Records of the Museums and Art Galleries of the Northern Territory Volume 16 December 2000 The Beagle, Records of the Museums and Art Galleries of the Northern Territory (formerly ‘Records of the Northern Territory Museum of Arts and Sciences’) ABSTRACTED IN ZOOLOGICAL RECORD AND BIOSIS EDITORIAL COMMITTEE D. Megirian Editor H.K. Larson Associate Editor P.N. Alderslade C.J. Glasby P.F. Murray R.C. Willan The Beagle is a refereed journal published by Academic Publications of the Museums and Art Galleries of the Northern Territory to disseminate the results of research in the following areas: Systematic and other studies of the terrestrial, marine and freshwater flora and fauna of the Northern Territory, tropical Australia, Southeast Asia and Oceania; Australian Aboriginal, Southeast Asian and Oceanic art, material culture and archaeology; Northern Territory and Indo-Pacific history and archaeology. The Beagle is publi.shed once or twice a year, depending upon the material available and budgetary considerations. All contributions to The Beagle arc peer reviewed by two referees and, where possible, at least one internationally-based referee is selected by the Editorial Committee. Whilst articles for The Beagle will normally be 10-50 pages in length, shorter communications, notes and review articles may also be acceptable. Longer articles, significant works or substantial revisions, which form integral studies, may be considered for separate publication as a Supplement. Prospective authors should contact the Editor, Academic Publications. The Beagle may be obtained by subscription or by exchange. Subscription rate for one year: individuals - $30.00, institutions - $40.00. Postal charges; within Au.stralia - $5.00; overseas - $ 10.00 (surface), $20.00 (airmail). Cheques or money orders should be make payable to ‘Museums and Art Galleries of the Northern Territory’. All subscriptions, back numbers and exchange enquiries should be addressed to The Librarian, Museum and Art Gallery of the Northern Territory, GPO Box 4646, Darwin, NT 0801, AUSTRALIA. AUTHORS OFFPRINTS Twenty-five offprints are provided free for each published paper. Additional offprints may be ordered when returning proofs. SUBMISSION OF MANUSCRIPTS A Guide to Authors is provided on the inside back cover of each volume. Contributions, including the original and two copies of each manuscript (including artwork), and a computer floppy disc or CD ROM should be sent to; The Editor, Academic Publications Museum and Art Gallery of the Northern Territory GPO Box 4646, Darwin NT 0801, AUSTRALIA. ©Museums and Art Galleries of the Northern Territory, 2000. ISSN 0811-3653 Printed by the Government Printer of the Northern Territory Northern Territory Government Department of Arts 5. Museums Front cover: Leptothecate hydroid colonies, Lytocarpia angulosa (left) and Nemertesia cylindrica (right), from Darwin Harbour (From J. E. Watson, pages 1-82) The Beagle Records of the Museums and Art Galleries OF THE Northern Territory Volume 16, December 2000 CONTENTS WATSON, J.E. - Hydroids (Hydrozoa: Leptothecatae) from the Beagle Gulf and Darwin Harbour, northern Australia.1 BRUCE, A. J. - The rediscovery of Leandrites stenopus Holthuis, 1950 (Crustacea: Palaemoninae), from Lucinda, Queensland.83 BRUCE, A. J. - Onycocaridella prima Bruce, 1981, a rare pontoniine shrimp from Darwin Harbour (Crustacea: Decapoda: Pontoniinae).89 BRUCE, A. J. - Biological observations on the commensal shrimp Paranchistus armatus (H. Milne Edwards) (Crustacea: Decapoda: Pontoniinae).91 VON HAGEN, H.-O.- Vibration signals in Australian fiddler crabs - a first inventory.97 BROWN, G. R. - Caetmthynnus, Nitidothynnus and Procerothynnus, new genera of Thynninae (Hymenoptera: Tiphiidae) from northern Australia.107 MACHIDA, Y. - A new bythitid genus and species, Acarobythites larsonae, from shallow rocky reefs off northern Australia (Pisces, Ophidiiformes, Bythitidae).123 KAILOLA, P. J. - Six new species of fork-tailed catfishes (Pisces, Teleostei, Ariidae) from Australia and New Guinea.127 MURRAY, P. and MEGIRIAN, D. Two new genera and three new species of Thylacinidae (Marsupialia) from the Miocene of the Northern Territory, Australia.145 CHATTO, R. and WARNEKE, R. M. - Records of cetacean strandings in the Northern Territory of Australia. 163 .. . ***'* <*»> 7^/fe . |.- f.ii'Vi •'> - ' ~ «-T-f-V ■’ & jPTi^jrflrf^/Sl 1^1 . jf-. i*>iSi|it's' i .>;£is^'^.l4» - "Jut ■ ?' !'..-v>'' 'V . 'S. niffcK;!* 6# B»r j» ^40f V-4 ■ ‘■C^i - H-Or... M' ' ' ^ '.*jM Jtrr IP r»ww>Jf t'iitsn*w%«i; ;4er. iQ' .Ijr -', - r.v ..-, ,4 ^ .‘t,A *4'' [. '■U» i*' # 4 . ‘^'' 4 >^CV'^' •*’ ■ "i ■ » •(« .^f . .I'^t . ■ . -1 * •—A ' * » .V *' ■.-«»-■ It;' V ^■-4 ’?'■' Ci^<- '!*> . >H.'.' v; Aa>®..*,j4h‘ ■■' Will* ;•■«' . »'9. ►’'I'f - r “ / Va 4'i t '''•.- '"V *■>.?; IfjC- lr,‘4n?i^i-» li,., " ■"••>!' .^F' V i;i-.4,!f*j.jtf Sf^prufj'' i;. ' TfAspiAiCITv '..'■»5!tri4»..ii. ':, :J .If ■j-<»@,*-j:.5'i«’*Ti'i auScA .. ,••'•*■ .t-i ■* •- ''^ If -^ • w .j(w|»tp '?»;»>.; vi^-i.>»;(r‘r'|^r -- ii4 ^•.iVi4?A<» ■' i' w*' f iiVit- ' r »• . >' i.l .TACJ*:^., J. i^yel ' .‘•*->-S- i* f ► 1 ‘T^f -• V ^ ' j I ,;t,. ..(x* , ■ti. ■ f'A >•. v/'= fj Ijf it..C't‘if. . 44V- .---.- 5^-'A.-’.- •-»'^ffj^ ”>eA' 4!!}^^ AAa 4 . 4iv^v i.^.4a .44%fr.''^ f« "T • f',. I iVr,- '•&i ... ■• 4 .''*''♦ Ar*'. i-'"'.,. s.*-.* ■'5 ■^> 3 a. g 3^0 O D. d. *0 ® ^ jC ^ Q Q O ^ g £ 3 H O O -s ^ O I •S e ^ •S §.f- E S .Q =- t= as ° r-; 3 f5 O § Q .2P iB b fa X .£ 3. C3 S Q. o ■S I ^ £ o o on o 00 .£ ul a. ^ -2 Q* g U . 0 ) o 3 ~o o o o c ^ •3 E ^ E iS o o — o . •o < >. .. 3-^. S U ^ 34 Beagle Gulf and Darwin Harbour hydroids wall straight to faintly sinuous or concave, contiguous with internode, base of adcauline wall with an inturned knot of perisarc, no true floor to hydrotheca, connection with hydrocladium wide. Margin thickened with two lobate lateral cusps and a much smaller adcauline cusp, aperture elliptical, opercular remnants obscuring most margins. Hydranths not well preserved, no caecum; approximately 16 tentacles. Gonotheca immature, balloon-shaped to irregularly cylindrical, inserted without pedicel between two cauline hydrothecae in mid to upper stem region, expanding a little apically to a blunt distal end, gonophore an aborted hydranth, perisarc extremely thin. Colour. Pale yellow-brown. Measurements (pm). Stem distance between successive hydrocladia on same side 1,060 -1,100 length of apophysis (distal side) 112 - 128 width of apophysis at distal node 144 - 160 Hydrocladium length of internode 296 - 336 width at node 120 - 128 Hydrotheca length of adnate adcauline wall 184 - 228 length of free adcauline wall (lateral view) 32 - 56 width between marginal cusps 88 - 112 Gonotheca length 600 width 256 Remarks. Although the cauline perisarc is com¬ paratively thick the stems are quite lax. Many hydrocladia are broken off at the distal apophysal node. The hydrothecal aperture is small and frequently obscured by adhering tissue and opercular remnants. Vervoort (1959) erected the var. alternata to accommodate specimens from west Africa with alternate hydrothecae; the present material has subopposite hydrothecae, thus conforming to Busk’s original description of the species. This first description of the gonotheca is based on the single immature gonotheca present in the sample; it is remarkably small and delicate in comparison with the sturdy trophosome. As there is no sign of ova it is probably male. Distribution. Tropical Indo-Pacific and Red Sea (Vervoort 1959, Mergner and Wedler 1977). Australian distribution: Torres Strait (Busk 1852) and tropical Queensland (Pennycuik 1959). Genus Thuiaria Fleming, 1828 Thuiaria operculata sp. nov. (Fig. 27A-D) Records and material. Holotype, NTM Cl2956, alcohol preserved material, NTM Cl2949, NTM C12971, MV F86921, microslides from holotype colony. East Arm port, concrete wall of berth, coll: J. E. Watson, depth 5 m, 20/8/1998, colony of several infertile stems. Paratype, NTM Cl2973, microslide, two infertile stems on hydrorhiza of Idiellana pristis, Stn 146. Description. Hydrorhiza thick, gnarled. Stems to 35 mm high, gracefully plumose, monosiphonic, proximal stem robust, perisarc thick, thinning distally, a distinct transverse node above base of stem, proximal cauline internode athecate, cauline and hydrocladial internodes thereafter thecate, nodes distinct to absent, slightly oblique to transverse, perisarc tumid above and below node. Hydrocladium proximal on internode, a hydrotheca deep in axil, one above, margin almost level with distal node, another in alternate position, opposite on stem. Hydrocladia alternate, long, upwardly directed at 60 - 70° to cauline axis, borne on a strong apophysis with one or two transverse distal nodes, internodes thereafter with two hydrothecae, nodes slightly oblique. Hydrothecae frontal on internode, subaltemate, facing forward, adnate adcauline wall smoothly convex, free part short, a small hook-shaped knot of perisarc at base of wall; abcauline wall contiguous with internode, straight to slightly sinuous, maximum width of hydrotheca in distal third, a bracket-like internal submarginal shelf just below abcauline margin. Aperture oval, facing upward with two large lateral lobes and a very small adcauline cusp; operculum a strong flap with outrolled rim, attached to abcauline submarginal shelf; perisarc of operculum thick. Hydranth with c. 10 tentacles, abcauline caecum visible in some hydranths. Colour. Live stems pale greenish-white. Measurements (pm). Stem diameter of basal node 176 - 208 length of internodes 560 - 680 diameter at node 112 - 168 Branch length of internode 400 - 520 diameter at node 72 - 96 abcauline length of apophysis from stem node 180 - 208 length of first athecate internode 60 - 72 Hydrotheca length of adnate adcauline wall 180 - 204 length of free adcauline wall 32 - 44 maximum width 96 - 100 Remarks. The abcauline operculum of one valve and hydranth with abcauline caecum necessitates referral of the species to Thuiaria. The operculum attached to the knuckle-shaped submarginal shelf is quite delicate in frontal view but in lateral view appears much thicker, its outrolled upper edge being quite clear. In contracted hydranths the small abcauline caecum is rather obscure, being evident in extended hydranths. 35 J.E. Watson Etymology. Named for the strongly developed opercular flap. Thuiaria plumularioides sp. nov. (Fig. 28A-F) Records and material. Holotype, NTM Cl2974, NTM Cl2975, MV F86923, microslides, from infertile colony on sponge, Stn 153. Paratypes, NTM Cl2976, microslide, colony from Stn 110. NTM C12977, microslide, colony from Stn 154. (All holotype and paratype material mounted on microslides: no preserved material remaining). Description. Hydrorhiza a knot of tubes entwining substrate. Stems to 25 mm high, straight, robust, pinnate, proximal part of stem ahydrocladiate but with two rows of opposite hydrothecae; ahydrocladiate paart of stem usually with a strong distal hinge joint. Cauline nodes of hydrocladiate part of stem distinct, slightly oblique to transverse; internode with three hydrocladia, one to three hydrothecae above hydrocladium, proximal one axillar, di.stalmost level with node. Cauline perisarc very thick proximally, thinning a little distally. Hydrocladia close set, alternate, long, arching gracefully upwards at an angle of c. 30“ to cauline axis from a strong apophysis with an indistinct transverse constriction and a pronounced distal hinge joint. Hydrocladial internodes long, nodes distinct hinge joints. Hydrothecae biseriate, frontal on hydrocladium, pro¬ ximal internodes bearing up to 10 pairs of hydrothecae, pairs becoming fewer distally; hydrothecae overlapping, facing upwards, sub-opposite on proximal region of hydrocladium, becoming opposite distally, each pair overlapping base of pair above. Hydrotheca long, tubular (lateral view), almost vasiform (anterior view), slightly tumid at base, narrowing just behind margin, adcauline walls joined, straight, a short convex length free of internode at a sharp outward distal bend; abcauline wall almost straight, a prominent abcauline submarginal shelf, a foramen at base of wall connecting with internode, base of hydrotheca rounded. Aperture oval, margin thickened with a pair of opposite laterally placed cusps and a rounded adcauline lobe; opercular valve attached to submarginal adcauline shelf. Hydranth with an abcauline caecum and about 10 tentacles. Colour. In life, pale yellow. Measurements (pm). Stem maximum diameter at proximal node 850 length of internode 960 -1,960 diameter at node 220 - 500 apophysis, length of abcauline wall 280 - 360 apophysis, width at transverse node 144 - 184 Hydrotheca length of adnate adcauline wall 184 - 240 length of free adcauline wall to margin 100 - 160 length of abcauline wall to top of margin 240 - 280 depth of margin, lateral view 60 - 64 width of margin, cusp to cusp. frontal view 68 - 80 Fig. 28. Thuiaria plumularioides .sp. nov. A, stem from holotype colony. B, hydrocladia. C, frontal view of hydrocladium. D, hydrocladium, lateral view. E, hydrotheca showing lobate margin. F, lateral view of hydrotheca showing submarginal abcauline thickening. Scale bars: A, 10 mm, B, 1,000 pm; C, D, 500 pm; E, F, 200 pm. 36 Beagle Gulf and Darwin Harbour hydroids Remarks. The species is referred to Thitiaria on the basis of the abcauline caecum in the hydranth and the hydrotheca having an abcauline operculum. Colonies may be sparingly branched, one stem of the holotype colony having remnants of a branch near the base. The overlapping hydrothecae and slightly tumid base of the hydrothecae are characteristic. Etymology. The species is named for its plumose morphology. Genus Thyroscyphus Allman, 1877 Thyroscyphus macrocytharus (Lamouroux, 1824) (Fig. 29A) Clytia macrocytharus Lamouroux, 1824: 647. -Lamarck 1837: 199. Campanularia marginata Bale, 1884: 54. - Bale 1888: 758. - Bartlett 1907: 62. - Levinsen 1913: 289. Laomedea marginata - von Lendenfeld 1885a: 404. Thyroscyphus marginatus - Bale 1914: 91. - Bale 1915: 245, 258. - Stechow 1924: 69. - Stechow 1925: 217. - Blackburn 1942: 112.-Watson 1973: 169. Thyroscyphus balei Calder, 1983:16. - Watson 1992: 220 . Thyroscyphus macrocytharus - Watson 1994: 156. - Watson 1996: 78. - Watson 1997: 517. Record and material. NTM Cl2987, MV F86893, microslides, sparingly fertile colony of four straggling stems on sponge, Stn 154. Description. Hydrorhiza reptant on substrate, hydrocaulus of same diameter as hydrorhiza, mono- siphonic, perisarc very thick. Stems to 10 mm high, internodes long, smooth, almost geniculate, nodes oblique, deeply incised. Hydrothecae alternate, distal on internode, borne on a moderately long pedicel either smooth or with one or two constrictions; hydrotheca campanulate, margin quadrate, conspicuously thickened, with four pointed cusps and four shallow emarginations; operculum of four equal valves. Gonotheca immature, top-shaped, borne on a short pedicel inserted in apophysis of stem beside hydrotheca. Colour. Pale brown to colourless. Measurements (pm). Stem length of internode 960 -1,320 width at node 168 - 256 Hydrotheca length of pedicel 100 - 240 depth, diaphragm to margin 680 - 792 diameter at margin 528 - 568 Remarks. The specimen conforms to the description of Thyroscyphus macrocytharus given by Watson (1994). Most hydrothecal pedicels of the present material are regenerated, some up to eight times. Thyroscyphus macrocytharus is endemic to Australia, usually occurring in shallow water; the present record from a depth of 30 m is the second deepest record for the species, the deepest being 137 m off southern Australia (Bale 1915). Distribution. The previous northernmost record of T. macrocytharus is from Geraldton, Western Australia (Stechow 1925). This is the first record of the species from tropical Australia. Thyroscyphus torresii (Busk, 1852) (Fig. 29B-C) Laomedea torresii Busk, 1852: 402. Thyroscyphus simplex Allman, 1888: 25. - Jiiderholm 1903: 273. - Jaderholm 1916: 5. - Stechow and Muller, 1923: 466. Campanularia torresii - Bale 1884: 52. - Kirkpatrick 1890: 604. Thyroscyphus regularis Jaderholm, 1896: 9. Thyroscyphus torresii - Jaderholm 1903: 273. -Stechow 1913: 12. - Jaderholm 1916: 5. - Stechow and Muller 1923: 466. - Redier 1963: 22. - Watson 1996: 78. Cnidoscyphus torresii - Splettstdsser 1929: 70-82, 125. - Pennycuik 1959: 156. - Vervoort 1941: 204. - Vervoort 1993: 104. Records and material. NTM Cl2887, alcohol preserved material, colony from Stn 110. NTM Cl2886 alcohol preserved material; NTM Cl2989, MV F86892, microslides, colony from Stn 13. NTM C12888, alcohol preserved material, colony from Stn 136. MV F86938, alcohol preserved material, colony from Stn 137. NTM Cl2988, microslide, colony from Stn 138. Many infertile colonies, the largest of 20 stems on shell and grit, Gymnangium longicorne and sponge. Other records. Stns 48, 67, 78, 87, 127, 136, 155. East Port channel and wharf pilings, small and large colonies. Port of Darwin, coll: J. E. Watson, depth 3 m, 16/8/1998. Plater Rock, many sparsely fertile large and small colonies, coll J. E. Wat.son, depth 3 - 10 m, 21/9/1999. De.scription. Hydrorhizal stolons tightly entwining substrate. Clusters of up to 20 pinnate stems to 100 mm high arising from hydrorhiza, basal stem region lightly fascicled by hydrorhizal stolons, stems thereafter monosiphonic, tubular, stiff and robust, perisarc of basal region very thick and smooth, thinning distally. Hydrocladia pinnate, alternate, well separated, long, no secondary branching, two alternate hydrothecae on stem between each branch and one axillar. Stem and hydrocladial internodes fairly long, nodes slightly oblique to transverse, marked by a notch in perisarc. Hydrocladial apophysis long, inclined upwards, proximal side marked by an indentation in perisarc in stem; first hydrocladial internode long, internodes shorter thereafter, nodes transverse, usually distinct. Hydrothecae alternate, pedicel short, thick, smooth or with several ridges; hydrotheca asymmetrical, abcauline wall straight, proximal third of adcauline wall convex, the curve straightening out distally, diaphragm distinct, down-curved, often asymmetrically tilted, hydrothecal margin circular, rim quadrate with four low. 37 J.E. Watson equidistant, sharply pointed cusps separated by four shallow emarginations; operculum thin, of four triangular flaps. Hydranths stubby, with at least 30 tentacles. Gonothecae immature, top-shaped, arising in proximal stem region beside a hydrotheca or on stem, hydrocladium or apophysis from which a hydrotheca has been shed; perisarc thin. Colour. Live colonies golden-yellow to golden- brown. Measurements (pm). Stem diameter at base distance between successive branches length stem internodes between successive hydrothecae Branch length first internode length succeeding internodes width branch at base width at node Hydrotheca length, including pedicel length of pedicel apophysis to diaphragm width of pedicel width at diaphragm diameter at margin width of marginal embayment Remarks. The hydrothecal pedicels of younger hydrocladia are usually smooth but in older stems are often ridged from repeated breakage and regeneration. There is no submarginal thickening of the hydrothecal rim as in Thyroscyphus macrocytliarus, only a faint line below the rim in some hydrothecae marking the junction of the operculum with the body. Several colonies are infested with Hebellopsis scandens, the hydrothecae of which are invariably close beside those of Thyroscyphus. The lower stem regions bearing developing gono¬ thecae are packed with globular white bodies, apparently developing gonophores which are visible through the stem perisarc. It is surprising that, as T. torresii is one of the most abundant hydroids in both dredged and scuba collections, only a few immature gonothecae were found. The only description and figure of the mature gonotheca of T. torresii is of a distally ridged gonotheca (Jaderholm 860 - 950 2,300 -2,800 800 -1,000 1,900 -2,040 700 -1,100 200 - 260 192 - 224 1,080 -1,200 120 - 152 160 - 192 240 - 280 472 - 576 232 - 320 Is of younger 1903). Distribution. Indonesia, China Sea, Mergui Archi¬ pelago (Vervoort 1993). Australian region: Torres Strait (Busk 1852, Kirkpartick 1890) Aru Sea (Stechow and Muller 1923), Fitzroy Island (Bale 1884), off Cape York (Allman 1888), Cape Jaubert, Western Australia (Jaderholm 1916), Queensland (Pennycuik, 1959). Thyroscyphus fruticosus (Esper, 1793) (Fig. 29D) Spongia fruticosus Esper, 1793: 188. Sertularellafruticosus-Thompson 1879: 100. Campanularia fruticosus - Marktanner-Turneretscher 1890: 205. Thyroscyphus vitiensis Marktanner-Turneretscher, 1890: 210. - Billard 1907: 343. - Jarvis 1922: 338. Lytoscyphus fruticosus - Bedot 1905: 51. - Stechow and Muller 1923: 465. - Stechow 1925: 215. Thyroscyphus fruticosus - Splettstosser 1929: 7, 122. - Billard 1933: II.- Vervoort 1941: 202. - Vervoort 1946: 306. - Millard 1952: 199. - Ralph 1961: 754. - Vervoort 1965: 35. - Vervoort 1967: 35. - Schmidt 1971: 35. - Millard and Bouillon 1973: 76. - Millard 1975: 323. - Mergner and Wedler 1977: 18. - Gibbons and Ryland 1989: 425. - Watson 1996: 78. Record and material. NTM Cl2889, MV F86945, alcohol preserved material; NTM Cl2990, microslide. Plater Rock, tall colonies on coral boulders and rock, coll; J. E. Wat.son, depth 5-8 m. 22/9/1999. Description. Colonies of up to 10 stems to 150 mm high. Hydrorhiza a tangled mass of tough tubes embedded in substrate. Stem and branches stiff, monosiphonic, proximal part of stem athecate; branching irregular, predominantly in one plane, usually almost perpendicular to stem, no division into nodes, tubular in section, each segment widening to a short, proximally swollen apophysis supporting a hydrotheca. Perisarc of stem and branches very thick and smooth. Hydrothecae strictly alternate on opposite sides of stem and branches, one in axil of branch; hydrotheca inclined upwards on a short thick pedicel, smooth or with one or two constrictions, tubular, asymmetrical, abcauline wall straight, adcauline wall convex, widest at proximal third, both walls narrowing slightly but distinctly behind margin; diaphragm distinct, marked by a circular shelf in perisarc tilted obliquely downwards; margin circular, rim thickened, an internal band of tissue below margin at narrowest point of hydrothecal body, margin sometimes with four almost imperceptible equidistant cusps, sometimes with one or two reno¬ vations. Operculum delicate, of four equal triangular valves. Gonothecae borne on a short, thick pedicel below hydrothecal apophyses on stem and branches, larger than hydrotheca, top-shaped, tilted downwards, perisarc thick, smooth to rather lumpy, distal end obliquely truncated, sealed by a plug of tissue. Gonophore (possibly female) distal in gonothecal cavity. Colour. Stem and branches of live colonies honey brown, hydrothecae and gonothecae rose pink; colony yellow-brown when preserved. Measurements (pm). Stem and branches diameter above hydrothecal pedicel 340 - 380 distance between successive hydrothecae on same side 1,480 -2,900 width below apophysis 400 - 540 38 Beagle Gulf and Darwin Harbour hydroids Fig. 29. A, Thyroscyphus macrocytharus: part of stem with developing gonotheca. B. C, Thyroscyphus torresii: B, fertile colony from Plater Rock. C, branch internodes with hydrothecae and young gonotheca. D, Thyroscyphus fruticosus: internodes from fertile colony from Plater Rock. Scale bars: A, C, D, 1,000 pm; B, 5,000 pm. width across distal node of apophysis 240 - 320 Hydrotheca length (diagonal) of adcauline wall. diaphragm to margin 1,300 -1,500 length of abcauline wall. diaphragm to margin 1,140 -1,200 diameter of margin 488 - 536 length (abcauline side) of pedicel. apophysis to diaphragm 280 - 320 width of pedicel above apophysis 208 - 240 Gonotheca length 1,460 -1,760 maximum width 840 - 940 diameter at margin 640 - 720 Remarks. Thyroscyphus fruticosus and T. torresii from the Beagle Gulf and Darwin share many characters, particularly in size of colony, habit and choice of habitat. The only reliable means of distinguishing between the species in situ is in the slightly more straggling colonies and pink colour of T. fruticosus while T. torresii is tidier in aspect and is honey brown in colour. Gibbons and Ryland (1989) reported very short (10-40 mm high) violet-coloured colonies of T fruticosus from rock and sandy habitats in Fiji. The pink colouration of T. fruticosus is lost in preservation, a reaction also noted by Millard (1975). Confusing morphological similarities of T. fruticosus with T. torresii have probably led to some past confusion of the species, for example Gibbons and Ryland (1989, fig. 40) depict asymmetrical hydrothecae, clearly those of T fruticosus, but with marginal cu.sps more resembling those of T torresii. In sterile material, the only reasonably constant morphological differences are the absence of internodes from stem and branches of T. fruticosus, the usually larger hydrothecae and the sometimes almost imperceptible narrowing of the adcauline wall behind the margin. Distribution. Mediterranean Sea, Indo-West Pacific Timor Sea, New Zealand. Australian distribution, north¬ western Australia (Watson 1996). Family Syntheciidae Marktanner-Ttirneretscher, 1890 Genus Syntheciiun Allman, 1872 Diagnosis. Colony with erect stem bearing hydro- cladia in opposite pairs. Stem and hydrocladia bearing hydrothecae in opposite pairs, the pairs always in the same plane forming two longitudinal rows. Gonothecae dioecious, springing from within hydrothecae. Remarks. Seven species of Synthecium are reported from Australia; Synthecium patulum (Busk, 1852), Synthecium orthogonium (Busk, 1852), Synthecium Allman, 1872 (see Jaderholm 1911), Synthecium campylocarpum Allman ISSS, Synthecium suhventricosum Bale, 1914 (recognised by Ralph (1958) as a small form of Synthecium elegans), Synthecium megalothecum Billard, 1924 (see Pennycuik 1959) and Synthecium dentigerum Jarvis, 1922 (see Watson 1969). Even with fertile material it can be difficult to differentiate between species of Synthecium. Size differences between hydrothecae in male and female colonies reported in some species, are a further complicating factor. Synthecium orthogonium and Synthecium campylocarpum have been confused in the past and as no type material of Synthecium patulum is known to exist (see Totton 1930) concepts of the species have varied widely. Bale (1888) reported Synthecium orthogonium (here considered to be S. campylocarpum, see discussion later) from off the coast of New South Wales and Billard (1925) synonymised S. orthogonium and S. campylocarpum in S. patulum. Synthecium orthogonium from Indonesia was identified by Vervoort (1941) as S. patulum while the description and figure of a gonotheca identified as S. patulum by Millard and Bouillon (1973, 1975) is clearly that of S. campylocarpum. The Beagle Gulf and Darwin collection includes two species, the most abundant of which is Synthecium campylocarpum-, the other, less common, is here referred to Synthecium orthogonium. Although Synthecium patulum is not represented in the present collection, the 39 J.E. Watson identity of other species rests largely upon establishment of the true identity of S. patiilum. For comparison, a redescription of S. patulum from southern Australia is given, based on extensive material in the author’s collection and holdings in the collection of the Museum of Victoria. Synthecium campylocarpum Allman, 1888 (Figs 30A-G, 31A-C) Synthecium campylocarpum Allman, 1888: 78. - Marktanner-Turneretscher 1890: 248. - Inaba 1890: 52- 54. - Farquhar 1896: 466. - Stechow 1913: 127. - Jaderholm 1919: 14.-Totton 1930: 169.-Ralph 1958: 347. - Yamada 1959: 52. - Hirohito 1969: 18. - Watson 1996: 78. Synthecium orthogonium (Busk, 1852). - Bale 1888: 767. - Bale 1924: 250. Records and material. NTM Cl2933, MV F86946, alcohol preserved material; NTM Cl2979, MV F86097, microslides, colony from Stn 111. NTM C12980, NTM Cl2981, microslides, colony from Stn 20. NTM Cl2983, microslide, colony from Stn 13. NTM Cl2934, NTM Cl2935 alcohol preserved material, colony from Stn 81. Colonies in poor condition on digitate sponges and Eunice tubifex. Plater Rock, large male colony on digitate sponge, coll J. E. Watson, depth 5 m, 22/9/1999. East Arm channel, large male colonies on Eunice tubifex, coll J. E. Watson, depth 3 m, 17/8/1998. Other records. Stns 21, no, 26. Description. Stems to 100 mm high. Hydrorhiza tubular, reptant on host, perisarc of hydrorhiza and stems thick. Stems wide proximally, narrowing distally, proximal internodes usually athecate, internodes thereafter long with one to three pairs of distal hydrothecae; hydrocladia opposite, distal on internode, given off at an angle of 60® - 90° to caulus above cauline hydrothecal pair; nodes transverse, distinct in younger parts of stems, indistinct to completely absent from older parts, sometimes a secondary node between primary nodes. Hydrocladia on a long vasiform apophysis, sometimes a transverse distal node below first hydro¬ thecal pair. Hydrothecae opposite, occupying almost or entire internode, nodes vestigial or absent, hydrothecae often in contact vertically, a little separated in front, in contact behind; hydrotheca tubular, adcauline wall smoothly convex, one fifth to one quarter of wall free of internode, abcauline wall straight or slightly tumescent proximally, becoming concave, curvature increasing distally, perisarc thinning a little towards margin. Margin circular, sinuous, facing slightly forward, parallel to hydrocladial axis or tilted slightly upward, rim outrolled, sometimes up to four renovations from within hydro¬ theca. Colonies dioecious, gonothecae borne on lower stem and hydrocladia. Male gonotheca long, irregularly pod¬ shaped, usually widest about middle, pedicel inserted Fig. 30. Synthecium campylocarpum from Beagle Gulf and Plater Rock: A, hydrocladiate part of .stem. B, hydrocladial internode. C, hydrotheca with everted marginal replications. D, male gonothecae. E, lateral view of female gonotheca. F, frontal view of female gonotheca. G, nematocyst (haplonemc?) from hydrocaulus. Scale bars: A, B, 500 pm; C, 200 pm; D - F. 1,000 pm; G, 50 pm. with a right-angled bend deep into hydrotheca, often splitting hydrotheca; perisarc of gonotheca rather thin, smooth to faintly undulated, orifice small, circular, operculum a thin sheet of tissue, gonophore elongate. Female gonotheca small, lenticular, one side usually flat, other side convex, perisarc thicker than male, six to eight transverse ribs crossing entire body, more deeply incised on flatter side, ribs fading out proximally and distally, orifice small, circular, on a short tubular neck, neck foreshortening as gonotheca matures. Cnidome of large /haplonemes with blunt ends, capsule 62.5 x 17.5 pm, tubule coiled, in tentacles and coenosarc of stems. Colour. Live colonies yellow; female gonophore orange. Measurements (pm). Hydrorhiza diameter 350 - 450 Stern length of internode (distance between hydrocladia) 1,700 -2,000 diameter at node 240 - 320 Hydrocladium maximum length 8,000 40 Beagle Gulf and Darwin Harbour hydroids length of apophysis to first hydrothecal pair 216 - 280 width across base of hydrothecal pair 296 - 336 Hydrotheca length of adnate adcauline wall 448 - 480 length of free adcauline wall 96 - 168 length of abcauline wall 256 - 368 diameter of margin 184 - 208 Gonotheca length of mature male 1,820 -2,200 maximum width of male 560 - 700 length of mature female 1,120 -1,600 width of female, frontal view 940 -1,100 width of female, side view 640 - 740 Remarks. Apical tendrils occur on several of the longer stems. On some older stems, hydrocladial apophyses are regenerated up to three times, each regrowth marked by a shoulder-like node. The large range of material examined revealed no difference in height of stems nor in size of hydrothecae between sexes. Bale (1888) referred to S. orthogoniiun a specimen from Port Jackson, on the eastern Australian coast, later finding nothing (Bale 1914, 1924) to distinguish it from 5. campylocarpum. His opinion was followed by many later authors (e.g. Thornely 1904, Nutting 1905, Jaderholm 1903, Jiiderholm 1916, Stechow and Muller 1923), none of whom described or figured their specimens. Ritchie (1911) assigned specimens from the eastern Australian coast to S. orthogonium noting that their dimensions did not agree very well with Billard’s (1910) measurements of the holotype of that species. His description of the hydrotheca with sinuate margin is undoubtedly of 5. campylocarpum. 1 have compared a microslide preparation of a syntype of Syntliecium campylocarpum from the Challenger collection, taken off the eastern Australian coast (Museum of Victoria collection registered number MV F58214), with a microslide specimen identified by Bale (1888) as S. orthogonium (specimen also held in the Museum of Victoria). The Challenger material is an almost complete fertile stem with young male gonothecae (Fig. 31D-F) and is almost identical with Bale’s specimen, thus supporting Ralph’s (1958) view that Bale’s specimen is not S. orthogonium but S. campylo¬ carpum. When examined in isolation infertile material of S. orthogonium may be confused with S. campylocarpum, especially in colonies of S. campylocarpum in which the hydrothecae are more outwardly bent than usual. They are best distinguished by the larger and more robust stems and distinctly sinuate margin of S. campylocarpum compared with the smaller, more delicate stems and sharply bent hydrotheca without marginal sinuosity of S. orthogonium. Distribution. Known from eastern subtropical to tropical northern Australia (Watson 1996) New Zealand, (Ralph 1958) Japan (Hirohito 1969). Fig. 31. SynthecUim campylocarpum, syntype, from Port Jackson, Challenger Expedition. MV F58214: A, part of fertile stem. B, hydrocladial intemode, hydrothecae with replicated margin with outrolled rim. C, male gonotheca. D, hydrocladial internode from a microslide specimen from Port Jackson, identified by Bale (1888), as Synthecium orthogonium held in Museum of Victoria. E, male gonotheca from same specimen. F, female gonotheca from same specimen. Scale bars; A, 5,000 pm; B, D, 500 pm; C, E, F, 1,000 pm. Synthecium orthogonium (Busk, 1852) (Fig. 32A-F, Table 4) Sertularia orthogonia Busk. 1852: 390. - Bale 1884: 88. - Billard 1910: 25. Not Synthecium orthogonium - Bale 1888: 767. - Bale 1924: 250. ISynthecium orthogonium - Jaderholm 1903: 289. - Jaderholm 1916:6.- Thornely 1904: 119. - Nutting 1905: 950. - Stechow and Muller 1923: 465. - Pennycuik 1959: 190.-Watson 1996: 78. Synthecium patulum (Bunk, 1852) - Billard 1925: 125. - Vervoort 1941: 199. Records and material. NTM Cl2936, alcohol preserved material, colony from Stn 136. NTM Cl2984, NTM C12985, MV F86890, microslides, infertile sparse colony on Gymnangium long iconic from Stn 87. NTM Cl2986, microslide, small fertile colony on aglaopheniid hydroid. Plater Rock, coll: J. E. Watson, depth 4 m, 21/ 9/1999. 41 J.E. Watson Description. Hydrorhiza tubular, reptant; stems short, to 35 mm high, lax, tubular, with two or three pairs of opposite hydrocladia, proximal stem region athecate, variable in length; diameter of stem diminishing and thickness of perisarc reducing distally; stem internodes long, nodes when present, a transverse constriction in perisarc; one or two pairs of hydrothecae about halfway to two thirds up internode; hydrocladia long and flexuous, opposite or single, given off at a slight upward angle from a strong shoulder-like apophysis. Hydrocladial nodes marked by a pronounced narrowing of perisarc or a transverse line, hydrothecae paired, distal on internode, proximal pair displaced slightly relative to one another; hydrothecae long, tubular, pairs may or may not be conjoined, adnate adcauline wall almost straight to faintly convex, free part bending sharply outwards and slightly forward perpen¬ dicular to hydrocladial axis for one third to half adcauline length, sometimes an internal thickening of perisarc at junction of adnate and free wall; abcauline wall slightly tumescent above base, almost straight to distal bend, then either straight or concave to margin, often an internal thickening of wall in bend. Margin parallel or almost parallel to hydrocladial axis, circular, rim strongly everted, some with up to seven marginal replications; perisarc of hydrotheca smooth. Gonotheca on a short pedicel issuing from a hydrotheca in lower stem region, gonotheca sausage¬ shaped, tapering distally, body with two rows of 10 deep corrugations along one side, corrugations smoothing out behind; distal end obtuse, no orifice evident. Colour. Live colony mauve, white or colourless when preserved. Comparison of Beagle Gulf specimens with holotype. 1 have compared specimens from the Beagle Gulf and Darwin Harbour with the holotype of Syntliecium orthogonium (Busk, 1852) from the type locality of Torres Strait, loaned by the Natural History Museum, London (Fig. 32 E, F). The NHM microslide preparation labelled "Sertularia orthogonia Australia, Rattles, Holotype, Busk 1852, Busk Coll: 99. 7. 1.6367” consists of two stem fragments, one 7 mm long and the other 5 mm long, in good condition but without hydrorhiza. There are several pairs of opposite hydrocladia, one pair per internode, given off from just below a node. The hydrocladia bear up to seven pairs of opposite hydrothecae, one pair per internode; adcauline walls of some pairs joined but others separated. Hydrothecae tubular, distal on internode and sharply bent outwards below node where adcauline wall becomes free. Margin circular and parallel to hydrocladial axis; margins of several hydrothecae arc sufficiently well preserved to show replication of the rim. Internal thickening of the abcauline wall is minimal, probably due to shrinkage overtime. The Beagle Gulf and Darwin specimens almost exactly conform with the type, the only difference being the somewhat shorter stem internodes, a variable character of little diagnostic value in the genus. Dimensions of the Beagle Gulf and Darwin specimens are compared in Table 4 with those of the holotype of S. orthogonium. Busk (1852) considered that the strong outward bend of the hydrotheca of Synthecium orthogonium clearly characterised the species but later speculated that S. orthogonium may be a variety of Synthecium patulum (Busk, 1852) from southern Australia. Bale (1884) reported and figured S. patulum from Port Phillip Bay in southern Australia but did not comment upon any supposed relationship with S. orthogonium, continuing to regard S. patulum a distinct species (Bale 1914a). Billard (1925) described and accurately figured S. orthogonium and its gonangium from Indonesia but, like many other authors, assigned it to S. patulum. Compa¬ rison of dimensions tabulated above and dimensions (Figs 32, 33) show that the free adcauline wall of S. orthogonium is much longer and the hydrothecal margin much narrower than that of 5. patulum. Fresh material of 5. orthogonium is readily distinguishable from S. patulum, colonies of 5. orthogonium being much smaller and more delicate than S. patulum. Although closely related, they are here regarded as distinct species. Distribution. On present knowledge, Synthecium orthogonium is a tropical species the only reliable records of which are from Torres Strait (Busk 1852) and Indonesia (Billard, 1910). Table 4. Comparison of measurements (pm) of Beagle Gulf and NHM Busk specimens of Synthecium orthogonium. Beagle Gulf, NHM Busk Darwin specimen Hydrorhiza diameter Stem length of intemode (distance 112 - 200 between hydrocladia) 800 - 2,800 1,880 - 1,960 diameter a! node 140 - 160 152 - 176 diameter lower stem 260 200 diameter thecate part of stem Hydrocladium 216 - 264 184 - 224 maximum length length apophysis to first 4,000 " hydrothecal pair width across base of first 108 - 272 ■ hydrothecal pair Hydrotheca length of adnate adcauline 256 - 320 wall 440 - 480 408 - 464 length of free adcauline wall length of abcauline wall 208 - 320 192 - 240 (diagonal) length of abcauline wall base 360 - 440 328 - 376 to bend length of abcauline wall bend 320 - 326 ■ to margin 136 - 160 - diameter of margin 136 - 160 120 - 152 length of outrolled rim Gonotheca 8 - 16 ■ length 1,300 - 1,400 - maximum width 360 - 376 - 42 Beagle Gulf and Darwin Harbour hydroids Synthecium patulum (Busk, 1852) (Fig. 33A-G, Table 5) Sertiilaria patula Busk, 1852, 390. - Bale 1884: 88. Synthecium patulum - Bale 1888: 766. - Hodgson 1950:18. - Watson 1975: 165.-Watson 1994: 66. ISynthecium patulum - Jarvis 1922: 332, 345. - Trebilcock 1928: 9. - Pennycuik 1959: 190. Not Synthecium patulum - Billard 1925: 125. - Vervoort 1941:199. - Millard and Bouillon 1973: 64. - Millard 1975: 12. - Vervoort 1987: 94. Description. Stems up to 50 mm high, occasionally with secondary branching; hydrorhiza tubular, reptant, perisarc of stems and hydrocladia fairly thick. Proximal ahydrocladiate stem segment variable in length, some stems with hydrocladia extending almost to base; stem internodes long with one or two pairs of opposite hydrothecae near middle of internode, node a constric¬ tion in perisarc. Hydrocladia up to 12 mm long, opposite, given off at an angle of 45° - 50° to stem from distal end of internode. Hydrocladial hydrothecae paired, distal on internode, not conjoined; internode somewhat constricted proximally, expanding to base of hydrothecae; hydro¬ thecae of first one or two pairs on internode a little displaced relative to one another. Hydrotheca tubular, expanding a little distally, perisarc smooth, adcauline wall almost straight proximally, gently convex distally, free part about one quarter total length of wall, base rounded with a small plug of perisarc; abcauline wall proximally tumescent, becoming concave to margin. Margin tilted upwards at an angle of 45° - 60° to hydrocladial axis, strongly sinuate, rim moderately everted, sometimes replicated several times. Hydranth small with about 10 tentacles. Two kinds of nematocysts present: - small ?mastigophores in tentacles, 9 pm long, discharged with difficulty, - large isorhizas 45 - 50 x 12 - 13 pm, tubule up to 500 pm long and 1 pm diameter, armed throughout with spirals of short bristles; in hydranth and scattered throughout coenosarc; easily discharged. Colonies dioecious, gonotheca arising on a long pedicel from deep within hydrotheca on stem and proximal parts of hydrocladia; male gonotheca long, straight to curved, elliptical, and llattened in frontal aspect; narrow in side view with up to eight deep, irregular corrugations fading proximally, corrugations shallower in frontal view, perisarc very thick; orifice circular, small, on a very short neck sealed by a plug of tissue. Female gonothecae on different stems from male, lenticular, shorter than male, with five or six irregular corrugations, orifice small, circular, on a short neck, sealed by a plug of tissue; gonophore one large or two small ova covered by a thick gelatinous pellicle, almost filling gonothecal cavity. Colour. Living colonies reddish-purple. Remarks. The above description is a composite from colonies collected by the author from several localities in Bass Strait, Victoria, Australia. Apical stolonisation from stems and distally from the hydrocladia commonly occur in some colonies, the stolons often giving rise to new stems. While there is some variation in diameter of the hydrothecal rim and in length of the adcauline wall throughout the range of colonies examined, the free wall is always one quarter to one fifth of total wall length. There is no difference in size of hydrothecae between male and female stems. Large nematocysts are very abundant in most colonies. Table 5. Measurements (pm) of specimens of Synthecium patulum from Bass Strait, southern Australia. Eastern Bass Western Port Port Phillip Bay Range of Strait dimensions Hydrorhiza, diameter Stem 300 300 maximum diameter 280 - 340 240 - 300 260 - 300 240 - 340 length of internode 2,060 -2,160 1,640 - 1,800 1,780 - 1,900 1,640 -2,160 diameter at node 260 240 - 260 240 - 280 240 - 280 Hydrocladium intcmodc length 608 - 672 552 - 560 600 - 672 552 - 672 diameter at node 152 - 168 144 - 152 152 - 168 144 - 168 Hydrolheca length of adnate adcauline wall 480 - 496 408 - 440 440 - 480 408 - 496 length of free adcauline wall 160 112 - 120 96 - 120 96 - 160 diagonal length abcauline wall 416 - 440 352 - 68 384 - 400 352 - 440 diameter at mtugin 192 - 216 144 128 - 184 128 - 216 Gonotheca width of male (lateral view) 360 - 460 360 - 460 width of male (frontal view) 560 - 600 560 - 600 length of female 960 - 1,100 1,100 - 1,300 960 - 1,300 width of female (lateral view) 480 360 - 400 360 - 480 width of female (frontal view) 520 - 620 460 - 560 460 - 620 43 J.E. Watson .2 c H Cfl £?. O * oc 3 E « "H. E K a ^1/5 0 E td S E •< -a o .2 S •n - C3 ui ‘5 ^ E O Ua o E o .3 . 56. Macrorhynchia quadriarmata sp. nov. A, holotype colony from Beagle Gulf. B, hydrotheca with truncated mesial nematotheca. C. hydrotheca with elongate mesial nematotheca. Fig. 57. A-C, Clytia ?warrenv. A, B, hydrothecae from colony from Beagle Gulf. D, anterior view of hydrotheca. E, apophysis of stem with proximal hydrocladial internode and C, gonotheca from same colony. D, E, Clytia linearis: D, colony from Beagle oblique hinge joint. F, cauline nematothecae. G, phylactocarp. H, distal end of phylactocarp Gulf. E, hydrotheca and gonotheca from same colony. Scale bars: A - C, E, 300 with three nematothecae. Scale bars: A, 10 mm, B, C, D, 200 pm; E - H, 300 pm. pm; D, 1,000 pm. Beagle Gulf and Darwin Harbour hydroids length of hydrothecal pedicel 200 - 880 diameter annulated hydrothecal pedicel 48 - 64 Hydrotheca length, diaphragm to margin 624 diameter at diaphragm 72 - 100 diameter at margin 336 Gonotheca length, excluding pedicel 488 - 640 distal diameter 120 - 200 Remarks. The specimens agree with dimensions of C. linearis given by Gibbons and Ryland (1989) and Migotto (1996). The perisarc is so frail that most hydrothecae collapse in mountant. The slight narrowing behind the distal end of the gonotheca shown by some authors (e.g. Gibbons and Ryland 1989, Migotto 1996) is not evident in the present material. There is a remarkable range in size of hydrothecae and if intergradations were not present on the same stem the extremes in size could be easily mistaken for two different species. The figured hydrotheca (Fig. 57E) was one of the largest found on the colonies. Distribution. Tropical to warm temperate oceans and Fiji (Gibbons and Ryland 1989), Brazil (Migotto 1996). Not previously recorded from Australia. Clytia sp. 1 Record and material. Infertile colony on aglao- pheniid hydroid, Stn. 61. Remarks. The few remaining hydrothecae are too badly crushed for identification. Clytia sp. 2 Record and material. Infertile colony on bryozoan, Stn. 38. Remarks. Stems simple, unbranched, annulated proximally and distally but the few remaining campa- nulate hydrothecae are crushed beyond recognition. ACKNOWLEDGMENTS I thank the Parks and Wildlife Commission of the Northern Territory and Director of the Museums and Art Galleries of the Northern Territory, Darwin, for providing the Beagle Gulf collection for examination; the CSIRO Centre for Research on Introduced Marine Pests for the invitation to participate in their 1998 survey of the Port of Darwin, the Museums and Art Galleries of the Northern Territory for providing facilities for SCUBA diving and laboratory space in 1998 and 1999, and the Zoologische Staalssamiung, Munich, Germany for loan of type and voucher material; Dr Dale Calder of the Royal Ontario Museum Canada for taxonomic advice. Dr Barry Russell, Dr Philip Alderslade and Dr Richard Willan of the Museums and Art Galleries of the Northern Territory and Dr Chad Hewitt of the CSIRO Centre for Research on Introduced Marine Pests for invaluable personal support. In particular, 1 sincerely thank Dr Wim Vervoort for his unfailing advice on difficult nomenclatural and taxonomic problems and constructive criticism of the manuscript. Lastly, I thank Karen Gowlett-Holmes of CSIRO for underwater companionship during field work in Darwin Harbour. REFERENCES Agassiz, A. 1862. Contributions to the natural history of the United States of America. Second monograph. Vol. IV. Little, Brown: Boston. Agassiz, A. 1865. North American Acalephae. Illustrated Catalogue of the Museum of Comparative Zoology at Harvard College. No. 2. Sever and Francis: Cambridge, Massachusetts. Allman, G. J. 1872. A monograph of gymnoblastic or tubularian hydroids. Part 2. Ray Society: London. Allman, G. J. 1877. Report on the Hydroida collected during the exploration of the Gulf Stream by L. F. de Pourtales, assistant United States Coast Survey. Memoirs of the Museum of Comparative Zoology at Harvard College 5(2): 1 - 66 . Allman, G. J. 1883. Report on the Hydroida dredged by H.M.S. Challenger during the years 1873-76. Part 1. Plumularidae. Report on the Scientific Results of the Voyage of H.M.S. Challenger (Zoology) 7(20): 1-54. Allman, G. J. 1885. Description of Australian, Cape, and other Hydroida. mostly new, from the collection of Miss H. Gatty. Journal of the Linnean Society, (Zoology) 19: 132-160. Allman, G. J. 1888. Report on the Hydroida dredged by H.M.S. Challenger during the years 1873-76. Part H. The Tubularinae, Coryniorphinae, Campanularinae, Sertularinae and Thalamophora. Report on the Scientific Results of the Voyage of H.M.S. Challenger 1873-76 (Zoology) 23: 1- 90. Bale, W. M. 1882. On the Hydroida of south-eastern Australia, with descriptions of supposed new species, and notes on the genus Aglaophenia. Journal of the Microscopical Society of Victoria 2: 15-48. Bale, W. M. 1884. Catalogue of the Australian hydroid zoophytes. Australian Museum: Sydney. Bale, W. M. 1887. The genera of the Plumulariidae with observations on various Australian hydroids. Proceedings of the Royal Society of Victoria 23: 73-110. Bale, W. M. 1888. On some new and rare Hydroida in the Museum collection. Proceedings of the Linnean Society of New South Wales 3(2): 745-799. Bale, W. M. 1894. Further notes on Australian hydroids, with dc.scriptions of some new species. Proceedings of the Royal Society of Victoria, (n.s.) 6: 93-117. Bale, W. M. 1913. Further notes on Australian hydroids. H. Proceedings of the Royal Society of Victoria 26: 114-147. Bale, W. M. 1914a. Report on the Hydroida collected in the Great Australian Bight and other localities. Biological Results of the Fishing Experiments carried on by F.I.S. "Endeavour" 1909-1914 2(\): \-62. Bale. W. M. 1914b. Report on the Hydroida collected in the Great Australian Bight and other localities. Biological Results of the Fishing Experiments carried on by F.I.S. “Endeavour” /909-/9/4 2(4): 164-188. Bale, W. M. 1915. Report on the Hydroida collected in the Great Australian Bight and other localities. Biological Results of 75 J.E. Watson the Fishing Experiments carried on by F.I.S. “Endeavour” 1909-1914 3(5): 241-336. Bale, W. M. 1919. Further notes on Australian hydroids - IV. Proceedings of the Royal Society of Victoria, (n.s.) 31(2): 327-361. Bale, W. M. 1924. Report on some hydroids from the New Zealand coast with notes on New Zealand Hydroids generally, supplementing Farquhar's list. Transactions of the New Zealand Institute 55: 225-268. Bandel, K. and Wedler, F. 1987. Hydroid, Amphineuran and Gastropod zonation in the littoral of the Caribbean Sea, Colombia. Senckenhergiana Maritima 19(1-2): 1-129. Bartlett. G. C. 1907. Notes on hydroid zoophytes. Geelong Naturalist, 2nd series 3(4): 35-45. Bedot, M. 1921. Notes systematiques sur les Plumularides. I. Revue Suisse de Zoologie 28: 311-356. Bedot. M. 1922. Les caracteres sexuels secondaires des Plumularides. Revue Suisse de Zoologie 29(4): 147-66. Bennitt. R. 1922. Additions to the hydroid fauna of the Bermudas. Proceedings of the American Academy of Arts and Sciences 57: 241-59. Billard. A. 1904a. Contribution a I’dtude des hydroides (multiplication, rdgdneration, greffes, variations). Annales des Sciences Naturelles, Zoologie (Series 8) 20: 1-251. Billard, A. 1904b. Hydroides recoltds par M. Ch. Gravier dans le golfe de Tadjourah. Bulletin du Museum d'Histoire Naturelle 10: 480-485. Billard, A. 1906 Hydroides. Expeditions Scientifiques du Travailleur et du Talisman. 8: 153-243. Billard, A. 1907. Hydroides de Madagascar et du sud-est de r Afrique. Arc/iivw de Zoologie Experimentale et Generale 7(4): 335-396. Billard, A. 1909. Revision des especes types d'hydroides de la collection Lamouroux conservde a I’lnstitut Botanique de Caen. Annales des Sciences Naturelles {Zoologie) 9: 307- 336. Billard, A. 1910. Revision d'une partie de la collection des hydroides du British Museum. Annales des Sciences Naturelles (Zoologie) 11(9): 1-67. Billard. A. 1911. Note preliminaire sur les especes nouvelles de Plumulariidae de I'expedition du Siboga. Archives de Zoologie experimentale et generale 8(5). notes et revue 3: Ixii-lxxi. Billard, A. 1913. Les hydroides de I’expedition du Siboga. 1. Plumulariidae. Siboga Expedition, 7a: 1-115. Billard, A. 1919. Note sur quelques especes nouvelles de Sertularella de I’expedition de “Siboga”. Archives de Zoologie experimentale et generale 58, notes et revue 1: 18-23. Billard, A. 1922. Note critique sur quatres espbces Sertularella. Revue Suisse de Zoologie 30: 103-113. Billard, A. 1924. Note critique sur divers genres et especes d’hydroides avec la description de trois especes nouvelles. Revue Suisse de Zoologie 31: 53-74. Billard, A. 1925. Les hydroides de I'expddition du Siboga. II. Synthecidae et Sertularidae. Siboga Expedition 7b: 117- 232. Billard, A. 1929. Note sur un genre nouveau et quelques especes d’Halecidac. Bulletin de la Societe Zoologique de France 54: 305-307. Billard, A. 1931. Hydroides de I’expedition du “Sylvana”. Bulletin du Museum National d’Histoire Naturelle 3(2): 248-250. Billard, A. 1933. Les hydroides des Golfes des Suez et d’Akaba. Memoires Presentes d I'Institut d'Egypte 21: 1-30 Billard, A. 1941. Note sur les Hydroides Hebella costata (Bale) et H. corrugata (Thornely). Bulletin de la Societe Zoologique de France 65:13-15. Billard, A. 1942. Note sur quelques especes et varietds nouvelles des genres Hebella et Hebellopsis (Hydroides). Bulletin de la Societe Zoologique de France 67: bl-lQ. Blackburn, M. 1938. The Hydrozoa of the Sir Joseph Banks Islands. Proceedings of the Royal Society of Victoria, (n.s.) 50(2): 312-328. Blackburn, M. 1942. A systematic list of the Hydroids of South Australia with a summary of their distribution in other seas. Transactions of the Royal Society of South Australia 66(1): 104-118. Boero, F. 1980. Life cycles of hydroids and hydromedusae: some cases of difficult interpretation. Memorie di Biologia Marina e Oceanograpliia 10 (SuppL): 141-147. Boero, F., and Bouillon. J. 1989. An evolutionary interpretation of anomalous stages in the life cycles of some Lepto- medusae (Cnidaria). In: J. S. Ryland and P. A. Tyler, (eds). Reproduction, Genetics and Distribution of Marine Organisms. Pp 37-41. Proceedings of the XXIII European Marine Biology Symposium. Boero, F., and Fresi, E. 1986. Zonation and evolution of a rocky bottom hydroid community. Ptibblicazioni della Stazione Zoologica di Napoli, Marine Ecology 7(2): 123-150. Boero, F., Bouillon, J., and Kubota, S. 1997. The medusae of some species of Hebella Allman, 1888, and Anthohebella gen. nov. (Cnidaria, Hydrozoa, Lafoeidae), with a world synopsis of species. Zoologische Verhandeligen, Leiden 310: 1-53. Borradaile, L. A. 1905. Hydroids. In: Gardiner, J. S. (ed.) The fauna and geography of the Maidive and Laccadive Archipelagoes. Vol 2. pp 836-845. Cambridge University Press: Cambridge. Bourne, G. C. 1890. Notes on the hydroids of Plymouth. Journal of the Marine Biological Association of the U.K. (New Series) 1: 391-398. Briggs, E. A. 1914. Hydrozoa from one hundred fathoms, seven miles east of Cape Pillar, Tasmania, Records of the Australian Museum 10(10): 285-302. Briggs, E. A. 1918. Descriptions of two new hydroids and a revision of the hydroid-fauna of Lord Howe Island. Records of the Australian Museum 12(3): 27-47. Briggs, E. A., and Gardner. V. E. 1931. Hydroida. Scientific Reports of the Great Barrier Reef Expedition 4(6): 181-196. Broch, H. 1910. Die Hydroiden der arktischen Meere. Fauna Arctica 5: 127-248. Buchanan,). B. 1957. The hydroid fauna of the Gold Coast. Revue de Zoologie et de Botanique Africaines 56: 349-372. Busk, G., 1852. An account of the Polyzoa and Sertularian zoophytes collected in the voyage of the Rattlesnake on the coast of Australia and the Louisiade Archipelago. In: MacGillivray, J. Narrative of the voyage of H.M.S. Rattlesnake, commanded by the late Captain O. Stanley, R.N., F.R.S. during the years 1846-1850. Vol. 1, Appendix 4. T. and W. Boone: London. Calder. D. R. 1983. Hydroida from estuaries of South Carolina, U.S.A: families Sertulariidae and Plumulariidae. Proceedings of the Biological Society of Washington 96( 1): 7-28. Calder, D. R. 1991. Shallow-water hydroids of Bermuda. The Thecatae, exclusive of Plumularioidea. Life Science Contributions, Royal Ontario Museum 154: 1-140. Calder. D. R. 1997. Shallow-water hydroids of Bermuda - (Superfamily Plumularoidca). Life Sciences Contributions, Royal Ontario Museum 161: 1-86. 