A NEW GENUS OF AMORBINI (HETEROPTERA: COREIDAE) FROM AUSTRALIA, WITH TWO NEW SPECIES H. BRAILOVSKY AND G.B. MONTEITH Brailovsky, H. & Monteith, G.B. 1998 06 29: A new genus of Amorbini (Heteroptera: Coreidae) from Australia, with two new species. Memoirs of the Queensland Museum 42(2): 379-385. Brisbane. ISSN 0079-8835. A new genus (Kurrajongia) containing two new species (K. armata and K. aploa) in the tribe Amorbini (Coreidae) is described. Both species feed on Brachychiton spp. (Sterculiaceae) in eastern Australia. Dorsal view illustrations of adults and nymphs and drawings of the male and female genitalia are provided, as well as a key and biological notes. □ Hemiptera, Coreidae, Amorbini. Brachychiton. H. Brailovsky, Instituto de Biologia UNAM, §70153, Mexico 04510 D.F.; G.B. Monteith, Brisbane 4101, Australia; 1 October, 1997. The tribe Amorbini (Hemiptera: Heteroptera: Coreidae) has 6 described genera in the Austra- lian region: Acroelytron Mayr (1 species), Amor- bus Dallas (15 species), Cneius Stal (1 species), Gelonus Stal (1 species), Kurnaina Distant (1 species), and Tambourina Distant (1 species). This paper describes a new genus of Amorbini with two new species, both of which are associ- ated with the common Kurrajong trees {Brachy- chiton populneus and B. diversifolius) in eastern Australia. They are moderately large (15-20mm) bugs with a speckled greyish appearance which occur in large clusters on the trunks of the trees. However, due to their camouflaged appearance they have been virtually overlooked by collectors until recent years when their host tree became known. The following abbreviations are used for insti- tutions where specimens are located: ANIC = Australian National Insect Collection, Canberra; AM = Australian Museum, Sydney; DPIQ = De- partment of Primary' Industries, Indooroopilly, Brisbane; IBUNAM = Instituto de Biologia (Coleccion Entomologica), Universidad Na- cional Autonoma de Mexico, Mexico City; QM = Queensland Museum, Brisbane; SAM = South Australian Museum, Adelaide; UQIC = Univer- sity of Queensland Insect Collection, Brisbane. Abbreviations used in the label data lists: GBM = G.B. Monteith; SF = State Forest; NP = National Park. All measurements are given in mm. Kurrajongia gen. nov. ETYMOLOGY. For the common name of Kurrajong given to the host trees of both species included in the genus. Kurrajongs are familiar trees of inland eastern Australia. TYPE SPECIES. Kurrajongia armata sp. nov. (Fig. I). Departmento de Zoologia, Apartado Postal Queensland Museum, PO Box 3300, South DIAGNOSIS. The new genus resembles Gelonus St&l in having head and pronotum wider than long, tylus unarmed and extending anteriorly to the jugae, tibiae sulcate, and the abdominal ster- nite VII of the 9 with plica and fissura. It also lacks the incrassate hind femora which are found in Amorbus but not in Gelonus (Steinbauer, 1995). In Kurrajongia the rostrum reaches the me- tastemum, with rostral segment 111 longer than 1 .30mm. The apex of the tylus reaches or extends beyond the antenniferous tubercles, the femora are unarmed, the eyes are large, slightly stylate and project beyond the lateral margins of the head. The pronotal calli are elevated and irregu- larly rounded at each side. In Gelonus the rostrum is shorter, reaching only on to the anterior part of the mesostemum, with rostral segment III shorter than 0.60mm. The tylus is short extending slightly beyond the antenniferous tubercles and the femora are armed. The eyes are hemispherical and not conspicuously projecting, with the calli almost flat. DESCRIPTION. Head. Width across eyes greater than length, non declivent, pentagonal, produced beyond antenniferous tubercles and dorsally flat; tylus laterally compressed and pro- jecting upward as an acute projection; jugae un- armed, apically subglobose and shorter than tylus; antenniferous tubercles prominent, obliquely truncate, unarmed; antennal segment I robust, thickest, slightly curved outwards and longer than head; segments II and III terete and robust; segment IV fusiform; antennal segment II the longest, segment III the shortest; ocelli raised on low tubercles; preocellar pits deep; postocellar pits deep or slightly excavated; eyes large. 380 MEMOIRS OF THE QUEENSLAND MUSEUM slightly stylate, hemispherical, projecting con- spicuously beyond lateral margins of the head; postocuiar tubercle relatively small; bucculae elongate, angulate anteriorly, not extending be- yond middle third of eyes, with external edge sinuate or entirely rounded; rostrum at least to anterior edge of metastemum and sometimes a little further; rostral segments 1 and IV subequal, segments II and III subequal and slightly shorter than I and IV; mandibular plate absent; frons with deep central longitudinal incision. Thorax. Pronotum trapeziform, clearly or slightly wider than long, moderately declivent; collar wide; frontal angles produced forward as medium to broad, short projections; anterior margin entire; anterolateral margins almost straight or slightly concave, with short and subacute spines or nod- ules; humeral angles rounded or acute and pro- duced into a short, laterally directed, conical tooth; posterolateral edge straight or sinuate; pos- terior edge straight; callus elevated, irregularly rounded at each side, divided along midline by a longitudinal furrow and ending on a deep almost circular central depression; disc striate and densely tuberculate. Anterior lobe of meta- thoracic peritreme elevated and auriculiform, posterior lobe sharp, small; propleura, mesopleura, and metapleura densely tuberculate; mesostemum with median sulcus, flanked by two elevated ridges. Legs. Fore coxal cavities slightly opened, middle cavity wide and conspicuously opened, and hind cavity open; femora unanued, densely tubercu- late; tibiae terete, sulcate. Scutellum. Triangular, length equal to or slightly greater than width; anterior third elevated, with anterior margin and posterior third flat; apex sub- acute; disc and apex densely tuberculate; anterior margin and anterolateral areas striate and pol- ished black. Hemelytra. Macropterous, almost reaching the apex of the last abdominal segment; costal margin emarginate; apical margin clearly sinuate with inner third concave and outer third straight; apical angle obtuse; hemelytral membrane elongate with several longitudinal veins, few of them bi- furcate. Abdomen. Connexival segments reflexed above margin of hemelytron at rest; posterior angles of connexival segments simple, not spinose; ab- dominal sterna with weak medial depression, ex- tending to middle third of stemite V. Female genitalia. Abdominal stemite VII with plica and fissura; plica narrow, rectangular, with outer margin straight to slightly concave; gono- coxae I conspicuously enlarged dorso-ventrally, larger than paratergite IX, in caudal view closed, in lateral view slightly convex; paratergite VIII triangular, with spiracle visible; paratergite IX square, larger than paratergite VIII. Spermatheca: Fig. 2A. Male genitalia. Genital capsule: Posteroventral margin emarginate (Fig. 2B, E), with edge rounded and slightly excavated on the midline. Parameres: Inner margin of the basal portion of the body with a lobe projecting anteriorly; inner margin of the middle third of the body flattened, extending anteriorly, and sometimes reaching the basal lobe; distal portion hook-like (Fig. 2C-D, F-G). Integument. Body surface rather dull, with long or short, decumbent to suberect, golden or silvery, bristle-like hairs. Head, pronotum, scutellum, cla- vus, corium, connexival segments, thorax, femora, tibiae, and abdominal sterna densely tu- berculate; pronotum and scutellum irregularly striate. KEY TO SPECIES OF KURRAJONGIA GEN. NOV. 1 . Humeral angles of the pronotum produced into an angulate tooth; pronotum clearly wider than long; second antennal segment with an apical, white ring; third antennal segment with no more than apical half pale; basal two segments of hind tarsus white, contrasting with very dark apical segment; parameres as in Fig. 2F-G K. armata sp.nov. Humeral angles of the pronotum rounded; prono- tum with length about equalling width; second antennal segment without white apex; third an- tennal segment with more than apical half pale; basal two segments of hind tarsus not contrast- ingly pale; parameres as in Fig. 2C-D. K. aploa sp.nov. Kurrajongia armata sp. nov. (Figs l,2E-G, 4A, 6) ETYMOLOGY. Named for the angular projections of the humeral angles of pronotum. MATERIAL. HOLOTYPE: QMT46015 d, Braemar SF via Kogan (Qld), 15-19.X.1979, G.B. Monteith, in QM. PARATYPES (nymphs (N) listed are not types) (Qld) : 2 9 9 , same data as holoty pe, in QM; 4 d d 3 9 9 , same locality, 4.iv. 1996, on B. populneum, GBM, in QM; 1 9, 40km S of The Lynd, 13.ix.l982, GBM, in IBUNAM; 12d(399 95N, Burnell R., 3km W Eidsvold, 14.iv. 1996, on B. populneiim, GBM, in QM; 396 6309 9 IN, Ml MoffattNP, Rotary Shelter Shed, 27.ii.I996, on B. popuineum, GBM, in QM, ANIC, SAM; 29 9, Mt Moffat NP, Kenniffs Lookout, NEW GENUS OF AMORBINI 381 13. xii.l987, GBM, Thompson & Yeates, in IBUNAM & UQIC; 3 $ 9 IN, Mt Moffatt NP, Mt Rugged Sum- mit, 27. ii. 1 996, on5. populneum trilobus ^GBM & C J. Burwell, in QM; 1 6 1 92N, Cracow, 13.iv.I996, on 5. populneum, GBM, in QM; I 9, 34km N of Taroom, 20. ix. 1997, GBM, on B. populneum, in QM; 4d 1 9, ‘The Amphitheatre’, Expedition Ra NP, 24- 26.ix.1997, D. & L.Cook, on B.populneum trilobus\ 1 6 1 93N, 25km N Monto, lO.iv. 1996, on B. popul- neum, GBM, in QM; 6c5' ^ 1 9 IN, Binjour Plateau, via Gayndah, I4.iv. 1996, on B. populneum, GBM, in QM; 1 6, 50km W Eidsvold, 14.iv.I996, on B. populneum, GBM, in QM; I 9, 5km S Bogantungun, 12.iv.I996, on B. populneum, GBM, in QM; 2N, Auburn River NP, via Mundubbera, 17-19.iv.l987, on B. populneum, GBM, in QM; 3c?dlN, 2ml WNW Beta, 30.iv.l957, Key & Chinnick, in ANIC: 2S6\9, Chinchilla, 14. xi.l988, D. Bunting, in DPIQ and IBUNAM; 1 9, ‘Allinga’, Chinchilla, I7.V.1984, G. Lithgow, in QM; 1 9 , Chinchilla town, I6.ii, 1984 , G. Lithgow, in QM. (QM Paralypes: QMT460 16-46049, 46051-46074, 46076-46128, 46133-46138). DIAGNOSIS. Kurrajongia armata is easily dis- tinguished from the only other species in the genus by its shorter, broader form and the charac- ters given in the key. The parameres (Fig. 2F-G) are also distinct. DESCRIPTION. MALE. Colouration. Dorsal: pale orange yellow with tubercles orange hazel; following areas black to dark red brown: sides of head in front of eyes, ocellar tubercle, two longi- tudinal bands running laterad of midline on the anterior and middle third of pronotal disc, two large spots on the posterior margin of pronotal disc, basal angle, anterior margin (except midline) and lateral margins of the anterior third of scutellum, irregular spots scattered on clavus and corium, and a square spot on the midline of connexival segments III to VII; apex of scutellum pale yellow; hemelytral membrane dirty am- barine, with veins, basal angle and some discoidal or irregular spotting dark brown; abdominal seg- ments bright orange yellow, with most of VII black; antennal segments I and II with ventral side black, and dorsal side red brown with irregular spotting yellow; apical one tenth of II yeIlow;seg- ment III with basal three fifths black, and distal two fifths yellow; segment IV dark red brown with basal yellow ring. Ventral: Pale orange yel- low with tubercles bright red brown to bright orange hazel; following areas black: apex of ros- tral segment IV, area around the metathoracic peritreme, anterolateral margins and spines of the thoracic pleura, most of mesostemum, scattered spots on abdominal sterna III to VII, and genital capsule and middle third of pleural sterna III to FIG. 1. Dorsal view of 9 Kurrajongia armata. VII; coxae hazel brown with external margin pale yellow; trochanters pale yellow with some ventral hazel brown spots; femur pale yellow with tuber- cles red brown, and two black irregular rings, one subbasal, the other subdistal; tibiae pale yellow, densely marked with red brown tubercles, with base and apex mostly black; hind tarsus segments I and II pale yellow and segment III dark hazel orange; anterior and posterior lobes of meta- thoracic peritreme, the area around each abdominal spiracle, and one or two discoidal or irregular spots on prothorax, mesothorax, and metathorax pale yellow to creamy yellow. Structure. Pronotum: Clearly wider than long; frontal angles produced forward as medium-sized projections; anterolateral margins emarginate, obliquely straight or slightly concave, with short and subacute spines; humeral angles acute, pro- duced into laterally-directed, conical teeth; pos- terolateral edge sinuate. Genital capsule. Posteroventral margin emargi- nate, with edge rounded, and slightly excavated on the midline (Fig. 2E). Parameres: Fig. 2F-G. FEMALE. Colouration. Similar to 6. Connexi- val segments VIII and IX black or red brown with 382 MEMOIRS OF THE QUEENSLAND MUSEUM FIG. 2. A-D, Kurrajongia aploa; A, $ spermatheca; B, S genital capsule; C-D, 6 parameres. E-G, Kurrajongia armata; E, 6 genital capsule; F-G, 6 parameres. posterior third orange yellow; dorsal segments VIII and IX black, with following areas dark orange yellow; longitudinal midline band, and lateral edges; genital plates pale orange yellow, with tubercles, and each edge of paratergite VIII and IX black to red brown. Variation: 1, femora pale yellow, with tubercles and irregular spots red brown, and with only one black subdistal ring; 2, abdominal stemite VII with fissura mostly black. Measurements (mm). 6 first, then 9. Head length: 1.80, 2.10; head width across eyes: 2.40, 2.62; interocular space: 1.35, 1.50; interocellar space: 0.47, 0.52; length antennal segments: I, 3.30, 3.40; II, 3.85, 3.95; III, 2.65, 2.80; IV, 3.40, 3.40; length rostral segments: I, 1.30, 1.60; II, 1.20, 1.30; III, 1.20, 1.40: IV, 1.40, 1.55. Prono- tum: Total length: 3. 10, 3.55; width across frontal angles: 2.05, 2.30; width across humeral angles: 4.35, 5.40. Scutellar length: 1.85, 2.20; width: I 75, 2.20. Total body length: 14.00, 16.90. DISTRIBUTION (Fig, 5). Widely distributed in central western Queensland from a little west of the Darling Downs, north almost 1200km to near The Lynd, SW of the Atherton Tableland. A sighting record by the second author is available for nymphs on Kurrajong trees at Granite Gorge, just W of Mareeba. Tliis would extend the known range of the species another 250km further north but requires adults for confirmation. Kurrajongia aploa sp. nov. (Figs 2A-D,3,4B) ETYMOLOGY. From aploos (Greek adjective mean- ing simple ), refers to the lack of spines on the humeral angles of the pronotum. MATERIAL. HOLOTYPE: QMT46014 d, Maiden- well (Qld), 18.iii.I996, on Brachychiton populneus^ GBM, in QM. PARATYPES (nymphs listed are not types) (Qld): 29 9 IN, same data as holotypc, in QM; 1 629 9, Bluff R. foothills, via Biggenden, l-7.i.I972, H.Frauca, inANIC;29lN, Cooyar, 18.iii.l996, on B. populneum, GBM, in QM, SAM; 26 639 94N, Mis- take Mtns via Cunninghams Gap, 10-12. iii. 1978, 1000-1 100m, on B. populneum, GBM and D. Booth, 1 (32 9 9 4N in UQIC and I <3 1 9 in IBUNAM; 1 d 1 9 , Gayndah, Masters, in AM; 1 9, no data,, in DPIQ; 1 6, Brisbane, 4.x. 1925, in IBUNAM; 19, Brisbane, 6.iii.l960, C.J. Paul, in QM; I d I 9, no data, in AM; IN, Nundubbermere Falls, 25km SW Stanthorpe, 2- 3.iv.l988, on B. populneum, GBM, in QM; Id, 6km N of Taroom, 2.iii.l991, G. Daniels, in UQIC; New South Wales: 66 <339 9,22km W of retreat (35km E of Manilla, 600m, 23.x. 1995, Schuh & Cassis, in AM and QM; 1<3, Tenterfield, 3. i. 1942, R. Pullen, in AM. (QM Paralvpes: QMT46050, 46075, 46129- 46132). DIAGNOSIS. Kurrajongia aploa can be recog- nized by its more elongate form, the humeral angles which are rounded and not produced, and characters of the antennae and hind tarsi given in the key. The parameres (Fig. 2C-D) are also dis- tinct. NEW GENUS OF AMORBINI 383 DESCRIPTION. MALE. Colouration. Dorsal: Pale orange yellow with tubercles orange hazel; following areas black to dark red: sides of head in front of eyes, ocellar tubercle, preocellar pits, postocellar pits, lateral margins of vertex, two longitudinal bands running laterad of midline of the anterior and middle third of pronotal disc, basal angle, anterior margin (except midline), and lateral margin of the anterior third of scutellum, and irregular spots on clavus and corium; prono- tal disc scattered with some creamy yellow spots; hemelytral membrane dirty ambarine, with veins, basal angle , and some discoidal or irregular spots dark brown; apex of scutellum yellow; connexi- val segments black with anterior angle and poste- rior margin yellow; abdominal segments bright orange yellow, with the greatest part of VII and lateral portion of VI black; antennal segments I and II, with ventral side black and dorsal side red brown with yellow marks; segment III yellow with basal third black; segment IV black to red brown, with narrow basal ring yellow. Ventral: Pale orange yellow with tubercles bright red brown to bright orange hazel; following areas black: apex of rostral segment IV, area around lobes of the metathoracic peritreme, anterolateral margins of the thoracic pleura, four discoidal spots on the anterior margin of abdominal stemite IV, and two on the sterna V and VI, and one row of discoidal spots near abdominal spiracles of sterna III to VII; anterior and posteror lobe of metathoracic peritreme, the area around each ab- dominal spiracle, and some discoidal or irregular spots on prothorax, mesothorax, and thoracic metapleura pale to creamy yellow; coxae and trochanters pale yellow, with some dark brown spots; femora dirty yellow, with tubercles and marks red brown, and with subdistal irregular black ring; tibiae dirty yellow, and densely cov- ered by dark red brown spots; hind tarsal seg- ments I and II hazel orange; segment III dark hazel orange. Structure. Pronotum slightly wider than long; frontal angles produced forward as broad projec- tion; anterolateral margins emarginate, almost straight, and nodulose; humeral angles rounded; posterolateral edge almost straight. Genital capsule. Posteroventral margin conspicu- osly emarginate, protruding, with edge rounded, and slightly excavated on the midline (Fig. 2B). Parameres: Fig. 2C-D. FEMALE. Colouration) Similar to 3 . Connexi- val segments VIII and IX black to red brown; dorsal segments VIII and IX black, with follow- ing areas dirty orange yellow: longitudinal band FIG. 3. Dorsal view of 3 Kurrajongia aploa. at midline of segment IX, and lateral portions of each segment; genital plates pale orange yellow, with basal angle of gonocoxae I, external edge of paratergite IX, and small spot near the spiracle of paratergite VIII black. Measurements (mm). 3 first, then $. Head length: 2.25, 2.55; width across eyes: 3.15, 3.25; interocular space: 1.75, 1.92; interocellar space: 0.57, 0.57; length antennal segments: I, 3.50, 3.85: II, 4.60, 4.85; III, 2.85, 3.20; IV, 3.40, 3 63; length rostral segments: I, 1.70, 2.00; II, 1.50, 1,75; III, 1.40, 1.65; Iv, 1.70, 1.90. Pronotum: total length: 3.75, 4.50; width across frontal an- gles: 2.35, 2.70; width across humeral angles: 3.90, 5.00. Scutellar length: 2.00, 2.30; width: 1.70, 2.00. Total body length: 17.10, 20.40, DISTRIBUTION (Fig. 5). This species is known from a series of localities, mostly at higher alti- tude, along the Dividing Range from Tenterfield in far northern New South Wales north to the Cooyar-Maidenwell district just north of Toowoomba in southern Queensland. Beyond that area there is one record 150km further north at lower elevations on the subcoastal ranges near Biggenden and another from 200km to the west 384 MEMOIRS OF THE QUEENSLAND MUSEUM FIG. 4. Last instar nymphs of Kurrajongia spp. A, K. armata. B, K. aploa. near Taroom. There are two early records from Brisbane which have not been repeated by recent collecting. Except for the Taroom locality the distribution of K. aploa is allopatric with respect to that of K. armata which is generally further inland. HOST PLANTS OF KURRAJONGIA SPECIES All host records for these bugs are from trees of Brachychiton Schott & Endl. (Sterculiaceae) which has been recently monographed by Guymer (1988). It includes trees commonly known as Bottle Trees and Kurrajongs. Brachy- chiton is considered to be one of the older en- demic components of the Australian flora and has 31 species of which 30 occur in Australia and 2 in New Guinea. Within Australia they occur across the north and east of the continent with one species in the central deserts. Centres of diversity are in the Kimberley (9 spp) and in Cape York Peninsula (10 spp). However records of Kurra- jongia are all from only two of the three non-de- ciduous species which form the Section Poecilodermis of Brachychiton. These are the familiar trees known as Kurrajongs and which are widely retained (or replanted) on farms in eastern Australia because of their value as drought relief fodder for stock. The species are the common Kurrajong, B. populneus (Schott & Endl.) R.Br. in Horsfleld, and the northern Kurrajong, B. di- versifolius R.Br. in Horsfield. Their distribution in eastern Australia and the records for Kurrajongia species are shown in Fig, 5. B. populneus is di- vided into two subspecies and the dotted line shows the ap- proximate boundary between them based on information in Guymer (1988) and more re- cent herbarium records pro- vided by that author. Almost all records for both species oi Kur- rajongia are from B. populneus but the two records from north Queensland {K. armata at The Lynd; presumed nymph of K armata from Granite Gorge) are from within the range of the closely related B. divers if olius. In the south the two subspecies of B. populneus were not all differentiated at the time of col- lection of the bugs. However it appears that all records of K. aploa are from B. populneus populneus while K armata occurs on both that subspecies and on B. p. trilobus Guymer. The single Taroom specimen of K. aploa is puzzling because there appears to be no Kurrajong trees at the locality. When the locality was revisited in Sept., 1997 the only Brachychiton found was the narrow leafed bottle tree, Brachychiton rupestris (Mitchell ex Bin- dley) Schumann but Kurrajongia was not found on that tree. Although there is a diverse flora of Brachychi- ton species available in Australia, it is very curi- ous that the two very distinct species of Kurrajongia which have arisen both feed on a very small taxonomic range of species. This may be an artefact of collecting because these bugs are very cryptic and require special search. However within the region of southern Queensland where the bugs are common on B. populneus there are at least five other Brachychiton species which have been examined by the second author without results. Brachychiton species in other areas may repay attention. BIOLOGY OF KURRAJONGIA SPECIES Both species are found on the trunks of the trees where they adopt a flattened posture against the rough bark which makes them very hard to see. A characteristic habit of both species, at least in late summer, is the formation of very large aggrega- NEW GENUS OF AMORBINI 385 tions of mixed adults and nymphs. These may number several hundred individuals and may measure 20cm in diameter. These aggregations form a quiescent patch on the trunk, ofen at about the level of the lower limbs of the tree. When disturbed they move en masse around the trunk and settle again. If disturbed vigorously they syn- chronously discharge their scent glands which provides a formidable group defence. Kurra- jongia armata has only been taken on the main trunks of the trees but K. aploa individuals are occasionally found on the smaller branches and FIG. 6. Living adult 9 of Kurrajongia armata on foliage of Brachychiton populneus. twigs where they adopt the same flattened posture lengthways on the twigs. Egg laying has not been observed though a single old egg, presumed to be of K. armata, was found in a bark crevice adjacent to an aggregation of that species. The nymphs are similarly coloured to the adults and have no aposematic colour or behaviour as seen, for example, in Amorbiis nymphs.. Last instars of both species are shown in Fig.4. They can be separated by the same prothorax, antenna and tarsal features described in the key to adults. ACKNOWLEDGEMENTS The following individuals and institutions pro- vided loans and other assistance; G. Cassis (AM), J. Donaldson (DPIQ) and Margaret Schneider (UQIC). Cristina Urbina (IBUNAM) prepared the adult dorsal view illustrations and Geoff Thompson (QM) drew the nymphs. Jeff Wright (QM) photographed the living adult. Special thanks are extended to the Consejo Nacional de Ciencia y Tecnologia, Mexico (CONACyT) for financial assistance to the senior author. Gordon Guymer of the Queensland Herbarium gave advice on taxonomy and distribution of Brachychiton. LITERATURE CITED GUYMER, G.P. 1988. A taxonomic revision of Brachychiton (Sterculiaceae). Australian System- atic Botany I: 199-323. STAL, C. 1865. Hemiptera Africana II. Stockholm; Norstedtiana: 1-200. STEINBAUER, M. 1995. The biogeography and host plant utilisation of eucalypt feeding Coreidae (Hemiptera: Heteroptera). PhD thesis. University of Tasmania, Hobart. 386 MEMOIRS OF THE QUEENSLAND MUSEUM EASTERN RANGE EXTENSION OF FORREST’S MOUSE {LEGGADlf^A FORRESTI) Memoirs of the Queensland Museum 42(2): 3S6. 199S:- Forrest’s Mouse {Leggadinaforresti) is generally restricted to arid inland areas of Australia (Reid & Morton, 1995; Watts & Aslin, 1981). Specimen records from the Queensland Museum mammal database indicate that the «istem-most registration of L for- resti is from ‘Hillgrove’ (19°38’S 145® 43’E). Regurgitated owl pellets collected from Mt Castor (23°30’25”S 147®0L5E’E), Gemini NP in central Qld during 1995 ex- tended the eastern range of the species by approximately 210km. Here, wc report the capture of two specimens of L forresti from properties near Rolleston (24°28’00”S !48®37’30”E), approximately 250km SE of Mt Castor and 300km E of ‘Hillgrovc’, where previous records of the species have been collected. Daring April, 1997 fauna surveys were completed in the forests and woodlands surrounding Rolleston, central Queensland. Two specimens of L forresti were captured in two separate tracts of open forest, dominated by Eucalyptus coolabah. Both specimens were captured in pitfall traps that consisted of three, 20-litre buckets, spaced 5m apart, set so that the lip of each bucket was flush with the ground. A 20m-long drift fence was erected dissecting the mouth of each bucket. A total of 96 pitfall trap-nights were undertaken during the fauna survey of the area. One 6 L. forresti was captured on ‘Sirius’ (24®25’25”S 148®36’30”E) and one 9 (QMJMU585) approximately 6.5km to the SE on ‘Hillview’ (24®28'40”S I48®38’ I0”E). Both areas were of riparian vegetation along Comet R. The soils at the sites were cracking clay alluvials. Other mammals recorded during the survey mostly reflect eastern assem- blages. They included; Planigale ingrami (coolibah forest, ‘Venus Downs’); Macropus parryi (edge of grassland and coolibah forest, ‘Venus Downs’); Melomys burtoni fsemi-de- ciduous vine forest, ‘The LakeT; Cams lupus dingo (grass- land, ‘The Lake’); Sminthopsis macroura and S. murina (coolibah forest, ‘Hillview'); Pseudomys delicatulus (silver- leafed ironbark woodland, ‘Tlie Crescent’); Chalinolobus pi- catus and C. nigrogriseus (creek in brigalow forest, ‘The Crescent’); Mus musculus, Nyctophilus geoffroyiy N. gouldi, Scotorepens greyii, Saccolaimus JlaviventriSy Mormopterus beccarii, Nyctinomus australis. Macropus giganteus and Trichosurus vulpecula (throughout study area); Petauroides volans (riparian zone along the Comet R., ‘The Lake’); Tachy- glossus aculeatus (coolibah forest, ‘Blockadirt’). Acknowledgements We thank Mr Ivan Eyies (‘Hillview’) and Mr Des Priddle (‘Sirius’) for allowing access onto their properties during the survey. Data records were supplied by H. Janetzki, Queens- land Museum. Literature cited REID, J.W.R. & MORTON, S.R. 1995. Forest’s Mouse. Pp. 555-556. In Strahan, R. (ed.) The Mammals of Austra- lia. (Reed Books: Chatswood). WATTS, C.H.S. & ASLIN, H.J. 1981. The rodents of Austra- lia. (Angus and Robertson.Sydney). Simon Banks, Allison Rushton, PPK Environment d Infra- structure, PO Box 2907. Brisbane 4000 d Steve Van Dyck. Queensland Museum. PO Box 3S00, South Brisbane 4101. Australia; 16 February, 1998. PONTONIINE SHRIMPS FROM MORETON BAY, QUEENSLAND (CRUSTACEA: DECAPODA: PONTONIINAE) A.j. BRUCE Bruce, A.J. 1998 06 29: Pontoniine shrimps from Moreton Bay, Queensland (Crustacea: Decapoda: Pontoniinae). Memoirs of the Queensland Museum 42(2): 387-398. Brisbane. ISSN 0079-8835. Three recently collected new species of pontoniine shrimps from Moreton Bay, southern Queensland, are described and illustrated. Periclimenes terangeri was associated with an unidentified actinarian, while the hosts of Onycocaris stradbrokei and Periclimenaeus myora, probably a sponge and a tunicate respectively, were not identified. These additional species increase the pontoniine fauna known from Moreton Bay to about 20 species. □ Natantia, Pontoniinae. new species, Moreton Bay, Queensland. A.J. Bruce, Crustacea Section, Queensland Museum, P.O.Box 3300, South Brisbane 4101, Australia: 4 July, 1997. The pontoniine shrimp fauna of Moreton Bay, Queensland (Qld), is of particular interest as it is nearing the southern limit of the Indo-West Pa- cific zoogeographical region. The first species recorded, Coralliocaris graminea, C. superba, Jocaste lucina, Periclimenes amymone and P. madreporae (as P. inornatusf all associates of branching corals from Myora, were reported by Patton (1966). Since then a further 13 species only have been recorded in published literature, making a total of 20 species. This is in marked contrast with the fauna of Heron I., at the southern end of the Great Barrier Reef, some 480km to the north, where the pontoniine fauna has received more intensive study and well over 100 species are known to occur. The discovery of three new species in Moreton Bay suggests that still further species may be expected to be present. A new species of Pontonia was also collected and will be described elsewhere. The type specimens are deposited in the collections of the Queensland Museum, In the descriptions, carapace length refers to the postorbital carapace length (POCL). SYSTEMATICS Class CRUSTACEA Order DECAPODA Latreille, 1803 Family PALAEMONIDAE Rafinesque, 1815 Sub-family PONTONIINAE Kingsley, 1878 Periclimenes terangeri sp.nov. (Fig. 1) ETYMOLOGY. Aboriginal, terangeri, Stradbroke Island. MATERIAL. HOLOTYPE; QMW21735, 1 ovig. 9; Polka Point, North Stradbroke 1. Qld, 27°30’S 153®23’E, 0.5m, coll. P. Davie and J. Short, 6 March 1996, in sea-grass. PARATYPE: QMW22275, 1 ovig. 9 (dissected), same collection data as holotype. DESCRIPTION. A small, slender Periclimenes species, of subcylindrical body form, generally similar to P. toloensis Bruce. Adult female. Rostrum (Fig. 1 A) slender, straight, slightly up-turned, c. 1.15 x carapace length, slender, acuminate, reaching to slightly beyond end of antennular peduncle, with 9 acute evenly spaced dorsal teeth, 1st tooth epigastric, 2nd situ- ated posterior to orbital margin, articulated, lat- eral carinae feebly developed, ventral carina obsolete, margin straight, with two small acute teeth on distal 3rd, with short plumose median setae proximally. Carapace smooth, glabrous, su- praorbital spines absent, orbit feebly developed, inferior orbital angle strongly produced, distally blunt in lateral view, antennal spine well-devel- oped, marginal, slightly exceeding inferior orbital angle, hepatic spine subequal to antennal spine, distinctly inferior to level of antennal spine, ante- rior to level of 1st dorsal rostral tooth, anterolat- eral angle of branchiostegite broadly rounded, not produced. Abdomen as in P. toloensis, smooth, glabrous, 3rd tergite not posterodorsally produced, pleura broadly rounded, 6th segment c. 1.65 X length of 5th, 1.9 X longer than anterior depth, 0.75 of carapace length, posteroventral angle bluntly pro- duced, posterolateral angle acutely produced. Tel- son (Fig. IN) c. I.l x 6th segment length, 4.0 x longer than anterior width, lateral margins straight, posteriorly convergent, with two pairs of well-developed dorsal spines, at c. 0.45, 0.65 of length, posterior margin c. 0.5 of anterior width. 388 MEMOIRS OF THE QUEENSLAND MUSEUM 2.0mni ABCDFH l.Omm IJKLN FIG. 1. Periclimenes terangeri sp. nov., ovigerous $ paratype (QMW22275), Polka Point, North Stradbroke 1. A, carapace and rostrum, lateral. B, antennule. C, antenna. D, eye, dorsal. E, 4th thoracic stemite. F, 1st pereiopod. G, same, chela. H, major 2nd pereiopod. I, same, chela. J, same, fingers. K, minor 2nd pereiopod, carpus and chela. L, 3rd pereiopod, propod and dactyl. M, same, distal propod and dactyl. N, telson. O, same, posterior spines. PONTONIINE SHRIMPS FROM MORETON BAY 389 angular, with acute median point, lateral spines (Fig. lO) subequal to dorsal spines, intermediate spines slender, c. 0.18 oftelson length, submediam spines robust, setulose, c. 0.45 of intermediate spine length. Antennular peduncle (Fig. IB) normal, as in P. toloensis, proximal segment c. 2.2 x longer than central width, with small acute ventromedial tooth, distolateral angle bluntly produced, with strong acute lateral tooth, extending well beyond distal margin of lobe, statocyst with irregular subcircular statolith, stylocerite slender, acute, reaching to c. 0,5 of segment length, intermediate and distal segments normal, combined lengths c. 0.7 of proximal segment length, upper flagellum with proximal 5 segments of rami fused, shorter free ramus with 5 segments, with 10 groups of aesthetascs, longer ramus and lower flagellum slender, filiform. Antenna (Fig. 1C) as in P. toloensis, with basicerite with well-developed lateral tooth, car- pocerite not exceeding proximal segment of an- tennular peduncle, c, 2.8 x longer than central width, reaching to c. 0.45 of scaphocerite length, flagellum well-developed; scaphocerite c. 3.3 X longer than central width, lateral margin straight with strong distolateral tooth, lamella broad, scarcely tapering, distally bluntly angular, far ex- ceeding distolateral tooth. Eye (Fig. ID) with large globular cornea, di- ameter c. 0.2 of carapace length, with dorsal accessory pigment spot, stalk c. 1.5 x longer than wide, 1.5 X longer than corneal diameter. Mouthparts of normal Periclimenes form, with- out special features. Third maxilliped with vestig- ial arthrobranch. Fourth thoracic sternite (Fig. IE) without median process, with low transverse lamina with subcircular median notch; 5th ster- nite similar, with u-shaped median notch. First pereiopods (Fig, IF) slender, merus reach- ing to middle of carpocerite, chela (Fig. IG) to end of scaphocerite, with palm subcylindrical, c. 2.4 X longer than deep, fingers slender, com- pressed, c. 0.8 of palm length, with small acute articulated tips, strongly hooked, cutting edges well-developed over distal half, entire, laterally situated; carpus c. 1.1 x chela length, subcylin- drical, c. 5.0 X longer than distal width; merus distinctly longer than carpus, c. 1.2 x chela length, uniform, 9.0 x longer than central width; ischium c. 0.6 of merus length; basis normal; coxa with well-developed setose ventral lobe. Second pereiopods (Fig. 1H,K) dissimilar, un- equal; major chela (Fig. H) c. 1.4 of carapace length, with palm subcylindrical, slightly com- pressed, smooth, c. 4.0 X longer than central width, fingers (Fig. IJ) robust, c. 0.55 of palm length, dactyl c. 4.5 X longer than deep, cutting edge unarmed, distal 2/3 of cutting edge lateral, entire, sharp, fixed finger similar, with single acute tooth proximally, cutting edge distally sharp, entire, tip missing; carpus c. 0.65 of palm length, 3.5 x longer than distal width, tapering proximally, unarmed; merus longer than carpus, equal to 0.95 of palm length, 0.65 of chela length, unarmed, 7.0 x longer than central width, subuniform: ischium c. 1.1 x merus length, 10 X longerthan central width; basis and coxa with- out special features. Minor chela (Fig. IK) c. 0.8 of carapace length, 0.65 of the major chela length; palm c. 3.5 x longerthan central width, subcylin- dicai, fingers c. 0.9 of palm length, more slender, proximal cuttings edge entire, tips acute, feebly hooked, carpus c. 0.95 of palm length, more slen- der; proximal segments similar to major chela. Ambulatory pereiopods slender; 3rd pereiopod with dactyl (Fig. IM) slender, c. 0.23 of propod length, simple, corpus c. 3.0 x longer than proxi- mal depth, compressed, tapering distally, dorsal margin convex, ventral margin sinuous, sharp, unguis distinctly demarcated, strong, curved, c. 0.65 of corpus length, deflexed; propod (Fig. IL) c. 0.65 of carapace length, 12 x longer than central depth, uniform, sparsely setose, distal 4th of ventral border with three pairs of long spines distally, c. 1.25 x propod width, single similar spine proximally, proximal three quarters of ven- tral border with two short spines, carpus 0.5 of propod length, unarmed; merus 0.95 of propod length, unarmed; ischium 0.6 of propod length, unanned, basis and coxa normal. Fourth and 5th pereiopods similar to 3rd. Uropod as in P. toloensis; protopod with small blunt distolateral lobe; exopod c. 3.0 x longer than wide, lateral margin with slender acute dis- tolateral tooth, with large mobile spine medially, dieresis feebly developed; endopod c. 0.9 of exopod length, c. 4.0 x longer than wide, MEASUREMENTS (mm). Holotype, QMW21735 9: carapace length, 2.4; carapace and rostrum, 4.9; total body length (approx.), 12.7; 2nd pereiopod, major chela, 3.5; minor chela, 2.1; length of ovum, 0.5. COLOURATION. No data. 390 MEMOIRS OF THE QUEENSLAND MUSEUM HOST. Not recorded. SYSTEMATIC POSITION. One of the Pericli- menes obscurus species group, most closely re- lated to P. toloensis Bruce, from which it may be distinguished by the following features: 1) more slender, acuminate rostrum, with two ventral teeth. 2) more elongate eyestalk, c. 1.5 x longer than wide, compared with c. 1.0 x . 3) 4th thoracic sternite with transverse ridge with subcircular median notch. 4) 1 St pereiopod chela slightly shorter than car- pus, distinctly shorter than merus, not longer than carpus. 5) propod of 3rd pereiopod armed distally with 3 pairs of long slender spines, length exceeding distal propod width, and a single long ventral spine, more proximal spines short, not with shorter spines, with two distal pairs only. 6) dorsal telson spines subequal to lateral pair of posterior telson spines, not minute, much smaller than lateral posterior spines. REMARKS. The holotype of P. toloensis was a S from 10-30m off Ap Chau, Hong Kong, and was possibly associated with a gorgonian host (Bruce, 1969; 1982). The host has subsequently been identified as Hicksonella princeps Nutting, 1910 (Moreton & Harper, 1995). As the present specimens are both adult 9 9 , sexual dimorphism may account for some of the morphological dif- ferences. The holotype of P. toloensis had a POCL of 2.4 mm; it was therefore of similar size to P. terangeri (POCL 2.7mm) and probably also fully adult. Periclimenes toloemis of Chace & Bruce (1993), from the Sulu Archipelago, from 38m, also an ovigerous 9, is particularly similar to P. terangeri in that the dorsal telson spines are not minute but it has only a single ventral rostral tooth and the spinulation of the ambulatory propod resembles closely that of P. toloensis. The specimens referred to P. indicus (Kemp, 1915), from intertidal anemones, Macrodactyla aspera, by Bruce (1983b) may also belong to the present species but are not available for re-exami- nation. The species reported from Moreton Bay by Wadley (1978) as Periclimenes {P.) nr ohscu- rus Kemp, 1922, may also belong to P. terangeri. Specimens provided by Dr Wadley have been examined but unfortunately no conclusions could be reached due to lack of many appendages. Onycocaris stradbrokei sp.nov. (Figs 2-3) ETYMOLOGY. From locality of capture, North Strad- broke 1. MATERIAL. HOLOTYPE: QMW21691, 1 9, stn QMP-6, Myora Reef, North Stradbroke 1., Qld, 27°29’S. 153°25’E., 9m, on patch reef. coll. P. Davie and J. Short, 5 March 1 996. ALLOTYPE; QMW22276 1 c3, same collection data as holotype. DESCRIPTION. Small shrimps of stout sub- cylindrical body form, generally similar to other Onycocaris species. Female with rostrum (Fig. 2AB, 3A) short and broad, about as wide as long, reaching to level of anterior margins of eyes, unarmed, distally acute, slightly upturned, without distinct dorsal carina, lateral carinae expanded, ventral carina well-de- veloped; orbital notch well marked, inferior or- bital angle extracorneal, broadly acute; anterolateral branchiostegite broadly rounded. Abdomen with pleura rounded, 6th segment with posterolateral angle acute, posteroventral angle larger, acute. Telson (Fig. 2J) 1.9 x longer than broad, lateral margins convex, convergent, dorsal spines small, subequal, marginal, at 0.55 and 0.8 of telson length, posterior margin broadly convex, without median point, c. 0.25 of anterior width, posterior spines (Fig. 3F) short, small, lateral spines similar to dorsal spines, c. 0.45 of intermediate spine length, intermediate spines c. 0.1 of telson length, 5 X than proximal width, with 3 submedian spines, feebly setulose, c. 0.6 of intermediate spine length. Antennular peduncle (Fig. 2C) short, exceeding carpocerite and scaphocerite; proximal segment c. 2 X as long as central width, with acute tooth distolaterally, without ventromedial tooth, stylo- cerite well-developed, acute, reaching to half seg- ment length; intermediate and distal segments short and broad, combined length c. 0.55 of proxi- mal segment length; upper flagellum short, feebly biramous, four proximal segments fused, distal part of 4th segment free, with 7 groups of aes- thetascs, dorsal ramus with 7 slender free seg- ments; lower flagellum short, with 12 slender segments Antenna (Fig. 2D) with basicerite stout, later- ally unarmed, with large protuberant process for antennal gland aperture; carpocerite well exceed- ing scaphocerite, c. 4.0 x longer than central width, flagellum short, filiform; scaphocerite small, c. 2 X as long as distal width, anterior margin broadly rounded, lateral margin straight PONTONIINE SHRIMPS FROM MORETON BAY 391 FIG. 2. Onycocaris stradbrokei sp. nov., 9 holotype (QMW21691), Myora Reef, North Stradbroke I. A, anterior carapace, eye, and left antennular peduncle, lateral. B, anterior carapace, eyes and antennal peduncles, dorsal. C, antennular peduncle. D, antennal peduncle. E, eye, dorsal. F, 1 st pereiopod. G, major 2nd pereiopod, lateral. H, 3rd pereiopod. I, same, propod and dactylus. J, telson. 392 MEMOIRS OF THE QUEENSLAND MUSEUM FIG. 3. Onycocaris stradbrokeisxi. nov., 9 holotype (QMW2I691), MyoraReef, North Stradbroke I. A, rostrum, lateral. B, mandible. C, 1st pereiopod, chela. D, major 2nd pereiopod, distal cutting edges of fingers. E, 3rd pereiopod, distal propod and dactyl. F, telson. posterior spines. with short stout distal tooth, not exceeding ante- rior margin of lamella. Eyes (Fig. 2E) contiguous in midline, anterior margins subquadrate, with oblique hemispherical cornea, feebly pigmented, stalk short, globular, width c. 1.25 of corneal diameter Mandible (Fig. 3B) with slender corpus; molar process robust, distally obliquely truncate, feebly dentate, sparsely setose, incisor process strongly reduced, non-dentate. Posterior mouthparts pre- sent no special features. Thoracic stemites un- armed; very narrow, longitudinally carinate between 3rd pereiopods. First pereiopod (Fig. 2F) long and slender, reaching beyond carpocerite by half merus, chela and carpus c. 0.65 of carapace length; chela (Fig. 3C) mooth, c. 4.0 x longer than central width, subcylindrical, slightly swollen centrally, fingers c. 0.35 of palm length, slender, very acutely pointed distally, feebly subspatulate with entire medial cutting edges, with row of stiff setae along lateral edges; carpus slender, c. 1.75 X chela length, 9.0 x longer than distal width, tapering proximally; merus 1.8 x longer than chela, 8.0 X longer than central width; ischium 1.33 x chela length, 4.5 x longer than central width; basis c. 0.8 of chela length; basis stout, with small distoventral process. Second pereiopods (Fig. 2G) well-developed, robust with large, stout, near subequal, vertically held chelae. Major 2nd pereiopod with chela c. 0.7 of carapace length, palm stout, smooth, com- pressed, broadly rounded posterodorsally, ventral border sinuous, c. 1.2 x longer than greatest depth, situated at c. 0.5 of length, tapering dis- tally, dactylus c. 0.66 of palm length, 3.3 x longer than proximal depth, sparsely setose, with PONTONIINE SHRIMPS FROM MORETON BAY 393 strong acute tooth distally, medially concave with lateral cutting edge, proximal half with 2 small, low blunt teeth, distal half (Fig. 3D) convex with c. 30 small rounded teeth, size increasing slightly distally, fixed finger similar, with 3 small teeth proximally, c. 40 teeth (Fig. 3D) along distal cutting edge, slightly smal ler than dactylar series, with well-developed submarginal flange distolat- erally, with c. 23 small rounded marginal teeth and blunt distal angle, dactylar cutting edge clos- ing into sulcus between flange and main cutting edge of fixed finger; carpus stout, c. 1 .7 x longer than greatest width, smooth, unarmed, distally broadened; merus stout, centrally swollen, with- out ventral tubercles, c. 0.6 of palm length, 1 .8 x longer than central width, with strong subacute distoventral tooth laterally; ischium slightly shorter than merus, tapering proximally, 2.0 x longer than distal width, ventrally smooth, with very strong acute distoventral tooth laterally; ba- sis and coxa robust, without special features. Mi- nor 2nd pereiopod similar to major; slightly smaller, chela c. 0.83 of carapace length, teeth along distal cutting edges of fingers slightly fewer and smaller. Third pereiopd (Fig. 21) short, robust, extend- ing beyond carpocerite by dactylus only; dactylus (Fig. 3E) strongly compressed, unguis distinct, acute, curved, c. 0.6 of dorsal corpus length, 2.5 X longer than basal width, without ventral den- ticles, corpus c. 1.2 X longer than greatest depth at 0.35 of length, dorsal and ventral borders strongly convex, ventral margin with large broad acute triangular distal tooth, with 8-9 small acute anteriorly inclined teeth proximally, central teeth larger than others; propod (Fig, 21) c. 0.28 of carapace length, stout, 3.7 x dactylar length, compressed, 3.7 x longer than central depth, with 2 pairs of stout distoventral spines, single similar ventral spine at 0.6 of ventral length; carpus 0.75 of propod length, 2.8 x longer than distal width, unarmed; merus l.I x propod length, 2.6 x longerthan central width, unarmed, ischium subequal to propod length, 2.3 x longer than distal width, unarmed; basis and coxa stout, without special features. Uropods without special features. Male generally similar to $, smaller and less robust, with relatively larger but similar chelae. Major chela c. 1.25 x carapace length, minor chela subequia to carapace length; dentition of fingers similar, major chela with 4 small teeth proximally on fixed finger. MEASUREMENTS (mm). Holotype, QMW21691 9, carapace length 4.25; total body length (ap- prox.) 12, major chela, 3.85; minor chela 3.6. Male allotype, carapace length 2.8, total body length (approx.) 8.3, major chela 3.4, minor chela 2 . 8 . COLOURATION. Unknown. HOST. Unidentified, presumably a sponge. SYSTEMATIC POSITION. Onycocaris strad- brokei is most closely related to O. bocki Bruce, 1992, and may be distinguished by the following characters: 1) dactylar cutting edge of the major 2nd pereiopod in O. stradbrokei has more numerous and distinctly smaller denticles than in O. bocki, of the same order of size as those on the fixed finger. In O. bocki there are less than 20 denticles, which are distinctly larger than those on the cut- ting edge of the fixed finger. 2) 3rd pereiopod dactyl has more numerous ventral denticles (8-9), shorter, less erect, more inclined distally, the distal accessory tooth broader, more triangular, and with 2 pairs of stout distoventral propod spines, rather than 6 ventral denticles, a more slender based, trapezoidal ac- cessory distal tooth, and a single pair of more slender distoventral spines with 3 ventral propo- dal spines in O. bocki. 3) 1st pereiopod is much more slender and elongate on O. stradbrokei than in O. bocki, with the carpus c. 1.75 x the chela length as opposed to 1.4 x,propod5.8 X longerthan central width, as opposed to 4.3 x in O. bocki. 4) scaphocerite falling clearly short of distal end of antennular peduncle in O. stradbrokei, as opposed to subequal in O. bocki. Carpocerite markedly longer than scaphocerite in O. strad- brokei, shorter in O. bocki. 5) orbital notches deeper in O. stradbrokei than in O bocki, so that tip of rostrum is level with anterior margins of eyestalks, with posterior part of eyestalk concealed. Eyes more fully exposed dorsally in O. bocki, (although this may possibly be due partly to post-mortem swelling). REMARKS. Onycocaris stradbrokei is a mem- ber of the O. analitica species group. This group has short toothless rostra and a strong distoventral tooth on the merus and ischium of the 2nd pereiopods; it includes O. aualitica (Nobili, 1904), O. amakusensis Fujino & Miyake, 1969, O. callyspongiae Fujino & Miyake, 1969, and O. o/igodentata Fujino & Miyake, 1 969. Onycocaris stradbrokei and O. bocki may be distinguished 394 MEMOIRS OF THE QUEENSLAND MUSEUM from other species by the non-denticulate ungues of the ambulatory dactyls. Onycocaris analitica and O. oligodentata appear to be particularly closely related or are possibly synonymous. The species O. amakusensis and O. oligodentata have also been recorded from Australian waters. Periclimenaeus myora sp.nov. (Figs 4-5) ETYMOLOGY. From locality of capture, Myora Reef MATERIAL. HOLOTYPE: QMW21706, ovig. 9 , stn QMP-3, Myora Reef, North Slradbroke L, Qld, 27°29’S. 153°25’E., 3m, on patch reef, coll. P. Davie and J. Short . 5 March 1996. DESCRIPTION. Small-sized pontoniine shrimp of stout subcylindrical, slightly compressed body form. Rostrum (Fig. 4ABC) slender, slightly de- pressed, short, c. 0.45 of carapace length, com- pressed without distinct lateral carinae, feebly up-curved, reaching to c. level of anterior corneal margins, falling well short of distal margin of proximal segment of antennular peduncle, dorsal margin with 3 acute teeth, 1st tooth well in ad- vance of posterior orbital margin, shorter and stouter than two slender distal teeth, ventral mar- gin convex, unarmed. Carapace smooth, depth subequal to length, without supraorbital spines or tubercles, antennal spine slender, acute, anterolat- eral angle of branchiostegite produced, rounded. Abdomen without special features, stout, pleura 1-4 expanded, broadly rounded, pleuron 5 reduced, 6th segment broad, depressed, c. as long as anterior width, with posteroventral angles large, acute, posterlateral angles smaller. Telson (Fig. 40) c. 2.0 X longer than 6th segment, 1.9 X longer than anterior width, lateral margins sublinear, posteriorly convergent, with two pairs of subequal, marginal dorsal spines, c. 0.07 of telson length, posterior margin c. 0.4 of anterior margin width, broadly convex, with three pairs of marginal spines (Fig. 5K) , lateral spines robust, slightly smaller than dorsal spines, intermediate spines stout, 0.16 of telson length, 1.8 x lateral spine length, submedian spines slender, sparsely setulose, 1.2 x intermediate spine length. Antennular peduncle (Fig. 4D) short, reaching to anterior margin of scaphocerite, proximal seg- ment c. 2 X length of central width, with small distolateral tooth, without ventromedial tooth, stylocerite acute, 3.0 X longer than wide, reach- ing to c. 0.5 of segment length, statocyst normally developed, with granular statolith; intermediate and distal segments normal, combined length c. 0.6 of proximal segment length; upper flagellum with rami fused for three proximal segments, with 3rd segment partially free, with 8 groups of aes- thetascs, longer free ramus with 6 segments, lower flagellum slender, short, with 9 segments. Antenna (Fig. 4E) with stout unarmed basicerite, coxal region with protuberant antennal gland opening, caipocerite short, slender, c. 5.0 X longer than distal width, subcylindrical, slightly flattened, not exceeding anterior margin of scaphocerite, flagellum short, filiform; scapho- cerite (Fig. 5B) small, reaching to end of anten- nular peduncle, distally broadened, 2.4 x longer than wide, lateral margin slightly convex, with small acute tooth distally at c. 0.8 of central length, anterior margin broadly rounded. Eye (Fig. 4F) with well pigmented oblique, hemispherical cornea, diameter c. 0.2 of carapace length, without accessory pigment spot, stalk short stout, c. 1 .3 x longer than wide, strongly flattened medially. Mouthparts of normal Periclimenaeus form, without special features. Mandible with corpus slender, incisor process (Fig. 5A) not expanded distally, with 4 main teeth, most medial tooth with smaller accessory denticle medially, molar proc- ess slender, obliquely truncate distally with mar- ginal fringe of short setae. Thoracic stemites narrow, unarmed. First pereiopods (Fig. 4G) normal, not particu- larly slender, merus reaching beyond carpocerite by chela, carpus and distal 3rd of merus, chela (Fig. 4H) with palm subcylindrical, slightly com- pressed, 2 X as long as deep, fingers subequal to palm length, broad, blunt, spatulate, with mi- nutely bidentate tips, distal cutting edges sharply carinate medially and laterally, entire; carpus subequal to chela length, 4.5 x longer than distal width, tapering proximally; merus c. 1.2 x car- pus length, 5.0 X longer than central width, sub- cylindrical, feebly bowed and tapering distally; c. 0.6 of merus length; basis normal; coxa without ventral process. Second pereiopods (Fig. 41) well-developed, unequal, dissimilar. Major chela (Fig. 4J) large, stout, c. 1.7 X carapace length, oval in section, not strongly compressed, 2.0 x longer than deep, tapering distally, smooth, glabrous, fingers c. 0.25 of palm length, sparsely setose, dactylus with dorsal margin semicircular, with stout acute tip, cutting edge with large molar process proxi- mally, distally minutely serrate (Fig. 5C), fixed finger with large fossa proximally, distal cutting PONTONIINE SHRIMPS FROM MORETON BAY 395 FIG. 4. Periclimenaeus myora sp. nov., ovigerous 9 holotype (QMW2 1 706), Myora Reef, North Stradbroke L A, carapace and rostrum, eye and antennal peduncles, lateral. B, anterior carapace, eyes and antennal peduncles, dorsal. C, anterior carapace and rostrum, lateral. D, antennular peduncle. E, antennal peduncle. F, eye, dorsal. G, Istpereiopod. H, same, chela. 