased in frequency and intensity (maximum cell density recorded as 5 x 10*cells/L in Jervis Bay February 1997). Over a one year sampling period from April 1997 to April 1998 at the CSIRO long-term coastal station off Port Hacking, Ajani et al. (1999) observed N. scintillans in 61% of samples and was absent from a very few samples in the autumn and winter months. The high frequency of occurrence for this species was unprecedented in these waters. This increase compared to earlier studies reinforces the need to investigate the global potential of Noctiluca as an indicator of coastal eutrophication (Porumb 1992; Qi et al. 1993; Lu et al. 1994; Huang and Qi 1997). The cyanobacterium Trichodesmium erythraeum ( Fig. 2a) is also a common ‘red tide’ organism in NSW coastal waters. This tropical/subtropical species produces episodic blooms that were historically reported as ‘sea sawdust’ during Captain Cook’s voyage through the Coral Sea (Cribb 1969). The filaments of this alga are united into small bundles that are just visible to the naked eye (~1mm). Blooms are most commonly seen in northern NSW waters in spring, summer and early autumn when the East Australian Current (EAC) transports these algal masses into NSW from Queensland waters. These blooms can appear yellow-grey in their early stages, while later they become a reddish- brown. Recent reports (Hahn and Capra 1992; Endean et al. 1993) suggest that this alga can produce compounds with a mouse intraperitoneal potency but this requires further investigation. Various T: erythraeum blooms have occurred in NSW waters with a particularly large bloom recorded on the south coast (Batemans Bay — south coast NSW) in early April 1998. Satellite imagery from NSW at his time showed unusually warm water throughout the NSW south coast area. This suggested that a strong manifestation of the EAC transported and conceivably triggered the extensive—T. erythraeum bloom in Batemans Bay (Clyde River estuary). The annual distribution of this species as reported by Ajani et al. (1999) from the Port Hacking 100m station also shows peak concentrations of this species in the coastal waters off Sydney in mid-April when surface waters were SIG, Gymnodinium sanguineum ( Fig. 2d), a large non-toxic dinoflagellate, has produced red water discolourations in NSW estuarine waters on several occasions (Hallegraeff 1991). Dakin and Colefax (1933) reported the first blooms in Sydney Harbour in July-August 1930, 1931 and 1932. Further blooms of this species have occurred in the Cooks River/ Alexandra Canal (Sydney) in October 1980, Lane Cove River in October 1981, Berowra Creek (a tributary of the Hawkesbury River estuary) in 1992 and 1994, and in North Harbour, Sydney in August 1995. This species has recently been renamed as Akashiwo sanguinea (Daugbjerg et al. 2000) Another dinoflagellate that is common in the plankton of Australian waters is Prorocentrum minimum ( Fig. 2f). This small ovate species can form extensive blooms and although it has been circumstantially linked with a shellfish-poisoning event in Norway in 1979 (venerupin poisoning), no toxic events have occurred in Australia to date (Hallegraeff 1991). In March 1995, this species bloomed in Berowra Creek. Each year this species becomes a dominant component of the phytoplankton in this reach of the Hawkesbury River. In offshore waters, maximum concentrations of this species have been observed in October (Ajani et al. 2001). Ararely reported ‘surf diatom’, Anaulus australis, was reported as causing oily slicks along Stockton Beach to Birubi Point (Hunter region) in July 1997 and again in Proc. Linn. Soc. N.S.W., 123. 2001 8 ALGAL BLOOMS IN N.S.W. November/December 1999 at Seven Mile Beach (Gerringong). These cells are able to rise to the surface and form dense accumulations by attaching themselves to wave- generated bubbles in high-energy surf zones. In most cases these accumulations disappear at night and reappear each morning. This species had only previously been reported along - the southern coasts of South Africa and Australia (McLachlan and Hesp 1984; Campbell 1996). Beginning in December 1992, the NSW coastal embayment of Jervis Bay was impacted for four weeks by an unprecedented, mono-specific bloom of the small (<10 uum) cosmopolitan coccolithophorid Gephyrocapsa oceanica ( Fig. 2e). It has been proposed that the combination of upwelled cool nutrient-rich slope water and an influx of warm EAC waters, providing enhanced upper layer temperatures and an oceanic algal seed, was the initial mechanism causing the milky green bloom (Blackburn and Cresswell 1993). The maximum cell density of 2 x 10’ cells/L (EPA unpublished) is greater than any previously recorded of this species in Australian waters. Deep purple, red or muddy looking blooms of the ciliate protozoan Mesodinium rubrum (=Myrionectra rubrum) ( Fig. 3d) are common in estuarine waters or sheltered embayments. These organisms contain cryptomonad algal symbionts. Blooms have usually been found to coincide with periods of warm and calm weather although no harmful effects have ever been recorded from such blooms (Bary 1953). Potentially Harmful Algal Blooms — Species harmful to marine organisms Some algal blooms can become so dense that they can cause death to fish and invertebrates due to either oxygen depletion or by abrasion damage to their gills (Table 2). A silicoflagellate Dictyocha octonaria ( Fig. 3h) was implicated as the causative organism in a fish kill which occurred in coastal waters off Newcastle in February 1993. Dead fish (especially Caranx sp.) were seen on beaches between Burwood Beach and Redhead and floating up to 3km offshore. While silicoflagellates are regularly seen in these waters in the spring and summer months (Ajani et al. 2001), a bloom event of this magnitude had never previously been recorded in NSW waters (Hallegraeff 1991). The non-toxic dinoflagellate Gonyaulax polygramma ( Fig. 2c) also has the potential to develop harmful anoxic ‘red tides’. Blooms in NSW have occurred in Sydney Harbour (Middle Head, July 1984), Darling Harbour (January 1993), Bate Bay (February 1993) and the most recent in Lake Macquarie in October 1993. Both colony-forming diatoms, Thalassiosira partheneia ( Fig. 2g) and T. weissflogii, have bloomed in NSW coastal waters. 7: partheneia bloomed in NSW coastal waters from August to September 1985 and was the dominant bloom species encountered during weekly sampling at the Port Hacking 100m station (Ajani et al. 2001). T. weissflogii bloomed in the Alexandra Canal in February 1986. An unidentified species belonging to the genus Thalassiosira bloomed at three locations in the Hawkesbury River in March 1999 - Berowra Creek (3 x 10° cells/L), Calabash Bay (3 x 10° cell/L) and Neverfail (5 x 10° cell/L). Despite oyster leases being nearby, no economic loss was reported from these blooms. In Tokyo Bay in 1951 a similar species, 7: mala, bloomed and damaged the gills of cultured bivalves resulting in significant economic loss (Takano 1956). Red-brown blooms of the dinoflagellate, Scrippsiella trochoidea ( Fig. 3a), have been reported in NSW coastal waters as early as the 1890s (Whitelegge 1891). This species has also caused water discolourations in the Hawkesbury River in March 1991 and Jervis Bay in 1994. Although non-toxic, this commonly occurring species can cause deoxygenation of the water and subsequent fish kills when it blooms in sheltered bays (Hallegraeff 1991). A bloom of Chaetoceros spp. occurred in January 1998 in Gunnamatta Bay, Port Hacking. No harmful effects were observed from this bloom, however large concentrations of some Chaetoceros spp. can potentially damage the gills of fish, which in turn produce mucus that induces hypoxia (oxygen deficiency in the body’s tissues) and Proc. Linn. Soc. N.S.W., 123. 2001 P. AJANI, G. HALLEGRAEFF AND T. PRITCHARD 9 hypercapnia (excessive amount of carbon dioxide to the blood) (Rensel 1993). The estuarine bloom coincided with maximum cell densities of this genus at the long-term station offshore from Port Hacking (Ajani et al. 2001). Figure 3a. SEM. Red-water dinoflagellate Scrippsiella trochoidea, 16-36 {1m long. Note tube-shaped apical pore on top of the cell and nearly equatorial (not displaced) girdle groove; b. LM. Chain-forming dinoflagel- late Alexandrium catenella, causative organism of paralytic shellfish poisoning. Individual cells 20-22 um long.; c.SEM. Red water dinoflagellate Alexandrium minutum, causative organism of paralytic shellfish poi- soning. Individual cells 24-29 im diameter. Note the hook-shaped apical pore on top of the cell and character- istic shape of the connecting first apical plate.; d. LM. Ciliate Mesodinium rubrum, 30 Um diameter, with two systems of cilia arising from the waist region.; e. LM.’Raspberry”-like cell of the fish-killing flagellate Heterosigma akashiwo (“akashiwo”’=red tide), containing numerous disc-shaped chloroplasts. Cell 11-25 um long.; f.LM. Undescribed flagellate resembling Haramonas. The cell surface is covered by numerous mu- cous-producing vesicles. Cells 30-40 um long. ; g. SEM. Small armoured dinoflagellate Dinophysis acuminata, causative organism of diarrhetic shellfish poisoning. Cells 38-58 um long.; h. SEM. Siliceous skeleton of the silicoflagellate Dictyocha octonaria, 10-20 um diameter; i. SEM. Small unarmoured, fish-killing dinoflagel- late Gymnodinium galatheanum, 15 um diameter. Proc. Linn. Soc. N.S.W., 123. 2001 10 ALGAL BLOOMS IN N.S. W. Table 2. Algal blooms recorded in New South Wales’s marine (M) and estuarine (E) waters - potentially harmful to marine organisms Location Bloom Taxa Sydney Harbour (E) Scrippsiella trochoidea' Middle Head, Sydney Harbour (E) Gonyaulax polygramma? NSW Coast (M) Thalassiosira partheneia’ Alexandra Canal (Cooks River, Sydney) (E) Thalassiosira weissflogii* Hawkesbury River (E) Scrippsiella trochoidea* Darling Harbour (Sydney) (E) Gonyaulax polygramma* Burwood Beach to Redhead(Newcastle) (M) Dictyocha octonaria* Bate Bay (M) Gonyaulax polygramma* Lake Macquarie (E) Gonyaulax sp° Lake lawarra (E) Gymnodinium cf. mikimotoi> Gunnamatta Bay (Port Hacking, Sydney) (E) Chaetoceros spp° Berowra Creek, Calabash Bay and Neverfail Thalassiosira sp° (Hawkesbury River) (E) Kianga Lake, NSW South coast (E) Chaetoceros spp.° 'Whitelegge 1890, 7 SPCC unpublished, * Hallegraeff 1991, * Hallegraeff 1995, ° EPA unpublished Potentially Harmful Algal Blooms - Toxic species Various algal species are potentially toxic to fish and humans ( Fig. 3 and Table 3). Heterosigma akashiwo (= H. carterae) ( Fig. 3e) a toxic raphidophyte, bloomed in Salamander Bay, Port Stephens (1980), Berowra Creek (November 1991, July 1995) and Sydney Harbour (November and December 1995). These potato-shaped cells contain numerous ejectosomes that readily discharge (especially upon preservation) making microscopic identification difficult. The toxicity of this species to fish was illustrated in the Puget Sound (USA) area where fish kills resulted in a considerable monetary loss to the salmonoid aquaculture industry (Rensel et al. 1989). The toxic mechanism of this species is still under investigation although data suggest that it is through the disruption of the fish gill lamellae. Hypotheses for this toxicity include possible production of high concentrations of superoxide radicals (Yang et al. 1995), the presence of heterotrophic bacteria which may explain the cell’s high toxin variability (Carrasquero- Verde 1999), or the presence of a neurotoxin which is similar to, but not identical to, brevetoxin (Kahn et al 1998) Another toxic raphidophyte, Chattonella cf. globosa, bloomed sporadically in Canada Bay, Sydney Harbour, from November 1996 to March 1997. Blooms of related species have caused significant mortality of cultured yellowtail and red sea bream in Japanese inland seas (Okaichi 1985) and implicated in the mass mortality of farmed bluefin tuna in Boston Bay, South Australia in 1996 (Marshall and Hallegraeff 1999). The production of superoxide radicals as the major mechanism of fish mortality is also hypothesised for this genus. Evidence for brevetoxin-like production is still being investigated. A raphidophyte flagellate, Haramonas dimorpha was identified and described from the mouth of the Daintree River, northeast Australia by Horiguchi in 1996. This genus is distinguished from other raphidophytes by the cell shape, the possession of a tubular invagination and a unique arrangement of chloroplasts. In December 1998, a closely related but as yet undescribed Haramonas species ( Fig. 3f), was implicated in a fish kill in Morrisons Bay, Parramatta River. Rainfall and proximity to a major stormwater canal were possibly contributing factors to bloom development in these waters. Another microalga associated with toxicity to fish is Gymnodinium galatheanum ( Fig. 31). Between June and July 1991 this species was implicated in two extensive fish Proc. Linn. Soc. N.S.W., 123. 2001 P. AJANI, G. HALLEGRAEFF AND T. PRITCHARD 1] Table 3. Algal blooms recorded in New South Wales’s marine (M) and estuarine (E) waters - potentially toxic to humans Location Bloom Taxa Port Hacking (M) Alexandrium catenella' Salamander Bay (Port Stephens) (E) Heterosigma akashiwo? Lake Illawarra (E) Gymnodinium galatheanum’? Berowra Creek (Hawkesbury River) (E) Heterosigma akashiwo? Newcastle Harbour (E) Alexandrium catenella? Parramatta River (Sydney) (E) Alexandrium catenella? Berowra Creek (Hawkesbury River) (E) Pseudonitzschia multiseries* Berowra Creek (Hawkesbury River) (E) Gymnodinium galatheanum? Jervis Bay (M) Prorocentrum sp, Dinophysis sp and Ceratium furca? Jan-95 Berowra Creek (Hawkesbury River) (E) Pseudonitzschia multiseries* Mar-95 Berowra Creek (Hawkesbury River) (E) Gymnodinium sp. Jul-95 Berowra Creek (Hawkesbury River) (E) Heterosigma akashiwo? Novy and Dec-95 Sydney Harbour (E) Heterosigma akashiwo?* Nov-96- Parramatta River (Sydney Harbour) (E) Chattonella globosa’ Mar-97 Ballina (M) Dinophysis acuminata? Dec-1997 Stockton Beach (M) Dinophysis acuminata?’ Mar-1998 Dec-98 Morrisons Bay (Parramatta River) (E) Haramonas sp. nov.* Apr 99- The Broadwater (Myall Lakes) (E) Mar 2000 Microcystis aeruginosa>* Oct-99 Wagonga Inlet, Narooma (E) Pseudonitzschia spp.* Nov-99 Parramatta River (Sydney Harbour) (E) Alexandrium sp. * 'Hallegraeff 1995, 7 SPCC unpublished, > EPA unpublished, * AWT unpublished kills in Lake Illawarra. Berowra Creek also experienced a bloom of this species in early 1994. G. galatheanum has been reported to produce toxins that are lethal to the eggs and larvae of certain fish. It is believed that the toxin produced is a haemolytic toxin, which affects the permeability of red blood cells. The physiological response of fish organisms to the toxin is believed to be the damage and necrosis of gill filament epithelial cells (Nielsen 1993). This species has recently been renamed as Karlodinium micrum (Daugbjerg et al. 2000). Dinophysis acuminata ( Fig. 3g) and D. tripos, both producers of diarrhetic shellfish poisoning (DSP), were implicated in the contamination of edible bivalves (Donax sp.) at Ballina, approximately 700 km north of Sydney, in December 1997 and in the Hunter area, approximately 200 km north of Sydney, in March 1998. Fifty-nine cases of gastroenteritis in humans were reported from Ballina and 23 cases were reported from the Hunter area following consumption of the bivalve. A mouse bioassay revealed a positive result for an unidentified DSP toxin and both species were found in the gut of the bivalve. Pectenotoxin DSP toxins have now been fully characterised (Quilliam et al., 2000). Peak concentrations of D. acuminata at the Port Hacking 100m station occurred in January (Ajani et al. 2001). Species belonging to the genus Pseudo-nitzschia have been implicated as the causative organisms of amnesic shellfish poisoning (UNESCO 1995; Hallegraeff 1994). Blooms of the toxic species P. multiseries were detected in Berowra Creek in 1993 and 1995 (5% of total phytoplankton biomass, EPA unpublished)(Table 3). A bloom predominantly of P. pseudodelicatissima ( Fig. 2h) was detected in Berowra Creek from October to November 1998. Although this species has been found to be toxic elsewhere Proc. Linn. Soc. N.S.W., 123. 2001 12 ALGAL BLOOMS IN N.S.W. (UNESCO 1995), analysis results from oysters from nearby leases showed no detectable concentrations of domoic acid, In October 1999, oyster leases in Wagonga Inlet, Narooma, were closed for harvesting due to a bloom of P. pseudodelicatissima, P. pungens and P. australis (toxic species) (Lapworth et al. 2000) (Table 3). P. pungens, a harmless species, also bloomed in Berowra Creek in July 1994 and again in February 1997 (Table 1). Many species of this genus have been observed at the 100m Port Hacking station (Hallegraeff and Reid 1986; Ajani et al. 2001). Scanning electron microscopy of samples collected from this station during 1997-1998 identified two potentially toxic species, P. australis and P. multiseries as well as the harmless species P. pseudodelicatissima, P. cf. subfraudulenta, P. pungens, P. subpacifica and P. heimii (Ajani et al. 2001). Two species of the dinoflagellate genus Alexandrium have been found to produce Paralytic Shellfish Poisoning (PSP) in NSW coastal waters - Alexandrium catenella ( Fig. 3b) and Alexandrium minutum (de Salas et al. 2000). A. catenella is thought to be the causative organism for the first record in Australian medical literature of human shellfish poisoning. This account is of wild mussels collected in February 1935 near Batemans Bay that produced typical PSP symptoms in mice (Le Messurier 1935). Since then, blooms of A. catenella have occurred in Port Hacking (1945) and in the Parramatta River in October 1993 (maximum cell count 9.5 x 10° cells/L, EPA unpublished). Samples of wild oysters collected during this time were found to contain more than 3 mg/kg of saxitoxin, which exceeded the statutory limit for Paralytic Shellfish Poisoning toxin concentrations of 0.8 mg/kg, prescribed under the NSW Food Act, 1989. Low levels of this toxin were also detected in harbour prawns. This species also bloomed in Newcastle harbour in September 1993. Toxic cysts (non-motile resting stage of some species) have been found in many NSW coastal estuaries and embayments. Cysts of Alexandrium have been reported in the gut of oysters from Port Stephens, a major NSW shellfish growing area (Richardson 1991). Sediment samples from oyster growing areas in Botany Bay in 1993 also contained resting cysts of the potentially toxic dinoflagellate genus Alexandrium (Lincoln-Smith and Smith 1993). It is unknown if these cysts were from potentially toxic species. A cyst survey of surface sediments in 1995 revealed the widespread occurrence of the toxigenic dinoflagellate Alexandrium catenella in the Lower Hawkesbury River, Sydney Harbour, Woolooware Bay, Botany Bay, and Port Hacking to Batemans Bay (Hallegraeff et al 1995). These results confirm Hallegraeff’s previous finding of this species at Batemans Bay in 1991. Cysts of’A. catenella were also found at low concentrations in surface sediments taken from several sites in Twofold Bay as part of an introduced marine pest survey in the Port of Eden by CRIMP (Centre for Research into Introduced Marine Pests) (1997). In 1992 Hallegraeff and Bolch examined the ballast water from a vessel which had its ballast tanks filled during a toxic dinoflagellate bloom in Japan. An estimated >300 million viable Alexandrium cysts were contained in this vessel upon arriving at the Port of Eden. This indicated that both diatoms and dinoflagellates that are not endemic to a region could be inadvertently introduced when their resistant resting stages are discharged with ballast-tank waters and the sediments of bulk cargo vessels. A. catenella and A. minutum ( Fig. 3c) have recently been detected in the Port of Newcastle during a similar survey (CRIMP 1999). Plankton samples during this survey also showed high numbers of vegetative cells of A. catenella in the water column, indicating that early spring might be a potential bloom period for this species. In addition, large numbers of A. minutum cysts were found at a dredge disposal site offshore from Newcastle. The most northern report of these species in New South Wales to date is Newcastle. In March 1994, a bloom of three dominant dinoflagellate taxa occurred in Jervis Bay: Prorocentrum sp, Dinophysis sp and Ceratium furca (EPA unpublished). There is no information available on maximum cell count, bloom colour, extent or duration. Proc. Linn. Soc. N.S.W., 123. 2001 P. AJANI, G. HALLEGRAEFF AND T. PRITCHARD 13 Species of the toxic cyanobacteria genera Anabaena and Microcystis are freshwater species that sometimes bloom in brackish/marine waters. Throughout 1999 and early 2000, Anabaena circinalis bloomed along the shoreline of the Myall Lakes, central NSW. Mircocystis aeruginosa increased in numbers towards the end of the bloom and low levels of toxins were detected in samples taken in early February 2000. Table 4. Unidentified algal blooms recorded in New South Wales’s marine (M) and estuarine (E) waters Location Taree (M) North Bridge (Sydney Harbour) (E) Evans Head (M) Seal Rocks (M) Maroubra Beach (Sydney) (M) Sydney Harbour (E) Brunswick Headland (M) Sydney northern beaches (M) Iron Cove Bay (Sydney Harbour) (E) North Head (Sydney Harbour) (M) Sydney northern beaches (M) Sydney northern beaches (M) Warriewood Beach (Sydney) (M) Dee Why Beach (M) Bloom Taxa Unidentified Species! Unidentified Species! Unidentified Species! Unidentified Species! Unidentified Species! Unidentified Species! Unidentified Species! Unidentified Species! Unidentified Species! Unidentified Species! Unidentified Species! Unidentified Species! Unidentified Species! Unidentified Species! Baragoot Lake, Bega (E) Five Dock, Sydney (E) Unidentified Species! Unidentified Species! ‘EPA unpublished Unidentified Blooms In addition to the blooms discussed above there have been a number of reports where algal blooms were observed for which no species identifications were reported/ recorded (Table 4). Reports of other toxic algae present in the plankton There are many species of the dinoflagellate genus Gymnodinium, of at least which ten are found in NSW coastal waters. Three species of this genus (G. catenatum, G. breve and G. mikimotoi) have been linked to toxic episodes elsewhere (PSP and NSP). There have been limited observations of these species in NSW coastal waters. A very localised cyst population of G. catenatum was found in Cowan Creek in 1995 (EPA unpublished). In 1998, this species was observed in the plankton from Calabash Bay, Berowra Creek. This was the first record of this species in the plankton of the Hawkesbury River. The dinoflagellate Alexandrium minutum has been identified in the Shoalhaven River in 1993 but to date has not bloomed in NSW waters. The presence of cysts of this species in sediments from New South Wales waters has been discussed above. During the recent sampling at the CSIRO long-term station off Port Hacking, other potentially harmful species were observed (Ajani et al. 2001). Dinophysis tripos was observed in 27% of samples collected throughout the year, D. fortii was present in 6% of samples and D. acuta 2%. D. caudata was also observed as ‘present’ in samples. These have all been implicated in diarrhetic shellfish poisoning (DSP) events throughout the world although toxin production is variable. Proc. Linn. Soc. N.S.W., 123. 2001 14 ALGAL BLOOMS IN N.S.W. Figure 4a. Recorded algal blooms in NSW marine and estuarine waters (1970 - present) with human population data. 6.4 | 6.2 f 6.0 5.8 | 5.6 15.4 POsZ | 5.0 | 4.8 _ 46 No. of blooms 1) (jo) Population (mil) 1970 1975 1980 1985 BAll bloom data @Noctiluca and Trichodesmium Pa Population (million) 1990 Figure 4b. Recorded algal blooms in NSW marine and estuarine waters (1970 - present) with annually averaged oxidised nitrogen concentrations. No. of blooms 1970 1975 1980 1985 1990 1995 BAI bloom data m@Noctiluca and Trichodesmium = Nox-N (ug/L) at PH50m (30m depth average) DISCUSSION This paper presents mainly ad hoc reports of bloom events in NSW marine and estuarine waters (Tables | to 4). It is likely that additional blooms have occurred and gone unreported. Bias in the data is therefore likely. The magnitude of this bias, and how significant the apparent changes in bloom reporting frequencies, are discussed below. Proc. Linn. Soc. N.S.W., 123. 2001 P. AJANI, G. HALLEGRAEFF AND T. PRITCHARD 15 Factors favouring higher reporting rates include increased awareness of environmental issues especially water quality, increased reliance on fisheries resources for food production, coastal population growth (including the commissioning of new ocean outfalls), public awareness of research activities and the utilisation of our coastal fringes for recreational and aquaculture purposes. It is not surprising therefore, that algal bloom reports have increased dramatically since the early 1900s ( Fig. 4). Although warm, calm weather and the presence of nutrient rich slope water during summer months increases the probability of visible blooms, high reporting rates from December to March may also be associated with an increased population of opportunistic recreational observers ( Fig. 5). In addition to this, the population of New South Wales has increased at a steady rate since 1970, reaching 6.3 million in 1998 ( Fig. 4a)(Bray 1995-1999). The diversion of sewage effluent in Sydney from shoreline outfalls to deepwater outfalls in 1990 may have affected phytoplankton populations although intrusions of nutrient-rich slope water have been identified as the principal factor leading to the development of marine algal blooms in these waters (Pritchard et al. 1999). Nutrient concentrations in NSW coastal waters vary considerably over time. Long-term trends in CSIRO data suggest possible increases in phosphate concentrations, (although phosphate data ceased to be collected in 1985). The nutrient ratio of dissolved inorganic nitrogen to dissolved inorganic phosphorus (DIN: DIP), and dissolved inorganic silicate to dissolved inorganic phosphorus (DIS: DIP) calculated from this long-term data set, however, show no clear trend to date. Annually averaged oxidised nitrogen concentrations from the CSIRO 50m long-term coastal station are shown in Fig. 4b. No clear pattern over time is evident. There were only four samples collected in 1988 and the low sampling frequency is believed to be the reason for the unusual peak in the annual average at this time. When the geographical distribution of blooms in NSW is examined, blooms are concentrated around major cities ( Fig. 6). This is likely due to a combination of the larger communities available for ad hoc observations and reporting and higher anthropogenic inputs. Figure 5. Annual distribution of recorded algal blooms in NSW marine and estuarine waters. 20 | No. of phytoplankon blooms Jul Aug Sep Oct Nov’ Dec _— Jan Feb Mar Apr May Jun ‘Toxic Blooms BNoctiluca scintillans oVarious Species Proc. Linn. Soc. N.S.W., 123. 2001 16 ALGAL BLOOMS IN N.S.W. Figure 6, Geographical distribution of blooms in NSW marine and estuarine waters Likewise, changes in the reporting and recording of algal blooms can introduce bias especially when most data come from ad hoc reports. For example, the EPA’s ocean nutrient and phytoplankton project solicited and facilitated reporting and recording of blooms from 1995 to 1998 (between Port Stephens and Jervis Bay) and inevitably contributed to the high reporting frequencies at these locations and during these years. Extensive Noctiluca scintillans blooms observed off Sydney and Port Kembla during January-February 1993 were recorded as a single bloom event while during January- February 1997, 11 individual reports of Noctiluca scintillans blooms were recorded (Table 1). There is evidence to suggest that many of the 1997 blooms were associated with a regional dynamic that caused three to four broad scale slope water intrusion ‘events’ Proc. Linn. Soc. N.S.W., 123. 2001 P. AJANI, G. HALLEGRAEFF AND T. PRITCHARD 17 Figure 7a. Algal secies distribution for recorded blooms in NSW marine and estuarine waters 8% 2% 3% 2% 2% 2% 16% g Noctiluca scintillans (harmless) @ Trichodesmium erythraeum (harmless) & Gymnodinium sanguineum (harmless) 5 Pseudonitzschia pungens (harmless) ug Mesodinium rubrum (harmless) a Gonyaulax polygramma (harmful to marine organisms) = Scrippsielia trochoidea (harmful to marine organisms) g Heterosigma akashiwo (potentially toxic) a Alexandrium catenella (PSP) 8 Other (see text) a Unidentified species Proc. Linn. Soc. N.S.W., 123. 2001 18 ALGAL BLOOMS IN N.S.W. Figure 7b. Algal species distribution for recorded blooms in NSW estuarine waters only. 20% W% AN 4% AX @ Gymnodinium sanguineum (harmless) Pseudonitzschia pungens (harmless) @ Mesodinium rubrum (harmiss) | @ Gonyaulax polygramma (harmful to marine organisms) m Scnppsiella tochoidea (harmful to marine organisms) @ Heterosigma akashiwo (potentially toxic) w Alexandnum catenella (PSP) tw Other (see text) 3 Unidentified species Proc. Linn. Soc. N.S.W., 123. 2001 P. AJANI, G. HALLEGRAEFF AND T. PRITCHARD 19 (Pritchard et al. 1997). In contrast, after the completion of data gathering for the ocean nutrient project, only 4 blooms were identified during the summer of 1998-99 and three of these were’ Trichodesmium erythraeum that has a tropical and sub tropical origin and is readily identifiable. When Noctiluca and Trichodesmium blooms are removed altogether from the bloom reports, the remaining reported blooms are predominantly estuarine blooms (Figs 7a and 7b). Water discolouration is also an important factor in the reporting of algal proliferations. Green algal blooms blend somewhat with the naturally changing spectra of coastal waters, and therefore are likely to go unreported. The more prominent bloom colours such as yellow, brown, red or milky (Hallegraeff 1991) are more ‘visible’ discolourations, and are therefore more likely to be noticed and recorded. Some species show no water discolouration but can be highly toxic at very low cell concentrations (Anderson 1995). In addition to this, many groups of algae remain poorly studied and it is likely that many more species will be found to bloom and more found to be toxic (Moestrup 1994). These ad hoc data sets, therefore, generally promote hypotheses and speculation about changes and differences in bloom events rather than providing the data set to test such hypotheses. Recognition of the limitations of the existing data emphasises the need to adopt more consistent monitoring/surveillance and recording protocols. Despite potential bias associated with ad hoc reports of visible blooms, there appears to be a compelling argument, supported by quantitative data (e.g. Hallegraeff and Reid 1986 and Ajani et al. 2001), to suggest that visible blooms in recent years have become dominated by Noctiluca scintillans with evidence to suggest that this was not the case two decades ago. In fact, forty-five percent of all recorded blooms in these waters were caused by the dinoflagellate Noctiluca scintillans ( Fig. 7a). Causal factors for this dominance remain unclear although large scale phenomena such as the El Nino Southern Oscillation may be a significant factor contributing to long term variability (Lee et al. 2001). (The ‘other’ group in these Figs 7a and 7b consists of novel bloom species, i.e. species that have been recorded as blooming only once to date). Long term data collection on phytoplankton populations and algal bloom events in NSW waters, and the use of emerging technologies such as molecular probes, remote sensing and physical/chemical/biological modelling, could be invaluable in the understanding of phytoplankton dynamics and their role in ecosystem health. ACKNOWLEDGMENTS The authors would like to thank EPA staff Mr Randall Lee, Mr Paul Rendell, Mr Sean Hardiman and Mr Kieran Horkan for their contributions. We are also grateful to Dr Peter Scanes and Dr Tony Church for reviewing this manuscript. Additional taxonomic assistance was provided by Prof. Ojvind Moestrup (Dept. of Phycology, University of Copenhagen, Denmark), Dr Steve Brett (Microalgal Services, Victoria) and Ms Caroline Lapworth (University of Tasmania). REFERENCES Ajani, P., Lee, R., Pritchard, T. and Krogh, M. (2001). Phytoplankton dynamics at a long-term coastal station off Sydney, Australia. Journal of Coastal Research (in prep.). Anderson, D.M. (1995). Toxic red tides and harmful algal blooms: a practical challenge in coastal oceanography. Rey. Geophysics, Suppl. U. S. National Report to the Int. Union of Geodesy and Geophysics. 1991-1994. pp. 1189-1200. Bary, B.M. (1953). Sea-water discoloration by living organisms. New Zealand Journal of Science and Technology. March 1953: 393-407. Bennett, G. (1860). ‘Gatherings of a naturalist in Australasia: being observations principally on the animal and vegetable productions of New South Wales, New Zealand and some of the Austral islands’. (London: John Van Voorst, Paternoster Row). Proc. Linn. Soc. N.S.W., 123. 2001 20 ALGAL BLOOMS IN N.S.W. Blackburn, S.I. and Cresswell, G. (1993). A coccolithophorid bloom in Jervis Bay. Australian Journal of Marine and Freshwater Research 44, 253-260. Bray, G. (1995-1999). ‘New South Wales Year Book No.s 75-79’. Australian Bureau of Statistics, NSW. Campbell, E.E. (1996). The global distribution of surf diatom accumulations. Rev. Chil. Hist. Nat. 69, 495- 450. Carrasquero- Verde, J.R. (1999). Role of associated bacteria in Heterosigma carterae toxicity to salmonoids. Aquatic. Toxicology 45, 19-34. Cribb, A.B. (1969). Sea sawdust. Queensland Naturalist 19, 115-117. CRIMP (1997). ‘Introduced species survey: Port of Eden and Twofold Bay, New South Wales’. CSIRO Centre for Research on Introduced Marine Pests. CRIMP (1999). “Introduced species survey: Newcastle, New South Wales’. CSIRO Centre for Research on Introduced Marine Pests. Dakin, W.J. and Colefax, A. (1933). The marine plankton of the coastal waters of New South Wales. 1. The chief planktonic forms and their seasonal distribution.’ Proceedings Linnean Society N.S.W. 58, 186-222. Dakin, W.J. and Colefax, A. (1940). The plankton of the Australian coastal waters off New South Wales. University of Sydney Monograph No.1 (Australasian Medical Publishing Co: Sydney). de Salas, M.F., van Emmerik, M.J., Bolch, C.J., Negri, A. and Hallegraeff, G.M. (2000). Genetic affinities of the Australian PSP dinoflagellates Alexandrium catenella, A.tamarense and A.minutum: introduced or indigenous? In “Harmful Algal Blooms 2000’ (Eds G.M. Hallegraeff, S.I. Blackburn, C.J. Bolch and R.J. Lewis). Proceedings of 9th International Conference on Harmful Algal Blooms. IOC. Paris. Daugbjerg, N., G. Hansen, J. Larson and O. Moestrup. (2000). Phylogeny of some of the major genera of dinoflagellates based on ultrastructure and partial LSU rDNA sequence data including the erection of three new genera of unarmoured dinoflagellates. Phycologia 39, 302-317. Endean, R., Monks, S.A., Griffith, J.K. and Llewellyn, L.E. (1993). Apparent relationships between toxins elaborated by the cyanobacterium Trichodesmium erythraeum and those present in the flesh of the narrow-barred Spanish mackerel Scomberomorus commersoni. Toxicology 31, 1155-1165. EPA (1995). ‘NSW State of the Environment 1995’. NSW Environment Protection Authority, Sydney. Graneli, E. and Moreira, M.O. (1990). Effects of river water of different origin on the growth of marine dinoflagellates and diatoms in laboratory cultures. Journal of Experimental Marine Biology and Ecology 136, 89-106. Hahn, S.T. and Capra, M.F. (1992). The cyanobacterium Oscillatoria erythraea- a potential source of toxin in the ciguatera food-chain. Food Additives and Contaminants. 9, 351-355. Hallegraeff, G.M (1991). ‘Aquaculturists guide to harmful Australian microalgae’. Fishing Industry Training Board of Tasmania/CSIRO. Division of Fisheries, Hobart. Hallegraeff, G.M. (1992). Harmful algal blooms in the Australian region.’ Marine Pollution Bulletin. 25, 186- 190. Hallegraeff, G.M. (1993). A review of harmful algal blooms and their apparent global increase. Phycologia. 32, 79-99. Hallegraeff, G.M. (1994). Species of the diatom genus Pseudonitzschia in Australian Waters. Botanica Marina 37, 397-411. Hallegraeff, G.M. (1995). Algal blooms in Australian waters. Water, July/August: 19-23. Hallegraeff, G.M. and Reid, D.D. (1986). Phytoplankton species successions and their hydrological environment at a coastal station off Sydney. Australian Journal of Marine and Freshwater Research 37, 361- SiUp Hallegraeff, G.M., Bolch, C.J., Blackburn, S.L. and Oshima, Y. (1991). Species of the Toxigenic dinoflagellate genus Alexandrium in southeastern Australian waters. Botanica Marina 34, 575-587. Hallegraeff, G.M. and Bolch, C.J. (1992). Transport of diatom and dinoflagellate resting spores in ships’ ballast water: implications for plankton biogeography and aquaculture. Journal of Plankton Research 14, 1067-1084. Horiguchi, T. (1996). Haramonas dimorpha gen. et sp. nov. (Raphidophyceae), a new marine raphidophyte from Australian mangrove. Phycological Research 44, 143-150. Huang, C. and Qi, Y. (1997). The abundance cycle and influence factors on red tide phenomena of Noctiluca scintillans (Dinophyceae) in Dapeng Bay, the South China Sea. Journal of Plankton Research 19, 303-318. Jeffrey, S.W. and Hallegraeff, G.M. (1990). Phytoplankton ecology of Australasian waters. In: “Biology of Marine Plants’ (Eds Clayton, M. N., King, R. J. pp. 310-348. (Longman Cheshire: Melbourne). Kahn, S., Arakawa, O. and Onoue, Y. (1998). Physiological investigations of a neurotoxin-producing phytoflagellate, Chattonella marina (Raphidophyceae). Aquaculture Research 29, 9-17. Lapworth, C.J., Hallegraeff, G.M. and Ajani, P.A. (2000) Identification of domoic-acid producing Pseudo- nitzchia species in Australian waters. 9th Int. Conf. Harmful Algal Blooms, Abstracts, p.159. Lee, R., Ajani, P., Krogh, M. and Pritchard, T.R. (2001). Resolving climatic variance in the context of retrospective phytoplankton pattern investigations off the east coast of Australia. Journal of Coastal Research (in press). Proc. Linn. Soc. N.S.W., 123. 2001 P. AJANI, G. HALLEGRAEFF AND T. PRITCHARD 21 Le Messurier, D.J. (1935). A survey of mussels on a portion of the Australian coast. Medical Journal of Australia 1, 490-492. Lincoln-Smith, M.P. and Smith, A.K. (1993). “A survey of cysts of toxic dinoflagellates in sediments in Port Jackson, Botany Bay and Port Kembla’. A report prepared for the Australian Quarantine and Inspection Service, Canberra, Australia. Lu, D., Zhang, Z., Zhu, G. and Lou, Y. (1994). Distributive pattern of Noctiluca scintillans in the Zhejiang coastal water area. Donghai-Mar Sci. Donghai Haiyang. 12,62-69. Marshall, J.A. and Hallegraeff, G.M. (1999). Comparative ecophysiology of the harmful algal Chattonella marina (Raphidophyceae) from South Australia. Journal of Plankton Research 21, 1809-1822. McLachan, A. and Hesp, P. (1984). Surf zone diatom accumulations on the Australian coast. Search, 15, 230- 231. Moestrup, O. (1994). Economic Aspects: “blooms”, nuisance species and toxins. In ‘The Haptophyte Algae’ (Eds. J. C. Green and B. S. C Leadbeater). Systematics Association Special Volume No. 51, pp. 265-285. (Clarendon Press: Oxford). Murray, S. and Suthers, I.M. (1999). Population ecology of Noctiluca scintillans Macartney, a red-tide-forming dinoflagellate. Journal of Marine and Freshwater Research 50, 243-52. Nielsen, M.V. (1993). Toxic effect of the marine dinoflagellate Gymnodinium galatheanum on juvenile cod Gadus morhua. Marine Ecology Progress Series 95, 273-277. Okaichi, T. (1985). Fish kills due to the red tides of Chattonella. Bulletin Marine Science 37, 772. Okaichi, T. and Nishio, S. (1976). Identification of ammonia as the toxic principle of red tide of Noctiluca miliaris. Bulletin of Plankton Society of Japan 23, 25-30. Oviatt, C., Lane, P., French, F. and Donaghay, P. (1989). Phytoplankton species and abundance in response to eutrophication in coastal marine mesocosms. Journal of Plankton Research. 11, 1223-1244. Pan, Y. and Rao, S. (1997). Impacts of domestic sewage effluent on phytoplankton from bedford Basin, Eastern Canada. Marine Pollution Bulletin 34, 1001-1005. Pritchard, T., Lee, R. and Ajani, P. (1999). Anthropogenic and oceanic nutrients in NSW’s dynamic coastal waters and their effect on phytoplankton populations. 14" Australian Coastal and Ocean Engineering Conference and the 7" Australasian Port and Harbour Conference — Coasts and Ports 1999. Porumb, F. (1992). On the development of Noctiluca scintillans under eutrophication of Romanian Black Sea waters. In ‘Marine Coastal Eutrophication’. Proceedings of an international conference, (Eds Vollenweider, R. A., R. Marchetti, R. Viviani). Bologna, Italy. 21-24 March, 1990, pp. 907-918. (Elsevier: Amsterdam). Pritchard, T.R., Lee, R.S. and Ayjani, P. (1997). Oceanic and Anthropogenic Nutrients and the Phytoplankton Response: Preliminary Findings. Pacific Coasts and Ports °97 Proceedings, V1. (CAE University of Canterbury, Christchurch, New Zealand). Qi, Y., Zhang, Z., Hong, Y., Lu, S., Zhu, C. and Li, Y. (1993). Occurrence of red tides on the coasts of China. In ‘Toxic Phytoplankton blooms in the sea’ (Eds T. J. Smayda and Y. Shimizu) pp. 43-46 (Elsevier: New York). Quilliam, M., Eaglesham, G., Hallegraeff, G., Quaine, J., Curtis, J., Richard, D., and Nunez, P. (2000). Detection and identification of toxins associated with a shellfish poisoning incident in New South Wales, Australia. In: ‘Harmful Algal Blooms 2000. Proc. 9th Int. Conf. Harmful Algal Blooms’. (Eds G.M.Hallegraeff, S.I.Blackburn, C.J. Bolch and R.J. Lewis). (IOC. Paris). Rensel, J.E. (1993). Severe blood hypoxia of Atlantic salmon (Salmo salar) exposed to the marine diatom Chaetoceros concavicornis. In: Toxic Phytoplankton Blooms in the sea. (Eds. T. J. Smayda and Y. Shimizu) pp 625-630. (Elsevier: New York). Rensel, J.E., Horner, R.A. and Postel, J-R. (1989). Effects of phytoplankton blooms on salmon aquaculture in Puget Sound, Washington: initial research. Northwest Environmental Journal 5, 53-69. Riegman, R., Noordeloos, A.M. and Cadee, G.C. (1992). Phaeocystis blooms and eutrophication of the continental coastal zones of the North Sea. Marine Biology 112, 479-484. Richardson, D.L. (1991). Some aspects of food and feeding in the Sydney Rock Oyster Saccostrea commercialis in Port Stephens. B.Sc. Honours thesis, University of New England, New South Wales. Smayda, T.J. (1990). Novel and nuisance phytoplankton blooms in the sea: evidence for a global epidemic. In ‘Red tides — Biology, Environmental Science and Toxicology’ (Eds T. Okaichi, D. Anderson and T. Nemots). (Elsevier: New York). Smayda, T.J. (1997). Harmful algal blooms: Their ecophysiology and general relevance to phytoplankton blooms in the sea.’ Limnology and Oceanography 42, 1137-1153. Sournia, A., Chretiennot-Dinet, M.J. and Ricard, M. (1991). Marine phytoplankton: how many species in the world ocean? Journal of Plankton Research 13, 1093-1099. Staver, K.W. and Brinsfield, R.B. (1996). Seepage of groundwater nitrate from a riparian agroecosystem into the Wye River estuary. Estuaries. 19, 359-370. Takano, H. (1956). Harmful blooming of minute cells of Thalassiosira decipiens in coastal waters in Tokyo Bay. Journal of Oceanographical Society of Japan. 12, 63-67. Taylor, F.J.R. (1990). Red tides, brown tides and other harmful algal blooms: the view into the 1990’s. In “Toxic Marine Phytoplankton’ (Eds. E. Graneli et al). (Elsevier: New York). Proc. Linn. Soc. N.S.W., 123. 2001 22 ALGAL BLOOMS IN N.S.W. UNESCO (1995). Manual on harmful marine microalgae. Eds. G.M. Hallegraeff, D.M. Anderson, A.D. Cembella and H.O. Enevoldsen. UNESCO (1996). Design and Implementation of some harmful algal monitoring systems. International Oceanographic Committee Technical Series No. 44. Velikova, V., Moncheva, S. and Petrova, D. (1999). Phytoplankton dynamics and red tides (1987-1997) in the Bulgarian Black Sea. Water Science and Technology. 39, 27-36. Ward, T., Butler, E. and Hill, B. (1998). ‘Environmental indicators for national state of the environment reporting ‘— estuaries and the sea’. State of the Environment (Environmental Indicator Reports), Department of the Environment, Canberra. Whitelegge, T. (1891). On the recent discoloration of the waters of Port Jackson. Recordings of the Australian Museum. 1, 179-192. Yang, C.Z., Albright, L.J. and Yousif, A.N. (1995). Oxygen-radical mediated effects of the toxic phytoplankter Heterosigma carterae on juvenile rainbow trout Oncorhynchus mykiss. Dis. Aquatic Organisms 23, 101-108. Proc. Linn. Soc. N.S.W., 123. 2001 The Coolangatta Latite Member and Associated Tuffs: Newly Identified Basal Units in the Gerringong Volcanics, Southern Sydney Basin, NSW GLEN R. BANN! AND BRIAN G. JONES School of Geosciences, University of Wollongong, Wollongong NSW 2522 Email: glenbann@ozemail.com.au and briangj@uow.edu.au ' Present address: Geo Bio Enviro Expeditions, 5 Mount Street, Aberdeen NSW 2336 Bann, G.R. and Jones, B.G. (2001). The Coolangatta Latite Member and associated tuffs: newly identified basal units in the Gerringong Volcanics, southern Sydney Basin, NSW. Proceedings of the Linnean Society of New South Wales 123, 23-37. Three newly identified stratigraphic units are described from the Broughton Formation in the southern Sydney Basin. They include the Back Forest Tuff Bed, which occurs near the base of the Westley Park Sandstone Member, the Koo-Lee Tuff Member, a laterally extensive pyroclastic ash bed in the middle of the sandstone member, and the Coolangatta Latite Member, a lava flow at the top of the sandstone member. These units form part of the Gerringong Volcanics and precede the previously identified lowest latite member, the Blow Hole Latite Member. The tuff beds consist of shallow marine, pyroclastic deposits, one of which is slightly fossiliferous. The Coolangatta Latite Member consists of a major lava flow at the top of, and a few lava-like lenses interstratified with, the Westley Park Sandstone Member. Evidence indicates a relatively proximal source from volcanoes ranging from mild Strombolian to violently explosive Vulcanian, Plinean and/or Surtseyan phreatomagmatic eruptions. The described deposits represent small or distant components of a much larger volcaniclastic apron surrounding a series of vents which were probably located to the southeast of Mount Coolangatta. Manuscript received 16 January 2001, accepted for publication 22 August 2001. KEYWORDS: Late Permian, Mount Coolangatta, latite, tuff beds, Gerringong Volcanics, Broughton Formation, Sydney Basin, marine, volcaniclastic, volcanism INTRODUCTION Mount Coolangatta, a prominent hill east of Nowra on the south coast of New South Wales (Fig. 1), is composed of early Late Permian (Kungarian-Kazanian) marine and volcanic rocks. These belong to the Berry and Broughton Formations of the Shoalhaven Group in the southeastern Sydney Basin (Table 1). This study provides a detailed stratigraphy of Mount Coolangatta, the only moderately exposed complete outcrop of the Westley Park Sandstone Member in the lower Broughton Formation. Stratigraphic mapping was achieved using an altimeter and 1:25,000 and 1:100,000 scale topographic maps (Sheet 9028). Particular attention was paid to volcanic features, such as lava flows, tuff deposits, and coarse volcaniclastic sandstone and clasts within the sedimentary succession. Fossils were observed and/or collected and identified at all available sites for use in palaeoenvironmental reconstruction. Proc. Linn. Soc. N.S.W., 123. 2001 24 BASAL UNITS IN THE GERRINGONG VOLCANICS Problems encountered in the field included the heavily eroded and weathered nature of many outcrops, the steep heavily vegetated gullies and the pervasive bioturbation that commonly obliterates any trace of primary sedimentary structures in the sedimentary succession. This paper defines three new stratigraphic units, which have been named in accordance with current lithostratigraphic nomenclature in Australia (Staines 1985). The new names have been registered with the AGSO Australian Stratigraphic Names Database, Canberra, ACT. 151° 00’ Wollongong __ ot SEA Be Back Forest mut Coolangatta South SOUTHERN SYDNEY BASIN Pacific @@§ Permian Lavas Figure |. Location map of the Mount Coolangatta area showing the extent of Late Permian shoshonitic lavas (after Carr 1983). Proc. Linn. Soc. N.S.W., 123. 2001 G.R. BANN AND B.G. JONES 25 Table 1 Permian stratigraphic units in the southern Sydney Basin, showing subdivisions of the Shoalhaven Group (modified after Carr, 1983) ILLAWARRA COAL MEASURES Cambewarra Latite Member Dapto Latite Member Jamberoo Sandstone Member Bumbo Latite Member Kiama Sandstone Member Blow Hole Latite Member Coolangatta Latite Member Koo-Lee Tuff Member Back Forest Tuff Bed Westley Park Sandstone Member Berry Formation Nowra Sandstone Wandrawandian Siltstone Snapper Point Formation Pebbley Beach Formation LATE PERMIAN Broughton Formation SHOALHAVEN GROUP EARLY PERMIAN TALATERANG GROUP Geological Setting The Sydney Basin (Fig. 1) is bounded to the east by the Palaeozoic southern New England Fold Belt and its southern extension, the Currarong Orogen (Jones et al. 1984) or Northumberland Ridge (Brakel 1984). To the west it non-conformably and unconformably overlies Early to Middle Palaeozoic magmatic and meta-sedimentary rocks of the Lachlan Fold Belt. The eastern orogen acted not only as a provenance for epiclastic material, but also as a centre for latite extrusions (Harper 1915; Raam 1964, 1969; Mayne et al. 1974; Jones et al. 1984; Shaw et al. 1991; Evans and Migliucci 1991; Veevers et al. 1994a). The Permian Shoalhaven Group (Table 1) forms the main sedimentary fill in the southern Sydney Basin. It consists predominantly of an alternating sandstone and siltstone succession deposited on an extensive open continental shelf subjected to glacio-eustatic sea level changes and variable rates of subsidence (Gostin and Herbert 1973; Herbert 1980, 1995; Jones et al. 1986; Tye et al. 1996; Bann 1998; Eyles and Eyles 1998). Development of a Late Permian volcanic island chain to the east of the present coastline brought about a drastic change in sediment provenance from western quartz-rich craton- derived sediment to volcaniclastic orogen-derived sediment from the east. The first documented sign of this emergent volcanic chain is a tuff emplaced in the upper Wandrawandian Siltstone at Green Point, northeast Jervis Bay (Runnegar 1980b; Shaw et al. 1991; Veevers et al. 1994a). However, Tye et al. (1996) suggested the orogen to the east may have initially emerged at the end of Snapper Point Formation deposition when the palaeo-environment changed from storm-dominated to longshore current-dominated. By the Late Permian, volcanic extrusive and associated volcaniclastic deposits swamped the southeastern Sydney Basin, with lava flows and intrusions typical of the shoshonitic rock association (Joplin 1968; Carr 1985, 1998). Volcaniclastic detritus is a major component in both the Berry Formation and overlying Broughton Formation that are present in the Mount Coolangatta area and represent the last phases of deposition of the predominantly marine Shoalhaven Group. Proc. Linn. Soc. N.S.W., 123. 2001 26 BASAL UNITS IN THE GERRINGONG VOLCANICS The predominantly dark grey fine-grained Berry Formation overlies the Nowra Sandstone gradationally and represents a continuation of the upper Nowra transgression (Le Roux and Jones 1994). It was deposited in a shallow marine environment between the craton in the west and an emerging orogen to the east. Anoxic conditions in and above the substrate are indicated by the sparse fossiliferous layers and low species diversity (Runnegar 1980a; Jones et al. 1986). Increasing volcanic activity in the eastern orogen produced a succession of volcanic-derived sandstones, known as the Broughton Formation, interspersed with latite flows of the shoshonitic Gerringong Volcanics (Harper 1915; Raam 1964; Carr 1983, 1984, 1985). These units represent a regressive phase following the deposition of the Berry Formation. Gerringong Volcanics Carr (1983, 1984) remapped and clarified the stratigraphic nomenclature of the Broughton Formation and Gerringong Volcanics, including a bibliography on the history of the naming of the stratigraphic units. He suggested, based on geochemical evidence, that the Blow Hole Latite Member, the first flow of nine recognised within the Gerringong Volcanic Facies, cropped out on Mount Coolangatta, rather than the Bumbo Latite as previously suggested by Harper (1915). Facer and Carr (1979) and Carr and Facer (1980) assigned K-Ar minimum ages of 253-238 Ma to the Gerringong Volcanics. Roberts et al. (1994), using zircon SHRIMP dates from Queensland and the Newcastle Coal Measures, and correlations based on brachiopod zones, suggested that the Gerringong volcanism is older than 253 Ma. Veevers et al. (1994) suggested a stratigraphic age of 260 Ma and, more recently, Veevers (2000) suggested 265 Ma for the tuff in the Wandrawandian Siltstone based on fossil evidence. Dating of the pyroclastic tuff beds described here from Mount Coolangatta could supply new insights into the timing and onset of volcanism within the southern Sydney Basin. SEDIMENTARY AND PYROCLASTIC FACIES IN THE BERRY AND BROUGHTON FORMATIONS Facies 1 Description: Laminated or massive, bioturbated, light to dark grey, muddy, fine- to medium-grained siltstone. A few areas show preservation of upper plane laminated and pelagic divisions of turbidite beds but most bed boundaries are difficult to discern due to pervasive bioturbation throughout the exposure. The trace fossil assemblage is characteristic of the Cruziana ichnofacies, although it is commonly overprinted by deep tiered Zoophycos burrows. Some micro-cross-bedding is preserved. Pyrite is commonly present in the dark fine-grained siltstone. Fossils and clasts and are either locally common, rare or absent. Large dropstones are absent. Spheroidally (onion-skin) weathered outcrops are a feature. Interpretation: The pervasive overprinting of the Cruziana-type assemblage in Facies 1 suggests slow rates of deposition of suspended fines in a low-energy marine environment around or just below storm wave-base. Most of the succession consists of material settled from suspension but a few micro-cross-beds reflect occasional weak bottom currents in a protected shallow marine environment. This facies represents the lowest depositional energy rates and the highest degree of reduction (anoxia) in the study area which may indicate that it was deposited during maximum relative sea level depths. This facies is only found in the Berry Formation. Facies 2 Description: This facies consists of massive fine- to medium-grained sandstone units. Bed thickness ranges from 0.20 m to >10 m, the average being approximately 0.8 m. No Proc. Linn. Soc. N.S.W., 123. 2001 G.R. BANN AND B.G. JONES 27 relationship between bed thickness and grain size was found. Trace fossils (Cruizana ichnofacies with Zoophycos assemblages), body fossils and clasts are often abundant, as is spheroidal (onion-skin) weathering. Pseudo-cross-lamination (cross-bedding) was sporadically discernible. Interpretation: This facies reflects deposition in an upper offshore environment between storm and fairweather wave-base. The massive nature of the unit reflects complete biogenic homogenisation or, alternatively, rapid deposition from mass-flows. Sediment transport probably occurred either via storm-generated density-current flows of sand from the nearshore environment or the sediment flows may have been triggered by seismic activity (i.e. earthquakes). Facies 3 Description: The facies consists of conglomerate beds and thin lenses with clast sizes reaching up to 15 cm in length. The matrix to clast ratio varies at different localities and the facies ranges from matrix- to clast-supported. Bed thickness varies and is generally less than 40 cm. Basal bed boundaries are often either erosional or diffuse. Clasts are predominantly volcanic in origin. Interpretation: The thin pebbly lenses and lag deposits representative of Facies 3 reflect periods of winnowing, probably during storm events. Fluctuating energy levels within the system caused in situ disturbance above storm wave-base. Facies 4 Description: This facies consists of poorly sorted fine- to coarse-grained sandstone that contains abundant volcaniclastic material, including euhedral plagioclase, subhedral augite and subangular to subrounded microlitic-rich volcanic fragments and clasts. Fossils are also present in this unit. Interpretation: The subangular to subrounded nature of the lithic component of this facies suggests that the fragments have undergone minimal transport. The presence of unaltered augite suggests that deposition took place rapidly (as augite is prone to weathering). Deposition probably occurred rapidly from a series of subaqueous debris flows originating from shallow marine bars, deltas or the subaqueous extension of lahar flows off a volcanic edifice. Facies 5 Description: In the Back Forest area on the southwestern flank of Mount Coolangatta, adjacent to the Shoalhaven River (elevation 30 m), a 2-4 cm thick fine-grained, hard, dark grey layer contains pumice and shard fragments. The unit is conformably interbedded with sandstone beds. The central part of this unit contains a very thin, single layer of coarse-grained lithic fragments up to 1 mm in size. Interpretation: Pumice and shards suggest a volcanic source (explosive eruption). The presence of lithic fragments may indicate that there was a minor depositional hiatus during which time winnowing occurred. Alternatively, the layer may represent a pulse in the eruptive episode with the lithic fragments being autoclastic in origin. The eruptive centre may have been some distance from the site of deposition. Facies 6 Description: Facies 6 consists of a2 m thick, fossiliferous, pyroclastic tuff bed, containing abundant glass shards (reticulite; Fig. 2) and a pseudoeutaxitic texture. Feldspar phenocrysts and biotite are rare. Warthia stricta have been identified within the tuff, preserved in bedding-parallel alignment (A. Wright, pers. comm. 1999). Proc. Linn. Soc. N.S.W., 123. 2001 28 BASAL UNITS IN THE GERRINGONG VOLCANICS Interpretation: This facies is interpreted as a submarine pyroclastic flow originating from a proximal phreatomagmatic (hydrovolcanic) eruption into a shallow marine environment. | 4 ¥ a : a a Figure 2. Chaotic glass shards in a cryptocrystalline vitreous matrix are a prominent feature in thin sections of the Koo-Lee Tuff Member. Plane polarised light. SEDIMENTOLOGY Berry Formation Outcrops of this formation are indistinctly bedded (often as a result of biogenic homogenisation) to flat-bedded, mid- to dark-grey siltstone and very fine feldspathic litharenite (Facies 1). Thin, fine-grained, light-grey to light-brown (i.e. siderite stained) sandstone lenses gradationally and conformably interdigitate with the grey siltstone along the western side of Mount Coolangatta, which represents the upper part of the Berry Formation. The beds become thicker as they grade into the overlying Broughton Formation. Bases of the sandstone beds sometimes contain clasts of intermediate volcanic origin. Primary sedimentary structures are difficult to ascertain, due to intense weathering and pervasive bioturbation. Lamination and micro-cross-bedding is evident in places in the small quarry east of Back Forest. Spheroidal or ‘onion-skin’ weathering, measuring up to 3 m in diameter, is a characteristic feature of the Berry Formation. Regional dip of the Berry Formation is about 2-3° to the northwest. Proc. Linn. Soc. N.S.W., 123. 2001 G.R. BANN AND B.G. JONES 29 Westley Park Sandstone Member Mount Coolangatta contains a complete succession of the Westley Park Sandstone Member, which can be seen to conformably overlie the Berry Formation in a small quarry at Back Forest. The Westley Park Sandstone Member consists of fine- to coarse-grained volcanic-derived litharenite throughout the entire succession, that changes colour from grey or greenish-grey at the base of the member to greenish-brown or pinkish-brown towards the top. Abundant body fossils are preserved at a few locations. Fossils include brachiopods (/ngelerella and a productid), bivalves (Megadesmus), digitate bryozoans and Stenopora crinata. Fossils preserved within one sandstone unit have been zeolitised, possibly caused by the effects of heat associated with volcanism (A. Wright, pers. comm. 1999). A thin tuff is interbedded within the lower part of the member east of Back Forest (56/288250E, 61401250N). On Mount Coolangatta, porphyritic igneous units are intercalated, extrusively and contemporaneously, between sandstone beds of the Westley Park Sandstone Member between about 50 m and 230 m (Fig. 3). Two thin conglomerate beds (Facies 3) overlie the tuff bed of Facies 6 at about 120 m and contain abundant rounded volcanic clasts. A greenish-grey, coarse-grained volcaniclastic sandstone (Facies 4) occurs on the northeast flank of Mount Coolangatta. It contains abundant volcanic clasts up to 1 cm long, which themselves contain plagioclase microlites and augite (3 mm). Brachiopods were also obtained from this site. A breccia unit immediately below the first latite flow marks the top of the Westley Park Sandstone Member. It contains large white tabular plagioclase grains and abundant volcanic fragments. The Westley Park Sandstone Member is approximately 190 m thick at Mount Coolangatta, exceeding the presumed 45 m suggested by Raam (1964, 1969) and Carr (1983). Kiama Sandstone Member For some distance above the Blow Hole flow the outcrop has been removed by weathering or concealed by vegetation and the lateral equivalents of the fluvial deposits and permafrost palaeosol reported from near Kiama by Retallack (1999) could not be observed. The latter palaeosol from near the base of the Kiama Sandstone Member contains abundant feldspar and pyroxene, and provides evidence of a humid frigid climate, with a boreal taiga type forest (Retallack 1999). Towards the summit of Mount Coolangatta (above 250 m) the Kiama Sandstone Member is at least 55 m thick, which is similar to the 53 m previously recorded at Jamberoo by Carr (1983). It is a characteristic pinkish brown to red (hematitic) silty marine sandstone showing some evidence of bioturbation. Brachiopods were identified within this unit, confirming its shallow marine origin. VOLCANIC ASH AND TUFF DEPOSITS Back Forest Tuff Bed A thin tuffaceous ash deposit (Facies 5, averaging about 3 cm thick) was found in a disused roadside quarry near Back Forest Bridge (56/288250E, 61401250N), conformably interbedded within fine-grained sandstone near the base of the Broughton Formation. It has been named after the Back Forest area. Fragments of pumice with vesicles were identified in thin-section, as are the remains of glass shards contained in a pseudo-eutaxitic texture. Tiny fragments of latitic material and angular anhedral phenocrysts of plagioclase are also present. This deposit could not be traced laterally due to soil and vegetation cover. The tuff deposit is evidence of an explosive volcanic eruption, previously undetected in the southern Sydney Basin. Proc. Linn. Soc. N.S.W., 123. 2001 30 BASAL UNITS IN THE GERRINGONG VOLCANICS 290 Kiama Sandstone Member pink-red medium-grained sandstone, POCny, edded, bioturbat Blow Hole Latite Member fine-grained foliated latite weathered contact zone Coolangatta Latite Member porphyritic latite with aligned lagioclace phenocrysts to 4cm fn Panesraned Bro aniaee ~7] Wavy contact “| conglomerate bands, volcanic clasts ‘;] Massive lithic sandstone Westley Park Sandstone Member oe fine-grained lithic sandstone / | porphyitic latite lens bioturbated fine-grained sandstone rare brachiopods A Koo Lee Tuff Member “| fine-grained lithic sandstone few fossils & basalt clasts | vesicular basalt, common fossils fine- to medium-grained poorly bedded lithic sandstone, bioturbated, rare ripples "| few coarse-grained lenses few porphyritic latite boulders “22: fine-grained lithic sandstone Back Forest Tuff Bed “:| fine- to coarse-grained lithic sandstone /] well bedded, minor bioturbation Berry Siltstone rare thin sandy interbeds Figure 3. A detailed composite stratigraphic section through the Westley Park to Kiama Sandstone Members, lower Broughton Formation, on the eastern and southern flank of Mount Coolangatta. Proc. Linn. Soc. N.S.W., 123. 