76 Beagle Gulf and Darwin Harbour hydroids Calder, D. R.. and Vervoort, W. 1998. Some hydroids (Cnidaria; Hydrozoa) from the Mid-Atlantic Ridge, in the North Atlantic Ocean. Zoologixche Verhandelingen, Leiden 319: 1-65. Campenhausen, B. von. 1896. Hydroiden von Ternate. In: Kukenthal, W. Ergehnisse einer zoologischen Forschiingsreise in den Molluken imd Borneo. Theil 2. Abhandlungen herausgegeben von der Senckenbergischen Naturfonschenden Gesellschaft. 23(2): 297-320. [Published separately as; Beitrage zur Systematik der Hydroiden. Frankfurt am Main. 1897.] Ciamician, J. 1880. Ueber Lafoea parasitica n. sp. Zeitschrift fiir Wissenschaftliche Zoologie 33: 673-676. Clarke, S. F. 1879. Report on the Hydroids collected during the exploration of the Gulf-Stream and Gulf of Mexico, by Alexander Agassiz, 1877-1878. Bulletin of the Museum of Comparative Zoology at Harvard College 5; 239-252. Clarke, S. F. 1907. The hydroids. In: Reports on the scientific results of the expedition to the eastern tropical Pacific, in charge of Alexander Agassiz, by the U.S. Fish Commission steamer “Albatross”, from October 1904, to March, 1905, Lieut.-Commander L. M. Garrett, U.S.N. commanding. Vlll. Memoirs of the Museum of Comparative Zoology, Harvard 35{\): 1-18. Colin, P. L. 1978. Caribbean reef invertebrates and plants. A field guide to the invertebrates and plants occurring on coral reefs of the Caribbean, the Bahamas and Florida. T.F.H. Publications: Jersey City, USA. Congdon, E. D. 1907. The hydroids of Bermuda. In: Contributions from the Zoological Laboratory, Syracuse University; also contribution from the Bermuda Biological Station for Research, 9. Proceedings of the American Academy of Arts and Sciences 42(8); 461-485. Cooke, W. J. 1977. Order Hydroida. In: Devaney, D. M. and Eldridge, L. G. (eds) Reef and shore fauna of Hawaii. Section I: Protozoa through Ctenophora. Revised edition. Special Publications of the Bernice P. Bishop Museum 64(1): 71-104. Cornelius, P. F. S. 1982. Hydroids and medusae of the family Campanulariidae recorded from the eastern North Atlantic, with a synopsis of genera. Bulletin of the British Museum (Natural History) (Zoology Scries) 42(2): 37-148. Cornelius P. F. S. 1995a. North-west European thecate hydroids and their medusae. Part 1. Introduction, Laodiceidae to Haleciidae. In; Barnes, R.S.K. and Crothers, J.H. (eds) Synopses of the British Fauna (New Series) 50: i-vii, I- 347. Cornelius P. F. S. 1995b. North-west European thecate hydroids and their medusae. Part 2. Sertulariidae to Campanulariidae. In: Barnes, R.S.K. and Crothers, J.H. (eds) Synopses of the British Fauna (New Series) 50: i-vii, 1-386. Cotton, B. C., and Godfrey, F. K. 1942. Idiellana, a new name for the preoccupied genus Idiella Stechow (Coelenterata - family Sertulariidae). Records of the South Australian Museum 7(2): 234. Crossland, C. 1904. The marine fauna of Zanzibar and British East Africa, from collections made by Cyril Crossland in the years 1901 and 1902. The Polychaeta, part 3, with which is incorporated the account made in the Maidive Archi¬ pelago in the year 1899. Proceedings of the Zoological Society of London 1: 287-330. Da Cunha, A. X. 1950. Nova contribuigao para o estudo dos Hidropdiipos das costas dc Portugal (Coiccga do Museu Bocage). Archivos do Museu Bocage 21:121-144. Dawydoff, C. N. 1952. Contribution a I’etude des invertebres de al faune marine bentique de I’lndochine. Bulletin biologique de la France et de la Belgique 37, suppl: 1- 158. (Coelenterata: 52-78). Deevey, E. S. 1954. Hydroids of the western Gulf of Mexico. In; Galtsoff, P. S. (ed.) Gulf of Mexico. Its origin, waters and marine life. Fishery Bulletin of the Fish and Wildlife Service of the U.S. 55: 267-272. Dollfus, R. Ph. 1933. Resumd analytique des Mdmoires T. XXL 1-6, presentds a I’lnstitut d’figypte dans la seance du 14 novembre 1932. Bulletin de Institut d'Egypte 15: 125- 157. Ellis, J. & Solander, D. 1786. The natural history of many curious and uncommon zoophytes, collected from various parts of the globe. White: London. Eschscoltz, J. F. 1829. System der Acalephen. Eine ausfiihrliche Beschreibung aller medusenartigen Strahltiere. Ferdinand Diimmler: Berlin. Esper, E. J. C. 1793. Die Pfianzenthiere in Abbildungen nach der Natur mit Farben erleuchtet. Vol 2. Bauer and Raspe: Niirnberg. Farquhar, H. 1896. List of the New Zealand Hydroida. Transactions of the New Zealand Institute 28: 459-468. Fleming, J. 1828. A history of British animals. J. Duncan, Bell and Bradfute: Edinburgh and London. FIdrez-GonzdIez, L. 1983. Inventario preliminar de la fauna hydroide de la Bahia de Cartagena y areas adyacentes. Boleti'n Museo del Mar, Bogota 11: 112-140. Fraser, C. M. 1938. Hydroids of the 1934 Allan Hancock Pacific expeditions. Allan Hancock Pacific Expeditions 4(1): 1- 104, pis 1-15. Fraser. C. M. 1943. Distribution records of some hydroids in the collection of the Museum of Comparative Zoology at Harvard College, with description of new genera and new species. Proceedings of the New England Zoological Club 22: 75-98. Fraser, C. M. 1944. Hydroids of The Atlantic Coast of North America. University Press: Toronto. Garci'a-Corrales, P., A. Aguirre Inchaurbe and D. GonzdIes Mora. 1980. Contribucion al conocimienlo de los hidrozoos dc las costas Espanolas. Parte HI. Sertulariidae. Boleti'n del Instituto Espahol de Oceanografta 6(296): 1-67. Genzano, G. N. 1992. La fauna de hidropolipos (Cnidaria) del litoral de Buenos Aires, Argentina. 1. Neotrdpica 38(100): 141-148. Gibbons, M. J., and Ryland, J. S. 1989. Intertidal and shallow water hydroids from Fiji. 1. Athecata to Sertulariidae. Memoirs of the Queensland Museum 27(2): 371-432. Gili, J-M. 1982. Fauna de cnidaris de les illes Modes. Treballs de la Institucid Catalana d'Historia Natural 10: 1-175. Gili, J-M., and Ballesteros, E. 1991. Structure of cnidarian populations in Mediterranean sublittoral benthic com¬ munities as a result of adaptation to different environmental conditions. In: Ros, J. D. and Prat, N. (eds) Homage to Ramon Margalef; or, why is there pleasure in studying nature. Oecelogi'a aquatica 10: 243-254. Gili, J-M. and Castelld, G. 1985. Hydroids of the north coast of Cape Creus (N. E. Spain). Miscelldnia Zooldgica 9: 7-24. Gmelin, J. F. 1791. In: Linne, C. Systema Naturae, ed. 13. Vermes. 1 (6): 3021-3910. Lip.siae. Gordon, D. P., d'Hondt, J-L., Watson, J. E. and. Jones, M. S., 1998. Discovery of the lost type of Caberea (Bryozoa) and the identity of Caberea pinnatu Lamouroux, 1816 (Hydroida). Journal of Natural History 32: 405-418. 77 J.E. Watson Gravely, F. H. 1927. The littoral fauna of the Krusadai Island in the Gulf of Manaar. Hydrozoa. Bulletin of the Madras Government Museum (n.s. Natural History) 1: 7-20. Gravier-Bonnet, N. 1979. Hydraires semi-profonds de Ma¬ dagascar (Coelenterata Hydrozoa), dtude systematique et ecologique. Zoologische Verliandelingen, Leiden 169: 3- 76. Gray, J. E., 1848. List of the specimens of British animals in the collection of the British Museum. Part I. Centroniae or radiated animals. British Museum: London. Hadzi, J. 1913. Poredbena hidroidska istraizivanja. I. Hehella parasitica (Ciamician). (S. dodat kom: Hebella brochi g. et sp. n.: Hebella (?) gigas Pieper). Rad Jugoslavenske Akademije Znanosti i Umjetnosti 198: 105-210. Hargitt, C. W. 1924. Hydroids of the Philippine Islands. Philippine Journal of Science 24: 467-507. Hartlaub, C. 1901. Revision dcr Senularella-Anen. Abhandlungen ausdem Gebiete der Naturwissenschaften, Hamburg 16(2): 1-143. Hartlaub, C. 1905. Die Hydroiden dcr magalhaensischen Region und der chilenischen KUste. Zoologische Jahrbucher (Suppl.) Zool. 6(3): 497-714. Hincks, T. 1868. A history of the British hydroid zoophytes. John Van Voorst: London. Hincks, T. 1874. On deep-water Hydroida from Iceland. Annals and Magazine of Natural History B(4): 146-153. Hirohito, Emperor of Japan, 1969. Some hydroids from the Amakusa Islands. Publications of the Biological Labo¬ ratory of the Imperial Household, Tokyo 9: 1-32. Hirohito, Emperor of Japan, 1983. Hydroids from 1/u Oshima and Niijima. Publications of the Biological Laboratory of the Imperial Household. Tokyo. (1983) 6: 1-83. Hirohito, Emperor of Japan, 1995. The hydroids of Sagami Bay. H. Publications of the Biological Laboratory of the Imperial Household, Tokyo, 1995: i-vii, 1-355 (English text); 1-244 (Japanese text), edited and annotated by M. Yamada. Hodgson, M. 1950. A revision of the Tasmanian Hydroida. Papers and Proceedings of the Royal Society of Tasmania 1949: 1-65. Inaba, M. 1890. Hydroids collected at Misaka, Miura, Soshu. Zoological Magazine 2: 245-341. Inaba, M. 1892. Soshu. Miura, Misaki ni oide edaru Hydroidea (Hydroids collected at Miura and Misaki in Soshu). Zoological Magazine 4: 93-101. Jaderholm, E. 1896. Ueber aussereuropiiische.Hydroiden des zoologischen Museums der Universitat Upsala. Bihang till Kungliga Svenska Vetenskabs-Akademiens Handlingar 21 (Afd4, 6): 1-20. Jaderholm, E. 1903. Aussereuropaischen Hydroiden im schwedischen Reichsmuseum. Arkiv for Zoologi 1: 259- 312 . Jaderholm, E. 1916. Results of Dr. E. Mjobcrgs Swedish scientific expeditions to Australia 1910 - 1913. XH. Hydroiden. Kungliga Svenska Vetenskaps-Akademiens Handlingar 52(12): 1-9. Jaderholm, E. 1917. Hydroids from the South Seas. Redogdreise for Norrkopings Hdgre Allmdnna Ldroverk Ldsaret 1916- 1917: 1-25. Jaderholm. E. 1919. Zur Kenntnis der Hydroidenfauna Japans. Arkiv for Zoologi 12 (9): 1-34. Jaderholm, E. 1920. On some exotic hydroids in the Swedish Zoological State Museum. Arkiv for Zoologi 13(3): 1-11. Jaderholm, E. 1923. Notes on hydroids from the Great Ocean. Meddelanden frdn Gdteborgs Musei Zoologiske Afdelning 27: 1-6. Jarvis, F. E. 1922. The hydroids from the Chagos, Seychelles, and other islands off the coasts of east Africa and Zanzibar. Transactions of the Linnean Society of London (Zoology) 18: 331-360. Jickeli. C. F. 1883. Der Bau der Hydroidpolypen. Morphologisches Jahrbuch, Leipzig 8: 373-416, 580-680. Johnston. G. 1837. A catalogue of the zoophytes of Berwickshire. History of the Berwickshire Naturalists’ Club 1: 107-108. Johnston, G. 1838. A history of the British hydroid zoophytes, Lizars, Edinburgh. Kirchenpauer, G. H. 1872. Ueber die Hydroidenfamilie Plumulariidae, einzelne Gruppen derselben und ihre Fruchtbehalter. 1. Aglaoplienia Lx. Abhandlungen aus dem Gebiete der Naturwissenschaften, Hamburg 5(2-3): 1-58. Kirchenpauer, G. H. 1876. Ueber die Hydroidenfamilie Plumularidae, einselne Gruppen derselben und ihre Fruchtbehalter 11. Plumularia und Nemertesia. Abhandlungen aus dem Gebiete der Naturwissenschaften. Hamburg 6: I- 59. Kirkpatrick, R. 1890. Reports on the zoological collections made in Torres Straits by Professor A. C. Haddon, 1888-1889. Hydroida and Polyzoa. Proceedings of the Royal Dublin Society (n.s.) 6: 603-626. Lamarck, J. B. P. A. de, 1816. Histoire natiirelle des animaux sans vertebres. Tome 2. Verdifere: Paris. Lamarck, J. P. B. A. de, 1837. Histoire natiirelle des animaux sans vertebres. 3rd edition. Bruxelles. Lamouroux, J. V. F. 1812. Extrait d'un memoire des polypiers coralligenes non entierement pierreux. Nouveau Bulletin des Sciences par la Societe Philomatique de Paris 3: 181-188. Lamouroux, J. V. F. 1816. Histoire des polypiers coralligenes flexibles, vulgairement nomm^s zoophytes. F. Poisson: Caen, France. Lamouroux, J. V. F. 1821. Exposition methodique des genres de I’ordre des polypiers, avec leur description et celle des principales especes, figures dans 84 planches: les 63 premieres appurtenant a I'historie natiirelle des zoophytes d'Ellis et Solander. Agasse: Paris. Lamouroux, J. V. F. 1824. Polypiers llexibles. In: Quoy, J. R. C. and Gaimard, J. P. Voyage aiitour du monde (execute sur les corvettes de S.M. I'Uranie et la Physicienne, pendant les annees 1817, 1819 et 1820, par M. Louis de Freycinet). Zoologie. Pillet, ATne: Paris: Leloup, E. 1932. Une collection d'hydropolypes appartenant a rindian Museum de Calcutta. Records of the Indian Museum 34: 131-170. Leloup, E. 1934. Note sur les hydropolypes de la Rade de Villefranche-sur-Mer (France). Bulletin du Musee royal d'Histoire Natiirelle de Belgique 10(31): 1-18. Leloup. E. 1935. Hydraires calyptoblastiques des Indes Occidentalcs. Memoires du Musee royal d'Histoire Natiirelle de Belgique 2(2): 1-73. Leloup, E. 1937. Hydropolypes et scyphopolypes sur les cotes de rindochine fran^aise. Bulletin du Mia^e royal d’Histoire natiirelle de Belgique 2{ 12): 1-3. Leloup, E. 1938. Quelques hydropolypes de la Bale de Sagami, Japon. Bulletin du Musee Royal d’Histoire Natiirelle de Belgique 14(28): 1-2. Leloup, E. 1939. Notes sur quelques Hydropolypes exotiques. Bulletin du Musee Royal d’Histoire natiirelle de Belgique 15(51): 1-19. Lendenfeld, R. von, 1885a. The Australian Hydromedusae. HI. Proceedings of the Linnean Society of New South Wales 9: 401-420. 78 Beagle Gulf and Darwin Harbour hydroids Lendenfeld, R. von, 1885b. The Australian Hydromedusae. IV. Proceedings of the Linnean Society of New South Wales 9: 467-492. Levinsen, G. M. R. 1913. Systematic studies on the Sertulariidae. Videnskahelige Meddeletser fra Dansk Naturhistorisk Forening i Kjohenhavn 64; 249-323. Linnaeus, C. 1758. Systema naturae. Ed. 10. Holmiae. Mammen, T. A. 1965. On a collection of hydroids from South India. 11. Suborder Thecata (excluding family Plumulariidae). Journal of the Marine Biological Association of India 7: 1-57. Mammen, T. A. 1967. On a collection of hydroids from South India. 111. Family Plumulariidae. Journal of the Marine Biological Association of India 7(2): 291-324. Marktanner-Turneretscher, G. 1890. Die Hydroiden des K.K. Naturhistorischen Hofmuseums. Annalen des K. K. Naturhistorischen Hofmuseums 5; 195-286. Mayal, E. M. 1983. Distribu^ao de hidroides (Hydrozoa, Thecata) na costa do estado de Pernambuco. Boletim de Zoologia, Universidade de Sao Paulo 6: 1-13. McCrady, J. 1859. Gymnopthalmata of Charleston Harbor. Proceedings of the Elliott Society of Natural History 1: 103-221. Medel, M. D., and Lopez-Gonzalez, P. J. 1996. Updated catalogue of the hydrozoans of the Iberian Peninsula and Balearic Islands, with remarks on zoogeography and affinities. In: Piraino,S., Boero, F.. Bouillon, J., Cornelius, P. F. S. and Gili (cds) J. M. Advances in Hydrozoan Biology. Scientia Marina 60(1): 183-209. Medel, M. D., and Vervoort, W. 1995. Plumularian hydroids (Cnidaria: Hydrozoa) from the Strait of Gibraltar and nearby seas. Zooligische Verhandelingen, Leiden 300: 1-72. Medel, M. D., and Vervoort, W. 1998. Atlantic Thyroscyphidae and Sertulariidae (Hydrozoa, Cnidaria) collected during the CANCAP and Mauritania-II expeditions of the National Museum of Natural History, Leiden, The Netherlands. Zooligische Verhandelingen, Leiden 320: 1-85. Mergner, H., and Wcdlcr, E. 1977. Uberdie Hydroidpolypenfauna des Roten Meeres und seiner Ausgange. "Meteor” Forschungs-Ergehnisse D 24: 1-32. Migotto, A. E. 1996. Benthic shallow water hydroids (Cnidaria, Hydrozoa) of the coast of Sao Sebastiao, Brazil, including a checklist of Brazilian hydroids. Zoologische Verhandelingen, Leiden 306: 1-125. Millard, N. A. H. 1958. Hydrozoa from the coasts of Natal and Portuguese East Africa. Part 1. Calyptoblastca. Annals of the South African Museum 44: 165-226. Millard, N. A. 11. 1962. The Hydrozoa of the south and west coasts of South Africa. Part 1. The Plumulariidae. Annals of the South African Museum 46: 261-319. Millard, N. A. H. 1968. South African hydroids from Dr. T. H. Mortensen’s Java-South Africa Expedition, 1929-1930. Videnskahelige Meddelelser fra Dansk Naturhistorisk Forening. 131: 251-288. Millard, N. A. H. 1971. Hydrozoa. In: Zinderen Bakker, E. M. van, Winterbottom, J. M. & Dyer, R. A., Eds. Marion and Prince Edward Islands: 396-408. Cape Town: A. A. Balkema. Millard, N. A. H. 1975. Monograph of the Hydroida of South Africa. Annals of the South African Museum 68: 1-513. Millard, N.A.H. 1978. The geographical distribution of southern African hydroids. Annals of the South African Museum 74 (6); 159-200. Millard, N. A. H. and Bouillon, J. 1973. Hydroids from the Seychelles (Coelenterata). Annalies dti Musee Royal de I'Afrique Centrale serie In-8“, Sciences Zoologiques 206: 1-106. Millard, N. A. H. and Bouillon, J. 1974. A collection of hydroids from Mofambique, East Africa. Annals of the South African Museum 65: 1-40. Millard, N. A. H. and Bouillon, J. 1975. Additional hydroids from the Seychelles. Annals of the South African Museum 69(1): 1-15. Mulder, J. F., and Trebilcock, R. E. 1914a. Victorian Hydroida with description of new species. Geelong Naturalist 6(1): 6-47. Mulder, J. F., and Trebilcock, R. E. 1914b. Victorian Hydroida with description of new species. Geelong Naturalist 6(2): 38-47. Neppi, V. 1921. Nuove osservationi sui polipi idroidi del Golfo di Napoli. Puhblicazioni della Stazione Zoologica di Napoli 3:1-31. Nutting, C. C. 1900. American hydroids. Part 1. The Plumularidae. Smithsonian Institution, United States National Museum, Special Bulletin A{\): 1-285. Nutting, C. 1904. American hydroids. Part 11. The Sertularidae. Smithsonian Institution, United States National Museum, Special Bulletin 4 (2): 1-325. Nutting, C. C. 1905. Hydroids of the Hawaiian Islands collected by the steamer “Albatross” in 1902. Bulletin of the United States Fish Commission for 1903: 931-959. Nutting, C. C. 1927. Report on Hydroida collected by the United States steamer “Albatross” in the Philippine region, 1907-1910. Bulletin of the United States National Museum 100: 195. Oken, L. 1815. Okens Lehrbuch der Naturgeschichte. Ill, Theil, Zoologie. Vol. I. Oken: Jena. Pena Cantero, A. L., Garcia Carrascosa, A. M., and Vervoort, W. 1998. On the species of Filellum Hincks, 1868 (Cnidaria: Hydrozoa) with description of a new species. Journal of Natural History 32: 297-315. Pennycuik, P. R. 1959. Faunistic records from Queensland. Part V. Marine and brackish water hydroids. Papers of the Department of Zoology of the University of Queensland 1; 141-210. Picard, J. 1952. Les Hydrozoaires des herbiers de Zosteracees des cotes fran?aises de la Mediterranee. Vie et Milieu (Suppl.) 2: 217-223. Pictet, C. 1893. Etude sur les Hydraires de la Bale d’Amboine. Revue Suisse de Zoologie I: I -64. Ralph, P. M. 1958. New Zealand thecate hydroids. Part 11. Families Lafoeidae, Lineolariidae, Haleciidae and Syntheciidae. Transactions of the Royal Society of New Zealand 85(2): 301-356. Ralph, P. M. 1961. New Zealand thecate hydroids. Part IV. The family Plumulariidae. Transactions of the Royal Society of New Zealand, Zoology 1(3): 19-74. Ramil, F., and Vervoort, W. 1992. Report on the Hydroids collected by the “BALGIM” expedition in and around the Strait of Gibraltar. Zoologische Verhandelingen, Leiden, 111: 1-262. Redier, L. 1963. Complement a la connaissance d’un hydraire des mers chaudes Lytoscyphus junceus (Allman, 1876). Cahiers du Pacifique 5:18-24. Redier, L. 1966. Hydraires et Bryozoaires. In: Contribution a Tetude des rivages coralliens d’apres les recoltes de Yves Plessis, en Oceanie (Mission Singer-Polignac). Cahiers du Pacifique 9: 78-122. Redier, L. 1967. Revision de la collection du Museum des hydraires de Lamouroux. Bulletin du Museum National d'Histoire Naturelle, 2me serie 39: 381-410. 79 J.E. Watson Rees, W. J., and Thursfield, S. 1965. The hydroid collections of James Ritchie. Proceedings of the Royal Society of Edinburgh 69; 34 - 220. Rees W. J. and Vervoort, W. 1987. Hydroids from the John Murray Expedition to the Indian Ocean, with revisory notes on Hydrodendron, Abietinella, Cryptolaria and Zygophylax (Cnidaria: Hydrozoa). Zoologische Verhandelingen, Leiden 237; 1-209. Rho, B. J. 1967. Marine hydroids from the West and South Sea of Korea. Papers of the Korean Cultural Institute 10; 341- 60. Rho, Boon Jo and Chang, S. R. 1974. On the classification and distribution of the marine benthic animals in Korea. I. Hydroids. Journal of the Korean Research Institute for Better Living, Ewha Womans University 12; 133-158. Riedl, R. 1970. Fauna und Flora der Adria, ed. 2. Verlag Paul Parey; Hamburg and Berlin. Ritchie, J. 1910. The marine fauna of the Mergui Archipelago, Lower Burma, collected by Jas. J. Simpson, M.A., B.Sc., and R. N. Rudmose-Brown, D. Sc., University of Aberdeen, February to May 1907. The Hydroids. Proceedings of the Zoological Society of London 1910; 799- 25. Ritchie, J. 1911. Hydrozoa (hydroid zoophytes and Stylasterina). In: Scientific results of the trawling expedition of H.M.C.S. “Thetis”. Memoirs of the Australian Museum 4(2); 807- 869. Rossi, L. 1950, Celenterati de Golfe di Rapallo (Rivieri Ligure). Bolletino dell’ Istituto e Museodi Zoologia della Universitd di Torino 1(A): 193-235. Ryland, J. S. and Gibbons, M. J. 1991. Intertidal and shallow water hydroids from Fiji. 2. Plumulariidae and Aglaopheniidae. Memoirs of the Queensland Museum 30(3); 525-560. Sars, G. O. 1874. Bidrag til kundskaben om norges hydroider. Forhandlinger i Videnskabs-Selaksbet i Kristiania (1873); 91-150. Schmidt, H. E. 1972. Some new records of hydroids from the Gulf of Aqaba with zoogeographical remarks on the Red Sea area. Journal of the Marine Biological Association of India 13(1); 27-51. Schuchert, P. 1997. Review of the family Halopterididae (Hydrozoa, Cnidaria). Zoologische Verhandelingen, Leiden 309; 1-162. Shepherd, S. A. and Watson J. E. 1969. The sublittoral ecology of West Island, South Australia: 2. The association between hydroids and algal substrate. Transactions of the Royal Society of South Australia 94: 139-146. Spencer, W. B. 1891. A new family of Hydroidea, together with a description of the structure of a new species of Plumularia. Transactions of the Royal Society of Victoria 2: 121-140. SplettstOsser, W. 1929. Beitrage zur Kennlnis der Sertulariiden Thyroscyphus Allman, Cnidoscyplius nov. gen., Parascyphus Ritchie. Zoologische Jahrbiicher, Abteilung fur Systematik, Okologie und Geographic der Tiere 58; 1- 134. Stechow, E. 1909. Beitrage zur Naturgeschichte Ostasiens: Hydroidpolypen der Japanischen Ostkiiste. I. Teil. Athecata und Plumularidae. Abhandlungen der Mathematisch- Physikalischen Klasse der Koninglichen Bayerischen Akademie der Wissenschaften (Suppl.) 1(6): 1-111. Stechow, E. 1913. Hydroidpolypen der Japanischen 0.stkuste. II Teil. Campanularidae, Halecidae, Lafoeidae, Campanulinidae und Sertularidae, nebst Erganzungen zu den Athecata und Plumularidae. Abhandlungen der Mathematisch-Physikalischen Klasse der Koninglichen Bayerischen Akademie der Wissenschaften Supplement- Band 2, 3: 1-162. Stechow, E. 1919. Zur Kenntnis der Hydroidenfauna des Mittelmeeres, Amerikas und anderer Gebiete, nebst Angaben iiber einige Kirchenpauer’sche Typen von Plumulariiden. Zoologische Jahrbiicher, Abteilung fur Systematik, Geographic und Biologic der Tiere 42: 1-172. Stechow, E. 1920. Neue Ergebnisse auf dem Gebiete der Hydroidenforschung. Sitzungsberichte der Gesellschaft fiir Morphologic und Physiologic, Miinchen 31: 9-45. Stechow, E. 1921a. Neue Ergebnisse auf dem Gebiete der Hydroidenforschung H. Miinchener medizinisches Wochenschrift 1921(1): 30. Stechow, E. 1921b. Neue Ergebnisse auf dem Gebiete der Hydroidenforschung HI. Miinchener medizinisches Wochenschrift 1921(28); 897. Stechow, E. 1922. Zur Systematik der Hydrozoen, Stromatoporen, Siphonophoren, Anthozoen und Ctenophorcn. Archiv fiir Naturgeschihte 8S: 141-155. Stechow, E. 1923. Die Hydroidenfauna der Japanischen Region. Journal of the College of Science, Tokyo Imperial University 44(8): 1-23. Stechow, E. 1924. Diagnosen neuer Hydroiden aus Australien. Zoologischer Anzeiger 59; 57-69. Stechow, E. 1925. Hydroiden von West- und Siidwestaustralien nach den Sammlungen von Prof. Dr. Michaclsen und Prof. Dr. Hartmeyer. Zoologische Jahrbiicher Abteilung fiir Systematik, Geographic und Geologic der Tiere 50: 191- 269. Stechow, E. and Miiller, H. C. 1923. Hydroiden von den Aru- Inseln. Abhandlungen herausgegeben von der Senckenberglschen Naturforschenden Gesellschaft 35(4): 459-478. Stranks, T. N. 1993. Catalogue of recent Cnidaria type specimens in the Museum of Victoria. Occasional Papers from the Museum of Victoria 6: 1 -26. Thompson, D'A. W. 1879. On some new and rare hydroid zoophytes (Sertulariidae and Thuiariidae) from Australia and New Zealand. Annals and Magazine of Natural History 3(5); 97-114. Thornely, L. R. 1900. The hydroid zoophytes collected by Dr. Willey in the southern seas. In: Willey, A., Zoological results based on material from New Britain, New Guinea, Loyalty Islands and elsewhere. Part IV. Cambridge University Press: Cambridge. Thornely, L. R. 1904. Report on the Hydroida collected by Professor Herdman, at Ceylon, in 1902. Report to the Government of Ceylon on the pearl oyster fisheries of the Gulf of Manaar, with supplementary reports upon the marine biology of Ceylon Part 2: 107-126. Thornely, L. R. 1908. Reports on the marine biology of the Sudanese Red Sea. 10. Hydroida collected by Mr. C. Crossland from October 1904 to May, 1905. Journal of the Linnean Society (Zoology) 31; 80-85. Thornely, L. R. 1916. Report on the Hydroida and Polyzoa collected by Mr. James Hornell at Okhamandel in Kattiwar in 1905-06. In: Hornell, J. Report to the Government of Baroda on the Marine Zoology of Okhamandel in Kattiwar, Part 11: 147-150. Totton, A. K. 1930. Coelenterata. Part V.- Hydroida. British Antarctic ("Terra Nova”) Expedition, 1910, Natural History Report, Zoology 5(5): 131-252. 80 Beagle Gulf and Darwin Harbour hydroids Trebilcock. R. E. 1928. Notes on New Zealand Hydroida. Proceedings of the Royal Society of Victoria 41: 1-31. Van Gemerden-Hoogeveen, G. C. H. 1965. Hydroids of the Caribbean: Sertulariidae, Plumulariidae and Aglaopheniidae. Studies on the Fauna of Curasao and Other Caribbean Islands 84: 1-87. Van Praet, M. 1979. Les types de polypes d’Hydraires conserves au Museum National d’Histoire Naturelle de Paris. Bulletin du MusOim National d’liistoire Naturelle de Paris 4me serie. Section A, 1: 871-940. Van Soest, R. W. M. 1976. A catalogue of the coelenterate type specimens of the Zoological Museum of Amsterdam. Beaufortia 25(323): 79-95. Vannucci-Mendes, M. 1946. Hydroida Thecophora do Brasil. Archivos de Zoologia de Sao Paulo 4(14): 535-598. Vannucci-Mendes, M. 1949. Hydrozoa do Brasil. Boletim da Faculdade de Filosofia. Ciencias e Letras, Universidade de Sdo Paulo. Zoologia 99(14); 219-266. Vannucci- Mendes, M. 1950. Rcsultados cientificas do Cruzeiro do “Baependi" e do “Vega” a Ihla da Trindade. Boletim do Instituto Paulista de Oceanografia 1(1): 81-96. Versluys, J. 1899. Hydraires calyptoblastiques recueillis dans la mer des Antilles pendant Tune des croisiires accomplies par le Comte R. de Dalmas sur son yacht CHAZALIE. Memoires de la Societe Zeiologiqite de France 12: 29-58. Vervoort. W. 1941. Biological results of the Sncllius Expedition. XI. The Hydroida of the Snellius Expedition. (Milleporidae and Stylasteridae excluded). Temminckia 6: 186-240. Vervoort, W. 1946a. Exotic hydroids in the collections of the Rijksniuseum van Natuurlijke Historic and the Zoological Museum at Amsterdam. Zoologische Mededelingen, Leiden 26:287-351. Vervoort. W. 1946b. Hydrozoa (C 1) A. Hydropolypen. Fauna van Nederlands 14; 1-336. Vervoort, W. 1959. The hydroids of the tropical coast of west Africa. Atlantide Report 5: 211-325. Vervoort, W. 1967. The Hydroida and Chrondrophora of the Israel South Red Sea Expedition, 1962. Bulletin of the Sea Fisheries Research Station Israel 43: 18-54. Vervoort, W. 1968. Report on a collection of Hydroida from the Caribbean region, including an annotated checklist of Caribbean hydroids. Zoologische Verhandelingen, Leiden 92: 1-124. Vervoort, W. 1972. Hydroids from the “Theta”, “Verna” and “Yelcho” cruises of the Lamont-Doherty Geological Observatory. Zoo/ogiic/ie Verhandelingen, Leiden 120: 1- 247. Vervoort, W. 1987. Evaluation of taxonomic characters in the Hydroida, particularly in the Thecata (= Leptomedusae). In: Bouillon, J., Boero, F., Cicogna, F. and Cornelius, P. F. S. (eds) Modern trends in the systematics, ecology and evolution of hydroids and Hydromedusae. Pp. 83-103. Clarendon Press: Oxford. Vervoort, W. 1993. Cnidaria, Hydrozoa, Hydroida: hydroids from the western Pacific (Philippines, Indonesia and New Caledonia) I: Sertulariidae (Part 1). In: Crosnier, A. (ed.) Resultats des campagnes MUSORSTOM, II. Memoires du Museum d’Histoire Naturelle 158: 89-298. Vervoort, W., and Vasseur, P. 1977. Hydroids from French Polynesia with notes on distribution and ecology. Zoologische Verhandelingen. Leiden 159: 1-98. Warren, E., 1908. On a collection of hydroids, mostly from the Natal coast. Annals of the Natal Museum 1: 269-355. Watson. J. E. 1973. Pearson Island expedition 1969 - 9. Hydroids. Transactions of the Royal Society of South .Australia 97(3); 153-200. Watson, J. E. 1975. Hydroids of Bruny island, southern Tasmania. Transactions of the Royal Society of South Australia 99(4): 157-176. Watson, J. E. 1992. The hydroid community of Amphibolis seagrasses in south-eastern and south-western Australia. In: Bouillon, J., Boero, F., Cicogna, F., Gili, J. M. and Hughes, R. G. (eds) Aspects of hydrozoan biology. Scientia Marina 56(2-3): 217-227. Watson, J. E. 1994a. Shallow water hydroids from eastern Bass Strait. Victorian Naturalist 111(2); 65-73. Watson, J. E. 1994b. New records and redescriptions of thecate hydroids from southern Australia. Proceedings of the Royal Society of Victoria 106: 147-162. Watson, J. E. 1996. Distribution and biogeographic relationships of the hydroid fauna of the Australian west coast: a preliminary account. In: Piraino, S., Boero, F., Bouillon, J., Cornelius, P. F. S. and Gili, J. M. (eds) Advances in hydrozoan biology. Scientia Marina 60(1) 75-83. Watson, J. E. 1997. The hydroid fauna of the Houtman Abrolhos Islands. Western Australia. In: Wells, F. E. (ed.) The marine flora and fauna of the Houtman Abrolhos Islands,Western Australia. Pp 503-546. Western Australian Museum: Perth. Watson, J. E. 1999. Hydroids (Hydrozoa: Anthoathecata) from the Beagle Gulf and Darwin Harbour, Northern Australia. The Beagle, Records of the Museums and Art Galleries of the Northern Territory 15: 1-21. Wedler, E. 1975. Okologische Untersuchungen an Hydroiden des Felslitorals von Santa Marta (Kolumbien). Helgoldnder Wissenschaftliche Meeresuntersuchungen 11-. 324-363. Weltner, W., 1900. Hydroiden von Amboina und Thursday Island. In: Semon, R. (ed.) Zoologische Forschungsreise in Australia und dem Malayischen Archipel, 5. Denk.schriften der Medizinisch-Naturwissenschaftlichen Gesellschaft zu Jena 8: 583-590. Whitelegge, T. 1899. The Hydrozoa, Scyphozoa, Actinozoa and Vennes of Funafuti. Memoirs of the Australian Museum 3; 371-394. Yamada, M. 1958. Hydroids from the Japanese Inland Sea, mostly from Matsuyama and its vicinity. Journal of the Faculty of Science, Hokkaido (Series 6) 14: 51-63. Yamada, M. 1959. Hydroid fauna of Japanese and adjacent waters. Publications from the Akkeshi Marine Biological Station 9: 1-101. Accepted 14 November 2000 81 J.E. Watson Appendix 1. Taxonomic index of lepthothecate hydroids from the Beagle Gulf Aglaophenia 57 AGLAOPHENIIDAE 57 alata sp. nov., Salacia 26 ambigua sp. nov., Macrorhynchia 70 amirantensis, Lafoeina 5 angiilosa, Lytocarpia 64 Antennella 45 Anthohebella 1 badia, Plumularia 51 bedoti, Plumularia 54 bidentata sp. nov., Salacia 26 bilamellata sp. nov., Dynamena 15 CAMPANULARIIDAE 73 CAMPANULINIDAE 5 campylocarpum, Synthecium 40 Clytia 73 Clytia sp. 1 75 Clytia sp. 2 75 cornuta, Polyplumaria 56 costata, Hebeliopsis 6 cylindrica, Nemertesia 49 darwinensis sp. nov., Anthohebella 8 decipiens, Sertularella 27 delicatula, Aglaophenia 57 diaphana, Sertularella 31 dichotomum, Hydrodendron 10 digitalis, Diphasia 14 Diphasia 12 Dynamena 15 dyssymetrum, Halecium 9 Filellum 5 flavidula sp. nov., Salacia 24 flexuosa, Monotheca 48 fruticosus, Thyroscyphus 38 Gymnangium 58 HALECIIDAE 9 Halecium 9 HALOPTERIDAE 45 Halopteris 46 Hebella 7 Hebeliopsis 6 hexodon, Salacia 21 Mans, Gymnangium 58 Hydrodendron 10 Idiellana 19 irregularis, Kirchenpaueria 57 Kirchenpaueria 57 Kirchenpaueriidae 57 LAFOEIDAE 5 Lafoeina 5 ?laterocaudata, Hebella 1 lepida sp. nov., Idiellana 20 linearis, Clytia 73 longicorne, Gymnangium 60 Lytocarpia 64 macrocytharus, Thyroscyphus 31 Macrorhynchia 67 mertoni, Dynamena 16 Monotheca 48 mutulata, Diphasia 12 Nemertesia 49 operculata sp. nov., Thuiaria 35 orthogonium, Synthecium 41 parasitica, Anthohebella 8 patulum, Synthecium 43 philippina, Macrorhynchia 67 Phoenicia, Macrorhynchia 68 phyteuma, Lytocarpia 65 pinnata, Sertularella 32 plagiocampa, Halopteris 47 Plumularia 51 PLUMULARIIDAE 48 plumularioides sp. nov., Thuiaria 36 polymorpha, Halopteris 46 Polyplumaria 56 pri.stis, Idiellana 19 quadriarmata sp. nov., Macrorhynchia 72 quadridens, Sertularella 28 quadridentata, Dynamena 15 Salacia 21 scabra, Plumularia 52 scandens, Hebeliopsis 6 secundaria, Antennella 45 Iserratuni, Filellum 5 Sertularella 27 Sertularia 33 SERTULARIIDAE 12 setacea, Plumularia 53 sinuosa, Salacia 22 spatulum sp. nov., Halecium 10 SYNTHECIIDAE 39 Synthecium 39 tetracythara, Salacia 23 Thuiaria 35 Thyroscyphus 37 torresii, Thyroscyphus 37 Tridentata 18 Tridentata sp. 18 trigonostoma, Sertularia 33 tubacarpa sp. nov., Plumularia 55 undulatum sp. nov., Gymnangium 62 unjinense sp. nov., Gymnangium 62 ?warreni, Clytia 73 82 The Beagle, Records of the Museums and Art Galleries of the Northern Territory, 2000 16: 83-88 The rediscovery of Leandrites stenopus Holthuis, 1950 (Crustacea: Palaemoninae), from Lucinda, Queensland A.J. BRUCE Research Associate, Museum and Art Gallery of the Northern Territory GPO Box 4646, Darwin NT 0801, AUSTRALIA abruce@ broad, net.au ABSTRACT A second specimen of the rare palaemonid shrimp Leandrites stenopus Holthuis, 1950, has been collected from Lucinda harbour, northern Queensland. The only previous specimen was collected on the Siboga Expedition in 1899, from Madura, Indonesia. This male specimen enables the generic placement to be confirmed and the geographic distribution to be considerably extended. Keywords: Leandrites stenopus, Crustacea, Palaemoninae, rediscovery, Queensland. INTRODUCTION The unique specimen of the small marine shrimp Leandrites stenopus Holthuis was collected in 1899 but was not described until 1950, half a century later. A further half century has elapsed before a second specimen was obtained. Despite innumerable marine faunistic surveys in tropical Indo-West Pacific waters during the last half century, it is still surprising that a species can remain so apparently rare. The present specimen was obtained during the course of a survey of the marine faunas of northern Queensland harbours carried out by the Department of Marine Biology and Aquaculture of James Cook University of North Queensland for Ports Corporation Queensland. Abbreviations used: CL, post-orbital carapace length; NTM, Northern Territory Museum, Darwin. SYSTEMATICS Leandrites stenopus Holthuis, 1950 (Figs 1-3) Leandrites stenopus Holthuis, 1950: 40-42, fig. 6. - Chace and Bruce 1993: 7-8. Material examined. IcT, Lucinda, stn Sh 62, 18° 31' S, 148° 19' E, 7 July-1999, modified Ockelman sled, 15 m, coll. F. Hoedt, NTM Cr.012794. Description. In poor condition, lacking both second pereiopods and some other appendages. Rostrum (Fig. lA) about 0.95 of CL, slightly exceeding antennular peduncle, exceeded by scapho- cerite, acute, tapering, horizontal, feebly developed lateral carinae, with 11 acute teeth dorsally, first tooth epigastric, second also situated on carapace, four small ventral teeth distally, numerous median sparsely plumose interdental setae dorsally, similar submedian marginal setae ventrally; inferior orbital angle (Fig. 3B) well developed, rounded, antennal spine lower, slightly submarginal, branchiostegal suture absent, branchiostegal spine similar to antennal, distinctly postmarginal, pterygostomial angle rounded, sparsely setose. Ophthalmic somite without bee ocellaire, with small pigment spot. Eye (Fig. ID) with large globular cornea, diameter about 0.25 of CL, feebly pigmented, with accessory pigment spot, stalk about 0.8 of corneal width, slightly longer than wide. Antennule (Fig. IB) with proximal segment of peduncle twice as long as width, distolateral angle produced, rounded, with small acute distolateral tooth, stylocerite acute, short, not reaching half segment length, ventromedial border with small acute tooth at half length, statocyst well developed, with large circular statolith; intermediate and distal segments very obliquely articulated, subequal in length; upper flagellum biramous, proximal 5 segments of rami fused, short ramus about 0.8 of CL, with 10 stout segments, lower ramus long, filiform; lower flagellum long, filiform. Antenna (Fig. 1B) with basicerite bearing small acute lateral tooth; carpocerite about 0.3 of scaphocerite length, subcylindrical, moderately compressed, merocerite short, ischiocerite with acute medial process; flagellum long, filiform; scaphocerite about 4 times longer than wide, maximal width at 0.33 of length, lateral margin straight with well developed distolateral tooth, not exceeding distal border of broadly rounded lamella. 83 A.J. Bruce Fig. 1. Leandrites stenopus Holthuis. Male, Lucinda, Queensland, NTM Cr.012794. A, carapace and rostrum. B, antennule. C, antenna. D, eye. E, first pereiopod, chela. F, same, fingers. G, third pereiopod, dactyl and distal propod. H, telson. I, uropod. Abdomen with sixth segment about 0.45 of CL, 1.3 times longer than deep; 2.3 times longer than fifth segment, posterolateral angle well developed, acute, posteroventral angle rounded, with small acute preterminal tooth, fourth pleuron posteroventrally rounded, fifth pleuron small, posteroventrally acute. Telson (Fig. II) about 0.6 times sixth abdominal segment length, 2.5 times longer than anterior width, lateral margins straight, posteriorly convergent, dorsal spines small, at about 0.5 and 0.8 of telson length, anterior pair smaller, about 0.75, and further apart then posterior pair, posterior width about 0.25 of anterior width, acutely pointed (Fig. 3J), lateral spines small. about size of anterior dorsal spines, marginal, medial spines well developed, stout, about 0.3 of telson length, subventral, submedian setae half length of medial spines, slender, sparsely setulose, subventral. Mandible (Fig. 2A) robust, without palp, with stout four-toothed molar process, incisor process with three large acute teeth distally. Maxillula (Fig. 2B) with distinctly bilobed palp (Fig. 3C), lower lobe with small distal tubercle bearing short hooked seta; upper lacinia with about 12 stout simple spines distally; lower lacinia with numerous shorter spiniform setae distally. Maxilla (Fig. 2C) with basal endite bilobed, lobes slender, subequal, with sparse slender setae distally, palp tapering 84 Palaemonid shrimp from northen Queensland Fig. 2. Leandrites stenopus Holthuis. Male, Lucinda, Queensland, NTM Cr.012794. A, mandible. B, maxillula. C, maxilla. D. first maxilliped. E. second maxilliped. F, third maxilliped. distally with single terminal plumose seta, proximal lateral border with short plumose setae, scaphognathite 2.7 times longer than wide, broad centrally, narrow anteriorly, with small posterior lobe. First maxilliped (Fig. 2D) with basal endite large, broad, distally rounded, distal and medial borders with numerous fine, mainly simple setae, coxal endite distinct, distomedially angular, sparsely setose, palp tapering, distally angular, with long preterminal plumose seta medially, short simple setae distally, exopod well developed, with numerous plumose setae distally, caridean lobe small, narrow, sparsely setose, epipod large, deeply bilobed distal lobe much larger, triangular, proximal lobe smaller rounded. Second maxilliped (Fig. 2E) with dactylar segment short, broad, about 3.0 times longer than width, 0.6 of propodal segment length, with numerous long serrulate spines along medial margin, propodal segment broadly expanded distomedially, rounded, with numerous long serrulate spines medially, carpus and ischiomerus normal, exopod well developed, with numerous plumose setae distally, ramus expanded proximolaterally, basis medially produced, rounded, with small sub-oval epipod laterally, with small multi-lamellar podobranch. Third maxilliped (Fig. 2F) slender, ischiomerus completely fused with basis, junction indicated by small medial notch, about 15 times longer than central width, distally 85 A.J. Bruce Fig. 3. Leandrites stenopus Hollhuis. Male, Lucinda, Queensland, NTM. Cr.0I2794. A, tip of rostrum. B, inferior orbital angle. C, maxillula, palp. D, first pereiopod, distal dactylus. E, first pleopod. F, same, endopod. G, second pleopod. H, same, endopod. I, same, appendix masculina. .1, telson, posterior spines. broadened to about double central width, with three small spinules distolaterally, medial margin sparsely setose, carpal segment subcylindrical, about 12 times longer than wide, sparsely setose medially, distal .segment about 0.6 of carpal segment length, with strong distal spine, medially with nine transverse rows of spiniform setae; endopod well developed, reaching to about 0.75 of ischiomeral segment length, with numerous plumose setae distally; coxa medially produced, sparsely setose, lateral plate well developed, rounded, with small lamellar arthrobranch proximally. Pleurobranchs 5. Fourth thoracic sternite with small acute medial process. First pereiopod as previously de.scribed. Slender, with chela (Fig. 1E) about 0.75 of propod length, 0.68 of meral length; chela with palm slightly compressed, about 3.5 times longer than depth, with six transverse rows of short serrate setae proximally, fingers (Fig. IF) sparsely setose, about 0.8 of palm length, similar, stout, feebly hollowed medially, about 5.5 times longer than proximal depth, with acute hooked tips, with small acute tooth (Fig. 3D) proximally, rest of cutting edge laminar, entire. Third pereiopod slender, dactyl (Fig. IG) about 0.33 of propod length, slightly curved, about 17 times longer than proximal depth, unguis feebly demarcated, about 0.12 of corpus length, with two slender setae at half dorsal margin length, smaller setae proximally, two 86 Palaemonid shrimp from northen Queensland sensory setae distolaterally, propod subequal to CL, with two small distoventral spinules, six smaller ventral spinules. Fourth and fifth pereiopods generally similar. First pleopod (Fig. 3E) with endopod (Fig. 3F) about half exopod length, sub-oval, about 2.75 times longer than central width, distal and lateral borders with short plumose setae, proximal medial border with three spiniform setae, with well developed appendix interna arising at 0.5 of medial margin, not exceeding distal border of ramus. Second pleopod (Fig. 3G) with endopod (Fig. 3H) about 0.95 of exopod length, with numerous long plumose marginal setae, appendices arising at about 0.3 of medial margin length, appendix masculina (Fig. 31) about 0.3 of endopod length, slender, subcylindrical, 12 times longer than proximal width with three slender serrulate terminal spines, about half appendix length, 2 shorter simple distolateral spines. Uropod (Fig. IFI) with rami distinctly exceeding telson, protopodite posteriorly acute; exopod 3.4 times longer than central width, lateral margin straight with small distal tooth, with mobile spine medially, diaeresis obsolete, distal lamella reduced. Measurements. CL approx. 3.4 mm. Colouration. Unknown. Habitat. Water temperature 25.4°C; salinity 34.9 ppt; sea floor firm, featureless sand/mud, 5.6 km offshore. Remarks. The specimen generally agrees closely with the original description. The type specimen has only three ventral rostral teeth, in contrast with four in the present specimen. The rostral dentition can therefore be 1 + 10/3-4. Not reported in the original description are the small preterminal accessory teeth on each of the fingers of the first pereiopods. No similar teeth have been reported in other palaemonid genera. Discussion. The original specimen was collected on March 8, 1899 and no further specimens have been subsequently described during the following century. The discovery of a second specimen, albeit damaged and incomplete, shows that the species is still extant and of wider distribution. The assessment of rarity in marine habitats remains an unsolved problem. A terrestrial species that has not been observed over a period of 50 years is deemed extinct (Baillie and Groombridge 1996). Li and Manning (1998) reported the presence of this species in the northern South China Sea but provided no further details. The slender build and appendages of this shrimp suggests that it is a free-living soft bottom species. The holotype was collected from 56 m on a radiolarian ooze substrate. Chace and Bruce (1993) commented that the generic position of this species could not be fully ascertained until a male had been collected, enabling the condition of the first pleopod to be determined. This has now been rectified and the presence of an appendix interna on the first pleopod endopod confirms that this species belongs in the genus Leandrites. The presence of a median sternal tooth on the fourth thoracic sternite is noted; a similar tooth is also present in Leandrites ceiebensis (De Man, 1881), a feature that is probably another character of generic value (Bruce 1987). It is found in some genera in both the Palaemoninae and the Pontoniinae and its phylogenetic importance is yet to be evaluated. The fingers of the second pereiopod chela in Leandrites stenopus appear to be unique amongst the Palaemoninae, with the small acute preterminal teeth. It remains possible that these may have been overlooked in other species as, although distinctive, they are very small and not readily discernible when the fingers are closed. Four species of the genus Leandrites are now known, all from the central Indo-West Pacific region, and a key for their identification is provided by Chace and Bruce (1993). The commonest and most widely distributed is L. ceiebensis, recorded from southern India, Indonesia and tropical Australian waters, from shallow brackish waters. The three other species are all south east Asian. Leandrites indicus Holthuis, 1950, is known from the type specimen from Sulawesi, Indonesia, and Vietnam (Nguyen Van Xuan 1992), where it is common in mangrove creeks and rivers; L. deschampsi (Nobili, 1903), from three specimens from the type locality, Singapore, and one further specimen from China (Liu et ai. 1990); L. stenopus is now definitely known from two specimens only, from off Java, Indonesia, and north east Queensland, Australia, and probably also the northern South China Sea. The present record also contrasts with the original discovery and indicates that the species may also occur in shallower water (15 m versus 56 m) and on non- radiolarian substrates. ACKNOWLEDGMENTS I am most grateful to Dr Frank Hoedt for the opportunity to examine the specimen on which this report is based, and Dr C.H.J.M Fransen, for drawing my attention to the abstract by Li and Manning. This study was carried out with the support of the Australian Biological Resources Study. REFERENCES Baillie, J. and Groombridge, B. 1996. The lUCN species survival commission 1996 red list of threatened animals. lUCN: Gland, Switzerland. Bruce, A.J. 1987. Records of three palaemonid shrimps new to the Australian Fauna. The Beagle, Records of the Northern Territory Museum of Arts and Sciences 4: 57-60. Chace, F.A., Jr., and A.J. Bruce. 1993. The Caridean Shrimps (Crustacea: Decapoda) of the Albatross Philippine 87 A.J. Bruce Expedition 1907-1910, Part 6: Superfamily Palaeinonoidea. Smithsonian Contributions to Zoology 543: i-vii, 1-252, figs 1-23. Holthuis, L.B. 1950. The Decapoda of the Siboga Expedition. Part XI. The Palaemonidae collected by the Siboga and Snellius Expeditions with remarks on other species. II. Subfamily Pontoniinae. Siboga Expedition Monograph 39a’: 1-268. Li Xinzheng and R.B. Manning. 1998. Palaemonid shrimps from the northern South China Sea, including new taxa and a species list from Chinese waters. Poster abstract. Proceedings and Abstracts, Fourth International Crustacean Congress, Amsterdam 20-24 July 1998, 202-203. Liu, J.Y, Liang, X.Q. and Yan, S.l. 1990. A study of the Palaemoninae. (Crustacea Decapoda) from China 1. Macrobrachium, Leander and Leandrites. Transactions of the Chinese Crustacean Society 2: 102-134. Man, J.G. de. 1881. Carcinological Studies in the Leyden Museum, N° 1. Notes from the Leyden Museum 3: 121-144. Nguyen Van Xuan. 1992. Review of the Palaemoninae (Crustacea: Decapoda: Caridea) from Vietnam {Macrobrachium excepted). Zoologische Mededelingen, Leiden 66 (2): 19-47. Nobili, G. 1903. Crostacei di Singapore. Bolletino dei Musei di Zoologico ed Anatomia comparativo della R Universita di Torino 18 (455): 1-39, 1 pi. Accepted 8 August 2000 The Beagle, Records of the Museums and Art Galleries of the Northern Territory, 2000 16: 89-90 Onycocaridella prima Bruce, 1981, a rare pontoniine shrimp from Darwin Harbour (Crustacea: Decapoda: Pontoniinae) A.J. BRUCE Research Associate, Museum and Art Gallery of the Northern Territory GPO Box 4646, Darwin NT 0801, AUSTRALIA abruce@broad,net.au ABSTRACT The second occurrence of the pontoniine shrimp Onycocaridella prima Bruce, 1981 is recorded. The single specimen was found in a sponge host collected from Darwin Harbour, Northern Territory, Australia. The species is otherwise known only from the type specimens from Heron Island, Queensland. Keywords. Onycocaridella prima, Crustacea, Pontoniinae, Darwin Harbour, sponge associate, Australia. INTRODUCTION In the course of a survey of the marine fauna of the Port of Darwin, carried out by marine scientists from CSIRO Centre for Research on Introduced Marine Pe.sts (CRIMP) and staff of the Museums and Art Galleries of the Northern Territory in 1998, a single specimen of a small pontoniine shrimp was obtained from a sponge host collected from harbour piles. The specimen was identified as Onycocaridella prima, a species previously known only from the two type specimens collected in 1979 from Heron Island in the Capricorn Islands of the Great Barrier Reef. The new specimen of this apparently rare shrimp is here described and compared with the type material from Queensland. Abbreviations used: CL, post orbital carapace length; NTM, Museum and Art Gallery of the Northern Territory. SYSTEMATICS Onycocaridella prima Bruce, 1981 (Fig. 1) Onycocaridella prima Bruce, 1981a: 243-250, figs. 1 - 6 . Material examined. 1 ovig. 9 , stn NTD FHI PI-3, Fort Hill Wharf, Port of Darwin Harbour, I2°28.322’S 130°50.826’E. depth 3-9 m, collected by hand from pile scrapings, 15 August 1998, coll. CSIRO CRIMP team, NTM CrO 12795. Description. A stout shrimp, with body sub- cylindrical, slightly compressed anteriorly, with relatively large abdomen. Generally as in original description. Rostrum about 0.22 of CL, acute, slender, depressed, slightly up-turned, unarmed. Carapace (Fig. lA) about as long as maximal depth, smooth, devoid of spines and teeth. Abdomen slightly compressed, with first three pleura broadly expanded. Chelae of second pereiopods similar, only slightly unequal; major chela (Fig. 1B) with fingers 0.47 of palm length; minor chela (Fig. 1C) fingers 0.57 of palm length, dactyls with sinuous entire cutting edges. Third pereiopods with propod stout, bearing two short stout distoventral spines; dactyl (Fig. 1D) with distinctly demarcated unguis, corpus with ventral cutting edge straight, without distal accessory tooth, with several small denticles proximally. Ova numerous, small. Measurements. CL 3.3 mm; length of ovum 0.8 mm. Host. Mycale sp., [Porifera: Mycalidae], possibly Mycale (Aegagropila) cf. obscura (Carter). Distribution. Previously reported only from the type locality. Heron Island. Capricorn Islands, Queensland, on the southern Great Barrier Reef, at 12 m depth. Remarks. The single specimen is complete and in good condition and in general agrees well with the original description. With a CL of 3.3 mm, it is considerably larger than both the type specimens (cf allotype 1 .8 mm; 9 holotype 1.75 mm) sugge.sting that both of these were juvenile specimens. With ova, the present specimen is clearly an adult. In the adult, the carapace length is less than the carapace depth, whereas in the juvenile types it is greater (Fig. IE), the rostrum is also more depressed, without a small preterminal dorsal tooth as in the holotype. In the holotype female the second pereiopod chelae are markedly unequal in size ( 2 . 0 ; 1 . 0 ), in the present specimen they are only slightly unequal (1.09:1.0), with the major chela 89 A. J. Bruce Fig. 1. Onycocaridella prima Bruce, ovigerous female, Darwin Harbour, NTM Cr012795. A, carapace, lateral. B, second pereiopod, major chela. C. same, minor chela. D, third pereiopod. distal propod and dactyl. E, comparison of carapace shapes in adult and juvenile (stippled) specimens to show relative profiles (not to scale). ACKNOWLEDGMENTS relatively smaller, 1.29 times the CL, as opposed to 2.0 times the CL in the holotype. The ambulatory dactyl of the third pereiopod is essentially as in the holotype but lacks the small distoventral accessory tooth and the distal ventral margin of the corpus appears unarmed. The absence of these small denticles may be merely the result of abrasion. The proximal ventral margin bears about six minute denticles. The host sponge has not been identified with full certainty. The preserved sample contained repre¬ sentatives of several sponge genera (J.N.A. Hooper, personal communication), but these included material referrable to a Mycale, possibly Mycale {Aegagropila) cf. obsciira (Carter). As the type specimens of O. prima were found in association with Mycale sulcata Hentschcl, it is most probable that this further Mycale was the host of the present specimens. The key to the genus Onycocaridella Bruce, accompanying the original description of O. prima remains valid (Bruce 1981a). The two other species of U;e genus, O. stenolepis (Holthuis, 1952) and O. monodoa (Fujino and Miyake, 1969) have also both been recorded in Australian waters, also from Heron Island (Bruce 1981b, 1983). 1 am most grateful to Helen Barnes for the opportunity to examine the specimen on which this report is based and to Dr J.N.A. Hooper for the identification of the sponge host. This study was carried out with the support of the Australian Biological Resources Survey. REFERENCES Bruce, A.J. 1981a. Onycocaridella prima gen. et sp. nov., a new pontoniine sponge-associate from the Capricorn Islands, Australia (Decapoda Caridea: Pontoniinae) Journal of Crustacean Biology 1(2): 241-250. Bruce, A.J. 1981b. Pontoniine shrimps of Heron Island. Atoll Research Bulletin 245: 1-33. Bruce, A.J. 1983. The pontoniine shrimp fauna of Australia. Australian Museum Memoirs 18: 195-218. Fujino, T. and Miyake, S. 1969. Studies on the genus Onycocaris with descriptions of five new species (Crustacea, Decapoda, Palaemonidae). Journal of the Faculty of Agriculture, Kyushu University 15: 403-448. Holthuis, L.B. 1952. The Decapoda of the Siboga Expedition. Part XI. The Palaemonidae collected by the Siboga and Snellius Expeditions with remarks on other species. II. Subfamily Pontoniinae. Sihoga Expedition Monograph 39al0: 1-252. Accepted 8 August 2000 90 The Beagle, Records of the Museums and Art Galleries of the Northern Territory, 2000 16 : 91-96 Biological observations on the commensal shrimp Paranchistus armatus (H. Milne Edwards) (Crustacea: Decapoda: Pontoniinae) A.J. BRUCE Research Associate, Museum and Art Gallery of the Northern Territory GPO Box 4646, Darwin NT 0801, AUSTRALIA abruce@broad.net.au ABSTRACT Some aspects of the biology of the commensal pontoniine shrimp, Paranchistus armatus (H. Milne Edwards), on the Great Barrier Reef, are reported. The species is an obligatory commensal living as heterosexual pairs only in the giant clam, Tridacna gigas (L.). Features of the morphology, fecundity, population structure, infection rate, and reproductive mechanisms are described. It is considered possible that the species shows indications of serial male protandrous hermaphroditism, not previously noted to occur in shrimps of the subfamily Pontoniinae. It may also be considered an endangered species as its host animal is under threat. Keywords: Paranchistus armatus, Pontoniinae, Decapoda, biology, possible hermaphroditism, Tridacna gigas commensal, Australia, Great Barrier Reef, endangered species. INTRODUCTION The pontoniine shrimps associated with bivalve hosts have attracted a certain amount of biological study, particularly the species of the genera Anchistus, Paranchistus and Conchodytes, found in association with hosts in the family Pinnidae (Johnson and Liang, 1966; Hipeau-Jacquotte 1974; Morton 1987). The species associated with the family Tridacnidae have received little attention, probably as the result of the clam’s legal protection on most coral reefs and a reluctance to sacrifice a significant number of these ho.sts. The shrimp Paranchistus armatus is unusual in the subfamily Pontoniinae, as it is one of the largest species known, and is found only in association with the giant clam Tridacna gigas (L.). The latter has a restricted distribution in the Malaysian-Western Pacific region (Rosewater 1965) and the shrimp, although first described by Henri Milne Edwards in 1837, has so far only been recorded from a relatively small number of localities within that region. The giant clam is now a protected species in Queensland waters, where spe¬ cimens can only be collected by special permit. The Queensland Fisheries Service has been carrying out an investigation of the biology of T. gigas, which has enabled numerous specimens of P. armatus to be examined without involving any unnecessary sampling of the host animal. Previous reports on these shrimps have generally been concerned with one to three individuals and the present collection of 164 specimens represents the first time that a small population, together with some data concerning the hosts, have been available for study. I am most grateful to Dr R.G. Pearson, of the Queensland Fisheries Service, Department of Primary Industry, and Dr V. Harriot, for this opportunity to report upon these specimens. Measurements (mm) refer to the postorbital carapace length (CL) of the shrimps. Representative specimens have been deposited in the collections of the Northern Territory Museum, Darwin, (Cr.008674) and the Queensland Museum, Brisbane (W25447). DESCRIPTION Paranchistus armatus (H. Milne Edwards, 1837) (Figs 1-3) Restricted synonymy: Pontonia armata H. Milne Edwards, 1837: 359. Anchistus biunguiculatus Borradaile, 1898: 387. Tridacnocaris biunguiculatus - Nobili 1899: 235. Anchistus oshimai Kubo, 1949: 26 Paranchistus biunguiculatus - Holthuis 1952: 13, 93-97, figs 36-38. Anchistus armatus - Bruce 1967: 564-568. Paranchi.stus armatus - Bruce 1975: 49-54, figs 1-3. Material examined, (i) 61 adult pairs, 5 ovig.