Memoirs of THE Queensland Museum Brisbane 30 June, 1974 Volume 17 Part 1 Volume 17 Part 1 Memoirs OF THE Queensland Museum Published by Order of the Board Mem. QdMus. 17 ( 1 ): 1-19. [1974] THE WOLF SPIDERS OF AUSTRALIA (ARANEAE: LYCOSIDAE): 2. THE ARENARIS GROUP R. J. McKay Queensland Museum ABSTRACT Four species of Australian lycosid spiders are defined as the ‘arenaris group’ due to the presence of a distinct tubercle on the outer curve of the fang of mature males. Lycosa celaenica is a synonym of L. arenaris, L. segregis is a synonym of L. pullastra, L. furcillata is assigned to this group of species, and L. lapidosa is described as new. In an attempt to arrange the Australian Lycosidae into a more natural classification, all species in which a distinct tubercle is present on the outer curve of the fang of mature males were placed together as forming a natural group of species. An analysis of this group indicates that the species are indeed related as some similarities can be observed in the shape of the epigynum, the structure of the male palp, and the internal genitalia of mature females. In addition to the morphological similarities, all species are found to occur in similar habitats such as river beds and in grassed areas near swamps or on watered lawns. A burrow consisting of an open shallow pit, or a webbed retreat among grasses or stones, is constructed by most species, rather than a well excavated burrow with a flap, door, or turret, as is frequently found in other members of the genus. The function of the tubercle on the fang of mature males is unknown. As this tubercle is not present on the fangs of the female or immature males, it may be associated with courtship, and used to ward off the fangs of the female prior to copulation. Lycosa pullastra has been observed during courtship and copulation, but at no time were the fangs employed in any manner to suggest the function of the tubercle. The mature males frequently cluster around a sexually mature female during the breeding season, as the population density is normally very high. Little intra-specific mortality occurs between males despite frequent contact ; the encounter is usually brief, although at times quite violent, and it is suggested that the tubercles on the fangs prevent a high mortality between courting males. The family Lycosidae is currently under review by the writer. This contribution is part 2 of a series of papers on Australian Wolf Spiders. An introduction to the study is given by McKay (1973). Lycosa arenaris Hogg, 1905 (Figure 1 , a-m) Lycosa pulvere-sparsa\ Hogg, 1896, p. 351, MacDonnell Ranges, Palm Creek, Bagot’s Creek, Paisley Bluff, Central Australia, [Not Lycosa pulvere-sparsa]. 2 MEMOIRS OF THE QUEENSLAND MUSEUM Lycosa arenaris Hogg, 1905, pp. 586-8, fig. 88, MacDonnell Ranges, (Northern Territory); Rainbow, 1911, p. 265; Bonnet, 1957, p. 2634. Lycosa celaenica Rainbow, 1917, p. 488, pi. 32, figs. 10 and 11, Farina, South Australia; Bonnet, 1957, p. 2637; McKay, 1973, p. 379. Varacosa arenaris: Roewer, 1954, p. 305; McKay, 1973, p. 381. Material Examined Syntypes: 2, SAM, labelled ‘'Lycosa arenaris Hogg, MacDonnell RNG. Horn Exp.’ in ink, ‘ Lycosa arenaris H. R. Hogg. 2 females Types. MacDonnell Range, Horn Exp.’ in pencil, and ‘F. 206' in pencil. A lectotype is here designated from this series. Lectotype: SAM 9 M with a well developed epigynum, in a tube with the original labels. Par a lectotype: SAM 9 P with the epigynum partly formed but in a penultimate condition, in the same tube as the lectotype. Holotype of Lycosa celaenica: SAM, 9 M, with abdomen detached labelled ‘ Lycosa caelonica Rainb. Farina, Mus. Exp. Int. 1916 T’ in ink (mis-spelt, = Lycosa celaenica ), ‘Farina to Spring, Mus. Exped. Interior 1916’ in pencil, a small faded printed label ‘TYPE’ and ‘ Lycosa caelonica Rainb. Farina to Spring TYPE’ in pencil on a fourth label. This specimen fits the description of L. celaenica well except that the carapace is 5*9 mm long and 4*3 mm wide (Rainbow has ‘5-8 mm broad, 4-3 mm long’, but this would not be possible for a Lycosa). The epigynum of the holotype is shown in Fig. 1 , g. Other Material: Western Australia: Avon River at Northam, 14.xii.1962, BYM, 4 $ M, 1 d 1 M, WAM 71-1000-4, 1 d M, WAM 71.988; Behn River at Old Argyle Downs station, 9.x. 1971, RJM, 1 $ M, 72-76, 24.x. 1971, 2 ? M, 1 d M, WAM 72-77-9; Broome Hill 15 miles east, 3.ii. 1961, BYM, 5 ? M, 3 d M, WAM 71-1546-53; Desperate Bay near Snag Island, 27.ii.1971, RJM, B. Ryle, 1 9 M, WAM 71-891; Fitzgerald River, 12.vii.1970, RJ M, 4 ¥ M, 8 d M, 4J, W AM 7 1 - 1 904-19 ; Fortesque River, 22. ii. 1 962, BY M, 2 ? M, 1 d M, 2J , W AM 7 1 - 1 554-8 ; Hooley Station at Kylan Pool, Sept. 1969, RJM, 1 J, WAM 71-946; Hotham River 71 mile peg Albany Highway, 27.x. 1961, BYM, 1 9 M, 2J, WAM71-989-91 ; Kojonup 17 miles east, 3.xi.l961, BYM, 6 9 M, 3 d M, WAM 71-1537-45; Maitland River, 23.ii.1962, A. R. Main, 4 9 M, 2d M, 1J, WAM 71-1513-20; Maitland River 30 miles south, 22.ii. 1962, A. R. Main, 4J, WAM 71-1521—4; Millstream Station, Fortesque River, 23, 25.ix.1969, RJM, 3 9 M, WAM 7 1 -942-4, 1 9 M, WAM 70-210; Mount Herbert at pool, 20. v. 1961, BYM, 1 9 M, WAM 71-1562; M u gum ba Creek, 26.ii.1963, A. R. Main, 1 9 M, WAM 71-1559; Murchison River at the Loop, 26.i. 1969, RJM, JG, P. Snowball, 3 9 M, 1 d M, WAM 69-335-40; Murchison River, 20. ii. 1962, BYM, 8 9 M, 4J, WAM 71-1525-36, 27.U969, RJM, JG, 1 9 M, WAM 71-1564, 8.vi.l970, W. K. Youngson, 1 ? M, 70-166; Ongei ap 85 miles east, 4.xi,1961, BYM, 1 9 M, WAM 71-1560; Ord River, 3.x. 1971, RJM, 1 9 M, WAM 72-166, 21.x. 1971, RJM, 2 9 M, WAM 71-153-4; Perth 80 miles east. 4.xi.l961, BYM 2 9 M, 6d M, WAM 71-992-9; Rudall River, 4.V.1971, RJM, R. W. George, 4 9 M, 2 d M, 4J, WAM 71-1241-50; Susetta River, 12.vii.1970, RJM, 2 9 M, WAM 70-218-9; Twirtup Creek, Fitzgerald River, 1 5.vii. 1970, AB, 6 9 M, 2d M, WAM 70-312-19’ Wittenoom Gorge Pool, 30.ix.1969, RJM, R. Dear, 2$M, WAM 69-1043^1; Yannarie River, 13.V.1972, RJM, 1 9 M, QM W3854. South Australia: Clayton Bore 33 miles north of Maree, 25.viii.1970, HB, W. D. L. Ride, 9 ? M, WAM 71-564-72; Kalamurina Homestead near Lake Eyre, 9,i.l902, Professor J. W. Gregory, 2 9 M, NM. Victoria: Snowy and Broadbent Rivers, xii. 1947, C. W. Brazenor, 1 9 M with young, NM. Misidentified as Lycosa pulveresparsa by Hogg (1896): Central Australia, W. A. Horn, 2 9 M, BM 1897,1.18. 11-12; Palm Creek, Central Australia, Horn Expedition, 2 9 M, Hogg Coll., BM 1924, III. 1. 977-8. Hogg’s (1896) record of L. pulveresparsa from Central Australia is in error, and was not rectified in his 1905 description of L. arenaris. Description Modified from Hogg: Female. Carapace yellow-brown with mixed black and white hairs on the cephalic part, and more posteriorly a wide white area constricted in the middle WOLF SPIDERS OF AUSTRALIA: 2 3 and extending on to the middle of the rear slope where it narrows abruptly as a thin white line ; margin of carapace with a thin, frequently broken, sub-marginal white line ; mandibles dark brown with pale brown hair; labium, maxillae, and sternum yellow-brown. Abdomen a Fig. 1 : Lycosa arenaris a-c, epigyna of WAM 69-340, WAM 69-335, and WAM 70-210; d-f, internal genitalia of WAM 70-312, WAM 70-313, and WAM 72-77; g, epigynum of holotype of L. celaenica; h, fang of mature male; i, median apophysis of male palp; j-k, unexpanded and expanded male palpal organ of WAM 71-988; 1-m, unexpanded and expanded palpal organ of WAM 70-314. 4 MEMOIRS OF THE QUEENSLAND MUSEUM dull brown, irregularly spotted with small patches of whitish hairs above; venter dull brown, the centre area rather paler than the sides. Legs and palpi yellowish with brown hairs and darker brown erect bristles ; a ring of paler yellow around femur, middle of tibia, and metatarsal joints. Anterior of carapace high and narrow with the clypeus about equal to or less than a diameter of the AM. Anterior row of eyes procurved, the AM larger than the AL and further apart than from the AL, PM less than a diameter apart. Three equal sized retromarginal cheliceral teeth. Labium almost as wide as long. The epigynum is generally broader than long with a narrow median guide, and a wide transverse guide curled around the base of the lateral margin. Three epigyna are illustrated (Fig. 1 , a-c). Legs slender, two spines above on tibiae III and IV, none on the first or second pairs ; on the underside the spines are long and stout. TABLE 1 : Measurements of Leg Segments of L. arenaris in mm From Hogg Leg Coxa Trochanter and Femur Patella and Tibia Metatarsus and Tarsus 1 2i 5 54 5 2 2 44 5 5 3 2 44 44 54 4 24 6 6 8 Palp n 3 2i n Variation: The venter of the abdomen of mature males and females may be of a uniform yellowish-brown without markings, light brown with vague darker bars or spots, or honey-brown with transverse, longitudinal, or criss-cross smoky-grey markings. Legs uniform brown, mottled, or with brown rings. The mature male has a tubercle on the outer surface of the fang (Fig. 1, h), penultimate males lack the tubercle. The eye measurements were found to vary in the ten specimens examined, each measurement is given below in Table 2 as a per cent of the total width of the first eye row. Hogg (1905, p. 587) describes the front row of eyes as straight, although his figure illustrates the anterior row as quite procurved ; all specimens examined by me had the anterior row of eyes procurved, and the distance AM:AM always wider than the distance AM:AL (not equidistant as described by Hogg). The clypeus is not twice the diameter of the AM as described by Hogg, but about equal to or much less than the diameter of the AM. The epigyna and the internal genitalia of three Western Australian females are illustrated (Fig. 1 , a-f). The male palpal organ is figured in the unexpanded (Fig. 1, j, 1) and partly expanded condition (Fig. 1, k, m) to show variation in the shape of the median apophysis illustrated separately in Figure 1 , i. Above the tip of the somewhat triangular blade of the median apophysis is the rather scoop-shaped membraneous secondary conductor in which the embolus lies in the unexpanded state. An unusual and very distinctive embolic guide (see WOLF SPIDERS OF AUSTRALIA: 2 5 TABLE 2: Eye Diameters and Interspaces of Lycosa arenaris Converted to Percent of the Total Width of the First Row of Eyes Regd No. Sex C.L. AM AL PM PL AM:AM AM:AL PM :PM AM:PM WAM 72.77 $ M 5-4 24 18 42 35 10 5 21 8 WAM 72.78 ?M 4-9 24 18 44 36 12 4 24 11 WAM 72.79 © Ol « op 3 Q © "O > as 1 19 <5 a % it U 2 u kI 3 1 8* |u ^ S3 T3 3 at 'S', §" JQ 6 a m '-I u 4 co © © -d- r- oo -d- m N N m ri ob m m u~, 4 © cn m oo m in in iri « i4444 JO x> ©©©©©©© t 4 - yr in rn - (N r~ o oo r~~ oo s © © © © © © © ® in oo l m mi © v-i rt 'O 00 mi © © © © © © © r- r 1 ' cn « 4 cs mi mi mi ^ rJ- oo © hi ib © ^ Tf Tf 1/10 ! w I •2 | « «> * -c •g 6 s g ° 5 « o T3 2 3 £ 3 O -o 3 O on o O, e E 3 z cd o . o h rt ta cd u JD CD D co rt J CO „ co ~ l-H a c/3 jd •y.. C ~ c 8 * 3 9 ? & o H nJ „ CD “ PC s c J o C > Cl S3 cd j> cd cd Cfi (j W M ■f— -f~ -I— +— ■+— -J— - +— +— +— +— +— +— ■2 '2 c c cd cd CO CO 03 iT 13 13 13 ^ •o c T3 c ■§ T3 G *§ T3 c *s T3 a a cd cd cd cd cd cd cd cd cd !-> «-i >-> „** C o a a C3 Q <3 CO Co a <3 CO a a <3 a <3 ri co © o © 1 -0 CO e CO 5 CO co © -c *-C: -g © O -© c> C g O s £ 5 co Co s £ e co 1 1 Co to £ Co Co | 03 £ £ £ £ J- J J s- J 'J J CJ a s •5 -C o V. *s -© -c. V ~c -C <1 3 zs a (Nc r it'~~r'-50 1, 2, 3, 5. Fine sand 50-25 4, 12, 14. <20 11, 15. Coarse sand <20 6, 7, 8, 9, 10, 13. Further perusal of data indicates that the most aberrant sites in the first, third and fourth groups are 1, 4 and 13 respectively. If required the fourth groups can be further subdivided by percentages of coarse sand into 6, 7, 8, with >60% and 9, 10, 13 with <60%. Water Movements North of Peel Island tides flood from and ebb to the north. Immediately south the tidal streams come from northeast and northwest round the island, leaving a ‘slacker’ area which includes area I. Conversely in area II the tidal currents are strong and during spring tides surface currents are 2-3 knots. No data are available on bottom currents. The area as a whole is protected from oceanic swells and waves are due solely to local winds. The most relevant data are the annual wind-roses for Brisbane given in the Official Year Book of the Commonwealth of Australia for 1970. At 9 a.m. the most frequent winds are southerly and southwesterlies, while at 3 p.m. they are most frequent in an arc from northeast to southeast. Newell (1971) also includes wind data from Sand gate covering January 1967 to December 1968 and states (p. 5): ‘A seasonal pattern emerges, with winds from north to east predominating between December and April, and from south to south- west between May and August. There is no obvious seasonal pattern in the occurence of strong or light winds.’ BENTHIC FAUNA OF SOFT BOTTOMS 79 The longest wind-fetches in the sampling area are in an arc from south to southeast and the wave-action is severest when these winds blow against a flooding tide from the north. This and the shallow depths makes sites 11, 12, 14 and 15 the most wave affected. Area I is the most protected, particularly from the north-northwest and north; next is area II especially from the north-northeast. From all other northern directions waves curve round Peel Island to affect all areas. Sources of Water in the Area: Newell (1971, p. 31) notes that the southern end of Moreton Bay is relatively isolated from the aspect of water exchanges and notes that \ . . the Bay water is divided into an eastern fringe, a western and southern fringe, and a central zone.' General observations supplemented by drogue data of Helbig (unpublished) shows that a stream of eastern or Pacific water enters the area from the South Passage via the Rainbow Channel with incoming spring tides. This will give the most oceanic influences in area I. The tidal flow into areas II and III is more from the western and central portions of the Bay which are under greater terrestrial influence. Newell (1971) has shown this includes greater dilution by flood waters and a greater temperature variation — it also includes higher turbidity and greater possibilities of pollution. Newell’s (1971) work did not encompass the occasional catastrophic effects of severe flooding from river run-off, which have been mentioned by Slack-Smith (1960) in a bio- logical connection and Stephenson (1968) in a brief hydrographical survey. The main influence of floods in the area is from the Logan-Albert River which discharges from the south. These effects will be greatest in area III, and least in area I. This topic is further discussed under seasonal changes (see below). SEASONAL CHANGES Dilution and Concentration No direct data are available on salinities within the area of investigation, but relative effects can be deduced from rainfall data and water discharge by the Logan-Albert Rivers. Unfortunately data upon flows in these rivers does not cover a sufficiently long and continuous period to be of present value. Rainfall data, kindly supplied by the Common- wealth Bureau of Meterology have been used. From the known rainfall pattern in the Logan-Albert catchment two weather stations were selected as typical of lower and higher rainfall areas; these are Beaudesert and Springbrook respectively. (In fact Springbrook is just outside the watershed but is still regarded as typical of the higher rainfall areas). Data upon monthly rainfalls at these sites from 1915 onwards are given in Table 3. From December to March is typically the wet season, and July to September is typically the driest part of the year. Analysis of the total available rainfall data from the above stations showed that the year before the study (1969) was fairly typical except for an increase in August recordings. The late autumn and winter of 1970 (May-August) and of 1 97 1 (April-June) were unusually dry and from this one would expect higher than usual salinities in the sampling data. The summer of 1970-71 (Dec. 1970-Feb. 1971) included two months of exceptionally heavy rainfall from which one would expect unusual summer dilution. Comparably ususual 80 MEMOIRS OF THE QUEENSLAND MUSEUM TABLE 3 Rainfall in 1/1 00th inch for Selected Sites in Catchment of Logan-Albert Rivers. Month Mean* (St. Dev.) Beaudesert 1969 1970 1971 Mean* (St. Dev.) Springbrook 1969 1970 1971 Jan 514(361) 238 796 801 1680(1376) 181 1116 1646 Feb 529(445) 99 487 1173 1809(1468) 1037 1964 3823 March 397(288) 75 470 143 2310(2978) 1197 1022 1743 April 241(212) 65 82 23 1039(817) 237 501 611 May 204(226) 838 63 88 828(681) 1643 126 147 June 245(286) 83 70 22 799(1081) 111 67 91 July 177(199) 90 64 108 603(842) 184 160 295 Aug 137(114) 527 42 155 390(339) 539 202 339 Sept 163(121) 127 274 148 408(306) 167 504 237 Oct 274(200) 543 314 109 613(475) 2188 987 243 Nov 312(233) 334 499 306 830(697) 1480 1405 416 Dec 478(298) 332 1204 244 1037(789) 398 3125 728 Total 3680 3351 4365 3320 11975 10462 11179 10234 * Means and standard deviations are for 1915-1970 inclusive. conditions, in both respects (low and high rainfalls), have not occurred since 1946-47, and for these reasons one might expect particularly accentuated seasonal patterns in the benthos. Temperature The most relevant available data are mean monthly temperatures at the Pile Light off the mouth of the Brisbane River, covering the period 1931-1950. TABLE 4 Water Temperature in °C at (Old) Pile Light. Month Mean Highest Lowest Jan, 25-6 27-8 23-9 Feb. 250 28 3 23-9 Mar. 25-0 26 1 23-3 Apr. 22-8 25-0 211 May 20-6 23-3 18 9 June 17 8 22-2 15 6 July 16 1 18 3 14 4 Aug. 16*7 18-9 14-4 Sept. 18 3 22-2 16*7 Oct. 211 22-8 19 4 Nov. 22-8 250 21-7 Dec. 250 26-7 22-8 Mean 21.1 22.8 20.6 BENTHIC FAUNA OF SOFT BOTTOMS 81 These data were kindly made available by the Commonwealth Bureau of Meteorology and comprise 9 a.m. readings at depths of 6-8 ft (ca. 2-3 m) below water level. Converted to °C they are shown in Table 4. No data are available during the years of the present study, but since there were no prolonged periods of unusual air temperatures, we may assume that the water temperature lay tolerably close to the mean values given above. SAMPLING PROGRAM Many authors have criticised the use of single 0- 1 m 2 grab samples, for example Holme (1950, 1953, 1964) and M. L. Jones (1961). Clearly the number of samples required depends upon the small-scale heterogeneity of the area and Ford noted as long ago as 1923 that a slight alteration of the position of the ship sometimes gave a striking change in fauna and in substratum. Preliminary work in the present general area indicated that twenty-five 0- 1 m 2 samples might be required at a given site. Sampling at this intensity for fifteen sites at each season was beyond the logistic possibilities, and would also have increased the risk of a second sample being collected from exactly the same area as a later one (see below). Eventually, following Steven (1930) and Longhurst (1958) five samples were taken at each site on each occasion. Sampling was repeated at intervals of three calendar months from March 1970 to December 1971, and with one exception (March 1970) collections were made within a week. Difficulties in operating occurred initially. The boat was positioned accurately by drifting from an upwind or uptide position, but like Day, Field and Montgomery (1971 p. 7) we found the grab ‘tended to strike the bottom on its side, and many attempts had to be made before the wire could be maintained sufficiently vertical for a reasonable sample to be obtained.’ This was overcome firstly by working only during neap tides, secondly by not operating in wind speeds exceeding 20 knots, thirdly by using an anchored boat, and fourthly by lowering the grab gradually until within about 2-3 m from the bottom. With this method, the grab worked successfully in the present range of depths and substrates, excepting that repeat samples were sometimes required in area III (see later). Problems of accurate relocation of sites arose as the work progressed. Although accuracy had been required it was discovered that only the boat anchor had been carefully positioned on each occasion. It was estimated that, owing to swinging in wind and tide, samples at each season were within an area of ca. 25 m 2 , but that over the entire program they lay within a circle of just under 20 m radius, i.e. an area of ca. 1000 m 2 . (On Figure 2 these circular areas are drawn to scale). Clearly some of the differences apparently due to seasons could be from sampling a different portion of the circular area. The inaccuracy of site relocation has a counter-balancing fortunate effect. It greatly reduces the chance of the grab collecting a presampled spot. On only one occasion was there evidence (from the physical appearance and faunistic contents of the sample) that this had happened in one of the quintuplieate samples. This was discarded and recollected. No evidence was obtained of exact resampling of a spot after a seasonal interval, and because of the large area involved this would be unlikely. The normal method of collecting the biota from grab samples was by on-board wet sieving, with the final apertures ca. 1-2 mm square. Particular care was taken over samples 82 MEMOIRS OF THE QUEENSLAND MUSEUM containing small specimens of bivalves and gastropods which fortunately were infrequent. The procedure was modified on grounds suspected to contain Branchiostoma because smaller specimens burrowed through the sieve holes, and in such cases the sample was hand sorted before the finer sieving. After collection all specimens were initially preserved in 10% formalin and within a month tranferred to the most appropriate final preservative. In summary the sampling program allows investigation of three types of spatial patterns (a) at a given site (and a given time) within the groups of quintuplicate samples whose exact spacings are unknown; (b) between the five sites within each area, which range from 150 to 700 m apart — an alternative is between topographic groupings as listed earlier; (c) between the three areas, whose incentres are from 1400 to 2430 m apart. It also allows a variety of temporal patterns to be investigated. These include seasonal (March, cf. June, cf. September, cf. December), annual 1970 cf. 1971), and possible differences immediately after a flood (March 1971 cf. remainder). BIOTIC DATA Identifications Chace (1969) comments on \ . . a major and virtually insuperable problem which plagues any ecological survey in tropical waters today; that is, the imperfect knowledge of the existing fauna , . This applies to Moreton Bay, despite systematic publications referred to in our dredging work (Stephenson, Williams and Lance, 1970). Because of anticipated difficulties in identification and in recognition of species, hydroids and amphi- pods were excluded from the investigation ; the former were also excluded on other grounds. Amphipods were present in smaller numbers than in comparable surveys elsewhere, but nevertheless their omission is unfortunate. The problems of identification were also responsible for the selection of a relatively coarse sieve (1-2 mm square) during collecting. This excludes the smaller species where the major taxonomic problems would lie. If these small species had been included a much more complex situation than that recorded would have been obtained. Shortage of space precluded retention of all specimens collected, so reference collec- tions were established with at least duplication of ‘species’ occurring twice or more. Initially species were segregated and given ‘working names’ with assistance from a reference collection of most species from the previous dredging work (Stephenson, Williams and Lance, 1970) housed in the Queensland Museum. This resulted in some over-splitting, and was resolved when group experts had identified collections. Over-splitting probably remains in the Nemertea. In one case ( Euclymene ) two or possibly three species are lumped because of uncertainty regarding early discarded specimens. Reference specimens, and when possible entire collections, were referred to group experts for identification. Without this assistance (as acknowledged later) the present study would have been very incomplete or even impossible. There remained a sizeable collection with only ‘working names' or incomplete identifications, some of which are likely to be in error. These have been deposited in the Queensland Museum and the attention of taxonomic experts is invited. BENTHIC FAUNA OF SOFT BOTTOMS 83 Groups Excluded In addition to those excluded because of taxonomic difficulty peneid prawns were eliminated ; they are known to be common at certain seasons in Moreton Bay but a grab is an inefficient device for their collection. The few specimens which were obtained were discarded. Acorn barnacles and limpets were recovered from superficial dead shells. These belong to the hard-bottom fauna and their distribution reflects the almost random recovery of large solid objects. All the hard-bottom fauna was excluded — serpulids, polyzoa, hydroids, most polyzoans and some algae. When species occurred on large dead shell fragments and also the substratum generally (e.g. some tunicates) they were recorded. Form of the Data : Usually the number of individuals of each species from each grab sample was recorded, but problems arose over colonial forms. Where the numbers of discrete colonies was known this \yas noted, but where this was not known — for example with Halophila where separate shoots arise from one or several stolons — we have only obtained data in a presence/absence form. Presence was noted as for a single individual. Material which was obviously dead was not retained. All colonies of the serpulid Filograna implexa (Berkeley) were excluded for this reason, as would have been most polyzoan and hydrozoan colonies but for their prior exclusion. In general, empty tubes of polychaetes were not recorded either because it was clear they had not been inhabited recently or because of uncertainty over the species. The exception was Chaetopterus variopedatus and when it appeared that tubes were currently inhabited, but the specimens had eluded capture, data were recorded. Complete tubes were never recovered (nor were intact specimens), and here two separate tube-ends were recorded as one worm and so was a single separate end. Species Obtained The 420 species are listed in Appendix I. The richness of this biota is difficult to compare precisely with that of most other surveys because of methodological differences. However, a certain number of local comparisons are possible. In our dredging work in Moreton Bay, we covered a much greater variety of habitats and a much greater area (ca. 775 km 2 against ca. 3 km 2 ) but obtained fewer species, 335 in all, and 25 of these are in groups excluded from the present survey. The difference mostly reflects the loss of specimens during dredging operations. Current work by Raphael and Stephenson in Bramble Bay near the mouth of the Brisbane River, using the present grab and sieves, and over an area of ca. 30 km 2 seems likely to produce ca. 180 species, thus indicating a relatively impoverished fauna. Hailstone (1972) in the lower reaches of the Brisbane River has noted 136 species from carefully sifted dredgings obtained from an area of ca. 1 km 2 (but of varied habitats). Comparisons from wider areas are more difficult but we can note 74 species from an area of ca. 4 km 2 in a single season at Sek Harbour, New Guinea, using the present technique (Stephenson and Williams, 1971) and 264 species from ca. 30,000 km 2 by Petersen (1914) in Danish waters (Stephenson, Williams and Cook, 1972). Summarising the above, we are clearly dealing with a species richness of a different order from the other authors quoted. 84 MEMOIRS OF THE QUEENSLAND MUSEUM At the level of actual species present, the present is a pioneer study in eastern Australian waters. Hence there are numerous new Queensland records and a considerable number of undescribed species. This renders comparisons difficult and premature, and the only relevant ones are with surveys in Moreton Bay. Comparison with the dredge survey of Moreton Bay as a whole (Stephenson, Williams and Lance, 1970) is difficult because of the unidentified species in the two surveys (e.g. sponges) and because some species were confused in the earlier study. For example we suspect at least five species of ophiuroids were recorded previously as Amphioplus sp. (This was due to damage of specimens during collection and examination of insufficient samples.) Within comparable taxonomic groups 115 species were common to the two surveys, 150 were obtained by dredge only and 275 by grab only. This confirms the relative inefficiency of dredge sampling and that the two methods sample different fractions of the biota. Comparison with incomplete data of Raphael and Stephenson indicates 70 species common to the two areas, 110 only at Bramble Bay and 350 only at Peel Island. Compari- sons are not affected with Hailstone’s data from the lower Brisbane River, much of which is incompletely identified. ANALYTICAL METHODS Fusion of Samples Analyses of small scale topographic patterning within quintuplicate samples was attempted, but because precise positions are unknown and only five samples are available at each site, analyses were restricted, and results are not detailed. There were indications of small scale aggregations in at least half of the more ubiquitous species. To reduce data to moderate proportions, results from quintuplicate samples are fused and in general represent numbers obtained from 0-5 m 2 samples. The presence/absence data upon such forms as Halophila now become meristic summations and hence all data are of substantially the same form. Choice of Method The main problem is to produce conceptual order within the 420 species obtained in the survey. It could be approached by what have become almost traditional methods of studying diversity, using one of the multiplicity of diversity measures based upon the proportions of species present for example Simpson (1949), Margalef (1951), McIntosh (1967), Edden (1971), Hurlbert (1971); or the widely used measures depending upon information theory, the Shannon and Brillouin measures (see Mac Arthur, 1955 ; Mac Arthur and MacArthur, 1961 ; Lloyd and Ghelardi, 1964; MacArthur and Wilson, 1967; Lloyd, Zar and Kar, 1968; Pielou, 1969; Whittaker, 1972). There is a comparably extensive literature upon measures of equitabllity or evenness, as quoted in Hurlbert (1971) and Whittaker (1972), and upon explanations of why diversity and equitability are higher in some situations than in others (see Pianka, 1966; Recher, 1972; and Whittaker, 1972). We believe that the best approach is by partitioning so that, for example, we can evaluate the contributions made by species in sites, seasons and years. There have been theoretical discussions of the partition of diversity information statistics for example by BENTHIC FAUNA OF SOFT BOTTOMS 85 Kullback, Kupperman and Ku (1962) and by Kullback (1968) and an ecological approach by Orloci (1968). Partitioning of information diversity has been explored in an ecological (rain-forest) problem by Williams et al (1973). There are two disadvantages as listed by Williams and Stephenson (1973); partitioning is incomplete and results seem ecologically unsatisfactory. An alternative method is to use a sum-of-squares partitioning as detailed in the last quoted paper and this we followed. Because it involves analysis of variance and euclidean distances (centred), results are likely to be unduly influenced by the more abun- dant species. In the earlier paper a cube-root transformation was used and incomplete investigations of effects of transformations suggested this was appropriate in the present case. Our original data is in four-dimensional form viz 1 5 sites x 4 seasons x 2 years x 420 species, and desirably we should seek patterns between all possible pairs of these dimensions. When analysis commenced, no method of handling four-dimensional data had been developed, so our data was first regarded as three-dimensional and in the form of 15 sites x 8 times x 420 species. (The times 1-8 run sequentially from March 1970 at three month intervals to December 1971). The three-dimensional technique is as described by Williams and Stephenson (1973) and leads to three two-way coincidence tables, of sites and species, times and species, and sites and times. Flexible sorting with 0 at —0-25 was used (see Lance and Williams, 1967; Stephenson, Williams and Cook, 1972). Critique of Method We should first comment on the advantages of this method over the two approaches used previously. The first used a qt x s matrix (where q = quadrats, t = times and s = species). Analysis by any of the usual techniques of grouping samples usually reveals via the dendrograms whether the major influence is due to q or to t, Interpretation of sub- ordinate groupings is difficult because they are variably homogeneous with respect to q and t attributes. Comparable difficulties arise if we make t separate analysis of q and s. As quoted in an earlier paper (Stephenson, Williams and Cook, 1972) referring to work by Hailstone (1972) we stated : ‘For each month of the year the data reveal site groups and species groups related through coincidence tables. But from month to month the site groups alter their composition and so do the species groups.’ Various sub-sets extracted from the present data had been separately analysed before the present main analysis began. These were the 55 species as related in Williams and Stephenson (1973), the ophiuroids (19 species) and tunicates (27 species). Details need not now concern us, the main conclusions were that the site-groupings obtained in each case were substantially similar, but the times-groupings were not. For example, with the tunicates the main picture was seasonal with grouping of times : 1 +5,2 + 6, and 3 + 7, but breaking down with 4 and 8. The 55 species and ophiuroids gave segregation of March data 1 + 5, with an annual pattern for the remainder, viz. 8 + 7 + 6, 4 + 3 + 2. With the 71 species selected as below yet another pattern emerged :7, 8 + 6 + 5 + l,4 + 3 + 2. It is evident that times-groupings give no overall conceptual picture. As in the case of qt x s matrices, groupings contained variable proportions of homogeneity with respect to two different variables. In the present case these are seasons (here m for months) and years (y). To resolve this difficulty one can again partition part of the data, converting 86 MEMOIRS OF THE QUEENSLAND MUSEUM times x species into the dimensional form of months x years x species (summed over all sites) i.e. m x y x s where m = 4 and y = 2. In the analyses which follow we commenced with a q x t x s matrix, and following the methodology of the previous paper affect data reduction, consider the relative import- ance of the different comparisons and analyse them in a preliminary way. This leads to two extensions, firstly of the species/sites relationships (s x q), and secondly to the species/ times ones (5 x /) which are expanded to three dimensions (s x m x >>). ANALYSES OF SITES x TIMES x SPECIES First Stages of Analysis These involve species reduction and considerations related thereto. We begin by summating and obtaining means (cube root data) of : (a) all species over all times at each site, and (b) all species over all sites at each time. Values are given in Table 5. TABLE 5 Mean Values, Species Summated (Cube Root Data) in Sites 1-15, and in Times 1-8, Site Mean n Times Mean n 1 0-490 1 0-814 2 0-473 2 1-583 3 0-456 3 1-593 4 0-671 4 1-394 5 0-457 5 0-832 6 0-883 6 1-188 7 0-785 7 1-769 8 0-775 8 1-140 9 0-749 10 0-739 11 0-672 12 0-664 13 0-840 14 0-806 15 0-852 The values in Table 5 are measures of total populations and are the transformed numerical equivalents of total biomass. These data already show some ‘ecological sense’, for example that (a) populations are least in the muddy sites 1, 2, 3, 5 and (b), a clear seasonal pattern exists with low populations in March, rising through June to a maximum in September and declining in December. Comparison of March 1971 (time 5) with March 1970 (time 1) fails to show the expected effects of the summer flood but there is possibly a delayed effect at June 1971 (time 6 cf. time 2). The second step leading to species reduction is to determine the contribution each species makes in terms of squared euclidean distance to the differences between all pairs of sites — i.e. to site classification. Based upon these contributions we can have a method of species reduction which is quantified and can be subject to a single decision. In the BENTHIC FAUNA OF SOFT BOTTOMS 87 earlier work, using 55 species, we showed an objective sum-of-squares chopping was much too drastic, replaced it by an arbitrary level and excluded species contributing less than 1% of the total variance. With the present full number of species (420) this would have left only 26 species. We reduced the level to 0-5% and retained 48 species. In a similar way the contributions of each species to a time classification were evaluated, and taking the 0*5% level we retained 52 species. In the analysis detailed below we considered the 7 1 species which were present in one or another of the above lists. They are listed in Table 6 with only the generic name given when a single species of a genus is in the main list (see Appendix). TABLE 6 71 Species List, the New Arbitrary Number is Given Followed by the Original Number in Parenthesis. 