L H Moon &Son Bookbinders / / OF THE WESTERN AUSTRALIAN Museum and Art Gallery Edited by the Director, BEENAED H. WOODWAED, F.G.S., C.M.Z.S. I. I. CONTENTS. Introduction, by the Editor - PAGE - 7 The Mammoth Cave, by L. Glauert, F.G.S. I I Tachyglossus aculeatus - - 13 Phascolomys hacketti - - 15 Phascolarctus cinereus - - 29 Sthenurus occidentalis )? - ■ 31 Museum Notes ■ - 37 Printed by Order of the Committee The Paragon Printing and Publishing Co., Hay Street, Perth. igio. o >Ci JPrice, as. tJtl 7 fe ■ ji- Mi: r::e I V Pr £ ENTRANCE TO THE CAVE. Tins illustration has been kindly lent by Eixcau Robinson, Esy , tbe Superintendent of the Caves, RECORDS OF THE WEST E IIN AIT ST R Alii AN Museum and kRT Gallery EoiTfiD BY THE DIRECTOR, BEEKAED H. AYOODAYAEl), F.G.S., C.M.Z.S. ^OXjTJIVEE I. I. Printed by Order of the Committee — BY — The Paragon Printing and Publishing Co.. Hay Street, Perth. 1910. THE COMMITTEE OF THE WESTEEN AUSTEALIAN MUSEUM AND AET GALLEEY Nominated by the Government. past /llicmbers: The Hon. Sir James G. Lee Steere, K.C.M.G., M L. A., Chairman, 1895-1903. His Honour Sir Edward Stone, 1895-1899. Hon. Sir George Shenton, Kt., 1895-1909. J. C. H. James, Esq., B.A., 1895-1898. Charles Harper, Esq., 1895-1908. Hon. H. B. Lefroy, 1899-1900. His Honour the Chief Justice Sir A. C. Onslow, 1899-1900. Hon. Adam Jameson, M.D., M.L.C., 1902-1902. iPreeent /Iftembera : 1895 . Hon. J. W. Hackett, M.A., LL.D., M.L.C., Chairman (1904). H. F. Harvey, Esq., M.R.C.S., L.S.A. M. F. A. Canning, Esq. 1902 . H. T. Kelsall, Esq., M.D., M.R.C.S., etc. 1904 . His Honour the Acting Chief-Justice Mr. McMillan. Hon. Henry Briggs, M.L.C., President Legislative Council. 1909 . T. H. Bath, Esq., M.L.A. The Rt. Rev. Dr. Riley, Bishop of Perth. director : Bernard H. Woodward, F.G.S., C.M.Z.S. (Curator 1889- igoi). THE WESTERN AUSTRALIAN MUSEUM AND ART GALLERY, RISE AND PROGRESS. 1889.— The Geological Museum, Perth, established to display the specimens collected by Mr. H. Page Woodward, F.G.S., and by his predecessors in the office of Government Geologist, and to receive the contents of the Fremantle Geological Museum, which had been founded in i88i, by the Rev. C. G. Nicolay. 1892. The scope of the Institution was widened to include general Natural History and Ethnology. The contents of the Museum of the Swan River Mechanics’ Institute purchased. That Institution was founded by Capt. John Septimus Roe, R.N., in i 860 . The title abbreviated to the Perth Museum and the connection with the Geological Department terminated. 1895. The Mineral Gallery built ; the collections placed under the control of a Committee ; the Art Collections commenced. 1897. The James Street frontage erected. The present title given to denote that the Institution was National, not merely local. 1899. The Bird Gallery opened. 1906. The Mammalian Gallery installed. 1908. The Beaufort Street wing opened as an Art Gallery on the 25th June by H.E. the Governor, Sir F. G. D. Bedford, G.C.B. FREE POPULAR LECTURES ON SCIENCE AND ART. These Lectures, commenced in 1905, are delivered on alternate Friday evenings during the winter months and have proved most successful. The sixth series opens on April 8th next with a Lecture on “ Some forward impulses in the History of Agricultural Progress,” by Prof. W. Lowrie, to be followed by Lectures on Art and Science by the leading authorities in the State on such subjects. A programme will shortly be published. RECORDS OF THE WESTERN AUSTRALIAN MUSEUM and ART GALLERY. T he time has arrived when it has become necessary to issue Records, as is done by the older Museums. These will deal almost entirely with the Natural History of this State. They will contain the results of Original Research, of Collecting Expeditions, and will also include brief notes on other matters connected with this Institution. Hitherto this information has been published through the medium of various Australian, British and foreign publications. BERNARD H. WOODWARD, Director. 1st January, 1910. FOSSIL MARSOPIALS OF WESTERN AUSTRALIA. By BERNARD H. WOODWARD, F.G.S., C.M Z.S., Director of the W.A. Museum and Art Gallery. INTRODUCTION. I N 1882, the late Mr. E. T. Hardman, F.G.S., the Government Geologist, reported the occurrence of Diprotodon bones in the Leonard River, Kimberley. In 1895 Mr. Arthur found in a gully near Lake Darlot, a portion of a lower jaw of this animal, which he presented to this Museum. The specimen was, unfortunately, so much weathered that it was valueless except as a record. In 1898 the Museum Committee sent an expedition to Lake Darlot, and in 1908 the Hon. Dr. Hackett, at his own cost, sent another party to this place to search for further specimens, but neither succeeded in the quest. In February, 1909, Mr. John Sharp, of Balladonia, Point Malcolm, on the South Coast, sent to the Museum a number of bones and a few teeth of Diprotodon australis that he had unearthed when sinking for water near his station, and has promised to send any more that may be found. These bones were so common in that district some years ago, that the settlers considered them of too little value to be worth preserving, but now that they have been awakened to their scientific interest, they will save and forward to Perth all future discoveries. In 1904, Mr, E. A. Le Soeuf heard that some bones had been found in the Mammoth Cave, Margaret River, so he proceeded to the spot and secured a large number of fragments from a cutting that had been made when a pathway was being formed under the direction of the Superintendent of the Caves, Mr. Edgar Robinson. 10 ] RECORDS OF W.A. MUSEUM. Amongst these bones Mr. Le Soeuf found portions of two jaw bones of Sthenums, which he retained, hoping to determine the species. The other specimens he returned to the Caves Board who presented them to the Museum. Unfortunately they contained nothing of interest except the posterior half of the right jaw of Sthenimts, which, though of no value in itself as it only carried two molar teeth, yet proved to be of the greatest interest, for when the Caves Board were enabled in, 1908 to donate the jaws above- mentioned it was found to complete one of them. (See Plate V.). In August, 1905, the writer was invited by the Hon. Dr. Hackett, who is Chairman of the Caves Board as well as of the Museum, to meet him at the Margaret River Caves to select sites for further exploration. After a careful examination he recom- mended that in the first place it would be advisable to thoroughly examine and excavate under the mass of stalagmite that had been partially removed in making the pathway in the Mammoth Cave. Nothing further, however, was done until February and March, 1909, when the Committee was able to obtain the services of Mr. Ludwig Glauert, F.G.S , Medallist in Geology and Demonstrator in Geology for five years in the University of Sheffield, who was then on the temporary staff of the Mines Department as palaeonto- logist to the Geological Survey. The Hon. the Minister for Mines gave him leave of absence, without pay, in order that he might undertake this exploration under the direction of the writer. The Caves Board gave permission and the Superintendent, Mr. Edgar Robinson, rendered all assistance possible, and Mr. Glauert in less than two months collected 2,000 bones or fragments of bones. These include remains of Diprotodon australis, Nototherium, new species of Sthenums and Fhascolomys, of Fhascolarctus cinereus, of Echidna aculeata and several species of wallabies ; descriptions of the latter follow in Part H.of these Records. The wallabies include some species still living, e.g., M. brachyurus ; others may be new and in any case will certainly throw considerable light on the conditions which prevailed in prehistoric times on this Continent, as well as upon the relationships of the Marsupials. The Caves Board has generously presented all these fossils to the Committee of the W.A. Museum. The first instalment of Mr. Glauert’s report, printed below, will give these Records a special interest to zoologists and palaeonto- logists. PALEONTOLOGY, GLAUERT. [II THl^ MAMMO'J’H CAAn^. By LUDWIG GLAUERT, F.G S., etc. The Mammoth Cave, in which were obtained the various animal remains about to be described, is one of the finest of the numerous limestone caverns to be found in the extreme south-west of this State, from Cape Naturaliste to Cape Leeuvvin. They occur in a strip of Pleistocene forarainiferal limestone, the so-called “coastal limestone,” resting upon the metamorphic rocks that form the range of hills running parallel to the coast-line in that district, at a distance of about four miles from the sea. At one time the gneissic slope extended uninterruptedly from the hills to the shore, but the strong south-west winds of the summer months soon gave rise to a chain of sand dunes which travelled inland and up the gentle escarpment of the hills. The material carried along by the wind consisted not only of the quartz grains which were derived from the disintegration of these rocks, but also included small shell fragments and the minute calcareous tests of foraminifera. Rain water, by virtue of the small amount of carbonic acid which it obtains from the atmosphere, is able to dissolve this lime and carry it in solution to the lower portions of the pervious beds. Owing to evaporation, this lime is re-deposited and in due course the loose incoherent sand becomes solidified, forming either a calcareous sandstone or a sandy limestone, according to the amount of lime present in the rock. The formation of caves, therefore, was accomplished without difficulty, and hastened by the fact that the rainfall on the western slope of the range is considerable, and drains through the limestone belt to the coast. The numerous winter streams that rise in the hills proceed uninterruptedly till the limestone belt is reached, when they mostly disappear, either in the mouths of caves, or in “ swallow-holes.” The stream that flows through the Mammoth Cave is one of the former ; it continues its course underground till it issues at the foot of the sea cliffs three miles to the westward. These subterranean watercourses are usually a succession of clefts and caverns, 12 ] RECORDS OF V/.A. MUSEUM. increasing considerably in width and height when conditions are favourable. As a general rule the upper contour of the underlying granite forms the bed of the stream, but a very hard band of the limestone will answer the same purpose, though not in so satisfactory a manner. We have an example of this in the Mammoth Cave, for the blocks of limestone in the large main chamber (the size of which suggested its name) afford distinct evidence that formerly there were two caves, one above the other. In course of time the percolating water so weakened the limestone that formed the roof of one cavern and the floor of the other, that it collapsed, uniting the two into one huge chamber, the floor of which is strewn with large masses most of them hundreds of tons in weight. It was on the upper surface of one of these blocks, quite 70 ft. above the stream which flows through the cave, that the numerous specimens were collected. To an observer standing at that spot it is evident that the remains cannot have been introduced by way of the present entrance, which is about 70 ft. lower, for many of them bear unmistakable evidence of water transport. It was therefore necessary to seek another explanation for the presence of the bones at such a height above the present cave floor, and as the features observed tend to show that the double chamber evidently had existed, that solution may safely be adopted. The material in which the bones were embedded, varied somewhat in nature ; it comprised two groups, the lower and older series consisted of a reddish, fairly coarse sand, containing fragments of wood and gastropod shells in addition to the bones, with occa- sional bands of black loamy soil about an inch in thickness, which seem to prove the existence of floods in days gone by. Layers of stalactite often enclosing bones, wood fragments, etc., and bearing casts of eucalyptus leaves were not uncommon, one of these layers completely covered the series, thus protecting the animal remains and accounting for their fine state of preservation. Above this there was another sandy bed which was yellowish in colour ; the bones it contained were much fresher in appearance and are undoubtedly much more recent than the remains preserved in the lower series. A layer of stalactite varying greatly in thickness covered the lot, and it was not until this layer was pierced that the presence of PALAEONTOLOGY. GLAUERT. [13 fossils was made known. This discovery suggested the advisability of further investigations, which having been undertaken under the auspices of the Western Australian Museum and the Caves Board, resulted in the collection of some thousand bones and the addition of several species to the known fauna of the State. Order MONOTREMATA. Fam. t. Echidnid,®. Tachyglossus (Illiger). Tachyglossns (Echidna) aculeatus (Shaw sp.) vav. typicus (Tho). Myrmecophagu aculeata, Shaw Ornithorhynchus hyslrix, Home . . Echidna hystrix, E. Geoff. ,, longiaculeaia.Ti-sdm. Tachyglossus anileatits, 111 . Echinopus hystrix, G. Fisch. Echidna aculeata, Garnot ,, aiistrallensis, Lesson Tachyglossus hystrix, Kaiip. Echidna australis. Lesson ,, acanthion, CoUett ,, aculeata oar. typica, Thomas > f i i t > II Nat. Misc. iii., pi, cix., 1792 Phil. Trans 1S02, p. 34S, pis. 10-12 Cat. Mus. p 224, 1803 Zool, i., p. 592, 1808 Prodr. Syst. Mamra p. 114, 1811 Zoogn. iii., p. 692, 1814 N. Bull, Soc. Philora., p. 45, 1825 Man. Mamm., p. 318, 1S27 Thierr. i., p. 255, 1835 Compl, a Buffon V,, pi 52, 1836 Forh. Vid. Selsk., 1884, No. 13, 1885 P.Z.S. 1885, p. 338, pi. 23, 1886 Cat. Marsup. Brit. Mus., p. 379. THE NATIVE PORCUPINE or HEDGE-HOG. One Monotreme is included in the list of specimens obttiined, but the only bone yet recognised is the right humerus of a Tachyglossus from the older deposit. It is almost perfect, so that its comparison with other specimens is not difficult. The collection of Monotremes in the Mammalian Gallery contains several Echidnee which are all somewhat smaller than the animal which yielded the bone now under review. The skeleton of E. aculeata is perfect, and gives ample opportunity for studying the typical bones. It is seen at once that the Caves’ specimen belongs to a larger animal, and that there are a few slight differences which are evidently only individual, as Owen’s figure of the humerus of the same species, then called E. hystrix (pi. 14 in Phil. Trans., 1884, part I, published in 1885) illustrates a bone which is identical with this, though slightly less in its dimensions. RECORDS OF IV. A. MUSEUM. 14J Tachyglossns [Echidna), aculeatus, var. typiciis, measurements : — Owen’s Mammoth Cave E. Hystrix. Specimen. Inches. Lines. Inches. Lines. Length . . . . . . 2 — 2 3 Breadth (proximal end) . . i — i 3 ,, (middle of shaft) .. — 4 — 5 ,, (distal end) . . i 9 2 — Thickness of middle of shaft — 3.J — It will be seen from the above table that there is an all round increase in size over the specimen that supplied Professor Owen’s measurements ; such an increase as would easily be accounted for by the greater size of an individual of the same species. A most careful examination of this specimen and a comparison with Owen’s three figures disclose two points of difference, the greater size, in regard to the whole bone, of the process known as the “ tricipital ” which forms the termination of the “ teretial ” or “ posterior tricipital ridge ” that runs along the radial border of the humerus from the ento-tuberosity, and secondly the greater comparative depth of the “ ulnar trochlea ” at the distal end of the bone. Both these features are of little importance as they would most likely become accentuated as the animal increased in age and size. It is an interesting fact that the days which saw the Dipvotodon and Nototheniim in the south-west of this State, also saw this ant-eater’s range much more extended than at the present time. RECORDS OF IV. A. MUSEUM. [15 Order MAKSUPIALIA. Sdb-Order DIPROTODONTIA. Fam. iii., Phascolomyid Purjun The incisor is narrow, sharp-edged, and deep antero-posteriorly,. the premolar, a little less in length than Mi, has a sharp edge that branches posteriorly into two diverging ridges. The four molars vary very little in size, the first one being the smallest; they have four tubercles or cusps in the shape of sharp curved crests which have the concave sides interiorly and the convex sides exteriorly.. All these characters are associated with herbivorous animals. RECORDS OF W.A. MUSEUM. 30] A careful search through the specimens in the Museum and the figures and descriptions to be found in the works of reference available, showed that without a doubt the two mandibles are those of a Koala, or Native Bear {Phascolarctus). See Owen's “ Extinct Mammals of Australia,” fig. ii, p. 90; fig. 5, p. 152; fig. 6, p. 153; fig- 3 i pl- XXXII. Also British Museum Catalogue of Marsupialia and Monotremata, pp. 209, 210, 212. The following measurements were taken in mm Antero-posterior length . . . : — Aged 100 Young 80 Height (coronoid) . . . . . . 70 49 Thickness behind M 4 . . 9-5 — Width, exterior of condyle to exterior of condyle 67 — Greatest width (coronoids) . . 74 53-5 Length of incisors II? II Thickness of do, . . . . . . 3-5 3-5 Depth at socket . . 7 6 Length of cheek series . . 35 38 ? Length of premolar . . 6.5 7 Greatest thickness of premolar ... . . 4-3 4-5 Length of Mi -M 3 . . 21-5 23 Length of M 1 -M 4 . . 28.5 31 ? Length of Mi . . . . . . . . 7 7'5 Thickness of Mi . . 5-5 5-5 Length of M 2 . . . . 7-25 7-5 Thickness of Mz . . . . . . 5-5 5-5 Length of M 3 . . . . 7-25 7-5 Thickness of M 3 . . 5 5 Length of M 4 . . .. 7-25 7-5 Thickness of M 4 . k - 5 5 Depth of jaw under P 4 .. .. 18.5 14 Depth of jaw under M 4 22 19-5 Distance of condyle above the alveolus of M 4 32 21 Length of symphysis .. 24 ? RECORDS OF W.A. MUSEUM. Car: Fam, Macropodid^e. Sub-Fam. Macropodin^. Sthemmis, Owen (1873)1 . Sihenurus — Owen .. Proc, Royal Society XXI., No. 141, p. 128, 1873 Owen .. Phil. Trans. Royal Society, p. 264, 1874 Lydekker .. Catalogue Fossil Mammals, Brit. Mus. , N.H. vol. V., p. 231, 1887 De Vis .. Proc. Linn. Society N.S. Wales, 2nd series, vol. X., p. 88, 1895 Protemnodon {partim) — Owen .. Proc. Royal Society XXI., No. 141, p. 128, 1873 Owen . . Phil. Trans. Roy, Soc., p. 274, 1874 Procoptodon — Owen .. Proc. Roy. Soc. XXI.. No. 145, p. 387, 1873 Owen .. Phil. Trans. Royal Society, p. 788, 1874 hydekker Catalogue Fossil Mammals Brit. Mus., N.H. vol. V., p. 231, 1887 This form being extinct, it has no vernacular name ; it belongs- to the kangaroo family. STHENURUS OCCIDENTALIS, Sp. Nov. Plate V. [Report originally prepared for the Caves Board on the specimens originally found by Mr. E, A. Le Souef, It was not printed, and has since been revised and extended.] The specimen about to be described consists of the right and' left horizontal rami, united in their natural position, with incisors,, premolars and molars all in place, and up in line. The hind lobe of the last molar, M4, showing traces of wear, the animal would be considered aged by Oldfield Thomas,^ and adult by C. W. De Vis.® The lower half of the right coronoid is present,, but of the left very little remains. It has been possible, however, to make the following, measurements : — Incisor. Length from base of enamel to the extremity of th& worn crown, 22mm. ; length of the cutting edge of crown, 13mm. ; thickness of crown, 6.5mm. ; vertical diameter at base of enamel,. 1 1 mm. ; transverse diameter, 7mm. Diastema. Length from posterior base of enamel of incisor to anterior edge of socket of premolar, 22.5mm. 1 Owen, Proc. Royal Soc. Lond. XXI., No. 141, p. 128, 1873. Owen, Phil. Trans. Royal Soc. Lond., p. 265, 1874. 5 O. Thomas, Catalogue of Marsupialia and Monotremata in the Collection of the British Museum (N.H.), p. 7. 1888. » C. W. De Vis, Proc. Linn. Soc. N.S. Wales (and ser.), vol. X., p. 79, 1895. 32 ] RECORDS OF W.A. MUSEUM. Cheek Teeth. Length of entire series (in situ) from anterior edge of P4 to hind edge of M4 60 to 6imm., Mi to M4 44.5mm., Mi to M3 32.5mm. Premolar (P4), antero-posterior dimension 17mm., summit of crown 12.5mm., greatest width of anterior moiety 7mm., of posterior gmm. First Molar (Mi)i, length 10.5mm., fore lobe 8.5mm., hind lobe 8.5mm. Second Molar (M2), length 11.5mm., fore lobe 8.5mm., hind lobe 9.5mm. Third Molar (M3), length 12.5mm., fore lobe lomm., hind lobe lomm. Fourth Molar (M4), length 11.5mm., fore lobe 9.5mm., hind lobe 9mm. The Mandihle, greatest length of jaw i6omm. (from the tip of the incisor to the hind margin of the coronoid) ; greatest depth in front of P4, 34mm. ; greatest depth at M4, 35mm. ; depth behind P4, 31mm. Thickness under P4, 15mm.; behind M3, i7-i8mm. Dental vascular foramen to 8mm. below the edge of diastema, slightly in advance of P4. A second smaller foramen below the base of posterior lobe of Mi. Description of the Specimen. The incisor much resembles the corresponding tooth of the S. atlas figured by Owen (Phil. Trans. 1874, pi. xxii., figs. 5 and 6), but has a relatively longer working surface, possibly on account of the greater age of the individual, and is more erect. The premolar is elongately oval in horizontal section, possess- ing a marked constriction at a point 7mm. from the anterior basal edges, particularly on the outer aspect, which almost divides the tooth into a fore and a hind lobe ; the oblique horizontal continua- tion of this vertical groove, inwards and backwards, gives a transversely bilobed appearance to the hinder part of the crown, and causes the lobe that represents the anterior portion of the outer aspect to form the whole of the inner surface. The posterior 1 This tooth corresponds to the D4 of Prof. Owen. The Mi of this authority agrees with the M2 of later authorities. In this table of measurements the width of the lobes of the molars is the extent of the cutting edges of the lobes. RECORDS OF IV. A. MUSEUM. [33 portion of the crown is shorter than the succeeding molar, and has a broad working surface with complex transverse ridging between the inner and outer simple trenchant edges. A basal ridge is present on the inner, fore, and outer aspects ; that on the front of the tooth is plain, but the other two each give rise to three swellings or ribs that pass upwards to the crown. A vertical fold in the enamel is present at the intero-posterior angle ; the hind surface of the tooth is perfectly smooth and convex, both vertically and horizontally. The Molars (M1-M4). The lobes are thin and almost straight, though slightly convex backwards, crests slightly concave, angles sharp, particularly on the inner edge ; from the outer angle of the hind lobe, a marginal fold sweeps downwards and forwards, almost closing the mid- valley ; this fold from the fore lobe forms the edge of the anterior talon. Similar branching marginal folds are present on the anterior surface of the inner angle of the hind lobe. There are several vertical folds on the fore surface of each lobe. On the hind surface they are more numerous (about nine), but are almost rudimentary. A basal ridge is present on the hind surface of each molar, that on Mi being little more than a slight swelling. As a general rule the ornamentation on the posterior molars is less complicated than on the anterior ones. A longitudinal link connects the fore lobe with the hind one. The Ramus is strong and powerful, it is convex vertically and slightly so horizontally. The anterior vascular foramen is slightly in advance of the premolar and of fair size, as in S. atlas of Owen, but the second one is placed under Mi instead of beneath one of the posterior molars. The under surface of the ramus shows a distinct upward arch, the posterior portion passing up in an unbroken curve to form the under aspect of the coronoid. Of the coronoid suffi- cient is present to show that the anterior margin of the ascending ramus ran at right angles to the line of the teeth ; the condyle is absent. The “ fenestral vacuity” takes the shape of a pouch, having its opening slightly below the level of the teeth, it contains the posterior outlet of the dental canal and a fenestral foramen communicating with the inner pouch of the ramus. The symphysis is anchylosed and the diastema very short. 34 ] RECORDS OF W.A. MUSEUM. Specific Characters. Longitudinal links continuous with the outermost of the in- cumbent folds, low but distinct, a second link lower and very indistinct in a worn tooth ; present in the mid-valley of all the molars. Posterior basal ridge absent in Mi and but faintly seen in the other molars. Mandible thick, symphysis anchylosed. Incisor inclined, posterior dental foramen below the level of the teeth, level with the ectalveolar groove. Anterior edge of coronoid process rising at right angles to the line of the teeth. Under surface of the mandible arched upwards. Diastema short. Ramus thinner than in 5 . areas (De Vis) and deeper than in 5 . atlas (Owen). Differences and Resemblances. S. areas (De Vis)^ has the same general outline as our form, the same depth of ramus and an anchylosed symphysis. On the other hand its ramus is much stouter, 22. 5-25. 8mm. against 17mm., and the molars broader. S. atlas (Owen). This species, according to De Vis, is dis- tinguished from the preceding by having a much more slender ramus and by the fact that the symphysis is not anchylosed. In shape, too, the ramus is “flat exteriorly, increasing in depth pos- teriorly. Lower contour line flat or arched upwards.” Against this, our form is thick, the exterior surface is convex and the lower contour line arched upwards ; the symphysis is anchylosed. Again,, the incisor of the new species is much more erect and the diastema appreciably shorter. In the older form the depths of the jaw are 26.1 to 28.5mm. and 29mm.-32.7mm., against 34 and 35mm. ; the teeth are smaller in the animal from the Eastern States. ADDITIONAL SPECIMENS. The collection contains numerous rami and portions of ramf belonging to members of this genus. They all belong to adult or aged (Thomas) animals, having; all the four molars in line and more or less worn. In consequence they give every opportunity for comparison with the type specimen 1 De Vis, loc. cit, pp. 89 and 96. RECORDS OF W.A. MUSEUM. [35 of Sthenums occidentalis the species I had the honour to examine, describe and name for the Caves Board. The most perfect specimen consists of the left ramus showing the condyle, the outer “ pouch ” the fenestral vacuity, the fenestral foramen, and the posterior outlet of the dental (vascular) canal ; but wanting the upper portion of the coronoid process. All the cheek teeth are present, though fairly worn ; of the incisors only the bases are preserved, the teeth themselves having most likely been lost in the course of the excavations. There is a complete anchylosis of the symphysis as in the type specimen. Of the right ramus we have all the cheek teeth in good condition, but not a trace of the ascending ramus, the coronoid process. All the measurements practically coincide with those of the type, there being only an occasional variation of less than imm. Judging by the amount of wear exhibited by the teeth, the animal evidently attained a considerable age before it met its death. A second specimen consists of the practically complete left mandibular ramus comprising the incisor, all the cheek teeth, practically all the coronoid, the condyle, and what is so often missing, the angle, in this case almost intact. The specimen is evidently a Stheniirus occidentalis, almost every measurement agreeing with the type. There are two additional specimens consisting of the four molars of the right and left ramus respectively, in which all the teeth show signs of wear, the dentine being exposed in Mi, and also slightly in M2. The whole series measure 47mm. against 45mm. in the type, but I have no doubt that these two belong to the same species, being simply the remains of a larger individual. There is no appreciable difference in the tooth sculpture between these and the type molars. 36 ] RECORDS OF IV. A. MUSEUM. Lastly, there is a specimen comprising the four molars, the last almost in place, the base of the coronoid and the posterior portion of the horizontal ramus which is of interest, as the teeth have a somewhat different type of ornamentation from that of the previous rami and are somewhat smaller. The ramus, too, is more slender and has a decided angle at the posterior lower end of the horizontal ramus in place of the graceful curve that characterises S. occidentalis. In these respects it resembles the right ramus of a young individual obtained in the cave when the deposit was discovered some years ago. These rami of young examples of S. occidentalis are of value, as they allow us to form a better idea of the appearance of fresh and unworn molars. The type specimen is the mandible of an old individual, rendering the accurate description of these teeth rather difficult. It is gratifying to be able to state that a careful examination of the teeth of these younger examples has confirmed the statements made in my report to the Caves Board of W.A. some months ago. The only additional feature to which attention may be drawn being the fact that the outer aspect of the “ external incumbent fold on the fore lobe which sweeps round to form the sharp ridge along the anterior talon of the tooth,” is ornamented with two or three vertical folds which are most prominent on the true molars. Mi to M3, and only rudimentary on M4. The angle at the base of the coronoid of the younger individuals to which reference was made above, is a character which loses its sharpness as the animal increases in size, for we have several intermediate stages from a young animal with its deciduous molars, D2-D3, still in position, to the aged individual that has had its last molar (M4) in use for some considerable time. As might be expected, these younger animals have a ramus which is considerably shorter, shallower, and much more slender than that of a full grown member of the same species. It must also be added that their teeth, too, possess a rather smaller “ crown only ” measurement, as they are comparatively speaking unworn and preserve all the sharp edges of the crowns or ridges of the lobes, the links, talons and folds. RECORDS OF W.A. MUSEUM. [37 MUSEUM NOTES. SOME IMPORTANT ADDITIONS TO THE COLLECTIONS RECEIVED SINCE ist JULY, 1909. Zoology. — Sir Ernest H. Shackleton, C.V.O., has presented a Crab-eating Seal, Lohodon carcinophagus, and an Emperor Penguin, Aptenodytes forstevi, collected on the South Polar Expedition. Mr. Gregory Mathews has sent, in exchange, specimens of the Tinamous, Timmus laiifrons, and of the Sand-grouse, Pteroclis bicinctus, examples of two orders of birds not hitherto represented. Botany. — Mr. J. H. Maiden, F.L.S., the Government Botanist of New South Wales, has given specimens of two new gums, Eucalyptus woodwardi and E. morrisoni, which were collected by Mr. Henry Deane, M.A., far to the East of Kalgoorlie during the Transcontinental Railway Survey. Mineralogy. — A series of Antarctic rocks, of lava and pumice from Mt. Erebus, forwarded by Professor T. W. E. David, of Sydney University, on behalf of the donor, Sir Ernest H. Shack- leton, C.V.O. Various W.A. rocks, ores etc., from residents in this State. Paintings. — The late Hon. Sir George Shenton, Kt., has bequeathed the oil-colours “ Evening on the Sussex Downs,” by A. F. Grace, and two seascapes. County Clare, Ireland. A Portrait of a Lady of the Court of Louis XIIL, by Phillippe de Champaigne (1602-1674), and the “Widow’s Acre,” by G. H. Boughton, R.A. (1834-1905), have been purchased on the recommendation of Sir James D. Linton, P.R.I. The Hon. Dr. J. W. Hackett has presented three oil paintings, a Sketch on the Normandy Coast, by J. W. Morrice, a Canadian artist ; the “ Reprimand,” by Henriette Browne (Madame Jules de Saux) (1820- 1904); and “ Loves me, loves me not,” by E. Phillips Fox. “A Feather fallen from the Wings of Cupid,” by Rupert C. W. Bunny, has been purchased. Messrs. Fox and Bunny are the first two 38 ] RECORDS OF W.A. MUSEUM. Australians to have attained the distinction of election as Associes de la Societe des Beaux Arts in Paris. The Gallery is indebted to Sir James D. Linton, P.R.I., for the present of a charming “ Portrait of a Lady ’’ by Sir Wra. Beechey, R.A. {1753-1839). Arts and Crafts. — The Imperial Academy of Science at St. Petersburg has sent out a very valuable collection of Russian porcelain, which includes a statuette, “ Lost in Thought.” These have been acquired in exchange for native weapons and other duplicate specimens. 5 N.B, — With the exception of the Frontispiece, the plates were made by Mr. H. J. Pether, the Government photolithographer, from photo- graphs taken by Mr. G. Pitt Morison, of the Museum. It has been found necessary to print each block on a separate page, therefore there are two each of Plates II., III., and V. Phascolumys haclietii (sp. Jiuv.j Plate IP, Fig. i,— TOP VIEW OF SKULL. , I < Kf ■J^^ MELBOURNE S3J4»W\ loches Phascolomys hachetti (sp. nov.) Plate II,, Fig. 2.— TEETH OF UPPER JAW. ■ r • ? 1 AL M U 5 E u'v! M t kB C L,' K N E Phascolomys hachetti (sp. nov.) Plate III., Fig. 3.— TOP VIEW OF MANDIBLE. S \ ■s,.- \ . - , .- Plate III , Fig. 4.— SIDE VIEW OF MANDIBLE, • 7 -' ; •V » r- /S .'- ••.' ' ■ • • .••'•■• .<• - •Vi’ liAriCnAl. VU3ELIY YELECEENE A? J C/ Mefres 0 ' 1 2 3 4 5 6 7 8 3 10 II 1 1 . ... ■ — ,-4- Inches 0 1 1 3 4 f : l , l „ 1 Phascolarctus cinereus (Goldf. sp.) Plate IV., Fig. 5.- SIDE VIEW OF MANDIBLE. ) tr)ch'is Stheuiii-tis occiilentalis (sp. nov.) Plate V., Fig. '6.— TOP VIEW OF MANDIBLE, Gentlrriofres i ! '« i I 1 t- t; •J3 f . y V M ? NE Stheniivus occidentalis (sp. nov.) Plate V., Fig, 7.— SIDE VIEW OF MANDIBI.E. K J L L' v'm a LB r F! NE - OF THE WESTERN AUSTRALIAN Museum and Art Gaixery Edited by the Director, BERNAED H. WOODWARD, F.G.S,, C.M.Z.S, ^OLTJLdlE] I. IT. CONTENTS : PAGE The Mammoth Cave (Contd.), by L. Glauert, F.G.S. - 39 Fossil Remains from Balladonia in the Eucla Division, by L. Glauert, F.G.S. - - - - 47 Determination of the Exact Localities where Cambrian Fossils were collected by E. T. Hardman in 1884, by L. Glauert, F.G.S. - - - - 66 Note, by the Editor - - - - - 73 Permo-Carboniferous Fossils from Byro Station, Murchison District, by L. Glauert, F.G.S. - - 75 Notes on Some Western Australian Fishes, by Allan R. McCulloch, Zoologist, Australian Museum - - 78 A Visit to Bernier and Dorr6 Islands during August and September, 1910, by Otto Lipfert - 98 List of Birds observed on Dorre and Bernier Islands, by Otto Lipfert - - - - - too Museum Notes. — The Trustees Act - - - 102 Recent Additions to the Collections - - 103 Printed by Order of the Trustees 'by The Paragon Printing and Publishing- Co., Hay Street, Perth. IfA r^RIOE, 3s. 6ci. RECORDS OF W.A. MUSEUM. [39 THE MAMMOTH CAVE {Continued). By LUDWIG GLAUERT, F.G.S,, etc. 1. MAXILLARY TEETH OF A NEW VARIETY OF NOTOTHERIUM. 2. REMAINS OF DIPROTODON AUSTRALIS (Owen). Order MARSUPIALIA. Sub-Order DIPROTODONTIA. Fam. PHALANGERID.E. Sub-Fam. Nototherin.e. Nototherium, Owen (1845)’^. N ototherium , sp. Syn. Zygomaiums, De Vis, Proc, Royal Soc., Queensl,, Vol. V., p. 114, i88g. NOTOTHERIUM. (Plate vi., Figs. 8, g, 10) The genus Nototherium was established by Prof. Owen in 1845, and comprises animals which are somewhat less in size than the better known genus Diprotodon. The animal closely resembled its larger companion, but also had many points in common with the Wombat (Phascolomys). In referring to the limb bones and skeleton, Lydekker“ states : “The limbs are of equal length ; the humerus has an entepicoudylar foramen, and closely resembles that of Phascolomys ; the olecranon is well developed, and the other limb bones and vertebrae are of the type of those of the latter genus. The structure of the feet is not fully known, but it apparently approximated to that of Phascolomys. This family (the N ototheriidce) connects the Phascolomyidw with the Diprotodontidce ; the cranium, limb bones, and vertebrae being 1 Rep. Brit. Ass Adv. Sci. for 1844, XIV., pp. 231-235. 1845; and Cat. Mamm. and Aves, Mus. Royal Coll Surg., p. 314, 1845. 2 Cat. Foss. Mamtn. Brit. Mus. ^N.H,), Part V,, page 161. 40 ] RECORDS OF W.A. MUSEUM. nearest to those of the former, the mandible showing characters common to the two, while the cheek-teeth are of the type of the latter. It is easy to see how the structure of the cheek-teeth could pass into that of the Phascolomyidce ; and it is not improbable that the two families may have diverged from a common ancestor. The structure of the humerus apparently indicates fossorial habits.” Two teeth from the upper jaw of a large Diprotodontoid marsupial were obtained from the lower stratum.* The one tooth is a premolar and the other a molar. They both belong to the left series. The bilobed molar (plate vi., figs. 8 and 9) has a rhomboidal outline, narrower posteriorly, with the angles obtusely rounded ; the anterior surface and the sides measure from 35 to 36mm. in linear extent, the length of the posterior face is 32mm. The two lobes of the tooth are parallel and en echelon, the hind lobe projects farther inwards and is rather shorter and narrower ; each lobe shows a slight swelling at the inner extremity, is convex on the anterior face and concave backwards. The crest of each lobe is obliquely truncated, the exposed dentine sloping anteriorly. As the fore lobe is more worn than the hind one it has developed a larger tract of bared dentine. Distinct anterior and posterior basal ridges are present, each terminates in a tuberosity at its outer end, and does not ascend the outer side of the lobe. The posterior ridge and tuberosity are less developed than the anterior ones. The inner continuation of the anterior basal ridge ascends the interior face of the lobe, gradually diminishing as it approaches the summit, on the posterior face of this lobe it descends to form a ridge closing the inner end of the mid valley, it then rises on the fore side of the hind lobe, subsiding towards the summit. A continuation of the posterior basal ridge rises upon the interior face of the hind lobe. On the outer aspect of the tooth a distinct isolated ridge closes the outer end of the mid valley. 1 Ante p. 12. RECORDS OF W.A. MUSEUM. [41 At a point in the mid valley of the tooth, about one-fourth from the outer border, a slight swelling or rudimentary link connects the two lobes, from this point the surface of the valley slopes gradually to the inner and outer confining ridges. The unworn enamel seen on the floor of the mid valley and near the anterior and posterior ridges is punctate. The tooth differs from the upper molars of the two species of Diprotodon, D. australis (Owen) and D. minor (Huxley) in having its lobes placed more obliquely across the tooth, in the inward pro- jection of the posterior lobe and the form of the anterior and posterior basal ridges. The upper molars of D. australis (Owen) are also much larger. Prof. Owen’s genus Nototherium possesses upper molars with obliquely directed lobes, anterior and posterior basal ridges^ enclosed mid valley and enamel punctate when unworn. The second upper molar (Mi of Owen’s monographs), as described and figured, agrees with the tooth under consideration in size, outline, and ornamentation, but the anterior basal ridge is stated to be “ continued at both ends upon the corresponding border of the anterior lobe^” without a tuberosity externally. In view of the close resemblance, this molar has been identified as belonging to a species of Nototherium and the presence of the tuberosity on the anterior basal ridge regarded as an individual or specific character. The premolar (plate vi., fig. 10) may be described as under:— Outline sub-triangular, the outer side being the longest. The anterior portion, or talon, is occupied by a cusp having the form of an equilateral triangle with one of the angles situated posteriorly and a slightly rounded side forming the anterior aspect. The crown of the tubercle is worn obliquely, so that the dentine is slightly exposed. This surface is considerably lower than the posterior portion of the tooth, showing that the opposed tooth of the mandible must have had a bulky anterior prominence such as shown in all Owen’s figures of D3 of the lower jaw of Nototherium. 1 Phil. Trans. Royal Soc., Parti, 1872, p. 68; and Ext. Mamm. Australia, p. 276, 1877. 42 ] RECORDS OF W.A. MUSEUM. The rest of the tooth consists of two ridges at right angles united in front and enclosing a simple valley behind. The outer ridge occupies about two-thirds of that side of the tooth and contains an anterior tubercle, the enamel of which is partly worn away, exposing a ridge-shaped surface of dentine having its apex posteriorly. The interior angle of the tooth is occupied by a tubercle which, too, is so worn as to expose the dentine; in this instance also triangular, but with the base situated intero-posteriorly and covering a much larger area than the other two patches already referred to. The surface of this patch is much lower than the others and slopes so that the base is considerably lower than the apex of the triangle. There is a slight swelling at the base of the posterior aspect which may represent a rudimentary cusp or tubercle. The anterior tubercle is divided from the posterior crown by two valleys, each falling from a slight bridge which connects this cusp with the rest of the tooth and each closed by a slight ridge. The valley enclosed by the two posterior ridges slopes very abruptly to the posterior margin of the tooth, where it also was enclosed by a confining cingulum or ridge which has its origin at the base of the anterior tubercle of the longitudinal ridge ; it runs along the base of the exterior, and at the extero-posterior angle rises half way up to the summit of the tooth, then it again descends to enclose the posterior valley, and appears to ascend to the present worn surface of the internal tubercle. It can then be traced down the anterior face, across the mouth of the antero-internal valley and into the mass of tubercle of the anterior talon. The fangs are typical of Dipvotoiion and Nototherium premolars. The premolars of Dipyotodon and Nototherium have led to a great amount of confusion. Phe milk teeth have been mistaken for permanent ones, and teeth of the one genus have been ascribed to animals of the other. The upper premolar (D3 of Owen) of Dipyotodon is not often found in the jaw as, being deciduous, it is shed as the animal approaches maturity. Prof. Owen figures a tooth from the lower jaw^ m its natural position taken from a sketch transmitted to him by a Dr. E. A. Hobson— which shows r Ext. Mamm. Australia, p. 204, 1877 RECORDS OF IV. A. MUSEUM. [43. that, like the permanent cheek teeth, it consists of two lobes separated by a mid-valley. From the state of wear of the crown, this great authority considered that the upper premolar also had two parallel lobes. The following tooth (D4 of Owen) has a somewhat similar shape, but is of greater size. Prof. Owen was so thoroughly convinced of the accuracy of his determinations and conclusions that he never agreed with the views expressed by Prof. Huxleyh As late as 1877 he states* *: “ The Diprotodon minor of Huxley is founded on the teeth of the species of Nototherium." It will be seen he did not fall in with the suggestion that there had been a second, smaller species of Diprotodon coeval with the larger D. australis, but stated most positively that the specimens to which Prof. Huxley referred were really species of Nototherium (N . niitchelli and N. victoria). At a later date De Vis, when attempting to identify and name the large number of Pleistocene marsupial remains in the Queens- land Museum, was able to extend our knowledge of the identity of these preniolars. In 1888 he sent a contribution to the Royal Society of Queensland “On Diprotodon minor (Huxley)*,” in which he summarises his conclusions in the following terms'*: “The premolars figured by Prof. Huxley are unmistakably teeth of Diprotodon. The distinctnes.s of the animal they represent from D. australis, affirmed with some reserve by Prof. Huxley, and practically without reserve by Sir R. Owen, is confirmed by fresh evidence. The differences between the three premolars made known are reconcilable, the difficulty raised by them les.s than that of admitiing three allied species in the same habitat. They represent one form, D. minor, which is a species, and not the female of D. australis. The genus therefore contains two Queensland species, D. australis (Owen) and D. minor (Huxley) The plate which illustrates the conclusion of De Vis shows the upper premolars of D. australis and of D. minor— this latter including the D. australis (?) of Prof. Huxley’s contribution of 1862. The Quart. Journal Geol. Soc., XVIII., p 422, et seq., 1862. ^ Loc. cit,, p. 511. “ Proc. Royal Soc. Queensland, Vol. V 1888 ; p. 38, 1889. * Loc. cit,, p. 44. 44 ] RECORDS OF W.A. MUSEUM. two parallel lobes are a striking feature in the premolar of Diprotodon and can be seen in every figure. As they are not present in the Mammoth Cave premolar, it is impossible to regard this tooth as belonging to this genus. Premolars of Notothenmu as well as the upper premolar of Macleay’s Zygomaturus trilohus, are figured by De Vis in the same volume.^ The Mammoth Cave premolar is unlike the Nototherium or Owenia^ teeth shown, and differs from the Zygomaturus tooth in the absence of the posterior cusp. In fact, the tooth has a triangular outline like N . iiierme shown in figure 5, plate xmi., of Owen’s “Extinct Mammals of Australia,” which it resembles more than the other Notothenmu premolars shown on plate lxxxviii., figures 11-17, 01' miniature on plates xxxvi. and xxxvii., or in full size in figures 3 and 4, plate XLiii. It differs, however, from Owen’s N . inernie in several important points. This author describes the upper premolar of that species as follows, on p. 277. The tooth “ is relatively smaller and less complex on the grinding surface than in the corresponding tooth in N mitchelli, the transverse and antero- posterior diameters are alike. The outer lobe or division has one coronal prominence upon which a slender triangular tract of dentine is exposed on the shorter, inner lobe ; an anterior and a posterior basal ridge bounds corresponding depressions divided by the con- fluence of the apices of the outer and inner divisions at the centre of the crown ; a short external basal ridge closes the concavity impressed upon the hind half of the outer surface of the crown. Owen described (loc. cit., p. 275) the tooth D3 of the upper jaw of N. mitchelli in the following terms. The tooth “ may be said to be two-lobed, but is divided in an opposite direction to that in the rest of the series, viz., into an outer and an inner, rather than a front and a back lobe. The working surface is sub-triangular in form, the angles obtusely rounded . . . The outer lobe or division is the chief one and constitutes the outer two-thirds, and the whole fore- and-aft extent of the tooth ; the outer side of its base swells out like part of a cingulum or ridge ; the summit is sub-compressed and seems to have been tritubercnlale ; the inner and lower division consists of a larger hind tubercle and a smaller front one It is r Proc, Royal Soc. Queensl., Vol. V., 1888, p. in, et seq., and plate , 1S89. Renamed “ Eiiovvenia,” as “Owenia” was previously occupied. RECORDS OF W.A. MUSEUM. [45 implanted by two roots, one behind the other, the posterior one being the larger and grooved anteriorly, as if preparatory to further transverse subdivision.” De Vis defines the upper premolar of Zygomaturus in these words': “ Its posterior two-thirds are occupied by a longitudinal ridge on the outer side, and two tubercles on the inner, the outer ridge and anterior tubercle being joined by a low link over which passes the longitudinal sulcus dividing the ridge from the tubercles. The anterior third supports a single large tubercle or rudimentary lobe.” The Mammoth Cave tooth is thus seen to resemble this premolar very closely, differing only in the state of development of the posterior tubercle of the hind lobe ; in the Zygomatuvus of De Vis it is prominently shown, but in the tooth from the Mammoth Cave it is only rudimentary. Mr. Lydekker states^ that the premolars of Nototherium are very variable, and therefore includes under one specific name Owen’s various forms of Nototherium as well as the fossils known as Zygomaturus trilobus by some authorities. This tooth is therefore classed at the upper premolar of Nototherium, and to show its close affinity to the Zygomaturus premolar the name Zygomaturus is entered as a synonym. There seems no doubt that the molar and the premolar from the Mammoth Cave have both belonged to the left upper tooth series of the same animal, a species of Nototherium. ^ Loc. cit., p. 114. * Ann. and Mag. Nat. Hist., series 6, Vol. III., p. 149, et seq., 1889. 46 ] RECORDS OF W.A. MUSEUM. Sub-Fam. Diprotodontin^,. Diprotodon, Owen (1838)1 Diprotodon australis, Owen (1838)“ THE DIPROTODON. There are many fragments of large bones which are very difficult to identify, but it is not at all unlikely that a number of them belong to this species, the largest Australian marsupial. Up to the present, those specimens which undoubtedly belong to D. australis consist of the head of a femur, several fragments of the ilium, including the socket which received the head of the thigh bone. It is quite possible that some of the other fractured and much decayed bones and bone fragments may belong to this animal, but there is so little material of value to work on that it would be too hazardous to express any opinion on the subject of their definite identity. Some of the portions of ribs certainly are of very great size, and seem to have belonged to an animal quite as large as the creature which yielded the bones from which the cast of the Diprotodon australis skeleton in the Mammalian Gallery was taken. A number of bones in a more perfect state of preservation but presumably all belonging to a smaller animal, have also been classed as bones of D. australis. They comprise radius, ulna, femur, clavicle, imperfect ribs, and cervical vertebrae — of these, the limb bones show all the Diprotodon features. The olecranon of the ulna is short, not prolonged as in Phascolomys (the Wombat), and the radius has a characteristic twist. The dimensions of the limb bones are as under : — Total length of ulna, including the olecranon . . 36 cna. „ „ radius .. .. .. 32-5010. ,, ,, femur .. .. about 46 cm. They are much smaller than the corresponding bones of D. australis, when fully grown, but evidently do not belong to the N ototherium whose teeth were found in the deposit, as the ulna and radius do not indicate fossorial habits. In Mitchell's “ Three Expeditions into Eastern Australia,” 2nd Edition, Vol. II., p. 362 (1838). 2 Loc. cit. RECORDS OF W.A. MUSEUM. [47 FOSSIL MARSUPIAL REMAINS FROM BALLADONIA IN THE EUCLA DIVISION. The Balladonia “Soak.” The Balladonia Soak, which has yielded the rich variety of fossilized animal bones and teeth about to be described, is situated at the foot of the outcrop of the Granite Rocks at Balladonia, in the Eucla Division of this State. These rocks cover an area of approximately 200 acres, and rise to a height of about 50 ft. above the surrounding limestone plain. During the rainy season the water from this catchment area drains into the encircling superficial deposits, where it is retained for some considerable period. In this, as in many other soaks, the loss by drainage and evaporation during the year is less than the quantity of water yielded by the annual rainfall, so that the store of water is practically permanent in character. In the winter time the rocks are surrounded by an expanse of water, which gradually decreases in area and becomes fringed by a border of swampy land, in which animals often become entombed in their efforts to reach the pools of water beyond. Carnivorous animals such as Thylacoleo and Sarcophihis would also make the country surrounding the water their haunt, and in addition to their own remains, would leave behind them bones of the animals upon which they preyed. That both Thylacoleo and Sarcophilus lived in the locality is proved by the presence of their bones and teeth among the other remains presented by Messrs. Wm. Ponton and John Sharp of Balladonia. Broken bones with undoubted tooth-marks upon them have been recognised, but up to the present it has not been possible to make those careful and painstaking investigations necessary to determine the identity of the animal which left the markings upon the fragments. When opportunity offers, however, experiments similar to those recorded 48 ] RECORDS OF W.A. MUSEUM. by Prof. Baldwin Spencer and Mr. R. H. Walcott, in the proceed- ings of the Royal Society of Victoria, Vol. XXIV., Part I., 1911, pp. 92-123, will be undertaken. The Museum and Science generally are greatly indebted to Messrs. Wm. Ponton and John Sharp for these collections, the value and significance of which cannot be estimated until all the specimens have been subjected to examination. Up to the present eight species of extinct Marsupials have had their range extended into Western Australia, and of these several do not appear to have been recorded in the adjoining State of South Australia. Order MARSUPIALIA. Sub-Order POLYPROTODONTIA. Fam. Dasyurid^e. Sarcophilus, Cuv. (1838) i- Sarcophilus laniaritis, Owen (1838). 2 Dasyurus laniarius — Owen .. Owen in Mitchell's “Three Expeditions into Australia,” 2nd edition, Vol. II,, p. 363. 1838. Sarcophilus laniarius — Owen . . Lydekker Cat. Foss. Mamm. Brit. Mus., Pt. V., p. 265, 1887. SARCOPHILUS LANIARIUS. This species is represented by eight specimens, the most perfect being a portion of the skull comprising the maxilla and part of the premaxilla with the right upper tooth series from I3 to M4 and showing a portion of the outline of the infraorbital foramen and the palatal vacuity. This specimen agrees in size with Prof. Owen’s figures on Plate V. of Ext. Mamm. Aust. The length of space occupied by Mi to M3 is 39mm. Four other fragments show portions of the upper tooth series; in every case the teeth are well worn, the animals to which they belonged were therefore adult or aged. The largest specimen has the infraorbital foramen perfect. A piece of the left lower jaw shows the four molar teeth in 1 F. Cuvier. Hist. Nat. des Mammiferes, pi, 70, 1838, “ Sir T. Mitchell's “ Three Expeditions into the Interior of Eastern Australia,” 2nd edition, Vol. II., p. 363, 1838, as Dasyurus laniarius. RECORDS OF W.A. MUSEUM. [49 position, one of them perfect and the others more or less damaged. The mandible agrees in dimensions with Owen’s figured specimens and exhibits the posterior predental foramen Ireneath the fore part of the second molar (M3) as in the type. The three remaining specimens, a right mandible with sockets of the four molars, the anterior portion of the right lower jaw with the teeth Pi, P3 and Mi and a fragment containing in situ the canine (C) and the two premolars (Pi and P3) bear a strong resemblance to the S. tirsinus described by McCoy they all show a posterior predental foramen below the anterior part of the first molar as is usual in this species, but in view of the opinion expressed by R. Lydekker ^ have been classed as S. lamamis (Owen). This species has not previously been recorded for Western Australia. Sub-Order DIPROTODONTIA. Fam. Phalangerid.®. Sub-Fam. Diprotodontin.®. Dipretodon, Owen (1838) » D. australis, Owen (1838) THE DIPROTODON. The abundance of remains of Diprotodon found in deposits similar to those of Balladonia would suggest that a number might be expected in the soaks at that locality. Although the bones presented by Messrs. Ponton and Sharp are all fragmentary, it has been possible to indentify many of them as belonging to this animal. They include fragments of ribs, vertebrae, shoulder blades, leg bones and the pelvis, as well as small portions of the maxilla and mandible containing the roots of cheek teeth. The first lot of specimens, consisting of teeth only, had been exposed to the surface for a considerable period, and were, therefore, 1 Prodromus of the Pal»ontology of Victoria, Decade VII., pp. 11-13 and figures on Plates Lxn, and lxiii., 1882. ’ Catalogue of the Fossil Mammals in the British Museum (N.H.) Part V., p. 265, 1887. » In Mitchell’s “ Three Expeditions into Eastern Australia,” Vol II., p. 362, 1838. < Loc. cit. RECORDS OF W.A. MUSEUM. 50J very much weathered. They were, without exception, the teeth of a full grown and aged animal, as they showed signs of having been in use for some time. In all of them the dentine is exposed to a marked extent, some even have all traces of the original mid-valley obliterated. The teeth indentified include the first incisor and the first, second and fourth molars of the upper jaw, and the incisors and first true molar (Mi) of the lower. The whole series of specimens so closely resemble the figures in Owen’s various papers * that -there is no doubt that they belong to Diprotodon australis. Remains of Diprotodon have been found in various localities. In 1883, the late E. T. Hardman '^ collected a leg bone in the Leonard River, near the Devil’s Pass in West Kimberley ; in 1892, the Elder Exploring Expedition’ obtained a fragment of a bone in the Great Victoria Desert ; Mr. F. R. Arthur found the greater part of a mandible to the west of Lake Darlot in 1895’; and in 1909 the writer had the good fortune to uncover a number of fairly perfect Diprotodon bones in the Mammoth Cave, near the Margaret River. Teeth however, are now recorded for the first time, and are a valuable addition to our knowledge of the West Australian form of the species, the identity of which is now beyond doubt. 1 In Phil. Trans. Royal Soc. 2 Hardman's Report on Geology of the Kimberley Districts 1884, p 21 " this bone was identified by the late Prof. McCoy as the long head of a femur of Diprotodon australis 2 Dr. K. C. Stirling, Director, National Museum, Adelaide, S.A. t This specimen is in the Museum Collection. RECORDS OF W.A. MUSEUM. [51 Sdb-Family PHASCOLOMYINvE. Phascolonus, Owen (1872). ^ Phaicolonus gigas, Owen (1859). Phascolomys gigas — Owen . . Encyclopaedia Britannica, 8th Edition, Vol. XVII., p. 175, 1859. ? Sceparnodon ramsayi— Owen .. Proc. Royal Soc. , XXXVI., No. 228, 1884, Phil. Trans. Royal Soc., Vol. CLXXV. pp. 245-248, 1885. GIANT WOMBAT. This animal is represented by a number of teeth, of which several are practically perfect. They include the right lower premolar (PM4), the right lower first molar (Mi), the left lower first molar (Mi), a lower molar which could not be definitely placed, and seven or eight fragmentary cheek teeth. In size these teeth are rather smaller than most of the examples figured by Prof. Owen in the Phil. Trans, of 1872, but the teeth shown in figure 6, plate XL., of that volume have the same dimensions as the Balladonia specimens, which, therefore, may be regarded as belonging to an animal of the same species. In life it must have equalled a tapir or donkey in size. The discovery of these remains considerably extends the known area of distribution of P. gigas, for the animal has not previously been recorded from Western Australia. Possibly the collection of further specimens may enable a final decision to be arrived at in reference to the relationship of Phascolonus and Sceparnodon. r Owen, Phil. Trans. Royal Soc., CLXII., p. 251, footnote p. 257, 1872 ; also Lydekker Cat. Foss. Mamm. Brit. Mus., part V., pp. 157-160,1887; raised to generic rank. 52 ] RECORDS OF IV. A- MUSEUM. Phascolomys, Geoff. (1803).! Phascolomys latifrons, Owen (1845).^ Phascolomys lasiorhinvs, Gould Lasior/iiiws m'coyi, Gray Phascolomys niger, Kreft't ,, lasiorhinus var, niger, Krefft Mamm. Austr. pis. Lix., lx,, 1863 Ann. Mag. Nat. Hist. (3) XI., p. 458, 1863 Mamm, Austr., text to plate v,, 1871 Proc. Zool. Soc., p. 796, 1872 THE HAIRY-NOSED WOMBAT. A number of Wombat teeth have been identified as belonging to an animal of this species. They mostly showed the working surface, the outline of which, together with the extent of the external enamel and the nature of the longitudinal grooving, render the determination a matter of comparative certainty. Several fragments of bone, including portions of the humerus, ulna, radius, and tibia, are undoubtedly phascolomine, and when compared with Prof. Owen’s figures in “ Extinct Mammals of Australia,” plate xcix., etc., they are seen to approach more nearly to the latifront type than to the more slender P. mitchelli. The P. hackeiti from the Mammoth Cave is larger in size, with teeth which are narrower in comparison to their length, antero-posteriorly, and without the faint longitudinal grooving. Phascolomys parvus, Owen (1872J. ^ TFIE DWARF WOMBAT. Four very worn molar teeth much smaller than those of adult P. mitchelli or P. latifrons were found amongst the second donation of specimens. They were at first taken to belong to a young P. mitchelli but there seems no doubt that they should be attributed to an example of the species, P. parvus of Owen figured in plate xix., figs. 6 and 7, and xx., figs. 6, 7 and 8 of the Philosophical Transactions, 1872, of which the founder of the species says on p. 193, “ with 1 Vide ante, p. 15, for synonymy. 2 Owen. Proc. Zool. Soc. for 1845, p. 82, 1845. 8 Owen. Phil Trans. Royal Soc. 1872, p. 193. RECORDS OF W^A. MUSEUM, [53 present evidence of the constancy of size, of the molar series of teeth in existing and extinct species of wombat, such series fully in place and well worn, having a longitudional extent of i inch 5 lines (36mm.) cannot be referred to a species with a longitudinal extent of molars never less than i inch 9 lines (44.5mm.) and usually more ; as e.g. in the Tasmanian Wombat, which is the smallest of the known existing species. ’ The four small molars of this collection when placed side by side measure 30mm. which, making allowance for the missing premolar, practically coincides with Prof. Owen’s measurement of I inch 5 lines for the whole cheek series. Diligent search among the material in this collection has failed to reveal the presence of any premolar teeth that might be referred to this species and none of the many fragmentary phascolomine bones identified show sufficient variation from the latifront type to warrant their inclusion under another species. This animal which has previously been recorded from New South Wales and Queensland must have had a range extending over the whole of the Continent. It is new to this State. Sub-Fam. : THYLACOLEONTIN^. Thylacoleo, Owen (1849-1852). Thylacoleo, Sp. THE MARSUPIAL LION. (Plates vtt., viii., Figs. 1-7). From the time that the first recognised remains of animals of this genus were described in the Philosophical Transactions of the Royal Society for 1859, there has been a great amount of discussion as to the nature of the food of this gigantic relation of the Phalangers. Prof. Owen stoutly asserted that the animal was a carnivore, pointing out the great resemblance between its sectorial cheek ‘ The P. ursinus Coll. Journey of the Fly, J. B. Jukes, has a tooth series 47mm. in length, equal to i inch 10 lines. » Owen, in Gervais's Zool. et Pal., Franc., ist edition. Part i, p. 192, 1849-52. 54 ] RECORDS OF W.A. MUSEUM. teeth and the highly-developed premolars of the Lion and other carnivora. Some authorities of repute, such as Falconer, Flower, Huxley, Krefft, and Boyd Dawkins, were of opinion that the animal was a vegetable feeder, “ being not much more carnivorous than the Phalangers of to-day,” and after years of discussion a compromise was suggested in the words of LydekkeP: “Fuller acquaintance with the anatomy revealed, however, its intimate kinship with the Phalangers, and when it was fully realised it was argued that Thylacoleo must be a purely vegetable eater. Many of the Cuscuses are, however, partly carnivorous in their habits, and in our own opinion it seems probable that in this respect their gigantic extinct cousin resembled them to a certain extent.” In Australia C. W De Vis, late of the Queensland Museum, and the late G. Krefft, of the Australian Museum, actively championed the two theories ; the former was greatly in favour of the “ carnivore ” theory, bringing forward as proof bones showing what were believed to be marks of the teeth of Thylacoleo ; the latter authority was, it seems, the original propounder of the theory which found the more general acceptance in spite of Owen and his followers. In support of his assertions, Owen declared that the highly- developed incisors had assumed the functions of the canines — a very unusual feature, which, however, in the opinion of the writer, is more easily explained than the presence of highly specialised premolars and degenerate molars in the cheek series of a vegetable- feeding animal, where the molar teeth are usually of such great importance in the task of crushing and pounding food. For a predatory animal, the well-developed canine teeth, separated by the incisor series, form an ideal mode of seizing, retaining, and killing prey ; a double grip is thus obtained, which is much more satisfactory than any hold that could be secured with the help of well-developed median incisors, such as those of the rodents and of Thylacoleo. This was a strong point brought forward in the arguments of Krefft and his supporters, for the examples of carnivorous rodents — 1 Lydekker, Lloyd’s Natural History, Handbook to the Marsupialia and Mono- tremata, p. 260, 1896. RECORDS OF W.A. MUSEUM. [55 the shrews atid the insectivorous hedgehog, which is known to kill young rabbits’, by means of its incisors — are by them regarded as exceptions to the general rule, and not to be compared with Thylacoleo, which was said to hunt the Dipvotodon, Nctothenum, and Giant Wombat, creatures equalling or surpassing itself in size. That Thylacoleo should be a vegetable feeder seems to be improbable, on account of its diminutive or suppressed molars; these are quite useless for the purpose of pounding and crushing, nor can these functions be performed by the abnormally developed sectorial premolars. As Prof. McCoy quite rightly observes (Prod. Pal. Viet., Dec. III., p. 8), the opponents of Owen’s theory seem to overlook this point when emphasizing the resemblances between Thylacoleo and the Rat Kangaroos, which have four, or occasionally three, well- developed molars behind each large premolar. Again, we have now abundant evidence pointing to the fact that Thylacoleo was in the habit of crushing the bones of animals which were its contemporaries. De Vis,* Anderson,’’ and Baldwin Spencer and Walcott * have described bones which bear un- doubted impressions of the upper and lower premolars of the animal. Crushed and broken bones, many bearing tooth-marks, are so plentiful in certain localities that it is quite out of the question to consider them the result of playful antics of the young of Thylacoleo ; they must rather be regarded as the refuse after a meal partaken of by this Marsupial Lion. In the writer’s opinion the animal obtained its food, not after the manner of the lion, tiger and dingo, but rather after the fashion adopted by the hyaena, acting as a scavenger and feeding upon the dead and dying. Thjs would explain the presence of its tooth marks on bones found in the 1 De Vis, C. W., “ On Tooth-marked Bones of Extinct Marsupials,’’ Proc, Lin. Soc., N.S.W., Vol. VIII., 1883/4, p. 187. ‘‘On a Femur, probably of Thylacoleo," Proc Roy. Soc. Q'land, Vol. III., 1886, p. 122. “ Remarks on a Fossil Implement and Bones of an Extinct Kangaroo," Proc. Roy. Soc. Viet., Vol XII., i88g. p. 81. ‘‘ Bones and Diet of Thylacoleo," Annals Q’land Museum, No. 5, 1900, p. 7. 2 Anderson, W. “ On Post-tertiary Ossiferous Clays near Myall Creek, Bingera," Rec. Geol. Surv., N.S.W., Vol. I., 1889/90, p. 116. 3 Baldwin Spencer, Dr. W., and Walcott, R. H., " The Origin of Cut on Bones of Australian Extinct Marsupials, ’’ Proc. Roy. Soc. Viet., Vol. XXIV., Pt. I, igii, p. 92. 56J RECORDS OF IV. A. MUSEUM. lakes and soaks, and if we regard it also as in the habit of forming its lair in caves, and dragging thither the more or less mutilated bodies of its victims, we could understand the discovery of gnawed bones and bone-bearing coprolites in cave breccias as recorded from Queensland and New South Wales. Geological Surveyor Anderson * described the tooth-marked bones in the following terms: “Occurring in the ossiferous portions of the clays were numerous small fragments of bones, generally a few inches in length, and chiefly pieces of the longer limb bones, which had been broken into fragments prior to their deposition in the clays. In almost every case the sharp fractured edges and angles of these fragments had been slightly rounded by attrition, but they were by no means so well water-worn as the pebbles which occurred along with them. The fragments of the thicker bones rarely showed an entire transverse section of the bone, which had not only been fractured transversely, but also longitudinally. In the case of fragments of the thinner bones, the transverse section is generally complete, the bone not having been fractured longitudin- ally. The large majority of the broken fragments show unmistakably the teeth-marks of some carnivorous animal, or animals. Most of these marks are, however, too fine to have been produced by the carnassial teeth of Thylacoleo, although there are some of them which seem large enough and coarse enough to have been so produced. The fragments of bones which show evidences of having been gnawed, are chiefly pieces of the shafts of the longer Mmb-bones and ribs. The teeth-marks occur singly along the surface of the fragments, corresponding marks being often present on the opposite surface, indicating the action of the teeth of both jaws on the bones. Generally, however, the marks are confined to one or both ends of the fragments, which often bears evidence of having been bitten sharp off, while close to the sharply bitten end the surface is furrowed with teeth-marks, showing that whatever the animal was, by the action of whose teeth the marks were produced, it had a similar habit to that of a dog, and other carnivora possess, of holding one end of the bone on the ground between the forepaws while it gnawed the opposite free end.” 1 LOC. Cit., p. 122. RECORDS OF W.A. MUSEUM. [57 The specimens identified as belonging to Thylacoleo are not numerous, they include however, the large upper and lower incisors, the third upper incisor, formerly regarded by Prof. Owen as the canine, the characteristic upper and lower PM4 and a fragment of the maxilla and malar bearing in situ a perfect left upper premolar (PM4) with the adjacant first molar, both these teeth are very much worn and evidently belong to a fully developed animal. In addition to these there is a small fragment of the left lower jaw with sockets of the last molar and a portion of the anterior border of the coronoid. These teeth and fragments, in showing signs of much wear, all suggest that the aged animal (or animals) to which they belonged was from i-4th to i-5th smaller than the animals described and figured by Prof. Owen and Prof. McCoy. The first upper incisors, plate vii., figs, i and 2, are mostly fragmentary ; they show that the perfect tooth is curved and compressed, and the enamelled area practically confined to the anterior and exterior surfaces, as in the wombat and the Rodentia. The greatest vertical extent is at the antero-exterior angle of the tooth. On the inner side there is a wedged shaped border confined to the anterior margin, and with its base at the cutting edge. On the posterior aspect the condition was originally similar, but continuous wear has removed much of the enamel which is now almost entirely confined to the cutting edge, a groove in the dentine, presumably caused by the friction of the lower incisor is most distinct. The distribution of the enamel is entirely different on the canine teeth of the Carnivora, which are completely covered on their exposed surfaces and therefore much more powerful than these incisors. The third upper inciSor in the collection closely resembles the tooth figured by Prof. Owen in Phil. Trans., 1871, Plate xi., fig. ii, which Krefft states is the third incisor but which according to plate xn., figs. 15-18, accompanying his article, should be one of the premolars This latter designation is evidently an error, for 1 Krefft. Ann. and Mag. Nat. Hist., Series 4. Vol. X., 1872, p. 172. 2 Log. cit., p. 182. 58 ] RECORDS OF IV. A. MUSEUM. KrefFt states ^ with reference to Prof. Owen’s determinations “ though he figures the small tubercular premolars with the nail headed crowns right enough.” The large upper premolars (Plate vi., fig. 3) has the typical outline ; it is slightly grooved vertically on the outside as well as the inside, the ridges becoming somewhat inflated towards the base. The enamel is worn down obliquely, particularly at the posterior region where the dentine is exposed ; the worn surface is on the inner side of the tooth. .\s is usually the case in Thylacoleo, the small, obliquely denuded first molar is situated on the inner side of the posterior extremity of the premolar. It has suffered much more than this latter, being worn down to the root on its inner surface ; the tooth is clearly the last of the upper cheek series. The first teeth in the mandibular series are the incisors ; of these, one, the right of a fairly young animal (plate viii., fig. 4), and another, the left of an older individual (plate vni., fig. 5, 6 and 7), are fairly perfect. Like all the other teeth from Balladonia, they are smaller than those figured by Prof. Owen. The right incisor has its posterior surface rather more curved than Prof. Owen’s figures ; it has the same number of serrated strengthening ridges, but the point of the tooth is considerably worn, so that an oblique, roughly circular patch of dentine about qn^tn. in diameter is exposed. It is unlikely that the origin of this worn tip was a fracture. Krefft in his paper states^: “In drawing a few of the lower incisors of Thylacoleo ... I noticed, to my astonishment, clear evidence of attrition on the inner side of several. There was no doubt about it, they had touched each other during the lifetime of the animal (as kangaroo teeth do), but generally at the tips only. In one specimen, however, the surface of the inner side was observed to be quite smooth to the extent of one inch on the lower margin. The ridge so prominent in young or immature specimens has totally disappeared, and my supposition that the jaws were loosely attached is clearly borne out.” No other authority appears to have referred to this feature. 1 Loc. cit. , p. 172. 2 Krefft. Ann. and Mag. Nat. Hist., Series 4, Vol. X., 1872, p. 181. RECORDS OF W.A. MUSEUM. [59 The Balladonia specimens both show this polished surface ; in the younger tooth it has only just commenced, but on the inner surface of the other, one of the serrated ridges has quite disappeared —the action seems to commence at the point of the tooth and extend along the anterior edge, and then backwards towards the posterior border. The irregular area of the worn and polished surface is shown at b in figs. 4 and 6 of plate vrii. The presence of these polished surfaces on the incisors of carnivorous or ossivorous animals is, I believe, unique. A carnivore would not require its lower median incisors to work against one another in a scissor-like manner, as do those of the phytivorous Kangaroo, and if traces of attrition are present an explanation must be sought in another direction. On examining the mandibles of the Dasyurida, among the marsupials, and of the Canida, which may be considered their Placental equivalents, it will be found that the two halves of the lower jaw are but loosely connected for a longer or shorter period of the animal’s existence. A number of these mandibles in the Museum Collection were carefully inspected, but no marks of attrition could be seen on the inner sides of the first incisors. It was, however, noted in these animals that during the process of crushing the food the upper cheek teeth are directly and alternately above the lower ones, hence any movement of the mandible due to the force used in crushing or masticating food would be outwards, that is to say, the two median incisors would be forced apart. In Thylacoleo, as restored by Prof. Owen, on the other hand, the relative positions of the large functional upper and lower premolars are such that the upper teeth pass outside the lower ones. 1 he crushing of food between the cheek teeth would there- fore tend to press the lower mandibles together, and cause the incisors to rub against one another. In this way, flat surfaces might be worn on the inner sides of the two median incisors. It is to be hoped that more abundant remains of this animal will be found in the near future, for, could an almost perfect skeleton be examined, many points which are at present difficult to interpret would be made clear. At various dates, odd skeletal bones have been determined as probably belonging to '1 hylacoleo, and suggestions made concerning 6o] gECORDS OF W.A. MUSEUM. the outline of the perfect animal. According to one of these, the Marsupial Lion, unlike the other Phalangers, would have progressed after the manner of Kangaroos and Wallabies, and not on all fours as Prof. Owen had determined. Our knowledge of the animal, its food and habits is very incomplete, and theories concerning its life- history must at present be partly conjectural, but the discovery of a more perfect skeleton would soon definitely determine the nature of the creature’s food. The other lower incisor, besides possessing the obliquely truncated tip and the worn surface on the inner aspect, shows distinct signs of attrition on its posterior face. For a distance of 12.6mm. near the outer edge a lenticular mass of enamel has been removed and a distinct hollow worn in the underlying dentine. Judging from the British Museum (Nat. Hist.) Cast M1958, this area ultimately becomes connected with the gradually increasing patch of dentine exposed at the worn tip. Of the lower functional premolars there is a fragment com- prising the anterior half of the enamelled crown, showing but little wear. Of the left mandibular P4, one tooth, still inserted in a fragment of the mandible, is almost perfect, the enamel coating of the antero-internal angle alone being absent ; the two other fragments are the anterior and posterior halves of the enamelled crowns, both showing longitudinal strips of dentine where the enamel has been removed by attrition. As was the case with the other teeth described, these worn premolars are less than the corresponding teeth figured by Prof. Owen ; in this instance, the Balladonia teeth are from .72 to .8 the size of those from the Eastern States. The discovery of more abundant remains in a better state of preservation will, no doubt, show that the Western Australian Marsupial Lion was distinct from the Eastern species, Thylacoho carnifex (Owen) of Victoria, and Thylacoleo oweni (McCoy)' of Queensland and New South Wales. Remains of Thylacoleo have not previously been recorded from Western Australia. 