2001 G.R. BANN AND B.G. JONES 31 Koo-Lee Tuff Member A 2 m thick volcanic ash deposit (Facies 6) was found in a gully on the eastern side of Mount Coolangatta at an elevation of 120 m (56/291250E, 6141300N; type section) and also in gullies on the northern side of the mountain at approximately the same elevation. It is named after the Koo-Lee property on which it occurs. It is interpreted to occupy a position about 85 m above the base, and within the middle of the Westley Park Sandstone Member. This places the occurrence of this second volcanic ash event well before the previously documented initial flow of the Blow Hole Latite Member in the Gerringong Volcanics. The volcanic ash consists of reticulite, which contains chaotic glass shards mixed in a cryptocrystalline vitreous matrix with subordinate euhedral-subhedral opaque grains (magnetite), rare subhedral plagioclase with a perthitic texture, and small pleochroic biotite (Fig. 2). In places a eutaxitic texture is exhibited. Pumiceous material, a strong foliation parallel to bedding, and a sparse shelly fauna are also present. Fossils found within the tuff have been identified as the small marine gastropod Warthia stricta (Dana, 1849). This particular fossil was identified in the middle of the bed; however, other fossil fragments can be recognised throughout the tuff. These small gastropods are one of three Permian Warthia species found in the area, and they have also been identified in Permian strata in New Zealand (Waterhouse 1967). The presence of the fossils indicates that the tuff was emplaced in a shallow marine environment probably by a flow mechanism or as a primary air-fall ash. LAVA FLOWS In 1849 Dana suggested that the igneous rocks high on Mount Coolangatta were ‘similar to the Kiama lavas’, which crop out at sea level 12 km to the north. In 1915 Harper suggested the Bumbo flow occurred on the northern side of Mount Coolangatta while the southernmost outlier of the Blow Hole flow was at Toolijooa (KG965505) some 10 km northeast of Mount Coolangatta. In this study two lava flows are recognised above the Westley Park Sandstone Member on Mount Coolangatta, and thin flows are present at lower stratigraphic levels within the member. Identification of a latite rock unit, 40 to 50 cm thick, was made low on Mount Coolangatta (elevation of approx 20 m) in a small gully covered with lantana (56/290675E 6141000N). This lower latite is conformable with the underlying and overlying sandstone strata. It is a dark greenish grey porphyritic latite with large preferentially eroded-out phenocrysts of plagioclase (labradorite) in a fine-grained groundmass. This same type of volcanic rock appeared again higher up on Mount Coolangatta, in a series of thin units interbedded with the sandstone (Fig. 3). Coolangatta Latite Member Immediately above the breccia at the top of the Westley Park Sandstone Member is a 30-40 m thick, dark grey, trachytic porphyritic latite extrusion, with plagioclase phenocrysts up to 4 cm in length (Fig. 4). This distinctive lower flow has been named the Coolangatta Latite Member since its major occurrence is on Mount Coolangatta. The type section (56/291300E, 6141500N) is located on Mount Coolangatta between 204 m and 226 m elevation (Fig. 3). The Coolangatta Latite Member has a similar texture to the porphyritic stage of the Bumbo flow (P. Carr, pers. comm. 1999), being pilotaxitic. In hand specimen the rock is dark grey and fine-grained with plagioclase and a dark mineral (iron oxide or augite) present. It exhibits a porphyritic texture with large tabular plagioclase (labradorite) phenocrysts up to 4 cm long, that exhibit excellent twinning according to albite, pericline and albite-Carlsbad laws. Smaller, skeletal, twinned phenocrysts exhibit a glomeroporphyritic texture. Part of the member has a pilotaxitic (trachytic) texture with microlites exhibiting a lineation or pronounced flow texture. Plagioclase microlites make up the majority of the groundmass with subordinate pyroxene Proc. Linn. Soc. N.S.W., 123. 2001 32 BASAL UNITS IN THE GERRINGONG VOLCANICS (augite) and opaque iron oxides (magnetite or/and hematite). Chlorite is present throughout the groundmass, as well as replacing mafic minerals (augite and possibly olivine). The fine-grained groundmass infiltrates cracks along twin planes in some of the large phenocrysts. This rock appears to be a latitic basalt. The underlying breccia would have formed as a basal breccia layer when the lava flowed into the shallow marine environment represented by the upper part of the Westley Park Sandstone Member. Figure 4. Coolangatta Latite Member showing large partly aligned plagioclase phenocrysts. The hammer handle is 35 mm wide. Blow Hole Latite Member The Blow Hole Latite Member consists of fine-grained slightly porphyritic latite occurring between 226 m and 242 m elevation near the top of Mount Coolangatta (Fig. 3). This latite unit exhibits a northwest-dipping flow foliation (Fig. 5) and an alignment of the phenocrysts present. However, it lacks the conspicuous large plagioclase phenocrysts found in the underlying flow. It is petrographically and geochemically the same as the remainder of the Blow Hole Latite Member (Carr 1984) and represents the southern extremity of this flow. Proc. Linn. Soc. N.S.W., 123. 2001 G.R. BANN AND B.G. JONES 33 Figure 5. Typical finely crystalline Blow Hole Latite Member showing flow foliation (dipping to the north). DISCUSSION The Permian flows recorded from the Mount Coolangatta area have a porphyritic or glomeroporphyritic texture that is characteristic of coherent lavas, syn-volcanic intrusions and derived clasts. This criterion is critical when distinguishing coherent facies (lava flows) from pyroclastic, resedimented volcaniclastic and volcanogenic sedimentary deposits (Walker 1973). Volcanic rock units of similar age may have disparate petrographies due to different cooling histories of magma from the same magma chamber (i.e. successive eruptions with and without large phenocrysts). Plagioclase phenocrysts in the two flows commonly have a lighter, more sodic-rich rim than their core, creating a corona, whilst the rarer alkali feldspars often have a more calcium-rich rim. This chemical zoning may be due to albitisation. Sericitisation is indicated by sericite inclusions in plagioclase cores while chlorite and calcite can replace both groundmass and phenocrysts, including plagioclase, olivine and augite. Proc. Linn. Soc. N.S.W., 123. 2001 34 BASAL UNITS IN THE GERRINGONG VOLCANICS Cas and Wright (1987) and McPhie et al. (1993) suggested that flat and slightly curved, platy glass shards, similar to those in the Koo-Lee Tuff Member, are of a Surtseyan type. Such shards are produced when sea water acts to catalyse explosive fragmentation of an already vesiculated, basaltic magma that is in a state of incipient magmatic explosive disruption. The abundant shards within the Koo-Lee Tuff Member suggests that a phreatomagmatic Vulcanian, Plinian or Surtseyan eruption to the south or southeast may have produced a surge that travelled across the water in a northward direction, depositing the Koo-Lee Tuff Member near the vent. Rapid deposition into a shallow marine environment is suggested by the homogenous vitric shard content (i.e very little sediment interbedded or mixed in), as well as the containment of Warthia (gastropod) and fossil debris within the middle part of the flow. The presence of Warthia suggests that the source location was proximal and the animals were buried either in situ or only transported a short distance. Ash fall deposits are often fossil- rich proximal to the source or wherever the tuff accumulation is thick (Lockley 1990). Some stratification in the tuff (i.e. variation in shard size or ‘bedding’ ) may reflect a progressive increase and/or fluctuation in eruptive vigour, or an interruption in the continuity of the eruption (Cas and Wright 1987; Scandone 1996), caused by temporary vent blockage, or rain showers during the eruption. Volcanic Activity The Back Forest Tuff Bed provides evidence for periodic volcanic activity during deposition of the lower Westley Park Sandstone Member. Incoming volcanic detritus led to the gradational change from the Berry Formation to the Westley Park Sandstone Member as volcaniclastic supply and proximity to volcano sources increased. A probable Surtseyan pyroclastic eruption deposited a thick ash-flow, the Koo Lee Tuff Member, in proximal channels and distally as an ash-fall or a pyroclastic flow. Less violent eruptions followed, with extrusion of fluidal lava (latite) that flowed down the volcano flanks and into the sea formed the Coolangatta and Blow Hole Latite Members. Mafic lava can flow significant distances downslope, and underwater (Walker 1989, 1993) where the lava then accumulates due to reduction in slope at the foot of the volcano. The Westley Park Sandstone Member at Mount Coolangatta is interpreted as a sedimentary apron about a volcanic vent(s) on the northwest edge of the magmatic arc in the Currarong Orogen. It may represent a small or more distant component of a larger volcaniclastic apron surrounding a substantial volcano tentatively located 40 km to the southeast (i.e. in or southeast of Jervis Bay). The greater thickness (i.e. >150 m) of the Westley Park Sandstone Member on Mount Coolangatta than elsewhere (maximum 45 m) suggests that the local units represent a wedge of sediment deposited during periods of high energy (shallower water) as the apron extended northwards. Abundant volcaniclastic material would have been shed off the volcano into the basin as lahars, debris flows and turbidites. The thick feldspar-phyric Coolangatta Latite Member flow was extruded onto unconsolidated wet sand and formed a breccia at the sediment-lava interface. The Blow Hole Latite Member was emplaced as a slightly younger flow before subsidence/ transgression occurred and the Kiama Sandstone Member was deposited. The relatively rapid rate of extrusion (i.e. 40 m of lava into a shallow marine environment) suggests a potential subaerial surface existed after the second flow. Evidence of subaerial weathering or sedimentation above the Blow Hole flow was not found (cf. Retallack 1999), however, benching near this contact on Mount Coolangatta indicates preferential erosion of the beds at this stratigraphic height. A transgression noted across the subaerial palaeosol on the Blow Hole flow near Kiama (Retallack 1999) suggests that the region had a low gradient, allowing reworking at the top of the flow that contributed to the deposition of the overlying marine Kiama Sandstone Member. Proc. Linn. Soc. N.S.W., 123. 2001 G.R. BANN AND B.G. JONES 35 Palaeogeography Sedimentation was drastically influenced by volcanoes to the south and southeast supplying voluminous volcanic material that overwhelmed the westerly- derived quartzose sediments from the Lachlan Fold Belt. The volcanoes may have aided mortality of fauna living in the shallow Permian seas at that time. Glacio-eustatic cycles, mainly due to Milankovitch cyclicity, have affected Permian sedimentation in the southern Sydney Basin (e.g. Tye 1995; Barry 1997; Bann 1998). In the Mount Coolangatta area, the interpretation of sea level changes and flooding events on sedimentation is complicated by local volcanicity (i.e. regressions before eruptions due to magma chamber bulge and subsequent transgressions after eruption). The increased amount of sedimentation in volcanic settings due to rapid erosion of pyroclastic detritus (Orton 1995) may have choked the basin and enforced a regression. Vents and flows are also likely to reduce wave activity, alter palaeocurrent trends and deflect craton-derived sediment to other areas of the basin. Other tectonic influences varied the rate of subsidence, perhaps by foreland loading to the east under orogen stresses (i.e. subduction) or changes in basin hinge zone location. CONCLUSIONS Subaerial and subaqueous volcanic events and contemporaneous erosion of the volcanic edifices featured during the mid Permian in the southern Sydney Basin. The eastern Berry Formation contains abundant volcanic detritus, shed as turbidite flows off the flanks of a southern volcano near Jervis Bay. The stratigraphic succession graded upwards conformably from the Berry Formation into the Broughton Formation as the influx of volcaniclastic sand increased. The Back Forest Tuff Bed within the lower Westley Park Sandstone Member contains latitic clasts, pumiceous material and glass shards, thus marking an explosive volcanic eruption. The Koo-Lee Tuff Member formed as a proximal, slightly fossiliferous pyroclastic vitric tuff that recorded a more violent Vulcanian, Plinian or Surtseyan phreatomagmatic eruption. Warthia fossils within the tuff bed reflect its submarine deposition. The Coolangatta Latite Member, a thick porphyritic mafic flow, records an early phase of lava activity within the Gerringong Volcanics that predates the Blow Hole flow. Penecontemporaneous sedimentation included considerable volcaniclastic sediment derived from rapidly eroding volcanic vents and flanks. Abundant fine- grained volcanic material dominated over the westerly-derived quartzose sediments of the Lachlan Fold Belt. Depositional mechanisms were mostly pyroclastic surges and flows, storm-generated turbidity currents, and debris flows triggered by seismic activity (earthquakes). ACKNOWLEDGEMENTS We acknowledge the School of Geosciences, University of Wollongong, for providing research facilities. We would also like to acknowledge useful discussions with, and comments from Kerrie Bann, Paul Carr, Pat Conaghan, Dick Flood, Stuart Tye, John Veevers and Tony Wright. We also thank the journal referees for their useful suggestions for improving the paper. Richard Miller drafted the figures. Proc. Linn. Soc. N.S.W., 123. 2001 36 BASAL UNITS IN THE GERRINGONG VOLCANICS REFERENCES Bann, K. L. (1998). Ichnology of the Early Permian Pebbly Beach and Snapper Point Formations, southern Sydney Basin. PhD thesis, University of Wollongong, Wollongong. Barry, S. C. (1997). Magma-sediment interaction in the Late Permian upper Broughton Formation of the southern Sydney Basin, N.S.W. BSc (Hons.) thesis, University of Wollongong, Wollongong. Brakel, A. (1984). Sydney Basin: Permian coal measures stratigraphy and sedimentation. /n Harrington H. J. and others, ‘Permian coals of eastern Australia’. Bureau of Mineral Resources, Geology and Geophysics, Australia, Bulletin 231, 9-41. Carr, P. F. (1983). A reappraisal of the stratigraphy of the upper Shoalhaven Group and lower Illawarra Coal Measures, southern Sydney Basin, N.S.W. Proceedings of the Linnean Society of NSW. 106, 287-297. Carr, P. F. (1984). Late Permian shoshonitic province of the southern Sydney Basin. PhD thesis, University of Wollongong, Wollongong. Carr, P. F. (1985). Geochemistry of Late Permian shoshonitic lavas from the southern Sydney Basin. In “Volcanism in eastern Australia with case histories from New South Wales’ (Eds F. L. Sutherland, B. J. Franklin and A. E. Waltho). Geological Society of Australia, N.S.W. Division Publication 1, 165- 183. Carr, P. F. (1998). Subduction related Late Permian shoshonites of the Sydney Basin, Australia. Mineralogy and Petrology 63, 49-71. 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. Cas, R. A. F. and Wright, J. V. (1987). ‘Volcanic Successions: Modern and Ancient’. (Allen and Unwin: London). Dana, J. D. (1849).’ ‘Geology’. Report of the United States exploring expedition during the years 1838-1842 under the command of Charles Wilkes 10, 481 pp. Evans, P. R. and Migliucci, A. (1991). Evolution of the Sydney Basin during the Permian as a foreland basin to the Currarong and New England Orogens. 25th Newcastle Symposium, Advances in the Study of the Sydney Basin, p. 27-29. (Department of Geology, University of Newcastle). Eyles, N. and Eyles, C. H. (1998). Facies and allostratigraphy of high-latitude, glacially influenced marine strata of the Early Permian southern Sydney Basin, Australia. Sedimentology 45, 121-161. Facer, R. A. and Carr, P. F. (1979). K-Ar dating of Permian and Tertiary igneous activity in the south-eastern Sydney Basin, N.S.W. Journal of the Geological Society of Australia 26, 73-79. Gostin, V. A. and Herbert, C. (1973). Stratigraphy of the Upper Carboniferous and Lower Permian sequence, southern Sydney Basin. Journal of the Geological Society of Australia 20, 49-70. Harper, L. F. (1915). Geology and mineral resources of the Southern Coalfield. Part 1. The south coastal portion. Geological Survey of New South Wales Memoir 7, 410 pp. Herbert, C. (1980). Depositional development of the Sydney Basin. Geological Survey of New South Wales Bulletin 26, 10-53. Herbert, C. (1995). Sequence stratigraphy of the Late Permian coal measures in the Sydney Basin. Australian Journal of Earth Sciences 42, 391-405. Jones, B. G., Gostin, V. A. and Dickins, J. M. (1986). ‘Excursion 1C Southern Sydney Basin’.12th International Sedimentological Congress Field Guide, Canberra, 41 pp. Jones, J. G., Conaghan, P. J., McDonnell, K. L., Flood, R. H. and Shaw, S. E. (1984). Papuan Basin analogue and a foreland basin model for the Bowen-Sydney Basin. Jn ‘Phanerozoic Earth History of Australia’, (Ed. J. J. Veevers) pp. 243-261. (Clarendon Press: Oxford). Joplin, G. A. (1968). The shoshonite association: a review. Journal of the Geological Society of Australia 15, 275-294. Le Roux, J. P. and Jones, B. G. (1994). Lithostratigraphy and depositional environment of the Permian Nowra Sandstone in the southwestern Sydney Basin, Australia. Australian Journal of Earth Sciences 41, 191-203. Lockley, M. G. (1990). How volcanism affects the biostratigraphic record. Geological Society of America, Special Paper 244, 1-12. Mayne, S. J., Nicholas, E., Bigg-Wither, A. L., Rasidi, J. S. and Raine, M. J. (1974). Geology of the Sydney Basin- a review. Bureau of Mineral Resources, Geology and Geophysics, Australia Bulletin 149, 229 Pp- McPhie, J., Doyle, M. and Allen, R. (1993). Volcanic Textures. A Guide to the Interpretation of Textures in Volcanic Rocks, 198 pp. Centre for Ore Deposits and Exploration Studies, University of Tasmania, Hobart. Orton, G. J. (1995). Facies models in volcanic terrains: times arrow versus times cycle. International Association of Sedimentologists Special Publication 22, 157-193. Raam, A. (1964). Geology of the Minnamurra-Gerroa area. BSc (Hons) thesis, University of Sydney, Sydney. Raam, A. (1969). Gerringong Volcanics. Journal of the Geological Society of Australia 16, 366-368. Retallack, G. J. 1999. Permafrost palaeoclimate of Permian palaeosols in the Gerringong Volcanic Facies of New South Wales. Australian Journal of Earth Sciences 46, 11-22. Proc. Linn. Soc. N.S.W., 123. 2001 G.R. BANN AND B.G. JONES 37 Roberts, J., Briggs, D., Claque-Long, J., Foster, C. and Maidment, D. (1994). Permian SHRIMP ages and correlations within eastern Australia. 28th Newcastle Symposium, Advances in the Study of the Sydney Basin, p. 24-30. (Department of Geology, University of Newcastle). Runnegar, B. (1980a). Marine Shoalhaven Group, southern Sydney Basin. Geological Survey of New South Wales Bulletin 26, 74-81. Runnegar, B. (1980b). Excursion guide to Montague Roadstead, Jervis Bay, southern Sydney Basin. Jn ‘Guide to the Sydney Basin’ (Eds C. Herbert and R.A. Helby). Geological Survey of New South Wales Bulletin 26, 525-530. Scandone, R. (1996). Factors controlling the temporal evolution of explosive eruptions. Journal of Volcanology and Geothermal Research 72, 71-83. Shaw, S. E., Conaghan, P. J. and Flood, R. H. (1991). Late Permian and Triassic igneous activity in the New England Batholith and contemporaneous tephra in the Sydney and Gunnedah Basins. 25th Newcastle Symposium, Advances in the study of the Sydney Basin, p. 44-51. (Department of Geology, University of Newcastle). Staines, H. R. E. (1985). Field geologist’s guide to lithostratigraphic nomenclature in Australia.’ Australian Journal of Earth Sciences 32, 83-106. Tye, S. C. (1995). Stratigraphy, sedimentology and tectonic significance of the Talaterang and Shoalhaven Groups in the southern Sydney Basin. PhD thesis, University of Wollongong, Wollongong. Tye, S.C., Fielding, C. R. and Jones, B. G. (1996). Stratigraphy and sedimentology of the Permian Talaterang and Shoalhaven Groups in the southernmost Sydney Basin, N.S.W. Australian Journal of Earth Sciences 43, 57-69. Veevers, J. J., Conaghan, P. J. and Powell, C. McA. (1994). Eastern Australia. Jn ‘Permian-Triassic Pangean basins and foldbelts along the Panthalassan margin of Gondwanaland’ (Eds J. J. Veevers and C. McA. Powell). Geological Society of America Memoir 184, 11-171. Veevers, J. J. (2000). Billion-Year Earth History of Australia and Neighbours in Gondwanaland, 388 pp. GEMOC Press, Macquarie University, Sydney. Walker, G. P. L. (1973). Lengths of lava flows. Philosophical Transactions of the Royal Society of London Series A2, 107-118 Walker, G. P. L. (1989). Gravitational density controls on volcanism, magma chambers and intrusions. Australian Journal of Earth Sciences 36, 149-165 Walker, G. P. L. (1993). Basaltic volcano systems. In ‘Magmatic processes and plate tectonics’ (Eds H.M. Prichard, T. Alabaster, N.B.W, Harris and C.R. Neary). Geological Society of London, Special Publication 76, 3-38. Waterhouse, J. B. (1967). Permian gastropods of New Zealand part!. Bellerophontacea and Euomphalacea. New Zealand Journal of Geology and Geophysics 6, 88-112. Proc. Linn. Soc. N.S.W., 123. 2001 , SL aaes iS cK ony eye Sak - NODS DGS +e eh ai ei “aac Hin} She Ipaait. at SR ti) ¥s. < 4 in F Peni conn Tpeaeal Te sane ee EO Oe eee 9 Ps blown Pith ak ey, recite ar tse hed Massie No ADTS $e ea) 1.7 ee ia 40 by! 40 inntndan Pyass: eer ine 5, ae e ie iA. SOS TIO aacasth mi 3 eb ele Ms ravi or isa? ie pes +3) ae TAL 4 ‘ae , 0 = honk op i eo bee ba Rea, 3,3 BAO Tipe oa ey Bes caltiqsiiates igual isan fA vitoerslenie Hite i aR Whe of th: implore CUTS $ a R23) Beat, fi AWE wh. Aubieraiis fe i addcsad is Rati up ys en q A, r Locktic Meat thi Buty mules inven affect tye bite goad fein & Grokigicn Sra ti Pare!) 2d4 a - ; Abiyate, fh), Muchwilinny ih) Bigg Wine, | aa ot Funidi, LP * “asd winia he Pi ita Fi, ae Aagip arrive fin): what H Nive wee Renan Croey-and. Sincyhsie's Lt mestsit : ue. ; i : MePhic. 1. Trivia MM). ond ‘Alte i, a iw Nien Pretisiess < ‘Gian Wilt Pods, [OR pi Cantre ve th ye Departs ai ihe retrii © haart, ae Baw, 5.0 Pucks rcidelebarrinaa: siachenateede “elon | ef Sealdimeatetugsen Special Pn 2 142-193, " i Gand, A ViGtk) Coding’ of thie Mitinmaaring, ore area. BSc uh tam A ge sae Vestine, Piavendl he: dba” Cie, tp U's vrnetbapet SeqhatowTNadeiber ch ASP - ~vie foal] Wales ‘ate ne a ah Ae “4 Ue adh, rie face Mee AIT a. rave The Middle Triassic Megafossil Flora of the Basin Creek Formation, Nymboida Coal Measures, New South Wales, Australia. Part 2. Filicophyta W.B. KeitH HoLMEs ‘Noonee Nyrang’, Gulgong Road, Wellington NSW 2820, Australia (Hon. Research Fellow, Geology Department, University of New England, Armidale, NSW 2351, Australia) Holmes, W.B.K. (2001). The Middle Triassic megafossil flora of the Basin Creek Formation, Nymboida Coal Measures, New South Wales, Australia. Part 2. Filicophyta. Proceedings of the Linnean Society of New South Wales 123, 39-87. Ferns and fern-like foliage comprise 25% of the collections from the Middle Triassic Basin Creek Formation at Nymboida. The diverse fern flora ranges in size from tiny forms a few centimetres in height, to large ferns with fronds equalling in length those of the largest extant treeferns. This paper describes ferns preserved as fertile fronds, fertile fronds with closely associated sterile fronds or sterile fronds of known affinities. The Marattiales is represented by eight taxa, the Osmundales by two taxa, the Filicales by three taxa and one taxon is of uncertain position. Ten new taxa are erected for fertile fronds and four species are ascribed to previously published material. New forms include Rhinipteris walkomii sp. nov., Asterotheca trullensis sp. nov., A. nymboidensis sp. nov., A. chevronervia sp. nov., A. diameson sp. nov., Herbstopteris colliveri (Herbst) gen. et comb. nov., Todites parvum sp. nov., Osmundopsis scalaris sp. nov., Hausmannia reticulata sp. nov. and Nymbofelicia aggregata gen. et sp. nov. A paper in preparation will describe many additional morpho-taxa of sterile fern fronds and fern-like foliage which cannot be systematically classified. Manuscript received 20 March 2001, accepted for publication 24 October 2001. KEYWORDS: Palaeobotany, Middle Triassic flora, Nymboida Coal Measures, Filicophyta, megafossil fertile ferns. INTRODUCTION In this second part of a series on the Nymboida Flora, ferns are described which are known from fertile fronds, fertile fronds with closely associated sterile fronds or sterile fronds that can be confidently assigned to known fern taxa. Ten new taxa are erected for fertile specimens and four species are ascribed to previously described material. Numerous other specimens comprising sterile fern-like foliage of doubtful affinities will be described in a future part of this series. Fossil ferns have been reported previously from the Nymboida Coal Measures by de Jersey (1958), McElroy (1963), Flint and Gould (1975), Retallack (1977), Retallack et al. (1977), Herbst (1977a, 1977b, 1978) and Webb (1982, 1983, 2001). Details of the Coal Mine and Reserve Quarries which were the source of the Nymboida Flora are provided in Holmes (2000). The present collections reveal further evidence in the shales and siltstones of autochthonous and semi-autochthonous AO TRIASSIC FLORA FROM NYMBOIDA — FILICOPHYTA assemblages derived from levee forests, floodplains and swamps somewhat similar to those reconstructed for the Molteno Formation of South Africa by Anderson et al. (1998) and MacRea (1999). The high energy sandstones and grits contain allochthonous assemblages of point bar and riverbank vegetation (Anderson and Anderson 1983; Retallack 1977). The Nymboida Sub-Basin palaeoenvironment was reconstructed by Retallack (1977) as a riverine floodplain. The rarity of exposed red-bed horizons in the Nymboida quarries suggests that a high water table prevailed during most of the period of deposition (Retallack et al. 1996). On the basis of the clay-mineral content in many of the sediments, McElroy (1963) suggested that the Sub- Basin represented a ponded environment. Ferns and fern-like foliage are common in most fossiliferous horizons at these two quarries. In my collections made over the last 35 years, ferns and fernlike material comprise approximately 25% of the catalogued material. Due to preservation, collecting and other biases, this figure does not necessarily represent the proportion of ferns in the living vegetation. In size, the Nymboida ferns range from tiny plants measuring only a few centimetres in length to fronds three metres or more long which equal in size some of the largest extant southern ferns such as Angiopteris, Cyathea and Dicksonia. Despite the climatic implications inferred by the palaeolatitude c. 60°S for the Nymboida Sub-Basin at the time of its deposition (Anderson and Anderson 1983; Scotese 1994)) the flora suggests a moist temperate climate. The number and diversity of ferns in the present collection is a remarkable demonstration of the recovery of this group of plants following the catastrophic end-Permian extinction event and the subsequent “coal gap” during the Early Triassic (Retallack et al. 1996). The classification adopted in this paper broadly follows that of Boureau (1975) and Meyen (1987). As noted by Meyen (1987), “the classification and systematics of the ferns is in a state of flux and in the fossil state where often only dispersed portions of fronds are available, the difficulties are even greater”. From my extensive Nymboida collections, it has been possible to link some sterile and fertile fronds. On rare occasions, these have been found found attached to the same frond. This paper and another in preparation will provide, for the first time, a comprehensive picture of the diversity of ferns occurring in Australia during the Middle Triassic. Due to the great dissimilarities between the Gondwana and Northern Hemisphere floras during the Middle Triassic (Archangelsky 1965; Meyen 1987), at least at species level, the Nymboida taxa are compared in most cases only with other Gondwana material. This series of papers, which deals only with the descriptive taxonomy of the plants present in the Nymboida Flora, will provide a reference base for future taphonomic, palaeoecological and allied studies. The Nymboida specimens were collected mostly from fallen blocks that had been blasted or mechanically excavated from the vertical quarry faces. Thus, the exact horizons from which the material originated is, in most cases, unknown. I have not tried to recreate vegetational palaeodemes in the sense of Anderson and Anderson (1983). Many blocks reveal two or more different taxa, so it may be possible to match matrices and taxa preserved to reconstruct plant associations such as those of Anderson et al (1998) and MacCrea (1999) for the Molteno Flora. As noted by Holmes (2000) the Nymboida fossil flora provides a unique window onto the mozaic of plant communities and successions that occupied an area of 2.5 hectares during a brief period of geological time c. 237 million years ago (Retallack et al. 1993). Most specimens are preserved as carbonaceous compressions in which the gross morphology is from well- to exquisitely preserved. However spores and cellular details have been destroyed by a tectonic heating event during the Cretaceous Period (de Jersey 1958; Hennelly, in McElroy 1963; Russell 1994). The institutions holding Nymboida fossil plant material are listed as — AM, Australian Museum, Sydney, NSW; UNE, University of New England, Armidale, NSW; QM, Queensland Museum, Brisbane, Queensland. Proc. Linn. Soc. N.S.W., 123. 2001 W.B.K. HOLMES 41 Specimens on the Figures are illustrated at natural size unless otherwise stated. Figures are located in a separate section at the back of this paper. SYSTEMATIC PALAEOBOTANY Phylum Filicophyta Order Marattiales Family Marattiaceae Genus Ogmos Webb 1983 Ogmos adinus Webb 1983, p.364. Figure 1A-C 1924 Taeniopteris (?Danaeopsis) crassinervis Walkom, Pl. 18, Figs | and 2 only. 1975 Taeniopteris crassinerva Flint and Gould, Pl. 2, Figs 7, 8. 1983 Ogmos adinus Webb, p.364, Figs 2-10. Description Fronds simple, large, to 400 mm long and 80 mm wide, elongate-spathulate to narrow elliptical, margin entire, midrib broad to 10 mm wide near base, but leaf base not known. Apex acuminate to broad obtuse or emarginate. Lateral venation very coarse, straight, attached at c. 80°-90° in basal half then becoming slightly more acute towards the apex. Veins simple or rarely once-forked; 5 to 12 per 10 mm. Material AMF113382- 113386, UNEF14125-6. Coal Mine Quarry, Nymboida. Discussion Ogmos adinus is a rare element at Nymboida having been collected on three occasions only. In one small shale lens the numerous leaves lying in close proximity or overlapping (Fig. 1B) suggested a growth form similar in life to that of the extant Asplenium nidus. The fertile material of Ogmos adinus from the Bryden Beds of the Esk Trough (Webb 1983) showed sporangia forming continuous single file rows from the midrib to the margin and parallel with the lateral veins. At Nymboida there are no fertile leaves present, but the size and venation pattern of the sterile leaves is closely similar to that of the sterile leaves illustrated by Webb (1983). Genus Marantoidea Jaeger 1827 Type species Marantoidea arenacea Jaeger 1827 Marantoidea acara Webb 2001 Figs 2A; 3A,B. 2001 Marantoidea acara Webb (this volume) Description Fronds are relatively large (Fig. 2A), with a rachis 6-7 mm wide from which opposite to subopposite pinnae diverge at angles of 45°-60°, 3-5 cm. apart. Individual pinnae are at least 18 cm long and 20-30 mm wide and taper gradually to an acute apex. At the base of each pinna the upper pinna margin is markedly contracted, whereas the lower margin is decurrent on the rachis. Midveins of pinnae are to 2 mm wide; secondary veins diverge at very acute angles but curve away almost immediately and run fairly straight and parallel to reach the margin at an angle of 60°-80° to the midrib. Most secondary veins fork once close to the midrib, rarely a second time; then anastomose with an adjacent vein near the margin, but this character is difficult to observe. Density of venation varies Proc. Linn. Soc. N.S.W., 123. 2001 42 TRIASSIC FLORA FROM NYMBOIDA — FILICOPHYTA from 12-14 veins per 10 mm in sterile fronds. In fertile material the veins are denser and vary from 18-20 per 10 mm. On fertile pinnae (Figs 3A, B) the underside of the leaf is covered with small (0.2 x 0.15 mm) ellipsoidal sporangia, with their long axes perpendicular to the lateral venation. Each sporangium has a prominent longitudinal slit, and no apical depression is visible. The sporangia are arranged in rows roughly parallel to the venation, but are not obviously differentiated into sori; spacing between rows is fairly uniform. Material AMF113387-113390, AMF113449. Coal Mine Quarry, Nymboida. Discussion The Nymboida material, which comprises both sterile and fertile fronds, complements that from the type locality in the Esk Trough, Queensland and was examined by Webb (2001) during his revision of the genus Marantoidea. Genus Rhinipteris Harris 1931 Type species Rhinipteris concinna (Presl) Harris 1931, p.58. Rhinipteris walkomii Holmes sp. nov. Figures 4A,B; 5A-C. Diagnosis Large dimorphic fronds; sterile portions tripinnate, fertile portions bipinnate, with both fertile and sterile pinnules sometimes occurring on same pinna rachis. Sterile pinnules with 4-6 well spaced lateral veins. Fertile pinnules with length to width ratio of 3-4:1; abaxial surface completely covered by synangia aligned in parallel rows both longitudinally and transversely. Description A fern with very large fronds estimated to be more than 3 metres in length; primary rachis to 5 cm wide near base; longitudinally striated (Fig. 5A). Fronds sterile and tripinnate or fertile and bipinnate; sometimes with sterile and fertile pinnae occurring on the same frond (Fig. 4A). Primary sterile pinnae bipinnate, broad-lanceolate to 100 mm long and 50 mm wide (Fig. 4B, 5A), overlapping adjacent pinnae; attached alternately to the broad primary rachis; proximal pinnae recurved; in the midportion of frond attached at right angles and at a decreasing angle distally. Secondary sterile pinnae alternate, linear, to 50 mm long and 8 mm wide; tapering distally and becoming pinnatifid; attached at 75°-90°. Proximal pinnules alternate, separated to a broad base; sometimes overlapping, oblong to circular, 2-2.5 mm wide and 3-6 mm long; attached at c. 60°, apex obtuse; becoming pinnatifid distally; midvein decurrent then decurving and running straight to pinnule apex; four to six lateral veins leave midvein but the details are usually obscured due to the thick texture of the pinnule lamina. Fertile pinnae once pinnate, linear, to 50 mm wide and to 200 mm long. The tertiary sterile pinnules have coalesced to form large coherent synangia-bearing fertile pinnules which are alternate to subopposite, broad-linear, 5-8 mm wide and 20-22 mm long, apex obtuse, sometimes slightly falcate (Figs. 4A, 5B,C). The pinnules are closely spaced or overlapping, attached at a high angle to pinna rachis. Each fertile pinnule bears rows of synangia, four on either side of the midvein, which completely cover the lamina surface except over the midvein. They are arranged in straight parallel rows both longitudinally and transversely. The synangia are closely spaced, about | mm in diameter, but the finer details of the sporangia are not preserved on present material. Proc. Linn. Soc. N.S.W., 123. 2001 W.B.K. HOLMES 43 Holotype AMF113393. Syntypes AMF113394-113399, AMF113461. Australian Museum, Sydney. Type Locality Reserve Quarry, Nymboida, Basin Creek Formation, Nymboida Coal Measures. Name Derivation walkomii - for my mentor, the late Dr A.B.Walkom, eminent Australian palaeobotanist, councillor for 53 years and editor for 39 years of the Linnean Society of New South Wales (Vallance 1975). Discussion This is one of the larger ferns in the Nymboida collection but has been collected only from a single horizon at the Reserve Quarry. That horizon has yielded two other large ferns; Asterotheca chevronervia described below and a fern known only from trimorphic sterile fronds which will be described in the next part of this series. The dimorphic form of the fertile and sterile portions of the fronds of Rhinipteris walkomii and the arrangement of the sporangia on the fertile pinnules distinguishes this fern from all others previously recorded from Gondwana Triassic assemblages. The present material is placed in Rhinipteris on the basis of the close similarity of its fertile pinnules with R. concinna (Presl) Harris (Harris 1931, pl.12, fig.1; Boureau 1970, fig.182). The genus was erected by Harris for both sterile and fertile fragments found in close association in the middle of Bed A in the Rhaetic “Lepidopteris Beds” of Scoresby Sound, Greenland. R. walkomii differs from R. concinna by its shorter fertile pinnules and by the regular arrangement of the synangia, both longitudinally and laterally. The reconstruction by Harris of a fertile pinnule (Harris 1931, text fig. 20F- incorrectly labelled R. nitida) shows irregular rows of synangia. No venation is preserved on the adaxial surface of fertile pinnules of R. walkomii to compare with Harris’s figure. The venation on the sterile specimens of R. walkomii also is not clear due to the apparent thick texture of the pinnules. Where seen, the lateral veins appear to be more sparse than in R. concinna, but there is a wide range of variation, as seen on the individual fragments illustrated by Harris (1931, text figs 20 B-D). Reinitsia ternerae Herbst et al (1998) from the Upper Triassic of Chile has fertile pinnules bearing two or three rows of synangia on either side of the midrib. R. ternerae differs from Rhinipteris walkomii by its sterile pinnules which are similar in form to its fertile pinnules, by the broad pinnule midvein and by the synangia being four to six sporangiate. From the shape and thick texture of the pinnules, sterile pinnae of R. walkomii could be confused with Lepidopteris spp. but they can be distinguished from that genus by the absence of pinnules on the rachis between the pinnae. Family Asterothecaceae Genus Asterotheca Presl, in Corda 1845 Type species A. sternbergii (Goeppert) Presl, in Corda 1845 Asterotheca is a genus erected for fertile ferns with pecopteroid pinnules which bear a line of adjacent synangia on the abaxial surface between the midvein and the margin. The synangia are composed of groups of sporangia conjoined at the base and dehiscing along an apical suture line. An annulus is absent. The genus includes species from both the northern and southern hemispheres which range in age from Carboniferous to Upper Jurassic. It is most probably not a natural genus. Similar ferns known only from sterile material are often placed in Pecopteris (Boureau 1970). Proc. Linn. Soc. N.S.W., 123. 2001 44 TRIASSIC FLORA FROM NYMBOIDA — FILICOPHYTA Asterotheca trullensis Holmes sp. nov. Figure 6A, 7A-C. 1975 Cladophlebis australis non (Morris) Seward, Flint and Gould, pl.3, fig. 5. Diagnosis Large fern with bi- (? tri-)pinnate fertile fronds. Pinnae long, linear, alternate to subopposite. Fertile pinnules with length to width ratio of c. 2-2.5:1; bearing a non- contiguous row of 12-16 tetra-sporangiate synangia to 0.75 mm in diameter around pinnule lamina, midway between midvein and pinnule margin. Description Figure 6A shows three frond segments which may be the primary pinnae of a larger tripinnate frond, or are individual bipinnate fronds arising from a common rhizomatous base. If bipinnate the fronds are estimated to have reached half a metre in length and 16 cm in width. The pinnae are alternate to sub-opposite. Their attachment to the primary rachis is decurved in the proximal pinnae, at right angles in the middle of the frond and becoming more acute towards the frond apex. Pinnae linear, to 90 mm long and 15 mm wide. Fertile pinnules alternate (Fig. 7A, B), attached at 55°-75°, margins parallel, apex rounded; basiscopically decurrent, to 9 mm long and to 4.5 mm wide, but usually less; becoming smaller distally and apically. Length to width ratio of 2-2.5:1. Midvein decurrent, thin, straight. Lateral veins at c. 45°, indistinct, one per synangium, once-forked. Synangia tetrasporangiate, c. 0.75 mm in diameter, c. 12-16 arranged as separate individuals in a single row around pinnule midway between midvein and pinnule margin. Usually one or two more synangia on basiscopic side of pinnule than on acroscopic side. Fossils represented in Figures 7A and 7B are external moulds of the abaxial surface of the pinnae. No fronds have been found which bear both fertile and sterile pinnules. On gross morphology and close association with fertile fragments, the sterile frond illustrated in Figure 7C is considered to represent the sterile form of A. trullensis. Pinnae are alternate, attached at c. 60°, broad-linear to 120 mm long and c. 25 mm wide for 2/3 their length then tapering distally to an acute apex. The sterile pinnules are similar in size but slightly more falcate than fertile pinnules with a length to width ratio of c. 2.5:1, attached alternately to the pinna rachis. The midrib is distinct, continuing almost to pinnule apex. Seven to ten pairs of lateral veins leave the midvein at c. 60°, forking once close to the midvein, diverging slightly and arching to meet the margin at a high angle. Holotype AMF113400. Australian Museum, Sydney. Type Locality Coal Mine Quarry, Nymboida. Basin Creek Formation, Nymboida Coal Measures. Other Material AMF113401-113405, AMF113437, AMF113441, AMF113450, AMF113467, UNEF13608, all from Coal Mine Quarry, Nymboida. Name Derivation trulla - Latin, basin, referring to the Type Locality within the Basin Creek Formation. Discussion A. nymboidensis Holmes (this paper), A. menendezii de la Sota and Archangelsky (1962), A. hilariensis Menendez (1957) and A. rigbyana Herbst (1977a) differ from A. Proc. Linn. Soc. N.S.W., 123. 2001 W.B.K. HOLMES 45 trullensis by their synangia, which cover all the surface of the pinnule lamina. A. hillae (Walkom) Herbst (1977a) and A. diameson Holmes (this paper) differ by their smaller sized synangia. Asterotheca nymboidensis Holmes sp. nov. Figures 8 A, 9 A-C, 10 A-C. 1975 Asterotheca hillae non Walkom, Flint and Gould, Pl.1, figs 1,2. 1975 Cladophlebis concinna non (Pres) du Toit, Flint and Gould PI.1, figs 7,8. 1977 Asterotheca menendezii non de la Sota and Archangelsky, Bourke et al. p.19, fig.3.3 Diagnosis Bipinnate fern with sterile pinnules oblong, apex rounded, with length to breadth ratio of c. 2-2.5:1. Lateral veins straight, once forked close to midvein. Fertile pinnules narrower with length to breadth ratio of 3-3.5:1, bearing seven to eight pairs of quadrisporangiate contiguous synangia c. 0.75-1 mm in diameter, arranged on either side of the midvein and filling all of the lamina space. Sterile and fertile pinnules sometimes on same pinna. Description Bipinnate, pinnae opposite to subopposite; attached to the primary rachis at c. 90° proximally and then at a decreasing angle distally, usually straight but sometimes decurving; to 100mm long and 18 mm wide from base to midpoint then decreasing in width towards the acute apex. Sterile pinnules (Fig. 8A) opposite to sub-opposite; closely spaced but free to the base where the basiscopic margins may be shortly confluent on the rachis; attached at c. 80°; margins parallel or slightly tapering, entire, apex acuminate to obtuse; pinnules straight to slightly falcate; length from 8-14 mm and width from 4-6 mm, smaller towards the apices of the pinnae and frond. The length to width ratio is c. 2-2.5:1. Midvein prominent, slightly decurrent on rachis and then running straight to the pinnule apex. Five to eight pairs of lateral veins are attached at c. 60°, forking close to the midvein, diverging then running almost straight and parallel to meet the margin at c. 45°. Very rarely a lateral vein will fork a second time. Fertile pinnules narrower than the sterile pinnules (Fig. 9 B,C), with a length to width ratio of 3-3.5:1, broadly decurrent, attached at c. 60°-90°. Seven to eight pairs of synangia are attached to either side of the pinnule midrib and occupy the whole surface of the lamina between the midvein and margin (Fig. 8A; 9A,C). Synangia (Fig. 1OA-C) c. 0.75-1 mm in diameter with a square or rosette form of arrangement of four, occasionally three or five, spherical to ovoid sporangia each c. 0.4-0.5 mm in diameter. Sporangial cell walls longitudinally elongate (Fig. 10C). Holotype AMF113408. Australian Museum, Sydney. Type Locality Reserve Quarry, Nymboida. Basin Creek Formation, Nymboida Coal Measures. Other Material AMF113406, AMF113409-113413, UNEF13396, UNEF13401, UNEF14122- 3. All from Reserve Quarry. Name derivation nymboidensis - from Nymboida. Proc. Linn. Soc. N.S.W., 123. 2001 46 TRIASSIC FLORA FROM NYMBOIDA — FILICOPHYTA Discussion Asterotheca nymboidensis was one of the larger and more common ferns in the Nymboida Flora. The form of the living plant is not known. The Nymboida material here described as A. nymboidensis was included with Scolecopteris australis Shirley (1898) (= Cladophlebis australis Walkom 1917) in Asterotheca menendezii de la Sota and Archangelsky (1962) by Herbst (1977a). A. menendezii differs from A. nymboidensis by the narrower, longer, more widely spaced fertile pinnules and by the greater number of synangia per pinnule. The number of synangia per pinnule in A. nymboidensis is c. 8-16; for A. menendezii, c. 18-24. The length to width ratio of pinnules of A. nymboidensis is c. 3:1 and for A. menendezii c. 4-5:1. A. nymboidensis differs from A. trullensis by its opposite to subopposite attachment of pinnae and pinnules, by the lesser length to width ratio of the sterile pinnules and by the contiguous synangia which fill the whole of the pinnule lamina. A. hillae (Walkom) Herbst (1977a) and A. denmeadii Walkom (Herbst 1977a), which are regarded as synonyms by Rigby (1977), differ from A. nymboidensis by the smaller synangia which occupy only a portion of the pinnule lamina. A. rigbyana Herbst (1977a) and A. hillae differ by the unforked lateral veins. Asterotheca sp. from the Sydney Basin (Herbst 1977a) differs by its trisporangiate synangia. A. truempyi Frenguelli (1943) from Argentina differs by the pinnules bearing less synangia and by the unforked lateral venation. A. hilariensis Menendez (1957) differs by its venation which was shown as once forked and strongly recurving by Menendez (1957) or unforked by Herbst (1977a). Asterotheca chevronervia Holmes sp. nov. Figures 11 A, 12 A-D. Diagnosis Large bi-(tri ?) pinnate frond; pinnules pecopteroid, closely spaced to overlapping; length to width ratio 2.5-3 : 1. Lateral veins opposite, straight, unforked, each pair forming a Shallow V shape. Fertile pinnules similar in shape to the sterile pinnules; eight to nine pairs of synangia fill the whole of the lamina surface between the midrib and pinnule margin. Description Complete fronds not known. The type specimen (Fig. 11A ) shows numerous pinnae, some of which are attached to smooth rhachises to 10 mm wide. Such fronds, when complete would have measured at least 2 m in length. Also on this specimen there are portions of axes to 40 mm in width. Perhaps these are the primary rachises of a tripinnate fern with fronds which may have exceeded 3 m long. Pinnae sub-opposite to alternate, attached at 50°-70°, closely spaced to overlapping; to 120 mm long and 10 mm wide; tapering distally. Pinnules oblong with rounded apex, entire, 1-2 mm wide, 3-6 mm long. Separated to the base but closely spaced to over-lapping; attached to the coarsely, longitudinally ribbed pinnae at 60°-80°; midvein straight, continuing almost to pinnnule apex. Eight to ten opposite pairs of unforked lateral veins are attached at c. 45°-60° and run straight to margin. Fertile pinnules similar in shape to sterile with c. eight pairs of synangia on either side of the midvein fill all the space between the midvein and lamina margin. Details of synangia are poorly preserved. Holotype AMF113414, paratypes AMF113415-113416, AMF113456-113458, AMF113462-113463. Australian Museum, Sydney. Type locality Reserve Quarry, Nymboida. Basin Creek Formation, Nymboida Coal Measures. Proc. Linn. Soc. N.S.W., 123. 2001 W.B.K. HOLMES 47 Name derivation chevronervia - referring to the chevron-like arrangement of the lateral veins. Discussion A.chevronervia is a rare fern having been collected from a single horizon only, where it was locally abundant. It is distinguished from the other Nymboida ferns by the unforked lateral venation. Sterile fronds closely comparable to A. chevronervia have been figured from the Molteno Formation of South Africa by Anderson and Anderson (1983, pl. 8) as Cladophlebis sp. “D”. A. hillae (Walkom) Herbst (Herbst 1977a) has somewhat similar venation to A. chevronervia but differs by the smaller synangia which occupy the outer portion only of the pinnule lamina. A. rigbyana Herbst (1977a) is similar to A. chevronervia but differs by the greater length to width ratio of the pinnules which are shorter and more widely spaced. A. truempyi Frenguelli from Argentina has pinnules with chevron-like venation similar to A. chevronervia. However the line illustration of a fertile specimen by Herbst (1977a, pl.1, fig. 3) shows only six or seven synangia on each pinnule. The photographed specimens of A. truempyi (Herbst 1977a, pl.3, figs 30, 31, 37) are not well enough preserved to allow comparisons. Asterotheca diameson Holmes sp. nov. Figure 14A Diagnosis Tripinnate frond, pinnae opposite. Pinnules small, length to width ratio of c. 1.5:1; bearing nine to ten separated tetrasporangiate synangia each c. 0.5 mm in diameter. Description Portion of a tripinnate frond probably exceeding 200 mm in length. Secondary rachis tapering from 2.5 mm to 2 mm in the 90 mm preserved. Pinnae opposite, attached at 75°-85° to rachis, 6-7 mm apart, overlapping, linear, 9 mm wide, length may exceed 60 mm. Pinnules closely spaced, alternate, attached to pinna rachis at 70°-85° by whole base, 3-4 mm long and 2.5 mm wide, margins parallel, apex rounded. Length to width ratio of 1.2-1.6: 1. Nine to ten tetrasporangiate synangia per pinnule, each c. 0.5 mm in diameter, well separated, aligned around pinnule midway between the midvein and the pinnule margin. Holotype AMF113420 and counterpart AMF113421. Australian Museum, Sydney. Type Locality Reserve Quarry, Nymboida. Basin Creek Formation, Nymboida Coal Measures. Name Derivation diameson - Greek - midway; referring to the position of the line of synangia between the midvein and pinnule margin. Discussion This is a rare fern, being known only from a single specimen and its counterpart. It is unusual in being preserved in a rare reddish-purple siltstone horizon which is normally unfossiliferous. On the edge of the type specimen there is a fragment of a second pinna which is aligned with the main illustrated pinna in a manner that strongly suggests that this was a tripinnate fern. Proc. Linn. Soc. N.S.W., 123. 2001 48 TRIASSIC FLORA FROM NYMBOIDA — FILICOPHYTA A. diameson differs from A. trullensis Holmes and A. nymboidensis Holmes by the shorter, straight pinnules and by the fewer and smaller synangia. A. rigbyana Herbst (1977a) is similar to A. diameson in pinnule size and shape but differs by the larger synangia which cover the whole of the pinnule surface. A. truempyi Frenguelli (1943) differs by its fewer and contiguous synangia. In synangial form and arrangement, the following four fertile fern fragments are distinct and have close affinities with Asterotheca. However, due to the small fragments available, they are illustrated and described briefly, without assigning them to species. Asterotheca sp. A Figure 13 A. Description A fertile pinna fragment showing portions of four pinnules with decurrent basiscopic bases. The pinnules vary in length and width from 13 x 5 mm to 12 x 6 mm and are probably attached at a high angle to the rachis. The tetrasporangiate synangia are closely spaced but not contiguous, c. 0.2-0.3 mm in diameter, aligned parallel to the margin and midway between margin and midvein. Material AMF113417, Coal Mine Quarry, Nymboida. Discussion The fertile pinnules on this specimen differ from all known Gondwana Asterotheca species. Asterotheca sp. B Figure 13 B. 1977b Reinitsia whitehousei Herbst in part, pl.1, fig.7; pl.2, fig.21, only. Description A portion of a fertile bipinnate frond with widely separated pinnae. The pinnae are all incomplete. The largest pinna fragment shows five pairs of alternate pinnules, with either decurrent or contracted basiscopic bases. The pinnules are broad-linear, entire, tapering slightly in the distal two thirds to a rounded apex; length c. 10 mm, width c. 4 mm; midvein decurrent, lateral veins perhaps once forking. From 26 to 34 poorly preserved synangia 0.5-0.75 mm in diameter form a single file around and close to the pinnule margin. Material QMF42305, on composite slab formerly registered UQF64174. Nymboida Colliery. Discussion This frond fragment was included by Herbst (1977b) in his Reinitsia whitehousei (see discussion below under Herbstopteris colliveri). Asterotheca sp. B does not belong in Herbstopteris. Asterotheca falcata de la Sota and Archangelsky (1962) and A. menendezii de la Sota and Archangelsky (1962) both have numerous synangia but differ from Asterotheca sp. B in synangial details and pinnule shape. Proc. Linn. Soc. N.S.W., 123. 200] W.B.K. HOLMES 49 Asterotheca sp. C Figure 13 C. Description Pinna fragments with alternate, broad, overlapping pinnules, c. 6 mm long and 3 mm wide, with decurrent midvein and c. 10 pairs of lateral veins that appear to fork close to the midvein. Some of the pinnules have a few poorly preserved synangia c. 0.5-0.8 mm in diameter spaced irregularly around the lamina margin. Material AMF113418. Coal Mine Quarry, Nymboida. Discussion These specimens may possibly be partially fertile or immature fertile pinnae of A. trullensis. Asterotheca sp. D Figure 13 D. Description Fertile pinna fragment with six pairs of oblong pinnules 8-9 mm long and 4-5 mm wide, with about 20 tetrasporous synangia forming a contiguous line around pinnule midway between the strong midvein and the pinnule margin. Material AMF113419. Coal Mine Quarry, Nymboida. Discussion This form is closest to A. trullensis but differs by the contiguous arrangement and greater number of synangia on the pinnules. Genus Herbstopteris Holmes gen. nov. Diagnosis Medium sized cladophleboid fern with bipinnate fronds radiating from a rhizotomous base. Primary and pinna rachises prominently longitudinally ridged. Frond broad elliptical with pinnae bearing falcate confluent pinnules which conjoin distally and apically to become lobate to entire. Proximal pinnules with decurrent midvein and once forked lateral veins. As pinnules coalesce, venation changes to a single vein entering each lobe, arching and forking once. Fertile pinnules with a single row of tetrasporangiate synangia on either side of pinnule midvein, with a basal synangium on decurrent portion of the pinnule lamina. Distally and apically as pinnules coalesce, the line of synangia follows parallel to the lobate or entire pinna margin. Discussion The genus Herbstopteris is erected for certain fertile fronds from Denmark Hill in the Late Triassic Ipswich Coal Measures of Queensland that have been placed previously in Thinnfeldia, Asterotheca or Reinitsea. Walkom (1917, pl.1, fig.3; pl.3, fig.3) described and illustrated the terminal portions of two pinnate fronds which he believed were the fertile leaves of Thinnfeldia (= Dicroidium). Based on abundant material from the Molteno Formation of South Africa Thomas (1933) demonstrated that Dicroidium was a seed- bearing pteridosperm. Townrow (1957) argued that Walkom’s fertile specimens were ferns which he placed in Asterotheca fuchsii (Zeiller) Kurtz (Kurtz 1921). [have examined Proc. Linn. Soc. N.S.W., 123. 2001 50 TRIASSIC FLORA FROM NYMBOIDA — FILICOPHYTA the material in the Natural History Museum (London) which were described in Townrow’s paper. It is a mixed collection of fern fragments from the Ipswich Coal Measures of Queensland and one from Brookvale in the Sydney Basin. Two specimens (NHMV4208 and V24467) are indeterminate Asterotheca spp., one (NHMV5950) is too poorly preserved to make comparisons; the Brookvale specimen (NHMV32112) is totally sterile and cannot be satisfactorily identified. Only two specimens (NHMV24632a and b), an apical portion of a fertile frond and its counterpart are similar to the material described and figured by Walkom (1917) as fertile Thinnfeldia. They are quite distinct from A. fuschii. These two specimens I include below in the new genus Herbstopteris. In a revision of some eastern Australian and Argentinean ferns Herbst (1977a) transferred Walkom’s specimens to a new species in the genus Reinitsia as R. colliveri Herbst. The genus Reinitsia was erected by Walkom (1932) for a fragment of a fern with long linear lobate pinnae with the proximal portion of the pinnae fertile. Tetrasporangiate synangia were arranged in a single file parallel to the lobate pinna margin. On the distal sterile portions of the pinnae the well-preserved lateral venation comprised single veins leaving the pinna rachis at an acute angle then branching irregularly two or three times into each lobe. This unusual form of venation is quite unlike Walkom’s (1917) ‘fertile Thinnfeldia’ and others that were included by Herbst (1977a) in Reinitsia. | am not aware of any other Gondwana Triassic fern with this irregular venation architecture although it is similar to that in the simple leaves of the pteridosperm Dejerseyia lobata (Jones and deJersey 1947; Anderson and Anderson 1983). Based on this difference in venation I now place Reinitsea colliveri in the new genus Herbstopteris. In the genus Asterotheca the fronds are bipinnate with the pinnules separated to the base. The lines of synangia are parallel from the base until they curve and join at the pinnules’ rounded apices.’ Herbstopteris has similar, usually four-sporangiate synangia but differs from’ Asterotheca by the pinnules being basally decurrent on the pinna rachis and by the line of synangia following around the base of the decurrent portion to form a U shape between adjacent pinnules. Distally and apically the pinnules coalesce to form lobed or entire pinnae while the line of synangia follows parallel to the lobed or entire margin, characters not present in Asterotheca. Additional and more complete material from Nymboida supports the erection of the new genus Herbstopteris. Name derivation Herbst - in recognition of the eminent Argentinean palaeobotanist Dr Raphael Herbst who has carried out extensive research on Gondwana ferns over the last forty years. pteris — Greek, fern. Herbstopteris colliveri (Herbst) Holmes comb. nov. Figures 15 A, B, 16 A-C, 17 A-D. 1917 Thinnfeldia feistmantelii non Johnston, Walkom in part, p.17 text fig.3, pl.1, fig.3 only. 1917 Thinnfeldia lancifolia non Morris, Walkom in part, p.21, pl.3, fig.3 only. 1957 Asterotheca fuchsii non (Zeiller) Kurtz, Townrow, p.22, pl.2, fig.A only. 1965 Asterotheca (Pecopteris) fuchsii non (Zeiller) Kurtz, Hill et al, pl. T3, fig.6. 1977b Reinitsia colliveri Herbst, p.24, figs 1-6, 16-20. 1977b Reinitsia whitehousei Herbst, figs 8-10, 22-25 only. Proc. Linn. Soc. N.S.W., 123. 2001 W.B.K. HOLMES 51 Description Small to medium sized fern with up to nine broad elliptic fronds of various sizes, radiating from an elongated rhizome c. 25 mm wide. Complete fronds estimated to range from 150 mm to 300 mm long. Portions of pinnae in the lower to mid portion of the frond and proximal to the main rachis bear well-developed pinnules which are conjoined by a narrow wing along the pinna rachis. Towards the distal portions of the pinnae and the apical part of the frond the pinnules coalesce to form lobate to entire pinnae. Proximal sterile pinnules Cladophlebis-like, decurrent, oblong to slightly falcate, margins entire, apex acuminate to obtuse, 4-8 mm long and 2-3 mm wide. Midvein straight, lateral veins well-spaced, forking once midway to the margin. As the pinnules coalesce distally and apically the venation changes to a single arching once-forked vein into each marginal lobe. On fertile portions of fronds, the mostly 4-sporangiate synangia, 0.5-1 mm in diameter, form a single row close to and parallel to the pinnule margin. Usually one or more synangia are attached around the decurrent basiscopic wing to form an open U- shape with the synangia on the adjacent pinnule; eight to twelve synangia per pinnule. As the pinnae taper distally and apically, the number of synangia on each pinnule decreases until they form an undulate or straight single file close and parallel with the margin of the coalesced pinnules. Material AMF113425-113431, Coal Mine Quarry, Nymboida. QMF42303, QMF42304, QMF42306. Nymboida Colliery, Nymboida. Discussion The Type specimen for H. colliveri is Geological Survey of Queensland Fossil Number 730 as selected by Herbst (19977b) for his Reinitsea colliveri. The Nymboida specimen (Fig. 15A,B) with both sterile and fertile fronds radiating from a common base is very important in demonstrating the habit of growth of this fern. Figures 15B and 16C show the change in form of a single frond from bipinnate basally and close to the main rachis then lobed to simply pinnate distally on the pinnae and apically on the frond. Other Nymboida specimens with only the apical portion preserved (Figs 16A,B and 17A-D) are closely similar to the Ipswich material of Herbstopteris colliveri. Based on material from the Nymboida Colliery, Herbst (1977b) erected a new species, Reinitsia whitehousei. His type material comprised a single slab (formerly catalogued as University of Queensland F64174) bearing a portion of a bipinnate-pinnate fertile frond associated with fragments of pinnae with elongated fertile pinnules bearing up to 34 synangia around each pinnule. Herbst considered all the material on this slab represented the range of variation within a single taxon. I have examined Herbst’s specimens which are now in the Queensland Museum. The bipinnate-pinnate frond and the elongate fertile pinnae are two separate entities. One of the latter pinnae (now catalogued QMF42305) is figured above as Asterotheca sp. B (Fig. 13B). The frond (QMF42306) and other fragmentary pinnae in Herbst’s collection (QMF42303-4) all fit comfortably within the range of variation of Herbstopteris colliveri and thus Reinitsea whitehousei is regarded as a later synonym of that species. Herbstopteris sp. A Figure 14 C. Description Two isolated parallel fertile pinna fragments, pinnules well-separated, to 7 mm long and 3.5 mm wide, bases strongly decurrent, bearing 14-20 synangia 0.25-0.5 mm in Proc. Linn. Soc. N.S.W., 123. 2001 52 TRIASSIC FLORA FROM NYMBOIDA — FILICOPHYTA diameter, irregularly spaced parallel to the pinnule margin, the basiscopic synangia following around the decurrent base. Material AMF 113423-113424. Coal Mine Quarry, Nymboida. Discussion This specimen differs from Herbstopteris colliveri by the broader, more separated pinnules and by the more numerous and smaller synangia, but as it is the only specimen available, it has not been formally named. Order Osmundales Family Osmundaceae R. Brown 1810. Genus Todites Seward 1900. Type species Todites williamsonii (Brongn.) Seward 1900. The genus Jodites was widespread in the Triassic and became cosmopolitan during the Jurassic and Cretaceous (Boureau 1970). It is based on fertile material where the Osmunda-like sporangia are attached to the pinnules to form an irregular mass, often covering the whole underside of the lamina. Leaves of similar form but known only in the sterile state are normally placed in the Cladophlebis genus. Todites have been recorded in the Triassic of Gondwana by Walkom (1928), Burges (1935), Holmes (1982), Anderson and Anderson (1983), Retallack (1983) and Herbst (1989). Todites parvum Holmes sp. nov. Figures 18 A- E. 21898, Triphyllopteris botryoides Shirley, p.20, pl.7, fig.1, middle and right hand specimens only. 21917, Coniopteris delicatula Walkom, p.6, text fig.3, pl.4, fig.2 Diagnosis A small bipinnate fertile frond; pinnules sub-opposite to alternate, triangular to elongated, margins lobed, broadly falcate, apex obtuse to acute. Sporangia circular, closely spaced, irregularly arranged on the whole undersurface of the pinnules. Description This is a rare fern element at Nymboida with only four specimens in the collections. Sterile fronds have not been recognised. The holotype (fig.18 A,B) is an apical portion of a bipinnate frond with opposite pinnae bearing closely spaced, alternate to sub-opposite pinnules. Pinnules in the proximal half of the lower preserved pinnae are triangular or broadly falcate, c. 2 mm wide and to 2.5 mm long, contracting to an obtuse or acute apex. Pinnules decrease in size distally on the pinnae and towards the apical portion of the frond. The pattern of venation can not be seen. About 15-20 closely spaced sporangia, each c. 0.25 mm in diameter cover the whole under surface of the pinnules. On other specimens (figs 18 C-E) the lamina tissue of the pinnules has not been preserved and the specimens show masses of sporangia at the position of the pinnules on either side of the pinna rachis. Holotype AMF 113432 and counterpart AMF113433. Australian Museum, Sydney. Proc. Linn. Soc. N.S.W., 123. 2001 W.B.K. HOLMES 2/3) Type Locality Coal Mine Quarry, Nymboida, Basin Creek Formation, Nymboida Coal Measures. Other material AMF113434-113436. Coal Mine Quarry, Nymboida. Name Derivation parvum - Latin - small, referring to the size of the pinnules. Discussion As most Todites spp. have dimorphic fronds it is difficult to make comparisons between taxa when only sterile or fertile fronds are present. Todites parvum is known only from fertile fronds. It is possible that the sterile fronds are also present in the collections but their affiliation with the fertile fronds is not, at present, recognised. Fertile and sterile fronds from the Esk Beds were assigned by Walkom (1928) and Hill et al. (1965) to the Northern Hemisphere Todites williamsoni. The Esk forms differ from 7: parvum by their greater size and by their orbicular shaped pinnules. Todites narrabeenensis Burges (1935) and Todites sp. Herbst (1989) have fertile pinnules of similar size to 7: parvum. However T. narrabeenensis differs by its pinnatisect pinnae with crenulate margins; Todites sp. differs by its circular pinnules and more numerous sporangia. Of the figured but undescribed Todites species of Anderson and Anderson (1983), Todites sp. A and Todites sp. B have larger, elongate, lobed pinnules and Todites sp. C has pinnules similar in shape to T: parvum but differs by the much larger size and more numerous tiny sporangia. T. pattinsoniorum Holmes (1982), 7: maoricus Retallack (1983) and T. baldonii Herbst (1989) are much larger in size and are known from both fertile and sterile material. Shirley (1898) described some fragments of fertile fronds together with a sterile pinna from Shorncliffe in the Ipswich Series of south-eastern Queensland as Triphyllopteris botryoides. The same specimens, together with more complete associated sterile fronds were assigned by Walkom (1917) to Coniopteris delicatula (Shirley) Walkom. Rigby (1977) expressed doubts that these leaves belong in Coniopteris. While the fertile fragments are described as being parts of a lobate frond, the individual pinnules are closely similar in size and sporangial arrangement to 7: parvum. Neither sterile fronds similar to Coniopteris delicatula nor other sterile fronds that could be affiliated definitely with T. parvum are known from Nymboida. Genus Osmundopsis Harris 1931 Type species Osmundopsis sturi (Raciborski) Harris 1931. Osmundopsis scalaris Holmes sp. nov. Figures 19A,B; 20A-C; 21A. Diagnosis Medium sized fern with bipinnate fronds radiating from a rhizomotous base; sterile pinnules cladophleboid, lateral veins well-spaced, once forked. Fertile pinnules with much reduced lamina, with c. 15-20 spheroidal or ovoid sporangia c. 1 mm in diameter, closely spaced on either side of the midvein. Fertile and sterile pinnae sometimes occurring on same frond. Description The holotype (Fig. 19A) is a beautiful and unique specimen. At least seven fronds, including two that are partially fertile, radiate from a common base to show the form of Proc. Linn. Soc. N.S.W., 123. 