t i! Li 4) BHI ‘" - j b i } Piet eee Weenicn i + thts eae i Os rceny a) ie oHoerni sk rt he ; if HH i pa i hlriats ih A vidas ne } 23 ist iat Hatt a Pe ASUNU Bates aed sp Magica ote cn UNE ER ee) ea URE Paes ta itis enti Ene coon e eta areata : : se PaO REESE RE ; 3 ROHS 0) Bisa caMANE AGT TRL Rep eT ae Reta auStale ene h a uenieneyt Husec ha rguipaouaeitee Sane ia ou tenet cram ; B naa mine iy a ae a . tithe yt i wists : e 2H Seo ys ote eres e? nud 243} 2 af ei sio NE m uh ee tte 3 o Rea ius EO cae Sry et Pas eta z Sgeteseeg exe Sars = Fy ' Ada ters aitiaed as } at as ' pili) We cans 5 Tat a8 ath ea ith BL As a OREERNANGH aie basse esos} rags (iar ited ih phat oas SEARLE ee i eaasctias UA eae ra tae Re: a ait ED pei a tr } i we He t i 3 tava rt a ase need nite it pk tisk aeiute at bt ie ras EY Ree 4 Y) wae She rae ; ace an re “7; oe =e ees eb eee (ieitet yi ge CH epee THE roe me aUNGS OF THE PINNEAN SOCIETY OF New SoutH WALES FOR THE YEAR Dt 4 5) VOL. L. WITH FIFTY PLATES and 242 Text-figures. SYDNEY: PRINTED AND PUBLISHED FOR THE SOCIETY BY THE AUSTRALASIAN MEDICAL PUBLISHING CO., LTD., Seamer Street, Glebe, Sydney, and SOLD BY THE SOCIETY. 1925-1926. +e CONTENTS OF PROCEEDINGS 1925 PART I. (No. 200). (Issued 17th April, 1925.) Pages. Presidential Address, delivered at the Fiftieth Annual Meeting, 25th MERE, USA, ly 1k, Jal, (Cranes, IMIG 216 o0 co oo o8 04 | oc i-xxxii Hlections and Announcements Mes MERE P ere Later) [DRG Ree Wepee hen MLE XXXiii 1kl@m, Weer Iles Saas oo 5650 oo 60 06 50 oo on v5 2CoghpoS 1 | Glendon Horizon §—----— | ____ 200ft above Muree Beds ! UREFEBEDS-—~—__ ea ~ 3000’ X Qa NS | : ta | = Hegeihe oN | = | ae 200. | =< BRANXTON BEDS INOWRA GRITS & ut Slon Meree Bo | co | so | = | NY | ) : ss fine ” Ff fy oa, Fo ft pen aauila AOrizon e Cacao JLLADULLA BEDS RETA GOALMEASURES V0 BEDS YOWER MARINE BEDS Up (ge aN COLUMN I COLUMN II Text-fig. 2.—Vertical sections of parts of the Permo-Carboniferous System, showing suggested correlation of Glendonite horizons in the northern and southern coalfields. in column i of text-figure 2. Column ii shows the equivalent stratigraphical succession in the South Coast District. Professor David, in his discussion of the Huskisson Beds, has correlated them with “the Zaphrentis horizon of the Branx- ton Beds (of the Hunter Coal-field) on the S.E. side of Black Creek. The latter BY IDA A. BROWN. 31 horizon is about 800 feet below the base of the Muree rock, and about 2,300 feet above the horizon of the Greta Coal Measures.” The position of the Wattle Ponds Creek horizon has been measured only with reference to the Muree rock, and may therefore also be equivalent to the Muswell- brook and Huskisson horizons. The difficulty of correlating definite horizons is due to the unequal develop- ment of the marine beds in the Southern and Northern Coalfields. The evidence in the Ulladulla District tends to confirm the inferences deduced in the paper by Professor David, Dr. Woolnough and others, with the exception of number (vi), which states that “The horizons of the glendonites being not far below, in some cases close to, the top of the highest beds of a Marine Series, where they are about to give place to fresh-water beds ey The Ulladulla Mudstones, in which the glendonites occur, are themselves the lowest beds of a marine series; the overlying level-bedded Nowra Grits are of the order of two thousand feet in thickness, and contain marine fossils at intervals right to the top of the series at Little Forest, a few miles north-west of Ulladulla at a height of 1,800 feet above sea-level, and almost that height stratigraphically over the glendonite beds. In conclusion, the writer wishes to thank Assistant-Professor W. R. Browne for his advice on the preparation of this note. EXPLANATION OF PLATES XIV-XV. Plate xiv. Sketch-map of the coast near Ulladulla, N.S.W., showing Glendonite horizons. Plate xv. 1. Group of Glendonites from Warden Head near Ulladulla. Photo: H. G. Gooch. 2. Glendonites in sitw on the rock platform at Warden Head, Ulladulla. 3. Large erratic of granite-gneiss embedded in rock platform at south end of Coller’s Beach, near Ulladulla. ON A REMARKABLE MODIFICATION OF THE EIGHTH ABDOMINAL SEGMENT IN LINDEHRA THESSALATELLA, WITH A DESCRIPTION OF THE MALE AND FEMALE GENITALIA. By ALFRED PuitportT, Assistant Entomologist, Cawthron Institute, Nelson, N.Z. (Communicated by Mr. HE. C. Andrews.) (Nine Text-figures. ) [Read 25th March, 1925.] The species under notice was described by Blanchard in 1852 (Hist. fis. y pol. Chile, Zool., vol. 7, p. 105). It is a well-known Australian species and is now established in New Zealand, at least in the North Island, and in Nelson in the South Island. In the following paper it is proposed to give a description of the male and female genitalia, there being some features of special interest in these structures. Text-figs. 1-4.