1 McCoy. Prod. Pal. Viet., Dec., III., p. 9, 1876. RECORDS OF W.A. MUSEUM. [6i Fam. Macropodid^b, Sub-Fam. MacropodiN/E. Macropus, Shaw (1790).! Macropus magister, De Vis (i895).2 Macropus titan, Owen {in parte) .. Owen. Phil. Trans., 1874, p. 248, et seq., plate XXII., figs, 17, 18. Lydekker Cat. Foss. Mamm. Brit. Mus., Part V,, p. 225, 1887, Lydekker Handbook Marsup. p. 254, 1896. MACROPUS MAGISTER. At the time of founding this new species, De Vis went fully into the points of resemblance and difference of M. titan (Owen) and this form of extinct Kangaroo. The specimen from Balladonia consists of a fragment of the left lower jaw containing one perfect molar, presumably M2, and the stumps of the two anterior teeth; the perfect tooth cannot be distinguished from the teeth Ma shown in figures 13 and 14, plate xxn., and figure 10, plate xxvi. of Phil. Trans., 1874, does it differ from several of the second molars of M. magister from Queensland, presented to this Museum by Mr. C. W. De Vis some thirteen years ago. The dimensions of the crown are 14.4 and 8.4 ; the crest of the hind lobe is somewhat worn, so that the tooth must have been in use for some time. The anterior talon has straight converging sides and a slightly sinuous, almost straight, anterior edge ; the lobes are thick, with rounded angles and rounded crests, having their convexity backwards ; the connecting links or bridges are slightly curved, the exterior face being concave. A vertical fold and a groove are present on the posterior aspect of the tooth. There is a bulging of the base and to erect plate. This is the first recorded specimen of this Queensland form for Western Australia. The fact that an undoubted M. titan was obtained in the Mammoth Cave in 1909 is of importance when considering the distribution of the two kindred species. 1 Nat. Misc., i. text to plate xxxiii , 1890. 2 Proc. Linn. Soc. N. S. Wales. 2nd Series, Vol. X., p. 120, 1895. 62] RECORDS OF W.A. MUSEUM. Macropus aiiak. Owen (1859). Proteinnodon anak, Owen (in parte) ,, 0^, Owen ,, roechus, Owen .. ,, mimas, Ovfen .. ,, antaeus, Owen .. Sihenurus brehus, Owen ,, atlas, Owen (in parte) Micropus mimas, Owen sp. . . ,, irrAas, Owen sp. .. ,, roechus, Owen sp. .. ,, iiHaA, Owen sp. Halmaturus anak, Owen sp. Phil. Trans., 1874, p. 275. l.c. 1874, p. 277. l.c. 1874, p. 281. l.c. 1874, p. 278. Ext. Mamm. Aust., p. 448, 1877. Phil. Trans., 1874, p. 272. l.c. 1874, p. 265. Flower, Cat. Vert. Anim. Mus. Royal Coll. Surg., Part II., p. 720, 1884. Lydekker, Cat. Foss. Mamra. Brit. Mus. (N.H.), Part V., p. 207, 1887. Lydekker, Cat. Foss. Mamm. Brit. Mus. (N.H.l, Part V., p. 212, 1887. Lydekker, l.c., p. 214. Troussart, Cat. Mamm. Tom. II , p. 1181, 1898-9. MACROPUS ANAK. Two fragmentary mandibles are conspicuous by the size of the teeth they bear and the limited depth of the horizontal ramus. The portion of the left jaw contains two molars( Ma and M3 in situ.) These teeth are almost perfect; they are but little worn, and therefore show the very faint verticle folds on the anterior aspects of the lobes. The other specimen — a part of the right ramus — shows the last molar rising from its alveolus, and the preceding one up in position ; this exhibits signs of wear on the posterior lobe, but the very faint vertical folds or grooves are still visible. The teeth are much larger than those of Sthenurus, and can readily be distinguished from the molars of Ihe living species of Macropus] they differ somewhat from all Owen’s figures of teeth of the species included in the above synonym, being very broad in comparison with their length. A specimen of Macropus anak (c. 112, Q.M. 10731) in the Museum Collection, presented by Mr. C. W. De Vis of the Queensland Museum some 13 years ago, very closely resembles the Balladonia fossils. Upon referring to a paper by this authority,^ “ A Review of the Fossil Species of the Macropodidae in the Queens- land Museum,” this species is defined, as follows: “Molars smooth, r Proc. Geol. Soc., XV., p. 185 (part), 1859. 2 Proc. Linn. Soc. N. S. Wales, 2nd Series, Vol. X., p. 75, 1895. RECORDS OF W.A. MUSEUM. [6i with rectilinear crests, feeble lines and sharp angles, upper molars without distinct anterior links, lower seldom without posterior talons. . . . These posterior basal talons are generally present as erect plates, raised rims, or a mere but decided bulging of the base.” This description applies to the specimens under consideration, and is made to embrace a number of Owen’s species, as De Vis found, after examining 330 specimens, that there was a good deal of variation, and that intervening forms made the gradations so minute that it was quite impossible to draw any hard and fast line of separation. Restates’: ‘‘ The degree of variation in the length of the cheek teeth found in this species is less than that shown by HalmaUirus rujicollis, and the premolar has a more restricted range of length than in most of the larger existing Wallabies. On the other hand, the width of the teeth and the depth of the mandible have a somewhat greater range of measurement than in living species, and in thickness the ramus is decidedly more variable ; but, as in all the dimensions, the extremes are reached by insensible gradations, excess even in the width of the teeth must be considered a peculiarity of the species. ... It is quite the most abundant macropod of its period.” The specimens from Balladonia resemble the mandible figured by Owen in Phil. Trans., 1874 (plate xxv., figures 7 and 8), as M. anak, and as regards the teeth they most nearly approach those of P. mimas of plate xxvi., figures i, 2 and 3, which they slightly exceed in width. The teeth of P. mimas (plate xxiv., figs, 13 and 14, and plate XXVI., figs. 4, 5 and 6), the P. roechus (plate xxvii., figs. 10, ii and 12), and the P. antaeus, figured on plate cx. of “ Extinct Mammals of Australia,” all show points of resemblance to the teeth from Balladonia, but differ in the relatively less width of the crowns. Remains of this species have been obtained in all the Eastern States and South Australia, but are now recorded for the first time from deposits in Western Australia. It was probably one of the largest Kangaroos, rivalling the huge M. titan, but giving pride of place to its contemporaries, the genera Procoptodon and Palorchestes which have likewise become extinct. 1 Loc. cit., p. 106. 64 ] RECORDS OF W.A. MUSEUM. Sthenurus, Owen (1873).! Sthenurus atlas, Owen (1838). 2 Macropiis atlas, Owen .. .. Mitchell's “ Three Expeditions into the Interior of Eastern Australia,” 2nd Ed., Vol. II., p. 359, 1838. Protemnodon anak, Owen (in parte) .. Phil. Trans., 1874, p. 275. STHENURUS ATLAS. There are several specimens which have been identified as belonging to this species. A small fragment consisting of the anterior portion of the left ramus with the root of the incisor, the diastema, and the permanent premolar rising from its alveolus, has this important tooth in perfect condition. The tooth closely resembles the rising premolar figured by Prof. Owen on Plate xxii. (figs. 5 and 6) in the Philosophical Transactions of 1874, has a somewhat different tooth sculpture, suggesting that this feature is subject to a certain amount of variation (see also Owen’s various figures of this premolar). The size of the diastema, the dimensions of the mandible, and the position of the predental foramen agree exactly with Owen’s drawings. The differences between the lower permanent premolar of this species and the corresponding tooth of 5 . occidentalis have been fully discussed,^ but are more readily discernible when specimens of the two species are placed side by side. Another example (in three fragments) comprises the major portion of the right horizontal ramus, with the base of the coronoid, the extent of the cheek teeth, part of the diastema, and the perfect incisor. The first and second molars, both much worn, are the only perfect cheek teeth present. The longitudinal links have disappeared, and the crown of the tooth is practically reduced to a flat grinding surface ; there is no trace of any verticle grooving or folding. A portion of the outer surface of the permanent prernolar is preserved, the other teeth are represented by roots only. The 1 For synonymy of the genus, vide ante, p. 31. 3 In Mitchell's "Three Expeditions,” 2nd Ed., p. 359, 1838. 3 Vide ante, pp. 32, 33 and 34. RECORDS OF W.A. MUSEUM. [65 incisor which has lost the tip of its root is very much worn, more than half the enamel on the inner side having been removed. The ower portion of the enamelled surface on this face is worn and polished by the friction of the other incisor, showing that the symphysis was lax and that, by a motion of the jaws, the two incisors worked against one another in a scissor-Iike manner. The jaw differs from S. occidentalis in its general outline, the contour of its inner and outer surfaces, the greater extent of the cheek teeth (P4 M4), which is 70mm. against 62mm. in the other species, and by the greater dimensions, both longitudinal and trans- verse, of all the individual teeth. On the other hand, this Balladonia animal has many points in common with the one whose broken mandible is figured by Owen (loc. cit., plate xxii., fig. 9, and plate xxiv, figs. 7 and 8). The extent of the tooth series and of P4-M2, the contour of the bone and the dimensions of the individual teeth are identical. The length of the tooth series and the measurements of the mandible do not come within the limits given by De Vis^ in his description of the species, but, as the specimen figured by Owen has not been excluded by this authority when he restricted the species, and has been allowed to stand by Lydekker’ in his catalogue, this jaw may be regarded as a true 5 . atlas. A third specimen, consisting of the greater part of the horizontal jaw with roots of the four molars, may be included ; it has an outline similar to the preceding, but has its lower margin perfect. This animal is new to the State. Its home has, up to the present, been considered to be in Queensland and New South Wales. The finds at Balladonia have greatly extended its range. 1 Proc. Linn. Soc , N. S. Wales, 2nd Series, Vol. X., p. g8, 1895. 2 Loc. cit., p. 233. 66 ] RECORDS OF W.A. MUSEUM. DETERMINATION OF THE EXACT LOCALITIES WHERE CAMBRIAN FOSSILS WERE COLLECTED BY E. T. HARDMAN IN 1884. By LUDWIG GLAUERT, F.G.S. In 1884, the late Mr. E. T. Hardman, F.G.S , visited the valley of the Ord River in the Kimberley district, where he collected a number of specimens which, together with many others obtained during that and the previous expedition into the Kimberleys, were mostly deposited in the Geological Museum, Fremantle, and the Swan River Mechanics’ Institute, Perth. The pieces of limestone which were consequently found to contain Cambrian fossils were placed in the hands of Mr. R. Etheridge, Curator of the Australian Museum, Sydney, for exam- ination. His manuscript notes, afterwards handed to Mr. A. H. Foord for incorporation in his “ Description of Fossils from the Kimberley District, Western Australia,”* first demonstrate the presence of Cambrian beds in this State. The fossils described are the Pteropod Salterella hardmani and the Trilobite Olenellus (?) forresii. The localities given with the published descriptions are “ Kimberley District ” for the former, and “ River south of Base Line, Kimberley District ” for the latter. These, it must be admitted, are very vague. Numerous attempts have been made to discover the spot where the specimens were collected but without avail. Mr. R. Etheridge stated in 1906“* that Mr. Hardman’s “ specimens were very poorly localised, and I have quite failed to identify his precise locality.” Dr. R. Logan Jack had an opportunity of t raversing a great deal of the country surveyed by Hardman, but he, too, was 1 Geol. Mag. Dec. III., Vol. VII., p. 98, et. seq. 2 Northern Territory of South Australia : N.-W. District Reports (Geological and General) resulting from the Explorations made by the Government Geologist and Staff during 1905 ; General Geology, p. 42, 1906. RECORDS OF W.A. MUSEUM. [67 constrained to admit that he had failed to trace the Cambrian limestone beds from which the Salterella and the Olenellus were collected. In fact, he considered the task so hopeless that he remarked in 1906^: “The palaeontological evidence is also exceed- ingly meagre and, scientifically speaking, ought to be ignored, as the two localities from which Hardman collected Cambrian fossils . . . have been defined too vaguely for identification as Kimberley District and River south of Base Line.” In his presidential address to Section C (Geology) at the Adelaide Meeting of the Australian Association for the Advance- ment of Science in 1907, Mr. A. Gibb Maitland gives as his opinion'': “Despite the fact of poor localisation of Hardman’s fossils, it may, I think, be taken for granted that Cambrian beds do occur somewhere in Kimberley about South latitude 18°. The recent discovery of Olenellus and Salterella in the limestones of the Daly River in the Northern Territory by Messrs. Brown and Basedow is of considerable geological importance, indicating a somewhat wide distribution of Cambrian strata in Northern Australia, and makes the solution of the Hardman puzzle almost imperative.” Mr. H. W. B. Talbot visited the Elvire and Ord rivers in his traverse of the route followed by the Canning Survey Party, 1907- 1909.’ In the course of his journey from Flora Valley Station to Wyndham he collected several pieces of fossiliferous limestone from the Ord River Station. I had the good fortune to handle these specimens for a few moments, and as far as I could judge, the fossils consisted of Salterella hardmani in a hard sub-crystalline limestone of a greyish colour. These specimens are now in the hands of Mr. R. Etheridge of Sydney, and when his report is available I have no doubt that my determination of the age of the rock and the nature of its fossil contents will be confirmed. \ The subject first came under my notice when preparing the “ List of Western Australian Fossils,” which was offered to the Government Geologist and published by him in Bulletin 36 (III. r “The Prospects of obtaining Artesian Water in the Kimberley District.” G. S. W. A. Bulletin, No. 25, p. 15, 1906. 2 Reo. Austr. Assoc. Adv. Soc., Adelaide Meeting, 1907, p 140, 1908. * G. Survey (W.A.) Bulletin, No. 39, Perth, 1910. 68 ] RECORDS OF W.A. MUSEUM. Palaeontological Contributions to the Geology of Western Australia, rgio). Some of the notes then made were recently discovered among other papers, and I decided to investigate matters more thoroughly as I now had the necessary leisure. It was found that, whereas the locality mentioned in the description of Salterella hardmani was correct, the record on the label attached to the Olenellus had been inaccurately copied. In a list of Western Australian Fossils in the Collection of the British Museum (Nat. Hist.) obtained for my assistance by the Acting Government Geologist, Mr. H. P. Woodward, F.G.S., in 1908, the locality for the Trilobite is given as “ River south of Base Line Camp ” and “Ord River” — that is to say, the Trilobite remains were collected in two localities, a fact confirmed by the following paragraph in Mr. Etheridge’s description.’ “ On the weathered surface of a similar limestone rock, and from the same locality as the head just described, there is a short spine probably belonging to the present species ; if so, it would be the telson. In another piece of limestone, similar to those containing the head and telson, there is a portion of a thoracic segment which agrees in form with the first two segments of an Olenellus ; this may also belong to the present species." The addition of the single word “ Camp ” simplifies the localisa- tion of the Cambrian beds from which the fossils were obtained. As the specimens collected in 1883 and 1884 and deposited in the Geological Museum at Fremantle are tabulated in the form of an appendix to Hardman’s 1884 Report, published in 1885, ® search was made, with the result that an entry, “Hb27. Limestone with lead, zinc, and fossils, from river south of Base Line Camp ; several specimens,®” was discovered. No “ Base Line Camp ” is referred to in the Report for 1883, but in that for the succeeding year, paragraphs 295® to 299 describe “ Section between J 34 and the Base Line Cump," and in paragraph 298 we read “ Base Line Camp, Z 27,” which an examination of Hardman’s map shows was placed on the left (north) side of the 1 Geol Mag. Dec , III., Vol. VII., p. 99, 1890. 2 Report on the Geology of the Kimberley District, Western Australia; Perth, 1885, p. 36. s Loc. cit., p, 32. RECORDS OF W.A. MUSEUM. [69 river, north of the Base Line W.B.-E.B.* This would account for the omission of the name of the river, for Hardman would, no doubt, consider this localisation sufficiently distinctive. In paragraph 302 of this Report, Hardman writes, “ Base Line Camp to Hardman Range. On this side of the river the prevailing rock is limestone, which crops out from under the alluvium and soil in low ridges, bearing a little N. of E. In the present section, only one of the ridges is visible. It is about half a mile wide, and has been traced for 10 or 12 miles in a direction to E.-N.E., about two miles from the river and parallel to it.” It may be that these limestone beds do not outcrop in the bed of the river close to Camp Z 27, for the only rocks mentioned in paragraph 298 as having been observed in the river bed between the Base Line Camp and J 34 are “ a few beds of red shales and micaceous mudstones.” The limestone may extend north of the patch mapped by Hardman, but it must be admitted that some of Hardman’s localisations are not very precise. As an example of this it may be mentioned that in paragraph 322 he states : “ Lead, zinc, etc., have been noticed in the form of Galena in the limestone rocks a little south of the Base Line (Ord River),” whereas in the Appendix we read that specimens “ Hb 27. Limestone with lead, zinc, and fossils” came from “River south of Base Line Camp." Therefore, as these minerals and fossils were collected either in the (Elvire) river south of Base Line Camp or (Limestone) ridge south of the Base Line (see specimen HB 18) it may, I think, be concluded that the Hardman Trilobite also came from one of these two localities, presumably from the former. As stated above, one of the Olenellus fragments was obtained from a locality some distance away from the Base Line Camp specimens. This is taken to be the one referred to as “Ord River” in the British Museum List of igo8, and is doubtlessly specimen HB 48 of Hardman’s Appendix, localised as “ River bed Ord, near J 38”^ In paragraph 304 Hardman writes: “About five miles below the junction of the Elvire and the Ord, limestone crops out in the river bed, and in low ridges through the alluvium to the 1 In Mr. Maitland's Presidential Address this line is incorrectly referred to as W.B.-E.F. ; loc. cit., p. 141. ^ Loc. cit., p. 37. 70 ] RECORDS OF W.A. MUSEUM. south, their general trend being E.N.E.” The position of these exposures is opposite J 38. The only other catalogued fossiliferous limestone from the river bed is “ limestone with fish palate, Camp Z 31, river bed Ord,” which obviously is not the specimen named by Mr. Etheridge. The specimen of Salterella hardmani in the Museum Collection is labelled HB 33, and was obtained from Mt. Panton. Hardman’s palaeontological observations in the field do not in any way lead to the belief that his locality south of Base Line or Base Line Camp refers to the Base Line laid down in 1883. Firstly, his list of specimens collected in 1883, in the Appendix of his Report for 1884, does not include a single fossil or fossiliferous specimen collected from or near Mt. Campbell, where the Base Line was situated'; and, secondly, the fossils observed in the rocks near that locality show a purely Carboniferous facies, for they include Poteriocrinus, Atkyris, and Lithodendron'^ in a good state of preservation. In the Report for 1884, numerous fossils are recorded from the Ord river district, including some purely Carboniferous genera, but there is some likelihood of Hardman’s field observations being at fault, for, unlike the fossils from the Western district, these latter were “usually so worn by the excessive action of the weather in this tropical climate that identification as to species is almost impossible in many cases. Of the genera mentioned, only one, Athyris, is more or less confined to the Carboniferous. LITHOLOGICAL EVIDENCE. In describing the Limestone Series (L) of the Ord District, Hardman says': — “ 123. This portion of the limestone country rises in a succession of low and almost imperceptible terraces into high tablelands. One of these extends to the east of the Ord, near the ‘ Cattle Station,’ and another to the north and east of the Negri River, where it is capped by Mt. Panton. 1 Mr. Gibb Maitland's Presidential Address, loc. cit,, p. 141. 2 Hardman’s 1883 Report (Perth, 1884), p. 16. 3 Hardman’s 1884 Report (Perth, 1S85), p. 17. i Loc. cit., p. 17. RECORDS OF W.A. MUSEUM. [71 “ 124. The Ord limestones are for the greater part hard and flaggy, rarely massive, usually grey in colour, sometimes sandy or magnesian, and seldom fossiliferous. In many parts of the district they are interbedded with red shales, marls, and sandstones, the former of which contain occasionally layers of gypsum, together with traces of rock salt. . . . “ 125. Over the great part of this country the limestone crops out in bare masses, cut through by numerous gullies and water- courses, along which the rock forms high cliff's and scaurs, showing the stratification — which dips at a very low angle in various directions — very distinctly.” Dr. R. Logan Jack states': “The strata [of the limestone] which at the south-western and south-eastern boundary of the formation dip to the north-east and north-west, are practically horizontal on the Ord River below its junction with the Elvire, and probably continue to be horizontal where they are covered, on the left bank of the Ord.” These descriptions enable us to define the known outcrops of Cambrian limestones of Western Australia as being hard and flaggy, usually grey in colour, either horizontally bedded or dipping at a very low angle in various directions. Evidence of confirmatory nature is found in the record of the explorations made by the Government Geologist of South Australia and his Staff in 1905. Mr. H. Y. L. Brown reports*: “ During the present examination of the belt of limestone, extending south- easterly from east of Mt. Litchfield to the Katherine River, was proved to be of Cambrian age by fossil evidence. Outcrops of the rock, two miles north of Noltenius Billabong, and about nine miles from the Daly River, consist primarily of a compact blue-grey and yellow sub-crystalline limestone, parts of which are rich in pteropod tests of Salterella, weathering slightly in relief. The beds are horizontal, and the line of outcrop trends north-westerly. No deep section is available, and, wherever encountered, the outcrops rise but a few feet above he surface. The physical features are low, banked, denuded tables, separated by horizontal joint planes, and piled one upon the other, the uppermost being very much smaller 1 Loc. cit., p. 26. Loc. cit., p, 14. 72 ] RECORDS OF iV.A. MUSEUM. than the lowest visible, which rests upon massive beds of the same rock belowd The exposed surfaces have been eroded in a remark- able manner by the atmosphere, and appear in the form of sharp, confluent, serrated ridges, grading downwards on all sides, the whole giving the effect of a model of mountain chains on a small scale. The rock decomposes comparatively readily, and produces a rich red clay soil.^ The beds overlie unconformably the schistose Pre-Cambrian rocks on the east and west. ... In lithological characters, this formation is similar to that covering large areas between the Katherine Station and Flora Falls; also to that at Jasper and Timber Creeks, off the Victoria River : between the Elsey and Daly Waters Stations on the Transcontinental Telegraph Line ; at Anthony’s Lagoon, Brunette Downs, Alex- andria Station, and other places. It is most probably continuous beneath the basalt, sandstone, and other later formations, The occurrence of Cambrian fossils near the Daly River and Alexandria Station prove that these widely separated expanses of limestone are identical in age.” CONCLUSIONS. Hardman collected his Cambrian fossils : — At River bed (Elvire) South of Base Line Camp Z27 (HB27). At river bed (Ord) 5 miles below its junction with the Elvire, opposite the Hill J38 (HB 48). And at Mt. Panton, Northern Territory (HB33).® Mr. H.W. B. Talbot collected Cambrian fossils [Salterella) at the Homestead, Ord river Station. The stratigraphigal results may be summarised as follows : — Cambrian rocks are exposed in Kimberley along the valleys of the Elvire and Ord rivers from near the North-western extremity of the Hardman Range to the Ord River Station, and so on to Mt. Panton (N.T.),< thence they will no doubt be found to 1 This admirably describes the outcrop in the Ord River Valley, and at Mt. Panton. 2 A marked feature round the Ord River Station homestead. 3 The “Limestone with Fish Palate,” is also of Cambrian age (/ostra p. 74). 4 An outlier of carboniferous age as a capping to Ml. Panton would explain the presence of carboniferous fossils at that locality. RECORDS OF W.A. MUSEUM. l73 extend to Jasper and Timber Creeks off the Victoria River, where the beds have been recognised by Mr. H. Y. L. Brown. In due course, when the mapping of the beds is extended, it it will probably be joined up to the Cambrian Limestone of Flora river, the Katherine river, and the Daly river valley, and the other localities further ea.st and south in the Northern Territory and Central Australia. 'I'he exact localisation of these Cambrian beds will have a pro- found effect upon the Geological mapping of this State. When the matter has been further worked out it will doubtlessly be found necessary to regard most of the beds marked “ S ” on the eastern half of Dr. Jack’s map of igo6 as of Ordovician age^. Beds mapped as “ D ” would become pre-Cambrian ^ to distinguish them from the still older metamorphics (M) which might be regarded as Archaean®. The Cambrian Limestone beds of Northern Australia, extend- ing from the Elvire river near the north-western extremity of the Hardman Range to the Daly river (Noltenius Billabong) and Alexandria Cattle Station in the Northern Territory are thus seen to present a remarkable uniformity of character throughout the course of their outcrops, and to be almost, if not quite, horizontally bedded wherever they appear at the surface. NOTE BY THE EDITOR. 19th March, igi2. I have just received the following communication from Dr. F. A. Bather, F.R.S., Assistant Keeper of Geology in the British Museum : — 1 Mr. Talbot has found similar beds to the south of those just mentioned, they extend from lat. 23deg. S. to 25deg. S-, with two smaller patches further south. 2 Mr. H. W. B. Talbot has shown that beds similar to these, and also horizon- tally bedded, extend southwards to lat. 23deg. S., long. i24deg. E. (Bulletin No. 39). 3 A band of these ancient metamorphic rocks was seen by Mr. Talbot to cross his route and extend from Mt. Methwin to Lake Nabberu, a little to the north of Wiluna. 74 ] RECORDS OF W.A. MUSEUM. Geological Department, British Museum (Natural History), Cromwell Road, London, S.W. 19th February, 1912. Memorandum to the Director of the Western Australian Museum and Art Gallery, Perth. The following is an exact copy of all labels, numbers, etc., attached to the specimens of Oleuellm forresti and Saltevella hardmani collected by Hardman in the Cambrian rocks of Kimberley, and preserved in the Geological Department of the British Museum : — (i 2346A) OleneUus ? forresti (Etheridge Jnr. M.S.) Foord Cambrian. " River south of Base Line Camp.” Kimberley, W. Australia. Figd. Geol. Mag. 1890, pi. iv., fig. 2a Presd. E. T. Hardman, 1886. S. of Base line HB27 (i 2346) Do. do. do. fig, 2 Presd. E. T. Hardman, 1886. River S. of Base line HB27 (I 2347) Oltntllus forresti (Etheridge Jnr. MS.) Foord Cambrian. Ord River, Kimberley District, W. Australia Figd. Geol. Mag. 1890, pl.iv., fig. 2b Presd. E. T. Hardman, 1886- Ord River Palaezoic H B 2/32 (g 7428) Saltevella hardmani. Spec. fig. in Geol. Mag. 1890, pi. IV., ff. la-b Presented by Gov. W. Australia per H. P. Woodward, Esq. No. I Ks Salterella hardmani (Ether.) Lr. Palaezoic Kimberley Exhibitor, E. T. Hardman RECORDS OF W.A. MUSEUM. [75 PERMO-CARBONIFEROUS FOSSILS FROM BYRO STATION, MURCHISON DISTRICT. By L GLAUERT, E.G S. A number of fossils of Permo-Carboniferous age, obtained in the course of sinking a bore at Byro Station, on the Byro plains south of the Carrandibby Range in the Murchison, were presented to the Museum by Mr. L. Landauer of Day Dawn. The majority of the fossils have the shell preserved, but others appear as more or less glossy internal or external casts; they include Spififer convolutus (Phillips), 5 . musaJtheyleHsis v. australis (Foord), Chonetes pratti (Davidson), Deltopecten subquinquelineatus (McCoy), and Aviculopecten sprenii (Johnston), Two new forms have been recognised; the Spirifer is represented by a number of specimens which render a diagnosis possible, but the Conularia is imperfect and fragmentary. BRACHIOPODA. Genus Spirifer. Spin fee byroeusis, sp. nov, Spirifer lata (?) .. Eth. til Geol Surv. W .A., Bulletin lo, p. 15, plate i., fig « (1903)- SPIRIFER BYROFNSIS. This species is represented by specimens more or less spindle- shaped, with sub-acute alar angles, beaks not prominent, rising but little above the cardinal edge, slightly incurved. Hinge line straight, cardinal area moderately wide, concave, transversely striated, as wide as the shell. Ribs radiating, numerous — about 40 on each valve — usually grouped in bundles of three, grouping becomes indistinct towards the wings — where the ribs are irregular and gradually fade away, — alar extremities curved, smooth wing- like. Lines of growth distinct towards the margin. Sinus and fold ribbed. Sinus of the ventral valve extends to the apex of the beak. 76] RECOliDS OF ir..i, MUSEUM. Dorsal valve: the cardinal area is rather narrow, cardinal process small, flattened and vertically striated, cardinal sockets long, grooved above, expanded distally. Denticles not observed. Dimensions of the shell, 37mm. by 103mm. The 5 . lata ? of Etheridge’ has a costate sinus, acutely extended and nearly smooth alations, which differ in several details from the shell under consideration. The sinus broadens much more rapidly and is less distinct, the alar angles are more acute, the beak is much more prominent, and the ribs lack the tendency to group them.selves in bundles. The fold of the dorsal valve is not grooved as in the 5 . lata from the Queensland Permo-Carboniferous,^ which shell also possesses plain sulcus and median fold, as well as stronger ribs. The shell differs from .S', latus (McCoy) in having fold and sulcus ribbed, in the number of ribs on each valve, and in possess- ing a cardinal area transversely concave. 5 . convolnlHS (Phillips) differs from the new species in the nature and extent of the ribbing, and in possessing much more attenuated (acute) alar angles. 5 . mnsaliheylensis v. australis (Foot'd), when young, is an alate form with ribs arranged in bundles, but its costation ismuch finer and the wings are not smooth ; it also has a much more prominent beak. There are five specimens (Nos. 1650-1654) in the collection. 'I'he two shells chosen as types (Nos. 1650 and 1651) show the features described above. GASTEROPODA PTEKOPODA. Genus Coniihiria. Conulayio, sp. nov. (?) c. f. C. Warthi .. Waagen, Pal. Ind., Series XIII., Salt Range Fossils, Vol. IV., p. 126, plate IV., fig. 6, plate v., fig. i (1889-1891 ). CON U LARI A, sp. nov. Three crushed fragments on a piece of ironstone (No. 1660) present the following features: — Elongated pyramidal .shell, quadrangular ; faces sub-equal (?), almost flat, slightly grooved in the middle, apical angle unknown, I Loc cit. Jack and Etheridge. Get)!, and Pal., Queensl., p. 229, 1892. RECORDS OF W.A. MUSEUM. [77 probably small ; folds sharp, narrow, punctate, slightly arched, with the concavity towards the apex, alternating at the side angles, interrupted in the middle of their length but rarely alternating, furrows wide — ten to twelve in the space of lomm., becoming narrower and more crowded towards the apex — side angles de- pressed, indistinctly preserved. This Conularia is nearly related to the two Australian species C. levigata (Morris)' and C. inormia (Dana)." It differs from both in having its thin, sharp folds punctate; the section of the shell is unknown, as the examples are all fragmentary, but one worn example suggests that the tube was rectangular in section. C. quadrisulcata (Miller)" has a much greater apical angle, with folds punctate but much more crowded than on the Byro Station fossils. Cnmihivia ( Waagen),^ collected by Dr. Warth in Chel Hill and at Dillur, can hardly be distinguished from the Byro plains shell in its general features, but a most careful examination with a lens failed to reveal the “ fine, somewhat irregular, plication that extends transversely over the spaces or valleys between the single ribs,” visible on the Indian fossil. The presence of the inter-costal tubercle at the angles of the tube could not be definitely ascertained owing to the imperfect nature of the fossil. The Conularia clearly shows a closer affinity to this Indian form than to the Permo-Carboniferous species of Eastern Australia. In his monograph. Dr. Waagen states that the Indian C. ivavthi is associated with the so-called Boulder Beds, which are now con- sidered to be the result of a glacial period of Permo-Carboniferous age. It is of the utmost geological and palaeontological importance to find that a closely allied, if not identical, species of Conularia has been obtained from strata intimately connected with the Western Australian Boulder Beds. The genus has not previously been recorded from Western Australian Sediments. > J Morris, in Strzetecki’s Physical Descriptions of N.S. Wales and Van Diemen's Land, p, zgo, plate xviii., fig, 9, s J D. Dana. Geol. Wilkes U.S. Expl. Exped., p. 709. plate x., fig. 8, 1849. •’ Sowerby's Min. Conch. Ill , p, toy, plate cclx., fig. 5, 1821. ^ Palreontologia Indica, Series XIII., Salt Range Fossils, Vol. IV., p. 126, plate IV., fig. 6 a, b, c, d ; plate v., fig. i a, b. 78 ] RECORDS OF W.A. MUSEUM. NOTES ON SOME WESTERN AUSTRALIAN EISHES. By ALLAN R. McCULLOCH, Zoologist, Australian Museum. Plates IX. to XIII. ; text-figures i, 2. The Australian Museum has received for identification a large collection of fishes from the Director of the Western Australian Museum and Art Gallery, while several smaller ones have been presented to the Trustees by Mr. A. Abjornssen, Chief Inspector of Fisheries, Western Australia. M'hey include a considerable number of new and little-known species, of which some are dealt with m the following pages. Others not previously recognised from the Western State are recorded. I wish to express my thanks to Mr. Bernard H. Woodward for enabling me to examine and write upon this extremely interesting collection, and also for various kindnesses connected with the publication of the paper. My thanks are also due to Mr. Abjornssen, who has made special efforts to secure several species I have particularly asked him for. GONOKRYNCHUS GREYI, Richardson. Rvnchana greyi, Richardson— Zool. Ereb. and Terr., 1845, p, 44. plate xxix., figs. 1-6. Gonorhynchus greyi, Gunther— Brit. Mus Cat. Fish, VII., 1.S68, p. 373 (part); Id., Castelnau, Proc. Zool. Soc. Viet., I., T872, p. 182 ; Id., Khinzinger, Arch, fur Nat., XXXVIII,, 1872, p. 42, and Sitzb. Akad, Wiss, Wien,, LXXX. i., 1879, p 415 : Id., Macleay, Proc, Linn. Soc. N. S. Wales, VI., i88t, p. 255 ; Id.. Johnston, Proc. Royal Soc. Tasm., 1882 (1883), p. 132, and 1890 (1891), p 37; /(f., Ogilby, Proc Linn. Soc. N. S. Wales, XXIV., 1899, p. 154 ; hi, Waite, Rec. Austr. Mus., IIP, 1900, p. 2t I. Gonoihynchus gonjrhynchus. Stead — Proc, Linn. Soc. N. S. Wales, XXXII , tgoS, p. 744, and “ The Beaked Salmon " (Dept. Fish. N. S. W.|, 1908, pp. 1-8, plate I ; Id., Zietz, Trans. Royal Soc, S. Austr., XXXII., igo8, P- 295 Gonoi kvruhtis gonorynchus, Waite — Trans. N. Zeal. Inst., XLII., 1910, p. 374. Gonorrynchus greyi, Ogilby — Ann. Queensl. Mus., No. 10, 1911, p. 34. ? Gonorrynchus forsteri, Ogilbj' — Loc, cit , and synonymy. RECORDS OF W.A. MUSEUM. \79 I have compared three Western Australian examples with four others from New South Wales, one from Victoria, and twelve from Lord Howe Island, and find them to be identical. They all differ from Richardson’s fig^ure, however, m the proportional lengths of the head and pectoral fins. My series includes specimens from 86-255mm. long. The pectorals are from 2^-2f in their dis- tance from the ventrals, not 35, as figured. The head varies from 4^-4f in the length to the hypural ; in the figure it is almost six times. The insertion of the dor.sal fin also is farther forward in my specimens than shown. It must be noted, however, that Stead figures New South Wales examples, the proportions of which approach Richardson’s specimens, while they also agree very well with Hector’s figure of one from New Zealand, which Ogilby has called G.forsteri. It therefore seems to me that the length of the body is very variable, as in many other elongate fishes, and that forsteri must be regarded as a synonym o{ gveyi. If this be accepted, the species is recorded from South-Western Australia, South Australia, Victoria, Tasmania, New Zealand, New South Wales, Lord Howe Island, and the Kermadecs. Specimens of G. parviiiianus (Ogilby),* of the same size as others of G. greyi, have very much smaller heads and pectoral fins, while the scales are cycloid instead of ctenoid. Waite^ considers the one to be merely the young (larval form) of the other, and the close association of the two over a wide geographical area lends consider- able support to the hypothe.sis. Specimens of the parviiiianus form are recorded from the following localities : — Lord Howe Island and Narrabeen, New South Wales; Kertnadec Islands®; Moreton Bay, Queensland. Another is in the Australian Museum from Manly, near Sydney. The type, which is the largest specitnen known, is 88mm. long. 1 Ogilby. Ann. (.)ueen,sl. Mus., No. 10, igii, p. 34. 1 Waite. Rec. Austr. Mus., V., 1904, p. 147, plate xvii., Eg. 3 8 Waite. Trans. N. Zeal. Inst , XLTI., igio, p. 374. 8oJ RECORDS OF W .A . MUSEUM. GYMNOTHORAX WOODWAKDI, sp. nov. Figure i. Muraena iiubila, Richardson— Zool Ereb. and Terr., Fishes, 1848, p. 81, part — specimen from Houtmans Abrolhos. Muraena richardsoHii, Gunther — Brit. Mus. Cat. Fish, VIII., 1870, p. 118— same specimen (not of Bleeker). Gymnothorax punctaiojasciatvs, Waite — Rec. Austr. Mus., VI., 1905, p. 58 (not of Bleeker). Head 2| in the trunk ; head and trunk ip in the tail. Snout 5I in tile head, mouth 2J, eye if in the snout, and almost equal to the interorbital space. Body compressed, snout pointed, the tip rounded. Teeth of adults uniserial in both jaws ; they are small anteriorly in the upper jaw, then large, and decreasing again backwards. In a small example there are one or two large canines inside the others near the middle of the jaw. Mandibular teeth decreasing regularly from front to back. One or two large depressible teeth on the median line of the mouth anteriorly; vomerine teeth uniserial, small, and mostly rounded. Gill-opening smaller than the eye. Origin of the dorsal, midway between the end of the mouth and the gill-opening. Colour. — Light brown after long preservation in spirits, darker posteriorly, with a wide meshed network of dark lines on the upper half of the body. A dark line near the back begins with the dorsal and follows it until it is lost on the tail ; this line is not very di.stincl in my smallest specimen. Some black lines extend from behind the mouth towards the gill-opening; head otherwise plain. Anteriorly the dorsal and anal fins are marked like the body, but posteriorly they are very dark with whitish margins. Described from five specimens, 325-72omm. long, from near Fremantle, Houtmans Abrolhos, and Pelsart Island. The type, which is 515mm. long, is from the latter locality, and is in the Western Australian Mu>eum. An eel in the British Museum, from Houtmans Abrollios, was identified by Richardson as his Muraena nubila, a.nd later by Gunther as M. richardsonii, Bleeker ; it is probably of the same species as the specimens described above. Mine differ from the figures of both nubila and richardsonii, however, in the arrangement of the dark marking RECORDS OF W.A. MUSEUM [8 1 Figs, i and 82] RECORDS OF W.A. MUSEUM. on the body, which is apparently very characteristic. Again, my smallest specimen is one of the three identified by Waite as Gymno- thorax punctatofasciatus, Bleeker ; I consider that its colour pattern separates it from these species also. COKYTHKOICHTHYS P0EC1L(3LAEMUS, Peters. Figure 2. Syii^niitluis f-oeciiolaenms, Peters — Monatsb. Akad. Wiss. Berlin, 1868 (i86g), p. 458: lei. Zietx, Trans. Roy. Soc. S. Austr. XXXII., 1908, p. 298. Syngnathus poekilulaemus, Duncker— Faun, Sudwest-Austr., II., 1909, p, 245 (see references). D. 28. P. 12. C. 10. Rings 19-1-48. Head in the trunk. Head and trunk 14 in the tail. Snout one third longer than rest of head. Eye less than one fourth as long as the snout, and 2 in postorbital portion of head. Snout with ridges but not serrated ; a median keel extends from the mouth to the occiput, but is very low anteriorly and is interrupted between the eyes. The supraorbital ridges commence some dis- tance before the eyes, and extend backward to the occiput; inter- orbital space concave. Nuchal keel low but distinct. Operculum with a prominent median keel. Body rings without spines. Trunk ii as deep as broad, its greatest depth being near the vent. Dorsal fin opposite the vent, occupying i|-l-6 rings; its base not raised above the level of the back. Median lateral ridge ending on the last body ring, and just below the origin of the upper caudal edge. Edges of the back extending to the end of the dorsal fin. Lower lateral edges of the body continuous with those of the tail. Caudal fin, small, shorter than the eye. Egg pouch very large, occupying i8 tail rings and containing 58 large eggs disposed in two, or in places, three rows. Colour. — Uniform brown in spirit, the snout with darker cross bars. Body with numerous minute ocelli. Described from a single example, 197mm. long, from Fremantle, and the property of the Western Australian Museum. RECORDS OF IV. A. MUSEUM. [83 SYNGNATHUS (YOZIA) TIGRIS, Castelnau. Plate, XI. ; Fig. 2. Syngnathus tigris, Castelnau — Proc. Linn. Soc. N. S. Wales, III , 1879, p. 397 : Id., Macleay, loc. cit., VI., 1881, p. 227; Id. Stead, loc. cii., XXXI., igo6, p. 428; Id., Duncker, Faun. Siidwest-Austr. II., igog, p. 245 D. 24-25. P. 15. C. 8. Rings 17 + 36. Head in the trunk. Head and trunk if in tail. Snout if- if in rest of head. Eye 2f-2-| in snout, and 2 in postorbital portion of head. Snout rugose but not serrated, with a median keel before the eyes which is more or less bifurcate between the eyes. Interorbital .space concave. Head granular, operculum with radiating lines, but without a median keel. Occiput and nape with a sharp, raised keel ; body rings without spines. Trunk deeper than broad, the breast more or less swollen ; depth between the upper and lower lateral edges of the trunk rather less than the length of the snout. Dorsal fin opposite the vent, occupying 2f-3 + 2-2l rings, its base elevated. Median lateral ridge continuous with the lower caudal edge; upper edge extending over i-i| body rings. Lower surface of the tail much broader than the upper. Caudal fin, large, two- thirds as long as the snout. Colours. — Light brown in spirits, each ring with a more or less distinct ocellus above the lateral ridge, and on the sides of the tail ; a dark-edged, semioval pearly spot on the edge of each body segment. Operculum with several oblique dark lines. Body with three darker cross bars, and tail with about seven more. Described from three specimens 265-280111111. long, from Port Jackson, the largest of which is the specimen figured (Reg. No. i. 12073). A single specimen from Fremantle differs only in having the head more rugose, the opercular markings broader, and nine instead of seven cross bands on the tail. According to Castelnau, the upper edge on the tail and the lateral line are continuous, but in all the specimens I have seen, the latter joins the lower edge above the vent. Through the kind- ness of Mr. Stead, I have examined the specimen he recorded from 84 ] RECORDS OF W.A. MUSEUM. the Hawkesbury River, while Mr. Ogilby informs me that accor- ding to his notes, the lateral ridge is always continuous yvith the lower caudal edge. I therefore regard the original description as incorrect in this detail. CYPSELURUS KATOPTRON, Bleeker. Exocoetus katoptron, Bleekev—i^ed. Tydschr. Dierk., III., p. 115 ; W., Gunther, Brit. Mus. Cat. Fish, VI,, 1866, p. 289; Id. Bleeker, Atl, Ichth., VI., 1871, p. 72, pi. CCXLVII , fig. 3. Cypsiluvus katoptron, Jordan and Seale— Bull. U.S. Bur. Fish, XXV., 1906, p. 211, fig. 16. A large specimen from Fremantle differs from Jordan and Seale’s excellent figure only in having a slightly larger eye, and thirteen instead of fourteen dorsal rays, but in both these details it is in agreement with Bleeker’s description. It differs from Gunther’s description of Exoccetus robustus ^ in having the iiiterorbital space concave instead of flat, and in the position of the ventral fin which is inserted farther back. Jordan and Seale consider C. robnstus and C. katoptron identical, notwith- standing that Gunther had Bleeker’s type before him for comparison. PARAPLESIOPS MELEAGRIS, Peters. Plate IX. Plesiops meleagris, Peters — Monatsb. Akad. Wiss. Berlin, 1869 (1870), p. 708, Rnppelia pvolongata, Castelnau — Res. J Fish. Austr. (Viet. Rec. Philad. Exhib.), 1875, p. 29 (not R. proloHgata, Cast., 1873). Paraplesiops meleagris, Boulanger— Brit. Mus. Cat. Fish., 1895, p. 339. A specimen from near Fremantle is in the Western Australian Museum, and another from the same locality was presented to the Australian Museum by Mr. A. Abjornssen. It is figured on plate I. (Reg. No. I. 1 1232). Though the species has not been recognised from Western Australia under this name, there can be no doubt that the second specimen identified by Castelnau as Ruppelia pvolongata, was really P. meleagris. 1 Giinther, loc. cit. RECORDS OF W.A. MUSEUM. [85 BOSTOCKIA HEMIGRAMMA, Ogilby. Plate X; Fig. 3. Bostockiahemigyanntia, Ogilhy — Proc. Linn. Soc. N. S. Wales, XXIV., 1899, p. 168. The collection includes a small specimen of this species from the Helena River, near Perth, and two others from the same locality are in the Australian Museum. The largest of these, 1550101. long, is the one selected for illustration. I have compared it with a co- type in the museum collection, so that, although its proportions differ somewhat from the smaller ones described by Ogilby, I have no doubt of its correct identification. EDELIA, Castelnau. Edelia, Castelnau — Proc, Zool. Soc. Vic., II.. 1873. p. 123 {vittata) ; Id., Ogilby, Proc. Linn. Soc. N. S. Wales, XXIV., 1899, p. 175. As all my specimens, including two examined by Ogilby, differ in some important details from that author’s definition of this genus, I have drawn up the following corrected diagnosis. The items in brackets are as stated by Ogilby. Body oblong, compressed. Scales, large, adherent, finely cili- iated, concentrically striated. Lateral line interrupted below the second dorsal (complete), the tubes few, irregular, simple, extending along the entire exposed surface of the scale. Head moderate, largely scaly, snout and lower jaw naked. Mouth with small oblique cleft; jaws equal, Premaxillaries protractile; maxillary exposed at distal extremity only, naked. Bands of slender villiform teeth in the jaws, a large patch on the vomer, and a few on the anterior part of the palatines (not seen by Ogilby) ; pterygoids and tongue smooth. Nostrils distant, simple. Eyes moderate, lateial, high. Preorbital denticulate. Preopercle entire. Operculum with two spines. Gill-openings wide, the membranes united in front, free from the isthmus; 5-6 branchiostegal.s. Psendobranchiae present. Gill-rakers moderate, few. Two dorsal fins connected at the base, the first with 7-8 spines and longer than the second ; the 86 ] RECORDS OF W.A. MUSEUM. second with i spine and about 9 rays. Anal short, with 3 spines and about 8 rays. Ventrals inserted behind the base of the pec- torals, close together, with a strong spine and 5 rays. Pectorals rounded, with 11-13 rays. Caudal slightly rounded. Vertebrae 13 + 15 (12 + 18). This genus is evidently closely allied to Nannoperca, Gunther. EDELIA VITTATA, Castelnau. Plate X.; Fig. 2. Edelia vittata, Castelnau— Proc. Zool. Soc, Viet, II., 1873, p. 124; Id. Ogilby, Proc, Linn. Soc. N. S. Wales, XXIV., 1899, p. 176. Edelia viridis, Castelnau — loc. cii., p. 125. There are twenty specimens in the Australian Museum, of which two from the Leschenault Inlet were received from Mr. A. Abjornssen. Two more from Donnybrook, in the County of Wellington, were some of those used by Ogilby in drawing up his description, while sixteen others without a definite locality were received from Mr. Albert Gale. GLAUCfOSOMA HEBRAICUM, Richardson. Glaucosoma hebraicum, Richardson — Voy. Ereb. and Terr., Fishes, 1846, p. 27, ■ pi, XVII. ; Id., Saville Kent, Nat, in Austr., 1897, p. 177, pi. XXX. Glaucosoma burgeri, Gunther— Brit. Mus. Cat. Fish, I., 1859, p. 211 (part). Fresh specimens of this species show striking dark longitudinal bands, the most prominent of which follows the lateral line, and is rather wider than the scales it covers. There are two broader ones between it and the back, each covering two rows of scales, while below it are three others which are still broader. A dark bar passes obliquely from the eye to the interopercle. The fins are dusky, but the tips of the anal and dorsal spines and the anterior margins of the fins are whitish, and there are oblique whitish bars on the ends of the caudal lobes. Although 1 have carefully compared my two specimens with Jordan and Thompson’s' excellent description and figure of G. burgeri, Richardson, I am unable to find any satisfactory differences 1 Jordan and Thompson, Proc U S. Nat. Mus., XXXIX., igii, p. 440. RECORDS OF W.A. MUSEUM. [87 between the two species that may not possibly be accounted for by variation with growth, etc. Both my specimens, however, have the fourth dorsal rays longest, so that the fins are angular in form instead of rounded. Minute palatine teeth are present in both, but are very indistinct and easily overlooked in one of them. As sug- gested by Jordan and Thompson, it will be necessary to compare specimens of both species before the characters separating the two can be positively determined. Both specimens were obtained near Fremantle, W. Australia. SILLAGO BOSTOCKII, Castelnau. Plate X., Fig. i. Sillago ciliata vel bostockii, Castelnau — Proc. Zool. Soc. Viet., II., 1873, p. 133. Sillago bostockii, McCulloch — “ Endeavour Report,” fishes (in press), pp. 60, 63. D. XL, 21-22; A. 19-22; P. 15-16; V. 1 ., 5; C. 17; 1 . lat. 69-74 i 6+14. Head 3^-3!, depth 5-5^ in the length to the hypural. Eye 5-51, snout 2f, pectoral if-if in the head. Interorbital width f-f the eye. Body elongate, compressed, the dorsal profile rather more arched than the ventral. Caudal peduncle much compressed, its depth almost equal to the postorbital portion of the head. Eye large, nearer the end of the operculum than the tip of the snout. Interorbital space flat. Preoperculum crenulate, a broad flat spine on the operculum. Mouth small, oblique, the maxillary a little longer than half the snout. Nostrils close together, near the eye, the anterior with a skinny lobe. A broad band of villiform teeth on each jaw, the outer ones of the upper jaw somewhat enlarged ; a broad horseshoe shaped band on the vomer. Gill-rakers short and thick, tubercular below, nine on the lower limb of the first arch. Scales finely ctenoid, extending forward to the nostrils on the upper surface of the head ; arranged in four rows on the cheeks. Snout and lower part of the head bare. Basal portion of the caudal fin densely scaly, while rows of small scales are present behind each ray of the other fins. Lateral line a little arched anteriorly, thence straight to the caudal peduncle, and continued on to the fin to the end of the middle rays. 88 ] RECORDS OF V/.A. MUSEUM. Dorsal fins separate or united by a low membrane ; the first is inserted well behind the ventrals, the second just in advance of the vent and terminating far behind the anal. Pectoral and ventral pointed, the outer ray of the latter sometimes produced. Caudal emarginate. Colour. — Sandy yellow, the upper parts closely speckled with minute grey dots. Both dorsal fins have from three to five longi- tudinal rows of large grey spots. Well preserved specimens have a dark (silvery ?) band from above the base of the pectoral to the caudal peduncle. No dark mark at the base of the pectoral. This is apparently the common whiting near Fremantle; a large number of specimens being included in the Western Australian Museum collection, while an excellent series was secured for the Australian Museum by Mr. Abjornssen. Specimen selected for illustration registered I. 11334. THERAPON CAUDAVITTATUS, Richardson. Datnta caudaviltata, Richardson — Voy. Ereb. and Terr., Fishes, 1848, p. 24, pi. XVIII , fig. 3-5. Therapon caudovitintus , Giinther — Brit. Mus. Cat. Fish. I., 1859, p. 284: Id., All. and Mad., Proc. Finn. Soc. N. S. Wales, I., 1877, p. 270 ; Id., Mad. loc. cit., II., 1878, p. 348; Id., Castelnau, loc. cit.. III., 1878, pp. 42 and 47 ; Id.. Kiunr, Sitzb. Ak. Wiss. Wien., LXXX., I., 1879, p. 350 ; Id,, Waite, Kec. Austr Mus., III., 1900, p. 210. Therapon caudovitlalus ? vel bostochii, Castelnau. — Proc. Zool. Soc. Viet., II., 1873, p. 128. I have examined nine specimens of this fish from Fremantle, W. Australia, and two from Murray Island, Torres Strait, but in none are the sub- and interoperculum serrated as described by Richardson. There are also 25 instead of 20 transverse series of scales, so that in both these details they agree better with his figure than his description. Castelnau proposed the name hostockii for Fremantle specimens of caudavittatus, which differed from the description given by Gunther in having the dorsal fin notched, and the last spines much shorter than the rays, This, however, is the normal form, and it is doubtful if Gunther has correctly described his specimens. RECORDS OF W.A. MUSEUM. [89 SPARUS SAKBA, Forskal. Pagius sarba (Forskal), Ogilby— Ed. Fish. N.S. Wales, 1893, P- 5 °' P^' XIV. Chrysophrys sarba. Stead — Ed. Fish, N. S. Wales, 1908, p. 78, pi. XLVII. Eight specimens are in the collection, from the neighbourhood of Fremantle, which agree in every way with others from Port Jackson. This species does not appear to have been recognised from the western coast of Australia. SPARUS AUSTRALIS, Giinther. Chrysophrys auslralis, Gunther — Brit. Mus. Cat. Fish. I., 18591 P- 494 ' Stead, Ed. Fish, N. S. Wales, 1908, p. 77, pi. XLVI. I am unable to separate four specimens, from Fremantle, from this species. It has been recorded from the Harvey River, Western Australia, by Gunther. PLATAX TEIRA, Forskal. Platax teria (Forskal), Jordan and Fowler — Proc. U. S. Nat. Mus., XXV., 1902, p. 256. A very small specimen, one and a half inches long, from Fre- mantle, is of interest as being apparently the first of the species recognised from Western Australia. LEPIDAPLOIS VULPINUS, Richardson. Cossyphiis vidpinns, Richardson — Proc. Zool. Soc., 1850, p. 71. Harpe vulpina, Waite— Rec. Austr. Mus., IV., 1902, p. 269, pi. XLIII. and VI., 1905, p. 70. Trochocopus rtifus, Macleay — Proc. Einn. Soc. N. S. Wales, III., 1878, p 35, pi. V., fig. 3. , , Cossyphus frenchii, Klunzinger — Sitzb. Akad. Wiss. Wien, LXXX., 1., 1879, p. 400; Id., Macleay, Proc. Linn. Soc. N. S. Wales, IX,, 1884, p. 46. Cossyphus aurifer, De Vis — Proc. Roy. Soc. Qld., I., 1884, p. 146. The collection includes a specimen from Abrolhos Island, which agrees very well with W^aite’s figure of Harpe vulpina. I have compared it with the types of Trochocoptis nifits, Macleay, with which it is identical. According to Mcleay’s description there are 45 scales on the lateral line, but in both his specimens there are only 36. This detail is referred to by Klunzinger, and as Trocho- copus has 45 or more, he placed the species in Cossyphus. Further, 90 ] RECORDS OF W.A. MUSEUM. there being already a Cossyphus ntfus, and as he did not recognise its identity with tlie earlier C. vulpinus, he changed the name to C. frencliii. Mr. Ogilby has kindly examined the type specimen in the Queensland Museum of Cossyphus aurifer, De Vis, for me, and writes that he considers it identical with the species figured by Waite. PSEUDOLABRUS PARILUS, Richardson. Plate. XII. Tautoga pavila, Richardson — Proc. Zooi. Soc., 1850, p. 70. Labrichihys parila, Castelnau — Proc. Zool. Soc. Viet., II., 1S73, p. 137. D. IX., II. ; A. III., 10 ; P. 13 ; V. L, 5 ; C. 12-13 ; 1 - lat. 26 ; 1 . tr. 3-4 -1-9. Height 3^ in the length to the hypural, and equal to the length of the head including the opercular flap. Eye 5, snout 3-3!^, caudal peduncle 1^% in the liead. Body moderately elongate, compressed, covered with large scales which extend on to the nape and the caudal fin, but not over the bases of the dorsal and anal. A single series of small and im- perfect scales from behind the eye to the cheek, and other large and irregular ones covering the operculum; head otherwise naked and closely pitted with minute pores. Preorbital narrower than the eye. A pair of strong canines in front of each jaw, those of the upper separated; sides with a single series of smaller canine-like teeth decreasing in size backwards, and a second inner series of very small ones anteriorly. Posterior canines present. Nostrils close together near the supero-anterior angle of the eye ; the anterior tubular, posterior simple. .Lateral line following the curve of the back over 20 scales, then bending downward, two rows to the middle of the caudal peduncle ; the ramifications of the tubules cover all the exposed portions of the scales except the extreme edges. Dorsal fin commencing above the hinder half of the operculum. Spines increasing in length backwards, the last 2^-3 in the head, and each topped by a prolongation of the membrane. Soft portion of the pin angular behind, the anterior rays longer than the pos- terior ; 2j in the head. Anal similar to, and terminating a little in advance of the dorsal. Pectoral i| in the head, the upper rays longest, margin rounded. Ventral pointed, second ray not quite reaching to the vent. Caudal rounded. RECORDS OR W.A. MUSEUM. [91 Colour. — Greenish, spotted with brown, the spots tending to form indistinct bands in one specimen but not in the other. Brown lines radiate from the eye and on to the operciilum. Broad darker markings enclosing light interspaces are distinct in one specimen on the lower parts of the head. Dorsal and anal with darker and lighter spots on the rays, and there is a larger spot between the first and second spines of the former. Described from two specimens, 200 and 225mm. long, in the Australian Museum (I. 11459 and 11461), received from Mr. A. Abjornssen who collected them at Fremantle. Castelnaii recorded this species from Port Jackson,* but it has not been again recognised from eastern Australia by any other author. I suggest that his specimen was merely one of the varia- tions of P. gymnogenys, Gunther, as I have a local example of that species with which his colour-notes agree very well. Having compared the specimens identified by Waite'* as P. ruber, Castelnau, with the two described above, 1 tliiuk it pos- sible that the former will prove to be either an older or sexual form ol P. piirilus. Though the differences between Waite’s figure and my own appear very great, some of the specimens nevertheless exhibit certain characters which are intemediate between the two extremes, while Castelnau’s notes on the colour of Western Aus- tralian examples of P. parilus agree as well with his own as with Richardson’s species. PSEUDOLABRUS BOSTOCKII, Castelnau. Plate XI ; Fig. i. Labrichthys tetrica, Giinther — Brit. Mas. Cat. Fish, IV., 1862, p. 116 (part). Labrichthys hostockii, Castelnau — Proc. Zool. Soc. Viet., IF, t873, p, 137; Id. Macleay, Proc. Linn. Soc. N. S. Wales, VI,, i88i, p. 85. Labrichthys hiserialis, Klunzinger — Sitzb. .Vkad. Wiss. Wien, LXXX. I,, 1879, p. 402. Pseudolabnis tetneus, Waite — Rec Austr. Mus., VI., 1905, p. 70 [nee Richardson) D. IX., ii; A. III., 10; P. 12; V. I.,5; C. 13-14; 1 . lat. 25-27 ; I. tr. 4-1-8. Height of body 3-3I in the length to the hypural, a little 1 Castelnau, Proc. Linn. Soc. N. S. Wales, III., 1879, p. 389. a Waite, Rec. Austr, Mus. IV., 1902, p. 185, pi. XXVIII. 92 ] RECORDS OF W.A. MUSEUM. greater than the length of the head without the opercular flap. Eye 4^-5, snout 3, caudal peduncle 2J in the head. Interorbital width I5 in the snout. Body moderately elongate, compressed, covered with large scales which extend forwards on to the nape, and on to the bases of ihe dorsal, anal, and caudal fins. Two oblique rows of small scales from behind the eye to the cheek, and other larger and irregular ones covering the operculum ; head otherwise naked. Preorbital either a little broader than or narrower than the eye. A pair of strong canines in front of each jaw, those of the upper being widely separated; sides with a single series of smaller canine-like teeth decreasing in size backward, and usually a second inner series of very small ones anteriorly. Posterior canines present, sometimes double. Nostrils placed close together near the supero-anterior angle of the eye ; the anterior tubular, posterior slit-like. Lateral line following the curve of the back over 19-20 scales, then bending downward two rows to the middle of the caudal peduncle ; anteriorly the tubes are much branched, simpler posteriorly. Dorsal fin originating over the hinder half of the operculum. Spines low, each topped by a prolongation of the membrane ; rays subequal, soft portion of the fin angular posteriorly. Anal similar to the dorsal. Caudal truncate, the tips usually a little produced. Upper rays of pectoral longest, the lower half of the fin rounded. Ventrals pointed, not reaching to the vent. Colour. — Red, darker aliove, each scale with a large carmine spot. A yellow band extends from above the base of of the pec- toral to the middle of the caudal peduncle. Dorsal black basally, then clear orange, and margined with a narrow violet line. Anal blood red with a broad violet edge, and with or without indications of a darker median band. Caudal orange with a darker edge. Pectorals and ventrals pink, the former with a black basal band. After long preservation almost all traces of the colour markings disappear, leaving only the darker fin markings. Described from six specimens, i6o-2oomm. long, one of which is from near Albany, another from Mandurah, and four from Fremantle. 1 am indebted to Mr. Abjornssen for beautifully pre- served examples of this species from which the accompanying figure has been prepared. RECOHDS OF W.A. MUSEUM. [93 I follow Klunzinger in regarding his P. hiserialis distinct from P. tetricus, Richardson, but consider that it is identical with P. bosfockii. Waite united the latter with P. tetricus, but they appear to differ in the arrangement of the scales on the cheeks and in the form of their fins. MUCOGOBIUS, gen. nov. Body oblong, compressed behind. Head a little compressed, scaleless, but with many horizontal and vertical raised mucous ridges, which also extend on to the body; no true barbels. Snout rounded, lower jaw the longer ; mouth oblique. Eyes large, close together. Opercles unarmed. A band of simple villiform teeth in each jaw, the outer ones somewhat enlarged; vomer and palatines toothless. Tongue rounded. Isthmus broad. Scales moderate, largest posteriorly, cycloid. Dorsal with six spines and about ten rays. Anal similar to the soft dorsal. Ventrals I. 5, united, not adnate to the belly. Pectorals pointed, without free silk-like rays. Caudal rather lanceolate, produced. Type. — Gohius mucosus, Gunther. MUCOGOBIUS MUCOSUS, Giinther. Gobius mucosus, (Gunther), Waite — Rec. Austr. Mus., VI., 1906, p. 200. The collection includes three from Fremantle, while another was collected by Mr. Abjornssen near Albany. I am unable to associate this species with any genus known to me, and therefore propose Mucogohius for it as above. SCORPAENA, Linnaeus. Scorpaeiui, Linnaeus— Syst, Nat., loth Ed., 1758, p. 266 (porous) ; Id., Jordan and Starks, Proc. U.S. Nat, Mus. XXVII,, 1904, p. 131. Sebastapistes, Gill, in Streets — Bull. U.S. Nat. Mus., No, 7, 1877, p. 62 (strongia) i Id., Jordan and Evermann, Bull, U.S. Fish., Comm., XXIII,, pt. i, 1905, p, 455. The genus Sebastapistes, is apparently distinguished from Scorpaena only by the armature of the preorbital and its smaller size. Some large Australian species have strong recurved spines on the 94l RECORDS OF W.A. MUSEUM. preorbital, and being unable to satisfactorily divide up those I have examined into the two genera, I prefer to regard Gill’s genus as a synonym of Scorpaena. The following is a key to the Australian species available to me. a. 50-55 rows of scales just below the lateral line. A more or less prominent median keel on the anterior portion of the interorbital space. cardinalis. aa. 45 or fewer rows of scales. b. Transverse hollow behind the eyes, distinct but shallow. c. Two prominent interorbital ridges ending in spines. Third dorsal spine generally longest. cruenta. cc. Inteiorbital ridges low, without spines. Fifth dorsal spine generally longest. d. Head and body with numerous tentacles. hynoensis. dd. Flead and body with but few tentacles. var. laotale. bb. Transverse hollow very deep. Interorbital ridges almost obsolete. sumpttwsa. SCORPAENA CARDINALIS, Richardson. Scorpiuua cardinalis, Richardson— Ann. Mag. Nat. Hist., IX., :842, p. 212 ; Id., Gunther, Brit. Mus. Cat. Fish., II., i860, p. 116. Scorpaena jacksoniensis, Steindachner — Sitzb. Ak. Wiss. Wien., LIII., I., 1866, p. 438, pi. III., fig. 2, 2a. Scorpaena crnenta, Ogilby — part, Ed Fi-h, N S. Wales, 1S93, p. 63, pi. XX. ; hi., Waite, Mem. Austr. Mus., IV., 1899, p. 99; Id., Stead, Ed. Fish. N.S. Wales, 1908, p 108, pi. LXXV. (not S. crnenta, Richardson). This .species, and not S. crnenta, Richardson, is apparently the common Red Rock Cod of the Sydney fishermen. I have exam- ined the specimens identified by Ogilby, Waite, and Stead as crnenta, and regard almost all of them as being cardinalis ; only two small ones from Port Jackson being the former species. Mr. Stead in- forms me that there is but one common species in the Sydney Markets, of which his specimens are representative, so that S. crnenta will probably prove to be a rare species here. Besides the two Port Jackson specimens, I hav'e examined several others from Tasmania, and find that they differ from cardinalis in having much larger scales, and in having the interorbital ridges ending in acute spines. Gunther' has united S', jacksoniensts, Steindachner, with S’, hynoensis, Richardson, but Klunzinger'^ has shown that this is 1 Gunther, Zool. Rec , 1866 (1867), p. 143. Klunzinger, Sitzb. Ak. Wiss. Wien, I.XXX. I., 1879. p 366. RECORDS OF MUSEUM. [95 incorrect. It differs in the form and disposition of its cephalic spines, smaller scales, general proportions, and colouration. I can find no difference between it and 5 . cardinalis. SCORPAENA SUM FTUOSA, Castelnau. Plate XIII. Scorpaena siimptuosa, Castelnau — Res. Fish., Austr. (Viet. Offic. Rec. Philad. Exhib.), 1875, p. 17 ; Id , Macleay, Proc. Linn. Sog., N. S. Wales. V., 1881, p. 432, D. XI., I. 9-10 ; A. III., 5 ; V. I. 5 ; P. r6 ; C. 13. Depth 2^-2|, head 2^ in the length to the hypural. Eye 4-4|-, caudal peduncle 3I- in the head. Dorsal profile highest at the base of the third dorsal spine, thence descending rapidly to the tail. Plead armed with strong spines and largely covered with glandular skin, only a few scales being present on the end of the operculum. Interorbital space deeply concave with a very low median ridge anteriorly, and defined posteriorly by a raised sinuous bone. A deep transverse hollow behind the eyes divided into four parts by three longitudinal ridges, the two exterior of which have each a small spine in front. Eye with a broad spine anteriorly, and two on the upper margin. Nasal spines simple or bifurcate. Preorbital with a rosette of sinuous bony ridges. A ridge with three or four spines extends across the cheek, and there are two more on the preopercular margin on the same line; below these the margin is armed with four more points. Superior opercular spine inclined upwards, the lower extending forwards as a prominent curved ridge. There is a small bifurcate spine behind the eye, and two larger ones with broad bases between it and the upper opercular margin. Nuchals large, each with two points; a very small spine between them and the operculum. Maxillary extending beyond the eye, and two-thirds as wide as it posteriorly. Bands of villiform teeth on the jaws, vomer, and pala- tines. Gill-rakers short, thick, and spiny. Scales large, cycloid, extending forward to just in front of the dorsal fin; six or seven between the lateral line and the twelfth dorsal spine, and about fifteen more to the vent. The lateral line is formed of about twenty-one tubes, and there are forty-four rows of scales directly below it. Skinny lobes are distributed at intervals all over the body, but are most numerous on the back and the lateral line. 96] RECORDS OF W.A. MUSEUM. First dorsal high, the third spine the longest and about two- thirds as long as the head ; eleventh spine very short, one third as long as the twelfth. Anterior rays of the soft dorsal sub-equal, the margin rounded posteriorly. Second anal spine longest, very strong and laterally grooved ; soft portion of the fin rounded. Pectoral reaching to, or not so far as the origin of the anal ; it has nine or ten .simple lower rays. Ventrals rounded, reaching to, or almost to the vent. Caudal rounded. Colour. — Yellowish or reddish with darker marblings. Lower parts of the head and body with numerous irregular dark spots. Spinous dorsal marbled with reddish brown, and with or without a large dark blotch on the hinder part. Soft dorsal, caudal, and anal with red and brown spots forming irregular rows. Pectorals also spotted, ventrals plain. Described from two specimens 310mm. and 300mm. long, the first from Fremantle, and the property of the Western Australian Museum, and the second from Albany, and in the collection of the Australian Museum. A third is also in the Australian Museum from Houtmans Abrolhos. This species is allied to 5. cardinalis, Richardson, but is at once distinguished by having only forty-four instead of fifty-five rows of scales below the lateral line, and in lacking the high median keel on the anterior part of the interorbital space. According to Castelnau his specimen had only ten spines in the first dorsal, but as this is an unusual number in the genus, and as my specimens agree in every other detail, I have no doubt that they are really S. sutnpiuosa. SCORPAENA' BYNOENSIS, Richardson. Scorpaena hyiioensis, Richardson — Voy. Ereb. and Terr., 184 ■, p. 22, pi. XIV., fig. 3-4 ; Id., Ktunzinger — Sitzb. Ak. Wiss Wien., l.XXX. I., 1879, p 366 (synonymy). Sebaslaptsles laotale, Jordan and Seale — Bull, U.S. Fish, Bur , XXV., igoC, p. 376, fig. 72 (variety). I have very carefully compared thirty-two examples of this species from various localities and find that they vary considerably in the development of the tentacles and cirri on the head and body. In one from Dunk Island, Queensland, the orbital tentacles are nearly twice as long as the eye, and other large ones are present on RECORDS OF IV. A. MUSEUM. [97 the spines and margins of the bones; in others from Murray Island, Torres Strait, all but the nasal tentacles are wanting. The first represents the form named hynoensis, while the others agree with laotalc, of which I have examined a co-type, but as my series exhibits every stage between the two, the latter must be regarded merely as a variety of the former. Of the thirty-two specimens, one has thirteen spines and nine dorsal rays, another eleven spines and ten rays, while all the rest have twelve spines and ten rays. The length of the posterior spine is variable, while the colour may be anything from ashy grey to brilliant brown and white, though the characteristic colour pattern is always more or less maintained. My specimens come from Dunk Island and Green Island near Cairns, Queensland ; Murray Island, Torres Strait ; Mapoon, Gulf of Carpentaria; Port Darwin and Western Australia. I regard the New South Wales and New Zealand records of this species as very probably incorrect since it is an inhabitant of coral reefs, etc., and is doubtless confined to the tropics. Gunther’s association of .S', jackscniensis, Steindachner, with bynoeusis seems to have been the cause of its first inclusion in the New South Wales lists, but this is shown to be incorrect (see ante). Ogiiby included it in his Edible Fishes of N. S. Wales,' but as there are no local specimens in the collection of the Museum, and as Mr. Stead in- forms me that he has not seen any specimens in the markets, I think its occurrence here needs verification. S. bellicosa, Castelnau,* from Nicol Bay, Western Australia, and Queensland is apparently very similar to .s. hynoensis, but is described as having prominent interorbital ridges, whereas they are low in Richardson’s species. Castelnau’s specimens may have been dried, as were many others in his collection, in which case the flesh shrinking from the bones would make the ridges appear more prom- inent, so that this difference is probably of little importance. 1 Ogiiby, Edible Fish, N. S. Wales, 1893, p. 65. 2 Castelnau, Res. Fish., Austr. (Viet. Offic. Rec. Philad. Exhib.), 1875, p. 17. 