1 ( 1) Discobotellina 37 (267) Gari venta 2 ( 29) N emertea — ‘ ‘pink” 38 (269) Later nula 3 ( 35) ? Euleanira 39 (270) Leptomya pura 4 ( 36) Leanira 40 (279) Malleus 5 ( 37) Sthenelais 41 (283) Neosolen 6 ( 58) Arabella 42 (285) Nucula astricta 7 ( 62) Eunice antennata 43 (288) Paphia gallus 8 ( 63) E. cf indica 44 (289) P. subrugata 9 ( 70) Onuphis sp. (“long gill”) 45 (297) Placamen tiara 10 ( 73) Chloeia 46 (306) Tellina lilium 11 ( 74) Eurythoe 47 (308) T. cf. solenella 12 ( 75) Glycera americana 48 (316) Tucetilla 13 ( 76) G. prashadi 49 (328) Golfingia 14 ( 77) Goniada 50 (330) Thermiste 15 ( 90) Nephtys 51 (340) Amphioplus depressus 16 ( 91) Amaeana 52 (341) Amphipholis loripes 17 ( 95) Loimia medusa 53 (342) Amphioplus sp. 18 (100) Pista sp.3 54 (343) Amphiura bidentata 19 (104) Terebellides 55 (344) A. cataphes 20 (123) Chaetopterus 56 (350) Ophiocantha 21 (129) Euclymene 57 (356) Ophionereis 22 (134) Magelona sp. 58 (360) Hypselaster 23 (139) Petaloproctus 59 (364) Protankyra 24 (145) Spiophanes 60 (367) Agnesia 25 (159) Alpheus distinguentus 61 (373) Eugyra 26 (164) Axius 62 (381) Molgula exigua 27 (165) Callianassa 63 (382) M. rima 28 (204) Carinapseudes 64 (383) M. sabulosa 29 (214) Columbella 65 (385) Polycarpa fungiformis 30 (221) Herpetopoma 66 (387) P. tinctor 31 (236) muricid 2 67 (390) Styela ramificata 32 (244) Dentalium 68 (394) Branchiostoma 33 (258) Cycladicama 69 (413) Pseudo codium 34 (262) Modiolus ostentus 70 (419) Halophila ovalis 35 (263) Ensiculus 71 (420) H. spinulosa 36 (265) Fulvia 88 MEMOIRS OF THE QUEENSLAND MUSEUM The noteworthy groups are polychaetes (22 spp.), bivalves (16), tunicates (8) and ophiuroids (7); the noteworthy exclusions are all of the sponges, coelenterates and crabs, and all but one of the algae. Relative Importance of Each Comparison Using the 71 species, mean variances per comparison, as shown in Table 7, permit assessment of the relative importances of each comparison. TABLE 7 Mean Variances per Comparison Nature of Variance attributable to : Comparison sites species times interaction inter-sites 1 629 0-905 0-406 inter-species 1-629 — 1-123 0-406 inter-times 0-905 1 123 — 0-406 These data show that, with respect to species, site-groupings are about 50% more important than times-groupings. The site-times combination, which concerns the ‘numbers’ equiva- lent of the total biomass, is about 80% of the species-times combination. Its magnitude indicates that there are important chronological changes in total populations, as discussed earlier. The interaction is the variance left in the system when any two of the major variables have been eliminated. It is the smallest value in the table and is not readily amenable to commonsense ecological interpretation so is neglected. Site-Species Relationships Classifying the sites by the 71 species (fused over all times) and taking this to the four group level gives: 1 = 11 +15, II = 6 + 7 + 8 + 9 + 10 + 13, III - 4 + 12 + 14, IV = 1 + 2 + 3 + 5. These are identical with those obtained in the preliminary analyses. It is evident that there is small scale patterning within the area of study. With area I of 0-24 km 2 there is one site-group and part of a second; within area III of 0*36 km 2 there are elements of three site-groups ; only area II of 01 6 km 2 appears biotically homogeneous. Further discussion of the site-groups is given later. To determine how these site-groups are characterised by species, reference was made to the 2-way table, referred to in Williams and Stephenson (1973) as the B table which gives single values — these are the means of site-centred and species-centred deviations. Perusal suggested that many species-groups derived in the analyses were heterogeneous, and we reverted to the tables with double values i.e. site-centred values used in site classification and species-centred values for the species classification. Using the latter the species-groups now appeared more homogeneous internally, but still insufficiently so in some cases. While many species-groups contained species which conformed ‘crisply’ to the site-groups in others the sorting was more ‘blurred’. A measure was sought of the extent to which species conformed to the site classification. BENTHIC FAUNA OF SOFT BOTTOMS 89 In general, the attributes used to define a numerical classification cannot be used as the basis for between-group tests of significance, since their differences have been optimized. It follows that the site-centred values could not be used for this purpose; it remains to consider whether the species-centred values can be used in their place. Unfortunately, the setting up of a two-way table implicitly assumes that the two sets of values are related, so that the between/within variance ratios will be conservative, not random, estimates. However, the variance ratio Fean still be used as a measure of the extent to which a given species defines the different site-groups, and the values associated with the usual probability estimates can be taken as convenient points on an arbitrary scale. We shall henceforth refer to such tests as ‘significance tests’ for the sake of brevity ; but it must be remembered that, though they are less biassed than would be the corresponding ratios obtained from the site-centred values, they are still biassed. Using the above test for four site-groups and 1 5 sites, F values are as follows : 0*05-3*59, 0 01-6-22, 0*001-1 1*56. Although these values are conservative they can be used to grade species with respect to site-group conformation. Four grades were recognised with F values as follows: VH > 11-56, H 1 1-56-6-22, M 6*21-3*59, and L <3-59. d a 2.0 Fig. 3: Dendrograms showing classification of times by species (a) and sites (d); euclidean distance type of coefficient. 90 MEMOIRS OF THE QUEENSLAND MUSEUM Of the 71 species classified, 22 were in the L grade, and this included species high on the list of contributions to site variance. Consideration of these and other species is given later. Times-Species and Times-Sites Relationships Using the 71 species list, four classifications were derived: (a) times by species, (b) species by times, (c) sites by times, and (d) times by sites. We consider first (a) and (d) above as shown in Figure 3. In both cases the nearest pair of seasons (months) are 1 and 5 i.e. the ‘normal’ March of 1970 and the ‘abnormal’ March of 1971 immediately after the flood. It is evident that the immediate effects of the flood were minimal both on the total populations present (dendrogram d) and on the actual species present (dendrogram a). In both cases the most isolated time is 7 (September 1971), suggesting a delayed flood effect. However in the times by sites analysis this time is characterised by a specially abundant total population. If there is a delayed flood effect it seems to have operated by allowing a transient increase in certain species which becomes evident after a delay of about nine months. In neither case does the first or second dichotomy reflect a clear seasonal grouping or a clear annual grouping to indicate which factor is the more important. For this and other reasons stated earlier the analyses were expanded. EXTENSIONS OF ANALYSES OF SPECIES-SITES Possibility of Alternative Site-Grouping Of the 7 1 species previously analysed it was considered that the 22 not fitting into the site-grouping earlier obtained, might conform to a different grouping. These species plus additional ones relatively high in the site variance hierarchy (= 34 in all) were used in a site reclassification. The site-groupings obtained were: 1 (very isolated), 12 + 14, 2 + 3 + 4 + 5, 13 + 15, 6 + 7 + 8 + 9 + 10+11. These do not reflect any obvious abiotic attributes nor an overall topographic pattern. Five of the six species highest in the variance hierarchy dominated the classification and gave almost individualistic distributions within the site-groups. While it is possible that another and more meaningful pattern would emerge if they were removed and the remainder re-analysed, it was considered more likely that yet more individualistic patterns would emerge. It was concluded that there is only a single important pattern of site-groupings. No further mention is made of the species which failed to fit the general site-groupings, unless they appear in the later times-groupings. One species which seemed to characterise the area as a whole, and which was one of the characterizing species of the earlier dredge study, is eliminated from the present groups. It is Leanira (36). Consideration of Additional Species The technique of testing species for significance of conformity to the major site- pattern was applied to all 420 species. Ninety-seven species conformed at the M grade or BENTHIC FAUNA OF SOFT BOTTOMS 91 4 Fig. 4: Dendrogram of sites classified by 97 spp., euclidean distance type of coefficient. The site-groups (I-1V) do not refer to the area groups in Figure 2. higher and these extended as low as 250th in the site-variance hierarchy. The matrix of 1 5 sites x 97 species was reclassified. Site-Groups As expected, the four original site-groups were generated (see Figure 4) and as stated earlier these agree with those obtained using sedimentary attributes. The correspondence goes below the four group level; thus in site-groups I and II the ‘aberrant’ sites by both systems are 1 and 4 respectively. Site-group IV can readily be subdivided by both systems which agree in having sites 6, 7 and 8 in one subgroup and 9 and 10 in the other. The only discrepancy concerns site 1 3 which links with 6, 7 and 8 in the dendrogram and with 9, 10 by sediments. 92 MEMOIRS OF THE QUEENSLAND MUSEUM 16 193 Fig. 5: Dendrogram of species-groups (involving 97 spp.) classified by sites, euclidean distance type of co- efficient. The 17 groups originally considered were reduced to the eleven which are ringed. Species-Groups and Two-way Tables After generating an excess of species-groups, 1 7 were considered as shown in Figure 5. Further fusions within the exact framework of the dendrogram destroyed the conceptual sense of the species groupings as revealed in two-way tables. This sense was retained by accepting groups at different dendrogram levels, as shown in Figure 5, with a condensed two-way table in Table 8. In other cases we have investigated, acceptance at different levels has been due to group-size dependence, a phenomenon inherent in sharply clustering strategies such as the flexible strategy here employed. (See Williams et al, 1971 for discussion of group-size dependence). In the present case higher fusions were possible when the groups contained the more abundant species and this is discussed later. Nine rows of Table 8 contain a single value distinctly higher than the remaining three, and here the species-groups characterize the site-groups in a positive way. Two rows contain two high values, and the relevant species-groups (189, 177) each positively charac- terize a pair of site-groups. There are only two possible species-groups with a single distinctly low value (189, 3) and in the discussion below we concentrate upon positive characterizations. BENTHIC FAUNA OF SOFT BOTTOMS 93 TABLE 8 Mean Deviations of Species-Groups (Species Standardized Values) in Site-Groups. Species-Groups are as in Figure 5. Species groups 1 1, 2, 3, 5 Site-groups and sites therein II III IV 4,12,14 11,15 6,7,8,9,10,13 185 -1-8 -0-9 -1-6 2-0 189 -4-8 -2-3 3-5 3-2 3 -2-0 9-3 4-4 -1-8 158 1-0 -01 —0-5 0-4 175 -01 1-8 -0-4 -0-7 171 3-1 0-4 -0-8 1-6 177 2-3 1-3 -1-3 -1-7 144 -0-9 -0-4 2-0 0-1 156 -0-9 -0-6 -0-4 1-0 170 -1-9 -0-3 2-5 0-9 178 -01 0-2 4-4 1-5 Perusal of Table 8 shows that fusions of species-groups with similar positive charac- terisations are possible — for example species groups 144, 170 and 178 all characterize positively site-group III. Such heuristic higher groupings do not conform to those of the dendrogram (Figure 5), and the possibility of another classificatory strategy for species was considered. The heuristic species-groupings were retained and those given in Table 8 fused as follows : A = 158 4- 171, positive characterization of site-group I B = 3 + 175, positive characterization of site-group II C - 144 + 170 + 178, positive characterization of site-group III D = 185 + 156, positive characterization of site-group IV E - 177, positive characterization of site-groups I and II F =189, positive characterization of site-groups III and IV. Characteristics of Species in Species-Groups Within each species-group, species were arranged in order of grades of conformity (VH, H and M) as defined earlier. Within these grades they were further graded in ‘impor- tance’. Importance values were estimated, within the 97 species, as the summated site- centred cube-root contributions of a given species to the site-classification. Values ranged from 79-2 to — 24-1 and were graded: H > 10, M 10 to 0, L < 0. Within the conformity- importance groups species were arranged in systematic order, with numbering as in the Appendix, and are given in Table 9. Interim Summary of Species-Sites Analyses Site-group III, comprising two sites (11, 15) and with a sediment of fine sand with a low percentage of mud, has the richest characterizing biota. This comprises the 24 spp. of 94 MEMOIRS OF THE QUEENSLAND MUSEUM TABLE 9 Characteristics of Species-Groups (Site- Analysis). Species Conformity Importance Species numbers Group grade grade (see Appendix) A VH L 80, 358 H L 85, 204,244, 311, 359. M fH 197, 420 1 L L 55, 66, 67, 126, 149, 214, 296, 303. B VH r H 364 i M 374 U 306 H 4 1 M 165 1 L L 277 M , r H 263 \ M 342 I Ll 18, 97, 160, 286, 309, 346. C VH j r H 129, 139 1 L L 109, 131, 255, 279, 395, 402, 419. H r H 62 4 M 283 \ u 120, 125, 210, 299. M J I M 97, 100 1 L L 171, 193, 199, 233, 253, 372, 393. D VH j r H 385 4 M 316, 382 1 Ll 132 H r H 75, 288, 350 4 M 221 1 Ll 54, 72, 107, 156, 184, 267, 381, 390 M r H 330 4 M 1, 145. i Ll 69, 70, 134, 140, 308, 356, 359, 366, 415. E VH H 37, 285 H L 270, 274 M H 90, 104. F VH H 63, 289, 297, 394 H H 35, 340, 386 M H 123 species-group A and the eight of species-group F. It also has the highest number of uniquely characterizing species with VH conformity — nine. Against the figures of 32 and 9 for site-group III there follow: site-group IV (coarse sand) — 28 and 4; site-group I (mud) — 23 and 2; and site-group II (fine sand with higher percentage of mud) — 19 and 3. The composition of the species-groups is considered later. BENTHIC FAUNA OF SOFT BOTTOMS 95 80 60 40 20 0 Fig. 6: Dendrogram of species classified by months. Species-groups with ‘internal homogeneity’ are ringed. EXTENSIONS OF ANALYSES OF SPECIES-TIMES Because the species-times relationships are of more interest than those of sites-times, expansion of times analyses were restricted to the former and now become a species x months x years situation in three dimensions. Relative Importance of Each Comparison Dealing with the original 71 species, mean variance per comparison is shown in Table 10. It will be noted that the total variance is half that of the species-times comparison listed in Table 7. The months x years comparison is approximately twice the importance of 96 MEMOIRS OF THE QUEENSLAND MUSEUM TABLE 10 Mean Variances per Comparison. Nature of Variance attributable to : Comparison months years species interaction Inter-months 0-270 0-077 0-055 Inter-years 0-270 — 0-129 0-055 Inter-species 0-077 0-129 — 0-055 the species x years, which in turn is almost double that of the species x months comparison. In other words, firstly changes in total populations with time are the most important elements and secondly that a monthly pattern of species common to the two years will be the ‘weakest’ of the three which emerge. This contrasts with the reasonable similarity of the month totals of species summated between the two years (see Table 5) and indicates that different species are replacing one another tolerably precisely from the months of 1970 to the equivalent months of 1971. Seasonal and annual associations of species are investigated below, meanwhile the interaction values (0-055) should be noted. While low in absolute terms they are high relative to the species-months comparison. Again no ecological interpretation of this interaction is attempted. Species-Months Relationships Only four months (1, March; 2, June; 3, Sept. ; and 4, Dec.) are involved. The months classification is of little interest, the dendrogram showing an initial dichotomy of 1 + 2 and 3+4. The similarity between 3 and 4 is slightly more than between 1 and 2. The dendrogram of the species classification (Figure 6) was taken to the 10 group level, and the seven groups containing the species with the largest deviations of species- centred values were internally homogeneous in that all species characterized the same month in the same way. These groups are ringed in Figure 6. However the remaining three groups, which contained most of the species (46/71) each included species with either two or three types of month characterizations. This heterogeneity was not resolved by a finer classification. Even more so than in the study of species characterizing sites, was it necessary to alter the computer classification, and in the present case it was discarded. Perusal of the values of species centred deviations showed, as with the species in site-groups, that some species had an outstanding positive deviation and hence positively characterized one of the four months. Others had negative characterizations and some appeared to have both positive and negative ones. A method of quantifying an ‘outstanding’ deviation was required, bearing in mind that tests of significance are not possible with only four entities. That adopted was the proportion of the total variance of differences of the species-centred values contributed by the selected species-in -month values. Values for the 71 spp. ranged from ca. 0-3 to 0-76 and were graded arbitrarily as follows: 0-3—0-49, L (17 spp.); 0-50-0*59, M (21 spp.); 0-60-0-69, H (22 spp.); and 0-70-0-76, VH. Species with a L grade were eliminated. BENTHIC FAUNA OF SOFT BOTTOMS 97 When one of the remaining species appeared to characterize two seasons, one positively and the other negatively, it was accepted if (a) the variance due to these two seasons was 0-95 or more of the total, and (b) several such species showed similar characterizations. On these bases eight species with double characterizations were accepted. It is appreciated that these criteria operate more stringently with respect to the month- groupings than those adopted for the site-groupings. This was done deliberately because in the present case there are not tests of significance and because there are fewer degrees of freedom. TABLE 11 Species, with Numbers from Appendix, which Characterize Months. Species Conformity Importance Species group grade grade numbers A H M 139 (+ve March) M H 123, 297 B VH L 165, 267 H H 73, 76 (+ve June) M f M 385, 37 1 l L 360 C VH L 1,373,413 H J f M 29, 104 (+ve Sept.) 1 l L 62, 100, 343, 367 M L 236, 241 D H L 145, 269, 306, 316 (+ve Dec.) M L 77, 283, 383 E VH J r H 289, 344 1 1 M 387 H 1 rH 35, 364 (— ve March) 4 M 330, 420 i LL 134, 263, 285, 342,419 F M M 75 (— ve June) G VH L 221, 279 H L 288 (— ve Dec.) M fH 90 1 i M 91 H* ( ve March) H r H 63, 394 M , IH 36 1 L L 74, 204, 328 (+ve Sept.) M L 244, 341 "■Double characterization — ve March, -f-ve Sept, grading restricted to outstanding cases. 98 MEMOIRS OF THE QUEENSLAND MUSEUM Using these criteria and re-sorting the groups obtained by computer classification gave eight species groups. Conformity values of each species were obtained as previously— precise values differ because they were derived on different assemblages, one of 97 spp. and the other of 71. In Table 1 1 these data are given with the months they characterize positively, negatively and both. Because only four months are considered and conformity grades are arbitrary the analysis was not extended beyond the 71 species of Table 6, and of the 54 species accepted 30 characterize months positively, 24 negatively, and 8 give double characterizations. The outstandingly ‘weak’ season is March with only three positive species and with 1 8 negative. It is followed by December with seven positive and five negative. The ‘strongest’ season is September with 19 positive and no negative species. Species- Years Relationships The interest is in species-groupings, and the computer classification dissects an array of numbers in which each species bears balancing positive and negative deviations in the species-centred values. Groups are tabulated below by positive characterizations, also listed are the range of deviations in each group, the number of species it contains and the arbitrary grades of conformity which were given. Because there is only a single degree of freedom these grades were applied stringently. TABLE 12 Number of Species of Different Conformity Grades which Characterize Positively the Years 1970 and 1971. 1970 Range No. spp. Range 1971 No. spp. Conformity grade 044- H 10 00-0-7 11 discard 15 - 2-5 7 1-1 - 2-6 6 L 3*1 - 3-9 10 3-0 - 3-6 7 M 41-5 4 9 42-5-5 6 H do 1 oo 5 9-8 1 VH Species are listed, with numbers from Appendix, in Table 13 for the upper three conformity grades, together with importance values obtained in the usual way. Months- Years Relationships Data are given as a double 2-way table in Table 14 and show that the 1971 biota is less than that of 1970, suggesting adverse effects of the flood. This conclusion is negatived by the data in Table 5 (second column) which deals with all species and not merely the 71 used for Table 13. The lowest value in Table 5 is for March 1970, before the flood; this is confirmed in Table 14. BENTHIC FAUNA OF SOFT BOTTOMS 99 TABLE 13 Species, with Numbers from Appendix, which Characterize Positively the Years 1970 and 1971. Species group Conformity grade Importance grade Species numbers A VH i H 63, 344, 350, 364 l L 343 (+ve 1970) H f H 73 < M 129, 159, 387 l L 37, 77, 360, 383, 390 M f H 340, 394 < M 385 l L 1, 62, 204, 265, 279 285, 381 B VH L 342 (+ve 1971) H r H 289 -< M 104 L L 308, 341, 367 M / H 35 l L 95, 145, 221, 236, 316, 382 TABLE 14 Two-Way Table of Months- Years; Upper Values Centred by Years, Lower Values by Months. Year Summations at Right, Monthly Ones Below. Months Year 1, March 2, June 3, Sept. 4, Dec. Summations 1, 1970 —0-5 43-8 -1-3 25-7 y t 67-7 — 102*2 44-8 430 14 3 4 ) > 2, 1971 0-5 -43-8 1-3 -25-7 y 2 -67-7 -67-3 -90 79-5 -3-2 Monthly Summations m, — 169-5 m 2 35-8 m 3 122-5 m, 11-1 4 CONSIDERATION OF SPECIES-GROUPINGS General Comparisons of Groupings There are two difficulties. The first is that the species-groups derived by the computer analyses did not precisely coincide with those required on conceptual grounds — it was by using the latter that groups were accepted. The second is that a test of conformity of species 100 MEMOIRS OF THE QUEENSLAND MUSEUM to entity-groupings which had some statistical significance could only be applied to the species-in-sites data. This allowed acceptance of more species in sites data (97) than in months (54) or years (38) data. A large fraction of the species which were above the M grade of conformity in the sites analyses were present in the other lists (39/97) and with lists of equal length the fraction would have increased. To exclude from site-classification all species with marked seasonal or annual patterns, as Petersen (1914) suggested, would greatly disturb the site- analyses and exclude much valuable data. On general grounds one might have suspected that species of VH conformity to a particular entity group would include many cases of faithful species, which in turn would be uncommon and have low importance values. Survey of the data did not support this expectation ; instead there are many cases of VH conformity coinciding with H importance. In these cases the diagnostic species are those of high dominance/constancy, and we have ‘Petersen type communities’ in the sense used in our previous paper (Stephenson, Williams and Cook, 1972). The difference is that these apply in an area sense (as did Petersen’s) but also in a season and a year sense. Post- Analytical Data Reduction Ninety-seven species can be accommodated in the site-groups. For each we have data in continuous form on the degree of conformity to the site-groups and on relative import- ance over all sites. Data involving this sort of detail are difficult to comprehend so both conformity and importance values have already been graded. As in earlier work (Stephen- son, Williams and Cook, 1972) further data compression is necessary for comprehension and below we eliminate all except those with VH conformity or H importance. These reduced data are given in Table 15 and are discussed below, comparable data for month- groups and year-groups are given in Tables 16 and 17. Species in Site-Groups Site-group I comprising the muddy sites is characterized uniquely by Leocrates (80) and Ophiura kinbergi (358) as VH conformity species and Rhizopa (197) and Halophila spinulosa (420) as H importance species. Ophiuroid ‘communities’ are known from muddy grounds from many parts of the world (see Thorson, 1957) and from our earlier work in Moreton Bay (Stephenson, Williams and Lance, 1970) and in New Guinea (Stephenson and Williams, 1971). This particular species has not been involved previously. Site-group II comprising fine sand with a relatively high percentage of mud is charac- terized uniquely by three species at VH conformity levels. These are Tellina lilium (306), Protankyra sp. (364) and Microcosmos (374). The H importance species are Ensiculus (263) and Protankyra (364). Here the coincident occurrence of Protankyra indicates that a dominant species is characterizing a community in the classical manner (see Petersen, 1914: Stephenson, Williams and Cook, 1972). Community indicators, comparable to some of the present species, occur elsewhere, for example the many Tellina communities listed by Thorson (1957). The present species T. lilium has moderate constancy for a site- group which is predominantly muddy north of the mouth of the Brisbane River (Raphael and Stephenson 1972). In the present case it is, however also associated with Protankyra. BENTHIC FAUNA OF SOFT BOTTOMS 101 TABLE 15 Species (in Groups) Characterizing Site-Groups* (condensed from Table 9). Positive Characterizations. Species group Selected species VH conf. H importance Site-group characterized A 80, 358 197, 420 I (sites 1, 2, 3, 5)— mud B 306, 364, 374 263, 364 II (sites 4, 12, 14)— fine sand, high mud C 109, 129, 131, 139, 255, 279, 395,402,419 62, 129, 139 III (sites 11, 15)— fine sand, low mud D 132,316, 382, 385 75, 288, 330, 350, 385 IV (sites 6, 7, 8, 9, 10, 13) —coarse sand E 37, 285 37, 90, 104, 285 I and II F 63, 289, 297, 394 35, 63, 123, 289, 297, 340, 386, 394 II and III TABLE 16 Species (in Groups) Characterizing Months-Groups* (condensed from Table 1 1). Species group Selected species VH conf. H importance Month characterized A — 123, 297 March (-(-ve) B 165, 267 73, 76 June (+ve) C 1, 373,413 — Sept (+ve) D — ' — Dec (+ve) E 289, 344 35, 364 March (— ve) F — — June (— ve) G 221,279 90 Dec (— ve) H — 63, 394 March (— ve), Sept (+ve) *Only VH conformity and H importance species are listed. Species numbers are from the Appendix. 102 MEMOIRS OF THE QUEENSLAND MUSEUM TABLE 17 Species (in Groups) Characterizing Year-Groups* (condensed from Table 13). Positive Characterization. Species group Selected species VH conf. H importance Year A 63, 343, 344, 350, 364 63, 73, 340, 344, 350, 364, 394 1970 B 342 35, 289 1971 *Only VH conformity and H importance species are listed. Species numbers are from the Appendix. This latter species, under the name of Leptosynapta was one of the species which, in the earlier dredge survey (Stephenson, Williams and Lance, 1970) characterized the peri- pheries of the extensive sloping areas of mud to the north of the Brisbane River. In that earlier survey no ‘ Protankyra grounds' were located in the Peel Island area. In the earlier survey Ensiculus was another characterizing species, but was not associated with the others now listed. It did characterize many of the areas near Peel Island, including the location of the present site-group. Site-groups I and II are collectively characterized by two species of VH conformity — Sthenelais (37) and Nucula astricta (285), and four species of H importance — the two just cited together with Nephtys (90) and Terebellides (104). The three polychaetes have almost cosmopolitan distributions and have been involved in community descriptions elsewhere (see Thorson, 1957). Two {Sthenelais and Terebellides ), like Protankyra in the dredge survey, characterized the muddy grounds to the north of the Brisbane River. Of these, one {Terebellides) is an important community indicator in the current work of Raphael and Stephenson. Again there is evidence of a ‘Petersen-type community’in the co-occurrence of Sthenelais and Nucula astricta under both conformity and importance headings. Site-group III comprising fine sand with a relatively low percentage of mud is uniquely characterized by nine species of VH conformity : sabellid 4 (109), the polyspecific Euclymene spp. (129), Isolda (131), Petaloproctus (139), Circe (255), Malleus (279), Acetabularia (395), Gracilaria verrucosa (402) and Halophila ovalis (419). Of these Euclymene and Petalo- proctus are of H importance, as is another species Eunice antennata (62). Many of these species also characterized site-groups in our dredge survey but were spread between several species groups and there is little correspondence. None of the species characterize the areas being studied by Raphael and Stephenson. Site-group IV comprising coarse sand is uniquely characterized by four species of VH conformity: Lygdamis (132), Tucetilla (316), Molgula rima (382), and Polycarpa fungi- formis (385). The H importance species are Glycera americana (75), Paphia gallus (288), Thermiste sp. (330) and Polycarpa fungiformis (385). Again several species are common to one or another of the species-groups derived in the dredge survey, and none are amongst the important species just north of the Brisbane River (Raphael and Stephenson 1972). BENTHIC FAUNA OF SOFT BOTTOMS 103 There remain several species (species-group F) characterizing site-groups II and III. The VH conformity species are : Eunice cf. indica (63), Paphia subrugata (289), Placamen tiara (297) and Branchiostoma (394). All four have H importance values as have lEuleanira (35), Chaetopterus (123), Amphioplus depressus (340) and Polycarpa pedunculata (386). Again most occurred in one or another of the groups from the dredge survey, but on preliminary results few appear important just north of the Brisbane river. Summarizing the species in sites analysis : (a) there are several cases of ‘Petersen-type communities’ in which species of VH conformity also have H importance values, (b) many of the characterizing species of the present survey also characterized site-groups in the earlier dredge survey (Stephenson, Williams and Lance, 1970) but there are only slight correspondences in detail, (c) several of the species characterizing the more muddy sites of the present survey also in the current work, are characterizing muddy sites just north of the Brisbane river. Several species which were intuitively thought to be characteristic of one area or another are not listed above. They include Discobotellina (1) which only has M conformity to site- group IV. Several species which occurred commonly and which might characterize the total area of investigation rather than sections of it are not listed. One such is Leanira (36), which did not conform to the major site pattern. Species in Month-Groups March has no species with VH positive conformity but two with H importance: Chaetopterus (123) and Placamen tiara (297). It is negatively characterized at the VH conformity level by Paphia subrugata (289) and Amphiura catephes (344), and at the H importance level by lEuleanira (35) and Protankyra (364). Additional species are con- sidered below. June only has positive characterization at the levels now considered ; the VH confor- mity species are Callianassa (165) and Gari venta (267) while the H conformity ones are Chloeia (73) and Glycera prashadi (76). September again only has positive characteristics with three species of VH conformity and none of H importance. The former are Disco- botellina (1), Eugyra (373), and Pseudocodium (413). Additional species are considered later. December has no positive characterizations at the prescribed levels and is negatively characterized at VH conformity by Herpetopoma (221) and Malleus (279) and at H im- portance by Nephtys (90). Finally, two species characterize March negatively and September positively at H importance ; they are Eunice cf. indica (63) and Branchiostoma (394). The results show only tenuous resemblance to the preliminary conclusions on season- ality from the earlier dredging work. In both studies Protankyra negatively characterized the warmer part of the year. In the previous work Discobotellina showed no seasonality, but now it does. This may relate to the different areas of sampling because the species is uniformly present in large numbers in the northern part of Moreton Bay but appears to occur in a patchy and spasmodic fashion near Peel Island. Studies by Stephenson and Rees (1965a, b) indicate that its normal life-span exceeds a year. There are perplexing aspects to the results. One would have expected the seasonal species to be predominantly small, to include many algae and, following Kott’s (1972) 104 MEMOIRS OF THE QUEENSLAND MUSEUM interpretation of our incomplete tunicate collections, to include many tunicates. In fact the species listed earlier include many of the larger forms, for example Chaetopterus , Placamen tiara , Paphia subrugata, Chloeia and Malleus. Moreover none of the above are highly mobile with the possibility of movement into and out of the area. This is the likely explanation of the seasonality of Callianassa which here occurs as juveniles and in which adults are likely to be more than one year old (Hailstone and Stephenson, 1961). A possible explanation of this paradox is in the existence of topographical micro- patterns with accidental concentration of sampling on certain patterns in certain seasons. As indicated elsewhere much more frequent sampling is required for adequate investigation of seasonality. Studies of populations of individual species which involve size measure- ments are also required; to date these have only done on Discobotellina (Stephenson and Rees, 1965a, b) and on intertidal populations of Callianassa (Hailstone and Stephenson, 1961). Species in Year-Groups The five species with VH conformity for 1970 are : Eunice cf. antennata (63), Amphiura bidentata (343), A. catephes (344), Ophiacantha (350) and Protankyra (364). Of these Eunice, A . catephes , Ophiacantha and Protankyra have H importance, as have Chloeia (73), Amphioplus depressus (340) and Branchiostoma (394). A single species characterizes 1971 with VH conformity — Amphioplus sp. (342) with two species of H importance — lEuleanira sp. (35) and Paphia subrugata (289). With the exceptions of Chloeia and Paphia all are small species and many are surface-dwelling ophiuroids whose populations might be expected to change from year to year. The surprising exclusions from the list are the smaller tunicates — these were below the arbitrary levels which were selected. DISCUSSION Introductory The objective of the present work was to investigate a suspectedly complex benthic biota and to attempt to resolve the complexity into a number of patterns in terms of area and of time. We use the term ‘complexity’ in deliberate avoidance of ‘diversity’ which means so many things to so many people. Fifteen stations were sampled in quintuplicate at each of four seasons for two years. The method of analysis with which we began was developed by use of a selected array of the present data, has already been described (Williams and Stephenson, 1973), and has been discussed earlier in the present paper. It involves three arbitrary elements. The first is the stringency of transformation (we used cube roots), the second the level of data reduction (this we modified, excluding initially species contributing less than 0*5% of the variance to the two initial analyses), and the third the sorting strategy (we used flexible). Site-Groups We first examined the site-groupings obtained using species as attributes (summed over all times). The past literature on soft-bottom benthos has sometimes stressed site- groupings and sometimes species-groupings, and there is uncertainty amongst ecologists generally as to which are the most important. Noy-Meir’s (1970) canvassing of opinion of non-marine ecologists suggested a preference for site-groups. If, following Dahl (1908), BENTHIC FAUNA OF SOFT BOTTOMS 105 communities are delimited biotopically, a view which appears (unforunately) to be gaining ascendency in ecological texts (see Stephenson, 1973), then clearly site-groups are at the core of the community concept. While we later concentrate on species-groupings, site- groupings are of great importance for two reasons. The first is that we can use their extrinsic abiotic attributes as criteria for choosing between classificatory methodologies (see Stephenson, Williams and Lance, 1970; Stephenson and Williams, 1971; Stephenson, 1973). The second is that we can use site-groups via two-way coincidence tables to link the