1, major 2nd pereiopod, medial. J, same, chela, lateral. K, minor 2nd pereiopod, chela. L, 3rd pereiopod. M, same, dactyl and propod. N, 5th pereiopod, dactyl and propod. O, telson. 396 MEMOIRS OF THE QUEENSLAND MUSEUM FIG. 5. Periclimenaeus myora sp. nov., ovigerous 9 hoiotype (QMW2I706), Myora Reef, North Stradbroke I. A, mandible, incisor process. B, scaphocerite. C, major 2nd pereiopod, distal dactylus. D, minor 2nd pereiopod, fingers. E, same, cutting edge of dactylus. F, same, fixed finger, cutting edge tooth. G, 3rd pereiopod, distal propod and dactylus. H, 4th pereiopod, same. I, 5th pereiopod, same. J, exopod of uropod, posterolateral angle. K, telson, posterior spines. edge entire, tip stout, acute; carpus 0.33 of chela length, stout, distally expanded, unarmed; merus stout, 0.9 of carpus length, 1.6 x longer than central width, slightly expanded centrally, un- armed, ventral surface without tubercles or denti- cles; ischium stout, subequal to carpus length, c. 2.0 X longer than distal width, tapering proxi- mally, unarmed; basis and coxa stout, without special features. Minor chela (Fig. 4K) well-de- veloped, stout, c. 0.66 of major chela length, 1.1 PONTONIINE SHRIMPS FROM MORETON BAY 397 of carapace length, with palm strongly com- pressed, c. 1.8 X longer than deep, smooth, sparsely setose, tapering slightly distally with fingers (Fig. 5D) curved medially, c. 0.4 of palm length, dactylus (Fig. 5E) 3.0 X longer than deep, dorsal margin concave, cutting edge sublinear, with c. 40 minute, slightly recurved acute teeth, extending distally onto acute tip, fixed finger (Fig. 5F) with stout acute hooked tip, cutting edge straight, longitudinally cannulate, with entire edges, small tuberculate tooth proximally on me- dial edge; proximal segments as for major chela, smaller, less robust. Ambulatory pereiopods stout, 3rd (Fig. 4L) ex- ceeding carpocerite by propod and dactylus (ap- prox.), dactylus (Fig. 5G) 0.2 of propod length, distinctly longer than width of distal propod, un- guis clearly demarcated, stout curved, c. 2.0 x longer than basal width, unarmed, corpus c. 1.5 X longer than basal width, compressed, tapering distally, with strong slender acute perpendicular tooth at 0.25 of ventral margin, distal ventral margin concave, 3 conspicuous sensory setae dis- tally; propod (Fig. 4M) 3.6 X longer than proxi- mal width, tapering distally, sparsely setose, with pair of stout distoventral spines, lateral spine c. 0.15 of propod length, medial spine smaller, car- pus c. 0.8 of propod length 2.7 x longer than distal width, unarmed; merus c. 1.3 X propod length, 3.0 x longerthan central width, unarmed; ischium 0,95 of propod length 2,5 x longerthan distal width, tapering proximally, unarmed; basis and coxa robust, without special features; 4th pereiopod generally similar to 3rd, slightly more slender, dactylus (Fig. 5H) less deep proximally, ventral margin less concave with only minute proximal denticle, propod with pair of slightly unequal distoventral spines; 5th pereiopod with dactylus (Fig. 51) longer than in 3rd, c. 0.2 of propod length, corpus 2.0 x longer than basal width, ventral margin feebly convex, unarmed; propod subequal to 3rd propod length, c. 5.5 x longerthan central width, with single distoventral spine and smaller preterminal spine. Urpods without special features, protopodite distolaterally blunt, exopod (Fig. 5J) with lateral margin entire, unarmed, with small distal tooth and slender mobile spine medially. MEASUREMENTS, (mm) 9 holotype (QMW- 21706), carapace length 1.8 total body length (approx.) 8.6; major chela, 3.1; minor chela 2.0, length of ovum, 0.55. COLOURATION. Unknown. HOST. Unidentified, presumably a tunicate. SYSTEMATIC POSITION. Periclimenaeus my- ora is most closely related to P. nobilii Bruce, 1974, and may be distinguished by the following characters: 1 ) rostrum with 3 dorsal teeth in P. myora (only 2 in P. nobilii). 2) chela of the 1st pereiopod is subequal to the carpus length (c. 0.8 of the carpus length in P. nobilii). 3) distal cutting edge of the major 2nd pereiopod dactylus is minutely denticulate (entire in P. nobilii. 4) cutting edge of the minor 2nd pereiopod dactylus is sublinear, with more uniform and nu- merous smaller teeth, about 40, (more sinuous with about 25 teeth, distinctly increasing is size distally in P. nobilii); chela ventrally glabrous (with numerous long setae in P. nobilii). 5) 3rd pereiopod with dactylus 0.22 of propod lengthened much longer than distal propod depth (only 0.17 x in P. nobilii) and subequal to distal propod width, ventral accessory tooth longer and more perpendicularly orientated; propod 3.6 X longer than greatest depth(more swollen in P. nobilii, c. 3. i) X longer than proximal width). 6) teison with dorsal spines marginal (subdorsal in P. nobilii), relatively smaller and shorter , c. 0.07 of the teison length (0.14 in P. nobilii), situated more posteriorly, anterior pair at c. 0.5 of teison length (0.3 in P. nobilii); lateral and inter- mediate posterior spines more robust, with more slender, longer submedian spines than in P. no- bilii. REMARKS. Periclimenaeus nobilii was de- scribed by Bruce (1974) on the basis of a single 9 collected from an unrecorded locality in the Red Sea by Dr Felix-Pierre Jousseaume in 1897, which had been examined by Nobili and identi- fied as ‘? anomalie de Coralliocaris hecate’’ but not reported in published literature. There have been few subsequent reports. It is not certain whether the cutting edge of the fixed finger of the minor 2nd pereiopod is longitudinally cannulate or not, but the illustration in Bruce (1 974, fig. 14d) suggests that it may be. Two specimens from New Caledonia were referred to P. nobilii by Bruce (1991), but I am no longer convinced that these are conspecific and consider that two separate but closely related species are represented, both asso- ciated with related but different host tunicates. Differences, though small, are most marked in the ambulatory dactyls and propods. These two spe- 398 MEMOIRS OF THE QUEENSLAND MUSEUM cies, together with P. nobilii and P. myora form a complex of closely related species probably all associated with related hosts of the tunicate fam- ily Didemnidae. PericUmenaeus nobilii has also been reported from La Reunion by Bruce ( 1 983a) and the status of this specimen should be re-ex- amined. ACKNOWLEDGEMENTS I am most grateful to Peter Davie and John Short of the Queensland Museum for the oppor- tunity to report on the specimens described herein. The preparation of this report was facili- tated by a grant from the Australian Biological Resources Survey. LITERATURE CITED BRUCE, A.J. 1969. Preliminary descriptions of sixteen new species of the genus Periclimenes Costa, 1844 (Crustacea, Decapoda Natantia, Pontonii- nae). Zoologische Medcdelingen 43(20): 2.53- 278. 1974. Observations upon some specimens of the genus PericUmenaeus Borradaile (Decapoda Natantia, Pontoniinae) originally described by G. Nobili. Bulletin du Museum national d’Histoire naturelle, Paris (3), no. 258. Zoology 180: 1557- 1583 (1975). 1982. ThePontoniine Shrimp Fauna of Hong Kong. Pp. 233-284. In Morton, B. & Tseng, C.K. (eds). Proceedings of the First International Marine Biological Workshop: The Marine Fauna of A NEW NAME, THINORA, PROPOSED FOR THE PREOCCUPIED NAME THORINA BRUCE (CRUS- TACEA: DECAPODA: HIPPOLYTIDAE) Memoirs of the Queensland Museum 42(2): 598. 1998:- A new generic name was proposed for the hippolytid shrimp Thor maldiven- sis Borradaile, 1915 (Bruce, 1997). Professor L.B. Holthuis has kindly pointed out that the proposed name, Thorina. is preoccupied and therefore invalid. The name 77?or/w was first employed by Stephensen (1944) for a new deep-sea caprellid amphipod, Thorina spinosa. A new name, Thinora, an anagram, is now proposed to replace the invalid name Thorina Bruce, 1997. The type species of Thorina Bruce is Thor maldivensis Borradaile, 1915. by monotypy. Hong Kong and Southern China, Hong Kong, 1980, 1. 1983a. A note on the pontoniine shrimp fauna of La Reunion. Bulletin of Marine Science 33(1): 165- 166. 1983b. The pontoniine shrimp fauna of Australia. Australian Museum Memoir 18; 195-218. 1 990. Additions to the marine shrimp fauna of Hong Kong. Pp. 61 1-648. In Morton, B. (ed.), Proceed- ings of the Second International Marine Biology Workshop; The Marine flora and fauna of Hong Kong and Southern China. Hong Kong, 1986, 2. 1991. Shallow water Palaemonoid shrimps from New Caledonia (Crustacea; Decapoda). Pp. 221- 279. In Richer de Forges, B. (ed.), Le Benthos des fonds meubles des lagons de Nouvelle-Cale- donie, 1. Etudes et Theses; Paris, ORSTOM. CHACE, F.A., Jr, & Bruce, A.J. 1993. The Caridean Shrimps (Crustacea: Decapoda) of the Albatross Philippine Expedition 1907-1910, Part 6: Super- family Palacmonoidea. Smithsonian Contribu- tions to Zoology 543: i-vii, 1-252. MORTON. B. & HARPER, E. 1995. An Introduction to the Cape d’Aguilar Marine Reserve, Hong Kong. (Hong Kong University Press: Hong Kong). PATTON, W.K. 1966. Decapod Crustacea commensal with Queensland branching corals. Crustaceana 10(3): 271-295. WADLEY, V.A. 1978. A checklist and illustrated key to the epibcnthic shrimps (Decapoda: Natantia) of Moreton Bay, Queensland. CSIRO Division of Fisheries and Oceanography 99: 1-24. Literature Cited BORRADAILE, L.A. 1915. Notes on Carides. Annals and Magazine of Natural History 15(8): 205-213. BRUCE, A.J. 1977. A new genus of hippolytid shrimp (Crus- tacea: Decapoda: Hippolytidae) for Thor maldivensis Borradaile. Memoirs of the Queensland Museum 42( 1 ): 13-23. STEPHENSEN. K. 1944. Crustacea Malacostrace VHI. Am- phipoda IV. Danish Ingolf Expedition 3(13): 1-51. A.J. Bruce, Crustacea Section, Queensland Museum. PO Box 5500, South Brisbane 4101, Australia: 20 March, 1998. A NEW SPECIES OF FOERSTERELLA DALLA TORRE (HYMENOPTERA: TETRACAMPIDAE) FROM AUSTRALIA CHRISTOPHER J. BURWELL Burwell, C.J. 1998 06 29: A new species of Foersterella Dalla Torre (Hymenoptera: Tetracampidae) from Australia. Memoirs of the Queensland Museum 42(2): 399-402. Brisbane. ISSN 0079-8835. The first Australian representative of Foersterella Dalla Torre, F. australis sp. nov. (Tetra- campidae) is described and distinguished from the similar F, scaposa Boucek from Papua New Guinea. The males of both species are the only known tetracampids with a five-seg- mented antennal funicle. Like other species of Foersterella, F. australis is a solitary egg-parasitoid of a tortoise beetle (Chrysomelidae: Hispinae), parasitising Cassida com- (Boheman). \I\ Hymenoptera, Tetracampidae, Foersterella, Cassida, Australia. Christopher J. Burwell, Queensland Museum, lia; 21 July, 1997. During the course of a study on the parasitoids of some tortoise beetles, Chrysomelidae: Hispi- nae (Cassidinae of some authors), from south-east Queensland (SEQ), an undescribed species of Foersterella Dalla Torre (Chalcidoidea; Tetra- campidae) was reared from the eggs of Cassida compuncla (Boheman). Cassida compuncta lays its eggs singly on the underside of leaves of Ipomoea cairica (L.) (Convolvulaceae) and a large series of the Foersterella sp. was obtained by sweeping the foliage of I. cairica at Long Pocket, Brisbane. Species of Foersterella are known from Europe, Africa, India, Papua New Guinea (PNG) and Australia and as far as known are all en- doparasitoids of the eggs of tortoise beetles (Boucek, 1988). There are no described species of Foersterella known from Australia although Boucek (1988) recorded one female specimen of an undescribed species from Western Australia (WA) in ANIC. This paper gives the first descrip- tion of an Australian species of Foersterella. Morphological terminology follows Boucek (1988). Acronyms for collections are as follows: ANIC, Australian National Insect Collection, Canberra; BMNH, Natural History Museum, London; QM, Queensland Museum, Brisbane; UQIC, University of Queensland Insect Collec- tion, Brisbane; USNM, United States National Museum, Washington. Order Hymenoptera Family Tetracampidae Foersterella Dalla Torre Foersterella contains F. reptans (Nees) and F. erdoesi Boucek from Europe and F. scaposa Boucek from PNG. Boucek (1988) noted unde- PO Box 3300, South Brisbane 4101, Austra- scribed species in Africa, India and Australia. Foersterella flavipes Forster (listed as the type species by most authors) is a junior synonym of F. reptans {BoyitQk, 1992), In Australia, Foersterella can be recognised using Boucek’s ( 1 988) key to the genera of Aus- tralasian Tetracampidae. Foersterella closely re- sembles Tetracampe Forster but can easily be separated by the greatly widened scapes of the males. In addition, the pilosity on the pronotum and mesonotum in Foersterella is sparse and in- cludes pairs of outstanding setae on the pronotum, median lobe of the mesoscutum and axillae (Boucek, 1988). In contrast, the pronotum and mesonotum in Tetracampe are regularly and rather densely pilose, without outstanding pairs of setae (Boucek, 1988). Both F reptans and F erdoesi are endoparasi- toids of the eggs in Cassida Linnaeus (Boucek & Askew, 1 968). The biology of F scaposa is un- known (Boucek, 1988). Further information on Foersterella, including a redescription, synonymies and biological and distributional information, is given by Boucek (1958), Boucek & Askew (1968) and Boucek (1988). Foersterella australis sp. nov. (Figs 1,2) MATERIAL. Holotype $ (QMT4I404), Long Pocket, Brisbane, 27°3 LSI 53°00’E. 1 2.xi. 1 994, C.J. Burwell. PARA TYPES: ( 1 75) Queensland; St Lucia. Brisbane, 27°30’S 153°0rE, 19, 13. ii. 1993, C.J. Burwell, cx egg Cassida (Boheman) (Hispinae) on Ipo- moea cairica', same data as holotype except 1 9 (9.vi.l991), 2d (29.X.1994), 21 d, 149 (3().x.l994). 17d. 69 (12. xi. 1994) (all UQIC); same data as holo- 400 MEMOIRS OF THE QUEENSLAND MUSEUM FIG. I. Foersterella australis sp. nov. paratypes. Long Pocket, Brisbane, 12.xi.l994, C.J. Burwell (UQIC). A, 9 head, anterior view. B, 9 head, dorsal view. C, 9 antenna. D, 6 antenna. E, 9 mesosoma, dorsal view. F, 9 pronotum and mesonotum, dorsal view. Scalebars = 0.1mm. type except ex egg Cassida compuncta (Boheman) on Ipomoea cairica, 19 (29.x. 1994), Id, 19 ( 1 2.xi. 1 994) (all UQIC); same data as holoty pe except. 10. ii. 1996, sweeping Ipomoea cairica, 5d, 59 (ANIC), 5d, 59 (BMNH), 5d, 59(USNM), 61 d, 199 (QMT4I324-41403). DIAGNOSIS. Male with antennal funicle com- posed of 5 segments. Median lobe of mesonotum with pair of outstanding, diverging setae and 5- 1 9, but usually 1 6 or fewer, additional short setae. Lateral lobe of mesonotum with 3-5 short setae and a longer, posterolateral seta directed medially. NEW SPECIES OF FOERSTERELLA 401 FIG. 2. Foersterella australis sp. nov. paratype 9, Lone Pocket Brisbane, 12.xi.l994, C.J. Burwell (UQIC). Forewing. Female. Length 0.75-1. 05mm, mean 0.91mm (n=24). Head, mesosoma and gaster black with green or brassy-green, metallic reflections. Scape and legs light yellow with apical tarsomeres brown. Pedicel and flagellum dark brown, first funicle segment often con- spicuously lighter. Wings hya- line, veins light brown. Head (Fig. lA, B) 1.2-1.35 X as wide as high and 1.5-1.55 wider than frontovertex. Eye height about 0.6 x head height. Malar space 0.5-0. 6 x eye height. Lateral ocelli closer to eye than each other. Lower ocular line bisecting antennal toruli. Eye with sparse setae a little longer than diameter of facets. Face and frontovertex delicately en- graved reticulate, scrobal grooves smooth. Antenna (Fig. 1C) with scape about 4 x longer than wide and combined length of pedicel and flagellum 1.25-1.35 X head width. Pedicel equal to or a little longer than combined length of first and second funicle segments. Funicle 6-seg- mented, increasing in width apically. Antenna weakly clavate, club unsegmented, longer than pedicel, a little more than 2.5 x longer than wide. Mesosoma (Figs IE, F) 1.25-1.40 x longer than wide, dorsal surface, except on propodeum, delicately engraved reticulate with sparse setae on raised papillae. Pronotum bell-shaped, collar not delimited. Pronotum with sparse setae but bare posteromedialiy, posterolaterally with a pair of outstanding, weakly converging setae about as long as pedicel. Mesonotum 2.75-3.5 x wider than long. Median lobe with reticulation trans- versely elongated, usually 6-16 (19 in one speci- men) short setae in anterior half to two thirds and medially to slightly posteromedialiy with a pair of outstanding, diverging setae about as long as those on pronotum. Lateral lobe with 3-5 but usually 4 short setae and a longer posterolateral seta directed medially. Axilla with a long seta on dorsal surface and 1-3 short setae. Scutellum 1.1- 1.2 X wider than long, with reticulation longitu- dinally elongated, with 2 pairs of long, converging setae, the anterior pair about as long as antennal club and situated about medially. Scutellum slightly depressed just posteriad of posterior pair of setae, forming transverse line. Propodeum smooth with 2-8 short setae an- teromedially, callus with numerous setae. Fore wing (Fig. 2) 2. 6-2. 8 x longer than wide. Submarginal vein with 4 setae on dorsal surface. Marginal vein 1.05-1.20, 5.0-5. 5 and 1.15-1.35 X as long as costal cell, stigmal vein and post- marginal vein respectively. Gaster ovate, slightly longer than mesosoma, 1.5- 1.8 X longer than wide in critical point dried specimens. Male. Length 0.74- 1. 02mm, mean 0.86mm (n=39). Similar to female except scape light brownish-yellow, brown dorsally. Pedicel, club and usually third funicle segment dark brown. Fourth and fifth funicle segments light brown, second segment light brownish-yellow, first seg- ment light yellow. Head 1.25-1.35 X as wide as high, 1.55-1.65 x as wide as frontovertex. An- tenna (Fig. ID) with scape greatly widened, 1.55- 1 .67 X longer than wide. Combined length of pedicel and flagellum 1.3-1. 4 x head width. Funicle 5-segmented, segments about subequal, except fifth a little wider. Antenna distinctly clavate, club about 3.5 x longer than wide. Mesosoma 1.20-1.45 x longer than wide. Mesonotum 3.05-3.55 x wider than long, with 5-16 short setae. Scutellum 1.05-1.15 x wider than long. Gaster about as long as mesosoma, 1.45-1.75 X longer than wide in critical point dried specimens. BIOLOGY. Foersterella australis is a solitary endoparasitoid in the eggs of the tortoise beetle Cassida compuncta which feeds on several spe- cies of Ipomoea L. and Polymeria calycina R. Br. (Convolvulaceae) in SEQ. The beetle lays its eggs singly, within membranous cases, on the under- side of leaves. Only eggs laid on /. cairica have been found parasitised by F. australis. A second species of tortoise beetle, Cassida diomma 402 MEMOIRS OF THE QUEENSLAND MUSEUM Boisduval, shares two larval-pupal parasitoids with C compuncta and although F. australis has not been found parasitising C diomma eggs to date, it is probably also a host. DISCUSSION There is a single female Foersterella from WA (4km SW Mining Camp, Mitchell Plateau, 14°52’S I25°50’E, 13.V.I983, I.D. Naumann, J.C. Cardale, ANIC) which closely resembles the specimens from SEQ. However until I see addi- tional material, particularly males, T am reluctant to assign the WA specimen to F. australis. Foersterella australis is closely related to F. scaposa Boucek from PNG. The antennae of males of both species are very similar and have only five funicular segments (Fig. I D), a synapo- morphy. Males of F. erdoesi and F. reptans have a 6-segmented funicle as do males of all other known tetracampids (Boutek 1988). Foersterella australis differs from F. scaposa by its less setose mesonotum. Each lateral lobe of the mesonotum of F. australis has at most 5 short setae and the median lobe at most 19 but usually 16 or less short setae (Fig. IF). The unique holo- type male of F, scaposa (in ANIC) has 14 short setae on the left lateral lobe and 25 short setae (one lost) on the median lobe of the mesonotum. The more densely setose mesonotum of F. scaposa is unlikely to be due to allometry as the body length of the holotype falls within the range of males of F. australis. In addition the mesonotal setae of F. scaposa are stronger and inserted on more prominent papillae than those of F. australis. Variation in mesonotal setation of F. scaposa is impossible to assess until additional specimens become available. At such time a re-examination of the separate status of F. scaposa and F. austra- lis may be necessary. ACKNOWLEDGEMENTS I thank Marlene Elson-Harris and Kylie Stumkat for preparing the scanning electron mi- crographs and Ian Naumann (ANIC) for the loan of specimens. Elizabeth Exley and David Yeates made useful comments on the manuscript. Geoff Thompson helped with the preparation of Fig. 2. LITERATURE CITED BOUcEK, Z. 1958. Revision der europaischen Tetra- campidae (Hym. Chalcidoidca) mit einem Kata- log der Arten der Welt. Revise evropskych Tetracainpidae s katalogcm druhu celeho sveta. Acta Entomologica Musci Nationalis Pragae 32: 353-404. 1 988. Australasian Chalcidoidca (Hymenoptera). A biosystematic revision of genera of fourteen families, with a reclassification of species. (C.A.B. International: Wallingford). 1992. A correction to a recent publication (Hym., Pteromalidae). Entomologist’s Monthly Maga- zine 128: 32. BOUCEK. Z. & ASKEW, R.R. 1968. Index of world Tetracampidae. Pp. 3-19. In Delucchi, V. & Re- maudiere, G. (eds). Index of Entomophagous In- sects 4. (Le Francois: Paris). A NEW TOOTH-PLATED DIPNOAN FROM THE UPPER DEVONIAN GOGO FORMATION AND ITS RELATIONSHIPS K.S.W. CAMPBELL AND R.E. BARWICK Campbell, K.S.W. & Barwick, R.E. 1998 06 29: A new tooth-plated dipnoan from the Upper Devonian Gogo Formation and its relationships. Memoirs of the Queensland Museum 42(2): 403-437. Brisbane. ISSN 0079-8835. The commonest dipnoan species from the Frasnian Gogo Formation are Chirodipterus australis Miles and Griphognathus whitei Miles. Less abundant species are Holodipterus gogoensis Miles and other holodipterids, Pillararhynchus longi Barwick & Campbell, and Gogodipterus paddyensis (Miles). One would have expected many more representatives of tooth-plated forms of the Dipterus-Xypc in the Late Devonian Gogo faunas. Recently a small number of individuals belonging to the tooth-plated group have been discovered, two of them being adults and one juvenile. They have been prepared to show how tooth-plates of this kind grew, and the outstanding features they produce in adults. They have been assigned to Adololopas moyasmithae gen. ct sp. nov. Histology of the loolh-plates has been investigated optically and by SEM. The cores of the teeth are composed of hypermineralised dentine with abundant dentine tubules; the pallial dentine is thick in the distal teeth where it is supported on a ring of pedestal dentine. The heels of the plates are composed of solid dentine of the kind that forms the core of the teeth. The basal pulp cavity is continuous beneath the tooth rows, but it is Joined to the underlying bone at well defined patches. A Late Devonian (Frasnian) Canadian genus Scaumenacia. has been studied with a view to adding further information to a wider study of tooth-plate evolution. Consideration is given to the reasons why long-headedness has to be further analysed before it can be used for taxonomic purposes. □ Devonian, Dipnoi, tooth-plates. Gogo Formation. K.S. PF Campbell dc R.E. Barwick. Department of Geology, Australian National University, Canberra. A.C.T. 0200, Australia; 5 January, 1998. Acid-etched nodules from the Gogo Formation occasionally contain dental plates carrying real teeth. These teeth are not associated with intact skulls, as is usual for Gogo dipnoans; one well- preserved body indicated that a complete skull should have been present, had it been well-ossi- fied. We have been reluctant to proceed with a description and taxonomic designation until more data on the skull roof became available. We have now etched most of the material available to us, and no well-ossified skull has appeared with teeth like those of the fragmentary material; conse- quently we are describing this material on the assumption that it will be some time before fur- ther specimens are discovered. The new informa- tion, though incomplete, is valuable for interpreting the Late Devonian history of the Dip- noi. Three specimens are available — an adult and a Juvenile held at the Geology Department, Aus- tralian National University (ANU) and an adult in the Western Australian Museum (WAM). The ANU adult was prepared from a nodule without first embedding it in plastic. As a result, all the parts are free and available for reassembly. The WAM specimen was split along the axis of the fish at the time of collection. Because of this, it was embedded in plastic prior to etching, so that it could be better held together during prepara- tion. Subsequently, the anterior part of one side of the preparation was treated with chloroform to loosen the plastic, and several isolated bones and a pair of tooth plates were obtained. These have proved especially valuable in understanding the histology of the tooth-plates. In the following discussion we have been able to provide separate descriptions of the macromorphology, micromor- phology and histology of the tooth-plates of our new species. SYSTEMATIC PALAEONTOLOGY Order SPEONESIDRIONINA Campbell & Barwick, 1990 REMARKS. We continue to reject cladistic analyses that use all the so-called synapomorphic characters as of equal value in establishing rela- tionships. We contend that a complete analysis of a group oftaxa is much more likely to be achieved by an examination of a number of related func- tional characteristics. Our 1990 analysis indicated that during the Early Devonian, dipnoans used three different methods of covering the palate and 404 MEMOIRS OF THE QUEENSLAND MUSEUM prearticular bones with dentine-bearing protuber- ances. Denison (1974: 32) indicated that early dip- noan types had very different dental apparatuses, and that there was little indication “that ‘plateless’ genera were derived from nor closely related to those with dental plates". We consid- ered that the three dental types referred to above, tooth-plated, dentine-plated and denticulate, are quite different in the organisation of other fea- tures in addition to the teeth, and once they were established they apparently did not alter from one type of organisation to another. This approach was attacked mainly because we did not regard the presence of denticles as always indicating a primitive condition, as though we had proposed the presence or absence of denticles in the mouth as the sole criterion of relationships. Although we used the dominance of denticles as a basis for naming one of the types, nowhere did we espouse the view that the presence of denticles is the sole defining feature of the group. To the contrary, we have pointed out that each type is organised so as to have a number of features correlated with the different types of food reduction (Campbell & Barwick, 1987, 1990; Barwick, Campbell & Mark-Kurik, 1997). One of the three types outlined above had tooth- plates formed by teeth that were arranged in rows; new teeth having been formed at the distal end of each row. Members of this group we assigned to the Order Speonesydrionina. The teeth were de- scribed as having a cover of enamel (enameloid) over a thin layer of pallial dentine, which covered a thick and dense layer of heavily mineralised dentine. Dental plates of this type we have pre- viously termed tooth-plated, as opposed to the dentine-plated types that had no true teeth at the plate margins (Smith & Campbell, 1987). In a further criticism of our basic approach, Kemp (1992) has pointed out that some of the genera we assigned to non-denticulate groups, do in fact have teeth in the early growth stages, but have lost them in adulthood. She referred to the Late Devonian lungfish fossils from the Coffee Hill Member, Catombal, near Orange, NSW in which juveniles, up to 3mm x 4mm, some of which may have been Chirodiptems, may have had teeth in the early growth stages. In 1994 she considered that ‘many lungfish of Devonian age, like Chirodipterus, previously regarded as having solid dental plates, are in fact built up from cusps that fuse at their bases to form tooth plates with radiating ridges. ... Chirodiptenis is definitely one of the tooth plated lungfishes’. Nowhere has she shown that Chirodipterus occurs in the Ca- tombal fauna, a difficult thing to do in the absence of skeletal features other than tooth-plates. Until her material is described, and evidence of the existence of Chirodipten4s of the C. australis type in the Catombal faunas is provided, we do not see that our analysis is in any way effected. FAMILY INCERTAE SEDIS REMARKS. In 1990 we separated late Palaeo- zoic tooth-plated dipnoans from Dipterus be- cause they had no separate K bone, no lateral line in bone J, no buccohypophysial canal, unossified otic part of the braincase, and a partly buried pectoral girdle. Adololopas has a separate K, a canal in J, and an unossified braincase, but it lacks a buccohypophysial canal opening, and its girdle is largely buried. Hence it does not fall easily into the Dipterus group or the alternative non-Dip- terus group on the characters defined in 1990. This view is supported by the elongate size of the cheek in the new genus Adololopas^ the peculiar position of the lateral lines on the posterior part of the head, and the advanced nature of the dental plates with large basal pulp cavities. In addition it has a lateral line emerging from bone I to run along the body, a feature not observed in any other tooth-plated form apart from Gnathorhiza. We do not have access to the external parts of the mandible and the anterior part of the mandible which would also enable us to clarify the phylo- genetic relationships. All this illustrates the diffi- culty of defining larger taxonomic groups on a number of characters which evolve considerably during time. Whatever the outcome, we now ac- cept that Adololopas cannot be placed definitely in the Dipteridae, but we will not make a new Family to include it, because many features of the genus cannot be determined on present material. Adololopas has a more advanced tooth-plate structure than Dipterus. Although Dipterus has for many years been regarded a member of the tooth-plated group, and forms assigned to it are very widely known throughout the world, the dental characteristics of specimens from the type locality are not very well preserved. Smith ( 1 984) described the teeth of Dipterus valenciennesi from Scottish specimens. No other Middle De- vonian genus showing the details of the tooth- plates has been described, but Elga Mark-Kurik has given us some incomplete Middle Devonian specimens. These are being described elsewhere in a separate paper (Campbell & Barwick, in press). These specimens provide an example of DEVONIAN TOOTH-PLATED DIPNOI FROM W.A. 405 the extent to which the plate structure had evolved during the Middle Devonian. Adololopas gen. nov. ETYMOLOGY. Greek adolos, true; Greek lopas, plate. TYPE SPECIES. A. moyasmithae sp. nov. from the Frasnian (Upper Devonian) of the Gogo Formation, Canning Basin, Western Australia. DIAGNOSIS. Endocranium poorly ossified. Dermal roofing bones not tightly sutured to- gether, cosmine covered; bone B long and nar- row, and with scalloped margins; bone I transmitting a lateral line canal from Yi to bone Z; no Y 2 preserved, but judging from the position of 8 and the shape of the opercular, it must have been present; bones 4, 5 and 8 unusually long for a tooth-plated form of this age; snout well-ossi- fied. Adult pterygoidal tooth-plates thick, with an open median suture and with 6-7 rows of teeth; new conical teeth added at ends of rows, and mounted directly on bone; mandibular tooth plates showing teeth added sufficiently deeply to be out of occlusion but growing up into occlusion as they increase in size; adult tooth-plates with thin primitive dentine and serrated edges along the posterior margins; palatal tooth-plates con- cave, and mandibular plates convex. Juvenile tooth-plates with six or seven rows of sharp teeth; continuous rings of enamel- or enameloid-cov- ered primitive dentine around the posterior, lat- eral and mesial edges. Parasphenoid with ploughshare shape, pointed posterior extremity and closed buccohypophysial canal lying in its anterior keel. Dentary narrow; angular bone with a deep notch posteriorly for articulation of the surangular. Body short and rapidly tapering; scales similar in form to those of Chirodiptenis australis, but their size diminishes rapidly to- wards the tail and the dorsal margin. REMARKS. This genus is noteworthy in having the lateral line canal running from bone I on the skull roof to the body. The only other genera in which this is known are Gnatharhiza andJarvikia (Campbell & Barwick 1990, figs 4, 5). This inter- pretation of Jarvikia has been questioned by Schultze & Marshall (1993), who claimed that the connection of the main lateral line of the body with that of the skull roof through bone 1 is unique for Gnatharhiza. The specimen that shows the detail for Jarvikia is the paratype of J. arctica Lehman (1959, pi. 20B, fig. 22). One of us, KSWC, has examined Lehman’s specimens, and considers that his photograph gives an incomplete impression of the lateral line canals. This inaccu- racy is also shown in his figure 22. Herein we produce new photographs of the same specimen showing the canals effectively (Fig. 2A,B). The canal runs posteromesially from the bone we interpret as Y into bone 1 (Campbell & Barwick, 1990, fig. 5). There the canal splits to produce a branch across bone B, and a second one back- wards lateral to the process on the posterior end of the bone I, where it Joins the main body lateral line canal. There is no other posteriorly directed canal from Y. The situation described is certainly unusual for the dipnoans, but we reject the state- ment of Schultze & Marshall (1993, p. 216) that our view on this feature is a ‘mistake in our morphological analysis’. The new photographs show that the canal is clearly present as we have indicated. The palate of Jarvikia is not well known; Lehman (1959) recorded that the genus did not have tooth plates, but probably had a denticulated palate. One of us, KSWC, has been able to confinn that Judgement by observation of the original specimens. The arrangement of the overlap between the pterygoids and the peculiar relationships of the parasphenoid, demonstrate X\\aiJarvikia and Adololopas are distantly related. We see no reason to group these genera together on the basis of the posterior lateral line exit from the skull roof On the other hand, the late Palaeo- zoic Gnatharhiza has totally different Y bones, and it has no lateral line canal in bone J; in addition it is unique in many other respects with which we dealt in 1990. As has been indicated above, the second feature distinguishing Adololopas from other cosmine- bearing genera in the Gogo fauna is the presence of real teeth (Fig. 4). This takes it out of the group of Chirodipterus and Pillararhynchus which also occur in the Gogo fauna. The external dermal bones of the mandible of Adololopas are inadequately known, preventing detailed comparison with the bones of Chirodip- terus and Pillararhynchus. However, the angular (Fig. 8 ) has a deep and distinctive notch for the surangular, which must have had a different shape from the surangulars of Chirodipterus and Pil- lararhynchus. We also infer that a splenial and post splenial were present in Adololopas, because of the shape and position of the anterior end of the angular. Thus it would be different from Pil- lararhynchus in this respect also. The arrangement of the scales is different from any other known Devonian dipnoan, in that they decrease dramatically in size a short distance behind the head and towards the dorsal margin. This indicates that, relative to its head size, the body was unusually short (Fig. 11). The wider 406 MEMOIRS OF THE QUEENSLAND MUSEUM significance of this feature is unknown, because so many dipnoans are known from heads only. Adololopas is much shorter in the body than the other Gogo dipnoans, Chirodipterus, Griphognathiis and Hoiodipterus in which we have adequate body preservation. Pillararhyn- chus is more poorly known. Diptenis also is long bodied. Finally, in connection with affinities, we note that Adololopas has a long cheek and a short snout. This type of structure was outlined by Schultze (1992), in his description of lowadip- terus, from the Givetian of Iowa, USA. It pro- vided the basis on which he discriminated two types of long headed dipnoans, one type such as Griphognathiis and Rhinodipterus which had long snouts, and the other such as lowadipterus and Hoiodipterus, which had the post-orbital part of the skull elongated. Whereas we have no diffi- culty in distinguishing different types of long- headedness, we are unaware of its significance. Adololopas has long bones 4 and 5 and, judging from the shape of the snout and the palate, it would have been short-snouted. According to Schultze’s criteria, it would be placed in the long- headed but short-snouted group. The problems raised by this assessment are dealt with below under the discussion at the end of the description. Adololopas moyasmithae sp. nov. ETYMOLOGY. In honour of Dr Moya Meredith Smith who has made a major contribution to the study of Palaeozoic dipnoan dentitions. MATERIAL. HOLOTYPE: ANU49213, an incomplete specimen with part of the skull preserved. PA- RATYPES: WAM93 1.1.221, a specimen with part of the body and some skull, mandible and tooth plates preserved; and ANU492I5, a small incomplete snout with the palatal and mandibular tooth-plates preserved; all from the Gogo I’ormation (Frasnian), Paddys Spring Valley. The two adults are so similar in structure of the tooth-plates and the cosmine on the external plates, that we see no reason to think that they are different species. The juvenile has tooth-plates in which the teeth are added at the ends of the tooth rows, successive growth of tissues at the tooth-plate margins similar to the adults, and with 6 or 7 tooth row's. All of these features are distinct from any other members of the other dip- noans in the fauna, that we can only consider all three specimens to be conspccific. DIAGNOSIS. As for the genus. DESCRIPTION. Skull Roof. Roofing bones are weakly joined at sutures, falling apart readily during etching (Figs 1, 4H,I, 5). Cosmine covers bones, except at the bone margins. Bone B is elongate, with a scalloped margin and with a pair FIG. 1. Adololopas moyasmithae gen. et sp. nov. An outline sketch of the skull roof and part of the cheek drawn from the hoIot>'pe, Fig. 4G-L The cheek is situated in relation to the skull because the lateral line in 4 is displayed in X-rays and indicates the approxi- mate position of the centre of ossification of X, and Y is placed to receive the lateral line from 1. Scale = 10mm. of centrally situated pit-lines; thickening for the attachment of the median crista is restricted to the anterior half of the plate. Bone I is transverse; posterior process on I for articulation of the ano- cleithrum is small; position of operculum and bone 8 indicate that Y 2 must have been present, and was loosely articulated (Fig. 3C); bone Z fitted into a groove in the posterior and postero- lateral margin of L Lateral line canal enters bone I anterolateral ly, runs to its centre of ossification, and then turns abruptly posterolaterally to exit into a loosely articulated Z (Figs 1, 5). Two pit-lines present on bone I, an elongate transverse one and a short posterolaterally oriented one, the two not meeting. Cartilage attaching dorsolateral crista to roof present only to the anterior part of I and extends on to posterior part of J. Anterolateral edge of I has a deep angular embayment, defining the edge of the Y i bone into which the lateral line passes. Bone A is transverse, roughly hexagonal, and carrying the occipital commissure; numerous sub-branches of the canal extend backwards as shown on the X-radiographs (Fig. 5); this com- missure continued through bone I as a short arc. Elongate bone J is preserved on both sides and carries a lateral line canal from its centre antero- laterally to an unpreserved bone; this must have been a bone K, which was a large bone whose outline is best defined on the left side; an elongate DEVONIAN TOOTH-PLATED DIPNOI FROM W.A. 407 FIG. 2. Jarvikia arctica Lehman. This is the specimen figured by Lehman (1959, PI. 20B). A, posterior part of the roof figured dorsal ly. B, enlargement of same, tilted to the right and laterally illuminated to show the lateral line canals, labelled Ic. The centres of ossifi- cation of bones B. I and Y are labelled. The lines at the back of the specimen were produced by needle scratches. The bones are labelled and the canal runs off the back of the skull as shown. The extension at the back of bone I is incomplete. Scales = 10mm. pit-line is also present on J. Bones C elongate and so shaped that any D present would have been small; other bones in front of Cs and on the cheek unknown, apart from some circumorbitals; lateral line canal present on bone 3; bone 2 short; bone 4 unusually long for a dipterid-Iike animal (Figs 1 , 4G), extends from orbit almost to the opercular. FIG. 3. Comparison between the lateral views of A, lowadipterus. B, Holodipterus. and C, Adololopas. lowadipterus halli is redrawn from Schultze (1992, fig. 10); Holodipterus is drawn from H. (Asthe- norhynchus) meemannae (Pridmore et al.. 1994). Adololopas moyasmithae is reconstructed from ANU49213. Note the difference in the positions of bones 4, 5 and 8. Lateral line enters 4 from an anterodorsal rather than a posterodorsal direction, indicating that X is situated further forward relative to the orbit than in other dipterid-like animals; contact be- tween X and J defined by projections on both sides of the specimen. Bone 8, which has been recognised by the angular bend in the lateral line and the finished edge against the opercular, is also longer than in other dipterids. Orbit 1 8% of length of skull from rear of B to front of C. We have two isolated snouts, an adult from ANU49213 (Fig. 4J, K) and juvenile ANU492I5 (Fig. 7A, B). Adult broad, and showing all the main features of other Devonian dipnoans of the 408 MEMOIRS OF THE QUEENSLAND MUSEUM FIG. 4. Adololopas moyasmithae gen. et sp. nov. holotype. A, B, ventral and dorsal views of the prearticulars and tooth plates. C, D, dorsal and ventral views of the pterygoids and tooth plates. pmla= pterygoid-mandibular ligament attachment. E, F, dorsal and ventral views of parasphenoid isolated during preparation. Note the foramen in the anterior end for the buccohypophysial canal. G, lateral view of the orbit and the cheek plates. H, I, ventral and dorsal views of the posterior part of the skull roof. The left side of I is incomplete, but the bones on the right have complete edges. Note the attachment of the dorsolateral crista beneath the anterior end of bone I and the posterior end of bone J on the left side of H. .1, K, posterior and anterior views of the isolated snout. Note the variation in the size of pores in the snout in K. Scales = 1 0mm. DEVONIAN TOOTH-PLATED DIPNOI FROM W.A. 409 FIG. 6. Adololopas moyasmithae gen. etsp. nov., holo- type. Isolated tooth plates from the anterior palate. A, median anterior plate showing the lines associated with new growth. B, C, pair of dermopalatine plates (in C the mesial anterior end is broken off); both show growth lines. D, E, another pair of plates of unknown origin from the same etch. Scales = 1 .0mm. FIG. 5. Adololopas moyasmithae. gen. et sp. nov. X- radiograph of the posterior part of the skull of the holotype, photographed in ventral view. The entry points of the lateral lines into the bones are marked with arrows. Note also the branching of the lateral line canal in bone A (see Figs 1 & 4H). Scale = 1 0mm. dental-plated types; external nares well separated, leaving only a small edge of tissue between them and the lateral labial pit; lateral line canals defined by the enlarged pores; juvenile snouts more highly arched than those of adults. Palate. Pterygoid thinnest towards the anterior margin, thickening gradually posteriorly; dorsal palatal process present but depressed, with a bev- elled dorsal surface inclined downwards laterally and a sharp crest along its mesial margin; bevelled surface without a periosteal layer, and exposing coarser textured bone than elsewhere (Fig. 4C); a rounded ridge adjacent to the mid-line increasing in height posteriorly and turning posterolaterally along the quadrate ridge; towards the mandibular articulation the quadrate ridge becomes narrower and sharper; lateral extremity turned down sharply towards the mandibular articulation; scar for pterygo-mandibular ligament well developed, and laterally placed on the quadrate. Included with ANU49213 are five small iso- lated palatal plates, one being a median anterior and two others being symmetrical dermopalatines (Fig. 6); the remaining two are possibly other anterior tooth plates (Miles, 1977, fig. 72); me- dian plate convex ventrally and carrying incre- mental layers of primitive dentine, forming marginal concentric outlines (Fig. 6 A); these out- lines do not continue around the entire margins without breaks; however, their distribution indi- cates that the median plate was not in contact with the adjacent pterygoid plates; central parts of all small plates include areas within which the den- tine is darkened by apparently hypermineralised dentine. Parasphenoid of ANU49213 well pre- served (Figs 4E,F); anterior face high and plough- share shaped; buccohypophysial canal situated in anterior corner, but does not open into the buccal cavity; posterior part of plate acutely triangular in outline, and with a long crest along its dorsal surface; central part of plate consisting of an open meshwork of bone. Neurocranium unossified except at the anterior margin where tubules related to the anterior sen- sory systems are present. Mandible. Mandible known from prearticulars and tooth plates of two adults and one juvenile, a small adult fragment and one adult angular. Lin- gual furrow of adult deep and barrel shaped (Figs 4A,B, 22A,B), slightly overhung by the tooth- 410 MEMOIRS OF THE QUEENSLAND MUSEUM FIG. 7. Adololopas moyasmithae gen. et sp. nov. A, B, snout of ANU492 1 5 in anterior and posterior views showing the high curvature. External naris labelled; pores for anterior commissure clear. C, internal view of broken dentar>' of the holotype. Note the edge of cosminc on the lateral and posterior edge, indicating the fact that it had no continuous bone Junction with the suitounding bones. Scales = 2.0mm. plates, and with a well rounded floor; junction between the two plates formed of bone without dentine; strong, rather sharp lateral projection of prearticular in front of the glenoid fossa (Figs 4B, lOB); pterygo-mandibular ligament attachment producing a small flange immediately medial to the glenoid fossa; between this flange and the back of the tooth plates is a deep gap for muscles connecting to the ceratohyal (Fig. 4A,B). Small isolated incomplete dentary (Fig. 7C) shows the posterior end completely isolated from the in- fradentaries, and having some small blisters of bone indicating that addition to the edge of the bone was taking place. Angular, represented by an isolated external bone of WAM93 1 . 1.22 1 , with an edge indicating that it bordered the adductor pit, and carries two lateral line canals (Fig. 8A,B); shape of this in- fradentary is unusual, having a deep posterior notch in which the lateral line emerges at its apex; lower lateral line enters the bone about 5mm from the posterior edge and runs fomard slightly, con- verging on the other canal; position of infraden- tary bone can be determined from the opening for the adductor fossa and the large upturned surface forming the contact with the process on the prear- ticular in front of the glenoid fossa; shape of the surangular notch and the position of the edge of the adductor fossa, indicate that the surangular was short; shape and position of angular suggests that both splenial and postsplenial were present to occupy the space between the mid-line and the angular. Pectoral Girdles. Two pectoral girdles, one more complete (Fig. 9A,B) from ANU49213, and the cleithrum only on WAM931.1.22I (Fig. 9C); three other smaller fragments are known. Cleithrum with a deeply concave external sur- face, its anterior rim standing high, sharply de- fined and with (Figs 9C, 1 1 A,B) or without (Fig. 9A,B) a large number of pustules; posterior half of external surface covered with moderately de- veloped rough surface features; ventral extremity acute and with an angular excavation for the articulation of the clavicle; branchial lamina standing at a high angle to the external lamina, concave on its anterior face, and carrying a num- ber of sharp projections into this concavity; dorsal anterior notch for surangular FIG. 8. Adololopas moyasmithae gen. et sp. nov. A, external and B, internal views of an isolated angular bone from WAM93 1.1.221. Position of mandibular and oral canals marked by arrows. Notch for surangular with a margin unaltered by etching. Opening for adductor muscles obvious at top of photograph. Scale = 10mm. DEVONIAN TOOTH-PLATED DIPNOI FROM W.A. 411 branchial -cleithral lamina subopercular 1 C broken clavicular branchial lamina FIG. 9. Adololopas moyasmithae gen. et sp. nov. A, B, lateral and anterior views of an incomplete right pectoral girdle of the holotype. In A and B the branchial lamina on the clavicle is incomplete. C, cleithrum and the subopercular 1 of WAM93 1.1.221. Note the difference in ornamental pattern between this specimen and that of A. Scales = 10mm. edge of lamina meeting the anterior edge of cleithrum about 8mm short of dorsal edge; ventral edge of branchial lamina thickened and bearing a distinct groove for the articulation of the clavicu- lar branchial lamina. Clavicle with a much less concave external lamina and a weaker anterior rim than the cleithrum; ornament well developed on the anterior rather than the posterior half; ventral end of external lamina with a cutaway edge that continues up into the branchial lamina, producing an over-lapped surface presumably for the interclavicle; branchial lamina incomplete, but extending nearer to ventral edge than in Chi- rodipterus australis; originally it continued up to meet the cleithral branchial lamina along its entire length. Operculars and Gulars. Opercular subquadrate in outline; leading edge almost straight ventrally (Fig. 10A,B), but turned posterodorsally on the upper edge; in anterior view, dorsal part of oper- culum bent medially; anterior edge of this in- turned part of bone abutts the posterior edge of the Y2 bone; the inward turning of the operculum demonstrates that the rotational edge lay ventral to the upper edge of bone 8; this articulatory edge with slight overlapped margin, indicating the po- sition of movement against the skull; posteroven- tral edge also slightly truncated; dorsal edge also overlapped; surface with strong Westoll lines. Subopercular 1 with a strongly overlapped dorsal edge and a slightly ventrally curved ventral edge (Figs 9C, lOB); remainder of surface cov- ered with cosmine apart from marginal resorp- tion. Subopercular 2 not well-known, only partly exposed (Fig. lOB), but with overlap edges on both dorsal and ventral edges. Principal gulars (Fig. lOB) generally ovate in outline, but with a strong point anterolaterally; about 18mm long and 1 8mm wide; anteromesial edge with a long undivided overlap which indi- cating a large median gular; lateral overlaps not completely clear, but indicating a submandibular and possibly a lateral gular; three Westoll lines on the surface; a transverse pitline 4mm long situated anteromesially on the plate. Pectoral Fins. External ornament of the right pectoral fm partly preserved on WAM93 1 . 