PROCEEDINGS of NEW SOUTH WALES VOLUME 130 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. The Society awards research grants each year in the fields of Life Sciences (the Joyce Vickery fund) and Earth Sciences (the Betty Mayne fund), 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. It holds field excursion and scientific meetings, including the biennial Sir William Macleay Memorial Lecture delivered 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 $456.00. The current subscription rate to the Proceedings is set at A$110.00 per volume. In recent years a volume consists of a single annual issue. Back issues of all but a few volumes and parts of the Proceedings are available for purchase. Prices are listed on our home page and can also be obtained from the Secretary. OFFICERS AND COUNCIL 2007/2008 President: Michele Cotton Vice-presidents: M.L. Augee, I.G. Percival, D.R. Murray Treasurer: 1.G. Percival Secretary: J-C. Herremans Council: A.E.J. Andrews, M.L. Augee, J.P. Barkas, M. Cotton, M.R. Gray, J-Cl. Herremans, D. Keith, R.J. King, H.A. Martin, E. May, D.R. Murray, Joelle Myerscough, I.G. Percival, J. Pickett, S. Rose, H.M. Smith and K.L. Wilson Editor: M.L. Augee Assistant Editor: Elizabeth May Auditors: Phil Williams Carbonara The postal address of the Society is: P.O. Box 82, Kingsford NSW 2032, Australia Telephone: (International) 61 2 9662 6196; (Aust) 02 9662 6196 E-mail: linnsoc @iinet.net.au Home page: http://linneansocietynsw.org.au Cover motif: Reconstruction of Palorchestes from the paper by B.S. Mackness, page 30 this volume. PROCEEDINGS of the LINNEAN SOCIETY of NEW SOUTH WALES For information about the Linnean Society of New South Wales, its publications and activities, see the Society’s homepage http://linneansocietynsw.org.au VOLUME 130 March 2009 u x aE on ihe . ris ry Tali sit i’ AS q 4 : WEAR ORCAS ELT LAT ab Om Aa Wart NO earch cles 4 hey upd rorterrrrd ter 4 SGP Se VIMO Ul Bao earteyeiag ese Arh. Ccmeate CD: ET PTE ert Museum Holdings of the Broad-headed Snake Hoplocephalus bungaroides (Squamata: Elapidae) JAMIE M. HARRIS AND Ross L. GOLDINGAY School of Environmental Science and Management, Southern Cross University, Lismore NSW 2480, Australia; Harris, J.M. and Goldingay, R.L. (2009). Museum holdings of the Broad-headed Snake Hoplocephalus bungaroides (Squamata: Elapidae). Proceedings of the Linnean Society of New South Wales 130, 1-19. The broad-headed snake Hoplocephalus bungaroides (Schlegel, 1837) is a highly endangered species endemic to the Sydney basin. We attempted to track down the whereabouts of museum specimens of this snake by contacting mainly Australian, European and North American curators of natural history museums and university herpetological collections. We received replies from 200 institutions, and from these we present details of 159 specimens from 27 museums in 11 countries reported to us as H. bungaroides. Countries include Australia (108 specimens), Germany (13), the United States (9), United Kingdom (7), France (4), Belgium (5), the Netherlands (5), Austria (3), Denmark (3), Italy (1), and Switzerland (1). At least 47 specimens are from the 19" Century, and accurate locality records were available for 98 specimens. Obviously, all of the specimens have value insofar as they may provide important biological data that will be useful to researchers working on the future conservation of this snake. Many of these specimens also provide important historical evidence of the species’ past distribution. Manuscript received 6 February 2008, accepted for publication 17 September 2008. KEYWORDS: conservation, distribution, museum, reptile, natural history, Krefft INTRODUCTION The broad-headed snake Hoplocephalus bungaroides is possibly the most endangered snake in Australia, with research indicating there are serious concerns for its future conservation (Shine and Fitzgerald 1989; Webb and Shine 1997, 1998a,b; Goldingay 1998; Shine et al. 1998; Goldingay and Newell 2000; Webb et al. 2002; Newell and Goldingay 2005). It has a highly restricted distribution within the Sydney basin where it is dependent on habitats characterised by sandstone cliffs, ridges and outcrops (Krefft 1869; Longmore 1989; Cogger 2000; Swan et al. 2004). This species is threatened by habitat loss through urbanisation, removal of bush rock for landscaping and ongoing degradation of rocky habitat caused by hikers and reptile poachers (Hersey 1980; Shine and Fitzgerald 1989; Cogger et al. 1993; Goldingay and Newell 2000; Webb and Shine 1998a, 2000; Newell and Goldingay 2005). The decline of H. bungaroides was noted as early as 1869 by Gerard Krefft (1830-1881), Curator and Secretary of the Australian Museum (Whitley 1961, 1969), in The Snakes of Australia (Krefft 1869), the first monograph published on Australian snakes. Krefft (1869) considered H. bungaroides (as its junior synonym Hoplocephalus variegatus) to be “very local” with specimens found only “in the immediate neighbourhood of Sydney”, that is, from Port Jackson to Botany Bay, on the shores of Middle Harbour, and at Lane Cove and Parramatta inlets. Krefft stated that this snake is “not so numerous as they were six or eight years ago” (i.e. around 1861-1863) and the decline was attributed to “their haunts having been invaded by the builder and the gardener”. Krefft also stated that “many hundreds” of H. bungaroides specimens had been distributed to unnamed “kindred institutions”. These statements sparked our curiosity, and subsequently we made considerable effort to locate these specimens. In so doing, we also aimed to gather information on all museum holdings of H. bungaroides because this may offer a rich source of data potentially useful to the future conservation of this endangered species. MUSEUM HOLDINGS OF THE BROAD-HEADED SNAKE MATERIALS AND METHODS We reviewed the annual reports of the Australian Museum for mention of reptile specimens received and exchanged by Krefft during his tenure as Curator and Secretary (1861-1874) (see Appendix 1). We also searched for information in the archives of the Australian Museum, including examination of Krefft’s correspondence and the ‘Exchange Register’ (pre 1874; series 58, Volume 1). Finally, we surveyed other museums and related institutions with herpetological collections, particularly those in Australia, Belgium, Czech Republic, France, Germany, India, Italy, Netherlands, Portugal, Spain, United Kingdom (UK) and the United States (US) since Krefft did send reptile specimens to these countries (Appendix 1). Museums in these and other countries were identified using online directories and also published lists in Leviton et al. (1980, 1985) and Roselaar (2003). Curators or collection managers were asked via email whether there were any H. bungaroides (or its synonym H. variegates) in their museums. If H. bungaroides was present, data were requested on numbers of specimens held; catalogue / registration numbers; collection locality; collector or donor name; collection date; and other details recorded with the specimens. Additionally, photographs of the specimens were requested to confirm that the correct identifications had been made. In relation to photographs, for one museum in France (Musée de Zoologie, Strasbourg) we received reports about two H. bungaroides specimens in their collection, but the photographs supplied did not reveal the striking appearance of H. bungaroides and we believe they represent the Stephens banded snake H. stephensii. We are confident about the identification of all other museum specimens listed, except for those at Zoological Museum, University of Liege (Belgium) because photographs of the five H. bungaroides in their collection were not supplied. RESULTS The annual reports of the Australian Museum for 1861-1874 did not provide details of “many hundreds” of H. bungaroides. Descriptions of species exchanges in these reports lack detail, and indicate, at a minimum, that H. bungaroides was definitely sent out to only three places (see Appendix 1). The reports mention that reptiles were shipped to a number of museums and specimen dealers in this period, but the specific composition of the shipments was generally not published. Recipients of Krefft’s reptiles included his colleagues in Mauritius (Victor de Robillard) and India (Richard Henry Beddome); one learned society (Royal Society of Tasmania); four specimen ‘dealers’ - J.C. Puls (Belgium), C.L. Salmin (Hamburg, Germany), Vaclav Fri¢ (Prague, Czech Republic), and Robert Damon (Weymouth, England); and at least nine museums, i.e. those in Hamburg and Berlin (Germany), Leiden (Netherlands), Madras (=Chennai, India), Milan (Italy), Paris (France), Madrid (Spain), London (UK), and Harvard at Cambridge (US). The pre-1874 Exchange Register in the Australian Museum archives (series 58, Volume 1) contained some inbound 1860s correspondence addressed to Krefft from dealers such as J.C. Puls and some museums, such as the Muséum National d’Histoire Naturelle, Paris and the Museum of Comparative Zoology, Cambridge. This Register also lists some, but not all, specimens sent on exchange by Krefft and also his predecessor George Bennett. These lists include an entry that a single H. bungaroides was sent to the Government Museum at Madras (=Chennai, India) (see Exchange Register p.16). Whilst this list is undated, it was probably the same consignment listed in the annual report for 1864 (see Appendix 1). The Exchange Register also itemised specimens dispatched to the Royal Society of Tasmania and H. bungaroides was absent from this list. By contacting museums directly, we located 159 specimens reported to us as H. bungaroides from 28 institutions in 11 countries (Table 1). Most specimens we found are held in Australia (108 specimens), but a considerable number are in Europe (43 specimens) and the US (9 specimens). Negative responses to our email enquiries were received from 174 institutions (see Appendix 2). There were also 74 other institutions that did not respond to our correspondence, despite more than one request (Appendix 2). We have compiled some detailed information on H. bungaroides specimens from many institutions in Australia, Europe and the US (see below). Australian collections The Australian Museum, Sydney (AM), has 77 H. bungaroides specimens (Table 1; Appendix 3) but none of these are designated type specimens (Shea and Sadlier 1999). Eight of these do not have any locality data and another 5 have an imprecise collection locality recorded as “Sydney”. There are 18 AM specimens collected at Waterfall, seven at Nowra, five at Long Bay, six at Royal National Park (NP) (including Bundeena), three at the Blue Mountains, three at Woronora Dam and two at La Perouse. Single AM specimen locality records were recorded for 20 locations (Appendix 3). Twenty four (31 %) of the Proc. Linn. Soc. N.S.W., 130, 2009 J.M.HARRIS AND R.L. GOLDINGAY Table 1: Specimens of Hoplocephalus bungaroides held in Australian and overseas museums. Institution Code n Australia Australian Museum, Sydney AM Vi Western Australian Museum, Perth WAM 3 Museum Victoria, Melbourne NMV 6 South Australian Museum, Adelaide SAMA 6 Queensland Museum, Brisbane QM 4 Northern Territory Museum, Darwin NTM 4 Australian National Wildlife Collection, Canberra ANWC 3 Macleay Museum, University of Sydney MMUS 3 Biological Museum, Australian National University ANU 2 Austria Museum of Natural History, Vienna NMW 3 Belgium Zoological Museum, University of Liege MZULG 5 Denmark Zoological Museum, University of Copenhagen ZMUC 3 France Muséum National d’ Histoire Naturelle, Paris MNHNP 4 Germany Museum fiir Naturkunde, Berlin ZMB 8 Senckenberg Natural History Museum, Frankfurt SMF 1 Zoologisches Museum, University of Hamburg ZMH 2 Zoologische Staatssammlung, Munich ZSM 2 Italy Museo Civico di Storia Naturale, Genoa MSNG i ~ Netherlands National Museum of Natural History, Leiden RHNH 5 Switzerland Naturhistorisches Museum, Basel NMB 1 United Kingdom Natural History Museum, London BMNH 6 Oxford University Museum of Natural History OUM 1 United States Field Museum of Natural History, Chicago FMNH 2 Museum of Comparative Zoology, Harvard University MCZ 3 National Museum of Natural History, Smithsonian Institution USNM 2 San Diego Natural History Museum SDNHM 1 University of Illinois Museum of Natural History UIMNH 1 Total 159 Proc. Linn. Soc. N.S.W., 130, 2009 MUSEUM HOLDINGS OF THE BROAD-HEADED SNAKE 77 AM specimens do not have collection dates, but presumably some of the undated specimens are very old and derive from the late 19" Century (Krefft’s era). The collection dates on the remaining 53 range from 1904 to 1996. The Macleay Museum (MMUS) holds three H. bungaroides that all are believed to be from the late 19" Century. One is from “Mount Wilson” but the collector and date are unknown. It was possibly John Anderson or James Cox since both of these zoologists made collections for MMUS in the Mount Wilson area (Fletcher 1929; Stuart Norrington pers. comm.). The only information with the two other MMUS specimens is that they were collected on the “coast near Sydney”. Hoplocephalus bungaroides specimens are also held in all other Australian mainland capital cities. The South Australian Museum, Adelaide (SAMA), has 6 specimens recorded on its collection register, but one of these (R00463) is now missing. This misplaced specimen is recorded as collected on 2 June 1915 at La Perouse and donated to the SAMA by the AM. Other SAMA specimens were from Kuringai Chase, Sydney and Woronora River. The Kuringai Chase specimens are reported to us as having been collected by “W. Irvine” in 1967. We enquired with William (Bill) Irvine (a well-known collector who still lives in Sydney) for details about these but he explained that his field notebooks from 40 years ago had now been destroyed. The Queensland Museum, Brisbane (QM), has four specimens: one from Waterfall; one from Nowra that was held in captivity for a period of time (Queensland Reptile Park); one was captive-bred; and another was confiscated by Queensland Parks and Wildlife Service in 1989. In Canberra, three specimens are in the Australian National Wildlife Collection (ANWC): one from about 1963-1964; the other two from around 1978- 1980 (J. Wombey pers. comm.). Collection localities are not available for any of these. Also in Canberra, the Museum at the Australian National University (ANU) has two specimens: Waterfall and Tiajuara Falls (22 km from Nerriga), although these have no dates or registration numbers. In Darwin, the Northern Territory Museum (NTM) has four specimens all from the 1970s and collected at Heathcote, Jarra Fall (Nowra), and Woronora Dam. In Melbourne, Museum Victoria (NMV) has six specimens. Four of these were registered sometime between 1900 and 1945, but collection dates are not available. Localities are Helensburg, Long Bay, Middle Harbour, and Coast Range at Botany Bay. The Middle Harbour specimen, at least, possibly originated from, or was known to, Krefft because this collection locality was specifically referred to by him (Krefft 1869). Two other specimens in NMV collected in 1975 are from Yal Wal (Nowra) and Royal NP. The Western Australian Museum (WAM) has three specimens, all from Woronora Dam in the 1960s and 70s. European collections In Germany, there are four museums with records of 13 H. bungaroides specimens. The Museum fiir Naturkunde, Berlin (ZMB) has eight specimens. Two of these were purchased from “Salmin”, a dealer in Hamburg who traded with Krefft. They are undated, but it is known from the Annual Reports that Krefft sent Salmin reptiles in 1866 (see Appendix 1). The ZMB also has three specimens labeled “Krefft” specifically. Another two specimens are from 1867 and donated by Richard Schomburgk (1811-1891). Schomburgk was Director of the Botanical Garden in Adelaide from 1865-1891. None of these seven specimens have specific point localities, i.e. either “Australia”, “New South Wales” or “Sydney”. The eighth specimen in the ZMB was donated by the Berlin Zoo on 12 September 1913, and the original collector and collection place are unknown. In the Zoologisches Museum, University of Hamburg (ZMH), there are two specimens: one from Krefft; the other with no collector details. These specimens are recorded as from “Sydney” and “Australia” respectively. The single specimen in the Senckenberg Natural History Museum, Frankfurt (SMF), from “eastern Australia” was donated in 1911 by “O. Frank”. We have no details on “O. Frank” or any other information on where he found his specimen. The Zoologische Staatssammlung, Munich (ZSM) had two H. bungaroides from “New South Wales” registered in 1920 and 1928, but these were destroyed during World War Two (D. Fuchs pers. comm.). Seven specimens were found in the UK. Six are preserved in the Natural History Museum, London (BMNH), and one in the Oxford University Museum of Natural History (OUM). One BMNH specimen was presented by the ‘Earl of Derby’ in 1847 (see also Ginther 1858; Boulenger 1896). This was Edward Smith Stanley (1775-1851), the 13th Earl of Derby. Two specimens in the BMNH derive from 1855. One of these was donated by the Zoological Society of London (ZSL), but the collector of this specimen is unknown. It was possibly John Gould, since he collected many specimens in Australia and also worked for ZSL. The second 1855 specimen is from the “collection of Captain Stokes”. This was John Lort Stokes, who was on the Beagle surveying expedition to Australia from 1837-1843. There is also a specimen in the BMNH registered 1859 that was presented by Proc. Linn. Soc. N.S.W., 130, 2009 J.M.HARRIS AND R.L. GOLDINGAY “Dr G. Bennett”. This was George Bennett, who was an early Curator of the Australian Museum from 1835-1841, and a Trustee of the Museum from 1853- 74. The other two BMNH specimens were purchased from Krefft and were registered on 16 June 1863. The only locality data with these specimens are “New South Wales” or “Australia”. The single specimen in the OUM was collected at “Sydney” by Francis Pascoe (1813 - 1893). Pascoe sailed to Australia in the Buffalo, captained by John Hindmarsh (first Governor of South Australia). After Pascoe’s death his large collection of zoological specimens was presented to the OUM by his daughter in 1909 (M. Nowak-Kemp pers. comm.). In France, the Muséum National d’Histoire Naturelle, Paris (MNHP) has four specimens. One of these (no. 7679) is the type of Alecto variegata (a junior synoymn for H. bungaroides), with locality given as “Australia”, collector/donor as Pierre Francois Kéraudren. This specimen is referred to by Schlegel (1837), Duméril et al. (1854), and Guibé and Roux-Estéve (1972). There is also a specimen from Port Jackson donated by “Quoy and Gaimard” (i.e. Jean René Constant Quoy and Paul Gaimard), collected some time prior to 1829 when these French naturalists visited Australia. The actual location data provided to us are for Middle Head. This specimen is also mentioned by Schlegel (1837), Duméril et al. (1854) and Guibé and Roux-Estéve (1972). Another MHNP specimen was collected from “Australia” by the French naturalist/specimen dealer Jules Pierre Verreaux some time in the early 1840s (also in Dumeril et al. 1854). According to the MNHP donations book, it was received in December 1846. The fourth MNHP specimen is a skull registered as no. 1991-4163. This specimen has no date, collector or locality details, but it is a different specimen to the above three, and it is believed to be from the same era, i.e. 19" Century (I. Ineich pers. comm.). In Austria, three specimens are preserved in the Museum of Natural History, Vienna (NMW). These are NMW 27699:1-3 and are dated between 1863 and 1877. There are no collector or donor names recorded with any of these, and the original label for these specimens indicates “West Australien” (=Western Australia). Photographs of the specimens supplied to the authors confirmed that the identifications are correct. However, the locality data is certainly erroneous. Other Australian snake specimens in the NMW collection were purchased from the dealer “Gerrard”, and it is possible that specimens with confused localities were sold by him, including these three H. bungaroides specimens. In Denmark, three H. bungaroides are preserved Proc. Linn. Soc. N.S.W., 130, 2009 in the collection of the Zoological Museum, University of Copenhagen (ZMUC). Two of these are dated 1862 and from “Sydney”, but no collector details are recorded for either specimen. The third from “Australia” was donated to ZMUC by “Dr Giinther” in 1867. In the Netherlands, the National Museum of Natural History, Leiden (RHNH), has five H. bungaroides specimens. One of these from “Nouv. Hollande” (Australia) was donated to RMNH by John Gould. Another two specimens recorded as from “Nouv. Hollande” are dated 1849 and were donated by “Frank”. This was probably G.A. Frank, a natural history dealer based in Amsterdam. A specimen from “Botany Head”, dated 1862, was received as a gift from the AM. The fifth specimen was also from the AM, but this has no date and no locality. Naturhistorisches Museum, Basel, Switzerland (NMB), has one specimen of H. bungaroides from “Australia”. It was donated in 1882 by Dr. Fritz Miller and is registered as no. 2188. Miiller apparently contributed many purchased or traded herpetological Specimens to the NMB in the years between 1880 and 1890 (R. Winkler pers. comm.). Advice received was that in this period, Miller worked voluntarily for the NMB and cared for the reptile, amphibian and fish collections. At the Zoological Museum, University of Liege, Belgium (MZULG), there are five H. bungaroides mentioned in the museum register. All arrived between 1856 and 1875 from specialised natural history shops (C. Michel pers. comm.). Three of these do not have localities, but two indicate “Melbourne”. If the latter two are truly H. bungaroides, then the recorded localities are also incorrect. However, as with the specimens from the NMW further study of MZULG specimens are also required to ascertain whether this is the case. In Italy, the Museo Civico di Storia Naturale, Genoa (MSNG), has one H. bungaroides (8687). The specimen was acquired in 1879 from the Godeffroy Museum of Hamburg (Germany), a private institution founded in 1860 by Johann Cesar Godeffroy (1831- 1885). The MSNG acquired specimens from the Godeffroy Museum by means of nine catalogues edited from 1864 till 1884 that listed duplicates put up for sale (G. Doria pers. comm.). No locality data are available for the specimen held at MSNG. North American collections In the US, there are five museums that together hold nine H. bungaroides specimens. The Field Museum of Natural History, Chicago (FMNH), has two specimens both collected at “Waterfall” in MUSEUM HOLDINGS OF THE BROAD-HEADED SNAKE the 1950s. One was collected by William Hosmer, a well-known herpetologist who worked as a field collector for the FMNH for many years and sold his Australian collection to that museum. It is known that the other FMNH specimen was collected by B Kaspiew, although we have no further information about this person. The Museum of Comparative Zoology, Harvard University (MCZ), has three specimens: one from “New South Wales”, received from Krefft in 1876; one from “Australia”, received from ““W. Keferstein” and registered in 1865; and one from “Gelle, Mt. Wilson, Blue Mountains”, received from the AM in 1914 (Loveridge 1934). The National Museum of Natural History, Smithsonian Institution (USNM), has a specimen catalogued in about 1872 with no locality details or collector name. The third from Sydney dated 1911 was received from “Julius Hurter”, a Swiss-American naturalist and early Curator of the St. Louis Academy of Sciences. The single specimen in the San Diego Natural History Museum (SDNHM) was originally sent there by the AM on exchange to Van Wallach and Richard Etheridge (San Diego State University) for Wallach’s studies on the visceral anatomy of the Australian Elapidae (see also Wallach 1985, 1998). A copy of the “specimen invoice form” shown to the authors was dated 19 January 1982 and indicates that this H. bungaroides was a “no data specimen”. The single specimen in the University of Illinois Museum of Natural History (UIMNH) has no location recorded with the specimen and was apparently “purchased from the AM” but the date for this transaction is unknown. It was originally catalogued into the very old zoology collection (<1943) and the Curator at the UIMNH suggested that it was probably from the 1920s judging by its very low “Z” catalogue number (006) (C. Philips pers. comm.) DISCUSSION Of the 159 specimens, accurate locality records were available for only 98 (62 %). The AM contributed 77 while another 25 institutions contributed the remaining 82. Several of the latter (detailed in the notes above) are highly significant: two records for Middle Head (dated <1829; 1935), two for Botany Bay (dated 1862; <1935), one for Long Bay (dated <1935), one for La Perouse (dated 1915), and three for Ku-ring-gai Chase NP (dated 1967). Four of the AM specimens (dated 1904/5) were from the same location at Long Bay as that above and two specimens (undated; 1895) were from the same locations at La Perouse as that above. Significant specimen records from the AM include those from the western side of the Blue Mountains (Bathurst: dated 1979; Ilford: dated ca. 1962), and from Mudgee (<1964). Other significant records are those from within the vicinity of Shoalhaven Formation geological outcropping along the western and north-western rim of the Sydney Geological Basin, the presumed limits of the species’ distribution. Whilst perhaps the species is absent there today, it gives a clear indication that some of this otherwise presumed habitat was in fact occupied by H. bungaroides. With many of these historical records collectors probably gave locations that covered wider districts or the specimens were allocated names of the centres they were brought to from the field. This is likely to be the case for the western records from Bathurst, north of Bathurst and Mudgee. The specimen locality data were mapped and contrasted with the 67 records in the Atlas of NSW Wildlife (Fig. 1). Two specimens from the AM (dated 1969) and one from SAMA (dated 1973) had as the locality data a site close to the location of the AM itself. We believe the co-ordinates for these three relatively recent specimens to be incorrect, and so excluded them from the map. The distribution of the museum records shows some concordance with the Atlas data. Both databases show aggregations of records in the Katoomba (Blue Mountains), Waterfall-Heathcote and Nowra (Shoalhaven) areas. Surprisingly, 37% of the records in the museum database are from Royal NP (28) and the adjoining Heathcote NP (8) and Garrawarra SRA (1). One location in Royal NP covering an area with a radius of 2 km contributed 23 specimens with collection dates spanning 1951- 72. These observations identify and confirm the currently known ‘hotspots’ of the distribution. We can also contrast Figure 1 with the only map previously published based on Australian Museum holdings (Longmore 1986; 50 specimens). There are about 15 museum records since 1986 including several for the Blue Mountains area (including Wollemi NP). Including these on our finer detail map gives it greater completeness as it includes Atlas records and non- AM museum records. Hoplocephalus bungaroides is reported from only a small geographic area, as evidenced from the locality data available from museum specimens (Fig. 1). Krefft (1869) reported H. bungaroides from Port Jackson, Botany Bay, Middle Harbour, Lane Cove and Parramatta, although as pointed out by Cogger et al. (1993), there have not been records from these areas for quite some time. These data indicate that the only museum specimen from Port Jackson was collected prior to 1829 by Quoy and Gaimard (MNHP 7678). Proc. Linn. Soc. N.S.W., 130, 2009 J.M.HARRIS AND R.L. GOLDINGAY Cessnock e llford » A ++ + + A Bathurst se Colo Heights *A~ th quithgow Records A museum a ee Kilometres + Wildlife atlas Fa 0 10 20 30 40 Map produced by Greg Luker, SCU GIS Lab, 22/8/2006 Figure 1. Geographic distribution of Hoplocephalus bungaroides as indicated by museum records and records in the Atlas of NSW Wildlife. Proc. Linn. Soc. N.S.W., 130, 2009 7 MUSEUM HOLDINGS OF THE BROAD-HEADED SNAKE At Botany Head, a specimen was collected in 1862, and ended up in Leiden, Netherlands, sent there by the AM (i.e. Krefft). There is also an AM specimen from Botany dated 1909 and another in NMV undated, but registered some time between 1900 and 1935. Middle Harbour museum specimens are in the NMV and MNHBP. It is likely that the Botany Bay and Middle Harbour specimens were known to Krefft, because these localities were specifically referred to by him (Krefft 1869). Of the 159 H. bungaroides specimens located, none had locality details recorded as Lane Cove or Parramatta. Thus, Krefft knew of H. bungaroides records from these locations, but it is uncertain whether he collected specimens from there. Krefft did undertake snake collecting in many places in the vicinity of Sydney. Rose Bay, Randwick, Manly, Coogee and Middle Harbour were reportedly principal localities (see correspondence between Krefft and Giinther in the archives of the AM). The annual reports of the AM are unequivocal in reporting that H. bungaroides specimens were sent to the Civic Museum, Milan (Italy), in 1865; R.H. Beddome (India) in 1867; and Berlin Museum (Germany) in 1871 (Appendix 1). The Exchange Register also indicates that one H. bungaroides was sent to the Madras Museum (now Government Museum, Chennai). In relation to the first of these, we made enquires with the museum in Milan (MSNM; Appendix 2), but H. bungaroides could not be found on the shelves or in the collections register. However, we found an H. bungaroides in Genoa, Italy (MSNG; Table 1), but this is dated 1879, and it is unknown whether this snake arrived at MSNG via the AM. In relation to Beddome, it is known that he was a naturalist and a British military officer posted to India. His zoological collection together with that of his son-in-law (G. C. Leman) was sold in 1935, and much of this material is now in the National Museums of Scotland (NMS); National Museum of Wales (NMW); and the Natural History Museum, London (BMNH). However, only the latter institution has H. bungaroides represented, and these specimens are all dated prior to 1863. Hence, the fate of the AM’s 1867 specimen sent to India is also unknown. The AM 4. bungaroides sent to Berlin in 1871 are still preserved in the ZMB. This museum has three H. bungaroides from the AM (Krefft), and another five specimens that arrived via other avenues. Unfortunately we were unable to confirm the presence or absence of H. bungaroides at the Government Museum, Chennai, because no advice was received in reply to our correspondence. The annual reports of the AM were quite vague in terms of the reptiles sent to de Robbillard in Mauritius; dealers Puls, Salmin, Fri¢ and Damon; and museums in Hamburg, Leiden, Madras, Paris, Madrid, London and Harvard (Appendix 1). Of these, we managed to track down H. bungaroides specimens collected/ donated by Krefft in Hamburg (ZMH), Leiden (RMNH), London (BMNH) and Harvard (MCZ). We can also confirm that Salmin received some H. bungaroides specimens (presumably from Krefft) because two from him were located in Berlin (ZMB). We found no evidence that other high-profile dealers such as Fri¢€ (Reiling and Spunarova 2005) received H. bungaroides from Krefft or anyone else. This review demonstrates the value of museum specimens as a source of information on species’ distribution (see also Shaffer et al. 1998). It’s widely known that much Australian material has made its way to 19" Century collections overseas, but the details of such holdings are still not easily accessible and so our contribution at least makes such distributional information available for H. bungaroides. Collectively, the museum data show specific records for Sydney’s urban areas - Botany Head, La Perouse, Long Bay, Botany, Concord West, Randwick, Middle Harbour and Port Jackson. These localities represent part of this species’ historical geographic range that has now been eliminated (see also Swan et al. 2004; Shine et al. 1998). Increasing our understanding of the historic distribution of H. bungaroides is of considerable importance because continued habitat clearing and fragmentation may eliminate this species from an area and without an historic record may lead to disagreement about whether an area is actually suitable for this species. For example, Hoser (1995) categorically refutes that H. bungaroides occurred in Ku-ring-gai Chase NP but three H. bungaroides specimens in the SAMA have collection details dated 1967 for that locality and there is no reason to doubt their authenticity. Recent surveys there (1998/9) failed to detect H. bungaroides (Newell and Goldingay 2005), suggesting it may now be locally extinct. The museum specimen localities provide a focus for increasing our understanding of the geographic range of A. bungaroides. There are three broad areas with aggregations of records: Katoomba (Blue Mountains), Waterfall-Heathcote and Nowra (Shoalhaven area). These areas also show aggregations of records in the Atlas of New South Wales Wildlife (Fig. 1). These may represent areas of highly suitable habitat for H. bungaroides. However, there is likely to be collecting bias evident with these data. For example, a few areas near Waterfall contribute 37% of all specimen locations, though records span a 27- year period. Recent detailed surveys in Royal NP (i.e. Proc. Linn. Soc. N.S.W., 130, 2009 J.M.HARRIS AND R.L. GOLDINGAY Waterfall) indicate that H. bungaroides is uncommon there (Goldingay 1998; Goldingay and Newell 2000; Newell and Goldingay 2005; Goldingay and Newell unpubl. data). The failure to detect H. bungaroides in recent surveys of national parks surrounding the Hawkesbury River where there are few historic records (Newell and Goldingay 2005) suggests that the species’ distribution is much more patchy than what might be predicted based on the presence of apparently suitable sandstone habitats. Further surveys of suitable habitat in areas without records need to be conducted. Records in the north-west of the species’ range (Bathurst: dated 1979; Ilford: ca. 1962) also highlight areas where further surveys need to be conducted. These represent the most western records of the species and a population in this area may show some genetic divergence and be of considerable conservation significance. The identification of museum holdings of H. bungaroides may be useful for a range of future research studies. This includes morphological research and further descriptions of diet based on stomach content analysis (e.g. Shine 198la,b, 1983; Keogh 1999). Furthermore, these specimens may provide a source of tissue samples for genetic studies that could contribute to an understanding of whether H. bungaroides has lost genetic diversity over time or if unique genotypes have been lost (see also Keogh 1998; Slowinski and Keogh 2000). Our collation here provides a record that will facilitate the use of specimens in this way. ACKNOWLEDGEMENTS We are grateful to the many curators and collection managers that responded to our enquiries. For data and information on museum holdings of H. bungaroides we acknowledge Ross Sadlier (AM), Russell Graham (ANU), John Wombey (ANWC), Colin McCarthy (BMNH), Maureen Kearney (FMNH), Paul Daughty (SAM), Carolyn Secombe (SAMA), Jose Rosado (MCZ), Raffael Winkler (MHNB), Stuart Norrington (MMUS), Ivan Ineich (MNHP), Giuliano Doria (MSNG), Nicole Kearney (MV), Christian Michel (MZULG), Marie-Dominique Wandhammer (MZUS), Franz Tiedemann (NMW), Paul Horner (NTM), Malgosia Nowak-Kemp (OUM), Andrew Amey (QM), Koos van Egmond (RHNH), Bradford Hollingsworth (SDNHM), Jens Kopelke (SMF), Chris Phillips (UIMNH), Ken Tighe (USNM), Rainer Guenther (ZMB), Alexander Haas (ZMH), Mogens Andersen (ZMUC) and Dieter Fuchs (ZSM). We thank Leoné Lemmer (Australian Museum Library) and Glenn Shea (University of Sydney) for assistance with identifying relevant literature, and for advice on Australian material held in overseas institutions. Further help was received from Van Wallach and Richard Etheridge. Stephen Sleightholme provided contact details Proc. Linn. Soc. N.S.W., 130, 2009 for some European museum curators, and Margaret Pembroke (Southern Cross University Library) tracked down some old literature. The map of distribution records was produced by Greg Luker (Southern Cross University GIS Lab). We thank David Newell and Ross Wellington as well as several anonymous referees for comments on an earlier draft of the manuscript. REFERENCES Boulenger, G.A. (1896). Hoplocephalus. In “Catalogue of the snakes in the British Museum (Natural History). Volume III., containing the Colubridae (Opisthoglyphae and Proteroglyphae), Amblycephalidae, and Viperidae’. Pp: 348-350. Cogger, H.G. (2000). Reptiles and Amphibians of Australia. 6" edition. Reed New Holland: Sydney. Cogger, H.G., Cameron, E.E., Sadlier, R.A., and Eggler, P. (1993). The action plan for Australian reptiles. Australian Nature Conservation Agency, Canberra, Australian Capital Territory. Dumeril, A.M.C., Bibron, G. and Dumeril, A. (1854). Erpétologie Générale ou Histoire Naturelle Complete des Reptiles. Paris: Roret Vol. 7(2) pp. 781-1536 [p. 1254-1258, pl. 76 b. fig. 1]. Fletcher, J.J. (1929). The Society’s heritage from the Macleays. Proceedings of the Linnean Society of New South Wales 54, 185-272. [269] Goldingay, R.L. (1998). Between a rock and a hard place: conserving the broad-headed snake in Australia’s oldest National Park. Proceedings of the Linnean Society of New South Wales 120, 1-10. Goldingay, R.L. and Newell, D.A. (2000). Experimental rock outcrops reveal continuing habitat disturbance for an endangered Australian snake. Conservation Biology 14, 1908-1912. ~ Guibé, J. and Roux-Estéve, R. (1972). Les types de Schlegel (Ophidiens) présents dans les collections du Muséum National d’ Histoire Naturelle de Paris. Zoologische Mededeelingen 47: 129-134. Giinther, A. (1858). Catalogue of colubrine snakes in the collection of the British Museum. London: British Museum. Pp. 213-214. Hersey, F. (1980). Broad-headed snake Hoplocephalus bungaroides. Pp. 38-40, in Endangered Animals of New South Wales, edited by C. Haigh. National Parks and Wildlife Service, Sydney. Hoser, R. (1995). The Australian broad-headed snake (Hoplocephalus bungaroides). The Reptilian Magazine 3(10), 15-27. Keogh, J.S. (1998). Molecular phylogeny of elapid snakes and a consideration of their biogeographic history. Biological Journal of the Linnean Society 63, 177- 203. Keogh, J.S. (1999). Evolutionary implications of hemipenial morphology in the terrestrial Australian MUSEUM HOLDINGS OF THE BROAD-HEADED SNAKE elapid snakes. Zoological Journal of the Linnean Society 125, 239-278. Krefft, G. (1866). On snakes observed in the neighbourhood of Sydney. Transactions of the Philosophical Society of New South Wales 1862- 1865, 34-60. Krefft, G. (1869). Broad-headed snake. In ‘The snakes of Australia; an illustrated and descriptive catalogue of all the known species’. Government Printer: Sydney. pp. 56-57 and Plate VI, figs 6, 6a, 6b. Leviton, A.E., McDiarmid, R., Moody, S., Nickerson, M., Rosado, J., Sokal, O., and Voris, H. (1980). Museum acronyms — second edition. Herpetological Review 11, 93-102. Leviton, A.E., Gibbs, R.H. Jr., Heal, E., and Dawson, C.E. (1985). Standards in herpetology and ichthyology: Part I. Standard symbolic codes for institutional resource collections in herpetology and ichthyology. Copeia 1985(3), 802-832. Longmore, R. (1989). Hoplocephalus bungaroides. In ‘Snakes: atlas of elapid snakes of Australia.’ Australian Flora and Fauna Series No. 7. Australian Government Printing Service, Canberra. p. 64. Loveridge, A. (1934). Australian reptiles in the Museum of Comparative Zoology Cambridge Mass. Bulletin of the Museum of Comparative Zoology (Harvard College) 77(6): 243-383. [290] Newell, D.A. and Goldingay, R.L. (2005). Distribution and habitat assessment of the broad-headed snake (Hoplocephalus bungaroides). Australian Zoologist 33, 168-179. Reiling, H. and Spunarova, T. (2005). Vaclav Fri¢ (1839— 1916) and his influence on collecting natural history. Journal of the History of Collections 17, 23-43. Roselaar, C.S. (2003). An inventory of major European bird collections. Bulletin of the British Ornithologists’ Club 123A, 253-337. Schlegel, H. (1837). Essai sur la Physionomie des Serpens. Partie Générale et Partie Descriptive. La Haye: Kips and Stockum 2, 477-478. Shaffer, H.B., Fisher, R.N., and Davidson, C. (1998). The role of natural history collections in documenting species declines. Trends in Ecology and Evolution 13, 27-30. Shea, G.M. and Sadlier, R.A. (1999). A catalogue of the non-fossil amphibian and reptile type specimens in the collection of the Australian Museum: types currently, previously and purportedly present. Technical Reports of the Australian Museum 15, 1- Dil. Shine, R. (1981a). Venomous snakes in cold climates: ecology of the Australian genus Drysdalia (Serpentes: Elapidae). Copeia 1981, 14-25. Shine, R. (1981b). Ecology of the Australian elapid snakes of the genera Furina and Glyphodon. Journal of Herpetology 15, 219-224. Shine, R. (1983). Arboreality in snakes: ecology of the Australian elapid genus Hoplocephalus. Copeia 1983, 198-205. 10 Shine, R. and Fitzgerald, M. (1989). Conservation and reproduction of an endangered species: the broad-headed snake, Hoplocephalus bungaroides (Elapidae). Australian Zoologist 25, 65-67. Shine, R., Webb, J.K., Fitzgerald, M. and Sumner, J. (1998). The impact of bush-rock removal on an endangered-snake species, Hoplocephalus bungaroides (Serpentes: Elapidae). Wildlife Research 25, 285-295. Slowinski, J.B. and Keogh, J.S. (2000). Phylogenetic relationships of elapid snakes based on cytochrome b mtDNA sequences. Molecular Phylogenetics and Evolution 15, 157-164. Strahan, R. (1979). Rare and curious specimens: an illustrated history of the Australian Museum 1827- 1979. The Australian Museum Trust. Swan, G., Shea, G. and Sadlier, R. (2004). A field guide to reptiles of New South Wales. 2nd Edition. Reed New Holland. Wallach, V. 1985. A cladistic analysis of the terrestrial Australian Elapidae, pp. 223—253 in Biology of Australasian frogs and reptiles (G. Grigg, R. Shine, and H. Ehmann, eds.). Royal Zoological Society of New South Wales, Chipping Norton. Wallach, V. 1998. The lungs of snakes, pp. 93-295 in Biology of the Reptilia (C. Gans and A. S. Gaunt, eds.), Vol. 19 (Morphology G). Society for the Study of Amphibians and Reptiles, Athens. Webb, J.K. and Shine, R. (1997). Out on a limb: conservation implications of tree-hollow use by a threatened snake species (Hoplocephalus bungaroides: Serpentes, Elapidae). Biological Conservation 81, 21-33. Webb, J.K. and Shine, R. (1998a). Using thermal ecology to predict retreat-site selection by an endangered snake species (Hoplocephalus bungaroides: Serpentes, Elapidae). Biological Conservation 86, 233-42. Webb, J.K. and Shine, R. (1998b). Ecological characteristics of a threatened snake species, Hoplocephalus bungaroides (Serpentes, Elapidae). Animal Conservation 1, 185-93. Webb, J.K. and Shine, R. (2000). Paving the way for habitat restoration: can artificial rocks restore degraded habitats of endangered reptiles? Biological Conservation 92, 93-99. Webb, J.K., Brook, B.W. and Shine, R. (2002). Reptile collectors threaten Australia’s most endangered snake, the broad-headed snake Hoplocephalus bungaroides. Oryx 36, 170-181. Whitley, G.P. (1961). The life and work of Gerard Krefft. Proceedings of the Royal Zoological Society of New South Wales 1958-59, 21-34. Whitley, G.P. (1969). Gerard Krefft (1830-1881) and his bibliography. Proceedings of the Royal Zoological Society of New South Wales 1967-68, 38-42. Proc. Linn. Soc. N.S.W., 130, 2009 J.M.HARRIS AND R.L. GOLDINGAY yno juas sordeds jo sisi] ou ‘AmnqioyUR_ ‘SUIT Il WOT, POAlooel sem (Saploimsung T=) ..Snjoselipa srypydaosojdopy ayxeuUs Yoe[q poyesoisea V,, “SOSULYOXO JO SI] Be SUIdsoy UdEq jou A[O}e] pey yFory yeu odes jenuue sty} Ul poyeys SoojsnI] JO pleog oy, popraoid a19M sasueLYyOXe JO S}sI{ OU PUR ‘poATodel o19M SaplOADSUNG "FT ON “SniyLineyy “SNOT LO “Pse][IGOY op ‘A Il 0} JUSS se sander 67 JO WONeT[09 V ‘Ad|PeIg “H'H IW Wop poataoel sem (sjosalspA “YY Se) Saploapsung “FF SUIPN{OUL SoxeUS SUIAT] JO UONIeT[00 WV ‘(SNJDBALADA “FT Se) Saplospsung ‘FT 210 10 9UO SUIpNyoUT « Ul[iog Ul unasnyy [eAOY SY} FO 10jDeM ‘S19}9q JOSSOJOIg,, 0} JUSS a1OM SaydaI JO UOTIET]OD B Inq ‘{paATedeI BIOM Saplos‘Dsung FY ON ‘[ro[vop eB] TyNowAaM “UOTURG VOqQOY IPI 0} JUSS OOM SoYsy pue spiig ‘soyydor ‘sjeurumew Jo suowmoods OSI pure ‘[s9[vop e] onBe1g ‘QI “A I 0} JAS (Sastoj10} 7] snjd) sender 17 Swnesny] YsHIIg 0} JUSS soysy pue sojydor Jo susuriseds gz] ‘AVISIOATUL) pleAILY ‘ASO[OO7Z sATyeredui0D Jo winesny] 0} JUAS dJaM sopdar poyloadsun ¢Z JOAOMOY ‘PpoATodeI OJOM Saplo/nsuNng FY ON ‘JNO JUSS O19M SoyIda1 OU PUR ‘PpaATIdOI 19M SaploADBUNG “ET ON ‘quas satoads ay} Jo AyUSpI JO SJoquINU sy} JO POprAold si UOTJVoIpUT ON ‘WINESNy[ UI[JOg OU} 0} JUES SEA SOYs pue sopdar ‘s]eUUeUT JO UONIETIOD & JN {poATodeI BOM SAp1O/DSUNG FY ON ‘(eIpuy ‘TeuuayD=) seIpey “S}So10,J JO 1OJeAIOSUOD, [elouZO ‘ewoppeg “HY urleyde_ 0} ques ATpoysodar sem (snjosaisva snpoydasojdoy se) usuttoads duo {paAtade1 d19M Saplospsung “FT ON ‘(sopydor ¢Z) uel UL UMesnyy dIAIZ ‘(soysy. pue sopyder p¢) unesnyy sted “(soysy pue soyydoe1 Q¢) Sinquiey ‘ules “TO IP “(e[Hder suo) prupeyy ye Winesnyy [eAOY ‘(soda QZ) elueusey Jo AJa1IN0g [eAOY dy} 0} JUSM soTdoy UdAIs JOU sem soapdar asoy) JO AIMUAPr 9Y} JNG SUINOSNW SNOTLIVA O} JNO yds d1oM sopydoy ‘soroods J1oyjOUe 10 sapIOADSUN SEA SIU} IOYJOYM Poje}s JOU SLM I JNQ “URULIOH “AA IJ] & WOIF poAtooos sem snjpoydasojdoy VW ‘(WINIS[ag Ul Jo[eap &) s[nd “Of IA & 0} pue “ummasny] sueg oy} 0} JUOS oJOM ‘saplospsung “FOU yng ‘soynde1 JOJO “GOT Ul OS[Y “Uk]IP Ul UNESNY] OIAID 0} JUAS seM (SNJDSaIIDA “FT Se) SaplosDSung ‘FT 9UO yNQ ‘PaATode1 SJOM SaplosDBUNg “FT ON ‘popnyour jou Ajuoredde sem saplospsung ‘YH Yysnoujye ‘seipeyy pure ueploy “sinquieyy Ul swWMasnUT 0} JUES o19M so[Nndoy “suOSIeg “J QUI[OIeD ssi & Ag payeuop sem (sjHSaLIDA +7) BYVUS popeoy-peolg V “UeUaIO_ I] & Ag poyeuop sem (s7jpsal1DA snyoydesojdopy] se) axeUus popesy-peolg V “poysijqnd jou arom sjiejap oytoads yng wMasnyl URTTeNSNY 9} 10F (Soroads QT ‘sayeus ¢¢ SuIpNjour) sojydo1 pojoo][0o YJory s}1odoy [wnuuy WO.y po}e]]09 sa}0N ‘saplospsung snjpydasojdozy ayeus papvosy-pvo.A IY) 0} 9dUd1EJo1 UO sIseyduo ue gould sajou oy], ‘sosuBYyoXe ofIdo1 Jo s[ivJop YIM PLST-LORT WNesny] ULTL.Qsny sy) Jo sj1odey [enuuy oy) Wor Usye) Soj0u ArevUIUINS :] xIpueddy pL81 eL81 CL81 IL81 OL81 6981 8981 L981 9981 £981 y98l €981 C981 1981 TBI I] Proc. Linn. Soc. N.S.W., 130, 2009 MUSEUM HOLDINGS OF THE BROAD-HEADED SNAKE BIZAIOT JO AUISIOAIUL) “WOTJIET[OD [eo1d0[007 - USpegsolM “SUNTWIUUILS oYOIPyeYyOsuossIMInjeN] USpeqso}:l - WNOISNUIS]JG Sio}s[ey-pueljoy D> - JoJsuny “OpunyzINjeN] Inj wunasnyy] SoyosTpeyIsoM, = snyry “Winosny] SI10}sTyINye NY = ci SloMyosuneig “WInesnyy SOYOSLIO}STYINJeN] SOYIITIeLIS yievuuseg GN oi uapsalq “opunyxJSL], Inj winesnyy soyorperys e, onseig ‘winasny] [euoTjeN Ss . yes}N1g ‘opunsxzINjeN Inj UNasnyy soysip}eR}S 2 ouIg “WInesny] UPIARIOT @ i oYMsj1ey “OpunyzINyeNy Iny wMesnyy soyorperis ayqndoay yoozD 5 s Z\IOH ‘opunyinjeny Inf unasnyy soyorjerys e ASO[OOZ JO winasnj] eo] Vy JO AjIsIOATU) = - JOAOUURH “WunesnuUsapueT SayosisyoRsiapalN, os unasny] O11ej}UGE [eAOY WN - sInqua}[y ‘WNnuURLINneyy] Wnesnyy soyorppunyinjyeNy se unesny] VIQUIN[OD Ysig [eAo yy Z, “ SINQOD “UInesny-OpunyINje Ny e umnasnyy eyIoq] Vy [PAO - Sizdio7] wWnesnwepunyinjeN, e daqangd “[RaNUO] ‘AVISIOATUL) [[L DOV “uMesnyy yyedpoy 5 x }pe}sjopny “WMosnyAy SoySst10}sTyIN}eN] x AIO}SIF{ [eINyeN] JO WNasnyy e109 PAON DN . Zuley] “WINasNy] SOYOSTIO}SIYAN}e | 2, BIQUIN]OD YsHIg JO AjISIOATUL) “UANESNIA] OeIGQOLIAA\ URMOD ros - AjeI}S|slosuosI[Ioy] ‘sIngounT NAVN be onjeN{ JO winasny] ueIpeueD & Fs uuog “sIus0y winesnyy upeueD a - }pe}sioq|ey Wnuvoulsy] winesnyy = BIOS “AIO\sTIH [eINJeN JO wWinesnyy [euoeNY oO - punwyI0g ‘opunyinjeNy Iny winesnyl vlivsng ° - uazjoyuosuey ‘pjay[y winesnyy = S[OSSNIg “SsoUdIOS [Injen JO aNjYsuy wuNIsjog [eAOY Ay - SINGSPIequ[IA\ ‘Winosny\-einp 7 USANT POISIOATUL) IYoT[ouIey “oIs0]OO7Z unosny a }pejsueg “WinosnwIsapueT soyosIssoy] : AJISIOAIUL) JUSH “apuNns.1o1q 100A wunesny * IVWISID “INjeN] op sneyy * S[oSsnuig ‘a1d0]007 ap winesny/y - yoniquieT “winesnyj-rouruin qd = IVUINOL, “o]JOINJeNY OA10}STFY.P Sasnyy Auvultasy umnisjog - BINOQSE.NS “d1d0[00Z op dasny\[ - ULIQUIOG NVYSsInjeN] 1oS19q].1V.10/\, c ISNO[NOT, “S][o1NjeNY O10}STEY.P UNasnyy e ZUIT “WINSsNUIsapuLT SOYOSIYdIo119}S0.10qO x UoNoOY “o][o1njJeNY O10}STF] Pp wWinasny| 2 ZeIQ) “WINoUUeOL WINesnuUsopue Ty] - DOIN ‘a][OINJVNY SILO}STFY.P Uunasnyy zn SINQZ|eS “Injen Jop sney ‘ SoJURN] “S][OANJVNY SILO}SIF{,.P UNasnypy BLysny 2; SO][LsIVJ] “O][OINJeNY OAO}STFY,P wunosnyj| . RIL] PT Op oasnj . UOAT “O[[OANJLNY OITO}SIFY,P UUNasNyY - So[PINJLN SPIOUSTD op OUTZUSSIV Oasnyy - UNINY “o][oINJVNY IITOSTFY.p WNasnyy a O[[IT [ens] uoloepuny ee d[QOUSIDH ‘o][o1NjeNY 9.110}SIF] Pp Unasnyy vujuos1y - uoliq ‘ssousI9g seq s}[novy ‘onbisojo0ozZ o110je10ge 7] - Kauphs Jo AjisioAtup) “winesnyy ASojoo7Z |[oMsey] - AoueBN ‘Saousldg sad 9}[Novy BI Op d130]007 ap omO}eI10gGeT é PIURUUSeY] JO A}ISIOAIUL) ‘UOT|DaT[OD AdoO[o0Z douRAy 2 ABOTOUYSST, JO 9jINjSUT ouINogey] [eAOY ‘UOIaT[OD ABojooZ “ NYAIN] JO AISIOAU) “wuNesny] [eoIs0;ooZ ‘: SUINOGIaJ\ JO AIsIOATUL) “WINaSNy] SSOLy, ¢ uinasny] AVIsIoATUp) B[AYSeAAL * puvjsucon() Jo ApisioAtup) ‘uinesnyy As0;007 2 ArOjSIF{ [RANeNY JO UNasnyy YsTUULF p puejsuq MoN JO AjIsioAtUy) ‘uunasnyp Ad0;007 puryluny P peqoy ‘Alo[eH jy pue winosny] ueruewse], = ArO\STF [RANJeNY JO Unosny] ULIUO}S_ n uojsoouney ‘Alo[eH iy pue winesnyl VIIO}DIA UdONd = Ipuelfi, ‘puelfi, Jo winosny\ ES AjIsIOATUP) o1renboryy “winasny] Soouatog [BoIsOO1g vIUuO}Sy ul[e.ysny u uonnjysuy u uonnjyysUy ysonboi uo UvYy} 9.101 331dsop Ajdo. jou pip (-) & Y{IAA SUOTNINSUT SUIMISNU ALoY) UT Poy 910A SUdUIDOdS saplossung “FZ OU VY) pordos (,.) & VIA poyxAvUl SUOINIHSUT :930N7 SA[day =y ‘auIeU UONIHSUI Vey) pue A1QUNOD Aq ATTVOHequvydye uo) IsAy LIPEUSNW IIA post] o1v Solu” ‘soUdpuOdsaa.109 [fvula Ano 0} Ajdo. jou pip 10 saplosnsung snjvydasojdopyy ay[eus popeoy-pvo.1g 94} JO SUdWAIDIdS P[oY jOU Op JE) SvaSADAO puUL LITV.QSNY Ul SUOLNIWSU] :7 xIpuoddy N — J.M.HARRIS AND R.L. GOLDINGAY ee ne eh sooualos Jo AWopeoy uvIssny ‘UIMasny] [ed150[007 AUISIOATUL) 9321S MODSO/| “LUNESNA] [29130[007 MOODSOJ] “WINISHA] UIMIE 3121S RISSHY “WUINEsNy] [LOISO[OOZ ueLIEqIS A}ISIOATUP) AOWIeUY ‘ATOYSTY] [eINjeNY JO wnesny| vISsny AjISIOAIU) 1eATOG-soqeg “UINasny] [ed1s0[00Z jsoleyong “winesnyy A10}S1H [RANIeNY NIQES “wunesnyy [eyUeyNIg BIULULOY RIQUILOD “UIMasny] AIO}SIF] [eINIEN RIQUILOD JO AjISIOATUL) ‘OISOTOdoONUY 2 Od1SO[00Z nasnyy UOgsTT JO AjISIOATUA) ‘[eINIeNY VIIO}SIEY Op [eUOIORN] Nasnyy pllopeyy ‘[eyouny op jediorunyy nosnyy OVO JO AYISIOAIUA) ‘[RINJeN VIIOISIEY op nosnyy UOgsTT ‘eIZO[00Z ap oUED [esnj}.10g Moyer “ApISIOAUP) URIUO][aISeL “LUNOSN] [RdISO;OOZ saouaIag Jo AWopeoy Ysijodg MP[IOIN JO AjISIOATUL) “UMNasNyy AIO\STH] [RANIeNY MesIeA ‘ASO[OOZ JO aN NSU] pure winesny[ puvlod O]SE JO AjisioATU) ‘UNMasnyy ASo;OoZ udsIog JO AjISIOATUP) “UOT}Oa]]0D ASo[007 JOBULARIS “SUI[OPAV YSISO[OOZ wnasnyj] JosUeAL}S wleypuory ‘Asojooeyory pue Aro}sip{ [eINyeN JO Winosnyy AGMAON Ulpounq “winasnyl 03210 pueppny ‘wnesny] pueppony yomnyoysiyO “winasnyy Aingiayued puelesZ Mon Wepio}suly “UINesny] [2913 0]007 ION JO AjISIOATUP) ‘UOT}IET[OD [eoIso[007Z WepioyOY ‘WepsoyOY WinosnuInNjeN, JYOIMseey] Wines] YOSOJsTYINNyeNy SpuLloyjoN SINOGUIOXN] ‘a][oINJeN SI1O}ST_Y pp [eUONeN sosnyy s.moquiexn] RSIY “VIA\eT JO wnesny] ATO}sI_] [eINjEN RIAL] OAYO], “Winasny] USING [eUOeN| “UOT}Oa][0D AS0[007 AIO}SIF [eINjeN JO Uinosnyy] eyesCO AUISIOAIUL) NYOYOL, “A1O}STF] [INN FO wNasnyy umnesny] ArO}sIP jernjeny NAsNAYLIDY uvder He ERP RE AUSIOAU) Wel[se_ “UONd9T[0D [eo1so[00Z pIssOy ‘ALOJSIF{ [PINJeN| JO WNosny [LLOUIAOI OUI, “AIOjSIH{ [RANWRNY FO winesnyy so[den Jo AjIsIaAUp) ‘Oo1s 0,007 Oosny| OULIA[e IP RIISIOAIUL) “ODISO[OOZ Oesny/| BUSI “OD1SO[O0Z Oasny/| ULIN], ‘einen SZUIIOG Ip sjeuoIsay Oosny/ I[VINJON YZUSIOG Ip s[BUOISOY Oosny/y eZUatdes eT, sWIOY JO AyISIOATUL) “VISO[OOZ Ip Oosny/y RUSO[Og ‘CUBO]O IP RUISIOAIUL) ‘eISO[OOZ Ip Oasny/[ vAOpE_ JO AjISIOAIUL] “eISO[OOZ Ip ossny/| eUSO]Og ‘voneAlas vuney vy Jod oyeuoizeN] O1NI1SU] [op Oosny| BUOIOA “OTRINJLN] VIIOJS IP OSIATD oasny| AUDA “A[LINJLN VIO IP OSIATD Oosny[ ULIIP] ‘O[LINIVN VIIO}S Ip OOIATD Oesny| OISOLL], Ip BLINN] VIIOJS IP OIA Oasnyjp| BSIq JO APISIOAIUL) ‘OLIOPIMIAT, [op 9 oTRANRN VIIO}S Ip Oosnyy SOUdIO][Yy ‘WInesny] AIO}STF] [LINJeN SOUSIO].Y sWIOY ‘ABO[OOZ JO winesnyy STATO RIOUTLD “AIOISIF] [RANVeN] OJUATVS oy} JO WNasnyy STATO ArO}SIF{ [PINJVN JO WNasny] IIAID OIe}OIO}UOP] OAONUTESeD leg Jo AjIsIoAluy) “WInesny] [eoIsoO]0OZ leg ulfgnd eset[oD Aur Uljqng “winesnyy ArO}sT [eINeN Alea] puvyjory TeqyeA ‘AISIOAUp) eIypuy “Wnasnyl [291d0[007 ByNO[VO “winesny] UeIpuy TeuUUSYD “Winesny] JUSWUUIOAODH) UOI9I[OD eIpuy Jo AoAINS [eoIso;007 IY] MON ‘ArO}STH [eINIeN JO wuNnesnypy [eUoneN Aequiog “uoysa][0D Ajeto0g Arojsipy [eAnjeN Avquiog jsodepng ‘uinesnyy ArojsIP jenjen Ueesuny, vipuy Axesunyy suo JO AyIsIoATU) “WInasny/] [eOIsO[00Z 999015) UdSUIQNY, JVjISIOATUP] “wuNesnyA] ASO[OOZ UWINasny] SOYOSIUIZIPSULIEL], pun Soyssiso[007 AUISIOALUP) JOUINT-UNIW] “WNyWSUy SoyosIs0[007, PIPMSHIOID “UmMasnyj] UsyOsIso[007 SINQoploH JO AjIsIOATU) “WUINssny] [POIsO[OOZ ussUIOH “UINasnyj [eO1s0;007 ARISIOATU)-S}YOoIQ] V-UeYSIIYO “Wnesny] [2d1d0[00Z7 YOO}JSOY JO AjISIOATUL) “WOT}D9][OD [eoIs0;007, ponunuos Aueu1es5 13 Proc. Linn. Soc. N.S.W., 130, 2009 MUSEUM HOLDINGS OF THE BROAD-HEADED SNAKE Ca SS RG RS DS ete CD SC Cs * i De ue OE AUISIOAIUA) SUNOA WeYysiig “WInasny/A] SOUSIDS oFIT QOUSIMET “UINasnyAy AIOISIFY [BINJeN] AjISIOATUE), Sesuey eUvIpUy ‘osoT[OD weypieg “winasnyy a100;] Ydasor AIO\SIH [RANJeN] JO wunasnyy OYepy JUIN [eINeN JO Winasny] UO\sNo}{ BIUAOFIeS ‘A1Ojsip [eanjyeN JO wnasnyy AeTpeA e215) Aro\sTpY [eINeN JO winasny] eIs1090 Alo\sipy{ [eANJeN FO wNasny] epr10] 4 ByuepYy ‘AIO}SIF{ [eINjeN JO Winesnyy yUequis, AIO}SIF{ [RANLN FO wNesny seT[eq SoJVIGO}IOA JO Winasny] ATISIOATUP) [[aUI0D AlO\stP [RINBN JO winasnyy puejars[D BUI[OIRD YNOS “UINesny] UO\sapIeYyD AVISIOAIUL) 9381S UOJBUTYSEA “UUINEsN] IouUOD ~yY septeyo AlO\SIF [VANJRN JO WINasny] sIdauIeD uoNdaT[0D Asojojodiayy sous Jo Awiopeoy erusojeD SIOUN]]] ‘PIOF yoo ‘A1OjsrY [eANyeN Jo wunasnyy coding aInj[ND pue ArojsTFY [enjeN] JO wWinesnyy syINg neMey “winesnyy doystg vjosouUIPy FO AjIsIOATUP) ‘ArO}STHY [eINWeNY FO wnasny] [[9q winesnyy ATISIOAIUp) 93%19 Avag uUNSny UOdIT]OD Asopojodiay] AjISIOAIU 9}e1g PUOZIIY YOK MON ‘AIO}STH [RANWeN JO winosnyy URoTIowy AIOYsTH [RANJRNY JO wnasnyy Bureqeil Vy vIydjape]iy_ ‘Soousiog jeinjeN Jo Awoperoy BILIOULY JO $9}¥}S poy AjISIOAU) Usapleqgy ‘uInesny] ASO[007Z Spoo7T JO AjISIOAIUA) “UOTJD9][O_D [eo1s0[007 Alay[eD iy pue umnesnyy AqID 19}S9910/\\ wunasny] AS0[007 sepunqd Jo AjIsIOATU) espliquieD ‘Aso[0oZ Jo umnasnyy AjIsIoATUy) pURlol] WIOYION UI wuNasnyy 10}S|) xossq “Uinosny\] Uoprep ONES Jojoxq ‘Winasny] [RWOWel Weg py [eAoy SdIAIOS Winasnyy SUIPRoYy SOIAIOG Splosoy pure sunasnyy AID YNowsI0g yynouwrAyg ‘AlayjeH iV pue wunesnyy AyD ynowA,g YsInquipy JO AjIsIOAUP) ‘UOMIET[OD A10\sTH [eINIeNY Ysinquipg ‘pueyjoog Jo sumnesny] [euOHeN Jjipred ‘soyeAA JO unasnyy [euoeN, Jojsoyoury] JO AjISIOAIUL) “WINasnyy Jo]soyouRy/\ suinesny] [OOdISATT uopuo7T ‘Aja100g ueouurT spoe7 ‘winasnyy AID spose UOpuoT “asaT]OD ssury [epucy ‘winoesny] [epuoyy SITYSI9]S9OMO]H “UINesnyA] opisAUNOD s1Oo/] UYOL BS AP DS PS * mre a ED aD BY yA Be SP SD YoIMsdy] “umnasnyy yorMsdy uOpuUOT ‘suossding Jo dda][0_D [eAOY ou} Je UNosNyy URLIO}UNFy PpuRpoos “‘Mosse[H ‘Alol]eH jLy pue winesnyy uewojunyy UOpUOT “wnssnyl UeWTUIO HY Alol[eD ily pue winesny] projoiay a[ISLOMAN “WUNasny] Yooouryy JaJSOYOUI\ “AdIAIAS sunesnyy AjuNOD ostysdurey UOPUOT B8a]]OD AjisioAtuE) “ASO;OOZ Jo winasnyy jUeIH aysy1eg “ASo;ooZ Jo wuinasnyy aJoD [oIslIg ‘AlaljeD yy pue umesnyy AyD [oIsg uoJYSIIg “AIOSI [eINyeN Fo unosnyy YOog (Ja]soyouRy]) UOWOg “sumMesny] UoO}TOg SMOIPUY 1§ JO AISIOATUL) “LUMasny] MoISINOg [[9q wopsury paytuy WINYJOTOS “WnosnuiNjeN Spuo,-ep-xneyD eT ‘a]joinjeN S110]sI}{ Pp sesny/] quuRsney ‘s180[007 ap sasny| Ug[eH 3S “uinesnuinyen] YoInZ JO AjISISAIUL) “WINesny [29150007 SINOQILY “WINesnyy AIO}STH [eANyeN Wlog “winasnyy A1O}SIP] jenjen, PADUDD “o][OINJEN SI1O}STE].p wunosny/y Ld REYAL AWS B.10Q9}0H “UInasnyy A1OjSIE] [eANJeN] 3810qQ9}0H AjISIOAIU) puny “wuiMesny] [RoIs0;00Z A1OjSIH{ [ene JO winasny] YstpaMms Wapaas PUISIO}VMIIA\ JO AjISIOATUE) “UMESNy] ASO[00Z Yyosoquay[ars JO AjIsIoATUP) “WuMesny] [2d1d0[00Z7 RUO}JOIg “WINasny] [BeASURIT, UMO], odeD “wnesny] ULOLIZYy yNOS winesny] Woqezi[q 0d UloJUOJWISO] g “WINesny] [PUOHeN SINQZILULIO}O1 J “WNesny [PIVEN Ag]loquiry “winasnyy 1OSs1g 9; ueqing “umesnyy ueging UMO}SWeYeIDH “wuMosnyy] Aueqyy BILIFY YINOS euelignly ‘A1ojsip [eINjeN] JO Winesny] URIUSAOTS BIUAAOTS Aofepieg “winesny] dyssuesg EPIEAOTS PLIpey ‘So[einjeN SeloualD op [euOIoRN Oasnyy BuoIOIeg “BOISO[OOZ Ip Oasnj| winesny] A10}STF] [VINJEN ,.SPISO[S] SINT, osuving ‘evoosny] UdIZ}USIZ INjeN OxEpyesuesng ulvds penuguos z xipueddy Proc. Linn. Soc. N.S.W., 130, 2009 14 J.M.HARRIS AND R.L. GOLDINGAY SEO OE BX ES I Ee EE * * AIOMSIFT [LINJLN JO UMosNyA] UOSOIO Jo AYISIOATUL) WINasny] 2101S BYSeAQON JO AjISIOATUP) uljsny ‘sexe] JO AjISIOATUL) ‘SUONOOT[OD ArO}SIH [eANJeN] sexo, AIOJSIFT [L.INILN JO Urnosnyy BIeqIeg vIULS ALOJSIFT [LINILN JO UMosny] VUOYLPO [GON Weg AjIsIoATU) a[eA ‘AIOISTEY [RAMJLNY JO tunesnyy Apoqeog umMasnyl AjISIOATU) 21219 USA SOOUSIOS [LINN JO Winasny] 97k}S BUTTOARD YON uinesnyl 972} JOA MON AIO\SIH [BANILN JO UNosny] OOTXA|A] MON styduroyy JO AyisIoAtuy) ‘Asojooz Jo umosnyy] OOIXO] MON JO “AlUP) ‘ASO[OIg Wo}soMUINOG Fo winesnyj| wnasnyy 91]qnd SaxNeAyAl ponuUods voLeUry JO Sojev}S pou, 15 Proc. Linn. Soc. N.S.W., 130, 2009 MUSEUM HOLDINGS OF THE BROAD-HEADED SNAKE Appendix 3: Hoplocephalus bungaroides specimens reportedly held in Australian and overseas museums. Note: the authors have not personally confirmed the identification of any of these specimens by examination. Records are arranged alphabetically by museum abbreviation (see Table 1) then numerically by registration number. Abbrevia- tions: Coll. = Collected; Confis. = Confiscated; Don. = Donated; NP = National Park; NSW = New South Wales; QNPWS = Queensland National Parks and Wildlife Service; ZSL = Zoological Society of London. Collection Date Museum Rego. No. Locality details Other details = AM R 1440 - Registered 30/08/1893 - AM R 1603 La Perouse - = AM R 1722 La Perouse Registered 14/04/1895 6/04/1900 AM R 2696 Mount Wilson Registered 10/05/1977 11/10/1904 AM R 3646 Long Bay Registered 15/05/1977 12 Apr 1905 AM R 3675 Long Bay Registered 15/05/1977 28/04/1905 AM R 3678 Long Bay Registered 15/05/1977 26/11/1905 AM R 3847 Long Bay Registered 18/05/1977 26/11/1905 AM R 3848 Long bay Registered 18/05/1977 16 Dec 1909 AM R 4619 Botany Registered 22/05/1977 - AM R 11179 Randwick Registered /04/1934 1/11/1959 AM R 15676 Waterfall Registered 27/11/1959 Aug 1959 AM R 18939 Waterfall Registered 30/11/1962 - AM R 18940 Waterfall Registered 30/11/1962 Apr 1962 AM R 18941 Mount Keira Registered 30/11/1962 - AM R 18942 Waterfall Registered 30/11/1962 - AM R 18943 Waterfall Registered 30/11/1962 - AM R 18944 Waterfall Registered 30/11/1962 - AM R 18945 Waterfall Registered 30/11/1962 - AM R 18946 Waterfall Registered 30/11/1962 - AM R 18947 Waterfall Registered 30/11/1962 = AM R 21071 Mudgee Registered 6/02/1964 2 Mar 1964 AM R 21219 Concord West Registered 6/03/1964 Feb 1969 AM R 30345 Springwood Registered 1/03/1971 8/09/1973 AM R 40309 Darkes Forest Registered 9/10/1973 2 May 1970 AM R 47415 Waterfall Registered 25/06/1975 22/10/1967 AM R 70034 Woodford Registered 1/02/1978 1966 AM R 74276 Royal NP Registered 16/06/1978 1966 AM R 74277 Royal NP Registered 16/06/1978 1971 AM R 74278 Waterfall Registered 16/06/1978 1971 AM R 74279 Sydney Registered 16/06/1978 1969 AM R 74280 Nowra Registered 16/06/1978 1969 AM R 74281 Sydney Registered 16/06/1978 1970 AM R 74282 Appin Registered 16/06/1978 1972 AM R 74283 Waterfall Registered 16/06/1978 Apr 1972 AM R 74284 Woronora Dam Registered 16/06/1978 2 Oct 1972 AM R 74285 Nowra Registered 16/06/1978 2 Oct 1972 AM R 74286 Nowra Registered 16/06/1978 2 Oct 1972 AM R 74287 Nowra Registered 16/06/1978 16 Proc. Linn. Soc. N.S.W., 130, 2009 2 Oct 1972 2 Oct 1972 2 Oct 1972 1967 1968 1969 197] 1971 1969 Oct 1978 1969 5 Aug 1978 2 Sep 1951 Jun 1963 5 Sep 1980 1979 17 Oct 1986 9 Feb 1996 1/01/1996 Jan 1980 Aug 1992 Aug 1992 Feb 1998 ~1963-1964 ~1978-1980 ~1978-1980 ~1855 ~1847 Proc. Linn. Soc. N.S.W., 130, 2009 > = > < An a A A ah Ss SS S Ss S > < zZ > PP Ss Ss zZ J.M.HARRIS AND R.L. GOLDINGAY R 74288 R 74289 R 74290 R 74291 R 74292 R 74293 R 74294 R 74295 R 74296 R 76338 R 82584 R 84381 R 92955 R 103159 R 103162 R 103711 R 107684 R 107685 R 107716 R 107717 R 107718 R 107719 R 107720 R 118644 R 125335 R 125414 R 128548 R 131075 R 131143 R 131144 R 131145 R 144614 R 144720 R 144876 R 147417 R 147418 R 150348 R 151978 RO1868 R0O5040 RO5041 1855.8.25.?? 1847.7.29.40 Nowra Nowra Nowra Helensburgh Royal NP Royal NP Waterfall Waterfall Sydney Area Colo Waterfall Colo Heights Waterfall Heathcote Mount Macleod Morgan Woronora Dam Bundeena Bundeena Stanwell Park Woronora Dam Waterfall or Heathcote Waterfall or Heathcote ~ 15km NE Bathurst on Road to Sofala Sydney Evans Lookout, Blue Mountains Hazelbrook, Terrace Falls Reserve Kangaroo Valley Captivity Linden, Glossop Road, Blue Mountains Linden, Glossop Rd., Blue Mountains Sydney Wollemi NP Waterfall Tiajuara Falls Australia Australia; Presented: Earl of Derby Registered 16/06/1978 Registered 16/06/1978 Registered 16/06/1978 Registered 16/06/1978 Registered 16/06/1978 Registered 16/06/1978 Registered 16/06/1978 Registered 16/06/1978 Registered 16/06/1978 Registered 30/10/1978 Registered 31/05/1979 Registered 14/05/1980 Registered 28/10/1981 Registered 28/10/1981 Registered 25/12/1981 Registered 7/04/1983 Registered 7/04/1983 Registered 7/04/1983 Registered 7/04/1983 Registered 7/04/1983 Registered 7/04/1983 Registered 7/04/1983 Registered 30/05/1986 Registered 28/03/1988 Registered 18/04/1988 Registered 31/12/1987 Registered 17/05/1988 Registered 19/05/1988 Registered 19/05/1988 Registered 19/05/1988 Registered 10/05/1996 Registered 15/05/1996 Registered 10/05/1995 Registered 10/05/1995 Registered 24/02/1998 Coll. H. Cogger Coll. Greg Mengden Capt. Stokes Collection Macgillivray collection ld ~1863 ~1863 ~1855 ~1859 ~1953-1956 6 Oct 1951 1876 1865 1914 <1837 Dec 1846 Jun 1836 1856-1875 1856-1875 1856-1875 1856-1875 1856-1875 1893 1893 1882 2 Aug 1863 12 Feb 1869 1877 ~1970s 9 Mar 1975 25 Nov 1972 1 Aug 1978 <1909 6 Jan 1959 30 Apr 88 4 Sep 1989 Aug 1862 No date 18 MUSEUM HOLDINGS OF THE BROAD-HEADED SNAKE BMNH 1863.6.16.50 BMNH 1863.6.16.55 BMNH 1855.10.16.109 BMNH 1859.6.30.10 FMNH 75118 FMNH 97310 MCZ R2525 MCZ R3642 MCZ R10282 MMUS RO501a MMUS ROSO1b1 MMUS RO501b2 MNHP 1991-4163 MNHP 3301 MNHP 7678 MNHP 7679 MSNG 8687 NMV D 4270 NMV D 4704 NMV D 51865 NMV D 51866 NMV D 65041 NMV R 12709 MZULG D.R.1883 MZULG R.E. 2657a MZULG R.E. 2657b MZULG R.E. 2657¢ MZULG R.E. 4221 MZUS 626 MZUS 627 NMB 2188 NMW 27699:1 NMW 27699:2 NMW 27699:3 NTM R1212 NTM R958 NTM R1115 NTM R1217 OUM OUM 4641 QM J52877 QM 347924 QM J49761 QM J61008 RMNH RMNH 1141 RMNH RMNH 1142 NSW; Purchased from: G. Krefft Australia Australia Australia; Presented: Dr G. Bennett Waterfall Waterfall New South Wales; received Nov 1870 Australia; Gelle, Mt. Wilson, Blue Mountains Mount Wilson coast near Sydney coast near Sydney Australia Port Jackson Australia - - type of Alecto variegata Australia Long Bay Middle Harbour Yal Wal, Nowra Royal NP Helensburg Coast Range, Botany Bay Melbourne Melbourne Queensland Queensland Australia Original Label “ West Australien” Original Label “ West Australien” Original Label “ West Australien” Jarra Fall, Nowra Woronora Dam Heathcote Woronora Dam Sydney Waterfall Nowra Botany Head, Sydney Museum Sydney Proc. Linn. Soc. N.S.W., 130, 2009 Registered 16 Jun 1863 Krefft ZSL B Kaspiew W. Hosmer Krefft W. Keferstein AM J. Verreaux Quoy and Gaimard Keraudren Acquired 1879 Registered 1900-1935 Registered 1900-1935 Registered 1900-1935 Registered 1900-1935 Rolle Rolle Fritz Miiller Coll. Graeme Gow Coll. Graeme Gow Coll. Graeme Gow Coll. Graeme Gow Coll. F.P.Pascoe Found under rock Captive specimen Confis. by QNPWS Captive bred gift of AM J.M.HARRIS AND R.L. GOLDINGAY - RMNH RMNH 1335 “Nouv. Hollande” (Australia) Gould 1849 RMNH RMNH 1336a “Nouv. Hollande” (Australia) Frank 1849 RMNH RMNH 1336b “Nouv. Hollande” (Australia) Frank 2/06/1915 SAMA R00463 La Perouse Don. AM; now missing 1967 SAMA R12099 Kuringai Chase W. Irvine 1967 SAMA R12100 Kuringai Chase W. Irvine 1967 SAMA R12101 Kuringai Chase W. Irvine 2/07/1971 SAMA R13433 Woronora River H. Ehmann Sep-73 SAMA R14116 Sydney G.N. Coombe 1980s SDNHM 63864 - sent on exchange by AM 1911 SMF 20532 eastern Australia Don. O. Frank ~1920s UIMNH 95151 - purchased from AM <1872 USNM 8050 - Catalogued about 1872 1911 USNM 56166 Sydney Coll. Julius Hurter 8 Aug 1964 WAM R53761 Woronora Dam G.F. Gow 9 Mar 1975 WAM R53762 Woronora Dam G.F. Gow 9 Mar 1975 WAM R53763 Woronora Dam G.F. Gow 1860s-1870s ZMB 4443 Sydney dealer Salmin 1860s-1870s ZMB 4444 Sydney dealer Salmin 1860s-1870s ZMB 5208 NSW Krefft 13 Sep 1913 ZMB 63510 Donated by Berlin Zoo - 1860s-1870s ZMB 63755 Australia Krefft 1860s-1870s ZMB 63756 Australia Krefft 1867 ZMB 63757 Recorded incorrectly as Adelaide Schomburgk 1867 ZMB 63847 Sydney Schomburgk 1868 ZMH R08213 514 Australia - 1861 ZMH R08212 763 Sydney Krefft Sep 1862 ZMUC R65270 Sydney - Sep 1862 ZMUC R65271 Sydney - ; Aug 1867 ZMUC R65272 Australia Don. Giinther 1920 ZSM 387/1920 NSW Destroyed in WWII 1928 ZSM 36/1928 NSW Destroyed in WWII Proc. Linn. Soc. N.S.W., 130, 2009 i SNe pati?’ Bray naa | ayleot ony arcane! AS ee & wien ten “uy | Qh ¥% etree, AY, istity Sadia pies ae eae ior aT WA Pe entered Liege aenaniey OORT sutely tee MD stihl Ante A “wero ay aes May" iA ) “a Ae | yw £# Cd } ‘fe ¥ byt WE END edinfed WAEMD clartle® yamadl) wha Kyte eg ME MAL rackklit Maser #2) (OVW id lywiy Otte i TPT mt hereon qt rex ia Mia bots rt y iy y 4 +e ‘ we! UME Nea Gin { 4 ‘ Kr ry) ; ” iy : " im LP ‘ oo) oe a DAMN BMNY hy a bs (Ri Levaw Ap Pullen bolt sunt eee Ne Wetqesagid’D tian A 4 edit ingest — oe Sa Wogan 2 Yara, Vital. sate Viusteiens nvdue' Pade 114) grok 49° ome ef “tam Neen 6. a * dpe (ethan Abad ma ; hee, Hava, 4 Ps Kc goto AMR Acs ol amet “ Trew ie anes ie ‘* “ as) ae OSTA ve POPOL EM ae serrate town esr REEL md apis, aA» iy eee A hw merci a Whewke, Vi dion =a ‘Daksa Cay piv) : vy pet art Soil weer: tts OREO UW Had une. pani j a petew FEE. 64 - (20K or Ajnctdlia,” gentnd wade iT Yoo TiGlas orread 7 incest ei Atel tama ety Fl lect (yur Agente eet ai (nite mono Ear ATK hong Baw -wanby2 iveh yup thy Mickite ir ep Bbeh y C0 @R5 nl Wal Mi eandeA HOSE ; Gs oGiadtiaryd baad OFRES 1 asad) Melerahury wiles s rerea re) Com TRongalfeymamy Bay | aatme DRS ohsbA ce bietimooed babicsA brat] AY WaT yonbv2 TKK Re NS cilenmanA hye RISPOR Le BN youl? fo C1Sa0a. Re a2 Melginen rsrby? ar czasy 54 Ohgtereatcred gantye (flan 2 “Tbe herent an oA ovkeat 1H Awkics the wav “WRITE ) ee 1 Trashed taal Aly 2 i alti” eeorec NR? a yin dbnipane ds A Te dj } Pri ag ex: hese 4, tage iit yd) Ties eal, Sia at, Gtacene Chin’ ROSK bree nie aides “ eT) Oonetien ed RYT Thaetdigais “« Sell irae: Chive a RII Wenn thing Calk Uhisewse Chow CHAS Bott Cok F. PPakgue Ce eal Wy ve tl Pond yader mach? a JETER i: Cagle ‘precy. ” pa ee Denis se Newel hon Mie 1 Sagi ne alle a Reconstructing Palorchestes (Marsupialia: Palorchestidae) - from Giant Kangaroo to Marsupial ‘Tapir’ B.S. MACKNESS School of Environmental Science and Management, Southern Cross University, PO Box 157, Lismore, New South Wales 2480, Australia. Mackness, B.S. (2008). Reconstructing Palorchestes (Marsupialia: Palorchestidae) - from Giant Kangaroo to Marsupial ‘Tapir’. Proceedings of the Linnean Society of New South Wales 130, 21-36. Since their initial description in 1873, palorchestid marsupials have been reconstructed in a variety of ways ranging from giant kangaroos, long-necked llama like-forms, bizarre okapians to their present popular image as quadrupedal marsupial ‘tapirs’. These reconstructions have resulted from an improved understanding of the phylogenetic position of Palorchestes, more complete fossil material and even the interpolation of supposed Australian Aboriginal renderings of these animals in Arnhem Land rock art. An examination of the timing of these different ‘views’ of Palorchestes has revealed that historical and social factors have also influenced how this animal has been visualized. Manuscript received 22 June 2007, accepted for publication 19 March 2008. KEYWORDS: history, Palorchestes, palorchestid, visual representation INTRODUCTION Attempts by vertebrate palaeontologists to reconstruct fossil animals are almost as old as the science that has informed such endeavours. In nineteenth century Europe, the French anatomist, Baron Georges Cuvier, gained a public reputation of being able to complete a “restoration from a single fossil fragment of complete skeletons of creatures long since extinct’ (Owen 1894:398). It appears, however, that Cuvier had only a marginal interest in attempting such reconstructions, dismissing them as too speculative (Coleman 1964, Outram 1984). Indeed, Cuvier didn’t publish any full reconstructions of prehistoric animals due primarily to his concern that such drawings would impact on his reputation as a scientist (Rudwick 1992). Across the channel, the so-called “British Cuvier’, Sir Richard Owen, earned similar accolades for his ability to reconstruct extinct animals from the most meager of remains. In one instance, Owen was said to have deduced the general form of the giant extinct New Zealand bird Dinornis from just “a six inch splint of bone with broken extremities” (Desmond 1975:101). Not all such palaeontological endeavours were so compelling however. When Cuvier was shown a tooth of the ornithischian dinosaur [guanodon, he identified the fossil as the upper incisor of a rhinoceros and later dismissed the metacarpal bones of the same animal as a species of hippopotamus (Delair and Sarjeant 1975). Owen’s work on Jguanodon was equally flawed. After being called on to supervise the sculpting of a life-size statue of the dinosaur, for the 1851 Great Exhibition of London, Owen not only posed the bipedal Jguanodon on all fours, but also placed its characteristic thumb spike on its nose (Desmond 1975). Although Cuvier was able to acknowledge his errors in identification before Mantell (1825) formally described /Jguanodon, Owen was not so fortunate. His anatomical faux pas were, and remain, highly visible thanks to the continued presence of the giant Iguanodon statue on its artificial island at Sydneham in London (Desmond 1975). In fact, almost a century and a half after its unveiling, Owen is still belittled over the anatomical inaccuracies of this reconstruction (Rudwick 1992) even though Owen was neither the first to reconstruct /guanodon nor the first to incorporate such maccuracies. Around 1835, for example, Mantell first visualized [guanodon as a type of a hypertrophied iguana (Williams 1991). Three years later, two further /gwanodon reconstructions were published in popular books on geology. George Nibbs completed a reconstruction as the frontispiece of George Richardson’s 1838 book, ‘Sketches in Prose and Verse’ while John Martin composed a gothic RECONSTRUCTING P4ALORCHESTES scene featuring three /gwanodon battling each other for Mantell’s, 1838 “Wonders of Geology’ (Rudwick 1992). Although significantly different from Mantell’s original iguana-like reconstruction, both followed his lead by picturing /gwanodon as a sprawling reptile with its thumb spike on its nose. While a paucity of fossil material has historically often been given as the reason for such errors in early reconstructions — in [guanodon’s case nothing more than a “few teeth and isolated bones” (Rudwick 1992:222) — other factors have also been implicated. At the time of Jguanodon’s discovery, the very concept of ‘dinosaur’ had not been formulated and the notion of extinct giant land reptiles was still novel (Delair and Sarjeant 1975:14). Further, given that there was also no demonstrated stratigraphic evidence that the Iguanodon fossils were anything olderthan Quaternary, it is perhaps not surprising that they were, at first, considered to be those of extinct mammals (Delair and Sarjeant 1975). Eventually, the existence of such giant land reptiles came to be accepted by scientists and even enshrined in the appellation Megal/osaurus or “great lizard’ — the formal name for the first of these creatures to be described (Buckland, 1824). As these giants had no living counterparts, they were understood using modern lizards as analogues and hence reconstructed as quadrupeds (Williams 1991). The first bipedal dinosaurs were not to be discovered for almost another two decades and on a different continent (Leidy 1858). As for the misplaced thumb spike, Mantell had originally indicated that the bone may be a dermal horn or tubercle but was convinced by unnamed authorities that the bone was a lesser horn of a rhinoceros (Delair and Sarjeant 1975). Even when Jguanodon was shown to be a giant reptile, it made more sense to place this ‘horn’ on the nose rather than on the hand given that there were no examples of similar thumb spikes in extant lizards. Desmond (1979, 1982), however, posits a deeper, political and perhaps even personal motives for Owen’s Crystal Palace reconstruction of /guanodon and the establishment of the taxonomic rank of Dinosauria (Owen (1841[1842]). This was to directly challenge the doctrine of Lamarckian transmutation, being espoused by many continental scientists and in England by his béte noir, Robert Grant of University College, London. Instead of giving the Crystal Palace statue the typical sprawling posture of all previous reconstructions, Owen stood his /guanodon erect like a mammal (Desmond 1982). By reconstructing it with such a modern stance, Owen hoped to discredit the doctrine of transmutation showing that present- day lizards and snakes represented a descent rather than an ascent as the ladder-like progression of the Dn, Lamarckian scheme demanded. Rupke (1994:133), however, contends that the establishment of the Dinosauria was nothing more than “the product of contemporary advances in taxonomic practices”. In Australia, the fossils of extinct giant marsupials, not dinosaurs, were the first to be studied and later reconstructed — primarily by overseas experts (Rich et al. 1985, Vickers-Rich and Archbold 1991). Among the earliest was Palorchestes, described by Owen (1873:387) as “the largest form of kangaroo hitherto found”. Its reconstructed skull was illustrated by Owen (1876) and then again in his seminal two volume work on Australian fossil mammals. In that work, Owen (1877) also provided a reconstruction of the country’s largest marsupial Diprotodon. As its feet were unknown at the time, the wily professor disguised these missing elements by hiding them in long grass. The foot bones were eventually found and described, almost a quarter of a century later, by Stirling and Zietz (1900). Modern reconstructions of Diprotodon differ little from the initial attempt by Owen except, of course, for the addition of the absent feet (Berganini 1964, Ruhen 1976, Quirk and Archer 1983). Other diprotodontid reconstructions have not been so readily accepted. The lack of recognizable postcranials of Zygomaturus meant that Gerard Krefft’s illustration of the animal, reproduced in Whitley (1966), was regarded as “curious speculation” by Archer (1984:677) while Lord and Scott’s (1924) reconstruction of the same animal was characterized as a “murky misconception” by Murray (1978:77), in spite of it being based on relatively complete fossil material (Scott 1915). The diprotodontoid Palorchestes, whilst being one of the first marsupials to be reconstructed, has also had the most varied reconstructions, being variously envisioned as a giant kangaroo (Owen 1876, Fletcher 1945); a gracile llama-like form (Bartholomai 1978); a bizarre okapian (Ford 1982); an elephantine- trunked quadruped (Flannery and Archer 1985); to its most recent guise as a marsupial ‘tapir’ (Quirk and Archer 1983) or ground-sloth-like creature (Long et al. 2003). Changes to how an animal has been reconstructed over time have normally been explained by reference to an increase in the availability of fossil material — “scientists of later periods have the benefit of more (and often better) specimens . . . than were available to their predecessors”(Rudwick 1992:220). The fossils of Palorchestes, however, are regarded as uncommon (Mackness 1995:606) or rare elements of fossil assemblages (Murray 1991:1106, Black 1997a:183), perhaps representing a solitary habit (Flannery 1983, Proc. Linn. Soc. N.S.W., 130, 2009 B.S. MACKNESS Flannery and Archer 1985, Black and Mackness 1999). The hypothesis that the extraordinary divergence in how Palorchestes has been reconstructed is due solely to changes in the amount of fossil material available has never been tested. Nor does such a suggestion allow for the influence of other factors even though these have been shown to have directly affected the visualization of other animals (Desmond 1979, Bakker 1988, Gould 1991, van Reybrouck 1998). This paper therefore seeks to systematically examine the major reconstructions of the marsupial ‘tapir’ Palorchestes, executed over the past 130 years, against the corresponding taxonomic understanding and available fossil material of the time in order to test the notion that changes in reconstructions of a particular animal result solely from improved fossil material and phylogenetic understanding and are independent of all other factor/s. The role played by palaeontological reconstructions in science communications is also discussed. MATERIALS AND METHODS Published reconstructions of Palorchestes from scientific and popular texts were digitally scanned and their main features rendered into line drawings. The taxonomic history of Palorchestes was chronologically arranged using summaries provided by Mahoney and Ride (1975) and Rich (1991). Details of fossils elements described were likewise listed in order of their publication following Woods (1958) and Rich et al. (1991), including those misidentifications that were used in the description of anatomical features of Palorchestes. Both these factors were compared against the line drawings of Palorchestes in order to ascertain whether there was any correlation between them. The possible effects of broader social and historical issues on each reconstruction were also considered. RESULTS Owen (1873) erected the genus Palorchestes on the basis of the anterior portion of a cranium, which included the rostrum. The holotype, collected by Dr Ludwig Becker from an unspecified deposit in Victoria, was named P. azael Owen, 1873. This locality has since been interpreted by Mahoney and Ride (1975) as the River Tambo in Gippsland. Owen assumed the animal was some sort of giant kangaroo as its cheek-teeth had longitudinal links between and in front of the transverse lophs (Archer 1984). These Proc. Linn. Soc. N.S.W., 130, 2009 features were later shown to have independently evolved in both palorchestids and kangaroos (Woods 1958). Nevertheless, Owen was convinced at the time that the new animal was a macropodid, a view reflected in his choice of its generic name, a conjunction of two Greek words which literally translate as ‘ancient leaper’ (Owen 1874:797). Two years later, Owen (1876) assigned further elements to P azael including a left and right mandibular rami, sacrum, caudal vertebra, innominate bone, femur, tibia, caleaneum and metatarsals, even though there was no field association with the holotype (Woods 1958). This same paper also contained the first published attempt to reconstruct Palorchestes in the form of an outline of its skull (Owen 1876, plate 20). The drawing (Fig. la), incorporated a realistic rendering of the holotype with a significant amount of the skull being inferred from extant kangaroos. This included the posterior portion of the cranium and the dentary. Surprisingly, although two mandibular fragments were assigned to Palorchestes in the same paper, they were not figured as part of the reconstruction but were used to justify the shape of the jaw as being most similar to Macropus, based on the changes in the depth of the fossil rami, rather than other extinct kangaroos such as Sthenurus and Protemnodon (Owen 1876). By reconstructing Palorchestes as a macropodid, Owen effectively obfuscated those features that would eventually come to be recognized as unique to palorchestids, such as the reduction of the nasals. Owen (1880a) described another species, P crassus from fluviatile deposits near Gowrie, south- east Queensland, on the basis of the symphyseal portion of a mandible with an anomalous condition in the molars of the right ramus. Lydekker (1887), however, found the condition absent in the left ramus and therefore synonomized P. crassus with P. azael. Woods (1958:182), in supporting Lydekker’s (1887) synonymy, further noted that the distortion originally described by Owen (1880a) was actually “postmortem fracturing, expansion and cementation with matrix”. A palorchestid palate from the Wellington Caves, New South Wales, named P rephaim by Ramsay (1885), was subsequently listed by both De Vis (1895) and Woods (1958) as P. azael. Consequently, the second valid palorchestid species to be described was P. parvus De Vis, 1895 from south-east Queensland. This new taxon appeared in De Vis’s (1895) paper on fossil macropodid jaws leaving no doubt that he shared Owen’s opinion that palorchestids were kangaroos. A premolar from Beaumaris Victoria identified by Hall and Pritchard (1897) as Palorchestes was later shown to belong to the Diprotodontidae (Stirton 1957). MB RECONSTRUCTING PALORCHESTES Figure 1. Historical reconstructions of Palorchestes from: a. Owen (1876); b. Fletcher (1945); c. Mur- ray (1978); d. Bartholomai (1978); e. Ford (1982); f. Quirk and Archer (1983). 24 Proc. Linn. Soc. N.S.W., 130, 2009 B.S. MACKNESS In 1912, the Trustees of the Australian Museum attempted the first three-dimensional reconstruction of Palorchestes using measurements from Owen and those from the mounted skins of living kangaroos (Fletcher 1945). The resulting sculpture stood almost three metres in height, even when posed in a resting position. Its imposing stature, when compared to that of living kangaroos, was said to have garnered much attention. This reconstruction was on display in the Museum for thirty-three years (Fletcher 1945). During the post-wars years, the higher classification of some mammal groups, including palorchestids, was reviewed by several workers. Simpson (1945) placed Palorchestes within the subfamily Macropodinae, following Owen’s lead, but the following year, Raven and Gregory (1946) moved it to the subfamily Sthenurinae. When Tate (1948) revised the kangaroos, he erected a new subfamily, the Palorchestinae, for Palorchestes. This meant that when the Australian Museum undertook a second supposedly more realistic reconstruction, taking into account “additional and important fossil remains” and to adopt “less misleading” assessments of how the animal should be modeled, Palorchestes was still thought of as a giant kangaroo (Fletcher 1945:363). The resultant model (Fig. 1b), was around 25% smaller than the 1912 original and photographed as the frontispiece of the Australian Museum Magazine (Fletcher 1945). Claims that this new museum model was the most accurate possible were somewhat tarnished however by errors in Fletcher’s (1945) accompanying text. He stated, for example, that Palorchestes was “first described in 1877 by Professor Sir Richard Owen, M.D., from the forepart of a crantum and portions of the jaw-bone with teeth” (Fletcher 1945:362-363) not in 1873 and based solely on a partial crantum as accepted by most other workers (Mahoney and Ride 1975, Mackness 1995, Black 1997a). Further, he interpreted the generic name Palorchestes to mean “the ancient dancer” (Fletcher 1945:362), even though Owen (1874a:797) specifically detailed its etymology. The greatest inaccuracies in the model, however, were to be exposed some thirteen years later. These were so significant that an embarrassed Australian Museum was forced to make a hasty and unceremonial disposal of their prized reconstruction (Archer 1984) with rumours still persisting that it is actually buried somewhere under Centennial Park in Sydney (M. Archer pers. comm.). The catalyst for the Museum’s precipitous action was a revision of Palorchestes by Woods (1958) who proposed that palorchestids were actually closer to diprotodontids than macropodids. The dentary of all Proc. Linn. Soc. N.S.W., 130, 2009 kangaroos possess a large mandibular foramen and masseteric canal. Both of these features were absent or suppressed in Palorchestes (Archer 1984). This meant that all the kangaroo-based reconstructions were incorrect and that palorchestids were most probably quadrupedal like other diprotodontids. Further, postcranials that had been attributed to Palorchestes in the past (e.g. Owen 1876, Gregory 1902, Scott 1916, Fletcher 1945) were shown by Woods (1958) to belong to either extinct kangaroos or wombats. The first undisputed palorchestid postcranial material was a series of caudal vertebrae of P. azael described by Bartholomai (1962), not in 1975 as claimed by Murray (1978). Five years after their description, a third palorchestid species, P painei Woodburne, 1967, was named from the Miocene Alcoota fauna of central Australia. Significantly, it showed the same extensive modifications to the rostral area that had been observed in P. azae/ and P. parvus by Woods (1958). In that same year, Stirton (1967) also formally recognized the Palorchestinae, which included Ngapakaldia and Pitikantia, as a subfamily within the Diprotodontidae. Archer and Bartholomai (1978) later raised this to familial status — the Palorchestidae. Further palorchestid postcranials were discovered in the seventies from a cave in the Wee Jasper area of New South Wales (Flannery and Archer 1985). These included a humerus and hindfoot which was subsequently prepared by the Australian Museum (Wells 1978). A humerus of P. azael was also reported from Victoria Cave, Naracoorte, South Australia by Wells (1975, 1978) along with phalanges and strange laterally-compressed scimitar-like claws, which Tedford of the American Museum of Natural History opined as being reminiscent of the extinct chalicotheres of the American Miocene. This led Wells (1978:109) to posit a tentative reconstruction of Palorchestes as “a large, quadrupedal grazing animal with longish limbs and plantigrade feet”. In the same year that Wells made his textual reconstruction, two new visual attempts were also published (Bartholomai 1978, Murray 1978). Both took account of Woods’s (1958) new phylogenetic understanding of palorchestids rejecting the earlier macropodid-based reconstructions. Murray’s (1978) sketch of a generalised Palorchestes (Fig. 1c), published in the specialist archaeological journal ‘The Artefact’, was based on the smaller Plio-Pleistocene palorchestid P. parvus. The reconstruction was part of a broader attempt to provide images of late Pleistocene fossil marsupials and a monotreme. Murray’s (1978, Fig. 12) sketch only included the head and shoulder D5 RECONSTRUCTING PALORCHESTES region, but a partial view of the entire animal was provided as part of a gallery of reconstructions (Murray 1978, Fig. 17). Following Woods’s (1958) re-description of P. parvus, Murray (1978:88) posited that Palorchestes would have had a “mobile upper lip indicated by the prominent pre-maxillary flange in the skull of P. parvus’’. It appears that Murray (1978:88) was also familiar with Fletcher’s (1945) article on the second model made by the Australian Museum as he repeated its error of interpreting the generic name of Palorchestes to mean ‘graceful dancer’. By contrast, Oakden’s scrapper board drawing of Palorchestes (Fig. 1d), for Bartholomai’s (1978) paper, was based primarily on the Miocene species P. painei. The catalyst for this reconstruction was the description of the cranium of P painei by Woodburne (1967); the preparation of further cranial material of the same species collected from the Waite formation during the 1974 Ray E. Lemley expedition of the Queensland Museum; and similar but less complete material of P azae/l and P. parvus held in the Queensland Museum (Bartholomai 1978:145). The reduction of the nasals, the elongation of the anterior of the palate and the presence of very large infraorbital foramina observed in these specimens led Bartholomai (1978) to postulate that all known species of Palorchestes probably had an extensive rhinarium or a tapir-like proboscis. Further, Bartholomai (1978) interpreted the narrow, deeply channeled mandibular symphysis as indicative of Palorchestes having had a long, flexible tongue. There were differences between the two reconstructions of Palorchestes, however, that could not be explained simply by the fact that they were based on different species. While Murray (1978:88) characterized Palorchestes as a ‘lightly buil[t] diprotodontid’, Bartholomai (1978) reconstruction was even more gracile with the longer neck making the animal look very llama-like. The position of the nares also differed, with those of Murray (1978) placed more posterior and superior to those in Bartholomai (1978). The latter was in line with Bartholomai’s (1978:148) assertion that Palorchestes may have possessed an “extensive rhinartum with anterodorsally directed nostrils”. Bartholomai’s (1978) Palorchestes was the first to feature a tapir-like trunk and also featured conspicuous vibrissae on the snout. By 1980, confirmation that the Wee Jasper material was indeed palorchestid came when a partial skeleton in the collection of the National Museum of Victoria was also shown to be that of Palorchestes (Flannery and Archer 1985). Although the Museum skeleton had no locality data, its association with some undisputed palorchestid teeth made the 26 specimen very important. Several of the bones in the skeleton had previously been labeled incorrectly by Scott (1916) as a giant species of wombat or wombat- like animal. Subsequently, other bones from Foul Air Cave at Buchan in eastern Victoria were also recognized as palorchestid. Given that the humerus of the Wee Jasper specimen was much smaller than the Buchan material, it was assumed that the Wee Jasper fossils represented P parvus while the Buchan bones were those of the larger P. azae/ (Flannery and Archer 1985). The identification of this additional postcranial material enabled a full reconstruction of Palorchestes as a quadruped. In 1981, Stahel produced a stipple drawing of an entire animal for an article published in a University newsletter (Archer 1981). This illustration was used the following year as the basis of a reconstruction (Fig. le) by Ibraham for an article in the science magazine ‘Omega Science Digest’ titled ‘The strange creatures of ancient Australia’ (Ford 1982). What is significant about both drawings is that they embodied a rather “chimeraesque’ understanding of Palorchestes, demonstrating a concomitant “high coefficient of weirdity” (Archer 1984:670). The overall body outline was rather ‘okapian’ with the hind-quarters lower than the front and the neck long and giraffid-like. The ‘bizarre herbivorous animal’ was Said to be as “large as a horse . . . [with] a trunk- like structure on its face .. . kangaroo-like teeth . . . [a] long giraffe-like tongue and . . . phenomenally huge sharp claws” (Ford 1982:84-85). These sharp koala-like claws were even thought, for a brief time, to represent an adaptation to climbing in trees like modern-day sloths but the idea was rejected when the huge size of Palorchestes became apparent (Archer 1984:670). These speculative views of Palorchestes were informed by palaeontologist Mike Archer who, just one year later, was involved in the production of another reconstruction that directly challenged many of the assumptions inherent in the ‘okapian’ model (Archer 1984). The rethink of how Palorchestes should be reconstructed was prompted by several factors including the identification of additional fossil elements and the opportunity to further refine or challenge aspects of previous reconstructions. The neck length of the Stahel and Ibraham reconstructions, for example, was deemed too long after the discovery that palorchestid cervical vertebrae were not elongate like that of giraffids (Archer 1984:670). Likewise, the size of the trunk was also thought to be over-inflated and consequently reduced with the tail likewise being shortened. These changes were encapsulated in a new rendering of Palorchestes which Archer (1984:670) Proc. Linn. Soc. N.S.W., 130, 2009 B.S. MACKNESS judged to be the “best” to date, acknowledging however that his opinion was biased, given his involvement in its formulation. The reconstruction, executed by Schouten (Fig. 1f), appeared in a book on prehistoric animals published by the Australian Museum (Quirk and Archer 1983). Schouten presented a composite view of the head and front feet of P. azae/ along with a full-view of the animal ripping bark from a tree. Beneath this illustration, a further sketch was provided to demonstrate how Palorchestes may have used its tongue to strip vegetation off branches. The body shape of Schouten’s Palorchestes was much more diprotodontid-like and its size more like that of a bull. The reconstruction also highlighted Palorchestes’s massive forearms; its rapier-like claws and tapir-like trunk. The text accompanying the new reconstruction was titled “unique trunked giant” and contained the first explicit connection between Palorchestes and Aboriginal people. Flannery (1983:54), who penned the text, suggested that Palorchestes may have been the inspiration behind the legend of the bunyip and that newly arrived Aboriginals may have had second thoughts about settling after seeing one of these giant marsupials. Further, Flannery (1983:54) claimed that Aboriginal people and Palorchestes had “co-existed in Australia between about 40 000-20 000 years ago”. In 1984, three different reconstructions of Palorchestes were executed by Murray, but in very different contexts. The first was a drawing of a generalized palorchestid (Fig. 2a) as part of a family tree of diprotodontoids presented in a children’s book ‘Australia’s prehistoric animals’ (Murray 1984a). Both Palorchestes and the mid-Miocene Neapakaldia were shown on the same blue branch representing the Palorchestidae (Murray 1984a). In contrast to his 1978 reconstruction of Palorchestes (Fig. 1c), however, Murray’s new depiction had a much longer tapir-like trunk. This interpretation was” justified with the inclusion of a diagram showing the similarities between the skull and trunk of a tapir and that suggested for Palorchestes. Murray’s illustration differed from Schouten’s (Fig. 1f) in having a longer tail but smaller body. Murray was also the first to explicitly use the term “tapir-like marsupial” (Murray 1984a:20). Murray’s second reconstruction was specifically of P. azael (Fig. 2b) and was published in a book on Quaternary extinctions. As with Ford’s (1982) characterization, Palorchestes was once again presented as a composite animal only this time it was said to have “tapir, chalichothere, pantodont and sloth-like features” (Murray 1984b:608). The “large kangaroo-like tail” of P azael was highlighted, citing Proc. Linn. Soc. N.S.W., 130, 2009 Bartholomai (1962) and a personal communication from the same author, while Archer and Bartholomai (1978) were quoted as the source of P. azae/ being “equipped with huge, curved, laterally compressed claws” (Murray 1984b:608). The overall body size of Murray’s P. azael was much more massive than his more generalized drawing (Fig. 2a) and featured a long flexible tongue. Fossil remains of P. azael were regarded by Murray (1984b) as not especially common but widely distributed, with specimens of P azael from Pulbeena Swamp in Tasmania, (54 200+11 000 - 4 500 yr BP) listed as a recent occurrence of the taxon (Banks et al. 1976). Flannery’s (1983) suggestion that Palorchestes and Aboriginal people lived contemporaneously was seemingly validated in 1984 when a large Aboriginal painting (Fig. 2c) was tentatively identified asa possible representation of the extinct marsupial (Murray and Chaloupka 1984). The painting, discovered in Deaf Adder Gorge, Arnhem Land in 1976, was part of a tradition called the Large Naturalistic Animal Style (sensu Chaloupka 1993), which included depictions of animals now extinct from the Australian mainland such as thylacines and Tasmanian devils (Calaby and Lewis 1977, Lewis 1977, Clegg 1978). Some of the features used by Murray and Chaloupka (1984) to identify the painting as Palorchestes included: 1) the considerable attention given to the tongue including small lines which were said to perhaps represent items of food such as leaves or insects; 2) the detail given to the claws and the angled calcaneal joint; and 3) a lack of ears. Two anomalous breast-like projections under the body were explained as “stylised attempts to show a long shoulder mane or shaggy long hair” (Murray and Chaloupka 1984:114). A smaller animal besides the larger painting was said to represent a joey of the extinct marsupial. Murray and Chaloupka (1984) compared the Palorchestes painting with those of introduced animals such as those found previously in Cape York (Trezise 1971) as well as a variety of megafaunal species. In suggesting that the painting represented a Palorchestes, Murray and Chaloupka (1984:115) were extremely circumspect however, stating that “maybe it [the painting] represents Palorchestes” but “it must be made very clear that the connection at present is of the most tenuous kind”. They even suggested that “there may not be much gained by attempting to compare this unique and intriguing painting with perhaps the most poorly known species in the megafaunal assemblages” (Murray and Chaloupka 1984:112). In spite of such tentativeness, however, and in spite of a serious challenge to both the methodology and assumptions used (Lewis 2a RECONSTRUCTING PALORCHESTES Figure 2. Further reconstructions of Palorchestes from: a. Murray 1984a; b. Murray 1984b; c. Arnhem Land ‘Palorchestes’ from Murray and Chaloupka (1984); d. Murray and Chaloupka (1984); e. Rich et al. (1985), f. Long et al. (2003). 