^, 2 9 , 4 cf, 3 immature, Arlington Reef, 16°45.0’S, 146° 00.0’E, from reef flat, 2 November 1978 to 12 November 1979. (ii) 4 pairs, 15 juveniles, Hastings Reef, 26 November 1979. (iii) 1 pair, I cf, 3 juveniles, Michaelmas Cay, 5 December 1979. 91 A. J Bruce Fig. 1. Paranchislus armatus (H. Milne Edwards). Arlington Reef, Queensland, Australia. A, left mandible. B, same, incisor process. C, maxillula. D, same, palp. E, same, upper lacinia. P, maxilla. G, first maxilliped, palp and endites. H, paragnaths, ventral aspect. I, same, left lateral. J, buccal region, left maxilla to third maxilliped removed. K. first pleopod, endopod. L, second pleopod, endopod, appendices. A-J, ovigerous female. KL,. male. Morphology. The present specimens agree closely with the previously published descriptions given by Kubo (1949, as Anchistus oshimai), Holthuis (1952, as P. biungidculatus), and Bruce (1975). In the large specimens, particularly the females, the hepatic spine is very small and could only be identified with considerable difficulty, even in dry specimens. The dorsal rostral dentition on small specimens is relatively conspicuous, in contrast to the condition in large females in which it is obsolete. The mouthparts correspond closely with the illustra¬ tions provided by Holthuis (1952). The maxillula has short simple spines along the inner aspect and simple setae on the outer side of the distal margin of the upper lacinia (Figs 1 C, E). The palp is bilobcd, with a short simple seta on the lower lobe (Fig. 1 D). The maxilla has the basal endite bilobed, with the distal lobe larger than the proximal and bearing six short setae, in contrast to five (Fig. I F). The palp is broad and flattened, without a subterminal seta but with a few short plumose setae on the proximal lateral border. The setae on the endites of the first maxilliped are simple. The coxal endite bears a single long slender seta. The palp is non-setose and the caridcan lobe of the exopod is fringed with short plumose setae. The dactyl of the second maxilliped is provided with numerous finely serrated spiniform setae Commensal shrimp of the giant clam Fig. 2. Paranchistus annatus (H. Milne Edwards), Arlington Reef, Queensland, Australia. Male, CL 8.0 mm. A, rostrum. B, chela of first pereiopod. C, fingers of right second pereiopod. D, fingers of left second pereiopod. E, posterior telson spines. Female, CL 14.0 mm. F, fingers of second pereiopod. G, posterior telson spines. (Fig. 1 J). The anteromedial border of the propodal segment bears long slender simple spines. The setae of the third maxilliped are simple (Fig. 1 G). The epipods of the three maxillipeds are deeply bilobed, sub- rectangular and oval respectively (Figs 1 K, L). The third maxilliped also bears a small six lamellar arthrobranch. The flagella of all exopods are well developed and broad, with numerous plumose setae along the margins of the distal third. The chelae of the first pereiopod are subspatulate, with the laterally situated cutting edges of the fingers fully finely pectinate (Fig. 2 B). The fourth thoracic sternite bears a low transverse ridge with a small median notch, and a similar but slightly large ridge is also present on the fifth sternite. The second pereiopods are generally subequal and similar in both males and females, but relatively larger in the former (Figs 2 C, D, F) The dactyl is generally armed with 3-4 small acute teeth on the proximal cutting edge and the fixed finger has 7-12, the most proximal of which may be very small. The telson is provided with three pairs of posterior spines and two pairs of small dorsal spines (Figs 2 E, G). The latter are situated at about 0.70 and 0.85 of the telson length. The dorsal spines are relatively longer in the males. The intermediate posterior spines are about 4.5 times longer than wide, about 0.1 of the telson length, and about twice the length of the lateral spines. The submedian spines are slender, feebly setulose and 0.4 - 0.7 of the length of the intermediate spines. Colouration. The body, antennal peduncles,second pereiopods and caudal fan are white, with the antennal flagella purplish. BIOLOGY Fecundity. Of the 73 females over 11 mm CL, all except four were ovigerous, and one of these appeared A. J Bruce to have just hatched its ova. The smallest ovigerous female, CL 11 mm, carried 1,550 ova, and the largest, CL 18mm, carried 8,901 ova. Intermediate size females carried the following numbers of ova: (i) CL 13 mm, 4646; (ii) CL 15 mm, 5638; (iii) CL 17 mm, 7,788. The ova are about 0.5 mm in length when freshly laid and 1.7 mm when about to hatch. Population structure. Of the 138 specimens from Arlington Reef, the males ranged in size from CL 7-11 mm and the females 8-18 mm. No Juveniles below CL 7 mm were found and the appendix masculina is re¬ cognisable in males of this carapace length. The distribution of carapace length is shown in Figure 3. It is noteworthy that no males occur with a CL of over 12 mm and that very few females occur with a carapace length below 11 mm, less than 3 %. As males are readily identifiable from CL, 7-11 mm, it appears that this population did not contain anyjuvenile females and suggests that males of this species may be a protandrous hermaphrodites, with the transition from male to female occurring at about CL 11 mm. In general the larger females are paired with the larger males but considerable variation exists and a larger sample is needed to clarify apparent discrepancies. The relationships are summarised in Table 1. Table 1. Range of carapace length (CL) in male.s paired with females of specified CL 9 CL cf CL, range n mean 1 1 8-11 5 9.8 12 7 - 9 2 8.0 13 9-11 6 9.6 14 7 - 10 7 8.2 15 9-11 15 10.5 16 8-11 14 9.7 17 9-11 13 10.4 18 10-11 3 10.6 Almost all specimens were found as male-female pairs. At Arlington Reef, 61 pairs were collected, together with 6 six ovigerous females without male partners and 2 non-ovigerous females, CL 16 and 8 mm. It is most probable that the males of the six ovigerous females were overlooked in dissection of the host clam, or escaped during its removal from the reef. Of the four cases where isolated females were found, three were small, CL 8 mm, and could have been without males or again, the males may have been overlooked during collection. In only one instance, at Arlington Reef, was more than a pair of shrimps found in the host clam. In this case a pair of small individuals, both of CL 10 mm were accompanied by an additional small male of 7.5 mm CL, the smallest identifiable male found. The female of this association was without ova. The two largest female shrimps (both 18 mm CL) were found in two of the largest host clams, with shell lengths of 83 cm and 95.5 cm. Two larger clams from Hastings Reef were sampled, valve length 100 cm and 102.5 cm. both with a breeding pair of shrimps. The smallest breeding pairs of shrimps were of 7.5 and 12 mm CL, and 10 and 11.5 mm CL in clams of 37.1 cm and 43.0 cm respectively. The smallest pair of shrimps, CLs 4.5 and 5.0 mm, were from Hastings Reef, from a clam of 22.2 cm valve length. Juveniles. The Arlington Reef population of P. armatus is remarkable for the apparent absence of juveniles, although sampling was carried out in the months November, December, January, July, August, September and October and ovigerous females were found at all times, so that breeding may occur throughout the year. Additional samples were also collected from Hastings Reef, where clams of a larger size and smaller size were sampled. At Hastings Reef two juveniles CL 4.5 and 5.0 mm, were found in clams of valve lengths Commensal shrimp of the giant clam 28 and 23 cm respectively. A pair with an ovigerous female were also found in a clam of valve length 43 cm. At Michaelmas Cay, eight additional small specimens of T. gigas were sampled, of which three contained juveniles shrimps CL 4, 4, and 5 mm, clam valve lengths 30, 31 and 27 cm respectively. One clam, valve length 34 cm, contained a single male, CL 7 mm, and another, valve length 37 cm., a small pair with an ovigerous female, CLs 7.5 and 12 mm. Two particularly large clams were also examined at Hastings Reef in November 1979, valve lengths 100 and 102 cm. Each contained a male-female pair of adult shrimp, CLs 11 and 15 mm and 11 and 17 mm. These were accompanied by 7 and 6 juvenile shrimp res¬ pectively, CLs 1.5, 4, 4, 4, 5.5, 7, 7 mm and 4, 4.5, 5, 6, 7, 7 mm in each case. The 1.5 mm CL specimen was the smallest found and is probably only a little over the first post-larval stage size. Autotomy. 89 specimens were examined in detail for evidence for autotomy and limb regeneration. In not a single example was there any evidence of limb loss, injury or regeneration. Host. All specimens were found in association with the giant clam Tridacna gigas (L.). Hosts sampled ranged in size from 22 - 102 cm in valve length. Small numbers of Tridacna crocea, T. derasa, T. maxima, T. squamosa and Hippopus hippopus were also sampled from Michaelmas Cay but all were without associated P. armatus (Pearson pers. comm.). The exact situation within the host clam was not observable or recorded. Infestation rate. Details are available of 91 clams from Arlington Reef. Of these, 84 contained associated shrimps, giving an infection rate of 92.3 %. Including clams from other localities, 118 clams were infested on 100 occasions, with an overall infection rate of 84.7%. Associated fauna. One specimen of Tridacna gigas, valve length 18.2 cm, also contained two specimens of the pinnotherid crab Xanthasia murigera White in addi¬ tion to Paranchistus armatus. No bopyrid parasitization of the shrimps was found. Distribution. Type locality: New Ireland, Papua New Guinea. Also known from Indonesia: Batanta; Mefour (Nobili 1899); Obi Latu (Holthuis 1952); Papua New Guinea: New Ireland (H. Milne Edwards 1837); Tubetube, Engineer Islands (Borradaile 1898); Hansa Bay (De Grave 1999); Australia: Undine Reef (McNeill 1968); Chapman Island (Bruce 1975); Cairns; Arlington Reef, Michaelmas Cay (Bruce 1983); Caroline Islands: Helen Atoll, Palau (Kubo 1949); Ngadarak Reef and Ngaianges Island (Miyake and Fujino 1968); Marshall Islands: Ujae Atoll (Holthuis 1953); Eniwetak Atoll (Rosewater 1965; Bruce 1975; Bruce 1979; Devaney and Bruce 1989); Bikini Atoll (Chace and Bruce 1993); Kiribati: Onotoa Atoll (Holthuis 1953). Remarks. Giant clams are notorious for their longevity and it may be safely assumed that each clam provides a home for many generations of commensal shrimp. No information is available on the life span of the infesting shrimps. Their relatively large size, in relation to other pontoniine shrimps, may indicate that they live longer than about 12 months as suggested for Periclimenes ornatus, a small pontoniine associate of Japanese sea anemones (Omori et al. 1994). Morton (1987) has discussed the biology of the commensal pontoniine shrimps, Anchistus custos (Forsskal) and Conchodytes rnonodactylus Holthuis, 1952, associated with the pinnid bivalve Pinna hicolor Gmelin. This host is comparatively short lived in comparison with T. gigas and the situation is not comparable. No information is available on the recruitment of these shrimps but it seems likely that they arrive as post-larvae from the plankton. In the present study the smallest individuals were little above the typical size of pontoniine post-larvae. The small ova indicate that there is no abbreviated larval development that would enable recruitment to be derived from the occupying pair of shrimps. The frequency of heterosexual pairs of adult shrimps only akso suggests that, in general, their presence suppresses or prevents further colonisation by juveniles. The high infestation rate suggests that replacement of shrimps is efficient and quick. The range of shrimps sizes indicates a continuous process without marked seasonality. Possibly the death of an adult female results in the change of the male partner into a female and the recruitment of a new shrimp from the larvae in the plankton, which would develop into a male. Death of a male would follow similar recruitment, with the post-larva also developing into a male. Protandrous and simultaneous hermaphroditism have been reported in a number of caridean shrimp families (Bauer and Holt 1998) but sex changes have so far not been recorded in the Palaemonidae. The absence of any signs of trauma or autotomy suggests that life in Tridacna gigas represents a particularly secure niche. This may be contrasted with the situation reported in Coralliocaris graminea (Dana) living in Acropora corals (Bruce 1976), where signs of damage to the shrimp were frequent. In this species a minimum of 23% of specimens examined showed signs of significant regeneration after serious damage. The species lives in small communities depending upon the size of the coral host. The incidence of damage increases with the size of the population, reaching up to 77% in the larger populations. This damage was attributed to intraspecific combat, particularly between females (Bruce 1976), factors that would not be present in the case of P. armatus. The giant clam, Tridacna gigas (L.), was formerly widely distributed in the Malaysian — northern Australian — Western Pacific- region (Rosewater 1965) but has recently been reported as “? extinct” in Taiwan, Vanuatu, Fiji, Guam, New Caledonia and the Northern Marianas (Wells 1997). Although unreported from these A. J Bruce localities, it is likely that P. armatiis was also present in these regions but is now similarly extinct. All species of Tridacnidae are listed in the Convention on International Trade in Endangered Species of Wild Fauna and Flora, APPENDICES I and II, as adopted by the Conference of the Parties, valid from 16 February 1995 (http:// www.iwec.org/citesl.htm). If T. gigas is considered to be an endangered species, logically its obligate “commensal” associate, P. annatiis, must be in equal danger. Much remains to be studied in the life histories of “commensal” shrimps such as Paranchistus annatus and most others. Information on their food and feeding habits is minimal, as is data on their larval stages and life history, mechanism of host colonisation, longevity and reproductive biology. The term “commensalism” is frequently a euphemism for ignorance of details of the lifestyle involved. ACKNOWLEDGMENTS I am grateful to Dr Richard Willan and Darryl Potter for information on the present status of giant clams. REFERENCES Bauer, R.T. and Holt, G.J. 1998. Simultaneous hermaphroditism in the marine shrimp Lysmata wurdemanni (Caridea: Hippolytidae): an undescribed sexual system in the decapod Crustacea. Marine Biology 132: 223-235. Borradaile, L.A. 1898. A revision of the Pontoniidae. Annals and Magazine of Natural History Series 7 2: 376-391. Bruce, A.J. 1967. The results of the re-examination of the type specimens of some pontoniid shrimps in the collection of the Museum National d’Histoire Naturelle, Paris. Bulletin dll Museum National d’Histoire Naturelle, Paris 2e Serie 39(3): 564-572. Bruce, A.J. 1975. Pontonia armata H. Milne-Edwards (Decapoda Natantia. Pontoniinae) - a correction. Criistaceana 29(1): 49-54. Bruce, A.J. 1976. A report on some pontoniinid shrimps collected from the Seychelle Islands by the F.R.N. Manihine, 1972, with a review of the Seychelle pontoniinid shrimp fauna. Zoological Journal of the Linnaean Society, London 59: 89-153. Bruce, A.J. 1979. A report on a small collection of pontoniine shrimps from Eniwetok Atoll. Crustaceana Supplement 5: 209-230. Bruce, A.J. 1983. The pontoniine shrimp fauna of Australia. Australian Museum Memoirs 18: 195-218. (1982). Chace, F.A. Jr. and Bruce, A.J. 1993. The caridean shrimps (Crustacea: Decapoda) of the Albatross Philippine Expedition 1907-1910, Part 6: Superfamily Palaemonoidea. Smithsonian Contributions to Zoology 543. De Grave, S. 1999. Bivalvia associated with Pontoniinae (Crustacea: Decapoda: Palaemonidae) from Hansa Bay, Papua New Guinea. Bulletinde I’lnstitut Royal des Sciences Naturelles de Belgique. Biologie 69: 161-177. Devaney, D.M. and Bruce, A.J. 1987. Crustacea Decapoda (Penaeidea, Stenopodidea, Caridea and Palinura) of Eniwetak Atoll. 17. In: Devaney, D.M., Reese, E.S., Burch, B.L. and Helfrich, P. (eds) The natural history of Eniwetak Atoll. Part 2. Biogeography and systematics. Pp. 221-233. U.S. Department of Energy: NTIS Energy Distribution Centre, Oak Ridge, Tennessee. Hipeau-Jacquotte, R. 1974. Etude des crevettes Pontoniinae (Palaemonidae) associeds aux molluscs Pinnidae it Tuldar (Madagascar). Archives de Zoologie Experimental et Generale 115(3): 359-386. Holthuis, L.B. 1952. The Decapoda of the Siboga Expedition. The Palaemonidae collected by the Siboga and Snellius Expeditions with remarks on other species. 2. The subfamily Pontoniinae. Siboga Expedition Monograph 39*"': 1-252. Holthuis, L.B. 1953. Enumeration of the decapod and stomatopod Crustacea from Pacific coral islands. Atoll Research Bulletin 24: 1-66. Johnson, D.S. and Liang, M. 1966. On the biology of the watchman prawn, Anchistus custos (Crustacea Decapoda Palaemonidae), an Indo-West Pacific commensal of the bivalve Pinna. Journal of Zoology, London 150: 433 455. Kubo, 1. 1949. A new species of the genus Anchistus. Bulletin of the Biogeographical Society of Japan 14: 26-29. McNeill, F.A. 1968. Crustacea Decapoda & Stoinatopoda. Great Barrier Reef E.xpedition, 1928-1929. Scientific Reports 7(1): 1-98. Milne Edwards, H. 1837. Histoire naturelle des Crustaces, comprenant I’anatomie, la physiologic et le classification de ces animaux. Vol. 2: pp 1-532; atlas, 1-32, pis 1-42. Paris: Librairie Encyclopedique de Rorel. Miyake, S. and Fujino, T. 1968. Pontoniinid shrimps from the Palau Islands (Crustacea, Decapoda, Palaemonidae). Journal of the Faculty of Agriculture, Kyushu Imperial University 10(3); 339-431. Morton, B. 1987. Temporal host segregation by Anchistus custos and Conchodytes monodactylus (Crustacea: Pontoniinae) and Pinna bicolor (Bivalvia: Pinnidae) in Hong Kong. Asian Marine Biology 4: 129-140. Nobili, G. 1899. Contribuzioni alia Conoscenza della Fauna carcinologica della Papuasia, delle Molucche e dell Australia. Annali del Museum Civico di Storia Naturale di Genova, Seiers 2 20(40): 230-282. Omori, K., Yanagisawa, Y. and Hori, N. 1994. Life history of the caridean shrimp Periclimenes ornatus Bruce associated with a sea anemone in south west Japan. Journal of Crustacean Biology 14(1); 132-145. Rosewater, J. 1965. The family Tridacnidae. Indo-Pacific Mollusca 1 (6): 347-393. Wells, S. M. 1997. Giant clams: .status, trades and mariculture, and the role of CITES in management. World Conservation Union (lUCN). Accepted 1 November 2000 96 The Beagle, Records of the Museums and Art Galleries of the Northern Territory, 2000 16: 97-106 Vibration signals in Australian fiddler crabs - a first inventory HEINRICH-OTTO VON HAGEN Fachhereich Biologic, Zoologie, Philipps-Universitdt D-35032 Marburg/Lahn, GERMANY ABSTRACT Vibration signals (often called sounds), which are components of the subterranean communication system of fiddler crabs, were recorded in 14 of the 17 confirmed Uca species of Australia and analysed by use of oscillograms. Signals of the related crab Heloechis cordiformis and two Asian species of Uca were used for comparisons. All crabs studied produce “bounces”, i.e. burst-pulsed sounds corresponding to laterally directed jerks of the whole animal. All species except H. cordiformis, U. dampieri and U. vomeris also emit “drumwhirls” or “rolls”, the familiar rapping sounds of the major chelipcd. Parameters of these drumwhirls were compared using aquarium sound recordings of males competing for a burrow. Indications of character displacement were found with regard to western and eastern samples of U. signata. The most advanced rapping signals were recorded in U. signata and U.flammitla and in the two Asian species, U. rosea and U. anmdipes', the most simple ones were recorded in U. longidigiium and U. polita. The two latter appear to have pre.served early stages of drumwhirl evolution, contributing to the assumption of an Australian origin of the genus Uca. Keywords. Crustacea, Brachyura, Ocypodidae, Uca and Heloecius, vibration signals, survey, early stages of evolution, Australian origin of Uca. INTRODUCTION Crabs inhabiting the semiterrestrial environment— mainly Sesarminae (Grapsidae) and the subfamilies of the Ocypodidae—are well-known examples of commu¬ nication by vibration signals. Various types of such signals have been reported in literature (see preliminary synopsis in von Hagen 1975), the most important ones being generated by percussion and stridulation. Most research on signal production, transmission and reception has been done in tiddler crabs {Uca) and ghost crabs iOcypode) (see Salmon 1983 for a review). It is not clear to date, how often vibration signalling arose independently in the two crab families mentioned. It is certain, however, that vibration signals are used in both agonistic and courtship situations and occur mostly underground or at least in or near the mouth of the crab’s burrow. Being short-range signals, vibrations appear to play a major role in reproductive isolation. At least in two West Indian sympatric and sibling species of Uca, vibration signals were found to be much more species- specific than the visual display (von Hagen 1984). As the signals can also have group-specific characters, they allow various phylogenetic inferences (von Hagen 1975, 1984), especially when these inferences are based on quantitative data. Until now, quantitative studies of Uca bioacoustics have mainly dealt with the European fiddler crab U. tangeri (von Hagen 1962; Altevogt 1963, 1970) and American species of both the Atlantic (Salmon 1965, 1983; Salmon and Horch 1972; von Hagen 1984) and Pacific coasts (Altevogt 1970; Muller 1989). R. Polivka (pers. com.) started a bioacoustic project on fiddler crabs of South East Asia, but did not complete and publish his studies. With respect to Australia, pertinent research seems to be lacking. In the course of two visits to Australia, the author had the opportunity to record subterranean vibration signals of nearly all Australian species of Uca and of the related ocypodid Heloecius cordiformis. The recordings were obtained using a special aquarium that allowed direct observation of the crabs during sound production (cf. von Hagen 1970, 1984). The oscillo¬ graphic analyses published here are meant to serve as a first survey or inventory and thus as a basis for more specific research in future. With a few exceptions, the vibration signals of Australian fiddlers are simple, especially when compared with the two Asian species included in this study. This result allows certain speculations: that some Australian Uca species have preserved an early stage of acoustic communication (early with respect to the whole genus) and that the genus probably originated in Australia. 97 H-O. von Hagen Table 1. Abbreviations, systematic grouping, scientific names and Australian distribution of the species and origin of the samples studied. Roman group numerals of Uca as in George and Jones (1982); in parentheses: subgeneric names from Crane (1975), invalid except A ustraluca. Distribution in Australia: C circum-tropical, E eastern, W western. Genus Heloecius: He H. cordifonnis (H. Milne-Edwards) E Brisbane Genus Uca, group IV ("Thalassuca"): dp U. dampieri Crane W Darwin vm U. vomeris McNeill E Bamaga, Brisbane Genus Uca, group III (Australuca): Ig U. longidigitum (Kingsley) E Brisbane pi U. polita Crane C Broome, Cairns, Darwin hs U. hirsutimanus George and Jones W Broome, Darwin sg U. signala (Hess) c Cairns, Darwin el U. elegans George and Jones w Darwin ss U. seismella Crane c Broome, Darwin Genus Uca, groups I and II {"Deltuca"): cp U. capricornis Crane W Bamaga, Darwin ds U. dussumieri (H. Milne-Edwards) E Bamaga, Cairns fl U. flammula Crane W Broome, Darwin cc U. coarctata (H. Milne-Edwards) E Brisbane rs U. rosea (Tweedie) - Penang and Malakka (Malaysia) Genus Uca, group V ("Celuca"): mb U. mjoebergi Rathbun W Broome, Darwin PP U. perplexa (H. Milne-Edwards) E Bamaga, Cairns an U. annulipes (H. Milne-Edwards) - Bombay (India) ANIMALS AND METHODS Species names and localities of the specimens are given in Table 1. The endemic Australian ocypodid crab Heloecius cordifonnis, lately placed in a subfamily of its own (Heloeciinae Tiirkay 1983), was studied in the northern part of its range, which ends at Rockhampton (Davie pers. comm). The present knowledge of the geographic ranges of the Australian Uca, as specified in von Hagen and Jones (1989), is also briefly indicated in the Table. Table 1 lists 14 of the 17 species of Uca confirmed for the Australian fauna (only U. crassipes, U. tetragonon and U. triangularis were not studied). As the discussion about the extent and proper naming of the species groups is not yet settled. Table 1 follows the grouping of George and Jones (1982) by Roman numerals; only U. elegans is transferred from group I to group III and the order of groups has been altered (see Discussion). In addition the subgeneric names from Crane’s monograph (1975) are listed, which are widely in use though mostly {except Austral uca) invalid because of Bott (1973). The abbreviations of species names (Table 1) are used also in Figures 1 and 2. The two Asian species, U. rosea and U. annulipes, included in the list, were collected by travelling students, and the sounds were recorded in Germany. The Australian species were studied at the places listed in Table 1 during July and August of the years 1986 and 1988. Outdoor recordings of subterranean signals of crabs are useful on certain occasions, but normally do not allow statements on how the sounds are generated (von Hagen 1975). Therefore, the signals were recorded indoors in aquaria (12 1, height 20.5 cm), which, using the technique of von Hagen (1970, 1984), allowed direct observation of subterranean activities of the crabs. Bricks were laid in the centre of the aquaria and only the narrow gap between these bricks and the plastic walls were filled with moist muddy sand from the habitat of the animals. The crabs were forced to dig their burrows within this thin vertical layer of soil and were observed, at very close quarters, through the wall, which had to be kept clean. Though also called “sounds”, the signals recorded act as substrate-borne vibrations; the air-borne 98 Vibration signals in fiddler crabs Table 2. Burst-pulsed sounds (“bounces”) measured in males of Heloecius cordiformis and four Australian Uca species (repetition rates calculated from means of preceding column). Species One bounce: seconds mean s.d. Sequences: maximum number of bounces One bounce plus subsequent silent interval: seconds mean s.d. Repetition rate: bounces per second Number of bounces measured (males) H. cordifonnis 0.30 + 0.10 3 0.64 + 0.24 1.56 22 (2) U. vomeris 0.12 + 0.04 15 0.39 + 0.19 2.56 23 (2) U. seismella 0.07 + 0.02 43 0.19 + 0.02 5.26 64 (5) U. coarctata 0.25 + 0.08 4 0.57 + 0.13 1.75 21 (3) U. mjoebergi 0.20 + 0.06 6 0.69 + 0.14 1,45 64 (5) components are only by-products. Therefore, a Briiel and Kjaer accelerometer (type 4368) was used as a vibration pick-up system. The accelerometer, screwed to a 6.5 cm probing rod, was placed close to the crab’s burrow and was connected to an Uher tape recorder (type 4200 Report Stereo 1C) via a small preamplifier designed by the Institute of Applied Physics, Marburg (Dr. K.H. Wittich). The recordings were analysed in Marburg by examination of oscillograms obtained from a Philips oscilloscope PM 3231 in combination with a camera (Tbnnies Recordine K-854). The soil temperature ranged from 23 to 27.5 “C during the recordings (including the ones in Marburg). All crab individuals tested were of medium size when compared with the figures of maximum carapace width for each species (contained in Fig. 2). No special attempts were made to detect signals of females, which tend to be rare and more faint, when present (von Hagen 1984). Thus all data given in this paper refer to males. Mean values of the signals recorded were compared by a version of Student’s t-test for heterogeneous variances and unequal sample sizes (L-test, Graf et al. 1966; cf. Zofel 1992). RESULTS Main types of sounds. Nearly all of the signals analysed in this paper were recorded in an agonistic context, i.e. the accelerometer was placed close to the mouth of a burrow occupied by a crab (“owner”). A conspecific (“intruder”) wandering about in the aquarium either detected the burrow spontaneously or was placed into the mouth of the burrow by the experimenter. In many of these situations one or both crabs emitted vibration signals, sometimes in an antiphonal manner (i.e. duetting with alternation). Sounds were either single ones or sequences of sounds (Fig. 1). Most of the species tested (though not all) produced the usual percussive or rapping sounds or “rolls” of the major cheliped, called “drumwhirls” in this paper and also by Salmon and Atsaides (1968) and by Crane (1975), who also speaks of “major-manus-drum”. All species emitted signals that are regarded as “additional components” of acoustic behaviour (von Hagen 1984) and are, like respiration noises, normally not specifically searched for by investigators. These signals, called “bounces” in this paper and “leg-flicking” by Salmon and Horch (1972), are “hoarse” burst-pulsed sounds that correspond to laterally directed jerks of the whole animal during its subterranean approach towards an opponent, often in a “testing” manner. These sounds are produced by the joint jerking motion of the flexed ambulatories and are rather irregulary shaped and spaced (e.g. Fig. 1 vm). By contrast, the drumwhirls of the major cheliped are highly stylized and “elaborate” signals (e.g. Fig. 1 sg) and are used not only in agonistic encounters, but also as courtship display. “Bounces” (burst-puLsed sounds). Examples of sequences of bounces produced by H. cordiformis and U. vomeris are given in Figure 1 (He, vm; cf. Table 2). Bounces seem to form the only acoustic signals detectable in these two species and U. dampieri (not figured), though in H. cordiformis the bounces may be superimposed by more evenly shaped short ambulatory “trills”, visible in the first and last signal of Figure 1 He and probably identical with the “ambulatory tapping” that Griffin (1968) observed in the field. Bounces were found in all species of Uca studied, though not figured in all cases. Bounces (most often a single one) frequently precede drumwhirls (Fig. 1 hSj, dSj, flj, cc, rs, mb, pp, an), especially in the species of group V (“Ce/HCfl”). Some species (in addition to H. cordiformis, U. vomeris and U. dampieri) produce bounces in short or long sequences. Such sequences occur at least in U. seismella, U. coarctata, U. mjoebergi and U. annulipes (Fig. I SS|, CC|, mb|, an^). Of these, the Australian species 99 H-O. von Hagen Fig. 1. Oscillographic samples of vibration signals of Australian Uca species, two Asian species {U. rosea, U. anniilipes) and Heloecius cordiformis. Explanation of species name abbreviations given, in same order, in Table 1. Scale bars 0.5 seconds. Following part of legend lists type and number of signals; B, bounces; D, drumwhirls (with number of strokes in parentheses): //c, and Hc^ 4 B, first and last superimposed by ambulatory trills. - vm, 8 B. - /g^, 2 D (1 each); /g^, D (2). -p/,, D 3 D (2 each). - hs^, B + D (15); hs^, 2 D (7 and 8). 100 Vibration signals in fiddler crabs Fig. 1 (cont.): Oscillographic samples of vibration signals. - sg^, D (35); sg^, D (14), (both from Darwin). - el^, D (9); el^, D (6). - 6 B; ss^, D (12); Mj, D (5). - cp^, D (4); cp^, D (4). - ds^, D (2); ds^, B + D (3), arrow points to first stroke. D (21);/Zj, B + D (3). - cc,, 2 B + D (6); ccj, B + D (3). - B + D (26); rs^, D (5) + B + D (3). - m/t,, 3 B + D (7); mh^, B + D (16). -pp,, B + D (6) + B + D (5); pp^, B + D (12). - an^, 2 B + D (14) + continuous ambulatory trill; cin^, B + D (10); an^, B + D (13) + B. 101 H-0. von Hagen Table 3. Rapping sounds of Australian male fiddler crabs (results of oscillographic analysis of tape recordings, cf. Fig. 1). W, E: Samples from western and eastern populations, respectively. Species of Uca One (complex) sound ("drumwhirl"): seconds One sound plus subsequent silent interval: seconds Strokes per sound: number Period (one stroke plus subsequent pause): milliseconds Number of sounds measured (males) mean s.d mean s.d mean s.d range mean s.d longidigitum 0.18 ± 0.16 1.33 ± 0.34 1.87 ± 0.87 (1-4) 159 + 34 23 (3) polita 0.12 + 0.12 1.39 ± 0.63 1.58 ± 0.77 (1-5) 163 ± 40 36 (4) hirsutimanus 0.45 ± 0.16 1.86 ± 0.75 10.09 ± 2.78 (5-16) 48 + 5 43 (4) signata (W) 0.90 + 0.52 2.55 ± 0.84 22.03 ± 11.34 (7-45) 41 ± 4 35 (2) signata (E) 1.03 + 0.41 2.99 ± 0.63 20.22 ± 7.09 (8-38) 53 ± 4 23 (3) elegans 0.36 ± 0.17 2.11 ± 1.07 7.30 ± 2.75 (2-11) 54 ± 12 20 (3) seismella 0.36 ± 0.14 1.17 ± 0.44 8.05 ± 2.58 (5-14) 49 ± 3 37 (4) capricornis 0.22 + 0.13 1.71 ± 0.41 3.10 ± 0.97 (2-5) 90 ± 23 20 (3) dussumieri 0.12 ± 0.04 3.24 ± 1.04 2.06 ± 0.44 (1-3) 98 + 25 16 (4) flammula 0.84 ± 0.53 2.36 ± 0.77 13.40 ± 8.89 (2-33) 69 ± 12 57 (5) coarctata 0.32 ± 0.19 1.91 ± 0.58 3.33 ± 1.88 (1-11) 136 ± 27 27 (3) mjoebergi 0.49 + 0.23 2.33 ± 0.80 9.42 + 3.71 (4-21) 56 ± 6 64 (5) perplexa 0.38 ± 0.23 1.85 ± 1.00 7.72 + 3.43 (3-15) 52 ± 7 54 (5) are included in Table 2, which presents some parameters measured. Table 2 is mainly meant to illustrate the speeialized charaeter of the bounces of U. seisniella (cf. Fig. 1 sSj). Long .sequences of bounces (only comparable with those of U. vomeris) are an essential part of its acoustic communication. U. seismella is also cons¬ picuous by its extremely low means and low standard deviations (differing from all other species, p < 0.001). “Drumwhirls” (rapping sounds). Oscillographic examples of rapping sounds are given in Figure 1 for 14 species including the Asian forms U. rosea and U. annulipes. Three Australian species, namely H. cordiformis, U. dampieri and U. vomeris, never emitted rapping sounds. Numbers of strokes listed in the legend of Figure 1 were verified by the running tape; the sudden extreme decrease of pulse amplitude (e.g. Fig. 1 hs„ el, mb,, an|) is due to reduced contact between chela and substratum. Time patterns of the Australian species are to be found in Table 3. The means given in this table are partly based on up to five males (see last column of Table 3) and different localities (Table 1). They have been united after testing their statistical compatibility. In U. signata, samples from Darwin and Cairns had to be treated separately (see below). Species with small numbers of sounds recorded (Table 3) reflect rarity of sound production {U. longidigitum, U. capricornis, U. dussumieri) or faintness of signals ((/. elegaus). Figure 2 and Table 4 were prepared to facilitate comparisons within Table 3. The results can be summarized as follows: I) Relations of body size and repetition rate of strokes. As the repetition rate of the rapping elements (strokes per second, calculated from the period in Table 3) might simply reflect the species-specific body size, the species were arranged according to the maximum carapace width recorded for Australia (Fig. 2, left scale and open circles). This arrangement is in accord with the systematic grouping of Table 1. As expected, the repetition rates (Fig. 2, right scale and filled circles) mostly correspond to the body size (though inversely): the broken line drawn at 17 strokes per second divides the larger and thus “clumsier” and slower species of group I and II from the smaller and more agile species of group III and V. Within this scheme the top positions are held by U. flamniula and the western U. signata sample, respectively. The two species confirm the advanced character of their acoustic signals also by the high number of strokes per sound (Table 3 and Fig. I sg|, n,). On the other hand, the repetition rates of U. longidigitum and U. polita do not fit the scheme at all: the two species have the lowest repetition rates, though they belong to the smaller species (middle and right sections of Fig. 2). The exceptional status of the two species is stressed by their low number of strokes per sound (Table 3 and Fig. I Ig, pi). For a weighting of these eharacters see below (Discussion). 2) In search of effects ofallopatry and sympatry. Most of the 17 confirmed Australian species of Uca are confined to one of the two northern zoogeographic provinces: the Dampierian in the west and the Solan- derian in the east, seen from the Torres Strait region (von Hagen and Jones 1989; cf. Table 1 of the present paper). Some of these allopatric species are still similar, i.e. they form inter-provincial pairs or groups of relatives. The 102 Vibration signals in fiddler crabs Maximum carapace width in Australia (mm) Repetition rate (strokes per second) Fig. 2. Body size (left scale, open circles) and repetition rate of rapping strokes (right scale, filled circles) in 12 species of Australian Uca (abbreviations of species names and Roman group numerals as in Table I). Values of U. signata separated according to a western (W) and eastern (E) sample. Maximum carapace widths in Australia after George and Jones (1982), except for U. seismella (Karumba sample of author). Repetition rate calculated from the period of strokes in Table 3. Table 4. Statistical comparison of allopatric and sympatric relatives {Uca spp.) and populations of U. signata (W western, E eastern) using two parameters of Table 3. ***, **, *: differences significant; p < 0.001, < 0.01, •cO.05, respectively; n.s. not significant (L-test). See Table 1 for geographic distribution of the species mentioned. Strokes Period of per sound strokes Allopatric relatives: U. capricornis VS. U. dussumieri n.s U. capricornis vs. U. coarctata n.s. *** U. flammula vs. U. coarctata U. mjoehergi vs. U. perplexa * ** Svmoatric relative.s: U. longidigitum vs. U. polita n.s. n.s. U. elegans vs. U. hirsutimanus * U. seismella vs. U. polita *** U. capricornis vs. U. flammula *** U dussumieri vs. U. coarctata Populations of U. sienata: U. signata (W) vs. U. signata (E) n.s. U. signata (W) vs. U. hirsutimanus *** U. signata (E) vs. U. hirsutimanus *** *** U. signata (W) vs. U. elegans *** U. signata (E) vs. U. elegans *** n.s. four species (U. polita, U. signata, U. seismella, U. triangularis) that occur in both provinces appear to have extended their range secondarily. They are one reason for the fact that close relatives occur not only in allopatric but also in sympatric distribution (Table 4). It is to be expected that allopatric relatives are rather “careless” of evolving species-specific vibration signals: there is no selection against retaining signals that antedate geographic separation. In fact three of the four allopatric pairs listed in Table 4 are similar (differences n.s. or p < 0.05 only) in one of the parameters chosen for comparison (strokes per sound and period of strokes). Sympatric relatives should reinforce selectivity of their communication system, and the frequency of significant differences in both parameters mainly appears to fit this concept (Table 4). The pair U. longidigitum vs. IJ. polita (Table 4) forms a puzzle, as both species are reported from Sandgate and Shorncliffe in Moreton Bay near Brisbane (George and Jones 1982). In 1986 the author found the two species at Shorncliffe on the same mud-flat opposite to the mouth of Nundah Creek, but not a single individual of U. polita in the many longidigitum-hahitats of the Brisbane River itself. Though both are members of the Solanderian province, the two species seem to meet rarely in the field so that their different time pattern of waving display (Pellikan 1990) may be sufficient as a behavioural isolating mechanism. 103 H-0. von Hagen The western and eastern samples of U. signata (from Darwin and Cairns) were treated separately throughout the paper, because they had clearly different periods of strokes (p < 0.001, Table 4). Each population of U. signata was compared with the two related western (Dampierian) species U. hirsutimanus and U. elegans. Indications of character displacement (i.e. the same two species differ more in sympatry than in allopatry) were not found between U. signata and U. hirsutimanus, but between U. signata and U. elegans (Table 4). The latter may be the closest relative of V. signata (Pellikan 1990). 3) Comparison with signals of two Asian species. The two Uca species of South East Asia that were included in this study fit easily into the systematic grouping of the Australian species (Table 1). However, within these groups the Asian species appear to be more specialized. The number of strokes in isolated sounds of U. rosea (e.g. 26 in Fig. 1 rs^ mean 17.06 ± 6.08, n = 17 drumwhirls of this long type) is only comparable with that of U.flammula (Table 3). There are shorter sounds in both species (Fig. I fl,, rs^), but only U. rosea can “switch” to long .sequences of short drumwhirls (up to 12 sounds measured, instead of up to 3 in U.flammula). Likewise, U. annulipes is more specialized than the two related Australian species U. mjoebergi and U. perplexa. The number of strokes per sound is similar to that of U. mjoebergi and U. perple.xa, but the period of strokes is more variable (cf. anj, with an^ and an^ in Fig. 1). In U. annulipes the drumwhirls are often followed (or preceded) by continuous burst-pulsed “trills” (Fig. 1 aOj) of the walking legs, lasting up to about ten seconds. These trills were also recorded in U. perplexa, but more rarely, and they appear to be missing in U. mjoebergi. DISCUSSION In explaining the Figures and Tables in the section on “drumwhirls”, a few conclusions were attempted that are more or less well supported. It is clear that body size and repetition rates of strokes are inversely related (with a few striking exceptions), but the second idea that sympatry and allopatry should inlTuence the signals needs to be tested by closer pair-by-pair examinations (cf. von Hagen 1984). A more detailed study of the U. longidigitum-U. polita pair and of the western and eastern populations of U. signata should be especially rewarding. Work on U. signata should also include samples from New Guinea, where the species occurs at least around Merauke, Irian Jaya (Leiden Mu.seum, Netherlands: e.g. No. D32522, D33526; von Hagen unpublished). It is also premature to make general conclusions on the advanced status of the signals of Asian fiddler crabs (see third section). The two species chosen (U. rosea, U. annulipes) might be specialized species by chance. There exist, however, some additional bioacoustic data: of U. paradussumieri Bott (= U. spinata Crane) and U. dussumieri from Malaysia and Indonesia (R. Polivka pers. com.). Judging only from their range of stroke numbers per sound (3-21 and 2-6, respectively) these Asian forms, too, appear to be more advanced than their Australian relatives. As is evident from examination of Figure I and Table 3 the sounds of Australian U. dussumieri as well as of U. capricornis and U. coarctata are among the simplest signals recorded. These species apparently hold a basal position within the “De//HCfl”-group, with regard to its Australian as well as to its Asian members. Only members of Australuca (group Ill), namely U. longidigitum and U. polita, are even simpler with respect to their signals: by their mean number of strokes (minimum in Table 3) and their long stroke period (maximum in Table 3, resulting in an exceptional low repetition rate. Fig. 2). Especially U. longidigitum (Fig. I, Ig), with the distribution pattern of a relict species (see George and Jones 1982), could have preserved the earliest phylo¬ genetic stage of drumwhirl invention in Uca, namely rapping as a by-product of waving the major cheliped. Within Australuca, U. longidigitum is the only species in which the length of one sound plus subsequent interval (1.33 ± 0.34, n = 23, Table 3) is statistically not separable from the length of one waving motion plus subsequent pause (1.22 ± 0.11, n = 36, according to motion pictures taken by the author and analysed by Pellikan 1990). Even in the aquarium U. longidigitum tended to switch from waving to rapping (and vice versa), the latter looking like a reduced or weak waving movement. Vibration components of visual display are well- known in Uca (Altevogt 1964), but in U. tangeri, a species that also possesses a complete system of drumwhirls. U. longidigitum seems to have but one gesture that serves two sen.sory channels - depending on its intensity or completeness. It is important to note that the nearly all Australian “Auslraluca”-group (which includes U. longidigitum) retains, as a whole, some other ancestral traits: above all, minor chelae with large teeth (Crane 1975) and waving display of females (von Hagen 1993). It is not clear, whether the “Thcdassuca”-gToup, which also has waving females (von Hagen 1993), holds a still more ancestral position with regard to sound evolution in Uca. The two Australian members, U. dampieri and U. vomeris, never emitted drumwhirls within the experimental design of the present study, though they should have: Crane (1975) recorded, in the field, underground rapping sounds in the close relatives U. hesperiae from East Africa and U. pacifensis from Fiji (see her Plate 47 A) and even in U. vomeris from New Caledonia; and Salmon (1984) describes a “shuttling” motion of the major chela in U. vomeris from Townsville, 104 Vibration signals in fiddler crabs used by intruders in the entrance of another crab’s burrow. A gesture of this kind was filmed in Cairns in a male of U. vomeris fighting on the surface of the ground (von Hagen, unpublished). Furthermore, the American narrow-fronts {Uca sensu stricto) are thought to be a branch of “Thalasstica" (Crane 1975), and they have genuine drumwhirls (e.g. von Hagen 1971, 1972). However, even without certainty in the case of U. dampieri and U. vomeris, Australia remains the continent where all stages of sound evolution are to be found: H. cordiformis without drumwhirls, U. dcimpieri and U. vomeris apparently without proper drumwhirls or normally not using them underground, U. longidigitum and U. polita with drumwhirls of a very early stage, U. capricornis, U. dussumieri and U. coarctata with simple sounds not much more progressive than those of U. polita, and after an intermediate stage two especially advanced positions, which evolved independently out ot'Australuca (U. signata) and “Deltuca” (U.flammula). Among the intermediate forms (not specified here) two species have, in addition to their drumwhirls, evolved peculiarities: U. seismella has specialized in long sequences of “hasty” bounces and U. perplexa (like U. anniilipes) in continuous ambulatory trills. Housing all stages of sound evolution, especially the early ones, and also the early stages of waving evolution (von Hagen 1993), Australia should be favoured as the continent where the genus Uca first evolved. The author is aware that directional change in behavioural complexity cannot be the only indicator in phylogenetic reasoning. A support at least by morpho¬ logical studies is necessary (Beinlich and von Hagen in prep.). However, this holds true also for other ap¬ proaches. Two related papers, which were published some years ago (Levinton et al. 1996, Sturmbauer et al. 1996), base their phylogenetic trees on certain analyses of 16S rDNA alone. Their inferences, above all an American centre of origin of Uca, should be treated with reserve, as they lack a proper morphological background for their discussions, which neglect for example, the most reliable synapomorphy in Uca systematics: the unique pleonal clasping apparatus identifying all broad-fronted species as a monophylum (Guinot 1979, Beinlich and von Hagen in prep.). The assumption that Uca first evolved in Australia is, furthermore, supported by an Australian origin of certain mangrove taxa (palynological evidence compiled in Specht 1981). As Uca is typically associated with mangroves, it is essential that Rhizophora, Avicemiia and Sonneratia first appeared in the late Eocene of West Australia and colonized Indonesia not before Oligocene/ Miocene - probably accompanied by fiddler crabs, which made South East Asia not their first, but their second centre of evolution. Crane, who considers “Deltuca” the subgenus “least specialized” (Crane 1975) and at first supposed a Malaysian origin of the genus (Crane 1956), finally called Australia’s northwest coast “the most conservative area, judging by the morphology of the local Uca" (Crane 1975). As the east coast has its ancestral forms as well (H. cordiformis, U. longidigitum, U. coarctata), one should extend Crane’s assessment to the whole continent. ACKNOWLEDGMENTS The author wishes to thank the following persons and institutions: the Northern Territory Museum (NTM, Darwin) and Mr N. Roche (Police Station, Bamaga) for generously providing a field vehicle; Dr A.J. Bruce and Dr J.R. Hanley (formerly of NTM), Mr P. Davie (Queensland Museum, Brisbane) and Mrs and Mr R. and I. Vela (Bamaga) for their kind and constructive help, e.g. in finding Uca habitats; Ms S. Dietrich and Mr P. Heimes (Marburg) for providing live U. rosea and U. annulipes', Mrs S. Gotzfried (Marburg) for many-faceted technical assistance. REFERENCES Altevogt. R. 1963. Akustische Epiphanomene im Sozialverhalten von Uca langeri. Verhandlungen der Deutschen Zoologischen Gesetlschaft in Wien 1962: 309-315. Altevogt. R. 1964. Ein antiphoncr Klopfkode und eine neue Winkfunktion bei Uca tangeri. Naturwissenschaften 51: 644-645. Altevogt, R. 1970. Form und Funklion der vibratorischen Signale von Uca tangeri und Uca inaequalis (Crustacea, Ocypodidae). Forma et/Mnct/o (Braunschweig) 2: 178-187. Bott, R. 1973. Die verwandlschaftlichen Beziehungen der Uca- Arlen (Decapoda: Ocypodidae). Senckenhergiana hiologica (Frankfurt) 54: 315-325. Crane. J. 1956. In Malaya. Looking for clues to fiddler crab evolution. Animal Kingdom 59: 131-136. Crane, J. 1975. Fiddler crabs of the world. Ocypodidae: genus Uca. Princeton University Press: Princeton. George, R.W. and Jones, D.S. 1982. A revision of the fiddler crabs of Australia (Ocypodidae: Uca). Records of the Western Australian Museum Supplement 14: 1-99. Graf, U., Henning, H.J.. and Stange. K. 1966. Formeln und Tahellen der mathematischen Statistik. Springer: Berlin. Griffin, D.J.G. 1968. Social and maintenance behaviour in two Australian ocypodid crabs (Crustacea: Brachyura). Journal of Zoology, London 156: 291-305. Guinot, D. 1979. Donnces nouvelles sur la morphologie, la phylogenese et la taxonomic des crustac^s decapodes brachyoures. Memoires du Museum National d’Histoire Naturelle A, 112: 1-354. Hagen, H.O. von 1962. Freilandstudien zur Sexual- und Fortpflanzungsbiologie von Uca tangeri in Andalusien. Zeitschrift fiir Morphologie und Okologie der Tiere 51:611- 725. Hagen, H.O. von 1970. Die Balz von Uca vocator (Herbst) als iikologisches Problem. Forma et functio 2: 238-253. Hagen, H.O. von 1971. Uca maracoani (Ocypodidae). Balz. Film E 1423 des Instituts fiir den Wissenschaftlichen Film Gottingen. Filmbeschreibung: 1-20. 105 H-O. von Hagen Hagen, H.O. von 1972. Uca mayor (Ocypodidae). Balz. Film E 1422 des Instituts fur den Wissenschaftlichen Film Gottingen. Filmbeschreibung: 1-19. Hagen, H.O. von 1975. Kla.ssifikation und phylogenetische Einordnung der LautauBerungen von Ocypodiden und Grapsiden (Crustacea, Brachyura). Zeitschrift fiir zoologische Systematik und Evolutionsforschung 13: 300- 316. Hagen, H.O. von 1984. Visual and acoustic display in Uca mordax and Uca burgersi, sibling species of Neotropical fiddler crabs. II. Vibration signals. Behaviour 9\: 204-228. Hagen, H.O. von 1993. Waving display in females of Ucapolita and of other Australian fiddler crabs. Ethology 93: 3-20. Hagen, H.O. von, and Jones, D.S. 1989. The fiddler crabs (Ocypodidae: Uca) of Darwin, Northern Territory, Australia. The Beagle, Records of the Northern Territory Museum of Arts and Sciences Cr. 55-68. Levinton, J., Sturmbauer, C. and Christy, J. 1996. Molecular data and biogeography: resolution of a controversy over evolutionary history of a pan-tropical group of inver¬ tebrates. Journal of Experimental Marine Biology and Ecology 203: 117-131. Muller, W. 1989. Untersuchungen zur akustisch-vibratorischen Kommunikation und Okologie tropischer und subtropischer Winkerkrabben. Zoologische Jahrhiicher Abteilung fiir Systematik Okologie und Geographie der Tiere 116: 47- 114. Pellikan, C. 1990. Verwandtschaftsforschung an indo- australischen Winkerkrabben der Untergattung Australuca sensu Crane. Diplomarbeit (unpublished MSc thesis). Philipps-Universitat: Marburg. Salmon. M. 1965. Waving display and sound production in Uca pugilator, with comparisons to U. minax and U. pugnax. Zoologica (N.Y.) 50: 123-150. Salmon, M. 1983. Acoustic ‘calling’ by fiddler and ghost crabs. Memoirs of the Australian Museum 18: 63-76. Salmon, M. 1984. The courtship, aggression and mating system of a "primitive” fiddler crab (Uca vocans: Ocypodidae). Transactions of the zoological Society of London 37: 1- 50. Salmon, M. and Atsaides, S.P. 1968. Visual and acoustical signalling during courtship by fiddler crabs (genus Uca). American Zoologist 8: 623-639. Salmon, M. and Horch, K.W. 1972. Acoustic signalling and detection by semi terrestrial crabs of the family Ocypodidae. In: H.E. Winn and B.L. Olla (eds) Behaviour of Marine Animals. Vol. 1: 60-96. Plenum Press: New York. Specht, R.L. 1981. Biogeography of halophytic angiosperms (salt-marsh, mangrove and sea-grass). In: A. Keast (cd). Ecological biogeography of Australia. Vol. 1: 575-589. W. Junk: The Hague, Boston, London. Sturmbauer, C., Levinton, J.S. and Christy, J. 1996. Molecular phylogeny analysis of fiddler crabs: test of the hypothesis of increasing behavioral complexity in evolution. Proceedings of the National Academy of Science of the United States of America 93: 10855-10857. Turkay, M. 1983. The systematic position of an Australian mangrove crab Heloecius cordiformis (Crustacea: Decapoda: Brachyura). Memoirs of the Australian Museum 18: 107-111. Zdfel, P. 1992. Statistik in der Praxis. 