1 .22 1 (Fig. 1IA,B); individual scales small, cosmine 412 MEMOIRS OF THE QUEENSLAND MUSEUM left mandibular plate opercular j subopercular 1 subopercular 2 FIG. \0. Adololopas moyasmithae gen. et sp. nov. A, an isolated operculum of holotype; anterior to the right. B, WAM93I.1.221, anterior part of specimen showing opercular, gular, subopercular I and 2 and mandibular plates. Scales = 10mm. covered but with a small overlapped edge on some individuals; preserved part of fin 22mm long; occasional scales preserved on the inside of the limb (Fig. 1 1 B); individual scales highly vari- able in outline; overlaps slight; probably only two or three rows of scales down the midline of the fin, but there may have been a smaller row on each side; no rays preserved. Scales. Scales thin, cycloid in shape, and almost equidimensional in outline; inner surface with concentric growth lines; exposed surface with cosmine; size of scales decreases rapidly in both dorsal and posterior directions; on the flanks only 100mm behind the head, the length of the cos- mine cover is only half of the forward scales (Fig. 1 1 A); scale overlap is large (Figs 1 lA, 12). Scales immediately posterior to the cleithrum show no sign of cosmine (Fig. 1 1 A,C). Like the cleithrum itself, they must have been buried beneath a thin layer of soft tissue. DESCRIPTION OF TOOTH-PLATES. Macro morphology of Juveniles. At the early growth stages, the palatal plates show very narrow bands of shiny (probably enamel-covered) dentine along the posteromesial edge, and these abruptly expand to form large flat surfaces anteromesial- ly (Figs 13, 14). These bands are continuous around the posterior edge of the plate to the lateral margins where the inner ones progressively dis- appear, or can be faintly traced turning inwards between the teeth. The two outermost bands are quite clearly in two discontinuous units (Fig. 13C,D). The posteriormost one is replaced by another band which continues around the an- teior margins of the plate, and carries the suc- cession of teeth upon its surface. The band swings back between the teeth on the anterior edges (Figs 1 3 A, 14C). There is no precise correspondence between the marginal bands and the addition of new teeth (Figs 13A,B). It is obvious that six teeth are added while four mar- ginal bands were growing (Fig. 13B). The situ- ation on the right plate is not so clear because of weathering, but the bands of tissue running across the plate surface are well marked (Fig. 13 A). There is no sign of clear resorption and replace- ment of any bands, but more mineralised dentine of the plate has grown into some of the inner bands in places. DEVONIAN TOOTH-PLATED DIPNOI FROM W.A. 413 FIG. I \ . Adololopas moyasmithae gen. et sp. nov. A, body of WAM931.I.22I; note the rapid reduction of scale size from front to back and from ventral to dorsal sides. B, cleithrum and right pectoral appendage; note the ornament on the cleithrum. C, enlargement of scales immediately behind the cleithrum; note that the more anterior scales lack cosmine and are covered with pustules. One cosmine-bearing scale in particular has pustules along its posterodorsal edge. Scales A-B = 10mm; C = 5mm. The lingual face of the juvenile mandibular tral and the oldest against the plate proper (Fig. plates stands almost vertically, and is covered by 1 5C,D). In the lingual furrow they have a distinc- a sequence of four layers of enamel-covered den- tive shape; their ventral edges are slightly curved tine, the most recently added of them being ven- except towards the anterior end where they bend right pectoral appendage 414 MEMOIRS OF THE QUEENSLAND MUSEUM FIG. 12. Adololopas moyasmithae gen. et sp. nov. A, computer reconstruction of lateral flank scales drawn on the basis of one scale in dark outline in the centre of the illustration; cosmine coloured black. Cosmine of adjacent scales grey, and surrounded by open areas which indicate bone covered with soft tissue. Twenty eight scales, which potentially were in contact with the main central scale, have been drawn in. B, computer reconstruction of lateral flank scale, cosmine-covered area in light grey, with the estimated pattern of edges of overlapping scales. Numbers refer to the potential number of overlapping scales at particular points on the scale. DEVONIAN TOOTH-PLATED DIPNOI FROM W.A. 415 FIG 13 Adololopas moyasmithae gen. et sp. nov. Palatal-tooth plates of the juvenile, ANU492I5; right palatal plate (from top to bottom) in anteroventral, medial ventral, ventral and marginal views. A, note the marginal teeth lying directly on bone, and layers of enamel-covered tissue adjacent to the mid-line and bending inwards behind the marginal teeth. B, C. bands of enamel-covered material around the posterior medial and lateral edges, show particularly well. D, shows the break between the posterior band of tissue and the tooth bearing band in front. Scales A-D=l mm. 416 MEMOIRS OF THE QUEENSLAND MUSEUM FIG. \A Adololopas moyasmithae gen. et sp. nov. Palatal tooth-plates of the juvenile ANU49215. A, marginal view showing teeth and marginal bone. B, C, two palatal tooth plates of the one specimen. B, slightly destroyed by erosion on the original specimen. Scales = 1mm. DEVONIAN TOOTH-PLATED DIPNOI FROM W.A. 417 FIG. 15. Adololopas moyasmithae gen. et sp. nov. Mandibular tooth-plates of the juvenile ANU49215. A, B, dorsal views of two plates; note the exposed bone in the mid-line. C, D, lingual views of the left and right plates showing the enamel-covered bands. E, view of the posterior lateral edge of A. Scales = 1mm. 418 MEMOIRS OF THE QUEENSLAND MUSEUM FIG. 16. Aclololopas moyasmithae gen. ct sp. nov. Outline drawings of the juvenile specimens of tooth plates in Figs 13-15. A-C, three views of the left palatal plate showing the marginal bands numbered 1-4. D, E, lingual views of the left and right mandibular plates, again showing the marginal bands 1-4. sharply upwards and then loop forwards on to the flanks of, or around, the last-foimed tooth. This gives the superficial impression that a new layer is formed contemporaneously with the addition of a new tooth, but that is not correct. As is shown by an examination of the anterolateral margin of the plate, new teeth are added directly on to bone and then a layer of dentine was deposited around the distal edges of the teeth. These bands were continued around the posterior margin of the plate, though some of the earlier formed ones were overlapped by the later ones. Only the last formed continues around the lateral edge of the plate, and it turns in behind the distal tooth on the third row and then continues in the furrows proxi- mal to the distal tooth on the second and first rows. Macromorphology of Adults. The adult palatal tooth plates of two specimens are known. The better preserved one (Figs 4C,D, I7A-D), has 6 well-formed rows of teeth with an indication of a seventh. In ventral view, the posteromedial edge of the plates against the median suture shows the remnants of four dentine bands (Figs 17A, 21). These bands have resorbed posterior edges, and are ingrown to various extents by columns of dark, apparently hypermineralised, dentine. The dark bands appear as slightly darker bands within the white dentine. These bands of tissue narrow anteriorly for about 4mm, where they become DEVONIAN TOOTH-PLATED DIPNOI FROM W.A. 419 indistinguishable. Anterior to them, the surface of the plate is fonned by bony tissue along the median edge of which there are few nan'ow elon- gate blisters of dentine which butt up against the mesial side of the tw'o most anterior teeth (Fig. 17D). These blisters become larger anteriorly. On the posterior margin of both plates three bands of dentine can be distinguished, the inner one strongly ingrown with columns of hypermin- eralised dentine; the outer one shows an irregular edge, indicating deposition not long before death. This outer band also contains a small number of pores as in C australis (Smith & Campbell, 1987, pi. 4, figs 24-26). The lateral margins of the plate (Figs 173,21) have a single narrow band of den- tine that continues forward to the fifth or sixth row of teeth where it turns inwards forming a ridge in each furrow proximal to the most distal tooth in each row. This most distal tooth on each row lies directly on bone (Fig. 1 73, C), and the concentric layer of enamel-covered tissue abutts the inner edge of the distal teeth, indicating that it did not form a continuous layer around the edge of the plate. Preparation of the paratype has allowed the dentine to fall free of the pterygoid (Figs 1 8A,3, 1 9A,3, 20A-C). A basal pulp cavity is continuous longitudinally beneath the tooth-bearing rows, except at the distal ends where there is bone beneath the outer edge of the last-formed tooth. Bone lies at the distal ends of the teeth and ex- tends slightly beyond it (Fig. 17B,C). Pterygoid bone joins the tooth plate between the tooth rows in thin bands. This bone Joins bone and coarse dentine that lies between the rows of teeth on the occlusal surface (Fig. 173). More detail of this arrangement is described below in the histology section. Along the lateral face of the plate the bone of the pterygoid also makes a continuous Junction with the dentine (Fig. 19A). Although the broken specimen is not completely clear, bone also meets the coarse dentine medial to the inner- most tooth row. The breadth of this bone in- creases posteriorly. Beneath the thick solid dentine part of the plate fomiing the heel, the dentine is braced against the pterygoid by irregu- lar discontinuous masses of bone, which become more closely spaced posteriorly where the den- tine thins out (Figs 18-20). The mandibular tooth-plates are convex, and contain six rows of teeth (Figs 43, 223, C); in each row, the last five or six teeth can be distin- guished; individual teeth have the same form as those on the palate. The plate is convex, and the most distal tooth lies directly on bone (Figs 103, 223, C). The left side of ANU49213 shows the growth series of tooth structure (Fig. 223). Along the left side of the plate is a continuous layer of enamel-edged tissue which runs forward to, and around, the last formed tooth in the sixth row (Fig. 223). At that point it turns inwards to form a boundaiy/ edge proximal to the distal tooth in the fifth row. It then continues across the plate inside the distal tooth in each row until the median row in which there are two teeth distal to it (Fig. 223). The bone on which the distal tooth sits is coarse at the base and fines up to the tooth base. As can be seen from the surface, the bone extends in- wards between the rows, and becomes progres- sively more ingrown by dark dentine proximally. Laterally the same type of tissue extends proxi- mally between the sixth row of teeth and the marginal ridge. A broken mandible (Fig. 22A), shows well the layers of reparative dentine along the lingual sur- face. These are in short sections in about four layers. Those at the dorsal edge are slightly eroded, and at the ventral edge are fresh, being the latest ones formed. As can be distinguished in this photograph, the bones lie below the dentine and form a slightly concave surface in the lingual furrow. HISTOLOGY OF ADULT TOOTH-PLATES DISTAL TOOTH. Enamel and Pallial Dentine. The tooth plates add considerably to our under- standing of the Late Devonian evolution of the marginal growth mechanisms, the pulp cavities and pulp canals, and the distribution of the vari- ous types of dentine. Vertical sections along the tooth rows (Fig. 23A,C) show a continuous basal pulp cavity which extends to the inner edge of the most distal tooth. At that point, the bone runs onto the outer edge of the second last tooth, making a firm connection between the bone and the outer edge of that tooth (Figs 23A,C, 24A,B, 25A,B). The distal tooth is covered with enamel except on the tritural surface where it has been worn away. The enamel layer extends over the outer surface of the distal tooth terminating deeper than the basal pulp cavity (Figs 24A,B, 25 A, 3, 26D). Beneath the enamel is a layer of dentine of vari- able thickness, dark in colour on SEMs, but light in transmitted light (Figs 23A-D, 24A-D, 25 A, 3, 26A-C,E,F). This is the pallial dentine layer, or mantle dentine of Weidenreich (1925) and Lund et al. (1992), which is found in all Palaeozoic dipnoans. For our usage of pallial dentine, see Campbell & Smith (1987: 166). Some ofthe pulp canals in the pallial dentine are continuous with 420 MEMOIRS OF THE QUEENSLAND MUSEUM FIG. \1 . Adololopas moyasmithae gen. etsp. nov. Pterygoid tooth-plates of the holotype. A, view of the posterior part of the palatal plates; note in particular the primitive dentine around the posterior edges of the plates and the successively generated layers of dentine along the median-posterior edges of the plates (arrowed). B, oblique view ofthc plates. C, anterior end of the same plates showing the teeth lying directly on bone and having enamel forming only around the teeth themselves; note also the down-turned part of the plate to the mandibular articulation, D, medial view of the left tooth plate which has been separated from the other plate; showing the articulation of tlie parasphenoid and the two small blisters of primitive reparative dentine towards the front of the plate. Scales = 10mm. DEVONIAN TOOTH-PLATED DIPNOI FROM W.A. 421 FIG. 1 8. Adotolopas moyasmithae gen. et sp. nov. An isolated left pterygoidal plate of WAM93I.I.221. A, The dorsal surface of dentine part of the plate showing the bases of the pulp canals. B, bone from which the dentine has been separated; A and B will fit together; B shows the positions of the pulp canals adjacent to the bone, and the white patches are the remains of the bone joining the pter>'goid to the dentine plate. Note the abundance of bone attachment under the heel of the plate. Scales = 1 0mm. those of the core dentine of the tooth, but others are directly connected with the underlying tissue on the inside of the tooth. We refer to this as pedestal dentine, and it will be discussed below (Figs 24A-D, 25A-B, 26C-D). At their tips, the pulp canals in the pallial dentine branch to form a thicket of dentine tubules which are not always obvious (Fig. 26F). Under crossed polarised light, this layer has a different pattern (Figs 24B,D, 25B, 26A) from the remainder of the dentine. The Core Dentine. In thin section, the core of the distal tooth is made of dark brownish tissue which forms a dense mass. It appears in weathered sur- faces as a dark material, and it is hard wearing; so to some extent it must be hypermineralised. In it the pulp canals are slightly bent, but they form continuous structures from the basal pulp cavity to the base of the pallial dentine (Figs 24A-D, 25A,B, 26A,B). The hypermineralised dentine was formed directly from the basal pulp cavity. Occasional branching of the pulp canals is ob- served, and interconnecting canals are common. Extending out at right angles to the canals are abundant dentine tubules (Figs 24A, 25A, 26D) which are more obvious nearer the pulp cavity, although more occlusally in the tooth they are occasionally visible, where they are narrower. FIG. 19. Adololopas moyasmithae gen. et sp. nov. Drawing of Fig. 18, showing the points of contact between the dentine and the pterygoid bone in black. Note that the two surfaces do not fit together because when the two were separated, the bony contacts sepa- rated part to one side and part to the other. Fine stipple refers to surface without attachment between the bone and dentine, and coarse stipple to parts of pterygoid not joined to dentine. Tubules are well shown on some SEMs where they are delicate structures which branch repeat- edly away from the canal and form a meshwork (Fig. 28B). As is shown in Fig. 27B & C, the crystal fibre bundles composing the dentine make a basket weave throughout the tissue (Smith, 1984), such as occurs in petrodentine. They are similar to the structure described by Campbell & Smith in the teeth of Holodipterus gogoensis (Campbell & Smith, 1987). We do not have enough material to cut horizontal sections through each tooth to determine the birefringence of the crystal fibre bundles. The abundance of tubules tliroughout the tissue indicates that it is not typical petrodentine, though it may be an intermediate stage in the evolution of petrodentine. In fact, it is a compact structure for which we can find no acceptable term among the many available ones. We will therefore refer to it as ‘a compact dentine’ in generic terms, not in- tending that this be taken as a new term for dentine morphology. Pedestal Dentine. Between the pallial dentine and the compact dentine of the distal tooth, is a layer of dentine which has pulp canals which are very 422 MEMOIRS OF THE QUEENSLAND MUSEUM FIG. 20. Adololopas moyasmithae gen. et sp. nov. A, occlusal view of pterygoidal tootn-platc. B. dorsal view of tooth-plate, but reversed so that it is in the same orientation as the occlusal view, to show the alignment of the bon> patches of the underside of the tooth plate as they correspond to the gaps between the surface ridges. C, bone of the pterygoid to show the surfaces to which the dentine plate was attached. Arrows indicate points of the superimposition of dentine on bone. Note that B is a mirror image of the underside of A, and it represents the joining suface with C. Arrows indicate the points of superimposition. distinct in some sections and weaker in others. This dentine is well shown in Figs 24A,C, 25A,B, 26A-D. It has the same colour as the underlying bone. It is always better developed on the inner edge of the distal tooth than the outer. The pulp canals in this dentine do not connect with the basal pulp cavity at their bases, but to- wards the occlusal edge of the tooth they are connected into the pallial dentine (Figs 24A, 25A, 26B-D), At the base they seem to grade into bone, which is distinguished by osteocyte spaces. Dentine tubules, com- ing off pulp canals, are not ob- vious in this dentine, but occasionally they can be de- tected. In Figs 24A,C, 26A-C this dentine is in contact with the core dentine of the pre- vious tooth, and cuts into the dentine pulp canals of that tooth. As seen in Figs 23 A, C, 26A-C, this dentine is lost, presumably by resorption, and it is not visible in any preced- ing tooth. Some pulp canals extend into the tissue between the dis- tal tooth and the proceeding tooth (Figs 23A,B, 24A-D, 26A-C). As can be seen in Figs 24A-D, 26C, which are optical photographs, and Figs 23A-B, which represents the SEM re- production of the same fea- ture, this layer of the most distal tooth lies occlusal to the enamel and pallial dentine of the second last tooth. The enamel (Fig. 23B, 24A-D) is left embedded in the matrix and it surmounts the bone on which it lies. The sharp bound- ary between this layer, and the underlying compact dentine which forms the core of the tooth, is well shown on Figs 23B, 26C. The SEM shows how the compact dentine does DEVONIAN TOOTH-PLATED DIPNOI FROM W.A. 423 hypermineralised core bone and dentine enamel reparative dentine layers attachment for pterygo-mandibular heel ligament dentine FIG. 2\ . Adololopas moyasmithae gen. et sp. nov. Oblique view of pterygoid plates of the holotype showing up the detail of bone, dentine and reparative dentine. Compare Figure I7B. Note particularly the presence of white tissue representing the enamel-covered surface around the teeth and forming a layer on top of bone between the distal teeth. not etch as easily as the outer dentine layers. The naming of this dentine is important. It is not interdenteonal in origin, it has poorly developed tubules, and occlusally it merges with pallial den- tine, but at the opposite end it grades into bone with no distinct boundaries. The function of this dentine is to provide an outline on which the distal tooth was deposited. We consider that it should be called ‘pedestal dentine’ (Smith, 1985: 629). Pedestal dentine is a primary dentine joining pal- lial dentine to the bone of attachment. MORE PROXIMAL TEETH. The second and third teeth from the outer edge of the plate have most characters in common with the last tooth. The second tooth overlaps the third tooth directly; most of this contact takes place between the com- pact dentine of the teeth, but towards the occlusal surface the pallial dentine of the two teeth are in contact (Fig. 23A,C). No sign of the pedestal dentine as described in the distal teeth can be seen in these older teeth. Presumably it was there origi- nally, but it has been resorbed to allow other dentine to develop in its place. The initiation of this resorption is visible in the pulp cavity (Fig. 23A,C). The compact dentines of the second and third teeth are contiguous, and in some places the only way of distinguishing them is the fact that the canals diverge slightly at the Junction. The pulp canals tend to be long and relatively straight, but they are interconnected irregularly (Fig. 24C-E). The fourth and more proximal teeth become less prominent as a result of wear (Figs 1 7B, 2 1 ). They are composed of continuous core (compact) dentine, but with a relict of pallial dentine in some places. Against the basal pulp cavity, it is possible to distinguish the base of the dentine columns in the core of the teeth (Fig. 24E-G). These ends are rounded in vertical section, an indication that there was continuous deposition around them by a layer of cells. The dentine formed in these positions has the same internal structure and col- ouring as the dentine higher up in the tooth; the only real difference is that the dentine has less obvious tubules away from the basal pulp cavity. 424 MEMOIRS OF THE QUEENSLAND MUSEUM - bands of reparative dentine bone in lingual furrow FIG. 22. Adololopas moyasmithae gen. et sp. nov. A, lingual view of the mandible of the holotype, with the right side of mandible removed; note the bone in the lingual furrow and the bands of reparative dentine along the basal edge of the dentine of the plate, with solid dentine dorsal to it. B, dorsal view of same with the two parts together; the specimen is tilted with the anterior upwards to show the anterior teeth against the bone. C, lateral view of WAM931.1.221 to show the curvature of the plate. Scales = 10mm. Dentine tubules are abundant close to the basal pulp cavity (Fig. 24E-G). Although the pulp ca- nals thin out towards the occlusal surface, we have been unable to distinguish any clear evi- dence of intratubular or peritubular dentine (Baume, 1980). There could be two possible causes — the thickness of the sections studied which cover up the boundaries of such layers, or to the fact that no such deposits are present, and the the pulp canals become wider towards the basal pulp cavity. We have not had sufficient material to cut hori- zontal sections of a tooth to observe the pattern of extinction under crossed polars such as occur in petrodentine. SEMs also show the fibre bundles in this dentine clearly, as on the right and left sides of Fig. 27B. The following features are also in- consistent with the dentine being petrodentine (Smith, 1 984); dentine tubules are present widely through the tissue — these are visible in both thin sections and in SEMs; the whole dentine is stained brown, whereas petrodentine is clear; and, the dentine is made up of tissue laid down at the base of each column (Fig. 23E-G), and is devoid of tissue that could be interpreted as having been laid down from a special layer of petroblasts. An isolated tooth-plate of WAM93 1.1.221 (Fig. 1 8), from which the dentine has been sepa- rated from the pterygoid bone, has been cut in an oblique transverse section about two teeth in from the margin. The section has been subjected to SEM examination (Figs 27-29), and the adjacent DEVONIAN TOOTH-PLATED DIPNOI FROM W.A. 425 compact dentine pallia! dentine compact dentine compact dentine dentine compact dentine bone bone showing resorption bone pallia! dentine basal pulp cavity A FIG. 23. Adololopas moyasmithae gen. et sp. nov. A, C, SEMs of two sections cut from a tooth row from WAM93 1 . 1 .22 1 ; note the early initiation of compact dentine in the last formed tooth surrounded by a mass of dentine supported at its base by bone and dentine which extend up from the basal pulp cavity. This outer layer is lost on much ofthe second teeth. Note also the large open space of the basal pulp cavity below the tooth row, and the structure of the supporting bone. B. enlargement of the junction ofthe last two teeth on A, as outlined on A. The enamel which extends dowm into the tissue is labelled; the junction between the two layers of compact dentine and the surrounding tissue is obvious. D, enlargement of the distal tooth of C; on the right side note how the pedestal dentine extends up around the compact dentine. Scales A, C = I mm; B = 0.1mm. D = 0.5mm. 426 MEMOIRS OF THE QUEENSLAND MUSEUM FIG. 24. Adololopas moyasmithae gen. et sp. nov. A, B, optical thin sections of the most distal tooth shown in an adjacent section in Fig. 23B; A is in plane polarised light and B in crossed polarised light. This shows the bone and the pedestal dentine extending up into the base of the tooth. C, D. the second tooth along the tooth row; C is in plane polarised light and D in crossed polarised light. Note the absence of pedestal dentine built into the tooth on the right side. E, an even more proximal tooth showing the basal pulp cavity at the base, and the long regular pulp canals which divide more extensively at the base than at the top of the tooth. F, G, two sections from the base of teeth to show the pulp canals running from the basal pulp canal, and also a number of dentine tubules extending into the tissue. Scales A-E = 0.25mm; F-G = 0.3mm. DEVONIAN TOOTH-PLATED DIPNOI FROM W.A. 427 enamel X pallial dentine pulp cavity enamel FIG. 25. Adololopas moyasmithae gen. et sp. nov. Optical photograph of the distal tooth of a second cut through WAM93 1 . 1 .221 . A. under single polarised light; B. under crossed polarised light. This specimen shows in better detail the bone and pedestal dentine as in Fig. 23. Scales = 0.25mm. parallel section has been cut for work on the optical microscope. These sections show much detail of the tissue in the valleys between the tooth rows and its relationship to the basal pulp cavity. As indicated above, the valleys between the tooth rows are filled with what appears to be coarse dentine in natural light. Across this whole section (Figs 27 A, 28A, 29), some of this inter-tooth-row dentine lies at the occlusal surface of the plate. The contact with the adjacent tissue of a tooth (compact dentine) is sharp, and there is no evi- dence of intergrowth of the intervening pallial dentine by pulp canals turned laterally from the pedestal dentine. The dentine in these valleys is formed of bands of tissue around pulp canals (Figs 27B,C, 28A, 29), and these bands are made of tissue like the compact dentine cores of the teeth. Many of the pulp canals are turned parallel with the occlusal surface, but deeper in the tissue the canaisjoin and form a more continuous layer. The boundary between the filling tissue and the underlying core tissue is shown in Figs 28A, 29. The space between the last and second last tooth on the right in Fig. 27A is shown in detail in Fig. 27B-C. It shows the fine structure more clearly than elsewhere on the section because it is the most recently formed unit between the tooth rows, and it lies over a patch where the tooth-plate surface is supported to the pterygoid bone as described above. This SEM shows no fine struc- ture betw'een the masses of dentine in the inter- tooth-rows, but some fine tissue must have been present in the original. Such material appears on the untreated specimen as whitish substance lying betw'een the darker dentine columns (Fig. I7B). Thin sections of this material (Fig. 30) show that it is composed of bone which does not show up well on SEMs. Figure 27B,C are also important because at the base of the column the remnants of pterygoid bone supporting the dentine, are seen. This bone represents the hard columns of bone which appear in Figs 18-19 between the open spaces below the radial tooth rows. Fig. 27 B,C show's a slight enlargement in which the dentine in the canals meet the bone, and shows the way in which the bone is surrounded by dentine growing over its surface away from the pulp cavity, in part, this provides an understanding of why the canals in the inter-tooth rows are oblique to the rows in the teeth, and it also shows the similarity in struc- 428 MEMOIRS OF THE QUEENSLAND MUSEUM DEVONIAN TOOTH-PLATED DIPNOI FROM W.A. 429 ture between the dentine in the tooth cores and the inter-tooth-row spaces. It also demonstrates why the boundary between the inter-tooth dentine makes a sharp boundary with the tooth dentine. Figure 28A shows the third tooth from the right in Fig. 27A, together with the tissue occupying the inter-tooth-row spaces. Because of the obliq- uity of the section in Fig. 27A, Fig. 28A is an older tooth than the one shown on the right side of Fig. 27A. Note the hard white compact dentine in the core of the tooth and the darkish layer on each side. An enlargement of the inter-tooth-row (Fig. 29) shows the sharp contact between the tooth dentine and the darker tissue between the tooth rows. The dentine in the inter-tooth valleys con- sists of largely globular masses separated by light coloured substances, as mentioned above. As can be seen from the growing edge of a plate, bone occupies the spaces between the tooth rows (Figs 17B, 21). With increasing age, dentine is depos- ited from pulp canals which grow up into the bone, with bone lying between dentine. Finally, the heel of the dentine plate is formed by the exclusion of bone, so that the heel is made entirely of compact dentine. Figure 30 shows stages of the development of this dentine. Bone, with the osteocyte spaces clearly preserved, is shown surrounded by den- tine and its pulp canals. The newly formed den- tine beginning to occupy the spaces between the bone is shown on Fig 27B, C. The dentine which forms in the valleys between the tooth rows is of the same type as the hypermineralised dentine (compact dentine) that forms the cores of the teeth. This new genus shows an intermediate stage in the development of a continuous pulp cavity be- neath the dentine, as is shown in Mesozoic and later species. The dental plate in Adololopas is still firmly attached to the pterygoid or the prear- ticular at numerous points (Figs 1 8, 1 9, 20). It still retains some bone and intergrowing dentine in the valleys between the tooth rows. Evolution in later lungfish reduced the bone, and the dentine of adjacent tooth rows Joined up laterally to make a continuous layer of dentine across the plate. Of course, associated with this change are the changes in dentine structure which have been described by many authors such as Denison (1974) and Smith (1984). REMARKS. The arrangement of the bones mak- ing up the posterior part of the skull roof is problematical, but enough is known to make the distinctiveness of the roof apparent. The postero- lateral edge of I carries an articulating groove indicating either that Z fitted around the comer of that bone, or a loose Y 2 was present. The shape of the operculum with its intumed dorsal edge is of great value in determining the presence of bones on the posterior part of the skull. Although Y 2 and Z were not preserved, it is possible to show that they were present originally because the opercu- lum fits neatly against the bone 8 and turns in- wards against the Y 2 and Z. On this reconstruction it becomes apparent that the genus was com- pletely different from Jarvikia and Gnathorhiza. A further complication is that the canal running forwards from I apparently enters a large bone which is situated too far posteriorly to be an X, and can only be a modified Yi. Judging from the embayments in the flanks of J, and from the position of the canal exiting anterolaterally from J into what normally is K, X must have had a relatively small contact with J, but the shape of its contact with 3 shows that it expanded laterally; K had a large contact with J (Fig. 1), and its anterior outline remains unknown. The position of K rela- tive to X and L is a matter of importance. In Dipterus Valenciennes K is a small bone be- tween X and L; but with the increased size of K and the anterior position of X in Adololopas^ L is further forward also. Unfortunately we do not have an X or a K to check if a lateral line canal Joins the two bones. Too few specimens are known to permit a discussion of the variability of the species except for one point- the ornament on the pectoral girdle. On WAM931 .1 .221 the anterior edge carries closely spaced pustules, but these are not present on ANU49213. Other features of both cleithrum and clavicle are closely comparable. Comparison should be made with the tooth plates of Scaumenacia curta Whiteaves, the best FIG. 26. Adololopas moyasmithae gen. et sp. nov. A, B, distal tooth as illustrated in Fig. 24A, B, emphasising details of dentine structure; A, is in crossed polarised light. C, an enlargement of B to show details of the tissue containing pulp canals but lying outside the enamel of both the teeth. Note also the resorption of the second last tooth, now occupied by pedestal dentine of the distal tooth. D, enlargement of the edge of the distal tooth shown in Fig. 25A; note the thick enamel layer, the pedestal dentine and the bone beneath it. E, contact between the distal tooth and the proceeding one, showing pallial dentine in both, but with the dentine tubules difficult to observe. F, contact between the second and third last teeth, showing the dentine tubules in the pallial dentine and the abundant tubules in the adjacent compact dentine. Scales A, B = 1mm; B = 0.25mm; D-F = 0.5mm. 430 MEMOIRS OF THE QUEENSLAND MUSEUM FIG. 27. Adololopas moyasmithae gen. et sp. nov. A, SEM of transverse section across the anterior part of WAM93 1 . 1.221 showing the teeth and the dentine filling the inter-tooth-row spaces. B, section through the last inter-tooth-row space outlined on the right of A showing the texture of the dentine in the teeth and the orientation of the pulp canals. A small amount of bone lies at the base of the dentine. C. an enlargement ot the space outlined on the basal part of B showing the dentine-bone contact, and the formation ot the new dentine around the bone spaces. Scale A= 10mm. DEVONIAN TOOTH-PLATED DIPNOI FROM W.A. 431 FIG. 28. Adololopas moyasmithae gen. et sp. nov, A, third tooth from the right in Fig. 27A and the infillings of the inter-tooth-row valleys. B. dentine close to the base of a pulp canal adjacent to the basal pulp cavity; note the branching nature of the dentine tubules. known of the Frasnian genera from elsewhere in the world. These have been described by Denison (1974), Smith et al. (1987), and Cloutier (1996). Denison thought that the plates of S. curta had grown in the same way as those of D. valencien- nesi, but their histology was entirely different. Unlike D. valenciennesi\ the specimen of S. curia sectioned by Denison (refigured here as Fig. 32E, F) had well-differentiated dentine and bone, sepa- rated by small resorption pulp cavities at the base of at least the most labial tooth; and the dentine canals, though branched, were much more regular than those of D, valenciennesi. Denison claimed to have observed a small patch of pleromic den- tine in the most proximal of the teeth in the section figured, but not in any of the others. We have examined the specimen and have been able to FIG. 29. Adololopas moyasmithae gen. et sp. nov. Enlargement of the space to the left of the third tooth from the right in Fig. 27A showing the contact (ar- rowed) between the compact dentine of the tooth core and that of the space between the teeth. Note the black gaps between the dentine that seem to have no tissue in them. I'hey are filled with thin bone which does not show up on SEMs, but is well shown in optical thin sections (Fig. 30). Scale = 0.5mm. confirm most of the points of Denison’s descrip- tion, although the mounting medium has crystal- lised, destroying the clarity of much of the slide. We have found a small patch of what Denison refers to as pleromic dentine, but we do not con- sider that it has any importance. Smith et al. (1987) have given a more complete account of S. curia. and have added more histological information. Cloutier (1996) has figured palatal plates of S'. curta which show clearly the suture between the two plates, the elongated lingual row on each plate, and the extension of the tooth rows to the heel of the tooth. In addition, his specimen is recorded as having 'small tubercles of enamel infrequently added posteriorly and posterome- 432 MEMOIRS OF THE QUEENSLAND MUSEUM FIG. 30. Adololopas moyasmithae gen. et sp. nov. Optica! sections cut parallel with the section in Fig. 21 A. A, B, a section in which the amount of dentine is only partly developed; C, D, a section in which the bone is more extensive. B and D are enlarged areas of A and C respectively. In A and C the pulp canals are surrounded by compact dentine, the dentine of adjacent canals being in contact. In D the pulp canals are not so common, and places where bone is surrounded by dentine to form a core, are well shown. Osteocyte spaces are present in the bone. Scale = 1mm. dially to repair the heel of the tooth plate’, and he comments that there is a thin layer of enamel covering most of the surface of the tooth plate. The first point indicates that the primitive pattern of denticles is maintained in this genus. The sec- ond point is difficult to accept because enamel is laid down as a tissue on the outer surface of a tooth, and should not be present in the intervening furrows between the tooth rows, unless the adja- cent tooth rows converge. Such does not seem to be the case. We also note that Cloutier’s fig. 6, seems to show a broad band of exposed bone mesial to the lingual row of teeth, but Smith et aL (1987, p. 315) comment that there are simple sheets of dentine in that region. In addition, they note that there is no bounding layer of enamel around the outer margins ofthe teeth. It seems that in these respects S. curta is closer to Adololopas than to later genera. We have been able to cut thin sections of two small topotypic tooth-plates of Scaumenacia curta given to KSWC by Denison. Both oblique transverse and vertical sections are available, one of the vertical sections being of the most labial tooth in a row. All sections show the usual enamel layer over a layer of pallial dentine. Low down on the tooth, ventral to the pallial dentine, is a coarse open tissue, as in Adololopas. This tissue is con- tinued downwards into the bone, and its detail is like that of the bone, (though without osteocyte spaces). We therefore refer to it as pedestal den- tine as in Adololopas. It is heavily stained with iron compounds, and the internal structure in some places is difficult to determine. On the other hand, it is possible to recognise dentine canals extending into it from the more medial part ofthe tooth (Fig. 32A, B). The core of the tooth is composed of hypermineralised dentine (referred to by Denison (1974) as trabecular dentine, and by Smith et al.,1987, as petrodentine) that has minimal iron-staining, and shows pulp canals like those figured by Denison (1974, fig. 5), i.e., they are essentially vertical but divide and rejoin to form a coarse reticulum. It is possible to recognise a few small dentine tubules extending from the dentine canals into this tissue, but the tubules are DEVONIAN TOOTH-PLATED DIPNOI FROM W.A. 433 FIG. 31. Lateral view of a head of Chirodipterus australis Miles. Note the shape of the lip outline which has been dotted in where bone is missing. Scale = 10mm. irregular in their distribution and they are much shorter and narrower than those in ordinary den- tine. In cross-polarised light, the cross pattern re- ported by Smi^ in petrodentine has been observed. In a section along a tooth row (Fig. 32E,F) the column of petrodentine extends almost up to the crest of the tooth, demonstrating that its deposi- tion began shortly after the trabecular dentine was formed in that region. In teeth that have com- pleted growing, the petrodentine lies directly on the basal bone. In more recently formed teeth there is a basal pulp cavity which is small and gives a space into which the petrodentine grows. This basal pulp cavity is small in the most recently added tooth in Denison’s specimen (Denison 1974, fig. 5A,B). In the ANU specimen it is large and has a clear opening into which growth may occur. In more proximal teeth, no basal pulp cavity is present. Presumably it has been filled by downwardly growing petrodentine. This is unlike the situation in Adololopas, in which the basal pulp cavities remain open under a number of teeth and into the more solid proximal parts of the plates. DISCUSSION OF LONG HEADEDNESS IN DIPNOANS Schultze (1992) pointed out that some De- vonian dipnoans are long-headed because of the length of the snouts and others because of the length of the posterior parts of the heads. This analysis makes many useful points, but it has some difficulties which we will discuss here be- cause Adololopas has a long cheek and should fall into the second of Schultze’s catego- ries. In our view the length of the head is related to a number of functional factors including gathering food, reduction of food, and the position of the gill chambers. The first problem to us is one of misinterpretation of data; e.g,, the reconstruction of the snout of Uranolophus wyom- ingensis (Schultze, 1992, fig. 5C). In our view, he shows the eye placed too far forward. This position is not explained in his text, but quite clearly the elongate bone lateral to Li on Schultze’s fig. 5C should be a bone 3, and the eye should be placed ventral to it. His value of 1^^ for this species is incor- rect. What is more, the posterior part of the head is more elongated than it should be in relation to the snout with which it is figured, a point that is illustrated by a comparison of the figures of Denison ( 1 968, fig. 3A) and Campbell & Barwick (1988b, fig. 2C). With the eye put into its correct position, this genus shows a greater comparison in skull outline to the adjacent figure of Soeder- berghia groenlandica in Schultze’s paper. Schultze’s (1992) discussion of Holodipterus is also erroneous in a number of respects. Details of the morphology of that genus are set out in Pridmore et al. (1994). For example, the palate has a denticulated surface with denticles on the parasphenoid (Pridmore et al., 1994, figs 8, 47, 49). The denticles are subject to renewal at several periods during the life history of an animal, and individual denticles are covered with intervening ‘primitive’ dentine during some phases of re- newal (Schultze, 1992: 42). This pattern is totally unlike that of Speonesydrion (Campbell & Bar- wick, 1983; 1984), a comparison suggested by Schultze. The marginal ridges are present, and are more obvious on some species, especially at some stages of resorption (Pridmore et al., 1994, figs 4, 29, 73, 96), but in other specimens they are ex- panded depending on how much resorption has taken place. Contrary to the statement by Schultze (1992: 42), the preorbital cheek has the same cutaway ventral edge as in Griphognathus, as was well shown by Campbell & Smith (1987, fig. 8). Schultze (1992) suggested that it is like the sur- 434 MEMOIRS OF THE QUEENSLAND MUSEUM FIG. 32. Scaumenacia carta Whiteave^. A-D, optical sections of ANU49092. A, vertical section of the most distal tooth showing the large basal pulp cavity surmounted by dentine, surrounded by dark tissue into which the pulp canals extend. B, enlargement of same, showing the absence ot dentine tubules and the cross-textured mass of dentine. C, part section of tooth rows cut from the same specimen; note that close to the tip ot the tooth the hard central dentine is present. D, enlargement ot the bottom Icfl tooth in C; note the pulp centres within the dentine. E. F. two views ot the section figured by Denison (1974, fig. 5) . The mounting medium has been rccrystalHsed making the detlnition worse than in the original figures. E is under single polansctnighl. and h is crossed polarised light. Note the absence of basal pulp cavity beneath any tooth. Scales A-(_ — Inim, u — 0.5mm; E, F = 1mm. DEVONIAN TOOTH-PLATED DIPNOI FROM W.A. 435 face of Chirodipterus, but in fact there are no similarities; it is difficult to see how such a con- clusion could be made. In comparing Fig. 33 in this paper with fig. 8 in Campbell & Smith (1987), we note that the ventral edge of the cheek in Holodipterus has a considerable gap between the edge of the cheek and the mandible. In addition, the marginal ridges on the palate are nothing like the ridges of Chirodipterus or Speonesydrion (Campbell 8c Smith, 1987; Campbell & Barwick, 1984) as suggested by Schultze (1992). Nowhere does Schultze’s work comment on the method of making marginal palatal ridges by resorption of the surface on the inner parts of the plate, or of the production of bosses on the palate by failure to resorb parts of denticulated surface during suc- cessive episodes. Although when Schultze wrote there was no evidence of basihyal-basibranchial bones in Holodipterus, but these are now avail- able Pridmore et al. (1994, figs 29, 32, 57). In view of Schultze’s discussion, we must emphasise that the presence of denticles in the mouth does not indicate a primitive dentition. Denticles are present in a variety of different genera, and we are in no doubt that they have evolved several times to meet different require- ments. We chose to use the term ‘denticulated’ to characterise one group of dipnoans, because that seemed to be an obvious feature of one group. This may have been an unfortunate choice of a term, but it does not permit a misinterpretation of the remainder of the discussion. Schultze also noted that Holodipterus was a food crusher in its feeding, whereas the denticu- lated group is not so. We point out that H,(H.) longi (Pridmore et ah, 1994) has a mandible like that of Griphognathus, its median area being re- duced in length. This mandible also has an adduc- tor pit that is similar to that in other species of Holodipterus, and its teeth have been almost com- pletely removed by resorption. In other words this species is different from the basic outline of H. gogoensis. The question we now have to ask ourselves is: does the presence of the teeth mean that the group is a tooth-plated form; or does the fact that no true tooth-plates are developed in Holodipterus, mean that the teeth were developed secondarily in a genus that was initially a ‘den- ticulated’ fonn? In connection with this argument we point out that in Holodipterus the teeth are different in histology from the teeth of other tooth-plated species (Campbell & Smith, 1987), and that they are resorbed periodically from the mesial part of the mouth, the evacuated surfaces becoming occupied with denticles. The latter view will be rejected by cladists because it is less parsimonious than the fonuer since it requires two developments of teeth within the dipnoans. However, such a parsimonious view is irrelevant, in that it obscures relationships based on biologi- cal similarities. We acknowledge that such a ‘par- simonious’ view will be adopted by some workers, but point out that parsimony applied in such circumstances is unacceptable. In this type of enquir>', parsimony may be used to choose between two hypotheses which have equal merit in explaining a set of observations. It should not be used to generate the hypotheses. In addition, we draw attention to the quote in the paper by Schultze & Marshall (1993): ‘And if you get only one tree, but you cannot explain it biologically - start all over again’. Vranolophus is said by some workers to be different from other denticulated genera in the form of the parasphenoid. Denison in his original paper indicated that it was long and thin, stretch- ing forwards almost to the anterior end of the pterygoids. Campbell & Barwick (1988b) at- tempted to show that the plate was short, rather like the parasphenoids in such other groups as Griphognaihus, and that the boundaries of the parasphenoid outlined by Denison were cracks in the palate. Schultze (1992) indicated that the in- terpretation of Denison was correct, reproducing a drawing in his fig. 1 and photographs in fig. 3 to support this view. Even so, he notes that ‘such a parasphenoid is unique within the dipnoans'. We wish once again to dispute his interpretation of the outline on 6 arguments. 1 ) So far as we are aware, the parasphenoid is a bilaterally symmet- rical median bone. Even in those genera in which the dorsal edge of the parasphenoid is expanded, the shape of both dorsal and ventral edges should be symmetrical relative to the midline. This is not so in the inferred outline of Denison. 2) If the shape is as shown by Schultze, 1 992, fig. 1 B, with the posterior outline skewed, this does not match the position of the dorsal edge as shown by Camp- bell & Barwick (1988b, fig. I E). 3) The pituitary is known to have a double origin — part of it is a ventral growth from the braincase, and the other from the pharynx. In IJ. wyomingensis the pitui- tary fossa (Campbell & Barwick, 1988b, fig. IE), is situated about one third of the length of the palate from the posterior end, which is its position in other Devonian dipnoans. What is more, the palate is thin, and the hypophysial stalk would have to be situated in approximately the same position. The organisation is very similar to that of Griphognathus whitei, in which the parasphe- 436 MEMOIRS OF THE QUEENSLAND MUSEUM noid is of normal Devonian type. 4) We simply controvert the view that other specimens of U. wyomingemis in which the palates are incomplete show that the details indicate an elongate paras- phenoid. 5) The palate is not thick, but it has thickened edges in certain places as is common in Griphognathus. Such thickenings are correlated with its peculiar mode of feeding (Campbell & Barwick, 1 988b, fig. 7A). This is the specimen on which Schultze (1992, fig. IB) is based, and in which the anterior end of the parasphenoid fol- lows the outline of the depression. 6. Comparison with outgroup species such as Diabolepis does nothing to add to the argument. What such argu- ment shows is that attention to comparative stud- ies with genera of doubtful affinity causes observers to view what they think are appropriate structures, rather than what is to be seen. It biases basic morphological work towards bad observa- tion. In summary we see no reason apart from cracks on the palate to assume that the parasphe- noid is long and narrow as suggested by Denison (1968) and Schultze (1992). We were mistaken in stating that the ‘denticu- lated’ group had no adsymphysial plate in the lower jaw of Griphognathus (Campbell & Bar- wick, 1 984, p. 1 63). Such a plate is not clear in G. whitei, but it is present in Griphognathus sculpta. This item will have to be removed from the fea- tures used to outline the ‘denticulated’ group (Campbell & Barwick, 1990). The group of Griphognathus, Rhynchodip- terus, Soederberghia and Fleurantia is not mono- phyletic; Schultze (1992) recognised this. lowadipterus cannot be related to Holodipterus as a long-headed form in which the the mandible is short. Compare the mandible of H.(H.) longi (Pridmore et al.,1994, fig. 33) with that of other specimens of Holodipterus figured elsewhere in the same paper. Holodipterus is a member of the so-called ‘denticulated’ group, which has become shorter headed than other members of that group as a result of the development of marginal teeth which are progressively resorbed. It is not possi- ble from available material of lowadipterus to indicate what its relationships are. Further evi- dence of the structure of its palate is necessary. Schultze (1992) was of the opinion that the longer posterior part of the skull permitted a larger mass of adductor mandibulae musculature to fit under the skull roof; this is certainly correct for lowadipterus which is also characterised by a very forward position of the eye. Comparison of Schultze’s (1992) figures of lowadipterus and Holodipterus (Fig. 3) show that the forward posi- tion of the eye relative to the median bones of the skull roof B and C, is not the only factor involved in providing a larger space for muscles, because the operculars of the two genera lie in a com- pletely different position relative to the Y bones. In dipnoans, the anterior edge of the opercular is situated against the quadrate, which defines the front wall of the gill chamber. We conclude, therefore, that in lowadipterus the gill chamber was situated further back relative to the Y bones than in Holodipterus. This is important because it suggests that the elongation of the posterior part of the head and the cheek is the result not only of the forward position of the eye, but also the posi- tion of the gill chamber. Consequently the func- tional significance of the position of the gill chamber has to be investigated as well as the space for adductor muscles. It seems probable that elongation of the head may be the result of more than one factor. In addition to muscle space and gill chamber position, we consider that elon- gation of the parasphenoid to roof over the buccal chamber, in which air was held prior to its being forced down into the lungs, could also be signifi- cant (Campbell & Barwick, 1988a). Information on the position and size of the parasphenoid in lowadipterus is not at present available, and so we cannot make a judgement about how it fits above the gill chamber. In summary, we see no evidence that the exten- sion of the posterior part of the head and the elongation of the cheeks, is likely to outline a distinctive group of dipnoans. ACKNOWLEDGEMENTS We are grateful to John Long for making avail- able one of the specimens of Adololopas. Elga Mark-Kurik gave us samples from Estonia for comparative work. Natalia Krupina allowed us to see photographs of Andreyevichthys. Thin sec- tions were made by Norman Fraser and John Vickers, and SEM replications were prepared by Roger Heady and Frank Brink. LITERATURE CITED BARWICK, R.E., CAMPBELL, K.S.W. & MARK- KURIK, E. 1997. Tarachomylax: a new Early Devonian dipnoan from Severnaya Zemlya, and its place in the evolution of the Dipnoi. Geobios 30(1): 43-75. BAUME, L.J. 1980. The biology of pulp and dentine. A historic, temiinologic-taxonomic, histologic- biochemical. embryonic and clinical survey. Monographs in Oral Science 8: 1-246. CAMPBELL, K.S.W. & BARWICK. R.E. 1983. Early evolution of dipnoan dentitions and a new genus DEVONIAN TOOTH-PLATED DIPNOI FROM W.A. 437 Speonesydrion. Memoirs of the Association of Australasian Palaeontologists 1; 17-29. 1984. Speonesydrion an Early Devonian dipnoan with primitive tooth plates. PaIaeoIchthyoloeia2; 1-48. 1987. Paleozoic lungfishes - a review. In Bemis, W.E., Burggren W.W. & Kemp, N.E. (eds) The biology of lungfishes. Journal of Morphology Suppl. I: 93-131, 1988a. Geological and palaeontological informa- tion and phylogenetic hypotheses. Geological Magazine 125:207-227. 1988b. Uranolophus: reappraisal of a primitive dip- noan. Memoirs of the Association of Australasian Palaeontologists 7: 87-144. 1990. Paleozoic dipnoan phylogeny: functional complexes and evolution without parsimony. Pa- leobiology 16(2): 143-169. (In Press) Middle Devonian dipnoan tooth-plates from Estonia. (Submitted). CAMPBELL, K.S.W. & SMITH, M.M. 1987. The De- vonian dipnoan Holodiptenis: dental form vari- ation and remodelling growth mechanisms. Records of the Australian Museum 39: 131-167. CLOUTIER, R. 1996. Dipnoi (Akinetia; Sarcoptery- gii). Pp. 198-226. In Schultze, H-P, & Cloutier, R. (eds) Devonian fishes and plants of Miguasha, Quebec, Canada. (Pfeil: Munchen). DENISON, R.H. 1968. Early Devonian lungfishes from Wyoming, Utah and Idaho. Fieldiana: Geology 17(4): 353-413 1974. The structure and evolution of teeth in lung- fishes. Ficldiana:Geology 33(3): 31-58. KEMP, A. 1992. Late Devonian lungfish microfossils from the Coffee Hill Member, Catombal Group, near Orange in New South Wales, Australia. P. 27. In Teng, W.C. & Li Jiaxing (eds) Abstracts from; International Symposium on Devonian System and its Economic Oil and Mineral Resources. 