28 Proc. Linn. Soc. N.S.W., 130, 2009 B.S. MACKNESS 1986, Mackness, unpublsihed data), the painting has been promoted as a credible example of megafauna depiction by Aboriginal artists (Chaloupka 1993, Flood 1997). A third Palorchestes reconstruction (Fig. 2d) by Murray appeared in his joint paper with Chaloupka on rock art. What was unique about the reconstruction was that certain features were specifically added to match the supposed Aboriginal representation of Palorchestes. The most obvious of these was a mane of long hair protruding below the line of the abdomen to match the anomalous projections of the painting (Murray and Chaloupka 1984). This feature was not present in any of Murray’s previous 1984 reconstructions. The ears were also placed so that they didn’t project beyond the outline of the head to likewise match the painting. In Murray’s generalised Palorchestes (Fig. 1a), the line of the ears was clearly shown projecting above the head. In support of such modifications, the authors restated Clegg’s (1981:313) assertion that “if a well executed drawing of potentially great antiquity best matches a good restoration of an extinct species , then that may well have been the target species”. While invoking this “Occam’s Razor of rock art analysis” as justification for their identification of a Thylacoleo drawing, Murray and Chaloupka (1984:115) regarded the evidence for the Palorchestes drawing as being “less satisfactory” however. While the reconstructions of Palorchestes by both Schouten and Murray featured relatively short tapir-like trunks and diprotodontid-like bodies, Knight’s (Fig. 2e) composite illustration of P. azael and P. parvus, published in Rich et al. (1985), featured much longer trunks, body shapes more reminiscent of myrmecophagids and rhinoceros- like tails. Knight actually completed the illustration in 1982, around the same time that the Stahel and Ibraham reconstructions were published. The text accompanying the illustration, by Flannery and Archer (1985), provided the first detailed description of palorchestid postcranials along with a sketch of the articulated arm bones and a rear view of the humerus. Flannery and Archer (1985) argued that the front legs of palorchestids were unusual, relative to other marsupials, because of a greatly enlarged area for the attachment of the pectoralis muscle which formed a high, hooked process. The ulna of both species was said to be almost solid with only a tiny marrow cavity. The nature of the articulation between the lower and upper arm bones in P. azael was such that it appeared to indicate an immobile elbow with the front legs being permanently locked in a partly flexed Proc. Linn. Soc. N.S.W., 130, 2009 position, strengthening the already massive forearms. The smaller P. parvus, however, appeared to have a slightly more flexibility in this joint. The authors also drew attention to the highly mobile fingers that each bore a massive, sharp, laterally-compressed claw similar to that of a koala but far larger. Flannery and Archer (1985) interpreted these claws as suitable for ripping, tearing or climbing but not for digging. By comparison, the authors considered the hindlimb of Palorchestes to be far less robust. The fourth and fifth toes were equipped with the same kind of massive claws seen on the fingers of the hands but toes two and three were reduced in size and syndactylous, perhaps used for grooming. Flannery and Archer (1985) also suggested that Palorchestes may have possessed a clawless opposable great toe similar to that seen in possums. Overall they suggested that Palorchestes filled a niche similar to that of elephants or the extinct ground sloths of the Americas, using its narrow and elongate tongue in conjunction with its trunk, to strip leaves off trees and bushes. Once again, an explicit connection was made between Palorchestes and Aboriginal people with the suggestion that the “exceptionally powerful forearms, massive claws and bizarre head would surely have been enough to have inspired the legend of the bunyip — or at least a few nightmares among Australia’s first Aboriginal inhabitants” (Flannery and Archer 1985:236). The composition of the Palorchestidae was challenged by Murray the following year with the description of the lamb-sized palorchestid Propalorchestes from mid-Miocene deposits of Bullock Creek Local Fauna, Northern Territory and several Oligo-Miocene sites at Riversleigh, Queensland. Doubts had previously been cast by Archer and Bartholomai (1978) and Archer (1984) about the monophyly of the Palorchestidae. Aplin and Archer (1987), in their review of marsupial systematics, had placed palorchestids in their present position within the Vombatiformes. A further reconstruction of Palorchestes (Fig. 3) was executed by James Reece for a popular book on prehistoric life by Mackness (1987). Reece combined the reconstructions of Schouten and Knight to produce a hybrid image that adhered to a by now standard formula for illustrating Palorchestes with a diprotodontid body, sharp claws and tapir-like trunk. Such visual codification, called conventionalization by Rudwick (1992) enabled those viewing the animal to instantly recognize it as Palorchestes. In 1990, Murray described another species of Propalorchestes and concluded that members of that genus were the plesiomorphic sister-taxon of 2 RECONSTRUCTING PALORCHESTES i) / Lt if i} y Pigia } i | 34 ae ‘ ~ \ V4 Les | | Ga a | | se } al | | / Ws I | / Af i ‘i / aa / | | > - att 2 | | S veg f % we / | fy be y} { fi # s oe L , | Ne @ i. 17 = ety | ol & | on | N ~~ | | { i i | | =, he r /) | tI | | \ , | i | { | on | oe \ | | | \ Hi | aa , : | ; \ get Ky y aN \ ¥ “a ae i 3 ‘a 1 _ | ; teen =| == i / ro ) / | SA, N \ ule ; | | et ee aie (eee a eee i ra } | ves \ LG ' | = ga! ee oe | | sine Saar | < ‘SS WN IN | Figure 3. Reconstructions of Palorchestes from Mackness (1987). Palorchestes while Ngapakaldia and Pitikantia should be regarded as primitive members of the Diprotodontidae (Black 1997a). Five years later, a new species of palorchestid, Palorchestes selestiae, was described from the early Pliocene Bluff Downs Local Fauna on the basis on an isolated M! (Mackness, 1995) with a fifth species, P anulus described just two years later by Black (1997a) from the early-late Miocene Encore Local Fauna, Riversleigh, again on the basis of an isolated M'. The most recently described palorchestid, P. pickeringi, was recovered by Piper (1996) from Pliocene and early Pleistocene deposits of Victoria. It is represented by a significant 30 amount of fossil material and has also possibly been identified from Queensland (Hocknull et al. 2007). By the last decade of the twentieth century, the term “marsupial tapir” had become firmly entrenched as the popular name for palorchestids (Murray 1991) even though alternative descriptors such as “marsupial tree-fellers” had been proposed (Flannery 1994). The visual codification of Palorchestes reconstructions continued to be refined with the most recent reconstruction of P. azael (Fig. 2f), executed by Anne Musser and published in Long et al. (2003), perhaps being the apogee of how the animal should be depicted. Musser’s illustration did not show an Proc. Linn. Soc. N.S.W., 130, 2009 B.S. MACKNESS exaggeratedly long tongue or a trunk capable of being bent back on itself as illustrated by Schouten. The forearms were shown to be immobile following Flannery and Archer (1985), while the tail was more like that proposed by Murray (1984b). The explicit connection between Palorchestes and the eutherian Tapirus was also being down-played with extinct ground sloths now being the dominant analogue. This suggestion, first raised by Archer (1984) and Murray (1991), was visually encoded by the depiction of Palorchestes walking on the sides of its feet or on its knuckles. Long et al. (2003) also included an illustration of the skull of P painei showing its fragmentary nature, linking the real with the inferred in a similar manner to that first employed in Owen’s (1876) first reconstruction almost a hundred and thirty years previously. DISCUSSION The veracity of palaeontological reconstruction is underpinned by a specific methodology which is supposedly deployed with each attempt to illustrate a prehistoric creature. Murray (1978:77) characterizes “serious” reconstructions as only those that are based on “detailed anatomical build up of soft tissues”. This requirement challenges most reconstructions as very few conform to such rigor. Schouten visualized this same process using Diprotodon as an example in Quirk and Archer (1983). It should be noted, however, that it would have been singly impossible for any one artist to have the detailed anatomical knowledge required to undertake similar soft tissue build ups of all the other animals illustrated in that work. Rudwick (1992:221) provided yet another outline of the methodology suggesting it occurs in the following sequence:- 1) the selection of suitable fossil bones for assembly of a partial skeleton of a particular individual; 2) the reconstruction ofa complete skeleton representative of the species, based generally on the remains of many individuals; 3) reconstruction of a generalised complete individual body with inferences about the animal’s unpreserved muscles and other soft parts, based partly on anatomical analogy with related living forms; 4) and finally inferences about the animal’s dynamic mode of life and habits, based partly on functional analysis of its anatomy and on physiological analogy with related living forms. Rudwick (1992:221) posits that the outcome of such a sequence is “a cascade of representations that are progressively bolder—yet still well-founded— reconstructions of the unobservable prehuman past Proc. Linn. Soc. N.S.W., 130, 2009 . . . progressing from the observed to the inferred, from the specific and contingent to the general and idealized”. Changes in successive attempts to portray the same animal are simply “attributed to the discovery of more and better specimens that are relevant to that reconstruction” (Rudwick 1992:220). Latour (1986:17), however, from whom Rudwick (1992) derived the notion of “cascade”, uses the term in a much different sense. For Latour (1986:17), the sequence of reconstructing a prehistoric animal results in a “cascade of ever simplified inscriptions [visual representations] that allow harder facts to be produced”. Therefore, it is the selection of bones from a collection to be used in the description of a new species or the reconstruction of a complete skeleton from bones held in several museums over a wide geographic locality that allow scientists to make “bolder” reconstructions. When a pile of individual elements are coalesced into a published type description or into an articulated form, they became a single entity of “the type of . . .” or “the skeleton of . . .” with all its associated eidetic qualities. This process of accumulation and simplification is only useful however when there is confidence that the meaning of each coalescence has been stabilized (Pinch 1985). If it hasn’t, then all subsequent layers that are built upon it risk collapsing like a veritable ‘house of cards’ should the underlying assumptions prove to be unstable or incorrect. Such was the case with Owen’s (1876, 1877) reconstruction of Palorchestes as a macropodid. While in hindsight, it may seem that Owen made a grave error in his classification of the animal, Fyfe and Law (1988:1) caution that“... both the processes that lead to the creation of depictions, and the way in which they are subsequently used, have to be studied in their historical specificity”. With Palorchestes, several factors mitigated against Owen recognizing its ‘true’ taxonomic affinities. The partial cranium used as the holotype, for example, lacked those features, such as the reduction and retraction of the nasals, which would eventually be regarded as autapomorphies for palorchestines. Indeed, it wasn’t until almost a century later, after Woods (1958) had revised the genus and Woodburne (1967) had described P. painei, that suitable material became available to elucidate such characters. The presence of longitudinal links between and in front of the transverse lophs, while used by Owen (1874) to justify Palorchestes as a kangaroo, has since been shown to be convergent with at least two zygomaturine genera — Maokopia Flannery, 1992 and a new, as yet unnamed, Plio-Pleistocene species from eastern Australia (Black and Mackness 31 RECONSTRUCTING PALORCHESTES 1999. Mackness, unpublished data) possessing similar links. Flannery (1992:325) postulates that the development of “anteroposteriorly directed linking is an adaptation to a more abrasive diet’. Similarly, it wasn’t until the early part of the twentieth century that Abbie (1939) demonstrated that the presence of the masseteric fossa was a feature that united all macropodids. The fossil rami described by Owen (1876) lacked this relevant portion. Archer (1984) rightly concluded that the absence of such a feature in palorchestids didn’t preclude the possibility that they were still a plesiomorphic sister group of kangaroos. It wasn’t until Murray’s (1986, 1990) description of Propalorchestes and detailed biostratigraphical research into the Riversleigh Local Faunas by Black (1997b) that the taxonomy of palorchestids obtained some sort of stability with many authors (e.g. Archer and Bartholomai 1978, Archer 1984, Murray 1990, Mackness 1995) having previously cast doubt about the phylogenetic make-up of the group. The first major rethink about how Palorchestes should be reconstructed was not so much a result of additional and better fossil evidence becoming available as required by Rudwick’s (1992) sequence, but rather a reassessment of existing museum material and a consequential re-interpretation of its phylogenetic affinities (Woods 1958). This conforms to Latour’s (1986) notion of a ‘cascade’ with Wood’s (1958) coalescence providing a stable platform for harder facts to be produced. When new fossil material was collected by Woodburne (1967) and Bartholomai (1978), it was therefore added to the already stable platform of ‘palorchestids as diprotodontoids’. In particular, Bartholomai’s (1978) interpretation that the rostral area of palorchestids may have supported a tapir-like proboscis or extensive rhinarium provided the basis for the interpretation of palorchestids as marsupial ‘tapirs’. The lack of unequivocal palorchestid postcranials, however, apart from those described by Bartholomai (1962), meant that only the head region was known well enough for Bartholomai (1978) and Murray (1978) to attempt reconstructions — except for one very generalized body view (Murray’s 1978, Fig. 17). Even after palorchestid postcranials had been discovered and identified from caves in New South Wales, Victoria and South Australia in the 1970’s, their lack of publication meant they were effectively unavailable for use in reconstructions except for those few who had access to the relevant museum collections and the detailed anatomical knowledge to interpret what individual elements represented. To this day, the only description of these fossils is the popular account by Flannery and Archer (1985) in Rich et al. (1995). Sy The temporal lag of almost a decade between the discovery of these fossils and their incorporation into reconstructions also suggests that the relationship proposed by Rudwick (1992) may not be as straight forward as first thought. While some delay is to be expected, to allow for the preparation, study and publication of fossils, the postcranials of Palorchestes were never published in a peer-reviewed journal. Further, the most diverse representations of Palorchestes occurred between 1981 and 1983 (acknowledging that Knight’s reconstruction was completed in 1982) after the concept of palorchestids as diprotodontoids was stabilized by Woods (1958). The various attempts at reconstruction may, in part, be due to scientists using them as heuristic devices to test various anatomical options. The fact that palaeontologists Archer and Flannery, supervised all these divergent ‘views’ of Palorchestes perhaps bears this out. Van Reybrouck (1998), in his study of Neanderthal reconstructions, suggests that the intellectual zeitgeist may also affect how an organism is visualized. The publication of the various reconstructions of Palorchestes coincided with what Tedford (1991:76) characterizes as the “coming of age” of Australian vertebrate palaeontology with many academic institutions launching indigenous study programs at that time. Concomitantly, it was also a time when attempts were being made to raise the profile of the discipline in order to attract new students to the nascent palaeontological programs being offered at Universities (Vickers-Rich and Archbold 1991, Tedford 1991); to raise funds for research; and to mobilize and educate the general public (Quirk and Archer 1983, Rich et al. 1985, Mackness 1987). Perhaps not surprisingly, these popular texts featured creatures with superlative values such as the oldest, the largest or in Palorchestes’s case, the weirdest (Archer 1984:670). Part of the reason Palorchestes came to be reconstructed in so many guises was its ‘weirdness’ when compared to other marsupials. As well as being co-opted as a ‘poster child’ to demonstrate the uniqueness of Australia’s past, Palorchestes was included in some seminal debates about Aboriginality concerning the interrelated topics of land rights, environmental management and the extinction of the megafauna. Questions about the antiquity of Aboriginal settlement of the Australian continent had followed the widespread availability of radiocarbon dates (Mulvaney and Kamminga 1999) and in particular the dating of the Lake Mungo burials. A date of more than 40 000 years became a “slogan for indigenous people” (Gillespie 2004:1) and mobilized in legal arguments about rights to land (Yunupingu Proc. Linn. Soc. N.S.W., 130, 2009 B.S. MACKNESS 1997). The contemporaneity of Aboriginal people and extinct megafauna was another plank in this argument with suggestions that Palorchestes was the subject of the bunyip legend (Flannery 1983, Flannery and Archer 1985) and its supposed representations in rock art (Murray and Chaloupka 1984) adding credence to such claims. While Owen (1880b) was amongst the first to implicate Aboriginal people and the extirpation of the Australian megafauna, the early eighties saw the emergence of a full blown debate on the issue (Horton 1979, 1980; Martin and Klein 1984), a subject that continues to provoke controversy two decades later (Flannery 1994, Horton 2000, Roberts et al. 2001, Wroe et al. 2004). Consequently, while fossil discoveries and reinterpretations of phylogenetic relationships have played an important part in the varied reconstructions of Palorchestes, other broader factors have also been implicated. No matter what these influences are, however, they only become relevant if a particular reconstruction continues to be deployed. Corrigan (1988) contends that every time someone reproduces a reconstruction it becomes imbued with power. The context of reproduction can also play an important part in how a reconstruction is judged. Schouten’s 1983 reconstruction of Palorchestes azael has, until recently, held sway not only because it supposedly best matched the fossil evidence and was the most sophisticated rendition (Archer 1984) but also because it appeared in a book published under the imprimatur of the Australian Museum, one of the nations leading scientific institutions. The most recent reconstruction by Musser in Long et al. (2003) has yet to gain the same widespread exposure of Schouten’s effort but it obviously has only been in circulation for a short time. Its eventual hegemony also rests on the acceptance of the ground sloth analogy, explicit in the reconstruction rather than the existing and long-standing marsupial ‘tapir’ model. Latour (1987:258) suggests that *. . to determine the objectivity or subjectivity of a claim [like that made by a scientific illustration] . . . we look not for their intrinsic qualities but all the transformations they undergo later in the hands of others’. Consequently, future reconstructions of Palorchestes will not just be judged by whether or not they best fit the palaeontological information available but also whether they are reproduced in wide enough contexts to be accepted. ACKNOWLEDGMENTS The author wishes to thank Bill Boyd, Southern Cross University; Sue Hand, University of New South Wales and Proc. Linn. Soc. N.S.W., 130, 2009 Errol Vieth, Central Queensland University for providing helpful comments on the manuscript. Greg Luker, Southern Cross University, undertook the line reproduction of the different Palorchestes reconstructions. Katarzyna Piper provided valuable access to her work on palorchestid marsupials. Glenda Kemmis, Southern Cross University Library sourced the many reference articles used in the paper. The study was supported, in part, by a postgraduate grant to the author from the School of Environmental Science and Management, Southern Cross University. REFERENCES Abbie, A.A. (1939). A masticatory adaptation peculiar to some diprotodont marsupials. Proceedings of the Zoological Society of London Series B. 109, 261-279. Aplin, K. and Archer, M. (1987). Recent advances in marsupial systematics with a new syncretic classification. In ‘Possums and Opposums: Studies in evolution’ (Ed M. Archer) pp. xv-lxxii. (Surrey Beatty and Sons and the Royal Zoological Society of New South Wales: Sydney). Archer, M. (1981). Hunting ancestors in the possum dreamtime. University of New South Wales Quarterly 22, 6-8. Archer, M. (1984). The Australian marsupial radiation. In “Vertebrate zoogeography and evolution in Australasia’ (Eds M. Archer and G. Clayton) pp. 633- 808. (Hesperian Press: Carlisle). Archer, M. and Bartholomai, A. (1978). Tertiary mammals of Australia: a synoptic view. Alcheringa 2, 1-19. Archer, M., Hand, S. and Godthelp, H. (1995). ‘Riversleigh: the story of animals in Australia’s ancient inland rainforests’. (Reed Books: Chatswood). Bakker, R.T. (1988). “The dinosaur heresies’ (Penguin Books: London). ; Banks, M., Colhoun, E. and Van de Geer, G. (1976). Late Quaternary Palorchestes azael (Mammalia, Diprotodontidae) from north-west Tasmania. Alcheringa 1, 159-166. Bartholomai, A. (1962). A new species of Thylacoleo and notes on some caudal vertebrae of Palorchestes azael. Memoirs of the Queensland Museum 14, 33-40. Bartholomai, A. (1978). The rostrum in Palorchestes Owen (Marsupialia: Diprotodontidae). Results of the Ray E. Lemley expeditions. Part 3. Memoirs of the Queensland Museum 18, 145-149. Berganini, D. (1964). ‘The land and wildlife of Australia’. (Life Nature Library: New York). Black, K. (1997a). A new species of Palorchestidae (Marsupialia) from the late middle to early late Miocene Encore Local Fauna, Riversleigh, northwestern Queensland. Memoirs of the Queensland Museum 41, 181-185. Black, K. (1997b). Diversity and biostratigraphy of the Diprotodontoidea of Riversleigh, northwestern Queensland. Memoirs of the Queensland Museum 41, 187-192. 33 RECONSTRUCTING PALORCHESTES Black, K. and Mackness, B.S. (1999). Diversity and relationships of diprotodontid marsupials Jn Archer, M., Arena, R., Bassarova, M., Black, K., Brammall, J., Cook, B., Creaser, P., Crosby, K., Gillespie, A., Godthelp, H., Gott, M., Hand, S.J., Kear, B., Kirkman, A., Mackness, B., Muirhead, J., Musser, A., Myers, T., Pledge, N., Wang, Y. and Wroe, S. (Eds) The evolutionary history and diversity of Australian mammals. Australian Mammalogy 21, 1-45. Buckland, W. (1824). Notice on the Megalosaurus, or great fossil lizard of Stonesfield. Transactions of the Geological Society 1, 390-396. Calaby, J.H. and Lewis, D.J. (1977) The Tasmanian devil in Arnhem Land rock art. Mankind 11, 150-151. Chaloupka, C. (1993). “Journey in time’. (Reed Books: Sydney). Clegg, J. (1978). Pictures of striped animals: which ones are thylacines? Archaeology and Physical Anthropology in Oceania 13, 19-29. Clegg, J. (1981). “Notes towards Mathesis art’. (Clegg Calendars: Balmain). Coleman, W. (1964). “Georges Cuvier: zoologist. A study in the history of evolution’. (Harvard University Press: Cambridge, Massachusetts). Corrigan, P. (1988). ‘Innocent stupidities’: de-picturing (human) nature. On hopeful resistances and possible refusals: celebrating difference(s) —c again. In ‘Picturing power: visual depiction and social relations’ Sociological Review Monograph 35. (Eds G. Fyfe and J. Law) pp. 255-281. (Routledge: London). Delair, J.B. and Sarjeant, W.A.S. (1975). The earliest discoveries of dinosaurs. Jsis 66, 4-25. Desmond, A.J. (1975). ‘The hot-blooded dinosaurs. A revolution in palaeontology’. (Blond and Briggs: London). Desmond, A.J. (1979). Designing the dinosaur. Jsis 70, 224-234. Desmond, A.J. (1982). “Archetypes and ancestors. Palaeontology in Victorian London 1850-1875’. (Blond and Briggs: London). De Vis, C.W. (1895). A review of the fossil jaws of the Macropodidae in the Queensland Museum. Proceedings of the Linnean Society of New South Wales 10, 75-133. Flannery, T.F. (1983). A unique trunked giant Palorchestes. In ‘Prehistoric animals of Australia’ (Eds S. Quirk and M. Archer) pp. 54-55. (Australian Museum: Sydney). Flannery, T.F. (1992). New Pleistocene marsupials (Macropodidae, Diprotodontidae) from subalpine habitats in Irian Jaya, Indonesia. Alcheringa 16, 321- Boe Flannery, T. (1994). ‘The future eaters’. (Reed Books: Sydney). Flannery T.F. and Archer, M. (1985). Palorchestes Owen, 1874. Large and small palorchestids. In ‘Kadimakara. Extinct vertebrates of Australia’ (Eds P.V. Rich, G. van Tets and F. Knight) pp. 234-239. (Pioneer Design Studio: Lilydale). 34 Fletcher, H.O. (1945). Palorchestes - Australia’s extinct giant kangaroo. Australian Museum Magazine 8, 361-365. Flood, J.M. (1997). ‘Rock art of the Dreaming: Images of ancient Australia’ (Angus andRobertson: Sydney). Ford, J. (1982). The strange creatures of ancient Australia. Omega Science Digest July/August, 84-87. Fyfe, G. and Law, J. (1988). Introduction: on the invisibility of the visual. In “Picturing power: visual depiction and social relations’ Sociological Review Monograph 35. (Eds G. Fyfe and J. Law) pp. 1-14. (Routledge: London). Gillespie, R. (2004). First and last: dating people and extinct animals in Australia. Australian Aboriginal Studies 1, 97-101. Gould, S.J. (1991). “Bully for Brontosaurus’. (Penguin Books: London). Gregory, J.W. (1902). Some remains of an extinct kangaroo in the dune-rock of the Sorrento Peninsula, Victoria. Proceedings of the Royal Society of Victoria 14, 139-144. Hall, T.S. and Pritchard, G.B. (1897). Note on a tooth of Palorchestes from Beaumaris. Proceedings of the Royal Society of Victoria 10, 57-59. Hocknull, S.A., Zhao, J., Feng, Y. & Webb, G.E. (2007). Responses of Quaternary rainforest vertebrates to climate change. Earth and Planetary Letters 264, 317-331. Horton, D.R. (1979). The great megafaunal extinction debate: 1879-1979. Artefact 4, 11-25. Horton, D.R. (1980). A review of the extinction question: man, climate and megafauna. Archaeology and Physical Anthropology in Oceania 15, 86-97. Horton, D.R. (2000). “The pure state of nature: Sacred cows, destructive myths and the environment’. (Allen and Unwin: Sydney). Latour, B. (1986). Visualisation and cognition. Thinking with eyes and hands. Knowledge and society: Studies in the sociology of culture past and present 6, \-40. Latour, B. (1987). “Science in action. How to follow scientists and engineers through society’. (Harvard University Press: Cambridge, Massachusetts). Leidy, J. (1858). Remarks concerning Hadrosaurus. Proceeding of the National Academy of Sciences 10, 215-218. Lewis, D.J. (1977). More striped designs in Arnhem Land rock paintings. Archaeology and Physical Anthropology in Oceania 12, 98-111. Lewis, D.J. (1986). Comment ‘The Dreamtime animals’: a reply. Archaeology in Oceania 2, 140-145. Long, J., Archer, M., Flannery, T. and Hand, S. (2003). ‘Prehistoric mammals of Australia and New Guinea’. (University of New South Wales Press: Sydney). Lord, C. and Scott, H.H. (1924). “A synopsis of the vertebrate animals of Tasmania’. (Oldham, Beddone and Meredith: Hobart). Lydekker, R. (1887). Catalogue of the fossil Mammalia in the British Museum (Natural History). Part 5. (Taylor and Francis: London). Proc. Linn. Soc. N.S.W., 130, 2009 B.S. MACKNESS Mackness, B.S. (1987). ‘Prehistoric Australia. 4 000 million years of evolution in Australia’. (Golden Press: Sydney). Mackness, B.S. (1995). Palorchestes selestiae, a new species of palorchestid marsupial from the early Pliocene Bluff Downs Local Fauna, northeastern Queensland. Memoirs of the Queensland Museum 38, 603-609. Mahoney, J.A. and Ride, W.D.L. (1975) ‘Index to the genera and species of fossil Mammalia described from Australian and New Guinea between 1838 and 1968’. Western Australian Museum Special Publication 6, 1-250. Mantell, G.A. (1825). Notice on the iguanodon, a newly discovered fossil reptile from the sandstone of Tilgate Forest in Sussex. Philosophical Transactions of the Royal Society of London 115, 179-186. Martin, P.S. and Klein, R.G. (1984). ‘Quaternary extinctions. A prehistoric revolution’. (The University of Arizona Press: Tucson, Arizona). Mulvaney, J. an Kamminga, J. (1999). ‘Prehistory of Australia’. (Allen and Unwin: Sydney). Murray, P.F. (1978). Australian Megamammals: Restoration of some Late Pleistocene fossil marsupials and a monotreme. The Artefact 3, 77-99. Murray, P.F. (1984a). “Australia’s prehistoric animals’. (Methuen Australia: North Ryde). Murray, P. (1984b). Extinctions downunder: a bestiary of extinct Australian late Pleistocene monotremes and marsupials. In ‘Quaternary extinctions. A prehistoric revolution’ (Eds P.S. Martin and R.G. Klein) pp. 600-628. (The University of Arizona Press: Tucson, Arizona). Murray, P. (1986). Propalorchestes novaculacephalus gen. et sp. nov., a new palorchestid (Diprotodontoidea: Marsupialia) from the middle Miocene Camfield Beds, Northern Territory, Australia. The Beagle. Occasional Papers of the Northern Territory Museum of Arts and Sciences 3, 195-211. Murray, P. (1990). Primitive marsupial tapirs (Propalorchestes novaculacephalus Murray and P. ponticulus (Marsupialia: Palorchestidae) from the middle Miocene Camfield Beds, Northern Territory, Australia. The Beagle. Occasional Papers of the Northern Territory Museum of Arts and Sciences 7, 39-51. Murray, P.F. (1991). The Pleistocene megafauna of Australia. In “Vertebrate Palaeontology in Australasia (Eds P. Vickers-Rich, J.M. Monaghan, R.F. Baird and T.H. Rich) pp. 1071-1163. (Pioneer Design Studio: Melbourne). Murray, P.F. and Chaloupka, G. (1984). The Dreamtime animals: extinct megafauna in Arnhem Land rock art. Archaeology in Oceania 19, 105-116. Outram, D. (1984). ‘George Cuvier. Vocation, science and authority in post-revolutionary France’. (Manchester University Press: Manchester). Owen, R. (1841[1842]). Report on British fossil reptiles: Part Il. Report of the British Association for the Advancement for Science (Plymouth Meeting). Pp. 60-204. > Proc. Linn. Soc. N.S.W., 130, 2009 Owen, R. [In Anon.] (1873). On the fossil mammals of Australia. Family Macropodidae. Genera Macropus, Leptosiagon, Procoptodon and Palorchestes. — Part IX. Proceedings of the Royal Society of London 21, 386-387. Owen. R. (1874). On the fossil mammals of Australia, Part IX. Family Macropodidae: Genera Macropus, Leptosiagon, Procoptodon and Palorchestes. Philosophical Transactions of the Royal Society of London 164, 783-803. Owen, R. (1876). On the fossil mammals of Australia, Part X. Family Macropodidae: mandibular dentition and parts of the skeleton of Palorchestes; additional evidence of Macropus titan, Sthenurus and Procoptodon. Philosophical Transactions of the Royal Society of London 166, 197-226. Owen, R. (1877). “Researches on the fossil remains of the extinct mammals of Australia; with a notice of the extinct mammals of England’. (J. Erxleben: London). 2 volumes. Owen, R. (1880a). Description of a portion of mandible and teeth of a large extinct kangaroo (Palorchestes crassus Ow.) from ancient fluviatile drift, Queensland. Transactions of the Zoological Society of London 11, 7-10. Owen, R. (1880b). XXIII. Description of some remains of the gigantic land-lizard (Megalania prisca, Owen), from Australia. — Philosophical Transactions of the Royal Society of London 171, 1037-1050. Owen, R. (1894). “The Life of Richard Owen’. (John Murray: London). 2 volumes. Pinch, T. (1985). Towards an analysis of scientific observations: the externality of evidential significance of observational reports in physics. Social Studies of Science 15, 3-37. Piper, K.J. 2006. A new species of Palorchestidae (Marsupialia) from the Pliocene and early Pleistocene of Victoria. Alcheringa Special Issue 1, 281-294. Quirk, S. and Archer, M. (1983). ‘Prehistoric animals of Australia’. (Australian Museum: Sydney). Ramsay, E.P. [In Anon.] (1885). Untitled. The Sydney Morning Herald 14878, 10. Raven, H.C. and Gregory, W.K. (1946). Adaptive branching of the kangaroo family in relation to habitat. American Museum Novitates 1309, 1-14. Rich, P.V., van Tets, G.F. and Knight, F. (1985). ‘Kadimakara. Extinct vertebrates of Australia’. (Pioneer Design Studio: Melbourne). Rich, T. H. (1991). Appendix II: Literature References to the Fossil Terrestrial Mammals of Australia. In “Vertebrate palaeontology in Australasia’ (Eds P. Vickers-Rich, JM. Monaghan, R.F. Baird and T.H. Rich) pp. 1058-1069. (Pioneer Design Studio: Melbourne). Rich, T.H., Archer, M., Hand, S.J., Godthelp, H., Muirhead, J., Pledge, N.S., Flannery, T.F., Woodburne, M.O., Case, J.A., Tedford, R.H., Turmbull, W.D., Lundelius, E.L. Jr., Rich, L.S.V., Whitelaw, M. J., Kemp, A. and Rich, P.V. (1991). Appendix I. Australian Mesozoic and Tertiary terrestrial mammal localities. In “Vertebrate 35 RECONSTRUCTING PALORCHESTES palaeontology in Australasia’ (Eds P. Vickers-Rich, J.M. Monaghan, R.F. Baird and T.H. Rich) pp. 1005- 1058. (Pioneer Design Studio: Melbourne). Roberts, R.G., Flannery, T.F., Ayliffe, L.K., Yoshida, H., Olley, J.M., Prideaux, G.J., Laslett, G.M., Baynes, A., Smith, M.A., Jones, R. and Smith, B.L. 2001. New ages for the last Australian megafauna: continent- wide extinction about 46 000 years ago. Science 292, 1888-1892. Rudwick, M.J.S. (1992). ‘Scenes from deep time: early pictorial representation of the prehistoric world’. (The University of Chicago Press: Chicago). Ruhen, O. (1976). “The day of the Diprotodon’. (Dai Nippon: Hong Kong). Rupke, N.A. (1994). ‘Richard Owen. Victorian naturalist’. (Yale University Press: New Haven). Scott, H.H. (1915). A monograph of Nototherium tasmanicum (Genus - Owen” Sp. nov). Zasmanian Department of Mines Geological Survey Records 4, 1-46. Scott, H.H. (1916). A note of “Palorchestes” as a Tasmanian Pleistocene genus. Papers and Proceedings of the Royal Society of Tasmania 1924, 53-58. Simpson, G.G. (1945). The principles of classification and a classification of mammals. Bulletin of the American Museum of Natural History 85, 1-350. Stirling, E.C. and Zietz, A.H.C. (1900). Description of the bones of the manus and pes of Diprotodon australis, Owen. Memoirs of the Royal Society of South Australia 1, 1-40. Stirton, R.A. (1957). Tertiary marsupials from Victoria, Australia. Memoirs of the National Museum of Victoria 21, 121-134. Stirton, R.A. (1967). The Diprotodontidae from the Ngapakaldi Fauna, South Australia. Bureau of Mineral Resources, Geology and Geophysics Bulletin 85, 1-44. Tate, G.H.H. (1948). Studies on the anatomy and phylogeny of the Macropodidae (Marsupialia). Results of the Archbold expedition, No. 59. Bulletin of the American Museum of Natural History 91, 237- 85k Tedford, R.H. (1991). Vertebrate paleontology in Australia: The American contribution. In ‘Vertebrate palaeontology in Australasia’ (Eds P. Vickers-Rich, J.M. Monaghan, R.F. Baird and T.H. Rich) pp. 46-83. (Pioneer Design Studio: Melbourne). Trezise, P. (1971). ‘Rock art of south-east Cape York’. (Australian Institute of Aboriginal Studies: Canberra). Van Reybrouck, D. (1998). Imaging and imagining the Neanderthal: the role of technical drawings in archaeology. Antiquity 72, 56-64. Vickers-Rich, P.V. and Archbold, N.W. (1991). Squatters, priests and professors: A brief history of vertebrate palaeontology in Terra Australis. In ‘Vertebrate palaeontology in Australasia’ (Eds P. Vickers-Rich, J.M. Monaghan, R.F. Baird and T.H. Rich) pp. 1-44. (Pioneer Design Studio: Melbourne). 36 Wells, R.T. (1975). Reconstructing the past. Excavations in fossil caves. Australian Natural History 18, 208- Ae Wells, R.T. (1978). Fossil mammals in the reconstruction of Quaternary environments with examples from the Australian fauna. In ‘Biology and Quaternary environments’ (Eds D. Walker and J.C. Guppy) pp. 103-124. (Australian Academy of Sciences: Canberra). Whitley, G.P. (1966). Some early references to the extinct marsupial Zygomaturus. Australian Zoologist 13, 228-231. Williams, J. (1991). Dinosaurs - The first 150 years. A brief history of dinosaur discovery. Geoscientist 1, 19-22. Woodburne, M.O. (1967). The Alcoota fauna, central Australia. An integrated palaeontological and geological study. Bulletin of the Bureau of Mineral Resources 87, 1-187. Woods, J.T. (1958). The extinct marsupial genus Palorchestes. Memoirs of the Queensland Museum 13, 177-193. Wroe, S., Field, J., Fullagar, R. and Jermin, L.S. (2004). Megafaunal extinction in the late Quaternary and the global overkill hypothesis. A/cheringa 28, 291-331. Yunupingu, G. (ed.) (1997). ‘Our land is our life. Land rights — past, present and future’. (University of Queensland Press: St Lucia). Proc. Linn. Soc. N.S.W., 130, 2009 A Basal Actinopterygian Fish from the Middle Devonian Bunga Beds of New South Wales, Australia. BRIAN CHOO School of Earth and Marine Sciences, The Australian National University, Canberra, ACT 0200, and Museum Victoria, PO Box 666E, Melbourne, Victoria 3001 (mail correspondance to latter address) bchoo@museum.vic.gov.au Choo, B. (2009). A basal actinopterygian fish from the Middle Devonian Bunga Beds of New South Wales, Australia. Proceedings of the Linnean Society of New South Wales 130, 37-46. A partial articulated skeleton of a basal actinopterygian fish is described from the Middle Devonian Bunga Beds of New South Wales. The specimen represents a new species and is questionably assigned as a congener of Howqualepis rostridens from the Middle Devonian of central Victoria. This represents the first record of an articulated postreranium of a Devonian ray-finned fish from New South Wales. The pectoral fin of Howqualepis is also redescribed based on a re-examination of Victorian material. The fin is broader in shape and less extensively unsegmented than previously recognised. The close similarity of the new form with contemporaneous taxa from Victoria and the Aztec Siltstone of Antarctica adds to an already wide body of evidence supporting a regionally endemic freshwater vertebrate fauna in the Middle Devonian of Eastern Gondwana. Manuscript received 7 Feb 2008, accepted for publication 22 October 2008. KEYWORDS = Actinopterygians, Bunga Beds, Devonian, fish, Howqualepis, New South Wales INTRODUCTION In stark contrast to their modern abundance and diversity, actinopterygians are a sparse component of most Devonian vertebrate assemblages. Australia is notable in producing some of the finest fossils of Devonian actinopterygians, the best known of which are exceptionally preserved specimens from the Frasnian Gogo Formation of northern Western Australia. Included within the assemblage are Moythomasia durgaringa (Gardiner & Bartram 1977, Gardiner 1984), the currently preoccupied “Mimia” toombsi (ibid), Gogosardina coatesi (Choo et. al, in press) and at least two additional forms (Choo, in prep). Southeastern Australian fossil sites have also produced a substantial amount of early ray-finned fishes. The first record of Australian Devonian actinopterygians consisted of the isolated scales of Ligulalepis toombsi from the Lower Devonian Taemas-Wee Jasper Limestones of New South Wales (Schultze 1968). A subsequently discovered braincase and skull-roof was assigned to this genus (Basden et al. 2000, Basden & Young 2001). Long (1988) described Howqualepis rostridens based on numerous specimens from the Givetian Mt Howitt fauna of central Victoria (age revised in Young, 1999). Adding to this Eastern Australian record is an incomplete but articulated fossil that was recently discovered by Gavin Young from the Middle Devonian Bunga Beds, near the shoreline at Bunga Beach in south coastal New South Wales. This represents the first discovery of an articulated Devonian actinopterygian postcrantum from New South Wales. Subsequent repeated searches failed to recover additional material of this form (Gavin Young, pers. com.). GEOLOGICAL SETTING The Bunga Beds represent a thinly bedded sequence of carbonaceous shale and sandstone that comprises the lowest section of an extensively fossiliferous Devonian sequence (Fergusson et al. 1979, Young 2007). Young (2007, figs 1, 2) provides and up to date account of the lithology, fossil assemblage and possible age of the Bunga Beds. The age of the unit is poorly constrained and probably older than the Late Devonian age stated in recent A DEVONIAN ACTINOPTERYGIAN FISH literature (Cas et al. 2000, Giordano and Cas 2001, Rickard and Love 2000). The dark shales of the Bunga Beds are highly fossiliferous with abundant plant material and uncommon vertebrate remains (Young 2007, fig. 3), possibly representing a deepwater lacustrine depositionalenvironment. The fossilfish fauna includes ischnacanthid acanthodians (Burrow 1996), several taxa of chondrichthyans including Antarctilamna prisca (Young 1982), originally described from the Givetian Aztec Silstone of Antarctica, and a possible tetrapodomorph sarcopterygian (Young 2007, table 1). The fossil ichthyofauna of the Bunga Beds seems impoverished due to the apparent absence of placoderms and dipnoans that are abundant in other southeastern Australian sites of a similar age. MATERIALS AND METHODS The fossil was recovered as a natural mould set within a matrix of dark shale. After collection, the Specimen was split into part and counterpart and bone remnants removed. Bone margins were exposed with mechanical preparation and the impressions of the fish were examined using a latex rubber cast whitened with ammonium chloride. For comparison, fresh latex casts were made from the following specimens of Howqualepis rostridens in Museum Victoria (MV) = P.160745A, P.160782A, P.160788, P.160792B, P.160811, P.160822A, P.160851B, P.160857. Abbreviations for actinpterygian dermal bones and other structures used in the text and figures are as follows: an.f, anal fin; Br. 1, 1st branchiostegal ray; Br. 7, 7th branchiostegal ray; c.ful, caudal (basal) fulcra; Clav, clavicle; Clth, cleithrum; cw.lep, cutwater of short lepidotrichial segments; d.lep, probable dorsal lepidotrichia; f.ful, fringing fulcra; nm, notochordal mass of caudal fin; Op, operculum; Sop, suboperculum; pec.f, pectoral fin; pel.f, pelvic fin; pseg, segmented posterior lepidotrichia on pectoral fin; tfr, terminal fringe of fine branching segments on pectoral fin; vhl, ventral hypochordal lobe of caudal fin; useg. unsegmented proximal lepidotrichia on pectoral fin. SYSTEMATIC PALAEONTOLOGY CLASS OSTEICHTHYES Huxley, 1880 SUBCLASS ACTINOPTERYGII, Cope, 1887 Family Howqualepididae Long, Choo and Young, 2008 38 Diagnosis (revised) Basal actinopterygians with an open spiracular slit bordered by the intertemporal, dermosphenotic and supratemporal. Intertemporal is very small (less than 1/3 the size of parietals). Pineal foramen present on anterior half of the median frontal contact. Dermosphenotic is elongate and tripartite. Suboperculum has a prominent anterodorsal process. Body form is elongate and fusiform. Squamation macromeric; scales are rhombic with linear ganoine ornamentation. Fringing fulcra are spine-like terminal sections of the anterior fin rays, lacking median contact between the hemilepidotrichia. Longest anterior pectoral fin rays are proximally unsegmented for over 60% of their length. Median scute series on dorsal and ventral surface do not extend anteriorly to reach the head. Remarks Diagnosis slightly modified from Long et. al (2008) to incorporate the revised description of the pectoral fin and fringing fulcra of Howqualepis presented below. Genus ?Howqualepis Long, 1988 ? Howqualepis youngorum sp. nov. Etymology After Professor Gavin Young (ANU) who discovered the holotype specimen and Mr Ben Young for conducting both the preparatory work as well as the key photography of the specimen. Repository The type and only known specimen is lodged in the collections of the Department of Earth & Marine Sciences, Australian National University, Canberra, represented in the text by the prefix ANU V. Holotype. ANU V2929a, b, an incomplete, partially articulated fish preserved laterally in part and counterpart. Consists of an incomplete opercular- gular series, cleithrum, clavicle, scales and all fins except the dorsal fin (Figs. 1-4). Collected by Gavin Young (ANU) from the Bunga Beds at Bunga Beach, south of Bermagui, New South Wales. Diagnosis A Howqualepis with more than 54 primary lepidotrichia on the anal fin and porous ornamentation on the cleithrum and clavicle. Proc. Linn. Soc. N.S.W., 130, 2009 B. CHOO Remarks Tentatively assigned to the genus Howqualepis. The extensive unsegmented pectoral lepidotrichia of ?Howqualepis youngorum sp.nov separates this taxon from all other Devonian actinopterygians except Howqualepis rostridens Long, 1988, Donnrosenia schaefferi Long, Choo and Young, 2008, and Tegeolepis clarki Newberry, 1888. ?H. youngorum differs from Donnrosenia in that the unsegmented fin-rays account for more than 75% of the total length of the pectoral fin. ?H. youngorum differs from Tegeolepis in possessing macromeric squamation, long-based pelvic fins and a segmented terminal fringe on the pectoral fin. Separable from H. rostridens in having porous (as opposed to entirely linear) ornament on the pectoral girdle and in having a larger anal fin (54+ vs 45 primary lepidotrichia). DESCRIPTION Overall body form ANU V2929 is preserved in lateral aspect (Fig. 1). The anterior part of the specimen terminates at an oblique breakage margin, with elements of the opercular-gular series and pectoral girdle preserved along with the pectoral fin (Fig. 2). 2.5 cm behind this is an incomplete pelvic fin with patches of squamation present above and to the rear of the fin (Fig. 3). The largest preserved segment comprises the rear section of the fish, including well preserved anal and caudal fins along with extensive squamation (Fig. 4). The preserved sections suggest a highly elongate, fusiform body form similar to that of Howqualepis rostridens (Long 1988) and quite unlike the more compact and robust form of “Mimia” or Moythomasia (Jessen 1968, Gardiner 1984). As preserved, the fossil measures slightly less than 12 cm from the anterior preserved edge of the clavicle to the posteriormost caudal scales. Assuming that the missing portions of the fish were of similarly proportions to that of Howqualepis rostridens, the complete fish would have measured about 14 cm from snout to caudal peduncle. Opercular-gular series A section of the dermal operculo-gular series of ANU V2929 is preserved in articulation and comprises Figure 1. ?Howqualepis youngorum sp. nov. a. photograph and b. line drawing of holotype (ANU V2929A) showing the entire preserved fossil in lateral view. The specimen is a latex cast whitened with ammonium chloride. Proc. Linn. Soc. N.S.W., 130, 2009 39 A DEVONIAN ACTINOPTERYGIAN FISH Figure 2. ?Howqualepis youngorum sp. nov. a. photograph and b. line drawing of the pectoral girdle and opercular-gular series of the holotype counterpart (ANU V2929B), c. photograph and d. line drawing of pectoral girdle, opercular-gular series and pectoral fin of the holotype (ANU V2929A). a posteroventral fragment of the operculum, a partial suboperculum, and at least seven branchiostegal rays (Fig.2). The anterior portions of most of these elements are missing, the preserved sections terminating at a margin of clean breakage, suggesting that a substantial portion of the fossil, possibly including the skull, was lost prior to collection due to weathering. The posterodorsal-most bone in the series is tentatively identified as the posterovental fragment of an operculum. It is an oblong bone bone, missing the dorsal and anterior margins. The bone surface is ornamented with short, posterolaterally directed linear ridges. The suboperculum is rectangular with a convex posterior margin. Ornament consists of short linear ridges that extend to near the posterior bone margin. At least seven branchiostegal rays are visible on ANU V2929b (Fig. 2). The first branchiostegal ray, whose dorsal margin is overlapped by the suboperculum, is more than twice as thick dorsoventrally as the other bones in the series. The 2nd ray is poorly preserved while the 3rd is narrower than the following two rays. Rays 6 and 7 are very narrow. Ornament on all bones in this series consists 40 of short rostrocaudally directed ridges with little evidence of the tubercular ornament present on the laterally facing branchiostegals of Howqualepis rostridens (Long 1988). Pectoral girdle A partial cleithrum and clavicle (Fig. 2) have a similar overall shape to those of most early actinopterygians. The cleithrum consists of an expanded ventral region with a slender vertically directed blade although the dorsal portion of this structure is missing. The bone is convex postiorly with a moderately deep embayment on the posterior margin for the insertion of the pectoral fin, similar to that of H. rostridens (Long 1988. Fig.27). The clavicle is triangular and overlaps the cleithrum posteriorly and is itself dorsally overlapped by the branchiostegal rays. Preserved sections of ornament on both the cleithrum and clavicle consists of limited areas of short ridges, particularly around the posterior margin of the clavicle and the vertical blade of the cleithrum, that are largely replaced by rostrocaudally oriented Proc. Linn. Soc. N.S.W., 130, 2009 B. CHOO rows of small pores over most of the remainder of the bone surface. This differs from the condition in Howgqualepis rostridens where the dermal surface of the corresponding area is covered ina mixture of ridges and raised tubercles with no porous ornamentation (Long, 1988. Fig.15). Donnrosenia has very similar ornamentation on the clavicle but has entirely linear ornamentation on the cleithrum (Long, Choo & Young, 2008. Fig.6). Moythomasia durgaringa and M. nitida also have porous ornamentation on the pectoral girdle, but restricted to the ventral faces of the cleithrum and clavicle (Choo, in prep) whereas pores are also present on the lateral surface in ?H. youngorum. Fins The pectoral fin (Fig. 2) is incomplete with no traces of the endoskeletal radial although the visible lepidotrichia are well preserved. The fin is elongate and triangular with more than 14 primary lepidotrichia present. As with H. rostridens and Donnrosenia, the anterior lepidotrichia are unsegmented for most of their length with secondary division restricted to the region near the fin margin. The trailing edge of the fin is not preserved and it is unclear if the posterior fin rays were fully segmented as in H. rostridens (see below). The fin reaches its maximum length at about the seventh primary ray, which is unsegmented for more than 75% of its length as in H. rostridens, longer than the c.65% unsegmented region in the fin of Donnrosenia (Long et.al, 2008). A short section of the leading edge is preserved with spine-like fringing fulcra formed by terminal branching of the leading fin rays. As with H. rostridens and Donnrosenia (see below) there is no medial contact visible between the distal hemilepidotrichia of each fringing fulcra on any of the fins. The pelvic fin (Fig. 3) is long-based and triangular. Its preserved lateral aspect and does not appear to be as elongate as in H. rostridens although it is unclear if a section of the posterior margin is missing. The fins are located approximately midway along the body between the pectoral and anal fins. Primary lepidotrichia are only preserved for the anterior half of the fin, comprising more than 22 rays suggesting more the 40 primary rays on the entire preserved section. These rays are evenly segmented along their preserved length. Slender spine-like fringing fulcra are present on the leading edge. The anal fin (Fig. 4a, b) is large and triangular in shape. At least 54 primary segmented lepidotrichia are present as opposed to c.45 fin rays on the anal fin of H. rostridens. It is unclear if the fin originally had a short posterior fringe trailing behind the main Proc. Linn. Soc. N.S.W., 130, 2009 Figure 3. ?Howqualepis youngorum sp. nov. a. pho- tograph and b. line drawing of the pelvic fin and associated squamation on ANU V2929A. triangular area of the fin as in H. rostridens. If this was the case then the complete fin would have probably had over 60 primary lepidotrichia. As in the other fins, shortened spine-like lepidotrichial segments form a serrated cutwater of fringing fulcra on the leading edge. As was the case in other known Devonian actinopterygians, the caudal fin (Fig. 4) was heterocercal in structure with a distinct posterior cleft separating the dorsal lobe (notochordal mass of the fin plus the dorsal hypochordal lobe) from the ventral hypochordal lobe. While little of its dorsal counterpart has been preserved, the ventral hypochordal lobe is elongate and triangular with c.40 primary lepidotrichia preserved. Spine-like fringing fulcra are present on the leading edge. The dorsal fin is not preserved in the holotype although a pair or large, isolated lepidotrichs preserved near the counterpart tail may have originated from that fin (Fig. 4d). Scales and squamation Articulated macromeric scales, scutes and basal fulcra are preserved from the caudal fin, extending 4] A DEVONIAN ACTINOPTERYGIAN FISH i Naeem lem Figure 4. ?Howqualepis youngorum sp. nov. a. photograph and b. line drawing of the anal and caudal fins of ANU V2929A, c. photograph and d. line drawing of the caudal fin of the holotype counterpart (ANU V2929B). forwards to above the anal fin (Fig. 4c, d). There are also isolated patches of scales preserved above and to the rear of the pelvic fins (Fig. 3). Very little of the scale ornamentation has been preserved. The visible scale types are described in accordance with the zonation terminology as proposed in in Esin (1990) and employed in Trinajastic (1999). Area C = flank scales extending from above the pelvic fins to above the anal fin. Scales are elongate and rectangular, with rostrocaudal length being at least twice the height of the scale. Ventral margin is gently convex. The disposition of the peg and socket articulation is unknown in the scales close to the pelvic fins and absent in the scales near the anal fin. Free field ornamentation is poorly preserved but individual scales show remnants of longitudinal furrows. Scales from near the front and rear of the field seem to have two or three serrations protruding along the caudal edge suggesting little or no rostrocaudal decrease in the number of serrations. Area D = scales anterior to the caudal fin and on the notochordal mass of the caudal fin. 42 Scales anterior to the caudal fin are rhombic in form, becoming smaller and increasingly elongate on the notochordal mass of the fin. Scales near area C have a gently convex ventral margin, becoming less prominent towards the caudal fin until the margin is completely straight at those scales near the caudal inversion. Peg and socket articulation is absent. The free field is smooth with no preserved traces of raised ornamentation. Posterior serrations range from two in scales near area C to none on those scales on the caudal fin. Area H = scales adjacent to the base of the anal fin. These scales are small, elongate rhomboids. Peg and socket articulation is not visible and probably absent. There is no evidence of ornamentation or posterior ridges. The only dermal scutes that have been preserved are an articulated series visible anterior to the dorsal caudal lobe and extending over the dorsal margin of the caudal fin (Fig 3b, c). Anterior to the caudal fin, the scutes are triangular plates with a caudally- Proc. Linn. Soc. N.S.W., 130, 2009 B. CHOO directed apex and are about three times longer than the adjacent flank scales. As the series progresses posteriorly over the notochordal mass of the caudal fin, the scutes narrow and spine-like with extensive overlap between the individual scutes. Redescription of the pectoral fin of Howqualepis rostridens Long (1988) described the pectoral fin of Howgqualepis rostridens as consisting of 25 primary lepidotrichia that are unsegmented for most of their extent, save for some secondary division near the fin margin. A complete pectoral fin was not figured and re-examination of this form has revealed the fin to be more extensive than previously recognised (Fig. 5). Additionally, the leading edge of the pectoral and other fins was described as having short, parallel rays similar to fringing fulcra, but not paired (ibid). A similar condition in Donnrosenia led to Long et. al (2008) to diagnose the Howqualepididae as possessing short spine-like lepidotrichia in lieu of true fringing fulcra. The anterior two-thirds of the fin consist of c.25 lepidotrichia that possess extensive proximally unsegmented sections that in some specimens display distal bifurcation. At the lateral margins, these primary rays branch into a fringe of narrow, segments. The relative length of the proximal rays to the segmented 5mm f.ful fringe is variable, with the unsegmented region accounting for between 75-90% of the length of the fin. There appears to be no correlation between the degree of distal segmentation and the size of the specimen. Posterior of the unsegmented rays are at more than 10 additional primary lepidotrichia that are segmented from base to margin, again displaying a variable degree of distal branching. The pectoral fin of H. rostridens was thus broader in shape and less-extensively unsegmented than has previously been described. In the majority of specimens, the delicate elements of the posterior rays and terminal fringe are scattered or missing, leaving only the thick unsegmented proximal sections in articulation. This configuration of the pectoral fin-rays is similar to that of a number of Carboniferous taxa including Rhadinichthys (Moy-Thomas & Bradley Dyne, 1938). On the leading edge of the pectoral fins of Howgqualepis rostridens, ?H. youngorum sp.nov and Donnrosenia, the terminal sections of the otherwise unsegmented marginal fin rays branch at least twice, to forming narrow spine-like elements that are not obviously paired. These elements are called “terminal lepidotrichia” in Cheirolepis (Pearson and Westoll, 1979) and Melanecta (Coates, 1998) or “cutwater lepidotrichia” in the Howqualepididae Figure 5. Pectoral fin of Howqualepis rostridens. a. photograph and b. line drawing of the fin of MV P.160857. c. photograph and d. line drawing of the fin of MV P.160851B. In this specimen, the posterior section has partially torn off and folded to be visible ventral of the anterior edge of the fin. Proc. Linn. Soc. N.S.W., 130, 2009 43 A DEVONIAN ACTINOPTERYGIAN FISH (Long, Choo and Young, 2008). In a recent study, such structures fall into Arratia’s “Pattern A” class of fringing fulcra, formed from overlapping branched projections of the anteriormost lepidotrichia (Arratia, in press), a condition found in all undisputed Devonian actinopterygians with the exception of Tegeolepis which appears to lack any sort of spiny cutwater (Dunkle and Schaeffer, 1973). The fulcra of Cheirolepis, which are of similar form to those of the Howqualepididae, comprise distally enlarged hemilepidotrichia that partially enclose their paired counterparts (Arratia, in press). The more obviously paired structures present in Moythomasia and “Mimia” (also falling within “Pattern A”) are the result of the terminal segments being of equal length and in medial contact. Given that the scheme proposed by Arratia (and adopted here) means that all Devonian fringing fulcra are in fact modified spine- like lepidotrichia (merely differing in the nature of contact between the hemilepidotrichia), the diagnosis of Howqualepidiae has been adjusted accordingly in the systematic description. DISCUSSION Long, Choo and Young (2008) erected the Howqualepididae, comprising Howqualepis rostridens from Mount Howitt, Victoria and Donnrosenia schaefferi from the Aztec Siltstone of Antarctica. ANU V2929 appears to represent a third taxon within this clade (Fig. 6). All three fish have an elongate body form with macromeric squamation; long-based pelvic fins; small fringing fulcra without medial contact between the distal hemilepidotrichia, and extensive unsegmented primary lepidotrichia that comprise most of the length of the pectoral fin. Among the other Devonian actinopterygians, only 7egeolepis clarki (Dunkle and Schaeffer, 1973) possesses extensive unsegmented pectoral lepidotrichia but is distinguished from the Gondwanan forms in lacking a terminal segmented fringe on the pectoral fins, in possessing micromeric squamation and having small, short-based pelvic fins. Assigning the Bunga Bed taxon to a genus is rendered difficult owing to the lack of key skull characters that are used to characterise Howqualepis rostridens from the similar Donnrosenia. For example, H. rostridens possesses an extremely long maxillary blade, a dentigerous rostral and small, dorsoventrally compressed premaxillae (Long 1988). Donnrosenia displays a short, deep maxillary blade, dorsoventrally prominent premaxillae, a small accessory operculum and much smaller teeth than Howqualepis (Long, 44 Choo and Young, 2008). ANU V2929 is considered to be closer H.rostridens in having more extensive unsegmented pectoral lepidotrichia and relatively smaller scales than Donnrosenia. The pectoral fins of ANU V2929 are more similar to that of H. rostridens in that both forms possess unsegmented lepidotrichia that account for over 75% of the maximum length of the fin. Those of Donnrosenia account for less than 70% of the maximum fin length (Long, Choo and Young, 2008. Hige7): Based on these anatomical similarities and pending the discovery of skull material for this taxon, ANU V2929 is tentatively assigned to Howqualepis. The Bunga Bed form is not conspecific with H. rostridens and is distinguished in having a larger anal fin with a greater number of primary lepidotrichia and in possessing porous dermal ornamentation of the pectoral girdle. The presence of a grade of Devonian actinopterygian so far found exclusively in Middle Devonian freshwater deposits of southeastern Australia and Victoria Land, Antarctica highlights the close biogeographical similarity between the fossil faunas of these two regions. The apparent absence of these ray-finned fishes in Devonian sites outside this area also adds to a growing body of fossil evidence that indicates a regionally endemic freshwater vertebrate fauna within Middle Devonian Eastern Gondwana. Similarities in key taxa of placoderms (Young 1988, Young and Long 2005), acanthodians (Long 1983, Young 1989, Young & Burrow 2004), chondrichthyans (Young 1982, 2007; Long & Young 1995) and dipnoans (Long 1992, 2003) have been well documented. ACKNOWLEDGEMENTS Thanks to Gavin Young and John Long for their supervision and helpful suggestions regarding this manuscript. Thanks also to Gloria Arratia for helpful discussion regarding fringing fulcra. Ben Young is commended for his excellent fossil preparation and photography. Proc. Linn. Soc. N.S.W., 130, 2009 B. CHOO Se os SS SSS SSS TN AS SS C4 Ceri pemyh— tt Sa 7] ASO Sy SSeS SESE SS a SSS SSS SS SSSA SEES cate Wat tere TAA Seewecween= Lis — mest: Sas tt cr ek wee Seo REL SoS LL a ch EEEREnawenensns= sy Ake Se SENS mm Figure 6. Comparison of the three known species of the Howqualepididae. Reconstructions presented in lateral view and are not to scale. Unknown parts of the anatomy are represented by dark grey areas. a. ?Howqualepis youngorum sp.novy., based on the preserved extent of the holotype with outline based on H. rostridens, c.14cm long. b. Howqualepis rostridens from Mount Howitt, Victoria (modified after Long, 1988). Size of specimens range from 20-50cm. c. Donnrosenia schaefferi from the Aztec Siltstone, South- ern Victoria Land, Antarctica (from Long et.al, 2008), c. 14cm long. REFERENCES Arratia, G. (in press). Identifying patterns of diversity of the actinopterygian fulcra Acta Zoologica (Stockholm). Basden, A.and Young, G.C. (2001). A primitive actinopterygian neurocranium from the Early Devonian Taemas Formation, Burrinjuck area, New South Wales, Australia. Journal of Vertebrate Paleontology, 21, 754-766. Proc. Linn. Soc. N.S.W., 130, 2009 Basden, A., Young, G.C., Coates, Mand Ritchie,A. (2000). The most primitive osteichthyan braincase? Nature, 403, 186-188. Burrow, C. J. (1996). Taphonomy of acanthodians from the Devonian Bunga Beds (late Givetian — early Frasnian) of New South Wales. Historical Biology 11, 213-228. Cas R. A. F., Edgar C., Allen R. L., Bull S., Clifford B. A., Giordano G. and Wright J. V. (2000). Influence of magmatism and tectonics on sedimentation in 45 A DEVONIAN ACTINOPTERYGIAN FISH an extensional lake basin: the Upper Devonian Bunga Beds, Boyd Volcanic Complex, south- eastern Australia. International Association of Sedimentologists Special Publications 30, 175-200. Choo, B., Long, J.and Trinajstic, K. (in press). A new genus and species of basal actinopterygian fish from the Upper Devonian Gogo Formation of Western Australia. Acta Zoologica (Stockholm). Coates, M. I (1998). Actinopterygians from the Namurian of Bearsden, Scotland, with comments on early actinopterygian neurocrania. Zoological Journal of the Linnean Society 122, 27-59. Cope, E. D. 1887. Geology and Palaeontology. American Naturalist, 1014-1019. Dunkle, D.and Schaeffer, B. (1973). Tegeolepis clarki (Newberry), a palaeonisciform from the Upper Devonian Ohio Shale, Palaeontographica Abteilung A 143, 141-158. Esin, D.N. (1990). The scale cover of Amblypterus costata (Eichwald) and the palaeoniscid taxonomy based on isolated scales. Paleontological Journal 2,90-98. Fergusson C. L., Cas R. A. F., Collins W. J., Craig G. Y., Crook K. A. W., Powell C. McA., Scott P. A. and Young G. C. (1979). The Late Devonian Boyd Volcanic Complex, Eden, N.S.W. Journal of the Geological Society of Australia 26, 97-105. Gardiner, B.G. (1984). The relationships of the palaeoniscoid fishes, a review based on new specimens of Mimia and Moythomasia from the Upper Devonian of Western Australia. Bulletin of the British Museum (Natural History) Geology, 37, 1-428. Gardiner, B.G. and Bartram, A.W.H. (1977). The homologies of ventral cranial fissures in osteichthyans, pp. 227-245. In S. M. Andrews, R. S. Miles and A. D. Walker (eds.), Problems in Vertebrate Evolution. Academic Press, London. Giordano G.and Cas R. A. F. (2001). Structure of the Upper Devonian Boyd Volcanic Complex, south coast New South Wales: implications for the Devonian — Carboniferous evolution of the Lachlan Fold Belt. Australian Journal of Earth Sciences 48, 49-61. Huxley, T. H. (1880). On the applications of the laws of evolution to the arrangement of the Vertebrata and more particularly of the Mammalia. Proceedings of the Zoological Society of London, 1880, 649-662. Jessen, H. (1968). Moythomasia nitida Gross und M. cf. striata Gross, Devonische palaeonisciden aus dem oberen Plattenkalk der Bergish-Gladbach- Paffrather Mulde (Rheinisches Schiefergebirge). Palaeontographica Abteilung A 128, 87-114. Long, J.A. (1988). New palaeoniscoid fishes from the Late Devonian and Early Carboniferous of Victoria. Memoirs of the Association of Australian Palaeontologists, 7, 1-64. Long, J. A.; Choo, B.and Young, G. (2008).A new basal actinopterygian fish from the Middle Devonian Aztec Siltstone of Antarctica. Antarctic Science 20, 393- 412. 46 Moy-Thomas, J. A.and Bradley Dyne, M. (1938) The actinopterygian fishes from the Lower Carboniferous of Glencartholm, Eskdale, Dumfriesshire. Transactions of the Royal Society of Edinburgh, 59, 437-480. Pearson, D. M.and Westoll, T. S. (1979) The Devonian actinopterygian Cheirolepis Agassiz. Transactions of the Royal Society of Edinburgh 70, 337-399. Rickard M. J. and Love S. (2000). Timing of megakinks and related structures: constraints from the Devonian Bunga-Wapengo Basin, Mimosa Rocks National Park, New South Wales. Australian Journal of Earth Sciences 47, 1009-1013. Schultze, H.-P. (1968). Palaeoniscoidea-Schuppen aus dem Unterdevon Australiens und Kanadas und aus dem Mitteldevon Spitzbergens. Bulletin of the British Museum (Natural History): Geology, 16, 342-368. Trinajstic, K. (1999). Scale morphology of the Late Devonian palaeoniscoid Moythomasia durgaringa Gardiner and Bartram, 1997. Alcheringa, 23, 9-19. Young, G.C. (1982). Devonian sharks from south-eastern Australia and Antarctica. Palaeontology 25, 817-843. Young, G.C. (1999). Preliminary report on the biostratigraphy of new placoderm discoveries in the Hervey Group (Upper Devonian) of central New South Wales. Records of the Western Australian Museum, Supplement 57, 139-150. Young, G. C. (2007). Devonian formations, vertebrate faunas and age control on the far south coast of New South Wales and adjacent Victoria. Australian Journal of Earth Sciences 54, 991-1008. Proc. Linn. Soc. N.S.W., 130, 2009 Fire and Habitat Interactions in Regeneration, Persistence and Maturation of Obligate-seeding and Resprouting Plant Species in Coastal Heath PETER J. MYERSCOUGH Institute of Wildlife Research, School of Biological Sciences, The University of Sydney, Sydney, NSW 2006. Email: pmyersco@bio.usyd.edu.au Myerscough, P.J. (2009). Fire and habitat interactions in regeneration, persistence and maturation of obligate-seeding and resprouting plant species in coastal heath. Proceedings of the Linnean Society of New South Wales 130, 47-61. After a fire in January 1991, populations of two obligate-seeding and two resprouting species were followed from seeds sown in dry heath and wet heath on Pleistocene beach sands in the Myall Lakes area. In each type of heath, there were four plots, each with ninety 25 X 25 cm quadrats in which seeds of the four species had been sown in various combinations and surface soil conditions. All four wet-heath plots burned again in January 1998, as did two of the dry-heath plots. The two obligate-seeding species were confined to their respective habitats early in the life cycle; Acacia ulicifolia to dry heath by lack of seeds and suitable conditions for seedling emergence in wet heath; Dillwynia floribunda to wet heath by failure of its seedlings to survive in dry heath. The two resprouting species were confined to their respective habitats in different ways; Banksia oblongifolia by failure of its seedlings to survive in dry heath; Banksia aemula by lack of suitable soil surface in wet heath for establishment of its seedlings. In both species of Banksia, seedlings require a lignotuber to survive their first fire, and may persist several years without appreciable growth. Manuscript received 11 August 2008, accepted for publication 17 December 2008 KEY WORDS: banksias, fire, heath, lignotubers, maturation, oskars, persistence, regeneration, resprouters, seeders INTRODUCTION Dispersal, survivalandreproductionofindividuals underlie patterns of distribution and abundance of species. Fire influences these processes in plant life histories, and moulds patterns evident in fire-prone vegetation across gradients in habitat. In fire-prone vegetation, species of seed plants tend to fall into two groups (Gill 1981), obligate-seeders, those whose adult plants die in fires that destroy their leaf canopies and regenerate after fire solely from seed, and resprouters, some of whose plants survive complete loss of their canopy in intense fires and resprout new canopies after fire from vegetative tissue that is protected from fire. Frequent fires may act selectively and reinforce the respective characteristics of these two groups of plants. In obligate-seeders, high production of seeds, in amount and early availability after fire, would be expected, with the seeds protected from burning, either by being contained in fire-resistant fruits or by dispersal to safe sites in soil and having dormancy that is only readily broken by stimuli connected with the passage of fire. In resprouters, seedlings would be expected to produce at an early stage vegetative parts that survive fire. Pate et al. (1990) showed that seedlings of obligate-seeding species devote much growth to their shoots and early seed production, while seedlings of comparable resprouting species devote a high proportion of their growth to underground tissues including fire-resistant vegetative storage organs. In seedlings of resprouters, production of fire-resistant vegetative tissue typically precedes seed production. In their life histories, seed production is usually considerably delayed compared with related obligate-seeding species. Obligate seeders probably have simpler relationships linking seed dispersal, germination and seedling establishment in particular environments, their regeneration niches (sensu Grubb 1977), to seed production than do resprouters. Resprouters, while passing through seed dispersal, germination and seedling establishment in particular environments, their regeneration niches, also have periods of persistence as vegetative plants, that though fire-hardy, FIRE AND HABITAT INTERACTIONS IN HEATH PLANTS may or may not become reproductive and produce seed. It is possible that resprouters may simply persist many years as small plants with little net growth. The persistence niche (sensu Bond and Midgley 2001) of resprouters may be wider than conditions under which the plants progress to seed production. The opportunity arose to observe through time seedlings of obligate-seeding and resprouting species after fire occurred across habitats in fire-prone coastal heath. Such heath occurs in south-eastern Australia on leached siliceous beach sands and dunes deposited during the Pleistocene from South Australia (Specht 1981) to sand islands such as North Stradbroke Island (Clifford and Specht 1979) off the south-eastern coast of Queensland. On the coast of New South Wales, they occur particularly north of Newcastle to the Queensland border (Griffith et al. 2003, Keith 2004). In the Myall Lakes area, heath occurs on a Pleistocene system of beach sands in the Eurunderee Embayment of Thom et al. (1992). On these sands, there is a catenary sequence of soils and vegetation with dry heath on ridges, wet heath on slopes and swamps in periodically waterlogged swales (Carolin 1970, Myerscough and Carolin 1986, Myerscough et al. 1995). Dry heath belongs to the Banksia _ serratifolia (aemula) Alliance of Beadle (1981) and Wallum Sand Heaths of Keith (2004), and wet heath to Beadle’s (1981) Banksia aspleniifolia (oblongifolia) Alliance and Keith’s (2004) Coastal Heath Swamps. Fire has occurred fairly frequently, but over two decades produced no detectable effect in changing the pattern of differentiation of vegetation across the sequence of habitats, though changes with time since fire were clearly evident in the vegetation within habitats (Myerscough and Clarke 2007). Carolin (1970) demonstrated that various species occupy characteristic ranges of habitat in the catenary sequence from the ridges to the swales. Myerscough et al. (1996) and Clarke et al. (1996) investigated in four species how occupancy of their ranges of habitat might arise through dispersal of seed and, after fire, germination and establishment of their seedlings. Seedlings of two species characteristic of wet heath, obligate-seeding Dillwynia floribunda and resprouting Banksia oblongifolia, did not survive in dry heath, despite their seeds occurring and germinating there (Myerscough et al. 1996). Seed of two species characteristic of dry heath, obligate-seeding Acacia ulicifolia and resprouting Banksia aemula, were at best rare in wet heath (Myerscough et al. 1996), and, unless the soil surface is artificially disturbed and seeds are buried, germination and seedling establishment did not occur (Clarke et al. 1996). In short, it was largely in regeneration niche (Grubb 1977) that these species 48 appeared to be segregated to their respective habitats, D. floribunda and B. oblongifolia to wet heath, and A. ulicifolia and B. aemula to dry heath (Myerscough et al. 1996, Clarke et al. 1996). Seedlings of the four species were observed beyond the phase of establishment. Establishment of seedlings of Acacia ulicifolia and Banksia aemula had occurred in wet heath, following experimental manipulation of the soil surface (Clarke et al. 1996). Early survival of seedlings of both obligate-seeding species was related to type of habitat, but in both resprouting species it was related to variation among plots within types of habitat (Clarke et al. 1996). In this paper, ongoing survival of seedlings of the two obligate-seeding species is examined in relation to type of habitat, while in the two resprouting species it is examined in relation to variation among plots within types of habitat. Fire recurred in six of the eight experimental plots seven years after the fire that immediately preceded the start of the experiment. Survival of seedlings of the two resprouting species through fire could thus be assessed. Benwell (1998) observed lignotubers in seedlings of Banksia aemula and B. oblongifolia in similar coastal heath and found their growth to be slow seven years after fire. Four years after our experiment started, lignotubers were observed on some seedlings of each of the two species. By using fire-proof tags and measuring sizes and positions of lignotubers, survival of seedlings through their first fire could be assessed in relation to size and position of their lignotubers, if indeed they had been formed. Auld (1987) had found in Angophora hispida that seedlings with buried lignotubers survived fire better than those lacking lignotubers or with them exposed above the soil surface. Ongoing observation of lignotubers and sizes of banksia seedlings was used to try to identify whether they were growing or merely surviving without net growth. One seedling of Banksia aemula was observed to flower and set fruit. It was thus possible to see whether a fire-proof stem was required for flowering and seed production as Bradstock and Myerscough (1988) found in juveniles of Banksia serrata. The questions investigated in this paper are: * Do patterns of early seedling survival in the two obligate-seeding species seen in relation to type of habitat continue into later stages, and how are these patterns related to flowering and seeding? * How are patterns of seedling survival in the two resprouting species related to characteristics of individual plots? ¢ What roles do formation, size and position of lignotubers play in the survival through fire of Proc. Linn. Soc. N.S.W., 130, 2009 P.J. MYERSCOUGH Table 1. Experimental plot locations and their transect, ridge position relative to coastline, habitat, and fire history since January 1991. Habitat Fire history Jan 1998 Nov 2006 WH Totally burnt Totally burnt WH Totally burnt Totally burnt Totally burnt but WH some scorched Totally burnt leaves present Totally burnt but Burnt but with WH some scorched some patches leaves present unburnt DH Totally burnt Totally burnt DH Totally burnt Totally burnt DH Unburnt Unburnt Plots GDA Transect Ridge 82° S29 E 2975S. T2W1 1125 2 Near ASH. S T2W3 21013 E T2 Far DO MOZS.° T3W1 22310 E T3 Near 29.083 S* T3W2 12.250'E T3 Mid 29.594 S* T2D2 71.040'E T2 Mid 29.500'S T2D3 71.026 IZ, Far DSS 4 T3D1 12302 E T3 Near 29.019'S ’ 32 22. 