98j RECORDS OF W.A. MUSEUM. A VISIT TO BERNIER AND DORRE ISLANDS During August and September, igio. BY OTTO LIPFERT Plates XIV. to XVI. The best time to visit Bernier and Dorre Islands is immediately after the rainy season, when the wild flowers are in full bloom. Parts of Dorre Island then resemble a brilliant flower garden ; however, the blooms do not last long, for during October the sun scorches everything except the salt bush and spinifex. There is little animal life ; even the sea birds are by no means abundant, and as the aborigines, for whose benefit these islands have been reserved, are to some extent dependant for food on the result of their success in hunting, the wallabies and bandicoots will soon become extinct. The Wallabies, of which three species are to be found, are not at all numerous. They are Lagostrophus fasciatus, the Banded Wallaby; Lagorchestes hirsutus, the Hare Wallaby, of which there are two. sub-species, hemievi and dorreae ; and Bettongia lesuemi, commonly called Lesueur’s Rat- Kangaroo.” The striped Bandicoot, Perameles hougainvillei, was reported some years ago by Mr. Shortridge as extinct. Fortunately this is not the case, for I was able to get twelve good specimens. The Mouse, Psettdomys (Gycmys) alhocinereus squalorttm, is plenti- ful. It has a beautiful soft whitish-grey fur, white feet and tail. The only other mammals were Bats, but as it was so early in the summer I only saw two or three, of which, thanks to Mr. Sheriefs I was able to obtain one Nyctinomus planiceps, Peters. The nesting season of the eagles and ospreys was well advanced. On Dorre Island I located three nests of Haliaetus leucogaster, the white-bellied Sea Eagle. One contained eggs, and the other two fledgelings ; while there were ten or more nests of Pandion leuco- ctphalus, the white-headed Osprey. Owing to the rocky nature RECORDS OF W.A. MUSEUM. [99 of the country, the nests are difficult of access. One sea eagle’s nest (Plate xiv., figs, i and 2) found on the east coast contained two eggs. It was a very bulky structure, the diameter on the top being 183 c.m., at base 240, height 52 ; the mould was only 65 wide by 12 deep. I'o reach the nest it was necessary to make a detour of at least a quarter of a mile to find a spot where one could climb down to the shore, about 40 feet below. Both eggs were slightly incu- bated. Two other nests of this eagle which I found on August 27th, contained young birds at least a fortnight old. On August 20th, during an excursion southwards, I found eight nests of the Osprey, of which several were old and deserted. It is more plenti- ful than the Sea-eagle, for five of these fine birds were in sight at once. A nest visited on August 7th was on a projecting cliff, and contained three young, two about a fortnight old, and the other about a week. It measured across the top 141cm., base i6ocm., mould 59cm., depth of same 6cm., height of nest 88cm. Both parent birds and their young were secured (Plate xv.) A remarkable find on this island was the nest with two young of Uroaetus audax (Plate xvi.), for owing to the total absence of trees, it was built on the side of a hill to secure a good outlook. It was situated about 15 yards from the top of the hill, and 35 from the foot. Across the top the nest measured 196 cm., at bottom 244 cm. The height on the upper side, top, was 53cm., and on the lower, 260cm. Besides eagles, there were large colonies of Cormorants, Sea- gulls and Terns ; these had quite finished their nesting. The breeding season of the smaller birds, Malums, Sencornis, Megaliirus, etc., had finished about the end of May. They were always to be found in family groups of two old and three or four young birds. These were not common, for one often tramped mile after mile without seeing a single bird. Bernier Island has the same desolate appearance as Dorre, perhaps even worse, as there are no flowers and the scrub is thicker. To travel a mile and a half per hour may be considered good work, and even that is very tiring. The animal life is nearly the same as on Dorre Island, with the exception of Megaltmis, which is not to be found here, and the Eagles and Ospreys are not so numerous as on Dorre Island. I lOOj RECORDS OF W.A. MUSEUM. found only one nest of Haliaetus lencogaster and four of Pandion Uucocephalus. On the north end Knock’s Island, numbers of Terns, both Sterna bergii and S. nereis had been breeding;. The Cormorants had no breeding place. As will be seen from the list appended, I secured a very good collection, and I desire to place on record my sincere thanks to Dr. Lovegrove and to Mr. Sheriefs for the valuable assistance they gave me. LIST OF BIRDS OBSERVED ON DORRE AND BERNIER ISLANDS. BY O. LIPFERT, September, 1910. Puffinns chlororhynchus B. Seen flying around steamer on returning to Carnarvon. Sterna bergii D.B. br. (young, fully fledged on 17th Sept.) Sterna nereis D.B. br. do. Larus novae-hollandiae D.B. br. Gabianus pacificus D.B. br. Arenaria inter pres B. Haematopns longirostris D.B. br. Haematopus fuliginosus D.B. br. Charadrius doniinicns B. Peltohyas australis B. Numenius cyanopus B. Visitor (seen by Dr. Love- grove). Limonites [Phobia) ruficollis ... B. Eupodotis australis B. Visitor only. Demigretta sacra D.B. br. Phalacrocorax hypoleiicus D.B. br. Large colony on Dorre Island. Circus assiinilis ... D. Only seen once near S. end. Uroaetus (Aquilla) audax D. br. Haliaetus leucogaster D.B. br. Cerchneis cenckroides D.B. Pandion Uucocephalus D.B. br. Podargus sp ? Cypselns pacificus Hirundo neoxeua Megalnnis siriatus Sericornis iiianilata Malurns beniieri Zosterops gonldi . . . Ptilotis soiioni Anthus australis ... RECORDS OF W.A. MUSEUM. [loi B. Only seen once on W. coast. B. Visitor during April, May and June. D.B. br. D. D.B. D.B. B. Visitor. D.B. D.B. D, — Dorre Island. B.— Bernier Island. br.— Breeding in August and September. 102 ] RECORDS OF W.A. MUSEUM. MUSEUM NOTES. THE TRUSTEES ACT (i Geo. V., 1911, No. 27). The Museum on its establishment in Perth in 1889, was directly under the control of the Hon. the Commissioner of Crown Lands ; in 1894 O'’ appointment of a Minister of Mines, it was trans- ferred to his charge; in 1895 it was placed under the control of a Committee of eight gentlemen nominated by the Gov^ernment. On the i6th February, 1911, assent was given to: — “An Act to provide for the appointment and incorporation of Trustees for the Public Library of Western Australia, and the Western Aus- tralian Museum and Art Gallery, and for purposes consequent on and incidental to that object.” This Act came into operation on the 20th November, 1911, by Proclamation of His Excellency the Governor, Sir Gerald Strickland, K.C.M.G. (Gov. Gazette No. 63, loth November, 1911). At the same time the Governor in Executive Council nominated twelve Trustees, of whom seven, the Hon. Henry Briggs, M.L.C. ; M. F. A. Canning, Esq. ; Hon. Sir J. Winthrop Hackett, Kt., M.L.C., etc.; Dr. H. F. Harvey; Dr. H. T. Kelsall ; The Hon. Mr. Justice McMillan ; and the Rt. Rev. Dr. Riley had been members of the Committee ; the remaining five are F. W. Burrows, Esq.; Walter Dwyer, Esq., M.L.A. ; R. S. Haynes, Esq., K.C. ; Sir Walter James, Kt., K.C. ; and H. S. King, Esq. On the 24th November, these Trustees met for the co-optation of two Trustees under Clause IV, of the Act, when the Hon. Sir Edward A. Stone, K.C.M.G., and W. Somerville, Esq., were nom- inated. On the 27th November, the Hon. Sir J. Winthrop Hackett, Kt., M.L.C., M.A., LL.D., Officier de VAcademie Ffangaise, was elected President of the Board. C. G. Morris, Esq., of the Lands Department has been appointed a Trustee in place of the late Mr. M. F. A. Canning. RECORDS OF W.A. MVSEUM. [103 RECENT ADDITIONS TO THE COLLECTIONS. Sculpture. — A cast of the Monasterboice Cross received in exchange from the National Museum, Dublin. The Laocoon pre- sented by the Hon. Sir J. Winthrop Hackett, and Augustus Caesar purchased. Paintings. — -An Australian Landscape “Summer,” Gumer- acha, S.A., igio-ii, by Will Ashton of Adelaide, purchased by the Committee aided by public subscription. “ Love under the Rose,” by Franceso Vinea of Florence, purchased on the recommendation of Sir J. D. Linton, P.R.I. ; “ Idleness,” by Rupert C. W. Bunny ; this and four water colours, “ The Nursery,” by Sir E. Waterlow, R.A., P.R.W.S. ; “ Old Harbour, Genoa,” by J. McWhirter, R.A,; “ A Windy Day,” by J. W. Herald, and “ Mauve and Greys,” by Francis E. James, A.R.W.S., were presented by the President, the Hon. Sir J. Winthrop Hackett, M.L.C., etc. Antiquities.— Repliquas of the Vaphio Bull Cups, of Bronze Daggers, and gold tiara and plaques of Aegean (Mycenean) origin. Zoology. — A specimen of Notovyctes typhlops, the Marsupial Mole from Wollal on the go-Mile Beach, on the North-west Coast, was presented by Mr. S. J. Pryon ; this is the second obtained in this State, the first being found at Joanna Springs, about midway between Wollal and the South Australian border. Skins of a kangaroo collected by Mr. J. P. Rogers in the McClintock Range, South-east Kimberley, in igog, were sent to Mr. Oldfield Thomas, who in the “ Annals and Magazine of Natural History,” June, igii, described them as a new sub-species of Wallaroo under the name of Macropus robustiis bracteator. The fur is longer and darker than that of M. r. ivoodwardi from Grant Range, West Kimberley. Mr. Rogers at the same time obtained an example of Phascogale penicillata pirata, not hitherto known to occur in this State, 104 ] RECORDS OF W.A. MUSEUM. Reptilia — Descriptions of some new species will be published in Part III. Free Popular Lectures on Science and Art. — The Eighth Series will be opened on Friday the 26th April next. These Lectures are attracting larger audiences every year, and it is hoped that the Government will provide a Lecture-room in the near future, for at present the Art Gallery has to be used for that purpose. Sca/e oT Centimefres. Nototheviiim, sp. Fig. 8 — Working surface of left upper molar. Fig. 9 — Outer aspect of same tooth. Fig. lo — Working surface of left upper premolar. Plate VI. j I i •; ! 1 ) } C .Pihrison. def. Thylacoleo, sp. Fig. I — Outer aspect of left upper incisor. Fig- 2— Inner aspect of same. Fig. 3 — Outer aspect of left upper functional premolar. a Unworn enamel. b Worn enamel. c Unworn dentine. d Exposed and worn dentine. Plate VII. i t '• • -IV.. . ■t ¥ \ V .'. ■ ■ V f - . , t ■ iv '. -: .V’- ; i ” Scd/e of Centimetres S PMcrfson de/ Thylacoleo, sp. Fig. 4 — Inner aspect of right lower incisor showing the polished and scratched enamel at b. Fig. 5 — Outer aspect of left lower incisor. Fig. 6— Inner aspect of same showing the large area of worn enamel at b. Fig. 7 — Upper view of the same tooth li Unworn enamel. c Unworn dentine. b Worn enamel. d Exposed and worn dentine. Plate VIII. . ■■ '•'"fi''' 1 v' ■ V, Paraplesiops niehagris, Peters. Plate IX. > M X! O 'o U o Pseudolabrus parilns, Richardson. Plate XII. >- i \ R. McCulloch, rf«'.] \ ] 1 Fig. I -Nest and Fg^gs of the White-bellied Sea Eagle. Fig. 2 — Distant view of the same Nest ; its position is indicated by the white cross. Haliaetus leucogaster. Plate XIV. n: Panihon leucocephahis. Nest of the White-headed Osprey. Plate XV. r; OF THE WESTERN AUSTRALIAN Museum and Art Gallery Edited by the Director, DEEMED H. WOODWAED, E.G.S,, C.M.Z.S. .^OLXJJVLE Z. ZZZ. CONTENTS. PAGE W. A. Echinoderms, by W. B. Alexander, M.A. - . - 105 The Crinoids collected by the Endeavour between Fremantle and Geraldton, by Austin Hobart Clark - - - 113 The Echinoderms of the W. A. Museum, by Hubert Lyman Clark, Ph.D. - - - 132 On a Collection of Reptiles and Batrachians from W. A., by Dene B. Fry - -- -- -- -- 174 Notes on some W. A. Fish, by Allan R. McCulloch - - - 211 Revision of the Freshwater Crayfishes of South Western Aus- tralia, by Allan R, McCulloch - 228 On a Specimen of Regalecus glesni, Ascan., obtained in W. A., by W. B. Alexander, M.A. ------- 236 A Brachiopod New to Australian Waters, by W, B. Alexander, • M.A. - - 239 A New Species of Embiid from W. A., by Dr. K. Friederichs - 241 The Mammoth Cave (Continued), by L. Glauert, F.G.S. - 244 Further Important Discoveries in the Mammoth Cave, by the Editor ---- - 252 Printed by Order of the Trustees BY The Paragon Printing and Publishing Co. Hay Street, Perth. 1914. 4 1 i I s A i . ^ ' 1 . > WESTERN AUSTRAIJAN ECHINODERMS BY W. B. ALEXANDER, M.A. In the two papers which follow (Crinoids, by A. H. Clark, and Echinoderms, by H. L. Clark) will be found descriptions of most of the Echinoderms in the Western Australian Museum. To prevent any misunderstanding it seems necessary to mention the circum- stances under which the papers were written. During June, 1912, the F.I.S. Endeavour was engaged in trawling operations off the Western Australian coast between Fremantle and Geraldton, and the Director obtained permission from the Federal authorities for the writer to accompany her on two of her trips to obtain specimens for this Museum. Perhaps the most striking feature of the hauls made by the Endeavour was the very large number of Crinoids which were brought up on many occasions. In number of individuals they surpassed all the other groups of Echinoderms put together. In view of the fact that Mr. Austin H. Clark, of Washington, had recently published a paper on the Crinoids of Australia, the collec- tion was sent to him for identification, and his report on this collec- tion constitutes the first of the papers which follow. In the meantime, Mr. H. Lyman Clark, of Harvard, had offered to determine the Echinoderms already in the Museum, and the remainder of the Echinoderms including all those obtained by the Endeavour, except the Crinoids, were therefore sent to him. A few species which had been named by the authorities of the Australian Museum in Sydney, and a few of the Endeavour speci- mens of the same species, together with some Crinoids which had previously been identified by Mr. A. H. Clark, along with those collected by the Hamburg Expedition, were not sent away. In the introduction to his paper dealing with this material Mr. Clark drew a number of conclusions as to the relative proportions RECORDS OF W.A. MUSEUM. io6] in which the different orders were represented in Western Austra- lian waters. As, however, the numbers he used were vitiated by the assumption that he had the whole of the Museum collection before him, with his permission the numbers have been removed from the paper as written by him and the true totals set out in this place. It seems better to consider the collection in two portions. 'I'he fir.st of these has been slowly accumulating for a number of years and is composed of specimens found on the shore in various parts of the State and on the leefs of the Abrolhos Islands. Excluding specimens too imperfect for complete identification, 46 species are represented distributed among the different classes thus; Crinoids 7, Asteroids 16, Ophiurans 2, Echini 15, Holothurians 6. The second portion of the collection consists of the specimens trawled by the Endeavour in depths varying from ig to 120 fathoms, on the slope of the Continental shelf ; most of these come from a depth of more than 50 fathoms. It contains 36 species distributed thus : Crinoids 12, Asteroids 7, Ophiurans 10, Echini 7. Only 7 species are represented in both portions of the collection. It is doubtful whether these numbers give a true idea of the real proportions in which the classes occur in Western Australian waters. The first portion of the collection is evidently a selected one, the larger and more striking Echini and Asteroids predominate. Holothurians have a commercial interest, whilst some of the Crinoids are very beautifully coloured. 1 he Ophiurans, which are mostly small and fragile, do not appeal to the ordinary individual for any of these reasons and are unrepresented. The collection which was made on board the Endeavour probably approximates much more closely to the true proportions existing amongst the forms found in deeper water, on a sandy bottom, though it is probable that here again the larger forms predominate unduly, as a trawl is not the best instrument with which to collect small and fragile specimens. The following table will give an idea of the comparison between these collections and that made by the Thetis on the coast of New South Wales. RECORDS Of W.A. MUSEUM. r 107 Collection in W.A. Museum. Endeavour Collection from W. Australia. Thetis Collection from New South Wales Percentage of recent Species known in each Class Crinoids 7 12 3 1 1 Asteroids 16 7 9 25 Ophiurans ... 2 10 18 34 Echini >5 7 15 12 Holothurians 6 8 18 Totals ... 46 36 53 100 As far as they go these figures seem to prove that on both the Western and Eastern coasts of Australia Echini are unusually numerous, while the number of Crinoids found in Western Aus- tralia is a long way above the normal. A list of all the identified Echinoderms in the Museum, with their localities when known, is attached. In addition to the sources already referred to, Dr. Michaelsen has presented a number of Ophiurans and Crinoids collected in this State by the Hamburg Expedition in 1905. These are included in the list. [The Classification followed is that of Bronn’s “ Tier-reichs ” for all the groups except the Crinoids. The later class are arranged according to Mr. A. H. Clark’s Monograph of the Recent Crinoids of Australia.] io8J RECORDS OF W.A. MUSEUM. \YESTEKN AUSTRALIAN LCHINODERMATA IN THE WESTERN AUSTRAEIAN AIUSEUAI, Class CRINOIDEA (Sea Lilies, Feather Stars). Order Comatulida. Family Comasteridae. Capillaster sentosa, Carp. Capillaskr multiyadiata, Linn. Comatulella brachiolata, Lamk. Comatula pnrpurea. Mull. Comatiila Solaris, Lamk. Comanthiis alternans, Carp. Coimnthns belli, Carp. Comantlms anmdata, Bell, Comanthus parvicirra. Mull. Coinanthus polycneniis, A. H. Clark Family Zygometridae. Zygometra elegans, Bell. Between Fremantle and Geraldton. Family Himerometridae. Aniphimetra discoidea, A. H. Clark. Between Fremantle and Geraldton Family Mariametridae. Dichrometra gyges, Bell. Inner Bar, Shark Bay. Family Tripiometridae. Tropiometra afra, Hartl. Between Fremantle and Geraldton. Family Calometridae. Neometm gorgonia, A. H. Clark. Between Fremantle and Geraldton. JVeoinetni conamiuis, A. ¥L. C\aT]i. ,, ,, ,, Family Thalassometridae. Ptilovietm macvonema, Miill. Off Geraldton. Between Fremantle and Geraldton. *? ? ? )> 5? n j) >5 M »» Abrolhos Islands. Port Hedland. Between Fremantle and Geraldton. RECORDS OF W.A. MUSEUM. [log Class ASTEROIDEA (Star Fishes.) Order Phanerozonia. Family Astropectinidae. Astropecten triseyiaius, Mull & Trosch, Garden Island. Lindiamaculata, Mtill&TrQsch. Between Fremantle and Geraldton. Family Pentagonasteridae. Pentogonastef stibarius, H. L. Clark. Between Fremantle and Geraldton. Tosia australis, Gray. Fremantle. I canaster longhnanus, Mobius Broome. Nectria ocellifera, Lamk. Between Fremantle andOeraldton. Stellaster incei, Gray . Carnac Island. Between Fremantle and Geraldton. Stellaster megaloprepes, H. L. Clark Port Hedland. Family Antheneidae. Anthenea tuberculosa. Gray. Family Pentacerotidae. Oreaster gracilis, Ltitken. Oreaster nodulosns, Perrier. Culcitaster anamesus, H. L. Clark. Family Asterinidae. Nepanthia brevis, Perrier. Fremantle, Cottesloe. Anseropoda rosacea, Lamk. Port Hedland. Astevina gunnii, Gray. Fremantle, Cottesloe. Order Cryptozonia. Family Linckiidae. Linckia tyloplax, H. L. Clark, Between Fremantle and Geraldton. Family Echinasteridae. Echinaster arcy status, H. L. Clark. Between Fremantle and Geraldton. Echinaster purptiveus. Gray. Fremantle, Broome. Echinaster vestitus, Perrier? Port Hedland. Plectasterdecanus, Miill & Trosch. Albany. no] RECORDS OF W.A. MUSEUM. Family Astp:riidae. Astevias polyplax, Mill! & Trosch. Between Fremantle and Geraldton. Asterias calamaria, Gray. Class OPHIUROIDEA (Brittle Stars). Order Zygophiurae. Family Ophiodermatidae. Pectinura discvyta, H. L. Clarli. Between Fremantle and Geraldton. Family Amphiuridae. Ophiactis savignyi, M.T. Amphiura constricta, Lym. Amphmra squamata, Chiaje. Family Ophiacanthidae. Ophiacantha clavigera, Koehler. Koombana Bay, Bunbury. Family Ophiocomidae. Ophiocoma hrevipes, Peters. Shark Bay. Clavigera ivendti, M.T. Shark Bay. Family Ophiotrichidae. Ophiothrix stelligera, Lym. Cockbiirn Sound, Fremantle. Between Fremantle and Geraldton. Ophiothrix spongicola, Stimpson. Between Fremantle and Geraldton. Ophiothrix hirsuta, M.T. Freycinet Reach, Shark Bay. Ophiothrix longipeda, Lamk. Abrolhos Islands. Ophiothrix hartmeyeri, Koehler. Sunday Island, Shark Bay. Ophiothela danae, Verrill. Surf Point, Shark Bay. Freycinet Estuary, Shark Bay. Champion Bay, Geraldton. Oyster Harbour, Albany. Order Streptophiurae. Family Ophiomyxidae. Ophiomyxa australis, Liitk. Between Fremantle and Geraldton. Order Cladophiurae. Family Astrophytidae. Asii'ogyinnotes catasticta, H. L. Clark. Off Jurien Bay. RECORDS OF W.A. MUSEUM. [Ill Ophiocreas me/amhaphes, H. L. Clark. Offjurien Bay. Ophiod'eas rhahdotum, H. L. Clark. Off Jurien Bay. Conocladus microconns, H. L. Clark. Astroboa ernac, Dod. Between Fremantle and Geraldton. Off Geraldton. Off Geraldton. Euryale aspera, Lamk. Class ECHINOIDEA (Sea Urchins). Order Cidaroida. Family Cidaridae. Tretocidaris bracteata, A. Ag. Between Fremantle and Geraldton. Goniocidavis tnbavia, Lamk. ,, „ ,, Phyllacimthus annuUfera, Lamk. Port Hedland. Phyllacanthus magnifiats, H. L. Clark. Between Fremantle and Geraldton. Order Diadematoida. Family Diadematidae. Centrostephanus tenuispiims , H. L. Clark. Between Fremantle and Geraldton. Family Temnopleuridae. Salmacis alexandvi, Bell. Off Geraldton. Salmacis sphaeroides, Linn. Port Hedland. Salmacis bicolor. As. Amblypneustes grisens, Blainv. Fremantle. Amblypnenstes grandis, H. L. Clark. Between Fremantle and Geraldton. Holopneustes porosissimus, Ag. Fremantle. Holopneiistes purpmescens, A. Ag. Family Echinometridae. Heltoct'daris armigem, A. Ag. Fremantle, Cottesloe. Heliocidaris erythrogravima,Y 'a\ . Echinometra mathaei, Blainv. 112 ] RECORDS OF W.A. MUSEUM. Order Clypeastroida. Family Clypeastridae. Clypeaster telurus, H. L. Clark. Between Fremantle and Geraldton. Family Laganidae. Laganum peroni, Ag. Swan River. Peronella aphnostina, H. L. Clark. Carnac Island. Order Spatangoida. Family Spatangidae. Linthia australis, Gray. Fremantle, Cottesloe. Echinocardiinn australe. Gray. Safety Bay. Breynia ausiralasiae, Leach. Abrolhos Islands, Broome. Class HOLOTHURIOIDEA (Sea Cucumbers, Beche-de-Mer). Order Actinopoda. Family Aspidochirotae. Actinopyga tniliaris, O. & G. Holotliuria atra, Jaeger. Family Dendrochirotae. Colocliints axiologus, H. L. Clark. Port Hedland. (,'olochirns qitadrangularis, Less. Cnlochirtts tuherculosns, Q. & G. Family Molpauiidae. Caudina tetvapora, H. L. Clark. Cottesloe Beach. RECORDS OF W.A. MUSEUM. fii3 THE CRINOIDS COLLECTED BY THE ENDEAVOUR BETWEEN FREMANTLE AND GERALDTON (Western Australia). By AUSTIN HOBART CLARK. PREFACE. Mr. Bernard H. Woodward, the Director of the Western Australian Museum and Art Gallery at Perth has recently honoured me with a request to examine a collection of comatulids or unstalked crinoids brought together by the Endeavour while working along the coast of Western Australia between Fremantle and Geraldton. Mr. Wilfrid B. Alexander accompanied the Endeavour at the time the collection described herein was made, taking careful colour notes on many of the specimens. These he has been kind enough to permit me to incorporate herein. Within the past two years the Australian crinoid fauna has received a considerable amount of attention; in 1911 there was published at Sydney a comprehensive monograph on the crinoids of Australia, including a historical introduction, a complete synonymy and a bibliography ; every Australian record of a crinoid or of a supposed crinoid is given. At the same time there was published a memoir upon the crinoid fauna of Australia west of iqo^E. long., based upon the collection made by the Hamburg West Australian Expedition under Drs. W. Michaelsen and R. Hartmeyer in 1905. In this memoir every record of a crinoid in the region under consideration is included, and the faunas of the various coasts of the continent are compared. In 1912 a paper on the crinoids preserved in the Naturhistorisches Museum, at Hamburg, appeared in which the following new records were included ; Comaster belli, from Houtman’s Rocks ; Conianthus samoana 1 14 ] RECORDS OF W.A. MUSEUM. (not previously known from Australia), from Houtman’s Rocks ; and Petasometra heliaiithoides (gen. et. sp. nov.) from Shark Bay ; other species were listed from localities where they were already known to occur, or from localities within their ascertained range. In another paper a new species of Zygometra, Z. punctata, was described from the Aru Islands where it had been collected by the Sihoga, and at the same time was recorded from Port Curtis, Queensland, and from Holothuria Bank in north-western Australia. In 1913 a supplement to the memoir on the crinoids of south- western Australia was published in which some additional specimens collected by Drs. Michaelseri and Hartmeyer were recorded and the information regarding the crinoid fauna of this region was brought up-to-date. THE ENDEAVOUR COLLECTION. The present collection contains representatives of twelve species included in nine genera and eight families and sub-families, as follows ; — Family: COM ASTERIDAE. Sub-family : Capillasterinae. Capillaster sentosa (P. H. Carpenter). Capillaster mnltiradiaia (Linne). Sub-family : Comactiniinae. Comatulella hrachiolata (Lamarck). Comatula purpurea (J. Muller), Sub-family : Comasterinae. ConiantJins (Vania) annulata (Bell). Comanthus (Vania) parvicirra (J. Muller). Family: ZYGOMETRIDAE. Zygometra elegans (Bell). Family: HIMEROMETRIDAE. Amphimetra discoidea (A. H. Clark). Family: TROPIOMETRIDAE. Tropiometra afra (Hartlaub). RECORDS OF W.A. MUSEUM. [II5 Family ; C ALOMETRIDAE. Neometra goygonia, sp. nov. Neometra conaminis, sp. nov. Family: TFIALASSOMETRIDAE. Ptilometra macronema (J. Muller). These species fall naturally into the following classes : — East Indian Species. Ccipillastev sentosa. Coniantlms annulata. CapillasUy multiradiata, Comanthns parvicirra. Australian Tropical Species. Comatula purpurea. Tropiometra afra. Zygometra elegans. ? Neometra gorgonia. Amphimetra discoidea. ? Neometra conaminis. South Australian Species. ComaUdella brachiolata. Ptilometra macronema. The family Calometridae was not previously known to occur on the Australian coasts, though it was to be expected that it would be discovered as soon as dredging operations were carried into sufficiently deep water. Capillaster sentosa also is new to the Australian coasts, though in view of its wide range in the East Indian region its presence here is not surprising. Tropiometra afra, though only known from Australia, has been previously reported only from Bowen, Queensland ; its occurrence on the west coast, and so far to the southward is therefore a matter of considerable interest. The known range of Capillaster multiradiata has been extended southward from Dirk Hartog Island, and the occurrence of Comatulella brachiolata, Comatula purpurea, Comanthus annulata, Zygometra elegans and Amphimetra discoidea in the vicinity of Perth, about which there had previously been some question, has been established. RECORDS OF W.A. MUSEUM. ii6] DESCRIPTION OF SPECIES COLLECTED. Fam. : Comasteridae, A, H. Clark. Sub-fam. : Capillasterinae, A, H. Clark. Genus: Capillastkr, A. H. Clark. CAPILLASTER SENTOSA, P. H. Carpenter. Comatula muUiradiata (part) i8i6. Lamarck, Hist. nat. des animaux sans vertebres, vol. 2, p. 533. Comatula (AUcfo) muUiradiata (part) 1849. J. Muller, Abhandl. d.k. preuss. Akad. d. Wiss. (1847), p, 261. Actinometra untosa, 1888. P. H. Carpenter, Challenger Reports, vol. 26, Zoology, p. 325, pi. Ixvi., figs. 4-6. Capillaster sentosa, igi2. A. H. Clark, The Crinoids of the Indian Ocean, p, 73. Differential Characters. Capillaster sentosa has the same carious arm structure as C. muUiradiata ; the IIBr series (the series following the first or “ radial ” axillary) are 4 fs + 4), and all the succeeding division series are 3 (2 4- 3) ; the first brachial of the free undivided arms, and the first ossicles following all the axillaries except the first (IBr or “radial” axillary), bear pinnules. It differs from C. muUiradiata in possessing a much larger centrodorsal, which bears longer and stouter cirri, with between 30 and 40 (commonly about 35) segments, and much more numerous arms, these ranging from thirty to, about eighty (most commonly between forty and sixty-five in number. Locality. Between Fremantle and Geraldton. Material. Two specimens; one of these is small, but typical, with seventy-three arms ; one of the IIBr series is 2, and the remain- ing nine are 4 (34-4) ; the following series are all 3 (24-3) ; the other has thirty-three arms 115 mm. long. ; there are nine IIBr series, all 4 (3 -1-4), and thirteen 1 1 IBr series, all 3 (24-3) ; the centrodorsal is typically large and hemispherical ; the cirri are 27 mm. to 30 mm. long, and are composed of 28-30 segments. Distribution. Capillaster sentosa ranges from the Maidive Islands eastward as far as the Philippines and the Moluccas, and southward to south-western Australia. Remarks. This is the first record for this species in Australia, though its occurrence there was to have been expected. RECORDS OF W.A. MUSEUM. [117 CAPILLASTER MULTIRADIATA (Linne). Capillaster wnlhradiata, 1911. A. H. Clark, The Recent Crinoids of Australia, p. 736. 1911. A. H. Clark, Ergebnisse der Hamburger sudwest-Aus- tralischen Forschungsreise, 1905, Bd. 3, Lief. 13, S. 445. 1912. A. H, Clark, The Crinoids of the Indian Ocean, p. 74. Locality. Between Fremantle and Geraldton. Material. Three specimens. One of these has twenty-five arms 1 10 mm. long ; there are eight IIBr. series present, four of them 4 (3 -h 4) and four 2 ; there are six I IIBr. series, five 3 (2 + 3) and one 2 ; there is one IVBr. series of 3 (2-1- 3) following a IIIBr. series of 2 : the cirri are XIII, 22-1-26, 15 mm. to 20 mm. long ; the second has twenty-five arms about 100 mm. long ; of the ten IIBr. series eight are 4 (3 - 1 - 4) and two are 2 ; the five IIIBr. series are 3 (2-1-3); the cirri are XVI, 20-1-21, 17 mm. long; the third is smaller and broken, but is similar to the others. Remarks. This species has not previously been recorded from further south than Dirk Hartog Island, where the German steamer Gazelle dredged it in 7 fathoms. Sub-fam. : Comactiniinae, A. H. Clark. Genus : Comatulella, A. H. Clark. COMATULELLA BRACHIOLATA (Lamarck). Comatula hrachiolata, 1911. A. H. Clark, The Recent Crinoids of Australia, p. 742. 1912. A. H. Clark, The Crinoids of the Indian Ocean, p. 79. Comatulella hrachiolata, 1911. A. H. Clark, Ergebnisse der Hamburger sudwest- australischen Forschungsreise, 1905, Bd. 3, Lief. 13, S. 447. Locality. Between Fremantle and Geraldton. Material. One fine specimen ; the centrodorsal is thick discoidal, the dorsal pole large, very slightly convex (almost flat), 5 mm. in diameter. The cirri are XVIII, 35-37 (usually the latter), 15 mm. to 18 mm. long ; they are arranged in one (with a partial second) irregular marginal row. All of the component segments are much broader than long. The cirri taper markedly for the first seven segments. RECORDS OF W.A. MUSEUM. ii8] The mouth is radial, situated at the base of the anterior arm pair. The arms are all of equal length and size, 85 mm. long, 2-5 mm. wide at the base, and 4-3 mm. wide at the broadest place, between the twelfth and fourteenth brachials. The terminal comb on the proximal pinnules is composed of fifteen teeth which are long and slender, shaped like an arrow head with the point truncated. All the arms bear ungrooved pinnules in equal numbers. In the proximal portion of the arms the pinnules on either side typically alternate, grooved and ungrooved ; further out tliere are two grooved pinnules between adjacent ungrooved pinnules, and toward the arm tips all the pinnules are grooved. There is a very great difference in the structure of the grooved and ungrooved pinnules, which is well shown in the more proximal portion of the arm where the two types regularly alternate. The grooved pinnules, after the first two segments, which are rather large, are slender, delicate and very flexible ; the ungrooved pinnules have slightly larger basal segments than the grooved and taper very gradually so that they are much stouter than the delicate grooved pinnules ; at first they lie horizontally, but in the distal half or third they curve dorsally into the form of a hook or spiral, exactly as do the cirri, forming tendril-like attachments all along the arm whereby the animal fixes each arm securely to the organisms on the sea-floor in addition to fixing its central portion by means of its cirri. The segments of the stout grooveless pinnules are produced dorsally into blunt rounded processes exactly resembling the dorsal convex swellings on the outer cirrus segments ; these are perfectly smooth, with no trace of spines; these processes are entirely absent from the slender grooved pinnules w'hich, instead, bear on the dorsal side of the terminal segments the long recurved spines characteristic of the pinnules of all the species of this family. The colour in life was purplish red, the centrodorsal and first seven segments of the cirri darker and more brownish, the distal portion of the cirri bright red. Remarks. Morphologically the first two segments of the pinnules are merely atrophied brachials, while the remaining RECORDS OF W.A. MUSEUM. [iig portion of the pinnules, including the third and succeeding segments is merely a tentacular body process, exactly comparable to the cirri, but carrying ambulacral structures on its ventral side. Each brachial originates as, and is fundamentally, an axillary ; one of the two derivatives from this axillary, after the formation of two ossicles, which are united to each other just as are the paired ossicles in the division series, abruptly ceases its development, while the other continues to increase in size, ils basal segment attaining the same diameter as the brachial upon which it rests. The atrophied branch from the original axillary stage of the growing brachial serves as the base from which there extends outward a long tentacular structure with no phylogenetic history, which forms within itself a series of skeletal braces as necessity requires, and which is in every way exactly comparable to a cirrus, which also is a long tentacular structure with no phylogenetic history forming within itself a series of skeletal braces as necessity requires, excepting only that it bears ambulacral structures along its ventral surface. Since pinnules beyond the second segment are merely elongate tentacular processes in which a skeleton is formed as needed, and cirri are also elongate tentacular processes in which a skeleton is formed as needed, it necessarily follows that the skeleton of the two sets of organs will be essentially identical, differing only in such modification as will enable the pinnule to carry ambulacral organs on its ventral side ; and, further, that if for any reason the pinnules are not supplied with ambulacral organs on their ventral side the difference between the cirri and the pinnules beyond the second segment will almost or entirely disappear. In this connection it is most instructive to see that in this specimen the ungrooved pinnules have approached so closely to the cirri in structure that they have taken upon themselves the performance of exactly the same functions. I20j RECORDS OF W.A. MUSEUM. Genus : Comaxula, Lamarck. COMATULA PURPUREA, J. Muller. Comahda purpurea, igii. A. H. Clark, The Recent Crinoids of Australia, p. 746, igii. A. H. Clark, Ergebnisse der Hamburger sudwest-Australischen Forschungsreise, igo5, Bd. 3, Lief. 13, S. 451. 1912. A, H. Clark, The Crinoids of the Indian Ocean, p. 81. Locality. Between Fremantle and Geraldton. Material. Four specimens ; one of these is typical, with VIII cirri, three in interradial pairs and two single ; another has XI cirri ; the cirri in the other two are more numerous than usual, but are evidently undergoing reduction toward the condition normal for the species, indeed in the larger the normal arrangement occurs on about four-fifths of the periphery of the ceiitrodorsal. In the largest specimen the anterior arms are about 100 mm. long. The colour in life was reddish purple. Sub-fam. : Comasxerinae, A. H. Clark. Genus ; Comanthus, A. H. Clark. Sub-genus ; Comanxhus, A. H. Clark. Specific Group : Vania, A. H. Clark. COMANTHUS (VANIA) ANNULATA, Bell. Comanthus (Vania) annulata, 1911. A, H. Clark, The Recent Crinoids of Aus- tralia, p. 757. igii, A. H. Clark, Ergebnisse der Hamburger sudwest- Australischen Forschungsreise 1905, Bd. 3, Lief. 13, s. 457. Comanthus annulata, 1912. A. H. Clark, The Crinoids of the Indian Ocean, p. 96. Locality. Between Fremantle and Geraldton. Material. Nine specimens; two of these are uniform light yellowish brown ; the other seven are yellowish brown, darkest on the calyx, division series and arm bases where they are thickly covered with small uniform round green spots ; according to the label these were dull green in life. The details of the specimens are as follows : (i) about forty- five arms about no mm. long; VI cirri ; (2) forty arms about no mm. long; X cirri; (3) about forty arms; one of the IIIBr. RECORDS OF W.A. MUSEUM. [I2I series is 7 (3 + 4, 6 + 7); the centrodorsal is a pentagonal plate slightly raised above the surface of the radials, with a few obsolete cirrus sockets about its periphery ; (4) about forty arms ; the centrodorsal is very thin discoidal, pentagonal in outline ; there are III cirri; (5) about thirty-five arms about 1 10 mm. long ; (6) about thirty-five arms about 90 mm. long, the centrodorsal is greatly reduced; there are III cirri ; (7) about thirty-five arms ; VIII cirri; the centrodorsal is greatly reduced : (8) about thirty arms ; V cirri ; (9) about twenty-five arms about 115 mm. long. Remarks. These specimens agree with those taken by the Hamburg West Australian Expedition at Shark Bay, and with others which I have examined from the vicinity of Perth. COMANTHUS (VANIA) PARVICIRRA, J. Muller. Comanthus (Vania) parvicirra, igii. A. H. Clark, The Recent Crinoids of Aus- tralia, p. 758. 1911, A. H. Clark, Ergebnisse der Hamburger sudwest- Australischen Forschungsreise 1905, Bd. 3, Lief. 13, s. 446. Comanthus pat vitirra, 1912, A. H. Clark, The Crinoids of the Indian Ocean, p, 97. Locality. Between Fremantle and Geraldton. Material. One typical specimen with twenty-two arms 70 mm. long ; of the ten IIBr. series seven are 4 (3-f 4) and three are 2 ; there are two IIIBr. series, one 2, developed internally, and one 4 (34-4), developed externally ; the cirri are VIII, 14, 7 mm. long. Remarks. This species has previously been recorded from Fremantle. Fam. : Zygometridae, A. H. Clark. Genus: Zygometra, A. H. Clark. ZYGOMETRA ELEGANS, Bell. Zygometra eleguns, 1911. A. H. Clark,, The Recent Crinoids of Australia, p. 762. 1911, A. H. Clark, Ergebnisse der Hamburger sudwest-Australischen Forschungsreise 1905, Bd. 3, Lief. 13, s. 458. 1912, A, H. Clark, The Crinoids of the Indian Ocean, p. 104. Locality. Between Fremantle and Geraldton. Material. Eight specimens ; the details of these are as follows: (i) forty-five arms no mm. long; of the ten IIBr. series 122 ] RECORDS OF W.A. MUSEUM. nine are 4 (3 + 4) and one is 2 ; of the twenty IIIBr. series seven- teen are 2 and three are 4 (3 + 4); the five IVBr. series are all 4 (3+4) 1 of them are developed on the outermost side of the ray, the fifth being by the side of one of these ; the centrodorsal is large, thick discoidal, the dorsal pole slightly concave, 5 mm. in diameter ; the cirri are 30 mm. to 35 mm. in length, and are composed of 44-47 segments ; Pd is rather slender, about 15 mm. long, composed of from twenty-seven to twenty-nine segments ; (2) forty arms about 100 mm. long; of the ten II Br. series seven are 4 {3 + 4) and three are 2 ; sixteen of the IIIBr. series are 4 (3-I-4) and two are 2 ; two are missing ; the cirri are 30 mm. to 33 mm. long, and are composed of 37-42 segments ; Pd is 18 mm long, rather slender, composed of thirty-seven segments ; (3) forty arms about 100 mm. long; the ten IIBr. series are 4 (3-I-4) ; twelve of the IIIBr. series are 2 and six are 4 (3-^4); the two IVBr. series are 4 (3 + 4) ; the cirri are about 30 mm. long and are composed of 38-39 segments; the elongate proximal pinnules are slender ; (4) two of the post-radial series are missing ; the three post-radial series present consist of eight arms each, two IIBr. and four IIIBr. series being present in every case; all of the IIBr. series are 4 (3-f4) ; nine of the IIIBr. series are 2 and three are 4 (3-44) ; (5) thirty-eight arms, with one IIBr. series missing ; the ten IIBr. series are 4(3 + 4); ten of IIIBr. series are 2, six being 4 (3+4); (6) thirty-six arms about 100 mm. long ; the ten IIBr. series are 4 (3-1-4) : nine of the IIIBr. series are 2 and seven are 4 (3 + 4) ; nine of the IIIBr. series are 2 and seven are 4 (3 + 4) the cirri are 25 mm. to 35 mm. long, composed of 34-42 segments ; (7) thirty-six arms about 100 mm. long ; five of the IIBr. series are 2 and five are 4 (3-f4) I eight of the III Br. series are 2 and eight are 4 (3 + 4) ; the cirri are 25 mm. to 30 mm. long, composed of 36-40 segments; (8) thirty-four arms about 100 mm. long; the ten IIBr. series are 4 (3+4); the fourteen IIIBr. series are 2 ; the cirri are 25 mm. to 30 mm. long, composed of 36-43 segments. The colour in fife is recorded as very variable. Remarks. Unfortunately both Zygornetra elegans and- Z. microdiscus vary very considerably in their arm structure, and examples of both may easily be found which possess more or less completely developed the arm structure of the other. For instance, the second specimen listed above has seven of the ten IIBr. series RECORDS OF W.A. MUSEUM. [123 4(3+4) and sixteen of the eighteen IIIBr. series 4 (3 + 4) ; this would indicate the species iiiicrodisms ; but there are only forty arms, and the character of the proximal pinnules and of the cirri is identical with that of the proximal pinnules and of the cirri of the eighth specimen, which is in every way a typical example of elegans. In Zygometra comata from the East Indies, Z. andvomeda from India and Z. punctata from north Australia and the Aru Islands the characters are very stable and variation is reduced to a minimum ; but in the larger forms, as in almost all comatulids with a very large number of arms, the arm structure becomes more or less uncertain and less reliable as a systematic guide, than the structure of the lower pinnules or of the cirri. There appear to be two definite and distinct structural types about which all of the large specimens of Zygometra centre, and it therefore seems most logical to recognise two species each with a definite pinnule and cirrus structure, and a definite average arm structure. Zygometra multiradiata I believe, from an examination of the type in London, to be merely an undeveloped specimen of Z. micro- discus. I would now refer to Z. microdiscus the specimens which I recorded from northwestern Australia, and from Lewis Island in the Dampier Archipelago. Fam. : Himerometridae, A. H. Clark. Genus: Amphimetra, A. H. Clark. AMPHIMETRA DISCOIDEA, A. H. Clark. Amphimetra discoidea, igii. A. H. Clark, The Recent Crinoids of Australia, p. 766 1911, A H. Clark, Ergebnisse der Hamburger sudwest-Australischen Forschungsreise 1905, Bd. 3, Lief. 13, s. 459. 1912, A. H. Clark, The Crinoids of the Indian Ocean, p. 112. Locality. Between Fremantle and Geraldton. Depth. 60-100 fathoms. Material, Six specimens. These are all large and well developed, the arms being between 185 mm. and 195 mm. long; the centrodorsal is from 6 mm. to 8 mm. in diameter, very broad, with a flat or more or less convex dorsal pole in the centre of 124 ] RECORDS OF W.A. MUSEUM. which there is sometimes to be seen a small pit ; the cirri are XX-XXXII, the longest with 37-44 segments, and 35 mm. to 40 mm. in length ; they are moderately stout and are composed of approximately sub-equal segments of which the longest (in the proximal portion) are from half again to twice as broad as long, and the distal are slightly shorter. Remarks. The ten armed species of the genus Amphimetra are as yet very imperfectly understood. A considerable number of them have been described, mostly based upon single specimens which in several cases are small and probably immature. Extensive material from any one locality commonly shows great variation in one or more of the characters ordinarily used in specific differentiation. As we know them at present these species appear to fall into three groups; (i) species in which the cirri are very stout with all of the component segments much broader than long, the distal being slightly longer than the proximal (typified hy A . milhevti) •, (2) those in which the cirri are comparatively slender with the proximal segments never more than twice as broad as long, usually about as long as broad, and always longer than the distal, and in which the dorsal spines on the cirrus segments are small, short and inconspicuous, developed only in the outer portion of the cirri (typified by A. discoidea) and (3) those in which the cirri, while in general resembling the cirri of the preceding group, possess large, long and conspicuous dorsal spines which are developed to well within the basal third (typified by A. jacquinoti). The specimens under consideration appear undoubtedly to be exceptionally large aud well developed examples of A. discoidea; typically discoidea has more slender and more tapering cirri in which the longer proximal segments are very nearly or quite as long as broad, but the distal are shorter, broader than long ; in typical inilherti the cirri are much stouter than in these specimens, the segments all being approximately of equal length, though the outer are a trifle longer proportionately, about four times as broad as long. RECORDS OF V/.A. MUSEUM. [125 Fam. : Tropiometridae, A. H, Clark. Genus : Tropiometra, A. H. Clark. TROPIOMETRA AFRA (Hartlaub). Tropiometra afra, igii. A. H. Clark, The Recent Crinoids of Australia, p. 780. 1912, A. H. Clark, The Crinoids of the Indian Ocean, p. 176. 1912, A. H. Clark, The Crinoids of the Natural History Museum at Hamburg, p. 28. Locality. Between Fremantle and Geraldton. Depth. 60-80 fathoms. Material. Two specimens ; one of these has an arm length of about 220 mm. ; the centrodorsal is thick discoidal, 10 mm. in diameter and 4 mm. high interradially ; the cirri are XXIII, 34-35, 35 mm. to 40 mm. long ; the other is similar, with arms 225 mm. long. The colour in life was dark purplish brown. Remarks. This species differs from T. macrodiscus of southern Japan in its smaller and more slender cirri ; T. afra, which is represented in the museums of the world by only three specimens other than the two described above, was previously known only from Bowen, Qeeensland, where two specimens were collected, more than fifty years ago, by the representatives of the famous Godeffroy company of Hamburg ; the third specimen was brought home from the “ South Pacific ” by the United States Exploring Expedition. Fam : Calometridae, A. FI. Clark. Genus : Nbometra, A. H. Clark. NEOMETRA GORGONIA, Sp. Nov. Locality. Between Fremantle and Geraldton. Depth. 80-120 fathoms. Material. Seven specimens. Description of the type specimen. The centrodorsal is discoidal, broad, with a broad and flat dorsal pole 5 mm. to 6 mm. in diameter ; the cirrus sockets are arranged in one and a partial second crowned and irregular marginal row. 126] RECORDS OF W.A. MUSEUM. The cirri are XIX, 39-50, 35 mm. to 45 mm. (usually about 40 mm.) in length, long, large and stout, with a pronounced taper in the distal half; the first nine or ten segments are half again to twice as broad as long (usually nearer the latter) the first segment being similar to those succeeding ; the segments following the ninth or tenth gradually become shorter, but at the tip of the cirrus slowly increase in length again ; the tenth has on the dorsal side just within the distal border an inconspicuous slightly elongated median tubercle which on the succeeding slowly transforms into a narrow and low, though prominent, median carination running the entire length of the segment ; on the fifteenth two small tubercles appear, one on either side of the median carination just within the distal edge of the segment ; these increase in size and elongate, after two or three segments, becoming prominent low narrow keels which resemble the median keel, though they are .slightly less in height and do not extend quite so far toward the proximal border of the ossicle; they are not quite parallel to the median keel, but converge slightly toward the proximal end of the segments ; distally all three carinate processes increase in height, especially the median, and a tubercle, which may be more or less elongate, usually appears just outside of the distal end of each of the lateral keels ; on account of the terminal taper of the cirri the opposing spine and terminal claw are rudimentary. The radials are short in the median line, but extend upward in the angles of the calyx in the form of broad processes with parallel or slightly converging sides which entirely and w'idely separate the bases of the IBr.i ; these processes are sharply truncated distally, and are not expanded or spatulate. The division series and arms extend outward almost hori- zontally from the calyx, as in Neometra sibogae. The IBr.i ; are short, oblong, four or five times as broad as long; the ventrolateral edge is produced into a thin border which, viewed dorsally, is seen to run from the distal edge of the interradial process of the radials to the distal lateral angles of the IBr.i where it disappears from dorsal view, being continued along the ventral side of the axillary and of the division series forming a deep trough in which the “soft parts” lie. The IBr.2 (axillaries) are broadly pentagonal, nearly twice as broad as long; the lateral RECORDS OF W.A. MUSEUM. [127 edges are little, if any, shorter than those of the IBr.i, making with them a broadly obtuse angle. The IIBr. and fllBr. series are 2, and all of the latter are developed. There are forty arms (thirty-nine in the type) 95 mm. long, resembling those of other species of the genus ; there is little or no overlapping of the distal edges of the brachials. Pi is 14 mm. long with twenty-eight segments, of which the first two are greatly enlarged, over twice the breadth of the succeed- ing, with the distal edge more or less convex, the third or third and fourth are slightly broader than long, and the remainder are sub- equal, slightly longer than broad ; as a whole the pinnule is proportionately longer and somewhat stouter and stronger, than is usual in the genus. P2 is 18 mm. long and is composed of nineteen segments; it resembles P3 but is very slightly less stout. P3 is 19 mm. long, very stiff and spine-like, composed of nineteen segments of which the first two are broader than long, slightly enlarged, with distal carinate processes the edges of which are straight and parallel to the longitudinal axis of the pinnule or nearly so, the third is nearly as long as broad, and the following are elongate with, after the eighth, produced distal edges which at the prismatic angles are provided with prominent spines. P4 is 14 mm. long and is composed of fifteen segments ; it resembles P3, but is less stout and the two first segments are less enlarged. P5 is 10 mm. long, composed of twelve segments ; it is more slender than P4. The distal pinnules resemble those of the other species of the genus ; they are very sharply triangular in cross section and their ambulacra are provided with very highly developed side and covering plates ; they measure 9 mm. in length, and are composed of fifteen segments of which the terminal four or five, upon which the ambulacra! grooves do not extend, are abruptly smaller than the preceding. The color in life is white, with purple bars and blotches on the pinnules. The specimens other than the type, (i) Forty-two arms; there are two IVBr. series, both developed on the inner side of 128] RECORDS OF W.A. MUSEUM. external IIIBr. series ; the dorsal pole of the centrodorsal is 6 mni. to 7 mm. in diameter ; the cirri are'XXII, 44-46, 35 mm. to 40 mm. long ; the colour is white, the outer part of the arms narrowly banded, and the pinnules broadly blotched, with purple ; (2) forty- one arms about 75 mm. long ; the cirri are XX, 39-44, 35 mm. to 40 mm. long; (3) about forty arms 65 mm. long; the cirri are about 30 mm. long; the colour is white, the arms beyond the division series narrowly and regularly banded with purple, the bands being continued on to the pinnules; (4) forty arms 60 mm. long ; dorsal pole of the centrodorsal is 5 ram. in diameter ; the cirri are XXI, 34-39, 25 mm. long ; (5) about forty arms, there are two IVBr. series ; (6) forty arms, similar to the preceding. Comparison with related species. Neometra gorgoma is very different from all of the other described species of the genus. As now known the species of this genus, on the basis of the number of their arms, fall into five groups : (i) species having from fifteen to twenty arms [aleoto) ; (2) species having twenty arms (multicolor spinosissima, conaminis) ; (3) species having thirty arms {acanthaster) ; (4) species having from thirty to forty arms (sihogae) ; and (5) species having forty arms (gorgonia). The only species with which this can possibly need comparison is N . sibogae. On the terminal twelve or thirteen cirrus segments in N. sibogae the high median carination is accompanied on either side by a usually more or less elongate tubercle which, however, is com- paratively small and inconspicuous. I did not notice this until my attention was called to the similar, but far more prominent, pro- cesses in N. gorgonia. In N. sibogae the edges, both proximal and distal, of the elements of the division series and the proximal brachials, and the distal edges of the brachials and of the cirrus segments are promi- nently everted, giving the animal a curiously ornate appearance ; in N. gorgonia the edges of the elements of the division series, brachials and cirrus segments are smooth, so that the general aspect of the two forms is strikingly different ; the cirri of N. gorgonia are very much longer than, in fact nearly twice as long as, the cirri of N. sibogae, and the three keels on the dorsal side of the outer segments are very much more strongly marked ; the cirrus segments in N. gorgonia are much more numerous than in N. sibogae. RECORDS OF W.A. MUSEUM. [129 NEOMETRA CONAMINIS, Sp. Nov. Locality. Between Fremantle and Geraldton. Depth. 80-120 fathoms. Material. Four specimens. Description of the type specimen. The centrodorsal is of medium size, discoidal, the dorsal pole flat, 4-5 mm. in diameter ; the cirrus sockets are arranged in a single more or less irregular marginal row (in one of the other specimens there is also a very deficient second row). The cirri are XIV (in the other specimens varying from XIV to XIX), 40-45, 25 mm. long ; the longer earlier segments are half again as broad as long to nearly as long as broad, and the shorter distal segments are about three times as broad as the median length, becoming longer again terminally ; the cirri are not particularly stout ; they taper slowly and gradually throughout their whole length ; on the sixth or seventh segment the median dorsal portion of the distal edge becomes slightly prominent ; this prominence rises in height and slowly extends itself proximally until on about the tenth there results a narrow median keel running the whole length of the dorsal surface ; on the succeeding segments this gradually increases in height, becoming the high thin median carination characteristic of the outer cirrus segments of the species of this genus ; the ventral surface of the cirri is rather narrowly rounded so that in cross section the cirri are seen to approach a rhombic shape, but with the four angles of the rhombic outline very broadly rounded. The radials are concealed in the median line, but extend upward in the angles of the calyx in the form of triangular processes of which the sides, which are about as long as the bases, are concave and the apices are truncated; the apices of these triangular processes entirely, though not very widely, separate the bases of the IBr.i. The IBr. i are very slightly trapezoidal, almost oblong, from three to four times as broad as long ; the ventrolateral edges, though sharp, are only very slightly if at all produced ; these ossicles occasionally bear an obscure low median keel ; the IBr. 2 (axillaries) are broadly pentagonal, slightly broader than long, the lateral edges usually slightly shorter than those of the IBr.i. 130 ] RECORDS OF W.A. MUSEUM. The twenty arms, which resemble those of related species, are from 85 mm. to 90 mm. long. Pi is 9 mm. long, composed of from twenty-five to thirty segments of which the first two are greatly enlarged and the remainder are slightly longer than broad ; P2 is 12 mm. long, with sixteen or seventeen segments which after the third become much elongated, those in the outer part of the pinnule having produced distal edges bearing prominent spines at the prismatic angles ; the pinnule is stiff and spine-like ; P3 is from 15 mm. to 17 mm. long with from sixteen to eighteen segments, resembling P2 but pro- portionately larger ; P4 is 17 mm. long with eighteen segments, similar to P3 ; P5 is 12 mm. long with fifteen segments, similar to the preceding ; P6 is 10 mm. long with fourteen segments ; the distal pinnules are about 11 mm. long. The colour in life is white with yellow bands on the arms and pinnules, more rarely on the cirri. The specimens other than the type. Three, all with twenty arms, and all resembling the type. Comparison with related species. The species of the genus Neometra at present known are seven in number ; they may be briefly diagnosed as follows: — Neometra alecto ; Fifteen to twenty arms, 60 mm. to 70 mm. long; cirri IX-XV, 39-46, 25 mm. to 30 mm. long (Philippine Islands ; 42-58 fathoms). Neometra multicolor ] Twenty arms, 60 mm. long, cirri XV, 35, 20 mm. long (southern Japan ; 20-110 fathoms). Neometra conaminis', Twenty arms, 85 mm. to 90 mm. long; cirri XIV-XIX, 40-45, 25 mm. long (southwestern Australia; 80-120 fathoms). N eometra spinosissima \ Twenty arms, 130 mm. long; cirri XI, 42-55, 25 long (Andaman Islands). Neometra acanthaster ; Thirty arms, 60 mm. long ; cirri 20 mm. long, ventrally carinate (Philippine Islands ; 49 fathoms). Neometra sihogae ; Thirty to forty arms, 70 mm. to 75 mm. long ; cirri XV, 31-36, 25 mm. long; edges of all the ossicles produced ; terminal cirrus segments with triple dorsal processes (Solor Strait, east of Flores ; 1 13 metres). RECORDS OF W.A. MUSEUM. [131 Neometm gorgonia ; Forty arms, 95 mm. long ; cirri XIX, 39-50, 35 rnm. to 45 mm. long ; edges of all the ossicles smooth ; all of the dorsal processes on the cirrus segments triple, the terminal often with two additional elements (south-western Australia; 80-120 fathoms). From an examination of the data given above it is evident that N. conaininis belongs to the group of species characterized by the possession of twenty arms, including Af. alecto, N. mtdticoloy , and N. spinosissima. The Japanese N. multicolor is much smaller than N . conaminis, and possesses shorter cirri with considerably fewer segments ; N . spinosissima from the Andaman Islands is much larger with pro- portionately shorter cirri which have more numerous segments ; N. alecto is of about the same size as N. conaminis, and possesses cirri with the same number of segments ; the cirri of N. alecto are proportionately noticeably longer and more slender than are those of N. conaminis, while the brachials and cirrus segments have slightly produced distal edges, these being smooth in N. conaminis. Fam : Thalassometridak, A. H. Clark. Genus ; Ptilometra, A. H, Clark. PTILOMETRA MACRONEMA, J. Muller. Ptilometra macronema, igii. A. H. Clark, Bull, du mus. d’hist. nat. de Paris, No. 4, 1911, p. 255, fig. iB, p. 256. 1911, A. H. Clark, The Recent Crinoids of Australia, p. 781. 1911, A. H. Clark, Ergebnisse der Hamburger sudwest-Australischen Forschungreise, 1905,8(1.3, Lief. 13, S. 461. 1912, A. H. Clark, The Crinoids of the Indian Ocean, p. 189. Locality. Off Geraldton ; “ very abundant.” Depth. 25-40 fathoms. Material. Ten specimens, all of medium size, the arms being 45 mm. to 55 mm. in length from the radials and the longest cirri about 45 mm. long, composed of 69-78 segments; six of the examples have twenty arms, two have nineteen, one has sixteen and one has fourteen. The colour in life is recorded as purple, the cirri red. In spirits the colour is yellowish-brown, the cirri becoming deep purple in the outer half. 132 ] RECORDS OF W.A. MUSEUM. THE ECHINODERMS OF THE WESTERN AUSTRALIAN MUSEUM. HUBERT LYMAN CLARK, Ph.D. Museum of Comparative Zoology, Cambridge, Mass., U.S.A. The authorities of the Western Australian Museum at Perth, having entrusted to me their collection of echinoderms for identifica- tion and study, I beg to offer here my sincere thanks for the honor they have thus done me. My obligation to Mr. Bernard H. Woodward, the director of the Museum, is particularly heavy, for his uniform courtesy and for the many favours he has shown me. The collection itself is not a large one but it is of extraordinary interest, not only for some of the hitherto undescribed species it contains, but also for the light it throws on the range of many East Indian and Australian echinoderms. The number of specimens sent me is only gg, but the number of species is more than half that, no less than 58 being represented. The collection is thus a selected one, usually only one or two specimens representing each form. There is thus no opportunity for me to comment on the variability of these West Australian species. The collection is notable for the large number of new and remarkable species it contains. There are no new crinoids, but five starfishes are undescribed, and one of these represents a new and remarkable genus, while three others are notably distinct forms. Of theophiurans, four are new, one representing a new genus and the other three being very different from any near allies. Of the echini, four, and probably five, are undescribed, and it is interesting to note that one of these is a cidarid, and another a centrechinid, the two oldest groups of Recent echini, while the other two are clypeastroids, a group particularly abundant and wide- spread in Tertiary times. Of the holothurians, certainly two, and RECORDS OF W.A. MUSEUM. [133 probably three, represent new forms, one of which may ultimately require a new generic name to emphasise its peculiarities. There are thus not less than 15 new species in this small collection, or more than a fourth of the total number. Of the others, at least 22 are not known from outside the Australian region. As 5 are not certainly identified, it follows that of 53 species, 37 or 70 % are characteristic of Australia, certainly a very notable fact. Of course, further knowledge of both the Australian and East Indian faunae, will alter these figures, perhaps materially, but they are at least suggestive of the remarkable echinoderm fauna which awaits further study on the western coast of Australia. Finally, the collection at hand is notable for the presence of many species whose occurrence on the western side of Australia was not previously known, and in some cases at least was hardly to be expected. Among the starfishes, the most notable case is that of Ansevopoda rosacea, previously known from Japan and the Bay of Bengal. Among ophinrans, there are two species, Pectinnra dyscrita and Ophiothrix spongicola, previously known only from New South Wales. Among echini, the occurrence of a Temnopleurus is par- ticularly remarkable, while the presence of Tretocidaris bracteata, previously known only from the East Indies and Japan, is most interesting. Holotypes of the new species described in this paper are in the Western Australian Museum at Perth, but cotypes of ten of them (wherever the holotype was not the only specimen) are now in the Museum of Comparative Zodlogy. CRINOIDEA. There are only five crinoids in the collection. One of these (No. 4922) is a damaged calyx with only the basal part of the arms, and hence is not identified, but it is certainly not one of the Comas- teridae. The other four represent four different species, but all belong to the single family Comasteridae. The chief interest of these crinoids lies in the fact that one of the species is new to the fauna of Australia, not being included in the list of “ Recent Crinoids of Australia ” published in 1911 by Mr. Austin Hobart Clark. 1 1 Clark, A. H., igii, “ The Recent Crinoids of Australia,” Mem. Austr. Mus., IV, pt. 15. RECORDS OF IV. A. MUSEUM. 134J COMATULA SOLARIS. Lamarck, 1816. Hist. Nat. Anim. s. Vert., vol. 2, p. 533. This species is well-known from Queensland and Torres Strait, and has been recorded from as far west as Holothuria Bank. Its occurrence on the western coast of Australia is therefore not remarkable. The specimen in the present collection is in what Mr. Clark (l.c.) calls the “broad-armed and generally robust phase” and is unicolor — almost black. Coast of West Australia. One specimen. No. 260. COMANTHUS ALTERNANS. Aciinometra aUernans, P. H. Carpenter, 1881. Notes from Leyden Mus., vol. 3, p. 208. This species has been recorded from the Philippines and from Port Molle, Queensland. Its occurrence, therefore, at the Abrolhos Islands, North-western Australia, is of no little interest. The specimen before me is small, having only 34 arms, each about 60 mm. long, which is an inaccurate way of saying that the tip of each ray is a little more than 60 mm. from the centre of the disk. There are two cirri, attached to the margin of a flat centre-dorsal less than 3 mm. in diameter ; the larger cirrus is broken ; the smaller has 15 very short, slightly swollen joints. In three of the rays the II Br series is 4 (3+4), the III Br is 2, and the IV Br is 4 (3-t-4) again ; in a fourth ray, on one side. III Br is 4 (3 + 4) and IV Br is 2 ; in the fifth ray, on both sides II Br is 2 ; and III Br is 2 also. The colour of this specimen is light wood-brown. On account of its small size and the fact that the two arms fail to show the regular alternation of the division series, I have hesitated in calling this specimen alternans, but I think there is really little doubt that it is a young example of that species. Abrolhos Islands, Western Australia. One specimen. No. 125. COMANTHUS PARVICIRRA. Aleclo parvicirra, J. Muller, 1841, Arch. f. Naturg., Jhrg. 7, vol. i, p. 145. Comanihus parvicirra, A. H. Clark, 1911. Mem. Aust. Mus., IV, pt. 15, p. 758. The occurrence of this species in the collection is quite to be expected since it has been previously recorded from Fremantle. Coast of West Australia. One specimen, without number. RECORDS OF V/.A. MUSEUM. [135 COMANTHUS POLYCNEMIS. A. H. Clark, 1910. Proc. U.S. Nat. Mus., vol, 36. p. 396, The specimen which I refer to this species is a small one, scarcely 120 mm. across, with 57 arms. The colour is a very deep olive- green, but the young arms and the cirri are pale brown or a dirty cream colour. There are 15 cirri, each with 15 joints, but they have the appearance of being fugaceous and the small thin centrodorsal adds weight to such an impression. Unfortunately, Mr. Clark’s description is rather unsatisfactory, no statement as to size, colour, or number of arms being made. There is a vague reference “to my previous list,” but I have not succeeded in finding the list to which he refers. It is possible, therefore, that I am wrong in referring this specimen to polycnemis (especially in view of the presence of numerous cirri) but the arrangement of the division series is a very noticeable character. Abrolhos Islands, Western Australia. One specimen. No. 124. ASTEROIDEA. The starfishes are a most interesting section of this collection, since five of the sixteen appear to be undescribed, and one of these represents a new and interesting genus. Of the sixteen species, only three are certainly known from outside of the Australasian region. LINDIA MACULATA. Miiller and Troschel, 1842. Sys. Ast. p. 77. A small specimen, with four of the seven arms broken, proves this widespread species to be a native of the West Australian coast. The colour in life is recorded as “buff with dark bands.” R.=r 80 mm. Between Fremantle and Geraldton, W.A., 8o-i2ofms.,No. 4919. TOSIA AUSTRALIS. Gray, 1840. Ann. Mag. Nat. Hist., vol. 6, p. 281. Since the type of this species was from Swan River, Western Australia, it is not strange that it should be represented in the present lot. From piles of an old jetty, Fremantle, W.A. Two specimens,. No. 6045. 136 ] RECORDS OF W.A. MUSEUM. PENTAGONASTER STIBARUS, ’ sp. nov. Plate XVII, Rays 5. R=52 mm., r=28 mm. R 2r. Interbrachial arcs well rounded. Rays broad and flat. Breadth at base, 29 mm. Disc large, very little arched ; vertical diameter only about 13 mm. Superomarginal plates, 40, that is, eight on each side or four on each margin of each arm ; the terminal pair the largest and much swollen ; the others are subequal and little swollen. Each plate is surrounded by a single, double or even triple series of minute, crowded, flat- topped granules, but the surface of each plate is perfectly smooth. Inferomarginals 60, that is, twelve on each side or six on each margin of each arm ; on each side of the starfish, the six median inferomarginals correspond exactly in position with the superomar- ginals above them, but the swollen terminal superomarginal overlies three inferomarginals, of which the antepenultimate (and not the penultimate nor the last) is the largest and is slightly swollen. The penultimate is distinctly smaller but is equally swollen, while the last is quite small and is almost or quite flat. The inferomarginals are smooth and surrounded by granules exactly as are the supero- marginals. The plates covering the abactinal surface are smooth, polygonal or rounded and surrounded, like the marginals, by a series of flat- topped granules. The largest are in the interradial areas and are four or five millimetres across ; the smallest are near the tips of the rays. Distally, the median radial series are slightly swollen, while proximally, in the type (the larger specimen), each plate bears a blunt low spine, or high tubercle. Two or three plates not in the median radial series, but near the centre of the disk, bear similar but small tubercles. In the smaller specimen (Ru-41 mm.) such a tubercle is indicated on only one plate, the dorso-central, while the five primary interradial plates are much larger and more conspicuous than in the type. In both specimens, the rounded triangular raad- reporic plate lies just distal to one of the primary interradials ; in the type, it is much swollen. The actinal surface is covered by polygonal plates similar to those of the abactinal side and like them surrounded by series of granules. Adjoining each mouth angle is a single plate, distal to 1 Gr. stibaros — compact, sturdy. RECORDS OF W.A. MUSEUM. [137 wtiich are a pair of plates of about equal size. Distal to them is a series of three or four plates. Theoretically this series is succeeded by one of six and that by one of eight plates, the median four of which abut on the inferomarginals. Actually, however, these last tvvo series are irregular in number, form, size and position of the plates. In any case, however, the plates which abut on the infero- marginals are the smallest and those next the mouth plates are the largest of the actinal intermediate plates. The adambulacral plates are about twice as wide as long. Their armature consists of a series of three stout, blunt, somewhat prismatic, furrow-spines, of which the middle one is either the largest or the smallest. This series is parallel to the furrow, and directly back of it, on the surface of the plate, is a second series of similar, but shorter and stouter spines. There may be either two or three spines in this series but three seems to be the typical number and when there are only two they are exceptionally stout. The remaining surface of each plate is covered by six to ten small, low, truncate, prismatic spines, arranged in series of three (rarely two or four) parallel to the furrow ; the outermost are very similar to the granules surrounding the actinal plates. The armature of the oral plates consists of six large marginal spines on each side of the “ jaw,” the two at the apex being the largest. On the surface of each jaw there are six to ten similar but smaller spines (three to five on each side) while the low keel of the jaw is concealed by about eight thick prismatic spinules arranged in a double series. Excavate pedicellariae, similar to those of P . pidchelhis and P. duheni, but smaller than those of the former and larger than those of the latter, occur on many of the abactinal plates. None were found on the actinal surface of either specimen. They occur irregularly scattered and may be present on any plate, but no plate carries more than one. They may have either two or three valves; in the smaller specimen they are mostly two- valved, but in the t}pe, the three-valved form predominates. Colour of type, very pale fawn ; the smaller specimen is brown- yellow. In life, the colour of both was “ bright orange.” Between Fremantle and Geraldton, W.A. 40-100 fms. Two specimens. No. 4916. The larger specimen is the Type. This species is obviously related to P. dubeni, Gray, but is at once distinguished by its much broader rays, covered by nearly 138 ] RECORDS OF W.A. MUSEUM. circular plates, in three-five longitudinal series, and the much larger terminal superomarginal plates. In diiheni when R=4i, r=:^only 17; the arm is 20 mm. broad at base and only ii mm. at a point half- way between tip and centre of disk ; there are 50-60 superomar- ginals. In stibarus with R=4i mm., r=23, br=26 and br at the middle of arm=20 mm. there are only 40 superomarginals. In dubcni, theabactinal plates on the rays are in only 1-3 series, swollen and much elongated parallel to the axis of the ray, but in stibarus, they are nearly or quite flat and more or less circular. From pulchellus, the type of the genus, stibarus differs greatly in the much closer plating of both surfaces, with smaller granules between the plates, and in the pedicellariae being abactinal instead of aclinal. If the abactinal tubercles are a constant feature of adult stibarus, they will serve as an additional specific character. Fisher has stated as a character of Pentagonaster that it is the plate of each marginal series which is largest and not the penultimate or antepenultimate. Examination of four specimens of pulchellus and three of dubeni shows that while it is the last of the superomarginals which is largest, of the inferomarginals, the last is smallest and the penultimate is largest, though sometimes approximately equalled by the antepenultimate. In the type of stibarus as already stated, the antepenultimate superomarginal is largest, but in the smaller specimen in only two of the ten possible cases is the antepenultimate ‘largest, in the other eight it is the penultimate which exceeds. This difference, however, is more im- portant on paper than in reality for in every series in the two specimens it is the fourth plate from the median interradial line which is enlarged and the apparent difference is due to whether one or two small plates are developed distal to it. The characteristic arrangement of the marginals in Pentagonaster may then be stated thus: The last superomarginal plate is enlarged and swollen and overlies the last two or three inferomarginals ; of these the penulti- mate, or sometimes the antepenultimate, is largest. 1 igii. Bull. U.S. Nat. Mus., No. 76, pp. 166 and 171. RECORDS OF W.A. MUSEUM. [139 NECTRIA OCELLIFERA. Asterias ocellifera, Lamarck, 1915. Anim. s. Vert. vol. 2 p. 553. Nectria ocellifera,. Gray, 1840. Ann. Mag. Nat. Hist. vol. 6, p. 287. (ocnlifera lap. cal.) The number of species in the genus Nectria has for many years been a matter of dispute. Perrier first called attention to the differences between the original specimens at Paris and the speci- mens in the British Museum. He ^ was sure they represented two different species and suspected that there was possible a third in the Paris Museum. Sladen, ^ however, on the basis of the Challen- ger material was inclined to think that there was only a single species, the differences to which Perrier called attention not being of specific value. Fisher, in a recent discussion of the genus, points out that the material in the M.C.Z. collection indicates the existence of two species. Very recently the M.C.Z. has received from Mr. Joseph Gabriel some well preserved specimens of Nectria from Victoria, and these, on comparison with two specimens from West Australia in the present collection, enable me to throw some light on the question. The West Australian specimens are essentially alike except that one has six rays, and is somewhat smaller than the other. These specimens agree well with Perrier’s description of ocellifera and I feel little hesitation in referring them to that species. The specimens from Victoria, on the other hand are obviously dif- ferent and seem to agree very well with Perrier’s description of his species ocellata which was based on Gray’s specimen in the British Museum. As all my specimens are well-grown and of approxi- mately the same size, it is clear that the differences are not in any way to be correlated with age, and I believe both species are fully entitled to recognition. The most striking difference between them is in the appearance of the dorsal paxilliform ossicles ; in ocellifera the top of each ossicle is covered by a group of 10-30 flat, irregularly polygonal, well-spaced granules, surrounded by a marginal crown of 10-25 large spaced, flaring flattened scale-like projections; in ocellata the top of each ossicle is much smaller and bears 10-30 rounded, well-spaced hemispherical grains, surrounded by a crowded series of 1 C. Perrier, 1876. Arch. Zool. Exp. vol. 5, pp. 1-6. 2 W. P. Sladen, 1889. C/irtl/sng'rr Asteroids, pp. 318-321. 3 W. K. Fisher, 1911. Bull. U.S. Nat. Mus., No. 76, pp. 163, 164. 