—Lindera tessalatella Blanch. 1.—Lateral view of terminal segments of male, showing normal position of genitalia. c, colon; p, penis; sac, saccus; w, uncus; v, valva; 7, seventh segment; 8, modified eighth segment. 2.—Modified eighth segment, obliquely lateral view. 3.—Tegumen, from beneath. sac, saccus; wu, uncus. 4.—Penis. d.e., ductus ejaculatorius. BY A. PHILPOTT. 33 The Male Genitalia. The abdomen, normally, shows only seven segments, the genitalia being withdrawn within the last one (Text-fig. 1). The ninth sternite forms a large saccus, the cavity of which receives the rounded basal part of the valvae; basally jt bears a small finger-like projection. The saccus merges, without articulation, into the ninth tergite (forming the upper part of the tegumen), which is thin and hoop-like, a short slender uncus springing from its apex (Text-fig. 3, also Text-fig. 1). The valvae.—The valvae (Text-fig. 5) are broad and rather concave, but much narrowed on the apical fourth. The upper part of the base is deeply excavate, leaving only a narrow portion of the costal margin projecting. The armature is feeble, consisting only of a series of short blunt spines round the edge of the narrow apical part, and a few weak hairs. The penis.—The penis (Text-fig. 4) is a blunt, slightly-curved organ. The aedeagus is not strongly chitinised and bears no armature.. Its base is curvedly truncate, forming a large aperture, into which the rather small ductus ejaculatorius enters. . The modified eighth segment.—Embracing the genitalia as they lie in their ordinary position is a strongly chitinised structure (Text-fig. 2, see also Text-fig. 1). It begins as a thin rod lying beneath the saccus, but, after passing under the valvae for a short distance, it divides into two branches which curve upwards, one on each side of these organs. Its upper portion is recurved, much broadened and divided apically into two prongs, the upper pair of which nearly meet dorsally. The valvae, tegumer and penis all lie within it and its function is evidently to act as a support for these structures, providing a strong, and at the same time elastic, control. It should also be noted that the lower prongs are so directed as to pass over the penis, thus giving additional protection. No such organ as the one here described had been previously met with by the writer in connection with the genitalia of the Lepidoptera. Though extremely puzzling at first sight as to origin and homology, it became a simple matter when it was recognised that here we had merely the eighth segment, or possibly the eighth sternite only, which had become modified into a spring-like band to support and protect the all-important terminal organs of the abdomen. The Female Genitalia. The copulatory opening, ductus bursa, and copulatory pouch offer no unusual features and need not be further referred to, but the ovipositor exhibits some interesting peculiarities. When in a position of rest the eighth and ninth seg- ments, which form an elongated ovipositing tube, are in great part withdrawn within the abdomen. The ninth is telescoped within the eighth, except for the swollen apex, and the eighth is, in turn, retracted within the seventh, only the inter-segmental membrane protruding. A ring of bristles marks the apex of the eighth segment, and apparently acts as a guide for the correct degree of with- drawal. When fully exserted these two segments, with their connecting membrane, make up about two-fifths of the whole length of the abdomen. The segmental rods (Text-fig. 6) are very long; in the normal position of the abdomen those of the ninth reach to just within the fourth segment, while those of the eighth extend almost to the third, their extremities being curved strongly upwards so as nearly to touch the dorsal wall of the segment. D 34 THE EIGHTH ABDOMINAL SEGMENT IN LINDERA TESSALATELLA. The ninth segment terminates in a rosette-shaped opening which can best be understood by reference to the figures (Text-figs. 7, 8 and 9). The dorsal part of this rosette consists of an apically bifid lobe, bearing a number of curved hairs. The function of these hairs, which are curved backwards and outwards, is probably, by engaging with the material among which the eggs are being laid, to hold the end of the abdomen in the desired position. There is a small lateral piece and a fairly large ventral one, divided into five lobes. Text-figs. 5-9.