280F T3 Mid Most or less DH unburnt - one edge slightly scorched Mostly unburnt — some lightly scorched patches * 30 X 5-m plot extends to left of marker post when facing inland; other plots extend to right of post. Dry heath (DH) and wet heath (WH). seedlings of the two resprouting species? * Do seedlings of the two resprouting species show appreciable net growth, and under what conditions may they do so? * Do patterns of seedling growth and survival give evidence of the longterm stability of the patterns observed in the vegetation across habitats of this coastal heath? METHODS AND MATERIALS Study area The heath studied was on sands of a Pleistocene beach system in the Euruderee Embayment of Thom et al. (1992). Twenty-four plots, 3 wet-heath and 3 dry-heath plots in each of 4 transects, were used by Myerscough et al. (1995) to analyse floristic variation in heaths across the system. Each plot was 30 X 5m with its longer sides parallel to the nearest beach ridge. Eight of the plots, two wet-heath and two dry-heath plots on each of the two central transects, were used Proc. Linn. Soc. N.S.W., 130, 2009 in the experiments of Myerscough et al. (1996) and Clarke et al. (1996). Each of these plots (Table 1) was divided into a grid of 150 square-metre cells. Ninety cells were randomly chosen and to each of these cells a 25 X 25 cm quadrat was randomly allocated to a particular experimental treatment. Experimental treatments, including placement of seeds of the four species of this study, are described in Myerscough et al. (1996). These ninety quadrats in each of the 8 plots were the areas in which seedlings that arose in 1991 were observed. Data collection Periodic counts of seedlings of Acacia ulicifolia and Dillwynia floribunda were maintained from 1991 until the fire of January 1998 burned six of the eight plots (Table 1). Between 1995 and 1997, due to the density of stems, especially in wet heath, it became increasingly difficult to count seedlings of D. floribunda and A. ulicifolia on the small quadrats. Since there was no seedling recruitment apparent during this period, where a greater number of seedlings 49 FIRE AND HABITAT INTERACTIONS IN HEATH PLANTS on a plot was recorded six months after the previous count, the greater number was taken to be correct. After January 1998 until October 2008, individuals of Acacia ulicifolia continued to be counted on the unburnt plots T3D1 and T3D2. In Banksia aemula and B. oblongifolia, all survivors of the 1991 cohort of seedlings were counted on the eight plots. In November 1995, surviving banksia seedlings were marked with fireproof metallic tags on stainless steel pins placed beside the seedlings. All seedlings of Banksia aemula were tagged. In B. oblongifolia, many more seedlings were then surviving, and in those quadrats where there was more than one seedling only one seedling in the 25 X 25 cm quadrat was randomly selected and tagged. The proportion of individuals tagged in November 1995 and the number of tagged individuals subsequently surviving were used to estimate the population of surviving seedlings of B. oblongifolia in each plot. When no tagged individuals had survived in a plot, it was assumed that the whole cohort of seedlings that had arisen in 1991 in the experimental quadrats of the plot had died. Lignotuber development was followed on each of the tagged seedlings, noting whether a lignotuber was absent or present. If present, its mean width was recorded from two measurements taken in two directions at right angles, and if it was not entirely buried, the height of its top above the soil surface was measured. After the fire of 1 January 1998, in March 1998 survival of the tagged seedlings was assessed. A seedling was scored as dead if it failed to resprout and live if it had resprouted. Most seedlings that resprouted had done so by March 1998, but a few resprouted later and were identified as alive when scored some months later. In all tagged seedlings, alive or dead, lignotuber presence or absence was noted, and, if present, its mean width was measured and whether its top was buried or exposed. The top was scored as exposed if its height above the soil surface was greater than | mm. Survival of seedlings through the fire was assessed in relation to habitat and lignotuber presence and exposure above the soil using 2 X 2 contingency tables and Chi square statistic. Growth of banksia seedlings between 1995 and 2007 was assessed from lignotuber width and plant height. In Banksia aemula, seedlings were deemed to have grown if in October 2007 they were found to have a lignotuber width of over 40 mm or a plant height of greater than 40 cm, while in Banksia oblongifolia seedlings with a lignotuber width of over 20 mm were deemed to have grown. Widths of lignotubers of Banksia aemula were not easily assessed in a consistent way through time for two reasons. Firstly, 50 although two measurements of width taken at right angles to each other were made on each occasion, not all lignotubers are radially symmetrical. Secondly, the lignotubers form with a thick bark, as in the sister species Banksia serrata (Beadle 1940, Bradstock and Myerscough 1988), and this bark may erode so that measured widths of lignotubers may lessen in time. Thus it is possible that some of the seedlings of Banksia aemula deemed not to have grown between 1995, or from when their lignotuber formed if it was later than 1995, may actually have grown slightly. Watertables were observed in the plots between 1991 and 1997, and their depths recorded as described in Myerscough et al. (1996). The fire of January 1998 prevented further observations, destroying tops of the plastic pipes used to observe depths to the watertable on 6 of the 8 plots. The depths given in Table 2 were measured on 23 September 1997, when the watertable was relatively high. To illustrate key floristic variation observed in 1990 across the plots, the nineteen most abundant species were selected from Appendix II of Myerscough et al. (1995) and listed in Table 2 in the order in which they were sorted in the TWINSPAN analysis given in Appendix I of Myerscough et al. (1995). The nineteen species included Banksia aemula, B. oblongifolia and Dillwynia floribunda. The other species, Acacia ulicifolia, whose seedlings were observed on the experimental plots was also included. The height of the canopy of each of the plots was recorded in September 2005 in ten randomly selected 1 X 1 m cells, except in T2W3 where inadvertently there were only nine cells. In each cell, the species of the tallest plant was noted. At the same time, the degree to which each surviving banksia seedling was shaded by surrounding vegetation was subjectively scored using a five-point scale of shade: 5, >95%; 4, 95-75%; 3, <75-25%; 2, <25% shaded; 1, seedling’s canopy unshaded. Nomenclature Nomenclature of plant names used follows Harden (1990, 1992, 1993 and 2002). RESULTS The plots differed floristically and in depths to the watertable (Table 2). Depths to watertable were greater in dry heath than in wet heath plots (F, =5.90 (p just >0.05)) and differed markedly among plots within habitats (F,,,=299.3 (p<0.001)). The habitats differed in plant species that provide significant cover. Both habitats had shrubs with appreciable Proc. Linn. Soc. N.S.W., 130, 2009 P.J. MYERSCOUGH Table 2. Experimental plots and mean (SE) depth (cm) to watertable and mean cover (“%) of twenty species (RS, resprouter; OS, obligate-seeder). Plot T2W1 T2W3 T3W1 T3W2 T2D2 T2D3 SIDI T3D2 2.0 10.6 2.8 18.9 43.3 26.3 108.8 38.5 Watertable OD CI CA" MIDE NOS MOGY OMOAR OS) Empodisma minus RS Gymnoshoenus sphaerocephalus 29 RS Leptospermum livesidgei RS 44 Banksia oblongifolia RS Dillwynia floribunda OS Epacris obtusifolia OS Xanthorrhoea fulva RS Lepyrodia interrupta RS 11 15 34 18 3 4 59 36 16 Darwinia leptantha OS Pseudanthus orientalis RS Persoonia lanceolata OS Kunzea capitataOS 1 9 5 14 ; 1 Dillwynia retorta OS Leptospermum polygalifolium RS 49 25 3 Leptospermum trinervium RS Acacia ulicifolia OS Banksia aemula RS 1 20 52 26 23 Melaleuca nodosa B RS Hypolaena 10 Vi 1S 4 fastigiata RS Epacris pulchella OS Proc. Linn. Soc. N.S.W., 130, 2009 5] FIRE AND HABITAT INTERACTIONS IN HEATH PLANTS cover such as the banksias, Banksia aemula in dry heath and B. oblongifolia in wet heath. Wet heath had more cover from resprouting monocotyledons such as Xanthorrhoea fulva than dry heath, and more cover from obligate-seeding shrubs such as Dil/lwynia floribunda and Epacris obtusifolia with sparsely branched, elongate ascending stems. Though similar in depth to the watertable, the wet heath plots T2W1 and T3W1 differed in plant cover. T2W1 had high cover of Empodisma minus and Gymnoschoenus sphaerocephalus. After the fire of January 1991, and sowing seeds in March 1991 under various treatments across the eight plots, as described in Myerscough et al. (1996), seedlings of Banksia aemula, B. oblongifolia, Acacia ulicifolia and Dillwynia floribunda differed in their patterns of survival across the plots (Table 3). In dry heath plots, seedlings of Dillwynia floribunda, though fairly numerous at six months, suffered heavy mortality and were completely absent after four years. They persisted in all wet heath plots with approximately 10% of the population observed at six months present six years later, with some plants observed to have flowered after three and half years. All the plants in the plots were killed by the fire of January 1998. In short, it was only in the wet heath plots that plants of D. floribunda survived and reproduced, doing so with little plot to plot variation apparent in their survival (Table 3). Seven and a half years after the fire in January 1998, D. floribunda was among emergent species in the canopy of the wet heath plots (Table 4). Some seedlings of Acacia ulicifolia survived from 1991 in each of the eight plots until the fire of January Table 4. Experimental plots and height (m) of canopy (mean (SE)), emergent species and relative shading (*RSh) of banksia seedlings (numbers in each category) in September 2005. Plot T2W1 T2W3 T3W1 T3W2 T2D2 T2D3 T3D1 T3D2 Ganon erent 1.50 e337) 1.38 1.09 2.08 1.93 DMS 1.63 (0.06) (0.06) (0.06) (0.07) (0.14) (0.16) (0.16) (0.08) @ Emergent Dif 3 Dfl2 Daria Dfl2 B.ae 3 B.ae4_ Bae4 B.ael Species - number TiS Io.15 1 JET JEG Ml te S LL GAM L.tr 4 L.tr 4 of contacts out of S.in 3 S.in 2 S.sp 1 D.re 2 10 (but out of 9 for Piaal las Pla3 T2W3) A.el | L.po | E.mi | E.mi 1 E.ob 2 E.mil B.ob 1 B.ob 1 B.ob 1 A.te | B fal Crem Mn | Wp | X fu | K.ca | B. aemula RSh 5 3 4 4 3 9 5 2 2 3 3 4 5 Sil 2 5 2 1 2 10 1 1 5 1 1 1 B. oblong- Rsh 5 folia 4 2 2 3) 1 1 1 2 2 1 @ A.el — Acacia elongata; A.te — Acacia terminalis; B.ae — Banksia aemula; B. ob — Banksia oblongifolia; B.fa— Boronia falcifolia; C.te — Calytrix tetragona; D.fl— Dillwynia floribunda; D.re — Dillwynia retorta; E.mi — Epacris microphylla; E.ob — Epacris obltusifolia: K.ca - Kunzea capitata; L.li— Leptospermum liversidgei; L.po — Leptospermum polygalifolia; L.tr — Leptospermum trinervium: M.n — Melaleuca nodosa; P.la— Persoonia lanceolata; S.in — Sprengelia incarnata; S.sp — Sprengelia sprengelioides; W.p — Woollsia pungens; X.fu — Xanthorrhoea fulva. * RSh: 5, >95%; 4, 95-75%; 3, <75-25%; 2, <25% shaded; 1, seedling’s canopy unshaded. 52 Proc. Linn. Soc. N.S.W., 130, 2009 P.J. MYERSCOUGH I I I G cE Os Og Cade L if 14 ¢ vl 69 IdeL 8C LE OV 8V L9 Ch Vell €dcl €l 91 v6 CUCL 8 8 Il Il II I] II Il I] I] I] IO. Stl 691 COC 6th OCh Ltr CM EL 6 6 6 6 6 Cl él 81 81 81 ral 98 i oils Gs Wir Wwe WAC IME L fl iL L fl fl (L Ol Ol Ol Ol €V 6V 09 v8 99 9LC LBC EMCL € 6 6 Cl cl 66 Stel WAN SOE SOG EG ESC IMCL py ofisuo]qo g Ol Ol Gl Cl Cl Gil tall €l vl Vall vi LI LI ial LI vv Iv 6S cae L I I it G € € v Ol 9 cv IGEL Ol Ol Cl Cl Cl Cl Gl cl Gl Gil Gl vv OV LV 8 VS igs VS €dcl I I I It I I C C G C C Gl LI {| LI CC GG CS CaCL VA VV cv LV LV LV SV 8V 8V OS IS 19 19 19 (69) 89 $9 teil CMEL Gil vl 8I OC OC Ke Ic GG GC €C €C Of O£ O€ [€ Of CE CE [MEL 9 9 9 9 9 9 9 9 9 9 9 6 6 6 Ol LV 4 LV EMCL V ¢ L 6 Ol cl Cl Cl tall €l el Ic IC GG CC LE 6 VC IMCL pinwap “g if I IL €CC cde L I € 68 IGEL € (G0) cdc I G 0cI CUCL 06 Icl 8Il Lcel OE COE 096 CMEL Ne €9 06 O€l coc Otc LOD IME ce 6£ OS C8 CLI 9SC SOS EMCL 8V 8S L8 LOI Sc O06C LOY IMCL ppunqisoyf ‘q I It I I € € € € € iS ¢ ¢ Cl vi cl SS Cae ¢ ¢ 8 8 8 8 Il Cl Cl el el VI VI VI 6l 6 6c CS IGE Ol Ol Il el Li LI eV €dcL C G C v Ol VI eV CUCL I I I V CC VA 99 CMEL € (e V All Of Se OL [MEL Ol Ol Ol LI CC 6£ (6S EMCL € ¢ [l GG €€ Gu VS IMCL pyofiayn y SS) OMS EES IES A SIS tell SCI IOI OS OL BOG SS Wy YS) OSS Sin iE MOG CS Tar out "1661 YAR] UL SUIMOS JdUIS (SAvdA) BUNTY YSNOAY) Spor [eJUIUITAIdX9 UO SUTATAINS SSUI[ Pes JO JOAQUIN *¢ 9IqUL 53 Proc. Linn. Soc. N.S.W., 130, 2009 FIRE AND HABITAT INTERACTIONS IN HEATH PLANTS 1998 burned six of the plots. Though survival varies considerably with plot, there is no clear pattern in this variation in relation to habitat or other characteristics of plots. In the two plots not burned in 1998, one plant continued to survive in T3D2 until it was fifteen and a half years-old, while in T3D1 five plants were still alive at 17.65 years (Table 3), four of them having fruited in 2008. In this plot, four-year-old plants flowered and fruited, and four-year-old plants were seen flowering on other plots (T2D3 and T2W3). In Banksia oblongifolia, some seedlings survived on each of the eight plots up to two years (Table 3). On dry heath plots, they had died out after 5 years on both plots where the watertable was deep (T2D2 and T3D1) but continued to survive in significant number on T2D3, the dry heath plot with the least depth to the watertable (Table 2). No seedling of B. oblongifolia survived the fire of January 1998 on a dry heath plot, but on each of the four wet heath plots some seedlings survived. On T2W1, the plot with high cover of Gymnoschoenus sphaerocephala and Empodisma minus (Table 2), no seedling survived twelve years, but on the other three wet heath plots some seedlings survived up to seventeen years (Table 3). In Banksia aemula, some seedlings survived on each of the eight plots up to nine and half years, including on the six plots totally burnt in the fire of January 1998. Their numbers were lowest on the two dry heath plots (T2D2 and T3D1) where the watertable was deep and cover of Leptospermum trinervium relatively high (Table 2). On T3D1, which had the deepest watertable (Table 2) and which was not burnt in 1998 (Table 1), the last survivor had died after ten years. On each of the other seven plots, at least one plant survived to seventeen years (Table 3). Among wet heath plots, there was heavier mortality of survivors of the fire of January 1998 on T2W1 and T3W1 (see years 7.64 to 17.65 in Table 3), plots with the shallowest watertable and highest cover of resprouting monocots (Table 2), than on T2W3 and T3W2. On T2W3 and T3W2, not only was the watertable deeper and the cover of monocots less (Table 2), but in September 2005 the surviving seedlings of Banksia aemula were less shaded (Table 4). There was one seedling of B. aemula on each of these plots that was unshaded (Table 4). On T2W3, one plant flowered at fourteen years and formed swollen follicles, and, after the fire in November 2006, six follicles appeared to have opened. This was the only banksia originating from seed in 1991 that was observed on any of the eight plots to have become reproductive. Across the wet heath plots, mortality from the fire of | January 1998 was much higher among seedlings of Banksia oblongifolia (83%) than among those of B. aemula (23%) (p<0.001). In both species of banksia, survival of seedlings on plots burnt in the fire of 1 January 1998 entirely depended on possessing a lignotuber; without a lignotuber no seedling survived (Table 5). Under comparable conditions in wet heath, the lignotubers of B. aemula survived better than those of B. oblongifolia. With the top of the lignotuber exposed, only 10% of seedlings of B. oblongifolia survived whereas 78% of those of B. aemula survived; with the lignotuber buried, 36% survived in B. oblongifolia and 95% in B. aemula. No tagged seedling of B. oblongifolia survived fire in a dry heath plot (Tables 3 and 5), while seedlings of B. aemula survived fire in both wet heath (WH) and dry heath (DH). The survival of B. aemula seedlings was much lower in DH (24%) than in WH (76%) not only because there was a higher proportion of seedlings without lignotubers in DH (24%) than in WH (7%) (p<0.001) but there was higher mortality of seedlings with lignotubers in DH (68%) than in WH (18%) (p<0.001). Burial of the lignotuber Table 5. Number of tagged banksia seedlings live or dead in March 1998 after fire of 1 January 1998 in relation to habitat and lignotubers. Species Banksia aemula Banksia oblongifolia Habitat Dry heath Wet heath Dry heath Wet heath Seedlings Live Dead Live Dead Live Dead Live Dead Lignotuber: absent 0 14 0 8 0 3) 0 5 present 14 30 93 21 0 10 19 117 Lignotuber top: buried 11 21 20 I 0 2 8 14 exposed 3) 9 We 20 0 8 1] 103 54 Proc. Linn. Soc. N.S.W., 130, 2009 P.J. MYERSCOUGH Table 6. Dimensions and relative shading (RSh) of grown tagged banksia seedlings in October 2007. Species Banksia aemula Banksia oblongifolia Dimension Mean Plant Relative Mean Plant Relative lignotuber height shading lignotuber height shading width (mm) (cm) (RSh)@ ~~ width (mm) (cm) (RShH)@ Dry heath plot T3D2 4] 43 3 Wet heath plots Al 14 2 3 16 2 T2W3 43 Sil 2 39 28 2) 74* Ue ] 33 76 ] T3W2 37) 72 1 * Plant first flowered in 2005. @ RSh: 1, seedling’s canopy unshaded; 2, <25%; 3, <75-25% shaded increased the chances of survival of seedlings, particularly in B. oblongifolia. In B. aemula, the extent of this was mediated by habitat. A higher proportion of lignotubers were buried in DH (73%) than in WH (18%) (p<0.001). Despite this, mortality of seedlings with buried lignotubers was much higher in DH (66%) than in WH (5%) (p<0.001), whereas seedlings with lignotubers exposed above ground suffered 75% mortality in DH and 22% mortality in WH (p<0.001). In short, though burial of their lignotubers enhanced survival of seedlings in both habitats, it was more effective in WH than DH though the proportion of seedlings with buried lignotubers was lower in WH than DH. Of those tagged banksia seedlings surviving to October 2007, appreciable growth was detected in relatively few (Tables 6 and 7), and most of these seedlings occurred in one wet heath plot, T2W3. Indeed, in this plot, two of the three surviving seedlings of Banksia oblongifolia, and four of the six surviving seedlings of Banksia aemula had grown, with one of them flowering in 2005 and producing an infructescence with a single swollen follicle. This individual was the only seedling to have had a lignotuber over 40 mm in width by March 1998; no others had achieved this by September 2005. In March 2007, it had four infructescences on which a total of six follicles had opened after the fire in November 2006. This was the only tagged banksia Proc. Linn. Soc. N.S.W., 130, 2009 seedling to have reached reproductive maturity. In all the other plots, there were only two tagged banksia seedlings that could be identified as having grown, both B. aemula, one on a dry heath plot, T3D2, and the other on a wet heath plot, T3W2. The rest of the surviving tagged banksia seedlings appeared to be simply surviving without net growth, and on the wet heath plot T3W2 such seedlings of B. aemula were particularly numerous (Table 7). In October 2007, all seedlings deemed to have grown were unshaded or <25% shaded (Table 6), except for the seedling on T3D2, a plot largely unburnt by the fire of November 2006 (Table 1). All the tagged banksia seedlings surviving in October 2007 had originated on quadrats sown in March 1991 with seed of their own species, except for three seedlings; a seedling of Banksia aemula on T2D3, another on T3W2 and a seedling of B. oblongifolia on T3W1 (Table 8). All six seedlings of B. aemula that had grown since their lignotubers were first recorded (Table 6) had each originated from seed sown and then shallowly buried (Table 8). In wet heath plot T3W2, the pattern of survival of the relatively numerous seedlings of B. aemula in October 2007 appears to reflect reasonably closely the original 4:6:4 ratio in March 1991 of seed buried: seed sown on disturbed surface: seed sown on undisturbed soil surface among the quadrats on the plot (Table 8). 55 FIRE AND HABITAT INTERACTIONS IN HEATH PLANTS Table 7. Dimensions (mean (S.E.)) of tagged banksia seedlings deemed not to have grown between first recorded presence of lignotuber (in November 1995 unless otherwise indicated) and October 2007. ~ Species NNNBaIsiaacs Ain) ain nD anksiatob oneriol cman Number Initial 2007 Plant | Number Initial 2007 Plant of plants lignotuber lignotuber height of lignotuber lignotuber height width width in plants width width m (mm) (mm) 2007 (mm) (mm) 2007 Dry heath plots T2D2 1 20 33 T2D3 10 15 19 (2) (2) T3D2 9 15 22 (1) (3) Wet heath plots T2W1 5 18 16 (2) (1) T2W3 Die 17 15 T3W1 14 23 DD (2) (2) T3W2 43@ 20 19 11 i 16 15 12 3 14 12 13 (3) (3) 12 ae 12 1] 8 * | plant first record of lignotuber in March1998; @ 4 plants first record of lignotuber in November 1996, and 4 in March 1998. DISCUSSION Fire and habitat interaction Fire and habitat variation interact in different ways across the four species of this study. The interaction is more complex in the two resprouting species than in the two obligate-seeding species. Of the two obligate-seeding species, Dillwynia floribunda has the more straightforward relation with habitat and fire. After fire, seedlings emerge from seeds whose dormancy has been broken by heat, as in Acacia ulicifolia (Auld and O’Connell 1991). Though its seedlings can appear in dry heath, they only survived to maturity in wet heath (Table 3). In wet heath, its soil seed-bank was found by Myerscough et al. (1996) to be abundant, survival of plants after six months was high (Table 3), it was 56 seen to be in flower three and a half years after fire and to be one of the emergent species in the canopy seven and a half years after fire (Table 4). It is one of a suite of obligate-seeding species with soil seed banks and similar sparsely branched erect stems with microphyllous leaves that emerge above resprouting monocotyledons characteristic of wet heath. Other such species are the heaths Epacris microphylla, E. obtusifolia, Sprengelia incarnata, S. sprengelioides, some of which occur with D. floribunda in fire-prone wet heaths on sandstones in the Sydney region (e.g., Keith and Myerscough 1993, Keith 1994, Keith et al. 2007a). Seedlings of Acacia ulicifolia arose in both wet and dry heath particularly after shallow burial of heat-treated seed (Clarke et al. 1996). Survival varied among plots in both wet and dry heath, but there Proc. Linn. Soc. N.S.W., 130, 2009 P.J. MYERSCOUGH Table 8. Number of tagged banksia seedlings live in October 2007 (grown: plants deemed to have grown; dwarfs: plants deemed not to have grown since their lignotubers were first recorded) in relation to how their seed was placed in March 1991 on or within the soil. Species Banksia aemula Banksia oblongifolia Seed placed Buried Surface Surface not Buried Surface Surface not disturbed disturbed disturbed disturbed Dry heath plots T2D2 dwarf 1 T2D3 dwarfs 2 5° 3 T3D2 dwarfs 1 4 4 grown | Wet heath plots T2W1 dwarfs 1 4 T2W3 dwarfs ] l 1 grown 4 1 1 T3W1 dwarfs 7 6 1 Oh i T3W2 dwarfs 12 I) 12% 2 grown i * includes one seedling that arose in a quadrat not sown with seed of that species. was at least one survivor in each plot immediately before the fire in January 1998 burned six of the plots (Table 3). Some plants were observed to flower at about three and a half years old, and, in an unburnt dry heath plot, plants flowered and set fruit until they were at least seventeen years old. The seedlings are, compared to those of Dillwynia floribunda, slow to gain in height, and are quickly overtopped in wet heath by resprouting monocotyledons, such as Gymnoschoenus sphaerocephalus in T2W1. A modest seed bank of A4.ulicifolia was shown to occur in dry heath but none was found in wet heath (Myerscough et al. 1996). Thus, in wet heath, lack of seed and, for any seed reaching it, scarcity of conditions for successful seedling emergence appear to exclude Acacia ulicifolia, and occurrence of the species is confined to dry heath where it has a soil seed bank, suitable conditions occur after fire for germination of seed and emergence of seedlings, and seedlings are less readily overtopped by other understorey species. Thus Dillwynia floribunda and Acacia ulicifolia are excluded from each other’s characteristic habitat Proc. Linn. Soc. N.S.W., 130, 2009 early in the life cycle, though at different stages; A. ulicifolia through lack of available seed and suitable safe sites (sensu Harper 1977) for any rare seeds present in wet heath, and D. floribunda apparently by lack of suitable growing conditions for seedlings in dry heath. In short, their respective distributions relate to their regeneration niches (sensu Grubb 1977). Beyond the regeneration stage, they need to reproduce successfully in their respective habitats, which observations in this study, while not detailed, indicate occurs, with some seedlings of each species in their fourth year probably contributing seed to the soil seed-bank. This study reveals that in this coastal heath the two resprouting species have three critical phases in their life cycle, regeneration, persistence and growth. What happens to individual plants as they enter and pass through each phase and make transition from one stage to the next depends on fire and habitat. This differs between the two species. Transition from seedling to persistent plant is made evident through fire, while that from fire-resistant but merely persistent plants 57 FIRE AND HABITAT INTERACTIONS IN HEATH PLANTS to plants growing toward reproductive capability is more gradual, presumably depending on success in garnering necessary resources. In the regeneration phase, patterns of seedling establishment between habitats (Myerscough et al. 1996) and among experimental treatments and plots within habitats (Clarke et al. 1996) differed between Banksia aemula and B. oblongifolia. Seedlings of B. oblongifolia that arose in dry heath were fewer and died earlier than on wet heath plots, and, though several survived on T2D3 for six and a half years, none survived the fire in January 1998 (Table 3). In contrast, survival of Banksia aemula occurred across both habitats, and on all the wet heath plots and on three of the dry heath plots there was at least one survivor after seventeen years (Table 3). Survival of seedlings of B. aemula was least on dry heath plots (T2D2 and T3D1) with low water tables (Tables 2 and 3). The fire of 1 January 1998 caused mortality among seedlings of both species, but mortality was much greater in B. oblongifolia than in B. aemula. Overall, in the wet heath plots, 77% of seedlings of B. aemula survived the fire while only 17% did in B. oblongifolia. In both species, to persist through the fire a lignotuber was essential (Table 5). Lignotubers had formed by four and a half years from the sowing of the seed on most of the seedlings that survived, but some had formed somewhat later (Table 7). One factor in the lower survival of seedlings of Banksia oblongifolia is the structure of the lignotubers its seedlings form. They are small and lack the thick corky bark of the larger lignotubers of the seedlings of B. aemula. In B. oblongifolia many of the unburied lignotubers formed completely above the ground surface while this did not occur in seedlings of B. aemula; in them, a lower part was at least in the ground. In both species, as Auld (1987) showed in seedlings of Angophora hispida, burial of the lignotuber enhanced survival of the seedlings, though again to a greater extent in Banksia aemula than in B. oblongifolia. In short, the lignotubers of seedlings of B. aemula appear to be better insulated than those of seedlings of B. oblongifolia, and the transition of seedlings through fire to the fire-resistant persistent phase is made with much less mortality in B. aemula than in B. oblongifolia. In each of the banksia species, very few of the surviving seedlings showed detectable growth between March 1998 and September 2007. Most of them appeared to be simply persisting without detectable net growth. They seem to be in a prolonged “sit-and-wait” state, ageing juvenile plants that Silvertown (1982) called oskars. In many plant communities, growth of 58 such oskars is restricted by lack of sufficient light. Though some shading occurs in the heaths, especially wet heath with abundant monocots in the understorey (Tables 2 and 4), lack of growth in these banksia seedlings is not solely related to light (Table 4). Indeed light was abundant at ground level for several weeks after fire, as occurred when the seedlings arose from seed in 1991 and immediately following their survival through the fire of 1 January 1998. If their growth is resource-limited, the critical resources are those in the soil. Water is probably readily available across the range of habitats, though water stress may be a factor in dry heath with deeper water tables (Table 2). The limiting resources are likely to be one or more of the mineral nutrients needed for plant growth. Previous work (Myerscough and Carolin 1986) has indicated that the sands on which these heaths occur are very low in mineral nutrients. Circumstantial evidence that shortage of mineral nutrients retarded growth of the seedlings comes from the seedlings that grew. All except two were from the wet heath plot T2W3. On this site, when holes were drilled to observe the watertable, a very consolidated coffee rock, B horizon, was reached at c. 0.5 m in three of the four holes. In other wet heath plots, B horizons were deeper and less consolidated. Data of Griffith et al. (2004) indicate that roots of seedlings of both species of banksia, particularly B. aemula, may grow down to B horizons fairly rapidly in similar heaths. It is thus possible that banksia seedlings on this plot could reach the B horizon relatively easily and extract nutrients from it. Furthermore, the plant that grew early and reached reproductive maturity was relatively near one of the holes that had pierced the B horizon to observe the watertable; the disturbance of the hole may have released nutrients that accelerated its growth. Incidentally, this individual flowered when it was less than a metre high (Table 6), showing no sign of requiring an elongated stem for flowering, as in Banksia serrata (Bradstock and Myerscough 1988). Its inflorescences were produced more or less sessile on a thickened main stem that was merely an upward extension of the thickening of the lignotuber. Other low-growing reproductive individuals of B. aemula on this sand system had a similar growth form, while taller growing individuals with trunks occur in sites such as T2D2 and T3D1, and, after fire, produce inflorescences on newly grown stems up to | m long. Growth leading to mature reproductive individuals, persistence of fire-resistant juveniles and regeneration in terms of establishment of seedlings appear as fairly distinct phases in the life cycles of B. aemula and B. oblongifolia. Each phase has its characteristic relations with habitat and fire that differ Proc. Linn. Soc. N.S.W., 130, 2009 P.J. MYERSCOUGH between the species. In B. oblongifolia, its restriction to wet heath is clearly evident at the regeneration phase (Myerscough et al. 1996), and, as seen in this study, should seedlings survive in a dry heath site they tend to be eliminated in the first fire and thus never enter the phase of fire-resistant juveniles (Tables 3 and 5). Fire-resistant juveniles of B. oblongifolia occurred in wet heath plots, but in the plot, T2W1, having survived the fire of 1 January 1998, they were eliminated (Table 3), probably shaded out under the high cover of Empodisma minus and Gymnoschoenus sphaerocephalus (Table 2). In the other three wet heath plots some continued to survive to seventeen years, but only clearly entering the growth phase in one, T2W3. In Banksia aemula, given availability of seed and modification of the soil surface (Myerscough et al. 1996, Clarke et al. 1996), seedlings arose and survived in all wet and dry heath plots. Ongoing survival was least in the two dry heath plots T2D2 and T3D1 (Table 3) with deep watertables. Transition through the fire of 1 January 1998 on six of the eight plots to persistent fire-resistant juveniles was made with high rates of survival, particularly in the wet heath plots (Table 3). The question arises as to whether the patterns seen at the regeneration stage in seedlings in relation to particular soil treatments applied at sowing of the seeds in March 1991 (see Clarke et al. 1996) were maintained or altered in subsequent survival. In T3W2, the plot with highest number of fire-resistant juveniles persisting at seventeen years, the indication is that the pattern seen in the regeneration phase in relation to soil surface disturbance and seed burial is retained at seventeen years in the persistence phase (Table 8). This plot incidentally was unique among the four wet heath plots in showing little effect of soil treatment and seed burial in numbers of seedlings Surviving at the regeneration phase (see Fig. 2 of Clarke et al. 1996); the three other plots all showed that the greatest number of seedings arose from buried seeds. All six of the plants that were deemed to have entered the growth phase had arisen from buried seed (Table 8). These findings give some insight into the status of populations of the two banksia species on the Eurundereee Pleistocene beach ridges. Firstly, they suggest that their population turn-over is very slow. Indeed, after seventeen years, there is little firm evidence of effective recruitment in either species. In Banksia aemula, only one surviving juvenile showed any evidence of growth in dry heath. While, in wet heath plots, there were numbers of persistent fire- resistant juveniles, a few of which grew, it was an artificial situation brought about by firstly unnaturally Proc. Linn. Soc. N.S.W., 130, 2009 increased availability of seed, relative to naturally occurring levels of seed (Myerscough et al. 1996), and secondly by burial of seeds which is unlikely to occur readily in nature in wet heath (Clarke et al. 1996). In Banksia oblongifolia, very few seedlings survived the fire on 1 January 1998 and persisted as fire-resistant juveniles. The only two that grew arose from seed that in one case had been buried and in the other from seed on a disturbed surface (Table 8). Casual observation of existing mature individuals of either B. aemula or B. oblongifolia suggests that over the seventeen years there was little if any mortality among them. The picture then is of populations of mature long-lived individuals into which there is little opportunity for recruitment of juveniles. Secondly, it appears that, though juveniles may persist several years in a non-growing state, they are limited by lack of resources for growth to progress to mature plants. It is probable that on most plots, the limiting resources are soil nutrients. In the case of one wet heath site with cover of Empodisma minus and Gymnoschoenus sphaerocephalus, lack of light may eliminate juveniles of Banksia oblongifolia, even though mature plants of the species had appreciable cover (Table 2). This suggests that either the current mature plants recruited as seedlings before E. minus and G. sphaerocephalus were so abundant in the site, or, if they were present and abundant, fire frequency was so high that shade from them did not eliminate juveniles of B. oblongifolia. Though the evidence indicates that, presently on these Pleistocene sand ridges, niches for effective regeneration, persistence and growth for these two resprouting species are rare, there must be periods when they are in colonising mode and these niches are more common. This would have been so for B. aemula when parts of Holocene dunes south of Mungo Brush between the Myall River and the sea were colonised by it. Their winged seeds, dispersed some distance in wind, as in those of Banksia serrata observed by Hammill et al. (1998), particularly in willy-willies as were seen to occur in the area of this study on the Pleistocene beach ridges after an intense fire in January 1991, appear well suited for the initial step in colonisation of new habitat. Selection and mode of regeneration The contrast is stark between the two obligate- seeding species studied in which in suitable habitat regeneration is followed very quickly by reproductive maturity of individuals, and the two resprouting species where formation of a fire-resistant lignotuber occurs early and fire-resistant individuals enter a period of persistence in which the majority in this 59 FIRE AND HABITAT INTERACTIONS IN HEATH PLANTS study showed no demonstrable growth toward maturity. As Keith et al. (2007b) have pointed out, the resprouters thus show all the characteristics of Grime’s (1979) stress-tolerators or Stearns’ (1976) K-selected species, while the obligate-seeders are examples of Stearns’ r-selected species. Whether, under selection, their breeding systems follow the suggestion of Heslop-Harrison (1964, Table IV, p. 200) that species with short life cycles, exemplified here by obligate- seeders, are more likely to be inbreeders while species with longer life cycles and slowly maturing adults, exemplified by resprouters, are more likely to be out- breeders would be interesting to establish. It is possible that paths to extinction may differ between obligate-seeders and resprouters. Resprouters may lose effective reproduction through seedlings and reachaterminable state ofa few mature long-persisting individuals, perhaps propagating as clones, while high levels of inbreeding may lead to extinction in some obligate-seeders. How far, in fire-prone habitats, general differences exist between obligate-seeders and resprouters in degrees of in and out-breeding, and thus levels of heterozygosity of individuals, is a question that is yet to be investigated. There is an indication in Table 2 that the species with high cover ten years from fire in both habitats are either strongly obligate-seeding or resprouting, with the possible exception of Pseudanthus orientalis. This would support the suggestion that, in vegetation subject to fairly frequent fires, as appears to have been so in these heaths (Myerscough and Clarke 2007), selection is strong for individuals and thus species to be either markedly obligate-seeding or strongly resprouting and against individuals and species that are neither markedly one nor the other. To establish this as a general rule would require further work. ACKNOWLEDGEMENTS The work began with support from an ARC Small Grant (1990-92) in collaboration with Nicholas Skelton and Peter Clarke. Beside Nicholas Skelton and Peter Clarke, Neil Tridgell, Ian Radford, Joan Myerscough, Andrew Denham, Alan Keating and James Myerscough each helped in the field, especially Neil Tridgell who accompanied me many times; I thank them all. I thank Tony Auld for the fire-proof tags used with the banksia seedlings, and Tony Auld and Andrew Denham for loan of callipers modified by Murray Ellis for measuring diameters of lignotubers. The work was done under licence from the Director of the New South Wales National Parks and Wildlife Service. Staff of Myall Lakes National Park helped with access to study sites, which is much appreciated. I thank an anonymous referee for constructive comments, and Jan Percival for help in getting the paper into correct electronic form for publishing. 60 REFERENCES Auld, T.D. (1987). Post-fire demography in the resprouting shrub Angophora hispida (Sm.) Blaxell: flowering, seed production, dispersal, seedling establishment and survival. Proceedings of the Linnean Society of New South Wales 109, 259-269. Auld, T.D. and O’Connell, M.A. (1991). Predicting patterns of post-fire germination in 35 eastern Australian Fabaceae. Australian Journal of Ecology 16, 53-70. Beadle, N.C.W. (1940). Soil temperatures during forest fires and their effect on the survival of vegetation. Journal of Ecology 28, 180-192. Beadle, N.C.W. (1981). “The vegetation of Australia’. (Cambridge University Press, Cambridge). Benwell, A.S. (1998). Post-fire recruitment in coastal heathland in relation to regeneration strategy and habitat. Australian Journal of Botany 46, 75-101. Bond, W.J. and Midgley, J.J. (2001). Ecology of sprouting in woody plants: the persistence niche. 7rends in Ecology and Evolution 16, 45-51. Bradstock, R.A. and Myerscough, P.J. (1988). The survival and population response to frequent fires of two woody resprouters Banksia serrata and Isopogon anemonifolius. Australian Journal of Botany 36, 415- 431. Carolin, R.C. (1970). Myall Lakes - an ancient and modern monument. The Proceedings of the Ecological Society of Australia 5, 123-129. Clarke, P.J., Myerscough, P.J. and Skelton, N.J. (1996). Plant coexistence in coastal heaths: between- and within-habitat effects of competition, disturbance and predation in the post-fire environment. Australian Journal of Ecology 21, 55-63. Clifford, H.T. and Specht, R.L. (1979). ‘The vegetation of North Stradbroke Island’. (University of Queensland Press, St Lucia, Queensland). Gill, A.M. (1981). Coping with fire. In “The biology of Australian plants’ (Eds. J.S. Pate and A.J. Mc Comb) pp. 65-87. (University of Western Australia Press, Nedlands). Griffith, S.J., Bale, C., Adam, P. and Wilson R. (2003). Wallum and related vegetation on the NSW North Coast: description and phytosociological studies. Cunninghamia 8, 202-252. Griffith, S.J., Bale, C. and Adam, P. (2004). The influence of fire and rainfall upon seedling recruitment in sand- mass (wallum) heathland of north-eastern New South Wales. Australian Journal of Botany 52, 93-118. Grime, J.P. (1979). ‘Plant strategies and vegetation processes’. (J. Wiley and Sons: London). Grubb, P.J. (1977). The maintenance of species richness in plant communities: the importance of the regeneration niche. Biological Reviews 52, 107-145. Hammill, K.A., Bradstock, R.A. and Allaway, W.G. (1998). Post-fire fire seed dispersal and species re- establishment in proteacous heath. Australian Journal of Botany 46, 407-419. Proc. Linn. Soc. N.S.W., 130, 2009 P.J. MYERSCOUGH Harden, G.J. (1990). “Flora of New South Wales, Volume 1°. (University of New South Wales Press, Sydney). Harden G.J. (1992). ‘Flora of New South Wales, Volume 3’. (University of New South Wales Press, Sydney). Harden, G.J. (1993). “Flora of New South Wales, Volume 4’. (University of New South Wales Press, Sydney). Harden, G.J. (2002). ‘Flora of New South Wales, Volume 2° (revised edition). (University of New South Wales Press, Sydney). Harper, J.L. (1977). ‘Population biology of plants’. (Academic Press: London). Heslop-Harrison, J. (1964). Forty years of genecology. Advances in Ecological Research 2, 159-247. Keith, D.A. (1994). Floristics, structure and diversity of natural vegetation in the O’Hares Creek catchment, south of Sydney. Cunninghamia 3, 543-594. Keith, D.A. (2004). “Ocean shores to desert dunes: the native vegetation of New South Wales and the ACT’. (Department of Environment and Conservation (NSW), Hurstville). Keith, D.A. and Myerscough, P.J. (1993). Floristics and soil relations of upland swamp vegetation near Sydney. Australian Journal of Ecology 18, 325-344. Keith, D.A., Holman, L., Rodoreda, S., Lemmon, J. and Bedward, M. (2007a). Plant functional types can predict decade-scale changes in fire-prone vegetation. Journal of Ecology 95, 1324-1337. Keith, D.A., Tozer, M.G., Regan, T.J. and Regan H.M. (2007b). The persistence niche: what makes it and what breaks it for two fire-prone species. Australian Journal of Botany 55, 273-279. Myerscough, P.J. and Carolin, R.C. (1986). The vegetation of the Eurunderee sand mass, headlands and previous islands in the Myall Lakes area, New South Wales. Cunninghamia 1, 399-466. Myerscough, P.J. and Clarke, P.J. (2007). Burnt to blazes: landscape fires, resilience and habitat interaction in frequently burnt coastal heath. Australian Journal of Botany 55, 91-102. Myerscough, P.J., Clarke, P.J. and Skelton, N.J. (1995). Plant co-existence in coastal heaths: floristic patterns and species attributes. Australian Journal of Ecology 20, 482-493. Myerscough, P.J., Clarke, P.J. and Skelton, N.J. (1996). Plant co-existence in coastal heaths: habitat segregation in the post-fire environment. Azstralian Journal of Ecology 21, 47-54. Pate, J.S., Froend, R.H., Bowen, B.J., Hansen, A. and Kuo, J. (1990). Seedling growth and storage characterisitics of seeder and resprouter species of Mediterranean-type ecosystems of S.W. Australia. Annals of Botany 65, 585-601. Silvertown, J.W. (1982). ‘Introduction to plant population ecology’. (Longman, London and New York). Specht, R.L. (1981). The sclerophyllous (heath) vegetation of Australia: the eastern and central states. In ‘Ecosystems of the world. Vol 9A. Heathlands snd related shrublands. Descriptive studies’ (Ed. R.L. Specht) pp. 125-210. (Elsevier, Amsterdam). Proc. Linn. Soc. N.S.W., 130, 2009 Stearns, S.C. (1976). Life-history tactics: a review of ideas. The Quarterly Review of Biology 51, 3-47. Thom, B.G., Shepherd, M., Ly, C.K., Roy, P.S., Bowman, G.M. and Hesp, P.A. (1992). Coastal geomorphology and Quaternary geology of the Port Stephens-Myall Lakes Area. Department of Biogeography and Geomorphology, ANU Monograph No. 6, Australian National University, Canberra. 61 FURY AND HLA BIT ARHDUDDE RAINE — PL pee bbs At Hert. \gatoenar nity vane MT COB A ab Mi asad iN: dausytt F964, we AS cepts, tA oF. wed) Yr ‘eRe o gral ied) | bare dg’ aie \ reratagt! wy ii i ‘yale Neg tv) ~wrtaley ee a | iene rts rah Bas ” nsehyhO WPT ARIE “APA NEAR pd S58 1 cme ites 4 pha lte ‘fond Li sa iid lil oindains sl rey, riots ney hansen PL ia WA). e ie 7 Thycanty t OME hire PPLE Ral RIS iy area FS CinePinnts any waite parte Tyee: oy Te mime te ay hry nny a pe ANE, by iy S hele Gath ey Re a ete eer ¢ asl Nevl A eedig tae Ip \ of] fae I el a) Vay mind |) Fix +, Dt ag ie Galt rk eat Why me ae rp TANK, rt hat ‘thw He euiny Sirgen hd tas iTiee ‘ i ie tian TOR, Fae vw Koel teil mee ei, Oh nti eine syetpoxe fiw ee iva! jaty #9 ae Py w4 real BS! ct an) At ‘Tey Het les ir ey We + a ‘ ay n Tie aig Jt iS abe HF He) ‘am phe nevis W? Chars o7 ed weeadtes a “ABS DMA: SID. EA ieH bik TU BREOM Hier ntlbe sy ORR RAMURY SIRNA e ar aA bind ee. t Shs yal! sepulert palate hon a PANE Aysabet, wcrrt cote aati Bo write ala ipo. wot No ea Nunbie dened seh iy pret amare SW! a {" re a diy attic ap se! W Conihel Mu, AAO) bint scala ab ies | SS: CHIBI eT ‘be MWasirag to pmay eno T FEA ly, ay ity Ae tah | SP Ra caved TRG Si io Chee iid rind tub pene Sia ete joe's ES? 7B ots ihe AKU Pela bd aod? ta 19% Baprnk Seth tat bak eve oom & NiCr S Bite area PR He Set ey: ‘ou tent OP ide yal. A naa dvrenl iene Hobie Rormiiil Micubs bag iia si8O: : tise ks i vier renin Ray ‘Henig sth | Rd ee ner ow ADL Kt rat siren ala oe t ; head Andries yy, J Sirah, © PE sicentont was” eastern weaiibenrsnats pnttegy ae i De eke sos. i Hiner Pr nd p (iRcey pce a WOR oie ine than DM sor, pie Si tila abuts Fade Sigh ahah au SHE kT Sitio tft Se HN aE ANS Pcotoginyl Socie nod aig welt re et Hn eRe Sead bind! eB caer Bae PERSE GR walt anaes T Hiseke Sy el aban a Oe BAN ea oS Td Malina NOL dette: (roee FL SRE baw U9 fade Fal aedheddy seth third Wd! storie Ve Sire base iin OS RAS "Gah st pas ty COeaernelared), cay. fiQ! 22 uorc8 | "a Ciel! ictal ia ee ie aa arsoniy iter Hilf Tacs oS Gs Aaa Stat ( ia Hedehiel, “Ania, eH ageRe Bat ‘SAHA fsa 7 Leb Chk Me CHEMO Police ae. Ea aahar sores “Ab ON arated Seas rae oaitll Aaah oi i nn ‘ghee 4 ‘it SoH} cH ‘es pee pret ®) pee pere spylesa \o. Yotirwtol it ri tila. 8 Swe Rob soi Seema OF ahd ili here HAR | FeO PNR a Vel BST Seb (Bay hh a wMS SSG abtichrier a! NOP re oO epee RIM, og Civile, SILT IEE Hidlpet Rigeh amok gis neightih bef etek or RortaaAP chery ts cour wha HANNS / held Wah e iis beth ee Tele Bite Ay SE SPARS) Site SPA SS tlh papraray OPES aa ih i Late Llandovery (Early Silurian) Dendroid Graptolites from the Cotton Formation near Forbes, New South Wales R. B. Rickarps!, A. J. WRIGHT? AND G. THOMAS? 'Department of Earth Sciences, University of Cambridge, Downing Street, Cambridge CB2 3EQ, U.K. (rbr1000@esc.cam.ac.uk); *School of Earth and Environmental Sciences, University of Wollongong, Wollongong NSW 2522 (tony_ wright@uow.edu.au) and Linnean Macleay Fellow; 3P.O. Box 130, Southland Centre, Victoria 3192. Rickards, R.B., Wright, A.J. and Thomas, G. (2009). Late Llandovery (Early Silurian) dendroid graptolites from the Cotton Formation near Forbes, New South Wales. Proceedings of the Linnean Society of New South Wales 130, 63-76. A well-preserved dendroid graptolite fauna of Early Silurian (late Llandovery: probable turriculatus graptolite zone) age is described from the Cotton Formation near Forbes, New South Wales. A possible rhabdopleuran hemichordate is described from Australia for the first time. The fauna consists of 13 taxa as follows: Dendrograptus sp. aff. D. avonleaensis, Dictyonema zalasiewiczi sp. noy., Dictyonema sp. aff. D. paululum australis, Dictyonema paululum australis, Dictyonema sp. aff. D. sp. cf. D. venustus of Bulman (?ssp. nov.), Dictyonema venustum, Dictyonema sp. ef. D. falciferum, Callograptus bridgecreekensis, Callograptus rigbyae, Callograptus sp. aff. C. ulahensis, Stelechocladia sp. cf. S. praeattenuata, Acanthograptus praedeckeri and ?Rhabdopleura sp. (? with zooids). The fauna is close in composition (although less diverse) and age to a dendroid fauna recently described from Bridge Creek near Orange, NSW, which was assigned to the slightly younger griestoniensis zone. Manuscript received 12 March 2008, accepted for publication 23 January 2009. KEY WORDS: Cotton Formation, dendroids, Early Silurian, Forbes, graptolites, New South Wales. INTRODUCTION The dendroid graptolites described here have been collected over many years by one of us (GT) from a quarry in the Cotton Formation at Cotton Hill near Forbes in western N.S.W. Fossils from these beds have been described by Sherwin (1974: graptolites) and Edgecombe and Sherwin (2001: trilobites). The described trilobite and graptolite faunas are from beds exposed in the quarry high in the upper part of the Cotton Formation (Sherwin 1974) and the graptolite fauna is correlated with the late Llandovery (Early Silurian) turriculatus graptolite zone (Edgecombe and Sherwin 2001). Despite the very nature of collections made in an active quarry, there seems little doubt that the bulk of the dendroid fauna and the graptoloids are from the same narrow horizon. The most similar known dendroid fauna was described by Rickards et al. (2003) from the Four Mile Creek district, south of Orange, NSW, and comparisons are made below with that fauna. AGE OF THE ASSEMBLAGE Although the Cotton Formation dendroid fauna (13 species and subspecies) is less diverse that described from the Bridge Creek localities in the Four Mile Creek district (24 species and subspecies) by Rickards et al. (2003), there can be little doubt that the two faunas are close in age. The largest assemblage at Bridge Creek, from locality F14, was referred by Rickards et al. (2003) to a horizon low in the griestoniensis graptolite Zone. The Cotton Hill fauna is assigned almost certainly to the stratigraphically lower turriculatus graptolite zone. Of the fauna we record here from the Cotton Formation, only Dictyonema zalasiewiczi sp. nov. and ?Rhabdopleura sp. have not been recorded from Bridge Creek at locality F1l4. Callograptus ulahensis Rickards et al., 2003 was recorded from a lower (gregarius Zone) assemblage at locality BF15 on Bridge Creek: the Cotton Hill Quarry species is referred to Callograptus sp. cf. C. ulahensis. EARLY SILURIAN GRAPTOLITES FROM NEAR FORBES Stelechocladia praeattenuata Rickards et al., 2003 was not recorded from F14 but occurs below (F19) and above (BF28, BF24 and BF 18), ranging from the gregarius Zone to the uppermost griestoniensis Zone. Sherwin (1973, 1974) referred the strata at Cotton Hill Quarry to the turriculatus Zone, with some levels probably earlier than this but without definite faunas. Sherwin (1970, 1973) also recorded Dictyonema spp. from the highest band of a group of beds yielding a likely turriculatus Zone fauna. Hence the two dendroid assemblages, from Cotton Hill (probable turriculatus Zone) and from Bridge Creek (griestoniensis Zone), are not dissimilar in age, the Cotton Hill fauna being about one graptolite zone lower. There is another difference between the two assemblages apart from a possible slight age difference and a diversity range, and that is that the Cotton Hill fauna is almost exclusively of slender, delicate species, often broken. In contrast, most of the species described by Rickards et al. (2003) from Four Mile Creek are robust, and are preserved in poorly bedded siltstone. The only robust form common to the two localities is Stelechocladia and at Cotton Hill it is known only from three small fragments showing distal, slender thecae. It is possible that the Cotton Hill assemblage lived in a quieter depositional environment, such as a lagoon, or further offshore. Edgecombe and Sherwin (2001) concluded that the laminated siltstones that dominate the formation were deposited in a ‘very calm’ environment, ‘most likely below storm surge wave base’. Associated graptoloids. Sherwin(1970, 1973) was the first to identify graptoloid species from the Cotton Beds, following the initial recognition of graptolites from this locality by Packham (1967). Sherwin (1970, 1973) recognised two faunas, an earlier assemblage (his fauna C) and a later assemblage (his fauna D) respectively from the east and west quarries on Cotton Hill: both are in the upper Cotton Formation. Fauna D, from the western, larger quarry, includes Dictyonema sp. (Sherwin 1973, fig. 10). Some mixing of faunas possibly occurred because collection was from large blocks on the quarry floor (Sherwin 1974, p. 149). It is this western quarry from which the present collection of dendroids came; the eastern, smaller, quarry has not so far yielded dendroid graptolites. The graptoloid assemblages were described in detail by Sherwin (1974) and, allowing for some possible mixing of faunas, the overall aspect is of a turriculatus Zone fauna, perhaps rather low in that horizon given the presence of Rastrites linnaei, Monograptus halli, and Monograptus sp. cf. M. sedgwickii. Thus the Cotton Hill quarry is at 64 a stratigraphically lower level than the Four Mile Creek (F14) locality which was mentioned in the preceding section and which is probably low in the griestoniensis Zone. Graptoloids occurring on the same rocks as the Cotton Hill dendroids described below include: Parapetalolithuspalmeus, ? Glyptograptus tamariscus, Monograptus andrewsi and Spirograptus turriculatus (Fig. 7b). The faunal lists given by Sherwin (1974) are fuller and much more reliable than the graptoloids at our disposal. Here we also record and illustrate (Fig. 7a) Parapetalothius palmeus (Barrande, 1850), a form not recorded by Sherwin (1974); its occurrence accords with his age attribution of the turriculatus zone. In their revision of Spirograptus, Loydell et al. (1993) assigned Sherwin’s (1974) Monograptus turriculatus (Barrande, 1850) to their new species Spirograptus guerichi. They further (Loydell et al. 1993, p. 924, text-fig. 7) stated that S. guerichi is “virtually confined to its biozone”, whereas S. turriculatus ranges through their turriculatus biozone into the crispus biozone; Sherwin (1974, p. 150) shows that both species occur in his fauna D. SYSTEMATIC PALAEONTOLOGY This benthic graptolite fauna has been assembled only by sustained and diligent collecting over many years by one of us (GT), as dendroids are rare at the locality. The preservation is of reddish brown graptolites against a very pale, fine-grained siltstone or mudstone. The specimens are often large but are mostly fragmentary, and there seems to be little in the way of burial distortion or twisting and no obvious tectonic deformation. Some specimens are preserved in three dimensions infilled, probably with goethite: in others the periderm is diagenetically flattened, but with some parts (e.g. stolons) pyritised. It is possible that in some instances pyritised zooids are present. Rarely stolons occur free on the bedding plane, the surrounding periderm having degenerated; this situation has been noted by Chapman et al. (1993) and Rickards et al. (2003). All specimens are deposited in the Australian Museum, Sydney, with numbers AM F123381-123428. Subphylum Pterobranchia Lankester, 1877 (nom. trans. Rickards and Durman 2006) Class Graptolithina Bronn, 1849 Order Dendroidea Nicholson, 1872 Family Dendrograptidae Roemer in Frech 1897 Proc. Linn. Soc. N.S.W., 130, 2009 R.B. RICKARDS, A.J. WRIGHT AND G. THOMAS Dendrograptus J. Hall, 1858 Synonymy aff. 2003 Dendrograptus avonleaensis n. sp.; Type species Rickards et al., pp. 312-3, figs 5A, 6A. Graptolithus hallianus Prout, 1851, subsequently designated by J. Hall (1862). Material AM F123381. Dendrograptus sp. aff. D. avonleaensis Rickards et al., 2003 Description Figures la, 3a The single specimen shows nine stipes and five Figure 1. a, Dendrograptus sp. aff. D. avonleaensis Rickards et al., 2003; AM F123381. b, Dictyonema sp. aff. D. cf. venustum Bulman, 1928; AM F123398. c, Dictyonema zalasiewiczi sp. nov., holotype AM F123402; d, Callograptus bridgecreekensis Rickards et al., 2003; AM F123403. e, Callograptus rigbyae Rickards et al., 2003; AM F123406. Black bars are dissepiments; black rods, arched in ventral views, are autothecal ventral processes. Scale bars 1 mm. Proc. Linn. Soc. N.S.W., 130, 2009 65 EARLY SILURIAN GRAPTOLITES FROM NEAR FORBES Figure 2. a-b, Dictyonema paululum australis Rickards et al., 2003, respectively AM F123382, AM F123383. c-d, Acanthograptus praedeckeri Rickards et al., 2003, respectively AM F123411, AM F123410. Scale bars 1 mm. branching points in mostly ventral view, but also shows autothecal profiles in places. Specimen almost flattened diagenetically, but periderm still slightly transparent. 15-20 autothecae in 10 mm, with a profile width of 0.2 mm, and a fairly simple aperture slightly arched ventrally as seen in the ventral view. Bithecae exceedingly inconspicuous, being tiny tubes opening externally alongside autothecal apertures and alternating along stipe. Branching of stipes may be in zones at 1-3 mm intervals; stipe lateral width 0.30 mm and dorsoventral width 0.40 mm. Remarks This specimen, probably representing the distal- 66 most parts of the colony, agrees closely with the type material of D. avonleaensis in most characters, especially the roughly zonal branching, autothecal nature and spacing and stipe dimensions. The type specimens from Bridge Creek (Rickards et al. 2003) had much of the proximal region preserved, and this is much more robust than the Cotton Hill material. Dictyonema J. Hall, 1851 Type species Gorgonia retiformis J. Hall, 1843, subsequently designated by Miller (1889). Proc. Linn. Soc. N.S.W., 130, 2009 R.B. RICKARDS, A.J. WRIGHT AND G. THOMAS Figure 3. a, Dendrograptus sp. aff. D. avonleaensis Rickards et al., 2003; AM F123381. b, Callograptus bridgecreekensis Rickards et al., 2003; AM F123403. c, Callograptus sp. aff. C. ulahensis Rickards et al., 2003; AM F123407. d, Dictyonema sp. cf. D. falciferum Bulman, 1928, AM F 123401. e, Dictyonema paululum australis Rickards et al., 2003; AM F123386. Scale bars 1 mm. Proc. Linn. Soc. N.S.W., 130, 2009 67 EARLY SILURIAN GRAPTOLITES FROM NEAR FORBES Figure 4. a, Dictyonema zalasiewiczi, sp. nov., holotype AM F123402. b, Dictyonema venustum Lapworth, 1881; AM F123397. c, Dictyonema paululum australis Rickards et al., 2003, respectively AM F123385. Scale bars 1 mm. Dictyonema paululum australis Rickards et al., 2003 Figures 2a-b, 3e, 4c, 7c Synonymy 2003 Dictyonema paululum australis n. subsp.., Rickards et al., p. 316, figs 7F-G, 9E, 12A. Material Twelve specimens, AM F123382-93, ranging from small fragments to almost complete colonies. 68 Description Probably fan-shaped rhabdosome of slender stipes, no indication of a conical colonial arrangement; colony with slender, parallel stipes, only approximately branching in zones, sometimes fanning out in rapid expansion. Stipes branch at intervals of 1.5-3.0 mm; stipe lateral width 0.20-0.25 mm proximally, 0.15 mm more distally; dorsoventral width 0.50-0.60 mm; stipe spacing 13-16 in 10 mm, stipe interspaces 0.50-0.60 mm. Autothecae denticulate, 18-20 in 10 Proc. Linn. Soc. N.S.W., 130, 2009 R.B. RICKARDS, A.J. WRIGHT AND G. THOMAS Figure 5. a, Dictyonema sp. cf. D. falciferum Bulman, 1928, partially preserved stolons in three portions of stipes where periderm is degenerate; AM F123400. b, Stelechocladia sp. cf. S. praeattenuata Rick- ards et al., 2003, AM F123408a. c, Acanthograptus praedeckeri Rickards et al., 2003, AM F123410. d-e, ?Rhabdopleura sp., respectively AM F123412-3; both exhibit stolons with possible preserved soft tissue (encysted zooidal attached). f, Callograptus sp. aff. C. ulahensis Rickards et al., 2003; AM F123407. Scale bars 1mm; stipple on Fig. a indicates possible attached soft parts. Scale bars 25 mm (a), 1 mm (b-f). mm; dissepiments slender, 0.05-0.10 mm, 14-20 in 10 mm. Dissepiments conspicuous because of their frequency; proximally they are more robust and perhaps sparser. Bithecal tubes seen in places but their apertural regions are difficult to discern; they may be of the type described by Bulman (1928) in D. falciferum where the bithecal apertural region hooks over the dorsal apertural region of the autotheca. Alternatively, they may grow short of the full hook (Fig. 3e); bithecal tubes 0.05 mm wide. Proc. Linn. Soc. N.S.W., 130, 2009 Remarks Rickards et al. (2003) considered the original material from Four Mile Creek probably had conical rhabdosomes but it seems more likely that they are fan-shaped. Bulman (1928) could not see the nature of the rhabdosome as a whole in the type subspecies, and he was particularly vague about the nature of the bithecae: otherwise the type subspecies is clearly close to the Australian form differing only as outlined by Rickards et al. (2003). Dictvonema paululum australis is the most common dendroid at Cotton Hill Quarry 69 EARLY SILURIAN GRAPTOLITES FROM NEAR FORBES Figure 6. a-b, Dictyonema sp. aff. D. paululum australis Rickards et al., 2003; respectively AM F123395, AM F123396. c, Callograptus rigbyae Rickards et al., 2003; AM F123405. d, Dictyonema sp. aff. D. sp. cf. D. venustum Bulman, 1928; AM F123398. Scale bars 1 mm. 70 Proc. Linn. Soc. N.S.W., 130, 2009 R.B. RICKARDS, A.J. WRIGHT AND G. THOMAS Figure 7. a, Parapetalolithograptus palmeus (Barrande, 1850) s.1., AM F123428. b, Spirograptus turricula- tus (Barrande, 1850), AM F 123427. c, Dictyonema paululum australis Rickards et al., 2003; AM F123384. d-e, Callograptus rigbyae Rickards et al., 2003; respectively AM F123404, AM F 123406. f, Acanthograp- tus praedeckeri, AM F123409. Scale bars 1 mm. Proc. Linn. Soc. N.S.W., 130, 2009 ral EARLY SILURIAN GRAPTOLITES FROM NEAR FORBES (and see also D. sp. aff. D. p. australis described below). Dictyonema paululum hanoverense Rickards et al., 2005 from the Late Silurian parultimus Zone near Neurea, N.S.W. differs in having an autothecal spacing of 28-30 in 10 mm and quite spinose ventral apertures. Dictyonema sp. aff. D. paululum australis Rickards et al., 2003 Figure 6a-b Synonymy aff. 2003 Dictyonema paululum australis subsp. nov.; Rickards et al., p. 316, figs 7F-G, 9E, 12A. Material AM F123394-6, 123415a-b. Description Nature of colony uncertain, possibly fan-shaped. Stipes with lateral width of 0.20-0.25 mm, and spaced at 20-22 in 10 mm, more or less parallel, and with interstipe spaces of 0.20-0.40 mm; branching roughly in zones every 1.0-2.5 mm. Autothecae spaced at 19- 20 in 10 mm; dorsoventral width uncertain but may be ca. 0.50 mm. Bithecae not detected. Dissepiments fine, spaced at ca. 20 in 10 mm. Remarks These specimens are superficially similar to those of the D. paululum australis material described in this paper, except that the stipes are more closely spaced and the interstipe spaces concomitantly narrow. There may be a temporal subspeciation factor involved here as the source level in the quarry for the specimens is uncertain; thus some of the D. p. australis specimens may be from older beds and others from the turriculatus level. Dictyonema venustum Lapworth, 1881 Figure 4b Synonymy 1881 Dictyonema venustum sp. nov.; Lapworth, pp. 171-2, pl. 7, fig. la-c 1928 Dictyonema venustum, Lapworth, emend; Bulman, pp. 61-3, pl. 5, figs 6-7, ?8, text-fig 34. 2003 Dictyonema venustum Lapworth, 1881; Rickards et al., pp. 315-6, figs 7A, 9D, 10B-D. Material An almost complete rhabdosome, AM F 123397, plus AM F123416-7. 2 Description Rhabdosome conical, reaching 8 mm x 8 mm; very proximal end missing though part of the holdfast may be present. Stipes with lateral width of 0.25-0.30 mm, dorsoventral width of 0.70 mm, and spaced at 16 in 10 mm. Interstipe spaces rectangular, up to 0.50 mm wide, and are bounded by stipes and dissepiments spaced at 5-8 in 10 mm. Dissepiments relatively robust, up to 0.15 mm thick. Autothecal spacing 16 in 10 mm; thecae appear to be denticulate but otherwise simple. Bithecal tubes present but relationships to autothecal apertures not seen. Remarks The specimen is very close to the type material redefined by Bulman (1928), differing only in a slightly closer spacing of the stipes. Dictyonema sp. aff. D. sp. cf. venustum Bulman, 1928 Figures 1b, 6d Synonymy aff. 1928. Dictyonema cf. venustum Lapworth, emend.; Bulman, pp. 62-3, pl. 5, fig. 8 (non 6-7). Material AM F123398; three other specimens (AM F123424-6) questionably assigned here. Description The large fragmental rhabdosome (AM F 123398) has 12 stipes preserved, spaced at 16 in 10 mm, with interstipe spaces of 0.10-0.40 mm, and spaced at ca. 1-6 in 10 mm. Lateral stipe width 0.25 -0.40 mm, usually nearer the latter. Autothecae unclear but may be spaced at ca. 20 in 10 mm with dorsoventral width of 0.50 mm. Remarks This specimen is very close to that figured by Bulman (1928, pl. 5, fig. 8) which he listed as D. venustum but he made it clear in the text that he placed it there only with reserve. As in the Cotton Hill quarry specimen the interstipe spacing 1s less and the stipes are more robust. The Girvan specimens illustrated by Bulman were said to come from communis zone beds (probably convolutus-sedgwickii zone in modern terminology); thus they may have come from pre- turriculatus Zone strata, and this is also possible in the case of the present specimen. Proc. Linn. Soc. N.S.W., 130, 2009 R.B. RICKARDS, A.J. WRIGHT AND G. THOMAS Dictyonema sp. cf. D. falciferum Bulman, 1928 Figures 3d, 5a Synonymy cf. 1928 Dictyonema falciferum n. sp.; Bulman, pp. 53-6, pl. 5, figs 1-3, text-figs 27-29. cf. 2003 Dictyonema falciferum Bulman, 1928; Rickards et al., p. 315, figs 5I, 8B, 9C, 10A. Material AM F123399-123401. Description Rhabdosome possibly fan-shaped (?conical), at least 25mm long and 18 mm broad, with numerous parallel stipes spaced at 14 in 10 mm, having stipe interspaces of 0.50-0.60 mm. Rectangular meshes are defined by stipes and conspicuous dissepiments spaced at 8-10 in 10 mm. Autothecal spacing 20 in 10 mm. Lateral stipe width 0.20-0.25 mm, and dorsoventral stipe width 0.50 mm. Autothecae appear to be simple denticulate but not spinose. Bithecal tubes present but their apertural regions unclear. Branching rather irregular, at 0.5-5.0 mm intervals. Remarks These specimens are closely similar to the specimens described from Four Mile Creek by Rickards et al. (2003) differing only in having a less regular branching pattern and slightly more parallel stipes. One specimen (AM F123400; Fig. 5a) has traces of preserved stolons. Dictyonema zalasiewiczi sp. nov. Figures Ic, 4a Material Holotype, AM F123402, an almost complete rhabdosome. Derivation of name After Dr. J. Zalasiewicz, University of Leicester, a leading graptolite worker. Diagnosis A Dictyonema species with 30-40 dissepiments in10 mm; stipes 0.2-0.5 mm wide and spaced at 0.2- 0.3 mm. Description Fan-shaped rhabdosome more than 30 mm long and over 20 mm wide, typified by its striking number of dissepiments, up to 40 per 10 mm, never less than 30. Dissepiments 0.05-0.10 mm across, often arched Proc. Linn. Soc. N.S.W., 130, 2009 distally, and quite frequently branching; commonly angled rather than normal to adjacent stipes, but also occur as closely spaced pairs. Stipes uniformly 0.20- 0.25 mm in lateral width, with branching every 2-2.5 mm proximally and more sparse distally, up to 6 mm. Branching occurs in broad zones. Stipes parallel and closely spaced, with interstipe spaces of 0.20-0.30 mm, similar to the lateral width, resulting in a stipe spacing of about 20 in 10 mm. Autothecal spacing difficult to discern in this dorsoventral view, but may be around 20 in 10 mm. Nature of autothecal apertures cannot be seen, except in one area where they appear to be denticulate or spinose. Bithecae not detected. Remarks This is a highly unusual and distinctive species because of the huge number of dissepiments. Bulman (1928, table II) gave only two species of Silurian dictyonemids with as many as 20 dissepiments in 10 mm (and none with this frequency in the Ordovician species; Bulman 1928, table I). Of Australian dictyonemids, Rickards and Wright (1997) and Rickards et al. (2003), for example, only once have dissepimental spacings as high as 30 in 10 mm been recorded, and that in some specimens of Dictyonema delicatulum barnbyensis from the middle to upper Ludlow; a few other Australian species have as many as 20 in 10 mm. Dictyonema paululum australis Rickards et al., 2003 is similar in having conspicuous dissepiments, but their spacing and that of the stipes is quite different. None of Boucek’s (1957) dictyonemids has high dissepimental spacings. Callograptus J. Hall, 1865 Type species Callograptus elegans J. Hall, 1865, by original designation. Callograptus bridgecreekensis Rickards et al., 2003 Figures 1d, 3b Synonymy 2003 Callograptus bridgecreekensis n. sp.; Rickards et al., p. 319, figs 14A, 15A-B. Material AM F123403. Description These 13 or so stipes are towards the distal end of a moderately-sized (8 mm x 5 mm) piece of rhabdosome; lateral stipe width of 0.50 mm most 13 EARLY SILURIAN GRAPTOLITES FROM NEAR FORBES proximally, and 0.20 mm at distal ends of stipes. Branching irregular, stipe spacing over 20 in 10 mm. No dissepiments. Autothecae not detected in this wholly dorsoventral view, but traces of bithecal tubes apparent. Callograptus rigbyae Rickards et al., 2003 Figures le, 6c, 7d-e Synonymy 2003 Callograptus rigbyae n. sp. Rickards et al., p. 319, figs 14B-C. Material Four almost complete colonies, AM F123404-6, 123414, plus AM F123419-123421. Description Fan-shaped or discoidal colony about 10 mm across, developed from a small holdfast. Up to 6 branching zones may occur in this short distance giving numerous peripheral stipes. Rare anastomosis of stipes. Interstipe spacing 0.50 mm; stipe spacing ca. 16 in 10 mm, lateral stipe width 0.20-0.30 mm. Autothecae spaced at 20 in 10 mm, and autothecal apertures bear a ventral spine up to 0.50 mm long. Dissepiments rare, and extremely fine. Bithecae occur, but their nature is unclear. Remarks The original specimens from Bridge Creek (Rickards et al. 2003, p. 319) were two colonies preserved in plan view. Two Cotton Hill specimens (Figs 7d-e) are more in profile. One (AM F123404a- b: Fig. 7d) shows the autothecae best and a short spine can be clearly seen. Bithecae were not detected in the original material. Callograptus sp. aff. C. ulahensis Rickards et al., 2003 Figures 3c, 5f Synonymy aff. 2003 Callograptus ulahensis n. sp.; Rickards et al., pp. 319-20, figs 16A, 17A. Material A small fragment of rhabdosome, AM F 123407, comprising nine stipes. Description The initial two parallel stipes branch after 3 mm, but thereafter branch at 1-1.5 mm intervals resulting in short, parallel stipes with lateral width 74 of 0.20 mm. Interstipe spaces ca. 0.50 mm, and stipe spacing ca. 20 in 10 mm. Autothecal spacing 20 in 10 mm; dorsoventral width may be 0.40-0.50 mm and thecal aperture may be denticulate. No dissepiments present. Remarks This specimen adds a little to the original description which was based upon two specimens (AM F114760 and 114780) from locality BF15, some 100 m S of the junction of Four Mile Creek and its tributary Bridge Creek (Rickards et al. 2003). The autothecae are not so clear in the Cotton Hill Quarry specimen, but the disposition of the stipes is more apparent. Family Stelechocladiidae Chapman et al., 1993 Stelechocladia Pocta, 1894 Type species — Stelechocladia subfruticosa Poéta, 1894, subsequently designated by Boucek (1957). Stelechocladia sp. cf. S. praeattenuata Rickards et al., 2003 Figure 5b Synonymy cf. 2003 Stelechocladia praeattenuata Nn. sp.; Rickards et al., p. 322, figs 17B, 19A-B. Material AM F123408a-b, and AM F123422-3. Description AM F 123408 is the distal end of a stelechocladiid with stipes spaced at 16 in 10 mm, some apparently laterally derived from nearby dominant stipes. Lateral stipe width from 0.20-0.40 mm, the more robust stipes being more proximal. Branching, where it occurs, is almost every mm, but long, unbranched portions also occur. Autothecae not seen. Remarks This form is almost certainly referable to S. praeattenuata, having the typical combination of dichotomous and “lateral” branching as well as the dimension of a distal part of that species’ rhabdosome. Lack of autothecal presentation, however, urges caution. Proc. Linn. Soc. N.S.W., 130, 2009 R.B. RICKARDS, A.J. WRIGHT AND G. THOMAS Family Acanthograptidae Bulman, 1938 Acanthograptus Spencer, 1878 Type species Acanthograptus granti Spencer, 1878, by original designation. Acanthograptus praedeckeri Rickards et al., 2003 Figures 2c-d, 5c, 7f Synonymy 2003 Acanthograptus praedeckeri n. sp.; Rickards et al., pp. 322-5, figs 17C-D, 19C, 20A- (not fig. 18A). 2003 Dictyonema warrisi; Rickards et al., fig. 18A (mislabelled). Material Three specimens, including one almost entire, small rhabdosome (AM F 123409, Fig. 7f): AM F123409-11. Description Twigs arranged at 8-16 in 10 mm, each 0.70- 1.00 mm long and comprising two or more thecae. Main stipes 0.40-0.50 mm wide laterally, and their ramifications fill all the space available to form a flabellate or fan-shaped colony. Branching occurs every 0.50-2.0 mm, usually 1.00-1.50 mm. Autothecal tubes 0.10 mm wide and do not seem to expand towards apertures. Bithecae may be not much smaller and may open near bases of twigs or on main stipe. Remarks The caption for Rickards et al. (2003, fig. 18A) wrongly states that the illustrated species is Dictyonema warrisi, really being Acanthograptus praedeckeri. Class Rhabdopleurina Fowler, 1892 Family Rhadopleuridae Harmer, 1905 Rhabdopleura Allman, 1869 Type species R. normani Allman, 1869. ?Rhabdopleura sp. Figures 5d-e Material AM F1123412-3; the latter has a fragment of Callograptus rigbyae on the reverse side (AM F123414). Proc. Linn. Soc. N.S.W., 130, 2009 Description The larger specimen (AM F123412: Fig. 5d) appears to have a basal thecorhiza from which arise about nine tubes with a diameter of 0.15-0.20 mm. Tubes distally less sclerotised. Suggestion of growth lines in places, especially on AM F123413 (Fig. 5e). In thecorhizal portion there are probably pyritised (non-goethitised) stolons and possibly also attached encysted zooids. Distal parts of tubes (coenecia) unoccupied and may represent free-standing parts of tubes. AM F123413 may also have pyritised stolons and zooidal remains. Remarks This form does not resemble tuboids such as Galeograptus and Cyclograptus which we have previously recorded from Australia (Rickards et al. 1995, 2003). Were it not for the uncertainty about the growth lines we would refer this to Rhabdopleura with more confidence. Rhabdopleura has not previously been recorded in Australian strata. ACKNOWLEDGEMENTS. RBR acknowledges support from Emmanuel College; the Department of Earth Sciences, University of Cambridge; and the Royal Society of London. AJW acknowledges with much gratitude the award of the Linnean Macleay Fellowship during 2006-7-8 by the Linnean Society of New South Wales, and the support of the School of Earth and Environmental Sciences, University of Wollongong. We are particularly grateful to Dudley Simon (Department of Earth Sciences, Cambridge University) for his skilful assistance in taking all the photographs and processing them. We also thank the reviewers for comments which considerably improved the original manuscript. REFERENCES Allman, G.T. (1869). On Rhabdopleura, a new genus of Polyzoa. Proceedings of the Royal Society of Edinburgh 6, 438-440. Barrande, J. (1850). Graptolites de Bohéme. Extrait du Systéme Silurien de la Bohéme. 74 pp. Published by the author, Prague. Bouéek, B. (1957). The dendroid graptolites of the Silurian of Bohemia. Sbornik Ustredniho Ustavu Geologického 23, 294 pp, Prague. Bronn, H.G. (1849). Index Palaeontologicus B, Enumerator Palaeontologicus. 980 pp. Stuttgart, E., Schweizerbart’sche. TD EARLY SILURIAN GRAPTOLITES FROM NEAR FORBES Bulman, O.M.B. (1928). In 1927-67. A Monograph of the British Dendroid Graptolites. Palaeontographical Society Monographs, i-ixiv, pp. 1- 97. Bulman, O.M.B. (1938). Graptolithina. In Handbuch der Paldozoologie, in O.H. Schindewolf (ed.), 2D, pp. 1-92. Chapman, A.J., Rickards, R.B. and Grayson, R. (1993). The Carboniferous dendroid graptolites of Britain and Ireland. Proceedings of the Yorkshire Geological Society 49, 295-319. Edgecombe, G.D. and Sherwin, L. (2001). Early Silurian (Llandovery) trilobites from Cotton Hill, near Forbes, New South Wales. Alcheringa 25, 87-105. Fowler, G.H. (1892). The morphology of Rhabdopleura allmani. In R. Leukarts (ed.) Festschrift cum Toten Gebiirstag, W.S. Engelmann, Leipzig. Frech, F. (1897). Lethaea geognostica; Theil |, Lethaea palaeozoica, | Band, Graptolithiden (Leif. 1-2 by F. Roemer; Lief. 3 by F. Frech), 544-684. Stuttgart. E. Schweitzerbart’sche. Hall, J. (1843). Geology of New York, IV, Survey of the Fourth Geological District. Natural History of New York. Geology of New York, 156 pp. Hall, J. (1851). New genera of fossil corals. American Journal of Science 11, 398-401. Hall, J. (1858). Descriptions of Canadian Graptolites. Geological Survey of Canada, Report for 1857, \\\- 145. Hall, J. (1862). New species of fossils from the investigations of the Survey. Wisconsin Geological Survey, Report for 1861, 1-18. Hall, J. (1865). Graptolites of the Quebec Group. Canadian Organic Remains, Geological Survey of Canada 2(i-iv), 151 pp. Harmer, S.F. (1905). The Pterobranchia of the Siboga Expedition, with an account of their species. Uitkomsten op zoologisch, botanisch, oceanographisch en geologisch gebied verzameld in Nederlandsch Oost-Indié 1899-1900 aan boord H.M. Siboga 12, Monographie 26, 132 pp. Leiden. Lankester, E.R. (1877). Notes on the embryology and classification of the Animal Kingdom: comprising a revision of speculations relative to the origin and significance of germ layers. Quarterly Journal of Microscopical Science 17, 399-454. Lapworth, C. (1881). On the Rhabdophora (Hopk.) or dendroid graptolites collected by Prof. Keeping in the Llandovery rocks of Mid Wales. Quarterly Journal of the Geological Society of London 37, 171-177. Loydell, D.K., Storch, P. and Melchin, M.J. (1993). Taxonomy, evolution and biostratigraphic importance of the Llandovery graptolite Spirograptus. Palaeontology 36, 909-926. Miller, S.A. (1889). North American geology and paleontology. Western Methodist Book Concern, Cincinnati, Ohio, 664 pp. Nicholson, H.A. (1872). Monograph of British graptolites, i-x, 133 pp. Blackwood and Sons, Edinburgh, London. 76 Packham, G.H. (1967). The occurrence of shelly Ordovician strata near Forbes, New South Wales. Australian Journal of Science 30, 106-7. Pocta, P. (1894). Systéme Silurien du Centre de la Bohéme, 8, pt. 1, Bryozoaires, Hydrozoaires et parte des Anthozaires. Prague, 230 pp. Prout, A.H. (1851). Description of a new graptolite found in the Lower Silurian rocks near the Falls of St. Croix River. American Journal of Science 11, 187-191. Rickards, R.B., Chapman, A.J., Wright, A.J. and Packham, G.H. (2003). Dendroid and tuboid Graptolites from the Llandovery (Silurian) of the Four Mile Creek Area, New South Wales. Records of the Australian Museum 55, 305-330. Rickards, R.B. and Durman, P.N. (2006). Evolution of the earliest graptolites and other hemichordates. In Studies in Palaeozoic Palaeontology, M.G. Bassett (ed.). National Museum of Wales, Geological Series 25, 5-92. Rickards, R.B., Hamedi, M.A. and Wright, A.J. (2001). New assemblages of graptolites, rhabdopleuran hemichordates and chitinous hydroids from the Arenig (Ordovician) of the Banestan area, Iran. Alcheringa 25, 169-190. Rickards, R.B., Packham, G.H., Wright, A.J. and Williamson, P.L. (1995). Wenlock and Ludlow graptolite faunas and biostratigraphy of the Quarry Creek district, New South Wales. Association of Australasian Palaeontologists 17, 68 pp. Rickards, R.B. and Wright, A. (1997). Graptolites of the Barnby Hills Shale (Silurian, Ludlow), New South Wales, Australia. Proceedings of the Yorkshire Geological Society 51, 209-227. Rickards, R.B., Wright, A.J. Morgan, E.J. and Farrell, J.R. (2005). Silurian graptolites from the Barnby Hills Shale and Hanover Formation, New South Wales, Australia. Proceedings of the Linnean Society of New South Wales 126, 153-169. Sherwin, L. (1970). Preliminary results on studies of graptolites from the Forbes district, New South Wales. Records of the Geological Survey of New South Wales 12, 75-6. Sherwin, L. (1973). Stratigraphy of the Forbes - Bogan Gate district. Records of the Geological Survey of New South Wales 15, 47-101. Sherwin, L. (1974). Llandovery graptolites from the Forbes district, New South Wales. Special Papers in Palaeontology 13, 149-175. Sherwin, L. (1976). The Secrets section through the Cotton Beds north of Parkes. Geological Survey of New South Wales, Quarterly Notes 24, 6-10. Spencer, J.W. (1878). Graptolites of the Niagara Formation. Canadian Naturalist 8, 457-463. Proc. Linn. Soc. N.S.W., 130, 2009 A Holocene History of the Vegetation of the Blue Mountains, New South Wales JANE M. CHALSON! AND HELENE A. MARTIN? '46 Kilmarnock St. Engadine, N.S.W. 2233 * School of Biological, Environmental and Earth Sciences, University of New South Wales, Sydney Australia 2052 (h.martin@unsw.edu.au) Chalson, J.M. and Martin, H.A. (2009). A Holocene history of the vegetation of the Blue mountains, New South Wales. Proceedings of the Linnean Society of New South Wales 130, 77-109. The Greater Blue Mountains Area has been inscribed on the World Heritage list for its exceptionally diverse Eucalyptus communities. Hanging swamps in this region, listed as ‘vulnerable ecological communities’, accumulate sediments that contain the palaeoenvironmental record. Seven of these swamps have been studied, revealing a history of the vegetation, climate and fire regimes. Palynological analysis of each swamp reveals a history of the surrounding vegetation. There are similarities and parallel changes between some of the swamps allowing generalities about the climate of the Holocene to be made. In the early Holocene, about eleven to nine thousand years ago (11-9 ka), the vegetation was more wooded and the climate was probably somewhat warmer and wetter. By the mid Holocene about 6-4 ka, trees were less dominant in the vegetation suggesting that the climate was probably drier. By 3-2 ka, wooded vegetation had mostly returned, and after 2 ka, Baeckea, Leptospermum, Kunzea and Melaleuca species increased somewhat, with further increases in European settlement time, possibly reflecting a reduction or thinning of the wooded canopy. Charcoal analysis of the accumulated sediments suggest that there was more fire in the early Holocene when trees increased the biomass. There was less fire through the mid Holocene when the biomass was lower, but it increased with the return to more wooded vegetation in the late Holocene. In particular, the woody shrubs of Baeckea, Leptospermum, Kunzea and Melaleuca increased with an increase in charcoal, probably because these shrubs benefit from a more open canopy, but they also grew on the swamps hence could deposit charcoal directly into the sediments. Charcoal values are particularly high after European settlement. It is possible that the disruption of Aboriginal burning practices allowed the increased growth of woody shrubs and hence a much greater fuel load. Manuscript received 21 May 2008, accepted for publication 17 December 2008. KEY WORDS: Blue Mountains, Climate change, Fire history, Palynology, Vegetation history. INTRODUCTION The Greater Blue Mountains Area was inscribed on the World Heritage List in December 2000. The Blue Mountains are a deeply incised sandstone plateau rising to over 1,300 m at its highest point. This plateau is thought to have enabled the survival of a rich diversity of plant and animal life by providing a refuge from climatic changes during the recent geological history. It is particularly noted for its wide representation of habitats, from wet and dry sclerophyll, mallee heathlands, as well as localised swamps, wetlands and grassland. Ninety one species of eucalypts are found in the Greater Blue Mountains Area and twelve of these are believed to occur only in the Sydney sandstone region (Australian Government, Department of the Environment and Water Resources, 2007a). The area has been described as a natural laboratory for studying the evolution of the eucalypts (Australian Government, Department of the Environment and Water Resources, 2007a). The steep terrain and sharp environmental gradients have allowed for major evolutionary change in some taxa, resulting in exceptional biodiversity, particularly within the eucalypt communities that dominate the place. Importantly, the evolutionary processes underpinning this diversity are believed to be ongoing, resulting in an evolutionary ‘laboratory’ that is exceptional in the world (Australian Government, Department of the Environment and Water Resources, 2007a). HOLOCENE HISTORY OF BLUE MOUNTAINS VEGETATION Peat formation on sandstone, the substrateofmost distribution and demonstrable threat has meant that of the Blue Mountains, is very unusual. The hanging _ these hanging swamps are now listed as ‘vulnerable swamps of the Blue Mountains are especially notable | ecological communities’ under the NSW Threatened and have lower sediment loads and accumulate Species Conservation Act of 1995 ((Australian organic matter more slowly than valley swamps and Government, Department of the Environment and swamps along watercourses. They are also easily | Water Resources, 2007b; Sullivan, 2007) eroded with any disturbance. The small geographic Seven swamps in an altitudinal sequence in | QLD | + Swamp, this study GN Other study sites N Ww 33° 00’ —- Be i pi Kings Waterhole Stay is QH2__— Urban areas Area Wee es sydney ers Road = IE ‘S0km Scale S cer a ke a 15 km = & eo S» 33° 15’ — x eee, * Newnes Sw. ihe fe Mountain : *%, Lithgow x Gooches Crater 33° 30' : § es Bell Se aaa aelarmss cee, *., ae 6, ce - ag i Burralow Cr.Sw a. Mt. Victoria ( __ Blackheath @ ee Katoomba : Lawson 7... Penrith BS e. ae wt {x Lakes Sw. : % ce Warrimoo Oval Sw. BSS Pre at loot been a 5 “‘Penmith SS Kings Tableland Sw. Glenbrook "77" coxs Jenolan R. n Notts Sw. \ LAS Warragamba Dam 7 T 7 I 150° 00’ 150° 15’ 150° 30’ 150° 45’ Figure 1. Locality map. 78 Proc. Linn. Soc. N.S.W., 130, 2009 J.M. CHALSON AND H.A. MARTIN the Blue Mountains (Fig. 1) were chosen for a palynological study and are described in Chalson and Martin (this volume). A method to identify Eucalyptus pollen to species was developed (Chalson and Martin, 1995) with the aim of revealing the history of the eucalypt communities of the region. At the beginning of the Holocene, 10,000 years ago, the climate was approaching that of today, but there have been changes through the Holocene (Allan and Lindsay, 1998). The history of the Holocene is thus the history of vegetation very like that of today. THE ENVIRONMENT Geology and geomorphology The Blue Mountains consist of a deeply dissected plateau rising from the Cumberland Plain in the east, along the Lapstone Monocline. Elevation is about 30 m in the east to over 1,000 m in the west. The sedimentary rock units are Triassic in age and curve upwards, from east to west, towards the edge of the Sydney Basin. In the east, Wianamatta Shale outcrops along the side of the Lapstone Monocline. West of the Monocline, the underlying Hawkesbury Sandstone Formation outcrops and further west, underlying the Hawkesbury Sandstone, the Grose Sub-Group of the Narrabeen Group outcrops. The Grose Sub-Group is divided into a number of formations and the ones encountered in this study are as follows: The Banks Wall Sandstone Formation, within which is found the Wentworth Falls Claystone Member, and the basal Burra-Moko Head Sandstone Formation, which is the most prominent cliff-forming unit in the Blue Mountains (Bembrick, 1980). The plateau surface is undulating with small creeks forming upland valleys. In areas where Hawkesbury Sandstone is the underlying rock type, the upland valleys progressively increase in gradient as they incise below the plateau surface and develop steeply inclined V-shaped gorges with only minor benching in the valley sides. To the west, where the Banks Wall Sandstone formation is the underlying rock type, the valley sides and floors slope gently and the streams do not incise but flow across a series of swamps and sandy peat deposits. Eventually, the streams cut through a sandstone layer into claystone or shale when a nickpoint (often a waterfall) is formed (Langford-Smith, 1976). The development of the swamps in these two areas varies enormously. The eastern region supports few swamps which are usually associated with large streams that have a central channel and flowing water. In the western region, there are more swamps and they are developed in broad shallow valleys with no Proc. Linn. Soc. N.S.W., 130, 2009 marked central stream but rather experience a general slow flow of water across the whole area (Langford- Smith, 1976). The climate Maximum temperatures in the Blue Mountains relate strongly to altitude. Average January maxima are highest at the lower altitudes, 29 °C at Richmond and lowest at the higher altitudes, 23 °C at Mt. Victoria. Average minimum temperatures generally decrease from east to west. The July minima range from 3.4 °C at Richmond to —0.8 °C at Lithgow (Table 1). Temperatures as low as —3 °C have been recorded from Katoomba (BoM, 2006; Bureau of Meteorology, 1979). Rainfall patterns relate to elevation and distance from the coast. The average annual rainfall increases from 806 mm at Richmond to 1424 mm at Newnes (Table 1). The driest months are usually July to September and the wettest are December to March (BoM, 2006; Bureau of Meteorology, 1979). Winds from the west or northwest dominate all the year, although there are significant easterly and northeasterly winds during the summer months of November to April. Fogs frequently occur on the higher Blue Mountains, with Katoomba and Mt. Victoria recording an average of 55 and 90 fog days per year, respectively (BoM, 2006). Frosts occur on 35 to 40 days of the year, mostly between April and November. Snow falls most frequently in July and August: Katoomba and Mt. Victoria have and average of 3 and 10 snow days per year, respectively (Bureau of Meteorology, 1979). Soils The quartz-rich sandstones in the area are low in most nutrients, and thus soil and alluvium derived from sandstones are low in nutrients. The soils are mainly lithosols and yellow podzolics with small areas of red and lateritic podzolic soils and sandy alluvial soils in the valleys. Most of the soils are moderately acidic, with pH values of 4.5 to 5. In rugged terrain, rock commonly lies near or at the surface. The soil fertility in the valleys may be higher because of the accumulation of organic matter (Chalson, 1991) Vegetation The vegetation is almost entirely dry sclerophyll woodland and open forest, the “Sydney Sandstone Complex’ (Keith and Benson,1988) with localised swamps in the valleys. There are small patches of tall open forest or wet sclerophyll in specially favourable habitats, such as protected gorges. Heathlands are found in the harshest environments. 79 HOLOCENE HISTORY OF BLUE MOUNTAINS VEGETATION Table 1. Climatic Averages. Stations are arranged according to altitude. Mean max. temp, hottest month, Station and altitude (m) (Jan.) °C 'Richmond, 19-20 + 29.5 *Penrith, 27 = *Springwood ~400 - *Kurrajong Heights, ~550 - *Lawson, 715 = *Wentworth Falls, ~900 Lithgow (Birdwood St.), 950 DNS Katoomba, 1030 23.1 Mt Victoria, 1064 23.0 *Blackheath PO, 1065 - Lithgow (Newnes Forest 332 Centre), 1050 f Mean Mean min. temp, annual coldest month, x rainfall, (June or July) °C mm 3.4 806 - 786 - 1076 - 1253 : 1260 1409 0.7 860 OES 1398 ley 1061 = 1145 -0.8 1072 1 From BoM (2007). 2 From Bureau of Meteorology (1979) 3 Average of Richmond RAAF and Richmond UWS Hawkesbury Open forest with Angophora costata, Eucalyptus piperita, E. agglomerata and Syncarpia glomulifera dominant is found in sheltered gullies with moist, well-drained soils on the Hawkesbury and Narrabeen Group sandstones. The understorey includes small trees of Allocasuarina torulosa and Acacia elata, with shubs of Hakea dactyloides, Pultenaea flexilis and Dodonaea triquetra. Tall open forest is restricted to the more sheltered gorges and is dominated by E. deanei with Syncarpia glomulifera, Acacia elata, Ceratopetalum apetalum, Callicoma serratifolia and Angophora floribunda. There is a distinctive riparian scrub of Tristaniopsis laurina and Backhousia myrtifolia along the larger water courses (Keith and Benson,1988), Woodland and low woodland with Corymbia gummifera, Eucalyptus sclerophylla and E. oblongata dominant is widespread on ridges and open slopes on shallow, well-drained soils of the Hawkesbury and Narrrabeen Group sandstones. E. punctata, E. piperita and Angophora costata may be present in the more sheltered sites. E. sclerophylla is particularly common on damper soils. The understorey is rich in shrubs of the Proteaceae, Myrtaceae and Fabaceae (Keith and Benson, 1988). There are other woodlands: the “Tablelands Grassy Woodland Complex’ with Eucalyptus dives, E. mannifera, E. eugenioides, E. pauciflora, E. rubida, E. aggregata and E. stellulata the common species. 