3. Auflage. G. Fischer: Stuttgart und Jena. Accepted 4 June 2000 106 The Beagle, Records of the Museums and Art Galleries of the Northern Territory, 2000 16: 107-122 Caetrathynnus, Nitidothynnus and Procerothynnus, new genera of Thynninae (Hymenoptera: Tiphiidae) from northern Australia G.R. BROWN Museum and Art Gallery of the Northern Territory GPO Box 4646, Darwin NT 0801, AUSTRALIA Current address : Department of Primary Industry and Fisheries GPO Box 990, Darwin NT0801, AUSTRALIA graham.hrown @ nt.gov.au ABSTRACT Three new genera are erected to accommodate nine new species of thynnine wasps: Caetrathynnus for C. greggi sp. nov. (type species), C galhinus sp. nov. and C. wesselensis sp. nov.; Nitidothynnus forN. purdiei sp. nov. (type species), N. eheneus sp. nov. and N. .spinulus sp. nov.; and Procerothynnus forP. centralianus sp. nov. (type species), P. arnhemicus sp. nov. and P. carpentarianus sp. nov. All the new species are described from males only, except Nitidothynnus spinulus and N. purdiei, which are known from both males and females. Keys to species are given, as is a discussion of coupling mechanisms in Nitidothynnus and the phenomenon of mixed species couples is considered in light of the biology of the subfamily. Most of the new species are recorded only from the Top End of the Northern Territory, although Caetrathynnus galhinus also occurs in adjacent areas of northern Western Australia and Procerothynnus carpentarianus also occurs in adjacent areas of north-western Queensland. Procerothynnus centralianus is recorded from central Australia while Nitidothynnus eheneus is known only from the holotype from northern Queensland. Keywords. Tiphiidae, Thynninae, new genera, Caetrathynnus, Chilothynnus, Nitidothynnus, Procerothynnus, Aspidothynnus, Australia, taxonomy, biogeography. INTRODUCTION Wasps of the subfamily Thynninae comprise a very diverse group in Australia which includes many undescribed species and genera (Brown 1998a). Some 600 species have been described, mostly from the vicinity of the large coastal cities in eastern, south¬ eastern and south-western mainland Australia (Given 1954; Salter 1954). Very few have been described from northern or inland Australia, and there are only 26 known from the Northern Territory. Adults are most commonly observed feeding on flowers (hence the common name “flower wasps”) where they often occur in copula. The main food source appears to be nectar, although other plant and insect secretions and exudates are also eaten (Tillyard 1926; Burrell 1935). Nectar from flowers of the plant family Myrtaceae, which includes eucalypts and tea trees, is particularly attractive, and may attract these wasps in large numbers at the height of the flowering season. Male thynnines are fully-winged, while females are smaller, wingless, and somewhat ant-like in appearance. Because females are wingless, and cannot therefore travel very far by themselves, they are dependant on males for food. Feeding occurs during mating as pairs couple for long periods during which the male carries the female to flowers. Pairs remain coupled while both are feeding, and there are morphological structures present that assist pairs to remain coupled for prolonged periods. Since the sexes are so different morphologically, it is the convention to describe each separately. Unfortunately in the de.scriptions that follow this is not possible for every species because of the lack of females. Females produce pheromones to attract males prior to mating. However, on rare occasions when a conspecific male never arrives, a female may couple with a non-conspecific male (Brown 1993). This is a common occurrence in this subfamily and is most probably a mechanism that ensures a hungry female is carried to a food source at the expense of successful mating. The latter may be of little concern to the survival of the species as a whole because unfertilised females of all Hymenoptera (ants, bees and wasps) produce male offspring which are haploid anyway (Naiimann 1991). This phenomenon of miscoupling is mentioned spe¬ cifically under the descriptions of Nitidothynnus purdiei and N. spinulus in this paper. The Thynninae has not been revised taxonomically since 1910 (Turner 1910) and the only critical examinations since then have been revisions for the 107 G.R. Brown genera Acanthothynniis Turner, Doratithynnus Turner, Encopothynniis Turner and Macrothynnus Turner (Brown 1987, 1989a, 1989b, 1995a). These revisions indicated that, in at least these genera, between 50% and 75% of the fauna currently represented in collections was undescribed and that many species were known from only one or two specimens. This suggests that more extensive collecting, especially in more remote areas such as the Northern Territory of Australia, will reveal even more new species. Currently there are 37 genera considered to be valid within the tribe Thynnini in Australia (Given 1954; Brown 1983, 1992, 1995b, 1997a, 1997b, 1998b). Studies by the present author indicate that these genera form three distinct clusters. The first cluster contains the genera Bifidothynmis Brown, Campylothynnits Turner, Catocheilus Guerin, Elidothynnus Turner, Guerinius Ashmead, Leptothynnus Turner, Lestricothynnus Turner, Lophocheilus Gudrin, Macrothynnus Turner, Megalothynnus Turner, Oncorhinolhynnus Salter, Pogonothynmis Turner, Thynnoides Gudrin, Thynnus Fabricius and Zaspilothynnus Ashmead. It is distin¬ guished from the other two generic clusters by numerous characters including the labrum weakly rather than strongly narrowed basally, the propodeum oblique in profile, tergite 7 longitudinally multicarinate (except in Bifidothynnus and Guerinius which have tergite 7 otherwise modified) and sternite 8 with prominent basal angles or spines in the male, and sternite 5 rugose and/ or carinate in the female. Most Australasian species are also larger in size than those found in the other two clusters. The second cluster inc\udes Acanthothvnnus Turner, Arthrothynnus Brown, Aspidothynnus Turner, Doratithynnus Turner, Encopothynniis Turner, Epactiothynnus Turner, Gymnothynnus Turner, Iswaroides Ashmead and Tmesothynnus Turner. It dif¬ fers from the subsequent cluster by having most metasomal segments strongly constricted (and sometimes spinose), and tergite 7 with a strong transverse apical Carina. The third generic cluster includes Aeolothynnus Ashmead, Agriornyia Gudrin, Ariphron Erichson, Chilothynnus Brown, Leiothynnus Turner, Neozelehoria Rohwer, Pentazelehoria Brown, Phymatothynnus Turner, Psammothynnus Ashmead, Tachynoides Kimsey, Tachynomyia Gudrin, Tachyphron Brown, and Zelehoria Saussure. This cluster is distinguished from other Australian Thynnini (and Thynninae) by the combination of: a relatively short, weakly protruding sternite 8 that is not broadened into lobes or spines at the base, and which may be armed apically with a narrow spine (with or without lateral spines); the metasomal segments are not strongly constricted or heavily sclerotized; a convex tergite 7 that is uniformly punctate with a medial impunctate area, but without a transversely carinate apical margin; and the basiparameres are not strongly developed ventrobasally such that the ventro- basal angle (in profile) is well separated from the basal ring. A key to genera in this cluster was given by Brown (1998a), although Tachynoides was omitted. That genus keys to Tachyphron in that key. The three new genera described here belong to the third cluster. Two of these were included in the key to genera by Brown (1998a) and referred to as genus A and genus B. Terminology follows Snodgrass (1941), Brown (1997a,b) and Naumann (1991). Relative terms relating to microsculpture are interpreted as follows: sparsely punctate = punctures greater than two puncture- diameters apart; punctate = punctures at most two puncture-diameters apart, but never confluent; closely punctate = punctures almost confluent; rugosely punctate = punctures partially confluent; finely punctate = punctures small and shallow; coarsely punctate = punctures large and deep; obscurely punctate = punctures small, sparse, shallow and only visible at certain angles. Abbreviations. Morphological characters: Tl-7, metasomal tergites 1-7; SI-8, metasomal sternites 1-8; POL:OOL, ratio of distance between posterior ocelli to distance between eye and posterior ocellus. Specimen repositories: AM, Australian Museum, Sydney; ANIC, Australian National Insect Collection, CSIRO, Canberra; BMNH, The Natural History Museum, London; MV, Museum of Victoria, Melbourne; NTM, Museum and Art Gallery of the Northern Territory, Darwin; OTTAWA, Agriculture Canada, Ontario; QM, Queens¬ land Museum, Brisbane; SAM, South Australian Museum, Adelaide; UQIC, University of Queensland, Brisbane; WAM, Western Australian Museum, Perth. SYSTEMATICS Caetrathynnus gen. nov. Type species. Here designated Caetrathynnus greggi sp. nov. Generic diagnosis. Male. Head, mesosoma and metasoma strongly polished with metasoma shagreened. Mandibles long, narrow and straight. Clypeus convex, strongly produced and broadly truncate apically. Antennal prominence long, narrowly U-shaped with margins slightly emarginate dorsally, sagittally sulcate, and with short oblique carina above antennal insertions. Frons strongly and broadly depressed lateral to antennal prominence. Antennae short and only just reaching propodeum, apical 6 segments weakly arcuate. Pronotum with anterior margin weakly raised not carinate, lateral margins strongly convergent anteriorly. Mid femur strongly and abruptly broadened at base. Metasoma 108 New thynnine wasp genera fusiform, segment 1 subpetiolate, about as wide as long; segments 3-4 widest, segments (apart from T2 and S2 anteriorly) not constricted. T7 convex, slightly produced and membranous apically, not transversely apical carinate. S8 subtriangular with a single narrow slightly upturned apical spine, narrowly emarginate at base of spine. Genitalia with basal ring cylindrical, not short; basiparameres suborbicular to subtriangular (viewed dorsally), not apically emarginate; parameres subparallel, of moderate length and width, apex subtruncate; cuspides of moderate length with opposing digitate digitus apically. Remarks. Caetratbynniis is most readily distin¬ guished from other thynnine genera (in particular, those of the third cluster) by the structure of the mandibles which are long, narrow and straight, and the clypeus which is broadly convex without a sagittal carina, and with the apical margin strongly produced and broadly truncate. The broadly convex and broadly apically truncate clypeus suggests a relationship with Psammo- thynnus and Zeleboria. However, in both of these genera, the mandible is shorter, broader and curved (as normally found in other genera) and the clypeus is shorter, more strongly convex and sagittally carinate. These differences in clypeal shape suggest that an enlarged clypeus may have arisen in Caetrathynnus independently to that of Psammothynnus and Zeleboria, and is supported by differences in the genitalia which differ from Caetra¬ thynnus by having the basal ring very short, the fused basiparameres truncate to emarginate dorsoapically and, in many species, the parameres with a ventroapical lobe. These two genera also lack a single medial spine on S8 which occurs in Caetrathynnus and many other Thynninae. Caetrathynnus superficially resembles Aspidothynnus Turner in that: the clypeus is broadly convex, and strongly and broadly produced apically without a sagittal Carina; and the metasoma is fusiform without spines on the tergites or sternites. However, Aspidothynnus belongs to the second generic cluster and as such differs by those characters as listed in the Introduction. Aspidothynnus also differs from Caetrathynnus by having the antennal prominence weakly developed rather than long, narrowly U-shaped with margins slightly emarginate dorsally; T1 wider than long rather than as approximately as wide as long; and metasomal segments with spots rather than lunulate marks. The three species of Caetrathynnus are very similar in appearance, but differ slightly in the punctation of clypeus, gena and pronotum, and the shape of the fore coxae, the depth of the anterior depression on S2 (typical of all Thynninae), the base of the apical spine of S8, and the genitalia. They are reliably distinguished by the genitalia. All three are recorded from the north of the Northern Territory. Caetrathynnus galbinus sp. nov. is recorded from near the Western Australian border while C. wesselensis sp. nov. is known only from the holotype from the Wessel Islands. C. greggi sp. nov. is the most widely occurring species, and it ranges from Darwin to Borroloola. Caetrathynnus was referred to as genus “A” by Brown (1998a) in a key to closely related genera. Etymology. The generie name is masculine and is derived from the Latin word caetra (also spelt cetra) which is a type of shield. It is a reference to the shape of the male clypeus. Key to males of Caetrathynnus, (Females are unknown.) 1. a Most sclerites predominantly yellow; prosterna black with yellow spot; propodeum with U-shaped lateral marks at least partially fused medially .C. galbinus sp. nov. b Sclerites predominantly black (may have exten¬ sive yellow marks but not distinctively more yellow than black); prosterna black without yellow spot; propodeum with U-shaped lateral marks not fused medially.2 2. a Apex of parameres truncate (Fig. 3); base of cuspis sinusoidal (in profile). .C. wesselensis sp. nov. b Apex of parameres rounded (Fig. 2); base of cuspis rounded (in profile).C. greggi sp. nov. Caetrathynnus galbinus sp. nov. (Fig. 1) Type material. HOLOTYPE cT - NTM (1911): Keep River NP, Garrandalng, 15°53’S, 129°03’E, Northern Territory, 30 April 1996, G.R. Brown. PARATYPES - Western Australia’. UQIC (3d'), 14 mi. (22.5 km) S of Kununurra, on Eucalyptus '/pruinosa, 5 March 1973, E.M. Exiey. Description of male. Body length 11 mm; fore wing 8 mm; hind wing 6 mm. Clypeus rugosely punctate, punctures shallow and vertically aligned, apical margin width to maximum width 1:2.0. POLiOOL 1:1.1. Frons closely and shallowly punctate. Vertex obscurely punctate. Gena shallowly and closely punctate. Pronotum sparsely and shallowly punctate, punctures more distinct posterolaterally. Mesoscutum and mesoscutellum sparsely punctate medially, closely punctate laterally. Metanotum punctate. Propodeum closely and shallowly punctate, punctures deeper laterally, impunctate anteriorly. Mesopleura closely to finely punctate. Fore coxae flat. T1 width to length 1:0.9. T2 weakly depressed anteriorly. Tergites and sternites closely and very shallowly punctate such that most tergites and anterior sternites almost appear impunctate, punctures coarser on posterior tergites. SI not medially raised. S2 strongly 109 G.R. Brown Figs 1-6. Caetrathvnnus spp., males: 1, C. galbinus sp. nov., genitalia, lateral and dorsal; 2, C. greggi sp. nov., genitalia, lateral and dorsal; 3, C. wesselensis sp. nov., genitalia, lateral and dorsal; 4, C. greggi sp. nov., habitus; 5, C. greggi sp. nov., S8; 6, C. greggi sp. nov., right mid trochanter and femur. A, basal ring; B, basiparameres; C, cuspis; D, digitus; E, aedeagus; F, paramere. Scale lines = 0.5 mm. no New thynnine wasp genera depressed anteriorly. S8 slightly emarginate at base of spine. Genitalia as in Figure 1: basiparameres sub- triangular (viewed dorsally) with each basiparamere truncate apically; parameres relatively short, subtruncate apically, weakly dorsally lobed (in profile) near base of aedeagus; cuspis subparallel basally, apically triangular, apex ending slightly beyond apex of digitus and outwardly curved, base rounded and slightly visible in profile. Colour. Black; yellow colouration as follows: mandibles (except apex), stipes, clypeus (except obscure medial anchor-shaped mark), large longitudinal mark (except very narrow sagittal line) above antennal insertion, inner and outer orbits of eyes with outer orbit almost continuous with inner orbits and extending discontinuously across vertex, anterior margin of pronotum broadly (very narrowly discontinuous medially), posterior margin of pronotum broadly, tegulae, mesoscutum with large spot on disc and horizontal mark above fore wing, large spot on disc of mesoscutellum, axillae, metanotum, metapleuron, broad lateral U-shaped mark on propodeum, mesopleura (except mesopleural groove), spot on prosterna, margins of mesosternal lamellae, coxae (except base of mid and hind coxae), outer surfaces of trochanters, femora and fore tibiae, dorsal surface of fore femora, Tl-6 (except anteromedially and posterolateral spot), .sagittal line on SI, S2 (except anteromedial and posterolateral spots), curved lateral mark on posterior margin of S3-6 (almost confluent medially); orange colouration as follows: legs (except coxae outer surfaces of trochanters, femora and fore tibiae, dorsal surface of fore femora). Margin of tegulae semitransparent. Wing membranes hyaline with veins orange to brown. Setae white. Distribution. Known only from the Northern Territory-Western Australian border near Kununurra. Remarks. This species is distinguished from other species of Caetrathynnus by the extent of the yellow colouration such that most sclerites are predominantly yellow, including the presence of a spot on the prosterna, and enlarged U-shaped lateral markings on the propodeum which are at least partially fused medially. It is also distinguished by having the fore coxae fiat, the clypeus with punctures vertically aligned, and the pronotum with punctures most distinct posterolaterally. Yellow colouration may be expanded on the clypeus, and on the propodeum so that marks are confluent medially, or slightly reduced on the pronotum, propodeum and metasoma. Black on the metasoma may be partially replaced by brown, or it may be expanded such that it is confluent with the black posterolateral spots on some segments. Etymology. The specific name is derived from the Latin word galbinus (= yellowish) and it refers to the extensive yellow colouration of the male. Caetrathynnus greggi sp. nov. (Figs 2, 4-6) Type material. HOLOTYPE cf - NTM (1895): 17 Mile, Virginia, I2°33’S, 131°02’E, near Darwin, Northern Territory, 28 April 1996, S. Gregg. PARA- TYPES - Northern Territory: AM (K122050), BMNH (E2000-27), MV (T17352), NTM (1896-1898), QM (T57838), SAM (1 21437), WAM (26555) {9cf), same data as holotype (3 NTM, 1 AM), 4 May 1996(1 BMNH, 1 MV), 19 May 1996 (I QM, 1 SAM), December 1997 (1 WAM); NTM (I899-I90I) (3cr), Arafura Swamp, I2“16’S, 124°59’E, 9 June 1996, G.R. Brown; NTM (1902-1903), OTTAWA (3cr), Berry Springs 50 km S Darwin, rainforest. Malaise trap, 4-27 December 1993, S. and J. Peck; ANIC (IcT), 12 km NNE of Borroloola, 15°58’S, 136°21’E, I November 1975, J.C. Cardale; NTM (1904) (Icf), McMillans Rd [Berrimah] near Darwin, eucalyptus woodland. Malaise trap, 1-25 December 1993, S. and J. Peck; NTM (1905), OTTAWA (2cf), Berrimah near Darwin, mixed eucalyptus woodland, 25 December 1993-10 January 1994, S. and J. Peck; NTM (1906) (Icf), Mindil Beach near Darwin, 13 May 1996, S. Gregg; NTM (1907) (IcT), Melville I., swamp W of Taracumbi Falls, 1I°35’S, 130°40’E, 4 October 1996, G.R. Brown and G. Dally; NTM (1908) (Icf), Melville I., Mirikau-Yunga Ck, ll°3rS, 130°4rE, 5 October 1996, G.R. Brown and G. Dally; NTM (1909) (IcT), Melville I., 11°35’S, 130°40’E, Malaise trap, 4-15 October 1996, G.R. Brown; NTM (1910) (Icf), Mt Mortgage, Humpty Doo, 12°35’S, 131 °05’E, 28 February-31 March 1992, A. Wells and J. Webber; ANIC(lcr), !9kmNE of Mt Cahill, 12°47’S, l32°5rE, 16 November 1972, J.C. Cardale. Description of male. Body length 8-12 mm; fore wing 6-9 mm; hind wing 4.5-6 mm. Clypeus rugosely and shallowly punctate, apical margin width to maximum width 1:1.6. POL:OOL 1:0.9. Frons closely and shallowly punctate. Vertex obscurely punctate. Gena shallowly and closely to rugosely punctate. Pronotum .sparsely and shallowly punctate, punctures more distinct on anterior and posterior margins. Mesoscutum and mesoscutellum sparsely punctate medially, closely punctate laterally. Metanotum punctate. Propodeum closely and shallowly punctate, punctures deeper laterally, impunctate anteriorly. Mesopleura closely to finely punctate. Fore coxae almost flat. T1 width to length 1:1.0. T2 weakly depressed anteriorly. Tergites and sternites closely and very shallowly punctate such that most tergites and anterior sternites almost appear impunctate, punctures coarser on posterior tergites. SI not medially raised. S2 depressed anteriorly. S8 emarginate at base of spine. Genitalia as in Figure 2: basiparameres broadly subtriangular (viewed dorsally) with each basiparamere rounded apically; parameres relatively long, rounded apically, distinctly dorsally G.R. Brown lobed (in profile) near base of aedeagus; cuspis subparallel basally, apically triangular, apex ending slightly beyond apex of digitus and outwardly curved, base rounded and strongly visible in profile. Colour. Black; yellow colouration as follows: mandibles (except apex), clypeus (except medial anchor¬ shaped mark), large longitudinal mark (except very narrow sagittal line) above antennal insertion, inner and outer orbits of eyes almost continuous dorsally, transverse line behind ocelli, margins of pronotum (anterior margin very narrowly discontinuous medially, and expanded ventrally), tegulae, mesoscutum with large spot on disc and horizontal mark above fore wing, central and anterolateral spots on mesoscutellum, axillae, disc and anterolateral spot on metanotum, metapleuron, broad lateral U-shaped mark on propodeum, anterior vertical mark and small mark and adjacent spot near mid coxa on mesopleura, margins of mesosternal lamellae, coxae (except bases), extreme apices of femora, dorsal line on fore femora, ventral margin of mid and hind femora, lunulate lateral line on posterior margin of T1-6 and S3- 5 usually or virtually contiguous medially; orange colouration as follows: legs (except coxae, apex of fore trochanter, extreme apices of femora, dorsal line on fore femora, ventral margin of mid and hind femora). Margin of tegulae semitransparent. Wing membranes hyaline with veins orange to dark brown. Setae white. Distribution. Coastal Northern Territory from Darwin to Borroloola. Remarks. This species is distinguished from Caetrathynnus galhimis by having less extensive yellow colouration such that most sclerites are predonimantly black. It is distinguished from C. wesselensis sp. nov. by having the apex of the parameres rounded and the base of the cuspis rounded in profile. It is distinguished from both species by having S8 more emarginate at the base of the apical spine, the gena with punctation more rugose, and the pronotum with punctation more distinct on the anterior and posterior margins. The yellow colouration may be expanded on the tergites to enclose a posterolateral spot, or reduced on the mesoscutum and mesoscutellum such that the medial spots are elongate marks. Black colouration on the trochanters may be replaced by orange. Etymology, This species is named after Steven Gregg of the Museum and Art Gallery of the Northern Territory who collected many of the type series, and who has a great love of natural history. Caetrathynnus wesselensis sp. nov. (Fig. 3) Type material. HOLOTYPE cT - ANIC : lUOl’S, 136°45’E, Rimbija Is., Wessel Islands, Northern Territory, 3-14 February 1977, T.A. Weir. Description of male. Body length 7 mm; fore wing 6 mm; hind wing 4 mm. Clypeus rugosely and shallowly punctate, apical margin width to maximum width 1:1.8. POL:OOL 1:0.9. Frons closely and shallowly punctate. Vertex obscurely punctate. Gena shallowly and closely punctate. Pronotum sparsely and shallowly punctate, punctures more distinct posterolaterally and ventrally. Mesoscutum and mesoscutellum sparsely punctate medially, closely punctate laterally. Metanotum punctate. Propodeum closely and shallowly punctate, punctures deeper laterally, impunctate anteriorly. Mesopleura closely to finely punctate. Fore coxae slightly convex. T1 width to length 1:1.2. T2 weakly depressed anteriorly. Tergites and sternites closely and very shallowly punctate such that most tergites and anterior sternites almo.st appear impunctate, punctures coarser on posterior tergites. SI not medially raised. S2 depressed anteriorly. S8 slightly emarginate at base of spine. Genitalia as in Figure 3: basiparameres subtriangular (viewed dorsally) with each basiparamere rounded apically; parameres relatively short, distinctly truncate apically, weakly dorsally lobed (in profile) near base of aedeagus; cuspis subparallel basally, apically triangular, apex ending slightly before apex of digitus and not outwardly curved, base sinuate and strongly visible in profile. Colour. Black; yellow colouration as follows: mandibles (except apex), clypeus (except obscure medial anchor-shaped mark), large longitudinal mark (except very narrow sagittal line) above antennal insertion, inner and outer orbits of eyes almost continuous dorsally, margins of pronotum (anterior margin narrowly discontinuous medially, and expanded and confluent ventrally), tegulae, mesoscutum with spot on disc and horizontal mark above fore wing, central and antero¬ lateral spots on mesoscutellum. axillae, disc and anterolateral spot on metanotum, spot on metapleuron, lateral U-shaped mark on propodeum, anterior vertical mark and small mark and adjacent spot near mid coxa on mesopleura, margins of mesosternal lamellae, apex of fore coxae, mark on inner and outer margins of mid and hind coxae apically, ventral surfaces and dorsal mark apical mark on femora, lunulate lateral line on posterior margin of T1-6 and S2-5 contiguous medially on tergites; orange colouration as follows: legs (except coxae, apex of fore trochanters, mid and hind trochanters, ventral surfaces and dorsal mark apical mark on femora). Margin of tegulae semitransparent. Wing membranes hyaline with veins orange to brown. Setae white. Distribution. Only known from a single specimen from Rimbija I., Wessel Islands off the coast of Arnhem Land, Northern Territory. Remarks. This species is distinguished from other species of Caetrathynnus by the apically truncate parameres and the straight lateral margins of the basiparameres. It also differs from C. greggi which has the base of the cuspis rounded rather than sinusoidal in profile and from C. galhimis which has more extensive 112 New thynnine wasp genera yellow colouration such that the sclerites are more yel¬ low than black including a yellow spot on the prosterna and the yellow U-shaped lateral marks on the propodeum fused medially. This species is also distinguished by the punctures on the pronotum being most distinct postero- laterally and ventrally, and the fore coxae which are slightly convex. This species is similar in colouration to C. greggi. Etymology. The species name is derived from the type locality. Nitidothynnus gen. nov. Type species. Here designated Nitidothynnus purdiei sp. nov. Generic diagnosis. Male. Head, mesosoma and metasoma (especially dorsally) strongly polished and mostly obscurely punctate (such that punctures are sparse, and very small and shallow giving the appearance of the integument being impunctate especially on the vertex and pronotum). Clypeus convex basomedially, not carinate, narrowly truncate. Antennal prominence double U-shaped, medially sulcate, above plane of clypeus. Antennae long, longer than combined length of head and mesosoma, flagellar (except basal) segments four times longer than wide, apical six segments weakly arcuate. Maxillary palps with apical 3 segments each longer than basal 3 segments, fourth segment longest. Pronotum with lateral margins strongly convergent anteriorly, anterior margin raised but not carinate, anterior truncation transversely carinate ventrally and produced laterally such that the pronotum is produced anterolaterally. Mesopleura with longitudinal median groove ventrally. Fore trochanters long and strongly narrowed over most of length from base. Metasoma elongate-fusiform, sclerites weakly sclerotized (except SI and S8) and at most slightly constricted (although more so basally on S3-5 and especially S2 and T2). T7 convex, becoming slightly produced and membranous apically, not carinate. S8 with lateral margins curved, trispinose with medial spine longest. Genitalia with basal ring of medium length, subcylindrical and (in profile) slightly curved; basiparameres subovate (viewed dorsally) and emar- ginate medially, apices of individual basiparameres produced and often acute; parameres long and narrow especially apically, curved and convergent apically. All species are extensively marked with yellow on a reddish orange background although most of the reddish orange is replaced with black on the meso- and metathorax in some specimens. Female. Mandible unidentate. Clypeus narrowly truncate to slightly emarginate, medially raised. Frons obscurely sagittally sulcate. Pronotum wider than long, anterior margin clearly defined, longer than posterior margin, anterolateral angles spinose, lateral margins carinate on posterior half ending in a small tooth anteriorly; disc slightly medially raised, raised area wider posteriorly. Mesoscutellum narrow. Mesopleura with slight dorsal surface. Propodeum projecting well above level of mesoscutellum, oblique posteriorly becoming flattened dorsally, apex directed anteriorly and narrowed. Fore femur concave on apical half of ventral surface. T1 vertically truncate and setose anteriorly; dorsal surface with carinate apical margin and curved impressed preapical line. T2 impunctate with three to five transverse carinae. T3-5 with impressed curved preapical line. Pygidium (T6) curved in profile, posterior surface long and narrow with carinate lateral margins, ventral margins narrowly rounded, slightly emarginate laterally. Sternites sparsely and shallowly punctate. Remarks. The genus Nitidothynnus is most readily distinguished from other genera of the generic cluster by the microsculpture of the head, mesosoma and metasoma which are distinctively polished and obscurely punctate (particularly diagnostic on the vertex and pronotum); the anterior truncated surface of the pronotum transversely carinate ventrally and produced laterally and visible laterally as a short truncated process; the antennae relatively long; and the fore trochanters long and strongly narrowed over most of their length from the base. A convex, non-carinate clypeus that is below the plane of the antennal prominence, and long apical maxillary palp segments are similar to these states in Neozeleboria. However, Neozeleboria has S8 subparallel to sub- triangular with the apex ranging from rounded to slightly emarginate but with only a single apical spine rather than with lateral margins curved with the apex trispinose. Antennae are characteristically long in Nitidothynnus with antennae longer than the combined length of the head and mesosoma, and most flagellar segments four times longer than wide. In most other genera of the generic cluster (except some species of Tachynomyia and closely related genera) they are shorter but distinctly longer than wide. This compares to the short antennae found in Procerothynnus in which the antennal segments are only slightly longer than wide, appearing virtually as long as wide. The three species placed in this genus are very similar in appearance and are also best separated by the male genitalia, although Nitidothynnus spinulus is distinguished by the presence of a preapical spine on the fore and hind trochanters. Only the females of A. purdiei and N. spinulus are known, and these are distinguished by the shape of the head and T6. Both species are widespread in the Top End of the Northern Territory, and have overlapping distri¬ butions, while N. ebeneus is only known from the holotype from north-eastern Queensland. Nitidothynnus was referred to as genus “B” by Brown (1998a) in a key to closely related genera. Etymology. The generic name is masculine and is derived from the Latin word nitidus (= shining), and is a reference to the shining head, mesosoma and metasoma of the male. G.R. Brown Figs 7-17. Nitidothynnus spp., males: 7, N. spinulus sp. nov., male carrying female (female except eye and antenna stippled); 8, N. purdiei sp. nov., fore trochanter; 9, N. spinulus sp. nov., fore trochanter; 10, N. purdiei sp. nov., cuspis, ventral; 11, N. spinulus sp. nov., cuspis, ventral; 12, N. ebeneus sp. nov., cuspis, ventral; 13, N. spinulus sp. nov., S8; 14, N. spinulus .sp. nov., prothorax lateral; 15, N. spinulus sp. nov., genitalia, lateral and dorsal; 16, N. purdiei sp. nov.. genitalia, lateral and dorsal; 17, N. ebeneus sp. nov., genitalia, lateral and dorsal. Scale lines = 0.25 mm. 114 New thynnine wasp genera Key to species of Nitidothynnus. (The female of N. eteneus is unknown) 1 . a Male (winged). 2 b Female (wingless). 4 2 . a Fore and hind trochanters with preapical spine on ventral margin (most conspicuous if viewed post¬ eriorly) (Fig. 9). N. spiniilus sp. nov. b Trochanters without preapical spine on ventral margin (Fig. 8 ). 3 3 . a Genitalia (viewed dorsally) with each basi- paramere with outer margin curved and ending in an obtuse point (Fig. 16) and cu.spis (Fig. 10) with ventral surface mostly horizontal becoming abruptly and strongly concave apically (excluding lamellate projection). N. purdiei sp. nov. b Genitalia (viewed dorsally) with each basi- paramere with outer margin sinusoidal and ending in an acute point (Fig. 17) and cuspis (Fig. 12) with ventral surface almost flat and sloping upwards towards apex, not abruptly truncated or concave apically (excluding lamellate projection). . N. ebeneus sp. nov. 4. a Head with posterolateral angles emarginate (Fig. 20); T 6 subparallel dorsally (Fig. 23). . N. spinulus sp. nov. b Head with posterolateral angles rounded (Fig. 21); T 6 broadened ventroapically(Fig. 22). . N. purdiei sp. nov. Nitidothynnus purdiei sp. nov, (Figs 8 , 10, 16,21-22) Type material. HOLOTYPEcT - NTM (1912): near Oenpelli reservoir, campsite, 12°23’S, 133°05’E, 27 November 1997, G.R. Brown and J.E. Purdie. PARA- TYPES - Northern Territory: NTM (1913-1916) (4cf), same data as holotype; NTM (1917) (I cf), below Oenpelli reservoir, 12°24’S, 133°05’E, 27 November 1996, G.R. Brown and J.E. Purdie; NTM (1918) (I cf), near Oenpelli reservoir, by sweeping, 12°22’S, I33°04’E, 27 No¬ vember 1996, G.R. Brown and J.E. Purdie; BMNH (E2000-27), QM (57840), NTM (1919-1920) (4cf), near OenpeWi, Aliosyncarpia forest, 12°23’S, 133°OrE, 28 November 1996, G.R. Brown and J.E. Purdie; NTM (1921) (Icf), near Oenpelli, Allosyncarpia forest track, 12°23’S, 133°02’E, 30 November 1996, G.R. Brown and J.E. Purdie; NTM (1922) (Icf), Oenpelli reservoir, I2°24’S, 133°05’E, 26 November 1997, G.R. Brown and J.E. Purdie; NTM (1923-1924) (2cr), Daly River rd, Adelaide R. crossing, I3°29’S, 13r06’E, 24 October 1996, G.R. Brown; NTM (1925-1926) {Id), Blackfellow Ck, Daly River rd, I3°32’S, 130°49’E, 19 November 1997, G.R. Brown; NTM (1927) (Icf), Groote Eyiandt, in pit fall trap in open forest, 17-23 June 1982, J. Majer; NTM (1928) (Icf), near Drovers Rest, Litchfield NP, 25 November 1993, G.R. Brown; NTM (1929) (Icf), Maningrida, rubbish tip rd, 12°06’S, I34°13’E, 13 June 1996, G.R. Brown; NTM (1930) (IcT), Mataranka, by sweeping, 1 January 1996, G.R. Brown; NTM (1931) (Icf), 18°37’S, 137°59’E, Border Waterhole near Musselbrook Reserve, 10-17 April 1995, G.R. Brown; NTM (1932) (Icf), UDP Falls, 18-19 July 1980, M.B. Malipatil. NTM (11166) (I 9 ) Virginia near Darwin, 12°33’S, 13r02’E, 28 October 1996, S.M. Gregg (mounted with male paratype of N. spinulus). Queensland: NTM 1933-1937 (5cf), Musselbrook Reserve near Musselbrook Ck, 18°33’S, 138°irE, 18 April 1995, G.R. Brown; ANIC (Icf), Holts Ck, 18°33’S, 138.11’E, 15 May 1995, at light, l.D. Naumann. Other material. Northern Territory: NTM (2cf), near Oenpelli re.servoir, by sweeping, 12°22’S, 133°04’E, 27 November 1996, G.R. Brown and J.E. Purdie; NTM (9cf), near Oenpelli reservoir, Allosyncarpia forest, 12°23’S, I33°02’E, 28 November 1996, 30 November 1996, G.R. Brown and J.E. Purdie; NTM (Icf), Oenpelli area, November 1996, G.R. Brown and J.E. Purdie. Queensland: NTM (4cf), Musselbrook Reserve near Musselbrook Ck, 18°33’S, HSMl’E, 18 April 1995, G.R. Brown. Description of male. Body length 4.5-8 mm; fore wing 4-7 mm; hind wing 2.5-4 mm. Clypeus shallowly and closely punctate, apical margin width to maximum width 1:4.3. POL:OOL 1:1.6. Frons, vertex, gena and pronotum obscurely sparsely punctate. Lateral margins of pronotum sightly curved and convergent anteriorly, anterior truncation with ventral transverse carina almost continuous medially. Mesoscutum closely punctate anteriorly and laterally, punctures more conspicuous than those on head and pronotum. Mesoscutellum and metanotum obscurely sparsely punctate. Propodeum transversely striate, longer than wide, lateral margins parallel-sided anteriorly, dorsal margin in profile evenly rounded down to metasoma. Mesopleura obscurely sparsely punctate. Fore coxae, long, conical, ventral surface flat. Fore trochanter (Fig. 8 ) not spined. Metasoma almost impunctate, T1 closely punctate, T 6 - 7 and S 6-8 with conspicuous setiferous punctures and setae long. T1 width to length 1:1.1. SI medially raised, rounded in cross-section, T2 weakly and S2 strongly depressed anteriorly, segments 1-5 Battened and wider than high. Genitalia as in Figure 16: apex of each basiparamere rounded dorsoapically; parameres narrow and narrowly subtriangular; cuspis (Fig. 10) lamellate apically, ventral surface mostly horizontal becoming abruptly and strongly concave apically (excluding lamellate projection). Colour. Reddish orange; yellow colouration as follows: mandibles (except apex), maxillary and labial palps, clypeus (except medial anchor-shaped mark and dorsolateral spots testaceous), antennal prominence, inner orbits of eyes extending to behind vertex, outer orbits sinusoidally extending across vertex and confluent with inner orbits, margins of pronotum (broadly so G.R. Brown Figs 18-23. Nitidothynnus spp., females: 18, N. spinulus sp. nov., mesothorax; 19, N. spinulus sp. nov., habitus; 20, N. spinulus sp. nov., head; 21, N. piirdiei sp. nov., head; 22, N. purdiei sp. nov., T6; 23, N. spinulus sp. nov., T6. Scale lines = 0.25 mm. ventrally), tegulae, mesoscutum with curved mark above fore wing and central subrectangular marks, me- soscutellum with anterolateral spot and central broad trident-shaped mark; disc and anterolateral line on metanotum, most of propodeum, mesopleuron (except curved branched medial mark and ventral surface), metapleuron (except extreme margins), coxae (except extreme bases), margins of mesosternal lamellae, apical margins of trochanters and inner surface of fore trochanter, dorsal and ventral margins of femora irregularly expanded on fore leg, outer surface of tibiae, lateral irregular spots on Tl-6 and Sl-2; black colouration as follows: ocellar triangle, posterior and posteroventral surfaces of head, anterior truncation and neck of pronotum, mesoscutum (except curved mark above fore wing and central subrectangular marks), mesoscutellum laterally and extreme anterior margin, metanotum (except disc and anterolateral line), extreme anterior margin of and pair of posteromedial spots on propodeum, prosterna, mesopleuron ventrally, extreme margins of metapleuron, SI, S8 and base of Tl; tarsi brown especially apically. Wing membranes hyaline, veins orangish brown. Description of female. Body length 3 mm. Head (Fig. 21 ) rounded, as wide as long, sparsely punctate with large semicircular closely punctate area occupying most of vertex, posterolateral angles strongly emarginate. T2 with 3 transverse carinae, apical 2 strongest, basal 1 may be obscured by Tl. Pygidium (T6) with lateral margins slightly diverging ventrally, and slightly medially and triangularly raised (Fig. 22). Colour. Pale yellowish brown; head (except appendages) greyer. Distribution. Northern Territory from Daly River south to Mataranka and east to the Queensland border. Remarks. The male of this species is distinguished from other species of Nitidothynnus by the structure of the genitalia which (when viewed dorsally) has each basiparamere with the outer margin curved and ending in an obtuse point (Fig. 16) and cuspis (Fig. 10) with the ventral surface mostly horizontal becoming abruptly and strongly concave apically. The female is distinguished by the structure of the head which has the posterolateral angles rounded (Fig. 21) and T6 broadened ventro- apically (Fig. 22). The only other known female of a species in the genus Nitidothynnus has the posterolateral angles of the head emarginate (Fig. 20) and T6 subparallel dorsally (Fig. 23). The yellow markings may be expanded on the posterior and anterior margins of the pronotum so that 116 New thynnine wasp genera they are confluent each side of the sagittal line, or confluent medially on SI. They may also be narrowly discontinuous medially on the anterior margin of the pronotum, reduced to a central spot on the pronotum or split into 2 spots on the mesopleuron and replaced with reddish orange, or reduced or absent on SI-2 and Tl. Much of the reddish orange on the mesoscutum and propodeum may be replaced with black. The apex of the metasoma may be slightly darker. The Musselbrook Reserve specimens tend to have the mesosoma darker than the majority of the other specimens, but fit within the range of colour variation of these specimens. There are no noticeable differences in the genitalia to Justify a separate species for the Musselbrook material. The specimen from Groote Eylandt is labelled as coming from a pit fall trap. This is unusual, but it is conceivable that a wingless female could have fallen into such a trap, and, once there, attracted the male by pheromones. I have never seen thynnines of either sex in pit fall traps, although the wingless females of the closely related family Mutillidae (“velvet ants”) are frequently caught in such traps. Etymology. This species is named after John Purdie who collected many of the type series, and who is a keen naturalist and a close friend. Mating biology. As indicated in the Introduction, mixed species pairs may be encountered occasionally. This is indicated in museum collections when a series consisting of a single species of male includes females of two different species that have both been collected in copula with this species of male. In the absence of morphological structures that may, for example, be adaptations to prolonged coupling, and therefore, associate pairs, sexes may be associated on the assumption that the majority of specimens couple correctly. This is the case with Nitidothynniis purdiei and N. spimilus. Females of Nitidothynniis purdiei and N. spinulus are associated with males on the basis of coincident collecting data for four males and one female of N. spinulus from Virginia, and two males and two females N. spinulus from Berrimah. The Virginia specimens include a pair collected in copula while the Berrimah females include a pair grasping each other and which had probably been in copula prior to being killed. The Berrimah pair is interesting in the way the male and female fit closely together. The male is clasping the female with its legs such that: the hind trochantcral spine fits into the groove before the apical carina on T2 of the female; the fore trochanteral spine is holding the fore femur of the female away from her body; and the male head is deflexed so that the female head is held closely between the mouthparts and the fore coxae of the male. The female is clasping the male with her mouthparts and has the fore femur overlapping the fore trochanter and femur of the male, and has the protruding propodeum slotted into the longitudinal ventral groove along suture of the mesopleura of the male. Contrary to this association of sexes is another female in which the head is rounded posterolaterally and T6 is broadened ventrally. This female was collected in copula at Virginia with a male of N. spinulus. The female is believed to be N. purdiei as it is the only other species which overlaps in distribution with N. spinulus (although there are no other coincident collecting records). As pairs couple for prolonged periods, there is the possibility that the shape of structures in the two sexes that are juxtaposed as a result of this coupling, may be correlated. The major differences in the shape of the female head, and the juxtaposition of the female head and male fore coxae in Figure 7 suggests that there may be a correlation between these structures that permit the association of sexes on morphological grounds. However, no differences in the structure of the male fore coxae could be found that would explain differences in the female head shape. Nitidothynniis ebeneus sp. nov. (Figs 12, 17) Type material. HOLOTYPF cT - QM (T57841): near Granite Gorge, 17°0rS. 145°20’F, 21 May 1989, 12 km SW Mareeba, Queensland, G. and A. Daniels. Description of male. Body length 7 mm; fore wing 5 mm; hind wing 4 mm. Clypeus with apical margin width to maximum width 1:5.0. POL:OOL 1:1.5. Anterior truncation of pronotum with ventral transverse carina broadly discontinuous medially. Fore trochanter not spined. Tl width to length 1:1.2. Genitalia as in Figure 17: apex of each basiparamere narrow and acute dorsoapically; parameres subtriangular, strongly narrowed beyond level of apex of cuspis; cuspis (Fig. 12 ) lamellate apically, ventral surface almost flat and sloping upwards towards apex, not abruptly truncated or concave apically (excluding lamellate projection). Otherwise as N. purdiei. Colour. Black; yellow colouration as follows: mandibles (except apex), maxillary and labial palps, clypeus (except testaceous medial anchor-shaped), antennal prominence, inner orbits of eyes extending to behind vertex, outer orbits of eyes sinusoidally extending across vertex and confluent with inner orbits, anterior and posterior margins of pronotum, tegulae, mesoscutum with curved mark above fore wing and central sub- rectangular marks, mesoscutellum with anterolateral spot and central broad inverted T-shaped mark, disc and anterolateral line on metanotum, propodeum with medial line strongly broadened anteriorly and lateral line strongly broadened posteriorly, mesopleuron (except curved branched medial mark and ventral surface), spot on metapleuron, fore and mid coxae (except extreme bases), outer margin of hind coxa, margins of mesosternal G.R. Brown lamellae, apical margins of trochanters, dorsal and ventral margins of femora, outer surface of tibiae, lateral irregular spots on Tl-5; reddish orange colouration as follows: gena and vertex dorsally, legs (except coxa and yellow marks on trochanters, femora, tibiae and tarsi) and metasoma (except yellow spots on T1 -5, T1, S1, base of S2 and S 8 ); S 8 dark brown. Wing membranes hyaline, veins pale orange. Distribution. Known only from near Granite Gorge, 12 km SW of Mareeba, North Queensland. Remarks. The male of this species is readily distinguishable from other species of the genus by the absence of reddish orange colouration on the mesosoma (excluding the legs) although this is based only on a single specimen. It is also distinguished by the digiti which are not visible ventrally and the cuspis (Fig. 12) which has the ventral surface almost flat and sloping upwards towards apex, not abruptly truncated or concave apically. The female is unknown. Etymology. The species name is derived from the Latin word ebeneus (= black) and refers to the relatively dark colour of the mesosoma. Nitidothynnus spinulus sp. nov. (Figs 7, 9, 1 1, 13-1.5, 18-20,23) Type material. HOLOTYPE cf - NTM (1954): Virginia near Darwin, 12°33’S, 131°02’E, Northern Territory, 28 October 1996 S.M. Gregg. PARATYPES - Northern Territory. NTM (1955-1958) (3d', 1 9 ), same data as holotype (Icf, 1 9 1955-1956), or dated 12 July 1997 (1 cf 1957) or 16 November 1997 (1 cT 1958); NTM, (1959-1962) (2d', 29 ), Berrimah near Darwin, 1-15 November 1996 (1 d" 1959), 16-30 November 1996 (1 cf, 29 1960-1962), A. Salvarani; NTM (1963-1965), OTTAWA, WAM (26556) (5d'), Darwin, eucalypt woodland. Malaise trap, (site 93-86), 1-25 December 1993, S. and J. Peck; NTM (1966-1967), QM (T57842) (3d'), Anbangbang Billabong, Kakadu NP, 12°52’S, 132°48’S, 10 June 1996(1 NTM 1966), 17 January 1998 (1 NTM 1967), 1 QM), G.R. Brown; ANIC , NTM (1968- 1973), OTTAWA ( 8 d'), Kapalga Research Stn, Kakadu NP, eucalypt woodland, Malaise trap, (site 93-117), 11- 25 December 1993, S. and J. Peck; NTM (1974-1975), OTTAWA (3d'), Kapalga Research Stn, Kakadu NP, North Point Rainforest, Malaise trap, (93-134), 24 December 1993-7 January 1994, S. and J. Peck; NTM (1976) (Icf), Keep River National Park, Gurrandaing, 15°53’S, 129°03’E, 30 April 1996, G.R. Brown; NTM (1977) (Icf), Keep River National Park, Gurrandaing, 15°53'S, 129°03’E, 29 April 1996, G.R. Brown. Other material. NTM (Icf), Keep River National Park, Gurrandaing, 15°53’S, 129°03’E, 29 April 1996, G.R. Brown. Description of male. Body length 5-7 mm; fore wing 4-6 mm; hind wing 3-4 mm. Clypeus with apical margin width to maximum width 1:4.8. Anterior truncation of pronotum with ventral transverse carina broadly discontinuous medially. Fore and hind trochanters with ventral preapical spine. Genitalia as in Figure 15: apex of each basiparamere acute dorsoapically; parameres narrow and narrowly subtriangular; cuspis (Fig. 11) lamellate apically. ventral surface oblique. Otherwise as N. piirdiei. Colour. Similar to Nitidothynnus purdiei except that at most S2 has a pair of pale spots, the apex of S1 is reddish orange, and the basal three tarsomeres are yellowish ventrally and brown dorsally with the apical two tarsomeres uniformly dark brown. Description of female. Body length 4 mm. Head (Fig. 20) rounded, slightly wider than long, sparsely punctate with semicircular closely punctate area on vertex, posterolateral angles strongly emarginate. T2 with 5 transverse carinae, apical 3 strongest, basal 2 may be obscured by T1. T3-5 with impressed curved preapical line preceded by a curved line of punctures. Pygidium (T 6 ) with lateral margins parallel (Fig. 23). Colour, Orangish-brown; coxae, trochanters, femora and mesosoma (except prothorax) slightly darker; tibiae and tarsi slightly paler. Distribution. Top End of Northern Territory between Darwin and Kakadu National Park. Remarks. This species is distinguished from all other species of Nitidothynnus by the presence of a preapical spine (most reliably viewed on the posterior surface) on the ventral margin of the fore and hind trochanters of the male. The female is distinguished by the structure of the head which has the posterolateral angles emarginate (Fig. 20) and T 6 which is subparallel dorsally (Fig. 23). The only other known female in the genus, N. purdiei, has the posterolateral angles of the head rounded (Fig. 21) and T 6 broadened ventroapically (Fig. 22). Etymology. The species name is derived from the Latin word for a spine, and refers to the preapical spine on the fore and hind trochanters. Procerothynnus gen. nov. Type species. Here designated Procerothynnus centralianus sp. nov. Generic diagnosis. Male. Head, mesosoma and metasoma polished. Clypeus closely and finely punctate, sagittally carinate, narrowly produced, convex medially, convexity extending medially almost to apical margin of clypeus. Antennal prominence V-shaped, sagittally sulcate, not carinate, not strongly raised, at level of clypeus. Antenna very short, not reaching back to propodeum, flagellar segments only slightly longer than wide, apical six segments slightly arcuate. Maxillary palp segments 2-6 subequal, segments 4 and 6 slightly longer. Pronotum with lateral margins curved and convergent anteriorly, anterior margin sharply raised and subcarinate. Metasoma fusiform, segments 3-4 widest, segments wider than high. T2 depressed anteriorly, T3 at most 118 New thynnine wasp genera slightly depressed anteriorly. T7 convex becoming sligh¬ tly produced and membranous apically, not carinate. SI broadly medially raised. S8 rounded apically, without spines. Genitalia laterally compressed; basal ring short and predominantly membranous with narrow, sclerotized ring basally; basiparameres fusiform (viewed dorsally) with apical margin sinusoidal, strongly rounded (viewed laterally); paramere subtriangular to subquadrate; aedeagus long and narrow. Remarks. Procerothynnus is distinguished from other genera of the generic cluster (and all other Aus¬ tralian Thynninae) by the combination of a medially convex clypeus that is sagittally carinate; very short antennae which do not reach back to the propodeum; S8 apically rounded and without spines; and the genitalia laterally compressed. A convex clypeus with a sagittal Carina and S8 that is not apically spinose suggest a relationship with Zelehoria, Psammothynnus and Chilothynmis. However, the former two have the clypeus strongly convex and broadly produced and apically truncate. Psammothynnus also has S8 apically e- marginate. Chilothynmis is the closest genus, but differs by having lateral spines on S8 (Brown 1997a), and together with all other Thynninae lack very short antennae and laterally compressed genitalia as found in Procerothynnus. The three species placed in this genus are similar in appearance and are most reliably identified from the male genitalia. However, head colour appears reliable in distinguishing P. arnhemicus sp. nov. from other species of Procerothynnus. All are known from relatively few specimens from relatively few locations; P. centralianus sp. nov. from four locations in the East MacDonnell Ranges in central Australia; P. arnhemicus sp. nov. from near Oenpelli in western Arnhemland and P. carpen- tarianiis sp. nov. from Musselbrook Reserve near the Northern Territory border in north-western Queensland. Like Nitidothynnus, nothing is know about the biology of Procerothynnus. All specimens were collected from habitats with loamy or sandy soil and near water courses (although all except the stream at Oenpelli, but including the stream at Musselbrook Reserve, were dry at the time at which specimens were collected). Etymology. The generic name is masculine and is derived from the Latin word procerus which means tall and slender, and is a reference to the shape of the male genitalia which are relatively high (in profile) and slender (viewed dorsally). Key to males of Procerothynnus. (Females are unknown.) 1. a Yellow marks above each antennal insertion at least partially confluent medially; mesoscutum above fore wing with at most a small yellow spot; genitalia with parameres broad and apex of basiparameres weakly sinusoidal (Figs 27, 29). . P. arnhemicus sp. nov. 1 b Antennal prominence with separate and discrete spot above each antennal insertion; mesoscutum with a longitudinal mark above fore wing; genitalia with parameres subtriangular and apex of basiparameres emarginate (Figs 25-7, 28).2 2. a Punctures on propodeum relatively deep, discrete and distinct; genitalia as in Figure 25; apical margin of basiparameres weakly sinusoidal, slightly produced medially and apicolaterally. . P. centralianus sp. nov. b Punctures on propodeum shallow, often not discrete; genitalia as in Figures 26; apical margin of basiparameres (Fig. 28) strongly emarginate, produced medially and strongly produced apico¬ laterally . P. carpentarianus sp. nov. Procerothynnus centralianus sp. nov. (Figs 25, 30) Type material. HOLOTYPE cf - NTM (1979): near Ghost Gum, 23°32’S, 134°23’E, Trephina Gorge Nature Park, Northern Territory, 26 February 1997, G.R. Brown. PARATYPES - Northern Territory. NTM (1980) (Icf), Ross River Highway 23°38’S, 134°18’E, 6.2 km W of Trephina Gorge Nature Park, 1 March 1997, G.R. Brown; NTM (1981) (IcT), road to Ruby Gap Nature Park, 23°29’S, 134°54’E, 2 March 1997, G.R. Brown; NTM (1982) (Icf), Corroboree Rock Conservation Reserve, 23MrS, 134°13’E, 1 March 1997, G.R. Brown. Description of male. Body length 6.5-7 mm; fore wing 5-6 mm; hind wing 4-5 mm. Frons shallowly punctate to shallowly closely punctate, punctures becoming aligned longitudinally on antennal pro¬ minence. Clypeus with apical margin width to maximum width 1:4.2. POL:OOL 1:0.9. Vertex and gena shallowly rugosely punctate. Pronotum closely punctate. Me¬ soscutum punctate to sparsely punctate, closely punctate near anterior margin. Mesoscutellum and metanotum sparsely punctate. Propodeum closely to rugosely punctate, punctures discrete. Mesopleuron closely to rugosely punctate becoming, punctate ventrally. Fore coxae with ventral surface almost flat, subparallel basally, subtriangular apically. T1 width to length 1:1.0. Tergites almost impunctate, punctures shallow, transverse and slit-like on most segments, deeper and rounder on T6-7. T3 not depressed anteriorly. Sternites almost impunctate but with punctures aligned in transverse straight or curved lines, S8 closely punctate. Genitalia as in Figure 25; apical margin of basiparameres weakly sinusoidal, slightly produced medially and apicolaterally. Colour. Black; yellow colouration as follows: clypeus (except medial anchor-shaped mark), mandibles (except apex), large spot above each antennal insertion, inner orbits continuous across vertex, outer orbits, margins of pronotum broadly confluent medially (except small central spot), tegulae, mesoscutum with curved mark 19 G.R, Brown Figs 24-30. Procerothynnus spp., males; 24, P. arnhemicus sp. nov., habitus; 25, P. centralianus sp. nov., genitalia, lateral and dorsal; 26, P carpentarkmus sp. nov., genitalia, lateral; 27, P. arnhemicus sp. nov., genitalia, lateral; 28, P. carpentarianus sp. nov., apical margin of basiparameres; 29, P. arnhemicus sp. nov., apical margin of basiparameres; 30, P. centralianus sp. nov., S8. Scale lines = 0.25 mm. 120 New thynnine wasp genera above fore wing and large medial subrectangular mark, mesoscutellum with large medial mark and smaller anterolateral spot, metanotum with disc, posterior margin narrowly and mark on anterior margin, metapleuron with large medial spot, broad sinusoidal transverse band on propodeum, three dorsal spots on mesopleuron, anterior spot largest, posterior spot near mid coxa smallest, coxae (except basally), apices of femora extending along ventral margin to base and along dorsal margin apically especially on fore leg, line on outer surface of tibiae, margins of mesosternal lamellae, lateral spot on Tl-6 and S2-5 (smaller or inconspicuous on sternites) extending medially as a curved line near posterior margin of sclerites, posteromedial spot on SI; orange colouration as follows: apex of mandibles, lateral parts of anchor¬ shaped mark on clypeus, legs (except coxae and yellow marks); much of T1 apically, T2-3 and S2-3, reddish orange. Wing membranes hyaline, veins pale orange to brown. Distribution. Eastern MacDonnell Ranges, Northern Territory. Remarks. This species is distinguished from other species of Procerothynniis by the presence of relatively shallow punctures on the frons, a weakly developed clypeal carina, and the genitalia (Fig. 25) which have the apical margin of basiparameres weakly sinusoidal and slightly produced medially and apicolaterally. Yellow colouration may be reduced such that marks and spots are smaller on some sclerites. In particular this colouration may be limited on the clypeus to the apical margin laterally, on the pronotum to the margins only, on the propodeum as lateral spots, on the tergites and sternites as a curved lateral line without a large lateral spot, and reduced on the legs especially the femora. The reddish orange colouration on the metasoma may be replaced with black as may some of the orange colouration on the legs. The Ruby Gap specimen is much darker than the other specimens but there is no noticeable difference in the genitalia. Etymology. This specific name refers to the central Australian distribution of this species. It is intended to be construed as adjectival. Procerothynniis arnhemiciis sp. nov. (Figs 24. 