—Lindera tessalatella Blanch. 5.—Valva, from within. 6.—Terminal segments of abdomen of female. 8 eighth segment; 9, ninth segment; i.m., inter- segmental membrane; s.7r., segmental rods. 7.—Dorsal view of ‘‘rosette.” 8, eighth segment; 9, ninth segment; i.m., inter- segmental membrane. 8.—Lateral view of “rosette.” 9.—Ventral view of “rosette.” Though the actual act of oviposition in Lindera does not appear to have been observed, it seems highly probable, judging from the structure of the ovipositing organ, that the eggs are laid in cracks or crevices, such as would be beyond the reach of a shorter structure. Mr. J. G. Myers has reared the moth from larvae found in old sacking and rubbish; doubtless the elongated organ would be suit- able for the deposition of eggs in the interstices of such material. NOTES ON AUSTRALIAN DIPTERA. No. v. By J. R. MALLocnH. (Communicated by Dr. E. W. Ferguson.) [Read 29th April, 1925.] In this paper I present keys for the recognition of the genera of Muscidae known to me from Australia. It is quite probable that there are some genera, and certain that there are many species, unknown to me, and I believe that the best method of interesting Australian students of Diptera in this rather difficult family is to make available to them data for the recognition of the known genera and species. From experience I judge that there is nothing tends more strongly to divert students from the intensive study of a group than a multiplicity of genera and species scattered throughout the literature of many countries, and in many languages, with no synoptic literature available for their recognition. I do not presume to say that this attempt to elucidate the eeoion genera is perfect, that anyone without experience in the work may sit down and in an off- hand manner name whatever species comes to hand. But the matter herein presented is the result of over twenty years’ study of the group and contains data which, in some cases, are made known for the first time in print. I have had ready for the press for some years a key to the genera of Muscidae of the world, but have delayed publishing it, as I desire to make a more extended study of some difficult groups, more especially those occurring in New Zealand. It would serve ‘ no good purpose, even for comparison, were I to include in this paper genera not yet known from Australia, so I confine my presentation strictly to those genera I have seen from my various correspondents in Britain, Australia, and elsewhere. Included in the family are all species that possess the following combination of characters: Second antennal segment with a longitudinal split at apex above, slightly to the outer side; spiracles of abdomen situated in the tergites at varying distances from the lateral margins, rarely in extreme margins; auxiliary vein of wing complete, ending in costa well in front of apex of first vein; postscutellum not convex; hypopleura without a vertical series of strong bristles below spiracle, sometimes with some fine hairs in this region; vibrissae present. The nearest related families are the Calliphoridae and Tachinidae on the one hand and the Dryomyzidae, Helomyzidae, and Sciomyzidae on the other. The group usually referred to as Scatophagidae or Cordyluridae I consider a subfamily referable to Muscidae, but I have seen no species from Australia that belongs here, the only genus referred to it definitely by a recent author, Tapeigaster Macquart, being, I think, an acalyptrate. Within the past two or three decades there has been considerable change in the opinions of workers on the calyptrates as to the limits as well as the number of families in this division. The more dogmatic definitions of these families have 36 NOTES ON AUSTRALIAN DIPTERA, V, been laid down by those workers who have confined their efforts to the species of a limited faunal region, and the increase in the number of families is credited to those who, covering several faunal regions, have endeavoured to apply the same differentiating characters to the entire assemblages as to the components of any given faunal region, with very unsatisfactory results. It appears to me, from a critical survey of a multitude of species from every part of the world, that we may be compelled to recognize but four families in this complex, Tachinidae, Calliphoridae, Oestridae, and Muscidae. There are several groups, such as are represented by the genera Sarcophaga, Metopia, Rutilia, Aulacocephala, and Mesembrinella, etc., which may serve as the nucleus of tribes or groups of lesser rank than families, even subfamilies, but, unless we are prepared to elevate the four groups above named to the rank of superfamilies, we must perforce admit that family rank, even for the most distinct of the others, is out of the question, viewing the matter from a comparative point of view. I present below a key to the four families listed above. Key to Families of Calyptrata. 