species-groups to the abiotic environmental features. We used the first in preliminaries to the present work and compared the site-groups which were generated by different transformations of three sub-sets of the present data. It was on this basis that a cube-root transformation was selected. As regards the second, there is an extraordinary resemblance between the site-groups generated by using species as attributes and the sediment classification of the sites. The correspondence is so close that the effects of other abiotic variables are completely masked. This almost exact relationship was not apparent when date for our earlier paper (Williams and Stephenson, 1973) were at hand. Sediment data for this earlier work were obtained at the start of the survey (March 1970), the data used in the present work was the average of values for March 1970 and December 1971. The differences between these two sets of values were greater than expected, and because sediment grades are so important it is evident that much more data should have been collected. This would have indicated whether sediments were changing on a short term basis (for example between spring and neap tides), on a long term basis (over several years) or whether the differences in the sets of data are due to different ‘random’ samplings of topographical micropattems. An original objective was to determine whether there were small scale area patterns in the area. They are clearly present. Within one area (I) of 0-24 km 2 there is one site-group and part of a second, and within another (area III) of 0-36 km 2 there are elements of three site-groups. Within our area it would have been desirable for the intersite distances to have been no greater than 0*25 km for adequate delineration of the kind of group we obtained. This indicates that close spacing is likely to be desirable in inshore sub-tropical areas with varying sediments and under some estuarine influences. Pearson (1970) has already suggested that the scale of inshore patterns is finer than offshore. We have already indicated (Stephenson, Williams and Cook, 1972) that Petersen’s (1914) stations were too distantly spaced to show coherent area patterns, and the present work confirms this opinion. A second kind of site-grouping was provided by the analyses — this is using times as attributes (with summation of species). It gives little conceptual advantage over considera- tion of the summed values of all species over all times in the sites — as in Table 5. The main conclusion is that the muddy sites (1, 2, 3 and 5) have the lowest total populations. Species in Sites As the present work progressed it became evident that interest was primarily focussed on species. It was equally evident that the ‘inverse’ classifications as revealed in the dendro- grams were not producing optimal sense. This led to the first important modification of 106 MEMOIRS OF THE QUEENSLAND MUSEUM the Williams and Stephenson (1973) methodology. In the previous paper in order to produce a single two-way table showing site-groups and their constituents and species- groups with their constituents, we had to marry classifications using two different sets of derived data. These were site-centred deviations as used in the site classification and species-centred deviations as used in the species classification. We did this, in tables desig- nated ‘B’, by taking the mean of the two differently centred values. In an attempt to resolve some of the problems of species classification we considered only species-centred values. This gave slightly greater conceptual sense, but difficulties remained. Their resolution, such as it was, was affected by concentrating on the species-centred values in site-groups obtained by using site-centred data. The computer classification we used is based upon the euclidean distances of species one from another (using cube-root data) and these distances are greater with the more abundant species. The less abundant species tend to be ‘lumped’ into what are virtually discard groups. The conceptual aim in a species/site situation is to determine which species characterize a site by predominantly occurring there (or by predominantly not occurring there). This predominance is not necessarily proportional to the abundances of the species, indeed on the assumption that species of high fidelity are relatively uncommon, the reverse might be expected. In the species-in-sites classification, species-groups derived by variance of differences were acceptable at different levels on the dendrogram, and only at lower levels for the less common species. It was necessary to redesign the upper parts of the hierarchy by common- sense methods. Neither the original strategy nor its commonsense substitute were satisfactory for deciding in doubtful cases whether a species was predominant in a particular site-group. Here a ‘test of significance’ was used — the ratio of between site-group variances to within site-group variances (the F test). The affinities between species on different branches of the dendrogram became obvious when this ratio was employed. The availability of an approximate test of significance of a given species to a pre- determined site-grouping is an important auxiliary advantage of the present methodology. It allowed us to reject species of low conformity to the grouping, and also to scan the total species list (beyond those with high site variance values) to pick out further conformers. As a result almost one quarter of the biota are known to conform to the site-groups we had specified. Having selected some species with high contributions to site-variance which did not conform to the original site-groupings we added species lower on the site-variance list and searched for an alternative site-grouping. None eventuated in the present case. There is no certainty that this would be so in other situations and it is quite possible that an area might contain sediment-sensitive and pollution-sensitive species which could generate alternative site-groupings. To be able to explore such possibilities is another advantage of the present technique. The tests of significance were used to produce grades of conformity of species to site- groups, and we have used the term ‘conformity’ throughout. In our previous numerical analyses of benthic data we have characterized species by the terms long used by terrestrial BENTHIC FAUNA OF SOFT BOTTOMS 107 botanists — dominance, constancy and fidelity. Previously the stress has been on dominance and constancy (see Williams, Stephenson and Lance, 1970; Stephenson and Williams, 1971 ; Stephenson, Williams and Cook, 1972). ‘Conformity’ combines the three concepts. Thus high positive conformity of a species to a site-group means that most individuals occur in that group. This is irrespective of the actual numbers there present. High positive conformity can occur with an uncommon species, in which case it equates to fidelity. It can also occur in an abundant species, in which case it is likely to equate to dominance/constancy. Another type of conformity occurs, although more obvious in later analyses. This is high negative conformity — it means that a species ‘avoids’ a site-group. To describe this in terms of negative dominance/constancy/fidelity introduces conceptual problems, so throughout we adhere to the terminology and concept of ‘conformity’. While dominance grades in a given site-group are thus discarded, relative abundances of species in a particular data matrix are readily obtained. They are the summations between site-groups of the site-centred deviations which are used in site-grouping. Because the term ‘dominance’ has a within site-group connotation we have used the term ‘importance’ for these summated values and graded them on an arbitrary basis. We might expect that the less common species with lower importance values would have the most discrete sorting into site-groups (i.e. show the highest fidelity) but the data did not show this — neither did it apply to the species in month-groups detailed later. Times Analyses The relative importance of times and sites with respect to species is indicated in Table 7 and is about 2:3. The times-sites comparison is a measure of the changes in total population (all species summated) and is less than that of species-times in the ratio of 4:5. The chrono- logical changes in total population through time are also shown in Table 5. Attempts to classify times by species were regarded as unsatisfactory because the dimension time involves two dimensions — months and years. The time-species data were expanded to three dimensions of species x months x years. This gives, overall, a partial partitioning in four dimensions. In general terms by successive paritioning of a matrix in which one dimension is species, there is no theoretical limit to the number of dimensions which may be explored. Within the species-times data, Table 10 shows that the greatest amount of variance is between months and years — i.e. in total population of all species summated, rather than in either of the combinations involving species. This indicates considerable replacement of one set of species by another set (or sets) as time proceeds. The species-year variance is almost double that of species-seasons indicating that annual changes in species composition are more important than seasonal ones. Data on the population changes from 1970 to 1971 are ambiguous (compare Tables 5 and 14) but it is clear that if there was any population reduction due to floods, there was a delay period of about six months, as evidenced the low values of June 1971. The smallest total population was in March 1970; then and previously there had been no obvious abiotic adversities. The highest total population was in September 1971 and because this followed an exceptionally dry winter, it may indicate that the salinities in the area are sub- optimal throughout. In summary explanations of differences in populations between 1970 and 1971 are tenuous. 108 MEMOIRS OF THE QUEENSLAND MUSEUM Analyses of the species-months data were concentrated on delimiting species-groups characterizing particular months. Because there are only four months to consider, tests of significance of conformity could not be made. It is plain that, for a satisfactory study of season changes more frequent sampling, for example 12 times annually, is required. We estimated conformity by using ratios of variance and obtained arbitrary grades; because they were arbitrary we did not extend the selection to species beyond the 71 species list of Table 6. Seventeen of the 71 species were eliminated because of very low conformity, and the remaining 54 already present more data than is readily comprehensible. Values of importance for the 54 species were obtained and have been cited (Table 13). The main difference in emphasis between the site-conforming species and the month- conforming species is that the latter contained a large fraction (24/54) giving negative characterization — almost exclusively of March. Analyses of species/years were purely in terms of species, and grades of conformity are arbitrary. They were made by using the points of dissection in an essentially continuous array which were revealed by the classificatory programme. The flexible sorting strategy which was used here proved its value — it optimizes discontinuties. Again the inadequacy of the data are apparent. Because annual effects are clearly important (in the present case more important than seasonal ones), to document their reality in terms of significance and to elucidate their causes would require data extending over possibly a decade. With the present data limitations each species which positively characterizes one year gives an equal negative characterization of the other. Comparisons of the numbers of positive and negative characterizations with those of other analyses are impossible. The most important general comment on the species-times data is that it allows us to think of species-groups in seasons and species-groups in years as well as species-groups in sites. We are no longer like Petersen (1914) bound to eliminate ‘seasonal animals' from our ideas of communities, or more strictly of associations (Mills, 1969; Stephenson, Williams and Cook, 1972). We return later to the contribution of these three kinds of species-groups to an understanding of the complexity of the area. The most important pragmatic consideration is that if a sampling programme is limited to a single set of samples taken ‘instantaneously’, a great deal of relevant information is likely to be missing. We understand that such sampling programmes are normally all that are required by authorities considering whether or not an engineering project is likely adversely to effect an environment — these are the ‘environmental impact’ statements. It would clearly be impossible, even with the present fairly extensive data, for us to predict in any detail the natural changes which might occur in our sampled area. To predict with any degree of confidence, the possible effects of superimposed changes might require an initial survey of ten years duration. Complexity and Diversity In the extensive literature upon diversity which has developed since the later 1950’s considerable attention has been paid to the comparison of diversities from different situations, and particularly from different latitudes. Recent reviews have been given by Recher (1972) and by Whittaker (1972) and critical earlier papers are by Fischer (1960), Klopfer and MacArthur (1969), Saunders (1968) and Thorson (1952). At the level of BENTHIC FAUNA OF SOFT BOTTOMS 109 species richness there is an immense literature in faunistic and floristic treatments of various taxa. There seems to be general agreement that diversity comparisons should either be within ecological units or between them. The units envisioned are sometimes habitats, and thus we have within-habitat a diversity and between-habitat j8 diversity. Sometimes diversity is considered within an ‘ecological community’ (Edden, 1971) and sometimes applying to a ‘community biota’ (Hendrickson and Ehrlich, 1971). The concepts and comparisons are valid when, and only when, there is agreement on what a habitat is or an ecosystem is. Delimitations of ‘habitats’ and ‘ecosystems’ involves human judgements which seem often to be made upon undefined bases. We suspect that a ‘habitat' judgement is usually made by the human eyes or by instrumental measurements of the abiotic features in a situation. If so it is directly comparable to biotopic concepts of the community which originated with Dahl (1908). These notions seem open to severe criticism in that we deter- mine the boundaries and hope that the organisms will respect them. The history of marine soft-bottom ‘communities’ which began with Petersen (1914) has not a tradition of accepting biotopic boundaries, probably because consideration of abiotic attributes produces continua. Boundaries have been set by the organisms, the ‘communities’ are biocoenoses. In fact as already stated by Mills (1969) and Stephenson, Williams and Cook (1972) the word ‘associations’ is more appropriate. If the ‘communities’ are though of as site-groups derived by using species as attributes, if there is a complex situation and if one is not to discard data on most of the species, one is forced into one or another method of numerical classification using computer techniques. Different methodologies produce different groupings and techniques of handling data are still evolving. Some advances were made in the preliminary paper using some of the present data (Williams and Stephenson, 1973) and others are contained in the present paper. No satisfactory technique exists in which there are statistically acceptable tests of the level at which, in a classificatory hierarchy, a site-group can be accepted or rejected. In comparisons between apparently similar ‘habitats’ at different latitudes it is possible that the sizes of site-groups which one would accept would differ. In more specific terms one explanation for increased tropical diversity could be the existence of topo- graphical micropatterns of site-groups. Another term for this is ‘patchiness’, and references to this are given by Pielou (1966) and Lloyd (1967). Both deal with the topic somewhat theoretically. In studies of soft-bottom benthos there appear to have been no investigations of the scales of patterning in comparable tropical, subtropical and temperate soft bottom environments. We believe our own investigations have been of a finer scale than others and we have revealed small-scale topographical patterning. We suspect that the real scale may be finer than that which we have revealed. Meanwhile, as Pearson (1970) has indicated, there may be obvious geographical change in patterning as one moves from inshore to offshore localities. This, as well as latitudinal changes, merits detailed investigation. There are many problems centred around the time factor in ecological surveys, including the methodological problems with which we have been concerned earlier in this paper. There are also conceptual problems, for example whether or not the ‘com- munity concept’ should or should not include chronological changes. Many of the earlier 110 MEMOIRS OF THE QUEENSLAND MUSEUM definitions of communities and associations exclude the time element deliberately or by implication. As stated in an earlier paper (Stephenson, 1973): ‘For example Mobius(1877) spoke of a community in which the total of species is mutually linked under the average external conditions of life. Hesse (1924) said that the groupings correspond with average prevailing conditions [The italics are not in the originals] ... It seems that the terrestrial botanists have adopted a similar view. Their associations are characterized by such things as constancy of species, and seasonal species are clearly inconstant on a continuous chronological scale’. Later definitions involve dynamic concepts, for example Resvoy’s (1924) mention of dynamic balance, and Emerson’s (1939) concept of a chronologically adjusting super-organism. Once the notion is established that there are seasonal changes and ‘seasonal com- munities’ we are in danger of conflict with one of the shibboleths of current thinking about diversity. This is the belief that high diversity occurs in situations of high stability. This danger of confrontation cannot be avoided if there is a component of complexity which is due to different annual ‘communities’, in which there is no evidence of regularity of re-occurrence. We have demonstrated that there are differences between one year and another in our own situation and that they are more important than seasonal differences. There is nothing intrinsically wrong with the argument that a situation showing annual instability should be more complex than one with annual stability. Indeed the argument appears as if it should run : biotic uncertainty equates to abiotic uncertainty ; whence diversity equates to an environment showing differences from year to year. To this argument we should add two important riders. The first is that, in our present situation, there is no clear evidence that the difference between the two years of the investigation was directly due to abiotic events. The second is that in discussing chronological instability in abiotic factors there are likely to be dif- ferences in scale. Gross instability is likely to lead to reduction in complexity, whereas instability within ‘acceptable limits’ may well increase it. The ‘acceptable limits’ will be in terms of the range and availability of organisms capable of immediate replacements in changed areas. What applies to an inshore situation with occasional reduction of salinities, or with changes in sedimentary patterns, can not be expected to apply to the terrestrial biota of an oceanic island devastated by a volcanic eruption. Just after the above was written the senior author was given by W. Sakai (private communication) results of a startling simple demonstration of the effects of instability within ‘acceptable limits’ causing an increase in diversity. Sakai worked on clumps of mussels and showed that moderate mechanical disturbance increased diversity. ACKNOWLEDGEMENTS We are deeply grateful to the following for identifications in the groups stated. Without this massive assistance species would have been confused or eliminated : Dr P. E. Gibbs (Marine Biological Lab., Plymouth, U.K.): Polychaetes — Amphinomids, Nereids, Gly- cerids, Opheliids, Orbinids; Mrs K. Hicks (Zoology Dept., University of Canterbury, N.Z.) : Polychaetes — Lumbrinerids ; Dr P. Hutchings (Australian Museum) : Polychaetes — Terebellids, Ampharetids; Dr R. B. Spies (Dept, of Fisheries & Wildlife, Melbourne): Polychaetes — Flabelligerids ; Mr B. M. Campbell (Queensland Museum): Crustacea — BENTHIC FAUNA OF SOFT BOTTOMS 111 general; Mr C. R. Smalley (Zoology Dept, University of Western Australia): Crustacea — Alpheids; Mr M. Johnston (Zoology Dept, University of Queensland): Crustacea — Cumacea; Dr D. F. Boesch (Virginia Institute of Marine Science, U.S.A.): Crustacea — Tanaidacea; Mrs H. King (Queensland Museum), Dr R. B. Wilson and Mrs S. M. Slack- Smith (Western Australian Museum), Dr W. F, Ponder (Australian Museum) : Mollusca — Gastropods, Bivalvia; Mr W. A. Green (4/11 Wilson Street, Berwick, Victoria): Sipun- culids; Dr A. N. Baker (Dominion Museum, N.Z.): Echinoderms — Ophiuroidae, Echinoi- dea; Dr D. L. Pawson (Smithsonian Institution, U.S.A.): Echinoderms — Holothurioidea; Dr P. Mather (Zoology Dept, University of Queensland): Tunicates; Mr O. E. S. Kelly (Zoology Dept, University of Queensland): Cephalochordata ; Dr A. B. Cribb (Botany Dept, University of Queensland) : Algae, Angiosperms. We are also grateful to Mr W. Sakai (Dept of Biological Sciences, University of California, Los Angeles) for permission to quote from results of his current experiments. APPENDIX I Species Obtained, with Original Arbitrary Numbers. Protozoa Foraminifera 1 Discobotellina biperforata Collins PORIFERA 2 Cushion Sponge 3 Sponge I 4 Sponge 2 5 Sponge 3 6 Sponge 4 7 Sponge 5 8 Sponge 6 9 Sponge 7 10 Sponge 8 1 1 Sponge 9 12 Sponge 10 1 3 Sponge 1 1 Cnidaria, Anthozoa Gorgonacea 14 ‘Whip coral’ 15 ‘Rusty wire’ gorgonid Pennatulacea 1 6 Scleroblemnon sp, 17 Sea pen Ceriatharia 18 cf. Cerianthus sp. Zoantharia 19 Sphenopus marsupialis (Gmelin) Actiniarla 20 Edwardsia sp. 21 Anemone 1 22 Anemone 2 23 Anemone 3 112 MEMOIRS OF THE QUEENSLAND MUSEUM Platyhelminthes POLYCLADIDA 24 Sp. 1 25 Sp. 2 Nemertea 26 ‘Black’ 27 ‘Brown’ 28 ‘Pale’ 29 ‘Pink’ 30 ‘Speckled’ 31 ‘Sunburnt’ 32 ‘White striped’ 33 ‘Red’ 34 ‘Red and white’ Annelida, Polychaeta Aphroditidae 35 ? Euleanira sp, 36 Leanira yhleni Malmgren 37 Sthenelais boa (Johnston) 38 Sigalionid 1 39 Sigalionid 2 40 ‘Small aphroditid’ 1 41 ‘Small aphroditid’ 2 42 ‘Small aphroditid’ 3 43 ‘Small aphroditid’ 4 44 ‘Small aphroditid’ 5 45 ‘Small aphroditid’ 6 46 ‘Small aphroditid’ 7 47 ‘Small aphroditid’ 8 48 ‘Small aphroditid’ 9 49 ‘Small aphroditid’ 10 50 ‘Small aphroditid’ 1 1 51 ‘Small aphroditid’ 12 52 ‘Small aphroditid’ 13 Nereid ae 53 Ceratocephala sibogae (Horst) 54 Nereis jacksoni Kin berg 55 Leonnates stephensoni Rullier 56 Platynereis insolita Gravier 57 Websterineris punctata (Wesenberg-Lund) Eunicidae 58 Arabella tricolor (Montagu) 59 Drilonereis sp. 1 60 Driloenreis sp, 2 61 Drilonereis sp. 3 62 Eunice antennata (Savigny) 63 Eunice cf. indica Kin berg 64 Eunicid (‘4 antennae’) 65 Lumbrineris latreilli Audouin and Milne Edwards 66 L. maxillosa (Ehlers) 67 L. mirabilis (Kinberg) BENTHIC FAUNA OF SOFT BOTTOMS 68 L. mucronata Ehlers 69 Marphysa sanguined Montague 70 Onuphis sp. (‘long gill’) 71 Onuphis sp. (‘short gill’) 72 Rhamphobranchium sp. Remaining Errantia 73 Chloeia flava (Pallas) (Amphinomidae) 74 Eurythoe cf parvecarunculata Horst (Amphinomidae) 75 Glycera americana Leidy (Glyceridae) 76 G. prashadi Fauvel (Glyceridae) 77 Goniada eremita Audouin and Milne Edwards (Glyceridae) 78 Hesionid 1 (Hesionidae) 79 Hesionid 2 (Hesionidae) 80 Leocrates cf. claparedii (Costa) (Hesionidae) 81 Phyllodocid 1 (Phyllodocidae) 82 Phyllodocid 2 (Phyllodocidae) 83 Phyllodocid 3 (Phyllodocidae) 84 Pilargid 1 (Pilargidae) 85 Pilargid 2 (Pilargidae) 86 Pilargid 3 (Pilargidae) 87 Syllid 1 (Syllidae) 88 Syllid 2 (Syllidae) 89 Syllid 3 (Syllidae) 90 Nephtys dibranchis Grube (Nephtyidae) Terebellidae 9 1 Amaeana trilobata (Sars) 92 Amphitrite rubra (Risso) 93 Lanice conchilega (Pallas) 94 Loimia ingens (Grube) 95 L. medusa (Savigny) 96 Lysilla sp. 97 Pista typha Grube 98 Pista sp. 1 99 Pista sp. 2 100 Pista sp, 3 101 Streblosoma gracile Caullery 102 Streblosoma sp. 103 Telothelepus sp, 104 Terebellides stroemi Sars 105 Trichobranchus gracialis Malmgren Sabellidae 106 Sabellid 1 107 Sabellid 2 108 Sabellid 3 109 Sabellid 4 1 10 Sabellid 5 1 1 1 Sabellid 6 1 12 Sabellid 7 113 Sabellid 8 114 Sabellid 9 115 Sabellid 10 116 Sabellid 11 114 MEMOIRS OF THE QUEENSLAND MUSEUM Remaining Sedentaria 117 Amphicteis gunner i (Sars) (Ampharetidae) 1 18 Anchenoplax sp. (Ampharetidae) 119 Armandia cf intermedia Fauvel (Opheliidae) 120 Armandia sp. (Opheliidae) 121 Bucher ta sp. (Capitellidae) 122 Capitellid 1 (Capitellidae) 123 Chaetopterus variopedatus Renier (Chaetopteridae) 124 Cirriformia ankylochaeta (Schmarda) Cirratulidae) 125 Coppingeria longisetosa Haswell (Flabelligeridae) 126 Dasybranchus caducus (Grube) (Capitellidae) 127 Diplocirrus cf capensis Day (Flabelligeridae) 128 Diplocirrus sp. 2 (Flabelligeridae) 129 Euclymene spp. (Maldanidae) 130 Haploscoloplos bifurcatus Hartman (Orbiniidae) 131 Iso Ida putchella Muller (Ampharetidae) 132 Lygdamis cf. indicus Kinberg (Sabellariidae) 133 Magelona cincta Ehlers (Magelonidae) 134 Magelona sp. (Magelonidae) 135 Maldanid (‘parchment-grit tube’) (Maldanidae) 136 Mesochaetopterus minutus Potts (Chaetopteridae) 137 Notomastus giganteus Moore (Capitellidae) 1 38 Pectinaria antipoda Schmarda (Pectinariidae) 139 Petaloproctus terricola Quatrefages (Maldanidae) 140 Piromis cf. arenosus Kinberg (Flabelligeridae) 141 Piromis sp. 2 (Flabelligeridae) 142 Polydora sp. (Spionidae) 143 Polyophthalmus pictus Dujardin (Opheliidae) 144 Pseudocapitella sp. (Capitellidae) 145 Spiophanes sp. (Spionidae) 146 Spionid 1 (Spionidae) 147 Spionid 2 (Spionidae) 148 Spionid 3 (Spionidae) 149 Sternaspis scutata (Renier) ( Stern aspidae) Uncertain Families 150 Polychaete 1 151 Polychaete 2 152 Polychaete 3 Arthropoda, Crustacea Stomatopoda 153 Squilla anomala Tweedie 154 S. fasciata de Haan Decapoda 155 Achaeus brevirostris (Haswell) (Majidae) 156 A. lacertosus Stimpson (Majidae) 157 Actaea savignyi (H. Milne Edwards) (Xanthidae) 158 Actumnus sp. (Xanthidae) 159 Alpheus distinguentus de Man (Alpheidae) 160 A. stephensoni Banner and Smalley (Alpheidae) 161 Alpheus sp. near pacificus (Alpheidae) 162 Alpheus sp. 1 (Alpheidae) 163 Alpheus sp. 2 (Alpheidae) BENTHIC FAUNA OF SOFT BOTTOMS 115 1 64 Axius glyptocercus von Marthens (Axiidae) 165 Callianassa australiensis Dana (Callianassidae) 166 Calmania prima Laurie (Xanthidae) 167 Ceratoplax truncatifrons Rathbun (Goneplacidae) 168 Chlorinoides longispinus (de Haan) (Majidae) 169 Cryptodromia unilobata Campbell and Stephenson (Dromiidae) 170 Cryptopodia queenslandi Rathbun (Parthenopidae) 171 Dorippe australiensis Miers (Dorippidae) 172 ? Globopilumnus sp. (Xanthidae) 173 ‘Hermit crab’ (Paguridae) 1 74 Hexapus granuliferus Campbell and Stephenson (Goneplacidae) 175 Hyastenus convexus Miers (Majidae) 176 H. diacanthus (de Haan) (Majidae) 177 Leucosia ocellata Bell (Leucosidae) 178 L. pubescens Miers (Leucosidae) 1 79 Leucosia sp. (Leucosidae) 180 Libystes paucidentatus Stephenson and Campbell (Portunidae) 181 Macrophthalmus sp. (Ocypodidae) 182 Myra affinis Bell (Leucosidae) 183 M. australis Haswell (Leucosidae) 184 Nursia plicata sinuata Miers (Leucosidae) 185 Parthenope harpax (Adams and White) (Parthenopidae) 186 Parthenope sp. (Parthenopidae) 187 Phalangipus australiensis Rathbun (Majidae) 188 Pilumnoplax sp. (Goneplacidae) 189 Pilumnus contrarius Rathbun (Xanthidae) 1 90 P. minutus de Haan (Xanthidae) 191 Pilumnus sp. (Xanthidae) 192 Pinnotheres spinidactylus Gordon (Pinnotheridae) 193 Pisidia dispar (Stimpson) (Porcellanidae) 194 P. cf. spinutifrons (Miers) (Porcellanidae) 195 Polyonyx transversus (Haswell) (Porcellanidae) 196 Raphidopus ciliatus Stimpson (Porcellanidae) 197 Rhizopa gracilipes Stimpson (Goneplacidae) 198 Scyllarus sordidus (Stimpson) (Scyllaridae) 199 Thalamita sima H. Milne Edwards (Portunidae) 200 Typhlocarcinops decrescens Rathbun (Goneplacidae) 201 T. tonsurata Griffin and Campbell (Goneplacidae) 202 Xenophthalmoides dolichophallus Tesch (Goneplacidae) 203 Xenophthalmus pinnotheroides White (Pinnotheridae) Remaining Crustacea 204 Whiteleggia stephensoni Boesch (Tanaidacea) 205 Cyclaspis sp. 1 (Cumacea) 206 Cyclaspis sp. 2 (Cumacea) 207 Heterotanais sp. (Tanaidacea) 208 Pomacuma australiae (Zimmer) (Cumacea) Mollusca Gastropoda 209 Adamnestia thetidis (Hedley) (Scaphandridae) 210 Bedeva hanlyi Angas (Muridae) 21 1 Cancellophora amasea Iredale (Cancellariidae) 212 Cerithiopsis (Notoseila) crotea Angas (Cerithiopsidae) 116 MEMOIRS OF THE QUEENSLAND MUSEUM 213 Clavus sp. (Turridae) 214 Columbella (Atilia) conspersa Gaskoin (Pyrenidae) 215 Conuber conica (Lamarck) (Naticidae) 216 Daphnella cheverti Hedley (Turridae) 217 Diali sp. 1 (Rissoidae) 218 Diali sp. 2 (Rissoidae) 219 Etrema catapasta Hedley (Turridae) 220 E. spurca Hinds (Turridae) 221 Herpetopoma atratus Gmelin (Trochidae) 222 Inquisitor cf. lassulus Hedley (Turridae) 223 I. metcalfei Angas (Turridae) 224 Latirus recuvirostris (Schub, Wagn.) (Fasciolariidae) 225 Mesophora bowenensis Laseron (Triphoridae) 226 Nassarius pictus (Dunker) (Nassariidae) 227 Natica cf. colliei Recluz (Naticidae) 228 N. vitellus L. (Naticidae) 229 Natica sp. (Naticidae) 230 Philine angasi Crosse and Fisher (Philinidae) 231 Pseudorhaphitoma cf. axicula (Hedley) (Turridae) 232 Pupa sp. (Acteonidae) 233 Reticunassa paupera Gould (Nassariidae) 234 Talopia morti Iredale (Trochidae) 235 Muricid 1 (Muricidae) 236 "Bedeva” fischerianus (Tapparone canefri) (Muricidae) 237 Nudibranch 1 (Nudibranchia) 238 Nudibranch 2 (Nudibranchia) 239 ‘Sea-hare’ 1 (Opisthobranchia) 240 ‘Sea-hare 1 2 (Opisthobranchia) 241 Triphorid (Triphoridae) 242 Trochid (Trochidae) 243 ? Daphnella sp. (Turridae) Scaphopoda 244 ‘ Dental ium’ Pelecypoda 245 Aerosterigma oxygonum Sowerby (Cardiidae) 246 Antigona lamellaris Schumacher (Veneridae) 247 Area ( Area) navicular is Bruguiere (Pinnidae) 248 Atrina (Servatrina) pectinata L. (Pinnidae) 249 Azorinus abbreviatus (Gould) (Solecurtidae) 250 Chama fibula Reeve (Chamidae) 25 1 C. pulchella Reeve (Chamidae) 252 Chlamys ( Annachlamys) leopardus Reeve (Pectinidae) 253 C. (Chlamys) gloriosa Reeve (Pectinidae) 254 C. (C.) grossiana Iredale (Pectinidae) 255 Circe sugillata Reeve (Veneridae) 256 Clementia strangei Reeve (Veneridae) 257 Crytomya cf. elliptica A. Adams (Myidae) 258 Cycladicama (Toralimysia) sp. (Ungulinidae) 259 Decatopecten (Decatopecten) strangei Reeve (Pectinidae) 260 Diplodonta (Zemysina) sp. (Ungulinidae) 261 Dosinia (Dosinia) cf. sculpta Hanley (Veneridae) 262 ‘Modiolus’ ostenatus Iredale (Mytilidae) BENTHIC FAUNA OF SOFT BOTTOMS 117 263 Ensiculus hilaris Iredale (Cultellidae) 264 Eufistulina sp, (Teredinidae) 265 Fulvia sp. (Cardiidae) 266 Gari (Gari) cf. simplex Sowerby (Psammobiidae) 267 Gari venta Iredale (Psammobiidae) 268 Gari sp. (Psammobiidae) 269 Laternula vagina (Reeve) (Latemulidae) 270 Leptomya pura Angas (Semelidae) 27 1 L. cf. pura Angas (Semelidae) 272 Limaria (Limaria) cf. delicatuta Iredale (Limidae) 273 Lunulicardia subretusum (Sowerby) (Cardiidae) 274 Macoma cf. donaciformis Deshayes (Tellinidae) 275 Macoma (Salmacoma) cf. vappa Iredale (Tellinidae) 276 Macoma sp. 1 (Tellinidae) 277 Macoma sp. 2 (Tellinidae) 278 Mactra ( Electromactra) angulifera Deshayes (Mactridae) 279 Malleus albus Lamarck (Malleidae) 280 Modiolus micropterus Deshayes (Mytilidae) 281 ‘ Modiolus ' cf. pulvillus Iredale (Mytilidae) 282 Musculus cumingiana Reeve (Mytilidae) 283 Neosolen correctus Iredale (Cultellidae) 284 Notocorbula hydropica Iredale (Corbulidae) 285 Nucula ( Leionucula) astricta Iredale (Nuculidae) 286 N. (L.) orekta Iredale (Nuculidae) 287 Ostrea cf. cristagalli L. (Ostreidae) 288 Paphia gallus Gmelin (Veneridae) 289 P. subrugata Iredale (Veneridae) 290 P. (Paphia) textile Gmelin (Veneridae) 291 P. (P.) undulata Bom (Veneridae) 292 P. (P.) cf. undulata Bom (Veneridae) 293 Phragmorisma sp. ? (Thraciidae) 294 Pinctada albina sugillata Reeve (Pteridae) 295 Pinna sp. (Pinnidae) 296 Placamen sidneyense (Menke) (Veneridae) 297 P. tiara (Dillwyn) (Veneridae) 298 Regozara flava L. (Cardiidae) 299 Scapharea (Cunearca) hubbardi (Iredale) (Arcidae) 300 Solecurtis sp. (Solecurtidae) 301 Solen vagina L. (Solenidae) 302 Spondylus wrightianus Crosse (Spondylidae) 303 Syndosmya sp. (Semelidae) 304 Tapes (Tapes) watlingi Iredale (Veneridae) 305 Tellina (Semelangulus) cf. brazieri Sowerby (Tellinidae) 306 T. (S.) lilium Hanley (Tellinidae) 307 T. (S.) semitorta Sowerby (Tellinidae) 308 T. (S.) cf. solenella Deshayes (Tellinidae) 309 Tellina sp. (Tellinidae) 310 T. ( Laciolena) texturata Sowerby (Tellinidae) 311 Tepidoleda sp. (Nuculanidae) 312 Tigamnona chemnitzi Hanley (Veneridae) 313 Timoclea (Glycodonta) cf. subnodulosa Hanley (Veneridae) 314 Trichomya hirsuta (Lamarck) (Mytilidae) 315 Trisidos trisidos L. (Arcidae) 118 MEMOIRS OF THE QUEENSLAND MUSEUM 316 Tucetilla tenuicostata (Reeve) (Glycymeridae) 317 Aloidid 318 Venerid 1 319 Venerid 2 320 Bivalve E 321 Bivalve H 322 Bivalve I 323 Bivalve J 324 Bivalve Z Polyzoa 325 Polyzoan Echiuroidea 326 Listriolobus bulbocaudatus Edmonds SlPUNCULIDAE 327 Aspidosiphon sp. 328 Golfingia longirostris (Wesenburg-Lund) 329 Sipunculus sp. 330 Thermiste sp. Phoronidae 331 Phronid 1 332 Phronid 2 Branchiopoda 333 Lingula tumidula Reeve Phylum Uncertain 334 species 1 335 species 2 336 species 3 337 species 4 Echinodermata Astroidea 338 Luidia sp. 339 Sea star Ophiuroidea 340 Amphioplus depressus (Ljungman) 341 Amphipholis loripes Koehler 342 Amphioplus sp. 343 Amphiura bidentata H. L. Clark 344 A. catephes H. L. Clark 345 A. magnesquama H. L. Clark 346 A. octacantha H. L. Clark 347 A. septemspinosa H. L. Clark 348 A, tenuis H. L. Clark 349 Amphiura sp. 350 Ophiacantha confusa Koehler 351 Ophiactis perplexa Koehler 352 O. savignyi Muller and Troschel 353 Ophiactis sp. 354 Ophiocentrus sp. 355 Ophionephthys sp. BENTHIC FAUNA OF SOFT BOTTOMS 119 356 Ophionereis stigma H. L. Clark 357 Ophiothrix stelligera Lyman 358 Ophiura kinbergi Ljungman Echinoidea 359 Brissopsis luzonica (Gray) 360 Hypselaster jukesii (Gray) 361 Salmacis belli Doderlein Holothurioidea 263 Holothuria spinifera Theel 363 Mensamaria intercedens (Lampert) 364 Protankyra sp. 365 Thyone papuensis Theel Tunicata 366 Adagnesia opaca Koti (Agnesiidae) 367 Agnesia glaciata Michaelson (Agnesiidae) 368 Aplidium sp. (Polyclinidae) 369 Ascidia aclara Kott (Ascidiidae) 370 A. sydneyensis Stimpson (Ascidiidae) 371 Botrylloides nigrum Herdman (Styelidae) 372 Cnemedocarpa floccosa Sluiter (Styelidae) 373 Eugyra moretonensis Kott (Molgulidae) 374 Microcosmus australis Herdman (Pyuridae) 375 M. exasperatus Heller (Pyuridae) 376 M. nichollsi Kott (Pyuridae) 377 M. spinifera Herdman (Pyuridae) 378 M, stolonifera Kott (Pyuridae) 379 Microcosmus sp. (Pyuridae) 380 Molgula diversa Kott (Molgulidae) 38 1 M. exigua Kott (Molgulidae) 382 M. rima Kott (Molgulidae) 383 M. mollis Herdman (Molgulidae) 384 M. sphaera Kott (Molgulidae) 385 Polycarpa fungiformis Herdman (Styelidae) 386 P. pedunculata (Heller) (Styelidae) 387 P. tinctor (Quoy and Gaimard) (Styelidae) 388 Pyura vittata Stimpson (Styelidae) 389 Styela plicata Lesuer (Styelidae) 390 S. ramificata Kott (Styelidae) 391 S. stolonifera (Herdman) (Styelidae) 392 Sycozoa pedunculata Quoy and Gaimard (Clavelinidae) Balanoglossidae 393 Glossobalanus hedleyi Hill Cephalochordata 394 Branchiostoma moretonensis Kelly Algae 395 Acetabularia caliculus Quoy and Gaimard 396 Asparagopsis taxiformis (Delile) Collins and Hervey 397 Ceramium sp. 398 Champia parvula (C. Ag.) Harv. 399 Chondria sp. 120 MEMOIRS OF THE QUEENSLAND MUSEUM 400 Dictyota dichotoma (Huds.) Lamx. var intricata (G. Ag.) Grev. 401 Gracilaria textorii (Suringar) De Toni 402 G. verrucosa (Huds.) Pap. 403 Gracilaria sp. 404 Griffithsia sp. 405 Hypnea cervicornis J. Ag. 406 Hypnea sp. 407 Laurencia rigida J. Ag. 408 Laurencia sp. 409 Martensia sp. 410 Microcoleus lyngbyaceus (Kuetz.) Crow 41 1 Polysiphonia macrocarpa Harv. 412 Polysiphonia sp. 413 Pseudocodium sp, 414 Solieria robusta (Grev.) Kyi in 415 Solieria sp. 416 Sporochnus cosmosus C. Ag. 417 S. harveyanus J. Ag. 418 Spyridia filamentosa (Wulf.) Harv. Angiosperma 419 Halophila ovalis (R. Br.) Hook F. 420 H. spinulosa (R. Br.) Aschers REFERENCES CITED (These exclude references to authors of species descriptions.) Birkett, L., 1953. Changes in the composition of the fauna of the Dogger Bank area. Nature, Lond. 171 : 265. Chace, F. A., 1969. Unknown species in the sea. Science, N, Y. 163: 3271. Dahl, F., 1908. Die Lycosiden oder Wolfspinnen Deutschlands und ihre stellung irn Haushalteder Natur. Nova Acta Leop. Carol. Dtsch Akad. Naturforsch. 88 : 1 74—6. Day, J. H., Field, J. G. and Montgomery, Mary P., 1971. The use of numerical methods to determine the distribution of the benthic fauna across the continental shelf of North Carolina. J. anim. Ecol. 40 : 93-123. Dean, D. and Haskin, H. H., 1964. Benthic repopulation of the Raritan river estuary following pollution abate- ment. Limnol. Oceanogr. 