27. 29) Type material. HOLOTYPE cT - NTM (1983): campsite near Oenpelli Reservoir, 12°23’S, 133°05’E, Oenpelli, Northern Territory, 27 November 1997, G.R. Brown and J.E. Purdie. PARATYPES - Northern Territory. ANIC , NTM (1984) {2cf), data as holotype; BMNH (E2000-27), NTM (1985-1986), WAM (26557) (3d'), Leaning Tree Lagoon, I2°43’S, I31''25’E, II December 1997, G.R. Brown and J.E. Purdie. Other material. NTM (Icf), Leaning Tree Lagoon, 12°43’S, 131 °25’E, 11 December 1997, G.R. Brown and J.E. Purdie. Description of male. Body length 6 mm; fore wing 4.5 mm; hind wing 3.5 mm. Frons closely to rugosely punctate. Clypeus with apical margin width to maximum width 1:4.3. POL:OOL 1:1.2. Vertex and gena shallowly rugosely punctate, punctures more discrete on gena. Pronotum closely punctate becoming impunctate ventrally. Mesoscutum punctate, punctures sparser medially. Mesoscutellum and metanotum sparsely punctate. Propodeum closely to rugosely punctate, punctures discrete. Mesopleuron closely to rugosely punctate becoming sparely punctate ventrally. Fore coxae with ventral surface convex, subtriangular. T1 width to length 1:1.0. Tergites almost impunctate, punctures shallow, transverse and slit-like on most segments, deeper and rounder on T7 although punctures sparser medially. T3 at most slightly depressed anteriorly. Sternites almost impunctate, closely punctate on S8. Genitalia as in Figure 27; apical margin of basiparameres (Fig. 29) weakly sinusoidal, not produced apically. Colour. As in P. centralianus except spots above antennal insertions partially confluent medially, mark above fore wing reduced to a small inconspicuous spot or absent, middle spot on mesopleuron small and posterior post absent, marks on most metasomal segments strongly curved, and metasoma without reddish orange colouration. Wing veins generally paler except subcosta. Distribution. Known only from the vicinity of Kakadu National Park, Northern Territory. Remark.s. This species is readily distinguished from other species of Procerothynniis by having separate yellow spots above each antennal insertion, at most a small yellow .spot on the mesoscutum immediately above the base of the fore wing, and the structure of the genitalia (Fig. 27), which have the apical margin of basiparameres (Fig. 29) weakly sinusoidal and not produced apically. Etymology. This specific name is derived from the type locality. It is intended to be construed as adjectival. Procerothynniis carpentarianiis sp. nov. (Figs 26, 28) Type material. HOLOTYPE cT - UQIC; Murrays Spring, I8°35’15”S, I38°04’28”E, 7km W of Mussel- brook Resource Center, Lawn Hill National Park, Queensland, 200 m, 10 May 1995, G. Daniels, M.A. Schneider. PARATYPES - Queensland’. UQIC (Icf), data as holotype dated 4 May 1995. Description of male. Body length 6 mm; fore wing 4.5 mm; hind wing 3.5 mm. Frons shallowly rugosely punctate. Clypeus with apical margin width to maximum width 1:4.0. POL:OOL 1:1.2. Vertex and gena shallowly punctate. Pronotum punctate becoming longitudinally rugulose ventrally. Mesoscutum punctate. Mesoscutellum and metanotum sparsely punctate. 121 G.R. Brown Propodeum closely to rugosely punctate, punctures shallow and not always discrete. Mesopleuron punctate. Fore coxae with ventral surface almost flat, subtriangular. T1 width to length 1:1.5. Tergites almost impunctate, punctures shallow, transverse and slit-like on most segments, deeper and rounder on T6-7 although punctures sparser medially. T3 not depressed anteriorly. Sternites almost impunctate, SI and S8 closely punctate. Genitalia as in Figure 26; apical margin of basiparameres (Fig. 28) strongly emarginate, produced medially and strongly produced apicolaterally. Colour. As in P. centralUinus except middle spot on mesopleuron small and posterior spot absent, yellow marks on sternites mostly absent, and metasomal segment 3 is darker with less reddish orange. Wing veins generally paler except subcosta. Distribution. Known only from two specimens from Musselbrook Reserve north-western Queensland on the border of the Northern Territory. Remarks. This species is easily distinguished from the other two species of the genus by shallower and less discrete punctures on propodeum, and the structure of the genitalia (Fig. 26), which have the apical margin of basiparameres strongly emarginate and produced medially and strongly produced apicolaterally (Fig. 28). Etymology. This specific name is derived from The Gulf of Carpentaria which is the broader Australian region from which the types were collected. It is intended to be construed as adjectival. ACKNOWLEDGMENTS I thank Steven Gregg and John Purdie for their interest in entomology in general and tiphiids in particular, the curators of the above li.sted institutions for the loan of material in their care, and the anonymous referees for comments on an earlier version of this paper. I also thank Dr Richard C. Willan for his comments and eye for detail. REFERENCES Brown, G.R. 1983. Pentazelehoria, a new genu.s of Australian Thynnini (Hymenoptera: Tiphiidae). Journal of the Australian Entomological Society 22: 61-64. Brown, G.R. 1987. Revision of the Australian genus Acantho- thynnus Turner (Hymenoptera: Tiphiidae). Journal of the Australian Entomological Society 26: 181-88. Brown, G.R. 1989a. Revision of the Australian genus Dora- tithynnus Turner (Hymenoptera: Tiphiidae). Journal of the Australian Entomological Society 28: 1-17. Brown, G.R. 1989b. The Australian genus Encopothynnus Turner (Hymenoptera: Tiphiidae). Journal of the Australian Entomological Society 28: 255-266. Brown, G.R. 1992. Bifidothynnus wubiniensis, a new genus and species of Australian Thynnini (Hymenoptera: Tiphiidae: Thynninae), Journal of the Australian Entomological Society 31: 215-217. Brown, G.R. 1993. A new species of Lestricothynnus with notes on miscoupling in Thynninae (Hymenoptea: Tiphiidae). Journal of the Australian Entomological Society 32: 197- 199. Brown, G.R. 1995a. Revision of the Australian wasp genus Macrothynnus Turner (Hymenoptera: Tiphiidae: Thy¬ nninae). Records of the Western Australian Museum 17: 267-275. Brown, G.R. 1995b. Tachyphron, a new genus of Australian Thynninae (Hymenoptera: Tiphiidae). Journal of the Australian Entomological Society 34: 241-246. Brown, G.R. 1997a. Chilothynnus, a new genus of Australian Thynninae (Hymenoptera: Tiphiidae) associated with orchids. The Beagle. Records of the Museums and Art Galleries of the Northern Territory 13: 61-71. Brown, G.R. 1997b. Arihrothynnus, a new genus of orchid- pollinating Thynninae (Hymenoptera: Tiphiidae). The Beagle, Records of the Museums and Art Galleries of the Northern Territory 13: 73-82. Brown. G.R. 1998a. Revision of the Neozelehoria cryptoides .species group of thynnine wa.sps (Hymenoptera: Tiphiidae): pollinators of native orchids. Australian Journal of Entomology yi\ 193-205. Brown, G.R. 1998b. The generic status of Catocheilus Gudrin and Hemithynnus Ashmead (Hymenoptera: Tiphiidae: Thynnini). General and Applied Entomology 28: 89-92. Burrell, R.W. 1935. Notes on the habits of certain Australian Thynnidae. Journal of the New York Entomological Society 43: 19-28. Given, B.B. 1954. A catalogue of the Thynninae (Tiphiidae, Hymenoptera) of Australia and adjacent areas. New Zealand Department of Scientific and Industrial Research Bulletin 109: 1-89. Naumann, l.D. 1991. Hymenoptera (Wasps, bees, ants, sawflies). Pp. 916-1000. In Naumann, l.D. (ed). The insects of Australia. A textbook for students and research workers. Melbourne University Press: Carlton. Salter, K.E.W. 1954. Studies on Australian Thynnidae. I. A check list of the Australian and Austro-Malayan Thynnidae. Proceedings of the Linnean Society of New South Wales 78(5-6): 276-315. Snodgrass, R.E. 1941. The male genitalia of Hymenoptera. Smithsonian Miscellaneous Collections 99: 1-86. Tillyard, R.J. 1926. Insects of Australia and New Zealand. Angus and Robertson: Sydney. Turner, R.E. 1910. Hymenoptera Earn. Thynnidae. Genera Insectorum 105: 1-62. Accepted 8 August 2000 122 The Beagle, Records of the Museums and Art Galleries of the Northern Territory, 2000 16: 123-126 A new bythitid genus and species, Acarobythites larsonae, from shallow rocky reefs off northern Australia (Pisces, Ophidiiformes, Bythitidae) YOSHIHIKO MACHIDA Department of Natural Environmental Science, Faculty of Science, Kochi University 2-5-1 Akebono-cho, Kochi 780-8520, JAPAN machida@cc.kochi-u.ac.jp ABSTRACT A new genus and species of the subfamily Bythitinae (Bythitidae), Acarobythites larsonae, is described from two specimens, 25.2 mm and 18.6 mm standard length, collected from pools on rocky reefs in the Northern Territory, Australia. This genus resembles Microhrotula Gosline, 1953 in the counts of pectoral and pelvic fin rays, branchiostegal rays and developed rakers on the anterior arch, and in having a normal pectoral peduncle, lower angle of preopercle without spines, sharp-pointed jaw teeth and vomerine and palatine teeth. However, it is clearly distinguishable from the latter in the following characters: head and body scalcless, relatively small eyes (diameter 17.8-18.0 times in head length), snout compressed, posterior end of maxillary rounded and lacking a ventrally-directed process near its postero- ventral corner, opercular spine flat and weak, 11-12 caudal fin rays and 42-43 total vertebrae. Keywords: Ophidiiformes, Bythitidae, Australia, new genus, Acarobythites. INTRODUCTION Cohen and Nielsen (1978) included 15 genera in the subfamily Bythitinae of the viviparous fish family Bythitidae. This subfamily is characterised by having dorsal and anal fins continuous with the caudal fin, and is separable from the free-tailed Brosmophycinae, the other member of the family (Cohen and Nielsen 1978). Since Cohen and Nielsen (1978), only Hastatohythites Machida, 1977 has been added to the Bythitinae (Cohen and Nielsen 1999). Recently, two small hythitid specimens were sent to the author for examination by Dr Helen K. Larson of the Museum and Art Gallery of the Northern Territory (NTM), Darwin, Australia. These specimens had dorsal and anal fins continuous with the caudal fin. Although they resembled the small-sized bythitine genus Microhrotula Gosline, 1953, known from Hawaii (Gosline 1953) and from Samoa and Vanuatu (Cohen and Wourms 1976), they were easily distinguishable from Microhrotula by the absence of scales on the head and body. The specimens are here described as a new genus and species. Measurements and enumeration follow Machida (1993). Standard length and head length are expressed as SL and HL, respectively. SYSTEMATICS Family Bythitidae Gill, 1861 Subfamily Bythitinae Gill, 1861 Acarobythites new genus 'Fype species. Acarobythites larsonae new species, by monotypy. Diagnosis. A genus of the subfamily Bythitinae (Cohen and Nielsen 1978: 42) with: head and body scaleless; anal fin origin at about mid-body; eye diameter 17.8-18.0 times in HL, snout compressed; posterior end of maxillary rounded, not sheathed by dermal cheek fold, without ventrally-directed process near postero-ventral corner; opercular spine flat and weak; no pore-bearing skin flap on upper angle of opercle; lower angle of preopercle smooth; pectoral peduncle normal, broader than long; no spine on cleithrum; all teeth sharp-pointed, some needle-like; teeth present in jaws and on vomer and palatines; developed gill rakers on anterior arch 3; pectoral fin rays 13-14; pelvic fin ray 1; caudal fin rays 11-12; branchiostegal rays 7; 12-13 precaudal vertebrae, with sharp-pointed neural spines; and 42-43 total vertebrae. Description. See species description below. Remarks. According to Cohen and Nielsen (1972, 1978), Cohen (1987), and Machida (1997), the following genera in the subfamily Bythitinae are known to have Y. Machida Table 1. Comparison between Acarobythites gen. nov. and Microbrotula Gosline, 1953. * = orbit diameter. Genus (Source) Acarobythites gen. nov. (this work) Microbrotula Gosline, 1953 (Gosline 1953; Cohen and Wourms 1976; Cohen and Nielsen 1978) Head squamation scales absent partly naked Body squamation scales absent complete Anal fin origin at about mid-body at about mid-body Eye diameter 17.8-18.0 in HL (6.4-15.4 in HL)* Snout compressed depressed Posterior end of maxillary rounded, not sheathed, without ventrally- directed process expanded, not sheathed, with ventrally- directed process Pectoral peduncle broader than long broader than long Opercular spine flat and weak sharp and needle-like Lower angle of preopercle smooth smooth Teeth sharp-pointed not all tiny and granular Palatine teeth present present Vomerine teeth present present Developed rakers on anterior arch 3 3-4 Pectoral fin rays 13-14 10-14 Pelvic fin ray 1 1 Caudal fin rays 11-12 4-6 Branchiostegal rays 7 7 Precaudal vertebrae 12-13 11-13 Total vertebrae 42-43 51-56 seven branchiostegal rays; Bythites (7-8 rays), Calamopteryx, Microbrotula, Saccogaster (7-9 rays), Stygnobrotulo and Thalassobathia. Although Cohen and Nielsen (1978) did not report the number of branchiostegal rays for Bellotia, this genus apparently differs from Acarobythites only in lacking pelvic fins. Among these genera having seven branchiostegal rays, two of them, Calamopteryx and Saccogaster, are well characterized by having an elongated pectoral peduncle (Cohen and Nielsen 1978). Stygnobrotula lacks palatine teeth, and Thalassobathia has two rays in each pelvic fin (Cohen and Nielsen 1978). Bythites has a scaleless head, completely scaled body, 24-30 pectoral fin rays, 0-1 developed gill rakers on the anterior arch, and the anal fin origin is positioned well behind the mid-body (Nielsen and Cohen 1973; Cohen and Nielsen 1978). Cohen and Wourms (1976) gave a revised diagnosis for Microbrotula, when they described a new species, M. randalli. They regarded small eyes, less than six times in head length, as one of the diagnostic characteristics of Microbrotula (Cohen and Wourms 1976). Orbit dia¬ meter ranges from 6.4 to 8.4 times and from 11.0 to 15.4 times in head length in M. randalli and in M. rubra, respectively (Cohen and Wourms 1976). Table 1 com¬ pares Acarobythites and Microbrotula. Acarobythites appears to be distinguishable from the latter by its naked head and body, small eye, compressed snout, posterior end of maxillary rounded, no ventrally-directed process at the postero-ventral corner of maxillary, opercular spine flat and weak, and 11-12 caudal fin rays. Etymology. The generic name is derived from the Latin, acaro (small), in reference to its small body, combined with Bythites (gender: masculine), one of the known genera in the family Bythitidae. Fig. 1. Acarobythites larsonae sp. nov., NTM S. 14665-026, holotype, 25.2 mm SL, from Australia. Scale bar indicates 5 mm. 124 New bythitid fish from Australia b Fig. 2. Lateral (a) and ventral (b) views of anterior part of head of NTM S. 14665-026, holotype of Acarobythiles larsonae sp. nov. Scale bar indicates 1 mm. Acarobythites larsonae new species (Figs 1-4) Type material. HOLOTYPE - NTM S. 14665-026, 25.2 mm SL, sex unknown, northwestern side of Field Island (12°04.20’S, 132°l9.3rE), Northern Territory, Australia, rock pools on reef, depth 0.3 m, substrate sand, mud, oyster covered rocks, coll. R. Williams and party, 5 June 1998. PARATYPE - NTM S. 10015-049, 18.6 mm SL, sex unknown, on silty sand with isolated coral clumps and rubble piles, off reef Hat on E side of Coral Bay, Cobourg Peninsula, Northern Territory, depth 3-4 m, coll. H. K. Larson, 17 Oct. 1981. Diagnosis. See generic diagnosis. Description. Data on the holotype given first, data on the paratype in parentheses when differing from holotype. Counts: dorsal fin rays 79 (73), anal fin rays 57 (53), caudal fin rays 12(11), pectoral fin rays 14 (13), pel vie fin ray 1, branchiostegal rays 7, developed gill rakers on anterior arch 3, pseudobranchial filaments 0, precaudal vertebrae 13(12), caudal vertebrae including urostyle 29 (32). Measurements in % SL: HL 21.4 (24.2), predorsal length 26.6 (29.3), preanal length 49.3 (46.2), body depth at dorsal fin origin 13.5 (15.9), body depth at vent 11.5 (13.7), head width 5.9 (8.6), pectoral fin length 8.7 (11.3), pelvic fin length 11.9 (13.7). Measurements in % HL: eye diameter 5.6 (5.6), snout length 22.2 (21.1), maxillary length 48.1 (48.9), fleshy interorbital width 18.5 (22.2). Head and body compressed, tail not tapered posteriorly (Fig. 1). Head small, slightly less than 1/2 preanal length. Snout round from lateral view, slightly projecting beyond upper jaw (Fig. 2a). Eye small, about 1/4 snout length. Anterior nostril tubular, just above upper lip. Posterior nostril small, rounded, slightly before anterior margin of eye, at mid-eye level. Mouth large, nearly horizontal. Maxillary extending backward far behind posterior margin of eye, its posterior end rounded, not sheathed by dermal cheek fold, without ventrally- directed process near its postero-ventral corner. Lower jaw included in upper jaw. Opercular spine flat and weak, completely covered by skin. Posterior and lower margins of preopercle smooth, completely covered by skin. Gill opening wide; gill membranes united slightly behind posterior margin of eye, free from isthmus. Pore-bearing skin tlap absent from just above upper angle of gill opening. A single pore with short tube in front of upper angle of gill opening. Infraorbital pores four; three large, slit-like pores between slightly behind anterior nostril and just below posterior nostril, a single, small pore above posterior end of maxillary. Supraorbital pore single, large, on underside of snout tip before anterior nostril (Fig. 2b). Two mandibular pores on each side; left and right anterior pores separated, left and right posterior pores united on ventral mid-line of head. A single, .small pore near lower angle of preopercle. All teeth sharp-pointed (Fig. 3). Upper jaw teeth uniserial, becoming longer anteriorly, needle-like. Lower jaw teeth biserial, inner teeth larger. About six teeth on vomer, uniserial, well separated, nearly equal to inner teeth on lower jaw. Palatine teeth about five, uniserial, slightly shorter than vomerine teeth. Tongue short, its tip bluntly pointed, free from mouth floor. Developed rakers on anterior arch three, short, about equal in length to eye diameter. Pseudobranchial filaments absent. 125 Y. Machida V V V y K V Fig. 3. Upper and lower dentition of NTM S. 14665-026, holotype of Acarobythites larsonae sp. nov. Scale bar indicates 0.5 mm. Dorsal fin originating above mid-length of pectoral fin. Origin of anal fin at about mid-body. Pectoral fin short, no separated or free rays. Pectoral peduncle normal, broader than long. No spine on posterior margin of cleithrum. Pelvic fins not reaching to below tip of pectoral fin, closely adjacent at their bases, inserted slightly behind posterior margin of preopercle. Head and body covered with mucous coating, completely devoid of scales. Tiny papillae sparsely present on anterior half of head. Lateral line indistinct. Tips of neural spines of precaudal vertebrae pointed; first neural spine short, 2nd spine longest, erect, 5th to 11th spines short, well depressed (Fig. 4). In alcohol, head and body uniformly creamy-yellow, slightly darker on dorsum of head and body, paler on belly. All fins and buccal cavity pale. Distribution. Known only from the type locality (Field Island) and Cobourg Peninsula, western Arnhem Land, Northern Territory, Australia. Etymology. Named larsonae after Dr. Helen K. Larson (NTM), who kindly sent me valuable bythitid and ophidiid specimens for study. ACKNOWLEDGMENTS I thank the staff of the Ichthyology Department, Museum and Art Gallery of the Northern Territory, Darwin, Australia, for the loan of the specimens described in this paper. REFERENCES Cohen, D.M. 1987. Notes on the bythitid fish genus Saccogaster with a new species from the Gulf of Mexico. Natural History Museum of Los Angeles County, Contributions in Science 385: 1-4. Cohen, D. M. and Nielsen, J.G. 1972. A review of the viviparous ophidioid fishes of the genus Saccogaster. Proceedings of the Biological Society of Washington 85: 445-468. Cohen, D.M. and Nielsen, J.G. 1978. Guide to the identification of genera of the fish order Ophidiiformes with a tentative classification of the order. NOAA Technical Report, NMFS Circular AM: 1-72. Cohen, D.M. and Wourms, J.P. 1976. Microbrotula randalli, a new viviparous ophidioid fish from Samoa and New Hebrides, whose embryos bear trophotaeniae. Proceedings of the Biological Society of Washington 89(5): 81-98. Gill, T. N. 1861. Catalogue of the fishes of the eastern coast of North America, from Greenland to Georgia. Proceedings of the Academy of Natural Sciences of Philadelphia 13 (Supplement): 1-63. Gosline, W.A. 1953. Hawaiian shallow-water fishes of the family Brotulidae, with the de.scription of a new genus and notes on brotulid anatomy. Copeia 1953(4): 215-225. Machida, Y. 1993. Two new genera and species of the subfamily Brosmophycinae (Bythitidae, Ophidiiformes) from northern Australia. Japanese Journal of Ichthyology 39(4): 281-286. Machida, Y. 1997. Hastatobythites arafurensis, a new bythitid genus and species from Australia (Bythitidae, Ophidiiformes). Ichthyological Research 44(4): 385-388. Nielsen, J. G. and Cohen, D.M. 1973. A review of the viviparous ophidioid fishes of the genera Bythites Reinhardt and Abythites new (Pisces, Ophidioidei). Steenstriipia, Zoological Museum, University of Copenhagen 3: 71-88. Nielsen, J. G. and Cohen, D. M. 1999. Family Bythitidae. In: Nielsen, J. G., Cohen, D. M., Markle, D. F. and Robins, C. R. (eds) Ophidiiform fishes of the world (Order Ophidiiformes}: an annotated and illustrated catalogue of pearlfishes, cusk-eels, brotulas and other ophidiiform fishes known to date. Pp. 94-135. FAO Fisheries Synopsis 125(18). Accepted 1 November 2000 Fig. 4. Positive image of X-ray photograph of NTM S. 14665-026, holotype of Acarobythites larsonae sp. nov. 126 The Beagle, Records of the Museums and Art Galleries of the Northern Territory, 2000 16: 127-144 Six new species of fork-tailed catfishes (Pisces, Teleostei, Ariidae) from Australia and New Guinea PATRICIA J. KAILOLA 19 Walkers Avenue, Newnham, TAS 7248, AUSTRALIA pkailola @ CompuServe, com ABSTRACT Six new species of Arius (Teleostei: Ariidae) are described from estuarine waters of southern New Guinea and northern Australia. Arius dioctes sp. nov. is pre,sent in both regions and is distinguished by its large maximum size, small eyes, fixed (non-depressible) jaw teeth, broad mouth, and yellow lower body. Arius insidiator sp. nov. has a much depressed body, long, angular ribs, small eyes covered by skin and fixed jaw teeth. It too has a yellow lower body. Arius pectoralis sp. nov. is a slender catfish with strong, numerous serrae along the hind margin of the pectoral spines and dark margins on the dorsal, pectoral and the caudal fins. Arius hainesi sp. nov. has fleshy lips, small mouth, partly restricted gill openings, short barbels, ridges in the skin on the back below the dorsal fin, and often a dark mouth and branchial chamber. The barbels of Arius hardenhergi sp. nov. are thick proximally. This species also has a fleshy snout, an extensive head shield with broad supraoccipital process and a long-based adipose fin. Ariuspaucus sp. nov. is a freshwater species very similar to Arius rnidgleyi Kailola and Pierce and inhabits the river systems draining into the Gulf of Carpentaria. Keywords: Ariidae; fork-tailed catfishes; new species; Australia; New Guinea; tropical marine; tropical freshwater. INTRODUCTION The isolation of the Sahul Shelf has enabled the evolution of many endemic species of fork-tailed catfishes (Ariidae). These fish inhabit the rivers, estuaries and coastal waters of tropical (northern) Australia and New Guinea. Earlier workers (e.g. Roberts 1978; Haines 1979; Coates 1983, 1988; Allen and Coates 1990) have remarked on the adaptive characteristics of these ariids. The Ariidae occurs worldwide in tropical and warm- temperate regions. It comprises at least 80 valid species and numerous nominal genera and species, many confined to particular geographic regions (west Africa, east Africa to India, South-east Asia, Australia-New Guinea, eastern Pacific, western Atlantic). The phylogenetic relationships of the family are under study. Thirty-five species of ariid catfishes inhabit Australian and New Guinea waters and only five of these, all marine catfishes (Arius thalassinus Riippell, 1837; A. bilineatus Valenciennes, 1840a; A. argyropleuron Valenciennes, 1840b, A. /ie//a (Valenciennes, 1840b) and A. polystaphylodon Bleeker, 1846), occur elsewhere: in South-east Asian, Indian and east African (A. thalassinus) seas (Kailola 1990b). Of the remaining 30 taxa, five species (A. nox Herre, 1935; A. solidus Herre, 1935; A. coatesi Kailola, 1990a; A. utarus Kailola, 1990a; and A. velutinus (Weber, 1908)) are endemic to northern New Guinea fresh waters, and 12 species are endemic to southern New Guinea fresh and saline waters (Kailola 1990b). Eleven more species are common to both northern Australia and southern New Guinea. Only two species - Arius rnidgleyi Kailola and Pierce and a new species - are endemic to Australia. The Ariidae are under-represented in collections. Most species attain moderately large size and it is sometimes not easy for field workers to keep intact material. Moreover, surveys have only been conducted since the mid-1970s in some of the areas which now reveal new species. The haphazard nature of ariid collecting in often difficult environments has created the paradox, however, that two of the endemic ariids from New Guinea described in 1913 by Weber (Doiichthys novaeguineae and Tetranesodon conorhynchus) are represented today by less than ten specimens. Yet the distribution of these taxa spans the most intensively collected aquatic habitats of southern New Guinea. I recognised nine Australo-Papuan ariid catfish taxa as new several years ago and three of them were described subsequently (Kailola and Pierce 1988; Kailola 1990a). Although I acknowledge the need of a phylogenetic review and comprehensive key to the species of this region, my objective here is to describe six new species. The other information will be published elsewhere. 127 P.J. Kailola METHODS Abbreviations (sucb as for bead length, standard length, gill rakers) and definition of ariid features and relevant osteology have been described in Kailola (1983). Method of counting and measuring are defined in Kailola (1983) and Kailola and Pierce (1988). The term ‘colour of fresh material’ refers to freshly dead (just caught) or chilled specimens except for A. insidiator, where ‘fresh colour' refers to frozen specimens. The term ‘serrae’ refers to large, regularly-spaced serrations. Institutional abbreviations follow Leviton et al. (1985) and are: AMS (Australian Museum, Sydney); CSIRO (Ian S.R. Munro Ichthyological Collection, CSIRO, Hobart); KFRS (Kanudi Fisheries Research Station collection. Port Moresby, Papua New Guinea); NTM (Museum and Art Gallery of the Northern Territory, Darwin); QM (Queensland Mu.seum and Art Gallery, Brisbane); USNM (United States National Museum, Smithsonian Institution, Washington); WAM (Western Australian Museum, Perth); ZMA (Zodlogische Museum, Universiteit van Amsterdam, Amsterdam). KFRS material of these new species identified ten years ago generally is not included in the type series as the present whereabouts and condition of the material is unknown. However, an exception was made for two new species because of the paucity of material. SYSTEMATICS Arius dioctes sp. nov. (Figs 1-2; Tables 1-2) Hexanematichthys sp - Kailola 1975: 41 (in part). Hexanematichthys sp ‘Y’ - Kailola and Wilson 1978: 42. Arius cf. stirlingi - Roberts 1978: 37, fig. 16a. Arius sp. - Maunsell and Partners 1982: 181.- Allen 1991: 56, photo 14. Arius (Hemiarius) species 1 - Kailola 1990b: 299. Hemiarius sp.l - Blaber, Brewer and Salini 1994: 163, fig 3a. Type material. HOLOTYPE - CSIRO C.3798, 430 mm SL, Norman River at Karumba (I7°29' S, 140‘’50' E), Queensland, D. J. Turner, 1969. PARATYPES - NTM S.l 1190-001, 2 (450-460 mm SL), Fog Bay (12°54’S, 130°I4’E), Northern Territory, P. Mundy, June 1983; AMS 1.15557-041,2 (103-172 mm SL), Gulf of Carpentaria at 17‘’26’S, 140‘’40’E., l.S.R. Munro, 23 December 1963; AMS 1.29292-001, 159 mm SL, Gulf of Papua (08°00' S, 145°00' E), S. Frusher, 11 July 1981; CSIRO H.5154-01, 1050 mm SL, Papua (Irian Jaya) at 04°49’S, 137°00’E, K. Hortle and A. Haris, 7 December 1996; NTM S.l4828-005, 95.1 mm SL, Papua (Irian Jaya) at 04°55'S, 137“15’E, K. Hortle and A. Haris, 28 May 1998; KFRS F.O4094, 200 mm SL, off Oreke River mouth (08°42' S,146°29' E), Papua New Guinea, J. Koaia, May 1973. Additional material examined (non-type). KFRS F.5627-01,350 mm SL, Tirere village, north side of Daru Island (09‘’05' S, 143°12' E), Papua New Guinea, April 1988; KFRS F.5626-03, 330 mm SL, Sagero, north side of Daru Island, Papua New Guinea, September 1988; KFRS F.5729-03, 166 mm SL, Daru Island. Papua New Guinea, 30 September 1983; KFRS F03992, 10 (84-89 mm SL), Pie River at Baimuru (07°33' S, 144‘’5r E), Papua New Guinea, 28 September 1971 (juveniles from mouth of an adult male. 104 cm SL); KFRS F04099, 328 mm SL, old Kukipi village, Lakekamu River mouth (08°12' S, I46°10 E), Papua New Guinea, 18 October 1971; CSIRO H.4957-01,4 (86-94 mm SL), Papua (Irian Jaya) at ()4°55’S, I37°03’E, K. Hortle and A. Haris, 28 May 1998; CSIRO H.4976-04, 325 mm SL. Papua (Irian Jaya) at 04‘’53’S, 136°55’E, 6 September 1997. Diagnosis. Teeth very strong, non-depressible; in narrow bands in jaws and in four patches arranged across front of palate. Snout prominent, jaw teeth exposed; mouth wide. Head somewhat venulose, head shield almost smooth. Eye small, free of head skin, 6-14 % HL; few (10-11) first arch gill rakers; no rakers on posterior face of first two arches; gill opening wide. Body orange-yellow; lips, inside mouth and body usually covered with orange or yellow mucus, and teeth frequently reddish. Description. D 1,7. P 1,11-12. A 17-21. GR (first arch) 10-11 of which 3-4 on upper limb. GR (last arch) 9-11. Vertebrae 49-50 (42-43 free). Body robust anteriorly, tapered, well compressed posteriorly. Prcdorsal profile straight becoming convex at nape. Snout prominent, well-rounded; inner lip margin scalloped. Jaws strong; mouth wide, curved, subterminal or inferior. When mouth closed, tooth band in upper jaw almost completely exposed, outer rows of lower jaw teeth also visible. Nostrils large, ovate, anterior one directly before posterior one. Eye ovate, moderately small, border completely free from head skin; eye dorsolateral, well before mid-head length. Gill opening wide, membranes broadly scalloped, meeting at sharp angle, leaving broad, free margin. All teeth sharp, strong, conical, not depressible. Premaxillary tooth band with 4-6 irregular series of teeth; mandibulary tooth band with 4-5 series of teeth, almost continuous across symphysis. Four patches of teeth in row across anterior of palate: two inner (vomerine) patches separated by narrow gap at midline, half as wide as outer, curved patches; 3-4 rows of teeth in each group. Palate smooth anteriorly, longitudinally creased posteriorly, with two long, low ridges ot epithelial tissue before branchial chambers. Head shield smooth anteriorly; several groups of sharp granules laterally, distinct parallel striae adjacent to distal third of dorsomedian head groove, striae New catfish from Australia and New Guinea Fig. 1. Arius dioctes sp. nov. lateral view, 200 mm SL paratype. Fig. 2. Arius dioctes sp. nov. (all 200 mm SL paratype) A, dorsal head view; B, ventral head view; C, arrangement of upper tooth patches; D, rubbing of head shield. 129 P.J. Kailola Table 1. Principal meristics of six new species, n = number; x = mean, SD = standard deviation; GR 1= rakers on first gill arch; GR 4 = rakers on last gill arch. Species variable Anal rays Pectoral rays Total GR 1 Total GR4 A. dioctes holotype 20 12 10 10 range, n = 9 17-21 11-12 10-11 9-11 X 18.7 11.6 10.3 10.0 SD 1.4 0.5 0.5 0.7 A. insidiator holotype 20 10 12 12 range, n = 4 19-23 10 11-12 11-13 X 20.5 10.0 11.5 11.5 SD 1.7 0.0 0.6 0.8 A. pectoralis holotype 18 8 15 15 range, n = 16 18-22 8-11 15-22 14-17 X 19.3 10.0 18.8 15.3 SD 1.2 0.9 1.8 1.1 A. hardenbergi holotype 18 10 14 13 range, n = 5 16-18 9-10 10-14 9-13 X 16.6 9.4 11.2 9.8 SD 0.9 0.5 1.6 1.8 A. hainesi holotype 20 10 32 36 range, n = 14 20-23 8-11 32-37 31-37 X 21.7 9.6 34.3 33.5 SD 0.7 0.8 1.4 1.7 A. paucus holotype 17 10 11 13 range, n = 16 16-18 9-10 10-11 11-14 X 17.3 9.9 10.7 12.8 SD 0.8 0.3 0.5 0.8 extending over supraoccipital process. Dorsomedian groove narrow, lanceolate, flat, extending from between nostrils to supraoccipital process base. Supraoccipital process with sharp, distinct keel, straight to slightly convex sides. Predorsal plate angular, striate. Numerous, anastomising venules over anterior two-thirds and sides of head, continuing onto shoulder. Rugose humeral process anteroventrally well ossified, forming narrow flange; acute, triangular shaft of process reaching one- quarter to one-third along pectoral fin spine. Axillary pore small and slit-like. Barbels thin distally. Maxillary barbel attains end of humeral process in juveniles, to well before opercular margin in adults. Mandibulary barbel reaches opercular margin in young, preopercular margin in adults. Mental barbel extends short distance behind eye. Chin barbel bases staggered. Gill rakers rigid, pungent, half to two-thirds length of opposing filaments. No rakers along posterior face of first two gill arches, 9-11 along posterior face of third arch. Low, muscular thickening posterodorsally on smooth arches. Fin spines thick, slightly curved, rounded, patterned with fine rugae and granules. Anterior margin of spines with distinct granules proximally, 6-10 antrorse, low serrae distally; posterior margin with 8-30 serrae or dentae. Short filament on spine tips. Longest dorsal ray 1.9-3.2 times longer than last ray. Pectoral fin low, extending to below dorsal fin. Ventral fin narrow in males, ending well before anal fin origin. Fin in females broad, attains anal origin or beyond (inner rays thicken and form pad of tissue in sexually mature individuals). Adipose fin moderately high, oblong, originating approximately opposite anal fin origin. Anal fin margin almost truncate, longest ray 2.1-3.6 times last ray. Caudal fin lobes broad-based, strongly tapered, upper lobe slightly longer than lower. Caudal peduncle shallow. Lateral line straight, oblique below dorsal fin, strongly upturned at tail base. Short lines and ascending rows of pores branch off from lateral line, numerous and extensive anteriorly. Colour of fresh material. Greenish or pale bluish grey above; pale orange or yellow on sides and below, less often cream or white. Bright yellowish orange mucus over all of body; lips and inside of mouth also yellow. Barbels and teeth reddish or yellow. Fins bluish grey, green or yellowish, dusky towards margins. Snout and head of some individuals blotched black. Colour of preserved material. Pinkish brown or tan above, cream or pale fawn below, top of head darker brown. Fins brown or light tan, margin of dorsal fin and caudal fin dark brown; upper aspect of pectoral fin, ventral fin and mid-anai fin rays darker brown. Maxillary barbel dark brown, others pale. Peritoneum grey. Comparisons. Arius dioctes is similar to A. stormii (Bleeker, 1858), a South-east Asian species growing to over 50 cm long (Weber and de Beaufort 1913). However, A. stormii possesses the following unique features: dorsal spine subequal to head length (versus equal to head length without snout in Arius dioctes)-, caudal peduncle depth 2.6-2.8 in its length (versus 1.9- 2.3 in Arius dioctes)-, anterior nostrils lateral to posterior ones (versus anterior to posterior nostrils in Arius dioctes)-, total gill rakers on the first arch 17-18 (versus 10-11 in Arius dioctes)-, serrae on front margin of dorsal spine directed upwards (versus directed downwards in Arius dioctes); fin spines very broad (versus moderately broad in Arius dioctes); adipose fin beginning before anal origin (versus opposite or behind in Aius diocte.s). The most phenotypically similar species to Arius dioctes in Australia and New Guinea is /\. armiger De Vis, 1884 (distinguished by its venulose head, small eye and strong teeth). The taxa can easily be distinguished however, on gill raker and anal fin meristics (A. armiger has GR 16-22, first arch; A 22-25) and maxillary barbel length (long maxillary barbel reaches at least to below dorsal fin). Arius dioctes has similar colouration and dentition to Arius insidiator sp. nov. However, the the form of the body easily distinguishes these two taxa. Distribution. New Guinea-, southern coast and rivers from Kamora River to the Otokwa River, possibly also 130 New catfish from Australia and New Guinea Table 2. Percent of head length (HL) and standard length (SL) for Arius dioctes sp. nov. and Arius insidiator sp. nov. n = sample size; X = mean; SD = standard deviation. Character holotype n Arius dioctes range X SD holotype n Arius insidiator range X SD Percent of HL head height 38.1 9 38.1 - 50.0 42.9 4.0 39.8 4 35.0 - 39.8 37.7 2.3 head width 66.3 9 61.3 - 71.3 66.3 2.9 64.1 4 61.9 - 64.9 64.4 2.0 eye diameter 6.8 9 5.7 - 13.2 8.8 2.5 6.0 4 5.6 - 6.8 6.0 0.5 mouth gape 47.3 9 44.7 - 52.3 47.4 2.5 44.3 4 42.8 - 45.8 44.0 1.2 intemostril distance 35.7 9 33.1 - 35.9 34.8 1.1 31.4 4 31.4 - 34.7 32.4 1.5 snout length 34.7 9 32.2 - 37.2 34.4 1.5 27.8 4 27.8 - 31.6 29.0 1.8 longest barbel length 54.4 9 40.7 - 77.4 61.3 13.2 32.4 4 32.4 - 52.0 40.1 8.5 bony interorbital width 33.1 8 26.6 - 35.4 31.7 3.2 25.8 4 25.8 - 27.8 26.5 0.9 occipital proc. length 28.1 9 25.5 - 34.4 29.6 2.6 28.8 4 28.8 - 36.9 33.0 3.5 occipital prix;. width 18.4 9 13.7 - 21.8 16.9 3.2 10.1 4 8.3 - 15.1 10.8 3.0 Percent of SL HL 31.4 9 31.4 - 34.7 32.5 1.1 32.6 4 29.8 - 32.6 31.1 1.2 head height 12.0 8 12.0 - 16.3 14.0 1.4 13.0 4 10.6 - 13.0 11.7 1.0 head width 20.9 9 19.5 - 22.9 21.5 1.0 20.9 4 18.8 - 21.1 20.0 1.1 eye diameter 2.1 9 2.0 - 4.0 2.8 0.7 1.9 4 1.8 - 2.1 1.9 0.2 mouth gape 14.9 9 14.2 - 16.5 15.4 0.7 14.4 4 13.0 - 14.4 14.0 0.6 intemostil distance 11.2 9 10.7 - 12.3 11.3 0.5 10.2 4 9.4 - 11.0 10.1 0.7 snout length 10.9 9 10.5 - 12.3 11.2 0.6 9.1 4 8.5 - 10.0 9.0 0.7 longest barbel length 17.1 9 14.1 - 24.7 19.9 4.2 10.5 4 10.5 - 16.4 12.5 2.7 bony interorbital width 10.4 8 8.5 - 11.4 10.3 1.1 8.4 4 7.9 - 8.4 8.2 0.3 occipital proc. length 8.8 9 8.7 - 10.9 9.6 0.8 9.4 4 9.4 - 11.2 10.2 0.8 predorsal length 39.4 9 39.4 - 42.8 41.1 1.1 37.3 4 35.4 - 37.3 36.3 0.9 length dorsal f. base 10.8 9 10.0 - 13.2 11.6 0.9 10.3 4 10.3 - 12.2 10.3 0.9 interdorsal length 22.8 9 21.6 - 25.7 23.6 1.3 28.8 4 25.1 - 28.8 6.0 0.4 length adipose f. base 10.4 9 8.0 - 12.3 10.4 1.4 9.0 4 9.0 - 11.1 10.0 0.9 length anal f. base 16.9 9 13.3 - 16.9 15.3 1.3 15.1 4 15.1 - 17.0 16.0 0.9 caudal ped. depth 6.5 9 6.5 - 7.3 6.9 0.3 6.3 4 6.2 - 7.1 6.5 0.4 caudal ped. length 13.8 9 11.2 - 17.7 14.4 2.0 14.5 4 13.3 - 15.0 14.3 0.7 pectoral spine length 19.5 8 17.3 - 21.3 19.6 1.6 15.0 4 15.0 - 16.4 16.0 0.6 dorsal spine length 21.8 8 17.1 - 23.2 21.1 2.4 16.8 4 15.6 - 18.3 16.9 1.3 further westward and in intermediate rivers; from there eastwards to the middle and lower river and delta of the Fly, throughout the Gulf of Papua rivers including the Kikori, Era, Pie and Purari systems and deltas, Vailala River mouth, Murua River, Lakekamu and Oreke Rivers. Australia: northern coast and rivers from the Adelaide and Alligator Rivers to the Gulf of Carpentaria and the Norman River. Ecology. Coastal and lower mangrove areas, estuaries into main rivers and lagoons; also freshwater. Remarks. In late 1976, this species was the second most important commercial species at Kikori (Papua New Guinea) after barramundi (Lales calcarifer (Bloch) (Haines 1979). Arius dioctes has been recorded to attain 1.2 m SL (Roberts 1978) and 19 kg whole weight (T. Coleman pers. comm.). Papuan villagers report that this species can attain 2 m in length. N. Haysom and T. Davis (pers. comm.) have collected 1-2 m, ‘up to 40 kg’ fish from the Alligator and Norman rivers. The 1050 mm SL paratype from Papua (Irian Jaya) weighed c. 15 kg when fresh. Etymology, dioctes, masculine. From the Greek, dioktes, meaning hunter or pursuer - in reference to the species’ apparent hunting ability, diet and dentition. Arius insidiator sp. nov. (Figs 3-4; Tables 1-2) Hexanematichthys sp - (in part) Kailola 1975: 41. Hexanematichthys sp ‘D’ - Kailola and Wilson 1978: 40, 42. Arius (Hemiarius) species 5 - Kailola 1990b: 306. Type material. HOLOTYPE - NTM S. 11189-001,350 mm SL, mouth of Wildman River (12°26' S, 132‘’09' E), Northern Territory, P. Mundy, 28 February 1984. PARATYPES - AMS 1.28960-001, 282 mm SL, Gulf of Papua (08‘’00' S, 145“00' E), S. Frusher, 1978; KFRS F.O3302, 188 mm SL, half mile inside Pai’a Inlet (07‘’39' S, 144°33' E), Papua New Guinea, F.R.V. Tagula, 6 May 1967; KFRS F.5526-01, 270 mm SL, off Kerema (07°58' S, 145‘’43' E), D. Whitten, D. Coates and R. Watson, 11 October 1983. Diagnosis. Head and anterior of body depressed; ribs long, angular, impressed in abdominal body wall. Eye covered with head skin; head smooth. Mouth wide, terminal, jaws upturned at symphysis; teeth fixed, cardiform, in 1-2 series on Jaws and palate; four elongate patches of palatal teeth. No rakers on posterior of first two arches; mandibulary barbel longest, reaching dorsal fin. A 19-23; total gill rakers (first arch) 11-12; total vertebrae 24+3+33. Fin spines slender, weak. Fresh P.J. Kailola Fig. 3. Arius insidiator sp. nov. lateral view, holotype Fig. 4. Arius insidiator nov. A dorsal head view; B, ventral head view (both of holotype); C, arrangement of upper tooth patches (188 mm SL paratype). New catfish from Australia and New Guinea colouration grey or yellowish, fins and mouth (bright) yellow or orange. Description. D 1,7. P 1,10. A 19-23. GR (first arch) 11-12, of which 5 on upper limb. GR (last arch) 11-13. Vertebrae 58-60 (51-54 free). Body moderately stout, anterior two-thirds depressed; angular pleural ribs of trunk and anterior caudal vertebrae apparent through body wall, forming a ‘shelf along lower sides. Predorsal profile almost straight. Snout evenly rounded; lips narrow, much thickened at corners, scalloped along inner margin. Jaws strong, lower jaw elevated at symphysis, slightly longer than upper; mouth terminal to superior, broad, curved; teeth not visible when mouth closed. Nostrils small, rounded, placed well forward on snout; anterior one directly in front of posterior one. Eye ovate, small, covered with head skin, almost dorsally situated and about four eye diameters before mid-head length. Gill opening wide, membranes meeting well forward; distal margin broad. Teeth fixed, strong, slightly curved, cardiform. Teeth on both palate and upper jaw arranged in two (rarely one) row(s) with distinct naked space between rows, teeth in inner row longer. One row of teeth on lower jaw; naked space at symphysis. Four elongate patches of teeth across palate: inner pair adjoining and half as wide as curved outer patches. Palate smooth; two long, low ridges of epithelial tissue before branchial chamber. Head shield completely smooth except for slightly roughened lateral margins. Dorsomedian head groove lanceolate, flat, beginning well forward on snout and almost reaching base of supraoccipital process. Process narrow, with straight sides, rounded from side to side. Sides of head smooth. Broad-based, triangular humeral process smooth, weakly ossified anteroventrally, its shaft short, oblique, extending one quarter distance along pectoral spine. Axillary pore small. Barbels thin, flattened. Maxillary barbel short, reaching only to preopercular margin. Mandibulary barbel long, extending past head or as far as anterior dorsal rays. Mental barbel reaches ventral head margin or beyond pectoral base. Chin barbel bases elearly staggered. Gill rakers stiff, moderately pungent, half as long as opposing gill filaments. Gill arches smooth; no rakers on posterior face of first two; 10-12 rakers along back of third arch. No thickened tissue postcrodorsally on arches. Fin spines feeble, rounded, pungent in smallest specimen; smooth or roughened along anterior margin, weakly serrated or roughened along posterior margin (smallest specimen with 8-9 serrae along posterior margin of pectoral spine). Dorsal fin truncate in outline, longest ray 2.3-2.8 times longer than last ray. Pectoral fin low on sides, its base in a horizontal plane; fin hind margin truncate above, concave below. Pectoral fin reaching to below dorsal. Ventral fin of both sexes broad. terminating well short of anal fin origin. Adipose fin above anterior two-thirds of anal. Anal fin margin almost straight, longest ray 2.6-3.1 times last ray in length. Caudal fin lobes broad and short. Caudal peduncle stout. Lateral line straight, curved dorsad at tail base. Numerous fine short lines emanate from lateral line, forming dense, branehing network anteriorly. Indistinct vertical series of pores ascend from lateral line over upper sides. Colour of fresh material. Pale grey to yellowish above, white below; colours well-separated. Barbels grey or cream; mouth ‘honey’-coloured. Dorsal, ventral, anal and pectoral fins bright yellow-orange; caudal fin dull yellow-orange. Colour of preserved material. Charcoal or dark lilac- brown above, upper jaw and undersides pale. Lower sides and underside of head cream or pinkish, colour extending dorsally along gill membranes; nostrils and eyes also in pale streaks. Dorsal, pectoral and ventral fins dusky yellow basally, charcoal distally or over dorsal aspect; anal and caudal fins dull yellow or pale orange. Peritoneum pale. Comparisons. This species is distinct in dentition, body form and colouration. Its colouration is similar to that of Arius dioctes and its strong teeth and small eyes link it to that species and to A. armiger De Vis, 1884. The only other species with a skin-covered eye is Doiichthys novaeguineae Weber, 1913 which also has an upturned mouth. Nevertheless, none of them have A. insidiator's flattened body, peculiar swimbladder form (flattened and elongate versus cardiform and inflated) and angular pleural ribs. Distribution. New Guinea: southern coast along the Gulf of Papua. Australia: northern coast east of Darwin. Probably more widespread (for example, individuals have been seined in the Fly River delta (C. Tenakanai, pers. eomm.). Ecology. Shallow mudflats and river mouths; muddy coastal waters. Remarks. The holotype is the largest specimen known. Etymology. Noun in apposition; from the Latin, insidiator, meaning ambusher or lurker. The dorsally placed eyes and depressed body form suggest that this species’ habit is to lie half-hidden in sediment. Its strong teeth and jaws indicate it is a predator. Although large fish scales and detritus were adhering to the gills of the largest specimen, the stomachs of all were empty. Arius pectoralis sp. nov. (Figs 5-6; Tables 1,3) Arius species 3 - Kailola 1990b: 397. Arius sp. 3 - Blaber, Brewer and Salini 1994: 168, fig. 3m. Type material. HOLOTYPE - AMS 1.27415-001, 226 mm SL, Chapman River, Queensland (14°56' S, P.J. Kailola Fig. S.Arius pectoralis sp. nov. lateral view, holotype. Fig. 6.Arius pectoralis sp. nov. (all of holotype) A, dorsal head view; B, ventral head view; C, arrangement of upper tooth patches, D, rubbing of head shield. 134 New catfish from Australia and New Guinea 141 °38' E), D.B. Carter, 21 December 1980. PARAT YPES - NTM S. 13004-001, 127 mm SL, Chapman River, Queensland, D.B. Carter, 5 March 1981; CSIRO A.3608, 112 mm SL, Norman River above Karumba, Queensland (17°29’ S, 140°50' E), l.S.R. Munro, 22 October 1972; CSIRO A.3609, 1 17 mm SL, same data as A.3608; CSIRO A.3610, 116 mm SL, same data as A.3608; QM 1.14917, 105 mmSL, Darwin Harbour (12‘’27’S, 130“46' E), November 1972; NTM S.10254-001,2 (94.5-101 mm SL), Mickett Creek, Shoal Bay, Melville Island, Northern Territory (12‘’2r S. 131°00' E), D. Grey, 16 December 1976; NTM S.10319-003, 56 mm SL, Shoal Bay, Melville Island (11°48' S, 130“39' E), D. Grey, 19 October 1972; NTM S. 10235-001, 90 mm SL, Shoal Bay, Melville Island, D. Grey, 1 August 1973; NTM S.11507-004, 145 mm SL, Ludmilla Creek, Darwin (12°27' S, 130°46' E), G. Cole and A. Howard, 19 December 1984; CSIRO H.5174-07, 4 (165-180 mm SL), Papua (Irian Jaya) at 04°52’S, 136°57’E, K. Hortle and A. Haris, 6 September 1997; CSIRO H.4937-02, 220 mm SL, Papua (Irian Jaya) at 04°52’S, ]36°57’E, K. Hortle and A. Haris, 6 September 1997. Additional material examined (non-type). CSIRO H.4220-03, 230 mm SL, Papua (Irian Jaya) at 04'’52’S, 136°55’E, K. Hortle and A. Haris, 8 February 1996; NTM S.14852-001, II (46-64 mm SL), Papua (Irian Jaya) at 04°52’S, I36°57’E, K. Hortle and A. Haris, 1 June 1998. Diagno.sis. Palatal teeth conical, in two oblique, oval patches; one on each side of palate, well-separated. Snout acute; lips fleshy, crenulate; lower jaw truncate. Head shield finely granular; dorsomedian head groove lanceolate, narrow distally. Eye large, 17-28 % HL. Large, flattened serrae along inner margin of pectoral spine; caudal fin lobes slender, acute. A 18-22; total first arch GR 15-22; rakers present along posterior face of all gill arches; vertebrae 18-l-6-t-27. Distinct, blackish margin on dorsal, pectoral and inner caudal fins. Description. D 1,7. P I, 8-11. A 18-22. GR (first arch) 15-22, of which 5-7 on upper limb. GR (last arch) 14- 17. Vertebrae 51 (44 free). Body moderately slender; predorsal profile straight. Snout moderately to slightly acute; lips fleshy, crenulate; short, transverse crescent often present on dorsum of snout between nostrils. Mouth subinferior, gape moderately large; upper Jaw somewhat acute medially, lower jaw almost truncate. Anterior nostril slightly medial to posterior one. Eye large, border free from head skin, dorsolateral and slightly before mid-head length. Gill opening moderately wide, concave at isthmus, margin moderately broad. Teeth in jaws fine, sharp, depressible, forming 7-8 irregular series in upper jaw and 4-6 series in lower jaw, bands either meeting or slightly separated at jaw midlines. Single large, oval patch of conical, blunt teeth on each side of palate anteriorly, obliquely situated, well- separated. Palate smooth or slightly papillose; two oblique, low to moderately developed epithelial ridges posteriorly on palate. Head shield finely granular. Dorsomedian head groove originates between nostrils and reaches supraoccipital process base. Groove flat, lanceolate anteriorly, narrow and straight posteriorly. Supra¬ occipital process triangular, straight-sided, with median keel. Triangular humeral process rugose, well-ossified anteroventrally, its shaft oblique, extending one-third along pectoral spine. Axillary pore small. Barbels flattened, moderately thick, fleshy. Maxillary barbel extends slightly beyond dorsal fin in juveniles, to humeral process in adults. Mandibulary barbel usually reaches pectoral fin spine base. Mental barbel reaches ventral head margin. Chin barbel bases slightly staggered. Gill rakers short, one-third to half length of opposing filaments. Rakers present along posterior aspect of all arches: 6-7 (upper limb only) to 11-14 (total) on first arch; 13-18 along second; 11-16 along third. Often some papillae on first two arches. Fleshy, scalloped epithelial folds moderately developed posterodorsally on first two arches. Fin spines strong, moderately long; sides with fine, longitudinal striae. Short filament on tip of spines. Anterior spine margin finely rugose with several sharp, antrorse serrae near tip; posterior (trailing) margin with low (dorsal spine) or large and flattened (pectoral spine) serrae: 8-11 along dorsal, 12-13 along pectoral. Last dorsal fin ray 2.9-3.4 times shorter than longest ray. Pectoral fin extends to below posterior dorsal fin rays. Ventral fin in males narrow, failing to reach anal origin by distance of half eye diameter; in females, broad-based, reaching to fourth anal ray (inner ventral fin rays thickened to form a pad in sexually mature female fish). Adipose fin rectangular, situated above middle of anal fin. Anal fin margin slightly concave, last ray 2.5-3 times shorter than longest ray. Caudal fin lobes slender, tapered, acute, upper lobe slightly longer. Caudal peduncle moderately stout. Lateral line straight along sides, oblique below dorsal fin, turned dorsad at tail base. Fine, short lines of pores diverge off length of lateral line. Colour of fresh material. Dark bluish brown or black above, white to cream below. Charcoal or black margin to fins (particularly caudal fin) and anterior two-thirds of anal fin also charcoal. Colour of preserved material. Dusky fawn to charcoal above, creamy below; colours well-separated. Maxillary barbel usually dark. Unpaired fins and dorsal aspect of pectoral and ventral fins dusky or charcoal; distinct charcoal or black margin to inner caudal, dorsal and pectoral fins, and charcoal anal fin. Peritoneum pale, occasionally with scattered, dark stipples. 135 P.J. Kailola Fig. 7. Anus hardenbergi sp. nov. A, lateral view of 74.3 mm SL paratype; B, lateral head view of 253.8 mm SL paratype. Fig. 8. Arius hardenbergi sp. nov. A, dorsal head view sketch of 253.8 mm SL paratype; B, rubbing of head shield, same paratype; C, arrangement of upper tooth patches, 74.3 mm SL paratype. 136 New catfish from Australia and New Guinea Table 3. Percent of head length (HL) and standard length (SL) for Arius pectoralis sp. nov. and Arius hardenbergi sp. nov. n = sample size; X = mean; SD = standard deviation. Character holotype n Arius pectoralis range X SD holotype n Arius hardenbergi range X SD Percent of SL head height 60.9 16 51.0 - 64.4 58.9 4.8 47.0 5 47.0 - 55.6 51.0 3.1 head width 68.9 16 63.5 - 73.5 69.6 3.0 73.0 5 71.9 - 76.2 71.8 5.4 eye diameter 19.4 16 17.1 - 28.0 23.2 3.5 10.1 5 8.4 - 11.3 9.6 1.2 mouth gape 40.2 15 29.5 - 43.6 38.8 4.8 40.9 5 40.9 - 54.1 50.8 5.6 intemostril distance 25.0 15 20.6 - 28.4 25.0 2.2 25.9 5 25.9 - 35.5 30.4 3.4 snout length 38.5 16 33.5 - 41.2 37.5 2.2 35.6 5 34.5 - 39.3 37.1 2.0 longest barbel length 79.9 15 78.0 -121.7 101.0 13.1 76.7 5 76.7 -110.5 86.6 23.2 bony interorbital width 29.4 15 26.3 - 33.0 29.8 2.0 31.7 5 29.9 - 34.3 32.1 1.9 occipital proc. length 28.2 15 28.2 - 38.2 33.9 2.5 - 4 26.9 - 41.6 31.5 7.3 occipital proc. width 19.4 15 18.7 - 26.0 21.8 2.8 - 4 5.4 - 34.1 18.2 12.0 Percent of SL HL 29.2 16 25.5 - 29.8 27.3 1.3 32.1 5 31.9 - 32.9 31.5 1.7 head height 17.8 16 14.4 - 17.8 16.1 1.0 15.1 5 15.1 - 17.9 16.5 1.1 head width 20.1 16 17.3 - 21.0 19.0 1.1 23.5 5 23.4 - 24.5 23.9 0.5 eye diameter 5.7 16 2.1 - 7.3 6.0 1.3 3.2 5 2.7 - 3.6 3.1 0.4 mouth gape 11.7 15 8.5 - 13.8 10.6 1.3 13.1 5 13.1 - 17.8 16.4 1.9 intemostil distance 7.3 15 5.7 - 8.0 6.8 0.6 8.3 5 8.3 - 11.3 9.8 1.1 snout length 11.2 16 9.1 - 11.7 10.2 0.8 11.4 5 11.3 - 12.7 12.0 0.7 longest barbel length 23.3 15 22.4 - 32.8 27.5 3.1 24.6 5 24.6 - 36.1 30.4 5.5 bony interorbital width 8.6 15 7.1 - 9.8 8.2 0.8 10.2 5 9.8 -11.0 10.4 0.5 occipital proc. length 8.2 15 8.2 - 10.6 9.3 0.8 - 4 3.0 - 14.1 8.5 4.5 predorsal length 37.9 16 35.2 - 40.0 36.9 1.5 42.0 5 39.2 - 42.0 40.5 1.3 length dorsal f. base 11.9 16 10.0 - 12.5 11.3 0.7 12.5 5 10.6 - 13.9 12.3 1.2 interdorsal length 27.7 16 16.5 - 33.7 28.2 3.7 26.2 5 14.3 - 26.2 18.5 5.1 length adipose f. base 8.5 16 6.7 - 11.5 8.5 1.3 12.1 5 11.7 - 18.9 16.0 3.7 length anal f. base 17.6 15 15.3 - 19.3 17.1 1.2 14.5 5 12.5 - 16.9 15.0 1.7 caudal ped. depth 8.0 16 7.1 - 8.7 7.9 0.5 7.6 5 6.9 - 7.8 7.4 0.4 caudal ped. length 15.9 16 15.2 - 20.2 17.0 1.2 14.8 5 14.6 - 17.8 15.9 1.3 pectoral spine length 19.8 15 19.5 - 26.0 21.4 1.7 21.8 3 17.3 - 21.8 19.9 2.3 dorsal spine length 21.7 14 19.0 - 27.3 21.9 2.0 15.3 4 15.3 - 18.8 16.9 1.5 Comparisons. Arius pectoralis is most similar to Arius berneyi (Whitley, 1941) and A. graeffei Kner and Steindachner, 1866, from which it can be distinguished by its more acute snout, palatal dentition (almost always a single patch on each side, compared with almost always two patches each side in A. graeffei and A. berneyi', also, palatal teeth finely conical instead of villiform in A. graeffei and A. berneyi), its strong pectoral spine serrae, very dark fin margins and slender caudal fin. Arius pectoralis differs from the only other taxon in the area with similar palatal dentition (Cinetodus froggatti (Ramsay and Ogilby, 1886)) by its wide gill opening and head shield form (restricted gill opening and very broad and high head shield in C. froggatti). Distribution. New Guinea', south coast and estuaries of Papua (Irian Jaya) from the Kamora to Otokwa rivers. Probably occurs more extensively, in pockets, including the Port Moresby to the Kempwelch River area in central southern Papua New Guinea. Australia: Darwin (harbour and Ludmilla Creek) to Karumba and the Chapman River in Queensland. Ecology. Coastal waters, estuaries and rivers within tidal influence; among mangroves. Remarks. There are no literature records of this species, which has probably been confused in the field and earlier reports for Arius berneyi or A. graeffei. The maximum recorded size for A. pectoralis is 393 mm FL (Papua (Irian Jaya) specimen). Etymology. From the Latin, pectoralis, meaning shoulder: refers to the strong serrae along the inner pectoral fin spine. Arius hardenbergi sp. nov. (Figs 7-8; Tables 1,3) Arius species 6 - Kailola 1990b: 455. Type material. HOLOTYPE - WAM P.29966-001, 260 mm SL, Manimeri River (02“06' S, 133°45' E), Bintuni Bay, Papua (Irian Jaya), G.R. Allen, 28 March 1989. PARATYPES - NCIP 436, 254 mm SL, Merauke (08°28'S, 140°20' E), Papua (Irian Jaya), J.D.F. Hardenberg, 4 August 1931; AMS 1.29291 -001,2 (59.5-74.5 mm SL), delta of Fly River (08‘’22' S, 142°40' E), Papua New Guinea, J. Watson, 16-22 July 1981; QM 1.26088, 56 mm SL, same data. Additional material examined (non-type). NTM S. 14873-001, 52 mm SL, Bamu River estuary, Papua New Guinea, at c. ()8°08’S, 143°35’E, 21 June 1993. 