2001 54 TRIASSIC FLORA FROM NYMBOIDA — FILICOPHYTA growth. Another specimen (AMF113447) shows fronds attached to a rhizome. The two fronds on the holotype which bear sterile pinnae near the base and apex with fertile pinnae in the middle portion of the primary rachis, are important in providing the key to allow affiliation of isolated fragments where only fertile or sterile material is preserved. Bipinnate fronds with opposite to sub-opposite, well-spaced pinnae are attached at c. 70°-80° to a slender primary axis to c. 50 cm in length. Sterile pinnae to 50 mm long and 20 mm wide, broad-linear, decreasing in width distally. Sterile pinnules (Fig. 19B, Fig. 20B, Fig. 21A) sub-opposite to alternate, attached at 70°-90°, adjacent to well-spaced, free to the base which may be slightly contracted or decurrent; 9-12 mm long and 3-4 mm wide, with a length to width ratio of 3:1. Lateral venation at c. 45° to midvein, well- spaced, once-forked. Fertile pinnae to c. 40 mm long and c. 10 mm wide. Fertile pinnules (Figs 19A, 20A, C) attached opposite to sub-opposite at 80°-90°; 6-8mm long, with much reduced lamina. Fifteen to twenty sporangia, spheroidal to ovoid, each c. 1 mm in diameter are closely spaced on either side of the midvein (Fig. 20C). Cell structure and annulus not preserved. Holotype AMF113468. Isotypes - fragments removed from the Holotype slab. AMF113438-9, AMF113444-6, AMF113452-55, AMF113469-75. Australian Museum, Sydney. Type Locality Coal Mine Quarry, Nymboida. Basin Creek Formation, Nymboida Coal Measures. Other Material AMPF113443, AMF113447. Coal Mine Quarry, Nymboida. Name Derivation scalaris - Latin - ladder-like, referring to the appearance and arrangement of the fertile pinnules. Discussion Andrews in Boureau (1970) queried the reasons given by Harris (1961) for the erection of the genus Osmundopsis. Andrews considered that the features of Osmundopsis coincided with those of the extant genus Osmunda. I consider the placing into a separate genus of fragmentary fossil material in which the finer details of structure may be missing, is preferable to assuming that the fossil is congeneric with extant material. In various features of gross morphology Osmundopsis scalaris resembles some extant osmundaceous ferns with dimorphic fronds on which the fertile pinnae have greatly reduced laminae. However, the lack of spores and the absence of finer details of the sporangia does not allow for closer comparison. Osmundopsis scalaris differs from O. sturi (Raciborski) Harris (1931, 1961) by the fewer and larger sporangia and from O. plectrophora Harris (1931) which is tripinnate in the fertile state. From the Upper Triassic Lashley Formation of Antarctica, Taylor et al. (1990) illustrated a fragment of a dimorphic bipinnatifid frond which bore on the same primary rachis both sterile pinnae with deeply dissected segments with Cladophlebis-type dichotomous venation and short non-laminate fertile pinnae with tightly compacted clusters of sporangia along the pinna midvein. The authors drew attention to the similarity of the frond morphology of the fossil to that of the extant fern Osmunda claytoniana. The Antarctic material, which was not formally described, has similar sterile pinnae to Osmundopsis scalaris but differs by its very short simple fertile pinnae. Proc. Linn. Soc. N.S.W., 123. 2001 W.B.K. HOLMES 55 Several specimens in the Nymboida collections show sterile fronds radiating from a common base. Some fronds show the primary and secondary rachises strongly grooved while others have smooth rachises. The rachises in life were probably kidney shaped in transverse section as in extant osmundaceous ferns. The appearance of the resulting fossil following compression depends on whether the fossil is a cast or mould of the upper or lower surface of the original plant. The smooth stemmed fossils represent a cast or mould of the upper convex surface - ridged or grooved stems would be casts or moulds of the concave lower surface. Order Filicales Family Gleicheniaceae Genus Gleichenites Goeppert 1836 Type species Gleichenites porsildii Seward 1926 Gleichenites wivenhoensis Herbst (1974) Figure 14 B. 1974 Gleichenites wivenhoensis Herbst p.79, figs 1,2, pl.9, figs 7,8, pl.10,fig.11. Description The two parallel pinnae illustrated on Fig. 14B suggest that they were from a bipinnate frond. These incomplete pinnae are 35 mm and 45 mm in length, tapering distally from 7 mm in width. The pinnules, which are separated to the base, are short, broad and falcate, c. 2.5 mm wide and 4 mm long, acuminate to obtuse, decreasing in size towards the pinna apex. They are attached to the pinna rachis at c. 45°. Basiscopic margin more convex than acroscopic margin; venation not preserved. Proximal fertile pinnules bear five to seven tetrasporangiate synangia which are aligned on either side of the pinnule midvein and occupy most of the lamina. The smaller distal pinnules bear less synangia. Material AMF113422, AMF11424. Reserve Quarry, Nymboida. Discussion In outline, sterile fragments of G. wivenhoensis could be confused with portions of Dicroidium leaves. G. wivenhoensis differs from other Gleichenites species by its soral characters (Herbst 1972, 1974, 1996; Herbst et al. 1998). Eboracia herbstii Rigby (in Playford et al. 1982), from the Middle Triassic Moolayember Formation of Queensland has long linear pinnae with fertile pinnules similar in size but more rounded than G. wivenhoensis. As the base of the pinnae of the Nymboida material is not preserved, comparisons cannot be made with Eboracia spp. in which the diagnostic character is a greatly enlarged basiscopic pinnule. Phipps et al. (2000) have described silicified gleicheniaceous sori as Gleichenipteris antarcticus. The poor preservation of the sori of Gleichenites wivenhoensis does not allow for comparison with G. antarcticus. Family Dipteridaceae Genus Dictyophyllum Lindley and Hutton 1834 Type species Dictyophyllum rugosum Lindley and Hutton 1834 Dictyophyllum davidii Walkom 1917, p.10. Figures 22A-E. 23A-C. Proc. Linn. Soc. N.S.W., 123. 2001 56 TRIASSIC FLORA FROM NYMBOIDA — FILICOPHYTA 1917 Dictyophyllum Davidi Walkom, p.10, pl.3, fig.2. 1924 Dictyophyllum rugosum L&H; Walkom, p.82, pl.21, fig.1. 1965 Dictyophyllum davidi Walkom; Hill et al. p.T6, pl.3, fig.1. 1975 Dictyophyllum davidii Walkom; Flint and Gould, p.71; pl.1 fig.3. 1982 Dictyophyllum davidii Walkom; Webb, p.85, figs. 6A, 7, 8.\ Description The Nymboida material, while containing specimens that agree well with the type specimen (UQF165) of Walkom (1917) from the Esk Formation of south-eastern Queensland, also includes a wider range of variation which is illustrated in Figures 22 and 23. Fronds are borne on a long slender primary rachis which may exceed 150 mm in length (Fig. 22A) and have a typical dipteridacean branching pattern in which the main rachis bifurcates into two arms although both arms are rarely preserved intact. Each arm bears five to eight pinnae attached in a pseudo-palmate arrangement. The pinnae are very shortly petiolate, lanceolate and of varying size and lobation, to 70 mm long and to 25 mm wide. The number of lobes and the depth of dissection is variable. Venation consists of a strong median vein with well-spaced secondary veins, one to each lobe, passing straight to the margin at c. 60° (Fig. 22E). Tertiary venation arises at high angles from the primary and secondary veins to form a fine network of square and polygonal meshes throughout the lamina. Fertile pinnae have groups of sori adjacent to the midvein and in the areoles formed by the tertiary veins (Fig. 23C). The sori contain irregular groups of c. 25 or more rounded sporangia each c. 1.5-2 mm in diameter. Details of the annulus are not clear. Material AMF113374-113381, AMF113450, AMF113459, AMF113460, UNEF13416. Coal Mine Quarry. Discussion Dictyophyllum davidii is the earliest reported member of the genus which had numerous species and a world-wide distribution in the Late Triassic and Jurassic (Boureau 1970; Herbst 1992). The Australian occurrences of Dictyophyllum davidii, which are restricted to the Anisian-Ladinian Nymboida Coal Measures and the Toogoolawah Group of the Esk Trough of Queensland, have been discussed in detail by Herbst (1975, 1979) and Webb (1982). D. davidii differs from D. bremerense and D. shirleyi from the late Triassic Ipswich Coal Measures by its generally smaller size and in the spacing and shape of the lateral lobes on the pinnae (Walkom 1917; Webb 1982). D. tenuiserratum and D. chihuiuensis from the Middle Triassic of Argentina (Herbst 1993) both differ from D. davidii by the frond lamina being deeply dissected into elongate lobes with serrate or denticulate margins. D. ellenbergii Greber in Fabre and Greber (1960) from the Late Triassic Molteno Formation of Lesotho and D. tenuifolium (Stipanicic and Menendez 1949: Bonetti and Herbst 1964; Herbst et al. 1998) from the Late Triassic of Argentina and Chile are somewhat similar in gross morphology to D. davidii but because of their geographical and time separation I consider they should be regarded as separate entities. Genus Hausmannia Dunker 1846 Type species H. dichotoma Dunker 1846 Hausmannia reticulata Holmes sp. nov. Figures 3C-E. Diagnosis Lamina sub-orbicular to reniform; margin entire; primary veins radiating from lamina base, forking and joining two to three times to form elongate areoles. Proc. Linn. Soc. N.S.W., 123. 2001 W.B.K. HOLMES 57 Description The holotype (Figs. 3C,D) is reniform, 25 mm long, 17 mm wide, with rounded lobes symmetrical about the base. Seven veins radiate from the base, forking and conjoining twice to form two transverse rows of broad elongate areoles. The paratype (Fig.3E) is 16 mm wide and 14 mm wide as preserved with basal lobes missing. The radiating veins fork and join two or three times to form two or three rows of areoles more elongate than the holotype. No secondary venation is preserved on either of the specimens. Holotype AMEF113391. Paratype AMF113392. Australian Museum, Sydney. Type Locality Coal Mine Quarry, Nymboida. Basin Creek Formation, Nymboida Coal Measures. Name Derivation reticulata Latin, net-like. Discussion This taxon is extremely rare. Two specimens only have been found during thirty five years of collecting. Despite the lack of preserved secondary veins at right angles to the primary veins, which in other species of Hausmannia form a fine network of quadrangular areoles over the whole lamina, the gross morphology of the Nymboida material suggests its inclusion in Hausmannia sensu Herbst (1992). With the exception of Hausmannia (Protorhipis) sp. cf. H. (P.) deferrariisii from the Jurassic of Queensland (Herbst 1979), which has eight primary veins that enter the deeply dissected lamina and branch and rejoin several times to form a network of decreasing sized meshes anastomosing veins, H. reticulata is the only other known species with anastomosing primary veins. Leaves referred to Chiropteris (Walkom 1925b; Du Toit 1927), Gingkophytopsis (Retallack 1980, 1983) and other new genera in studies yet to be published by Herbst in Argentina and Anderson and Anderson in South Africa, have laminae with reticulate but finer venation. The more elongate character of the meshes and the cuneate to digitate outline of the lamina, distinguish those leaves from H. reticulata. Incertae sedis Nymbofelicia Holmes gen. nov. Nymbofelicia aggregata Holmes gen. et sp. nov. Figures 24 A-D, 25 A-C. Combined Diagnosis Bipinnate fertile frond with opposite pinnae bearing linear lobed pinnules. Pinnules with straight median vein and well-spaced lateral veins, one to each lobe and once broadly forking. Fertile pinnules bearing circular sori each formed from a loose aggregate of 10-15 spheroidal sporangia centred below the fork of each lateral vein. Description The type specimen is a fragment of a bipinnate frond (Fig. 24A) with the preserved portion of a smooth surfaced primary rachis decreasing from 9 mm to 8 mm over a distance of 10 cm - thus suggesting that the complete frond was at least one metre in length. Pinnae opposite, attached 20 mm apart, at 50°-60° to the primary rachis, 20-22 mm wide, length to c. 70 mm. Pinnules closely spaced, slightly decurved or at right angles to the Proc. Linn. Soc. N.S.W., 123. 2001 58 TRIASSIC FLORA FROM NYMBOIDA — FILICOPHYTA pinna rachis (Fig. 24B-D), attached by whole or slightly contracted or decurrent base, linear with acuminate apex, 10-15 mm long, 3 mm wide, margin lobate, c. 7-9 lobes on either side of proximal pinnules, midvein fine and straight with one lateral vein to each lobe leaving midvein at c. 45° and once broadly forked (24D, 25A). Fertile pinnules bearing circular sori c. 1-1.5 mm in diameter centred in each lobe below the fork of the lateral vein. Sori composed of a cluster of c. 10-15 spherical sporangia 0.2-0.4 mm in diameter (Figs 24C, 25B, C). No annulus observed. Holotype AMPF113448, Australian Museum, Sydney. Type Locality Coal Mine Quarry, Nymboida. Basin Creek Formation, Nymboida Coal Measures. Other Material AMF113476-7, AMF113479-80. Name Derivation Nymbofelicia - contrived - for Nymboida fern aggregata - Latin - in clusters; referring to the arrangement of the groups of sporangia. Discussion Nymbofelicia aggregata is based on a single fertile fragment and a few sterile fragments. The genus Chansitheca was erected for fertile ferns from the Palaeozoic of China (Halle 1927; Boureau 1970) with pinnules bearing rounded sori comprised of from 8-16 sporangia, each with a distinct annulus — a feature not seen on Nymbofelicia. Chansitheca argentina was described by Herbst (1963) for a fertile fern fragment from the Upper Triassic of Patagonia. The specimen had pecopteroid pinnules bearing irregular ovoid sori composed of 8-12 pedunculate sporangia. An annulus was not apparent. Herbst also listed other differences to Chansitheca sensu stricta. Chansitheca argentina may be generically similar to Nymbofelicia but differs from N. aggregata by the broader pinnules and the irregular arrangement of the sori which are composed of fewer sporangia which form elongated ovals along the lateral veins. Illustrated sterile specimens of Cladophlebis mendozaensis (Geinitz) Frenguelli (Frenguelli 1947, pl. 11 fig. 7; Retallack et al 1977, fig. 5B; Herbst 1978, pl. 1, fig. 4) and Cladophlebis johnstonii Walkom (Hill et al. 1965, pl.T2, figs 4,5; Jain and Delevoryas 1967) have elongate lobed pinnules similar in outline to Nymbofelicia aggregata but differ by the lateral veins being twice forked. Todites maoricus Retallack (Retallack 1981, fig. 1, fig. 11A; 1983, fig. 3A-C) also has lobed pinnules similar to N. aggregata but differs by the twice forking lateral veins and by the contiguous sori covering most of the pinnule surface. ACKNOWLEDGEMENTS I wish to thank my family for assistance and co-operation over many years; Mrs Adela Romanowski for providing many of the photographic prints; the curators at the Natural History Museum, London, the Australian Museum, Sydney, the Queensland Museum, Brisbane, the University of Queensland and the University of New England for providing access to their fossil plant collections; Dr H.M. Anderson for much assistance; Dr R. Herbst for helpful comments; the Director of the National Botanical Research Institute, Pretoria, South Africa for providing research facilities; the two anonymous referees who provided valuable comments and suggestions, and the Joyce Vickery Research Fund for some financial support to examine the Queensland material. Proc. Linn. Soc. N.S.W., 123. 2001 W.B.K. HOLMES 59 REFERENCES Anderson, J.M. and Anderson, H.M., (1983). ‘Palaeoflora of Southern Africa. Molteno Formation (Triassic) Volume |. Part 1. Introduction. Part 2.‘Dicroidium’. (A.A. Balkema. Rotterdam). Anderson, J.M., Anderson, H.M. and Cruickshank, A.R.I., (1998). Late Triassic ecosystems of the Molteno/ Lower Elliot biome of southern Africa. Palaeontology 41, 387-421. Archangelsky, S., (1965). Taphofloras paleozoicas y eomesozoicas de Argentina. Boletin de la Sociedad Argentina Botanica 10, 247-291. Bonetti, M.R.I. and Herbst, R., (1964). Dos especies de Dictyophyllum del Tridsico de Paso Flores, Provincia Neuquen, Argentina. Ameghiniana 3, 273-279. Boureau, E., (1970). ‘Traité de Paléobotanique. Vol. 4. Pt.1. Filicophyta’. (Masson et Cie. Paris). Boureau, E., (1975). ‘Traité de Paléobotanique. Vol. 4. Pt.2. Pteridophylla’. (Masson et Cie. Paris). Bourke, D.J., Gould, R.E., Helby, R., Morgan, R. and Retallack, G.J., (1977). Floral evidence for a Middle Triassic Age of the Gunnee Beds and Gragin Conglomerate, near Delungra, New South Wales. Journal and Proceedings of the Royal Society NSW 110, 33-40. Burges, N.A., (1935). Additions to our knowledge of the flora of the Narrabeen Stage of the Hawkesbury Series in New South Wales. Proceedings of the Linnean Society NSW 60, 257-264. Cairncross, B., Anderson, J.M. and Anderson, H.M., (1995). Palaeoecology of the Triassic Molteno Formation, Karoo Basin, South Africa — sedimentological and palaeontological evidence. South African Journal of Geology 98, 452-478. de Jersey, N.J, (1958). Macro- and micro- floras from north-eastern New South Wales. Journal and Proceedings of the Royal Society of NSW 92, 83-89. de la Sota E.R. and Archangelsky, S., (1962). Dos nuevas especies de Asterotheca de la Serie Tridsica ‘El Tranquilo’, Provincia Santa Cruz. Ameghiniana 2, 113-119. du Toit, A.L., (1927). The fossil flora of the Upper Karoo beds. Annals of the South African Museum 22, 289- 420. Faber, J. and Greber, C., (1960). Presence d’un Dictyophyllum dans la flore Molteno du Basutoland (Afrique Australe). Bulletin Société Géologique de France, Series 7, 2, 178-182. Flint, J.C.E., and Gould, R.E., (1975). A note on the fossil megafloras of the Nymboida and Redcliff Coal Measures, southern Clarence-Moreton Basin. Journal and Proceedings of the Royal Society NSW 108, 70-74. Frenguelli, J., (1943). Contribuciones al conocimiento de la flora de Gondwana superior en Argentina. 14. Asterotheca truempyi n.sp. Notas Museo La Plata 8 Paleontologia 60, 421-430. Frenguelli, J., (1947). El genero “Cladophlebis’ y sus representantes en la Argentina. Annales de Museo de La Plata n.s. 2, 4-74. Harris, T.M., (1931). Fossil flora of Scoresby Sound, East Greenland. Meddelelser om Grgnland 85, 1-102. Harris, T.M., (1961). “The Yorkshire Jurassic Flora. 1. Thallophyta‘— Pteridophyta’. (British Museum (Natural History ). London). Herbst, R., (1963). Chansitheca argentina n. sp. del Triasico Superior de Sanat Cruz, Patagonia. Ameghiniana 3, 108-112. Herbst, R., (1972). Gleichenites potrellos n. sp. de Triasico medio de Mendoza (Argentina), con comentarios sobre Gleicheniaceae fosiles de Argentina. Ameghiniana 9, 17-22. Herbst, R., (1974). Note on two Triassic plants from Queensland, Australia. Proceedings of the Royal Society of Queensland 85, 79-84. Herbst, R., (1975). Consideraciones sobre las Dipteridaceae (Filicales) de Gondwana. Acta Congr. Argentina, Paleontologia, Biostratigrafia. 1, 525-535. Herbst, R., (1976). Sobre Marattiales (Filicopsida) Tridsicas de Argentina y Australia. Parte 1. El genero Asterotheca. Ameghiniana 14, 1-19. Herbst, R., (1977a). Sobre Marrattiales (Filicopsida) Tridsicas de Argentina. Parte 1. El genero Asterotheca. Ameghiniana 14, 1-18. Herbst, R., (1977b). Sobre Marattiales (Filicopsida) Triasicas de Argentina y Australia. Parte 2. Los géneros Danaeopsis y Reinitsia. Ameghiniana 14, 19-32. Herbst, R., (1978). Revision de las especies Australasicas de Cladophlebis (Bgt). Facena Corrientes Argentina 2, 3-28. Herbst, R., (1979). Review of the Australian Dipteridaceae. Proceedings of the Linnean Society of NSW (1978) 103, 7-21. Herbst, R., (1989). La flora tridsica de Grupa el Tranquilo, Provincia de Santa Cruz (Patagonia). Parte 2. Filicopsida. Ameghiniana (1988) 25, 365-379. Herbst, R., (1996). New species of Gleichenites (Gleicheniaceae, Filicales) from the Upper Triassic of Argentina and Chile. Palaeobotanist 43, 67-72 Herbst, R., Melchor, R. And Troncoso, A., (1998). Las pteridophyta y el paleoambiente de la Formacién La Ternera (Tridsico Superior), en Quebrada La Cachivarita, 111 Region, Chile. Revista Geologica de Chile 25, 85-107 Hill, D., Playford G. And Woods, J.T., (1965). ‘Triassic Fossils of Queensland.’ (Queensland Palaeontolographical Society. Brisbane). Proc. Linn. Soc. N.S.W., 123. 2001 60 TRIASSIC FLORA FROM NYMBOIDA — FILICOPHYTA Holmes, W.B.K., (1982). The Middle Triassic flora from Benolong, near Dubbo, central-western New South Wales. Alcheringa 6, 1-33. Holmes, W.B.K., (2000). The Middle Triassic megafossil flora of the Basin Creek Formation, Nymboida Coal Measures, New South Wales, Australia. Part 1. Bryophyta, Sphenophyta. Proceedings of the Linnean Society of NSW 122, 43-68. Jain, R.K. and Delevoryas, T., (1967). A Middle Triassic flora from the Cacheuta Formation, Minas Petréleo, Argentina. Palaeontology 10, 564-589. Jones, O.A. and de Jersey, N.J., (1947). The flora of the Ipswich Coal Measures — morphology and floral succession. Department of Geology, University of Queensland Paper n.s. 3, 1-88. Kurtz, F., (1921). Atlas de Plantas Fésiles de la Republique Argentina. Actas de la Academia Nacional de Ciencias Cordéba 7, 129-153. MacRea, C., (1999). ‘Life Etched in Stone‘— Fossils of South Africa’. (Geological Society of South Africa, Johannesburg). McElroy, C.T., (1963). The geology of the Clarence—Moreton Basin. Geological Survey of N.S.W. Memoir 9, 1-172. Menendez, C.A., (1957). Asterotheca hilariensis n.sp. de Tridsico superior de Hilario, San Juan, Argentina. Ameghiniana 1, 25-30. Meyen, S.V., (1987). ‘Fundamentals of Palaeobotany’. (Chapman and Hall, London). Phipps, C.J., Axsmith, B.J., Taylor, T.N. and Taylor, E.L., (2000). Gleichenipteris antarcticus gen. et sp. nov. from the Triassic of Antarctica. Review of Palaeobotany and Palynology 108, 75- 83 Playford, G., Rigby, J.F. and Archibald, D.C., (1982). A Middle Triassic flora from the Moolayember Formation, Bowen Basin, Queensland. Geological Survey of Queensland, Publication 380, 1-52. Retallack, G.J., (1977). Reconstructing Triassic vegetation of eastern Australia: a new approach for the biostratigraphy of Gondwanaland. Alcheringa 1, 247-277. Retallack, G.J., (1981). Middle Triassic megafossil plants from Long Gully, near Otematata, north Otago, New Zealand. Journal of the Royal Society of New Zealand 11, 167-200. Retallack, G.J., (1983). Middle Triassic megafossil marine algae and land plants from near Benmore Dam, southern Canterbury, New Zealand. Journal of the Royal Society of New Zealand 13, 129-154. Retallack, G.J., Gould,R.E. and Runnegar, B., (1977). Isotopic dating of a Middle Triassic megafossil flora from near Nymboida, north-eastern New South Wales. Proceedings of the Linnean Society of NSW 101, 77-113. Retallack, G.J., Renne, P.R. and Kimbrough, D.L., (1993). New radiometric ages for Triassic floras of south- eastern Gondwana. In, S.G. Lucas and M. Moriales, Eds.,“The Non-marine Triassic’. New Mexico Museum Natural History Science Bulletin 3, 415-418. Retallack, G.J., Veevers, J.J. and Morante, R., (1996). Global coal gap between Permian- Triassic extinction and Middle Triassic recovery of peat-forming plants. Geological Society of America, Bulletin 108, 195-207. Rigby, J.F., (1977). New collections of Triassic plants from the Esk Formation, south-east Queensland. Queensland Government Mining Journal 78, 320-325. Russell, N.J., (1994). A palaeothermal study of the Clarence-Moreton Basin. Australian Geological Survey Organisation Bulletin 241, 237-276. Scotese, C.R., (1994). Early and Late Triassic palaeographic maps. In.- de V.Kleinig, Ed., “Pangea: Palaeoclimate, Tectonics and Sedimentation during Accretion, Zenith and Breakup of a Supercontinent’. Geological Society of America Special Paper 288, 7 Shirley, J., (1898). Additions to the fossil flora of Queensland. Geological Survey of Queensland Publication 1, easy. Stipanicic, P.N. and Menendez, C.A., (1949). Contribuci6n al conocimiento de la flora fésil Barreal (Provincia de San Juan) 1. Dipteridaceae. Boletin de Informaciones Petroleras 291, 44-73. Taylor, T.M., Taylor, E.L., Meyer-Berthaud, B., Isbell, J.L. and Cuneo, N.R., (1990). Triassic osmundaceous ferns from the Allan Hills, southern Victoria Land. Antarctic Journal 1990 Review 25, 18-19. Thomas, H.H., (1933). On some pteridospermous plants from the Mesozoic rocks of South Africa. Philosophical Transactions of the Royal Society. Series B, Biological Sciences 222, 193-265. Townrow, J., (1957). On Dicroidium, probably a pteridospermous leaf, and other leaves now removed from this genus. Transactions of the Geological Society of South Africa 60, 1-36. Vallance, T.G., (1978). Arthur Bache Walkom 1889-1976. Memorial Series Number 25. Proceedings of the Linnean Society of NSW 102, 148-155. Walkom, A.B., (1917). Mesozoic floras of Queensland. Pt.1 (cont.) The flora of the Ipswich and Walloon Series. (c) Filicales. Queensland Geological Survey Publication 257, 1-46. Walkom, A.B., (1924). On fossil plants, near Bellevue, Esk. Queensland Museum Memoir 8, 77-92. Walkom, A.B., (1925). Notes on some Tasmanian Mesozoic plants. Part 2. Papers and Proceedings, Royal Society of Tasmania (1925), 63-74. Walkom, A.B., (1928). Fossil plants from the Esk district, Queensland. Proceedings of the Linnean Society of NSW 53, 458-467. Walkom, A.B., (1932). Fossil plants from Mount Piddington and Clarence Siding. Proceedings of the Linnean Society of NSW 57, 123-126. Proc. Linn. Soc. N.S.W., 123. 2001 W.B.K. HOLMES 61 Webb, J.A., (1982). Triassic species of Dictyophyllum from eastern Australia. Alcheringa 6, 79-91. Webb, J.A., (1983). A new plant genus, possibly a Marattealean fern from the Middle Triassic of eastern Australia. Memoir of the Association of Australasian Palaeontologists 1, 363-371. Webb, J.A., (2001). Marantoidea acara sp. nov., anew Marattealean fern from the Middle Triassic of eastern Australia. Proceedings of the Linnean Society of NSW 123, Legends for figures Figure 1.(A-C). Ogmos adinus. (A) AMF113382, X0.85. (B) AMF113383, group of superimposed leaves, X0.6. (C) AMF113384, X0.5. Scale bar = 1 cm. Figure 2.(A). Marantoidea acara. AMF113387, distal portion of sterile frond. Scale bar = 1 cm. Figure 3.(A-B). Marantoidea acara. Fertile pinnae. (A) AMF113379, X2. (B) AMF113390, X2. (C—E) Hausmannia recticulata. (C,D) Holotype, AMF113391. (D) X2), (E) AMF113392, X2. Scale bar = 1 cm. Figure 4.(A-B). Rhinipteris walkomii. (A) Holotype, AMF113393 X2. (B) Paratype, sterile foliage, AMF113394. Scale bar = 1 cm. Figure 5.(A—-C). Rhinipteris walkomii, Paratypes (A) AMF113395, base of primary rhachis. (B) AMF113396, fertile pinnae. (C) AMF113397, sterile and fertile pinnae. Scale bar = | cm. Figure 6.(A). Asterotheca trullensis. Holotype. Fertle fronds, AMF113400. Scale bar = 1 cm. Figure 7.(A-C). Asterotheca trullensis. (A,B) Fertile pinnae, X2. (A) AMF113400. (B) AMF113451. (C) Sterile frond, AMF113478. Scale bar = 1 cm. Figure 8.(A). Asterotheca nymboidensis. (A) AMF113409. Frond bearing both sterile and fertile pinnules. Scale bar = | cm. Figure 9.(A—-C). Asterotheca nymboidensis. All X2. (A) AMF113408, Holotype, showing fertile and sterile pinnules on same pinna. (B) AMF113410, fertile pinnae. (C) AMF113411, fertile pinnae on primary rachis. Scale bar = 1 cm. Figure 10.(A—C). Asterotheca nymboidensis. All AMF113412. (A) X15. (B) X30. (C) X60, showing longitudinally striated sporangial walls. Scale bar = 1 mm. Figure 11.(A). Asterotheca chevronervia. Holotype, AMF113414. Scale bar = 1 cm Figure 12.(A-D). Asterotheca chevronervia. (A-C) Portions of the Holotype, AMF113414 (A) X1. (B-C) Pinnae with sterile and fertile pinnules, X2. (D) AMF113415, sterile pinnae showing chevron-like lateral venation on the pinnules. Scale bar = 1 cm. Figure 13.(A). Asterotheca sp A, AMF113417, X2. (B) Asterotheca sp.B, QMF42305, X2. (C) Asterotheca sp.C, AMF113418, X2. Asterotheca sp.D, AMF113419, X2. Scale bar = 1 cm. Figure 14.(A). A. Asterotheca diameson. AMF113420, Holotype, X1.2. (B) Gleichenites wivenhoensis. AMF 113422, X2. (C) Herbstopteris sp.A, AMF113423. Scale bar = 1 cm. Figure 15.(A,B). Herbstopteris colliveri. AMF113425. (A) Fronds radiating from stout rhizome or trunk. (B) Apical portion of fertile frond. Scale bar = 1 cm. Figure 16.(A-C). Herbstopteris colliveri. (A) AMF113426, X2. (B) AMF11342, X2. (C) AMF113429, X2. Scale bar = | cm. 7 Figure 17.(A-D). Herbstopteris colliveri. (A) AMF113430, X2, (B) AMF113431, X2, (C) AMF113427, X10. (D) AMF113428, X10. Scale bar = 1 cm for (A,B), | mm for (C,D). Figure 18.(A-E). Todites parvum. (A,B) AMF113432, Holotype. (A) X10. (B) X2. (C) AMF113434, X2. (D) AMF113435, X2. (E) AMF113436, X2. Scale bar = 1 cm for (A, C-E), 1 mm for (B). Proc. Linn. Soc. N.S.W., 123. 2001 62 TRIASSIC FLORA FROM NYMBOIDA — FILICOPHYTA Figure 19.(A,B) Osmundopsis scalaris. (A) AMF113468, Holotype, fern with dimorphic fertile and sterile fronds. (B) AMF113438, sterile pinna to show venation, X2. Scale bar = | cm. Figure 20.(A-C). Osmundopsis scalaris. (A) AMF 113468, Holotype, portion of fertile frond. (B) AMF113469, sterile pinna, X2. (C) AMF113443, fertile pinna, X10. Scale bar = 1 cm for (A,B), 1 mm for (C). Figure 21.(A). Osmundopsis scalaris. AMF 113447, three sterile fronds radiating from ?common base. Scale bar = 1 cm. Figure 22.(A-E). Dictyophyllum davidii. (A) AMF113394, with long primary rachis. (B) AMF113375. (C) AMF113376. (D) AMF113377. (E) AMF113378, showing tertiary venation, X2. Scale bar = 1 cm. Figure 23.(A-C). Dictyophyllum davidii. (A) AMF113379. (B) AMF113380. (C) AMF113381, fertile frond, X2. Scale bar = 1 cm. Figure 24.(A-D). Nymbofelicia aggregata. (A-C) AMF113448, Holotype. (A) X1. (B) X2.5. (C) X10. (D) AMF113476, sterile pinna, X1. Scale bar = 1 cm for (A,B and D), 1 mm for (C). Figure 25.(A-C). Nymbofelicia aggregata. AMF113448, Holotype. (A) pinna showing once forked venation, X20. (B) fertile pinna, X20. (C) fertile pinna, X40. Scale bar = 1 mm. Proc. Linn. Soc. N.S.W., 123. 2001 W.B.K. HOLMES FIGURE 1 63 TRIASSIC FLORA FROM NYMBOIDA — FILICOPHYTA FIGURE 2 W.B.K. HOLMES FIGURE 3 Sat 65 TRIASSIC FLORA FROM NYMBOIDA — FILICOPHYTA 66 NRC - FIGURE 4 W.B.K. HOLMES , ¥ yy. ict lieth Page FIGURE 5 67 TRIASSIC FLORA FROM NYMBOIDA — FILICOPHYTA 68 FIGURE 6 W.B.K. HOLMES FIGURE 7 TRIASSIC FLORA FROM NYMBOIDA — FILICOPHYTA FIGURE 8 W.B.K. HOLMES FIGURE 9 71 TRIASSIC FLORA FROM NYMBOIDA — FILICOPHYTA FIGURE 10 W.B.K. HOLMES FIGURE 11 TRIASSIC FLORA FROM NYMBOIDA — FILICOPHYTA 74 FIGURE 12 W.B.K. HOLMES 75 FIGURE 13 TRIASSIC FLORA FROM NYMBOIDA — FILICOPHYTA 76 6 PE bv PAE FIGURE 14 W.B.K. HOLMES i, j 4 &% ‘ 4 FIGURE 15 78 TRIASSIC FLORA FROM NYMBOIDA — FILICOPHYTA FIGURE 16 79 W.B.K. HOLMES FIGURE 17 80 TRIASSIC FLORA FROM NYMBOIDA — FILICOPHYTA FIGURE 18 81 W.B.K. HOLMES FIGURE 19 2 TRIASSIC FLORA FROM NYMBOIDA — FILICOPHYTA FIGURE 20 W.B.K. HOLMES ae FIGURE 21 83 84 TRIASSIC FLORA FROM NYMBOIDA — FILICOPHYTA FIGURE 22 85 W.B.K. HOLMES 4 a Ay y os wd © fe ot et FIGURE 23 TRIASSIC FLORA FROM NYMBOIDA — FILICOPHYTA 86 FIGURE 24 87 W.B.K. HOLMES FIGURE 25 The Diet of the Pilliga Mouse, Pseudomys pilligaensis (Rodentia: Muridae) from the Pilliga Scrub, Northern New South Wales ELIZABETH A. JEFFERYS AND BARRY J. Fox School of Biological Science, University of New South Wales, Sydney NSW 2052 Jefferys, E.A. and Fox, B.J. (2001). The diet of the Pilliga mouse, Pseudomys pilligaensis (Rodentia: Muridae) from the Pilliga Scrub, northern New South Wales. Proceedings of the Linnean Society of New South Wales 123, 89-99. The contents of the digestive tracts from five Pilliga mice ( Pseudomys pilligaensis), collected over twelve years, were processed. Seed was the most abundant food in the diet and was represented by five or six seed types in Spring/Summer and accounted for 95% of the diet, with leaf contributing less than 2%. InWinter, seed contributed 62% of diet, represented by more than twelve seed types with leaf contributing 38% of the diet. These results indicate that P pilligaensis is a specialised granivore, although it may be more opportunistic in winter. Manuscript received 21 May 2001, accepted for publication 19 September 2001. KEYWORDS: diet, granivore, Pilliga mouse, Pseudomys pilligaensis. INTRODUCTION The Pilliga mouse, Pseudomys pilligaensis (Rodentia: Muridae) was first captured in 1975 and was thought then to be an unusually small specimen of the New Holland Mouse (Pseudomys novaehollandiae). However, material collected over the following years revealed significant morphological differences between the animals from the Pilliga and other Conilurine rodents and a new species was described by Fox and Briscoe (1980). Briscoe, et al (1981) used electrophoretic analysis to clearly demonstrate genetic differentiation between Pseudomys pilligaensis and related Pseudomys species. Very little is known about the Pilliga mouse except that it is terrestrial and appears to live in burrows, often in sandy substrate. It apparently has an extremely restricted distribution, and is known only from the Pilliga Scrub including Pilliga State Forests. This area is situated north of Coonabarabran in northern New South Wales. It is a forest on an isolate sand substrate, supporting a mixed Callitris-Cypress-Pine-Eucalyptus forest and heath communities (Fox and Briscoe 1980). All known individuals captured up until 1988 appear to have been within a 50 km radius. The status of the Pilliga mouse has been listed as rare, limited, vulnerable and perhaps at risk but possibly under estimated (Strahan 1995). The Pilliga mouse is nocturnal and inhabits areas with sparse ground cover. The Pilliga mouse is similar in appearance to the New Holland Mouse, Pseudomys novaehollandiae which has a coastal distribution and is restricted to sandy areas. Both are nocturnal and inhabit areas with sparse ground cover. Proc. Linn. Soc. N.S.W., 123. 2001 90 DIET OF THE PILLIGA MOUSE novaehollandiae which has a coastal distribution and is restricted to sandy areas. Both are nocturnal and inhabit areas with sparse ground cover. In captivity, animals live in family groups. The breeding season ( in captivity ) extends at least from October to mid-February and the gestation period is between 24 and 31 days. Young are born with the incisors erupted and eight laboratory-born litters had a modal size of three young (Fox and Briscoe 1980). There are no published data on the diet of this species. This study examined the diet of five wild-caught individuals of Pseudomys pilligaensis, using microscopic analysis of faecal samples which were collected over a twelve year period. Any knowledge of the natural diet and cues for habitat selection are crucial in understanding the biology of any rare native rodent (Cockburn 1981a,b). This information could substantially contribute to the understanding of its habitat requirements and to the conservation management of this rare and endangered species. STUDY SITES The five individuals were all captured from localities within the Pilliga Scrub area (Table 1). The specimens came from habitats in mixed cypress-eucalypt forest and woodland. The dominant canopy species were red stringy bark (Eucalyptus macroryncha), scribbly gum (Eucalyptus rossii) and black cypress pine (Callitris endlicheri) with Blakely’s red gum (Eucalyptus blakelyi) and rough barked apple (Angophora floribunda) in creek lines, which also had stands of bottle brush (Callistemon) and tea tree (Leptospermum) in the understorey. Apart from creek lines the sparse understorey comprised heath-type shrubs similar to the She-Oak (Casuarina distyla) dominated heath adjacent to the forest. More detailed descriptions of the vegetation and a plant species list are provided by Fox and Briscoe (1980). The Pilliga State Forests have a very unpredictable physical environment with annual rainfall around 625mm with a summer maximum. Variation in annual rainfall is great, with periods of both great drought and extreme wet being common (Norris et al.1991; Hart 1995). METHODS The digestive tracts of five spirit specimens of Pseudomys pilligaensis were carefully removed from the body cavity and opened using fine scissors and fine jewellers forceps. Three of the specimens contained faecel pellets only (BE. 791103; BE. 791105 and NPWS 1-88) while the remaining two specimens contained material in all sections of the digestive tract (A.M. M10438; and A.M. M10602) (see Table 1,specimens No. 1, 3 and 4 , described by Fox and Briscoe 1980). Hence to be consistent, only the faecal pellet data were used in comparing the diets. The contents of each section of the tract were flushed with 75% alcohol and placed into separate containers and categories ( i.e. caecum, stomach, large intestine). Faecal pellets when present were carefully removed from the rectum. Thus one sample consists of five faecal pellets that were randomly selected from one mouse. Each sample was washed through a column of five different sieves with 0.58, 0.41, 0.26, 0.15, and 0.13mm mesh. Any material passing through the smallest sieve was collected on filter paper was divided into six sub samples. Each subsample was transferred into a labelled test tube to allow materials to be homogeneously mixed. The material was then transferred from each tube with a pipette to a correspondingly labelled slide (25.4 x 76.2mm) which was placed on a hotplate at 30C to dry. After all moisture on the slides was evaporated the slides were removed from the hotplate, placed on a flat surface and slowly covered with modified Berlese mounting medium (Luo 1993). The completed slides were then left Proc. Linn. Soc. N.S.W., 123. 2001 91 E.A. JEFFERYS AND B.J. FOX JOIAIOS JJIPIIM pue syieg [euOTRN = SMdN ‘XOJ Aueg = qq ‘Wnosnyy uelensny = WY ‘sioquiny onsojeyeD WITT sulids 8861 1498S 61 DAIOSOY OINEN CSI [Id 88-1 “SMdN ‘S (4.87 o6r1 “S.IS ol €) xour|d sulids 6L61' AON ool) olfeypeurunr p SOII6L “AP ‘YP (4.87 o6tI ‘S.IS ol €) xour gd sulids 6L61' AON oor) oif[eyyeurunt | COII6L AG '€ (4.91 o6tI ‘S.IE o0€) Jase[ys 4 OVUM, LL6t Aqne JSOIOJ IS “| eBIIIId ZOMOIIN NV “7 (4.87 o6tI ‘S.1S ol £) asoyV OVUM 9L61 AIne pI YooID o[eypeuruarl | SErOlIN WV ‘I “su0T “eT “ON onsoyeyeg I0}D9][OD uoseas po}d9][09 93e8q — uol}eD0'T / ‘ON uouttsads Bye UONI][OD pure sisuansiyjid sXuopnasg JO sonyesoy7T “[ 9192]. Proc. Linn. Soc. N.S.W., 123. 2001 92 DIET OF THE PILLIGA MOUSE undisturbed for 2-3 days until the mountant set hard. This modified medium not only clears the material but also has an ability to preserve and retain the original appearance of the faecal pellet material, whether that be fresh green material or old yellow mature plant material. The amount of material on each slide was standardised using a scoring (frequency- of-occurrence) technique as described by Johnson 1982; Luo 1993; Fox, Read, Jefferys and Luo 1994 . Estimation of Dietary Composition Generally the food material in the samples of P pilligaensis could not be identified to species and was classified as leaf, stem, seed, insect, pollen, sand granules or unidentified (following the protocol of Fox, Read, Jefferys and Luo 1994). As rodents tend to masticate their food very finely, positive identification of food items (e.g. epidermal tissue) is extremely difficult. Some small pieces of plant material could be identified to genus by comparison with the plant slide collection in the ecology laboratory at the University of New South Wales. All six slides from each sample were scored because different dietary items were best represented on slides with the appropriate sized particles. A total of 50 fields of view at 40 x magnification were scored for each of the six slides from each sample. The frequency of occurrence of each food category was recorded from its presence or absence in each field. Next the number of scores or frequency for each food item was summed over all slides and the percentage of each category in the diet, from 300 fields of view, was calculated as P.=(f/2f;) x 100% where P; is the percentage and f; the frequency of i” food item. Although some authors have calculated frequency percentages as the number of fields in which a food category occurred out of 100 fields, then converted them into relative percentage of occurrence (e.g. Sparkes and Malechek 1968; Cockburn 1980; Carron , Happold and Bubela. 1990), the above equation determines the relative percentage of occurrence of each food item irrespective of the number of fields scored. Detailed discussion of the techniques we have used are presented in Luo 1993; Fox, Read, Jefferys and Luo 1994; Luo , Fox and Jefferys 1994. RESULTS Table 2 shows the percentage dietary composition of the five faecal samples of the Pilliga mice collected in the wild. Overall plant material (stem, seed, and leaf) was the most abundant food in the diet of Ppilliagensis, accounting for more than 94% of its total diet. There was no pollen present in any of the samples. The bulk of the plant material was seed, over 80%, and 22 seed types were recognised, 12 of which were monocots and 10 seed types recognized as dicots. Table 3 shows seed types found in the diet of Pseudomys pilligaensis. The seed types 16 to 22 ocurred only at trace levels. The seeds were finely masticated but some particles were large enough to make some general identifications. The Monocot seed material appears to come from the genera of Cyperus, Lepidosperma and Schoenus, (Cyperaceae); Poa, Juncus, Aristida, Cymbopogon, Dichelachne , Digitaria, Entolasia, Eragrostis, Imperata, and Triodia spp. (Poaceae) all of which are found in the Pilliga Scrub area. Leaf material found in the diet varied from 1% in spring to 52% in winter. Stem material varied from 4% in spring to 1% in winter Sample No. 3 (BF. 791103) contained small traces of leaf of the plant species Aotus subglauca, family Fabaceae. Most of the Proc. Linn. Soc. N.S.W., 123. 2001 E.A. JEFFERYS AND B.J. FOX sulds suds sutids AyTuo 99%.) A]uo 99%.) v 0 I I 66 66 sodA} poos p= ‘sadA} poas 9 = GONG LS OSI G 1 JOyUIM G VC 0 9L sodA} poos ¢][ = We Al WS JoyUIM ATUo 90e) (6S LV sodA} pods Z[ = 6I-LVZ101-8°9°S €-1 NOSVdS LOUSNI AVAT WHLS dads SddAL dads ON uoulsads ‘sojduues |e Ul yUdSoId SBM PUL ‘S[eNPIAIPUI Sisuavsi]]Id ‘q IAI JO SOI9CJ OY} Ul SUID}I JOIP Jo UONISOduI0D oBeIUDIIOg *Z 91GUL Proc. Linn. Soc. N.S.W., 123. 2001 DIET OF THE PILLIGA MOUSE 94 Table 3. Seed types found in the diet of Pseudomys pilligaensis Seed Type _|Class sd Family | Gentuss/speeciess |Monocotyledon | Poaceae 2 |Monocotyledon | Poaceae CCs‘Digiitaria_sp. 3 |Monocotyledon | Poaceae CP sp. 4 | Monocotyledon___—| Poaceae CC Stipa_ sp. 5 [ienossivieion | eesiiies 6 IBWTorieic ty Jel ee Berassitype 7 Montcotyledoniayi nian | aa NN Perassitype 8 Monocotyledonias Kaass =| Teenie OP) al Werassitype 9 [Borat coyl ecto ripen mee | een eI Werissitype 10 |Monocotyledon | Cyperaceae | Cyperus sp. in | Monocotyledon | Cyperaceae_ i Cyperus sp. 12 |Monocotyledon | Juncaceae | Juncus sp. 13 |Dicotyledon | Goodeniaceae | Goodenia_sp. 14 | Dicotyledon | unknown family 15 | Dicotyledon | unknown family 16 |Dicotyledon | Portulaceae 17 | Dicotyledon | “Myrtaceae Anngophora sp. 18 |Dicotyledon | Solanaceae | Solanum sp. 19 EDicotyledonty Semana NE MMalavaccact 5 20 [Desikiot 2 || 21 svc 2 Se iio aa 22 Dain 7 ikon 7 Saas 123. 2001 N.S.W., Soc. LINN. Proc. E.A. JEFFERYS AND B.J. FOX 95 leaf material was Monocot and was associated with the seed eaten (i.e. glumes and lemma). All of the leaf material in sample No. 1 AM.M10438 was Monocot material and is either Poaceae or Cypeaceae family. Some of the leaf cells were identified as belonging to the family Poaceae (grasses) by the dumb-bell shaped silica bodies in the cells. Dicot leaf was observed. Fungal spores were present in one individual (No. 4 BF. 791105) but as only two single spores were present it would appear that this was accidentally ingested and does not appear to be part of their diet. Dirt and sand granules were present in all samples but in insignificant proportions. However as there is no evidence of any root material or any soil invertebrates, this suggests that the mice were probably not digging to acquire any of the food in their diet at these times. Insect material was found in all five individuals but only four of the samples were able to be identified. Larval skin from the same species of Lepidopteran caterpillar was found in both diets from spring (BF. 791103 and BE. 791105) winter (AM. M. 10602). Hymenopteran material identified as ant (Formicidae) alate wings was found in winter (AM. M. 10602) and spring (NPWS. 1-88), and a section of ant gaster (Formicidae) appears in spring (BF. 791103). A section of leg of Orthopteran (grasshopper type) was found in winter. (AM. M.10602) so they appear to be consuming similar material over this time period. Pseudomys pilligaensis appears to have a strictly granivorous diet during the spring/summer months (September to November), consuming from 91% to 95% of its diet as seed. This seed material is mainly yellow to brown and appears to be mature dried seed from 4 to 6 species. One type of seed (No.6, a grass type) appears in four of the five samples, and it appears in the diet in the years 1976, 1977 and 1979. However in the winter months they broaden the diet to include a much greater range of seed types and a higher proportion of leaf material. DISCUSSION It should be noted that the sample size is very small and any results will only be a limited appraisal of what the individuals were eating at the time of the sampling. The five samples represent one juvenile and four adults and were collected over a number of years. Winter samples comprised one adult in July 1976 AM. M.10438 and one juvenile July 1977 AM. M.10602. What we have classified as the spring/summer samples comprised two adults collected in November 1979, i.e. BE. 791103 and BE. 791105 and one adult from September 1988. Hence it is only possible to compare the diet for spring/summer and winter in a very broad manner. The results from the other sections of the digestive tract were very similar to those from the faecal pellets. This was also found to be the case with Pseudomys novaehollandiae diet analysis by Thompson (1980), with stomach contents having a close similarity to faecal pellets. Insects do not seem to be a very important part of the diet, as they only range from 0 to 6% of the total diet. Insects, especially ants and termites, are readily available in this habitat (Rolls 1981; Hart 1995 ) and would provide more calories per gram than seeds (Golley 1961, Reichman 1975, Redford and Dorea 1984). The insect material in all of the diets could possibly have been accidentally ingested as they all are associated with the parts of the plants that were consumed. However, it would appear that these rodents are deliberately selecting plant material instead of insects at these times of the years. A comparison of the diet of Pseudomys pilligaensis with those of other species of Pseudomys in both spring/summer and winter (Table 4) shows that P. pilligaensis like all other species demonstrated a drop in the percentage of seed from spring/summer to the winter diet, but the magnitude of the shift differs. Pseudomys novaehollandiae is almost entirely dependent on seed (97%) in summer but drops to 45% in winter, while P apodemoides drops from 86% in summer to 55% in winter ( Table 4). These changes contrast with the Proc. Linn. Soc. N.S.W., 123. 2001 96 DIET OF THE PILLIGA MOUSE Table 4, A comparison of diet items in the faeces of P. pilligaensis with those for five other species of Pseudomys in (A) Spring/summer and (B) winter. Values are the mean percentage occurence +standard error. Sources: Cockburn (1980); ?Thomson (1980); *Cockburn (198 1a), *Cockburn (198 1b): Fox et al (1994); Luo and Fox (1994). A) Spring/Summer diet items Species Leaf Stem Seed Fungi Insect Other Total (n=samples) Pseudomys NZ 13 95.0 0 2.0 0 100.0 pilligaensis tale) +0.3 4273 +0 +2.0 +0 (3) Pseudomys ' LS) - 96.9 0 0.6 0.6 100.0 novaehollandiae +0.9 stelle) +0 +0.6 +0.6 (5) Pseudomys 7 4.94 = 81.2 - 3.8 10.0 99.9 novaehollandiae +1.2 acl) +0.3 al 0) (35) Pseudomy * 4.9 4 @ 85.8 0 1.0 1.0 100.1 apodemoides stella) +0.7 +0 +0.2 +0.4 (15) Pseudomys* 1.4 ~ 58.6 19.8 12.6 4.9 99.9 fumeus (23) Pseudomys° 44.6 0.3 44.2 0.3 Tod 3.4 100.1 oralis te) +0.1 a4, a5 03) tele (29) Pseudomys® 4.6 Bi ail 34.2 22a 3.0 4.3 99.9 gracilicaudatus +1.9 a8) +329 seni a0) 9) +1.4 (25) a Leaf and stem conbined Proc. Linn. Soc. N.S.W., 123. 2001 E.A. JEFFERYS AND B.J. FOX 97 B) Winter diet items Species Leaf Stem Seed Fungi Insect Other Total (n=samples) Pseudomys 38.0 1.25 61.5 0 1.0 0 100.0 pilligaensis +14.0 +0.5 +14.5 +0 +1.0 +0 (2) Pseudomys' 18.9 44.6 38) 17.0 15.7 100.0 novaehollandiae +0.9 +8.9 58 )58) +14.0 +5.6 (7) Pseudomys? ily a 54.7 — 7.3 26.6 100.1 novaehollandiae +0.6 +0.3 +0.7 acl) (22) Pseudomys?* Sala a 558) 10.4 nee 15.0 100.0 apodemoides ae) +1.4 +0.6 atoll +1.8 (15) Pseudomys* 1.6 - 26.7 55.0 6.6 10.2 100.1 fumeus (23) Pseudomys° 77.0 6.5 11.6 el 28} ie) 100.0 oralis ap eS +1.8 +0.6 +0.4 +0.8 (29) Pseudomys® WA DOR lel 22.8 1.3 3.4 9919 gracilicaudatus +1.5 aeile// stale aE IL) +0.6 +0.9 (36) a Leaf and stem combined larger species, P oralis which has seed as 44% of its diet in summer and 12% in winter, the lowest use of seed in winter, but consumes 77% leaf in winter. Pseudomys gracilicaudatus shows little change in seed consumption from summer (34%) to winter (31%), but this holds true for all diet items so that this species shows the least seasonal change. All the other species increase the amount of leaf or stem in the diet in winter but P. oralis shows by far the greatest change from 45% in summer to 77% in winter. Pseudomys pilligaensis appears to be closest in diet to P novaehollandiae with its dependence on seed (95%) in summer to 62% in winter. The magnitude of shift in the diet is the lowest for the genus. Pseudomys_pilligaensis had the highest percentage of seed (62%) in its winter diet , which is followed closest by P novaehollandiae with 45% and P apodemoides with 55%. These observations may be a reflection of differences in the availability of seed in the habitats of these species, as P pilligaensis occupies heath and open forest, as does P novaehollandiae, while P. oralis occupies forest, P fumeus subalpine heathland and woodland, with P apodemoides found only in mature heath. Proc. Linn. Soc. N.S.W., 123. 2001 98 DIET OF THE PILLIGA MOUSE CONCLUSIONS Pseudomys pilligaensis appears to be similar to other Pseudomys studied (Cockburn, 1980, 1981la, 1981b; Thompson 1980; Luo and Fox 1994 and Fox ,Read, Jefferys and Luo, 1994) which broaden their diet in winter to incorporate a greater range of seed types and a higher proportion of other plant material (e.g. an increase in stem, leaf or sporangia). It is not clear whether this reflects an active choice by P. pilligaensis to select for increased amounts of leaf and a wider range of seed types or a default selection caused by lack of alternative dietary items in such a restricted habitat. These results indicate that P. pilligaensis is a granivore, as the diet is dominated by seed ( over 50%) both in spring/summer and winter. This is consistent with Kerley and Whitford’s (1994) definition “The term granivore,... .... should refer to animals whose diet is dominated (>50%) by[seed].” However P. pilligaensis could be operating in a general opportunistic manner (i.e. interpreting opportunistic species as one that is utilizing an unpredictable environment, differently over the seasons and years) considering the very restricted habitat that the species survives in today. These results are important as they provide the only information to date on the diet of P. pilligaensis in the wild. As the species is classed as rare, limited and vulnerable (Strahan 1995) this information may contribute to better understanding of its habitat requirements and conservation management. ACKNOWLEDGEMENTS We would like to thank Jenny Taylor and Marilyn Fox for constructive comments on a draft manuscript. We thank the Australian Museum for access to the two specimens from their collection and the National Parks and Wildlife Service for access to the specimen from their collection. Funding for this research was provided in part by grants from the Australian Research Council, most recently from (A1/9700994) REFERENCES Briscoe, D.A., Fox B.J. and Ingleby S. (1981) Genetic differentiation between Pseudomys pilligaensis and related Pseudomys . Australian Mammalogy 4, 89-92. Carron, P.L., Happold, D.C.D., and Bubela, T.M. (1990). Diet of two sympatric Australian subalpine rodents, Mastocomys fuscus and Rattus fuscipes. Australian Wildlife Research 17: 479-89. Cockburn, A. (1980). The diet of the New Holland Mouse (Pseudomys novaehollandiae) and the house mouse (Mus musculus) in a Victorian coastal heathland. Australian Mammalogy 3, 31-4. Cockburn, A. (1981a).Diet and habitat preference of the silky desert mouse, Pseudomys apodemoides (Rodentia) Australian Wildlife Research 8, 475-07. Cockburn, A. (1981b). Population regulation and dispersion of the smoky mouse, Pseudomys fumeus 1. Dietary determinants of microhabitat preference . Australian Journal of Ecology 6, 231-54. Fox, B.J. and Briscoe , D. A. (1980) Pseudomys pilligaensis: a new species of Murid rodent from the Pilliga Scrub, northern New South Wales. Australian Mammalogy 3, 109-126. Fox B.J., Read ,D.G., Jefferys E.A. ,and Luo, J. (1994) Diet of the Hastings River Mouse (Pseudomys oralis). Wildlife Research. 2,491-505. Golley, F.B. (1961) Energy values of Ecological Material . Ecology 42, 581-584. Hart D.M. (1995) Litter and decomposition in the Pilliga State Forests, New South Wales, Australia. Australian Journal of Ecology 20, 266-272. Johnson , M. K. ( 1982 ) Frequency sampling for microscopic analysis of botanical compositions. Journal of Range Management 35, 541-2. Kerley, G.I.H. and Whitford W.G. (1994) Desert-dwelling small Mammals as Granivores: Intercontinental Variations. Australian Journal Zoology 42, 543-555. Proc. Linn. Soc. N.S.W., 123. 2001 E.A. JEFFERYS AND B.J. FOX 99 Luo J. (1993). A study of the diets of Rattus lutreolus and Pseudomys gracilicaudatus (Rodentia:Muridae) and an investigation of methodology in mammalian dietary analyses . Ph.D. Thesis, The University of New South Wales. Luo, J. and Fox, B.J. (1994). Diet of the eastern chestnut mouse (Pseudomys gracilicaudatus): Il. Seasonal and successional patterns. Wildlife Research 21, 419-31. Luo. J. Fox. B.J. and Jefferys E.A. (1994) Diet of the Eastern Chestnut Mouse (Pseudomys gracilicaudatus) . 1. Composition, Diversity and Individual Variation Wildlife Research 21, 401-417. Norris, E.H., Mitchell, P.B. and D.M. Hart. (1991) Vegetation change in the Pilliga forests: a preliminary evaluation of the evidence. Vegetatio 91, 209-218. Redford, K.H. and Dorea J.G. (1984) The nutritional value of invertebrates with emphasis on ants and termites as food for mammals. J. Zool. Lond. 203, 385-395. Reichman, O.J. (1975) Relation of Desert Rodent Diets to Available Resources.Journal of Mammalogy 56, no.4 .731-751. Rolls, E. C. (1981) A million Wild Acres. Nelson. Melbourne. Sparks, D. R., and Malechek, J.C., (1968). Estimating percentage dry weight in diets using a microscope technique. Journal of Range Mangement 21, 264-78. Strahan, R. (ed.) (1995) The Mammals of Australia .Reed Sydney. Thompson, P. C. (1980) Population dynamics and feeding ecology of the New Holland mouse Pseudomys novaehollandiae (Waterhouse,1843) and the house mouse Mus musculus Linneaus, 1758. M.Sc. Macquarie University, Sydney. Proc. Linn. Soc. N.S.W., 123. 2001 e-. | RO Lah eke RAE Se | (sebdiniM: giinobust sheared pee vib angen Ueovind oft teed lacus seat noiingenat dh apathenih to ets (aviobd ignoeny eeBetb Ni oh debgiato ade mIRETIGN hsyfsedt ae 3 aerate ar rhs aah: § ren qpattes Gursives i: ida ly. Sere Mea aD ened © galiigeaenstr re ihe es. ate Clase ak phe We néined reais Pinta 3995) thy, sailormation may i back inseestanding OF iti TEC CIE AILS BPS CEE VEE sretiase eet ’ Pits Sn Ee aon Vio ils) he in ea Fara Re, Tie hee Acute lien: Pisiwssett FOr Ae ene Baits Sgrviie bor soe we PV Re 1S Ns Fri AL Try I ” Ra y a ‘ara d i, as i ; we, PL, bbs ie en ee oy ar n F tite ke Se FIT KY (ie | i My ny Oh eG i gat Ph eng real trae. al a the | Ada hip pone gd . a9 size fa ct | ‘ ¥ PA Pe Seas i ly Val - . “i « @ ata L ay ee ek Lae A cx ox Pate a ae ‘ d xsbec A me F +? , ties ey ar aterrninaalé VW ASSTIMADIOW mrsiery ; Crh fi artentingy ‘Exe ey @, 234 ' e ye b thea) i hd, Pir iat yWil ifs Nib? : i wi web tine TS. A CLIN Popa BW eee eOe VRE OT Mb Soilacing five Nera, porte Mew Soin Water. Aiton Mnvanaioey’ 2105-72 hia Al. eed 2) elite me LA, ada, 7 iad) Pay 1 ey 40 ver igh River MM und ret Waa coer: 2.4%) pine cee FALL) Bat alors of i caumpaaed Mateste! j ui “uv aa ft Y Sa y . > U Pig V7 ee } 1 + ol iv air. = cre sion Fi Migs’? hau bie tl iis eye ¥ att uth, Wale A sient, tei wade Ah faba) PR, Wate TIE 1, Me Ow ¢ HRD 1! Brebnc’ y OL al Pir Geb ecrgidc | mehily eA ot bai on coupe Reese Masieaowenl D6, 844-2 a sii i ; hotei AM ood Weiland MU} oet' Wad (Segaet -iSp altiau j aa j Mauinnath ihe Sjeaningieg ag tt os seetilind Jinmemyl Cahogy €Z, 24 149)) 100, FA 42.4 oe 901 5ont Land Surface Rehabilitation Research in Antarctica KEVIN KIERNAN AND ANNE MCCONNELL School of Geography and Environmental Studies, University of Tasmania, Tasmania 7005 Kiernan, K. and McConnell, A. (2001). Land surface rehabilitation research in Antarctica. Proceedings of the Linnean Society of New South Wales 123, 101-118. Ice-free ground surfaces in the Australian Antarctic Territory are sensitive to damage by artificial disturbance. Natural processes appear generally inadequate to heal the resulting scars over human time scales and substantial ongoing environmental impacts may accrue where melting of subsurface permafrost is triggered. Studies of some rehabilitation projects at sites where significant ground disturbance had been caused during geoscientific research indicate that although specific site conditions are critical to the approach taken, environmental harm can be reduced provided maximum advantage is taken of the opportunities to minimise and manage impact at each of the project design, environmental review, site selection, operational and rehabilitation phases. These sites provide a useful analogy for larger disturbances caused by infrastructure development. Manuscript received 1 August 2001, accepted for publication 21 November 2001. KEYWORDS: Antarctica, Vestfold Hills, environmental impact, geoenvironment, geoconservation, land rehabilitation. INTRODUCTION Ice-free land surfaces in the Australian Antarctic Territory can be particularly susceptible to environmental damage (Schofield 1972, Parker and Howard 1977). Anthropogenic disturbance resulting from some research and engineering activities can produce scars that take a very long time to heal (Campbell and Claridge 1987). Where a permafrost layer becomes exposed, meltwater discharge, severe channelisation, erosion and slope instability may result (French 1976, Burgess et al. 1992). Present expertise and experience in Antarctic land surface rehabilitation is very limited and the lack of significant vegetation deprives land managers of an important rehabilitation tool generally available elsewhere (McVee 1973). As a party to the Protocol on Environmental Protection to the Antarctic Treaty or “Madrid Protocol” (Antarctic Treaty Consultative Parties 1992), Australia has legal obligations to protect the Antarctic environment. There is also a public expectation that environmental protection in Antarctica be very strict, although this has not always been the reality and a legacy of scars already exists in some areas. Apart from infrastructure development, the most obvious potential for scarring of the landscape arises from geoscientific research that involves excavation to allow investigation and sampling of subsurface materials. The perceived potential for environmental harm contributes to occasional advocacy that geoscientific research in Antarctica should be restricted (Graham 1997). Proc. Linn. Soc. N.S.W., 123. 2001 102 ANTARCTIC REHABILITATION Figure |. Localities mentioned in text. The numbers indicate localities referred to in Table 2. 68°30’S Proc. Linn. Soc. N.S.W., 123. 2001 a Y = > = & De, = Sy © W > 7) © = K. KIERNAN AND A. McCONNELL 103 Recent research in the Vestfold Hills has revealed that geoscientific research accounts for nearly half the recorded sites of persistent human impact on the physical environment outside Davis Station limits (Fig. 1). Comparisons between rehabilitated sites and others where little if any rehabilitation had been attempted revealed that natural processes alone are generally insufficient to heal the damage. This paper details the specific environmental protection and remediation strategies that facilitated the reduction in environmental harm observed at the rehabilitated sites reported upon in that study (Kiernan and McConnell 2001). We describe and evaluate planning and site management designed to reduce environmental impacts, and surface rehabilitation methods employed. We compare site conditions immediately subsequent to the rehabilitation work with the results of site monitoring undertaken just under four years later. Physical impact of geoscientific excavations Ice-free Antarctic land surfaces are characterised by a lack of significant vegetation, a relative scarcity of water, a very slow rate of soil material movement, a ground surface that comprises unconsolidated material that is commonly overlain by a desert pavement of lag gravels and coarse sands, and the presence of subsurface permafrost (Campbell and Claridge 1987). Excavation inevitably involves disturbance of the natural stratigraphy, especially if waste material is replaced in the pit. Unnatural change to surface contours, at whatever scale, by definition damages the natural geomorphology. Natural soil properties may be compromised through soils adjacent to an excavation becoming contaminated by spoil, especially if it is redistributed by the wind (Campbell and Claridge 1987). Some research requires that the pristine condition of the Antarctic environment is maintained, for example understanding the accumulation of soil nitrate from atmospheric circulation, but local soil contamination has been demonstrated (Campbell and Claridge 1987, Claridge et al. 1995). Excavation can also interfere with ongoing natural processes, as when removal of a desert pavement lag exposes underlying finer material to wind erosion. Any failure to adequately reconsolidate spoil replaced in a pit may result in progressive settling that ultimately creates a depression. Alternatively, a difference in bulk density relative to the undisturbed material surrounding the pit may leave fill prone to differential permafrost development. Impacts on aesthetic values may include changes to form, line, colour and texture of a landscape (United States Forest Service 1974, Laurie 1975) and such changes may be particularly evident in open Antarctic environments where masking vegetation is absent. Some researchers may consider backfilling of research excavations to achieve little because the disturbance is permanent and difficult to conceal (Campbell and Claridge 1987). The permanent impact of excavation upon the natural stratigraphy is impossible to undo in any practical or useful way even if harm might be reduced were material from different strata stockpiled separately and replaced in the order in which it was originally encountered (Campbell et al. 1993). However, other potential adverse effects including long term changes to natural processes may be minimised if appropriate techniques are developed. Some geomorphological and visual impacts may be reduced by efforts to reproduce the original shape, texture and form of the ground. Contamination of soils adjacent to pits can be minimised by stockpiling spoil on cover sheets or placing it in bags so it is not left exposed to wind. Campbell et al. (1993) have shown that some shallow excavations and vehicular tracks can persist for more than 30 years, but that some types of impact can recover more quickly where there are repeated freeze-thaw cycles, and to a lesser extent wind action. Improved understanding of the degree of repair that is desirable and achievable is important in order to reduce environmental harm caused by research. Research excavations can also provide a small scale analogy for the larger disturbances caused by the provision of infrastructure for science, tourism or other purposes, and the information obtained may help inform planning and remediation of larger projects.. Proc. Linn. Soc. N.S.W., 123. 2001 104 ANTARCTIC REHABILITATION Table 1. Assessment criteria for rapid visual evaluation of terrestrial environmental impacts, and scoring system. Items A-K are from Campbell et al. (1993); items L-P are from Kiernan and McConnell (2001). Impact assessment criteria A. O. Proc. Disturbed surface stones Stone impressions Boot imprints Visibly disturbed area Colour difference (Munsell units) Other surface impressions (eg. equipment) Walking routes Foreign objects Fuel spills Biological disturbance Cumulative impact (scale 1-10) Stratigraphic disturbance . Morphological or texture change Rock cairns Paint marks Other marks (eg. stakes) Linn. Soc. N.S.W., 123. 