9: 551-63. Edden, Anne C., 1971 . A measure of species diversity related to the lognormal distribution of individuals among species. J. exp. mar. Biol. Ecol. 6 : 199-209. Emerson, A., 1939. Social co-ordination and the super-organism. In T. Just (Ed.), ‘Animal and Plant Com- munities.’ Proc. Conf. Amer. Midi. Nat. 21 : 182-209. Fischer, A. G., 1960. Latitudinal variations in organic diversity. Evolution 14 : 64-81. Ford, E., 1923. Animal communities of the level sea-bottom in waters adjacent to Plymouth. J. mar. biol. Ass. U.K. 13 : 164-224. Hailstone, T. S,, 1972. ‘Ecological Studies on the Sub-Tidal Benthic Macrofauna at the Mouth of the Brisbane River.’ Ph.D. thesis (Unpublished), Zoology Department, University of Queensland. Hailstone, T. S. and Stephenson, W., 1961. The biology of Callianassa (Trypaea) australienensis Dana 1852 (Crustacea, Thalassinidae). Pap. Dep. Zool. Univ. Qdl (12): 259-85. Hendrickson, J. A., Jr. and Ehrlich, P. R., 1971. An expanded concept of ‘species diversity’. Notulae Naturae 439 : 1-6. Hesse, R., 1924. ‘Tiergeographie auf Okologischer Grundlage.’ Pp. i-xii, 1-613. (Fisher: Jena). Holme, N. A., 1950. The bottom fauna of the Great West Bay. J. mar. Biol. Ass. U.K. 29: 163-83. 1953. The biomass of the bottom fauna in the English Channel off Plymouth. J. mar. Biol. U.K. 32 : 1-49. 1964. Methods of sampling the benthos. Pp. 171-260, in ‘Advances in Marine Biology’, Vol. 2, (Academic Press: New York). BENTHIC FAUNA OF SOFT BOTTOMS 121 Hurlbert, S. H., 1971. The nonconcept of species diversity: a critique and alternative parameters. Ecology 52: 577-86. Jones, M. L., 1961. A quantitative evaluation of the benthic fauna off Point Richmond, California. Unix. Calif. Publ. Zool. 67: 219-320. Kitamori, R., 1950. Studies on the benthos of Tokyo Bay (II). On the distribution and seasonal change of the benthos (Japanese and English abstract). Bull. Jap. Soc. scient. Fish. 16: 275-80. Klopfer, P. H. and MacArthur, R. H., 1961. On the causes of tropical species diversity: niche overlap. Am. Nat. 95: 223-6. Kott, P., 1972. Some sublittoral ascidia in Moreton Bay and their seasonal occurrence. Mem. Qd Mus. 16 (2): 233-60. Kuhlmorgen-Hille, G., 1965. Qualitative und quantitative Veranderungen der Bodenfauna der Kieler Bucht in den Jahren 1953-65. Kieler Meeresforsch. 21: 167-91. Kullback, S., 1968. ‘Information Theory and Statistics 5 . Pp. 1-399, (Dover: New York). Kullback, S., Kupperman, M. and Ku, H. H., 1962. Tests for contingency tables and Markov chains. Techno- metrics 4 \ 573-608. Lance, G. N. and Williams, W. T., 1967. A general theory of classificatory sorting strategies, I. Hierarchical systems. Comput. J. 9: 373-80. Lloyd, M„ 1967. Mean crowding. J. anim. Ecol. 36: I 30. Lloyd, M. and Ghelardi, R. J., 1964. A table for calculating the ‘equitability 5 component of species diversity. J. anim. Ecol. 33: 217-25. Lloyd, M., Zar, J. H. and Karr, J. R., 1968. On the calculation of information-theoretical measures of diversity. Am. Midi. Nat. 79: 257-72. Longhurst, A. R., 1958. An ecological survey of the West African marine benthos. Fish. Publ. Colon. Off. pp. 1 102. MacArthur, R. H., 1955. Fluctuations of animal populations, and a measure of community stability. Ecology 36: 533-6. MacArthur, R. H. and MacArthur, J. W., 1961. On bird species diversity. Ecology 42 : 594—8. MacArthur, R. H. and Wilson, E. 0., 1967. ‘The Theory of Island Biogeography 5 pp. 1-203 (Univ. Princeton Press: Princeton). McIntosh, R. P., 1967. An index of diversity and the relation of certain concepts to diversity. Ecology 48 : 392^104. Margalef, R., 1951. Diversidad de especies en las commumcades naturales. Publicaciones de Instituto de Biologia Aplicada, Universidad de Barcelona 6: 59-72. Maxwell, W. G. H., 1970. The sedimentary framework of Moreton Bay, Queensland. Aust. J. mar.freshw. Res. 21 : 71-88. Mills, E. L., 1969. The community concept in marine zoology, with comments on continua and instability in some marine communities: a review. J. Fish. Res. Bd Can. 26 (6): 1415-28. Mobius, K., 1877. ‘Die Auster und die Austerwirtschaft 5 , Berlin. (Transl., 1880, The Oyster and Oyster Culture. Rept U.S. Fish. Comm. 1880: 683-751). Moore, H. B., 1933, A comparison of the sand fauna of Port Erin Bay in 1900 and 1933. Proc. malac. Soc. Lond. 20: 285-94. Newell, B. S,, 1971. The hydrological environment of Moreton Bay, Queensland, 1967-1968. Dix. Fish. Oceanogr. Tech. Pap. No. 30, 35 pp,, (C.S.I.R.O. : Melbourne). Noy-Meir, I., 1970. ‘Component analysis of semi-arid vegetation in southeastern Australia.’ Pp. i-ix, 1-371, Appendix 1-15. Ph.D. Thesis (Unpublished), Australian National University, Canberra. Orloci, L., 1968. Information analysis in phytosociology: partition, classification and prediction. J. Theoret. Biol. 20: 271-84. Parker, R. H., 1955. Changes in the invertebrate fauna, apparently attributable to salinity changes in the bays of Central Texas. J. Paleont. 29 (2): 193-211. Pearson, T. H., 1970. The benthic ecology of Loch Linnhe and Loch Eil, a sea-loch system on the west coast of Scotland. I. The physical environment and the distribution of the macrobenthic fauna. J. exp. mar. Biol. Ecol. 5: 1-34. Petersen, C. G. J., 1914. Valuation of the sea. II. The animal communities of the sea bottom and their importance for marine zoogeography. Rep. Dan. Biol. Sta. 21 : 1-44, Appendix 24 pp. 122 MEMOIRS OF THE QUEENSLAND MUSEUM Pianka, E. R., 1966. Latitudinal gradients in species diversity: a review of concepts. Am. Nat. 100: 33-46. Pielou, E. C., 1966. Species-diversity and pattern-diversity in the study of ecological succession. J. Theor. Biol. 10: 370-83. 1969. ‘An Introduction to Mathematical Ecology.’ Pp. viii + 286, (Wiley, Interscience: New York). Poulsen, E. M., 1951. Changes in the frequency of large bottom invertebrates in the Limfjord in 1927-50. Rep. Dan. biol. Sta. 53: 17-35. Raymont, J. E. G., 1949. Further observations on changes in the bottom fauna of a fertilised sea-loch. J. mar. biol. Ass. U.K. 28: 9 19. Raphael, Y. Irene and Stephenson, W., 1972. The macrobenthic fauna of Bramble Bay, Moreton Bay, Queens- land. I. Preliminary studies. Kept to Comm. Dept Works and Qd Dept Coord. General, Oct. 1972, 32 pp. Recher, H. F., 1972. Bird species diversity: a review of the relation between species number and environment. In ‘Quantifying Ecology’. Proc. ecol. Soc. Aust. 6: 135-52. Resvoy, P. D., 1924. Zur definition des Bioconose — Begriffs. Russk. gidrobiol. Zh. 3: 204-209. Sanders, H. L., 1968. Marine benthic diversity: a comparative study. Am. Nat. 102: 243-282. Sandisson, E. E. and Hill, M. B., 1959. The annual repopulation of Lagos Harbour by sedentary marine animals. In Sears, Mary, ‘Reprints of the 1st International Oceanographic Congress, 31st Aug-12th Sept, 1959’. Pp. 584-5, (AAAS : Washington), Segestr&le, S. G., 1960 Fluctuations in the abundance of benthic animals in the Baltic area. Comment, biol. 23(9): 1-19. Simpson, E. H,, 1949. Measurement of diversity. Nature , Lond. 163: 688. Slack-Smith, R. J., 1960. An investigation of coral deaths at Peel Island, Moreton Bay, in early 1956. Pap. Dep. Zool. Univ. Qd 1 : 21 1-22. Stauffer, R., 1937. Changes in the invertebrate community of a lagoon after disappearance of eel grass. Ecology 18: 427-31. Stephenson, W., 1968. The effects of a flood upon salinities in the southern portion of Moreton Bay. Proc. R. Soc. Qd 80: 19-34. Stephenson, W., 1973. The validity of the community concept in marine biology. Proc. R. Soc. Qd 84 (7): 73- 86 . Stephenson, W. and Rees, May, 1965a. Ecological and life history studies of a large foraminiferan ( Discobotellina biperforata Collins, 1958), from Moreton Bay, Queensland. I. Life cycle and nature of the test. Pap. Dep. Zool. Univ. Qd 2(10): 207-23. 1965b. Ecological and life history studies of a large foraminiferan ( Discobotellina biperforata Collins, 1958), from Moreton Bay, Queensland. II. Aquarium observations. Pap. Dep. Zool. Univ. Qd 2(12): 239-58. Stephenson, W. and Williams, W. T., 1971. A study of the benthos of soft bottoms, Sek Harbour, New Guinea, using numerical analysis. Aust. J. mar. Freshw. Res. 22: 11-34. Stephenson, W., Williams, W. T. and Cook, S. D., 1972. Computer analyses of Petersen’s original data on bottom communities Ecol. Monogr. 42 (4): 387-415. Stephenson, W., Williams, W. T. and Lance, G. N., 1970. The macrobenthos of Moreton Bay. Ecol. Monogr. 40: 459-94. Steven, G. A., 1930. Bottom fauna and the food of fishes. J. mar. Biol. Ass. U.K. 16: 677-705, Thorson, G.,1952. Zurjetzigen Lage der marinen Bodentier-Okologie, Verh. dtsch. zool. Ges. 1951 : 1 15-37. 1957. Bottom communities (sublittoral or shallow shelf). In Hedgpeth, J. W., (Ed.), ‘Treatise on Marine Ecology and Palaeontology' Vol. I., Ecology, pp. 461-534. Mem. Geol. Soc. Amer. 67. (1963 reprint). Tulkki, P., 1965. Disappearance of the benthic fauna from the basin of Borholm (S. Baltic) due to oxygen deficiency. Cah. biol. mar. 6: 455-63. Ursin, E., 1952. Changes in the composition of the bottom fauna of the Dogger Bank area. Nature , Lond. 170: 324. Whittaker, R. H., 1965. Dominance and diversity in land plant communities. Science, N. Y. 147: 250-60. 1972. Evolution and measurement of species diversity. Taxon 21 : 213-51. Williams, W. T., Lance, G. N., Dale, M. B. and Clifford, H. T., 1971. Controversy concerning the criteria for taxonometric strategies. Comput. J. 14(2): 162-5. Williams, W. T., Lance, G. N. , Webb, L. J. and Tracey, J. G., 1973. Studies in the numerical analysis of complex rain-forest communities. VI. Models for the classification of quantitative data. J. Ecol. 61 (1): 47-70. BENTHIC FAUNA OF SOFT BOTTOMS 123 Williams, W. T., and Stephenson, W., 1973. The analysis of three-dimensional data (sites x species x times) in marine ecology. J. exp. mar. Biol. Ecol. 11: 207-27. Yamamoto, G., 1952. Seasonal changes of benthonic communities and the succession in the benthos caused by the production of the scallop (Japanese with English abstract). Sci. Rep. Tohoku Univ. Sci. (4) 19 (4): 302-14. Mem. QdMus. 17(1): 125^19. [1974] A MICIBDELLA AND MICOBDELLA GEN. NOV. OF EASTERN AUSTRALIA (HIRUDINOIDEA : HAEMADIPSIDAE S.L.) Laurence R. Richardson 4 Bacon St, Grafton, N.S.W. ABSTRACT Two new genera and three new species are described. Amicibdella niger sp.nov. : 5-annulate; contrast stripes lateral in the paramedian fields and along the lines of the supramarginal sense organs; nephropores in a dark band in the marginal fields; northeastern Queensland. Categories are defined for individual variations in pattern, and three forms of tertiary variation described for A. niger. Micobdella gloriosi sp.nov. : 4-annulate, contrast stripes in the dorsal median field, lateral in each inter- mediate field extending into the line of the supramarginal sense organs; nephropores in a dark band in the marginal fields; annuli of the midnephric somites equal in length; southeastern Queensland, Micobdella auritus sp.nov.: annuli of midnephric somites unequal in length; Sydney, N.S.W. Micobdella sp.?, (?) Tasmania. Modification of the paramedian and intermediate palisades of dorsoventral muscles due to the ventral reduction of xxiii to xxvii, is briefly described. Topographically defined pattern is proposed as a guide to generic separations among species exhibiting close morphological similarity. This paper provides two new genera for eastern Australian land-leeches. One is 5-annulate, known only in the type species in northern Queensland ; the second, 4-annulate, based on a species in southern Queensland, is represented by a second species in Sydney, N. S. W. , and a specimen recorded as from Tasmania. The latter locality is possibly doubtful. The two genera are defined in a combination of external and internal characteristics. Each has a distinctive pattern, as in other genera I have defined in this manner. Preserved specimens of the 5-annulate show marked variation in individual pattern. In describing this, I differentiate the principle forms of individual variation in pattern as known to me in australian Haemadipsidae, some forms having such high frequency as to be characteristic of the species and systematic in value, as in this 5-annulate. The Haemadipsidae occur in the Oriental, Australian, Oceanic, and Malagasian zoogeographic divisions. As known the family has been small, containing (Soos, 1967) 3 1 species in 9 genera, three of the genera based on australian species with a fourth genus in Papua based on an oriental species. Elsewhere, I will add two other new genera for Papua. Below, I give indications for other new genera in eastern Australia. The diversity here is adequate for the testing of the principles on which the family had long been systematized. One such principle has been that pattern is so highly variable as to be without syste- matic value (v. Blanchard, 1917). This was formulated in earlier practice in which ‘genera’ were characterized essentially on simple external meristic morphology. Such genera are 126 MEMOIRS OF THE QUEENSLAND MUSEUM heterogenous in content, present zoogeographic anomalies, and are zoologically inade- quate (Soos, 1967; Richardson, 1969, 1971). I review elsewhere {Mem. Nat. Mus. Viet. 35) in more detail, the development of the concept of ‘genus’ in the Haemadipsidae. Briefly, early genera ( Haemadipsa Tennent 1859, Oriental; Chtonobdella Grube, 1866, Australian) were defined on the nature of the ocular arch, the total number of annuli, and the number of annuli between the genital pores. Then, following Whitman, from 1893 to 1897, by Blanchard on the number of annuli in the fully annulate somite ( Mesobdella , 3-annulate, Neotropical; Planobdella , 7-annulate and Phytobdella, 6-annulate, Oriental; Philaemon, 4-annulate, Australian) and species separated on the number of annuli between the genital pores or other detail in general somital annulation. Blanchard (1917) reduced Chtonobdella to synonymy under Haema- dipsa , both being 5-annulate, although Haemadipsa was based on a trignathous leech. Blanchard’s system (1917) continues to dominate the greater part of the Haemadipsidae (v. Soos, 1967). A sound basis for departure from that system was provided by Miss A. M. Lambert (1898) in her closely detailed account of the morphology of Philaemon pungens, showing this to be duognathous, the postcaeca terminating in distinctive organs, now the lambertian organs, as also (1899) the same characteristics in two 5-annulate species in Australia. Using this, Harding (1913) defined Idiobdella (Seychelle Is.) : 5-annulate, duognathous, 11 ^annuli between the genital pores, lacking auricles and lambertian organs; Augener (1931) revalidated Chtonobdella as duognathous; Moore (1938) defined Tritetrabdella (Malay Peninsula): 4-annulate, trignathous, lacking lambertian organs; Nybelin (1943), Nesophilaemon (Juan Fernandez): 4-annulate, duognathous, lacking lambertian organs. Idiobdella , Tritetrabdella, Nesophilaemon , are monotypic. Adding to these criteria the annulation of the somites anterior and posterior to the fully annulate series, and the morphology of the auricle, I established (1969) Neoterrabdella (Northern Territory): 4-annulate, duognathous, lacking lambertian organs, exceptional in having xxiv 3-annulate, xxv 2-annulate, and the auricles on xxiv to xxvii. Neoterrabdella is monotypic. I demonstrated also the distinctive morphology of the reproductive systems in the Haemadipsidae, and on this basis removed Nesophilaemon, and later (1971) Mesobdella from this family. Criteria of systematic significance additional to those above, are introduced in the present paper : jaws armed with an edentulous cutting ridge ; salivary gland papillae on the jaws of the 5-annulate ‘ Geobdella' whitmani Lambert 1899; a count of the annuli posterior to xxiii a 2 , since xxiv commonly lacks detectable somital sense organs and the annulation of these two somites cannot always be determined with full confidence ; the form, location, and the relative lengths of the lambertian organs and their ducts; the relative lengths of the two limbs of the primary loop of the female median region; and topographically defined pattern. In various combinations, these have generic value. I show here for the first time in the Haemadipsidae, a separation of species on the relative lengths of the annuli in the mid-nephric somites. With progress toward a more precise concept of genus and of species in the Haemadip- sidae, variation in pattern in the individuals of a species can be closely analysed, categorized, A MICIBDELLA AND MICOBDELLA 127 and utilized systematically. Additional to those genera which can be separated on morphological grounds, there are in the eastern australian Haemadipsidae assemblies of 4- and of 5-annulate species which are closely similar morphologically and cannot be separated in this manner. Each assembly contains groups of species, the groups exhibiting differences in the topography of pattern of the same order as the differences in the distinctive patterns of those genera which have been defined on combinations of external and internal morpho- logical characteristics. This indicates that the knowledge of the systematic morphology of the Haemadipsidae is still incomplete ; but I continue unable to locate and fill these gaps. Unless there is acceptance of the principle that topographically defined pattern is in itself a reliable indication for the separation of genera among these assemblies in eastern Australia, the result will be a division of the Haemadipsidae here into genera, some defined on morphological terms and each with its distinctive pattern ; and others in which pattern will be diversified and without systematic meaning as in the zoologically inadequate ‘genera’ of Blanchard. This paper brings us to the point where the acceptance of the principle is warranted. Amicibdella gen. nov. Derivation : amicus , friendly ; bdella, a leech, m. Haemadipsidae; duognathous; lacking salivary gland papillae; somites ix to xxii, 5-annulate (total 14); viii and xxiii, 4-annulate; xxiv, 2-annulate; 7 annuli posterior to xxiii a 2 ; auricles posterior to xxiv, the margin lobed ; wide contrast stripes lateral in the paramedian fields, narrow contrast stripes along the lines of the supramarginal somital sense organs ; nephropores included in a wide dark band extending across the marginal and submarginal fields and including the submarginal and ventral intermediate somital sense organs; teeth minute, a nearly uniform row of about 95; pharynx terminates at viii/ix; lambertian organs, posterior, each about twice the length of its duct; genital pores, xi b 5 /b 6 , xiii bj/b^ reproductive systems, haemadipsoid : anterior region of male paired ducts reflected as a single primary loop in the median splanchnic chamber, a sperm duct on the procurrent or on both limbs; ejaculatory bulbs, present; median regions, hemimyo- meric, the male a micromorphic atrium, the female formed on a posteriorly directed loop, the limbs equivalent in length, and the loop extended posteriorly as an oviducal glandular sac. Type Species: Amicibdella niger sp. nov. as below. Amicibdella is characterized by: the absence of a contrast stripe in the dorsal median field ; by the presence of a contrast stripe lateral in each paramedian field from in v to xxvii, the stripe including the intermediate sense organs along the posterior half of the body; by a narrow contrast stripe along and restricted to each line of supramarginal somital sense organs from in vi to in xxiv; by a dark band occupying the marginal and ventral intermediate fields from in v to in xxv, the band including the submarginal sense organs along its length, the ventral intermediate sense organs anteriorly, and the nephropores which are in the marginal field. 128 MEMOIRS OF THE QUEENSLAND MUSEUM It differs in general somital annulation from a 5-annuiate genus in Papua, {Mem. Nat. Mus. Viet. 35) which has viii to xxiii 5-annulate (total 16); vii, incomplete 5-annulate; the lateral margin of the auricle, straight. The general external meristic morphology and the internal morphology of Amicibdella is the same as in the other 5-annulates of eastern Australia. The majority of these have a contrast stripe in the dorsal median field from in iii (or shorter in some, from in ix) to in xxii or xxiii ; a stripe lateral in the paramedian field from in v to in xxiv ; some with one stripe in the medial half of the dorsal intermediate field, others with this stripe and a second stripe in the lateral half of this field, from in vi to in xxiii, xxiv, or xxv; the stripes of the paramedian and intermediate fields complete to much broken represented only by short to longer patches; a stripe in the ventral half of the marginal field, usually complete from the velum to the auricle, expanding into the dorsal half of this field as a lobe on the nephro- poric annulus, the lobe including the nephropore which is in the marginal field. Leeches in this assembly are known to me from southern Queensland, south to beyond Sydney, N.S.W. Moore (1944) and myself (Richardson, 1967, 1968a, b; Richardson and Hunt, 1968) were misled by Blanchard (1917) to identify leeches in this assembly as Chtonobdella limbata Grube, 1 866. 'Geobdella' whitmant Lambert 1899 of southern Queensland is the only described leech in this assembly. I find it has salivary gland papillae on the jaws, not known previously in the Haemadipsidae. The median dorsal stripe terminates in xxi; a single stripe in each intermediate field, commonly broken into patches commencing and terminating on a 2 ; a stripe in the marginal field from the velum to the auricle, has straight edges, fills the field, and includes the nephropores. A new genus will be provided elsewhere for whitmani. The 5-annulates lacking a dorsal median stripe are known to me only from north- eastern Queensland, and from eastern New South Wales south of Sydney. Those in New South Wales are Chtonobdella limbata Grube 1866. I have studied the types and recently collected material. ‘ Geobdella ’ australiensis Lambert 1899 is a synonym of limbata. In C. limbata (v. Lambert, 1899) the dorsal pattern is restricted to the posterior end of the body, from xx/xxi to xxiii/xxiv : a single patch (gold in life) on each side of the midline on the contiguous annuli of xxi and xxii, also on these annuli for xxii and xxiii, and also on the posterior annuli of xxiii ; a pair of patches in both intermediate fields at the same levels as the anterior and middle patches in the paramedian fields, together forming distinct transverse rows ; a stripe (red in life) lateral in each paramedian field from xx/xxi into the anterior half of xxiii ; erratic white patches in xxiv ; a stripe in the ventral half of the marginal field from the velum to xxiii/xxiv, expanding dorsally on the nephroporic annuli as a lobe including the nephropore. Amicibdella niger sp. nov. (Figures 1, 2, 3) Holotype : Preserved, 520 mm long. Herberton Range, nortn of the main road from Herberton to Atherton, North Queensland, map reference (1 : 100,000) CA 319850; Grassy clearing in rain-forest; Alt., 2,000 m; May 4, 1973; Collector, W. Whiteman; Per J. W. Winter. Dissected. Deposited: Queensland Museum, G5310. Paratype; Preserved, 38 5 mm long. Herberton Range, north Queensland, map reference (1:100,000) CA 3 10980; grassy forest grading into wet sclerophyll; alt., 1,200 m; May 29, 1 973; collector, J. Winter. Dissected, right ventrolateral jaw removed, mounted separately. Deposited: Australian Museum, Sydney, W4305. A MICIBDELLA AND MICOBDELLA 129 Description of Holotype (For details of the teeth, refer to the paratype description.) General Form: Moderately large; maximum extension in life, 55*0 mm. Preserved, extended, 52 0 mm long; elongate tapering cylindrical, widest along the posterior half of the body, the width equal to the depth, reducing gradually anteriorly in width and depth to the narrow anterior sucker, more abruptly posteriorly to the sucker which is about two-thirds the maximum width of the body. Total length, 52-0 mm; the width and depth, 2 0 mm at v/vi; 4*5 mm at x/xi, 10*0 mm from the tip of the velum; 6*0 mm at xiv/xv, 20*0 mm from the tip; 6*5 mm at xvii/xviii, 30 0 mm from the tip; 7 0 mm at xx/xxi, and posteriorly to 45 0 mm; reducing to the basis (4 0 mm) of the posterior sucker which is 5*0 mm in diameter. Colour: In life: the dorsum black with a pair of wide-spaced longitudinal contrast stripes, golden along the greater length of the body, pale green at the anterior and posterior ends, extending onto the dorsum of the posterior sucker to enclose a black postanal patch; these stripes expanding briefly at regular intervals and wider on the posterior portion of the body. Lateral to these golden stripes, a wide reddish brown band maculated with dark brown on the posterior portion of the body extends onto the venter, the upper and lower borders of the band maculated. A very narrow gold stripe with straight margins divides the reddish brown band along the length of the body. The venter, light brownish, immacu- late between the reddish brown bands. Preserved in formalin: The black of the dorsum diminishes in intensity; a narrow pale brownish median dorsal band with irregular dark margins becomes vaguely indicated ; green disappears from the paired dorsal stripes which, as also the outer stripes reduce to yellow; the paired reddish brown bands reduce to a pale brown with obvious dark brown maculation along both margins; the venter becomes a pale yellow. Pattern (Figs. 1, A-E): The wide median dorsal black band fills the ocular arch in ii to iv ; in v posteriorly to in xxvii, the band includes the lines of the paramedian somital sense organs, extends across the medial half of the paramedian fields, and is represented by the postanal patch on the dorsum of the sucker. The inner paired wide contrast stripes occupy the lateral half of the paramedian fields from in v to ix, progressively posterior to this expanding on a 2 , then on b 2 , a 2 , b 5 , to contact and then to include the sense organs of the intermediate lines from xv posteriorly, and to briefly enter the intermediate fields. The outer paired narrow contrast stripes occupy and are restricted to the lines of the supramarginal sense organs from in vi to xxiv/xxv dividing the side paired bands into dorsal and marginal elements. The dorsal element occupies the dorsal intermediate field from in vi to xxiii with the intermediate somital sense organs included marginally on the anterior portion of the body. The marginal element is defined as such from in vi to xxiv/xxv; occupies the marginal field, includes the nephropores which are in this field, the lines of both submarginal and ventral intermediate sense organs and the ventral intermediate field on the anterior portion of the body, but only the lateral half of this field on the posterior portion of the body. Annulation (Figs. 1, A-F): In life, moderately areolate; extended, preserved, areolae weakly defined. Somital and intersomital furrows equivalent; somital limits not directly recognizable; somital sense organs, each enclosed in a small white disc, generally 130 MEMOIRS OF THE QUEENSLAND MUSEUM Fig. 1 : Amicibdella niger gen. et sp. nov. Holotype. A, general form and dorsal pattern, showing contrast stripes in the paramedian fields, B, lateral aspect showing somital annulation of somites i to x, and detailed topography of pattern. C, the same, somites xxii to xxvii. D, the same, somites xvii and xviii. E, ventral aspect, somites xi into xiii. F, right auricle, dorsal view. Somites and somital ganglia indicated by roman numerals; somital limits, by broken lines; annuli, ‘a 2 \ etc. ; somital ganglia represented at relative size. Scales in millimeters, 1-0 mm, or as indicated. Abbreviations: an.gr., annular groove; at., atrium; au., auricle; cr., crop; dm.r., dorsomedian muscular ridge; ej.b., ejaculatory bulb; f.p., female pore; inm., intermediate somital sense organ; j., A MICIBDELLA AND M ICO B DELLA 131 obvious as transverse and longitudinal series; sensillae, minute white points, central in the areolae, nephropores, located in the marginal field, generally obscure minute apertures in the posterior half of b 2 immediately dorsal to the level of the submarginal sense organ on a 2 ; width of the dorsal median field in xvi = half the width of the somite. Somites ii and iii, uniannulate with the 1st and 2nd pairs of eyes, each eye in an ocular areola, the furrow ii/iii extending between these areolae, and iii with the 1 st pair of para- medians; the 1st nephropores on the margin of the velum, lateral to the 2nd pair of eyes; iv, 2-annulate between the ocular areolae containing the 3rd pair of eyes; v, 2-annulate above, a^ with the 4th pair of eyes = a 3 , aja 2 /a 3 extending into the intermediate line, and uniannulate v forming the lateral and ventral margins of the sucker; vi, with the 1st complete series of somital sense organs, 3-annulate above, aj = a 2 with the 5th pair of eyes < a 3 , aj/a 2 extending into the ventral intermediate field, and vi 2-annulate below, > a 3 ; vii, complete 3-annulate, a, = a 2 = a 3 (= viii a t ) ; viii, 4-annulate, a { = a 2 slightly > b 5 = b 6 ; ix to xxii 5-annulate (total 14); ix, b, =b 2 bfi-the second nephropores on b 2 ; x, bj = b 2 < a 2 = b 5 = b 6 , as also xi and xii; xiii, bj — b 2 slightly < a 2 = b 5 = b 6 ; xiv to xvi, b ] = b 2 < a 2 > b 5 — b 6 ; xvii to xix, bj = b 2 < a 2 slightly > b 5 slightly > b 6 ; xx to xxii, b, = b 2 < a 2 = b 5 = b 6 ; xxiii, bj = b 2 < a 2 > a 3 , the 16th nephropores on b 9 ; xxiv, 2-annulate, a^ = a 3 , an intermediate sense organ on the left, a^the last annulus complete across the venter; xxv-vii, incomplete, transversely abbreviated, uniannulate; anus at the posterior margin of a supernumerary annulus. Auricles, small, formed along the lateral edges of xxvi and xxvii ; margined anteriorly by xxv ; the lateral edge of the auricle with an anterior flange on xxvi separated from a posterior flange on xxvii by an open arch ; the flange, roofing over a well formed chamber little deeper than the groove elsewhere between the posterior somites and the dorsum of the sucker. Dorsum of the sucker with 5 concentric rings of areolae ; ventral surface with a central papillate area about J of the diameter of the sucker, and radiating muscular ridges dividing to terminate as about 80 at the margin. Genital pores, xi b 5 /b 6 ; xiii b 1 /b 2 . Central Nervous System (Fig. 2A) : The ventral component of the anterior ganglionic mass is compact, situated at the level of the anterior annuli in vii, with ganglion vii narrowly spaced from the ventral component and situated in the posterior annulus of vii. Ganglion viii is widely spaced from* vii, posterior in somite viii at the level of b 5 /b 6 ; ganglion ix, widely spaced from viii and at the level of a 2 /b 5 in ix. Body Wall and Muscular System: All three muscular layers in the body wall are individually distinct in dissection. jaw; l.o., lambertian organ; m.p., male pore; nepr., nephropore; n.r., connectives of anterior ganglionic masses; ov., ovary; ovd.s., oviducal glandular sac; p.c., postcaecum; pm., paramedian somital sense organ; ph., pharynx; pr., prostate; pr.l., procurrent limb; re., rectum; re.L, recurrent limb; sbm., submarginal somital sense organ; sl.g.d., aggregated salivary gland ducts; sp.d., sperm duct; spm., supramarginal somital sense organ; te., testis; v.d., vas deferens; v.gl.m., ventral glandular mass of salivary glands; v.inm., ventral intermediate somital sense organ; vl.r., ventrolateral muscular ridge; v.pm., ventral paramedian somital sense organ. J 32 MEMOIRS OF THE QUEENSLAND MUSEUM j. Fig. 2: Amicibdella niger gen. et sp. nov. Holotype. A, ventral aspect, somites v to ix opened by a median longi- tudinal incision to show the somital relationship of the ventral component of the anterior ganglionic mass and of ganglia vii to ix. B, pharynx opened along midventral line to show jaws, internal muscular ridges, column of aggregated salivary gland ducts. Inset, profile of median end of jaw. C, crop, engorged, caecation somites xviii, xix; lambertian organs; intestine; rectum. D, anterior region of male paired ducts, displaced laterally the broken lines indicating the original positions; male median region; female repro- ductive system. For abbreviations, etc., see Fig. 1. The paramedian palisades of dorsoventral muscles are obvious along the length of the crop only as clusters of strands at each intersomital level. In the intestinal region, the palisades are represented by three spaced flat pairs of strongly muscular bands, between the intestine and the postcaeca, and distinctly dorsoventrally oblique. A MICIBDELLA AND MiCOBDELLA 133 The bands appear to be at the intersomital levels: xx/xxi, xxi/xxii, xxii/xxiii. Due to the greater development of these somites on the dorsal aspect, relative to the ventral aspect, with the dorsum longer than the venter, the dorsal end of each band is attached to the body wall at a level anterior to the ventral attachment, and since the ventral aspect narrows more rapidly than the dorsal, the dorsal ends are more widely spaced than the ventral. In this way, the bands appear to be radiating from a limited ventral origin close to the base of the sucker, and more widely spaced at their dorsal insertions, as though possibly functional when the animal is erect. The bands are adherent to the wall of the intestine in the preserved specimen. The intermediate dorsoventral palisades are composed of a uniformly spaced series of individual strands. Alimentary Tract (Figs. 2, B, C): The lower surface of the velum is smooth with a subcentral trifid opening, the apex ventral. The wall of the chamber of the sucker internal to this opening, is smooth, and terminates as a ridge forming the anterior wall of a distinct continuous annular groove, housing the jaws, and the posterior wall of the groove formed by the anterior end of the pharynx. The jaws are transverse, subhorizontal, small, the base about 0-9 mm long, and the anterior end of the pharynx also carries a low obtusely triangular dorsomedian muscular pad and a similar pad in the ventromedian position, the pads much smaller than the jaws, the width not half of the length of the base of a jaw. The entrance to the pharynx is very small, restricted, no wider than the dorsomedian pad ; the wall weakly muscular, and the lumen tapering. Each jaw and the muscular pads continue posteriorly as low undivided primary internal ridges on the wall of the pharynx. There are no salivary gland papillae on the jaws. The pharynx terminates at viii/ix. The extrinsic radial musculature is an obvious system in vi, vii, viii, and in ix. The engorged compartments of the crop obscure the distribution of the salivary glands. Very thick bands of aggregated ducts enter the jaws. The caecation of the crop can only be assessed as each compartment with a pair of caeca at the anterior, and a second pair at the posterior levels ; the pairs, equivalent ; but this may not be correct since the postcaeca originate from the lateral aspects of the middle portion of the compartment in xix. The postcaeca extend in the paramedian splanchnic chamber into xxiv. The lambertian organs are ventral to the postcaeca; each elongate cylindrical, about 2-5 mm long and 0*5 mm wide; posterior in position, the anterior end at xxi/xxii; each longer than its duct which is about TO mm in length and connects subterminally to the ventral face of the postcaecum. The lambertian organs are lined with a longitudinally rugose epithelium. The intestine is compressed, tapering tubular, and connects terminally to the rectum at xxiii/xxiv. Reproductive Systems (Fig. 2D): Assessed as adult, male gravid. Typically haemadipsoid ; ejaculatory bulbs present. 