137 P.J. Kailola Diagnosis. Four patches of conical teeth across front of palate: rounded vomerine patches well-separated; outer patches oval or crescentic, 2-3 times larger than inner patches. Snout fleshy, overhanging mouth. Low papillae on palate and posterior face of gill arches; rakers rarely on posterior face of first two gill arches. Barbels thick proximally, thin distally. Head shield very granular, extensive posteriorly; triangular supraoccipital process short, broad. Eye small, dorsolateral, 8-12 % HL. A 16- 18; total gill rakers (first arch) 10-14; vertebrae 19+5+26. Adipose fin large, base longer than anal fin base. Fresh colouration dark grey; inner, dorsal aspect of paired fins charcoal blue. Description. D 1,7. P 1,9-10. A 16-18. GR (first arch) 10-14, of which 3-5 on upper limb. GR (last arch) 9-13. Vertebrae 50-52 (43-46 free). Body robust. Head broad, depressed; predorsal profile almost straight, slightly convex at nape. Snout rounded to slightly acute, prominent; lips moderately thick and fleshy, inner margin crenulate. Tiny, fine papilla-like structures on snout and around mouth in all specimens. Mouth very broad, curved, more inferior than sub¬ terminal; one-quarter to half of upper jaw tooth band exposed when mouth closed. Nostrils rounded, anterior one slightly lateral to posterior one. Eye small, not completely free of head skin, situated dorsolaterally, well to slightly before mid-head length. Gill opening moderately wide, united membrane concave over isthmus, margin broad and free. Teeth conical, tips sharp or compressed; very slightly depressible; embedded in thick tissue. Five to eight irregular series of teeth in premaxillary band. 4 to 6 in lower band; narrow edentulous space at symphysis of lower jaw. Four tooth patches across front of palate, separated by narrow spaces. Vomerine patches rounded; outer patches large, oval, at least twice as large as vomerine patches. Palate smooth or with few scattered low papillae; two low oblique ridges of epithelial tissue before branchial chamber. Head shield extensive, very granular, granules low, coarse in large individuals. Ridges of striae flank dorsomcdian head groove posteriorly; broader striae extend over supraoccipital process. Groove lanceolate, beginning on snout, extending to base of supraoccipital process; groove flat anteriorly, narrow and deeply excavated posteriorly. Supraoccipital process broad, triangular; sides almost concave; strong median keel. Sides of head slightly venulose; small, oval, naked space in head shield above operculum. Humeral process broad- based, heavily ossified anteroventrally; its triangular shaft extending one-third distance along pectoral fin spine. Axillary pore moderately large. Barbels flat, thick proximally, thin and wisp-like distally. Maxillary barbel extends to base of pectoral fin or to below dorsal fin base (juveniles). Mandibulary barbel reaches pectoral fin base. Mental barbel ends midway between eye and pectoral fin base. Bases of chin barbels moderately staggered. Gill rakers well-spaced, half as long as opposing filaments. Numerous, low papillae along back of first two gill arches (fewer in larger fish); no rakers on poste¬ rior face of first arch; rarely 1-3 rakers posterodorsally on second arch; 9-10 rakers on posterior face of third arch. Narrow pads of epithelial tissue on gill arches posterodorsally, best developed on second. Fin spines robust, tips sharp. Outer spine margin with low granules and 1-3 serrae distally; posterior margin with 3-5 serrae (dorsal spine) or 5-10 large antrorse serrae (pectoral). Longest dorsal ray 1.9-2.2 times last ray. Pectoral fin extends to below mid-dorsal fin. Ventral fin short of or reaches anal fin origin (condition in mature females unknown). Adipose fin oblong, long-based, beginning noticeably before anal fin origin, opposite all of anal fin. Anal fin margin concave, longest ray 2.4- 3.2 times last ray. Caudal fin moderately short, lobes broad, tapered. Caudal peduncle moderately deep. Lateral line turned dorsad at tail base, elevated below dorsal fin. Low, short lines of pores diverge from length of lateral line, more abundant anteriorly. Colour of fresh material. Brown to dusky mauve above, pale yellow or fawn below; iridescent gold on sides. Fins dusky mauve or brown. Barbels pale yellow. Colour of preserved material. Light brown or tan, paler below; fins brown; barbels fawn. Peritoneum pale. Comparisons. Arius hardenhergi can be distin¬ guished from the similar A/7 mx proximus (Ogilby, 1898) by its larger adipose fin (12-19% SL, compared with 6- 12% in A. proximus). mouth shape, eye diameter (8-12.5% HL, compared with 14-28% in A. proximus), extent of head shield, supraoccipital process shape, and swimbladder form (margins smooth versus scalloped in A. proximus). The supraoccipital and mouth shapes, paired fin colouration and adipose fin size of A. hardenhergi are similar to those of Cinetodus (Kailola 1990b). However, the taxa belonging to that genus have rakers along the posterior face of the first two gill arches, different palatal dentition and more restricted gill opening. Cinetodus carinatus (Weber, 1913) can be confused with A. hardenhergi but C. carinatus has more gill rakers on the first gill arch (15-19) and more and .stronger .serrae along the inner pectoral spine. Distribution. New Guinea: south coast. Vogelkop Peninsula to the Fly River mouth. Ecology. Muddy, shallow coastal waters and tidal rivers. Remarks. On a label attached to the largest specimen from Merauke, is written 'Arius uniformis Hardenberg’. Hardenberg did not publish a description, and there is no unpublished manuscript at the ZMA. This name is therefore not available. This species probably attains a larger size, although the maximum size specimen 1 have measured is 254 mm New catfish from Australia and New Guinea SL. It could have been mistaken in collections for several other species such as Arius proximus (Ogilby) and members of the genus Cinetodus. Etymology. I name this species after J.D.F. Hardenberg, not only because he recognised that the species was new but also as an acknowledgement of his insightful contributions to Indo-Australian ichthyology in the mid-twentieth century. Arius hainesi sp. nov. (Figs 9-11; Tables 1,4) Nedystoma dayi - Tortonese 1964: 24. Nedysloma dayi (in part) - Kailola 1975: 42. Nedystoma sp. - Liem and Haines 1977: 25. Arius species 4 - Kailola 1990b: 459. Arius sp. 4 - Blaber, Brewer and Salini 1994: 168, fig. 3n. Type material. HOLOTYPE - NTM S. 11507-001, 304 mm SL, Ludmilla Creek (I2“27' S, 130“46' E), Darwin, Northern Territory, G. Cole and A. Howard, 19 December 1984. PARATYPES - AMS 1.25995-001, 204 mm SL, Baimuru (07°33’ S, 144“5r E), Papua New Guinea, A.K. Haines, 12-14 September 1974; QM 1.22657, 228 mm SL, Morowan (07'’35’ S,I44‘’58' E), Ini Island, Papua New Guinea, A.K. Haines, 15 May 1975; AMS 1.27414-001, 237 mm SL, same data, 237 mm SL (cleared and stained); USNM 288553, 135 mm SL, Gulf of Papua (OSW S, I45‘’00’ E), S. Frusher, 13 July 1981; AMS 1.25996-001, 136 mm SL, Gulf of Papua, A.K. Haines, May 1975; CSIRO C.3799, 187 mm SL, Norman River at Karumba (17“29' S, I4()‘’50' E), Queensland, D.J. Turner, 1969; NTM S. 10190-002, 2 (88.0-88.2 mm SL), Mickett Creek (12“2r S, 131°00' E), Melville Island, Northern Territory, D. Grey, 26 June 1975 (larger specimen cleared and stained); CSIRO H.4545-01,2 (120.4-125.9 mm SL), Papua (Irian Jaya) at 4°56.6’S, 137°03.2’E, K. Hortle and A. Haris, 4 September 1996; CSIRO H.4937-03, 222.5 mm SL, Papua (Irian Jaya) at 4°52.rS, I35°57.7’E, K. Hortle and A. Haris, 6 September 1997; CSIRO H.5252-01, 2 (210-234 mm SL), Papua (Irian Jaya) at 4°53.8’S, 136°54.6’E, K. Hortle and A. Haris, 6 September 1997. Additional material examined (non type material). KFRS F.O3990, 181.5 mm SL, Murua (07‘’52' S, 145°47' E), Matupi River, Papua New Guinea, October 1971; KFRS F.O4093, 3 (115-193 mm SL), off Oreke River mouth (08°42' S, I46'’29' E), Papua New Guinea, P. Kailola and J. Koaia, 3-4 May 1973; CSIRO H.4958- 01, 12 (33-42 mm SL), Papua (Irian Jaya) at 4°52.4’S, 136°57’E, K. Hortle and A. Haris, 1 June 1998 (mouth juveniles); CSIRO H.5174-06, 210 mm SL, Papua (Irian Jaya) at 4°52.6’S, 136°56.5’E, K. Hortle and A.Haris, 6 September 1997; WAM P.29965-002, 94 mm SL, Manimeri River (02‘’06' S, 133°45’ E), Papua (Irian Jaya), G.R. Allen, March 1989; AMS 1.27413-001,2 (108-151 mm SL), Arehava (07‘’35' S, I44“57' E), Papua New Guinea, A.K. Haines, September 1974. Diagnosis. Palate without teeth (but autogenous tooth plates present); jaw teeth slender, in short, oblong band. Lips fleshy, thin; mouth small, lower jaw truncate. Barbels thin, short, maxillary barbel reaching only just beyond eye; bases of chin barbels close together. Eye large, 14-24 % HL, lateral. Dorsomedian head groove elongate, deep posteriorly. Gill rakers along posterior P.J. Kailola Fig. 10. Arius hainesi sp. nov. A, dorsal head view (228 mm SL paratype), B, ventral head view (holotype). of all arches; gill opening slightly restricted; two large epithelial flaps on palate posteriorly and double folds on upper limb of first two gill arches. A 20-23; total gill rakers (first arch) 32-37. Longitudinal furrows antero- dorsally on body. Fin spines thin, long, slender; adipose fin short-based, over posterior two-thirds of anal; ventral fin pad of sexually mature females scalloped and tapered. Fresh colouration dark grey above and iridescent purple; palate and branchial chamber often purplish brown to charcoal. Description. D 1,7. P I, 8-11. A 20-23. GR (first arch) 32-37, of which 10-12 on upper limb. GR (last arch) 31- 37. Vertebrae 49-50 (43-44 free). Body moderately elongate, cylindrical; head slightly depressed. Predorsal profile straight, abruptly convex at nape. Snout slightly rounded or truncate, curved ventrad, overhanging subinferior mouth. Lips thin, fleshy or ‘rubbery’, inner aspect spongy and papillose. Mouth gape moderately wide, lower jaw truncate or horizontal; all of premaxillary tooth band visible when mouth closed. Nostrils large, rounded, anterior nostril slightly median to posterior one. Shallow groove usually present on snout between posterior nostrils. Eye large, rounded. Fig. 11. Arius hainesi sp. nov. A, rubbing of head shield (228 mm SL paratype), B, arrangement of upper tooth patches (204 mm SL paratype), C, ventral fin of mature female showing pad of epithelial tissue. prominent, situated laterally at or slightly before mid¬ head length, margin free of head skin; lateral ethmoid prominent before eye. Gill openings somewhat restricted ventrally, membranes attached broadly to isthmus, margin concave. Teeth in jaws slender, long, depressible; very fine; embedded in tissue; in 5-6 series. Tooth band (both jaws) interrupted at symphyses by edentulous space. Pre¬ maxillary bands elongate-oval. No teeth on palate. Palate almost smooth or with low, scattered papillae. One large pair of epithelial tissue folds hangs from palate into anterior part of branchial chamber. Head shield rugose or finely granular. Granules conspicuous, usually .sharp, arranged in series along dorsomedian head groove and along edge of shield in larger specimens. Dorsomedian groove begins at level of nostrils and continues to supraoccipital process base; groove flat anteriorly, thence deeply excavated; conspicuous. Rugose supraoccipital process triangular with straight sides; sharp median keel prominent. Naked space in head shield above gill opening. Large venules on sides of head. Lateral head skin and ‘shoulder’ skin develop 7-10 longitudinal furrows or ridges with age: most apparent in individuals longer than about 150 mm SL. Humeral process triangular, indented above, heavily ossified anteroventrally, shaft oblique, acute; granules arranged in series along process. Moderate sized, slit¬ like axillary pore present. Barbels thin, rounded proximally, flattened and wisp¬ like distally. Maxillary barbel reaches to eye or halfway between eye and gill opening (juveniles); mandibulary barbel extends to below middle of eye; mental barbel not or just reaches front eye margin. Chin barbel bases close together, almost transversely aligned. Gill rakers long, equal to gill filament length. Shorter rakers along posterior face of all arches: 32-36 along first arch, 33-40 on second. 30-34 on third. Curious large double fold (or pouch) of epithelial tissue links upper limb of first two arches. Fin spines long, slender, rather compressed: sides smooth to finely striate; anterior border with sharp granules or ridges. Low serrae along distal part of dorsal fin spine hind border and 25-35 sharp, large retrorse serrae along pectoral fin spine hind border. Last dorsal fin ray 3.0-3.7 times in longest ray. Pectoral fin reaches just beyond dorsal fin. Ventral fin narrow in males, reaching second or third anal fin ray; broad in females, extending to about fifth anal fin ray; mature females with curiously-shaped thickening along dorsal aspect of sixth ray consisting of series of lobes and proximally a short, lateral horizontal process (Fig. 11C). Adipose fin short- based. oblong; above middle of anal fin. Outer margin of anal fin truncate (adult) to deeply concave (juveniles), last ray 2.8-3.2 times in longest ray. Caudal fin lobes narrow, tapered, upper lobe slightly the longer. 140 New catfish from Australia and New Guinea Caudal peduncle moderately deep. Lateral line elevated below dorsal fin, curved dorsad at tail base. Thin, dorsally-directed branchlets of lateral line conspicuous; vertical series of fine pores diverging from lateral line extend over upper body surface. Colour of fresh material. Iridescent charcoal-purple on head and back, otherwise dark grey or brown above, white below. Sharp division between dorsal and ventral colours. Fins pale. Lining of mouth and branchial chamber either white to cream, or purplish brown to charcoal. Colour of preserved material. Dark brown or charcoal grey over upper two-thirds of head and body, pale orange-fawn or dusky grey below. Both lips pale. Dorsal, adipose and caudal fins dusky tan with broad, brown margins. Upper aspect of pectoral fin dark grey or brown; upper ventral fin, anterior anal fin rays and anal fin margin dusky. Palate and branchial chamber pale to dark brown or charcoal. Barbels brown. Comparison.s. Arius hainesi superficially re.sembles Nedystoma dayi (Ramsay and Ogilby, 1886) in having a toothless palate, epithelial folds around and before the anterior gill arches, and many gill rakers. However, the Table 4. Percent of head length (HL) and standard length (SL) for Arius hainesi .sp. nov. n = sample size; x = mean; SD = standard deviation. Character holotype n Arius hainesi range X SD Percent of HL head height 54.8 14 45.6 - 65.1 55.2 5.5 head width 68.1 14 52.3 - 74.6 66.4 5.7 eye diameter 14.1 14 14.1 - 23.8 18.7 2.5 mouth gape 34.0 14 27.5 - 37.9 32.2 2.8 intemostril distance 18.4 14 15.6 - 19.1 18.2 0.9 snout length 43.3 14 38.2 - 43.3 40,6 1.7 longe.st barbel length 34.2 14 28.9 - 55.8 42.1 7.7 bony interorbital width 28.0 14 27.0 • 42.6 30.1 3.7 occipital proc. length 28.0 14 22.9 - 50.0 30.3 6.6 occipital proc. width 12.1 14 12.1 - 20.5 16.6 2.3 Percent of SL HL 32.1 14 27.7 - 33.0 30.4 1.9 head height 17.6 14 15.0 - 18.4 16.7 1.2 head width 21.9 14 17.3 - 21.9 20.1 1.4 eye diameter 4.5 14 4.5 - 6.6 5.7 0,6 mouth gape 10.9 14 8.4 - 12.1 9.8 l.l intemostil distance 5.9 14 4.9 - 6.0 5.5 0,4 snout length 13.9 14 10.8 - 13.9 12.4 l.l longest barbel length 11.0 14 9.5 - 15.8 12.7 1.9 bony interorbital width 9.0 14 8.0 • 12.8 9.2 1.2 occipital proc. length 9.0 14 7.5 - 13.9 9.1 1.5 predorsal length 40.8 14 36.5 - 42.2 39.6 1.6 length dorsal f. base 11.6 14 9.9 - 12.3 11.2 0.6 interdorsal length 28.1 14 25.9 - 31.0 28.8 1.5 length adipose f. base 6.2 14 4.5 - 9.4 6.7 1.4 length anal f. base 17.0 14 17.0 - 19.9 18.5 1.0 caudal ped. depth 8.3 14 7.1 - 8.5 7.9 0.4 caudal ped. length 15.4 14 13.9 - 17.6 15.8 1.1 pectoral spine length 23.0 11 19.2 - 25.1 22.7 2.0 dorsal spine length 23.4 11 21.4 - 26.4 23.9 1.4 shapes of the posterior head groove and mature female ventral fin, the short barbels (maxillary harbel 29-56 % HL, compared with 46-76 % HL in N. dayi), longer snout (38-43 % HL, compared with 30-36 % HL in N. dayi) are among distinguishing characters. The two species also apparently occupy different habitats {N. dayi is almost strictly freshwater). Tortonese’s (1964) ‘Ne¬ dystoma dayi’ from Katau (09°08' S, I42°56' E) must be referable to this species. Distribution. New Guinea: southern coast from Kamora to Otokwa rivers (Papua (Irian Jaya)) and from Katau (west of Fly River delta) to Aird Hills and Oreke River (Papua New Guinea); possibly continuous distribution and more extensive westwards. Australia: from vicinity of Darwin to the southern Gulf of Carpentaria. Ecology. Predominantly in marine waters of the lower estuaries, along the coast and off river mouths. In Papua New Guinea the species is common locally (such as the Era River - Morowan area) (Haines 1979) although in central-southern Papua (Irian Jaya), A. hainesi is common in the estuaries. Remarks. Arius hainesi attains a maximum size of 320 mm FL (Papua (Irian Jaya) material). Etymology. Named for Alan K. Haines, who worked in fisheries in Papua New Guinea between 1972 and 1976. He undertook several surveys of river systems entering the Gulf of Papua (notably the Piirari) and de¬ veloped a sound appreciation of those river systems and their aquatic fauna. Through his ability to distinguish the species and his faithful recording of the biology and ecology of the fork-tailed catfishes in the Gulf rivers, Alan made a significant contribution to our knowledge of this fish family in the Australian and New Guinea region. Arius paiiciis sp. nov. (Figs 12-13; Table 5) Arius midgleyi Kailola and Pierce, 1988; 75 (in part). Type material. HOLOTYPE - QM 1.12910, 326 mm SL, Flinders River near Maxwellton (20°47' S, 142°43' E), Queensland, H. and M. Midgley, 14 October 1974 (paratype of Arius midgleyi). PARATYPES - QM 1.12757, 310 mm SL, Flinders River near Maxwellton, H. and M. Midgley, 14 October 1974 (paratype of Arius midgleyi)-, QM 1.16730, 2 (315-329 mm SL), Flinders River near Maxwellton, H. and M. Midgley, October 1974 (paratypes of Arius midgleyi)-, QM 1.11364, 205 mm SL, Forest Home Station (18°15' S, 143°02' E), Gilbert River, Queensland, T.C. Marshall, 24 September 1953 (paratype of Arius midgleyi)-, AMS 1.25315-001 (previously IB.3159/2882), 171 mm SL, same data (paratype of Arius midgleyi); QM 1.11990, 145.5 mm SL, Mitchell River (16°00' S, 142°20' E), Queensland, 8 September 1959, (paratype of Arius midgleyi); QM I. 16735, 240 mm SL, Hodgson River (14°54' S, 134‘’33' E), Northern Territory, H. and M. Midgley, 17 September 141 P.J. Kailola CO Fig. 12. Range of total gill raker count (first arch) of Arius midgleyi and A. paucus sp. nov.(species combined) and 57 specimens of Arius leptaspis (Bleeker) from northern Australia and southern New Guinea. Arius paucus sp. nov. counts are at the left of the figure. 1979 (paratype of Arius midgleyi)-, QM 1.16738, 2 ( 327 mm SL and 151.5 mm HL), Mannaburoo Hole (16°05' S, 135°22' E), Limmen Bight River, Northern Territory, H. and M. Midgley, 29-30 August 1979 (paratypes of Arius midgleyi)', NTM S. 12070-001,2 (298-315 mm SL), same data (paratypes of Arius midgleyi)-, QM 1.16737, 310 mm SL, same locality, H. and M. Midgley, 1 September 1979 (paratype of Arius midgleyi)-, NTM S.12083-001, 331 mm SL, Wilton River (13“46' S, 134°28' E), H. and M. Midgley, 25-27 September 1979 (paratype of Arius midgleyi)', NTM S. 1775, 257 mm SL. Wollogorang Station (17° 13' S, 137°57' E), Northern Territory, D. Howe, 15 June 1974 (paratype of Arius midgleyi)-, CAM F.36. 273 mm SL, same data (paratype of Arius midgleyi). Additional material (non-type). QM 1.16741, 2 (123.5-131 mm HL), McArthur River (24°46'S, 143°44' E), Queensland, H. and M. Midgley, June 1975; NTM S.12077-001, 107.5 mm HL, Roper River (14°46' S, 134°01' E), Northern Territory, H. and M. Midgley, September 1979; NTM S.l 1153-001, 110 mm HL, Mainoru River(14°0r S, 134°03’E), Northern Territory, H. and M. Midgley, September 1979. Diagnosis. Gill rakers total 10-11 on first gill arch, 11-14 on last arch. Eye diameter as percentage of head length 9-15. In addition, the characters of Arius midgleyi Kailola and Pierce: sleek body; strong jaws; jaws upturned slightly at symphyses, mouth broad; snout truncate in profile; head oblong, its width averaging 66% HL. Supraoccipital process narrow with parallel margins. Numerous fine, sharp teeth on palate in transverse band of four oblong groups. No rakers on posterior aspect of gill arches. A 16-19; vertebrae 20+7-8+28-29. Barbels thin and short, rarely reaching beyond pectoral fin base and less than 25% SL. Comparisons. Arius paucus differs from A. midgleyi in gill raker numbers (10-11 on first arch in A. paucus. versus 15-17 on first arch in A. midgleyi) and eye size (8.9-15.3% HL in A. paucus versus 12.9-21.8 % HL in A. midgleyi) (Table 5). No other morphometric characters distinguish these species, although juvenile colouration may be important. Furthermore, there is a clear disjunction in range of the two species: A. midgleyi distributed in the west (the Victoria, Katherine, Daly, Ord, remaining Kimberley and northern river systems including the Alligator); and A. paucus distributed in the east (the Roper and Flinders River systems and all other rivers draining into the Gulf of Carpentaria). Distribution. River systems of the Roper (Limmen Bight, Roper, Hodgson. Wilton and Mainoru), McArthur, Tooganginie Creek, Robertson and Calvert (rare); south¬ east to the Flinders, Gilbert, Staaten and Edward River systems (Strathgorden Lagoon) on Cape York Peninsula (Midgley 1979, 1980, 1981, 1982, 1983, 1984, pers. comm.). Remarks. Arius paucus and A. midgleyi are most closely related to Arius leptaspis (Bleeker) and less so, to Arius latirostris Macleay, a Papuan endemic. The distinguishing features of these fourtaxa were de.scribed Table 5. Comparison of key characters between Arius paucus sp. nov. and A. midgleyi. n = number; x = mean; SD = standard deviation; GR 1 = rakers on first gill arch; GR 4 = rakers on last gill arch. Specie.s variable Eye diam. % HI. Total GR 1 Total GR4 A. midgleyi holotype 13.3 16 18 range, n = 14 I2.9-2I.8 15-17 16-19 X 17.1 15.8 17.3 SD 3.2 0.7 0.9 A. paucus holotype 9.1 11 13 range, n = 16 8.9-15.3 10-11 11-14 X 11.0 10.7 12.8 SD 1.7 0.5 0.8 142 New catfish from Australia and New Guinea 22-1 20 18 c 0> o 0 ) E * 12 - O a> uj 10- 8 -* 100 200 SL(mm) 300 Fig. 13. Comparison of eye size between Arius midgleyi (hollow circles) and A. paucus sp. nov. (solid circles). by Kailola and Pierce (1988). Arius paucus and A. midgleyi are common and widespread in river systems of northern northern Australia, and attain a large size (to about 28 kg and 1.3 m total length; A. midgleyi) (Kailola 1990b). Etymology. From the Latin, paucus, meaning few, or less. Refers to the new species having fewer gill rakers and a smaller eye than does A. midgleyi. ACKNOWLEDGMENTS 1 am grateful to the many collectors and collection managers of institutions who have sent me material. 1 appreciate the more recent support and assistance of Helen Larson (NTM), Grant West, Alastair Graham and Peter Last (CSIRO). Kent Hortle (P.T. Freeport Indonesia, Papua (Irian Jaya)) enabled my access to west New Guinea material and facilitated transfer of material. I thank Hamar Midgley, Gerry Allen and Alan Haines for their insights. REFERENCES Allen, G.R. 1991. Field guide to the freshwater fishes of New Guinea. Christensen Research Institute Publication 9: Madang, Papua New Guinea. Allen, G.R. and Coates. D. 1990. An ichthyological survey of the Sepik River system, Papua New Guinea. Records of the Western Australian Museum Supplement 34: 31-116. Blaber, S.J.M., Brewer, D.T. and Salini, J.P. 1994. Diet and dentition in tropical ariid catfishes from Australia. Environmental Biology of Fishes 40: 159-174. Bleeker, P. 1846. Overzigt der Siluroieden, welke te Batavia voorkomen. Natuur-en Geneeskundig Archief voor Neerlands Indie 3(3): 135-184. Bleeker, P. 1858. De visschen van den Indischen Archipel beschreven en toegelicht. Vol. I. Siluri. Verhandelingen der Natuurkundige Vereeniging in Nederlandsch Indie 4. 1-370 p. Batavia, Lange & Co. (also published as: Ichthyologiae Archipelagi Indici Prodomus Vol.l, Siluri. Lange & Co.: Batavia). Coates, D. 1983. The biology of fork-tailed catfishes (Ariidae) from the Sepik River. Papua New Guinea Department of Primary Industry Fisheries Research Report 83-19. Coates, D. 1988. Length-dependent changes in egg size and fecundity in females, and brooded embryo size in males, of fork-tailed catfishes (Pi.sces: Ariidae) from the Sepik River, Papua New Guinea, with some implications for stock assessments. Journal of Fish Biology 33: 455-466. De Vis, C.W. 1884. Fishes from South Sea Islands. Proceedings of the Linnaean Society of New South Wales 8: 445-457. Haines, A.K. 1979. An ecological survey of fish of the lower Purari River system, Papua New Guinea. In: Purari River (Waho) Hydroelectric Scheme, Environmental Studies. Volume 6. Office of Environment and Conservation and Department of Minerals and Energy, Papua New Guinea. Pp 1-102. Herre, A.W.C.T. 1935. New fishes obtained by the Crane Pacific Expedition. Publications of the Field Museum of Natural History, 335, Zoology Series 18(12): 383-438. Kailola, P.J. 1975. A catalogue of the fish reference collection at the Kanudi Fisheries Research Laboratory, Port Moresby. Research Bulletin, Department of Agriculture, Stock and Fisheries, Port Moresby 16: 1-277. Kailola, P.J. 1983. Arius graeffei and Arius armiger, valid names for two common species of Australo-Papuan fork-tailed catfishes (Pisces: Ariidae). Transactions of the Royal Society of South Australia 107(3): 187-196. Kailola, P.J. 1990a. A review of the freshwater fork-tailed catfishes (Pisces: Ariidae) of northern New Guinea with description of two new species. Records of the Western Australian Museum Supplement 4: 1-30. Kailola, P.J. 1990b. The catfish family Ariidae (Teleostei) in New Guinea and Australia: relationships, systematics and 143 P.J. Kailola zoogeography. Unpublished PhD thesis. University of Adelaide. Kailola, P.J. and Pierce, B.J. 1988. A new freshwater catfish (Pisces: Ariidae) from northern Australia. Records of the Western Australian Museum 14(1): 73-89. Kailola, P.J. and Wilson, M.A. 1978. Trawl fishes of the Gulf of Papua. Papua New Guinea Fisheries Research Bulletin 20. Kner, R. and Steindachner, F. 1866. Neue fische aus dent Museum der Herren Joh. C. Godeffroy und Son in Hamburg. Sitzungsherichteder Mathematisch-Naturwissenschaftlichen classe der Kaiserlichen Akademie der Wissenschaften, Wien 54( 1): 356-395. Leviton, A.E., Gibbs Jr, R.H., Heal, E. and Dawson. C.E. 1985. Standards in herpetology and ichthyology: part 1. Standard symbolic codes for institutional resource collections in herpetology and ichthyology. Copeia 1985(3): 802-832. Liem, D.S. and Haines, A.K. 1977. The economic significance and economic importance of the mangrove and e.stuarine communities of the Gulf Province, Papua New Guinea. In: Purari River (Waboj Hydroelectric Scheme Environmental Studies. Volume 3, pp 1-35. Office of Environment and Conservation and Department of Minerals and Energy, Papua New Guinea. Maunsell and Partners 1982. Working paper 16: Biological and heavy metal reconnaisance survey of the Ok Tedi - Fly River system. In: Ok Tedi Environmental Study. Volume 6: Ok Tedi - Fly River Acpiatic Survey: Heavy Metal review. Pp.57-289. Midgley, S.H. 1979. The Roper River system. Limmen Bight River system, Rosie Creek system in the Northern Territory. A biological resource study of fresh waters conducted during August-September 1979. Report to the Fisheries Division of the Northern Territory Department of Primary Production. Midgley, S.H. 1980. The Daly River and its principle tributaries Katherine River, King River, Dry River, Flora River, Fergusson River, Fish River. Douglas River in the Northern Territory. A biological resource study of fresh waters conducted during August-September 1980. Report for the Fisheries Division of the Northern Territory Department of Primary Production. Midgley, S.H. 1981. The Victoria River and its tributaries Angalarri River and its tributary the Ikymbon River. Armstrong River, Camfield River, Wickham River and its tributary the Humbert River, West Baines River and its tributary Waterloo Creek, East Baines River, Neave Creek and Jasper Creek. The Fitzmaurice River. The Keep River and its tributary Chinaman Creek in the Northern Territory. A biological resource study of fresh waters conducted during August-September 1981. Report for the Fisheries Division of the Northern Territory Department of Primary Production. Midgley, S.H. 1982. The Lake Eyre drainage. The James River and Lome Creek. The inland drainage. The Playford River, Brunette Creek, Corella Creek. Creswell Creek and Newcastle Creek. The Gulf of Carpentaria drainage. The McArthur River, Wearyan River, Foelsche River, Robinson River, Calvert River and Nicholson River systems in the Northern Territory. A biological resource study of fresh waters conducted during August to October 1982. Report for the Northern Territory Fisheries and Industrial Research Development Trust Fund. Midgley, S.H. 1983. The rivers of Arnhem Land. The Phelp River, Rose River, Walker River, Koolatong River, Giddy River, Goyder River, Wilton River, Mainoru River, Cadell River, Liverpool River and Goomadeer River, and the Towns River immediately to the south-east of Arnhem Land in the Northern Territory. An account of a biological resources study of fresh waters conducted during July to August 1983. Report of a private study by Midgley and Midgley, with the assistance of the Northern Territory Fisheries Division of the Northern Territory Department of Primary Production. Midgley, S.H. 1984. The Reynolds River. Finniss River, Adelaide River, McKinley River, Mary River, South Alligator River and Karns Creek in the Northern Territory. A biological resources study of fresh waters conducted during August, September 1984. Report to the Northern Territory Fisheries and Industrial Research Development Trust Fund. Ogilby, J.D. 1898. New genera and species of fishes. Proceedings of the Linnaean Society of New South Wales 23(3): 280- 299. Ramsay, E.P. and Ogilby, J.D. 1886. A contribution to a knowledge of the fish fauna of New Guinea. Proceedings of the Linnaean Society of New South Wales, ser.2, 1(1): 8 - 20 . Roberts, T.R. 1978. An ichthyological survey of the Fly River in Papua New Guinea with descriptions of new species. Smithsonian Contributions to Zoology 281: 1-72. Riippell, W.P.E. 1835-38. Neue Wirbelthiere zu der Fauna von Abyssinien gehorig. Fische des Rothen Meeres. Frankfurt. 148 p.. 33 pis. Tortonese, E. 1964. Contribute alio studio sistematico e biogeografico dei pesci della Nuova Guinea. Annals Museo Civico Storia Naturel Genova 75: 13-98. Valenciennes, A. 1840a. pp i-xx, 1-344. In: Cuvier, G. and Valenciennes, A. Histoire naturelle des poissons, 14. F.G. Levrault, Paris. Valenciennes, A. 1840b. pp i-xxiv, 1-397. In: Cuvier, G. and Valenciennes, A. Histoire naturelle des poissons, 15. F.G. Levrault, Paris. Weber. M. 1908. Sus.swasserfische aus Niederlandish Sud- und Nord-New Guinea. Nova Guinea. Resultants de I 'Expedition Scientifique Neerlandaise a la Nouvelle-Guinee en 1907 et 1909 sous les auspices de Dr H.A. Lorents 5 (2): 201- 267. Leiden: E.J. Brill. Weber, M. 1913. Die Fische der Siboga E.xpedition. Leyden. Weber, M. and de Beaufort, L.F. 1913. The fishes of the Indo- Australian Archipelago, 2. Leiden: E.J. Brill. Whitley. G.P. 1941. Ichthyological notes and illustrations. Australian Zoologist 10(1): 1-50. Accepted 2 November 2000 144 The Beagle, Records of the Museums and Art Galleries of the Northern Territory, 2000 16: 145-162 Two new genera and three new species of Thylacinidae (Marsupialia) from the Miocene of the Northern Territory, Australia PETER MURRAY' AND DIRK MEGIRIAN^ 'Muaeum of Central Australia, PO Box 2108, Alice Springs, NT0871, AUSTRALIA peter, mu tray @ nt.gov. au ^Museum and Art Gallery of the Northern Territory, GPO Box 4646, Darwin, NT0801, AUSTRALIA dirk.megirian@nt.gov.au ABSTRACT Three new species and two new genera of Thylacinidae from the Northern Territory of Australia bring the total number of known mid and late Tertiary species to 11 in eight genera. Tyarrpeciniis rothi gen. et sp. nov. from the Alcoota Local Fauna (Waite Formation), and Ninibacinus richi sp. nov. and Mutpiiracimis archibaldi gen. et sp. nov. from the Bullock Creek Local Fauna (Camfield Beds), are relatively small and amongst the more plesiomorphic members of the family. Phytogeny reconstruction using cladistic methods and biochronological data indicate that a major radiation occurred within the family in pre-Miocene times. Specialisation in the form of dental carnassiali.sation appears to have evolved in parallel in at least two crown groups, one of which includes the recently extinct ‘Tasmanian wolf, Thylacinus cynocephalus. The other, together with all the lineages represented by unspecialised species, apparently did not survive into the late Miocene. However, insufficient data are available to show what the regional, let alone continental, pattern of succession was during the Miocene. Keywords. Marsupialia, Thylacinidae, Miilpuraciniis archibaldi gen. et sp. nov., Tyarrpecinus rothi gen. et sp. nov., Nimbacinus richi sp. nov., Alcoota Local Fauna, Bullock Creek Local Fauna, Camfield Beds, Waite Formation, phylogeny, evolution, Miocene, Australia. INTRODUCTION The Thylacinidae is a family of dasyuromorphian marsupials, the last member of which, the Tasmanian wolf {Thylacinus cynocephalus (Harris)), was extin¬ guished by bounty hunters in the 1930s. Pliocene and Pleistocene fossil .species have all been synonymised with T. cynocephalus (Dawson 1982), and the fossil record of the family was only extended into the late Miocene with the discovery at Alcoota of Thylacinus potens Woodburne, 1967. Thylacinus potens provided the first indication that thylacinids were a radiated family, as the species is derived dentally in ways that precludes its direct ancestry of T. cynocephalus. In more recent years, the fossil record has been greatly expanded by new discoveries at Alcoota, Bullock Creek and Riversleigh (Fig. 1), spanning the ?late Oligocene to late Miocene (Muirhead and Archer 1990; Wroe 1996; Muirhead 1997; Murray 1997; Muirhead and Wroe 1998). On the basis of the known diversity of thylacinids in the central and northern Australian deposits, it might be anticipated that similar diversity will eventually be found in correlative deposits in the Lake Eyre and Tarkarooloo Basins of South Australia. So far, only a single premolar of unknown affinity has been listed from Lake Ngapakaldi, South Australia (Kutjamarpu Local Fauna. Wipijiri Formation of the Lake Eyre Basin) (Archer 1982). Hypotheses of phylogeny within the family are currently in a state of flux as each new discovery adds to the diversity of fossil forms available for analysis. Tyarrpecinus rothi gen. et sp. nov. (Alcoota LF, late Miocene), Mutpuracinus archibaldi gen. et sp. nov. and Nimbacinus richi sp. nov. (both Bullock Creek LF, mid Miocene) represent three new species and two new genera of Thylacinidae, and raise the total number of known Tertiary species to 11 in eight genera. The new species are comparatively small members of the family (small and medium-dog sized, compared to the wolf¬ sized T. cynocephalus, for example). Cladistic analysis of dental characters resolves them as plesiomorphic members of the family, closely related to certain previously described forms (Muirhead 1992, 1997; Wroe 1996; Muirhead and Wroe 1998). Excellent preservation of the maxilla of Mutpuracinus and a complete dentary of Nimbacinus richi sp. nov. fill some important gaps in 145 Murray and Megirian Fig. 1. ?Late Oligocene and Miocene localities that have produced thylacines. our knowledge of Tertiary thylacinid morphology. The three new species and other recently reported late Oligocene and Miocene finds exemplify a gradual structural succession towards the genus Thylacinus. The current consensus is that thylacinids are dasyuromorphians derived from a dasyurid ancestor, although a possible affinity with South American borhyaenoids was a matter of prolonged debate from 1903 into the 1980s (e.g. references in Archer 1982; Muirhead and Wroe 1998). We accept argument presented by Wroe (1996) and Muirhead and Wroe (1998) for the inclusion of Muribacinus gadiyuli Wroe, 1996, and Badjcimis tumhulli Muiihead and Wroe, 1998, in the Thylacinidae (rather than in the Dasyuridae), and offer no additional observations on the issue regarding the thylacinid status of the similarly plesiomorphic Miitpuracinus archibaldi gen. et sp. nov. Other than to remark on the assumption that the metaconid reduces uniformly along the tooth row in thylacinids, we support the re-diagnosis of the family presented in Muirhead and Wroe (1998). METHODS Adult dental terminology (Flower-Luckett) recognises PI-3, Ml-4 (Flower 1869, Luckett 1993); all Archer (1978, 1982) terminology in cited literature has been converted. Specimens are Northern Territory Museum (NTM) Palaeontological collection unless designated QMF (Queensland Museum, Fossil collection). The thylacinid-producing formations referred to are not precisely dated, and their ages can at present only be loosely expressed in terms of the standard geological timescale. Con.sequently, modifiers such as ‘early’ and ‘late’ are left uncapitalised when attached to the standard Tertiary epochs so as not to imply a geochronological precision that does not exist. Stage-of-evolution biochronological concepts and terminology follow Megirian (1994). Key to anatomical abbreviations. A-E Stylar cusps AC Anterior cingulum or precingulum ALV Alveolus ANP Angular process CCR Centrocrista CON Condyle CRN Reception notch for canine END Entoconid HLD Hypoconulid HYD Hypoconid IFO Incisive foramen lOF Infraorbital foramen MAE Mandibular foramen MCE Metaconule MCR Submasseteric crest MEC Metacone MED Metaconid MEF Mental foramen MFO Masseteric fossa MJS Maxillo-jugal suture PAC Paracone PAD Paraconid PAF Palatal fenestra PCD Precingulid PCL Protoconule PMF Posterior mental foramen PMS Premaxillo-maxillary suture PRC Protocone PRD Protoconid PSD Parastylid SYSTEMATIC PALAEONTOLOGY Order Dasyuromorpbia Gill, 1872 Superfamily Dasyuroidea Goldfuss, 1820 Family Thylacinidae Bonaparte, 1838 Miitpuracinus gen. nov. Type species. Miitpuracinus archibaldi sp. nov. by monotypy. Diagnosis. Small thylacinid with strong expression of stylar cusps on M''^; large, posteriorly directed P^; infraorbital foramen not in contact with jugal and situated over posterior root of M'. Size similar to holotype of Muribacinus gadiyuli (QMF30386); lower position of zygomatic root, more anterior position of infraorbital foramen relative to the teeth (i.e. over anterior root of M'); more posteriorly-directed premolar crowns; larger posterolingual cuspule on P’; M^ narrower relative to length; larger stylar cusp D on M'"^ and presence of stylar 146 Fossil thylacines from the Northern Territory cusp C and E on M'’^; paracone more reduced relative to metacone on M'"^; M 4 metaconid more reduced; narrower cingulids and talonids on lower molars. Differs from Badjcimis turnhuUi in which the jugal closely approaches the infraorbital foramen opening above P'; the upper premolar crowns are more slender, taller; M' lacks stylar cusps on anterior lobe, preparacrista parallel to long axis of tooth row and with minute stylar cusp D. Differs from Nimhacinus dicksoni Muirhead and Archer, 1990, which is significantly larger; the infraorbital foramen is within the jugal and situated over anterior root of M^; in which conules are more strongly expressed and distinctly longer than Differs from Wabulacinus ridei Muirhead, 1997, Ngamalacinus timmulvaneyi Muirhead, 1997, and species of Thylacinus in showing less reduction of metaconids on M 2 - 4 ; absence of carnassial notch in cristid obliqua and less reduetion of entoconid and protocone. Etymology. Miitpura is the tribal designation of the Aboriginal people living in the Camfield district [= Tjambutjamhulani - Tindale 1974] + kynos (Gr.) ‘dog’. Mutpuracinus archihaldi sp. nov. (Figs 2-4, Table I) Diagnosis. As for genus. Type material. HOLOTYPE - NTM P907-3 (Blast Site), left maxilla with P--M'‘, missing canine and P'. PARATYPE - NTM P9612-5 (Top Site), left dentary fragment with M 3 . 4 , missing ascending ramus and anterior part of horizontal ramus. Referred material. All NTM: P9464-120 (Top Site), right premaxilla with I'"' alveoli; P9464-119 (Top Site) posterior fragment of horizontal ramus of right dentary preserving submasseteric crest, part of condyle and part of angular process; P87108-10 (Blast Site), isolated right M4. Type locality. ‘Blast Site’, Bullock Creek, Northern Territory, Australia. Low limestone hill 1 km east of type section of Camfield Beds; ISDSl’E, 17° 07’S; Wave Hill 1:250 000 Map Sheet, SE/52-8. Lithic unit, fauna and age. Camfield Beds, Bullock Creek Local Fauna, middle Miocene on the basis of marsupial stage-of-evolution biochronology (Woodburne et al. 1985; Murray et al. 2000). Both fossil quarries (sites) listed above are in the vicinity of the type locality (Murray and Megirian 1992: fig. 2). Description. Premaxilla (Fig. 2A,B). Edentulous, 13.0 mm anteroposteriorly by 11.5 mm dorsoventrally, right premaxillary fragment P9464-I20 is missing the ascending process. Four incisor alveoli are present: I' is largest; are narrower; and 1 “ is slightly smaller than I' but larger than P'^. The incisor arcade is nearly straight and less acutely angled relative to the midline than in Dasyurus hallucatus (Dasyuridae), for example. A large reception socket for the lower canine measures 4.5 mm anteroposteriorly by 6.5 mm dorsoventrally. The anterior margin of the right incisive foramen indicates a narrow, slot-like structure. Maxilla (Fig. 3A,B). P907-3 preserves alveoli for upper canine and P' and remainder of left cheek tooth Table 1. Measurements (mm) of cheek teeth. L= length; W = width; AW = anterior width; PW = posterior width. Mutpuracinus archihaldi gen, et sp. nov.lower cheek teeth Pi P2 Pi Ml M2 M., M4 L W L PW L AW PW L AW PW L AW PW L AW PW L AW PW P9612-5 - 5.7 3.6 2.9 5.9 3.3 2.3 P87108-10 - - - 5.3 2.9 2.2 upper cheek teeth P' P^ P’ m' M^ M^ M^ L W L PW L AW PW L AW PW L AW PW L AW PW L AW PW P907-3 - - 4.5 1.8 5.5 2.4 3.0 5.8 4.1 6.2 5.7 5.0 6.5 5.4 5.7 7.0 6.0 4.7 3.5 Nimbacinus richi sp. nov. lower cheek teeth Pi P2 Pi Ml M2 M., M4 L w L PW L AW PW L AW PW L AW PW L AW PW L AW PW P%12-4 5.0 2.0 6.1 2.8 7.2 2.1 3.1 7.0 3.3 4.2 8.3 4.5 4.4 8.2 5.0 4.0 8.7 4.2 2.5 P8695-92 - - 5.8 2.8 - - - 6.2 2.9 3.2 - - - - 4.3 3.6 7.2 - 3.0 P904-7 - 7.4 4.1 4.3 8.0 4.5 4.1 7.9 4.1 2.9 P85553-3 4.5 1.7 6.0 2.8 - - - 6.9 - 4.0 - - - - - - - - - Tyarrpecinus rothi gen. et sp, . nov. upper cheek teeth P' p* P’ m' M^ M^ M^ L W L PW L AW PW L AW PW L AW PW L AW PW L AW PW P9821I - - 6.5 2.5 - - - 4.8 7.0 7.3 - - - 7.7 7.9 9.8 - - - Murray and Megirian Fig. 2. Mutpuracinus archibaldi gen. et sp. nov. Referred right preinaxilla (P9464-120) in A, lateral, and B, occlusal views. Paratype left dentary fragment (P9612-5) in C, lateral view; D, medial, E, lateral and F, occlusal, views of molars M3.4, Scale bars in mm. 148 Fossil thylacines from the Northern Territory PAF Fig. 3. Mutpuracinus archihaldi gen. ct .sp. nov. holotype left maxilla (P907-3) in: A. lateral view; B, occlusal view; and C, detailed occlusal view of molars M'“'. Scale bars in mm. row; premaxillary contact preserved anterodorsally and nasal contact preserved dorsally; posterior (lacrimal) process missing; jugal process intact. Canine alveolus planoconvex, about 3.5 mm wide by 5.5 mm antero- posteriorly. Alveolus for P' situated immediately posterior to the canine alveolus, about 3.5 mm long, oriented slightly obliquely to succeeding premolars. A diastema 4.0 mm long separates P' from P^. Crown of P- 4.6 mm long, separated from P^ by narrow, but distinct, slot that, while only 0.7 mm wide, is suggestive of rudimentary diastema. P^ much larger than P‘ with crown projecting well below occlusal line of the other cheek teeth. Labial margin of molar row gently arcuate. Interdental embrasures present between M'becoming increasingly larger and distinct posteriorly. Anterior margin of posterior palatal fenestra preserved at the level of PVM'. Palatal processes thin posteriorly, thickening anteriorly in front of fenestral margin towards PL Jugal contact a shallow, crescentic depression extending to within about 2.5 mm of infraorbital foramen dorsally 149 Murray and Megirian Fig. 4. Comparison of maxillae of A, Mutpuracinus archihaldi gen. et sp. nov. and B, Muribacinus gadiyuli drawn to the same scale, showing differences in the position of the infraorbital foramen relative to M'. (B reversed: after Wroe 1996: fig. I.) and about 4.0 mm ventrally. Infraorbital foramen large, about 5.0 mm vertically and 2.5 mm mediolaterally (Fig. 4), opening behind well-defined circular infraorbital fossa, about 12 mm in diameter. The posterior margin of the foramen is aligned with the distal root of M'. A shallow depression extending anterodorsally from infraorbital fossa is bounded dorsally by very distinct, sinuous groove terminating about 12 mm above level of canine alveolus. External surface of the canine alveolus bulges conspicuously outwards from surrounding contours of maxilla, resulting in a wide concavity in the vertical profile of specimen. Lateral surface of alveolar bulge rugose, with sharply defined posterior crest. Incisors, Alveoli only (Fig. 2A,B): I' largest, broadly oval, 2.0 mm wide, 2.1 mm anteroposteriorly; F triangular, 1.5 mm wide, 1.8 mm long, slightly wider than P; P narrow, slot-like; 1.3 mm wide, 2.1 mm long; P, ovo-triangular, 1.8 mm wide, 2.0 mm antero¬ posteriorly. Canine. Alveolus only (Fig. 2A,B), 5.5 mm antero¬ posteriorly; 3.5 mm mediolaterally; planoconvex shape; root socket 14.0 mm deep, angled -30° to palatal plane. Premolar, P' (Fig. 3A,B). Alveolus only, 3.5 mm anteroposteriorly, anterior root socket 1.3 mm diameter; angled about 12° from plane of P^’^. (Fig. 3A,B). Crown 4.5 mm long, 1.8 mm maximum width; low anterior basal crest, concave anterior margin, tip directed posteriorly; slight posterolingual expansion and small posterobasal cuspule; posterior root larger than anterior. (Fig. 3A,B). Much larger than P^, 5.5 mm anteroposteriorly by 3.0 mm posterior width; low anterobasal crest; concave, rounded anterior margin, tip directed posteriorly; convex, crested posterior margin; low, well-developed posterolingual cuspule developed on thick basal cingulum, terminating in elevated postero¬ median cuspule; posterolabial cingulum poorly developed. Molar, M' (Fig. 3C). Paracone situated much more labially than metacone, very short transverse preparacrista; centrocrista nearly parallel to axis of tooth row; metacone large, premetacrista nearly transverse. Talon broad antero¬ posteriorly and short transversely; small metaconule and protoconule present; protocone large. Stylarcusp B closely adpressed to paracone; higher but slightly smaller than the latter; vestigial stylar cusp C situated at base of stylar cusp D, a large conical cuspule occupying about half the posterior lobe of the crown. Well-developed ectoflexus present, buccal occlusal profile of crown distinctly lobate. Low swelling in the position of stylar cusp E (heavily worn); weak precingulum present with shallow inter- proximal notch for PL M- (Fig. 3C). Larger, more equi-triangular crown; paracone extends much further lingually than in M'; short, transverse preparacrista extends to stylar cusp B; stylar cusp C well-developed, about equal in size to B; stylar cusp D large, conical, occupies slightly less than half of metastylar lobe of crown, slightly smaller but nearly equal in height to metacone; small, distinct stylar cusp E situated midway between D and metastylar tip. Metacone much larger than paracone. Talon broad, but more V-shaped than in M', conules small, protocone narrow. Precingulum well-developed with notch for metastyle of M'. M^(Fig. 3C). Larger than M^ in occlusal aspect, significantly wider transversely, equal in length labially and slightly longer posterolingually; stylar cusp B reduced; stylar cusp C well-developed but low; stylar cusp D large, triangular, but considerably reduced in height and transverse width; tiny cuspule present in position of stylar cusp E. Ectoflexus strong, parastylar and metastylar tips slender, lobate. Preparacrista long, straight, transverse; paracone less reduced relative to metacone; postmetacrista longer than in M^. Talon narrow, distinctly V-shaped, small conules present, more distinct than in M-; protocone reduced. Precingulum longer than in M^ with V-shaped accommodation notch for M’ metastyle. M''(Fig. 3C). Tip of parastyle situated lingual to tip of M’ metastyle; preparacrista slightly convex, about equal in length to M^; metacone greatly reduced, Protoconule better developed than on other molars; metaconule absent; talon narrow; preprotocrista weak. A tiny stylarcusp is situated midway between parastylar tip and base of metacone. Meristic gradients, M' "' (Fig. 3C). Preparacrista: increases posteriorly, M^"'* about equal; postmetacrista: Fossil thylacines from the Northern Territory increases posteriorly except M'' where absent; paracone height: increases posteriorly M''^, M"* equals or slightly lower than M^; metacone height: increases posteriorly, except where reduced; protocone height: decreases posteriorly; talon surface: increases posteriorly M'^ decreases posteriorly stylar cusp B size: decreases M'■*; stylar cusp C: increases M'"^, absent M"*; stylar cusp D: decreases M' \ absent M''; stylar cusp E: decreases M''^, possibly absent M'*. Dentary (Fig. 2C). Medial surface flat, digastric fossa faint, shallow; subalveolar fossa, lateral crest well- defined. Masseteric fossa deep with prominent shelf-like submasseteric crest. Depth of horizontal ramus below M 4 14.0 mm; thickness of inferior border at same level, 5.5 mm. A/j (Fig. 2C-F). Paraconid and protoconid bulky; metacristid nearly perpendicular to long axis of crown; talonid distinctly narrower than trigonid; metaconid large, higher than paraconid, bulky at base. Entoconid small but distinct; as high as hypoconulid; hypoconulid merges labially with postcingulid. Hypoconid lowest cusp on crown; pre- and postcingulids present but not strongly developed. Cusp heights: protoconid> metaconid> paraconid> entoconid= hypoconulid> hypoconid. M^fFig. 2C-F). Cusps of trigonid more slender, talonid greatly reduced compared to Mj; metaconid and paraconid similar in height to M 3 but more slender. Entoconid reduced to thickening of preentocristid, lower than hypoconulid; hypoconulid high pointed cusp, continuous labially with postcingulid. Pre- and postcingulids not strongly developed. Cusp heights: protoconid> metaconid> paraconid> hypoconulid> entoconid> hypoconid. Meristic gradients M3.4 (Fig. 2C-F). protoconid, paraconid and metaconid decrease in bulk posteriorly; talonid decreases in width posteriorly; entoconid decreases in size posteriorly; angle of paracristid to long axis of crown more acute posteriorly. Etymology. Named after Ian Archibald in honour of his contributions to the natural hi.story of the Northern Territory. Niinbacinus Muirhead and Archer, 1990 Type species. Niinbacinus dicksoni Muirhead and Archer, 1990, by original designation. Niinbacinus riclii sp. nov. (Fig. 5, Table 1) Niinbacinus dicksoni - Muirhead and Archer 1990: fig. 3 (in part). Diagnosis. Medium-sized thylacinid with narrow diastemata between premolars, P 3 much larger than P 2 ; lower molars with well-developed metaconids and entoconids. Differs from Niinbacinus dicksoni in possessing well-developed metaconids on M 2-4 in conjunction with reduced metaconid on Mi and large conical entoconids on M 1 . 3 ; differs from Ngamalacinus timinulvaneyi and Badjcinus turnbulli in lacking a camassial notch in the hypocristid, and from Wabulacinus ridei Muirhead, 1997, and species of Thylacinus in having large metaconids. Type material. HOLOTYPE - NTM P9612-4 (Top Site), right dentary with P1-M4; canine, incisors missing. PARATYPE - NTM P85553-3 (unrecorded quarry), fragment of right dentary with P 1.2 and M| (also a paratype of N. dicksoni Muirhead and Archer, 1990). Type locality. ‘Top Site’, Bullock Creek, Northern Territory (additional data as for Mutpuracinus archibaldi). Referred material. NTM P8695-92 (Blast Site), horizontal ramus of left dentary retaining P 2 , Mi, M 3.4 (M 3 lacking paraconid); P904-7 (Top Site), left dentary fragment retaining M 2 - 4 . Lithic unit, fauna and age. Camfield Beds, Bullock Creek Local Fauna, mid Miocene (additional information as for Mutpuracinus archibaldi). Description. Dentary (Fig. 5A,B). P9612-4 is about 25% smaller than Thylacinus inacknessi Muirhead, 1992, but relatively stouter anteriorly and more bowed along inferior border; anterior margin of ascending ramus more erect, condylar notch wider, more invasive. Diastema between P 2 -P 3 very short, approximately 1.4 mm; large mental foramen under posterior root of Pi, small posterior mental foramen under posterior root of Mi. Masseteric crest well developed, flange-like; masseteric fossa relatively shallower and condyle; condyle transversely elongate (14.0 mm wide), slightly flattened dorsally. Mandibular foramen situated 22.0 mm from distal edge of condyle; depth of ramus below M 4 is 15.6 mm; thickness of inferior border at same level is 7.0 mm. Incisors, I/.jiFig. 5A,B). Three incisor alveoli present anteromedial to canine alveolus, arranged in a triangle, I 3 ventrolaterally, I 2 lingually and L above I 2 but aligned vertically with LTs socket smallest (1.3 mm x 1.5 mm) followed by I 2 (1-3 mm x 2.5 mm) with L largest (1.5 mm X 3.0mm); lateral width of all three sockets about 3 mm, vertical extent about 4.8 mm. Canine (Fig. 5A,B). Alveolus only (7.4 mm vertical, 4.0 mm mediolateral) oval socket, anterolaterally directed. Preinolars, P1.3 (Fig. 5A,B). P| oriented obliquely (approximately 15°) to long axis of other premolars, much smaller than P 2 (5.0 mm long, 2.0 mm wide) with low, recurved protoconid and very elongated posterior shelf lacking distinct cuspule. Protoconid of P 2 much higher, longer than Pi; small anterobasal and posterobasal cuspules present (6.1 mm long, 2.8 mm wide); P 3 , anterobasal cuspule more defined, posterior shelf more talonid-like and elevated distally with small postero- lingual and large posterobasal cuspules (7.2 mm long, 3.1 mm wide). Protoconids of all premolars directed posteriorly. Murray and Megirian Fig. 5. Nimbacinus richi sp. nov. holotype right dentary (P9612-4) in: A, lateral view; B, occlusal view; C, molars M 1-4 in occlusal view; and D, molars Mm in lateral view. 152 Fossil thylacines from the Northern Territory Molar, M/(Fig. 5A-D). Subrectangular crown, rather truncated anteriorly and markedly so posteriorly; talonid wider than trigonid; short paracristid slightly angled lingually, paraconid low, broad, conical with short preparacristid terminating in small parastylid; protoconid central, crown shallowly waisted on either side; posterolingual corner of talonid forming right angle; entoconid prominent, posteriorly-directed, conical cusp about 1.5 mm in diameter, separated from base of metaconid by deep, transverse groove; entoconid much larger and higher than hypoconulid; hypoconid large, triangular, lowest cusp on talonid. Small metaconid represented by low, broad triangular bulge on postero¬ lingual side of protoconid. Pre- and postcingulids thick, well developed anterolabially, terminating above middle of anterior root; postcingulid ascends to just below hypoconulid. Cusp heights; protoconid > metaconid > entoconid > paraconid > hypoconulid = hypoconid. MpCFig. 