140 ] RECORDS OF W.A. MUSEUM. ^5-30 grains, a little higher than, but not essentially different from, those on the top. As a result of this difference in the ossicles the madreporic plate in ocellata is plainly visible and the papulae seem larger and more conspicuous than in ocellifera where also the mad- reporic plate is more or less concealed. The marginal plates are more numerous in ocellata than in ocellifera ; in a specimen of the former having 70 there are 23 superomargiiials on one side of a ray, while in an ocellifera with R:^83, there are only 18. The actinal intermediate areas are much more closely granulated in ocellata than in ocellifera, in the latter the separate plates are quite distinctly indicated by the groups of granules which they bear. The figures given by Sladen in the Challenger Report are all illustrations of ocellata and it seems to me probable that he had not seen ocellifera. Whether the large specimen in the M.C.Z. collection (No. 1932) upon which Fisher made his anatomical investigations is ocellata or represents a new species, I am unable to decide. It is much larger than any ocellata I have seen, and its peculiarities may be a matter of age, but I confess that I think it likely it represents a third species. The dorsal paxilliform ossicles are close fitting polygons unlike any that I have seen in other Nectrias. On the other hand, the larger specimen of ocellata have pedicellariae and occasionally show inter-marginal papulae in the interbrachial arc (though, it must be added, very rarely), while both of these features, particularly emphasised by Fisher, seem to be wanting in the specimens of ocellifera. It seems to me that neither is a constant generic character. W hether ocellata and ocellifera have distinct geographical ranges remains to be determined. The exact locality whence Lamarck’s type came from is not known. The British Museum specimens described by Gray and by Perrier, the Challenger material and all of the specimens in the M.C.Z. are from south-eastern Australia or lasmania, and these are all ocellata. The two specimens before me from West Australia are ocellifera. So far as the evidence goes, then, the areas occupied by the two species do not overlap. The West Australian specimens have the following data : Between Fremantle and Geraldton, W.A., 60-100 fms. Colour orange. No. 4914. Two specimens. RECORDS OF W.A. MUSEUM. [Hi STELLASTER MEGALOPREPES, ' sp. nov. Plate XVIII. Rays 5. R=i2o mm. r. = 45 m. R. = 2fr. Br at base = 4o mm. ; at half-way point, 24 rnm. Interbrachial arcs well rounded. Disklarge, not highly arched, yet distinctly convex; vertical diameter about 20 mm. Entire abactinal surface covered by a closely granulated membrane, obscuring but not actually concealing the out- lines of the underlying plates. Superomarginals, 22 on each side of ray ; median interradial pair about 12 mm. high by 4 mm. wide, but distally the height decreases with little change in width, the fifteenth plate from the interradius being about 6 mm. high by 3.5 mm, wide (or long). At the tip of the ray the 19-21 superomarginals of the two sides are in contact abactinally but the twenty-second pair are so small, they are completely separated from each other by the twenty-first pair being in close contact with the rather large terminal plate. Abactinal ends of superomarginals (except distalmost) rounded wedge-shape, with a group of papulae on either side. No spines, tubercles or pedicellariae on any supero- marginals. Abactinal skeletal plates small, largest in interradial areas, where the}' may measure nearly five millimetres across. Median radial series consists of about forty plates and runs to a point about nine or ten millimetres proximal to tip of ray, where meeting of superomarginals of opposite sides puts an end to abactinal plating. On each side of median radial series, at base of arm, are two parallel series of somewhat smaller plates ; the one adjoining the median series runs about as far as the fifteenth superomarginal while the outer one ends at the eleventh or twelfth. Excepting thirty (more or fewer) distal, medial radial plates and about ten or a dozen distal plates in each of the two adjoining series, all abactinal plates carry sharp conical spines ; the largest are situated one at the base of each ray on the first of the tnedian radial plates ; these five spines are about five millimetres high and two in diameter, at base ; most of the spines are two millimetres high or less, but all if uninjured are very sharp. Papulae numerous but rather small, in groups of 3-20, all over abactinal surface ; largest groups between upper ends of superomarginal plates. Madreporic plate, 7mm. long by 6 wide, rounded triangular, its proximal margin 12 mm. from centre of disk. Arms very distinct. Gr. megaloprepes — magnificent, splendid. 142] RECORDS OF W.A. MUSEUM. Actinal surface, covered, like abactinal, with a closely granu- lated membrane, but outlines of plates quite distinct in dried specimen. Inferomarginal plates, 22 on each side of a ray, corres- ponding exactly in position to superomarginals, and approximating them in size; the distal margin of each is greatly modified by the series of spines borne thereon. Beginning at tip of ray (to pass from simple to complex conditions), each inferomarginal plate carries at or below the middle of its distal margin a stout, flattened, bluntly pointed spine, rather longer than the plate and lying appressed to the surface of the ray and parallel with its long axis. On plates 4-8 a second similar but smaller spine is present, below (f.«. actinal to) the first; on plates g-ii, a third spine still smaller arises below the second ; on plates 12-14 there is a fourth ; on plates 15-18, a fifth ; and on plates 19-22, a sixth. On plates 20-22 a second series of two or three spines appears, proximal to the first, at the base of the second and third spines of that series. With increase in number, there is a change of position and a marked increase in length of these marginal spines, so that on the median interradial inferomarginals, each of which bears 8 or 9 spines, the original first spine is 7 or 9 mm. long, and is borne on the outer proximal corner of the plate. Actinal intermediate plates rather few (25-30) aside from the actino-lateral series, which reaches scarcely to the middle of the ray ; very rarely a small spine, similar to those on the inferomar- ginals may be seen on one of the distal intermediate plates. Adam- bulacrai plates, about as long as wide ; armature in two series, parallel to furrow ; an inner series of 6 or 7 slender spines, median longest, adoral and aboral, shorter and subequal ; and an outer of 2 (or rarely 3) very broad thin and flat, truncate spines, erect on surface of plate ; these flat spines are 3-4 mm. long and i-i'5 mm. wide. Oral plates with about 7 spines on each side, of which the innermost are longest and stoutest, and are distinctly prismatic; surface of each jaw with about 3 large, flat, thin spines similar to those in outer adambulacral series. Pedicullariae abundant and of two kinds. Scattered irregularly over the abactinal surface are small hivalved pedicellariae (using Fisher’s terminology), with jaws about half a millimetre long. Less common, and generally situated near a spine, are tall, 2-valved spatnlate pedicellariae, with jaws nearly a mllimetre high. Bi- RECORDS OF W.A. MUSEUM. [143 valved pedicellariae sometimes occur with three jaws and they often have the jaws as high as wide. It is easy to follow the transition from bivalved to spatulate pedicellariae, in this starfish. Both kinds of pedicellariae occur on the inferomarginal plates, but the bivalved are much the more common. Large spatulate pedicellariae occur on many adambulacral plates usually at the adoral end between the two series of spines. Groups of 8-20 large bivalved pedicellariae cover the surface of the actinolateral plates and give them a rough appearance ; some of these pedicellariae have jaws more than a millimetre long. The colour of this fine starfish in life is said to have been “buff above, purple star around mouth.” The dry specimen is light reddish-buff above and reddish-white below; the oral plates and first four or five actinolaterals on each side of each ambulacral furrow are dull rose-red, but more or less of the centre of each actinolateral plate is whitish, the area increasing on the more distal plates ; inferomarginal and adambulacral spines white. Off Port Hedland, W.A-. One specimen. No. 4030. Type, This beautiful starfish is quite distinct from any other member of the genus, yet seems to be a true Stellaster. The inferomarginal spines and the numerous sharp conical spinules on the abactinal surface are characteristic, while the adambulacral armature and the coloration add important distinctive marks. OREASTER GRACILIS. Liitken, 1871. Vid. Med. f, 1871, p, 260. This West Australian specimen equals in size that in the British Museum, described by Bell, for R. = 220 mm., but there are only 25 or 26 marginal plates as against 30 in that specimen. There are no data with the present specimen other than the getieral statement that it is from West Australia. OREASTER NODULOSUS. Pentaceros nodiilosus, Perrier, 1876. Arch. Zool. Exp., vol. 5, p. 53. Orfaster Hodulosus, Bell, 1884, Proc. Zool. Soc., London, p. 66. Although this specimen is much larger than either of those mentioned by Bell, for R. — 120 mm. as against 70 in his larger specimen, there are only two more marginal plates, 19 instead of 17 144 ] RECORDS OF W.A. MUSEUM. on each side of a ray. This species is notable, judging from the present individual, for its smooth surface, the plates being flattened, the tubercles rounded and the granulation so even that at a little distance the specimen looks water-worn, an illusion which examination with a lens dispels. The papular areas are small but very sharply defined. There are 14 or 15 tubercles on each median radial ridge ; the largest, which is about 13 mm. in diameter and 8 mm. high, is at the radial angle of the disk, the others being succes- sively smaller distally. There are no data with the specimen, but it is from West Australia. CU LCITASTER, ' gen. nov. Form stellate, but rays short and disk disproportionately large. Marginal plates concealed, except on terminal third of ray. No large terminal marginals. Abactinal skeleton, coarsely reticulate with numerous large papular areas, regularly arranged in sixes or sevens around each plate. Entire animal covered by a closely granulated skin. No large spines or tubercles, Actinal intermediate areas very large covered with a flat pavement of polygonal plate, arranged in very regular series. Bivalved and spatulate pedicel- lariae present, at least actinally. Type species — Culcitastev anamesus sp. nov. This remarkable genus is so perfectly intermediate between Oreaster and Culcita, when seen from above, one might find justification for putting the present species in either of those genera. The actinal surface, however, is more distinctive and makes it desirable, if not positively necessary, to establish a new genus. One can easily imagine the rays of an Oveaster gracilis being shortened and the disk enlarged and puffed out until its appearance would be very much like that of Culcitaster, in its dorsal aspect, and if the process continued until the virtual disappearance of the ray, there would be difficulty in distinguishing such a specimen from Culcita schnideliana. But a glance at the actinal surface would be sufficient to distinguish Culcitaster, for no Oreaster or Culcita, now known, has any such regularly tessellated intermedi- ate areas as characterise this new genus. 1 CiUcitii, a well-known genus of starfishes aster a common termination for starfish genera. RECORDS OF IV. A. MUSEUM. [145 CULCITASTER ANAMESUS, ^ sp. nov. Plate XIX. Rays 5. R.= 185 mm. r — no mm. Br. at base=70 mm. Br. at 20mm. from tip of ray, 28 mm. Disk circular somewhat swollen^ at least 200 mm. in diameter, and projecting beyond superomar- ginals in all interradial areas. Entire animal covered with a closely though rather coarsely granulated skin, which conceals many of the marginal plates but through which most of the abactinal plates are discernible. These plates seem to be rounded or stellate and are united with each other by narrow radiating plates, six or seven to each central plate. All the space between these radiating plates is occupied by papulae, so that the papular areas, each with dozens of papulae, are more or less triangular in shape and are arranged in groups of six or seven around each primary abactinal plate. On the rays, distal to the disk, one can distinguish at least three longi- tudinal series of plates and hence the papular areas have a linear arrangement. Six or eight superomarginal plates on each side of the tip of the ray are easily recognisable; the last four or five are high and narrow and at least on two arms, the two distalmost pairs meet abactinally in the median line. Terminal plates relatively very small. Tip of ray turned up so far that as in Culcita, the ambulacral grooves extend on to the abactinal surface. Actinal surface flat, tessellated, covering-membrane not con- cealing the outlines of the plates. Distal to each mouth angle is a large rhomboidal plate about 18 mm. across; from its two distal sides extend the series of conspicuous actinolateral plates, at first nearly square but soon becoming evidently wider than long. Distal to the large rhomboidal plate is a similar but smaller plate, its two proximal sides in contact with the first actinolateral plates ; from its two distal sides extend series of plates adjoining and parallel to the actinolaterals. Distal to the second rhomboidal plate is a similar but smaller plate from whose distal sides, series parallel to the actinolaterals again arise. Distal to the third rhomboidal plate is a pair of narrow plates lying side by side, from the distal ends of each of which a series of four or five similar plates extends to the margin of the area. Adjoining each of these series are parallel rows which I Gr. anamesos — in the middle ; in reference to its intermediate position. 146] RECORDS OF W.A. MUSEUM. extend from the actinolaterals (beginning with the third) to the margin. At the margin of each interradial area, the regular serial arrangement of actinal plates is interrupted by the intercalation of small rounded or polygonal plates, some of which also crowd in between the lower ends of the inferomarginals. Adambulacral plates short and not very wide ; there are about two to each actinolateral plate. Armature in a double series ; inner of 7-9 rather slender spines, parallel to furrow, median longest and adoral and aboral shortest and subequal ; outer of two or three very short blunt spines about 2 mm. long and 1-2 mm. thick, on actinal surface of plate, parallel to furrow, median (or adoral of two) longest and stoutest. Oral plates with about a dozen spines on each mar- gin ; proximal very large, 7 or 8 mm. long, 3 or 4 mm. thick at tip ; distal spines smaller : distalmost grading into inner series of first adambulacral plate ; each oral plate bears on its surface 3 or 4 very stout, low spines, similar to those of the outer adambulacral series. Pedicellariae abundant on actinal surface, but not observed in this specimen abactinally. On adambulacral plates are very heavy spatulate pedicellariae with two jaws ; these are at adoral end of plate and may be one, two or three in number ; if there are two or three, one is usually much the largest. Scattered all over the actinal interradial areas, but most abundant on the radial ends of the actinolateral plates are bivalved pedicelloriae with jaws a millimetre wide or less. Colour above, light grey ; actinal surface more yellow-brown. There are no data with this extraordinary starfish, but it is said to have been taken on the coast of West Australia. It is difficult to understand how so large and conspicuous a form should have so long been undescribed, but lean find nothing in the literature which would warrant the belief that specimens had ever been seen by European zoologists. The swollen circular disk beyond which project the short stumpy rays give the animal a very odd appear- ance, while the regular actinal plating covered with a closely granulated membrane is remarkably distinctive. There can be little doubt that the genus is intermediate be- tween Oreaster and Culcita. RECORDS OF W.A. MUSEUM. [147 LINCKIA TYLOPLAX, ‘ sp. nov. Plate XX. Rays 5, R. = 150 mm. r. 25 mm. R. = 6r. Br. at base = 27 mm. Br. at half-way point = 18 mm. Disk small, but much elevated vertical diametre about 20 mm. Rays tapering to a blunt point. Whole body surface covered by a granular membrane; granules largest near the centres of abactinal plates smallest on papular areas. Abactinal skeleton composed of 3-5 irregular series of rounded plates, with smaller plates scattered among them. Larger plates, nearly all swollen into rounded or flat-topped knobs, 2-5 mm. in diameter and about 2 mm. high. Between the plates are papular areas, 2-3 mm. across, with -numerous small papulae. Madreporic body large, 7 mm. in diameter, about 15 mm. from centre of disk. Marginal plates fairly distinct, especially near tip of ray. Superomarginals about 43 on each side of a ray, the basal ones at least, knobbed. Inferomarginals of about the same number but less distinctly knobbed. Intramarginal papular areas well developed, nearly to tip of ray. Two series of actinal intermediate plates extend nearly or quite to tip of ray ; actinolateral series adjoining adambulacral plates much larger than the second series, except near tip of arm. Papular areas present on actinal surface not only between inferomarginals and actinal intermediate plates but also between the two series of the latter, even on the actinal interradial areas, Adambulacral plates small ; armature characteristically Linckian ; each plate bears two short, thick, blunt spines on its furrow margin and a larger tubercle-like spine, 2 mm. long by i mm. thick, on its actinal surface. Armature of oral plates similar and equal to that of two adambulacrals. No pedicellariae. Colour, light brown above, darkest on knobs ; actinal surface nearly white. Between Fremantle and Geraldton, W.A., 80-120 fms. Two specimens. No. 4931. The larger specimen is the Type. The presence of actinal papulae would almost warrant a new genus for this interesting starfish. Its abactinal aspect is much like some species of Nardoa, but the adambulacral armature pre- cludes its inclusion in that genus. It does not seem to be very nearly related to any other previously known member of the family. 1 Gk. Uidos — a knob ; plax — a plate. 148 ] RECORDS OF W.A. MUSEUM. ASTERINA GUNNII. Gray, 1840. Ann. Mag. Nat. Hist., vol. 6, p. 289. The occurrence of this species on the West Australian coast is quite to have been expected. All the specimens have six rays. From piles of old jetty, Fremantle, W.A. Two specimens. No. 6044. Without data, two specimens in poor condition. Nos. 146 and 148. ANSEROPODA ROSACEA. Asterias rosdco/s, Lamarck, 1816, Anim. s. Vert. vol. 2, p. 558, par. 3. Anserofoda rosacea, Fisher, 1906. Bull. U.S. Fish. Comm, for 1903, p. 1089. This is one of the notable starfishes in the collection, for its occurrence off West Australia is very interesting, even if not sur- prising. The present specimen is 168 mm. across, and thus some- what smaller than the specimen described by Muller and Troschel (1842) but larger than the one so finely figured by Koehler in his account (igio) of the shallow-water starfishes of the Indian Museum (PI. XX). The individual from West Australia is, in its dry condition, dull, deep cream-colour with not very numerous, well scattered small spots of deep purple on the upper surface. It is remarkable for having 16 rays instead of the typical number, 15. From Port Hedland, W.A. No. 4029. ECHINASTER ARCYSTATUS, ^ sp. nov. Plate XXI. Rays 5. R.=:I30 mm. r."2o mm. R.— 6.5r. Br. at base =24 mm. Br. at half-way point— 18 mm. Disk small ; vertical diameter about 20 mm. Rays rounded, tapering to a rather blunt point. Abactinal skeleton and that of sides of rays forming a very distinct net- work with meshes 4-10 mm. in diameter, and occupied by 10-60 papulae. The skeletal ridges carry numerous, but well- spaced, bluntly pointed spinelets about a millimetre high. These 1 Gr. arhustatos — surrounded with nets, in allusion to the conspicuous reticu- lations of the abactinal skeleton. By an unusual typographical error in Hinds and Noble’s Classic Greek Dictionary, 1901, p. 102, I was led to write the word acrystata in naming a brittle star in 1911, Bull. U.S. Nat. Mns. No, 75, p. 145. Under the circumstances, the name given is obviously a typographical error and the brittle-star should be known as Amphiurct arcysiata. RECORDS OF IV. A. MUSEUM. [149 are most numerous and sharpest near tips of rays. Madreporic plate small, only 3 mm. across, and situated only 7 or 8 mm. from centre of disk. Adambulacral plates short and numerous ; each bears a pointed spine, deep in the furrow, and two blunt spines on the furrow margin ; these two spines are 1.5-2 mm. long, about .5 mm. wide and are somewhat flattened ; the aboral is a trifle smaller and stands further back from the furrow margin. Oral plates each with four marginal .spines similar to those of the adambulacral plates, but somewhat larger. Actinal surface with numerous papulae everywhere. Actinal intermediate areas without spines. Opposite the fifth adambulacral plate, there begins a very distinct series of small sharp spines which runs to the tip of the ray. The fifth adambulacral spines are 5 mm. from the first spine of this series, but at the tip of the ray the distance between the two series is little more than a millimetre. The whole starfish is covered by a skin which is particularly thick and noticeable orally. On the basal half of the arms, very distinct furrows run out at right angles to the long axis of the ray between the adambulacral plates for some distance beyond the actinolateral series of spines just described. In the dry specimen these furrows are very conspicuous because of their lighter colour. There ate of course no pedicellariae. Colour of dry specimen, reddish-brown. In life the colour is said to have been purple. Between Fremantle and Geraldton, VV.A. One specimen. No. 4918. Type. If we are to distinguish Othilia as a separate genus from Echinaster, on account of the actinal papulae, then this species belongs to Othilia, for the actinal papulae are a very noticeable feature. But the type of Othilia is Edhinaster spinosus of Brazil, and the present species does not resemble that species in general appear- ance so much as it does some of the East Indian Echinasters. For the present therefore, I have concluded to ignore Othilia and describe this specimen as an Echinaster. It does not seem to be very closely related to any of the previously known species, but it must be granted that Echinaster is a perplexing genus and specific differentiation is not very complete within it. Individual variation RECORDS OF W.A. MUSEUM. 150] is considerable and there are few characters which seem well to separate the species. A revision of the described species is much needed. ECHINASTER VESTEIUS. Ophidiaster {}) vestUus, Perrier, 1869. Arch, Sci. Nat., vol. 12, p. 254. Echinastey vssiitus, Perrier, 1875. Arch. Zool. Exp,, vol. 4, p. 372. This specimen has been compared with specimens from Mauritius and Zanzibar labelled vestitus by Perrier himself, but owing to its poor condition, due to a peculiar scurfy encrustation all over the surface, I am not wholly satisfied that it is identical with them. There are 5 rays about 120 mm. long, and nearly 20 mm. in diameter at base ; they are nearly cylindrical but taper somewhat to a blunt point ; r^r6 mm. The abactinal surface and sides of the rays are covered with papulae and numerous small spines, while the skeleton itself is pretty well concealed by the over- lying skin. The spinelets are rarely a millimetre long. The adam- bulacral plates carry only two spines each, one deep in the furrow and a larger blunt, or even slightly clavate spine on the margin. Papulae are not present below what seems to be an inferoraarginal series of spines ; between this series and the adambulacrals there is a more or less well-marked series of somewhat smaller actino- lateral spinelets. Colour, bright yellow-brown. Off Port Hedland, W.A. One specimen. No. 4031. The peculiar encrustation on this specimen prevents a satisfac- tory study of its characters. The nature of this encrustation I have not been able to determine, but it appears to be organic. PLECTASTER DECANUS. Echinasier decauns, Muller a.n 6 Troschel, 1843. Arch, f, Naturg, Jhrg. g, vol. i, p, 114. PUctaster decanns, Sladen, 1889. Challenger Asteroids, p. 535. The occurrence of this species on the south-western coast of Australia is not surprising, though it is a considerable extension of its known range. Albany, W.A. One specimen, in poor condition. No. 4859. RECORDS OF W.A. MUSEUM. [isr ASTERIAS CALAMARIA. Gray, 1840, Ann. Mag. Nat. Hist., vol. 6, p. 179. This species has not been recorded from West Australia hitherto, though its occurrence there is quite to have been expected if Gray’s original specimens were really from such widely separated places as Mauritius and Australia. No data. One specimen, No. 133. ASTERIAS POLYPLAX. Asteracanthioji polyplax, Muller a.n6 Tioschel, 184^. Arch. f. Naturg. Jhrg. 10, vol. I, p. 178. Asterias polyplax, Perrier. 1875. Arch. Zool.'.Exp., vol. 4, p. 327. This is one of those perplexing starfishes, which it is almost impossible to distinguish from Asterias, in a broad sense, and which nevertheless seems to belong in the Stichasteridae if that family is to be recognised. As I very much question the validity of this latter family, it seems to me better to keep polyplax m the genus Asterias, until that assemblage of species is properly broken up into its component parts. As both Verril and Fisher are now at work on this problem, the name Asterias polyplax may well be used for a few years more for the present Australasian species. Between Fremantle and Geraldton, 80-120 fms. One speci- men, No. 4917. 152 ] RECORDS OF IV. A. MUSEUM. OPHIUROIDEA. Although four of the nine species of ophiurans appear to be undescribed and one of these requires the formation of a new genus, the collection from the Western Australian Museum is remarkable for what it does not contain, to nearly as great a degree as for what composes it. The ophiurans of West Australia have recently been the subject of a paper by Koehler i based on the collection made by Michaelsen and Hartmeyer in 1905. That collection contained 28 species, of which only three were considered as previously undes- cribed ; one of these was an Ophiacantha and the other two belonged to Ophiothrix, both large, diversified and widespread genera. A fourth species was subsequently described as new by Doderlein an astrophyton of the genus Astroboa. Of the 28 species, only 3 are in the collection before me. Of the 14 genera collected by Michaelsen and Hartmeyer, only 4 are in this collec- tion; such common genera as Amphiura, Ophiactis, Ophionereis, Ophiocoma and Ophiacantha are entirely unrepresented. On the other hand the two collections contrast with each other sharply in the fact that only one of Michaelsen’s and Hartmeyer’s species was a Euryalid while fi ve of the nine species before me represent that group. It would be hard to collect two series of Ophiurans from the same region which would difler more strikingly from each other than do the two under discussion. Nor is it easy to suggest any reason for such an extraordinary difference. Possible further studies, made on the ground, may exphiin the matter.® PECTINURA DYSCRITA. H. L. Clark, igog. Thetis Echinoderms. Mem. Aust. Mus., vol. 4, p. 534. This species, previously known only from New South Wales, is represented by a single specimen in poor condition. Between Fremantle and Geraldton, W.A. One specimen. No. 4929. 1 Koehler, 1907. Die Fauna Siidwest-Australiens ; Ophiuroidea, vol. i, pp. 241-254. 2 Doderlein, T911. Tiber japanische nnd andeie Euryalae, p. 82. 3 The explanation of the difference between these two collections appears to be simple. The collection sent to Mr. Clark was entirely composed of speci- mens trawled by the Endeavour , chiefly in depths approaching too fathoms. Drs. Michaelsen and Hartmeyer obtained their specimens by dredging in comparative!}' shallow water in shelten d localities, — W. B. A. RECORDS OF IV. A. MUSEUM. • [153 OPHIOTHRIX SPONGICOLA. Stimpson, 1855. Proc. Acad. Nat. Sci. Philadelphia, vol. 7, p. 385, This is another of the New South Wales species, which might naturally be expected in West Australian waters. P'or an account of its more recent history see Thetis Echinoderms, Mem. Aust. Mus. vol. 4, p. 546. Between Fremantle and Geraldton, W.A. One specimen in poor condition. No. 4930. OPHIOTHRIX STELLIGERA. Lyman, 1874. Bull. M.C.Z., vol. 3, p. 237. This species, which was taken by Michaelsen and Hartmeyer at five stations, is represented by a rather large specimen, with disk 8 mm. across, and the most nearly complete arm from 32-35 mm. long. It is in fairly good condition. It seems to have been taken with the previous species, as it was in the some vial and bears the same catalogue number. Between Fremantle and Geraldton, W.A. One specimen. No. 4930. OPHIOMYXA AUSTRALIS. Liitken, i86g. Add. ad Hist. Oph., pt. 3, p. 45. This is the second of the three species in this collection, which were also taken by Michaelsen and Hartmeyer ; they found it at four stations. It was also taken by the Thetis and by the Challenger, so it may naturally be considered one of the commonest Australian brittle stars. Between Fremantle and Geraldton, W.A. One specimen in poor condition, No. 4928. ASTROGYMNOTES, ^ gen. nov. Disk and arms covered with a skin, which apparently contains no calcareous plates, except numerous rounded granules on the upper surface of the arms. No radial shields on upper or under arm-plates are visible. No madreporite. Teeth well developed, but oral papillce rudimentary. Arm spines and tentacle scales both present and easily distinguishable from each other. r Gr. aster — a star; gymtwtes — nakedness, in allusion to the absence of plates on disk and arms. 154 ] RECORDS OF W.A. MUSEUM. Type species, Astrogymnotes catasticta, sp. nov. This interesting new ophiuran, noticeable for being jexamerous, appears to be one of the sub-family Astroscheminae as restricted and defined by Doderlein in igii. But it is easily distinguished from the other members of that family by the absence of both upper and under arm plates, and the presence of both tentacle scales and arm spines. ASTROGYMNOTES -CATASTICTA, ' sp. nov. Plate XXII. Rays 6, rarely 7. Disk about 9 mm. across ; rays about 45 mm. long. Entire animal covered by a smooth skin, which is perfectly bare, except on the upper surface of the arms and the adjoining portions of the disk, where it is more or less crowded with minute circular bits of lime ; about a dozen series of these occupy the width of the arm. Radial shields not visible but indicated through the skin by short radial elevations, a pair at the base of each arm ; these are about 3 mm. long. No upper or under arm- plates to be seen even in a dried specimen. Teeth well formed, in a vertical series of five ; oral papillae rudimentary, about three on each side of each jaw. No madreporic or oral shields present. Arm-spines present, one on each of the three basal segments, but two on each of the remaining segments ; they are less than a milli- metre long, cylindrical, blunt and subequal. Tentacle-pores evident, each one guarded by a flat, nearly circular tentacle scale. Genital slits small and oblique, about equal to the length of an arm-joint ; two in each interradius. Colour yellow-brown, indistinctly speckled on radial areas of disk, and conspicuously spotted all over the lower surface of disk and arms with pale yellowish white. In dry specimens the colours are dull, the spotting is not so noticeable and the cal- careous granules give the upper surface of the arms a whitish cast. Off Jurien Bay, W.A., 80-100 fms. Ten specimens. No. 4,924. It is a real satisfaction to have such a good series of this remarkable ophiuran for study. It is probably as nearly related to Astroschema as to any known genus, but still retains a distinc- tion between arm spines and tentacle scales. The arms are also 1 Gr. catasiictos — spotted. RECORDS OF W.A. MUSEUM. [155 relatively much shorter than is usual in that genus. Several of the specimens show by their unequal development that reproduction l>y fission is normal for the species. OPHIOCREAS MELAMBAPHES, ^ sp. nov. Kays 5. Disk about 6.5 mm. in diameter, with arms about 120 mm. long and 2 mm. thick. Entire animal covered with a thick smooth skin, through which, in dried specimens, the radial shields and side arm-plates are revealed. The radial shields are about 3 mm. long by one 'mm. broad, and are perfectly smooth, though slightly swollen. There are no calcareous granules in the skin of the dorsal surface on either disk or arms. Teeth 4 or 5, the lowest often displaced and somewhat deformed. Oral papillae wanting. No oral shields or madreporite. First two pairs of the very small tentacle-pores bare, but all subsequent pores guarded by two spine-like tentacle-scales of which the inner is a trifle the longer and is about half a millimetre long. Genital slits very small, less than a millimetre long, oblique, cresentic ; two in each interbrachial space. Colour, deep purple or purplish black above and on sides, but actinal surface of disk and arms dull cream colour. Off Jurien Bay, W.A., 80-100 fms. Two specimens. No. 4,925. The smaller specimen is the Type. Although this species resembles 0 . silogae, Koehler, in many characteristic features, it is distinguishable from that species at a glance by its different proportions, different tentacle-scales and strikingly different colour. Doderlein does not consider the charac- ter by which Astroschema differs from Ophiocreas sufficiently con- stant to warrant the maintenance of the two genera, but I am inclined to think that while our definitions will need to be readjusted, the two groups had better be retained. The present species is a typical Ophiocreas so far as its external covering is concerned. 1 Gr. melambaphes — dark-dyed. 156 ] RECORDS OF W.A. MUSEUM. OPHIOCREAS RHABDOTUM, ’ sp. nov. Rays 5. Disk 5 mm. in diameter, with arms about 78 mm. long. Very similar to the preceding species, but arms shorter and more slender, radial shields shorter and less prominent and teeth sharper and more regular. Only the first pair of tentacle-pores lacks tentacle-scales. Colour dull yellow, speckled and streaked on disk with blackish : a broad blackish stripe runs the length of the arm in the median line abactinally, but under the lens, even this stripe is found to be finely speckled with yellowish ; just alrove the tentacle-scales there is on each side of the arm, a very narrow and often indistinct blackish stripe. Off Jurien Bay, W.A., 80-100 fms. Two specimens. No. 4,926. The larger specimen is the Type. As this form was taken at the same station with the preceding, I think it quite possible that it is only a colour phase of that species. But the colour difference is so marked, and the two specimens of each species are so distinctly characterised thereby, I have felt it was more satisfactory to designate them by different names. Further investigation on the West Australian coast is necessary for a correct settlement of the question.'* CONOCLADUS MICROCONUS. ^ sp. nov. Plate XXV, Rays 5, but in the type specimen there are 6, Disk, 35 mm. in the type, in the smaller specimen 27 mm. in diameter, with arms about go or 100 mm. long and 10 mm. wide at base; height of arm near disk, 5-6 mm. Radiating wedges of disk separated from each other by five (in the type, six) narrow deep grooves, which are 1 Gr. j'Riitotos — striped. 2 The two supposed species of Ophiocreas were obtained together m large quantities attached to one particular species of Hydroid, with which their arms were so much intertwined that it was very difficult to remove them, except in fragments. Whilst I entered them under two numbers, I made a note at the time that they were “ probably colour varieties of one species ” — VV. B. A. Gr, mihfos — little ; konos — cone ; in reference to the small size of the cones on the disk. RECORDS OF W.A. MUSEUM. [157 practically filled up by the small, rough cones of the disk plates. Each wedge is covered by a rough, uneven pavement of plates, grains and granules, rarely smooth, but usually bearing a ridge, lump or small cone. These cones are of very diverse sizes and shapes ; they are rarely a millimetre high and are very seldom pointed; they commonly terminate in a group of 3-15 little spinules. This irregular and very rough pavement extends out on the area so that there is no line of division between the latter and the disk. None of the cones are enlarged nor have they any defi- nite arrangement. The arms branch about eight times ; the first division being about 22-27 rnm. from the disk ; the resulting branches after each division are often very unequal. Beyond the second fork the branches become long and very slender and are covered by alternating paired half-circles of granules and glassy booklets. Oral surface of disk and arms covered with small roundish flat granules, most numerous on the mouth angles and bases of the arms. Tentacle-pores small ; first pair (not counting buccal tentacles) well within disk, with no tentacle-scales. Each succeeding pore is more or less concealed by a slight ridge on its adoral side, which carries 4 (rarely 3 or 5) short, slightly curved peg-like spines, rather more than half a millimetre long. Each spine is compressed and its terminal margin divides into 3-5 little glassy spinelets. Each mouth angle carries a cluster of twenty or more spiniform teeth and similar but shorter papillae. Genital slits small, hardly 2 mm. long, and more or less concealed. Madre- poric plate distinct, hardly 2 mm. across. Colour, uniform light brown ; in life, “ dull brown.” Between Fremantle and Geraldton, W.A., 80-120 fms. Two specimens, No. 4921. The larger specimen, although 6-rayed, is selected as the Type. The occurrence of a new species of Conocladus in West Aus- tralian waters is most interesting, the two previously known species having been reported only from New South Wales. As Doderlein figii, liber Japanische und andere Euryalae, p. 68) has pointed out Conocladus is in certain respects a very primitive form mostly nearly allied to Astroconus anstralis, Verr. This discovery of a third well-marked species would seem to indicate that Australia is emphatically the home of these primitive Eurylids and even sug- gests that it may have been the ancestral home of the whole order. 158 ] RECORDS OF W.A. MUSEUM. The West Australian species is very easily distinguished from its New South Wales congeners by the entire absence of large cones or tubercles on the disk. ASTROBOA ERNAE. Doderlein, igir. Uber japanische und andere Euryalae, p. 82. It is interesting to find specimens of this species in the collec- tion since it was hitherto known only from the holotype, a specimen 22 mm. across the disk. One of the present specimens is only 15 mm. across the disk while the others are nearly 40. But I have nothing to add to Doderlein’s careful description. The colour of these individuals in life is said to have been “ buff or leaden.” In alcohol, they are a peculiar shade of purplish brown, which is hard to name. On drying, they become very light, almost a dirty white, with a purplish cast. Off Geraldton, W.A., 29 fms. Three specimens. No. 4923. ECHINOIDEA. Although the collection of Echini is remarkable for the large number of species it contains, yet eleven species which were in the Thetis collection from New South Wales are not represented here. Several of these are common littoral forms of wide distribution and will probably be found hereafter on the West Australian coast. As already noted, four and possibly five of the nineteen species are new to science, and belong in groups of more than usual palaeontologi- cal interest. PHYLLACANTHUS MAGNIFICUS, ’ sp. nov. Plate XXVI. Test nearly spherical, 92 mm. in horizontal diameter and 72 in vertical ; hence v.d.=.78 h.d. Longest primary, about 75 mm. in length, 10 mm. in diameter near base and 5 mm. in diameter at tip. Interambulacral plates 10 in each column, all, except sometimes, the uppermost with a long,' stout primary spine. Interambulacra 1 The origin and significance of this name are obvious. RECORDS OF W.A. MUSEUM. [159 43 mm. wide ; ambulacra nearly 13 mm. ; hence interambulacra more than three times as wide as ambulacra. But median interam- bulacral space only ii mm. wide, and hence only one-fourth of interambulacrum. Median ambulacral area less than 6 mm. wide, and hence less than half the ambulacrum. Abactinal system 28 mm. in diameter and actinostome the same. Genital plates moderately large, about ii mm. wide by 7 mm. high, but madreporic genital very large, nearly 16 mm. wide by ii mm. high. Oculars small and low ; ocular V insert, and I nearly so ; in the smaller specimen, all the oculars are exsert but V is nearly in. Genital pores large, near centre of plate, surrounded by a circle of about a dozen broad flat spinelets larger than the others on the plate ; as these spines are closed over the pores, they form a con- spicuous conical elevation. Ocular pores small near distal margin of plate. Interambulacra, median areas of ambulacra, abactinal system and actinostome, densely covered with small flattened bluntly pointed spinelets and pedicellariae. Secondary spines of interambulacra in circles of 15-20 around the bases of the primary spines ; they are broad, flat and truncate, about 8 mm. long by 2 mm. broad. Along the margin of the median area of each ambula- crum is a series of narrow, flat spines, about 4 mm. long and .5 mm. wide, which lie flat across the poriferous areas. Between these marginal series are four or five series of smaller spines or spinelets similar to those on the interambulacra. Pores of a pair rather large wider than high, separated by a space wider than the width of a pore and connected by the groove characteristic of the genus. Pedicellariae present in great numbers. The tndentate show little diversity in size, but are chiefly actinal in position and on the interambulacra. Their valves are about two millimetres long, but the blade is only about .25 mm. wide, so they are very slender ; the valves meet only at or near the tip. There is a single vertical ridge extending the length of the blade in the median line on its inner surface similar to those found in the tridentate pedicellariae of P. iinperialis, but rather more prominent ; it is sharply serrate, more coarsely so than the margins of the blade. The small globifevoiis pedicellariae are abundant almost everywhere and are very similar to those figures by Mortensen for P. Imperialis. Their valves are about .30 mm. long. The large globiferous pedicellariae are very common, particularly on the ambulacra ; the valves are .80-. 90 mm. RECORDS OF W.A. MUSEUM. i6oJ Ion", but the stalk is little more than half that ; it has no “ limb.” The valves are shaped very much like those of the small ^lobiferous pedicellariae, but the terminal opening is surrounded by coarse, curved teeth. This opening is very variable in size and form ; it may be less than one-fourth the length of the valve or it may be nearly two-fifths ; it is often of some peculiar shape and not rarely is divided vertically into two openings ; even when the opening is short there is little tendency towards a “snout-like” blade. The large globiferous pedicellariae vary very little in size and do not seem to intergrade at all with the small ones. Primary spines very stout, rough with small rounded granules, which soon become more or less completely concealed by a spongy alteration in the outer layer of the spine and the profuse growth of bryozoa, sponges and other organisms which cover the old spines. At the tip, the series of rounded granules are arranged longitudi- nally and pass into ridges which surround the blunt end of the spine. Young spines are bluntly pointed but become more and more flaring with age and the actinal primaries are particularly notable for their stoutness and flaring tips. The thickness of some of these at tip may be equal to one fourth or even almost one-third the total length of the spine. There are 24-36 longitudinal ridges around the tips of the larger primaries. The collar is low, seldom over 3 mm. in height. Colour, deep red-brown, so far as secondaries, miliaries and pedicellariae ar-e concerned ; young primary spines yellow-brown, but rapidly darkening with age ; the collar remains yellow-brown throughout life, but the rest of the spine, where not covered by foreign growths, has a marked purplish-red shade. Between Fremantle and Geraldton, W.A. Two specimens, No. 4935- The larger is the Type. The type specimen of this fine new species is one of the largest and most perfectly preserved cidarids I have ever seen. The other specimen is 75 mm. h.d. and 51 v.d., so that v.d. is little more than .66 h.d. There are 8 or g interambulacral plates in each column. In other particulars the specimens differ little from each other While the relationship to imperialis is obvious, this species is easily distinguished by the remarkable actinal primaries and the large number of coronal plates. To no other of the Recent species of the genus does it show any close relationship. RECORDS OF W.A. MUSEUM. [i6i PHYLLACANTHUS ANNULIFERA. CidaHtes annuUfera, Lamarck, i8i6 Anim. s. Vert., vol. 3, p. 57. Phyllacanthus annuUfera, A. Agassiz, 1872. Rev. Ech., Pt. i, p. 150. There is a beautiful cidarid in the collection which seems to belong to this species although the coloration is different from that of any I have seen. The test is 30 mm. i.d., and the primaries, which are very thorny and quite free from foreign matter are about 40-45 mm. long ; the collar is 4 mm. high. The ocular plates are all insert, but II and III are only barely so. The test is cream-colour, or almost pure white on the bare median ambulacral area ; the miliary and secondary spines are almost white, or at least very light coloured, with a broad longitu- dinal stripe on their upper (or outer) surface of bright brown, which is darkest on the smallest spines and palest on the largest ; actinally the brown is almost brownish-red. The primary tubercles and the collars of the primaries are bright pinkish lavender. The primary spines are light coloured with a markedly greenish cast, and with almost eight bands of dull purplish-red ; these bands are broken on the sides of the spine and are faint on the lower surface ; the thorns on the actinal primaries are red, often with white tips, and the primaries close to the actinostome have the entire tip red. Port Hedland, W.A. One specimen, No. 4026. TRETOCIDARIS BRACTEATA. Dorocidaris bracleaia, A. Aga§siz, 1879. Proc. Amer. Acad., vol. 14, p. 197. Tretocidaris bracteata, H. L. Clark, 1907. Bull, M.C.Z., vol. 51, p. 206. These specimens are larger than any previously known, having h.d.=35 mm. The primaries are 50 mm. or more in length, and are 4 mm. broad near the base ; they are somewhat flattened there and the longitudinal ridges are very prominent ; these latter are broken up into flattened truncate or sharp teeth, making the spines conspicuously and coarsely rough. The collar on the primaries is very low, which is one of the best characteristics of the species. Another useful character is the spotting of the primaries, at least near base, with longitudinal series of red-brown dots. The pink secondary spines are also a noticeable character, those of the inter- ambulacra in these specimens being evidently tipped with yellowish. The large globiferous pedicellariae are very uncommon but are to RECORDS OF W.A. MUSEUM. 162] be found abactinally. The heads are less than a millimetre in length, but the stalk is nearly two millimetres ; the latter has a con- spicuous “ limb,” the branches of which are about .2 of a millimetre long. The valves resemble closely those of T. affinis. The triden- tate pedicellaria are very slender, the valves, which are hardly .05 mm. wide, range from .9 to 1.25 mm. in length and meet only near tip. The small globiferous pedicellariae have valves ranging from .25 to .55 mm. in length and are provided with an end tooth. Between Fremantle and Geraldton, W.A., 60-100 fms. Two specimens. No. 4,933. The occurrence of this distinctly East Indian species off West Australia is notable and hardly to be expected. These large speci- mens bear a striking superficial resemblance to Phyll acanthus annul- fem, but even hasty examination distinguishes them. The low collar and the red-brown spots, not to mention the pedicellariae, are sufficiently marked differences to be obvious to even a casual observer. GONIOCIDARIS TUBAKTA. Cidariks lubariii, Lamarck, 1816. Anim. s. Vert., vol. 3, p. 57. Gonioiidaris tnharia, Liitken, 1864. Vid. Med. f, 1863, p. 137. The occurrence of this species on the west coast is interesting, even though the specimens are small (20 mm. h.d.). Between Fremantle and Geraldton, W.A., 100 fms. Two specimens, No. 4,938. CENTROSTEPHANUS TENUISPINUS, > sp. nov. Plate XXVI. This species is so nearly related to C. rodgersii of New South Wales that an extended description would be superfluous. The test is essentially alike in the two species, but there are more coronal plates in the one from West Australia. The actinostome is notice- ably smaller and the abactinal system is somewhat larger. Thus in an eastern specimen, 84 mm. h.d., there are 16 interambulacral plates in each colnmn, the actinostome is 36 mm. across and the abactinal system is 18 mm., while in a western specimen of the same size, there are 18 interambulacral plates in each column, the 1 Gr. tenuis = slender + spinus = a spine. RECORDS OF V/.A. MUSEUM. [163 actinostome is 32 mm. across and the abactinal system is 24 mm. When other adult specimens (i.e., over 60 mm. L.d.) are taken into account, we find that these differences sum up to about this; in the eastern species the abactinal system is .50-. 60 of the actinostome, while in the western form it is .65-. 75. The most obvious character of the new species, however, is seen in its slender spines ; the largest primaries are from 1.3 to 1.6 mm. in diameter where thickest (near base) and as they are 75-80 mm. long, they appear very slender as compared with rodgersU ; in rcdgenii the primaries are 2-3 mm. in diameter, at least near base, and seldom exceed 75 mm. in length. This difference in the primary spines gives the two species totally different facies. There is also a noticeable difference in colour, in the specimens I have seen, though it may not prove a constant one; in rodgeysii, the colour is more or less purple, sometimes almost black, again deep crimson rarely brownish-red; in tenuispinus, the colour is reddish-brown or dull greenish, with no trace of purple. The pedicel lariae in the two species do not seem to differ ex- cept in relative frequency. Thus in specimens of rodgeysii, slender tridendaie pedicellariae seem to be very rare. Mortensen did not find them, and I have only found one on six specimens from New South Wales. But in tenispinns they are very common all over the test and the valves may exceed 3 mm. in length. On the other hand, the stout tridentate pedicellariae with curved valves seem to be rather uncommon in tenuispinus. Between Fremantle and Geraldton, W.A. Two specimens. No. 4,936. The light-coloured one is the Type. It is of course possible that larger series of specimens than are available to me, will show that this supposedly new species is only a form of rodgersii, but the general appearance is so different, I have little hesitation in giving it a new name. And I am confirmed in this by finding that the specimens of C. rodgersii recorded by A. Agassiz (Rev. Ech. pt. i, p. 98) from “ Houl man’s Abrolhos” are the slender spined western form and not typical rodgersii-, at any rate this is true of the specimen in the M.C.Z. collection. This specimen is young (only 47 mm. h.d.) but when conrpared with a specimen from Port Jackson, 43 mm. h.d-, its smaller actinostome and its much more slender primary spines are noticeable. 164 ] RECORDS OF JV.A. MUSEUM. SALMACIS ALEXANDRI. Bell, 1885. Proc. Linn. Soc. N.S.W., vol. g, p. 505. A very fine specimen of this species is in the collection from off Geraldton, W.A., 29 fras., No. 4932. It is deep rose-purple, the spines tipped with white. SALMACIS SPHAEROIDES. Echhuis sphaeroides, Linne, 1758. Sys. Nat. ed. 10. p. 664. Salmaceris sphaeroides, Loren, 1887. Ech. Linn. p. 69. There are two fine specimens (Nos. 4027 and 4028) from Port Hedland, W.A. ; in one the test has a greenish cast and the bases of all the spines are conspicuously dark green, in the other the green colour is confined to the spine bases and is reduced to a mini- mum there ; as a result of this seemingly slight difference the speci- mens look quite unlike. TEMNOPLEURUS, sp. ? There are a couple of bare tests of a tenmopleurid (No. 5007) in the collection, with no data other than that they are from Fre- mantle Beach, West Australia, which cannot be referred to any known species. I think they undoubtedly represent a temnopleurus and probably an undescribed species, but I cannot see that any- thing is gained by giving them a name. They are 24 or 25 mm. h.d. and 13 or 14 v.d. The colour is dull green, with the median ambul- acral and interambulacral areas cream-colour, gradually widening actinally, so that the whole lower surface is of that light shade. The base of one primary spine remains attached to the test, and it is pale red in colour. Until the spines and pedicellariae can be examined, this species may well be nameless. AMBLYPNEUSTES GRANDIS. H. L. Clark, 1912. Mem. M.C.Z., vol. 34, p. 329. There is a single specimen (No. 4932) in the collection. It is about 60 h.d. and has lost most of its spines. It is from off Geraldton, 29 fms. RECORDS OF W.A. MUSEUM. [165 AMBLYNEUSTES GRISEUS. Echinus griseus, de Blainville, 1825. Diet. Sci. Nat. Oursin, vol. 37, p. 81. Amblyoneusles griseus, L. Agassiz, 1841, Intro. Mon. Scut., p. IX. A rather small specimen (No. 5008) of this species is in the collection. From Fremantle Beach, W.A. HOLOPNEUSTES POROSISSIMUS. L. Agassiz and Desor, 1846. Ann. Sci. Nat. (3) vol. 6, p. 364. A specimen from Fremantle Beach, bearing the same number as the specimen of the preceding species from the same place (viz. 5008), proves to be one of this species. HELIOCIDARIS ARMIGERA. Sicongylocenirotus armiger, A. Agassiz, 1872. Bull, M.C.Z., vol, 3, p. 55. Heliocidaris armiger, H, L. Clark, 1912. Mem, M.C.Z,, vol. 34, p. 350. Although the primary spines in these specimens are not quite so stout as in the type specimen, they are sufficiently so to distin- guish them at a glance from their nearest ally, H . erythyogramma. None of the specimens is large (h.d. ranges 25-38 mm.) but all seem adult. The longest primaries do not exceed 15 mm., and their thickness is from 1.5 to 2 mm. Fremantle Beach, W.A. One specimen. No. 5006. Fremantle, W.A. One specimen. No. 149. Cottesloe Beach, W.A. One specimen. No. 5036. HELIOCIDARIS ERYTHROGRAMMA, Echinus cry thvogrammus, Valenciennes, 1846. Voy. Venus,, Zooph pi. VII, fig. i. Heliocidaris erythrograiuma, L. Agassiz and Desor, 1846 (3) Ann. Sci. Nat., vol. 6, p. 371 {eurythrogrammus lap. cal.) Although these specimens are a little larger than those of armigera, the spines do not exceed 15 mm. in length, but few of them are as much as i mm. in thickness, so that they seem longer than they are. There are no data with the two specimens, but they bear the numbers 171 and 179, and are said to be from West Australia. ECHINOMETRA MATHAEI. Echinus mathaei, de Blainville, 1825. Diet. Sci. Nat. Oursin, vol. 37, p. 94. Echinometra mathaei, de Blainville, 1830, Diet. Sci, Nat., Zooph., vol. 60, p. 206. The two specimens of this very common and widely distributed sea-urchin have no data with them. They bear the numbers 165 and 167 and are said to be from West Australia. RECORDS OF W.A. MUSEUM. i66J CLYPEASTER TELURUS/ sp. nov. Plate XXIII. Test broadly oval, widest posteriorly, very flat, 96 mm. loop, 89 mm. wide and 13 mm. high ; its breadth is thus nearly .93 of its length, while its height is less than .14. Test thin and fragile, its margin only 3 mm. thick. Abactinal system at centre of test, but the latter slopes more abruptly posterior to the apex than it does anteriorly or laterally. Posterior interradial margin distinctly depressed below posterior radial margins. Lower surface slightly but very uniformly concave : the slope begins very near or at the margins and the mouth which is perfectly central, is nearly 4 mm. below (i.e. above, of course !) the lateral margins. Petaloid area 49 mm. long and 24 mm. wide. Anterior or unpaired petal 26 mm. long, 12 mm. wide, rather broadly open at distal end, the converging poriferous areas each about 2 mm. wide. Anterolateral petals 22 mm. long, ii mm. wide, and nearly closed (open by i mm.) poriferous area about 2 mm. wide. Posterolateral petals 23 mm. long, 12 mm. wide, well open (by 3 mm.) : poriferous area 2 mm. wide. Ridges between pore-pairs of unpaired petal, each with a single series of six or more primary tubercles. Median area of petals not at all abovate but as wide at middle as anywhere. Anal system, trigonal with rounded angles or oval, about 4.5 mm. broad by 3.75 mm. long ; its distal margin 10 mm. from distal mar- gin of test, or more than .20 of the long radius. Madreporic body small, only 2,5 mm. across. Genital and ocular pores indistinguis- able. Auricles well developed, well separated. Abactinal primary spines very small. Scarcely a millimetre long, thickened at tip. Actinal primaries 3 mm. long (those about mouth 4 mm.) terete and bluntly pointed. Miliary spines minute and abundant, slightly thickened at tip. Pedicell- ariae rather scarce, except around mouth and anus where tridentate are fairly common. Valves of tridentate rather stout, the blades broad, meeting only at tip ; in the largest ones seen the valves are about .30 mm. long. Ophicephalous pedicellariae very scarce, small but not peculiar. Colour of test, abactinally, dull purplish brown of a light shade ; actinally the test has a slight greenish cast; abactinal spines under 1 Gr. telouros — remote ; in reference to the unusual distance of the anal system from the margin of the test. RECORDS OF W.A. MUSEUM. [167 a lens, nearly white ; actinal spines whitish, the largest ones with a faint broad band of purplish, near middle. Between Fremantle and Geraldton, W.A. One specimen, No. 4937. Type. This interesting new species is in many particulars like C. rotundus, A.Ag. from the west coast of Mexico and Central America, but there are some important differences. Chief of these is the position of the anus which in is rarely more than 3 mm. from the margin of the test. In rotundus, too, the poriferous areas of the petals are much wider in specimens of the same size ; thus in a specimen of rotundus of the size of the telums above described, the poriferous area would be about 3 mm. wide, or nearly 50 % broader. The test is much more fragile and the margin is thinner in telurus than in rotundus. PEKONELLA APHNOSTINA’ sp. nov. Plate XXIV, Test somewhat elongated, abruptly narrowed posterior to middle, very flat, finely and evenly granulated ; length 137 mm. ; greatest breadth, slightly anterior to mouth, 112 mm.; breadth equals less than .82 of length; 15 mm. back of mouth, width is only 95 mm. and 30 mm. back of mouth it is only 88 mm.; at anus, it is 55 mm. Apex of test coincides with centre of madreporic body, 64 mm. from anterior margin of test. Mouth directly beneath apex. Apical-oral diameter, 13 mm. Test thinnest at margin where it is scarcely 3 mm. thick. Oral surface flat ; mouth scarcely at all sunken. Auricles fused into a single stout piece on each interambulacrum. Anal system small, about 4 mm. in diameter, its distal margin about 6 mm. from test margin ; it is covered with small spine bearing plates. Petaloid area, about 72 mm. long and 68 mm. broad. Anterior or unpaired petal, 36 mm. long ; antero-lateral, 32 mm. ; postero- lateral, 37 mm. Each petal is about 10 mm. wide. Unpaired petal broadly open (by 4.5 mm.) ; paired petals open by about 3 mm. Madreporic body about 4 mm. across. Genital pores 4, there being none in posterior interradius. Primary spines 3-4 mm. long actin- 1 GR.=Afhno, of a sudden to make narrow, in reference to the shape of the test. > 68 ] UECORDS OF W.A. MUSEUM. ally, much smaller abactinally, terete, nearly smooth, those about mouth stoutest. Miliary spines minute, abundant, similar to those of P. lesiienvi. Pedicel lariae very scarce ; the single tridentate that was found, was similar to those of lesuem'i. Colour abactinally, light reddish brown ; the red is very marked when the test i-s moistened ; actinally the colour is more yellow- brown. Carnac Island, near Fremantle, W.A., Dec. ii, igog. One specimen, No. 3936. Type. I have been greatly in doubt as to whether this specimen represents a new species or is a peculiar individual variant of P. ksueuri, Agass. Comparison with numerous specimens from Queensland and from the East Indies has failed to satisfy my doubts, but as lesuciin has hitherto been found only on the eastern coast of Australia and does not reach so far south as riass Strait, I have decided to give this West Australian specimen a new name indicative of its strikingly peculiar shape. Compared with a speci- men of from Queensland, of the same length, the following peculiarities are noted : — 1. The test is remarkably narrow; its greatest width is not much over 80 % of its length, while in lesueun it is distinctly over 90 %. 2. The test is abruptly narrowed back of the mouth; its width 30 mm. back of mouth is less than 65 % of its length while it is more than 78 % in lesueuri at the same point. 3. The petals are shorter, wider and much more open at the tip ; in lesueuri the petals are open only i mm. or less. 4. The test is thinnest at the margin ; in lesueuri the test is a trifle swollen at the margin and is thinnest several millimetres proximal to margin. 5. The test seems to be more finely and uniformly granu- lated than in lesueuri. While the last three of these characters are certainly more or less variable In and hence of doubtful value, the combination of the five ill this Carnac Island specimen gives it a general appear- ance totally unlike that of any Peronella I have ever seen. The collection of a few more specimens on the West Australian coast would show whether aphnostina is a valid species or not. RECORDS OF IV. A. MUSEUM. [169 LINTHIA AUSTRALIS. Desoria australis, Gray, 1851. Ann. Mag. Nat. Hist. (2), vol. 7, p. 132. Lintht a australis, A. Agassiz, 1872. Rev. Ech. pt. I, p. 138. A single bare test (No. 5005) from Fremantle Beach shows that this is a West Australian species, although it was previously known only from South-western Australia and Tasmania. It is still a rarity in Museums and specimens with spines are greatly to be desired. ECHINOCARDIUM AUSTRALE. Gray, 1851. Ann. Nat. Hist (2), vol 7, p. 131. It is not at all surprising to find this widely distributed species in the collection. Safety Bay, W.A. Three specimens. Nos. 239, 240, 241. BREYNIA AUSTRALASIAE. Spataugus australasiae, Leach, 1815, Zool. Misc., vol 2, p. 68. Breyiiia australasiae. Gray, 1851. Ann. Mag. Nat. Hist. (2), vol. 7, p. 131. The larger of these bare tests (No. 4562) measures 120 mm. long, by 100 mm. wide and 60 mm. high, showing that Breynia grows to a larger size than has hitherto been known. Although these speci- mens differ from those taken at Lord Howe Island, N.S.W., by the very characters on which Gray based his Breynia desorii, other specimens in the M.C.Z. collection are intermediate and I do not feel satisfied that desorii is a valid species. Abrolhos Islands, W.A. One specimen. No. 4562. Fremantle Beach, W.A. One specimen. No. 5004. HOLOTHURIOIDEA. The holotburians unfortunately are not in nearly as good con- dition as the dry echinoderms, and it is impossible to identify one- third of them. They seem to have been in formalin and several are more or less decalcified. Nevertheless all are of interest for we have hitherto known almost nothing of the holotburians of the western half of Australia — indeed we know little enough ol those of the eastern coast. Of the six identified species, three belong to the genus Pentacta (=Colochirus) and one of these seems to be a remarkable new species, which, should the characters shown by these specimens prove real and constant, might well be given a genus of its own. The other new species is an apodous holothurian of the genus Caudina. lyo] RECORDS OF W.A. MUSEUM. CAUDINA TETRAPORA, ' sp. nov. Tentacles 15, each with 4 slender, sharply pointed digits, the terminal pair the longer. Calcareous ring well developed, about 2 mm. broad (high) ; posterior prolongations of radial pieces very conspicuous, nearly 4 mm. long. Stone-canal single, well-developed. Polian vessel single, long and cylindrical. Genital glands rather sparingly dichotomously branched. Respiratory-trees very fully developed. Body not very stout, passing gradually and not abruptly into the tail. The tip of the tail is damaged in both specimens, so the condition of the anal papillae could not be deter- mined. Calcareous particles knobbed buttons perforated ivith four holes and with more or less conspicuous pro- jections around the margin (text, fig. I) ; they are about .1 of a millimetre long. Most of the particles are of compara- tively regular form, but more or less variation from the typical condition may be noted ; buttons with fewer than four holes are more frequent than those with more. Colour white or pale brown. Length about 75 mm., of which about one-third is tail ; diameter at middle of body 15 mm. : diameter near tip of tall, about 3 mm. Cottesloe Beach, W.A., July 6, 1912. One specimen. No. 5035. Type. Abrolhos Islands, W.A. One speci- men, No. 229. This species is readily distinguished from the other members of the genus by its perfectly distinctive calcareous particles. It is no doubt most nearly related to C. chilensis, the common New Zealand species, as might have been expected from its geographical propinquity. Text-Figure i. Calcareous particles from body- wall of Caudiiia tetrapora. x 350. a. A typical button seen from the outer surface. h. A similar button from the side, t. A button with long marginal projections. 1 Gr. having four pores, in reference to the calcareous particles of the skin. RECORDS OF W.A. MUSEUM. [lyr MOLPADIA, sp. ? A specimen from Cottesloe Beach [No. 233 (2983)] is in such - poor condition that its identification is impossible, but the stout calcareous rin^ with comparatively short posterior prolongations on the radial pieces is suggestive of Molpadia rather than Caudina, and the fifteen tentacles and absence of tube-feed, taken in connec- tion with its general appearance, show that it is certainly a mol- padid. The tail is missing : the remaining body measures about 40 mm. long by 17 in diameter. COLOCHIRUS QUADRANGULARIS. Holothuria qmdranguiaris. Lesson, 1830. Cent. Zool., p. go, Colochirns quadrangularis, Selenka, 1868. Zeit, f.w.zool,, vol. 18, p. 112. A single specimen, well preserved and about 55 mm. long, but without data, other than “ West Australia,” is the sole representa- tive of this species. COLOCHIRUS TUBERCULOSUS. Holothuria tuberculosa, Quoy and Gaimard, 1833. Voy. de I’Astrolabe, Zool. vol. 4, p. 131. Colochirus tuheiciilosa, Semper, 186S. Reis, Arch, Phil.: Holothnrien, p. 239. A specimen about 50 mm. long, and without other data than the simple “West Australia,” represents this species. COLOCHIRUS AXIOLOGUS," sp. nov. Plate XXV, Tentacles 10, large and arborescent, the two ventral ones much smaller than the other eight. Calcareous ring only moderately stout, about 3-4 mm. broad (high), with no posterior prolongations, Polian vessel single, rather large and nearly spherical. Stone-canal single, in the dorsal mesentery. Genital glands unbranched, 40-50 mm. long, in a thick tuft on each side of the mesentery, near the middle of the body cavity. Respiratory trees short, but well- developed and much branched. Ambulacra! appendages nearly or quite confined to the ventral ambulacra; each of these ambulacra at the middle of the body carries about eight longitudinal series of well- formed, rather large pedicels, the entire band being about 10 mm. wide. As the ends of the body are approached, the series of 1 Gr, Axiologos^rems.rksRAe. 172] RECORDS OF W.A. MUSEUM. pedicels rapidly, almost abruptly decrease to only four or three and then disappear altogether, so that there are no pedicels near either mouth or anus, even in the ventral ambulacra. Dorsal ambulacra (except to some extent near mouth) and all interambulacral areas, entirely free from pedicels or appendages of any kind. Body wall thick (in the type which has been preserved in alcohol, it is 2-3 mm thick) but entirely without calcareous deposits of any kind ; a few minute calcareous rods are to be found in the finer branches of the tentacles, and the anus is guarded by five conspicuous calcareous teeth. Interambulacral areas somewhat pointed and projecting at both their anterior and posterior ends, forming valve-like folds which, in the contracted condition, conceal both mouth and anus. Colour rose-purple, dull in the type and passing into brown on the tentacles, but rather bright in the other specimen, which might perhaps better be described as purplish-rose ; in this specimen the tentacles are fully contracted and drawn into the body cavity and the neck-skin thus protected is bright purple, which is perhaps the natural colour of the entire animal in life. The form of this species is notable and is better shown in the type, which is approximately 90 mm. long. The distance from mouth to anus along the mid-dorsal interambulacrum is however less than 70 mm. while along the mid-ventral ambulacrum it is 210 mm. The girth of the body is 190 mm. Port Hedland, W.A. Two specimens, Nos. 4032 and 4033. No. 4033 is the Type. It is difficult to decide whether the absence of calcareous particles in the body-wall is the natural condition or is due to decalcification. The type is so well preserved it is hard to believe it has been decalcified, but the other specimen was preserved in formalin and has the appearance of having been decalcified. If decalcification has occurred artifically it is hard to understand why the anal teeth and calcareous ring should persist apparently unin- jured. On the other hand, the species of Colochirus, hitherto known, have an excessive amount of calcareous matter in the body- wall and its absence would therefore be an extraordinary specific character, should it prove to be natural and constant. The form of the body in these two specimens and the absence of pedicels on the dorsal surface are also extraordinary characters and should further RECORDS OF W.A. MUSEUM. [173- collecting produce additional specimens showing essentially the same combination of remarkable features, I should consider the species entitled to separate generic rank. ACTINOPYGA MILIARIS. Holothuria miliaris, Quoy and Gairaard, 1833, Voy. de I’Astrolabe, Zool., vol.4,. p, 137. MiiUeria miliaris, Brandt, 1835. Prod. Desc. Anim., p, 74 (et auct.) Two large specimens (Nos. 218, 219) without data, except “ West Australia, ” are in the collection. The generic name Miilleria has so often been shown to be preoccupied, so far as holo- thurians are concerned, there can be no justification for its further use instead of Brown’s suggested substitute, Actinopyga. STICHOPUS, sp. ? Three large holothurians (No. 4939) from between Fremantle and Geraldton seem to belong to the genus Stichopus, but they are so contracted and distorted and the body surface is so rubbed that it is impossible to determine them satisfactorily. They do not seem to be 5. variegatus and I think they probably represent an undes- cribed species. HOLOTHURIA ATRA Jaeger, 1833. De Holoth., p. 22, These specimens which range from 100 mm. to 250 mm. in length seem to be identical -w'ith similar individuals from the Philippine Islands, and I see no reason to doubt that they are atra. There are no data with them but they are from “ West Australia ” and bear the Nos. 213, 215 and 216. Three specimens. HOLOTHURIA. sp. ? There is a small, decalcified holothurian (No. 251) from “Pelsart Island, Abrolhos, W.A.,” which unfortunately cannot be determined. Museum of Comparative Zoology, Cambridge, Mass., U.S.A., June \st, 1913. 174 ] RECORDS OF W.A. MUSEUM. ON A COLLECTION OF FEOM WESTERN AUSTRALIA By dene B. fry Junior Assistant, Australian Museum, Sydney. Published by permission of the Trustees of the Australian Museum, Plates XXVII-XXVUI. and Text Figures i-ii. In continuation of the arrangement with Mr. Bernard H. Woodward, Director of the Western Australian Museum, Perth, the Trustees have received a large collection of Reptiles and Batrachians from that State for identification. The present paper forms a report on those species contained which are of special interest, some being regarded as new, whilst others are re-described or recorded from Western Australia for the first time. LACERTILIA (Lizards). Diplodactyliis woodwardt, sp. nov. ,, lucasi, nom. nov. I'eropns variegata, var. punctata, var. nov. OpMoseps repens, sp. nov. Amphibohmis sciitulafus, Stirling and Zietz. Egernia formosa, sp. nov. Lygosoma (Rhodona) pictunitum, sp. nov. CryptoblepJiarns rhodonoides, L. & F. OPHIDIA (Snakes). Liasis olivaceus. Gray ? Demansia psamiuophis, var. reticulata, Gray. ,, modesta, Gunther. ,, appnts, Gunther. ,, nnchalis, Gunther. Psendechis australis, Gray. Furina bimaculata, D. & B. RECORDS OF W.A. MUSEUM. [175 BATRACHI A (Frogs and Toads). Limnodynastes ornatns. Gray. ,, dorsalis, Gray, var. typica. Crinia geovgiana, Bibr., var. siolata, Cope. ,, leai, Fletcher. Phractops australis, Gray. Helioporus albopunctatus, Gray. Myobatrachus gouldii, Gray. Hyla caerulea, White. ,, rubella, Gray. ,, adelaideusis, Gray. ,, latopalmata, Gunther. ,, uasuta. Gray. I wish to express my thanks to Mr. Woodward for his kindness in allowing me to examine this collection, and for undertaking the publication of my report. LACEirriLIA (Lizards). DIPLODACTYLUS WOODWARDI, Sp. nov. Plate XXVII,, Fig. I. and Text Fig. i. Head elongate oviform, very convex, three-fourths as broad as long (the length of the head is measured from the tip of the snout to the ear opening). Snout rounded, one-third longer than the orbital diameter, as long as the distance between the eye and the ear orifice. Ear opening hardly distinguishable, very minute. Loreal region grooved ; forehead not grooved. Rostral twice as broad as high, highest in the middle line where it is nicked ; a small groove above, extending on the upper third of the rostral. Nostril situated between the first upper labial, a supranasal, and three small postnasals ; supranasals separated from one another by two enlarged hexagonal granules; postnasal granules not enlarged; granules bordering supranasals enlarged, hexagonal. The posterior border of the upper eyelid (seen under a lens) bears six to seven small, 176] RECORDS OF W.A. MUSEUM. soft conical spines. Labials ii, anterior largest. Mental longer than broad, broader in front than behind, slightly longer than the adjoining labials. Angle of the mouth slightly in front of the hinder border of the eye. A small fold from below the eye to Text Fig. i. — Diplodactylus woodwnrdi, sp. nov. (from type). a. Dorsal view of tip of snout, enlarged. b. Dorsal view of tip of fourth toe, enlarged. c. Ventral view of tip of fourth toe, enlarged. behind the angle of the mouth ; another, possibly due to preserva- tion, above the ear opening. Scales of head granular, very convex, uniform, those near the labials enlarged. Scales of back like those of the head, slightly larger than those of the sides. Belly scales flatter than dorsal scales, with a tendency, like those of the sides, to overlap. Scales of limbs uniform. Scales of tail squarish, larger than body scales, arranged in transverse rows. Digits cylindrical, not depressed at tips ; inferiorily with transversely oval granules; two enlarged, elongate, slightly diverging plates at the apex. Colour (Spirits) : — Back light brownish, with eight or nine alternating darker brown bands consisting of a network enclosing occasional white spots. Large white spots on the sides between the bands. A band of reticulations on the loreal region and behind the eyes, connecting with its fellow in the occipital region. Upper surface of limbs with faint reticulations. Tail with alternating brown bands dorsally connected with each other laterally. Under surfaces uniform creamish. Total length (from tip of snout to tip of tail) 51 mm. ' This handsome species is allied to D. alboguttatns, Werner, ^ 1 Werner — Fauna Siidwest-Austr., II, 1910, p. 462, fig. 4. RECORDS OF W.A. MUSEUM. [177 recently described from Denham on the Peron Peninsula, Shark Bay. The following short definition will distinguish the two species. D. alboguUatus, Werner. Snout a little longer than the distance of the eye from the ear opening. Ear opening obliquely elliptical. Rostral almost twice as broad as high, rectangular. Mental not longer than the adjoining labials. Nasal opening in contact with the rostral, one upper labial and four nasals, the upper (supranasals) in contact mesially. D. woodwardi, sp. nov. Snout as long as the distance between the eye and the ear opening. Ear very minute. Rostral twice as broad as high, roughly pentangular, nicked above. Mental slightly longer than the adjoining labials. Nasal open- ing not in contact with the rostral, and the supranasals are separate mesially by two hexagonal enlarged granules. There are also colour differences which, however, are better shown by a comparison of the two figures. Locality.