1. Hypopleura without a vertical series of strong bristles below spiracle.............. Aaa neem Me PH oot Moet No ta MS ei maa ae er MUSCIDAE (incl. GASTROPHILIDAE) Hypopleura with one or more vertical series of bristles below spiracle .......... 2 2.) Mouths parts vestigial ayy. ce cydaeses beer hhh vbr seaeanetcw dies eae Ree eee OESTRIDAE MIGUIA ARES WEN GleweloMacl, AMIMCHOME .5.5¢cscncacncocsoccodstucccaucaebcacase 3 3. Postscutellum fot conspicuously convexly developed ............ CALLIPHORIDAE Postscutellum prominently convexly developed ..................... TACHINIDAE There does not appear to me to be any good reason, based upon structure, for the separation of Tachinidae from Dexiidae of authors. The absence or presence of hairs on the arista, the only character that is used generally for the separation of these groups, being unreliable for that purpose, I have included them both under Tachinidae in the above key. I present below a key for the recognition of the subfamilies of Muscidae known to me from Australia. Key to Subfamilies of Muscidae. 1. Sixth wing-vein complete, faint apically, but extending to margin of wing; thorax with but three strong pairs of postsutural dorsocentral bristles; hind tarsus with an outstanding bristle at base below on basal segment; eyes of males narrowly separated above, of females separated by about one-third of the head width, interfrontal stripe in this sex generally with a pair of cruciate bristles about the middle; hind tibia with at least two long posterodorsal bristles; scutellum with some erect soft hairs below apically ............ ANTHOMYIINAE Sixth wing-vein not distinctly traceable to margin of wing; other characters not BAS: Above: dni COCON TL. SOR AIAG 5 a See arses ial POON Menace tebe eis eetonts chs dan eatee aire tete eee 2 2. Pteropleura with a group of hairs in centre; eyes of males and females separated by about one-third of the head width, the interfrontalia without “cruciate bristles ; palpi dilated apically, usually conspicuously so; parafacials with some hairs on their sentire’ Tenethy” cezscvasous asm ecceco sare ee ow ters ke mene ROR MeL onewel ol aiosic dsaothe eye oh eneretere LisPINAE Pteropleura bare, or, if haired, the eyes of males are not as widely separated as those of females; in no case are the palpi conspicuously dilated, and the parafacials are bare below bases of antennae .....................+22--+e- 3 3. Sixth wing-vein very short, the seventh more or less distinctly curved forward round its apex; hind coxae usually with one or two fine hairs at apex above; thorax with three pairs of postsutural dorsocentral bristles; each orbit in females and in males with wide frons with two bristles on upper half directed outward over eyes, or the upper one slightly backwardly directed ................ FANNIINAE BY J. R. MALLOCH. 37 Sixth wing-vein sometimes very short, but when it is, the seventh is straight and never curved forward round its apex; hind coxae bare at apices above; orbits not with two upper bristles curved outward over eyes in either sex ........... 4 4. Frons in both sexes one-third of the head width, each orbit with one long backs wardly directed bristle on upper half; thorax with but one pair of presutural dorsocentral bristles; lower stigmatal bristle directed downward .... COENOSIINAE Frons of male narrower than that of female, or, if equally wide, the orbits have each two backwardly-directed bristles on upper half; there are two pairs of presutural dorsocentrals present, or the lower stigmatal bristle is not directed downward .. 5 5. Proboscis heavily chitinised, fitted for piercing, the labellae not enlarged or spongy ; arista with long hairs on upper side, bare or pubescent below; lower calyptra as in Phaoniinae; fourth wing-vein bent forward at or before middle of its apical SOCULOTIMMMET pane n rene rate oot eT Aone) cet Als: SMC Ata eewes sorthe Pabst atcuaBete mute a eae he STOMOXYDINAE Proboscis not heavily chitinised, or fitted for piercing, the labellae generally large and more or less spongy, with chitinous rods or teeth; arista with the hairs almost invariably equally long above and below .................-.