5A-D). Trigonid somewhat compressed, much longer and slightly wider than talonid; paracristid slightly less obtuse than in Mi; metaconid much larger and more differentiated from protoconid; small carnassial notch present in metacristid; entoconid slightly larger and higher than in Mi; extending high above smaller, flat, triangular hypoconulid and closely adpressed to parastylid of M 3 . A short preentocristid developed. Hypoconulid extending distolabially a short distance beyond the entoconid where it is accommodated by distinct C-shaped notch between precingulid and parastylid. Hypoconid a broad, triangular cusp much lower than hypoconulid. Crislid obliqua ascending base of protoconid for short distance. Cusp heights: protoconid > metaconid > paraconid > entoconid > hypoconulid > hypoconid. Afj(Fig. 5A-D). Protoconid more slender than in M 2 , paraconid higher, slightly more erect, metaconid also higher, more differentiated, but more slender than in preceding molar; talonid shorter and narrower relative to trigonid than in M 2 . Entoconid reduced to small spur¬ like cusp on anterolingual margin of enlarged, triangular hypoconulid; hypoconulid projects distally into accommodation notch in M.i. Postcingulid reaches labial side of hypoconulid; hypoconid situated much lower than in M 2 and considerably reduced; cristid obliqua ascends posterior face of protoconid about halfway up to carnassial notch. Cusp heights: protoconid > metaconid > paraconid > hypoconulid > entoconid > hypoconid. A/.i(Fig. 5A-D). Paraconid directed slightly more to lingual side than in M 3 ; para- and metacristids blade¬ like, with deep carnassial notches; metaconid greatly reduced though fully differentiated from protoconid; talonid width about half that of preceding molar; entoconid reduced to thickening of the preentocristid, hypoconulid formed as in preceding molar, but smaller, succeeded anteriorly by strong posthypocristid; po.stcingulid vestigial. Hypoconid small, situated low on posterolabial margin of talonid basin; talonid basin open labially; cristid obliqua strongly developed, ascends posterior face of protoconid to base of carnassial notch in metacristid. Cusp heights: protoconid > paraconid > metaconid > hypoconulid > entoconid > hypoconid. Meristic gradients, M 1.4 {Fig. 5A-D). Protoconid decreases in bulk posteriorly; paraconid increases in height and obliquity to long axis of crown posteriorly; metaconid increases markedly in bulk from Mi to M 2 then gradually decreases; metaconid gradually increases in height to M 3 then decreases in M^; entoconid increases in size and height to M 2 then decreases M 3 to M 4 ; hypoconulid increases in height po.steriorly; hypoconulid increases in size posteriorly to M 3 then decreases in M 4 ; talonid increases in width from M, to M 2 then decreases from M3 to M4. Remarks. Association of referred specimens of lower molars with the uppers in Nimhacimis dicksoni was problematic (Muirhead and Archer 1990). Muirhead and Archer (1990) note several differences between the hololype Mi (QMF16802, ‘Henk’s Hollow’, Carl Creek Limestone) and the Bullock Creek Local Fauna specimen (P85553-3). The latter specimen shows a stronger entocristid and a more angular posterolingual corner, resulting in greater posterior crown width and a more rounded contour of the anterolingual surface. Although very worn, the entoconid of P85553-3 (difficult to see in their illustration) is somewhat larger and more distinct than that of the holotype. The same differences are more emphasised in less-worn P9612-4 in which a large, conical entoconid is present. These distinctions appear to be consistent and therefore not readily attributable to individual variation. More important however, is the status of paratype QMF16809 (Site D locality), identified as a right M 2 (their M 3 in Archer (1978, 1982) terminology). If this specimen, that has a greatly reduced metaconid, represents N. dicksoni (as they argue at some length), then P96I2-4, which has a large metaconid on M 2 , could not represent this species. While a species- level distinction could be made on the basis of this technicality alone, the subsequent discovery of several similar-sized thylacinid species increases the probability that QMFI6809 might not represent N. dicksoni, and therefore a systematic distinction made on this basis could be phylogenetically misleading. While the differences in M| morphology alone are sufficient to indicate the existence of another species of the genus Nimhacimis, this proposed revision leaves the states of the M 2.4 metaconids of N. dicksoni in doubt. Etymology. Named in honour of Thomas Rich for his many important contributions to vertebrate pa¬ laeontology in Australia. Tom introduced us to the Bullock Creek Local Fauna of the Northern Territory in 1984. 153 Murray and Megirian Tyarrpecinus gen. nov. Type species. Tyarrpecinus rothi sp. nov. by monotypy. Diagnosis. Medium-sized thylacinid with transversely narrow, elongate M' and strong ectoflex and elongation of metastylar spur on M^; differs from Muribacinus gadiyuli and Mutpuracinus archihaldi in larger size, narrower M', more elongate and slender metastylar wing; wider angle between preparacrista and postmetacrista; more closely approximated paracone and metacone on differs from Badjcinus turnbulli in form of premolar crown; in possessing stylar cusps and preparacrista on the anterior lobe of M'; more reduced stylar cusp B and relatively larger stylar cusp D on M’; differs from Nimbacinus dicksoni and Ngamalacinus timmulvaneyi in having a more slender, elongated M' with straighter centrocrista and more reduced parastylar spur; metastylar spur more elongate; narrower trigone basin, metacone larger relative to paracone; conules where present, extremely reduced; differs from Wabulacinus ridei in more transverse orientation of preparacrista and retention of stylar cusps B and D on M'; differs from Thylacinus macknessi in reduction of precingulum but retention of stylar shelf, otherwise showing conservative Thylacinus-Vikc apomorphies. Etymology. Eastern Arrente dialect tyarrpa ‘cracked’ -I- Kynos (Gr.) ‘dog’. Tyarrpa, pronounced char-puh, (Henderson and Dobson 1994), is in reference to the state of preservation of the type material. Tyarrpecinus rothi sp. nov. (Fig. 6, Table 1) Diagnosis. As for genus. Type material. HOLOTYPE - NTM P98211 (Main Pit), fragment of left maxilla with P\ M''^'; M' and M’ complete but assembled from fragments; M^ and M‘' represented by talons only; approximately 75 small fragments consisting of isolated roots and bone and enamel fragments. Type locality. Alcoota Scientific Reserve, Northern Territory, Australia. 22°52' S, 134‘’52' E; Alcoota 1:250,00() Map Sheet, SF/53-10. Lithic unit, fauna and age. Waite Formation, Alcoota Local Fauna, late Miocene on the basis of marsupial stage-of-evolution biochronology (Woodburne et al. 1985; Murray et al. 1993, Murray et al. 2000). Description. The Tyarrpecinus rothi sp. nov. holotype, P9821I, was reconstituted from a con¬ centration of small fragments of bone and teeth that may represent the contents of a crocodilian coprolite. Many of the fragments show chemical erosion and are coated with a layer of calcite. Part of the left maxilla has been assembled and, except for some conspicuous cracks and a few missing chips of enamel, the M‘ and the M^ have been adequately restored for description. Maxilla. Maxilla contains P- and anterior alveolus of P^. A 4.5 mm long section of the diastema anterior to P^ does not reach the P‘ alveolus. Palatal process extends to median palatal suture and is about 7.0 mm wide immediately anterior to P^. P’ alveolus situated immediately posterior to P- and no diastema present. Anterior root socket of P^ is about twice diameter of that of P^. Premolar, P-. P- crown low and elongate basally. Cusp points posteriorly and anterior profile slightly convex. Posterior margin of crown distinctly concave. Surface of posterior shelf obscured by an obdurate calcite encrustation. Posterior root about twice as wide mesio- distally as anterior root. Crown length about 6.5 mm; width, 2.5 mm. P\ Alveolus only; at least 7.3 mm total length (hypertrophied); anterior root socket about 3.0 mm long and 2.0 mm wide; posterior root socket at least 4.0 mm long, width unknown. Molar, M‘. Reconstructed from three fragments; heavily worn and thickly coated with calcrete. Crown elongated, transversely narrow; short, distinct pre¬ cingulum pre.sent; deep cleft divides crown into two lobes, shallow but distinct ectoflex. Stylar cusp B about same size as paracone; metacone about half again larger than paracone. Postmetacrista .straight, angles slightly lingual to postprotocrista. Preparacrista short, extends transversely to long axis of crown to stylar cusp B. Centrocrista obtuse, essentially parallel to tooth row; stylar cusp D much smaller than metacone, closely situated distolabially and connected to it by low crest with conspicuous groove labial to metacone and mesial to stylar cusp D; small stylar cusp E heavily worn, but faintly visible. Talon short transversely with U-shaped profile. A very small metaconule represented. Cusp heights: metacone > stylar cusp B > stylar cusp D > paracone > protocone. M\ Talon only, larger than .M' talon and more V- shaped. Basin shallow, heavily worn on mesial side; conules absent; crest of base of metacone confluent with postprotocrista. MT Complete crown and roots, reconstructed from four fragments, moderate wear (slight misalignment of fragments due to missing slivers of dentine at contacts). Precingulum short, complete; parastylar spur much shorter than metastylar spur, strong ectoflexus; stylar cusp B connected to paracone by convex, mesially- angled preparacrista; stylar cusp C reduced, but present; low elongated stylar cusp D produces conspicuous labial bulge; distinct, elliptical thickening ol enamel in region of stylar cusp E, but no definite cusp structure visible. Metacone much larger and higher than paracone. Angle of centrocrista about 90“; angle between preparacrista and postmetacrista slightly wider than in Nimbacinus dicksoni and Ngamalcinus timmulvaneyi. Talon narrow. 154 Fossil thylacines from the Northern Territory Fig. 6 . Tyarrpecinus rothi gen. et sp. nov. holotype left maxilla (P98211). Maxillary fragment in A, lateral and B, occlusal views; C, reconstructed molar tooth row (M,. 4 from left to right) in occlusal view. Scale bars in mm. V-shaped, very faint protoconule evident, metaconule absent. Cusp heights: metacone > paracone > stylar cusp D > ?stylar cusp E = stylar cusp B > stylar cusp C > protoconule > protocone. Talon only; greatly reduced compared to M^; narrow, V-shaped, with low protoconule. Etymology. Named in honour of Karl Roth for his contributions to the natural history of Central Australia and long career as Custodian of the Alice Springs branch of the Museums and Art Galleries of the Northern Territory, from which he recently retired. PHYLOGENY RECONSTRUCTION Muirhead and Wroe (1998) analysed shared, derived character states amongst five thylacinid genera to 155 Murray and Megirian reconstruct phylogeny in the family. Of 32 characters analysed, 24 are dental, with the remainder pertaining to the cranium. As a preliminary analysis, the states of the three newly described species plus Muribacinus gadiyuli Wroe, 1996, and Thylacinus megiriani Murray, 1997, were scored and added to their matrix (Table 2A - necessarily omitting the eight cranial characters) and analysed with Hennig86 version 1.5 (Farris 1988). One Table 2. Thylacinid character expressions for phylogenetic analysis: A, following Muirhead and Wroe (1998), cranial characters 1, 2 and 4-9 omitted; B, preferred matrix as discussed in the text. Character definitions for matrices A and B given with Figures 7 and 8 respectively. A 3: •a c ^ ^ Jv *2? ^ 53 c ^ ^ S 5 2 ^ 5 . I C5 ^ ^ Sr ^ ,5 *S *5 S a f-, t-, h, s. 3 0 0 0 1 7 0 1 7 0 7 1 1 I 10 0 1 1 1 7 2 1 2 I 1 2 3 3 11 0 1 1 1 7 2 2 3 3 4 4 4 4 12 0 0 0 0 7 0 0 1 1 0 1 2 2 13 0 0 0 0 7 2 1 2 1 2 2 3 3 14 0 0 0 0 ? 1 1 2 2 2 2 2 2 15 0 1 1 1 7 1 1 2 2 2 2 2 2 16 0 0 0 0 7 3 0 1 2 1 1 1 1 17 0 1 1 1 7 1 1 2 2 2 2 2 2 18 0 0 0 0 1 1 0 7 0 0 0 0 0 19 0 1 1 1 1 1 2 7 3 4 4 7 4 20 0 1 1 2 0 1 0 7 4 3 3 7 3 21 0 0 0 0 0 0 1 7 0 0 0 7 0 22 0 0 7 0 0 0 0 7 1 0 0 7 0 23 0 0 7 0 1 0 7 7 7 1 1 ? 1 24 0 1 1 1 1 1 2 7 3 4 5 7 5 25 0 0 0 0 1 1 1 7 0 0 0 7 0 26 0 0 0 0 0 0 1 7 0 1 1 7 1 27 0 1 1 1 2 1 2 7 1 3 3 7 3 28 0 0 0 0 0 1 1 7 0 0 0 7 0 29 0 1 1 1 2 1 2 7 2 3 3 7 3 30 0 0 7 1 1 0 7 7 1 1 1 7 1 31 0 ? 0 0 0 0 1 7 1 I 1 7 1 B a? u 0 2 ^ OT JS s: U Ct- 1 0 0 1 2 2 2 2 7 3 3 4 7 4 2 0 0 0 0 0 1 1 0 0 0 0 0 0 3 0 0 0 0 0 0 0 1 1 1 1 1 1 4 0 0 0 0 0 0 0 0 1 1 0 0 0 5 0 0 0 0 0 0 0 1 0 0 1 1 1 6 0 0 0 0 0 0 0 0 1 1 0 0 0 7 0 0 0 0 0 0 0 0 0 0 0 I I 8 0 0 0 0 0 0 0 1 1 1 1 1 1 of eight equally parsimonious hypotheses selected for discussion purposes is shown as Figure 7A, and the Nelsen consensus tree is shown as Figure 7B. In Figures 7A and 7B, the previously analysed species branch consecutively in the same order as illustrated in Muirhead and Wroe (1998). Of the new species, Mutpuraciniis archibaldi aligns with the plesiomorphic Muribacinus gadiyuli at the base, and Tyarrpecinus rotlii branches off after Ngamalacinus ridei and Wabulacinus. The emphasis in Muirhead and Wroe (1998) on ordered multistate characters, postulating broad morphoclines for the majority of character states, tends to relegate resulting trees to a hierarchy of structural grades. Notwithstanding, there are some promising synapomorphic states that suggest clade formation within the consecutively branched dendrogram. These include the synapomorphic occurrence of a carnassial notch in the hypocristid of Badjcinus and Ngamalacinus (Character 28 in Table 2A) and mutual loss of stylar cusp B accompanied by extreme reduction of the stylar shelf in Wabulacinus ridei and Thylacinus macknessi (Character 11); which, in combination with the strongly morphoclinal expression of metaconid reduction (Characters 18-19) result in the more resolved hypo¬ theses (Fig. 8 based on the character matrix in Table 2B) preferred here. Muirhead and Wroe (1998) divide the character of metaconid reduction into two sets. Character 18, in which a differential absence of metaconid on M, and presence on the other molars is scored autapomorphically among thylacinids i'or Badjcinus, and Character 19 in which the size of metaconids on M 2 -t is scored as an independant multistate set. The recognition of a unique differential reduction of the metaconid in Badjcinus is apparently based on the assumption of uniformly reducing metaconids in Nimhacinus dicksoni, as indicated by the D-Site M 2 paratype, QMFI6809. However, Nimbacinus richi shows a similar degree of differential metaconid expression to that of Badja- cinus, in which the cusp is nearly obsolete on Mi, but well-differentiated and large on M 2 . 3 , then reduced on M 4 . Thus, while the extent of metaconid reduction on M| in Badjacinus is more advanced than in Nimbacinus richi, we do not consider the character of differential expre.ssion of the metaconid to be unique to Badjcinus (within the Thylacinidae). The states ot metaconids in N. richi also negate the proposition that the expression gradient of the metaconid along the molar row is typically uniform in thylacinids. In any case, the differential reduction of metaconids in Badjcinus turnbulli, Nimbacinus richi and Ngamalcinus timmulvaneyi may be the rule rather than the exception, and as such, the basic similarity of the differentially reduced states of the metaconids in these species is probably more meaningful for thylacinid phylogeny than the autapo- morphic absence of a metaconid on M' in Badjcinus. Fossil thylacines from the Northern Territory A Dasyurid outgroup Muribacinus gadiyuli Mulpuracinus archibaldi Badjcinus turnbulli Nimbacinus dicksoni Nimbadnus lichi Ngamaladnus timmulvaneyi Wabuladnus ridei Tjarrpednus rothi € Thylacinus macknessi i— - Thyladnus potens 1— Thyladnus megiriani B Thyladnus cynocephalus € Dasyurid outgroup Muribadnus gadiyuli Mutpuradnus archibaldi Badjcinus turnbulli Nimbacinus dicksoni Nimbacinus richi Ngamaladnus timmulvaneyi Wabuladnus ridei Thyladnus macknessi Tjarrpednus rothi Thyladnus potens Thyladnus megiriani Thyladnus cynocephalus Fig. 7. A, one of eight equally parsimonious trees depicting ihylacinid phylogeny using character matrix after Muirhead and Wroe (1998) (see Table 2A of this work); B, Nelsen consensus of the eight equally parsimonious trees (Consistency Index = 0.73; Retention Index = 0.85: Hennig 86 version 1.5 (Farris 1988)). Abridged definitions of character expressions (see Muirhead and Wroe 1998 for details). 3, infraorbital foramen; 0 , not bounded by jugal; 1 bounded by jugal. 10 , size of paracone: 0 , equivocal plesiomorphic state; 1 , slight reduction; 2, significant reduction; 3, extreme reduction. 11, stylarcusp B; 0, well developed; 1, slight reduction; 2, greater reduction; 3, more reduced yet; dramatically reduced [Muirhead and Wroe (1998) include the stylar shelf and stylar cusp D in this morphocline: actual loss or extreme reduction of stylar cusp B in thylacinids is confined to Thyladnus macknessi and Wabuladnus ridei (Muirhead 1992, 1997).] 12, anterior cingulum: 0, complete; 1, incomplete. 13, protocone and conules: 0, well developed; 1, slightly reduced; 2, significant reduction; 3, conules lost. 14, length postmetacrista: 0, as in dasyurids; I, slight elongation. 15, angle centrocrista: 0, acute; 1, more obtuse than in dasyurids; 2, colinear. 16, direction of preparacrista: 0, perpendicular; 1, slightly oblique; 2, parallel to long axis; 3, directly anterior to paracone. 17, angles preparacrista and po.stmctacrista: 0 , narrow; 1 , slightly widening; 2 , wider yet. 18, size of melaconid in Mi relative to Mj^: 0 , uniform reduction; 1, variably absent on Mi, present M 2 .<; 19, size metaconid M 24 : 0, plesiomorphic condition; 1, some reduction; 2 more reduced; nearly lost; 4, lost. 20, size entoconid: 0, large; 1, slight reduction; 2, greater reduction; 3, very reduced: 4, absent. 21. morphology entoconid: 0, plesiomorphic condition; 1. laterally compressed. 22. diastema PI-2: 0, present; 1, absent. 23, diastema P2-3: 0, present; 1, absent. 24, hypoconulid / hypoconulid notch: 0, present; I, slightly diminished; 2, obvious reduction; 3, more reduced yet; 4, extremely reduced; 5, notch absent. 25, posterior cingtilid and hypocristid: 0, separated; 1, joined. 26, cristid obliqua carnassial notch: 0, absent; 1, present. 27, angle hypocristid: 0, parallel to transverse axis of dentary; I, moderate angle; 2, intermediate; 3, pronounced. 28, carnassial notch in hypocristid: 0, absent; 1, present. 29, termination of cristid obliqua: 0, low; 1, higher; 2, stronger, higher; 3, principle posterior shearing crest. 30-31, size P3, two pathways: 00, 01, plesiomorphic; 10 apomorphic [see Muirhead and Wroe (1998) for explanation.] 32, length M4: 0, shorter than M3; longer than M3. The inclusion of Muribacinus gadiyuli and Mutpu- racinus archibaldi present a fairly complete picture of the morphocline for metaconid reduction in thylacinids: Muribacinus gadiylui shows gradual posterior en¬ largement of the metaconids ( 0 ); Mutpuradnus archibaldi shows a slight reduction in the size of M 4 metaconid (I), (M |.2 unknown); Ngamaladnus timmulvaneyi shows marked reduction of metaconid on Mi and M4; Nimbacinus richi shows further reduction of metaconid on M| with no further reduction on M 4 than in the former species, and Badjcinus turnbulli shows complete loss of metaconid on M| with no further reduction on M 4 than in the former species ( 2 ); while the metaconid is nearly obsolete on all molars of Wabuladnus ridei and Thyladnus macknessi (3) and lost on all other species of Thyladnus (4). Resolution of clade formation is obscured by the establishment of several interrelated morphoclines (widened angle of preparacrista and postmetacrista, straightened centrocrista, elongation of postmetacrista, stylar shelf and cusp reduction) expressing progressive states of carnassialisation in the Thylacinidae. These overwhelm the otherwise compelling evidence for at least two parallel trends in development of longer shearing crests while simultaneously suppressing the relatively few synapomorphic states that might otherwise have been recognised by the algorithm. While it is true that there is a general trend towards increased car- nassialization signalled among these genera, and that the structural changes that characterise these trends are essentially indistinguishable (the possible exception being T. macknessi - Muirhead 1992; 72), the disjunct nature of the character combinations indicate that the similarities in the states of carnassialization between Thyladnus macknessi, Wabuladnus ridei and the comp- 157 Murray and Megirian _r J LJ h-' Dasyurid outgroup Muiibacinus gadiyuli Mutpuradnus archibaldi Nimbadnus dicksoni Nimbadnus richi Badjdnus tumbulli Ngamaladnus timmulvaneyi Wabuladnus ridei Thyladnus macknessi Tjarrpednus rothi Thyladnus patens Thyladnus rnegiriani Thyladnus cynocephalus -c B Dasyurid outgroup Muribadnus gadiyuli Mutpuradnus archibaldi Nimbadnus dicksoni Nimbadnus richi Badjdnus tumbulli Ngamaladnus timmulvaneyi Wabuladnus ridei Thyladnus macknessi Tjarrpednus rothi Thyladnus potens Thyladnus meginani Thyladnus cynocephalus Fig. 8. Dendrograms (A. more, and B, less, resolved) depicting preferred hypotheses of thylacinid phylogeny based on a restricted suite of characters (Table 2B) which do not include the autapomorphous and discontinuous characters of Table 2A. (Hennig 86 version 1.5 (Farris 1988); 2 equally parsimonious trees; Consistency Index = 1.0; Retention Index = 1.0). Definitions of character expression. 1, pattern of metaconid reduction: 0, slight, uniform; 1, slight M 4 ; 2, differential, marked on M|, slight M 4 ; 3, near obsole.scence on all molars; 4, entirely lost on all molars. 2, carnassial notch in hypocristid: 0, absent; I, present. 3, elongation of postvallum, shearing surfaces: 0, slight; 1, significant. 4, stylocone B: 0, present; 1 , extremely reduced or lost. 5, paracone reduction / hypertrophy of metacone: 0, slight - moderate; 1, conspicuous. 6 , ectoflexus; 0, strong expression; I, weak expression. 7, precingulum: 0, pre.sent, strong; 1, reduced - absent. 8 , entoconid: 0 , distinct; 1 , reduced - absent. arably derived species of Thyladnus are phylogenetically uninformative homoplasy. As Muirhead (1992, 1997) and Muirhead and Gillespie (1995) point out, both Wabuladnus ridei and Thyladnus macknessi show progressive states of carnassialization (loss of stylar cusps, extreme reduction- loss of stylar shelf, lengthening, straightening of crests) in combination with several autapomorphies and retained plesiomorphies (relatively unreduced paracones). Whereas in Tyarrpecinus and other members of Thyladnus, carnassialization has proceeded in con¬ junction with plesiomorphous retention of the stylar shelf, stylar cusps and apomorphous reduction of paracone and/or enlargement of metacone. In other words, the crown genera share an ancestor with states close to Nimbadnus, rather than, as the Muirhead and Wroe (1998) cladogram implies, direct, successive sister relationships with Wabuladnus ridei and Thyladnus macknessi. Tyarrpecinus rothi is therefore considered to represent the plesiomorphic sister-taxon of the Thyla- cinus clade (T. potens + T. rnegiriani + T. cynocephalus, but not T. macknessi) (Fig. 8A,B), as it is too derived (more reduced M' paracone and precingulum) to have given rise to Thyladnus macknessi and probably Wabuladnus ridei (greater paracone reduction on M') while lacking highly derived states of stylar cusp and stylar shelf reduction/loss of the Thyladnus clade. Tyarrpecinus rothi shows incipient to moderate development of Thyladnus-iike states in the considerable reduction of the parastylar spur, reduction of the precingulum, elongation of the postmetacrista; mesial obliquity of the preparacrista, reduction of the paracone, extreme reduction to loss of conules and reduction of the trigone basin on M^. Resemblance to particular species of Thyladnus are evident, such as strong ectollex present in T. potens and T. rnegiriani M^. There is also the question of the expression of a tiny stylar cusp E which appears to be pre.sent in the form of an elliptical thickening of the stylar crest of in both Tyarrpecinus and Thyladnus rnegiriani (Murray 1997). Although Muirhead and Wroe (1998) have concluded that the tiny stylar cusp on the M^ of Thyladnus cynocephalus represents a distally-shifted stylar cusp D, we prefer to wait for additional data. Other than some minor shifts in emphasis on some of the character states, the new data present no significant points of disagreement with recently proposed hypo¬ theses of thylacinid phylogeny (Muirhead and Archer 1990, Muirhead 1992, Muirhead and Gillespie 1995, Wroe 1996, Murray 1997, Muirhead and Wroe 1998). However, the problem of association of the upper and lower molars of Nimbadnus dicksoni creates an ambiguity in character states of a pivotal group. The D- Site M 2 (QMFI6809) has a much narrower paraconid and much smaller metaconid than the equivalent molar of the Bullock Creek LF Nimbadnus (P9612-4), so we are certain that it represents a different taxon, probably Fossil thylacines from the Northern Territory another genus, in concert with the current level of systematic discrimination. However, the Bullock Creek LF specimens (P85553-3 and P9612-4) are sufficiently close to the holotype M 2 (QMF16802) to indicate a species-level distinction, and accordingly, we have inferred that the unknown M 2-3 of N. dicksoni are as similar to N. richi as is the Mi. Although poorly represented, the Tyarrpecinus rothi morphotype is sufficiently generalised to represent the structural stage that may have given rise to all Thylacinus species except T. macknessi, and perhaps to Wabulacinus ridei. On the basis of present information, Thylacinus macknessi branched off from a less derived form than Tyarrpecinus. Taking the few synapomorphies between Wabulacinus and Thylacinus macknessi at face value, they become, more or less by default, sister taxa in a minor clade (Fig. 8A,B) characterised by progressive, predominantly homoplasious camassialization. Muirhead and Archer’s (1990) observations on some strong si¬ milarities between N. dicksoni and T. potens Woodburne, 1967, become more pertinent here, because, as they suggest, the well-developed ectoflexus in N. dicksoni may constitute a synapomorphic condition. With the additions of Mutpuracinus archibaldi, Tyarrpecinus rothi and Thylacinus megiriani, all of which display strong ectoflexus on M^, we are inclined to favour the interpretation of a symplesiomorphy retained in a conservative lineage culminating with the genus Thylacinus. STRATIGRAPHIC PALAEONTOLOGY AND BIOCHRONOLOGY The newly described Bullock Creek and Alcoota LF species brings the total number of ?late Oligocene and Miocene thylacinids to 11 species in eight genera, all but Nimbacinus and Thylacinus being monospecific. No species are represented by abundant material, and thylacinids are presently of minimal use in intraformational biocorrelation and of no use in intracontinental correlation (Table 3). Even at the generic level, a high degree of faunal regionalism is indicated, which follows the general pattern observed by Rich (1991) for the Australian mid Tertiary vertebrate record. There is stratigraphic evidence of sympatry for: Mutpuracinus archibaldi + Nimbacinus richi (‘Top Site’, Camfield Beds); Ngamalacinus timmulvaneyi + Wabulacinus ridei (‘Camel Sputum Site’, Carl Creek Limestone); Thylacinus macknessi + Muribacinus gadiyuli (‘Gag Site’, Carl Creek Limestone) and Thylacinus potens + Tyarrpecinus rothi (Waite Formation - two excavation sites, ‘Main Pit' and ‘Paine Quarry’, sampling the Alcoota LF horizon) (Table 4). Although the ‘Top Site’ (Camfield Beds) assemblage is drawn from several lithofacies (Murray and Megirian 1992), and therefore possibly different beds, P9215-4 (Nimbacinus richi) and P9215-5 (Mutpuracinus archibaldi) were found lying one upon the other in the same block of limestone. The only biostratigraphic evidence of temporal succession amongst thylacinids comes from super¬ position in the Waite Formation of Thylacinus megiriani (Ongeva LF) over Thylacinus potens and Tyarrpecinus rothi (Alcoota LF) (Murray et al. 1993; Megirian et al. 1996; Murray 1997; Megirian 2000), but a broader interpretive perspective is provided by zygomaturine stage-of-evolution biochronology (Woodburne et al. 1985; Megirian 1994; Murray etal. 2000) (Table 5). The as yet un-named species of Neohelos from the Camfield Beds and Carl Creek Limestone ‘Gag’ and ‘Henks Table 3. Stratigraphic di.stribution of ?late Oligocene and Miocene species of Thylacinidae. and biocorrelation of fossil a.s.semblages. Formation (geographic location) IxKal Fauna Site / Quarry Species Waite Formation (Alcoota) Ongeva South Quarry Thylacinus nwgiriani Alcoota Paine Quarry Thylacinus potens Alcoota Main Pit Tyarrpecinus rothi Camfield Beds (Bullock Creek) Bullock Creek Top Site Mutpuracinus archibaldi Bullock Creek Top Site NimhaciniLS richi Bullock Creek Blast Site Ninthacinus richi Carl Creek Limestone (Riversleigh) Camel Sputum Ngarmilacinus tinunulvaneyi Inabeyance Ngamalacinus timmulvaneyi Camel Sputum Wabulacinus ridei D-Site Nimbacinus dicLsoni Henks Hollow Nimlxicinus dicksoni Gag Muribacinus gadiyuli Gag Thylacinus macknessi Nevilles Garden Thylacinus nuicknessi Mikes Menagerie Thylacinus mm:knessi White Hunter Badjcinus turnbulli 159 Murray and Megirian Table 4. Stratigraphic evidence of sympatry amongst Miocene thylacinids. Formation Local Fauna Site Species Waite Formation Ongeva South Quarry Thylacinus megiriani Alcoota Paine Quarry Thylacinus patens Main Pit Tyarrpecinus roihi Camfield Beds Bullock Creek Top Site Mutpuracinus archibaldi Top Site Nimbacinus richi Blast Site Nimbacinus richi Carl Creek Limestone Camel Sputum NgamaUicinus timmulvaneyi Camel Sputum Wabulacinus ridei White Hunter Badjcinus turnbuUi Gag Muribacinus gadiyuli Gag Thylacinus macknessi Inabeyance Ngamalicinus timmulvaneyi Henks Hollow Nimbacinus dicksoni D-Site Nimbacinus dicksoni Mikes Menagerie Thylacinus macknessi Nevilles Garden Thylacinus macknessi Hollow’ sites (Murray et al. 2000) is here identified as "Neohelos sp. nov.’. The following conclusions can be drawn from Table 5. 1. The apparent contemporaneity of Nimbacinus richi (Camfield Beds) and Nimbacinus dicksoni (Carl Creek Limestone) in Neohelos sp. nov. time suggests that these were allopatric sibling species. 2. Thylacinus macknessi lived at Riversleigh from Neohelos tirarensis to Neohelos sp. nov. time, which Murray et al. (2000) suggest may have spanned the latest Oligocene to mid Miocene. If ‘D-Site’ Nimbacinus dicksoni is actually conspecific with ‘Henks Hollow’ N. dicksoni as proposed by Muirhead and Archer (1990), then this taxon may have had a similar temporal range. Alternatively, the ‘D-Site’ Nimbacinus material may represent a different species, as discussed in the Systematics section above, which could conceivably represent the ancestor of one or both of Nimbacinus dicksoni and N. richi. 3. Five species of thylacinid may have co-existed at Riversleigh during N. tirarensis time, supporting the conclusions of Muirhead (1997), Muirhead and Wroe (1998) and of this study that a major phylogenetic radiation of thylacinids occurred before the ?late Oligocene (= pre N. tirarensis time). 4. No potential ancestor-descendent relationships have been identified amongst the 11 known Miocene thylacinid species; Mutpuracinus archibatdi is identified above as having structural states that may reflect the ancestral conditions giving rise to Nimbacinus dicksoni, but a phyletic succession from M. archibaldi to N. dicksoni is not consistent with biochronological indications in that Mutpuracinus archibaldi post-dates the earliest record of Nimbacinus dicksoni. Estimating changes in thylacinid diversity through the Miocene is problematic due to different taphonomic processes at the very few known fossil- producing localities. Remains of small taxa are rare in the ephemeral channel and overbank deposits of the Waite Formation, and small terrestrial animals are sparse in the billabong and channel deposits of the Camfield Beds (Murray and Megirian 1992). Per¬ sistent sampling over many years produced the three new species reported here. In contrast, the small-scale cave-entrance/perched springline, fluvial tufa and associated pond deposits of the N. tirarensis biochron of the Carl Creek Limestone are noted for their preservation of small and medium-sized animals (e.g. Archer e/a/. 1989, 1991; Megirian 1992, 1997). The apparent decline in diversity in the Carl Creek Limestone from five species during Neohelos tirarensis time to two species during Neohelos sp. nov. time, with no species reported yet from apparently younger zygomaturine biochrons, may also reflect preservational and sampling biases. The association of quite highly advanced and very conservative forms in the Carl Creek Limestone during N. tirarensis time (e.g. Thylacinus macknessi -i- Muribacinus gadiyuli) reflects the extent of independent (parallel) evolution of lineages which emerged in the Paleogene, including the crown group comprising the Waite Formation species and culminating in the recently extinct ‘Tasmanian wolf’, T. cynocephalus. It appears, from absence in the later fossil record, that by the late Miocene (Kolopsis torus time), all the conservative lineages as well as the quite advanced Thylacinus macknessi + Wabulacinus ridei clade, had become extinct, but the pattern of succession is not clear. 160 Fossil thylacines from the Northern Territory Table 5. A, thylacinid biochronology in terms of zygomaturine stage-of-evolution (Murray et al. 2000): Neohelos tirarensis stage = latest Oligocene / early Miocene; Neohelos sp. nov. stage = middle Miocene; Kolopsis torus stage = late Miocene; K. yperus stage = latest Miocene. * No Neohelos has been described from Carl Creek Limestone White Hunter Site, which is here tied in by simple marsupial biocorrelation, B, using burramyids (Brammall and Archer 1997), macropodids (Cooke 1997: table 1), phascolarctids (Black and Archer 1997) and the diprotodontoid Bematherium anguhtm (Black 1997: fig.l). White Hunter clusters with sites containing N. tirarensis, using Bray and Curtis agglomeration analysis and flexible UPGMA (Un-weighted Pair Group Method, Arithmetic average) algorithms in PATN (Belbin 1994). A Formation Local Fauna Site Thylacinid species Zygomaturine stage-of-evoIution (Murray et al. 2000) Waite Formation -unconformity- Ongeva South Quarry Thylacinus megiriani Kolopsis yperus (-t- Kolopsis torus) Waite Formation Alcoota Paine Quarry Thylacinus potens Kolopsis torus Waite Formation Alcoota Main Pit Tyarrpecinus rothi Kolopsis torus Camfield Beds Bullock Creek Top Mutpuracinus archibaldi Neohelos sp. nov. Camfield Beds Bullock Creek Top Nimbacinus richi Neohelos sp. nov. Camfield Beds Bullock Creek Blast Nimbacinus richi Neohelos sp. nov. Carl Creek Limestone Gag Muribacinus gadiyuli Neohelos sp. nov. Carl Creek Limestone Gag Thylacinus macknessi Neohelos sp. nov. Carl Creek Limestone Henks Hollow Nimbacinus dicksoni Neohelos sp. nov. Carl Creek Limestone Camel Sputum Ngamalacinus timmulvatieyi Neohelos tirarensis Carl Creek Limestone Camel Sputum Wabulacinus ridei Neohelos tirarensis Carl Creek Limestone Inabeyance Ngamalacinus timmulvaneyi Neohelos tirarensis Carl Creek Limestone D-Site Nimbacinus dicksoni Neohelos tirarensis Carl Creek Limestone Mikes Menagerie Thylacinus macknessi Neohelos tirarensis Carl Creek Limestone Nevilles Garden Thylacinus macknessi Neohelos tirarensis Carl Creek Limestone White Hunter Badjeinus turnbulli (Neohelos tirarensis)* B 53 & Site Burramys briityi Bulungamay delicata Wabularoo naughtoni Ganawamayi sp2 Balbaroo gregoriensis Nambaroo sp5 Nowidgee matrix Litokoala kanunkaensi Nimiokoala greystanesi Bematheriui angulum Neohelos sp. nov. Neohelos tirarensis Gag Henks Hollow Mikes Menagerie Camel Sputum Nevilles Garden 1 1 1 1 I 1 1 1 1 1 1 1 1 1 1 1 I 1 1 1 1 1 1 1 1 1 Inabeyance D-Site White Hunter 1 1 1 1 1 1 1 1 1 1 1 1 * REFERENCES Archer, M. 1978. The nature of the molar-premolar boundary in marsupials and interpretation of the homology of marsupial cheekteeth. Memoirs of the Queensland Museum 18: 157- 164. Archer, M. 1982. A review of Miocene thylacinids (Thylacinidae: Marsupialia), the phylogenetic position of the Thylacinidae and the problem of apriorisms in character analysis. In: Archer, M. (ed.) Carnivorous marsupials. Pp. 445-476. Royal Zoological Society of New South Wales: Sydney, Australia. Archer, M., Godthelp, H., Hand, S.J. and Megirian D. 1989. Fossil mammals of Riversleigh, northwestern Queens¬ land: preliminary overview of biostratigraphy, correlation and environmental change. Australian Zoologist 25(2): 29-65. Archer, M., Hand S.J. and Godthelp, H. 1991. Riversleigh. The story of animals in ancient rainforests of inland Australia. Reed Books Pty Ltd: Sydney, Australia. Belbin, L. 1994. PATN, pattern analysis package users guide. CSIRO Division of Wildlife and Ecology, Lyneham ACT, Australia. 161 Murray and Megirian Black, K. 1997. Diversity and biostratigraphy of the Di- protodontoidea of Riversleigh, northwestern Queensland. Memoirs of the Queensland Museum 41(2): 187-192. Black, K. and Archer, M. 1997. Nimiokoala gen nov. (Mar- supialia, Phascolarctidae) from Riversleigh, northwestern Queensland, with a revision of Litokoala. Memoirs of the Queensland Museum 41(2): 209-228. Bonaparte, C.L.J.L. 1838. Synopsis vertcbratorium systematis. Nuovi Annual, Science and Nature, Bologna 2: 105-133. Brammall, J. and Archer, M. 1997. A new Oligocene-Miocene species of Burramys (Marsupialia, Burramyidae) from Riversleigh, northwestern Queensland. Memoirs of the Queensland Museum 41(2): 247-268. Cooke, B.N. 1997. Biostratigraphic implications of fossil kangaroos at Riversleigh, northwestern Queensland. Memoirs of the Queensland Museum 41(2): 295-302. Dawson, L. 1982. Taxonomic status of fossil thylacines (Thylacinus, Thylacinidae, Marsupialia) from late Quaternary deposits in eastern Australia. In: Archer, M. (ed.) Carnivorous marsupials. Pp 445-476. Royal Zoological Society of New South Wales: Sydney, Australia. Farris, J.S. 1988. Hennig86. Version 1.5. Port Jefferson Station: New York. Flower, W. 1869. Remarks on the homologies and notation of teeth in the Marsupialia. Journal of Anatomy and Physiology 3: 262-278. Gill, T. 1872. Arrangement of families of mammals with analytical tables. Smithsonian Miscellaneous Collection 2: 1-98. Goldfuss, G. 1820. Handhuch der Zoologie. Abteilung II. J. L. Schrag: Nuremburg. Henderson, J. and Dobson, V. 1994. Eastern and Central Arrente to English dictionary. lAD Press: Alice Springs, Australia. Luckett, P. 1993. An ontogenetic assessment in dental homologies in the therian mammals. In: Szalay, F., Novacek, M., and McKenna, M. (eds) Mammal phylogeny: Mesozoic differentiation, multituberculates. monotremes, early therians. and marsupials. Pp 182-204. Springer-Verlag: New York. Megirian, D. 1992. Interpretation of the Miocene Carl Creek Limestone, northwestern Queensland. The Beagle. Records of the Northern Territory Museum of Arts and Sciences 9(1): 219-248. Megirian, D. 1994. Approaches to marsupial biochronology in Australia and New Guinea. Alcheringa 18: 259-274. Megirian, D. 1997. The geology of the Carl Creek Limestone. Unpublished PhD thesis, Flinders University of South Australia, Adelaide. Megirian, D. 2000. Report on shallow angering at the MAGNT Alcoota Fossil Reserve, June and August, 1998. Museums and Art Galleries of the Northern Territory Research Report 7. Megirian, D., Murray, P.F. and Wells, R.T. 1996. The late Miocene Ongeva Local Fauna of central Australia, The Beagle. Records of the Museums and Art Galleries of the Northern Territory 13: 9-38. Muirhead, J. 1992. A specialised ihylacinid, Thylacinus macknessi, (Marsupialia: Thylacinidae) from Miocene deposits of Riversleigh, northwestern Queensland. Australian Mammalogy 15: 67-76. Muirhead, J. 1997. Two early Miocene thylacines from Riversleigh, northwestern Queensland. Memoirs of the Queensland Museum 41(2):367-377. Muirhead, J. and Archer, M. 1990. Nimhacinus dicksoni, a plesiomorphic thylacine (Marsupialia: Thylacindae) from Tertiary deposits of Queensland and the Northern Territory. Memoirs of the Queensland Museum 28(1): 203-221. Muirhead, J. and Gillespie, A. 1995. Additional parts of the type specimen of Thylacinus macknessi (Marsupialia: Thylacinidae) from Miocene deposits of Riversleigh, northwestern Queensland. Australian Mammalogy 18: 55- 60. Muirhead, J. and Wroe, S. 1998. A new genus and species, Badjcinus turnhulli (Thylacinidae: Marsupialia) from the late Oligocene of Riversleigh, northern Australia, and an investigation of thylacinid phylogeny. Journal of Vertebrate Paleontology 18(3): 612-626. Murray, P.F. and Megirian, D. 1992. Continuity and contrast in middle and late Miocene vertebrate communities from the Northern Territory. The Beagle, Records of the Northern Territory Museum of Arts and Sciences 9( 1): 195-218. Murray, P.F. 1997. Thylacinus megiriani, a new species of thylacine (Marsupialia: Thylacinidae) from the Ongeva Local Fauna of central Australia. Records of the South Australian Museum 30(1): 43-61. Murray, P.F., Megirian, D. and Wells, R.T. 1993. Kolopsisyperus sp. nov. (Zygomaturinae, Marsupialia) from the Ongeva Local Fauna: new evidence for the age of the Alcoota fossil beds of central Australia. The Beagle, Records of the Northern Territory Museum of Arts and Sciences 10(1): 155- 172 Murray, P.F., Megirian, D., Rich, T.H., Plane, M., Black, K., Archer, M., Hand, S. and Vickers-Rich, P. 2000. Morphology, systematics and evolution of the marsupial genus Neohelos Stirton (Diprotodontidae, Zygomaturinae). Museums and Art Galleries of the Northern Territory Research Report 6. Rich, T.H. 1991. Monotremes, placentals, and marsupials: their record in Australia and its biases. In: Vickers-Rich, P., Monaghan, J.M., Baird, R.F. and Rich, T.H. (eds) Vertebrate palaeontology of Australasia. Pp 893-1069. Pioneer Design Studio and Monash University Publications Committee: Melbourne, Australia. Tindale, N.B. 1974. Aboriginal tribes of Australia. University of California Press: Berkeley. Woodburne, M.O. 1967. The Alcoota Fauna, central Australia. An integrated palaeontological and geological study. Bureau of Mineral Resources Bulletin 87: 1-187. Woodburne, M.O., Tedford, R.H.. Archer, M., Turnbull, W.D., Plane, M.D., and Lundelius Jr, E.L. 1985. Biochronology of the continental mammal record of Australia and New Guinea. Special Publication of the South Australian Department of Mines and Energy 5:347-363. Wroe, S. 1996. Murihacinus gadiyuli (Thylacinidae: Mar¬ supialia), a very plesiomorphic thylacinid from the Miocene of Riversleigh, northwestern Queensland, and the problem of paraphyly for the Dasyuridae (Marsupialia). Journal of Paleontology 70(6): 1032-1044. Accepted 21 November 2000 162 The Beagle, Records of the Museums and Art Galleries of the Northern Territory, 2000 16: I63-I75 Records of cetacean strandings in the Northern Territory of Australia RAY CHATTO' AND ROBERT M. WARNEKE^ 'Parks and Wildlife Commission of the Northern Territory Palmerston, NT0831, AUSTRALIA ray. chatto@nt.gov. an ^ 1511 Mt Hicks Road, Yolla, TAS 7325, AUSTRALIA ABSTRACT Fifty-seven cetacean strandings are here documented for the Northern Territory (NT) coastline. This total includes events discovered in the course of Parks and Wildlife surveys, chance encounters reported by a variety of observers, and records held by various other government agencies and institutions. A total of 35 of these records are identified to species, seven to genus (Globicephala, Balaenopiera), and the remainder can be listed only as ‘dolphin’ (3), ‘beaked whale’ (l)or ‘not known’ (II). Of 26 species known to occur in Australia’s tropical waters only 12 had previously been confirmed for the NT, and two others were provisionally listed on the basis of specimen material requiring further analysis. Strandings reported here include 10 of those 12 species and verify the two previously listed as uncertain (Globicephala macrorhynclius, Kogia simiisj. The total confirmed list for NT waters (within three nautical miles of the coast) is further expanded to 16 species, now including the melon-headed whale Peponocephala electro (2 strandings) and killer whale Orcinus orca (sighting). A further species, the .sei whale Balaenoptera borealis (trawled carcass), can also be added to the Commonwealth waters (within two hundred nautical miles) off the NT coast. Keywords, cetaceans, whales, dolphins, strandings. Northern Territory, Australia. INTRODUCTION With the inclusion of offshore islands and estuary systems, the Northern Territory coastline is over 10,000 km in length. For the most part it is of low relief, consisting mainly of sand beaches and mangrove flats, interrupted by headlands which seldom rise above 30 m. Because much of this coast is remote and sparsely inhabited, discoveries of stranded cetaceans are highly fortuitous events, and often their remains are quickly lost or scattered because under tropical conditions, carcasses rapidly disintegrate. Aboriginals do encounter beached cetaceans from time to time, but difficulties of communication and traditional concerns mean that few of these incidents are likely to be reported. A potentially important source of sightings are Coastwatch Bights, which are conducted along the entire Northern Territory coast on a fairly frequent basis, but the primary objective is surveillance of human activities. Recently stranded single large whales, small groups and schools are detected from time to time, and arc reported, but decomposing and partly buried carcasses are likely to be passed unnoticed. Over the past ten years one of us (RC) has conducted numerous surveys of terrestrial wildlife in coastal areas on the ground and from the air. Incidental encounters with stranded cetaceans were investigated as a matter of routine, but were usually limited by the means then in hand and unfortunately the salvage of specimen material was generally impractical. This paper summarises these direct observations on beached cetaceans or their remains, as well as reliable reports received by wildlife authorities from other observers, and documented records held both by the Museum and Art Galleries of the Northern Territory (MAGNT), Darwin, and by the National Cetacean Stranding Database (NCSD) main¬ tained by Environment Australia, Canberra. Such a summary of cetacean strandings in the Northern Territory has not been attempted in the past. METHODS Information on strandings from all sources other than newspapers, which have not been searched, has been collated according to the following format: Species, Date, Locality (including latitude/longitude) and Comments. The primary date given is qualified by the term ‘stranded’ (which is either the known date of stranding or an estimated recent date if the specimen(s) was found dead in fresh condition), or the term ‘found’ in cases where specimens were decomposing or skeletal; otherwise it is simply the date a report was received. Doubtful circumstances are explained under Comments. Locality is given as a latitude and longitude and then described with reference to named topographical features, although in some instances this and the coordinates can only be approximations. The systematic order of the records listed in this paper follows Bannister et al. (1996). Chatto and Warneke Specimens retained by MAGNT and referred to in this paper are listed using a reference number commen¬ cing with ‘U’. Records taken from the NCSD and used in this paper are listed by NCSD followed by their number on that database. Other abbreviations used in this paper include: NT (Northern Territory), PWCNT (Parks and Wildlife Commission of the Northern Territory), DPIF (De¬ partment of Primary Industries and Fisheries) and NT News {Northern Territory News). SYSTEMATIC LIST OF CETACEANS STRANDED IN THE NORTHERN TERRITORY Family Delphinidae Sousa chinensis - Indo-Pacific hump-backed dolphin Date. Eound August 1948. Locality. I2°15’S, I36°54’E. Yirrkala, near Gove. Comments. Johnson (1964) reports four mandibles, likely of three different animals, that were picked up off the beach by Aboriginal children during the American- Australian scientific expedition to Arnhem Land. Measurements and teeth counts are given in Johnson (1964). Date. Stranded 23 November 1985. Locality. l2°2i’S, I30°52’E. Casuarina Beach, Darwin. Comments. Single carcass washed ashore in fresh condition, male, length 2.08 m. Skull, mandibles and teeth collected by B. Freeland, PWCNT. Condylobasal length 510 mm, width 185 mm and height 175 mm; 30 teeth alveoli in one maxillary row and 32 teeth alveoli in one mandible. Photographs in PWCNT file PI999/ 629. Skull and mandibles in MAGNT, registered # U254. NCSD# 180. Date. Found 28 October 1988. Locality. 13°10’S, 130°07’E. Channel Point, opposite North Peron Island Comments. Skull, lacking mandibles, condylobasal length 505 mm, width 220 mm and height 190 mm; 30 tooth alveoli in maxillary row. Collected by D. Neal. No further details. Skull in MAGNT, registered # U528. Date. Stranded c. 2 March 1996. Locality. 1I°20’S, 132°07’E. Knocker Bay, Cobourg Peninsula. Comments. Single carcass, in fresh condition, no external injuries, c. 2 m in length and weight 69 kg. Found 3 March by A. Withers, PWCNT. Collected entire by MAGNT and has been processed (Temporary taxi¬ dermy ref. no. 342) to be accessioned as a disarticulated skeleton. Date. Stranded 28 October 2000. Locality. 12°20’S, 130°53’E. Lee Point beach, Darwin. Comments. Single carcass washed ashore in rea¬ sonably fresh condition, female, length 2.1 m. There were no obvious external injuries except a quite deep cut at the base of the tail, which is sometimes indicative of being caught in a mesh net. Thirty-one teeth in one maxillary row and 29 teeth in one mandible. Photographs (nos 6761-6756) and further measurements in PWCNT file PI999/629. Entire specimen collected by MAGNT; had not yet been registered at time of writing. Tursiops truncatiis cf. aduncus - bottlenose dolphin Remarks. It is likely that all Tursiops in NT waters are referable to the subspecies aduncus (Banister etal.). This species group was under taxonomic review at the time of writing and aduncus may be elevated to full species rank in the future (C. Kemper, South Australian Museum, pers. comm.). Date. Found 22 July 1972. Locality. 14°53’S, 135°43’E. Maria Island, Gulf of Carpentaria. Comments. Partially weathered skull, lacking mandibles, condylobasal length 420 mm, width 205 mm, height 160 mm; 25 (+?1) teeth alveoli in one maxillary row. Collected by D. Howe. No further details. Skull in the MAGNT, registered # U3955. Date. Found 6 October 1972. Locality. 11°00’S, 136°46’E. Cape Wessel, Mar- chinbar Island. Comments. Partially weathered skull, lacking mandibles, condylobasal length 420 mm, width 210 mm and height 170 mm; 24 (+?1) teeth alveoli in one maxilla. Collected by W. Dodd. No further details. Skull in the MAGNT, registered # U3956. Date. Found c. 1977 during the ‘dry’ season (May to September). Locality. 1I°47’S, 132°34’E. Mouth of Minimini Creek, south of Cobourg Peninsula. Comments. Carcass of an adult, badly decomposed. Found by F. Woerle, PWCNT. Carcass left on site. No further details. Date. Found 5 June 1985. Locality. 11°26’S, 132°58’E. North-east of Mur- genella. Van Diemen Gulf. Comments. Partially weathered skull, lacking mandibles, condylobasal length 430 mm, width 210 mm and height 160 mm; 24 teeth alveoli in one maxillary row. Collected by W. H. Butler, 5 June 1985. No further details. Skull in MAGNT, registered # U24L Date. Found 30 November 1993. Locality. 15°30’S, 136°57’E. Urquhart Island, Sir Edward Pellew Islands. Comments. Bleached disarticulated skeleton, including skull and mandibles. Found by R. Chatto, 30 November 1993, lying among rocks. Condylobasal length of skull c. 420 mm, mandibular tooth count (from 164 Whale and dolphin strandings in the Northern Territory alveoli) 24 + 24; wear on Ihe only remaining tooth indicates animal was fully mature. Measured and photographed, but no material collected. Photographs (nos 5253-57) in PWCNT file PI999/629. Date. Found 29 September 1994. Locality. 15°30’S, 136°55’E. Pearce Islet, Sir Edward Pellew Islands. Comments. Bleached skeleton, including skull and mandibles, found on beach by R. Chatto. Condylobasal length 400 mm; 22 (+2?) teeth alveoli in one maxillary row and one mandibular row on one side. Photographed and measured but no material collected. Photographs (nos 5284 and 87) in PWCNT file PI999/629. Date. Found May 1998. Locality. 12°24’S, 136°55’E. 10 km south of Cape Arnhem. Comments. Weathered skull on beach; no other skeletal remains. Found by M. Stevens, PWCNT. Skull on display in the PWCNT office, Batchelor, NT. Stenella attenuata - pan-tropical spotted dolphin Date. Stranded. 20 October 1999. Locality. 11°10’S, 132°09’E. Black Point, Cobourg Peninsula. Comments. Single adult female 2.15 m in length stranded alive; emaciated. Examined same day by PWCNT staff and euthanased on veterinary advice. Field necropsy revealed old trauma to head, but no other obvious injuries. Histology of internal organs showed changes consistent with prolonged starvation, but no evidence of disease. Whole specimen secured near site for later retrieval of skeleton by MAGNT. Photographs and report in NT News, 6 November 1999. Photographs (nos 6536-54) and measurements in PWCNT file PI 999/629. Stenella longirostris - spinner dolphin Date. Found 5 June 1985. Locality. 11° 26’S, 132° 58’E. North-east of Murgenella, Van Diemen Gulf. Comments. Weathered partial skull, lacking man¬ dibles and maxillae; condylobasal length >370 mm, width 150 mm and height 130. Collected by W. H. Butler on 5 June 1985. No further details. Specimen compared with other S. longirostris skulls in MAGNT for identification. Skull in MAGNT, registered # U242. Date. Stranded c.l9 April 1994. Locality. 11°I7’S. 132°43’E. Templer Island, off Croker Island. Comments. Single carcass, in fresh condition, no external injuries. Found by R. Chatto, 19 April 1994, on beach. Length of body 1.5 m, tip of upper Jaw to apex of melon 150 mm, height of dorsal fin 150 mm; 43+ maxillary teeth and 45+ mandibular teeth on one side. No material collected. Photographs (nos 5240, 41,45, 47, 49, and 51) in PWCNT file PI999/629. Date. Found September 1998. Locality. 11° 08’S, 132° lO’E. Just east of Smith Point, Cobourg Peninsula. Comments. Five decomposing carcasses found by A. Withers, PWCNT, over a 30 m section of beach adjacent to a broad tidal flat with rapid tidal flux. Carcasses left on site. Single vertebra collected for display at Black Point Ranger Station, Gurig National Park. No further details Peponocephala electra - melon-headed whale Date. Stranded, from 21 to 26 March 1996. Locality. 11°45’S, 135°54’E. Eastern end of Elcho Island. Comments. A mass stranding of 40 animals, occurred over six days. Examined and photographed by R. Chatto; one complete skull collected, condylobasal length 440 mm, width 250 and height 190 mm; 25 teeth alveoli in one maxillary row and 24 teeth alveoli in one mandibular row. Photographs (nos 4775, 77-79, 80, 82-85 and 91) in PWCNT file PI999/629. Complete skull in MAGNT, registered # U4438. See Chatto (2000c) for a detailed description. Date. Found 8 October 1996. Locality. 15°35’S, 136°3rE. North-west side of West Island, Sir Edward Pellew Islands. Comments. Weathered skull, lacking mandibles, found on beach by R. Chatto. Condylobasal length 300 mm, width 200 mm, 19+ teeth alveoli on either side. Photographed and measured but no material collected. Photographs (nos 5264 and 75) in PWCNT file P1999/ 629. Identification confirmed with reference to MAGNT # U4438. Pseudorca crassidens - false killer whale Date. Found 1968. Locality. 