2001 1 none visible 0 none visible none visible <5 m? none visible none visible not visible none visible 0 none visible none visible disturbance not visible negligible negligible none none none Severity and extent of impacts (class) 2 few <10 just visible just visible 5-10 m? weak contrast weakly visible weakly defined few <10 faintly distinguished <1 mm weakly distinguished within on unit just evident rare or small rare or small rare or small 3 many 10-25 distinct distinct 20-100 m? moderate contrast distinct moderately defined some 10-25 visible 1-5 m? clearly visible disturbance within two units moderate very change moderately common or large moderately obvious moderately obvious 4 abundant >25 fresh fresh >100 m? strong contrast very fresh strongly defined many >25 very obviou >5 m disturbed & very obvious multiple units obvious very common & obvious very obvious very obvious K. KIERNAN AND A. McCONNELL 105 GENERAL METHODOLOGY In 1996 we became involved in a geoscientific research program in the Vestfold Hills that included excavation of a number of small soil investigation pits on moraines and larger excavations at Marine Plain and Heidemann Bay (Fig. 1). The purpose of the Marine Plain excavations was to investigate sediments and ground ice conditions, and to evaluate evidence for reported soil development. The Heidemann Bay excavation involved re-evaluation of evidence for a claimed phase of climatic warming (Hirvas et al. 1993) that, if correct, had major implications for understanding the evolution of Antarctic environments and global climate change. The excavations were undertaken only after prior environmental evaluations and approval by the Australian Antarctic Division, and were planned to minimise adverse impacts consistent with achieving the scientific goals. Concerted efforts were made to achieve a high level of ground surface rehabilitation. The system developed by Campbell et al. (1993) was employed to record rapid visual estimates of the remaining impacts, supplemented by some additional criteria (Kiernan and McConnell 2001) including the use of Standard Rock Colour Charts (Goddard et al. 1948) (Tables 1 and 2). Photo-monitoring was initiated and record made of any physical evidence suggestive of artificially accelerated melting of permafrost. This assessment process was repeated just under four years later. The condition of some sites not effectively rehabilitated by previous researchers prior to advent of the Madrid Protocol was also recorded on both occasions to permit assessment of the relative efficacy of deliberate rehabilitation compared to natural processes that might effect repair (Table 2). REHABILITATION TECHNIQUES AND RESULTS Marine Plain Marine Plain is floored by a veneer of glacial sediment over Pliocene marine diatomite. The rocks are subject to intense salt weathering in this very arid environment. Permafrost occurs below ~1 m depth and the local landforms have evolved due to very slow progressive melting of ground ice (Kiernan et al. 1999). Terrain produced by this process is known as periglacial thermokarst because the resulting depressions give the topography an appearance similar to that of conventional limestone karst. Significant geohazards associated with thermokarst are well documented from the northern hemisphere and include the risk of accelerated subsidence, slumping and the discharge of meltwater to the surface where disturbance of the seasonally-thawed “active layer” so modifies the thermal condition of deeper permafrost as to accelerate its melting (French 1976). Environmental issues at Marine Plain also include the sensitivity of the ground surface to trampling. A thin crust, possibly gypsum, occurs over the loose, powdery surface horizons of the diatomite, and this crust 1s crushed by foot-fall, in the same manner as very thin flowstone in a limestone cave. This releases a plume of diatomite dust which in windy conditions facilitates soil contamination risks of the kind alluded to by Campbell and Claridge (1982) and also leaves a persistent sharply-defined, colour-contrasting footprint. Some observed footprints appeared to have persisted since at least the previous summer and hence were at least a year old. Un-rehabilitated pits monitored in January 1997 and December 2000 showed little evidence of healing by natural processes (Table 2, site 1) (Kiernan and McConnell 2001). Proc. Linn. Soc. N.S.W., 123. 2001 106 ANTARCTIC REHABILITATION Table 2. Impact assessment (in 1997 & 2000) of some excavation sites, including some data from Kiernan and mcConnell (in press). See Table 1 for criteria and scoring system. No data = nd. assessment criteria A B Cc D E F G Jab. oll J K L MONE One site Marine Plain la 4 4 4 3 nd 4 3 3 1 1 4 2 4 1 1 4 1b 4 4 2 3 nd 4 yy 3 1 1 3 y) 4 1 1 4 2a 4 ” 1 4 4 2 1 1 1 4 3 4 1 1 1 2b 4 4 1 3 nd 4 2 2 1 1 4 3 4 1 1 1 2c 4 2 1 3 4 2; 1 1 1 4 3 4 1 1 1 3a 4 2 D 3 3 2 3 1 1 1 3 3 2 1 1 1 3b 1 1 | 1 1 1 1 1 1 1 3 3 2 1 1 1 4a 3 2; 2 3 nd 2 yy 1 1 1 2 3 1 1 1 1 4b 1 1 1 1 nd 1 1 1 1 1 1 3 1 1 1 1 Heidenmann Valley 5a 3 3 Ak 2 nd 4 1 1 1 1 4 2) 4 1 1 1 5b 3 3 1 D) nd 4 1 1 1 1 4 D)} 4 1 1 1 6a 3 3 3 2 nd 4 1 1 1 1 4 yy 4 1 1 1 6b 3 3 3 2 nd 4 1 1 1 1 4 D 4 1 1 1 Ta 3 1 1 3 1 D} 2 1 1 1 1 1 1 1 1 1 7b 1 1 1 1 i 1 1 1 1 1 1 1 1 1 1 1 8a yD 1 D 1 1 1 2 1 1 1 1 2 1 1 1 1 8b 1 1 1 1 1 1 1 1 1 1 1 py) 1 1 1 1 9a 4 Dy 1 4 le 2 1 1 1 2) 4 1 1 1 1 9b 1 1 1 1 1 1 1 1 1 1 1 4 1 1 1 1 Sites: Marine Plain: 1= small un-rehabilitated pits believed to have been excavated in 1995-96 (a) condition in January 1997, (b) in December 2000. 2= Big Ditch site (a) un-rehabilitated condition prior to re-distrubance in 1997, (b) immediately after reha- bilitation attempt in January 1997, (c) in December 2000. 3= 1997 scarp trench (a) immediately after rehabilitation in March 1997, (b) in December 2000. 4= 1997 pit above main trench (a) immediately after rehabilitation in January 1997, (b) in December 2000. Heidemann Valley: 5= old un-rehabilitated soil pit on Stinear Moraine (a) in March 1997, (b) in December 2000. 6= old un-rehabilitated soil pit on moraine on south side of Heidemann Bay (a) in March 1997, (b) in December 2000. 7= Section of previously un-rehabilitated bull-dozed track re-used in 1997 (a) immediately after rehabilita- tion attempt in March 1997, (b) in December 2000. 8= Cross-country route between old bulldozed track and trench site (a) immediately after rehabilitation in March 1997, (b) in December 2000. 9= 1997 trench (a) immediately after rehabilitation in March 1997, (b) in December 2000. Proc. Linn. Soc. N.S.W., 123. 2001 K. KIERNAN AND A. McCONNELL 107 Big Ditch (Table 2, site 2) Big Ditch is a natural elongate depression across northern Marine Plain. A substantial excavation had already been undertaken on the depression margin by earlier researchers, hence re-excavation here avoided creation of a new disturbance, and also provided an opportunity to experiment with repair of a site where a significant impact had already been caused with no effective rehabilitation. The earlier disturbance had involved a cut ~5 m wide, 2 m high and penetrating ~5 m into the side of a natural scarp. Initial observations in January 1997 revealed a bulge in the foot of the old excavated face, and a deposit of fine sediment and surface salt that extended downslope, suggesting accelerated permafrost melting had resulted from the original disturbance. The 1997 research required cleaning-back the previously-excavated face by ~10 cm to reveal a vertical section 50 cm wide and 135 cm high. Excavation, sampling and rehabilitation were completed in one day. It proved impracticable to do more than seek to ensure our re-disturbance was no more visually evident than was the much larger artificial face and hole in which it was sited. The original large excavation could not be filled and the overall slope restored because the spoil from the earlier excavation had spilled downslope and been redistributed by wind and meltwater such that little could be retrieved without risking significant additional disturbance. An attempt was made to reduce visual impact by using the very small volume of spoil that could be retrieved to mimic the natural slopes to either side. Efforts were made to reproduce the slope angle, colour and texture of the face of the earlier excavation, itself very conspicuously artificial. Old discarded scraps of latex and hessian were also removed. Some larger rocks retrieved from the earlier spoil were distributed along the top of the face to mimic the adjacent pattern of glacial boulders spilling onto the upper scarp. Immediately after rehabilitation on 24 January 1997 the new disturbance was difficult to distinguish from the earlier excavation (Fig. 2a). The newly-disturbed surface was a mixture of unweathered sediments from various depths. The colour contrast between the newly disturbed and previously disturbed surfaces was moderate at two Munsell units, but this did not reflect the real visual impact of the original excavation on the broader landscape. Because the depression in the active layer caused by the earlier excavation was not filled, the potential remained for continued permafrost melting. When the site was monitored on 11 December 2000 some further movement of the lower part of the face by solifluction was evident. Discolouration of downslope areas by salt and fine sediment was more pronounced than in 1997 (Fig. 2b). These observations were interpreted as indicating that thawing of the permafrost had been accelerated by renewed disturbance of the active layer. Hence, the structural and cosmetic rehabilitation attempted had not been sufficient to redress the effects of the re-disturbance superimposed upon the original un-rehabilitated excavation. Scarp trench (site 3) A trench 22 m long and 30-40 cm wide with a series of benches along its length was excavated down the face of a scarp 9.5 m high (Figs. 3a and 3b). The area disturbed was minimised by restricting pedestrian access to the line of the trench. Placing the large volume of spoil on a ground sheet to reduce the spread of lighter coloured material was not practical on the steep slope, and additional degradation was judged likely if spoil was carried from the top to the bottom and then vice versa. For these reasons, spoil was placed to one side so that sediments of different kinds would not become mixed in a pile at the foot of the slope, and so the broad composition and colour of the material excavated could be replaced as closely as possible to its original position. Small calibre scree was stockpiled separately and larger rocks were also segregated. Excavation, sampling and rehabilitation were accomplished within two days. Proc. Linn. Soc. N.S.W., 123. 2001 108 ANTARCTIC REHABILITATION Figure 2. Big Ditch site (site 2): (a) in January 1997, immediately after disturbance and rehabilitation of the old excavation site; and (b) in December 2000, showing fine sediment and salt deposited by meltwater still being released from the permafrost. ASE ie 2 Proc. K. KIERNAN AND A. McCONNELL 109 An attempt was made to minimise alteration of the natural geomorphology by seeking to (1) minimise changes to the insulating properties of the disturbed active layer that might otherwise trigger permafrost melting and slope instability; (2) mimic an old lake shoreline on the slope, and (3) mimic the natural distribution of scree. Only material that had been removed from the excavation was used to refill the trench. The natural morphology and surface rock distribution were reproduced as closely as possible with respect to (1) swathes of fines that extended down the slope, (2) a distinct break of slope above a clay zone; and (3) a steep slope facet in rocky upper areas of the scarp. An attempt was also made to restore the aesthetic values by identifying and focussing upon what were judged to be the dominant site-specific visual elements, namely (1) the natural profiles of the lower slope, both longitudinal and lateral; (2) the bouldery texture of the top of the slope; (3) the surface colour pattern, involving lighter-coloured sediment upslope and a distinct darker band downslope; (4) the presence of cobbles lower down the slope; and (5) the presence of gravelly scree that commenced at half height on the slope. During rehabilitation each bench within the trench was used as a foundation upon which returned spoil could be placed, to ensure that any later settling would occur in small localised units rather than being sequentially transmitted down the full length of the trench. Larger rocks were placed towards the outer edge of each bench to retain fine material and the fill was consolidated on a bench-by-bench basis and manually compacted. Restoration was undertaken from the top downwards rather than from the bottom upwards, so that each bench had to be made stable in its own right. The retained scree was utilised in an attempt to mimic the natural veneer of lag gravels in an effort to obtain as natural a surface texture and appearance as possible. Replacement of the original surface clasts weathered side up helped further reduce the visual impact. As the rehabilitation proceeded, periodic checks were made on its appearance from a distance so that adjustments could be made. At the conclusion of this work on 24 January 1997 a satisfactory mimic of the natural form and texture of the slope had been achieved. However, marked differences between the colours of the surface sediment and exposed subsurface sediments precluded immediate satisfactory visual restoration, the colour contrast between the weathered natural surface and the disturbed subsurface diatomite being very high (up to four Munsell value units) (Fig. 3c). Differences in colour between surface and subsurface glacial clasts contributed to the visual contrast after rehabilitation. However, while this site remained visually evident when viewed from close up, it was less conspicuous from a distance of more than ~200 m because the colour was reasonably similar to that of natural swales that occur locally down the face of the scarp. When monitored on 11 December 2000 the trench site was superficially indistinguishable from its appearance prior to excavation (Fig. 3d). There was no evidence of any slumping or downslope flow, suggesting that construction of the subsurface rehabilitation was adequate. Nor was there any evidence of lateral subsidence that might be attributable to melting of ground ice, nor of any discharge of fine material or salts onto the surface. This suggests the thickness and thermal characteristics of the deliberately re-consolidated infill provided a reasonable mimic of the original active layer. The small volume of fine subsurface dust that discoloured the ground surface immediately after initial rehabilitation was no longer evident, presumably having weathered or been removed by the wind, the latter implying diffuse contamination downwind. The scatter of small calibre gravel applied to the surface during rehabilitation could not be differentiated from the natural material to either side of the disturbed area. An equally satisfactory result was obtained after rehabilitation of an associated soil pit at the crest of the scarp (Table 2, site 4). Proc. Linn. Soc. N.S.W., 123. 2001 110 ANTARCTIC REHABILITATION Figure 3. Marine Plain scarp trench (site 3); (a) prior to excavation; (b) during excavation, with structural rehabilitation of top section initiated. Figure 3a. Figure 3b. Proc. Linn. Soc. N.S.W., 123. 2001 K. KIERNAN AND A. McCONNELL 111 Figure 3 continued. Marine Plain scarp trench (site 3); (c) immediately after rehabilitation immediately after rehabilitation in January 1997; and (d) and in December 2000. Proc. Linn. Soc. N.S.W., 123. 2001 112 ANTARCTIC REHABILITATION Heidemann Bay Heidemann Bay penetrates ~2 km into the lower part of Heidemann Valley (Fig. 1). The undulating valley floor is generally below 10 m altitude and is mantled by glacial sediment including many large surface boulders. Old un-rehabilitated soil pits at two locations showed negligible natural recovery between March 1997 and December 2000 (Table 2, sites 5 and 6). The research undertaken in 1997 involved re-use by a heavy excavator and other vehicles of an old, previously un-rehabilitated lightly bulldozed track (Table 2, site 7) and beyond this a short cross-country route (site 8), to facilitate excavation of a large trench (site 9). The original proposal involved excavation right across the head of the Bay to allow assessment of the extent and continuity of the key deposits. However, this trench would have been ~500 m long, 4-6 m deep and 3-4 m wide, entailing a significant environmental impact. Prior to our departure for Antarctica a more environmentally appropriate compromise was reached involving a much shorter trench of ~50 m length, with provision for four smaller pits if needed. This was approved by the Australian Antarctic Division and relevant Minister following an environmental assessment that included public input and attempts to elicit responses from environmental groups. A decision was subsequently made in the field to reduce the size of the excavation even further to a single trench 20 m long. This still permitted the original scientific objectives to be achieved. The precise site selected was on a Stretch of raised gravelly beach superimposed upon the glacial sediments. This was chosen because it would not be necessary to remove large surface boulders that would be difficult to replace without their disturbance being evident and even greater impacts being generated in shifting them. This site was also selected because the adjacent areas already bore visible artificial impacts, primarily vehicle tracks across the valley floor, and a large stone quarry 200m east of the beach on the valley edge. The trench was dug using an excavator with a5.5 m arm. Snow cover reduced ground impact on the access route, which was also selected to avoid large boulders. To minimise ground disturbance while the digging proceeded, the uppermost material removed was used to construct a pad on which the excavator was positioned. The final trench measured 20 m long, ~2-3 m wide and 4.0-4.5 m deep. The width was kept to a minimum and reflects essentially the width of the bucket (plus about 0.5 m). The excavator was left at the site for the duration of the work to minimise the number of passes along the access route, to create additional shelter from the wind, and to form part of a barrier that was constructed to prevent wildlife falling into the trench while it was unattended. About 300 m? of sediment was removed and stored on the upwind side of the hole to maximise trapping back into the pit of any sediment blown from the heap. Even when great care is taken during excavations, fines from dry soils can be distributed by the wind and a large area become contaminated (Campbell and Claridge 1987) but the melting of recent light snow and the damp environment on the coast reduced this hazard considerably. Once the permafrost table was reached the melting of ground ice helped further dampen and stabilise the spoil, and there was little wind at the time of the excavation. Examination, recording and sampling of the sediments was accomplished in 1.5 days. The operation was a race against melting of the permafrost and consequent loss of trench wall stability, and also against the weather, a major factor in planning and executing any proposed activity in Antarctica. Minor slumping due to permafrost melting was evident by the evening on which documentation and sampling was completed, and had worsened slightly by the next morning, with the risk of environmental damage increasing the longer the trench was kept open. Because rehabilitation would also have been made much more difficult had the trench become filled by snow it was re-filled immediately, despite a gathering blizzard with rising Proc. Linn. Soc. N.S.W., 123. 2001 K. KIERNAN AND A. McCONNELL 113 Figure 4. Heidemann Bay trench (site 9) (a) during rehabilitation under blizzard conditions; (b) linear depression formed in rehabilitated surface between March 1997 and December 2000. POE Figure 4a. Proc. Linn. Soc. N.S.W., 123. 2001 114 ANTARCTIC REHABILITATION winds of 40-45 knots and blowing snow (Fig. 4a). This might have posed the risk of sediment being blown considerable distances had the moist condition of the spoil not by now become even more pronounced due to continued permafrost melting. The position of the trench and direction of the wind meant that any dust generated would be carried into the sea rather than contaminate soil surfaces, but after re-filling there was no evidence of dust on the snow any further than 5 m downwind. The excavator was used to compact the fill and to rake the surface. The sediment fitted back in the hole with negligible surface mound remaining, a result attributed to some mass loss having occurred through melting of ground ice. No work was undertaken the next day as the blizzard continued. The following day was devoted entirely to rehabilitation, initially involving work with the excavator, which was then driven from the site, again with the ground impact reduced by a layer of new snow. Manual rehabilitation of the ground surface using spades, mattocks and rakes initially involved digging snow out of ruts made by the excavator so that these could be smoothed. This snow was later scattered across the rehabilitated area to maximise the available soil moisture and enhance the potential for frost heaving. This work was made much easier by the fact that insufficient time had elapsed to allow the disturbed ground to compact and harden. Care was taken to avoid smoothing the ground to such an extent that the rehabilitated surface appeared flatter and smoother than the natural surface surrounding it. Some retained glacial rocks were scattered back across the site with their weathered upper surface uppermost. Efforts were made to smooth the transition between disturbed and undisturbed ground to soften the contrast in colour and texture. These same procedures were employed along the full length of the access route (sites 7 and 8). The impact of two passes by the excavator proved visually less than that caused by several traverses by a four wheel drive vehicle used to carry other equipment and heavy materials to the site. The hand-tool phase of the rehabilitation process took ~2.5 person days. At the conclusion of the rehabilitation work on 8 March 1997 an acceptable mimic of the form and texture of the original surface had been achieved. The colour contrast between disturbed and undisturbed sediment matrix was low to moderate at 1-2 Munsell value units. Rock colour comparisons also revealed only moderate contrast at two Munsell value units. The remaining visual impact varied according to the distance from which the site was viewed. The colour contrast proved relatively inevident from close up where there was no immediately juxtaposed undisturbed and disturbed ground, a minor contrast was more evident from a slightly greater distance, but from beyond ~100 m distance it was difficult to discern. We were encouraged that when we photographed the rehabilitation from the air a week later a helicopter pilot who had previously overflown the trench repeatedly at low altitude was unable to identify where it had been. When monitored on 14 December 2000 the trench site could not be located visually on the ground without using photographs. Once the target area had been located accurately a slightly darker colouration and slight relative scarcity of surface rocks was discerned where spoil had been stockpiled. A subtle linear depression a few centimetres across, less than 1 m long and up to 5 cm deep was found to have formed along the line of the trench and to have been partly filled by in-washed fine sediment (Fig. 4b). Though visually similar to nearby natural features, this depression may reflect either settling of the spoil or frost cracking, also conceivably triggered by the disturbance. Neither the trench site nor access route were readily discernible upon overflying the site at ~50 m altitude. That part of the older vehicular track that was re-used and rehabilitated in 1997 (site 7) was just discernible at ground level, but the precise location of the rehabilitated cross-country access route (site 8), traversed by the excavator, 4WD vehicles and pedestrians, could no longer be discerned (Figs 5a and 5b). The un- rehabilitated parts of the older bull-dozed track remained conspicuous. Proc. Linn. Soc. N.S.W., 123. 2001 K. KIERNAN AND A. McCONNELL 115 Figure 5. Heidemann Bay cross-country access track (site 8); (a) before rehabilitation in March 1997; (b) in December 2000. The trench site (site 9) is immediately behind and to the right of the researcher. Figure Sa. Figure 5b. Proc. Linn. Soc. N.S.W., 123. 2001 116 ANTARCTIC REHABILITATION DISCUSSION These results demonstrate that long term impacts of ground disturbance can be minimised if maximum advantage is taken of the opportunities to review and modify procedures at each of the project design, environmental review, site selection, operational and rehabilitation phases. Careful planning, siting, management and rehabilitation of any excavation are important. Confining traffic to defined access routes, and where possible to snow covered ground rather than sediment surfaces, reduces potential damage to the ground. At Heidemann Bay the construction of a pad from which the excavator could work considerably reduced potential surface disturbance. At Marine Plain, minimising trampling of the soft diatomite in the first place proved much more effective than any rehabilitation afterwards. Care is required in stockpiling spoil, useful approaches include placing it on a sheet to prevent contamination of adjacent surfaces, or in a bag to prevent it being redistributed by the wind. The shape of the disturbance is also important. At Heidemann Bay the linear disturbance of the old bulldozed track was less easily rehabilitated than was the more irregularly-shaped disturbance at the trench site, even though the magnitude of the works at the latter was vastly greater. Our evidence suggests that a better result is possible with immediate diligent rehabilitation than if rehabilitation occurs after a significant delay. Even the site of our main trench in the thermokarst at Marine Plain (site 3) appeared to have achieved a greater degree of stability, and was visually less evident, than some disturbances that were decades older but which had been subject to minimal or no rehabilitation. We conclude that it is critical that rehabilitation is properly designed and programmed, and is executed immediately following disturbance. Rehabilitation must be viewed as an integral part of a project, not just an obligatory and token exercise achieved by merely shovelling a few spadefuls of spoil back into a hole. Rehabilitation can be aided by the availability of photographs of the site taken prior to disturbance. The likely success of different rehabilitation strategies is highly dependent on site conditions. The long term impact of any artificially accelerated melting of permafrost appears strongly influenced by the degree to which the ice is segregated within the sediments and the nature of those sediments. Only limited segregation of ground ice was evident at Marine Plain but the acceleration of its melting by artificial disturbance of the active layer at the Big Ditch site (site 2) was nevertheless sufficient to trigger accelerated thermokarstic processes in the silty diatomite. In contrast, while a comparable degree of ice segregation was revealed in the Heidemann Bay trench we found no evidence of thermokarst processes affecting the poorly sorted glacial sediments. Most of the colour contrast left after excavation derives from unweathered matrix material brought to the surface. Whereas raking proved a very satisfactory way to diffuse edge contrasts at Heidemann Bay where the colour contrast between surface and unearthed subsurface material was moderate, the strong colour contrast between the surface and subsurface material at Marine Plain meant that employment of the same technique there would have served only to broaden the area of visually obvious disturbance. Rock colours vary considerably due to differences in rock type, but subsurface rocks are commonly unweathered and coated by unweathered matrix material, hence the colours of mineral constituents are often not as dramaticlaly brought out as with surface rocks. Regular and effective monitoring to allow assessment of the impacts of ongoing activities, the verification of predicted impacts and the early detection of unforseen effects are also important and are required under the Madrid Protocol (Lyons 1993). Monitoring also enables rehabilitation methods to be assessed and improved for the future. Respondents to the Heidemann Bay public consultation process stressed the need for long term monitoring. However, there must be an appropriate mechanism to ensure monitoring can occur. A formal monitoring process for disturbed sites is required, rather than opportunistic, ad hoc monitoring. In establishing a formal monitoring processes and setting rehabilitation Proc. Linn. Soc. N.S.W., 123. 2001 K. KIERNAN AND A. McCONNELL 117 and monitoring requirements, the relative roles of the proponent of the disturbance, and the management agency and its various staff, needs to be realistically appraised and taken into account. For effective monitoring to occur there must be both a commitment and resourcing. Monitoring must also be properly designed and programmed. It is generally beyond the capacity of individual scientists who may never be able to return to Antarctica. We were fortunate in being able to undertake monitoring ourselves in December 2000 while transiting through Davis, but the circumstances that permitted this were uncommon. While our proposal for monitoring of the 1997 excavation was endorsed by the agency responsible for management of the area, up until December 2000 no monitoring had occurred. Presumably the agency either did not have the resources to conduct the monitoring or had not considered it a priority, even though the Heidemann Bay trench was probably the largest research excavation ever undertaken in East Antarctica. Monitoring should employ objective criteria such as soil colour charts which provide an effective means of measuring visual impact, and careful photo-monitoring. Monitoring techniques must not themselves create an additional environmental risk or impact. Documentation of the rehabilitation and monitoring results is also critical for ongoing monitoring and assessment. An adequate database on disturbances and rehabilitation measures is essential. Our capacity to evaluate recovery of sites disturbed prior to 1997 was seriously impeded by the lack of reliable information concerning the age, location, dimensions and any rehabilitation history of earlier excavations. CONCLUSIONS The case studies presented here demonstrate the value of careful, planned, expeditious and monitored rehabilitation of disturbed sites in Antarctica, a highly sensitive environment that is deserving of the most stringent environmental protection and care. The relatively small disturbances reported here also provide a potentially useful analogy for larger scale activity associated with the provision of infrastructure in Antarctic environments, such as the significant ground impacts associated with Australia’s Antarctic bases. A legacy of past disturbance exists in Antarctica, not all present-day activities are environmentally benign, and pressures on the Antarctic environment may well increase in future. An unsuccessful 1988 proposal to establish a centre in the Vestfold Hills to cater for up to 16,000 tourists a year hints at future possibilities (Martin 1996:257). Developing a higher level of expertise and experience in Antarctic land surface rehabilitation is warranted. ACKNOWLEDGMENTS The monitoring was made possible by an Antarctic Science Grant and other assistance from the Australian Antarctic Division. We very gratefully acknowledge the contribution of Sel Peacock whose skill in operating the excavator and attention to detail during the original work at Heidemann Bay was outstanding. Peter Corcoran, Pene Greet, Melissa Giese, Steve Richards, Sarah Mills and Noel Ward provided assistance in the field. Eric Colhoun, Mick Brown and Bruce Chetwynd commented on earlier drafts. REFERENCES Antarctic Treaty Consultative Parties (1992). Protocol on Environmental Protection to the Antarctic Treaty - Final Report of the Eleventh Antarctic Treaty Special Consultative Meeting, Madrid, 22-30 April 1991; 17-22 June 1991; 3-4 October 1991. (Ministerio de Asuntos Exteriores: Madrid). Proc. Linn. Soc. N.S.W., 123. 2001 118 ANTARCTIC REHABILITATION Burgess, J.S., Spate, A.P. & Norman, F.I. (1992). Environmental impacts of station development in the Larsemann Hills, Princess Elizabeth Land, Antarctica. Journal of Environmental Management 36, 287-299. Campbell, I.B., Balks, M.R. and Claridge, G.C.C. (1993). A simple visual technique for estimating the impact of fieldwork on the terrestrial environment in ice-free areas of Antarctica. Polar Record 29 (171) 321-328. Campbell, I.B. and Claridge, G.G.C. (1982). The influence of moisture on the development of soils of the cold deserts of Antarctica. Geoderma 28, 221-238. Campbell, I.B. and Claridge, G.G.C. (1987). ‘Antarctica, Soils, Weathering Processes and Environment’. (Elsevier: Amsterdam). Claridge, G.G.C., Campbell, I.B., Powell, H.K.J., Amin, Z.H. and Balks, M.R. (1995). Heavy metal contamination in some soils of the McMurdo Sound region, Antarctica. Antarctic Science 7(1), 9- 14. French, H.M. 1976. ‘The Periglacial Environment’. (Longman: London). Goddard, E.N., Trask, P.D., De Ford, R.F., Rove, O.N., Singlewald, J.T.Jr. and Overbeck, R.M. (1948). Rock Colour Chart. Geological Society of America: Boulder, Colorado). Graham, A. (1997). Antarctic policy review. The Tasmanian Conservationist 256, 22-23. Hirvas, H., Nenonen, K and Quilty, P. (1993). Till stratigraphy and glacial history of the Vestfold Hills area, East Antarctica. Quaternary International 18, 81-95. Kiernan, K and McConnell, A. (2001). Impacts of geoscience research on the physical environment of the Vestfold Hills, Antarctica. Australian Journal of Earth Sciences 48, 767-776. Kiernan, K., McConnell, A. & Colhoun, E. (1999). Thermokarst landforms and processes at Marine Plain, Princess Elizabeth Land, East Antarctica. INQUA XV International Congress, 3-11 August 1999, Durban, South Africa. Book of Abstracts, 98. Laurie, M. (1975). “An Introduction to Landscape Architecture’. (Elsevier: New York). Lyons, D. (1993). Environmental impact assessment in Antarctica under the Protocol on Environmental Protection. Polar Record 29(169), 111-120. McVee, C.V. (1973). Permafrost considerations in land use planning. In ‘Permafrost. North American Contribution to Second International Conference’ pp. 146-151. (National Academy of Sciences, Washington DC). Martin, S. (1996). ‘A History of Antarctica’. (State Library of NSW Press: Sydney). Parker, B.C. and Howard, R.V. (1977). The first environmental impact monitoring and assessment in Antarctica. The Dry Valley Drilling Project. Biological Conservation 12, 163-177. Schofield, E. (1972). Presenting the scientific value of cold desert ecosystems: Past and present practices and a rationale for the future. In Conservation problems in Antarctica (Ed. B.C. Parker) pp. 193-227 (Virginia Polytechnical Institute, Blacksburg, Virginia). United States Forest Service (1974). National Forest Landscape Management Volume 1, Agriculture Handbook 434 (United States Forest Service: Washington DC). Proc. Linn. Soc. N.S.W., 123. 2001 Tertiary Echinoids from Papua New Guinea I.D. LINDLEY Department of Geology, Australian National University, Canberra, A.C.T. 0200 Lindley, I.D. (2001). Tertiary echinoids from Papua New Guinea. Proceedings of the Linnean Society of New South Wales 123, 119-139. Tertiary echinoid faunas from four different environments are described from three localities in Papua New Guinea (PNG). The fauna from the Middle Miocene Langimar Beds at Aseki village, Morobe Province, is dominated by near-surface, sand dwelling clypeasteroid echinoids Echinodiscus bisperforatus Leske, 1778 and Laganum depressum Lesson in L. Agassiz, 1841. The fauna from the Upper Oligocene-Lower Miocene Padowa Beds in the Sagarai valley, Milne Bay Province, is dominated by sand and sandy-mud burrowing spatangoid echinoids Brisaster latifrons (A. Agassiz, 1898) and Brissopsis ?luzonica (Gray, 1851). The rich echinoid fauna from the Lower Pliocene Kairuku Formation on Yule Island, Central Province, includes sea-grass meadow dwelling, and highly turbulent, shallow-water dwelling forms. The fauna includes the clypeasteroid echinoids L. depressum and fibulariid (?)gen. et sp. nov., a temnopleurid echinoid Zemnotrema macleayana (Tenison-Woods), a phymosomatoid echinoid stomechinid (?)gen. et sp. nov., and spatangoid echinoids B. latifrons and Ditremaster sp. indet. Five of the eight described echinoid genera from the Tertiary of PNG still live in waters of the Indo-Pacific today. Manuscript received 5 July 2001, accepted for publication 21 November 2001. KEYWORDS: Echinoidea, Echinodiscus, Laganum, Brisaster, Brissopsis, Temnotrema, Ditremaster, fibulariid, stomechinid, Tertiary, Papua New Guinea. INTRODUCTION Echinoids are a not uncommon component of Tertiary faunas of Papua New Guinea (PNG), but little systematic work on them has been completed. Tenison-Woods (1878) provided the first description of PNG Tertiary echinoids, describing the temnopleurid Zemnechinus macleayana Tenison-Woods, 1878 and noting the occurrence of the clypeasteroid Peronella decagonalis Lesson in A. Agassiz, 1872. This material was probably collected from the Lower Pliocene Kairuku Formation on Yule Island, northwest of Port Moresby. Jack and Etheridge (1892) provided a further description of Tenison-Woods’ (1878) material. Echinoids were also noted by Maitland (1892) from lateral equivalents of the Kairuku Formation near Delena, 3 km south-southeast of Yule Island. Chapman (1914) recorded indeterminate echinoid plates and spines from a limestone on the eastern side of Bootless Inlet, southeast of Port Moresby. This limestone, known as the Bootless Inlet Limestone, is of Upper Oligocene-Lower Miocene age (Tertiary Letter Stage lower Te) (Pieters 1978). Further consideration of Yule Island and Delena echinoid faunas appeared in reports of the Anglo-Persian Oil Company, when F. Chapman and I. Crespin (in Montgomery 1930) noted the cidaroids Phyllacanthus javanus K. Martin, 1885 and P sundaica K. Martin, 1885, the clypeasteroids Laganum sp., L. depressum Lesson in L. Agassiz, 1841, L. bonani Klein, 1734, L. elongatum L. Agassiz, 1841, Arachnoides placenta (Linnaeus), A. cf. placenta, Peronella sp. and Fibularia sp., Proc. Linn. Soc. N.S.W., 123. 2001 120 TERTIARY ECHINOIDS FROM P.N.G. the spantangoids Eupatagus pulchella (sic) (Herklots), Brissopsis sp. and Hemiaster sp., and the camarodonts Pleurechinus javanus (Martin) and Echinus cf. stracheyi. Numerous fragmentary echinoid remains were observed in surface and subsurface Tertiary formations in western Papua and New Guinea during the exploration work of Carne (1913), the Anglo-Persian Oil Company (1920-1929) and the Australasian Petroleum Company (1937- 1961). Although several cladistic classifications have recently been proposed for echinoids (Jensen 1981; Smith 1984), and the clypeasteroids in particular (Wang 1984; Mooi 1990a, 1990b), the classsification used herein follows that of McCormick and Moore (1966). Cladistic attempts at classification, such as Mooi (1990a), have relied heavily on characters usually only evident in extant forms, including external appendages, Aristotle’s lantern anatomy and test structure, and are not readily applicable to the fossil specimens used in this study. Collection details are provided in the Appendix. All specimens are housed in the Department of Geology, Australian National University, Canberra. Figure 1. Papua New Guinea, showing Aseki, Sagarai valley and Yule Island echinoid localities. BISMARCK SEA PAPUA NEW GUINEA Aseki @ Yule Island Sagari Valley CORAL SEA STRATIGRAPHY AND ECHINOID FAUNAS The echinoids described in this paper were collected at Aseki village in Morobe Province, the Sagarai valley in Milne Bay Province, and Yule Island in Central Province (Fig. 1), and represent four contrasting facies controlled assemblages. The echinoid fauna from Aseki was collected from a rubble of lithic sandstone developed on the Langimar Beds (Fig. 2). The Langimar Beds crop out in a 120 km belt straddling the flanks of the Owen Stanley Ranges, west of Wau and Bulolo (Smit et al. 1974). The formation consists of conglomerate, sandstone, interbedded marl, mudstone and calcarenite, passing southwards to silty mudstone, siltstone and sandstone with interbedded biohermal limestone, interpreted by Smit et al. (1974) as a shelfal facies. Abundant foraminifera in calcareous beds in the formation indicate a Middle Miocene (Tertiary Letter Stage lower Tf) age (Dow et al. 1974). The fauna is dominated by the Proc. Linn. Soc. N.S.W., 123. 2001 [.D. LINDLEY 121 Figure 2. Aseki locality, Menyamya district, Morobe Province. mE 4 10 000 91 90 000 mN =n v.! ——) Bulolo Windowi 91 80 000 mN \ Landing ground @ Collection locality Gamodoudou e@ 4 Proc. Linn. Soc. N.S.W., 123. 2001 fae Ulo Ulo Mine Creek \e 122 TERTIARY ECHINOIDS FROM P.N.G. clypeasteroids Echinodiscus bisperforatus Leske, 1778 and Laganum depressum Lesson in L. Agassiz, 1841. Both species are today widely distributed throughout the tropical and sub-tropical Indo-West Pacific. The Aseki echinoids are preserved in a fine-medium grained sand, typical of the sandy habitats occupied by modern clypeasteroids. In their life position, they probably burrowed in the uppermost sand layer, from which they obtained their food by particle-sieving (Seilacher 1979). Today, living clypeasteroids occupy extreme variants of the sandy habitat, as intertidal browsers, epibenthic deposit feeders in sea-grass meadows, surf dwellers, and stationary suspension feeders in the protected sands of bays (Seilacher 1979). Echinoids from the Sagarai valley were collected from exposures along a tributary of the Sagarai River (Fig. 3). Smith and Davies (1972) mapped these gently dipping tuffaceous sandstone exposures as the Padowa Beds. Delicately preserved echinoid tests are abundant at this locality and occur with gastropods, pelecypods, solitary corals and “larger” foraminifera. Foraminifera indicate an Upper Oligocene-Lower Miocene age (Tertiary Letter Stage Te) for the formation (D.J. Belford in Smith and Davies 1973). The burrowing spatangoids Brisaster latifrons (A. Agassiz, 1898) and Brissopsis ?luzonica (Gray, 1851), with deep anterior grooves, appear to have been dominant in the sand- sandy mud substratum of this locality. Excellent preservation is a function of their infaunal mode of life (Kier 1977). Both species are known in the Indo-Pacific today. Brissopsis luzonica is a common form in tropical and sub-tropical waters (Clarke and Rowe 1971; De Ridder 1986), whereas B. /atifrons appears to be restricted to sub-tropical and cooler waters (Mortensen 1951). The Yule Island specimens were collected from friable bioclastic limestone of the Lower Pliocene (zone N18-N19/20: Haig et al. 1993) Kairuku Formation. This inner- neritic carbonate sand facies includes coral-rich beds and Marginopora-rich sands, interpreted to have accumulated in sea-grass meadows (Haig et al. 1993). The rich echinoid fauna is dominated by infaunal irregular forms including clypeasteroids L. depressum and an unnamed fibulariid, and spatangoids Ditremaster sp. indet. and B. latifrons. Epifaunal echinoids are typically small and include the temnopleurid Temnotrema macleayana (Tenison-Woods, 1878) and an unnamed stomechinid. Some of the Yule Island echinoids were members of a predominantly infaunal sea-grass community, confined to coarse-grained sand in shallow water. A highly turbulent, nearshore niche of creviced rock with pockets of sand was inhabited by 7’ macleayana, the stomechinid and the fibulariid. The test of 7? macleayana is thick and strong and has a well- developed sutured plating, enabling it to withstand increased impact loading (Smith 1984), and allowing it to live in a highly turbulent shallow-water habitat. The stomechinid has a flattened test with a low down ambitus, an adaptation giving stability in currents on either rocky or sedimentary substrata. The fibulariid is distinctive for its well-developed marginal frill spines, and Mellita lata H.L. Clark, 1940, a common Carribean clypeasteroid that lives in sand of the breaker zone, is useful for comparative purposes. This species uses its frill spines for burrowing and, by bending them down, to reduce shifting (Seilacher 1979). Tertiary echinoid faunas of the surrounding region have been described from the Indonesian archipelago (Jeannet and Martin 1937) and Barrow Island, off the Pilbara coastline of northwestern Australia (McNamara and Kendrick 1994). The Barrow Island limestones represent the most northerly surface exposure of Miocene marine deposits in Australia (McNamara and Kendrick 1994), and their echinoid fauna is therefore of significance in relation to Tertiary faunas in PNG. McNamara and Kendrick (1994) noted that the Barrow Island fauna is typically tropical in nature, having strong affinities with those from the Miocene deposits of India and Java. By contrast, the strong links between Indo-West Pacific Miocene faunas do not exist with the well-documented faunas from the Miocene of southern Australia (McNamara and Kendrick 1994). All of McNamara and Kendrick’s (1994) recorded genera still live in the Barrow Island area today. Proc. Linn. Soc. N.S.W., 123. 2001 I.D. LINDLEY 123 There are no genera in common between the PNG and Barrow Island echinoid faunas, and this faunal mismatch may, to some extent, be influenced by facies differences. However, affinities with the Mio-Pliocene of the Indonesian archipelago are stronger, with Echinodiscus, Laganum and Temnotrema identified in both faunas. Five of the eight described echinoid genera from the Tertiary of PNG still live in waters of the Indo-West Pacific today. SYSTEMATIC PALAEONTOLOGY Class ECHINOIDEA Leske, 1778 Subclass EUECHINOIDEA Bronn, 1860 Superorder ECHINACEA Claus, 1876 Order TEMNOPLEUROIDA Mortensen, 1942 Family TEMNOPLEURIDAE A. Agassiz, 1872 Genus TEMNOTREMA A. Agassiz, 1863 Type species Temnotrema sculptum A. Agassiz, 1872, by original designation. Temnotrema macleayana (Tenison-Woods) Figs 4d-f, 5 Synonymy Temnechinus Macleayana Tenison-Woods, 1878, p. 126; Etheridge 1889, p. 173, 178; Etheridge 1892, p. 209, 214; Jack and Etheridge 1892, p. 691. Temnechinus Macleayi (sic) Tenison-Woods; Tate 1894, p. 213, 214. Temnechinus Macleaya (sic) Tenison-Woods; Carne 1913, p. 17. Pleurechinus javanus (Martin); F. Chapman and I. Crespin in Montgomery 1930, Wo Stee Dicoptella agassizi Lambert and Thiéry; Lambert and Jeannet 1935, p. 34. Dicoptella agassizi var. tenuis Jeannet in Lambert and Jeannet 1935, p. 34. Dicoptella agassizi var. elevata Jeannet in Lambert and Jeannet 1935, p. 34. Dicoptella leupoldi Jeannet in Lambert and Jeannet 1935, p. 38. Dicoptella tobleri Jeannet in Lambert and Jeannet 1935, p. 39. Dicoptella cf. tobleri Jeannet in Lambert and Jeannet 1935, plate 2: figs 13-15. Dicoptella javana Jeannet in Lambert and Jennet 1935, p. 40. Temnotrema macleayana (Tenison-Woods); Philip 1969, p. 235. Description Test small, 13 mm diameter, hemispherical with oral surface sunken around peristome, thick and strong. Details of apical system unknown. Ambulacra at ambitus are about 2/3 as wide as interambulacra. Ambulacral plates compound, composed of three individual plates each bearing pore pair; pores in straight series, small, equal sized. A prominent elevated primary tubercule is present in the middle of each plate; tubercules are imperforate and noncrenulate. Primary tubercules form a distinct vertical series; each tubercule surrounded adapically by a semi-circular row of secondary tuberculation. Large deep, elongate-ovoid sutural pits excavated at end of horizontal suture between adjacent plates. Interambulacral plates about equal in height to opposite ambulacral plate. Each possesses a prominent imperforate, noncrenulate primary tubercule; elevated, occurring in the middle of each plate. Tubercules form two regular vertical series. Irregular secondary tubercules surround each primary tubercule on all plates. Deeply excavated elongate-ovoid sutural pits present at medial and adradial end of all horizontal sutures; pits spaced well apart on oral and apical Proc. Linn. Soc. N.S.W., 123. 2001 124 TERTIARY ECHINOIDS FROM P.N.G. surfaces. Peristome moderate, 1/4 of horizontal diameter; gill slits indistinct; other details of peristome unknown. Remarks Temnotrema A. Agassiz, 1863 is a Miocene-Recent temnopleurid with many species. The Recent species of Zemnotrema (numbering at least 13) are, like most temnopleurids, warm water forms and are distributed throughout the Indo-Pacific (Red Sea to Hawaii, Japan to Australia) (Mortensen 1943; Fell and Pawson 1966). Fossil species of Temnotrema are known from the Miocene of Java and Borneo and the Pliocene of Indonesia (Fell and Pawson 1966). Several species with strong affinities to TZ. macleayana have been described from the Miocene of Java (Lambert and Jeannet 1935; Philip 1969: ( 2335). The Yule Island species of Jemnotrema was originally assigned to Temnechinus by Tenison- Woods (1878) because of affinities with Zemnechinus lineatus Duncan, 1877, from the Pliocene Sandringham Sand, Port Philip Bay, and Temnechinus globosus Forbes, 1852, from the Pliocene of England. 7: /ineatus has subsequently been designated as the type species of Otholophus Duncan, 1887, and some writers (Lambert and Thiéry 1910) have referred T. globosus to Dicoptella Lambert, 1907, now regarded as a synonym of Temnotrema (Mortensen 1943). Etheridge (1889, 1892) and Tate (1894) expressed doubt about Tenison- Woods’ (1878) generic assignment. Etheridge (1892: 214) considered that the test ‘does not possess the typical excavations along the sutural margins of plates seen in all true forms of Temnechinus, nor are the ambulacral plates confluent’. Figure 4. Regular echinoids. Stomechinid (?)gen. et sp. nov. Lower Pliocene, Yule Island, Central Province. 4a-c, UPNG F1184, aboral, oral and lateral views. Temnotrema macleayana (Tenison- Woods). Lower Pliocene, Yule Island, Central Province. 4d-f, UPNG F1181, aboral, oral and lateral views. Bar scale = 0.5 cm. Proc. Linn. Soc. N.S.W., 123. 2001 I.D. LINDLEY 125 Figure 5. Temnotrema macleayana (Tenison-Woods). Lower Pliocene, Yule Island, Central Province. 5a,b, UPNG F1181, plating diagrams at ambitus for ambulacrum, interambulacrum. Mortensen (1943: 246) noted that details of spines, the pedicellariae, periproct and colour are of importance in identifying tests of Recent temnopleurids, and for fossil species, where details of tuberculation and the shape of sutural pits provide species distinction, identifications are very unreliable. Philip (1969) echoed these concerns, noting a preoccupation of previous work on fossil temnopleurids “with minor details of the sculpture or ornament for the purposes of taxonomic discrimination - which features are acknowledged to be exceedingly variable in adequately known species’. Palaeontologists have answered this problem of variation ‘by naming extremes of variation in an assemblage’ (Philip 1969). The nature of sutural plating for the Yule Island species of Temnotrema is very similar to that described for Temnechinus by Forbes (1852). The arrangement of pore pairs is regarded as a significant distinction between both forms. The pore pairs of 7. macleayana are arranged in a straight series, contrasting with those of 7Zemnechinus, with pairs arranged in oblique arcs of three. Philip (1969) noted a twofold subdivision of fossil temnopleurids using test morphology, viz. forms with a sculptured test, and those with sutural pits. He noted that all Australian Tertiary temnopleurid species are sculptured forms, with the pitted forms well represented in the Indo-Pacific. TZ macleayana clearly falls within the Indo-Pacific realm. Smith (1984) interpreted the evolution of sutured plating as an important advance that allowed an increase in shock-resistance capabilities of the test, allowing echinoids to invade highly turbulent shallow-water habitats. Proc. Linn. Soc. N.S.W., 123. 2001 126 TERTIARY ECHINOIDS FROM P.N.G. Material UPNG F1181, a complete test from the Kairuku Formation, Yule Island, Central Province, PNG. Lower Pliocene. Figure 6. Stomechinid (?) gen. et sp. nov. Lower Pliocene, Yule Island, Central Province. 6a,b, UPNG F 1184, plating diagrams at ambitus for ambulacrum, interambulacrum. Order PHY MOSOMATOIDA Mortensen, 1904 Family STOMECHINIDAE Pomel, 1883 Genus UNCERTAIN (?nov.) Stomechinid (?) gen. et sp. nov. Figs 4a-c, 6 Description Test small, 20 mm diameter, subhemispherical, height of test about 1/2 of its diameter. Ambitus relatively low, rounded pentagonal outline. Oral surface distinctly concave. Apical system small, with central periproct. Ambulacra at ambitus are about 1/ 2 width of interambulacra. Ambulacral plates compound, trigeminate, pore pairs in arcs of three. A prominent elevated primary tubercule in present in the middle of each plate, forming vertical series. Tubercules imperforate, noncrenulate, as large as interambulacral primaries. Secondary tubercules, not as large as primaries, with a tendency to align in a vertical series on medial side of primary tubercule. Interambulacral plates about equal in height to opposite ambulacral plate. Each possesses a central prominent imperforate, noncrenulate primary tubercule; surrounded by annulus of smaller secondaries, inturn flanked by about four larger secondaries. Peristome large, about 1/3 horizontal diameter of test; gill slits are distinct; other details of peristome unknown. Proc. Linn. Soc. N.S.W., 123. 2001 I.D. LINDLEY 127 Remarks The Yule Island specimen with its large peristome (typical of all stirodont orders), imperforate primary tubercules and trigeminate ambulacral plates can be clearly placed in Order Phymosomatoida. The specimen’s noncrenulate primary tubercules, with ambulacral primaries as large as interambulacral primaries, and a large peristome with distinct gill slits readily places it in Family Stomechinidae. The stomechinids include 21 genera ranging from the Lower Jurassic-Recent (Fell and Pawson 1966). The low hemispherical test with rounded pentagonal ambital outline is distinctive, but formal description of the single Yule Island specimen is deferred until more material is available. Trochalosoma Lambert, 1897 from the Late Cretaceous of Jamaica has a similar ambital outline with trigeminate ambulacral plates, but the test is flattened and wheel-shaped (Fell and Pawson 1966). Similarly, both Gomphechinus Pomel, 1883 from the Late Cretaceous of North Africa and Madagascar, and Phymechinus Desor, 1856 from the Middle Jurassic-Late Cretaceous of Europe, are polyporous forms that possess an ambitus that is low down, but their test outline is circular (Fell and Pawson 1966). The noticeably low hemispherical profile of the Yule Island test, with a low down ambitus with subrounded outline, is similar to that described for Holectypus depressus (Leske), a Jurassic irregular echinoid (Smith 1984). Its test shape was interpreted by Smith (1984: 101) to be an adaptation giving stability in currents on either rocky or sedimentary substrata, or an adaptation for burrowing. The lack of any difference in size or shape between oral and apical tubercules on the Yule Island test suggests that burrowing was unlikely. Material UPNG F1184, a complete test from the Kairuku Formation, Yule Island, northwest of Port Moresby, Central Province, PNG. Lower Pliocene. Superorder GNATHOSTOMATA Zittel, 1879 Order CLYPEASTEROIDA A. Agassiz, 1872 Suborder SCUTELLINA Haeckel, 1896 Family ASTRICLYPEIDAE Stefanini, 1911 Genus ECHINODISCUS Leske, 1778 Type species Echinodiscus bisperforatus Leske, 1778, by subsequent designation of ICZN, 1950; Recent, Indo-Pacific. Echinodiscus bisperforatus Leske, 1778 Fig. 7d Synonymy Echinodiscus bisperforatus Leske, 1778, p. 196. Echinodiscus bisperforatus truncatus (L. Agassiz), H.L. Clark, 1914, p. 72. T. Mortensen (1948), A Monograph of the Echinoidea 4(2), Clypeasteroida, p. 406 and p. 410, lists the previous synonymies. Description Test of medium size 65 mm x 70 mm, low discoidal with semi-circular outline, mostly truncate at the posterior end. Apical system is central; petals straight, narrow and short, extending slightly less than one-half distance to margin; distinctly closed distally; pore zones slightly less than the width of interpore zones. Details of apical system unknown. Two closed lunules or slits in the posterior ambulacra, narrow, about as long as the petals. Details of oral surface unknown. Proc. Linn. Soc. N.S.W., 123. 2001 128 TERTIARY ECHINOIDS FROM P.N.G. Figure 7. Clyperasteroid echinoids. Fibulariid (?) gen. et sp. nov. Lower Pliocene, Yule Island, Central Province. 7a-b, UPNG F1183, aboral, oral views. Bar scale = 0.5 cm; 7c, UPNG F1183, detail of apical system showing ambulacral position of genital pores. Bar scale = 0.25 cm. Echinodiscus bisperforatus Leske, 1778. Middle Miocene, Aseki village, Morobe Province. 7d, ANU 60549, aboral view. Bar scale = 1.0 cm. Laganum depressum Lesson in L. Agassiz, 1841. Lower Pliocene, Yule Island. 7e, UPNG F1182, aboral view. Bar scale = 0.5 cm. CU A bd ay LADO: Proc. Linn. Soc. N.S.W., 123. 2001 1.D. LINDLEY 129 Remarks The author follows Clark and Rowe’s (1971) key which, for Echinodiscus, used lunule length relative to test radius or petal length as a diagnositic character. The possession of closed posterior lunules of a similar length to petals clearly indicates assignment of the Aseki specimen to Echinodiscus bisperforatus Leske, 1778. E. tenuissimus (L. Agassiz, 1847), a similar species in many respects, possesses lunules shorter than petals. Although Mortensen (1948: 409) observed that the length of lunules varies very considerably within this species, it is useful to note that the lunules of the Aseki specimen are at most about as long the petals, a diagnostic character of var. truncatus (L. Agassiz, 1841). The status of Echinodiscus bisperforatus truncatus is unclear. Clark and Rowe (1971) did not recognise var. truncatus in their monograph of extant Indo-West Pacific echinoderms, whereas De Ridder (1986) described its occurrence in New Caledonia. Mortensen (1948) described an Indo-West Pacific distribution for both the main form and the var. truncatus, from the Red Sea and East Coast of Africa to New Caledonia, with no record from the Philippines and farther north. Mortensen (1948) did not record any fossil species of Echinodiscus Leske, 1778, but Durham (1966) noted a Miocene-Recent range for the genus. The amended range clearly includes the Aseki specimen. Material ANU 60549, a complete test from the Langimar Beds, headwater course of a small tributary of the Kapau River, Aseki village, Morobe Province, PNG. Middle Miocene. Suborder LAGANINA Mortensen, 1948 Family LAGANIDAE A. Agassiz, 1873 Genus LAGANUM Klein, 1734 Type species Laganum petalodes Link, 1807, by original designation. Laganum depressum Lesson in L. Agassiz, 1841 Fig. 7e Synonymy ?Echinarachnius conchatus M’ Clelland, 1840, p. 181. Laganum multiforme K. Martin, 1880, p. 3. Laganum depressum Lesson in L. Agassiz, 1841, F Chapman and I. Crespin in Montgomery 1930, p. 57. Laganum boschi Jeannet and Martin, 1937, p. 253. T. Mortensen (1948), A Monograph of the Echinoidea 4(2), Clypeasteroida, p. 313, lists the previous synonymies. Description Test flattened with rounded pentagonal outline; length of test varies from 28 to 32 mm, the length exceeding the breadth; broadest at the antero-lateral petals. Test of F1182 is weakly reenteringly curved in the interambulacra. Edge of test is inflated forming a broad margin; inside inflated margin the test is somewhat sunken, then rises to the apical system. The oral surface is weakly concave. Apical system central and raised; pentagonal with apices opposite interambulacra. Petaloid area is large, about 2/3 the test length, the petals reaching the marginal inflated region. Genital pores five, located at adapical end of interamulacra. Petals are relatively broad, closed distally. The anterior petal of F1180 is slightly longer than others. Pore pairs consist of inner, smaller circular Proc. Linn. Soc. N.S.W., 123. 2001 130 TERTIARY ECHINOIDS FROM P.N.G. pore and outer, larger elongated pore. Interporiferous zone is covered by scattered primary tubercles amongst miliary tubercles. Tuberculation on remainder of aboral surface and on oral surface consists of similarly scattered primary tubercules amongst numerous miliary tubercles. The periproct of F1180 near posterior edge of test, c. 25 per cent of distance from mouth to posterior edge; transversly elongate. Details of peristome and ambulacral furrows unknown. Remarks Laganum is an Eocene-Recent clypeasteroid (Durham 1966) with numerous fossil and extant species (Clark 1938; Jeannet and Martin 1937; Mortensen 1948; De Ridder 1986). Laganum depressum Lesson in L. Agassiz, 1841 is a well known Recent species widely distributed throughout the tropical and sub-tropical Indo-West Pacific (Mortensen 1948). Fossil L. depressum are recorded from the ?Miocene-Pleistocene (Mortensen 1948). The ?Miocene occurrence of the species is from Fiji (Mortensen 1948). Laganum multiforme K. Martin, 1880 and Laganum boschi Jeannet and Martin, 1937, both described from the Pliocene of Java by Jeannet and Martin (1937), are considered as identical to L. depressum by Mortensen (1948). The Yule Island specimens are identical in almost all respects to Recent specimens examined by the writer from the south coast of New Britain, PNG, and the Philippines. The fossil tests are comparatively thinner than Recent tests, probably a result of compaction by enclosing sediment during lithifaction. Material Three tests: two complete tests UPNG F1180 and F1182 from the Kairuku Formation, Yule Island, northwest of Port Moresby, Central Province and ANU 60556, a poorly preserved test from the Langimar Beds, tributary of the Kapau River, Aseki village, Morobe Province, PNG. Middle Miocene-Lower Pliocene. Family FIBULARIIDAE Gray, 1855 Genus UNCERTAIN (?nov.) Fibulariid (?) gen. et sp. nov. Figs 7a-c Synonymy Peronella decagonalis Lesson in A. Agassiz, 1872-74; Tenison-Woods 1878, p. 126; Etheridge 1889, p. 173, 178; Etheridge 1892, p. 209, 215; Jack and Etheridge 1892, p. 692; Tate 1894, p. 213, 214; Carne 1913, p. 17 [non Laganum decagonale (Blainville, 1827)]. Peronella sp., F. Chapman and I. Crespin in Montgomery 1930, p. 57. Echinodiscus lesueuri: Jeannet and Martin 1937, p. 254 [non Peronella lesueuri Valenciennes in L. Agassiz, 1841]. Description Test markedly flattened with elliptical outline; length of 35 mm; broadest at the anterolateral petals; edge of test is weakly inflated forming a narrow margin. The apical system raised and slightly anterior; composed of single large madreporite plate; stellate with apices opposite interambulacra. Genital pores five, located midway along sides of central plate, at adapical end of ambulacra (Fig. 7c). Petaloid area is large, about 2/3 test length. Petals are relatively narrow and closed distally; the anterior petal is slightly longer than the others. Plates of petals apparently simple, running across half the width of the petal. Pores small, about equal sized, conjugate. Interporiferous area is fairly narrow, covered by scattered primary tubercles and numerous miliary tubercules. Most of the Proc. Linn. Soc. N.S.W., 123. 2001 I.D. LINDLEY 131 remainder of aboral surface is covered by scattered coarse primary tubercles and fine miliary tubercles; densely spaced primary tubercles are present around edge of test on the weakly inflated margin. Pattern of tuberculation on the oral surface is similar to that of the aboral surface. Oral surface shallowly concave; ambulacral furrows absent. The periproct is transversely elongate; nearer the posterior margin than the mouth. Remarks There are no food grooves on the well preserved oral surface of the Yule Island test and although, as pointed out by the anonymous reviewer, the flattened body and well developed petals are typically laganid, the specimen is tentatively assigned to Family Fibulariidae. Durham (1966) and Smith (1984) attached considerable importance to food grooves and the nature of their branching in their classification of the clypeasteroids. Clypeasteroids lacking food grooves were assigned to Family Fibulariidae. By contrast, Mooi (1990a) gave little attention to food grooves in his cladistic classification of the order. The fibulariids according to Durham’s (1966) classification include a diverse group of species, typically small, but with very variable test shape and petals, ranging from the Paleocene-Recent (Kier 1982). Durham (1966) and Kier (1982) regarded many of the characters of fibulariids as primitive and, with their relative stratigraphic position, the family was interpreted to represent the ancestral stock of other clyperasteroids. Mooi (1990a) presented an opposing view, that the fibulariids were highly specialised forms not at all representative of an ancestor to the clypeasteroids. Fossil fibulariids from the region include Fibularia gregata Tate, 1885 and Scutellina patella Tate, 1891 from the Eocene-Miocene and Eocene, respectively, of Australia (Tate 1891) and Echinocyamus sp. and Fibularia rhedeni Jeannet, 1937 from the Lower Miocene and Oligocene-Miocene, respectively, of the Indonesian archipelago (Jeannet and Martin 1937). McNamara and Kendrick (1994) noted the presence of poorly preserved Fibularia sp. from the Middle Miocene of Barrow Island, off the Pilbara coast of Western Australia. Although the Yule Island specimen is tentatively assigned to Family Fibulariidae, generic assignment of this single specimen is problematical and circumstances do not permit collection of additional material. Two characters appear to distinguish it from other fibulariids. Firstly, the position of the genital pores midway along the sides of the large stellate madreporite plate, a condition found in some apical systems of the monobasal Clypeasteroidea (Melville and Durham 1966: U229), does not appear to be present in the fibulariids. Secondly, the test’s size is significantly larger than for other fibulariids. Tenison-Woods’ (1878) specimen measured about 25 mm diameter and the present specimen 35 mm. The small size of the fibulariids allows food to be passed directly to the mouth by tube feet, obviating the need for oral ambulacral food grooves along which food particles are transported to the mouth in mucus strings (Seilacher 1979; Smith 1984). This method of food gathering was also used by early clypeasteroids. The presence of frill spines in an outer margin position, indicated by densely spaced tubercules around the edge of the Yule Island test (Figs 7a,b), served to sieve sand and also probably act as a steering device during burrowing (Seilacher 1979). Material UPNG F1183, a complete test from the Kairuku Formation, Yule Island, northwest of Port Moresby, Central Province, PNG. Lower Pliocene. Proc. Linn. Soc. N.S.W., 123. 