1994. New insights into the Devonian lungfish genus Chirodipterus. P. 9. In Turner, S. (ed.) Ichthyolith Issues, April, 13. LEHMAN, J-P. 1959. Les dipneustes du Devonien superieur du Greenland. Meddelelser om Gron- land 160(4): 1-58. LUND, R., BARTHOLOMEW, P. & KEMP, A. 1992. The composition of the dental hard tissues of fishes. Pp. 35-71. In Smith, P. & Tchernov, E. (eds) The structure function and evolution of teeth. (Freund: Tel Aviv). MILES, R.S. 1977. Dipnoan (lungfish) skulls and the relationships of the group; a study based on new species from the Devonian of Australia. Zoologi- cal Journal of Linnean Society of London 61: 1-328. PRIDMORE, P.A. CAMPBELL, K.S.W. & BAR- WICK, R.E. 1994. Morphology and phylogenetic position ofthedipnoansofthe Upper Devonian of northwestern Australia. Philosophical Transac- tions of the Royal Society of London, B 244- 105-164. SCHULTZE, H-P. 1992. A new long-headed dipnoan (Osteichthyes) from the Middle Devonian of Iowa. Journal of Vertebrate Palaeontology 12(1)- 42-58. SCHULTZE, H-P. & MARSHALL, C.R. 1993. Con- trasting the use of functional complexes and iso- lated characters in lungfish evolution. Memoirs of the Association of Australasian Palaeontologists 15:211-224. SMITH, M.M. 1984. Petrodentine in extant and fossil dipnoan dentitions: microstructure. Proceedings of the Linnean Society of NSW 107: 367-407. 1985. The pattern of histogenesis and growth of tooth plates in larval stage of extant lungfish. Journal of Anatomy 140(4): 627-643. SMUTI, M.M. & CAMPBELL. K.S.W. 1987. Com- parative morphology, histology and growth of the denial plates of the Devonian dipnoan Chirodip- terus. Philosophical Tran.sactions of the Royal Society of London, B 3 1 7; 329-363. SMITH, M.M., SMITHSON, T.R. & CAMPBELL, K.S.W. 1987. The relationships of Uronemns: a highly modified dipnoan with highly modified tooth plates. Philosophical friinsactions of the Royal Society of London, B 317: 299-327. WEIDENREICH, F. 1925. Uberden Schmeltzder Wir- beltiere und seine Beziehnungen zum Zahnbein. Zeitschrift fuer Anatomic und Entwick- lungsgeschte 79: 292-351. 438 MEMOIRS OF THE QUEENSLAND MUSEUM CONSUMPTION OF JDIOSPERMUMA USTRAUENSE SEEDS BY THE MUSKY HAT K.\NG VROO, HYP- SIPRYMNODON MOSCHATUS (Marsupialia: Potoroi- dae). Memoirs of the Queensland Museum 42(2): 428. / 998:- Information on the feeding ecology of Hypsiprymnodon moschatus Ramsay, 1876. a primitive diurnal marsupial, is scant. Our present knowlege has mainly come through the PhD research of Andrew Dennis (unpbl.,. lames Cook Univer- sity), carried out on the Atherton Tablelands. NE Queensland. The current study is based on our daylight observations of Musky Rat Kangaroos (April 1 6 to May 4. 1997) in the Cooper Ck catchment. Cape Tribulation NP. During this period. Musky Rat Kangaroos were twice observed eating the seeds of Idiospermum australiense. Consumption of these seeds causes cattle deaths (Everist, 1981). Idiospermum australiense is relatively common in the study site, about 1 Om from Cooper Ck in an area of lowland complex mesophyll vine forest, and often grows in small stands of 5- 1 0 plants. The ground beneath these plants is littered with seeds of all ages; 100 seeds were counted in the particular stand studied. Other stands occur throughout the surrounding rainforest. The fruit ranges from 80-!00mm in diameter. It has a brown pericarp which splits to reveal a seed with 3-6 large cotyledons. April is a month of low fruit abundarjce in tliis area; fruiting species include candlenut {Aleurites moluccana or A excelsa ) and the fig {Ficus variegata). During 1 06 hrs of observation, we recorded 44 instances of feeding at the site. On April 18 at 2:25pm. a Musky Rat Kangaroo was seen to pick up an Idiospermum seed witli its forelimbs, then chew it for about 30 seconds, before dropping it and hopping away, this was repeated at 3:35pm. The same seed was chewed on both occasions. It appeared to be a fairly old seed with no pericarp. Approxi- mately one fiftli of the seed had been removed and charac- teristic teeth marks were evident. Idiospermum seeds at several stands were examined to determine the percentage that had been chewed. To establish the age of seeds, we recorded whether or not the pericarp was intact and whether it was fresh or brown {the dark colour is indicative of older seeds). The state of the seed; untouched, chewed, or scratched wa.s also recorded. Freshly chew ed seeds are almost white; they darken as they oxidise over a few days (presumably by polyphenol-oxidase activity). We examined 375 seeds iuid found lhal 51% were chewed, 40% were untouched, and 9% were scratched. Seeds with only their pericarps chewed were not observed during this study. In most cases the seed had been chewed where the pericarp was already broken. Most chewed seeds had only a small percent- age of the cotyledons removed, and these seeds were of all ages. Older seeds were more heavily chewed. The chew marks were similar to those made by the Musky Rat Kangaroo that we had previously observed. We seeded the stand of Idiospermum with 20 Aleurites moluccana seeds to see whether H. moschatus showed a preference for either of the two species. A. moluccana seeds were picked from a fallen branch using gloves to avoid biasing the animal's choice by touching them with bare hands. The seeds were set out at 7:00am and collected at 5:00pm for 5 days. The first day, 10 of the 20 seeds were chewed or removed. None were touched during the following 4 days. The presence of the A. moluccana seeds did not seem to deter H. moschatus from feeding on the Idiospermum .seeds. On the third day of seeding, 10% of the Idiospermum seeds were freshly chewed (7/70). On the fifth day, 4% (3/72) were freshly chewed . The total percentage chewed rose from 48% before we seeded to 57% on the fifth day. It seems reasonable to suggest that H. moschatus were not feeding on A. moluc- canna seeds in preference to Idiospermum seeds. We also observed H. moschatus eating aitd dispersing the large seeds of the cycad Lepidozamia hopeii, another species assumed toxic to most mammals (Everist, 1981). A Musky Rat Kangaroo was observed (May 4) picking up, chewing, and then running off with a L. hopeii seed. This occurred 3 times in a 3-hour period. Several of the seeds left on the ground were also chewed. Behaviour similar to this obsen^ation had been observed by the landowners on previous occasions (P. Hewitt, pers. comm). Acknowledgements For their hospitality, we thank Prue and Neil Hewitt (own- ers of the Cooper Ck property on which the observ'ations were made). Allison Drew (a School for International Training student) was supported in part through the Australian Tropical Research Foundation and the Cape Tribulation Tropical Re- search Station. Literature Cited EVERIST S-L. 1981. Poisonous Plants of Australia. (Angus & Robertson: Sydney), A. Dre%v, c/o School for International Training, Holloways Beach, Cairns 4870, Australia & H.J. Spencer, Cape Tribu- lation Tropical Research Station, PMB 5 Cape Tribulation 4873, Australia: 1 1 August 1997. ABERRANT PLUMAGES IN BIRDS OF PARADISE (PARADISAEIDAE) CLIFFORD B. FRITH AND DAWN W. FRITH Frith, C.B. & Frith, D.W. 1 998 06 29: Aberrant plumages m birds of paradise (Paradisaeidae). Memoirs of the Queensland Museum 42(2): 439- 443. Brisbane. ISSN 0079-8835. More than 6,000 bird of paradise (Paradisaeidae) skins were examined in museum collections. Aberrant plumage was recorded in 1 2 of the 42 species, and less than 1 % of the total sample. Most aberrant individuals exhibited partial albinism in showing a few white feathers, or a patch of them, in otherwise normal plumage. Two specimens were predominantly white. Two specimens of Cnemophilus macgregorii, two of Ptiforis magnificus, and one of Cicinnurus respublica showed other aberrant plumage. All aberrancies are described and discussed, and several of these are illustrated. □ Paradisaeidae, aberrant plumage, partial albinism. Clijford B. Frith and Dawn \V. Frith, Honorary Research Fellows of the Queensland Museum, ‘Prionodura ’, P.O. Box 581, Malanda, Queensland 4885, Australia: 1 0 December 1997. During recent visits to the majority of world plumages of this species and of congeneric museums (see acknowledgements, and Frith & Loria’s Bird of Paradise C. /or/'ae were known to Frith, 1997b) holding significant numbers of birds of paradise, we examined all specimens of the Paradisaeidae. One of us (CBF), measured 5,677 of the sexed skins. Studies resulting from this work to date include the description of some previously unknown plumages (Frith, 1987, 1996; Frith & Harrison, 1989), reassessment of some taxa (Frith & Frith, 1996a,b, 1997b), a review and summary of biometrics (Frith & Frith, 1997a; Frith, 1997) and a review of a classifica- tion of the phylogeny and systematics of the family (Frith & Beehler, in press). In addition to these more quantitative studies, note was made of specimens exhibiting aberrant plumage of any kind. The most commonly found type of aberrant plumage was partial albinism; which may involve some, to most, of the otherwise normally-pig- mented plumage of a bird being white (whole or part feathers may be white). This paper descri bes and discusses bird of para- dise specimens showing visually noticeable aber- rant plumage characters. Where registration or catalogue numbers include a prefix on the speci- men label these are included but in remaining cases only the number appears. In the following, capitalised and/or numbered colours are those of Smithe ( 1 975) while uncapatilised ones are those we consider more helpful than other options. The sequence of genera and species and names used below are those presented in a forthcoming re- view of the systematics of the family (in Frith & Beehler, in press). CRESTED BIRD OF PARADISE Cnemophilus macgregorii DeVis, 1890. $ and immature S be generally olive-green until a distinctive grey Juvenile plumage was found in both species (Frith, 1 987; Frith & Harrison, 1989). Two speci- mens of C macgregorii were subsequently found to be in a previously unrecorded, aberrant ‘fawn’ plumage (see Frith, 1996). As the fawn plumage in this taxon has not been illustrated, an unsexed specimen CGI 898 1381 from the Astrolabe Mts, Papua New Guinea (Museum National d’ Histoire Naturelle, Paris) is shown here (Plate lA). The pale bill (normally black or blackish) of this specimen strongly supports the conclusion that the plumage is aberrant and not merely faded or foxed. It is best described as non-eumelanic schi- zochroism (Harrison, 1985). Two of 183 speci- mens examined (1.1%) had aberrant plumage. MACGREGOR’S BIRD OF PARADISE Mac- gregoriapulchraDeM\^. 1897. Specimen 1305 in the Staatliches Museum fiirNaturkunde, Stuttgart shows partial albinism in having a few white feathers in its abdomen plumage. It is labelled as 6 but is quite possibly a 9 in view of its small size (see Frith & Frith, 1997b). One of 76 speci- mens examined (1.3%) had aberrant plumage. PARADISE CROW Lycocorax pyrrhopterus (Bonaparte, 1851). Specimen 60787, an adult 9 in the Forschungsinstitut und Naturmuseum Senckenberg, Frankfurt is a partial albino. It dis- plays typical plumage except for a single pure white feather in its mantle. Specimen 6921, an adult 9 in the Museum Zoologicum Bogoriense, Bogor is a partial albino. It has a large patch of 440 MEMOIRS OF THE QUEENSLAND MUSEUM ABERRANT PLUMAGE IN BIRDS OF PARADISE 441 white feathers in its central abdomen plumage. Two of 224 specimens examined (0.9%) had ab- errant plumage. CURL-CRESTED MANUCODE Manucodia comrii Sclater, 1 876. Specimen 784692, an adult 9 in the American Museum of Natural History, New York has several white feathers in its throat, chest and abdomen plumage. One of 79 speci- mens examined (1.3%) had aberrant plumage. SPLENDID ASTRAPIA Astrapia splendidissima Rothschild, 1895. Specimen 342102, a 9 in the American Museum of Natural History shows par- tial albinism. Its head, nape, throat and upper chest are white with a few black flecks. The upperwing (i.e., greater coverts) is white as are the inner primaries and secondaries plus a tertial on one side. There are a few odd white feathers in the breast, abdomen and vent. The legs and feet are conspicuously piebald with all of the claws pale. Specimen 302958, an undoubted 9 (its oocytes being drawn on the label) in the American Mu- seum of Natural History has a single iridescent magenta (2) adult (3-like feather in upper breast - also several larger lower back feathers have half the vane white. Two of 204 specimens examined (1%) had aberrant plumage. STEPHANIE’S ASTRAPIA Astrapia stephaniae (Finsch& Meyer, 1885). Specimen 1900.2.19.16, an adult 6 in the Natural History Museum, Tring is a partial albino in showing several white feath- ers in the lower green throat and crown plumage, plus one white feather in the nape. Specimen 12.478, an adult 3 in the Zoologische Staats- sammlung, Munchen is in normal plumage, save a single white feather in its crown. Two of 291 specimens examined (0.7%) had aberrant plum- age. MAGNIFICENT RIFLEBIRD Ptiloris rnagnif- icus (Vieillot, 1819). An unnumbered adult 3 in the Rijksmuseum van Natuurlijke Historic, Leiden is aberrant. It lacks almost all colour, there being a mere trace of iridescent purple-blue be- tween its fore-eyes and nostrils only. This bird is a washed-out matte, pale-smoky, brownish-grey above (somewhat reminiscent of Semioptera). The wings and tail are paler (see Plate I B). There is a single teardrop-sized, and shaped, spot of typically normal iridescence on one central tail feather (12mm above its tip). The crown feather- ing is typically scale-like, but is matte, smoky, brownish-grey and is soft in texture. The breast shield is normal in shape and structural stiffness. It is slightly glossy, brownish-grey with no irides- cence. The tract of feathers between the sides of the breast shield and mantle are near velvet black, being dark brownish-grey. Remaining under parts are blackish-brown (closest to 121), with no sign of the usual claret red or green sheens. Quill shafts of primaries, secondaries and rectrices are white, clearly emphasizing this abnormal plum- age. A Dutch publication noted the bird had cer- tainly not been preserved in fluid (BQttikofer, 1 895) and in any event, iridescence becomes evi- dent in fluid-preserved specimens when dried out (M. LeCroy, in litt.). This specimen is most ap- propriately described as demonstrating dilution or leucism (C.J.O. Harrison, in litt.). Specimen CGI 930 No. 440, an adult 3 in the Museum National D’Histoire Naturelle, Paris has the rear chin and throat feathers of its otherwise typically iridescent breast shield glossy-blackish each with a pale, off-whitish or very pale buff, centre and broad outer edging (see Plate I C). Two of 524 specimens examined (0.4%) had aberrant plumage. PALE-BILLED SICKLEBILL Drepanornis bruijnii Oustalet, 1880. Specimen 10239, an im- mature 3 in the Staatliches Museum fur Natur- kunde, Stuttgart shows partial albinism. The terminal two thirds of the second outermost pri- mary of its right wing are pure white save that its central shaft is darkly pigmented. One of 151 specimens examined (0.7%) had aberrant plum- age. MAGNIFICENT BIRD OF PARADISE Cicin- mrus (Diphyllodes) magnificus (Pennant, 1781). Specimen Cl 1289, III, 174, an adult 3 in the Staatliches Museum fiir Tierkunde, Dresden is a FIG. 1. A. unsexed and mounted Cnemophilus maegregorii Museum National d’Histoire Naturelle specimen CGI 898 138L B. adult 3 Ptiloris magnificus Rijksmuseum van Natuurlijke Historic unnumbered specimen. C, throat and breast detail of adult 3 P. magnificus Museum National d'Histoire Naturelle CGI 930 No. 440. D, dorsal view of adult 3 3 Staatliches Museum fiir Tierkunde Cl 7848 Paradisaea minor (ki'i) SMTCl 1289. MI, 174 Cicinnurus (Diphyllodes) magnificus (right). E, dorsal view, and F, ventral view, of (left) adult 3 Staatliches Museum tiir Naturkunde SNHM 15522 Cicinnurus (Diphyllodes) respublica and (right) a typically normally-plumaged adult d (SNHM 44502) of that species in the same collection. 442 MEMOIRS OF THE QUEENSLAND MUSEUM partial albino. It is white throughout except that its underparts (chin and throat excluded) are nor- mal iridescent green and it has a few brownish back and wing feathers and one or two mustard yellow tertials (Plate ID). Specimen 454366, an adult S in the American Museum of Natural History is also a partial al- bino. It has several white feathers in its upper- parts, and a few in the underparts. The primaries and secondaries and most of crown, nape and cape is white. Its tarsii and claws are whitish and the toes are piebald. Part of this specimen is illustrated by a colour photograph in Purcell & Gould (1988: 75, pi. 52). Specimen 454367, an adult 3 trade skin lack- ing the central rectrices (from unknown locality) in the American Museum of Natural History, is almost entirely white . It has one almost normally pigmented secondary with scattered blackish feathers (tipped in green) in its ventral plumage, more so in the flanks. Specimen 678406 in the American Museum of Natural History, a flat, legless and distorted adult 3 trade skin (lacking the central rectrices), has plumage similar to that of 454366, above. Four of 600 specimens exam- ined (0.7%) had aberrant plumage. WILSON’S BIRD OF PARADISE Cicinnurus (Diphyllodes) respublica (Bonaparte, 1850). Specimen SNHM 15522, an adult 3 in the Staatliches Museum fiir Naturkunde, Stuttgart is unique in that all its underparts, and most of the upperparts, are a deep rich fawn colour (27). However, the normal yellow cape and the red of the central back (bordered posteriorly by a black line) and wing patch are present. The back and secondary coverts are, however, less red and more orange than usual. The bare head skin is not black (as in all other dry skin specimens) but dark brown, suggesting it may not have been the usual blue colour in life. The usually dense and plush black head and throat feathers are silvery-buff in this bird. The central tail ‘wires’ are missing, presumably moulted. This individual’s appear- ance fits that of non-eumelanic schizochroism (Harrison, 1985) (see Plate IE & F). One of 94 specimens examined (1.1%) had aberrant plum- age. LESSER BIRD OF PARADISE Paradisaea minor Shaw, 1809. Specimen SMTC 17848, an adult 3 in the Staatliches Museum fiir Tierkunde, Dres- den is extensively a partial albino. It is entirely glossy white save for a pale yellow wash on the crown. It is even paler, on mantle and wing-cov- erts. It has a single pale brown tertial, wing and tail feather (Plate ID). One of 565 specimens examined (0.2%) had aberrant plumage. BLUE BIRD OF PARADISE Paradisaea {Para- disornis) rudolphi (Finsch, 1885). Two colour photographs of a live partial albino 9 P. rudolphi, taken under controlled conditions in New Guinea, appear in Gilliard (1953: 441). This bird is in typical plumage but shows a white patch of ter- tials, a few white primaries, secondaries and lesser upper wing coverts, and a white terminal third to its outer left tail feather. None of the 90 museum specimens examined showed schizo- chroic characters or those of any other aberrant plumage. DISCUSSION Birds of paradise showing even the smallest amount of aberrant plumage are few and represent less than 1% of the above species samples com- bined. In examining more than 6,000 museum skin specimens (including unsexed individu- als) and in reviewing the literature only the above individuals came to our notice. Clearly, most of these examples of aberrant plumage involve indi- viduals showing partial albinism. In most cases there are no more than a few white feathers, or a patch of them, in otherwise typically normal plumage. Given the large sample size, partial albinism is clearly rare, while a true albino (en- tirely white plumage throughout with associated pale iris) individual remains unrecorded, in the Paradisaeidae. This said, it is possible that such unusual looking individuals were specifically sought or purchased by field collectors and trad- ers because of their greater value as novelties. The single adult- (J like feather in an otherwise normally-plumaged 9 A, splendidissima is pos- sibly an example of the well-known phenomenon of an older 9 individual bird showing some sign of adult 3 nuptial plumage. Of the above documented abnormal bird of paradise plumages, those of Cnemophilus macgregorii, Ptiloris magnificus and Cicinnurus respublica are the most noteworthy, 'fhe aberrant plumage of the P. magnificus specimen is appar- ently the result of a reduction in the black pigment eumelanin which results in the bird having a distinctly washed-out or ‘faded’ appearance but with no tendency toward a ‘warming’ of appear- ance by fawn colouration. The above aberrant plumages of both C macgregorii and C respub- lica are noteworthy, however, as their clearly ‘fawn’ and ‘warm’ colouration is generally simi- ABERRANT PLUMAGE IN BIRDS OF PARADISE 443 lar to the perfectly normal but ‘faded’ plumage appearance of the Standardwing Bird of Paradise Semioptera wallacii. It has been suggested that this odd but perfectly typical plumage morphol- ogy of 5. wallacii may be due to its island isola- tion in the absence of other polygynous birds of paradise (Graves, 1 995). This scenario could also be true of C. respublica which shows the most similarly ‘fawn’ aberration (although it does share its island ranges with the Red Bird of Para- dise Paradisaea rubra). The authors would be interested to learn of any additional examples of aberrant plumages in the Paradisaeidae. ACKNOWLEDGEMENTS We thank the following persons and institutions for access to specimens: Dean Amadon, Chris Blake, Walter Bock, Joel Cracraft, Mary LeCroy, Manny Levine, Lester Short and The Frank M. Chapman Memorial Fund, American Museum of Natural History, New York, USA; Rene Dekker and Peter Van Dam, Rijksmuseum van Natuurli- jke Historie, Leiden, Netherlands; Eric Pasquet, Museum National D’Histoire Naturelle, Paris, France; Siegfried Eck, Staatliches Museum fur Tierkunde, Dresden; D.S. Peters, Martina Kusters and Karin B5hm, Forschungsinstitut und Natur- museum Senckenberg, Frankfurt; Claus Konig, Staatliches Museum fur Naturkunde, Stuttgart and Josef H. Reichholf, Zoologische Staat- ssammlung, Munchen, Germany; Mohammad Amir and Darjono, Museum Zoologicum Bo- goriense, Bogor. Gary Graves, Colin Harrison and Mary LeCroy most kindly and helpfully com- mented on a draft of this note. LITERATURE CITED BUTTIKOFER, J. 1 895. Einige bemerkungen iiber neu angekommene paradiesvOgel. Notes from the Leyden Museum 17: 36-40. FRITH, C.B. 1987. An undescribed plumage of Loria’s Bird of Paradise Loria loriae. Bulletin of the British Ornithologists’ Club 107: 177-180. 1996. Further notes on little-known plumages of the Crested and Loria’s Birds of Paradise Cnemophi- lus macgregororii and C. loriae. Bulletin of the British Ornithologists’ Club 1 16: 247-251. 1997. Huia {Heteralocha acutirostris, Callaeidae) - like sexual bill dimorphism in some birds of paradise (Paradisaeidae) and its significance. No- lornis44: 177-184 FRITH, C.B. & BEEHLER, B.M. 1998. The Birds of Paradise. (Oxford University Press: Oxford). (In press). FRITH, C.B. & FRITH, D.W. 1996a. Description ofthe unique Parotia lawesii x Paradisaea rudolphi hybrid bird of paradise (Paradisaeidae). Records of the Australian Museum 48: 111-116. 1996b. The unique type specimen of the bird of paradise Lophorona superba pseudoparotia Stresemann 1934 (Paradisaeidae); a hybrid of Lophorina suberba X Parotia carolae. Journal fur Omithologie 137: 515-521. 1997a. The taxonomic status of Paradigalia carun- cidata intermedia (Paradisaeidae) with notes on the other Paradigalia taxa. Bulletin of the British Ornithologists’ Club 1 17: 38-48. 1997b. Biometrics of birds of paradise (Aves: Paradisaeidae) with observations on interspecific and intraspecific variation and sexual dimor- phism. Memoirs of the Queensland Museum 42: 159-211. FRITH, C.B. & HARRISON, C.J.O. 1989. An unde- scribed plumage of the Crested Bird of Paradise Cnemophilus magregorii. Bulletin of the British Ornithologists’ Club 109: 137-139. GILLIARD, E.T. 1953. New Guinea’s rare birds and stone age men. National Geographic Magazine 103:421-488. GRAVES, G.R. 1995. Sequence of plumage evolution in the Standardwing Bird of Paradise. Wilson Bulletin 107: 371-3. HARRISON, C.J.O. 1985. Plumage, abnormal. In Campbell, B. & Lack, E. (eds) A Dictionary of Birds. (Poyser: Calton). PURCELL, R.W. & GOULD, S.J. 1988. Finders Keep- ers. (Hutchinson Radius: London). SMITHE, F.B. 1975. Naturalist’s Color Guide. (Ameri- can Museum of Natural History: New York). A MATURE CUPULE OF T£Pra5'77?05U5'(CZEKAN0WSKIALES) FROM THE LATE TRIASSIC OF QUEENSLAND H.T. CLIFFORD AND N. CAMILLERI Clifford, H.T. & Camilleri, N. 1998 06 29: A mature cupule of Leptostrobus (Czekanowskiales) from the Late Triassic of Queensland. Memoirs of the Queensland Museum 42(2): 445-447. Brisbane. ISSN 0079-8835. A single bivalvate cupule of Leptostrobus is reported from Upper Triassic sediments of the Ipswich Coal Measures. This record is the oldest known for the genus. □ Leptostrobus, Triassic, Australia. H. Trevor Clifford and Natalie Camilleri, Queensland Museum, PO Box 3300, South Brisbane 4101, Australia; 21 February 1998. The Czechanowskiales were first recognised as occurring in Australia when Jones & de Jersey (1947) recorded Czekanowskia tenuifolia (Johnston) from Late Triassic Ipswich Coal Measures. The species had originally been de- scribed by Johnston (1887) from Lord’s Hill near Hobart. At that date they were regarded as be- longing to Baeira a genus of Ginkgoales. On the basis of its epidermal structure the species B. tenuifolia was transferred to Czekanowskia by Jones & de Jersey (1947). The similarity of B. tenuifolia, as it was then known, to Czekanowskia microphylla had been noted by Walkom (1925), indicating an overlap in gross morphology of the two species. In addition to Czekanowskia Jones & de Jersey (1947) also recorded a single presumed cupule- bearing axis of Leptostrobus in the same deposit. The cup-shaped organs are much smaller than the cupule described here and so may be regarded as immature. Although Czekanowskia and Leptostrobus have not been encountered in organic connection, their frequent joint occurrence is taken to indicate that they represent the foliage and cupules of the same taxon. STRATIGRAPHY The specimen described was exposed by the splitting of fossiliferous, siltstone float close to a dirt track beginning at the end of Thomas St, on the W side of ‘Castle HilF, Blackstone. On the basis of lithology the specimen is considered as being in situ. Whether these siltstones represent a horizon within the Rhaetian Aberdare Conglomerate or the Camian Blackstone Formation is difficult to determine with certainty, because the two are separated by an unconformity, and siltstones within the Aberdare Conglomerate are lithologi- cally similar to some beds in the Blackstone For- mation (Cranfield et aL,1976). Nonetheless, it is certain the siltstone float is Upper Triassic (de Jersey, 1975). SYSTEMATIC PALAEOBOTANY Division Gymnospermae Order Czekanowskiales Harris & Miller Leptostrobus Heer Leptostrobus cookii sp. nov. (Figs 1-2) ETYMOLOGY. For Alex Cook in recognition of his enthusiasm for all things geological. MATERIAL. Holotype: QMF37262a, (convex) Ips- wich, Queensland 27°37’39"S, 152M7’58"E; QMF37262b, (concave) counterpart. DIAGNOSIS. Coalified cupule consisting of two similar oppositely placed valves obovate in out- line. Internal surface with slightly raised longitu- dinal ribs. Seeds and ovules missing. DESCRIPTION. The cupule is completely coali- fied and no cell detail is available, T^e scales are obovate in outline, 16mm long and 12mm broad. The thickening of the valve can be estimated from the fractured surfaces. Towards its apex the valve is about half as thick as at its base (£5-3. 0mm). The inner surface of one scale is concave and that of the other is convex which suggests that the latter has been pushed into the former during the process of fossilization (Figs I, 2). DISCUSSION Well-developed cupules of Leptostrobus have not been reported previously from Australia al- though its associated foliage Czekanowskia (as Baiera) has been known for more than a century. Since the Ipswich area, in particular, has been intensively collected over a long period it is clear the cupules are rare. 446 MEMOIRS OF THE QUEENSLAND MUSEUM 1cm FIG. I . Internal surfaces of the valves of a mature cupule of the holotype of Czekanowskia cookii sp.nov. A, QMF37262a; B, QMF37262b (surface convex due to compression). As judged by the rotting vegetation on the floor of a small lake at J.C. Slaughter Falls in the Mt Coot-tha Reserve near Brisbane this situation is perhaps to be expected. Although leaves of Euca- lyptus and Corymbia are abundant on the lake- FIG. 2. Diagrammatic representation of a mature cu- pule of the holotype of C cookii sp. nov. based on the valves illustrated in Fig LA, valve QMF37262a. B, valve QMF37262b. C, hypothetical reconstruction of cupule. X = valve margin, f = cracks in wall of valve. bed the woody fruits of the two genera are rare (Clifford, 1990). Seed-bearing structures of any kind are likewise rare in the Ipswich beds which were also laid down in freshwater. Aside from Leptostr obits ^ the only other seed-bearing structures recorded in the sediments are Fraxinopsis major and a possi- ble corystosperm megasporo- phyll (Hill et al., 1966; Jones & de Jersey, 1947). In his original description of Czekanowskia tenuifolia Johnston (1887) refered to abundant associated reproduc- tive structures in the material. He provided detailed descrip- tions and some figures but these are difficult to interpret and so will not be considered further. The presence of Czekanowskiales in the Trias- sic of Australia has been long over-looked. In the paper in which the Order was established (Harris & Miller, 1974) no reference was made to the work of Jones & de Jersey (1947) which was also apparently unknown to either Taylor & Taylor (1993) or Spome (1974) who claim the Czekan- owskiales extended only from the Jurassic to the Cretaceous. According to Max Banks (pers. comm.) the foliage reported from the New Town shales at Lord’s Hill are Ladinian and so are slightly older than the Ipswich deposits which were laid down about 235MA (Harland et al., 1989). The geologi- cal range of the Czekanowskiales therefore is to be extended back from the early Jurassic to the Middle Triassic. ACKNOWLEDGEMENTS We thank Sarah and Carole Thomas for alerting us to the site where this specimen was found. John McKellar of the Geological Survey of Queens- land provided helpful comments. LITERATURE CITED CLIFFORD, H.T. 1990. The historian is a prophet in reverse. Review of Palaeobotany and Palynology 64: 5-11. CRANFIELD, L.C., SCHWARZBROCK, H. & DAY, R.W. 1976. Geology of the Ipswich and Brisbane 1:250 000 Sheet Areas. Geological Survey of Queensland. Report 95: 1-176. MATURE CUPULE OF LEPTOSTROBUS 447 de JERSEY, NJ. 1975. Miospore Zones in the Lower Mesozoic of southeastern Queensland. Pp. 159- 172. In Campbell. K.S.W. (ed.) Gondwana Geol- ogy, Papers presented at the 'fhird Gondwana Symposium, Canberra, Australia 1973. (Austra- lian National University Press: Canberra). HARLAND, W.B., ARMSTRONG, R.L., COX, A.B., CRAIG, L.C., SMITH, A.G. & SMITH, D.G. 1989. A geologic time scale 1989. (Cambridge University Press: Cambridge). HARRIS, T.M. & MILLER, J. 1974. The Yorkshire Jurassic Flora IV, 2. Czckanowskiales. (British Museum Natural Histoiy: London). HILL, D., PLAYFORD, G. & WOODS, J.T. 1966. Triassic Fossils of Queensland. (Queensland Pa- laeontographical Society: Brisbane). JOHNSTON, P.M. 1887. Fresh contribution to our knowledge of our knowledge of the plants of Mesozoic Age in Australia. Papers and Proceed- ings of the Royal Society of Tasmania 1886: 160-182. JONES, O. A. & de JERSEY, N. J. 1 947. The flora of the Ipswich Coal Measures - morphology and floral succession. Papers of the Department of Geology, University ofQueensIand 3(3): 1-88. SPORNE, K.R. 1974. The morphology of Gymno- sperms. (Hutchinson University Library: Lon- don). TAYLOR, T.N. & TAYLOR, E.L. 1993. The biology and evolution of fossil plants. (Prentice Hall: Edgewood Cliffs). WALKOM, A.B. 1925. Notes on some Tasmanian Mesozoic plants. Part 1. Papers and Proceedings of the Royal Society ofTasmania 1924: 73-89. 448 MEMOIRS OF THE QUEENSLAND MUSEUM FIRST RECORD OF CA YTONIA IN AUSTRALIA Mem- oirs of the Queensland Museum 42(2): 448. 1 998:~ An axis bearing two rows of shortly stalked ovule-bearing cupules and attributable to Caytonia has been observed in Lower Jurassic sandstones belonging to the Marburg Formation, at Durikai in SE Queensland. Support for the identification is provided by impressions oX Sagenopteris rhoifolia in tl>e same beds (Hill et al., 1966). Though not as yet found in organic connection Caytonia and Sagenopteris regularly co-occur and are usually regarded as being the seed-bearing organ and foliage of the same taxon (Taylor & Taylor, 1993). Division PTERIDOSPERMAPHYTA Order CAYTONIALES H.H. Thomas Family CAYTONIACEAE Pilger & Melchior Caytonia tierneyi sp. nov. Etymology. For Paul Tierney the collector. Material Holotype; QMF39005, Durikai 38'’12'S, 15U37’E. Paratype: OMF3588L same locality. Diagnosis. An unbranched axis bearing two rows of alternat- ing co-planar, shortly stalked, transversely elliptical cupules each bearing several ovules. Description. Axis at least 1 L5mm long bearing two rows of cupules each about 1.0-1. 2mm x 2.0mm, borne on stalks up to 1mm long. Both axis and cupule stalks have a median groove on their abaxial surfaces. (Figs 1, 2). Remarks. The new species differs from others of Caytonia in that the cupules are elliptical rather than spherical and are arranged alternately instead of oppositely or suboppositely on their axis (Thomas, 1925). FIG. 2. Mould of an inflorescence of Caytonia tierneyi sp. nov., Paratype QMF35881. Scale bar in mm. Acknowledgements Natalie Camilleri and Alex Cook are thanked for the pho- tographs. Literature Cited HILL, D.. PLAYFORD, G. & WOODS, J.T. 1966. Jurassic fossils of Queensland. (Queensland Palaeontographical Society; Brisbane). TAYLOR, T.N. & TAYLOR, E.L. 1993. The biology and evolution of fossil plants. (Prentice Hall: Edgewood Cliffs). THOMAS, H.H. 1925. Caytoniales. a new group of angiospermous plants from the Jurassic rocks of York- shire. Philosophical fransactions of the Royal Society ofLondon 2 13B: 299-363. H. Trevor Clifford, Queensland Museum, PO Box3300. South Brisbane 4101, Australia: 10 March, 1998. FIG. 1. Inflorescence oX Caytonia tierneyi sp. nov. with one cupule containing ovules. Holotype QMF39005. Scale bar Imm. FRASNIAN GASTROPODS FROM THE BONAPARTE GULF BASIN WESTERN AUSTRALIA ALEX G. COOK Cook, A.G. 1998 06 29: Frasnian gastropods from the Bonaparte Gulf Basin, Western Australia. Memoirs of the Queensland Museum. 42(2): 449-457. Brisbane. ISSN 0079- 8835. Eight taxa of gastropods are described from the Late Devonian (Frasnian) Westwood Member of the Hargreaves Formation, Bonaparte Gulf Basin, Western Australia. Of these Ningbingia robertsi gen. et sp. nov., Aglaoglypta veeversi sp. nov., Pseudomphalotrochus bonapartensis sp. nov., Westwooditrochus weestwoodensis gen. et. sp. nov., and Plagiothrya curlyjigga sp. nov. are new. The remaining taxa are left in open nomeclature. Pseudompha- lotrochus, Aglaoglypta and Plagiothrya relate strongly to European and/or North American Devonian taxa. □ Devonian, Frasnian, gastropods, Western Australia. AlexG. Cook, Queensland Museum, POBoxSSOO, South Brisbane 4 101, Australia: 6 August 1997. The Westwood Member of the Hargreaves For- mation is an abundantly fossiliferous unit which crops out near Westwood Ck, in the vicinity of Cape Dommet, Bonaparte Gulf Basin, WA. The member contains a diverse assemblage of algae, corals, stromatoporoids, brachiopods, conodonts, fish plates, bivalves and gastropods, of which, only the brachiopods (Roberts, 1971) and cono- donts (Druce, 1969) have received taxonomic attention. The most recent stratigraphic and sedimen- tologic summaries of the Hargreaves Formation are given by Mory & Beere (1988) and Mory (1990) who both indicated reefal and back-reefal depositional environments for this unit.The age of the Westwood Member was give as Frasnian by Druce (1969) and Roberts (1971) and sub- sequent studies (Mory & Beere, 1988; Mory, 1990) have confirmed this age. This paper de- scribes the modest gastropod faunule from the Westwood Member. All gastropods described in this work come from outcrops on, or adjacent to, tidal flats near Westwood Ck. These are: QML1035 - Westwood Ck, Cape Dommett, tidal flats, 14°5L7’ S, 128*^30. UE. Medusa Banks 1:250000 Geological Sheet SD 52-10. Collected Jell & Cook, 1996. QML 1 037 - 400m to NE of QML 1 035, amongst pinnacles of reefal limestone and oncolitic beds. Collected Jell & Cook, 1996. Phylum MOLLUSCA Class GASTROPODA Order ARCHAEOGASTROPDA Theile, 1925 Suborder BELLEROPHONTINA Ulrich & Scofield, 1897 Superfamily BELLEROPHONTOIDEA M’Coy, 1851 Family BELLEROPHONTIDAE M’Coy, 1851 Subfamily TROPIDODISCINAE Knight, 1956 Ningbingia gen. nov. ETYMOLOGY. For ‘Ningbing’ Station, upon which the material was collected. TYPE SPECIES. Ningbingia robertsi sp. nov. DIAGNOSIS. Large members of subfamily, with narrow, rapidly expanding shell, with wide um- bilicus, and an ornament consisting of strongly arcuate ribs. REMARKS. The narrow crested selenizone, the arcuate ribbing and the shell form identifies the material as belonging to the Tropidodiscinae. It is most similar to Tropidodiscus, but is distin- guished by its large size and the more arcuate ornament on the whorl face, particularly near the umbilicus. The type species of Tropidodiscus, Bellerophon curvilineatus Conrad, from the Ono- danga Limestone of New York is a moderately large member of the genus (and subfamily). The genotype has a diameter of approximately 28mm (Knight, 1941), contrasting strongly with the tiny species known from Early and Middle Devonian faunas in eastern Australia. For example T. cen- trifugalis (Chapman) of Tassell (1982) from the Receptaculites Limestone (Emsian) of New South Wales is close to one third of this size. A small tropidodiscine from the Burdekin Forma- tion (Givetian) of Queensland is only 2.5mm in diameter (Cook, 1 997). Ningbingia is up to 52mm in diameter, and is proportionately thicker than the genotype, it also possesses strong, arcuate 450 MEMOIRS OF THE QUEENSLAND MUSEUM TABLE 1. Shell measurements for gen. et sp. nov. Specimen Diameter (mm) Width (mm) OMF35010 18.1 9.5 OMF35059 38.5 17.6 OMF35061 26.5 17.5 OMF35072 52.1 25.5 OMF35n8 32.0 18.5 QMF35121 26.0 10.5 ornament in contrast to the fine growth lines of Tropidodiscus curvilineatus (Conrad). Nin&bingia robertsi sp. nov. (Fig. lA-E) ETYMOLOGY. For Professor John Roberts. MATERIAL. Holotype: QMF35059. Paratypes: QMF35054-35058. 35060-35066, 35069, 35071- 35074, 35088-35090, 35116-35118, 35121, 35010 all from QMLI035. DIAGNOSIS. As for genus. DESCRIPTION. Shell medium sized to large, sharply narrow, involute, rapidly expanding shell, up to 38.5mm in diameter and 25.5mm wide (Table 1). Umbilicus up to 1/3 of shell diameter. Selenizone on a crest, narrow, flanked by two weak cords. Ornament consists of strong arcuate ribs curving strongly into the umbilicus, and on the last whorl becomes more arcuate towards the aperture. Family BELLEROPHONTIDAE M’Coy, 1851 Aglaoglypta Knight, 1942 REMARKS. Aglaoglypta Knight 1942 was erected upon Bellerophon koeneni Clarke, from the Late Devonian (Frasnian) “Naples” Fauna of New York. Knight (1942) did not provide a list of species included in his concept of the genus, and subsequently little attention has been paid to the taxon. Despite the suggestion (Knight, 1942) and the assertion (Knight et al., 1 960) that Aglao- glypta represents a subgenus, it is retained here at generic status due to the highly diagnostic pustu- late ornament. Species here placed in Aglaoglypta are, apart from the type; Bellerophon tubercula- tus Ferrusac & d’Orbigny of d’Archaic & de Vemeuil (1842), B. maera Hall, from the Che- mung of New York and B. neleus Hall & Whit- field ( 1 877) from the Devonian of Nevada. Clarke (1903) listed B. striatus Ferrusac & d’Orbigny, B. tubercidatiis d’Orbigny (a differing authorship attributed than by d’Archaic & de Vemeuil) and B. alutaceus d’Orbigny as those Devonian Bel- TABLE 2. Shell measurements for Aglaoglypta veeversi sp. nov. Specimen Diameter (mm) Width (mm) QMF35539 13.5 14.0 OMF350I6 14.5 14.0 QMF35540 17.5 15.5 OMF35542 13.4 12.7 QMF35568 15.0 14.3 lerophon species with pustulose ornament. I have been unable to access original descriptions of B. alutaceus, B. striatus, or B. tuberculatus, so my assessment of the genus is incomplete. The genus requires a review, by an author with access to the early 19th century literature and the European material. Aglaoglypta veeversi sp. nov. (Fig. IF-J) ETYMOLOGY. For Professor J.J. Veevers. MATERIAL. Holotype: QMF35539. Paratypes: QMF35016, 35540-35560 all from QML1035. DIAGNOSIS. Ornament subdued, pustulate, ar- ranged along fine growth lines. DESCRIPTION. Small, spheroidal, involute shell, with very narrow umbilicus. Up to 1 7.5mm diameter, 15.5mm in width (Table 2). Ornament of fine growth lines with papillae arranged in rows upon them. Weak, low crest bearing seleni- zone. Aperture moderately flared where pre- served. REMARKS. The species differs from the type in having weaker pustules which occupy less space on the whorl surface. Aglaoglpyta tuberculatus (Ferussac & d’Orbigny) of d’Archaic & de Vemeuil from the Devonian (Eifel) of Paffrath has similar arrangement and strength of the pus- tules (based on the figure and limited descrip- tion). A. maera (Hall) also has stronger ornament, and a slightly more recurved lip than the Bona- parte species. Suborder MACLURITINA Cox & Knight, I960 Superfamily EUOMPHALOIDEA de Koninck, 1881 Family EUOMPHALIDAE de Koninck, 1881 Pseudomphalotrochus Blodgett, 1992 Pseudomphalotrochus bonapartensis sp. nov. (Fig. 2) ETYMOLOGY. From the Bonaparte Gulf Basin. MATERIAL. Holotype: QMF35087. Paratypes: QMF35067, 35068, 35070, 35075, 35076, 35079- GASTROPODS FROM THE BONAPARTE GULF BASIN 451 FIG. 1. A-E. Ningbingia robertsi gen. et. sp. nov. A, B, Holotype QMF35059. A, xl, B, xl.5. C, D, Paratype QMF35010 X 2. E, Paratype QMF35061 x 1. F-J, Aglaoglypta veeversi sp. nov. F, G, H Holotype QMF35539, X 2.5. 1, Paratype QMF35540 x 2. J, Paratype QMF3554I x 2.5. 35085, 35094-35098, 35100-35104, 35107, 35109- 35115, 35120, 35122-35127 all from QML1035. DIAGNOSIS. Medium sized, turbiniform, me- dium to high-spired, phaneromphaious with cir- cumbilical nodose cord. DESCRIPTION. Shell medium-sized to large, turbiniform, phaneromphaious, with a variable spire height (Table 3). Sutures impressed, whorls embrace below mid-whorl. In early whorls the whorl profile is rounded, however, in later whorls the whorl profile consists of a sub rounded, shal- low sloping subsutural surface bordered abapi- cally by a weak angulation, near vertical slightly rounded midwhorl, similarly bounded, and a rounded lower whorl profile giving way to an umbilicate base. The umbilicus occupies up to 1/3 of the base and a circumbilical nodose cord is present. Surface of the shell has adhered skeletal fragments on midwhorl, on the angulation on the upper mid whorl or has shallow pits on the mid whorl face. Growth lines are fine, numerous and nearly orthocline. Aperture is quadrate, wider than high with a slight parietal invagination. Early whorls are septate, with a few closely spaced septa in a few specimens. Generally the early whorls are missing. 452 MEMOIRS OF THE QUEENSLAND MUSEUM FIG. 2 Pseudomphalotrochus bonapartensis sp. nov. A-C, Holotype QMF35087 xl.5. D-G, Paratype QMF35I04, xl.8. H-L Paratype QMF35075 x 1.8. K, Paratype QMF35I02 x 1.8. REMARKS. Blodgett (1992) erected this genus for large gradate omphalotrochids with a circum- bilical ridge on the basal angulation. The type species P. linsleyi Blodgett, 1992 from the Eife- lian of Alaska has a more prominent angulation on the base, but shares the circumbilical cord. The Bonaparte species is most closely allied with P. hoffmanni (Linsley & Yochelson, 1973, from the late Eifelian Rogers City Limestone, Michigan. They share similar whorl profiles, and both have GASTROPODS FROM THE BONAPARTE GULF BASIN 453 TABLE 3. Shell measurements for Pseudomphalotro- chus bonapartensis sp. nov. Specimen Height (mm) Width (mm) OMF35087 15.0 21.2 OMF35I03 19.5 23.0 OMF3508I 20.9 22.5 OMF35100 23.6 28.7 OMF35104 17.8 23.0 QMF35125 22.3 27.5 a circumbilical cord, although P. hoffmanni lacks nodes on that cord. Also included in Pseudom- phalotrochus by Blodgett (1992) was Cirrus leonhardi D’ Archaic &. DeVemeuil, 1 842, which has a nodose upper whorl angulation. It is impos- sible to discuss this taxon without reference to closely related species of euomphalids. Linsley & Yochelson (1973) reviewed a number of Strapa- rollus taxa including S.(S.) hoffmanni, assigned to Pseudomphalotrochus by Blodgett (1992), with material adhered to the shell, collectively referring to them as the ‘Devonian carrier shells’. Importantly they remarked on the variability in relative spire height in both S, (StraparoUus) and S. (Euomphalus). The variation in whorl profile present in P. bonapartensis is also recorded in S. (?E.) incrustatus Linsley & Yochelson, from the Middle Devonian of Utah. It could be argued that this species should be placed in Micromphalus Knight, 1 945, given the high spire and the angular faces on the whorl profile. However the W.A. specimens have the upper whorl profule sinus, and a wide rather than a narrow umbilicus, a narrow umbilicus being diagnostic for Microm- phalus. T \\q presence of this genus in faunas of this age is not unexpected and highlights affinities to faunas in North America and Europe. Suborder INDET. Superfamily PALAEOTROCHOIDEA Knight, 1956 Family PALAEOTROCHIDAE Knight, 1956 Westwooditrochus gen. nov ETYMOLOGY. For Westwood Ck, the type locality, and trochus referring to the trochiform shell. TYPE SPECIES. Westwooditrochus westwoodensis sp. nov. DIAGNOSIS. Medium sized, squatly turbini- form, with flattened base, cryptomphalus with callus pad, nodose cord just below suture, another on angulation at whorl periphery situated well below mid-whorl; growth lines opisthocline on upper whorl face. REMARKS. The lower cord and angulation, combined with the unusual orientation of the growth lines define placement within the family, but the material cannot be accomodated in Pa~ laeotrochus Hall, 1879 which lacks the upper cord or Turbinopsis Grabau & Shimer, 1909 which not only lacks the upper cord but possesses strong ridges on the upper whorl face. The base of the Bonaparte species is more flattened than other members of the family and possesses an umbilical plug. Westwooditrochus westwoodensis sp. nov. (Fig. 3A-F) ETYMOLOGY. For Westwood Ck, the type locality. MATERIAL. Holotype: QMF35014. Paratypes: QMF35009, 3501 1 all from QML1037. DIAGNOSIS. As for genus. DESCRIPTION. Shell thick, medium sized, rap- idly expanding, squatly turbiniform. Sutures im- pressed; whorls embrace at periphery which is well below the mid-whorl. Whorl profile angular, with nodose carina at periphery. Upper whorl surface is about 2/3 of the profile, steep (approx. 45^^); adjacent to the suture it bears a spiral row of nodes. Below the periphery the lower whorl pro- file is an inclined surface itself bounded abapi- cally by a weaker nodose spiral cord. Base flattened with an additional, weaker cord near the outer margin. An umbilical pulg occupies more than half of the base. Growth lines are fine and numerous, gently curved and opisthocline on the upper whorl surface, with a slight sinus, nearly straight across the peripheral angulation and strongly prosocline on the base. Suborder NERITOPSIN A Cox & Knight, 1960 Superfamily NERITOIDEA Rafinesque, 1815 Plagiothyra Whidbome,1892 TYPE SPECIES. Monodonta purpurea d’Ar- chaic & de Vemeuil, 1842 from the Middle De- vonian of Pafffath, Germany. Plagiothyra curlyjigga sp. nov. (Fig. 4A-G) ETYMOLOGY. From the vernacular of P.A. Jell for fossil snail; ‘curly-jigger’. MATERIAL. Holotype: QMF35045. Paratypes: QMF35046-35052, 35105, 35106, 35108, 35119, all from QMLI035. DIAGNOSIS. Large member of genus with elon- gated nodes on the strong spiral ornament. 454 MEMOIRS OF THE QUEENSLAND MUSEUM FIG. 3. A-F, Westwooditrochus bonapartensis gen. et sp. nov. A, B, Paratype QMF35009 x 1.5. C, D, Holotype QMF35014 X 1.5. E, F, Paratype QMF3501 1 x 1.5. G, H, Murchisoniidae gen. et sp. indet.; G, QMF35078 x 1.5; H, QMF35093 X 1.5. DESCRIPTION. High-spired, dextral, turbiniform, thick-shelled and anomphalous. Sutures moderately impressed, whorls embrace just below periphery situated slightly below mid-whorl. Whorl profile rounded with widened mid-whorl area. Whorl face adorned by up to 13 nodose spiral cords. On the holotype the cords are sporadically interrupted such that they possess elongated nodes, present- ing a dashed-line appearance. On other specimens the nodes are more rounded. Aperture rounded with invagination near the suture. The holotype possesses a small fold on the upper part of the parietal lip as it approaches the suture. Proto- conch present on the holotype is simple, smooth and dextral. REMARKS. The single parietal fold, combined with the strong spiral, nodose ornament places the material in Plagiothyridae Knight, 1956. It is most similar to Plagiothrya Whidbome, 1 892 and is placed there given the singular parietal fold. The type species Monodonta purpurea d’ Archaic & de Vemeuil, 1842, from the Middle Devonian of Pafffath, Germany is only 22mm high, in stark contrast to the Bonaparte species, and the genus lacks the strongly elongated aspect to nodes on the spiral cords. Dirachis atavus of Whidbome GASTROPODS FROM THE BONAPARTE GULF BASIN 455 FIG. 4. A-G, Plagiothyra curlyjigga sp. nov. A-C, Holotype QMF35045 x 2. D, Paratype QMF35046 x 2. E-F, Paratype QMF35047 x2. G, Paratype QMF35051 x 2, H-J, Naticopsis {Naticopsis) sp. H, I, QMF35013 x 1. J, QMF35109 x 1, K, Murchisonia (Murchisonid) sp. QMF35015 xl.5. (1892), from the Middle Devonian of Chudleigh is only 13mm high, has similarly elongated nodes on the spiral cords, but possesses two folds on the inner lip. The species is clearly closely related to both Dirachis and Plagiothrya. 456 MEMOIRS OF THE QUEENSLAND MUSEUM TABLE 4. Shell measurments for Murchisoniidae gen. et sp. indet. Specimen Height (mm) Width (mm) OMF35077 22.7 13.5 OMF35078 25 12 OMF35092 14.9 9.5 OMF35093 20 10 Family NERITOPSIDAE Gray, 1847 Subfamily NATICOPSINAE Miller Naticopsis M’Coy, 1844 Naticopsis (Naticopsis) M’Coy, 1844 Naticopsis (Naticopsis) sp. (Fig. 4H-J) MATERIAL. QMF35I09, 35538 from QML 1037, QMF35013 from QML1035. DESCRIPTION. Small to medium sized, natici- form, rapidly expanding shell, maximum height 22mm, maximum width 17mm. Suture strongly impressed, channelled in largest specimen (QMF35013). Whorls embrace high on whorl face. Whorl profile rounded with midwhorl slightly below periphery. Aperture ovate with vertical axis longer. Growth lines very fine, slightly prosocline. REMARKS. Poor preservation, and the small number of specimens, renders specific diagnosis impossible. The material possesses the charac- teristic globular form and growth lines of the subgenus. Suborder MURCHISONIINA Cox & Knight, 1960 Family MURCHISONIIDAE Koken, 1896 Murchisonia d’Archaic & de Vemeuil, 1841 Murchisonia (Murchisonia) d’Archaic & de Vemeuil, 1841 Murchisonia (Murchisonia) sp. (Fig. 4K) MATERIAL. QMF35015 from QML1037. DESCRIPTION. A very high-spired, narrow, and multi-whorled shell which is partially weathered from limestone. It is 40.7mm high, with a basal width of 13.6mm. At least 9 whorls present but the apex is obscured. Suture impressed and whorls embrace well below mid whorl. Whorl profile angular with mid-whorl peripheral angu- lation bearing a selenizone Hanked by two cord. A very weak spiral thread is present on the lower whorl surface. REMARKS. I am unable to nominate a species based on the single poorly preserved specimen. Murchisonidae gen. et sp. indet. (Fig. 3G, H) MATERIAL. QMF35077, 35078, 35091-93. DESCRIPTION, Medium-sized, anomphalus, high spired, up to 25mm high and 13.5mm wide (Table 4). Suture weakly impressed. Whorl pro- file flattened to very weakly rounded. Weak spiral thread present on the lowermost whorl face, just above the suture. LITERATURE CITED ARCHAIC, E.J.A. D’ & VERNEUIL, E.P. D’ 1841. Note sur Ic genre Murchisonia. Bulletin de la Societe de France 12: 154-160. 1842. On the fossils of the older deposits in the Rhenish Provinces, preceded by a general survey of the fauna of the Palaeozoic rocks, and followed by a tabular list of the organic remains of the Devonian system in Europe. Transactions of the Geological Society of London 6: 303-410. BLODGETT, R.B. 1992. Taxonomy and palaeobio- geographic affinities of an early Middle Devonian (Eifelian) gastropod faunule from the Livengood Quadrangle, east-central Alaska. Palaeontogrpa- hica Abt A. 221:125-168. CLARKE, J.M. 1903. Some new Devonic fossils. New York State Museum Bulletin 107: 153-291. COOK, A.G. 1997. Gastropods from the Burdekin For- mation, Middle Devonian, north Queensland. Memoirs of the Queensland Museum 42( I ) 37-49. DRUCE, E.C. 1969. Devonian and Carboniferous conodonts from the Bonaparte Gulf Basin, north- ern Australia and their use in international corre- lation. Bureau of Mineral Resources, Geology and Geophysics Bulletin 98: 1-243. GRABAU, A.W. & SHIMER, H.W. 1909. North American Index Fossils. (A.G. Seiler: New York). HALL, J. 1879. Geological Survey of the State of New York; Palaeontology Volume 5, Part 2, containing descriptions of the Gasteropoda, Pteropoda and Cephalopoda of the Upper Helderberg, Hamilton. Portage and Chemung Groups. (Charles Van Ben- thuysen & Sons: Albany). HALL, J. & WHITFIELD, R.P. 1877. Paleontology Part II. Pp. 199-302. In King, C., Report of the Geological Exploration of the fortieth parallel V4. Professional Papers of the Engineer Department U.S. Anny No. 18 (Government Printing Office: Washington). KNIGHT, J.B. 1941. Paleozoic gastropod genotypes. Geological Society of America Special Papers 32. 