— I have examined only a single young example from Western Australia. Type. — In the W.A. Museum. DIPLODACTYLUS LUCASI, nom. nov. Diplodactylus hilineatus, Lucas and Frost, Proc. Roy. Soc. Viet., Ser. 2, XV., 1903, p. 146 (not Diplodactylus biUneatus, Gray, Cat. Liz. Brit, Mu.s., ist ed., 1845, p. 149, and Zool. Erebus & Terror, Reptiles, pi. XV, fig. 3.) The name Diplodactylus hilineatus, Lucas and Frost, is antedated by the same name proposed by Dr. J. E. Gray for a Gecko which Dr. Boulenger now regards as synonymous with Phyllodactylus ocellatus, Gray. As it becomes necessary to propose a new name for Messrs. Lucas and Frost’s species, I 'have much pleasure in associating the name of Mr. A. H. S. Lucas with it. It is worthy of remark that Diplodactylus michaelseni, described and figured by Dr. Werner \ bears a general resemblance to Dr. Gray’s figure of Diplodactylus hilineatus, but differs materially in structural characters. 1 Werner — Fauna Siidwest-Austr., II, 1910, p. 460, fig, 4. 178 ] RECORDS OF W.A. MUSEUM. Diplodactylus michaelseni, Werner, was described from Denham, while Messrs. Lucas and Frost’s type of D. bilineatns came from Carnarvon. The type specimen of Dr. Gray’s D. bilineatns {Phyl- lodactylus ocellatus, Gray, fide Boulenger was described from Houtman Abrolhos, and recorded by Dr. Gunther from Champion Bay, and by Dr. Werner ® from Boorabbin, Coolgardie Goldfield. PEROPUS VARIEGATUS, Bum. and Bibr., van PUNCTATUS, var. nov. Gehyra variegata. Gray, Lucas and Frost, Rep. “Horn” Sci. Exped. Centr. Austr., II, 1896, p. 124, pi. IX, fig. 3 (part only). ? Gehyra variegata. Gray, Werner, Fauna Siidwest-Austr., II. 1910, p. 467 (part only). In the collection is a single female example of this widely dis- tributed species from the Strelley River, Pilbara. It presents the colour variety noted and figured by Messrs Lucas and Frost. In all structural characters it agrees with the variable P. variegatus, but the colour pattern is so different that I propose to distinguish it under the varietal name of punctatus. The following short diagnosis will serve to define this form. Upper surfaces russet brown, lighter on the snout, with trans- verse rows of silvery and yellowish spots. The spots are more abundant on the limbs. Black spots may be distributed all over the dorsal surface or they may be confined to two rows, one each side of the vertebral column, between the rows of lighter spots. Sometimes two faint streaks of brown on the loreal region, the upper continuing behind the eye to above the ear opening. OPHIOSEPS REPENS, sp. nov. Text Figures 2 and 3. Snout projecting, slightly hooked and trilobed ; a little more than three times the diameter of the eye. Eye considerably longer than its distance from the mouth, surrounded by a ring of small scales, of which the posterior are the largest. Portion of the rostral seen from above is once and one half as long as its distance from the frontal, once and a quarter as broad as long, as long as the 1 Boulenger — Brit. Mus. Cat. Liz., 2nd ed., I, 1885, p. 93. a Gunther — Ann. Mag. Nat. Hist., Ser. 3, XX, 1867, p. 49. 3 Werner — loc. cit. p. 456. RECORDS OF W.A. MUSEUM. [179 diameter of the eye ; the portion seen from below longer than broad and slightly narrower behind than in front. Nasals apparently fused with the first supralabial of each side, in contact behind the rostral ; a little shorter than the praefrontals, their suture being the Text Fig. 2. — Ophioseps repens, sp, nov. From type. a. Side of head, enlarged. b. Dorsal view of head, enlarged. c. Ventral view of head, enlarged. same length as that of the praefrontals. Nostril large, not visible from above ; a very distinct groove runs from its posterior edge to the lower anterior border of the praefrontal. Praefrontals a little larger than the nasals, a little more than half the length of the frontal, forming a suture in the median line and with the first supralabial (the second supralabial if we count the first as fused with the nasal). Frontal six-sided ; broader in front than behind; almost once and a half as long as broad, as long as its distance from the end of the snout. Supraocular narrow, extending down behind the eye and meeting the third supralabial. A pair of parietal shields larger than the adjoining scales, in contact behind the eye with the third supralabial. A small praeocular ; no postocular. i8o] RECORDS OF W.A. MUSEUM. Four supralabials distinct ; the third very high and bandlike, in contact behind the eye with the supraocular and the parietal scale, the fourth smallest, squarish. Mental large, nearly twice as broad as long. Four sublabials, the anterior pair the largest and sepa- rated from each other in the median line by a small shield which is followed by three others, the middle one of which is the smallest. Headshields thickly dotted with minute papillae. Scales smooth. Text Fig, 3. — Ophioseps repens, sp. nov, From type. Lateral view of anal cleft. Typhlops-like, in twelve series round the body. Praeanals not enlarged. No external rudiments of limbs. Tail one third as long as the body, terminating in a circular scale. Colour (Spirits) : — Body light greyish brown above, slightly lighter beneath. Longitudinal rows of faint spots on the dorsal and lateral scales, these spots fainter and the rows more numerous on the sides. Headshields with faint brown marks. Tail yel- lowish, the rows of spots of a reddish colour. Total length 145 mm.. Tail 45 mm. This new species is distinguished from 0 . nasuta, Bocage, mainly by the longer snout, which is hooked in profile ; the much larger nostril and the presence of a very distinct groove which con- nects with the praefrontal, not with the first supralabial ; the different proportions and shapes of the headshields, which in 0. repens are covered with minute papillae, and the smaller number of scales round the body. The two species are here compared side by side. RECORDS OF W.A. MUSEUM. [i8i Ophioseps repens. Ophioseps nasutus. I. Snout slightly hooked and trilobed : more than three times the diameter of the eye. I. Snout not hooked, broadly rounded. Two and one half times the diameter of the eye, 2. Portion of the rostral visible from below longer than broad. 2. Portion of the rostral visible from below broader than long. 3. Nostril large, the groove connecting it with the praefrontal. 3. Nostril small, the groove, when pre- sent, connecting it with the first supralabial. 4. Praefrontals and nasals seen from above longer than broad. 4. Praefrontals and nasals seen from above much broader than long. 5 Fontal broader in front than behind, just as long as its distance from the tip of the snout. 5. Frontal broader behind than in front, more than twice as long as its distance from the tip of the snout. 6. Supraocular in contact with the third supralabial behind the eye. 6. Supraocular separated from the third supralabial by a postocular or several small scales. 7, Headshields with minute papillae. 7. Headshields quite smooth. 8. Scales in twelve series round the body. 8. Scales in fourteen series round the body. Dr. Werner shows^ that there is considerable variation in O. nasuta. In his Aprasia brevirostris ^ which he has later shown * to be synonymous with 0. nasuta, the small scales which surround the eye are fused into a narrow band, and the portion of the rostral seen from above is sometimes half as long as its distance from the frontal while Jensen’s figure ^ of O. nasuta, shows it to be almost as long as its distance from the frontal. The nasal cleft may be present or absent. In Jensen’s specimens and in Du Bocage’s type ® there is a postocular, while in Dr. Werner’s types of A. brevirostris there are two small undifferentiated scales behind the eye. The praeanals may or may not be enlarged. This remarkable variation within the one species is only in keeping with the extraordinary individual variation of other members of the family Pygopodidae, to which family Dr. Werner has referred this genus. The above proposed new series however, 1 Werner — Das Tierreich, Lief 33, igi2, p. 26. 2 Werner — Fauna Siidw.-Austr, II, igog, p. 266, figs. ® Werner — Das Tierreich, Lief 33, 1912, p. 26. ^ Jensen — Vid. Meddel., 1899, p. 317, pi. iii. ■5 Bocage — Jorn. Sc. Lisb., IV, 1873, p. 321. RECORDS OF W.A. MUSEUM. 182 ] differs too much to be considered a degree in variation of the type species, and in some respects is close to Aprasia pulchella, Gray. Locality. — A single specimen from W.A. without exact locality is in the collection. Type. — In the W.A. Museum. AMPHIBOLURUS SCUTULATUS, Stirl. & Zietz. Amphibolunis scutulaius, Stirling & Zietz, Trans. Roy. Soc. S. Austr., XVI., 1893 p. 165, pi. VII., figs. I and 2. Snout almost twice as long as the diameter of the eye opening, with projecting canthus rostralis and very concave lores. Nostril slightly nearer the eye than the tip of the snout, directed upwards and backwards, pierced in an enlarged nasal situated below the angle of the canthus rostralis. Rostral small, broader than long, roughly hexagonal. Tympanum large, vertically elliptic but slant- ing slightly forwards, its vertical diameter equal to that of the eye opening. Upper head scales sub-equal, smallest on the supra- orbital region, mostly tricarinate, the keels meeting at the posterior extremity of each scale. A superciliary ridge with elongate uni- carinate scales. Scales much smaller on the temporal region. Scales bordering the upper lip, regular, not enlarged. A ridge of enlarged scales runs from beneath the eye to above the tympanum. Several small folds of skin on the neck bearing groups of small conical spines. Gular scales small, flat, sometimes faintly keeled, enlarged and hexagonal towards the lower labials. Mental longer than broad, slightly larger than the rostral. Scales of the upper surfaces strongly keeled, very slightly enlarged towards the verte- bral region. A very distinct nuchal crest continuous behind with a vertebral series of slightly enlarged scales, or a distinct dorsal crest. Scales of limbs larger than body scales. Ventral scales keeled, larger than dorsal scales. Fore limb moderate ; hind limb long, the toes of the adpressed limb reaching to the eye or the loreal region ; the length of the foot equals that of the fore limb. The male with fifty to fifty-three praeanal and femoral pores forming an uninter- rupted series on each side. Tail depressed at the base, with a serrated ridge not distinguishable on the posterior third of its length. RECORDS OF W.A. MUSEUHf. [183 Colour (Spirits) : — Upper surface of head rusty brown with black spots on the keels of the scales. Sides of head lighter, shad- ing to yellow, sometimes with several fine black temporal streaks. A single median series of large black spots on the neck and anterior part of the back, breaking up into a double series of dorso-lateral spots towards the posterior part of the body. These spots some- times enclose (towards the posterior part of the body) a light area of yellowish or brownish with a variegated centre. External to this series of spots is a light reddish, dark-edged band of lozenge-shaped spots, or a series of definite dark-centred light-edged ocelli. A series of about six narrow, light ashy-grey transverse bars cross the back, dividing the ocelli or separating the lozenge-shaped spots. Sides yellowish, reticulated with greyish or black. Tail sometimes with a distinct lateral band of lozenge-shaped spots and a light dorsal band, or variegated brownish. Numerous light ashy-grey bands cross the tail, completely dividing the dark lateral band. Under surfaces uniform yellowish, or more or less densely clouded with black on the throat, chest and abdomen. Limbs reddish- brown with yellow lines enclosing more or less regular shaped spots. Under surfaces sometimes clouded with black. Feet and hands rusty-red. The affinities of this lizard are with A. cristatiis, Gray, and the differences have been summed up by Messrs. Stirling and Zietz as follows : — •“ All the scales are much smaller, the head scales not rugose as in A. cristatus, but smooth and strongly keeled. The dorsal scales are not intermixed with larger ones, but increase in size towards the vertebral line. The compressed spines of the nuchal crest are shorter.” To these might be added the marked difference in colour marking. Recorded from between Fraser Range and Queen Victoria Springs. Localities. — A single adult female from Mt. Malcolm is in the collection. 1 have also examined a specimen presented to the Aus- tralian Museum by Mr. A. H. S. Lucas, from Kalgoorlie. Both these and the original description have been utilised in the prepara- tion of the above description. RECORDS OF W.A. MUSEUM. 184] EGERNIA FORMOSA, sp. nov. Plate XXVII, Fig. 2 and Text Fig. 4. Head rather broad. Head shields smooth. Rostral once and a half as high as broad, pentagonal, forming an arcuate suture with each nasal. Nasals slightly swollen, separated or just in contact with the rostral. A curved groove behind the nostril. Frontonasal slightly broader than long. Praefrontals moderate, forming a short median suture or just separated by the frontal. Frontal once and a third to once and a half as broad as long, separated from, or narrowly in contact with, the frontonasal ; in contact with the first and second supraoculars. Four supraoculars, second largest ; six to eight supraciliaries. A group of imbricate prae- and postoculars. Frontoparietals forming a median suture; in contact with the second, third and fourth supraoculars. Interparietal almost twice as long as broard, as long as or longer than the frontal. Parietals separated by the interparietal. Three pairs of enlarged nuchals Text Fig. i\.—Egernia Jormosa, sp. nov. a. Side view of head, enlarged. b. Dorsal view of head, enlarged. c. Tip of snout of a specimen showing variation in shields RECORDS OF W.A. MUSEUM. [185 normally. Labials ^ or the seventh upper and the seventh lower largest, the fifth and sixth below the orbit. Three enlarged temporals. Ear opening oval, with two to four projecting lobules anteriorly, its vertical diameter almost as long as the eye opening. Twenty-eight smooth or feebly straited scales round the body, those on the verterbral line slightly enlarged, those on the sides smallest. Praeanals not or but feebly enlarged. Limbs moderate; when adpressed they just meet or slightly overlap. Toes slender, com- pressed ; eighteen to twenty lamellae under the fourth. Tail cylindrical or very slightly depressed at the base, a little longer than the head and body ; no series of transverse scales dorsally except in reproduced portions. Colour (Spirits) : — Dark or yellowish olive above. Head shields with irregular dark brown markings. Two dark brown bands which sometimes coalesce, start on the parietals and extend on to the back, where they break up into spots. Numerous trans- versly arranged yellow spots on the back and tail. A dark brown band starts on the loreal region and extends on to the side where it breaks up into spots. Labials yellow, sometimes barred with brown Throat reticulated with brown. Under surfaces uniform yellowish. Length of largest specimen, from snout to vent, 95 mm. Egernia formosa has the general appearance of E. striolata,^ Peters, but in reality is widely separated from that species. In Dr. Werner’s key ^ to the species of the genus Egernia it must be placed in the division I. A., the species of which are characterised by the cylindrical tail and the smooth scales. Of the five species in this division it shows affinity with the first two only, Egernia luctuosa, Peters* and E. lauta, de Vis.* From the former it is dis- tinguished by the presence of a curved groove behind the nostril and in having twenty-eight scales round the body ; from E. lauta it also differs by the possession of a post-narial groove, while there are no infraoculars in E. formosa and much longer limbs. The separation of the palatine bones by the palatal notch is characteristic of Egernia.® Though this condition is maintained in all my specimens yet the fleshy intergument covering them over-reaches their edges so that they appear to be in contact in the median line. 1 Peters — Mon. Berl. Acad., 1870, p. 642 (TropidoUpisma striolaium). 2 Werner — Fauna Siidwest-Austr., II, 1910, p. 472. 3 Peters — loc. cit. 1866, p. 90 {Cyclodus [Omolepida] luctuosus). i de Vis — Proc Linn. Soc., N.S.W., 2nd ser., II, 1887, p. 813. 3 Boulenger — Brit. Mus. Cat. Liz., 2nd ed., Ill, 1887. p, 134. RECORDS OF IV. A. MUSEUM. 1 86] I have also found the same structure in E. striolata and E. whitei. A knife passed between the fleshy lobes, however, shows the true condition of the bones. Localities- — Eight specimens are before me, six adults and two young. The Type, together with three other specimens, was presented to the Trustees of the Australian Museum by Mr. W. D. Campbell, who collected them at Perth. There are also three other specimens collected by the same gentleman at Boulder. A single specimen, figured on PI. XXVII, Fig. 2, and Text Fig. 4, is in the Collection forwarded by Mr. Woodward, but is unfortunately with- out data. Type: — In the Australian Museum, Sydney. Reg. No. R. 3058. A co-type is deposited in the Western Australian Museum- LYGOSOMA (RHODONA) PICTURATUM, sp. nov. Plate XXVII. Fig. 3, and Text Figure 5. Head small, sub-cuneiform. Snout rounded, acutely produced in profile, three to four times as long as the obital diameter, with a slightly projecting labial ridge. Eye small ; the lower eyelid with a transparent disc. Ear visible, minute, but generally covered with scales. Rostral twice as broad as high, highest in the mid- line; forming an arcuate suture with each nasal. Nostril pierced in the centre of a large swollen nasal which forms a moderate suture with its fellow. Fronto-nasal once and one-quarter to once and three-quarters as broad as long, forming a broad, curved suture with the frontal. Praefrontals small and widely separated ; some- times united with the posterior loreal and in contact with the second and third, or the third upper labial. Frontal large, much broader than the supraocular region ; a little longer than broad ; almost as long as its distance from the end of the snout ; in contact with the first supraocular. Supraoculars two, first much the larger. Three or four supraciliaries, the first one sometimes very long. Upper eyelid represented by a row of very small lubricate scales between the supraciliaries and the orbit. A group of small, irregular prae- and postocular scales. Frontoparietals separated by the interparietal ; about as large as the first supraocular, larger than the praefrontals. Interparietal moderate, narrowly in contact RECORDS OF W.A. MUSEUM. [187 with the frontal. Parietals rather narrow, forming a suture behind the interparietal. Two to four pairs of indefinitely enlarged nuchals. Temporal shields three, the two bordering the parietals large. Text Fig. 5. — Lygosama {Rhodona) picturaiimi, sp. nov. a. Dorsal view of head, enlarged. b. Ventral view of head, enlarged. Labials f , the fourth upper entering the orbit, the fifth largest. Twa elongate loreal shields, one, the posterior, sometimes united with the praefrontal. Eighteen to twenty polished scales round the body. Praeanals distinctly enlarged. Fore limbs very minute, represented by a scarcely visible papilla situated in a depression, or by a styli- form rudiment which may be as long as the eye. Hind limbs didactyle as long as, or nearly as long, as the distance between the eye and the fore limb ; contained almost four, to nearly five times in the distance between the axilla and the groin. Toes compressed ;. twelve to fourteen lamellae under the longest. Tail as thick as the body. Colour (Spirits) : — Buff-coloured above. Head shields edged with brown. Two to four rows of spots confluent into distinct lines on the back and tail. A dark brown lateral band runs from the nostril, along the side, to the end of the tail. Uniform yellowish beneath. Length of largest specimen, from snout to vent, 92 mm. This species presents two more or less distinct varieties which may be separated as follows : — Variety A (typical form). — Rows of spots forming four lines down the back. Fore limb a distinct styliform rudi- ment at least as long as the eye. Twenty scales round the body. RECORDS OF W.A. MUSEUM. t88] Variety B. — Only two distinct lines down the back. Fore limb represented by a minute papilla, situated in a small depression, below the general surface. Eighteen to twenty scales round the body. Lygosoma picturatum is allied to L. gevrardi, Gray, * but also appears to possess affinities with L.walheri, Blgr. ® The three species may be distinguished as follows : — Lygosoma gervardi, Gray. Three supraoculars, second largest. Five supraciliaries. Frontoparietals distinct. Yellowish above with a dark lateral and a broad dorsal band. Lygosoma pictiiratum, sp. nov. Two supraoculars, the first much the larger. Three or four supraciliaries. Frontoparietals distinct. Buff coloured above, with a dark lateral band and two to four rows of spots confluent into lines on the back and tail. Lygosoma walkeri, Blgr. Three supraoculars, second largest. Five supraciliaries. Frontoparietals fused into a single shield. Greyish above, each scale with a black dot, which is largest on the fourth scale from the mid-dorsal line. Localities. — I have examined nine specimens of this new species. Two, including the Type, were presented to the Trustees of the Australian Museum by Mr. W. D. Campbell, who collected them at Boulder. A third specimen was collected by the same gentleman at Perth. Five other specimens are without locality, and differ somewhat from the typical form in having only two lines down the back. In the collection forwarded by Mr. Woodward is a single large specimen, figured on Plate XXVII, fig. 3, and Text fig. 5. This specimen is unfortunately without data. Type. — In the Australian Museum at Sydney, Reg. No. R. 3101. A co-type is in the W.A. Museum. 1 Boulenger— Brit. Mus. Cat. Liz., 2nd ed.. HI, 1887, p. 335, pi. XXVII, fig- 3 - 2 Boulenger — Ann. Mag. Nat. Hist. (6), VIII, 1891, p. 405. RECORDS OF IV. A. MUSEUM. [189. CRYPTOBLEPHARUS RHODONOIDES. L. &F. Ablepharus rhodonoides, Lucas and Frost, Proc. Linn. Soc. N.S. Wales, XXI,, 1896, p. 281. ? Ablepharus distinguendus, Werner, Faun. Siidwest. Austr., II., 1901, p. 490. The resemblance between this small Crypioblephanis and Lygo~ soma (Rhodona) fragile, Gunther, is very striking indeed, rendering the specific name proposed for it by Messrs. Lucas and Frost a very appropriate one. It was originally described from Mildura, a town- ship on the Victorian side of the Murray River. I am now able to extend its range to both Western Australia and New South Wales. In the Western Australian Collection is a specimen from the Strelley River. The Australian Museum Collection contains numerous specimens from the following localities : — Perth and Boulder, Western Australia ; Mildura, on the Murray River, Victoria (one of the types); Darling River, Moloch, Bindogmidra and Narra- mine, on the Western Plains of New South Wales. A careful examination of all my specimens shows that this species is not a variable one, indeed I can find no point in which they differ from the original description, or from the co-type speci- men in the Australian Museum. It seems to me very probable that the scink described by Dr. Werner as Ablepharus distinguendus really belongs to this species. Dr. Werner’s description is very meagre however, and I cannot be sure of this. The only difference that I can detect is in the number of scale rows, which in his new species are eighteen in number whereas in all my specimens of C . rhodonoides they are in twenty. A. distinguendus is stated to be very closely allied to Cryptoblepharus muelleri, Fischer, with which species Messrs. Lucas and Frost also compare C. rhodonoides. RECORDS OF W.A. MUSEUM. 190] OPHIDIA. (Snakes.) LIASIS OLIVACEUS, Gray ? Liasis olivacens (Gray), Boulenger, Brit. Mus. Cat. Sn., I., 1893, p. 79, pi. IV., fig. 2. There is a large skin in the collection which differs so much from Boulenger’s description that it should perhaps be regarded as a distinct species. With our limited knowledge of the variations of this snake however it is safer to merely record these variations and await additional material. The differences will be seen from the following table. L. olivacens. L. olivacens. ? V. 349-361. V. 357 - S. C. I00-I02. S.C. loi. Scales in 69-75 rows. Scales in 56-64 rows. . Labials Labials Rostral broader than high. Rostral higher than broad. Eye larger than its distance from the Eye smaller than its distance from the mouth. mouth. Frontal ij as long as broad. Frontal i| as long as broad. A small azygous shield (often absent) A large well defined six-sided azygous separates the praefrontals. shield separates the praefrontals. Internasals more than half as long as Internasals half as long as anterior anterior praefrontals. praefrontals. There are five lower labials pitted, the pits being decidedly stronger than those figured by Boulenger. The parietals are in contact behind the frontal and are considerably larger than he shows them. DEMANSIA, Gray. Deviansia, Gray, Zoo!. Miscell., 1842, p. 54 (for Elaps psammophis, Schleg.) Diemaiisia, Gunther, Cat. Col. Sn., 1858, p. 254. Diemenia, Gunther, Ann. Mag. Nat. Hist., (3), XII., 1863, p. 350. et auctormn. The generic name Diemenia used by recent authors is unten- nable. It was originally spelt Deinansia by Dr. Gray but was RECORDS OF W.A. MUSEUM. [191 afterwards altered to Diemansia and Diemenia by Dr. Gunther. Diemenia has been used by all recent authors but the name as originally spelt must be used. The following pages include descriptions and notes on the Western Australian species of the genus, based on the material forwarded by Mr. Woodward together with numerous specimens in the collection of the Australian Museum. Key to the Western Australian specis of Demansia. A. Scales in fifteen rows D. psammophis var. reticulata. Gray. B. Scales in seventeen rows. c. Portion of rostral visible from above measuring about half its distance from the frontal. V. 154-178, S.c. 38-51 pairs D. modesta, Gunther, p. cc. Portion of rostral visible from above about three quarters as long as its distance from the frontal. Snout broadly rounded. Head with brown smudges ; back with scattered brown spots. V. 215-225, S.c. 52-63 pairs, D. afinis, Gunther, p. ccc. Portion of rostral visible from above as long as its dis- tance from the frontal. Head tapering, sides nearly straight, snout more or less truncate. V. 184-224, S.c. 49-66 pairs D nuchalis, Gunther, p. DEMANSIA PSAMMOPHIS, Schl., var. reticulata, Gray. Diemenia psammophis, var. reticulata, Boulenger, Cat. Sn. Brit. Mus. III., 1896, p. 323- Diemenia reticulata, Lucas and Frost, Kept. “ Horn” Sci, Expdn., II., 1896, p. 147. Diemenia psammophis, Werner, Fauna Sudwest-Austr., II., 1909, p. 257. Scales 15 rows on anterior half, 13 on posterior half of body. Oc. I -I- 2 or 3. Temp. 2 + 2. V. 176-197. S.c. 65-79 pairs Several specimens of var. reticulata Ate in the collection. They are all a light olive with a very distinct reticulated pattern. The hinder third of the body is somewhat lighter, shading to yellowish brown at the tip of the tail. A black, yellow edged, streak passes from eye to eye round the rostral, and another larger one beneath the eye passes obliquely downwards and backwards. A yellow spot on the praeocular ; labials yellow. Yellowish or creamy white beneath 192 ] RECORDS OF W.A. MUSEUM. Localities. — This well marked variety is confined to Central and Western Australia. Messrs. Lucas and Frost record it from Alice Springs and Charlotte Waters, Central Australia, and Dr. Werner from Northam, W.A. In the Australian Museum are specimens from Perth and Strelley River, Pilbara. Several speci- mens of D. psammophis from Bourke and Moree, western N.S.W., belong to the typical variety, while one from North Australia is the form described by Macleay as D. papuensis. DEMANSIA MODESTA, Gunther. Cacophis modesta, Gunther, Ann. Mag. Nat. Hist. (4), IX, 1872, p. 35, pi. III,, fig. C. Ftirina ramsayi, Macleay, Proc. Linn. Soc. N.S.W., X, 1885. p. 61. ,, I, Lucas and Frost, Kept. '■ Horn ” Sci. Expdn., II, 1896, p. 149. ? Brachysoma sulherlandi, de Vis, Proc. Roy. Soc. Q'land., I, 1884, p. 139. ? Pseudelaps sutimlandi, Boulenger, Cat. Sn. Brit. Mus., Ill, i8g6, p. 320. ? Psendelaps sulherlandi, Longman. Mem. Q'land. Mus., I, 1912, p. 24. Diemenia modesta, Boulenger, Cat. Sn. Brit. Mus., III., i8g6, p. 320. Scales 17 rows. Temp, i -|- i or 2, V. 154-178, S.c. 38-51 pairs. This species is at once distinguished from its congeners, D. textilis and D. affinis, by its smaller number of ventral and sub- caudal plates. In the specimens before me, all of which are young, the dark cross bands vary considerably according to the age of the specimen. They become indistinct, in some cases quite obsolete, at a somewhat earlier stage than do the bands in young examples of D. textilis. I have not seen any fully adult examples of this species, but, judging from Dr. Boulenger’s description, the cross bands in some specimens remain throughout life, as in exceptional cases in D. textilis. I have examined the three specimens from Milparinka, western N.S.W,, labelled types of Fufina ramsayi, Macleay, and find they are identical with the young of D. modesta. There are some dis- crepancies in Macleay’s description. According to that author the ventrals are 162 and the subcaudals 38 pairs in the largest speci- men, whereas I find them to be 178 and 43 pairs respectively. In the second largest specimen they read V. 165, S.c. 49 pairs, and in the smallest, V. 169 I, and S.c. 45 pairs. In the latter specimen, which is apparently just hatched, the frontal is twice as broad as the supraocular, and proportionately large. This feature I have observed in very young snakes of other species. RECORDS OF P/.A. MUSEUM. [193 Localities. — D. modesta, Gunther, is recorded from the follow- ing localities : — Perth ; the North-West (types, Gunther, 1872, and Bonlenger, 1896) ; Geraldton (Boulenger, i8g6), Milparinka, Bar- rier Range, Western N.S.W. (types of Purina ramsayi, Macleay, Austr. Mus.) ; Crown Point, Horse Shoe Bend, Finke River, Central Australia (Lucas and Frost, 1896); Norman River, North Queensland (types of Bfachysoma sutherlandi, de Vis, 1884, Q’land. Mus.) ; Lawlers and Boulder, W.A. (Austr. Mus,). DEMANSIA TEXTILIS, D. & B. Dieinmia textilis, Boulenger, Cat. Sn. Brit, Mus,, III., 1896, p. 325, Although there are several mentions of D, textilis occurring all over Australia, I know of no authentic record of its occurrence in Western Australia. Krefft ^ gives “ Australia generally ” as the habitat of Diemenia superciliosa, as it was then called. Again he says, — “ A species which . . . ranges from the East to the West Coast, and perhaps extends all over the whole continent.” Speak- ing of D. nuchalis, Waite ® is of the opinion that “ it is possible that the Brown Snakes {D. textiUs) recorded from Western Australia are referable to this species.” DEMANSIA AFFINIS, Gunther. Plate XXVII, Fig. 4, and Text Fig. 6a. Pscudonajaaffinis, Giinther, Ann. Mag. Nat. Hist., (4), IX, 1872, p. 35, pi. IV, fig. C. ,, ,, Macleay, Proc. Linn. Soo. N.S.W., II, 1878, p. 29. ,, ,, Lucas and Frost, Kept. “ Horn " Sci. Expdn., H, 1896, p. 148, Diemenia michaUs (part), Boulenger, Cat. Sn. Brit, Mus., HI, 1896, p. 326. Head broad ; snout not truncate but broadly rounded. Eye moderate. Rostral broader than deep, the portion visible from above almost as long as its distance from the frontal ; internasals two-thirds to four-fifths the length of the praefrontals ; frontal broader than the supraocular, once and a half as long as broad two- thirds the length of the parietals ; nasal entire or semi-divided. Two or three postoculars; temporals i -f- 2. Six upper labials, last very large, third and fourth entering the eye. Scales in seventeen to twenty-one rows. V. 215-225. S.c. 52-63 pairs. 1 Krefft — Austr. Vertebr., Foss, and Recent, Sydney, Feb. 1871, p. 54. ■I Krefft — Vertebr. Lower Murray, 1865, p. 31. 3 Waite — Australian Snakes, i8g8, p. 51. RECORDS OF W.A. MUSEUM. 194J Colour (Spirits) : — Uniform brownish above, lighter on the head and the sides. Head with dirty smudges and a few scattered, irregularly placed dark brown spots. Sides and back with irregu- larly distributed, more or less abundant dark brown spots, which in many cases exactly cover one scale. Under surfaces uniform yellowish olive, or finely peppered with grey. Throat yellowish or dark bluish grey. Total length oflargest specimen, 1525 mm., tail, 250 mm. I have examined four specimens of this Demansia and think it will prove constantly distinct from D. nuchalis. After examining Dr. Giinther’s type specimen Dr. Boulenger placed this species in the synonomy of D. nuchalis, possibly regarding such slight differ- ences founded on a single specimen as not of specific value. With four additional specimens at hand however, all of which differ in the same characters from D. nuchalis, I propose to raise D. afinis to specific rank. When placed side by side the two forms appear Text Fig. 6. а. Demansia affinis, Gthr. (After Gunther, from type specimen, reduced.) б. Demansia nuchalis, Gthr. (After Gunther, reduced.) Strikingly different, but a closer examination shows that the differ- ences are mainly in the general appearance, the outline of the head (see Text figs. 6a and 6b), and the colouration. Also, Messrs. Lucas and Fro.st record a specimen in which the scales are in twenty-one rows, the greatest number recorded for the genus. Localities. — Demansia affinis in known from the following localities: — Australia (type, Gunther, 1872). Reedy Creek, George RECORDS OF IV. A. MUSEUM. [195 Gill Range, Central Australia (Lucas and Frost, 1896). Western Australia (W. A. Mus.) ; S.W. Australia (Austr. Mus.) King George Sound (Macleay Mus.) DEMANSIA NUCHALIS, Giinther. (Fig. 6b.) 182. 219. Pseudoiiaja nuchalis, Gunther, Cat. Col. Sn., 1858,-9, 227. Gunther, Proc. Zool. Soc., 1863, p. 17, fig. McCoy, Ann. Mag. Nat. Hist., (3), XX., 1867. p. Krefft, Sn. Austr., 1869, p 44, pi. XII., fig 13. Macleay, Proc. Linn. Soc., N.S.W., II., 1878, p. de Vis, Proc. Roy. Soc. Q’land., I., 1884, p, 58. Lucas and Frost, Kept. " Horn”Expdn., II., 1896, p. 148. Diemenia aspidorhyncha, McCoy, Prodr. Zool. Vic., 3., 1879, p, 13, pi. XXIII., fig. 4. Diemenia nwhalis, Boulenger, Cat. Sn. Brit. Mus., HI., 1896, p. 326. ,, ,, Werner, Fauna Siidwest-Austr., II., 1909, p. 257. Pseudelaps hancrofti, de Vis, Ann. Q’land. Mus. X., 1911, p. 25. ,, ,, Longman, Mera. Q’land. Mus., I., 1912, p. 24. Head tapering and truncate at the snout, the sides almost straight in the adult but somewhat curved in the young. Eye mod- erate. Rostral broader than deep, the portion visible from above may be a little longer than its distance from the frontal or not quite as long ; internasals shorter than the praefrontals ; frontal consider- ably broarder than the supraocular, once and a half to almost twice as long as broad, a little shorter to a little longer than its distance from the tip of the snout, two-thirds to four-fifths the length of the parietals. Nasal entire or semi-divided, in contact with a single praeocular two or three postoculars ; temporals i 2 or 3 ; six upper labials, third and fourth entering the eye, sixth very large. Scales in seventeen to nineteen rows. V. 184-224; anal divided; S.c. 49-66 pairs. Colour (Spirits) : — Var. A. Body uniform light or dark brown above. A. nuchal collar may be present. Under surfaces greenish grey or yellowish ; throat slightly darker A melanotic form, in which the head, neck and anterior parts of the body are rich brown or black, occurs in this variety. (Vars. A. and C. of Lucas and Frost; vars. A. and B. part only ? of Boulenger ; vars. 1 , II, III of Werner.) 196] RECORDS OF W.A. MUSEUM. Var. B. Body light or dark brownish above, variegated with black or dark brown. The scales may be edged with darker (edged with lighter according to McCoy in D. aspidorhynclia) , or there may be bands of darker colour or zig-zag lines of dark brown. These markings, if not confined to the posterior two-thirds of the body are most distinct in that region. One or two black nuchal bands are rarely present. Ventrals olive, greyish or yellowish, uniform or with round dark spots on those of the posterior moiety. Throat some- times peppered with grey. A melanotic form also occurs in this variety. (Vars. B. and E. of Lucas and Frost ; var. B. part only ? of Boulenger ; vars. IV and V of Werner). Total length of largest specimen . (Krefft), 1600 mm., tail 300 mm. Demansia nuchalis is an extremely variable species both in colouration and in structural characters. The youngest specimen I have seen is sixteen inches in length, in which the colouration is like that of some of the adult specimens. This suggests that the young of this species, unlike the young of D. textilis, and D. modesta, may not have dis.tinct black bands round the body. I have examined two melanotic specimens which do not otherwise differ from the typical form, and, as this form is present in both varietie.s it would appear to be quite sporadic in its occurrence. Localities. — This species is known from the following locali- ties : — Swan River and Geraldton ; North-Western Australia ; Port Essington, Northern Territory (Boulenger, 1896); Edel Land, Baba Head and Rottnest Island, W.A. (Werner, igog) ; Alice Springs and betiveen Laurie’s Creek and Glen Edith, Central Australia (Lucas and Frost, 1896); Port Darwin, Northern Terri- tory (Madeay, 1878); Stannary Hills, Central Eastern Queensland {de V\s, igji, as Pseiuhlaps biincrofli). In the Australian Museum are specimens from Perth, 80 miles south of Perth, and Beverley, W.A. ; Hermidale, near Nyngan, Western N.S.W., and Yandem- bah, Riverina, N.S.W. RECORDS OF W.A. MUSEUM. [197 PSEUDECHIS AUSTRALIS, Gray. A very large specimen is in the collection which is of interest as the characters exhibited show an overlapping between P. denisonioides, Werner and P. australis, Gray. The ventrals are i8g and the subcaudals 40 + (54 in all), thus agreeing almost exactly with the scaling of P. denisonioides, V. 189, S.c. 34 + xf (54 in all). In P. denisonioides the internasals are one-third as long as the praefrontal, a condition exhibited by none of the specimens of P. australis that I have seen. The frontal is half as long as the parietal in Werner’s new species, but in my specimenit is only three- fifths as long, the normal for P. australis being two-thirds the length of the parietals. Other differences in Werner’s species are a single anal and the greater extension of the rostral on the dorsal surface of the snout. I have examined a number of specimens of P. australis and find that this species varies very considerably. P. denisonioides, hoveever:, differs so much that the above could not be regarded as uniting the two species, but shows at least that the scaling overlaps. The scale formulae of P. australis should now stand; — V. 189- 220; Anal 2, S.c. 8-41 single -f 14-54 paired. Total 53-70. Tem- porals 1-2 -t- 2. PURINA BIMACULATA, D. and B. Text Fig. 7. Furina bimaciilata. Dum. and Bibr., Erp. Gen., VII, 1854, p. 1240. ,, ,, Jan., Rev. and Mag. Zool., 1859, p. 125, pi. VI. Pseudelaps bimamlaliis, Jan., Icon. Gen., Livr. 43, 1873, pi. V., fig. 2. Fnrina himacidata, Uoulenger, Cat, Su Brit, Mus,, III, 1896, p. 406 (see synonomy). ,, ,, Werner, Fauna, Siidw.-Austr., II., 1909, p. 262. There are two specimens of this rare snake in the collection. The largest 317 mm. long and has 181 ventrals, 30/29 subcaudals, 2 anals, 1 + i temporals, the first very long. The smaller specimen which is 230 mm. long differs somewhat from Boulenger’s descrip- 1 Werner — Fauna Siidw.-Austr., II, 1909, p. 25S. RECORDS OF IV. A. MUSEUM. 198] tion in having 217 ventrals, 20/19 subcaudals, and in being nuiformly coloured on the body. The two characteristic black bands on Text Fig. 7, Fitrina limaculata, D. and B. Dorsal view of head enlarged. The upper head shields of the nape and head are present in both specimens, but the smaller one has a black blotch on the rostral. With these variations and those noted by Werner the scale formulae should now stand: — Sc. 15 rows, V. 181-217, S.c. 20/19 to 30/29. Anal 2. Labials f. Temporals i -1 r-2. the larger specimen are figured. BATKACHIA. (Frogs and Toads). Key to the We.stern Australian Frogs And Toads. A. Toes free or fringed only, ' the fringe sometimes forming a very membrane at the base of the toes. B. Inner metatarsal tubercle large, shovel-shaped. C. Teeth in two transverse series behind the choanae. D. A large parotoid-like gland on the calf of the leg. Brown above with dark brown spots or bands and a light dorsal stripe. Limnodynastes dorsalis, Gray, var. typica. DD. No parotoid-like gland on the calf of the leg. Back beautifully variegated with symetrical grey- ish and yellowish marks. Rarely a dorsal band. Limnodynastes ornatns, Gray. CC. Teeth in two groups between the choanae. E. Back granulated. Brown or purplish-black above, uniform or with lighter spots. Heleioporus albopunctatiis, Gray. 1 In the case of Limmdvnastes ornatns the toes may be fringed only, or as much as two-thirds webbed. This species is therefore placed in both divisions of the Key, A. and A. A. RECORDS OF WA. MUSEUM. [199 BB. Inner metatarsal ivibercle moderate or small] not or but little larger than the other. F. Under surfaces G. A large wlnie spot on the hinder side of the thighs. H. A large swollen parotoid gland on the shoulder. Two rather large sub-equal meta- tarsal tubercles. Dark spotted beneath. Uperoleia marmorata. Gray. GG. No large white spot on the hinder side of the thigh. I, Smooth above. No parotoid gland. Abroad dark brown band starts between the eyes and extends to the sacral region where it bifurcates. Beneath spotted with darker. Crinia lead, Fletcher. II. Upper surfaces with prominent smooth warts. An indistinct subcircular parotoid gland on the shoulder. Brown or olive above, lighter on the forehead and a few large patches on the back. Beneath with faint marks Psendophryne guentheri, Boulenger. FF. Under surfaces coarsely wrinkled, not granular. J. Tympanum hidden. Habit extremely stout. Limbs excessively short. Toes not longer than the fingers. Myohntrachus gouhhi, Gray, FFF, Under surfaces granular. K. Tympanum slightly distinct. Vomerine teeth present or absent. Belly immaculate. Hinder side of thighs carmine. L. Back with warts or folds. Middle of back dark brown, A broad dorso-lateral yellowish band, sometimes divided by fine lines. Crinia georgiana, Bibr., var. stolata, Cope. LL. Back quite smooth. A dark lateral streak. Crinia georgiana, Bibr., var. affinis. Giinth. 200 ] RECORDS OF W.A. MUSEUM. KK. Tympanum quite hidden. Vomerine teeth none. Belly dark spotted or marhled. Hinder side of thighs dayk, uniform or variegated. M. Snout rounded, as long as the orbital diameter. Sub-articular tubercles rather small. A slight tarsal fold. Crinia signifera, Girard. AA. Toes distinctly ivebbed from one third to fully so. N. Toes not more than- half webbed. A large oval, com- pressed, inner metatarsal tubercle. O. Teeth between the choanae, P. Habit stout. Head rather pointed. Back with large warts. Toes short, one third to one half webbed Brownish variegated above ; sometimes a fine dorsal stripe. Phractops hrevipalmatus, Gunther. PP. Habit stout. Head very large-, snout broadly rounded. Back finely granulated and with fine dorso-lateral fold. Toes short, broadly fringed, about half webbed. Greyish peppered with black above. Hinder side of thighs uniform. Head variegated. Young spotted above. Phractops australis, Gray. PPP. Habit rather slender. Head long and pointed. Back warty, wdth longitudinal folds. Toes long nearly entirely webbed. Uniform or marmorated brown above. A canthal streak. Hinder side of tlie thighs and groin with black and white spots. Phractops alboguttatus, Gunther. OO. Teeth in two transverse series behind the choanae. Q. Tympanum very indistinct. Head high; snout very short. Back beautifully variegated. Limnodynastes ornatns. Gray. NN. Toes webbed to or almost to the discs. Inner metatarsal tubercle not shovel-shaped, generally not different from the outer when present. R. Vmgers at least one-third webbed. TOIsqs very large. RECORDS OF W.A. MUSEUM. [201 S. Uniform green above (bluish in spirits). Vomerine teeth in two groups on a level with the hinder edge of the choanae. Hyla caeruka, While. SS. Variegated brown above ; hinder side of thighs black marbled. Vomerine teeth in two groups between the choanae. Hyla peronii, Bibron. SSS. Finger discs two-thirds the tympanum ; discs of toes small. Front and hinder side of thighs, groin and axilla, excessively blotched with black. Hyla peronii, Bibr., var. rct/iw, de Vis. RR. F'ingers free or ivebhed at the base. Discs generally small, sometimes not enlarged. T. Fingers webbed at the base. Discs distinctly enlarged, as large as the tympanum or only half so. U. Discs of fingers as large as the tympanum. Head broader than long ; snout rounded. Hinder side of thighs and groin with large purplish black spots. Hyla ewingii, D. and B., var. calliscelis, Ptrs. UU. Discs of fingers distinct, only half the tympa- num. Head longer than broad ; snout prominent. Hinder side of thighs dark, uniform or with white spots. Hyla adelaidensis, Gray. TT. Fingers quite free. Discs of fingers never more than half the size of the tympanum, sometimes scarcely enlarged. Generally a small outer meta- tarsal tubercle. V. Habit moderate. The tibiotarsal articulation of the outstretched limb reaches the nostril or not so far. Wk Hinder side of thighs bluish or greenish, uniform. A broad dorsal and a lateral stripe. Under surfaces uniform creamish- white. Hyla aurea, Lesson, var. typica. 202 ] RECORDS OF W.A. MUSEUM. WW. Hinder side of thighs black with white spots. Belly and under surfaces of thighs with dense black and brown reticulations (absent in young). Hyla aurea. Less., var. cyclorhynchus, Blngr. VV. Habit very slender. The tibiotarsal articulation of the outstretched limb reaches the tip of the snout or far beyond. X. Back smooth or with very faint smooth ivarts. Hinder side of thighs with very accentuated black markings. Hyla latopalmata, Gunther. XX. Back with longitudinal folds and dark with light brown stripes. Hinder side of thighs longitudinally striped with black or brown. Snout and toes excessively long, the tibio- tarsal articulation of the outstretched limb reaching far beyond the tip of the snout. Hyla nasuta, Gray. LIMNODYNASTES ORNATUS, Gray. Limnodynastes ornatus, Boulenger, Brit. Mus. Cat. Batr., 2nd ed., 1882, p. 262. .. ,, Spencer, Kept. “Horn” Sci. Exped. Centr. Austr., II., 1896, p. 156, pi. XII,, figs 3 and 4 and pi. XV., figs. 18-25. Fletcher, Proc. Linn. Soc. N.S.W., XXII. , 1S97 (1898), pp, 676 and 682. Locality. — Napier Broome Bay, Northern coast of W.A. One specimen. LIMNODYNASTES DORSALIS, Gray, var. TYPICA, Fry. Lhnnodynastes dorsalis, Boulenger, Brit. Mus. Cat. Batr., 2nd ed,, 1882, p, 261, ,, ,, Fletcher, Proc. Linn. Soc. N.S.W., XXIL, 1897 (tSgS), P- 675. ,, ,, var. typiea, Fry, Rec. Austr. Mus., X., 1913, p. 24, pi, II, fig. 2. Ten specimens are in the collection. I have utilised these in an already published account of the variations of L . dorsalis. RECORDS OF W.A. MUSEUM. [203 CRINIA GEORGIANA, Bibr., var. STOLATA, Cope. (Plate XXVIII., Fig. i.) Crinia georgiana var stolaia, Boulenger, Brit. Mus. Cat. Batr,, 2nd ed., 1882, p. 264. . Vomerine teeth in two indistinct, convergent groups, behind the level of the choanae. Tongue rather large, elongate oval and entire. Tympanum slightly distinct. Fingers and toes with a slight fringe. Subarticular tubercles well developed ; palmar and plantar surfaces with numerous distinct granules. Two distinct metatarsal tubercles. Tarsal fold very indistinct. Upper surfaces with longitudinal folds and warts. Under surfaces coarsely granular. Colour (Spirits). — Back with a broad dark median area which is variegated with darker. A yellow dorso-lateral band, divided by numerous thread-like lines, starts behind the eye and bifurcating, runs to the groin. Sides of body and head variegated and striped. Legs barred and beautifully variegated with light and dark brown. Groin and hinder side of thigh bright carmine. Under surfaces with a few faint grey spots, most distinct on the throat. Total length from snout to vent ... 24 mm. Width of head ... ... ... 8.5 mm. Length of head to level of tympana 8 mm. Length of the outstretched hind limb 34 mm. This is a very beautiful variety and is very distinct. '1 he specimen figured on Plate XXVIII, fig. i, agrees in all essentials with Bell’s figure ^ of CysttgnaiJiiis georgimus, but I regret that I am unable to refer to Cope’s original description ^ of Crinia stolata. Locality.— A single specimen from the Margaret River in the collection. CRINIA LEAI, Fletcher. (Plate XXVIII., Figs. 2 and 2a. Crinia Uai, Fletcher, Proc. Linn. Soc. N.S.W., XXII., 1897 (1898), p. 677. Vomerine teeth in two distinct, convergent groups behind the choanae, or they may be entirely wanting. Tongue rather large, elongate oval and entire. Tympanum quite hidden. Fingers 1 Bell— Zool. '* Beagle ” Kept,, 1843, p. 33, pi. XVI.. fig. 4. 2 Cope — Journ Acad. Philad. (2), VI., 1867, p. 201. 204 ] RECORDS OF W.A. MUSEUM. cylindrical, not fringed, first shorter than second. Toes with a distinct fringe or devoid of a fringe. Subarticular tubercles low and flat, rather indistinct. Palmar and plantar surfaces devoid of granules. A small inner metatarsal tubercle. No tarsal fold. Upper and lower surfaces smooth except for a granulated area on the under surface of the thighs. Colour (Spirits). — Dark or light blackish-grey above, forehead lighler. A broad black, sometimes light edged band starts between the eyes and runs to the sacral region, bifurcating to a varying extent. This band may be represented by spots only. A dark streak on the canthus rostralis. Lips spotted. Legs and arms transversely barred with black, sometimes absent. A dark, light- edged streak or triangular spot on the hinder side of the thighs near the anus. Under surfaces with numerous blackish-grey spots, the shank, tarsus and metatarsus being much darker, sometimes blackish. Total length from snout to vent ... 22 mm. Width of head ... ... ... 8 mm. Length of head ... ... ... 8 mm. Length of outstretched hind limb ... 34 mm. Locality. — A single specimen is in the collection from the Margaret River, which agrees well with Mr. Fletcher’s description. PHKACTOPS AUSTRALIS, Gray. Chiroleptes austialis, Boulenger, Brit, Mus. Cat. Batr., 2nd ed., 1882. p. 269, ,, ,, Fletcher, Proc. Linn. Soc. N.S.W,, XXII,, 1897 (1898), pp. 67S and 682. There is a single example in the collection which differs from eastern specimens in having a smoother back, slightly narrower head, and a very accentuated canthal and temporal streak. Locality. — Napier Broome Bay. HELIOPORUS ALBOPUNCTATUS, Gray. Text Figs. 8 b. and 10. Ueleioponis albotnnctalns, Boulenger, Brit, Mus. Cat. Batr., 2nd ed., 1882, p. 271. ,, ,, Fletcher, Proc, Linn. Soc. N.S.W,, XXII., 1897 (1898), p. 678. I find a small but constant structure in this species which appears to have hitherto escaped notice. At the anterior corner of RECORDS OF W.A. MUSEUM. [205 the eye (see Text Fig. 8b) is a small fold of integument, beset with three or four conical papillae, which when the eye is closed, fits over the anterior junction of the upper and lower eye-lids, This structure is apparently of secondary importance during burrowing operations which have unfortunately never been observed in this species. However, in allied burrowing forms, and in particular Philo cry phis, flavoguttatus and Limnodynastes dorsalis, var. dumerilii burrowing has been carefully observed and there is no reason to believe that it differs to any extent in H. albopunctaius. The large eyes which under normal conditions protrude considerably, can be withdrawn well into the orbits and even to below the general surface of the head. This is done during the burrowing process by the two species mentioned above, just before the frog’s head disappears beneath the surface of the earth, and it will be found if the mouth be opened and examined at this stage that its cavity is nearly filled by the downward bulge of the orbits. The upper eyelid has little power movement, but by the contraction of the eye into the orbit it is drawn considerably over the eye, while the thin semi-transparent lower lid is drawn up under its outer edge. When so closed, this serrated flap of skin b. Heliofonis albopmctatus, Gray Side view of head, enlarged. 206] RECORDS OF IV. A. MUSEUM. wliich is a direct downward contiimation of the upper eye-lid, would function as a valve to prevent the small particles of grit from entering the eye during the frog’s passage through the earth. In PMlocyylhns flavogutfatus (Text-fig. 8a) an homologous structure occurs, but differs from that of H. albopunctatus (Text-fig. 8b) in being larger and semi-circular in outline, with several nicks on its free edge. It is also to be regarded as of an accessory nature as it is not a direct continuation of the upper eyelid, but quite separate from that structure. Helioporns pictus and Limnodynastes dorsalis, which are both borrowers, possess no such apparatus, but in the former, the anterior free edge of the upper eye-lid overlaps the lower to an unusual extent, and apparently serves the same pur- pose. In Pomatops valvifera, a New Guinea representative of the Family Engystomatidae Dr. Barbour ‘ records an interesting development of this nature. The upper eye-lid is in the form of a flap of skin which extends for some distance anterior and posterior to the eye, and which is sufficiently developed to allow of its laying down so that it covers the whole eye. By analogy Dr. Barbour regards this frog as a borrower. The two specimens in Mr. Woodward’s collection are from the Margaret River. They are the hosts of a great number of the maggot larvae of some Dipterous insect, probably Batrachomyia.^ Each of these maggots causes a gall-like swelling of considerable size and a frog so infested presents a remarkable sight. Mr. Fletcher ® has suggested that “if the generic definition of Heleioporus be amended in respect to the tympanum” his genus Philocryphus might be regarded as synonymous with it. He also notes the very different form of the sternum but apparently did not value it as a generic character. The widely different nature of the sterna of these tw'O species (Text-Figs, ga and loa) seems to me to justify their generic separation, and with such a tangible external character as a distinct tympanum, as opposed to a hidden one in Helioporns, Mr. Fletcher’s genus appears to be very well founded. The chief differences in the sternal apparatus lie in the sternal plate itself. In Philocryphus (Text Fig. ga) it is a densely calcified plate with a slight median ventral ridge ; it is produced backwards into two cylindrical, diverging, bony horns nearly as long as the sternal 1 Barbour — Proc. Biol. Soc. Wash,, XXIII, igio, p. 89, pi. i. 2 Skuse — Proc. Linn. Soc., N.S.W. (2), IV, p. 172. ® Fletcher — Proc. Linn. Soc., N.S.W., XXII,, 1897 (iSgS), p. 678. RECORDS OF W.A. MUSEUM. [207 Text Figs, g, 10 and ii. Text fig. 9. — Philocryfhus flavoguttatus, Fletcher. Text fig. 10. — Helioporus alhopnnctatus , Gray. Text fig. II. — Helioponis pictus, Ptrs. a. Ventral view of sternal apparatus. h. Dorsal view of terminal phalanx of fourth toe. c. Lateral view of same. d. Dorsal view of sacral vertebra. plate, and separated from one another at their tips by a distance equal to their length. In Helioponis albopunctatus (Text Fig. loa), the sternum is a weak, semi-ossified plate, slightly nicked posteriorly; the ossification is confined to two longitudinal tracts one each side of the median line. In Philocryphus the free edge of the left or ventral-most epicoracoid cartilage is slightly calcified and considerably thickened ; it forms a prominent ridge for the attachment of the anterior section of the pectoralis major muscle. No such ridge is discernible on the epicoracoid of H. albopunctatus, nor do the two epicoracoids overlap to nearly the same extent as in Philocryphus. Another difference is apparent in the shape of the omosternum. In H. albopunctatus, it is a simple narrow band of cartilage, while in Philocryphus it is expanded anteriorly into a semi-circular plate. The omosternum, however, shows considerable variation in this (see Text Fig. iia, H. pictus) 2o8] RECORDS OF W.A. MUSEUM. and other Cystignathoid genera, notably Liranodynastes, so that only specific importance can be attached to it. In H. pictus, it bears a much closer resemblance to that of Philocryphus than does H . alhopunctaUis. Philocryphus is a much larger and stouter form than its western ally, and has much more powerful arms. In correlation with this development we find the strongly calcified sternum with its ventral ridge, and the downturned thickened edge of the left epicoracoidal cartilage ; the pectoral muscles which find attachment at these ridges are the largest I have observed in any Australian frog. The sternal apparatus of H. pictus (Text fig. iia) differs from that of H. in minor joints only. The most important difference is the broader sternal plate which shows no sign of ossifi- cation. The omosternum is expanded anteriorly and is not unlike that of Philocryphus, The two genera may be characterised as follows: — HELIOPORUS, Gray. Pupil erect. Tongue subcircular, slightly nicked behind. Vomerine teeth between the choanae. Tympanum concealed. Toes sometimes as much as half-webbed. Outer metatarsals united. Omosternum cartilaginous ; sternum a cartilaginous or semi-ossified plate, nicked behind. Sacral vertebra dilated. Ter- minal phalanges stout, simple. PHILOCRYPHUS, Fletcher. Pupil erect. Tongue subcircular, slightly nicked behind Vomerine teeth between the choanae. Tympanum quite distinct Toes with a thick basal w'eb. Outer metatarsals united. Omo- sternum cartilaginous; sternum a strongly calcified plate w’ith a slight median ventral ridge and two strong xyphisternal horns. Sacral vertebra dilated. Terminal phalanges moderate, simple. MYOBATRACHUS GOULDH, Gray. Myobatraclins govldii, Boulenger, Brit. Mns. Cat. Batr., 2nd ed., 1882, p. 329. ,, ,, Fletcher, Proc. Linn. Soc. N.S.W., XXII, 1897 (1898), pp. 680 and 681. I have examined four specimens of this species and find that they differ in several points from Dr. Boulenger’s description. The RECORDS OF IV. A. MUSEUM. [209 pupil is in every case horizontal, and the tympanum is hidden. The under surfaces are coarsely wrinkled, and although not granular, could hardly be described as smooth. An examination of further specimens will be necessary to decide whether this condition is natural. Localities. — -Two specimens are in the collection. In the Australian Museum are two specimens taken by Mr. H. Richards at the Harvey, eighty miles south of Perth. HYLA CAE RULE A, White. Hyla caerulea, Boulenger, Brit. Mus. Cat, Batr., 2nd ed., 18S2, p. 383. Fischer, Zool. Gart., XXIV., 1883, pp. 21 and 55 (habits). Fletcher, Proc, Linn. Soc. N.S.W., XXII , 1897 (1898), pp. 669 and 681-2. ,. ,, Carman, Bull. Mus Comp. Zool., XXXIX., 1901, p, 14 (note only). ,, ,. Boulenger, Zool. Jahrb., Supplm,, XV., Bd. I., 1912, p. 2ir. ,, irrorala.de Vis, Proc. Roy. Soc, Q'land., I., p, 128. Boulenger, Ann, Mag, Nat. Hist. (5). XVI., 1885. p, 387 {=H. infnifrenala ?) ,, ,, Fry, Rec. Austr. Mus,, IX., 1912, p. too (=caeruka). Locality. — A single specimen is in the collection from Napier Broome Bay. HYLA RUBELLA, Gray. Hyla rubellii, Boulenger, Brit. Mus. Cat. Batr., 2ud ed., 18S2, p. 405. ,, ,, Spencer, Rept. ‘'Horn” Sci. Expdn.. II., i8g6, p. 170, pi. XV.. figs. 26-28. Fletcher, Proc. Linn. Soc. N.S.W., XXII.. 1897 (1898), p 669. Locality.— A single specimen from Napier Broome Bay. HYLA ADELAIDENSIS, Gray. Hyla adelaidensis, Bonlenger, Brit, Mus. Cat, Batr., 2nd ed, 1882, p. 408. „ ,, Fletcher, Proc. Linn, Soc., N.S.W., XXII., 1897 {1898), p. 681. Locality. — Several specimens are in the collection from Harvey. HYLA LATOPALMATA, Giinther. Hyla latopalmaia, Boulenger, Brit. Mus. Cat. Batr., and ed., 1882, p. 414, pi. XXVI,, fig. 4. „ ,, Fletcher, Proc. Linn. Soc., N.S.W, , XXII. ,1897 (1898) pp. 681-2. ,, ,, Fry, Rec. Austr. Mus., X., p. 20, pi. I and Fig. 12. Locality. — Four specimens from Napier Broome Bay. RECORDS OF W.A. MUSEUM. 2I0] HYLA NASUTA, Gray. Hyla nasiita, Boulenger, Brit. Mus. Cat. Batr., 2nd ed., 1882, p. 413. ,, ,, Fletcher, Proc. Linn. Soc. N.S. W., XXII, 1897 (1898), p. 682. ,, ,, Kampen, Nova Guinea, V., 1909, p. 35. ,, peninsulae, de Vis, Proc. Roy. Soc. Q’land., I, 1884, p. 129. ,, ,, Boulenger, Zool Rec., 1885, Rept., p. 24 {— nasiiia.) ? ,, sevioni, Bottger, in Semon’sZool. Forsch., V., 1894, p. it 2, pi. V, fig. i. Locality. — Three specimens from Napier Broome Bay. Col- lected by Mr. G. F. Hill. This species has not been previously recorded from Western Australia. RECORDS OF V/.A. MUSEUM. [211 NOTES ON SOME WESTERN AUSTRALIAN FISHES. ■ — BY ALLAN R. McCULLOCH, Zoologist, Australian Museum. Plates XXIX.XXXL, and Fig, i, The specimens on which the following notes are based are part of the collections referred to in my previous paper dealing with fishes from Western Australia. Most of them are from South- western Australia, but a few from Port Hedland add a little to our knowledge of the fauna of the great North-western Coast, which is, at present, largely unknown. Family RHINOBATIDAE. RHINOBATUS BANKSII, Miiller and Henle. Rhinobatus hanksii, Waite, Mem. Austr. Mus , IV., pt. I., 1899, p. 38, pi. III. Id., Zietz, Trans. Roy, Soc. S, Austr,, XXXII, igo8, p. 292. A young male, 580 ram. long, from Western Australia, does not differ from Port Jackson specimens of this species. R. ianhsU has been identified in South Australia by Zietz. Family CLUPEIDAE. ETRUMEUS JACKSONIENSIS, Macleay. Plate XXIX. Etrumeiis jacksoniensis, Macleay, Proc. Linn. Soc. N.S. Wales, III., 1878, p. 36, pi. IV, fig, i, and loc. cil., IV., 1879, p. 382. Id., Ogilby, Ed. Fish, N.S. Wales, 1893, p. 186. Id., Zietz, Trans. Roy. Soc. S. Austr., XXXIl,, igo8, p. 294. D.21 ; A.ii? P.16 ; Y.g; C.iy. Head 4.45 in the length to the hypural ; height before the dorsal fin 5 54 in the same, and 1.24 in the head. Snout and eye subequal in length, the latter 3.06 in the 1 McCulloch. — Rec. W. Austr. Mus,, I., pt. 2, 1912, pp. 78-97, pis. IX-XIII. 212 RECORDS OF IV. A. MUSEUM. head. Interorbital width equal to the depth of the caudal peduncle, 3.83 in the head. Mandible 2, maxillary 2.87, highest dorsal ray 1.53, and pectoral 1.53 in the head. Body elongate, subcyliudrical and slightly compressed. Head deeper than broad, pointed and flattened above. A thick, trans- parent membrane extends from the snout, over the eye to the preoperculum. Jaws subequal. Maxillary almost entirely exposed, only the upper edge slipping under the preorbital ; it reaches back almost to below the anterior orbital border, and though broad, is scarcely expanded posteriorly. Hinder margin of preoperculum sloping very obliquely backwards and downwards, the angle sharply rounded. The opercular margin forms an obtuse angle posteriorly. No teeth are visible, even under a lense, but the jaws and bones of the palate are a little rough to the touch. Gill openings large, the membranes free from the isthmus. Origin of dorsal almost midway between the tip of the snout and the vertical of the end of the anal. Ventrals Inserted behind the tip of the adpressed dorsal and somewhat nearer the hypural than the pectorals ; the first ray is more than one-third as long as its distance from the anal. Anal very small, its origin much nearer the hypural than the ventrals. Pectorals opening horizontally, the anterior rays as long as those of the dorsal ; their hinder borders are a little emarginate. Scales of moderate size, cycloid, and they appear to have fine longitudinal striations on their exposed surfaces. They are nearly all missing in my specimen, but there seem to have been eighteen on the back before the dorsal fin, and about fifty-six between the hinder margin of the head and the hypural. The bases of the dorsal and anal fins are enveloped in scaly sheaths, and the pectorals and ventrals each have a long flap composed of axillary scales. The colour appears to have been dark blue on the back, sharply separated from the silvery sides and belly. Tips of the jaws blackish. Described from a single specimen, 220 mm. long, from Albany. A few specimens of this species have been taken in Port Jackson, and it is recorded from South Australia by Zietz. RECORDS OF W.A. MUSEUM. [213 Family ENGRAULIDAE. ENGRAULIS ANTIPODUM, Gunther. EngrauUs eiicnisi chains var. aniipodum, ’Gunther, Brit. Mas. Cat. Fish., VII., 1868, p. 386. Eight examples from Freshwater Bay, Swan River, do not differ from Tasmanian specimens in the Australian Museum. Family PLOTOSIDAE. CNIDOGLANIS MEGASl OMUS, Richardson. Ploiosus megaslontus, Richardson, Zool. “ Erebus and Terror,” Fishes, 1845, p. 31, pi. XXI., hg. 1-3. Cnidoglanis megnsloma, Gunther, Brit. Mus. Cat. Fish., V., 1864. p. 27. Id.t Klunzinger, Sitzb, Akad. Wiss. Wien., LXXX., i., 1879, p. 410. Cnidoglanis bostockii, Castelnau, Proc. Zool. Soc. Viet., II., 1873, p. 140. A young example, 177 mm. long, from the Swan River, does not differ from Port Jackson specimens of this species. It also agrees very well with Castelnan’s description of C. bosloclm, so that Klnnzinger’s suggestion of the identity of that species with C. megasioinns is proved correct. Family MURAENIDAE. GYMNOTHORAX PRASINUS, Richardson. Gymnothorax prasinus, Waite, Rec. Austr. Mus., V., pt. 3, 1904, p. 144. Two specimens from Fremantle agree in every detail with others in the Australian Museum from near Sydney. Family ATHERINIDAE. ATHERINA PIN GUIS (Lacepede) Ogilby. Atherina pingiiis, Ogilby, Mem. Qld. Mus., I., pt. I, 1912, p. 38, pi. XII., fig. i. Two examples are in the collection from Fremantle. Waite examined specimens from the same locality, and recorded them as A. laennosa, Forster. ' 1 Waite, Rec. Austr. Mus., IV., 1902, p. 180. 214 ] RECORDS OF W.A. MUSEUM. Family POLYNEMIDAE. Genus POLYDACTYLUS, Lacepede. POLYDACTYLUS (ELEUTHERONEMA) TETRAD ACTYLUS, Shaw. Polynemtis tetradaclyhis, Gunther, Brit. Mus. Cat. Fish, II., i860, p. 329. Id , Day, Fish. India, 1878, p. 180. Id., Macleay, Proc. Linn, Soc. N.S. Wales, VIII, 1883, p. 203. Kent, Gt. Barrier Reef, 1893, p. 287, pi, XLVI, fig. 2, and Nat. in Austr. , 1897, p, 168-9. Polynemus coccus, Macleay, Proc, Linn, Soc. N.S. Wales, II., 1878, p. 354, pi. IX, fig I- Having examined the type of P. coccus in the Macleay Museum, I have to support Klunzinger's opinion that that species is synony- mous with P. tetradactylus. A large specimen is preserved from Fremantle. Family GADIDAK. PHYSICULUS BARBATUS, Gunther. Physictilus barbatiis, Zietz, Trans. Roy. Soc. S. Austr., XXXIII,, 1909, p. 266. Id., McCulloch, Zool. Res. “ Endeavour,” pt. I, 1911, p. 38 — synonymy, Albany. One specimen, 345 mm. long. Family PEMPHERIDAE. Genus LEPTOBRAMA, Steindachner. LEPTOBRAMA MULLERI, Steindachner. Leptobrama mi'illeri, Steindachner, Sitzb, Akad. Wiss. Wien, LXXVIII, i,, 1878,. p. 388, and Denks. Akad. Wiss. Wien, XLI, i., 1879, pi. Ill, fig. i. Id., Klunzinger, Sitzb. Akad. Wiss. Wien, LXXX, i., 1879, p, 381. Ncopempheris rmnsayi, Macleay, Proc. Linn. Soc. N.S. Wales, V., 18S1, p, 517, pi. XIV. NeopempJieris pectoralis, Ramsay and O.gilby, Proc. Linn. Soc, N.S. Wales, (z), II., 1887, p. 563. A comparison of the type of Neopcmphei'is ramsayi with Steindachner’s figure of Leptobrama innlleri leaves no doubt as to the identity of the two species. Mr. Ogilby has examined several specimens from Moreton Bay and informs me that he considers A. pectoralis, may be synonymous with N. ramsayi. I have compared the two type specimens and find only a few differences between them which are probably not specific. Mr. Ogilby will give details in his paper in the memoirs of the Queensland Museum. RECORDS OF W.A. MUSEUM. [215 A fine specimen, 297 mm. long, from Port Hedland, is in the Western Australian Museum collection. Family SERRANIDAE, EPINEPHALUS MEGACHIR, Richardson. Efinephahis migachiv, Boulenger, Brit. Mus. Cat, Fish,, (2 ed.), 1895, p. 219. Port Hedland. One specimen, 230 mm. long. PSAMMOPERCA WAIGIENSIS, Cuvier and Valenciennes. Psammopeica vaigiensis, Boulenger, Brit. Mus. Cat. Fish, (2 ed.), I, 1895, p. 365. A small specimen, 210 mm. long, from Geraldton, differs from larger ones in the Australian Museum from Torres Strait in having the snout more pointed, and its upper profile more concave. Family PSEUDOCHROMIDIDAE, GNATHYPOPS INORNATUS, Ramsay & Ogilby. Plate XXX. Opisthognathus inoriiatus, Ramsay and Ogilby, Proc. Linn. Soc. N.S. Wales, (2), II., 18S7, p. 561. A very large specimen, 485 mm. long, from Port Hedland, differs from the type in having a much smaller eye, the head very much more tumid, and the pectorals more rounded. The scales are obsolete anteriorily and the whole head, body and fins are covered with soft, plicated skin ; on the anterior portion of the ventrals and the margin of the spinous dorsal the skin is particularly thick and convoluted. All these differences are doubtless due to the age of the specimen. I have figured one of the two type specimens, which are in bad condition; only the spots on the body are copied from the larger example. The following is a key to the Australian species of Gnathypops, Gill, and Mevogyimns, Ogilby, all of which are represented in the Australian Museum. 2i6] records of W.A. MUSEUM. a. Outer row of teeth considerably larger than the others. Scales extending forward to below origin of soft dorsal. Gnaihypops. b. Body and fins light with numerous small dark brown spots ----- rnacnlaius . ' bb. Body and fins dark, sometimes with irregular scattered dark spots - - . . . inornaius bbb. Head with small, body with large incomplete brown rings, dorsal, anabaud caudal with oblique bars, the former with a large black spot anteriorily. dannniensis. ’ aa. Outer row of teeth scarcely larger than the others. Scales not extending forward beyond middle of pectorals. Mcrogyinnus. c. Scales very small, covering the greater part of the sides and belly - . . exiiiiiiis. * cc. Scales larger, confined to hinder half of body, belly naked - - - jacksoniensis. “ I have compared the types of Batvcwhus punctiilatus, Ramsay “ and Opisthognathus macnlatiis, Alleyne and Macltay * and find them to be the same species. Family GERRIDAE. Genus PAREQUULA, Steindachner. PAREQUULA MELBOU RNENSIS, Castelnau. Gerres melboimteiisis, Castelnau, Proc. Zool. Soc. Viet,, I., 1872, pp. 158, 245, and lot. cil., II,, 1873, p. 37. /i.. Macleay, Proc, Linn.’ Soc. N.S. Wales, V., 1S81, p. 3S0. Pareqiiula licornis, Steindachner, Deiiks. Akad. Wiss. Wien, XLI, i,, 1879, p. 8. Id., Klunzinper, Sitzb., Akad, Wiss, Wien, LXXX,, i., 1879, p. 380. Chthaiimlopleryx melbnurnensis , Ogilby, Proc. Zool., Soc,, 1S87, p, 616, figure. Id., McCulloch, Zool, Res. " Endeavour ” I , pt. i., 1911, p. 63. Having compared the descriplion of Parequula bicovnis with the specimens on which Ogilby based his genus ChthamaJoptevyx, f have no doubt that the two names refer to the one species. According to Steindachner the scales are ctenoid (ciliated) along the edge and 1 Alleyne and Macleay, Proc. Linn. Soc. N.S, Wales, I, 1877, p. 280, pi. IX, fig. 3. 2 Macleay, Proc. Linn. Soc. N.S. Wales, II, 1878. p. 355, pi. IX, fig. 3. 3 Ogilby, Proc. Roy. Soc. Qld., XXI, 1908, p. 18. ^ Macleay, Proc. Linn. Soc., N.S, Wales, V, 1881. p, 570. « Ramsay, Proc. Linn. Soc. N.S. Wales, VIII, 1883, p. 177. RECORDS OP' W.A. MUSEUM. [217 on the outer surface near the margin, while Ogilby described them as cycloid. This discrepancy is explained by the fact that they appear cycloid to the naked eye, the very minute teeth l^eing only visible under a microscope ; some scales, also, particularly in young specimens, are quite smooth. Nine specimens, 142-185 mm. long, are in the collection from Albany. Family CHEILODACTYUDAE. DACTYLOPHORA, de Vis. Dactylophoia, de Vis, Proc. Linn. Soc. N.S. Wales, Vlll., July 1883, p. 284 {D. seiiiimaciilcilci, de Vis). Psilocraninm, Macleay, Proc. Linn, Soc. N.S.W., VIII, Feb,, 1884, p. 439 (F. Ci)Xii , Macleay). This genus is apparently distinguished from all others of the family Cheilodactylidae in having the cheeks naked. In its short and high anal fin, large scales, and general form, it appears to be related to Chirodactylns, Gill. ^ De Vis has described the teeth as arranged in sevei al rows in the upper jaw, and in one in tlie lower. Mr. Ogilby has, at my request, re-examined the type specimen in the Queensland Museum, and finds a band of villiform teeth in the upper jaw, which is much broader in front than laterally, and three series of small curved teeth in the lower jaw. In full-grown specimens the teeth become much more numerous. DACTYLOPHORA NIGRICANS, Richardson. Cheilodactyliis nigricans, Richardson, Proc. Zool. Soc., 1850, p. 63, and Ann. Mag. Nat. Hist , (2), VII, 1851, p. 279. Chilodactylus nigricans, Gunther, Brit. Mus. Cat. Fish, II., :S6o, p. 79. Id., Gill, Proc. Acad. Nat. Sci. Pliilad., 1S62, p. 118. Id., Gunther, Ann. Mag. Nat. Hi.st.. (3) XX, 1867, p. 59 . Id., Canestrini, Arch. Zool. I'Anat (2), I., 1869. P- i 55 - < Castelnau, Proc. Zool. Soc. Viet., I., 1872, p. 75. Id., Macleay, Proc. Linn. Soc. N.S. Wales, \'.. 1881, p. 423. Id., Johnston, Proc. Roy. Soc Tasm., iSgo (1S91), p. 31. Id., Waite, Rec. Austr. Mus , \H., 1905, p. 63 CliilodaUyins nebiiiosus, Klunzinger, Arch. Naturg., XXXVIII, i., 1872, p. 2C, and Sitzb. Akad. Wi.5S. Wien., LXXX, i , 1897, p. 3C4. Id., Steindachner, Sitzb. Akad. Wiss. Wien,, LXXXVIII, i., 1884, p. 1078, pi- II. , fig. I, Id., Macleay, Proc. Linn. Soc. N.S. Wales, IX, 1884, p. 17, Dactylophoia seviiinaciilata, de Vis, Proc, Linn. Soc, N.S.VVales, VIII 1883, p. 284, Id., Macleay, Proc. Linn. Soc. N.S. Wales, VIII., 1884, p. 441. Psilocraninm coxii, Macleay, loc. cit., p. 440, pi. XXII. Psilocraninm nigricans, Macleay, loc. cit., p. 441. 1 Gill, Proc. Acad. Nat. Sci. Philad,, 1862, p. iig. RECORDS OF W.A. MUSEUM. 218] D. XV-XVI. 24-26 ; A. I II. 9-10 ; P. 9 + 5-6 ; V.I.5 ; C.14-15 ; I. lat. 48-51. Body moderately elongate, deeper in the young than in the adult, 3.24-4.10 in the length to the hypural. Head 3.3- 4.10 in the same. Eye 4.38-6.23 in the head, and i 61-2.23 in the snout. Snout 2.71-3, greatest breadth 2.11-1.77, caudal peduncle 3.56-2,84, longest pectoral ray 1.05-1.1, sixth dorsal spine 2.45- 2.59 in the head. Eye i. 07-1. 47 in the interorbital width. Body covered with large cycloid scales, which become extremely small on the chest, and form a sheath at the bases of the dorsal and anal fins. Postorbital portion of head and opercles covered with small scales ; remainder of the head naked. Fleshy eye- opening almost equal to the interorbital width in young specimens, much narrower in adults. Nostrils close together, in the hinder half of the snout ; the anterior has a short skinny lobe. Lips very thick, maxillary reaching to below the anterior or posterior nostril A band of small cardiform teeth in each jaw which is broad in front, but becomes very narrow on the sides ; vomer and palatines toothless. Spinous dorsal a little shorter than the soft, and its margin is a little arched ; the sixth spine is usually the longest, the others decrease regularly in length. Anterior dorsal rays equal to or higher than the longest spine ; they become regularly shorter, and the margin of the fin is straight. Third anal spine longer, but much weaker than the second ; anterior rays much higher than those of the dorsals, the posterior ones very short, so that the margin is a little excavated. Pectoral, with five or six simple rays, the fifth from the bottom the longest, reaching backwards to above the vent in young specimens and not so far as the ends of the ventrals in adults. Caudal forked. Lateral line almost straight from the operculum to the upper portion of the caudal peduncle ; it is formed of very small scales intercalated between the larger ones of the body, each of which bears a simple or bifurcate tube. The colour markings are very distinct in the young, less so in adults. They consist of six broad brown bands, with darker edges, which descend from the back and run obliquely forward on the sides ; a seventh less distinct one is present on the caudal peduncle. In addition, irregular rows of large dark spots are present on the lower parts of the sides. Two brown bands extend backwards RECORDS OF W.A. MUSEUM. [219 from the eyes, and one below it ; the spaces between them are silvery. All the fins are dusky with narrow white edges ; the soft dorsal may bear several rows of rather large brown spots, while several still larger ones are present on the caudal. I have examined fifteen specimens, including a well graduated series, varying in length from 163-575 rnm. from the tip of the snout to the end of the middle caudal rays. Of these, four are from South Australia. Seven, including the type of Psilocranium coxii are either from the neighbourhood of Sydney, or purchased in the Sydney markets, and one is from the Melbourne market. The three others are said to have come from the Clarence River, New South Wales, but I do not consider the evidence of their capture so far north as satisfactory. Macleay compared his type of Psilocranium coxii with the four South Australian specimens mentioned above which he identified as Chilodactylns nigyicans and Dactylophora semimacnlata} He noted its generic identity with Richardson’s species, but did not recognise that they were all merely growth stages of the one species. I forwarded one of the smallest of these to Mr. Ogilby for comparison with the type of De Vis’ Dactylophora, and he informs me that it cannot be separated from that species. He also agrees that it is the young of C. nigricans. Klunzinger has also described this young stage as C. nebidosa, and Steindachner has given a beautiful figure of it under that name This may be compared with the very rough figure of Psilocranium to show the difference between the young fish and the adult. Family KYPHOSIDAE. KYPHOSUS SYDNEYANUS, Gunther. Pimelepterus sydneyantts, Ogilby, Ed. Fish. N.S. Wales, 1893, p. 40, pi. XVI. Kyphosus sydiieyaniis, Zietz, Trans. Roy. Soc. S. Austr., XXXIII, 1909, p. 267. A single specimen, 235 mm. long, from Houtman Abrolhos, is similar to others from Port Jackson. Zietz has recorded the species from St. Vincent Gulf. 1 See Macleay, Proc. Linn. Soc. N.S. Wales, VIII., 1883, p. 441. 