80 The ‘Snow Gum Woodland’ has FE. pauciflora, E. dalrympleana, E. rubida and E. stellulata dominant (Keith and Benson, 1988). Open heath communities have Eucalyptus stricta, Allocasuarina nana and Leptospermum trinervium, Phyllota squarrosa, Eriostemon obovalis, Epacris reclinata, Dracophyllum secundatum and Gleichenia rupestris dominant. Phyllota squarrosa and Eriostemon obovalis are common in montane heaths whereas Phyllota phylicoides and Eriostemon hispidula are common on the Lower Blue Mountains heath. Many other smaller shrubs are found in these heath communities (Keith and Benson,1988). Closed heath or ‘Newnes Shrub Swamps’ have Leptospermum lanigerum, Baeckea linifolia, Grevillea acanthifolia and Xyris ustulata dominant. They are found in shallow valleys above 1,000 m elevation in swamps, with poorly drained, acid and sandy peat soils. There is a ground cover of sedges including Baloskion australe, Empodisma minus, Lepyrodia_ scariosa, L. anathria, Lepidosperma limicola and small shrubs (Keith and Benson,1988). Closed sedgeland, the ‘Blue Mountains Sedge Swamps’, have Gymnoschoenus sphaerocephalus, Lepidosperma limicola, Xyris ustulata and Baeckea linifolia dominant. These sedge swamps are found at lower altitudes than the closed heath swamps and occupy steep-sided basins (the ‘hanging swamps’). They are intermittently waterlogged and have shallow Proc. Linn. Soc. N.S.W., 130, 2009 J.M. CHALSON AND H.A. MARTIN sandy soils. Many sclerophyllous shrubs form an open heath (Keith and Benson,1988). For a full description of the specific vegetation found at each site, see Chalson and Martin (this volume). Human Occupation The Blue Mountains, especially the lower part, was highly favourable to the hunter-gatherer, (Stockton,1993a). Movement was relatively easy on the ridges, water was not scarce while flora and fauna suitable for food were both plentiful and varied. The rivers were also a source of rock types used for tool making. Campsites with an abundance of worked stone were particularly common in the Lower Blue Mountains. In the Upper Mountains, there were fewer campsites than in the Lower Mountains, but their concentration of flaked stone showed that they have been equally well used. The Central Mountains reveal many rockshelter sites where there were fewer stone artifacts than the Upper and Lower Mountains. However, there was a high concentration of rock art, engravings, paintings and axe grinding grooves. This suggests that the Upper and Lower Mountains were used for survival but the Central Mountains were more of religious and ritual significance (Stockton, 1993a). It is generally presumed that the climate in the Blue Mountains was too severe for year-round occupation during the ice age. However, protected sites such as the rock shelters would have been livable, especially if protected from the bitter westerly winds. (Stockton, 1993b). The oldest signs of occupation in the Blue Mountains were found at Kings Tableland, Wentworth Falls with the oldest date of 22,240 years BP. Walls Cave at Blackheath and Lyre Bird Dell, Leura both yielded dates of more than 12,000 years BP. There were other sites, e.g. Hazelbrook, to 7,200 years BP, Springwood Creek Rock Shelter, from 8,500 years BP up to European times and open sites, e.g. Jamison Creek. Evidence from the Nepean River, at the foot of the Blue Mountains suggests human occupation could go back to 40,000 years BP. In all, there were over 700 Aboriginal sites in the Blue Mountains (Stockton, 1993b; Attenbrow, 2002). With the coming of Europeans, both Europeans and Aborigines avoided each other and early travelers in the Mountains rarely saw any Aborigines. Settlers followed the first crossing of the Mountains in 1813 by Blaxland, Lawson and Wentworth (Breckell, 1993) After some skirmishes about the land the settlers had taken, Aborigines and Europeans co-existed, though not without racist incidents (Smith, 1993). Proc. Linn. Soc. N.S.W., 130, 2009 METHODS Seven swamps in an altitudinal sequence were chosen for study and they are described in Chalson and Martin (this volume). A study of the pollen in surface samples from swamps (Chalson and Martin this volume) provides insights that assist in the interpretation of the pollen spectra from the sediments. The description of the vegetation at each site is also presented in Chalson and Martin (this volume). The swamps were systematically probed to identify the area where accumulating sediments were the deepest, using a Russian D-corer (Birks and Birks, 1980). The sediments and stratigraphy were described using the terminology of Birks and Birks (1980) Samples for radiocarbon dating were taken from a pit where possible, otherwise with repeated use of the D-corer until sufficient sediment was acquired. The standard radiocarbon dates were calibrated using the CalPal (Version March 2007) program. Samples of sediment were taken from the core every 10 cm, or where it was thought there could be a critical change, every 5 cm. For pollen preparations, the core sediments were spiked with Alnus of a known concentration, treated with hydrochloric and hydrofluoric acids to remove siliceous material (Birks and Birks, 1980), oxidised with Schultz solution (a saturated solution of potassium perchlorate in nitric acid), cleared in 10% potassium carbonate and the residue was mounted in glycerine jelly (Brown, 1960). Reference pollen was treated with standard acetolysis (Moore et al., 1991) and also mounted in glycerine jelly. Pollen was identified by comparing grains from the core with a collection of reference pollen. Special attention was paid to pollen of the family Myrtaceae which may be identified to species following the method in Chalson and Martin (1995). Pollen was counted along transects across the slides and tests showed that a count of more than 140 grains adequately sampled the residues. The counts were presented as percentages of the total count and pollen concentrations were calculated for the most abundant pollen groups. Percentages are relative and a change in a single pollen group will affect percentages of all the other groups, but presenting both percentages and concentrations will reveal fluctuations in individual pollen groups. The abundance of charcoal retained on a 150 um sieve, as part of the palynological preparation, was estimated subjectively on a scale of 0 to 8. Counts of microscopic charcoal for a swamp at Kings Tableland showed that the two methods gave similar results, although the microscopic charcoal was more variable (Chalson, 1991). 81 HOLOCENE HISTORY OF BLUE MOUNTAINS VEGETATION RESULTS Burralow Creek Swamp Burralow Creek Swamp, at 33° 32’S, 150° 36’ 38”E and 310-330 m altitude, is situated in a narrow V-shaped valley and follows the course of the creek for some 3.5 km. The substrate is Hawkesbury Sandstone, but Wiananatta Shale outcrops on the surrounding ridge-tops. The upper reaches of Burralow Creek drain urban areas and farmland areas. An isolated farm adjacent to the swamp was incorporated into the Blue Mountains National Park. Weed growth from this farm is confined to a small area and has not spread into the adjacent bushland. Stratigraphy: Sediments were recovered to a depth of 310 cm. Clayey peat was found down to 10 cm, humic clay at 15-50 cm and humic sandy clay at 60-70 cm. Sand was encountered at 80-260 cm and clay/sand at 260-310 cm. The radiocarbon dates are presented in Table 2. Swamp vegetation and surface pollen: Species of Kunzea and Leptospermum were dominant on the swamp but Restionaceae, Cyperaceae and Selaginella Species were also present (Chalson and Martin, this volume). Surface sample pollen from the swamp (Chalson and Martin, this volume) showed appreciable Leptospermum/Baeckea and a considerable amount of Restionaceae or Cyperaceae in some samples. The fern spore content was low. The pollen record: The pollen spectra from the sediments is presented in Figs 2A, 2B and has been divided into the following zones: 310 to140 cm, no pollen recovered. Zone E, 130 cm, age ? > 1,200 cal yr BP (see Fig. 3 for estimated ages). Angophora floribunda, Eucalyptus spp. and possibly Casuarinaceae pollen were the most abundant of the possible arboreal groups. There was a moderate representation of Poaceae and Selaginella (Fig 2A) and other shrubs and herbs were present in low frequencies (Fig 2B). 120-110 cm, no pollen recovered. Zone D, 100-90 cm, age c. 1,200 —1,000 cal yr BP. . This zone had a very high proportion of Selaginella spores and low proportions of everything else, including tree pollen. The pollen concentrations showed a similar pattern to that of the percentages which revealed a change in the whole pollen spectrum, not only reflecting the addition of a large number of Sellaginella spores to spectra otherwise like that in zone E. Zone C, 80-60 cm, age c. 1,000-800 cal yr BP The Sellaginella content had decreased considerably when compared with the zone brlow, There was a high proportion of Casuarinaceae and Myrtaceae, including Eucalyptus species and the Poaceae content was low. Zone B, 50-20 cm, age c.8700-250 cal yr BP. The Casuarinaceae content had increased and was the highest for the profile. Eucalyptus species and Angophora floribunda were well represented and Leptospemum juniperinum was present in low frequencies. There was a moderate content of Poaceae and Cyperaceae, with a diversity of fern spores. Sellaginella content was minimal. Zone A, 15-0 cm, age c. 250-present, cal yr BP. European Pinus was found in this Zone and there was a high content of L. juniperinum. There was some change in the Eucalyptus species, Casuarinaceae declined. and the Poaceae content was moderate, when compared with the zone below. Charcoal content was low to moderate through most of the profile, with a somewhat higher content at the base of Zone A, the zone of European influence. Table 2. Radiocarbon ages for Burralow Creek Swamp Depth (cm) Material dated Laboratory no. 15-20 Humic clay SUA-2607 50-60 Humic sandy clay SUA-2608 80-90 Sand SUA-2609 95-105 Sand SUA-2610 125-135 Sand SUA-2611 82 Radiocarbon years _—_ Calibrated age (cal yr (yr BP) BP.) 250 +50 3484130 830+60 848+70 1,070450 1068+50 820+50 818450 660455 688450 Proc. Linn. Soc. N.S.W., 130, 2009 J.M. CHALSON AND H.A. MARTIN ‘xipusddy aas ‘uoneja80 oy} ul od4} uatfod ay} Jo vd.41n0s a[quqoid 104 ‘vajoods uatjod dureMg Yae1D MOVING “GZ “WZ SoINSI as soysuiesB ,OL X OF pajunos saiods pue %OZ FS Be feos te Ril feasrvaa] = id t J — : ta es "WS Z | pues ie al Wad [~~ se uajjod jeyo{ = wns ualjod je}oL ee - pe OSF889 | | ey ik it = | : | | | ca a eae TRE area st ieee > a= aaa { | | OZL ? 1 ua}iod oN | 001 = al - 08 | 09 (a ees ov besapn cl Pike : = Teves se ry rr Tan] 77) eesuensnenl =] | fad, ~ » O @ : § £ ££ Pf ff fee 2 g < S > & § § § & €é& Cen eee Si eee g 25h 2 y ga ae sy oF “so o1dosso9e | 2 ¥ g 4 é a S. > § iS a $ é & € G< & oe f S Sos see y & # i f se SS 8 . ri) : 7 Ss S é s $ y Ve ® a 4% yy" S ¥ 83 Proc. Linn. Soc. N.S.W., 130, 2009 84 ponunuos Z ainsi HOLOCENE HISTORY OF BLUE MOUNTAINS VEGETATION Depth (cm) OOL lo) Some, tic) i= a =) a 28 nig 2 8 a c i= o oO = = ph ™ a = Q ° o 4 P= @ ago 3 2 = ig) er oO a = oO C2) et io} = ° 3 | oo ne) A = 5 = 3 oO & ® te 9) . 2 = roy = rs a = a rom 5 a @ Proc. Linn. Soc. N.S.W., 130, 2009 J.M. CHALSON AND H.A. MARTIN History of the vegetation: Initially, more than 1,200 cal yr BP, there was a mixed tree cover of Myrtaceous species and possibly Casuarinaceae with a moderate Poaceae understorey. Selaginella, was prominent on the swamp. A period of possibly a reduced tree cover followed, with an expanded swamp area with abundant Selaginella about 1.2-1.0 cal ka. Alternatively, if the swamp area was larger, the trees may have been further away, hence they contributed less pollen to the spectrum. The tree cover increased and Selaginella was much reduced by about 1-0.8 cal ka. At this time, the clay content of the sediments increased, perhaps indicating a less energetic water flow. Casuarinaceae became prominent about 0.8-0.25 cal ka with less Myrtaceae, although a diversity of species was identified. Simultaneously, Se/laginella decreased while Cyperaceae and Poaceae increased. In the European zone, there was some change in Eucalyptus species and a big decline in Casuarinaceae while Leptospermum juniperinum became prominent. Fire was a constant factor in the environment, especially in the early part of the European zone. Warrimoo Oval Swamp Warrimoo Oval Swamp, at 33° 43’ 21.44’S, 150° 36’ 58.35”E and 190-200 m altitude, is situated in a V-shaped valley with a stream flowing through it. The substrate is Hawkesbury Sandstone, but Wiananatta shale outcrops on the surrounding ridge-tops. Substantial urban areas occur within a kilometre from the swamp and weed invasion is considerable. Stratigraphy: Total depth recovered was 250 cm. The top 20 cm was peat, then sandy peat down to 50 cm. A layer of sand was found between 50 and 90 cm, then sandy silt down to 200 cm, then sand down to 250 cm when coring stopped (Fig. 4A). The radiocarbon dates are given in Table 3. Swamp vegetation and surface pollen: Species of Baeckea, Kunzea and Leptospermum were dominant on the swamp. Cyperaceae, Juncaceae and Gleichenia species were also present (Chalson and Martin, this volume). The pollen spectra from the surface samples (Chalson and Martin, this volume) contained appreciable Melaleuca, Baeckea/Leptospermum and Gleichenia species. The pollen record: The pollen spectra from the sediments are shown in Figs 4A, 4B. Zone B, 250-130 cm, c. 4,700-2,200 cal yr BP (for estimated ages, see Fig. 5). Abundant Gleichenia denoted this zone, The Myrtaceae content was low, with some of the pollen identifiable to genus/species. There was a consistent content of Casuarinaceae and Haloragis, and Poaceae was almost entirely absent. Zone A, 120 cm to surface, c. 2,200-present cal yr BP. There was very little Gleichenia,, together with an increase in the Myrtaceae and Casuarinaceae content, when compared with the zone below. The Poaceae, Cyperaceae and Restionaceae content was higher and the pollen flora considerably more diverse when compared the preceding zone. Pinus was found down to a depth of 20 cm, thus denoting the European influence, where Baeckea/Leptospermum species increased and Casuarinaceae decreased. The charcoal content was consistently very low in zone B (4.7-2.2 cal ka) and higher in zone A (2.2 cal ka to present). History of the vegetation: From about 4.7-2.2 cal ka, myrtaceous species and Casuarinaceae dominated open vegetation communities. The swamp supported abundant Gleichenia. About 2.2 cal ka, the tree cover of the dryland vegetation increased, with Eucalyptus spp and Leptospermum spp. becoming more diverse and abundant. Casuarinaceae was also more abundant. Gleichenia declined dramatically, but this change was not accompanied by any visible change in the sediments. Fire appears to have been a rare feature of the environment when Gleichenia was dominant. With the change to a more diverse flora and increase of Leptospemum in the swamp community after 2.2 cal ka, fire was more common, particularly in the Table 3. Radiocarbon ages for Warrimoo Oval Swamp Depth (cm) Material dated Laboratory no. 15-25 Peat SUA-2603 120-130 Sandy silt SUA-2604 160-170 Sandy silt SUA-2605 240-250 Sand SUA-2606 Proc. Linn. Soc. N.S.W., 130, 2009 Radiocarbon years Calibrated age (cal (yr BP) yt BP) 730480 738480 2,190+80 2,248+100 2,880+70 3,088+110 4,060+80 4,668+140 85 HOLOCENE HISTORY OF BLUE MOUNTAINS VEGETATION Se ee ee SPS Seis ° (=) 3 o oS ro) ro} S oO =) o o ' ' b Y n ' 1 oo a mt i ar peice > Peps My q 4 a eRe TET aa ee | H : = "s rs i I S ALL an 5 = : = beers eet | Ses, eee = Con Crap Oo Ye) ~~ ae 1 th = T ieee "Or 35 Be pape 7p See 8] 8 - | oe TPT S hes 9 Bees : of Ly, = “yo, Peo =] 17 o Y i} are iF 1 | i] A] if) i 1] no = Pap * + mop do \} ~ Nagy, eg Yay, Oey Cas, a 4 r {1 5) OS¢., "Pree ‘2 4 1 2 | 1 \ vee bee a oy 7 . So LVS g bro Phy ety ana prong gts Mame MNase 6 2 % 3 area er aes re =) oe \\- — icc ' a sty aT ay a8 2 hp a pr a yg ty 59 1 ET TTT SE ra nag Rig ii i Ming” i Poy : 3 ‘ sNcsonthtaseckiny cles Maar 7 pa) sa a OW Mhoatiet faa naa igg tical Ae | ie jms Cae, 0 : Sip, it , “ag Si = z + 7 team Ae? pe - : ala { *n | r oe Sl Kis I a. | i Af TT Po we q 7 ae he Ero il HI “Tt a wil Ch bios: ee meee) a banca ig. ee Von : To, Ls 4 7 it pera Ae ieee SYS aS SER ator: ie a {— *} | T esl Coa} En A i coat ey, ee, : (a) emg pg egypt 7 o) a a Tape Sen, C856 : “a SS herp aa) ee Pore, mie 3 | rel =z oe ( | | | i - —8 L rs) w 8 ao ah trig Sy a ee tt a = a . ® ae 5 pt i Rote’ OvlFg99'F *xipuoddy 99s ‘uo -8}930A oy} Ul od} UaTjod 34} JO 9d.1NOS s[quqoid 10,7 *e.ajoods uayjod durwas [vaAQ OOWLLILAA “Ap ‘VP Soins 86 Proc. Linn. Soc. N.S.W., 130, 2009 J.M. CHALSON AND H.A. MARTIN ® & rd Sts 2 gf e £ ey 2 § ® © > Fs 2 S 7 2. cians SF 3 # se SF&F EEFEEESB FF SS FE TES EL ES se ST te GS Se GS < 0 Fite dy, Fl

¢ v a f V9 89 Proc. Linn. Soc. N.S.W., 130, 2009 HOLOCENE HISTORY OF BLUE MOUNTAINS VEGETATION ponulyjuod 9 sins Depth (cm) hh — wn So oO ° ° fon) ro) Oo = wie re E 3 ra ny oO z = g a S Es = <4} BA = tS} =) s =~ i“) = =) = o— ) A = e) “< o 8 a gz 8 Fy a 3 o = 2. 3 iS) e. =! 5 os a 90 Proc. Linn. Soc. N.S.W., 130, 2009 J.M. CHALSON AND H.A. MARTIN Table 5. Radiocarbon ages for Ingar Swamp Depth (cm) Material dated 30-40 Humic clay with roots 120-130 Sandy humic clay 140-150 Sandy clay Laboratory no. Radiocarbon years (yr BP.) BETA 20942 105.1+0.8% modern BETA 20943 6,460+100 BETA 20944 6,2204100 Calibrated age (cal. yr BP.) Modern (<43) 7,428+90 7,188+90 unidentifiable. There was moderate Casuarinaceae and Poaceae. The pollen record: The pollen spectra from the sediments is shown in Fig. 8A, 8B and has been zoned thus: Zone C, 150-130 cm, c. 7,000 cal yr BP (see Fig. 9 for estimated ages). Abundant Restionaceae marked this zone. Eucalyptus piperita, other Myrtaceae and Casuarinaceae were prominent and there was a moderate content of Poaceae. Zone B, 120-40 cm, c. 7,000-?2,200 cal yr BP. There was greater diversity here and more of tree/large shrub pollen, viz. E. piperita, Angophora, Melaleuca & & é ae SB hig@ak ef e é & é av & @ & £ f e oO & e — ~— é o B) FS 1 ceded? ¢G |b & LF Gb us §f See ds £ Ss F Q & sé & fou 4 ris se fy : = —— - me pil a 207 x 2 - : ; hoe Modern a 40 60 80 B 100 420 4 ue — 7,428490 —' ike be 7,188+90 “_- ets = waee cont) Ss Peat - Aw Figures 8A 8B. Ingar swamp pollen spectra. For probable source of the pollen type in the vegetation, see Appendix. Scale ram font 20% 10X10" grains/cc Proc. Linn. Soc. N.S.W., 130, 2009 ES Fe E | ~'s\"| Sand Silt | Clay M Roots i HOLOCENE HISTORY OF BLUE MOUNTAINS VEGETATION Restionaceae decreased but Cyperaceae, Selagiella and Gleichenia increased slightly. In the European zone, there was a Slight decline in Casuarinaceae and an increase in the swamp species of Restionaceae and Gleichenia. Fire was relatively rare about 6 cal ka, but increased through time, to a peak in the European period. Kings Tableland Swamp Kings Tableland Swamp, at 33° 45” 47” S, 150° 22’ 43” E and about 780- 790 m altitude, is located in the floor of a steeply sloping small valley off Queen Victoria Creek. The valley floor steepens abruptly below the swamp and a waterfall cascades over a small cliff. The Banks Wall Sandstone Formation underlies the swamp and the Wentworth Falls Claystone outcrops near the base of the swamp. An area of development is found less than 1 km to the west where exotic conifers have been planted in the gardens. Stratigraphy: The core sampled 220 cm of sediments which were peat down to 10 cm, then peaty sand at 15-20 cm, humic sand at 30-40 cm, peaty silt at 50 cm, humic sand at 60-90 cm, clay/sand at 100-120 cm and sand at 130-220 cm. Radiocarbon dates are given in Table 6. 140 Figure 8 continued styphelioides and Casuarinaceae, when compared with the zone below. There was a little more Poaceae but less Restionaceae than in the zone below. Selaginella, and to a lesser extent, Gleichenia, were moderate in the base of the zone. 0 European zone Zone A, 40-0 cm, c. 2,200-0 cal yr BP to modem. Pinus was found down to 20 cm, marking European settlement. The dryland 50 flora was similar to the zone below, but tree species declined with European influence. Restionaceae and Gleichenia were more abundant than in the zone below. 400 There was very little charcoal in the basal zone C, increasing in zone B and reaching a maximum in the European zone A. Zone A SEEaalaeeearasad Ke ; ; Zone C ‘ ; 0 “waz 4 6 History of the vegetation: Before 7 cal ka , Age (k yr) ] Radiocarbon date the vegetation was relatively open, but after @ Calibrated date about 6 cal ka, the tree cover increased, especially Casuarinaceae. On the swamp, Figure 9. Ingar Swamp summary diagram 92 Proc. Linn. Soc. N.S.W., 130, 2009 J.M. CHALSON AND H.A. MARTIN Table 6. Radiocarbon ages for Kings Tablelands Swamp Radiocarbon years (yr Calibrated age BP.) (cal. yr BP.) a 1.045+0.008 x Modern (< 33) modern 1,210+70 1,208+90 2,410+70 2,578+130 9,040+80 10,208+130 Depth (cm) Material dated Laboratory no. 15-20 Peaty sand SUA 2656 50-60 Humic sand SUA 2657 80-90 Humic sand SUA 2658 155-160 Fine sand SUA 2659 The swamp vegetation and surface _ pollen: Leptospermum species were dominant, but Gleichenia and sclerophyllous shrubs were also found on the swamp (Chalson and Martin, this volume). In the surface samples, the Myrtaceae content was low but Casuarinaceae was well represented (Chalson and Martin, this volume). The swamp taxa Restionaceae, Selaginella and Gleichenia were also well represented and the introduced Pinus was abundant. The pollen record: The pollen spectra from the sediments (Figs 10A, 10B) have been zoned thus: Zone C, 200-90 cm, c. ?>12,000-3,800 cal yr B P (see Fig. 11 for estimated ages). The Myrtaceae content was low and Casuarinaceae content moderate (Fig. 10A). Sclerophyllous shrubs and Restionaceae were well represented (Fig. 10B). Gleichenia and other fern spores were moderate. Eucalyptus deanei was found in the basal part of the zone and Banksia in the upper part. Zone B, 80-30 cm, c. 3,800 cal yr BP to modern. This zone had some very high pollen concentrations which mirrored the spectra of the percentages, suggesting that the high concentrations result from slow sediment accumulation rather than the increased input of any one (or more) particular pollen type(s). The Myrtaceae pollen proportion remained low but the Casuarinaceae representation had increased, when compared with the zone below. The proportion of Restionaceae and Gleichenia had decreased, but Cyperaceae and Selagiella had increased, in comparison with the zone below. Sclerophyllous shrubs were also well represented in this zone. Zone A, 0-25 cm, modern. Pinus was found here, delimiting the European zone. The myrtaceous content had increased a little, especially Melaleuca. Casuarinaceae and Restionaceae decreased somewhat but Gleichenia increased considerably, when compared with the zone below. The charcoal content was low to moderate in zones C Proc. Linn. Soc. N.S.W., 130, 2009 and B, and increasing in the modern zone A. History of the vegetation: The dearth of myrtaceous taxa, predominance of Casuarinaceae and the diversity and relative abundance of the shrubby taxa suggests a heathland, given that the two species of Casuarinaceae found in the region today, A//ocasuarina distyla and A. nana, are shrubs/small trees. The swamp flora was dominated by Restionaceae throughout, with Gleichenia becoming prominent in modern times. Myrtaceae remains low until modern times, suggesting the surrounding vegetation remained relatively open. The charcoal content was relatively low until modern times, suggesting less fire activity, or lesser fuel to burn, until European times. Katoomba Swamp Katoomba Swamp, at 33° 43’ 03” S, 150° 19’ 18” E and 950 m altitude, is located in a small, shallow valley which is a tributary of Gordon Creek (Chalson and Martin, this volume). The Banks Wall Sandstone Formation underlies the swamp and the Wentworth Claystone Member outcrops near the base of the swamp, probably impeding drainage. - This swamp is surrounded by urban development. There is evidence of drainage ditches and a sealed road runs across the swamp. Much of it is (or has been) used for yards for light industry and horse paddocks. Housing extends to the edge of the swamp. Stratigraphy: Two cores were necessary to recover sediments spanning the whole of the Holocene. Core 1 consisted of (1) dark greyish brown or dark brown silty clay/humic clay/clay with roots, 0-20 cm, then (2) dark greyish brown, black, or dark grey silty or sandy clay at 25-80 cm, followed by (3) dark grey sand at 85 cm, (4) dark grey clay at 90 cm, (5) dark greyish brown or dark grey sandy or silty clay at 95- 115 cm, (6) dark grey sand at 120 cm and (7) dark grey sandy clay at 125-130 cm. The stratigraphy of core 2 consisted of (1) dark greyish brown, dark grey or dark brown silty clay at 0-30 cm, then (2) dark grey or dark brown sandy clay, 93 HOLOCENE HISTORY OF BLUE MOUNTAINS VEGETATION charcoal ge 1 4 7 a ia) yee = = a ~~ rs a H id ~ = ies, wu Oe we siay gdp s Ofna aa ae aiewip cop wan xgowent att Skasa oh ie ese nee oan 155 x10° 163x107} F B 80 . > THe on tne ge Pn 2 Serer Sree see < scores ie C 270 xi0* | 2,578+130 - 100 lee oe C Eee a 120 = = : : ie Wee . i 140 ee 3 - ; — 160 i - . = : 10,2084130. 180 _ : —— e — Le is a. 7 E | Sand [7] Sit [-=_] Clay boats Figures 10A, 10B. Kings Tableland Swamp pollen spectra. For probable source of the pollen type in the vegetation, see Appendix. Proc. Linn. Soc. N.S.W., 130, 2009 94 J.M. CHALSON AND H.A. MARTIN SED Figure 10 continued 35-40 cm, followed by (3) dark greyish brown sand at 42-48 cm and (4) dark grey clay or sandy clay at 50- 55 cm. Radiocarbon dates are presented in Table 7. The swamp vegetation and surface pollen: The moss Dawsonia, and species of Cyperaceae and Juncaceae were dominant on the swamp. Kunzea and Leptospermum species were also dominant and many sclerophyllous shrubs were found on the edge of the swamp, but the natural vegetation was highly disturbed here (Chalson and Martin, this volume). Poaceae (both native and introduced species) was the dominant pollen type in the surface samples, reflecting the urbanisation and the disturbance at the site. Pinus pollen was also present in appreciable amounts. Total Myrtaceae pollen was moderate and Casuarinaceae pollen was low. The swamp taxa, Restionaceae, Cyperaceae, Selaginella and Proc. Linn. Soc. N.S.W., 130, 2009 Gleichenia were present in low proportions (Chalson and Martin, this volume). The pollen record: Pollen recovery from the cores was good and some very high concentrations were found, especially in the clay (Figs 12A, 12B). The cores were zoned thus: Core 2, Zone D, 55-0 cm, c. 12-11,000 cal yr BP (see Fig. 13 for estimated ages). The Myrtaceae content was low but Eucalyptus oreades and E. pauciflora had been identified. Casuarinaceae and Poaceae representation was moderate and Restionaceae was high (Fig. 12A). Asteraceae/Tubuliflorae and Ericaceae were prominent amongst the herbs and shrubs (Fig. 12B). The charcoal content was moderate throughout. 95 HOLOCENE HISTORY OF BLUE MOUNTAINS VEGETATION 1 nm Nay ie | Ny 1] Zone A |4 ‘a. | £ ios it * at = 255 5 ZoneB i: t= y n | , a | oO | 4 1 1004 OF | 25 | Zone C i * 150+ ae ? re i ” 6 Age (k yr) i Radiocarbon date @ Calibrated date Figure 11. Kings Tableland Swamp summary diagram. Core 1, zone C, 130-75 cm, c. 6,200-4,000 cal_yr BP (for estimated ages, see Fig 13). The Myrtaceae representation was very low, lower than in the zone below, and Eucalyptus species were not recorded from most samples. Casuarinaceae representation was low also, Poaceae was moderate and Restionaceae high, all fairly similar to the zone below. Core 1, zone B, 70-30 cm, c. 3,100-?1,500 cal yr BP. The Myrtaceae content had increased and Eucalyptus oreades was present through the zone, and this was the most notable difference when compared with the zone below. Casuarinaceae abundance was moderate and the Poaceae representation had decreased when compared to the zone below. Restionaceae abundance was a little less than in the zone below, decreasing further towards the top of the zone. Haloragis and Grevillea acanthifolia were prominent amongst the herbs and shrubs. Coren zonesAce25-0ucmcan eo 00Rcalmyme Pato present. Pinus was consistently present, denoting the European zone. Total Myrtaceae and Casuarinaceae Table 7. Radiocarbon ages for Katoomba Swamp Depth (cm) Material dated Laboratory no. Core 1,125-130 Sandy clay Beta 24545 Core 2, 0-5 Silty Clay Beta 24547 Core 2, 50-55 Sandy Clay Beta 24546 96 representation were low, decreasing somewhat from the base, but E. oreades and A. floribunda were found throughout the zone. Poaceae pollen increased markedly from the base of the zone but Restionaceae was very low at the very base, then virtually absent from the rest of the zone. Cyperaceae increased a little and Asteraceae/Liguliflorae was present throughout the zone. The charcoal content was very low in zone C, then low through the rest of the core, with an occasional moderate value. History of the vegetation: There was an open or sparse tree cover about 11-12 cal ka. By 6-5 cal ka, the site appears to have been almost treeless. About 4 cal ka, E. oreades returned to the site which became wooded once again. Restionaceae was dominant on the swamp and Poaceae was moderately common until 3 kyr BP, after which, both declined. In the European zone, Poaceae increased dramatically, no doubt reflecting urbanisation. At the same time Restionaceae decreased and almost vanished from the swamp. E. oreades remained dominant but it Radiocarbon years Calibrated age (yr BP) (cal. yr BP) 5,450+80 6,288+100 11,0304£130 12,998+120 10,570£100 12,558+170 Proc. Linn. Soc. N.S.W., 130, 2009 J.M. CHALSON AND H.A. MARTIN ‘x1pusddy aas ‘u01e}030A ayy Ut 9dA} UDTJ[Od 94} JO 9d4N0s a[quqoad 104 “v.aQd0ds Uatjod dulEMG equiooyey “G71 “VZI SdINSI] s}ooy gy Kelp ae WS Z ] pues }e9 “, OLXShy ° sions 7 ve ceeence cert nau manila ay a eens on | | | od1#8s6" rAUAE | Se MKB esseenezey 01 x8h: Sail Tee os - 2 | ree Sw val SOLK Lg -- ee PO hae reek oe see $e dere : | | ‘ Or ts vente QEMQZ cree meee ece seach vette esos ne eae camel geteeee Ty te remem el). ga Sl RE et eet et ce ; al ae ytie ee eceeme peer as wien ‘ gay pOb XOL 2 SE OLKGE SF yOLXOTT - wes a - vm mare x y Fat ae 1) a] aes nrenees <3 OLX9CP + “eae x seer Benes apse DS hea | * 02 aed AE ODMR to Se ey orue wea eeeecestrgd ahve vate neon cmos mena ie trices sean Bae wrsainhd auwism ace j;—_" minsins ai at [| 0 gaa OP RA Sma Cetae bee ree oases ajeos a —! if ip ipcatraciaaed i =, 0LXG¢ Be S ghy po onteritan as: raat oH | 0 OZ1LF866 cl ieee o0L#89z'9 weap d pODXP PL sonm sae oe macnn nd moe a eee pee ie ae 2 ctanseh we pabeeet ii 5k od bon rarer com | POD Kit anes tease A= Sees peeeer wee © 7 ra WA So/SuIEUB ,OL x OL = eg rie oe a es ees Ey Oz —| 5 ,OLXOS! we es se Zing neue eee mee aremeg » - ammo rid OLX PEP mci eesse aes] PL OLXZO) “on “Ae ee ~ eeeren Lx Bea sek ees ord | panel <,0.x£0P eR arp Re Py 01xoel ceeetatiser asterd <3, OLXG'9 = { OL ac OLXQT) emcee ate) A nee ee oem EY cease = 0Lxoge=~ “= maw ee etn] SO LNOL | asses mene om een OL XSL. rena =, %0C Ss —w ; ajeas Pol hoy eee eee |e aPertcre mec aes 3.0 aac, rye Cae pOLXCOZ -—- « pe cases j§,0LXBZL Ap 0s wereae rou hans ad weiss eee nee see | q onan 09 ov _—— ~n ee rer ce i myn en — = rie > Nir moe pores cet tk oz i Nee OLX £97 “nese nese == oe an =|4,OLXP = 24 ,0XEe2 SS —_ saa Se ane pce ae | AYO J oa, fBooreyo didossoJoEyy 97 Proc. Linn. Soc. N.S.W., 130, 2009 HOLOCENE HISTORY OF BLUE MOUNTAINS VEGETATION Figure 12 continued decreased, along with Casuarinaceae in the time of the Europeans. Fire activity was low to moderate through most of the time. Newnes Swamp Newnes Swamp, at 33° 22’ 57” S, 150° 13’ 20” E and 1,060 m altitude, is located in a shallow hanging valley with pine plantations in close proximity. Regular burning maintains fire breaks for the young pine plantations. The swamp is underlain by the Burra-Moko Head Sandstone Member of the Banks Wall Sandstone Formation which has thin claystone interbeds, and it is likely that one of these clay layers impedes drainage and hence maintains the swamp. Swamp stratigraphy: The core sampled 90 cm of sediment. Clay or peat with roots was found down 98 to 20 cm, then sandy clay down to 35 cm, followed by sand to 55 cm, then sandy clay with roots down to 65 cm, then silty clay to 75 cm, and finally sand or sandy clay in layers to 90 cm. Radiocarbon dates are presented in Table 8. The swamp vegetation and surface pollen: Banksia and Kunzea were dominant and Baeckea, Leptospermum, other sclerophyllous shrubs, Cyperaceae and Poaceae were also present on the swamp (Chalson and Martin, this volume). There was appreciable Myrtaceae pollen in the surface samples, but Restionaceae and Gleichenia were dominant in the surface pollen spectra. Pinus was present but not abundant. (Chalson and Martin, this volume). The pollen record: Pollen recovery from the core Proc. Linn. Soc. N.S.W., 130, 2009 J.M. CHALSON AND H.A. MARTIN 50 Depth (cm) 100 0 2 6 Age (k yr) J Radiocarbon date @ Calibrated date Figure 13. Katoomba Swamp summary diagram. was good and there was some exceptionally high concentrations, especially in the clay at 60-70 cm. The core was zoned thus (Figs 14A, 14B): Zone D, 90-55 cm, c. 11,000-7,5 00 cal yr BP (see Fig. 15 for estimated ages). Myrtaceae pollen was low, but Eucalyptus pauciflora/rubida had been identified. Casuarinaceae was also low at the base of the zone, increasing upwards (Fig 2A). Asteraceae/ Tubuliflorae and Chenopodiaceae were prominent amongst the herbs and shrubs (Fig. 14B). Poaceae and Restionaceae were well represented. Zone C, 50-40 cm, c. 7,500-1,800 cal yr BP There was very little Myrtaceae pollen, with only one record of a Eucalyptus species. Casuarinaceae pollen increased, Haloragis was moderate and Poaceae and Restionaceae were reduced when compared with the Table 8. Radiocarbon ages for Newnes Swamp Depth (cm) Material dated Laboratory no. 20-25 Sandy clay SUA 2648 35-40 Sandy clay SUA 2649 50-55 Sand SUA 2650 77-83 Silty clay SUA 2651 87-93 Sand SUA 2652 Proc. Linn. Soc. N.S.W., 130, 2009 preceding zone. Zone B, 35-25 cm, c.1,800-?1,000 cal yr BP. Melaleuca representation was significant, Casuarinaceae had decreased, the shrubs were well represented, and Poaceae and Restionaceae remained low when compared with the previous zone. Zone A, 20-0 cm, ?1,000 cal yr BP to present. Melaleuca continued to be the most significant of the Myrtaceae, Styphelia and Haloragis were appreciable, Poaceae remained low and Restionaceae was somewhat greater than the zone below. Pinus was present throughout the zone, denoting European activity. The charcoal content was moderate in zone D, extremely low in zone C, and moderate to high in zones B and A. Radiocarbon years Calibrated age (cal. yr (yr BP) BP) 1,090+70 1,098+80 1,930+£70 1,948+80 6,650+100 7,588+80 9,820+90 11,398+130 9,640+80 11,038+160 99 HOLOCENE HISTORY OF BLUE MOUNTAINS VEGETATION y, ~ “era, ae * 2 Ys oth | . 8 Sera er a ee o oO Oo oS i= o ve oO Sot 3 ee oT ok es 2 a rs a > if e _ ae ra) 8 j 7 = eS | If “ip o BS rh bee, eg, 8 < a8] ) a | | | | | : By, : a Sa, r - — : 00) | c=) Ou. = Uig Soy Sy oH Yay ss Mo, = 10 : Me oe SS of jen E Mf : “% ; [ re Arey hy y o > “a5 = * oe b secret oereore as : ih i [= ES Vee, pes Pop, ‘ h ! i t E i Sy, 43,000-year vegetation and fire history from Lake Baraba, New South Wales, Australia. Quaternary Science Reviews 25, 3003-3016. Boland, D.J., Brooker, M.I.H., Chippendale, G.M., Hall, N. H., Hyland, B.P.M., et al. (2002). ‘Forest Trees of Australia’ 4" Edition. (Thomas Nelson: Australia) 687 pp. BoM, (2006). Commonwealth Bureau of Meteorology Website (http://www.bom.gov.au). Accessed August 2006. Breckell, M. (1993). Shades of grey: early contact in the Blue Mountains. In “Blue Mountains Dreaming, the Aboriginal Heritage’ (Ed. E. Stockton) pp. 114-121 (Three Sisters Publication Winmalee NSW). Brown, C.A. (1960). ‘Palynological Techniques’ (C.A. Brown, Baton Rouge) 188 pp. Bureau of Meteorology (1979). ‘Climatic survey Sydney Region 5 New South Wales’. Depatment of Science and the Environment, Commonwealth of Australia (Australian Government Publishing Service, Canberra) 141 pp. CalPal (Version March 2007). Cologne Radiocarbon Calibration and Palaeoclimatic Research Package. (http://www.calpal.de/). Accessed May 2008. Chalson, J.M. (1991). The late Quaternary vegetation and climatic history of the Blue Mountains, NSW, Proc. Linn. Soc. N.S.W., 130, 2009 J.M. CHALSON AND H.A. MARTIN Australia. PhD Thesis, University of New South Wales (unpubl.) Chalson, J.M. and Martin, H.A. (1995). The pollen morphology of some co-occurring species of the family Myrtaceae from the Sydney region. Proceedings of the Linnean Society of New South Wales 115, 163-191. Chalson, J.M. and Martin, H.A. (2008). A 38,000 year history of the vegetation at Penrith Lakes, New South Wales. Proceedings of the Linnean Society of New South Wales 129, 97-111. Chalson, J.M. and Martin, H.A. (this volume). Modern pollen deposition under vegetation of the Blue Mountains, New South Wales. Proceedings of the Linnean Society of New South Wales 130, 111-137. Donders, T.H., Haberle, S.G, Hope, G., Wagnera, F. and Visschera, H. (2007). Pollen evidence for the transition of the Eastern Australian Climate system from the post-glacial to the present-day ENSO mode. Quaternary Science Reviews 26, 1621-1637. Entwisle, T. (2005). She-oak up in smoke. Nature Australia, Spring 2005 28(6), 72-73. Keith, D.A. and Benson, D.H. (1988) The natural vegetation of the Katoomba 1:100 000 map sheet. Cunninghamia 2, 107-143. Kohen, J. (1995). ‘Aboriginal Environmental Impacts’. (University of New South Wales Press, Sydney) 160 pp. Langford-Smith, T. (1979). Geomorphology. In “An Outline of the Geology and Geomorphology of the Sydney Basin’ (Ed. D. Branagan), pp 49-58. (Science Press, Marrickville). Moore, P.D., Webb, J.A. and Collison, M.E. (1991). “Pollen Analysis’. (Blackwell Scientific Publications, Oxford) 216 pp. Moy, C.M.., Seltzer, G.O., Rodbell, D.T. and Anderson, D.M. (2002). Variability of El-Nifio/Southern Oscillation activity at millennial timescales during the Holocene epoch. Nature, 420, 162-165. Robbie, A. and Martin, H.A. (2007). The history of the vegetation from the last glacial maximum at Mountain Lagoon, Blue Mountains, New South Wales. Proceedings of the Linnean Society of New South Wales 128, 57-80. Rose, S. and Martin, H.A. (2007). The vegetation history of the Holocene at Dry Lake, Thirlemere, New South Wales. Proceedings of the Linnean Society of New South Wales 128, 15-55. Sainty, G.R. and Jacobs, S.W.L. (1981). “Water Plants of New South Wales’ (Water Resources Commission New South Wales, Sydney) 550 pp. Smith, J. (1993). Katoomba’s fringe dwellers. In “Blue Mountains Dreaming, the Aboriginal Heritage’ (Ed. E. Stockton) pp. 122-135. (Three Sisters Publication Winmalee NSW). Stockton, E. (1993a). Archeology of the Blue Mountains. In ‘Blue Mountains Dreaming, the Aboriginal Heritage’ (Ed. E. Stockton) pp. 23-54 (Three Sisters Publication Winmalee NSW). Proc. Linn. Soc. N.S.W., 130, 2009 Stockton, E (1993b). Aboriginal sites in the Blue Mountains. In “Blue Mountains Dreaming, the Aboriginal Heritage’ (Ed. E. Stockton) pp. 55-62 (Three Sisters Publication Winmalee NSW). Sullivan, L. (2007). Small victory in long battle to save hanging swamps. Sydney Morning Herald Letters, 17 October 2007. Ward. D.J., Lamont, B.B. and Burrows, C.L. (2001). Grasstrees reveal contrasting fire regimes in eucalypt forest before and after European settlement of southwestern Australia. Forest Ecology and Management 150, 323-329. 107 HOLOCENE HISTORY OF BLUE MOUNTAINS VEGETATION APPENDIX A Pollen type name on the pollen diagrams and the probable source in the vegetation. Name of pollen type. Major pollen groups (A diagram) Podocarpus Pinus Angophora/Corymbia Eucalyptus/Melaleuca Melaleuca styphelioides Leptospermum/Baeckea Tristaniopsis Unidetified Myrtaceae Casuarinaceae Poaceae Restionaceae Cyperaceae Selaginella Gleichenia Other fern spores Probable source in the vegetation and ecological inference. From PlantNet (2007) Probably Podocarpus spinulosus: sclerophyllous shrub/small tree Pinus sp(p). Introduced: Pollen input from urban/forestry areas. Species within the two genera: sclerophyll woodland Species within the two genera sclerophyll woodland/forest Melaleuca styphelioides: moist stream bank habitat Species within the two genera: ?mainly swamp communities Tristaniopsis spp : moist habitats in sclerophyll communities All pollen types not identifiable further Casuarina, Allocasuarina sp(p): A. distyla and A. nana in this study Native and exotic species in the family: open situations, dryland and swamp species All species in the family: swamp and dry land species All species in the family: swamp and dry land species All species in the genus: damp sites, edge of swamp Gleichenia sp(p): damp sites, edge of swamp Other ferns: many possible species Names of shrubs and herbs (B diagrams) Grevillea acanthifolia G. sphacelata Grevillea Hakea Persoonia pinifolia Persoonia Symphionema montanum Banksia Other Proteaceae Acacia Styphelia Monotoca Other Ericaceae Rutaceae type Pimelea Plantago Haloragis Other tricolporate grains Podocarpus Micrantheum Myriophyllum Asteraceae/Liguliflorae Asteraceae/Tubuliflorae Chenopodiaceae Shrub: swampy areas, sand or peat Shrub: heath, dry sclerophyll forest Grevillea sp(p): sclerophyllous understorey Hakea sp(p): sclerophyllous understorey Shrub: heath, dry sclerophyll forest Persoonia sp(p): sclerophyllous understorey Shrub: heath or dry sclerophyll forest, wet or dry situations Banksia sp(p): sclerophyllous understorey Other taxa in the family: sclerophyllous understorey All species in the genus Styphelia sp(p): sclerophyllous understorey Monotoca sp(p): sclerophyllous understorey Other taxa in the family: sclerophyllous understorey All taxa in the family sclerophyllous understorey Pimelea sp(p): sclerophyllous understorey Plantago sp(p): native and introduced herbs Haloragis/Gonocarpus sp(p): Damp sites, sclerophyllous understorey Probably shrubs and herbs Probably Podocarpus spinulosus: sclerophyllous shrub/small tree Shrub: heath and dry sclerophyll forest, sandy infertile soils Mainly aquatic herbs, also on damp ground around water bodies Fenestrate-grained taxa in the subfamily Liguliflorae: herbs Echinate-grained taxa in the subfam. Tubuliflorae: shrubs and herbs Ruderals, salt tolerant 108 Proc. Linn. Soc. N.S.W., 130, 2009 J.M. CHALSON AND H.A. MARTIN APPENDIX B Myrtaceae Pollen type name on the pollen diagrams and the probable source in the vegetation. Name on the pollen diagrams Angophora costata Probable source in the vegetation and ecological inference. From PlantNet (2007) Deep sandy soils on sandstone Angophora floribunda A. costata x floribunda Angophora Baeckea/Leptospermum Callistemon Eucalyptus deanei E. eugenioides E. fibrosa E. oblonga E. oreades E. pauciflora/E. rubida E. piperita E. racemosa Eucalyptus type B Eucalyptus type C Eucalyptus type D Eucalyptus type K Eucalyptus type M Eucalyptus/Melaleuca Kunzea Leptospermum juniperinum L. polygalifolium Melaleuca ericifolia M. styphelioides Melaleuca type B Melaleuca type C Melaleuca Myrtaceae type C Myrtaceae type D Unidentified Myrtaceae Usually on deep alluvial soils Some species in swamp/moist habitats, also dryland species Dry sclerophyll communities, some swamp species Tall wet forest, sheltered valleys, deep sandy alluvial soils Dry sclerophyll or grassy forest, on deep soils Wet or dry sclerophyll forest, on shallower, somewhat infertile soils Dry sclerophyll woodland, on extremely infertile, sandy soils Wet or dry scleropnhyll forest, on poor skeletal or sandy soils Grassy or dry sclerophyll woodland, on cold flats. Dry sclerophyll forest/woodland, moderately fertile, often alluvial sandy soils Dry sclerophyll woodland, on shallow infertile soils ) ) ) For definition of Eucalyptus pollen types, see Chalson (1991) ) ) Species within the two genera: sclerophyll woodland/forest Understorey sclerophyll forest, moist depressions Swamp, heath and sedgeland, on sandy peat soils Dryland habitats and moist depressions Heath and dry sclerophyll forest, streambanks and coastal swamps Moist situations, often stream bank habitats ) ) For definition of pollen type, see Chalson (1991) For definition of Melaleuca pollen types, see Chalson (1991) ) ) For definition of pollen types, see Chalson (1991) All myrtaceous pollen types not identifiable further Proc. Linn. Soc. N.S.W., 130, 2009 109 SEMEN RUA fel ath Ce lt eal saath mst dagen sid maces st we en — Drench 3 eee castes iat Lise ap cata nt ne “ome vat Oli: a hich ELeTIIS es 1 ? Siac ater ms vs ; y sented hordes cle 4 tt a or alge on i a | : ; piper ae eine wi ied hil eer o. 1 LOOTED ) OF whee AW erty pw be er anal aoe yet NM Yrb als 2 RPA DAH LE asic istry” ae ala honk “3 einoge terns Sele OT ieee {Pec mahiy’ PHP os ND athe SCC | ys oreo tt rite ybose beta inesttiay' foeen ey an ea i el ae nih eens abiue qaob te: eeny Vente ib Wh ainsty ag. a ocak wale preven line skiirrsiiny Huby Senad TOWwG ERAS Tat eae tS ehh FT ete a9) ew sat! hee sear ETH iy au as ahiow' vbnat ,chirtstet yi PRS ty erred H gigas Hoe mg "WHO? viotiae 76 [ndekaale 1d ES hy Ivins i ww” me hams de a tee Baie tay ene Re iit fear “rh thy vorank 208. Res, Axprallyy sN Savi y weatho obit. vloneist bili Wisk tbh Wy eT eerie Kare vee ic vest: SEM MR a i ihe Pies ey a ation olin mi pst Hey a eee he leas er na os. | hc ( uke mingle ay ot eH hrasatnih 5 2 ea ania (FOR GREY sere exgy! ANG PUERRDN Ay ASH Se Co nas aati slid, "Snake jonah diy eboney tt Finmeat eq eet Cneviltedaphd): scleriphy dans uinkeuorey® M anh) savienihe fest onetlbotaye 1h yuh ee RB cel OW SRY mihi eLGe wonalnls a ani mE xoteaniinsh Pedi bias Ai ea Su ifobebe! J \ Wf i Were 4 vee gaclor liye dabg espii dO, tiaslaabee, bp, thsad LOTUS 0 ae eee Alot TAR AHURA DITA , aa WE? AWE. brs estrada ONG Lao a> sey BA ERS Pres he “SStlolisiye pSaalhiwt | x} mati THRE Hite OMS ano aulic salaly | ahiohlaneige Hy 1 SP): MTCTO Pay thous anisherstg Gl wach isis si ry’ a eel gahay age Mm Llale oh te) Bae { (Oe T pneielog > 598 oa nolan, i RITA L 2) Maal st er D940 ounle waeyyy ( eh ep att tO He bia Be ty Vide tht fred ot he [ev (ts TT “alg fv : (fRO4) mosien) ou? gp Sa RANGA? i oh Cl oqyt sad poe aidedisaHhy Hoet'ita ss Be eee 0d aaa seat ani) i me, incarceration iarnnes op ee ye balhe — sana rah! SEP RA T PiObahity % inka how bere, pataltiy Qader ics aprpiainlanen ach neha thee rh vt ge Saath hn sity ane ery ye! eneaptey} taht ar. wANC Ls waterrle Shute ; ,, = - ; ~ ee ‘parotid bresrts, dss cat deri. growl aren Ww mor! i ert ene Lwutifionas a wy ray et winnd SN U “one wubrtie nily Livihiiarae tebe pe 1 q 7 r Dy. ico ip ' a. THB siwns fi¢ho nate yes ee, ti aa 1 he sear titite. } nbudifion he! snaibeland 4 Piberpa ag aia ehcp oder égald tobmedit: ” ane: Ns ead pad, nie eer Sarwan — a oti A hae tn Here _ a cer ha ll A ee Modern Pollen Deposition Under Vegetation of the Blue Mountains, New South Wales JANE M. CHALSON! AND HELENE A. MARTIN? '46 Kilmarnock St. Engadine, N.S.W. 2233 * School of Biological, Environmental and Earth Sciences, University of New South Wales, Sydney Australia 2052 (h.martin@unsw.edu.au) Chalson, J.M. and Martin H.A. (200x). Modern pollen deposition under vegetation of the Blue Mountains, New South Wales. Proceedings of the Linnean Society of New South Wales 130, 111-137. Pollen was extracted from surface samples of swamp sediments and soils under various types of vegetation in the catchments of these swamps. The pollen assemblages in these surface samples were compared with the floristic composition of the vegetation to provide a means of interpreting the assemblages of fossil pollen retrieved from the swamp sediments. The surface pollen assemblages reflected the local vegetation, indicating more/less tree cover, swamp and/ or adjacent dryland environment and local flora diversity. All the evidence pointed to very local deposition and little long distance dispersal of pollen. A number of different units may be defined within the one major vegetation type, dry sclerophyll forest/woodland 1n this case, but the floristics of the units are too similar to allow discrimination of them from their modern pollen assemblages. Manuscript received 21 May 2008, accepted for publication 17 December 2008. KEYWORDS: Blue Mountains, local pollen deposition, long distance pollen dispersal, modern pollen deposition, pollen spectra. INTRODUCTION Pollen is deposited in sediments by the contemporaneous vegetation, but a number of factors affect the representation of each taxon in the sediments so that it is not possible to relate a fossil pollen assemblage in a deposit directly to the vegetation that produced it. Pollen productivity, dispersal and preservation are the main factors that influence representation of a taxon, and each of these factors are in turn influenced by the local environmental conditions. Pollen deposited from under known plant communities, however, may be used to characterize that community and hence assist in the interpretation of pollen spectra recovered from swamp sediments. The nature of pollen deposition of individual taxa may also be deduced from the surface pollen spectra. Sites for a study of the history of the vegetation were chosen from swamps in an altitudinal sequence in the Blue Mountains (Fig. 1). These sites are situated on a relatively uniform substrate, sandstone, within dry sclerophyll woodland/open forest. Observations of modern pollen deposition are reported in this paper, and the Holocene history of the vegetation from the swamps is reported in Chalson and Martin (this volume). THE STUDY SITES The Blue Mountains are a deeply dissected plateau rising from the Cumberland Plain in the east. The plateau surface is undulating and small creeks form upland valleys. Where the underlying rock type is Hawkesbury Sandstone, the upland valleys become incised and develop into V-shaped gorges. In the west where rock type is Banks Wall sandstone, the valley sides and floors slope gently and the streams flow through a series of swamps (Chalson, 1991). The swamps chosen for study are as follows (see Fig. 1) and the species found at each site are listed in Appendix 1: Burralow Creek Swamp, at 33° 32’S, 150° 38’E and 310-330 m altitude, is anarrow swamp that follows the creek for some 3.5 km. The upper end of the swamp is 2 km southeast of Kurrajong Heights. The core site is | km downstream from the northern end. There are few cleared areas near the swamp, the nearest being over 2 km away. The vegetation around Burralow Creek is open forest, woodland and swamps (Keith and Benson, MODERN POLLEN DEPOSITION IN THE BLUE MOUNTAINS 33° 30’ sy Mt. Victoria \ oy be L Grose ® pleekheath bos Mediow Bath & Study area ney Burralow k Sw. * oe eee ~A y, S =i Springwood & ie ‘ Katoomba | awson sj Qa 2 Penrith 33° 45’ ~ pees Jenolan R. ~ 150° 00’ 15° 0 Scale 15 km © & Urban area Katoomba Sw:-* Wood NympHs Dell Warrimoo Kings Tablelands Sw. 5 ~~" Notts Sw. ~~, \ -, ae > Ce ae Sw. Oval Sw. Murphys Glen a es Ingar Sw. Glenbrook — Yoo Warragamba Dam WP agieg 30 150° 45) %e Swamp, this study ve Other study site & Other surface sample Figure 1. Locality map 1988). Angophora bakeri, A. costata, Corymbia eximia, Eucalyptus eugenioides, E. multicaulis, E. paucifiora and E. radiata are locally dominant with a few kilometers of the swamp. The surface of the swamp supports an open heathland of Leptospermum polygalifolium, L. trinervium and _ Eleocharis sphacelata. Nomenclature follows Harden (1992; 1993; 2000; 2002) and PlantNet (2006) 2 Warrimoo Oval Swamp, at 33° 43’ 21.44”S, 150° 36’ 58.35”E and 190-200 m altitude, is approximately 1.5 km east of Warrimoo Post office and 0.4 km south of Warrimoo Oval. There are substantial urban areas within a kilometer of the swamp and weed invasion is considerable. The vegetation is mainly woodland with some open forest and swamp communities (Keith and Benson, 1988). Locally, Angophora bakeri, Proc. Linn. Soc. N.S.W., 130, 2009 J.M. CHALSON AND H.A. MARTIN Eucalyptus pauciflora and E. radiata are dominant. The swamp surface supports an open heathland with Leptospermum spp. Notts Swamp, at 33° 48’ 35.44” S, 150° 24’ 27.66” E and about 682 m altitude is approximately 12 km south-southeast of Wentworth Falls and to the west of Notts Hill. The lower third of the swamp is used as a market garden, but there is no sign of disturbance or weed invasion at the study site. There is no indication of European activities in the catchment upstream of the study site and the nearest settlement is some 7 km to the north-northeast. The major plant community is open woodland and there is a little open forest and some swamps (Keith and Benson 1988). Eucalyptus eugenioides, E. multicaulis, E. piperita, E. racemosa and E. sieberi are locally dominant. The swamp supports a closed sedgeland of Gymnoschoenus sphaerocephalus, Leptospermum trinervium and Baloskion australe. Ingar Swamp, at 33° 46’ 11.65” S, 150° 27’ 22.92” E and 584m altitude, is approximately 8 km southeast of Lawson. European settlement is some five km to the northeast, along the highway, and includes some very large, old conifer trees. The vegetation is mainly woodland with Corymbia gummifera, Eucalyptus oblongata, E. piperita, E. pauciflora, and Angophora costata dominant locally. Open forest in gorges along the creeks is dominated by E. eugenioides, E. sclerophylla, Tristania neriifolia and Angophora costata. The swamp community is a closed sedgeland of Gymnoschoenus sphaerocephalus, Leptocarpus tenax, Baumea sp., Chorizandra sp., Baloskion australe and, towards the edge, Hakea teretifolia, H. dactyloides and Leptospermum lanigerum. Kings Tablelands, at 33° 45’ 47” S, 150° 22’ 43” E and about 780-790 m altitude, is located in small valley off Queen Victoria Creek. It is about 0.6 km east of Queen Victoria Memorial Hospital near Wentworth Falls. An urban area is found less than | km to the west where exotic conifers have been planted in the gardens. The vegetation is mainly open forest around the study site, with woodland on the ridges and closed sedgelands in the swamps (Keith and Benson, 1988). Locally, Eucalyptus dives, E. oreades, E. sieberi and E. piperita are dominant in the open forest and Corymbia gummifera, E. racemosa and E. sieberi are dominant in the woodland. On the exposed plateau to the northeast, the dominants in an open heathland are Allocasuarina distyla, E. ligustrina, E. stricta, Proc. Linn. Soc. N.S.W., 130, 2009 Banksia serrata and Hakea teretifolia. The dominants on the swamp are Leptospermum juniperinum and L. grandiflorum. Katoomba Swamp, at 33° 43’ 03” S, 150° 19’ 18” E and 950 m altitude, is 1 km east northeast of Katoomba Post Office and 1 km west of Leura Post Office. This swamp is surrounded by urban activity, with drainage ditches and a sealed road running across the swamp. Much of the swamp is (or has been) used for yards for light industry and horse paddocks. Housing extends to the edge of the swamp. Most of the area around the swamp has been cleared but there are a few remnant pockets of Sandstone Plateau Forest (Keith and Benson, 1988) remaining. Eucalyptus acmenoides, E. oreades, E. stellulata, E. oblongata and E. sieberi are dominant. The understorey is problematic as the remnant stands are heavily weed infested. Little remains of the original vegetation over the swamp surface and species of Poaceae are predominant. A small patch of swamp edge vegetation forms a dense thicket of Leptospermum juniperinum and L. scoparium. Newnes Swamp, at 33° 22’ 57” S,150° 13’ 20” E and 1,060 m altitude, is within a forestry area with pine plantations. Regular burning maintains fire breaks. Woodland communities are found around the swamp (Benson and Keith, 1990) but the shrub layer has been much reduced by frequent burning. Shrubs remaining on the swamp include Leptospermum trinervium and Grevillea acanthifolia. A ground cover of grasses is found in all but the wettest areas where Juncaceae and Restionaceae are dominant. METHODS The vegetation units at each site were determined from maps in Benson (1992), Keith and Benson (1988) and Benson and Keith (1990). Each site was visited, the vegetation checked with the maps and as many species as possible were identified in each of the vegetation units. Since palynology cannot reveal the structure of the vegetation, the focus of survey was on the species list. Dominance was determined subjectively from the abundance of the species Samples from the surface of the soil, or where possible, from moss polsters, were collected from the centre of the swamp, the swamp edge and the plant communities adjacent or local to, the swamp sites. Samples were taken from at least 100 m away from community boundaries where possible. The sample types and vegetation are listed in Table 1 and the 113 MODERN POLLEN DEPOSITION IN THE BLUE MOUNTAINS Table 1 Surface samples used for pollen spectra presented in Figs 2 and 3. Codes for vegetation map units are from Keith and Benson (1988). ; Vegetation . STIRS Vegetation e Sample material sample no. map unit Burralow Creek 1 Open sedgeland mid-swamp 28a Soil 2 Open sedgeland mid-swamp 28a 0 cm core 3 Swamp fringe 28a Soil 4 Low Woodland 10ar Soil 5 Open forest 10ag Soil Warrimoo Oval 6 Closed sedgeland mid-swamp 26a Soil 7 Closed sedgeland mid-swamp 26a 0 cm core 8 Closed sedgeland swamp fringe 26a Soil 9) Low woodland 10ar Soil Notts 10 Closed sedgeland mid-swamp 26a Soil 11 Closed sedgeland swamp fringe 26a Soil Ingar 12 Closed sedgeland mid-swamp 26a Soil IL) Closed sedgeland swamp fringe 26a Soil 14 Low woodland 10ar Soil 15 Low woodland 10ar Soil Kings Tableland 16 Closed sedgeland mid-swamp 26a 0 cm core iW, Closed sedgeland swamp fringe 26a Soil 18 Low woodland 10ar Soil 19 Low woodland 10ar Soil 20 Open forest 91 Soil Dil Open forest 91 Soil 22 Open heath Dalits Soil Katoomba 73 Closed sedgeland mid-swamp 26a Soil 24 Closed sedgeland swamp fringe 26a Soil 1) Open forest 91 Soil 26 Open forest 91 Soil Newnes ZT Closed heath mid-swamp 20a Moss 28 Closed heath swamp fringe 20a Moss 29 Woodland 10f/lla Moss 30 Woodland 10f/lla Moss 31 Woodland 10f/lla Soil 32 Woodland 10f/lla Soil 33 Open heath 21d Soil 34 Open heath ZN Soil 35 Forest 10f Soil 36 Forest 10f Soil Murphys Glen 37 Tall open forest 6c Soil 38 Tall open forest 6c Soil Wolgan 39 Open woodland lla Soil 40 Open woodland lla Soil Wood Nymphs Dell 4] Open forest 10ag Soil Medlow Bath 42 Open forest 91 Soil 114 Proc. Linn. Soc. N.S.W., 130, 2009 J.M. CHALSON AND H.A. MARTIN study sites are shown in Fig. 1 Six to ten sub-samples were taken from each plant community over a transect of approximately 20 m. The sub-samples were mixed together to reduce the possible over-representation of any one species due to close proximity to an individual plant (Chalson, 1991). The samples were treated with hydrochloric and hydrofluoric acids to remove siliceous material (Birks and Birks, 1980), oxidised with Schultz solution (a saturated solution of potassium perchlorate in nitric acid), cleared in 10% potassium carbonate and the residue was mounted in glycerine jelly (Brown, 1960). Pollen was identified by comparing the grains with reference pollen treated with standard acetolysis (Moore et al., 1991). Grains were counted along transects across the slides and tests showed that a count of 140 grains adequately sampled the residues. The counts of each pollen type were presented as percentages of the total count on the pollen diagrams. RESULTS Fig. 2 presents the pollen spectra from vegetation on the swamp surface and at the edge of the swamp, and Fig 3. presents spectra from the dry-land communities in the surrounding vegetation. Table 2 presents the name on the pollen diagram, the probable source of the pollen in the vegetation and ecological inference. Preservation, although adequate, was not good enough for the identification of Eucalyptus species beyond broad groups (Chalson and Martin, 1995). The pollen from moss polsters may be better preserved than that from the soil, but moss polsters were not common and usually dried out severely in the forest environment, hence soil samples were usually collected in all but the dampest areas. Exotic Pinus is present in all samples (Figs 2A, 3B) and values are highest at sites near urban areas (Kings Tableland, Katoomba). Surprisingly, Pinus values are not high at Newnes, in the forestry area with pine plantations, but the pines were very young at the time of this study. Angophora/Corymbia and Eucalyptus/Melaleuca have been identified in low frequencies in some of the samples which were better preserved. Melaleuca styphelioides has been identified in some of the swamp samples (Fig. 2A) where counts may be high. M. styphelioides was not found during the survey of the vegetation, but it may be grown in gardens. The Proc. Linn. Soc. N.S.W., 130, 2009 highest count at Warimoo Oval Swamp is close to substantial urban areas. Leptospermum/Baeckea has been identified from some swamp samples (Fig. 2A) where counts may be considerable. Leptospermum spp. are often dominant in the swamp communities (see Appendix 1) The unidentified Myrtaceae group is larger than the other groups of Mytaceae and counts from the swamp samples are the lowest of all. The woodland or forest samples from the borders of the swamp (Fig. 2A) all have higher counts than the swamp samples. Frequencies in samples from the dry-land vegetation (3A) are much higher than those from swamps. Lack of specific identification was generally due to poor preservation. Casuarinaceae frequencies are usually low, with a few higher values. The highest value (Fig. 3A) comes from heathland vegetation. Poaceae frequencies are generally low and the high values are associated with urbanisation and disturbance (Katoomba, Fig. 2A). Restionaceae frequencies are variable but most of the high values are found in the swamp samples. Cyperaceae has not been recorded from many samples, and where it is present, frequencies are generally low, with the few higher frequencies being found in the swamp samples. Selaginella is present in a few samples and appreciable frequencies may be recorded in some swamp samples. Gleichenia may be present in appreciable frequencies in some swamp samples also. Other fern spores are usually recorded in low frequencies and are more common in the dry-land samples. Table 2 also lists the likely environmental indication of the pollen groups on the diagrams, but this is difficult, given that a group may include many possible species. For example, the families Restionaceae and Cyperaceae include both swamp and dry-land species, but the species in the vegetation and patterns of high pollen frequencies on the diagrams may indicate the nature of the environment when considered together. Thus the species of Restionaceae and Cyperaceae found in the local vegetation (Appendix 1) are almost entirely species of swamps or damp places (Table 2). DISCUSSION There are many indications that the pollen recovered from the surface samples was produced mainly by the local vegetation and thus the pollen spectra can indicate the type of 115 MODERN POLLEN DEPOSITION IN THE BLUE MOUNTAINS iy eo] @ eu] a5 Ro tty oe? § é g Ss % 0 & G @ & & @ Se g = @ a oY s y fo} Lg & C) oy 2 & @ a a fr io) 72 oP g § 2 e e 8g & &F § & é g § <« << 2 2 IY YL te) RS ae : N SoS & & J é eo é é on ee ek : : eves © Burralow Creek Swamp 1 Mid swamp S — 2 Mid swamp , 3 Swamp edge 4 Weadiand | eee oe 10 Mid swamp t | 11 Swamp edge 12 Mid swamp z | 13 Swamp edge 18 Woodland 16 Mid swamp iz L Swamp edge a 21 Forest os a m7 N fh + HW Swamp edge 26 Forest ae: Mid swamp | Swamp edge { Woodland | mh o oo ~ 6 Mid swamp fal 7 Midswamp [~ ca 8 Swamp edge 9 Woodland ia ~~ a Berea | direc tee Warrimoo Oval Swamp Notts Swamp — Ingar Swamp | eee Kings Tableland Swamp fay oF ATTA, MaapepD er mee sae RES ie, ae Rye oe Ge i I Katoomba Swamp Toles Bel al dele TT jaar ase - th, ay te ee Newnes Swamp % ee | | aos Scale —— ee count a Figure 2A. The pollen spectra from plant communities associated with swamps within major pollen groups. The Sample number (extreme left hand side) refers to the sample in Table 1. Proc. Linn. Soc. N.S.W., 130, 2009 116 J.M. CHALSON AND H.A. MARTIN & & Scale . 20% of total count 2 Ss = e QD 5 Ca ve rd % & ® $ S ¢3 F) £ ~ 4 @ & So 2 8b 9 FE EE SSE GET ES ®2¢ F S&S Fs SF EFESESES FEA BEEE OL LIF EES LETC LLEP TIE Burralow Creek Swamp ab 7 Mid swamp | | | i L j } ! ' 2 Mid swamp | | is \ | r 3 Swamp edge | 4 Woodland |= at ed | oodian Pee Se dL) 4 b Warrimoo Oval Swamp 6 Mid swamp | + | L iL | 7 Mid swamp = i | | | 8 Swamp edge H aeety | 9 Woodland | - | L L lj r [ | Notts Swamp 10 Mid swamp | L | | | { ' | 41 Swamp edge | | | | i [ f | | | [ | Ingar Swamp 12 Midswamp | ] | | | 13 Swamp edge | | | | L { L L | r i ; ees Kings Tableland Swamp 18 Woodland ‘ | i i 16 Mid swamp | r | - | | | [ 17 Swamp edge ig | ri - oP Perks 21 Fores | ee Re de bai Katoomba Swamp 23 Mid swamp | l | Ie 24 Swamp edge i { foot || ae ea Newnes Swamp 27 Mid swamp | i F r | ae ae 28 Swampedge F | le j 1 | 29 ‘Noodland = le ror 4 br bk b | L Figure 2B. The pollen spectra from plant communities associated with swamps within low frequency taxa. The Sample number (extreme left hand side) refers. to the sample in Table 1 vegetation from which it came. For example, the Myrtaceae pollen content (Figs. 2A, 3A), is lowest from swamp sites, intermediate from the dry-land communities bordering the swamps and highest from the woodland and forest sites away from the swamps, thus inferring a parallel approximate tree cover. Swamp samples contain much higher pollen frequencies of Restionaceae and/or Cyperaceae than the dry-land sites, although both of these families contain swamp and dry-land species. The species of Restionaceae recorded in the vegetation (Appendix 1) are found on wet and poorly drained soils and in damp to wet heaths (PlantNet, 2007). Most of the species of Cyperaceae, on the other hand, are found in fresh water swamps and swampy areas (Sainty and Jacobs, 1981; PlantNet, 2007), although one dry-land species is also recorded (Appendix 1). Thus high frequencies Proc. Linn. Soc. N.S.W., 130, 2009 of Cyperaceae probably indicate swamps which are more permanently waterlogged than swamps with high frequencies of Restionaceae. Both Selaginella and Gleichenia are found in wet places, on the edge of swamps and streams (PlantNet, 2007). The pollen of sclerophyllous shrub taxa (Figs 2B, 3B) are usually found sporadically and in very low frequencies, indicating under-representation and very localised distribution. These findings are in accord with other studies of surface pollen assemblages which indicate very localised distribution of pollen (Dodson, 1983; Kodela, 1990). Kershaw and Strickland (1990) found that, in a 10 year pollen trapping experiment, most pollen came from within 10 m of the trap. These study sites are all contained within small valleys where some barrier impedes drainage of the 117 MODERN POLLEN DEPOSITION IN THE BLUE MOUNTAINS s ¢ e S) § oe © S & ~ op 5 = @ g &$ s &8 & £ Fe ® & © § £ v § gs Lo §& & §€ & S S RY Q + S FF LN WG > & a) > § eé ree & & ? g P #8 ff etd & 3 i ee Be \ ‘ ‘ | aoe! th aati Woodland 10ar Woodland 10ffita 29 i 30 | | 31 jae 32 | Open Woodland 11a 39 ae AE cial Saal Be Ao 3 Open forest 10ag 5 -—— at | | [ pf ier es = |I| |E a Ss Open forest 9i Tall Open Forest 6c {0a ae Open Forest 10f = ARE E| res F 7 ic} 3 3 Open Heath 2%f, 21d, 21c 22 Scale a 33 pe fe i rier = 34 20% of total count j Figure 3A. Pollen spectra associat- ed with dry-land plant communities within major pollen groups. 1 The sample number refers to the sam- ple in Table 1. Codes for the vegeta- tion map units are from Keith and Benson (1988) Proc. Linn. Soc. N.S.W., 130, 2009 118 J.M. CHALSON AND H.A. MARTIN € § og v * Scale 1, 20% of total count Bo Si S £& cS S $s < & &¢ @ Be anc ghesirescom Sle o ae 7 % oe ¢ Sea askS ao G @ © o § 2LLESS ESa SF EBES F § SESS EE FFE S 5 LL oe & & § ELLES SF EEESESS FEE Bes eiatate te ieakodik ut fo an eee t Day => | ieee take | ae Woodland 10ar | 1 at fr Ha se Wa Ba | | 14 | ela : | | ; | | 15° Bhi i r r 19 | | | | lest t F F Woodland 10f/i1a 29 [> Be We SOC Pelee ce 31 r | LF Poy t Ravel gb hie PLE A re La Open Woodland 11a Sat tiiie fe Abe seprte Hey nies oF sn . | arse aa cr Open forest 10ag | | — Pot |. \ ie i la. doal er | ror | L nen, hee Cake eee ace er ws AE stad - | ares ; - + Herald | i it ones ee Oo socaenpeccor i = o ® m 2 | Open Forest 6c ill ee pen Forest 10f Or ae ee ee | mF io sorrel sal psligee aaa ie Open Heath 21f, 21d, 21c iat. fk TER PE y+] Figure 3B. Pollen spectra associated with dry-land plant com- munities within low frequency taxa. 1 The sample number refers to the sample in Table 1. Codes for the vegetation map units are from Keith and Benson (1988) NN NY =| & Oo mo oa RN aie te. fe ee eee eet ; a oy es ea Tt wae ee | stream and maintains the swamp (fora full description assemblage. While this may happen, it has been found of the sites, see Chalson and Martin, this volume). It that very little pollen is transported into the site so may be argued that pollen can be transported a long __ that the assemblage truly reflects the local vegetation distance by a stream, to be deposited with the local (Chmura and Liu, 1990). Proc. Linn. Soc. N.S.W., 130, 2009 119 MODERN POLLEN DEPOSITION IN THE BLUE MOUNTAINS Table 2. Pollen type name on the pollen diagrams (Figs 2, 3) and the probable source in the vegetation. Name on the pollen diagrams 2A and 3A Podocarpus Pinus Angophora/Corymbia Eucalyptus/Melaleuca Melaleuca styphelioides Leptospermum/Baeckea Tristaniopsis Unindetified Myrtaceae Casuarinaceae Poaceae Restionaceae Cyperaceae Selaginella Gleichenia Other fern spores Names on 2B and 3B Grevillea acanthifolia Grevillea Hakea Persoonia Symphionema montanum Banksia Proteaceae Acacia Styphelia Monotoca Ericaceae Rutaceae Pimelea Plantago Haloragis Asteraceae/Liguliflorae Asteraceae/Tubulifiorae Chenopodiaceae 120 Probable source in the vegetation and ecological inference. From Plantnet (2007) Probably Podocarpus spinulosus: sclerophyllous shrub/small tree Pinus sp(p), Introduced: Pollen input from urban/forestry areas. Species within the two genera: sclerophyll woodland Species within the two genera : sclerophyll woodland/forest Melaleuca styphelioides: moist stream bank habitat Species within the two genera: ?mainly swamp communities Tristaniopsis spp : moist habitats in sclerophyll communities All pollen types not identifiable further Casuarina, Allocasuarina sp(p): A distyla and A. nana in this study Native and exotic species in the family: open situations, dryland and swamp species All species in the family: swamp and dry land species All species in the family: swamp and dry land species All species in the genus: damp sites, edge of swamp Gleichenia sp(p): damp sites, edge of swamp Other ferns: many possible species G. acanthifolia: sclerophyllous understorey Grevillea sp(p): sclerophyllous understorey Hakea sp(p): sclerophyllous understorey Persoonia sp(p): sclerophyllous understorey S. montanum: heath or dry sclerophyll forest Banksia sp(p): sclerophyllous understorey Other taxa in the family sclerophyllous understorey All species in the genus Styphelia sp(p): sclerophyllous understorey Monotoca sp(p): sclerophyllous understorey Other taxa in the family: sclerophyllous understorey All taxa in the family: sclerophyllous understorey Pimelea sp(p): sclerophyllous understorey Plantago sp(p): native and introduced herbs Haloragis/Gonocarpus sp(p): Damp sites, sclerophyllous understorey Fenestrate-grained taxa in the subfamily Liguliflorae: herbs Echinate-grained taxa in the subfam. Tubuliflorae: shrubs and herbs Ruderals, salt tolerant Proc. Linn. Soc. N.S.W., 130, 2009 J.M. CHALSON AND H.A. MARTIN It is unfortunate that the Myrtaceae species cannot be identified in most cases, since the vegetation units are defined on their species of Myrtaceae. Most Myrtaceae grains are small and thin-walled (Chalson, 1991; Chalson and Martin, 1995) and the preservation may not be good enough to preserve this fine detail which would distinguish the species. The result is that there are large counts of unidentified Myrtaceae. The alternate wetting and drying at the soil surface in these sclerophyll forests are not ideal conditions for pollen preservation. The forests, woodlands and heaths defined by Benson (1992), Keith and Benson (1988) and Benson and Keith (1990) are structural units within one major vegetation formation and share many species, although the abundance of a particular species may vary. The pollen assemblages cannot denote structure of the vegetation and the floristics of these units are too similar to allow any differentiation, especially as the Myrtaceae pollen is so poorly preserved. For practical purposes, the surface pollen assemblages can denote major vegetation formations (Birks and Birks, 1980; Moore et al., 1991), more/less catchment tree cover, swamp and/or adjacent dry-land environments and local floral diversity. ACKNOWLEDGEMENTS We are indebted to the Joyce W. Vickery Research Fund of the Linnean Society of NSW, the River Group Fund of the Federation of University Women, and the Penrith Lakes Development Corporation for financial assistance with this project. Our thanks go to Dr. John Turner, the National Parks and Wildlife Service of NSW and the Forestry Commission of NSW for assistance. To the many friends, relatives and colleagues who gave unstinting help and encouragement, our heartfelt gratitude. REFERENCES Benson, D.H. (1992). The natural vegetation of the Penrith 1:100 000 map sheet. Cunninghamia 2(4), 502-662. Benson, D.H. and Keith D.A. (1990). The natural vegetation of the Wallerawang 1:100 000 map sheet. Cunninghamia 2(2), 305-335 Birks, H.J.B. and Birks, H.H. (1980). ‘Quaternary Palaeoecology’ (Edward Arnold, London) 289 pp. Brown, C.A. (1960). ‘Palynological Techniques’ (C.A. Brown, Baton Rouge) 188 pp. Chalson, J.M. (1991). The late Quaternary vegetation and climatic history of the Blue Mountains, NSW, Australia. PhD Thesis, University of New South Wales (unpubl.) Proc. Linn. Soc. N.S.W., 130, 2009 Chalson, J.M. and Martin, H.A. (1995). The pollen morphology of some co-occurring species of the family Myrtaceae in the Sydney region. Proceedings of the Linnean Society of New South Wales 115, 163- Ii. Chalson and Martin (this volume). A Holocene history of the vegetation of the Blue Mountains, New South Wales. Proceedings of the Linnean Society of New South Wales 130, 77-109. Chmura, G.L. and Liu, K-B. (1990). Pollen in the lower Mississippi River. Review of Palaeobotany and Palynology 64, 253-261. Dodson, J.R. (1983). Modern pollen rain in southeastern New South Wales, Australia. Review of Palaeobotany and Palynology 38, 249-268. Harden, G.J. (1992, 1993, 2000, 2002). ‘The Flora of New South Wales, Vol. 3, Vol. 4, Vol.1 (revised edition) and Vol. 2. (revised edition)’, respectively. (University of New South Wales Press: Sydney). Keith, D.A. and Benson, D.H. (1988). The natural vegetation of the Katoomba 1:100 000 map sheet. Cunninghamia 2(1), 107-145. Kershaw, A.P. and Strickland, K.M. (1990). A 10 year pollen trapping record from rainforest in northeastern Queensland, Australia. Review of Palaeobotany and Palynology 64, 281-288. Kodela, P.G. (1990). Modern pollen rain from forest communities on the Robertson Plateau, New South Wales. Australian Journal of Botany 38, 1-24. Moore, P.D., Webb, J.A. and Collison, M.E. (1991). ‘Pollen Analysis’. (Blackwell Scientific Publications, Oxford). PlantNet (2007). National Herbarium website (http:// plantNet.rbgsyd.nsw.gov.au) Accessed April 2007 Sainty, G.R. and Jacobs, S.W.L. (1981). “Waterplants of New South Wales’ (Water Resources Commission of N.S.W., Sydney) 550 pp. 121 MODERN POLLEN DEPOSITION IN THE BLUE MOUNTAINS APPENDIX. Species found in the vegetation on and around the swamps. Nomencla- ture follows Harden (1992; 1993; 2000; 2002) and Plantnet (2006). Vegetation map units are from Keith and Benson (1988) D, dominant. *, introduced species. BURRALOW CREEK SWAMP Species BRYOPHYTES Sphagnaceae Sphagnum sp. PTERIDOPHYTES AND ALLIES Adiantaceae Adiantum aethiopicum Blechnaceae Blechnum ambiguum B. cartilaginum Dennstaediaceae Pteridium esculentum Gleicheniaceae Gleichenia dicarpa G. microphylla Osmundaceae Todea barbara Selaginellaceae Selaginella uliginosa ANGIOSPERMS, DICOTYLEDONS Apiaceae Platysace ericoides P. lanceolata P linearifolia Xanthosia pilosa Apocynaceae Parsonsia straminea Araliaceae Polyscias sambucifolia Asteraceae Cassinia aculeata C. aureonitens Casuarinaceae Allocasuarina nana Ceratophyllaceae Ceratophyllum demersum Cunoniaceae Callicoma serratifolia Dilleniaceae Hibbertia acicularis H. bracteata 122 Wood- land 10ag Edge Mid swamp swamp 28a 28a + + + + + + + + + + Proc. Linn. Soc. N.S.W., 130, 2009 J.M. CHALSON AND H.A. MARTIN Elaeocarpaceae Elaeocarpus reticulatus Tetratheca thymifolia Ericaceae Epacris paludosa E. pulchella Leucopogon hookeri Euphorbiaceae Ampera xiphoclada Phyllanthus hirtellus Fabaceae, Faboideae Bossiaea obcordata Dillwynia floribunda D. retorta Gompholobium huegelii Pultenaea tuberculata Fabaceae, Mimosoideae Acacia falciformis A. myrtifolia A. obtusata A. ptychoclada A. terminalis Goodeniaceae Dampiera stricta Goodenia dimorpha G. heterophylla G. ovata Lamiaceae Prostanthera violacea Lauraceae Cassytha melantha Lobeliaceae Pratia purpurascens Loganiaceae Mitrasacme pilosa Meliaceae * Melia azedarach vat. australasica Menyanthaceae Villarsia exaltata Myrsinaceae Rapanea howittiana Myrtacae Angophora bakeri A. costata A. floribunda Proc. Linn. Soc. N.S.W., 130, 2009 + ~ - + + + ~ + + + + + + D 123 MODERN POLLEN DEPOSITION IN THE BLUE MOUNTAINS Baeckea linifolia Corymbia eximia Eucalyptus eugenioides E. multicaulis E. pauciflora E. radiata Kunzea capitata Leptospermum polygalifolium L. trinervium Melaleuca linariifolia Tristania neriifolia Oleaceae *Ligustrum sinense Notelaea longifolia Pittosporaceae Billardiera scandens Proteaceae Banksia ericifolia B. serrata Hakea teretifolia Lambertia formosa Persoonia laurina P. levis P. linearis P. mollis P. oblongata Petrophile pulchella Ranunculaceae Clematis aristata Rhamnaceae Cryptandra amara Rutaceae Eriostemon hispidulus Sapindaceae Dodonaea pinnata D. triquetra Stackhousiaceae Stackhousia viminea Thymelaeaceae Pimelea ligustrina Violaceae Viola hederacea ANGIOSPERMS, MONOCOTYLEDONS Cyperaceae Baumea juncea 124 + D D D D + D + + + + + Proc. Linn. Soc. N.S.W., 130, 2009 J.M. CHALSON AND H.A. MARTIN Baumea sp. ar ar Chorizandra sp. + zi Eleocharis sphacelata + + Lepidosperma longitudinale tt Schoenus sp. a Lomandraceae Lomandra glauca a L. longifolia + Phormiaceae Dianella caerulea =f Restionaceae Leptocarpus tenax ar Baloskion fimbriatum a5 Smilacaceae Smilax australis = S. glyciphylla +f WARRIMOO OVAL SWAMP Open Edge Mid PTERIDOPHYTES AND ALLIES Adiantaceae Adiantum diaphanum a Dennstaediaceae Pteridium esculentum als ais Gleicheniaceae Gleichenia dicarpa + + ANGIOSPERMS, DICOTYLEDONS Apiaceae Actinotus minor als Platysace lanceolata a5 at P linearifolia ar Ericaceae Brachyloma daphnoides a Dracophyllum secundum + Epacris paludosa a 7 Fabaceae, Faboideae Bossiaea heterophylla “F a *Cytisus scoparius ar Daviesia ulicifolia ate +r Dillwynia phylicoides a Gompholobium huegelii + + G. latifolium ae Hovea linearis + Proc. Linn. Soc. N.S.W., 130, 2009 125 MODERN POLLEN DEPOSITION IN THE BLUE MOUNTAINS Mirbelia rubifolia Fabaceae, Mimosoideae Acacia falciformis A. ptychoclada A. rubida A. terminalis Goodeniaceae Dampiera stricta G. ovata Lobeliaceae Pratia purpurascens Myrtacae Angophora bakeri Baeckea linifolia Eucalyptus notabilis E. pauciflora E. radiata Kunzea capitata Leptospermum grandifolium L. polygalifolium L. trinervium Polygalaceae Comesperma defoliatum C. ericinium Proteaceae Banksia ericifolia B. oblongifolia B. serrata Grevillea laurifolia G. mucronulata G. phylicoides Hakea salicifolia Isopogon anethifolius I. prostratus Persoonia laurina P. myrtilloides P. pinifolia Rutaceae Boronia microphylla Thymelaeaceae Pimelea glauca P. ligustrina Violaceae Viola hederacea 126 + +++ +++ + 4+ 4+ 4+ 4+ 4+ + + + + ~) Proc. Linn. Soc. N.S.W., 130, 2009 J.M. CHALSON AND H.A. MARTIN ANGIOSPERMS, MONOCOTYLEDONS Cyperaceae Baumea juncea ar Eleocharis sphacelata + Juncaceae Juncus remotiflorus ate Lomandraceae Lomanara filiformis ssp coriacea ae L. longifolia ai L. obliqua a5 Phormiaceae Dianella caerulea + Restionaceae Leptocarpus tenax +f ap NOTTS SWAMP Open -e/ « Mié Species forest swamp 10ar 26a PTERIDOPHYTES AND ALLIES Dennstaediaceae Pteridium esculentum a Gleicheniaceae Gleichenia dicarpa + Selaginellaceae Selaginella uliginosa +f ANGIOSPERMS, DICOTYLEDONS Apiaceae Actinotus forsythii a Platysace lanceolata + P linearifolia + Ericaceae Epacris paludosa a Lissanthe sapida ate Euphorbiaceae Poranthera microphylla ats Fabaceae, Faboideae Bossiaea heterophylla ste Phyllota squarrosa 7° Platylobium formosum + Fabaceae, Mimosoideae Acacia melanoxylon al A. obtusata =i A. obtusifolia a A. stricta a Proc. Linn. Soc. N.S.W., 130, 2009 WF MODERN POLLEN DEPOSITION IN THE BLUE MOUNTAINS Myrtacae Eucalyptus aggregata E. dives E. ligustrina E. pauciflora E. piperita E. sclerophylla Kunzea capitata Leptospermum juniperinum Proteaceae Banksia oblongifolia B. serrata Grevillea phylicoides Hakea teretifolia Isopogon prostratus Persoonia laurina P. linearis Petrophile pedunculata Rutaceae Boronia microphylla ANGIOSPERMS, MONOCOTYLEDONS Cyperaceae Baumea rubiginosa Carex sp. Gahnia sp. Iridaceae Patersonia sericea Juncaceae Juncus remotiformis Phormiaceae Dianella caerulea Poaceae Entolasia marginata Poa sp. Restionaceae Baloskion australe Leptocarpus tenax (Labill.) INGAR SWAMP Species GSeoeagvs + ++ + + + + + Open forest 10ar Wood- 6cTall Edge Mid PTERIDOPHYTES AND ALLIES Adiantaceae Adiantum aethiopicum 128 land open swamp swamp 10ag forest 26a 26a + Proc. Linn. Soc. N.S.W., 130, 2009 J.M. CHALSON AND H.A. MARTIN Dennstaediaceae Pteridium esculentum Dicksoniaceae Calochlaena dubia Gleicheniaceae Gleichenia dicarpa G. microphylla Osmundaceae Todea barbara Selaginellaceae Selaginella uliginosa ANGIOSPERMS, DICOTYLEDONS Apiaceae Actinotus forsythii Platysace lanceolata P. linearifolia Casuarinaceae Allocasuarina distyla Cunoniaceae Bauera rubioides Callicoma serratifolia Ceratopetalum apetalum Dilleniaceae Hibbertia acicularis Elaeocarpaceae Elaeocarpus reticulatus Ericaceae Brachyloma daphnoides Dracophyllum secundum Epacris paludosa Leucopogon esquamatus L. hookeri L. lanceolatus Lissanthe sapida Euphorbiaceae Ampera xiphoclada Fabaceae, Faboideae Bossiaea heterophylla B. obcordata Daviesis alata D. ulicifolia Dillwynia philicoides D. retorta Glycine clandestina Hovea linearis Proc. Linn. Soc. N.S.W., 130, 2009 + + 129 MODERN POLLEN DEPOSITION IN THE BLUE MOUNTAINS Phyllota phylicoides P. squarrosa Platylobium formosum Pultenaea divaricata P. flexilis P. incurvata P. tuberculata Fabaceae, Mimosoideae Acacia echinula . melanoxylon . obliquinervia . obtusata . obtusifolia . stricta DS TX JX EX Bx SS . suaveolens Goodeniaceae Dampiera stricta Goodenia bellidifolia G. dimorpha G. ovata Haloragaceae Gonocarpus chinensis ssp verrucosus G. longifolius Myrtacae Angophora bakeri Backhousia myrtifolia Baeckea diosmifolia Corymbia eximia Eucalyptus agglomerata E. dalrympleana E. dives E. obliqua E. oreades E. pauciflora E. radiata E. sieberi Kunzea capitata Leptospermum grandifolium L. juniperinum L. polygalifolium L. scoparium L. trinervium Melaleuca linariifolia Syncarpia glomulifera 130 + + + + + + + + + - + + + + + + + + + + + + + D + D + + + D Proc. Linn. Soc. N.S.W., 130, 2009 J.M. CHALSON AND H.A. MARTIN Proteaceae Banksia ericifolia B. oblongifolia 3 af B. serrata + + Grevillea aspleniifolia + G. laurifolia G. phylicoides =F Hakea propinqua H. sericea or H. teretifolia ai af Isopogon prostratus ai ar Lambertia formosa + Lomatia myricoides Persoonia acerosa ate P. laurina + P. levis ate P. linearis =F +r P. pinifolia 1° Petrophile pedunculata als Ranunculaceae Clematis aristata Rhamnaceae Cryptandra amara Rutaceae Boronia microphylla a Thymelaeaceae Pimelea ligustrina ANGIOSPERMS, MONOCOTYLEDONS Cyperaceae Baumea rubiginosa Carex sp. + aie Chorizandra cymbaria Eleocharis sphacelata cle Gahnia sieberana a Gahnia sp. Gymnoschoenus sphaerocephalus Lepidosperma longitudinale Iridaceae Patersonia sericea ar Juncaceae Juncus remotiformis Luzuriagaceae Eustrephus latifolius Phormiaceae Dianella caerulea ats Proc. Linn. Soc. N.S.W., 130, 2009 ‘31 MODERN POLLEN DEPOSITION IN THE BLUE MOUNTAINS Poaceae Entolasia marginata Poa sp. Restionaceae Baloskion australe Empodisma minus Leptocarpus tenax Smilacaceae Smilax australis KINGS TABLELAND SWAMP Species PTERIDOPHYTES AND ALLIES Dennstaediaceae Pteridium esculentum Gleicheniaceae Gleichenia dicarpa GYMNOSPERMS Cupressaceae Callitris muelleri ANGIOSPERMS, DICOTYLEDONS Apiaceae Actinotus forsythii Platysace lanceolata Casuarinaceae Allocasuarina distyla Allocasuarina nana Ericaceae Dracophyllum secundum Epacris paludosa Fabaceae, Faboideae Bossiaea heterophylla Daviesia alata D. ulicifolia Hovea linearis Phyllota squarrosa Pultenaea divaricata Fabaceae, Mimosoideae Acacia obtusata A. stricta A. suaveolens A. terminalis 132 Open forest 91 + + D D D + Open Edge Mid heath swamp swamp GE 26a 26a + Proc. Linn. Soc. N.S.W., 130, 2009 Myrtacae Corymbia eximia C. gummifera Eucalyptus deanei E. oblonga E. pauciflora E. piperita E. sclerophylla E. stellulata E. stricta Kunzea capitata K. ericoides Leptospermum grandifolium L. juniperinum L. polygalifolium Olacaceae Olax stricta Proteaceae Banksia ericifolia B. oblongifolia B. serrata B. spinulosa Grevillea phylicoides Hakea dactyloides H. salicifolia H. sericea Isopogon anemonifolius I. prostratus Lomatia silaifolia Persoonia laurina Petrophile pedunculata Thymelaeaceae Pimelea ligustrina J.M. CHALSON AND H.A. MARTIN ANGIOSPERMS, MONOCOTYLEDONS Lomandraceae Lomandra glauca KATOOMBA SWAMP Species BRYOPHYTES Dawsoniineae Dawsonia sp. Open forest Upper 91 Open forest Lower 91 Edge swamp 26a Proc. Linn. Soc. N.S.W., 130, 2009 MODERN POLLEN DEPOSITION IN THE BLUE MOUNTAINS PTERIDOPHYTES AND ALLIES Blechnaceae Blechnum cartilaginum Dennstaediaceae Pteridium esculentum Gleicheniaceae Gleichenia dicarpa Lycopodiaceae Lycopodium deuterodensum ANGIOSPERMS, DICOTYLEDONS Araliaceae Polyscias sambucifolia Asteraceae Arrhenechthites mixta Bracteantha bracteata Cunoniaceae Callicoma serratifolia Ericaceae Epacris paludosa Fabaceae, Faboideae Bossiaea rhombifolia Daviesia latifolia Fabaceae, Mimosoideae Acacia obtusata A. suaveolens Myrtacae Callistemon citrinus Eucalyptus obliqua E. oblonga E. sclerophylla E. squamosa Kunzea capitata K. ericoides Leptospermum polygalifolium L. trinervium Oleaceae *Ligustrum sinense Polygonaceae *Acetosella vulgaris *Rumex obtusifolius Proteaceae Banksia spinulosa Grevillea mucronata Isopogon prostratus Lomatia myricoides 134 + Proc. Linn. Soc. N.S.W., 130, 2009 J.M. CHALSON AND H.A. MARTIN Persoonia laurina Petrophile pedunculata Rutaceae Boronia microphylla ANGIOSPERMS, MONOCOTYLEDONS Cyperaceae Caustis flexuosa Juncaceae Juncus remotiformis Lomandraceae Lomandra obliqua Phormiaceae Dianella caerulea ae Poaceae Poa sp. NEWNES SWAMP Open Species rey 91 PTERIDOPHYTES AND ALLIES Blechnaceae Blechnum cartilaginum ata Dennstaediaceae Pteridium esculentum + Gleicheniaceae Gleichenia dicarpa a” ANGIOSPERMS, DICOTYLEDONS Apiaceae Platysace lanceolata F Asteraceae Arrhenechthites mixta ate Helichrysum scorpioides aL Olearia sp. aff. chrysophyplla ate Casuarinaceae Allocasuarina nana Dilleniaceae Hibbertia dentata ay Ericaceae Brachyloma daphnoides Epacris obtusifolia E. paludosa ae Lissanthe sapida fe Monotoca scoparia Proc. Linn. Soc. N.S.W., 130, 2009 Wood- land lla Edge swamp 20a + Mid swamp 20a jam ore) CA MODERN POLLEN DEPOSITION IN THE BLUE MOUNTAINS Euphorbiaceae Ampera xiphoclada Fabaceae, Faboideae Daviesis corymbosa D. ulicifolia Gompholobium grandiflorum G. latifolium Phyllota phylicoides P. squarrosa Platylobium formosum Fabaceae, Mimosoideae Acacia elata A. linifolia A. longifolia A. melanoxylon A. suaveolens Goodeniaceae Dampiera stricta Myrtacae Baeckea diosmifolia Eucalyptus acmenoides E. aggregata E. deanei E. notabilis E. oreades E. racemosa E. sclerophylla Kunzea capitata Leptospermum juniperinum L. polygalifolium Proteaceae Banksia spinulosa Grevillea acanthifolia G. aspleniifolia G. phylicoides Hakea salicifolia H. teretifolia Petrophile pedunculata Ranunculaceae Clematis aristata Rhamnaceae Cryptandra amara Rutaceae Boronia microphylla 136 + ++ 4 SGeevwes o + + + + D + + + D Proc. Linn. Soc. N.S.W., 130, 2009 J.M. CHALSON AND H.A. MARTIN Santalaceae Exocarpos strictus + Thymelaeaceae Pimelea glauca oF P. ligustrina a ANGIOSPERMS, MONOCOTYLEDONS Cyperaceae Lepidosperma laterale + Iridaceae Patersonia sericea + + Juncaceae Juncus remotiformis a Lomandraceae Lomanadra filiformis ssp coriacea tr or L. filiformis ssp filiformis ats L. glauca Ewart 4 Phormiaceae Dianella caerulea + Poaceae Entolasia marginata Poa sp. + Restionaceae Empodisma minus Leptocarpus tenax vi 1 Proc. Linn. Soc. N.S.W., 130, 2009 onounibre aonkvers ee iy cOVNA, cotbint aR, ae rancuelNR, hewel oomnly, eWlenincen: — ymooainnt eulrreng BUSH PANTY BIE a snethe ew rons) Silurian Rhynchonellide Brachiopods from Yass, New South Wales Desmonp L. Strusz Department of Earth and Marine Sciences, Research School of Earth Sciences, Australian National University, Canberra, Australia 0200 (dstrusz@ems.anu.edu.au), and Research Associate, Australian Museum, 6 College Street, Sydney, NSW 2010. Strusz, D.L. (2009). Silurian rhynchonellide brachiopods from Yass, New South Wales. Proceedings of the Linnean Society of New South Wales 130, 139-146. Rhynchonellide brachiopods are rare in the Silurian sequence at Yass. In this paper two species are described, one new species Agarhynchus australe being abundant at just one locality in the late Wenlock or earliest Ludlow Yass Formation. The other species, tentatively assigned to Tuvaerhynchus, is known from only a few specimens of late Wenlock to Ludfordian age. Manuscript received 19 October 2008, accepted for publication 21 January 2009. KEYWORDS: Agarhyncha australe, Ludlow, rhynchonellide, Silurian, Tuvaerhynchus, Wenlock, Yass INTRODUCTION Rhynchonellide brachiopods were recognised in early accounts of the stratigraphy of the Yass Syncline, but none has ever been described. Jenkins (1879, p. 26) recorded Rhynchonella from what (using modern terminology) was probably the basal Bowspring Limestone at locality GOUS7, and Mitchell (1887, p. 1201) listed the same genus from pebbles in the Sharpeningstone Conglomerate at Bowning (two specimens, described in this paper). Shearsby (1912, pp. 110-112) in his more detailed account of the succession north of Yass then noted the presence of possible Rhynchotreta and Camarotoechia at two localities, one within the Douro Volcanics, the other in the Yass Formation. The latter is in the same area along Derringullen Creek from which both of the species described in this paper were collected by Dr R.S. Nicoll and myself in 1982. However, other than at that locality, rhynchonellides are rare (only six usable specimens) in the Yass sequence. Only two taxa can be recognised. The first, Agarhyncha australe n. sp., occurs at only the one locality (on Derringullen Creek), just below the Cliftonwood Limestone Member of the Yass Formation, but is there in some numbers. Agarhyncha Havliéek, 1982, is otherwise known from the Wenlock and Ludlow of the Czech Republic. The other Yass rhynchonellide occurs in very low numbers at a few localities from the Yass Formation to the Yarwood Siltstone Member of the Black Bog Shale, and in pebbles in the Sharpeningstone Conglomerate. It is tentatively referred to the genus 7uvaerhynchus Kul’kov, 1985, from the Wenlock of Tuva. This raises some problems concerning provinciality which cannot be properly assessed until better material from Yass becomes available, enabling more confident identification. For a diagrammatic representation of Yass stratigraphy and ages, refer to Strusz (2002, fig. 1). Localities are detailed in that publication, with additions in Strusz (2003, 2005). SYSTEMATIC PALAEONTOLOGY Classification The classification followed is that of Savage et al. (2002). Measurements and symbols All linear measurements are in millimetres, and unless otherwise specified are as defined by Williams and Brunton (1997); the following symbols are used for these measurements: Ls, Ws, Ts — maximum shell length, width, thickness. Wh — hinge width. L(Wmax) — length to widest part of shell. SILURIAN BRACHIOPODS FROM YASS Repositories The repositories for the specimens studied are shown by the following prefixes to their catalogue numbers: AMF - macrofossil Museum, Sydney. ANU - Department of Earth and Marine Sciences (Research School of Earth Sciences), Australian National University, Canberra. CPC - Commonwealth Palaeontological Collection, Geoscience Australia, Canberra. collection, Australian Phylum BRACHIOPODA Class RHYNCHONELLATA Williams, Carlson, Brunton, Holmer and Popov, 1996 Order RHYNCHONELLIDA Kuhn, 1949 Superfamily RHY NCHOTREMATOIDEA Schuchert, 1913 Family TRIGONIRHYNCHIIDAE Schmidt, 1965 Subfamily TRIGONIRHYNCHINAE Schmidt, 1965 Genus Agarhyncha Havliéek, 1982 Type species Terebratula famula Barrande, 1847, by original designation; Ludlow, Bohemia. Diagnosis Subpentagonal to subcircular outline; biconvex to globose profile. Beak suberect to erect; foramen with minute deltidial plates. Fold and sulcus well defined, broad, anterior commissure uniplicate; tongue rectangular, serrate. Costae coarse, rounded, simple, but umbones smooth. Dental plates very short. Dorsal median septum thin; septalium with cover plate anteriorly; crura close to septum posteriorly (Savage p. 1052 in Savage et al. 2002). Agarhyncha australe sp. nov. Figs 1-5, Table 1 Diagnosis Relatively large biconvex species of Agarhyncha with smooth non-sulcate umbones, sulcus often weak anteriorly, ribs only moderately developed, medially concave dental plates, impressed ventral muscle field, raduliform crura, long dorsal median septum. Material Holotype CPC39529, paratypes CPC39530- 39592, all from locality GOU49. 