---++--, 6 6. Lower calyptra rounded apically, its inner basal margin well separated from the basal angles of scutellum; fourth wing-vein never angularly bent forward near middle of its apical section; if curved apically, the beginning of the curvature is generally well beyond middle of apical section ................... PHAONIINAE Lower calyptra more or less transverse at apex, its inner basal margin close to or underlying the basal angle of scutellum; fourth wing-vein conspicuously bent forward at or close to middle of its apical section, sometimes angularly so 5 CRORE. HERERO LCT Oe PIC one CRCIONOrO erence er CR IerE: CrOtn cl Mer ao ro: one e CUE ieer Noten eerie SerCt cic ras MUSCINAE Subfamily ANTHOMYIINAE. This subfamily I consider the most generalized of the family. It is essentially northern in its distribution, being abundantly represented in Europe and North America, extending its range into the Arctic Regions. Many of the species are phytophagous, some of them doing serious damage to root-crops and cereals. So far I have seen but three genera from Australia and these may be distinguished by means of the key presented below. Key to genera. 1. Propleura haired in centre, i.e. below humeral angle and above propleural bristle in front of spiracle; species densely whitish-grey pruinescent, with conspicuous black spots on dorsum of thorax and abdomen .............. Anthomyia Meigen ELODIEUGAs DAREGH IM CONthers soi citcews eee cy sve cy cuebe ay dlareueiec shut) Re LeTe nV hs ME ac hoo soe es 2 2. Mid tibia without a ventral median bristle ...................... Hylemyia R.-D. Mid tibia with a distinct median ventral bristle ......................- Egle R.-D. Genus ANTHOMYIA Meigen. I have seen but one species of this genus from Australia, vicarians Schiner. Stein considered this species to be a synonym of illocata Walker, but I think he was in error. I have published a key to the species of the genus known to me (Annals and Magazine of Natural History, 1923, 267). Unfortunately the last section of caption 16 referring to vicarians was accidentally omitted in printing the key. This should read: “No dark spot at base of antennae in male; base of scutellum black .. vicarians Schiner”’ The specimens I have seen are from Eccleston and Maroondah, N.S.W. Genus HytemyiA R.-D. I have recently published a key to the known Australian species of this genus. Genus HGLe R.-D. But one species of this genus is known to me from Australia. As it is of some economic importance, the larvae feeding in various root-crops, i give a short description of it to facilitate its identification. 38 NOTES ON AUSTRALIAN DIPTERA, V, An entirely black species, the interfrontalia of the female usually rufous in front; wings hyaline. Calyptrae white. Halteres yellow. Thorax in male with three broad black vittae, the spaces between them not conspicuously pale; in the female the thorax is much more densely gray pruinescent and the vittae are not very noticeable. Abdomen in male gray pruinescent, with a broad central vitta and a narrow apical and basal fascia on each tergite black. Eyes very narrowly separated in male, in female separated by fully one-third of the head width, interfrontalia in female with a pair of cruciate bristles; arista pubescent. Thorax with 2 + 3 pairs of dorsocentral bristles and one or two pairs of presutural acrostichals. Abdomen in male depressed. Fore tibia with one anterodorsal and one posterior bristle near middle; mid tibia with one ventral, one anterodorsal, and two or three posterior bristles; hind femur in male with a series of long bristles on anteroventral and another on basal two- thirds of posteroventral surface; hind tibia with one or two anteroventral, four to six anterodorsal, and three posterodorsal bristles. First posterior cell of wing slightly narrowed apically. Lower calyptra protruded. Length, 5-5.5 mm. One pair, Melbourne, Victoria, 14th Nov., 1923. Subfamily LIsprnae. I have already published a generic key to this subfamily (THESE PROCEEDINGS, xlviii, 1923, 606). Subfamily FANNIINAE. There are but two genera of this subfamily known to me from Australia. I give a diagnosis below for their identification. A. Eyes of male separated by less than one-fifth of the head width; entirely black species, or at most only the base of abdomen faintly translucent yellowish See paysite yee re fuels ethalah os ecto idosecla setae) tegte ee ces oye Ulbon Seuah ante ceielepieaetrsl cite is: Slee Meroe ROR eer eee Fannia R.-D. AA. Eyes in both sexes separated by one-third of the head width; legs and abdomen CARWALY LY. ClUO WS lattoa a ereed is, state to prret tees aantetet ate tt SETI cogtndchee Sea meter aes Euryomma Stein. Genus Fannia R.