220j RECORDS OF IV.. I . MUSEUM. TEPHKAEOPS TEPHRAEOPS, Richardson, Orenidens tephraeops, Richardson, Zool. " Erebus " and “Terror,” Fishes, 18/17, p. 6g, pi. LXI, figs. 1-2. Tephraeops yichardsonii, Giiniher, Brit. Mus. Cat, Fish, I., 1859, p. 432. Id., Klunzinger, Sitzb. Akad. Wiss. Wien., LXXX, i,, 1879, p. 356. Id., Macleay, Proc. Linn. Soc. N.S. Wales, V., 1881, p. 410. Tephraeops tephraeops, Waite, Rec. Austr. Mus,, VI,, 1905, p. 63 I count about 84 pores along the lateral line and 116 directly above it. Some notes on the teeth are given under Melainbapkes. Fremantle. One specimen, -290 mm. long. MELAMBAPHES, Gunther. Melainbapkes, Gunther, Ann, Mag. Nat. Hist., {3), XI, 1863, p. 115 (M. nigroris, GuiUher = G/rella zebra. Rich.; not Glyphisodon nigroris, Cuv. and Vat,) Girellichlhys, Klunzinger, Arch, haturg,, XXXVIII, i., 1872, p. Z2 (G. zebra. Rich.) Neotephroeops, Castelnau, Proc. Zool. Soc. Viet., I., July, 1872, pp, 68, 248 (AT, zebra, Richardson). Ihis genus differs from Tephraeops only in having the whole operculum scaly, the body scales somewhat larger, and the anal fin with an even margin, the anterior rays not being produced beyond the general line of the fin. The teeth of the two are similar. Richardson described the vomer and palatines as toothless in Tephraeops, while Klunzinger found a group of small teeth on the vomer in In my specimens of both genera, there are two or three minute, isolated, fixed teeth on the vomer, and a similar patch on'the hinder end of each palatine bone; besides these, there are numerous microscopic setiform, dermal teeth, which are quite soft, surrounding the stronger fixed ones. All these teeth are very difficult to find, and might be easily overlooked. MELAMBAPHES ZEBRA, Richardson. CrenuUns zebra, Richardson. Zool,, “ Erebus” and “Terror," Fishes. 1847, P- 70 - Tf/Zirurw/s Gunther, Brit- Mus. Cat. Fish,, I, 1839, p. 432. Id., Ogilby, Cat, Fish. N.S, Wales, 1886, p, 18. Melainbapkes nigraris, Gunther, Ann. Mag. Nat. Hist., (3) XI, 1863, p. 115. (Not Glyphisodon nigroris, Cuvier and Valenciennes.) Girella zebra, Steindachner, Sitzb. .-Vkad Wiss, Wien., LIII, 1866, p. 430, pi VI, fig 2. Neotephroeops zebra, Castelnau, Proc. Zool. Soc. Viet., I, 1872, p. 6g. Id. Macleay, I'roc. Linn. Soc. N.S. Wales, V., 1881, p. 410. Girellichthys zebra, Klamlngee, Arch. Naturg., XXXVIII, i.,1872, p. 22, and Sitzb, Akad. Wiss. Wien, LXXX, i. 1879, p. 356. . RECORDS OF W.A. MUSEUM [22 C Mr. C. Tate Regan has, at my request, very kindly compared Gunther's type specimen oi Melambnphes mgroris wth Steindachner’s figure of Girella zebra, and writes to say that he has no doubt they represent the same species. Castelnau {loc. cit. p. 68) has given reasons for supposing that Gunther’s determination of the Victorian fish as G/y/AiWoK Higf'om, Cuvier and Valenciennes, is incorrect. I fully agree with this conclusion, particularly as the latter species is said to generally resemble G. vakti and G. beiigaleiisis, which are very different in appearance to MehimbapJtes. 1 have examined four specimens from Albany, 180-290 mm. long. Of these one has thirteen dorsal spines, two have fourteen, and one has fifteen. There are 64-70 pores along the lateral line, and go-ioo .scales immediately above it. A skin received from Count Castelnau as Neotephroeops zebra, is also in the Australian M useu m . Family CHAETODONTIDAE. HOLACANTHUS (CHAETODONTOPLUS) PERSONIFER, sp. nov. Plate XXXI. D. XIII. 19 ; A. III. 19 ; V.I.5 ; P.19 ; C.15. Body more elongate than is usual in the genus, the depth before the ventrals 2.3 in the length to the hypural. Head 4.7 in the sanie.^ Eye 3f in the head, as long as the snout without the upper lip. Interor’uital space convex, about half as wide again as the eye. Suborbital bone a little wider than the eye. Maxillary vertical, lower jaw the longer. Posterior preopercular margin finely dentate, lower entire ; the spine is very large, longer than the width of the interorbital space. Teeth cardiform, tricuspid, the median cusp much larger than the others ; they are arranged in about five rows in each jaw of which the outer ones are the longest. Entire head, body and fins, with the exception of the ventral rays, covered with minute, strongly ctenoid scales which are very irregularly arranged. Lateral line distinct anteriorly, scarcely 1 The anterior portion of this specimen having been partially cleaned out and refilled with straw, the head is somewhat distorted, and these measurements may not be quite accurate. ■222] RECORDS OF W.A. MUSEUM. traceable posteriorly ; it originates above the operculum and is curved upwards towards the back, with which it appears to run nearly parallel to the end of the dorsal fin. Dorsal commencing above the end of the operculum, the first spine rather short, the next two increasing rapidly in length, the others becoming gradually longer to the last. The margin of the fin is almost straight, the soft portion broadly rounded behind. Anal of similar form to the dorsal. Upper pectoral rays longest ; the margin of the fin rounded. First ventral ray a little produced, and somewhat longer than the pectoral fin. Caudal slightly emarginate. Colour. — Head grey, with large yellow spots. The anterior portion of the body and chest is bright yellow shading into white, this colour forming a sharp line with the deep black of the rest of the body which forms a backward curve from the first dorsal spine to the ventral fin. The black colour projects forward behind the pectorals on to the coracoid bone, and the basal half of the pectoral fin is also of this colour. Ventrals yellow with white tips. Dorsal and anal black like the body, as is the greater part of the caudal which bears a lunate, yellow, submarginal band with a narrow blackish border. Described from a single example, ago mm. long, from Shark Bay. 1 am indebted to Mr. W. B. Alexander for notes on the life- colours of this specimen. The colour pattern distinguishes this species from all others of the genus known to me, though I have been unable to compare it with H. zebra, H. regtnae, and H. caiidibicolor of Lienard. ^ It appears to be nearest allied to H. conspicillatus, Waite. ^ HOLACANTHUS DUBOULAYI, Giinther. Holacanthus dnhonlayi, McCulloch, Rec. Austr. Mus. IX., pt. 3, 1913, p. 360, pi. XIV. Port Hedland. Two specimens, 130-140 mm. long. 1 Lienard in Sauvage, Hist. Madagascar, XVI , 1891, pis. XXIX., XXXII. . and XXXIV. 2 Waite, Rec. Austr. Mus., III., 1900, p. 203, pi. XXXV. RECORDS OF W.A. MUSEUM [223 Family CARANGIDAE. TRACHINOTUS RUSSELII, Cuvier and Valenciennes ? Trachynoius russellii, Day, Fish. India, 1878, p. 233, pi. LI, b, fig. 3. Trachiiiotiis rtisselli, Stead, Ed. Fish N.S.Wales, igo8, p.29, pi. LXII. D.VII. 24 ; A. II. 1 . 20. Head 3.53, depth 2.24 in the length to the hypural. Eye a little shorter than the snout, 4.3 in the head. Maxillary reaching to below the anterior portion of the pupil. Pectoral 1.3, ventral 2 in the head and reaching to the vent. Dorsal lobe a little shorter than the base of the soft portion of the fin ; anal lobe longer, longer than the base of the fin. Caudal lobes, measured from the hypural, almost equal to the height of the body; 2.26 in the length. An indefinite dark spot above the origin of the lateral line followed by four others placed just above the line, the last very indistinct. Dorsal and anal lobes dark as are the outer margins of the caudal. One specimen, 265 mm. long, from Bernier Island. An examination of the specimens of T. russelii available to me indicates considerable variation in this species and particularly in the depth of the body, length of the fin-lobes, etc., but the West Australian example has a larger head and more obtuse snout that others from Port Jackson and Lord Howe Island. Ogilby has suggested ^ that the records of T. russelii and T. baillonii, Lacepfede, from Australia require verification. It is of interest to note, therefore, that I have collected the latter at Murray Island in the Torres Strait, while I cannot separate Port Jackson and Lord Howe Island examples from T. russelii. Family BLENNIIDAE. BLENNIUS TASMANIANUS, Richardson. Bletutius tasmaniamis, Waite, Rec. Austr. Mus., VI., 1906, p. 205, pi. XXXVI., fig. 5. Id.. Hall, Proc. Roy. Soc. Tasra., 1902, p. i. Seven examples of this extraordinarily variable species from Fremantle, do not differ from some Tasmanian specimens in the Australian Museum. Length 47-77 mm. I Ogilby, Proc. Roy. Soc. Qld,, XXI, 1908, p. 15. 224] RECORDS OF W.A. MUSEUM. Family BATRACHOIDIDAE, Genus PSEUDOBATRACHUS, Castelnau ? ? Pseudohairachus, Castelnau, Res. Fish. Austr. (Viet. Offic. Rec. Philad. Exhib.), 1875 , p. 24 (P. sti'ialus, sp. nov.) Batrachomoeiis, Ogilby, Ann. Qld. Mus., No. 9, 1908, p. 46 (B. minor, sp. nov. —Batrachus dnhius, Shaw). I follow Ogilby in regarding Psendobatrachts and Batrachomoeiis, as probably identical, but prefer to use the earlier name until they can be shown to be distinct. PSEUDOBATRACHUS DUBIUS, Shaw. Lophius dubius, Shaw in White, Voy. N.S. Wales, 1790, p. 265 and plate. Batnxclius dubius, Richardson, Voy. " Erebus ” and “ Terror ’’ Fishes, 1844. p. 16, pi- X, figs. 1-2. Id., Gunther, Brit. Mus. Cat. Fish, III, 1861, p. 169, and .'tnn. Mag. Nat. Hist. (3), XX, 1867, p. 61. Id., Castelnau, X-’roc. Linn. Soc. N.S. Wales, III, 1879, p, 353. Id., Macleay, Proc. Linn. Soc. N.S. "Wales, V. 1881, p. 572. Id., Ogilby, Cat. Fish. N.S. Wales, 1886, p. 31. Batrachoides dubius,Waite, Mem. N.S. Wales Nat. Club, 1904, p. 54. Thalassophrynecocca, de Vis, Proc. Linn, Soc. N.S. Wales, IX, 1884, p. 547. Batrachonioeus coccus, Ogilby, Ann. Qld. Mus., No. 9, 1908, p. 49. Bairachomoeus minor, Ogilby, Ann. Qld. Mus., No. 9, p. 47. Batrachui trispinosus, Kner, Reise " Novara,” Zool., I., :865, p. 189 (not of Gunther). [? Not Batriuhus dubius, Alleyne and Macleay, Proc. Linn. Soc. N.S. Wales, L, 1877, p. 335. Id., Macleay, loc. cit., VIII, 1883, p. 267. Opsamis dubius, Jordan and Seale, Bull. U.S.Fish. Bur., XXV, 1906, p. 416 after Macleay. Fseudebatrachus striatus, Castelnau, Res. Fish. Austr. (Viet. Offic. Rec. Phliad. Exhib.), 1875, p. 24. Bali'achus striatus, Macleay, Proc. Linn. Soc. N.S. Wales, V., 1881, P- 574-] The first notice of this fish consists of a few lines of descriptive matter, accompanied by a very defective figure. No definite locality is given for it, though judging from White’s narrative, it was almost certainly collected in, or very near. Port Jackson. Only one species of the Batrachoididae occurs near Sydney, where it is not uncommon, and is doubtless identical with that figured by Shaw. Ogilby ^ has placed B. dubius in the synonymy of Coryzichtliys diemensis, Le Sueur, but I see no reason to accept this especially as that species is not so far known from New South Wales. Ogilby described specimens from Moreton Bay as Batrachomoeiis minor, and while recognising their probable identity with the 1 Ogilby, Ann. Qld, Mus , No. 9, 1908, p. 51. RECORDS OF IV. A ^ MUSEUM. [225 Figure I. PseudohatracJius duhitis, Shaw 226] RECORDS OF MUSEUM. Batrachus dubius of most authors, though not of Shaw, he considered they differed in having a narrower interorbital space and shorter tentacles. I have compared two specimens received from him as his species, with others from Port Jackson and find them identical in every way. fn a well graduated series of fourteen specimens, 45-320 mm. long, the interorbital width increases regularly with size, being much narrower than the eye in young specimens, and considerably wider than it in adults. The tentacles are always short aiui thick. In Richardson’s figure, by which Ogilby has probably been guided, they are shown much too long and straggling. There is a young specimen 102 mm. long, in the Australian Museum from Garden Island, Port Jackson, caught and beautifully preserved by ray friend Staff Paymaster P. B. Stevens, R.N. It shows the form and arrangement of the tentacles and colour particu- larly well, though the former are proportionately larger than in older examples, and is figured above. Mr. Ogilby has recently informed me he is now sure that Thalassopkyyne coeca, de Vis, is merely the adult of the estuarine form which he described as Batrachomoeiis minor. He also agrees that both are synonymous with P. dubius, de Vis’ specimen being a large example from deep water. A single large specimen is in the Western Australian Museum from Fremantle. Macleay recorded specimens from Torres Strait and New Guinea, but the records need verification, since his speci- mens are no longer in the Macleay Museum and his identifications of all specimens of this family were very faulty. Family ANTENNARIIDAE. ANTENNARIUS UROPHTHALMUS, Bleeker. Chironectes caitdhnaculaius , Richardson, Zool. "Erebus” and "Terror,” Fishes, 1848, p. 125, pi. LX , figs. 8-9 (perhaps not C. caiidimoculatus, Ruppell.) Aniennarius caiidiinaciilaiiis, Bleeker, Atl. Ichth., V., 1865, p. 15, pi. CXCVII. fig. 6. Aniennarius urophthalmus Bleeker, Nat. Tyd. Ned. Ind., II., 1851, p. 488, Id., Gunther, Brit. Mus. Cat. Fish., III., 1861, p. 192. Id., Macleay, Proc. Linn, Soc N.S. Wales, II., 1878, p. 356. Id , Klunzinger, Sitzb. Akad. Wiss. Wien., LXXX, i., 1879, p. 388. A specimen from Broome, 120 mm. long, agrees very well with Richardson’s and Bleeker’s figures of this species. It has not previously been recorded from Western Australia. RECORDS OF IV. A. MUSEUM. 227 PTEROPHRYNE HISTRIO, Linnaeus. Antennarius marmoratus, Gunther, journ. Mus, Godeffroy, V,, 1876, p. 162, pi. C. fig. a. One specimen from Western Australia without definite locality. Length 67 mm. Family TETRAODONTIDAE SPHEROIDES PLEUROGRAMMA, Regan. Tetvodon hypselogemion, Steindachner, Sitzb. Akad. Wiss, Wien, LIII, 1S66, p. 478, Id-, Gunther, Brit. Mus. Cat. Fish., VIII, 1870, p. 277 (part). Id., Macleay, Froc. Linn. Soc. N.S. Wales, VI. 1881, p. 337. Id , Waite, Rec. Austr. Mus., V.. 1903, p. 38 (not 2 '. hvpselo^eneion, Bleeker). Tetrodon pleiirogramma, Regan, Froc. Zool. Soc., 1902(1903), II., p. 300, pi. XXIV fig. 2. A large example, 165 mm. long, from Fremantle, differs from smaller ones from Eastern Australia only in having the white spots on the back larger and more distinct. The Australian Museum collection includes specimens from near Sydney, New South Wales; Moreton Bay, Queensland ; and Lord Howe Island. Family BALISTIDAE. ABALISTES S L'ELLARIS, Bloch and Schneider, var. PHALERATUS, Richardson. Batistes phakratus, Richardson, Stokes’ Discov. in Austr., I., 1846, p. .484, pi. V., fig- 4 - Batistes slellatus, Gunther, Brit. Mus. Cat. Fish., VIII., 1870, p, 212. Two young specimens from Port Hedland, 110-115 mm. long, differ from Indian specimens of the typical stellans in some details of colour marking. Instead of the small light spots on the body, the sides bear numerous large angular grey spots, and the blackish marking on the upper parts is formed of similar darker spots placed closer together. The first dorsal is largely black, and the second dorsal, anal, and caudal are marked with large greyish spots and bands, the last named fin also having a blackish margin. This form is evidently a variety of A. and was described and figured from Western Australia by Richardson as B. phaleratus. 228] RECORDS OF W.A. MUSEUM. EEVISION OF THE FRESHWATER CRAIFISHI'IS OF SOUTH- WE STEEN AUSTEALIA. BY ALLAN R. McCulloch, Zoologist, Australian Museum. Plates XXXII to XXXV. Most of the material on which the following notes are based, was received in January 1912 from Mr. Bernard H. Woodward, Director of the Western Australian Museum. At my request, he very kindly collected together a large series of specimens of all sizes from various scattered localities. These prove to belong to three species of the genus Cheraps, and may be reasonably supposed to be representative of the freshwater crayfishes of South-western Australia. I am indebted, also, to Mr. W. B. Alexander for information on other specimens, since received by the Western Australian Museum. There are also a few examples in the Australian Museum which were collected near Perth and at Albany by Mr. A. Abjornssen, Inspector of Fisheries, Western Australia. While engaged on the preparation of these notes I learnt that Mr. Geoffrey Smith was to read a paper before the Zoological Society of London on the crayfishes of Australia, and I preferred to avoid confusion by waiting until it had been published.^ The greater part of the material available to him was collected in Tasmania and Victoria, though he had a few specimens from South-western Australia. Those I have examined belong to the same species as identified by him, but having a large and apparently representative series of each, I have been able to describe and figure some noticeable variations in their specific characters. 1 Smith, Proc. Zool. Soc., 1912, pt. I., pp. 144-171, pis. XIV-XXVII. RECORDS OF V/.A. MUSEUM. [229 CHERAPS, Erichson. Cheraps, Erichson, Arch, fur Naturg., XII. i., 1846, p. loi. Von Martens, Monastb. Ak. Wiss. Berlin, 1868 (1869), p. 616. Chaeraps, Smith, Proc, Zool. Soc,, 1912, p. 165. (Not Cheraps, Huxley. Proc. Zool. Soc., 1878 (1879), p. 768.) The genus Cheraps was first defined by Erichson as a sub-genus of Astacus, Milne Edwards, for his C. preissii, but his definition does not include some of the most important characters by which the several genera of the family Parastacidae are distinguished. Huxley, in 1868, raised it to the rank of a genus, and described the branchial structures in detail, but he only had a specimen from the Yarra River, Victoria, which he could not, with certainty, identify as a true Cheraps. Taking into consideration Smith’s recent investigations on the Crayfishes of Australia, it is almost certain that Huxley’s specimen was Astacus bicariiiatiis, Gray, which Smith made the type of his Parachaeraps. Therefore, Cheraps of Huxley, but not of Erichson, is identical with Parachaeraps. Smith has again defined Cheraps, his definition being based on C. quiiiqiiecarinatus, Gray, C. teuuimanus. Smith, C. quadricarinatas, Von Martens, and C. intermedius, Smith, but unless these can be shown to be generically identical with the lost C. preissii, their claim to inclusion in Cheraps cannot be proved. For reasons given below, however I regard C. intermedins as scynonymous veith. C. preissii ; this conclusion be accepted, it follows that Smith’s definition correctly applies to Cheraps, Erichson. CHERAPS PREISSII, Erichson. Astacus (Cheraps) preissii, Erichson, Arch, fur Naturg., XII. i., 1846, p. loi. Id., Von Martens, Monatsbr. Ak. Wiss. Berlin, 1868 (1869), p. 617. Astacopsis preissii, Haswell, Cat. Austr. Crust., 1882, p. 177. Chaeraps intermedins. Smith, Proc. Zool. Soc., 1912, p, 168, pi. XXIV., fig. 2, and pi. XXVII.. fig. 34. C. preissii was very imperfectly described from a specimen taken in South-Western Australia and has not again been recog- nised. ^ This unique specimen is apparently lost since Dr. Von t I consider the specimens from Victoria which Ortmann (Zool. Jahrb. VI., 1891, p. 8, pi. I., fig. I.) identified as C. preissii to be almost certainly Parachaeraps bicarinaitis, since it is improbable that a species occurring in the fresh waters of South-western Australia would also be found in Victorian rivers. Some notes on the distribution of P. bicarinatus, are given under the heading of that species. 23 °] RECORDS OF W.A. MUSEUM Martens was unable to find it in the Berlin Museum in 1868, and Professor A. Brauer, Director of that Institution, has very kindly informed me that it is not there now, and that he does not know where it is. For the purpose of this paper, I have examined over one hundred crayfish from several widely separated localities in South- western Australia. Most of these were received from Mr. Wood- ward who made special efforts to secure good series of as many different species as possible. They represent only three species — C. qiiinqnecarinatus, C. tennitnamts, and C. intermedins and hence I suppose that no other occurs in that part of the Continent. There- fore, the lost C. preissii is identical with one of these, and for reasons given below, I believe that one to be C. intermedins. According to Erichson the margins of the rostrum are entire while he does not note the presence of any keels on the carapace. In C. teniiimanns the rostrum is armed with several strong lateral teeth, and the carpace bears three keels in addition to the two extending backwards from the rostrum. C. qiiinqnecarinatus also has five keels. C. interniedius, on the other hand, has only two which are more obtuse and less elevated than those of the other species, while the teeth of the rostrum are sometimes so small as to be easily over- looked ; in addition to these facts, such other characters as Erichson has given, agree very well with my specimens of C. intermedins . Erichson’s specimen was only three inches five lines long, and the chelae measured one inch in length and five lines in breadth. These proportions agree much better with Smith’s C. intermedins than with another which I propose to separate as var. angiisius, in which the chelae are much narrower, particularly in young specimens. The following are the principal characters of the species as exhibited by eighteen specimens 95-161 mm. long from the tip of the rostrum to the end of the telson. Carapace finely pitted above, minutely tubercular on the lower parts of the sides, the tubercles a litile larger anteriorly ; a series of slightly enlarged tubercles on the sides behind the cervical groove. Two obtuse keels, each terminating anteriorly in a very blunt spine or tubercle. Rostrum almost horizontal or obliquely deflexed, not reaching the end of the antennal scales; it is flat above and the margins are defined by very narrow, scarcely raised keels. RECORDS OF W.A. MUSEUM. [231 It ends in a spine and there are two, inconspicuous, lateral serra- tions on each side near the tip. Abdomen with pits, but otherwise smooth. Chelipeds massive, the hand either half, or usually more than half as broad as long. It is smooth and rounded above with minute, rather scattered pits which become larger and more crowded on the outer surface. Inner margin raised, with seven to nine serrations which form rounded tubercles in large specimens ; a tomentose patch inside this margin. Fingers either meeting along their whole length or more or less widely gaping. Mobile finger with a rounded tubercle about the middle of its length, which may be either extremely small, or large and followed by several smaller ones. Immobile fingers with a row of larger or smaller tubercles along the basal half. Wrist with a large blunt internal spine and sometimes a second smaller one at its base. The colours are faded, but the chelae appear to have been purplish with many lighter spots. Of the eighteen specimens, eleven are from Kojonup ; six were collected in a billabong at Harvey, Harvey River; and one from Guildford, near Perth. Mr. Alexander has also seen it from streams outside the Mammoth Cave and inside the Calgardup Cave, in the cave district. South-western Australia. The specimens from the latter were living absolutely in the dark. CHERAPS PREISSII. var. ANGUSTUS, var. nov. Plate XXXII. This variety differs from the typical form only in having more slender chelipeds and legs, and a slightly narrower carapace. The chelae are particularly narrow in my smallest specimen, and in this respect differ greatly from others of the typical form of about the same size in which they are almost similar to, those of the adults. The colour, after long preservation, is a dark violet with light bluish areas on the sides and under parts. The plate illustrates the only three specimens I have seen which were collected for the Australian Museum by Mr. A. Abjorns- sen, near Albany. They are 141, 102, and 78 mm. long from the tip of the rostrum to the end of the telson, and the largest and smallest are females. 232 ] RECORDS OF W.A. MUSEUM. CHERAPS QUINQUECARINATUS, Gray. PLATE XXXIII. Astacus quutque-carinatus, Gray, in Eyre, Journ. Exped, Centr. Austr., I, 1845, p, 410, pi. Ill, fig. 3. Astacus (Cheraps) qiiinquecarimtus. Von Mai tens, Monatsbr. Akad. Wiss. Berlin, 1808 ( 1869), P- 6i6- Astacepsis quinqne-carinaius, Haswell, Cat. Austr. Crust. 1882, p. 176. Chaeraps quinque-carinutus, Smith, Proc. Zool. Soc., 1912, p. 165, pi. XXIII and XXVII, fig. 25-29. I have examined thirty-eight specimens of all sizes between 45 and 131 mm. long, from the tip of the rostrum to the end of the telson, and find but little variation in the essential characters of thi.s species. There are some differences in the form of the chelae, which consist chiefly of an alteration in the shape of the fingers, they being narrow and somewhat pointed in some, and much broader and more obtuse in others. A tomentose patch on the upper surface of the hand, near the inner margin, is usually present, but may be absent. Carapace finely pitted above ; large specimens w'ith some very minute tubercles on the lower anterior parts of the sides. A row of enlarged tubercles behind the cervical groove. Back with five keels, of which two are liackward extensions of the lateral margins of the rostrum. The median keel arises between these last, and may run liackward to the cervical groove or terminate well before that point ; the outer keels form two sharper or blunter points anteriorly. Rostrum either horizontal or rather obliquely deflexed not reaching the ends of the antennal scales ; its sides are elevated, leaving the upper surface more or less concave. It ends in a spine, and there are usually two, rarely one or three denticulations on each side near the tip. Its length and breadth is subject to a little variation. Abdomen pitted but otherwise smooth. Chelipeds rather massive. The hand is comparatively elongate, its breadth being always distinctly less than half its length. It is smooth and rounded above with small scattered pits, which are a little larger and more crowded on the outer surface ; the two largest specimens have a narrow, irregular groove along the middle of the upper surface. The inner margin is raised with seven to nine serrations ; usually there is a tomentose patch just within this margin. Fingers either meeting along their entire length, or with a larger or smaller gape between them. They are armed with one or RECORDS OF W.A. MUSEUM. [233 more small rounded tubercles on their inner margins which are scarcely developed in the smaller examples. Wrist with a sharp internal spine, and often several smaller ones near its base. Such specimens as retain any traces of colour appear to have been purplish with minute pale dots on the chelae. The material described above was collected at the following localities in South-western Australia; — Chidlow’s Well, north-east of Perth ; near Perth ; Cannington, south of Perth ; seven miles above Harvey, Harvey River ; Korijekiip, Harvey River. According to Mr. Alexander, this species, which is known as the gilgie, is probably restricted to the smaller streams which do not dry up in the summer. In addition to the above noted localities, he has identified specimens from the Vasse River. CHERAPS TENUIMANUS, Smith. Plates XXXIV & XXXV. Chaeraps teiiuimaiiiis, Smith, Proc. Zool. Soc., 1912, pt. I., p. 166, pi. XXII., and pi. xxvn., fig. 30. A well graduated series of forty specimens, ranging from 57-296 mm. long from the tip of the rostrum to the end of the telson, shows considerable variation in the form of the chelae and rostrum, as well as in the armature of tlie hitter. The chelae are always much more slender in young specimens than in adults, while large specimens always have the fingers bent more or less obliquely inwards. The teeth on the inner margin of the palm are scarcely developed in small specimens, bur become very prominent in adults ; there are usually seven or eight, but may be as few as three. Though these alterations are correlated with growth, yet specimens of the same size are not always equally changed, some retaining their juvenile characters longer than others. Some different forms of chelae are illustrated on plate XXXV. The rostrum has from three to six denticulations on either side which, as often as not, are not paired but more or less alternate. Its length and breadth are also variable ; it may reach well beyond the antennular peduncle, or only to the middle of th-e last joint, but it never attains the tip of the antennal scale. A few of the most striking variations shown in my series are figured on plate XXXV. 2341 RECORDS OP W.A. MUSEUM The following are the principal characters of the species. Carapace finely pitted above ; large specimens have numerous sub-acute tubercles on the back and sides behind the cervical groove which can be traced more or less distinctly in all but the smallest specimens. There are usually four sharp spines on the sides directly behind the cervical groove. Back with five keels, of which two are backward extensions of the lateral margins of the rostrum. The median keel rises between these last and may rim back to the cervical groove, or terminate well before that point. The outer keels each end in an acute spine anteriorly. Rostrum nearly horizontal or obliquely smooth in younger specimens but with numerous large and rounded tubercles in adults. Chelae slender in young, massive in adults ; the breadth varies from more than four to about two-and-a-half in the length, according to age. They are smooth and rounded above, with minute scattered pits which are more crowded on the outer surface. The inner margin is not raised upwards, and is usually provided with seven or eight tubercular denticulations. There is no tomentose patch on the upper surface in any of my specimens. Fingers either meeting along their whole length or with a greater or smaller gape between them. They each have a large tubercle on the basal halves o their inner margins, with one or two smaller ones behind them. Wrist with a moderately large internal spine. This large and handsome species is apparently confined to the rivers of South-western Australia. Of the series examined the greater number were obtained in tlie Harvey River, seven miles above Harvey; one is from Korijekup, Harvey River ; one from Margaret River; one from Balingup Brook; and three from the Warren River. Mr. W. B. Alexander informs me that it is known as the Marron, and that he has also seen it from Kojonup. Genus PARACH AERAPS, Smith. PARACHAERAPS BICARENATUS, Gray. Parachaeraps hicarinatus, (Gray) Smith, Proc Zool. Soc. 1912, pt. I., p. 163, pi XXI., and pi. XXVI., figs. 15-24. Smith has described specimens of this species from Victoria, and states that “ it is widely distributed all over Central, Northern, and Western Australia and in Queensland.” Unfortunately, the RECORDS OF IV. A. MUSEUM. only definite localities he gives are a few in Eastern Australia. I am unable to find any records of its occurrence in Western Australia, though since it lives in the deserts of Central Australia, and at Port Essington it possibly extends into the central and northern parts of the Western State. As already stated it is not represented in a series of more than one hundred specimens from several widely separated localities in South-western Australia, its place being apparently taken by Cheraps pyeissii which resembles it both in appearance and in habits. 236J RECORDS OF W.A. MUSEUM. ON A SPECIMEN OF REGAT.ECUS GLESNE. ASCAN.. OBTALNED IN AYESTERN AUSTRAEIA. BY — W. B. ALEXANDER, M.A. Plate XXXVIII. On January 8th, 1913, the Museum obtained through Captain A. Abjornssen, Inspector of Fisheries, a fine specimen of an Oar- fish. It had been found on the shore of Rottnest Island a few days previously in a dying condition by a boy named Backhouse. As this curious fish is so rarely met with, as some doubt exists as to whether there is only one cosmopolitan species of Oar-fish, or whether there are a number of closely allied forms, and as this is the first authenticated record for the west coast of .'\ustralia, it seems worth while to give a short description of the present example. Before doing so it will be well to notice briefly the records of previous occurrences of Regalecns in Australasian waters. The first example seems to have been obtained at Nelson, N.Z., in i860, and since then .some 15 other examples have been recorded from New Zealand. Of these, three are of special interest since they were made types of new species. One obtained at New Brighton near Christchurch in 1876 was described by von Haast as R. pacificus (Trans. N.Z. Inst. X., p. 246) whilst one obtained at Moeraki in Otago Harbour was described by Parker as R argenteus (Trans. N.Z. Inst., XVI, p. 284). Another, also obtained in Otago Harbour was made the type of R . pavheri by Benham (Trans. N.Z. Inst., XXXVI, p. 198). Other New Zealand specimens have been referred to R. gladiiis and R. grillii. 1 can only find references to four previous occurrences of Regalecns in Australia, all from the East or South-east. The first obtained in Bass Strait in 1878 was described and figured by McCoy (Prodr Zool. Viet., vol. IP, p. i6g) under the name of R. banhsi. The second, found on the beach in the Tweed River District of Queensland, was made the type of a new species, R. masferii, by de Vis (Proc. Roy. Soc. Q’land, 1892, p. 109). I he RECORDS OF tV.A. MUSEUM. [237 third, obtained at Shark Beach in Port Jackson, N.S.W., in 1899, was described under the name oi R. glesne by E. R. Waite (Rec. Austr. Mus., Ill, p. 163), whilst it appears that a fourth was stranded below the lighthouse at Cape Everard, in Victoria, in 1896 (Melbourne Leader, Aug. i, 1896, p. 7). The present is there- fore the fifth Australian specimen to be recorded in print, but from information obtained by Mr. O. Lipfert it would appear that a specimen was washed up on the beach at Bunbury, W.A., some ten or twelve years ago. Mr. Lipfert’s informant states that he was on the beach with one or two other men when they observed the fish swimming on the surface, that as it neared the shore a shark bit the middle pottion of the body right through, that the two halves were soon afterwards washed upon the beach and that he had the long oar-bones in his possession for many year.s after- wards. The only specimen of Regalecns which has previously been recorded from the Indian Ocean was obtained at Vizagapatnam, in March, 1788. It. remains the unique specimen of Regalecus nissellit, Shaw, characterised especially by the presence of a distinct caudal fin. As it was only 2 feet 8 inches long it is possible that R. nisselln was founded on a very young example and that the caudal fin is lost in later life. The tail is, however, so fragile that very few, if any, other examples have been obtained in which it is perfect. Most recent authorities who have dealt with the genus have agreed that it is better to regard all the examples known, with the possible exception of if. russellii, as forms of one variable species, if. glesne, Ascan. The present example falls within the limits of v^ariability assigned to this species by Goode and Bean in their “Oceanic Ichthyology ” in all respects, except the smaller number of dorsal rays, but it would not agree with any one of the other species described from Australasia. As in most other examples our specimen was not perfect, as the fragile crest, the long ventral fins and the tip of the tail were all broken. The two former breakages had evidently occurred when the fish was washed ashore, and portions of the crest and oars were obtained with the fish. The tail, however had been broken at some former period, within a few inches of the tip and had subsequently healed. RECORDS OF W.A. MUSEUM. 238] The specimen is a comparatively small one, the total length being nine feet seven inches. The following features are those in which different examples of Regalecus vary : — Height from ijio. to 1/24 of the length (Goode and Bean) ; the present example was 8 ^ inches in height and g ft. 7 ins. long, hence the height was 1/13 of the length. Length of the head contained from 16 to 20 times in the length of the body (Goode and Bean); the present example had a head 7 inches long, or 1/16 of the length of the body. Teetli minute or absent (Goode and Bean); in this ca.se they are absent. Diameter of eye 4 to 6 times in length of head (Goode and Bean) ; the eye has a diameter of inches or 5.6 times in length of head. Dorsal rays 275 to 400 (Goode and Bean) ; our specimen has only 205 dorsal rays, and the small missing portion of the tail could not possibly have borne more than another 20. Pectoral rays ii to 14 (Goode and Bean) ; they are 12 in this example. Goode and Bean also state that the skin has numerous bony tubercles, Parker point.s out in his de.scription of R. avgenteus that the tubercles which appear to be bony as long as the skin is moist, disappear when it is allowed to become completely dry. I found that this was the case with the present specimen. The bases of the rays of the crest also agree precisely in their arrangement and relative thickness with those so caerfully described by Parker (loc. cit. and Trans. N.Z. Inst., XX, p. 20.) On the other hand, the shapes and relative sizes of the oper- cular bones differ very markedly from those figured by Parker, and indeed from those in all the other figures which I have been able to find. The accompanying photographic illustration will show this better than I can describe it (cf. Benham, P.Z.S., igo6, p. 544.) Unfortunately I was absent when the specimen reached the Museum and the sex was not determined nor the contents of the stomach noted. A plaster cast was prepared and the missing fins added and coloured from the details given by Clarke (Trans. N.Z. Inst., XXX, p. 253). The head and skin are preserved in the Museum collection, { Registered No. P 23). RECORDS OF' IV. A. MUSEUM [239 A BRACHIOPOl) NEW TO AUSTRALIAN WATERS. — BY — W. B. ALEXANDER, M.A. On July 26, 1913, I visited Cottesloe Beach to look for marine animals which might have been washed up lay the stormy weather of the earlier part of the week. My search was well rewarded by the discovery of a Brachiopod shell in good condition which I was able to identify from Davidson’s work on “ Recent Bracbiopoda” as Terehratulina radiata. Reeve. The species is characterised by the pair of ribs which run for- wards along the dorsal valve, separated by a deep groove. On the ventral valve the middle line is occupied by a strong rib, with a groove on each side. The deltidial plates in my specimen are united, thus completing the foramen ; Davidson states that this condition “ appears to be the exception and not the rule.” My specimen is II lines long, 8 broad and 6 deep; these dimensions agree with those given by Davidson except that for breadth, which he gives at 9 lines. He states, however that he had seen “ a great many specimens of the shell, some as wide as long, others longer than wide”, so that the width is a very variable feature. It is possible that this species is really a variety oiTerebratnlina cancellata, Koch, which appears to be the only Brachiopod previously recorded from Western Australia, from which country the type- specimens were said to have been brought previous to 1843. This latter species occurs on the coasts of New South Wales, Tasmania and South Australia, and one would suppose that if T. radiata were only a variety of that species it would have been found with it in the other States. Davidson says on this subject : “ Mr. Lovell Reeve considers this to be a good and well-marked species, and in this statement I feel disposed to concur. I have seen a great many specimens of the shell ... all presenting a more or less well-marked l)ipll- cation. It is a smaller shell than T. cancellata, its nearest ally.” 240] RECORDS OF W.A. MUSEUM. The latter attains a length of i inch 7 lines. With regard to habitat, Davidson states : “ Mr Cuming possesses two or three specimens, all exactly alike, procured, he fancies, from the dredgings of Sir 1 £. Belcher in the Strait of Corea. I have seen and possess a number of specimens of this shell, which Mr Sowerby assures me were obtained near the Cape of Good Hope, its probable habitat.” Since Davidson’s work was published 25 years ago, it is quite possible that the species may have been found in the interval. I have looked through the papers dealing with Australian Brachio- poda by Dr. Verco and Mr. Hedley, neither of whom records this species in the lists for other parts of Australia. I think, therefore, that I can safely claim TerehratuHna radiata, Reeve, as a species new not only to this State, but to the whole of Australia. RECORDS OF W.A. MUSEUM. [241 A NEW SPECIES OF EMBIIl) FROM MT.STERX AUSTRALIA — BY — Dr. K. FRIEDERICHS (Apia, Samoa). (The English translation has not been revised by the Author). The following description is founded on two dry examples which I received for identification from the Western Australian Museum at Perth. OLIGOTOMA HARDYI, n. sp. Male. Winged. Posterior branch of the radial ramus ^ not forked in either wing, rather strongly developed, but not reaching the tip of the wing ; the same reniark.s apply to the median, whilst the cubital is absent. Anal present. The wing- veins which are present, as well as the line of the absent cubital, are brown-edged ; so that, apart from the anal, five double brown longitudinal lines are present and the wings look rather darkly striped to the naked eye. Transverse veins numerous. In the forewing six between the anterior margin of the wing and the first radial branch (these are only weak, the rest are stronger) ; four between the first and second radial branches ; three between the latter and its successor ; and two in the space to the median. In the hindwing the corres- ponding numbers are; three, four, two, two. As in the other species the number of transverse veins varies. 'I'his species may be said to have much better developed wing-veins than all other known oecies. A’. .ennae with 21 segments. Eyes large, projecting, kidney- shaped, broader than long (as seen from above). The whole head about times as long as the eye. Lateral borders of the head converging posteriorly and slightly rounded. 1 The terminology employed is that of Enderlein’s Monograph of the Embiidae. 242 ] RECORDS OF W.A. MUSEUM. Prothorax somewhat narrower than the posterior margin of the head, the apotom separated off by a distinct constriction. Meso- and ineta-notum naked (the rest of the body covered with stiff hairs.) Legs without peculiarities (so far as I could ascertain from these dry specimens). Front tarsi with the normal oval metatarsus, convex above,*for use in spinning. Extremity of the abdomen; loth tergite divided longitudi- nally ; the right half produced into a long pointed process, the left with a short round termination. The appendage of the gth sternite (penis ?) projects straight back and appears to be bifid ; its anterior portion is hidden by the middle part of the loth tergite. Basipodite of the left cercus flattened, produced inwards ; the first joint of the left cercus strongly club-shaped, posterior half produced inwards and toothed on that side ; the second joint short, thick and cylindrical. (It is difficult to determine the precise structure of the extremity of the abdomen in dry specimens). Colour. — Dark brown, meso and meta-notum and parts of the legs lighter; the first segments of the antennae yellowish brown. Dimensions. — Specimen i — length ii| mm. Specimen 2 — length 10 mm. Wings the same size in both specimens — length ii-io mm., breadth 2|-2j mm. Habitat. — Western Australia. Captured in Perth by Mr. G. H. Hardy, now at the Tasmanian Museum, Hobart. They flew into a room in company with several other individuals attracted by the light of a lamp in June, 1912. Remarks. — This species does not agree with the diagnosis of the genus Oligotoma given in either of the Monographs (those of Krauss and Enderlein). Yet, there is no doubt that O. hardyi is an Oligotoma. It differs from Enderlein’s diagnosis in the strong development of the median and the posterior branches of the radial ramus, and in the absence of a process on the left half of the lolh tergite. From Krauss’ diagnosis it differs in possessing teeth on the much-thick- ened first joint of the left cercus. O. heymonsi, End., also differs from this diagnosis in having the same organ toothed, whilst in 0 . greeniana, End., the left process of the loth tergite is also absent. RECORDS OF W.A. MUSEUM. 243 Whilst in Sydney I had the opportunity of examining the type (male) of Froggatt’s Australian Embiid (0. gimieyi) in the Agricul- tural Museum. It is a dry specimen without an abdomen. 1 he neuration characterises the species as an Oligotoma. From the pre- sent species it may be distinguished by its smaller size and lighter colour (the whole body is light brown, the head no darker than the thorax) as well as by the wing- veins. The (unforked) posterior por- tion of the radial ramus is only well-developed at its proximal end, the remaining part beii}g only faintly indicated. The same is true of the median and the cubitus. The eye as seen from above is almost circular (as in Enderlein’s figure of 0. sanndersi). It is possible that it may be one of the cosmopolitan species (sanndersi or latreillei) but a reliable identification of this specimen is obviously out of the question. 244 ] RECORDS OF IV. A. MUSEUM. THE MAMMOTH CAVE [Continued). By LUDWIG GLAUERT, F.G.S., etc. ORDER MONOTREMATA. Earn. Tachyglossidae. Zaglossus Gill (1^77). Pi'oechiditii, Gervais .. Osteog. Monot. Viv et Foss., p. 43, 1877. Acauthoglossiis, Gervais .. Comptes Rendus, Baris LXXXV No. ig, p. 838, 1877. Bruijnia, Dubois .. .. Bull. Soc. Zool. Franc. VI for 1S81. No. 6, p. 267-270, 1882. ZAGLOSSUS HACKETTI, sp. nov. HACKETT’S GIANT ECHIDNA. Plates XXXVI-XXXVII. The bones about to be described were collected in the Mammoth Cave, where they where found intermingled with the remains of Sthetiunis, Macvopus, etc., in the older portion of the deposit first examined in 1909. They comprise the atlas vertebra, the clavicles and episternum, the pelvic girdle, two femora, a tibia and a radius. The limb bones are all twice the length of those of the living species of Echidna [Tachyglossus acnleatus) and in addition are stouter in proportion. The bones of the trunk show a similar superiority in dimensions. Fossil Echidnas of large size have been known in Australia for many years. In 1868“ Krefft described the proximal end of a humerus which he named and in 1884’’ Prof. Owen examined the cast of a more perfect example (E. ramsayi) which is now regarded as identical with the former species, to which have also fieen ascribed the other remains of large fossil Echidnas in the Australian Museum, Sydney. In 1895^ Mr. W. S. Dun fully described the imperfect skull and atlas vertebra of a larger animal, which he named Echidna (Pi'oecJiidna) vohnsta. 1 Ann. Record Science and Industry for 1876, p. clxxi, which appeared on May 5th, 1877. Gervais Osteog. Monotremes Viv. et Foss,, p. 43, was published on Nov. 30th, 1877, hence Gill’s name has priority over that proposed by Gervais. Fide Palmer, Ann Mag, Nat. Hist, (4) Vol. i, p. 113 (1863). » Phil. Trans, 1884, p, 273. ■i Rec. Geol. Surv, N.S, Wales, vol. IV, part 3, p. 121 (1895). RECORDS OF W.A. MUSEUM. [245 Description of the Remains. The alias vertebra is perfect, its inferior arch completely ossified and the transverse processes fused to the rest of the vertebra. Its neural arch is well developed antero-posteriorly, it has a steeply inclined anterior surface and a gradually sloping posterior one. The neural .spine is represented by an anteriorly directed tubercle from the base of which a rounded ridge extends to the anterior border of the transverse processes. The greatest width of the anterior surface of the arch is but 6.5 mm. compared with 10.5 in E. (P ) rohustn. The extent of the posterior surface, measured from the tip of the neural spine to the posterior margin is 15 mm., the arch is not as flat as in Dun’s specimen and is much more slender, although it helps to form a larger concavity for the reception of the odontoid process of the axis. The inferior arch is slender and agrees with Dun’s figures, ^ but the curve is not quite so regular, resembling Tachyglossiis in this respect. Both the anterior zygapophyses are large, irregular in outline, with an anterior lobe, their lower end passes imperceptibly into the inferior arch. They are concave vertically and almost flat horizontally. The posterior zygapophyses are ear shaped and possess well-marked posterior lobes ; downwards they merge gradually into the inferior arch, their apex projects from the lateral mass to a marked degree ; they are but slightly concave vertically and horizontally. Tlie trans- verse processes are complete, they extend outward and backward with a slight downward curve and a well marked ridge on their external border ; there is a slight concavity in the anterior region of their upper surface, but the main mass of each process is convex above and below. The area bounded by the neural and inferior arches is deeper and broader than in E. (P.) robusta. The extreme width of the vertebra, including the transverse processes is 59 mm. and the heighth to the tip of the neural tubercle 33 mm. The united clavicles and episternum form a bow-shaped mass whose constituents are completely fused, rendering it impossible to distinguish the individual clavicles and the upper part of the T-shaped episternum. It is roughly trihedral in section, the well marked antero-superior ridge gradually pas.sing downwards to the I Loc. cU. pi. XI, figs. 5 and 6. 246] RECORDS OF W.A. MUSEUM. origin of the vertical member of the episternum. The inferior and the postero-superior angles are more rounded and constant in position. The imperfect pelvic girdle comprises two fused sacral verte- brae and part of the right innominate bone with the complete, perforated articular cavity and the adjacent portions of the ilium and ischium. The sacral vertebrae, the second and third of the series, have marked intervertebral cavities and their neural spines fused together at their tips, the neural arches also send out bony processes forwards and backwards to completely cover the neural canal. There are the usual lateral openings for the passage of the nerves. The anterior vertebra is attached to the ilium by the usual lateral growths from centrum and neinal arch and the posterior one by the pleurapophysial ossification only. The vertebrarterial passage on the right side is larger than the neural canal. The ilium is trihedral its fusion with the other components of the innominate bone complete, its acetabular border short and concave, with the prominence for the attachment of the rectus (extensor) muscle of the leg forming a prominent compressed sub- limate area close to the rim of the acetabulum — in the Australian Tachyglossus this surface is markedly triangular. The acetabulum is large, more ovate than in Tachyglossus, somewhat flattened on the dorsal surface, incompletely ossified but with an entire margin ; its dimensions are 40 mm x 34 mm. The pubis has a well- marked pectineal process which is comparatively more developed in Tachyglossus. Very little of the broad flat ischium is preserved. Both femora are present in the collection. The complete right one (plate XXXVI, fig. i) is short and broad, flattened from before backwards with a well defined head that rises but little above the great trochanter on a short and indistinct neck; in this it differs from the femur of T. a. ineptiis (plate XXXVI, fig. 2), and from the fossil Z. [Echidna) oweni specimen F13580 of the Australian Museum, represented in this collection by a cast, whose heads rise considerably above the trochanter. The great trochanter is much more massive than in the living form and gives rise to a strong, rough, projecting ridge that extends along two-thirds of the shaft, where it ends abruptly. The lesser trochanter on the inner side of RECORDS OF W.A. MUSEUM. [247 the shaft is situated more distally then in the living Echidna, rising sharply at a point 20 mm. from the head, from this point a well- defined ridge runs to the distal. end of the shaft. At the longitudinally compressed, and transversely expanded distal extremity, the trochlear articular surface is well marked, in front it is slightly curved transversely and convex vertically, posteri- orly it is divided into two condyles by a wide and deep intercondylar notch, which in comparison is deeper and broader than in the living Echidna. The tibia (plate XXXVII) is- straight, laterally expanded at the proximal and distal ends with a relatively slender shaft ; the anterior convexity of the proximal end of the shaft (the tuberosity for the tendons of the extensor muscles) is separated from the lateral expansions by two shallow grooves rvhich gradually diminish as the shaft becomes more slender, till finally the anterior surface becomes a regular convexity. At the distal end of the bone and at some distance above it on the posterior aspect (plate XXXVII, fig. 3) the concavities and convexities for the reception of the bones of the ankle joint and of the adjacent fibula are very prominent ; they show no material difference in arrangement from the corresponding features in the Echidna T. a.ineptus represented in the Museum Collection. The fore limb is represented by the right radius only. The bone is more slender than the tibia, has its heads much expanded laterally ; these are excessively developed in directions at right angles to one another. The oval hollow for the reception of the humerus is deep with a marked thickening at its posterior margin. The distal end has two concavities separated by a convex ridge to fit into the convexities of the scapholunar, and the small sesamoid bone developed in the tendon of the flexor carpi radialis, the concavities are sub-equal, but in Tachyglossus acideatus inepUis, theliving species of Western Australia, the outer one is much the larger. The interosseous ridge and the tracts for the attachment of the muscles are also situated more distally, but the most noticeable difference between the species is the extreme robustness of the fossil form. The minimum antero-posterior diameter of the radius of the living species is 3.5 mm., whereas, in the bone of the fossil species 248 ] RECORDS OF IV. A. MUSEUM. which is approximately twice the length, the least diameter in the same direction is 9.5 mm., the other dimensions of the shaft are in proportion. On account of this striking characteristic noted in all the bones at present collected, and of its excess in size over the largest known forms, it is proposed to regard this animal from the Mammoth Cave as the type of a new species to which the name of Zaglossiis hachetti has been given, in honour of Sir J. Winthrop Hackett, K.C.M.G,, etc., the President of the Board of Trustees, as a slight acknowledge- ment of his generous support which alone rendered the exploration of these caves possible. Order MARSUPIALIA. Sub-Order POLYPROTODONTIA. Family Dasyuridae Sub-Family Dasvdrinae. Thylacymts, Temxn. (1827) 1 THYf.ACYNUS CYNOCEPHALUS, Harris (1808) ^ THE TASMANIAN WOLF OR TIGER, An interesting specimen from the upper layer “ in the Mammoth Cave consists of an almost complete left mandible with four cheek teeth in position. Of these teeth two premolars, p3 and P4 have, been subjected to a certain amount of wear ; the molars mi and m2 are missing, but m3 is present, rising from its alveolus, and the last cheek tooth m4 is stilt in its formative cavity. The general outline of the ramus is similar to that of TAy/firj'nws; it is slender and is slightly curved horizontally with its convexity downwards. The coronoid is short and delicate with a thickened anterior margin, the angle of the jaw is large, broad and prominent. The condyle is strong, broad transversely and has its upper surface below the line of the crests of the teeth ; it is separated from the coronoid by a deep notch. 1 Mon : Mammalogie, Vol. I, 30 Mon : pp XXIII, 23-24 footnote, 267, pi. 7, figs. 1-4, on pages 60-65 of the same volume the spelling is Thylacimis. Trouessart Cat. Mamm. Vol. II, p, 1216(1898/9) does not include the form Thylacimis in the synonymy of the genus Thylacynns, regarding it as a misprint, Trans. Linn. Soc. IX, p. 174, pi. 19 (1808), 8 vide ante p. 12. RECORDS OF W.A. MUSEUM. \M9 The dentition is imperfect, it is possible to recognise the sockets of three incisors, one canine, one small premolar in advance of the two still present and two molars between these two teeth and the rising m3. The preniolars resemble the corresponding teeth of Thylacynus cynocephalns ; they are much longer than broad, sharp and pointed, the blade is curved slightly backwards and inwards, the posterior ledge cuspless. P4 is considerably larger than P3. The molar 1113 is slender and narrow ; it consists of a sharp, slightly incurved central cusp, a small anterior cusp and a less developed posterior one associated with which is a rudimentary postero-internal swelling. The posterior cusp is broader than the anterior one and has a subquadrate, not triangular outline. The two main secondary cusps are in contact with the central one on the outer surface. This tooth also more nearly approaches the type of the genus Thylacynus than that of the closely related Sarcophilus. On account of its general outline the jaw cannot be regarded as that of Sarcophilus, whilst the crowded nature of the teeth, which however may be regarded as an individual character in a young animal, distinguishes it from TAy/rtcy/zj/s, to which it has neverthe- less been ascribed on account of the numerous resemblances the fossil bears to a typical 'Fasmanian Tiger or Wolf, Thylacynus cynocephalns. A fragment of an atlas vertebra and an imperfect humerus of Thylacynus probably of the same animal were found associated with the mandible. A fragment of the skull with several cheek teeth in situvfa.s obtained in the Museum Cave in November, 1912. SARCOPHILUS, Cuv, (1838). 1 SARCOPHILUS HARRISI, Boitard (1842).^ THE TASMANIAN DEVIL. Some time ago Mr. T. Connelly, the Caretaker of the Margaret River Caves, whilst undertaking explorations in the Bride’s Cave, -» 1 Vide ante, p. 40 footnote. 1 ajardin des Plantes, p. 290, 1842, A* Oldfield Thomas, Proc. Biol. Soc. Washington, vol. XXV, pp, 115-118. June 2gth, 1912. 250 ] RECORDS OF W.A. MUSEUM. which is some three or four miles south of the Mammoth Cave, discovered an unfamiliar skull lying uncovered upon a large talus of coarse sand and fragments of limestone and stalactite. When Mr. W. B. Alexander, M.A., of this Museum, visited the locality in October, 1912, he noticed the striking difference between this skull and those of known indigenous mammals of the district, and therefore made a rough sketch of the specimen, as he was unable to remove it from the Cave. This sketch was subse- quently shown to me and I was able to report that the remains were those of some species oi Sarcophilus. In November, 1912, I was instructed to proceed to the Caves in order to obtain this specimen and to collect, among other things, as many remains as possible from this and other Caves in the neighbourhood within the time at my disposal. Soon after my arrival in the area, the Bride’s Cave wa.s visited and the skull removed from its precarious position. A large amount of sand and debris in the vicinity of the discovery was carefully sifted and examined, with the result that a small fragment of the lower jaw, with a molar in situ and several limb bones, presumably of the same animal, are now' in the Museum Collection. The bones are all in a fragile state and are not at all mineral- ised, suggesting that the animal was entombed within compar- atively recent times. The bones adhere to the tongue as do those of the extinct Phascolomys hacketti from the Mammoth Cave, with which they may perhaps be considered contemporaneous. The skull is not quite perfect, the left zygomatic process and the left auditory bulla are damaged, and the incisors and canines wanting. An interesting feature in connection with this specimen is the abnormal outline of some of the bones of the nasal region ; the left and right nasals and the right maxilla and premaxilla being the bones affected. Apparently at seme period of the animal’s life the bones on the right side of the muzzle were crushed and broken, parts of the right nasal being lost. In consequence of nature's attempt to repair the damage, a secondary growth of the maxilla extends to the median line of the skull, thus separating the anterior and posterior portions of the broken nasal and confining the area of the upper extension of the premaxilla, a detached piece of which seems to occupy the position of a lost fragment of the nasal. RECORDS OF V/.A. MUSEUM. [251 The skull is larger than any preserved in this Museum but is slightly smaller than one from Tasmania, of which the measure- ments are given on p. 261 of the B. M. Catalogue of Marsupialia and Monotremata ; on the other hand, it has a longer nasal bone, a larger palatal foramen, a wider construction and a much smaller 1114 than the specimen in the British Museum, points of difference which may be safely regarded as due to individual variation and the greater age of the Tasmanian animal described by Mr. Thomas. In addition to this cranium the Bride’s Cave has yielded a fragment of the right mandible with an imperfect m3 in situ and the incomplete socket of m2 and the following limb bones, right humerus, right ulna, right femur, the distal end of the left femur and the left fibula. All these bones represent an animal that is not yet full-grown. Remains of the Tasmanian Devil have been found in parts of Eastern Australia and the animal itself has been recorded recently from Tooborac beyond Kilmore about 63 miles from Melbourne Mvhere a large female was killed in September, 1912, and sent to the National Museum, Melbourne. It would therefore appear that this creature, long regarded as extinct on the Australian mainland, may still inhabit certain wild and secluded districts, rarely, if ever, visited by man. Measurements of the large male skull in the Collection of the British Museum (Natural History) from Tasmania, and of the skull from the Bride’s Cave, Margaret River, South- Western Australia, with the percentage proportions of the latter to the former ; — B.M. Cat. p. 261. ’cwe.® Proportions. Basal length 123 I14 92.7 Greatest breadth 104 99 approx 95 I Nasals, length . . 51 52 101.95 ,, greatest breadth , . 20 19 5 97-5 Constriction, breadth 17.5 17 5 100.0 Palate, length 75 70 933 ,, width between outer corners of M3 62 53 7 86.6 Palatal foramen 67 j -3 c immaterial depends upon age Basi-Cranial axis 38 35 5 93 4 Basi-Facial . . ... 85 79 92.94 Facial index 224 222.5 99 3 Teeth length, p3 6 8 f >5 95 6 ,, P 4 ,, ,, mi-3 352 32 5 923 ,, breadth, m4 . 95 6. 52.0 1 J. A. Kershow, Vic. Nat. XXIX, p. 76, 1912. 252 ] RECORDS OF W.A. MUSEUM. MAMMOTH CAVE. The discoveries made by Mr. Glauert in 1909 and again in 1912 were of such great scientific interest that we were most anxious that the investigations should be continued, and therefore applied for permission from the Tourist Department in whose charge the Caves are placed. This was granted at once and every facility given Mr. Glauert to continue the work. We then appealed to the public for funds who responded generously. The President, The Hon. Sir J. Winthrop Hackett, K.C.M.G., who had defrayed the entire cost of the 1912 expedition, headed the list, and we were enabled to instruct Mr. Glauert to proceed to the Mammoth Cave and continue his researches for the space of three months. During this time he unearthed and forwarded to the Museum about 10,000 bones and fragments of bones, many consisting of jaws and teeth in an excellent condition of preservation. The genera represented com- prised : — Nototherinm, Phascolomys, Phascolarctus, Sthenurns, Macropus, Bettongia, Dasyurus, Thalacomys, Perameles, Isoodon, and Tachyglossus. There were also specimens of Muridae and of reptiles and birds. Of special interest are the skeletal bones of Nototherinm which show that the animal bears a much closer resemblance to Diprotodon than is generally supposed to be the case. There are two perfect skulls of Sthenurus. A description of these will shortly be published in the fourth part of the Records which will complete Volume I. Editor. Pentagonastev stibarus. Plate XVH. Stellastev tnegalopyepes. Plate XVIII, ' i ii ii 1 i 'V 1i il .1! 51 Calcitasiey anainesus Plate XIX. Echinastey ay cyst at us. Plate XXI. i M Astvogyuinoies catastlcia. Plate XXII. \TiO^ ■ A •i Clypeastev teluni$. Plate XXIII, Peronella aphnostina. Plate XXIV. /J ii t ii 1 '-' M 1 i ' * . I i i Conodadiis microconus. Colochirus axiologus. Plate XXV^ Phyllacanihus iiiagni ficus. Centrostcphamts tcnuispinns. Plate XXVI D. B. Fry, del] [H. Barnes, photo Plate XXVII. Fig. i—Diplodaciylus woodwardt, sp. nov. Twice natural size (from type). Fig. -i—Egerma formosa, sp. nov. About natural size. Fig. 3 — Lygosomn (Ehodotia) picturatum, sp. nov. About natural size. Fig. 4 — Demansta affif its, Gunihex. Natural size. ’■iAl 6 NH 6 Plate XXVIII, -CriHia georgiana, Gray, var. stolaia, Cope. Twice natural size, and 2 a. — Crima leai, Fletcher. Twice natural size. j i a o d § 00 £3 Plate XXIX, HA77o I 1 I I li V Giiathypops inornaUts, Kanisay and Ogilby. Plate XXX. •'-Ov • > I Holacanthns personifer, McCulloch. Plate XXXI. i A. R. McCulloch, del.] Cheraps preissii, var. angustus. Plate XXXII. } ■% % < V i / ■ mm- Cheyaps quinqnecarinahts. A. R. McCulloch, del.] Plate XXXIII. 3 ] A. R. McCulloch, del.] Cheraps tenuimanus, Smitli. Plate XXXIV. ' i' A. R. McCulloch, del.] Cheraps tenuimaniis. Smith. Plate XXXV. — V / ,1 4 . 1 G. P. Morison, del.] Fig. I. Zaglossus Jiacketti, sp. nov., right femur, anterior aspect. Fig. 2. Tachyglossiis aculeatus ineptus, Thos., right femur, anterior aspect. Letter Referenxes h. Head of the femur. gt. Great trochanter. It. Lesser trochanter. Scale — Three-quarter natural size. Plate XXXVI. G P. Morison, del.] Zaglossus Iiacketli, sp. nov. Fig. I. Right tibia, anterior aspect. Fig. 2. Right tibia, internal aspect. Fig. 3. Right tibia, posterior aspect. Letter Reference — i.m. Internal malledus. Scale — Natural size. Plate XXXVII. i RECORDS .V, OF THE WESTERN AUSTRALIAN Muse DM and Art Gallery VOLUME PART I A Review of the Land Mollusca of Western Australia by Tom Iredale. Printed by Order op the Trustees OP THE Public Library, Museum and Art Gallery OF Western Australia, Perth, August 1st, 1939. PERTH: Bt Abihobiti : Peed. Wm. Simpsoh, Goteenmeht Pbistee. g 2314/39 1939. RECORDS OF THE WESTERN AUSTRALIAN MUSEUM AND ART GALLERY. Volume II. Part I. A REVIEW OF THE LAND MOLLUSCA OF WESTERN AUSTRALIA.* By Tom Iredale, Coneholog'ist, The Australian Museum, Sydney, New South Wales. Bt! Permission of the Trustees of the Austnilian Museum. Forty odd j'oars ago E. A. Smith of the British Museum published an account of the land shells of Western Australia, and since then numerous additions are available for record, and the present essay bi’ing's our statistical knowledge down to date. Twenty year’s ago Iledley listed the species, then known, in his Preliminary Index but no detailed I’evision was attempted. It must be emphasised that this review is intended to serve as a basis for AVestern Australian workers, and conseqirently most attention has been j)aid to the taxonomy and discrimination of the named forms. It is important in this grouir to know the topogra|)hy, geogra))hy, geology and ])otany, a combina- tion I have termed Geozoology, of the. State to d(>fine exactly the status of the molluscs collected, all these factors being of delinit(‘ influence in the evolution of the mollusc. It may be pointed out that in over one hundred years of history of the State no local naturalist has made any major contri- bution to our knowledge of this group though probably no more profitable group could be selected, the problems being innumerable and all of them interesting. Smith noted “The report is limited to those species occurring west of long. 129°E., the boundary line between AVest Australia and South Australian teiritory. Its extreme length from north to south is 1,2.50 miles and 800 trom east to wo.st, and it embraces an area of 1,000,000 square miles.” The limits given are those of Western Australia, as politically known, but the area does not agree with that provided by tin* AV.A. Government, which in the Year Book before me reads 975.920 square miles. I have indicated the natural divisions of Australia as shown by the Land Mollusca, and it will be seen from the map here reproduced that within the unnatural political boundaries of AA^estern Australia, three faunulae are represented, t-wo restricted and one intrusive element. The Autoehthonian Eaunula restricted to the south-west corner, termed the Leeuwinian Area, is characterised by the very peculiar Bothriembryontid development. AAhth it are a few Endo- dontids, the only Heliearionid we.st of Victoi'ia, and a (perhaps) Bhytidid, no Helicids save through intrusion. On the other hand the north-west iiart of Australia, called the Dampierian Sub-Area, is inhabited by a Caurine Eaunula which is of very peculiar facies related to the molluscan fauna of ,* 'A®® published by the Royal Society of Western Australia in Volume Tom Irbdale. the East Indies rather than with that of the adjacent territories. These are peculiar Ilelieids, and no Bothriembryoutids occur save as strag'g'lers into the southern point. There is no He.limrion and no Rhytidid while the so-called Chloritids are peculiar, and two Microcysluls occur. These two faunulao are vei'y peculiar in their composition, and the third is that of the Contralian Area, known as the Eremian or Eyrean Eaunula, an intrusive series. This faimula ranges across the whole of the centre of zVustralia and runs into the south-west, meeting' members of the Autoehthonian Eaunula. It fui'thei' readies the coast to the north of the Leeuwiuian Area, but so far it has not been I'eeogiiised in eouneetiou with the Caurine Eaunula. This inland area of Western Australia has not been explored for molhises so that probably many species exi.st. While those may be related to the known mem- bers of the Eremian Eaunula it is possible that the novelties may oiler shells of ditfereut form. Possibly ihe vegetation will offer most assistance and iu this respect “The Soils of Australia in relation to Yegetatioii and Climate” (Commonwealth C.S.I.H. Bulletin Xo. .12, 1931) by Prescott is woi'thy of intensive study. Examination of the vegetation ma];) therein included suggests that the groups of Bothriembryoiitid molluscs are distributed in aecordaiiee with the i botanical data. i The following account as to the rainfall, the mo.st important item in the ■ life history of a snail, is taken from an ollieial source. The rainfall varies throughout the South-West division, ranging from an annual average of 10 inches in the heavily timbered portions of the coastal districts to 18 inches at Geraldton and nearly 10 indies at Yalgoo, from .36 inches at Albany to 14 indies at lsraeli((> Bay. and from 44 inches in Ihe Darling Ranges to 21 at Toodyay, 16 at Xortham, and 10 at Southern Cross. A similar diminu- tion oeenrs eastward of Bunhury, starting with a register of .37 inches. The climatic conditions are. vastly different in the tropical parts of Western Aus- tralia. Tleat'y tropical rains fall from the end of Xovember lo the end of Dlardi, with rarely a wet day during the rest of the year. For instance the averag(‘ at Wyndhani is nearly 27 indies of which 24H fall in Xovember to March, another D ■> in October and Ajiril, tlie total from Dlay to Septcinher inclusive being less than 1 inch. On the other hand that of Perth is only .34 indies, Decemher, .lainiarv', February averaging nearly Vo-ineh each month, April and May nearly 1 inch each while IMay, ■Tune, July and Aug’ust liave from & to 7 indies each month. The rainfall at Albany is a little lighter in the winter months, but a little heavier in the slimmer, totalling a little more Hiaii 39 indies in the year. In this coiinedioii it may he recorded that ai)])arently most lYestern Anstralian snails develop a strong epiphragm for use during the long dry spiells. 1 have long hoped to jirovide a ground work for .stmleiils of zVustraliaii land-shells as these are very abniidaiit, varied in form and scientifically very valuable. To this end I have iinblished A Basie List of the Band Mollnsea of Australia in the Australian Zoologist ( Yol. YIIT., pj). 287-334, Dlch 12, 19.37; Yol. IX., pp. 1-39, pis. I.-IIB, Xov. 12, 1937; and Yol. TX., pp. 83-124, pis. Xll.-XIIT., Xov 30, 1938). I later preiiared “An zViinolated Cheek List of the Land Shells of South and Central Australia” which has apiicared in the South Anstralian Xaturalist (Yol. XVTIT., iiji. 6-.17, |ils. l.-II., Sep- tember 30, 1937). The in'esent essay leaves Queensland and New South lYales to be as completely treated, and these T have in hand now. X"n section, how- ever, iirovides the delightful complications present in the IVestern Anstralian A Hkview oe the Land JIollusca of AYesterx Australia. Fauna and I have to thank sincerely Mr. L. Glauei't for the oiiportunity of examining’ so much material. My thanks arc also here tendered to my eolleag'ues, Miss Joyce Allan, who has ])repared the excellent draivings which accompany this report, and Mr. G. C. Glutton for the photographs of the Bothriembryontid forms. L— 11 1 R T OR I C A L A C G 0 t’NT , While it be possible that some of the earlier visitors such as A'laming, Pelsart, Damjiier or Vancouver may hai'e carried back landshells to Europe, the first real note was made by Peron in 1807 who recorded that on June 29, 1801, at Bernier Island, also Dorre and Uirk Hartog’s Islands, in Shark Bay, “Two siiecies of landshells extremely numerous, but all dead, occuiiicd great stretches of the interior of the island, one was a small species of Helix, the other belonged to the genus Bulirnns of J\l. de Lamarck.’’ Thus from this source P<'russac included in his Prodrome Helix costulaia, H. me'’.ones and H. tonilus. The former was not described and figured but localised as “Le jiort du Roi George, Peron: la bale des Chiens-lMarins, Gaudicho,’’ the second as “La Nouvelle-Hollande,” the last as “La Nouv.-Hol- laude, voyage de I’eron.” It will b(‘ nop'd that another colled or Gaudicho (Gaudichaud) is here menfioiied. 0. Gaudichaud was apothecary on the Uraiiie, on which vessel (,)uoy and Gaimard were the naturalists: thi.s vessel visited Shark Bay September 12-27, 1818, and was shipwrecked at the Falk- land Islands on its way home and the collections made lost: but aiiparently Gaudichaud managed to save some of his treasures. Tjamarck published with a description in 1822 BnlmiiH inflatits citing as synonymous Ferussae’s H. costulaia. Aiiolher species named and figured by Ferussac, He'ix prunum, and whose locality was given as “Les Terres Australes, voyage dc Peron” has been credited lo Western Australia, but examination of the figure suggests an Eastern Australian form. Then the English explorer Captain P. P. King, who Avas very interested in natural history, collected land-shells at King George’s Sound on January 20, 1818. Gray, of the British Museum, described a species naming it after King in 1820, King’s own account not being pub- lished until 1826. In the latter year the “Astrolabe” another French explor- ing A-essel, was in Western Australian A\-aters, and the brilliant and enthusiastic naturalists Quoy and Gaimard were on board, inteni to make up for their losses in the Franie shipwreck. The land-shells secured umre this time safely taken to Paris and Avere carefully described and beautifully illustrated in one of the most splendid works on natural history yet issued. An unknown collector, pi’obably Hobert Brown, provided Gray of the British Museum Avith some material, and it has only retamtly been recognised as coming from the Recherche Group. A little later a German collector. Dr. L. Preiss, visited South-W(‘st Australia in 18.