1942. Four new genera of Paleozoic Gastropoda. Journal of Paleontology 16; 487-488. 1945. Some new genera of Paleozoic Gastropoda. Journal of Paleontology 19: 573-587. 1956. New' families of Gastropoda. Washington Academy of Science Journal 46(2): 41-42. GASTROPODS FROM THE BONAPARTE GULF BASIN 457 KNIGHT, J.B., COX. L.R., KEEN, A.M., BATTEN, R.L., YOCHELSON, E.L. & ROBERTSON, R. 1960. Systematic Descriptions. Pp. 1 169-1331. In Moore, R.C. (ed.) Treatise on invertebrate paleon- tology. Part I. Mollusca 1 . (Geological Society of America & University of Kansas Press; Lawrence, Kansas). LINSLEY, R.M. & YOCHELSON, E.L. 1973. De- vonian carrier shells (Euomphalidae) from North America and Germany. United Staes Geoligical Survey Professional Paper 824; !-25. MORY, AJ. & BEERE, G.M. 1988. Geology of the Onshore Bonaparte and Ord Basins in Western Australia. Geological Survey of Western Austra- lia Bulletin 134: 1-184. MORY, A. J. 1990. Bonaparte Basin. Western Australia Geological Survey Memoir 3: 380-415. ROBERTS, J. 1971. Devonian and Carboniferous gas- tropods from the Bonaparte Gulf Basin, north- western Australia. Bureau of Mineral Resources, Geology and Geophysics Bulletin 122. 2. vols. TASSELL, C.B. 1982. Gastropods from the Early De- vonian "Receptaculites" Limestone, Taemas, New South Wales. Records of the Queen Victoria Museum Launceston 77: 1-59. WHIDBORNE, G.F. 1889-92. A monograph of the Devonian fauna of the South of England, Vol. I, the fauna of the limestone of Lummaton, Wolbor- ough, Chircombe Bridge and Chudleigh. (Pa- laeontographical Society: London). 458 MEMOIRS OF THE QUEENSLAND MUSEUM HEIDECKERNEMA GEN. NOV. A REPLACEMENT NAME FOR THE DEVONIAN TL'RRETELLID TRINEMA COOK. Memoirs of the Queensland Museum 42(2): 458. 1998:- Trinema, with type species T. heideckeri Cook, 1997 was erected for a turretellid gastropod from the Middle Devonian Burdekin Formation. This name is invalid as the generic name is occupied by a thecamoeban protistan Trinema Diijardin, 1841. I propose the new name Heideck- ernema to replace Trinema Cook, 1997. Tlie name is in honour of Eric Heidecker. I thank Joan Thorne (Zoological Record) for bringing this oversight to my attention. Literature Cited COOK, A.G. 1997. Gastropods from the Burdekin Formation, Middle Devonian, north Queensland. Memoirs of the Queensland Museum 42:37-49. DUJARDIN, F. 1841. Histoire naturelle des Zoophytes- Infusoires. (Paris) A.G. Cook, Queensland Museum, PO Box 3300, South Bris- bane, Queensland 4101, Australia: 1 February 1998. TWO NEW SPECIES OF RAMPHOTYPHLOPS {Sq\JMAkTK\ TYPHLOPIDAE) FROM QUEENSLAND PATRICK J. COUPER, JEANETTE A. COVACEVICH AND STEPHEN K. WILSON Couper, PJ., Covacevich, J.A. & Wilson, S.K. 1998 06 29: Two new species of Ramphoty- phlops (Squamata: Typhlopidae) from Queensland. Memoirs of the Queensland Museum 42(2): 459-464. Brisbane. ISSN 0079-8835. To Australia’s Ramphotyphlops-r'\e\\ herpetofauna are added R. robertsi sp. nov. and R. aspina sp. nov. The former is known only from liie holotype from Shipton’s Flat, near Cooktown, northeastern Queensland. A midbody scale count of 22 separates it from all but five of its Australian congeners. From these it is easily distinguished by its strongly dichromatic pattern. Ramphotyphlops. aspina sp. nov. is known from the holotype and one paratype from the Barcaldine area, central Queensland. This species is very distinct, lacking the caudal spine which appears to be present in all other Australian Ramphotyphlops spp. □ Squamata, Typhlopidae, Ramphotyphlops, P.d Couper, J.A. Covacevich S.K. Wilson. Brisbane 4101, Australia: 21 November 1998. Australia’s snakes are relatively well-known taxonomically and, to a lesser degree, ecologi- cally. However, one family, Typhlopidae, re- mains poorly-known on both fronts. Relative neglect of Australia’s highly diverse Ramphoty- phlops spp. results from a combination of factors. Many species are small, and similar superficially. Most are grey-brown/brown and have high ven- tral scale counts and head shields which are diffi- cult to delineate. Hence, examination of specimens is time-consuming. Further, speci- mens of Ramphotyphlops are difficult to collect so, for many species, there are only small series in museum collections. Several species are known only from type material. To the already Ramphotyphlops-x'xoh Austra- lian herpetofauna we add two new, highly-dis- tinctive species. Prior to their discovery, 34 species of Ramphotyphlops were recognised from Australia, 16 from Queensland. Most Queensland species were described in the late 19th century in pioneering works by early ‘giants’ of herpetology (e.g., A. Boulenger, J. E. Gray and G. Peters); two date from the only Australia-wide revision of the genus (Waite, 1918); and two have been de- scribed recently (Ingram & Covacevich, 1993). Increasingly, it has become apparent that Austra- lia’s tropics and subtropics harbour many unde- scribed species (e.g., Storr, 1981; Ingram & Covacevich, 1993; Shea & Horner, 1997). One of the new species is unusual in being markedly dichromatic. The other appears to be unique amongst Australian Ramphotyphlops spp. in lacking a caudal spine, a feature regarded as consistently present in the genus (e.g., Cogger, Queensland, Australia. Queensland Museum, PO Box 3300, South 1994; Storr et al., 1986). Like many other Ram- photyphlops spp., both are based on small samples — the first (from near Cooktown, northeastern Queensland [NEQ]) on a single specimen; the second (from the Barcaldine area, central Queensland [CQ]) on two specimens collected close to 80 years apart. While recognising that description of these species from such scant ma- terial precludes an account of variation, it seems reasonable to proceed because many more years may elapse before more specimens of either can be located. The collector of the NEQ specimen is a naturalist who has lived near the collection locality for 40+ years and knows the area inti- mately. That he has seen only one specimen of this species in that time evidences the dilficulty in obtaining additional material. The specimens of these new species are small and slightly dam- aged. To avoid further damage, dissections to determine sex/maturity/state of male genitalia have not been undertaken. We follow Shea & Homer (1997) in assigning these new species to Rampho- typhlops on the basis of geography (Robb, 1 966). MATERIALS AND METHODS Snout-vent length was measured with a stand- ard 30cm ruler. The rostral/head width ratio was calculated from head drawings of the holotypes. All other measurements were taken using Mitu- toyo electronic callipers. The species accounts follow the format of Shea & Homer (1997). The presence of a caudal spine in all currently recognised Australian species (34) except the new species from the Barcaldine area was con- firmed by examination of specimens of the species 460 MEMOIRS OF THE QUEENSLAND MUSEUM FIG. 1. Ramphotyphlops robertsi sp. nov. (Holotype QMJ63736) Above, dorsal view of head. Centre, lateral view. Below^ ventral view. listed in Appendix 1. One type description: R. tovelli (Loveridge, 1945); extrapolation from Cogger (1986), Storr et a!. (1986) and Aplin & Donnellan (1993): R. australis (Gray 1845), R. central is ^iorx^ 1984,7?. endoterus{'^?i\Xt, 1918), R. hamatus StoiT, 1981, R. howi Storr, 1983, R. kimberleyensis Storr, 1981, R. leptosoma Robb, 1972,/?. leucoproctus (fiouXQWgtx, 1889), R. mar- garetae Storr, 1981, R. micromma (Kinghom, 1929), /?. pilbarensis Aplin & Donnellan, 1993, R. pingiiis (Waite, 1 897), R. troglodytes Storr, 1981, M’aitii (Boulenger, 1895) and R. yampi- ensis Storr, 1981. Whether the presence or ab- sence of a caudal spine is related to sexual or ontogenetic differences, or damage, was assessed in a sample of 24 R. nigrescens which contained adults of both sexes, and juveniles. All speci- mens, including those desiccated or damaged, have a prominent caudal spine. Removal of the distal scale containing the caudal spine from one specimen (QMJ57932) required considerable force. Once this was done, the tail of this speci- men was compared with those of the two speci- mens from the Barcaldine area (QMJ7, J51541). The tails of both Barcaldine specimens were in- tact, being capped by a smooth, highly-polished distal scale. By comparison, the tail of OMJ57932 terminated in dull, dermal tissue. Ramphotyphlops robertsi sp. nov. ^(Figs 1 & 2) ETYMOLOGY. Named for Lewis Roberts, eminent naturalist and Honorary Consultant of the Queensland Museum, collector of the holotype. MATERIAL. HOLOTYPE: QMJ63736, Romeo Ck, via Shipton’sFial(15°50’S, 145°14’E) NEQ, 10 Feb, 1983. DIAGNOSIS. Ramphotyphlops robertsi sp. nov. is easily distinguished from its congeners by the following characters combined: strongly dichro- matic pattern (in alcohol, dorsum puiplish brown, venter white); 22 midbody scale rows; 556 ven- trals; a bluntly-rounded snout; rostral circular from above; nasal not completely divided by na- sal cleft. Just visible from above and joining sec- ond supralabial below. DESCRIPTION. Head smoothly-rounded above and in profile; dorsal surface of rostral large and circular (58% head width); ventral lobe of rostral narrower (32% head width), tapering caudally, lateral margins concave; nasals broadly separated by prefrontal; prefrontal slightly larger than fron- tal; supraoculars broadly separated by prefron- tal/frontal junction; nostrils inferior, near apex of snout, equi-distant from rostral and preocular, opening laterally; nasal cleft beginning at second supralabial, extending across rostroventral mar- gin of nostril, then curving towards the rostral and TWO NEW SPECIES OF RAMPHOTYPHLOPS FROM QUEENSLAND 461 FIG. 2. Holotype of Ramphotyphlops robertsi (QMJ63736). terminating 2/3 the distance from nostril to ros- tral, just visible from above; preocular narrower than nasal and subequal to ocular; eye small only just visible, beneath preocular/ocular suture, and immediately below preocular/ocular/supraocular junction; posterior edge of ocular overlaps three equal scales (one parietal and two postoculars); four supralabials — first smallest and overlapped by rostral lobe of nasal; second supralabial larger overlapped by rostral lobe and caudal lobe of nasal and preocular; third supralabial larger, over- lapped by preocular, strongly overlapping ocular; fourth supralabial much the largest, elongate, overlapped by ocular; mental subequal to post- mental; infralabials three — second largest; mi- crotubercles of head shields most dense on nasal scute and lower surface of rostral; glands visible beneath posterior margin of nasals; tail with ter- minal spine. Midbody scale rows 22; ventral scales 556; subcaudal scales 12; SVL 285mm; body width 4.3mm (1.5% SVL); tail length 4.7mm (1.6% SVL); head width 2.9mm (1.0% SVL). Colour in alcohol, strongly dichromatic, dor- sal and ventral colours sharply delineated; dorsum dark purplish brown (11 scale rows); rostral cream; venter uniformly cream (11 scale rows). COMPARISON. Ramphotyphlops robertsi can be confused with only Ramphotyphlops spp. which have 22 midbody scale rows and a nasal cleft beginning at the second supralabial [from Australia - R. australis, R. hamatus, R. kimber- leyensis, R. pilbarensis, R. polygrammicus and R. troglodytes', from PNG/eastem Indonesia - R. de- pressus (Peters, 1880), R. flaviventer (Peters, 1864), R. olivaceus (Gray, 1845) and R. willeyi (Boulenger, 1900)]. It can be separated from the Australian spp. by its strongly dichromatic col- our/pattem. R. robertsi is further separated from R. australis by a considerably higher ventral scale count (556 vs 278-357); from R. hamatus and R. pilbarensis by snout shape in profile (rounded vs angular); from R. kimberleyensis and R. poly- grammicus by the shape of the rostral from above (circular vs elongate) and from R. troglodytes by 462 MEMOIRS OF THE QUEENSLAND MUSEUM FIG. 3. Ramphotyphlops aspina sp. nov. (Holotype QMJ51541). Above, dorsal view of head. Centre, lateral view. Below, ventral view. the nasal cleft (not completely dividing nasal scale vs completely dividing nasal scale). From the PNG/eastern Indonesia spp. (except R. flaviventer) it can also be separated by its strongly dichromatic colour/pattem. From R. flaviventer it is readily distinguished by the shape of the rostral from above (circular vs oval). This character also separates it from R. depressus. R. robertsi is fur- ther separated from these two species by the con- dition of the nasal cleft (visible from above vs not visible from above); from R. olivaceus by its rostral scale (without a transverse keel vs with a transverse keel) and from R. willeyi by its snout profile (smoothly-rounded vs a rounded wedge) (McDowell, 1974; Wallach, 1996). HABITAT. Open forest dominated by Blue Gum, Eucalyptus tereticornis, Brown Bloodwood, E. trachyphloia and Yellow Stringybark, E. acmenoides. REMARKS. The type specimen was found emerging from a fallen Eucalyptus branch which had been hollowed-out by termites, and sub- sequently recolonised by small black ants. Both the snake and the ants were ‘agitated’, suggesting that the branch may have fallen to the ground recently. The snake was seen feeding on ant eggs, which were regurgitated after its capture (L. Roberts, pers. comm.). Ramphotyphlops aspina sp. nov. (Figs 3 & 4) ETYMOLOGY. Latin, a-, without and spina, a spine. TYPE MATERIAL. HOLOTYPE: QMJ51541, Mar- got Stn, 20km N Barcaldine (23®27’S, 145°I6’E) CQ, donated 1990 by Tim Pulsford. PARATYPE: QMJ7, Coreena Stn, Barcaldine (23^1 7'S, 145°24’E) CQ, reg- istered 26 May, 1911, donated W.C. Miller. DIAGNOSIS. Ramphotyphlops aspina lacks a caudal spine. It is further distinguished from other members of the genus by the following characters combined: 1 8 midbody scale rows; 403-428 ven- trals; snout bluntly-rounded from above, rounded and flattened laterally; rostral elongate from above; nasal not completely divided by nasal cleft, clearly visible from above and joining sec- ond supralabial below. DESCRIPTION. Head rounded and slightly-flat- tened in profile; head bluntly rounded dorsally, slightly ‘squared’; rostral large and elongate above (37% head width), slightly tapered caudal ly; ventral lobe of rostral narrower (24% head width) tapering caudally, lateral margins parallel; nasals broadly separated by prefrontal (QMJ5 1 54 1 ) or in point contact (QMJ7); prefron- tal slightly larger than frontal (QMJ51541) or smaller than frontal (QMJ7); supraoculars broadly separated by prefrontal/frontal Junction; TWO NEW SPECIES OF RAMPHOTYPHLOPSmOU QUEENSLAND 463 FIG. 4. Holotype of Ramphotyphlops aspina (QMJ5 1541). nostrils inferior, near apex of snout, equi-distant from rostral and preocular, opening laterally; na- sal cleft beginning at second supralabial, extend- ing across rostroventral margin of nostril, passing dorsally and slightly rostrally to terminate 2/3 to 3/4 the distance from nostril to rostral, clearly visible from above; nasal wider than preocular; preocular about as wide as ocular; eye small and distinct beneath ocular/supraocular junction, im- mediately posterior to preocular/ocular/su- praocular Junction; ocular caudally overlaps two postoculars and a larger parietal; first supralabial smallest, overlapped by rostral lobe of nasal; sec- ond supralabial larger, overlapped by rostral lobe and caudal lobe of nasal and preocular; third supralabial larger, overlapped by preocular but strongly overlapping ocular; fourth supralabial much the largest, elongate, overlapped by ocular; mental narrower than postmental; infralabials three - third the largest; microtubercles of head shields densest on lower surfaces of rostral and nasals; glands not visible below margins of head shields; tail lacking terminal spine. Midbody scale rows 1 8; ventral scales 403-428; subcaudal scales 10-16; SVL 230-275mm; body width 3. 6-3. 7mm (1.3- 1.6% SVL); tail length 1.1 -2.3% SVL; head width .94-1.1% SVL. Measurements of holotype: ventral scales 403; subcaudal scales 16; SVL 230mm; body width 3.8mm; tail 5.5mm; head width 2.6mm. Colour in alcohol, pale pinkish/tan above and below; each scale with a darker edge. COMPARISON. Only Australian species of Ramphotyphlops have 18 midbody scale rows. Ramphotyphlops aspina can be confused with those Australian species having 18 midbody scale rows and a nasal cleft beginning at the second supralabial {R. affinis, R. grypus, R. guentheri, R. howi, R. leptosoma, R. margaretae, R. micromma and R. chamodracaena). It is separated from these species in lacking a caudal spine. R. aspina can be separated further from R. affinis and R. grypus by its snout profile (rounded vs angular); from R. guentheri, R. margaretae and R. chamodracaena by its nasal cleft (clearly visible from above vs 464 MEMOIRS OF THE QUEENSLAND MUSEUM not visible from above) and from R. howi, R. leptosoma and R. micromma by its nasal shield (not completely divided vs completely divided). HABITAT. The only two collection localities for R. aspina are in ‘sheep country’ of Queensland’s central downs. Most of the area in which R, aspina was found has been cleared of open eucalypt forest and replaced by introduced pasture grasses. ACKNOWLEDGEMENTS We are grateful to Paul Homer (Northern Ter- ritory Museum, Darwin) for examining speci- mens in his care; Ross Sadlier (Australian Museum, Sydney) for providing a listing of AM Ramphotyphlops specimens from CQ and NEQ; Glenn Shea (Department of Veterinary Anatomy, University of Sydney) for examining the types of R. minimus and R, yirrikalae\ Tim Pulsford (Queensland Department of Environment) for providing collection details, and Jeff Wright for preparing the photographs. We are indebted to Lewis Roberts for collecting specimen QMJ63736 and bringing it to our attention. LITERATURE CITED APLIN, K.P. & DONNELLAN, S.C. 1993. A new species of blindsnake, genus Ramphotyphlops (Typhlopidae, Squamala), from northwestern Australia, with a redescription of R. hamatus, Storr, 1981. Records of the Western Australian Museum 16(2): 243-256. COGGER. H.G. 1 986. Reptiles and amphibians of Aus- tralia. 4lh ed. (Reed: Sydney). 1994. Reptiles and Amphibians of Australia. 6th ed. (Reed: Sydney). INGRAM. G..I. & COVACEVICH, J.A. 1993. Two new species of striped blindsnakes. Memoirs of the Queensland Museum 34(1): 181-184. LOVERIDGE. A. 1945. A new blind snake (Typhlops tovelli) from Darwin, Australia. Proceedings of the Biological Society of Washington. 58: 1 1 1-112. McDOWELL, S.B. 1974. A catalogue of the snakes of New Guinea and the Solomons, with special ref- erence to those in the Bernice P. Bishop Museum, part 1. Scolecophidia. Journal of Herpetology 8(1): 1 57. ROBB, J. 1 966. The generic status of the Australasian typhlopids (Reptilia: Squamata). Annals & Maga- zine of Natural History Series 13(9): 675-679. SHEA. G.M. & HORNER, P. 1997. A new species of Ramphotyphlops from the Darwin area, with notes on two similar species from northern Australia. The Beagle, Records of the Museums and Art Galleries of the Northern Territory 13; 53-60. STORR, G.M. 1981 . The genus Ramphotyphlops (Ser- pentes: Typhlopidae) from Western Australia. Re- cords of the Western Australian Museum 9(3): 235-271. STORR, G.M., SMITH, L.A. & JOHNSTONE, R.E. 1986. Snakes of Western Australia. (Western Australian Museum: Perth). WAITE, E.R. 1918. Review of the Australian blind snakes (Family Typhlopidae). Records of the South Australian Museum 1(1): 1-34. WALLACH, V. 1996. The systematic status of the Ramphotyphlops flaviventer (Peters) complex (Serpentes: Typhlopidae). Amphibia - Reptilia 17(4): 341-359. APPENDIX 1 Specimens of examined: Queensland Mu- seum: R. qffinis (Gray/l845) - QMJ 11630, 5262, 23870, 44501 ; R. bituberculatus (Peters, 1 863) - QMJ33377, 33378, 33380; R. braminus (Oaudin, 1803) - QMJ32880, 32970, 39690; R. broomi (Bouienger. 1898) - QMJ2954, 20315, 47503; R. chamodracaena Ingram & Covaccvich. 1993 - QMJ28082, 3 1963, 39673, 40233 holotype, 4 1 550, 51980; R. di\ersus (Waite, 1894) - OMJ2943 holotype; R. grypus (Waite, 1918) - QMJ23943, 27510, 39596; R. guemheri (Pe- ters, 1865) -QMJ 2266; R. ligaius (PeXers, 1879)-QMJ57653, 62775; R. nigrescens (Gray, 1845) - QMJ2875, 2884, 3447, 5731, 6636, 7346, 10577, 12348, 20696. 22700, 22706, 35333, 35649, 38527, 43754-55, 43780, 44049, 45255. 46130, 46316, 54538, 54856. 57932: R polygrammicus (Schlegel, 1839) - QMJ5748. 59885, 60625, 61 588; R proxh ww.9(Waite. 1893)-QMJ2936, 10927. 14212,38741:7?. silvia Inaram & Covacevich, 1993 -QMJ8521. 23620, 27387 holo- tvpe, 31576, 31577, 31579. 35872. 43785, 46128. 60852; R. linguirostris (Peters, 1867) - QMJ29747, 44502, 56903, 58642; R. weidii (Peters, 1867) - QMJ22670. 23965, 43373, 47964. Northern Territory Museum: - R. minimus • NTMR7521, 9874. R. nema Shea & Horner 1997 - NTMR16047. 21665 holotype. 34110. Australian Museum: R. minimus’* (Ktnghom, 1929) AMR9692 holotype; R. yirri- kalae (Kinghorn, T942) - AMRI238I holotype. * The holotype of R minimus possesses a caudal spine. However, one of two additional specimens currently assigned to this taxon in the collection of the Northern Territory Museum lacks a spine (P. Homer, pers. comm.). TWO NEW SPECIES OF SAPROSCINCUS (REPTILIA: SCINCIDAE) FROM QUEENSLAND PATRICK J. COUPER AND LAUREN D. KEIM Couper, P.J. & Keim, L.D. 1998 06 29: Two new species of Saproscincus (Reptilia; Scincidae) from Queensland. Memoirs of the Queensland Museum 42(2): 465-473. Brisbane ISSN 0079-8835. Saproscincus basiliscus (Ingram & Rawlinson, 1981), a small skink endemic to the rainforests of eastern Queensland, is a composite of three species. These are readily diagnosed by head scalation (siiboculars; contact between second supralabial and lower preocular), measurements (SVL; length of original tail) and body scale counts (midbody and para- vertebrals). Two species are restricted to northeastern Queensland — S. basiliscus ( 1 6®02’S -19®28’S) and S, lewisi sp. nov. (I5°04’S - I6®08’S). The third, S. hannahae sp. nov., is confined to mideastern Queensland (20°15'S - 2U32'S). Published genetic data show that these distributions are in accord with those of some other vertebrate and invertebrate taxa. □ Saproscincus hannahae, Saproscincus lewisi, Scincidae, rainforests, Queensland. Patrick Couper & Lauren Keim, Queensland Museum, PO Box 3300, South Brisbane 4101, Australia: 24 October 1997. Saproscinctts Wells & Wellington, 1984 occurs in moist forest habitats of eastern Australia (Cog- ger, 1994). It currently contains seven species: S. basiliscus (Ingram & Rawlinson, 1981); S. chal- lengeri (Boulenger, 1887); S. czechurai (Ingram & Rawlinson, 1981); S. mustelinus (O’shaugh- nessy, 1874); 5'. Wells & Wellington, 1985; S. spectabilis (De Vis, 1 888) and S. teiradactylus (Greer & Kluge, 1980). Three of these {S. basilis- cus, S. czechurai and S. tetradactylus) belong to a distinct ‘northern’ lineage defined by the fol- lowing synapomorphies: small size (maximum SVL = 47mm); a peculiar arrangement of nuchal scales; an elevated number of premaxillary teeth (13 or more); and a constant clutch size of two (Greer, 1989). An examination of Saproscincus basiliscus specimens from the collections of the Queens- land, Australian and South Australian Museums shows that this species, as currently recognised, contains three morphologically distinct popula- tions, two of which represent undescribed taxa. The nominate form occurs in northeastern Queensland (NEQ) from Roaring Meg Valley (16°02’S, 145°21’E) to Mt Elliot (19‘'28’S, 146°57’E); one new species occurs from Mt Webb National Park (15°04’S, I45°01E) to the Cape Tribulation area (16°08’S, 145‘=’27’E) NEQ; the other, from Mt Dryander (20°15’S, 148‘^33’E) to East Funnel Ck (2r32’S, 149°09’E) mideastem Queensland (MEQ). MATERIALS AND METHODS All measurements were taken using Mitutoyo electronic callipers. Supraciliaries, supralabials, infralabials, and subdigital lamellae on the 4th toe were counted on both sides of specimens exam- ined. The following mensural characters have been used: snout-vent len^h (SVL); axilla to groin (AG); tail length, vent to tip (TL); forelimb, axilla to tip of longest digit (LI); hindlimb, groin to tip of longest digit (L2); head length, tip of snout to anterior margin of ear (HL); head width, measured level with the posterior margin of the parietals (HW); snout, tip to anterior margin of orbit (S); eye to ear, posterior margin of orbit to mid, anterior margin of ear (EE). Premaxillary teeth were counted for QMJ58017 and 62437. Clutch size was recorded for QMJ25137, 25229, 24648, 53481 & 57899-900. SYSTEMATICS The new species described here can be assigned to the ‘northern’ lineage of Saproscincus by the following synapomorphies: maximum SVL 50mm (see remarks for 5. basiliscus); the postero- lateral edge of each parietal scale is usually bor- dered by a nuchal and two temporal scales (in some cases the posterior temporal scale is fused with the nuchal scale); 13 premaxillary teeth; and a constant clutch size of two (Greer, 1989). They are readily separated from S. tetradactylus by the number of digits on each forelimb (5 vs 4). From S. czechurai they are distinguished by the appear- ance of the breeding males (breeding colours not pronounced vs side of head and neck jet black 466 MEMOIRS OF THE QUEENSLAND MUSEUM FIG. 1 . Saproscincus hannahae sp. nov. (paratype, QMJ639 1 1 ; QM photographic collection, image ref. NU908), Finch Hatton Gorge, EungellaNP, MEQ. (B. Cowell) with some lighter speckling) and the shape of the snout (rounded vs pointed). Saproscincus hannahae sp. nov. (Figs 1 & 2) Lampropholis basiliscus Ingram & Rawlinson, 1981 (part). ETYMOLOGY, Named for Hannah Couper. MATERIAL. HOLOTYPE: QM.I57034, Chelmans Rd, Eungella NP (2r00’S, i48''34T) MEQ. PA- RATYPES: Queensland Museum. QMJ57899-901. Mt Dryander (20°15’SJ48°33'E); QMJ57008-I0, Vine Ck (20°15’S, 148°32’); QMJ35664, 58017, Mt Diyander (20°15’S, 148°33^E); QMJ49759, Conway NP, 20.17,148.45; QMJ49709, Conway NP (20°18’S. 148°46T); QMJ32758, 32760, 32766-71, 32780-83. 32794, Brandy Ck {20°2rS, 148°43’E); 58490, Re- pulse Ck, Conway (20^23 ’S, 147°46'E); QMJ57030, Puritan Bay (20‘^29’S. 148°52’E); QMJ59342-43, 61478-82, 61484-85, 61487-90, Mt Macartney (20‘^50’S, 148°34’E); QMJ49615, 49712, Eungella NP, Palm Valley Track (20°55’S, 148°30’E); QMJ49755, 63911, Eungella NP. Finch Hatton Gorge (20°55’S, 148^30'E); QMJ49756, Eungella NP (20®55’S, 148°30’E); QMJ56277, small knoll immedi- ately WNW ofMt Ossa township (20®55'S, MSMS’E); QMJ53508, Mt Ossa/Ossa Ck, via Mirani (20'’56’S, 148“49'E); QMJ53566-67, 53582-84, 5361 1-12, Cof- fee Ck, Mt Jukes, via Mackay (20°59’S, 148°57’E); QMJ53435-39, St Helens Gap, via Mt Charlton (2r00’S, I48°43’E); QMJ53408-09, 53422-24, Mt Charlton, foothills, via Mt Charlton (2l°0rS, I48°44’E); QMJ5348I. Mt Blackwood NP, via Mackay (21®02^S, 148'’56’E); QMJ32602. 32605, 34000-006, 34036, 34038-40, 34042, 34047. 34062, 34066, 34069, 34092-93, 34095-100, Finch I lattonNP (2U06’S, I48°38’E); QMJ49722, Eungella NP, Palm Valley track (2r09’S. I48®30’E); QMJ51 126, 62770, Eungella NP, Broken R (2riO’S, 148'^30’E); QMJ57067, Teemburra Ck, 8km S. of Finch Hatton (2ri3’S, 148°36’E); QMJ53497, 53601, 53614, 53618-21. E Funnel Ck, I2-15km SW Sarina (2r32’S,149°09’E). Australian Museum. AMR47868, Box Ck, Mt Dryander (20°14’S, 148°3rE); AMR87087-95, 9.15km W of Cathu SF office (20°5 1 ’S, I48°38’); AMR87096, 7.4km E of Rd to Mandalay, via Airlie Beach-Shute Harbor Rd (20°17’S, 148°47’E); AMR87097-I03, St Helen’s Gap, 3.7km N of Mt Charlton by Rd (2r00’S, 148°42’E); AMRl 11572, Broken R, Eungella NP (21®I0’S. 148°30’E). All localities are in MEQ. DIAGNOSIS. Saproscincus hannahae sp. nov. can be confused only with S. lewisi sp. nov. and, to a lesser degree, with S', basiliscus. It does not occur in sympatry with either species. It can be distinguished from the former by its smaller size (max SVL = 38.4mm vs 42.3mm); length of the original tail (mean 130% SVL vs 161% SVL); NEW SAPROSCINCUS FROM QUEENSLAND 467 and the arrangement of the second supralabial and lower preocular scales (overlapping and in con- tact [Fig. 2] vs not in contact, 85% of specimens examined, or only in point contact). From S. basiliscus it is distinguished by its paravertebral count (42-49 vs 50-59); subocular scalation (continuous series of subocular scales in contact with granules of lower eyelid always ab- sent vs only rarely absent (Fig. 3), 3% of speci- mens examined); smaller size (max SVL = 38.4mm vs 49.8mm); length of original tail (mean 130% SVL vs 163% SVL), and midbody scale count (mean 22 vs 26). DESCRIPTION. SVL (mm) 17.7-38.35 (mean = 32.3, N = 109). Proportions, (%SVL): AG = 41.4-56.6 ( mean = 49.1, N= 100); TL = 113.2- 146.6 (mean = 130.0, N = 29); LI = 24.6-33.3 (mean = 28.9, N = 100); L2 = 34.0-45.5 (mean = 39.3, N= 100); HW= 12.8-16.7 (mean = 14.6, N = 100); HL = 20.4-27.0 (mean = 22.5, N = 100); S = 8.2-1 1 . 1 (mean = 9.7, N = 1 00); EE = 6.7-1 0.4 (mean = 8.5, N = 100). Nasals widely spaced, rostral and frontonasal in broad contact; preffon- tals narrowly to moderately separated; maximum length of frontal 1.2- 1.7 x maximum width (mean = 1.5, N = 101); frontal contacting fronto- nasal, prefrontals, first two supraoculars and fron- toparietals; supraoculars 4, second the largest; supraciliaries 6-8 (mean = 7. 1 , N = 205), first the largest; frontoparietals paired (rarely ftised, or partially fused) and distinct from interparietal; enlarged nuchal scales 0-3; loreals 2; preoculars 2, the lower being the largest; presuboculars 2, the first being the largest; supralabials 6-7 (mean = 6.05, N = 212), fourth subocular when six, fifth subocular when seven; infralabials 5-6 (mean = 5.99, N = 21 1); postmental contacting two in- fralabials on each side; palpebral disc moderate, less than half the lower eyelid; ear opening small, round or vertically elliptic. Midbody scale rows 20-24 (mean = 22.3, N = 1 1 4); number of scales in a direct line from mental to anal (inclusive) 48-57 (mean = 52.7, N = 75); para- vertebral scales, from anterior-most nuchal to posterior margin of hindlimb 42-49 (mean = 47.0, N = 1 1 7); lamellae beneath fourth toe 1 6-22 (mean = 1 8.6, N = 1 95), with a medial groove on distal portion. Dorsal surface: fawn to reddish brown. Some scales with dark spots, forming broken, longitu- dinal lines on back. Lateral surface: flanks paler, weakly to heavily marked by dark flecks arranged in longitudinal rows. Sides clearly demarcated from dorsal surface by a prominent dorsolateral zone which runs as a broad brown bar between FIG. 2. Holotype of Saproscincus hannahae sp. nov. (QMJ57034). A, lateral view of head; B, dorsal view of head. the eye and forelimb, then breaks into a row of dark spots along the upper flanks. In some speci- mens, this zone is bordered dorsally by a fine pale stripe. The sides merge evenly with the ventral colours. A distinct light spot is present at the posterior base of the thigh. Ventral surface: white, with dark flecks on throat and tail. Lower surfaces of hands and feet pale brown or whitish. Head: conspicuously marked by dark spots, which begin between the eyes and spread posteriorly to form a ‘V’ shaped marking on the frontoparietal and parietal scales. Lips barred. The measurements and scale counts for the holotype (QMJ57034) are as follows: SVL = 33.68mm, AG = 17.34mm, TL = 49.37mm, LI = 9.39mm, L2 = 13.77mm, HW = 4.94mm, HL = 7.69mm, S = 3.30mm, EE = 2.69mm. Maximum length of frontal 1.4 x maximum width; supra- ciliaries 7; enlarged nuchal scales 2; supralabials 6; infralabials 6; midbody scale rows 22; number of scales in a direct line from mental to anal (inclusive) 54; paravertebral scales 48; lamellae beneath fourth toe 20/18. DISTRIBUTION. Confined to rainforests of the Central Mackay Coast Biogeographic region (as defined by Stanton & Morgan, 1977), The broad distribution for this species extends from Mt Dry- ander, via Proserpine (20®I5’S, 148®33’E) to E Funnel Ck, via Sarina(2 r32’S, 149®09’E) MEQ. 468 MEMOIRS OF THE QUEENSLAND MUSEUM FIG. 3. Variation in the subocular scales of Saproscincus basiliscus. A, a continuous series of enlarged subocular scales, holotype (QMJ34409); B, enlarged subocular scales fail to meet beneath the orbit (QMJ52830X PO = preocular, 4 = 4th supralabial. Disjunct populations occur at Mt Dryander (20°15’S, 148‘^33’E), Conway Ra. (20“27’S, 148°44’E), Clarke Ra. (2nO’S, 148®30’E), MtOssa/ Mt Charlton area (21°00’S, 148‘^43’E), Mt Jukes (2r02’S, 148°57’E), Mt Blackwood (2r02’S, 148°56’E) and in the Connors Ra. (2r32’S, 149®12’E)(Fig. 4). HABITAT. Complex notophyll and mesophyll vine forests. HABITS. A secretive litter-skink which always remains close to cover. This species appears to be most common along creek margins. CONSERVATION STATUS. Specimens of S. hannahae sp. nov. have been collected at more than 20 rainforest sites in coastal MEQ (20°15’S - 21®32’S). This species is common and well represented in existing reserves. REMARKS. Recognition of 5. hannahae sp. nov. is of note zoogeographically. Prior to the separa- tion of S. hannahae from 'S. basiliscus' sensu lato, the latter was the only reptile ‘species’ to span the ‘Burdekin Gap’ — the dry corridor which has separated faunas of NEQ (Wet Tropics Biogeographic Region) and MEQ (Central Mackay Coast Biogeographic Region) rainforest ‘islands’ for between one million years (e.g., three bird spp., Joseph et al., 1993) and, it is estimated, seven million years [e.g., Saltuarius cornntus (Ogilby, 1892V'^5'. salehrosus (Co- vacevich, 1975), C. Schneider, in press]. Thus, description of S. hannahae brings to six [Phyllu- rus isis Couper et al., 1993; Phyllurus nepthys Couper et al., 1 993; Phyllurus ossa Couper et al., 1993; Eulamprus ampins (Covacevich & McDonald, 1980); Eulamprus luteilateralis (Co- vacevich & McDonald, 1980) and S. hannahae], the number of obligatory rainforest reptile species confined to the Central Mackay Coast. Ingram & Rawlinson (1981) recognised differ- ences between the populations of S. basiliscus (as Lampropholis basiliscus) occurring in MEQ and NEQ. While the two populations were similar in colour, pattern and aspects of scalation, the MEQ specimens had a ‘tendency to be slightly smaller NEW SAPROSCINCUS FROM QUEENSLAND 469 FIG. 5. Saproscincus lewisi sp. nov. (holotype, QMJ62440; QM photographic collection, image ref. NQ819), Shipton’s Flat, NEQ. (J. Wright) and “stubbier”, and to have slightly lower mid- body and toe lamellae counts’. They regarded these differences as clinal, rather than warranting specific or subspecific recognition. Conse- quently, specimens of S. hannahae w ere included in the type series of L. basiliscus. The following specimens are paratypes of both species: QMJ32602, 32605, 32758, 32760, 32766-71, 32780-3, 32794, 34000-006, 34036, 34038-40, 34042, 34047, 34062, 34066, 34069, 34092-3, 34095-100. AMR47868, 87087-103. Saproscincus lewisi sp. nov. (Figs 5 & 6) Lampropholis basiliscus Ingram & Rawlinson, 1981 (part). ETYMOLOGY. Named for Lewis Roberts, an Honor- ary Consultant of the Queensland Museum. MATERIAL. HOLOTYPE: QMJ62440 (ethanol pre- served ), Parrot Ck, Shipton ’ s Flat ( 1 5°48 ’ S, 1 45® 1 5 ’ E) NEQ. PARATYPES: Queensland Museum. QMJ32354-55, Mt Webb NP (I5®04'S, 145®07’E); QMJ24648, Helenvale, 13km S, Three Jims Ck, Bloomfield Rd (15°42’S. 145®I7’E); QMJ60889, Big Tableland (15®43’S, 145°17’E); QMJ60887, O’Keefe Ck, Big Tableland (15®42’S, 145”16’E); QMJ24918, Home Rule, Slaty Ck, on banks (15®44’S, 145®18’E); QMJ25 137-38, Home Rule to Intake Falls, Home Rule S of Cooktown (15°44’S, 145®17’E); QMJ25229, 25308, Home Rule (I5°44’S, !45®17’E); QM.T25257, Home Rule, nr Home Rule Camp, 30km S Cooktown (15°44’S, 145®17’E); QMJ25265, Mt Hedley slopes (15®44'S, 145®16’E); QMJ25289, Home Rule, nr, on Track to Granites (15®44’S, 145°18’E); QMJ25204, Granite Ck to Cedar Bay, on track (15®45’S, 145°20’E); QMJ24848. Mt Hartley, nr Home Rule, S of Cooktown (15°46’S, 145®19’E); OMJ17902. 62432-39, 62461, Shipton’s Flat. 38km S Cooktown (15®48’S, 145®! 5’E); QMJ 17903, Shipton’s Flat, 32- 48km S Cooktown (15®48’S, !45°16’E); QMJ27135, 27141, Shipton’s Flat. Site 36 (15®48’S, 145®16’E); QMJ25301, 12 Mile Scrub, 60-121m, Gap Ck (15®50’S, 145®I9’E); QMJ27258 12 Mile Scrub, Gap Ck, ca. 30km S Cooktown (15®5rS, 145®2rE); QMJ39436, Bloomfield (15°57’S, 145°20’E); QMJ49599, Roaring Meg R flats, China Camp, via Bloomfield (I6®02’S, I45®18’E); QMJ41514, Cape Tribulation. 2.5km W (I6®05’S, 145®27’E); QMJ42311, Cape Tribulation (16®05’S, 145®29’E); QMJ49547. Monkhouse Oliver, Mosaic Ck, TR165, Cape Tribulation area (16®I2’S, I45°25’E). Austra- lian Museum. AMR26833, Big Tableland (15®43’S, 145®I7’E); AMR26783, Home Rule, Slaty Ck. on banks ( 1 5®44’S, 1 45® 1 8'E). South Australian Museum. 470 MEMOIRS OF THE QUEENSLAND MUSEUM FIG. 6. Hoiotype of Saproscincus lewisi sp. nov. (QMJ62440). A, lateral view of head; B, dorsal view of head. SAMR9770, Big Tableland (15°43’S, I45‘’17’E). All localities are in NEQ. DIAGNOSIS. Saproscincus lewisi can be con- fused with only S. hannahae sp. nov. and, to a lesser degree, with S. basiliscus. Of these, it is geographically isolated from S. hannahae, and may occur in sympatry with S. basiliscus in the vicinity of Roaring Meg (1 6°02’S, 145®2 1 ’E) and Cape Tribulation (I6®08’S, 145°27’E). It can be distinguished from the former by its larger size (max SVL = 42.3mm vs 38.4mm); length of the original tail (mean 161% SVL vs 130% SVL), and the arrangement of the second supralabial and lower preocular scales (not in contact, 85% of specimens examined [Fig. 6], or only in point contact vs overlapping and in contact). From .S', basiliscus it is distinguished by its smaller size (max SVL 42.3mm vs 49. 8); paraver- tebral count (46-50 vs 50-59); and midbody count (mean 22 vs 26). DESCRIPTION. SVL(mm) 23.48-42.34 (mean = 35.6, N = 39). Proportions, (%SVL): AG = 44.3- 54.2 (mean = 47.6, N = 38); TL = 147.5-173.8 (mean = 160.6, N = 8); LI = 26.7-33.3 (mean = 30.3, N = 38); L2 = 36.5-44.1 (mean = 41 .1, N = 38); HW = 1 1 .8-16.9 (mean = 13.9, N = 37); HL = 20.7-25.5 (mean = 22.7, N = 38); S = 8.4-1 1 .3 (mean = 9.7, N = 38); EE = 7.4-9.7 (mean = 8.3, N = 38). Nasals widely spaced, rostral and fron- tonasal in broad contact; prefrontals narrowly to moderately separated; maximum length of frontal 1.2- 1.6 X maximum width (mean = 1.5, N = 38); frontal contacting frontonasal, prefrontals, first two supraoculars and frontoparietals; supraocu- lars 4, second the largest; supraciliaries 5-8 (mean = 6.8, N = 78), first the largest; frontoparietals paired and distinct from interparietal; enlarged nuchal scales 0-2; loreals 2; preoculars 2, the lower being the largest; presuboculars 2, rarely 1 , the first being the largest (mean = 1.98, N = 78); supralabials 6-7 (mean = 6.1, N = 78), fourth subocular when six, fifth subocular when seven; infralabials 5-6 (mean = 5.9, N = 78); postmental contacting 1-2 infralabials on each side (mean = I. 96, N= 78); palpebral disc moderate, less than half the lower eyelid; ear opening small, round or horizontally elliptic. Midbody scale rows 22-24 (mean = 22.4, N = 39); number of scales in a direct line from mental to anal (inclusive) 52-59 (mean = 53.8, N = 37); para- vertebral scales, from anterior-most nuchal to posterior margin of hindlimb 46-50 (mean = 47.9, N = 39); lamellae beneath fourth toe 1 9-25 (mean = 22.2, N = 7 1 ), with a medial groove on distal portion. Dorsal surface: bronze to mid-brown with a few scattered, dark flecks. Lateral surface; flanks uni- form brown, or marked with dark flecks arranged in longitudinal rows. Sides often sharply demar- cated from dorsal surface by a prominent dorso- lateral zone which runs as a broad brown bar between the eye and forelimb, then continues as a fine line along the upper flanks, or breaks into a row of dark flecks. In some specimens, this zone is bordered dorsally by a fine pale stripe. The sides merge evenly with the ventral colours. A distinct light spot is present at the posterior base of the thigh. Ventral surface: white, with dark flecks on throat and tail, and occasionally on the belly. Lower surfaces of hands and feet pale brown or whitish. Head: marked by dark and pale spots. These sometimes form a weak ‘V’ shaped marking on the frontoparietal and parietal scales. Lips barred. Measurements and scale counts for the hoio- type (QMJ62440) are as follows: SVL = 34.74mm, AG= 15.55mm, TL = 60.15mm, LI = I I. 58mm, L2 = 15.59mm, HW = 4.8lmm, HL = 8.20mm, S = 3.57mm, EE = 3.07. Maximum length of frontal 1.5 x maximum width; supra- ciliaries 7/8; enlarged nuchal scales 1; su- pralabials 6; infralabials 6; midbody scale rows 22; number of scales in a direct line from mental NEW SAPROSCINCUS FROM QUEENSLAND 471 TABLE L A comparison of the measurements and scale counts from De Vis’ (1888) description of Mo~ coa spectabilis with those of each of the putative ‘syntypes’. S. spectabilis S. basiliscus De Vis J244 J255 J 19742 J 19743 Total length fmm) 134 133 88.01 49.5 103.4 Head length (mm) 10.5 10.5 9.4 8.3 9 Head width tmm) 6 6 5.9 5 5.2 Body tmm) 42 39.7 33.7 29.7 32.4 Fore Limb fmm) 10.5 13.3 14.7 11.7 12.8 Hind limb tmm) 18 18.3 19.8 16.1 17.5 Tail tmm) 81.5 82.8 44.91 11.5 62 Suoraciliaries 7 7 7 7 7 Midbodv rows 22 22 28 28 26 Subdigital la- mellae (4th toe) 20-24 23-25 21-23 19-20 20 to anal (inclusive) 52; paravertebral scales 48; lamellae beneath fourth toe 23/25. DISTRIBUTION. From Mt Webb NP (I5®04’S, 145®07’E) to the Cape Tribulation area(16®08’S, 145‘’27’E), NEQ (Fig. 4). HABITAT. Complex notophyll vine forests. HABITS. A secretive litter-skink which always remains close to cover. This species appears to be most common along creek margins. CONSERVATION STATUS. Specimens of S. lewisi sp. nov. have been collected at more than 20 rainforest sites in coastal NEQ (15®04’S - 16®08’S), This species is common and well rep- resented in existing reserves. REMARKS. Saproscincus lewisi sp. nov., ranges from Mt Webb (1 5®04’S, I45'’07’E) to Cape Trib- ulation (Ib'^OS’S, 145®27’E) NEQ, 5. basiliscus from Roaring Meg (16®02’S, 145°2I’E) to Mt Elliot (19“28’S, 146‘^57’E) NEQ, so there is at least a 15km zone of sympatry for these two species. The former is largely ‘northern’ and the latter largely ‘southern’ in relation to what has been termed the ‘Black Mountain Barrier’ (a low, discontinuity in the mesothermal rainforests of the uplands — nearKuranda, 16®49’S, 145®38’E, NEQ), a zone of separation for many obligatory rainforest species of snails, frogs, lizards and birds (Moritz et al., 1996). For the frogs and reptiles at least, this separation is in the order of 2-6 million years, based on genetic studies on: Litoria genimaculata (Horst, 1 883); Liioria nan- notis (Andersson, 1916); Litoria rheocola Liem, 1974; Carphodactylus laevis Giinther, 1897; Saltiiarius cornutus (Ogilby, 1892); and Grjype- toscincus queenslandiae (De Vis, 1890) (Schneider et al., in press). The most likely expla- nation for sympatry of S. lewisi and S. basiliscus is secondary contact, following reconnection of rainforests. Ingram & Rawlinson (1981) included speci- mens ofS. lewisi in the type series of 5*. basiliscus (as L. basiliscus). The following specimens are paratypes of both species: QMJ17902-03, 24648, 24848, 24918, 25137-8, 25204, 25229, 25257, 25265,25289,25301,25308,27135,27141,27258, 32354. AMR26783, 26833. Saproscincus basiliscus (Ingram & Rawlinson, 1981) sensu stricto DIAGNOSIS. With the recognition of S. hanna- hae sp. nov, and 5. lewisi sp. nov., the following changes apply to the description of S. basiliscus: SVL = 20.14-49.79mm (mean = 38.6, N = 257). (% SVL) T = 14 1 .9- 1 86.5 (mean = 1 63 .0, N = 39); L2 = 36.5-48.4 (mean == 42.9, N = 247). Midbody scale rows 22-29 (mean = 25.9, N = 254); Paravertebral scale rows 50-59 (mean = 53.5, N = 255); Subocular scales enlarged and continuous 61% (Fig. 3A); small (distinctly larger than granules of lower eyelid) continuous 36%; enlarged suboculars not forming a continu- ous series beneath eye 3% (N = 259) Fig. 3B. For material examined see Appendix 1. DISTRIBUTION. Broad distribution for this species extends from Roaring Meg (16°02’S, 145®2I’E) to Mt Elliot (1928’S, 14657’E), NEQ (Fig. 4). CONSERVATION STATUS. Throughout most of its range, S. basiliscus is a common species that is well represented in existing reserves. However, its status north of the ‘Black Mountain Barrier’ (16®49’S, 145®38’E) is difficult to assess from existing collections. REMARKS. The definition provided by Greer (1989) for the ‘northern’ lineage of Saproscincus requires the following change to maximum SVL: 50mm (previously 47mm). There has been some confusion regarding the type series of M spectabilis De Vis, 1888. Covacevich (1971) identified four syntypes (QMJ244, 255, 19742-3) for this taxon. Sub- sequent authors followed her interpretation of the type material (Cogger et al., 1983; Wells & Wel- lington, 1985; Sadlier et al., 1993). Sadlier et al. (1993) found the syntypes to be composite, consisting of one Saproscincus galli Wells & Wellington, 1985 (QMJ244) and three 5'. basilis- 472 MEMOIRS OF THE QUEENSLAND MUSEUM TABLE 2. A comparison of the paravertebral scale counts, and the presence (Y) or absence (N) of a continuous series of enlarged subocular scales for the types of L. basilisciis (QMJ3449), S. hannahae (QMJ57034), S. lewisi (QMJ62440) and the putative types of M. spectabilis. * = one side only. Taxa Paravertcbrals Siiboculars Tvnes L basiliscus 55 Y S. hannahae 48 N S. lewis i 48 Y Putative tvnes M spectabilis QMJ244 54 N QMJ255 57 Y QMJ 19742 56 Y OMJI9743 54 Y* cus (QMJ255, 1 9742-3). They concluded that the designation of QMJ 19743 as the iectotype of M spectabilis by Wells & Wellington (1985) meant ‘...that Mocoa spectabilis becomes a senior syno- nym of, and the available name for the taxon previously known as Saproscincus basiliscits\ This interpretation was refuted by Ingram (1994) who argued that De Vis' description of M specta- bilis was based on a single specimen (QMJ244). He concluded that Mocoa spectabilis is a sen- ior synonym of the taxon previously known as Saproscincus galli Wells & Wellington, 1985 ... Also, S. basiliscus (Ingram & Rawlinson, 1981) remains valid’. We concur with Ingram (1994). Re-examination of the putative type/s for M. spectabilis confirms that QK4J244 was, at the very least, paramount in De Vis’ (1888) description (Ta- ble 1). Re-examination of the syntypes of M. specta- bilis (OMJ255, 19742 and 19743) shows that these specimens agree in all respects with S. basiliscus. They can be assigned readily to the nominate form by their paravertebral scale counts and the presence of an enlarged series of subocu- lar scales (Fig 3A, Table 2). The collection local- ity associated with these specimens (Gympie) is obviously in error. ACKNOWLEDGEMENTS This work was funded partly by a grant from the National Rainforest Conservation Programme, ad- ministered by the Qld Dept of Environment and Heritage (now the Qld Dept of Environment). We are grateful for assistance in the preparation of this paper from; the Queensland Museum, Ross Sadlier and Allen Greer (Australian Museum), Adrienne Edwards (South Australian Museum^ Glenn Shea (University of Sydney), Chris Schneider and Michael Cunningham (University of Queens- land), Greg Harold, Steve Wilson, Jeff Wright and Lewis and Charlie Roberts. Special thanks to Jeanette Covacevich, Paul Homer and Ross Sad- lier for helping to improve the manuscript. LITERATURE CITED COGGER, H.G. 1994. Reptiles and Amphibians of Australia. 6th ed. (Reed: Sydney). COGGER. H.G., CAMERON, E.E. & COGGER, H.M. 1983. Amphibia and Reptilia. Zoological Cata- logue of Australia 1: 313pp. COVACEVICH, J. 1971. Amphibian and reptile type specimens in the Queensland Museum. Memoirs of the Queensland Museum 16(1 ): 49-67. DE VIS, C.W. 1 888. A contribution to the herpetology of Queensland. Proceedings of the Linnean Soci- ety of New South Wales 2(2): 81 1-826. GREER, A.E. 1989. fhe Biolog>' and Evolution of Australian Lizards. (Surrey Beatty and Sons Pty Ltd: Chipping Norton, Sydney). INGRAM, G.J. 1994. The holotype of Mocoa specta- bilis De Vis, 1888. Memoirs of the Queensland Museum 35(1): 34. INGRAM, G.J. & RAWLINSON, P. 1981. Five new species of skinks (genus Lampropholis) from Queensland and New South Wales. Memoirs of the Queensland Museum 20(2): 311-317. JOSEPH, L., MORITZ, C. & HUGALL, A. 1993. A mitochondrial DNA perspective on the historical biogeography of midcastern Queensland rainfor- est birds. Memoirs of the Queensland Museum 34(1): 201-214. MORITZ, C., SCHNEIDER, C., CUNNINGHAM, M., HOSKIN, C. & HUGALL A. 1996. Genetic di- versity, conservation value, and evolutionary processes of wet tropics fauna. P. 27. In, Anon, World Heritage Tropical Rainforests Conference - handbook and abstracts. (Wet Tropics Manage- ment Authorit)': Cairns). SADLIER, R. A., COLGAN, D.J. & SHEA. G.M. 1993. Taxonomy and distribution of the scincid lizard Saproscincus challengeri and related species in southeastern Australia. Memoirs of the Queens- land Museum 34(1): 139-158. SCI INEIDER, C.J., CUNNINGHAM, M. & MORITZ, C. (in press). Comparative phylogeography and the h istory of endern ic vertebrates i n the Wet T ropics rainforests of Australia. Molecular Ecology. STANTON, J.P. & MORGAN. G. 1977. The rapid selection and appraisal of key and endangered sites. The Queensland case study. University of New England, School of Natural Resources Re- port, No. PR4. WELLS, R.W. & WELLINGTON, R.C. 1985. A clas- sitlcation of the Amphibia and Reptilia of Aust- ralia. Australian Journal of Herpetology, Supplementary Series no. 1. (Australian Biologi- cal Services: Sydney). NEW SAPROSCINCUS FROM QUEENSLAND 473 APPENDIX 1 Specimens of Saproscincus basiliscus examined in this study (259). Queensland Museum. QMJ5 8441, Roaring Meg Valley (16‘*02’S, !45®2rE); OMJ60732-33, Thornton Peak. N on Crib track from Daintree R. (Ib^Ob’S, 145®20’E); QMJ41727. Ml Hemmanl, 6km SW Cape Tribulation (I6°07’S, 145°25’E); QMJ439I5. Hilda Ck, Thornton Peak, base 145"23’n); QMJ49696 Oliver Ck. nr road crossing (16‘‘I2’S, 145'75’E); QMJ54323, Mt Molloy, 3km N (16“39’S, 145"20’E); QMJ52928, Kuranda (I6°49^S, 145°38’E); QMJ5574L Kuranda (I6®49’S, 145°38’E); QMJ6I037. Saddle Mtn (16‘’49*S, 145“40’E); QMJ39429, Kuranda, 6km SE (16“52’S, I45‘*40’E); QMJ39430-31, 39433, Kuranda, 6km SB (I6'‘52’S, 145”40’E); QMJ51951, NorthSouth Bell Peak Saddle, Malbon Thompson Ra (17°06’S, 145°54’E); QMJ48274-76. DanbullaSF, Kauri Ck. Mt Haig Rd (17°07’S, 145‘*38’H); QMJ48224. Danbulla SF, 1.5km along Kauri Rd {17‘‘08’S, I45°37’H); QMJ12145, Atherton Tbld, Tinaroo Dam (17°10’S, 145“33’E); QMJ49548, Severin, Boar Pocket (!7°irS. I45“40’E); QMJ6I04L Danbulla Sci. Res. (17°12’S. 145'’4’E); QMJ4971I. Gadgarra SF (17“I6’S. 145°4’E); QMJ49733. Gadgarra SF {\7^WS, I45‘’4rE); QMJ12158, L Eacham (I7“17’S, 145"37’E); QMJ49702, 1. Eacham (!7°I7’S, 145°37’E); QMJ49708, L Eacham NP (I7°17’S, 145‘’37’E); QMJ6 1038, Graham Range (1 7° 17’S, I45“58’E);QMJ39867- 68, Bellenden Ker Ra, 0.5km S Cable Tower No 7 (IT^’S, I45°52’E); QMJ45920, Malanda (17*’2rS, I45°36’E); QMJ39866, Bellenden Ker Landing, Russell R at landing (17°22’S, 145°58’E); QMJ45917, Russell R. cave site (I7“22’S, 145“53’E); QMJ6I040, Upper Plalh Rd (17'73’S. i45°28’E); QMJ59161 Topaz, Westcott Rd (I7°24'S, 145°41 ’E); QMJ4793 1-32. 47960, Mt Bartle Frere (17°24^S, I45“49’E); QMJ49700, Russell R. headquarters (17“24’S, 145"46’E); QMJ 12146-47, 49704, 49710, Crater, Atherton Tbld (I7°26’S, 145'^9'E): OMJ49703, 49706. Boonjee. 6.5km ESE Lammers Hill (I7“26’S, 145M6’E); QMJ59i2L Hughes Road, Topaz ( 1 7‘’26'S, 1 45°42’E); QMJ59659 Tower nr Crater NP (I7°27’S, 145'^29'E); QMJ308I0. Bartle Frere (17°27’S, I45°53’E); QMJ59649, Polly Ck (Hasenpusch) {17‘^28’S, 146"0rH); QMJ25450. Flying Fish Pt (17"3’S, 146°05’E); OMJ5892L nr Kennedy Hwy crossing of upper Barron R, 25km S of Atherton (17^32'S, l45‘"3rH); QMJ17435,Innisfail(17'’32’S, 146‘’0rE);QMJ61 058-59, Ml Fisher (Kjelberg Rd) ( 1 7‘’32\S, 1 45°33'E); OMJ4391 3, Innis- fail, 25km W (17®37’S. 145‘’48’H); QMi49697, Millstrcam NP (I7°39'S, 145‘‘27’E); QMJ34408-I0, 47637, Charmillan Ck,viaRavenshoe(I7“43*S. 145^*3 rE);QMJ481 73-74, Wal- ter Hill Ra, Charappa Ck drainage. Suttees Rd (17°43’S. 145°4rE); QMJ60892 Tully Valley, frog site (17“45’S, 145°35’E); QMJ48246-47, Forestry 'H’ Rd, via Tully (17M5’S, 145^39’E); OMJII161, 12159. Tully Falls (17®46’S. 145‘’34'E); QMJ48209, Billy Ck Bridge SF 758. vicinity of bridge (17°49’S, I45M7’E); QMJ30811, Mission Beach (\r52% I46®06’E); QMJ514I4, Kirrama Ra, BryceHcnry Logging Area (I8”01\S, 145^36’E); QMJ48373, Kirrama SF, Jennings Logging Area (ISW’S, 145°37’E); OMJ48369, Kirrama Ra, 2km N Kirrama Forest Stn (I8°rS, I45®43’E);OMJ48340, Kirrama Ra, Douglas Ck (i8°13 S, I45°48’E); QMJ48348-49, Kirrama Ra. Kennedy Falls Logging road(18M3’S, 145‘'48’E);QMJ48362, Kirrama SF, Yaccabine Ck (18°13'S. I45“46’H); QMJ25022-59. 25061, Herbert Gorge (I8°i4’S, I45“32’E); OMJ25825-75, Cardwell (18®I6’S, 146“01’H); QMJ45866, Hinchinbrook Is, Scraggy Pi (18°17’S, Mb^Ob’E); QMJ48303, Mt MacAlistcr, Cardwell Ra (18“18’S, I45‘*56’E); OMJ483I2. Cardwell Ra, Upp Broadwater Ck, Valley 145®56’E); QMJ26375-76. Hinchinbrook Is, Ramsay Bay (I8‘’19’S, I46®18’E); QMJ26330-32, Hinchinbrook Is (18®22’S. Hb^’lS’E); QMJ44I62, 44165, 44168, 44195-97. 44220, 44233. Hinchinbrook Is. Upp Gayundah Ck (18'’22,!46‘’15’E); QMJ44169-72, 44190. 44200-13. 44216, Hinchinbrook Is. Gayundah Ck (18‘72’S, Ub^D'E); QMJ5I409-10, Mt McAlister. 11km S (18®23’S, I45°56’E); QMJ48293, Mt Graham, 8km N Abergowrie (I8“24’S, I45‘’52’E): QMJ5 141 1-13, Cardwell Ra, Dalrymple Track {I8“24^S, 146“04’E); OMJ45545, 45548, Mt Diamantina (!8‘*25'S, 146“I7’E); OMJ61304, Sword Ck catchment (I8“27'30”’S, I45‘’42’30"E); QMJ51415, Seaview Ra, Istck past GarawaltCk Rd Xing (18‘’43’S, I45M4'E); OMJ497I5, 49719-20. 49723, 49725-27, 49729-31 Mt Spec (I8“57’S. 146“! LE); QMJ497I6, Mt Spec, Little Crystal Ck (18*57’S, 146”irE); QMJ29668, Mt Spec, Paluma (19“00’S I46°12'E); QMJ49718. 49724. Paluma (I OW’S, 146®I2’E); QMJ52995, 52997, Mt Halifax, summit ridge (I9°06’S, I46‘»22'E); QMJ53032, Mt Halifax, 250m SE (19"06’S. 146‘^22’n); QMJ53039-40, Mt Halifax (I9®06’S, I46°22’E); QMJ5304L Ml Halifax. 300m SE (19°06'S, 146'72’E); OMJ52996, Mt Halifax, 2.5km SE summit (I9®07’S, 146®2rE); QMJ53034. Ml Halifax. 400m SE (!9“07’S 146'’22’E); QMJ53035-36, Mt Halifax. 