134 MEMOIRS OF THE QUEENSLAND MUSEUM Testes, saccular, the anterior pair at xiii/xiv, the posterior pair at xxii/xxiii, total 10 pairs; vasa efferentia, very short, connecting laterally to the white, thin-walled, tortuous vas deferens extending anteriorly in the paramedian splanchnic chamber to the middle of xii ; reducing then to a thin-walled narrow tube which passes through the paramedian palisade, enters the median chamber, and is developed as a posteriorly directed primary loop in this chamber; a much folded sperm duct on the procurrent limb of the left loop, and on both limbs on the right loop ; the relationship, tandem. Each sperm duct terminates in a small opalescent, muscular, fusiform ejaculatory bulb, reducing abruptly into a short narrow ejaculatory duct entering the anterior aspect of the atrium. The atrium, minute, entirely ventral to the nerve cord, thin -walled, and amyomeric, micromorphic. The whole female system is ventral to the anterior region of the male paired ducts. The single pair of thin-walled saccular ovaries are posterior in xii ; each continuing as a long narrow, thin-walled transparent oviduct; the oviducts joining to form the median region. There is no obvious female atrium. The median region is formed on a posteriorly directed primary loop reflecting at xiii/xiv. The two limbs of the loop are essentially equal in length; the initial recurrent limb, a wide thick-walled tube; the terminal procurrent limb, wider than the recurrent limb, more strongly muscular, terminates at the genital pore. The oviducal glandular sac originates from the posterior aspect of the elbow of the loop, extends into the posterior half of xiv, as an inflated thin-walled sac with a glandular lining. The prostatic tissue forms a large glandular mass ventral to the nerve cord, completely investing the male atrium, the ejaculatory ducts, and extends onto the terminal portion of the bulbs. There is no indication of albumin glands investing the median region of the female system. Description of Paratype General Form: At rest, 25-0' mm long, maximum width, 60 mm; full extension, horizontal, 45-0 mm, erect 35-0 mm. Preserved, 38-5 mm. Colour : As in the type. Pattern: As the type, excepting the posterior end of the contrast stripe in the para- median field is narrower but still includes the somital sense organs of the intermediate lines. Annulation: Distinct supramarginal sense organs immediately lateral to the 2nd, 3rd, and 4th eyes. Nephropores, as in the type excepting nearly exactly at the level of the submarginal sense organs. General somital annulation, as in the type, with a 2 distincty the longest annulus in xii to xxii, bj = b 2 < a 2 > b 5 = b 6 ; xxiv, 2-annulate. Auricle, the eave more extended than in the type ; the notch deeper, wider ; but the chamber beneath the eave no deeper than the groove lateral to xxiv. Alimentary Tract : The anterior margin of the right ventrolateral jaw is 0-9 mm long ; the medial portion, of the margin, convex, the lateral portion almost straight; teeth, A MICIBDELLA AND MICOBDELLA 135 95 to 97, acutely conical and crowded along the medial half, progressively wider spaced reduced to be granular along the last quarter of the row; 0016 mm tall at the medial end, the height diminishing uniformly and gradually along the row to be 0*014 mm in the middle of the row. Differing from the type: the dorsomedian internal muscular ridge subdivided into two ridges, one wider; well-defined compartments on the crop in xiii to xviii, each with a single pair of long-based caeca originating along the middle level ; xix, with long-based postcaeca extending posteriorly to xxiii/xxiv. Lambertian organs as in type : the anterior end at xxii a 2 , the posterior end at xxiii/xxiv, and the duct short, its length no more than the width of the organ. Reproductive Systems: As in type, excepting: the sperm ducts short; the primary loops on the anterior region of the male paired ducts, parallel in relationship; primary loop of the female median region reflecting in the middle of xiii, consequently the recurrent and procurrent limbs, short, but equal in length; the posterior end of the glandular sac at the level of ganglion xiv, and the sac relatively longer. Distribution As known, northeastern Queensland, Cairns to Townsville, coastal into the Tablelands. Variation in Pattern (Fig. 3) For descriptive purposes, individual variation in pattern in the Haemadipsidae can be differentiated as : A. Primary Variation: a deviation from a precisely defined topography, a band or contrast stripe extending into the lines of sense organs or into adjacent fields for one or a few somites; erratic; commonly asymmetric, occasionally symmetric ; of such high frequency in some species as to be characteristic. B. Secondary Variation: differing degrees in length or interruption and in- completeness of contrast stripes, or a combination of both; conforms to the defined topography; asymmetric, or some measure of symmetry when interruptions occur over the same annuli in somites; of such high frequency in some species as to be characteristic. C. Tertiary Variation: forms of pattern appearing during growth; not con- stantly correlated with size; conforms to the defined topography; typically symmetrical ; of such high frequency in some species as to be characteristic. D. Pseudo variation : artefacts appearing in preserved specimens. In Amicibdella niger primary variation has been seen in few specimens : the supra- marginal sense organs lateral to the narrow contrast stripe on a few mid-body somites, the stripe then in the intermediate field — Caribou Falls, 45 0 mm, xvii b 5 to xix b 2 ; a small gold patch within the ocular arch. Secondary variation is rare, one specimen with a break in a supramarginal contrast stripe for three somites on one side. Pseudovariation is common, many preserved specimens showing a very pale narrow median band along the length of the body, the edges vaguely defined, the band related to 136 MEMOIRS OF THE QUEENSLAND MUSEUM Fig. 3: Amicibdella niger. Tertiary variation in individual pattern in three specimens from Caribou Falls (v. Additional material, 3), shown at the same size for direct comparison. The lines of the paramedian and intermediate sense organs; stipple indicating the pale brown of the body of bands in the median, para- median and intermediate fields; black, the dark brown margins on these bands, the dark brown line medial to the paramedian sense organs, and dark brown areolae including these sense organs; contrast stripes of the paramedian fields, white. A, 28 0 mm, Unistriate along the length of the body; above, anterior region of the body back into xi ; middle, xvii and xviii; below, posterior region of body, xxii to xxvii. B, 310 mm, the same, Catenulate along the length of the body. C, 45 0 mm, the same, Bistriate fused anteriorly, Catenulate on the posterior somites. A MICIBDELLA AND M ICO BD ELLA 137 the absence along this line of the very dark internal botryoidal tissue and a thin body wall. Tertiary variation has a very high frequency in A. niger , commonly very obvious in preserved specimens; only occasionally detectable in the living specimen. It has three forms: (a) Unistriate: a wide pale band extends across the dorsal median field to terminate in the paramedian field with a narrow dark margin. This form is general in specimens less than 30*0 mm and occurs also in larger specimens. In others, less than 30-0 mm to full size, a second dark line is present. This is medial to the line of paramedian sense organs and is of the width of the line lateral to them. This provides two forms, one or the other, or in combination in the one specimen ; the resultant, essentially symmetrical. (b) Bistriate : the two dark lines parallel, the paramedians in a pale band between them; the dark lines may widen, the pale band narrow, or fuse to include the sense organs. (c) Catenulate: the paramedian sense organs each in a dark areola; the two lines converging and connecting to the areolae, diverging between the areolae, and ‘chain-like* in appearance. Amicibdella niger : Additional Material In all of the following, the annuli of the midnephric somites show differences in length with a 2 distinctly long. (1) 3 specimens, 16 0, 20- 0, 48-0 mm. Ridge above Christie’s Pocket, due W. of Thornton’s Peak, Q., Rainforest; June 16, 1973; Collector, J. W. Winter. 160 mm, unistriate; 200 mm, catenulate along the greater length of the body; 48 0 mm, bistriate anteriorly, the paramedian organs each in a dark areola, then catenulate back to xxi. (2) 5 specimens, 18 0, 25 0, 32 0, 34-0, 52 0 mm. Townsville, Q. ; August, 1968; collector, M. J. Grice. 1 8*0 mm, catenulate in xv to xviii ; 25*0, catenulate ix to xvii ; 32 0 mm, bistriate, the dark lines broad, female pore in xiii b 2 ; 34 0 mm, as 32*0 but catenulate from in xix to in xxvi; 52*0 mm, engorged, catenulate xviii to in xxiv. Colour in life : As in type. (3) 6 specimens, 28*0, 28 0, 30*0, 31*0, 45*0, 53*0 mm. Caribou Falls, near Lake Eachem, N. Queensland ; August 31, 1969; collector, North Queensland Naturalists Club, Cairns. Colour in life, as in type. Pattern (Fig. 3). Preserved in alcohol, the paired contrast stripe is defined below only by a maculate line separating the margin from the venter. (4) Two specimens, 21-0, 32 0 mm. Tinaroo Creek, near Mareeba, N. Q. ; Alt., 3,000 ft. ; collector, North Queensland Naturalists Club, Cairns. In life, 32*0 mm, generally dark brown, a paler brown band in the median field ; bistriate, the lines partly fused from xii to xv, catenulate from xv to xxiv; the stripe of the supramarginal line extending briefly into the marginal field ; venter, sparsely maculate. 138 MEMOIRS OF THE QUEENSLAND MUSEUM (5) Four specimens, 19-0,29-0, 50-0, 57 0 mm. No locality; August 11,25, 1968; collector. North Queensland Naturalists Club, Cairns. 19*0 mm, viii to xii unistriate, xii to xvi bistriate, xvii to xxii catenulate ; 29-0 mm, paramedian sense organs in brown areolae, bistriate, partly fused from xi to xv, catenulate to xxiv; female pore, xiii b 2 ; 50-0 and 57-0 mm, unistriate for the length of the body. The 50 0 mm specimen has 4 small opaque white subspherical bodies, each about 0-20 mm in diameter, visible through the ventral body wall. They are erratic anatomically, and have the general appearance of parasitic cysts. (6) One specimen, 29-0 mm. Broomfield Swamp, Atherton Tableland ; in cleared rainforest, near a swamp; June 3, 1969; collector, B. Hyland; per. P. Ogilvie. Preserved, very pale brown, the dorsal band, greyish, unistriate with some very weak indications of the dark line medial to the paramedian sense organs. (7) One specimen, 29-0 mm. Forest Patch on Roaring Meg, North Daintree River, N. Queensland; June 17, 1973; collector, J. W. Winter; preserved, brown, pattern complete, catenulate viii to xxiii. (8) Two specimens, curved, approx. 17-0, 23-0 mm. Grassy woodland, 800 to 1,200 ft. Gold Hill, end of ridge W. from Thornton Peak and Enterprise Mine; June 18, 1973; collector, J. W. Winter; preserved in formalin, pattern complete. 17-0 mm, bistriate vii to xv, then catenulate; a few supramarginals lateral to the stripe in the midbody; 23-0 mm, catenulate viii to xi and xix to xxiii, bistriate between xi and xix. Micobdella gen. nov. Derivation : micere, to move to and fro with a rapid motion ; bdella , a leech, m. Haemadipsidae ; duognathous; lacking salivary gland papillae; somites viii to xxiii 4-annulate (total 16); xxiv, uniannulate; 6 annuli posterior to xxiii a 2 ; auricles, small, posterior to xxiv, the margin very weakly lobed ; a contrast stripe in the dorsal median field from ii/iii into xxvii ; a contrast stripe lateral in the intermediate field from in vi to in xxvii also includes the line of the supramarginal sense organs, and has straight edges; 1st nephropores lateral on iv, 2nd to 16th nephropores ventral in the marginal field, and included in the band in this field ; jaws edentulous, each armed with a continuous low cutting edge; pharynx terminates at viii/ix; crop compartments each with a single pair of wide based caeca at the median level ; lambertian organs in xxi to in xxii, elongate cylindrical, the organs shorter than the ducts; genital pores, xi b 5 /b 6 , xii b 5 /b 6 ; reproductive systems, haemadipsoid : anterior regions of male paired ducts reflected in a single primary loop in the median splanchnic chamber, a sperm duct on the posterior halves of both limbs of the loop, ejaculatory bulbs present, weakly muscular; median regions, hemimyomeric, the male a micromorphic atrium, the female formed on a posteriorly directed loop, the limbs of the loop equal in length, and the loop extended posteriorly as an oviducal glandular sac. Type Species: Micobdella gloriosi sp. nov. as below. Other Species : Micobdella auritus sp. nov. A MICIBDELLA AND M ICO B DELLA 139 Micobdella is characterized in the combination of a general somital annulation as in Philaemon and other 4-annulates of eastern Australia ; a contrast stripe along the length of the dorsal median field; a contrast stripe along the length of each intermediate field including the line of supramarginal somital sense organs ; the jaws armed with a continuous cutting edge. With the exclusion (Richardson, 1969) of the neotropical Nesophilaemon Nybelin 1943 from the Haemadipsidae, the previously established 4-annulate genera in the family are: Philaemon Lambert 1898; Tritetrabdella Moore 1938; Neoterrabdella Richardson 1969. Tritetrabdella is based on a species from the Malay Peninsula : viii to xxii 4-annulate ; trignathous, the jaws with 45 teeth; no salivary gland papillae; no lambertian organs. The pattern is not topographically defined. Neoterrabdella is based on a species from the Northern Territory: viii to xxiii 4- annulate ; xxiv 3-annulate ; 8 or 9 annuli posterior to xxiii a 2 ; large lobed auricles formed on xxiv to xxvii ; jaws, duognathous, with teeth ; pattern : longitudinal contrast stripes — the middle half of the dorsal median field from ii/iii to xxvii, on each line of paramedian sense organs from in viii (or shorter, in ix, or in xii) to xxiv a 2 , median in each paramedian field from in vi to xxiv a 2 , median in each intermediate field from in ix to in xxiv a 2 , and in each marginal field from the velum onto the auricle ; nephropores in the marginal fields ; no salivary gland papillae ; no lambertian organs. This combination has not been seen in any 4-annulates from eastern Australia or Papua. Philaemon is based on specimens probably from southern Victoria, possibly from Tasmania, the type species being Philaemon pungens as described by Lambert (1898). From Lambert, the characteristics are: viii to xxiii 4-annulate; xxiv,uniannulate; 6 annuli posterior to xxiii a 2 ; genital pores, xi b 5 /b 6 , xii b 5 /b 6 ; auricles, posterior to xxiv; duogna- thous, with ‘some seventy or more small teeth' ; no salivary gland papillae (v. pi. xi. figs. 10 to 12); lambertian organs in xx and xxi, the organ about half the length of the duct; pattern : a pale brown band filling the median field from in vi a 2 to in xxvii; contrast stripe (green in life) along each line of paramedian sense organs from in vi a 3 to xxiii/xxiv, the stripe distinctly and regularly narrowed on a 2 in each somite; a stripe in each marginal field at least from in x to the auricle ; location of the nephropore, ?. The combination of characteristics as in Philaemon is found in 4-annulates from North Queensland to Tasmania. These constitute a complex assembly, as yet confusingly similar in pattern and external meristic morphology, such that I continue unable to sort them with any confidence. One from Mount Glorious, southeast Queensland, differs in having also a contrast stripe in the lateral half of the intermediate field, the stripe extending to include the supra- marginal sense organs; genital pores, xi b 6 , xii a 2 . It is brown in general colour, small, the largest specimen to date, 14-0 mm long. Additional to the above, a new genus in Papua is based on a troglobitic 4-annulate lacking cutaneous pigment and pattern: the 1st nephropore on viii a } ; viii to xxiii 4- annulate ; 6 annuli posterior to xxiii a 2 ; lambertian organs in xxi and xxii, the organ much longer than the duct (as in 5-annulates) ; recurrent limb of the female median region, short, about half the length of the procurrent limb, etc. 140 MEMOIRS OF THE QUEENSLAND MUSEUM Micobdella gloriosi sp. nov. (Figures 4, 5) Holotype: Preserved, 55 0 mm long. Mt Glorious, SE. Queensland; rainforest; Feb. 28, 1972; collector, A. Hiller; per C. Wallace. Dissected, jaw removed, mounted separately. Deposited, Queensland Museum, G5309. Paratype: Preserved, total length 55 0 mm. Same locality, date, collector, as type. Dissected. Deposited Australian Museum, Sydney, W4306. Description of Holotype General Form (Fig. 4 A): Preserved, elongate, slightly depressed ; widest posteriorly, the width diminishing gradually anteriorly to the base of the narrow anterior sucker; more rapidly posteriorly to the base of the posterior sucker which is nearly as wide as the widest portion of the body. Total length, 55-0 mm; the width at v/vi, 3 0 mm; the width and depth, 3 0 mm at ix/x, 8 0 mm from the anterior tip of the animal; 4 0 mm and 3-0 mm at xi/xii, 14 0 mm; 6 0 mm and 5-0 mm at 20 0 mm ; 8*0 and 5-0 mm, at 30 mm ; 7*5 mm and 5 0 mm, at xx/xxi, 40*0 mm; the base of the posterior sucker, 5*0 mm at 53*0 mm; and the sucker 7*0 mm wide and 9*0 mm long. Colour : In life : the dorsum a deep dark brown sparsely maculate with black so as to be dusky ; divided into a pair of dark bands by a continuous longitudinal brightly yellow median stripe along the full length of the body, and a single pair of continuous longitudinal brightly yellow stripes. The paired stripes divide the dorsum from the light brownish immaculate venter. Dorsum of the posterior sucker, immaculate pale grey. Preserved in formalin : the dorsum diminishes to a pale brown excepting along the margins of the bands which are dark brown ; contrast stripes diminish to off-white or pale cream ; the venter to dusky faintly brownish pale grey. The narrow lower margin to the lateral stripe weakens to become broken, even difficult to detect. Pattern (Fig. 4, B to E) : Preserved : the median contrast stripe is continuous from ii/iii, immediately behind the 1st pair of eyes into xxvii, reaching to and enclosing the anus. It almost fills the median field from ii/iii to v ; elsewhere uniformly restricted to the middle half of the median field which is completed by the narrow medial dark margins of the dorsal bands. The medial margin of this band includes the line of the paramedian sense organs ; the paler portion of the band occupies the paramedian field and the line of the intermediate sense organs ; and the outer dark margin of the band is in the intermediate field, close to the line of intermediate sense organs in the pregenital region, progressively further from this line as the field widens along the genital and postgenital regions, with the greater part of the medial half of the field then occupied by the paler portion of the band. The single pair of contrast stripes are narrow ; commence immediately lateral to the 5th eye on vi, enclose the supramarginal sense organ close to the edge of the stripe. The stripe continues posteriorly in this relationship, occupying the lateral half of the inter- mediate field, the line of the supramarginal sense organs just within the outer edge of the stripe. A MICIBDELLA AND M ICO B DELLA 141 Fig. 4: Micobdella gloriosi gen. et sp. nov. Holotype. A, general form and dorsal pattern, contrast stripes in the median and intermediate fields. B, left dorsolateral aspect, somites i to ix, showing somital annulation and topography of pattern. C, the same, for somites xxii to xxvii. D, ventral aspect, somites xi and xii. E, Left lateral aspect, somites xxiii to xxvii, dorsum and ends of ventral muscular ridges on sucker, topography of pattern, and morphology of auricle. For abbreviations, etc., see Fig. 1, 142 MEMOIRS OF THE QUEENSLAND MUSEUM In this way, both edges of the paired contrast stripes are straight, and the stripe uniform along the greater length of the body. The paired contrast stripes divide the dorsal pattern from the marginal and ventral pattern, which is uniform excepting the narrow dark margin dorsal in the marginal field. The dark margins of the bands narrow behind xxiii. The paired stripes continue posterior to this, reducing in width and terminate enclosing the supramarginal sense organ on xxvi. There is no indication of pattern on the dorsum of the posterior sucker. Annulation (Fig. 4, B-E): Interannular and intersomital furrows equivalent; somital limits not directly recognizable ; no obvious couplets or triplets of annuli ; somital sense organs, in life everywhere encircled in white and obvious, preserved, frequently obscure, detectable with difficulty; sensillae, obvious as small white points, relatively few in all fields; annuli, strongly areolate in life, areolae detectable only in the ocular somites in well extended specimens; nephropores obvious in life as small dark points central in the length of a 1 , in the marginal field just dorsal to the level of the submarginal sense organ on a 2 . Fully extended, ii/iii is the first recognizable furrow, weak, restricted to the median field, with the 1 st eyes anterior to it ; iii, uniannulate with the 2nd pair of eyes ; iii/iv, stronger, terminating in the intermediate lines; iv, uniannulate, with the 3rd pair of eyes and 1st detectable paramedian sense organs; v, 2-annulate, a^ = a 3 , the 4th pair of eyes in a } a 2 as also the 1st detectable supramarginal sense organs, a^/aj terminating at the level of the submarginal line, and v forming the posterior portion of the surface of the velum and extending ventrally as the lateral and ventral margin of the sucker; no defined dorsolateral lobe on the margin; the 1st nephropore on the lower surface of the velum close to the lateral margin at the level of v a^/a^ vi, 3-annulate above, < a 2 < a 3 , a 2 with the 5th pair of eyes and the 1st complete set of somital sense organs, a 1 /a 2 extends into the ventral intermediate field, and below this, vi 2-annulate, a^ > a 3 ; vii, complete 3-annulate, a x < a 2 < a 3 (a 3 = viii a } ) ; viii to xxiii 4-annulate (total 1 6) ; viii, a x slightly > a 2 = b 5 = b 6 ; 2nd nephropore on ix a x ; ix to xx, a } = a 2 = b 5 = b 6 , with b 6 slightly shorter than b 5 but the dilference is so small that b 6 is not directly recognizable by length alone; xxi, a i < a 2 = b 5 = k 6 , as also xxii; xxiii, a x = a 2 slightly > b 5 = b 6 ; 6 annuli posterior to xxiii a 2 ; xxiv to xxvii, uniannulate; xxiv, complete across the venter, but reduced, and carrying minute white dorsal paramedian, slightly larger intermediate and supramarginal somital sense organs; xxv and xxvi with minute paramedians and large obvious inter- mediate sense organs; xxvii, with minute paramedians, followed by a supernumerary annulus nearly divided by the anus. Auricles distinctly small; the eave, narrow, differentiated as a thick flange on the edge of xxv, longer than a thin flange on xxvi, the two separated by a notch ; no differentiated eave on xxvii ; the groove between the eave on xxv and the dorsal surface of the sucker not appreciably deeper than the groove beneath xxiv. In life, the 17th nephropore elevated on a papilla below the notch in the eave; preserved, the papilla fully retracted and the nephropore a minute aperture on the inner wall of the groove. Dorsum of the posterior sucker divided into about five concentric rows of areolae ; A MICIBDELLA AND MICOBDELLA 143 the venter with a prominent clamp, the surface divided into radiating muscular ridges which subdivide to end as about 80 at the margin of the sucker. Body Wall and Musculature : The body wall thin, only the inner layer of longi- tudinal muscle strands detectable in dissection. Paramedian palisades of dorsoventral muscles represented along the crop by small clusters at the intersomital levels. Posterior to xix/xx, four pairs of strong bands on each side of the intestine, serial, apparently intersomital, the pairs reducing in length posteriorly ; the dorsal ends attached to the body wall anterior to and more widely spaced than the attachment to the ventral body wall where the ends are closely approximated. In this way the bands appear to rise from a common ventral attachment and radiate to individual spaced attachments on the dorsal body wall. Intermediate palisade, uniformly spaced strands back to xx. Posterior to this, three pairs of distinct bands, much narrower than the paramedian bands, are closely approxi- mated in their ventral attachment and more widely spaced dorsally. Central Nervous System (Fig. 5 A) : The anterior ganglionic mass has right and left small dorsolateral components well spaced from the compact bilobed ventral component which innervates the velum, at least somites iii and iv, v and vi. Ganglion vii in intimate contact with the ventral component and widely spaced from viii which is at the level of viiib 5 . Alimentary Tract (Fig. 5, A-C): The lower surface of the velum inflated, smooth, with a median subcentral trifid aperture; the rim of the aperture forming the anterior edge of a wide deep annular groove on the inner wall of the chamber of the sucker, the posterior wall of the groove formed by the anterior end of the pharynx ; the groove, un- divided excepting partially ventrally by a ventromedian thickening on the end of the pharynx. Duognathous; dorsomedian and ventromedian muscular pads on the end of the pharynx; the ventrolateral jaws housed in the annular groove, horizontal, the obtuse convex anterior margin 0-75 mm long, the profile at the median end, obtusely convex, as tall as wide (05 mm) at the base. The right ventrolateral jaw removed. Under high power, no indication of teeth; the edge armed with a thin low ridge, a cutting edge, 073 mm long, highest (0 03 mm) at the medial end, reducing gradually in height along its length to be 0*02 mm in the middle. No salivary gland papillae on the jaws. The entrance to the pharynx, narrow, restricted, not half the length of the anterior margin of the jaw, and closely embraced by the connectives between the dorsal and ventral components of the anterior ganglionic masses. The wall of the pharynx, thin ; internal muscular ridges present as undivided dorso- median, ventromedian, and ventrolateral ridges; no dorsolateral ridges. Salivary glands, sparse aggregations in vii, viii, and ix, and a compact ventral mass on each side posterior in viii ; the ducts joining on each side to form heavy columns of aggregated ducts terminating in the jaws. 144 MEMOIRS OF THE QUEENSLAND MUSEUM The pharynx supported by an obvious system of extrinsic radial muscles, terminates at viii/ix. The crop thin-walled, narrowly tubular without obvious compartmentation in ix and x; then partly filled with blood, the compartments each extended laterally with a single pair of wide based simple caeca at the median level in xi to xviii ; xix, with postcaeca originating from the anterior half of the compartment, the postcaeca entering the para- median splanchnic chambers and extending posteriorly into xxiii ; the outer aspect of the postcaeca, lobed. Lambertian organs located in the paramedian chambers ventral to the postcaeca; the organs from in xxi to in xxii ; elongate cylindrical, continuing as a very narrow duct connecting terminally to the postcaeca ; the organs distinctly shorter than the ducts, about one third the length of the duct. The crop terminates at xix/xx, connecting terminally to the tubular intestine which shows no indication of anterior caeca, and joins at xxii/xxiii to the end of the tubular tapering rectum. Reproductive Systems (Fig. 5D): Assessed as male mature. Typically haemadipsoid ; ejaculatory bulbs present. Genital apertures, xi b 5 /b 6 , xii b 5 /b 6 . Testes, saccular, the anterior pair at xiii/xiv, the posterior pair at xxii/xxiii, total 10 pairs; vasa efferentia, short, connecting laterally to the white tortuous vas deferens ex- tending anteriorly in the paramedian splanchnic chamber, reducing in width in xiii, and entering the median chamber at xi/xii, developed here as a simple posteriorly directed loop, the two in parallel; a much folded sperm duct occupying the posterior half of the recurrent limb of the loop and of the procurrent limb ; terminal portion of the procurrent limb narrowly tubular, terminating as a small elongate opalescent muscular ejaculatory bulb ; the bulb continues as a very short ejaculatory duct, the two ducts entering indepen- dently into the anteroventral aspect of the male atrium ; male atrium thin walled, small, standing just above the level of the ventral cord, amyomeric, micromorphic. The single pair of small saccular ovaries, posterior in xii, each continuing as a short narrow delicate oviduct; oviducts joining just anterior to the female pore to form the median region ; no distinct female atrium ; median region formed on a posteriorly directed primary loop reflecting in xiv, the two limbs of the loop essentially equal in length; the initial recurrent limb, thick walled, narrower than the strongly muscular procurrent limb ; the posterior wall of the elbow of the loop expanded as a thin-walled glandular sac extending into xv. Prostatic tissue, a large glandular mass in xi a 2 to xi/xii, adherent to the ventral body wall, enclosing and concealing the atrium, the ejaculatory ducts and the terminal ends of the bulbs. There is no glandular tissue recognizable as albumin glands. Distribution Eastern Queensland : Cairns ; Mount Glorious, Brisbane. A MICIBDELLA AND MICOBDELLA 145 viii, jaws, pharynx, aggregated salivary gland ducts and ventral glandular mass as exposed by a median incision along the ventral aspect. B, jaw and cutting ridge, arrow marks medial end. C, caecation of crop, somites xviii and xix, lambertian organs and ducts, intestine, rectum. D, E, F, Micobdella auritus sp. nov. Holotype. E, jaw with cutting ridge, arrow marks medial end. Inset, minute teeth of the medial end of the anterior margin of the jaw. F, right lateral aspect xxiv to xxvii, and the auricle. 146 MEMOIRS OF THE QUEENSLAND MUSEUM Additional Material One specimen, 34-0 mm, Under a log, Road A, about two miles from Tinaroo Lake, near Atherton, N. Q. ; Nov. 7, 1971 ; collector, North Queensland Naturalists Club, Cairns. Sent alive, dead and dessicated on arrival. Placed on damp filter-paper, recovered form, colour and pattern. Colour, as type. Preserved, dorsal bands, pale brown with narrow black margins; contrast stripes, cream ; venter white, immaculate, with an incomplete narrow pale brown margin in the marginal field. Pattern, general somital annulation, genital pores, nephropores, auricles, as type. Somites xvii, xviii, the annuli essentially subequal in length, a x and b g slightly < a 2 = b 5 , but not regularly so along the midnephric series. Micobdella auritus sp. nov. (Figures 5 E, F) Holotype. Total length, 67 0 mm. Killara, Sydney, N.S.W. (Moist valley); July 7, 1973; collector, Glen Hunt; per Dr P. Hutchings. Dissected, one jaw removed, mounted separately. Deposited, Australian Museum, Sydney, W5537. Description of Holotype General Form: A large 4-annulate land -leech, the size and general form as in M. gloriosi. Colour : In life : margin of the velum pale grey ; dorsum of the velum and of the body, intense black; contrast stripes, golden, tinged with green anteriorly; somital sense organs obscure ; venter, dark red, immaculate, excepting large dark brown maculae just ventral to the outer paired stripes, each macula restricted to an annulus; dorsum of the posterior sucker, immaculate pale grey, the venter off-white. Preserved in formalin : the dorsum remains black excepting posterior to xxiii which becomes pale reddish ; the green tinge fades from the contrast stripes which become pale yellow ; the venter and the maculae diminish in intensity. Pattern: As in M. gloriosi : a median dorsal contrast stripe with sharply defined parallel edges occupies the middle third of the median field from ii/iii to in xxvii, uniform in width along the greater length of the body, slightly wider anterior to viii ; a single pair of contrast stripes with sharply defined parallel edges, each occupying the lateral half of the intermediate field and the line of supramarginal sense organs from in vi a 2 , lateral to the 5th eye, to xxiv/xxv; nephropores below the stripe close to the line of submarginal sense organs. Annulation : General somital annulation as in M. gloriosi. Somital annulation differs : iv, 2-annulate between the ocular areolae; vi, a l < a 2 < a 3 above; viii, a x > a 2 < b 5 = b 6 ; ix, a Y < a 2 < b 5 = b 6 ; x, a ] = a 2 < b 5 = b 6 ; xiii to xvi, a x = a 2 = b 5 > b 6 ; xvii, xviii, xix, a x < a 2 > b 5 = b 6 ; xx, xxi, a x = a 2 < b s > b 6 ; xxii, xxiii, a x = a 2 > b 5 > b 6 ; xxiii b 5 , last annulus complete across the venter, b 6 reduced to a very thin ridge. A supernumerary annulus divided by the anus. A MICIBDELLA AND MICOBDELLA 147 The relative lengths of the annuli in ix to xx are distinct, and fully recognizable. Auricles, short, as an eave formed on the lateral ends of xxv and xxvi, terminating at xxiv/xxv and xxvi/xxvii, divided by a notch ; the eave covering a groove no deeper than the groove beneath the end of xxiv ; the eave more prominent than in gloriosi. Genital pores, xi b 5 /b 6 , xii b 5 /b 6 . Alimentary Tract: Velum, annular groove, housing of the jaws, entrance to the pharynx, the pharynx, crop, and intestine, as in M. gloriosi . Jaws, compressed, taller, more convex along the anterior margin than in gloriosi . The anterior margin 1-4 mm long; minute, obtuse, low conical, blunt teeth 0 006 mm high, spaced at intervals of 0 0 1 mm, detectable along the medial portion for a distance of 0-4 mm, and a low cutting ridge of about the same height along the rest of this margin. The teeth are of such small size and stand at varying angles as though non-functional. No salivary gland papillae. Lambertian organs in xxi and xxii, elongate cylindrical, each much shorter than its duct. Reproductive Systems : As in M. gloriosi Assessed as male gravid. The sperm ducts occupy almost the entire length of both limbs on the primary loop. The two loops are parallel in relationship. M. auritus differs from gloriosi in having a 2 and/or b 5 distinctly long in somites xiii to xxiii; in having minute (? non-functional, vestigial) teeth along a short portion of the anterior margin of the jaw additional to the cutting edge ; and in being black on the dorsum, the black remaining through early preservation. Distribution Known only from the holotype. This species has not been represented in the many other collections I have examined from south of Sydney, the Northern Rivers Region in N.S.W., or Queensland. Additional Material Micobdella sp.? One specimen, 44 0 mm long. Tasmania. National Museum of Victoria, Melbourne, G852. (There is no other information). Large, heavy-bodied; preserved in alcohol; faded; unsuitable for dissection. General form, pattern and topography of pattern, general somital annulation, as in Micobdella. Narrow brown lines define the edges of the contrast stripes, the median from in iii to in xxvii; the dorsal edge of the paired stripes from in viii (left), in ix (right), to in xxiv; venter immaculate. The topography of the pattern is shown clearly in xxiii as : a contrast stripe occupying the middle half of the dorsal median field ; the paired contrast stripes as in the lateral half of the intermediate field with the supramarginal sense organ immediately within the edge of the stripe ; the nephropores in the marginal field immediately dorsal to the level of the submarginal sense organ. 148 MEMOIRS OF THE QUEENSLAND MUSEUM The auricle, very small, a longer flange on xxv separated by a notch from a small flange on xxvi ; the eave compressed and hardly recognizable. Somites xv to xxi, a x = a 2 = b 5 > b 6 ; xxii, = a 2 = b 5 < b 6 ; xxiii, a t > a 2 > b 5 = b f . In this G852 resembles auritus and differs from gloriosi. It differs from the 4-annulates previously recorded from Tasmania, Philaemon pungens Lambert 1898 and P. grandis Ingram 1957, both having a contrast stripe along the lines of the paramedian sense organs. Although unsuitable for dissection, the jaws grossly contracted and unusable for detail, the combination of the somital annulation and the distinctive topographic pattern of Micobdella indicates that the specimen can be assigned with reasonable confidence to this genus. I have had no other material from Victoria or Tasmania assignable to Micobdella ; but with the experience of a single specimen from Sydney as the only evidence in many collections of this genus outside of Queensland, its occurrence in Tasmania cannot be seen as improbable. G852 is not fully documented. Lambert’s material of ‘ Geobdella austr aliens is’ and ‘ Geobdella ’ whitmani carry the collection numbers G831 to 833 on printed labels with writing in the same hand as for G852, which may be only a coincidence of cataloguing such as happens with the use of a new label. At least there is this evidence that material had come to the National Museum from southern Queensland and from central eastern N.S.W. For this reason, it seems best to accept the record of Micobdella as in Tasmania, as provisional. ACKNOWLEDGEMENTS This study initiated with collections sent to me by the North Queensland Naturalists Club, by Mr P. Ogilvie of the Forestry Department of Queensland, and Dr M. J. Grice, then of Townsville and now of Brisbane. These showed novelty which was not resolvable until I had extended my experience of the Haemadipsidae in eastern Australia. More recently, others have come to me from Mrs C. Wallace, the Queensland Museum, and associates ; from Mr J. W. Winter, National Parks Zoologist for Queensland ; and from Dr G. Hunt of Sydney through Dr P. Hutchings, Australian Museum. The majority were sent as live specimens, the later collections resolving the difficulties in the apparent changes in pattern following preservation. I do most warmly acknowledge my debt to these persons and their associates ; as also to the Directors of the Queensland Museum, the Australian Museum, and the National Museum of Victoria, to Mr B. Campbell and Mrs C. Wallace, Dr P. Hutchings, and Dr B. J. Smith of these institutions for the opportunity to study material coming to and in the collections of their institutions; to Miss E. Pope and Dr J. C. Yaldwyn, both formerly of the Australian Museum, for their help in previous years ; and to the Librarians, New England University, Armidale, the Australian Museum, and C.S.I.R.O. Melbourne, for assistance with literature; as also Professor M. C. Meyer, University of Maine, Orono, for assistance with difficult literature. This study was supported by an award from the Australian Research Grants Com- mittee. A MICIBDELLA AND M ICO B DELLA 149 LITERATURE CITED Augener, H., 1931. Resultats scientifiques du voyage aux Indes Orientales Neerlandaises — . Mem. Mus. Roy. d’Hist. Nat. Beige. (2) 6: 1-12. Blanchard, R., 1917. Monographic des Hemadipsines (Sangues terrestres). Bull. Soc. Pathol. Exot. 10 (7): 640-75. Grube, A. E., 1866. Landblutegel aus Sudaustralien. Jahresb. Schles. Ges. vaterl. Cultur. 