‘19-4(), and M<’nke issued an account of the mollusca in 1848. At the same 1im(> Gilln'rt, the famous ornitho- logical assistant of Gould, secui'ed some shells and these were tran.sferred to the British Museum by Gould, and have been regarded as having been col- lected by that gentleman, who, however, never' I’eached Western Australia during his famous travels. Simultaneously Captain Stokes Avas exploring the West Coast and he and his men AA’cre notable collectors, Dring, Wickham and Ineo all coming under notice in connection with land-shells. Pnfortunatelv the exact localities Avere not preserved by those to AAdiom the shells passed and thus in some cases it has been a matter of great difficulty in determining these at the present time. It must be remembered that this A\ms the day of sailing vessels winch called in and sheltered at places not now commonly resorted 4 Tom Irbdalb. to, but when intensive eolleetinn; is carried out it may become easy to fix the exact place as the species are comparatively well defined. When Eyre crossed from South to West Australia he noted shells scat- tered about on the ground but that does not concern us as none apjiears to have been collected. It must be recalled that he was succoured by Captain Eossiter of the French wlialer Alississi])))! on June 2, 1841. That particular captain happened to be a very zealous amateur conchologist, and later settled in Sydney, when the well-known professional conchologist John Brazier mar- ried hi.s sister. Later Brazier circulated shells collected by his brother-in-law, who had secured them on the islands of the Recherche Archipelago when whaling about them. Until this date all the sjiecies had been described in Europe, and a few collected by Dr. Bacon in the Swan River Settleitient were named by Benson, but now Australian eouchologists became qualified, and Cox, Hedley, Angas, ' Brazier, Tate arid Hedley have all made additions to our knowledge. Still the spell of the e.xtraliinital worker held, although he was ignorant of local con- ditions and heedless of geography and history. Thus E. A. Smith in the Zoology of the Voyage of the Erebus and Terror dealt with two plates (that had been prepared for Stokes) under the impression that they had something to do with that voyage. Fortunately the species figured were so distinct that he could not make many mistakes, hnl he managed to confuse matter.s a little. Smith, however, assisted onr study, when he prepared the first complete account of Western Australian land-shells based npon a collection made by the well-known (‘ntoniologi.st, J. J. Walker,* wlien he was chief engineer on H.M.S. Penguin, surveying off the Xorth-West coast of Australia. It is unfor- tunate for Australian students that Walker’s collection was made on almost inaccessible islands, l)iit bis discoveries opened a new vista for the forms were of novel facies ami now provide the most intriguing problem in our science. W. W. Froggatt, a professional entomological collector, went to Derby and explored the Xa|)ier and Oscar Ranges, and brought back some twenty species of land and fresb-water inolliises. Some of these were described by Cox and others by the great French conchologist Anecy, but no complete study of the collection was atleinpted. The bulk went into the .Macleay Mnsenm and lias been here utilised. Years later Dr. Herbert Basedow wimt into the Xapier Ranges, and an excellent collection was briefly recorded by Hedley. The dominant Western Austi'alian group, Bothripmbi'iion, was carefully mono- graplied by Pilsbry, who had nuicb material sent him for the purpose by Cox, but the localities iirovided by the latter were untrustworthy and misleading. Kobelt, a year later, also catalogued the grouj), practically following Pilsbry. Some odd siiecies have been named by P'nlton, Preston, Gude, Odlnier and Thiele. The two latter dealt with collections mad(' by Swedish expeditions to the Xortli-West, and German expeditions to the South-West, respect iv'oly. Mr. Sidney W. Jackson visited Western Australia in 1!)12, and working in the Bow River district, south coast, made a line collection which was acconi- jianied by excellent field notes. This series iticlnded many novelties and was examined by Hedley hut the results were never published. About ton simeics were secured, all of them new, and suggest great i>ossibilitics to some local enthusiast. The only rocenl collectors have been Mr. L. Glauert, the Director of the Museum, to whom 1 am greatly indebted for the oiiportunity of studying the material in that Alnsenm, npon which this review is primarily based: Mr. P. R. Bradshaw, of Tambelhip, and Ylr. E. Sedgwick, of Xangeenan, each of whom have sent me very interesting shells. * Died February. 1939. A Review of the Land Mollusca of Western Australia. 5 It is my great hope that this account will arouse someone in Western Australia to study their own mollusean fauna, which suggests more problems than almost any other in the world. Prior to Smith in 1894 some thirty-five species had been recorded: in his essay Smith added eighteen, that is half as many again. Hedley listed the fauna in 1916 as eighty, that is almost exactly half as many again. This list has increased the number in a similar propor- tion, and yet I conclude that the number on record is only a small proportion of the molluscs living. A rejiresentative collection will be made available at the Perth Museum for students, and I will be only too pleased to assist any such in any way that I possibly can. PHYLUM MOLLUSCA. Class GASTROPODA. Subclass PROSOBRANGHIA. Order PECTINIBRANCHIA. FAMILY HELICINIDAE. Small conical operculates, solid, spire short, whorls flattened, periphery subkeeled, umbilical area covered by a callus ]iad. Genus PLEUEOPOMA Molleiulorff 1893. 1893 — Pleuropoma Mollendorff, Ber. Senekenb. Nat. Ges. Frankfiirt, 1893, p. 140, July. Orthotype TTeUrina dicliroa Mollendorff. 1905 — Aphanoconia Wagner, Denksch. K. Akad. Wis.sen. Wien, Jlath. Nat. Class., Vol. LXXVIL, )). 388. Logotype Iredale, Austr. ZooL, Vol. VIIL, ]i. 291, 1937, Heli'^ina rerecunda Gould. 1905 — ReticMlata, Sculpta, Dichroa id ib., sectional names only with Tauto- types. 1909 — Albooinctft Wagner, Ryst. Conch. Cab. (Mart. & Chemn. ), ed. Kuster, Bd. 1., Abth. 18, heft ccxxv., lief 53.5, p. 1G9. Tautotype Helicina alhocincta Hombron & Jacquinot. 1909 — Sphaeroconia id. ib., heft ccxxviii., lief 538, p. 189. Logotyire Iredale, Austr. Zook, Vol. VIIT., p. 292, 19.37, Helicina sphaeroconus M61- lendorff. Pleuropoma walkeri Smith 1894. Plate I., fig. I. 1894 — Helicina walJceri Smith, Proc. Malac. Soc. (Loud.), Yol. I., p. 99, pL YIL, fig. 26, June. Queen’s, Baudin and Parry Islands, North- West Australia. 1909 — Aphanoconia {Spli.acroconia) bandinensis Wagner, Syst. Conch. Cab, (Hart. & Chemn.), ed. Ku.ster, Bd. I., jVblli. XVIII., p. 217, pi. 43, figs. 11-13, as of Smith errore, new name for wallceri Smith. Baudin Island, N.W.A. Shell small, depressedly conical, spire short, whorls llattened, last whorl subkeeled, mouth lunate, outer li]i slightly thickened, umbilical area covered by a callus which extends across to outer lip. Coloration fawnish with a whitish peripheral band, and sometimes a paler base, some shells uni- color fawn. Apical whorl smooth, adult ivhorls three and one half, dull, with faint growth .striae but no spiral striation. The umbilical pad is 6 Toil I redale. roughened, the columella short, straight, the operculum thick, horny. The specimens before me ft’orn Baudin Island, presented by the collector J. J. Walker, are apiiarontly slightly smaller than the type measured by Smith, ivho gives “Diam. ma.j. o, alt. 3.5 mm.”: these are about 4 mm. in breadth by 3 mm. in height. FAMIl.Y CYCLOPHOEIDAE. This family does not occur in Western Australia, though two species were described by Benson as having been collected at the Swan Biver by Dr. Bacon. As this collector also worked in India it is a]jparent that the localities became confused, and the Cyc.lophorids described originated in India. The names were Ciiclostoma liricinctum and C. orbiculatmn Benson, Ann. Mag. Nat. Hist., Ser. IT. , Vol. XT., ji. lOfi, February, and they were figured by Reeve, Conch. Icon. Vol. XI IT., C udoplwrus, Vol. XX., s]i. 100 and 101, Aug. 1801. They can now be comiiletely dismis.sed fi'om this fauna. Subclass PULMONATA. This subclass includes all the land snails save those with an operculum, and as only one of the latter comes into our review, the remaining members of our fauna here follow’. Numerous smaller divisions are utilised in classify- ing this large number. Order STYLOMMATOPHORA. The subclass is separated into twt) Orders, but only one is represented in this fauna, but this has been subdivided into many large groups, but as thej’ seem somewhat artificial, only the families are here mentioned. FAMILY VEETIGINIDAE. This familv name wuis previously used for all Australian pupoid shells, but as now restricted, no representative is yet known from Western Aus- tralia, but as the s])ecies are very small and difficult to find and so little searching has yet been done in this land it cannot be definitely stated that the family is absolutely absent from the State. FAMILY GASTEOCOPTIDAE. The minute Pupoid shells of Australia were placed in one family in the Basic List, following Pilsbry's IMonograph, but ujion reconsidering the siu'cies in connection with the South Australian forms I regretted that action. T merely noted my objection in my account, but here allocate the Western Au.s- tralian shells in a. more scientific manner. It became (|uile obvious that we were dealing with diverse stocks, and then it was seen that Pilsbry himself had been dubious of his own action. Steesiberg, a European authority, has given an excellent account of the anatomy of these difficult minutiae, and separated the l^alaearctic forms into many families. Following this worker, the Australian species w’ould be arranged in four families, and this Is much nearer the truth. The few Western Australian species then fall into three families : Shell very small, pupoid, with small mouth with comj)lex armature; species dextral or sinistral. Ga.stroco))tidae. Shell small, but larger than ])reeeding, wdth no armature in mouth save a tubercle or base of shell adjacent to outer lip. Pu]ioididae. A Review of the Land Mollusca of Western Australia. 7 Shell very small, stouter than first named, mouth a; mature not so com- plex and fokls dilf'ereiilly placed. Pupillidao. Genus- AUSTEALEINULA Pilsbry 1916. 1916 — Australbinida Pilsbry, Man. Conch. (Tryon), 2nd Ser., Vol. XXIV. (pt. !K5), p. II., Dec. 18: id. (pt. 94) pp. ].o5/166, July 18, 1917. Drtliotypc (iaHtrocopln Pilsbry. Small Pupoid shells, de.xtral, rarely sinistral, faintly striate, with the small mouth almost closed by teeth blocking the aperture: a large eolumellar tooth; on the base of the slidl one to three, called the parietal teeth, one of which is generally larger than the others ; and on the inside of the outer Tip three or four, called ]ialatal, teeth. .Vll the shells are small, up to 3 mm, in length, and must be examined microseoiheally for differentiation. Australbinula wallabyensis Smith 1894. 1894 — Pupa wallabi/ensis Smith, Proc. Malac. Soc. (Lond.), Vol. I., p. 97, June. Mast Wallaby Island, Houtman’a Abrolhos (Walker). 1916 — Bifldaria u'dlliibijensif; Hedley, .lourn. Roy. Soc. IVest. Aust., Vol. I., p. 217 (68 in se]).). 1917 — Gastrocopta iraUabiiensin Pilsbrv, Man. Conch. (Trvon), Ser. 2, Vol. XXIV., pt. 94, p. 171, July is. Unfortunately this ,S|)ecies has not yet been figured, and no specimens from the Abrolhos are available. Smith described it in great detail as fol- lows: “Shell dextral, cylindrical, pellucid, white, narrowl,y riinate; whorls five, very convex, sutures profoundly impressed, very obli(iuely striate, the last whorl as(ending at the aperture; sjiiix' cylindrical, obtuse to the apex; aiierture very small, ringent, oiu' fourth the total lengfb of the shell, fuii'n- ished with seven une(|ual teeth (three very small jiariotals, the middle one largest, one largi* eolumellar lamelliforin tooth, three large lamelliform palatals) ; jieristome lightly expanded, margins apiiroaching, joined by a thin callus. Length 2.5, diameter 1 mm. Ilab., East Wallaby Island, Houtman’s Abrolhos (IValker). The teetli of this species are very characteristic, and block ii|) the aperture to a considerable extent. The three parietal are'mfaeh smaller than the rest, and situated close together, the central one being the largest. That on the columella is large, lamdliform, and prominent. The central palatal tooth is narrow, long, and extends inwards; those on each side of it are shorter, but a trille more ])rominent perhaps. P. M acdonnetli. Brazier, from Kitzroy Island, N.E. Australia, is a little shortei- than this species and has more feeble armature in (he .aperture.” Pilsbry added: “Mr. B. B. Woodward, who ha.s kindly examined the ty|ie specimen for me, states that the colunu'llar lamella ascends within. The spi'cies is thtn'idore related to the ])oIymorphic G. lariipi^ita. It differs by the unusually .small aperture (only one-fourth the total length, while in lurapinta it is at least one-t;hi:rd), and the narrower contour of the shell.” Australbinula complexa sp. nov. Plate 1. Fig. 7. Pilsbry ((Man. Conch. (Tryon), Ser. 11. , Vol. XXIV. (pt. 94), p. 170, July 18, 1917) proposed Gafttrocoptu larapinta deserti for a form of pupoid shell, picked out of series sent from Central Australia by Professor Tate as 8 Toil Ihkdale. ^‘mooreuna: \ arious Ceiili'aliiui loiialitios wore confused, and Pilsbry added Porrest Biver, North-Western Australia, an indiscreet addition. However he figured the speeiinens so that separation is ea.sy. His ili'st measurement of his “degerti” is length 2.25, diameter above aperture 1.15 mm., and that agrees with his figure 3 on plat(‘ 2, which is here selected as type. His figure 1 shows a more slender .shell also without locality. Mr. E. Sedgwick has sent from Nangeenan via Merredin shells recalling this flg’ure 1 in form, lint with the whorls even less rounded and the tip more ohtiise. The angular lamella is more pronounced and there is a notable basal fold present. The type measures 3 mm. in length by 1.25 mm. in breadth. Australbinula helmsiana sp. nov. Plate 1. Fig. 2. 1917 Gastrocopta laraphitii deserti Pilsbry, Man. Concli., (Tryon), Ser. II., Yol. XXIY., p. 171, pi. 30, fig. 1, July 18. Specimens only from Forrest River, North-Western Australia (R. Helms). At the place quoted Pilsbry recorded that these .specimens, one of which he figured was “a little larger, 2.5 to 2.55 mm., long, diameter 1.1 mm., or 2.4. 1.2 mm. There is no brownish tint excejd what may be caused by the dried animal. There is a small iiifraparietal nodule in some examples, wuinting in others. The eolumellar lanielln has a rounded outline and ascends obliquely inward. 1 , 0 wer parietal plica i.s strong, the u|)per smaller and far shorter. There is no basal fold. The lip ex]»ands broadly.” Shell minute, cylindrical, dextral, whorls round, sutures dee]), spii'e obtuse, sculpture delicate oblique striae, mouth with outer li]) exi)anded, aperture with intrusive teeth as noted above. Australbinula mooreana Smith 1894. Plate 1. Fig. 4. 1894 — Pnpn mooreana Smith, Proe, Malac. Soc. (Loud.), Yol. I., p. 97, pi. YII., fig. 25, June. Roebuck Bay, North-West Australia. 1916 — Bifidarin mooreana Hedley, Journ. Roy. Soc. West Austr., Yol. I, p. 217, (p. 68 in se))ai'ate). 1917 — Gastrocopta mooreana Pilsbi'v, Man. Conch. (Tryon). Ser. 2, Yol. XXIV. (|)t. 94), p. 16(1, ])!. 30, fig. 4 (copied from Smith): Yol. X.KA^I. (pt. 104), p. 230, pi. 24, tig. 6, 7, Nov. 1921: pai'atypieal specimens figured, “Easily distingnisluHl b.v the .sculpture.’’ Striae irregular, interrupted by raalleation or shallow uiu'ven ijitting the parietal lamella, is not coiinecled with the aji.gular lamella which is straight, the i)arietid not very long an:l only slightly curved. There are only two palatal folds, the lower being larger and almost taking the )ilaoe of the basal which is missing. The eolumellar lamella is deei)ly set and straight. Pei'hai)s nearer fi'CuJnea than Gastrocopta the angular laminae being present hnt not con- nected: this feature may he of value in either connection, hence to indicate this abnormality the new snhgeneric name Westralcopta is here introduced. Pilsbry's account of pnraty])es is here added for reference: “The shell re- sembles G. larapinta in contour, but has a special character in the sculpture as seen under the microscope, the striae being made irregular, and in places interrupted by malleatinn or shallow, uneven pitting, producing sometimes a sort of reticulation, not uidike that of some Nesopupae {Tndopupa). The A Review of the Land Molldsca of Western Australia. 9 straight angular lamella joins the lip weakly or scareeljL It does not connect with the parietal lamella, which is high but not very long. The columellar lamella is high on the columella, and ascends very slightly inward, being nearly horizontal. There are two short palatal folds, the lower being larger and somewhat more immei’sed. No basal fold. Peristome reflected and thickened moderately within. Length 2.2, diam. to edge of lip 1.2 mm.; bYz whorls.” PAJIILY PUPOIDIDAE. The small pupoids assigned to this family are generally larger than the remaining Australian pujioid shells, and are either sinistral or dextral, rather stoutly built, not conspicuously sculptured, and with the mouth rather square the outer lip e.xpandcd, and a parietal lamina, which lies adjacent to the junction of the outer liin ( leans THEMAPUPA Iredale 19:10. lO.'iO — Thenidpupa Iredale, Yict. Naturalist, Vol. 47, p. 120, November. IIai)]otype Pupa beltiana Tate. The definition above given aiijlies to the Australian members of the family. Themapupa contraria Smith 1894. 1894 — Pupa contraria Smith, Proe. llalac. Soc. (Lond.), Vol. I., p. 96, June. East Wallaby Island, Iloutman’s Abrolhos (Walker). 1916 — Pupokles coiilmriiis Hedley, Journ. Roy. Soc. West Au.str. Vol. I., p. 217 (68 in sep.). 1921 — Pupoidcs cnntrarhis Pilsbry, Man. Couch. (Tryoii), Ser. 2, Vol. XXVr. (pt. lO.'l), p- 144 (not. jfl. 15, figs. 9, 10 eremicola) , August 4. As this shell has not yet been figured, and we have no Abrolhos material, Smith’s description is reproduced : “Shell sinistral, rimatc, fuscous horny, obliquely very finely striate ; whorls 51 / 2 , convex, separated by a deep suture, the last whorl scarcely broader than the one preceding, ascending at the aperture; s]un' elongate, convex, ])yramidal, subglobose at the apex; aper- ture one-third the total length; peristome white, expanded, margins joined by a thin callus, a tubercle near the insertion of the lip. Length 4.5, breadth 2 mm. Aperture 1.5 mm. long. This species may possibly prove to be the sinistral form of P. adehtidae Ad. and Angas, which is only known to me by the description.” P. adelaidae is obviously a larger shell, as is eremicola with which the present species was later associated. Themapupa sinistralis Pilsbry 1921. 1894 — Pupa pacifira Smith, Proc. Malac. Soc. (Lond.), Vol. I., p. 96, June. Cassini T., N.W. Australia (J. J. Walker). 1921 — Pupoides paeificus fonu sinistralis Pilsbrv, Man. Conch. (Trvon), Ser. II., Vol. XXVI. (pt. 103), p. 144, Aug. 4. When Smith recorded the East Coast pacifica from North-West Aus- tralia he noted: “All the specimens from Cassini Island are sinistral; other- wise they resemble the normal form.” Pilsbry, in accepting this determination, observed that numerous speci- mens from all the other localities admitted furnished dextral specimens alone and recorded the Cassini Island shells as a form sinistralis^ and this name may be used sjiecifieally to keep this shell under review. 10 Toil Ieedalk, Tliemapupa lepidiila A. Adams and Angas 1804. plate I., figs. 5 and 5a. 1864 — BuUminns {ChondriiJa) lepidula A. Adams and Angas, Proe. Zool. Soe. (Loud.), 1804, in 38, June 24. Shark Bay, W.A. Pig’d. Cox, Austr. Land Slmils, )). 69, pi. XIX., f. 14, May, 1868, from a painting nf the type by Angas. 1894 — Pupa lepidula Smith, Proe, M.iiae. Sue. (LomL), 1894, p. 90, lane, cites ‘‘Cox, op. eit.. p. 69 (nnfigiired).'’ 1916 — Pupoidcf^ iepidtdiis Hcdley, Journ. Roy. Soc. "West. Austr., Vol. I., p. 217. 1921 — Pupoides pacijicnn Pilsbry, Man. Conch. (Tryon), Ser. 2, Vol. XXVI. (pt. 103), ]). 141. August 4 (part). This is a very distinct species judging from Cox’s figure of the type which is here reproduced. Pilsbry rather carelessly included it in the synonymy of 7'. pacificm, observing, “It came from well within the known range” of that 8pcci('s, whereas it was well outside. Then he figured a speci- men from Porrest River, near Wyndham, which was not much like Cox’s figure, and noting that the lip was narrower concludes, “If this proves con- stant they might be sejiarated as a race lepidula.” The original description is here added as no one has collected land-shells at Shark Bay in recent years: •‘Shell turrited, iniiifoiiu, iimbilicate, thin, glos.sy, semipellucid, corneous; whorls 5 strongly convex, longitudinally striate. Aperture rotund-ovate; peristome interi'upted, white, broadly re- flected; outer lip lu’ovided above with a small, white, tuberculiform callus. Length 2, rvidth 1 line. This little species differs from C. ndelaidae in being semipellucid, shining and of a horn colour. The whorls, moreover, are much more strongly conve.x.” In view of the fact that s])ecies have been confused ixnder the name pacificuet, it is somewhat amusing to record a recent conclusion. Rens-h (Zool. .Jahrbudi. fSyst.), Jena Bd. 63, heft I., p]). 1-130, Apl. 12, 1932), led astray by a desire to recognise affinities without regard to local condi- tions, has ])!oposed to admit one species PupodJes coenopictus Hutton, with an old-worhl I’ange as follows: — l\c. coenopiefus Hutton. India, Ceylon, Afghanistan. P.c. seiiegalenfiif! iMorelet. West Africa. P.e. conneclenn nov. Sumba. P.c. pacificus Pfeiffer. Australia, Islands of Torres Straits. The idea of introducing a. new subspecies, in an amalgam of such a composition, does not seem in accordance witli his own theory of widespread “races.” These “races” appear to he superspeeies of other malacologists. Themapupa anapaciflea s]). uov. Plate L, fig. 9. 1874 — BuUmus pacif.eus Smith, Zool. Prehus & Terror, Moll., p. 3, pi. IV., f. 6. Pigeon T. X.IV. Australia (Richardson). Xot B. pacificus Pfeiffer, Proc. Zool. Soc. (Loiuh), 1846, p. 31. 1894 — Pupa parifica Smith, Proc. Malae. Soc. (Loud.), Vol. I., p. 96, Juno. Pigeon I., near IVallaby 1., lloutman’s Abrolhos, West Australia; Koebuck Bay and Baudin L, X.W. Australia (J. J. Walker). A Review oe the Land Mollusca of Western Australia. 11 1921 — Pupoides pacijicim Pilsbry, Man. I’onc-Ii., Svr. 2, Vul. XXVI. (i)t. lO.i), p. 141, pi. 15, flg’. 14 only, Aug. 4. Forrest River, East Kimberley District (liicluinl Helms). The Western shell has been associated with pacificus by Smith, and Pilsbry, although the latter noted the discontinuous distribution, and tigured as representative of a Xortli Queensland island shell, siieeimens from Narrabri, inland New South Wales, Forrest River, West Australia, and ap- parently only one from Fating- Island, almost in the south of Queensland. The last named, as would naturally be concluded, appears to be the nearest to Ihe true paci/ica. The Forrest River shell is here named. Shell small, pupoid, dextral, whorls convex, sutures deeply impressed, umbilicate, apex obtuse, coloration pale brown, outer lip white. The apical whorls are smooth, the adult whorls obli(iuely lincdy striated with threads, in some cases well marked, the mouth rather small. Length 4.25 mm.: breadth 2 mm. Themapupa dirupta sp. nov. Plate I.. (Ig. 8. Mr. W. W. Froggatt collected a few specimens of a puiiioid shelli in the Rarrier-Napier Range, and these break u[' the “pacifica’’ pihantasy. They are detlnitely of that association, but also perfectly distinct, being smaller than any of the so-called forms, measuring 3.5 mm. length by 1.5 mm. in breadth. The whorls are Hatter than in the previous sprecies though the .sutures are deep, the shell darker coloured, the sculpitnre weaker, the mouth comparatively larger, the shell having; a more squat aiijiearance. This is apjiarently the lirst inland represeulali\e in Australia of the ‘‘pacified” series. Themapupa heltiana Tate 1894. Plate 1., h'g's. 3 and fi. 1894 — Pupa beltiinui Tate, Trans. Roy. Soc. South Austr., AMI. XVIII., p. 191, November. Central Australia. Fig’d Rep). Horn Sci. E.xpted. Centr. Austr. (pt. II.), /tool., pr. 204, p. XA^IIL, flg. 15, Feb. 189(1 (as dextral form of P. contraria Smith). 1921 — Pupoides coiitrarins heJlianus Pilsbry, Alan. Conch. (Tryon), Ser. 2, \'ol. XXA"I. (p)t. 103), p). 145, pil. 15, figs. 5, 7, 8, August 4. Aprparently there is deflnite loc.-d variation in this species as -well as individual variation, and a good series from Cardanumbi .shows all the spreci- mens to be broader than cotypes of Itelliana. The.se may be regarded as a subspiecies T. h. contexta nov. The tyj'c measuring 0 mm. in length by 2.75 mm. in breadth, .some even being broader, whereas the typje of belliana was only 4.5 mm. in length by scarcely 2 mm. in breadth, piaratypes lieing more slender still. A series from Nangeenan via Alerredin collected by Air. E, Sedgwick, shows a more conical form than any of Pilsbry’.s excellent lig'ures, recalling a de.xtral lig. 9 (contra da). The .shells are variable in size but all .show the same shape, the whorls flatly rounded, the sutures deep), the angular nodule rathei- small, and may be called T. b. asserta, snbsp. nov., the type measuring 5.5 mm. in hniglh and 2.5 mm. in breadth. FAAIILY PUPILLIDAE. The usage of northern families for southern shells will pirobably later be as amusing to onr sneccssors as son;e of our predecessors’ wmrk is to our- selves. Nevertheless it is necessary to have some refuge' for these wails, and therefore the above is here used w-ith reservation. 12 Tom Jkbdale. Genus OMEGAPILLA Jicdale 193 1 . 1937 — OmcgapiUa Iredale, Aiistr. ZooL, Vol, p. dOl, March 12. Ortho- type Pupa nelsoni Vox. Small pupoid shells, dextral or siiiistral, month small, teeth three or four, differently arranged from those of other Pupoids, tliough ohxiously correla- tive. Hedley placed this in the European genius, \ ertigo, while Ihlsbry in- cludes it in the Piilaearctic genus, PupUJa, under a section, PrimiptipUlUf based on a Caucasian shell, but later discarded this in favour of GiblmUnopaii Germain, proposed for an “Eniieid” shell from the Island of Keunion. I reiterate my conclusion that the association of southern Australian Pupoids with those from Europe and South Africa is basically unsound, and should not be recognised by Australian conchologists. Omegapilla occidentalis sp. nov. Plate I., tig. 10. 1894 — Papa linvohiensis Smith, Proc. IMalae. Soc. (Loud.), ^’ol. I., p. 961, June. Pigeon Island, near Wallaby Island {Dr. Kichardson, in British Museum), and East Wallaby Islatul, Houtman's Abrolhos (Walker). 1916 — Vertigo Jincolnensis Hedley, Journ. Roy. Soc. IVest Austr., Vol. 1., j). 217 (p. 68 in separate). 1921 — PupiUa aii.'itndis Pilsbry, Man. Conch. (Tryon), Ser. 2, A^ol. XXVI. (pt. 104), p. 218. Smith remarked “This species is described (by Cu.x (Mon. Austr. Land Shells, jn 80, pi. XV, fig. 16, May, 1868; Port Lincoln, South Australia)], as having only a single lootli or tubercle in tlie aperture. The specimens collected by Dr. Richardson and Mr. Walker have a soeond biisal tubercle as indicated in Cox's tigurc, and a third far within uj)on the columella. It is jjossible that in (he examples examined by Dr. Cas the denticles were only feebly developed, or they may even have been overlooked, being rather indistinct.” Many specimens from Rottnost I. arc much smaller than South Austra- lian lincolneims or australis, and have the teeth placed much furt'ier back and less noticeable, the parietal tooth most pronounced, the columellar one smaller and more hidden, the liasal one well inside. EAi\IILV SUCCINEIDAE. This family is at present allowed a world-wide range, but this is ques- tionable, and recent researches into Brifisli forms show distinct groups defin- able in that small compact area. 1 have therefore introduced the generic name Austrosnccinea for the common southern Australian type as the animal has been shown to differ from Ibat of the northern group whose shell is most like. For a second peculiar Australian group I have proimsed Arborcinea: both these occur in Western Australia but there may bo a third later separated. Gemis ALTSTROSUCCINEA Iredalc 1937. 1937 — Austrosaccinea Iredale, Austr. Zook, \ ol. VIII., p. 30t, Meh. 12, Orthotype Succiyiea australis Ferussac. The waxy appearance and peculiar form make these molluscs recognis- able at sight.' Subbulimoid in form with a very .short spire and open oval A Review of the Land Mollusca of Western Atjstkalia. 13 mouth, imperforate, texture very thin, the outer lip fragile, sculpture of rude growth radials, soiiietiines with line subordinate lining. Austrosuccinea scalarina Pfeiffer 1861. Plate I., fig. 11. 1861 — Succinea scalarina Pfeiffer, Proc. Zool. Soc. (Lond.), 1861, p. 28, May 1. King George’s Round, South-M'e.st Australia. Pig’d. Cox, Mon. Austr. I.and Shells, p. 80, pi. XX, tig. 19, May, 1868, from a painting of the type by Angas. “Shell ovately conical, sealarine, rather solid, irregularly rugosely plicated, slightly .shining, reddish; spire elongated, rather acute; whorls convex, last slightly exceeding the spir<‘, somewhat attenuated at the base; columella receding', nearly straight, forming with the peristome an indistinct angle; aperture oblique, oval, scarcely angular ahove, peristome simple, colnmellar margin slightly rellected above.’’ Such is the description of a Succinea, and is here reproduced as a standard. The shell is reddish when alive but commonly fades to horny, the solidity is usually missiTig, thin and fragile better describing the species; length 15 mni. ; breadth 7.5 mm. ; length of aperture 9 mm. Austrosuccinea contenta sp. nov. Plato I., tigs. 12 and 13. 1813 — Succinea nblonga “Dr.” Menke, Moll. Xov. lloll. Spec., p. 6 (Apl. 11), nom, nud., in cracks of limestone rocks not far from the sea near Perth. 1844 — Succinea ohlongu IMenke, Zeitschr. fiir. Malak. (Menke), 1884, p. 56, Apl. 10. New Holland specimens described. Not Succinea ohlonga Draparnaud, Tab. Moll. Prance, p. 56, 1801. Probably there are many si)ecies in the Perth district, but obviously Menke’s name was used for the common one. This has a short spire with a rather swollen body whorl, the spire shorter than that of scalarina and the body whorl a little broader; the .shell is not so rugose, the early whorls being much smoother, the inner lip is a little thickened and a slight body glaze connects with the outer lip. The length of the type is 11 mm., the breadth 7 mm., the length of the aperture 8 mm. Specimens from Rottnest Island app)ear to differ in their narrower form, less swollen body whorl, longer spire and may represent a distinct subspecies, A. contenta isolata .subsp. nov., but this will be more fully discussed later. The type measures 11 mm. in length, 6 mm. in breadth and the aperture 6.5 mm. in length. Austrosuccinea caurina sp. nov. Plate I., fig. 16. 1895 — Succinea scalarina Hedley, Proc. Malae. Soc. (Lond.), Vol. I., p. 260, July. Pitzroy River, Noi'th-West Australia. 1898 — Succinea scalarina Anccy, Proc. Linn. Soc. N.S.W., Vol. XXII., (189^), p. /7/, June 4. Leunard River, 100 miles inland from Derby, North-West Australia (W. W. Proggatt). This north-western species which has been confused with the south- western one has a general superficial resemblance to that, but is smaller, nar- 14 Tom Ieedalb. rower anil lacks the rugose plications. This reads ven- similarly to the de- scription ot strigiUuta but the shape is different. Shell elongate, rather narrow, spire moderate, attenuate, mouth oval not much expanded, senljiture weak, early whorls almost smooth. The whorls are rounded, but not strongly convex, the sutures fairly deep. There is very little body glaze connecting the inner and outer lips. Length of type 12 mm., breadth (>.2,5 mm., length of aijerture 7 mm. Austrosuccinea strigillata A. Adams and Angas 1864. 1864 — Succii ea sirigillata A. Adams and Angas, Proc. Zool. Soc. (Lond.), 1864, ]). 38, June 24. Shark Bay, West Australia. Fig'll. Cox, Mon. Anstr. Land Shells, [i. 89, pi. XV., Ilg. 5, May, 1868. No specimens are available from Shark Bay so the original ileseriiJtion is here given: “Shell ovately conical, thin, semi-))ellucid, pale horny yellow; spire scarcely eiinalling the aperture; apex papillary; whorls 3%, strongly convex, longitudinally finely and minutely striated; aperture oblong orate; left margin with a long thin callus deposit, right simple. Length fi lines, breadth 21/2 lines. A species somewhat allied to Siicchivu striijnta Pfr., from South Australia, having the same papillary ape.x; but ditfering in being smaller, thinner, of a lighter colour, and in the whorls being less strongly rugose.” Smith (Proc. Malac. Soc. (Loud.), Vol. 1., p. 97, June, 1894) has re- corded Succinea scalarina from K. Wallaby L, llontman’s Ahrolhos, W.A., and given a figure but rvhether this be from the type of scalarina or from an Ahrolhos shell is not stated. At |)resent as no specimens are availal)le the identity of the island form cannot he ascertained. Austrosuccinea aridicola sp. nov. Plate 1., fig. 15. Shells collected at Boulder by W. D. Campbell are unlike any of the coastal forms, but still more unlike the Centraliaii interioris. Shell small, elongate, spire long, month short of moderate expansion, whorls very rounded, sutures very deep. The shells are dead, the imriostracnin missing, the sculp- tui'e weakly strigose, the slrigatinns still notable on the early whorl. The spire is almost as long as the a]ierture, the latter being rather a rognlar oval, the inner lip conueeting with a body glaze to tlie outer lip, which is thin a.s usual. Length 1(1.5 nun., breadth 6 mm., length of aperture 5.5 mm. Austrosuccinea eexi Finlay 1927. Plate I., fig. 17. 1927 — Sucemea coxi Finlay, Trans. New Zeal. Inst., Vol. 57, p. 521, Jan. 19, new name for ISOg — Siiccinca aperta Cox, Alon. Austr. Laud Shells, p. 90, pi. XVII., fig. 6, May. King (leorge’s Sound, IVest Australia (Masters). Not Sncciiiea aperta Lea, Trns. Amer. Phil. Soc., Vol. VI., p. 101, 1838. This species is recngnisabli' at siglit by the large month which is the bulk of the shell, the sjtire being only one-third the length of the aperture. From conchological features alone it deserves separation subgenerically as Cerinasota nov., and it is fairly certain that the anatomy will necessitate a higher value. A Review of the Land ^[ollusca of Western Australia. 15 Genus AEBOKCINEA 1 redale IB.IT. 1937 — Arbormtea Iredalc, Austr. ZooL, Vol. VIII., p. 308, Mch. 12. Ortho- type Sitccinea euc(dnpti Cox. The tre(‘ living Suecineids have very different habits from those on the ground whieh 1‘reiiuent damp plaee.s, and it has been recently argued that field observations must be taken into account in the taxonomic arnteil among those alone. The shell features or Arborcinea include a short s]iire, swollen body whorl and a rather bold sculpture with a truncated thickened columella. Arborcinea menkeana Pfeiffer 1850. I'latc I., fig. 14. 1843 — SU'Ccinea nmphib'm “Dr.” Menke, Moll. Nov. Holl. Spec., p. 6, Apl. uom. nud. “Under the Bark of Euealypts.” District Hay, West Australia. 1844 — Siiccinra amphibia Menke, Zeitschr. fiir Malak (Menke), 1844, p. 55, Apl. 10. Specimens described. Not iS'. amphibia Drai'arnaud, Tab. Moll. Prance, p. 55, 1801. 1850 -Smcc/wcu mmlicatia Pfeiffer, Zeitschr. fur Malak (Menke), 1849, p. 110, Jan. 1850, fide t’.D.S. New Holland (L. Preiss) = amphibia Menke, i.e., from Hay as above. Fig’d. Pfeiffer, Syst. Conch. Cab. (Mart. & Chemn.), ed. Kuster, Bd. I., Abth XL. p. 45, pi. 4, fig's. 36, .37, 38. 1855. 1930 — Surcinaa hrevi.^isimn Thiele, Die Fauna Sudwest Anstraliens, Bd. V., lief 8, p. 587, ]il. IV., f. 66. Collie, South-West Australia. TIic description indicates the genns thus “Shell ovately-elliptical, thin, distinctly striated, pellucid, shining, horny; spire short, papilliform, suture deep; whorls 2’','h, the last but one very convex, the last attenuated at the base; columella somewhat callous, regularly curved; aperture slightly oblique, regularly oval; peristome simple, thin, margins approaching. Length 31/2 lines; breadth 1^, lines.” FAMILY BOTHRIEMBRYONTIDAE. Prohahly the most intriguing laud shells in Australia are the bulimoid forms inhabiting the South-West corner. A large number of species and races has developed, and ]irobably only a tithe has been described. It is un- fortunate these have not yet heen studied hy anyone conversant with local conditions, and it is certain that ti;ey will provide future students with much re.scarch. No more exciting subject could be chosen hy the student, hut the unravelling of the many problems will necessitate much investigation. The present essay, also hy an outsider, is the re.sult of oi'er a doFum years of eon- •sideration, the specimens having been handled from every viewjjoint, gpologi- eal data, botanical data, meteorological data and even soil conditions have been brought under review. The first S])eeies were described from the collec- tions of the French naturalists; a little later similar shells taken hV Britisl IG Tom Ieedale. explorers being' made known. A Gerniaii naturalist then collected some speci- mens, and these, of course, were dealt with hy German concholog'ist.s. Some time later Cox, the Sydney conchologist, listed the species in his Monog’i'a]di, Init was fontent to indicate the various varieties only. Hedlcy attempted to ai’rauge the forms hut did not ])ublish his conclusions. A J'eal attempt to systematise them was made hy the inimitable American malaeolog'ist, Pilshry, in the “Manual of Concholog'y.” To assist him, Cox sent him a large collection hut unfortunately the bulk of that collection was labelled ‘Tving George .s Sound," the name covering all the adjacent locality within a radiu.s of a hnndred miles or so. Tlius Pilshry was baffled with the great variation in the shells from this locality, it is indeed iortmiate that the Cox collection, including the specimens handled hy Pilshry, is now in the care of the Australian IViuseum, and has been utilised in tliis study. A year or so later the German monograidicr, Kobelt, dealt with this grou]), and in the main followed Pilshry, hut did not realise tlie value of exact local- mes. in the future the scitmee of geozoology will probably play a great part in the elucidation of tills gi'Diiji, as the species ,and even liigher group- ings appear to ho living in government hy their geological and botanical environment, with, of cour.se, the rainlall being a matter of inpiortanee. file Hollinembryontids vary in size from about 2.') mm. to 50 iiini. in length, fi'oiii very narrow to plump in fonii, and from almost smooth to heavily graiiose si-ulpture. It is iirobahle that future investigators will dis- cover anatomical features in the animal to assist them, hut at present only the shells are available for examination. The notable variations are easily seen, as dii.i', indutns, melo, onslowi, hingii and gratwlckii. According to Pilshry’s rcsc-arclies these apiiearcd to show different apical features so that as a guide for future investigators I introduced suhgeiieric uanies, Bartogembrj/on for uiiKlowi, and tiutagembrjion for gratwicki Tlu' type of Bolhii»mbrgon i.s mclo, and Unrlogcnihriion seems to ho an offshoot from thi.s source, while Batagembrgoii may be a specialised iiroduct of the bingii root. However, da.i; and indiittis, the two largest forms, are thus left nameless, and while the latter may he related to the melo branch, it is now difl’crentiated clearly, and is her(> .subgenorically named Dialembryon. The source of du.i- is at iireseiit indeterminable, and a siihgeiius Ponembryon is here [iroposed to distinguish it. As Sutagembrgon is so almoriual, a suhgeniis Telembryon, is introdiu-eil for the normal Idngii series. By the usage of these subgeneric names, and the allotment of the named siiecies will secure greater accuracy and lead to a better knowledge of this interesting series of land-shells. It must he i)ointed out, however, that though dux and indutus are in no danger of confusion, (here are some shells that appear to link together bingii and undo, hut hiTter material and more local knowledge might easily dissipate the clouded outlook of the extra limital struggle for light in this delightful molluscan group. The Bothriembryontids are i)raetically confliied to the south-west corner of Western Australia, which I have called the Leeuwinian Aren, and con- ■siitute a I'emarkalile ehunent of the Autochtlioniaii Faunula. Xortliw narrow umbilicus. The coloralion is commonly unicolor, i)ah‘ fawn, sometimes with a sub- sutural reddish band and red longitudinal streaks bul never with a red cir- eumbilical patch. Dead shells are white as the coloration lies in a very fine periostraeum. The apex is rather flattened, of two whorls, the sculpture of fine punc- tuation or thimbling as it is sometimes termed, this being coarser than that of the Swan River bulla; the adidt sculiiture consists of radial growth lines, irregular and wavy, becoming less marked on the last whorl. On the early whorls there is a faint concentric sculpture also seen but this is never notable and usually vanishes early. Length 26 mm., breadth 15.5 mm. This de- scription is drawn up from apecimens from the type locality, where a groat deal of colour variation is not seen. As related below, many shells have hee.i regarded as answering to the original nominations, hut as now restricted the above apjtlies exactly. The confusion in connection with the common King George’s Sound shell has lasted since its discovery. A similar shell was found in the Perth district, and another not a great deal unlike at Shark Bay. Api)arently Ferussac distinguished these, and Lamarck confused them, and then Q'uoy A Review oe the Land ilOLLUSCA op Westekx Australia. 1!> and Gaimard re-recognised Ferussae’s form. Menke, however, could not understand them, and added some more s])oeies. Collections did not solve the problems, as the localities whence the shells came were not accurately dis- tinguished. Hence we have the great authority on land-shells of eighty years ago, Pfeiffer (Mon. Ilelic. viv., Vol. IV., ji. 477, 18.b9). allowing one species under the name melOj but arranging six varieties, rvliich he did not system- atically name, but used phrases only. The first words were afterwards re- corded as varietal names, and these read casUir/enn, Ininas, iinijor, alhiclus, the fifth without a distinctive adjective to begin witli, and the sixth, tenim, cit- ing' [ihysode.i Menke as equivalent. No localities wei'e known for these dif- ferent variations, and it is now impossible to determine these varieties. Cox mentioned Menke and Pfeitfer, and that he eonid add more varieties, but did not give any names. Hedley arranged the shells in the Australian Museum and gave names to two colour Mirielies as was the custom thtu, but did not publish these when he listed the species under the name Lipnnis inflatus Lamarck, with varieties, melo Q. & G., physodes Menke, castaiicus Pfeiffer, hulla Tdenke and rhodoslonta Gray. Pilsbry introdueed the new generic name Bnlhriemhryon, aiid using inflatns as the specific name, admitted as varieties melo Q. & G., rastieneus Deshayes, and added var. maevliferus and var, conispira, allowing piysodcs, India and liiodostonia as different although he could not recognise the two last-mentioned. Kobelt did not know what to do, so recorded most of these following Pilsbry’s descriptions, but not admitting his valuation. His account is too confused to need much discus- sion, his descriptions and figures being liereaftei' allotted as far as ptossible. Bothriemhrjmn castaneus Pilsbry 19(11). Plate II., fig. 2. imi—Botliriernhryon mflMus var. castanens Pilsbry, IMan. Couch. (Tryon), Ser. 2., \ ol XIH., p. . 1 , p|. T., figs. 11, 18, A|iril 2;i, ex Ile.shayes MS. (citing Vol. VIIL, p. 24.0, Lamarck, Hist. 2ud cd., where it does not occur), giving George Sound, We.st Amstralia— Doubtful Island South- We.st Australia. ’ Quoy and Gaimard figured as a variation of their Helix melo, a nar- rower brown shell with a broad white peripheral band. This variation was included by De.shayr.s, but he did not assign it any name, writing, “var. (ietas) eastanea : vitta eineta alba.” Pilsbry used the name castaneus as a varietal one, eop\iug Quoy and Gaimard’s figure, but de.seril)ing a .sjieeimeu .sent by Dr. Cox, wTiich he also figured. I am selecting the latter as the tvpe of Pilsbry s castaneus, as it was one of a series from Doubtful Island, and his specimen is here reflgiired. The Doubtful Island shells are constant in coloration and size, the sculidure being a little stronger, and the size a little less. Length 2.) nim., breadth 12.5 mm. Bothriembryon rhodostomus Gray 18.34, Plate IL, figs. 3-7. 1834 — rhodostomus Gray, Zool. Roe. (Loud.), 18.34, p. 67, November 25. New Holland (probably collected by Robert Brown at Goose Island, Recherche Group, M'.A.). Not Bithmus rhodostoma Reeve, and later authors. imO—Bothriernbryon inflatiis var. maruliferus Pilsbry, Man. Conch. (Tryon), Ser. II., Vol. XllL, p. 5, pi. L, figs. 12-14, Apl. 23. “King Gpor<^e’ Sound, Soutli-A est Australia.'’ = Recherche Group based on°a shell collected by Rossiter. 20 Tom Iredale. 1901 — Bothriembri/on martensii Kobelt, Syst. Conch. Cab., (Mart. & Chemn.'i cd. Kustcr, Bd. 1., Abth. 13 (2), p, 704, pi. 112, figs. 3, 4, (sheet dated 21 VI. 1901). Now Holland. Throng'll lack of specimens this species has been overlooked, a different species being ligun'd by Reeve, but with doubt, yet lie has been followed. If the description had lieen studied, the shell would have been recognised as it is really iiuite good: “Shell ovate, perforate, solid, striate, whitish marbled with I'ose, cuticle thin, olivaceous, suture delicately circulated; whorls with two oliscure posterior bands; throat roseate; ]ieristome a little thickened; axis anteriorly deep brown. Axis Ih'i; diam. % unc.” The size and proportions do not fit any other species than the following. AVhen it is realisi'd that Robert Brown was on (loose Island, one of the Recherche Croup, where this kind of slu'll is common, its recognition is easy. Reeve figured a diftVrenl shell under Cray’s name, but he doubted the a.ssocia- tion, and he was right in that doubt, as his specimen almost certainly came from the I’ei'th di, strict. Rossiter also collected specimens in the Recherche Grouji, and one of these was named by Pilsbry as the Coxian locality “King George's Sound” misled him. The remainder of Rossiter’s specimens in the Brazier collection are labelh'd “Recherche Group.” One of the most interesting revelations of Australian conchologieal science was made by Tlr. A. P. Basset Mull, who visited the Recherche Group, with the assistance of Messrs. Henry Grant and J. H. AVright of the Aus- tralian Aluseum. Many s])ecimens were collected on the mainland at Esper- ance, and on the eastern islets of the Archipelago. They found that each locality showed a definite variety, and these are illustrated and diagnosed here. At first sight these recall the King Gc'orge Sound melo, but are larger, and restricting rhodoMomns to the western islands of the R(‘cherche Archi- pelago, those from the eastern ones are here described. The Guntim Island shells may be called B. rhodostomus hullianus subsp. nov. The groiuul colour is pale brownish cream, a reddish brown band below the sulure follow(>d by a ])ale band, and then another reddish brown band above the periiihery with a similar red brown patch around the umbilicus. The aiierture g(‘net'ally (lale creamy white inside, sometimes purplish, the columbella pale or dark purplish brown, rarely white. A strong wrinkling below the suture gives the subsutural red band a streaked appear- ance. In form, they are r'oundly elongate, the spire being about equal in length to the aperture. Some thirty si'ecimens are still available, of which two are unicolor, the others constant. The \ariation in measurements is; 35 mm. in length by 13 mm. in breadth (type), ’H by 20, ,)4 by 1/ and 31 by 20 mm. One shell from Rob Island (Mondrain) is smaller with the ground colour pinkish white, lu'avily strc'aked with purplish brown, the aiierture and columt'lla imi'iilish, and ]>robably conu's nearc'st. true vhodoHomus. A long;*]' series from Charlev Island shows moia' variatioji, the darker shells being brownish cream, longitudinally heavily streaked with dark brown, some are less heavily mai’kod, the .streaking being fewc!’, and bands showing through so that one almost agre(>s with the Gunton Island shell, ana some are even unicolor. In the other direction, the sti'caks tend to coalesce until the last whorl is imactically uniform dark purplish brown. The aper- ture varies in coloration, according to the exterior, from white with a rose columella, and outer lip internally, through pale reddish to dark purple. In shape some are comparatively plump while others have the spire lengthened. A Review of the Land Mollusca of Western Australia. 21 but in none is the spivo shorter than the aperture. Measurements read length, 40 mm. by breadth, 20 mm. (typo), 38 by 22, 38 liy 20, 3.') by 20 to 32 by 20. These may be called B. r. grantianus subsp. nov. Another fine serie.s from Kab))it Island (not Rabbit Island in King George’s Sound) are similar in size and form to the preceding, but are paler throughout, and are here named B. r. wrightianus subsp. nov. The markings are always paler, and pallid shells are more' numerous, while these arc^ gimerally smaller, all the largest one.s being streaked. Onh’ one a])proaches the normal Gunton Island form coloration, and the measurements of the streaked shells are length 36 mm. by breadth, 21.5 mm. (type), .'Ui by 20, and 35 by 21 mm., while the pale unicolor shells measure 34 by 21, 34 by 20 and 33 by 17 mm. The shells from Woody Isle are a little .smaller, with a jiale cream zone below suture, and a pale patch on liase, the whorl being brown streaked with dark purplish brown, the streaks almost coalescing on the last whorl and forming a broad band. The measurements are, length, 31 nun. by 17 mm., and 30 mm. by 18 mm. This may be called B. r. perspectus subsp. nov., and this leads to the mainland Esperanee form. The mainland shells, collected commonly at Esperanee, recall melo, but arc generally lar,gcr and paler. Pilsbry has figured this as maciilifents, fig. 14, but as his figs. 12 and 13 of maenlif erus are alike in colouring, and are larger and are part of a lot from the Recherche group, nuicuNf nriis becomes a synonym of rhodo^tomus. Some hundreds of shells were collected at Esperanee crawling on grass and bushes after rain, and these are quite constant, although the streaks vary in number, the colouring is pah', and none is dark as the island forms, and this is here described as a distinct species. Bothriembryon esperantia sp. nov. Plate TI., fig. 8. Shell plumply elongate, not as broad as hnlla, with less sculpture, the spire conical as long as the a|)erture, whorls rounded, the columella reflected over the umbilicus but still leaving a notable chink, shell solid. Coloration greyish white s]iarsely flamed -with irregular streaks of dull pinkish brown, the ground colour dominating the coloration. A))e\' flnc'ly punctate, a little more eh'vated than that of typical melo, the adult scul])lure rough, irregular radial growth ridges with very little cross scul])ture and no decussation notable in some cases; in other rather coarsely granulose as the one described and figured by Pilsbry (p. 6, pi. 1., fig. 14) which is here refigured, the form and coloration being diagnostic. Length 23 mm., breadth 15 mm. The largest measures 26 mm. by 16 mm. Bothriembryon balteolus sp. nov. Plate IT., fig. 9. Many shells from the Esperanee Malice Belt district, 50 miles south of isTorseman, Madui'a, Salmon Gums, are all dead and agree in showing a strong banded coloration. ” In form the shells recall huUa, but the s])ire is not so exsert and they must be near to rliodostomnff geograidiically. The shells ha\ e the s]nre defin- itely less elate than the Esperanee coastal species, with the seul)iture more regular almost producing a cancellation on the penultimate ivhorl and shoulder of the last w'horl. The ground colour is ebalky-wdiite with a nai'row purple peripheral band and a large purjdc circumbilical patch. The columella is 22 'J'OM Iredale. reflected almost covering up the umbilicus, the outer lip thin, and a glaze connects the inner lip to the outer lip across the body whorl. The apex is eroded but shows the coarse pitting of the mclo series. iMeasurements — ■ length 21 mm., breadth 15 mm. (Esperance ilallee Belt). The range of this form inland would be interesting, as we might get intermediate localities between this and the Centralian sycnceri. From Newman Kocks shells are a litth' more globose, spire shorter, and the colours reversed, being pale bi'own with a peripheral white band recalling cdi^taneus, but differing in sculidure. Bednall (Trans. Roy. Hoc. South Austr., Vol. NVI., p. (Ki, Dec., 1892), re- corded Bulimiis wielo Q. & (r. from the Eraser Kang<>, from dead shells alone: this report may refer to specimens of this species. Bothriembryon serpentinus si), nov. Plate II., tig. 10. A large series of shells collected alive by Mr. L. Glauert at Serpentine Palls, Darling Range, all agree in form an. 15. 1900 — Bothriembryon baeovi id. ib., p. 16, pi. 2, fig. 42. ’1901— Bothriembryon haconi Kobelt, Syst. Conch. Cab. (Mart. & Chemn.), ed. Kuster, Hd. L, Abth. 13, Theil 2, heft CLXV., lief. 463, p. 770, pi. 112, fig. 18; pi. 116, fig, 6, sheet dated 8-7-1901. 1901 — Bothriembryon bulla id. ib., heft CLXVIIL, lief. 467, p. 777, pi. 113, figs. 15-16, sheet dated 10-7-1901. 1991— Bothriembryon rhodostoma id. ib., ]i. 786, pi. 114, fig 6, sheet dated 28-7-1901. 1902 — Bothriembryon darns id. ib. heft CLXXIII., lief 473. )). 930, p]. 131 figs. 1-2, sheet dated 8-4-1902. Probably Australia. Shell elongate, jilump, spire conical, short, last whorl plump, not obese, shell thin, whorls rather flattened, last whorl very large, mouth broad oval! elongate, outer lip thin, columella strai,ght, reflected, hiding umbilicus which however, still remains open but narrow. Coloration somewhat varied a greenish aiipearance prevailing: the background jiale green finely streaked with reddish brown and a jiale peripheral band overrun however by the streaking: sometimes the gnen prevails and there is almost a unifoi'n^ shell but this is rare. At others the red is predominant and the shell takes on a 24 Toil Iredale. reddish tone, but always the coloration is characteristic. The apex is punc- tate, the thiinbling tiiier tlian in the southern meio : the longitudinal irregular ribbing is cut into nodules by the concentric lining and there is commonly a notable deep line just below the suture which is never seen in tnelo, the sculpture being much more ])ronounced in this species. Some specimens elongate Avith age and thus we get Keeve’.s ‘'rhodostoma” and Pilsbry’s conispira. The normal shell measures 2.5 mm. in length by 1G.5 mm. in breadth, conispira being 25 by 14.5 mm., and others up to 28 mm. by 10.5 mm. Bothriemhryon bradshawi sp. nov. Plate II., tig. 14. A good series, collected by Mr. F. K. Bradshaw at Tambellup, north ol the Stirling Ranges, and sontli of Broome Hill, provides an interesting problem. The shell is small, thin, sjjire and mouth about equal in length, spire conical, mouth a little iullated so that it ajjpears someAvhat intermediate between the Idngii ami melo series. The apex is a little more elevated than in the mdo form, but not so elate as in the kinijii group, and is finely punctate, the sciilptiu'e fading away so that no vaiix-like division is seen. The adult seuljdiu'e consists of fine sloping radials being more rude as the shell develops, and contiiming on to the base of the last whorl. On the first adult whorl a tew widely spaced (‘onceiitric lines arc* seen which soon vanish. The early coloration is whitish mottled with reddish brown, the motllings massing so that the coloration of the last Avhorl resembles that of some of the well-colored forms of kingii. Length, 1SI.5 nun., breadth, 12 mm. Bothriemhryon irvineanus sp. nov. Plate 11., fig. 15. A series, collected by Mrs. Irvine, a very well-known IVest Australian shell lover, at Cape Naturaliste, along with B. naturalistanim, but probably occupying a distinct station ecologically, is here named. The shell is thin, elongale, ot the mdo form, but nari'ower and showing no perforation, recalling serpent in us, hut broader, the spire a little shorter than the aperture, columella twisted. Coloration uniform with no red circumbilical patch. The n()ex is coarsely punctate r('calling that of kingii^ but broader, more elevated than that of bulla; adult whorl strongly rudely railially ribbed crossed by concentric ditches iiroducing a strong snbnodulose effect, the radials persisting strongly on to the body whorl but concentiic ditches disappear on base. Length of type, 2G nun. ; breadth, 16 mm. Bothriemhryon richeanus sp. nov. Plate II., fig'. 16. A series labelled “Cape Riche, King George’s Sound, S.IV. Australia’^ by Brazier, proves that the locality “King George’s Sound’’ was used for the whole of that south-west l)lock, as Cape Riche is sixty miles distant, and is a well known landmark. The shells are quite distinct, being strongly grannlose recalling Iceuwinensis and hrazlerp but are iiarroAver than the former, and broader than the latter. They belong to the kingii series, hut are broader with the A Review of the Land Mollusga op Western Australia. 25 spire not attenuate and about the length of the aperture. The apical whorls are .strongly punctate, the adult ivliorls rounded, the sculpture being rather strongly nodulose, the nodules irregular, but squarish, and persisting but weakening on to the base of the body whorl. The shell is thin and conse- quently the outer lii> is thin, the columella reflected, leaving so minute a chink that the shell appears imperforate. The coloration is a dirty fawn with indistinct longitudinal streakings of dull red; the red cireumbilical patch is only seen in one specimen. The type measures 24 mm. in length, 13 mm. in breadth, the aperture 12 mm. Bathriemhryon leeuwinensis Smith 1894. Plate II., figs. 17 and IS. 1894 — BuUmus (Lipunts) leeuwinensis Smith, Proe. Malac. Soe. (Loud.), Vol. I., p. 94, ])1. VII., fig. 27, June; Cape Leeuwin, South-West Australia. 1900 — Bothriemlirifon leeuwinensis Pilsby, Man. Conch. (Tryon), Ser. 2, Vol. XIIL, p.' 13, pi. 3, fig. 49, A pi 23. 1901 — Botiiriemhrijon leeuwinensis Kobelt, Syst. Conch. Cab. (Mart & Chemn.),' ed. Kuster, Bd. 1., Abth. XIIL, Theil 2, heft CLXV., lief 403, p. 768, pi. 112, fig. 4 (sheet dated 8 VII. 1901). Smith’s comparison reads: “B. me'o is a broader and more bulky species, and B. kinaii is more acumitialc above, less granular, and exhibits more coloration both e.xternally and within Ihe aperture. The umbilicus is more closed than in the present species, and is surrounded by a brown zone.” The notable feature is the strong granulation and this aiipears to connect it with the kingii serie.s. Mr. A. W. B. Powell, of the Auckland Museum, has given mo three specimems collected at F'linder’s Bay by Chnneut L. Wragge, and these are to]iotype.s of the species. The apical features jirove it to be refer- able to the kinfiH series fi'om which at sight it appears very different. The shell is elongate, rather stout, the spire about equal to the np(>rtnre, but not attenuate, the whorls apiireciably shouldered. The apex is elevated, finely punctate, the |)unctalions separable until th(' end of two whorls, where a definite small varix can be seen. The adult scul])ture consists of coarse granu- lation, rough separate radials being cut into oblong lozenges by concentric deep irregular lines, the unevenness of the' radials givitig a nodulose sugges- tion. This continues on to the upper half of the last whorl, the rough radials alone being observed on the lower half. The columella is white, reflected, but leaving a notable umbilical chink behind; a white glaze connects the inner lip with the outer lip which is thin. The coloration is a uniform creamy white. The length of the figured shell is 29 mm., the breadth 15 mm. A specimen from Lake Cave, Margaret River, generally agrees but is much narrower, measuring 28 mm. by 13 mm. Another series from Margaret River is short and bi'oad, recalling the melo series, but the apex seems to Ix'long to (he kinijii series. The mouth and aperture are about equal, and the sculpture is similar to that of the above, but much weaker, the concentric lines disajipearing on the body whorl. The columella is reflected but hnves an open chink, and the outer lip is thin. The coloration is dirty whitt' with flesh stia'aks. There aiiiiears to b(! much con- fusion in the ai'ea from Cape Leeuwin to Cape Naturaliste, and for the present this form is being only named subspeeifically Bothriembryon leeuwinensis eventus subsp, nov. ; the type measuring 23 mm. in length by 14 mm. in breadth. Tom Iredalb. 2 () Bothriembryon costulatus Lamarck 1822. Plate II., tig'. 19. 1822 — Helix costulata Lamarck, Hist. Anim. s. Vert., Vol. VI., pt. 2, p. 122^ Apl. as synonym of inflatun infra, ex “Daudeb, Xo. 405,’’ which re- fers to 1821 — Helix (Corh]o':ina) costulata Perussae, Tabl. Syst. Anim. Moll. Lima(;ons, p. 58, Jan.; p. 54, June. Xomen nndnm. “Le port du Roi (leorge, dans la Xoiivelle-Hollande, Peron ; la bale des (Ihiens- IMarins, Gaudicho.” Shark Bay is here selected as ty|)e locality. 1822 — Balimits injlaUis Laimirck, Hist. Anim. s. Vert., Vol. VL, pt. 2, ji. 122, A))l. Nouvdle llollande. Not Bulimus in flatus Olivier, Voy. Othoman, Vol. 11. , p. 356, 1801. 1838 Bulnnus costulatus Potiez et ilichaiid, Galeric des Molusques Mus. Douai, \()1. 1., |). 138, pi. XIL, figs. 17-18, October, ex Cochlogena costulata Perussae. 1841 BuIiidus inflafus Ilelessert, Kceuell Co(]. Lamarck, pi. 28, flg. 1. Laniarck’.s .shell figured. 1864 — Bulimus' onslowi Cox, Cat. Austr. Laud Shells, p. 24. Dirk Hartog- I., West Australia (Onslow). Pigs. Cox, Mon. Austr. Land Shells, p. 74, pi. XIII., f. 13, May, 1868. 1892 — Li pants onslowi Medley, Kcc. Austr. Mus., Vol. II.. 29, Aug. 1894 — Bulimus [Lipams) onslowi Smith, Proe. Malae. Soe. (Loud.), ^Ml. I., p. 95, ])1. VI I., lig. 28 ()iot fig'. 29), June. Dirk Hartog Island, AVi'st Australia (J. J. Walker). 1900 — Bofliricmhri/on onslowi Pilsbry, Man. Conch. (Tryon), Ser. 11., Vol. -Xlll., j). 11, ])1. IW, figs. 43, 44, 48; Ajtex, I'l. IV., fig. 75, A])!. 23. 1901 — Bothriemh ri/on onslowi Kobelt, Syst. Conch, Cab. (Mait. & Chemn.), ed. Kuster, Bd. 1., Abih. XITL, Theil 2, lief 463, p. 769, pi. 112, fig. 15-16, (p. 765 dated 8 VI 1. 1901). The identity of l.amarckV, spei^ies .“oems to be obvious though hitherto overlooked. His descriidion reads; “B. testa ovata, ventricosa, jierforata, longitudinaliter striata, siptalide alba; spira obtusiuscala ; labro margine subrefiexo. Habile dans la X'ouvelle llollande. Mon cabinet. Longueur, jires d’un ponce.’’ This agrees very wc'll tvith the Shark Bay shell, the words “ventricosa . . stjualide alba . . spira obtusiuscula” being characteristic of that form. Perussae had jiretiousjy named the shell costulatus. and Lamarck re.jected this, renaming it injlatiis, but unfortunately the latti'r had been previously used by Olivier, so that renr sae's name comes into u e. .\s localities Fernssac gave King George’s Sound and Sharks Bay, and though the fornun' locality has b('en aiaepted for Lamarck's iiiflalus, the lignre given by Delessert is of the Shark Bay shell. This figure measures 27 mm. by 17 nun., and Shark Bay shells \ ary from 2() to 28 mm. by 17 mm., smaller ones 21 nun. by 16 mm., and 2] mm. by 15 mm. Smith’s figure of onslowi measures 24 mni. by 16 nmi., and Cox’s measurements of oitslowi are equal to 21.2 mm. by 15 mm. Peron records from Bernier (and other) isiand(s) in Shark Bay: “Two species of land shells extremely numerous, but all dead, occupied great A Review of the Land Molliisca op Westehx Australia. 27 stretches of the interior of the island; one was a siuall species of Helix, the other belonged to the genns Bullmun of M. de J.amarck.’’ Ferussac gave as the collector of liis 8hark Bay shells, ‘‘Gaudicho’’ ; this refers to C. Gaiidi- chaud, the apothecary on the lAanie, which visited Shark Bay in 1818. In further contirination, Botiez and Michaud are cited, as they figured the Shark Bay shell under the name Rulimus contulatus Ferussac, and acknow- ledged the as.sistance of Feinssac. The shell i.s roundly oval, the spire short, obtuse, less in length than the aperture, which is oval, outer lip thin, rather solid. The coloration of all the dead shells seen is white, but tlu" living shell is orange, longitudinally flamed with dull streaks. Aiie.x flattened not distinguished finnilv from adult whorls, the sculpture be ginning as faint wrinkled lines, the succeeding whorls sculptured with tine concentric lines cutting; the faint growth lines into small lozenges which vanish helow the periphery of the last whorl. Colunuhla short, rellocted tri- angularly over the umbilicus, which however is not dosed. Bothriembryon minor Pilsbry 1900. Plato IT., fig. 20. 1900 — Bothriembri/on onshnvi var. minor Pilsbry, Man. Conch. (Tryon), 2nd Sot., Vnl. XITI., p. 12, Apl. 2.3, for Smith, Proc. Malac. Soe. (Loud.), Vol. I., p. 9.1, pi. VTI., tig. 29, June. Dirk Hartog I., West Australia (J. J. Walker). 1901 — Bothriembrijon melo var. hrirtogensis Kobelt, Syst. Conch. Cab. (Mart. & Chemn. ), ed. Kuster, Bd. L, Ahth. 13, Theil 2, p. 770, pi. 112, flg. 15, dated June 19. Dirk Ilartog’s I., W.A. Smith wrote: “otislou'i .... Five other examples are considerably .smaller, averaging only 15 to 18 mm. in length. They are a trifle less globose, and more strongly granular just below the suture.’’ Pilsbry com- mented: “Apparently adult specimens of this small form before me are even smaller than the dimensions given by M. Smith, two measuring: Alt. 13.5 diam. 9, longest axis of aperture 8 mill, whoils D/gl «ihI alt. 14.5, diain. 8.2, longest axis of aperture 8 mill., whorls a trifle over 5. They are strongly granose below the suture, and striped longitudinally with pale reddish or grey and opaque white.'’ Kobelt later also named this form, and it seems common and easily .se])arahle and not a dwarf of '^onslowi.” Collecting might solve the problem as there is no similar instanee of dwarfing in this family. Bothriembryon whitleyi sp. nov. Plate IT., fig. 21. A nice series, collected by my colleague, Mr. G. P. Whitley, at Gerald- ton, is composed of white dead shells. They are, however, quite fresh and do not appear to have been coloured. Shell small, dead wliifc ralher shin- ing subglohose ])erforate, perforation hidden by reflected columella. Apex of two whorls, finely ])unctato, a subvarix seiiarating these from the adult four whorls which are longitudinally radially wrinkled, the racUals crossed by concentric line.s, wliieh produce a subnodulose a])pearanee, the base of the last whorl smooth. The mouth is oval, the outer lip thin, the columella rather broadly reflected, hiding the umbilicus, but leaving it quite oi)en. Height 16 mm., breadth 12.5 mm. This is a plumper shell than minor, and differs in coloration. 28 Tom ] redale. BottrieinlDryon perobesus sp. nov. Plato TT., flg. 22. One specimen from the month of the Moore River obviously belongs to the ‘‘on^loxci” series^ bnt is even more globose than the most inflated typical “onslou'i.” Tt is a live shell, and has lh(‘ same style of coloration as the onsloici forms, but in the living shell the ground colour is nale yellowish green, the longitudinal banding a rich brown. The shell is stout, the spii'e very short and somc'what flattened, the body whorl taking up the bulk of the shell which consists of four adult whorls and two a|)ical whorls, the latter being somewhat strongly tu'oded. The columella is l)road, white, strongly reflected, hiding the narrow umbilicus, which, how'evor, still remain.s 0 ]ien. Only a, slight subeancellate scupltftre ap]iears below the suture, the she'I otherwise being smooth, the growth lines scarcely showing u|). The height of the shell is 25 mm. while its breadth is ft) mm. the length of the aperture being 15 mm., its bread 8 mm. The annual rainfall is between 15 and 20 inches. Bothriembryon indutus iNtenke 1843. Plate TI., fig. 23. 1842 — BtiUtnus iurlutiis ilenke, Moll. Xov. Roll. Spec., p. 6, Apl. Darling Range and Mt. Eliza, West Australia. Fig’d. Cox, Alon. Austr. Land Shells, ji. 73, pi. 13, fig. 10, IMay, 1868. Piksbry, Man. Conch. (Tryon), Ser. 11., Vol. Xlll., ]). 13, pi. 3, fl.gs. o8-Cl, Apl. 23, 1 !)()(). Kobelt, Syst. Conch. Cab. (ilart. & Chemm), ed. Kuster, Bd. I., Ab(h X]lI.,' Theil 2, heft CLXVIII., lief. 467, p. 783, 1)1. 114, figs. 2, 3, 7, sheet dated 28-Xri.-lfl07. 1877 — TSulimus ponsonli; Angas, Proc. Zool. Soe. (Loud.,) 1877, p. 70, pi. xxvi., fig. 1. M^estern Australia (J. Gould). 1001 — Vanda '? pon.-^ntihui. Kobelt, Syst. Conch. Cab. (l\Iart. & Chemn.), ed Kuster, Hd. L. Abth. XI if., Theil 2, heft CLXVIII., lief, 467, )). 785, pi. 114, fig, 5, sheet dated 28-VI I.-lflOl. Angas, figure copied. Although this species is one of the most distinct it was overlooked for some time but is now well known from its form and coloring as well as size. Two notable colour variations occixr dull greenish yellow with dark growth period stop marks, and reddish bi’own with jiale yellow growth period sto)) marks. Apparently these shells stop growth, and the inside of the outer lip takes on a dilTereut colouring, and when the shell restarfs, it leaves a very distinct coutra.sting recoT’d of the stoppage. The shell is very elongate, stout, regularly narrowed about twice as long as broad, and while the colum- ella is twisted a slight umbilical chink is present, the spire about equalling the aperture in length. The ajiex is imnctate but usually eroded, and no varix can be seen, the adult whorls show a faint nodulose retieulatioji on upper whorls, the last wdiorl only with obscure growth marks. The specimen figured is fi'om King's Park, Perth, and measures 40 mm. in length by 19 mm. in breadth, wdiile smaller ones measure 34 mm. by 16 mm. The dead shells are white as the coloration lies in the rather thick periostracum which clothes the shell. ITedley (Proc. Malac. Soc. (Lond.), Vol. T, ]i. 260, 1895), wrote: “Judg- ing from the figure of VuHwas PonsouJiii. and from the travels of its col- lectoi’, I am disposed to deny that it is a Liparus, or that it comes from Western Australia, but hold it rather to be a variety of Vanda atomata col- lected in Xew South Wales.” There is now little doubt that Gilbert collected the specimens in the neighbourhood of Perth, and that the name is an abso- lute synonym of indatus. A. Review of the Land AIollusca of Western Australia. 29- Bothrienibryon glauerti sp. iiov, Plate IL, %. 24. A very distinct form belonging to the indutns series, but easily distin- guished by its shape, the very short spire and the swollen body whorl, which is more than two-thirds the bulk of the shell. There is a gios.sy green perio- stracum which .shows the eharaeterislie "indultm'’ growth stages in darker relief. The a|)e.x is elevated, and the incurved earliest iiorliou i.s coarsely anastomosely wrinkled, a coarse irregular pitting succeeding, with finally a fine wrinkling striae forming; with a strong lens this sculpture can be traced on the body whorl, and more notably on tlu' earlier whorls, suiisutural wrink- ling being [iresent. The columella is twisted, white, a white glaze crossing the body whorl, the aperture within being bluish white, the outer lip thin; there is no umbilical fissure left, the shell appearing imperforate though in the very juvenile stage a small chink may be noted. The tj’pe measures 38 mm. in length by 22 mm. in breadth, the length of the body whorl lieing 30 mm. and that of the aperture 21 mm. The locality is Stirling Ranges, which is included in the 20 to 25 inches annual rainfall belt. Bothriembryon fuscus Thiele 1030. Plate II., fig. 25. 1930 — Bothriembryon fuscus Thiele, die Fauna Slid- West Australiens, Bd. V., lief 8, ]3. 588, pi. IV., fig. 08. Torbay, South-West Australia. Thiele described a very immature s| ecimen which appears to be a repre- sentative of a very large series collected by Mr. Sidney W. Jackson many years ago in the karri foresLs at Nornalup Inhd., Deep River, which has a rainfall of 35 to 40 inches annually, which is tlu' same as at Toi'bay. The Denmark shells are here described, and if these should lie later |)roien dis- tinct, as Thiele’s figure is not e.xactly agreeable they may be called franki suggested by Mr. Jackson. Shell largo, elongate, narrow, spire and aperture about equal, latter oval, columella slightly thickened, approssed, commonly closing umbilicus, but very rarely this may be I’etained as a chink. Coloration varying from straw to dark brownish yellow, streaked with darker brown, especially along the growth lines. Sometimes the juveniles are uniform straw, at others dark chocolate; again shells may be found bearing a broarovides the dark growth streaking, a feature of all the indutrs series. The ape.x is elevated, niinuli'ly wrinkling sculptured, the wrinkles fading into growth ridge.s, the whorls showing no distinct separa- tion but succeeded by a fine concentric lining which never becomes jirominent and soon disappears. The growth lines are flattened so that they are more notable as colour streaks than ridges, and the faint concentric lines can only bo seen with a lens, the body whorl appearing smooth. Length (type of franki) 40 mm., breadth 21 mm., length of aiierture 20 mm. These” shells are found in hazel scrub up high on hazel trees and h'aves, up to 40 feet on the limbs. Bothriembryon kingii Gray 1825. Plate IL, figs. 26-28. 1825 — Bulimus kingii Gray, Annals Philos (Thomson), A'ol. XXV., (2 Ser.,, Vol. IX.), p. 414, June. Xew Holland (Capt. King) = King George’s Sound, South-West Australia. 30 Toil Ieedalb, 1826 — Ileli.r hulinnis Ivin”', Xarr. Sui'v. Coasts Austr., Yol. 1., p. 12, “1827” — Apl. 15, 1826. Near Bald Hoad, King George’s Sound. 1828 — Helix kingii Wood, Siippl. Inde.v Test, p. 22, pi. 7, Helix, tig. 27, (prel. Hay 17), New Holland. Brit. Mils. Type figured. 1822— J7e/*,r trilineata Qnoy & Gaimard, Voy. Astro!., Zool., Vol. II., p. 107, pi. IX., tigs. 1-3. Bald Head, King George’s Sound, S.W.A. 1848 — Bulimux triliiieiitiis R(>eve, Coneh. Iron., Vol. V., pi. XLVIII, sp. & f. 310, Xovemher, as ol (I. & G. New Holland. 1849 — BnJimus IrUineutux Reeve, Conc-h. Icon., Vol. V., ])1. LIX., sp. & f. 397, January, ex Q. & G. “Port George,’’ New Holland. 1864 — (BiiliniHs) (inoi/i Cox, Cat. Austr. Land Shells, p. 23, new name for Reeve’s fig. 397, not Reeve’s fig. 310. 1868 — Bnlimus kingi Co.x, Mon. Austr. Land Shells, p. 75, pi. XIII., fig. 7, ilay. 1892 — Liparus kingi & var. trilineatiis Hedley, Ree. Austr. Mus., Vol. II., p. 29, Aug. 1900 — Botkriemhrgon kingii Pilshry, Man. Coueli. (Tryon), Ser. II., A’ol. XIII., p. 7, pi. 2, figs. 21-28, apex, iil. 4, fig. 77-78, Apl. 23. 1900 — BotJiriemhrgon phgsoidex var. humilix Pilshry, iMan. Coneh. (Tryon), Ser. II., Vol. XIIL, p. 10, pi. 11. , figs. 33-34, Apl. 23. King George’s Sound, M'est Australia, 1901 — Bothriemhrgon tiiUneatus Kobelt. Syst. Coneh. Cab. (Mart. & Chemn.), ed. Kuster, Bd. L, Abth. XIIL, Theil 2, heft CLXV., lief. 463, |). 7(i6, pi. 112, figs. 5-9, (sheet dated 8, VIL, 1901). 1901 — Boihriemhryon h'ngii Kobelt, id. ib., lull CLXVIIL, lief 467 ]i. 779, pi. 113, figs. 19-23, pi. Ill), figs. 14-15, (sheet dated 10 VIL, 1901). 1901 — Bollirienihrgon /digsoidex var. humilix Kobelt, id. ib., p. 790, pi. 116, figs. 9-10, (sheet apex elevated, the mouth elongate, shorter than the siiire, the columella reflected and appressed with a twist, obliterating the umbilicus, outer li|) thin. The shell is thin and deli- cate, and the coloring is a dirty while slrealu'd more oi' less thickly with pale brown, rarely of a reddish tinge. The streaks are commonly notable as in the typo of kingii, and are as commonly almost obsolete as in the type of trilineahix. Some are almost clear white', othei’s clear ymllowish, and in some cases the streaks coalesce, and an almost uniform brown body whoil i.s formed. The thinness and shape are however constant as is the lack of any defined sculpture, the rude flattened radials showing no cross sculpture. The apex ; consists of two whorls, the ti|) elevated, minutely punctate, the punctations j finally running into lines: the adult whorls follow without a varix, but the A Keview of the Land Mollusca of Western Australia. 31 change can be seen. The columella is rellecled, closing' the umbibcus as a general rule, but in some cases as in trilineatun typical, as hero shown, the swelling of the body whorl allows the retention of a slight chink indicating that the juvenile is perforate. A circumihlical red patch is ir ually present in this group. The two speciinen.s figured are from the type locality, and represent normal lingii and trillHCdtnu. the former measuring 24 mm in length, by 13 mm. in breadth, the Initer lieing '29 mm. in length, by 15 mm. in breadth. The ty])e of Pilsbry’s var. humilis is also here ligured, and it measures 17.5 mm. in hmgth, by 10 mm. in breadih. 