900m SE (19‘‘07’S. 146®22'E); OMJ46774-75, Bluewaler Ra, N of Townsville (19°irS, Ub^JJ'E); OMJ52828, 52831-33, Mt Elliot (I9°28’S, 146“57’E); QMJ52830, Mt Elliot, 2km NNW of peak (I9'’28’S. I46°57’E): QMJ528I4, 52835, 52837, Mt Elliot. 2km NW of peak (19”28’S, 146‘’57’E); QMj52836. Mt Elliot, 4km NNW of peak (I9°28’S. I46®57'E) OMJ59880 Cannabullen PIcateau. Australian Museum. AMR56563-64, 56571, 56573 southern base of Thornton Peak (I6“I0’S, 145"23’E); AMR56589, 57129, Hilda Ck, southern base of Thornton Peak (Ib^IO’S, I45®23'E); AMR59329 Thornton Peak ( 1 6“ 1 0’ S, 1 45®23 . AMR87080-8 1 , on ridge running E-W between Thornton Peak and Daintree R (16°09’S, 145®21E); AMRI20420-22, Australian Museum- Queensland Museum rainforest survey site nr Thornton Peak (Ib'^IO’S, 145®23’E). South Australian Museum. SAMR22390, Kuranda (16“49’S 145®38’E). 474 MEMOIRS OF THE QUEENSLAND MUSEUM A FURTHER DIAGNOSTIC CHARACTER AND SOME BASIC BIOLOGICAL INFORMATION FOR LERISTA COLLI VERl. Memoirs of the Queensland Mu- seum 42(2): 474. }998:-\'\\e. limb reduced, Queensland skink Lerista colliveri was recently recognised as distinct from L. allanae (Couper & Ingram, 1992; see also Shea, 1993 and Covacevich ct al., 1996) on the basis of two characters, one morphological and one colour. Morphologically, L. colliveri differs from L. allanae in having the front limb represented by a nubbin versus, at most, a shallow depression. In colour pattern, L colliveri differs from L. allanae in having continu- ous, dark longitudinal lines dorsally and laterally, and ventral scales with or witliout dark flecks versus longitudinal lines of dark spots dorsally and laterally, and ventral scales dark edged. Recent work has revealed a second morphological character distinguishing the two species, about which there had been some residual taxonomic concern (Covacevich et af , 1996), and provides an opportunity to make some general observations on L. colliveri, the better known of the two species. We report this new information in this note. Speci- mens examined in addition to those reported in Couper & Ingram, 1992 are as follows: L. colliveri - QMJ59904-05, 61265-66, 61270. and L allanae - SAMR2823 (Shea, 1993). The new diagnostic character for the two species oi Lerista is the number of infralabials contacted by the postmental: one (bilaterally in 31 specimens) in L. colliveri vs usually two (bilaterally in 8 specimens; 1, bilaterally in 1 specimen - QMJ 12232) in L. allanae. The two species also differ somewhat in the way they appear to fuse supraciliaries. Both species have six as the maximum number of supraciliaries. but when there are five continuous supraciliaries (i.e., no fusions with the supraocu- lars), this is achieved tlirough the fusion of supraciliaries one and two (unilaterally, AMRI 13538), two and tliree (unilater- ally, AMRl 13526), four and five (unilaterally, AMRI 13516), or five and six (bilaterally, AMRI 13532) in L colliveri (totd of five cases in 25 specimens) and through the fusion of supraciliaries one and tw'o in L. allanae (all five cases of supraciliary fusion in the seven specimens in which the su- praciliary area is undamaged, ()MJ6040, 6238, 6429-30, 12232). Thus L colliveri appears to be much more variable in the way it fuses supraciliaries than does L. allanae. L. co/Z/veW has seven premaxillar>' teeth (n=25; total), 9-12 maxillary teeth (mean = 10.2, n = 25; left side), and 12-15 dentary teeth (mean = 13.2, n = 25; left side). There is a significant positive correlation between head length (tip of snout to centre of external car opening), and the number of maxillaiy' teeth (r = 0.41, P = 0.049) and dentary teeth (r = 0.72, P < 0.001), and between snout-vent length and the number of maxillary teeth (r = 0.41, P = 0.045) and dentary' teeth(r = 0.66.P <0.001). In L. colliveri, the least squares regression of log head length (tip of snout to centre of external ear opening) on log snout-vent length has the equation log head length = 0.63 log snout-vent length -0.29 (r^ = 0.96. n = 15, P < 0.001) for males, and log head length = 0.59 log snout-vent length -0.24 (r^ = 0.97. n = 10, P < 0.00 1) for females. The 95 percent confidence interval (±0.08) for the slope of the regression for each sex (0.63 and 0.59, respectively) is well below the isometric value of 1.00, indicating that head length becomes proportionately shorter as snout-vent length increases. The slopes of the two regression lines are not significantly dift'ercnl (F = 0.55, df = I, 21, P = 0.46). However, the elevations arc significantly different (F = 1 1.22, df = 1, 22. P = 0.003), indicating that males have relatively longer head lengths than females. In L colliveri, the least squares regression of log rear limb length on log snout-vent length has the equation log rear limb length = 0.57 log snout-vent length -0.32 (r^ = 0.86, P < 0.00 1 ) for males, and log rear limb length = 0.43 log snout-vent length -0.08 (r^ = 0.80, P <0.001) for females. The 95 percent confidence interval (± 0.14 and ±0.17, respectively) for the slopes of the regression (0.57 and 0.43, respectively) for males and females fall well below the isometric value of 1.00, indicating that rear limb length becomes proportionately shorter as snout-vent length increases. The slopes of the two regression lines are not significantly different (F = 2.25, df = 1,21. P = 0.15). However, the elevations are significantly different (F = 5.58, df= 1, 22, P = 0.03), indicating that males have relatively longer rear limbs than females. In L. colliveri, the least squares regression of log hind limb length on log head length for both sexes combined has the equation log hind limb length = 0.83 log head length + 0.01 ( r^ = 0.84, P < 0.001). The 95 percent confidence interval (± 0.15) for the slope of the regression (0.83) falls below tlie isometric value of 1.00, indicating that hind limb length becomes proportionately shorter as head length increases, as it does as snout-vent length incretises (above). The sexes were combined because the slopes and the elevations were not significantly different between tlie sexes (F = 0.87, df = 1, 21, P = 0.36, and F = 1 .08, df = 1 , 22, P = 0.3 1 , respectively). Data on clutch size is available for four specimens of L. colliveri. 'fhree specimens (AMRI 13527, 113533, 113538) collected in the vicinity of Red Falls, the Basalt Wall, N of Charters Towers (latitude; i9®56’S) in the period 10-12Sept., 1984 contain a single yolking follicle in each ovaiy , and one specimen (QMJ33I28) collected at Battery Stn (latitude; 19®26’S)on 12July, 1981 contains a single shelled egg in each oviduct. These data suggest that the species is oviparous with a clutch size of two, and that egg laying may occur during the tropical dry season (QMJ33128). Acknowledgements We thank G. Shea for critically reviewing the manuscript. Literature Cited COUPER, P.J. & INGRAM, G.J. 1992. A new species of skink of Lerista from Queensland and a re-appraisal of L allanae (Longman). Memoirs of the Queensland Museum 32(1): 55-59. COVACEVICH, J.A., COUPER, PJ. & MCDONALD, K R. 1996. Lerista allanae (Scincidae: Lygosominae); 60 years from exhibition to extinction? Memoirs of the Queensland Museum 39(2): 247-256. SHEA, G.M. 1993. New record of Lerista allanae (Squamata:Scincidac). Memoirs of the Queensland Museum 33(1): 220. A.E. Greer, The Australian Museum, 6 College St. Sydney 2000 & P.J. Couper. Queensland Museum, PO Box 3300, South Brisbane 4101, Australia; 27 January 1998 . REPTILE DIVERSITY AT RISK IN THE BRIGALOW BELT, QUEENSLAND J.A. COVACEVICH, P.J. COUPER AND K.R. McDONALD Covacevich, J.A., Couper, P.J. & McDonald, K.R. 1998 06 29: Reptile diversity at risk in the Brigalow Belt, Queensland. Memoirs of the Queensland Museum. 42(2): 475-486 Brisbane. ISSN 0079-8835. Much of the extensive Brigalow Belt (BB) of Queensland has been cleared, primarily for grazing and agriculture. As a result, the conservation status of the region’s natural ecosystems is poor: 65 of 163 ecosystems of the region are classed ‘endangered’ or ‘of concern’. Reptile diversity and species endemism in the region are high, but many species appear to be ‘at risk’, presumably from clearing. One hundred and forty-eight species of reptiles occur in the BB, 13 of these are confined or virtually confined to the region; and, for a further 14 species, the BB is a significant portion of their ranges in Queensland. Type localities of 49 nominate reptile species are in the BB of Queensland. (These, vital in species delineation and attempts to maintain diversity, have been overlooked in management). At only 10 of 28 discrete type localities for reptiles in the region are there still good stands of native vegetation in which a full complement of reptile species might flourish. Under lUCN Species Survival Commission definitions, one species {Lerista allanae) is critically endangered; one {Anomalopus mackayi) is endangered; eleven {Rheodytes leukops, L vittata. Diplodactylus taenicauda, Delma lahialis, Paradelma orientalis, Anomalopus hrevicollis. A. leuckartii, Denisonia maculata. Purina barnardi. F. dunmalli Hemiaspis damelii) are vulnerable; one {Acanthophis antarcticus) is of lower risk; and four (Delma plebeia, Egernia rugosa, Lerista cinerea, Menetia sadlieri) are data deficient. Further, five species appear to be at risk on a local, BB scale: Chlamydosaurus kingii, Tympanocryptis lineata, T. pinguicolla, Tiliqua rugosa and Aspidites ramsayi The BB is an example of how a region should not be managed to maintain biodiversity. Extensive clearing of native forests, overgrazing, intensification of agriculture and proposed significant water infrastructure for further development raise serious concern about mainte- nance of biodiversity in this region, unless integrated nature conservation strategies are enacted urgently. □ Brigalow Belt, reptiles, biodiversity, conservation. Queensland. J.A. Covacevich. P.J. Couper, Queensland Museum, PO Box 3300, South Brisbane 4101, Australia: K.R. McDonald. Department of Environment, PO Box 834, Atherton 4883, Australia: 18 November 1997. Clearing of brigalow lands for crops and pasture has reaped great economic benefits but it has also had severe negative impact on native ecosystems ... flora and fauna...’ Johnson (1 996). The Brigalow Belt (the Northern and Southern Brigalow Regions) encompasses about 20% of Queensland, roughly between the 500 and 750mm isohyets, in tropical (south of Towns- ville) and subtropical Queensland (Fig. 1) and northern New South Wales (Stanton & Morgan, 1977; Thackway & Cresswell, 1995). Prior to European settlement, the Brigalow Belt (BB) was a complex mosaic of woodlands (e.g., dominated by Eucalyptus crebra and E. drepano- phylla; E. populma and E. brownii; E. melano- phloia; E. coolabah, E. tereticornis and E. tessellaris); forests (e.g., of E. tereticornis and E. camaldulensis); shrubby open forests (e.g., domi- nated by Acacia harpophylla., sometimes mixed with E. cambageana, Cadellia pentasyeis, A. cambagei, A. shirleyn or A. rhodoxylon\ diverse semideciduous vine thickets; and open grasslands (e.g., of Dichanthium serviceum). The BB supports a highly diverse vegetation, classified in 163 regional ecosystems of 36 vege- tation provinces (P. Sattler, pers. comm.). However, because of long European settlement Q^ix, 1994) and concomitant large-scale clearing of native vegetation for grazing, farming and mining, rep- resentation of its habitat diversity, particularly on fertile soils, in conservation reserves is poor. Some conservation reserves (national parks, conservation parks and resource reserves) have been set aside for the protection of plant/animal diversity of the region as well as spectacular scenic features like Carnarvon Gorge. These reserves range in size from less than 2ha to nearly 300,000ha and account for 2.07% of the BB, most of which is associated with rugged sandstone areas. Other native forests are partly protected in 476 MEMOIRS OF THE QUEENSLAND MUSEUM FIG. 1 . Main towns/cities and river systems of Queensland’s Brigalow Belt. State Forests (2,622,239ha) and Timber Reserves (140,472ha). Many BB habitats are now at risk and are of con- cern to land managers, wild- life/diversity managers and conservationists. The numbers of regional ecosystems classi- fied as endangered (32) and of concern (33) (P. Sattler, pers. comm.) highlights the diffi- culty confronting attempts to maintain the region’s biodiver- sity. /4. /7a/*o/?/rv//a communities, semideciduous vine thickets and native grasslands, particularly, have been extensively modi- fied and are now extremely poorly protected in reserves, largely because they occur/oc- curred on the richer soils. This problem has been highlighted many times (e.g., Sattler & Webster, 1984; Gasteen, 1985, 1987; Sattler, 1986, 1993; Davie etal., 1994; and Johnson, 1996). In a context of widespread re- placement of native vegetation of the BB with crops, introduced pastures and accompanying plant and animal pests, it is not surprising that the native verte- brates, especially species re- stricted to the region, would be in decline. Extinct bird and mammal species once occur- ring solely or mainly in the BB include: Psephotus pulcher- rimus, Notomys mordax and Conilurns albipes\ species extinct from the BB but still extant be- yond this region include Pseudomys australis and Macrotis lagotis\ species whose populations have declined significantly and are now threatened in- clude Neochina ruficauda ruficauda, Erythrotri- orchis radiatus, Lasiorhinus krefftii and Ony’chogalea fraenata (Gordon, 1984; Garnett, 1992; Lee, 1995 and Maxwell et al., 1996). Reptiles of the BB have received little attention until recently (e.g., Gordon, 1984; Covacevich et al., 1988; Czechura& Covacevich, 1985; Cogger et al, 1993; McDonald et al, 1991). Significant collections of reptile specimens from the region have been deposited in museums (notably the Queensland and Australian Museums [Brisbane, Sydney, Australia] and the California Academy of Sciences in San Francisco, USA). Some gen- eral surveys of areas in the BB have included reptiles (e.g., Kirkpatrick, 1968; James, 1974; Crossman & Reimer, 1986; Russell et al, 1992; and Horsup et al, 1993) and the conservation status of some reptile species from the area has been assessed (Gordon, 1984; Covacevich et al, 1988; Czechura & Covacevich, 1985; McDonald et al, 1991; Cogger et al, 1993; Covacevich & Couper, 1996; and Covacevich etal, 1996a) but no methodical review of the distributions of ail BB reptile species and their conservation status has been pubished. In 1994, financial support to review museum and literature records and conduct field work to refine knowledge of the distribution and conser- vation status of rare or threatened BB reptile REPTILE DIVERSITY IN THE BRIGALOW BELT 477 species was received from the Australian Nature Conservation Agency (now Environment Austra- lia). Lerista allanae (Longman, 1937), a species believed to be very rare, received special atten- tion. All specimen records in the Australian Mu- seum, the Queensland Museum and the California Academy of Sciences and all relevant records in the few BB surveys which have included reptiles were examined. Field work was undertaken in 1995 and 1996, in areas selected in an attempt to add to or clarify existing records: Emerald- Capella-CIermont district, central BB, 1 3-24 Feb- ruary, 1995 (JAC, PJC); Inglewood-Tara-Yuleba district, southern BB, 13-27 March, 1995 (JAC, PJC); Emerald-Clermont-Collinsville district, cen- tral and northern BB, 20-29 April, 1 995 (JAC, PJC, KRMcD); Capella district, central BB, 9-16 Feb- ruary, 1996 (PJC, JAC). In all areas, standard hand-collecting, spot-lighting, and pit- trapping methods were employed. Pit-trapping has been shown to be a valuable sampling method for small species (e.g., Hannah & Smith, 1995; Hobbs et al., 1994). Voucher specimens from the field work have been added to the Queensland Mu- seum reference collection. They bear registration numbers QMJ59368-424, QMJ595 16-576, QMJ59781-870,QMJ61407-423 and QMJ6 1426-461. Major findings (published) of this study are: maintenance of reptile diversity in the BB appears to be unlikely in the face of further clearing of native vegetation (Covacevich, 1996); Lerista al- lanae is at least critically endangered and may be extinct (Covacevich et al., 1 996a); Aspidites raw- ^^^/(Macleay, 1 882) is vulnerable in the southern portion of the BB (Covacevich & Couper, 1996); and, despite extensive clearing in the Emerald district, a highly diverse array of small reptiles survives in a disturbed Callitris forest close to Emerald (Deer, 1996). Further, incidental obser- vations made during the study have enhanced knowledge of Nephrurus asper, Gehyra dubia and Vermicella annulata (Couper, 1996; Couper et al., 1995; and Couper et al., 1996) and of the previously unknown donor of the holotype of Lerista allanae, Maida Allan (Covacevich & Couper, 1997). REPTILE DIVERSITY OF QUEENSLAND'S BRIGALOW BELT One hundred and forty-eight species of reptiles are known from Queensland’s BB (Appendix 1). The occurrence of these species has been mapped from collection/sight localities which have been listed in an unpublished report lodged with Envi- ronment Australia (Canberra) and in the Library, Queensland Museum (Brisbane), accession num- ber 97/22378, Covacevich et al., 1996b. The ranges of 13 species are confined to, or virtually confined to the BB: Rheodytes leukops Legler & Cann, 1980; Diplodactylus taenicauda De Vis, 1886 (Fig. 2); Paradelma orientalis (Gunther, 1876); Anomalopus brevicollis Greer & Cogger, 1985; Anomalopus mackayi Greer & Cogger, 1985; Ctenotus ingrami Czechura & Wombey, 1982; Lampropholis mirabilis Ingram & Rawlinson, 1981; Lerista allanae (Longman, 1937); Lerista vittata Greer, McDonald & Lawrie, 1983; Menetia sadlieri Greer, 1991; Denisonia maculata (Steindachner, 1867); Pu- rina dunmalli (Worrell, 1955); Hemiaspis da- melii (Gunther, 1876). For a ftirther 14 species, the BB is a significant part of their ranges in Queensland: Chelodina expansa Gray, 1867; Gehyra catenaia Low, 1979; Oedura monilis De Vis, 1888; Saltuarius salebrosus (Covac- evich, 1975); Anomalopus leuckartii (Weinland, 1862); Egernia modesta Storr, 1968; E. rugosa DeVis, \ %%%\ Eulamprus sokosomaGveex, 1992; Glaphyromorphus punctulatus (Peters, 1871); Lerista fragilis (Gtinther, 1 876); Menetia timlowi Ingram, 1977; Hoplocephalus bitorquatus (Jan, 1859); Pseudechis guttatus De Vis, 1905; and Suta dwyeri (Worrell, 1956) =S. spectabilis dwyeri (Worrell, 1956). CONSERVATION STATUS Given that much of the BB has been cleared, that 40% of its regional ecosystems are classed endangered or of concern, that several bird and mammal species formerly from the area are ex- tinct and that populations of many bird and mam- mal species of the region have declined seriously, it is predictable that reptile species of the BB would be at risk. Species confined to or occurring predominantly in the region are of most concern. Prior to this study, three authorities had afforded special conservation status to reptile species from the BB in Queensland ( McDonald et al., 1991; Cogger et al., 1993; Queensland Nature Conser- vation (Wildlife) Regulation 1994). Table 1 sum- marises the status of rare and threatened species of reptiles of this region according to these authors. In November, 1994, the 40th meeting of the lUCN Council of the lUCN Species Survival Commission prepared L.. new definitions for Red List categories ...’ of threatened species. In the light of these, following our field work and review of reptile specimen records in museums and rele- vant literature, the current conservation status of 478 MEMOIRS OF THE QUEENSLAND MUSEUM FIG. 2. One of 15 reptile species currently at risk in Queensland’s Brigalow Belt. Diplodactylus taeni- cauda is vulnerable. Specimen collected (and re- leased) near Yuleba, SCQ, 21 March. 1995, during extensive reptile surveys in the region. (J. Wright, Queensland Museum) species of special concern at the national/intema- tional level can be assessed. Under the lUCN (1994) definitions one species is critically endan- gered; one is endangered; 10 are vulnerable; two are at lower risk, near threatened; and four are data deficient (Table 2). Under the lUCN definitions, one BB reptile species previously classed rare or threatened (McDonald el aT, 1991; Cogger et ah, 1993; Nature Conservation (Wildlife) Regulation, 1994) does not qualify for special conservation status. Chelodina expansa of rivers and lagoons in and beyond the BB is apparently not presently at risk. Another species {Lampropholis mirabi- lis), formerly classed rare and threatened (McDonald et ah, 1991; Cogger et ah, 1993; Nature Conservation (Wildlife) Regulation, 1994) warrants discussion. L. mirabilis clearly is a vulnerable (VU D2) species: population very small or restricted ... acute restriction in its area of occupancy ... thus prone to the effects of human activities (or stochastic events) This species is known from only three localities (Magnetic h, Cape Cleveland and Mt Elliot, all near Towns- ville, NEQ, in the north-east part of the BB over an area considerably smaller than lOOkm^. The full known range of this species is secure in national parks. Thus, given that this species is not prone to any identifiable threats, we do not regard it as being vulnerable. A further 5 reptile species appear to be at risk in the BB: Chlamydosaunis kingii Gray, 1825 is primarily a species of northern and coastal Queensland. Two museum specimens (QMJI5592 - 97km E of Clermont, 22°48’S I48°3UE; QMJ33566 - Coorara stockroute, ca 55km ENE of Emerald, 23®05’S 148®25’E) attest to the occurrence of this species in the BB. These were registered in 1968 and 1990 respectively. Reliable, more recent sight records have been made in the BB: 20 Feb., 1992, Epping Forest 22®19’S 146®45’E, (G. Porter, Department of En- vironment, pers. comm.); 13 Feb., 1995 Taunton, 23®30’S 149°15’E (G. Porter, pers. comm.); Feb. and Oct., 1991 Wonga Hills, 26°06’S 150*’45’E (Russell et al., 1992). C. kingii is a species well- known and frequently-reported elsewhere. Re- cords of this species in the BB appear both to be and to have been scant, and most of the few localities from which it is known have been modi- fied for farming or grazing. Epping Forest and Taunton, where specimens of C. A/>?g/7 were seen in 1992 and 1995, are small national parks. This affords some hope for the survival of this species in the BB. Tympanocryptis lineata Peters, 1863 was classed by Gordon (1984) in the first review of fauna (^mammals, primarily) of the BB as ‘a lizard of the blue grass downs of central Queensland’. ‘Natural grasslands, downs...’ he noted ‘... were greatly favoured by early settlers and have been subject to heavy grazing since first settlement...’. Since these surveys of the 1970s and the Gordon (1984) review, much of the area near Emerald where T. lineata occurred has been developed for agriculture. Museum specimens of T. lineata have not been collected recently: 2 April, 1976, Emerald Downs 23°29’S 148‘^08’E (QMJ3685 1); 4 Jan., 1976, Selma 23°3rS 148^02’E (QMJ36849); 5 Oct., 1976, 22 Sept., 1976, 5 Oct, 1976, Iona 23°32’S 147°57’E (QMJ36853, QMJ36848, QMJ36852); Nov., 1977, nr Emerald 23^32’S 148"I5’E (QMJ36847);10, 19, 22 Aug., 1929, 9 Sept., 1929 Retro 20m W of Capella 22®52’S 147°54’E (Cal. Acad. 77623-77626). REPTILE DIVERSITY IN THE BRIGALOW BELT 479 TABLE 1 . Rare and threatened species of reptiles from Queensland’s BB according to three recent authorities. Species Authority McDonald et al., 1991 Cogger et al., 1993 Nature Conserv- ation (Wildlife) Regulation 1994 Chelodina expansa — rare or insufficiently known — Emvdura macQuarii — rare or insufficiently known _ Rheodvtes leukops very restricted, poorly known vulnerable vulnerable Diplodacwlus faenicauda — rare or insufficiently known rare Delma labialis very restricted, rare vulnerable vulnerable Paradelma orientalis restricted to specific habitats, vulnerable vulnerable vulnerable Anomalopus brevicollis restricted to specific habitats, poorly known, occurring in reserves — rare Anomalopus mackavi restricted to specific habitats, poorly known vulnerable vulnerable Ctenotus eiitaenius restricted to specific habitats, poorly known, occurring in reserves — — Esernia rusosa restricted to specific habitats, rare rare or insufficiently known vulnerable Lampropholis mirabilis very restricted, rare, occurring in reserves rare or insufficiently known rare Lerista allanae restricted to specific habitats, rare endangered endangered Lerista cinerea very restricted, poorly known rare or insufficiently known rare Lerista vittata very restricted, poorly known vulnerable vulnerable Menetia sadlieri — rare or insufficiently known rare Acanthophis antarcticus — rare or insufficiently known rare Denisonia maculata restricted to specific habitats, vulnerable, occurring in reserves vulnerable vulnerable Purina barnardi — rare or insufficiently known rare Purina dunmalli restricted to specific habitats, vulnerable, occurring in reserves vulnerable vulnerable T. pinguicolla Mitchell, 1948 was described as a subspecies of 7. lineata. Concern about prospects for survival of ‘T. lineata pinguicolla' has been expressed recently because of ‘substantial loss and modification of native grasslands in south- eastern Australia’ (Osborne et aL, 1993). Current taxonomic review of Tympanociyptis spp. indi- cates that T. pinguicolla is a distinct species (A. Greer, pers. comm., Nov., 1997). In southern Queensland, T. pinguicolla appears to have an extremely restricted range. It has been recorded only near the southeastern boundary of the BB, in the Oakey district. Museum holdings of this spe- cies are scant and not recent: no date, Bougeen near Oakey 27°34’S 15r27’E (OMJ8604); Dec., 1978, 30km S of Oakey 27‘’39’S 15r36’E (QMJ34744). There have been no recent reports of sightings of T. pinguicolla in the southern BB, despite considerable field work in and near appro- priate localities. The case of T. pinguicolla at the north-eastern limit of its range in the BB of Queensland parallels a major conservation prob- lem highlighted by Daugherty et a!., 1990 for Sphenodon spp. In reviewing the conservation/ management of these species, they showed how a lack of awareness of species/‘taxonomic ne- glect’ could lead to extinction. Tiliqua rugosa (Gray, 1825) is a large, conspicu- ous skink. Many museum specimens of this spe- cies, frequently obtained as ‘road-kills’, are held from the BB (e.g., Ingram & Raven, 1991). This is an easily-identified species reported often. In Queensland it has been recorded subtropically and subcoastally in the BB and mulga lands im- mediately to the west of the BB. The last BB specimen added to the Queensland Museum col- lections is QMJ58785 (Mungailala 26°27’S 147®33’E, Dec., 1993). No specimens were seen during our field work in the BB. This included intensive work on Retro Station and the lands which were formerly part of this station, a known former locality of T. rugosa (Cal. Acad. 76652- 76653, 8, 14 August, 1929). Retro has been cleared almost entirely of native vegetation. Whether T. rugosa populations have declined in recent years has not been shown conclusively. However, probable local declines have been suggested also by other herpetologists who have worked in the BB (S. Irwin and C. Eddie, pers. comms). Aspidites ramsayi (Macleay, 1882) has recently been shown to be vulnerable in the southern BB (Covacevich & Couper, 1996). 480 MEMOIRS OF THE QUEENSLAND MUSEUM TABLE 2. Conservation status of reptile species at risk in the BB of Queensland, under lUCN Species Survival Commission definitions of 1994. Species Conservation Status Lerista allanae ‘critically endangered’ ‘(CE) C2a’ ‘Population estimated to number less than 250 individuals ... A continuing decline...’ Anomalopus mackayi ‘endangered’ ‘(EN) Ale’ ‘Population reduction ... of over 50% over the last 10 years ... based on ... decline in quality of habitat...’ Rheody>les leukops Lerista vittata ‘vulnerable’ ‘(VU) D2’ ‘Population restricted ... characterised ... by an acute restriction in its area (typically less than 100km2) ... the number of locations (typically less than 5)...’ Paradelma orientalis Diplodactylus taenicauda Anomalopus hrevicoUis Anomalopus leuckartii Denisoma maculata Furina barnardi Purina dunmalli Hemiaspis damelii ‘vulnerable’ ‘(VU)AIc’ ‘Population reduction ... decline in area of extent of occurrence and ... quality of habitat ...’ Acanthophis antarcticus Delma tabialis ‘lower risk, near threatened’ ‘(LR) 2nf ‘... close to qualifying for vulnerable ...’ Delma plebeia Egernia rugosa Lerista cinerea Menetia sadlieri ‘data deficient’ ‘inadequate information ... appropriate data on abundance and/or distribution is lacking ... possibility that future research will show ... threatened ... is appropriate’ Our field work and a study arising from it (Deer, 1996) have shown that reptile diversity can re- main high in protected reserves and state lands of the BB (e.g., Dipperu N.P./I6 species; Yuleba SF/25 species); reptile diversity can remain high in the face of considerable disturbance of a forest (e.g., Fairbaim Dam remnant Callftris forest with 24 small reptile species); and predictably, that agricultural lands are depauperate in terms of reptile diversity. Although Pogona barbata (Cu- vier, 1829), Tiliqua scincoides (White ex Shaw, 1790), Antaresia macuiosus (Peters, 1873) and Pseudonaja textilis (Dumeril, Bibron & Dumeril, 1854) to remain conspicuous (?common) on and near roads in agricultural areas and fringing strips of forest and weeds, agricultural lands apparently support fewer species than reported previously from these areas prior to clearing (e.g., Slevin, 1955; Covacevich et ai., 1996b). DISCUSSION Biodiversity has declined in the BB due to clearing of native vegetation. Only 2.07% of the region is set aside expressly for protection of biodiversity; a further 4% is afforded some pro- tection in state lands. Forty percent of the regional ecosystems are known to be at risk, with most Brigalow systems endangered. One reptile spe- cies is critically endangered, possibly extinct; 16 are at risk; and 5 are of concern locally. There will be additions to the list of critically endangered- vulnerable species (13) and more species will be recognised as being at risk if the threatening proc- esses responsible for past demises are not modified. In return for clearing much of the BB there have been many gains. L.. Originally a sheep raising district, the fertile soils now support extensive grain-growing, chiefly wheat but including barley, maize, oats and oilseeds; irrigated land in the west produces cotton. Dairy and beef cattle are also important, and oil, natural gas and coal are natural resources ...’ (Dawson, 1990). L.. In 1982, (Emerald’s) population was about 4,500. Today, Emerald is a boom town. Its wealth is derived from cattle, cotton, oranges, wheat, sorghum, mung beans, sunflower seeds, and coal and gold mines. In the principal catchment area of Emerald, there are now 50,000 people, most of whom must be living well, because the average annual household in- come in the area is $45,000 ...’ (Hay, 1995). Large financial investments are proposed to continue to increase productivity. This will involve further clearing and intensification of land use. In 1995, the Queensland Government endorsed a plan to investigate three water infrastructure develop- ment projects in the Fitzroy Basin ... pipeline from Eungella Dam to the central Queensland coalfields(;) a proposed dam on the Dawson River to provide urban, industrial and irrigation water ... to existing and new land ... to be devel- oped (; and) ... a proposed dam on the Comet River to provide urban, industrial and irrigation supplies to lands adjacent to the Comet, Nogoa and MacKenzie Rivers ...’ (Queensland Depart- ment of Primary Industries, undated, 71995). A recent report of the Queensland Government’s REPTILE DIVERSITY IN THE BRIGALOW BELT 481 Water Infrastructure Task Force (1997) recom- mends for the BB, eight Category 1 water supply projects suitable for continuation or commence- ment of detailed planning and assessment; three for regional planning and assessment studies; two for groundwater projects; and one general pro- posal for water supply research, catchment man- agement and on-farm water conservation. The Task Force also recognised ‘All infrastructure development projects should be subject to satis- factory outcomes of normal impact assessment studies (including environmental, economic, so- cial and cultural impacts) and negotiation of sat- isfactory funding arrangements prior to recommendation for funding by Government Further, all Australian state governments have re- cently (July, 1996) signed ‘National Principles for the Provision of Water for Ecosystems’ following representations to the Australian and New Zea- land Environment and Conservation Council and the Agriculture and Resource Management Council of Australia and New Zealand. A major goal of these principles is providing water for the environment ... to sustain and, where neces- sary, restore ecological processes and biodiver- sity of water dependent systems The construction of dams will lead to increased pres- sures on the reptiles (and other species) of the BB, through progressive intensification and expan- sion of agriculture and associated infrastructure. This impact will not only occur in areas directly affected, ie the irrigation areas. The size of this investment will encourage development across the whole region. This, in turn, will threaten re- maining significant remnants of native vegeta- tion. By comparison, proposals to address nature conservation in the BB have not yet been accepted as urgent. IMPORTANCE OF TYPE LOCALITIES. In any area as assailed by clearing as is the BB, type localities, critical in species delineation, become important in the preservation of diversity. This principle applies particularly where species are narrowly confined or virtually-confined to a region as are many of the reptile species of the BB. The type locality is ‘... the place where the population occurs from which the type specimen [i.e., the specimen on which the description and name of any species are based] was taken ... The “type” affords the standard of reference that de- termines the application of the scientific name [ ] ours (Rules of Zoological Nomenclature in Mayr, 1969). Some BB type localities for reptile species have been published (Covacevich, 1996). The complete list of BB type localities for reptile species is extensive (Table 3). Many were poorly- defined in early type descriptions; many are now cities or towns; and some have been cleared for agriculture and/or grazing. Only a few support anything like the vegetation present when the type specimens were collected. Those where good tracts of native (and near native) vegetation remain are highlighted in Table 3. LESSONS FROM THE BRIGALOW BELT. The BB in Queensland is a fine example of how to diminish diversity — encourage clearing of native forests, through taxation and other incen- tives; fail to secure in conservation reserves rep- resentatives of all vegetation types in the region; recognise late, after several species have become extinct or are in decline, that a major problem in maintaining diversity exists; and proceed with plans for further development even though diver- sity is already known to be assailed. As attempts progress to improve biodiversity protection in the BB, lessons from other settings become relevant: 1) Research/management attention focussed narrowly on a particular habitat/region or on rare species is, inevitably, costly for other habitats/re- gions/species (e.g., Voous, 1977; Covacevich & McDonald, 1993). 2) ‘We should pay more attention to habitats rather than concentrating heavily or solely on spe- cial-interest taxa’ (Czechura& Covacevich, 1985). 3) Although habitat protection is the best ap- proach to species protection, it cannot guarantee species survival and habitat destruction guaran- tees loss of species (e.g., Ingram & McDonald, 1993; Covacevich et al., 1996). 4) ‘... fauna protection authorities ... have tended to approach reptile conservation by ... restrictions on ... collecting and keeping ... (so) much emphasis has been placed on rare and en- dangered species, a flawed approach, which al- lows statutory protection of habitat and mandatory requirements that wildlife authorities monitor populations ...’ (Rawlinson, 1981, in Czechura & Covacevich, 1985). 5) Perhaps most important is the finding that ‘... Species-rich areas (‘hot spots’) frequently do not coincide for different taxa, and many rare species do not occur in the most species-rich (areas) ...’ (Prendergastetal., 1993). In the Queensland BB it is apparent that small, remnant patches of native vegetation support many reptile species. However, a paucity of data makes it impossible to determine whether or not these areas can sustain reptile populations in the 482 MEMOIRS OF THE QUEENSLAND MUSEUM TABLE 3. Queensland BB type localities for nominate reptile species. Type localities where good tracts of native vegetation remain are highlighted *. Warrawee Stn, 60kmSE20°24' 146°40’* Lerista cinerea Greer, McDonald & Lawrie. 1983 Mt Cooper Stn, 90kni SE 20^31 ’S I46”47’E Lerista vittata Greer, McDonald & Lawrie, 1983 Bowen Port Denison) Aspidiotes melancephalus Krefft 1864 = Aspidites melanocephaliis (Krefiu 1 864) Bulliwallah Stn, on Bclyando R., via Clennont * Faranus bulliwallah Worrell, 1956 = Varanus mertensi Glauert, 1 95 1 Barmounl 80km NW Marlborough 22°32' 149°06 MEO* Menetia timlowi Ingram, 1977 Batheaston Stn 22°26'S 148°47’E* Gehyra catenata Low, 1 979 Lake Elphinstone Tropidolepistm striolatum Peters, 1 870 = Egernia striolata 1870) lioplocephalm frenatus Peters, 1 870 = (Peters. 1863) Retro, Capella Rhodona allanae Longman, 1937 = Lerista allanae {Lm^mdx\. 1937) Peak Downs Diplodactylus steindachneri Boulcngcr, 1885 Nephniriis asper Gmihcr, 1876 Delma orientalis Gunther, 1 876 = Paradelma orientalis (Giinther. 1 876) Rhondona fragths Gunther, 1876 = Lerista frdsiJis (Giinther, 1 876) Lygosdma heterodactylum (liiiuhcr, 1876 = Glaphyromorphus punctulatus (Peters- 187!) Central Queensland Grammatophora mennis De Vis. 1888 = Cterwphorus nuchalis {Dt Vis, 1884) neighbourhood of Rockhampton 1 fonlocephalus nigrostriatus Krefft. 1 8o4 = khinoplocephalus nigrostriatus 1864) Hoplocephalus nigrostriatus Kreftt. 1 8o4 = khinoplocephalus ni^rosfriatus\\\xcn\ 1864) near Rockhampton Denisonia ormita Krefft, 1869 = Dcnhionia maculata (Steindachner. 1 867) Rockhampton f/eteropus schmeltzii Peters. 1 867 = Carlia schmeltzii (Peters, 1867) Pseudechis scute llatus Peters. 1 867 = Oxy’uranus scutellatus (I’elers. 1867) Hinulia fasciolata Gunther. 1 867 = Eremiasanacifasciolatiis (GiinlJier. 1 867) Tiliqua longicauda De VTs. 1 888 = Cvclodomorphus gerrardii (Gray, 1845) (designated by Wells & Welltiiaion. 1985) Phyllodactylm anomalus Peters, 1867 = lleteronotia binoei (Peters, 1867) Pseudechis scutellatus Peters, 1867 = ()x\'uranus scutellatus (Peters, 1867) Ty'phlnps (Onchychocephalus) unguirostns part Peters, 1 867 = Ramphotyphlnps tinguirostris (Peters. 1867) restricted to Rockhampton Hoplocephalus maculatus Sleindacnner, 1867 = Denisonia maailafa (Steindachner. 1 867) Rockhampton and Peak Downs Hoplocephalus dametii Giinther, 1 876 = Hemiaspis damelii (Gunther. 1 876) Rockhampton, 1 km E of Frenchville 23‘’22’S 150'^32’E Anomalopus brevicollis Greer & Cogger, 1985 Fitzrov R.. 63km N & 2'2km E of Duaringa, eleva- tion 40m, 23°09’S I49°55’E Rheodytes leukops Legler & Cann, 1980 19km W of Emer- ald 81 .5km E of Boeantunaan* Oedura attenboroughi Wells & Wellington, 1985 15 miles S Duarinaa Glyphodon barnardi Kinghorn, 1939 = Purina barnardi (Kinanorn. 1939) State Forest, foot- hills ofMt Larcom 23°49‘S l5r22’E* Heleropus mimdusDe Vis. 1885 = Carlia munda (De Vis. 1885) Gyrandra. Dawson R. (as Upper Dawson R.) Lygosoma hancrofti Longman, 1916 = Anomalopus leuckartii (Weinland. 1862) Port Curtis* Simotes australis Krefft, 1 864 = Smoselaps australis (Krefft. 1864) Monto Phy'llunis salebrosus CovaccVidy 1975 = Satuarim salebrosus (Covacevich. 1 975) Eidsvold. Upper Burnett R.* Varanus punctaius orientalis Fry, 1913 = Varanus tristis (Schleael, 183*9) Burnett River * Diemenia maailiceps Uoettaer, 1 898 = Demansia vestigiata (De Vis, 1884) Lygosoma verreauxii biunguiculata Oudemans, 1894 = Anomalopus leuckartii (Weinland. 1 862) Chinchilla Diplodactylus taenicauda De Vis, 1886 I^ernia whitei modesta Slorr. 1968 = tsernia modesta (SlOTT. 1968) Chinchilla. Darlina Downs Heteropus vertebralis Dc Vis, 1 888 = Carlia mundivensis (Broom, 1 898) Mowen (?=Morven) central railway Typhlops diversus Waite, 1 894 = Ramphotyphlops diversus (Waite, 1 894) Glenmorgan* Glyphodon dunmalll Worrell. 1955 = Purina dunmalli (Worrell, 1955) Denisonia d\^’\‘eri Worrell. 1956 = Sutadw\'eri (Worrell, 1956) near Surat Hoplocephalus ornatus Dc Vis. 1 884 = Denisonia devisi (Waite & Longman. 1920) Cecil Plains. SEO Pseudechis suttata De Vis. 1905 Moombah Sin.. ca 64km W West- mar, ca 60km NE St George. SEQ27°59^ I49^I8'E* Ctenotus ingrarni Czechura & Wombey. 1982 long temi. Our recent field work in the BB has shown that reptile diversity in some areas, all still supporting native vegetation, is high. However, despite the intensity of this work and success in finding reptiles, few of the 20 species at risk in the BB were encountered. In Dipperu NP, none of these species was found; in Inglewood SF one {Acanthophis antarcticiis) was found and one {Paradelma orientalis) has been found since (Schultz & Eyre, 1997); in Callitris forest near REPTILE DIVERSITY IN THE BRIGALOW BELT 483 Emerald, one {Diplodactylus taenicauda) was common (Deer, 1 996); and an isolated population of Aspidites ramsayn is well-known in the Yuleba district. These results superficially support the findings of Prendergast et al., 1993, that L.. A limited number of species-rich areas do not guarantee effective conservation for rare and re- stricted organisms, many of which occur outside species-rich areas ...L However, they are far from a thorough assessment of the question of the coincidence of diversity hotspots and rare species in the BB. LITERATURE CITED ANONYMOUS, 1 997. New rules to protect endangered habitats. The Courier Mail, 10 July 1997, p. 7. BRANSDON, A. 1995a. farmers plan protest. Central Queensland News. 3 May. p. 1 1995b. Graziers in limbo after tree protest. Central Queensland News. 5 May. p. I . COLLIE, G. 1995. Clearly the hot topic. Courier Mail. 1 July, p. 33. COGGER. H.G., CAMERON, E.E., SADLIER, R.A. & EGGLER, P. 1 993. The action plan for Austra- lian reptiles. (Australian Nature Conservation Agency: Canberra). COUPER, P.J. 1996. Neprurus asper (Squamata: Gekkonidae): sperm storage and other reproduc- tive data. Memoirs of the Queensland Museum 39(2): 487. COUPER, P.J., COVACEVICI I, J.A. & McDONALD, K.R. 1996. A Bandy Bandy with a difference. Memoirs of the Queensland Museum 39(2): 242. COUPER. P.J., COVACEVICH, J.A. & WILSON, S.K. 1995. Sap feeding by the Australian gecko Gehyra dubia. Memoirs of the Queensland Mu- seum 38(2): 396. CO VACEVICH, J.A. 1 996. Realities in the biodiversity Holy Grail: prospects for reptiles of Queensland's Brigaluw Biogeographic Region. Proceedings of the Roval Society of Queensland 106(1): 1-19. COVACEVICI 1, J.A. & COUPER, P.J. 1996. Aspidites ramsayi (Boidae) in the Brigalovv Biogeographic Region of Queensland: occurrence, conservation status and possible bilby associations. Memoirs of the Queensland Museum 39(2): 243-246. COVACEVICI!, J.A. & COUPER. P.J. 1997. Maida Allan. Pp. 29-30. In McKay, J. (ed.). Brilliant Careers (Queensland Museum: Brisbane). COVACEVICI L J.A. & McDONALD. K.R. 1 993 . Dis- tribution and conservation of frogs and reptiles of Queensland rainforests. Memoirs of the Queens- land Museum 34(1): 189-199. COVACEVICH, J.A., COUPER, P.J. & McDONALD, K.R. 1996a. Lerista allanae (Scincidae: Lygosomi- nae): 60 years from exhi bition to extinction? Mem- oirs of the Queensland Museum 39(2): 247-256. 1996b. Reptiles of Queensland’s Brigalow Bio- geographic Region: distributions, status, conser- vation. Unpublished Report to the Australian Na- ture Conservation Agency, Canberra. (Project 504) COVACEVICH, J., DUNMALL, W. & SORLEY, J. 1988. Reptiles. Pp. 265-273. In Scott, G. (ed.). Lake Broadwater. The natural histoiy- of an inland lake and its environs. (Darling Downs Institute Press: Toowoomba). CROSSMAN. D.G. & REIMER, D.S. 1 986. Mammals, birds, reptiles and amphibians of the Taroom Shire, central Queensland. Queensland Journal of Agricultural and Animal Sciences 43(1); 55-72. CZECHURA. G.V. & COVACEVICH. J. 1985. Poorly known reptiles in Queensland. Pp. 472-476. In Grigg, G., Shine, R. & Ehmann, 1 1, (eds). Biology of Australasian frogs and reptiles (Surrey Beatty & Sons: Chipping Norton). DAUGHERTY, C.H., CREE, A., HAY, J.M. & THOMPSON. M.B. 1990. Neglected taxonomy and continuing extinctions of tuatara {Spheno- c^o«). Nature. 347(6289); 177-179, 13 Sept., 1990. DAWSON, S. (ed.) 1 990. The Penguin Australian Encyclo- paedia. (Penguin Books Australia Ltd: Ringwood). DEER, R. 1996. Reptile diversity in a Callitris forest in central Queensland’s brigalow belt. Memoirs of the Queensland Mu.seum 39(2): 390. DAVIE, J.S., BARRY, S. & MORGAN, G. 1994. Plan- ning for nature conservation in rural environ- ments: the Brigalovv Belt in central Queensland. Pp. 69-89. In I lynes, R.A. & Johnson, R.W. (eds). Rural Queensland: A sustainable future. The ap- plication of geographic information systems to Imid planning and management. Proceedings of a symposium held 23-24 Nov., 1989. (The Royal Society of Queensland: Brisbane). GARNETT, S. (ed.) 1992. Threatened and extinct birds of Australia. Royal Australasian Ornithologists Union Australian National Parks and Wildlife Service RAOU Report 82. (Royal Australasian Ornithologists Union: Moonee Ponds). GASTEEN, J. 1985. The Brigalow lands of eastern Australia - agricultural impact and land use po- tential versus biological representation and stabil- ity. Pp. 45-49. In Gaslecn, J., Henry, D. & Page, S. (eds). Agriculture and conservation in inland Queensland. (Wildlife Preservation Society of Queensland: Brisbane). 1987. The brigalow lands of Australia: agricultural im- pact and landusc potential versus biological rep- resentation and stability. In Werren, G. & Kershaw, A.P. (eds), The rainforest legacy: Aus- tralian National Rainforests Study Vol. 1. Special Australian Heritage Publication Series No. 7(1). (Australian (iovernment Publishing Service: Canberra). GIBSON, N.V. 1995. An open letter to Wayne Goss. Central Queensland News. 17 May 1995, p. 4. GORDON, G. 1984. Fauna of the Brigalow Belt. Pp.61- 70. In Bailey, A. (ed.). The Brigalow Belt of Australia. The proceedings of a symposium held at the John Kindler Memorial Theatre, Queens- 484 MEMOIRS OF THE QUEENSLAND MUSEUM land Institute of Technology. (The Royal Society of Queensland: Brisbane). HANNAH, D.S. & SMITH. G.C. 1995. Effects of pre- scribed burning on reptiles in southeastern Queensland. Memoirs of the Queensland Museum 38(2): 529-31. HAY, J. 1995. The tidiest town in the west. Sunday Mail Magazine, 10 Sept., p. 9. HOBBS, T.J., MORTON, S.R., MASTERS, P. & JONES, K.R. 1994. Influence of pit-trap design on sampling of reptiles in arid spinifex grasslands. Wildlife Research 21: 438-490. HORSUP, A., JAMES, C. & PORTER, G. 1993. Ver- tebrates of dry rainforest of south and midcasiern Queensland. Memoirs of the Queensland Museum 34(1): 215-228. INGRAM, G.J.& MCDONALD, K.R. 1993. An update on the decline of Queensland’s frogs. Pp. 297-303. In Lunney, D. & Ayers, D. (eds) Herpetology in Australia. A diverse discipline. (The Royal Soci- ety of New South Wales with Surrey Beatt>' & Sons: Mosman & Chipping Norton). INGRAM, G.J. & RAVEN, R.J. (eds) 1991. An atlas of Queensland’s frogs, reptiles, birds and mammals. (Queensland Museum: Brisbane). lUCN COUNCIL 1994. lUCN Red List Categories. Prepared by the lUCN Species Survival Commis- sion as approved by the 40th meeting of the lUCN Council, Gland, Switzerland, 30 Nov., 1994. (lUCN Council: Gland). JAMES, J. W. 1 974. A sur\'ey of the vertebrate fauna of Southwood National Park. Technical PaperNo. 1. (Queensland Department of Forestry: Brisbane). JOHNSON, R.W. 1996. The impact of clearing on Brigalow communities and consequences for con- servation. Unpublished conference presentation, 'Conservation outside nature reserves’, the Uni- versity of Queensland. Februaiy', 1 996. Abstract published. In Anon, (cd.) Con.scrvation outside nature reserves, a conference to look at nature conservation in landscapes managed primarily for other uses. (Centre for Conservation Biology, the Universitv of Queensland: St Lucia.) KIRKPATRICK, T.H. 1968. Mammals, birds and rep- tiles of the Warwick district, Queensland. 3. Rep- tiles and general conclusions. Bulletin 447. Queensland Journal of Agricultural and Animal Sciences 25: 235-241. LEE, A.K. 1995. The action plan for Australian rodents. Australian Nature Conservation Agency Endan- gered Species Program Project 130. (Australian Nature Conservation Agency: Canberra). MAXWELL, S., BURBIDGE, A.A. & MORRIS, K. (eds) 1996. The 1996 action plan for Australian marsupials and monotremes. Wildlife Australia Endangered Species program Project Number 500. (Australian Nature Conservation Agency: Canberra). MAYR, E. 1969. Principles of systematic zoology. (McGraw-Hill Book Co.: New York). McDonald, k.r., covacevich, j.a., ingram, G.J. & (30UPER, P.J. 1991. The status of frogs and reptiles. Pp. 338-345. In Ingram, G.J. & Ra- ven, R.J. (eds). An atlas of Queensland’s frogs, reptiles, birds and mammals. (Queensland Mu- seum: Brisbane). NATURE CONSERVATION (WILDLIFE) REGU- LATION 1994. Queensland Subordinate Legisla- tion 1 99 1 No. 474, Nature Conservation Act 1 992. NIX, H. 1994. The Brigalow. Pp. 198-233. In Dovers, S. (ed.) Australian environmental history. Essays and cases. (Oxford University Press: Oxford). OSBORNE, W.S., KUKOLIC, K. & WILLIAMS, K.D. 1993. Conservation of reptiles in lowland native grasslands in the Southern fablelands of New South Wales and the Australian Capital Territory. Pp. 151-158. In Lunney, D. & Ayers, D. (eds) Herpetology in Australia. A diverse discipline. (Royal Zoological Society of New South Wales: Mosman). PRENDERGAST, J.R., QUINN, R.M., LAWTON, J.H., EVERSHAM, B.C. & GIBBONS, D.W. 1993. Rare species, the coincidence of diversity hotspots and conservation strategies. Nature: 365, 23 Sept, 1993:335-337. QUEENSLAND GOVERNMENT WATER INFRA- STRUCTURE TASK FORCE 1997. QUEENSLAND DEPARTMENT OF PRIMARY IN- DUSTRIES. Undated, 71995. Proposed Fitzroy Basin Development Project. Water ... securing the future. Leaflet (Queensland Department of Pri- mary IndusU'ies: Brisbane). RAWLINSON, P.R. 1981. Conservation of Australian amphibian and reptile communities. Pp. 127-38. In Banks, C. & Martin, A.A. (eds) Proceedings of the Melbourne Herpetological Symposium. (Zoo- logical Board of Victoria: Melbourne). RUSSELL, M.J., YOUNG, R.A., DORRICOTT, K.E., ALCOCK, K.M., GLOSSOP, W.J., PITT, K.M., McKILLIGAN, N.G. & McCONNELL, P. 1992. Planning and monitoring of habitat retention in Queensland. Unpublished Report. Project Report 39, for the Au.stralian National Parks and Wildlife Service Endangered Species Program: Canberra. SATTLER, P. 1986. Nature conservation in Queens- land: planning the matrix Proceedings of the Royal Society of Queensland 97: I -21. 1993. Towards a nationwide biodiversity strategy: the Queensland contribution. Pp. 313-325. In Moritz, C. & Kikkawa, J. (eds). Conservation Biology in Australia and Oceania. (Surrey Beatty & Sons Pty ltd: Chipping Norton, Sydney). SATTLER, P. & WEBSTER, R.J. 1984. The conserva- tion status of Brigalow (Acacia harpophylla) communities in Queensland. Pp. 149-160. In Bailey, A. (cd.). The Brigalow Belt of Australia. (Royal Society of Queensland: Brisbane). SCHULTZ, M. & EYRE, T.J. 1997. New distribution and habitat data for the pygopodid, Paradelma orientalis (Gunther, 1876). Memoirs of the Queensland Museum 42(1): 212. REPTILE DIVERSITY IN THE BRIGALOW BELT 485 SLEVIN, J.R. 1955. Notes on Australian amphibians. Proceedings of the California Academy of Sci- ences Series 4 28(8); 355-392. STANTON, J.P. & MORGAN, G. 1977. The rapid selection and appraisal of key and endangered sites. The Queensland Case Study. University of New England, School of Natural Resources Re- port, No, PR4. TRACKWAY, R. & CRESSWELL, l.D. (eds) 1995. An interim biogeographic regionalisation for Aus- tralia; a frame work for setting priorities in the national reserve system co-operative program, version 4. (AuslralianNatureCon.servation Agency; Canberra). VOOUS, K.H. 1977. Three lines of thought for consid- eration and eventual action. Pp. 343-7. In Chan- cellor, R.D. (ed.) World Conference on Birds of Prey. (International Council for Bird Preserva- tion; Vienna). WILLIAMS, B. 1997. Row over plan for tree-clear permits. The Courier Mail, 25 July, p. 7. APPENDIX 1 Reptile species of the Brigalow ‘Belt’ of Queensland. Family Chelidae Che lodina expansa Gi^y, 1857 C. longicoUis (Shaw, 1 802) Elseya dentata (Gray, 1 863) E. irwini Cam. 1997 E. latisternum Gray, 1 867 E. mydura krefftii {Gray, 1871) E. macquarii (Gray, 1831) Rheodytes leukops Legler & Cann, 1980 Family Gekkonidae Diplodactylus conspicillatus Lucas & Frost, 1 897 D. steindachneri Boulenger. 1885 D. taenicauda De Vis, 1886 D. tessellatus (GOnther, 1 875) O. vittatus Gray, 1 832 D. williamsi Kluge, 1963 Gehyra catemta Low, 1979 G. diibia (Macleay, 1877) G. variegata {DwmtxW, 1836) Heteronotia binoei Gray, 1 845 Nephrurus asper Gunther, 1 876 N. milii (Bory de St Vincent, 1 825) Oedura castelnaiiil\{Qm\mX. 1889 O. lesueurii (Dum^ril & Bibron, 1836) O. marmorata Gray, 1 842 O. monilis De Vis. 1 888 O. rhombifer Gray, 1 845 O. robusta Boulenger, 1 885 O. tryoni De Vis, 1885 Rhynchoedura ornata Gunther, 1 867 Saltuarius salebrosus (Covacevich, 1975) Family Pygopodidae Delma inornata Kluge, 1974 D. labialis Shea, 1987 D. plebeia De Vis, 1888 D. tincta De Vis, 1 888 Lialis burtonis Gray . 1835 Paradelma orientalis (GUnther, 1 876) Pygopus mgricepsV\SQ\\QX, 1882 Family Scincidae Anomalopus brevicollis Greer & Cogger, 1985 A. Zewcfa/r/;/ (Weinland, 1862) A. mackayi Greer & Cogger, 1985 A. verreauxii Dumiiril, 1851 Carliajarnoldae Covacevich & Ingram, 1975 C. mu^a (De Vis, 1885) C. mimdivensis (Broom, 1 898) C. pecforalis (De Vis, 1885) C. schmeizii fPeters, 1 867) C. tetradactyia {O'Shaughmssy, 1879) C. vnm(Dc Vis, 1885) Cryptoblepharm carnabyi (Storr, 1976) C. plagiocephalusiCocXtau. 