140 Horizon Topmost O’Briens Creek Member, Yass Formation. Age Probably Homerian (late Wenlock), possibly earliest Gorstian (early Ludlow). Description Juvenile shells (taken as Ws <6.0 mm - see Fig. 4b) lenticular, biconvex to ventribiconvex, elongate lacriform to lozenge-shaped (mean juvenile Ls/Ws 1.10, mostly 1.0-1.2), generally relatively thin (mean juvenile Ts/Ws 0.43, mostly 0.35- 0.50). Adult shells (Ws >6 mm) subtriangular to subpentagonal, biconvex to slightly ventribiconvex, largest shells globose (mean adult Ts/Ws 0.55, max. 0.85). Maximum observed width 12.2 mm; length about equal to width (mean adult Ls/Ws 1.01, mostly 0.9 - 1.1). Dorsal fold and ventral sulcus appear at lengths of 3-4 mm, generally shallow, but variably developed anteriorly in larger shells; tongue when developed trapezoidal. Ventral beak suberect, sharp (especially in juveniles), usually small but in some shells extended posteriorly. Foramen mesothyrid (Fig. lh), delthyrium wide, deltidial plates narrow, disjunct. Umbones smooth, ribs appearing at Ls from 2.5 to 5 mm, initially faint. Ribs anteriorly rounded- angular, simple, generally low (especially laterally); margins of sulcus defined by pair of relatively well developed ribs, sulcus contains 1-3 ribs (2-4 on fold); 2-5 ribs on each flank. Shell generally thin-walled. Dental plates short, upright to gently convergent ventrally, somewhat concave medially. Ventral muscle field elongate, moderately impressed into slightly medially thickened shell, may be divided by very low myophragm. Dorsal median septum long (at least Ls/2), posteriorly supports small V-shaped septalium (Fig. 3) which is open posteriorly, covered mid-length to anteriorly (see Fig. 2, especially CPC39544, sections 1.2 to 1.8 mm). Outer hinge plates wide, flat in narrow zones between crural bases and inner socket ridges, moderately thick medially and generally strongly thickened beneath sockets. Sockets widely divergent, large; inner socket ridges robust, outer socket ridges merged with valve walls. Crural bases strong, triangular; crura calciform, curved somewhat towards ventral valve. No cardinal process. Remarks This form differs from leiorhynchids in its generally thin-walled shell which is mostly not globose, in its only moderately impressed ventral Proc. Linn. Soc. N.S.W., 130, 2009 DL. STRUSZ Figure 1. Agarhyncha australe; a-g, growth series of paratype shells in dorsal aspect, CPC39535, 39536, 39537, 39540, 39539, 39541, 39538; h-l, holotype CPC39529 in dorsal, lateral, ventral, posterior and an- terior aspects; m-p, paratype CPC39532 in dorsal, lateral, ventral and anterior aspects, a partly decorti- cated relatively wide shell with anteriorly well developed fold; q-s, paratype CPC39533 in dorsal, lateral and ventral aspects, a posteriorly decorticated shell with low convexity, few subdued ribs; t, paratype CPC39543, a large shell in ventral aspect, with 4 anteriorly strong ribs in sulcus; u-w, paratype CPC 39534 in dorsal, lateral and ventral aspects, a posteriorly decorticated large shell showing local crush- ing, presumably before lithification of the enclosing sediment; x, paratype CPC 39542, a large relatively wide shell in ventral aspect; y, paratype CPC39592, a dorsal internal mould (see Fig. 3). All x4, scale bar 5 mm. Locality GOU49, Yass Formation, O’Briens Creek Member immediately below Cliftonwood Limestone; probably Late Wenlock. Proc. Linn. Soc. N.S.W., 130, 2009 [4] SILURIAN BRACHIOPODS FROM YASS CPC39544 OOO OL Yi ie * U2 14 CPC39545 Figure 2. Agarhyncha australe; selected serial sections of paratypes CPC39544, 39545; distances from posterior ends in millimetres. Scale bar 2 mm. 2mm Figure 3. Agarhyncha australe; paratype dor- sal internal mould CPC39592 enlarged to show septalium and long but low median septum. Scale bar 2 mm. muscle field, and a cover plate on the septalium. From rhynchotrematids it differs in its smooth umbones, distinct dental plates and lack of a cardinal process. It shares important features with the Trigonirhynchiidae. Among trigonirhynchiids Astua Havlitek, 1992 (Lochkovian, Bohemia and central Asia) differs in stronger ribs, fold and sulcus, an emarginate anterior commissure, and internally in lacking a cover-plate on the septalium. Oxypleurorhynchia Plodowski, 1973 (Pridoli, Carnic Alps) is dorsibiconvex, with coarse ribs and pronounced fold and sulcus extending from the umbones. Virginiata Amsden, 1968 (Llandovery to Ludlow, N. America, China and Siberia) lacks fold and sulcus, but its ribs extend from the beaks; it also differs in being more elongate, having a posterior cover-plate on the very small septalium, robust cardinalia, and a short dorsal median septum. The new species is referred to the Bohemian Wenlock to Ludlow genus Agarhyncha on the basis of its 142 Ls = 1.08Ws Ts = O.48\/s Ws mm ) ead (CRs 4 8 12 Figure 4. Agarhyncha australe; a, length (Ls) and thickness (Ts) plotted against width (Ws). The di- vergence from the overall means at widths above 6 mm is just noticeable on these plots; b, thick- ness (Ts) plotted against width (Ws) on log-normal coordinates; in this plot the change in growth pa- rameters at a width of about 6 mm is quite clear. smooth umbones, short dental plates, and medially to anteriorly covered septalium. The Ludlow-age type species, Agarhyncha famula (Barrande, 1847) is smaller (Ws to c. 9.6 Proc. Linn. Soc. N.S.W., 130, 2009 D.L. 8= ~—L(Wmax) mm LiWmax) = 0.61Ls Figure 5. Agarhyncha australe; plot of length to greatest width (L(Wmax)) against length (Ls), showing only weak variability. mm), more globose, with in some cases anteriorly truncated margins, often posteriorly elongate ventral beak, stronger ribs which may be flattened and grooved marginally, shallower sulcus, ribs at least faintly developed umbonally, high dorsal median septum, and rod-like crura. The Wenlock species A. agason Havliéek in Havliéek and Storch, 1990 is of comparable size and outline, but has more and stronger, more angular ribs, especially medially. The other Bohemian Ludlow species, A. chuchlensis Havliéek in Havliéek and Storch, 1990 is wider (Ls/Ws_ 0.83-0.95), with generally subpentagonal outline, low, rounded beak, weakly ribbed umbones, anteriorly well developed fold and sulcus, more ribs, ventral muscle field not impressed but dorsal adductor field with fine lateral bounding ridges, rod-like crura, and somewhat shorter dorsal median septum. None of the Bohemian species shows medially concave dental plates. Family ORTHORHYNCHULIDAE Cooper, 1956 Genus Tuvaerhynchus Kulkov, 1985 Type species Tuvaerhynchus khalfini Kul’ kov in Kul’kov et al., 1985, by original designation; Wenlock, Tuva. Diagnosis Small with subpentagonal to subrectangular outline and dorsibiconvex profile. Beak suberect; delthyrium with disjunct deltidial plates. Fold and sulcus strong, narrow, well defined, from umbones; anterior commissure uniplicate; tongue high, trapezoid, dentate. Costae numerous, simple, angular. Dental plates short, vertical, close to valve wall. Septalium short, wide; hinge plates concave, slope medially; cardinal process septiform, thin; crura short, curved sharply ventrally (Savage p. 1081 in Savage et al. 2002). Proc. Linn. Soc. N.S.W., 130, 2009 STRUSZ Tuvaerhynchus? sp. Fig. 6, Table 2 Material Yass Formation: GOU47, CPC 39595 and 1 very uncertain fragment; GOU49, CPC 39596. Barrandella Shale Member, Silverdale Formation: GOU2a, CPC 39593. Lower Black Bog Shale: KF, ANU46537. Yarwood Siltstone Member, Black Bog Shale: GOU28, CPC 39594. Horizon uncertain: Bowning, “Upper Conglomerate”, Mitchell Collection AMEF28588, 133959 - presumably (following Mitchell 1887) from pebbles in the Sharpeningstone Conglomerate, derived from an older horizon. Stratigraphic distribution Yass Formation to Yarwood Siltstone Member, Black Bog Shale Age Late Wenlock? to early Ludfordian Description Available specimens are few, and mostly poorly preserved. Best are a steinkern from the Mitchell Collection, and a small shell from GOU47. Both are dorsibiconvex, with rounded outline; the small shell (CPC 39595, Ws 6.2 mm) is longer than wide (Ls/Ws ca 1.16), the steinkern (AMF28588, Ws ca 12 mm) transversely oval (Ls/Ws ca 0.8). They are globose - in both cases Ts/Ls is about 0.7. They are strongly ribbed, the ribs starting at the beaks. CPC 39595 has a shallow ventral sulcus with 3 ribs, there being 5 ribs on each flank. AMF28588 has a well developed fold and sulcus, forming a high trapezoidal tongue anteriorly; the sulcus contains 3 ribs, the flanks 5 ribs each, and the fold is formed of 2 ribs which split once. The other Mitchell Collection specimen, AMF133959, is an incomplete flattened internal mould with 4 ribs in the sulcus, 6 on each flank. Inter- rib furrows extend as short marginal spines. None of the specimens shows clear details of the ventral beak, and so the presence and nature of a delthyrium cannot be demonstrated. Large teeth are supported by fairly short but distinct dental plates which are somewhat convergent towards the valve floor. Details of the ventral muscle field are not known. Dorsal median septum long, low, fine, continuous with linear cardinal process which arises from a shallow septalium which is either sessile or nearly so. Crural bases robust, crura unknown. Discussion Among Silurian rhynchonellides, the general SILURIAN BRACHIOPODS FROM YASS Ls Ld Ws CEE39529"; OF me SION OF CPC39530 5975. Se, (6.0, CPC39534 9.0 - 8.9 CPC39538 Be) a SM) CPC39543 9:9 - 9.6 Ts L(Wmax) Ls/Ws _ Ts/Ws 3.3 3.4 0.97 0.52 39) 3.3 0.98 0.65 5.0 4} On 0.56 es) 4.0 1.18 0.50 - De) 1.03 - Table 1: Agarhyncha australe: dimensions in mm and proportions of holotype (*) and selected para- types. Measurements in italics are best estimates for damaged specimens. Ls Ws Ts AMEF28588 DT NZ FO) CPC39593 NOS Qn 8 CPC39595 TO re ANU46537 iS Sai) - L(Wmax) Ls/Ws _ Ts/Ws 4.8 0.81 0.58 55 HT 0.56 4.0 LAG 0.82 55) 0.94 - Table 2. Tuvaerhynchus? sp.: dimensions in mm and proportions of selected specimens. Measurements in italics are best estimates for damaged specimens. shell form and strong simple ribbing of this form, coupled with distinct but short dental plates and a linear cardinal process on a sessile or near-sessile septalium, points to the Orthorhynchulidae (whose genera are also united by possessing an open or near- open delthyrium). Orthorhynchula Hall and Clarke, 1893, has dental plates fused to the valve walls, and a low fold. The Tasmanian Ordovician Tasmanella Laurie, 1991, has a high fold, but differs in its fused dental plates, and a short, high dorsal median septum supporting a raised septalium. Twvaerhynchus is closest morphologically, but in the absence of details of delthyrium, deltidial plates, and crura, generic identity cannot be certain. In the absence of that certainty, palaeobiogeographic speculation on this possible link between the Tuvaella and Retziella Faunas of Rong et al. (1995), and thus the Mongolo- Okhotsk and Sino-Australian Provinces of the Uralian-Cordilleran Region, is pointless. ACKNOWLEDGEMENTS I thank Jan Percival and Norman Savage for reviewing this paper, and Prof. Brian Kennett, Director of the ANU Research School of Earth Sciences, for continuing provision of facilities in the Department of Earth and Marine Sciences within that School. Serial sectioning was made possible by the loan of a Croft Parallel Grinder from the Geological Survey of New South Wales. Photography by H.M. Doyle (formerly of Geoscience Australia) and the author. 144 REFERENCES Amsden, T. W. (1968). Articulate brachiopods of the St. Clair Limestone (Silurian), Arkansas, and the Clarita Formation (Silurian), Oklahoma. Paleontological Society Memoir 1. Barrande, J. (1847). Uber die Brachiopoden der silurischen Schichten von Bohmen. Naturwissenschaftlichen Abhandlungen. I Band. W. Haidinger, Wien, 357- 475. [not seen, fide Havliéek, 1982]. Cooper, G.A. (1956). Chazyan and related brachiopods. Smithsonian Miscellaneous Collections 127. Hall, J, and Clarke, J.M. (1893). An introduction to the study of the genera of Palaeozoic Brachiopoda. Palaeontology of New York 8 (2), 1-317. Havli¢ek, V. (1982). New genera of rhynchonellid and camerellid brachiopods in the Silurian of Bohemia. Véstnik Ustredniho tistavu geologického 57, 365- 372. Havlicek, V. (1992). New Lower Devonian (Lochkovian- Zlichovian) rhynchonellid brachiopods in the Prague Basin. Sbornik Geologickych Véd, Paleontologia 32, 55-122. Havliéek, V. and Storch, P. (1990). Silurian brachiopods and benthic communities in the Prague Basin (Czechoslovakia). Rozpravy Ustredniho Ustavu Geologického 48. Jenkins, C. (1879). On the geology of Yass Plains. Proceedings of the Linnean Society of New South Wales, series 1, 3, 21-32. Kuhn, O. (1949). ‘Lehrbuch der Palaozoologie’. (E. Sweizerbart’sche Verlagsbuchhandlung, Stuttgart). Proc. Linn. Soc. N.S.W., 130, 2009 DL. STRUSZ Figure 6. Tuvaerhynchus? sp.; a-e, CPC39593, a slightly crushed lenticular shell in dorsal, lateral, ven- tral, posterior and anterior aspects (locality GOU2a, Barrandella Shale Member, Silverdale Fm, late Gorstian); f-j, CPC39595, a shell in dorsal, lateral, ventral, posterior and anterior aspects, the dorsal umbo worn and revealing the median septum (locality GOU47, topmost O’Briens Creek Member, Yass Fm, probably Late Wenlock); k-o, AMF28588, a wide and very globose steinkern in dorsal, lateral, ven- tral, posterior and anterior aspects, the ventral beak broken (Mitchell Collection, from pebble in Sharp- eningstone Creek Conglomerate); p, AMF133959, somewhat crushed ventral internal mould (source as AMEF728588). All x4, scale bar 5 mm. ee Bl A Proc. Linn. Soc. N.S.W., 130, 2009 SILURIAN BRACHIOPODS FROM YASS Kul’kov, N.P., Vladimirskaya Ye.V. and Rybkina N.L. (1985). Brakhiopody i biostragrafiya verkhnego ordovika 1 silura Tuvy. Trudy Institut Geologii i Geofiziki, Sibirskoye otdeleniye Akademiya Nauk SSSR 635. [Russian] Laurie, J.R. (1991). Articulate brachiopods from the Ordovician and Lower Silurian of Tasmania. Memoir of the Association of Australasian Palaeontologists 11, 1-106. Mitchell, J. (1887). On some new trilobites from Bowning, N.S.W. Proceedings of the Linnean Society of New South Wales, series 2,2, 435-440. Plodowski, G. (1973). Revision der Brachiopoden-Fauna des Ober-Siluriums der Karnischen Alpen, 2. Rhynchonellacea aus den Zentralkarnischen Alpen. Senckenbergiana lethaea 54, 65-103. Rong J.-Y., Boucot, A.J., Su, Y.-Z. and Strusz, D.L.,1995. Biogeographical analysis of Late Silurian brachiopod faunas, chiefly from Asia and Australia. Lethaia 28, 39-60. Savage, N.M., Mancefiido, M.O., Owen, E.F., Carlson, S.J., Grant, R.E., Dagys, A.S. and Sun D.-L., 2002. Rhynchonellida. In ‘Treatise on Invertebrate Paleontology, Part HH, Brachiopoda, revised, volume 4: Rhynchonelliformea (part)’ (Ed. R.L. Kaesler), pp. 1027-1376. (Geological Society of America, Denver and University of Kansas Press, Lawrence). Schmidt, H. (1965). Neue Befunde an Paldozoischen Rhynchonellacea (Brachiopoda). Senckenbergiana Lethaea 46, 1-25. Schuchert, C. (1913). Class 2. Brachiopoda. In ‘Text- book of Palaeontology, volume 1, part 1, edition 2, translated and edited by Charles R. Eastman’ (Ed. K.A. von Zittel) pp. 355-420. (MacMillan and Co., Ltd, London). Shearsby, A.J. (1912). The geology of the Yass district. Reports of the 13" Meeting of the Australasian Association for the Advancement of Science, Sydney 1911, 106-119. Strusz, D.L. (2002). Brachiopods of the Orders Protorthida and Orthida from the Silurian of the Yass Syncline, southern New South Wales. Alcheringa 26, 49-86. Strusz, D.L. (2003). Late Silurian strophomenate brachiopods from Yass, New South Wales. Alcheringa 27, 1-35. Strusz, D.L. (2005). Late Silurian pentameride brachiopods from Yass and Molong, New South Wales. Alcheringa 29, 205-228. Williams, A. and Brunton, C.H.C. (1997). Morphological and anatomical terms applied to brachiopods. In “Treatise on Invertebrate Paleontology, Part H, Brachiopoda, revised, volume 1: Introduction’ (Ed. R.L. Kaesler) pp. 423-440. (Geological Society of America, Boulder, and University of Kansas Press, Lawrence). Williams, A., Carlson, S.J., Brunton, H.C., Holmer, L.E. and Popov, L. (1996). A supra-ordinal classification of the Brachiopoda. Philosophical Transactions of the Royal Society, London, Series B, 351, 1171-1193. 146 Proc. Linn. Soc. N.S.W., 130, 2009 Cortinarius Fr. Subgenus Cortinarius in Australia A.E.Woop: School of Biological, Earth and Environmental Sciences, University of New South Wales, UNSW Sydney, NSW, 2052, Australia. Wood, A.E. (2009). Cortinarius Fr. subgenus Cortinarius in Australia. Proceedings of the Linnean Society of New South Wales 130, 147-155. Three new species within Cortinarius subgenus Cortinarius from Australia are described, each belonging near a different species, but differing significantly from the type variety in all cases. They represent distinct species — C. jenolanensis, C. kioloensis and C. hallowellensis. Manuscript received 15 October 2008, accepted for publication 4 February 2009 Keywords: agarics, Cortinarius, distribution, mushrooms, new varieties, toadstools INTRODUCTION Cortinarius subgenus Cortinarius is characterised by the presence of fleshy carpophores, with a cap that is frequently squamulose, large conspicuous cheilocystidia and vacuolar, mostly violet, pigments. The spores show both a suprapilar plage, usually flattened and often more or less smooth. There have been scattered records of this subgenus, particularly C. violaceus from Australia. This species was reported from Victoria by Cooke (1892) and this report was carried forward by McAlpine (1895) and Brittlebank (1940). Cleland (1933, 1934) did not record the species, nor did Grgurinovic (1997) record it from South Australia. Shepherd and Totterdell (1988) recorded the species from the Australian Capital Territory, New South Wales and Victoria. Young (1994) also recorded the species from New South Wales and Victoria. This species was also recorded from Western Australia by Griffiths (1985), Hilton (1988) and Syme (1992) and more recently was fully described by Bougher and Syme (1998). All these records are for C. violaceus, in some cases with uncertainty being expressed as to whether the collections are identical with the European species. Recently a new species, Cortinarius austroviolaceus has been described from Tasmania by Gasparini (2001). There have been some recent studies on C. violaceus in Europe and now two species are widely recognised, C. violaceus and C. hercynicus (Brandrud 1983; Brandrud et al. 1989-1998). The study by Moser (1986) of some collections from the SW-Pacific area has added four more species to the subgenus C. atroviolaceus, C. subcalyptrosporus, C. atrolazulinus and C. paraviolaceus. In view of the diversity of taxa of the subgenus in the SW Pacific, the suggestion has been made that they represent the descendants of a Gondwanan species of possibly ancient origin (Gasparini, 2001). However the subgenus has not been reported from Tierra del Fuego (Horak 1979) or in other areas of South America (Moser and Horak 1975). Cortinarius violaceus s.s. Montagne, (from Chile, see Horak, 1979) is a different, unrelated species, Cortinarius gayii Horak (see Horak, 1979, p. 396 with full description). There has been considerable discussion over many years as to whether Cortinarius violaceus is a single species in Europe or whether several taxa at some close level (species, subspecies or variety) are involved. Some claim that over a large number of collections, a continuous variation can be found between the two main forms. However many now recognise two distinct forms, though the level at which they should be considered is also disputed. The view taken here (following Moser (1983), Horak (2005), Breitenbach & Kranzlin (2000) and Knudsen &Vesterholt (2008)) is to recognise two separate species from Europe as follows : Cortinarius violaceus with spores (12)13-16(17) x 7-8(8.5) um, elliptic to amygdaliform, verrucose, cap mostly 6-14 cm, under deciduous woods; Cortinarius hercynicus with spores (12)13- 16(17) x 7-8(8.5) um, broadly ellipsoid to subglobose, strongly verrucose, cap mostly 5-10 cm, under coniferous woods (spruce, pine, sometimes mixed woods). CORTINARIUS Fr. SUBGENUS CORTINARIUS IN AUSTRALIA Most records are only from the latter part of the twentieth century (May and Wood, 1997). The records are probably accurate because of the distinctive characteristics of Cortinarius violaceus s.1., but they give no information as to which of the currently reported species are intended. Later records indicate that the subgenus is widespread throughout most of Australia, but that it is not collected frequently. Studies of DNA sequences of various species of Cortinarius concluded that there were grounds for considering the creation of two separate genera (Hoiland and Holst-Jensen, 2000). A later study of DNA sequences for a large range of Cortinarius species (Garnica et al, 2005) supported the Cortinarius clade, without any further additions of any closely related groups or species. Bougher and Syme (1998) used the epithet C. violaceus with some reservations for their local collections. Chambers et al. (1999) compared DNA from New South Wales material with reported sequences from Northern Hemisphere collections of Cortinarius violaceus, and reported that the local material while close, belonged to a different taxon and noted ‘a careful revision of Australian Cortinarius violaceus collections is clearly required’. Unfortunately, voucher material of these collections has not yet been available. Examination of material from mainland Australia has demonstrated close similarities to the European species but with some clear differences. All the Australian material does not belong to a single species but represents four different taxa of which three are new. The differences described below clearly indicate three distinct taxa, related to previously described species. The differences are sufficient to require the creation of three new species. MATERIAL AND METHODS Material was mounted in 5% KOH solution and stained with Congo Red. Specimens are housed in the J.T. Waterhouse Herbarium, University of New South Wales (UNSW), except for Western Australian material, which is in the Western Australian Herbarium (PERTH). The collections at UNSW all have extensive field notes and colour photographs taken under standard conditions. Spore measurements indicate the range of sizes found in the various collections. Where spore sizes are included in brackets, they indicate that the spore sizes were more than one measuring unit (0.3 um) beyond the range for all other spores. The value Q 148 represents the mean length:breadth ratio of the spores. Measurements of Q were averaged for a collection and where a range is quoted it represents the range across collections. Measurements of the spores exclude the apiculus and the ornamentation. Measurements of cystidia indicate length and maximum width. Measurements of the basidia exclude the sterigmata. Colours are usually followed by an annotation from Maerz and Paul (1950) and have a format such as 10D3. All colour comparisons were made under natural light. The figures show the microscopic features at standard magnification: spores x2000, cystidia and basidia x1000. The scale bar represents 10 um at x2000 magnification. Key to the SW-Pacific species of Cortinarius subgenus Cortinarius 1. Average basidiospore length less than 10 um, cheilocystidia not capitate. . Beate nad 1* Average basidiospore jenothyeh more than 10) um 2. Cheilocystidia 50-140 x 10-25 um, pleurocystidia scarce, 40-100 x 10-18 um, lanceolate. UGeIOee A. Seon. C. atroviolaceus 2* Cheilocystidia 30-48 x 4-7 um, pleurocystidia ABSENT, . PATNI... .OaGee 1.C. jenolanensis 3. Cheilocystidia capitate..........C.austroviolaceus 3* Cheilocystidia not capitate or absent...............4 4. Spores with visible perispore ..............::cceeeceeeee Pe tah teeth C. subcalyptrosporus 4* Spores without visible perispore....................5 5. Cheilocystidia absent, pleurocystidia rare...... ...C. paraviolaceus SiCheilocystidiaipresentss-..ce-eeee a ee ee 6 6. Spores large, at least up to 12 um ee ee up to 16 um in length... wee es 6* Spores smaller, at most up 1b 12 um lene 2 slender, Q 1.86, cheilocystidia lageniform, 45- TOK 12-20 Ree. be eee eC: aly Olagul nus 7. Spores ellipsoid to amygdaliform................. 8 7* Spores broadly ellipsoid to subglobose........... 9 8. Spores elongate Q=1.87, width narrow, 6.3- 7.5 um; cheilocystidia 50-60 x 10-12 um, pleurocystidia frequent, similar.................. Eee ...2.C. hallowellensis 8* spore ptortee Q- 1. 56, rid broader 7.5-8.5 um; cheilocystidia 35-80 x 15-25 um, pleurocystidia frequent, similar..... C. violaceus 9. Spores 11-13 x 8-9 um, Q=1.45; cheilocystidia 55-80 x 14-19 um, lasenitornas..2) 22 9oe, Aon. ae C. hercynicus Proc. Linn. Soc. N.S.W., 130, 2009 A.E. WOOD 9* Spores 11-14 x 8.1- 9.3 um, Q=1.40; cheilocystidia 45 — 120 x 14-17 um, lageniform PE SOMO are, A LUd, OLA eat 3.C. kioloensis 1. Cortinarius jenolanensis Wood, sp. nov. (Fig. 1: a-d) Pileo usque ad 4 cm lato, convexo, demum plano, obscure violaceo, sicco, subtiliter fibrillo-squamoso. Lamellis obscure violaceis, brunnescentibus. Stipite ell 5-6 cm longo, 5-8 mm crasso, sicco, appresse fibrilloso, pallidiori violacea. Sporis 8.4 — 10.2 x 5.7- 6.9 um, Q=1.55, ellipsoideis, subtiliter verrucosis, cheilocystidiis sparsis, lageniformis 30-48 x 10- 14 wm, absentibus pleurocystidiis, absentibus pileocystidiis. Hyphis fibuligeris. Habitato in humo in silvis Eucalyptus mixtis. Pileus to 4 cm, hemispherical at first, then convex to flat convex and finally plane, very finely to a little coarsely radially fibrillose, deep violet, dry, not i! Figure 1. Cortinarius jenolanensis (UNSW 88/107) : a. basidiome (x 1); b.spores; c.basidia; d. cheilo- cystidia; Cortinarius kioloensis (UNSW 83/781) e. basidiome( x 1) f. pileocystidia. Proc. Linn. Soc. N.S.W., 130, 2009 CORTINARIUS Fr. SUBGENUS CORTINARIUS IN AUSTRALIA hygrophanous. Lamellae broadly adnate to slightly decurrent, thin, crowded, with one to two series of lamellulae, deep violet then deep ferruginous, margin concolorous. Stipe 50-60 x 5-8 mm, central, firm to tough, equal to slightly swollen below, sometimes slightly tapering at the base, upper part cap coloured or slightly paler, lower part a little paler with base pale violet, with no obvious basal mycelium, and no clear zone of velar remains. The only velar remains were a few scattered appressed fibrils throughout with only small areas or groups. Aroma There is no apparent aroma. Spores 8.4-10.2 x 5.7-6.9 um, mean 9.44 x 6.09 um, mean Q = 1.55, oval, suprahilar depression not clearly present and not clearly smooth, ornamentation low to very low, a little blunt, not anastomosing. Basidia 25-32 x 11-14 um, clavate, four-spored; clamp connections present. Cheilocystidia fairly sparse, variously lageniform (some somewhat irregular) 30-48 x 10-14 um, pleurocystidia absent. Pileal cuticle a loose layer of narrow hyphae, each 4-7 um diameter, not encrusted with pigment, mainly radially arranged and repent, a few a little irregularly loose and more or less upright with rounded terminal cells but not specialised as pileocystidia. Below this layer was a densely packed layer of parallel hyphae, the layer about 40-50 um thick with individual hyphae of 4-10 um diameter. Below this layer was a layer of interwoven hyphae, somewhat compact, of pale golden hyphae with individual hyphae of 5-8 um diameter. Habitat On soil in eucalypt sclerophyll forest. Commentary This species is different from all the species described by Moser (1987) because of the smooth pileus, different structure of the cuticle, absence of pleurocystidia, without amorphous deposits and also by being of smaller general size and lacking aroma. It is close to the typical forms of Cortinarius atroviolaceus but differs in having slightly smaller spores which are more finely rough and lack a clearly visible plage, the complete absence of pleurocystidia and smaller cheilocystidia. It may be that Corner Collection RSNBB 5258B, noted by Moser(1987), which has finer ornamentation on the spores and smaller cheilocystidia, also represents this species. Cortinarius austroviolaceus is also close, but that 150 species has cheilocystidia that are regularly slightly capitate and are more variable otherwise, and it also has a different cuticle with occasional lanceolate (lageniform) terminal cells. (See Moser 1987, pp 139,140). Material Examined NSW : Jenolan Caves, Binda Cabins, Eucalypt woodland, 30.4.88, A.E.Wood et al. (UNSW 88/107) Holotype; ACT, Canberra, Tidbinbilla Nature Reserve, Eucalypt woodland, 16.5.92, A. E. Wood et al. (UNSW 92/121). 2. Cortinarius kioloensis Wood, sp. nov. (Fig. 1: e,f; 2: a-c) Pileo usque ad 6 cm lato, convexo, demum plano, obscure violaceo, sicco, fibrilloso-squamoso. Lamellis obscure violaceis, brunnescentibus. Stipite 8-12 cm longo, 15 mm crasso, basi clavatus usque ad 30 mm crasso, sicco, appresse fibrilloso, pallidiori violacea. Sporis 11.1 - 13.5 x 8.1 — 9.3 (10.5) um ellipsoideis. verrucosis, cheilocystidiis lageniformis, 45-120 x 14 —19 um, pleurocystidiis sparsis, lageniformis 45-113 x 15 —26 um, pileocystidiis cylindricis vel fusiformis 35-60 x 13 —28 um. Hyphis fibuligeris. Habitato in humo silvis Eucalyptus mixtis. Pileus to 6 cm diam., rounded convex at first, then rounded umbonate to convex, finally almost plane with age, strongly fibrillose to a little tomentose to finely squamulose, more adpressed with age, deep violet (48H11—12), becoming blackish with age, dry, not hygrophanous. Lamellae narrowly to broadly adnate to slightly sinuate, thin to moderately thick, somewhat spaced, one or two sets of lamellulae, dark violet at first, then gradually deep ferruginous, margin concolorous. Stipe central, firm, solid, bulbous at base, 8-12 x 1.5 cm, base 3 cm, dry, mostly with clear fibrillar velar zone and scattered fibrils below, violet above, somewhat paler than cap, a little paler below (to 46E6 - 17E4), basal bulb globose, concolorous. Flesh whitish to pale violet, outer layer of stem dark violet, deep violet at apex of stipe. KOH (5%) on cap bright red. Aroma Clearly absent even when quite young and fresh; one collection with slight aroma of wood shavings (but not camphor wood). Spores 11.1-13.5 x 8.1-9.3 (10.5) um, mean 12.5 x 8.8 um, Q = 1.37-1.46, grand mean Q = 1.42, Proc. Linn. Soc. N.S.W., 130, 2009 A.E. WOOD c Figure 2. Cortinariu kioloensis (UNSW (83/781) : a. spores; b. basidia; c. cheilocystidia ovoid to elliptic, suprahilar depression not marked but present in some cases, but not clearly smooth, ornamentation moderate, coarse, blunt, with some slight anastomosing. Basidia 35—50 x 10—12 um, four- spored; clamp connections present. Cheilocystidia abundant, ventricose to lageniform, 45-120 x 14-19 uum; pleurocystidia sparse but clearly present, similar to cheilocystidia, but with some a little fusoid, 45-113 x 15—26 um. Pileal cuticle a layer of loose hyphae with upturned terminal cells which are somewhat inflated, swollen or cylindrical, 35-60 x 13-28 um; subcuticular layer of subcellular cells, 30-40 um diameter, walls not coloured, below this a narrow Proc. Linn. Soc. N.S.W., 130, 2009 layer of somewhat inflated, closely packed hyphae, 20-25 um diameter, with coloured contents, below this the context was of loosely arranged somewhat inflated hyaline hyphae, 15—25 um diameter. Habitat On soil in eucalypt sclerophyll forest. Commentary This species is different from the typical forms of Cortinarius violaceus and C. hercynicus and from all the other species described by Moser (1986). It is distinct because of the different habit, absence of CORTINARIUS Fr. SUBGENUS CORTINARIUS IN AUSTRALIA aroma, relative scarcity of pleurocystidia, presence of pileocystidia and spores which are without a well- differentiated plage and have less well developed wall ornamentation. There are also some slight differences in the size and shape of the spores. In this species, the size and shape are nearer to that found in Cortinarius hercynicus rather than that found in Cortinarius violaceus but the shape seems distinctly different from that of spores of Cortinarius hercynicus in that the spores are broadly ellipsoid rather than distinctly amygdaliform. Because of all these features it is regarded as a distinct taxon and is described as a new species of Cortinarius near to C. hercynicus. Material Examined NSW: Sydney, Scotland Island, Eucalypt woodland, 22.6.80, S. Lowry, (UNSW 80/268); Batemans Bay, Kioloa State Forest, Eucalypt woodland, 19.5.83, A. E. Wood & J. J.Bruhl, (UNSW 83/781) Holotype; Sydney, Royal National Park, Eucalypt woodland, 5.6.83, F. K. Taeker, (UNSW 83/923);Batemans Bay, Kioloa State Forest, Higgins Creek, Eucalypt woodland, 15.5.84, A. E. Wood & N. B. Gartrell, (UNSW 84/495); Sydney, Royal National Park, Couranga Track, Eucalypt woodland, 28.5.86, F. K. Taeker, (UNSW 86/254); Sydney, Boronia Park, Eucalypt woodland, 27.5.90, R. Kearney, (UNSW 90/197); Hazelbrook, James Park, Eucalypt woodland, 30.5.92, A.E.Wood et al., (UNSW 92/206); Springwood, Sassafras Gully, Eucalypt woodland, 16.4.94, A. E. Wood et al.,(UNSW 94/47); Sydney, Sydney Harbour National Park, Bradleys Head, 7.6.98, B. J.& N. W. Rees, (UNSW 98/25); Sydney, Lane Cove Bushland Park, Gore Creek, Eucalypt woodland, 7.6.98, B. J. & N.W. Rees, (UNSW 98/28). Authentic material from Sweden (Femsjo) was collected and at first was identified as Cortinarius violaceus. However later detailed examination clearly showed that it was a typical example of Cortinarius hercynicus and the following microscopic details are added for this collection (as Cortinarius hercynicus var hercynicus) Spores 12.6-15.0 x 8.4-9.3 um, mean 13.47 x 8.94 um, Q=1.51, spores elliptic, only vaguely amygdaliform, with only some spores showing a slightly flatter supra-hilar depression, but that mostly not smooth, ornamentation moderate, a little broad and only slightly blunt. Cheilocystidia frequent 75-85 x 13-19 uum, narrowly lageniform, pleurocystidia sparse but clearly present, lageniform, somewhat more variable, 152 50-90 x 12-20 um. Pileal cuticle of closely packed and interwoven hyphae, layer 100-200um deep, individual hyphae 5-7 um diameter, without any terminal cystidia (Fig. 3). Material Examined : SWEDEN: Femsjo, woodland, 2.9.79, M.M.Moser & A.E.Wood, in UNSW(UNSW 79/29). 3. Cortinarius hallowellensis Wood, sp. nov. (Fig. 4) Pileo usque ad 6 cm lato, convexo, demum plano, obscure violaceo, sicco, subtiliter fibrilloso-squamoso. Lamellis obscure violaceis, brunnescentibus. Stipite cylindrico vel clavato, 4-7 cm longo, 10-15 mm crasso, basi leviter, sicco, fibrilloso violacea. Sporis 11.1- 12.0 x 6.3-7.5 um, ovoideo-ellipsoideis, verrucosis, cheilocystidiis fusiformis vel lageniformis, 50-60 x 9-13 um, pleurocystidis fusiformis, 50-60 x 9-13 um, absentibus pileocystidiis. Hyphis fibuligeris. Habitato in humo in silvis Eucalyptus mixtis. Pileus to 3.4-6.0 cm, rounded convex at first, flattening at maturity, finely radially fibrillose, very dark violet brown (16F4), not hygrophanous. Lamellae broadly adnate to adnate, thin, a little spaced, dark violet (16B5), more rusty with age, with two series of lamellulae. Stipe cylindrical to clavate, with a swollen base 3.7-7.0 x 1.0-1.5 cm, solid, dry, dark violet (16B4) with fine cobweb veil, rapidly disappearing (after Bougher & Syme 1988, colours from Kornerup & Wanscher, 1978). Spores 11.1-12.0 x 6.3-7.5 um, mean 11.49 x 6.81 um, mean Q = 1.69, oval to elliptic, occasionally vaguely amygdaliform, with occasionally a slight supra-hilar depression, but not visibly smooth, ornamentation moderate, coarse, blunt. Basidia cylindrical to clavate, 40-55 x 10-12 um, four-spored, clamp connections present. Cheilocystidia plentiful, narrow lageniform to fusoid, 50-60 x 9-13 um, pleurocystidia sparse, but clearly present, similar to cheilocystidia, but mostly fusiform 50-60 x 10-12 um. Pileal cuticle with a surface layer 35 -50(80) um deep, a thin layer of loosely arranged hyphae, individual hyphae 2.5- 5 um diameter, mainly repent, with no erect hyphae and no differentiated terminal cells, without wall encrustation, some walls with pale golden walls; below this a layer of closely packed cylindrical hyphae of the trama (35-50 x 7-10 um, some a little larger and a few pseudoparenchymatous cells present). Proc. Linn. Soc. N.S.W., 130, 2009 A.E. WOOD Figure 3. Cortinarius herycynicus (UNSW 79/29) : a. spores; b. basidia; c. cheilocystidia; d. pleuro- cystidia Commentary This species is different from the typical Cortinarius violaceus in that this species has oval to elliptic spores (Q = 1.69), rather than amygdaliform spores, the cuticle does not produce pileocystida, the cheilocystidia are narrower to fusiform and the general habit is much smaller. Hence it is regarded as a close, but distinct species. Material Examined WA: Denmark, Mount Hallowell Reserve, Proc. Linn. Soc. N.S.W., 130, 2009 Eucalypt woodland, 22.5.93, K. Syme. (PERTH 0550 6794), Holotype. Collection PERTH007775665 also seems to be this species. However it was collected in a Pinus radiata plantation. It has spores with size 12-13.8 x 6.6- 7.5 um, mean 13.14 x 7.02 um, Q = 1.87, spores ovoid to elliptic, some vaguely amygdaliform, supra-hilar depression sometimes slightly present, but never clearly smooth. Cheilocystidia abundant, narrow lageniform to narrow fusiform 85-110 x 10- 153 CORTINARIUS Fr. SUBGENUS CORTINARIUS IN AUSTRALIA int oD Oo d Figure 4. Cortinarius hallowellensis (PERTH 0550 6794) : a. spores; b. basidia; c. cheilocystidia; d. pleurocystidia. 12 um, pleurocystidia abundant, narrow fusiform of the same dimensions. Pileal cuticle a thin, scarcely differentiated layer 20-30 wm deep, composed of narrow hyphae, 2-5 um diameter, the surface slightly more loosely arranged, but with no special terminal cells and no upturned cystidia and then the underlying tissues gradually becoming more densely packed. This collection has slightly larger spores and slightly longer cystidia, but does not otherwise differ from the previous collection. In the absence of further collections, this is left as another collection of Cortinarius hallowellensis. This leaves the question as to whether this form is a local form which has transferred to the introduced host or whether it was introduced with the exotic species, and may occur elsewhere. Much more extensive collecting may allow this question to be answered. 154 Material Examined WA:. North of Jarrahdale, Pinus radiata plantation, 2.6.76, M. Durack. ( PERTH 00775665). ACKNOWLEDGEMENTS The advice and encouragement of the late Professor M.M.Moser is_ gratefully acknowledged. Grateful appreciation is given to New South Wales National Parks and Wildlife Service and State Forests of New South Wales for permission to collect specimens from areas under their control. Thanks are also due to PERTH herbarium for the loan of specimens. Proc. Linn. Soc. N.S.W., 130, 2009 A.E. WOOD REFERENCES Bougher, N. L. & K. Syme, (1988). ‘Fungi of Southern Australia’. (University of Western Australia Press, Perth ). Brandrud, T. E. (1983). Cortinarius subgen. Cortinarius (Agaricales) in the Nordic countries, taxonomy, ecology and chorology - Nordic Journal of Botany 3, 577-592. Brandrud, T. E. et. al (1989-1998) Cortinarius. Flora Photographica. (Cortinarius HB, Matfors, Sweden). Breitenbach, J. &F. Kranzlin, (2000). Fungi of Switzerland. Vol 5. Cortinariaceae. (Lucerne,Switzerland). Brittlebank, C. C. (1940). Catalogue of Australian Fungi. (Published by the author ) Chambers, S. M. & N. A. Sawyer & J.W.G. Cairney (1999). Molecular identification of co-occurring Cortinarius and Dermocybe species from southeastern Australian sclerophyll forests. Mycorrhiza 9, 85-90. Cleland, J. B. (1934). ‘Toadstools and -Mushrooms and Other Larger Fungi of South Australia’. 1. (Government Printer, Adelaide). Cooke, M. C. (1892). “Handbook of Australian Fungi’. (Williams & Norgate, London.) Garnica, S., Weiss, M., Oertel, B., Oberwinkler, F., (2005). A framework for phylogenetic classification in the genus Cortinarius (Basidiomycota, Agricales) derived from morphological and molecular data. — Canadian Journal of Botany 83 (11), 1457 — 1477. Gasparini, B. (2001). A Contribution to the knowledge of Cortinarius and allied genera of Southern Tasmania, Australia. 1. Cortinarius subgenus Cortinarius. Australasian_Mycologist 20 (1), 49-54. Grgurinovic, C. A. (1997). “Larger Fungi of South Australia’. (Botanic Gardens and State Herbarium and Flora and Fauna of South Australia Handbooks Committee, Adelaide). Griffiths, K. (1985). ‘A Field Guide to the Larger Fungi of the Darling Scarp and the South West of Western Australia’. (Published by the author). Hilton, R. N. (1983). A census of the larger fungi of Western Australia. Part 11 — Journal of the Royal Society of Western Australia 70, 111-118. Hoiland, K. & A. Holst-Jensen, (2000). Cortinarius phylogeny and possible taxonomic implications of ITS rDNA sequences. - Mycologia 92(4), 694-710. Horak, E. (1979). Fungi, Basidiomycetes Agaricales y Gasteromycetes Secotioides. Flora Criptogamica de Tierra del Fuego. Tomo X1 Fascicle 6 (Buenos Aires, Argentina). Horak, E. (2005). ‘ Rohrlinge und Blatterpilze in Europa.’ (Elsevier, Munchen). Knudsen, H. & Vesterholt, J. (Eds).(2008). “Funga Nordica’. (Nordsvamp, Copenhagen). Kornerup, A. & Wanscher, J. H. (1978). ‘Methuen Handbook of Colour’. (Methuen, London). Proc. Linn. Soc. N.S.W., 130, 2009 Maerz, A. & M. R. Paul (1950) “A Dictionary of Colour’. 2nd Edition. McGraw —Hill, New York. May, T. W. & A. E. Wood, (1997) Catalogue and Bibliography of Australian Macrofungi. 1. Basidiomycetes p.p. Fungi of Australia Volume 2A. Australian Biological Resources Study. Canberra. 348 pp. McAlpine, D. (1895). Systematic Arrangement of Australian Fungi, (R. S. Bain, Government Printer, Melbourne). Moser, M. (1983). Keys to Agarics and Boleti (Polyporales, Boletales, Agaricales, Russules) (Roger Phillips. London). Moser, M. M. (1986) 1987. Cortinarius Fr. subgenus Cortinarius 1n the SW-Pacific area. Sydowia 39 : 138-147 Moser, M. M. & E. Horak, (1975). Cortinarius Fr. und nahe verwandte Gattungen in Sudamerika, Beih. Nova Hedwigia 52: 1-628. Shepherd, C. J. & C. J. Totterdill (1988). Mushrooms and Toadstools of Australia. (Inkata Press. Melbourne). Syme, K. (1992) Survey of the Larger Fungi of the Two Peoples Bay Nature Reserve. Denmark Environment Centre, Denmark. Young, A. M. (1994) Common Australian Fungi. (New South Wales University Press, Sydney). 155 _ ConTINARIL AY PF: sti RC vamewcnten USINA AUS Sool a seein “OHO Tle 5 Ds Aystnaliat Ato? wot DH winidoM cvtiba pe . baw sugolmis's {FOP pa bos OF Da, Be ET Yauch [ender A onl YeuAdo AeptergoildiG | AS oat attecA to Liat |e strpvirnal Met ensian) obi vsirabio8 Bo pia i neifeugerd Pa } Aa) phe w re | durgotet} URL) ) tein) AIM tahoe T smears viata, ane em east i mellow: ene \ isto bai xaieguscn htor pee M, rsen dicvitenid A colar saab on {ovaling,S ml i Ht hte Ai dvdr’ wens’ hie OF ive’. .eoam we aKwe av or i ni 74 ine 24 zaioniiws. lastinlbars Le eae Ais8 cchnatiot ai cogmmtisd otbaew tee olen Lime ie raiaiadiA erg tiem.oninrteuy (eet) f As (umiadie wean vee tat) pon Ton choot . newt dy lot won adit to fev (tet Rw do iTramowena: Panel! pirat oR eroaei dn asta} drmerstoKh a he wavy gel wnnaleendeae onion (Doe MA { vonbu® bayr'l 4 NAB Frist Uae frend | } | fry oy } j j ; / \ ; a \ | , | @y. \ 1} AJ \ ri \ } ( "> C4 } f Vapeneh a Contieandts hat Lrwellvresia (PERTH acai ETO ae Billed A PAE d.. (ear yiiie rita abumient, drow filer of ae Heal caticle a teat, pearcels iveted layer SUI) ate deen, ..compored SUT ALS apemial i eet < aT at LY Orne ie eie, Ane a+ 4 y ittenwed, Dat meter, thst witly.no } Mme. Cysin Ho one Wer the Onova amore denacly i Hh CSHSCS On PHik bly IT Siporss eid eiga? is (et oe ae rary ie aundattani 3 bwial fener wiht Nas or whether 1 waa iyibeaa Mana woth, Wie eaatic art ria isceur hice ce eeerve adlectmng ony We enwintend, j : ee PULTE OT PILES Vrs ce Reeder tis) ST ar 3S paneer) te the wintebed hoe eperitem: ary Thigy Meh ert? PER Cee art mot j iu callovtven), Ti Qe pore oF anhselien at: has inf ‘ynvdnor fs “aban eh * | j , 7 ae evil nlperey > (NS - habbon ental . tghas A en i, sisal } dae oti % 1) tS ia a a ried tote) Be Anite rae. wey as agi Soc ak styiatalod, amine vat as-Gatiow) wu pre ee pane ait DS rane paisa sf dihalben-+-R—rigaloy AA, sinh diaeie Yo igre yoyiat” ATOR TA 9 orvunh nushedtaH ssid bie anobwO ey ‘ellen tye rie i igh in sbivid bal q re Aeber) Ae . ‘one ‘ cri) sb ie —" pie fe rite isiy 3 re “OT theres bh rvored YG Gk ekvenie> (0000 )\c5e0nl-elol AD A De Jo: RnodnoGM 2imonoxs! aidigang bas ynogotydle ODO: (MAE ONRIE Didi tol RCs. ani | Yableabiag Avatzoymnnibiesdl inn en ob Bharani Derek mahypetowe’ os Nene Meet} «) oigtaeetls iK ov ont ey UA ee ee oe Mew South phason , wf iQirindtiei Svbeepaniatid Bewerpnilade be 48 “tor penmnisiith lo motlee saciimcteottog M control: Maou LBOOGjx dead wl § nt Wks ie lot axes: imenforgudasgod epoca eal ay nautitioM’ (ATOL) Hi, ger ie adbio.t — Sag le ta Wi, phat gun (PE Late Ordovician Strophomenide and Pentameride Brachiopods from Central New South Wales IAN G. PERCIVAL Geological Survey of New South Wales, Department of Primary Industries, 947-953 Londonderry Road, Londonderry, NSW 2753, Australia (ian.percival@dpi.nsw.gov.au). Percival, I.G. 2009. Late Ordovician Strophomenide and Pentameride Brachiopods from central New South Wales. Proceedings of the Linnean Society of New South Wales 130, 157-178. Strophomenide and pentameride brachiopods are described from shelfal environments (BA 3) flanking islands of the Macquarie Arc during the Late Ordovician (latest Sandbian to early Katian stages). Most of the strophomenoid genera recognized are new, monotypic, and hence endemic, although the occurrence of a new species of Shlyginia is indicative of affinities with Kazakhstan. Taxa described include the strophomenid Geniculomena barnesi gen. et sp. nov., the rafinesquinid Testaprica rhodesi gen. et sp. nov., glyptomenids Resupinsculpta cuprafodina gen. et sp. nov., Paromalomena zheni sp. nov., and Platymena? sp., and the plectambonitoid Shlyginia rectangularis sp. nov. Review of the generic assignment of Oepikina? walliensis Percival, 1991 suggests that this species is better placed in Murinella Cooper, 1956. Relatively rare pentameride brachiopods are represented by only a few specimens, including an unnamed species of Parastrophina, and a species tentatively referred to Eoanastrophia. Manuscript received 1 December 2008, accepted for publication 16 February 2009. KEYWORDS: brachiopod, Late Ordovician, Macquarie Arc, new genera, pentameride, strophomenide INTRODUCTION Late Ordovician strophomenide brachiopods are well-represented in limestones and sandstones deposited around volcanic islands forming the Macquarie Arc in central New South Wales, with most of the fauna having previously been described over the past three decades (Percival 1979a, 1979b, 1991; Percival et al. 2001). For various reasons (including rarity of specimens, and insufficient knowledge of morphological features needed to characterize new species), several additional strophomenide taxa have remained undocumented. This paper aims to address this deficiency in order to present a more complete picture of the fauna to underpin future analyses of biogeographic relationships. In addition, species of Late Ordovician strophomenides previously tentatively ascribed to Oepikina by Percival (1979b) from the vicinity of Gunningbland, and Percival (1991) from the Licking Hole Creek area, near Cliefden Caves (Figure 1), are reassessed in order to clarify their systematic position. The opportunity is also taken to describe some rare examples (represented by just a handful of specimens) of Late Ordovician pentameride brachiopods. Both genera recognized are left in open nomenclature as all specimens are incomplete. However, the presence in the fauna of two additional camerelloids is significant and worthy of documentation as only one species of pentameride brachiopod, Didymelasma inconspicua Percival, 1991, had previously been described from contemporaneous rocks of the region. Except for specimens of Testaprica rhodesi gen. et sp. nov. and Platymena? sp. which were found in fine-grained sandstone in the upper Gunningbland Formation of late Eastonian (Ea3-4) age, the brachiopods described here are silicified, having been recovered from residues of limestones dissolved in dilute hydrochloric acid. These limestones are of early Eastonian age, equivalent to the latest Sandbian or earliest Katian of international usage. Details of the stratigraphic succession and tectonic context within the Macquarie Arc in central NSW are provided by Percival and Glen (2007), and only a brief summary of the age and correlation of these strata (Figure 2) is given here. BRACHIOPODS FROM CENTRAL NEW SOUTH WALES BeeeT Gate o LSI... # BILLABONG : id CREEK unningbland ‘ LIMESTONE e PARKES e FORBES e Eugowra j | NEW SOUTH WALES $ i ie Bathurst EE ~ oo N CANBERRAe N Molong | ~REEDY CREEK / LIMESTONE / Cudal ° aw ~ BOWAN PARK CARGO CREEK SHEGREA LIMESTONE , \ Cargo ORANGE e | __- REGANS CREEK "7 LIMESTONE CANOMODINE — Y ‘ LIMESTONE e Canowindra eae Licking Hole Creek — CLIEFDEN CAVES ~— LIMESTONE SUBGROUP ® Mandurama ® Woodstock Figure 1. Locality map showing sites in central New South Wales yielding Late Ordovician brachiopods described in this paper. Outcrop of main Upper Ordovician limestone units shown in black; localities (L37, L51) in overlying Upper Ordovician clastic-dominated units are shown by spots. Stratigraphic setting Cliefden Caves and Licking Hole Creek areas, east flank of Molong Volcanic Belt In the Cliefden Caves area of central New South Wales (Webby and Packham 1982) and the Licking Hole Creek area adjacent to the west (Percival 1976), a well preserved Late Ordovician carbonate-dominated sedimentary succession formed on an eroded volcanic island setting, represented by the Walli Volcanics. The Cliefden Caves Limestone Subgroup includes the Fossil Hill Limestone at the base, succeeded by the massive Belubula Limestone which is itself overlain by the Vandon Limestone. Biostratigraphic evidence from conodonts, trilobites, corals and stromatoporoids, and brachiopods, demonstrates that the Fossil Hill Limestone (and equivalents in the Licking Hole Creek area), and the lower part of the Belubula Limestone, were deposited in the earliest Eastonian (Eal); the remainder of the limestone succession is of Eastonian 2 age, which corresponds to the basal Katian stage. 158 The strophomenide biofacies characterizes Benthic Assemblage 3 (BA 3) throughout these limestone deposits, which is interpreted as occupying open shelf environments in well-circulated shallow to moderate water depths (Percival and Webby 1996). Representative brachiopods of this biofacies have been largely documented by Percival (1991); further species described herein include Geniculomena barnesi, Resupinsculpta cuprafodina, Paromalomena zheni, Shlyginia rectangularis, and Parastrophina sp. Additionally Oepikina walliensis Percival, 1991, described from the basal Belubula Limestone in the Licking Hole Creek area, is reassessed and assigned to Murinella. Regans Creek Limestone, southeast of Cargo, east flank of Molong Volcanic Belt The Regans Creek Limestone, mapped by McLean (1974), is a relatively small exposure of limestone that is contemporaneous with the Cliefden Caves Limestone Subgroup. Proc. Linn. Soc. N.S.W., 130, 2009 I.G. PERCIVAL BOLINDIAN Bot Bo2 Bo3 Bo4 Bod EASTONIAN wee 1, 2,3 au 1, 2,3 mmm 1, 2,3 mes 1, 2 UPPER ORDOVICIAN SANDBIAN Gi2 GISBORNIAN Gil Murinella walliensis Geniculomena bamsei Resupinsculpta cuprafordina Paromalomena zheni Shlyginia rectangulans Testaprica rhodesi Platymena? sp. mu 27,3 Parastrophina sp. 3 Bowan Park asses le Saeetell 4 ANGULLONG VOLCANICS MALACHIS HILL t+ FORMATION MALONGULLI FORMATION | GUNNINGBLAND FORMATION Jee eee voy te Je | sieeg a | aeons | CAVES LIMESTONE Sleleigaboed SUBGROUP BILLABONG CREEK LIMESTONE Se SS al es CARGO WALLI et VOLCANICS VOLCANICS = g 7) S & (base early Wy | (base unknown) (base unknown) Darriwilian) 2009_02_0042 Figure 2. Stratigraphic levels at which Late Ordovician brachiopods described in this paper occur in central New South Wales. Numerals associated with approximate ranges refer to numbered stratigraph- ic columns to the right. Note that Parastrophina sp. also occurs in the Checkers Member in the upper Regans Creek Limestone (not shown on this diagram). HIRN. = Hirnantian stage. : The Checkers Member in the upper part of the Regans Creek Limestone yields a silicified fauna comparable to that in the Trilobite Hill Limestone Member of the Vandon Limestone at Cliefden Caves, although diversities are considerably lower. To the brachiopods described from this level by Percival (1991) can now be added Parastrophina sp. Bowan Park area, west flank of Molong Volcanic Belt The geology of the Bowan Park area has been described in detail by Semeniuk (1970, 1973). Limestones of the Bowan Park Subgroup (including in ascending order, the Daylesford Limestone, Quondong Limestone, and Ballingoole Limestone) overlie the Cargo Volcanics, and are in turn overlain by Proc. Linn. Soc. N.S.W., 130, 2009 the Malachis Hill Formation (Fig. 2). The succession at Bowan Park differs from that on the southwestern MVB (in the Cliefden Caves area) where late Eastonian (Ea3) age sediments are represented by the graptolitic Malongulli Formation above the Cliefden Caves Limestone Subgroup, whereas carbonate deposition (Ballingoole Limestone) occupied this interval in the Bowan Park area. The Quondong Limestone contains abundant marine invertebrate faunas of the strophomenide biofacies (Percival 1991), comparable in age and diversity with those in the Trilobite Hill Member of the Vandon Limestone (Cliefden Caves Limestone Subgroup) and like that unit clearly belongs to BA 3 (i.e. shelfal). Additional species described herein from the Quondong Limestone include Resupinsculpta BRACHIOPODS FROM CENTRAL NEW SOUTH WALES cuprafodina, Paromalomena zheni, Shlyginia rectangularis, Parastrophina sp. and Eoanastrophia? sp. Gunningbland area, Junee-Narromine Volcanic Belt The Billabong Creek Limestone was shown by Pickett and Percival (2001) to extend from southeast of Gunningbland in a broad arcuate band trending northwestwards to north of the Parkes- Broken Hill railway, then northeast to exposures on “Kirkup” property (Figure 1). Conodonts from the “Kirkup” section, of early Darriwilian (Da2) age (Zhen and Pickett 2008), are the oldest dated fossils in the Billabong Creek Limestone. Younger conodont and coral assemblages from the type section of the formation on “Nelungaloo” property, southeast of Gunningbland, range in age through the late Darriwilian, Gisbornian and earliest Eastonian (Pickett and Percival 2001). Outcrops in and adjacent to Billabong Creek at the southern extremity of the limestone belt are rich in silicified fossils, particularly brachiopods (including Geniculomena barnesi, Resupinsculpta cuprafodina, Paromalomena zheni and Shlyginia rectangularis, described herein, and a diverse fauna documented by Percival 1991) and trilobites (Webby 1973, 1974), of Eastonian 2 age (Pickett and Percival 2001). These upper beds of the Billabong Creek Limestone correlate with the Quondong Limestone at Bowan Park, and the Trilobite Hill Limestone Member of the Vandon Limestone in the Cliefden Caves Limestone Subgroup (Figure 2). The Billabong Creek Limestone is apparently conformably overlain by the Gunningbland Formation, although the actual boundary is unexposed. The outcrop belt of the Gunningbland Formation consistently lies immediately west of the arcuate trend of the Billabong Creek Limestone exposures (Pickett and Percival 2001). Shallow excavations and exposures in ploughed fields on “Currajong Park”, “Sunnyside” and “New Durran” properties in the Gunningbland district reveal that the Gunningbland Formation predominantly consists of siltstone, shale, and fine- to medium-grained sandstone, together with minor fossiliferous limestones. Most of the Gunningbland Formation is of late Eastonian (Ea3) age, determined from graptolites in siltstones, and conodonts including Taogupognathus tumidus in limestone lenses. The limestones also contain a coral-stromatoporoid assemblage corresponding to the contemporaneous Fauna III (McLean and Webby 1976, Webby and Morris 1976). Two brachiopod faunas, elements of which were described by Percival (1978, 1979a, 1979b), are recognised. Brachiopod Fauna C, of Ea3 age, is present 160 in the lower part of the formation on “New Durran” property. The presumed latest Eastonian age of Fauna D (Percival 1992), occurring in strata on “Currajong Park” property, was confirmed by the presence of graptolites of Ea4 age in the uppermost beds of this section (Pickett and Percival 2001). A diverse trilobite fauna has recently been described from this upper part of the unit (Edgecombe and Webby 2006, 2007), associated with the brachiopods Testaprica rhodesi gen. et sp. nov. and Platymena? sp. which are documented herein. This completes description of the brachiopod fauna collected from the Gunningbland Formation over more than three decades; two other genera (Christiania sp., Ptychopleurella? sp.) are represented in the upper part of this unit by single specimens of ventral valves, which do not warrant description until further material is forthcoming. Systematic palaeontology Type material (designated MMF), comprising specimens described and illustrated or listed herein, is curated in the palaeontological collections of the Geological Survey of New South Wales held at Londonderry in western Sydney. Some specimens labeled SUP, including material of Murinella walliensis and an external mould of the ventral valve of Testaprica rhodesi, were transferred from the Geology Department of the University of Sydney to the Australian Museum, Sydney in the mid- 1980s (these are awaiting renumbering). For brevity, authorship of taxonomic hierarchy above genus level is not cited in the References; these bibliographic sources are listed in the revised (2 edition) Treatise of Invertebrate Paleontology, Part H: Brachiopoda Volume 3 (Williams et al. 2000). Phylum Brachiopoda Duméril, 1806 Subphylum Rhynchonelliformea Williams, Carlson, Brunton, Holmer and Popoy, 1996 Class Strophomenata Williams, Carlson, Brunton, Holmer and Popov, 1996 Order Strophomenida Opik, 1934 Superfamily Strophomenoidea King, 1846 Family Strophomenidae King, 1846 Subfamily Furcitellinae Williams, 1965 Geniculomena gen. nov. Type species (by monotypy): Geniculomena barnesi gen. et sp. nov. Diagnosis Dorsally geniculate planoconvex to weakly concavoconvex furcitellin with unequally Proc. Linn. Soc. N.S.W., 130, 2009 I.G. PERCIVAL parvicostellate ornament lacking rugae; teeth and Diagnosis sockets without crenulations; dorsal myophragm As for genus. absent; septa associated with dorsal muscle field Etymology are less strongly developed than single continuous Genus name in reference to geniculate dorsal median ridge. valve profile and broadly crescent-like shell outline; species name honours David Barnes, photographer Geniculomena barnesi gen. et sp. nov. in the NSW Department of Primary Industries, in Fig. 3 A-N appreciation of the assistance he has provided to me Figure 3. Geniculomena barnesi gen. et sp. nov. A — B: interior and exterior of dorsal valve, holotype MMF 44915. C — E: interior, exterior and lateral profile (dorsal side uppermost) of dorsal valve, MMF 44916. F—H: interior, exterior and anterior profile (dorsal side uppermost) of dorsal valve, MMF 44919. I — J: interior and exterior (bearing heliolitid coral) of dorsal valve, MMF 44917. K — L, O — Q: one incomplete individual shell, which disarticulated during acid dissolution of limestone matrix; K — L: ex- terior and lateral profile (dorsal side uppermost) of dorsal valve, MMF 44918a; O — Q: exterior, interior and lateral profile (ventral side uppermost) of ventral valve, MMF 44918b. M: interior of dorsal valve, MMF 44920. N: interior of dorsal valve, MMF 44921; note distortion on anterolateral margin, probably indicating repaired injury. Scale bar below C represents one cm. A— E, I— L, O— Q from L24, Trilobite Hill Limestone Member of Vandon Limestone, upper Cliefden Caves Limestone Subgroup at Licking Hole Creek near Walli; F — H from L135 (east of Copper Mine Creek, near Cliefden Caves) in Trilobite Hill Limestone Member of Vandon Limestone, upper Cliefden Caves Limestone Subgroup; M — N from L143, upper Billabong Creek Limestone at Billabong Creek road crossing south of Gunningbland. Proc. Linn. Soc. N.S.W., 130, 2009 161 BRACHIOPODS FROM CENTRAL NEW SOUTH WALES over the past decade in preparing many illustrations of fossils for publication. Material Five dorsal valves, mostly entire, and one partial ventral valve with corresponding partial dorsal valve (disarticulated), all material silicified. Holotype is dorsal valve MMF 44915; paratypes include dorsal valves MMF 44916, MMF 44917, MMF 44919, MME 44920 and 44921, and ventral valve MMF 44918a and corresponding dorsal valve 4491 8b. Localities Type locality is L24 (Licking Hole Creek area), in Trilobite Hill Limestone Member of Vandon Limestone, upper Cliefden Caves Limestone Subgroup; also found in same stratigraphic unit at L135 (east of Copper Mine Creek, near Cliefden Caves); also occurs at L143 in upper Billabong Creek Limestone, at Billabong Creek road crossing south of Gunningbland [full details of these localities are given by Percival 1991]. Description Shell planoconvex to very weakly concavoconvex (rarely ventribiconvex, e.g. Fig. 3G), becoming dorsally geniculate when fully grown; transversely subquadrate with maximum width either at, or immediately anterior to, hingeline; lateral and anterior margins broadly curved. Shell of moderate size, ranging in length from 12 to 16 mm, and in width from 23 to 29 mm in largest specimens; length to width ratio 0.55 -0.80. Ornament unequally parvicostellate, with every fourth or fifth rib accentuated; rugae lacking; exterior of the sole ventral valve assigned to this species is almost entirely devoid of ornament, but this may have been eroded prior to fossilization. Ventral interior (described from an incomplete valve) shows robust oblique teeth supported by low plates for approximately three-quarters length; dental plates extend anteriorly to bound triangular diductor scars flanking (but not enclosing) narrower median pair of adductor scars separated by low median ridge not extending forward of muscle field which occupies three-eighths valve length. Mantle canals prominent, of lemniscate type with anteriorly divergent vascula media not enclosing vascula genitalia. A distinct but low subperipheral rim defines a dorsally-deflected marginal band approximately one-seventh of valve length extending around entire lateral and anterior valve margin. Details of interarea and delthyrium not known. Dorsal interior with Type A strophomenoidean cardinalia consisting of twin cardinal process lobes 162 extending just posterior to hingeline and convergent above a hollow, with narrow, widely divergent socket ridges recurved posterolaterally at extremities; sockets short but deep; no crenulations visible on socket ridges. Notothyrial platform poorly developed, lacking myophragm; low median septum extends from immediately in front of cardinal process lobes to terminate at about half valve length, separating moderately conspicuous pair of adductor scars which are bounded by weaker side septa; short transmuscle septa barely visible or lacking. Mantle canals apparently lemniscate, poorly expressed, except for vascula genitalia in largest specimen. A variably defined subperipheral ridge is sometimes developed slightly posterior to dorsally-directed geniculation of marginal band. Dimensions Holotype MMF 44915: length 12.0 mm, width 19.0 mm; paratypes MMF 44916: length 13.1 mm, estimated width 23 mm; MMF 44917: length 15.5 mm, width of specimen (incomplete) 18.8 mm; MME 44919: length 16.0 mm, width 23.5 mm; MMF 44920: length 13.5 mm, estimated width 22.5 mm; MMF 44918a (vv): length 15.9 mm, estimated width 29 mm. Discussion Geniculomena is assigned to the subfamily Furcitellinae, rather than the Strophomeninae, due to the presence of a moderately well-defined dorsal muscle field in some specimens, although muscle bounding ridges, side septa and transmuscle septa are somewhat variably developed and may be barely discernible in other examples depending on degree of silicification. Dorsally geniculate genera similar to Geniculomena are more typical of furcitellins rather than strophomenins. Dactylogonia Ulrich and Cooper, 1936 (and its synonym Cyphomena Cooper, 1956) appears to closely resemble Geniculomena in general morphology, but Dactylogonia is readily distinguished by its much stronger development of transmuscle and side septa in the dorsal valve. The new genus lacks the characteristic rugate ornament of Bellimurina Cooper, 1956, and differs internally in absence of a forked anterior termination to the dorsal median ridge. Although Geniculina R6émusoks, 1993, from the latest Ordovician (Hirnantian) of the Baltic region, is broadly similar to Geniculomena, the new genus apparently lacks the prominent posterolateral oblique rugae developed on the ventral valve of Geniculina. Nor have crenulations been observed on the teeth and socket ridges of Geniculomena, whereas these are Proc. Linn. Soc. N.S.W., 130, 2009 I.G. PERCIVAL characteristic of at least four species of Geniculina (e.g., ROOmusoks 2004, pl. IX fig. 12, pl. XI fig. 6). The median septum in Geniculomena is a single ridge that extends from the cardinal process and is rather more prominent than the side septa, unlike the arrangement in Geniculina that has strong side septa and a stout myophragm which bifurcates at its anterior extremity. The multicostellate ornament of Maakina Andreeva, 1961 (in Nikiforova and Andreeva 1961), from the early Katian of the Siberian Platform, is quite different from that of Geniculomena. Internally, the absence of a dorsal median septum and presence of crenulations on the socket ridges in Maakina are additional features clearly distinguishing these two genera. Distribution Early Eastonian (Ea2), equivalent to basal Katian; presently monotypic and known only from limestones of the Macquarie Arc in central NSW. Murinella Cooper, 1956 Type species: Murinella partita Cooper, 1956 Murinella walliensis (Percival, 1991) Fig. 4 A-G Synonymy Oepikina? walliensis Percival, 1991: p.147, fig. 14.20-28. Discussion Two species with Oepikina-like morphology have previously been described from the Late Ordovician of central NSW. One form from the Gunningbland Formation was tentatively referred to Oepikina? sp. by Percival (1979b), and a new species Oepikina? walliensis was described by Percival (1991) from the Licking Hole Creek area, occurring in strata equivalent to the basal Belubula Limestone. In their revision of the superfamily Strophomenoidea, Rong and Cocks (1994, p.694) noted that the cardinalia of O? walliensis was “of the Strophomena group”, presumably implying that in their view the species was a strophomenin rather than a furcitellin. Zhan et al. (2008) observed that these two subfamilies are difficult to separate using the revised Treatise classification (Cocks and Rong 2000). Rong and Cocks (1994, text-fig. 3) also presented a well-illustrated comparison between the dorsal cardinalia of the type species of Strophomena and Murinella. Although a reclassification of O? walliensis on the basis of cardinalia alone might therefore be superfluous, the comments by Rong and Cocks (1994) have prompted a reassessment of other possible generic affinities of Figure 4. Murinella walliensis (Percival, 1991). A— E: Holotype SUP 68516, exterior of conjoined valves, dorsal, ventral, posterior profile, anterior profile and lateral profile respectively. F: fragment of dorsal valve interior showing cardinalia, SUP 68523. G: interior of ventral valve, SUP 68518. Both scale bars represent 1 cm (that beneath F pertains only to this specimen; the shorter scale bar applies to specimens A-E and G). All specimens from basal Belubula Limestone at Licking Hole Creek, near Walli. Proc. Linn. Soc. N.S.W., 130, 2009 BRACHIOPODS FROM CENTRAL NEW SOUTH WALES this species. The holotype of O? walliensis is here refigured, together with a paratype fragment showing the cardinalia and a ventral valve interior. Reasons given by Percival (1991) for provisionally assigning this species to Oepikina include poorly developed septa in the dorsal valve, and presence of a relatively small ventral muscle field enclosed by low bounding ridges. Both these features are atypical of Oepikina, whereas they are characteristic of the similar genus Murinella Cooper, 1956. Although a distinguishing feature of the type species of Murinella, M. partita Cooper, 1956, is the extension of the median septum anterior to the ventral muscle field, not all species show this (e.g. M. muralis Cooper, 1956 and M. semireducta Cooper, 1956). In retrospect, O? walliensis accords best with Murinella, and it is here designated M. walliensis (Percival, 1991). Other features supporting this reassignment include the relatively large pseudodeltidium and prominent subperipheral rim in the dorsal valve of M. walliensis. Furthermore, the cardinalia definitely conform to the Murinella model. A species of Murinella has also been described from the lower limestone member of the Benjamin Limestone in Tasmania by Laurie (1991). That species, M. magna, is distinguished by its much larger dimensions, and in having a median septum extending forward of the ventral muscle field, compared to M. walliensis. Oepikina? sp from Gunningbland is known only from one specimen (Percival 1979b, fig. 1.12), which clearly shows the presence of Type A cardinalia (sensu Rong and Cocks 1994). In all other features this dorsal valve is definitely Oepikina-like, with strong side septa, but the absence of a corresponding ventral valve continues to prevent a confident assignment to that genus. The external mould supposedly of a dorsal valve (SUP 62569), mentioned but not figured by Percival (1979b, p.183), is now considered to be a ventral valve of Testaprica rhodesi (see below) rather than being related to Oepikina. Family Rafinesquinidae Schuchert, 1893 Subfamily Rafinesquininae Schuchert, 1893 Testaprica gen. nov. Type species (by monotypy): Zestaprica rhodesi gen. et sp. nov. Diagnosis Convexo-concave to convexo-planar rafinesquinin similar to Rhipidomena but with 164 prominent subparallel side septa in dorsal valve; other septa and median ridge subdued or lacking. Testaprica rhodesi gen. et sp. nov. Fig. 5 A-H Diagnosis As for genus. Etymology Genus name derived from testa (Latin): shell, and apricum (Latin): a sunny spot, in reference to the occurrence of this brachiopod adjacent to “Sunnyside” property; species named in honour of Julie and John Rhodes, former owners of “Sunnyside” and “Currajong Park” properties at Gunningbland, who kindly provided access to collect on their land, and who also recognised and donated several important brachiopods and trilobites for scientific description. Material Holotype: MMF 36806a and b, dorsal valve internal mould and external mould of corresponding ventral valve. Paratypes: MMF 36798a and b, dorsal valve internal and external moulds; MMF 36801 and MME 36805, both external moulds of dorsal valves; MMF 36813, dorsal valve internal mould; SUP 62569 ventral valve external mould. Localities All specimens from upper Gunningbland Formation on “Currajong Park”, Gunningbland at locality L51 [see Percival 1979a for full details] with exception of MMF 36813, collected from locality L48 situated in immediately underlying beds in the same formation on this property. Description Large convexo-concave to convexo-planar shells up to 40 mm wide and 30 mm long, with maximum width attained at or immediately anterior to hingeline; anterolateral and anterior margins very broadly rounded. Length to width ratio varies between two- thirds and almost three-quarters. Ornament finely and evenly multicostellate, lacking rugae; costellae slightly curved on lateral flanks; occasional concentric growth discontinuities may be present, but concentric filae lacking. Ventral valve weakly concave, becoming almost planar anteriorly; interarea low, catacline to weakly apsacline, with small pseudodeltidium. Details of interior unknown. Dorsal valve strongly convex; interarea very low with delicate chilidial plates (poorly preserved on Proc. Linn. Soc. N.S.W., 130, 2009 I.G. PERCIVAL Figure 5. Testaprica rhodesi gen. et sp. nov. All specimens from upper beds of the Gunningbland For- mation on “Currajong Park”, Gunningbland. A — D: Holotype, MMF 36806a and b; A: exterior mould of ventral valve (on left, 36806a) and interior mould of corresponding dorsal valve (on right, 36806b); B: latex replica taken from this specimen; C: enlargement of posterior region of latex replica of dorsal valve; D: latex replica of exterior of ventral valve, tilted to better show ornament and interarea. E: inte- rior mould of dorsal valve, MMF 36798a. F: latex replica of dorsal valve, MMF 36813. G: latex replica of exterior of dorsal valve, MMF 36805. H: latex replica of exterior of dorsal valve, MMF 36801. Both scale bars represent 1 cm (that below C pertains only to this enlargement). available specimens). Cardinalia consisting of small transmuscle septa barely visible; muscle bounding cardinal process with pair of discrete peg-like lobes ridges not present and muscle field not impressed. above low notothyrial platform, with very short, Mantle canals not discernible. straight socket ridges extending obliquely; median ridge either very short or not developed; prominent Dimensions subparallel pair of side septa, low and thin, extend MME 36806a, b (holotype): DV internal mould to between one quarter and one third valve length; and VV external mould L= 26.3 mm, hinge W= 39.3 mm; Proc. Linn. Soc. N.S.W., 130, 2009 165 BRACHIOPODS FROM CENTRAL NEW SOUTH WALES MMF 36798a, b: DV internal and external moulds L= 28.4 mm, W= 39.3 mm; MME 36801: DV external mould L= 22.4 mm, spec W= 25.3 mm, W= 29.6 mm; MME 36805: DV external mould L= 16.0 mm, W= 22.0 mm; MME 36813: DV internal mould L= 17.6 mm; SUP 62569: VV external mould L= 13.5 mm, W= 17.5 mm. Discussion This monotypic genus has cardinalia of Type B (sensu Rong and Cocks 1994), with small discrete cardinal process lobes that are not continuous with a median ridge, and which are also definitely disjunct from the socket ridges (the latter being straight and oblique, rather than recurved laterally towards the hingeline as in strophomenids). Clearly then, its affinities lie with the rafinesquinids. The only previously described rafinesquinin brachiopod with a convexo-concave valve profile is Rhipidomena, which is of generally comparable size. However, dorsal valves of the 5-6 species of this genus known from North America (Cooper 1956), are never quite as convex as is Testaprica, and the latter is not resupinate as is commonly the case with Rhipidomena. In possessing prominent side septa T- rhodesi differs from all North American Rhipidomena species, and is further distinguished by its relatively poorly developed median ridge and transmuscle septa (although there is some variation in the strength of these features). These distinctions in total appear to be of generic significance, so that despite the absence of ventral interiors the establishment of a new genus is warranted. Equally prominent side septa are also characteristic of Lateriseptomena Zhan, Jin, Rong, Chen and Yu, 2008, known from two species of late Katian age from Zhejiang Province, south-east China. However, Lateriseptomena has Type C (glyptomenid) cardinalia, and furthermore has a _ planoconvex to biconvex profile, so is apparently not closely related to Testaprica. The concavo-convex profile of Dirafinesquina Cocks and Zhan, 1998, from Upper Naungkangyi Group equivalent strata (probable Katian age) in the Southern Shan States of Burma, readily distinguishes this genus from Testaprica; the few known dorsal interiors of Dirafinesquina also lack the characteristic side septa of the new genus. Distribution Presently known only from the Gunningbland Formation (upper part) in vicinity of Gunningbland village, between Parkes and Bogan Gate, central west 166 NSW; late Eastonian (Ea3-4) i.e. Katian. Family Glyptomenidae Williams, 1965 Subfamily Glyptomeninae Williams, 1965 Resupinsculpta gen. nov. Type species (by monotypy): Resupinsculpta cuprafodina gen. et sp. nov. Diagnosis Resupinate glyptomenin displaying weak rugation on exterior of both valves; teeth and socket ridges occasionally crenulate. Resupinsculpta cuprafodina gen. et sp. nov. Fig. 6 A-P Diagnosis As for genus. Etymology Genus name in reference to resupinate profile and finely engraved appearance of ornament (resupinus: L bent back; insculptus: L engraved); species name in reference to Copper Mine Creek, the type locality (cuprum: L copper; fodina: L mine or pit). Material Holotype MMF 44923 (conjoined valves); paratypes include MMF 44924 (ventral valve), MMF 44925 (dorsal valve), MMF 44926 (ventral valve), MMF 44927 (dorsal valve), MMF 44928 (ventral valve), and MMF 44929 (conjoined valves). All specimens are silicified. Localities Type locality L135 (east of Copper Mine Creek, near Cliefden Caves), in Trilobite Hill Limestone Member of Vandon Limestone, upper Cliefden Caves Limestone Subgroup; also found at L138 (“Quondong”, Bowan Park, east of Cudal) in Quondong Limestone, Bowan Park Subgroup; and at L144 in upper Billabong Creek Limestone beside the road crossing Billabong Creek, south of Gunningbland [full details of these localities are given by Percival NSO: Description Shell relatively small, length up to 12 mm and width to approximately 18 mm; outline subquadrate initially, becoming transverse and slightly auriculate when fully grown with maximum width at hinge line; length two-thirds width in these largest specimens. Proc. Linn. Soc. N.S.W., 130, 2009 I.G. PERCIVAL Figure 6. Resupinsculpta cuprafodina gen. et sp. nov. A— D: Holotype conjoined valves, MMF 44923; A: exterior of ventral valve; B: exterior of dorsal valve; C: lateral profile (dorsal valve uppermost); D: posterior profile (dorsal valve uppermost). E — F: exterior and interior of ventral valve, MMF 44924. G — I: exterior and interior of ventral valve, and enlargement of delthyrium to show crenulated teeth, MMF 44926. J — K: exterior and interior of dorsal valve, MMF 44925. L — M: interior and exterior of dorsal valve, MMF 44927. N: exterior of ventral valve, MMF 44928. O — P: conjoined valves, ventral and dorsal exteriors respectively, MMF 44929. Scale bar represents 1 cm for whole figure (except I, which is a five-times enlargement of H). A— F, J — K from L135 (east of Copper Mine Creek, near Cliefden Caves): in Trilobite Hill Limestone Member of Vandon Limestone, upper Cliefden Caves Limestone Subgroup; G —-I, N — P from L138 (“Quondong”, Bowan Park, east of Cudal) Quondong Limestone, Bowan Park Subgroup. Ventral valve with sharply pointed beak; profile initially weakly convex, becoming resupinate in largest specimens; dorsal valve planar posteriorly, gently to moderately convex anteriorly in adults; whole shell very compressed dorsoventrally. Ornament unequally parvicostellate, commonly with 3-4 finer costellae between accentuated ribs, with indistinct rugae developed posteriorly. Ventral interarea low, apsacline, with wide delthyrium covered apically by pseudodeltidium. Delicate teeth, crenulated in one specimen (Fig. Proc. Linn. Soc. N.S.W., 130, 2009 61), supported by thin subparallel dental plates that terminate immediately in front of teeth. Muscle field indistinct, apparently very short, not enclosed by ridges. A weak subperipheral rim is present in one specimen. Mantle canals not visible. Dorsal interarea very low, orthocline to weakly anacline; notothyrium entirely occupied by cardinal process lobes; chilidial plates either lacking or extremely weakly developed. Cardinalia consist of small paired cardinal process lobes fused to long, straight, widely divergent socket ridges (which are BRACHIOPODS FROM CENTRAL NEW SOUTH WALES finely crenulated in at least one specimen, Fig. 6L) with slightly curved terminations; cardinal process lobes extend very slightly posteriorly of hingeline and anteriorly overhang a concavity in place of notothyrial platform; median ridge short, very low; side and transmuscle septa absent. Muscle field and mantle canals not visible. Dimensions Nearly all specimens are incomplete; a juvenile conjoined shell MMF 44929 is 6.2 mm long and 7.7 mm wide; the largest shell (holotype, MMF 44923) is 11.6 mm long and 18.5 mm wide. Discussion The new species presents a conundrum as regards its generic affinities. It has Type C (glyptomenin) cardinalia, and conforms in almost all respects with the characteristics of Glyptomena, except for the resupinate profile of larger shells. Smaller shells are planoconvex and thus more similar to the typical concavoconvex profile of Glyptomena. As shell profile is often used to distinguish genera in strophomenides, it seems reasonable to establish a new genus within the glyptomenines based on the resupinate character. Furthermore, crenulated teeth and socket ridges as seen in Resupinsculpta cuprafodina are apparently rare in glyptomenines; Rong and Cocks (1994) only mentioned their occurrence in Mjoesina, which was doubtfully assigned to the family (Cocks and Rong 2000), but is now regarded more likely to be a rafinesquinid (Cocks 2005). The indistinct rugae present in the posterior region of the exterior of both valves of R. cuprafodina are lacking in species of Glyptomena, but the distinctively dorsally geniculate Glyptomenoides Popov and Cocks, 2006 (which is otherwise generally similar to Glyptomena) also displays irregular rugae. Most comparable of other strophomenids is possibly Longvillia Bancroft, 1933, which also is resupinate; however, Longvillia has Type A cardinalia and is therefore not closely related to the new genus. Distribution Only known from limestones of early Eastonian (Ea2) age, equivalent to the earliest Katian Stage, in the Macquarie Arc, central NSW. Paromalomena Rong, 1984 Type species: Platymena polonica Temple, 1965 Paromalomena zheni sp. nov. Fig. 7 A-V 168 Diagnosis A species of Paromalomena distinguished by its prominent pseudodeltidium with a minute foramen at the apex, and lacking conspicuous external rugae. Etymology This species is named in honour of my colleague Dr Yong-Yi Zhen, in recognition of his extensive palaeontological studies in the Ordovician of both Australia and China. Material Holotype is MMF 44932 (ventral valve); paratypes include MMF 44930 (dorsal valve), MMF 44931 (ventral valve), MMF 44933 (ventral valve), MME 44934 (conjoined valves), MMF 44935 (dorsal valve), MMF 44936 (ventral valve), MMF 44937 (conjoined valves), MMF 44938 (dorsal valve), MMEF 44939 (dorsal valve), MMF 44940 (dorsal valve), MMF 44941 (dorsal valve), MMF 44942 (dorsal valve), MMF 44943 (ventral valve), MMF 44944 (ventral valve), MMF 44945 (ventral valve), and MMF 44946 (conjoined valves). All specimens are silicified. Localities Type locality is L138 (“Quondong”, Bowan Park, east of Cudal) in Quondong Limestone, Bowan Park Subgroup; also occurs at L24 (Licking Hole Creek area, Walli) in Trilobite Hill Limestone Member of Vandon Limestone, upper Cliefden Caves Limestone Subgroup; and at localities L143 and L144 in upper Billabong Creek Limestone, in vicinity of Billabong Creek road crossing, south of Gunningbland [full details of these localities are given by Percival 1991]. Description Shells generally small and thin, not exceeding 7.5 mm in length and 9.2 mm in width, with subquadrate to subrectangular outline; hingeline straight and wide, in all but one specimen just slightly narrower than maximum valve width which is approximately coincident with midlength, anterior margin broadly rounded; length:width ratio ranges from 0.65 to 0.88, with average of 0.77 for 18 specimens. Profile generally planoconvex, to weakly concavoconvex with tendency to geniculation dorsally in largest specimens; a subtle sulcus may develop in anteromedian sector of dorsal valve, with corresponding weak fold in ventral valve. Ornament finely and equally parvicostellate, lacking rugae; occasional concentric growth discontinuities may be present. Ventral interarea apsacline, with relatively wide delthyrium at least half Proc. Linn. Soc. N.S.W., 130, 2009 I.G. PERCIVAL Figure 7. Paromalomena zheni sp. nov. A — B: exterior and interior of ventral valve, MMF 44931. C: inte- rior of dorsal valve, MMF 44935. D: exterior of ventral valve, MMF 44936. E — F: exterior and interior of dorsal valve, MMF 44930. G: interior of ventral valve, holotype MMF 44932. H: interior of ventral valve, MMEF 44933. I — J: conjoined valves, ventral and dorsal exteriors respectively, MMF 44934. K: dorsal exterior of conjoined valves, MMF 44937. L: interior of juvenile dorsal valve, MMF 44938. M: interior of juvenile ventral valve, MMF 44944. N — O: conjoined valves, ventral and dorsal exteriors respectively, MME 44946. P—Q: exterior and interior of dorsal valve, MMF 44939. R: interior of ventral valve, MMF 44945. S: exterior of dorsal valve, MMF 44940. T: interior of dorsal valve, MMF 44941. U: interior of dorsal valve, MMF 44942. V: exterior of ventral valve, MMF 44943. Scale bar represents 1 cm. A — L from L138 (““Quondong”, Bowan Park, east of Cudal) Quondong Limestone, Bowan Park Subgroup; M, Q-V from L143, upper Billabong Creek Limestone at Billabong Creek road crossing south of Gunning- bland; N — O from L24, Trilobite Hill Limestone Member of Vandon Limestone, upper Cliefden Caves Limestone Subgroup at Licking Hole Creek near Walli. to three-quarters covered by prominent high convex Ventral interior: Pedicle foramen about pin- pseudodeltidium; a minute pedicle foramen is present hole size, encased in callus at extreme posterior of at apex of pseudodeltidium. Dorsal interarea barely _delthyrial cavity. Small teeth supported by receding evident, considerably lower than that of ventral valve; dental plates, below which extend anteriorly chilidial plates (if present) extremely delicate. divergent, subparallel or slightly convergent lateral Proc. Linn. Soc. N.S.W., 130, 2009 BRACHIOPODS FROM CENTRAL NEW SOUTH WALES muscle bounding ridges that rapidly decline in height and do not enclose muscle field anteriorly; diductors surround adductors that are embedded in shallow subcircular pit on low median ridge. Muscle field occupies approximately one-third valve length and less than one-quarter width. Mantle canals not observed. Dorsal interior: Cardinalia of glyptomenin type (Type C), with very delicate cardinal process lobes joined to fine, short socket ridges that diverge and curve to extend subparallel to hingeline; notothyrial platform absent; low, broad median ridge is barely developed in some larger specimens, otherwise lacking; side and transmuscle septa never developed; muscle scars not clearly defined. Mantle canals not discernible, due to thinness of shell material that reflects external costellae. Dimensions Valve length ranges from 3.2 mm to 7.5 mm, and valve width ranges from 4.3 mm to 9.2 mm (measurements from 18 individuals; no significant difference between ventral and dorsal valves). Holotype (ventral valve MMF 449372) is 7.5 mm long and 9.2 mm wide; majority of specimens cluster in the range of 5.0-6.5 mm long, and 5.5-8.5 mm wide. Discussion This new species shares many morphological characteristics with the cosmopolitan Late Ordovician (late Katian — Hirnantian) genus Paromalomena including shell profile and ornament, development of fold and sulcus anteriorly, and in most internal details. It differs from described species mainly in having a conspicuous pseudodeltidium, and in lacking a large chilidium and external rugae. Paromalomena typically occurs in deepwater settings (BA 4-6) in distinctive faunal associations such as the Foliomena fauna (e.g. Neuman 1994) and the younger Hirnantia fauna (e.g. Temple 1965). Like these species, P. zheni is quite thin-shelled, but unlike them it occurs in considerably shallower environments (BA 3) and is somewhat older (earliest Katian). Unlike species of Glyptomena, the new species has a furcitellin-like ornament (i.e. equally parvicostellate), and is generally planoconvex rather than concavo-convex, except in largest specimens. P. zheni 1s readily distinguished from Resupinsculpta cuprafodina, the other glyptomenin with which it is associated in the same strata in central NSW, by the latter’s resupinate profile, unequally parvicostellate ornament and presence of rugae. Glyptomenoides species differ in having an unequally parvicostellate ornament with rugae