-D. I have described one species, australis, and recorded another, canicularis L., from Australia. There is a third species known to me, but I have only a rather poor male specimen of it on hand, so defer publishing an identification meanwhile. Genus HurYomMMaA Stein. I have already dealt with the only known species of this genus from Australia, peregrinum Meigen, in a previous paper in this series. Subfamily CoENOSIINAE. This subfamily is a rather poorly defined one. I have been compelled to drop Atherigona Rondani and Pygophora Schiner as members of it and have transferred them to Phaoniinae. Ultimately it may be necessary to do the same with the other genera, in which case the group would bear the oldest name, Coenosiinae. I have seen but two genera from Australia, which are distinguished as below. A. Hind tibia with a long anterodorsal bristle near middle and a much shorter one on anteroventral surface distinctly apicad of it ............ Coenosia Meigen. AA. Hind tibia with two long bristles near middle, one on the anterodorsal and the other on the anterior surface, the bases of which are quite close together bbsroyatotdrorcr DPR R ANE RORHES Steaveim theauemmestar Use wettinen le aon “Sto eneeorala wy Gidea Ocmotorae aso Caricea R.-D. I have dealt with the species of these genera known to me in the preceding paper in this series. BY J. R. MALLOCH. 39 Subfamily PHAONTINAE. This is the best represented subfamily in Australia, as it also is in Africa and South America. I give below a key for the identification of the Australian genera known to me at present. 10. ial. 12. 13. 14. 15. 16. Key to genera. (2 FETO Puma eRe Merete ancp crass, sbesavsh as; aro Mee ee ee en RPA ELA: Hee eedn a ea ae eutibe alee ie 2 IPtEropleunal wale ye MY OAM bieg ip ens, 5, = cree RA ec eR ea cnet oe) SP at eae on ere wile ete cations 24 Mindswineoveinwhanrevatebase Pasi. Sede clieteeetares sists ota oh aud alts (seta eae Paros tees eat Se eae 3 Third wing-vein with some minute setulae at base .............2..0002ceeeeuee 22 Mid tibia with a strong ventral bristle near middle ................ Anaclysta Stein. Mid tibia without a distinct ventral bristle near middle ...................... 4 IVINIC Smee eer en Voie tru -rttars My ets clear d his neces Bh ee gdh bad Rete nad Bee sole Bess es en eee 5 JECIRAWIOS | 5 ci a peta: OeC RE SIO se Caen IE MEP Togs eh tarictic ican gp eee Ra ON Geh ae Sn eng UR Tmo 14 Fore femur with an elongate depression or concavity on apical half of ventral surface at the basal extremity of which there is at least one sharp thorn or STOIC Mlaey sce Fone tae 1 ct ore cus ered on. Aisdsey shan), Ba OP acm eR kay sos ohewel am woke obs Hydrotaea R.-D. Fore femur without a distinct depression or concavity and thorn as above ...... 6 Eyes separated by about one-third of the head width ......................0% 11 Eyes separated by not more than one-fifth of the head width ................. 7 Hind tibia with one outstanding bristle on posterodorsal surface distinctly beyond middle, which is noticeably longer than diameter of tibia .. Phaonia R.-D. Hind tibia with at most one or two short setulae on posterodorsal surface, none of which is noticeably longer than diameter of tibia .......................- 8 Auxiliary vein of wing gradually approaching costa, first vein connecting with the latter beyond inter cross-vein ; halteres black FE Me ee eee TRO 9 Auxiliary vein with a rounded curvature near apex, its junction with costa almost a right angle, first vein connecting with costa before or above inner CLOSSAV GLI sapien Wee sire hese ont oe A A TE eee Gas Seay oP Ng oe cee 10 Eby WoOpletnawbane) wpis faethe oe Dies ek Siac EE ee. Se Ophyra R.-D. Hypopleura with fine hairs on upper margin in front of spiracle ............ CNOMOTE BY Oo ane AUER OO) CHOTO TG RON ORC ETE ROG CRONE CERCIE SECRETE: En RPMS SOR EE meat tri rc Australophyra Malloch. Fourth wing-vein not or very inconspicuously curved forward on apical section ; prescutellar acrostichals not noticeably proximad of posterior pair of GCOESOCENERATS iy Pie iiccite be tee ccna eee eee inicio zay al ees eepa eeee ned eirs Helina R.