1836) C. virgatiis (Garman, 1901) Ctenotus allotropis Storr, 1 98 1 C. ew/aemiAS Storr, 1981 C. hebetior %ion, 1978 C. ingrami Czechura & Wombey, 1982 C. panther 'mus Peters, 1 866 C. robustus Storr, 1 97 1 C. strauchii (Boulenger, 1 887) C. taeniolatus (Shaw, 1 790) Cyclodomorphus gerrardii (Gray, 1 845) Egernia cmninghami (Gray, 1 832) E.frerei GUnther, 1897 E. modestaStoxx, 1968 E. rugosa De Vis, 1888 E. striolata Peters, 1 870 Eremiascincus fasciolatus (Gunther, 1 867) E. richardsoni (Gray, 1 845) Eulamprus brachysoma (LOnnberg & Andersson, 1915) E. martini Wells & Wellington, 1985 E. quoyii (Dum^ril & Bibron. 1839) E. sokosoma Greer, 1 992 E. tenuis Gray. 1 83 1 Glaphyromorphus punctulatus (Peters, 1871) Lampropholis delicata (De Vis, 1888) L. mirabilis Ingram and Rawlinson, 1981 Lerista allanae {Loxxgman, 1937) L cinerea Greerm McDonald & Lawrie, 1983 L fragilis (Gunther, 1 876) L mueileri (V'lschcx, 1881) L. pimctatovittata (Gunther, 1867) L. vittata Greer, McDonald & Lawrie, 1983 Lygisaurus foliorum Dc Vis, 1 884 Menetia greyii Gray, 1 845 M. Greer, 1991 M. z/w/ow/ Ingram, 1977 Morethia boulengeri (Ogilby, 1890) M taeniopleura (Peters, 1 874) Proablepharus kinghorni (Copland. 1947) Saiphos equalis (Gray, 1 825) Tiliqua multifasciata Sternfeld, 1919 T. rugosa (Gray, 1827) T. scincoides (Shaw, 1 790) Family Agamidae Amphibolurus burnsi (Wells & Wellington, 1985) 486 MEMOIRS OF THE QUEENSLAND MUSEUM A. gilberti (Gray, 1842) A. (Shaw, 1790) A. nobbi WiUen ,1972 Chlamydosaurus kingii Gray, 1 825 Ctenophorus mtchalis (De Vis, 1885) Diporiphora australis (Steindachner, 1 867) Physignathus lesueurii (Gray, 1831) Pogona barbata{Q\i\\^T, 1829) Tympanocryptis lineata Peters, 1863 T. Mitchell, 1948 Family Varanidae Varanus gouldii {Gr?iy, 1838) V. mertensi G\iA\QTi, 1951 V. panoptes Storr, 1 980 V. .vcfl/tfm Merlcns, 1941 V. (Schlegel, 1839) V. varius 1790) Family lyphloptdae Ramphotyphlops ajfmis (Boulenger, 1889) R. bitubercidatus 1863) R. (Waite, 1918) R. (Peters, 1879) R /7/'av/>ww.v (Waite, 1893) R. unguirostris (Peters, 1 867) R. wiedii (Peters, 1 867) Family Boidac Antaresia maculosa (Peters, 1 873) Aspidites melanocephalus (Krefft, 1 864) A. ramsay i (Mac\cay, 1882) Morelia spilota {Laceptde, 1804) Family Colubridae Boiga irregularis (Merren, 1 802) Dendrelaphis punctidata (Gray, 1 827) Tropidonophis mairii (Gray, 1841) Family Elapidae Acanthophis aniarclicus (Shaw, 1 794) Cacophis harriettae Krefft, 1 896 Demansia papuensis (Macleay. 1877) D. psammophis (Schlegel, 1837) D. torqi4ata (GOnther , 1 862) D. vestigiata (Dc Vis, 1884) Denisonia devisi Waite & Longman, 1920 D. Steindachner, 1867 Purina barnardi 1939) F. (s'/WeAwa (Schlegel. 1837) F. (Worrell, 1955) Hemiaspis damelii (Giinther, 1 876) Hoplocephalus bitorquatus (Jan, 1 859) Notechis scutatus (Peters, 1861) Pseudechis australis (Gray, 1842) P. guttatus De Vis, 1905 P. porphyriacus (Shaw', 1 794) Pseudonaja nuchalis Gunther, 1 858 P. textilis (Dumeril, Bibron, & Dumeril, 1854) Rhinoplocephalus boschmai (Brongersma & Knaap-van Meeuwen, 1961) R. nigrescens (GMUcr, 1862) R. nigrostriatus {Krefft, 1864) Simoselaps australis Krefft, 1864 Suta dwyeri CWorreW, 1956) S, suta (Peters, 1863) Vermicella annulata {Gray , 1841) TWO NEW SEMI-AQUATIC SPECIES OF DIPLOTREMA (OLIGOCHAETA: MEGASCOLECIDAE: ACANTHODRILINAE), WITH A RE-DESCRIPTION OF THE TYPE SPECIES G. R. DYNE AND B.G.M. JAMIESON Dyne, G.R. & Jamieson, B.G.M. 1998 06 29; Two new semi-aquatic species of Diplotrema (Oligochaeta; Megascolecidae: Acanthodrilinae), with a re-description of the type species. Memoirs of the Queensland Museum 42(2): 487-493. Brisbane. ISSN 0079-8835. Two new species of the genus Diplotrema sharing a suite of morphological characters and a semi-aquatic mode of life are described. Together with the previously described D. tyagarah. they are placed in a species-group the nominate species of which is the generic type-species, D. fragilis. The latter taxon, of pivotal importance in early phylogenetic schemes owing to an erroneous determination of the condition of the male pores, is redescribed from the syx\\yxies.n Earthworms, Megascolecidae, Diplotrema, semi-aquatic. G.R. Dyne, Environment Australia, PO Box 787, Canberra ACT 2600, Australia: B.G.M. Jamieson, Department of Zoology, University of Queensland, St Lucia 4072. Australia; 4 May 1997. The significance of the type-species of Diplo- trema, D. fragilis Spencer, in the inferred phylo- geny of the Megascolecidae proposed by Stephenson ( 1 930) has been outlined by Jamieson (1971) and Dyne (1978). More recent studies (Jamieson & Dyne, 1976; Dyne, 1984) have shown that the genus is far more widespread and speciose than previously believed, and that dis- tinct species-groups are recognisable. The two new species described here confirm the concept of a species-group including the type- species of the genus and a number of morphologi- cally similar species inhabiting a similar ecotype. The collection of these species at sites either subject to periodic inundation, or pennanently below the water-table, testifies to their semi- aquatic propensities. This represents exploitation of a niche largely unfilled by other Australian megascolecids. SomQ Spencer iella species are known from swampy habitats, however (Dyne, 1984). SYSTEMATICS Members of the Diplotrema fragilis species- group share the following character set: pros- tomium epilobous, nephropores in the vicinity of cd, slender body form, S pores slightly presetal, intersegmental markings restricted to the region of segments XVI-XX, genital seta follicles strongly developed in VII, VIII or IX, often with associated glands, penial setae fine and delicate with little or no ornamentation, pharyngeal tufted nephridia absent, large glandular mass (here termed ‘ventral glands’) located under the nerve- cord in XVII-XTX, spermathecae divided into a subspherical ampulla and short, blunt digitiform diverticulum bearing numerous sperm chambers on its inner surface (D. helonoma diverges some- what from this latter state). It should be noted, however, that most of these unifying characteristics are symplesiomorphies: only the possession of pharyngeal tufting, genital seta glands and ventral glands represent probable synapomorphies. Five new species of Diplotrema from south and central Queensland were recently partly de- scribed and sketched (Blakemore, 1997). Four of these differ from the fragilis group in lacking anterior nephridial tufting. A fifth, D. ambrosen- sis, has anterior tufts, but, as for the other species, the form and sculpturing of the penial, and geni- tal, setae is not described, considerably impeding comparison, and no reference is made to ventral glands. As the penisetal follicles are described as ‘enormous’, it is unlikely that D. ambrosensis belongs to the fragilis group. KEY TO SPECIES OF THE DIPLOTREMA FRAGILIS SPECIES-GROUP 1 . Dorsal pores present 2 Dorsal pores absent D. proserpinensis sp. nov. 2. Prostatic glands long, extending at least into seg- ment XXI D. fragilis Sx>Qncex Prostatic glands not extending beyond XX 3 3. Gizzard rudimentary; first dorsal pores in the vi- cinity of 19/20; conspicuous glands associated with the genital setae D. helonoma sp. nov. Gizzard well-developed; first dorsal pores in the vicinity of 7/8; no conspicuous glands associ- ated with the genital setae. . . . D. tyagarah Dyne 488 MEMOIRS OF THE QUEENSLAND MUSEUM Diplotrema fragilis Spencer, 1900 (Fig- 1) Diplotrema fragilis Spencer, 1 900: 3 1 -32 pi . 4, figs 4-6. Diplotrema fragilis^ Sweet, 1900; 1 14-5 pi. 14, fig. 6. Diplotrema fragilis’, Michaelsen, 1907: 142. Plutelliis (Diplotrema) fragilis’, Michaelsen, 1916: 61. Diplotrema fragilis, 1971: 100-102. MATERIAL. Many specimens examined, I closely (condition poor). SYNTYPES; National Museum of Victoria G3L Approx. 25°37’S, 15U37’E., Gayndah, Sept., 1891- Qld. DESCRIPTION. Length = 3 1 mm. Width = 2.0mm. Segments =111. Form cylindrical, anterior end blunt and rounded, slightly bulbous, pigmentless buff in alcohol. Prostomium closed, epilobous 1/2; first dorsal pore (perforate) in 9/10. Setae 8 per segment, closely paired, the ventral setal cou- ples of XVIII present, those of XVII and XIX modified as enlarged penial setae, those of VIII and IX as genital setae. Nephropores not visible. Clitellum annular, weaker ventrally, over XIII- XVI, dorsal pores obscured, intersegmental fur- rows and setae only sporadically visible. 6 pores minute orifices equatorial, in XVIII. 6 field an approximately circular, somewhat tumid area ex- tending longitudinally from the setal arc of XVI to that of XIX, and laterally, in XVIII, beyond b-Wnes, small, indistinct, transversely oval papil- lae (the prostatic porophores) present in XVII and XIX, at the sites of the modified ventral setal couples. 9 pores minute orifices in a pair of considerable papillae in front of a setae, on XIV. Spermathecal pores on a pair of elliptical, slender papillae, in 7/8, 8/9, with centres in b. Accessory markings: a pair of homy ridges pos- teriorly in XIV, median to a (evidently misiden- tified by Spencer as the 9 pores); a median transversely elliptical prominence in 1 7/1 8 is pos- sibly a normal feature of the species. Segment VIII glandular, and appreciably widened. Septa: 5/6 slightly, 6/7 and 7/8 moderately strongly, and 8/9 and 9/10 very strongly thick- ened, the succeeding three septa slightly thick- ened, remainder thin. Pharyngeal glands well-developed, forming several tiers of flat- tened, shortly palmate lobes in HI, anteriorly. Dorsal blood vessel single, continuous to the pos- terior portion of the pharynx, flattened and adher- ent to the intestine in XIV posteriorly; last hearts in XIII, those of X-XIII large, and increasing slightly in size posteriad, supra-oesophageal ves- sel present, but its limits not detemiinable be- cause of maceration. Gizzard large, glossy, but thin-walled, in V. Oesophagus narrow, VI-XIV, FIG. Diplotrema fragilis Spencer, 1900. Right sper- matheca of segment IX. Syntype {NMVG3 1). calciferous glands or other elaborations absent; intestine commences with abrupt expansion in XV, reaching full width in XVI, a definite ty- phlosole lacking. Nephridia exclusively holonephric, commenc- ing in 11, each invested in a high peritoneum and forming a regular, simple lobe on each side, in each segment; ducts long and avesiculate, enter- ing the parietes in c-lines. Pharyngeal tufting absent. Holandric: large, thick-walled spermatic fun- nels present in X and XI; seminal vesicles 2 pairs of compact, iobulated, and fairly small masses in XI and XII. Prostatic glands 2 pairs of flattened, tubular organs, the anterior pair extending to XXII, the posterior set to XXIIL The glands are sinuous, with each bend contiguous with the next, ducts abruptly demarcated, slender, glossy, and running almost straight medianly. Anterior pros- tates approximately twice the size of the posterior pair, though the length of the ducts is similar; the ducts enter a median, internal glandular mass extending from XVII to XIX, and corresponding with the external manifestation of the 6 field. Two penial seta follicles enter the glandular mass in the vicinity of the prostatic ducts, each with several setae; the setae slender, narrowly pointed, basally often expanded, gently curved or slightly sinuous, with the ectal 1/4 curved through as much as a right angle, the ectal end simple, or with a crumpled appearance; minute notching present on the shaft, mostly almost inappreciable. Length mature seta = 1. 4mm (mean of 2), midshaft di- ameter = 40]im. Ovaries moderately stout, long, linguiform lobes of several large, irregularly ar- ranged oocytes; funnels moderate-sized, simple, also in XIII. Spermathecae 2 subequal pairs, in VIII and IX, each with a medianly directed ovoid, or spherical ampulla and a lateral, equisized or somewhat longer, iridescent concavo-convex TWO NEW SPECIES OF DIPLOTREMA 489 diverticulum. The ampulla and diverticulum share a short narrow duct; length right sper- matheca of IX (base of ampulla to pore) = 0.64mm. A battery of genital setae present in segments VIII and IX, located behind each sper- mathecal duct; the setae gently curving, with a broad point, omam.ented with the deep, elongate, crescentic notching characteristic of such setae in acanthodriles, from very near the tip to slightly less than halfway down the shaft, length mature seta = 9.98mm, midshaft diameter = 40pm (mean of2). REMARKS. There can be no doubt that the re- examined specimen is conspecific with the mate- rial described by Spencer (1900) as Diplotrema fragilis, the type of a new genus. The specimen presumably contributed to the description. Points of agreement in Spencer's account which confirm the identity of the re-examined specimen are: the epilobous prostomium, the large oval midventral glandular patch in XVI-XIX, and the swollen, glandular appearance of segment VIII. With re- spect to the position of the 6 and prostatic pores, however, Spencer’s account is erroneous, a de- parture from his usual accuracy, possibly because of the small size of the species. No credence can now be placed on his observation of only a single pair of prostates with pores behind the 6 pores on XVIII; the re-examined specimen has two pairs of prostatic pores, on XVII and XIX, with the 24mm, 1 13mm. Width (midclitellar) = 2.8mm. Segments ?, 182 (H - posterior amputee, PI). Pigmentless buff in alco- hol, clitellum with a pinkish tinge, particularly in freshly preserved material. Prostomium epilo- bous 1/3, closed. First dorsal pore 19/20 (H, PI), dorsal pores generally indistinct. Setae regularly disposed in 8 rows, uniform in size with the exception of the ventral setae of XVII and XIX, which are enlarged as penial setae, and those of VIII and IX, modified as genital setae; ventral setal couples of XVlll present. Nephropores vis- ible as minute depressions immediately anterior to c in each segment, by a distance slightly ex- ceeding cd. Clitellum annular, obscuring the intersegmen- tal furrows, though the setae remain prominent; clitellum developed over 1/3 XII-XVl (more so 490 MEMOIRS OF THE QUEENSLAND MUSEUM female pore prostatIc pore male pore prostatic pore •genital marking FIG. 2. Diplotrema helonoma sp. nov. Male genital field. Holotype (ANICGD.95. 1 44. 1 ). dorsally, where it extends to beyond 1/2 XVII). S pores 2 orifices at the midpoints of a pair of narrow, ill-defined (shallow) seminal grooves linking the prostatic porophores of a side; the 6 pores are located barely presetally on XVIIL in be, closer to b. Dependent on the state of muscular contraction during fixation, the d field (defined laterally by the seminal grooves, anteriorly by a rhomboidal tumescence, and posteriorly by an- other marking), may be either generally concave, which sharpens the relief of the porophores, and the ridge joining the d pores, or of fairly unifonn contour. The fonner condition is presumably as- sumed during copulation, when the genital ori- fices would need to be raised for adpression to a partner. The porophores are coincident with the penial seta openings; the setae may be protruded in preserved specimens (H, P4), or withdrawn (PI, P3); in the former case, each porophore would appear to have only a single seta protruding at any one time. $ pores a pair of pre-setal slits, roughly aligned with ab, on XIV. Spermathecal pores 2 pairs in 7/8 and 8/9, each with a prominent presetal lip, and with posteriorly adjacent tumes- cences associated with the genial seta follicles. Accessory markings: a large, unpaired median rhomboidal (or elliptical) swelling in 16/17 (de- veloped to a greater or less extent in all clitellate specimens examined); a similar marking in 19/20; the d field (divisible into 4 depressed areolae in contracted specimens) is also glandu- lar, occasionally (P4), with pore-like impressions. Septa: 5/6 thickened, 6/7-9/10 moderately mus- cularised, remainder thin. Dorsal blood vessel single, continuous onto the pharynx; last hearts in XIII, the commissurals in X-XIII large, and latero-oesophageal; the supra-oesophageal vessel is traceable from VI to XIV, attaining maximal size in VI-IX, where its increased importance apparently corresponds to a decline in the size of the dorsal vessel. The remaining commissurals decrease in size anteriad, and are dorso-ventral only. Gizzard small, soft and readily compressed, though with a slight muscular sheen, in V; oe- sophagus moderately wide, vascular, in VI-XVII, calciferous glands or conspicuous outpouchings absent. Intestinal origin in XVII, dilating gradu- ally posteriad, typhlosole lacking. Holonephric: nephrostomes conspicuous on long necks in ab\ avesiculate ducts open in line with c-setae, entering the parietes slightly anterior to the setal arc; nephridial bodies are enveloped in thin peritoneum, appearing as regular wafer- like units in each segment. Anterior tufting is absent. Holandric: 2 pairs of brightly iridescent sper- matic funnels are present in X and XI, and 2 pairs of large, acinous seminal vesicles in IX and XII, the latter pair conspicuously the larger, and occu- pying virtually the entire ventral portion of that segment. Prostatic glands 2 pairs of coiled, tubu- lar organs which encroach on posterior segments to a greater or lesser degree, penetrating the septa. The anterior pair is always the larger of the two. The prostatic ducts are short, straight and slightly muscular. The entire prostatic region is covered ventrally with a glandular mass that appears to be an internal manifestation of the S field, and which may be responsible for secretions neces- sary for successful copulation. The prostatic ducts and penisetal follicles pass through this mass prior to discharging to the exterior. Penisetal fol- licles (consisting of conjoined a and b setal com- ponents) are equipped with little copulatory musculature; the setae are moderately long and gently arcing, the ectal 1/8 ornamented with ir- regularly spaced, rounded, deep scallops of vari- able diameter with somewhat jagged overhangs; these become more scattered and shallower en- tally. Length of mature setae = 1.9mm, midshaft diameter = 36.2pm (mean of 3). Ovaries, consisting of flabellifonn clusters of oocytes, and medium-sized ovarian funnels, pre- sent in XIII; no ovisacs demonstrable. Sper- mathecae 2 subequal pairs in VIII and IX, TWO NEW SPECIES OF DIPLOTREMA 491 FIG. 3. Diplotrema helonoma sp. nov. Left spermathe- cal of segment IX. Paratype I. discharging anteriorly in their segments; each consisting of a bipartite ampulla: a clavate or irregular saccifonn portion, connected by a broad neck to a subspherical component. A small, hemi- spherical ‘diverticulum’ is sessile on the latter, opposite to the point of attachment of the clavate section of the ampulla; a medium-sized duct is shared by the ampulla and diverticulum; the latter is brightly iridescent, containing numerous intra- mural sperm chambers. Length right spermatheca of IX = 1. 7mm (base of ampulla to pore). Con- spicuous genital seta follicles are associated with the spermathecae; these are surrounded by promi- nent, discrete, whitish glands, of which there is one per follicle in IX, and two in VIII. These glands are prostate-like, both in overall morphol- ogy and histologically, with a definite central lumen leading to small ducts discharging at the point of emergence of the genital seta from the body wall; their function is unknown. The setae are fairly straight, ornamented over the ectal por- tion of the shaft (including the somewhat swollen apical region) with conspicuous longitudinal flut- ing; this appearance is due to the alignment of long roughly parallel scallops with fairly smooth edges. Length of mature seta = 1 .34mm; midshaft diameter = 35. 1 pm (mean of 3). REMARKS. Diplotrema helonoma is the geo- graphically closest species to D. fragilis. The similarities in overall morphology shared by the two taxa leaves little doubt as to their phyloge- netic propinquity. Points of divergence include the much reduced gizzard in D. helonoma, differ- ences in the position of the seminal vesicles and the dorsal pores and the conspicuous develop- ment of genital seta glands in D. helonoma. Diplotrema proserpinensis sp. nov. (Figs 4, 5) ETYMOLOGY. Named for the type locality. MATERIAL. HOLOTYPE: QMG5449. Myrtle Creek, between Cannonvale and Proserpine, (20°23’S,148°36’E), CEQ; frequent in mud in the bot- tom of and in the bank of a stagnant backwater. Coll. B.G.M. Jamieson, 15 Jul 1966. PARATYPES 1-3: QMG2887, 2892. 2893, collection data as for holotype. OTHER MATERIAL; QMLGH2888, collection data as for holotype. DESCRIPTION. Length - 103, 82mm. Width = 1.5mm. Segments ^225, 175 (H, PI). Form cy- lindrical, anterior end slightly bulbous over sev- eral segments, pale in alcohol. Prostomium pro-epilobous, dorsal pores absent. Setae closely paired throughout, ventral setal couples of XVIII present, those of XVIl and XIX modified as en- larged penial setae, often protmding through the prostatic pores; those of VIII replaced by genital setae, situated on strongly protuberant papillae. Nephropores not externally recognisable. Clitellum not developed. S pores seen slightly presetally, lateral of ^-lines, in shallow seminal grooves that link the porophores of a side; pro- static pores 2 pairs, on medianly inclined dome- shaped papillae at the ab loci of XVII and XIX. The ventral body surface is depressed within the area circumscribed by the four prostatic papillae and extending as far as intersegment 16/17 and 19/20, its margins raised on each side as a longi- tudinal ridge along which the seminal grooves run. $ pores minute, presetal in a or lines of XIV. Spermathecal pores 2 pairs, in 7/8 and 8/9, seen as elliptical areas, in /)-lines. Accessory markings: large, glandular pad, ap- proximately circular in outline occupies most of XX between the a setae, and encroaches some- what into XIX. This marking is developed only in the holotype to any significant degree, very faintly in PI , and not seen in any other specimens examined. Septa: none strongly muscularised, but those an- terior to and excluding 9/ 1 0 somewhat thickened. Dorsal blood vessel single, last hearts in XIll; of all the commissurals in VII-Xlll, only those in XII and XIII are large and heart-like. Further vascular details not determinable owing to the small size of the species and pronounced bleach- ing. Gizzard large, in V; oesophagus narrow, moniliform, in VI-XIX, lacking pouching or 492 MEMOIRS OF THE QUEENSLAND MUSEUM FIG. 4. Diplotrema proserpinensis sp. nov. Male geni- tal field. Holotype (QMG5449). calciferous glands. Intestine commences in XIX or XX, a well-developed oesophageal valve pre- sent at 18/19 or 19/20; typhlosole absent. Nephridia present at least as far forward as VIII, stomate, their avesiculate ducts entering the pa- rietes in a single series on each side, in c-lines; phar>'ngeal tufting absent. Holandric: 2 pairs of very large, iridescent sper- matic funnels, and sperm masses, free in X and XI; seminal vesicles lobed, in XI and XII, those of XII filling the segment longitudinally, those of IX smaller. Prostatic glands slender tubes, those of XVll slightly tortuous, and extending into XIX or XX; those of XIX much smaller, tortuous, extending into XX; each with a slender duct. Penisetal tbilicles well-developed, each extending into its next posterior segment, four or more setae to a bundle. Tlie setae curved gently, or through as much as a semicircle, ornamented over the distal sixth by a series of widely-separated, minute, distally directed teeth, some or all of which are mem- bers of transverse circlets; the distal extremity of the setae are 'laterally’ widened, and tenuinally form a bulb with or without a small, apical point. Length mature seta = 0.75mm, midshaft diameter = 27.3pm (mean of 2). Ovaries, each with several strings of large oo- cytes, and funnels, present in XI 11, ovisacs absent. Spermathecae two pairs, in VIII and IX, the pos- terior pair almost twice as large as the anterior. Each with a large, somewhat irregular ampulla, and a stout duct, at the ectal end of which, me- dially, is a single, large, ovoid diverticulum which may be almost as large as the ampulla. Length left spermatheca of IX = 0.66mm. Large genital seta follicles present in VIII, apparently associated with a glandular manifestation; the setae fairly straight, the ectal 1/3 ornamented with a series of longitudinal scallops, giving a serrated appear- ance. Length mature seta = 0.75mm (mean of 3), midshaft diameter = 27.3pm. REMARKS. The absence of dorsal pores in this species may be connected with its semi-aquatic mode of existence, since this loss is also seen in the primarily limnic families Aliuroididae, Bi- wadrilidae, Almidae, Lutodrilidae, Spargano- philidae and the marine littoral megascolecid Pontodrilus (see Jamieson, 1988). It is presum- ably a response to the superfluity of coelomic fluid lubrication in aquatic habitats. D. corni- gravei, which also lacks dorsal pores, is not nec- essarily related to members of the D. fragilis species-group. The disappearance of dorsal pores in the former species, recorded 'from swampy ground’ (Jackson, 1931), has probably occurred convergently. Whereas the gut contents of D. helonoma con- sist almost entirely of soft, partially digested or- ganic matter, the upper alimentary tract of D. proserpinensis was found to contain, inter alia, numerous irregular quartz grains (especially in FIG. 5. Diplotrema proserpinensis sp. nov. Left sper- matheca of segment IX. Holotype (QMG5449). TWO NEW SPECIES OF DIPLOTREMA 493 the gizzard). A dietary divergence may therefore account for the differential development of the gizzard in the two species. LITERATURE CITED BLAKEMORE, R.J. 1997. Two new genera and some new species of Australian earthworms (Acantho- drilidae, Megascolecidae: Oligochaeta). Journal of Natural History 3 1 : 1785-1848. DYNE, G.R. 1978. A new species of Microscolex (Diplotrema) from New South Wales. Proceed- ings of the Linnean Societ>' of New South Wales 103(1): 37-41. 1984. Systematics and zoogeography of Australian megascolecoid earthworms. Unpubl. PhD thesis University of Queensland, St Lucia. JACKSON, A. 1 93 1. The Oligochaeta of Western Aus- tralia. Journal of the Royal Society of Western Australia 17: 71-136. JAMIESON, B.G.M. 1971. A review of the megascole- coid earthworm genera of Australia. Part II - The subfamilies Ocnerodrilinae and Acanthodrilinae. Proceedings of the Royal Society of Queensland 82(8): 95-108. 1988. On the phylogeny and higher classification of the Oligochaeta. Cladistics 4: 367-410. JAMIESON, B.G.M. & BENNETT, J. 1979. New spe- cies of Acanthodrilinae and a new genus of Pe- rionychini (Oligochaeta: Megascolecidae) from New Caledonia, their phylogeny and Zoogeogra- phy. Bulletin de la Musee National de Paris 4e ser., 1, 1979(2): 353-403. JAMIESON, B.G.M. & DYNE, G.R. 1976. The acan- thodriline earthworm genus Microscolex (Diplo- trema) in the Northern Territory of Australia Australian Journal of Zoology 24: 445-76. MICHAELSEN, W. 1907. Oligochaeten von Austral- ien. Abhandlungen aus dem Gebiete der Natur- wissenschaften, Hamburg. XXX Band, 1 Heft. 1910. Oligochaten von verschiedenen Gebieten. Mitteilungen aus dem Naturhistorischen Mu- seum in Hamburg 27, 2: 47-169. 1913. Die Oligochaten von Neu-Caledonien und den benachbarten Inselgruppen. Pp. 173-280. In Sarasin, F. & Roux, J. (eds) Nova Caledonia, Zoologie, vol. I L. HI, No. 5. (C.W. Kreidals: Weisbaden). 1916. Results of Dr E. Mjoberg’s Swedish scientific expedition to Australia 1910-1913. XIII Oligo- chaten. Stockholm Kungliga Svenska Vetenskap- sakademiens Handlingar 52( 1 3): 1 -74. 1921. Zur Stammgeschichte und Systematik der Oligochaten, insbesondere der Lumbriculiden. Archiv fur Naturgeschichte 86(1920), Abt. A. SPENCER, W.B. 1 900. Further descriptions of Austra- lian earthworms. Part I. Proceedings of the Royal Sociey of Victoria 13 (ns)(l): 29-67. STEPHENSON, J. 1930. The Oligochaeta. (Oxford University Press: Oxford). SWEET, G. 1900. On the structure of the spermiducal glands and associated parts in Australian earth- worms. Journal of the Linnean Society, Zoology 28: 109-139. 494 MEMOIRS OF THE QUEENSLAND MUSEUM FIRST RECORD OF BI LLER’S ALBATROSS IN QUEENSLAND. Memoirs of the Queensland Museum 42(2) 494. ] 99S> Buller’s Albatro.ss (Diomedea bulleri Rotlischild, 1893) has not previously been documented from Queensland (Blakersetal., I984;Slorr. 1984: Lindsey. 1986; Marchanl& Higgins, 1990), and is not represented in the eollections ofthe Queensland Museum (S. Van Dyck, pers. comm ). Further- more, 10 our knowledge, no sight records ofthe species exi.st for the Stale. We report here the first Queensland record of D. bulleri, based on a description and skull of a beach-washed specimen from North Stradbrokc I.. SFQ. During 1991, a dead albatross was found stranded on Frenchmans Bch, Point Lookout (27“25’S. 133“32’E) by a local resident, G. Blackburne. The collector examined the bird, which was in fresh condition, then removed the head to assist with identification. Features noted, including a grey head and yellow upper and lower bill margins, matched those described by Simpson (1976) for Buller’s Albatross (G. Blackburne, pers. comm.). [Although a separate account of/). bulleri does not appear in this publication, distinguishing characters for the species are provided in the text for the Grey-headed Albatross (/). chry^sostoma Forster, 1785)]. Notes detailing the appearance of the bird’s plumage and bare parts were not made, but the skull, without its bill plates, was retained. The skull wtis shown to ICG during May 1993. by which time the collector could not recall the exact date ofthe find. Because of its significtmce, the specimen was subsequently obtained and registered in the Queensland Museum as 030971. To confirm the original identification, it was for- warded to the Museum of New Zealand (MNZ) for comptiri- son with skulls of (he four Australasian mollymawk species in the same general size range, namely Buller’s, Grey-headed. Yellow-nosed (A chlororhynchos Gmelin. 1 789) and Black- browed {IX melanophris Temminck. 1828) Albatrosses. The Point Lookout skull matches specimens of Bullcr’s Albatross well. It is larger than D. chlororhynchos skulls, and ofthe two other species similar in size and shape to 030971, the skull of D. chrysostoma is usually larger apin and possesses a Hat section along the top of the culminicorn, whereas that of A melanophris has litrger nasal entrances and smaller orbital ridges (A. Tennyson, in litt., 3.10.97). Buller’s Albatross, a non-breeding migrant from New Zea- land, has been recorded often in SE Australia (Lindsey, 1986; Marchanl & Higgins, 1990), but is considered to be uncom- mon off the ea.st coast N of Eden (37®04’S, 149®54’E; D.W. Eades in Marchanl & Higgins. 1990). Holmes (1977) docu- mented two birds offCoffs I larbour (30®1 8’S, 1 53'’09’E) from 1974-1976, tmd Lindsey (1986) stales A bulleri is known N to Byron Bay, NSW (28“38’S, I53“38’E) but provides no details of specific records. The beach-washed specimen from Point Lookout thus extends the species’ distribution further northward. That A bulleri would eventually be added to Queensland’s avifauna was not unexpected given the rela- tively close proximity of its reported occurrence in northern NSW. The timing of available records from this region, e.g., .lune and July in Holmes (1977), suggests Buller’s Albatross may be an occasional winter straggler to southern Queensland. Acknowledgements GeotT Blackburne kindly provided the Buller’s Albatross skull iUid information about his initial discovery. The efforts of Alan Tennyson, Collection Manager (Birds) at the MNZ, in verifying the identity of the skull are greatly appreciated. We thank Rory' O’Brien (Museum of Victoria) for comments on the manuscript. Literature Cited BLAKERS, M., DAVIES, S.J.J.F. & REILLY. P.N. 1984. The Atlas of Australian Birds. (Melbourne University Press: Melbourne). HOLMES, G. 1977. The ecology of petrels in mid-northern New South Wales. Australasian Seabird Group News- letter 8: 20-35. LINDSEY, T.R. 1986. The Seabirds of Australia. (Angus & Robertson: North Ryde). MARCHANT. S, & HIGGINS, P.J. 1990. Handbook of Aus- tralian, New Zealand & Antarctic Birds. Vol. 1 , Ratites to Ducks, Part A. (Oxford University Press: Mel- bourne). SIMPSON, K. 1976. Grey-headed albatross. Pp. 36. In Frith, H.J. (ed.). Reader's Digest Complete Book of Austra- lian Birds. (Reader’s Digest Services: Sydney). STORR, G. 1984 Revised list of Queensland birds. Records ofthe Western Australian Museum, Supp. No. 19. I.C. Gynther, Department of Environment. EO Box 155, Al- bert St. Brisbane 4002, Australia David A. Stewart. Depart- ment of Environment. PO Box 42. Kenmore 4069, Australia: 15 December 1997. 57?ra/?/^C///5(PHIDOLOPORlDAE) AND RETELEPRALIA (CHEILOPORINIDAE): TWO NEW GENERA OF INDO-PACIFIC BRYOZOA DENNIS P. GORDON AND PETER W. ARNOLD Gordon, D.P. & Arnold, P.W. 1998 06 29: Btyorachis (Phidoloporidae) and Retelepralia (Cheiioporinidae); Two new genera of Indo-Pacific Bryozoa. Memoirs of the Queensland Museum 42(2); 495-503. Brisbane. ISSN 0079-8835. Two new genera of ascophorine Cheilostomalida arc recognised from northeastern Australia and the Norfolk Ridge. Biyorachis (family Phidoloporidae), with erect, unbranching or scarcely branching stems, has a colony form somewhat intermediate between that of Reteporellina and Chevron. Two species, both new, are ascribed to Bry>orachis. Retelepralia^ with a sole included species, also occurs in Mauritius. ITovisionally included in the Cheiioporinidae, it is characterised by relatively large, vcntricose zooids interconnected by short tubular processes. There is also a striated median gymnocystal strip frontally. □ Bryozoa, Btyorachis, Retelepralia. taxonomy, deep-sea. Dennis P. Gordon, National Institute of Water Atmospheric Research (NIWA), P.O. Box 14-901 Kilbirnie. Wellington. New Zealand: / land, 70-84 Flinders Street. Townsville 4810, In the course of the authors’ respective studies on the bryozoan faunas of the Norfolk Ridge and NE Australia, recent collections from deeper water have yielded new taxa common to both areas. These comprise two new species repre- senting a new genus of the lace-coral family Phi- doloporidae, and a third species, previously known only from Mauritius, representing a new genus of Cheiioporinidae. The new phidoloporid genus is only the second in the family occurring on soft sediments and appears significant in its unique colonial mor- phology as transitional between a shelf/slope ge- nus of rigid-erect species and a deep-sea genus whose species are found on terrigenous- foraminiferal oozes. The new cheiloporinid is unusual in the family, and among lepralioid asco- phorines generally, in having encrusting zooids linked by tubular connections. The purpose of this paper is to describe the new taxa, and assess their taxonomic and bio- geographic affinities. MATERIALS AND METHODS Material was collected in the course of various oceanic cruises. Specimens from the Norfolk Ridge were collected during cruises of New Zea- land Oceanographic Institute (NIWA) research vessels Taranui and Tangaroa. Specimens from the Coral Sea were collected during three Cidaris expedition cruises of R.V. Franklin. Selected specimens were soaked in domestic bleach (hypochlorite solution) to remove organic tissues from the zooidal skeletons. These were ^eter W. Arnold, Museum of Tropical Queens- Australia; 31 March 1997. washed thoroughly in water and dried prior to carbon or gold coating for scanning electron mi- croscopy. Dimensions were calculated from the scale bars in SEM photographs and checked by calibration against graticule measurements using light microscopy. SYSTEMATICS Class GYMNOLAEMATA Allman, 1856 Order CHEILOSTOMATIDA Busk, 1852 Family PHIDOLOPORIDAE Gabb & Horn, 1862 Bryorachis gen. nov. ETYMOLOGY- The generic name is derived partly from the phylum name Bryozoa. and rachis, a Greek feminine noun meaning backbone, alluding to the ap- pearance of the stems. TYPE SPECIES. Bry^orachis pichoni sp. nov. DIAGNOSIS. Colony erect, rod-like and un- branched, or with limited branching basally and laterally, with distinct frontal and ab frontal sur- faces. Stems 3-5-serial, the axial and adjacent longitudinal series of zooids opening frontally, additional series originating abfrontally but di- rected laterally. Zooidal peristomes projecting, the paired lateral ones giving the stems the ap- pearance of vertebral columns: peristomial ori- fices lack spines but may be denticulate. Primary' orifices suborbicular with a broad shallow sinus, the distal arch beaded. Autozooids with a de- pressed granular-tubercular area frontally in which a pair of areolar pores are recessed proxi- mally. Avicularia absent from the axial row of zooids, present adventitiously on most of the 496 MEMOIRS OF THE QUEENSLAND MUSEUM others except at proximal ends of stems, with short rounded or triangular rostrum. Ovicells cu- cullate, smooth and imperforate, widely open frontally and non-labellate. Abfrontal kenozooids small, disposed at regular intervals adjacent to the dorsal wal Is of the lateral zooids and often bearing small avicularia. Bryorachis pichoni sp. nov. (Figs 1,2 A, B) ETYMOLOGY. For the chief scientist on the Ci- daris expeditions, Professor Michel Pichon (Ecole Pra- tique des Hautes Etudes, Universite de Perpignan, France). MATERIAL. HOLOTYPE: QMG21274, J-shaped col- ony, Cidaris 3 Stn 6.4, 11 Nov. 1992, 10°00.82’S, 145°00.15’E, 1777m, Calypso dredge, coll. M. Pichon, R.A. Birtles. P.W. Arnold, in alcohol. PARATYPES; QMG21275, L-shaped colony, same locality as holo- type, mounted on SEM stub; QMG2I276, 4 addi- tional fragments, in alcohol. ADDITIONAL MATERIAL: Cidaris I Stn 20.3, 10 May 1986, 17°46.5’S, I47®48.8’E, 1223-1224m, Ockelmann epibenthic sledge, coll. M. Pichon, R.A. Birtles, P.W. Arnold, I specimen in alcohol (QMG21277), 1 mounted on SEM stub (QM21278). Cidaris 1 Stn 35. IE, 13 May 1986, 16®50.7^S, 147°08.rE, 1604m, Smith-Macintyre grab, coll. M. Riddle, D. Alongi, 1 specimen in alcohol (QM21279). Cidaris 3 Stn 6.3. 1 1 Feb. 1992, 10°01.2’S, 145°01.5’E, I777-1779m. Ockelmann epibenthic sledge, coll. M. Pichon, R.A. Birtles, P.W. Arnold, 1 specimen mounted on SEM stub(QM2I280). DESCRIPTION. Colony (known only from frag- ments) erect, rod-like, maximum length 8.0mm, generally unbranched but limited branching basally can create roughly L- or J-shaped colo- nies; 1 colony has a short lateral branch. Stems 3-5-serial, 0.54-0.69mm wide between peris- tomial projections, 1.9-2. 5mm at peristomial projections. Zooids in the axial longitudinal row 0.89- 0.98mm long and 0.33-0. 52mm wide. Frontal shield in these zooids with an extensive granular- tubercular depressed area, 0.38-0.53mm long and 0.21-0.33mm wide, at the proximal end of which is a pair of recessed areolar pores, 0.06-0. 09mm wide, best seen in oblique view. Peristomes sub- tubular, projecting, especially in lateral zooids, the secondary orifice 0.23-0.27mm diameter with an irregular rim; with labial pore (spiramen) and median suture inset in a groove. Primary orifice suborbicular, the distal arch of the anter beaded, with broad sinusoid poster; no condyles. Auto- zooids flanking axial series of zooids similar to them but the granular-tubercular area not signifi- cantly depressed and lacking areolar pores, the orifices facing about 40° away from main axis; both these zooids and the axial ones bear ovicells. Most lateral autozooids originate abfrontally at an oblique angle but the peristomes on each side are directed laterally in opposite pairs; labial pore and groove visible frontally, but also with a distinct groove and pore abfrontally; not seen to bear ovicells. Ovicells recumbent, cucullate, 0.25-0.35mm long and 0.28-0.35mm wide, smooth and imper- forate and widely open frontally; no labellum or lateral slits but the sides of the ovicell descending somewhat proximally. Avicularia adventitious, borne mostly on the zooids flanking the axial series, occasionally on the most lateral zooids, never on the axial ones; somewhat oval in shape with a variable rostrum (rounded-triangular to acute) and complete cross- bar, positioned adjacent to the peristomes of the axial autozooids. Abfrontal kenozooids roughly triangular, infill- ing axillary region between dorsal walls of auto- zooids; surface granular, usually with 2 small pores and occasionally an avicularium like the frontal ones. Ancestrula with globular base, orientated at right angles to main growth axis. Peristomes of the first 3-5 zooids (including ancestrula) directed to base of colony, with a reversal in orientation of subsequent zooids to form a single growth axis or short branches. REMARKS. Among the various specimens taken at different stations, the most significant variation is in the form of the avicularia. Specimens from Cidaris 1 Station 20.3 (17°46’S, 1223-1224m) have more acutely triangular mandibles on both frontal and abfrontal surfaces than in specimens from Cidaris 3 Stations 6.3 and 6.4 (c. I0°01’S, 1 777-1 779m). The opesial region of the avicular- ium is contained in a more prominently exserted base in specimens from Cidaris 1 Stn 20.3. FIG. 1. A-D, Bryorachis pichoni gen. et sp. nov. (registered material from the Coral Sea). A, frontal view of L-shaped paratype colony (QMG21275), x 23. B, central part of stem with ovicelled zooids, x 44. C, frontal view, tilled to show' the characteristic paired proximofrontal areolar pores, x 63. D, abfrontal view of distal end of stem showing arrangement of peristomes and avicularia, x 37; (B-D, QMG21278). NEW INDO-PACIFIC BRYOZOA 497 498 MEMOIRS OF THE QUEENSLAND MUSEUM Bryorachis curiosa sp. nov. (Fig. 2C,D) MATERIAL. HOLOTYPE. NIWA (N.Z. Oceano- graphic Institute) registration no. H-67L from NZOI Stn GIO. 32°09.5’S, 168°I5.0’E (southern part of Nor- folk Ridge NE of Wanganella Bank, just inside the Australian EEZ of Norfolk L). 970m, coll. 29 Sept. 1966 during the Noumea Cruise of R.V. Taranui. Only known specimen, coated frontally for SEM examina- tion, DESCRIPTION. Colony erect, rod-like, un- branched, the sole fragment 6.0mm long, the proximal end rounded. Stem 5-serial, 0.65- 0.71mm wide between peristomial projections, 1.1 9- 1.50mm at peristomial projections. Autozooids in the axial longitudinal row 0.77- 0.94mm long. Frontal shield of these zooids with a conspicuous granular-tubercular depressed area, 0.33-0. 39mm long and 0.27-0. 36rnm wide, at the proximal end of which is a pair of recessed areolar pores each up to 0.09mm wide, visible frontally but especially conspicuous when viewed obliquely. Zooidal peristomes subtubu- lar, scarcely evident on the axial zooids, most developed on the lateral ones, the secondary' ori- fice 0.22-0. 24mm wide with a U-shaped pseudos- inus that is not further constricted to become a closed labial pore and suture. Primary orifice suborbicular, the distal arch of the anter beaded, with broad sinusoid poster; no condyles. Auto- zooids flanking axial autozooids similar to them, granular-tubercular over most of the surface but lacking the depressed area and areolar pores; per- istomes facing slightly away from main axis. Ovicells not present. Avicularia adventitious, borne on most autozooids except the axial series, with a very narrow opesial region separated by a complete crossbar from the rounded rostrum; po- sitioned adjacent to the peristomes of the axial autozooids. Abfrontal surface granular-tubercular; keno- zooids somewhat pyriform to rounded- triangu- lar, situated on the 'backs’ of the most lateral zooids, adjacent pairs of kenozooids separated by an irregularly zigzag median groove coursing down the back of the stem. Each kenozooid typi- cally with an avicularium identical to the frontal ones. REMARKS. The sole colony fragment of B. cu- riosa is somewhat worn, and lacked membranes and soft tissue when collected. It may have been transported; the peristomes of all the zooids are eroded distally, although it apppears that the per- istomes of the axial series of autozooids did not project as far in life as those of the marginal zooids. Despite the condition of the specimen, it is clearly congeneric with B. pichoni. Both spe- cies share the following characteristics: same col- ony morphology, i.e., having the appearance of a vertebral column; the axial series of autozooids lack avicularia and have a depressed frontal area with two proximal areolar pores; regularly dis- posed abfrontal kenozooids with adventitious avicularia; same shape of primary orifice. Bryorachis curiosa differs from the type spe- cies in having a rounder, more conspicuous de- pressed area frontally, in some zooids bordered by an inner rim; one zooid has two depressed areas in one and it appears that this and the bor- dering rims in other zooids are the product of some kind of frontal budding, although for what purpose is unclear. It is conceivable that 1-2 rhi- zoids could issue from the areolar pores, descend basally, and anchor the colony to the substratum, but we have no evidence of rhizoids in either species. Other differences between the two spe- cies include the lack of a closed labial pore in B. curiosa and rounder avicularian rostra. ECOLOGY. B. curiosa is known from only one station, at 970m depth on the southern part of the Norfolk Ridge NE of Wanganella Bank. The sole specimen was taken in a 1 0-min pipe-dredge sam- pling which yielded a meagre collection of sand and gravel with pteropod shells, stylasterid hy- droids, solitary corals, and bryozoan fragments. Notwithstanding the small size of the collection, the many broken fragments give evidence of an impressive diversity of erect growth fonns among the several genera of stylasterids and bryozoans. The latter include Crassimarginatella vincularia Gordon (Calloporidae), Nellia tenella (Lamarck) (Quadricellariidae), Btyopastor sp. (Chaperii- dae), Poricellaria sp. (Poricellariidae), Atelesto- zoum sp . , Cellar ia s pp . , Euginoma s p . , Syringotrema sp. (Cellariidae), Siphonicytara sp. (Siphonicytaridae), Raxifahia sp. (Bifaxarii- I 'lG. 2. A-B. Bryorachis pichoni gcw. cl sp. nov. (registered material from the Coral Sea). A. elose-up ofovicelled /ooid ((9MG21 275). X 95. B. pdmary orifice, showing blunted distal beading and the shallow arcuate proximal rim, X 492. C-D. Bryorachis curiosa gen. et sp. nov. (registered material from NIWA Sin G 10. Norfolk Ridge). C'. part oflransported stem show ing arrangement of zooids and the subcircular depressed area ol an axial zooid with a secondary rim from apparent frontal budding. X 70. D, entire transported stem. X 17; (C-D. NIWA holotype 11-671). NEW INDO-PACIFIC BRYOZOA 499 500 MEMOIRS OF THE QUEENSLAND MUSEUM dae), Plagiopora recens Gordon (Catenicellidae), Kladapheles gammadeka Gordon (Lepraiielli- dae), Mawatariiis inexpectabilis (Gordon), Mawatarius sp. (Mawatariidae), and an undescribed genus of Celleporidae. The only encrusting or nodular bryozoans were species of Smittina (Smittinidae) and Celleporina (Celleporidae). The diversity of erect forms anchored by rootlets probably reflects the primarily unstable soft-sedi- ment habitat. The few encrusting or nodular spe- cies would be limited to small areas provided by gravel particles, pteropod shells, and fragments of corals and bryozoans. Family CHEILOPORINIDAE Bassler, 1936 Retelepralia gen. nov. ETYMOLOGY. The name is constructed of rete, a Latin neuter noun meaning net or network, and Lepralia, a well-known abandoned genus of cheilostomate br>'o- zoans from which the adjective lepralioid is derived. TYPE SPECIES. Lepralia mosaica Kirkpatrick, 1888. DIAGNOSIS. Colony encrusting, tending to pluriserial. Autozooids large, linked quincun- cially by short tubular processes. Frontal shield lepralioid, highly convex, evenly and densely pseudoporous, with or without a smooth thin ridge in the midline longitudinally. Orifice dis- tinctly divided into anter and poster, the latter broad, straight or barely convex, separated from anter by a pair of stout condyles. No oral spines. No avicularia. Ovicells relatively short, en- dooecium granular-tubercular and imperforate, ectooecium membranous, the combined orifice larger than that of autozooids and closed by the zooidal operculum. Retelepralia mosaica (Kirkpatrick, 1888) (Fig. 3A-D) Lepralia mosaica, Kirkpatrick. 1888: 79, pi. 8, fig. 6. Hippopodina mosaica, Hayward 1988; 319, fig. 6a,b. MATERIAL. NIWA: A small infertile colony from NZOI Station 190, 29°25.0’S, 168°05.6’E, Norfolk Ridge SE of Phillip I., 71m depth, coll. 23 July 1975 during the Norfolk Ridge Cruise of R.V. Tangaroa. Museum of Tropical Queensland: QMG2I28I. a rela- tively large fertile colony on SEM stub, from RV. James Kirby Stn 1467. 18®57’S, I46°58’E. mid-shelf, central Great Barrier Reef, 24m depth, coll. 9 Aug. 1986 using Ockelmann epibenthic sledge by R.A. Birtles and P. Arnold. The Natural History Museum, London, BMNH reg. no. 1934.10.6.20, Lepralia mosaica Kirkpatrick, 1888, from Mauritius. DESCRIPTION. Colony encrusting, forming relatively small planar patches less than 1 cm across, the zooids generally arranged quincun- cially or, depending on the substratum, somewhat pluriserially. Zooids large, 0.77-0.92mm long and 0.57-0.72mm wide, each connected to its neighbours bye. 5-7 short tubes around the latero- proximal margins basally; these separated by uniporous septula from the parent zooids. Frontal shield lepralioid-cryptocystidean, densely and evenly pseudoporous with small tubercles be- tween the pores; the shield arching convexly from the midline to join the basal wall on each side laterally; at the place where the frontal and basal walls meet may be a slight gymnocystal ridge, representing a thin, upwards extension of the basal wall. This ridge is continuous with a thin median strip of calcification that extends along the midline of each zooid towards the orifice; a series of transversely parallel ridges and grooves may occur along the median strip. Orifice longer (c. 0. 1 8mm) than wide (c. 0. 1 5mm), with a large, evenly rounded anter separated from a broad poster by a pair of stout condyles, the proximal rim of the poster gently convex or nearly straight. The orificial rim comprises a thin gymnocystal strip that is continuous with the one in the zooidal midline. No oral spines. No avicularia. Ovicell subglobose, relatively small compared to the overall volume of the autozooid, the en- dooecium granular-tubercular and imperforate, the ectooecium membranous. The frontal open- ing of the ovicell larger than autoozoidal orifices, widely arching above the anter of the maternal zooid such that its sides extend to the level of the condyles; thus the ovicell is fully sealed when the zooidal operculum is closed. The edge of the opening and the lateral margins of the ovicell are bordered by a thin gymnocystal strip where the membranous ectooecium attaches that parallels another on each side of the orifice coursing disto- laterally to the edge of the zooid. The ovicell illustrated (Fig. 3A,C) was raised and free of the substrate on a terminal autozooid so there could FIG. 3. A-D, Retelepralia mosaica (Kirkpatrick). A, several zooids of a colony, including one ovicelled zooid at the colony margin, x 67. B, close-up of a zooid showing well-developed striations along the median strip, X 10. C, close-up of ovicellular endooecium and fertile orifice, X 123 (A-C, QMG21281 from mid-shelf, GBR). D, three zooids, one quite worn, x 42 (unregistered colony from NIWA Stn 190, near Phillip 1., Norfolk Ridge). NEW INDO-PACIFIC BRYOZOA 501 502 MEMOIRS OF THE QUEENSLAND MUSEUM be no contribution to the frontal shield of the ovicell from a distally placed autozooid. Ancestrula unknown. REMARKS. Examination of part of Kirkpa- trick’s Mauritian material of Lepralia mosaica demonstrates conspecificity with our specimens. Whereas neither Kirkpatrick (1 888) nor Hayward (1988) illustrated or mentioned the median gym- nocystal strip on the frontal shield, it is neverthe- less present in the loaned Mauritian specimen. In zooids in which the outer epitheca is present, however, it can be difficult to detect, which ex- plains its not having been noticed previously. ECOLOGY. No information is available on the provenance of Kirkpatrick’s material other than that it was collected from Mauritian waters. In the Australasian region, Retelepralia mosaica occurs at shallow shelf depths. At NZOI Stn 190 SE of Phillip I., R. mosaica was taken at 71m depth by a medium Agassiz trawl from a bottom of bio- genic rubble dominated by a species of pectinid bivalve, with octocorals, gastropods and hermit crabs, Asterodisciis (Asteroidea), and a variety of echinoids. The sole colony of R. mosaica encrusts bryozoans and foraminiferans on a shelly substra- tum. The Barrier Reef material was collected from soft sediment with calcareous green algae, coralline-algal rhodolilhs, seagrass, and exten- sive biogenic rubble. The site had the greatest species richness and species diversity of molluscs and echinoderms of any along a cross-shelf tran- sect (Thirties & Arnold, 1988). Retelepralia mosaica grew on the skeleton of a discoidal foraminiferan Marginoporavertebralis dc Blain- ville, which was also overgrown by an unidenti- fied cyclostomate bryozoan and cheilostomate bryozoans of the genera Trypostega and Exechonella. The similarity of habitats between the GBR and NZ sites, particularly, and the failure to recollect R. mosaica on coral reefs of Mauritius (Hayward, 1 988) suggest that this species may be characteristic of biogenic rubble on an otherwise soft-sediment seafloor. DISCUSSION Bryorachis is a significant new addition to the family Phidoloporidae inasmuch as it has colonial and zooidal morphologies transitional between Reteporellina Harmer, 1933, a genus ranging from the shallow subtidal to shelf and slope depths, and Chexron Gordon, 1989, an unusual genus of small V-shaped colonies occurring on deep-sea oozes. Reteporellina comprises a num- ber of species of erect, mostly open-branched bryozoans, some fairly small. The colonies are characteristically fixed to a hard substratum. Among the characteristic features of the genus are the generally projecting subtubular peristome with an irregular rim and an open or closed pseudosinus, as well as the ovicell with conspicuous lateral processes and median labellum. The peristomial characters of Bryorachis accord well with those of Reteporellina, but not the ovicellular (or indeed colonial) ones. On the other hand, Bry’orachis has a significant number of characters in common with Chevron. These include: similar peristomial fea- tures (i.e., either a pseudosinus or closed labial pore and suture) and primary orifice, a frontal granular-tubercular area, similar adventitious avicularia, and, in one of the two Chevron spe- cies, an identical ovicell. We would argue, based on morphological congruence and geographical distribution, that Reteporellina represents the most generalised taxon of the three considered here, with increasing specialisation shown by Bry’orachis and Chevron. It would be useful to know precisely how colonies of the latter two genera are anchored to the substratum. Rootlets seem likely, but we have no evidence for these as yet. The affinities of Retelepralia are not certain. In the absence of ovicells, Hayward (1988) included Lepralia mosaica in the genus Hippopodina (Hippopodinidae). The absence of avicularia would be unusual in Hippopodina, which also has perforated ovicells, and, now that the ovicell is known, the combined characters seem more sup- portive of a relationship with the genera of Cheiloporinidae. These have a similar orifice and, typically, small or very reduced imperforate ovicells. On the other hand, the tubular connec- tions and median frontal gymnocystal strip in Retelepralia are hardly characteristic of the fam- ily so the question of affinity remains open. ACKNOWLEDGEMENTS We thank Mary Spencer Jones, The Natural History Museum, London, fora loan of Mauritian material of Retelepralia mosaica. and Bob Riel and Heather Winsor of the Electron Microscope Unit, James Cook University, for help to P.A. with SEM photography. Collection of Australian material was partly funded by Marine Science & Technology Research grants to Professor Michel Pichon (Universite de Perpignan) and Alastair Birtles (James Cook University). DPG’s research was carried out as part of NIWA’s Marine Tax- onomy Programme, funded by the N.Z. Founda- tion for Research, Science & Technology (Contract CO 1421). NEW INDO-PACIFIC BRYOZOA 503 LITERATURE CITED ALLMAN, G.J. 1856. A Monograph of the Freshwater Polyzoa, Including all Known Species, both Brit- ish and Foreign. (The Ray Society: London). BASSLER, R.S. 1936. Nomenclalorial notes on fossil and Recent Biyozoa. Journal of The Washington Academy of Sciences 26: 156-162. BIRTLES, R.A., & ARNOLD, P. 1988. Distribution of trophic groups of epifaunal echinoderms and mol- luscs in the soft sediment areas of the central Great Barrier Reef shelf. Pp. 325-332. In Choat, J.H. et al. (ed.). Proceedings of the Sixth International Coral Reef Symposium, Vol. 3. (James Cook University: Townsville). BUSK, G. 1852. An account of the Polyzoa, and sertu- larian zoophytes, collected in the voyage of the Rattlesnake, on the coasts of Australia and the Louisiade Archipelago. Pp. 343-402, pi. 1 (Ap- pendix No. IV). In MacGillivray, J. (ed.). Narra- tive of the Voyage of H.M.S. Rattlesnake, Commanded by the late Captain Owen Stanley ... 