developed, and furthermore are quite distinct internally from P. zheni which lacks a stout myophragm and transmuscle septa. Distribution Limestones of early Eastonian (Ea2) age, equivalent to the earliest Katian Stage, in the Macquarie Arc, central NSW. Platymena Cooper, 1956 Type species: Platymena plana Cooper, 1956 Platymena? sp. Fig. 8 A-D Material MMF 36804, external mould of ventral valve; MMEF 36810, internal mould of dorsal valve; MMF 44968, internal mould of dorsal valve (not figured). Figure 8. Platymena? sp. A — B: latex replica of ventral valve exterior and interarea of conjoined valves, MMF 36804. C — D: latex replica and corresponding internal mould of dorsal valve, MMF 36810. All specimens from upper beds of the Gunningbland Formation on “Currajong Park”, Gunningbland. Scale bar represents 1 cm. 170 Proc. Linn. Soc. N.S.W., 130, 2009 1.G. PERCIVAL Locality All three known specimens from sandstones in upper Gunningbland Formation on “Currajong Park’, Gunningbland at locality L51 [see Percival 1979a for full details]. Description Transverse auriculate shell with maximum width at hingeline; lateral and anterior margins broadly rounded; profile apparently weakly concavo-convex, with median ventral fold; periphery of both valves dorsally geniculated. Length: width ratio 0.53 (ventral valve), 0.56 (dorsal valve). Ornament unequally parvicostellate, with 2-3 finer costellae separating relatively strongly accentuated costellae; very fine crowded concentric filae are just visible interstitially between costellae; three faint oblique rugae developed on posterolateral flanks. Ventral valve with low apsacline interarea, and narrow pseudodeltidium extending entire height of interarea. Interior details of ventral valve unknown. Dorsal valve interarea very low, orthocline to weakly anacline, with small, apparently complete chilidtum. Delicate cardinal process lobes are continuous laterally with fine, broadly divergent socket ridges; notothyrial platform beneath cardinal process is barely thickened above valve floor, extending anteriorly as a short, low median ridge; transmuscle septa very poorly developed. Musculature and mantle canals not deeply impressed; muscle field extends no more than one-third valve length. Broadly rounded subperipheral rim slightly raised above dorsal valve floor, geniculate dorsally in anterior portion; width of subperipheral rim greatest in posterolateral corner of valve. Dimensions MME 36804 (VV): length 14.8 mm, width 27.7 mm, MME 36810 (DV): length 17.1 mm, specimen width 26.8 mm; estimated complete width 30.8 mm; MME 44968 (DV): length 18.6 mm, width 28.8 mm. Discussion Lack of knowledge about interior details of the ventral valve prevents conclusive identification of this species as either Platymena or Glyptomena. In establishing both genera, Cooper (1956, p.882) commented upon differences between them, remarking on the flatness of the dorsal valve and thickened marginal region in Platymena. The delicate cardinalia and socket ridges, and weakly developed to barely perceptible septa in the dorsal muscle field Proc. Linn. Soc. N.S.W., 130, 2009 are more reminiscent of Glyptomena, and although no dorsal valve exteriors are known for the Gunningbland species, the sole internal mould seems to suggest a weakly concave (rather than planar) profile. However, the presence of a relatively prominent subperipheral rim is more characteristic of Platymena, to which this species is tentatively assigned. Distribution Gunningbland Formation (upper part) in vicinity of Gunningbland village, between Parkes and Bogan Gate, central west NSW; late Eastonian (Ea3-4) i.e. Katian. Superfamily Plectambonitoidea Jones, 1928 Family Leptellinidae Ulrich and Cooper, 1936 Subfamily Leptellininae Ulrich and Cooper, 1936 Shlyginia Nikitin and Popov, 1983 Type species: Shlyginia declivis Nikitin and Popov, 1983 Remarks In addition to describing S. printhiensis from Molong, NSW, the first species of Shlyginia known from outside Kazakhstan, Percival (in Percival et al., 2001) reviewed all six species previously attributed to this genus. All are similar with respect to general characteristics of the dorsal valve interior, whereas there is a wide variation in the size and disposition of the ventral muscle field. The type species, S. declivis, has a widely divergent ventral muscle field extending to about one-third valve length (Nikitin and Popov 1983, pl. 3, fig. 4; Cocks and Rong 2000, fig. 208, 3b — same specimen). In Shlyginia fragilis (Rukavishnikova, 1956) the ventral muscle field extends for about one- third valve length (Rukavishnikova 1956, pl. 2, fig. 18; Popov et al. 2002, pl. 6, figs 22, 25). Shlyginia extraordinaria (Rukavishnikova, 1956) has a very large ventral muscle field extending beyond mid valve length, in which the muscle impressions are conjoined medially for much of their length (Popov ef al. 2000 pl. 3, fig. 19; Popov and Cocks 2006, pl. 4 figs 22-23). The ventral muscle field of S. perplexa Nikitin and Popov, 1996 is much reduced, occupying no more than one-quarter to one-fifth valve length (Nikitin and Popov 1996, fig. 4 F-G). The NSW species S. printhiensis has a ventral muscle scar confined to the posterior third of the valve, whereas in the new species described below, the ventral muscle field just reaches (but never exceeds) half valve length. Excluded from Shlyginia is S. solida Nikitin and Popoy, 1984; the sturdy, apparently tubular dorsal BRACHIOPODS FROM CENTRAL NEW SOUTH WALES median septum of this species indicates that it belongs in Mabella Klenina, 1984. Also referred to Mabella on this same criterion is Dulankarella namasensis Klenina, 1984 (and its synonym D. subquadrata Klenina, 1984), previously assigned to Shiyginia by Nikitin and Popov (1996). Shlyginia rectangularis sp. nov. Fig. 9 A-X Diagnosis Transversely rectangular, dorsoventrally compressed Shlyginia with distinctive V-shaped incision at posterolateral extremities of ventral valve; muscle scar extending to midlength of ventral valve; 2-3 pairs of discrete nodes present on platform of dorsal valve laterally between muscle field and peripheral rim. Etymology Referring to rectangular outline. Material Holotype MMF 44959 (ventral valve); paratypes include MMF 44947 (conjoined valves), MMF 44948 (ventral valve), MMF 44949 (dorsal valve), MME 44950 (dorsal valve), MMF 44951 (conjoined valves), MMF 44952 (ventral valve), MMF 44953 (ventral valve), MMF 44954 (ventral valve), MMF 44955 (dorsal valve), MMF 44956 (dorsal valve), MME 44957 (ventral valve), MMF 44958 (dorsal valve), MMF 44960 (dorsal valve), and MMF 44961 (dorsal valve). All specimens are silicified. Localities Type locality is L142 (Paling Yards Creek section at “The Ranch”, Bowan Park), in Quondong Limestone, Bowan Park Subgroup; also found in same horizon at L138 (“Quondong”, Bowan Park, east of Cudal); occurs also at L24 (Licking Hole Creek area, Walli) in Trilobite Hill Limestone Member of Vandon Limestone, upper Cliefden Caves Limestone Subgroup; and at L143 in upper Billabong Creek Limestone, from outcrop in Billabong Creek at road crossing, south of Gunningbland [full details of these localities are given by Percival 1991]. Description Transversely rectangular shells with long, straight hingeline, lateral margins nearly straight and parallel to slightly convergent anteriorly, with broadly rounded anterior margin. Dorsoventrally compressed, planoconvex profile; maximum convexity close to anterior margin; ventral valve flattened medially, 172 becoming broadly sulcate anteromedially in largest specimens. Valve length between 4.4 and 8.8 mm, width 5.7 to 13.7 mm; length:width ratios in 13 specimens ranging from 0.55-0.69, with average of 0.61; maximum width at hingeline with slightly auriculate, posterolateral extremities in best preserved specimens, otherwise widest in posterior third of shell. Ornament finely unequally parvicostellate, very faintly impressed except for accentuated costellae, rarely lamellose peripherally in largest specimens. Ventral interarea low, apsacline, with upper third of delthyrium covered by small deltidium; dorsal interarea much lower, anacline, with very fine, paired chilidial plates flanking trifid cardinal process. Ventral valve interior: teeth small, unsupported by dental plates. Muscle field moderately to deeply impressed, adductor scars confined to a small median depression deep within delthyrium; diductors much larger, moderately divergent anteriorly, distinctly separated medially by fine ridge, and extending to mid valve length. Mantle canals of lemniscate type, with moderately strongly impressed vascula media and weaker vascula genitalia (sometimes not visible). Narrow, linear median depression extending from muscle field nearly to anterior margin of valve appears to exactly coincide with dorsal median septum. Dorsal valve interior: Anterior edge of hingeline thickened towards lateral extremities. Cardinalia typically leptellinine, trifid with prominent central ridge flanked by finer oblique lateral ridges, supported on a low thickened notothyrial platform. Socket ridges short, bladelike and pointed oblique to hingeline. Muscle field well defined by bounding ridges extending anteriorly from ends of socket ridges; muscle field bisected obliquely by low ridges that may represent proximal traces of vascula media. Solid ridge-like median septum, not expanding anteriorly, is separated from front of notothyrial platform by shallow depression; septum rises sharply and extends to approximately 0.8-0.85 valve length to merge with edge of barely undercut platform margin. Two to three pairs of discrete nodes are present on platform lateral to muscle field. Mantle canals beyond muscle field rarely impressed, possibly saccate. Dimensions Holotype MMF 44959 (ventral valve) is 7.2 mm long and 11.5 mm wide. Paratype MMF 44947 (conjoined valves) measures 7.4 mm in length, 12.1 mm in width, and 2.0 mm in thickness. Lengths of 12 other paratypes range from 4.4 mm to 8.8 mm, with most between 5.0-7.5 mm long; widths of 13 paratypes range from 5.7 mm to 13.7 mm, most are 9-12 mm wide. There is no appreciable difference Proc. Linn. Soc. N.S.W., 130, 2009 I.G. PERCIVAL Figure 9. Shlyginia rectangularis sp. nov. A— C: exterior of conjoined valves, ventral and dorsal respec- tively, and posterior profile (dorsal valve uppermost), MMF 44947. D — F: exterior, interior and lateral profile (posterior to left) of ventral valve, MMF 44952. G — I: exterior, interior and anterior profile of ventral valve, MMF 44948. J — L: conjoined valves, ventral and dorsal exteriors and posterior profile (ventral valve uppermost) respectively, MMF 44951. M: interior of ventral valve, MMF 44953. N: inte- rior of dorsal valve, MMF 44950. O: interior of dorsal valve, MMF 44955. P: interior of ventral valve, MMF 44954. Q: interior of dorsal valve, MMF 44958. R: interior of juvenile dorsal valve, MMF 44956. S —T: interior and exterior of dorsal valve, MMF 44960. U: interior of ventral valve, holotype MMF 44959. V: interior of ventral valve, MMF 44957. W: interior of dorsal valve, MMF 44961. X: interior of dorsal valve, MMF 44949. Scale bar represents 1 cm. A— C, G —I, N, X from L24, Trilobite Hill Limestone Member of Vandon Limestone, upper Cliefden Caves Limestone Subgroup at Licking Hole Creek near Walli; D — F, J—L, M, O, P, R from L135 (east of Copper Mine Creek, near Cliefden Caves), in Trilobite Hill Limestone Member of Vandon Limestone, upper Cliefden Caves Limestone Subgroup; Q, U, V from L142 (Paling Yards Creek section at “The Ranch”, Bowan Park), in Quondong Limestone, Bowan Park Subgroup; S — T from L138 (“Quondong”, Bowan Park, east of Cudal), Quondong Limestone, Bowan Park Subgroup; W from L143, upper Billabong Creek Limestone at Billabong Creek road crossing south of Gunningbland. between measurements of dorsal and ventral valves. shaped incisions at the posterolateral extremities of the ventral valve interior, and the presence of nodes on Discussion the platform lateral to the dorsal muscle field — serve Two distinctive morphological features — the V- to distinguish S. rectangularis from all other known Proc. Linn. Soc. N.S.W., 130, 2009 BRACHIOPODS FROM CENTRAL NEW SOUTH WALES species of Shlyginia. The function of the V-shaped incisions is not clear, although one likely explanation is that they interlock with corresponding thickened parts of the hingeline in the dorsal valve to strengthen articulation of the valves when open. Containment of the dorsal muscle field by bounding ridges is another characteristic feature of S. rectangularis. The large ventral muscle field of the new species is comparable only with that of S. extraordinaria which also has a similar trapezoidal outline, being noticeably widest at the hingeline. However, the dorsal interior of S. extraordinaria, illustrated by Popov et al. (2000, pl. 3, figs 18-20) and Popov and Cocks (2006, pl. 4, figs 25- 26), exhibits a much less robust median septum than does S. rectangularis. Unlike both S. extraordinaria and the other NSW species S. printhiensis, the new species lacks a well-defined marginal rim in the ventral valve; the median septum of S. rectangularis is also relatively much longer than that of S. printhiensis. Distribution Limestones of mid-Eastonian (Ea2) age, equivalent to basal Katian, throughout the Macquarie Arc in central NSW. Order Pentamerida Schuchert and Cooper, 1931 Suborder Syntrophiidina Ulrich and Cooper, 1936 Superfamily Camerelloidea Hall and Clarke, 1895 Family Parastrophinidae Schuchert and LeVene, 1929 Parastrophina Schuchert and LeVene, 1929 Type species: Atrypa hemiplicata Hall, 1847 Parastrophina sp. Fig. 10 A-G Material One fragmentary dorsal valve (MMF 44962) from L147, three ventral valves (all incomplete) MMF 44963-44965 from L24, and one partial ventral valve (MMF 44966) from L138 (doubtfully attributed). Localities Vandon Limestone (Trilobite Hill Limestone Member), Cliefden Caves Limestone Subgroup at locality L24, Licking Hole Creek, Walli; Checkers Member of Regans Creek Limestone at locality L147, “Red East”, Regans Creek southeast of Cargo; ventral valve from Quondong Limestone, Bowan Park Subgroup at locality L138, “Quondong”, Bowan 174 Park, east of Cudal is doubtfully attributed | full details of localities given by Percival (1991)]. Description Ventral valve: convex, smooth externally on posterior and lateral flanks, with shallow sulcus developed anteriorly, bearing 2-3 costae to form a weakly plicate anterior margin; internally with large subparallel dental plates extending to valve floor, bounding narrow, deep, parallel-sided sessile spondylium extending to approximately two-thirds valve length, supported anteriorly by very short median septum which barely extends beyond anterior edge of spondylium. Dorsal valve: smooth, convex posteriorly with prominent umbo (anterior part of valve not preserved); cardinal process lacking; deep narrow septalium present bounded by thin walls anteriorly convergent on to low thin median septum that extends anteriorly for an unknown distance; alate plates present. Dimensions Dorsal valve MMF 44962 L= (incomplete), width estimated at 20 mm. Ventral valve MMF 44963 L=4.5 mm, full width unknown. Ventral valve MMF 44966 W= 12.5 mm (incomplete), estimated width about 20 mm. 7.5 mm Discussion The available material, although incomplete, is assigned to Parastrophina rather than to the externally similar Camerella on the basis of the presence of alate plates in the sole dorsal valve. The ventral valve from the Quondong Formation at Bowan Park has the same smooth exterior, at least posteriorly, and similar dimensions to the other specimens. However, it is only doubtfully attributed to the same species, as evidence that the spondylium is supported above the valve floor at the front is lacking (this part of the shell being broken away). Alternatively, if the dental plates rest unsupported on the valve floor then this specimen may be better placed in Stenocamara Cooper, 1956. Numerous species of Parastrophina have been described, from North America (Cooper, 1956), Kazakhstan (Sapelnikov and Rukavishnikova 1975; Nikitin et al. 1996; Popov et al. 2002; Nikitin et al. 2006) and elsewhere, but it is difficult to make accurate comparisons between those (particularly when described from serial sections) and the sparse and incomplete silicified material from NSW. Distribution Rare in limestones of early Eastonian (Ea2) age, Proc. Linn. Soc. N.S.W., 130, 2009 I.G. PERCIVAL equivalent to earliest Katian, in the Macquarie Arc of central NSW. Eoanastrophia Nikiforova and Sapelnikovy, 1973 Type species: Eoanastrophia antiquata Nikiforova and Sapelnikov, 1973 Eoanastrophia? sp. Fig. 10 H-I Material One specimen, an incomplete dorsal valve, MMF 44967. Locality Quondong Limestone, Bowan Park Subgroup at locality L138, “Quondong”, Bowan Park, east of Cudal [full details of locality given by Percival Geom: Description Dorsal valve entirely costate with angular ribs, occasionally with intercalated costellae; internally with short septalium supported on long high median septum; very small sockets; crura present (preserved only on left-hand side of specimen); no cardinal process. Ventral valve not available for description. Dimensions Specimen 11.4 mm long and 9.4 mm wide (both dimensions incomplete). Discussion Similarly strongly costate parastrophinid genera include Eoanastrophia Nikiforova and Sapelnikov, 1973 and Maydenella Laurie, 1991. The latter genus, from the late Middle Ordovician Upper Cashions Creek Limestone in Tasmania, has a sessile septalium resting on the valve floor that is bounded by long subparallel hinge plates, whereas in Eoanastrophia the hinge plates converge onto a septum which Figure 10. A — G: Parastrophina sp. A — B: exterior and interior of partial dorsal valve, MMF 44962, specimen broken during photography; from Checkers Member of Regans Creek Limestone at locality L147, “Red East”, Regans Creek southeast of Cargo. C — D: interior and exterior of partial ventral valve, MME 44963; from Vandon Limestone (Trilobite Hill Limestone Member), Cliefden Caves Limestone Subgroup at locality L24, Licking Hole Creek, Walli. K — G: two interior views (the first slightly tilted to show dental plates extending to valve floor) and exterior of ventral valve, MMF 44966, doubtfully attrib- uted to this species; from Quondong Limestone, Bowan Park Subgroup at locality L138, “Quondong”, Bowan Park, east of Cudal. Scale bar representing 1 cm applies to all specimens in this figure. H -—I: Eoanastrophia? sp., exterior and interior of dorsal valve (interior view slightly tilted to better show septum supporting septalium), MMF 44967, from Quondong Limestone, Bowan Park Subgroup at local- ity L138, “Quondong”, Bowan Park, east of Cudal. Proc. Linn. Soc. N.S.W., 130, 2009 175 BRACHIOPODS FROM CENTRAL NEW SOUTH WALES supports the septalium (Laurie 1991, p. 85; Carlson 2002, p. 955-958). On this basis, the NSW specimen is most like Eoanastrophia, although as only one valve is known, the generic assignment is necessarily tentative. Distribution Presently known only from the one locality in the Quondong Limestone. ACKNOWLEDGMENTS David Barnes (NSW Department of Primary Industries) expertly prepared the photographic illustrations, and Cheryl Hormann drafted Figures 1 and 2. Reviews by Robin Cocks (Natural History Museum, London) and an anonymous referee greatly facilitated polishing of the manuscript for publication. Published with the permission of the Director, Geological Survey of New South Wales, NSW Department of Primary Industries. This paper is a contribution to IGCP Project No. 503: Ordovician Palaeogeography and Palaeoclimate. REFERENCES Bancroft, B.B. (1933). Correlation tables of the stages Costonian-Onnian in England and Wales. 4pp. Printed and published by the author, Blakeney, Glouchestershire, England (not sighted; fide Williams et al. 2000). Carlson, S.J. (2002). Suborder Syntrophiidina. In ‘Treatise on Invertebrate Paleontology, Part H. Brachiopoda (Revised), Vol. 4’ (Eds A. Williams, C.H.C. Brunton and S.J. Carlson) pp. 929-960. (The Geological Society of America: Boulder, and The University of Kansas Press: Lawrence). Cocks, L.R.M. (2005). Strophomenate brachiopods from the Late Ordovician Boda Limestone of Sweden: their systematics and implications for palaeogeography. Journal of Systematic Palaeontology 3, 243-282. Cocks, L.R.M. and Rong, J.-y. (2000). Order Strophomenida. In “Treatise on Invertebrate Paleontology, Part H. Brachiopoda (Revised), Vol. 2’ (Eds A. Williams, C.H.C. Brunton and S.J. Carlson) pp. 216-348. (The Geological Society of America: Boulder, and The University of Kansas Press: Lawrence). Cocks, L.R.M. and Zhan, R.-b. (1998). Caradoc brachiopods from the Shan States, Burma (Myanmar). Bulletin of the Natural History Museum, London (Geology) 54, 109-130. Cooper, G.A. (1956). Chazyan and related brachiopods. Smithsonian Miscellaneous Collections 127, 1245 pp. + 269 pl. Edgecombe, G.D. and Webby, B.D. (2006). The Ordovician encrinurid trilobite Sinocybele from New 176 South Wales and its biogeographic significance. Memoirs of the Association of Australasian Palaeontologists 32, 413-422. Edgecombe, G.D. and Webby, B.D. (2007). Ordovician trilobites with eastern Gondwanan affinities from central-west New South Wales and Tasmania. Memoirs of the Association of Australasian Palaeontologists 34, 255-281. Klenina, L.N. (1984). Brakhiopody 1 biostratigrafiya srednego 1 verkhnego ordovika Chingiz- Tarbagatayskogo megantiklinortya. [Brachiopods and biostratigraphy of the Middle and Upper Ordovician of the Chingiz-Tarbagatai Meganticlorium]. In Klenina, L.N., Nikitin, I.F. and Popov, L.E. ‘Brakhiopody 1 biostratigrafiya srednego 1 verkhnego ordovika khrebta Chingiz’, pp. 6-125. (Nauka, Alma- Ata). Laurie, J.R. (1991). Articulate brachiopods from the Ordovician and Lower Silurian of Tasmania. In P.A. Jell (ed.) “Australian Ordovician Brachiopod Studies’. Memoirs of the Association of Australasian Palaeontologists 11, 1-106. McLean, R.A. (1974). The geology of the Regans Creek area, near Cargo, central New South Wales. Proceedings of the Linnean Society of New South Wales 98, 196-221. McLean, R.A. and Webby, B.D. (1976). Upper Ordovician rugose corals of central New South Wales. Proceedings of the Linnean Society of New South Wales 100, 231-244. Neumann, R.B. (1994). Late Ordovician (Ashgill) Foliomena fauna brachiopods from northeastern Maine. Journal of Paleontology 68, 1218-1234. Nikiforova, O.I. and Andreeva, O.N. (1961). Biostratigrafiya Paleozoya Sibirskoi Platformi, vol. |. Stratigrafiia Ordovika 1 Silura Sibirskoi Platformy i ee Paleontologichesko Obosnovanie (Brakhiopody) [Biostratigraphy of the Paleozoic of the Siberian Platform, vol. 1. Ordovician and Silurian stratigraphy of the Siberian Platform and its paleontological basis (Brachiopoda)]. Vsesoiuznyi Nauchno- Issledovatel skii Geologicheskii Institut (VSEGED), Trudy 56, 1-412. Nikiforova, O.I. and Sapelnikoy, V-.P. (1973). Nekotorye dreynie pentameridy Zeravshanksogo khrebta. [Some fossil pentamerids from the Zeravshan Range]. In Sapelnikov, V.P. and Chuvashoy, B.I. (eds), Materialy po paleontologii srednego Paleozoia Uralo — Tian’shan’skoi oblasti. Akademiya Nauk SSSR, Ural ’ski Nauchnyi Tsentr, Institut Geologii I Geokimii, Trudy 99, 64-81. Nikitin, I.F. and Popov, L.E. (1983). Sredneordovikskie ortatsei 1 plektambonitatsei severnogo Priishimya 1 basseyna reki Akhzhar v Tsentral’nom Kazakhstane (Brakhiopody). [Middle Ordovician orthids and plectambonitaceans from northern Priishimiya and the River Azkhar in central Kazakhstan (Brachiopoda)]. Ezhegodnik Vsesoyuznogo Paleontologicheskogo Obshchestva 26, 228-247. Proc. Linn. Soc. N.S.W., 130, 2009 I.G. PERCIVAL Nikitin, I.F. and Popov, L.E. (1984). Brakhiopody bestamakskoy i sargaldakskoy svit (srednii ordovik). [Brachiopods from the Bestamak and Sargaldak Formations (Middle Ordovician)]. In Klenina, L.N., Nikitin, I.F., and Popov, L.E. “Brakhiopody i biostratigrafiya srednego 1 verkhnego ordovika khrebta Chingiz’, pp. 126-166. (Nauka, Alma-Ata). Nikitin, I.F. and Popoy, L.E. (1996). Strophomenid and triplesiid brachiopods from an Upper Ordovician carbonate mound in central Kazakhstan. A/cheringa 20, 1-20. Nikitin, I.F., Popov, L.E. and Bassett, M.G. (2006). Late Ordovician rhynchonelliformean brachiopods of north-eastern Central Kazakhstan. In Bassett, M.G. and Deisler, V.K. (eds) “Studies in Palaeozoic Palaeontology’. National Museum of Wales Geological Series No. 25, pp. 223-294. (National Museum of Wales, Cardiff). Nikitin, I.F., Popov, L.E., and Holmer, L.E. (1996). Late Ordovician brachiopod assemblage of Hiberno- Salairian type from Central Kazakhstan. Geologiska Foéreningens i Stockholm Férhandlingar 118, 83-96. Percival, I.G. (1976). The geology of the Licking Hole Creek area, near Walli, central western New South Wales. Journal and Proceedings of the Royal Society of New South Wales 109, 7-23. Percival, I.G. (1978). Inarticulate brachiopods from the Late Ordovician of New South Wales, and their palaeoecological significance. A/cheringa 2, 117-141. Percival, I.G. (1979a). Ordovician plectambonitacean brachiopods from New South Wales. A/cheringa 3, 91-116. Percival, I.G. (1979b). Late Ordovician articulate brachiopods from Gunningbland, central western New South Wales. Proceedings of the Linnean Society of New South Wales 103, 175-187. Percival, I.G. (1991). Late Ordovician articulate brachiopods from central New South Wales. In P.A. Jell (ed.) “Australian Ordovician Brachiopod Studies’. Memoirs of the Association of Australasian Palaeontologists 11, 107-177. Percival, I.G. (1992). Ordovician brachiopod biostratigraphy of central-western New South Wales. In B.D. Webby and J.R. Laurie (eds) “Global Perspectives on Ordovician Geology (Proceedings of the 6th International Symposium on the Ordovician System)’, pp. 215-229. (Balkema, Rotterdam). Percival, I.G. and Glen, R.A. (2007). Ordovician to earliest Silurian history of the Macquarie Arc, Lachlan Orogen, New South Wales. Australian Journal of Earth Sciences 54, 143-165. Percival, I.G. and Webby, B.D. (1996). Island benthic assemblages: with examples from the Late Ordovician of Eastern Australia. Historical Biology 11, 171-185. Percival, I.G., Webby, B.D. and Pickett, J.W. (2001). Ordovician (Bendigonian, Darriwilian to Gisbornian) faunas from the northern Molong Volcanic Belt of central New South Wales. Alcheringa 25, 211-250. Proc. Linn. Soc. N.S.W., 130, 2009 Pickett, J.W. and Percival, I.G. (2001). Ordovician faunas and biostratigraphy in the Gunningbland area, central New South Wales. Alcheringa 25, 9-52. Popov, L.E. and Cocks, L.R.M. (2006). Late Ordovician brachiopods from the Dulankara Formation of the Chu-Ili Range, Kazakhstan: their systematics, palaeoecology and palaeobiogeography. Palaeontology 49, 247-283. Popov, L.E., Cocks, L.R.M. and Nikitin, I.F. (2002). Upper Ordovician brachiopods from the Anderken Formation, Kazakhstan: their ecology and systematics. Bulletin Natural History Museum London (Geology) 58, 13-79. Popov, L.E., Nikitin, I.F. and Cocks, L.R.M. (2000). Late Ordovician brachiopods from the Otar Member of the Chu-Ili Range, south Kazakhstan. Palaeontology 43, 833-870. Rong, J.-y. (1984). Brachiopods of latest Ordovician in the Yichang district, western Hubei, central China. Stratigraphy and Palaeontology of Systemic Boundaries in China: Ordovician-Silurian boundary 1, 111-176. Rong, J.-y. and Cocks, L.R.M. (1994). True Strophomena and a revision of the classification and evolution of strophomenoid and ‘strophodontoid’ brachiopods. Palaeontology 37, 651-694. R6omusoks, A. (1993). Four new brachiopod genera of the subfamily Oepikinae (Strophomenacea) from the Ordovician of Estonia. Proceedings of the Academy of Sciences of the Estonian SSR, Geology 42, 48-57. R6dmusoks, A. (2004). Ordovician strophomenoid brachiopods of northern Estonia. Fossilia Baltica 3, 151 pp. + 34 pl. Rukavishnikova, T.B. (1956). Brakhiopody ordovika Chu- Tliyskikh gor. [Ordovician brachiopods of the Chu-Ili Range]. Trudy Geologicheskogo Instituta Akademii Nauk SSSR 1, 105-168. Sapelnikov, V.P. and Rukavishnikova, T.B. (1975). “Verkhneordovikskie, Siluriiskie i Nizhnedevonskie Pentameridy Kazakhstana’ [Upper Ordovician, Silurian and Lower Devonian pentamerids of Kazakhstan]. Akademia Nauk, Ural’skii Nauchnyi Tsentr, Institut Geologii i Geofiziki, 227 pp. + 43 pls. (Izdatel’stvo “Nauka”, Sverdlovsk). Schuchert, C. and LeVene, C.M. (1929). New names for brachiopod homonyms. American Journal of Science 17, 119-122. Semeniuk, V. (1970). The Lower-Middle Palaeozoic stratigraphy of the Bowan Park area, central-western New South Wales. Journal and Proceedings of the Royal Society of New South Wales 103, 15-30. Semeniuk, V. (1973). The stratigraphy of the Bowan Park Group, New South Wales. Journal and Proceedings of the Royal Society of New South Wales 105, 77-85. Temple, J.T. (1965). Upper Ordovician brachiopods from Poland and Britain. Acta Palaeontologica Polonica 10, 379-427. Ulrich, E.O. and Cooper, G.A. (1936). New genera and species of Ozarkian and Canadian brachiopods. Journal of Paleontology 10, 616-631. BRACHIOPODS FROM CENTRAL NEW SOUTH WALES Webby, B.D. (1973). Remopleurides and other Upper Ordovician trilobites from New South Wales. Palaeontology 16, 445-475. Webby, B.D. (1974). Upper Ordovician trilobites from central New South Wales. Palaeontology 17, 203- Dar, Webby, B.D. and Morris, D.G. (1976). New Ordovician stromatoporoids from New South Wales. Journal and Proceedings of the Royal Society of New South Wales 109, 125-135. Webby, B.D. and Packham, G.H. (1982). Stratigraphy and regional setting of the Cliefden Caves Limestone Group (Late Ordovician), central-western New South Wales. Journal of the Geological Society of Australia 29, 297-317. Williams, A., Brunton, C.H.C. and Carlson, S.J. (eds). (2000). ‘Treatise on Invertebrate Paleontology, Part H. Brachiopoda (Revised), Vol. 3’. (The Geological Society of America: Boulder, and The University of Kansas Press: Lawrence). Zhan, R.-b., Jin, J., Rong, J.-y., Chen, P.-f. and Yu, G.- h. (2008). Strophomenide brachiopods from the Changwu Formation (Late Katian, Late Ordovician) of Chun’an, western Zheijiang, south-east China. Palaeontology 51, 737-766. Zhen, Y.-y. and Pickett, J.W. (2008). Ordovician (Early Darriwilian) conodonts and sponges from west of Parkes, central New South Wales. Proceedings of the Linnean Society of New South Wales 129, 57-82. 178 Proc. Linn. Soc. N.S.W., 130, 2009 Rare Fossils (Conulata; Rostroconchia; Nautiloidea) from the Late Ordovician of Central New South Wales IAN G. PERCIVAL Geological Survey of New South Wales, Department of Primary Industries, 947-953 Londonderry Road, Londonderry, NSW 2753, Australia (ian.percival@dpi.nsw.gov.au). Percival, I.G. 2009. Rare fossils (Conulata; Rostroconchia; Nautiloidea) from the Late Ordovician of central New South Wales. Proceedings of the Linnean Society of New South Wales 130, 179-191. Four decades of detailed palaeontological investigations into highly fossiliferous Upper Ordovician strata of the Macquarie Arc in central New South Wales has revealed several unique specimens which in some instances represent the only known examples of phyla or subphyla in this region. Conulariids have not previously been reported from Ordovician rocks in NSW; here is documented Conularia sp., known from one specimen found in the Fossil Hill Limestone, and several microscopic specimens of different genera (including Metaconularia? sp., and the new genus and species Microconularia fragilis) from deep water allochthonous limestones (Malongulli Formation, and Downderry Limestone Member of the Ballingoole Limestone). The first Ordovician rostroconch mollusc from NSW is described from a solitary individual of Eopteria, from the top of the Malongulli Formation. A coiled nautiloid tentatively identified as Plectoceras from the Gunningbland Formation, again represented by a single specimen, is also described and illustrated. Manuscript received 1 December 2008, accepted for publication 16 February 2009. KEY WORDS: Conulariid, Late Ordovician, Macquarie Arc, Nautiloid, Rostroconch INTRODUCTION Rare fossils, often represented by unique specimens, can sometimes be overlooked in systematic documentation of a fauna, particularly if they are not spectacular in appearance or preservation. Yet such fossils, even when fragmentary or incomplete, by their very presence can be quite significant biogeographically. Despite intensive collecting over more than thirty years (and in some cases around four decades), the examples described in this paper are the only known specimens of conulariids, a rostroconch mollusc, and a genus of tarphyceratid nautiloid that have been found in Upper Ordovician rocks of the Macquarie Arc in central New South Wales. Their uniqueness well qualifies them to be described and illustrated for the first time. Stratigraphic setting The Cliefden Caves Limestone Subgroup and the overlying Malongulli Formation occur in the Walli area, between Mandurama and Canowindra in central NSW (Figure 1). Outcrop of these units has been mapped in detail south of the Belubula River by Webby and Packham (1982) in the vicinity of Cliefden Caves, and by Percival (1976) in the Licking Hole Creek area, adjoining to the west. Webby and Packham (1982) established the stratigraphic nomenclature of the Cliefden Caves Limestone Subgroup, comprising three formations (in ascending order: Fossil Hill Limestone, Belubula Limestone, Vandon Limestone), with the first and last of these subdivided into a number of members. Conularia sp. 1s represented by a single specimen (described herein) collected by G.H. Packham in the late 1960s from the Taplow Limestone Member of the Fossil Hill Limestone in the section west of the “Boonderoo” shearing shed (Webby and Packham 1982, fig. 3, p.302). No other material of this species has been found in this or any other level in the Cliefden Caves Limestone Subgroup, despite intensive palaeontological investigation of the area over the past four decades. The age of the Fossil Hill Limestone is early Eastonian (Eal), equivalent to latest Sandbian in the middle Late Ordovician. The Taplow Limestone Member was deposited in shallow RARE FOSSILS FROM THE LATE ORDOVICIAN Bogan Gate L5ty, a * BILLABONG CREEK ~ LIMESTONE e e PARKES Gunningbland e FORBES ® Evugowra NEW SOUTH WALES Bathurst | . SYDNEY < 400km N Molong | SEEDY CREEK i LIMESTONE U Cudal e ORANGE e Ww — sowan PARK CARGO CREEK pamRP ci LIMESTONE , \ Cargo : _— REGANS CREEK ~ LIMESTONE ! ¢ CANOMODINE — \ - LIMESTONE ® Canowindra Sia Licking Hole Creek — CLIEFDEN CAVES +— LIMESTONE ~~ SUBGROUP 6 Mandurama ® \Voodstock Figure 1. Locality map showing sites in central New South Wales yielding the Late Ordovician fossils described in this paper. Outcrop of main Upper Ordovician limestone units shown in black; localities (L37, L51) in overlying Upper Ordovician clastic-dominated units are shown by spots. turbulent water interpreted as Benthic Assemblage (BA) 2 in depth (Percival and Webby 1996); the conulariid was obtained from skeletal grainstones in the middle to upper part of the member, overlying Tetradium cribriforme coral banks. A solitary outcrop of allochthonous limestone at the top of the Malongulli Formation on the north- east flank of Malongulli Trig (Percival 1976) directly overlies graptolitic shale of early Bolindian age (Bol, Zone of Climacograptus uncinatus), equivalent to the latest Katian stage of the Late Ordovician. A very diverse fauna — including stromatoporoids (Webby and Morris 1976), radiolaria (Webby and Blom 1986), sponge spicules (Webby and Trotter 1993), and brachiopods including lingulates (Percival et al. 1999), strophomenoids and orthoids (Percival 2005), accompanied by numerous fragments of the nautiloid Bactroceras latisiphonatum Glenister, 1952 (Stait et al. 1985) — is known from acid-processed 180 residues of the limestone. Also present in the residues are extremely rare conulariid remains, including a single microscopic conulariid specimen designated as Microconularia fragilis gen. et sp. nov. and fragments of a separate conulariid with distinctive pustulose ornamentation, and a unique specimen of an articulated rostroconch identified as Eopteria sp. which, although fragmentary, is recognizable as the first (and only) known example of this Class in the Upper Ordovician of NSW. The conulariid material and the rostroconch are described herein. The allochthonous limestone is interpreted as having been initially deposited as periplatformal ooze on the upper slope (Webby 1992) in BA 4 water depths, prior to being displaced (after lithification) downslope to its present BA 5 setting. Thus the conulariids occurring at this level lived at considerably greater depths than the larger Conularia sp. from the Taplow Limestone Member. Proc. Linn. Soc. N.S.W., 130, 2009 I.G. PERCIVAL The Bowan Park Limestone Subgroup in the area east of Cudal (Fig. 1) spans a similar age range to the Cliefden Caves Limestone Subgroup and the lower part of the Malongulli Formation, i.e. early (Eal) to late (Ea 3-4) Eastonian, as indicated by conodonts studied by Zhen et al. (1999). The Bowan Park area was mapped in detail by Semeniuk (1973) who established the internal stratigraphy of formations and members in use today. One specimen of Microconularia fragilis gen. et sp. nov. is known from residues of the Downderry Limestone Member (of the Ballingoole Limestone at the top of the Bowan Park Subgroup), which is interpreted as a submarine channel-fill deposit emplaced at water depths approximating BA 4 environments. The Downderry conulariid therefore occupied a comparable habitat to that of the conspecific example from allochthonous limestone at the top of the Malongulli Formation. Upper Ordovician rocks in the Gunningbland area, west of Parkes on the western side of the Macquarie Arc (Pickett and Percival 2001), include the Billabong Creek Limestone (the upper part of which is correlative with the Cliefden Caves Limestone Subgroup), and the overlying clastic- dominated Gunningbland Formation which was deposited contemporaneously with the Malongulli Formation, though in slightly lesser water depths. Faunas of the Gunningbland Formation are dominated by trilobites (Edgecombe and Webby 2006, 2007) and brachiopods (Percival 1978, 1979a, 1979b, 2009). Stait et al. (1985) previously described two coiled nautiloids from this formation, including a single specimen each of Paradiscoceras dissitum and an indeterminate tarphyceratid. The specimen of Plectoceras? sp. described herein is the best preserved tarphyceratid nautiloid known from this level (a further fragmentary coiled nautiloid is documented by illustration only). These specimens, of late Eastonian (Ea3) age, equivalent to early Katian, are externally similar to slightly younger tarphyceratids documented by Percival et al. (2006). Systematic palaeontology Type material, comprising specimens described and illustrated or listed herein, is curated in the palaeontological collections of the Geological Survey of New South Wales (designated MMMC for microfossil specimens, and MMF for macrofossils). For brevity, authorship of taxonomic hierarchy above genus level is not cited in the References; these bibliographic sources are listed in Leme et al. (2008) for conulariids, Pojeta and Runnegar (1976) for rostroconchs, and Furnish and Glenister (1964) for tarphyceratid nautiloids. Proc. Linn. Soc. N.S.W., 130, 2009 Phylum Cnidaria Class Scyphozoa Goette, 1887 Subclass Conulata Moore and Harrington, 1956 Order Conulariida Miller and Gurley, 1896 Suborder Conulariina Miller and Gurley, 1896 Family Conulariidae Walcott, 1886 Conularia Miller, in Sowerby 1821 Type species: Conularia quadrisulcata Miller, 1821 Conularia sp. Fige2 Material A single incomplete specimen, MMF 44969a-b, represented by a natural cast and an associated partial external mould. Description The sole specimen includes the upper two- thirds (approximately) of one individual, extending 25.1 mm in length from the top of the apertural lobes; maximum width immediately below aperture is 9.1 mm. Cross-section quadrate, profile steeply pyramidal with planar to very slightly convex faces (slightly distorted in preservation) that gently taper apically, with apical angle estimated to be 8°; apex not preserved. Ornament consists of narrow, gently arched transverse ribs (23 per cm) that are defined by pair of closely-spaced parallel ridges, separated by interspaces up to three times as broad as the ribs; interspace ridges barely visible on one face (Fig. 21). Midline variably expressed, either as a very narrow ridge (suggestive of an internal carina) across which the transverse ribs meet in opposition (Fig. 2H), or a vertical discontinuity across which ribs alternate (Fig. 21). Transverse ribs are almost everywhere non- tuberculate except for isolated section of one face (Fig. 2H). Corner sulcus flat-bottomed, with transverse ribs continuous between adjacent faces. Apertural lobes triangular in outline and broadly convex in profile, with continuation of midline; individual lobes are near vertical in orientation, surrounding a large open aperture. No internal features preserved. Discussion Conulariids are very rare in the Ordovician of Australia; only a single species has previously been described from Tasmania by Parfrey (1982), who established a new genus and _ species, Tasmanoconularia tuberosa, based on a solitary partially fragmented specimen (nevertheless with excellent surface detail) preserved in the Westfield Sandstone of the Florentine Valley. The brachiopod 181 RARE FOSSILS FROM THE LATE ORDOVICIAN Proc. Linn. Soc. N.S.W., 130, 2009 182 I.G. PERCIVAL fauna of this unit (Laurie 1991; age revised by Rong et al. 1994) contains species of Hirnantia, Kinnella, Eospirifer, Cryptospira, Onniella? and Isorthis (Ovalella), which are representative of the Hirnantia fauna of Hirnantian (latest Ordovician) age. Parfrey (1982) distinguished Tasmanoconularia from Conularia and other genera included in the subfamily Conulariinae Walcott, 1886, by virtue of the Tasmanian conulariid having a distinct corner furrow which interrupted the continuity of the majority of transverse furrows between adjacent faces. This characteristic suggested affinities with the Paraconulariinae Sinclair, 1952. Subsequent opinion (Van Iten and Vyhlasova 2004) and cladistic analysis (Leme et al. 2008, page 652) has concluded that Tasmanoconularia is most likely identical with Conularia, with Leme et al. (2008) advocating that all previously-proposed families and subfamilies of conulariids (with the exception of the Conulariidae Walcott, 1886) be regarded as invalid. Comparison of C. tuberosa (Parfrey, 1982) with C. sp. reveals significant differences in size and ornamentation, sufficient to easily distinguish the two forms. The Tasmanian species is considerably wider, attaining a width estimated at 15 mm in an incomplete specimen, and is much more sharply tapering towards the apex than is the NSW species. The ribs of C. tuberosa are crowded together (35-38 per cm) whereas those of C. sp. are considerably less crowded (23 per cm). Both species are finely tuberculate, C. tuberosa conspicuously so; although C. sp. appears to be almost exclusively devoid of tubercles, this is most likely an artifact of preservation, as they are present in one small area (Fig. 2H) of a face that is less weathered. Conularia is a long-ranging cosmopolitan genus with numerous species; furthermore, the cladistic analysis of Leme et al. (2008) suggests that several other genera should probably be regarded as synonyms of Conularia. Comparison of the NSW species with others assigned to Conularia or its synonyms seems to be of doubtful value until the genus as a whole is revised. Coarsely crystalline calcite infilling the sole specimen of C. sp. has destroyed definitive evidence of carinae and ridges internal to the corners and midline (although it is possible the midline is strengthened internally by a carina — see Fig. 2H). Such features are significant criteria distinguishing genera and species of conulariids (Van Iten 1992, Jerre 1994, Leme et al. 2008), and their absence hinders comparisons with established taxa. It is appropriate here to compare C. sp. with Late Silurian conulariids revised or newly described from central NSW by Sherwin (1970), as these forms are closest in age and geography. Mesoconularia webbyi Sherwin, 1970 has an identical apical angle of 8° and generally comparable dimensions; however, transverse ribs on this species are more than twice as crowded as are those on the Late Ordovician C. sp., and are always offset across the midline. Paraconularia packhami Sherwin, 1970, has a very similar apical angle and spacing of transverse ribs compared to the older Conularia sp., but in P. packhami the arched transverse ribs are disjunct and apically depressed at the midline, whereas in Conularia sp. the transverse ridges are evenly convex toward the aperture and may be both continuous and alternating across the midline. Distribution Only known from the Taplow Limestone Member of the Fossil Hill Limestone, Cliefden Caves Limestone Subgroup; early Eastonian (Eal) age, equivalent to latest Sandbian. Microconularia gen. nov. Type species (by monotypy): Microconularia fragilis gen. et sp. nov. Diagnosis A microscopic conulariid with non-tuberculate widely-spaced transverse ribs, lacking a midline; corners rounded, without furrows. Discussion Most Ordovician conulartids are more than 25 mm in length, with only two previously-described species being less than one-tenth this (Leme et al. 2003, fig. 5). Size would not normally be considered Figure 2 (LEFT). Conularia sp. A— D: Four faces of internal cast, MMF 44969; E, G, views of corners of this specimen; F, detail of area of corner outlined on E, showing continuation of transverse ridges across corner sulcus; H, detail of area of face outlined on B, note minute nodes present on four transverse ridges adjacent to midline in lower part of enlargement, and continuation of majority of transverse ridges across midline; I, detail of area of face outlined on D, showing disjunct transverse ridges at midline, and suggestion of interspace ridges in upper part of enlargement. Scale bar in centre of upper row applies to A-E and G; scale bar beneath F applies only to the three enlargements. In both instances the scale bar represents five mm. From Taplow Limestone Member of Fossil Hill Limestone, near Cliefden Caves. Proc. Linn. Soc. N.S.W., 130, 2009 183 RARE FOSSILS FROM THE LATE ORDOVICIAN as a distinguishing generic character, but in the case of conulariids there seems to be a clear dichotomy between those forms commonly found in inner and outer shelf environments in a variety of lithologies (where the overwhelming majority are macrofossils), and other taxa that are generally known only from fragmentary remains or microfossils recovered in acid-insoluble residues of limestones. The latter may range from relatively shallow to moderate water depths (BA 2-3), such as those described from Silurian limestones of central NSW (Bischoff 1973) and the island of Gotland, Sweden (Jerre 1993), to deep water (BA 4) settings as interpreted for the forms described here. One genus is less than 2 mm long, and appears to be quite distinct from many described macrofossil conulariids in lacking a definite midline, and in not developing furrows along the corners. Certainly these characteristics seem to qualify for differentiation at genus level, and hence the new genus Microconularia is proposed. Teresconularia Leme et al., 2003, from the Lower Ordovician Santa Victoria Group of the Cordillera Oriental, northwestern Argentina, shares with Microconularia the attributes of minute size (length 1.4 mm) and rounded corners lacking a sulcus. However, the Argentine genus is considerably more widely expanding than is Microconularia, and the latter genus bears much coarser transverse ribs with strongly angular profiles. The ornament of Yeresconularia is very fine and crowded by comparison. There is no evidence of a midline on the faces of Microconularia, whereas in Teresconularia a midline is present, albeit very faintly, being marked by a slight deflection of the otherwise confluent transverse ribs. Climacoconus pumilus (Ladd, 1929), most recently described and illustrated from the Upper Ordovician Maquoketa Formation of northeastern Iowa by Van Iten et al. (1996), is another unusually tiny conulariid up to 2.5 mm in length. Although it resembles Microconularia in its low apical angle and coarse transverse ribs, the two genera are readily distinguished by the pronounced midline and corner sulcus of C. pumilus. The maximum length of Eoconularia loculata (Wiman, 1895), from the Silurian Hemse Beds of Gotland, is estimated by Jerre (1994) at 10 mm, approximately 6-7 times as large as Microconularia. It resembles the new genus in lacking a midline, and has a similar gradually tapering shape and coarse transverse ribs. However, the presence of a corner sulcus in E. /oculata distinguishes it from Microconularia. The distinctive internal septa of E. loculata have not been observed in the two known specimens of the new genus. 184 Microconularia fragilis gen. et sp. nov. Fig. 3 Diagnosis As for genus. Etymology Genus name in reference to the microscopic size of the test; species name in reference to the thin and fragile nature of the specimens. Material Holotype MMMC 4388 from L37, allochthonous limestone at top of Malongulli Formation, head of Sugarloaf Creek on northeast flank of Malongulli Trig, near Cliefden Caves; paratype MMMC 4389, a fragmentary specimen from the same locality as the holotype; paratype MMMC 4390 from Downderry Limestone Member of the Ballingoole Limestone, Bowan Park Subgroup, near Malachis Hill at Bowan Park. Description Test minute, less than 2 mm in length, very gradually tapering with apical angle of the order of 1-3°; cross-section quadrate, with flat to slightly concave faces ornamented with relatively coarse Figure 3. A—C. Microconularia fragilis gen. et sp. nov. A — B: Holotype, MMMC 4388, view show- ing corner, and lateral view of face. C: Paratype MMMC 4390. Scale bar represents one mm. Holo- type from locality L37, allochthonous limestone at top of Malongulli Formation on flank of Malongul- li Trig; earliest Bolindian (Bol) age. Specimen C from Downderry Limestone Member of the Ball- ingoole Limestone, Bowan Park Subgroup, near Malachis Hill at Bowan Park, late Eastonian (Ea3) age. Proc. Linn. Soc. N.S.W., 130, 2009 I.G. PERCIVAL widely and evenly spaced transverse ribs separated by interspaces of similar length; 14-19 ribs per mm; ribs have angular profile (where not eroded) and are gently and evenly arched; midline lacking. Corners of test rounded without any furrow; transverse ribs from adjacent faces are not continuous around corners, but appear to be offset and alternate so that a rib passes abruptly into an adjacent interspace. Apertural lobes and apex not preserved in available specimens. Internal features unknown. Dimensions Holotype MMMC 4388: maximum width 0.3 mm. Paratype MMMC 4390: length 1.7 mm, maximum width 0.4 mm length 1.5 mm, Discussion This exceptionally rare conulariid is represented in two localities, both in allochthonous limestones of BA 4 original depositional depth (inferred on the basis of associated faunas) that have been redeposited downslope. Age of these horizons is reasonably contemporaneous (late Eastonian to early Bolindian). Distribution Deepwater strata of late Eastonian (Ea3-4) to earliest Bolindian (Bol) age, equivalent to Katian Stage, in central NSW. Metaconularia Foerste, 1928 Type species: Conularia aspersa Lindstrém, 1884 Metaconularia? sp. Fig. 4 Figure 4. A— D Metaconularia? sp. A: exterior fragment, MMMC 4391. B: exterior fragment, MMMC 4392. C — D: interior and exterior of fragment, MMMC 4393. Scale bar represents one mm. All speci- mens from locality L37, allochthonous limestone at top of Malongulli Formation on flank of Malongulli Trig; earliest Bolindian (Bol) age. Proc. Linn. Soc. N.S.W., 130, 2009 185 RARE FOSSILS FROM THE LATE ORDOVICIAN Material Fragments with tuberculate ornamentation are uncommon in residues of acid-etched limestones at locality L37, allochthonous limestone at top of the Malongulli Formation, head of Sugarloaf Creek on northeast flank of Malongulli Trig, near Cliefden Caves. Three representative specimens, MMMC 4391 — 4393, are illustrated. Description All material consists of incomplete fragments of the test, displaying distinctive coarse and fine tuberculate ornamentation on the exterior surface. The fragments are flat to gently convex, sometimes bearing shallow sulci or furrows, and presumably represent portions of the faces of the theca. Largest fragment observed is 6.3 mm in length. Tubercles are irregularly distributed, generally crowded along shallow furrows in the shell surface (occasionally, a furrow is underlain by a septum on the interior surface of the test) and more scattered on the flanks adjacent to the furrows. A subtle to moderately strongly expressed longitudinal and _ transverse arrangement of the tubercles into columns and rows is often discernable; the rows may be oblique (Fig. 4B) or perpendicular (Fig. 4D) to the main axis of the specimen. Many (if not all) of the tubercles are hollow, observable where the tips have been eroded. Transverse ridges and interrods are lacking, and sharply defined midlines are not present. Corners between faces are unknown in the available material. Internal septa are low narrow linear features, not twinned; remainder of interior surface is smooth. Discussion The tuberculate ornament and absence of transverse ridges on the faces readily distinguishes these fragments from specimens of Microconularia with which they are associated in the acid-etched residues. Where septa are present internally, their surficial expression is a crowding of tubercles along a shallow linear depression or furrow; there is no development of a deep narrow midline such as is seen in conulariid fragments from the Silurian age Boree Creek Formation of central NSW (Bischoff 1978, pl. 1, fig. 1la-b). Jerre (1993) discussed and figured several conulariid fragments with comparable tuberculate ornamentation that he referred to Metaconularia aspersa (Lindstrém, 1884) from the Silurian of Gotland. Bischoff (1973) also recognized similar fragments from both Silurian (Bischoff 1973, pl. 2, figs. 15 and 17) and Ordovician (pl. 3, fig. 11) horizons, but did not attribute these to genera. Metaconularia 186 ranges from the Middle and Late Ordovician (Van Iten and Vyhlasova 2004, fig. 14.1) through the Silurian (Leme et al. 2008). The specimens from NSW are too incomplete for definitive identification, so they are tentatively assigned to Metaconularia pending collection of more entire material. Distribution Recovered only in residues of allochthonous limestone at top of Malongulli Formation; earliest Bolindian (Bol) age, equivalent to Katian. Phylum Mollusca Cuvier, 1797 Class Rostroconchia Pojeta, Runnegar, Morris and Newell, 1972 Order Conocardioidea Neumayr, 1891 Superfamily Eopterioidea Miller, 1889 Family Eopteriidae Miller, 1889 Eopteria Billings, 1865 Type species: Eopteria typica Billings, 1865 Diagnosis Eopteriid with a prominent anterior snout that lacks radial ribs (Pojeta et al. 1977, p. 26). Eopteria sp. Fig. 5 Material Figured specimen MMF 44970a, comprising a fragmentary pair of silicified conjoined valves. Several shell fragments definitely attributable to this specimen were also picked from the same limestone residue, as was the isolated posterior extremity MMF 44970b here figured (Fig. 5C). Although the latter cannot with certainty be assigned to the main specimen due to absence of intervening shell, there is a very high probability that it was broken from that specimen. Description Valves moderately biconvex, maximum length at hingeline; inflated medial third of valves extends from prominent umbo to ventral margin, and bears about a dozen rounded radial ribs spaced 2-3 per mm; raised ribs not present on snout, which instead bears shallow radial grooves becoming more widely spaced towards anterior extremity. Posterior third of valves also marked with shallow radial grooves. Faint closely spaced concentric growth lines are present on anterior and posterior flanks. Few internal features visible due to fragmentary preservation; however, a prominent Proc. Linn. Soc. N.S.W., 130, 2009 LG. PERCIVAL Figure 5. A— C Eopteria sp., conjoined specimen MMF 44970. A: left valve; B: right valve; C: fragment of posterior, viewed slightly obliquely. Scale bar represents one mm. Specimen from locality L37, alloch- thonous limestone at top of Malongulli Formation on flank of Malongulli Trig. internal ridge is present on the interior of the right valve, trending anteroventrally to intersect the valve margin. Internal shell surface smooth. Presence of pegma not verifiable. Dimensions The specimen is incomplete, with length of 6.7 mm, and height of 6.0 mm. The separate posterior extremity is 2.1 mm in length. Estimated maximum dimensions of the complete individual would be 6-7 mm in height, and at least 9-10 mm in length. Discussion This was the specimen from NSW referred to by Popov et al. (2003, p.177, pers. comm. by I.G. Percival), in discussion of the palaeogeographic setting of their new species Eopteria aiteneria from the Late Ordovician (Hirnantian) Angrensor Formation of north-eastern central Kazakhstan. That species was also described from a single damaged shell, though it is more complete than the NSW specimen which is of almost identical dimensions. However, the two are not conspecific, the most significant point of difference between them being the characteristic lunulate comarginal ornament developed on the anterior snout of E. aiteneria. The anterior snout of the NSW species instead bears several shallow radial grooves that possibly define a series of flattened wide ribs progressively decreasing in amplitude away from the inflated umbo. The medial strongly ribbed part of E. aiteneria is sharply bounded by carinae, particularly posteriorly, whereas the NSW species is more evenly rounded with a relatively gradual change in slope from the median region to the adjacent flanks. Proc. Linn. Soc. N.S.W., 130, 2009 The presence of an internal ridge in the right valve of E. aiteneria cannot be verified as the interior of this species 1s unknown. The only other known Late Ordovician species of Eopteria is E. conocardiformis Pojeta and Runnegar, 1976 from the Little Oak Formation of Alabama and the High Bridge Group of Kentucky, which is characterized by an elongation of the anterior snout. It also appears to be considerably more inflated than the species from NSW. Cope (2004) assigns an early Late Ordovician (Sandbian equivalent) age to this species. Thus Eopteria sp. from NSW, of late Katian age, 1S significant in partly bridging the gap between the Laurentian and Kazakhstan occurrences, where previously no species referable to this genus were known (Cope 2004, fig. 20). ; Although Eopteriasp.can bereadily distinguished from these Late Ordovician (and older) species, it would be unwise to establish a new species based on such fragmentary material, and so the specimen is left in open nomenclature. Distribution Recovered only in residues of allochthonous limestone at top of Malongulli Formation at locality L37; earliest Bolindian (Bol) age, equivalent to late Katian. Class Cephalopoda Cuvier, 1797 Subclass Nautiloidea Agassiz, 1847 Order Tarphycerida Flower, in Flower and Kummel, 1950 Family Plectoceratidae Hyatt, 1894 187 RARE FOSSILS FROM THE LATE ORDOVICIAN Genus Plectoceras Hyatt, 1894 Description Type species: Nautilus jason Billings, 1859 Conch exogastric, planispiral, tightly coiled with three whorls all in contact, gently expanding from 6-7 Plectoceras? sp. mm height in inner whorls to attain 16.5 mm in height Fig. 6 at body chamber which is approximately 21 mm in length; whorl expansion rate (WER) 1.78. Exterior Material with moderately coarse rounded ribs directed apicad Specimen MMF 44971, represented by a and forming a wide V-shape at midline; 3-4 ribs in composite cast, mostly decorticated; this specimen 10mm. Chambers are rounded subquadrate in cross was longitudinally sectioned for study. section, moderately inflated and gently impressed Figure 6. A— E Plectoceras? sp., MMF 44971. A— C: conch (prior to sectioning) in lateral view, and two whorl profiles (slightly rotated) showing ornament and camerae; D — E: longitudinal section through conch, D off-centre, and E sagittal, with internal features inked-in for clarity. Note in E, remains of sip- huncle (ventral in position) infilled with calcite immediately above body chamber. F. fragment of exterior of indeterminate tarphyceratid nautiloid, MMF 44972. Scale bar represents one cm. Both specimens from Gunningbland Formation at locality L51, “Currajong Park”, Gunningbland. 188 Proc. Linn. Soc. N.S.W., 130, 2009 I.G. PERCIVAL dorsally. Camerae behind body chamber gently flexed towards aperture and spaced up to 4.3 mm apart, narrowing to about 2 mm apart in inner whorls where they extend straight across venter. Siphuncle only partly visible as a calcite-filled tube with diameter of 2.4 mm, ventral and marginal in position; septal necks and connecting rings not preserved. Dimensions Maximum diameter of conch 60 mm; width of body chamber 20.0 mm. Discussion Due to its poor internal preservation, with septal necks absent and the siphuncle inadequately preserved, identification of this specimen to genus level remains tentative at this time pending the discovery of additional better-preserved material. The ventral position of the siphuncle and strongly ribbed coiled conch invites comparisons with tarphyceratids. Of Middle to Late Ordovician genera, the most similar to the Gunningbland specimen appears to be Plectoceras, whichis represented by numerous species in North America. Frey (1995) observed that these species fell into two major groups, one comprising forms that are generally smaller in diameter in which the whorls remain in contact, contrasting with the second group of generally larger conchs (including the type species) in which the final whorls became disjunct. Affinities of the Gunningbland species lie with the first of these species groups. The relatively low WER is also similar to that of Tarphyceras, but that predominantly Early Ordovician genus is typically nearly smooth externally (B. Kréger, pers. comm.). Two species of ribbed coiled nautiloids of latest Ordovician (Hirnantian) age from the Morkoka River region of the Siberian Platform were identified as Tarphyceras? by Balashov (1962). Illustrations of one of these, 7? morkokense Balashov, 1955, clearly show a ventral submarginal siphuncle. Dimensions of the type specimen (Balashov 1955, pl. XLII fig. 3a-b; refigured by Balashov 1962, pl. XLVI fig. 3) are very similar to those of Plectoceras? from Gunningbland. Stait et al. (1985, fig. 10) documented an incomplete external cast of a strongly ribbed coiled nautiloid from immediately overlying beds in the Gunningbland Formation, which, in the absence of any internal features, could only be referred to an indeterminate tarphyceratid. This specimen is very possibly congeneric with the one described here as Plectoceras? as it shares comparable dimensions and external features. An additional fragmentary exterior of a similar unidentified coiled nautiloid with coarse Proc. Linn. Soc. N.S.W., 130, 2009 ribbing is illustrated (Fig. 6F). Slightly younger tarphyceratids, again with prominent ribs, have been found in the Jingerangle Formation (of early Bolindian age, i.e. latest Katian) near Quandialla, about 95 km south of Gunningbland (Percival et al. 2006), but as all are preserved as moulds the position of the siphuncle and other internal features is unknown. Distribution Gunningbland Formation (upper part), “Curra- jong Park” property at Gunningbland; late Eastonian (Ea3-4), equivalent to early Katian. ACKNOWLEDGMENTS Without permission from landholders to collect specimens from their properties, none of these rare fossils would ever have come to light; for allowing access I thank the Dunhill family of “Boonderoo”, the McLarens of “Liscombe Pools”, and John and Julie Rhodes, former owners of “Sunnyside”. Technical assistance provided by Gary Dargan (NSW Department of Primary Industries) enabled preparation of the polished section of the tarphyceratid nautiloid from Gunningbland. I am grateful to Sue Lindsay (Australian Museum, Sydney) for facilitating SEM imaging of the microconulariids. David Barnes (NSW DPI) expertly prepared the photographic illustrations, and Cheryl Hormann (NSW DPI) drafted Figure 1. Heyo Van Iten, Bjorn Kroger and Leonid Popov provided very helpful advice on the identifications of the conulariids, nautiloid and rostroconch respectively. Reviews by two anonymous referees of the entire manuscript were most useful in fine-tuning it for publication. This paper is a contribution to IGCP Project No. 503: Ordovician Palaeogeography and Palaeoclimate. Published with the permission of the Director, Geological Survey of New South Wales, NSW Department of Primary Industries. REFERENCES Balashov, Z.G. (1955). Klass Cephalopoda. Otryad Nautiloidea [Class Cephalopoda. Order Nautiloidea]. In Nikiforova, O.I. (ed.) Polevoi Atlas, Ordovikski i siluriiski fauny Sibirskoi Platformy, 87-104. (VSEGEI: Moscow). Balashov, Z.G. (1962). Nautiloidei Ordovika Sibirskoi Platformy [Ordovician Nautiloids of the Siberian Platform]. /zdatelstvo, Leningradskogo Universiteta, 205 pp. Billings, E. (1859). Nautilus jason. Canadian Naturalist and Geologist 4, 464. Billings, E. (1865). ‘Palaeozoic Fossils: containing description and figures of new and little known species of organic remains from Silurian rocks. 189 RARE FOSSILS FROM THE LATE ORDOVICIAN Volume 1’. Geological Survey of Canada. 426 pp. (Dawson Brothers, Montreal). Bischoff, G.C.O. (1973). On the nature of the conodont animal. Geologica et Palaeontologica 7, 147-174. Bischoff, G.C.O. (1978). Internal structures of conulariid tests and their functional significance, with special reference to Circonulariina n. suborder (Cnidaria, Scyphozoa). Senckenbergiana lethaea 59, 275-327. Cope, J.C.W. (2004). Bivalve and rostroconch mollusks. In ‘The Great Ordovician Biodiversification Event’ (Eds B.D. Webby, F. Paris, M.L. Droser and I.G. Percival) pp. 196-208. (Columbia University Press, New York). Edgecombe, G.D. and Webby, B.D. (2006). The Ordovician encrinurid trilobite Sinocybele from New South Wales and its biogeographic significance. Memoirs of the Association of Australasian Palaeontologists 32, 413-422. Edgecombe, G.D. and Webby, B.D. (2007). Ordovician trilobites with eastern Gondwanan affinities from central-west New South Wales and Tasmania. Memoirs of the Association of Australasian Palaeontologists 34, 255-281. Foerste, A.F. (1928). American Arctic and related cephalopods. Denison University Bulletin, Journal of the Scientific Laboratories 23, 1-110, 29 pls. Frey, R.C. (1995). Middle and Upper Ordovician Nautiloid Cephalopods of the Cincinnati Arch region of Kentucky, Indiana and Ohio. U.S. Geological Survey Professional Paper 1066-P, 1-126, 22 pls. Furnish, W.M. and Glenister, B.F. (1964). Nautiloidea— Tarphycerida. In “Treatise on Invertebrate Paleontology, Part K. Mollusca 3’ (Eds C. Teichert, et al.) pp. K343-K368. (Geological Society of America: New York, and University of Kansas Press: Lawrence). Glenister, B.F. (1952). Ordovician nautiloids from New South Wales. Australian Journal of Science 15, 89- 91. Jerre, F. (1993). Conulariid microfossils from the Silurian Lower Visby Beds of Gotland, Sweden. Palaeontology 36, 403-424. Jerre, F. (1994). Anatomy and phylogenetic significance of Eoconularia loculata, a conulariid from the Silurian of Gotland. Lethaia 27, 97-109. Ladd, H.S. (1929). The stratigraphy and paleontology of the Maquoketa Shale of Iowa. Jowa Geological Survey Annual Report 34, 307-448. (not seen — fide Van Iten et al. 1996). Laurie, J.R. (1991). Articulate brachiopods from the Ordovician and Lower Silurian of Tasmania. In P.A. Jell (ed.) Australian Ordovician Brachiopod Studies. Memoirs of the Association of Australasian Palaeontologists 11, 1-106. Leme, J.M., Heredia, S., Rodrigues, S.C., SimGes, M.G., Acefiolaza, G.F. and Milana, J.P. (2003). Teresconularia gen. nov. from the Lower Ordovician of the Cordillera Oriental of Salta (NW Argentina): the oldest conulariid (Cnidaria) from South America. 190 Revista Espanola de Micropaleontologia 35, 265- DP. Leme, J.M., Simdes, M.G., Marques, A.C. and Van Iten, H. (2008). Cladistic analysis of the suborder Conulariina Miller and Gurley, 1896 (Cnidaria, Scyphozoa; Vendian — Triassic). Palaeontology 51, 649-662. Lindstrom, G. (1884). On the Silurian Gastropoda and Pteropoda of Gotland. Svenska Vetenskaps- Akademiens Handlingar 19, 1-250. (not seen — fide Jerre 1993). Parfrey, S.M. (1982). Palaeozoic conulariids from Tasmania. Alcheringa 6, 69-75. Percival, I.G. (1976). The geology of the Licking Hole Creek area, near Walli, central western New South Wales. Journal and Proceedings of the Royal Society of New South Wales 109, 7-23. Percival, I.G. (1978). Inarticulate brachiopods from the Late Ordovician of New South Wales, and their palaeoecological significance. Alcheringa 2, 117-141. Percival, I.G. (1979a). Ordovician plectambonitacean brachiopods from New South Wales. Alcheringa 3, 91-116. Percival, I.G. (1979b). Late Ordovician articulate brachiopods from Gunningbland, central western New South Wales. Proceedings of the Linnean Society of New South Wales 103, 175-187. Percival, I.G. (2005). Late Ordovician deepwater (BA 4) brachiopods of circum-Pacific terranes. In Abstracts for the Second International Symposium of IGCP503 on Ordovician Palaeogeography and Palaeoclimate, June 15-18, 2005. Insight (a Milwaukee Public Museum series in Natural History) 2, 24-25. Percival, I.G. (2009). Late Ordovician Strophomenide and Pentameride Brachiopods from central New South Wales. Proceedings of the Linnean Society of New South Wales 130, 157-178. Percival, I.G., Engelbretsen, M.J. and Brock, G.A. (1999). Distribution of Middle to Late Ordovician lingulate brachiopods in New South Wales. Acta Universitatis Carolinae, Geologica 43, 347-350. Percival, I.G. and Webby, B.D. (1996). Island benthic assemblages: with examples from the Late Ordovician of Eastern Australia. Historical Biology 11, 171-185. Percival, I.G., Zhen Y.-Y. and Pickett, J.W. (2006). Late Ordovician faunas from the Quandialla-Marsden district, south-central New South Wales. Proceedings of the Linnean Society of New South Wales 127, 235- DIS). Pickett, J.W. and Percival, I.G. (2001). Ordovician faunas and biostratigraphy in the Gunningbland area, central New South Wales. Alcheringa 25, 9-52. Pojeta, J. jr., Gilbert-Tomlinson, J. and Shergold, J.H. (1977). Cambrian and Ordovician rostroconch mollusks from Northern Australia. Bureau of Mineral Resources, Geology and Geophysics, Bulletin 171, 54 pp. + 27 pls. Proc. Linn. Soc. N.S.W., 130, 2009 I.G. PERCIVAL Pojeta, J. jr. and Runnegar, B. (1976). The paleontology Zhen Y.-Y., Webby B.D. and Barnes C.R. (1999). Upper of rostroconch mollusks and the early history of the Ordovician conodonts from the Bowan Park Phylum Mollusca. United States Geological Survey succession, central New South Wales, Australia. Professional Paper 968, 88 pp. + 54 pls. Géobios 32, 73-104. Popov, L.E., Cope, J.C.W. and Nikitin, I.F. (2003). A new Ordovician rostroconch mollusc from Kazakhstan. Alcheringa 27, 173-179. Rong J.-y., Zhan R.-b. and Han N.-r. (1994). The oldest known Eospirifer (Brachiopoda) in the Changwu Formation (Late Ordovician) of western Zhejiang, east China, with a review of the earliest spiriferoids. Journal of Paleontology 68, 763-776. Semeniuk, V. (1973). The stratigraphy of the Bowan Park Group, New South Wales. Journal and Proceedings of the Royal Society of New South Wales 105, 77-85. Sherwin, L. (1970). Silurian conularids from New South Wales. Records of the Geological Survey of New South Wales 11, 35-41. Sowerby, J. (1821). ‘The mineral conchology of Great Britam; or coloured figures and descriptions of those remains of testaceous animals or shells, which have been preserved at various times, and depths in the earth’. 194 pp. (W. Arding Co., London). Stait, B.A., Webby, B.D. and Percival, I.G. (1985). Late Ordovician nautiloids from central New South Wales, Australia. Alcheringa 9, 143-157. Van Iten, H., Fitzke, J.A. and Cox, R.S. (1996). Problematical fossil cnidarians from the Upper Ordovician of the north-central USA. Palaeontology 39, 1037-1064. Van Iten, H. and Vyhlasova, Z. (2004). Conulartids. In “The Great Ordovician Biodiversification Event’ (Eds B.D. Webby, F. Paris, M.L. Droser and I.G. Percival) pp. 119-123. (Columbia University Press, New York). Webby, B.D. (1992). Ordovician island biotas: New South Wales record and global implications. Journal and Proceedings of the Royal Society of New South Wales 125, 51-77. Webby, B.D. and Blom, W.M. (1986). The first well- preserved radiolarians from the Ordovician of Australia. Journal of Paleontology 60, 145-157. Webby, B.D. and Morris, D.G. (1976). New Ordovician stromatoporoids from New South Wales. Journal and Proceedings of the Royal Society of New South Wales 109, 125-135. Webby, B.D. and Packham, G.H. (1982). Stratigraphy and regional setting of the Cliefden Caves Limestone Group (Late Ordovician), central-western New South Wales. Journal of the Geological Society of Australia 29, 297-317. Webby, B.D. and Trotter, J. (1993). Ordovician sponge spicules from New South Wales, Australia. Journal of Paleontology 67, 28-41. Wiman, C. (1895). Palaeontologische Notizen 1—2. Bulletin of the Geological Institution of Upsala 3(2), 1-9. (not seen — fide Jerre 1994). Proc. Linn. Soc. N.S.W., 130, 2009 191 veocget st ROL err I ‘al tea {thee pean otha seeggnad suntan w », eed reas Ay pa, ae eee yi ~Y he in aeari ne ah i a nd thei The "nt m rence 4 tae (PLB PE Nar Seypharow). Seni Sugeber ged el alee any Sep WA fuk Un ‘ erie a OS 4a ih iy Leet ian Pi Aa ti , F fn wad ek Ny ¥ r 1 “f "ot i ; * i” rv ¥ ay i. we wey AH) hs ie as al “We ii ee ' ra . Lod U vt 4 " ot At pr ry 1 die Wi ™~ 4 Hier, Bald i Sounds Wieh 4 41 i B, CC bat ati * Vins erticy , “4 yt ve wticitry pal te ser oa i ee 1 : BAT a arp tay ae hg ¥ oh'va Sh ‘, it, ny i Ti Wl ee ia it 4:08 at | i a je am 1 ECW ate oe gbargaiat Hepp ce: r ane? Loker’ So et nv Tienes Hi i ted, on DP tay Ceri firey : nepal ‘4 fetralrs ay ihe Aveoehelon Of Abarnimem ‘ in * hI Pe eMe ; * ; ei S.. ochenpiben Cy Breed 1 Ki cartels | ET) aie rad, dP Ct f pil eire ally abl tit Pivevat fie |, SHOT i wdo vinings oH tic Com tAieog alia NW Arypesctiow): . trad yh, xvarierial wrist niet fromm, Sabet | Naericn 0, if 9 stein “ihe BAA pes an td ae » Mens Mihara, a wal’ Boos vedio Oey roy NY wversstrond den sun ts taaiapreiabenntl " yetaphigts sro dh Anoka : rio itine Heitey stT To Mere, thi sha it wey? aa ety: ER He segeabarnanaleeh set patel Ue y depamgsabtirde aah aren ye a vefipetine then") bush: oq ok-aalgea Aiba 193 Ct Feat ated wnt 1 st iN Ase, ah iestehneunos tap oo é wi Ne aero araanaiey?. a 6 aaah, 2 val: £80. APR TRY, Yeonwetteal ode heeiee ae wy tenn 1 iccavhaibevsiaxoahin gas aid ia. er art we ap a pray ep vita ytd ete pts 2 fatica MI, an bat a fe} 1 dda da lh sei el ria sealfelunl aaamaa shan} ee al Se Soatiy Wake PAL Rrmapeioarlalty, avy (RPO. KPa kero ah | peal rae Coie sie v4 “Yor He Semon Porcine eh Pieters aba} ane tao RE 7 (dn Waa yok a 290°0N Ride a aay wie a emai icra a BLL hace wort jam ively RTT heat yah, sautaghepuiciotatn Wil eh Hi, haan yop Loonie ago yy} es = tow ion ot AOROL nA, Wo redthk ; vali ‘iconanedon ant omelet CGAP) eb etgehayneycus) oa a cl sahuneesgcel niblaie 0 te. fi aiephhiliad ity yeaa ins. taal eope iy o ehh ahha wih UE manors | stance 3 Uiditin aera! ites Mast yey pnb in’ Otway moan, , VO Vek ee ano reign, ie sie eh in iG “or bana meet nny fot 4 sai Sivne A og Devonian Marine Invertebrate Fossils from the Port Macquarie Block, New South Wales JOHN PickeTT!, Davip OcH* AND EvaAN LEITCH? 1. Geological Survey of New South Wales, NSW, Department of Primary Industries, W.B. Clarke Geoscience Centre, 947-953 Londonderry Road, Londonderry, NSW 2753, Australia (picketj@bigpond.net.au); 2. Parsons Brinckerhoff, Level 27, Ernst & Young Centre, 680 George Street, Sydney, NSW 2000; GPO Box 5394 Sydney, NSW 2001, Australia (doch@pb.com.au); 3. Department of Environmental Sciences, University of Technology, Sydney, P.O. Box 123, Broadway, NSW 2007, Australia (Evan.Leitch@uts.edu.au). Pickett, J.W., Och, D.J., and Leitch, E.C. (2009). Devonian marine invertebrate fossils from the Port Macquarie Block, New South Wales. Proceedings of the Linnean Society of New South Wales 130, 193- 217. Two assemblages of rugose and tabulate corals, with accessory stromatoporoids and chaetetids, are described from the Touchwood and Mile Road Formations of the Wauchope — Port Macquarie district of northeastern New South Wales. Both assemblages are derived from allochthonous limestone clasts, except that the Mile Road fauna is accompanied at the same level by branching tabulate corals occurring in the matrix, indicating probable contemporaneity. The fauna from the Touchwood Formation indicates an Early Devonian (Emsian) age. Macrofossils from the Mile Road Formation indicate a broad Middle Devonian, probably Givetian age; conodonts accompanying the coral assemblage yield a precise age in the upper part of the early Givetian varcus Zone. Geographic affinities of the assemblages are typically eastern Australian, so that if terranes are represented in the block, these were not remote. Stratigraphic and structural relationships of the units are discussed. The name Mile Road Formation is formally defined. Manuscript received 27 November 2008, accepted for publication 16 February 2009. Key words: chaetetids, conodonts, Devonian, Emsian, Givetian, Mile Road Formation, Port Macquarie Block, Rugosa, stromatoporoids, Tabulata, Touchwood Formation. INTRODUCTION Immediately west of Port Macquarie, some 350 km north of Sydney, Palaeozoic units of the New England Fold Belt are exposed in a series of narrow belts delimited by NNE-striking faults (Fig.1) (Leitch, 1980; Roberts et al., 1995). Stratigraphic relationships between the units and their relative ages are not clear, so indications of age are particularly important in geological interpretations of the area. The ages of these rocks have been little constrained by published biostratigraphic data, with the only firm determinations those yielded by conodonts of Middle-Late Ordovician age from chert in the structurally dismembered Watonga Formation (Och et al., 2007), earlier attributed a Silurian or Devonian age on the basis of meagre conodont and radiolarian faunas (Ishiga et al., 1988). Unpublished reports by Pickett (1985, 1991) presented evidence for the Devonian age of limestone from two other units, the Touchwood Formation (Leitch, 1980) and the Mile Road Formation (Taylor, 1984, unpublished; Roberts et al., 1995). The present article is principally based on the re-examination of material described by Pickett, augmented by additional collecting. This has led to some refinement of the initial results, tectonically valuable biogeographic information, and a new stratigraphic interpretation. A formal description of the Mile Road Formation is included as Appendix 1. STRATIGRAPHIC UNITS The Mile Road Formation has only been recognized in the southern part of the wedge of rocks bounded by the Sancrox, Cowarra and Sapling Creek faults (Fig. 1) where it comprises interbedded fossiliferous siltstone and sandstone, containing blocks of coralline limestone and silicic tuff. The rocks form a sequence at least 1500 m thick, dipping steeply DEVONIAN MARINE INVERTEBRATE FOSSILS s dncd Lak¢ Innes (377 rp 485000 Legend Geological Boundaries Ue Clarence 4 Mapped Sample Localities ;Moreton - Sc \ leet Inferred iy Lithostratigraphic Units HRD cco Tablelands ! Recent Sediments Display Boulders at Dam ee’ aoe al eee Camden Haven Group New England] med Karikeree Metadolerite Fald Belt ~—4 ANY 8 ee iene A Thrumster Slate Roads — —— =Faits [| water ee N Boulders near Spittway MRF (Taylor's Locality) Soh Aushralia ve fat os Mingaletta Formation Watonga Formation A = Kilometers 6 Os 14 1 ee cas, Lesa. facia cer 7 — 1 _¥ Port Macquarie Serpentinite GDA 1994 - MGA Zone 56 Figure 1. Geological map of Wauchope — Port Macquarie area. mostly to the west, and younging in this direction, from northeast away from the Cowarra Fault to almost based on meagre data from near the intersection of north close to the fault. Sandstone is volcaniclastic Cowarra Access Road and the Mile Road (GR 478900 and of silicic, probably dacitic, provenance, with 6514400, Grants Head 1:25 000 sheet). Strike ranges abundant detrital plagioclase and vitric and felsitic 194 Proc. Linn. Soc. N.S.W., 130, 2009 J. PICKETT, D. OCH AND E. LEITCH lithic grains, and uncommon monocrystalline quartz grains. Many grains are angular and little abraded suggesting the rocks include abundant little modified ash. Finer grained rocks are of similar composition. Scattered coral, brachiopod and crinoid fossils are locally prominent in the clastic rocks some of which are extensively bioturbated (cf. Fig. 7C). Coralline limestone occurs as blocks embedded in fossiliferous sandstone and siltstone and locally (GR 478400 6514100) as weathered-out boulders up to about 1 m across that may originally have been derived from autochthonous lenses beyond the outcrop area. Silicic tuff is prominent in the lower part of the formation where it forms hard grey beds up to at least 0.25 m thick. It is of stmilar composition to the sandstones but distinguished by the presence of well-preserved shard structures in which the original glass has been replaced by fine-grained quartzofeldspthic aggregate. Euhedral plagioclase grains are widespread although broken angular grains are also common. The Zouchwood Formation is exposed between the Lake Innes and Innes Estate faults (Fig.1) from where it was described by Leitch (1980) as consisting of a sequence of siltstone, sandstone, paraconglomerate, basalt breccias and andesite at least 600 m thick. Much of the stratigraphically lower sedimentary part of the formation here is thin- bedded and consists predominantly of simply graded grey sandstone and darker horizontally laminated siltstone. Rare paraconglomerate beds, up to at least 20 m thick and the upper part of which are simply graded, contain clasts of basalt and andesite, slabs of bedded intraformational material and cobbles of coralline limestone. Further west, between the converging Sancrox and Cowarra faults north of the Mile Road Formation, Taylor (1984) mapped thin bedded siltstone, some radiolarian-bearing, graded and massive sandstone, chert and andesitic breccia as Touchwood Formation. He considered these rocks were faulted against the Mile Road Formation, an interpretation followed by Roberts et al. (1995). The contact between the two units is unexposed and occurs in a region of very little outcrop. Although it may be a fault, on the basis of structural and younging indications in both units, we favour interpretation as a stratigraphic contact, with Touchwood overlying the Mile Road (but see below under Discussion). Like those of the Mile Road Formation, Touchwood sandstones are volcaniclastic but differ in being of more mafic provenance. Abundant detrital components are lathwork and microlitic lithic grains and plagioclase; felsitic and vitric grains and quartz are uncommon, and calcic clinopyroxene is widespread but mostly only in small amounts. Proc. Linn. Soc. N.S.W., 130, 2009 FOSSIL LOCALITIES The fossils described in this article come from three localities. The first (HRD) lies within the Touchwood Formation in its type section (Leitch, 1980), the material coming from a disused quarry on the eastern side of Aston Street, north of the Hibbard — Port Macquarie road (Hastings River Drive) at GR 490000 6522560 (m), Port Macquarie 1:25,000 sheet (9435-2S). The material was originally collected by Erwin Scheibner, and is supplemented by samples taken by the present authors; Leitch’s (1980, p. 278) first mention of fossils is restricted to reporting rugose and tabulate corals. The second (MRF) is within the informally named “Mile Road Formation” of Taylor (1984) in a creek-bed in wooded country west of Forest Road at GR 478400 6514100 (m), Grants Head 1:25,000 sheet. The material was originally collected by Michael Taylor, and his formation name is formalised herein. Locality MRF could not be re-located using the information supplied by Taylor, but a general search led to a third locality (SC) in the bed of Sarahs Creek south of the ford on an unnamed forestry track at GR 478500 6514100 on the Grants Head 1:250,000 sheet (9434-1N). Here the mudstones of the Mile Road Formation dip 65° to 145°, and contain abundant fragments of the branching tabulate coral Thamnopora over a stratigraphic interval of possibly 30 m; near the middle of this interval there are also larger blocks of limestone made up of large colonies of massive favositid and heliolitid corals, the largest with maximum dimensions of c. 700 x 450 mm. The broken fragments of Thamnopora which occur in the matrix indicate that the larger blocks were derived penecontemporaneously. During the construction of Cowarra Damanumber of large blocks of allochthonous limestone were uncovered, the largest of which are now on display at the picnic area near the dam wall. The assemblages in these blocks indicate that their source is the same as that of the blocks originally collected by Taylor, but the assemblages they contain are much richer. In addition to the small assemblage originally reported by Pickett (1985) and supplemented herein, the blocks include large colonies of a large species of Spongophyllum, Syringopora sp., Heliolites sp., a cystiphyllid, a large solitary rugosan and Sguameofavosites sp., as well as brachiopods. Because of the display situation, none of this material could be collected. The display boulders were obtained from a locality now covered by the dam wall at GR 477650 6514250 (Grants Head sheet), and more material near the spillway at GR 477950 6514250. All these localities within the Mile Road Formation are roughly aligned ina WNW 195 DEVONIAN MARINE INVERTEBRATE FOSSILS — ESE direction, suggesting an episode of slumping of limy material during deposition of what is probably the older part of the formation. The environmental setting of all localities is similar, in that the fossils are allochthonous, being derived from clasts in slump deposits. At locality HRD the limestone clasts are small, the largest observed being about 35 cm in maximum dimension; some of the soft-sediment clasts in this deposit exceed a metre in maximum dimension. All the limestone and fossil clasts from the Mile Road Formation however are considerably larger, suggesting that their source lay much closer than in the case of the Touchwood Formation. AGES OF THE OCCURRENCES The occurrences of coralliform taxa are listed in Table 1. Those forms only identified in the field (marked with an asterisk) are not used for age determination. Detailed discussion is supplied in the Systematics section, under “Remarks” for each of the relevant taxa. Touchwood Formation. Significant for the age of this unit are Xystriphyllum cf. mitchelli minus, known only from the mid-Emsian perbonus-gronbergi Zone, Acanthophyllum sp., whose congeners are restricted to the Emsian in eastern Australia, and Sterictophyllum sp., whose genus 1s typically Pragian. Phillipsastrea Table 1. Occurrences of coral taxa in the Mile Road and Touchwood Formations. HRD Ashton St Quarry Touchwood Fm MRF SC Cowarra Dam Sarahs Creek Mile Road Fm Mile Road Fm Chaetetes sp. X Coenostroma sp. * Endophyllum cf. columna Hill Acanthophyllum sp Xx Xystriphyllum cf. mitchelli minus Parker = Phillipsastrea cf. maculosa Hill | x Sterictophyllum sp. ¥ Favosites salebrosa Etheridge fil. Pachyfavosites sp. Squameofavosites squamuliferus Etheridge fil. Cladopora sp. X Thamnopora randsi Jell & Hill Alveolites sp. A Alveolites sp. B Heliolites daintreei group IV Jones & Hill = *Spongophyllum sp. *Syringopora sp. *? Squameofavosites sp. *Heliolites sp. *cystiphyllid *large solitary rugosan Xx x Xx Wstonbor Ne sind : Proc. Linn. Soc. N.S.W., 130, 2009 J. PICKETT, D. OCH AND E. LEITCH maculosa 1s known from Pragian and Emsian strata, and the Sguameofavosites squamuliferus group is typically Early Devonian, although it does range down into the uppermost Silurian. Some further slight support for an Emsian age is indicated by the occurrence of the stromatoporoid Coenostroma. In summary, the assemblage is taken to indicate a later Early Devonian age, with a high probability of its being Emsian. Mile Road Formation. The coral assemblages from this unit are less reliably indicative of age than those of the Touchwood Formation. The best indicator is probably Endophyllum cf. columna, which suggests a mid-Givetian age. Thamnopora randsi, on the other hand, is only known reliably from the mid-Emsian, whereas Favosites salebrosa is apparently more typical of Eifelian strata. A small amount of material, offcuts from the original collection of Michael Taylor, was digested in acetic acid (Geological Survey of NSW sample C880), and yielded material of conodont species which indicate a precise age: Polygnathus linguiformis klapperi Clausen et al., 1979, Polygnathus linguiformis weddigei Clausen et al., 1979, Polygnathus hemiansatus Bultynck, 1987 and Icriodus difficilis Ziegler et al., 1976. The area of overlap of the ranges of these species, as given by Bultynck (1987, fig. 9) lies in the upper part of the lower varcus Zone, of early Givetian age. These taxa are illustrated in Fig. 2. DISCUSSION In the Touchwood Formation the dated material all occurs as clasts, and in the Mile Road Formation at least some of the dated material occurs as blocks embedded in a clastic matrix, and none has been shown unequivocally to be autochthonous. Thus the dates provide a maximum age for the units. The presence of fossils in the matrix as well as in blocks in the Mile Road Formation suggests the blocks are penecontemporaneous and hence the Givetian age is taken as that ofat least part ofthe Mile Road Formation. For the Touchwood Formation the interpretation is more equivocal. The limestone here occurs only as clasts which are restricted to a single bed that is a debris flow or the product of a high density turbidity current. Fossils are absent from the surrounding rocks. There is no record of Devonian limestone clasts in any of the Carboniferous or Permian units in this region, and the rocks are of a more mafic provenance than any of the latter units but similar to those of the Frasnian Birdwood Formation of the Yarras district some 25 km further west (Roberts et al., 1995). This suggests the age of the formation lies within the Emsian - Frasnian range, and on the basis of its stratigraphically overlying the Mile Road Formation can be further restricted to Givetian - Frasnian. However, the rocks north of Cowarra Dam mapped as Touchwood Formation have so far yielded no fossils, and the age suggested by the assemblage from the type area is older (Emsian) than that of the possibly underlying Mile Road Formation (early Givetian). Thus either the assemblage from the Touchwood formation in its type area does not yield a true age for the formation or, as originally interpreted by Taylor (1984), the contact between Mile Road Formation and Touchwood Formation north of Cowarra Dam is faulted. It is noteworthy that the aspect of all fossil Figure 2. Conodonts from the Mile Road Formation, all x10 except A, x7.5, and E, x20. A—K, Pa elements of Polygnathus hemiansatus Bultynck, 1987, B and C are oral and oblique views of the same specimen, E is a juvenile. F, Pa element of Polygnathus linguiformis weddigei Clausen et al., 1979. G, Pa element of Polygnathus linguiformis klapperi Clausen et al., 1979. H, icriodiform element of Icriodus difficilis Ziegler et al., 1976. Proc. Linn. Soc. N.S.W., 130, 2009 7 DEVONIAN MARINE INVERTEBRATE FOSSILS assemblages is typically Australian. Several taxa are ascribed to Australian species (Endophyllum cf. columna, Xystriphyllum cf. mitchelli minus, Phillipsastrea cf. maculosa, Favosites salebrosa, Thamnopora randsi), and the Squameofavosites squamuliferus group is very common in Early Devonian assemblages throughout eastern Australia. The genus Sterictophyllum is not known outside Australia. Thus although the rocks have been displaced along with the rest of the Hastings Block (e.g. Cawood and Leitch, 1985) their original location was well within the Australian province, probably from a southern continuation of the Tamworth Belt (Roberts and Geeve, 1999). It is also worth noting that in the latter a change in sediment provenance from a region in which intermediate and silicic volcanism was widespread to one dominated by mafic volcanic occurred in the Middle Devonian (Cawood, 1983), a change similar to that which occurred between deposition of the Mile Road and Touchwood Formations. SYSTEMATIC PALAEONTOLOGY The material discussed below is held in the collections of the Geological Survey of NSW, indicated by the prefix MMF. Specimens prefixed AM or AMF are held in the Australian Museum, Sydney. Literature citations for authors of taxa above the family level are not cited in the references; they may be found in Hill (1981). Phylum PORIFERA Grant, 1836 Class 7?DEMOSPONGIAE Sollas, 1885 Order uncertain Family CHAETETIDAE Milne-Edwards & Haime, 1850 Genus Chaetetes Fischer von Waldheim in Eichwald, 1829 Type species Chaetetes cylindraceus Fischer von Waldheim, 1829. Remarks The taxonomy of the fossil group informally known as chaetetids has been in a state of flux since the recognition that certain Recent sponges have a chaetetid morphology, although their spicular morphology indicates that they are demosponges (e.g. Ceratoporella: Hartman & Goreau, 1972; Acanthochaetetes: van Soest, 1984). In the last overview of chaetetids as a taxonomic group (Hill, 1981) they were regarded as tabulate corals; in recent treatises on Porifera (Hooper & Van Soest, 2002; Finks et al., 2004) they have been largely ignored, at least in terms of updated taxonomy: of the twenty- nine available generic names given by Hill (1981) in her review of the “Order” Chaetetida, only two are mentioned in Finks et al. (2004) and three in Hooper & Van Soest (2002). On the other hand, modern genera of “sclerosponges” with chaetetid morphology receive more exhaustive treatment. It is clear from Hill’s (1981) introductory remarks that she regarded the group as polyphyletic, but she has also rendered the service of bringing together those names relating to a particular group of morphologies. In the past, the vertical tubes of chaetetids have usually been referred to as corallites. In view of their highly probably sponge nature this seems inappropriate, so they are here referred to as calicles, the term favoured for similar features in Ceratoporella (e.g. Hartman & Goreau, 1972). Chaetetes sp. Figure 3 D-G Material Two specimens, MMF 32039 and 32040, with six thin sections. Locality HRD, probably Emsian. Description The species forms small, compact masses reaching at least 5 cm in diameter and 3 cm in height. The shape appears to have been more or less hemispherical, but some thin sections show a surface which bears low mamelons about 7 mm in diameter, Figure 3 (RIGHT). Spongiomorphs. A-C, topotype specimen of Litophyllum konincki (Etheridge & Foord, 1894), MMF884, Reid River Limestone, Reid Gap, S of Townsville, Queensland. A, B, transverse and longitudinal sections, x6; C, detail of B showing vertical trabeculae, x20 approx. D-G, Chaetetes sp., Touchwood Formation, locality HRD. D, transverse section, MMF32039b, x 3; E, longitudinal sec- tion of specimen with irregular surface, MMF32040a, x3; F, transverse section, MMF32039a, x10; G, longitudinal section, MMF32039c, x10. H-K, Coenostroma sp., Touchwood Formation, locality HRD. H, K, tangential and longitudinal sections, MMF44850, x4.5; J, detail of K showing microstructure of micropillars, x15. 