-D. Fourth wing-vein rather conspicuously bent forward on apical _ section; prescutellar acrostichals very noticeably proximad of posterior pair of GOESOCENULRALS tyes ereaey cre eles lavel suspelichere nay srova, shisiwelo on slice ec iiciev oysicueye Antipodomyia Malloch. Costa not extending to apex of fourth wing-vein .............. Neohelina Malloch. Costa GRAINS HO) BYOEDS OLE IROUNAT AD, WALA oooocnb donno coon onendnobeub nono 12 Hind tibia with only three bristles, one on each of the following surfaces: anterodorsal, anteroventral, and posterodorsal; presutural and anterior postsutural dorsocentral bristles much shorter than usual, hardly distinguish- ablestrom~ephediscalshairspandsserulaena ee aeimciceieeeeeieree Atherigona Rondani. Hind tibia with at least four bristles, two on each of the following surfaces: posterodorsal and anterodorsal; all dorsocentral bristles long and strong .. 13 Each orbit with the two upper bristles directed backward, the anterior one of these not much longer than posterior one and not placed close to the bristle hgh DUR M KOE ANCE AS ey manent Os ee ih ener ee Oa SURE een oe era a ee eMC Cae eos Lispocephala Mik. Each orbit with the upper two bristles directed backward, the anterior one much longer than the posterior one and much closer to the bristle anterior to it WHIM Me WS tO We jOoswecioe Om GoococccocaueaccocunoUuesouS Pygophora Schiner. IMGcnrOmuaNiG, TL A, TENIe Ge GANGA IWMSUIES seoscocscn0ng00d00sanGndd0bbabG 15 Interfrontalia without a pair of cruciate bristles, sometimes with scattered fine TVET! yack i eee ae A te oteiosy Oe nos a eS clay o SED Yai titel arate tree ee aioe lateua ee iL'7/ Hypopleura with hairs on upper margin in front of spiracle .. Australophyra Malloch. TEIN fog Tbh eh Love) ce hetaa ctor Oc iG. bicuct cee Le Giotencieacy duc Ceoent ere ALcR REC tcentaic cect cet cine nechoEr akAncr co anicneernr st 16 Ocellarmtrians le hiechilyscslossyvan black mae ae ieee oleia cl oreo = aie Ophyra R.-D. Ocellar triangle more or less shining, not highly glossy .......... Hydrotaea R.-D. 40 NOTES ON AUSTRALIAN DIPTERA, V, 17. Hind tibia with one outstanding bristle on posterodorsal surface beyond middle, which is noticeably longer than tibial diameter; dorsocentrals all long DN Re IR Were) NTE TPT ae ees “Siac Trees ......... hirtibasis Malloch. Thorax with 4 pairs of postsutural dorsocentral bristles ............ regina Malloch. Thorax with 3 pairs of postsutural dorsocentral bristles .................... 9 Thorax black, without blue tinge, with rather dense whitish pruinescence; hind tibia with two anterodorsal bristles and no outstanding anterior setulae; lower calyptra HAIG MEVAFO WIE WHEW WISDEN) oe ooot oonbowoadeboudcebooeuode calyptrata Malloch. Thorax and abdomen blue or green, with whitish pruinescence, that on abdomen GHA; 915) 006 daa eee Shon ROOM RCRD Te ickca Gace RTC aa tani ern Corie tech Cocme ri eeN colons Granth cackeN uCA EC IANCROaGTC 10 Thorax blackish-blue, abdomen blue; hind tibia with three or more anterodorsal bristles; wing without brown margins to veins ............ caerulescens Stein. Thorax and abdomen green, the former pruinescent, the latter without whitish pruinescence; hind tibia with two anterodorsal bristles; wing-veins margined with brown, the cross-veins most conspicuously so ............ howei Malloch. Thorax with 4 pairs of postsutural dorsocentral bristles .................... 12 Thorax with 3 pairs of postsutural dorsocentral bristles ..................... 3 Eyes of male very closely contiguous .................... TUR ts nigrescens Stein. Eyes of male separated by more than the width of third antennal segment ........ Ea Nee Oe CEO EEG De SEO CRER E er CUE M ERIS PER eoereG Na pach GiomROrOTE Rearcin nigrohalterata, n. sp. Legs entirely black; margins of calyptrae fuscous in male, yellowish in female SRE Ge eNDA ONS OTIC CRON RCRDIOR CROPS CHONG IMCS ECC CCE OES HCCC ne are Ervcue Oreo ores castigata, n. sp. Tibiae tawny, remainder of legs black, margins of calyptrae yellowish in both SOROS MT nicssrsu sy otcecetenaece dees ans CORE OER E CI CREc cree oNo ROeO MERE OR IORCk ree simulata Malloch. Thorax testaceous yellow, with or without distinet blackish vittae; sternopleurals 1: 2 oh OP Cte Coo ey REED Bate aT OI OTE DIRE COORG Oo CROMER ETC GMS PROIET aT Dee Shy SE A eed Geet otc oot iotcenid aicacicke ontere 15 ALM OR AK DlACK. en Owm eRe: Vell O Wares tyscty sas mois Steet ee Ch oes) ep Ssasd Sees Sek Sas aey Sy ew on New oe arom t ote 16 Longest hairs on arista much shorter than width of third antennal segment; thorax with one pair of strong presutural acrostichals; head black; legs yellow, tarsi LSC) iota raecte clan ce Oret ov Oncec arc 6 Gincd alo te Dera c. o B. oftaae oat One pe act eon IDRC imitatrix Malloch. Longest hairs on arista longer than width of third antennal segment; thorax without SELON Se DLeSULUnA ACL OStichale briStlesmenieiieiie citi ieieciacicl cae eiaieia iene ieien limonene 16 42 16. 17. 18, 18). 20. 21. bo BO 24. 26. 