2001 132 TERTIARY ECHINOIDS FROM P.N.G. Figure 8. Spatangoid echinoids. Ditremaster sp. indet. Lower Pliocene, Yule Island, Central Province. 8a-c, UPNG F1186, aboral, posterior and lateral views. Brisaster latifrons (A. Agassiz, 1898). Lower Pliocene, Yule Island, Central Province. 8d-f, UPNG F1179, aboral, oral and lateral views. Bar scale = 0.5 cm. Proc. Linn. Soc. N.S.W., 123. 2001 I.D. LINDLEY 133 Superorder ATELOSTOMATA Zittel, 1879 Order SPATANGOIDA Claus, 1876 Suborder HEMIASTERINA Fischer, 1966 Family HEMIASTERIDAE Clark, 1917 Genus DITREMASTER Munier-Chalmas, 1885 Type species Hemiaster nux Desor, 1853, by subsequent designation of Cotteau, 1887; Eocene, France. Ditremaster sp. indet. Figs 8a-c Description Test small with ovoid outline; sub-globular, rather high, vaulted. Test outline weakly notched at the frontal ambulacrum. Apical system subcentral, ethmophract, with madreporite not separating posterior oculars. Two gonopores, although genital 5 in F1186 appears to be perforated by two pores. Mortensen (1948: 316) noted similar genital pore doubles in clypeasteroid species. Frontal sinus slight. Petals well developed, straight, closed, somewhat sunken, the posterior pair very short, about 1/ 3 length of anterior ones. Pores small, elongate and conjugate. Scattered granules present in interporiferous and pore zones. Peripetalous fasciole well developed but no other fascioles. Periproct at upper end of curved posterior. Area within peripetalous fasciole covered by densely spaced tubercules. Remainder of aboral surface covered by scattered tubercules. Details of peristome unknown. Other details of oral surface unknown. Remarks The possession of closed paired petaloid ambulacra and a well developed peripetalous fasciole, and no others, indicates assignment to Family Hemiasteridae. This family includes 15 genera ranging from the Upper Cretaceous-Recent, the modern forms mainly mud-dwellers (Fischer 1966). The sub-globular shape of the test, with a faint frontal sinus, two gonopores, and the very short posterior ambulacral pair suggest assignment to Ditremaster Munier-Chalmas, 1885. Known only from fossil, the genus ranges from the Eocene-Pliocene (Fischer 1966). It has not been possible to review Lambert and Thiéry (1924), who apparently referred thirty species to this genus (Mortensen 1950), and specific referral is deferred. Both Mortensen (1950) and Fischer (1966) have figured the type species, D. nux (Desor) from the Eocene of France. The test of the type species has a sub-circular outline, distinct from the ovoid outline of the Yule Island tests. Material Two tests: UPNG F1186, an undeformed virtually complete specimen and F1185, a weakly deformed incomplete specimen, both from the Kairuku Formation, Yule Island, Central Province, PNG. Lower Pliocene. Family SCHIZASTERIDAE Lambert, 1905 Genus BRISASTER Gray, 1855 Type species Brissus fragilis Diiben and Koren, 1844, by original designation; Recent. Proc. Linn. Soc. N.S.W., 123. 2001 134 TERTIARY ECHINOIDS FROM P.N.G. Figure 9. Spatangoid echinoids. Brissopsis ?luzonica (Gray, 1851). Upper Oligocene-Lower Miocene, Sagarai valley, Milne Bay Province. 9a,b, ANU 60551, aboral and oral views of internal cast. Brisaster latifrons (A. Agassiz, 1898). Upper Oligocene-Lower Miocene, Sagarai valley, Milne Bay Province. 9c-e, ANU 60550, aboral, oral and lateral views of internal cast. Bar scale = 0.5 cm. Proc. Linn. Soc. N.S.W., 123. 2001 I.D. LINDLEY 135 Brisaster latifrons (A. Agassiz, 1898) Figs 8d-f, 9c-e Synonymy Schizaster latifrons A. Agassiz, 1898, p. 81. Hemiaster sp., F. Chapman and I. Crespin in Montgomery 1930, p. 57. T. Mortensen (1951), A Monograph of the Echinoidea 5(2), Spatangoida II, p. 289, lists the previous synonymies. Description Test of small to medium size, broadly rounded almost as broad as long, F1179 measuring 34 mm x 30 mm and ANU 60550 measuring approximately 60 mm x 51 mm; test rather low, highest towards posterior, oral side gently convex. Apical system posterior; this area not well preserved to show details of genital pores; posterior petals short, about 1/3 the length of the anterior ones; frontal ambulacrum broad and moderately to deeply excavated, forming a distinct notch in anterior end of test; anterolateral petals are gently curved. Details of periproct unknown, apparently located on truncated posterior end of test. Peristome close to anterior end of test shallowly sunken, with the sunken peristomal region continuing directly to frontal notch. Peripetalous fasciole well developed in F1179; a lateral fasciole passes posteriorly to meet with an anal fasciole, that is partially preserved in this specimen. Both oral and aboral surface with uniform covering of tubercules; no large (primary) tubercules present. Remarks Although the apical system is not well preserved, the general shape of the test, and the arrangement of the petals and preserved fascioles distinguishes these specimens as members of Family Schizasteridae. The low shape of the test clearly distinguishes it from other schizasterids as being typical of Brisaster Gray, 1855. Brisaster includes eight species ranging from the Oligocene-Recent (Mortensen 1951; Fischer 1966). Recent species of the genus are distributed throughout the northern Atlantic, northern Pacific and South African seas from bathymetric depths of 40 to 1,300 m (Mortensen 1951). The Sagari valley and Yule Island tests are clearly distinguishable from Oligocene Brisaster maximus Clark, 1937, a large species from Oregon with a test about 100 mm long and straight anterior petals (Mortensen 1951). However, both tests have several characters shared with Recent Brisaster latifrons (A. Agassiz, 1898), as described by Mortensen (1951), including low test height posteriorly, very short posterior petals and a broad frontal ambulacrum. Material Two tests: UPNG F1179, an undeformed virtually complete specimen from the Kairuku Formation, Yule Island, Central Province; and ANU 60550, an incomplete internal cast from the Padowa Beds, Wabalam Creek, northern Sagari valley, Milne Bay Province, PNG. Upper Oligocene-Lower Pliocene. Proc. Linn. Soc. N.S.W., 123. 2001 136 TERTIARY ECHINOIDS FROM P.N.G. Suborder MICRASTERINA Fischer, 1966 Family BRISSIDAE Gray, 1855 Genus BRISSOPSIS L. Agassiz, 1840 Type species Brissus lyrifer Forbes 1841, by subsequent designation of Desor, 1858; Recent, Gulf of Mexico. Brissopsis ?luzonica (Gray, 1851) Figs 9a,b Synonymy Kleinia latior Gray, 1851, p. 133. T. Mortensen (1951), A Monograph of the Echinoidea 5(2), Spatangoida II, p. 397, lists the previous synonymies. Description Test of moderate size, length about 55 mm, width 44 mm, height about 22 mm; with an ovoid outline. Test outline weakly notched at the anterolateral ambulacra. Frontal notch distinct; the frontal depression deep. Details of posterior end not preserved. The anterior petals are distinctly sunken and divergent, with a tendency to parallel proximally. Pore pairs are preserved in ambulacrum IV, with outer series distinctly larger. A narrow ridge formed at the proximal end of the interambulacra | and 4 appears to separate the anterior and posterior petals. Posterior petals are not preserved. The frontal ambulacrum is sunken. Other details of this ambulacrum not known. Interambulacra consist of typical alternating series of plates; other details unknown. Details of apical system unknown. Peristome near anterior end of test, deeply sunken; sunken region continuing to frontal notch; other details unknown. Details of periproct unknown. Details of fascioles unknown. Remarks Brissopsis L. Agassiz, 1840 is an Oligocene-Recent micrasterid and a discussion of its many species is found in Mortensen (1951) and McNamara et al. (1986). At least 10 fossil species are known from the Tertiary of Europe, the Middle East, southeast Asia, Australasia and Central America. Recent species number at least 17 and are widely distributed in both the Atlantic and Indo-Pacifc, with shallow water (12 to 45 m) and deep water (130 to 2,980 m) forms known (Mortensen 1951). Mortensen (1951: 378) considered that the arrangement of posterior petals, whether divergent or to a varying extent parallel or confluent proximally, to be an important diagnostic character amongst species of Brissopsis. The curvature of the anterior petals of the Sagari valley test is consistent with the development of semi-circular anterior and posterior petals on each side of the test. This petal arrangement is distinctive in Brissopsis tatei Hall, 1907, Brissopsis crescenticus Wright (1855) and Brissopsis luzonica (Gray, 1851) (Hall 1907; Mortensen 1951; McNamara et al. 1986). B. tatei is an Early-Middle Miocene species from Victoria and South Australia and B. crescenticus, and a nominate form var. syriaca Vautrin, 1933, are known from the Miocene of Malta and Syria (McNamara et al. 1986). B. luzonica is a Recent species widely distributed throughout the Indo-West Pacific, including off the northern coasts of Australia (Mortensen 1951; McNamara et al. 1986). The poor preservation of the Sagarai valley test does not permit a thorough consideration of affinities with closely allied species, and the specimen is tentatively referred to B. luzonica because of more or less conspicuous notches at the antero-lateral ambulacra and a deep anterior groove. Nichols (1959) noted that a deep anterior groove, providing a channel for the passage of food currents from the aboral surface to the mouth, is related to the nature of Proc. Linn. Soc. N.S.W., 123. 2001 I.D. LINDLEY 137 the substratum inhabited. Forms with deep anterior grooves inhabit sand or sandy mud, where it is difficult to maintain a tube from the aboral surface round the anterior ambitus to the oral surface, and presumably it is necessary for spines to arch over the groove and make a roof to it (Nichols 1959). Material ANU 60551, an incomplete internal cast of a test with some details of ambulacrum IV from the Padowa Beds, Wabalam Creek, northern Sagari valley, Milne Bay Province, PNG. Upper Oligocene-Lower Miocene. ACKNOWLEDGMENTS The specimen of Echinodiscus bisperforatus was kindly presented to the author in 1978 by the Rev'd Dieter and Ruth Geisler, then of the Aseki Mission of the Evangelical Lutheran Church of Papua New Guinea. Echinoid material from Yule Island, Central Province, was kindly provided by Russell Perembo, Department of Geology, University of Papua New Guinea. Hugh Davies, also at the Department of Geology, University of Papua New Guinea, is thanked for his interest in the project. Most species assignments were discussed at length with Ken Campbell, but ultimate responsibility rests with the author. A draft of the manuscript was reviewed by Ken Campbell. The comments of an anonymous reviewer improved the manuscript. Photography of specimens and dark room work were completed with the assistance of Richard Barwick. Bev Allen, librarian at the Australian Geological Survey Organisation, and Helen Taylor, Pacific Collections librarian at the Australian National University are acknowledged for their effort in locating reports of the Anglo-Persian Oil Company. This research was completed while the author was a Visiting Fellow in the Department of Geology, Australian National University, and David Ellis, Head of Department, is thanked for the provision of departmental facilities. REFERENCES Anglo-Persian Oil Company. (1930). “The oil exploration work in Papua and New Guinea conducted by the Anglo-Persian Oil Company on behalf of the Government of the Commonwealth of Australia’. 4 vols. (Harrison and Sons: London). Australasian Petroleum Company Proprietary. (1961). Geological Results of Petroleum Exploration in Western Papua 1937-1961. Journal of the Geological Society of Australia 8, 133p. Carne, J.E. (1913). Notes on the occurrence of coal, petroleum, and copper in Papua. Bulletin of the Territory of Papua 1, 116p. Chapman, F. (1914). Description of a limestone of Lower Miocene age from Bootless Inlet, Papua. Journal and Proceedings of the Royal Society of New South Wales 48, 281-301. Clark, A.M. and Rowe, F.W.E. (1971). Monograph of shallow-water Indo-Pacific echinoderms. Trustees of the British Museum (Natural History) Publication 690. Clark, H.L. (1938). Echinoderms from Australia: An account of collections made in 1929 and 1932. Memoirs of the Harvard Museum of Comparative Zoology 55, 596p. De Ridder, C. (1986). Les échinides. In ‘Guide des étoiles de mer, oursins et autres échinodermes du lagon de Nouvelle-Calédonie’ (Eds A. Guille, P. Laboute and J.-L. Menou) pp. 22-53. (Institut Francais de Recherche Scientifique pour le Développement en Coopération. Collection Faune Tropiclae 25). Dow, D.B., Smit, J.A.J. and Page, R.W. (1974). Wau, 1:250,000 Geological Series - Explanatory Notes. (Bureau of Mineral Resources, Geology and Geophysics and Department of Lands, Surveys and Mines: Canberra and Port Moresby). 33p. Durham, J.W. (1966). Clypeasteroids. In ‘Treatise on Invertebrate Paleontology, Part U, Echinodermata 3’ (Ed. R.C. Moore) pp. U450-U491. (Geological Society of America and University of Kansas Press: Lawrence). Etheridge, R. Jr. (1889). Our present knoweldge of the palaeontology of New Guinea. Records of the Geological Survey of New South Wales 1(3), 172-179. Etheridge, R. Jr. (1892). Our present knowledge of the palaeontology of New Guinea. In “British New Guinea’ (J.P. Thomson). pp. 208-215. (George Philip and Son: London). Fell, H.B. and Pawson, D.L. (1966). Echinacea. In “Treatise on Invertebrate Paleontology, Part U, Echinodermata 3’ (Ed. R.C. Moore) pp. U367-U440. (Geological Society of America and University of Kansas Press: Lawrence). Proc. Linn. Soc. N.S.W., 123. 2001 138 TERTIARY ECHINOIDS FROM P.N.G. Fischer, A.G. (1966). Spatangoids. In ‘Treatise on Invertebrate Paleontology, Part U, Echinodermata 3’ (Ed. R.C. Moore) pp. U543-U628. (Geological Society of America and University of Kansas Press: Lawrence). Forbes, E. (1852). Monograph of the Echinodermata of the British Tertiaries. Palaeontographical Society 36p. Haig, D.W., Perembo, R.C.B., Lynch, D.A., Milner, G. and Zammit, M. (1993). Marine stratigraphic units in Central Province, Papua New Guinea: Age and depositional environments. In ‘Petroleum Exploration and Development in Papua New Guinea: Proceedings of the Second PNG Petroleum Convention, Port Moresby’ (Eds. G.J. Carman and Z. Carman) pp. 47-60. (PNG Chamber of Mines and Petroleum: Port Moresby). Hall, T.S. (1907). Four new echinoids from the Australian Tertiary. Proceedings of the Royal Society of Victoria 19, 47-53. Jack, R.L. and Etheridge, R. (1892). ‘The Geology and Palaeontology of Queensland and New Guinea’. (Government Printer: Brisbane). Jeannet, A. and Martin, R. (1937). Ueber Neozoische Echinoidea aus dem Niederlaendisch-Indischen Archipel. Leidsche geologische mededeelingen 8(2), 215-308. Jensen, M. (1981). Morphology and classification of Euechinoidea Bronn, 1860 - a cladistic analysis. Videnskabelige Meddelesar Dansk Naturhistoriske Forening i Kjobenhavn 143, 7-99. Kier, P.M. (1977). The poor fossil record of the regular echinoid. Paleobiology 3, 168-174. Kier, P.M. (1982). Rapid evolution in echinoids. Palaeontology 25, 1-9. Lambert, J. and Jeannet, A. (1935). Contribution a |’ étude des Echinides tertiaires des iles de la Sonde: Mémoires de la Société Paléontologique Suisse 56, 1-76. Lambert, J. and Thieéry, P. (1910). Essai de nomenclature raisonnée des échinides. pp. 81-160. (Libraire Ferriére: Chaumont). Lambert, J. and Thiéry, P. (1924). Essai de nomenclature raisonnée des échinides. pp. 385-512. (Libraire Ferriére: Chaumont). Lindley, I.D. (1991). Initial Geological Observations in P.A. 958 - Dawa Dawa River, Milne Bay Province. Report for Highlands Gold Limited. (Geological Survey of Papua New Guinea). Maitland, A.G. (1892). Geological Observations in British New Guinea in 1891. In Appendix M “Papua New Guinea Annual Report 1891-1892’. pp. 53-85. McCormick, L. and Moore, R.C. (1966). Outline of Classification. In ‘Treatise on Invertebrate Paleontology, Part U, Echinodermata 3’ (Ed. R.C. Moore) pp. U295-U297. (Geological Society of America and University of Kansas Press: Lawrence). McNamara, K.J. and Kendrick, G.W. (1994). Cenozoic Molluscs and Echinoids of Barrow Island, Western Australia. Records of the Western Australian Museum, Supplement 51, 50p. McNamara, K.J., Philip, G.M. and Kruse, P.D. (1986). Tertiary brissid echinoids from southern Australia. Alcheringa 10, 55-84. Melville, R.V. and Durham, J.W. (1966). Skeletal Morphology. In ‘Treatise on Invertebrate Paleontology, Part U, Echinodermata 3’ (Ed. R.C. Moore) pp. U220-U257. (Geological Society of America and University of Kansas Press: Lawrence). Montgomery, J.N. (1930). Part V-A contribution to the Tertiary geology of Papua. In ‘The oil exploration work in Papua and New Guinea, conducted by the Anglo-Persian Oil Company on behalf of the Government of the Commonwealth of Australia, 1920-1929, Volume 4’. pp. 3-85. (Harrison and Sons: London). Mooi, R. (1990a). Paedomorphosis, Aristotle’s lantern, and the origin of the sand dollars (Echinodermata: Clypeasteroida). Paleobiology 16, 25-48. Mooi, R. (1990b). Progenetic miniaturization in the sand dollar Sinaechinocyamus: implications for clypeasteroid phylogeny. In ‘Echinoderm Research’ (Eds C. DeRidder, P. Dubois, M.-C. Lahaye and M. Jangoux) pp. 137-143. (A.A. Balkema Publishers: Rotterdam). Mortensen, T. (1943). A Monograph of the Echinoidea 3(2), Camarodonta I. C.A. Reitzel, Copenhagen. 553p. Mortensen, T. (1948). A Monograph of the Echinoidea 4(2), Clypeastroida. C.A. Reitzel, Copenhagen. 471p. Mortensen, T. (1950). A Monograph of the Echinoidea 5(1), Spatangoida I. C.A. Reitzel, Copenhagen. 432p. Mortensen, T. (1951). A Monograph of the Echinoidea 5(2), Spatangoida II. C.A. Reitzel, Copenhagen. 593p. Nichols, D. (1959). Changes in the chalk heart-urchin Micraster interpreted in relation to living forms. Philosophical Transactions of the Royal Society of London, Series B, 242, 347-437. Philip, G.M. (1969). The Tertiary echinoids of south-eastern Australia IV. Camarodonta (2). Proceedings of the Royal Society of Victoria 82, 233-275. Pieters, PE. (1978). Port Moresby-Kalo-Aroa, 1:250,000 Geological Series - Explanatory Notes. (Bureau of Mineral Resources, Geology and Geophysics and Department of Lands, Surveys and Mines: Canberra and Port Moresby). 55p. Seilacher, A. (1979). Constructional morphology of sand dollars. Paleobiology 5, 191-221. Smit, JA.J., Dow, D.B., Pitt, R.P.B. and Pillinger, D.M. (1974). Wau Sheet SB55-14, Papua New Guinea 1:250,000 Geological Series. (Bureau of Mineral Resources, Geology and Geophysics and Department of Lands, Surveys and Mines: Canberra and Port Moresby). Proc. Linn. Soc. N.S.W., 123. 2001 1.D. LINDLEY 139 Smith, A. (1984). ‘Echinoid Palaeobiology’. (George Allen and Unwin: London). Smith, I.E.M. and Davies, H.L. (1972). Samarai Sheet SC56-9, Papua New Guinea 1:250,000 Geological Series. (Bureau of Mineral Resources, Geology and Geophysics and Department of Lands, Surveys and Mines: Canberra and Port Moresby). Smith, I.E.M. and Davies, H.L. (1973). Samari, 1:250,000 Geological Series - Explanatory Notes. (Bureau of Mineral Resources, Geology and Geophysics and Department of Lands, Surveys and Mines: Canberra and Port Moresby). 21p. Tate, R. (1891). A bibliography and revised list of the described echinoids of the Australian Eocene, with descriptions of some new species. Zransactions and Proceedings of the Royal Society of South Australia 14, 270-282. Tate, R. (1894). Note on the Tertiary fossils from Hall Sound, New Guinea. Proceedings of the Linnean Society of New South Wales 9, 213-214. Tenison-Woods, J.E. (1878). On a Tertiary Formation at New Guinea. Proceedings of the Linnean Society of New South Wales 2, 125-128. Wang, C.C. (1984). New classification of clypeasteroid echinoids. Proceedings of the Geological Society of China 27, 119-152. APPENDIX Collection details Aseki locality: Collection site is in a tributary of the Kapau River, immediately east of the airfield and Lutheran Mission at Aseki village, Morobe Province, PNG. Grid Reference 111878 Aseki 1:100,000 Sheet 8183. Site is upstream of the track from Aseki to Window1. Sagari valley locality: Collection site is in the vicinity of Ata Ata village on the northern side of the Sagari River valley, southwest of Alotau, Milne Bay Province, PNG. Site is in the lower reaches of Wabalam Creek, about 540 stream metres downstream from its junction with Petula Creek. The reader is referred to map labelled Dwg. 958/5/2 in Lindley (1991) for additional details (collection site at survey station 813 on map). Yule Island locality: Collection site is on the southeastern coastline of Yule Island, Central Province, PNG. Grid Reference 493258 Kubuna 1:100,000 Sheet 8280. Proc. Linn. Soc. N.S.W., 123. 2001 a | CNTR ea Pt a PFEAN). Roosters Ohl tee: Ye ieneet ttenalt yrelatatyest - apse t engi) Bas Papo Rioag, ater Snare tendiie: Contd Prostate, Haya New, Cataon ron pres} ihe Sciiagtbl drteasehéctne t bitictrn fidiwe suse lithestemen endl tn Tabiiaitien Hdammesd Pads oltead hob thus Sine 9d etide sf Ataod, VG (Peiciae ezasih ‘edt Cap hilivascaasW bn, saleweseAl Agoige wer rasan st Pon hierseta rte ae ; WesasiS, 627A uals: iio ey nn: Gait Foe sl ipeo eee 19: A253 J BIS-EES Pralatl "Racst erste asront: f 8 rates Figure Saanllipstsas 0 secotadanit re ee mart . S261) TS eats 7 piaqites Broun, Hien ~ arciadishe worn 129, =! tsk, e Pensa se fs vey y Mey ‘iil Hox cn 89 want icity Alen Pha. rT PE) & , onthe, Je nai? Fit : ‘Gel & Bie Y ane any a: Thiéey ri axes s dee ena de | is soe GHD eel Aras Shanes SER ON i matt daw as Kaeaienors Wis sheuciass 9000, SG i i tl aie bid Mid i, ery eat wi ‘og . iD a ain walls 1S ripest WO , rene inal isi: 16, Mfc sees nt. E ies goterbi 7? vali Boiscila)! sliprsbddte spizelal 2 Seg on toes Hi Veith S NI Ot: vont net! She yeiomtey o2 seadihen 00, ow ee itt Ave thee austin Veieifle. ? a ‘ Lit ei Ss a Winotonay fo Teas oe Uertelotrels Faleunnth fi ¥ ae . ne See nat * ae \ ‘ aly ree ‘ B is , RILEY, ie } ior uti io ee THe are a PANU 5S, ap | T he on 7 wi MN ie it oh : be hte ‘Esl ‘ sy ALG) Ma)" rey ive hela oie if ehusitaliy HRR- LOND. oping 4 8S CaN t os Who’, Re “aw Peeaerce Ms, Pee, MONET mragh. 1s gen Ge Be ial aletiprs ; Z i f ‘ : j 150) Ls H j , ry HT I a x tes VE FG yang Ne wictenysd i nha? iettiedd 4 “sy ‘ j : ) Loi hy hme abet wi}, My (ener, | wi ’ PchaLis e _ r lw Vee i wMunicnees Ay OA A tioctirentis a the Bhi meonben S75). Saatereae a due firebneny PajOsti/ Mowe nity OF ie kal arden, Sts, Soe haveniaes: f ro itepwel Capen Minot O) (18S), Changes ochalk beattverthin: MWieniney tl eth > id velation (p HY a VA¢cagey Higa Trade iliiag a FT the Avewal Sedicvt) ial Le brag Ne V fa i, 7A e jA7sR AT a ; Pale, OM At06R, TRE Tettine’y eulinduls efeet) coon? Auihin Cama don ar ibe ; AY Anwel Soclen i 4 3.97 { Petey. PY. Ca) Pelt Mayeeaty Male Arn 24(Liso enlowirar Sera: beplanoiory Nate rv Oriel benmmicmk | hohkiey at {le Mer Bar ak ran Luts areas ai a% Cavteers aif Pint Meiuiiyy, SEs . a, anne . Rpilessver, J 4% | Lage Myron (ere shoe’ nae Galilee Dota hilo & hy ae My Oa halt LAD: Pee DA) Bia, BR ap Peg es 119 F) Phe Sheet sues Ah Peal Newt i:250,080 Genie pial Sorite, (orem of Mippera Picardie (Ca, Coubemy duit Aepihiyy orn tLonds, Survey ed Mini 4 Aniweri's fad Yad Monty) Ie vers Loeb. Sweet foie, abadl) Dear A Review of the Genus Kobonga Distant with the Description of a New Species (Hemiptera: Cicadidae) M.S. Mou.ps! AND K.A. KOPESTONSKY? ‘Australian Museum, 6 College St, Sydney NSW 2010, Australia email: maxm @austmus.gov.au *Hiram College, Hiram Ohio 44234, U.S.A. Moulds, M.S. and Kopestonsky, K.A. (2001). A review of the genus Kobonga Distant with the description of a new species (Hemiptera: Cicadidae). Proceedings of the Linnean Society of New South Wales 123, 141-157. Distinguishing characters for the genus Kobonga (Hemiptera: Cicadidae) are listed. Kobonga apicans sp.n. is described from the Northern Territory and Kobonga oxleyi (Distant) and Kobonga fuscomarginata (Distant) are new combinations. A key to the six species is provided and known distributions for all species are extended and recorded in detail. Manuscript received 4 October 2001, accepted for publication 24 October 2001. KEYWORDS: Cicada, Cicadidae, Kobonga. INTRODUCTION Distant (1906a) erected the genus Kobonga to accommodate Cicada umbrimargo Walker and later in the same year Melampsalta godingi Distant was transferred to Kobonga (Distant 1906b). A third species, K. froggatti Distant, was described in 1913. Kobonga clara Distant has since been removed from Kobonga as a junior synonym of Cicadetta spinosa (Goding and Froggatt) (Moulds 1990). Colour figures of K. froggatti and K. umbrimargo were provided by Moulds (1990) in addition to brief diagnostic descriptions and notes on the distributions and biology of these two species. However, K. godingi was not discussed by Moulds and nothing has been published on this species beyond its original description. In this paper we confirm the identity of K. godingi, recognise Cicadetta oxleyi and Pauropsalta fuscomarginata as belonging to Kobonga and describe a new species that is closely allied to K. froggatti. Kobonga is redefined and we re-describe the five previously named species and include the first study of male genitalic structures in Kobonga. The following abbreviations are used below: AM, Australian Museum, Sydney; ANIC, Australian National Insect Collection, Canberra; BMNH, The Natural History Museum, London; JTM, collection of J.T. Moss; MV, Museum of Victoria, Melbourne; MSM, senior author’s collection; SAM, South Australian Museum, Adelaide. For each species, measurements (in mm), are given as arange and mean, and include the smallest and largest of available specimens. Proc. Linn. Soc. N.S.W., 123. 2001 142 THE GENUS KOBONGA (CICADIDAE) Family CICADIDAE Subfamily Tibicininae Tribe Cicadettini Genus Kobonga Distant, 1906 Kobonga Distant, 1906a: 387; Distant, 1906b: 163, 177; Ashton, 1912: 27; Ashton, 1914: 351; Schulze, Kiikenthal and Heider, 1926-40: 1731; Kato, 1932: 187; Neave, 1939: 831; Metcalf, 1944: 155; Metcalf, 1947: 163; Kato, 1956: 70,79; Burns, 1957: 666; Metcalf, 1963: 271; Dugdale, 1972: 877, 878, 880; Duffels and van der Laan, 1985: 299; Moulds, 1990: 129. Type species Cicada umbrimargo Walker, 1858, by original designation. Included species apicans sp.n., froggatti Distant, 1913; fuscomarginata (Distant, 1914), comb.nov.; godingi (Distant, 1905); oxleyi (Distant, 1882), comb.nov.; umbrimargo (Walker, 1858). Distribution Southern third of Western Australia, the central region of Northern Territory, much of South Australia except the north-east quarter, western Victoria, western NSW and eastern half of Queensland south of Townsville. Diagnosis A full diagnosis will be provided as part of a generic review of Australian cicadas by the senior author (in prep.). Distinguishing characters The fore wing ambient vein is continuously infuscated along apical cells 3-6 and often beyond, in addition to infuscations at the bases of apical cells 2 and 3, at the extremities of the longitudinal veins and in some species also elsewhere. The fore wing apical cells are much shorter than the ulnar cells in all species except K. froggatti. Tymbals possess five ribs, the first three long and dominant, the shorter fourth rib often broken and the fifth very short. The male genitalia have uncal lateral lobes that are flat with an out-turned rim along upper margin to apex, this rim in ventral view distinctly stepped inwards apically. The pygofer basal lobes in K. umbrimargo, K. froggatti and K. apicans possess a characteristic, broad, rounded, lateral swelling, low in profile and development mostly on inner surface. The aedeagus is trifid with a fleshy endotheca and a well developed basal hinge. Key to Species of Kobonga ji Hind wing with an infuscated border following ambient vein ...............: eee 2 - Hind wine border withoutintuscationer. 1-2 eee ere eee 3) ep Head at back of eyes yellowish or light brown; dorsal area immediately anterior of pronotal collar bearing a pair of dull yellow fascia that meet, or nearly meet, on LU iETO NTT coptveh edie btn dott ah Sree bie pak hard tety Alrel obs ate at dere hee ence apicans sp.n. - Head at back of eyes black; dorsal area immediately anterior of pronotal collar CHEERY JEL LACK eas Tepe eter tree cece eee nee eres froggatti Distant Proc. Linn. Soc. N.S.W., 123. 2001 M.S. MOULDS AND K.A. KOPESTONSKY 143 3: Length of fore wing greater than 29mm; distal end of fore wing radial cell not beyoudimiddensthivoliwang ie wets c ect eese ee eee umbrimargo (Walker) - Length of fore wing less than 27mm; distal end of fore wing radial cell beyond MIG Gle Me thro fewest Leas MN ES SCL Sa ie 2h are eee Ei RY PL 4 4. Fore wing crossvein m not infuscated; five apical Cellilishunnvhnaitvelgwyairn Gee esac oes teeeee es abeceeeees.ceetcets nceeeneees fuscomarginata (Distant) - Fore wing crossvein m partly or entirely infuscated; six apical Cellist waaay swan eee ATE ae Te STS EO Sy TE A Da Soros ye an 5 5u Infuscation on fore wing along margin broad, usually (but not always) of near even width, its width always consuming infuscation around APLC ALCS Te ee a EE AS SEA ERE IT OS, oxleyi (Distant) - Infuscation on fore wing along margin scalloped, never of near even width, the infuscation on apical cell 7 extending inwards far beyond remainder of infuscation BONS WASP aN SUT See a ER Ve Lae ceees godingi (Distant) Kobonga umbrimargo (Walker) (Figs 1, 2, 11) Cicada umbrimargo Walker, 1858: 32; Distant, 1905: 270. Melampsalta umbrimargo (Walker): Stal, 1862: 484; Distant, 1892: 67; Goding and Froggatt, 1904: 632. Kobonga umbrimargo (Walker): Distant, 1906a: 388; Distant, 1906b: 177; Kirkaldy, 1907: 309; Ashton, 1914: 351; Kato, 1932: 187; Burns, 1957: 666-667; Metcalf, 1963: 272-273; Duffels and van der Laan, 1985: 300; Ewart, 1989a: 80; Moulds, 1990: 129-130, pl. 21, figs 7a,7b. Kobonga umbrimago (Walker) [sic]: Ashton, 1912: 27. Misspelling. Type Syntype female labelled as follows:(1) Swan/R and [18]53/50 on reverse (handwritten india ink on circular label); (2) Type (circular, machine printed BMNH label with green ‘Walker’ border); (3) Melampsalta (handwritten india ink in unknown hand); (4) C. umbrimargo/Walker (underlined and handwritten india ink on thicker than usual card in unknown hand); (5) BMNH (E)/# 651009 (BMNH specimen database number). Examined, in BMNH. Walker (1858) described only the female but did not designate a type nor state the number of specimens he had. The single specimen known to have been a Walker type is thus regarded as a syntype. Material examined WESTERN AUSTRALIA: 1 male, Arthur River, 6 mi. E Darkan, 28.1.1971, G.A. Holloway and H. Hughes; | female, Walyunga National Park, 35 mi. NW Perth, 7.1.1971, G.A. Holloway and H. Hughes; 1 female, King George’s Sound (no other data); 1 female, Perth, G.H. Hardy, No 933, K43348. (AM). 2 males, 1 female, Geraldton, 1914, Clark; 1 female, Perth, 15.xii.[19]05, Lowe, ex W.W. Froggatt Collection; 1 male, 1 female, Perth, 1917, Clark, ex W.W. Froggatt Collection; 1 male, 2 females, Cape Riche, 1,9,1.1940, K.R. Norris (ANIC). 6 males (1 genitalia prep. KO4), 3 females, Kalbarrie, 21.xi.1978, M.S. and B.J. Moulds; 1 female, Northampton, 25.1.1974, D. and N. McFarland; 1 male, Howatharra Nature Reserve, Moresby Rg, 30km NNE of Geraldton, 28:33S 114:40E, 7.xii.1973, N. McFarland; 5 males, 1 female, 8km E of Dongara, 24.xi.1978, M.S. and B.J. Moulds; 1 male, Perth, 3.1.1961, M.S. Moulds; 3 males, 1 female, Mandurah, 1.1991, M.S. and B.J. Moulds; | female, Pinjarra, 1.1990, A. Johnson; Proc. Linn. Soc. N.S.W., 123. 2001 144 THE GENUS KOBONGA (CICADIDAE) | male, Corrigin, 3.x11.1985, M.S. and B.J. Moulds; 3 males, Narrogin, 4.x11.1985, M.S. and B.J. Moulds; 2 males, 1 female, Oldfield River at road crossing, E of Ravensthorpe, 4.xii1.1978, M.S. and B.J. Moulds; 2 males, Jerramungup, 21.x.1977, 6.xi.1983, K. and E. Carnaby; | male (genitalia prep. KO3), Tone Bridge [Tone River], 23.1.1975, K. and E. Carnaby; | male, Pallinup River x-ing, 30km W of Gnowangerup, 9.xi1.1985, M.S. and B.J. Moulds; 1 female, Lancelin, 7.1.1979, K. and E. Carnaby; 1 male, South Tammin, 7.1.1979, T.M.S. Hanlon; | male, Yellowdine, 28.1.1979, T.M.S. Hanlon; | male, Mount Ragged, 17.xi1.1995, M.S. and B.J. Moulds and K.A. Kopestonsky; 2 females, near Queen Victoria Rock, 31:17S 120:56E, 23.xii.1995, M.S. and B.J. Moulds and K.A. Kopestonsky; 4 males, | female, Lake Douglas, 12km SW of Kalgoorlie, 13.i1.1989, 20.1.1991, M.S. and B.J. Moulds; 1 male, Zanthus, Trans Aust. Railway, 22.1.1991, M.S. and B.J. Moulds; 1 female, 55km ESE of Kimba, on Kimba/Iron Knob road, 12.xii.1995, M.S. and B.J. Moulds and K.A. Kopestonsky (MSM). 1 male, Cue, H.W. Brown; | female, Geraldton, x1.1920, J.W. Mellor [Mellor coll. via Capt. S.A. White coll.]; 1 female, Kuminim, E.F. du Boulay (SAM). 1 female, West Midland, 27.xii.[19]49, I.M. [coll. A.N. Burns]; 1 male, Camboon Park, 22.1.[19]51, I.M. [coll. A.N. Burns]; 1 female, Toodyay, x1i.[19]51, R. McMillan; 1 female, Kenwick, 28.1.1960, McTurton [coll. A.N. Burns]; 1 male, Glen Forrest, 15.i.[19]50, I.M. [coll. A.N. Burns] (VM). SOUTH AUSTRALIA: | male, Wilpena Figures 1-6. Kobonga spp., male genitalia, lateral view: (1) umbrimargo, Kalbarrie, WA, lateral view; (2) same, ventral view; (3) apicans, Witchety Bore, NT, lateral view; (4) froggatti, Agnew, WA, lateral view; (5) godingi, Inglewood, Vic, lateral view; (6) oxleyi, Barcaldine, Qld, lateral view. Proc. Linn. Soc. N.S.W., 123. 2001 M.S. MOULDS AND K.A. KOPESTONSKY 145 Pound, Flinders Ranges,. 19.1.1976, M.S. and B.J. Moulds; | male, Alligator Gorge Nat. Pk. near Wilmington, 17.1.1976, M.S. and B.J. Moulds (MSM). 4 females, Musgrave Rg., 25 mile Bore, 9,12.11.1966, P. Aitken and N.B. Tindale; 1 female, Orroroo, 9.1.[18]91, Bimbrick; 2 males, 1 female, Calpalanna W.H. Cons Pk. Wedina Well, Eyre Pen. 30.xi.1986, J.A. Forrest; 1 female, Eyria [sic] Pen., v.1897, E.M. Thring [ex Capt. S.A. White coll.]; 1 female, Ooldea, 1897, R.T. Maurice, [Wimecke coll.]; 1 female, Willowie East, xii.1984, A. McIntyre, Mt Burr Ecological Survey Group (SAM). | male, Orroroo, 1.1.[19]42, J.T. Gray [coll. ALN. Burns] (VM). VICTORIA: 1 male (genitalia prep. KO2), 1 female, Little Desert, near Kiata, 11.1.1976, M.S. and B.J. Moulds (MSM). 1 male, Nyah, 10.xi.[19]09, McLeach; | male, Kiata, 20.xi1.[19]46, C. McCubbin (VM). NEW SOUTH WALES: 2 females, Round Hill, near Euabalong, 1.1970, Jim Rich (AM). 1 female, 3km S of Matakana, 12.xii.1982, T.M.S. Hanlon (MSM). QUEENSLAND: 1 female, Clermont, Dr. K.K. Spence (AM). 1 female, 24km N of Miles, 21.1.1991, A. Sundholm; 1 male, 2 females, nr Alpha, 23:37:32S, 146:38:21E, 30.xi1.2000, M.S. and B.J. Moulds (MSM).Rg., 25 mile Bore, 9,12.11.1966, P. Aitken and N.B. Tindale; 1 female, Orroroo, 9.1.[18]91, Bimbrick; 2 males, 1 female, Calpalanna W.H. Cons Pk. Wedina Well, Eyre Pen. 30.xi.1986, J.A. Forrest; 1 female, Eyria [sic] Pen., v.1897, E.M. Thring [ex Capt. S.A. White coll.]; 1 female, Ooldea, 1897, R.T. Maurice, [Wimecke coll.]; 1 female, Willowie East, xi1.1984, A. McIntyre, Mt Burr Ecological Survey Group (SAM). | male, Orroroo, 1.1.[19]42, J.T. Gray [coll. A.N. Burns] (VM). VICTORIA: | male (genitalia prep. KO2), 1 female, Little Desert, near Kiata, 11.1.1976, M.S. and B.J. Moulds (MSM). 1 male, Nyah, 10.xi.[19]09, McLeach; 1 male, Kiata, 20.x11.[19]46, C. McCubbin (VM). NEW SOUTH WALES: 2 females, Round Hill, near Euabalong, 1.1970, Jim Rich (AM). 1 female, 3km S of Matakana, 12.xi1.1982,T.M.S. Hanlon (MSM). QUEENSLAND: | female, Clermont, Dr. K.K. Spence (AM). 1 female, 24km N of Miles, 21.1.1991, A. Sundholm; 1 male, 2 females, nr Alpha, 23:37:32S, 146:38:21E, 30.xi1.2000, M.S. and B.J. Moulds (MSM). Redescription MALE (figured by Moulds 1990, PI.21, fig.7) Head.- Black and orange brown; orange brown along all or much of dorsal midline, expanded at posterior end; an orange brown blotch adjacent to each eye never reaching anterior of head; orange brown at back of eyes to varying extent; orange brown along most anterior margin of antennal plate; below orange brown in part adjacent to eyes and anteriorly on lorum which, on some individuals dominates the black. Postclypeus black and orange brown, the orange brown forming a border ventrally but sometimes with black extending into the orange brown along transverse ridges; also dominantly orange brown dorsally and at most anterior point. Anteclypeus black and orange brown to variable degree, usually black with median or paramedian orange brown fascia on about basal three quarters. Rostrum orange brown tending black apically, reaching apices of mid coxae. Antennae with basal segment orange brown, otherwise usually black or nearly so. Thorax.- Pronotum orange brown with black markings; black fascia either side of dorsal midline, greatly expanded laterally at anterior end and to a far lesser degree at posterior end a little before pronotal collar; usually black irregular border to lateral plates, and sometimes broken irregular black markings in or near sutures; pronotal collar orange brown, dorsally usually with black anterior border to near lateral angles and usually a black external border anterior of lateral angles. Mesonotum black and orange brown or pinkish brown, the black often dominant and forming the following markings: a dorsal pair of obconical fascia based on anterior margin and meeting on midline, a paramedian pair of obconical fascia much longer than dorsal pair and reaching to or nearly to anterior arms of cruciform elevation, a marking occupying area between anterior arms of cruciform elevation that tapers forward usually to a sharp point on midline and sometimes reaches between the pair of dorsal obconical markings and fuses with them; cruciform elevation orange brown. Proc. Linn. Soc. N.S.W., 123. 2001 146 THE GENUS KOBONGA (CICADIDAE) Figures 7-10. Kobonga species, adults, dorsal view: (7) K. godingi, male; (8) K. oxleyi, male; (9) K. fuscomarginata, female; (10) K. apicans, male. Wings.- Hyaline. Fore wing with distal end of radial cell not reaching beyond mid length of wing; bold infuscation overlaying basal veins of apical cells 2 and 3, variable in extent but always prominent; bold infuscation along ambient vein from apex to vein CuA,, this infuscation extending part way along each adjoining vein but following entire perimeter of cell 7; rarely infuscated at base of apical cell 5; usually tinted brown, often weakly, at distal ends of apical cells between vein infuscations; basal cell weakly tinted amber; basal membrane usually crimson, sometimes tending pale orange; costa orange brown to reddish brown. Hind wing often without infuscation but sometimes lightly infuscated along vein 3A with some marginal diffusion at distal end; plaga crimson or sometimes tending pale orange or yellowish. Legs.- Light to dark brown, considerably variable between individuals, often with areas of black and usually with a blurred pair of black fascia along much of length of mid and hind femora. Opercula.- Muddy pale yellow, sometimes a little black at base; outer margin fringed with long silver cilia. Meracantha broad, near an equilateral triangle, muddy yellow with black at base. Abdomen.- Tergites orange brown to pinkish brown with a broad black anterior margin to each segment variable in extent between individuals, sometimes reaching to lateral extremity and always covering at least half of tergite 8; tergites 2-7 with a black irregular patch of variable extent near lateral extremities close to posterior margin, this patch on some individuals merged with the black following anterior margin; ventral surface of tergites 3-7 yellowish brown to reddish brown with narrow muddy yellow posterior margin. Sternites I and II often substantially black with muddy yellow Proc. Linn. Soc. N.S.W., 123. 2001 M.S. MOULDS AND K.A. KOPESTONSKY 147 or brown posterior margin; sternites III-VII muddy yellow to yellowish brown each with a black anterior blotch on midline tending reddish at margins, these forming a broad, broken central fascia; sternite VIII muddy yellow to yellow brown, sometimes with black basally. Genitalia (Figs 1, 2) Pygofer upper lobe in lateral view broad basally, tapering to rounded apex, variable in size and proportions between individuals; basal lobe with apex rounded and a lateral, low, rounded swelling developed to varying degrees, also variable between individuals. Uncal lateral lobes with edging rim terminating a little above level of ventral margin of lobe. Aedeagus with endotheca long, from its visible base extending three quarters or more to apex of pseudoparameres. FEMALE (figured by Moulds 1990, P1.21, fig.7a) Similar to male. Abdominal segment 9 in dorsal view a little longer than an equilateral triangle; black and orange brown; a pair of paramedian black fascia that sometimes continue to dorsal spine; suffused black laterally on basal half to varying degrees; apical spine at least black at apex, sometimes entirely black; a black spot mid laterally on distal half. Ovipositor sheath terminating near level with apical spine. Figures 11-12. Distribution of Kobonga species: (11) umbrimargo and fuscomarginata; (12) froggatti, apicans, oxleyi and godingi. e K. umbrimargo @ K. fuscomarginata Bit Se taal seg le to e K. froggatti @ K. apicans aK. oxleyi @K. godingi POS att Proc. Linn. Soc. N.S.W., 123. 2001 148 THE GENUS KOBONGA (CICADIDAE) MEASUREMENTS n= 10 males, 10 females (includes largest and smallest of available specimens). Length of body: male 23.1 - 26.5 (24.9); female 25.7 - 29.1 (27.3). Length of fore wing: male 29.3 - 34.8 (32.2); female 33.8 - 38.6 (36.0). Width of head: male 7.1 - 8.3 (7.8); female 8.0 - 9.3 (8.5). Width of pronotum: male 7.9 - 9.0 (8.7); female 8.4 - 10.3 (9.4). Distinguishing characters Most similar to K. godingi from which specimens can be separated by having the distal end of the fore wing radial cell not reaching to mid length of wing; that of godingi extends clearly beyond mid length. In addition umbrimargo is a larger insect with a fore wing length greater than 29 mm; that of godingi never reaches 29mm. Distribution (Fig. 11) Throughout much of the southern half of mainland Australia, primarily inland in areas receiving less than 750 mm annual rainfall. It appears to be most common in Western Australia where specimens have been taken in many localities south of the Murchison River and east to Zanthus. It is widespread in South Australia except for the north-eastern quarter but is the most common to the east and west of Spencer Gulf. Records from the eastern States are sparse; there are records only from north-western Victoria, near Euabalong in central New South Wales and from Clermont, Alpha and Miles in Queensland. Adults have been taken from late October to early March. Specimens recorded as this species by Goding and Froggatt (1904) from Ardrossan and Gawler are K. godingi (see comments under that species). Habitat Myrtaceous trees, especially Eucalyptus spp. where adults mostly frequent the upper trunk and branches. Song Unknown. Kobonga godingi (Distant) (Figs 3, 7, 12) Melampsalta godingi Distant, 1905: 270. Kobonga godingi (Distant): Distant, 1906b: 177; Burns, 1957: 666; Metcalf, 1963: 272; Duffels and van der Laan, 1985: 300; Moulds, 1990: 129. Melampsalta umbrimargo (Walker): Goding and Froggatt, 1904: 632, pl. xviti, fig. 12 (partim, misident., localities Androssan and Gawler only). Note on synonymy Goding and Froggatt (1904) quoted a slightly edited version of Walker’s original description of umbrimargo but added two further locality records, Androssan and Gawler. However, their figure of a fore wing (P1.18, fig.12) suggests they may have misidentified the specimens representing these additional localities. Distant (1905) described Goding and Froggatt’s Androssan and Gawler specimens as a new species, godingi, and not as a new name for umbrimargo Goding and Froggatt as erroneously stated by Metcalf (1963); there is in fact no nominal species umbrimargo Goding and Froggatt as distinct from umbrimargo Walker. Type Male syntype labelled as follows: (1) Melampsalta/ Godingi/Dist/type (handwritten india ink possibly in Distant’s hand); (2) Type (circular, machine printed Proc. Linn. Soc. N.S.W., 123. 2001 M.S. MOULDS AND K.A. KOPESTONSKY 149 label with red border); (3) Distant Coll./1911-383 (machine printed); (4) No 4. (handwritten, india ink); (5) Melampsalta/umbrimargo/Walk/Queensland (handwritten, india ink, in an unknown hand but appearing not to be that of Distant, Froggatt or Masters). Examined, in BMNH. Distant (1905) apparently based his description on at least two males as he describes only the male and gives Androssan and Gawler as localities. Only the male above could be identified as belonging to the type series. The fifth label detailed above gives the locality as Queensland but Androssan and Gawler are both in South Australia. However, there is no reason to doubt the association of the first four labels with this specimen as it clearly fits Distant’s description; the fifth label stating Queensland is considered erroneously attached. Further, this species is unknown from Queensland. Material examined Type as above and the following: NEW SOUTH WALES: | female, W. Wyalong, 23.xi.1963, J.C. Le Souef (MSM). VICTORIA: | male, 5km SW of Inglewood, 29.x1i.1989, K.L. Dunn (MSM). SOUTH AUSTRALIA: 1 female, no locality apart from State, 1904, Mac. Dup., ex W.W. Froggatt Collection (ANIC). 7 males, 5 females, S. Australia (or S. Aust.) (1 male labelled umbrimomargo, sic) (MM). STATE UNKNOWN: 1 male, ex W.W. Froggatt Collection (ANIC). Redescription Similar to K. umbrimargo but differing as follows. MALE (Fig. 7) Head.-Usually black at back of eyes rather than orange brown. Thorax.-Prothorax normally with extensive irregular black blotches following sutures of lateral plates. Wings.-Fore wing with distal end of radial cell beyond mid length of wing; always with an infuscation at proximal end of apical cell 5; never weakly tinted brown at distal end of apical cells between vein infuscations. Abdomen.-Tergites usually dominantly black rather than usually half black and half orange brown. Genitalia (Fig. 3).-Similar to K. umbrimargo but with pygofer basal lobes lacking an obvious, low, rounded, lateral swelling; uncal lateral lobes with edging rim terminating at level of ventral margin of lobe; ventral support of aedeagus much longer, its length almost equal to that of endotheca. FEMALE Similar to male. Abdominal segment 9 recurved below apical spine, sometimes to such an extent as to project distally to level of spine apex. MEASUREMENTS n = 8 males, 7 females (includes all of available specimens). Length of body: male 18.1-20.6 (19.05); female 16.6-20.0 (18.53). Length of fore wing: male 21.4-24.6 (23.05); female 21.5-26.0 (23.60). Width of head: male 5.6-6.3 (5.89); female 5.5-6.3 (5.87). Width of pronotum: male 6.1-7.2 (6.56); female 6.0-7.2 (6.51). Distinguishing characters Most similar to K. umbrimargo from which it is best distinguished by the characters listed under that species. Proc. Linn. Soc. N.S.W., 123. 2001 150 THE GENUS KOBONGA (CICADIDAE) Distribution (Fig. 12) Known only from West Wyalong in southern inland New South Wales, Inglewood in western Victoria and from South Australia in the vicinity of Vincent Gulf at Gawler and Androssan. Specimens have been taken in November and December. Habitat Unknown. Song Unknown. Kobonga oxleyi (Distant), comb.n. (Figs 4, 8, 12) Melampsalta oxleyi Distant, 1882: 131; Goding and Froggatt, 1904: 655; Distant, 1906b: 174; Ashton, 1912: 26; Burns, 1957: 660. Cicadetta oxleyi (Distant): Weidner and Wagner, 1968: 149; Duffels and van der Laan, 1985: 290; Ewart, 1988: 182, pl.2, fig.D; Lithgow, 1988, 65; Ewart, 1989a: 80; Moulds, 1990: 147, pl.21, fig.1; Ewart, 1998: 60. Type Holotype female labelled as follows: (1) Peak-Downs./Mus. Godeffroy./No. 17618 (partly machine printed, partly handwritten); (2) M./oxleyi/Dist. (handwritten with lined border); (3) Z.M.H./Hamburg (machine printed with lined border); (4) No. 17618./Melampsalta/oxleyi/Dist./ P Downs (handwritten on a machine-printed label with a heavy black outer border and light inner border and stating Museum Godeffroy Hamburg); (5) Type (machine printed on red label). Examined, in Universitat Hamburg, Zoologisches Institut und Zoologisches Museum. Material examined Type as above and the following: QUEENSLAND: 1 male, ‘Glendon’, W of Mackay, 20-29.i1.[19]84, J.T. Moss; 1 male, Cunnamulla, J.T. Moss (JTM). 3 males, Reid River, 50km S of Townsville, 6.1x.1992, M.F. Braby; 1 male, 1 female, six miles N of Main Range Ck, approx 56m S of Sarina, brigalow scrub, Bruce Highway, 10.xi1.1973, A. and M. Walford-Huggins; 1 male, 2 females, Clermont, iv.1988, S. Lamond; 5 males, 2 females, Barcaldine, 10.11.1981, M.S. and B.J. Moulds; 2 males, 1 female, 5km N of Mourangee Hsd, nr Edungalba, 8.x1i.1984, 2.1.1988, E.E. Adams; | female, 3.5km N of Mourangee Hsd, nr Edungalba, 24.xii.1988, E.E. Adams; 1 male, 3km E of Mourangee Hsd, nr Edungalba, 3.1.1988, G. van Moolenbroek; | male, Edungalba, nr Duaringa, 22.1.1982, M.S. and B.J. Moulds; 2 males, | female, base, Mount Scoria, 6km S Thangool, 24:32S 150:36E, 11.11.1991, G. and A. Daniels, C. Burwell; 1 male, 56km S of Rolleston, 20.xii.1983, M.S. and B.J. Moulds; 1 female, 6km N of Taroom, 25:36S 149:46E, 1.x.1991, 200m, mv lamp, G. Daniels; 4 males, 35km SSE of Roma, 23.xi.1986, M.S. and B.J. Moulds; 1 female, Lake Broadwater Nat Park, Dalby, 25.x.1986, R. Eastwood (MSM). NEW SOUTH WALES: 1 male, ‘Woodbine’ 17km W of Goolgowi, 11.xi1.1992, M. Coombs (MSM). Redescription Similar to K. umbrimargo but differing as follows. Proc. Linn. Soc. N.S.W., 123. 2001 M.S. MOULDS AND K.A. KOPESTONSKY ily! MALE (Fig. 8; figured by Moulds 1990, P1.21, fig.1) Head.- Orange brown areas tending green on live specimens; head below with minimal black. Thorax.- Nearly always lacking the anteriorly-projecting black marking based between anterior arms of cruciform elevation but always possessing a small black spot near distal end of each anterior arm of cruciform elevation. Wings.- Fore wing with distal end of radial cell clearly beyond mid length of wing; marginal infuscation usually band-like and usually of an even dark intensity; often infuscated at base of apical cell 5; basal cell hyaline; costa green in life, tending yellowish brown on dried specimens. Hind wing plaga white or whitish. Abdomen.- Tergites with black dominating. Genitalia (Fig. 4).- Similar to K. godingi but basal lobe of pygofer broader in lateral view. FEMALE Similar to male. Abdominal segment 9 substantially orange brown, black markings confined to a small mid lateral spot on distal half and a narrow basal band that joins a pair of paramedian fascia that never reach the apical spine. MEASUREMENTS n= 10 males, 10 females (includes largest and smallest of available specimens). Length of body: male 16.7-20.1 (18.65); female 18.0-20.3 (19.43). Length of fore wing: male 20.3-24.9 (22.78); female 23.2-25.4 (24.22). Width of head: male 5.2-7.0 (5.8); female 5.4-6.4 (6.01). Width of pronotum: male 5.3-7.5 (5.96); female 5.5-6.9 (6.24). Distinguishing characters Most similar to K. fuscomarginata from which it differs in the much broader infuscation on the fore wing margin which always incorporates to some degree cross vein m; there are also 6 marginal cells to the hind wing instead of 5. It is possible that K. oxleyi and K. fuscomarginata are one and the same species (see comments under K. fuscomarginata). Distribution (Fig. 12) The eastern half of Queensland south from near Townsville in areas receiving less than 800 mm annual rainfall, and from inland southern New South Wales near Goolgowi. Adults occur from September (Ewart 1998) to April. Habitat Open forest, including brigalow and casuarina woodlands, where adults inhabit foliage (Ewart 1988, 1998). Song Ewart (1988, 1998) provide oscillograms of the calling song. He describes the call as particularly distinctive, consisting of alternating ‘chirping’ and ‘clanging’ phrases peaking at 7-10 kHz. Kobonga fuscomarginata (Distant), comb.n. (Figs 9, 11) Pauropsalta fuscomarginatus Distant, 1914: 63-64; Ewart, 1989b: 293. Melampsalta fuscomarginata (Distant): Burns, 1957 653. Pauropsalta fuscomarginata (Distant): ?7Dugdale, 1972: 877 (uncertain identification); Duffels and van der Laan, 1985: 301; Moulds, 1990: 131. Proc. Linn. Soc. N.S.W., 123. 2001 152 THE GENUS KOBONGA (CICADIDAE) Types Syntype male and female labelled as follows: (1) Yarrawin/N.S.W/20.11.13/ W.W.F. (handwritten india ink); (2) 1914.122 (handwritten india ink referring to BMNH register entry); (3) 84 (handwritten india ink); (4) Pauropsalta/fuscomarginata/type Dist. (handwritten india ink possibly in Distant’s hand); (5) (syntype male only) Type/H.T (circular, machine printed BMNH label with red border). The male is also labelled BMNH (E)/# 651019 (BMNH specimen database number) and the female BMNH (E)/# 651020. Examined, in BMNH. Distant (1914) based his description on at least two specimens as he gives the BMNH and AM as depositories. The latter could not be found and is presumed destroyed. It is interesting that the female is labelled as a type, apparently in Distant’s hand, as the original description gives male measurements only and there is no specific mention of a female. Material examined Type as above and the following: QUEENSLAND: | male, Cunnamulla, x.1941, N. Geary (AM). NEW SOUTH WALES: 1 female, 30km S of Lightning Ridge, 27.xii.1988, M.S. and B.J. Moulds (MSM). VICTORIA: 1 female, Grampians, 1905, Hill, ex W.W. Froggatt Collection (ANIC). Redescription Similar to K. oxleyi but differing as follows. MALE Wings.- Fore wing marginal infuscation narrow, never incorporating crossvein m. Hind wing with 5 apical cells. FEMALE (Fig. 9) Similar to male. Abdominal segment 9 similar to that of K. oxleyi. MEASUREMENTS n= 1 male, 2 females (includes all of available specimens). Length of body: male 16.8; female 17.1-17.4 (17.25). Length of fore wing: male 20.2; female 21.3-22.0 (21.65). Width of head: male 5.3; female 5.6-5.9 (5.75). Width of pronotum: male 5.3; female 5.9-6.0 (5.95). Distinguishing characters Very similar to K. oxleyi but the extent of the fore wing infuscation is far less. The five apical cells to the hind wing of fuscomarginata appear to distinguish it from K. godingi which has six apical cells. It is possible that specimens of fuscomarginata are aberrant individuals of oxleyi but a lack of material makes a conclusive decision difficult. Specimens of K. umbrimargo show considerable differences in the extent of the fore wing marginal infuscation, the extremes almost equalling the difference shown between fuscomarginata and oxleyi. Reduction in the number of hind wing cells is also not uncommon in the Cicadettini, also suggesting the possibility that the 5 hind wing cells of the four known specimens of fuscomarginata may be aberrations. Distribution (Fig. 11) Known only from Cunamulla in far south-western Queensland, near Lightning Ridge and Yarrawin (type locality) in western New South Wales and from the Grampians in the central west of Victoria. Distant (1914) gives the type locality as ‘504 miles west from Sydney but the male syntype is labelled ‘Yarrawin’. There is a pastoral property by this name some 50 Proc. Linn. Soc. N.S.W., 123. 2001 M.S. MOULDS AND K.A. KOPESTONSKY 153 km SE of Brewarrina and this is presumably the type locality. There are records for October and December only. Habitat Unknown. Song Unknown. Kobonga froggatti Distant (Figs 5, 12) Kobonga froggatti Distant, 1913: 490; Ashton, 1915: 91; Distant, 1915: 53; Burns, 1957: 666; Metcalf, 1963: 272; Duffels and van der Laan, 1985: 300; Moulds, 1990: 129, pl. 21, fig.5. Kobonga castanea Ashton, 1914: 351, pl.7, figs 5,5a; Ashton, 1915: 91; Distant, 1915: 53. Types Kobonga froggatti. Female syntype labelled as follows: (1) Australia/W.W. Froggatt/1913-364 (machine printed); (2) Cue, W.A./H.W. Brown (machine printed); (3) 71.W.W.F./1913 (handwritten india ink); (4) Kobonga/frogatti/type Dist. (handwritten india ink possibly in Distant’s hand); (5) Type (circular, machine printed BMNH label with red border); (6) BMNH (E)/# 651011 (BMNH specimen database number). Examined, in BMNH. Distant (1913) did not state the number of females upon which he based his description. Only the specimen above has been identified as belonging to the type series. Kobonga castanea. Male syntype, labelled (1)’Cue, W.A./H.W. Brown (machine printed); (2) Ashton coll./K. castanea co-type’. Examined, in AM. Ashton (1914) did not state the number of males upon which he based his description. Only the specimen above has been identified as belonging to the type series. A female labelled “Type’ in SAM is not considered to be part of the syntype series as the description appears to be based on males only. Material examined WESTERN AUSTRALIA: | male, 1 female, Cue, H. Brown, No. 71, ex W.W. Froggatt Collection (ANIC). 2 males, 60km W of Sandstone, 18.1.1989, M.S. and B.J. Moulds; 5 males, 1 female, 25km E of Sandstone, 17.1.1898, M.S. and B.J. Moulds; 28 males (3 genitalia preps KO1, KO10, KO11), 24 females, 17.1.1989, Agnew, 27:59S 120:41E, M.S. and B.J. Moulds; | male, 65km SE of Leinster, 28:20S 121:05E, 16.1.1989; M.S. and B.J. Moulds; 2 males, 18km N of Leonora, 16.1.1989, M.S. and B.J. Moulds; 3 males, 2 females, Leonora, 15.1.1989, M.S. and B.J. Moulds (MSM). 3 females, Cue, H.W. Brown (one labelled as Type, see above) (SAM). SOUTH AUSTRALIA: 6 males, 7 female, Stuart Hwy, 16km N of Tarcoonyinna Ck x-ing, 5.11.1984, M.S. and B.J. Moulds; 1 female, 65km S of Wintinna Hsd,. 5.11.1984, M.S. and B.J. Moulds; 1 male, 30km S of Mt Willoughby Stn, 6 Feb. 1984, M.S. and B.J. Moulds; 1 female, Coober Pedy, 23.xi1.1988, S. Lamond (MSM). Redescription MALE (figured by Moulds 1990, PI1.21, fig.5) Head.-Black, usually a small inconspicuous yellowish brown area visible under magnification on antennal plate adjacent to postclypeus and on midline near posterior margin; postclypeus black with lateral margins broadly yellowish brown, the colour often reaching to distal end and the black extending into yellow along transverse ridges. Proc. Linn. Soc. N.S.W., 123. 2001 154 THE GENUS KOBONGA (CICADIDAE) Anteclypeus black. Rostrum black sometimes tending brown basally. Antennae black or nearly so, sometimes brown at distal end of basal segment. Thorax.- Pronotum dark reddish brown to near black; a broad jet black fascia along dorsal midline usually encompassing a narrow muddy yellow fascia for much of its length, the black gently tapering distally with both anterior and posterior ends expanded laterally but most pronounced on latter; pronotal collar dull mid yellow between lateral angles, anterior of lateral angles black or nearly so but sometimes with areas of yellowish brown. Mesonotum dark reddish brown; a dorsal pair of obconical black fascia based on anterior margin and meeting on midline; a paramedian pair of similar but larger fascia that reach to extremities of anterior areas of cruciform elevation; black between anterior arms of cruciform elevation, often with ill-defined margin; cruciform elevation black and dark brown, the anterior arms nearly always partly or entirely dark brown. Wings.- Hyaline. Fore wing with distal end of radial cell not reaching beyond mid length of wing; bold infuscation overlaying basal veins of apical cells 2 and 3; bold infuscation along ambient vein from apex to vein CuA,, the infuscation extending part way along each adjoining vein except CuA,,; basal cell hyaline or very nearly so; basal membrane grey, usually dark. Hind wing with bold infuscation along ambient vein except on leading edge; plaga muddy white. Legs.- Black and yellow; fore femur with a prominent dull yellow fascia mid laterally on outer surface and a smaller yellow fascia almost dorsal; mid and hind coxae, trochanters and femora dull yellow on their inner surfaces; hind tibia brown at distal end. Opercula.- Muddy yellow, partially black at base and around inner margin; outer margin fringed with long silver cilia. Meracantha greater than equilateral triangle. Abdomen.- Tergites dominantly black, tergite 2 with posterior margin dark reddish brown dorsally, tergites 3-8 with posterior margins reddish brown tending yellow at extreme edge, coloring on 8 broadest and dominantly yellow. Sternites brownish yellow with broad black midline, sternite VII with the black tapering to terminate short of distal margin, sternite VIII blackish only at base. Genitalia (Fig. 5).-Similar to K. umbrimargo but with median lobe longer and uncal lateral lobes with edging rim terminating at level of ventral margin of lobe. FEMALE Similar to male. Abdominal segment 9 in dorsal view near an equilateral triangle; black dorsally and laterally, brown or brownish yellow ventrally and sometimes apically except for black apical spine. Ovipositor sheath projecting marginally beyond pygofer apical spine. MEASUREMENTS n= 10 males, 10 females (includes largest and smallest of available specimens). Length of body: male 24.8 - 31.7 (29.0); female 31.0 - 36.3 (33.25). Length of fore wing: male 32.8 - 37.9 (35.7); female 36.5 - 40.6 (38.95). Width of head: male 8.3 - 9.4 (8.95); female 8.8 - 10.2 (9.6). Width of pronotum: male 9.1 - 10.6 (9.9); female 9.6 - 12.1 (10.8). Distinguishing features In general appearance most similar to K. apicans, especially in wing markings. Best separated from K. apicans by the characters listed in couplet 2 of the Key to Species. Distribution (Fig. 12) Western Australia between Cue and Leonora and in South Australia from near Chandler to Coober Pedy. The lack of records from localities between these two widely separated areas of distribution is almost certainly a consequence of inadequate collecting. Proc. Linn. Soc. N.S.W., 123. 2001 M.S. MOULDS AND K.A. KOPESTONSKY 155 The localities, Moir’s Rock and Merredin recorded for this species by Moulds (1990) should be referred to K. apicans, a species described in this paper. Adults have been taken in December, January and February, but adult emergence is strongly influenced by rainfall with most appearing immediately after heavy summer rain. Habitat Mulga, Acacia aneura, and other shrubs. Song. Unknown. Kobonga apicans sp.n. (Figs 6, 10, 12) Types Holotype male, 70km E of The Three Ways, nr Tennant Creek township, Northern Territory, 21.1.1984, M.S. and B.J. Moulds (AM). Paratypes:- WESTERN AUSTRALIA: 2 males (1 genitalia prep. KO12), Moir’s Rock, 42km NNW of Salmon Gums, 32:39S 121:25E, 3.1.1987, G. and A. Daniels (MSM). 1 female, Dedari, 20.1.[19]34, H.W. Brown; 1 male, Merredin, 15.xii.[19]60, L. Willis [coll. A.N. Burns] (MV). NORTHERN TERRITORY: 2 males, | female, (1 male genitalia prep. KO6), 200km N of Tennant Creek township, Stuart Hwy, 4.iv.1984, K. and E. Carnaby; | male (genitalia prep. KO8), 8 females, same data as holotype; | male, | female, Taylors Creek, 47km N of Barrow Creek township, 22.i1.1984, M.S. and B.J. Moulds; 1 female, (genitalia prep KOS5), Witchetty Bore, Mt Allan Stn, 7.1.1977, G. Griffith (MSM). SOUTH AUSTRALIA: 1 male, Musgrave Ra, 25ml Bore, 10.1i.[19]66, P. Aitken and N.B. Tindale (SAM). Description MALE (Fig. 10) Similar to K. froggatti but differing as follows. Head.-Eyes set back hard against pronotum (those of froggatti tend to protrude with a gap between); back of eyes yellowish or light brown, never black. Thorax.- Yellow pronotal midline more pronounced and adjacent to the posterior end of this fascia a pair of laterally directed yellow fascia that touch on midline; pronotal collar yellow between lateral angles as for K. froggatti but the yellow considerably narrowed by black along anterior margin; cruciform elevation substantially muddy yellow. Legs.- Similar to K. froggatti but distal end of mid and hind femora with distinct yellow lateral patch. Abdomen.- Tergites rarely entirely black on basal half, sometimes dominantly reddish brown. Genitalia (Fig. 6).-Similar to K. umbrimargo but uncal lateral lobes with the edging rim terminating at level of ventral margin of lobe. FEMALE Similar to male. Abdominal segment 9 long, in dorsal view much longer than wide; dominantly muddy yellow, black or near black laterally but not reaching distal margin; the lateral black area usually with a black finger projecting distally in subdorsal region; apical spine black or near black. Ovipositor sheath long, projecting at least 1mm beyond pygofer apical spine. Proc. Linn. Soc. N.S.W., 123. 2001 156 THE GENUS KOBONGA (CICADIDAE) MEASUREMENTS n=4 males, 10 females (includes largest and smallest of available specimens). Length of body: male 23.4 - 24.7 (24.0); female 29.4 - 33.2 (30.95). Length of fore wing: male 29.1 - 30.2 (29.8); female 32.8 - 36.6 (36.6). Width of head: male 7.3 - 7.4 (7.33); female 8.0 - 8.9 (8.35). Width of pronotum: male 8.1 - 8.8. (8.3); female of = OZ (©:5)). Etymology A modification of the latin apricatio meaning basking in the sun, and referring to the species apparent liking for the sun and a hot climate. Distinguishing Characters Very similar to K. froggatti from which it is best separated by the characters listed in couplet 2 of the Key to Species. Females of K. apicans differ from all other Kobonga species in having abdominal segment 9 very elongate and the Ovipositor sheath very long so that it extends at least 1 mm beyond the pygofer apical spine. Distribution (Fig. 12) Inland Northern Territory south from the Tennant Creek district, Musgrave Ranges in the north of South Australia, and inland districts of south-western Western Australia between Coolgardie, Merredin and Salmon Gums. The absence of records across Western Australia is almost certainly an artifact of inadequate collecting. There are records for December, January, February and April but adult emergence is probably restricted to periods immediately following heavy rains. Habitat Tall shrubland in which adults are probably associated with Acacia spp. Song Unknown. ACKNOWLEDGEMENTS Dr C.N. Smithers provided helpful comments on the manuscript. For the gift of specimens we are grateful to the collectors of specimens as listed above. We thank the following for the loan of specimens in their care: Ms J. Forrest (SAM); Ms M. Humphrey (MM); Ms C. McPhee, (MV); Ms J. Margerison-Knight (BMNH) and Mr T. Weir (ANIC). Mick Webb (BMNH) kindly assisted with additional data on types. The photographs for Figs 7-10 were taken by Stewart Humphreys (AM) and the digital plate was assembled by Shane McEvey (AM). REFERENCES Ashton, H. (1912). Catalogue of the Victorian Cicadidae in the National Museum, Melbourne. Memoirs of the National Museum of Victoria 4: 23-29, pl.IV. [Separate, paginated 1-7.] Ashton, H. (1914). Catalogue of the Cicadidae in the South Australian Museum with descriptions of several new species. Transactions and Proceedings of the Royal Society of South Australia 38: 345-358, pl.XVIL. Ashton, H. (1915). Synonymic notes on a recent catalogue of Cicadidae in the South Australian Museum. Transactions of the Royal Society of South Australia 39: 91. Burns, A. N. (1957). Check list of Australian Cicadidae. Entomologischen Arbeiten aus dem Museum Georg Frey 8(2): 609-678. Distant, W.L. (1882). On some undescribed Cicadidae from Australia and the Pacific region. Proceedings of the Zoological Society of London 1882(1): 125-134, pl. VII. Proc. Linn. Soc. N.S.W., 123. 