198 Proc. Linn. Soc. N.S.W., 130, 2009 J. PICKETT, D. OCH AND E. LEITCH whereas others have a smooth surface. The skeleton is for the most part recrystallised, but some areas reveal it to have been composed of fine, near-vertical monacanthine trabeculae about Proc. Linn. Soc. N.S.W., 130, 2009 0.05 mm in diameter (Fig. 3G). The trabeculae are united to form walls defining subrounded calicles about 0.2 mm in internal diameter; wall thickness at the mid-point is 0.1 — 0.15 mm. In some areas the 199 DEVONIAN MARINE INVERTEBRATE FOSSILS calicles are interconnected uniserially, but in a rather meandering pattern (Fig. 3F); in longitudinal section these appear as pores in the walls. The calicles are traversed by fine, rather sagging tabulae c. 0.02 mm thick, generally separated by a distance greater than the width of the calicle, though this is not always the case; they number about eleven in 5 mm. The tabulae display a marked tendency to occur at similar levels in adjacent calicles, and may even be continuous through the mural pores. Remarks Chaetetids have been reported from Australia in a number of publications. Etheridge & Foord (1884) described Amplexopora konincki from Reid Gap, south of Townsville, north Queensland (Reid River Limestone, Emsian); Etheridge (1899) reported the species from Tamworth in NSW and erected for it the genus Litophyllum. Etheridge’s specimen of L. konincki from Tamworth (Australian Museum specimens AM3940, 3941, Moore Creek, near Tamworth, presumably from the Eifelian Moore Creek Limestone) is too recrystallised to show details of wall structure; it is impossible to recognise whether or not there were trabeculae. It does show rare connections between calicles. A topotype specimen (MMF884), rather recrystallised, shows a microstructure of vertical trabeculae similar to those of the Port Macquarie material and of other species of chaetetids (Fig. 3C). However, the tabulae are crowded (23 in 5 mm) and the spaces between them always less than the calicle diameter. Connections between the calicles are rare. I can see no reason for separating Litophyllum from Chaetetes itself. Chapman (1918) described Ch. stelliformis from Early Devonian Loomberah Limestone of the Tamworth area and, in 1920, Ch. spinuliferus from an Early Carboniferous Limestone in the Parish of Mooroowarra, i.e. near Somerton, NSW. Most of the material reported with this locality information derives from the hill known as Watts, Babbinboon (Visean; cf. Campbell, 1957; Pickett, 1967; Moore and Roberts, 1976), but, in spite of intensive collecting, the species has not been found there again. The type specimen in the Museum of Victoria (P73813, with a longitudinal section; the transverse section is apparently lost) is clearly a favositid of the squamuliferus group, revised by Philip (1960), though not included in his revision; the age of the specimen is therefore most probably Early Devonian, and the locality data given by Chapman erroneous, since there are no outcrops of Early Devonian rocks within the Parish of Mooroowarra. The species stelliformis 200 is now considered a tabulate coral, Squameofavosites (Hill, 1950; Philip 1960). Pohler (1998) reported Pachytheca cf. abdita Yanet, 1972 (in Breyvel’ et al., 1972) from a stromatoporoid bioherm in the Moore Creek Limestone Member of the Yarrimie Formation (Eifelian), but the material was not illustrated or described. It may be that this is the same form as Etheridge’s (1899) Litophyllum konincki. Material of Chaetetes (MMF44896-7) from the Uglovka Formation in Uglovka quarry, Russia (upper Serpukhovian) is interesting in that one of the specimens grew with a smooth surface, while the other bore abundant mamelons, just as in the Port Macquarie material, suggesting that this apparent dimorphism was a regular feature of chaetetids. Hill (1981) also included desmidoporids and lichenariids in the order Chaetetida, and the genera Desmidopora and Lichenaria have both been reported from Australia (Etheridge, 1902; Fitzgerald, 1955; Hill, 1955, 1957). These occurrences are either Ordovician or Silurian; they differ considerably from the present material, and their taxonomic status is not discussed here. Class STROMATOPOROIDEA Nicholson and Murie, 1878 Order SYRINGOSTROMATIDA Bogoyavlenskaya, 1969 Family COENOSTROMATIDAE Waagen and Wentzel, 1887 Genus Coenostroma Winchell, 1867 Type species Stromatopora monticulifera Winchell, 1866. Coenostroma sp. Figure 3 H-K Material MMF44860 from locality HRD. Description Specimen fragmentary, but in excess of 32 mm wide and 9 mm high. Surface apparently smooth and undulose. Coenostromes dominant, varying widely in thickness from 0.05 to 0.25 mm 8 — 11 in 2 mm, separating galleries 0.07 — 0.13 mm high, and which are subrounded to rather wider than high, consistently on the same level, generally discrete in longitudinal section, but occasionally joined laterally over six or more adjacent galleries. The transverse section shows a single walled tube 0.6 mm in diameter which may be a syringoporellid corallite. Coenosteles strongly Proc. Linn. Soc. N.S.W., 130, 2009 J. PICKETT, D. OCH AND E. LEITCH superimposed, up to 20 observed in a vertical series, appearing rather meandrine in tangential section. Microstructure reticulate, of clearly defined micropillars which are normal to the surface, appearing as dark spots in tangential section. Remarks Coenostroma species do not form a conspicuous element of eastern Australian Early and Middle Devonian faunas, as far as they are known (e.g. Webby et al., 1993; Webby & Zhen, 1993, 1997), the only published report being Coenostroma sp. from the Early Emsian (dehiscens Zone) Buchan Caves Limestone and Heath’s Quarry, Buchan, Victoria (Webby et al., 1993). The present material differs from this in its more crowded coenostromes, coenosteles which are more strongly superposed, and apparently also in the prominent micropillars of the coenosteles. Phylum COELENTERATA Frey and Leuckart, 1847 Class ANTHOZOA Ehrenberg, 1834 Subclass RUGOSA Milne Edwards and Haime, 1850 Order STAURIIDA Verrill, 1865 Family ENDOPHYLLIDAE Torley, 1933 Genus Endophyllum Milne-Edwards & Haime, 1851 Type species Endophyllum bowerbanki Milne-Edwards & Haime 1951. Endophyllum cf. columna Hill, 1942a Figures 4 A-B, 5A Material MME2921 2a, 29213a, with two thin sections. Locality MRF. Description Corallum cerioid, exceeding 10 cm in diameter. Epitheca 0.4 — 0.5 mm thick, showing strong median dark line. Maximum corallite diameters are 7 — 10 mm. Septa 18 — 22 in each order, the major septa extending well into the tabularium and sometimes almost reaching the axis. Minor septa also enter the tabularium but inside the presepiments are only about halfas long as the major septa. Even in young corallites both orders are interrupted peripherally by up to four rows of steep to almost horizontal presepiments, some of the inner ones bearing septal crests corresponding to both orders of septa. Tabulartum 4.5 — 6.0 mm wide, with tabulae which are flat or slightly concave near the axis, but turned strongly down and then back up again in the outer tabularium; 9 or 10 tabulae in 5 mm. Proc. Linn. Soc. N.S.W., 130, 2009 Remarks The Queensland species Endophyllum columna Hill most nearly approaches the present material in corallite dimensions, though it is generally slightly larger, in both corallite diameter (10 — 22 mm) and tabularium diameter (6 — 9 mm), and the wall thickness is rather less (0.05 — 0.15 mm). Of the other Australian species of Endophyllum still referred to that genus, EF. je//li Zhen, 1994 has a much wider tabularium (10 mm), E. giganteum Zhen & Jell, 1996 has much larger corallites (24—40 mm), and E. banksi Jell & Hill, 1970a has much larger corallites and more than twice as many septa. Endophyllum columna occurs in the upper part of the Burdekin Formation and the lower beds of the Cultivation Gully Formation, and is ascribed a mid- Givetian age by Zhen and Jell (1996). Family PTENOPHYLLIDAE Wedekind, 1923 Genus Acanthophyllum Dybowski, 1873 Type species Cyathophyllum heterophyllum Milne-Edwards & Haime, 1851. Acanthophyllum sp. Figure 4C Material MMEF32041, locality HRD. Description The single specimen is a somewhat oblique thin section of an eroded corallite near 10 mm in diameter. In spite of the obliquity of the section the tabularium appears to be oval rather than round. There are an estimated 28 major septa; both orders of septa are thickened in the dissepimentarium, being thickest in its central part. Near the epitheca they are quite thin. Minor septa only just reach the tabularium. Septa are smooth and strongly trabeculate and in their thickest parts they show a clear zone of trabecular divergence. Major septa extend almost to the axis; they are straight in the dissepimentartum but become wavy in the tabularium. There is a degree of bilaterality of septa coinciding with the long axis of the section, and at its margin, on this axis, lies a very short septum, possibly the counter septum, situated between a major septum on one side and a minor septum on the other. The dissepimentarium accounts for about half the radius of the corallite, the estimated diameter of the tabularium being 6 mm. 201 DEVONIAN MARINE INVERTEBRATE FOSSILS Remarks as the distinguishing character between the subgenera Most acanthophyllids described from Australia Acanthophyllum and Neostringophyllum, although have a calyx which is either bell-shaped or inverted _ this differentiation has not always been supported conical. Strusz (1966) took these two calical shapes 202 Proc. Linn. Soc. N.S.W., 130, 2009 {PF PICKEMIS DOCH ANDIE. IE EIRCH (e.g. Hill, 1981). The only material showing pronounced fusiform dilatation of the septa in the dissepimentarium has been referred to the related species Acanthophyllum clermontense (Etheridge, 1911) and A. kennediense Yu & Jell 1990, both of which are much larger than the present specimen, which is not necessarily a fully grown individual. If the smaller size is a reliable indication, it comes closest to the material from the Garra Formation referred to A. aff. clermontense by Strusz (1966). The Queensland reports of A. clermontense are from the Emsian (perbonus to inversus Zones; Mawson & Talent, 2003) Douglas Creek Limestone and the late Emsian Mount Podge Limestone (Zhen, 1995); A. kennediense is from the older, Lochkovian to Pragian Shield Creek Formation (Yu & Jell, 1990). Most of Struzs’s material from the Garra Formation comes from the upper levels, so the age is probably late Emsian (Mawson & Talent, 2000). Genus Xystriphyllum Hill, 1939 Type species Cyathophyllum dunstani Etheridge, 1911. Xystriphyllum cf. mitchelli minus Pedder, 1970a (in Pedder et al., 1970a) Figure 4D Material MMEF32042, MMF44865 from locality HRD. Description One specimen is a small piece of a cerioid colony which is too thin to permit preparation of a thin section, but the other has yielded a transverse section. The weathered surface shows about 20 corallites more or less in cross section. Corallites range in diameter from 4.2 mm to 5.8 mm and have 16 — 18 septa in each order. The major septa reach or almost reach the axis, but do not appear to interdigitate. Remarks In size and septal number the specimen falls within the ranges of the three smallest species of Xystriphyllum known from Australia. Xystriphyllum insigne Hill, 1940a, from Limestone Siding, Silverwood, Queensland, has diameters in the range 2 — 4 mm with 12 — 13 septa of each order; the corallites of X. mitchelli minus Pedder, 1970 (in Pedder et al., 1970a), from the Taemas Limestone, Wee Jasper, N.S.W. (mid-Emsian) are less than 6 mm in diameter, with no more than 20 septa of each order; and X. parvum Yu & Jell, 1990 has 12 — 15 septa and diameters of 4— 4.5 mm. Yu & Jell (1990) indicate a Lochkovian to Pragian age for X. parvum; Mawson & Talent (1989, fig. 2) suggest an age in the pesavis — sulcatus Zones, which is in direct agreement with that of Yu & Jell. If weight is given to the septal number in determining the species, then the present material comes closest to X. mitchelli minus. This form is known only from the Emsian Taemas Limestone, from a level within the perbonus-gronbergi Zone (Pedder et al., 1970a; Mawson & Talent, 2000). Family PHILLIPSASTREIDAE Hill, 1954 Genus Phillipsastrea d’Orbigny, 1849 Type species Astrea (Siderastrea) hennahi Lonsdale, 1840. Phillipastrea cf. maculosa Hill, 1942¢ Figures 4E, 5B Material MMF44866, a single fragment from locality HRD, from which only a longitudinal section could be prepared. Description The slide shows longitudinal sections of one tabularium of an astraeoid or thamnastraeoid coral, 5 mm in diameter and bounded on either side by strongly thickened, trabecular fans of a septal stereozone and its associated ring of horseshoe dissepiments. The fans are 1.5 — 2.0 mm wide. Septa are robust even in the outer dissepimentarium, and the dissepimentarial profile indicates that the everted calyces were raised Figure 4 (LEFT). Rugose corals from the Touchwood and Mile Road Formations. A, B, Endophyllum cf. columna Hill, 1942, Mile Road Formation, locality MRF, transverse and oblique longitudinal sections, MME29213a and 29212a respectively, x1.6. C, Acanthophyllum sp., Touchwood Formation, locality HRD, oblique section, MMF32041, x3.5. D, Xystriphyllum cf. mitchelli minus Pedder, 1970, Touchwood Formation, locality HRD, transverse section MMF44865, x4.3. E, Phillipsastrea cf. maculosa Hill, 1942, Touchwood Formation, locality HRD, longitudinal section, MMF44866, x3. F, G, Sterictophyllum sp., Touchwood Formation, locality HRD, transverse and longitudinal sections, MMF44861, x4.5. Proc. Linn. Soc. N.S.W., 130, 2009 203 DEVONIAN MARINE INVERTEBRATE FOSSILS Figure 5. Detail of material from Figure 4. A, Endophyllum cf. columna Hill, 1942, Mile Road Formation, locality MRF, showing details of septa and budding corallite (centre) MMF29213a, x 2.8. B, Phillipsast- rea cf. maculosa Hill, 1942, Touchwood Formation, locality HRD, MMF44866, showing long major septa and details of traabecular fans and horseshoe dissepiments, x 8.8. only a millimetre or so above the general level of the dissepimentarium. Trabeculae stout, 0.1 — 0.45 mm in diameter. Tabulae incomplete, the tabularial floor more or less flat or somewhat raised axially, the rather confused nature of the section suggesting that the major septa extend close to the axis. Remarks Tabularia are rather larger than those of the type material (“about 3 mm”), but in its general robustness the specimen is much closer to P maculosa than any other Australian species currently referred to the genus. The tabularia of Bensonastraea praetor Pedder, 1966 are similar in dimensions, but that genus has strongly vepreculate septa, of which the present material gives no indication; the septa of B. praetor are also less robust than those of the Port Macquarie specimen. The tabularia of P. carinata Hill, 1942a are only 3 mm wide and, as the name implies, the septa are strongly carinate. P. oculoides Hill, 1942d, from the Garra Formation, has tabularia similar in width to those of the present specimen, but the septa of that species are so short that major and minor septa are of nearly the same length, and the tabulae are concave or nearly horizontal (see also Wright, 2008). 204 Phillipsastrea currani Etheridge, 1892, as redescribed by Pedder (in Pedder et al., 1970a), has tabularia up to 4 mm in diameter, short major septa and horseshoe dissepiments which are not continuously developed. Finally, the recently described P. scotti Wright, 2008, is also close to the present form, but the Port Macquarie material is too scant for confident attribution to either this species or P. maculosa. Phillipsastrea maculosa is known from its type locality in the Sulcor Limestone (Emsian, serotinus Zone, Mawson & Talent, 2000), from the Liptrap Formation at Waratah Bay (Hill, 1954; Emsian, perbonus-gronbergi Zone, Mawson & Talent, 2000), the Coopers Creek Limestone at Tyers in Victoria (Philip, 1962; Pragian, sulcatus to pireneae Zones, Mawson & Talent, 1994b) and the Late Emsian (serotinus Zone) Mount Podge Limestone Zhen, 1995). Phillipsastrea scotti is also of serotinus Zone age. Suborder CYATHOPHYLLINA Nicholson, 1889 Family CYATHOPHYLLIDAE Dana, 1846 Genus Sterictophyllum Pedder, 1965 Proc. Linn. Soc. N.S.W., 130, 2009 J. PICKETT, D. OCH AND E. LEITCH Type species Cyathophyllum cresswelli Chapman, 1925. Sterictophyllum sp. Figure 4 F-G Material A single specimen MMF44861 from locality HRD, with a transverse and a partial longitudinal section. Description Corallum solitary, apparently cylindrical, with a maximum diameter of 14.7 mm. Septa long, strongly radial, of two orders, forming a marginal stereozone about 3 mm wide, in which the trabeculae are clearly visible. Septa 25 in each order, the major septa reaching the axis, where they are carinate; minor septa long, entering the tabularium. Septa of both orders taper abruptly after leaving the stereozone. The imperfect longitudinal section shows no details of the tabulae, but shows the numerous, steeply inclined dissepiments inside the stereozone, and the carinate septa near the axis. The tabulartum is 5.5 mm wide. Within the stereozone sections of laterally-growing trabeculae appear as dark spots; in the inner dissepimentarium they are only slightly inclined towards the axis. Remarks The present specimen is smaller than the maximum diameters quoted for any of the Australian species referred to Sterictophyllum, although a single specimen cannot give any impression of the range of variation. The stereozone is thicker than in the other species (S. creswelli (Chapman, 1925) — 2 mm; S. vallatum Pedder, 1965 —2.5 mm; S. pridianum (Philip, 1962) — 1.5 — 2.5 mm); in S. vallatum, however, the major septa do not reach the axis. On the basis of its relative dimensions, the present form appears to be closest to S. pridianum. (A fourth species, Mictophyllum trochoides Hill, 1940b, type species of Cavanophyllum Pedder, 1964, has been included in the genus by Jell & Hill, 1969, but has major septa which are somewhat contorted at the axis, lacks the pronounced stereozone, and is much larger than all the others. It is not further considered here). All these species are Early Devonian in age. The type species (sensu stricto) is known only from its type locality in the Lilydale Limestone at Lilydale, Victoria (Pragian, kindlei — pireneae Zones; Mawson & Talent, 2000); both the other species come from the Limestone phase of the Coopers Creek Formation (Pragian, sulcatus to possibly dehiscens Zones; Mawson & Talent, 2000). Proc. Linn. Soc. N.S.W., 130, 2009 Subclass TABULATA Milne-Edwards & Haime, 1850 Family FAVOSITIDAE Dana, 1846 Genus Favosites Lamarck, 1816 Type species Favosites gothlandicus Milne-Edwards Haime, 1850. and Favosites salebrosa Etheridge, 1899 Figure 6 A-B Synonymy 1899 Favosites basaltica var. salebrosa Etheridge, p. 166, pl. 21 figs 3-5, pl 27, figs 1-2. 1937 Favosites salebrosa Etheridge; Jones, p. 95, pl. 14, figs 2-6. 1940 Favosites salebrosus Etheridge; Hill and Jones, p. 197. 2002 Favosites sp. aff. F: salebrosus Etheridge; Pohler, p. 19, figs SA-D Material MME 44857 from the Mile Road Formation, locality SC. Description The material is from fragments ofa large, massive colony, exceeding 70 x 45 cm in original dimensions. Corallites range in diameter from 0.65 to 0.83 mm, with a mean at 0.73. Wall thickness ranges from 0.05 to 0.14, mean 0.06. Mural pores are 0.2 — 0.3 mm in diameter and at least 0.6 mm apart. Septal spines are neither conspicuous nor frequent, projecting 0.2 mm from the wall. There are 14 — 18 complete tabulae in 10 mm. Remarks The material accords well with the sections of the lectotype (AMF 4288. sections AM 47A, B) from the Woolomol Limestone, in portion 38, parish of Woolomol, northwest of Tamworth, N.S.W., in which I have measured rare corallites with a diameter of as much as 0.9 mm. Jones (1937) reports the species from what is probably the Cavan Bluff Limestone at Taemas, without illustration or nomination of material; this equates to the middle part of the Cavan Formation of Pedder et al. (1970a), of early Emsian (dehiscens Zone) age. For the type locality, neither Hill (1942c), Brown (1942) nor any of the publications of the Macquarie University group (e.g. Mawson and Talent, 1994a; Pohler, 2001) provides information which helps age determination. However, this limestone outcrop, 205 DEVONIAN MARINE INVERTEBRATE FOSSILS a hoe f x hat ol Figure 6. Tabulate corals from the Touchwood and Mile Road Formations. A, B, Favosites salebrosa Etheridge, 1899, Mile Road Formation, locality SC, transverse and longitudinal sections, MMF44857, x5. C, ?Pachyfavosites sp., Touchwood Formation, locality HRD, predominantly longitudinal section, MMF44863, x4. D, E, Squameofavosites squamuliferus (Etheridge, 1899), transverse and longitudinal sections, MMF32037, x3.7. in adjacent portions, is the type locality for some of the sponges described by Pickett (1969), which are also characteristic of the lowermost beds of the Timor Limestone to the southeast. For this interval Pedder et al. (1970a) have determined an earliest Eifelian age, so it is probable that the so-called Woolomol Limestone is more or less coeval. On the other hand, Pohler (2001, p. 96) indicates that all favositids from the Tamworth district examined by her are Emsian in age, and later (Pohler, 2002) describes Favosites aff. FE. salebrosus from the Emsian Sulcor Limestone, but 206 her material forms cylindrical branches of at least 3 cm diameter, in contrast to the type material, which is massive, and certainly the present material, which forms large masses. Genus Pachyfavosites Sokolov, 1952 Type species Calamopora polymorpha vat. tuberosa Goldfuss, 1826. Proc. Linn. Soc. N.S.W., 130, 2009 J. PICKETT, D. OCH AND E. LEITCH ?Pachyfavosites sp. Figure 6C Material A single specimen, MMF 44863, from locality HRD, with two thin sections. Description Corallum massive, original form unknown. Surface possibly with raised areas. Corallites four to six sided, 1 — 1.3 mm in diameter. Walls immensely thickened, so that the lumen diameter is 0.2 — 0.7 mm, and composed of large bundles of calcite fibres. Mural pores prominent, about 0.2 mm in diameter. Tabulae complete, more or less horizontal, irregularly spaced, possibly reach as many as 14 in 5 mm. Remarks This species is much more thickened than the type, or any other species referred to the genus by the Russian school, such as P markovskyi (fide Sokolov, 1962), as the lumen may be all but occluded. In this respect it is similar to the mature stages of Riphaeolites Yanet in Sokolov, 1955 (tentatively included in the family Cleistoporidae by Hill, 1981), but the present material shows no indication of the early, less thickened favositid stage characteristic of that genus . Pachyfavosites has been reported from Australia by Pohler (2002), who illustrates both P. rariporosus Dubatolov, 1963 and P tumulosus Yanet, 1965 from the Emsian Sulcor Limestone Member of the Yarrimie Formation near Tamworth, NSW; neither of these species has the intense thickening of the Port Macquarie material. Riphaeolites is restricted to a single doubtful record from an unspecified Emsian limestone (Sulcor?) from the Tamworth area (Pohler, 1998), unaccompanied by either description or illustration. Genus Squameofavosites Chernyshev, 1941 Type species Favosites hemisphericus var. bohemica Pocta, 1902. Squameofavosites squamuliferus (Etheridge 1899) Figure 6 D-E Material MME32027, with three thin sections; locality HRD. Description The single specimen is a fragment of a cerioid Proc. Linn. Soc. N.S.W., 130, 2009 colony 30 x 15 mm in diameter and c. 30 mm high. Corallite diameter ranges from 1.0 to 1.25 mm, diameters in the lower range being more common. Wall thickness is variable, from 0.1 to as much as 0.27 mm. Squamulae, though present, are not obvious, the longest one observed being only 0.1 mm in length. There are 33 — 40 tabulae per cm. Mural pores have a diameter close to 0.2 mm, but the preservation is such that measurements are imprecise. The vertical distance between their centres is 0.7 — 0.9 mm. Remarks Forms which may be referred to the squamuliferus group in Australia make a fairly homogeneous assortment (cf. Philip, 1960). The range of variation described by Philip (1960, notably figs 2, 3) suggests continuous variation between most of these forms. All of the material described by Philip (1960), and the type material of the various taxa involved, derives from strata of Early Devonian age; in central western New South Wales the group ranges down into Late Silurian strata (Pickett & Ingpen, 1990; Pickett & McClatchie, 1991). Philip (1960) referred the taxa in this group either to Squameofavosites grandiporus (Etheridge, 1890) or to “formae” within Squameofavosites squamuliferus (Etheridge, 1899), these latter comprising some eight subspecific units designated by the first eight Greek letters, and for the first five of which names in the species category are available (bryani Jones, 1937; nitidus Chapman, 1914; stelliformis Chapman, 1918; australis Chapman, 1907; ovatiporus Hill & Jones, 1940). The present specimen fits within the range reported for forma bryani (Jones, 1937), the type locality of which is in the Taemas Limestone at Good Hope, NSW, and which Philip reports from the “Tyers River Limestone”; both these localities are Emsian, though the lack of precise localities makes it difficult to assess the age more accurately. Family PACHYPORIDAE Gerth, 1921 Genus Cladopora Hall, 1851 Type species Cladopora seriata Hall, 1851. Remarks Following a revision of the type species by Oliver (1963), Hill (1981) restricts the genus to those species whose calices are lozenge-shaped in their mature portion. This definition excludes most of the Australian species previously included in the genus. Since the restricted material of this study does not 207 DEVONIAN MARINE INVERTEBRATE FOSSILS allow the observations necessary for more rigorous treatment, forms with a consistently rounded calyx are also treated here. Cladopora sp. Figure 7 A-B Material MMF 23038, a single fragment of a branching colony, with one oblique thin section; locality HRD. Description The colony is cerioid and branching, the branch 4 mm in diameter. Corallites very small, many in Figure 7. Tabulate corals from the Touchwood and Mile Road Formations. A, B, Cladopora sp., Touch- wood Formation, locality HRD, oblique sections of branches. A, MMF32028d, x7.5; B, MMF32038b, x7.8. C, D, Thamnopora randsi Jell & Hill, 1970. Mile Road Formation, locality SC, random sections, MMF44858, x2.5. Note bioturbation burrows in C. E, F, Alveolites sp. A., Touchwood Formation, locality HRD. E, transverse section, MMF32038c, x4.8; F, longitudinal section MMF32038a, x5. G, H, Alveolites sp. B, Mile Road Formation, locality MRF, portions of large specimen MMF29213a with areas of trans- verse (G) and longitudinal (H) orientation, x5. 208 Proc. Linn. Soc. N.S.W., 130, 2009 J. PICKETT, D. OCH AND E. LEITCH the cross-section of a branch (at least 18 along a diameter), rounded in transverse section axially, but lozenge-shaped towards the margin, 0.35 — 0.4 mm in maximum diameter, vertical at the axis and curved gradually towards the surface, which they reach at an acute angle. Walls thick relative to size of corallites, without any obvious thickening towards the surface. Tabulae not observed; there is some indication that the calices were back-filled with lamellar calcite rather than by tabulae. Septal spines not observed. Remarks Among the Australian species referred to Cladopora, C. foliata (Jones,1941) is encrusting; C. gippslandica (Chapman, 1907), originally described as a bryozoan, and redescribed by Philip (1962), does not appear to have the lozenge-shaped calices of the present form; three species from Victoria (Talent, 1963) (lemaitreae, corrigia, surculus) are all described as having rounded apertures. I have examined the type specimen of Cladopora mirabilis (Etheridge, 1917) (AMF899; 4 thin sections) from the Reid River Limestone at Reid Gap, south of Townsville, Queensland. This species forms branches 2 —3 mm in diameter in which the axial corallites are subrounded, becoming lozenge-shaped only towards the periphery. In longitudinal section they curve gently towards the periphery, without geniculation. There are 6 — 7 corallites along a diameter. Mural pores are common, tabulae rare, and the wall displays a prominent median dark line. The present material differs markedly from C. mirabilis in the much smaller and more crowded corallites. Genus Thamnopora Steininger, 1831 Type species Thamnopora madreporacea Steininger, 1831 (= Alveolites cervicornis de Blainville, 1830). Thamnopora randsi Jell and Hill, 1970b Figure 7 C-D Material Two large blocks of mudstone containing abundantly branching coralla, MMF 44858 (2 thin sections), MMF 44869, Mile Road Formation, locality SC. Description Corallum branching, bifurcating, branches cylindrical, 6 — 14 mm in diameter, mature branches being generally in the upper range. In transverse section there are 8 — 10 corallites along the median Proc. Linn. Soc. N.S.W., 130, 2009 plane ofa mature branch. Diameter of mature corallites 1.2 — 1.5 mm, their combined walls about 0.2 mm in thickness near the axis, and 0.6 — 0.8 mm in the outer stereozone, which is 2.5 — 3 mm wide. In the inner parts of the branches the walls show a conspicuous median dark line, but this becomes much more diffuse in the outer stereozone, where the stereome may show a lamination parallel to the surface. Near the axis the corallites are parallel to the branch, but turn outwards without geniculation to reach the sides of the branch at about 45°. Mural pores have a diameter of 0.2 — 0.3 mm, and are rather funnel-shaped in the stereozone. They occur in a single series on the faces of the corallites. Calices are 3 — 4 mm deep. Septal spines are rare, < 0.1 mm in length, conical, not trabeculate, and occur both at the axis and in the stereozone. Latex replicas from natural moulds show no sign of septal ridges in the calyces. Tabulae are complete, lie closer together than the width of the corallite, usually 4 in 2 mm. Remarks The present material accords well with that described by Jell and Hill (1970b), though the branches are slightly thinner (14 mm as against 15 mm in the types), and the corallites open slightly more obliquely to the sides of the branches. A significant similarity is the way the median dark line becomes less obvious towards the exterior, and the presence of growth lamination in these areas. Thamnopora plumosa Jones, 1941 has much stouter branches with nearly twice as many corallites across the median plane, and the thickening is less conspicuous. Thamnopora foliata Jones, 1941 is laminate; 7: meridionalis (Nicholson and Etheridge, 1879) is more delicately branched; T. crummeri (Etheridge, 1899) is closer, but has fewer corallites across the median plane and the difference in the amount of thickening between the axial and outer zones is less pronounced (I have examined the sections of the holotype, AM 3981 and 4687). The Victorian species, T alterivalis (Chapman, 1914), T. angulata Hill, 1950 and 7 tumulosa Hill, 1950 all have significantly thinner branches, the largest reaching only 7 mm. Thamnopora randsi 1s known so far only from its type area, the Douglas Creek Limestone, of Clermont, Queensland (mid-Emsian, perbonus to inversus Zones; Mawson & Talent, 2003). Family ALVEOLITIDAE Duncan, 1872 Genus Alveolites Lamarck, 1801 209 DEVONIAN MARINE INVERTEBRATE FOSSILS Type species Alveolites suborbicularis Lamarck, 1801. Alveolites sp. A Figure 7 E-F Material Four indifferently preserved — specimens, MMEF32038, 44862, 44867 and 44868, with six thin sections, all from locality HRD. Description All the material is of small fragments, the largest being less than 2 cm in maximum diameter. Corallites usually crescentic, apparently reclined, up to 0.4 mm high and 0.8 — 1.0 mm wide. Squamulae, septal spines and mural pores not observed. Tabulae 0.25 — 0.4 mm apart. Alveolites sp. B Figure 7 G-H Material Three specimens, MMF 29213 — 29315, with three thin sections, all from locality MRF. Description Corallum moderately large, exceeding 10 cm in maximum dimension. Corallites generally crescentic, but occasionally polygonal, 0.3 — 0.5 mm high and 0.5 — 0.7 mm wide. Occasional short septal spines occur on the side of the corallite away from the curved surface. The wall thickness varies considerably within the corallum, some areas having corallites whose walls reach only 0.02 mm, but for the most part the walls are thickened, reaching 0.1 or even 0.2 mm in extreme cases. The non-thickened areas pass rather abruptly into thickened areas, and individual corallites may lie partly in each of the two. There are occasional mural pores. Tabulae are at right angles to the direction of growth of the corallite, thin even in the thickened parts of the corallum, complete, and 0.2 — 0.6 mm apart. Remarks For all that the Australian literature refers to some twenty species of A/veolites (Pickett, 1999), the genus is not well documented in this country, either morphologically or stratigraphically. Six species (caudatus Hill, 1954; intermixtus Lecompte, 1939; multiperforatus Salée, 1916 (in Lecompte, 1933); saleei Lecompte, 1933; swborbicularis Lamarck, 1801; tmida Hinde, 1890) are known from Late Devonian strata in Western Australia. A further two were established by de Koninck (1876; obscurus, rapa) and, as the type material was destroyed by fire and details of the type localities are vague, it is probably better that the names be allowed to languish; his other three reports are unillustrated and, based as they are on external features alone, should be regarded as dubious. Chapman’s (1921) species regu/aris and victoriae were regarded as species of Favosites by Philip (1960). The status of the remaining eight taxa is not necessarily sound. The holotype of the only Silurian species, A/veolites piriformalis Etheridge, 1921 from the Yass district, has never been traced, and its internal structure is inadequately known. The holotype of 4. queenslandensis Etheridge & Foord, 1884, from the Emsian Reid River Limestone at Reid Gap, south of Townsville, has not been traced and the species has not been redescribed since the original publication. Hill et al. (1967) illustrated, without description, forms referred to A. sp. ex gr. fecundus (Salée, 1916) and A. sp. nov. aff. /emniscus Smith, 1933, of which the first has a branching corallum and the second does not show the areas of thin- and thick-walled corallites of A. sp. B. from locality MRF. A/veolites stamineus Hill, 1950 from the Emsian Murrindal Limestone at Buchan, Victoria, is a distinctive, thinly encrusting form. Neither of the forms referred to A, suborbicularis Lamarck, 1801 or A. sp. nov. aff. A. hemisphericus (Chernyshev, 1937) by Brihl & Pohler (1999) shows areas of thin- and thick-walled corallites, apart from the thinner-walled basal layer of A. suborbicularis. Finally, the material referred to A. sp. aff. A. taenioformis Schliiter, 1899 by Philip (1962) forms encrusting layers no more than 4 mm in thickness. The material described here as A/veolites sp. A is too scant for proper identification, and that described as Alveolites sp. B does not appear to be the same as any Australian forms so far reported. Order HELIOLITIDA Frech, 1897 Family HELIOLITIDAE Lindstrém, 1876 Genus Heliolites Dana, 1846 Type species Astraea porosa Goldfuss, 1826. Heliolites daintreei Nicholson & Etheridge, 1879 group IV Jones & Hill, 1940 Figure 8 Material Two specimens, fragments of much_ larger Proc. Linn. Soc. N.S.W., 130, 2009 J. PICKETT, D. OCH AND E. LEITCH Figure 8. Tabulate corals from the Mile Road Formation, locality SC. A, B, Heliolites daintreei Nicholson & Etheridge, 1879, group IV Jones & Hill, 1940. A, transverse section MMF44856, x5.3; B, longitudinal section MMF44859, x5.7. colonies, MMF 44856, 44859, from locality Sarahs Creek. Description Corallum massive, large. Tabularia consistently 1.2 mm in diameter, but ranging up to 1.5 mm. No areola is developed, but the tabularia are surrounded by 19 —20 tubules of varying size; tubules throughout the coenenchyme range from 0.2 to 0.5 mm in diameter. Tabularia separated by 3 — 10 tubules. Septa 12, laminar, apparently without axial spines, reaching about halfway to the axis. There are thin horizontal zones in which the skeleton is slightly thicker; these are about 2 mm thick and 7 — 9 mm apart. The tabulae are 0.6 — 1.1 mm apart, and the diaphragms about 11 in 5 mm. Proc. Linn. Soc. N.S.W., 130, 2009 Remarks Since the review of Australian Silurian and Devonian heliolitids by Jones and Hill (1940) no other overview of the group has been attempted. There is a clear need for any proper study of the group to be based on extensive material, allowing population studies. Here we simply follow the work of Jones and Hill. Heliolites daintreei, as conceived by Jones and Hill (1940), is an enormously variable species ranging from the Late Silurian to the Early Devonian, even the four informal “groups” not demonstrating reliable stratigraphic range. The present material, as it appears to lack axial spines on the septa, does not fit comfortably in any of the taxa recognised by them. Dil DEVONIAN MARINE INVERTEBRATE FOSSILS ACKNOWLEDGMENTS We thank Mike Neville of the NSW Department of Commerce for information on the source of the limestone olistoliths on display at Cowarra Dam. David Barnes and Yong-yi Zhen helped with photography and plate assembly. JWP is particularly grateful to Ruth Mawson for her generous help with the conodont determinations. Michael Taylor originally discovered the limestone fossils in the Mile Road Formation and first recognized the Touchwood Formation west of the Cowarra Fault. The Birpai Aboriginal Land Council graciously granted permission to collect material on their land. REFERENCES Breyvel’, M.G., Bogoyavlensaya, O.V., Breyvel’, I.A., Khodalevich, A.N., Shurygina, M.V., & Yanet, F.E.M. (1972). Kishepolostnye 1 brakhiopody zhivetskikh otlozheniy vostochnogo sklona Urala. Nedra, Moscow, 264 p. Brown, I.A. (1942). The Tamworth Series (Lower and Middle Devonian) near Attunga, N.S.W. Journal and Proceedings of the Royal Society of New South Wales 75, 165-176. Briihl, D., & Pohler, S. (1999). Tabulate corals from the Moore Creek Limestone (Middle Devonian: Late Eifelian-Early Givetian) in the Tamworth Belt (New South Wales, Australia). Abhandlungen der Geologischen Bundesanstalt 54, 275-293. Bultynck, P. (1987). Pelagic and neritic conodont successions from the Givetian of pre-Sahara Morocco and the Ardennes. Bu/letin de I Institut royal des Sciences naturelles de Belgique, Sciences de la Terre 57, 149-181. Campbell, K.S.W. (1957). A Lower Carboniferous brachiopod — coral fauna from New South Wales. Journal of Paleontology 31(1), 34-98, pl. 11-17. Cawood, P. A. (1983). Modal composition and clinopyroxene geochemistry of lithic sandstones from the New England Fold Belt (east Australia): a Paleozoic fore-are terrain. Geological Society of America Bulletin 94, 1199-1214. Cawood, P. A., & Leitch, E. C. (1985). Accretion and dispersal tectonics of the southern New England Fold Belt, eastern Australia, in Howell, D. G., ed. “Tectonostratigraphic terranes of the circum-Pacific region’, Circum-Pacific Council for Energy and Mineral Resources Earth Science Series, Number 1, 481-492, Houston, Texas. Chapman, F. (1907). New Silurian fossils of eastern Victoria, Part 1. Records of the Geological Survey of Victoria 2, 67-80. Chapman, F. (1914). Newer Silurian fossils of eastern Victoria - Part III. Records of the Geological Survey of Victoria 3, 301-316, pl. 46-61. Chapman, F. (1918). Appendix 2. Note on a new species ZW of Chaetetes. In Benson, W.N., The geology and petrology of the Great Serpentine Belt of New South Wales. Part VII. The geology of the Loomberah district and a portion of the Goonoo Goonoo estate. Proceedings of the Linnean Society of New South Wales 43, 392-394, pl. 42. Chapman, F. (1920). Appendix. Lower Carboniferous limestone fossils from New South Wales. In Benson, W.N., Dun, W.S., & Browne, W.R., The geology and petrology of the Great Serpentine Belt of New South Wales. Part IX. The geology, palaeontology and petrography of the Currabubula district, with notes on adjacent regions. Section B. Palaeontology. Proceedings of the Linnean Society of New South Wales 45, 364-367, pl. 24. Chapman, F. (1921). New or little-known Victorian fossils in the National Museum. Part XXV. Some Silurian tabulate corals. Proceedings of the Royal Society of Victoria 33, 212-225, pl. 9-11. Chapman, F. (1925). New or little-known fossils in the National Museum. Part XX VIII. Some Silurian rugose corals. Proceedings of the Royal Society of Victoria 37, 104-118, pl. 12-15. Chernyshev, B.B. (1937). Siluriyskie i devonskie Tabulata Mongolii 1 Tuvy. Trudy Akademii nauka SSSR, Mongol skogo Komiteta, 30(6), 1-31, pl. 1-4. Chernyshev, B.B. (1941). Siluriyskie i nizhnedevonskie korally basseyna reki Tarei (yugo-zapadnoy Taymyr). Trudy vsesoyuznogo Arkticheskogo Instituta 158(5), 9-64. Clausen, C.-D., Leuteritz, K., & Ziegler, W. (1979). Biostratigraphie und Lithofazies am Siidrand der Elsper Mulde (hohes Mittel- und tiefstes Oberdevon; Sauerland, Rheinisches Schiefergebirge). Geologisches Jahrbuch AS51, 3-37, | pl. Dana, J.D. (1846). Structure and classification of zoophytes. “U.S. Exploring Expedition during the years 1838 — 1842 under the command of Charles Wilkes, U.S.A.’ 7, x + 740 p. Lea & Blanchard, Philadelphia. de Blainville, H.M.D. (1830). Zoophytes. In Dictionnaire des Sceinces naturelles 60. 1-546 (not seen). D’Orbigny, A. (1849). “Note sur des polypiers fossiles.’ 1-12. Victor Masson, Paris. Dubatoloy, V.N. (1963). “Pozdnesiluriyskie i devonskie tabulyaty, geliolitidy 1 khetetidy Kuznetzkogo basseyna.’ 193 p. Nauka, Moscow. Duncan, P.M. (1872). Third report on the British fossil corals. Report 41 Meeting British Association for the Advancement of Science, Edinburgh (1871), 116-137. Dybowski, W.N. (1873). Monographie der Zoantharia Sclerodermata Rugosa aus der Silurformation Estlands, Nord-Livlands und der Insel Gotland. Archiv fiir Naturkunde Liv-, Ehst- und Kurlands, ser: 1,5, 257-414, pl. 1-2. Eichwald, C.E. von (1829). “Zoologia specialis quam expositis animalibus tum vivis, tum fossilibus potissimum Rossiae in universum, et Poloniae in specie, in usum lectionum.’ Vol. 1, vi+ 314 p., 5 pl. J. Zawalski, Vilna. Proc. Linn. Soc. N.S.W., 130, 2009 J. PICKETT, D. OCH AND E. LEITCH Etheridge, R., Jr (1890). Descriptions of Upper Silurian fossils from the Lilydale limestone, Upper Yarra district, Victoria. Records of the Australian Museum 1, 60-67, pl. 8-9. Etheridge, R., Jr (1892). Descriptions of four Madreporaria Rugosa - species of the genera Phillipsastrea, Heliophyllum and Cyathophyllum - from the Palaeozoic rocks of New South Wales. Records of the Geological Survey of New South Wales 2, 165-174, pl. 11-12. Etheridge, R., Jr (1896). Description of a small collection of Tasmanian Silurian fossils presented to the Australian Museum by Mr A. Montgomery, M.A., Government Geologist, Tasmania. Report of the Secretary of Mines Tasmania, xli-xlviii, | pl. Reprinted 1897, Papers and Proceedings of the Royal Society of Tasmania 1896, 29-46, pl. 1. Etheridge, R., Jr (1899). On the corals of the Tamworth district, chiefly from the Moore Creek and Woolomol Limestones. Records of the Geological Survey of New South Wales 6, 151-182, pl. 16-38. Etheridge, R., Jr (1902). Additions to the Middle Devonian and Carboniferous corals in the Australian Museum. Records of the Australian Museum 4, 253-262, pl. 37-40. Etheridge, R., Jr (1911). The Lower Palaeozoic corals of Chillagoe and Clermont. Part 1. Publications of the Geological Survey of Queensland 231, 1-8, pl. A-D. Etheridge, R., Jr (1917). Descriptions of some Queensland Palaeozoic and Mesozoic fossils. 4, Vetofistula, a new form of Palaeozoic Polyzoa, allied to Rhabdomeson Young & Young, from Reid’s Gap, near Townsville. Publications of the Geological Survey of Queensland 260, 17-29, pl. 4. Etheridge, R., Jr (1921). Palaeontologia Novae Cambriae Meridionalis — Occasional descriptions of New South Wales fossils. No. 8. Records of the Geological Survey of New South Wales 10, 1-11, pl. 1-7. Etheridge, R., Jr & Foord, A.H. (1884). On two species of Alveolites and one of Amplexopora from the Devonian rocks of northern Queensland. Annals and Magazine of Natural History (5) 14, 175-179, pl. 6. Finks, R.M., Reid, R.E.H., & Rigby, J.K. (2004). “Treatise on Invertebrate Paleontology, ed. C. Teichert. Part E, Porifera (Revised).’ Volume 3. Geological Society of America and University of Kansas. Fitzgerald, J.K. (1955). A new tabulate coral from New South Wales. Journal of Paleontology 29, 1057-1059, 1 text-fig. Gerth, H. (1921). Die Anthozoen der Dyas von Timor. Paldontologie von Timor 9(16), 65-147, pl. 145-150. Goldfuss, G.A. (1826). ‘Petrefacta Germaniae.’ 1, 1-76. Armz & Co., Diisseldorf. Hall, J. (1851). New genera of fossil corals from the report by James Hall, on the palaeontology of New York. American Journal of Science, series 2, 11, 398-401. Hartman, W.D., & Goreau, T.F. (1972). Ceratoporella (Porifera: Sclerospongiae) and the chaetetid “corals”. Transactions of the Connecticut Academy of Arts and Sciences 44, 133-148. Proc. Linn. Soc. N.S.W., 130, 2009 Hill, D. (1939). The Middle Devonian rugose corals of Queensland, I. Douglas Creek and Drummond Creek, Clermont district. Proceedings of the Royal Society of Queensland 50, 55-65, pl. 4-5. Hill, D. (1940a). The Middle Devonian rugose corals of Queensland, II. The Silverwood — Lucky Valley area. Proceedings of the Royal Society of Queensland 51, 150-168, pl. 2-3. Hill, D. (1940b). The Lower Devonian rugose corals of the Murrumbidgee and Goodradigbee Rivers, N.S.W. Journal and Proceedings of the Royal Society of New South Wales 74, 247-276, pl. 9-11. Hill, D. (1942a). The Middle Devonian rugose corals of Queensland, III. Burdekin Downs, Fanning R., and Reid Gap, North Queensland. Proceedings of the Royal Society of Queensland 53, 229-268, pl. 5-11. Hill, D. (1942b). The Lower Devonian rugose corals from the Mt. Etna Limestone, Qld. Proceedings of the Royal Society of Queensland 54, 13-22, pl. 1. Hill, D. (1942c). The Devonian rugose corals of the Tamworth district, N.S.W. Journal and Proceedings of the Royal Society of New South Wales 76, 142-164, pl. 2-4. Hill, D. (1942d). Middle Palaeozoic rugose corals from the Wellington district, N.S.W. Journal and Proceedings of the Royal Society of New South Wales 76, 182-189, pl. 5-6. Hill, D. (1950). Middle Devonian corals from the Buchan district, Victoria. Proceedings of the Royal Society of Victoria 62, 137-164, pl. 5-9. Hill, D. (1954). Devonian corals from Waratah Bay, Victoria. Proceedings of the Royal Society of Victoria 66, 7-118, pl. 6-9. Hill, D. (1955). Ordovician corals from Ida Bay, Queenstown and Zeehan, Tasmania. Papers and Proceedings of the Royal Society of Tasmania 89, 237-254, pl. 1-3. Hill, D. (1957). Ordovician corals from New South Wales. Journal and Proceedings of the Royal Society of New South Wales 91, 97-107, pl. 2-4. Hill, D. (1981). ‘ Treatise on Invertebrate Paleontology, ed. C. Teichert. Part F, Coelenterata, Supplement 1, Rugosa and Tabulata.’ 2 vols. Geological Society of America and University of Kansas. Hill, D., & Jones, O.A. (1940). The corals of the Garra Beds, Molong district, New South Wales. Journal and Proceedings of the Royal Society of New South Wales 74, 175-208, pl. 2-8. Hill, D., Playford, G., & Woods, J.T. (1967). ° Devonian fossils of Queensland.’ Queensland Palaeontographical Society, Brisbane. 1-32, pl. 1-15. Hinde, G.J. (1890). Notes on the palaeontology of Western Australia, 2. Corals and Polyzoa. Geological Magazine (3) 7, 194-204, pl. 8, 8a. Hooper, J.N.A., & Van Soest, R.W.M., eds (2002). ‘Systema Porifera: a guide to the classification of sponges.’ 2 vols. Kluwer Academic/Plenum Publishers, New York. Ishiga, H., Leitch, E.C., Watanabe, T., Naka, T., & Iwasaki, M. (1988). Radiolarian and conodont 23 DEVONIAN MARINE INVERTEBRATE FOSSILS biostratigraphy of siliceous rocks from the New England Fold Belt. Australian Journal of Earth Sciences 35(1), 73-80. Jell, J.S., & Hill, D. (1969). Devonian corals from the Ukalunda district, north Queensland. Publications of the Geological Survey of Queensland 340 (Palaeontological Papers No. 16), 1-27, pl. 1-9. Jell, J.S., & Hill, D. (1970a). The Devonian coral fauna of the Point Hibbs Limestone, Tasmania. Papers and Proceedings of the Royal Society of Tasmania 104, 1-16, pl. 1-6. Jell, J.S., & Hill, D. (1970b). Revision of the coral fauna from the Devonian Douglas Creek Limestone, Clermont, central Queensland. Proceedings of the Royal Society of Queensland 81(10), 93-120, pl. 3-8. Jones, O.A. (1937). The Australian massive species of the coral genus Favosites. Records of the Australian Museum 20, 79-102, 11-16. Jones, O.A. (1941). The Devonian Tabulata of Douglas and Drummond Creeks, Clermont, Queensland. Proceedings of the Royal Society of Queensland 53(3), 41-60, pl. 1-3. Koninck, L.G. de (1876-77). “Recherches sur les fossils paléozoiques de la Nouvelle-Galles du Sud (Australie).’ 1-373, Atlas (pl. 1-4, 1876; pl. 5-25, 1877) Brussels. [Re-issued in 1877-1878 in Mémoires de la Société royale des Sciences de Liége (2) 6(2), 1-140, pl. 1-4 (1877); 7, 1-235, pl. 5-24 (1878). An English translation, Descriptions of the Palaeozoic fossils of New South Wales, differently paged but with the same plates, was issued in 1898 as Memoirs of the Geological Survey of New South Wales, Palaeontology 6, \-xiii + 1-298, pl. 1-24. Lamarck, J.B.P.A. de M. (1801). “Systeme des animaux sans vertebres.’ villi + 432 p. The author, Paris. Lamarck, J.B.P.A de M. (1816). ‘Histoire naturelle des animaux sans vertebres.’ 2, 1-568. The author, Paris. Lecompte, M. (1933). Le genre A/veolites Lamarck dans le Dévonien moyen et supérieur de |’ Ardenne. Mémoires du Muséum royale d'Histoire naturelle de Belgique 55, 1-50, pl. 1-4. Lecompte, M. (1939). Les tabulés du Dévonien moyen et supérieur du bord sud du Bassin de Dinant. Mémoires du Muséum royale d'Histoire naturelle de Belgique 90, 1-229, pl. 1-23. Leitch, E.C. (1980). Rock units, structure and metamorphism of the Port Macquarie Block, eastern New England Fold Belt. Proceedings of the Linnean Society of New South Wales 104(4), 273-292. Lindstrém, G. (1876). On the affinities of the Anthozoa Tabulata. Annals and Magazine of Natural History, ser. 4, 18, 1-17. Lonsdale, W. (1840). In Sedgwick, A., & Murchison, R.L., On the physical structure of Devonshire, and on the subdivisions and geological relations of its older stratified deposits. Transactions of the Geological Society of London, ser. 2, 5, p.697. Mawson, R., & Talent, J.A. (1989). Late Emsian — Givetian stratigraphy and conodont biofacies — carbonate slope and offshore shoal to sheltered 214 lagoon and nearshore carbonate ramp — Broken River, North Queensland, Australia. Courier Forschungs- Institut Senckenberg 117, 205-259. Mawson, R., & Talent, J.-A. (1994a). The Tamworth Group (mid-Devonian) at Attunga, New South Wales: conodont data and inferred ages. Courier Forschungs-Institut Senckenberg 168, 37-59. Mawson, R., & Talent, J.A. (1994b). Age of an Early Devonian carbonate fan and isolated limestone clasts and megaclasts, east-central Victoria. Proceedings of the Royal Society of Victoria 106, 31-70. Mawson, R., & Talent, J.A. (2000). The Devonian of eastern Australia: stratigraphic alignments, stage and series boundaries, and the transgresssion-regression pattern re-considered. Courier Forschungs-Institut Senckenberg 225, 243-270. Mawson, R., & Talent, J.A. (2003). Conodont faunas from sequences on or marginal to the Anakie Inlier (Central Queensland, Australia) in relation to Devonian transgressions. Bulletin of Geosciences 78(4), 335-358. Milne-Edwards, H., & Haime, J. (1850). A monograph of the British fossil corals, p. 1— Ixxxv, 1-71, pl. 1-11. Palaeontographical Society Monographs, London. Milne-Edwards, H., & Haime, J. (1851). Monographie des polypiers fossils des terrains paléozoiques. Archives du Muséum Nationale d'Histoire naturelle 5, 1-502, pl. 1-20. Moore, D., & Roberts, J. (1976). The Early Carboniferous marine transgression in the Merlewood Formation, Werrie Syncline, New South Wales. Journal and Proceedings of the Royal Society of New South Wales 109(1-2), 49-57. Nicholson, H.A., & Etheridge, R., Jr (1879). Descriptions of Palaeozoic corals from northern Queensland, with observations on the genus Stenopora. Annals and Magazine of Natural History (5), 4, 216-226, 265- 285, pl. 14. Och, D.J., Percival, I.G., & Leitch, E.C. (2007). Ordovician conodonts from the Watonga Formation, Port Macquarie, northeast New South Wales. Proceedings of the Linnean Society of New South Wales 128, 209-216. Oliver, W.A. (1963). Redescription of three species of corals from the Rockport Dolomite of New York. U.S.G.S. Professional Paper 414G, 1-9, pl. 1-5. Pedder, A.E.H. (1964). Two new genera of Devonian tetracorals from Australia. Proceedings of the Linnean Society of New South Wales 88, 364-367, pl. 19. Pedder, A.E.H. (1965). A revision of the Australian Devonian corals previously referred to Mictophyllum. Proceedings of the Royal Society of Victoria 78(2), 201-220, pl, 30-34. Pedder, A.E.H. (1966). The Devonian tetracoral Haplothecia and new Australian phacellophyllids. Proceedings of the Linnean Society of New South Wales 90, 181-189, pl. 6. Pedder, A.E.H., Jackson, J.H., & Philip, G.M. (1970a). Lower Devonian biostratigraphy in the Wee Jasper Proc. Linn. Soc. N.S.W., 130, 2009 J. PICKETT, D. OCH AND E. LEITCH region of New South Wales. Journal of Paleontology 44, 206-251, pl. 37-50. Pedder, A.E.H., Jackson, J.H., & Ellenor, D.W. (1970b). An interim account of the Middle Devonian Timor Limestone of north-eastern New South Wales. Proceedings of the Linnean Society of New South Wales 94(3), 242-272, pl. 14-24. Philip, G.M. (1960). The Middle Palaeozoic squamulate favositids of Victoria. Palaeontology 3, 186-207, pl. 30-40. Philip. G.M. (1962). The palaeontology and stratigraphy of the Siluro-Devonian sediments of the Tyers area, Gippsland, Victoria. Proceedings of the Royal Society of Victoria 75, 123-246, pl. 11-36. Pickett, J.W. (1967). Lower Carboniferous coral faunas from the New England District of New South Wales. Geological Survey of N.S.W., Memoir Palaeontology 15, 1-38, pl. 1-20. Pickett, J.W. (1969). Middle and Upper Palaeozoic sponges from New South Wales. Geological Survey of N.S.W., Memoir Palaeontology 16, 1-25, pl. 1-11. Pickett, J.W. (1985). A mid-Devonian limestone near Wauchope. Geological Survey of New South Wales - Unpublished Palaeontological Report 1985/4 (GS 1985/118). Pickett, J.W. (1991). Fossil corals from the Touchwood Formation near Port Macquarie. Geological Survey of New South Wales - Unpublished Palaeontological Report 1991/1 (GS 1991/105). Pickett, J.W. (1999). “OZCORALS: a Bibliography and index of fossil corals from Antarctica, Australia, New Guinea and New Zealand.’ Microsoft Access® database, downloadable from website of Association of Australasian Palaeontologists. Pickett, J.W., & Ingpen, I.A. (1990). Ordovician and Silurian strata south of Trundle, New South Wales. Quarterly Notes of the Geological Survey of New South Wales 78, 1-14. Pickett, J.W., & McClatchie, L. (1991). Age and relations of stratigraphic units in the Murda Syncline area. Quarterly Notes of the Geological Survey of New South Wales 85, 9-32. Poéta, F. (1902). Anthozoaires et Aleyonaires. In: Barrande, J., ed., “Systeme silurien du centre de la Boheme.’ Part 1, 8(2), 1-347. Prague, Paris. Pohler, S.M.L. (1998). Devonian carbonate buildup facies in an intra-oceanic island arc (Tamworth Belt, New South Wales, Australia). Facies 39, 1-34. Pohler, S.M.L. (2001). Paleoecology, biostratigraphy and paleogeography of Favositidae (Tabulata) from the Emsian to Middle Devonian Tamworth Group (New South Wales, Australia). Senckenbergiana lethaea 81(1), 91-109. Pohler, S.M.L. (2002). Favositidae (Tabulata) from Emsian to Middle Devonian limestones of the Tamworth Group (N.S.W., Australia). Paldontologische Zeitschrift 76(1), 1-19. Roberts, J. & Geeve, R. (1999). Allochthonous forearc blocks and their influence on an orogenic timetable for the Southern New England Orogen, in Flood, Proc. Linn. Soc. N.S.W., 130, 2009 P. G., ed. “Regional Geology Tectonics and Metallogenesis New England Orogen’, 105-114, University of New England, Armidale. Roberts, J. , Leitch, E.C. , Lennox, P.G. , & Offler, R. (1995). Devonian-Carboniferous stratigraphy of the southern Hastings Block, New England Orogen, eastern Australia. Australian Journal of Earth Sciences 42(6), 609-634 Schliiter, C. (1899). Die Anthozoen des rheinischen Mitteldevons. Abhandlungen zur geologischen Specialkarte von Preussen und den Thiiringischen Staaten 8(4), x + 207, pl. 1-16. Smith, S. (1933). Sur les espéces nouvelles d’ Alveolites de |’Eifelien inférieur du Nord de la France et de la Belgique. Annales de la Société géologique du Nord 57, 134-145. Sokolov, B.S. (1952). Tabulyaty paleozoya evropeyskoy chasti SSSR, chast’ 4, Devon Russkoy platformy 1 zapadnogo Urala. Trudy Vsesoyuznogo Neftyanogo Nauchno-issledovannogo geolog-razvedochnogo Instituta (VNIGRI), n.s. 62, 1-292. Sokolov, B.S. (1955). Tabulyaty paleozoya evropeyskoy chasti SSSR. Vvedenie: Obshchie voprozy k sistematiki i istorii razvitiya tabulyat. Trudy Vsesoyuznogo Neftyanogo Nauchno-issledovannogo geolog-razvedochnogo Instituta (VNIGRI), n.s. 85, 227 p., 90 pl. Sokoloy, B.S. (ed.) (1962). “Osnovy Paleontologu. Tom 2, Gubki, arkheotsiaty, kishechnopolostnye, chervi.’ 485 p. Izdatel’stvo Akademii Nauk SSSR, Moscow. Steininger, J. (1831). ‘Bemerkungen iiber die Versteinerungen, welche in dem Uebergangs- Kalkgebirge der Eifel gefunden warden.’ 1-44, Trier. Strusz, D.L. (1966). Spongophyllidae from the Devonian Garra Formation, New South Wales. Palaeontology 9, 544-598, pl. 85-96. Talent, J.A. (1963). The Devonian of the Wentworth and Mitchell Rivers. Geological Survey of Victoria Memoir 24, 1-118, pl. 1-78. 4 Taylor, M.A. (1984). ‘Geology of the Middle Devonian to Early Triassic of the Wauchope district, N.S.W.’ B.Sc Honours thesis, University of Sydney. 158 pp. Torley, K. (1933). Ueber Endophyllum bowerbanki M. Ed. U. H. Zeitschrift der deutschen geologischen Gesellschaft 85, 630-633. Van Soest, R.M.W. (1984). Deficient Merlia normani Kirkpatrick, 1908, from the Curagao reefs, with a discussion on the phylogenetic interpretation of sclerosponges. Bijdragen tot de Dierkunde 54(2), 211-219. Webby, B.D., Stearn, C.W., & Zhen, Y.-y. (1993). Lower Devonian (Pragian — Emsian) stromatoporoids from Victoria. Proceedings of the Royal Society of Victoria 105(2), 113-185. Webby, B.D., & Zhen, Y.-y. (1993). Lower Devonian stromatoporoids from the Jesse Limestone of the Limekilns area, New South Wales. Alcheringa 17, 327-352. Webby, B.D., & Zhen, Y.-y. (1997). Silurian and Devonian clathrodictyids and other stromatoporoids from the 215 DEVONIAN MARINE INVERTEBRATE FOSSILS Broken River region, north Queensland. A/cheringa 21, 1-56. Winchell, A.N. (1866). The Grand Traverse Region: A report on the Geological and Industrial Resources of the Counties Antrim, Trand Traverse, Benzie, and Leelanaw in the Lower Peninsula of Michigan. Dr Chase’s Steam Printing, Ann Arbor. 97 p. Winchell, A.N. (1867). Stromatoporidae: their structure and zoological affinites. Proceedings of the Americal Association for the Advancement of Science 15, 91 -99. Wright, A.J. (2008). Emsian (Early Devonian) tetracorals (Cnidaria) from Grattai Creek, New South Wales. Proceedings of the Linnean Society of New South Wales 128, 83-96. Yanet, F.E. (1965). Mikrostrukturnye osobennosti stenok eifel’skikh 1 zhivetskikh tabulyat 1 khetetid Urala. In Sokolov, B.S., & Dubatolov, V.N., eds. Tabulatomorfnye korally devona i karbona SSSR, Trudy I simposiuma po izucheniya iskopaemykh korallov 2, 12-24. Nauka, Moscow. Yu Chang-min & Jell, J.S. (1990). Early Devonian rugose coral fauna from the Shield Creek Formation, Broken River Embayment, North Queensland. Memoirs of the Association of Australasian Palaeontologists 10, 169-209. Zhen, Yong-yi (1994). Givetian rugose corals from the northern margin of the Burdekin Basin, north Queensland. A/cheringa 18, 301-343. Zhen, Yong-yi (1995). Late Emsian rugose corals of the Mount Podge area, Burdekin Basin, north Queensland. Alcheringa 19, 193-234. Zhen, Yong-yi, & Jell, J.S. (1996). Middle Devonian rugose corals from the Fanning River Group, north Queensland, Australia. Palaeontographica A242(1- 3), 15-98. Ziegler, W., Klapper, G., & Johnson, J.G. (1976). Redefinition and subdivision of the varcus-Zone (Conodonts, Midle - ?Upper Devonian) in Europe and North America. Geologica et Palaeontologica 10, 109-140, pl. 1-4. 216 Proc. Linn. Soc. N.S.W., 130, 2009 J. PICKETT, D. OCH AND E. LEITCH APPENDIX 1. Mile Road Formation Definition: The name Mile Road Formation is applied to interbedded fossiliferous siltstone and sandstone, containing blocks of coralline limestone and possibly autochthonous limestone lenses, and silicic tuff, mapped as stratigraphically underlying the Touchwood Formation in the eastern part of the Hastings Block. Synonymy: The unit was first recognised by Taylor (1984) who termed it the Mile Road Formation. The unit was referred to as the Mile Road beds by Roberts et al. (1995). Derivation of name: Named from the Mile Road that traverses part of the unit in the Cowarra State Forest (GR 478700 6514700 Grants Head 1:25 000 sheet). Distribution: The Mile Road Formation is known only from the eastern Hastings Block where it has been recognised in the southern part of a slender wedge bounded by the Cowarra, Sapling Creek and Sancrox faults. It is mapped over an area of about 7 km’. Type section: Neither Taylor (1984) nor Roberts et al. (1995) designated a type section although the latter authors specified a type locality on the Cowarra Access Road (GR 479000 6514800 to 478900 6513100, Grants Head 1:25 000 sheet). This locality lies nearly along strike and encompasses only the lower part of the formation. We suggest that the type section be that extending northwest from the Cowarra Access Road at GR 478800 6513800 (base) along a tributary of Sarah Creek to GR 478300 6514300 (top) (Grants Head 1:25 000 sheet). Stratigraphic relationships: Neither base nor top of the unit is exposed. It is truncated downwards by the Cowarra Fault and is here interpreted as being stratigraphically overlain by the Touchwood Formation | — 2 km south of the Oxley Highway. Thickness: A maximum preserved thickness of between 1500 and 2000 m is estimated based on the mapped width of the unit and the assumption of an overall steep northwest dip and consistent northwest direction of younging. Content: Medium to thick bedded volcaniclastic siltstone and sandstone of intermediate-silicic Proc. Linn. Soc. N.S.W., 130, 2009 provenance, locally bioturbated and/or fossiliferous with crinoids, brachiopods and corals. Widespread breccias/conglomerates with coralline limestone clasts to c. | m set in a coarse sandy matrix. Grey hard massive silicic tuff interstratified with epiclastic rocks. Age and correlation: A small conodont assemblage from probably penecontemporaneously derived allochthonous blocks gives a precise age of the upper partofthelowervarcus Zone, early Givetian. Significant taxa are Polygnathus linguiformis klapperi Clausen et al., 1979, Polygnathus linguiformis weddigei Clausen et al., 1979, Polygnathus hemiansatus Bultynck, 1987 and Icriodus difficilis Ziegler et al., 1976. Additionally the blocks contain an abundant macrofauna of rugose and tabulate corals, spongiomorphs and brachiopods; the branching tabulate coral Thamnopora, occurring in the bioturbated matrix, suggests strongly that the blocks are penecontemporaneous, and that the conodont assemblage indicates a real age, at least for that part of the Formation. DAG, faabdardoaners ft noe eta instid nahh vate j ; rei nope ME ders tat ebocaidtonyd shins tik - pene, elt RAR "Na selina R all eure aT hiats (ic seine b, tT, Hy ay EB aba) a ie tae one in: ; hi f Bande gent fu idole nee on Ran ahi ats ‘Yh tergies setae i eae A oububaiah lege nt ‘aiteniorg’ weal by toad aii arveneitetotal ls he Bey 7 ree “ra ptyet yh be OG DIS f MOVE YAbtanenionh HERS pops Heyes aitiagie: tush vache Soa weoweemoeaMeNdi BL sale! FN eee arstiylh us LA ayabs, te '. tht y' rai Heal ee a iN eR ARS Oy. | \ ad Te Diath ‘| fd ipo eh at 1 Viaonieaseata | Coe), 9a Ty hve Os hbé Arp fitty At Hl FSD ap ats\Aorns +H washcoainth real hile ie sf3 fatty, nahi el att " deena i, ‘waren snwhesten beth ‘id yt TE Mt i afta ul Beto byte: SRO lrg wet ie to anebttad, Apmetiar’d orti bath. -lpeheganiocthl Pil AM a iain bisaelivstentd sith oat tat Aye cS Hee OMIM sensor anh r i; wpe un ats tania § CORE Si eas Sivelf i e . ; inane ot aitt Th Panty pet a cri hens si sine “itt 7 seals? Wd ponligrea bit enw Saw’ wT ALL ‘Wm det) gnileqse enero!) sdlvd bebagad a i =y lave if sae satan i viol ‘iio pin 1 eon ile sided Jeb “betgatte Tut is im Wee invotens. ad? a coianrtoy boowstoucll ta Ui oth a be alt Pony ey mA: SM sett ai boarrat odw wipertia ie eed | Thine wif aly oa Oo barton 7 fli bigot Shih itt mat boenet ited ho: nines Sede arate) teks ri Vitti bel to Tne] eet (toni OD 25 Fahy ‘ered oor a At rwautal at soit t0% bao bitty sit alii wh th suedw A50lH eyaiieall matees only, antral, rotyrale, fh to Meq modding ot th bazingesst tort Vey BOE KB TVG St ai: Bh - — caer te OR, wih yer: (peel) weal vet inition saoltuse ! galt gundite avitosd oy! & barniieob (200% yeredire.) oxtl nin iilsool ed 6 han 2p ard WORE oy OOREIED COURTS AO) baer | yasotmdT Agoda 00028: 7 bsol ajacit a) Yio waedaqmmosty bun odie aools oykve: yt ort ad) Hoggue OW! .oonanta outs to dra in St thot) owrthon gaibmshcatadlied ite abe (some!) GORE LEA OORBTD: A ty hoot pesogiee Feith CURLED hey lad aVh RO A doen die? tow rhitist “" bsoid NO EE: i belt eu . ts} chai “yend petto ‘oghndlbin aiden Dutra pateopurnt 2c 7 Leweepaeo: er sant ei aes god et bos alin erwok, ons ht yd ninhove Weotlge na ice | (a) 0 a aoe ty como: Bini Pepiaiany cadence A rest wiht Au ana Namie sid m berate’ Bae G08" 5 han atenalis: ‘bidbnad ey we eit cai -eceitsea 4a); ee eit Age Determination and Growth in the Male South African Fur Seal Arctocephalus pusillus pusillus (Pinnipedia: Otariidae) Using External Body Measurements C. L. Stewarpson’, T. Prvan*, M. A. MEYER? AND R. J. RitcHie** 'Botany and Zoology, Australian National University, Canberra, ACT, Australia. (Present address, Fisheries and Marine Sciences Program Bureau of Rural Sciences, The Department of Agriculture, Fisheries and Forestry, CANBERRA ACT 2601 Australia). "Department of Statistics, Macquarie University, NSW 2109, Australia. *Marine and Coastal Management (MCM), Rogge Bay, Cape Town, South Africa. *School of Biological Sciences, The University of Sydney, NSW 2006, Australia. *Corresponding Author (rrit3 143 @usyd.edu.au) Stewardson, C.L., Prvan, T., Meyer, M.A. and Ritchie, R.J. (2009). Age determination and growth in the male South African fur seal Arctocephalus pusillus pusillus (Pinnipedia: Otariidae) using external body measurements. Proceedings of the Linnean Society of New SouthWales 130, 219-244. Morphology, relative size and growth of the South African fur seal or Cape fur seal, Arctocephalus pusillus pusillus, from the coast of southern Africa are described and comparisons made to data available on the closely related Australian fur seal (Arctocephalus pusillus doriferus) and the New Zealand fur seal (Arctocephalus forsteri). Useful information can be gained from body measurements of seal carcasses provided canine teeth are extracted for aging. External body measurements (12 linear variables) were examined in relation to standard body length (SBL) and chronological age (y) using linear regression and non-linear least squares fitting as appropriate. Animals ranged from < 1 month to > 12 y. Of the 149 animals in the study, 39 were animals of known-age based on tagging; 34 were aged from highly reproducible counts of incremental lines observed in the dentine of upper canines (i.e., range 1-10 y); 10 were identified as adults > 12 y (1.e., pulp cavity of the upper canine closed); and 66 were not aged. At birth, male South African fur seals are 35% (c. 69 cm) of their mean adult size. At puberty, they are 57% (c. 113 cm). The foreflippers measure 25—26% (c. 18 cm) of standard body length (SBL) in pups, and 24% (c. 48 cm) of SBL in adults. The hind flippers are considerably shorter, measuring 19% (c. 13 cm) in pups, and 14.5% (ce. 29 cm) in adults. Axillary girth is usually about 57-67% of SBL. Growth of SBL was rapid during the early postnatal period with a significant growth spurt occurring at the onset of puberty (2-3 y). The rate of growth slowed significantly between 6 and 7 y. Social maturity was reached at about 9 to 10 y. Growth slowed thereafter. The mean SBL for aged males >10 y and unaged animals > 200 cm was 199 cm. Relative to SBL, facial variables and the fore/hind limbs scaled with negative slope relative to SBL or were negatively allometric; tip of snout to genital opening scaled with positive slope; and tip of snout to anterior insertion of the foreflipper was positively allometric. Relative to age, body variables scaled were negatively allometric. SBL was found to be a ‘rough indicator’ of age and age group. The growth kinetics of juvenile and adult the South African fur seal and the Australian fur seal are best described by the logistic and double exponential (Gompertz) models rather than the exponential von Bertalanffy model. Australian fur seals grow at a faster rate but asymptotic maximum sizes are similar in South African and Australian fur seals. Manuscript received 21 May 2008, accepted for publication 17 December 2008. Key words: Allometry, Arctocephalus pusillus doriferus, Arctocephalus pusillus pusillus, Australian fur seal, body, growth, growth curve modelling, pinnipeds, South African fur seal. evolutionary links within and between populations of INTRODUCTION the same species and between species. Growth and body-size estimates can be used for monitoring the Data on the physical growth of pinnipeds is effects of population pressures and changes in the important to understanding the biology, ecology and quality of the habitat of marine mammals (Bester BODY MEASUREMENTS OF SOUTH AFRICAN FUR SEALS and Van Jaarsveld, 1994). Within the Otariidae (fur seals and sea lions) quantitative descriptions of growth in body length based on animals aged from tooth structure, or on animals of known-age (1.e., animals tagged or branded as pups), are available for several species of fur seals and sea lions including the Australian fur seal (Arctocephalus pusillus doriferus) (Arnould and Warneke, 2002) which is very closely related to the South African fur seal (Wynen et al., 2001); the New Zealand fur seal (Arctocephalus forsteri) (Dickie and Dawson, 2003; McKenzie et al., 2007), the subantarctic fur seal (Arctocephalus tropicalis) (Bester and Van Jaarsveld, 1994), the Antarctic fur seal (Arctocephalus gazella) (Payne, 1979; Krylov and Popov, 1980; McLaren, 1993), the Northern fur seal (Callorhinus ursinus) (Scheffer and Wilke, 1953; Bychkov, 1971; Bigg, 1979; Lander, 1979; McLaren, 1993; Trites and Bigg, 1992, 1996) and the sea lions, Eumetopias jubatus, the Steller sea lion (Fiscus, 1961; Thorsteinson and Lensink, 1962; Calkins and Pitcher, 1983; Loughlin and Nelson, 1986; McLaren, 1993; Winship et al., 2001), and Otaria byronia, the South American sea lion (Rosas endle 1993): Physical growth in the northern fur seal and Steller sea lion have been studied in the most detail and is based on the largest number of animals of known age. The general growth curve for the Northern fur seal and the Steller sea lion is presumably representative of all highly polygynous male otartids. Male pups of Northern sea lions measure c. 66 cm at birth and grow at a steady rate (Scheffer and Wilke, 1953; Trites and Bigg, 1992, 1996). Growth is claimed to increase suddenly at 3-4 y (puberty) and slows soon after attamment of social maturity (McLaren, 1993). Estimated asymptotic length is about 189 cm for males > 4 y, and is reached by c. 12 y in most animals (McLaren, 1993). Growth curves of the Steller sea lion are basically similar in shape and also claimed to best fit a logistic rather than exponential saturation curve (Winship et al., 2001). Asymptotic maximum size of the Steller sea lon is much larger than fur seals: maximum size of males is about 3 m and 700 kg at about 12 y. The limited information on growth in body size available for South African fur seals was based on measurements that were aged physiologically (cranial suture age) rather than chronologically (y) (Rand, 1956). Unfortunately, in South African fur seals cranial sutures are not a very reliable guide to age (Stewardson, 2001; Stewardson et al., 2008). Comparisons will be made to data available on the Australian fur seal (Arnould and Warneke, 2002), the New Zealand fur seal (Dickie and Dawson, 2003; 220 McKenzie et al., 2007) and the subantarctic fur seal (Bester and Van Jaarsveld, 1994). Apart from studies by Scheffer and Wilke (1953) and Payne (1979) information on the relative growth of external body measurements of other fur seals is scant, e.g., axillary girth vs. standard body length, length of limbs vs. standard body length. Here we examine the body measurements of 149 male South African fur seals, Arctocephalus pusillus pusillus, from Southern Africa. Specific objectives were to: (i) describe the general morphology of the animal; (11) quantify growth of body measurements (12 variables) relative to standard body length (n = 134 animals) and chronological age (7 = 83 animals), (iii) determine if standard body length (SBL) is a useful indicator of age, (iv) compare three commonly used models for the growth kinetics of South African fur seals compared to Australian fur seals (exponential saturation curve or von Bertalanffy curve, Logistic curve and the double exponential or Gompertz curve) (Zullinger et al., 1984; Zeide, 1993). MATERIALS AND METHODS Collection of specimens South African fur seals were collected along the Eastern Cape coast of South Africa between Plettenberg Bay (34° 03’S, 23° 24’E) and East London (33° 03’S, 27° 54’E), from August 1978 to December 1995, and accessioned at the Port Elizabeth Museum (PEM). From this collection, 110 males were selected for examination. Apart from specimens collected before May 1992 (n = 38), all specimens were collected by the first author. PEM animals were aged based on dentition (n = 32), some PEM animals were aged using dentition growth rings, animals designated >12 y (n = 10) were animals with 12 growth rings in their teeth but their pulp cavities were closed and so no more growth rings could be deposited and so were at least 12 y old but could have been older. One animal (PEM2238) was collected NE of the study area, at Durban. Measurements from 39 males from Marine and Coastal Management (MCM), Department of Environment Affairs and Tourism, Cape Town were also available. These measurements were from animals that had been tagged as pups, and were therefore of known-age (1-13 y). MCM seal specimens are accessioned as MCM followed by a number. The accession numbers of all the animals used in the present study are listed in Appendix 1. The full data set is accessible in the public domain (Stewardson, 2001). Proc. Linn. Soc. N.S.W., 130, 2009 C.L. STEWARDSON, T. PRVAN, M.A. MEYER AND R.J. RITCHIE Body measurements Standard necropsies were performed and biolog- ical parameters recorded, based on recommendations of the Committee on Marine Mammals, American Society of Mammalogists (1967). Upper canines were collected for age determination. The skull is probably the most useful part of a seal carcass to retain for later study but it is not always possible to arrange for the skull of a dead seal to be retained. Nuisance seals are sometimes culled to satisfy the concerns of aquaculture and fisheries interests. From humane considerations, permits for such culls usually specify that the animals are fatally shot in the head, which ruins the skulls for morphological studies, but teeth for aging can usually be retrieved (Thorsteinson and Lensink, 1962; Pemperton et al., 1993; Winship et al., 2001; Arnould and Warneke, 2002; McKenzie et al., 2007). Body measurements of seal carcasses are most useful if canine teeth are extracted for aging. Measurements (12 variables) were taken to the nearest 5 mm (0.5 cm) using a flexible tape measure or vernier callipers as appropriate (Figure 1). Although body weight and blubber thickness were recorded, these measurements were not included in the analysis because they can vary according to physiological condition, e.g., body condition is influenced by seasonal fluctuations in food supply, illness or injury, and breeding condition. The blubber of Australian fur seals is known to vary seasonally with a maximum in late austral spring (Arnould and Warneke, 2002). Apart from specimens collected before May 1992, all PEM measurements were recorded by the first author. The majority of MCM measurements were recorded by the third author. Age determination The age of animals was estimated from counts of Growth Layer Groups (GLGs) observed in the dentine of thin tooth sections (Payne, 1978; Oosthuizen, 1997; Oosthuizen and Bester, 1997; Stewardson et al., 2008). Upper canines were sectioned longitudinally using a circular diamond saw. Sections were ground down to 280-320 pm, dehydrated, embedded in resin and viewed under a stereomicroscope in polarised light (Oosthuizen, 1997; Oosthuizen and Bester, 1997). Each section was read by one individual five times, without knowledge of which animal was being examined (repeated blind counts) similar to Payne (1978). Ages were rounded off to the nearest birth date. The median date of birth was assumed to be | December (Shaughnessy and Best, 1975), which is similar to the mean date of birth for Antarctic fur seals (Payne, 1978). The median of the five readings was used as an estimate of age. Outliers were discarded as reading errors. Proc. Linn. Soc. N.S.W., 130, 2009 Currently, examination of tooth structure is the most precise method of age determination in pinnipeds (McCann, 1993), including South African fur seals (Oosthuizen, 1997; Oosthuizen and Bester, 1997). However, this method can only be used in South African fur seals < 12 y. At about 12 y of age, closure of the pulp cavity terminates tooth growth and no further growth rings are formed. Arnould and Warneke (2002) claim that growth rings could be distinguished in male Australian fur seals up to 16 y and a similar upper limit of about 15 y was found in the Antarctic fur seal by Payne (1978). Payne (1978, 1979) also found that useful ages could be estimated from growth lines in the cementum of the teeth of Antarctic fur seals (A. gazella) but this method was not attempted in the present study. Ofthe 149 animals in the study: (1) 39 were known- age MCM animals; (11) 34 were aged from counts of incremental lines observed in the dentine of upper canines, i.e., range |—11 y; (111) 10 were identified as adults = 12 y (pulp cavity of the upper canine closed); (iv) 66 were not aged but could be classified into subadults and adults based upon SBL; allowing for (1), (11) and the problem animals mentioned in (iii) above, there was a total of 73 animals of known age available for modelling of growth vs. age. For this study, the following age groups were used: pup (< | month to 6 months); yearling (7 months to 1 y 6 months); subadult (1 y 7 months to 7 y 6 months); and adult (= 7 y 7 months) (Table 1). Very old animals of known-age were not available for examination. Estimated longevity is c. 20 y, based upon the lifespan of zoo animals and known life-spans of other fur seals (Wickens, 1993). Australian male fur seals (A. pusillus doriferus) have a lifespan of about 20 years but female Australian fur seals are known to live to over 20 y (Arnould and Warneke, 2002). The New Zealand fur seal (A. forsteri) (McKenzie et al., 2007) and the Steller sea lion (Eumetopias jubatus) (Winship et al., 2001) both have similar lifespans of about 20 y for males and well in excess of 20 y for females. Australian Material The South African fur seal data on SBL vs. age were compared to published material from Arnould and Warneke (2002) on Australian fur seals. Data were read off the graphs in their published paper (Arnould and Warneke, 2002) with an accuracy of the SBL readings of about + | cm. Fits of their data were then compared to similar data for South African fur seals from the present study using the same statistical software. Jia \h BODY MEASUREMENTS OF SOUTH AFRICAN FUR SEALS Snout & SS —T \\ 1 } \\ xX NN Poe NES >~ 7 (' " \ \ Rees Sm 1 ; \ 2 le | Lowe § Lars Wey \ \ 4-9 | jek ae Nee A es . | ae aa f {ieee S ok i=) el nell cmmastiemsenticennesticemedlienssntieenes tl anmnesthacam dices ’cocenetiteeelimemtianeticnmatiemes tant °,” litecnsesticmae’ianeatienn lt emactinnaticnns!ammatiemmd tensa onmaalicmdicnun!cmentemeeedlmeatiemntieenaltcnctemadimmeticaed a a as Oe er er ar nr wr car ee ce acer arr ae ae eee ae eee se es ~ I os ee ae | He teach | | \ | \ | | | , \ | ' i | | \ | 4 MA | } Pee ee tte tf \ | ee \ | iN hey’ Biblicous rd / \ / \ I / \ | . / | | Opening \ ¢ fe ee \ \ \ I \ I ; Anus \ I {—- —— cca Stat gt and beg | Tail } —_ . ym en wen es JA evel Veo / Vea / \ | / \ / \ \ 1 Lo \ Al Figure 1: Diagram of a male South African Fur Seal showing how individual body measurements were taken. All measurements were taken with the animal lying on its back. B1, Circumference of head at canine; B2, circumference of head at eye; B3, tip of snout to centre of eye; B4, tip of snout to centre of ear; B5, tip of snout to angle of gape; B6, standard body length or SBL (straight line from tip of snout to tip of tail with animal lying on its back); B7, ventral curvilinear length (tip of snout to tip of tail over body curve); B8, tip of snout to genital opening; B9, tip of snout to anterior insertion of the foreflipper; B10, length of foreflipper (anterior insertion to tip of first claw); B11, axillary girth; and B12, length of hind flipper (anterior insertion to tip of first claw). All body measurements were made in cm. DU) Proc. Linn. Soc. N.S.W., 130, 2009 C.L. STEWARDSON, T. PRVAN, M.A. MEYER AND R.J. RITCHIE Table 1: The age distribution of Male South Af- rican Fur Seals. Pups were defined as animals < 1 month old. Animals 1-10 y: 37 MCM animals were of known-age; 34 PEM animals were aged from counts of incremental lines observed in the dentine of upper canines. Animals > 12 y: 2 MCM animals were 13 y; 10 PEM males were > 12 y, i.e., the pulp cavity of the upper canine was closed. Statistical analysis Body variable expressed in relation to standard body length Growth in body measurement, relative to standard body length (SBL), was calculated as follows: body measurement (cm)/SBL (cm) x 100% As the variance of the ratio estimate is difficult to validly estimate, particularly on small samples, percentages must be interpreted with caution, i.e., both y and x vary from sample to sample (Cochran, 197 Te pals3): Body length as an indicator of age The degree of linear relationship between log body measurement (log SBL) and age (y) was Proc. Linn. Soc. N.S.W., 130, 2009 calculated using the Spearman rank-order correlation coefficient. Linear discriminant analysis can be used to classify individual seals into mutually exclusive age groups based on seal body length. The dependent variable (y) is the age group and the independent variable seal body length (x) is the feature that might describe the age group. For each age group we can determine the mean of seal body length (x, ) and for each seal we compute the Mahalanobis distances of the body length (x) to the mean seal body length of age group 7: EQUATION 1 Di(x)=—2|x, S"x=4%, 87x, |+ x7S x where, S is the pooled sample variance matrix. Since we are dealing with univariate data we have xe =X; , X' = X and S being the pooled sample covariance. The term in square brackets is the linear discriminant function. We allocate an observation (x) to the age group (pup, yearling, sub adult, adult), which gives the smallest calculated Mahalanobis distance. This is equivalent to allocating the observation (x) to the age group which has the largest linear discriminant function value (Anderson, 1984). Growth Models The most commonly used growth models (SBL vs. age) for post-natal growth of marine mammals are the exponential saturation curve, known as the von Bertalanffy model, the logistic ‘curve and the double exponential or Gompertz model (Zullinger et al., 1984; Trites and Bigg, 1992, 1996; Zeide, 1993; Winship et al., 2001; Arnould and Warneke, 2002; McKenzie et al., 2007). In most cases where these equations have been used, a time base adjustment (moving the x-axis) has been used to optimise the fit but this is not a good statistical procedure. No attempt is usually made to estimate the errors of the fitted parameters. In the present study, the models have been expressed in forms where the unknowns were the asymptotic maximum size, the apparent pup size (P) and an exponential constant. Models are for post-natal growth; they are not intended to model the growth of suckling pups and the apparent pup size (P) does not necessarily reflect the actual birth size: 223 BODY MEASUREMENTS OF SOUTH AFRICAN FUR SEALS EXPONENTIAL SATURATION OR VON BERTALANFFY CURVE EQUATION 2 Y=(E, -P)(1-e")+P or Y=E,, -E,e™+Pe™ where, E isthe asymptotic maximum size, Pis the apparent pup size, kis a exponental growth constant tis tim e. LOGISTIC EQUATION EQUATION 3 E = ! iE ‘ | 1+ | Serie. 