1846-1850; including discoveries and surveys in New Guinea and the Louisiade Archipelago, etc., to which is added the account of Mr E.B. Ken- nedy’s expedition for the exploration of the Cape York Peninsula [including Mr Carron’s narra- tive]. Volume 1. (T.W. Boone: London). GABB, W.M. & I lORN, G.FI. 1 862. The fossil Polyzoa of the Secondary and Tertiary formations of North America. Journal of the Academy of Natural Sci- ences of Philadelphia 5: 1 1 1-179, pis 19-21. GORDON, D.P. 1989. The marine fauna of New Zea- land: Biyozoa: Gymnolaemata (Cheilostomida Ascophorina) from the western South Island con- tinental shelf and slope. Memoirs of the New Zealand Oceanographic Institute 97: 1-158. HARMER. S.F. 1933. The genera of Reteporidae. Pro- ceedings of the Zoological Society of London 1933:615-627. HAYWARD, P.J. 1988. Mauritian cheilostome Bryo- zoa. Journal of Zoology (London) 215: 269-356. KIRKPATRICK, R. 1888. Polyzoa of Mauritius. An- nals and Magazine of Natural History, Series 6, 1 : 72-85. 504 MEMOIRS OF THE QUEENSLAND MUSEUM ASSESSMENT OF HORSEBACK SURVEYS OF KANGAROOS. Memoirs of the Oueenshnd Musem 42(2): 504. 19%:- Ground surveys of kiingaroos form the basis of many studies on kangaroo habitat use, population dynamics and social organisation. They provide estimates of population composition and size, and the latter can be used to calibrate aerial counts. Walked line transects have been shown by Southwell (1994) to return accurate estimates of population size. Less time-consuming surveys using vehicles along tracks are frequently biased and not always possible, but may. nev- ertheless, provide useful indices of kangaroo numbers. An alternative is to conduct surveys on horseback. Line transect methodology involves determining the dis- tances to objects from the transect line. The decay in sighta- bility away from tlie line is then modelled, allowing an estimate of density to be calculated. In order of importance, the technique carries four assumptions (Buckland et al., 1993); 1, Animals on the transect line are always seen 2, Animals do not move before being counted, and none are counted twice 3, Distances and angles are measured without error 4, Sightings are independent events. Compared with foot surveys, horseback surveys may be more likely to meet these assumptions because kangaroos may flush later and not as far, and the observer can concentrate more on sighting animals than on negotiating the transect. During Sept. 1993. foot and horseback surveys were com- pared along identical transects at two sites in central-western Queensland: Idalia NP (24“53'S. I44“46’H) and Terrick Ter- rick, a neighbouring sheep grazing property. At both sites, vegetation ranges from grassland to acacia woodland, with some areas of dense regrowth and fallen timber. On each site, four transects, each 5km long, were walked by a single ob- server navigating using a sighting compass. Transects on Idalia were walked twice to boost .sample size. For each transect, the two survey methods (horseback and foot) were separated by 24h. the order varying. Distances to kangaroos were determined with a rangefinder and angles from the transect line to each animal dctemiincd using a sighting compass, allowing perpendicular distances to be calculated. For each kangaroo species, data vvcrc pooled across each site for analysis iuid were tinalysed using the computer program DISTANCE (Buckland et al.. 1993) as described bv Clancy etal.(1997). The two survey methods returned similar estimates of density for the three common species counted (Red Kanga- roos. Kiacropus ntfus. Common Wallaroos. M. robusius. and Eastern Grey Kangaroos, M. gigantens) (Table 1 ). The dispar- ity for wallaroos on Terrick Terrick may be the result of heaping at zero distance during the horseback surveys, result- ing in an overly steep sightability curve and inflated density estimate. At such high densities, observer saturation and TABLE 1 . Density estimates (D) and associated sample sizes (n) for horseback and foot surveys at the two sites. Site Kangaroo species Horseback Foot D n D n Idalia Red 13.89 (143) 15.67 (104) Idalia Wallaroo 21.08 (116) 22.75 (111) Terrick Terrick Red 18.44 (84) 16.53 (68) Terrick Terrick Wallaroo 85.16 (371) 59.97 (202) Terrick Terrick Eastern Grey 7.81 (25) 8.59 (55) Idalia foot survey ooooooooooo <£> o FIG. 1 . Histograms of perpendicular distances of sightings of red kangaroos along transect lines conducted on foot and on horseback at Idalia NP. reactive movement can also lead to inaccurate estimates (Southwell, 1994). Though not quantified, kangaroos ap- peared to flush later and not as far when confronted by an observer on horseback compared w ith one on foot. Horseback surveys general ly yielded larger sample s izes (Table 1 ), which usually leads to improved modelling of sightability (Buckland et al., i993). This larger sample size is reflected in much flatter sightability curves for horseback surveys compared with foot surveys (Fig. I ), presumably resulting from the higher vantage point, Surv'cying on horseback was approximately twice as fast as walking on foot and observers were less fatigued. Such advantages of horseback surveys must be contrasted with the additional costs of transporting and maintaining horses, and having observ'ers competent with horses. .Some habitats will be unsuitable for horseback surveys, such as steep terrain and thick vegetation, and fences create barriers, requiring access to gates. Acknowledgements We thank Glen Edwards and John Michelle for acting as observers on foot and horseback surveys, respectively. Com- ments from Stuart Cairns improved an earlier draft. This work w as funded by Queensland Department of Environment. Literature Cited BUCKLAND, S.T.. ANDERSON, D.R., BURNHAM, K.P. & LAAKE, J.L. 1993. Distance Sampling. (Chapman and Hall: London). CLANCY, T.F., POPLE, A.R. & GIBSON. L A. 1997. Com- parison of helicopter line transects with walked line transects for estimating densities of kangaroos. Wild- life Research 24; 397-409. SOUTHWELL, C. 1994. Evaluation of walked line transect counts for estimating macropod density. Journal of Wildlife Management 58: 348-56. A.R. Pople, Department of Zoology. University of Queensland. St Lucia 4072, Australia O.IV. Maag. Queensland Depart- ment of Environment. Mail Sendee 394. Warwick 4370. Attstra- lia; 14 February^ 1 998. NEW SYNONYMS AND NOMENCLATURAL CHANGES IN THE ANT GENUS POLYRHACHISVK. SMITH (HYMENOPTERA: FORMICIDAE: FORMICINAE) RUDOLF J. KOHOUT Kohout, RJ. 1998 06 29: New synonyms and nomenclatural changes in the ant genus Polyrhachis Fr. Smith (Hymenoptera: Formicidae: Formicinae). Memoirs of the Queensland Museum 42(2); 505-531. Brisbane. ISSN 0079-8835. Twenty-eight former subspecies are raised to species: P. andromache semitestacea Emery; P. arcuata denselineata Viehmeyer; F. arcuspinosa waigeuensis Donisthorpe; P. atropos circum- data Viehmeyer; P. atropos tersa Viehmeyer; P. conops simpla Santschi; P. conops spinifera Stitz; P. continua procera Emery; P. costidata radicicola Dahl; P. dahli cincta Viehmeyer; P. hippomanes hortensis Forel; P. hippomanes lucidula Emery; P. labella obliqua Stitz; P. mayri obesior Viehmeyer; P. mucronata bismarckensis Forel; P. rasteUata goramensis Emery; P. relucens decipiens Roger; P. relucens Utigiosa Emery; P. saevissima romanovi Santschi; P. santschii campbelli Mann, P. schang alata Forel; P. schang excitata Viehmeyer; P. schang leviuscula Viehmeyer; P sexspmosa esuriens Emery; P. sumatrensvi striatorugosa Mayr; P. textor aequalis Forel; P. thrmax javanica Mayr and P. villipes noesaensis Forel. 47 synonyms of species-group names are proposed (senior names cited first): P andromache Roger = P. proximo Donisthorpe; P. orcw<3/a(LeGuillou)=P. arcuatacontinentis?oxe.\\P. awr/ra Emery = P. schlagmhaufeni Viehmeyer = P. schlaginhaufeni longispina Viehmeyer; P batesi Forel = P distincta Karavaiev; P bicolor Fr. Smith = P bicolor concolor Forel; P caecilae Forel = P. punctata Karavaiev; P carbonaria Fr. Smith = P orsyllus dentulata Stitz = P orsyllus subcarinata Emery; P. caulomma Viehmeyer = P caulomnia parallela Viehmeyer; P conops Forel =P conops stitziSanischi: P. continua Emery =P continua revocata Viehmeyer; P creusa Emery^ = P creusa distinguenda Karavaiev; P cyrus Forel = P urania Forel; P decipiens Roger =P restiti^ta Viehmeyer = P. restitutaconclusa Viehmeyer; P. dives Fr. Smilh=P dives shviensis Santschi; P. doddi Donisthorpe = P townsvillei Donisthorpe; P esuriens Emery = P arcuspinosa Donisthorpe = P jiixtaspinosa Donisthorpe; P goramensis Emery = P. coronala Santschi; P hermione Emery = P. daemeli sulcativentris Forel; P leviuscula Viehmeyer = P schang laurae Menozzi; P limbata Emery = P verticalis Santschi; P Utigiosa Emery = P sericata pruinosula Santschi; P marginata Fr. Smith = P rotimdiceps Karavaiev; P melpomene Emery - P dolichocephala V iehmeycr; P moesta Emery = P paromalus tobias Forel ; P nigropilosa Mayr = P nigropilosa conophfhalma Emery; P niimeria Fr. Smith = P schizospina Karavaiev; P. obtusa Emery = P arcuata ariiana Karavaiev; P queenslandica Emery' = P. dahli unisculpta Viehmeyer; P radicicola Dahl = P karawaicM'i Santschi; P. rLxosa Fr. Smith = P lycidas Fr. Smith; P rufofemorata Fr. Smitli = P. biroi Forel = P biroi bidentata Stitz ^ P biroi paprika Forel; P sculpturata Fr. Smith = P. sumatrensis hamulata Emeiy; P sericeopubescens Donis- thorpe = P inusitata Kohout; P solmsi Emery = P. alexandri Karavaiev; P. spinifera Stitz = P continua hirsutula Emery; P striata Mayr = P. striata assamensis Forel; P striatorugosa Mayr = P striatorugosaexophthalmaVore\;P. vestitaPr. Smith=P vestitaunicolorPmery fP. vigilans Fr. SmiHi = P. pyrgops Viehmeyer and P villipes Fr. Smith =P sumatrensisVr. Smith=P striata tritschleri Forel. P. hosei Donisthorpe, formerly synonymised with P inermis Fr. Smith, and P mentor Forel, formerly synonymised with P limbata Emery, arc reinstated as valid species. P olybria Forel, originally described from a queen and placed in the subgenus Myrmhopia Forel, is recognised as a member of the nominal subgenus Polyrhachis and the identity' of the worker caste established. Material of three unavailable names is referred to valid species (valid names cited first): P. andromache Roger for P sericata pruinosida harmsi ’ Karavaiev; P.foreli Kohout for ‘P. relucens decipiens papuana ' Emery and P. obliqua Stitz for ‘P relucens Utigiosa aloseana ’ I 'orel. I'ypes of P bihamata minor Karavaiev, P hooliri Lowne. PJurii Karavaiev, P pallescens Mayr and P ypsilon synacantha Santschi, which were previously reported inaccessible or missing, presumed lost, have been located and their specific identity, including previously proposed synonymy, confirmed. P. eudora Fr. Smith and P. latreillei (Guerin-Meneville) are declared species inquirendae. Lectotypes are designated for P creusa distinguenda Karavaiev and P hermione Emery. □ Hymenoptera, Formicidae, Polyrhachis, taxonomy, nomenclature, synonymy. Rudolf J. Kohout, Queensland Museum, PO Box 3300, South Brisbane 4101, Australia; 7 June 1997. 506 MEMOIRS OF THE QUEENSLAND MUSEUM During visits to European and British museums in 1990, partly funded by an Australian Entomo- logical Society Research/Travel Grant (Kohout, 1991),! was able to study a number of Polyrhachis types, including some that were previously con- sidered lost. These studies, with those of a later European survey in 1992, revealed a number of species-level synonyms. Some, relevant to the Australian fauna, have been discussed elsewhere (Kohout, 1994; 135). In 1994 I visited the Zoo- logical Institute of the Ukrainian Academy of Sciences in Kiev in order to examine the little- known types of Polyrhachis species described by Karavaiev. Their subsequent comparison with types of species formerly described by Emery, Forel, Fr. Smith, Stitz and Viehmeyer helped to clarify their specific identity and to unravel their extensive synonymies. The following nomenclatural changes were compiled too late for inclusion in the catalogues of Bolton (1995) or Dorow (1995). I extend sin- cere thanks to both these colleagues for making their then unpublished computer-based manu- scripts available for this study. The treatment of species is arranged alphabeti- cally within relevant subgenera and/or species- groups, which generally follow the classification of Dorow (1995). Despite some criticism (Hung, 1967) the subgenera oi Polyrhachis provide use- ful foci for sorting of species in this otherwise large and difficult genus (Kohout & Taylor, 1 990: 510; Dorow, 1995: 7). Publication dates and the spelling of species epithets and author’s names follow Bolton (1995). The measurements (in mm) and indices follow those of Kohout (1990: 499): TL = Total length (the neccessarily composite measurement of the entire ant); HL = Head length (the maximum meas- urable length of the head in perfect full face view, measured from the anterior-most point of the clypeal border or teeth, to the posterior-most point of the occipital margin); HW = Head width (width of the head in perfect full face view, measured immediately in front of the eyes); Cl = Cephalic index (HW x 100/HL); SL = Scape length (length of the antennal scape, excluding its condyle); SI = Scape index (SL x 100/HW); PW = Pronotal width (width of the pronotal dorsum measured at the bases of the pronotal spines, or across the humeri in species without spines); MTL =Metathoracic tibial length (maximum measurable length of the tibia of the hind leg). Where a holotype is described herein as ‘unique’, this infers that there was only a single specimen available for that species description. Abbreviations for institutions (with the names of cooperating curators) are: ANIC = Australian National Insect Collection, Canbeira, Australia (Drs R.W. Taylor, S.O. Shattuck); BMNH = Natural History Museum, London, U.K. (Barry Bolton); BPBM = Bernice P. Bishop Museum, Honolulu, Hawaii, USA (Dr G.Vl. Nishida); BSMP = Bureau of Science, Manila, Philippines; CAS = California Academy of Sciences, San Francisco, CA, USA; HNHM = Hungarian Natural History Museum, Bu- dapest, Hungary (Dr J. Papp); DEIE = Deutsche Entomologische Institut, Eberswalde, Germany (Dr A. Taeger); IRSN = Institut Royal des Sciences Naturelles de Belgique, Brussels, Belgium (Drs P. Grootaert, P. Dessart); IZAS = Institute of Zoology, Ukrainian Academy of Sciences, Kiev, Ukraine (Dr A.G. Radchenko); MCSN = Museo Civico di Storia Naturale ‘Giacomo Doria’, Genova, Italy (Drs R. Poggi, V. Raineri); MHNG = Museum d’Histoire Naturelle, Geneva, Switzerland (Drs C. Besuchet, I. LobI); MNHN = Museum National d’Histoire Naturelle, Paris, France (Dr J. Casevitz Weulersse); MNHU ^ Museum fur Naturkunde, Humboldt- Universitat, Berlin, Germany (Dr F. Koch); NHMB = Naturhistorisches Museum, Basel, Switzerland (Dr M. Brancucci); NHMW = Naturhistorisches Museum, Wien, Austria (Drs M. Fischer, S. Schodl); NRMS = Naturhistoriska Riksmuseet, Stockholm, Sweden (Drs K-J. Hedqvist, F. Ron- quist); OXUM = University Museum, Oxford, UK (Dr C. O’Toole); QMBA = Queensland Museum, Brisbane, Australia (Drs G.B. Monteith, C.J. Bur- well); RMNH = Nationaal Natuurhistorisch Mu- seum, Leiden, The Netherlands (Dr Ing. C. van Achterberg); SMF = Forschungsinstitut Sencken- berg, Frankfurt am Main, Germany (Dr J.-P. Kopelke, W.H.O. Dorow); UDSB = Universita De- gli Studi Di Bologna, Italy (Prof Dr Maria M. Principi); USNM = US National Museum, Wash- ington, DC, USA (Dr D.R. Smith); ZSSM = Zoolo- gische Staatssammlung, Munchen, Germany (Dr E. Diller). SYSTEMATICS Genus Polyrhachis Fr. Smith, 1857 Subgenus Campomyrma Wheeler, 1911 Polyrhachis creusa Emery, 1897 Polyrhachis creusa Emery. 1897: 577. Holotype worker. Type locality: New Guinea, Ighibirei (L. Loria), MCSN (examined). Polyrhachis (Campomyrma) creusa distinguenda Karavaiev, 1927: 7, fig. 1. Syntype workers. Type locality: Indonesia, Ambon I. (W. Karavaiev #3007), IZAS, QMBA (examined). Syn. nov. NEW SYNONYMS IN THE ANT GENUS POLYRHACHIS 507 (For full synonymy citations see Kohout & Taylor, 1990: 514). REMARKS. Karavaiev described P. creusa dist- inguenda from five workers ‘samtlich einzeln aber nahe von einander gesammelt’. The original series comprised specimens in two distinct size classes, but Karavaiev considered them conspeci- fic, since he described them as ‘major’ and ‘mi- nor’ workers of a single species. However, examination of four syntypes shows that the type series includes two distinct species. The ‘major’ workers of the P. creusa distinguenda syntype series (#3007) resemble the syntypes of P. creusa chlorizans Forel from Ralum, Bismarck Archi- pelago so closely that they appear to be part of the same series. Kohout & Taylor (1990: 514) synonymised P. creusa chlorizans with P. creusa Emery, and this synonymy is now confirmed. The ‘minor’ workers of the P. creusa distinguenda series (#2452) are, on the other hand, identical to voucher specimens (ANIC, QMBA) compared formerly by Taylor (1989: 25) with the holotype of P. insularis Emery. I consider them conspecific. LECTOTYPE DESIGNATION. I have designated a ‘major’ worker specimen from the Karavaiev collection (IZAS) as the lectotype of P. creusa distinguenda and the other ‘major’ specimen (QMBA) as paralectotype. Both are in good con- dition. They bear identical sets of three labels, as follows: - Amboina, Karavaiev, 3007; - 3007. Coll. Karavaiev; - Syntypus ? Polyrhachis creusa distinguenda Karaw. (red tag). Subgenus Myrmatopa Forel, 1915 Polyrhachis alata Forel, 1904 stat. nov. Polyrhachis gracilis x,alataYoxt\, 1904: 177. Syntype workers. Type locality; Indonesia, Sulawesi, Patuhuang (H. Fruhstorfer), MHNG (examined). Polyrhachis gracilis Emery; Ford, 1909: 232. Junior synonym of P. schang Forel. Polyrhachis schang var. alata Forel; Emery, 1 925: 181. REMARKS. Direct comparison of a P. alata syntype with the syntypes of P. schang Forel and P. gracilis Emery shows them to be closely re- lated. The two latter taxa are undoubtedly conspe- cific, but the general appearance of the P. alata type sets it well apart. The P. gracilis syntypes are more-or-less uniformly brown, with only the mouth parts, pronotal collar, appendages, and apical portion of the gaster a shade lighter. The P. alata syntype is distinctly bicoloured — mostly light yellowish-brown, with the occiput, lateral and ventral portions of the head, antennae (except the apical funicular segments), lateral portions of mesosoma, front coxae, legs (except trochanters) and most of the petiole and gaster diffusely blotched medium to dark reddish-brown. The masticatory margins of the mandibles, denticles of the truncated median portion of the anterior clypeal margin, frontal carinae, mesonotal and lateral propodeal margins and the petiolar spines are narrowly bordered with very dark brown. This conspicuous colour pattern is particularly distinct in recently collected specimens (Sulawesi Utara, Dumoga-Bone Nat. Park, nr Danau Mooat, 1 100m, 1 viii 1985, N. Stork). Besides its colour, the P. alata type differs from the other relevant types in its larger size (HL 1.53-1.79 in P. alata versus HL 1.40-1.50 in P. gracilis), with higher and more acute lateral margins to its mesonotum, and longer, more slender petiolar spines. Polyrhachis excitata Viehmeyer, 1913 stat. rev. Polyrhachis excitataVKhxm^yQX, 1913: 147, fig. Holo- type worker. Type locality: Indonesia, Sulawesi (in copal), MNHU (examined). Polyrhachis schang var. excitata Viehmeyer; Viehmeyer, 1914:48, fig. 8. REMARKS. P. excitata stands closest to P. ulys- ses Forel and shares with that species distinct, posteriorly raised lateral margins to the mesono- tum, and relatively long petiolar spines. However, the P. excitata type differs from P. idysses speci- mens in a number of characters, including the strongly raised frontal carinae, the more convex and shorter propodeal dorsum, the less divergent petiolar spines, and rather dense sculpturation of the head, mesosoma and gaster. Sculpturing in P. ulysses in distinctly finer, notably on the head and over the anterior portion of the pronotal dorsum. The characters separating P. excitata from P. schang are even more pronounced. They include the lack of highly raised frontal carinae in P. schang, its generally finer sculpturation, lighter colour and distinctly smaller size. I consider that Viehmeyer’s action in reducing P. excitata to a subspecies of P. schangv'pe queen. Type locality: Burma, BMNH (examined). Polyrhachis bicolor var. concolor Forel, 1910: 129. Syntype workers, queen, male. Type locality: Phil- ippines, Luzon, Manila (C.S. Banks), BSMP (types destroyed - see Baltazar. 1966: 282^ MHNG (ex- amined). Syn. nov. REMARKS. P, bicolor was first reported from Australia (Northern Territory, Darwin) by Clark (1941: 69) and subsequently included by Kohout & Taylor (1990: 51 1) in their list of Australian Polyrhachis s^QCXQS. The identification of Clark’s specimens has, however, never been verified. Re- cent collecting (by A.N. Andersen, H. Reichel, R.J. Kohout) has provided abundant additional material, including several queens. Their com- parison with the holotype queen of P. bicolor (BMNH) finally establishes that the Australian specimens are conspecific with that species. Sub- sequent study of several infraspecific forms of P. bicolor in the Forel collection (MHNG) demon- strates that the characters used by Forel to sepa- rate P. bicolor concolor diwd P. Z)/co/or are trivial. I consider them as representatives of a single species. In the course of this study 1 have also examined those infraspecific taxa of the P. bicolor-com\Aex described by Karavaiev; namely A bicolor atro- castanea,P. bicolor aurata,P. bicolor erecta 2 ind P. bicolor weyeri. I consider that most, if not all, represent valid species. However, the P. bicolor species-group, as delimited by Dorow ( 1 995; 49), contains a number of other unresolved taxa and their treatment is beyond the scope of this study. Resolution of the complex will require detailed examination of all specific and infraspecific named entities, types of which are not presently available. Their taxonomy will therefore be dealt with separately. Polyrhachis dives species-group Polyrhachis dives Fr. Smith, 1857 Polyrhachis dives Fr. Smith, 1857: 64. Holotype worker. Type locality: Singapore (A.R. Wallace). BMNH (examined). Polyrhachis (Myrmhopla) dives var. siwiensis Sant- schi. 1932: 20. Holotype queen. Type locality: New Guinea. Siwi (Foret) (Prince Leopold), IRSN (ex- amined). Syn. nov. (For full synonymy citations see Bolton. 1995: 347 and Dorow. 1995:52.) REMARKS. I have compared the holotype queen of P. dives siwiensis with numerous P. dives queens from a wide distributional range, and be- lieve that the characters given by Santschi do not justify the separate status of the former as a sub- species of P. dives. The slightly larger head and the rather short propodeal spines sets the queen of P. dives siwiensis somewhat apart, but these dif- ferences seem to be taxonomically insignificant. In fact, the holotype compares relatively well with a somewhat aberrant queen collected in N Qld in a nest of 'normaF P. dives workers. I am therefore confident in considering the holotype queen of A dives siwiensis to be a variant individ- ual, and consequently a synonym of A dives. Polyrhachis mucronata species-group Polyrhachis batesi Forel, 1911 Polyrhachis batesi Forel, 1911:301. Holotype worker. Type locality: Indonesia, Seram L (Bates), ZSSM (examined). Polyrhachis (Myrmhopla) distincta Karavaiev, 1927: 40, llg. 18. Syntype w'orkers. Type locality: Indone- sia, Ambon I. (W. Karavaiev), IZAS, QMBA (ex- amined). Syn. nov. REMARKS. This is a straightforward synonymy. Karavaiev was evidently unaware of Forel’s prior available species name. Polyrhachis bismarckensis Forel, 1901 stat. nov. Polyrhachis mucronata var. bismarckensis Forel, 1901: 33. Holotype worker. Type locality: Bismarck Archipelago, Wunakokur (= Vunakokor [Mt Var- zin], New Britain) (F. Dahl), MNHU (examined). 516 MEMOIRS OF THE QUEENSLAND MUSEUM REMARKS. Comparison of the P. mucronaia and P. mucronata bismarckensis holotypes clearly shows them to be representatives of dis- tinct, separate species. The pronotal dorsum is relatively wide, with less prominent humeral teeth and distinctly more divergent propodeal spines in P. bismarckensis. However, the most obvious character separating these taxa involves the structure of their petioles. In P. mucronata the dorsolateral petiolar spines are rather slender and evenly curved, embracing the base of the first gastral segment. They have between them a pair of very short, posteriorly directed intercalary teeth. In P. bismarckensis the dorsolateral spines are relatively thick (thought not as massive as in P. lucidula Emery or P. ridleyi Forel — see below) and slightly curved at their bases; posteri- orly the spines are almost straight, though slightly divergent, with their extreme apices distinctly bent outwards. The acute, obliquely rised interca- lary teeth of P. bismarckensis are distinctly longer than those of P. mucronata. Polyrhachis hortensis Forel, 1913 stat. nov. Polyrhachis hippomams var. hortensis Forel, 1913: 138. Syntype workers, queen, males. Type locality; Indonesia, Java, Buitenzorg (= Bogor) (H.v. Biittcl- Reepen), MHNG (examined). REMARKS. Comparison of the P. hippornanes and P. hippornanes hortensis types shows them to differ widely in a number of characters, including a distinctly more slender body and relatively much longer propodeal and petiolar spines in P. hippornanes. The head and mesosoma are also more densely punctate and opaque than in P. hippornanes hortensis., where the dorsal sculptur- ing is distinctly more shallow, with some shine evident on the occiput, at the bases of the propodeal and petiolar spines and over the gastral dorsum. The propodeum in P. hippornanes hortensis is relatively narrow, but less so than in P. hippornanes. It is armed with rather short, straight, divergent spines. The petiole of P. hip- pomanes hortensis has the lateral spines distinctly shorter, and there are two very short intercalary teeth, which are lacking in P. hippornanes. I am confident on these bases that P. hortensis repre- sents a distinct species separate from P. hippo- manes. Polyrhachis lucidula Emery, 1893 stat. nov. Polyrhachis hippornanes lucidula Emery, 1893: 254 (footnote). Holotype worker. Type locality: Burma, Tenasserim, Malewoon (L. Fea). MCSN (exam- ined). REMARKS. I have directly compared the holo- types of P. hippornanes and P. hippornanes lu- cidula and consider them to represent separate species. They differ in general appearance and in a number of other characters, including a dis- tinctly more slender body, with relatively thin and long spines in P. hippornanes. The head and mesosoma are also rather densely punctate and opaque. In P. lucidula\he body is wider, superfi- cially sculptured and shiny. P. lucidula is obvi- ously closely related to P. ridleyi Forel, and like that species it has remarkably massive petiolar spines. However, these two taxa differ in general appearance, with P. lucidula being consistently larger and considerably more shiny than P. ridleyi. Examination of modem specimens from Suma- tra and Java, which are comparable to P. lucidula, demonstrates some variability, notably in size. Even the largest specimen fails to attain the size of the holotype (HE i.47 versus 1.53). However, all these specimens share an apparently distinc- tive, shiny appearance and colour pattern, involv- ing a black body, dark brown antennae and light reddish-brown legs. In spite of their smaller size, the Sumatran specimens are very similar to the P. lucidula holotype, and I consider them to repre- sent a conspecific population. Polyrhachis moesta Emery, 1887 Polyrhachis hippornanes var. moesta Emery, 1887; 237. Holotype worker. Type locality: Indonesia, Sumatra, Sungei-Bulu (O. Beccari), MCSN (exam- ined). Polyrhachis paromalus tobiasVox^\. 191 1: 391. Holo- type worker. Type locality: Indonesia, Sumatra (Moesch), MHNG (examined). Syn. nov. Polyrhachis moesta Emery; Wang & Wu, 1991: 599. Raised to species. REMARKS. Forel noted the similarity between P. paromalus tobias and P. hippornanes moesta, but he had evidently not compared relevant speci- mens. 1 have compared the holotypes of both taxa and confidently declare them to be conspecific and taxonomically inseparable, despite some triv- ial differences, such as fractionally smaller hu- meral teeth and slightly less elevated propodeal spines in the P. paromalus tobias holotype. Polyrhachis armata species-group Polyrhachis caecilae Forel, 1912 Polyrhachis caecilae Forel, 1912: 76. Holotype worker. Type locality: Indonesia, Sumatra, In- drapura (Tritschler), MHNG (examined). NEW SYNONYMS IN THE ANT GENUS POLYRHACHIS 517 Polyrhachis (Myrmhopla) punctata Karavaiev, 1927; 36. Hololypc worker. Type locality: Indonesia, Java, Buitenzorg (Karavaiev #2893). Syn. nov. REMARKS. The holotype of P. punctata is con- specific with that of P. caecilae. The only appar- ent difference between them is the somewhat more solidly built m.esosoma, with more down- ward curved pronotal spines, in P. punctata. The dorsal edge of the petiole between its lateral spines is furnished with three subequal interca- lary teeth, with the middle tooth situated dis- tinctly more forward than the other two, fonning in dorsal view a clearly defined inverted W’. The colour of the body and appendages of these holo- types is essentially identical except for the gaster, which is distinctly more reddish in P. caecilae. The specific epithet stands as P. caecilae in the published original description, and that is for- mally the correct name for this species. The tag on the holotype, however, reads 'caeciliae\ and P. caecilae has been listed incorrectly under that name by all subsequent authors, including Emery ( 1 925 : 1 94), Dorow ( 1 995 : 47) and Bolton ( 1 995 : 345). P. caecilae is closely related to P. gestroi Em- ery (1900: 714) and P. gestroi rufiventris Forel (1911:391). However, as Forel (1912:77) noted in the original description, P. caecilae differs from both species in having 3 intercalary teeth on the petiole, while there are only 2 in P. gestroi Emery and P. gestroi rufiventris Forel. Polyrhachis romanovi Santschi, 1928 stat. nov. Polyrhachis (Myrmhopla) acantha dichroa Karavaiev, 1927: 33. Holotype worker. Type locality; Indone- sia, Java, Bantam (W. Karavaiev), iZAS (exam- ined). Junior primary homonym of P. dichroa Forel. 1893. Polyrhachis (Myrmhopla) acantha st. romanovi Sant- schi, 1928: 139. Replacement name. Polyrhachis saevissima romanovi Santschi; Dorow, 1995. REMARKS. 1 have examined and directly com- pared the unique holotypes of P. romanovi and P. saevissima Fr. Smith, 1860, together with the types oiP. acantha Fr. Smith, 1 860 and P. acasta Fr. Smith, 1 860, both of which were synonymised with P. saevissima by Bolton (1974: 178). I con- clude that P. romanovi is obviously related to P. saevissima and bears a close resemblance, which is most direct with specimens meeting the P. acasta diagnosis. This similarity is accentuated by the almost identical colour pattern of the rele- vant specimens. However, the two species differ in several details, including the outline of the mesosoma, which is relatively high and evenly curved in P. romanovi^ while in P. saevissima and its cohorts (see below) the profile from the sum- mit towards the propodeal spines is more-or-Iess sinuate, or almost straight, with a shallow depres- sion at the promesonotal suture. The pronotal spines of P. romanovi are somewhat longer and more slender than those of P saevissima, while the propodeal spines are shorter and the posterior face of the petiole distinctly more convex, in fact almost 'inflated’. The colouration ofP. romanovi is distinctive, with the gaster and appendages light reddish-brown, in contrast to P. saevissima, specimens of which are totally black, apart from those formerly assigned to P. acasta (see above and under P. saevissima below). Comparison of the above specimens with recently collected ma- terial shows that these characters, including the colour scheme, are constant in specimens throughout the whole range of their distribution. It thus appears that assignment of specific rank to P. romanovi is appropriate. I have also studied the types of P. saevissima, P. acantha and P. acasta (excluding their infras- pecific forms) together with modem material from Sulawesi. Comparison shows P. saevissima to be relatively somewhat more slender than the others, with a distinctly narrower mesosomal dor- sum (notably across the bases of the pronotal spines). The propodeal spines are also relatively more slender, and stand somewhat isolated. In P, acantha and P. acasta the bases of the spines are more broad and more-or-Iess medially contigu- ous. In profile the propodeal spines in P. acasta are strongly raised and virtually straight, while they are distinctly less elevated in the other two taxa. Those of P. saevissima are almost straight, while those of P. acantha are gently and evenly bent downwards. The petiolar spines in P. acasta closely embrace the first gastral segment, while in P. acantha the semicircle they form is dis- tinctly more narrow. It is even more so in P. saevissima. Uniformly reticulate sculpturation of the mesosomal dorsum in P. acantha and P. acasta is almost hidden by abundant appressed pubescence, but this is rather dilute in P. saevis- sima. The colour of the body in these taxa is black, with the legs very dark brown to black, except in P. saevissima, where the apical portions of the tibiae are a shade lighter, and in P. acasta, where the trochanters, femora and tibiae are reddish- brown. 1 consider these three taxa to be very similar and almost certainly closely related. They appear to form numerous interlocking popula- tions throughout their shared distributional range. As a result of these studies, and despite some 518 MEMOIRS OF THE QUEENSLAND MUSEUM reservations, I currently follow Bolton (1974: 178) in considering/^, saevissima, P. acantha and P. acasta to be putatively conspecific. I believe that the taxonomy of these entities could be com- petently resolved only following detailed study of worker-associated queens and consideration of the various (currently unavailable) relevant types of infraspecific forms. Bolton (1974: 179) also suggested that the pu- tatively closely related P. diaphanta Fr. Smith, 1 86 1 could be a junior synonym of P. saevissima. However, concurrent examination of the relevant types has shown P. diaphanta obviously to be a distinct species. Its most characteristic feature is the sculpturation of the mesosoma, which is rather densely and deeply rugose-punctate, while merely uniformly reticulate in P. saevissima. The pronotal spines in P. diaphanta are short and thick, hardly longer than wide, while in P. saevis- sima they are distinctly more slender, and almost twice as long as wide at their bases. In P. dia- phanta the mesosoma in profile rises strongly to the centre of the pronotum, then descends gently in a virtually straight, uninterrupted line towards the abrupt, almost vertical declivity. In contrast the outline of the mesosoma in P. saevissima is more-or-iess sinuate, with a shallow depression at the promesonotal suture, and a gently raised mesonotum. The propodeal spines in P. dia- phanta are relatively long and widely divergent, almost horizontal in profile and gently curved downwards. Their broadly contiguous bases bluntly divides the propodeal dorsum from the declivity. The petiolar dorsum bears a pair of strong, horizontal, relatively long, semicircular spines and two short, acute, dorso-posterioriy directed intercalary^ teeth. The petiolar spines in P. diaphanta embrace the first gastral segment, while in P. saevissima the semicircle formed by the spines is much narrower. On the inner surface of each spine in both species there is a distinct furrow, which runs from the base, adjacent to the intercalary teeth, almost to the tip. The colour of the body in P. diaphanta is black, with the legs very dark brown to black, in contrast to P. saevis- sima, where the colour of the legs ranges from relatively light reddish-brown to dark brown or black, with the tibiae a shade lighter. The anten- nae in P. saevissima are uniformly very dark brown or black, while in P. diaphanta Xht funicu- lus is distinctly bicoloured. The first segment is black, save for a narrow, light, apical band. The width of such light coloured bands progressively increases through the series of more distal seg- ments, with the five apical segments entirely light yellowish-brown. P. diaphanta appears to be rather uncommon. I have seen only one other specimen closely comparable to its type. Polyrhachis sexspinosa species-group Polyrhachis esuriens Emery, 1897 stat. nov. Polyrhachis sexspinosa var. esuriens Emery, 1897; 591. Syntype workers, queen. Type locality: New Guinea (L. Loria), MCSN, QMBA (examined). Polyrhachis sexspinosa var. esuriens Emery; Bolton, 1975: 12. Junior synonym of P. sexspinosa (La- treille, 1802: 126). Polyrhachis (Myrmhopla) arcuspinosa Donisthorpe. 1941; 140. fig. 1. Syntype workers. Type locality: New Guinea, Mt Nomo, S of Mt Bougainville (L.E. Cheesman), BMNH, QMBA (examined). Syn. nov. Polyrhachis (Myrmhopla) juxtaspinosa Donisthorpe, 1949; 417, fig. 2. Syntype workers. Type locality: New Guinea, Maffin Bay (E.S. Ross), CAS, BMNH (examined). Syn. nov. REMARKS. Bolton (1975: 12) considered P. sexspinosa esuriens, P. arcuspinosa and P. jux- taspinosa (and several other names), to be syn- onymous under P. sexspinosa (Latreille). I have examined the syntypes of these three taxa, in comparison with the voucher specimen (‘para- digm’) of P. sexspinosa in QMBA (Kohout & Taylor, 1990: 519-520) and I am confident that, while they are clearly conspecific, they are not synonymous with P. sexspinosa. The syntypes of these taxa differ from the P. sexspinosa paradigm in a number of characters, including distinctly more gracile stature and more slender spines. The petiole has a flat, anteriorly sloping dorsum and more-or-less horizontal, barely elevated, posteri- orly directed spines, situated at the apex of the posterior face of the segment. The propodeal spines are very variable in length and degree of elevation, though mostly strongly inclined poste- riorly and either parallel or diverging with their tips curved downwards. In P. sexspinosa the body is generally more robust and the petiole has a distinctly convex dorsum, with spines rising obliquely from their bases, which are situated well below the apex of the dorsal convexity. The propodeal spines appear less variable, and are mostly vertical or inclined forwards or back- wards, and generally straight with only their tips slightly curved. The mainly golden pubescence of P. esuriens (including P. arctt.'ipinosa and P. jia- taspinosa) almost hides the underlying sculptura- tion, except on the propodeal dorsum, where it is usually somewhat diluted, exposing the charac- teristic reddish colour of the body. In contrast, the pubescence in P. sexspinosa is off-white to sil- very, and the body mostly black. NEW SYNONYMS IN THE ANT GENUS POLYRHACHIS 519 Polyrhachis waigeuensis Donisthorpe, 1943 stat. nov. Polyrhachis (Myrmhopla) arcuspinosa waigeuensis Donisthorpe, 1943; 467. Syntype workers. Type locality: New Guinea, Waigeu L, Cape Nok (L.E. Cheesman), BMNH (examined). Polyrhachis (Myrmhopla) arcuspinosa waigeuensis Donisthorpe; Bolton, 1975: 12. Junior synonym of P. (Latrcillc, 1802: 126). REMARKS. Comparison of the syntypes of P. arcuspinosa waigeuensis with types and other specimens of closely related taxa of the P. sexspi- /?(95a-complex indicated that they represent a ‘good’ species. P. waigeuensis stands between P. esuriens and P. sexspinosa, sharing with P. esuriens a flat-topped petiole with barely ele- vated, posteriorly directed spines, and with P. sexspinosa a relatively robust body. However, P. waigeuensis differs from both these species (and others of the complex) by a highly convex prono- tal dorsum. Subsequent examination of abundant recent material, including several nest series (PNG, West Sepik Prov., Pes Mission, c. 12km WSW of Aitape, 03*^1 US, 142“15’E,31 vii-3 viii 1984; Tor- ricelli Mts, Lumi, 03°28’S, 142‘'02’E, 4-13 viii 1984; Madang Prov., Naru, Gogol Riv., 20km SW ofMadang,05°2rS, l45°4rE,22 viii 1984 [all RJ. Kohout]), shows the pronotum ranging from mod- erately to highly convex, even almost dome-shaped in some individuals. However, despite this variabil- ity, it seems always to be higher than in other species of the complex, and this provides a dependable character, setting waigeuensis well apart. 1 thus recommend its separate specific status. Subgenus Myrma Billberg, 1820 Polyrhachis continua species-group Polyrhachis conops Forel, 1901 Polyrhachis conops Forel. 1901: 28. Syntype workers, queens, males. Type locality: Bismarck Archipel- ago, Herbertshohe (F. Dahl), MNIIU, MHNG, QMBA (examined). Polyrhachis (Myrma) conops var. bismarckensis Kara- vaiev, 1927: 46. Holotype worker. Type locality: Bismarck Archipelago, 7MNHU, 71ZAS (location of type unknown). Junior primary homonym of P. mucronata bismarckensis Forel, 1901: 33. Polyrhachis (Myrma) conops var. stitzi Santschi, 1 928: 139. Replacement name. Syn. nov. REMARKS. The holotype of P. conops stitzi cannot be found in the Karavaiev collection (IZAS), or in the Stitz collection (MNHU). It appears to have been lost. How- ever, during a recent visit to MNHU 1 located a small bottle containing six unidentified specimens referable to tlie P. continua species-group, labelled ‘Ralum Dahf. The series comprised two woricers and four alate queens. all callows at various stages of pigmentation. However, the woifcera and two of the queens were suitable for mounting. Their later comparison with Karavaiev’s brief description, strongly indi- cates that they are in fact identifiable as P. conops stitzi. Judging from the original description (‘Gasterseg- mente, vom dritten an, dunkel ockerfarben’), the P. conops stitzi holotype was also callow. It seems reason- able to conclude that the aforementioned specimens are the remainder ofthe series from which Stitz supplied one specimen to Karavaiev (1927: 46). Their comparison with the syntypes ofP. comps shows these forms to be essentially identical and undoubtedly conspecific. Polyrhachis continua Emery, 1887 Polyrhachis continua Emery, 1887: 235, pi. 4. fig. 21. Holotype worker. Type locality: Indonesia, Ternate, Aqui Conora (O. Beccari), MCSN (examined). [Specimens from Paumomu River (= Angabanga Riv.), New Guinea, L. Loria (Emery det. & coll., MCSN) are wrongly labelled as types]. Polyrhachis continua var. revocata Viehmeyer, 1913: 151, Syntype workers. Type locality: Indonesia, Sulawesi (in copal), MNHU (examined). Syn. nov. REMARKS. Comparison of the P. continua holotype with P. continua revocata syntypes shows them to be very similar. In both the clypeus is almost straight, with only a shallow depression at the distinctly medially truncate anterior mar- gin. The frontal carinae are markedly close to- gether, leaving the central area rather narrow, and scarcely widened towards the occiput. However, in P. continua, the somewhat flattened pronotal spines are shorter, relatively less divergent and minutely emarginated at their bases, while P. continua revocata has longer, distinctly more di- vergent spines, which have their dorsolateral bor- ders continuous posteriorly in uninterrupted alignment with the pronotal margins. The petiolar spines are also slightly longer than in P. continua. The head and mesosoma in P. continua are very finely, more-or-less regularly, striate-punctate. The sculpturation in P. continua revocata is somewhat more distinct and more regular. How- ever, when the types are compared to other New Guinean specimens referable to P. continua, it is clear that the characters putatively separating these forms fall well within the limits of variation represented in the material. I therefore conclude that the names P. continua and P. continua revo- cata together refer to a single biological species. Polyrhachis procera Emery, 1897 stat. nov. Polyrhachis continua procera Emery, 1897: 581. Syntype workers, queens. Type locality: New Guinea, Haveri (L. Loria), MCSN (examined). 520 MEMOIRS OF THE QUEENSLAND MUSEUM REMARKS. Emery (1897) separated P. procera from P. continua on the basis of its larger size and the more pronounced striation of its head and thorax (‘... sono piu grandi e con striaturadel capo e del torace piu marcata’). However, direct com- parison of the types evidences differences in a number of other characters. In P. procera the clypeus in profile is gently sinuate, with the ante- rior margin vaguely obtuse medially (entire in modem specimens) and the medial carina distinct for most of its length. In contrast the clypeal border in P. continua is almost straight, with only a shallow depression at the anterior margin, which is distinctly medially truncate. The prono- tal spines in P procera are relatively long and more slender, with the lateral margins of mesono- tal dorsum clearly converging anteriorly, and its posterior angles broadly rounded. The leading edges of the antennal scapes of P. procera carry a few hairs, which are totally lacking in P. con- tinua. Comparable pilosity in both species is lim- ited to the head and gaster, where it is rather diluted dorsally. P. procera is black with dark reddish-brown appendages, with the tibiae and apical segments of the funiculi a shade lighter. In P. continua the colour scheme is similar, except that the appendages are generally light reddish- brown. Examination of the types and other avail- able material clearly demontrates that P. procera is a ‘good’ species distinct from P. continua. Polyrhachis sericeopubescens Donisthorpe, 1941 Polyrhachis (Myrma) sericeopubescens Donisthorpe, 1941: 61. Synt>'pe worker, queen. Type locality: New Guinea, Japen L, Ml Baduri (for ? ); Mt Eiori (for 9 ) (L.E. Cheesman). BMNH (examined). Polyrhachis inusitata Kohoul, 1989: 513. figs 4. 5, 9. Holotype, paratype workers. Type locality: Austra- lia, Qld. Cape York Peninsula, Iron Range (G.B. Monleith & D. Cook), QMBA. ANIC. Syn. nov. REMARKS. I have examined a syntype worker of P. sericeopubescens and the holotype and two paratypes of P. inusitata. Despite some relatively minor differences they are remarkably similar. Both taxa appear to be very rare. The only known specimens are their respective type series. The type localities are separated by a great distance, and no specimens have been recorded from inter- vening areas. However, because of their similarity, I believe these two nominal taxa to represent sepa- rate populations of a single biological species. Polyrhachis simpla Santschi, 1928 stat. nov. Polyrhachis (Myrma) conops simplex Karavaiev, 1927: 45. Syntype workers, queens, males. Type locality: Indonesia, Aru Is, Wammar I. (W. Kara- vaiev), IZAS, QMBA (examined). Junior primary homonym of P. simplex Mayr, 1862: 682. Polyrhachis (Myrma) conops st. simple Santschi, 1928: 139, Replacement name. REMARKS. I have examined syntypes of P. sim- pla together with syntypes of P. continua Emery, P. continua hirsutulaEmQry\P. conops Forel and P. conops spinifera Stitz, and believe P. simpla to represent a species separate from the others. Its type stands closest to those of P. conops spinifera, sharing with them a relatively wide pronotal dor- sum. However, the lateral pronotal margins in P. conops spinifera are distinctly raised, while those of P. conops simpla are rather flat. The lateral margins of the mesonotal and propodeal dorsa in P. conops simpla are more strongly converging posteriorly, with the pilosity much diluted, pro- ducing a rather graceful appearance, in compari- son with the more robust and rather hairy P. conops spinifera. Further examination of abun- dant modem material confirms these characters to be constant, justifying the elevation of P. conops simpla to specific rank. Polyrhachis spinifera Stitz, 1911 stat. nov. Polyrhachis conops var. spinifera Stitz, 1 91 1 : 376, fig. 22b. Syntype workers. Type locality: New Guinea, Tana (Moszkowski), MNHU. MHNG (examined). Polyrhachis continua var. hirsutula Emery, 1911: 256. Syntype workers. Type locality: New Guinea, R. Digul (= ?Digoel Riv.), MCSN (examined). Syn. nov. REMARKS. Direct comparison of P. conops spinifera and P. continua hirsutula syntypes shows them to be very similar, undoubtedly rep- resenting a single species. They differ from P. conops and P. continua by possessing a markedly wider pronotal dorsum and rather abundant semierect to erect pilosity, which is almost com- pletely lacking in these other species. Stitz (191 1: 376, fig. 22c) also described a very similar taxon, P. conops cuspidata, in the same article as P. conops spinifera. He specified the shape of the petiole, with shorter and divergent petiolar spines and a rather narrow Thorax’ as the main characters separating P. cuspidala from P. spinifera. Comparison of the syntypes of these nominal taxa to investigate their relative status has proved somewhat inconclusive. Examination of numerous specimens of various P. continua- group species, including several nest series of the closely allied P. simpla Santschi (see above), has shown a degree of variability in all of the charac- ters given for P. cuspidata and P. spinifera by Stitz, but never sufficient to bridge the difference NEW SYNONYMS IN THE ANT GENUS POLYRHACHIS 521 in width of the pronotal dorsum separating their types. 1 therefore hesitate to consider them con- specific. Only study based on material additional to that now available could establish the true status of P. conops ciispidata. Polyrhachis relucens species-group Polyrhachis andromache Roger, 1863 Polyrhachis hector Fr. Smith, 1859; 142. Holotype worker. Type locality: Indonesia, Aru Is (A.R. Wal- lace), OXUM (examined). (Junior primary homo- nym ofP. hector Ft. Smith, 1857: 62). Polyrhachis andromache Roger. 1863b: 8. Replace- ment name. Polyrhachis (Myrma) proximo var. semirufipes Donis- thorpe, 1943: 469. Holotype worker. Type locality: New Guinea. Waigeu, Camp Nok (L.E. (iheesman), BMNH (examined). Syn. nov. 'Polyrhachis (Myrma) sericata pruinosula var. har- msi' Karavaiev, 1 930: 2 1 2. Workers. Original local- ity: New Guinea (F. Weyer), IZAS (examined). Unavailable name. (For full reference citations see Kohout, 1988: 430.) REMARKS. Direct comparison of the P. proximo semirufipes and P. andromache holotypes, to- gether with a single worker of the ‘’P. sericata pruinosula harmsC original series, shows them undoubtedly to be conspecific. Within the whole range of distribution (which extends from the eastern Indonesian islands, throughout New Guinea, and southwards to Cape York Peninsula in northern Australia) the general appearance of P. seems rather constant. Variability appears to be confined to the color of the rather smooth, appressed pubescence, which ranges from silvery to a rich golden colour. Polyrhachis decipiens Roger, 1863 stat. nov. Polyrhachis decipiens Roger, 1863a: 156. Synlype workers. Type locality: Indonesia, Batjan I., MNHU (examined). Polyrhachis relucens decipiens Roger; Emery, 1897: 580. Polyrhachisrestitiita'^^\Q\\mQy