11 is obviously referable to this series, but differ.s from tlie typical form in showing a coarser longitudinal radial ribbing, cut by a few concentric lines on the eaT'lier Avhorls and on the shoulder of the last whorl. There is an umbilical chink, and no red circum- bilical patch, and somewhat similar shells have been seen from the Pallinup River, South Coast. Bothriembryon notatus sp. nov. Plate II., fig. 29. An excellent series from the Pallinup River, south coast, recalls B. jaclfscni, but the shells are narrower and have a paler ground colour. Shell thin, elongate, of the Ungii style, but the spire not so attenuate, and the mouth is e(|ual to the spire in length. Coloration greyish-white longi- tudinally flamed with red brown, the markings varying in intensity, some showing few marks being almost iniieolor, while others have the markings running together so as to produce an almost uniform dark shell. The red circumbilieal ])alch is always present. The apex is of the typical kingii form, the succeeding whorls sculptured with wid(‘ longitudinal ribbing, which is most pronounced below the suture producing a puckered collar. This per- sists on to the body whorl, where otherwise the ribbing becomi's obsolete. The columella is reflected so as to hide the umbilicus, which, however, re- mains open as a chink. Length 24 mm., breadth 11 mm., length of ai)erture 12 mm. Bothriembryon jacksoni sp. nov. Plate II., fig. 30. A very interesting form of the kinfjU series was collected by i\Ir. Sidney W. Jackson at Dee)) Hiver, Krankland River, Nornalup, to the west of King George Sound, on ti'ces and shrubs in the hazel scrub among the karri. It lacks the acuminate .spire having a rather swollen body whorl with a conical short s|)ire, the texture thin, the colouring bright fawn with deep red brown streaks, and a deep red brown circumbilieal |iatch. The apex is finely regularly jninctate throughout the two whorls, the sculpture on the subsequent whorls being conqiosed of rude radials crossed lightly on the earlier whoi’ls by line concentric lines forming a line nodulation at places, but irregular and becoming obsoleti' on the body whorl. The columella is twisted, white, and retlectcd leaving an umbilical chink. The length of the type is 27 mm., the breadth 1.5 mm., length of si)ire 13 mm. Bothriembryon max-wePi Kobelt 1901. Plate 1 1., fig. 31. 1901 — “Bothriembrgon” ma.ni-elli Kobelt. Sy.st. Conch, Cab. ( Hart. & Chemn.), ed. Kuster, Bd. 1., Abth. XIII., Theil 2, heft CLXVIII., lief 467, )). 781, pi. 112, figs. 4, 5, 17, sheet dated 28 YII. 1901, ex Brazier M.S. No definite localitv = “Doubtful Island King George’s Sound, W.A.” .32 Tom Iredale. 1868— BuJimus limji Cox, iJoii. Austr. Land Shells, p. 75, pi. 19, fig. 12, 12a, Hay. Doubtful 1., W. Australia (Ma.sters). im6—Bothrifmbr!jon kingii Pilsbry, l\Iau. Conc-h. (Tryon), Ser. II., Vol. XI 11., pi. 2, ttg. 29 only, Apl. 23. Although Cox figured this form ho did not mention it in the text, and Pilsbry also figured a .S])eeimen whieh is here rofigured, and then Kohelt brought in a Brazier iMS. name tor this sliell. There are many specimens in the Australian Hiiseiim hearing Brazier’s M8. name, and localised as Doubtful Island, hut U’hicli Doubtful I. is intended must he found out by local collectors On the labels here it is said to be “Doubtful Island, King George’s Sound,’’ and many shells of the vi(do series are also so lalxdled, one set being the castaneus form (Pilsbry ’s fig. 18)., and another set agreeing with Pilsbry’s figure 3. It there he three distinct series of Both n'enihn/on living on Doubtful Island, the ])lace is in need of close investigation. The shell is a small thin representative of the kingii form, but is smaller, nari’ower and not so atten- ! uate as the tyi.dcal shell. It is always of a fawn colour, rarely streaked i, with reddish and the red circumbilical 2 >atch is usually absent. The apex i is elevated, normally i)unctate, th(> punctations linking u]) into lines, the adult scul]}ture easily separated, rough radial ribbing develo})ing below the suture Avhere it persists through growth. The radials become obsolete with ' age, while a regular series of concentric lines ai)i.H‘ars on the earlier whorls, none of which cuts the ra- ture regular wavy growth lines rvith transverse cross lining but the rvholo seuli)tur(* very weak: the iiuckering below the suture is well marked and a little of the longitudinal radials remains on the body whorl. Golumella twisted and api^ressed so that no umbilical fissure remains, the shell appearing imperforate. A thin glaze crosses from the inner lip to the edge of the oiiter which is thin. Length of type 24 mm., breadth 12 mm. The locality is in the annual rainfall 15-20 inches belt. Bothriembryon sayi Pfeiffer 1847. Plate II., figs. 33 and 34. 1847 — BuJimna sai/i Pfeiffer, Proc. Zool. Soe. (Loud.), 1846, p. 114, Jan. 26, 1847. Locality unknown = Cai'e Ereycinct, West Australia. Eig’d Reeve, Conch. Icon., Vol. V., pi. LXV., sp. and fig. 458, 1849. A Review of the Land Mollusca of Western Australia. 33 1900 — Bothriembrijon Idngii var. solidus I’ilsbi’y, Man. Conch. (Tryon), Ser. n., \‘ol. XIII., p. 9, pi. II., fig-. 28, April 23. West Australia = ilarg-aret River. Specimens from Cape Freycinet agree so closely in their features with the figure and flcseription of sagi that they can be so named, while solidus PiLsbry aiii)ears to be merely a large local variation of this species. The typical shell i.s small, with an attenuate spire with the whorls flat- tened, rather solid, with the und)ilicus always left a little open. The colora- tion is whitish with a few pale real streaks at times. There is no civciimbilical patch in this series, other wi.se there is no disen'pancy at all. A rather notable feature is the thickening of the columella and the strong white glaze connecting the inner and outer li])s, the latter having th(' edgi' stouter than usual. The tyi)e of solidus, here figured, is much larger and more solid, and be- longs to a series from the Margaret River, having the same conical attenuate spire with flattened whorls, and the longitudinal sculpture a little stronger. Bothriembryon naturalistarum Kobelt 1901. Plate II., figs. 35 and 3G. 1901 — Bothriemhriion kiiioii var. tiatnridistariun Kobelt, Syst. Conch. Cab. (Mart. & Chemn.), ed. Kuster, Bd. 1., Abth. Xlil., Theil 2, li ft CLXVIII., lief. 407, p. 781, pi. 11.3, figs. 22-3, sheet dated 28-YII.- 1901. Caiie Xaturaliste, IVi'st Australia. 1900 — Bothriemhriion kinijii Pilsbry, Man. (Tmch. (Tryon), 8er. II., Vol. XIII., p. 9, pi. 2, fig. 30 only, A))!. 23. Cape Xaturali te, IX. A. Pilsbry wrote: “Fig. 30 is a small specimen from Cape Xaturalist. There are 5 whorls, the last inflated below, almost imperforate. It i.s beauti- fully streaked with opa(|ue, glossy white on a cortu'ous and inirplish brown ground. Alt. 20, diam. 11, longe.st axis of aperture 10, 7 mill. The aperture is decidedly more obliipu' than in the tyjiical foi-m from King (leorge Sound. It is not unlikely that this will prove a geographic race or variety. From the Cox collection.” Kobelt gave this the varietal name mil urulisto.rum. The series from which Pilsbry selected the figured .sb.ell is before me, and I am refiguring his specimen. Again it is immature, but numbers collected by Tom Carter at the same locality show that it develoiis into an elongate shell very like that from Margaret Ri\er, which 1 regard as sagi, but it is thin and has the mouth more e.xpanded, is smooth, and measures 25 mm. in length by 13 mm. in breadth. Local naturalists must study these shells and fix the status of the named forms. Bothriembryon revectus sj). nov. Plate 1 1 ., fig. 37. IVhile collecting numerously B. jacksoiii at Deep River and Bow River, Mr. Sydney IV. -Jackson also found alive under debris on ground in dense hazel scrub at the Bow Rix'er some .shells of the kinijii series, but sculptured after the style of hrazieri and leemvifiensis. The narrow shell here selected as type measures 23 mm. in length by 11 mm. in breadth; a broader phell measures 23.5 mm. in length by 12.5 mm. in breadth. The sjiire about equals the aperture in length, and is not attenuate. The columella is appressed so that barely a chink remains, and there is a red eireumbilical jiatch, while the 34 T02X Iredale, apex is of the I'iiirjii style. The sc-nl|)tui'e is siihnodulose, the radials iiTeg'iilar and rather ill-delined, wliile the coiieentrie lines vary in strength, but the sculpture is no(al)Ie in evei'y ease. The eoloratioii is fawn, more or less llam.-d ith red brown, the shell thin and the surface lustrek'ss. A similar shell was found on .small bushes” “in jarrah forest, Mt. h’lank- land," three months later by Jackson. Smith’s record of pli/ii^oidrs (Proe. i\Ialac. Soc. (Lond.), Vol. 1., p. 95, pi. VIL, figs. 30-31, June, 1894) may refer to shells of this form. Bothriembryon brazieri Angas 1871. Plate II. , fig. 38. 1871— (Lipariifi) brazicvi Angas, Proc. Zool. Soc. (Lond.), 1871, p. 19, |)1. I., fig. 28, June 12. Sinclair's (eri'or for Sfirling) Range, King George’s Sound, West Australia. 1873 — Buhnnift (lAjiavui^) bncicri Brazier, Proc. Zool. Soc. (Lond.), 1872, |). S07, correction of error of locality name. 1892 — Lipanif! brazieri Medley, Hec. Austr. ilus., Vol. IL, p. 29, Aug. 1900 Bothri(iml)r!ion pliijsoides var. brazieri Pilsbrv, llan. Conch. (Tryon) Ser. 2, Vol. XllL, p. 10, pi. 2, figs. .38-40,' Apl. 23. imi—Bothriernbri/on brazieri Kobelt, Syst. Conch. Cab. (Wart. & Chemm), ed. Kuster, Bd. I., Abth. XIIL, Theil 2, heft CL-WMI., Lef. lOl’ p. 778, ])1. 113, ligs. 17-18, (sheet dated 10, Vll, 1901). this e.vtracmd inary developtnent of the lunpii series is sliorfer than usual, not so acuminate in the s|iire, the Avhorls more, r'ounded. The coloration is a brownish fawn with a texv red radial streaks and a. bright red circumbilical patch. The sculpture is coarsely granular, the rude longitudinal ribs being cut into nodules by deep concentric lines, and this sculpture continues on to the base of tlu' body whoil. The apex is of the kinijH st.vle, linely imiictate and stop})ing abruptly at the oul of two whorls. The columella is white, appressed leaving a small chink. The shell is thin, and the figured one measures 19 mm. in length by 10 mm. in breadth. Bothriembryon gratwicki Cox 1899. Plate IL, fig. 39. 1899 — Biilimiinis (jratu'icki Cox, Proc. Linn. Roc., N.S.W., Vol. XXIV., p. 43)5, fig. in text, December 9. 50 miles East of Israelite Bay, South-West Australia. 1900 — bBdhrienibrpoa prafu'icki Pil.sbry, Man. Conch. (Tryon). Rer. 2, AMI. XllL, p. 11, pi. 2. tigs. 31, 32, ])1. 45, figs. 2-4, Apl. 23. 1901 — Boiliriend)ri/ov gratwicki Kobelt, Ryst. Conch. Cab., (Alart & Chemn.), ed. Kuster, Bd. I., Abth. XIIL,' Theil 2, heft CLXVIIL, lief. 407, p. 789, pi. 110, fig. 718, (sheet dated 2, VIII, 1901). This interesting shell is elongated, narrow, spire long, pointed, longer than a])erture, which is narrowly oval, solid, chalky, umbilicate. The colora- tion is chalky white. Strongly sculptured tlu’oughout, thereby differentiating this species from all others save brazieri. The .seulihure consists of elevated irregular roundc'd ribs, a])i>ai'ently intensified growth lines, over-run with concentric lines forming a subnodulose ornament, the nodules irregular in size and form. A Keview of the Land Mollusca of Western Australia, 35 The apex is coarsely wrinkly striate around the upper part of the whorls, the lower rcticulately i)ittcd, coiisistiii" of two full whorls, the adult sculpture abruptly forniiug’ tliereafter, almost a varix intervening. The oolumella is strongly reflected, hut does not hide the umbilicus, and joins the outer lip, which is thin, by means of a glaze across the body whorl, almost freeing- the mouth. Length of figured shell, 27 mm. ; bi-eadth 12 mm. The locality is in th(> annual rainfall 10-15 inches bolt. Bothriembryon dux Pfeiffer 1801. Plate IP, tig. 10. 1861 — BnUmiis du.r Pfeitf'er, Proc. Zool. Soe. (Loud.), 1861, p. 21, May 1. King fleorge’s Sound, South-West Australia. Pig’d. Cox, Mon. Austr. Land Shells, p. 71, pi. XI II., fig. 1, pi. XVIII., fig. 16, May, 1868. 1892 — Liparuf! (hi.i- Iledley, Pec. Austr. Mus. Vol. II., p. 29, August. 1891 — Bulimutt (Lipurus) du.r Smith, Proc. INIalac. Soc. (Loud.), Vol. 1, p. 91, June. 1900 — Bothricyyibrnon du.r Pilsbry, Man. Conch. (Tryon), Ser. IT., Vol. XTII., p. 3, pi. 3, fig. 62, Apl. 23. 1901 — Botlirirrydjri/OM du.r Kobelt, Syst. Conch. Cab., (Mart & Chemn.), ed. Kuster, Pd. 1., Theil 2, heft CLXV., lief. 163, p. 763, ph 112, figs. 1-2, sheet dated 21 ; VI. : 1901. This niagnilieent species stands alone in its large size, its white colour, and its solidity. The ajiex is regularly punctate, the scul])ture is of rough radials crossed on the earlier whorls by fine concentric lines, the cross sculpture vani.sliing on the shoulder of the last whorl. The columella is broadly reflected, concealing an open umbilicus, and the outer lip is thin. The spire is about the same length as the aperture. The shell figured measures 60 mm. in length by 35 nim. in breadth. Odd speeimens ai-e known from Norseman, Duiidas, Salmon flnras, Balladonia, Esjieranee Mallee Belt, and it thus seems to be an inhabitant of the drier inland belt and not coastal although all the early collected speeimens are labelled King George’s Sound. Bothriembryon barretti Iredale 1930. Plate TP, figs. 41 ami 42. 1930 — Botlirieyyihii/on barreUi Iredale, Vii't. Xntnralist, Vol. XLVII., p. 119, fig. in text, Xov. Nullarbor Plain, S.W. Australia. 1879— Bidimus indutuf! var. paUidufs Tate. Trans. Proc. Roy. Soc. Adelaide, South Austr., 18i8-9, ]?. 134. Ruiula Plah-au, Xullarbor Plain, South Australia. Not B. palUdus C. H. Adams, Proc. Post. Soe. Nat. Hist., AMI. IT., p. 12, 1845. 1900— Bothriembri/on inrhifm var. pallidus Pilsbry, Alan. Couch. (Tryonl, Ser. II., AMI. XIIT., p. 15, pi. 3, figs. 63, 64, 65. Apl. 23. 1901 — Bothriembriion indutus var. palliduyy Kobelt, Syst. Conch. Cab. (Mart. & Chemn.), j>d. Kuster, Pd'. I., Ahth. XIII., Theil 2, heft CLXA'III., lief 467, p. /84, pi. 114, fig. 8 (sheet dated 28-AMI.-1901). Pilshry’s figures copied only. 36 Tom Iredale. The original description of barretti reads: ‘‘Shell elongate, dirty white, aperture about equal to length of spine (spire). The apex is sculptured with a curious fine wrinkling, which develops into line radials which are later crossed by revolving lines. This causes a. mat-like appearance of rather longitudinal nodules, wTiich, however, disa])pear uijon the base. A young e.x- aiuple shows the spiral sculpture iiiucli more prominently than in the adult. There is an umbilical chink pi'csent, and the outer lip is sharp. Whorls, six; length, 27 mm. ; breadth, 15.5 mm. Th(' type of B. barretti was collected near Hampton inside the Western Australian boundary. 'rat('’s locality, ‘‘Bunda Plateau” also cro.sscs the boi'dcr line, and in a series sent to Pilsbry he ligured two variations, the general label being ‘‘Kucla.” Among the shells before' me now there are also two varieties, and it is now seen that the coaslal shells are narrower than the more inland ones. I’pon referring to the rainfall again, that is given as being 20-25 inches on the coast, and only 15-20 inches in the inland area. The narrow form was figured by Pilsbry, f. 63, and his ispecinn'ii is here religured as type of a sub- species, B. b. indictus nov., the shell measuring 30 mm. in length by 15 mm. in width, the series coming from Eucla. Bothriembryon distinctus sp. nov. Plate IT., fig. 43. A series of shells from Cardanumbi, west of Eyre, is of great interest, the shells having an acuminate sjiire with a swollen body wlioii. This shape is not .seen in any other series, the Itingii being elegantly slender, while the melo forms are regularly oval. The coloration is dirty white in the dead shells examined. The apex is not elevated, wrinkled, Ihe vvi'inkliug being succeeded by a punctation, no subvarix vi.sible, the adult sculpture' being ol' the usual rough radials, some- what subdued, and crossed by tine concentric lines, irr('gnlarly forming squarish nodules, Avhich disa])pear aItog('ther on the body whorl, only faintly discernil)l(! on the penultimate one, and never very strong on any of the earii('r whorls. Although the umbilicus is open, the bi'oadly rellected columella hides it, the mouth is rather large, the outer lip thin. The spii'e is longer than the aperture but the body whorl is large and dominant. The h'ligth of the type is 27 mm., breadth 15.5 min., length of the aperture 12 mm., the breadth of the pi'iiultimate whorl !) mm. The locality is in the annual I'ainfall 10-15 inches belt, and from the 70- mile tank at Balladonia similar shells have been sent along with barretti. As above stated, this sjiecies does not fit into any of the named sub- generic groups so is here differentiated as Celatembryon subgen. nov. FAMILY LAOMIDAE. Though this family name is continued, it may need revision as all the Australian forims have unarmed mouths, and a distinct shell apjiearanco, the Xeozelanic typical Lrioma having teeth in the aperture, and a different shell texture. However our shells appear to be of Paralaomid alliance, and that is a Neozelanic group with .suggested relationship to Luoma. Gemis WESTRAIiAOMA gen. nov. Ordinary looking Paralaomids in general appearance, but with the protoconch concentrically spirally striate. The shells are small, generally under 3 mm. in breadth, and 1.50 mm. in height, depressedly conical with A Review oe the Land Mollusca op Western Australia. 37 comparatively largo apical whorls, the ajiex, as above, concentrically spirally striate, the adult sculpture radial threads, with larger radial ridges, more or less distant, and usually well marked, rarely with concentric striation. The umbilicus is com|iaratively wide and open, the mouth circular, the lip thin. The typo of Wef which is more notable in the undulical cavity, which is about one-third the breadth of the base. Shell subdiscoidal, spire little elevated, ape.x finely concentrically striate, month subcircular, outer lip thin, columella a lillle reflected. Breadth 2 mm.; height 1.1 ram. Nangeennn via .Merredin, W.A., collected by E. Sedgwick. Genus INSULLAOMA Trcdale 1937. 1937 — I'lisullaomd Iredale, South Austr. Naturalist, Yol. XV III., p. 19, Sept. 30. Ilaplotype Parahtoma riddle! Iredale. This name was introduced .snbgenerically for the South Australian forms of Pandooma, as these had the apical whorls spirally striate or even lirate, whereas the fyi)ical form had the apex smooth or radially fiiudy striate. Thu shell se](‘cted as type of IfisuUiionid^ riddle!, was soTiiewhat elevated, umhiHciis narrow, finely reticulately sculptured, and had the a|)ox boldly strongly oon- cientrieally lined. Of course this cannot be seen withoul a .sli'ong lens. More- over, in this ease there was a pei'i|)heral subkeeling. The recognition of a Western Australian form agreeing in all these features has led to the admis- sion of this group geuerieally. Insullaoma predicta sp. nov. Plate 1., fig. 22. Anotln'r of the discoveries of Mr. Sidney W. Jackson at Bow River-, South ConsI, especially as it appears to l)e a true Ttisiillaoiiia. Shell small, subdiseoida], spire short, brown, undriliens narrow. The apical two whorls are large and bulbous, and are coarsely s])irally lirate: the adult whorls are closely flue radiate, with a line eoncenlrie striation ei'os.sing it, the last whorl subkeeled. The mouth is subcircular, outer lip tliiu, columella a little re- flected, the umlrilicus oiien, narrow, less than one-third the width of the base. Bi-eadth 1.25 mm. : height .6 mm. Found alive under karri bark at foot of karri tree on edge of wattle scrub. Genus G-KATILAOMA gon. nov. Ty]re G-. cara sjr. nov. Plate 1., fig. 20. This beautiful little shdl collected by Tlr. Sidney IV. Jackson at the Bow River, South Coast, suggests an elevated relation of Westralaoma, as it is sub- conical with similar radials, but has a cross sculpture of crosslining, the umbilicus being nan-owed. As Westralaoma is constant in form it seems better to name the present form separately than r-efer it to that genus, and thus spoil the homogeneity of the group. A Ke\-iew of the Land Mollusca of Western Australia. 89 Shell very .small, siil;coiiical, whorls a little rounded, sutures lightly im- pressed, last whorl deseeiidiiig rather' rairidly, mouth sirbeireirlar, oblique, outer lip thin, eolumella rather' broadly relleeted, obscirr'ing- a little the narrow open do('p umbilierrs. Color'atioir pale browrrish. Aprex very finely eon- ceritrieally striate, aborrt two whorls, adult whor'ls thr'ee arrd a half, sculpture with fine growth striae and thr'srr ar'e crossed by ver'y obscure eorreentrie striatrou. A ferv distant large atr'iae develop but they carr scar-ccly be called ridges: these rrurrrber about lwenty-tr\-e to llrir'ty birt they are irregular'ly spaced and never- pr'omirierrt. Breadth 1 rrrrrr. : height .8 mm. Found ali\’e under' karr'i bark in dense wattle ser-ub. FAMILY DIPNELICIDAE. This family was introduced for arr irrteresting' shell from Hammock Island, South Austr-alia, and it was r-emarked “It has not yet been discovered on the mainland, which srrggests that it is a I’elict form of great age.” Mr. Sidney W. -laeksorr collected, at the Bow Kiver arrd Deep River, specimens of another cur-rorrs shell, arrd though it ditfer's in detail it seems to recall the South Australian shell rrr sortre respects. It apparerrtly does not fit into any of the other families, such as Char'opidae, Lamnidae, Flanrmulinidae or Rhytididae. Germs ANNOSELIX gen. nov. Tyjre A. dolosa-sp. nov. Plate I., lig.s. 24 and 27. A very curious little shell, i-egrrlarly broadly conical, the base convex, the pori|)hery keeled, umbilicus deep, very rnirrow, the mouth squarish, the outer lip thirr, columella. Coloration brown, base paler browrr, but sometimes flamed with reddish brown. One apical whor] apparently smooth but obscurely radially str-iate, the str'iae developing on the adult whorls Avithorrt any' intervening varix. Thera are four and one-half adult whorls regularly increasing, whorls flattened, sutures scarcely inurressed. The r-adial striae develop slowly and become very numerous and fine, while a delicate concentric striation also forms, j)rodueing a very fine semi-cancellate appearance. On the ba.se the concentric spirals generally predominate. Breadth, 7., 5 mm. ; height 5 mm. The type is from Deej) River, and some specimens from Bow River. South Coast, show that, in some cases, the radials, every now and then, strengthen into ridges and create a fringe at the periphery and continue strongly on to the base. This fringe sticks out a little, and presents a saw- like appearance to the peripheral keel. FAMILY CHAROPIDAE. This family has a few members in Y'estc-rn Australia, but there may' be many' more, as they are minute and difficult to find. Seven species have been described and while one is rejected, a few are added. One eastern genus, Femarjera, based however on a western shell, i.s admitted, Init the remainder of the species ajipear to be of endemic origin. All the species so far seen have no armature of the mouth, and all are openly- umbilicate. The features of the family are the small size of the shell, the depressed helicoid form, com- monly- discoidal, with an adult sculpture of numerous radial ridges of varying strength. 40 Toil lEEDALE. (ieniis PEKNAGERA Trcdale 1933. 1933 — Pemageru Iredale, Hoc. Au.str. Mas., Vol. XIX., p. 53, Aug. 2. Ortho- type Helix (ubanensiti Cox. Shell siriall, somewhat elevated for this family with a. very wide cav- ernous umbilicus, the radial sculpture being rather coarse. Whorls loosely coiled, sutures deeji, almost canaliculate, atiex radially striate, tlu' tip smooth. The mouth is subcircular, outer lip thin, columella straight, .scarcely reflected at all. Pernagera albanensis Cox 3808. Plate I., llg. 23. 1868 — Helix alhancnxis Co.x, Proc. Zool. Soc. (Loud.), 1867, p. 723, April 3. King George’s Sound, We.st Australia (Ma.sters). Fig’d. Cox, Mon. Austr. Land Shells, p. 15, pi. IV., tig. 2, May, 1868. Shell small, almost subglobose, spire conical, whorls rounded, loosely coiled, almost seiiarated by a canaliculate suture, the last whorl descending, other characters as for the genus. Coloration fawn, irregularly rayed with reddisli. Ajie.x tip smooth, then radiately striate, the striae developing into riblets which are iiunKu'ous and fairly regularly spaced, about .seventy-five on last whorl, the interstices finely radially striate, no concentric lines apparent, t'mbilicus funnel shaped, about one-half the width of the shell. Breadth 5 mm. ; height 3 mm. Pernagera lena sp. nov. Plate III., fig. 3. A very pretty li ttle .shell, col haded hy Mr. Sidney W. Jackson at the Bow River, recalls allxineuxis in miniature. Shell .small, subdepressed, sirire .slightly elevated, whorls rounded, sutures deej), almost eanaliculate, last whorl deseending a little, mouth subcircular, eoliimella straight, scarcely refleetcil, umbilicus wide, fniiiiel shaped exposing all the whorls. Coloration pale brownish with broad red (lames. A|)e.x smooth at tip, then finely radially striate, the striae developing into ridges wliich are regular and numerous, numbering about one hundred and twenty on the last whorl, the interstices striate. Umbilicus about half the widtli of the base of the shell. Breadth 3 mm.; height 1.5 mm. Under hark and leaves on the ground in Hazel scrub. Gciiiis LUINODISCUS Iredale 1937. 1937 — Luinodisenx Iredale, Austr. Zook, Vol. Vlll., ]>. 331, Mch. 12. Ortho- t.vpe Helix cnprea Cox. “Stnall flattened eharopid sludls, .spire not sunken, umbilicus wide, mouth small, thin, seulpture of regular fine radials, protoconch s^nrally lirate.” While the type of Luinodixcux is a small shell, other species, referred for the present to this group, ai'e large and more bulky. Luinodiscus cupreus Cox t868. Plate ITT., fig. 5. 1868 — Helix, cnprea Cox, iVlon. Austr. Laud Shells, p. 22, pi. XIT., fig. 9, May. King George’s Sound, West Australia ((Masters, A. M.). 1875 — Helix (Charopa) nnpera Brazier, Proc. Linn. Soc., N.S.W., Vol. I., p. 18, Apl. 27. King George’s Sound, W.A. (Masters). A Review oe the Land Mollusca of Western Australia. 41 Shell di.seoi(lal, spire not sunken, whorls rounded, sutures deeply im- pressed, last whorl doscondiug a little, mouth subcircular, lip thin, columella not reflected, straight, umliilicus very wide. The apex is eoncentrically striate, the adult whorls closely finely ribbed, ribs about eighty in number, interstices striate. Coloration fawn. Breadth 2..') mm. ; height 1.5 mm. Luinodiscus sublestus Benson 1853. 1853 — Helix snblesta Benson, Ann. Mag. Nat. Hist., Scr. 2, Vol. VI., p. 30, Jan. 1. Freemantle, Swan River, West Australia. rig’d. Reeve, Conch. Icon., Vol. VI., pi. 174, sp. 1177, Oct. 1853. Fig’d. Cox, Mon. Austr. Laud Shells, p. 1C, pi. XI., fig. 10, May, 1868. 1930 — Charopa liedleyi Thiele, Die Fauna Siidwest Austr., Bd. W, lief 8, p. 587, pi. IV., fig. 67. Brunswick Benson naturally gave no details as to the protoeoneh features nor the number of ribs. It wa.s differentiated from ‘'H. luloidea, Forbes, by its narrower umbilicus, and the want of concavity in the spire; from H. cygnea by the flr.st mentioned feature, and the sculpture.” Thiele does not compare his new species with this, and generally there/ is agreement, the protoconch being described as spirally sculptured, its size being given at 3.75 mm. in breadth, 1. 8 mm. in height, with the umbilicus I mm. wide. Benson’s description is here given a.s the species needs fixation : “Shell rather broadly umbilicate, orbiculately depressed, above red horny, below horny, very minutely costulatdy striate above, rather flattened, sutures im- pressed, whorls four, the last a little depressed, rounded, aperture vertical, roundly lurate, peristome acute, umbilicus perspective. Diam. major 3, minor 21 / 2 , axis 1 mill.” Specimens from the Swan River, collected by R. Helms, measure up to 4 mm. in breadth by 2 mm. in height. These generally agree that the spire is flattened, but not depressed, the umbilicus wide, open, perspective, fully one-third the width of the base. The apex appears to be smooth but under a high power is seen to be finely concentrically striate, the adult sculpture of radial ridges shows between ninety and one hundred on the last whorl, the intestice.s being crossed by very fine eoueentrie threads, only discernible by a high power. iS])ecimons from Pe)ipermint Grove arc consistently smaller but other- wise there appears to be no definable distinction. Luinodiscus cygneus Benson 1853. Plate I., fig. 25. 1853 — Helix cygnea Benson, Ann. Mag. Nat. Hist., Ser. 2, Vol. VI., p. 30, Jan. 1. Perth, Swan River, West Australia (Dr. J. F. Bacon). Fig’d. Reeve, Conch. Icon., Vol., VI L, pi. 174, sp. 1182, Oct. 1853. Pig’d. Cox, Aloti. Austr. Land Shells, p. 16, pi. 12, fig. 3. May, 1868. 1930 — Charopa mprea Thiele, Die Fauna Sudwest Austr., Bd. V., lief 8, p. 587. Fremantle, S.W.A. Not Helix cuprea Cox 1868 nt supra. Contrariwise, Benson separated this from H. suhlesta by the more distant ribs, wider umbilicus, colour and larger size from 77. Tuloiclea, Forbes, of the eastern coast of Australia, by its more distant ribs and w'ant of concavity on the upper side. Diam. major 4, minor SVz, axis llj, mill. 42 Toil Ieedale. Tliiek', again, o\'erlookiug tliis species, referred specimens to cuprcu will: the size .'ijo nim. by 2 lum., the protocouch finely sculptured spirally, and the adult ivhorls decussate. The size and the sculpture refer the specimens to this species, rather than to cuprea. Benson’s deseri]ition is here offered for reference: “Shell broadly umbilicate, orbiculately depressed, horny, fui'- nislied with rather rwnotc' olili(iue eostulate radials; spire scarcely a. little convex, sutures excavate, apex planatc; whorls four and one-half, convex, last rounded, aiiertui'c' suljvertical, roundly lunate, peristome acute, Timbilicus perspective.’’ Reeve (igui'ed the type specinum tlu' same year, and added to the above description “decussated, beneath the lens, in the interstices with very minute spiral striae” and showed this character in his illustration. Luinodiscus repens sp. nov. Plate I., fig. 21. Another of Mr. Riflney W. .Jackson’s discoveries at the Bow River and Deep River, this species is the largest yet found of this group, measuring 6 mm. in breadth by 2..') mm, in height. The shell is fawn flamed with red. Apex large, of two whorls, finely spirally lirate, about fifteen lirao counted from above, almost a varix intervening before the adult radial sculp- ture begins. This, on the first adult whorls, consists of about one hundred regular line radials, the interstiees being finely concentrically striate, almost giving the effect of fine beading to the radials. The shell is tlat-top]H‘d, the mouth large, wldle the uml)ilicus is narrow and deep, measuring about one fourth of the basal breadth. Luinodiscus tumidus Odhner 1917. Plate IIP, fig. 2. 1917 — Endoclonta (CJiaropa) tumirla Odhnei', Kungl. Rvensk., Vetensk., Haiull. Bd. 52, No. l(i, |). 72, ])1. 3, figs. 72-74, 8(>i)t. 19. Noonkau- bali, Fitzroy River, Xorth-lVest Australia. Odhuer's description was somewliat brief, hut from a i)araty])e more eletails can he given. Shell discnidal, spire concave, umbilicus moderately wide (narrow, half the width of the last whorl, Odhner) mouth open lunate no teeth in a])erture outer lip Ihin columella straight slight callus joining ]i|is but outer lip descending a little so that it does not reach nl)ove the level of the last whorl, sutures d(>ep, almost excavate, whorls well rounded. Tin- jii)i('al whorls ai’c two, ending in a varix, ami are sculplnred with taint concentric striae which are overridden by distant radial lirac of the same charactm' as the adult sculpture. This consists of regularly closely .spaced ridges, about eighty on the first adult whorl and continuing similarly closely i)acked to the end of the last whorl. The interstices a)'e very finely, regularly, closely, concentrically striate. J’lie ty[)e measured 3.9 mm. in breadth, the heiglit 2 ram., the sjxa-i- men above deseiibed being vei'y slightly snuiller. It is eiii'ious tliat the majority of \V('stern jVnstralian C'haroi)ids so far found have the ai'ieal whorls eoueeiitrically striate, the striae varying in strength, thus while cupreus has the striae well marked almost lirae, snhlestus has the whorls almost smooth, the eoneentric striatioiis being seen only by a strong lens and then they are obscure. Tin' present species has the apex A Review of the Land IMollusca of Western Austraeta. 43 somewhat diffei’ciitly sc-ulptured, the eoncenti’ic striae being overiidden by distant radials, ilie t'orei'unnors of the adult soiilptnre. As the shell is also coneave above, a feature not seen in any southern shell, a new subgenus Corinomala being introdueed with A', tnmida as type. Genus EPINICIUM gen. iiov. Type E. restifer sp. nov. Plate 1., fig'. 36. A very beautiful shell sent by Mr. Glauert from Serpentine attracted by' the bold ridges, recalling those of the much smaller Eastern Kgilomen. Shell small, subdiscoidal, spire a little el(‘vateeciiuens, lollecled by W. W. Froguatt in the Bar- rier Range, all show an engraved umbilicus, a feature not recorded bv the veiw accurate Smith in his descripition of IF. Unna, though he carefully ex- amined the umbilical features. Yet this is a feature of many inembiu’s of this family. The Barrier Range species is smaller, loss elevated, the umbilical ridge less notable and the umbilical cavity is filled with a gum-like matter. It measures 7 by 4.25 mm. against 8.5 by 5 mm. The umbilicus is also quite open, and though narrow would not be termed “very contracted.” A Review of the Land Mollusca of Western Australia. 45 FAMILY HELICARIONIDAE. Thi.s family is based on Tasmanian and Eastern Australian inolluses, which, thoug'h slugliko in form, still retain a thin semi-circular thin shell. Many species are known, extending as far north as Cape York, and cxtra- Ihnital species north of that place have been included. None is known from North-West Austr'ulia, while a single siiecies is included from South-M est Australia, although the group is absent from South Australia. The local form has a shell (pnie unlike that of the Tasmanian and A^ic- torian species, but apparently has a black animal like that of the latter, as Quoy and Gaiuutrd recorded it under the name \'itrina nigra, which they in- troduced on account of the black colour of the animal. Genus LUINARION Iredale 193.3. 1933 — Lttinarion Iredale, Rec. Austr. Mus., Vol. XIX., p. 38, Aug. 2. Haplo- type Helicarion thomsoni Ancey = Vitrina eastanea Pfeiffer. Luinarion was introduced as a subgenus of Helicarion, but it seems to stand further apart, as the shell is so unlike that of other Australian Heli- carionids that it has not been recognised up to the present although described eighty years ago. Shell with the spire a little elevated, smooth, somewhat clepressedly glo- bose, mouth large, open, subcircular, outer lip, sinuate, I'eceding ba.sally, base convex, columella arched, a little reflected. Shell fragile. Luinarion castaneus Pfeiffer 1853. Plate III., fig. 1. 1832 — Vitrina nigra Quoy and Gaimard, AYy. de I’Astroh, ZooL, Vol. II., p. 130, part only (AYestern Port, A^ictoria) and King George’s Sound, AVest Australia. 1853 — Vitrina caslanra Pfeiffer, Mon. Ilelic. Anv., A^ol. III., p. 5, (pref. Alay) Australia. 1854 — Vitrina ca»tanea Pfeiffer, Proc. Zool. Soe. (LoncL), 1852, ]). 56, Alch. 22, 18.54. Australia. 1854 — Vitrin-i castaina Pfeiffer, Syst. Conch. Cab. (Alart. & Chemn.), od. Kuster, Rd. L, Abth. XI., p. 24, pi. 6, figs. 1-4. Australia (in my collection). 1862 — Vitrina eastanea Reeve, Conch. Icon., Vol. XIIL, pi. VI., sp. 37, May (ex AYrreaux in Mu.s. Cum. “chestnut olive”). 1868 — Vitrina eastanea Cox, Mon. Austr. Land Shells, p. 84, pi. XIA^., fig. 11, May, coiiic'd from Reeve (colour all wrong). 188.5 — Helicarion eastaneiis Tryon, Alan. Conch., Ser. II., AYl. I., p. 169, pi. 38, fig. 41, July 3. Had copy of Pfeiffer’s coloration. 1889 — Helicarion tliomsoni Ancey, Le Naturaliste, 1889, p. 19. Geographe Bay, Soufh-AA’est Australia. 189.5 — Helicarion thomsoni Iledley, Proc. Alalac. Soc. (Lond.), Vol. I., p. 260. 1910 — Helicarion thomsoni Hedley Journ. Roy. Soe. AA^est Austr., AYl. L, p. 220 (71 in separate). Quoy Gaimard wrote “Le port du Roi Georges nous a fourni des individus plus petits, (than the Victorian species), vivant sous les arbres. 46 Tom Ieedalb. loin do 1 ('an douct*. ’ ^Yltliougli Cox rc'cordod this noto in the ('ai'li( 3 st cata- logue of our Laud Slndls l)y himself in 1864, Smith in 1894 wrote “No species of Vitnna, Tlelictirion .... are yet known from this region.” This was corrected by Medley, who pointed out Ancey’s description of 1889, but also o\ erlooked tifuoy & Gaimard’s notice. A)iparently everyone has omitted notice of Pfcili.er’s CKstanea^ which, published in 18,53, is undoubtedly the present species. FAMILY CHLORITIDAE. This family comprises many shells covered with a periostraeum bearing hairs but, through lumping, species without such n covering are commonly included. TTu' typical Chloritis is a large llattened umbilicate shell with a recurved outer lip, and the apex granulose. It is essentially of northern origin, and species oc(air throughout East Australia, as far south as Victoria and along' the north coast. TIorvcvf'r many speems ha,v(' Ijeen referred to this group, whose claims are very doubtful and a redistribution is sorely needed. Gude reviewed the series, and went so far as to group all the Australian s])ecies, whatever their form, under AustrocUloritis, a somewhat peculiar conclusion. Oeiuis DAMOCHLOEA hedale 1938. li)3S^I)amncJtIora Iredale, Austr. Zool. Yol. IX., p. 97, Nov. 30. Orthotype Helix millepunctata Smith. This generic name was iniroduced for some North-Western species, which Smith placed under Hclir, using as a suhgenns Chloriiis, the type species being described as minutely punctate and with the form of delensertiana, while revtUiiJiniiii was subgrauulos(', clothed with a thin scabroirs epidermis and the form unlike. It is pi'obable that these are not at all closely related, and in order to stress this point the snbgeneric name Perochlora is intro- duced, the apertural characters of the tyiic, rectilabnim, reading quite dif- ferently. Damochlora millepunctata Smith 1894. 1894 — Helix {Chloritw) miJJe punctata Smith, Proc. Tlalac. Soe. (Loud.), I'ol. I., j>. 88, pi. VII., tig. 11, June, llaudin Island, North-West Australia. Smith's description reads: “Shell orbicular, widely umbilicate, thin, horny, snlqx'llncid ; whorls five, slowly increasing, everywhere minutely juuictate, coin'ex, sutures jirofoundly impressed, .sculptured with thin ()bli((ue]y arcuate growth-striae, la.st whorl rounded at the iieripherj', scarcely descend- ing in IT'onl : ai»erlure obli(ine and broadly Innate, si)ire very .short, ralher ob- tuse to the apex: peristome thin, very little thickened, narrowly expanded and r('flected; coluniellar edge very obli(|ne, a litlle arcuate, dilated a little above where it joins the whorl. Diam, maj. 16, min. 13 mm.; alt. 8.5 mm. Aper- ture 6 high, 7 broad. Hah. — Baudin Island, N.W. Australia. This species has much the general aspect of H. deleasertiana. It is, however, flatter, and, of course, quite different in .sculpture. ” Damochlora cassiniensis Smith 1894. 1894 — Heli.r (GJiloidtis) millepunctata var. raiisiniensis Smith. Proc. Malac. Soe. (Lond.), Vol. L, p. 88, pi. VII., fig. 12, June. Cassini Island, North-West Australia. Smith briefly described this; “Shell smaller than the typical form, nm- bilicns a little narrower, aperture more contracted, peristome a little more A Review op the Land ]\[ollusca op Western Axtstralia. 47 thickened. Diain. maj. 14.5, min. 10.5 mm.; alt. 7 mm. Aiierturc 5 high, 5 I -oad. Hab. Cassini Island, N.W. Australia. Although smaller and differing f: om the type in the i)oints referred to, it seems advisable to consider this form as a variety, rather than as a distinct species.” Damochlora rectilabrum Smith 1894. 1894 — lleli.r (CJiIoritis) rectilahnim Smith, Proc. Alalac. Soc. (Lond.), A ol. I., p. 88, pi. VII., fig. 14, June. Parry Harbour, X.W. Australia. The descriiition given by Smith reads: “Shell orbicular, dc|iressed, um- bilicate, pallid bi'ownish, subgranulate, clothed with a thin scabrous epidermis; spire a little elate, obtuse to the aiiex; whorls five slowly increasing, convex, suture.s profoundly impn'ssed, last whorl rounded at the peiiplu'ry, very slightly keeled, de.scending slightly in front; aporturi' almost horizontal, lunate; peristome thin, narrowly expanded and reflected, margins coming together, the basal edge straightened a little towards the coluuiellar dilation. Diam. maj. 13, min. 11 mm, alt. (1 nun. Aiperture 4 high, 5.5 broad. Hab. Parry Harbour, N.W. Australia. A dead siiecimen devoid of epidermis does not appear nearly so granular as fresh living examples, although traces of the granulation, esiieeially ui)on the apical whorls, are di.scernible.” Cemis KIMBOEAGA Tredale 193:!. 1933 — Kimhcmtrjd Iredale, Rec. Aust. Mus., Yol. XIX., p. 50, Aug. 2. Ortho- type Chlnrith micromphala (hide. Gude named a large series of shells under CMoritis many of which ap- pear to be not closely related to the genotype of CMoritis, while some, such as the present series, seem to have very little affinity. Shell globose, thin, elevated si)ire, mouth large, subcircular, outer lip a little expanded, thin, columella broadly reflected, umbilicus deep, open, apex smooth, shilling, willi faint growth radials only. Xhi yxu'iosi racum discernible and sculpture of fine growth radials crossed by fine concentric lines. The apex is large, a little eccentric, a half whorl smooth, another whorl finely striate radially, and three adnlt whorls only. The related, .superficially, forms Glorengenia and FargJogenia have the apex small and regularly coiled, and the flr.st whorl and a half are succeeded by five adult whorls. Kimboraga micromphala Gude 1907. Plate IIP, fig. 12. 1907 — CMoritis micromphala Gude, Proc. Malac. Roc. (Lond.), Vol. VII., p. 231, pi. XXI., fig. 6, April 3. Barrier Range, Xorth-West Aus- tralia. The generic details above given cover the species; tlie coloration is (lale fawnish, and the noun measures 28 mm. in breadth and 18 mm. in height, tho aperture being about 11 mm. high and broad, the outer lip descending appreciably in front. [Parglogenia forrestiana Angas 1875. 1875 — Helix forrestiana Angas, Proc. Zool. Soc. (Lond.), 1875, p. 389, pi. XIW., figs. 3, 3a. North-West Australia. The locality “North-West Australia,” almost certainly refers here to an extra-limital locality, as, from the description, the shell is very close to, if not identical with pelodes Pfeiffer pseudoprunum Pilshry. The last named ■48 Tom Iredale. was also close I'i bod from novth-wostorii Australia, by wliieli Filsbry iuteiided Port Darwin, whioh is politieally in the Xorthern Territory. An earlier name still inav bo siihgriinosa Le (iiiillou, and tbo .slioll has also boon known as prmmm Fornsaae, whiob has boon shown to be very ditTorent. Misled by tlie ineorroft assooiatioti, llodloy eonfu.sod the shell, Kimhnragn mirrow phala, and thn.s inoluded forreationa in his W.A. list.] Genas TOKEESITRACHIA I redale l!t33. 1933 — Torresitrachia Iredale, Rec. Austr. ilus., Vol. XIX,, p. 55, Aug. 2. Orthotypo eiuleavonrensis Brazier. Shell snbdisooidal, sjriro a little obtusely elevated, citreous, whorls rounded, sutures impressed, periphery rounded, mouth aubeireular, open, outer lip thin, reflected all round, a little thickened basally, columella almost straight, a little relleeted, umbilicus narrow, deej), ojien, showing coiling and not obscured by colnmellar reflection. Coheration greenish white, unieolor. Ap)e.\ smooth, adult sculpture close radial ribbing almost ridges above, base smooth save for growth striae. This Xorth Queensland style of shell appears to travel along the north coast into Xorth-West Australia. Torresitrachia bathurstensis Smith 1894. Plate 3, Fig. 2. 1894 — Hfili.f {Trachia) Smith, Proe. Malac. Soc. (Loud.), Yol. I., p. 93, pi. VII., tig. 2(1, June. Heywood Island and Bathurst Island, King Sound, X.IF. Australia. A curious complication is seen in the tact that the specific name is taken from one island, and the type shell came from the other; these islands lie at each side of the Sound and the species are probably distinct on each island. Curiously enough the specimen before me, collected by IMr. J. J. Walker, is a'jtaratype, but it has no definite locality. This specimen is very like the species of Tnrrenifrachia but can l)e se])ar- ated by the liner, closer, and more regular character of the costulation. The tyj)e from lieywood Island measured 15.5 mm. in breadth by the height omitted, while the Bathurst Island shell was 13.5 min. in breadth, again no height being given. The paratype abovementioned and bere figured measures 14.5 mm. in breadth and is 8.5 mm. in height. The ribbing on the base is stronger than on (he base of the (Queensland shells, but is not as strong as on the uppei- surface nor is it regular. Torresiti'achia monticola s]). nov. Plate IIP, fig. 13. The species of Torresit rachia are found on the islands of Torres Strait and along the north coast of (Queensland. The occurrence of similar shells on the islands otf the Xorth- \Ve«t Coast was unexpected, but not so surprising as the recognition of a form among the shells collected by W. W. Froggatt in the Barrier (Xa])ier) Range, many miles inland. This siiecies is flattened, snbdisooidal, wborls rounded, openly narrowly umbilicate, mouth subcircular, open, lips thin I’ellected, columella arcuate. The apex is smooth, a subvarix sej)ai’ating this from the adult sculpture which consists of deep well separated very regular costulations, which con- tinue onto the base where how'cver they are less marked. Coloration trans- A Eeview op the Laxd Mollusca of Westers Australia. 49 luecnt and glassy. The umbilicus is narrow, deep, only about one eighth the width of the base yet the previous whorling may be seen therein. The outer lip is reflected all round, though thin, and the columella is little thickened. Breadth 1.5..') mm.; height 9.5 mm. The costulations are a little stronger than in the coastal shell, the spire a little more elevated and the umbilicus a little wider. Genus BAUDINELLA Thiele 1931. .19,31 — Buudhiella Thiele, Handbuch syst. Weicht., Vol. I., p. (185. Ortho- type, Hdi.r haiidinensiH Smith. 1933 — Gonohnudinia Iredale, Hec. xlusfr. Mus., Vol. XIX, p. 55, Aug. 2. Ortholype, lleliic- baudinands Smith. Thiele’s Handbuch was not available when I introduced Gonohaudinia^ and Thiele’s subordination of his subgenus Bandinella to Augasella is difficult to understand. Small, stout, widely umbilicate with contracted mouth, it resembles no other Australian shell. Shell diseoidal, spire flattened, whorls few, last descending a little, mouth small, broader than high, outer li[) thickened and reflected, with a ditch behind, umbilicus wide and open, nearly half the width of the base. The apex is granulosc of about two whorls, the adult sculpture developing without any varix intervening: this consists of stout radial ribs with wide interspaces, the ribs running across the upper surface in a .sigmoid manner, then with an even curve into the umbilical cavity where twenty-one may be counted on the last whorl. Along the upirer part of the last whorl there is a deep depression which causes the ecceutrieitj' in the lib development, and this deioression appears in the outer lip as a tooth. Baudinella baudinensis Smith 1893. Plate III., tig. 9. 1893 — Helix {Gonostoma) httudlnends Smith, The Conchologist, Vol. II., p. 97, fig. in text, lleh. 25. Baudin I., Xorth-West Australia (J. J. Walker). The generic characters are, at ])resent, sufficient to enable the recognition of this bizarre little shell, which measures (1 mm. in breadth by 3 mm. in height. There is a stout epiphragm tin the aperture of the shells collected alive. Genus SETOBAUDINIA Tredalo 1933. 19;i3 — Setohaudinia Iredale, Rec. Austr. ilus., Vol. XIX., p. 55, Aug. 2. Orthotype Helix coUInjjii Smith. The detailed specilie account following shows that this differs from the preceding genus in the essential distinction that the shell is covered with a 2 iilose jocriostracum. Otherwise it is larger, lacks the chai’acteristie sculiiture, and has a more open mouth. It is almost certain that it is not closely related. Setohaudinia collingii Smith 1893. 1893 — Helix {Gonostoma) collingii Smith, The Conchologist, Vol. II., p. 98, fig. in text, ilch. 25. Baudin I., Xorth-West Australia (J. J. Walkei') . As no shells are available in this case, Smith’s account reads: “Shell flatly diseoidal, with the spire only very little raised above the body- whorl, rather oiienly umbilicated, light browm above, and jiale beneath. When in 50 Ton Ihedale. fresh (“ondition the surfae<' is eoveied with a thin, shortly ])ilosc e))idermis. Worn shells exhibit iiuiumeTablc minute ])unctures showing: where the short delieate setae have been. Whorls TVo, regularly and rather slowly increasing, moderately convex, and separated by a deep sutural Hue, besides tlu‘ punc- tures, exhibiting tine lines of growth; last whorl I'ouuded at the jieriphery, only very leebly dellexed close to the aperture, and exhibiting a slight depres- sion above about the middle of the uijper inargi)! of the ]ieristome. Aperture somewhat triangular in outline, but with rounded angles, almost horizontal in position; peristome narrowly relleetcd above, more broadly expanded along the basal margin, espieeially over the umbilicus. A conspicuous tubercle er prominence occurs on the innin- edge of the basal margin, and a less pro- nounced one within the upper margin correspiondiiig to the slight depression upon the outer surface of the whorl. Breadth 10 mm.; height 4 mm.” Genus WESTEALTRACHIA tredale 1933. 1938 — Westraltracliid Iredale, Hoc. Austr. IVIus., Vol. XIX, ]). 55, xAug. 2. Orthotype Trachitt froggatti Ancey. This geiiei'ie name was intrf)duced for a uundoer of species, which Ancey had, with great doubt, referred to Tracliia. The species are flattened helicoids with low spires and keeled or subkecled periphery, narrow umbilicus and trans- verse mouth sometimes basally flattened and indi.stinetly toothed. Although generally smooth and shining, the tj'pe is granulose and dull, and it may bo that the smooth sjiecies should constitute a separate snhgenns, which may be called Zygotrachia the species, 11'. alterna. being named as type. Westraltrachia froggatti Ancey 1898. Plate 111., fig. 10. 1898 — Tracliia froggatti Ancey, Proc. Limn Soc. X.S.W., Vol. XXII., p. 774, pi. XXXA'I., lig. 2, June 4. Oscar Hange, 100 miles inland from Derby, North-West Australia. (W. W. Proggatl). Slndl small, sidjconical, deiiressed, stout, spire a little devated, wdiorls | flattened, sutures little impressed, last whorl acutely keeled, mouth oblique, outer lip expanded and a little recurved. A))ex radially finely subgranose, adult sculpture flattened coarse granulation finer on the base. Westraltrachia derbyi Cox 1892. Plate III., fig. 6. 1892 — TTrlix [lladra) derhyi Cox, Proc. Linn. Roc. N.R.W., Rev. 2, A'ol. VI., p. 56(1, pi. XX.. figs. 4, 5, I\Iay 2:1, Derby District, North-West Aiisiralia (W. W. Proggatt). 1894 — ihdix (Tracliia) dcrbyaiia Rmith, Proc. Malac. Roc. (Lond.), Vol. I., ]). 92, id. VTI., fig. 19. .tune. Burner (error pro Barrier) Ranges, Derby, X(n-lli-West Australia. Shell .small, snbdiscoidnl, spire a little elevated, whorls little rounded, sutures impressed, urabilieatc, umbilicus narrow, deep, columella reflected, mouth open, outer lip. a little reflected with a slight snlmodulation on base. Coloration whitish marbled with brown flames above, a post peripheral brown band, rest of base dirty white, shell shining. Apex smooth, adult sculpture faint growth lines only. Breadth 11 mm.; height 6 mm. Collected on the Barrier (— Nainer) Range by W. W. Proggatt. A Review op the Land Moelusca of West’brn Australia. 51 Westraltracliia orthocheila Ancey 1898. Plate TIT., fig'. 8. 1898 — Tracliia orthocheila Anccy, Pi'oe. Linn. Soe. N.S.W.^ Yol. XXII., p. 774, ]il. XXXVT., fig'. 4, June 4. Oseai' Range, 100 miles inland from Derby, North-West Australia (W. W. Froggatt). Shell a litlle larger than the preeeding, .subdiseoidal, spire scarcely elevated, whorls .scarcely rounded, sutures impressed, umliilicus narrow, deep, columellar i-dleclion small, outer lip thin a, little rellected, mouth open with no basal subnodulation nor flattening. Coloration whitish marlfled above, with a brown band below the iieriphery, the rest of the base white, shell shining. Apex smooth, and only tine growth lines seen on the adult whorls. Breadth 13-14 mm., height 61/4-7 mm. The Oscar Range is twenty miles distant from the south end of the Napier (or Barrier Range), and its shell fauna appears to be different although this species is very close to derhyi. Westraltracliia alterna sp. nov. Plate III., fig. 17. Shell large for this genus, subdiseoidal, s)ure a little elevated, whorls slightly rounded, sutures imiiressed, last whorl descending a little, .subkceled, mouth flattened hoilzontally. The apical whorls, two in number, arc smooth, the adult four and a half sculplured only with regular striae, which become more distant on the last whorl. Outer lip I'diocted, base of lip flattened and almost nodulo.se, columella broad, angularly joining the base of the lip, reflected. Umbilicus narrow, ojien, half hidden by the reflection of the columella. Coloration horny, banded with dull browui above and below the periphery, base almost white. Breadth 17 mm., height 9 mm. Collected on the Barrier Range (i.e., Napier) by W. W. Froggatt. Westraltrachia increta sp. nov. Plate III., fig. 16. Shell large, not as largo as preceding, conical, spire elevated, whorls rounded, last subkeeled a little descending at the aperture, the mouth flat- tened horizontally. Colour pale brownish whit(> banded above and below the periphery with broad browm bands, the base dirty wdiite. The two apical whorls are smooth, the adult four and a half have only subobsolete striae vanishing on the base. The umbilicus narrow, almost concealed by reflected columella, which is broad and almost angularly meets the ba.se which is flat- tened and almost subnodulose. Breadth 1.5.5 mm., height 9.5 mm. Collected by W. W. Froggatt on the Barrier Napier) Range. Genus QUISTRACHIA geii. Dov. Type Trachia monogramma Ancey. Shell thin, subdiseoidal, s]ure a little elevated, wdiorls rounded, umbilicus narrow', persiiective, outer liji thin, a little reflected, subcircular open mouth, columella rather broadly reflected. Although hitherto classed with the preceding this shell is unlike any of the Westraltrachia in texture and form, and is nothing like Rhagada. Toil IliEDALE. Quictracaia monogramma Ancey 1898, Plate III., fig. 20. 1898 — Tracliia monogramma Aneoy, Proe. Linn. Soc. N.S.W., Vol. XXII., p. 775, pi. XXXVI., fig. 9, .June. Oscar Range, N.W.A. (W. W. Proggatt). In addifion to the characters mentioned above, the shell is horny brown with a darker brown on peripheral band. The apical whorls are .smooth as is the rest of the shell save for indistinct growth lines, the whole surface slightly shining (“oleoso,” Ancey wrote). The figured shell measures 17 mm. in breadth by 10 mm. in height, while the type, a unicum, measured 15.5 by 8.5 mm. [Helix australis Menke 1848. 1848 — Helix (ReliceUii) australifi Menke, Moll. Xov. Holl. Spec., p. 6 (Apl. 11), among limestone rocks at Mount Kliza near the Swan River. 1858 — Helix australis Pfeiffer, Syst. Conch. Cab. (Martini & Chemnitz), ed. Kuster, Bd. I., Abtli 12, pt. 2, ]>. 276, pi. 123, figs. 7, 8 (after June, 1852). Specimen from Menke Collection figured. 1852 — Helix australis Reeve, Couch. Icon., Vol. All., pi. 131, sp. 808, Oct. Swan River, New Holland. )Mus. Cuming. This species has caused a lot of trouble, yet it seems certain that it is merely a South African shell mixed with the IVestern Australian collection of Dr. L. Preiss. Why this has remained on the list, when this imssibility was known is curious. Reeve noted, “.X small striiH'd si)ecies of European aspect,” while, of the South African slull he wrote, “Closely allied to a well- knowni European form.” Menke himself compared it with a South African shell, and Benson recorded that •‘this South African representative could not be refound by Dr. Bacon.” Smith questioned the identity of Reeve'S specimen wnth Menke’s species, and ])laced the shell under Xerophila, a European subgenus. So it may be omitted from Western xiustralian lists as it is unlike any local shell in size aiid form.] FAMILY XANTHOMELONTIDAE. The most notable mollusc in North Australia is a large thick globose shell, for which many years ago the descriptive name, Xaiitlinmelon, was in- troduced. It is common about Port Darwin, and goes eastward along the coast round the Cape York Peninsula and down toAvards Moreton Bay. As far as yet is known this form does not occur in Western Australia, an extraordinary result from superficial knowledge. In llu' interior of Australia, apparently degenerate relations live, and these penetrate into Western Aus- tralia in the mid and southern areas. This series, named ttinumeloii , ajpAears to be characteristic of the Centraliaii Area ; different foians, sjAecies or genera being develo])ed on each Range. Gciuis SINUMELON Ircdale 1930. 1930 — Sinumelon Ircdale, Viet. Naturalist, XMl. XLVIL, p. 120, Xov. Ilaplo- type Helix nullarboriea Tate. 1932 — Nutnbadistes Cotton and Godfrey, South Austr. Naturalist, Vol. XIII., lip. 169-170, Aug.-Septeniber 30. Orthotype Helix bitaeniata Cox flindersi Angas. These desert living Xanthomelontids, ranging all through the interior in suitable places, enter into southern Western xVustralia, and ai)parently all the A Review of the Land Mollusca of Western Australia. 53 subgroups appear there. Sliells small for the family, .subglobose. more or less openly umbilieate, no definite seuliiture, longitudinal growth ridges over- ridden by granulation, mouth subcircular, open, the columella thickened and reflected. Sinumelon nullarboricum Tate 1879. Plate 111., lig. 19. 1879 — Helix imMarbanca Tate, Trans. Proe. Phil. Hoc. Adelaide, South Austr., 1878-9, p. 133, pi. VI., ff. 1 a-b, ex p. 126, worn niid. Bunda Plateau, Nullarbor Plain, South Australia. Shells from Eucla agree in detail with ])araty[)es of Tate’s species. These are subglobose, dirty white, sculjitured with rough radials, spire a little elevated, umbilicus narrow, columella thickened, outer liji thickened. The two apical whorls are .smooth and shining, but show under a lens very fine radials towards the suture. The rude irregular radials show dents on the last whorl, which ajipear conci'iitrically as if they were broken rough incised lines, but line incised concentric lines may bo seen on the base of some specimens subordinate to the denting. The columella is reflected, almost con- cealing umbilicus, which is encircled by a ridge. Height 14 mm., breadth 16 mm. Sinumelon datum sp. nov. Plate 111., lig. 18. 1895 — Uelix angasiana Hedley, Proe. Malac. Soc. (Lond.) Vol. I., ji. 260, July. Eucla, Western Australia. Not Sinumelon godfreiji Iredale, Rec. Austr. Mus., Vol. XIX, p. 52, Aug. 2, 1933, new name for Helix angasiana Pfeiffer, Journ. de Conch., Vol. X., p. 228, pi. X., fig. 2, July 1, 1862: near Lake Torrens, South Australia. Shell subglobose, broader than nullarboricum, with the spire less elevated, the sutures deeper, the umbilicus wider and more open, the columella more curved and refbuled, the outer lij) reflected but not thickened. Colora- tion dirty white. The one and a half a))ical whorls have more prominent radial sculpture easily seen with a lens, the adult sculpture being rough radials with granu- lose subordinate sculpture, the grains lengthening on the last whorl into broken radials. The dents are less marked and even more concentrically .arranged. Breadth, 19 mm.; height 14 mm. Type from Eucla. Shell from kladura larger and broader but obviousl.v conspecific. Sinumelon kalgum sp. nov. Plate ITT., fig. 25. From Hannans, Kalgoorlie and Lake Kalgoorlie dead shells (which were probably coloured in life) appear to be relations of Ihe fodinale series. They arc stouter than datum, and have not the deep sutures of that species, and the sculpture differs. The shell is subglobose, stout, spire short, sutures impressed but not subcanaliculate as appear those of datum. The apical whorls are not so markedl.v striate, while the radials are finer and the deli- cate granulation present never becomes radial, and there appear to be no ■signs of coneentrie striae nor any denting. The mouth is more vertical and 54 Toil Iredale. more circular, while the umbilicus is more hidden though still open: the | columella thickened with the outer lip reflected. Breadth, 20 mm.; height, 17.5 mm. Bednall's record of focVnalis (Trans. Boy. Soc., South Austr., Vol. XVI., p. 63, Dec. 1892), from between Victoria Spring and Fraser Range 1 may refer to this species. Sinumelon lennum sp. nov. Plate III., fig. 21. I 1892— 7/e/(.r ((iiihijias) perinfl(ila Bediiall, Trans. Roy. Soc. South Austr., Vol. XVI., p. 62, December. Cnvatiagh Range. About three miles south of Camp 58, Victoria Desert.. Betwe(>n Fias:'r Range and Yilgarn noldti<'lds. Western Australia. [ Xot lleU.r pcriulUOn PfeilTer, Proc. Zool. Soc. (Bond.), 1863, p. 528, i Apl. 20, 1864, .MacDotmell Ranges, Central Australia. i Representatives of the perivllata series occur throughout the interior of South- AVest Australia varying in size and form according to locality. i Shell sul)globose, siiire short and conical, whorls rounded, sutures im- pressed, umbilicus ^'('^y narrow hidden by I’efloction of columella, mouth large and subcircular. Coloration green. Apex finely granulated, adult whorls 4 V 2 , sculptured with faint radial growthlines, rvitli irregular granules cover- j ing'the whole surface tending to lengthen into beaded radials. Breadth, 24 mm.; height, 21 mm. (type from Boulder). Specimens from Aladura are broader with the same height, the spire being shoi'ter and the body whorl more sw’ollen; those measure 26 mm. in Imeadth by 21 mm. in height, and may l)e called S. 1. mutuum subsp. nov. Sinumelon vagente sp. nov. Plate ITT., fig. 24. From Alt. Singleton, inland from Geraldton, comes a relation of perinjlata, more elevated and less inflated than lenmim, with a stronger granose sculiiture. Sludl subglobose, spire short, whorls rounded, sutures impressed, umbilicus very narrow almost concealed by expansion of columella, mouth large, round. out(‘r lip thin. Coloration brownish green. Apex finely radially granulosc', adult wliorls four, sculptured with fine irre.gular growth radials w’hieh are. overridden by a fine granulation. Columella strongly refiected and united to the outer lip by a thin callus. Breadth, 21 mm.; height, 19 mm. Genua PLEUKOXIA Anee.v 1887. lj ;87 — Pletiro.rid Aiicey. Conch. Evchange, Vol. II.. pt. 3, p. 38, Septem- ber, new name for 2804 Ancp^sella Angas, Proc. Zool. Soc. (Bond.). 1863, 521, A]fl. -0, 1864, ex A.' Adams AIR. Haidotype Belir cprtopleiira Pfeiffer. Xot Anr/afiiplla Crosse, .Toiirn. de Conch., Vol. XII., p. 50, footnote, ,Ian. 1, 1864. These shells, referred on account of anatomical details, to the family Xanthomelontidae, are very nnlike typical members of that family in every conchological feature. A Kevi?]w of the Laxd Moi^lusca of Western Australia. 00 The type has the spire depressed, the shell rtiscoidal, the umbilicus very wide and open, the mouth circular, lip a little rellected; the apex is granosely radiate, the radial sculi ture developing- into ribs, the grains continuing as a .subordinate feature. The south-western species, represented by olifjopleura, have the apex smooth, siiire elevated, shell subdiscoidal, sometimes sub- globose, the mouth circular, with the lijis continuous, the umbilicus narrow. The sculpture is similar but coarser, and this series is separated as a new subgenus, Angasietta. The Gantheaume Hay specie.s, P. ahstaiiHj however, has the apex coarsely granular, spire elevated, the umbilicus moderately wide, the mouth circular, lip.s continuous, expanded rather broadly and constricted behind, and is therefore snbg'enerically differentiated as Gantomia nov. Pleuroxia polypleura Tate 1899. Plate III., fig. 27. 1899 — Angasella politplciira Tate, Trans. Roy. Soc. South Austr., Vol. XXII 1., p. 246, pi. VI., figs. 2a-c, Doe. Bunda Plateau, Great Aus- tralian Bight, South Australia. 1879 — Helix eyrtophura Tate, Trans. Proc. Phil. Soc. Adelaide, South Austr., 1878-9, |). 120, not of Pfeiffer, 1862. 1895 — Ilelii- tyvlopleutut Tledley, Proc. Malac. Soc. (Bond.), Yol. I., p. 260, July, not of Pfeiffei', Journ. de Conch., ^'ol. X., p. 227, pi. X., fig. 4, July 1, 1862. Shell subde])ressed, flattened, Avidely umbilicate, with about sixty sig- moid threadlike ribs, the inter.sjiaces coarsely granular, the apex large and smooth, breadth 18 mm., height 10.5 mm. Specimens from the road between Madura and Muiidrabilla agree in size and sculplure and one is here figured. The umbilicu,s is better described as narrow and open, not much more than one-fifth the width of the base, while the large open .subcircular mouth has the thin lip reflected, and the lips are connected by a strong body callu.s. Shells collected at XeAvman Rocks are similar in sha]H‘ and sculpture but much smaller with the umbilicus a little wider, the granulation finer. The largest measures 13.5 mm. in lireadth and 8 mm. in height, and is subspeci- fically named Pleuroxia polypleura elflna nov. Pleuroxia commenta sp. nov. Plato HI., fig. 20. Specimens collected by Mr. Charh's Barrett, the tamed Victorian natural- ist and writer, on the Nullnrbor Plains were recorded as P. polypleura. Bet- ter knowledge allows their descrii)tion as distinct, and the exact locality proves to be Hamilton Tablelands, inside the West Australian boundary. Shell small, subde])ressed, spire a little elevated, sutures impressed, whorls rounded, last Avhovl flattened above and tlien rounded, a little descend- ing in front, mouth large, subeircular, outer lip a little rcdlectod, umbilicus narrow, deep, open. The eoloration of the living shells is a dirty broA\mish Avhite. Very similar in general appearance and size to P.p. elfina, but a little more depressed, and Avith mueli coarser sculpture. The ribs are much stronger, more distant, and the granulation almost obsolete, the ribs num- bering forty to forty-five. The mouth and umbilical features are very simi- lar to those of the jireceding. The shell measures 13 mm. in breadth by 7 mm. in height. 56 Toil Iredale. Pleuroxia oligopleura Tate 1894. Plate III., %. 28. 1894— JTatlra oUgopleura Tate, Trans. Roy. Soc. South Aiistr., Vol. XVIII., p. 193, Nov. Eyre’s Sand Patch, 160 miles west from Enda, West Australia (Adcock). IHm—AngaseUa oUgopleura Tate, Rep. Horn. Sci. Exped. Centr. Anstr., pt. 11., Zool., ]). 219, pi. XIX., fifj. 39, Eeb. “Flinders’ Range, South Australia,” error only, through interchange of localities with H. trilpenenms only. “Similar to II. cry! o pleura (sic) hut the i)lications sharper, higher, and about one-third less in number (35 to 40) ; the outer lip is thin, and the whorl is more constricted behind it. Diameters, 14.15 and 12; height, 8; height of aperture 6 mm.” Topotypes agreed with this diagnosis but shells from 70 Mile Tank ea.st of Ralladonia are a little larger and flatter, and jirobably belong to the same subspecies that occurs on the Hamilton Tableland and at Cardanmnbi, west of Eyre. This subspecie.s is altogether larger, more de- pressed the last whorl flattened above and rounded below, the iieriphery sub- keeled, and the rib sculpture mucli more pronounced and the ribs only about thii’ty in number, interstitial granulation obsolete. The coloralion of the living shell is brownish white, and the type measures 18 mm. in breadth and 8.5 mm. in height, the subsjieeies being named Pleuroxia cligopleura numba nov. Pleuroxia gascoynensis Smith 1894. 1894 — Helix (Trachia) gavcoijnensis Smith, Proe. Malac. Soc. (Loud.), Vol. 1., p. 93, pi. \'I1., fig. 13, June. Gascoym- District, Western Aus- tralia (II. P. Woodward). “Shell depr(>ssed, orbiculai', broadly and openly umbilicated, whitish, rather solid; whorls four, convex, sutures deeply impressed, closely and minutely granulated, scul])lured with oblique rugae or irregular rugose plicae, the two a|)ical smoother, the last whorl rounded or subangulate at the peri- phery, descending- consjiicuously in front, the underside rugose; spire flat, apex obtuse, a liltl(‘ eh’vated; apertur(> subhoriy.ontal, rounded; [leristome continuous, apjiressed to the last whorl, narrowly (>xpanded, the lower margin broadly dilated. Diam. maj. 12.5, min. 10 mm.; alt. 5.5 mm. Aiierture 4 high, 4.5 broad. It is similar in form to II. cyrtopleura , Pfr., from South Aus- tralia, but differs in size and sculjiture. It is smaller, flatter, and has the peri- stome more decidedly continuous. Rather a ;)owertul lens is necessary to observe the fine gramdation which covers the entire surface.” Pleuroxia abstans sp. nov. Plate III., fig. 22. A series labelled “On chalk, Murchison House, Gantlieaume Bay,” con- sists of dead shells, smaller than the preceding, the largest being only 10 mm. in breadth and barely 5 mm. in height, and having the apex coarsely granu- lose. Shell small, subdiscoidal, spire, scarcely elevated, .sutures deep, whorls a little rounded, last whorl descending in front, umbilicate, umbilicus wide, open, mouth oblique, subcircular, chalky white. Apex half whorl strongly granu- lated, not differentiated Horn four adult whorls, the granules massing to form radial ribs. On the last whorl these ribs are rvell marked closely set ridges. A Eeview of the Land Mollusca of Western Australia. 57 somewhat irregular in sparing, the intorstiee.s somewhat obscurely very finely grained. The ridges continue into the umbilicus which is open and about one- third the diameter of the base. The mouth is subcircular, the outer lip thick- ened and reflected, an antelabial ditch jirosent; the columella is arched, re- flected a little, and connects with the outer liji by a strong callus. [Pleuroxia radiata Hedley 1905. 1905 — Xunthomeloii radiatum Iledlev, Trans. Kov. Soc. South Austr., Yol. XXTX., p. 163, pi. XXX., figs. 4, 5, 6, December. Mt. Davies, Tom- kinson Range, and Yusgrave Range.s, t'entral Australia. The Tomkinson Range (‘xtends across I he border into Western Australia, but Mount Davies is on the Kouth Australian side. The Mann Range is also only a short distance on the wrong side of the West Australian border, and in the same iraper, Hedley recorded Thersites basedoivi, desc'ribed from the Mus- gravo Ranges, and now placed in the family Hadridae, and Xmithomelon asperrimum, now regarded as Glijptorbagada, but with a distinct subgenus Eximicyrhugadu , was described from the Mann Ranges. These records suggest many novelties from the multitude of ranges indicated on the map running south-west from the Tomkinson Range to Mt. Margaret, and thence south to Kalgoorlie.] FAMILY RHAGADIDAE. The notabh' Caurine faunula is characterised by many species of mol- lusca of a solid chalky a])pearance, many with coloured bands, of normal helicoid aspect, quite unlike those from the rest of Australia. The typical RJiagada is a small solid striped Helicoid with the umbilieu;-: closed by appression of th(‘ columella; the undiilicus is narrow and open in the .juvenile but rarely opi'u in the adult, sometimes showing a chink but com- monly coni])lelely closed. The only sculj)ture is growth lines and the mouth is roundly' Innate', a subbasal tuberdi' '|)resent. Many' species ai’e larger, .some more globose, and some with I'egular sculpture aboi’e but the facies of all resembles each oilier. This family is utilised lentatively to include some sfiecies which are not solid, coming from Ihis Dainpierian Sub-Area. It is strange that so far no similar shells have been found in the Xorthern Terri- tory. The earlie.sf Rhagadoid shells w'ere collected by Ptu'cn and Lesueur, and later the collectors with Stokes secured similar shells. Fifty years ago that fine collector and entomologist W^. W. Froggatt collected many- sjiecimens in the Barrier and Oscar Ranges. A little later an e.xtraordinary addition to our knowledge was made by another entomologist, J. J. Walker, whose official post was engineer on the surviw'ing vessel Penguin. A|)parently paiMies were landed on \arious islets betwemi Broome and Darwin, and this industrious collector in his search for insects secured a large number of curious land shells. Owing to the donation of this collection to the Brilish Museum the curator of molluscs (R. A. Smith) examined them and found so much novelty that he prepared a List of the Land Shells of Western Australia, the onl'v monographic attempt made. Twenty years ago Dr. H. Basedow explored the Na|)ier Range and again many land shells were procured. Although the gen- eral locality “Xapier Range” was the same for Froggatt’s and Basedow’s col- lections Uiere was .so much discrepancy in the material that it became impera- tive to investigate their collecting grounds a,s otherwise recorded. The localities visited by W. W. Froggatt are given in the Proc. Linn. Soc., N.S.W., Ser. 2, Yol. lY., pt. 2, p. 199, Sept. 20, 1889, as follows:— (1) 58 TO.U IlfEDALE. Ironstone Ridge, 25 miles South-East of Yeeda Station, Eitzroy River; (2)’ ilount iMarniion; Ci) iloniit Xortli Creek, Napier Range; (4) Lennard River Gorge, Xapi(M' K’angc; (5) Harrier ttange Homestead, Xa[)ier Range; and (6) Oscar Range, north-east side. On tile other hand an account of Has(‘dow’s E.\]iedition has a’ppeared in the Transactions of tlie Royal GeogT-iiiiliical Society of Austrahisia, South Australian liranch, Vol. XVIIl, pp. 10.5-295, ,Tnly 17, 1918. From Hie route tliereon given wo tind that Dr. Hasetlow jiassed through tlie Barker Gorge in the middle of tlie Xapier Range, then retraced his steps and went along the %ve.stern side northwards to another gaj) where he fornid the limestone caves which h(‘ later named 1\ angalinnya Caves. Thus Basedow’s localities are (piite distinct Irom those of Froggatl though the general name Barrier or Xapier Itange was used by both. All I'h'oggatt’s shells were collected in the southern end of the Range and in the Oscar Range which liis some twenty miles to the south-east. I have continually obseri'cel that our land molluscs mn.st be studied in conjunction with geograiihical, geological and climatic conditions. This in- stance provides a striking e.xamjile as the variation seen in the two cidh'ctions (with the same locality label) was not under.sood until the gi'ograjthy of the district was known. In a geological sketch plan which accom|ianies Base- dow’s account, the Xapier Range is shown as Devonian limestone, the aiiproach from Derby as Permo-Carboniferous, while the Kimberley block to the north- ■svard ajrpears as Cainbro-Ordoviciau, a pre-Cambria u sector irdervening ju.st north of the Xapier Range. A peculiar note is the showing of a .small ]iatch of Pcrmo-Garbouiferous to the south-east of Wyndham whence come some peculiar snails. With regard to tlu' A'ariation seen in the Xa])i(‘r Range collec- tion.s other factors, sindi as climatic, may hav(' interfered as this aspect needs consideration. Genus EHAGADA Allnu's 1861. 18()1 — Ithagadd Allnn-K, Die Heliceen, 2nd ed. (IMartens), p. 108, “1860.” Drthotyp(‘ I fell. - it'mga Gray = Pfeiffer. The restricted gi'oup of true Ehagada comprises small shells, of llattened Helicoid aspect, but chalky, rather shining, spire little elevated, sutures ini- ])ressed, whorls rounded, mouth .subcircular, outer lip thin but a little re- flected, inner base showing a slight tubercle, columella curved, apjiressed generally closing the umbilicus in the adult, a narrow perforation always visible in the juvenile stage. Sculpture of obscure, radials only. Coloration white with a few coloured bands. Rhagada torulus Eerussac 1819. Plato IV., fig. 1. 1819 — Heli.r toriditfi Eerussac, Hist. Moll. livr. 6, pi. XX VIE, figs, .'i, 1, Xov- ember; Syst. Tnbl. Mist. i\Ioll., )>. 34, .Ian. p. 30, .Tune 1821. Xew Holland (Peron) = Shark Bay, W.A. The oidy locality whence Peron could have secured a shell such as Eerussac figured is Shark Bay, and we find that, dealing with the natural history of Bernier Island, with which he associated that of Dorre and Dirk Hartog’s Isles, he mentioned “two species of land shells extremely numer- ous, but all dead, occuiiied great stretches of the interior of the island. One was a small species of Helix.” A liEviEw OF THE Land Mollusca op Western Australia. 59 Sniitli sugg'ost('(l lliiit rrinij(( Gray might bp sviioiiymoiis with tliis species, writing: It seems to me jiroljable that tlie //. tor::'Hs yer., is identical with this si)ecies {rcliu/a). It was collpct(‘d by I’eron (hiring one of the early Frimch voyages, but the exact |tart of Australia, wliere lie obtained it is not stated. Accepting this suggestion Hedh'v used toridnx, and I utilised it in the Ba.sie List, but in the moi’e detailt'd examination necessary for this account 1 found too many diserepaucu's, and now record ioruJns as available for a Shark Ray shell, whicli is not at present in our collection. Deshayes’ descrijition agrees vvilh the ligure which sliows a subglobose, subpei’torate shell, smooth, whit(' with one anl(‘-|)eri])heral brown band : the sutures deep, the whorls convex. The illustration is of a shell of the Shark Bay form, being more eonv(>.\ than the more northern rehiga series. Rhagada reinga Pfeiffer 1840. Plate IV., fig. 2. 1846 — Ileh.r reinga Pfeill'er, Symb. Helie., 5'ol. III., ]>p. .41, 50, 7.S ex Gray Ms. New Zeal., t. 1, f. 11, 12. New Zealand, type in Pfeiffer Coll., No. 459. 1848— l/c/cr reinga PfViffm', i\ton. llelic. viv., Vol. I., p. 289, cites Chemn., od. II., Helix, N. 44.'i, t. 73, f. 8-9. 1851 — Helix reinga Pfeiff'ei', Syst. Conch. Cab., (Mart. & Clnmin.), ed. Kuster, Bd. I., Abth. XIII., Teil 2, p. 52, id. 73, fig. 8-9. 1852 — Helix reinga IP'eve, Conch. Icon.. Vol. VII., pi. 128, sp. 772, Oct. 1890 Hieiix reinga Pilshry, .Man. Conch. (Trvon), Ser. II., V