2001 M.S. MOULDS AND K.A. KOPESTONSKY 157 Distant, W. L. (1892). On some undescribed Cicadidae, with synonymical notes. Annals and Magazine of Natural History 9(6): 313-327. Distant, W.L. (1905). Rhynchotal notes.—XXXV. Annals and Magazine of Natural History 16(7): 265-281. Distant, W.L. (1906a). Some undescribed species of Cicadidae. Annals and Magazine of Natural History 17(7): 384-389. Distant, W.L. (1906b). A synonymic catalogue of Homoptera. Part 1. Cicadidae. British Museum, London. 207 pp. Distant, W. L. (1913). Undescribed species of Cicadidae. Annals and Magazine of Natural History 12(8): 485- 491. Distant, W. L. (1914). Some undescribed Cicadidae. Annals and Magazine of Natural History 14(8): 61-65. Distant, W. L. (1915). On some Australian Cicadidae. Annals and Magazine of Natural History 16(8): 50-53. Duffels, J. P. and van der Laan, P. A. (1985). Catalogue of the Cicadoidea (Homoptera, Auchenorhyncha) 1956-1980. Series Entomologica, Vol. 34. 414 pp. Junk, Dordrecht, Netherlands. Dugdale, J. S. [1972]. Genera of New Zealand Cicadidae [Homoptera]. New Zealand Journal of Science 14(4): 856-882. [For note of date of publication see Fleming, C. A., 1984, Records of the National Museum of New Zealand 2: 192.] Ewart, A. (1988). Cicadas (Homoptera). Pp. 180-201 in Scott, G. (ed.), Lake Broadwater. The natural history of an inland lake and its environs. Darling Downs Institute Press, Toowoomba. Ewart, A. (1989a). Cicada songs - song production, structures, variation and uniqueness within species. News Bulletin of the Entomological Society of Queensland 17(7): 75-82. Ewart, A. (1989b). Revisionary notes on the genus Pauropsalta Goding and Froggatt (Homoptera: Cicadidae) with special reference to Queensland. Memoirs of the Queensland Museum 27(2): 289-375. Ewart, A. (1998). Cicadas, and their songs, of the Miles-Chinchilla region. Queensland Naturalist 34: 54-72. Goding, F. W. and Froggatt, W. W. (1904). Monograph of the Australian Cicadidae. Proceedings of the Linnean Society of New South Wales 29(3): 561-670, pls XVIII, XIX. Kato, M. (1932). Monograph of Cicadidae. San Sei Do, Tokyo. 450 pp., 32 pls. [In Japanese but partly supplemented by English. ] Kato, M. (1956). The biology of cicadas. Bulletin of the Cicadidae Museum. 319 pp., 16 pls. Kirkaldy, G. W. (1907). Some annotations to M. Distant’s recent Catalogue of the Cicadidae. [Hem.] (1). Annales de la Societe entomologique de Belgique 51: 303-309. [Separate paginated 1-7.] Lithgow, G. (1988). Insects collected in and around the Chinchilla Shire. Pp. 63-72 in Hando, R. (ed.), Going bush with Chinchilla Nats -twenty years of field observations. Chinchilla Field Naturalists’ Club, Chinchilla. [Second edition, 1997] Metcalf, Z. P. (1944). List of journals and topical index. A bibliography of the Homoptera (Auchenorhyncha) 2: 1-186. Metcalf, Z.P. (1947). The center of origin theory. Journal of the Elisha Mitchell Scientific Society 62: 149- 175, pls 23-41. Metcalf, Z. P. (1963). General catalogue of the Homoptera. Fasc. 8. Cicadoidea. Part 2. vi, 492 pp. Moulds, M.S. (1990). Australian cicadas. New South Wales University Press, Kensington. x, 217 pp., pls 1- 24. Neave, S.A. (1939). Nomenclator zoologicus. 2: 1-1025. Schulze, FE., Kiikenthal, W. and Heider, K. (1930). Nomenclator animalium generum et subgenerum. 3(13): 1619-1778. Stal, C. (1862). Synonymiska och systematiska anteckningar 6fver Hemiptera. Ofversigt af Kongliga Svenska Vetenskaps-Akademiens Forhandlingar, Stockholm 19: 479-504. Walker, F. (1858). List of the specimens of homopterous insects in the collection of the British Museum. Supplement. British Museum, London. 369 pp. Weidner, H. and Wagner, W. (1968). Die entomologischen Sammlungen des Zoologischen Staatsinstitut und Zoologischen Museum, VII. Teil, Insecta IV. Mitteilungen aus den Hamurgischen Zoologischen Museum und Institut 65: 123-180. Proc. Linn. Soc. N.S.W., 123. 2001 | add “remem naa caavoN eae 29 Late vADg: ices faSeags reat 5A Sn vr ene bi doen oabibioiD badsesbut red @ten @3) -t8-08 (2t Sener ¥ monte he Oey athe HG Ae nici pve any ae sa eee < rule ao oti ea teuea Me eee faba Very inal Be ty rine rromwast Pra shiek if ist at dsichay Gt diaspiad. 1OS-083 tag pram ess D sg CaN 5 Be 0 YY et oa gins ee: on » ih Beane 8 he aro nn I ec peta ie BO * 23 | Papa ane OEY ORES MTT? Snahtssae eae Fach ateie sere: (eam Romo } negeor? bak yihiboo nth: Ae) gumsy beh Ao sot puanowives {68 RE RBS onsthy it na ‘Bitola ait to eiamalh bneaes90h) oo sureties inizege rire EPS; © coiga: slitdamd asi ail to..eg nee tte See, aahen {SURI oid aebiheot > goiledeu, gargs BORE) WoW x ey cases TBS ayes aan lees io iar rah peer eit tage oot Ww Aetinon bel ’ Lai; Wis DEY Sey tr arterial cheery t Git Salen ; rie ters at Bes 7 : mite et Taint spanaanatl rac) xiguroniiyy Sh. a zie ‘bres one ‘a, ae sre Sb Sb o2 ove Deeb ‘SAUIA GAY fo ates Ysa igi Yo raiase ‘et Chol} ES, Seng b-ES aly Bt Hg ER ers hed wsbiobssia 4 cent, nrsingsnelh oily lo Sag goles inease) AEA Dy SS it -f dq, a TiSax Sis angi est inrevia U cole Mnoe walt ages rasifercreash, AOROT) Eat, : BE } AC KNOY ‘| Bene tet adios anv aaigel AREA A HED mwervateaduer 19 iniesensy, tinlooenG wt) Pile ora y. 2 Skis brn doen aa Or roN erH a -enabieeeg igi i igis mh met Pe AOR Shae (fan ie i Macirages Pane naiazigoloo’, ShorPsaiiet’ sh Ben nf reeaigh ¥ z Saris ii i 481240 (eae at Ue Viceormn ¢ i alee sik tins Rataoniil i vere sn Ageteieede ny ‘i Migiuive wl Hie Nativogn! Amérniny kets thy fhe PPO eel TY | Seperate Ong hades hid Askin 1007S), Canitiogue of the Cleadidae jo se a “ith am | mri tina 8 Viernes teh coll thas oF pew aoboes: Trehianctipns pol Mrrceadinns of ie Rien Society af Goeth Aue WM), PAV —_ dito, 1.259045). Byounvinde reese of econ oa (i & ion he (nt ie mrae fecha iptid Vinenp OF the Royal Sdeies f Seu ifs Agerall 3h; pions +. (7997). Cheek Ge of d Lair ier eodabhe Meth falas pial folds nt: hema vbr tn wes dem af Poa Oct) Oe ay ‘ Cmment, WA. (10P2) Tho ald bndeteribed Oyaliitaw dren Anintielia GM ie Paciie cas i avin san | ah i Sti lety Cf LAE id THAZi. } ta be ota pi Vil OOELAOL Sat 201502, Mae 20H A New Cuttlefish, Sepia grahami, sp. nov. (Cephalopoda: Sepiidae) from Eastern Australia AMANDA REID 6 Sturt Place, Bulli, NSW, 2516. Email: mandy.reid @ optusnet.com.au Reid, A. (2001). A new cuttlefish, Sepia grahami, sp. nov. (Cephalopoda: Sepiidae) from Eastern Australia. Proceedings of the Linnean Society of New South Wales 123, 159-172. Anew cuttlefish, Sepia grahami, sp. nov. (Cephalopoda, Sepiidae) from eastern Australia is described. The species is found from between southeast of Yamba (29°33’S 153°25’E) to Tathra (36°40’S 150°03’E) in New South Wales at depths ranging from 2.5—84 meters. The presence of a pair of distinctive ‘eyespots’ on the dorsal mantle has led to its misidentification as Sepia mestus Gray, 1849 and because of similarities in the cuttlebone it may also be confused with the commercially important species Sepia rozella (Iredale, 1926). Distinguishing features are provided to enable the reliable identification of mature specimens of this new species. Manuscript received 2 July 2001, accepted for publication 21 November 2001. KEY WORDS: Sepiidae, Sepia grahami, Sepia mestus, Sepia rozella, new species, Australia. INTRODUCTION Australia is endowed with the richest cuttlefish fauna in the world. Of the 113 nominal species, one third occur in Australian waters. The new species described here brings the total to 37 species found in Australian waters. The species, from eastern Australia, was discovered among collections held in the Australian Museum and the Museum Victoria. It occurs sympatrically with two species with which it has been confused: S. mestus Gray, 1849 and S. rozella (Iredale, 1926), though based on the number of specimens held in museum collections and examined in fishmarkets in Sydney, it appears that this new species occurs in lower numbers than S. mestus and S. rozella. MATERIALS AND METHODS This work was based on museum material. All material studied is listed in the ‘Material examined’ section below. Institutional acronyms used are: AM, Australian Museum, Sydney, Australia; MV, Museum Victoria, Melbourne, Australia. Measurements and indices used throughout this paper are primarily those given in Roper and Voss (1983), using dorsal mantle length (ML) as a size standard. Some additional measurements are used, and these with the definitions listed by Roper and Voss (1983) are given in Reid (2000). Parts of the club and arm sucker rims are described using the terminology of Nixon and Dilly (1977) while nomenclature for the radula follows Nixon (1995). Beaks were described following Clarke (1986). Diagrammatic illustrations of measurements and terminology used for key structures are shown in Reid (2000). Proc. Linn. Soc. N.S.W., 123. 2001 160 NEW CUTTLEFISH FROM EASTERN AUSTRALIA Measurements were made either using dial callipers, or an eyepiece micrometer inserted in a stereo microscope. All measurements are expressed in millimetres (mm). Measurements and counts for individual specimens and ranges of arm length indices, arm sucker diameter indices, and arm sucker counts are presented in the tables; ranges for all other characters appear in the text. In the species description and tables, the range of values for each character is expressed as: minimum—mean—maximum (standard deviation, SD). Values for each sex are given separately. Measurements for structures that were clearly distorted or broken were not attempted, and these, in addition to missing and unknown values, appear as a dash (—) in the tables. Ranges for specific character traits given with the species description do not, therefore, always refer to the total number of specimens examined. For examination of arm and club sucker rims, suckers were removed from the middle of designated arms and the tentacular club, mounted in glycerine jelly and viewed using a compound microscope. Radulae and beaks were dissected from the buccal mass, and soaked for approximately 30 minutes in a warm, saturated potassium hydroxide solution, then radulae were cleaned using forceps and a fine brush. Radulae were mounted in glycerol, and the new, unused portion was examined. All characters refer to both sexes unless stated otherwise. The species description was generated using DELTA (DEscription Language for TAxonomy) software (Dallwitz 1980; Dallwitz et al. 1993; Partridge et al. 1993). Taxonomy Sepia grahami, sp. nov. (Figs 1-5, Tables 1-3) Material examined Type material: Australia: New South Wales: Holotype: 1° (43.8 mm ML), Clarence River, 29°33°S 153°25°E—29°32’S 153°25’E, 53-49 m, 30 May 1995, coll. K. J. Graham (MV F87757). Paratypes: 1% (65.5 mm ML), 1 2?(56.8 mm ML), SE of Yamba, 29°33°S 153°25°E-29°34’S 153°24’E, 51-49 m, 30 May 1995, coll. K. J. Graham (AM C306769); 29 (55.4, 55.7 mm ML), NE of Wooli, 29°48°S 153°27 E-29°49’S 153°26E, 70-65 m, 18 May 1995, coll. K. J. Graham (AM C306773); 1° (49.3 mm ML), 19(47.4 mm ML), off Wooli, 29°49°S 153°27°E-29°48’S 153°26°E, 69-22 m, 18 May 1995S, coll. K. J. Graham (AM C306389). Other material examined: New South Wales: 1° (55.0 mm ML), 1? (49.7 mm ML), Clarence River, 29°33’S 153°29°E-29°32’S 153°29’E, 61-56 m, 26 Jun 1995, coll. K. Graham, NSW Fisheries (MV F88564); 1° (30.3 mm ML), 12(48.9 mm ML), Clarence River, 29°38°S 153°24°E-29°39’S 153°23’E, 48-44 m, 11 Apr 1996, coll. K. Graham, NSW Fisheries (MV F88565); 1° (30.9 mm ML), 12 (35.8 mm ML), off Newcastle, 32°45°S 152°15°E-32°44’S 152°15’E, 84-82 m, 10 Oct 1995, coll. K. J. Graham (MV F87758); 3% (44.1-56.7 mm ML), 1? (53.3 mm ML), Sydney Harbour, Parsley Bay, 33°51°S 151°16E, 4.0-2.5 m, 8 May 1976, coll. J. Paxton, D. Hoese, B. Russell and D. Blake (AM C204479); 292 (55.4, 80.7 mm ML), Jervis Bay, approx. 35°0’S 150°45’E, 22-23 May 1971, (AM C152595); 1° (45.4 mm ML), 19 (60.4 mm ML), off Tathra, 36°36°S 150°1°E-36°44’S 150°5’E, 59-37 m, 7 Mar 1994, coll. K. J. Graham on FRV ‘Kapala’ (AM C304884); 22 (57.6, 81.6 mm ML), off Tathra, 36°40°S 150°3°E-36°43°S 150°2°E, 48-46 m, May 1994, coll. K. J. Graham (AM C304890). Diagnosis Male and female arms subequal in length. Arm suckers tetraserial. Hectocotylus absent. Tentacular club with 4—5 suckers in transverse rows (usually four, rarely five); Proc. Linn. Soc. N.S.W., 123. 2001 A. REID 161 Figure 1. Sepia grahami, sp. nov.: (a) funnel organ, female 35.8 mm ML (MV F87758), scale bar 2 mm; (b) funnel locking (left) and mantle locking (right) cartilage, male 56.7 mm ML (AM C204479), scale bar 2 mm; (c) sucker rim arm 2, portion of toothed half, paratype male, 65.5 mm ML (AM C306769), scale bar 0.05 mm; (d) sucker rim, portion of non-toothed half, specimen as in (c); (e) club, paratype female, 55.7 mm ML (AM C306773), scale bar 2.0 mm; (f) club sucker rim, portion of toothed half, paratype female, 56.8 mm ML (AM C306769), scale bar 0.04 mm; (g) club sucker rim portion of non-toothed half, specimen as in (f). Proc. Linn. Soc. N.S.W., 123. 2001 162 NEW CUTTLEFISH FROM EASTERN AUSTRALIA TABLE | Measurements (mm), counts and indices of seven male Sepia grahami, sp. nov. Museum MV F87757 AM AM AM C304884 AM C306389 AM AM C306769 Reg. no. holotype C204479 C204479 paratype C204479 paratype ML 43.8 44.1 45.3 45.4 49.3 56.7 65.5 MWI 71.0 47.6 60.7 50.9 62.1 58.0 58.8 AMHI 7 9.5, 8.6 12.6 = 12.3 8.7 VMLI 90.4 87.5 88.5 86.6 84.2 90.3 100.9 FWI 8.0 11.3 8.4 14.5 11.2 13.2 iIS57/ Fila HES) 6.1 6.4 4.0 6.1 4.4 9.6 Fllp 5.0 6.1 5.7 4.0 5.5 4.6 3.1 FuLlI 38.8 37.9 B75 33.0 34.5 31.7 39.2 FFul 16.0 15.9 Wet) 13.2 18.3 17.6 21.4 HLI 39.5 31.7 31.3 37.7 41.6 33.2 42.0 HWI 55.3 46.7 43.3 45.8 51.1 41.4 42.1 EDI 15.3 12.0 13.7 15.4 15.2 12.2 12.8 ALI 43.4 38.5 35.3 37.4 48.7 38.8 55.0 ALI2 41.1 36.3 35.3 37.4 46.7 35.3 53.4 ALI3 42.2 35.1 37.5 39.6 53.8 34.4 45.8 ALI4 52.5 39.7 39.7 39.6 46.7 40.6 67.2 ASIn1 IES 7 1.36 1.32 1.32 1.32 1.23 1.30 ASIn2 1.37 1.36 1.32 1.32 1.42 1.41 1.37 ASIn3 1.37 1.36 1.32 1.32 ES 2 1.41 1.37 ASIn4 1.60 1.59 1.55 1.43 1.42 1.41 1.53 ASC1 120 124 104 110 118 126 125 ASC2 110 116 110 116 134 126 130 ASC3 122 130 118 130 147 132 152 ASC4 147 152 152 160 154 176 166 CILI 12.3 13.6 11.7 13.2 15.0 12.7 14.4 CIRC 4 4 4 5 4 5 4 TrRC 20 14 17 20 19 15 18 CISI 0.91 1.13 0.88 1.10 1.22 1.06 1.37 ClSId 0.57 0.79 0.55 0.70 0.81 0.53 0.76 CISIv 0.46 0.68 0.44 0.66 0.61 0.71 0.61 GiLC = 24 23 26 = 26 25 GiLI 28.3 30.8 33.3 25.3 31.8 28.2 31.6 SpLI 11.2 = Vell 10.1 = 6.0 9.5 SpWI 6.12 = 4.29 6.52 = 4.41 4.84 CbL 44.4 43.4 44.7 45.2 56.6 56.8 72.6 CbWI1 42.3 ~ - 41.8 39.9 40.8 39.5 CbBI 10.8 = = 10.8 10.4 10.9 11.4 SLI 5.6 = = 6.6 5.1 6.2 6.5 StZI 56.3 - = 57.3 61.0 60.0 62.0 LoLI 85:1 - - 33.0 22.8 34.5 30.7 LoL/StZ (% 62.4 = = 57.5 37.4 SS 49.6 suckers differ only slightly in size, 3-4 suckers slightly enlarged. Cuttlebone spine with ventral keel; sulcus shallow, narrow, indistinct; anterior striae inverted U-shaped; inner cone limbs raised to form rounded ledge posteriorly, thickened, yellowish or ochre- coloured. Mantle with dorsal ‘eyespots’. Description Counts and indices for individual specimens are given in Tables 1 and 2; ranges for arm length indices, arm sucker diameter indices and arm sucker counts are shown in Table 3. Small to moderate-sized species; ML males 43.8—50.0-65.5 (SD, 8.2), females 47.4-61.0—81.6 (SD, 12.0). Mantle oval; MWI males 47.6—58.4—71.0 (SD, 7.6), females 56.3-63.5-79.9 (SD, 7.2); dorsal anterior margin triangular, obtuse; mantle extending anteriorly to level of middle of eyes (approximately); AMHI males 8.6—LO.9—13.7 (SD, 2.2), females 9.6—12.4—14.4 (SD, 1.8). Ventral mantle margin emarginate, without distinct lateral angle; VMLI males 84.2—89.8-100.9 (SD, 5.4), females 86.0-88.0—90.7 (SD, 1.5). Fins wide; FWI males 8.0—11.8—15.7 (SD, 2.9), females 8.4—12.3—15.1 (SD, 2.5); anterior origin posterior to mantle margin; FlJa males 4.0-6.3—9.6 (SD, 1.9), females 3.2—6.2— 11.4 (SD, 2.8); fins rounded posteriorly; narrow gap between fins; FIIp males 3.1—4.9— 6.1 (SD, 1.1), females 4.3-6.6—8.8 (SD, 1.4). Funnel short, robust, broad; extends to Proc. Linn. Soc. N.S.W., 123. 2001 A. REID 163 Figure 2. Sepia grahami, sp. nov.: (a) upper beak, lateral view, female 47.4 mm ML (AM C3063839), scale bar 2 mm; (b) lower beak anterio-lateral view, specimen as in (a); (c) radula, paratype female, 47.4 mm ML (AM C306389), scale bar 0.20 mm; (d) male reproductive tract (testis not shown), paratype 49.3 mm ML (MV F82464), scale bar 4 mm (AAG, appendix of accessory gland; AG, accessory gland; CC, ciliated canal; DDC, distal deferent canal; GO, genital orifice; MG, mucilaginous gland; SG, spermatophoric gland; SS, spermatophoric sac (containing spermatophores); T, testis; VD, vas deferens); (e€) spermatophore, male 49.3 mm ML (AM C3063839), scale bar 0.40 mm (CB, cement body; EA, ejaculatory apparatus; SR, sperm reservoir); (f) spermatophore, enlargement of cement body, specimen as in (e), scale bar 0.20 mm. b Wea: J\ist\ SAAG Proc. Linn. Soc. N.S.W., 123. 2001 164 NEW CUTTLEFISH FROM EASTERN AUSTRALIA Figure 3. Sepia grahami, sp. nov.: (a) cuttlebone, dorsal view, female 60.4 mm ML (AM C304884), scale bar 10 mm; (b) cuttlebone, ventral view, specimen as in (a) (IC, inner cone); (c) arrangement of skin papillae dorsal and ventral to eye (E), male 56.7 mm ML (AM C204479), scale bar 16.2 mm; (d) dorsal mantle showing position of dorsal eyespots (larger posterior pair) and small granular orange spots (smaller anterior pair), female 81.6 mm ML (AM C304890), scale bar 10 mm. Proc. Linn. Soc. N.S.W., 123. 2001 165 A. REID 0687050 NIV 2 66E OVE L°87 v'6C G9E = L°C9 0°c9 c99 £9 0'6S = g9 se = 6S 6s = Lil £01 86 Tor TOI v8y Ley O'tY ler v'6e = 6S 69 VLS Cc C9 19s cv = = 88 COL = OTE 9'SE V6e a OEE + 0S'0 0s'0 cs'0 c9'0 vs'0 vs'0 9°50 0s'0 69°0 0L':0 cL'0 £9°0 cll 801 6eT 90'T 801 801 61 LI 0c (G6 81 LI v v Vv v Vv v 6Tl Vel Lot Lel 61 Oe! psi p81 081 9LI 181 891 9r1 ssl Chl (Gal vel IST orl Orl al Orl bri cr 9EI 8cl OIl Oc! OT Oe! 9E 1 6r'I 9S'1 8S'l vr 9c I vel cel 9c L9'I vr 97 I vol OTT 9ST 8ST c9o'T 9 T vol OTT 9S'I 0s T br 9c T veVv Lv 06S 9S TOS vCY 6 0v T8e 9'8P TTS 6S cer LE V9E very LY CLS 09r CLE y9E LLY T Is CLS 6 9F SIT Vl 901 vol 6Cl mal 8 '6r 6 SP ees CS L6v 0'8r CEE STE a4 00r OLE OLE ssl L'0C =] ie a! Lit OTE 6 EE 9'9€ = 9'8E Se ES 97 6v ce ve 617 OL 19 OES ce 06 of vl Ll 6c! st Ctl Lil vL8 L'06 cL8 688 0°98 v'88 ell 9TI = vel = 6€l g°99 8s 6 6L £9 Leg v9 L'08 v'09 OLS 89° Lss pss S6SCSIDWNV P88P0E0 WV O068P0E0O NV 69L900E0 NV ELL9DED WV £41900 NV ‘aou ‘ds ‘1upyp18 vidag a{eWId} DUIU JO SODIPUI puke s}UNOD ‘(WIU) syUDTOINSeOY/] 7 ATaVL 6LPb0CO WV 6869060 NV %) ZIS/TOT 1101 IZIS IIs 1aq9 IM9) 790 1a3q IT!9 OTD ATSIO PISIO ISIO OWL oulo Im10 vOSV €OSV ZOSV IOSV pulsv EuSV ZUISV IUISV vITV €I1V ZITV ITV “ou “JOY umosnyy Proc. Linn. Soc. N.S.W., 123. 2001 166 NEW CUTTLEFISH FROM EASTERN AUSTRALIA Figure 4. Cuttlebones, dorsal and ventral views. Sepia grahami, sp. nov., paratype AM C306769, 72.9 mm CbL: (a) dorsal view, (b) ventral view. Sepia mestus Gray, AM C152722, 80.5 mm CbL: (c) dorsal view, (d) ventral view. Sepia rozella (Iredale), MV, 113.7 mm CbL: (e) dorsal view, (f) ventral view. Scale bars 1 cm. Proc. Linn. Soc. N.S.W., 123. 2001 A. REID 167 Figure 5. Distribution of Sepia grahami, sp. nov. (large circles); arrow indicates the type locality. Tathra e Proc. Linn. Soc. N.S.W., 123. 2001 168 NEW CUTTLEFISH FROM EASTERN AUSTRALIA TABLE 3 Sepia grahami, sp. nov.; ranges of arm length indices (ALI), arm sucker diameter indices (ASIn) and arm sucker counts (ASC) of seven mature males and nine mature females. min. = minimum, max. = maximum, SD = standard deviation. Males Females min. mean max. SD min. mean max. SD ALI1 335.3) 42.4 55.0 Toll 36.4 44.2 Syl) eS ALI2 353 40.8 53.4 6.9 36.4 43.3 57/5) 6.4 ALI3 34.4 41.2 53.8 6.8 36.6 44.0 53.9 6.1 ALI4 39.6 46.6 (SZ 10.3 37.5 47.9 59.2 8.3 ASIn1 13 LY 137/ 0.04 1.16 1233 1.56 0.14 ASIn2 32) 137 1.42 0.04 1.16 1.36 1.62 0.18 ASIn3 132 1.38 eS 2) 0.07 1.24 1.40 1.67 0.14 ASIn4 1.41 1.50 1.60 0.08 1.26 1.47 1.69 0.13 ASC1 104 118 126 8 110 Da 144 11 ASC2 110 120 134 10 126 139 146 6 ASC3 118 133 IS 12 134 149 184 15 ASC4 147 158 176 10 168 / 186 6 level of anterior rim of eye; FuLI males 31.7—36.1—39.2 (SD, 3.0), females 26.0—34.3— 39.0 (SD, 4.2). Funnel free portion approximately half funnel length; FFul males 13.2— 17.1-21.4 (SD, 2.5), females 11.7—16.0—20.7 (SD, 3.4). Funnel organ dorsal elements inverted V-shaped with low medial swelling and small papilla in front; ventral elements oval with acute anterior tips (Fig. 1a). Mantle-locking cartilage curved, with semicircular ridge; funnel-locking cartilage with depression which corresponds to ridge (Fig. 1b). Head short; HLI males 31.3—36.7-42.0 (SD, 4.6), females 29.5-35.7-43.6 (SD, 4.3); head broad, narrower than mantle; HWI males 41.4-46.5—55.3 (SD, 5.1), females 42.0— 48.1-53.3 (SD, 3.9). Eyes moderate size; EDI males 12.0—13.8—15.4 (SD, 1.5), females 10.2-12.1-16.4 (SD, 1.8); ventral eyelids present. Male and female arms subequal in length. Arm length index (ALI) of longest arms in males (ALI4) 39.6-46.6—67.2 (SD, 10.3), ALI of longest arms in females (ALI4) 37.5-47.9-59.2 (SD, 8.3). Protective membranes in both sexes narrow; normal, not thickened. Distal arm tips in both sexes not markedly attenuate. Arm suckers tetraserial in both sexes. Suckers in males normal in size (not greatly enlarged); similar to females arm suckers in size. Chitinous rims of arm suckers with elongate rectangular teeth on distal half of inner ring (Fig. Ic), teeth absent on proximal half of ring (Fig. 1d); infundibulum with 8-9 rows of hexagonal processes, inner 4—5 (variable) rows with elongate rounded pegs, pegs becoming smaller towards periphery of sucker; peripheral sucker rim processes radially arranged, elongate, without pegs. Sucker counts range from 104-186; females with higher average counts than males. Hectocotylus absent. Tentacular club similar length in males and females; CILI males 11.7—13.3— 15.0 (SD, 1.2), females 11.6—13.6—16.7 (SD, 1.5). Club slightly recurved; short; sucker- bearing face flattened (Fig. le). Club with 4-5 (usually four, rarely five) suckers in transverse rows, CIRC males 4—4—5 (SD, 0.5), females 4-4-5 (SD, 0.3); 14-22 suckers in longitudinal series, TrRC males 14-18-20 (SD, 2), females 17—18—22 (SD, 2). Suckers differing slightly in size, 3-4 suckers toward posterior end of club very slightly enlarged; distal tip of club without pair of enlarged suckers; CISI males 0.88—1.10—1.37 (SD, 0.17), females 1.03—1.12—1.39 (SD, 0.11); dorsal and ventral marginal longitudinal series of suckers differing slightly in size; dorsal marginal longitudinal series of suckers slightly larger than those in ventral marginal series; CISId males 0.53-0.67—0.81 (SD, 0.12), Proc. Linn. Soc. N.S.W., 123. 2001 A. REID 169 females 0.50—0.64—0.74 (SD, 0.09); ClSIv males 0.44—0.59-0.71 (SD, 0.11), females 0.42—0.54—0.67 (SD, 0.07). Sucker dentition: half inner ring circumference in both sexes with elongate rectangular teeth (Fig. 1f), remaining half with blunt projections (Fig. 1g); infundibulum with seven rows of hexagonal processes, innermost with elongate rounded pegs, pegs smaller towards periphery of sucker; at periphery, processes smaller, elongate- rectangular, without pegs (similar to arm suckers). Swimming keel of club extends well beyond carpus. Dorsal and ventral protective membranes not fused at base of club; joined to stalk; dorsal and ventral membranes differing in length, dorsal membrane extends beyond carpus along stalk, ventral membrane terminates at posterior end of carpus; approximately equal width; dorsal membrane forms shallow cleft at junction with stalk (Fig. le). Gills with 23—26 lamellae per demibranch; GiLC males 23-25-26 (SD, 1.3), females 26. Gill length: GiLI males 25.3—29.9—33.3 (SD, 2.8), females 31.0—37.1—-50.7 (SD;,7.2): Buccal membrane without suckers. Upper beak (Fig. 2a) rostrum sharply pointed, short, length approximately equal to width, cutting edge slightly curved; hood high above crest posteriorly; crest curved, lateral wall shallowly indented posteriorly; wings and hood narrow and short; jaw angle slightly less than 90 degrees, slightly acute; hood and crest dark brown. Lower beak (Fig. 2b), rostrum protrudes only slightly, cutting edge straight; hood low on crest; crest straight, no indentation on lateral wall edge; lateral wall edge angled posteriorly, not perpendicular to crest; hood and wings, width broad; hood notch deep, broad; wings widely spaced; crest narrow; rostrum pigmented dark brown, wings dark brown on inner margin only, rest of wing light brown, crest dark brown. Radula (Fig. 2c) homodont; rhachidian teeth with truncate bases, slender, triangular, sides straight; first lateral teeth similar length and width to rhachidian teeth, asymmetrical with mesocone slightly displaced toward centre of radula; second laterals longer than first, distinctly curved on lateral margin, with broad heels; marginal teeth elongate with long tapered and curved mesocone. Digestive tract: (not illustrated) paired salivary glands approximately one-third length of buccal mass; paired digestive glands large, located close together, with narrow, elongate triangular lobes posteriorly, ducts connect digestive glands near midline with caecum, ducts with branched attached pancreatic tissue; oesophagus runs dorsally along median junction of digestive glands, joins sac-like stomach immediately posterior to digestive glands; caecum disc-like, grooved in a blunt V-shape anteriorly, surface lining finely pleated; intestine undifferentiated; ink sac very large, elongate; anal flaps well-developed. Male reproductive tract (Fig. 2d): testis on left posterior side of viscero-pericardial coelom; at distal end, convoluted vas deferens opens into broad, cone-shaped mucilaginous gland, then narrower, curved, spermatophoric gland. Close to junction with lobe-shaped accessory gland and gland appendix, delicate ciliated canal joins spermatophoric gland; distal deferent canal connects appendix of accessory gland to spermatophore storage sac; genital orifice opens dorsal to left gill in anterior end of mantle cavity. Spermatophores (Fig. 2e, f): cement body bipartite; aboral end cylindrical, rounded posteriorly, connects to sperm reservoir via narrow duct; oral end of cement body cylindrical, approximately three-quarters length of, and very slightly narrower than aboral end, connects to aboral end via short neck; middle tunic commences along aboral part of cement body; ejaculatory apparatus coiled, extends into oral dilation of spermatophore. Spermatophores 3.4—-4.5— 6.2 mm long (SD, 1.1), 0.2-0.2—0.3 mm wide (SD, 0.1); SpLI 6.0—8.9-11.2 (SD, 2.1); SpWI 4.3-5.2-6.5 (SD, 1.0). Smallest male with well-developed spermatophores in spermatophoric sac 43.8 mm ML. Female reproductive tract: (not illustrated) ovary hangs from dorsal wall of posterior viscero-pericardial coelom. Oviduct thin-walled, continuous with body cavity; distally with thickened, glandular walls (oviducal glands). Nidamental glands in mature animals occupy large portion of ventral side of mantle cavity. Accessory nidamental glands Proc. Linn. Soc. N.S.W., 123. 2001 170 NEW CUTTLEFISH FROM EASTERN AUSTRALIA anterior to nidamental glands. Eggs spherical, 3.6—4.9-5.7 mm diameter (SD, 0.95); EgDI 4.5-7.6—10.2 (SD, 2.5). Smallest female with well-developed eggs in ovaries 55.7 mm ML. Subdermal cartilaginous layer between cuttlebone and skin absent. Cuttlebone (Figs 3a, b and 4a, b) length approximately equal to mantle length; outline oval; CbL males 43.4—-52.0-72.6 (SD, 10.8), females 51.8-60.1—83.5 (SD, 10.4); CbWI males 39.5—40.9—42.3 (SD, 1.2), females 39.4—43.1—48.4 (SD, 2.9); cuttlebone not strongly convex in lateral view; CbBI males 10.4—-10.9-11.4 (SD, 0.4), females 9.8—10.6— 11.7 (SD, 0.7). Bone bluntly rounded anteriorly; bluntly rounded posteriorly; not strongly recurved ventrally. Dorsal surface creamy white (very slight pinkish tinge); evenly convex; entire surface calcified with reticulate granulose sculpture concentrated posterio-laterally and posteriorly in irregular longitudinal ridges (Figs 3a and 4a); posterior end of bone not covered with smooth glaze-like substance. Dorsal median rib absent; lateral ribs present, indistinct. Chitin surrounds rim of cuttlebone outer cone. Spine present, short; SLI males 5.1—6.0—6.6 (SD, 0.6), females 3.8—5.5-6.5 (SD, 1.1); spine straight, directed dorsally; with ventral keel; fine, radiating ribs between outer cone and spine; ventral notch at base of spine absent. Dorso-posterior end of cuttlebone without median longitudinal ridge anterior to spine. Striated zone concave; StZI males 56.3—59.3-62.0 (SD, 2.4), females 58.7-62.1—66.2 (SD, 2.7); striated zone extends laterally to inner cone, not separated from outer cone by smooth marginal zones. Last loculus flat; LoLI males 22.8—31.2—35.1 (SD, 5.0), females 28.7— 33.9-39.9 (SD, 4.1); last loculus at midline half length of striated zone, LoL/StZ(%) males 37.4—-52.9-62.4 (SD, 9.8), females 43.4-54.8-63.6 (SD, 8.2); loculus extends posteriorly as very narrow margin on each side of striated zone for about half its length. Sulcus extends entire length of cuttlebone; shallow, narrow (very indistinct); not flanked by rounded ribs. Last loculus with shallow median indentation, not very pronounced. Anterior striae inverted U-shaped. Limbs of inner cone extend anteriorly to approximately two-thirds length of striated zone; inner cone lateral limbs not separated from outer cone by two distinct smooth zones. Inner cone limbs narrow, strap-like anteriorly, broaden posteriorly; raised to form rounded ledge posteriorly (Figs 3b and 4b); thickened (yellowish, or ochre-coloured); without calcareous ribs radiating into outer cone. Outer cone calcified; narrow anteriorly, broadens posteriorly; lateral limbs flared ventro-laterally; limbs forming thin rim ventral to spine. Body papillae present (not visible all specimens); dorsal mantle with longitudinal row of up to six ridges along each side, close to fins; ventral mantle with longitudinal row of approximately six ridges along each side close to fins and scattered papillae ventral to these. Head papillae present; positioned dorsally and laterally; prominent ear-shaped lobe ventral to eye; second large lobe dorsal to eye; small papillae posterior to large ventral lobe (Fig. 3c). Additional small papillae scattered over head. Arm papillae present, same as those on head. Colour (alcohol preserved specimens): head, arms, and dorsal mantle pale buff pinkish-brown (see Remarks); paired dark-brown dorsal eye spots present and pair of distinct orange spots anterior to these (Fig. 3d). Fins pale; without markings at base. Ventral pigment very pale with few evenly scattered chromatophores. Ridges orange-pink in colour. Type locality Australia: New South Wales, 29°33°S 153°25 E-29°32’S 153°25’E. Distribution Australia: New South Wales, from south-east of Yamba 29°32°S 153°25°E— off Tathra 36°44’°S 150°5’E (Fig. 5); depth range 2.5—84 m. Proc. Linn. Soc. N.S.W., 123. 2001 A. REID 171 Etymology This species is named in honour of Ken Graham from NSW Fisheries. Ken has made an enormous contribution to our understanding of the diversity of cephalopods in eastern Australian waters through his collection of specimens and lodgement of this material in Australian museums. Ken collected almost all specimens of this new species. Remarks A single large male specimen (AM C306769, 66.5 mm ML) is patterned dorsally with fine transverse bars over the dorsal mantle. Though faint, the pattern is of the ‘zebra’-type, typical to that seen during courtship displays in a number of male sepiids (Hanlon and Messenger 1996). Preserved specimens of S. grahami have been misidentified as S. mestus. This is largely due to the presence of the dark dorsal ‘eyspots’, which are present in both species. This may also lead live S. mestus to be confused with S. grahami. The body is pinkish-brown in preserved S. grahami specimens, and dark purple in S. mestus. Sepia mestus is known commonly as ‘the red cuttle’ because the live animals are often deep reddish in colour. It may be that live S$. grahami are paler than S. mestus, but until an accurate identification of a live specimen it made, this remains to be seen. The cuttlebone is narrower in S. grahami, than S. mestus (Fig. 4a—d). The cuttlebone widths of ten mature male and ten female S. mestus have been measured and found to be: CbWI males 43.0-46.6—-50.2 (SD, 5.1), females 48.5—51.1—52.5 (SD, 1.3). While there is some overlap in the measurements shown above for S. grahami, this comparison may assist in identification. The cuttlebone of S. mestus does not have a thickened, raised and yellowish-ochre inner cone as seen in S. grahami. In addition, the club has 5—6 suckers in each transverse row, while there are usually only four, and rarely five suckers in transverse rows in S. grahami. The arms are relatively longer in S. grahami than in S. mestus (measurements for S. grahami shown in italics): males ALI] >35.3 v. <35.0, ALI2 >39.6 v. <37.6; females ALI1 >36.4 v. <34.4, ALI2 >36.4 v. <34.9, ALI3 >36.6 v. <34.9, ALI4 >37.5 v. <37.1. Finally, S. grahami has a greater number of suckers on the ventral arm pairs than does S. mestus. This is most obvious on arms 4: males ASC4 >/47 vy. <112; females ASC4 >/86 v. <149. (Values for S$. mestus from Reid, unpublished data.) The cuttlebone of S. grahami shares some similarities with that of S. rozella, however, the inner cone is distinctly pink and the sulcus very deep in mature S. rozella (compare Fig. 4a, b with Fig. 4e, f), while the inner cone is yellowish, or ochre- coloured in S. grahami. Juvenies of the two species are very difficult to identify because the sulcus is not well developed in small S. rozella specimens. The dorsal surface of the cuttlebone (with the exception of the outer cone) is usually pinkish in S. rozella, while it is white in small S. grahami specimens. No other differences between juveniles of the two species have been found. ACKNOWLEDGMENTS I wish to thank Ian Loch for the loan of specimens from the Australian Museum and Dermot Henry and Melanie Mackenzie for access to facilities at the Museum Victoria. Proc. Linn. Soc. N.S.W., 123. 2001 172 NEW CUTTLEFISH FROM EASTERN AUSTRALIA REFERENCES Clarke, M.R. (1986). ‘Handbook for the Identification of Cephalopod Beaks’. (Oxford Science Publications: New York). Dallwitz, M.J. (1980). A general system for coding taxonomic descriptions. Taxon 29, 41-46. Dallwitz, M.J., Paine T.A. and Zurcher, E.J. (1993). ‘User’s Guide to the DELTA System: a General System for Processing Taxonomic Descriptions’. (CSIRO Division of Entomology: Canberra). Gray, J.E. (1849). ‘Catalogue of the mollusca in the collection of the British Museum. Part I. Cephalopoda Antepedia’. (British Museum: London). Hanlon, R.T. and Messenger, J.B. (1996).’‘Cephalopod Behaviour’. (Cambridge University Press: Cambridge, UK). Iredale, T. (1926). The cuttle-fish “bones” of the Sydney beaches. (Phylum Mollusca” Class Cephalopoda).—Australian Zoologist 4, 186-196. Lu, C.C. (1988). A synopsis of Sepiidae in Australian waters (Cephalopoda: Sepioidea). In ‘Systematics and Biogeography of Cephalopods’ (Eds N.A. Voss, M. Vecchione, R.B. Toll and M. Sweeney) pp. 159-190. Smithsonian Contributions to Zoology, Number 586. Nixon, M. (1995). A nomenclature for the radula of the Cephalopoda (Mollusca) — living and fossil. Journal of Zoology. London. 236, 73-81. Nixon, M. and Dilly, P.N. (1977). Sucker surfaces and prey capture. In ‘The Biology of Cephalopods’ (Eds M. Nixon and J.B. Messenger) pp. 447-511. Symposia. Zoological Society of London. Vol. 38. Partridge, T.R., Dallwitz M.J. and Watson, L. (1993). ‘A Primer for the DELTA System’. (CSIRO Division of Entomology: Canberra). Reid, A.L. (2000). Australian cuttlefishes (Cephalopoda : Sepiidae): the’ ‘doratosepion’ species complex. Invertebrate Taxonomy 14, 1-76. Roper, C.F.E. and Voss, G.L. (1983). Guidelines for taxonomic descriptions of cephalopod species. Memoirs. National Museum Victoria 44, 48-63. Proc. Linn. Soc. N.S.W., 123. 2001 Silurian Biostratigraphy of the Cadia Area, South of Orange, New South Wales R.B. Rickarps!, I.G. PERcIvAL”’, A.J. SIMpsON*” AND A.J. WRIGHT’. 'Department of Earth Sciences, University of Cambridge, Cambridge CB2 3EQ, UK; *Geological Survey of New South Wales, P.O. Box 76, Lidcombe, NSW 2141; 3Division of Environmental and Life Sciences, Macquarie University, NSW 2109; and *School of Geosciences, University of Wollongong, NSW 2522; *Research Associate, Centre for Palaeobiology and Ecostratigraphy, Macquarie University, NSW 2109. Rickards, R.B., Percival, I.G., Simpson, A.J. and Wright, A.J. (2001). Silurian Biostratigraphy of the Cadia Area, South of Orange, New South Wales. Proceedings of the Linnean Society of New South Wales, 123, 173-191. New Silurian fossil discoveries in the vicinity of Cadia Mine indicate ages younger than shown on recent maps. Limestone, intersected in drill core immediately above an unconformable contact with Late Ordovician volcanics of the mine sequence, yielded an early Wenlock conodont fauna including Pterospathodus amorphognathoides, P. procerus and P. rhodesi, together with Kockelella ranuliformis. A diverse shelly fauna of late Wenlock to early Ludlow aspect, dominated by brachiopods, is present in a slumped mudstone on the mine access road. South of the mine, in Rodds Creek valley, Silurian rocks are shown to occur as infaulted slices along the Werribee Fault. Limestone pods in this area contain conodonts (Coryssognathus dubius) indicative of a Ludlow age; a graptolite fauna from nearby siltstones includes Monograptus flemingii warreni, M. flexilis, Monoclimacis flumendosae flumendosae and Cyrtograptus ex. gr. C. rigidus, and is assigned to the lundgreni- testis Biozone (late Wenlock). The youngest graptolite assemblage (Pridoli) occurs in siltstones, tentatively correlated with the Wallace Shale, exposed in a shallow excavation east of Cadia Mine. This fauna, which includes Dictyonema sherrardae mumbilensis, Acanthograptus aculeatus neureaensis, Pristiograptus shearsbyi, P. cf. P. dubius, Monograptus parultimus minutus, M. microdon aksajensis and M. cf. M. yassensis, is younger than all known graptolite faunas from the nearby Four Mile Creek area, and provides the first Australian record of Monograptus microdon. Manuscript received 19 October 2001, accepted for publication 21 November 2001. KEY WORDS: Silurian, biostratigraphy, graptolites, conodonts, brachiopods INTRODUCTION Cadia Mine, situated 20 km SSW of the city of Orange in the central west of New South Wales (Fig. 1), is a significant producer of gold and copper ore hosted within a Late Ordovician volcanic succession. The source of the mineralising fluids is believed to have been associated with igneous intrusions of latest Ordovician or earliest Silurian age. Overlying Silurian strata are not mineralised and, therefore, have received less attention from geologists, other than in thesis studies and regional mapping projects. Until recently, Silurian sedimentary rocks overlying the mine sequence were assumed to belong to the Cadia Coach Shale, of early-middle Llandovery age (Jenkins 1978), and Proc. Linn. Soc. N.S.W., 123. 2001 174 SILURIAN BIOSTRATIGRAPHY OF CADIA Figure 1. Locality map covering 8630-N Canowindra 1:50 000 (NE corner) and Millthorpe 8731-3-S 1:25 000 (SE corner) topographic sheets, prepared from an aerial photograph of the Cadia mine area supplied by Newcrest Mining Ltd, showing positions of palaeontological sample sites C1885, C1890, C1891, W906, W910, and macrofossil locality on mine access road. (G7 Cadiangullong z MIDDLE BELL LAKE , Muella Junction Pool Bloodwood a” Vospers Pool Wattle Pool i LES Revh Beal BELLS i Rays Pool S Ge, Steves Pool GIDGEE LAKE Nellyvale km Figure 1. Map of Bells Creek showing property boundaries, the eight pools and the terminal salt lakes. MATERIALS AND METHODS Study Area Bells Creek is 23 km long and flows northwest from slightly elevated stony hills towards Cuttaburra Creek, an interdistributory stream from the Warrego to the Paroo River in far north-western NSW. Bells Ck’s path is blocked by quaternary dunes so that it terminates in three salt lakes and a number of pools have developed, largely on bends, in the lower gradient downstream reaches (Fig.1). These pools range in size from 85-500m long and are all less than 1m deep (Table 1). While most contained water for most of 1998, generally they dry for a large portion of each year and perhaps all year during droughts. During 1995 to 2000 their relative persistence was Upper Crescent> Lower Crescent>> Rods> Steves> Vospers> Junction> Wattle>> Rays (all names of local validity only). Observations on Bells Ck during 1998-90 suggest the creek flows after c. 25mm of rain Over a day or two. Such flows are slow and may not reach the lower pools. Larger flows generated by rainfall >50-100mm scour the pools and eventually fill the terminal lakes. Annual rainfall for the area averages 310mm per year, while evaporation is c. 2600mm (Bell 1972; Timms 1997a). Rainfall figures for 1998 are from Bloodwood homestead, a few kilometres from the pools. In 1998 the pools (see Fig 1 and Table 1) first filled after 66mm of rain on 20 and 21 April. Rainfall on 21 June, during 5-23 September and on 11 November produced similar or smaller flows to the April event. A large flow occured during July/early August following 172.5 mm of rain 18-27 July, when the creek ran for 18 days and up to 52 cms deep (R. Barden, pers. comm.). Most pools were low before this major flow (Rays Pool had dried). After this flow, levels receded at different rates, Rays Pool dried between September and November, Steves, Rods, Lower Crescent and Wattle Pools dried between November and January the next year, leaving just Upper Crescent, Vospers and Junction Pools persisting through till January from the April filling. The lower two pools (Vospers and Junction) were aided in their persistence by connection to Gidgee Lake, a terminal salt lake. Proc. Linn. Soc. N.S.W., 123. 2001 195 B.V. TIMMS SUIMOTJIOAO pure [[NJ usyM IWdoq, ‘poyigur Aypenjsed jood pue Apnis 3ulinp payrys pog weons Apues ‘Apnjs Jo Suruurseq ye suOTsUSWTC, 68-LL ¢°8 8I-€ 6 00909-0381 OcOrl ¢°0 81 OLT uorsune V6-SL ¢'8 €c-0 el 00L99-916 C968 9'0 OI OSt stodson L8-EL 18 89¢-79 rol TveI-L9l ILV v0 8 O€T Ope 8-99 CL 8rr-6r I1é SS¢-08 vel L0 LI OOS JUSISOT) JOMO'T 6-TL SL vry-0v SVC ce6-VOl £9C 60 CC Olv quaosarD Jaddy C8-CL CL 796-87 Ie 9SL-90I 8Se v0 cl OCC spod e818 c8 LSI-8€ 86 SSI-IOT vel c0 6 OIl 4 SAVY C8-L9 GL CIC-6€ trl OVI-ve SL 9'0 ¢ 68 S9A}$ osurl uvoul osuel uvoul osuel uvoul .yidoq WIPIM. y3ueT] (N.LA) (osm) (ut) Hd AMpIginy, Ayanjonpuod 9Z1§ ]ood ‘sjood yaaID s]jog IY3Ie Jo somnyeoy [eorMayo pure [eorshyd owiog | 9qQRI, Proc. Linn. Soc. N.S.W., 123. 2001 196 LIMNOLOGY OF CREEK POOLS IN THE PAROO Sampling Sampling commenced 28 April following the filling of the pools around 22 April. Subsequent visits were made on 3 & 21 May, 9 & 27 June, 16 July, 6 August, 26 September, 25 November and 21 January. On each occasion a surface water sample from the deepest area was taken for immediate measurement of temperature, conductivity, pH, and later measurement of turbidity. Instruments used were a mercury thermometer, a HANNA HI 8633 conductivity meter, a HANNA HI 8924 pH meter and a HACH DR/2000 Spectrophotometer method 8237 for turbidity. Depth was read from a graduated staff installed in each pond. Zooplankton was collected with a net of mesh size 159um mounted on a pole and with a rectangular aperture 30 x 15 cm. Collections were made for one minute (2-5 minutes if plankton was sparse) in the deepest area of each pond. In the laboratory species were identified and a random sub-sample of 200 individuals counted. The remainder of the sample was scanned for rare species and these added to the count as 0.1. To simplify the data, percentage occurrences were averaged for the 8-10 samples from each pool and summarized into occurrences in the early (28 April, 3 & 21 May), middle (9 & 27 June, 16 July, 6 August) or late stages of filling (26 September, 25 November, 21 January). Macroinvertebrates were sampled with a rectangular net of aperture 30 x 15 cm and mesh size 1mm. This was swept through the pond for 15 minutes. Animals caught were examined in a white tray and abundance recorded as r = 1 individual in whole collection, x =2-10 individuals, xx = 11-100, xxx =c. 100- c. 1,000, xxxx = c. 1,000 - 10,000, and rarely xxxxx =>c. 10,000. As for zooplankton samples, data were simplified by averaging, this time on a log scale, with the number of ‘x’s converted to integers. Although this method is approximate, it provides a rapid assessment of relative numbers. The notation used for occurrences in the early, middle and late stages of the filling cycle for zooplankton was also used for macroinvertebrates. Statistical methods Similarities in invertebrate community structure among the pools and the average situation for four wetland types (freshwater lakes, hyposaline lakes, claypans, and vegetated pools) at Bloodwood (from Timms and Boulton in press) were compared using a combination of classification and ordination approaches: hierarchial clustering and non- metric multidimensional scaling (NMDS). To ordinate the data, log, (x+1) transformed data was used and the subroutine MDS in PRIMER, employing 50 random starts to minimise the risk of erroneously accepting solutions trapped in local minima (Clarke and Warwick 1994). Spearman Rank Correlations (Zar 1984) were used to assess the influence of various environmental variables on species richness in the ponds. RESULTS Physical and Chemical Features. Recorded water temperatures ranged from 7.8 to 35.1°C, with the average recorded temperature for each visit ranging from c. 10°C in winter to c. 32°C in summer (Table 2). Most pools had fresh water (conductivity <400umS/cm), a slightly alkaline pH and were moderately turbid (200-300 FTU)(Tables 1 and 2). The two saline pools (Vospers and Junction) were different, not only in higher conductivity but their pHs were higher and turbidities lower. In general, the conductivity and pH of pools increased over time whereas turbidity reached a maximum sometime after filling (Table 2). These changes were greatest in Vospers and Junction Pools which commenced as turbid freshwater pools but finished as clear saline pools. Proc. Linn. Soc. N.S.W., 123. 2001 197 B.V. TIMMS 6V 08 8rr 09¢ OLC 10€ c9I cOl ccl OLA ut Aypriqin |, (6661) UOYNog sMoT[oy JUSWIUSISSR,, Se = (ks 8 CSE 69 GiG= HOG I'v C7-AON Gil OI 98 GVGC— Gril 8°61 97-das CL €8 €8 NG =o/E 4) 191 90-3sny Cl £6 6S TIL -6 COI gt-Ing CL 08 pL Gc 28 (onal L7-unt Gil GI: 09 LN AGG Db 6 EI 60-uNf tL 06 OL GOG-BL (nal 1Z-ARIN SL L8 EL CC AEN 102 €0-ARIN 9°9 OOI BL GG.- Or [Li gc-idy w9/Suin Ul sud ds ueoul Hd Ayianonpuoj ~=—s ut ydagq ,O1jeI Joyepoid Dp UE sfood [Te I0J [OOg JUD9NSII_D JOMO'T UI sasuKYo [eUOSeaS -UOU /JOJepoig onjesodurd} 1978 A aed ‘siojoueied duos UI sasuLYO [eUOsRaS °Z IIGUL Proc. Linn. Soc. N.S.W., 123. 2001 198 LIMNOLOGY OF CREEK POOLS IN THE PAROO Aquatic plants During the 1998 filling aquatic plants did not appear till many months after initial filling and were only common during the last third of the filling (summer). Composition varied from pool to pool, with Marsilea drummondii common in Steves Pool, Eleocharis pallens in Rods Pool, Nitella subtilissima in the Crescent pools, and Glossostigma diandrum then Lepilaena bilocularis in Vospers and Junction Pools. Other species present in some pools included Callitriche stagnalis, Diplachne muelleri, Marsilea angustifolia, Mimulus repens, Myriophyllum verrucosum, Ottelia ovalifolia, and Vallisneria gigantea. Zooplankton Thirty-eight taxa, some identified only to generic level, were caught in the eight pools (Appendix 1). This list does not include rotifers, the larger taxa of which were not common in the net zooplankton. The two largest and most persistent pools, Upper and Lower Crescent, had the most species, while the most intermittent pool, Rays, had by far the least (Fig. 2a). Many species were littoral strays appearing only rarely in collections; they were encountered more regularly in the larger pools which had more aquatic plants. Also some eulimentic species, such as Calamoecia spp. and Ceriodaphnia spp., and most conchostracans were restricted to the larger pools, further enhancing their species richness (Appendix 1). Momentary species richness varied in the same pattern between the pools as did cumulative species richness (Fig. 2a). Species turnover percentages (calculated by expressing as a percentage the number of different species since the last visit/number of species present) averaged 35% and were highest in the two pools, Vospers and Junction, which changed from fresh to saline during the study (Fig. 2a). All pools were dominated by Boeckella triarticulata throughout most of their existence while Microcyclops sp. and/or Mesocyclops sp., Daphnia angulata and Moina micrura were dominant at some stage (Appendix 1). The larger pools also had Calamoecia spp important, while the two saline pools had additional species, Apocyclops dengizicus, Daphnia n.sp., Daphniopsis queenslandensis, and various ostracods dominant during their saline phase (Appendix |). Large branchipods were unimportant, except in claypan- like Steves Pool, and also to a lesser extent in the next two pools downstream. Species composition changed during the life of the pools (Appendix 1). Almost invariably the colonizers were various large branchipods, B. triarticulata, Microcyclops sp., Moina micrura, and also Calamoecia lucasi in the larger pools. Except for M. micrura and many large branchipods these persisted into the middle period of the life of the pool and were joined by Mesocyclops sp., Daphnia spp. and sometimes ostracods. The saline pools started differentiating at this stage with the appearance of Daphniopsis queenslandensis, Cyprinotus sp. and other ostracods. The last successional stage varied between the pools, with simplification of species structure to just the core dominants in the smaller pools, the reappearance of M. micrura together with Ceriodaphnia spp., Simocephalus spp., various chydorids and ostracods in the larger pools, and in the two lower pools the appearance of many saline species, particularly ostracods. Ordination of the pool data with averages from habitat types taken from a similar study (Timms and Boulton, in press) suggested the two upper pools had similarities to claypans, three middle pools behaved like freshwater lakes, while the lower two pools had some similarities to hyposaline lakes (Fig. 3a). There was little affinity of the creek pools with vegetated pools despite the appearance of many plants late in their life. Macroinvertebrates Approximately 86 taxa were collected from the eight pools (Appendix 2). This list includes large ostracods, the large cladoceran Simocephalus and large branchipods accounted for in the zooplankton collections, but is incomplete in that benthic invertebrates, particularly chironomids, were inadequately sampled and some of the taxa probably Proc. Linn. Soc. N.S.W., 123. 2001 Junction. = Vospers, and 8 Wattle, 7 Lower Crescent, 6 Upper Crescent, 5 Steves, 2 = Rays, 3 = Rods, 4= Chart 1 Figure 2a. Mean momentary species richness (bars with diagonal stripes) and cumulative species richness (clear bars) of zooplankton in the eight pools. Also shown are species turnover percentage for each pool. Key to pools: 1 (a) Zooplankton B.V. TIMMS a6e}JUa0I198g J8AOUIN]| Salseds saiseds jo siaquinn Yu Proc. Linn. Soc. N.S.W., 123. 2001 200 LIMNOLOGY OF CREEK POOLS IN THE PAROO Numbers of Species € | ov os . Jd . ddd . Cu; s]00d .§ MM .§ Ml .JWvtdda . Wd Species Turnover Percentage Proc. Linn. Soc. N.S.W., 123. 2001 | 09 Sa}eIqaWaAuloIoeW (q) OL [ ueyo ‘uoTjoune [ :sjood 0) Aay “Jood yore Joy a8e}UIdIEd JaAOUIN) SaIdeds are UMOYS p ‘spoy = ¢ ‘skvy = Z ‘SoAaqg = Q ‘JWIOSaID IaMOT= ¢ ‘JUSOSaID Jaddq L eTWeM OSTY “s[Ood 1YBta ay} UT sayeIQOIOAUIOIOLU [LION] JO (Seq IeIJO) ssaUYoLI satoads SANL[NUINS pue (sodis [eUOSeIP YIM Sieq) SsoUYoTI sotdads AreyUSWIOUL URI *GZ BINT Q pure ‘siadso,, B.V. TIMMS 201 Figure 3. Ordination diagrams of the pools (numbers as in Figure 2) and averages for four wetland types (F = freshwater lakes, S = hyposaline lakes, P = claypans, and V = vegetated pools) from Timms and Boulton (in press) based on (a) zooplankton assemblages and (b) macroinvertebrate assemblages. 2 (a) Zooplankton 1.5 (b) Macroinvertebrates Proc. Linn. Soc. N.S.W., 123. 2001 202 LIMNOLOGY OF CREEK POOLS IN THE PAROO represent more than one species. Fifteen species are considered rare as only one or two individuals were found throughout the study. MSR and CSR varied by up to 3x between the larger more persistent pools (Upper and Lower Crescent) and the smaller (Steves and Rays) and saline pools (Vospers and Junction)(Fig. 2b). A few species, all hemipterans, dominate in each pool —Micronecta sp., Agraptocorixa eurynome, and Anisops spp. Other common taxa include Branchinella spp.. Triplectides australis, Eretes australis, Allodessus bistrigatus, Antiporus gilberti, Berosus spp., Aedes sp., Culex sp., Eylais sp., Hydrachna sp., Glyptophysa gibbosa and Gabbia australis. Large branchipods were most common in Steves Pool, Upper and Lower Crescent Pools, adult beetles in these pools and Rods Pool, beetle larvae in Upper and Lower Crescent Pools, and snails in the four central pools (Rods to Wattle). Species composition changed though the life of the pools (Appendix 2). Some species, mainly widespread common ones as listed above, were present whilever there was water in the pool. Others, were present only in the early, middle or late stages in the life of the ponds. Examples of early species include the large branchipods, Berosus spp., Chironomus sp., and Aedes sp., the middle group include Culex sp., many beetles, chironomids and mites, and species appearing late include mayflies, odonates, various bugs such as Ranatra and Hydrometra, caddises, and some beetles. The two saline pools behaved similarly to this model, but with few large branchipods initially and the presence of large saline ostracods in the late stages. Species turnover averaged 28% and was higher in the two saline pools (Fig. 2b). For much of the life of the pools non-predators were more common than predators (Table 2), though predators were comparatively abundant initially and particularly in the drying phases. Ordination of the pools using macroinvertebrate presence and abundance showed the two upper pools were similar to claypans, the next four pools aligned with freshwater pools and also to some extent with vegetated pools and the two lower pools had affinities with hyposaline lakes (Fig. 3b). The match of these two lower pools is the least convincing perhaps because early in their existence they were dominated by freshwater species rather than hyposaline lake species. Vertebrates Tadpoles of at least three species, including Cyclorana platycephala, Notoden bennettii and probably Lymnodynastes tasmaniensis were most abundant in Steves Pool, present in the next four pools, and rare in the lower three pools. No fish were present during the study or have ever been seen over five years of general observations. Waterbirds did not utilize the pools to any great extent. Over 8-10 visits none were ever seen on Steves Pool, or Rays Pools and few were seen on Rods, Wattle, Vospers and Junction Pools. The two Crescent Pools often supported a few Wood Duck, Grey Teal, Hoaryheaded Grebe and uncommonly Pinkeared Duck, Pacific Black Duck, Whitenecked and Whitefaced Herons, and Yellowbilled Spoonbills. DISCUSSION Physical and chemical environment The pools of Bells Creek have many characteristics shared with other wetlands in the Paroo. They range from fresh to saline, clear to turbid, and are alkaline, shallow and intermittent (Timms 1993, 1997a, 1997b, 1999; Timms and Boulton in press). Environmentally the saline pools are similar to hyposaline lakes and the others to smaller freshwater lakes (e.g. Lake Freshwater) and larger Black Box swamps (e.g. Tredega Proc. Linn. Soc. N.S.W., 123. 2001 B.V. TIMMS 203 Swamp, both nearby)(Timms 1997a). Certainly all pools are very different physically and chemically from claypans, vegetated depressions, riverine waterholes, and the more turbid lakes. The pools are also different from other wetlands (except riverine waterholes) in that they may be scoured by creek flows from time to time. While these may wash some organisms out, they also replenish water supplies and extend the existence of the pools as they did for example during 1998. Being on the course of a creek means that at least the larger pools seem to be more reliably present that most other waters in the area, but they are not permanent like the big waterholes along the much larger Paroo River or some of the hydrologically advantaged larger lakes like Lake Numalla (Timms 1999). There are few other creeks in the lower and middle Paroo catchment (within NSW) with similar pools. Jaensch (1999) indicates the possibility of many such creeks in the upper catchment in Qld, but their physical and chemical environment are unknown. Community structure The dominant and common invertebrates of Bells Creek are the same as elsewhere in the Paroo. Unlike many of the wetland types in the Paroo (Timms and Boulton in press), there does not appear to be any taxa endemic to the pools or occurring in greater abundance. The pools are, however, important breeding sites for their inhabitants, especially for corixids, notonectids, beetles, mosquitoes, snails, and amphibians. Compared to other intermittent wetlands nearby, especially claypans and vegetated depressions, these pools were not important for large branchipods during 1998. Those that were present were of mixed origin, some normally living in claypans, others in vegetated depressions, indicating that many were washed in either as eggs, juveniles or adults (Sanders 1999). Monitoring records for Lower Crescent Pool for the period 1988- 2000 (author, unpublished data) suggests that large branchipods are sometimes abundant. Whether they are absent or present depends on further flows once the eggs have hatched — the numerous flows during 1998 would have washed young away. Supporting this intrepretation is the greater abundance of large barnchipods in Steves Pool which is not located on the main channel and their increased abundance in 1998 in pools immediately below it. Mean momentary species richness and cumulative species richness varies widely between the pools (Fig. 2). In Spearman’s rank correlations between these measures of richness and pool area, pool persistence during 1998, average pool permanency 1995- 2000, relative abundance of aquatic vegetation and salinity, suggested all factors, except salinity, contribute to MSR and CSR(Table 3). The first four factors are interrelated (Table 3) and all measure habitat availability, which has often been correlated with the number of species present (e.g. Fryer 1985). Interestingly species richness is not much higher in nearby relatively large freshwater lakes than in the largest pools (areal differences of 10- 100x). Lake Freshwater is 17 ha in area and 96 species of macroinvertebrate have been recorded over 13 years (Timms 1998b and unpublished data). This suggests a pool persisting for a few months and c. 1 ha in area is large enough to support most local species to be expected in that type of wetland. Normally salinity is negatively correlated with species richness (e.g. Timms 1993), but in the two pools that became saline, richness is relatively high because both freshwater and saline species were found as the pools progressed from fresh to saline systems. This is a further expression of the observation that saline lakes that change markedly in salinity during their existence have a remarkable array of invertebrates (Williams et al. 1990; Timms 1998b). According to Jaensch’s (1999) classification of wetlands in nearby south-western Qld based entirely on vegetation, Bells Ck is a wooded watercourse and distinctive from 19 other types. While physicochemical and faunal characteristics of Paroo wetlands typically parallel gross vegetation characteristics (Timms 1999; Timms and Boulton Proc. Linn. Soc. N.S.W., 123. 2001 LIMNOLOGY OF CREEK POOLS IN THE PAROO 204 Table 3. Spearman correlation matrix for the relationships between species richness (momentary and cumulative) and pool size, persistence in 1998, average permanency, vegetation and salinity. ae 5 5 3 5 s = 5” 5” 3 : : : Z Ss) S 2 2 2 > Cumulative SR 1.000 Momentary SR 0.731 1.000 Area of Pool 0.898 0.619 1.000 Persistence 0.778 0.476 0.905 1.000 Permanency 0.778 0.690 0.643 0.571 1.000 Vegetation 0.970 0.786 0.833 0.667 0.810 1.000 Salinit 0.072 0.071 0.262 0.467 0.333 0.000 1.000 123. 2001 N.S.W., Proc. Linn. Soc. B.V. TIMMS 205 in press), correlations apparently break down at the microscale in creeks. In a comparsion of macroinvertebrate presence and abundance with other wetlands in the Paroo, Bells Ck pools do not represent an extra type of wetland, but modifications of distinctive existing types (Timms and Boulton in press). Ordinations (Fig.3) of both the zooplankton and littoral invertebrate components suggest many of the pools (Upper and Lower Crescent, Wattle and possibly Rods) are similar to freshwater lakes of the area. Steves Pool has some affinities with claypans, while the position (claypan-like or freshwater lake-like) of Rods and Rays Pools varies with the group analysed. The two pools that become saline as they age (Vospers and Junction) have some likeness to hyposaline lakes. In conclusion, while the gross environment of Bells Creek is distinctive and homogeneous according to Jaensch’s (1999) classification, the individual pools are somewhat heterogeneous and not distinctive. Succession Species composition changed markedly over time in the pools as conditions changed. Because of the persistence of many of the dominant species during the life of the relative long existence of the pools, succession is not as distinct as in the nearby claypans (Hancock and Timms unpublished data) or vegetated depressions (author unpublished data). Nevertheless, except for the initial appearance of predators as in other Paroo intermittent waters (Timms 1997a; Hancock and Timms unpublished data), there is a change from dominance by filter/collecting feeding groups to dominance by predators, as in claypans and temporary waters in general (e.g. Lake et al.1989). Succession is also not as distinct as in claypans because filter feeding large branchipods are not as important in the pools as in the pans. On the other hand the development of aquatic vegetation during the existence of the pools, means many species associated with vegetation (e.g. cladocerans, ostracods, beetles) appear later in the succession. Succession in the pools is also masked by seasonal changes and the effect of flows through them. Compared with the semipermanent lakes in the vicinity, the appearance of Daphnia spp in the plankton in the middle (winter-spring) stages would be expected, as would the presence of Moina micrura and Ceriodaphnia cornuta in warmer months. Similarly among littoral macroinvertebartes, at least some beetles and mites are summer visitors to the area. The effect of flows is harder to determine, but the appearance of some species (e.g. Culex sp. and Berosus spp.) thought to be colonizers in the middle stages of the existence of the pools would be associated with the large flow in late July and the rejuveniation of the pools. The flows are probably also important in promoting similarity among the pools, despite their differing appearance due to varying development of aquatic vegetation. The first flow is also probably responsible for the low large branchipod numbers as hatchlings would be washed away (see above). The decline in species richness towards the final stages observed in claypans (Hancock and Timms unpublished data), was not as evident in the creek pools, because of the appearance of many species (e.g. odonates, Ranatra dispar) with longer development times in the pools. Also many species were associated with the development of vegetation in the pools in the middle and late stages, a feature absent in the claypans. Species turnover was greater in the zooplankton (35.6% + 6.0) than among the macroinvertebrates (28.1% + 4.4), and the final assemblage contained fewer species than earlier assemblages compared to macroinvertebrates, due largely to the shorter life cycles of zooplankters. These relationships were only partially seen in the claypan successions, because of truncation by early dryness (Hancock and Timms unpublished data). Proc. Linn. Soc. N.S.W., 123. 2001 206 LIMNOLOGY OF CREEK POOLS IN THE PAROO The two pools (Vospers and Junction) that became hyposaline as they matured, had higher mean species turnovers (45% for zooplankton and 34% for macroinvertebrates) than the other pools because of the demise of strictly freshwater species and the appearance of species tolerant to salinity. Not surprisingly initial and final assemblages, except for a few widespread tolerant species, were quite different in these pools (Appendices 1 and 2). As such they are typical of many lakes in the Paroo which are wide ranging in salinity (Timms 1998b). ACKNOWLEDGEMENTS I am greatful to Ron and Sue Barden, former landowners of Bloodwood Station and Ray and Wilga Bremner owners of Muella Station for permission to study and stay on their properties. I am also indebted to Steve Campbell, former manager of Muella Station for assistance, to numerous field assistants, to Andrew Boulton for the ordinations and to Marty Hancock for computing assistance and comments on the manuscript. REFERENCES Bell, F.C. (1972). Climate of the Fowlers Gap - Calindary area. In‘Lands of the Fowlers Gap- Calindary Area, New South Wales.’ Fowlers Gap Arid Zone Research Station Research Series No. 4, pp41-67. (University of New South Wales, Sydney). Boulton, A.J., 1999. Why variable flows are needed for invertebrates in semi-arid rivers. In: ‘A Free-flowing River: The Ecology of the Paroo River.’ (Ed. R.T. Kingsford) pp. 113-128. National Parks and Wildlife Service, Sydney. Clarke, K.R. and Warwick, R.M. (1994).’*Change in marine communities: An approach to statistical analysis and interpretation’. (Plymouth Marine Laboratory: Plymouth). Fryer, G.,1985. Crustacean diversity in relation to the size of water bodies: some facts and problems.’ Freshwater Biology 15, 347-361. Hancock, M. A. and Timms, B.V. (submitted). Ecology of four turbid clay pans during a filling-drying cycle in the Paroo, semi-arid Australia. Hydrobiologia Jaensch, R. (1999). “Wetlands of south-western Queensland.’ (Environment Protection Agency, Brisbane). Kingsford, R.T. (ed) (1999). ‘A free-flowing river: the ecology of the Paroo River’ (NSW National Parks and Wildlife Service, Hurstville). Kingsford, R.T., and Porter, J.L., 1999. Wetlands and waterbirds of the Paroo and Warrego Rivers. In: ‘A Free-flowing River: The Ecology of the Paroo River.’ (Ed. R.T. Kingsford) pp 23-50. National Parks and Wildlife Service, Sydney. Lake P.S., Bayly, I.A.E. and Morton, D.W., 1989. The phenology of a temporary pond in western Victoria, Australia, with special reference to invertebrate succession. Archiv fur Hydrobiologie 115, 171-202. Sanders, P.R. (1999). Biogeography of fairy shrimps (Crustacea: Anostraca) in the Paroo, northwestern Murray- Darling Basin. Honours Thesis, University of Newcastle, NSW. Timms, B.V. (1993). Saline lakes of the Paroo, inland New South Wales, Australia. Hydrobiologia 267, 269- 289. Timms, B.V. (1997a). A comparison between saline and freshwater wetlands on Bloodwood Station, the Paroo, Australia, with special reference to their use by waterbirds. /nternational Journal of Salt Lake Research. 5, 287-313. Timms, B.V. (1997b). A study of the wetlands of Currawinya National Park. A report to the Queensland Department of Environment, Toowomba, Queensland. (University of Newcastle, Newcastle). Timms, B.V. (1998a). Spatial and temporal variation in inundation of wetlands in the Paroo catchment of the Murray-Darling. In ‘Wetlands in a Dry Land: Understanding for Management.’ (Ed. W.D.Williams). pp 51-66. (Environment Australia, Canberra). Timms, B.V. (1998b). Further studies on the saline lakes of the eastern Paroo, inland New South Wales, Australia. Hydrobiologia 381, 31-42. Timms, B.V. (1998c). A study of lake Wyara, an episodically filled saline lake in southwest Queensland, Australia. International Journal of Salt Lake Research 7, \\3-132. Timms, B.V. (1999). Local runoff, Paroo floods and water extraction impacts on the wetlands of Currawinya National Park. In ‘A free-flowing River: the Ecology of the Paroo River’. (Ed. R. T. Kingsford). pp 51- 66. (NSW National Parks and Wildlife Service, Hurstville). Proc. Linn. Soc. N.S.W., 123. 2001 B.V. TIMMS 207 Timms, B.V. and Boulton, A.J. (in press). Typology of arid-zone floodplain wetlands of the Paroo River (inland Australia) and the influence of water regime, turbidity and salinity on their aquatic invertebrate assemblages. Archiv fur. Hydrobiologie Williams, W.D., Boulton, A.J. and Taaffe, R.G. (1990). Salinity as a determinant of salt lake fauna: a question of scale. Hydrobiologia 197, 257-266. Zar, J.H. (1984). ‘Biostatistical Analysis’ (Prentice-Hall: Englewood Cliffs, New Jersey). Proc. Linn. Soc. N.S.W., 123. 2001 LIMNOLOGY OF CREEK POOLS IN THE PAROO 208 x Ke KK * * * * ta KK KK KK KR KK OK KK KR KKK KKK KK OK ta Se Pr Re je goussaid Kx KK KK OK e) {UOISsdONsS JO o8e)s Kx KK KK jsourpunge SPIDAY quownqd sisua1sojoia snjpydasouig preyory vignp resu piuydppoisay sIeg pinusoo viuydvpo1say AddBIIS sisuapuvjsuaanb sisdoiuydvg ‘dsu piuydoq yoqoH vioaloadpiuydvg yoqoyH vjpydan1suo] piuydvg yWoqoH Vvjnsuv viuydvg TYSAOWIIg Muyaig sisdouojvyT ‘ds vuadozyo¢ (uryyosedaq) sna1z18uap sdojaXoody ‘ds sdojokoosapw ‘ds sdojoko0s01y Apeig 1spon] pldaowvjvy7 AjAeg Disaquvd vId20UDIDI UOSWIOY], VIDININADIADIJAyIAOg soytpodados podados tydneu podados ][WH pue rsoueds sisuanpasnv sdoisy, ‘dsu vjjaiayjsauav) ‘ds pluayjsauav) ‘dds sisdoppuwiT ‘dsu snoiz400q ‘ds vippuwry] ‘dds = wjjauryouvig saroadsg “sosuvYO [BUOSROS PLOIG pue dOURLpUNQe d8vI}Ud010d UvOU :s[ood jYSTa dy) UI UOJYyULTdOOZ URaDRISNID “| xIpuaddy N.S.W., 123. 2001 Proc. Linn. Soc. 209 B.V. TIMMS ‘uoTjOUNs= g ‘sIodsoA = / INEM = 9 ‘WWdsaID JaMO'T = ¢ “UddSaI_D Joddy = p ‘spoy = ¢€ ‘skvy = Z ‘S9Ad}§ =] :spood 0} apoo . ‘(Arenuer - Joquiajdag) aye] ‘( ysnsny - sung) o[pprw ‘spood ay) Jo ot] oy) UT (ARI ‘[Edy) Apes = 9 :apoo : “%VSTX = XXXX *%CT - LOL = XXX “OL - VL = ¥¥ “%T - 70 = X “%10> = 1 :2p00 , I X 9 ‘ds pooeso Xx Ge I X q ‘ds poornso Xo eX Sx uw X v ‘ds pooenso 5 uae ju XX Joyyoeq oq vsojnqgo]s siadko0u0814 Me aX I X (uewideyD) vpipuajds siadkoojyKy GRAS. jw XX ‘ds sidkoyay X Xx I XX ‘ds swudkoosajapyy Xx Xx I XX ‘dds siidkonig > Geek wa XX ‘ds snjou1sdk&a Xx I I ‘ds pyjaidk) x Ux Ay X I I ‘ds pisuojauuag YX x I I ‘ds snxosinajd UOTSSa9ONsS souepunqe 8 PR ey ey AC To : JO o8v\s sess ote Pelee ts Je g0uasaid Proc. Linn. Soc. N.S.W., 123. 2001 LIMNOLOGY OF CREEK POOLS IN THE PAROO 210 xX xX x 4 xX xX x ».4 x x xX ».4 ».4 xX 4 x ».4 x x x xX xX 4 x ».4 xX 4 xX 4 xX ».¢ x x xX 4 xX x xX xX 4 x 4 4 x x xX ».4 ca x KK OK xX x Ke KK i —————————— Sc OT Se Vee ¢ ul aouesaid i (EEE C * x KK RK I C = —_—= —_—= = — uOIssa0ons Jo 938%1S XXxxX KH KH K < * KK KK HK KK RK ,sourpungqe ISVIOAYV ‘dds njaau0s1y (sAJaS) ND] pI;npsoonuaz] (sa)stouLIng) sisuandnd xpuvimazy ‘ds sapoonjdiq a}euopo ueiajdostue polyuseprun (SAJaS) wnsnauosyjKsa UOLASDOYJUDX (preA]]LL) supisav Sajsajo1jsny (skJaS) snsojnuuD sajsajo1sny (a8e\sa]) 1p4AVA]]11 SIUPODIOUDUSD J, ‘ds uoaoj]) Jayyoeg eq vsojngo]s siadko0u0s14 (uewideyD) vpipuajds siudxoojuKp ‘ds swudkongq reau ‘dds = snjpydasoung [[@H pue soousdg sisuayp.ssnv sdoly ‘ds snaouaT Aruoyx, snuidsiaavd sisdopouuny (preg) myo. sisdopouuny ‘dsu snoi24009 ‘ds pippuwiy] ‘dds = pypjaluayjsouavn) (Apeig) viupan) pisayjsauav) ‘dds pyjauiyouvig yoos] pruospreyoL parynueprun yooo] pruoydissoys payynueprun wurupayAqeyd pepo poynueprun satoads ‘sosuvyo [BUOSvaS PROG pUk dOULPUNe SO] URIUT :s}OOd JYSIO OY} Ul SayvAqo}IOAUIOIORI\ *7 XIpusddy 123. 2001 N.S.W., LINN. SOc. Proc. B.V. TIMMS Kx KK OK x KK KK x LL x KK RK Kx KK KK 9 ~ KK KKK OK x KK KR KK KK KK OK £ x xX x x x ».4 x x x x x x x ».4 x x xX xX x x x x x ».4 ».4 x x ».4 xX xX »4 x ».4 ».4 ».4 x xX x x x SW. ul douasaid Kx KK OK xX C x KK KK x *K a a * [wo UOISsadoNs Jo o8e1S (6 HH KK KOH KK o * xXx XxX ,souepunge OSPIDAY ‘ds sajsiqk) searey prosnAp dnois lulssaprg oeArey] pros Ap ‘ds snsodyuy seas] prosnAp (RIO) sniojnopuyynu snosisdousays Aveo por Sjpanins snjupyy yeyQ. imoy snsodvsapy dieys 9 snjpsyas snjiydosov7T (1OIZNONUOY) suDsaja snapkyddy (UOSYOIIA) siD1snv Ssajasq JIOAYQ snypjoundisy sajsiqk) YIR[O Y4ag18 snosodyuy (RID) snjvs1481g snssapo]y]y ‘ds snjdyoy syuegq snoypasnd¢ Sapyoa,dis ‘ds pulojon ‘ds s1Ja9aQ uvIo)diuoy pIAOIOIUW porfyuoprun via}diuay pryoA polyyuoprun ‘ds pajawospay] uopuruoyy 10dsip vaypupy preqpuny] imuupiuauaiys sdosiuy Aprexary ypis sdosiuy eH snjoss sdosiuy eH snjosvz109¢ Sdosiuy geH vivjoundiaspd px1s090]dv138y geH suosfysy ox1A o20jdD1SV Apjeyiryy awoucinapxis0s0jdps3sy soroads Proc. Linn. Soc. N.S.W., 123. 2001 LIMNOLOGY OF CREEK POOLS IN THE PAROO x 9 £ orate Mo OX xe =X > aoe Gi x, OX KAY OK X Me mec SX Xx a. ix Xx Xx Xx Xx x X x X Ker AXE OX MM ANS SK Xx Xm = OX. Xx Xx Sal way Si o-eivee ie ul douaso1d x KK OK (6 x Ke KK SER e---- 3 I I I w I I 7 UOIssaoons JO 9388)S Pe ee ay ,souepungqe IdVIODAV uewidoly vsoumnsd piukdoja0yD asNyS siigvjou vikwsaqv]qy soroads paryyuaprun svArey] pryiios 9]190q pruor[NoINd polyyueprun (snoLiqey) sijp.snv sajnauiq ‘ds snaydasdg searey] prayosieds Keaqory, snjpydaatqnjd snayosad¢ AYs[nyos}oyy s7jv.ysnv snyroipxH] q ‘ds umouyun ovary pryrydorpAy e ‘ds umouyun ovary pryrydoipAy ‘ds snpiydospay searey prytydoipAy ‘dds snsosag sere] pytydorpAy WOWAYIO,P snjpuisivuiu sndojousajy (Ava ov) Snanusdd snudovivg orreunte, Duidsiaasg snpiydoupay Ava TORN, SUSI] SAADIO]AH (uinqyor[g) snjpad«]9 saapyr0] aH] Aeaporyy, Sdao1nopul snayooug (uinqyoR[g) sisuaida snay2oug ‘dds snsosag ‘ds snupdydnisy snuds palyQuaprun seal] prosyAp ‘ds sajoaudypjd Jearey plosynAp ‘ds snjupyy dear] prosnAp ‘ds snuodpsapw searey prosnAp ‘ds snapdyddy dearey prosyAp ‘ds sajaaq Bare] plosnAp so1sadg 123. 2001 Linn. Soc. N.S.W., Proc. ioe) AN B.V. TIMMS ‘uonounr= g ‘siodso, = / INVA = 9 ‘UIOSIID IOMOT = ¢ 4WUdDSIID Joddy = p ‘spoy = ¢ ‘skvyY = Z ‘SOAS =] :sfood 0} apoo ¢ ‘(Arenuer - Jaquiajydas) aye] “( isnsny - une) apppru ‘sfood ay) Jo oy1] oy) UT (Avy ‘[LIdy) Ayres = 9 :apod Z aouepunge URoUl 30] [G¢°7< = XXXX ‘CZ - ['] = XXX ‘T - 17'0 = XX ‘Z'0 - 90'0 = X ‘S0'0> = 1 :8pod | eax SX jus SVSUY 2 SUIRPY 1qWOdIMaU DJAsAOPIS] Rei Cane Crue Gal iP male Gere eee) XXX (pinoy) vsogqis vskydojdd]}H Xo Seen ee Xen no jus XXX UOAIT, S7]D4sSnDDIGGDyH KENT Sexe. SEX I I ‘dds = pisauuiyT X X ju I ‘dds = puyopapay] Xo SX NGS NO eX OX eee) XX ‘dds sinjaq XO Xe XS EXO XS ex [ule me ‘dds snjuopojdiq XS" TENS « SN BX, jw x ‘dds snanuaisy SN Key X I J aval] prpesAd poynuoprun Mom x Nene Kapok 3 a dwAIL] pluege) polyynuoprun Xe x I J dvAIe] prAwoNe.s polynusprun igen BOG 6) I VAIL] Plolpnd poynuoprun X Xe gee. uw I q ‘ds sajaydouy Xe Xe CX jw I eds sajaydouy Node = Nae EX XS EX Xx uw XX ‘ds xajnd Nae RNS UNG NS SEGRE Xx =) XX ‘ds sapay oe ae 8) I AIL] prluosodo}je.199 poryynusprun Nala Xie eX ae Kee jw Xx asnyg 4afignu wnjipadi{jog Mien ) I ‘ds snsupjcuvjosvg I I ‘ds sadipuajosziq Kel EX. ul I JaJJloy wnssopiasiss snuouosyIdosK) Xx 3 J q:ds snwouosmy) X X wo I eds snwouosy) UOISSo9ONS oourpunge Bey th Soho CRE hs = € ; Jo 93815 SPIDAV eoigeds £ ul douasoid Proc. Linn. Soc. N.S.W., 123. 2001 A New Marattialean Fern from the Middle Triassic of Eastern Australia JOHN A. WEBB Department of Earth Sciences, La Trobe University, Bundoora, Victoria 3086 john.webb @ latrobe.edu.au Webb, J.A. (2001). A new marattialean fern from the Middle Triassic of eastern Australia. Proceedings of the Linnean Society of New South Wales 123, 215-224. The marattialean genus Marantoidea Jaeger is revised, and Danaeopsis Heer is shown to be a junior synonym of Marantoidea. Of the species formerly assigned to Danaeopsis, those that can be confirmed as marattialean are transferred to Marantoidea; many of the others are probably pteridosperms. The new species Marantoidea acara is described from the Middle Triassic Toogoolawah Group and Nymboida Coal Measures of eastern Australia. It is distinguished by the small size of its sporangia. The Triassic distribution of marattialeans in Gondwana indicates that these ferns had a similar temperature preference in the past to that today, and have probably had a megathermal-mesothermal distribution throughout their history. Manuscript received 10 February 2001, accepted for publication 18 July 2001. KEYWORDS: Palaeobotany, Pteridophyta, Marattiales, Middle Triassic, Eastern Australia. INTRODUCTION The marattialean ferns can be traced back to the Carboniferous, and they have a more or less continuous fossil record from then to the present (Taylor and Taylor 1993). There are 6 extant genera, two of which, Angiopteris and Marattia, grow in northeast Australia (Camus 1998). In the Mesozoic of the Northern Hemisphere (Laurasia), marattialean ferns were often abundant and formed a significant element of these floras, e.g. Late Triassic of northern China (Chen et al. 1979), Early Jurassic of Japan (Kimura et al. 1992), and Middle Jurassic of Yorkshire (Harris 1961; Hill 1987). The Middle-Late Triassic Danaeopsis-Symopteris flora of northern China is partly named after the marattialean fern Danaeopsis (Kimura and Ohana 1990). By contrast, marattialeans were a minor component of Mesozoic Gondwanan floras. They were rare in the Jurassic and Cretaceous, e.g. there are only 2 records, one of them doubtful, from Australian strata of this age (Dettmann and Clifford 1991). In the Triassic of Gondwana marattialeans were more common, but apart from a few localities where they were abundant, they were only ever a small element of these floras. A number of Triassic marattialean genera have been recorded in Gondwana: one from South Africa (Anderson and Anderson 1985), four from South America (Herbst 1977a, 1977b, 1988), and seven from Australia: Asterotheca (Herbst 1977a, Holmes this volume), Marattiopsis (Playford et al. 1982), Rhinipteris (Holmes this volume), Danaeopsis (Walkom 1917, 1928), Ogmos (Webb 1983), Eboracia (Playford et al. 1982), and Herbstopteris (part Rienitsia of Herbst Proc. Linn. Soc. N.S.W., 123. 2001 216 NEW TRIASSIC MARATTIALEAN FERN 1977b; Holmes this volume), although the affinities of the last four are tentative (Dettman and Clifford 1991, and disucssion below). Danaeopsis is reassessed in this paper, which describes a new species of marattialean fern from the Middle Triassic of eastern Australia. GEOLOGICAL SETTING In southeast Queensland and northeast New South Wales there are two Middle Triassic basins, the Esk Trough and Nymboida Basin respectively. The Esk Trough contains up to 2350 m of nonmarine conglomerate, sandstone, shale and minor coal, with interbedded andesitic lavas and tuffs (Cranfield et al. 1976). These strata are divided into the Esk Formation, Neara Volcanics and Bryden Formation, there being some degree of lateral equivalence between all three. Palynofloras from drillholes in the southern portion of the basin indicate that the Esk and Bryden Formations are Anisian-Ladinian in age (de Jersey 1975; Helby et al. 1987), confirmed by radiometric dates from lavas within the sequence (236-242+5 Ma, Webb 1982; probably Anisian using the time-scale of Young and Laurie 1996). The Esk Trough strata thin abruptly on the eastern side of the basin against the West Ipswich Fault (O’Brien et al. 1994). The Nymboida Basin lies to the east of the southern extension of the West Ipswich Fault, and is, therefore, probably separate from the Esk Trough; cover sediments of the Late Triassic-Jurassic Clarence-Moreton Basin obscure the relationship between the two Middle Triassic basins. The Nymboida Basin contains the Nymboida Coal Measures, approximately 1070 m of nonmarine conglomerate, sandstone and shale with interbedded basic volcanics (McElroy 1963, Holmes 2000); the uppermost, thickest unit is the Basin Creek Formation. Attempts to extract identifiable palynofloras from these sediments have been fruitless (de Jersey 1958); the palynomorphs are partly carbonised due to the high temperatures that have affected the coal measures (VR 0.95-0.97; Russell 1994). The macroflora has been correlated with that in the Esk Trough strata, and therefore regarded as Middle Triassic (Flint and Gould 1975). This is verified by radiometric dating of an interbedded basalt flow as probably Anisian (230.7+0.4 Ma; Retallack et al. 1993, using the time-scale of Young and Laurie 1996). TAXONOMY Genus Marantoidea Jaeger 1827 1827 Marantoidea Jaeger, p. 28 1864 Danaeopsis Heer, p. 54 Type species Marantoidea arenacea Jaeger 1827, from the Triassic of Germany; by original designation. Diagnosis (Expanded from Schimper 1869) Fronds large, pinnate; pinnae long and lanceolate; secondary veins diverge from midrib at very acute angle, then immediately curve away and run towards the margin at 60°-80° to midrib; each secondary vein usually bifurcates close to the midrib, may fork again and then anastomose near pinna margin. On abaxial side of fertile specimens, biseriate sori, arranged parallel to the lateral venation, extend from midrib to margin; sporangia small, free, circular or elliptical in outline, with longitudinal furrow marking line of dehiscence. Discussion There has been some confusion over the correct name and circumscription of this genus. Jaeger (1827) erected the genus and species Marantoidea arenacea, which Proc. Linn. Soc. N.S.W., 123. 2001 217 WEBB A J IOs juaoe[pe ugemjaq suroeds uorssaidap [eorde (MAF) OZ 1g | [evorsea90 (1918) FT 01 ZI - 21R S191 01 ee S101 01 an quanbay ZI} OI - [euoIse990 reas [euose900 [euorseo00 SOsOWO]seue £1 01 (wu 0] /SUI9A) Aytsuop uoneuaA ATepuosas be ae sak Te yuaumoap ude aidoosiseq a (wo) I 8I< OI< 0 ysuay] “xew ‘(Ryep OU = “p'U) aIqQuTTeAR SI [eLIOJeUL OpIVJOJ YOIYA\ J0J satoads vaprojupivpy Jo uostredu0| “| aqR I, CE} CZ SE 0} 02 Sy 1 CI 9 01 LT 8S 0} £7 OE % SC CC 91 ST G9 01 CP E IPI “XPUL erpesny SISSCLLL, 2[PPUN PISSNY disseny, Ape UR|syeze yy OISSIL], buly) OISSPLL], IVT RunuasIy IISSeL], eT RISsny ‘Uapams IISSeL], eT elysny OISSPLL], O18] adoing OISseL], 18 -2[PPI] (Ayteooy ‘ade) goue1ns00 ‘Aou ‘ds qqany vivoD (TL61) AojAed 2 eAoUeIY “AOU “QUIOD ([ 16] AOTARG 2 vAOWeIYY) vIIUoYyjad "pw (7661) YE “AOU “QUIOD (Z66] YG) vipusinWa ‘p (TS61) 22°] 9 2zS (QL61) UM % ORS] ‘TT “(9S61) 9zS (C961) O8SL “(9E6T) Ud “AOU “GUID (JEGT UPd) 14j/DY ‘W (8861 “QLL61) 18q19H "AOU ‘QUIOD ([Z6] APH) vpunsaf w (661) 81oqnoN ‘(1Z61) tH “AOU “QUOD ([ ZH] I[PH) vpunsaf Ww (6061) Jassery “AOU “QUIOD (6(6] Jossery X9 CQR] INS) sisuazun] “pW (7061) ipreyINeT] (6981) Jeduityog ‘(p98T) YuoyoS “AOU “qUIOD (/Z8] Ja8eef) vaovUAID ;W SQOUDIOJOI pue saroads ., 123. 2001 Proc. Linn. Soc. N.S. 218 NEW TRIASSIC MARATTIALEAN FERN Presl (1838) renamed Taeniopteris marantacea. In 1845 Presl erected Danaeopsis for a species of modern fern; Heer (1864) used the same name for a new genus with the type species D. marantacea (Presl) Heer, but this is clearly a later homonym. Krasser (1909) and Halle (1921) argued that M. arenacea should be discarded in favour of Danaeopsis marantacea because the former name had fallen into disuse, the type species of Marantoidea 1s sterile whereas that of Danaeopsis is fertile, and the name Marantoidea was derived from the supposed resemblance of the fossil to the extant monocot Maranta. Nevertheless, under Article 51 of the International Code of Botanical Nomenclature (Greuter et al. 1993), “an alteration of the diagnostic characters or of the circumscription of a taxon does not warrant a change in its name”, the name Marantoidea is valid and has priority over later homonyms. A number of species of Danaeopsis and Marantoidea are known, but only seven of these, for which fertile pinnae have been described, can be validly assigned to Marantoidea (Table 1). The differences between many of these species are small, and further study could show some of them to be synonymous. It should be noted that of the five species of Danaeopsis described by Brik (1952), only one, D. angustipinnata Brik 1952, includes fertile material, but the holotype of this species is sterile; the fertile specimens differ in pinna width and vein density from the holotype, and are referable to D. emarginata Brik 1952. Of the other species previously referred to Danaeopsis, D. sp. cf. cacheutensis Kurtz 1921, described by Frenguelli (1938), probably belongs to a new marattialean genus, but better preserved material is necessary before this can be erected. D. cacheutensis Kurtz 1921, D. rajmahalensis Feistmantel 1877, D. hughesii Feistmantel 1882 and D. gracilis Lele 1962 are pteridosperms (Lele 1962, Herbst 1977b, Retallack 1977). Probably three separate species have been described as D. hughesii; the Indian species (Feistmantel 1882; Lele 1962) is different from the Australian specimens (Walkom 1917, 1928), and both differ from the material from China and Kazakhstan (P’an 1936, Brik 1952). The two species described by Prinada and Turutanova-Ketova (1962), D. rarinervis Turutanova-Ketova 1962 and D. taeniopteroides Turutanova-Ketova 1962, are both sterile and could be cycadophytes. Amongst the extant marattialean ferns, Marantoidea is closest to Archangiopteris, but differs in that the latter has well-spaced sori that do not extend to either the midrib or pinna margin (Bower 1926). Marantoidea acara Webb sp. nov. Figs 1A-D, 2 Diagnosis Anastomoses in secondary veins rare; sporangia very small (0.2 mm x 0.15 mm), with no spacing between adjacent sori. Description Fronds are relatively large, with a short petiole and a rachis 6-7 mm wide, from which opposite to subopposite pinnae diverge at angles of 45°-60° every 3-5 cm (Figs 1A,B). Individual pinnae are 22-32 mm wide and at least 18 cm long and taper gradually to an acute apex. At the base of each pinna the acroscopic pinna margin is markedly contracted, whereas the basiscopic margin is decurrent on the rachis (Figs 1B,2). Midveins of pinnae are up to 3.5 mm wide; secondary veins diverge at very acute angles but curve away almost immediately and run fairly straight and parallel to the margin at an angle of 60°-80° to the midrib. Most secondary veins fork once close to the midrib, rarely to occasionally a second time, in which case they anastomose with an adjacent vein near the margin. Density of venation is 12-14 veins per 10 mm in sterile fronds. In fertile material the veins are denser, 18-20 per 10 mm. On fertile pinnae the abaxial surface of the leaf is covered with small (0.2 x 0.15 mm) ellipsoidal sporangia (Fig. 1C), with long axes perpendicular to the lateral venation. Each sporangium has a prominent longitudinal slit, and no visible apical depression (Fig. 1D). The sporangia are relatively well-spaced and are arranged in rows roughly parallel Proc. Linn. Soc. N.S.W., 123. 2001 J.A. WEBB 219 Figure 1. Marantoidea acara sp. nov., UQF71052, from UQL4224, holotype, fertile frond. A,B: adaxial sur- face, note decurrent pinnae bases. C,D: abaxial surface, note arrangement and morphology of sporangia. Proc. Linn. Soc. N.S.W., 123. 2001 220 NEW TRIASSIC MARATTIALEAN FERN Figure 2. Marantoidea acara sp. nov., AMF 113388, from UQL3745, sterile frond; note scale bar in top left corner (divisions in cm). Proc. Linn. Soc. N.S.W., 123. 2001 J.A. WEBB 221 to and overlying the venation, but are not obviously differentiated into sori; spacing between rows 1s fairly uniform. Holotype UQE 71052, Palaeontological Collection, Queensland Museum (formerly housed in Department of Earth Sciences, University of Queensland). Type Locality UQL 4224, hillside, 594893 Somerset Dam 1:50,000 sheet, Bryden Formation, Toogoolawah Group, Esk Trough, Queensland; Anisian — Ladinian. Other Material UQEF 71053 from the type locality and AMF 113387-113390, AMF 113449 from UQL 3745, Coal Mine Quarry, Basin Creek Formation, Nymboida Coal Measures; Anisian — Ladinian. Etymology acares - Greek for small or tiny. A reference to the small sporangia. Discussion Sterile foliage of M. acara is readily identifiable, despite the lack of sporangia, by its similarity in size, shape and venation to fertile fronds. The adaxial surfaces of fertile specimens show these features clearly. Sterile fronds of M. acara resemble some other Marantoidea species (Table 1), but fertile specimens are readily distinguished by the considerably smaller sporangia. PALAEOCLIMATIC SIGNIFICANCE Living marattialeans are exclusively tropical and subtropical, i.e. megathermal- mesothermal using the broad-scale temperature regime terminology of Nix (1982). In eastern Australia the most widely distributed living species extends to only ~ 30°S (Camus 1998). In the Middle Triassic marattialeans had a much broader distribution; the Esk Trough and Nymboida Basin lay at about 60°S (Veevers 2000), and the other Gondwana locations where marattialeans have been recorded (El Tranquilo Basin, South America; Karoo Basin, South Africa) were at similar palaeolatitudes. Overall, the Triassic climate was considerably warmer than at present (Frakes 1979), and in the Middle Triassic of the northern hemisphere, mesothermal floras characterised by pteridophytes and cycadophytes extended to 70°N (Ziegler et al. 1993). The Middle Triassic climate at 60°S in eastern Australia was probably humid and mesothermal, based on the diverse flora, containing many large-leaved species (Balme et al. 1995). The extensive development of fluvial sedimentation, with some coal measures (Yeates et al. 1986), indicates a more humid climate and higher water tables than in the Early Triassic, when red-beds were deposited across eastern Australia. Thus the occurrence of marattialeans in these relatively high latitude eastern Australian Triassic floras is consistent with the humid megathermal- mesothermal distribution of extant representatives of this family. In the Late Triassic and Jurassic of the northern hemisphere, the highest diversity, humid mesothermal floras containing numerous fern genera, including marattialeans, lay at 40-50°N (Ziegler et al. 1993). The relative scarcity of Gondwanan marattialeans of this age may reflect the fact that there are relatively few megafloral localities known from equivalent southern palaeolatitudes. Proc. Linn. Soc. N.S.W., 123. 2001 222, NEW TRIASSIC MARATTIALEAN FERN ACKNOWLEDGEMENTS Keith Holmes kindly provided access to specimens of M. acara from Nymboida, and stimulated me to write this paper, which would have otherwise languished unpublished. Geoff Playford supervised my PhD thesis on Triassic megafloras, and I am grateful for his encouragement and assistance. REFERENCES Anderson, J.M. and Anderson, H.M. (1985). ‘Palaeoflora of southern Africa. Prodromus of South African megafloras Devonian to Cretaceous’. (A.A. Balkema: Rotterdam). Balme, B.E., Kershaw, A.P. and Webb, J.A. (1995). Floras of Australian coal measures. In ‘Geology of Australian Coal Basins’ (Eds C.R. Ward, H.J. Harrington, C.W. Mallett and J.W. Beeston, J.W.), Geological Society of Australia Coal Geology Group Special Publication 1, 41-62. Bower, F.O. (1926). “The ferns (Filicales), vol 2’. (Cambridge University Press: Cambridge). Brik, M.I. (1952). ‘Iskopaemaya flora i stratigrafiya nizhnemezozoyskikh otlozheniy Basseyna R. Ilek’. (Gosgeolizdat: Moscow). Camus, J.M. (1998). Marattiaceae. In ‘Flora of Australia, Volume 48, Ferns, gymnosperms and allied groups’, pp. 109-112. (CSIRO: Melbourne). Chen Y., Duan, S. and Zhang, Y. (1979). New species of Late Triassic plants from Yanbian, Sichuan. I.’Acta Botanica Sinica 21, 57-63. Cranfield, L.C., Schwarzbock, H. and Day, R.W. (1976). Geology of the Ipswich and Brisbane 1:250,000 sheet areas. Geological Survey of Queensland, Report 95, 1-176. De Jersey, N.J. (1975). Miospore zones in the Lower Mesozoic of southeastern Queensland. In “Gondwana Geology. Third Gondwana Symposium, Canberra, 1973’ (Ed. K.S.W. Campbell), pp. 159-172. (Australian University Press: Canberra). De Jersey, N.J. (1958). Macro- and microfloras of northeastern New South Wales. Journal and Proceedings of the Royal Society of New South Wales, 92, 83-89. Dettmann, M.E. and Clifford, H.T. (1991). ‘The ecology and evolution of Australian fossil ferns: implications for fossil fuels’. Commonwealth of Australia National Energy Research, Development and Demonstration Program. End of Grant Report Number 1184, 1-178. Feistmantel, O. (1877). Fossil flora of the Gondwana System. Jurassic (Liassic) flora of the Rajmahal Group in the Rajmahal Hills. Palaeontologia Indica, series 2, 1(2), 53-162. Feistmantel, O. (1882). Fossil flora of the Gondwana System. The fossil flora of the South Rewa Gondwana Basin. Palaeontologia Indica, series 12, 4(1), 3-52. Flint, J.C.E. and Gould, R.E. (1975). A note on the fossil megafloras of the Nymboida and Red Cliff Coal Measures, southern Clarence-Moreton Basin, N.S.W. Journal and Proceedings of the Royal Society of New South Wales 108, 70-74. Frakes, L.A. (1979). ‘Climates throughout geologic time’. (Elsevier: Amsterdam). Frenguelli, J. (1938). La florula Jurasica de Paso Flores. Revista Museo La Plata, new series 1, 67-108. Greuter, W., Barrie, F.R. and McNeill, J. (1993). International Code of Botanical Nomenclature (Tokyo Code). Regnum vegetabile 131, 1-389. Halle, T.G. (1921). On the sporangia of some Mesozoic ferns. Arkiv Botanik 17, 1-28. Harris, T.M. (1961). ‘The Yorkshire Jurassic flora. I. Thallophyta-Pteridophyta’. (British Museum (Natural History): London). Heer, O. (1864). ‘Die Urwelt der Schweiz, Part 1’. (F. Schulthess: Zurich). Helby, R., Morgan, R. and Partridge, A.D. (1987). A palynological zonation of the Australian Mesozoic. Association of Australasian Palaeontologists, Memoir 4, 1-94. Herbst, R. (1977a). Sobre Marattiales (Filicopsida) triasicas de Argentina y Australia, Parte 1. El genero Asterotheca. Ameghiniana 14, 1-18. Herbst, R. (1977b). Sobre Marattiales (Filicopsida) triasicas de Argentina y Australia, Parte 2. Los generos Danaeopsis y Rienitsia. Ameghiniana 14, 19-32. Herbst, R. (1988). La Flora triasica del Grupo El Tranquilo, Provincia de Santa Cruz (Patagonia). Parte II. Filicopsida. Ameghiniana 25, 365-379. Hill, C.R. (1987). Jurassic Angiopteris (Marattiales) from North Yorkshire. Review of Palaeobotany and Palynology 51, 65-93. Holmes, W.B.K. (2000). The Middle Triassic megafossil flora of the Basin Creek Formation, Nymboida Coal Measures, N.S.W., Australia. Part 1: Bryophyta, Sphenophyta. Proceedings of the Linnean Society of New South Wales 122, 43-68. Holmes, W.B.K. (2001). The Middle Triassic megafossil flora of the Basin Creek Formation, Nymboida Coal Measures, N.S.W., Australia. Part 2. Filicophyta. Proceedings of the Linnean Society of New South Wales 123. Jaeger, G.F. von (1827). ‘Ueber die Pflanzenversteinerungen welche in dem Bausandstein von Stuttgart vorkommen’. (J.B. Metzler’schen Buchhandlung: Stuttgart). Proc. Linn. Soc. N.S.W., 123. 2001 J.A. WEBB 223 Khramova, S.N. and Pavlov, V.V. (1971). Nekotoryye paporotniki iz verkhnetriasovykh otlozhenty Timano- Pechorskoy oblasti. Nauchno-Issledovatelskii Institut Geologii Arktiki, Uchenye Zapiski, Paleontologiia i Biostratigrafiia 32, 71-77. Kimura, T., Lebedev, E.L., Markovich, E.M. and Samylina, V.A. (1992). Macroflora of eastern Asia and other circum-Pacific areas. In ‘The Jurassic of the Circum-Pacific’ (Ed. G.E.G. Westermann)’ World and Regional Geology 3, 311-316. Kimura, T and Ohana, T. (1990), Triassic-Jurassic plants in Japan. In ‘Pre-Cretaceous terranes of Japan’ (Eds K. Ichikawa, S. Mizitani, I. Hara, S. Hada, and A. Yao), Publication of IGCP Project 224, 371-379. Krasser, F. (1909). Die diagnosen der von Dionysius Stur in der obertriadischen Flora der Lunzerschichten als Marattiaceenarten unterschiedenen Farne. Sitzungsberichte Kaiserliche Akademie der Wissenschaften in Wien, Mathematisch-Naturwissenschaftliche Klasse 118, 13-43. Kurtz, F. (1921). Atlas de plantas fosiles de la Republica Argentina. Actas Academia Nacional de Ciencias en Cordoba 7, 129-153. Lele, K.M. (1962). Studies in the Indian Middle Gondwana flora, 2. Plant fossils from the South Rewa Gondwana Basin. Palaeobotanist 10, 69-83. Leuthardt, F. (1904). Die Keuperflora von Neuewelt bei Basel, Teil 2. Kryptogamen. Abhandlungen Schweizer Palaeontologische Gesellschaft 31, 25-47. Li, P.-C., Tsao, C.-Y. and Wu, S.-C. (1976). ‘Mesozoic fossils from Yunnan, China, Fasc. 1. Mesozoic plants from Yunnan’, pp. 87-160. (Institute of Geology and Palaeontology Academia Sinica: Nanking). McElroy, C.T. (1963). The geology of the Clarence-Moreton Basin. Geological Survey NSW, Geology Memoir 9, 1-172. Neuberg. M.F. (1959). Paleobotanischeskogo obosnovanie triasavhky uglenosnykh otlozheniy Perchorskogo Bassenya. Doklady Akademiya Nauk S.S.S.R. 127, 681-684. Nix, H. (1982). Environmental determinants of biogeography and evolution in Terra Australis. In ‘Evolution of the flora and fauna of arid Australia’ (Eds Barker W.R. and Greenslade, P.J.M.) pp. 47-66 (Peacock Publications: Frewville). O’Brien, P.E., Korsch, R.J., Wells, A.T., Sexton, M.J. and Wake-Dyster, K. (1994). Structure and tectonics of the Clarence-Moreton Basin. Australian Geological Survey Organisation Bulletin 241, 195- 216. P’an, C.H. (1936). Older Mesozoic plants from North Shensi. Palaeontologia Sinica, series A 4(2), 1-76. Playford, G., Rigby, J.F. and Archibald, D.C. (1982). A Middle Triassic flora from the Moolayember Formation, Bowen Basin, Queensland. Geological Survey of Queensland Publication 380, 1-52. Presl, K. (1838). In von Sternberg, C., “Versuch einer geognostich-botanischen Darstellung der Flora der Vorwelt, 2 vols’. (Leipzig and Prague). Presl, K. (1845). ‘Supplementum Tentaminis Pteridographica’. (Prague). Prinada, V.D. and Turutanova-Ketova, A.I. (1962). Triasovie khvoshchi i paporotiniki bashkiri. Paleontologicheskiy Zhurnal 1962 (3), 111-122. Retallack, G.J. (1977). Reconstructing Triassic vegetation of eastern Australasia: a new approach for the biostratigraphy of Gondwanaland. Alcheringa 1, 247-277. Retallack, G.J., Renne, P.R. and Kimbrough, D.L. (1993). New radiometric ages for Triassic floras of southeastern Gondwana. In ‘The non-marine Triassic’ (Eds S.G. Lucas and M. Morales), New Mexico Museum Natural History Science Bulletin 3, 415-418. Russell, N.J. (1994). A palaeogeothermal study of the Clarence-Moreton Basin. Australian Geological Survey Organisation Bulletin 241, 237-276. Schenk, A. (1864). Beitraege zur Flora der Vorwelt. Paleontographica 11, 296-308. Schimper, W.P. (1869). ‘Traite de Paleontologie Vegetale, Vol. 1’. ( J.B. Bailliere: Paris). Sze, H.C. (1956). Older Mesozoic plants from the Yenchang Formation, northern Shens1. Palaeontologia Sinica, new series A 54, 111-217. Sze, H.C. and Lee, H.H. (1951). Notes on a Rhaetic species Danaeopsis fecunda Halle from the Yenchang Formation of Kansu. Scientific Records (Academia Sinica) 4, 85-91. Taylor, T.N. and Taylor, E.L. (1993). ‘The biology and evolution of fossil plants’. (Prentice-Hall: New Jersey). Tsao, C.Y. (1965). Fossil plants from the Siaoping Series in Kaoming Kwangtung. Acta Palaeontologica Sinica 13, 510-539. Veevers, J.J. (Ed.) (2000). ‘Billion-year earth history of Australia and neighbours in Gondwanaland’. (Gemoc Press: Sydney). Walkom, A.B. (1917). Mesozoic floras of Queensland, Part 1 (continued). The flora of the Ipswich and Walloon Series. (c) Filicales etc. Geological Survey of Queensland Publication 257, 1-67. Walkom, A.B. (1928). Fossil plants from the Esk district of Queensland. Proceedings of the Linnean Society of New South Wales 53, 458-468. Webb, J.A. (1982). Triassic radiometric dates from eastern Australia. In ‘Numerical dating in stratigraphy’ (Ed. G.S. Odin) pp. 515-521. (John Wiley: New York). Proc. Linn. Soc. N.S.W., 123. 2001 224 NEW TRIASSIC MARATTIALEAN FERN Webb, J.A. (1983). A new plant genus, possibly a marattialean fern, from the Middle Triassic of eastern Australia. Memoirs Association Australasian Palaeontologists 1, 363-371. Yeates, A.N., Mulholland, S., Helby, R., Price, P.L., Day, R.W., McKellar, J.L. and Webb, J.A. (1986). ‘Correlation chart of Australian Triassic units’. BMR and AMIRA Palaeogeographic Maps Project. Young, G.C. and Laurie, J.R. (1996). ‘An Australian Phanerozoic timescale’. (Oxford University Press: Melbourne). Ziegler, A.M., Parrish, J.M., Yao, J., Gyllenhaal, E.D., Rowley, D.B., Parrish, J.T., Nie, S., Bekker, A. and Hulver, M.L. (1993). Early Mesozoic phytogeography and climate. Philosophic Transactions of the Royal Society of London B 341, 297-305. Proc. Linn. Soc. N.S.W., 123. 2001 Diagenesis of the Organic Matrix in Anadara trapezia During the Late Quaternary: Preliminary Findings MArGARET J. WHITELAW!, Barry D. Batts!, CoLiIn V. MuURRAY-WALLACE* AND CHRISTOPHER R. McRAE! ‘Department of Chemistry, Macquarie University, North Ryde, NSW 2109; ?School of Geosciences, University of Wollongong, Wollongong, NSW 2522 Whitelaw, M.J., Bats, B.D., Murray-Wallace, C.V. and McRae, C.R. (2001). Diagenesis of the organic matrix in Anadara trapezia during the Late Quaternary: Preliminary findings. Proceedings of the Linnean Society of New South Wales 123, 225-234. The concentration of amino acids found in the soluble organic matrix of modern and fossil shell of Anadara trapezia is quantified in this study. Results indicate that 91% of the total amino acids present in the soluble organic matrix are lost within 4,000 years, at an average rate of 3.25 x 10° pmol/ug per year. Over the next 2,000 years a further 6.3% are lost at an average rate of 4.49 x 10% pmol/ig per year and for a further 119,000 years 0.9% of material was lost at an average rate of 1.07 x 10° pmol/ug per year. Aspartic acid, glutamic acid, glycine and alanine were found to be present with the highest concentrations having a mean concentration of 28.3 pmol/.g. These amino acids were also found to be more readily hydrolyzed from the soluble organic matrix. After 4,000 years their concentrations had dropped to be within the same range as all amino acids remaining in the organic matrix with a mean concentration of 1.3 pmol/ug. Amino acid concentrations remained at this level for the next 119,000 years with little further losses. The degree of racemization from L form to D form, i.e. the D/L ratio of amino acids was found to be related to concentration and amino acids such as aspartic acid, with high initial concentration also show faster racemization rates. Aspartic acid hydrolyses and racemizes at a faster rate than other amino acids leading to the hypothesis that the more aspartic acid molecules present in the protein strand, the greater the chance that aspartic acid will be in the optimum position for hydrolysis and racemization. Manuscript received 14 February 2001, accepted for publication 22 August 2001. KEYWORDS: mollusc, organic matrix, conchiolin, diagenesis, amino acid racemization, Anadara trapezia. INTRODUCTION Researchers from several scientific disciplines are studying the protein found in the shell matrix of molluscs. Material scientists have examined the way these organic matrix proteins, secreted by molluscs, regulate calcite and/or aragonite crystal formation within shells (Weiss et al. 2000). In particular, research has focused on biomineralization processes and their relevance to the synthesis of high performance nano-composite materials (Keith et al. 1992; Belcher et al. 1998). Geologists and archaeologists have used the degree of racemization of amino acids in shell and bone as a dating tool (Hare and Abelson 1968; Murray-Wallace 1993; Kimber et al. 1994; Harada and Handa 1995; Rutter and Blackwell 1995; Johnson and Miller 1997). Increasing amounts of D form Proc. Linn. Soc. N.S.W., 123. 2001 226 ORGANIC MATRIX IN ANADARA TRAPEZIA amino acids in fossil molluscan shell is an indication of “time-since-death”. The rate of racemization is known to be genus dependent and these preliminary results are part of an ongoing study into the reasons for this dependency. All living organisms retain their amino acids in the L-form by enzyme action. When an organism dies, the enzyme suppression ceases and racemization reactions begin and will continue until equilibrium is attained. This process is highly sensitive to temperature and accordingly may have duration of 200 ka to 10 Ma (Miller and Brigham-Grette 1989). At equilibrium D/L equals 1 for enantiomers and approximately 1.3 for diastereoisomers (Miller and Brigham-Grette 1989; Murray-Wallace 1993). Stratigraphic correlation (Belperio et al. 1984; Murray- Wallace et al. 1999), phylogenetic studies (Degens and Spencer 1967; Robbins and Ostrom 1995), palaeotemperature studies (Cann and Clarke 1993; Miller et al. 1995) and the detection of reworking of fossils from older into younger sedimentary deposits (Murray- Wallace 1993; Rutter and Blackwell 1995; Wehmiller et al. 1995; Miller et al. 1997), represent other research applications of the racemization reaction of amino acids. Knowledge of how the organic matrix breaks down over time is important when considering racemization of amino acids as a dating tool. It has long been established that temperature and the availability of water affect the rate of racemization in fossils, and that variations in the degree of racemization have been found between genera of the same age (Williams and Smith 1977; Lajoie et al. 1980; Wehmiller 1980; Rutter and Blackwell 1995; Roof 1997). The variation in the degree of racemization between genera of the same age is thought to be due to the presence of different proteins, characterized by more stable peptide bonds (Degens and Spencer 1967; Hare and Abelson 1968; Wehmiller 1984; Kaufman et al. 1992; Goodfriend et al. 1997). It has been established that amino acids at N-terminal positions of proteins tend to racemize fastest and that when the amino acid is in an internal position the rate of racemization 1s affected by the amino acids on either side (Mitterer and Kriausakul 1984; Mitterer 1993; Qian et al. 1995; Goodfriend 1997). The relative position of an amino acid in the protein chain will thus determine the rate of racemization. This study examines the breakdown of the soluble organic matrix in the shell of one species of mollusc, the estuarine bivalve mollusc Anadara trapezia, over a period of 125,000 years. As diagenesis occurs, amino acids are hydrolyzed from the protein chain becoming free amino acids. In this study larger peptides and proteins have been extracted and amino acid concentrations have been determined. The concentration and composition of amino acids retained in protein and peptides over 12 kDa in size in fossil samples, is compared with the results for modern specimens. As routine studies have not quantified the concentrations of amino acids in fossil shell with any precision, this report is adding to knowledge in this field. MATERIALS AND METHODS Many current methods used to identify the amino acid composition of shell proteins involve demineralizing the shell using 6M HCl. The resultant solution is then analyzed for amino acid composition and concentration. The acid hydrolyses the intact proteins and peptide strands, as well as demineralizing the shell. Researchers then analyze the total acid hydrolysate and at times, free amino acids (Powell et al. 1989; Powell et al. 1991; Kaufman et al. 1992). Analysis of total amino acid hydrolysate includes — amino acids from the hydrolysis of the organic matrix, — free amino acids from earlier diagenesis, — free amino acids formed from the chemical breakdown of other amino acids, for example, Serine > Glycine + Alanine (Akiyama 1980), together with Proc. Linn. Soc. N.S.W., 123. 2001 M.J. WHITELAW, B.D. BATTS, C.V. MURRAY-WALLACE AND C.R. McRAE 227 — free amino acids that have diffused into the shell over time, that is, foreign organic material. This depends on the integrity of the shell matrix and the species concerned and studies have shown little in the way of non indigenous amino acids diffusing into shell (Miller and Hare 1980). To overcome the problem of hydrolysis of protein associated with the use of 6M HCl, the method used in this study involves the gentle demineralization of the shelly material using a 10% solution of EDTA at pH 8 to remove Ca**. The sample of ground shell is placed in a dialysis tube with a nominal pore size of 12 kDa allowing all free amino acids to diffuse through the tubing. Only intact proteins and larger peptide strands are retained (Wheeler et al. 1987; Halloran and Donachy 1995; Murray-Wallace et al. 2000). Selected samples were chosen in an attempt to quantify the rate of change or loss of protein residues in shells during early diagenesis using one species, the bivalve Anadara trapezia. Modern shells were analyzed along with fossil samples of known age. Modern samples were collected live from Wallagoot Lake, near Eden, on the South Coast of New South Wales (Table 1). They were immediately frozen and then shucked prior to cleaning and grinding. Fossil samples of Anadara trapezia were collected from several geological coastal deposits in southern Australia (Table 1), where the age and origin of the deposits are well established (Murray-Wallace et al. 2000). Fossil shell samples were collected from sites, which today have similar mean annual air temperatures as it is likely that materials of the same age will have experienced similar diagenetic temperatures (Table 1). Table 1. Location of collection sites for modern and fossil samples used in this study and codes assigned to each fossil sample. Age* (years) Code Location 0 Modern Wallagoot Lake, South Coast, NSW 3,880 +60 AWP Wilson Memorial Park, Koona Bay, Lake Illawarra, (SUA-3059) NSW 6,280+60 Att Tom Thumb Lagoon, Wollongong, NSW (SUA-3058 6,800+70 Akb Kully Bay, Lake Illawarra, NSW (SUA-3102) ; 125,000 Al Largs, NSW 125,000 Awg Watson’s Gap, Chiton Rocks, SA *Note: Conventional radiocarbon ages have been corrected for the marine reservoir effect for southern Australian ocean surface waters, (-450+35 years, Gillespie and Polach (1979)) and converted to sidereal years using the revised calibration program of Stuiver and Reimer (1993). Proc. Linn. Soc. N.S.W., 123. 2001 228 ORGANIC MATRIX IN ANADARA TRAPEZIA Shells for amino acid concentration analysis were cleaned by boiling 10 minutes in 30% H_O, followed by boiling for 10 minutes in 50% bleach (2.6% sodium hypochlorite by weight). They were then washed in Milli-Q™ water and scrubbed lightly. Grinding of shells was achieved using a Retsch Muhle electric mortar and pestle. Finely powdered shell (20 g), was placed in Sigma 43 mm dialysis tubing with a nominal pore size of 12 kDa; 100 mL of 10% EDTA solution at pH 8 was added and the suspension placed in a 3 L conical flask with 1.5 L of the same EDTA solution. The flask was shaken in a Bioline orbital shaker at 180 rpm at 21°C, until demineralization was complete (about 3 days). The EDTA solution in the flask was completely replaced with fresh EDTA solution once during the process. The contents of the dialysis tubing were centrifuged at 20,000 g for 30 minutes at 10°C in a Sorvall RC-5B Superspeed centrifuge. The soluble matrix in the supernatant was decanted and the insoluble matrix frozen and stored for future investigation. The soluble matrix was then placed in dialysis tubing and washed exhaustively in Milli-Q™ water at 4°C in an electric shaker. The water was frequently changed to remove the EDTA. The content of the tubing was again centrifuged at 20,000 g at 10°C for 30 minutes and the supernatant liquid containing the soluble matrix was freeze-dried. After purification, 5mg/mL samples of the soluble matrix were acid hydrolyzed and analyzed for amino acid concentration and composition. Amino acid analysis was performed by the Australian Proteome Analysis Facility (APAF) using F- moc (9-Fluorenylmethyl chloroformate) chemistry, a sensitive method that allows detection of amino acids at low levels (down to 1 ug). The detection of F-moc derivatised amino acids is via fluorescence using the GBM Aminomate system. Nitric acid and hydrochloric acid were trialed as demineralizing agents with the results showing that some hydrolysis of the shell protein occurred. EDTA, although a gentler reagent, has disadvantages. EDTA is difficult to remove completely from the final protein sample and interferes with the amino acid analysis. A blank sample was run and the results showed that peaks at the phenylalanine and tyrosine positions were produced and to a lesser extent at other amino acid wavelengths. This was overcome by deleting phenylalanine and tyrosine from the results and subtracting the blank (EDTA) values from the sample values for other amino acid peaks (Table 2). Analysis for the extent of racemization followed the methods documented in Murray-Wallace (1993). Analysis of the N-pentafluoropropionyl-2 propyl esters was undertaken using a Hewlett-Packard 5850 Series II gas chromatograph fitted with a coiled, 25m capillary column with the stationary phase Chirasil-L- Val. RESULTS Protein was extracted from the shells, purified and analysed for amino acid concentration and composition. Analysis showed that the loss of amino acids was rapid over the first 4,000 years from the onset of racemization, and maintained a relatively stable concentration from 4,000 years to 125,000 years (Fig.1). Over the first 4,000 years 91% of original material was lost at an average rate of 3.25 x 10° pmol/g per year. Over the next 2,000 years a further 6.3% was lost at an average rate of 4.49 x 10% pmol/ug per year. Over the final 119,000 years, 0.9% of material was lost at an average rate of 1.07 x 10° pmol/ug per year. The curve for the average depletion rate of amino acids is shown in Figure 3 and has the equation y = 12.221x 7! with an R? of 0.9783. In the modern shells three groups of amino acids are apparent based on their relative concentrations (Fig. 1). A group with high concentrations (mean of 28.3 pmol/ ug), a group with moderate concentrations (mean of 15.7 pmol/ug) and a third group with quite low concentration level (mean of 1.4 pmol/ug). After 4,000 years the low- level amino acids have completely disappeared and therefore disregarded in this study, Proc. Linn. Soc. N.S.W., 123. 2001 229 M.J. WHITELAW, B.D. BATTS, C.V. MURRAY-WALLACE AND C.R. McRAE 080 9LT SsI9 sky L10 9L 1 67+ skT 010 9L'I I'6 sk 6S°7 Sy SOLOL «(NOT IL'0 Svc OIS'SS nay €7'0 Sv'T €9'°61 nay 79°T SSII pOOIZ Al 7L'0 SS Il 7199 Jit €10 SS II 19 IZ II 97 p10 S8P8I eA 88°0 p10 E99 LIN 7S'0 p10 9S OF LIN €0'T GGll GLYh — wil Be 671 16697 IK], Beal 67 I OT vEl IK], S001 887 TSOSL Old LOE 887 68° TET Old Eel 88°C L8°7OI Old p71 ISI LLb6 IV 0€'0 Gale Gaze BV aml) Is] 8611 eV 9r'I 991 6011 UL €70 951 L&I YL 80°0 9S ee YL Isl €cL0 ILEll 3IV 970 €L0 907 3IV 90°0 €L'0 ESS 3I1V els I€91 8elor AID 9¢'1 I€9L €7 Eel AID LL‘0 I€9l =e rl AID Sol TSE Pics 1S all) TSE EC Jag 60°0 TS'€ El Ol Jag 88°0 p97 9689 NID 970 197 1617 nD 010 19°C 8°6 no 00°I €67 I'8L dsy 1€0 €67 7097 dsy S10 £67 Ie vl dsy snjjowd yuejg odues qyuy snyoud yurejg os[dures IV snyjowd yurjg © ajduies SMV 70 9L1 TS8I sAT Is'0 9L 1 996 sk] I9'L 9L'I STL SkT 6S°0 Gn 2) ae Ol'l Svc OV 'S8 noy $86 Src p0'IPL nay 09°0 SSI 8h 9S atl 9L'0 SS Il 99°89 Tl ESL SII Ss 9L¢ Il 130 t10 1609 [eA IVT v10 SOEs eA I€0Il +10 Iv€Ll eA 770 GON bie 28M, re0 671 96°97 IK], 60°0 67 I 91°8 IAL, 79'S 887 Serr ld SIT 887 6r 16 Old SSIE 88° €8 167 Old 80 IST 7EOe eV ETT ISI +6'€6 BV CiSCameeel'Gul €8°C8LI1 eV 610 9ST 8ST WL rai 9S1 8g°S8 WL ORCE OS 66'€S6 YL 970 EL0 60% 3IV LOT €L0 I€T8 3IV OTL €L'0 Or LES 31V p61 I€9I s9I9T AID LLY I€9I 87 HLe AID 9rST IE9L 6L'S7O6I AID 810 TSE POLI 1S 60°7 TSE 67 09I 12g GES CF 97 SPIT Jag LT0 p97 ITET MD 61 p97 =LI8bl nD 7697 = H9'T Sr 1ZO7 nD 770 €67 86 dsy 99°| €67 61L7I dsy 969€ £67 97 SLLTZ dsy snjjoulg yuelg osjdwes Vv snyjowg yuejq adues dMV snjowg yuelg gjdures UIIpO| ‘sonjeA o[dures oy} Wos sonjea yuRTq oy) SuNOeNQns Aq pourejqgo 3r1/joud ut ‘uone.nUIDUOD pajsn[pe oy} SI 9a} UWINTOS puke UNI SeM VLG JO a[duies yuryq & Udy plov OUTUIR YORO OJ poute}go UOT}eUIOUOD JUoIedde dy} SMOYS UUIN]OS puodsas oy J, ‘aTduies sri¢y tod jowd ur uonenusou0d ay) SMOYS UUUNIOS sIJ IYJ, ‘Sofduues [JOYS XIS oy} JoJ UMOYS oIe SUONeNUDDUOD ‘7 IIQVL, Proc. Linn. Soc. N.S.W., 123. 2001 230 ORGANIC MATRIX IN ANADARA TRAPEZIA 40.00 - EF; = 35.00 E E3000 & 2500 E 20.00 = 3 15.00 g = 10.00 z Sree 5.) = E000 Modem 3880 6280 125,000 Age (in years) Figure 1. Depletion of amino acids with time. Aspartic acid shows a higher concentration in fossil samples relative to the other amino acids. There are three groups of amino acids, Group A with high initial concentrations, Group B with moderate initial concentrations and Group C with low initial concentrations. Amino acids are listed from top to bottom on the figure. It is to be noted that there are three amino acids with very similar concentration, isoleucine, arginine and lysine, seen as a single dark line in Group B. Group A Group B Group C Aspartic acid Serine Methionine Glutamic acid Threonine Histidine Glycine Valine Alanine Leucine Lysine Isoleucine Arginine Table 3. Concentrations of four amino acids and the corresponding D/L ratios for those amino acids from modern and fossil samples Age Amino acid concentration Amino acid D/L ratio (years) (pmol/g) (in fossil samples) Asp Glu Ala _ Val Asp Glu Ala_ Val 0 3996) 26:92, 23H/5— 103i - - - - 3,880 GOR 194. S123e5 Malt 0.42 0.12 0.22 0.06 6,280 O22 02778 O35 Sie 0r8il OAS 0:18) S1O%32) OFZ 6,800 100 0.88 1.24 2.46 NAS 2027/0330) ON 125,000 0.31 0.26 0.30 0.88 0.55 0.36 0.60 0.30 Proc. Linn. Soc. N.S.W., 123. 2001 M.J. WHITELAW, B.D. BATTS, C.V. MURRAY-WALLACE AND C.R. McRAE PPE }I| but for all the other amino acids, after 4,000 years, their concentrations are within the same range, (mean of 1.34 pmol/ug) and although low, are still detectable. Amino acid concentrations remain at this level for the next 120,000 years with very little further losses. D/L ratios are reported for the enantiomeric amino acids alanine, valine, glutamic acid and aspartic acid (Table 3). D/L ratios for the four amino acids are reported for all fossil samples. The ratios show an interesting relationship with the original concentration of each amino acid in modern shell. As expected, with each of the four amino acids, the D/L ratio increases with time as amino acids racemize and approach equilibrium. The concentration of amino acids in the unhydrolyzed organic matrix is at the same time decreasing (Fig. 2). Figure 2. Composite graph which illustrates the depletion of four amino acids from the organic matrix, (shown as hatched lines) and the corresponding increase in D/L ratio of those amino acids, with time. iS) 40 + fa & 3 & = 35 + om See. 30! oa sa == e —) 95+ mm o a 8 e 8 Bie 25 Sa | 15 + — = =| OD « & 3 R 10 + = ms 5+ 5 a 0-4 s ¥ So Age (in years) for the amino acids Aspartic acid, Glutamic acid, Alanine and Valine There is a correlation between the D/L ratio and the original concentration of a given amino acid in the modern shell. Aspartic acid was originally present in Anadara trapezia with the highest concentration. After some 4,000 years since the onset of racemization, the concentration dropped to levels comparable to most other amino acids in the sample and yet the aspartic acid D/L ratios in fossil samples tend to be higher than for the other amino acids (Fig. 3). The high concentration of aspartic acid found in the modern organic matrix suggests that there is a greater chance that an aspartic acid residue will be in the terminal position of the protein and hence in a position where racemization may readily occur. Hence we hypothesize that the higher the concentration of an amino acid the greater the chances of its occurrence at terminal positions during diagenesis and hence the more likely that racemization will occur thus accounting for the higher rate of racemization experienced by that acid (in this case, aspartic acid). This hypothesis is in agreement with the finding that acidic amino acid residues are found in every biomineralized system examined to date (Weiner et. al 1983). DISCUSSION The concentration of aspartic acid falls dramatically over 4,000 years possibly indicating that aspartic acid is hydrolyzing rapidly from the protein and forming free amino acids. These free amino acids are undergoing racemization over this period of time. Proc. Linn. Soc. N.S.W., 123. 2001 232 ORGANIC MATRIX IN ANADARA TRAPEZIA Figure 3. Average depletion curve. When the average of all the amino acid concentrations is plotted against time, it is possible to plot a power curve that fits the data very well with an R?> 0.97. The equation of the curve is y=12.2x**° where x represents time and y is concentration. wm 16 = 14 4 Bean a aw\ Wiccal 25221 a ag Be R’ = 0.9783 =S Ss 5 6- — 3g a ay O 0 0 4000 6000 125,000 Time (years) One sample, Akb, dated at 6,800+ 70 years, was found to have higher concentrations for all amino acids when compared with another sample Att, of similar age, (6,280+60 years), and geological setting. Both samples related to the same set of geological processes: quiet water, coastal lagoon which formed after the sea level rose to its present position 7,000 years ago (Young et al. 1993). It is speculated that the preservation of shells collected in Kully Bay, Lake Illawarra (Akb) was better than conditions experienced at Tom Thumb Lagoon, Wollongong (Att) thus retarding the hydrolysis of the conchiolin proteins. This could be due to differing local environmental conditions or individual variations in the sample shells. The extent and effect of boring animals in shell samples could account for discrepancies, as would other natural variations. An alternate interpretation for the discrepancy would be that whilst in situ, low molecular weight peptides and free amino acids were preferentially lost from perhaps only one shell in the sample through in situ leaching. As the amino acid concentrations are derived using several shell samples and the D/ L ratios are calculated for the total amino acid hydrolysate, the loss of lower molecular weight peptides and free amino acids could produce this discrepancy. The results for Akb are shown in Table 2 but for clarity are not included when comparing amino acid depletion with time. CONCLUSIONS | Hydrolysis of the shell protein is initially rapid with over 90% of original material lost within 4,000 years, at an average rate of 3.25 x 10° pmol/ug per year. The remaining protein is relatively stable and this study shows that over the next 120,000 years only 0.9% of material was lost at an average rate of 1.07 x 10° pmol/g per year. Proc. Linn. Soc. N.S.W., 123. 2001 M.J. WHITELAW, B.D. BATTS, C.V. MURRAY-WALLACE AND C.R. McRAE 233 2 Regardless of whether amino acids are present in high or moderate concentration in the protein of modern shell, their concentrations fall to approximately the same level after 4,000 years, (with a mean of 1.34 pmol/ ug), from the onset of diagenesis. As hydrolysis proceeds and the protein breaks down into peptide strands, amino acids will increasingly be found in optimum positions for racemization and the D/L ratio for each amino acid will increase. Graphs of D/L ratios of individual amino acids versus their concentration in unhydrolysed organic matrix show these expected trends. 3 If an amino acid is present in high concentrations in modern shell it will also show a higher rate of racemization in fossil samples. The rate of increase of the D/L ratio for aspartic acid was found to be much greater than that of the other amino acids considered in this study. As there is a higher concentration of aspartic acid molecules in the peptide strand, we suggest that there is a greater chance that aspartic acid will be found at the optimum N-terminal during hydrolysis and so racemize faster. REFERENCES Akiyama, M. (1980). Diagenetic decomposition of peptide-linked serine residues in fossil scallop shells. In “Characterization of proteins,peptides and amino acids’. Quaternary Biogeochemistry of Amino Acids. Conference Papers (Eds P.E. Hare, T.C. Hoering and K. King) pp. 115-120. (Wiley: New York). Belcher, A.M., Hansma, P.K., Stucky, G.D. and Morse, D.E. (1998). First steps in harnessing the potential of biomineralization as a route to new high-performance composite materials.” Acta Mater 46, 733- 736. Belperio, A.P., Smith, B.W., Polach, H.A., Nittrouer, C.A., DeMaster, D.J., Prescott, J.R., Hails, J.R. and Gostin, V.A. (1984). Chronological studies of the Quaternary marine sediments of northern Spencer Gulf, South Australia. Marine Geology 61, 265-296. Cann, J.H. and Clarke, J.D.A. (1993). The significance of Marginopora vertebralis (Foraminifera) in surficial sediments at Esperance, Western Australia, and in last interglacial sediments in northern Spencer Gulf, South Australia. Marine Geology 111, 171-187. Degens, E.T. and Spencer, D.W. (1967). Palaeobiochemistry of molluscan shell proteins. Comparative Biochemical Physiology B. 20, 553-579. Gillespie, R. and Polach, H.A. (1979). The suitability of marine shells for radiocarbon dating of Australian prehistory. In ‘Proceedings Ninth International Conference on Radiocarbon Dating’. (Eds R. Berger and H. Suess) pp. 404-421. (University of California Press: California). Goodfriend, G. Flessa, K.W. and Hare, P.E. (1997). Variation in amino acid composition among shell layers in the bivalve Chione from the Gulf of California. Geochimica et Cosmochimica Acta. 61, 1487- 1493. Halloran, A.B. and Donachy, J.E. (1995). Characterization of organic matrix macromolecules from the shells of the Antarctic scallop, Adamussium colbecki. Comparative Biochemistry Physiology B. 2, 221- 231. Harada, N. and Handa, N. (1995). Amino acid chronology in the fossil planktonic foraminifera, Pulleniatina obliquiloculata from the Pacific Ocean. Geophysical Research Letters. 22, 2353-2356. Hare, P.E. and Abelson, P.B. (1968): Racemization of amino acids in fossil shells. In “Carnegie Institute of Washington Yearbook’. Vol. 66. pp. 526-528. Johnson, B.J. and Miller, G.H. (1997). Archaeological applications of amino acid racemization. Archaeometry. 38, 265-287. Kaufman, D.S., Miller, G-H. and Andrews, J.T. (1992). Amino acid composition as a taxonomic tool for molluscan fossils: an example from Pliocene-Pleistocene Arctic marine deposits. Geochimica et Cosmochimica Acta. 56, 2445-2453. Keith, J.A., Stockwell, S.A., Ball, D.H., Muller, W.S., Kaplan, D.L., Thannhauser, T.W. and Sherwood, R.W. (1992). Characterization of the complex matrix of the Mytilus edulis shell and the implications for biomimetic ceramics. Hierachically structured materials. Proceedings Materials Research Society. 255, 3-8. Kimber, R.W.L., Kennedy, N.M. and Milnes, A.R. (1994). Amino acid racemization dating of a 140 000 year old tephra-loess-paleosol sequence on the Mamaku Plateau near Rotorua, New Zealand. Australian Journal of Earth Sciences. 41, 19-26. Lajoie, K.R., Wehmiller, J.F. and Kennedy, G.L. (1980). Inter- and intrageneric trends in apparent racemization kinetics of amino acids. In ‘Quaternary mollusks. Biogeochemistry of amino acids. Conference Papers’. (Eds P.E. Hare, T.C. Hoering and K.King Jr.). pp. 305-339 (Wiley: New York). Proc. Linn. Soc. N.S.W., 123. 2001 234 ORGANIC MATRIX IN ANADARA TRAPEZIA Miller, G.H. and Brigham-Grette, J. (1989). Amino acid geochronology: resolution and precision in carbonate fossils. Quaternary International .1, 111-128. Miller, G.H. and Hare, P.E. (1980). Amino acid geochronology: Integrity of the carbonate matrix and potential of molluscan fossils. In “Biogeochemistry of Amino Acids’. (Eds P.E. Hare, T.C. Hoering and K. King). pp. 415-451. (Wiley: New York). Miller, G.H., Magee, J.W. and Jull, A.J.T. (1997). Low-latitude glacial cooling in the Southern Hemisphere from amino-acid racemization in emu eggshells. Nature. 385, 241-244. Mitterer, R.M. (1993). The diagenesis of proteins and amino acids in fossil shells. “Organic Geochemistry’ (Eds M.H. Engel and S.A. Macko). pp. 739-753. (Plenum Press: New York). Mitterer, R.M. and Kriausakul, N. (1984). Comparison of rates and degrees of isoleucine epimerization in dipeptides and tripeptides. Organic Geochemistry. 7, 91-98. Murray-Wallace, C.V. (1993). A review of the application of the amino acid racemization reaction to archaeological dating. The Artifact. 16, 19-26. Murray-Wallace, C.V., Belperio, A.P., Bourman, R.P., Cann, J.H. and Price, D.M. (1999). Facies architecture of a last interglacial barrier: a model for Quaternary barrier development from the Coorong to Mount Gambier coastal plain, southeastern Australia. Marine Geology .158, 177-195. Murray-Wallace, C.V., Beu, A.G., Kendrick, G.W., Brown, L.J., Belperio, A.P. and Sherwood, J.E. (2000). Palaeoclimatic implications of the occurrence of the arcoid bivalve Anadara trapezia (Deshayes) in the Quaternary of Australasia. Quaternary Science Reviews. 19, 559-590. Powell, E.N., King, J.A. and Boyles, S. (1991). Dating time-since-death of oyster shells by the rate of decomposition of the organic matrix. Archaeometry 33, 55-68. Powell, E.N., Logan, A., Stanton Jr., R.J. and Davies, D.J. (1989). Estimating time-since-death from the free amino acid content of the mollusc shell: a measure of time averaging in modern death assemblages? Description of the technique. Palaios. 4, 16-31. Qian, Y., Engel, M.H., Goodfriend, G.A. and Macko, S.A. (1995). Abundance and stable isotope composition of amino acids in molecular weight fractions of fossil and artificially aged mollusk shells. Geochimica et Cosmochimica Acta. 59, 1113-1124. Robbins, L.L. and Ostrom, P.H. (1995). Molecular isotopic and biochemical evidence of the origin and diagenesis of shell organic material. Geology. 23, 345-348. Roof, S. (1997). Comparison of isoleucine epimerisation and leaching potential in the molluskan genera Astarte, Macoma, and Mya. Geochimica et Cosmochimica Acta. 61, 5325-5333. Rusenko, K.W., Donachy, J.E. and Wheeler, A.P. (1991). Purification and characterization of a shell matrix phosphoprotein from the American oyster. In “Surface reactive peptides and polymers’. (Ed K.W. Rusenko) pp. 107-124. (American Chemical Society). Rutter, N.W. and Blackwell, B. (1995). Amino acid racemization dating. In ‘Dating Methods for Quaternary deposits’. (Eds N.W. Rutter and N.R. Catto) pp. 125-164. (Geological Association of Canada, Geotext2). Stuiver, M. and Reimer, P.J. (1993). Extended '4C data base and revised calibration 3.0 “C age calibration program. Radiocarbon. 35, 215-230. Wehmiller, J.F. (1980). Intergenetic differences in apparent racemization kinetics in mollusks and foraminifera: implications for models of diagenetic racemization. In “Biogeochemistry of amino acids’. (Eds P.E. Hare, T.C. Hoering and K. King Jr.) pp. 341-355. (Wiley: New York). Wehmiller, J.F. (1984). Relative and absolute dating of Quaternary molluscs with amino acid racemization: evaluation, applications and questions. In ‘Quaternary dating methods’. (Ed W.C. Mahaney) pp. 171-193. (Elsevier Science:Amsterdam). Wehmiller, J.A, York, L.L. and Bart, M.L. (1995). Amino acid racemization geochronology of reworked Quaternary mollusks on U.S. Atlantic coast beaches: implications for chronostratigraphy, taphonomy and coastal sediment transport. Marine Geology. 124, 303-337. Weiner, S., Traub, W. and Lowenstam, H.A. (1983). Organic matrix in calcified exoskeletons. In ‘Biomineralization and biological metal accumulation’. (Ed. P. Westbrock and E.W. de Jong), pp. 205-224. (Dordrecht, Netherlands: Reidel). Weiss, I.M., Kaufmann, S., Mann, K. and Fritz, M. (2000). Purification and characterization of Perlucin and Perlustrin, two new proteins from the shell of the mollusc’ Haliotis laevigata. Biochemical and Biophysical Research Communications . 