43: 66. Harding, W. A., 1913. On a new land-leech from the Seychelles. Trans. Linn. Soc. Lond. Zool. (2) 16 (1): 39-43. Lambert, A. M., 1898. The structure of an australian land-leech. Proc. Roy. Soc. Viet. 10: 211-35. 1899. Description of two new species of australian land-leech, with notes on their anatomy. Proc. Roy. Soc. Viet. 11: 156-63. Moore, J. P., 1938. Leeches (Hirudinea) principally from the Malay Peninsula, with description of new species. Bull. Raffles Mus. 14 : 64-80. 1944, Leeches in the British Museum, mostly Haemadipsinae from the South Pacific, with description of new species. Ann. Mag. Nat. Hist. (11) 11: 383-409. Nybelin, O., 1943. Nesophilaemon n.g. fur Philaemon skottsbergi L. Johansson aus den Juan Fernandez Inseln. Zool. Anz. 142 : 249-250. Richardson, L. R., 1967. The suitability of australian land-leeches as a source of experimental material in biological researches. Aust. J. Sci. 30 (3): 107. 1968a. An annotated list of australian leeches. Proc. Linn. Soc. N.S. W. 92 (3): 227-45. 1968b. Observations on the australian land-leech, Chtonobdella limbata Grube 1866 (Hirudinea: Haemadip- sidae). Aust. Zool. 14(3): 294-305. 1969. On a distinctive new subequatorial quadrannulate land-leech, and related matters, Aust. Zool. 15 (2): 201-13. 1971. The relationship of the terrestrial jawed sanguivorous g. Mesobdella to the neotropical hirudiniform leeches (Hirudinoidea). Proc. Linn. Soc. N.S.W. 95 (3): 215-20. Richardson, L. R. and Hunt, P. J., 1968. Trypanosomes in the crop of an haemadipsid leech. Aust. J. Sci. 30 (9) : 374-5. Soos, A. 1967. Identification key to the leech (Hirudinoidea) genera of the world, with a catalogue of the species, iv. Family Haemadipsidae. Acta Zool. Hung. 13 (3-4): 417-32. Tennent, J. E. 1859. ‘Leeches’. P. 500 in: Ceylon. An account of the Islands. Vol. 1. (London). Mem. QdMus. 17(1): 151-68, pis. 4, 5. [1974] A REVIEW OF THE LITORIA NANNOTIS SPECIES GROUP, AND A DESCRIPTION OF A NEW SPECIES OF LITORIA FROM NORTHERN QUEENSLAND, AUSTRALIA (ANURA: HYLIDAE) David S. Liem Department of Zoology, University of Queensland ABSTRACT The identity of Hyla dayi Gunther was discussed and re-examined in the light of extensive study on taxonomy and biology of northern Queensland frogs. It is proposed that the original description of Hyla dayi was based on a northern Queensland Nyctimystes. Litoria dayi non Gunther is described and is named rheocolus. A new species of Litoria from northern Queensland similar to rheocolus and nannotis is described, and is named nyakalensis. Tadpoles of rheocolus and nannotis are described, and the male calls of rheocolus and nyakalensis are discussed. Litoria nannotis species group consists of three species, rheocolus, nyakalensis, and nannotis, all restricted to northern Queensland. Phylogenetic relationships amongst these three species are discussed; it is proposed that rheocolus is the most primitive, nannotis the most specialized, and nyakalensis is intermediate and probably the immediate ancestral stock of nannotis. It is suggested that the Litoria nannotis species group evolved in northern Queensland from a New Guinean ancestral stock which crossed the Torres Strait land bridge in the Pleistocene. During the course of taxonomic and biological studies on northern Queensland frogs I collected an undescribed Litoria which is closely related to Litoria dayi auctorum and Litoria nannotis. Because little is known on the biology of the above species group, the doubtful identity of Hyla dayi Gunther, and erroneous description of nannotis tadpole, it is desirable to review the whole species group which occurs in northern Queensland. MATERIALS AND METHODS Observations and drawings were carried out under a Wild stereoscopic microscope. Techniques and abbreviations of measurements not given in an earlier paper (Liem and Hosmer, 1973), are as follows : interorbital width (IO) is the narrowest width of the fronto- parietal, upper eyelid width (EW) is the widest section of the upper eyelid, tympanum width (TW) is the width of the tympanum measured along its horizontal diameter, and eye diameter (ED) is measured along its horizontal diameter. DSL is abbreviation for David S. Liem collection, and QM for Queensland Museum collection. 1 52 MEMOIRS OF THE QUEENSLAND MUSEUM SYSTEMATICS Key to Species of the Litoria nannotis Species Group 1 . Head rounded ; IN /EN ratio more than 0-970 ; web on 1 st toe on or just beyond proximal subarticular tubercle ; ventral surfaces of posterior portion of body cream with reddish brown tinge • nyakalensis Not as above . . . . . . . . . . . . . . • . . . ■ • 2 2. Snout shorter than eye diameter; snout bluntly rounded ; loreal region strongly concave ; prepollex enlarged ; nuptial pad large with coarse spinules ; single outer metacarpal tubercle; adult more than 40 mm in SV length . . . . nannotis Not as above rheocolus Litoria rheocolus sp. nov. (Figures 1 ,4C , D ; Plate 3 A) Hyta dayi (non Gunther, 1897): Loveridge, 3935, p. 38; Copland, 1957, p. 20; Tyler, 1968, p. 502. Holotype: Adult male, QM 122631 from Kuranda, 50 km west of Cairns, N. Queensland, collected 21 May 1972 by David S. Liem. Paratypes: 15 <$— QM J22632-4, DSL 6187,6588, collected together with the holotype; QM J22635-9, from Bloomsfield Rd, 50 km SE. ofCooktown, collected 14 October 1972; QM 322640-42 from Cape Tribulation, 100 km N. of Cairns, collected 11 October 1972; DSL 6522 from Henrietta Creek, Palmerston N. P., collected 17 October 1972; DSL 6517 from the Boulders, 5 km W. of Babinda, collected 10 October 1972. 5 9 — QM J22643-4 from Cape Tribulation, collected 11 October 1972; QM J22645, DSL 6547, 6554 from Shiptons Flat, 50 km S. ofCooktown, collected 14 October 1972; all localities are in N. Queensland; all specimens were collected by David S. Liem. Other Material: 56 specimens — Amos Bay, 45 km S. Cooktown, QM J22383-86; Shiptons Flat, 50 km S. Cooktown, QM J 17848; Bloomsfield Rd, 55 km S. Cooktown, QM J 17847, DSL 6536-8, 6540, 6542-3, 6545, 6550-2; Cape Tribulation 100 km N. Cairns, DSL 6176, 6181, 6191, 6548, 6527, 6530-1,6533; Kuranda, 50 km W. Cairns, DSL 6606, 6173, 6175, 6178-80, 6183, 4925-30; Davies Creek Rd, 15 km SE. Mareeba, QM J 18031 ; Tinaroo Creek Rd, 10 km SE. Mareeba, DSL 4932 ; Millaa Millaa Falls, DSL 6532^1; Henrietta Creek, Palmers- ton N. P., DSL 6518-21, 6523-4; The Boulders, 5 km W. Babinda, DSL 4839-45, 6514-16. Eleven tadpoles at stage 25 from Kuranda, 50 km W. of Cairns. All localities are in N. Queensland. Diagnosis: A moderate sized treefrog, ranging from 27 0 to 37-8 mm in SV length. It could be distinguished from nyakalensis and nannotis by the following combination of characters: 1, snout bluntly pointed; 2, prepollex normal, not enlarged; 3, nuptial pad small, spinules fine; 4, forearm of males normal, not robust; 5, webbing on outer margin of 3rd finger reaches base of penultimate phalanx. Furthermore it differs from nyakalensis by its narrow IN/EN ratio (less than 0-970), cream ventral surfaces without reddish brown tinge, and from nannotis by its smaller size (adults less than 40 mm), two outer metacarpal tubercles, snout larger than eye diameter, absence of keratin spinules on chest of males, and the males call. Description of Holotype: SV length 30-6 mm; TL 16-8 mm, 0-549 of SV length; THE LITORIA NANNOTIS SPECIES GROUP 153 HW 11-0 mm, 0-359 of SV length ; EN 3-3 mm; IN 3 0mm; IN/EN ratio 0-909; IO 3-1 mm; EW 2-6 mm; I0/EW ratio W92; TW 1-5 mm; ED 3-1 mm; TW/ED ratio 0-405. Tip of snout bluntly pointed; canthus rostralis sharp, angular in cross-section; loreal region slightly concave; dorsal view of head bluntly pointed (Fig. 1C ), rounded blunt in profile (Fig. IE ); tympanum small; supratympanic fold present; tongue subovate, slightly nicked posteriorly; vomerine teeth in short, slightly oblique, oval rows just behind the level of the posterior edge of the choanae; upper jaw toothed ; males without vocal sac. Forearm not robust ; a row of whitish tubercles present on lateral edge of forearm ; distal segment of fingers expanded into disc; with ventro-marginal groove; 3rd finger disc larger than the diameter of the tympanum; length of fingers from shortest to longest 1 -2-4-3; phalangeal formula of hand 2-2-3-3; basal segment of 1st finger not enlarged and its disc moderate in size; subarticular tubercles roundish, one each on 1st and 2nd fingers, and two each on 3rd and 4th fingers ; two small supernumerary tubercles on base of metacarpal of 3rd and 4th fingers ; two oval (small medial and an elongated lateral) outer metacarpal tubercles on base of the palm, and an elongated inner metacarpal tubercle on the proximo-ventral inner surface of the metacarpal of 1st finger. Fingers moderately webbed: between 1st and 2nd fingers free of web; web between 2nd and 3rd fingers reaching just beyond the proximal subarticular tubercle by a narrow fringe; webbing between the two outer fingers reaching by a narrow fringe the 2nd sub- articular tubercle of 3rd finger, and just beyond the 2nd subarticular tubercle of 4th finger. Nuptial pad small, and consists of small fine darkbrown keratinous spinules. Hindlimbs moderately long; distal segment of toes expanded into disc, with a ventro- marginal groove; length of toes from shortest to longest 1-2-5-4-3; phalangeal formula of foot 2-2-3-4-3 subarticular tubercles roundish, one each on 1st and 2nd toes, two each on 3rd and 5th toes, and three tubercles on 4th toe; a row of minute supernumerary tubercles is present on metatarsal of 5th toe; an oval inner and a small round outer metatarsal tubercles are present ; foot webbing reaching base of disc of 1 st toe, outer margins of 2nd and 3rd toes, and inner margin of 5th toe; web reaches by a narrow fringe the base of disc of 2nd and 3rd toes; web reaches halfway down the penultimate phalanx of 4th toe. Cloacal opening at mid-level of the thighs; skin of dorsum shagreen, with small conical tubercles scattered on the head, upper eyelids, pectoral region, and throat; postero- ventral portion of femur granular; skin fold across chest absent. Pupil horizontal, roundish to oval ; palpebral venation on lower eyelid absent. Colour: In life, dorsal colour slate; proximal portion of thighs dirty brown; sides of body greyish brown ; ventral surfaces of throat, body and anterior portion of thighs cream without pinkish tinge; palm of hands, ventral surfaces of tarsus, plantar surfaces of foot brownish slate ; posterior and postero-ventral of femur, and postero-ventral of upperarm slate tinge with tan ; margins of throat lightly spotted with brown. In preservative, dorsal colour slate, brownish tinge not distinct, and ventral surfaces cream white. Description and Variation: There is no marked difference amongst the males: fourteen male paratypes are 28-9-32-4 mm in SV length (mean 30-85 ± 3*981 mm); TL/SV 0*522-0-589; HW/SV 0*330-0-389; IN/EN 0-867-0-968; lO/EW 1-000-1-270. 154 MEMOIRS OF THE QUEENSLAND MUSEUM Male paratypes and other male specimens have similar finger webbing as the holotype ; web between the two outer fingers reaching on or just barely touching the 2nd subarticular tubercle; web on outer margin of 3rd finger reaches approximately two thirds down the length of the proximal phalanx in 5 specimens (DSL 6183, 6173, 6175, 6178, and 4930). Foot webbing is slightly variable: toe webbing in the majority of males examined reaches just the base of disc on outer margin of 2nd and 3rd toes. Females are larger than males: nine females including the paratypes ranging from 32-7-37-8 mm in SV length (mean 34*01 ± 2-181 mm); TL/SV 0-534-0-569; HW/SV 0*320-0-365; IN/EN 0-882-0*970; IO/EW 1*067-1*147. In some females the toe webbing is less extensive, e.g. in two specimens it reaches outer margin of 3rd finger approximately two-thirds down the length of the proximal phalanx, and in two others the web reaches three-quarters the length of the proximal phalanx of the 1st toe. In females the super- numerary tubercles of palm and sole are distinct, arranged in rows along the metacarpals and metatarsals; inner and outer metatarsal tubercles are small and rounded. In two females (DSL 6525-26) the conical tubercles on back are large. Colour : In life most males have similar colouration as the holotype ; in some males however (QM J22635, J22637-8, J22642), the fore and hindlimbs are dark grey with brow- nish tinge. In QM J22632, J22642, the dorsal colour is brownish with a vague broken inverted triangular marking; in others (QM J22633, and DSL 6522) the fore and hindlimbs are greyish with distinct dark spots ; ventral surfaces of hands and feet are grey with dark markings. In preservatives the following specimens are ashy grey dorsally : QM 122643^1, J22641, DSL 4927, and 4930. In life the ventral surfaces of body and throat are spotted with brown pigments or lightly dusted on margins of throat. Female colouration differs from the male; dorsal background colour brownish tan with a distinct black inverted triangular marking on the back, stretching from between the eyes down to the coccygeal region ; dorsal surfaces of fore and hindlimbs are brownish or grey, spotted with dark pigments, or with irregular vague cross-bands; background colour of ventral surfaces cream; throat, pectoral region, antero- and postero-ventral surfaces of thighs heavily dusted with darkbrown pigments; one individual (QM J22644) has a uniform cream whitish ventral surface dusted with greyish-brown on postero-ventral surfaces of the thighs. Mating Call: Regular repeated long drawn single note call, easily distinguished from the rasping call of nyakalensis. Details of the call structure will be reported elsewhere. Life History: Amplexus pairs were observed from November to March; males call throughout the year except during cold winter nights ; amplexus axillary. Males call from rocks or boulders in creeks or from vegetation overhanging water along streams and creeks. Eggs pigmentless, ovidiameter 1-4-1 -8 mm; eggs are laid in compact gelatinous clumps under rocks in water. Tadpoles at stage 25 ranging from 21T-29-0 mm in snout-tail length (ST); tail height (TH)/tail length (TL) ratio 0-375-0-414; tail tip rounded; proximal half of dorsal and ventral tail fins are thick, and taper off posteriorly; muscular tail strong, and it is twice the height of the tail fins at base of tail (Fig. 4C) ; labia expanded and forms a large sucker- mouth (Fig. 4D); mouth width (MW)/body length (BL) ratio 0-512-0*702; the sucker- THE LITORIA NANNOTIS SPECIES GROUP 155 mouth consists of two continuous rows oflabial papillae, one row along the edges of the labium, and another one on the inner surface of the mouth close to the labial tooth rows; additional labial papillar rows are present on the posterior portion of the labia, and at the anterior angles of the mouth (Fig. 4 D). Labial teeth consist of two entire upper and three entire lower labial tooth rows (II/III); the inner rows are stronger than the outer rows. Fig. 1: Litoria rheocolus. A, ventral surface of hand (DSL 6187); B, dorsal surface of hand (DSL 6518); C, dorsal surface of head (DSL 6157); E, ventral surface of foot (QM J22645); F, lateral view of head (DSL 6517). Line equals 1 mm. 156 MEMOIRS OF THE QUEENSLAND MUSEUM Upper jaw M-shaped ; the lower jaw is V-shaped ; both jaws are serrated (Fig. 4 D) ; spiracle opens at the end of a tube on ventro-lateral side of the body ; anal opening median. In life, body ground colour sandy ; abdomen black and body darkbrown ; the tail is transparent with reddish tinge ; muscular tail lightly dusted with pale brown pigments. There is no marked differences between rheocolus and nannotis tadpoles. Tadpoles live in fast flowing water current and are difficult to remove when clinging on rocks or other substrates. Efficient adaptation to fast water current is shown structurally by the large sucker-mouth, flattened head and body, and the strong muscular tail. Habitat : Litoria rheocolus is restricted along rocky fast flowing creeks and streams in rainforest as well as in wet sclerophyl forest. It ranges from sea level (Cape Tribulation and The Boulders) up to 1200 m above sea level (Mt. Lewis). This species is synchrono- patric with Litoria nyakalensis, nannotis , eucnemis , lesueuri , chloris , and infrafrenata, Nyctimystes hosmeri, tympanocryptis , Mixophyes schevili, Taudactylus acutirostris, rheophilus , Hylarana daemeli, Sphenophyrne pluvialus, Cophixalus ornatus and exiguus. Distribution : It is distributed throughout the rainforest and wet sclerophyl forest from Palmerston N.P. (near Innisfail)*in the south of Mt. Finigan (S. of Cooktown) in the north. L. rheocolus has yet to be discovered along the Cape York Peninsula rainforest chain. Etymology : The specific name is derived from the Greek rheos meaning creek or stream, and the latin suffix, colus meaning an inhabitant; it refers to the habitat of the frog. Discussion In 1 897 Gunther described and named a treefrog Hyla dayi from a single specimen collected from Mt. Bartle Frere, northern Queensland by W. S. Day. The type probably lodged in the Tring Museum is now apparently lost (Tyler, 1968; H. G. Cogger and A. G. C. Grandison, in litt.). Loveridge (1935) and Copland (1957) each refer one specimen collected to this species, and both add degrees of variation to the original description. Straughan (1966), studying 1 6 specimens, accepted Copland’s modified description but noted variable webbing between fingers. Tyler (1968) using Loveridge and Copland specimens designated these Hyla dayi (Gunther) whilst noting that they differed from the original description in extent of finger webbing, tympanum size, SV length and dorsal colouration — differences which he con- sidered unimportant. Included in a collection of practically all known northern Queensland frogs made personally during the past few years are approximately 50 specimens which correspond with Hyla dayi of Loveridge (1935), Copland (1957), Straughan (1966), and Tyler (1968), but careful examination convince me that these are not Hyla dayi Gunther. The specimens studied come from various localities in northern Queensland, as far south as Mt. Spec, north of Townsville, and as far north as Mt. Finigan, south of Cooktown, and included specimens from part of Mt. Bartle Frere close to the type locality of Hyla dayi Gunther. Specimens came from as low as sea level up to approximately 1000 m above sea level. Distribution of characters which were used in the original description of dayi , amongst the small-eared northern Queensland treefrogs are presented in Table 1. As seen from THE LITORIA NANNOTIS SPECIES GROUP 157 Table 1, SV length, dorsal colouration, and finger webbing are important and critical in establishing the status of the original description of dayi . It is unfortunate that Gunther (1897) did not define exactly the extent of webbing on the 3rd finger; he did not specify whether it reaches the base or the distal end of the pen- ultimate phalanx of the 3rd finger. From his statement ‘fingers strongly webbed’, apparently the web on the 3rd finger extends beyond the base of the penultimate phalanx, probably close to the distal end. If this is true then the web of dayi holotype differs from dayi auctorum, but is similar to Nyctimystes. Live or preserved specimens of dayi auctorum, nannotis , and the undescribed Litoria are slate or olive with dark markings, but never uniformly brown. Females of dayi auctorum are brown dorsally, but they always have a distinct large inverted triangular black marking which extends from the level of the eye to the coccygeal region; Loveridge’s (1935) specimen has a similar pattern and most likely it is a female. Amongst northern Queensland small-eared treefrogs, only females of nannotis, Nyctimystes hosmeri and probably also of tympanocryptis reach more than 50 mm in SV length. SV length of dayi holotype is substantially larger than dayi auctorum. Although dayi holotype is within the size range of nannotis females, they differ in other respects (see Table 1). Although texture of skin depends on the state of preservation, I believe that the large tubercles of dayi auctorum should have been detected were they present on the holotype. Based on the above evidence I am convinced that dayi holotype was based on a Nyctimystes female; since uniform brown colour variant is present in tympanocryptis , hosmeri and vestigea, (The validity of vestigea is discussed by Liem, in MS). This makes specific identification impossible. Litoria nyakalensis sp. nov. (Figure 2 ; Plate 4) Holotype: Adult male, QM J22624 from Henrietta Creek, Palmerston N. P., N. Queensland, Australia, 800 m above sea level, collected 17 October 1972 by David S. Liem. Paratypes : Nine males, QM J22625-8, DSL 6502, 67 1 9 collected together with the holotype : QM J22629 30, DSL 6174, collected 29 November 1971 from Beatrice Creek, Palmerston N. P., N. Queensland. All specimens were collected by David S. Liem. Other Material : Besides the type series, two other specimens from Beatrice Creek, Palmerston N. P. were studied (DSL 4320-1). Diagnosis: A moderate sized robust treefrog, males ranging from 29-7-32-3 mm (mean 31-6 ± 2-975 mm) in snout-vent length. It could be distinguished from rheocolus and nannotis by the following combination of characters: 1 , snout rounded; 2, IN/EN ratio more than 0-980; 3, first finger enlarged, its disc small; 4, web on 1st toe on or just beyond the proximal subarticular tubercle; 5, inner tarsal fold absent or indistinct; 6, ventral surfaces of forearms, hindlimbs, and part of the pectoral region are washed with reddish pink ; 7, males have a rasping call. It is distinguished from rheocolus by the robust forearms, and large nuptial pad with coarse spinules; from nannotis by the smaller size (less than 40 mm in SV length), longer snout (snout longer than eye diameter), twc outer netacarpal tubercles present, moderate sized prepollex, absence of keratinous spinules on chest, throat, and sides of head, and the males call. 158 MEMOIRS OF THE QUEENSLAND MUSEUM Description of Holotype: SV length 32-3 mm; TL 17-5 mm, 0-542 of SV length; HW 12-0 mm, 0-372 of SV length ; EN 3-4 mm ; IN 3- 1 mm ; IN/EN ratio 0-912 ; IO 3-5 mm ; EW 3-0 mm; IO/EW ratio 1-167; TW 1-4 mm; ED 3-9 mm; TW/ED ratio 0-359. Tip of snout rounded; canthus rostralis rounded in cross-section and it is distinct (Fig. 2 A) ; loreal region slightly concave ; dorsal view of head rounded ; it is bluntly rounded in profile (Fig. 2C); tympanum small, partly hidden under the skin; supartympanic fold present ; tongue subovate, free posteriorly ; vomerine teeth in a lightly oblique short oval rows just behind the level of the choanae ; vocal sac absent ; upper jaw toothed. Forearm robust; a row of whitish tubercles is present on lateral edge of the forearm ; distal segment of fingers expanded into large disc, with ventro-marginal groove ; length of fingers from shortest to longest 1 -2-4-3; phalangeal formula of hand 2-2-3-3; first finger enlarged, but its digital disc is small (Fig. 2D). Fingers slightly webbed: web between 2nd and 3rd fingers reaches the proximal subarticular tubercles; web between 3rd and 4th fingers reaches halfway down the proximal phalanx of 3rd finger, and the 2nd subarticular tubercle of 4th finger by a narrow fringe. Nuptial pad large with coarse black keratinous spinules (Fig. 2B); subarticular tubercles roundish, one each on 1st and 2nd fingers, and one each on 1st and 2nd fingers, and two each on the two outer fingers; a row of super- numerary tubercles is present on metacarpal of 3rd and 4th fingers ; a pair of outer meta- carpal tubercles are present on the proximo-lateral portion of the palm, and an elongated tubercle on the proximo-ventral surface of the metacarpal of the 1st finger (Fig. 2D). Hindlimbs moderately long; distal segment of toes expanded into disc, with ventro- marginal groove; length of toes from shortest to longest 1-2-5-4-3; phalangeal formula of foot 2-2-3-3; one subarticular tubercle each of 1st and 2nd toes, two each of 3rd and 5th toes, and three on the 4th toe; supernumerary tubercles are present on ventral surfaces on metatarsal of toes ; inner metatarsal tubercle oval ; outer metatarsal tubercle indistinct or absent. Toes extensively webbed; web reaches halfway down the proximal phalanx of 1st toe by a narrow fringe; web reaches just below base of discs on outer margins of 2nd and 3rd toes, and it reaches by a narrow fringe the proximal subarticular tubercle of 2nd toe, and the 2nd subarticular tubercle of 3rd toe; web reaches the 3rd subarticular of 4th toe ; it reaches the base of the disc of 5th toe. Cloacal opening at mid-level of the thighs; skin of dorsum with conical tubercles, scattered on the head, upper eyelids and back; dorsal surfaces of fore- and hindlimbs smooth with some scattered tubercles on tibia and heels; ventral surfaces of abdomen, chest, throat, and posterior ventral portion of the thighs granular; skinfold across chest absent. Pupil horizontal, oval ; palpebral venation on lower eyelid absent. Colour: Dorsal colour of head and body slate; fore- and hindlimbs greyish brown, with indistinct dark markings ; ventral surfaces of body and thighs cream, reddish pink on ventral surfaces of fore- and hindlimbs. Palm and sole uniformly grey to slate colour without dark markings. In life, dorsum brownish olive or greyish brown with dark olive irregular markings on extremities ; eyes brown with black pupil ; ventral surfaces cream and reddish pink on fore- and hindlimbs as well as the pectoral region. THE LITORIA NANNOTIS SPECIES GROUP 159 Fig. 2: Litoria nyakalensis. A, dorsal view of head (QM J22624, holotype); B, dorsal surface of hand (QM J22626); C, lateral view of head (QM J22624, holotype); D, ventral surface of hand (QM J22624, holo- type); E, ventral surface of foot (QM J22625). Description and Variation: Six males from Henrietta Creek, Palmerston N.P. are 29-7-32*3 mm in SV length. Five males from Beatrice Creek Palmerston N.P., N. Queens- land ranging from 30*8-33*1 mm in SV length. Because there is no marked differences between these two samples, their ratios are combined: TL/SV ratio 0*520-0*560; HW/SV ratio 0*330-0-377; IN/EN ratio 0-978-1 029; IO/EW ratio 0 999-1-133; TW/ED ratio 0-326-1-029. There are slight variations in regard to the extend of finger webbing : in four specimens the web reaches only one-third down the length of the proximal phalanx on outer margin of 3rd finger, and not quite reaching the 2nd subarticular tubercle of 4th finger (QM J22629, J22625-6, DSL 6502). In QM J22625-7, J22629, DSL 4320-1, 6502, the web reaching the proximal subarticular tubercle of the 1st toe whereas in QM J22625-6, J22630, DSL 4320-1, 6502, the web barely reaches the base of disc on outer margins of 2nd and 160 MEMOIRS OF THE QUEENSLAND MUSEUM 3rd toes; in QM J22627 it reaches halfway down the penultimate phalanx of 3rd toe. in DSL 4320-1 the outer metatarsal tubercle is obscured, and in DSL 4320 the super- numerary tubercles of the foot area are absent or indistinct. In life, the dorsal ground colour of nyakalensis varies from uniform slate colour to greyish brown, some specimens with dark irregular markings. In QM J22625-6 and J22628, the dorsal ground colour of fore- and hindlimbs are greyish brown, and is lighter than the dorsal colour of the body. In most individuals the ventral colouration is uniformly cream with reddish pink on forearm, pectoral region, thighs and tibia; in QM J22628-30, DSL 6174, 6502, 6719, the throat, pectoral region, lateral region of abdomen, antero- and postero-ventral surfaces of the thighs are heavily pigmented with brown markings washed with reddish brown pigmentation. The eye is brown, and with black pupil. Mating Call: Regular repeated rasping single note call, easily distinguishable from the long drawn single note call of rheocolus. Details of the call structure will be presented elsewhere. Life History: Little is known about the life history of nyakalensis ; mating call of males were heard from October to March. Females of this species are yet to be discovered, probably similar to rheocolus females. Tadpoles unknown; judging from the substantial similarities amongst the Litoria nannotis species group, nyakalensis probably has similar sucker-mouth tadpoles like the other two species. Habitat: Litoria nyakalensis occurs along creeks in rainforest and wet sclerophyl forest ; it perches on rocks or on vegetation overhanging creeks. This species is synchrono- patric with Litoria rheocolus , eucnemis , lesueuri , and chloris , Nyctimystes tympanocryptis , Mixophyes schevili, Taudactylus acutirostris, Sphenophryne pluvialis, Cophixalus ornatus and exiguus. Distribution: At present L. nyakalensis is known from three localities, Henrietta Creek and Beatrice Creek in Palmerston N.P., and Tinaroo Creek Rd, 15 km south of Mareeba, N. Queensland. Etymology: The specific name refers to Nyakali, an aboriginal tribal group who once lived in the vicinity of the type locality. Discussion The new form of Litoria is intermediate between rheocolus and nannotis. In addition to characters listed in Table 1 , the new form differs from rheocolus in the following respects : slightly larger IN/EN ratio, rounded snout, rounded canthus rostralis, first finger of males enlarged, moderate sized nuptial pad with large coarse spinules, males with robust forearm, less extensive webbing on 1st toe, and ventral surface cream with reddish brown tinge on posterior portion of the body. It differs from nannotis by the smaller S V length, slightly larger 1U/EN ratio, tympanum always distinct, slightly concave loreal, smaller prepollex in males, less extensive webbing on first toe, two outer metacarpal tubercles, and ventral surface cream with reddish brown tinge of posterior portion of the body. Litoria nannotis does not call ; the rasping male call of the new form could readily be distinguished from the monotonous long drawn single note of rheocolos. THE LITORIA NAN NOT IS SPECIES GROUP 161 Lltoria nannotis (Andersson) (Figures 3, 4 A, B; Plate 3B) Hyla nannotis Andersson, 1916, p. 16; Tyler, 1965, p. 94. Hyla obsoleta : Loveridge, 1935, p. 50 (part). Hyla obsoleta nannotis : Copland, 1957, p. 55. Litoria nannotis : Tyler, 1971, pp. 351, 353. Material Examined: (31) Shiptons Flat, 50 km S. Cooktown, QM J 17861-3; Mt. Spurgeon, QM J5570; Mt. Lewis, 100 km N. of Cairns, DSL 5015-16; Tinaroo Creek Rd, 10 km SE Mareeba, DSL 4931 ; Atherton Tableland, QM J17017; Palmerston N.P., DSL 4261-2; Mt. Spec, 120 km N. of Townsville, DSL 4996-5001, 5003, 5005-10, 5012-14, 6198, 6306, 6343, 6559-60. All localities are in northern Queensland. Diagnosis: A moderately large sized treefrog, adult males ranging from 40- 1-47-9 mm, and adult females from 49-1-54-7 mm in SV length. It differs from rheocolus and nyakalensis by the following characters: 1, short snout (shorter than eye diameter); 2, strongly concave loreal; 3, extremely enlarged prepollex; 4, keratin spinules are present on chest, margins of throat, and sides of head; 5, males do not call; 6, adults more than 400 mm in S V length ; 7, only one outer metacarpal tubercle is present. It differs from nyakalensis by its smaller IN/EN ratio (less than 0-970), and it is further distinguished from rheocolus by its evenly rounded canthus rostralis. Description of Holotype: See Andersson (1916), and additional information in Tyler (1965). Description and Variation: Although Tyler (1965) gave an adequate account of variation, it is noteworthy to add more in the light of fresh material collected from new localities outside the range of published reports. Males range from 40*1— 47-9 mm in SV length; TL/SV 0-514-0-584; HW/SV 0-342- 0-397; IN/EN 0-896-0-960; IO/EW 1-043-1-275; TW/ED 0-303-0-315 or obscured. Females are larger: 49T-54-7 mm in SV length; TL/SV 0-510-0-570; HW/SV 0*351— 0-379; IN/EN 0-793-0-963; IO/EW 1-070-1-195; TW/ED 0-303-0-378 or obscured. Tip of snout bluntly rounded ; snout shorter than eye diameters ; loreal region strongly concave ; canthus rostralis evenly rounded in section ; tympanum barely visible or hidden under the skin; tympanum is distinct in 6 (DSL 6560, 5000, 5015-16, 5010, 4996) out of 26 individuals studied. Only one large outer metacarpal tubercle is present (Fig. 3D); inner metacarpal tubercle on base of thumb elongated; prepollex extremely enlarged (Figs. 3D and E); supernumerary tubercles are present on palm, on metacarpals, and on the phalanges ; a row of tubercles along lateral side of lower forearm present. Fingers moderately webbed; web between 2nd and 3rd fingers reaching their proximal subarticular tubercles; web between 3rd and 4th fingers reaching halfway down the proximal phalanx of 3rd finger, and between half to three-quarters down the length of the proximal phalanx of the 4th toe. Webbing of the foot reaching the base or barely reaching the base of disc of 1st, 2nd, 3rd, and 5th toes; it only reaches the 3rd subarticular tubercle (or just beyond it) of 4th toe (Fig. 3 A) ; web on inner margins of 3rd and 4th toes reaches the base of discs by a narrow fringe; metatarsals are separated by webbing. Supernumerary tubercles are arranged in rows on the metatarsals; ventral surfaces of tarsus heavily tubercular; outer metatarsal 162 MEMOIRS OF THE QUEENSLAND MUSEUM I ig. 3: Litoria nannotis. A, ventral surface of foot (DSL 5016); B, dorsal view of head (DSL 6960); C, lateral view of head (DSL 6960); D, ventral surface of hand (DSL 5016). E, dorsal surface of hand (DSL 5016). THE LITORIA NANNOTIS SPECIES GROUP 163 tubercle small and rounded. Tyler (1965) noted that in the holotype of nannotis , an internal vocal sac is present, but he failed to mention whether a vocal sac was present in other males. There is no trace of this vocal sac in any specimens examined. The report of the presence of a vocal sac in nannotis holotype is probably erroneous ; short openings observed were probably artifacts. Dorsal surface shagreen, with scattered tubercles on body, head and dorsal surfaces of the limbs ; large tubercles are present on upper eyelids, posterior portion of eyes, tym- panic region, on elbow and heel, on posterior and postero-ventral surfaces of the thighs. Ventral surfaces of the body (including the throat) are granular; only the anterior and antero-ventral surfaces of thighs, and inner surfaces of tibia and tarsus are smooth. Adult males are easily distinguished from females by their smaller size, robust forearm, extremely large prepollex, heavily spinulated nuptial pad on the proximo-dorsal side of the thumb, and black tipped spinules on inner surfaces of forearm, base of forearm, along ventral and ventro-posterior region of the eyes, loreal region, and scattered on ventral surfaces of the thighs and inner surfaces of the tarsus. Colour: In life, dorsal colour uniformly slate, or yellowish olive background with dark irregular markings (DSL 6343, 6559-60); in some individuals the extremities are washed with purplish colour (DSL 5010, 5005, 4997-8) ; dorsal colour of some individuals are pale slate (DSL 5000, 5007, 4999). Ventral surfaces of males cream; throat dusted with brown (DSL 6559, 6306, 5012). Ventral surfaces of females cream, and heavily pigmented with brown on the throat and ventral surfaces of the thighs; abdomen in four females (DSL 6343, 5000, 5014, 4999) is uniformly cream. In preservatives the purplish brown colouration of the throat, ventral surfaces of fore- and hindlimbs disappeared ; the majority of specimens are slate dorsally. Mating Call: No call has been associated to nannotis; males do not call. Life History: Pigmentless eggs (ovidiameter 2-7-3-4 mm) are laid in gelatinous eggmasses under rocks in water. Eggmasses laid by amplexus pairs were raised, and hatched tadpoles were substantially different than those described by Tyler (1965), and Martin and Watson (1971). Large tadpoles described by Tyler (1965) were undoubtedly those of Mixophyes schevili (will be reported elsewhere); W. Hosmer (pers. comm.) who collected tadpoles (AMNH 67115) upon which Tyler (1965) based his description agreed that they were not nannotis but the common Mixophyes schevili tadpoles. Tadpoles of nannotis are similar to rheocolus ; the former have sucker-mouth, very strong muscular tail, and narrow tail fins (Fig. 4 A) : spiracle opens at the end of a tube on ventro-lateral left side of the body ; anal opening median. The sucker-mouth consists of two complete papillar rows, one along the edge of the labium, and another close to the labial tooth rows; the posterior half of the latter consists of a single row whereas its anterior half consists of two papillar rows; in addition to the above rows two or more rows of papillae are present on the posterior of the labium between the inner and outer rows (Fig. 4B). The labial tooth row consists of two entire upper and three entire lower rows (formula II/III); the inner labial tooth rows are stronger than the outer ones. Jaws are strong, lower jaw is V-shaped, and the upper jaw is slightly indented anteriorly (Fig. 4B). In life, body ground colour sandy, with dark abdomen; tail with yellowish tinge and 164 MEMOIRS OF THE QUEENSLAND MUSEUM numerous dark-brown blotches on muscular tail. Like rheocolus tadpoles they are highly adapted to fast flowing water current. Tadpoles of nannotis differ from rheocolus by their darker tail colouration and their yellowish tinge. Habitat: This species occurs along rocky fast flowing rainforest or wet sclerophyl forest creeks ; L. nannotis is uually restricted near waterfalls. It is synchronopatric with Litoria rheocolus , nyakalensis, eucnemis, lesueuri, infrafrenata, and chloris , Nyctimystes hosmeri , tympanocryptis, Mixophyes schevili, Taudactylus acutirostris , rheophilus, Hylarana daemeli , Sphenophryne pluvialis, Cophixalus ornatus and exiguus. Remarks: Litoria nannotis is restricted in rocky stream habitats with fast flowing water current, behind waterfalls. Unlike other frogs, adults and juveniles form a colony amongst large boulders behind waterfalls. These aggregations were observed during the day as well as at night. On various occasions I observed that individuals perch close together (sometimes on top of each other) in a comer of a boulder. Loud background noise of Fig. 4: A, B, Litoria nannotis. A, lateral view of tadpole, B, ventral view of body. C, D, Litoria rheocolus. C, lateral view of tadpole. D, mouth parts. THE LITORIA NANNOTIS SPECIES GROUP 165 rushing waterfall probably resulted in the loss of the male call; consequently colony aggregations keep males and females close together for breeding or social contacts. The large number of females collected in one place on Mt. Spec, suggests that females do not wander far from males and from their microhabitats behind waterfalls ; large numbers of females collected at one spot is unusual in frogs. Distribution : Before the present study, nannotis was known only from the Atherton Tableland (Andersson, 1916; Loveridge, 1935; Copland, 1957; Tyler, 1965). My recent survey in northern Queensland reveals that it has a much wider distribution : it extends in the south to Mt. Spec (120 km NW of Townsville) and in the north to Mt. Finigan (50 km SE of Cooktown), N. Queensland. It is also found on Mt. Lewis, Mt. Spurgeon, Shiptons Flat, Tinaroo Creek Rd (15 km SW of Mareeba), Palmerston N.P., and Tully’s Falls, all in northern Queensland. Its presence along the northeastern rainforest chain of Cape York Peninsula has yet to be discovered. Phylogenetic Relationships within the Litoria nannotis Species Group Three species, rheocolus, nyakalensis and nannotis are placed in the Litoria nannotis species group. It is characterized by the following combination of characters: 1, moderate to large sized rheocolous treefrog; 2, tympanum small, less than 0*4 the width of eye diameter; 3, prepollex small to large; 4, moderate to very coarse nuptial pad spinules; 5, fingers moderately webbed — web on 3rd finger reaches base of penultimate phalanx or less ; 6, vomerine teeth in short or long rows, behind level of choanae; 7, eggs unpigmented, individually encapsulated; 8, eggs laid in a clump under rocks in water; 9, sucker-mouth tadpoles with median anal opening; 10, labial tooth row formula II/III. Satisfactory interpretation of phylogenetic relationships requires knowledge (or reasonable assumption) of primitive and derived states. This method had been used by Ozeti and Wake (1969), and Liem (1970, and MS). I adopted Marx and Rabb’s (1970) criteria for establishing primitive and derived states. Nine characters utilized for deducing phylogenetic relationships amongst species of the Litoria nannotis species group are presented in Table 2. The best phylogenetic tree obtained from these characters is presented in Figure 5 (construction of tree see detail in Liem, 1970 and MS). Even scanning through Table 2, it is obvious that nyakalensis shares more states with nannotis than with rheocolus, and this is shown by the best tree (Fig. 5). Based on the above evidence I consider rheocolus the most primitive, and the ancestral stock of Litoria nannotis species group; nannotis is the most specialized, whereas nyakalensis is intermediate between rheocolus and nannotis . Litoria rheocolus which is the most common and widely distributed species and lives in generalized creek habitats in contrast to the restricted waterfall microhabitat of nannotis or the restricted distribution of nyakalensis , lends support to the above hypothesis. Relationships between Litoria nannotis species group and New Guinean species is unclear. Its restricted distribution in northern Queensland rainforest and wet sclerophyl forest, and its ‘rheocolous’ habits, strongly suggests that Litoria nannotis species group is a recent arrival into northern Queensland from New Guinea. Probably its ancestral stock crossed the last Torres Strait land bridge in the Pleistocene together with Rana daemeli, Nyctimystes, and other microhylids (Liem and Hosmer, 1973). TABLE 1 Distribution of Character States of Hyla dayi Gunther, Litoria dayi (non Gunther), L. nyakalensis sp.nov., L. nannotis (Anderson), 166 MEMOIRS OF THE QUEENSLAND MUSEUM THE LITORIA NANNOTIS SPECIES GROUP 167 1 24 5 6 8 9 NANNOTIS Fig. 5: Suggested phylogenetic tree of Litoria nannotis species group. Numbers along nodes are number of derived states (see Table 2). TABLE 2 Character Distribution of Litoria nannotis Species Group. Character Character State* Speciesf Snout 0, longer than eye 1, 2 1, shorter than eye (1) 3 Tympanum 0, distinct 1, 2 1, indistinct or obscure (2) 3 Canthus rostralis 0, angular 1 1, rounded (3) 2, 3 Inner tarsal fold 0, absent/indistinct 2 1, present (4) 1,3 Outer metacarpal tubercle 0, two 1, 2 l,one(5) 3 Prepollex 0, moderate size 1, 2 1, very large (6) 3 Nuptial pad spinules 0, moderately fine 1 1 , coarse (7) 2, 3 Chest keratin spinules 0, absent 1, 2 1 , present (8) 3 Male call 0, present 1, 2 1, absent (9) 3 * Primitive state is coded 0 and the derived state 1. Numbers in brackets are derived states number and they are also used in the phylogenetic tree in Fig. 5. t 1, L. rheocolus ; 2, L. nyakalensis ; 3, L. nannotis. 