4, M, I / F ! where,E isthe asymptotic maximum size, é % E,, oe aes te eee cree | a 1 | is a scaling constant, i Pis the apparent pup size, kis an exponental constant, tis time. DOUBLE EXPONENTIAL OR GOMPERTZ EQUATION - EQUATION 4 a er | wa E... = [xP )- uF Cie where, E isthe asymptotic maximum aze, [Ln(P)-LnfE, jlisa scaling constant Pis the apparent pup size, kis an exponential constant, tis time. For Equations 2, 3 and 4 the incremental component of growth (sat) 1S; EQUATION 5 gon Ms = ge The approxim ate error for E ners (AE pew his: , Ti SH Races AE aonth = “f (anlet jar (AP } where, dh APisthe error of the apparent pup size. 224 13 the error of the maximum body size, The growth of suckling pups would be expected to be governed by a different growth curve and so the apparent pup size (P) is an abstraction. There are also statistical limitations of the models. Three (3) unknowns have to be fitted. It is much more difficult to fit an equation with 3 unknowns than one with 2 unknowns. The characteristics of the underlying function can also give rise to difficulties; the logistic equation, in particular, is notoriously difficult to fit (Zullinger et al., 1984). The equations cannot be adequately fitted if there is an insufficient amount of data to clearly indicate curvature towards an asymptotic maximum. The errors of the fitted parameters can be estimated using matrix inversion methods (Johnson and Faunt, 1992). However, most attempts to use such growth curves on mammals and growth of trees have not used enough data points, resulting in the asymptotic errors being so large that the estimates of the fitted parameters are not useful (Zullinger et al., 1984; Zeide, 1993). Most previous attempts to fit various types of exponential saturation curves have used data where the equations have been simplified by using a fixed estimate of the initial condition at t = 0 (the apparent pup size), hence simplifying the equations to equations with only two unknowns (Australian fur seals - Arnould and Warneke, 2002; New Zealand fur seals — McKenzie et al., 2007; Steller sea lion — Winship et al., 2001). Least squares fitting routines assume that the error in the dependent variable is normally distributed and independent of the magnitude of the independent variable. In many biological situations this assumption is not valid because the error of the dependent variable increases with increasing magnitude of the independent variable. A constant relative error is often a more realistic assumption to make for biological data. The usual procedure to deal with situations is to log/log transform the data and then use a least squares fitting procedure on the transformed data. In the present study, we found no great improvement in the curve fits (in terms of correlation r) using log/log transformed data. Plots of residuals vs. predicted Y-values did not indicate a systematic increase in the size of the residuals as the predicted Y-value increased. No log/log transform was needed. Bivariate allometric regression The relationship between value of body measurement and: (i) SBL and (11) age (y), was investigated using the logarithmic (base e) transformation of the allometric equation, Proc. Linn. Soc. N.S.W., 130, 2009 C.L. STEWARDSON, T. PRVAN, M.A. MEYER AND R.J. RITCHIE y = ax’, which may equivalently be written as log y = log a+ b log x. ‘Robust’ regression (Huber M- Regression) was used to fit straight lines to the transformed data. The degree of linear relationship between the variables was calculated using the Spearman rank-order correlation coefficient, r (Gibbons and Chakraborti, 1992). This is a non- parametric procedure. Since the log-transformation is monotonic you get the same value for r on transformed or untransformed data. It is important to note that the regression equations relating to overall growth are not used on body measurements that are likely to vary with seasonal variations in body condition that are known to occur in this species (e.g. Rand, 1956). For example, body girth or weight would be inappropriate parameters to use in such analyses. Statistical tests of hypotheses about model parameters are only validifthe model assumptions hold (1.e., errors are independently and identically normally distributed, with zero mean and with a variance (o”) (Weisberg, 1985, p. 24, 156). The standard approach is to first examine the residual values versus fitted plot. If this is a random scatter about zero then it is valid to assume the model is adequate and proceed to check the normality assumption. In the present study, the following tests for checking for normality were used: (i) Anderson-Darling, (ii) Ryan-Joiner and (111) Kolmogorov-Smimov (Cochran, 1977). We used the following test statistic to test one of the hypotheses given below about the slopes of the fitted lines: EQUATION 6 — _ b=) SE(b) where, bis our estimate of the slope using robust 0 regression and SE ( b ) is the standard error of b . Under the null hypothesis the test statistic T has a ¢ distribution with - 2 degrees of freedom (df). The following hypotheses were tested: H,: b= 1 (isometric) versus H,: b # | (either positively or negatively allometric); H,: b > 1 (positively allometric); H,: b < 1 (negatively allometric). Statistical Software Statistical analysis and graphics were implemented in Minitab (Minitab Inc., State College, Proc. Linn. Soc. N.S.W., 130, 2009 1999, 12.23), Microsoft Excel 97 (Microsoft Corp., Seattle, 1997) and S-PLUS (MathSoft, Inc., Seattle, 1999, 5.1). The EXCEL 97 routines for non-linear least squares fits and calculation of the asymptotic errors of the fitted parameters for the von Bertalanffy, Logistic and Gompertz equations (Equations 2, 3 and 4) are available from Dr R.J. Ritchie (rrit3 143 @usyd. edu.au) upon request. Terminology A juvenile is a weaned pup that has not yet achieved adult size. Puberty is when reproduction first becomes possible (production of sperm in quantity), and social maturity is the age when the animal reaches full reproductive capacity (physically able to establish and maintain a harem). Sexual development of male South African fur seals is discussed elsewhere (Stewardson et al., 1998). RESULTS Age determination based on dentition (intra- observer variability) Counts of GLGs (growth layer groups) in canine teeth were found to be highly reproducible. Of the 34 PEM animals for which GLGs were counted, 14 (41%) had all five readings equal; 16 (47%) had one reading out of 5 different from the mode; and 4 (12%) had 2 readings out of 5 different from the mode. Age determination (variability between known- age and canine aged animals) Standard body length (SBL) was selected to investigate whether MCM (animals of known-age) and PEM (canine aged animals) animals were similar with respect to age. When comparing the (robust) regression line for SBL on age for MCM animals with SBL on age for PEM animals, partial t-tests indicate that age is important (t = 7.07, p < 0.001), even after adjusting for group and age-group interaction; but they provide little information on group (f= -0.82, p = 0.42) and age group interaction (¢ = 0.87, p = 0.58), hence one straight line can be fitted to the data. These statistical conclusions were verified by examining graphical displays of fitted values and residuals. Thus PEM and MCM animals were not significantly different with respect to age distribution. This conclusion is supported by the sequential F test, provided the sequence of terms added sequentially (first to last) was: (i) none (i.e., fitting a line parallel to the x axis); (ii) age (F = 817.69, p < 0.001) (one straight line); (iii) museum (1.e., MCM and PEM) (F = (0.0659, p = 0.7984) (two parallel lines); (iv) age x 225 BODY MEASUREMENTS OF SOUTH AFRICAN FUR SEALS museum interaction (F = 0.1883, p = 0.6661) (two lines not necessarily parallel). Bivariate allometric regression Regression statistics for body measurements on SBL and age (1-10 y) are given in Appendix 3 and 4. Overall, correlation coefficients were moderately to strongly positive, i.e., most points on the scatter plot approximated a straight line with positive slope, r => 0.70. Exceptions included tip of snout to centre of eye (B3) with age and SBL (r = -0.008 and r = 0.15 respectively); tip of snout to angle of gape (BS) with age (r = 0.56); circumference of head at canine (B1) with age (r = 0.59). Although correlation coefficients indicate that linearity was reasonably well approximated for most variables by log-log transformations, a linear relationship did not necessarily best describe the relationship. In the present study, we have attempted to fit more complex models in the case of SBL vs. age with the specific aim of comparing our growth curves with those found for the Australian fur seal (Arnould and Warneke, 2002)(see below). Growth of body variables Most variables were significantly positively correlated with each other, r = 0.68 (Appendix 2). Exceptions were: (1) tip of snout to centre of eye (B3) with all variables; (i1) circumference of head at eye (B2) with tip of snout to angle of gape (B5) (r= 0.61); and (111) circumference of head at canine (B1) with tip of snout to angle of gape (BS5) (7 = 0.63). Circumference of head at canine (B1) Growth of circumference of head at canine (B1) was variable relative to age, r = 0.59 (Appendix 4). Overall growth expressed negative allometry relative to SBL and age (Appendix 3, 4), increasing by 57% at 10 y relative to pups (RTP) (Table 2). Growth increment decreased with increasing SBL until about 7 y (c. 15% of SBL) (Table 3). The mean B1 of males > 10 y (including unaged animals > 200 cm and of indeterminate age > 12 y) was 31.8 + 1.2 cm (n=S). The maximum-recorded value was 35.0 cm (animal MCM3017, SBL 209 cm, 12 y 11 months). Circumference of head at eye (B2) Growth of circumference of head at eye (B2) was rapid during the early postnatal period and continued to increase until at least 13 y. Overall growth expressed negative allometry relative to SBL and scaled with negative slope relative to age (6 = 0.12) (Figures 2a, b; Appendix 3, 4), increasing by 65% at 10 y (RTP) (Table 3). Growth increment decreased with increasing SBL until about 7 y (c. 22% of SBL) (Table 2). Mean B2 of males > 10 y (including unaged animals > 200 cm and of indeterminate age > 12 y) was 45.8 + 1.8 cm (n = 6). Maximum recorded value was 53.0 cm (animal PEM676, SBL 197 cm). Tip of snout to centre of eye (B3) Growth of tip of snout to centre of eye (B3) was highly variable relative to age, r= -0.008, and SBL, r= 0.15 (Appendix 3, 4). Growth increment decreased with increasing SBL until about 9 y (c. 5% of SBL) (Table 2). Mean B3 of all males > 10 y (including unaged animals > 200 cm and of indeterminate age > 12 y) was 10.4 + 0.6 cm (7 = 10). Maximum recorded value was 14.4 cm (animal PEM2194, SBL 194 cm). Table 2 (Pages 227-228): Summary statistics for body variables (B1—B12), according to age (y) and age group of male South African Fur seals. Data presented as mean body measurement in cm + S.E., followed by coefficient of variation in round brackets, and body variable expressed as a percentage of SBL. Maximum value of each variable (males of unknown-age) is also presented. Variables: B1, Circumference of head at canine; B2, circumference of head at eye; B3, tip of snout to centre of eye; B4, tip of snout to centre of ear; B5, tip of snout to angle of gape; B6, standard body length (SBL); B7, ventral curvilinear length; B8, tip of snout to genital opening; B9, tip of snout to anterior insertion of the foreflipper; B10, length of foreflipper; B11, axillary girth; B12, length of hind flipper. Variable B3 was poorly correlated with body variables and age (Appendices 1, 2, 3 and 4), therefore has been excluded from further analysis. B7 was shown to be a poor indicator of SBL, therefore was exclud- ed from further analysis. B11 may be influenced by seasonal change and illness, therefore was excluded from further analysis. Sample size (n) is the number of dentition-aged and known-age (tagged) animals. Sample size given in square brackets where this does not equal total sample size. The data summary includes calculations of the mean of each variable + S.E. for the 7 largest males (> 200 cm) of known or unknown-age; maximum value in square brackets, followed by sample size. 226 Proc. Linn. Soc. N.S.W., 130, 2009 C.L. STEWARDSON, T. PRVAN, M.A. MEYER AND R.J. RITCHIE p=ulo7sT] 6S #0 TLI 19 — (+6) [ell LrFLLLi =(W) STFS 907 —(9°9) 09 F681 a AO) [rl ©7F8OLI — (3°) [e] S€ FO 191 —(¢L1) [Le] SEFLIEL — (79) [sl] PE FS LSI —(°2) l6l eI =8°Stl —(97@) [7] STS 9E1 —(F 11) [gs] 0S FEI —(0'8) Or STII — (Sp) 61 #8 £6 —(r'L) TF 3°06 = (0) STF E69 9d 69 %b'9 (L'b1) lol] +0 F601 %S'°9 (ZS) SOFS EI %0°9 (L'9) POFIII %b9 (88) POFLOl %8°9 (061) [s] 8066 %TL (11) [tr] TOF €6 %E9(L 91) SOFE6 %O'L (9'L) [6] €0F7OI %8°9 (L' +1) [e] L0 +98 %6'L (8'L) €0F66 %OL(E rT) 9098 MES (p11) vOFSL Mrs (ETI) COFLL %EOI (S91) LOFUL sd €L %9 TT (OID) SOFE OC %6 IT (rE) 90 9b %6 OI (61) 70 +007 %0°TI (16) 80 FS 07 %L IT (C6) LOF68I %9'EI (S01) €0 F081 MOLI (LL) pOoFTsl %0° ET (78) SOF E6I Mp El (CTT) 60F88I %Y rl (6L) SOFE SI %7 ST (8'S) POFITLI %6'ST (LL) SOF6TI %IST (9'L) vOFLEL %OLI (MO LOFS8TI pa €=ul6+T] COFOFI 79 %eS (P81) [st] pO 16 MES (STI) 60FT' II %7S (L'07) le] T1456 %Ib (TST) [r] LOFTS %L'S (C91) 90F8'8 %L"L (S81) [Le] €0 +86 %79 (9°81) [L190 +06 %CL (OFT) [s]SoFrO0l %9°9 (FST) L0*66 %9'L (LL1) SOF16 %9'6 (TET) [rp] 71 = LOI %8'OI (S Sz) [rl € 1 $71 %0'6 (C81) [L190 E8 %IEl (791) 60F16 ed 89 %9 17 (S01) [st] VL +T6E %OIT(S LI) SSS bb %6 1 (LL) [el S1 +L 6¢ %9 17 (8°S) OLFPLE %b IT (S01) [s] 81 ¥9'8€ %79T (6'8) ler] SOF STE %ETC (CL) LOIS OFL +E %EST (O'L) SOFTLE %8'9T (6'L) irl pL F8'rE %L LT (8'p) SOFEVE %C-8z (E'S) SOFTTE %BTE (L' 11) 91 + 8'0E %6 OE (Z'EL) [LI r LF 6LZ %L ve (E01) VIF be cd p=uloet] 90F8'II 89 %0'SI (CTI) [r1] 60+ £97 %EST (LSI) SEFSTE %L rl (FL) [el 11 + 9°92 %T~SI (Ch) [rl] $0 +092 %8'+I (9'6) [s]O1L CH %6 LI (56) ler] €0 + 9°E7 %I-SI (901) lol] 80 #L'€Z %0'LI (8'L) 90 F6 be %6 LI (r'€) [Ir] +0 + O'F~ %9 61 (9'L) 90 ¥ Ibe %L6I (8') SOF TC %S CC (TCI) T1+1I %OTT(S TL) 60 +961 bre (E11) T1L¥691 Id Si II Ol € (u) ezis ojduies ¢=ulose] OCFETE el Ol mG (A) o8y [siayorrg Ul onyeA “xeUT] WO YNZ < SOTPUI [[v 1OF Uv 18}0], yNpVy ynpeqns SUIpIvO, dng dnois os yal Proc. Linn. Soc. N.S.W., 130, 2009 BODY MEASUREMENTS OF SOUTH AFRICAN FUR SEALS Table 2 continued Age group Pup Yearling Subadult Adult Total Mean for males > 200 cm [max. value in brackets] Age (y) =i 10 13 31.3 +2.0 [35.0] n=3 Sample size (n) 3 10 11 45 17 73 44.34+3.2 [50.0] n=3 B7 70.9 + 3.6 (8.7) — 95.3 + 3.9 [3] (7.1) - 149.7 +2.7 [3] @2Q)= 155.3 +5.2 [3] Gy 158.5 + 4.3 [5] (6.0) - 155.2+2.6 [11] 6S)= 166.0 + 2.1 [5] Qs3)= 185.8 + 3.4 [4] Cu 203.7 + 4.9 [3] (4.2) - — [0] 182.0 + 4.9 [12] (9.2) - 29 91.0 + 3.4 [98.0] n=4 B8 55.6 + 1.7 (5.3) 80.2% 75.9 +2.2 (8.1) 83.7% 79.6 + 2.4 (6.8) 84.9% 98.1+42.1 (4.9) 87.0% 107.2 + 3.6 [8] (9.5) 86.2% 124.5+4.8 (8.7) 84.4% WGA 22 M7) (4.2) 87.2% 3-5 2e 25) (6.3) 84.9% 115.7 + 2.9 [44] (16.5) 86.0% 136.6 + 3.1 (5.5) 85.8% 152.6+ 2.6 (3.8) 89.6% 159.3+5.4 (6.8) 87.1% 178.5 + 3.5 (2.8) 86.4% 151.6 + 3.8 (10.2) 87.4% 72 49.0 + 2.7 [55.0] n=4 B9 31.7+0.9 (4.8) 45.7% Al = 1.7 (11.6) 45.3% 37.7+£0.8 (4.6) 40.2% 48.9+2.1 (9.5) 43.3% 52.5 + 1.7 (9.8) 42.6% 62.7442 (14.9) 40.5% 65.5 + 3.4 (16.3) 43.6% 71.8+2.1 (9.6) 45.8% 59.2+2.0 (22.2) 43.4% 76.6 + 2.2 (7.1) 50.1% 83.8+5.8 (15.6) 48.4% 87.8 + 8.5 (19:5) 48.1% 91.5+3.5 (5.4) 44.3% 83.142.9 (14.1) 47.9% 73 135.0 + 34.0 [169.0] n=2 B10 17.6+ 1.6 (16.2) 25.4% 22.4+ 1.2 (15.0) 24.7% 23.5+0.4 (4.3) 25.1% 27.4+£1.4 (11.2) 24.3% 30.1+ 1.1 (11.0) 24.5% 35). 7/ 2 IA! (9.0) 23.9% 33.6 + 0.9 [9] (7.7) 23.0% 34.7+ 1.1 (10.1) 22.4% 31.6 + 0.7 [44] (15.3) 23.6% 35.2+1.6 (11.5) 21.3% 40.6 + 0.9 (5.0) 24.1% 40.3 + 2.1 [3] (10.2) 22.0% 48.4 + 3.6 (10.5) 23.4% 39.5 + 1.3 [16] (13.5) 22.7% 72 28.8 + 1.4 [29.2] n=3 Bll 39.6 +3.5 (15.5) 57.1% 53.14£4.6 (24.4) 58.5% 58.2431 (11.8) 62.0% 73.9 +2.0 (6.2) 65.5% 80.2+2.3 (8.5) 64.7% 85.8 + 0.8 [2] (1.2) 62.8% 91.4+2.1 [9] (6.8) 62.7% 100.2 + 3.1 [7] (8.3) 64.4% 83.2+2.4 [37] (17.5) 63.8% 90.6 + 4.6 [2] (7.2) 57.5% 114.5 +2.9 [4] (5.1) 67.0% 111.9+6.9 (12.2) 61.2% — [0] 108.7+4.1 [10] (12.0) 62.8% 58 B12 13.3£0.7 (9.4) 19.2% 15.1+£0.4 (8.0) 16.6% 16.0 + 0.6 (8.3) 17.0% 18.1408 (9.3) 16.0% 18.6 + 0.5 (8.8) 15.2% 21.7+1.3 [3] (10.1) 15.0% 21.2 +0.4 [9] (5.9) 14.5% 23.6 +0.7 [10] (9.1) 15.2% 20.2 + 0.5 [41] (15.0) 15.3% 26.0 + 1.0 (9.7) 15.8% 28.2413 (10.1) 16.5% 27.1 + 1.6 (11.7) 14.8% 27.7£1.5 (7.7) 13.4% 27.1 £0.46 (9.8) 15.3% Proc. Linn. Soc. N.S.W., 130, 2009 69 228 C.L. STEWARDSON, T. PRVAN, M.A. MEYER AND R.J. RITCHIE 89 €L [BIOL — 0°98 C el [RG acOIle 10r -/A9Cll SiS vr -v98I VL ‘L€91 1G Hf V Ol Pe we bees Spee Wie co [rv] pee ech [rl] €8-VIIl | vSl-S O€1 £6 :-9791 | LIL -S VLI 19 {Or € 8-106 18-072 ey (EE S 6 Pere OO eo Le] [s] aL [s] [s] Lot] “ofc 66) pic: [3] 9. “c- ‘9: Lot] lor] [6] [6l oe ae (6l [6] apes tee er) a op nee lesisiso com = voneoige | cor ove leo ea I) | Bec cececsl ee ee S| 6 891 -0 £9 86-696 | S CI- -6 07 € 1-9 90--L IV | . ‘ . . © . . 6 . [3] [2] . 6 . . © . . 4 . . 4 . Of -S 6E 66:CIL VL-8S9 6 OIG ZA as CVI -18et 99-1 SS | JES) ok, (OY 68 -9 CV 6 14 Gale Cces 9°91 87S 9°67 -€'VS TET VOL €07 :L' C9 SOM LO | OSI lS Cin oll Ce 1S OTE S te J LS 6 1S OEE OR ell Ol 617: EL VE-ESE 07:76 88 -L'97 TOI 0°87 LL 9% iS ¢ | Wapeqns EET CEI WAG WILE 667 -6'67 SOE COE 6 0€ -6 0€ il it COI “S91 191 4 PSI of SM 8 [ | Surpieox - - ; - - ea (W) | (4) dno oZIS (TdS) 94 cd tas ca De | Fee ay o3y “SISATVUL JY) WIOA S[VUTIUL IUOS JO UOISN[IXO OY} 0} INP 9ZIs s[ dures [v}0} [enbo you svop Siu) o10ya [u] sjoyov1q a1enbs ul UdAIs ozIs ajduleg -oAQQT x ['’4/(' A —’A)] “ort ‘suMM]OD JULATA1 JY) JO opis puvy JYSII ay} UO pojyussaid aie “(T-} = av) 1vaXk snorAaid ay) 0) JAQLIAA YIMOIS 107 Sane ‘dnd z sv azIs UMOUY 94} SI QA puL CT ‘OT “6 ‘8 ‘ZL ‘9 ‘S ‘FE ‘ZT =} ELOY %OOT X [ACA — *A)| Corr ‘SUUIN[OD JUVAIII.A JY} JO OPIS puvy Jo] 94} UO pajuasoAd 9.18 0.19Z ISU 0} JATIJLIIA YJMOAG OJ SIN[VA *PIPAOIIA JOU IAA S[RUIUL ISI} JO Z[ 10J STS “9't ‘SUOIIPUOD YSNOA JO asnevd0q TAS P1090. 0} a[qissod sve jou svAd JI (YD}vI-Aq) vas Jv PoANSvoU S[eUTIUL 10,7 ‘s[VUITUL (posdse}) 95v-UMOUY puL pase -u0l UEP Jo AaquInu ay} SI (U) 9ZIs aJduIKS *(Z B]quUI 99S) SISATeUL WOA, PIpNIXo 919M [[q pue Lg ‘Cg SorqulAVA *7Z 9Iquy, 105 sv (ZL q-Tq) Seyquriva *‘pojuasaid os[e Si (oSv-UMOUYUN JO SopVUl) o[quIIvA YIva JO INVA WINUIIXLJA] *S}OYIVAG Ul UOHLLIVA JO JUIIDYJIOD & YJIM “YS F UD Ul 91¥ SJUIWIAINSvOU [TY Soy aeak snoraaid ay) uso. (11) pue $A WO ‘0.192 98v ye (1) :sjvas ny UVdILAPY YINOS a[vUl Jo USWIOANSvaU ApoOg JO aN[vA ULI JY} 0} VANLIAA (71 G—-1q) Sorquiaea Apog Ul YYMOAD :¢ [Quy WY, Proc. Linn. Soc. N.S.W., 130, 2009 BODY MEASUREMENTS OF SOUTH AFRICAN FUR SEALS Tip of snout to centre of ear (B4) Growth of tip of snout to centre of ear (B4) was rapid during the early postnatal period and continued to increase until at least 13 y (Table 2 and 3). Overall growth expressed negative allometry relative to SBL and scaled with negative slope relative to age (b = 0.04) (Figures 3a, b; Appendix 3, 4), increasing by 70% at 10 y RTP (Table 3). Growth increment decreased with increasing SBL until about 7 y (c. 12% of SBL) (Table 2). The mean B4 of all males > 10 y (including unaged animals > 200 cm and of indeterminate age > 12 y) was 22.7 + 0.8 cm (n = 7). The maximum-recorded value was 25.2 cm (animal MCM3125, SBL 204 cm, 13 y). Tip of snout to angle of gape (B5) Growth of tip of snout to angle of gape (B5) was variable relative to age, r = 0.56 (Appendix 4). Overall growth scaled with negative slope relative to SBL (6 = 0.64) and expressed negative allometry relative to age (Appendix 3, 4), increasing by 55% at 10 y RTP (Table 3). Growth increment decreased with increasing SBL until about 7 y (c. 6% of SBL) (Table 2). The mean BS5 of all males > 10 y (including unaged animals > 200 cm and of indeterminate age > 12 y) was 13.2 + 0.7 cm (n = 7). The maximum recorded value was 15.0 cm (animal PEM676, SBL ISS eran), Standard body length (B6 or SBL) Growth of SBL (B6) was rapid during the early postnatal period with a significant growth spurt between 2 and 3 y (two sample t test: p-value = 0.008; df = 5). The rate of growth slowed significantly between 6 and 7 y (two sample t test assuming unequal variances: p-value = 0.011; df= 9). A weak growth spurt was observed at 9 and 10 y but could not be examined statistically, 1.e., this secondary growth spurt may be attributed to sampling error. Growth increased by 164% at 10 y RTP (Table 3). Considering that the 13 y old males measured 206.5 + 2.5 cm (n = 2), and mean SBL of all males > 10 y and/or unaged animals > 200 cm was 197 = 4.1 cm (n = 15), growth appears to slow after attainment of social maturity (Table 2). Tip of snout to genital opening (B8) Growth of tip of snout to genital opening (B8) was rapid during the early postnatal period and continued to increase until at least 13 y (Table 2 and 3). Growth increased by 186% at 10 y RTP (Table 3). In subadults and adults, mean value remained at about 86% of SBL (Table 2). Overall growth scaled with weak positive slope relative to SBL (6 = 1.04) and negative slope relative to age (6 = 0.02). The maximum recorded value for parameter B8 was 184.0 cm (animal PEM2256, SBL 198 cm). The mean B8 of all males > 10 y, including unaged animals > 200 cm) was 171.143.4cm (n=7). Tip of snout to anterior insertion of the foreflipper (B9) Growth of tip of snout to anterior insertion of the foreflipper (B9) was rapid during the early postnatal period and continued to increase until at least 10 y (Table 2 and 3). Overall growth expressed positive allometry relative to SBL, and negative allometry relative to age (Figure 4a, b; Appendix 3, 4). Growth increased by 177% at 10 y RTP (Table 3). Mean SBL of all males > 10 y, including unaged animals > 200 cm was 94.2 + 3.1 cm (m = 7). Maximum recorded value for B9 was 110.0 cm (animal PEM2374, SBL 186 cm). Length of foreflipper (B10) Growth of length of foreflipper (B10) was rapid during the early postnatal period and continued to increase until at least 13 y (Table 2 and 3). A significant growth increment was evident between 4 and 5 y (two sample t test: p-value = 0.015; df= 8). Overall growth scaled with negative slope relative to SBL (b = 0.89) and age (b = 0.07). Growth increased by 129% at 10 y RTP (Table 3). Growth increment decreased with increasing SBL until about 6 y (c. 23% of SBL) (Table 2). The mean length of flipper (B10) of all males > 10 y, including unaged animals > 200 cm was 47.2 + 1.9 Figure 2a, b (right): Bivariate plot of log circumference of head at canine (cm) on: (a) log SBL length of seal (cm) and (b) age (y). PEM animals, open squares; MSM animals, closed triangles. Figure 3a, b (right: Bivariate plot of log tip of snout to centre of ear (cm) on: (a) log length of seal (cm) and (b) age (y). PEM animals, open squares; MCM animals, closed triangles. Figure 4a, b (right): Bivariate plot of log tip of snout to anterior insertion of the foreflipper (cm) on: (a) log length of seal (cm) and (b) age (y). PEM animals, open squares; MCM animals, closed triangles. 230 Proc. Linn. Soc. N.S.W., 130, 2009 C.L. STEWARDSON, T. PRVAN, M.A. MEYER AND R.J. RITCHIE Ee sce £ L=)) c 2 > 2) (s) a | wes | oF vv ov Or gE gE ve oF vy ov OF ge 9 ve (w9) Jaddiyjes0) 0} jnous U4 (w3) seddijjas0j 0} jNOUS UT a4 <0 E < om OO (=) 2 omoo o L > O «e405 o L 8 O fa <@« OO = = aa <4 F <4 <4 | a4 <4 r EO) oO Ci co) fe) Re) c ae ov ge gE ve ee ge 9€ ve ce (ud) efe ye peay Jo sduaJaywiNdJIO U7 (wis) aXe ye peasy Jo sousia}WINIIIO U7 Proc. Linn. Soc. N.S.W., 130, 2009 12 10 Age (y) Age (y) Age (y) 23k BODY MEASUREMENTS OF SOUTH AFRICAN FUR SEALS Ln length of hind flipper (cm) 4.2 44 46 Ln body length (cm) 48 5.0 5.2 Ln length of hind flipper (cm) Age (y) Figure 5a, b: Bivariate plot of log length of hind flipper (cm) on: (a) log length of seal (cm) and (b) age (y). PEM animals, open squares; MCM animals, closed triangles. cm (v = 8). The maximum recorded value for B10 was 55.0 cm (animal PEM1560, SBL 201 cm). Length of hind flipper (B12) Growth of length of hind flipper (B12) was rapid during the early postnatal period and continued to increase until at least 8—9 y (Table 2 and 3). Overall growth scaled with negative slope relative to SBL (b= 0.81) and expressed negative allometry relative to age (Figures 5a, b; Appendix 3, 4), increasing by 103% at 10 y RTP (Table 3). Growth increment decreased with increasing SBL until about 4 y (c. 15% of SBL) (Table 2). The mean B12 of all males > 10 y, including unaged animals > 200 cm was 28.7 + 0.9 cm (n = 7). The maximum recorded value was 32.0 cm (animal PEM1890, SBL 192 cm, = 12 y). Body length as an indicator of age In animals 1-10 y, growth in SBL was highly positively correlated with age (y) (r = 0.96, n = 56) (Appendix 4). After fitting the (robust) straight line model of age on standard body length, graphical displays of residuals and fitted values were examined, and the straight line model was found to be adequate. Thus, the following equation can be used as a ‘rough indicator’ of absolute age for animals 1—10 y. Age = - 6.54 + 0.0087 x SBL, n = 56 The coefficient of variation (100 x s/X) in SBL for young males I—5 y (17.2%) was considerably higher than 1n older males (8-10 y, 6.9%; => 12 y, 5.3%). ps3) Body length as an indicator of age group Linear discriminant analysis was used to classify seals of unknown age into one of the four age groups (pup, yearling, subadult, adult) based on body length. Performing linear discriminant analysis using the body length data where the age group is known we get the following four linear discriminant functions of the form y = mx+b: VA = OO =12 y (known to be at least 12 y but could not be aged more definitively due to the limitations of the dentition aging method). Percentage of animals correctly classified into age group is given in brackets. The overall percentage correctly classified: (3+7+23+21)/70 x 100% = 77.14%. Pups: All 3 pups have been correctly classified. Yearlings or juveniles: 1 yearling was incorrectly classified as a pup and the rest of the juveniles (7) have been correctly classified. Subadults: 9 subadults were classified as yearlings, 23 subadults were correctly classified and 5 subadults were classified as adults. Adults: one (1) adult was incorrectly classified as being subadult and the rest (n = 21) were correctly classified. True Group Juvenile or Predicted Group Ea Yearling(7 mo to | Salt art geese aie 35) Total mo) 610) to 7 y 6 mo) 7 mo) — 3 (100%) 1 (12.5%) 0 Juvenile or F wens 0 7 (87.5%) 9 (24.3%) 0 16 Subadult 0 0 23 (62.2%) 1 (4.5%) 24 Adult 0 0 5 (13.5%) 21 (95.5%) Total 3 37 22 classified as a subadult. The overall percentage correctly classified was calculated from adding up all the correctly classified animals and then dividing by the total number of animals multiplied by 100%; that is, (3+7+23+21)/70 x 100% = 77.14%. Body length is therefore useful in discriminating between different age groups but some groups such as yearlings and subadults can be difficult to correctly classify. Curvilinear Body length as an indicator of SBL Curvilinear body length (CBL) was found to be approximately 10.0 cm longer than SBL (SBL: 146.7 + 5.6; CBL: 157.1 + 6.2, nm = 50 using paired samples only). However, CBL was greatly influenced by the quantity of food in the stomach and by the degree of post-mortem bloating. For example, CBL was 20-25 cm longer than SBL in 5 animals that had been dead for several days, or had consumed large quantities of fish; therefore, CBL was not considered to be a useful substitute for SBL. Growth Curve Models Figure 6 shows a non-linear least squares fit of the Logistic model (Equation 3) to SBL vs. age for male Proc. Linn. Soc. N.S.W., 130, 2009 South African fur seals compared to curve fits on data from a previously published study on the Australian fur seal (Arnould and Warneke, 2002). Non-linear fits were also made using the exponential saturation + constant or von Bertalanffy model (Equation 2), and the Gompertz or double exponential equation (Equation 4). Table 5 shows the statistics of the curve fits. The correlations for all three models are very high (r > 0.94). Tests for significant differences in the fitted parameters were done using t-tests assuming equal variances or assuming unequal variances as appropriate (Cochran, 1977). The Australian fur seal data fits to the von Bertalanffy model quite well (Table 5), however the model does not appear to be suitable for the South African fur seal data. The fit for the South African fur seal data gives a fitted curve that is very close to linear and gives an unrealistically high estimate of the asymptotic SBL of over 270 cm. South African fur seals have a lower apparent pup size and exponential constant and the growth rate is lower than for the Australian fur seals (Table 5). The fits using the logistic (Equation 3) and the Gompertz (Equation 4) equations are more similar to 233 BODY MEASUREMENTS OF SOUTH AFRICAN FUR SEALS 250 200 150 Standard Body Length (SBL) cm 50 SBL data for Australian fur seals SBL of South African fur seals 0 5 10 15 Age (y) Figure 6: Growth kinetics of male South African fur seals (Arctocephalus pusillus pusillus) (closed circles) compared to male Australian fur seals (A. pusillus doriferus) (open squares). Age in y and SBL in cm. Curves fitted using the logistic model (Equation 3). each other than those made using the von Bertalanffy model (Equation 2) and both models give more realistic estimates of asymptotic maximum size for both the Australian and South African fur seals. Table 5 shows that the significant differences in the model parameters between the South African and Australian fur seals are the apparent pup size (P) and the exponential constant (k). The significant differences in (k) values reflect a slower growth rate from a lower initial (P) in South African fur seals. The asymptotic maximum SBL is about 214-232 cm based on the logistic and Gompertz models. These two models agree that the asymptotic maximum SBL is not significantly different in South African and Australian fur seals. Overall, the logistic curve (Figure 6, Equation 3) seems to be the most satisfactory growth model, based upon the high correlation coefficients of least-squares fits to the data and the lowest relative errors of the fitted parameters. 234 ——e« Predicted SBL for Australian fur seals DISCUSSION Age determination Dentition-age estimates of the South African fur seals were considered to be reliable, with inconsistencies among _ readings mitigated by repeated estimates, following a set protocol of procedures and double-blind tests (Payne, 1978, 1979; Doubleday and Bowen, 1980; Arnbom et al., 1992; McCann, 1993: Oosthuizen, 1997; Oosthuizen and Bester, 1997; Arnould and Warmeke, 2002). Nevertheless, the limitations of dentition-based estimates of the ages of seals are apparent, particularly for old animals where the pulp-cavity has filled and they can no longer be aged by growth rings. There is a need for more life-history and morphometric data based on animals tagged as pups. Predicted SBL for South African fur seals Body size Arctocephalus pusillus is the largest of the fur seals with the South 20 African subspecies (4. pusillus pusillus) tending to be slightly smaller than the Australian subspecies (A. pusillus doriferus) (Stewardson et al., 2008). Comparison of growth curves for the two populations show that the Australian fur seal grows at a faster rate than the South African variety but asymptotic maximum sizes are very similar (Table 5). Male SBL ranged from 66 to 243 cm. The largest animal in the collection (PEM952) was measured in 1980 at Kings Beach, Port Elizabeth, by V. Cockcroft and A. Bachelor. This is of similar length to an unusually large male (SBL 241 cm) measured by Rand in 1946 (Rand, 1949). The largest animal measured by the first author was 203 cm in 1994 (PEM2201). The largest individual in the data set used by Arnould and Warneke (2002) for the Australian fur seal was a 15 y old bull 224 cm long (close to the asymptotic maximum sizes estimated using the Logistic and Gompertz models for the Australian fur seal). The slightly larger mean size reached by male Australian vs. the South African populations of Arctocephalus pusillus may or may not be genetically based. Stewardson et al. (2008) pointed out that the present South African population has largely recovered to pre-exploitation levels, Proc. Linn. Soc. N.S.W., 130, 2009 C.L. STEWARDSON, T. PRVAN, M.A. MEYER AND R.J. RITCHIE Table 5: Growth Kinetics Models of Male South African fur seals (Arctocephalus pusillus pusillus) com- pared to Male Australian fur seals (A. pusillus doriferus). The SBL vs. age data were fitted to the exponential saturation model (Equation 2), the Logistic model (Equation 3, Figure 6) and the double exponential of Gompertz model (Equation 4). Student’s t-tests were performed to test if the fitted parameters for the South African and Australian fur seals were significantly different (Cochran, 1977). Preliminary F-tests showed that in most cases the vari- ances could be assumed to be equal (p > 0.05). This assumption could not be accepted in the cases of the asymptotic maximum sizes and incremental growths determined using the exponential saturation model. The t-test for the case of unequal variances was used for comparing the asymptotic maximum size and incremental growth of the two varieties of fur seal estimated using the exponential saturation model. Growth curve fits for the South African fur seal are based on 73 animals of definitive age (denti- tion-aged animals with an undefined age >12 y are excluded). The data for the Australian seals (n = 69) were redigitized from Arnould and Warneke (2002). 5a. Exponential saturation growth model seeea! Maximum SBL (cm) Pup Size at Birth | Lifetime Incremental Growth (cm) Population Australian fur seal n= 69, r= 0.9423 -0.1703 139.4 + 0.0223 ay OD + 8.54 South African fur seal n= 73, r=0.9524 -0.0799 VAM 203.5 + 0.0197 ge 3.117) 32 3.55 P<0.001 Significance P<0.001 P= 0.0678 n.s. P=0.1928 ns. 5b. Logistic growth model Maximum SBL Growth k Pup Size at Birth | Lifetime Incremental (cm) (y') (cm) Growth (cm) Population Australian fur seal n= 69, r= 0.9443 South African fur 139.6 seal ee + 9.697 + Q. n= 73, r= 0.9495 a Significance jon (OL OATS) tess p < 0.001 p= 0.1796 ns. 5c. Double exponential or Gompertz growth model -0.3023 124.2 + 0.0268 ae) // IIS) Maximum SBL (cm) Growth k Pup Size at Birth | Lifetime Incremental (y") (cm) Growth (cm) Population Australian fur seal n=69, r= 0.9437 130.2 + 6.569 224.2 -0.2352 94.04 9-138 + 0.0242 a3 uliO7 South African fur seal n= 73,r=0.9528 158.3 + 14.99 p = 0.0954 n.s. Proc. Linn. Soc. N.S.W., 130, 2009 DBS Misys} -0.1599 74.01 + 14.48 + 0.0207 + 3.844 Significance P= VOLTS nes: p= 0.019 p < 0.001 BODY MEASUREMENTS OF SOUTH AFRICAN FUR SEALS whereas the Australian population is still rapidly increasing and have not yet reached a steady population. At birth, male South African fur seals are about 35% (c. 70-80 cm) of their mean adult size which is about 197 cm based upon the mean adult size, SBL for animals >10 y including unaged animals > 200 cm. At puberty they are about 57% (c. 112.8 cm at 3 y) of their mean adult size. Although axillary girth varies with body condition, it is usually about 57-67% of SBL. The foreflippers are relatively long measuring 25-26% (c. 18 cm) of SBL in pups, and 24% (c. 48 cm) of SBL in adults. The hind flippers are considerably shorter measuring 19% (c. 13 cm) of SBL in pups, and 14.5% (c. 29 cm) of SBL in adults. Body shape Male South African fur seals are exceptional swimmers and divers, and haul out on land to rest, moult and breed. Body shape and general physiology have been modified to accommodate the demands of both marine and terrestrial environments (Bryden, 1972). For example, bulls spend most of their life at sea, hauling out to moult (predominantly February and March), rest, and reproduce (establish territories and breed from late October to late December/early January). The body is streamlined with a rounded head and a relatively short snout; small external ear pinnae (narrow and pointed); a small tail positioned between the hind flippers; a retractable penis that can be withdrawn into a cutaneous pouch; and modified fore/hind limbs (flippers). The strong fore limbs have been modified into elongated flippers for propulsion through the water (forceful strokes towards the body) and terrestrial locomotion (palm extends laterally with the flipper bending between the two rows of carpal bones). Characteristic features include predigital cartilage, a long first digit, reduced fifth digit, rudimentary nails and hairless palms. Unlike the foreflippers, which are the primary appendage used for propulsion through the water, the smaller hind flippers have been modified for terrestrial locomotion (soles extend laterally with the flipper bending forward at the ankle). Characteristic features include predigital cartilage; long grooming claws on digits 2-4; enlargement of digits one and five; and hairless soles. Function and growth Overall growth in SBL was similar to that of other highly polygynous male otariids including Arctocephalus gazella and Callorhinus ursinus, with 236 rapid early postnatal growth; a sudden increase in body size at puberty; and a reduced rate of growth soon after attainment of social maturity (McLaren, 198) South African fur seals pups are born on land between October and late December (Rand, 1956; Rand, 1967; Shaughnessy and Best, 1975). Newborn pups are 70-80 cm long at birth (c. 35% of mean adult length), which agrees with the apparent pup size estimates from the present study shown in Table 5. In November (when the majority of pups are born), mean length and weight is about 76 cm and 5.986 kg for males, and 73 cm and 5.487 kg for females (Rand, 1956). By April, mean length and weight is about 82.0 cm and 19.183 kg for males, and 84 cm and 15.147 kg for females (Rand, 1956). Table 5 shows that the estimated pup size at birth derived from the exponential and logistic growth curve models are not significantly different from the actual measurements given by Rand (1956). When juveniles gain their permanent teeth (June) they disperse to deeper water for short periods, supplementing their milk diet with solids (Rand, 1956). During this period they learn foraging skills while accompanying their lactating mothers to sea. Most animals feed independently at 9-11 months (Rand, 1956). There is a decline in body weight soon after weaning (Rand, 1956). Most males attain puberty between 3-4 y, as evidenced by the presence of sperm in the epididymis of some animals at 2 y 10 months (Stewardson et al., 1998). The onset of puberty (2—3 y) is associated with a sudden increase in body size (present study). It is thought that puberty is attained when seals reach a certain threshold size in body weight, with slower- growing animals reaching puberty later than faster- growing animals (Laws and Sinha, 1993). Although pubertal males produce sperm, they do not have the ability to acquire and maintain a harem (Stewardson et al., 1998). Small body size and inexperience prevents young males from gaining the high social status required for a breeding male. Growth in SBL continues to increase steadily until about 6 y. In animals = 7 y, growth continues to increase but at a slower rate (Tables 5 and 6, Figure 6). Social maturity is attained at about 9-10 y and appears to be associated with a weak secondary growth spurt in body size (present study). At this age, large body size has a direct advantage in competitive interactions with rival males, including intimidatory display without actual fighting, and an indirect effect through the presence of large stores of fat which enable large males to remain on territory for up to 40 days (Rand, 1967; Wartzok, 1991). Successful Proc. Linn. Soc. N.S.W., 130, 2009 C.L. STEWARDSON, T. PRVAN, M.A. MEYER AND R.J. RITCHIE bulls may hold harems multiple times over a two to three year period but are likely to die before reaching reproductive senescence (see Stewardson et al., 1998). Growth in body size slows soon after attainment of social maturity (present study). Growth of length of the foreflippers continued to increase until at least 13 y, with a significant increase in length at 4-5 y (present study). This increase may partially reflect changes in swimming and/or diving behaviour, with older animals presumably diving to deeper depths in search of prey. Growth of the smaller hind flippers slowed much earlier (8—9 y) than growth of the foreflippers (as is also found in the case of the Australian fur seal; Arnould and Warneke, 2002). The maximum sizes of the fore and hind flippers found in the present study for the South African fur seal are similar to the asymptotic sizes found in the fore and hind flippers of the Australian fur seal (Arnould and Warneke, 2002). No special development of the foreflippers or hind flippers associated with locomotion was reported in Arctocephalus gazella, i.e., a more or less constant rate of growth from age one to 7 (Payne, 1979). Body length as an indicator of age SBL could not be used reliably to assign a seal toa particular age because there was considerable overlap between year classes, especially among middle-aged animals. Similar findings have been reported in other species of pinnipeds (e.g., Laws, 1953; Bryden, 1972; Bengston and Sniff, 1981). However, SBL was found to be a ‘rough indicator’ of age for animals 1—10 y, and of age group (Table 4). The curvilinear models (von Bertalanffy, logistic and Gompertz models) shown in Table 5 could also be used to estimate age from SBL but inspection of Figure 6 clearly shows that they would not be reliable for estimating the age of animals greater than about 10 y. In male South African fur seals, postnatal growth is rapid with a significant growth spurt at the onset of puberty (2-3 y) and a weak growth spurt at social maturity (9-10 y). Body size continues to increase but at a slower rate between 6 and 7 y, and then growth slows soon after the attainment of social maturity. Growth was a differential process and not simply an enlargement of overall size. Relative to SBL, facial variables and the fore/hind limbs scaled with negative slope relative to SBL or were negatively allometric; tip of snout to genital opening scaled with positive slope; and tip of snout to anterior insertion of the foreflipper was positively allometric. Relative to age, body variables scaled with negative slope or were negatively allometric. SBL was found to be a ‘rough indicator’ of age and of age group. Proc. Linn. Soc. N.S.W., 130, 2009 Model Growth Curves for Male South African and Australian fur seals Further information is needed on older animals of known-age in order to more accurately estimate asymptotic maximum size (see Figure 6 and Table 5). In the present study, low sample size at the intermediate ages, and the absence of very old animals of known- age (15-20 y), made it difficult to determine a more exact shape of the growth curve. Published growth curves are also available on the male Australian fur seal (Arnould and Warneke, 2002, n = 69), male and female New Zealand fur seals (Dickie and Dawson 2003, males n = 64), male New Zealand fur seals (McKenzie et al., 2007, n= 86), subantarctic fur seals (Bester and Van Jaarsveld, 1994) and the male Steller sea lion (Winship et al., 2001, n = 203). The breeding/ non-breeding status of the animals in the present study was not known. Breeding bulls are thought to be larger in size than non-breeding bulls of the same age; therefore, the growth pattern of male fur seals may be more complex than implied by the models used in the present study. For example, the data of Arnould and Warneke (2002) is based on males shot at a breeding colony and so has many large males of breeding status. The males in the present study are mainly based on dead or dying animals found stranded on the coastline and incidental drownings from trawling. The study area of the present study was a seal feeding area rather than a breeding colony. Figure 6 and Table 5 show that the von Bertalanffy, Logistic and Gompertz models suggest that the kinetics of growth in SBL vs. age is different in the two subspecies. South African fur seals seem to have a smaller apparent pup size and a slower growth rate than the Australian fur seal once living independently. The skulls of male Australian fur seals are significantly larger than South African male seal skulls (Stewardson et al., 2008), however the often repeated statement that Australian fur seals are consistently larger in body size than the South African variety (Pemperton et al., 1993; Arnould and Warneke, 2002; Stewardson et al., 2008) is not supported by the values for the asymptotic maxima of the logistic and Gompertz curve fits shown in Table 5 and by inspection of Figure 6. However, for our South African material, the average SBL of all males > 10 y and/or SBL > 200 cm was 197 + 4.1 cm (n = 15) is significantly smaller than a similar calculation for Australian males (211 + 1.5 cm, n = 17) using the data of Arnould and Warneke (2002). This might more accurately reflect differences in the types of populations sampled in the study by Arnould and Warneke (2002) — a breeding colony, and in the present study — a feeding population probably with DB] BODY MEASUREMENTS OF SOUTH AFRICAN FUR SEALS many non-breeding males. In any case, more data on age-tagged old males is needed to better define the growth kinetics of the species. After weaning, the Australian variety seems to grow faster and reaches maturity earlier than the South African variety. For example, the data of Arnould and Warneke (2002) includes one individual that exceeded 200 cm when only 5 and a half years old. Such growth kinetics are consistent with what would be expected of a rapidly increasing population, not limited by natural resources, recovering from severe depletion. The Australian population has not yet reached pre-exploitation population size whereas the South African population is today close to pre- exploitation levels (Pemperton et al., 1993; Arnould and Warneke, 2002). Differences between the two varieties might therefore reflect differences between a well-fed expanding population and a population in steady-state with limited resources (Stewardson et al., 2008), rather than a genetic difference. Such a proposition would predict that as the Australian fur seal population approaches the carrying capacity of its niche, a reduction in the average pup size, growth rate of the pups and the growth rate of independent animals would be expected. Winship et al. (2001) working with a data set of 203 aged male Steller sea lions were also able to show that the Logistic and Gompertz models were better (in terms of sum of squares residuals and correlation r) than the von Bertalanffy model for describing the growth kinetics of seals. The logistic and Gompertz models are also a very good fit to the growth kinetics of male New Zealand fur seals (Dickie and Dawson 2003; McKenzie et al., 2007). The Logistic and Gompertz exponential constants found for the New Zealand fur seal are comparable to those found in the Australian fur seal due to a similar lifespan and similar relative sizes of the pup to the adult. A great deal of effort has been spent in discussing the relative merits of growth curves in biology but often the data sets are too small for this time and effort to be justifiable (Zeide, 1993). The von Bertalanffy model was not a satisfactory fit for the South African fur seal data because it gave an unrealistically high estimate of the asymptotic maximum size of 276 cm, which is well above the largest recorded SBL of 241 cm for a South African fur seal. The fitted equation shows very little curvature, due to a lack of accurately aged very old animals (Figure 6). Previous attempts to model the growth kinetics of seals have generally reached the conclusion that the von Bertalanffy model tends to give imprecise overestimates of the asymptotic maximum size (Zullinger et al., 1984; 238 Trites and Bigg, 1992; Bester and Van Jaarsveld, 1994; Winship et al., 2001; Arnould and Warneke, 2002; McKenzie et al., 2007). The logistic and Gompertz models are both more realistic than the von Bertalanffy model for post-natal mammalian growth because they both have a point of inflection: mammals grow exponentially while pups and juveniles, then linearly as a subadult and finally growth decreases asymptotically when reaching maturity. Both are very popular for modelling growth in mammals but Zullinger et al. (1984) points out that bothmodels tend to bothunderestimateandimprecisely estimate the maximum body size of mammals. Plots of the fit to the Gompertz model are almost identical to those made using the Logistic model but Table 5 and Figure 6 show that the asymptotic body size is not as precisely defined as in the case of the Logistic model (Equation 4). This is particularly the case for the South African fur seal data, due to the lack of very old animals in the data set. The growth models suggest there are significant differences in how the maximum size is achieved in the two populations; independent juvenile and adult Australian fur seals have a very significantly higher growth rate than in the case of South African fur seals. Thus the Australian fur seal achieves maximum size earlier than the South African fur seal and tends to grow larger, at least under current population densities. Caution is needed in interpreting changes in growth kinetics of seals over time or differences between different species or populations. The population history of the Northern fur seal is similar to that of most fur seals: extreme depletion by the beginning of the 20" century, followed by recovery but for reasons that are not clear-cut population numbers and growth kinetics have varied considerably since their initial recovery in about 1940. Trites and Bigg (1992, 1996) have found that the growth kinetics of the Northern fur seal (Callorhinus ursinus) has varied over time on the Pribilof Islands off Alaska but were cautious about attributing it to changes in population pressure on resources or other environmental effects. Trites and Bigg (1992) state that higher growth rates and body size seem to correlate with lower total densities of animals but migration effects between different colonies could be a complicating factor. Conclusion The classification criteria for age and age group developed in this study will be particularly useful when canines are not available for age determination, e.g. behavioural studies, census counts and where animals are drugged for mark/recapture studies. Proc. Linn. Soc. N.S.W., 130, 2009 C.L. STEWARDSON, T. PRVAN, M.A. MEYER AND R.J. RITCHIE Removing postcanines for aging live animals as done by Payne (1978, 1979) might not be possible under some jurisdictions. Tagging of live animals, should be encouraged wherever possible, because of the lack of data on development and longevity of most species of fur seal impacts on the development of rational management policies. Information presented in this study contributes to earlier descriptions of the South African fur seal (Rand, 1956), and provides new information on body growth according to age (y) that 1s useful for comparisons with the Australian fur seal. ACKNOWLEDGEMENTS We wish to express our sincere appreciation to the following persons and organizations for assistance with this study: Dr V. Cockcroft (Port Elizabeth Museum), Dr J. Hanks (WWE-South Africa) and Prof. A. Cockburn (Australian National University) for financial and logistic support; Mr B. Rose (Oosterlig Visserye, Port Elizabeth) who enabled us to collect seals from his commercial fishing vessels; Dr J.H.M. David (Marine Coastal Management, Cape Town) for use of measurements from known-age animals; Mr H. Oosthuizen (MCM) for assistance with age determination; Mr S. Swanson (MCM) for assistance with data extraction of MCM specimens; Mr N. Minch (Australian National University) for photographic editing; Dr C. Groves for constructive comments on an earlier draft of this manuscript. This paper was compiled on behalf of the World Wild Fund for Nature - South Africa (project ZA - 348, part 1a). The South African fur seal data presented in the paper is abstracted from a PhD thesis presented to the Australian National University by C. L Stewardson in 2001. REFERENCES Anderson, T.W. (1984). An Introduction to multivariate statistical analysis, 2nd ed. (John Wiley and Sons Publ., New York). Arnbom, T.A., Lunn, N.J., Boyd, I.L. and Barton, T. (1992). Aging live Antarctic fur seals and southern elephant seals. Marine Mammal Science 8, 37 — 43. Amould, J.P.Y. and Warneke, R.M. (2002). Growth and condition in Australian fur seals (Arctocephalus pusillus doriferus) (Carnivora: Pinnipedia). Australian Journal of Zoology 50, 53-66. Bengston, J.L. and Sniff, D.B. (1981). Reproductive aspects of female crabeater seals (Lobodon carinophagus) along the Antarctic Peninsula. Canadian Journal of Zoology 59, 92-102. Bester M.N. and Van Jaarsveld A.S. (1994). Sex specific and latitudinal variance in postnatal growth of the Proc. Linn. Soc. N.S.W., 130, 2009 subantarctic fur seal (Arctocephalus tropicalis) Canadian Journal of Zoology 72, 1126-1133. Bigg, M.A. (1979). Preliminary comments on body length in female northern fur seal. In ‘Preliminary Analysis of Pelagic Fur Seal Data Collected by the United States and Canada during 1858-1974’. Report to 22nd Annual Meeting of the North Pacific Fur Seal Commission, pp. 171-179. (National Marine Fisheries Service, NOAA, Seattle, Washington). Bychkoy, V.A. (1971). Age variation of body length and weight of fur seals of Seal Island. Ekologiva 1971 (1), 101-105. [Engl. transl. Soviet Journal of Ecology 2, 81-83.] Bryden, M.M. (1972). Growth and development of marine mammals. In “Functional anatomy of marine mammals’, vol. 1, (Ed. Harrison, R.J.), pp. 2-79. (Academic Press Publ., London). Calkins, D. and Pitcher, K.W. (1983). Population assessment, ecology and trophic relationships of Steller sea lions in the Gulf of Alaska. In ‘Environmental Assessment Program’. U.S. Department of Commerce, NOAA, 19, pp. 445-546. Cochran, W.G. (1977). Sampling techniques, 3rd ed. (John Wiley and Sons Publ., New York). Committee on Marine Mammals (1967). Standard variables of seals. Journal of Mammalogy 48, 459-462. Dickie G.S. and Dawson S.M. (2003). Age and reproduction in New Zealand fur seals Marine Mammal Science 19, 173 — 185. Doubleday, W.G. and Bowen, W.D. (1980). Inconsistencies in reading the ages of harp seal (Pagophilus groenlandicus) teeth, their consequences, and a means of resulting biases. North Atlantic Fisheries Organization, Scientific Council Research Document 80/X1I/160, Serial No. N247. Fiscus, C.H. (1961). Growth in the Steller sea lion. Journal of Mammalogy 42, 218-223. Gibbons, J.D. and Chakraborti, S. (1992). Nonparametric statistical inference, 3rd ed., (Marcel Dekker Inc Publ., New York). Johnson, M.L. and Faunt, L.M. (1992). Parameter estimation by least squares methods. Methods in Enzymology 210, 1-37. Krylov, V.I. and Popov, L.A. (1980). On comparative morphology of Antarctic seals of King George Island. In ‘Morskie Mlekopitayushchie,’ pp. 40-51. Moscow: VNIRO [In Russian. ] Lander, R.H. (1979). Fur seal growth. In “Preliminary analysis of pelagic fur seal data collected by the United States and Canada during 1958-1974. Report to 22nd Annual Meeting North Pacific Fur seal Commission, pp. 143-170. Seattle, Washington: National Marine Fisheries Service, NOAA. Laws, R.M. (1953). The elephant seal (Mirounga leonina Linn.). 1. Growth and age. Falkland Islands Dependencies Survey Scientific Reports 8, 1-62. Laws, R.M. and Sinha, A.A. (1993). Reproduction. In ‘Antarctic seals research methods and techniques’ IB9 BODY MEASUREMENTS OF SOUTH AFRICAN FUR SEALS (Ed. Laws, R.M.), pp. 228-267. Cambridge University Press Publ., London). Loughlin, T.R. and Nelson, R. (1986). Incidental mortality of northern sea lions (Eumetopias jubatus) in Shelikof Strait, Alaska. Marine Mammal Science 2, 14-33. McCann, T.S. (1993). Age determination. In “Antarctic seals, research methods and techniques’ (Ed. Laws, R.M.), pp. 199-227. (Cambridge University Press Publ., London). McKenzie, J., Page, B., Goldsworthy, S.D. and Hindell, M.A. (2007). Growth strategies of New Zealand fur seals in southern Australia. Journal of Zoology 272, 377 — 389. McLaren, I.A. (1993). Growth in pinnipeds. Biological Review 79, 1-79. Oosthuizen, W.H. (1997). Evaluation of an effective method to estimate age of Cape fur seals using ground tooth sections. Marine Mammal Science 13, 683-693. Oosthuizen, W.H. and Bester, M.N. (1997). Comparison of age determination techniques for known-age Cape fur seals. South African Journal of Zoology 32, 106-111. Payne, M.R. (1978). Population size and age determination in the Antarctic Fur seal Arctocephalus gazella. Mammal Review 8, 67 — 73. Payne, M.R. (1979). Growth in the Antarctic fur seal Arctocephalus gazella. Journal of Zoology (London) 187, 1-20. Pemperton D., Kirkwood, R, Gales R. and Renouf, D. (1993). Size and shape of male Australian fur seals, Arctocephalus pusillus doriferus. Marine Mammal Science 9, 99 — 103. Rand, R.W. (1949). Studies on the Cape fur seal Arctocephalus pusillus pusillus 3. Age grouping in the male. Progress report submitted November 1949, Government Guano Islands Administration, Department of Agriculture, Union of South Africa. Rand, R.W. (1956). The Cape fur seal Arctocephalus pusillus pusillus (Schreber): its general characteristics and moult. Sea Fisheries Research Institute Investigational Report, South Africa 21, 1-52. Rand, R.W. (1967). The Cape fur seal Arctocephalus pusillus pusillus 3. General behaviour on land and at sea. Sea Fisheries Research Institute Investigational Report, South Africa 60, 1-39. Rosas, F.C., Haimovici, M. and Pinedo, M.C. (1993). Age and growth of the South American sea lion, Ofaria filavescens (Shaw, 1800), in Southern Brazil. Journal of Mammalogy 74(1), 141-147. Scheffer, V.B. and Wilke, F. (1953). Relative growth in the northern fur seal. Growth 17, 129-145. Shaughnessy, P.D. and Best, P.B. (1975). The pupping season of the Cape fur seal, Arctocephalus pusillus pusillus. Sea Fisheries Branch South Africa, pp. 1-8, 1975. (Unpublished internal report) Stewardson, C.L. (2001). “Biology and conservation of the Cape (South African) fur seal Arctocephalus pusillus pusillus (Pinnipedia: Otariidae) from the Eastern Cape Coast of South Africa’, Thesis (Ph.D.), 240 Australian National University. http://thesis.anu.edu. au/public/adt-ANU20030124.162757/index.html Stewardson, C.L., Bester, M.N. and Oosthuizen, W.H. (1998). Reproduction in the male Cape fur seal Arctocephalus pusillus pusillus: age at puberty and annual cycle of the testis. Journal of Zoology (London) 246, 63-74. Stewardson, C.L., Prvan, T., Meyer, M.A. and Ritchie, R.J. (2008). Age determination and growth in the male South African Fur Seal Arctocephalus pusillus pusillus (Pinnipedia: Otariidae) based upon skull material. Proceedings of the Linnean Society of NSW 129, 207-252. Thorsteinson, F.V. and Lensink, C.L. (1962). Biological observations of Steller sea lions taken during an experimental harvest. Journal of Wildlife Management 26, 353-359. Trites, A.W. and Bigg, M.A. (1992). Changes in body growth of northern fur seals from 1958 to 1974: density effects or changes in the ecosystem? Fisheries Oceanography 1, 127-136. Trites, A.W. and Bigg, M.A. (1996). Physical growth of northern fur seals (Callorhinus ursinus): seasonal fluctuations and migratory influences. Journal of Zoology (London) 238, 459-482. Wartzok, D. (1991). Physiology of behaviour in pinnipeds. In “Behaviour of pinnipeds’ (Ed. Renouf, D.), pp. 236-299. (Chapman and Hall Publ., London). Weisberg, S. (1985). Applied linear regression, 2nd ed., (John Wiley and Sons Publ., New York). Wickens, P.A. (1993). Life expectancy of fur seals with special reference to the South African (Cape) fur seal. South African Journal of Wildlife Research 23, 101 — 106. Winship, A.J., Trites, A.W. and Calkins, D.G. (2001). Growth in body size of the Steller sea lion (Eumetopias jubatus). Journal of Mammalogy 82, 500-519. Wynen, L.P., Goldsworthy, S.D., Insley, S.J., Adams, M., Bickham, J.W., Francis, J., Gallo, J.P., Hoelzel, A.R., Mailuf, P., White, R.W.G. and Slade, R. (2001). Phylogenetic Relationships within the Eared Seals (Otariidae: Carnivora): Implications for the Historical Biogeography of the Family. Molecular Phylogenetics and Evolution 21, 270-284. Zeide, B. (1993). Analysis of growth equations. Forest Science 39, 594-616. Zullinger, E.M., Ricklefs, R.E., Redford K.H and Mace G.M. (1984). Fitting sigmoidal equations to mammalian growth curves. Journal of Mammalogy 65, 607-636. Proc. Linn. Soc. N.S.W., 130, 2009 APPENDIX 1 South African fur seals (n = 149) examined in this study. Animals were collected from the coast of southern Africa between August 1978 and September 1997. Accession Numbers of Specimens used in the Present Study. Port Elizabeth Museum (PEM), Port Elizabeth, South Africa PEM603 PEM676 PEM877 PEM928 PEM1159 PEM1698 PEM1895 PEM2007 PEM2020 PEM2049 PEM2087 PEM2186 PEM2203 PEM2257 PEM2401 PEM2414 PEM605 PEM824 PEM886 PEM951 PEM1453 PEM1706 PEM1999 PEM2008 PEM2021 PEM2051 PEM2131 PEM2188 PEM2238 PEM2257 PEM2403 PEM2415 PEM607 PEM828 PEM888 PEM952a PEM1507 PEM1879 PEM2000 PEM2009 PEM2036 PEM2052 PEM2132 PEM2191 PEM2248 PEM2348 PEM2404 PEM2454 PEM608 PEM834 PEM889 PEM958 PEM1560 PEM1882 PEM2002 PEM2010 PEM2045 PEM2053 PEM2137 PEM2194 PEM2252 PEM2359 PEM2405 PEM2455 PEM658 PEM852 PEM898 PEM975 PEM1587 PEM1885 PEM2003 PEM2013 PEM2046 PEM2054 PEM2140 PEM2197 PEM2253 PEM2374 PEM2406 PEM2458 PEM661 PEM874 PEM916 PEM 1073 PEM1696 PEM1890 PEM2004 PEM2014 PEM2047 PEM2081 PEM2141 PEM2198 PEM2254 PEM2379 PEM2409 C.L. STEWARDSON, T. PRVAN, M.A. MEYER AND R.J. RITCHIE PEM670 PEM875 PEM917 PEM1135 PEM1697 PEM1892 PEM2006 PEM2015 PEM2048 PEM2082 PEM2143 PEM2201 PEM2256 PEM2400 PEM2411 Marine and Coastal Management (MCM), Dept of Environment Affairs and Tourism, Cape Town, South Africa MCM1565 MCM3586 MCM4577 MCM4989 MCMS5001 MCMS5135 Proc. Linn. Soc. N.S.W., 130, 2009 MCM1786 MCM3587 MCM4584 MCM4991 MCMS5002 MCMS5136 MCM2763 PEM3589 MCM4585 MCM4992 MCMS5005 MCMS5142 MCM2795 MCM3636 MCM4595 MCM4996 MCMS021 MCMS5145 MCM3017 MCM4023 MCM4597 MCM4998 MCMS5022 MCM3125 MCM4365 MCM4985 MCM4999 MCM5133 MCM3582 MCM4388 MCM4987 MCM5000 MCMS5 134 241 BODY MEASUREMENTS OF SOUTH AFRICAN FUR SEALS APPENDIX 2 Spearman rank-order correlation coefficients for log body variables of male South African Fur seals. Variables are as for Appendix 1. Pups were excluded from the analysis. p < 0.001 unless otherwise stated in square brackets. * Significant at 2% level (2-tailed). ** Significant at 1% (2-tailed). Sample size (n) in brackets. BI B2 B3 B4 BS B6 B7 BS B9 B10 Bil B12 Bl 1 082* 0.12 0.74* 0.63 0.77* 0.84% 0.76 0.71* 0.71* 0.82* 0.72* -99 “OR OO si -97 -94 aSF -54 -96 -98 -96 81 -93 -85 B2 0.82* 1 0.17 0.76* 0.61* 0.62* O.81* 0.78% 0.73% 0.74% 0.86 —0.72* “986 5102) [OAM Awe LOO -97 -90 -57 Oy 49)060 400+180 61:0 + 16°0- 76 (Za) soddry pury fo yysu0T VN VN VN (100>49)7%60 4004680 810 = 160- £6 (01a) Joddryor0; Jo ysusT Looo=4 €61<9:'H (100>49) £60 SOOFITT 770 F EE I- 66 (6) Joddiyor0y oy} Jo UONTOSUT JOLIO}UR 0} jnoUS Jo dry VN VN VN (100>%)660 100401 L0'0 + SE0- 16 (gq) sutuodo [ey1U08 0} nous Jo diy, VN WN VN (100>%8L0 4004690 7C'0 ¥ 780" 6 (Sq) odes Jo a[S3ue 0} ynous Jo diy, 1l00>d 16 1>49:'H (100>4) P80 €00FES0 p10 ¥0€0 €6 (pq) Iwo JO x99 0} nous Fo diy, — - —su (91°) =) S10 — - L8 (¢€q) 249 JO OUD 0} JnoUS Jo dry, loo>d 988 1>49:'H (100>4)780 10:0 +050 810 + 601 06 (Zq) 240 Je peoy JO soueJEFUIMIAT 10>4 Sg 1>4:H (100>DLL0 +100FLr0 810 + 680 L8 (Tq) ourueo ye peoy Jo soUorEFUIMIIID (d) Apiqeqorg ~P, Se CO) Same dOIS: els ao an ers] er g]qeliva quepuedaq AAJOWIOTLV UOISSIAGIA AVIUT'T (Z 9[QUI, UI So}OUO0OF 998) sISATeUR WOLF popnjoxe [][ gq pue Lg so]qeueA “o[qeordde you 4soq “arr “jour JOU 919M (q) aUT] dy} Jo odoys oy} jnoge sosoyjodAy jso} 0} posnber suonduinsse jopoul “WN “‘WLOYIUSIS JOU UOISSOIdOI ILOUT] “JOAQMOY “JOU OIOM E_ I[GUIIBA IOJ suondunsse Jopoyw ‘¢gq wos yede (poyie}-Z) [OAd] %] OU} Je JULOIUTIS ov SUOTLIOIIOD |] YW JUSIOYJoos UONRIEU09 Jop1o-yues uewtvedg “I (Q]] = U “ol “papso0s -91 JOU dJOM sa[eUl poSeun/pose ¢] Wo sT|gs pue ‘sishyeue wo popnyoxe o19M sdnd) poproser TqS YIM sjeultue poseun pue pose Oj si (U) ozIs o[dureg *s[voS IN| ULILAPY YING [vu Jo (WD) YSUaT Apog [ves So] uo (Wd) yUdWOANSvaW Apog So] 10 AjoWOT]e pue ‘s}Ud1DYJo09 UOHL[IA109 19p10-yuRA ULULIvads ‘suoyEenbea aur yy sI¥.48 sarenbs jsvoy .Jsnqoy, € XIGNHddV 243 Proc. Linn. Soc. N.S.W., 130, 2009 BODY MEASUREMENTS OF SOUTH AFRICAN FUR SEALS APPENDIX 4 ‘Robust’ least squares straight-line equations, Spearman rank-order correlation coefficients, and allometry for log body measurement (cm) on age (y) of male South African Fur seals. Sample size (n) is the number of skulls with body variable and age recorded (only animals 1—10 y were included in analysis, 1.e., n = 68). r, Spearman rank-order correlation coefficient. All correlations are significant at the 1% level (2-tailed), except for B3. Model assumptions were met for variable B3; however, linear regression not significant. NA, model assumptions required to test hypotheses about the slope of the line (b) were not met, 1.e., test not applicable. Variables B7 and B11 excluded form analysis (see footnotes in Table 2). Dependent variable (B) Circumference of head at canine (B1) Circumference of head at eye (B2) Tip of snout to centre of eye (B3) Tip of snout to centre of ear (B4) Tip of snout to angle of gape (B5) Standard body length (B6) Tip of snout to genital opening (B8) Tip of snout to anterior insertion of the foreflipper (B9) Length of foreflipper (B10) Length of hind flipper (B12) Total Sample size (n) 63 63 Sil 68 64 56 67 68 67 64 68 Intercept + S.E. -2.59 + 0.50 -2.63 + 0.43 2.67 + 0.02 2.03 + 0.03 4.45 + 0.02 -1.28+0.14 3.56 + 0.03 3.10 + 0.03 2.64 + 0.02 Linear regression Slope + S.E. 0.17 + 0.021 0.12+0.01 0.04 + 0.004 0.04 + 0.005 0.08 + 0.003 0.02 + 0.001 0.10 + 0.005 0.07 + 0.005 0.07 + 0.004 nO) Alternative P hypothesis 0.59 (p < 0.01) IH 2 [Bl 0.69 (p < 0.01) NA -0.008 (p = 0.95) ns 0.69 (p < 0.01) NA 0.56 (p < 0.01) 1s re Sa 0.96 (p < 0.01) NA 0.93 (p<0.01) NA 0.90(p<0.01) H,:B<1 0.82 (p<0.01) NA 0.93(p<0.01) H,:B<1 Allometry df 61 NA 66 NA 62 Probability (p) p<0.01 NA p< 0.01 NA p < 0.01 Proc. Linn. Soc. N.S.W., 130, 2009 244 Linnaeus: King of Natural History PauL ADAM! AND ELIZABETH May? ' School of Biological, Earth and Environmental Sciences, University of NSW, Kensington, NSW, 2052; * School of Biological Sciences, University of Sydney, NSW, 2006 Adam, P. and May, E. (2009). Linnaeus: king of natural history. Proceedings of the Linnean Society of New South Wales 130, 245-250. Linnaeus’ legacy was far more encompassing than taxonomic. We argue that, while the systematic recording of species remains fundamental to modern ecological concerns, Linnaeus also laid the foundation for other major areas of ecology, including comparative biogeography, plant demography, and comparative anatomy. Manuscript received 18 October 2008, accepted for publication 21 January 2009. KEYWORDS: Biodiversity, Linnaeus, natural history, natural philosophy, taxonomy. INTRODUCTION Linnaeus is one of the towering figures in the history of biological science. He is remembered today chiefly for his introduction of the binomial system of nomenclature and for his taxonomy. However, if we consider his skill as an observer of the relationships between the plants and animals he classified, and their environment, he should also be regarded as one of the earliest practicing ecologists. He passed on these skills to his many students, including those who travelled the world collecting and classifying organisms and became known as the ‘apostles’. In eighteenth-century Sweden, Linnaeus was accorded high status, and this great respect continued to hold internationally until the twentieth century. Lord Rutherford’s throw-away comment that ‘(a)ll science 1s either physics or stamp collecting’ reflected the marginalization of taxonomy and natural history within science as technological advances in physics, chemistry and engineering attracted funding and support. These advances also revolutionised biology, permitting breakthroughs in physiology, biochemistry and the molecular sciences, but in this brave new world the diversity and distribution of organisms lost their attraction as fields of study. It was only towards the end of the twentieth century when environmental issues became such a major theme in politics and with the public that there was renewed interest in the study of biodiversity, and a need to bring new techniques and approaches to ‘old fields’. By then, many of the essential skills underpinning the study of biodiversity were already in decline. (Biodiversity itself is a word of recent origin — first coming to the fore with the publication of Wilson [1988].) The taxonomic side of Linnaeus’ achievements was outstanding. While his sexual system did not long survive as a basis for plant classification, Linneaus had grasped the potential for classification to be predictive and ‘natural’, even if his particular approach had its limitations. He had recognized the importance of the hierarchical approach and provided a nomenclatural system that was functional and, importantly, had practical application to the large number of new species that were being discovered outside Europe. It was the first classification system that was accessible to the non-specialist, with the work encapsulated in handbooks that were ‘small enough to be carried into the field’ (Koerner, 1999, p40). In the year 2007, we celebrated the tercentenary of Linnaeus’ birth. It also happens to be the anniversary of Georges-Louis Leclerc, Comte de Buffon, the author of the Histoire Naturelle. Conniff (2007), in his article aptly titled Happy Birthday, Linnaeus, argues that Buffon should be regarded as at least the equal of Linnaeus, and in particular suggests that Buffon had a superior understanding of habitat, anticipating the development of ecology as a science. Linnaeus and Buffon were mutually fierce critics, and Buffon was undoubtedly also a major figure in the history of science: he had better geological insight than Linnaeus, and was closer to having an evolutionary perspective. However, to suggest that Linnaeus’ natural history LINNAEUS: KING OF NATURAL HISTORY was not ecologically focused is far from accurate. A major part of Linnaeus’ teaching was based on field excursions: he had a very comprehensive knowledge of the local flora of southern Sweden, and while his sexual system was not appropriate for higher taxonomic ranks, Linneaus’ species concepts have largely stood the test of time. We would argue that Linnaeus’ taxonomic work was firmly underpinned by a deep understanding of natural history and that natural history in turn provided the basis for ecology (Mayr, 1997; Blunt, 1971). The ecological insights of Linnaeus are clearly seen in his botanical ‘text book’, Philosophia Botanica (Linnaeus, 1751). Koerner (1999) notes that ‘...he described many of the mechanisms of species interdependence, as Charles Darwin noted on reading his Oeconomia naturae of 1749’ (p15). LINNAEUS AND NATURAL HISTORY ‘Natural history’, the advancement of which is the prime objective of the various Linnean Societies around the world, has a very long history. Early hunter- gatherer societies could not have survived unless members possessed what we might consider to be an innate understanding of natural history, including the ability to recognise different sorts of food and to distinguish between the edible, the toxic and the dangerous. Cave paintings provide, in a tangible form, evidence for knowledge of natural history. In classical times, plants and animals were seen as sources of medicines or as an element in a broader natural philosophy, and the Greeks and Romans left a documentary record which at least in part survives to this day and that would have been known to Linnaeus. Knowledge of natural history would have been current amongst the broader population, the majority of which lived in rural environments and were intimately dependent on the natural world for survival, but in ‘academic’ circles natural history was increasingly associated with medicine. For hundreds of years herbalists recycled the writings of classical authors, without making original observations and with the claims becoming more fanciful on each retelling. The Renaissance then brought a new curiosity about the world and more organized scientific inquiry, although the importance of the links to medicine continued, and old myths still retained currency. Linnaeus himself was Professor of Medicine and Botany at Uppsala University, and in some institutions close links between the two disciplines survived until the twentieth century. The recent growth of interest in alternative medicine suggests a need for revitalizing 246 the links to scientific botany, and ethnobotany has been given a new impetus as a field of study by the regime for rewarding traditional owners of knowledge and resources, which was established by the United Nations Convention on Biological Diversity 1992. From the late 16 century onwards there was a considerable interest in collecting and studying ‘curiosities’ of all kinds, and some of the collections of natural history objects that were assembled were large. Some of the more academic natural historians associated with these collections were distinguished scientists whose work has stood the test of time. An example is John Ray, who was the first to draw a distinction between Monocots and Dicots and who is commemorated in the still existing Ray Society and in the name of the herbarium at the University of Sydney. Some of the impetus for collecting was stimulated by the increasing numbers of exotic specimens being sent back to Europe from wider exploration. Linnaeus was very much part of this natural history tradition and although he did not travel beyond Europe, he actively encouraged his students to do so, and he was familiar with non-European plants both in the form of herbarium specimens and in gardens. The non- European species he described famously included bananas, which would then have been regarded as very exotic. Linnaeus’ own exploration was closer to home and included his early expedition to Lapland. Although there are suggestions that his account of his travels is somewhat exaggerated! (Koerner, 1999), it established his reputation as an explorer and natural historian. Lapland in the eighteenth century was at the edge of the world and for many Europeans would have been regarded with as much trepidation as Africa. Even today it remains one of the few wilderness areas in Europe (Ratcliffe 2006). Once he was established as a senior academic in Uppsala, field teaching became an essential and popular part of Linneaus’ teaching. His excursions attracted large numbers of students and were organized with almost military precision (Blunt 1971). Most attention was paid to the flora, although any matter of natural history interest was open for study and comment. Many of the localities around Uppsala that were visited on excursions still support the same species today, so that it requires no great stretch of the imagination to visit sites today and see what Linneaus’ students would have seen, and to experience the same excitement of first encountering a wet meadow full of snakeshead fritillaries (Fritillaria meleagris) or a dry calcareous esker with a spring abundance of Pasque flowers (Anemone pulsatilla). Proc. Linn. Soc. N.S.W., 130, 2009 P. ADAM AND E. MAY LINNAEUS AND ECOLOGY William Stearn (in Appendix I of Blunt, 1971) recognised that Linnaeus has been variously declared ‘a pioneer ecologist, a pioneer plant-geographer, a pioneer dendrochronologist, a pioneer evolutionist...’ but considered that the ‘most influential and useful of his contributions to biology undoubtedly is his successful introduction of consistent binomial specific nomenclature’. Itis true that Linneaus’ contribution to ecology and plant geography is rarely acknowledged within these disciplines, and the recognized founding fathers were all much more recent. Nevertheless his Philosophia Botanica contains many ecological insights, which were in the published literature and were dormant seeds for many decades. Given the very large number of students who attended Linneaus’ classes, and the wide circulation of his publications, the ecological perspective he developed must have been assimilated into the perceived wisdom of the day, and when ecology and plant geography developed as separate disciplines, Linneaus’ ideas would have been part of the assumed background. Today the major concerns of ecology include the identification and evaluation of biodiversity. Of the three generally accepted levels of biodiversity, Linneaus was ignorant of genes, but he clearly recognized the need to document species, and recognized that species occupied habitats. In fact, he devotes part of the Philosophia to discussing the main habitats (communities) in Sweden. He also indicated what notes should be made on field excursions. Unfortunately, the details on many herbarium labels in current collections fail to provide any ecological information. Linnaeus’ advocacy of the systematic recording of habitat data was part of his approach to cataloguing information and these features are easily accommodated in modern databases. If Linnaeus were alive today, he would undoubtedly be active in the development of bioinformatics and the creation and manipulation of databases. The omissions of the past cannot be corrected but today’s collectors should be encouraged to record much more than is often the case. Regrettably, ecologists are often amongst the worst offenders when it comes to a lack of detail associated with voucher specimens. Linneaus was well aware of the variability displayed by some species and devoted Chapter IX of the Philosophia to a discussion of ‘Varieties’. He urged against giving taxonomic recognition to environmentally determined phenotypic variation, as he recognized that a variety of diseases and insect attack could cause abnormalities in plants (Philosophia Proc. Linn. Soc. N.S.W., 130, 2009 section 312), displaying evidence of very careful observation. He also pointed out that variation could be correlated with soil conditions and microclimate and advocated an experimental approach (Philosophia section 316: “Cultivation is the mother of very many varieties and is the best means of testing varieties’), foreshadowing by a century and a half experimental taxonomy (genecology), which enjoyed its heyday in the second half of the 20" century. Chapter XI of Philosophia (entitled ‘Sketches’) contains much ecological material. Section 334 ‘The native locations of plants relate to region, climate, soil and ground’ contains a very succinct introduction to ecology and biogeography (as well as some rather strange views about geology). The discussion about the relationship between latitude and flora gives hints of the ideas subsequently developed in greater detail by Alexander von Humboldt. The relationships between soil types and the plants they support also introduce topics that formed a major part of ecological research in the twentieth century. Section 335 provides an overview of phenology and indicates that Linnaeus was well aware that factors such as temperature and day length were involved in controlling flowering, although it was to be many years before physiological understanding of the mechanisms involved was achieved. Even on botanical excursions students recorded the plant species eaten by particular animal species ‘while watching the botanical specimens disappear at the moment they realized that they needed to identify them’ (Koerner 1999 p49). Chapter V of Philosophia (Sex) includes observations on annual seed production of individual plants, probably the first scientific exploration of plant demography. The essential feature of the Philosophia is the importance of observation, and it is remarkable how much was achieved using lenses and microscopes that today would be regarded as woefully inadequate. While Linnaeus was a creationist, the recognition of variation suggests that he was not as rigidly so as he is usually portrayed - he certainly recognized that the appearance of species could change. A synthesis of Buffon’s and Linnaeus’ ideas could have accelerated the development of evolutionary theories, well ahead of the publications of Darwin and Wallace in the mid 19" century. The Philosophia is also strongly focused on the utilisation of plants, not just as medicines but for a whole range of purposes. The 365" (and final) article states: ‘The economic use of plants is of great utility to the human race.’ (Linnaeus, 1751). One of the major justifications for biodiversity conservation is the maintenance of the ecosystem services that LINNAEUS: KING OF NATURAL HISTORY biodiversity supports. This is a concept that would clearly have found favour with Linneaus, and the sorts of observations he advocated are needed to document ecosystem processes. He recorded details of the trophic interactions between organisms and had an appreciation of the recycling of materials, noting people used churchyard soil for growing cabbages, hence ‘human heads ... turn into cabbage heads’ (Koerner, 1999, p83). Robert MacArthur (1972) famously wrote that ‘to do science is to search for repeated patterns’ and stressed the importance of natural history as the starting point for ecological research. MacArthur pointed out that not every natural historian was a scientist (in terms of approach and method, not necessarily profession) and not all ecologists were natural historians, but we would agree with him that most of ecology has its roots in natural history: even theoretical mathematical ecology starts with ideas that are ultimately based on field observation. Underwood (2007) has recently observed ‘one of the great joys of experimental ecology is that natural history is SO important in the development of explanatory models’. COLLECTING BIODIVERSITY: PRESERVING BIODIVERSITY? In the eighteenth, nineteenth and early twentieth century much natural history involved collections; and many large collections of, for example, insects, bird eggs, or plants were made both by, or for, major institutions and individual collectors. Many people who subsequently became famous scientists in other fields (for example Macfarlane Burnet - Sexton 1999) were avid collectors in their youth. Charles Darwin himself was an avid beetle collector in his college days (preferring ‘beetling’ to mathematics - Desmond and Moore 1991). (Another suggestion for Linnaeus’ mis-representation of his travels in Lappland was mis-calculation (Selander, 1947 in Koerner, 1999). Perhaps he shared Darwin’s aversion to mathematics?). The making of collections taught the need for careful observation, systematic recording of data, and provided in-depth understanding of particular groups of organisms. Collectors and recorders were not just the clergy and the landed gentry (or their spinster siblings); there was, at least in the United Kingdom, a very strong working class element of miners and factory workers, who, in their very limited free time, spent many hours completing arduous hikes and making major finds of often taxonomically-challenging organisms. 248 This tradition of the extremely skilled amateur was never as strong in New South Wales as it was in the United Kingdom (for reasons that perhaps require the attention of a social historian). Certainly, there have been some very gifted amateurs, but their interests tended to be restricted to groups such as birds or flowering plants; there was not the same interest or expertise shown in, for example, cryptogamic plants as was the case in the United Kingdom. Given the dearth of professionals in Australia, this means that there are major components of biodiversity about which we remain still basically ignorant. Today collection is frowned upon, and in many cases (such as the collecting of bird eggs) it is properly illegal. The shift away from collection partly reflects lack of opportunity given an increasingly urbanized population, rejection of a “stamp collecting’ approach to science and greater concern for conservation. Certainly, it would not be possible to condone egg- collecting or capturing and killing native vertebrates outside specially-approved scientific licences, but nevertheless it is probable that both school and university students are missing out on what previously had been important educational experiences. The old collections remain of continuing value, providing comparative material for taxonomic studies, as well as evidence of changing distributions or of environmental change. For example, the ability to measure long-term trends in thickness of the shells of raptor eggs was extremely important in drawing attention to the effects of the new post-war organic agricultural chemicals (Ratcliffe 1967, 1970; Olsen and Olsen 1979). Sadly, curation of historical collections in some of our greatest museums 1s being eroded as funding for the care of collections, in the management schemes of modern-day directors, is coming a distant second behind promotion of exhibitions for entertainment’s sake without the underlying scholarship being obvious. Even in the absence of collections, skilled amateurs in the United Kingdom have been able to systematically record distributions of large numbers of taxa at the national scale. These spatially explicit data are of enormous value for monitoring environmental change. Data of this sort would be very difficult to collect (and mordinately expensive) if we had to rely on professionals, yet they will be crucial to our monitoring of biodiversity. In Australia the omithologists have pioneered systematic recording of species distribution at the continental scale, harnessing the skills and enthusiasm of amateurs and professionals. The differences in distribution and abundance between the two editions of the Bird Atlas (Blakers et al. 1984; Barrett et al. 2003) provides Proc. Linn. Soc. N.S.W., 130, 2009 P. ADAM AND E. MAY compelling evidence for the impacts of environmental change, and the recording program will continue into the future. Other areas of natural history have not been so well served. Teaching in the field is still a component of many University courses but is under great pressure because of cost, occupational health and safety issues (which can create bureaucratic nightmares), the large number of students with part-time jobs who find it difficult to attend courses at weekends or during vacations, and because of the decline in the number of academic staff with knowledge of many groups of organisms. The long-term future of field teaching is very uncertain and this will have major consequences for our ability to produce graduates capable of addressing biodiversity issues. Peter Marren (2002, 2005) has written on a number of occasions, pointing out the loss of expertise in UK institutions and the decline in the numbers of skilled amateurs. At the same time there is an increase in the membership of NGO conservation societies, indicating wide support for natural history, but the deep engagement with some particular field within natural history is less common. There may be many and varied reasons for this, but Marren suggests one may be that, with the modern pressures on many people, and the absence of time, natural history had become a spectator rather than participant sport. The quality and expertise shown in recent TV natural history programs 1s such that rather than encouraging participation they suggest that we already know everything. This is an idea that needs further exploration. The achievements of Linneaus and _ his students were remarkable and the detail of their observations made with minimal technological aids, was particularly remarkable. Few students today would have the capacity or patience to make similar investigations. Experimental science needs to be underpinned by substantial bodies of observation in order that appropriate hypotheses can be generated and tested. Linnaeus’ legacy of observation has been built upon for the last 250 years, but the capacity to continue to do so is being lost. Contrary to the impression left by Conniff, we would argue that Linnaeus also laid the foundation for other major areas of ecology, including comparative biogeography (long before van Humbolt), plant demography, and comparative anatomy. His legacy was far more encompassing than taxonomic, even though the systematic recording of species remains absolutely fundamental to modern ecological concerns. Proc. Linn. Soc. N.S.W., 130, 2009 REFERENCES Barrett, G.W., Silcocks, A.F., Barry, S., Cunningham, R.D. and Poulter R. (2003). ‘The new atlas of Australian birds’. (Birds Australia [Royal Australasian Ornothologists Union]: Hawthorn East, Victoria) Blakers M., Davies S.J.J.F. and Reilly, P.N. (1984). “The atlas of Australian birds’. (Royal Australasian Ornithologists Union and Melbourne University Press: Carlton, Victoria) Blunt, W. (1971). ‘The Compleat Naturalist. A Life of Linnaeus’. (Collins: London). Conniff, R. (2007). Happy Birthday, Linnaeus. Natural History 115: 42-47 Desmond, A. and Moore, J. (1991). ‘Darwin’. (Michael Joseph: London). Koerner, L. (1999). “Linnaeus: Nature and Nation’. (Harvard University Press: Cambridge, Massachusetts). Linnaeus, C. (1751). ‘Philosophia Botanica’. (Translated by Stephen Freer.) (Oxford University Press: Oxford). MacArthur, R.H. (1972). “Geographical ecology: patterns in the distribution of species’. Harper and Row: New York. Mayr, E. (1997). ‘This is Biology. The Science of the Living World’. (Belknap-Harvard University Press: Cambridge, Massachusetts). Marren, P. (2002). “Nature Conservation - A Review of the Conservation of Wildlife in Britain 1950-2001’. (Harper Collins: London). Marren , P. (2005). “The New Naturalists- half a century of British natural history. 2nd Edition’. (Harper Collins: London). Olsen, P. and Olsen, J. (1979). Eggshell thinning in the peregrine, Falco peregrinus (Aves: Falconidae), in Australia. Australian Wildlife Research 6, 217-226. Ratcliffe, D.A. (1967). Decrease in eggshell weight in certain birds of prey. Nature (Lond) 215, 208-10. Ratcliffe, D.A. (1970). Change attributable to pesticides in egg breakage frequency and eggshell thickness in some British birds. Journal of Applied Ecology 7, 67-115. Ratcliffe, D.A. (2006). ‘Lapland: a natural history’. (Yale University Press: New Haven, Connecticut). Selander, S. (1947). Linné 1 Lule Lappmark. Svenska Linnéséllskapets Arsskrift. XXX, 9-20 Sexton, C. (1999). ‘Burnet, a life’. (Oxford University Press: Melbourne). Underwood, A.J. (2007). Logic, design, analyses, interpretations: essentials of ecological experimentation. British Ecological Society Bulletin 38, 24-27. Wilson, E.O. (ed) (1988). ‘Biodiversity’ (National Academy Press: Washington DC). 249 LINNAEUS: KING OF NATURAL HISTORY (Endnotes) ' He variously reported traveling up to three times 250 further than he really did and spoke of long periods spent with the native Sami, when in fact it was a few weeks. The fact that he was being paid by the mile for his journey by the Science Society of Uppsala is cited by way of partial explanation (Koerner 1999). Proc. Linn. Soc. N.S.W., 130, 2009 Erratum Wright, A.J. (2008). Emsian (Early Devonian) tetracorals (Cnidaria) from Grattai Creek, New South Wales. Proceedings of the Linnean Society of New South Wales 128, 83-96. Figure 4 on page 92 of the above paper was published without scale bars. It is reproduced here: Ca ee Ma te a og > ‘ ~* ’ “) oe. Wo gs oe Cpeyene" -.< Sa a ee kat * ‘ , MMF transverse view ; b b) Figure 4a-b. Trapezophyllum grattaiensis sp. nov., holotype. a, MMF 34186b =5 mm. 34186a, longitudinal view. Bar scales if wanotesy viele nae sipert'n sa ihe cubes! Ae, ihe rey Kieee