267, 17-21. Wheeler, A.P., Rusenko, J.W., George, J.W. and Sikes, C.S. (1987). Evaluation of calcium binding by molluscan shell organic matrix and its relevance to biomineralization. Comparative Biochemistry and Physiology. 87B, 953-960. Williams, K.M. and Smith, G.G. (1977). A critical evaluation of amino acid racemization to geochronology and geothermometry. Origins of Life. 8, 91-144. Young, R.W., Bryant, E.A., Price, D.M., Wirth, L.M. and Pease, M (1993).Theoretical constraints and chronological evidence of Holocene coastal development in central and southern New South Wales, Australian Geomorphology.7, 3\7-329. Proc. Linn. Soc. N.S.W., 123. 2001 The Insect Assemblage Visiting the Flowers of the Subtropical Rainforest Pioneer Tree Alphitonia excelsa (Fenzl) Reiss. ex Benth. (Rhamnaceae) GEOFF WILLIAMS !*? AND PAUL ADAM 7 ‘c/o Department of Entomology, Australian Museum, 6 College Street, Sydney NSW 2000 ? School of Biological Sciences, University of New South Wales, Sydney NSW 2052 Williams, G. and Adam, P. (2001). The insect assemblage visiting the flowers of the subtropical rainforest pioneer tree Alphitonia excelsa (Fenzl) Reiss. ex Benth. (Rhamnaceae). Proceedings of the Linnean Society of New South Wales 123, 235-259. Alphitonia excelsa is a bisexual, protandrous, pioneer rainforest tree. Anthesis and nectar production are diurnal. Populations studied on the Mid-North Coast of New South Wales flower between January and March. Alphitonia excelsa is dependent upon insects for pollen transfer. Flower-visiting insect assemblages are dominated by Hymenoptera, Coleoptera, and especially Diptera but vary over time, and geographically. Most of the visiting insects were 6 mm or less in size. Approximately 200 genera, from 116 families, were recorded from A. excelsa flowers. This fauna comprises taxa that, currently, are known within the region only from A. excelsa, and species shared with other mass-flowering rainforest trees. Aculeate wasps were a conspicuous element of the anthophilous insect fauna visiting A. excelsa in a littoral rainforest remnant at Harrington. Introduced honey bees, Apis mellifera, were active at blossoms at all study sites, but visitation varied over the 3 seasons of study. Manuscript received 15 August 2001, accepted for publication 21 November 2001. KEYWORDS: insect assemblages, biodiversity, Alphitonia excelsa, pollination, subtropical rainforest, rainforest restoration. INTRODUCTION Although Australian rainforests have been subject to an increasing number of ecological studies much recent work has concentrated in the tropics (e.g., Harrington et al. 1997, Laurance 1997, Kitching et al. 2000), and documentation of the biodiversity of subtropical rainforests is still relatively poor. We investigated the insect fauna associated with the flowers of the small to medium-sized pioneer tree Alphitonia excelsa (Fenzl) Reiss. ex Benth. (Rhamnaceae) (Figs. 1, 2). Alphitonia excelsa occurs in the Northern Territory, Western Australia, Queensland, and New South Wales (extending to Mt Dromedary on the far south coast) (Harden 1990). Isolated populations occur inland in the Pilliga Scrub and the Nandewar Ranges (Mt Kaputar National Park) on the New South Wales northern slopes. In New South Wales A. excelsa occurs commonly on the margins of littoral rainforest, in floodplain rainforest remnants, submontane subtropical rainforest, wet sclerophyll forest and also occasionally in dry sclerophyll forest. This distribution indicates that the species is a successful coloniser of a broad range of forest subformations and soil types. Proc. Linn. Soc. N.S.W., 123. 2001 236 INSECTS VISITING A RAINFOREST TREE The genus Alphitonia consists of approximately 20 species distributed in Australia, Malesia and Polynesia. Four of the 6 species occurring in Australia are endemic (Harden 1990). Only the flower-visiting insects of A. petriei Braid & C. White have been previously investigated (Irvine and Armstrong 1988). This species occurs in Queensland and uncommonly in northern New South Wales (Harden 1990). The anthophilous insect visitors, floral structure, floral longevity and flowering phenology of A. excelsa in northern New South Wales were studied from 1991 to 1994. Opportunistic observations on flowering phenology continued until 2001. Comparisons were made with the insect assemblages visiting sympatric mass-flowering rainforest species. Pioneer trees occupy margins and canopy gaps and can occur as senescent emergents above the canopy layer of late-phase regenerating rainforest. Colonising rainforest trees form a distinct subcanopy within wet sclerophyll forests. Pioneer species generally recruit from a broad spectrum of pollinators and are “usually self-compatible, often autogamous, or reproduce significantly by vegetative means” (Gross 1993, see Baker 1955). In the case of subtropical rainforest pioneers their ability to recruit pollinators from adjacent plant communities, or from the sclerophyllous canopy stratum in wet sclerophyll forests, has been little investigated. The insect fauna associated with flowering Alphitonia will contain species filling a variety of roles, and not all will be pollinators. However, the assemblage will contain pollinators, and within this subset individual species will vary in their efficiency in effecting pollen transfer. We have previously demonstrated that apparently smooth-bodied, and large, insects visiting rainforest flowers (including those of A. excelsa) may carry pollen (Williams and Adam 1998), and House (1985, 1989) has documented pollen movement by small-sized insects between tropical dioecious rainforest trees. Consequently, members of the assemblage may, even if incidentally, contribute to pollination but the relative contribution of different species was not studied. MATERIALS AND METHODS Investigations were undertaken at three locations; in a littoral rainforest remnant at Harrington (31°52’30”S, 152°41’00”E), and in mixed subtropical rainforest - wet sclerophyll forest at Kenwood Wildlife Refuge (31°44’45”S, 152°31°30”E) and at Lorien Wildlife Refuge (31°45’00”S, 152°32’30”E). The latter two sites were situated on the Lansdowne-Comboyne Escarpment (Williams 1993), approximately 18 km north of Taree on the north coast of New South Wales. This study was part of a larger investigation of the pollination ecology of lowland subtropical rainforest (Williams 1995). Insect voucher specimens were deposited primarily in the Australian Museum, Sydney, and the Australian National Insect Collection, CSIRO, Canberra. Flower characteristics Flower opening, structure, and presence of nectar were determined in the field and by examination of individual florets using a light microscope. Florets were examined under ultra violet light to investigate presence of nectar guides. Flower Phenology Observations on flower opening were undertaken at Harrington and Lorien Wildlife Refuge. In addition to A. excelsa, all tree species were censused each month, over three seasons (1991-94). At Harrington two marked transects ~ 400-500 m long, separated by 2.5 kms, were traversed. Casual observations of the Harrington A. excelsa population continued to 2001. At Lorien Wildlife Refuge a single 600-700 m transect was walked. Observations of A. excelsa at the Kenwood site were restricted to the period of sampling during the 1991-92 season only, and only related to the two trees being Proc. Linn. Soc. N.S.W., 123. 2001 G. WILLIAMS AND P. ADAM 13H) sampled. Owing to time constraints we did not record the numbers of trees annually flowering within each population. However, weekly field observations subsequent to and during sampling at Harrington and Kenwood provided finer time scale data on length of flowering for the trees being sampled. Visitation Periodicity The number of flowers produced by individual trees is one of many biotic and abiotic factors that can influence recruitment of pollinators (Dafni, Lehrer and Kevan 1997). Variation in the numbers of flowers produced by individual trees was assessed by estimating the ratio of blossoms (buds) to leaf surface (BLS), and by a relative estimate of the percentage of available buds (PAB) on each of the trees sampled. These BLS and PAB ratios are a measure of potential attraction defined by floral resource availability, and can indicate peak periods of foraging activity over the period of flowering by individual trees. The measures are discussed in Williams (1995) and Williams and Adam (1997), however, it needs to be emphasised that the PAB value is a relative measure, because the number of initially available buds diminishes due to fertilisation, floret abortion, herbivore attack, and dislocation by wind and other disturbances generally. Peak BLS values occur at 2 50%, being a consequence of the physical presentation of flowers and leaves. Composition of the visiting fauna The taxonomic composition of diurnal anthophilous faunas was assessed using repeated standardised hand-netting of A. excelsa trees (at Harrington and Kenwood Wildlife Refuge), to provide an indication of spatial variation in visiting taxa and their abundance and to assess the daily and seasonal variation in foraging patterns. Opportunistic night spot-lighting of trees in flower was undertaken at Harrington in 1991. Four trees (2 x 2) were sampled over three years (1991-93) at Harrington and during 1992 at Kenwood Wildlife Refuge. The same trees were sampled each year at the Harrington site. At each site the trees being sampled were more than 20 m apart. Observations of the insects visiting flowers were undertaken on an additional 2 trees at Kenwood (in 1992), and more than 17 trees at Harrington (1991-1993). The insect species observed on these trees did not obviously differ from the trees sampled. Although these additional trees were not sampled, owing to the sample load generated by the broader pollination study (see Williams 1995), occasional insects were individually collected to determine pollen loads (see Williams and Adam 1998). Day-flying insects visiting inflorescences of A. excelsa were sampled using a long hand-held net with a cloth bag that could be quickly detached. Insects could not exit through the cloth wall. Ten inflorescences (= 1 aggregate/composite sample), of similar developmental stage from each of two trees, were sampled by quickly placing the net over each inflorescence and shaking briskly to dislodge all insects. The mouth of the net was closed by rotating the handle to minimise loss of fast-flying insects. This was repeated for the same 2 trees per site, morning and afternoon (i.e. 4 aggregate samples, or 40 subsamples per sampling day) during the 1992 and 1993 seasons, and morning sampling only for the 1991 flowering season. This sampling was carried out approximately once a week, over the entire period of flowering, so that insect diversity and abundance can be related to changes in recruitment cues and potential floral resources (using BLS and PAB values). Most samples were collected during humid and hot (> 28° Celsius) days, such weather conditions being the seasonal norm at the study sites. Sampling was avoided during cooler periods, during and shortly after rain, and during overcast periods; which are conditions that reduce insect activity and abundance, or reduce floral cues (e.g., by diluting nectar and scent plumes). Consequently, there was occasional variation in sampling time between days. Proc. Linn. Soc. N.S.W., 123. 2001 238 INSECTS VISITING A RAINFOREST TREE Following the collection of each aggregate sample the net bag was detached, placed within a plastic container and sprayed with commercial pyrethroid insecticide and sealed for 10-20 minutes. Contents were then emptied into labelled petri dishes for later sorting, measurement and identification. Insects were assigned to arbitrary size classes, and the number of insects (abundance), and estimated number of ‘taxa’, as recognisable taxonomic units or morphospecies per insect order or suborder, were recorded for each aggregate sample. The behaviour of individual species on inflorescences was observed, but no attempt was made to characterise the efficiency of individual taxa as pollinators. Comparison of fauna with that of other flowering rainforest trees To gauge whether mass-flowering rainforest trees recruit insects from a shared local (site-specific) assemblage, anthophilous insects collected from A. excelsa at Harrington were compared with diurnal insects visiting Acmena smithii (Poiret) Merr. and Perry (Myrtaceae), Euroschinus falcata J.D. Hook (Anacardiaceae), Scolopia braunii (Klotzsch) Sleumer (Flacourtiaceae), Guioa semiglauca (F. Muell.) Radlk. and Alectryon coriaceus (Benth.) Radlk. (Sapindaceae) at the same site. These are mass-flowering rainforest species that produce numerous open and actinomorphic flowers of a generally similar size and colour to A. excelsa. Acmena smithii, E. falcata, S. braunii, G. semiglauca and A. coriaceus flower earlier in the year than A. excelsa, but individual species may not flower each season (Williams 1995). The sampling methodology was the same as that for A. excelsa, however, the reduction in flowering episodes reduced the sample base for these species. RESULTS Flower characteristics The small white flowers of A. excelsa (Figs. 1, 2) are massed in terminal or axillary cymes and conform to the general entomophilous floral syndrome (Williams and Adam 1994). They are not obscured by foliage. The flowers are protandrous, strongly scented, with an odour reminiscent of urine, and do not possess nectar guides. Sepals 5; petals 4 or 5, rarely 6, reduced and spoon-shaped (spathulate). Stamens antepetalous, 4-5 in number, rarely 6. Style 2-, sometimes 3-lobed, developing after anther dehiscence. There are no obvious nectariferous basal glands, but the receptacle is ‘fleshy’. Ovaries are green, petals and filaments white, the filaments being slightly translucent. The sepals and receptacle are greenish-white. Petals and stamens are erect by the time the anthers dehisce, and recurved and held laterally as the ovary develops. Pollen is cream to greenish in colour, translucent, and slightly sticky. At flower opening, and when anthers dehisce, the stamens are encapsulated by the petals (which at anthesis are not fully spathulate in shape). When the petals surrounding the dehiscent anthers are depressed (for example, by a visiting insect) the sticky pollen grains are exuded, in an elongated mass, clear of the encapsulating petals onto flower visitors. Following anther dehiscence, the stamens are carried ‘free’ or away from the developing ovaries as the encapsulating petals gradually recurve. Nectar is produced at full flower opening, which is diurnal and occurs between 1030-1300 hours EST. Nectar production ceases upon apparent pollination/fertilisation. Stigmatic lobes become brown following pollination. Mean flower longevity 7 (range 5-9) days (number of flowers sampled = 10). Flowering phenology Flowering at all sites was restricted to January - March with peak flowering (highest BLS/PAB levels in sampled trees) in February, approximately 2-4 weeks after onset of flowering. However, even at the coarse monthly census scale, records indicate spatial, and suggest interseasonal, variation in flowering (Williams 1995). Proc. Linn. Soc. N.S.W., 123. 2001 G. WILLIAMS AND P. ADAM 239 Figure 1. Polistes wasp on Alphitonia excelsa flowers. Figure 2. Alphitonia excelsa floret, c. 4.5 mm diameter. Proc. Linn. Soc. N.S.W., 123. 2001 240 INSECTS VISITING A RAINFOREST TREE The individual A. excelsa trees sampled at Harrington flowered for approximately 7-8 weeks in 1991, and 6 weeks in 1992 and 1993. Populations at Harrington flowered every year of observation (1991-2001). Length of flowering by the Kenwood population (in 1992) was approximately 6 weeks. However, during three seasons (1991-94) of observation at Lorien Wildlife Refuge A. excelsa flowered only in the 1991-92 season, for approximately 6 weeks (Williams 1995). Observations during 1991-94 coincided with a period of drought (Williams 1995) and flowering episodes at Lorien during this period may have been affected by water stress. Visitation Periodicity Differences in the number of individuals and taxa recorded in Tables 1-5 suggest temporal variation in insect recruitment, however, this may reflect change in the size and composition of the available pool of insects. The data also indicate variation in numbers of insects and numbers of species visiting individual trees. However, we do not know whether these results represent patterns of daily foraging behaviour by individual taxa, fluctuation in the pool of available pollinators, differential response to recruitment cues entrained by variation in the availability of resources, or environmental influences. At Harrington in 1991 more individual insects were captured on Tree 2 than Tree 1 during peak phase flowering. Numbers of individuals peaked on the 25th January at both trees, approximately 3 weeks after onset of flowering. Coleoptera comprised a greater proportion of individuals on Tree 1, than Tree 2, while Diptera and Hymenoptera comprised a greater proportion of insects visiting Tree 2 (Table 1). Species at Tree 1 comprised a higher proportion of Coleoptera, and although richness peaked approximately 3 weeks after onset of flowering levels were spread more evenly than Tree 2 throughout the period of flowering (Table 1). The greatest number of species at Tree 2 occurred approximately 2-4 weeks after flowering onset, and comprised mainly Hymenoptera and Diptera (Table 1). Visitation patterns at Harrington in 1992 varied between trees and between morning and afternoon samples (Table 1). Numbers of individuals and species were generally greatest at mid-phase flowering, approximately 2-3 weeks after onset of flowering. On Tree 1 Diptera dominated early phase flowering but steadily reduced in proportion throughout the season. There were no parallel trends on Tree 2. Greatest number of individual Hymenoptera occurred on both trees during peak availability of floral resources, approximately 2-3 weeks after flowering onset. There were no clear trends in numbers of species within individual insect orders, however, there was an overall peak in number of species approximately 2-3 weeks after flowering onset. Visitation in the final 1993 sampling season at Harrington was characterised by early phase Hymenoptera and Diptera peaks. There continued to be variation between trees in numbers of individuals and species (Table 1). At the Kenwood site Hymenoptera and Coleoptera occurred in fewer numbers than Diptera, but there was no obvious pattern (Table 2). Total numbers of individuals and species (as means) were plotted against increments in blossom:leaf surface (BLS) and percentage of available bud (PAB) values. Numbers of individuals and species, measured against BLS ratios peaked in mid phase flowering (Fig. 3). However, measured against PAB values, there was a distinct late phase peak, but with no clear trend over the period of flowering (Fig. 4). The late phase peak may be interpreted as a concentration of insects on diminishing floral resources at the conclusion of flowering, and lack of alternative mass-flowering trees late in the season (Williams 1995), or changes in the regional pool of available insects. Proc. Linn. Soc. N.S.W., 123. 2001 241 G. WILLIAMS AND P. ADAM uly-O1> ‘dds -arput ‘dds uly-O1> OOT-Oe> Ol ‘dds OOT-0C LI 9] 06<-0¢ OL<:0b> I 81 91 6C 08<-0€ ‘dds = -atput vi cl cl ‘dds OS<-0S vL 6v Or>-0€ SI cI el “AIPUI uet cc I vl ‘dds OS<-09 ec cv OS-0r ‘dds “07 O0t>-07 ‘dds S-0E< I>:0€< Son[ea qVd:S Td .B, RXR) ‘OSTU (UWI, exe} “OSTU ®, elaydoajog (W, elojdoajoD &, eioidiq ul, eloidiq &, elojdouswAY ul, eiajdousurAY ¢ 904A, sonea qVd-S'1d B, BXP) “OSTUI UW, BX} ‘OSTU 8, e1aydoajog ul, elojdoajod &, elaydiq Ww, eiaidiq &, eiaj}douswAH UW, eiojdouswAY T 994 1661 ‘(BULIOMOTJ Poystuly = ,ULT, ‘sojdures uoousaye = .e, ‘SUIUIOUI = WI, ,‘soroads Jo soquinu = ,-dds, ‘s[eNPIAIpUI JO JoquINU = ,“AIPUI, “SuTUIOU! UT ATUO pa[duwes 1661) €661-166] ‘UOISUTILY ‘so[dures UoOUIAIe pUk SUTUIOUI UL Pa}d[[09 S}JOISUT UdIMJOq SODUSIOTJIG “| IGRI Proc. Linn. Soc. N.S.W., 123. 2001 Table 1 continued 1992 Tree 1 Hymenoptera ‘m’ Hymenoptera ‘a’ Diptera ‘m’ Diptera ‘a’ Coleoptera “m’ Coleoptera ‘a’ misc. taxa ‘m’ misc. taxa ‘a’ BLS:PAB values Tree 2 Hymenoptera “m’ Hymenoptera ‘a’ Diptera ‘m’ Diptera ‘a’ Coleoptera “m’ Coleoptera ‘a’ misc. taxa “m’ misc. taxa ‘a’ BLS:PAB values 1993 Tree 1 Hymenoptera “m’ Hymenoptera ‘a’ Diptera “m’ Diptera ‘a’ Coleoptera “m’ Coleoptera ‘a’ misc. taxa ‘m’ misc. taxa ‘a’ BLS:PAB values Tree 2 Hymenoptera “m’ Hymenoptera ‘a’ Diptera “m’ Diptera ‘a’ Coleoptera “m’ Coleoptera ‘a’ misc. taxa “m’ misc. taxa ‘a’ BLS:PAB values Proc. Linn. Soc. N.S.W., 123. 2001 INSECTS VISITING A RAINFOREST TREE 3 Feb indiv. spp. 4 4 15 9 80 14 21 13 8 3 7 3 2 2) 0 0 0:<10 4 3 4 4 55 24 31 IS) 7 5) 5 3 D 2 1 1 0:<10 28 Jan indiv. spp. 63 17 10 9 89 21 10 8 19 10 1 1 0 0 1 1 40:20 13 5 17 10 32 14 87 16 3 3 6 4 2 2 4 2 <20:<10 12 Feb indiv. spp. 15 24 «4 62 16 Bil ils) 16 «611 6 3 4 3 2 2) 50:30 73). 9) 26 9 124 27 35m aul SS 10 7 9 5 2 2 40:30 4 Feb indiv. spp. ail) ap.) 2 MW mM) —Y) 29 4 8 1 0 0 0 0 7 5 ee a 2 2 | | 2 2) 2 | 0 0 0) 0) <20:<20 20 Feb indiv.spp. 68 14 99 20 43 20 24 15 Sie) 1S) © 3 3 1 1 >50:>60 106 30 30.11 60 27 a M0) Sil 8 13 6 8 5 0 0 >50:>50 11 Feb indiv.spp. 3)5) 9» KO) Dp) Sil 20 13 9 8 8 6 mM) 2 4 4 0 0 40:30 7 & 10 6 Oa | 1 7 4 3 | y 2 28 Feb 6 Mar indiv. spp. _ indiv. spp. 28 7 9 7 26 15 19 10 43 20 21 17 50 DD) Is 9 23 7 10 6 22 13 6 2 4 4 4 4 5 5 2 2 50:10 30:>99 21 5) nS 16 6 V3 2 3 2 6 4 8 6 3 2 >60:>95 40:>95 1 Mar indiv. spp. By) 7 7 11 8 13 11 4 4 5 3 7 7 3 it <10:80 1] 5 10 8 1] 8 14 11 5) 2 3 3 3 2 3 3 <10:80 — i) Pe BD NAA WwW OF <10:Fin — N YN WAW OM FW <20:Fin G. WILLIAMS AND P. ADAM 243 Table 2. Differences between insects collected in morning and afternoon samples, Kenwood Wildlife Refuge, 1992 (‘indiv.’ = number of individuals, ‘spp.’ = number of species;‘m’ = morning, ‘a’ = afternoon samples). 6 Feb 13 Feb 22 Feb 29 Feb Tree 1 indiv. spp. indiv. spp. indiv. Spp. indiv. spp. Hymenoptera ‘m’ 23 14 15 14 14 12 32 19 Hymenoptera ‘a’ 23 14 13 1] 24 11 16 5 Diptera ‘m’ 139 23 180 28 114 24 515 44 Diptera ‘a’ 43 15 62 17 104 21 131 21 Coleoptera ‘m’ 4 4 6 5) 5 4 13 11 Coleoptera ‘a’ fl 4 8 HT 3 3} 10 6 misc. taxa ‘m’ 8 5 9 7 8 3} 7 6 misc. taxa ‘a’ 2 2 5 4 4 4 8 7 BLS:PAB values <30:<5 <30:<20 40:50 30:>90 Tree 2 Hymenoptera ‘m’ 56 30 2D, 19 BS 18 13 11 Hymenoptera ‘a’ 110 50 304 51 49 26 39 18 Diptera ‘m’ Wy 22 139 32 97 20 85 27 Diptera ‘a’ 289 45 370 54 109 22 165 35 Coleoptera ‘m’ 23 16 38 14 14 8 30 11 Coleoptera ‘a’ 24 19 29 12 4 4 20 11 misc. taxa ‘m’ 6 6 9 7 11 7 9 6 misc. taxa ‘a’ 12 6 13 5 5) 5 10 7 BLS:PAB values 40:20 40:>60 30:>90 30:>90 Composition of the visiting fauna No vertebrates, including nocturnal mammals, were observed visiting A. excelsa flowers during this study. Although several species of passerine birds (e.g., Zosterops lateralis, Meliphaga lewinii) foraged for insects amongst foliage and seed clusters, no birds visited flowers. Pollination is achieved principally or solely by insects. Day-time flower visitors were largely Diptera, Hymenoptera and Coleoptera which collectively constituted more than 96 percent of the fauna sampled at Harrington and Kenwood Wildlife Refuge (Table 3). Miscellaneous taxa were principally Hemiptera but also included Blattodea, Thysanoptera, Lepidoptera, Psocoptera, Orthoptera and Collembola. There were considerable differences in the proportion of individual insects and insect orders collected between the two sites, and between seasons at the Harrington site. Flower visitors at Harrington were dominated by Hymenoptera (~37 %: range 27.1 - 46.8) and Diptera (~49 %: range 27.8 - 52.6) (Table 3). Diptera were dominant at Kenwood and averaged approximately 72 % of the diurnal anthophilous fauna. These values compare with those reported by Ireland and Griffin (1984) who record that, on average, approximately 30% of individual insects, visiting Eucalyptus muelleriana A. Howitt (Myrtaceae) in East Gippsland, Victoria, were Diptera and approximately 37% were Hymenoptera. In contrast, Hingston and Potts (1998) found, although there was variation between their study sites, that approximately 68% of all individuals (including introduced Apis mellifera) visiting flowers of Eucalyptus globulus Labill. in eastern Tasmania were Hymenoptera and only 4% were Diptera. Proc. Linn. Soc. N.S.W., 123. 2001 number of insects 244 INSECTS VISITING A RAINFOREST TREE Figure 3. Mean number of individuals and taxa in 10% increments of BLS. 40 — - - — - 35 30 25 4 20 15 10 | ) Lhe 0-10 11-20 21-30 31-40 41-50 51-60 61-70 71-80 81-90 Figure 4. Mean number of individuals and taxa in 10% increments of PAB. 91-100 | LR a en re 40 35 30 25 | 20 | number of insects Thh ebb. 0-10 11-20 21-30 31-40 41-50 51-60 61-70 71-80 81-90 Proc. Linn. Soc. N.S.W., 123. 2001 91-100 individuals Ospecies @ individuals Ospecies G. WILLIAMS AND P. ADAM 245 Table 3. Number of individual insects collected in netted aggregate inflorescence samples (number of days sampled and number of aggregate samples taken each day given in parentheses with total insects collected; Harrington 1991 sampled only in morning, Harrington 1992-3 and Kenwood 1992 sampled morning and afternoon). Harr. % of Harr. % of Harr. % of Kenwood % of 1991 total 1992 total 1993 total 1992 total Tree 1 tot. insects 169 (8:1) 886 (6:2) 530 (4:2) 1557 (4:2) tot. Coleoptera 52 30.8 149 16.8 94 17.8 56 3.6 tot. Hymenoptera 63 37.3 304 34.3 222 41.9 160 10.3 tot. Diptera 47 27.8 401 45.3 199 S75 1290 82.9 tot. misc. taxa U 4 32 3.6 15 2.8 51 3.3 Tree 2 tot. insects 327 (7:1) 1195 (6:2) 302 (4:2) 2266 (4:2) tot. Coleoptera 26 8 194 16.2 31 10.3 182 8 tot. Hymenoptera 153 46.8 324 Dale 99 32.8 628 Pel tot. Diptera 137 42 629 52.6 158 52.3 1381 60.9 tot. misc. taxa 11 3:5 48 4 14 4.6 IS 3.3 Total insects 496 2081 832 3823 Although Kenwood was sampled during one season the number of individuals collected was much greater than that collected during individual seasons at Harrington (Table 3). Spot-lighting at Harrington indicated that small Diptera (e.g., Nematocera) and Coleoptera (e.g., Anthicidae, Scirtidae, Scarabaeidae - Melolonthinae) were active on inflorescences at night. Overall, the majority of insects occurred within the <3 mm - 6 mm size range (Tables 4, 5), however, there was a substantially larger proportion of small-sized fauna at the Kenwood site (97.4 %). The proportion of small-sized insects collected in samples at Harrington did not vary substantially between years (1991: 83.2 %, 1992: 80.9 %, 1993: 86.3 %). Insect families and genera identified from A. excelsa flowers at Harrington and Kenwood Wildlife Refuge are listed in Appendix 1. A list of species is given in Williams (1995). All taxa could be determined to family level, and most could be determined to genus. A small proportion of the collected insects (<5%) could not be determined to genus, and these are cited as numbers of species present in Appendix 1. All taxa are able to contact anthers and stigma. Data in Appendix | indicate seasonal and spatial variation in the insect fauna. Although numbers of individuals collected in netted samples from Kenwood Wildlife Refuge were substantially greater than those collected in any of the three seasons at Harrington, A. excelsa flowers at the later site were visited by a much larger number of taxa. Collectively 116 families, comprising approximately 200 genera (the determination of some genera cited in Appendix 1 is uncertain), were recorded from Harrington and Kenwood Wildlife Refuge. One hundred and four families were collected from inflorescences sampled at Harrington over the total study period (Appendix 1). Seventy-one families were recorded from the Kenwood site, sampled over a single season (1991-92). Seventy-eight families were recorded from the Harrington site in the same season. Proc. Linn. Soc. N.S.W., 123. 2001 246 INSECTS VISITING A RAINFOREST TREE Coleoptera comprised 23 percent of families, Diptera comprised 26 percent of families, and Hymenoptera comprised 30 percent of families collected at Harrington. Comparative values from Kenwood Wildlife Refuge were Coleoptera 31, Diptera 28 and Hymenoptera 25 percent of families respectively. At Harrington there was a high number of aculeate wasp taxa (Williams and Adam 1995) (Appendix 1) in the families Pompilidae, Scoliidae, Sphecidae, Vespidae and Tiphiidae. Only single species of Rhagigaster (Tiphiidae), Scolia (Scoliidae), Sphex, Sphodrotes and Tachysphex (Sphecidae) occurred at Kenwood Wildlife Refuge. The fly fauna at Harrington was particularly rich in Platystomatidae (9 spp.), Stratiomyidae (6 spp.), Sepsidae (6 spp.) and Tachinidae (10 spp.), many species of which were unrecorded from mass-flowering rainforest trees elsewhere in the region (see Williams 1995). Nine butterfly, 5 ant and 7 tachinid genera were collected at Harrington, but none were collected or observed visiting flowers at Kenwood. There was variation in the native bee fauna, but only Homalictus (Halictidae) and Heterapoides (Colletidae) were shared between Harrington and Kenwood. The native apid Trigona carbonaria was only recorded from Kenwood Wildlife Refuge. The introduced honey bee Apis mellifera occurred at both sites. Apis mellifera represented approximately 25 percent of total insects collected at Harrington in 1992, but only approximately 2 percent of insects collected in 1991 and 7 percent of insects collected in 1993 at the same site. Of the 375 insects collected in the 9-12 mm size class at Harrington in 1992, 317 were A. mellifera. Honey bees represented 2.5 percent of all insects collected at Kenwood in 1992. Data on bee visitation is discussed in more detail in Williams and Adam (1997). Flower-dependent insect species peculiar to A. excelsa at Harrington, and not recorded elsewhere in the region (see Williams 1995, pers. observations), included the specialised rhipiphorid Macrosiagon sp., the calliphorid Stomorhina melastoma, the syrphid Dideopsis sp. and a number of aculeate wasps (Williams and Adam 1995). The records for Dideopsis sp. and Stomorhina melastoma extend their known distribution south from Indonesia, Papua New Guinea and Queensland (Kurahashi 1989, Thompson and Vockeroth 1989). Four buprestid species in the nectivorous genus Castiarina (C. acuminata, C. neglecta - Harrington, C. producta, C. oblita - Kenwood) were collected from Alphitonia excelsa flowers, and represent new adult host plant records for these taxa. Castiarina acuminata collected from Guioa semiglauca at Harrington (see Appendix 2) also represents a new adult host record. A number of species collected from A. excelsa also visit the flowers of regional non-rainforest plants (G. Williams unpublished data), however, these generally flower no later (and normally earlier) than A. excelsa. These include the Coleoptera Castiarina neglecta (Leptospermum — Myrtaceae), Torresita cuprifera (Leptospermum, Melaleuca — Myrtaceae, Jacksonia — Fabaceae, Ceratopetalum gummiferum — Cunoniaceae), Ophidius histrio (Baeckea — Myrtaceae), Eupoecila australasiae (Kunzea, Melaleuca — Myrtaceae), Polystigma punctatum (Melaleuca), Glycyphana brunnipes (Syncarpia, Melaleuca — Myrtaceae), Phyllotocus australis (Hakea — Proteaceae, Actinotus— Apiaceae, Leptospermum) and Cheiragra ruficollis (Leptospermum). Species from additional higher taxa await identification. Comparison of fauna with other flowering rainforest trees Insect species shared between flowering Alphitonia excelsa, Acmena smithii, Euroschinus falcata, Scolopia braunii, Guioa semiglauca and Alectryon coriaceus at Harrington are listed in Appendix 2. More exhaustive sampling is likely to expand this list. A full list of species collected from each of these plants, and other rainforest trees sampled in the region, is given in Williams (1995). Months and years in which A. smithii, A. coriaceus, G. semiglauca, E. falcata and S. braunii flowered and were sampled are given in Appendix 2. Unlike A. excelsa populations, that flowered heavily each year at Harrington, flowering of these species was more variable and occurred, generally, over shorter time spans (Williams 1995). Proc. Linn. Soc. N.S.W., 123. 2001 G. WILLIAMS AND P. ADAM 247 Table 4. Size distribution of individual insects collected in netted samples (‘m’ = morning, ‘a’ = afternoon). Harrington 1991 (15 daily aggregate samples; no afternoon samples collected) Hymenoptera total insects mean range Diptera total insects mean range Coleoptera total insects mean range misc. orders total insects mean range Total percent <3mm 3-6mm Harrington 1992 (12 daily aggregate samples) Hymenoptera total insects mean range Diptera total insects mean range Coleoptera total insects mean range misc. orders total insects mean range Total (m+a) percent (m+a) <3mm 1270 61.4 3-6mm m a 15 23 1.3 1.9 0-3 0-5 111 45 9.3 3.8 0-47 0-14 34 37 2.8 3.1 0-8 0-8 15 8 1.3 0.7 0-4 0-2 247 19.5 6-9mm m a 59 3.9 0-29 3 0.2 0- 1 0.07 0-1 63 13.1 6-9mm m a 43 59 3.6 4.9 0-18 0-19 14 6 I OS) 0-7 0-2 1 1 0.1 O.1 0-1 0-1 4 0.3 0-2 128 6.2 9-12mm m a 13 0.9 0-7 2 0.1 0-1 15 3.1 9-12mm m a 183 184 15.3 15.3 0-70 0-67 1 2 0.1 0.2 0-1 0-1 4 1 Os — Obil 0-2 0-1 375 18.1 >12mm 0-1 0-1 0-3 30 1.5 Proc. Linn. Soc. N.S.W., 123. 2001 248 Table 4 continued Harrington 1993 (8 daily aggregate samples) Hymenoptera total insects mean range Diptera total insects mean range Coleoptera total insects mean range misc. orders total insects mean range Total (m+a) percent (m+a) INSECTS VISITING A RAINFOREST TREE <3mm m a 50 31 6.3 3.9 1-21 1-6 159 124 20 15.5 1-74 0-76 63 40 ORS AS 0-28 0-20 12 10 1.5 1.3 0-5 0-4 489 60.6 3-6mm m a 73 48 9.1 6 0-19 0-13 39 26 49 3.3 0-14 0-8 8 6 1 0.8 0-3 0-2 4 3 0.5 0.4 0-2 0-2 207 25.7 6-9mm m a 39 iS 4.9 1.9 0-10 0-6 4 4 05 0.5 0-1 0-3 3 } 0.4 0.3 0-2 0-1 1 0.1 0-1 68 8.4 Kenwood Wildlife Refuge 1992 (8 daily aggregate samples) Hymenoptera total insects mean range Diptera total insects mean range Coleoptera total insects mean range misc. orders total insects mean range Total (m+a) percent (m+a) Proc. Linn. Soc. N.S.W., 123. 2001 <3mm Wiis 52 22.1 44 11-40 7-134 1340 1173 167.5 146.6 82-488 92 85 PES OG 2-28 1-13 46 52 501 (0:5 2-8 2-5 3317 91.6 3-6mm m a 3 20 4b AS) 0-2 0-7 56 45 7 5.6 43-409 36 23 As) WY 0-14 0-7 20 8 DS | 0-6 0-2 211 5.8 6-9mm m a 1 6 0.2 0.8 0-1 0-3 1 0.1 1-27 0-12 3 2 0.4 0.3 0-1 0-2 13 0.4 9-12mm m a 20 16 2.5 D 0-9 0-6 1 0.1 0-1 1 0.1 0-1 1 0.1 0-1 39 4.8 9-12mm m a 28 47 3.5 5.9 0-15 1-7 0-1 2 3 0.3 0.4 0-2 0-1 1 0.1 0-1 81 2) >12mm 0-1 0-1 >12mm 0-1 0.03 G. WILLIAMS AND P. ADAM 249 Table 5. Size distribution of mean number of species collected in netted samples (total numbers are not given because taxa are shared between individual samples; ‘m’ = morning, ‘a’ = afternoon). Harrington 1991 (15 daily aggregate samples; no afternoon samples collected) <3mm 3-6mm 6-9mm 9-12mm >12mm m a m a m a m a m a Hymenoptera mean 351) ils} 0.7 0.7 0.1 range 0-8 0-3 0-3 0-5 0-1 Diptera mean 4.8 1.1 0.1 range 0-11 0-4 0-1 Coleoptera mean 1.9 1 range 0-8 0-3 misc. orders mean 0.6 0.3 0.07 0.1 0.1 range 0-2 0-2 0-1 0-1 0-1 Harrington 1992 (12 daily aggregate samples) <3mm 3-6mm 6-9mm 9-12mm >12mm m a m a m a m a m a Hymenoptera mean 33+ Sho 1.3 1.5 14 2.1 1.4 1.3 OO Os) range 1-10 0-8 0-3 0-4 0-5 = 0-5 0-6 0-3 0-11 0-2 Diptera mean 13.1 10.8 3.6 1.6 1 0.5 0.1 0.2 0.4 range 3-20 5-19 0-47 0-5 0-5 0-2 0-1 0-1 0-3 Coleoptera mean 3.8 2.8 2.1 1.8 0.1 0.1 range 3-5 1-6 0-7 0-7 0-1 0-1 misc. orders mean 1.8 1.5 1 0.5 0.3 0.3 0.1 0.2 range 0-3 0-5 0-3 0-2 0-2 0-1 0-1 0-1 Proc. Linn. Soc. N.S.W., 123. 2001 250 Table 5 continued INSECTS VISITING A RAINFOREST TREE Harrington 1993 (8 daily aggregate samples) Hymenoptera mean range Diptera mean range Coleoptera mean range misc. orders mean range <3mm m a 4 3.6 lz 1-5 5.5 5.4 1-10 0-9 2.8 1.6 0-5 0-3 1.4 0.8 0-5 0-2 3-6mm m 1.6 0.5 0-2 a 2.4 0.5 0-1 0.4 0-2 6-9mm m a 1.8 1.4 0-4 0-2 0.5 0.4 0-1 0-2 0.4 0.3 0-2 ~=0-1 0.1 0-1 Kenwood Wildlife Refuge 1992 (8 daily aggregate samples) Hymenoptera mean range Diptera mean range Coleoptera mean range misc. orders mean range Proc. Linn. Soc. N.S.W., 123. 2001 <3mm m a 15.9 20.4 11-28 9-45 30.1 26 19-36 15-48 6.4 6.6 3-11 1-13 4 3.8 1-6 2-5 3-6mm m 0.3 2.1 0-5 1.8 0-3 a 1.9 6-9mm m a 0.2 0.7 0-1 0-2 0.1 0-1 0.4 0.3 0-1 0-2 9-12mm m a 0.8 0.8 0-1 0-1 0.1 0-1 0.1 0-1 0.1 0-1 9-12mm m a 0.9 1 0-2 1 0.3 0-1 0.3 0.4 0-2 0-1 0.1 0.1 0-1 0-1 >12mm 0.1 0.1 0-1 0.1 0-1 0.1 0-1 >12mm G. WILLIAMS AND P. ADAM 251 Although the sample base from the additional plants is limited by the variation in flowering episodes data in Appendix 2 indicate that numerous taxa are shared between local assemblages of mass-flowering trees. This shared pool includes species (obligates) (e.g., within the families Buprestidae, Cantharidae, Mordellidae, Scarabaeidae, Tiphiidae, Colletidae, Halictidae, Arctiidae) that possess mouthparts adapted to feeding from flowers, and taxa with no obligate morphological adaptation to feeding from flowers (e.g., Chrysomelidae, Coccinellidae, Dermestidae, Melyridae). Shared taxa also include Diptera from the families Calliphoridae, Lauxaniidae and Platystomatidae. These possess mouthparts adapted to feed from a variety of liquids. Interestingly, Trigona, Lasioglossum (Halictidae), Megachilidae and Colletidae- Euryglossinae bees were not recorded from any of the plant species sampled at Harrington but occur elsewhere in the region (Williams and Adam 1997). The scarabaeid beetle genus Phyllotocus, which can occur prolifically on Guioa semiglauca (F. Muell.) Radlk. (Sapindaceae) at Harrington, and Acmena smithii (Poiret) Merr. & Perry and Waterhousea floribunda (F. Muell.) B. Hyland (Myrtaceae) elsewhere in the region, occurred only rarely at the Harrington site on A. excelsa (Williams 1995). Butterflies and cetoniine beetles were collected on Alphitonia excelsa flowers at Harrington but were not collected from other rainforest trees at the site. These included Junonia villibe callibe, Hypolimna bolina nerina, Danaus hamata hamata (Nymphalidae), Delias nysa nysa (Pieridae), Candalides absimilis, Deuodorix epijarbas diovis, Erysichton lineata lineata (Lycaenidae), Graphium eurypylus lycaon (Papilionidae), Polystigma punctatum and Eupoecila australasiae (Scarabaeidae - Cetoniinae). These species are widely distributed in northern New South Wales and have been recorded elsewhere in the study area (Williams 1995, G. Williams and J. Brown unpublished data). Cetoniinae and butterflies undertake frequent interplant movements and are likely to make important contributions to out-crossing. DISCUSSION The slightly sticky pollen of A. excelsa is not easily displaced from the anthers, so it is unlikely that wind pollination occurs. The protandrous pattern of flower opening, and the recurving of petals which articulate the stamens away from the developing ovaries, may militate against self-fertilisation or interference with stigmatic function by self-pollen. Flower longevity is considerably longer than the average of less than two days recorded by Stratton (1989) for Costa Rican tree species, but is similar to the mean flower longevity of most subtropical trees investigated by Williams (1995). There is no obvious adaptation of flowers to visits by specialised insects (e.g., long-tongued bees) and the shallow, readily accessible, perianth allows visitation by numerous insects, the majority of which are less than 6 mm in length. Pollination in A. excelsa appears to be a flexible general enthomophilous system in which the contributions made by individual insect orders and lower rank taxa vary spatially and temporally. Primack (1978) considered that unspecialised floral syndromes may be an adaptation to highly variable pollinator assemblages. Unspecialised flower structures permit visits by a broad range of pollinators, and this flexibility in the use of available pollination vectors facilitates colonisation of new areas (Primack 1978). However, we know little about the cues, environmental or otherwise, that influence pollinator abundance and diversity at A. excelsa populations, and within and across different seasons. Many of the insects visiting A. excelsa also visit flowers of other species. Successful pollination in plants utilising a pool of generalist pollinators is likely to be enhanced if competition between plant species for the same pool of insects is reduced. On the Mid-North coast of New South Wales most lowland rainforest tree species have ceased flowering by the beginning of summer, with the majority of species flowering Proc. Linn. Soc. N.S.W., 123. 2001 252 INSECTS VISITING A RAINFOREST TREE from October to December (Williams 1995, G. Williams unpubl. data). Flowering of A. excelsa populations is relatively synchronous and occurs after this spring-summer flowering peak. Populations flower for approximately 6-8 weeks. Alphitonia excelsa is the last widespread lowland rainforest tree in the region to flower in abundance during summer. This may increase the chances of pollination but also substantially extends the availability of floral resources to nectar and pollen-dependent species (and species preying on them). Alphitonia excelsa, over three seasons at Harrington, provided the greatest number of species collected during our broader study of lowland subtropical rainforest (Williams 1995), and the late season flowering, when few other flowers are available, may explain this abundance. The majority of insect visitors to A. excelsa are mainly Diptera and Hymenoptera but whether these are efficient as pollinators is not known. Williams and Adam (1998) document pollen loads from large-sized Coleoptera, Diptera and Hymenoptera visiting A. excelsa indicating the potential, of at least a subset, of insect visitors to transport pollen. Orders, families and genera varied in daily, seasonal and geographic abundance and diversity on A. excelsa flowers and there were marked differences between the composition of faunas visiting A. excelsa populations at Harrington and Kenwood Wildlife Refuge. Hingston and Potts (1998) record geographic variation in insect assemblages visiting flowering Eucalyptus globulus populations in Tasmania. Although there is temporal heterogeneity in occurrence of insects at A. excelsa blossoms, there is a general peak in abundance of anthophilous taxa and individuals approximately 2-4 weeks after the onset of flowering. Although A. excelsa has not been previously studied, Irvine and Armstrong (1988) discuss pollination of the related A. petriei in tropical Queensland. This species is also a pioneer, mass-flowering, bisexual protandrous rainforest tree. Unlike A. excelsa, A. petriei flowers from September to November (Irvine and Armstrong 1990), during the tropical dry season. Irvine and Armstrong (1988) found that A. petriei has a flexible entomophilous pollination system (similar to that suggested by the floral visitors to A. excelsa) dominated by Diptera and Coleoptera, and to a lesser degree wasps. Apis mellifera was also an active visitor to A. petriei flowers. Irvine and Armstrong (1988) recorded ten Coleoptera species from limited observations on A. petriei, and stated “In terms of cantharophilly [beetle pollination syndrome], it is the specialised inflorescence type, attracting predominantly medium- sized, non-flower-damaging, nectar/pollen-feeding beetles...”. Although we collected 28 beetle families on A. excelsa, only Buprestidae, Cantharidae, Elateridae, Lycidae, Mordellidae, Rhipiphoridae and Scarabaeidae included taxa that are considered specialised blossom visitors. Representatives from these families were generally uncommon, and did not approach the diversity of Coleoptera encountered on mass-flowering Myrtaceae elsewhere in the study region (Williams 1995). Coleoptera are numerous on a number of mass-flowering tree species in late spring and early summer in the region (Williams 1995). In addition to A. petriei, tropical Australian rainforest pollinator faunas, dominated generally by combinations of Diptera, Hymenoptera and Coleoptera, are documented from Flindersia brayleyana F. Muell. (Flindersiaceae) (Irvine and Armstrong 1988, 1990), Neolitsea dealbata (R. Br.) Merr., Litsea leefeana (F. Muell.) Merr. (Lauraceae), and Diospyros pentamera (Wools and F. Muell.) F. Muell. (Ebenaceae) (House 1985, 1989). House (1989) found that Diptera were the most abundant visitors to flowers of dioecious N. dealbata, L. leefeana and D. pentamera, but that Coleoptera were the most important in carrying pollen to pistillate D. pentamera trees. Generalist insect pollination systems are widely recorded from a number of plant communities (e.g., Bawa 1990, Herrera 1988, Moldenke 1975, Petanidou and Ellis 1993, Primack 1978) and have also been recorded from individual plant species (e.g., Ervik and Feil 1997, Kato 2000,). However, rainforest invertebrate assemblages dominated by Diptera, Hymenoptera and Coleoptera are not Proc. Linn. Soc. N.S.W., 123. 2001 G. WILLIAMS AND P. ADAM 253 restricted to anthophilous insects. For example, Basset and Kitching (1991) collected approximately 42,000 arboreal arthropods (which are likely to have included a proportion of anthophilous taxa) in tropical Queensland rainforest using composite ‘malaise-window’ intercept traps designed to minimize taxonomic bias in the fauna being sampled; 27% of individuals were Coleoptera, 41% were Diptera and 11.3% were Hymenoptera. The putative flexible pollination system of A. excelsa would permit it to colonise rainforest margins and regenerating, previously cleared, landscapes without reliance on specialised pollination mutualists. An ability to draw upon a wide taxonomic range of polylectic pollinators, and a lack of dependence on any one taxon, or subset of pollinator taxa, make such species suitable for ‘nurse crop’ plantings in the restoration of rainforest remnants and corridors (although species used in rehabilitation plantings should naturally occur in the landscape subject to restoration). The late season flowering of A. excelsa provides an extension of floral resources to flower-dependent faunas surviving in small lowland rainforest remnants. This late season availability of resources may be important to the conservation of relict insect populations surviving in forest fragments where individual tree species may maintain small populations, and where floral resources are likely to be spatially and temporally limited. ACKNOWLEDGEMENTS We thank Drs David McAlpine, Shane McEvey, Daniel Bickel, Courtenay Smithers (Australian Museum, Sydney), Ken Walker (Museum of Victoria, Abbotsford), Terry Houston (Western Australian Museum, Perth), Graham Brown (Museum and Art Gallery of the Northern Territory, Darwin), and John Lawrence, Elwood Zimmerman, Ian Naumann, Laurence Mound and Don Colless (C.S.I.R.O., Canberra) for identifying a number of the insects. Dr Caroline Gross (University of New England, Armidale) is thanked for suggestions and comments on an earlier version of this paper. Thusnelda Lehner kindly helped with fieldwork. REFERENCES Baker, H.G. (1955). Self-compatibility and establishment after ‘long-distance’ dispersal. Evolution 9, 347- 348. Basset, Y. and Kitching, R.L. (1991). Species number, species abundance and body length of arboreal arthropods associated with an Australian rainforest tree. Ecological Entomology 16, 391-402. Bawa, K.S. (1990). Plant-pollinator interactions in tropical forests. Annual Reviews in Ecology and Systematics 21, 399-402. Dafni, A., Lehrer, M. and Kevan, P.G. (1997). Spatial flower parameters and insect spatial vision. Biological Reviews 72, 239-282. Ervik, F. and Feil, J.P. (1997). Reproductive biology of the monoecious understorey palm Prestoea schultzeana in Amazonian Ecuador. Biotropica 29, 309-317. Gross, C.L. (1993). The breeding system and pollinators of Melastoma affine (Melastomataceae); a pioneer shrub in tropical Australia. Biotropica 25, 468-474. Harden, G.J. (ed.) (1990). Flora of New South Wales. Volume 1. (New South Wales University Press, Kensington). Harrington, G.N., Irvine, A.K., Crome, F.H.J. and Moore, L.A. (1997). Regeneration of large-seeded trees in Australian rainforest fragments: a study of higher order interactions. In Tropical forest remnants: ecology, management, and conservation of fragmented communities. (Eds. W.F. Laurance and R.O. Bierregaard). (The University of Chicago Press, Chicago). Herrera, J. (1988). Pollination relationships in Southern Spanish Mediterranean shrublands. Journal of Ecology 76, 274-287. Hingston, A.B. and Potts, B.M. (1998). Floral visitors of Eucalyptus globulus subsp. globulus in eastern Tasmania. Tasforests 10, 125-173. House, S.M. (1985). Relationships between breeding and spatial pattern in some dioecious tropical rainforest trees. Ph.D. thesis. Australian National University, Canberra. House, S.M. (1989). Pollen movement to flowering canopies of pistillate individuals of three rainforest trees in tropical Australia. Australian Journal of Ecology 14, 77-94. Ireland, J.C. and Griffin, A.R. (1984). Observations on the pollination ecology of Eucalyptus muelleriana Howitt in East Gippsland. The Victorian Naturalist 101, 207-211. Proc. Linn. Soc. N.S.W., 123. 2001 254 INSECTS VISITING A RAINFOREST TREE Irvine, A.K. and Armstrong, J. (1988). Beetle pollination in Australian tropical rainforests. Proceedings of the Ecological Society of Australia 15, 107-113. Irvine, A.K. and Armstrong, J. E. (1990). Beetle pollination in tropical forests of Australia. In Reproductive ecology of tropical forest plants. Volume 7. (Eds. K.S. Bawa and M. Hadley). (Unesco, Parthenon, Carnforth). Kato, M. (2000). Anthophilous insect community and plant-pollinator interactions on Anami Islands in the Ryukyu Archipelago, Japan. Contributions from the Biological Laboratory, Kyoto University 29, 157-252. Kitching, R.L., Orr, A.G., Thalib, L., Mitchell, H., Hopkins, M.S. and Graham, A.W. (2000). Moth assemblages as indicators of environmental quality in remnants of upland Australian rainforest. Journal of Applied Ecology 37, 284-297. Kurahashi, H. (1989). Family Calliphoridae. Chapter 109. In Catalog of the Diptera of the Australasian and Oceanian Regions (Ed N.L. Evenhuis). (Bishop Museum Press, Honolulu, and E.J. Brill, Leiden). Laurance, W.F. (1997). Hyper-disturbed parks: edge effects and the ecology of isolated rainforest reserves in tropical Australia. In Tropical forest remnants: ecology, management, and conservation of fragmented communities. (Eds. W.F. Laurance and R.O. Bierregaard). (The University of Chicago Press, Chicago). Moldenke, A.R. (1975). Niche specialisation and species diversity along an altitudinal transect in California. Oecologica (Berl.) 21, 219-242. Petanidou, T. and Ellis, W. (1993). Pollinating fauna of a phryganic ecosystem: composition and diversity. Biodiversity Letters 1, 9-23. Primack, R.B. (1978). Variability in New Zealand montane and alpine pollinator assemblages. New Zealand Journal of Ecology 1, 66-73. Stratton, D.A. (1989). Longevity of individual flowers in a Costa Rican cloud forest: ecological correlates and phylogenetic constraints. Biotropica 21, 308-318. Thompson, F.C. and Vockeroth, J.R. (1989). Family Syrphidae. Chapter 51. In Catalog of the Diptera of the Australasian and Oceanian Regions (Ed N.L. Evenhuis). (Bishop Museum Press, Honolulu, and E.J. Brill, Leiden). Williams, G.A. (1993). Hidden rainforests: subtropical rainforests and their invertebrate biodiversity. (New South Wales University Press, Kensington). Williams, G.A. (1995). Pollination ecology of lowland subtropical rainforests in New South Wales. Ph.D. thesis. University of New South Wales, Kensington. Williams, G.A. and Adam, P. (1994). A review of rainforest pollination and plant-pollinator interactions with particular reference to Australian subtropical rainforests. The Australian Zoologist 29, 177-212. Williams, G.A. and Adam, P. (1995). Records of aculeate wasps from flowering subtropical rainforest trees. The Australian Entomologist 22, 51-58. Williams, G.A. and Adam, P. (1997). The composition of the bee (Apoidea: Hymenoptera) fauna visiting flowering trees in New South Wales lowland subtropical rainforest remnants. Proceedings of the Linnean Society of New South Wales 118, 69-95. Williams, G.A. and Adam, P. (1998). Pollen loads collected from large insects in Australian subtropical rainforests. Proceedings of the Linnean Society of New South Wales 120, 49-67. APPENDIX 1 List of insect families and genera collected from A. excelsa at Harrington (1990-93) and Kenwood Wildlife Refuge (1991-92) (genera not determined for all families) (numbers in columns indicate number of species recorded in individual seasons). Families Genera Harr Harr Harr Ken 90-91 91-92 92-93 91-92 BLATTODEA Blattellidae Ectoneura 1 COLEOPTERA Anobiidae Anthicidae Anthicus 1 1 1 1 ?Anthicus 1 Attelabidae Auletobius 1 1 Buprestidae Castiarina 2 2 zy Cisseis 1 Melobasis 1 Neocuris ] Torresita 1 Cantharidae Chauliognathus 1 ] Carabidae Sarcothrocrepis 2 Cerambycidae Tillomorpha I Proc. Linn. Soc. N.S.W., 123. 2001 Chrysomelidae Cleridae Coccinellidae Corylophidae Curculionidae Dermestidae Elateridae Euglenidae Hydrophilidae Lathridiidae Lycidae Melyridae Mordellidae Nitidulidae Phalacridae Pythidae Rhipiphoridae Scarabaeidae Scirtidae Staphylinidae Tenebrionidae G. WILLIAMS AND P. ADAM Cryptocephalus Ditropidus ?Ditropidus Monolepta Lemidia Amidellus Cryptolaemus Egleis Epilachna Orchus Rhizobius ?Rhizobius Rodolia Sericoderus Balanerhinus Baris Cyttalia Meriphus Neolaemaraccus Anthrenocerus ?Anthrenocerus Thaumaglossa Megapenthes Ophidius Aderus ?Aderus Pseudohydrobius Cortinicara Porrostoma ?Porrostoma Carphurus ?Carphurus Dicranolaius Helcogaster ?Helcogaster Neocarphurus Hoshihananomia Mordella Mordellistena ?Mordellistena Tomoxia ?Tomoxia Epuraea Litochrus ?Litochrus Parasemus ?Phalachrus Paromarteon Macrosiagon Cheiragra Eupoecila Glycyphana Heteronyx Phyllotocus Polystigma Pseudomicrocara ?Pseudomicrocara Scirtes ?Scirtes genus near Anotylus ?Apellatus Ecnolagria Nocar Ommatophorus 255 1 1 2 1 >/=4 1 5 1 1 1 1 1 1 1 D: 1 1 1 1 1 1 1 1 1 1 1 >1 1 1 1 1 1 1 1 1 1 1 2 1 1 1 1 1 1 1 >2 1 1 1 >1 >1 >2 1 1 1 1 >1 >1 1 >1 1 1 1 1 1 1 1 1 1 1 1 2 1 2 1 1 1 Proc. Linn. Soc. N.S.W., 123. 2001 256 COLLEMBOLA Entomobryidae DIPTERA Bibionidae Bombyliidae Calliphoridae Cecidomyiidae Ceratopogonidae Chironomidae Chloropidae Culcidae Dolichopodidae Drosophilidae Empididae Ephidridae Lauxaniidae Milichidae Muscidae Mycetophilidae Nemestrinidae Phoridae Platystomatidae Psychodidae Rhagionidae Sarcophagidae Scatopsidae Sepsidae Stratiomyidae Syrphidae Tabanidae Tachinidae ?Lepidosira Bibio Geron Ligyra Pseudopenthes Chrysomya Paramenia Stomorhina Apotropina Krakatauia Drosophila Leucophenga Nesiodrosophila Homoneura Melanina ?Melanina Sapromyza Steganopsis Trypetisoma Musca Cyclopsidea Duomyia Euprosopia Microepicausta Pogonortalis Rivellia Chrysopilus Sarcorohdendorfia Australosepsis ?Lasionemapoda ?Parapalaeosepsis Sepsis Acanthasargus Damaromyia Hermetia Odontomyia Dideopsis Eristalinus Mesembrios ?Xanthogramma Scaptia ?Tabanus Austrophorocera Blepharella ?Blepharella Blepharipa Prosena Rutilia Proc. Linn. Soc. N.S.W., 123. 2001 INSECTS VISITING A RAINFOREST TREE Ne ee — — = — SBN ee >1 257 G. WILLIAMS AND P. ADAM x Xx ‘ds sndadOuadyqUuy ¢ X Xx ‘ds snaaouadyjup depHsowog Xx ‘ds snuiysaupjpg oepluolnoing Xx Xx X ‘ds snsapoo1sag oeprydoyA1og Xx 12]2qa0y DIIDPOY X Xx X x Xx ‘ds sniqoziyy x Xx SNIADAOASIU ADAU ‘ds snigoZzIYy IEPI][OUIIDOD X pynosniuiu; pjdajouopy Xx X Xx sypajsnv pjdajouop Xx K x ‘ds snpidoajiq oeprljowosAly) x stuuadianyf¢ snyjousoynvy) oeplieyjue) Xx DIDUIWNID DULADIJSDD ovprsoidng Vad LdOd TOO IOOV Wash [eq TWISV eIgS psjaoxa piuoj1yd]p €6-1661 JEN uel ‘vsjaoxa piuopydiy svoovuUleyYy 76 (0661 99d “AON 76 “0661 ‘AON “21YJUs Duawoy “WWUuSy ‘DINDIBIWAS DOINL) “WI3St) avooeyA\| C6" 166] “Ue 0661 ‘AON “390 “Unb vIdojoozg “e.1GS ‘Snaav1409 UOKAJIA]P “IOV QvddVIJINOIL] J ovooepuides 16 ‘0661 (99d “AON ‘BIDIIvf snu1yosoing “yeyq IeOORIPIVIVUY ‘(SULIOMOTZ JO 9dUONbas UI poyqr} soroods yueld ‘vsyaoxa ‘py 10J yd90x9) po[dures sivod pue syyuow ‘sorseds jueyd :Aoy “UOJSULLICH 3e SddIj JSoIOJUIeI SULIOMO]J-SSEW INYO pue Dsj]adxXa D1IUOJIYd] Pp Ud9MjoqQ Poreys sotdeds jo9sSU] c XIGNHddV Proc. Linn. Soc. N.S.W., 123. 2001 INSECTS VISITING A RAINFOREST TREE 258 * * pajz0p Spj4OUuosog ‘ds pjsnva1da0A,d1y ‘ds piduong pjaid piduong ‘ds pwosyadasy AajJSDBOUDJaU Sisdounsajs sdaa14381u pzAuosdv¢ ‘ds pjiydosoiq 1OUIG]DIDUL DINDIDYDAY ‘ds pu1ysowuojgy ‘ds pluawuviavdg ‘ds u0sa4) AOJDIIUAL O101G ‘ds pADIOAIIWMOpNas_ ¢ ‘ds pavo0AJMOpNnasd sypajsnv snz0jo1]4yd ‘ds xAuosajayy SIJOIYNA DABDAIOYD ‘ds snaav]Oyd ¢ ‘ds snwuasvavd ‘ds SnayIONT SNIDINIDULZ SNAYIONT ‘ds pavandy ‘ds DjJaps4ow sispqusnsun¢ Snanydivr0an ‘ds snanydavy ‘ds DADIIUIAOD ‘ds SndapPy ¢ sepHyeulo}sAjeyd oepriuexne seprTydosoiq oeprpodoysijoq seproydiy[e) oepryAquiog oepluoigig VudLdia ePHIOS aeproeqeivos aeplioeyeyd Sep! [NPVIN Sepl|]9P1oW aeplAjayy oepuplyyey] aepruapsng Proc. Linn. Soc. N.S.W., 123. 2001 259 G. WILLIAMS AND P. ADAM Kx KK OK ac x Kx KKK RK ca Kx KK RK Kx KKK Kx KK KK ‘ds pipup ‘ds s1yaDysd]Og psopid¢ siyopysdjog Oy? s1yaDYysA] Og DJDUIIOABIU DIDBUAA ‘ds xaWACWOPIAT ‘ddsj‘ds sajspsojpuad) SNUBISDBIU SNJI1]DUOLT Is4opullf snjo1jpUox SNJDISIAIg SNJIOAdOIAT apiajxa Avau ‘ds sapiodpsajayy snjjasopignu snanjdydup pAéafyjau sidp SES 21 SOc 1QOYAAOYJUDX DIAOG2]27 DNBIUOI¢ DIAOG2]AZ SISUAAPUDLY ADAU “dS 1dJSDB1IBDYY SISUBIUDUSD] ¢ Snuudyjoyoody SnJDIpIMip snuudyjoydsouiqg AOJOI1g DUI, ‘ds xaydsdyovy], ‘ds snyjayjsoyjuvop ‘ds piupaq ‘ds purynyjods ‘dS DUWDABOYIUDY ¢ syuissip sisdagy ‘ds pifséopuapysOIID, aepinory VuUALdOddaT SEPIOIULIO] BOPIOOIWLIO J oepnoleH aepHay[OD avpidy voplody aepidsa,, oepriydty aeprooyds oeplueaq Vad LdONANAH oepniyday, oeprydidg sepisdag ovpiseydoores Proc. Linn. Soc. N.S.W., 123. 2001 s ine dinesienereats ino Abus man qe = i ~ wagitean® nena Bodvotigesahetalp i a os, ait mina aah a webatih aa en alee, ne ae ae ! 27 eta - sehieainae’ ‘Sabioumot rc ix aye age Wap ge. > FE OREIN 5 = BATION : ae sal meearpcornacd ) _ARTVTOOPEES BOOK REVIEW THE FORGOTTEN NATURALIST In search of Alfred Russel Wallace John G. Wilson (2000) Australian Scholarly Publishing, Kew, Victoria RRP $34.95 The key to this book is the subtitle; it is the story of the author’s travels along some of the pathways trod by one of the great naturalists of all time. Read that way it is a pleasant travelogue with a theme of particular interest to anyone interested in natural history. There are some biographical details of Alfred Wallace and some interesting snippets that arose in the course of the author’s travels (such as the origin of the term “antimacassar” which comes from Macassar Oil used by Victorian gentlemen on their hair. The source of that oil was of course Macassar). This book is not an in depth analysis of the roll played by Wallace in the shaping our understanding of evolution. His tremendous insight based on field observations and a flash of genius, said to have occurred while in his sick bed with malaria, led him to see the role of survival of the fittest in shaping new species. There are occasional hints in the text, even in the title, that Wallace’s place in history has not been recognised, however I think this is not the case. As an undergraduate I was certainly introduced to Wallace, and no Australian biologists would fail to recognise the biogeographical significance of Wallace’s line. I suspect, although it is not spelled out, that the author might believe that Wallace is forgotten as the co-founder, or even by precedence the founder of the modern understanding of evolution and the role of “survival of the fittest”. Wallace himself would and did dispute that, as would any practicing scientist. Each of us may hope to have made a contribution to our field of study, but we know full well that any advance is based on groundwork laid by those before us. The ideas of neither Wallace nor Darwin came “out of the blue” in a flash of genius. Both were mightily influenced by the writings of Maltheus and the geology of Lyell. As modern biologists build their work on the shoulders of Darwin, so Darwin and Wallace continued along a path started by others, going right back to Aristotle who was well aware of the roll of the environment in the manifestation of physical characters. The concept that progress occurs in sudden leaps due to a flash of genius is part of the folklore of science but far from an accurate view of the workings of scientists. And it is not good enough to simply have the idea. It needs to be expostulated and supported. Nothing in Wallace’s life as set out in this book indicates that he would have had the patience to amass the material on which Darwin built his argument. None the less, Wallace’s article “On the law which has regulated the introduction of new species” published September 1855 in Annals and Magazine of Natural History was truly remarkable, setting out clearly that evolution had taken place and outlining an early version of “gradualism”. I agree with John Wilson that it was the most important contribution to evolutionary theory before the publication of Darwin’s “On the origin of species”. It was however not “world shattering” as the world of natural scientists was at that time very small. In those days scientific society meetings, such as of the Royal and Proc. Linn. Soc. N.s.w., 123. 2001 262 THE FORGOTTEN NATURALIST Linnean Societies, were focal points of active discussion of issue of the day as well as a venue for the presentation of “read” papers. It is inconceivable that Darwin’s ideas were not bandied about at such gatherings. Darwin did not keep them secret and had clearly been discussing his developing thoughts with the major scientific figures of the day, especially the geologist Lyell and the botanist Hooker. Therefore the earth shattering aspect of Wallace’s “Law Paper” was not the ideas themselves as much as their appearance from an amateur naturalist in Indonesia. Lyell, and to a lesser extent Hooker, urged Darwin to publish. When he did so the basic ideas were supported by careful exposition and wide observations. None of this is meant to detract from Alfred Russel Wallace as one of the great naturalists of all time. And much in this book is of particular interest because it shows the tremendous effort and enthusiasm that Wallace put into field work. After returning from the far east he was to put that enthusiasm into a number of causes, many of which seem well in advance of his time. Some of the most interesting material in the book “The Forgotten Naturalist” concerns these activities. There are some odd statements in the book, such as “Tarsiers are primates not lemurs” (lemurs are primates) and “animals and birds”. However the author does not claim to be a biologist, although some tight editing by a scientific editor would have been useful. But basically this is a travel book and, as I said at the beginning, should be read as such and will provide much enjoyment. M.L. Augee 31 October 2001 Proc. Linn. Soc. n.s.w., 123. 2001 a snes: Sobietie, ere fs 5 wenne jot the presentation af ADEN. ae > Tiel trandied show al-nnch patos Daj dd cl 2. . beta discussing, is sovelopitig | thoughhy: with: iN ene pientific fiswes of bee "© Sepecially the seologiet Lyell and the beranint Mopker “Therefore ihe earth shat gypect of Wallace's "Law Paper” was nol the Aides thessmaeftes as much ac shear a > fen an ameter naturale ip: Bidexesia,: Lyell pad we ieieerenwit Hooker, 4 | ikea - ihe Aina yt nie ime, Ka ome ct te most contesting mate era tt nthe $ book The: a cen ‘erricertis these Activities: om am ae a “Phare ue, sone! end siaberoulits mt . Caynars, are brates} and: “anmials nad. Say? Y Pe, io palilish, When hé did 40 a basic: ee ed att std ne be ee sali ie Srasten $ pia firiniats ot author ae i pa, WeuBN mnne. tight esting vite isa peer book ae a1 I said : .7a4 / eine tre low Oc cay. 2O1= i re ae Teter hen: qo Bent (bere a AO! pean i Aras Feet Vella 20, ae i tao nes 225 M.J. WuiteELAw, B.D. Batts, C.V. Murray-WaLLACE AND C.R. McRae Diagenesis of the organic matrix in Anaaara trapeziaduring the Late Quaternary: Preliminary Findings 235 G. WiLLAms AND P. ADAM The insect assemblage visiting the flowers of the subtropical rainforest pioneer tree Ajphitoniia exce/sa (Fenzl) Reiss. ex Benth. (Rhamnaceae) 261 Book REviIEw The forgotten naturalist - in search of Alfred Russel Wallace by John G. Wilson The Linnean Society of New South Wales publishes in its proceedings original papers and review articles dealing with biological and earth sciences. Intending authors should contact the Secretary (PO Box 137, Matraville NSW 2036, Australia) for instructions for the preparation of manuscripts and procedures for submission. Instructions to authors are also available on the society’s web page (http://www.acay.com.au/~linnsoc/welcome.html). Manuscripts not prepared in accordance with the society’s instructions will not be considered. PROCEEDINGS OF THE LINNEAN SOCIETY OF N.S.W. VOLUME 123 PRT Se ie eA Ey TEL RW ANSGAR Issued 22 December 2001 CONTENTS 1 23 39 89 P. AJANI, G. HALLEGRAEFF AND T. PRITCHARD Historic overview of algal blooms in marine and estuarine waters of New South Wales, Australia. ; G.R. BaNN AND B.G. JONES The Coolangatta latite member and associated tuffs: newly identified basal units in the Gerringong volcanics, southern Sydney Basin, NSW W.B.K. Homes The Middle Triassic megafossil flora of the Basin Creek Formation, Nymboida Coal Measures, New South Wales, Australia. Part 2. Filicophyta E.A. JEFFERYS AND B.J. Fox The diet of the Pilliga mouse, Pseudomys PiligeoteX ore: Muridae), from the Piue. Scrub, northern New South Wales . K. KiERNAN, K. AND A. MCCONNELL Land surface rehabilitation research in Antarctica 1.D. LINDLEY Tertiary echinoids from Papua New Guinea M.S. Moutps AND K.A. KOPESTONSKY A review of the genus Aobonga Distant with the description of a new species (Hemiptera: Cicadidae). A. REID Anew cuttlefish, Seova graharm, sp. nov. (Cephalopoda: Sepiidae) from eastern Australia R.B. Rickarps, |.G. Percivat, A.J. Simpson AND A.J. WRIGHT Silurian biostratigraphy of the Cadia area, south of Orange New South Wales B.V. Timms Limnology of the intermittent pools of Bell’s Creek, Paroo, arid Australia, with special eerie to biodiversity of invertebrates and succession. J.A. Wess Anew marattialean fern from the Middle Triassic of eastern Australia Printed by Southwood Press Pty Ltd, 76-82 Chapel Street, Marrickville 2204 == 0 — + ——— ue Ranlenins ein wat un eg ene Senos