168 MEMOIRS OF THE QUEENSLAND MUSEUM ACKNOWLEDGEMENTS For the privilege of studying specimens in their care I am indebted to Dr H. G. Cogger (Australian Museum), and Miss J. Covacevich (Queensland Museum). This work was supported by A.R.G.C. grants (1971-72), and Department of Zoology vote 1970-72. Dr M. C. Bleakly read the MS, and gave helpful suggestions and criticisms. LITERATURE CITED Andersson, L. G., 1916. Results of Dr E. Mjoberg’s Swedish Scientific Expeditions to Australia, 1910-1913. IX. Batrachians from Queensland. K. svenska Vetensk Akad. Hand/. 52 (9): 1-20. Copland, S. J., 1957. Australian tree frogs of the genus Hyla. Proc. Linn. Soc. N.S. W. 83: 9-108. Gunther, A., 1897. Description of new species of lizards and of a treefrog from North Queensland. Novit. zool. London 4 (3): 403-6. Liem, D. S., 1970. The morphology, systematics, and evolution of the Old World treefrogs (Rhacophoridae and Hyperoliidae). Fieldiana: Zoology 57: 1-145, 71 figs. Liem, D. S., in MS. Morphology, systematics, and evolution of Australo-papuan Leptodactylid and Hylid frogs (Anura : Leptodactylidae and Hylidae). Submitted for publication, Liem, D. S. and W. Hosmer, 1973. Frogs of the genus Taudactylus with description of two new species (Anura, Leptodactylidae). Mem. Qd Mus. 16 (3): 435-57. Loveridge, A., 1935. Australian amphibia in the Museum of Comparative Zoology, Cambridge, Massachusetts. Bull. Mus. comp. Zool. Harvard 78 (1): 1-62. Martin, A. A. and G. F. Watson, 1971. Life history as an aid to Generic Delimination in the family Hylidae. Copeia 1971 (1): 78-89. Ozeti, N. and D. B. Wake, 1969. The morphology and evolution of tongue and associated structure in sala- manders and newts (Family Salamandridae). Copeia 1969 (1): 91-123. Straughan, I. R., 1966. Queensland frogs. Species recognition. Unpublished Ph.D. Thesis, University of Queens- land. 241 pp. Tyler, M. J., 1965. Taxonomic studies of some hylid frogs of Australia and New Guinea. Proc. Zool. Soc. London, 145(1): 91-106. 1968. An additional Australian species of the hylid frog genus Nyctimysles with notes on Hyla dayi Gunther. Arkiv For Zoologi 20 (22): 501^4. 1971. The phylogenetic significance of vocal sac structure in Hylid frogs. Univ. Kans. Pubis. Mus. nat. Hist. 19(4): 321-60. Plate 3 A: Holotype of Litoria rheocolus (QM J22631). B : Litoria nannotis (DSL 5001). THE LITORIA N ANN OTIS SPECIES GROUP Plate 3 Plate 4 Holotype of Litoria nyakalensis (QM J22624). THE LITORIA NANNOTIS SPECIES GROUP Plate 4 Mem . QdMus. 17 ( 1 ): 169-74, pi. 5. [1974] A NEW SPECIES OF THE LITORIA BICOLOR SPECIES GROUP FROM SOUTHEAST QUEENSLAND, AUSTRALIA (ANURA :HYLIDAE) David S. Liem Department of Zoology, University of Queensland ABSTRACT A new species of treefrog of the Litoria bicolor species group is described from south-east Queensland. It is distinguished from other species of that group by the male mating call, broader head and snout, presence of dark brown spots and re- ticulations on dorsum, and a distinct brown purplish femoral streak bordering the dorsal green colour of the thighs. It is closely related to Litoria glauerti (Copland), and it occurs along coastal sandy areas. Early in 1971 I collected a series of small green treefrogs from a Freshwater Lake, Cooloola, and these appear to be an undescribed species closely related to Litoria glauerti. Subsequently similar frogs were collected from Fraser Island by Rick Shine, and from Lake Coolamera in Cooloola by Peter Ogilvie. Subsequent visits to Cooloola convinced me that these frogs represented an undescribed species. During the course of this study I have examined Litoria bicolor and L. glauerti as defined by Straughan (1969). Most of the specimens were collected by the author and are now deposited in the Queensland Museum (QM). Techniques and abbreviations of measurements are given elsewhere (Liem and Hosmer, 1973; Liem, MS). Head length (HL) is distance from tip of snout to angle of jaws. Mouth width of tadpoles is distance between most lateral portions of labial papillar row, and body length (BL) of tadpoles is distance between tip of snout and anal opening. The following ratios were computed, TL/SV, HW/SV, HW/HL, IN/EN, and TW/ED, but only HW/HL, IN/EN ratios and SV length were useful for distinguishing the three species of this species group. SYSTEMATICS Litoria cooloolensis sp. nov. (Figure 1 ; Plate 5) Holotype: Adult male, QM J22646 from Lake Coolamera, Cooloola, SE Queensland, collected on 27 September 1972 by David S. Liem. Paratypes: Eight adult males QM J22647-51 and DSL 6402, 6448, 6730, collected together with the holotype by David S. Liem. Other Material: Besides the holotype and paratypes, eleven other specimens were examined: Freshwater 170 MEMOIRS OF THE QUEENSLAND MUSEUM Lake, Cooloola (DSL 4531-39); and from Coomboo Lake, Fraser Island (DSL 5273-75). Fifteen tadpoles and eggmasses from Coolamera Lake were studied. Fifteen individuals of Litoria bicolor from North Queensland were studied: Atherton Tableland, QM J 12409 10; Ravenshoe, QM J10590-91; Mt. Molloy, QM J19501-B2; Dimbullah, QM J16996-98, QM J 1 70 1 2 ; Yorkey Knob, QM J16992; Mareeba, QM J 1 70 16; Palm Island, QM J5306, J5309; and from Rockhampton, QM J 15728. Sixteen specimens of Litoria glauerti from southeast Queensland were studied : Tin Can Bay, DSL 4542, 4544, 4549; Montville near Nambour, DSL 6403-04; Samford, near Brisbane, DSL 4271, 4273, 4277, 4279, 5173, 5176, 5230, 5249, 5251; Mt. Glorious, DSL 5223-24. Diagnosis : A small sized green treefrog’ less than 32 mm in SV length. It is distin- guished from other Australian treefrogs by the following combination of characters : 1 , small size; 2, absence of vomerine teeth; 3, dorsal background colour green with dark- brown spots and reticulations ; and 4, posterior surface of thigh bright orange separated from the dorsal green colour by a distinct brown purplish femoral streak. It is distinguished from glauerti and bicolor by the longer SV length, broader head (HW/HL ratio more than 1054), broader internarial width (IN/EN ratio more than 1-044), absence or indistinct maxillary cream streak, absence of brown loreal streak, and the peculiar diphasic male call. Description of Holotype: SV length 24- 0 mm; TL 12-7 mm, 0*529 of SV length; HW 7-7 mm, 0-321 of SV length ; HL 6-6 mm ; HW/HL ratio M 67 ; EN 2-3 mm ; IN 2-4 mm ; IN/EN ratio 1 044; ED 2-9 mm; TW 1-7 mm; TW/ED ratio 0-586. Tip of snout bluntly pointed (Fig.lA); snout slightly protruding beyond lower jaw; canthus rostralis evenly rounded ; loreal region straight, slightly sloping outward ; dorsal view of head bluntly pointed ; tympanum more or less distinct ; supratympanic fold weak or absent; tongue oval, free posteriorly; vomerine teeth absent; upper jaw toothed. Vocal sac with two slit-like openings. Distal segment of fingers expanded into disc ; disc of 3rd finger as large as tympanum ; length of fingers from shortest to longest 1 -2-4-3. Fingers with vestigeal webbing; web between 3rd and 4th fingers reaches their 2nd subarticular tubercles. Subarticular tubercle present, one each on 1st and 2nd fingers, two on 4th finger, and three on 3rd finger. There are three oval outer metacarpal tubercles, and an elongated inner one; supernumerary tubercles present on metacarpals and on the palm (Fig. IB). Hindlimbs moderately long; distal segment of toes expanded into disc with ventro- marginal groove; length of toes from shortest to longest 1-2-3-5-4; subarticular tubercles simple, one each on 1st and 2nd toes, two each on 3rd and 5th toes, and three on 4th toe; supernumerary tubercles on plantar region weak or indistinct. Inner metatarsal tubercle oval, outer one absent. Foot extensively webbed ; it reaches base of disc of inner margin of 5th toe, and outer margin of 2nd and 3rd toes; it reaches halfway down the proximal phalanx of 1st toe, and the 3rd subarticular tubercle of the 4th toe. Dorsal surface shagreen; abdomen and postero-ventral surfaces of thighs coarsely granular ; throat lightly tubercular. Pupil horizontal, oval. Adult males have finely spinulated nuptial pad on base of thumb. Colour: In life, dorsal colour green with dark-brown spots and reticulations ; a narrow barely visible dark-brown loreal streak runs from anterior angle of eye to nostril, just along the ventral margin of canthus rostralis ; a cream maxillary streak runs from below the eye to base of forearm ; dorsal green colour of femur is bordered ventrally by a brown purplish THE LITORIA BICOLOR SPECIES GROUP 171 femoral streak meeting along a sharp line; posterior surface of thigh orange, bordering the femoral brown purplish streak along a sharp line. Ventral surfaces of body and limbs cream. In preservative the dorsal green colour change into bluish green; orange patch on posterior surface of thigh disappears and the femoral brown purplish streak is retained. Description and Variation: Fifteen adult males: SV length 24-68 ± 1-480 mm (range 22-9-26*1 mm); TL length 10*9-13-6 mm, 0-452-0-535 of SV length; HW 7-0-8-5 Fig. 1. Litoria cooloolensis. A. Dorsal aspect of head (holotype, QM J22646), B. ventral aspect of hand (QM J22651), C. mouth of a tadpole, D. lateral aspect of a tadpole at stage 25, E. ventral aspect of foot QM J22651). Line equals to 1 mm. 172 MEMOIRS OF THE QUEENSLAND MUSEUM mm; HW/SV ratio 0-286-0-332; HL 6-6-7-8 mm; HW/HL ratio 1-039-1*167; EN 1 -6-2*5 mm; IN 2- 1-2-5 mm; IN/EN 1-044-1-313; ED 2-6-34 mm; TW 1-3-1 *7 mm; TW/ED 0457-0-586. Females are larger than males, 29-2-30 0 mm in SV length; TL 14-3-15-5 mm; TL/SV ratio 0-4974)-530; HW 9-0-9-2 mm; HW/SV ratio 0-300-0-315 ;HL 8 2-8-5 mm; HW/HL ratio 1-082-1-098; EN 2-5-3 0 mm; IN 2-9 mm; IN/EN 1-160-1 166; ED 34-3-6 mm; TW 2-0 mm; TW/ED 0-556-0-588. There are no marked differences in measurements amongst the males ; females are significantly larger than males. Some individuals show less extensive foot webbing: web along outer margin barely reaching the base of disc of 2nd toe in two individuals (QM J22647, J22649); one individual (QM J22649) lacks the cream maxillary streak. Individuals from Freshwater Lake, Cooloola are darker; dark- brown spots and reticulations on dorsum and the brown loreal streak are barely visible; they also lack or with very small cream maxillary streak ; foot webbing is less extensive in one individual (DSL 4534); in one individual (DSL 4538) the throat is extensively dusted with brown pigments. The brown purplish femoral streak is consistently present in all individuals studied and so is the orange patch on posterior surfaces of the thighs. Mating Call : A diphasic call, a short creak followed by a rattling noise at the end (call structure will be reported elsewhere). With little practice one can easily distinguish by ear, cooloolensis from glauerti and bicolor calls. Life History: Males call from August to May, but they may also call during warm winter nights. Amplexus pairs were observed during or after rain; amplexus axillary. Small jelly encapsulated eggs are deposited on submerged vegetation; eggs with light brown animal and cream vegetal hemispheres, similar to glauerti and most other Australian green treefrogs. TABLE 1 Distribution of Characters useful for Distinguishing Species of Litoria bicolor Species Group. Characters L. bicolor L. glauerti L. cooloolensis SV length (males) 22-10± 1-687 (N = ll) 23*54 ± 1*519 (N- 13) 24-68 ± 1-481 (N = 15) HW/HL ratio 0-846-0-912 0-855-0-971 1 054-1-167 IN/EN ratio 0-864-0-957 0-863-0-926 1 044-1-313 Web on 1 st toe halfway penultimate phalanx base of disc i to 2 penultimate phalanx Dorsal colour bronze and green uniform green green with brown spots and reticulations Brown loreal streak present present indistinct/absent Brown head streak present present absent Maxillary cream streak prominent prominent small/absent Tympanum colour brown brown green Upperarm colour pale yellow pale yellow/orange orange Colour of posterior of thighs pale yellow orange orange Brown purplish femoral streak absent indistinct/absent present THE LITORIA BICOLOR SPECIES GROUP 173 Tadpoles have a typical Australian hylid habitus, blunt snout and a pointed tail. Fifteen tadpoles at stage 25 have the following dimensions: ST length 25 0-29 0 mm, TL/ST 0-643-0-680, TH/TL ratio 0-444-0-500, M W/BL ratio 0-300-0-750. Tail is relatively high ; muscular tail narrow ; spiracle opens at the end of a tube on ventro-lateral left side of the body; anal opening dextral. Mouth moderately large, directed antero-ventrally; labial papillar rows consist of 1-2 layers, more at angle of jaws and is interrupted at its anterior portion (Fig. 1C). Labial tooth row formula 1,1/111, similar to some populations of glauerti tadpoles; jaw moderate in size and serrated. Unlike the spectacular coloured glauerti tadpoles (will be reported elsewhere), cooloolensis tadpoles are dull: tail fins transparent with dark brown reticulations ; head, body, and base of tail with cream (light in preservatives) and brown markings (Fig. ID). The dull colouration of the cooloolensis tadpole distinguishes it from the multicoloured tadpoles of glauerti. Habitat: Litoria cooloolensis is restricted in sandy fresh water lakes along coastal SE. Queensland and on Fraser Island. It occurs in wallum habitats and lowland rainforest. It is synchonopatric with Litoria peroni, olongburensis , Bufo marinus, Crinia signifera, Uperoleia laevigata , Pseudophryne bibroni, Platyplectron dumerili, and an undescribed Litoria . Distribution : At present it is known from only three localities : Freshwater Lake and Coolamera Lake in Cooloola, and in Coomboo Lake on Fraser Island, SE. Queensland. Etymology : The specific name refers to the locality where the holotype was collected. Status and Relationships Only two species, Litoria bicolor (Gray) and L. glauerti (Copland), amongst the Australian treefrogs recognized, show any potential resemblance to the new form. Species of the Litoria bicolor species group are characterized by small size (less than 32 mm in SV length), dorsal green background colour, absence of vomerine teeth, vestigeal finger webbing, and a characteristic diphasic mating call. The only other Australian treefrog with- out vomerine teeth is Litoria microbelos from northern Queensland; it differs from L. bicolor species group by its smaller size, less extensive foot webbing, and the brownish dorsal colour. Morphologically Straughan (1969) distinguished bicolor from glauerti by its narrower head width (HW/HL ratio less than 0-92), and more or less larger tympanum (TW/ED ratio 0-667-0-750). My sample however, did not show any significant differences in both ratios: HW/HL 0-846-0*912 and TW/ED 0*507-0*592 in bicolor as opposed to 0-855-0*971 and 0-500-0*636 for glauerti. In the present study L. glauerti sample came from SE. Queensland whilst Straughan (1969) used North Queensland sample and probably this attributed to the difference. Because bicolor and glauerti are sympatric in North Queensland the difference between these two species in North Queensland is probably due to character displacement. Significant characters for distinguishing bicolor , glauerti and the new form are listed on Table 1 . In SV length bicolor is not significantly different from glauerti at p < 0 05 (t = 2*118, DF = 3), however the new form differs significantly from either glauerti at p < 0-05 (t = 2-163, DF = 25) or from bicolor at p < 0*01 (t — 0*054, DF = 21); further- more the new form is distinguished from the two other species by a broader head (HW/HL ratio more than 1-054), and broader internarial distance (IN/EN ratio more than 1-044). 174 MEMOIRS OF THE QUEENSLAND MUSEUM Because the new form differs from bicolor and glauerti in SV length, HW/HL ratio, and IN/EN ratio, tadpole, colouration, and the male call structure, it merits specific recognition, and I propose to name it cooloolensis. Its restricted occurrence in coastal SE. Queensland and on Fraser Island, and the presence of derived characters (presence of brown spots and reticulations on dorsum and purplish brown femoral streak, absence or indistinct loreal and head streaks in cooloolensis ), suggests that cooloolensis is a derived species, probably it evolved from the widely distributed and generalized glauerti through isolation along SE Queensland coastal pocket during the Pleistocene similar to speciation of Crinia tinnula (Straughan and Main, 1966). Key to the Species of the Litoria bicolor Species Group 1. IN/EN and HW/HL ratios more than 1 000; dorsal colour green with darkbrown spots and reticulations; purplish brown femoral streak bordering the dorsal green colour of the thigh . . . . . . . . . . . . cooloolensis Not as above . . . . . . . . . . . . . . . . . . . . 2 2. Dorsal bicoloured, bronze and green ; web on 1 st toe halfway down the penultimate phalanx ; posterior of thighs pale yellow . . . . bicolor Dorsal colour uniformly green or brownish; web of 1st toe reaches base of disc; posterior of thighs orange . . . . . . . . . . . . glauerti ACKNOWLEDGEMENTS I wish to thank Miss Jeanette Covacevich of Queensland Museum for the privilege to study specimens under her care, and Rick Shine and Peter Ogilvie for the valuable gifts. This work was supported by ARGC grants (1971-72), and part by the Department of Zoology Vote 1970-72. Dr M. C. Bleakly read the manuscript. LITERATURE CITED Liem, D. S., 1974. A review of the Litoria nannotis species group, and a description of a new species of Litoria from north Queensland, Australia (Anura: Hylidae). Mem. Qd Mus. 17(1): 00-00. Liem, D. S. and W. Hosmer, 1973. Frogs of the genus Taudactylus with description of two new species (Anura, Leptodactylidae). Mem. Qd Mus. 16(3): 435-57. Straughan, I. R., 1969. The Hyla bicolor complex (Anura, Hylidae) in North Queensland. Proc. R. Soc. Qd, 80(5): 43-54. Straughan, I. R. and A. R. Main, 1966. Speciation and polymorphism in the genus Crinia Tschudi (Anura, Leptodactylidae). Proc. R. Soc. Qd, 78(2): 11 28. Plate 5 Litoria cooloolensis (holotype, QM J22646). THE LITORIA BICOLOR SPECIES GROUP Plate 5 Mus. Qd Mem. 17(1): 175-89. [1974] A REVIEW OF THE AUSTRALIAN CRABS OF FAMILY HIPPIDAE (CRUSTACEA, DECAPODA, ANOMURA) Janet Haig Allan Hancock Foundation, University of Southern California, Los Angeles, California, U.S.A. ABSTRACT Five species of Hippidae are recorded from Australian waters. Keys, diagnoses, and illustrations are given for their identification, and the Australian literature of the family is reviewed. Mastigochirus quadrilobatus, Hippa adactyla, H. pacifica , and H. celaeno are tropical species and widely distributed in the Indo-West Pacific; Hippa australis is a warm-temperate Australian endemic. The Hippidae are a small family of anomuran crustaceans of worldwide distribution, adapted for burrowing in sand. Because of this habit they are known in some areas as mole crabs or sand crabs, but in Australia no common name has been applied to the family as a whole. Hale (1927, p. 97) used the name ‘southern mole-crab’ for Hippa australis , and McNeill (1958, p. 491) referred to the tropical species collectively, under the name H. adactyla, as ‘turtle crab’. Hippa, the largest of the three genera comprising the Hippidae, is represented in the tropical Indo-West Pacific by at least ten species, most of which were the subject of an excellent revision by De Man (1896). The literature on the genus is in considerable con- fusion, however, because many identifications were based on an earlier revision by Miers (1878). Miers considered most of the nominal species of Hippa to be synonyms of a single, variable species, Remipes testudinarius {adactyla of some authors). Another source of confusion is the fact that the species-complex to which Miers applied the name testudinarius does not include the species to which that name properly belongs. For a discussion and partial solution of this problem, see Haig (1970). Records in the literature give the impression that three species of Hippidae occur in Australia : Mastigochirus quadrilobatus Miers, from northern Queensland ; Hippa australis Hale, confined to temperate waters ; and H. adactyla Fabricius, widespread in the Australian tropics. Griffin and Yaldwyn (1968, p. 171) included H. adactyla among examples of decapod Crustacea that are widely distributed in Australia. This picture, however, proves to be somewhat inaccurate and misleading. The present study, in which I have used De Man’s revision for the first time for the identification of Australian Hippa, has shown that there are at least three tropical species of that genus in Australia and that H, pacifica (Dana) is the common and widely distributed one, while H. adactyla appears to be relatively rare in the area. 176 MEMOIRS OF THE QUEENSLAND MUSEUM Measurements mentioned in the text refer to carapace length (cl.), and were taken in the midline with a dial caliper. In order to clarify the status of the Hippidae in Australia it was necessary to examine some material on which earlier records were based ; I also examined unrecorded specimens from several sources. Museums in which material is deposited are abbreviated in the text as follows: A. M. Australian Museum, Sydney B. M.N.H. British Museum (Natural History), London M. U.Q. Museum of the Department of Zoology, University of Queensland, Brisbane N. M.V. National Museum of Victoria, Melbourne Q.M. Queensland Museum, Brisbane S.A.M. South Australian Museum, Adelaide W.A.M. Western Australian Museum, Perth Z.M.H. Zoologisches Museum, Hamburg Family HIPPIDAE Diagnosis: Carapace oval, strongly convex, with lateral extensions which cover all but the first pereiopods. Telson of abdomen elongate, lanceolate. Mandibles reduced, non-functional as feeding organs. Third maxillipeds without an exopodite; merus broadened. First pereiopods non-chelate. Dactyl of pereiopods 2-4 curved and flattened. Key to the Genera of Hippidae 1 . Antennal flagella very long ; dactyl of first pereiopods oval and lamellate Emerita (Nine species, including three from the Indo-West Pacific ; not reported from Australia) Antennal flagella short ; dactyl of first pereiopods not o val and lamellate . . 2 2(1). Dactyl of first pereiopods greatly elongated and multiarticulate Mastigochirus Dactyl of first pereiopods styliform, not multiarticulate . . . . Hippa Mastigochirus Miers, 1 878 Mastigopus Stimpson, 1 858, p. 230. Type-species : Mastigopus gracilis Stimpson, 1 858, by original designation. Mastigochirus Miers, 1878, p. 321. New name for Mastigopus Stimpson, 1858, preoccupied by Mastigopus Leuckart, 1853; ipso facto the same type-species: Mastigopus gracilis Stimpson, 1858. Mastigochirus quadrilobatus Miers, 1878 (Figure 1) Mastigochirus quadrilobatus Miers, 1878, p. 322, pi. 5 fig. 8 (type-locality: Guimaras, Philippine Is.); 1884, p. 280 (Prince of Wales Channel). Henderson, 1888, p. 39 (off Booby I., Flinders Passage). Remipes adactylus: Grant and McCulloch, 1906, p. 33 (Port Curtis). Not Hippa adactyla Fabricius. [?] Emerita analoga: Stephenson, Endean, and Bennett, 1958, p. 269 (Low Is.). Not E. analoga (Stimpson). Material: Western Australia: Three specimens, Roebuck Bay, Mrs B. Grey, B.M.N.H, 1932,1 1.30.61-62. THE AUSTRALIAN HIPPIDAE 177 Female, Broome, sand bar opposite jetty, R. W. George on ‘Dorothea’, 16.x. 1962, W.A.M. 8-68. Queensland: Male, two females. Albatross Bay near Weipa, Gulf of Carpentaria, H. Foley, vi.1962 to iii.1963, A.M. P14138. Three females, mouth of Embly R., Weipa, 200 yds offshore, 15 ft on sand, high tide, E. Gamberg, 29.U962, W.A.M. 311-62. Three females, Port Curtis, presented ii. 1907 by Mrs F. E. Grant, A.M. G5759 (with label ‘Remipes testudinaria’). Female, 1 mile SE of Skirmish Pt, Moreton Bay, 2 fm on sand ripple, Dep. Zool. Univ. Qd, 14.viii.1967, Q.M. W2988. Diagnosis: Carapace covered with wavy, transverse lines. Frontal margin with four teeth; outer pair narrow, triangular, and projecting well beyond inner pair; latter close-set, rounded or triangular. No teeth on lateral margins. A row of about 20-24 shallow, seti- ferous pits near each lateral margin. First pair of legs longer than carapace plus dorsally visible portion of abdomen ; dactyl heavily clothed with long hairs, and composed of about 20 articles. Measurements: Male 9-1 mm, non-ovigerous females 6-2 to 11*6 mm, ovigerous females 9*5 to 12-8 mm. Review of Australian Literature: Mastigochirus quadrilobatus was collected off Queensland by the ‘Alert’ (Miers, 1884) and the ‘Challenger’ (Henderson, 1888). Grant and McCulloch (1906) reported three specimens of Remipes adactylus, dredged at Port Curtis in 7 fm. The collections of the Australian Museum contain no Hippa that match these data; however, the three female Mastigochirus from Port Curtis, A.M. G5759, are almost certainly the basis of Grant and McCulloch’s record. The record by Stephenson et al. (1958) of Emerita analoga is undoubtedly erroneous; E. analoga (Stimpson) is a western American species, and none of three Indo-West Pacific members of the genus has been found in Australian waters. The collections from the 1954 Low Isles survey are housed in the Department of Zoology, University of Queensland, but the material on which this record was based cannot be located (Dr W. Stephenson, pers. comm.). It was probably Mastigochirus quadrilobatus, the only Australian hippid with a superficial resemblance to Emerita. Remarks : This species has one congener, which also inhabits the Indo-West Pacific. Mastigochirus gracilis (Stimpson) is distinguished from M. quadrilobatus by the presence of three teeth instead of four on the frontal margin of the carapace, and several teeth on the lateral margins. Distribution: Waltair coast, India; Guimaras, Philippine Is.; and Australia from Roebuck Bay, W.A., around the northern coast and southward to Moreton Bay, Qd. To depth of about 8 fm. Hippa J. C. Fabricius, 1787 Hippa J. C. Fabricius, 1787, p. 329. Type-species: Hippa adacty la Fabricius, 1787, by subsequent designation ofRathbun, 1900, p, 301. Remipes Latreille, 1804, p. 126. Type-species: Remipes testudinarius Latreille, 1806, by subsequent mono- typy- 178 MEMOIRS OF THE QUEENSLAND MUSEUM Remarks : Most of the Indo-West Pacific members of this genus were treated by De Man (1896, 1898). Since it is highly probable that additional species of Hippa may yet be encountered in Australian waters, his revision should be consulted for the identification of any material that does not agree with the illustrations and diagnoses given in the present paper. c D Fig. 1: Mastigochirus quadrilobatus (Stimpson). A, animal in dorsal view; B, carapace, lateral view; C, left antenna; D, dactyl of right second pereiopod. Scales A, B = 5 mm; C = 0.5 mm; D — 1 mm. THE AUSTRALIAN HIPPIDAE 179 Key to Hippa known from Australia 1. Frontal margin of carapace with two (occasionally three) median and two lateral lobes or teeth ; each lateral margin of carapace with a submarginal row of shallow, setiferous pits . . . . . . . . . . . . . . . . 2 Frontal margin of carapace with a broad median lobe, lateral lobes scarcely developed ; no row of setiferous pits paralleling lateral margins H. australis 2 (1). All frontal lobes about equally projecting; concave margin of dactyl of legs 2 and 3 cut into an obtuse angle . . . . . . . . . . . . . . 3 Lateral lobes of frontal margin much more strongly projecting than median pair ; concave margin of dactyl of legs 2 and 3 cut into a right angle H. adactyla 3 (2). Submarginal row with about 3(M0 setiferous pits ; antennal flagellum normally with two articles . . . . . . . . . . . . . . H. pacifica Submarginal row with about 23-28 setiferous pits ; antennal flagellum normally with a single article . . . . . . . . . . . . H. celaeno Hippa adactyla J. C. Fabricius, 1787 (Figure 2 ; Plate 6) Hippa adactyla J. C. Fabricius, 1787, p. 329 (type-locality: ‘in Oceano australi’). Haig, 1970, p. 293 (CofTs Harbour; neotype designated). Remipes testudinarius Latreille, 1806, p. 45 (type-locality: ‘in Oceano Australasiae.); 1819, p. 141 (‘Nouvelle- Hollande’). Desmarest, 1823, p. 285 (‘Nouvelle-Hollande’); 1825, p. 175 (‘Nouvelle-Hollande’). Guerin, 1825, p. 281 (‘nouvelle Hollande’). H. Milne Edwards, 1837a, p. 114, footnote, pi. 42 figs. 1, la-j (‘Nouvelle-Hollande’); 1837b, p. 206, pi. 21 figs. 14-20 (‘Nouvelle-Hollande’). Haswell, 1882, p. 151 (no new records). De Man, 1896, pp. 461, 463; 1898, pi. 33 figs. 50, 50a^:. Remipes [ testudinarius ]: Latreille, 1817, p. 28 (type-locality restricted to ‘Nouvelle-Hollande’). Remipede tortue: Latreille, 1817, pi. 2 fig. 2. Remipes denticulatifrons White, 1847, p. 57 (nomen nudum; type-locality: Phillippine Ils.). Remipes testudinarius var. denticulatifrons Miers, 1878, p. 318 (in part), pi. 5 fig. 2. Hippa testudinarius var. denticulifrons : Thomassin, 1969, pp. 150, 161, 164, 172, figs. 6, 7a, 8a, pi. 5, [non] Remipes testudinarius : White, 1847, p. 57. Miers, 1878, p. 316, pi. 5 fig. 1. Ortmann, 1892, p. 537. [non] Remipes adactylus : Grant and McCulloch, 1906, p. 33. [non] Hippa adactyla : Stephenson, Endean. and Bennett, 1958, pp. 269, 274. McNeill, 1968, p. 39. Material: Large dry specimen lacking most of carapace, Dunk I., Qd, presented by E. J. Banfield and registered ii. 1920, A.M. P4660. Female, neotype, 25T mm cl., McCauley’s Beach, CofTs Harbour jetty, N.S.W., low tide, C. Bowden, A.M. P17384. Female, 26-5 mm cl., and soft-shelled juvenile, 7-5 mm cl., same data as neotype, A.M. PI 5825 and PI 5826. Male and female, Australia, Z.M.H. K5053. Soft-shelled specimen, Australia, Lidth de Jeude, 1866, N.M.V. Male, old collection, Q.M. W381. Diagnosis : Carapace densely covered with sharply serrate, transverse lines. Frontal margin five- toothed; outer pair narrow, triangular, and sharp-pointed, and in adults projecting well beyond inner ones; inner pair rounded; between them a small median denticle, broadly triangular and scarcely produced. A row of 50-55 shallow, setiferous, slightly elongate pits near each lateral margin, forming a narrow band. Antennal flagellum with 3-6 articles, the number increasing with age. Dactyl of second and third legs deeply falcate, distal and proximal portions of the concave margin meeting at a right angle. 