11°13’S, 131 °59’E. Trepang Bay, Cobourg Peninsula. Comments. Bleached skull and mandibles, condy¬ lobasal length 620 mm, width 350 mm and height 260 mm; nine teeth in each maxillary row, eight in each mandible. Collected by H. J. Frith, CSIRO. No further details. Skull and mandibles in the Museum and Art Galleries of the Northern Territory (MAGNT), registered # U2056. Date. Found 8 October 1996. Locality. 15°35’S, 136°31’E. North-west side of West Island, Sir Edward Pellew Islands. Comments. Weathered skull, lacking mandibles, found on beach by R. Chatto. Condylobasal length 560 mm, width 330 mm, height 260 mm, c. 9-10 pairs of teeth in each Jaw. Photographed and measured, but no material collected. Photographs (nos 5267, 69, 71-72) in PWCNT file PI999/629. 165 Chatto and Warneke Globicephala macrorhynchus - short-finned pilot whale Date. Stranded 18 March 1999. Locality. 11°11’S, 130°22’E. Cape Van Diemen, NE tip of Melville Island. Comments. Three adults and two juveniles on beach, reported by Coastwatch. When examined next day by R. Chatto, only the smallest was still alive. It was euthanased and all specimens were left on site. Reported in NT News, 19 March 1999. Photographs (nos 5798- 5804) in PWCNT file PI999/629. See Chatto (2000d) for a detailed description. Globicephala sp. - pilot whale Date. Found 1983. Locality. 12°05’S, 135°22’E. Jigaimarea Point, Howard Island, NW Arnhem Land. Comments. A school of 30-50 ashore. Source of report not stated, no further details. NCSD # 168. Date. Found 1984. Locality. 12°02’S, 134°58’E. Yabooma Island, off Millingimbi, NE Arnhem Land. Comments. A school of 20 animals reported to have stranded - one died and the remainder escaped. Source of report not stated, no further details. NCSD # 169. Date. Found 1984. Locality. 12°02’S, 134°58’E. Yabooma Island, off Millingimbi, NE Arnhem Land. Comments. Three animals reported to have stranded - one died and two escaped. Source of report not stated, no further details. NCSD # 170. Date. Stranded c. 31 March 1985. Locality. 15°24’S, 136°13’E. Rosie Creek, SW Gulf of Carpentaria. Comments. Group of eight animals seen on 1 April 1985 comprising six adults and two calves; one large male still alive. Some whales on beach, some above high tide mark and some washed inland of tree line by cyclone surge. Source of report not stated, no further details. NCSD# 173. Date. Found 10 April 1985. Locality. 16°04’S, 137° 17’E. Myoorlka Island, mouth of the Robinson River, SW Gulf of Carpentaria. Comments. Single animal on beach, partly covered by sand and weed. Source of report not stated, no further details. NCSD # 174. Orcaella brevirostris - Irrawaddy dolphin Date. Found 16 July 1948. Locality. I2°15’S, 136°41’E. Melville Bay, near Gove. Comments. Johnson (1964) reports two skulls of unknown sex, one with a partial body skeleton, were picked up at an Aboriginal campsite during the American-Australian scientific expedition to Arnhem Land. The skulls were likely to have been the remains of dolphins eaten by Aboriginals and caught in Melville Bay. Photographs and some skull measurements are given in Johnson (1964). Date. Found 12 June 1992. Locality. 12°50’S, 130°16’E. Fog Bay, 70 km SW of Darwin. Comments. Single carcass, decomposing, c. 1.5 m in length, with bulbous head partly eaten by dingos. Found and examined by R. Chatto. No material collected. Photographs in PWCNT file PI999/629. Date. Found August 1997. Locality. 12°37’S, 130°31’E. Indian Island, Bynoe Harbour. Comments. Weathered skull on beach, no other skeletal remains. Collected by PWCNT staff. Skull on display in the PWCNT Office, Batchelor, NT. Date. Found 27 May 1999. Locality. 13°22’S, 136°08’E. North-east side of Isle Woodah, Gulf of Carpentaria. Comments. Collapsed and dehydrated carcass, skin and flippers still present, body length estimated 1.5 m. Skull separate, clean, mandibles lacking, 18 tooth alveoli in each maxilla. Found by R. Chatto, no material collected. Photographs (nos 6110 and 6111) in PWCNT file PI999/629. Date. Stranded 1 October 1999. Locality. 12°46’S, 130°22’E. Dundee Beach, Fog Bay. Comments. Single adult female 1.93 m in length, washed ashore dead. Examined same day by PWCNT staff; six 5 cm cuts in a line along its belly. These were likely to have been done by a person with a knife, probably alter death to cause the animal, which may have been removed from a fishing net, to sink. Estimated to have died 1-2 days previously. Necropsy indicated that the animal was healthy and in good condition prior to death. Specimen currently held frozen by MAGNT (Temporary taxidermy ref. no. 1366) to be accessioned as a disarticulated skeleton. Photographs (nos 6409-14) and measurements in PWCNT file PI999/629. Dolphin, unidentified Date. Found June 1983. Locality. 13°06’S, 130°07’E. Just north of Channel Point, opposite North Peron Island. Comments. Partly decomposed carcass lying in intertidal zone, estimated length c. 2.5 m. Found by A. Donati, Department of Primary Industries and Fisheries (DPIF). Specimen left at site, no further details. Date. Stranded 13 April 1985. Locality. 13° 50’S, 136° OO’E. South of Cape Barrow, west of Groote Eylandt. Comments. Found alive, but rescue not attempted. Possibly Tursiops but definitive details not recorded. NCSD# 175. Date. Found 17 May 1985. 166 Whale and dolphin strandings in the Northern Territory Locality. 13°39’S, 129°48’E. c.l7 km south of Cape Scott, 160 km SW of Darwin. Comments. One animal ashore and a second floating near shore c. 2 km to west. Source of report not stated, no further details. NCSD # 179. Family Ziphiidae Ziphiiis cavirostris Cuvier’s or goose-beaked whale Date. Found 1969. Locality. 11°22’S, 132°18’E. Cobourg Peninsula. Comments. Weathered skull, lacking mandibles, condylobasal length 910 mm, width 480 mm and height 500 mm. Collected by D. Lindner, PWCNT. No further details. Photographs in PWCNT file PI999/629; skull in MAGNT, registered # U2055. Date. Found 30 July 1998. Locality. 12°02’S, 134°57’E. Yabooma Island, off Millingimbi. Comments. Bleached skeletal remains including skull, lacking mandibles, found high on beach adjacent to low shrubs by R. Chatto. Condylobasal length c. 1000 mm, width c. 500 mm, height c. 400 mm; c. 3 m portion of vertebral column lying in situ. Photographed and a single vertebra collected; retained in Darwin office of PWCNT. Photographs (nos 5499, 5500 and 5501) in PWCNT file P1999/629. Identified by G. J. B. Ross, who commented that it was probably a mature female, based on fusion of the epiphyses to the centra of the vertebrae and on skull characters. Skull collected by R. Chatto on 18 November, 2000; deposited at MAGNT (not registered at time of writing). Beaked whale, probably Ziphius cavirostris Date. Found 2 August 1984. Locality. 12°03’S, 134°55’E. A small island (possibly called Rakuna), just south of Yabooma Island off Milingimbi. Comments. Single animal, length 5.18 m stranded alive but date uncertain. Reported by the Northern Land Council to PWCNT. When examined by P. Whitehead, PWCNT, on 2 August, the carcass had decomposed to the point where the skull and mandibles were ‘clean’ except for some skin at the very tip; no sign of erupted teeth or alveoli and deterioration of the genital/anrd area precluded determination of sex. Specimen photographed and measured, but no necropsy was performed and no material was collected because of the totemic signi¬ ficance of the animal to the Aboriginal traditional owners. Those who saw the whale alive identified it as Z. cavirostris with reference to the illustration of the Indo- Pacific form in Watson (1981:118). Measurements and five photographs (un-numbered) on PWCNT file PI999/629. NCSD #31. Family Physeteridae Physeter macrocephalus - sperm whale Date. Stranded c.l7 November 1980. Locality. 11°39’S, 133°22’E. Off north-west side of South Goulburn Island, on a reef c. 2 km offshore. Comments. Freshly dead when found by a fisherman on 18 November 1980. Lodged among mangroves on South Goulburn Island when investigated by MAGNT staff on 24 November 1980, estimated length 13 m. Reduced to a skeleton by late December. Substances from rotting carcass had killed the mangroves over a radius of 100 m (1. Archibald, MAGNT, pers. comm.). Aboriginal traditional owners would not allow any part of the specimen to be collected. NCSD # 166. See also Hodgkinson (1987) for further details regarding the Aboriginal issues concerning the stranding. Date. Stranded c. 1987. Locality. 14°13’S, 129°25’E. Beach near Port Keats. Comments. Bleached skull observed from aircraft by R. Chatto, 15 December 1992. Advised by Sgt. Kevin Winzar of Port Keats police that the whale washed 9 km to the NNW about 1987. It was not examined and was left to disintegrate on site. Sgt. Winzar photographed the skull on 17 December 1992 and measured the width at 120 cm and height at 70 cm. No further details. Photographs in PWCNT file PI999/629. Date. Stranded 22 January 1993. Locality. 12°2rS, 130°52’E. Casuarina Beach, Darwin. Comments. Single, adult male stranded alive, 15.4 m in length. Grounded in shallows and died soon after. Skeleton collected for display by MAGNT (file BP2000/ 793); not registered in scientific collection. Photographs and report in NT News, 23 January, pp 1,14,15. See also Chatto (2000a) for more detailed description. Date. Stranded between 3 and 6 August 2000. Locality. 11°34’S, 133°05’E. 1 km south of Brodgen Point, east of Murgenella. Comments. Freshly dead male lying in the intertidal zone, length c. 15 m. Found by P. Luton, PWCNT, on 6 August 2000. It had not been there on 3 August 2000. Examined and photographed by R. Chatto on 17 August 2000; no external injuries; 24+ teeth in one mandibular row, several broken. Carcass left on site. Site revisited by R. Chatto on 17 November, 2000. Skull (minus all teeth), some vertebrae and a rib bone still present. Additional measurements and photographs taken, and rib collected and deposited in MAGNT (not yet registered at time of writing). Photographs (nos 6742- 48 and 6844-52) and measurements in PWCNT file PI 999/629. Family Kogiidae Kogia simus - dwarf sperm whale Date. Stranded 19 August 1995. 167 Chatto and Warneke Locality. 12°22’S, 130°5rE. Nightcliff Beach, Darwin. Comments. Adult male, length 2.1 m, stranded alive late afternoon. Towed out to sea and released. Stranded next day on Mindil Beach, where it died during rescue attempts. Photographs and report in NT News, 21 August 1995, p 1. Necropsy performed, photographs and measurements taken; details in PWCNT file PI999/629. Specimen currently in MAGNT freezer (Temporary taxidermy ref. no. 126); to be accessioned as a disarticulated skeleton. See Chatto (2000b) for a detailed description. Family Balaenopteridae Balaenoptera musculus - blue whale Date. Found 19 December 1980. Locality. 12°04’S, 131°18’E. Beach on the east side of Cape Hotham. Comments. Specimen in skeletal condition, estimated length c. 25 m. Found by a fisherman who removed and sold some vertebrae as garden seats. Remainder of skeleton collected by MAGNT staff some months later and put on public display at MAGNT; not registered in scientific collection. Identification confirmed by S. Van Dyck, Queensland Museum, on the basis of skull measurements. NCSD # 165. Megaptera novaeangliae - humpback whale Date. Found 1981. Locality. 11°59’S, 135°49’E. Napier Peninsula, NE Arnhem Eand. Comments. Found dead with a hole in its side, estimated length 10 m. Source of report not stated, no further details. NCSD# 167. Balaenoptera sp. Date. Stranded January 1996. Locality. 13°10’S, 136°17’E. Western point of entrance to Myoola Bay, Gulf of Carpentaria. Comments. Carcass washed ashore, estimated length 10+ m. Found February 1 and photographed by Waka, an Aboriginal traditional owner from the Banyalla community. Carcass lodged near mangroves in an advanced state of decomposition, but numerous throat grooves still evident. R. Chatto visited site with Waka in May 1999, at high tide, but no remains could be seen. No further details. Photograph in PWCNT file P1999/629. Date. Stranded c. 19 April 1999. Locality. 12°24’S, 136°55’E. 15 km south of Cape Arnhem, near Gove. Comments. Decomposed carca.ss, estimated length 10+ m washed ashore on or just before 19 April 1999. Dhimurru Fand Management Aboriginal Corporation provided photographs and video taken 10 May 1999, which show a flipper of moderate length, numerous throat grooves and the skull missing. Remains left on site, and were still present but very decomposed when examined and photographed by R. Chatto on 27 May 1999. No further details. Photographs (nos 6117-6120) in PWCNT file PI999/629, copy of video in Darwin office of PWCNT. Family and species unknown Date. Found February 1984. Locality. I5°40’S, 137°00’E. Vanderlin Island, Sir Edward Pellew Islands. Comments. A group of six whales ashore. Reported by Coastwatch whose observers supposed that they could have been driven ashore by cyclone ‘Kathy’. No further details. Date. Stranded 1 August 1984. Locality. I2°02’S, 134°56’E. Yabooma Island off Millingimbi. Comments. Alive when seen by Coastwatch; length estimated at 4.5 m. Coastwatch observers commented that the weather was rough at the time and described the whale as black with white marks on the head, a colour pattern suggestive of Z. cavirostris. No further details. NCSD# 171. Date, Found April 1985. Locality. 15°36’S, 137°I0’E. 2 nautical miles SSE of Cape Vanderlin, Sir Edward Pellew Islands. Comments. Two animals, possibly a mother and calf. Observer in aircraft thought they were too large for dolphins. Source of report not stated, no further details. NCSD# 172. Date. Found 16 April 1985. Locality. 15°02’S, 135°34’E. North of mouth of Limmen River, Gulf of Carpentaria. Comments. Single animal on beach, dead. Reported by Coastwatch. No further details. NCSD # 176. Date. Found 16 April 1985. Locality. I5°36’S, 136°33’E. West Island, Sir Edward Pellew Islands. Comments. At least one animal on beach, dead. Re¬ ported by Coastwatch. No further details. NCSD # 177. Date. Found 1985 during ‘dry’ season (May to September). Locality. ll‘’20’S, 132°56’E. Just east of De Courcy Head, north of Murgenella. Comments. Decomposing carcass, estimated length 5 m. Observed from aircraft by A. Donati, DPIF. No further details. Date. Found 9 May 1985. Locality. 16°00’S, 137°12’E. 2 nautical miles north of Robinson River mouth, SW Gulf of Carpentaria. Comments. Single animal, decomposing, estimated dead for 2-3 weeks. Length said to be approximately 6- 7 m. Reported by Coastwatch, who said it appeared to be unlike Globicephala and was possibly a baleen whale. However, this is very small for such. No further details. NCSD# 178. Date. Found pre 1990. 168 Whale and dolphin strandings in the Northern Territory Locality. I2°54’S, 130°20’E. Just south of Finniss River mouth. Comments. Skeletal remains of large whale, one rib c. 3 m in length, possibly either a large Physeter or a large baleen whale. Reported to R. Chatto in June 1999 by Aboriginal traditional owner. No further details. Date. Found early 1990s, in month of February. Locality. 12°13’S, 136°15’F. North-east part of Arnhem Bay. Comments. Bleached skeletal remains of a whale c. 5 m in length, but possibly longer. Observed from aircraft by M. Stevens, PWCNT. Reported to R. Chatto, June 1999. No further details. Date. Stranded December 1994. Locality. 11° 57’S, 134° lO’F. West Point, near Maningrida. Comments. Farge individual washed ashore, left untouched. Reported to R. Chatto by Northern Fand Council in 1995. Site visited by R. Chatto on 17 November, 2000. One mandible (no teeth alveoli; curved length 2.97 m), on vertebra and two ribs remaining. The specimen can now be confirmed as a baleen whale. The vertebra was collected and deposited in MAGNT; not yet registered at time of writing. Photographs (nos 6866- 72)and measurements in PWCNT file PI999/629. Date. Stranded January 1998. Locality. II°()2’S, 132°35’F. North-east side of Croker Island. Comments. Five small black whales stranded on beach; four c. 1.5 m and one c. 1.0 m in length. Report by Aboriginal traditional owners via Northern Fand Council to R. Chatto stated that the animals “had a dorsal fin but a blunt nose, not like normal dolphins”. Two were returned to the sea but the others persistently re-stranded, and were left on site. One had cuts to the belly area. No further details. DISCUSSION There is scant information on the occurrence of cetaceans in NT waters. This unsatisfactory situation is the result of many factors, including the remoteness of much of the Territory’s coastline, the sparse and non- randomly distributed human population, the sporadic nature of human activity in inshore waters, poor communication, historical disinterest and continuing low levels of scientific inquiry. These latter two issues are due in part to the cost and logistical difficulties of survey and salvaging specimens from remote areas. Thus when Banister et al. (1996) prepared the first comprehensive review of Australia’s cetaceans, they could confidently list for the Northern Territory only 12 (inclusion of Mesoploclon layardii in their Table I is an error) of the 26 species known to occur in Australia’s tropical waters. They provisionally included two others on the basis of uncertain identification of specimen material. Records based on strandings, accumulated from a variety of sources and summarised here, include ten of those 12 species, while the two provisionally included {Globicephala macrorhynclnts, Kogia sirniis), are now confirmed. The melon-headed whale Peponocephala elecira, can now also be added to the Northern Territory list. Two species listed from the NT by Bannister et al. (1996), but yet to be recorded by strandings are the rough-toothed dolphin Steno brenadensis and common dolphin Delphinus delphis. In addition, another species can now be confirmed for NT waters and an additional species to Common¬ wealth waters off the NT coast. Confirmation of the killer whale, Orcinus orca, in NT waters is based on a sighting made on 7 April 1999 at Yirrkala(12° 15’S, 136° 54’F) by C. Fane, and reported by the Dhimurru Fand Management Aboriginal Corporation. At least one animal was seen from shore at around 12.30 h approx¬ imately I km out to sea and heading south. The animal was described as 3-4 times the size of a dolphin, strikingly marked with black on top and white below, and with a very large dorsal fin. (The dolphins most frequently seen in this area are Sousa cbinensis and Tursiops truncatus cf. aduncus, i.e. individuals about 2.5 m in length (Chatto, pers. obs.).) One old Aboriginal lady said that she had not seen this species for 20 years, but does remember seeing them from time to time when her children were young. Confirmation of the sei whale, Balaenoptera borealis, in Commonwealth waters off the NT coast is based on a sample of baleen from a dead specimen c. 15 m in length, trawled up from 113 m by a Thai fishing vessel on 1 September 1989 at 10° 21’S, 133° 55’F, approximately 160 kilometres north-east of Croker Island. The specimen of baleen was donated to MAGNT where it is on display but it has not been registered in the scientific collection. Of the total series of 57 stranding events reported on here (Table 1), 35 were identifiable to species, seven to genus {Globicephala, Balaenoptera), and the remainder could be described only as ‘dolphin’ (3), ‘beaked whale’ (1) or ‘unknown’ (II). This series is loo small and the data available for most events too meagre for intra- and inter-species compa¬ risons. Similarly, there are too many confounding factors and biases to search for any patterns in geographic distribution of the 57 events. For instance, there is no way of telling whether any of the singletons found dead were alive when they beached, or if they died at sea and their floating carcasses had been transported by current, wind and tide over considerable distances before they eventually washed ashore. However, it is appropriate to consider the results of this study in terms of the known distributions of these 13 species and their habitat preferences, seasonal movements or migrations. We have grouped them according to broad oceanographic categories based on 169 Chatto and Warneke Table 1. Species, number of stranding events and number of individuals per event Species Common name No. of stranding events No. of individuals per event Balaenoptera musculus blue whale 1 1 Balaenoptera sp. baleen whale sp. 2 1 (x2) Globicephala macrorhynchus short-finned pilot whale 1 5 Globicephala sp. pilot whale sp. 5 1,3,8, 20,30-50 Kogia siina dwarf sperm whale 1 1 Megaptera novaeangliae humpback whale 1 1 Orcaella brevirostris Irrawaddy dolphin 5 1 (x4), 2 Peponocephala electro melon-headed whale 2 1,40 Physeter macrocephalus sperm whale 4 1 (x4) Pseudorca crassidens false killer whale 2 1 (x2) Sousa chinensis Indo-Pacific hump-backed dolphin 5 1 (X 4), 3 Stenella attenaata pan-tropical spotted dolphin 1 1 Stenella longirostris spinner dolphin 3 1, 1,5 Tiirsiops truncatus cf. aduncus bottlenose dolphin 7 1 (x7) Ziphius cavirostris Cuvier’s beaked whale 2 1 (x2) 'beaked whale' beaked whale sp. 1 1 'dolphin' dolphin sp. 3 1 (x3) 'unknown cetacean' unknown cetacean sp. 11 1 (x7), l+, 2,5,6 water temperature and commonly used to define distributions, as follows. Tropical - subtropical. Includes two inshore species (Indo-Pacific hump-backed dolphin and Irrawaddy dolphin), and two offshore species (melon-headed whale and the spinner dolphin), both of which range into coastal waters. The Indo-Pacific hump-backed dolphin is restricted to coastal, inshore and estuarine habitats throughout northern Australia, ranging south along the WA coast to about 24°S, and to Moreton Bay in Queensland, at 27°30’S (Bannister et al. 1996). Stragglers have been recorded as far south as Sydney (Llewellyn et al. 1994). This species is not known to be migratory, but individuals may move about a great deal along-shore and between islands. It appears to be loosely social in small groups (Ross et al. 1994) and is not prone to strand. This species is often seen in very shallow water in large tidal variation areas and does not become stranded (Chatto pers. obs.). G. Ross (pers. comm.) has also seen this species school fish up onto sandbanks and then slide up to grab them without becoming stuck. Thus all five NT records listed here probably represent mortality at sea rather than active strandings. The Irrawaddy dolphin has a similar but slightly more restricted distribution in northern Australia, occurring on the WA coast north of Broome (18° S), throughout the NT and to Gladstone (23° 50’S) on the Queensland coast (Banister et al. 1996). It is usually encountered in groups of less than 6 animals, but groups of 10-15 animals have been reported (Marsh et al. 1989). As this species is not prone to strand, the five NT records listed here probably also represent mortality at sea. The spinner dolphin inhabiting NT waters has recently been described as a subspecies - Stenella longirostris roseiventris (Perrin et al. 1999). It is a small form restricted to the shallow inner waters of South-east Asia, including the Gulf of Thailand, Timor and Arafura Seas, and similar waters of Indonesia, Malaysia and northern Australia; it is not known to be migratory. In deeper oceanic waters it is replaced by the typical and larger pelagic form, S. 1. longirostris (Perrin and Gilpatrick 1994). Little is known of the occurrence of spinner dolphins in northern Australia, even though they were commonly caught by accident in the Taiwanese gillnet fishery for sharks in the Timor and Arafura Seas (Harwood and Hembree 1987). Stenella /. roseiventris is known, however, to feed on shallow water benthic and reef dwelling fish, squid, cuttlefish and shrimp (G. Hembree, cited in Perrin et al. 1999). More southern records, to Bunbury in WA (30°19’S, Bannister et al. 1996) , from the Great Barrier Reef and in NSW to Uranga (30°31’ S, Llewellyn et al. 1994) appear to be of the pelagic form. The close grouping of the three NT records listed here over less than 1 degree of longitude, and including a mass stranding, may indicate a centre of activity for S. 1. roseiventris in that region, as there is no reason to suspect it may be attributable to greater observer activity. The melon-headed whale is a highly social deep-water species, occurring in large schools and often associated with other oceanic dolphins and small whales (Leather- wood and Reeves 1983; Perryman et al. 1994). Schools have been sighted over the continental slope off northern NSW (Dawbin et al. 1970) and several mass strandings have occurred in southern Queensland (Bryden et al. 1997) and NSW (Dawbin et al. 1970; Smith 1997). The small .school that stranded on Elcho Island in the NT in 1996 may have become disorientated when navigating in unfamiliar shallows, however little is known about 170 Whale and dolphin strandings in the Northern Territory this species which may not be restricted to deep waters only (Ross pers. comm.). Tropical - temperate. Four ecologically very different species fit this general category - pan-tropical spotted dolphin, bottlenose dolphin, false killer whale and dwarf sperm whale. The spotted dolphin occurs circumglobally in deep offshore waters, most frequently where sea surface temperatures exceed 25° C and overlie a sharp thermo- cline at depths less than 50 m (Perrin and Hohn 1994). It is often encountered in large schools and is often associated with spinner dolphins. In Australia, spotted dolphins are known to occur north of Augusta in WA (38° 20’S) and north of Sydney on the east coast (34° GO’S, Bannister etal. 1996). Their activity patterns in Australian waters are not known, but elsewhere, seasonal inshore-offshore movements occur in spring (Leatherwood and Reeves 1983). The timing of the one NT record of a compromised adult fits this pattern. The adiincus form of bottlenose dolphin is very widely distributed in tropical, subtropical and temperate Australian waters, ranging south to about 32° S. Where it occurs sympatrically with the hump-backed dolphin it tends to be active slightly further offshore, but often in depths of less than 10 m, and may range to about 10 km beyond the shelf (Bannister et al. 1996). Because bottlenose dolphins navigate confidently in shallow and confined waters, the seven NT records, all singleton events, probably represent mortality at sea. This appears to be the case with the typical Inincalus form in Victoria where many events involve aged, diseased, injured and very young animals (Warneke, unpublished data). The false killer whale is an oceanic species, rarely approaching land except where the continental shelf is narrow. In Australia, false killers have been widely recorded by some sightings and strandings in all slates. Strandings occur throughout the year, but the majority of mass events occur from May to September on south and south-east coasts, indicating a seasonal movement inshore or along the continental shelf (Bannister et al. 1996), possibly associated with abundances of prey. The two singleton events in the NT are unrevealing, but probably represent mortality at sea. The short-finned pilot whale occurs throughout tropical and warm temperate regions of Indian, Pacific and Atlantic Oceans, but the southern limits of its range are poorly known, partly because of the difficulty of distinguishing it from the closely similar long-finned pilot whale, Globicephala melas, at sea. The short-finned pilot whale appears to be widely distributed off Australia’s northern coast where it ranges into con¬ tinental seas. Records from temperate waters on the southern coast (South Australia, Tasmania) probably reflect the influence of the south-flowing, warm Leeuwin Current and/or East Australia Current (Bannister et al. 1996). Because G. melas is restricted to temperate waters (the most northern record in Australia is of a straggler at Point Lookout in Qld (27° 26’S, Paterson 1986) it is reasonable to assume that the five NT ‘pilot whale’ records are of this species, making six in total. The high proportion of school strandings (5) in this series and their wide geographic spread indicate considerable activity in NT waters and that this species is at high risk when navigating near the shore in those areas. The large and fast tidal variations of the NT may also be influential in this regard. The dwarf sperm whale is primarily oceanic and occurs world wide within about 40° N and S of the equator. The frequency and distribution of strandings in some regions appears to correlate with continental shelf width and important current systems, and suggests that the species might be more abundant in the warmer parts of its range (Caldwell and Caldwell 1989). It feeds primarily on benthic and demersal organisms. Dietary studies in southern Africa (Ross 1984) indicate that Juveniles and immatures are active over the outer continental shelf and slope where suitable prey is often more abundant and that adults feed in deeper waters beyond the shelf. Comparative data suggest that dwarf sperm whale is more active nearer to coasts than the pygmy sperm whale, Kogia breviceps, but in Australia the former strands very rarely by comparison, with single records only in WA, SA, NSW and now the NT. Tropical - subantarctic. Cuvier’s beaked whale occurs world-wide within this broad latitudinal range. It is a true oceanic species, normally ranging far from land. It is thought to be resident in some areas, e.g. off Japan, where it is most common in waters deeper than 1000 m (Nishiwaki and Oguro 1972). In the Australian region it is known from a few strandings in all mainland states, but many more (15) in Tasmania (Bannister et al. 1996). Strandings have been recorded in all months except September to January, but this indication of seasonality of occurrence may be an artefact of the small series of Just 12 dateable events. No pattern of seasonality has been detected in strandings elsewhere (Heyning 1989). The two certain and one probable NT events listed here suggest that the species is not uncommon in the region, but otherwise are unrevealing. Tropical - Antarctic. Sperm whales range widely in all oceans and tend to concentrate in the vicinity of steep continental shelves, oceanic islands and sea mounts where favoured prey are more abundant. Breeding females form stable nursery schools, while adult males are solitary or associate in temporary bachelor schools. Judging by the frequency of strandings, sperm whales are common in Australian waters. The majority of events have occurred in the south-east and most involved single animals, 75% of which were male. In striking contrast mass strandings are rare, except in Tasmania (15 events, Warneke, unpublished data). The series shows a strong seasonal bias, with most singleton events occurring in Chatto and Warneke spring to autumn, and mass events in January - March, September and October. While this pattern accords with the known generalised open ocean movement of the species southwards in summer (Rice 1989; Bannister et al. 1996), very little is known of its activities near-shore. The four NT singleton events listed here of known date indicate that sperm whales, probably only adult males, are present in the region during summer. Two forms of blue whale are recognised - the world wide ‘true’ blue, Balaenopterci miisculus intermedia, and the southern hemisphere pygmy blue, B. in. brevicauda, which is most abundant in the Indian Ocean (Yochem and Leatherwood 1985). Both forms migrate between warm water breeding grounds at low latitudes and cold water feeding grounds at high latitudes - the blue penetrating to the Antarctic ice edge, whereas the pygmy blue normally feeds north of 55° S. In recent years blue whales have been encountered relatively clo.se to the Australian coast at various places, mainly in southern waters from Rottnest Island WA to southern NSW; the most northern records being sightings off the Dampier Archipelago in WA at about 20° S (Bannister et al. 1996) and a stranding at Couti Uti in Qld (22°20’S, Paterson 1986). The single confirmed record for the NT extends the known range in the Australian region by 8 degrees of latitude. At the time this paper was going to print, a 21 m blue whale washed up dead on to Troughton Island (13°45’S, 126°09’E) in NW Western Australia on November 7, 2000 (Chatto pers. obs.). Two relatively discrete Australian populations of humpbacked whales undertake a seasonal north-south migration similar to that of the blue whale, but between breeding grounds on either side of the continent at about 15-20°S and feeding grounds at about 60-70°S; some individuals may remain in the tropics, e.g. in Torres Strait (Bannister et al. 1996). Migrating animals generally pass close to the coast at various points, and stragglers may enter bays and inlets where they navigate confidently in confined shallow waters, e.g. in Victoria (Warneke 1995a). Consequently strandings are rare and apart from live animals entangled in netting they almost invariably involve individuals that have died at sea from natural causes. The NT stranding appears to be one such natural event. Other species likely to strand in the Northern Territory. Finally, it remains to consider what species might in future strand in NT waters and to emphasise the importance of investigating as many events as possible to obtain authentic identifications and to document little known aspects of the species’ biology. Bannister et al. (1996) lists 14 species that are widely distributed in the tropical region and are likely to range into the tropical waters of the Australian region, but which have not been authentically recorded in the NT. We provide brief comment on each of these, as well as the rough-toothed dolphin, killer whale and Bryde’s whale not included in the review above. The rough-toothed dolphin, Steno bredanensis, is widely distributed in deep offshore waters in tropical to warm temperate latitudes, including the Indo-Australian Archipelago, but it is not known to be numerous. It is social and gregarious, and associates with pilot whales and bottlenose dolphins (Leatherwood and Reeves 1983). It was listed for the NT by Bannister et al. (1996). Risso’s dolphin. Grampus griseiis, is abundant and cosmopolitan in tropical to warm temperate waters (Leatherwood and Reeves 1983) and although an oceanic species it is frequently seen over the continental slope and ranges onto the shelf in parts of Australia (e.g. Corkeron and Bryden 1992). The striped dolphin, Stenella coeruleoalba, is widely distributed in deep waters in tropical to warm temperate latitudes, ranging to the outer edge of the continental slope. It is social and gregarious, and is known to be migratory in some regions (Perrin et al. 1994). In Australia, southernmost records are strandings and may be related to the south-flowing warm Leeuwin and East Australia Currents - at Augusta WA (34° 20’S, Bannister et al. 1996) and at Woolongong NSW (34° 23’S, Llewellyn et al. 1994). The common dolphin, Delphiniis delphis, is very widely distributed and abundant in tropical to temperate latitudes, but there are very few records from Australian tropical waters (Bannister et al. 1996), which may reflect its preference for areas of high topographical relief and upwellings (Evans 1994) and the very wide continental shelf around northern Australia. The species is often encountered over the narrow continental shelf in south¬ eastern Australia, where many singleton strandings have been recorded (Nicol 1987; Kemper and Ling 1991; Warneke 1995b); mass strandings are rare, except in Tasmania. The species Delphiniis iropicalis is more likely to occur in waters off the NT coast than D. delphis (Ross pers. comm.). Fraser’s dolphin, Lagenodelphis hosei, is abundant and widely distributed in oceanic tropical waters (Perrin, Leatherwood and Collet 1994) and could be expected in waters off the NT coast (Ross pers. comm). This species occasionally strays into temperate regions, where single and mass strandings have occurred, e.g. in NSW (Llewellyn et al. 1994) and Victoria (Warneke 1995c). The pygmy killer whale, Feresa attenuata, is primarily tropical in distribution; records from higher latitudes being associated with warm boundary currents (Ross and Leatherwood 1994). In Australia it is known from sightings in north-eastern waters and several strandings in NSW (Llewellyn et al. 1994) and WA (Bannister ct a/. 1996). The killer whale, Orcinus area, is cosmopolitan, occurring from the equator to polar seas. In Australia 172 Whale and dolphin strandings in the Northern Territory this species is often seen along the continental slope and on the shelf, and has been recorded from all states, but more commonly in the southern waters (Bannister et al. 1996). Killer whales navigate confidently in very shallow waters, consequently live strandings are rare. Several singleton events in Victoria and Tasmania involved very old and juvenile individuals and one neonate that had died at sea, or were moribund when they beached (Warneke, unpublished data), ie they were instances of natural mortality. Similar events can be expected in the NT. Longman’s beaked whale, Indopacetiis pacificus (Moore \ 96i){-Mesoplodonpacificushongmm 1926), is known only from two skulls, one from Mackay, Qld, at 21‘’09’S and the other from Danane, Just north of the equator in Somalia — points which suggest an Indo- Pacific distribution. At an estimated length of c. 7.5 m it is by far the largest me.soplodontid. Intriguingly during the past three decades there have been numerous sightings of large beaked whales very similar in form to the southern bottlenose whale Hyperoodon planifrons (see below), in tropical waters of the Indian and Pacific Oceans. A recent review of these sightings and some associated photographs by Pitman et al. (1999) indicates that this ‘tropical bottlenose whale’ differs from H. planifrons in several respects and that it is almost certainly the elusive I. pacificus. Any report of a large beaked whale stranded in the NT should be promptly investigated. Blainville’s beaked whale, Mesoplodon densirostris, is an oceanic species widely distributed in tropical to temperate waters world wide. In Australia it is known only from strandings, in most of the southern states and from WA and Qld south of 20° lO’S (Mead 1989). Records in NSW, Victoria and on the west coast of Tasmania may be of stragglers ranging southwards in the warm Leeuwin and East Australia Currents. The southern bottlenose whale, Hyperoodon plani¬ frons, is included here only on the basis of the proximity of a record (the type specimen) from Lewis Island in the Dampier Archipelago (20° 35’S) in WA. The normal range of this species is considered to be south of 30“ S to the Antarctic ice edge (Bannister et al. 1996). Although commonly and widely encountered in deep oceanic waters (Ross 1984) this species appears to avoid continental seas and strands infrequently (Baker 1983). The pygmy sperm whale, Kogia breviceps, has a similar world wide distribution to its congener K. siniiis, but tends to remain in deeper waters further offshore (Caldwell and Caldwell 1989). However, in Australia K. breviceps strands much more frequently and most records are concentrated on the east and south-east coasts, especially in NSW (38 records to 1992, Llewellyn et al. 1994), possibly because of the relatively narrow continental shelf in much of that region. Two rorquals - minke whale, Balaenoptera acuto- rostrata, and Bryde’s whale, B. edeni, are known from tropical waters. In the Australian region their status and movements are not well understood, and the situation is further complicated by the presence of two forms of each species. An ‘inshore’ Bryde’s whale is confined to the tropics, whereas a larger ‘offshore’ form ranges into temperate waters, where it has been recorded as far south as Bass Strait (Dixon and Frigo 1994). Similarly, a diminutive form of minke occurs mainly in the tropical - subtropical region north to at least 12“ S, although there are records as far as 58-65“ S, while a ‘dark shoulder form’, B. bonaerensis, ranges not quite so far north in winter as the diminutive form (recorded at 21° S), but migrates to the Antarctic ice edge during summer (Banni.ster et al. 1996). Two other migratory rorquals - the fin whale, B. physalus, and the sei whale, B. borealis, overwinter in subtropical to temperate waters in the Australian region, but they tend to remain well offshore and it is not known whether either species ranges into NT waters (Bannister etal. 1996); in any case they rarely strand. ACKNOWLEDGMENTS Environment Australia, Canberra, provided data from the National Cetacean Stranding Database. Several strandings were reported by Coastwatch, Darwin, the Northern Land Council, Darwin, and the Dhimurru Land Management Aboriginal Corporation, Nhulunbuy, NT. Access to specimens and/or additional reports of strandings were provided by Paul Horner, Ian and Jared Archibald of the Museums and Art Galleries of the Northern Territory; Stuart Woerle, Mark Stevens and Alan Withers of the Parks and Wildlife Commission of the NT; Bryan Walsh, Frank Woerle and Peter Whitehead, formerly with the Parks and Wildlife Commission of the NT; Adrian Donati of the Department of Primary Industries and Fisheries of the NT; and Sergeant Kevin Winzar, Port Keats Police. Dr Catherine Kemper, South Australia Museum, Dr Stephen Van Dyck, Queensland Museum and Dr Graham Ross, Australian Biological Resources Survey, Canberra, confirmed some of the identifications. Dr Peter Whitehead, Dr John Woinarski and Dr David Cheal of the Parks and Wildlife Commission of the NT, Palmerston, commented on the manuscript. REFERENCES Baker, A.N. 1983. Whales and dolphins of New Zealand and Australia: an identification guide. Victorian University Press: Wellington, New Zealand. Bannister, J.L., Kemper, C.M. and Warneke, R.M. 1996. The action plan for Australian cetaceans. Australian Nature Conservation Agency, Wildlife Australia Endangered Species Program, Project Number 380: Canberra. 173 Chatto and Warneke Bryden, M.M., Dawbin, W.H., Heinsohn, G.E. and Brown, D.H. 1997. Melon-headed whale, Peponocephala electra, on the east coast of Australia. Journal of Mammalogy 58: 180- 187. Caldwell, D.K. and Caldwell, M.C. 1989. Pygmy sperm whale Kogia breviceps (dc Blainville, 1938): dwarf sperm whale Kogia sima Owen, 1866. In: Ridgway, S.H. and Harrison, R. (eds) Handbook of marine mammals. Vol 4 River dolphins and the larger toothed whales. Pp 235-260. Academic Press: London. Chatto, R.E. 2000a. Whale strandings in the Northern Territory I. Great sperm whale Physeter macrocephalus on Casuarina Beach, Darwin. NT Naturalist 16: 14-15. Chatto, R.E. 2000b. Whale strandings in the Northern Territory II. Dwarf sperm whale Kogia sima. NT Naturalist 16: 15- 16. Chatto, R.E. 2000c. Whale strandings in the Northern Territory III. A mass stranding of melon-headed whales Pepono¬ cephala electra on Elcho Island. NT Naturalist 16: 16-18. Chatto, R.E. 2000d. Whale strandings in the Northern Territory IV. A mass stranding of short-finned pilot whales Globicephala macrorhynchus on Melville Island. NT Naturalist 16: 19-20. Corkeron, P.J. and Bryden, M.M. 1992. Sightings of Risso’s dolphin. Grampus griseiis (Cetacea: Delphitiidae), off Fraser Island, Queensland. Australian Mammalogy 15: 129- 130. Dawbin, W.H., Noble, B.A. and Fraser, F.C. 1970. Observations of the electra dolphin, Peponocephala electra. Bulletin of the British Museum of Natural History 20: 173-201. Dixon, J.M. and Frigo, L. 1994. The cetacean collection of the Museum of Victoria. An annotated catalogue. Australian Deer Research Foundation: Melbourne. Evans, W.E. 1994. Common dolphin, white-bellied porpoise Delphinus delphis Linnaeus, 1758. In: Ridgway, S.H. and Harrison, R. (eds) Handbook of marine mammals. Vol 5. The first book of dolphins. Pp. 191-224. Academic Press: London. Harwood, M.B. and Hembree, D. 1987. Incidental catch of small cetaceans in the offshore gillnet fishery on northern Australian waters: 1981-1985. Report to the International Whaling Commission. SC/38/SM14. Heyning, J.E. 1989. Cuvier’s beaked whale Ziphius cavirostris G. Cuvier, 1823. In: Ridgway, S.H. and Harrison, R. (eds) Handbook of Marine Mammals. Vol 4. River dolphins and the larger toothed whales. Pp 289-308. Academic Press: London. Hodgkinson, F. 1987. Frank Hodgkinson’s Kakadu and the Arnhem Landers. Weldons Pty Ltd: McMahon’s Point, NSW. Johnson, D.H. (1964). Mammals of the Arnhem Land expedition. In: Specht, R.L. (ed.) Records of the American-Australian scientific expedition to Arnhem Land. Vol. 4. Zoology. Pp. 427-515. Melbourne University Press: Victoria. Kemper, C.M. and Ling, J.K. 1991. Whale strandings in South Australia (1881-1989). Transactions of the Royal Society of South Australia 115(1): 37-52. Leatherwood, S. and Reeves, R.R. 1983. The Sierra Club handbook of whales and dolphins. Sierra Club Books: San Francisco. Leatherwood, S., Reeves, R.R., Perrin, W.F. and Evans W.E. 1982. Whales, dolphins, and porpoises of the eastern North Pacific and adjacent Arctic waters. NOAA Technical Report NMFS Circular 444. Llewellyn, L., Ellis, M., Martin, J. and Ferguson, A. 1994. Atlas of New South Wales wildlife: marine mammals and reptiles. NSW National Parks and Wildlife Service: Sydney. Longman, H.A. 1926. New records of Cetacea with a list of Queensland species. Memoirs of the Queensland Museum 8: 266-278. Marsh, H., Lloze, R., Heinsohn, G.E. and Kasuya, T. 1989. Irrawaddy dolphin Orcaella brevirostris (Gray, 1866). In: Ridgway, S.H. and Harrison, R. (eds) Handbook of marine mammals. Vol 4. River dolphins and the larger toothed whales. Pp. 101-118. Academic Press: London. Mead, J.G. 1989. Beaked whales of the genus Mesoplodon. In: Ridgway, S.H. and Harrison, R. (eds) Handbook of marine mammals. Vol 4. River dolphins and the larger toothed whales. Pp. 349-430. Academic Press: London. Moore, J.C. 1968. Relationships among the living genera of beaked whales, with classification, diagnoses and keys. Fieldiana Zoology 53: 206-298. Nicol, D.J. 1987. A review and update of the Tasmanian cetacean stranding record to the end of February 1986. University of Tasmania, Centre for Environmental Studies, Envi¬ ronmental Studies Working Paper 21. Nishiwaki. N. and Oguro, N. 1972. Catch of Cuvier’s beaked whales off Japan in recent years. Scientific Reports of the Whales Research Institute 24: 35-41 Paterson, R.A. 1986. A list of specimens of the order Cetacea in the Queensland Museum. Memoirs of the Queensland Museum 22(2): 309-311. Perrin. W.F. and Gilpatrick, J.W. 1994. Spinner dolphin Stenella longirostris. In: Ridgway, S.H. and Harrison, R. (eds) Handbook of Marine Mammals. Vol 5. The first book of dolphins. Pp. 99-128. Academic Press: London. Perrin, W.F. and Hohn, A.A. 1994. Pantropical spotted dolphin Stenella attenuata. In: Ridgway, S.H. and Harrison, R. (eds) Handbook of marine mammals. Vol 5. The first hook of dolphins. Pp. 71-98. Academic Press: London. Perrin, W.F, Wilson, C.E. and Archer II, F.I. 1994. Striped dolphin Stenella coeruleoalba Meyen, 1833. In: Ridgway, S.H. and Harrison, R. (eds) Handbook of marine mammals. Vol 5. The first hook of dolphins. Pp. 129-159. Academic Press: London. Perrin, W.F, Leatherwood, S. and Collet, A. 1994. Fraser’s dolphin Lagenodelphis hosei Fraser, 1956. In: Ridgway, S.H. and Harrison, R. (eds) Handbook of marine mammals. Vol 5. The first hook of dolphins. Pp. 225-240. Academic Press: London. Perryman, W.L, Au, D.W.K.. Leatherwood, S. and Jefferson, T.A. 1994. Melon-headed whale Peponocephala electra Gray, 1846. In: Ridgway, S.H. and Harrison, R. (eds) Handbook of marine mammals. Vol 5. The first book of dolphins. Pp. 363-386. Academic Press: London. Pitman, R.L., Palacios, D. M., Brennan, P.L.R., Brennan, B.J., Balcomb, K.C. and Miyashita, T. 1999. Sightings and possible identity of a bottlenose whale in the tropical Indo- Pacific: Indopacetus pacificusl Marine Mammal Science 15(2): 531-549. Rice, D.W. 1989. Sperm whale Phy.seter macrocephalus Linnaeus, 1758. In: Ridgway, S.H. and Harrison, R. (eds) Handbook of marine mammals. Vol 4. River dolphins and the larger toothed whales. Pp. 177-233. Academic Press: London. Ross, G.J.B. 1984. The smaller cetaceans of the south east coast of southern Africa. Annals of the Cape Provincial Museums (Natural History) 15(2): 173-410. Ross, G.J.B, Heinsohn, G.E. and Cockcroft, V.G. 1994. Humpbacked dolphins Sousa chinensis (Osbeck, 1765), Sousa plumhea (G. Cuvier, 1829) and Sousa teuszH (Kukcnthal, 1892). In: Ridgway, S.H. and Harrison, R. 174 Whale and dolphin strandings in the Northern Territory (eds) Handbook of marine mammals. Vol 5. The first book of dolphins. Pp. 23-42. Academic Press: London. Ross, G.J.B. and Leatherwood, S. 1994. Pygmy Killer Whale Feresa atteniiata Gray, 1874. In: Ridgway, S.H. and Harrison, R. (eds) Handbook of marine mammals. Vol 5. The first book of dolphins. Pp. 387-404. Academic Press: London. Smith, P. 1997. Management manual for marine mammals in NSW. NSW National Parks and Wildlife Service: Sydney. Warneke, R.M. 1995a. Humpbacked whale. In: Menkhorst, P.W. (ed.) Mammals of Victoria. Distribution, ecology and conservation. Pp. 302-303. Oxford University Press and Department of Conservation and Natural Resources: Melbourne. Warneke, R.M. 1995b. Common dolphin. In: Menkhorst, P.W. (ed.) Mammals of Victoria. Distribution, ecology and conservation. Pp. 282-283. Oxford University Press and Department of Conservation and Natural Resources: Melbourne. Warneke, R.M. 1995c. Fraser’s dolphin. In: Menkhorst, P.W. (ed.) Mammals of Victoria. Distribution, ecology and conser¬ vation. P. 286. Oxford University Press and Department of Conservation and Natural Resources: Melbourne. Watson, L. 1981. Sea Guide to whales of the world. Hutchinson: London. Yochem, P.K. and Leatherwood, S. 1985. Blue whale Ba- laenoptera musculus (Linnaeus, 1758). In: Ridgway, S.H. and Harrison, R. (eds) Handbook of marine mammals. Vol 3. The sirenians and baleen whales. Pp. 193-240. Academic Press: London. Accepted 14 November 2000 175 i GUIDE TO AUTHORS A comprehensive style guide is available from the Editor on request (e-mail dirk.megirian@nt.gov.au, or at the Editorial postal address provided on the inside front cover). Manuscripts not in The Beagle style will he returned to authors for revision. Three hard copies of word-processed manuscript plus a 3(4” computer floppy disk or CD ROM (MS/IBM DOS compatible) should be submitted in the first instance. Manuscripts should be in English, double-spaced throughout, and have a margin of at least 3 cm on the left-hand side. Text should be printed on one side of good quality A4 bond paper. Where appropriate, articles should conform to the sequence: Title, Author’s name and address. Abstract, Keywords, Introduction, Materials and Methods, body text under appropriate headings. Discussion, Conclusions, Acknowledgments, References. The Title should be concise and informative. The Abstract should not exceed 150 words, and should state concisely the scope of the work and give the principal findings. Keywords to facilitate information retrieval, of up to 15 in number, should be chosen to outline the main subjects covered. The Introduction, including a review of literature, should not exceed what is necessary to indicate the reason and significance of the work, and to provide the essential background. Abbreviations used may be explained at the end of the Introduction, or placed separately in a Materials and Methods section. The International System of units should be used. In the descriptive text, numbers from one to nine should be spelt out and numerals used for numbers over nine. For associated groups, numbers should be expressed consistently, e.g. 5 to 10, not five to 10. Systematic papers must conform to the International Code of Zoological Nomenclature (4th Edition, 1999) and should follow the recommendations of the Commission. Synonymies should be given in the short form (taxon author, date: page) and the full reference cited in the References section at the end of the paper. Subsequent citations of taxa given in synonymies should he separated from bibliographieal details by a short dash (-) as in the following example: Bougainvillia balei Stechow, 1924: 58. - Stechow 1925: 199, fig. B. - Watson 1996: 78. (Note that a comma is only u.sed between the original author and date of the taxon.) Full bibliographic details of taxa used in the text (i.e. taxon author, date) must also be included in the References. TABLES Tables should be numbered with arabic numerals and accompanied by a caption. Horizontal rules are inserted only above and below column headings and at the foot of the table. Footnotes to tables should be kept to a minimum and be reserved for specific items in columns. All other explanatory material should be incorporated in the caption. ILLUSTRATIONS Line drawings, maps, graphs and photographs are cited as ‘figures’ and are to be numbered consecutively with Arabic numerals for interspersion through the text. The author’s name, abridged title of paper and figure number must be indicated on the reverse side of all illustrations. Captions should be typed together on a separate page, or pages, at the end of the text. Black and white photographs must be sharp, of high contrast on glossy paper, and mounted on flexible card. Colour illustrations may be accepted, and in all but the most exceptional cases, the author will be asked to bear costs of colour production. Authors wishing to publish in colour should correspond with the Editor before submission. Digital images are acceptable, and are preferred. However, authors intending to submit digital images should correspond with the Editor before submission about acceptable file formats and other technical details. Drawings must be on drawing film or good quality flexible card with appropriate lettering inserted. Figures .should be suitable for scanning; i.e. not requiring any additional artwork or image processing by the journal to make them publishable. Maps, anatomical drawings, photomicrographs, etc. should be supplied with a labelled scale bar. Lettering and numbering on figures should be in a clear, sans serif font (e.g. Arial, Helvetica), and of a suitable pitch that ensures that when the figure is reduced to page or column size, text will be not less than 8 pitch. Line-work should be of sufficient thickness to withstand reduction. An allowance for a c. 25% reduction is desirable for drawings, as this results in a crisp reproduction. CITATIONS AND REFERENCES Citations of literary sources within the body of the text should include the author, year of publication and, where appropriate, page or figure references, e.g. Roth (1896: 23), (Roth 1896), (Roth 1896, 1898), (Roth 1896; Smith 1915; fig. 1). Citations should be ordered chronologically. Note that commas are not used in bibliographical citations (e.g. Roth 1896). References should be arranged alphabetically and chronologically at the end of the paper. Minimal capitalisation should be applied to article and book titles. Journal titles must be given in full and wherever possible in BIOSIS format. Where an author has cited more than one work by the same author in any one year, the references should be appended with the letter (a), (b), etc. The following examples show the style to be followed: Brake, B., McNeish, J. and Simmons, D. 1979. Art of the Pacific. Oxford University Press: Wellington, New Zealand. Callen, R.A. 1984. Clays of the palygorskile-sepiolite group: depositional environment, age and distribution. In: Singer, A. and Galan, E. (eds) Palygorskite-sepioUte occurrence, genesis and uses. Pp 1-38. Elsevier: Amsterdam. Crowley, L.M. 1949. Working class conditions in Australia, 1788-1851. Unpublished PhD thesis. University of Melbourne. Sadlier, R.A. 1990. A new .species of scincid lizard from western Arnhem Land, Northern Territory. The Beagle. Records of the Northern Territory Museum of Arts and Sciences 7(2); 29-33. The Beagle Records of the Museums and Art Galleries OF THE Northern Territory Volume 16, December 2000 CONTENTS ^ WATSON, J.E. - Hydroids (Hydrozoa; Leptothecatae) from the Beagle Gulf and Darwin Harbour, northern Australia. 1 BRUCE, A. J. - The rediscovery of Leandrites stenopus Holthuis, 1950 (Crustacea: Palaemoninae), from Lucinda, Queensland.83 BRUCE, A. J. - Onycocaridella priina Bruce, 1981, a rare pontoniine shrimp from Darwin Harbour (Crustacea: Decapoda: Pontoniinae).89 BRUCE, A. J. - Biological observations on the commensal shrimp Paranchistus armatus (H. Milne Edwards) (Crustacea: Decapoda: Pontoniinae).91 VON HAGEN, H.-O.- Vibration signals in Australian fiddler crabs - a first inventory.97 BROWN, G. R. - Caetrathynnus, Nilidothymnis and Procerothynnus, new genera of Thynninae (Hymenoptcra: Tiphiidae) from northern Australia. 107 MACHIDA, Y. - A new bythitid genus and species, Acarobythites larsonae, from shallow rocky reefs off northern Australia (Pisces, Ophidiiformes, Bythitidae). 123 KAILOLA, P. J. - Six new species of fork-tailed catfishes (Pisces, Teleostei, Ariidae) from Australia and New Guinea. 127 MURRAY, P. and MEGIRIAN, D. Two new genera and three new species of Thylacinidae (Marsupialia) from the Miocene of the Northern Territory, Australia. 145 CHATTO, R. and WARNEKE, R. M. - Records of cetacean standings in the Northern Territory of Australia. 163