180 MEMOIRS OF THE QUEENSLAND MUSEUM Review of Australian Literature : Remipes testudinarius was described by Latreille (1806) from material collected by Peron and Lesueur at an unspecified Australian locality. This record was subsequently repeated several times (Latreille, 1817, 1819; Desmarest, 1823, 1825; Guerin, 1825). H. Milne Edwards’ diagnoses and illustrations (1837a, 1837b) may have been based on other than type-material. Haswell (1882) recorded Remipes testudinarius in his catalogue of Australian Crustacea, but had no material and merely translated H. Milne Edwards’ (1937b) diagnosis. Haig (1970) selected as the neotype of Hippa adactyla a specimen of Remipes testudinarius from New South Wales, thereby reducing the latter name to synonymy. Remarks : This species is easily distinguished from all other members of the genus by the shape of the dactyl of the second and third pairs of legs. In the form of the frontal margin of the carapace it resembles Hippa admirabilis (Thallwitz), which is recorded from southern New Guinea and should be looked for in suitable habitats along the northern Australian coast. Distribution : From the west coast of Madagascar eastward through the East Indian Archipelago, north to Japan and east to the Marquesas Is. (for confirmed records see Haig, 1970, p. 294). In Australia the only precise localities on record are Dunk I. and Coffs Harbour on the east coast. Fig. 2: Hippa adactyla Fabricius. A, frontal part of carapace; B, carapace, lateral view; C, left antenna; D, dactyl of right second pereiopod. Scale A = 4 mm; B = 6 mm; C, D = 2 mm. THE AUSTRALIAN HIPPIDAE 181 Hippa pacifica (Dana, 1852) (Figure 3) [?] Remipes testudinarius : White, 1847, p. 57 (Australia). Ortmann, 1892, p. 537 (E. Australia). Not R. testudinarius Latreille = Hippa adactyla Fabricius. [?] Remipes marmoratus Jacquinot in Jacquinot and Lucas, 18[?], pi. 8 figs. 22-26; 1853, p. 97 (type-locality: Raffles Bay, Australia). Remipes pacificus Dana, 1852, p. 407 (type-localities: Fiji and Sandwich = Hawaiian Is.); 1855, pi. 25 figs. 7a-g. Miers, 1877, p. 74 (Australia, at least in part). De Man, 1896, pp. 462, 476; 1898, pi. 33 figs. 53, 53a-c. Remipes testudinarius : Miers, 1878, p. 316 (in part: Australia), [?] pi. 5 fig. 1. Not R . testudinarius Latreille — Hippa adactyla Fabricius. Hippa adactyla : McNeill, 1968, p. 39 (Murray I., Torres Str., in part). Not H. adactyla Fabricius. Hippa pacificus : Thomassin, 1969, pp. 157, 161, 164, 172, figs. 7c, 8c, 10, pi. 7. Material: Western Australia: Ten males, seven females, Pt Vlamingh, Rottnest I., in coarse sand at water’s edge, W.A.M. 24/9-36 and 30/44-36. Three specimens. Cape Vlamingh, Rottnest I., in coarse sand at edge of water, low tide, B.M.N.H. 1938.2.24.1-3. One specimen, Rottnest I., Hymnus, ii. 1920, W.A.M. 9612. Female, Pt Quobba near Carnarvon, P. Barrett Lennard, x.1959, W.A.M. 117-68. Female, Pt Cloates, W.A.M. 282-52. Six males, four females, near Yardie Creek Homestead, North West Cape, on beach, Alf Sneed, v. 1959, W.A.M. 67-62. Male and female, between Cape Dupuy and Cape Malouet, Barrow I., intertidal rock flat, W. A. Mus-U. S. Nat. Mus. Barrow I. Exped., 13. ix. 1966, W.A.M. 96-68. Female, Yampi Sound, in beach sand, G. H. Robinson, x.1959, W.A.M. 60-62. Queensland: Sixteen males, forty-seven females, Murray I., Torres Str., C. Hedley and A. R. McCulloch, viii-x. 1907, A.M. P2937, Male and female, Hope I. near Cooktown, A. R. McCulloch, viii. 1906, A.M. P17395. Fig. 3: Hippa pacifica (Dana). A, frontal part of carapace; B, carapace, lateral view; C, left antenna; D, dactyl of right second pereiopod. Scale A = 4 mm ; B = 6 mm ; C, D = 2 mm. 182 MEMOIRS OF THE QUEENSLAND MUSEUM Female, Michaelmas Cay off Cairns, presented xi. 1926 by C. Hedley, A.M. P8743. Male, Orpheus I., A. A. Cameron, 3.vii.l940, Q.M, W1 109. Female, Townsville, found in surf, A. Collins, 8.ix. 1923, Q.M. W109. Female, Wistari Reef near Heron I., Capricorn Grp, under the sand, A. A. Cameron, x.1939, A.M. PI 1164. Female, Caloundra, C. Y. Ross, 29.iv.1946, Q.M. W1677. Female, Redcliffe, F. Stott, 16.vi.1949, Q.M. W1706. New South Wales: Female, Corindi Beach, 30 miles S. of Grafton, intertidal sandy beach, J. Knight, 20.ii. 1964, A.M. PI 580 1 . Female, McCauley’s Beach, CofFs Harbour, low tide, presented by C. Bowden and registered 26.ix.1957, A.M. PI 2973. Lord Howe I.: Female, old collection, no data, A.M. PI 5806. Female, W. S. Thomson, collected before 1902, A.M. G4072. Female, presented xi. 1 923 by R. Baxter, A.M. P6764. Diagnosis: Carapace covered with wavy, transverse lines. Frontal margin with four rounded lobes, outer pair slightly more projecting than inner; lobes of inner pair separated by a shallow concavity; rarely a small median denticle. A row of about 30-40 shallow, setiferous, slightly elongate pits near each lateral margin. Antennal flagellum normally with two articles. Dactyl of second and third legs with a concave margin, distal and proxi- mal portions of this margin meeting at an obtuse angle. Measurements: Males 9 0 to 17-7 mm, non-ovigerous females 11*3 to 24-9 mm, ovigerous females 14-5 to 27-5 mm. Review of Australian Literature: White (1847) listed six Australian specimens of Remipes testudinarius in the collections of the British Museum, and Miers (1877) noted that there were specimens of Remipes pacificus in the collections of that institution. Miers’ Australian material, which would have included that listed by White, was probably based at least in part on Hippa pacifica ; but until these specimens can be re-examined their status must remain uncertain. Miers (1878) published a revision of the Hippidea in which he combined most of the nominal Indo-West Pacific species of Hippa , including H. pacifica, under the name Remipes testudinarius. The record from E. Australia by Ortmann (1892), who followed Miers’s broad interpretation of R. testudinarius , may have been based on Hippa pacifica. McNeill (1968) referred to notes made in 1907 by A. R. McCulloch on Hippa adactyla from Murray I., Torres Str. Eighty-one specimens collected by C. Hedley and A. R. McCulloch on that occasion are in the collections of the Australian Museum; 63 of them proved to belong to H. pacifica and the rest to H. celaeno. Jacquinot (in Jacquinot and Lucas, 1853) described a new species, Remipes marmoratus, with Raffles Bay as the type-locality. Miers (1876, p. 59) included it in his catalogue of the Crustacea of New Zealand, citing the Raffles Bay locality; but Thomson (1899, p. 169) pointed out that Raffles Bay is on the northern coast of Australia, and deleted R. marmora- tus from the New Zealand faunal list. De Man (1896) did not include Jacquinot’s species in his revision of Indo-West Pacific Remipes, and its status has remained unknown. During a visit several years ago to the Paris Museum I examined the type-material, consisting of four soft-shelled specimens; although I did not compare them critically with material of Hippa pacifica, I noted that they agree with that species in the number of setiferous pits near the lateral margin of the carapace and in having a two-segmented antennal flagellum. The illustration of R. mar- moratus was published before the desctription and probably earlier than 1852, since Dana THE AUSTRALIAN HIPPIDAE 183 (1852, p. 408) mentioned the species in connection with his original description of R. pacificus. Should careful comparison of the two species prove them to be synonymous, the unused name marmoratus might have to be suppressed to insure the stability of pacificus. Remarks: From the study of Remipes testudinarius (probably = Hippa pacifica, at least in part) from the Philippine Is., Estampador (1939, p. 355) concluded that these animals are either predators or scavengers. H. pacifica was later reported feeding on Physalia on Hawaiian beaches (Bonnet, 1946, as Emerita pacifica ). Matthews (1955, p. 382) also observed H. pacifica preying on Physalia in the Hawaiian Is., and discussed the mechanics of selective feeding in this species. No observations have been published on feeding of Hippa in Australian waters ; but the label with ten specimens of H. pacifica from North West Cape, W.A., listed above, indicates that the animals were “attracted by fish.’ Hippa pacifica is most closely related to H. ovalis (A. Milne-Ed wards), which differs chiefly in the form of the frontal lobes, and in having a three-segmented flagellum and about 45-55 setiferous pits near the lateral margins of the carapace (see De Man, 1896 and Thomassin, 1969). Distribution : Hippa pacifica is more widely distributed than are any of its congeners. It is apparently absent from the Red Sea, where all verified records refer to either H. picta (Heller) or H. celaeno (De Man) (Lewinsohn, 1969, pp. 172, 173). Most records from the E. and S. coasts of Africa under the names of testudinarius and adactyla probably refer to H. ovalis (A. Milne-Edwards), but I have seen material of H. pacifica from Tanzania. There are confirmed records throughout the tropical Indo-West Pacific as far east as the Hawaiian and Gambier Is. The species has also reached the west coast of the Americas where it occurs in the tropics from Mexico to at least as far south as Panama, and on various outlying islands (Clipperton, Revillagigedos, Cocos, Galapagos); Efford (1972) showed that all W. American records under the names adactyla, denticulatifrons, and testudinarius are based on material of H. pacifica. In Australian waters the species is now shown to range from Rottnest I., W.A., around the northern coast and southward to Coffs Harbour, N.S.W. Hippa celaeno (De Man, 1896) (Figure 4) Remipes celaeno De Man, 1896, pp. 462, 483 (type-localities: Makassar and Amboina); 1898, pi. 33 figs. 55, 55a-e. Hippa adactyla : Stephenson, Endean, and Bennett, 1958, pp. 269, 274 (Low Is.). McNeill, 1968, p. 39 (off Low Is.). McNeill, 1968, p. 39 (off Low Is.; Murray L, Torres Str., in part). Not H. adactyla Fabricius. Material: Queensland : Two males, sixteen females, Murray I., Torres Str., C. Hedley and A. R. McCulloch, viii-x. 1907, A.M. P7552 and P14455. Male and female, Hope I. near Cooktown, A. R. McCulloch, viii. 1906, A.M. PI 5863. Two females, off Low Is., Great Barrier Reef Exped. 1928-29, B.M.N.H. 1937.9.21.367-368. Dry specimen, Low Is., wet beach sand just above reef flat, 1954 Low Is. Survey, 14. viii. 1954, M.U.Q. Male, three females. Horseshoe Bay near Bowen, Dep. Zool. Univ. Qd, 20. v. 1966, W.A.M. 6-68. Female, Heron I., Capri- corn Grp, Dep. Zool. Univ. Sydney, viii. 1960, A.M. PI 5800. Diagnosis: Carapace covered with short, somewhat wavy, tranverse lines. Frontal margin with four rounded or rounded-triangular lobes, all about equally projecting, outer 184 MEMOIRS OF THE QUEENSLAND MUSEUM Fig. 4: Hippa celaeno (De Man). A, frontal part of carapace; B, carapace, lateral view; C, left antenna; D, dactyl of right second pereiopod. Scale A = 2 mm; B — 5 mm; C, D = 1 mm. pair somewhat broader than inner ones. Anterior part of lateral margins rather markedly and evenly concave. A row of about 23-28 shallow, setiferous, slightly elongate pits near each lateral margin ; this row sharply diverging from margin posteriorly. Antennal flagel- lum normally with a single article. Dactyl of second and third legs with a concave margin, distal and proximal portions of this margin meeting at an obtuse angle. Measurements: Males 6 0 to 9-1 mm, non-ovigerous females 7-8 to 16-0 mm, ovigerous females 110 to 16-9 mm. Review of Australian Literature : This species was not reported previously from Australian waters. The specimens collected at Low Is. during the Great Barrier Reef Expedition of 1928-29, and one from the 1954 Low Is. Survey, all of which had been reported as Hippa adactyla, were re-examined and found to be H. celaeno. As I have noted under the account of H. pacifica, McNeill’s (1968) record of H. adactyla from Murray I. was based on both H. pacifica and H. celaeno which were collected together at that locality by C. Hedley and A. R. McCulloch. Remarks: Hippa celaeno may be easily recognized by the abrupt concavity of the anterior portion of the lateral carapace margin, and by the strong divergence from that margin of the last few setiferous pits. Only one other species, H\ picta (Heller), has both a one-segmented antennal flagellum and fewer than 30 pits in the submarginal row. THE AUSTRALIAN HIPPIDAE 185 Distribution: Red Sea, East Indian Archipelago, Bismarck Archipelago, New Caledonia, and now reported from the east coast of Australia from Torres Str. to Capricorn Grp, Qd. Hippa australis Hale, 1 927 (Figure 5 A-F) Hippa australe Hale, 1927, p. 97, fig. 94 (type-locality: Comey Pt at mouth of Spencer Gulf, Australia); 1928, p. 97 (Cottesloe). Material: South Australia: Cast exoskeleton, holotype, Comey Pt, Yorke Peninsula, mouth of Spencer Gulf, Mrs Isobella Klem, 1.x. 1923, S.A.M. C994. Western Australia: Female, beach near Forest Grove, close to Black Rock, W.A.M. 540-31. Female, Bob’s Hollow, Calgardup, W.A.M. 151-40. Female, beach near Margaret R., 1954, W.A.M. 4-68. Female, Gnoocardup, W.A.M. 37-44. Female, Bunker Bay, in fine beach sand at tide level, R.W. George, S.iii. 1959, W.A.M. 68-62, Eleven males, four females, beach W. of Harvey, W.A.M, 4/18-36. Two females, beach near Harvey, W.A.M. 213/4-35. Three males, twenty-four females, Harvey, received from Snell, W.A.M. 3-68. Female, Cockburn Sd, D. McCorkill, iii. 1958, W.A.M. 64-62. Female, Naval Base, Cockburn Sd, W.A.M. 326-33. Female, Port Beach, Fremantle, G. Riley, 24.xi.1963, W.A.M. 2-68. Male, Leighton Beach, W.A.M. 33-45, Twenty-one males, twenty females, Cottesloe, W.A.M. 10174/93. Female, City Beach, W.A.M. 288-45. Female, Scarborough Beach, presented 27.xi.1963 by R. G. Mason, W.A.M. 5-68. Female, Triggs I., C. Tillbrook, iii. 1959, W.A.M. 65-62. Diagnosis: Carapace covered with shallow, setiferous pits, nearly absent anteriorly, elsewhere arranged in more or less even, transverse rows; anterolaterally these close-set, forming a series of oblique lines near lateral margins. Front with a low, broad, obtusely triangular rostrum; orbits shallowly concave, outer orbital lobes rounded, obscure. Longer flagellum of antennule with about 32-40 articles, shorter with 5-7 (usually 6) in males, 7-12 (usually 8) in females. Antennal flagellum with two articles, the proximal one elongate. Concave margin of dactyl of second and third legs curved, distal and proximal portions not meeting at a sharp angle. Measurements: Males 5-7 to 10*2 mm, non-ovigerous females 6-4 to 18-2 mm, ovigerous females 8-5 to 17T mm. Remarks: Hale (1928, p. 98) suggested that Hippa australis might be identical with H. truncatifrons (Miers), a species known from China and Japan. Several years ago, at my request, Dr Isabella Gordon compared specimens of the Australian species with the female holotype of H. truncatifrons (B.M.N.H. 1936.8.25.1) and pointed out several diffrences. Later I was able to verify these by examination of four Japanese specimens of H, truncatifrons sent by Dr Sadayoshi Miyake. Although closely related, the two species may be distinguished by the following characters: 1. In Hippa australis the low, triangular lobe is flanked on either side, to the inside of the eye, by a weakly developed, rounded lobe (Fig. 5 A). There is no trace of such lobes on the rostral area in H. truncatifrons (Fig. 5G). 2. In Hippa australis the longer flagellum of the antennule is elongate and more or less gradually tapering toward the tip (Fig. 5C), while in H. truncatifrons it is much shorter and less tapering (Fig. 5H). The shorter antennular flagellum may be consistently longer in H. australis than in H. truncatifrons, but the articles have been SBln 186 MEMOIRS OF THE QUEENSLAND MUSEUM counted in a very small sample of the latter species. In females (figured) the usual number of articles in H. australis is 8, in H. truncatifrons 6. In male H. australis there are usually 6 articles in the shorter flagellum ; there are only 4 in the single male H. truncatifrons that I have examined. Fig. 5: A-F. Hippa australis Hale. A, frontal part of carapace; B, carapace, lateral view; C, left antennule; D, left antenna; E, F, dactyl of right first and second pereiopods, respectively. G-I. Hippa truncatifrons (Miers) from Tosa Bay, Shikoku, Japan. G, frontal part of carapace; H, left antennule; I, dactyl of right first pereiopod. Scales A, E, G, I = 2 mm; B = 4 mm; C, D, F, H = 1 mm. THE AUSTRALIAN HIPPIDAE 187 3. The propodus and dactyl of the first pereiopods are more slender in Hippa australis (Fig. 5E) than in H. truncatifrons (Fig. 51). Hippa was placed on the Official List of Generic Names in Zoology and designated as feminine (International Commission on Zoological Nomenclature, 1963, p. 18). Therefore, the original spelling australe is changed to australis in this paper to agree in gender with Hippa. Distribution: The holotype of Hippa australis is a cast exoskeleton, which was picked up on the beach in South Australian waters. All subsequent records for the species are from temperate Western Australia. DISCUSSION Mastigochirus quadrilobatus, Hippa adactyla, H. pacifica, and H. celaeno are members of the tropical Indo-West Pacific fauna, and are widely distributed in the Indian and western Pacific oceans. The ranges of H. adactyla and H. pacifica extend to eastern Polynesia, and the latter species has crossed the Pacific faunal barrier to become established on the west coast of the Americas. With the exception of H. pacifica , the group of about eight species of Hippa to which the tropical Australian forms belong — all having two or three median frontal lobes or teeth, and a row of setiferous pits paralleling each lateral margin — has no representatives outside the Indo-West Pacific region. Mastigochirus , with two species in the Indian and western Pacific oceans, is an endemic Indo-West Pacific genus. The fifth species treated in this paper, Hippa australis , is an Australian endemic and is confined to warm-temperate waters in the south and southwest part of the continent. It belongs to a second group of Hippa species, characterized by the presence of a single median frontal lobe and the lack of a submarginal row of setiferous pits. Its closest relative is H. truncatifrons, a subtropical Indo-West Pacific species that is apparently restricted to southern Japan and adjacent parts of the Chinese mainland. The other three members of this group are tropical with one representative each in the eastern Pacific, central and western Atlantic, and eastern Atlantic. The division of the Australian hippid fauna into a southern temperate and a northern tropical component is consistent with the distribution of Australian marine decapods in general (Griffin and Yaldwyn, 1968, p. 1 68). Of the tropical species, Mastigochirus quadrilo- batus and Hippa pacifica occur in suitable sandy habitats on both west and east sides of the continent and to the north, while H. adactyla and H. celaeno appear to be restricted to the east side. However, the distribution of these latter species may prove to be much wider when more collecting has been done. Thomassin (1969, p. 172), in a study made on the west coast of Madagascar, found that the distributions of Hippidae in that area are influenced by wave action, salinity, and texture of the substrate. Hippa adactyla tolerates water of reduced salinity and lives in coarse sediments in the infralittoral zone; H. pacifica tends to be distributed according to age, with young postlarval stages (to 5 mm cl.) usually found in fine sands on the outer slope of the reefs, while adults live in coarser sediments, always in the region of breaking waves. The ecological requirements of the other Australian species have not been investi- gated. 188 MEMOIRS OF THE QUEENSLAND MUSEUM ACKNOWLEDGMENTS Without help from many sources, this study would never have been completed. My sincere thanks go to the following persons, who offered assistance when I examined material at their institutions, sent material for study at the Allan Hancock Foundation, or provided various kinds of information : D. J. G. Griffin and the late F. A. McNeill, Aus- tralian Museum; J. C. Yaldwyn, Australian Museum (now at National Museum, Welling- ton); A. Neboiss, National Museum of Victoria; Helene M. Laws, South Australian Museum; R. W. George, Western Australian Museum; B. M. Campbell, Queensland Museum; W. Stephenson, University of Queensland; Isabella Gordon and R. W. Ingle, British Museum (Natural History); J. Forest, Paris Museum; G. Hartmann, Hamburg Museum ; and S. Miyake, Kyushu University. I also wish to thank F. Munger, who took the photographs from which I prepared some of the illustrations ; J. S. Garth for reading the manuscript; and the administration of the Allan Hancock Foundation for providing working space. Travel and examination of material in Australia were made possible by Grant GB-6900 from the National Science Foundation. Work on this project was continued at the Allan Hancock Foundation with partial support from NSF Grant GB- 16386. LITERATURE CITED Bonnet, D. D., 1946. The Portuguese man-of-war as a food source for the sand crab ( Emerita pacified). Science, N. Y. 103: 148-9. Dana, J. D., 1852. Crustacea. ‘United States Exploring Expedition during the years 1838, 1839, 1840, 1841, 1842 under the command of Charles Wilkes, U.S.N.’ Vol. 13, Part 1, pp. i-viii, 1-685. (Philadelphia.) 1855. Crustacea, Atlas. Ibid., Vol. 14, pp. 1-27, pis. 1-96. Desmarest, A. G., 1823. Malacostraces. Diet. Sci. Nat. 28: 138-425. 1825. ‘Considerations generates sur la classe des crustaces, et description des especes de ces animaux, qui vivent dans la mer, sur les cotes, ou dans les eaux douces de la France. ’ pp. i-xix, 1-446, pis. 1-56. (Paris and Strasbourg). Efford, I. E., 1972. The distribution of the sand crabs, Hippa strigillata (Stimpson) and Hippa pacifica (Dana) in the eastern Pacific Ocean (Decapoda, Anomura). Crustaceana 23: 1 19-22. Estampador, E. P., 1939. Studies on the anatomy of Remipes testudinarius Latreille, with some observations on its relative morphological affinity. Philipp. J. Sci 69: 347-69, pis. 1-4. Fabricius, J. C., 1787. ‘Mantissa insectorum sistens eorum species nuper detectas adjectis characteribus genericis differentiis specificis, emendationibus, observationibus.’ Vol. 1, pp. i-xx, 1-348, (Copenhagen.) Grant, F. E., and McCulloch, A. R., 1906. On a collection of Crustacea from the Port Curtis district, Queens- land. Proc. Linn. Soc. N.S. W. 31: 2-53, pis. 1-4. Griffin, D. J. G., and Yaldwyn, J. C., 1968. The constitution, distribution and relationships of the Australian decapod Crustacea. A preliminary review. Proc. Linn. Soc. N.S.W. 93: 164-83. Gue'rin-Me'neville, F. E., 1825. Remipede: p. 281. In Latreille, P. A., ‘Encyclopedic methodique. Histoire naturelle. Entomologie, ou Histoire naturelle des crustaces, des arachnides et des insectes.’ Vol. 10. (Paris.) Haig, J., 1970. The status of Remipes testudinarius Latreille, and designation of a neotype for Hippa adactyla J. C. Fabricius (Decapoda, Hippidae). Crustaceana 19: 288-96. Hale, H. M., 1927. ‘The crustaceans of South Australia.’ Part 1, pp. 1-201. (Government Printer: Adelaide.) 1928. Some Australian decapod Crustacea. Rec. S. Aust. Mus. 4: 91 104. Haswell, W. A., 1882. ‘Catalogue of the Australian stalk- and sessile-eyed Crustacea.’ pp. i-xxiv, 1-326, pis. 1-4 (Australian Museum: Sydney.) Henderson, J. R., 1888, Report on the Anomura collected by H.M.S. Challenger during the years 1873-76. ‘Report on the scientific results of the voyage of H.M.S. Challenger during the years 1873 76 under the 189 command of Captain George S. Nares, R.N., F.R.S. and the late Captain Frank Tourle Thomson, R.N.’ Zoology, Vol. 27, Part 1, pp. i-xi, 1-221, pis. 1-21. International Commission on Zoological Nomenclature, 1963. Opinion 643: Idotea Fabricius, 1798, and Mesidotea Richardson, 1905 (Crustacea, Decapoda); validation under the Plenary Powers. Bull. Zool. Norn. 20: 18-25. Jacquinot, H., and Lucas, H., 1842-53. Crustaces: pis. 1-9. In Hombron, J. B., and Jacquinot, C. H. (eds.), ‘Voyage au Pole Sud et dans l’Oceanie sur les corvettes 1’ Astrolabe et la Zelee.’ Zoologie, Atlas. (Paris.) 1853. Crustaces: pp. 1-107. Ibid., Zoologie, Vol. 3. Latreille, P. A., 1804. Tableau methodique des crustaces. Nouv. Diet. Hist. Nat. 24: 123-7. 1806. ‘Genera crustaceorum et insectorum secundum ordinem naturalem in familias disposita, iconibus exemplisque plurimis explicata.’ Vol. 1, pp. i-xviii, 1 302, pis. 1-16. (Paris & Strasbourg.) 1817. Les crustaces, les arachnides et les insectes. In Cuvier, G., ‘Le regne animal distribue d’apres son organisation, pour servir de base a 1’histoire naturelle des animaux et d’introduction a 1’anatomie comparee.’ Vol. 3, pp. i-xxix, 1-653, and Vol. 4, pis. 12-14. (Paris.) 1819. Remipede, Nouv. Diet. Hist. Nat. (nouv. ed.) 29: 140-1, Lewinsohn, Ch., 1969. Die Anomuren des Roten Meeres (Crustacea Decapoda: Paguridea, Galatheidea, Hippidea). Zool. Verh. Leiden 104: 1-213, pis. 1, 2, McNeill, F. A., 1958. Hermit crabs and allies. Aust. Encyclopaedia 4: 489-92. 1968. Crustacea, Decapoda & Stomatopoda. Scient. Rep. Gt Barrier Reef Exped. 7(1): 1-98, pis. 1, 2. Man, J. G. de, 1896. Bericht iiber die von Herrn Sehiffscapitan Storm zu Atjeh, an den westlichen Kiisten von Malakka, Borneo und Celebes sowie in der Java-See gesammelten Decapoden und Stomatopoden. Vierter Theil. Zool. Jb., Syst. 9: 459-514. [Remipes: pp. 459-90.] 1898. Ibid., Sechster (Schluss-) Theil. Zool. Jb., Syst. 10: 677-708, pis. 28-38. [Remipes: pis. 33, 34.] Matthews, D. C., 1955. Feeding habits of the sand crab Hippa pacifica (Dana). Pacif. Sci. 9: 382-6. Miers, E. J., 1876. ‘Catalogue of the stalk- and sessile-eyed Crustacea of New Zealand.’ pp. i-xii, 1-136, pis. 1-3. (London.) 1877. Account of the zoological collection made during the visit of H.M.S. ‘Peterel’ to the Galapagos Islands. Crustacea. Proc. Zool. Soc. Lon. 1877: 73-75, pi. 12. 1878. Revision of the Hippidea. J. Linn. Soc. (Zool.) 14: 312-36. pi. 5. 1884. Crustacea: pp. 178-322, pis. 18-34. In ‘Report on the zoological collections made in the Indo-Pacific Ocean during the voyage of H.M.S. “Alert” 1881-2.’ (British Museum: London.) Milne Edwards, H,, 1837a. Les crustaces. In Cuvier, G,, ‘Le regne animal distribue d’apres son organisation, pour servir de base a 1’histoire naturelle des animaux, et d’introduction a 1’anatomie comparee.’ (Disciples ed.) pp. 1-278 and Atlas, pis. 1-80. (Paris.) 1837b. ‘Histoire naturelle des crustaces, comprenant I’anatomie, la physiologie et la classification de ces animaux.’ Vol. 2, pp. 1-531, and Atlas, pp. 1-32, pis. 1-42. (Paris.) Ortmann, A., 1892. Die Decapoden-Krebse des Strassburger Museums, mit besonderer Beriicksichtigung der von Herrn Dr. Doderlein bei Japan und bei den Liu-Kiu-Inseln gesammelten und z. Z. im Strassburger Museum aufbewahrten Formen. V. Theil. Die Abtheilungen Hippidea, Dromiidea und Oxystomata. Zool. Jb., Syst. 6: 532-88, pi. 26. Rathbun, M. J., 1900. The decapod crustaceans of West Africa. Proc. U. S. Natn. Mus. 22: 271-316. Stephenson, W., Endean, R., and Bennett, I., 1958. An ecological survey of the marine fauna of Low Isles, Queensland. Aust. J. Mar. Freshw. Res. 9: 261-318, pis. 1-11. Stimpson, W., 1858. Prodromus descriptionis animalium evertebratorum, quae in Expeditione ad Oceanum Pacificum Septentrionalem, a Republica Federata missa, Cadwaladaro Ringgold et Johanne Rodgers Ducibus, observavit et descripsit. Pars VII. Crustacea Anomoura. Proc. Acad. Nat. Sci. Philad. 10: 225-52. Thomassin, B., 1969. Identification, variability et ecologie des Hippidea (Crustacea, Anomura) de la region de Tulear, S.W. de Madagascar. Reel Trav. Stn Mar. Endoume (fasc. hors ser.) Suppl. 9: 135-77, pis. 1-7, Thomson, G. M. 1899. A revision of the Crustacea Anomura of New Zealand. Trans. Proc. N. Z. Inst. (n. ser.) 31: 169-97, pis. 20, 21. White, A., 1847. ‘List of the specimens of Crustacea in the collection of the British Museum.’ pp. i-viii, 1-143. (British Museum: London.) Plate 6 Hippa adactyla Fabricius, neotype. McCauley’s Beach, Coffs Harbour, N.S. W. (AM PI 7384). Scale in millimetres. THE AUSTRALIAN HIPPIDAE Plate 6 OBITUARY Henry Hacker 1876-1973 Henry Hacker, who had charge of the Queensland Museum’s entomological collection from 191 1 to 1943, was bom at Slip, Walthamstow, Essex, England on 31st January, 1876 and died at Brisbane on 21st September, 1973, He started his study of entomology at the British Museum and it is likely that he also acquired a grounding in bee-keeping in England. However, he soon embarked on a wandering, adventurous life. He came to Australia and gold fever seems to have caught him. He vistited Kalgoorlie and Coolgardie in Western Australia, and in November, 1898 took out a Miner’s Right at Charters Towers. About 1 897-8 he began collecting Australian insects. He served as a Trooper in the Boer War and took his discharge at Durban on 27 November 1901, giving his trade as miner and his intended place of residence as Johannes- burg. He was still in the Transvaal in June 1902 but later that year he made the first of three major collecting trips to parts of Australia which, previous to his visits, were entomo- logically unknown and of which he wrote interesting accounts. The first was a bicycle trip from Adelaide 300 miles by road to Port Augusta and then a further 258 miles northwest to Tarcoola (now on the Trans- Australia Railway) following camel pads, tracks about 2 feet wide. When riding he was so embedded in bundles that people used to wonder what it was coming along. For meat he depended on emus and rabbits which were present in thousands. His return journey was in winter and the beetles collected were left with Rev. T. Blackburn in Adelaide and the new species described by Blackburn, Sloane, Carter and Lea. ( Ent . Soc. Qd. Minutes, 27 March, 1935). In August 1904, Hacker took out a Miner’s Right at Cairns. He travelled by boat to Port Stewart and cycled to Coen which he made his base. The bicycle was of little use for riding but was invaluable for transporting equipment and stores. After a mishap he joined the frame together with hardwood pegs and fencing wire and used it for hundreds of miles more. He got two sawyers to fell a small patch of scrub for him, and he visited this every night with a hurricane lamp. It was during the wet season, and although he had many wettings he made a good collection from this patch. There were dangerous episodes with sunstroke and with a group of aborigines at a spot where a man had been speared. From his collections, Lea, Sloane and Carter described 51 new species of beetles. {Ent. Soc. Qd Minutes, 8th August, 1932, 27 March 1935). He was in Brisbane in 1907 but having business in Cloncurry took ship to Townsville and train to Charters Towers, whence he set out on 7 February to bicycle over 500 miles to Cloncurry, collecting beetles along the way. The weather was exceedingly hot and dry but four days beyond Richmond (where the railway ended) the rain fell in torrents. ‘It was impossible to ride or even push my bicycle through the wet black soil, so I shouldered it at sunrise and started to walk to the next stopping place, Fisher’s Creek, a distance of forty 192 MEMOIRS OF THE QUEENSLAND MUSEUM OBITUARY 193 miles. ’ With the help of a little riding he reached it at midnight and arrived at Cloncurry about midday on 20 February. Lea reported that Hacker sent him from this trip 128 species of beetles, probably more than half of them new to science, ( Tasm . Nat . 1 (2) : 12-13 1907). In 1909 Hacker married Constance Callaghan (d. 1970), sister of a school-teacher friend. They had a happy family life and reared three sons and two daughters. They settled in the Upper Mulgrave valley south of Cairns, but their rubber farm was unsuccessful and they came to Brisbane. Perhaps it was of this period that Hacker wrote: ‘Many years ago in N. Queensland I successfully worked large flowering blood-woods with the aid of a pair of climbing irons. Taking some lunch and a bottle of water I used to sit among the branches from early morning till about midday. The net was attached to a long bamboo so that all the branches could be reached. In catching large Buprestidae and Cetonidae, a tap on the underside of the branch with the rim of the net was sufficient to cause them to tumble in. Such fine species as Calodema plebejus and regalis , Metaxyorpha hauseri and gloriosa, besides many fine Stigmodera, large Cetonidae, and innumerable smaller insects were collected by this method.’ ( Ent . Soc. Qd. Minutes, 27 March 1935). He sold (probably in 1910) his private collection of about 6000 species of Australian Coleoptera to the National Museum, Berlin. He applied for a position on the staff of the Queensland Museum citing A.M. Lea, C. French, A. J. Turner, H. Tryon, W. Froggatt and T. Sloane as all being acquainted with his work, and on 7 March 191 1 he was appointed to the Museum staff. As an entomologist he undoubtedly had skill and enthusiasm, and nowhere is this skill and enthusiasm more evident than in the collections and the Memoirs of the Queensland Museum. He was most outstanding in his collecting ability and in his powers of observation. Many of his contributions to the Memoirs consisted of notes and observations on the collection, life histories, and habits of Australian insects. In these he often passed collecting hints on to other entomologists, for example, in Vol. 6, page 107 he writes ‘ to take note of all the dams and waterholes in the district to be collected over, and visit them in November or December. About that time they will be nearly dry and the mud around the edges visited by many fine wasps and bees. ’ His other papers in the Memoirs , one of which was written in conjunction with T. D. A. Cockerell, concern the taxonomy of various Hymenoptera and Homoptera. Throughout the Memoirs there are papers by such authors as Lea and Cockerell in which Hacker’s name constantly appears as the collector of many new species. Many other new species have been described by other authors and in other journals. During Hacker’s era not only were the collections increased but he encouraged specialists to identify his material. He would sort the material into conspecific groups and gave every group numbers. One or two specimens from each species group were forwarded to the specialist concerned who would return a list of identifications with the corresponding numbers. In such a system one would expect errors, but in fact, Hacker’s eye for a species is proving very accurate when the material is re-examined by today’s specialists. His collecting ability and knowledge of insect life histories was also evident in the museum displays he prepared for the instruction and enjoyment of the public. Both the collection and displays still have examples of his skill at preserving caterpillars by inflation 194 MEMOIRS OF THE QUEENSLAND MUSEUM and drying. Another of his skills was also seen in these displays, notably photography, an interest stimulated by Tom Marshall, a young Museum Assistant who often accom- panied Hacker on collecting trips to the Mt. Glorious area. The displays contained clear and sharp photo-micrographs of fleas, lice and mosquitoes which Hacker produced using equipment improvised from cardboard cylinders and rubber bands. Further evidence of his photographic skill can be seen in the plates accompanying his papers in the Memoirs. In 1929 Robert Veitch arranged for Hacker to be seconded to the Department of Agriculture and Stock to identify and build up its reference collection, while continuing to work one or two days a week at the Museum. Though eventually he became an officer of the Department, this arrangement continued until his retirement on 30 June 1943. When the serious bee disease, American foulbrood, was identified at Clear Mountain in 1931, none of the senior entomologists in the Department was experienced in beekeeping. Hacker was appointed an Inspector of Apiaries in February 1931 and was actively engaged on this for about a year. The beekeepers found him an unusual character to deal with, with his small goatee beard and rather dapper English-style suit, waistcoat and tie, which he wore even to the roughest spots. He was very knowledgeable about bees, particularly in relation to diseases and pests, and the youthful Charlie Roff found it very interesting and instructive to ac- company him. On other occasions when he was issuing certificates for export of queen bees, John Weddell went with him. Both found it a very exciting experience, for at this time Hacker travelled with a motorcycle and sidecar in which was the apiary equipment as well as the passenger. He would ‘go like hell’ irrespective of the kind of road, never slackening at comers and, coming back into town among the slow traffic, would point at a gap between two vehicles and somehow pass through. For the Department Hacker produced several extension articles on bees, the most important of these was the Department’s first extension Bulletin for beekeepers, ‘An introduction to beekeeping’. The basic information in it on the life history of the honey bee is as sound now as when it was written, though the equipment described is outdated. It was fortunate indeed for Queensland and for entomology that in days when jobs for entomologists were few Henry Hacker was able to obtain a post at the Queensland Museum where he could give all his time to entomology. His interest was extended to other orders besides Coleoptera by the need to build up the Museum’s collection, and it was further stimulated by visits from and correspondence with leading specialists from other states and overseas. He was one of the founders of The Entomological Society of Queensland in 1923, and at its meetings, by his papers, exhibits and informed comments on the exhibits of others, he was able to pass on much of his great knowledge to younger entomologists. By consent of the Society’s Council, parts of its tribute to him (Marks, E. N., 1973. Ent. Soc. Qd News Bull. 100: 13-16) have been reproduced here. Elizabeth N. Marks Queensland Institute of Medical Research, c /- Department of Entomology, University of Queensland. Edward Dahms Queensland Museum CONTENTS McKay, R. J. The Wolf Spiders of Australia (Araneae:Lycosidae):2. The Arenaris Group . , McKay, R. J. k The Wolf Spiders of Australia (Araneae:Lycosidae):3. A Coral Shingle In- habiting Species from Western Australia McKay, R. J. .... The Wolf Spiders of Australia (Araneae:Lycosidae):4. Three New Species from Mount Kosciusko, N.S.W. Archer, M. Apparent Association of Bone and Charcoal of Different Origin and Age in Cave Deposits Covacevich, J. The Status of Hyla irrorata De Vis 1884 (Anura:Hylidae) Wallace, C. A Numerical Study of a Small Group of Acropora Specimens (Scleractinia: Acroporidae) Archer, M. Some Aspects of Reproductive Behaviour and the Male Erectile Organs of Dasyurus geoffroii and D. hallucatus (Dasyuridae : Marsupialia) . . Wood, P. A. and King, Helen A Holocene Molluscan Fauna from Maroochydore, Queensland Stephenson, W., Williams, W. T., and Cook, S. D. The Benthic Fauna of Soft Bottoms, Southern Moreton Bay Richardson , Laurence R. Amicibdella and Micobdella gen. nov. of Eastern Australia (Hirudinoidea- Haemadipsidae) Liem, David S. A Review of the Litoria nannotis Species Group, and a Description of a New Species of Litoria from Northern Queensland, Australia (Anura:Hvlidae) Liem, David S. 3 3 A New Species of the Litoria bicolor Species Group from Southeast Queensland Australia (Anura:Hylidae) Haig, Janet A Review of the Australian Crabs of Family Hippidae (Crustacea, Decapoda Anomura) Obituary— Henry Hacker 1876-1973 . . Page 1 21 27 37 49 55 63 69 73 125 151 169 175 191