27. NOTES ON AUSTRALIAN DIPTERA, V, Thorax with 4 pairs of postsutural dorsocentral bristles; head yellow, third antennal segment and sides of frons black; thorax with three broad varicoloured dark vittae; apices of femora and tarsi black ................... fuscoflava Malloch. Thorax with 3 pairs of postsutural dorsocentral bristles; head entirely black; thorax with a fuscous dorsocentral vitta; femora yellow, tarsi black .................. Thorax with 4 pairs of postsutural dorsocentral bristles ..................... 18 Thorax with 3 pairs of postsutural dorsocentral bristles ....................- 23 Longest hairs on arista longer than width of third antennal segment; legs tawny, tarsi fuscous; cross-veins of wings not noticeably clouded; fore tibia without a median posterior bristle; fore femora of female normal ................ 19 Longest hairs on arista very much shorter than width of third antennal segment; at least a part of femora in addition to tarsi fuscous in males; inner cross-vein and sometimes also the outer one distinctly clouded; fore tibia with a median posterior bristle; fore femora in female with a scale-like downward extension at apex on anterior side and a large flat bare area at base on same side .... 21 Antennae and palpi black; presutural acrostichal bristles absent or very weak Peo AEP Cea RCO aS ore CIO or onetey os cut ora tat ONS CE Ie REI O Gach o/aneh antarctica Macquart. Antennae partly, palpi entirely yellow, Fac. «.= « a. ce + len cies ics ener ieee) eters 20 Thorax with one pair of long presutural acrostichal bristles; hypopleura pilose below BDIN ACTS. He aha creieted sre totsesaeret ora te tenets ete an sega cclstenectencscetenehRee ciate nme nee adversa, Nn. sp. Thorax without presutural acrostichals; hypopleura bare .... vandiemani Malloch. Legs black; hind tibia on male with a series of very long bristles on anteroventral and another on posterior surface; fore femur of female with the apical scale- like extension short and inconspicuous ................ poeciliventris Malloch. At least the tibiae tawny; hind tibia of male with from two to four short antero- ventral bristles and no series as above; fore femur of female with the apical scalleslike ‘extension very distinct, .6...445 tates ce ere ee Sue en oe Renin 22 Femora almost entirely black in male, only the extreme apices yellowish, in female the apices are broadly yellowish; fifth sternite of male with normal hairs; fore femur in female much thicker than usual, with the flat basal area on anterior side covering nearly the basal half, the apical anterior scale very large GIs ORCI OORT oT ECR aCe Sholay Sito ee MoE Ot ico louae acd Bactrg a addita Walker. Femora of male broadly yellowish at apices, those of female entirely yellow; fifth sternite in male more densely haired than usual; fore femur in female but little thickened, the oval bare flat area at base on anterior side not one-fourth of the femoral length, apical scale not very large .............. piliventris Malloch. Wings with a fuscous spot close against third vein just beyond outer cross-vein; both cross-veins broadly clouded; longest hairs on arista distinctly shorter than width of third antennal segment; fore tibia with a median posterior bristle; knlobsizo £whlalliteresiibr owaleracisios cieinicleie)- ioe) nee cn iene trinubilifera Malloch. Wing without a spot in first posterior cell as above; cross-veins clouded or clear eee eye Hiacee Dlaliuvs 3) Grd oh kao Mbit ais, Rica aiies Guapo ee eie len sy obs, cue eusicts Sa kaet oun care Cross-veins of wings conspicuously clouded ..................2-c cece eee eeeee 25 CHOSE VEINS OF WANS moe iMmCeeeiolhy, Clowielecl sooonsos coda dono ne de oD doe OOD HOGS 26 Longest hairs on arista much shorter than width of third antennal segment; fore tibia with a median posterior bristle; presutural acrostichal bristles absent OCR AEE OME re mR TO ROT CRO Cn Oo CREO oF Omron Ot OO AL Cla cd OED victoria Malloch. Longest hairs on arista about as long as width of third antennal segment; fore tibia without a median posterior bristle; thorax with one pain of presutural ACLOSLICH Al /DLISHIEST rac ch eee eee i CORSE Ee ee eee simulata Malloch. Bristles on anteroventral surface of hind femur extending from base to apex; neither the median posterior bristle on fore tibia nor posterior setulae on hind tibia present; no strong presutural acrostichals present ............ micans Malloch. Bristles on anteroventral surface of hind femur confined to apical half, or, if present on almost the entire length, there is a bristle at middle of posterior surface of fore tibia or there are setulae on posterior surface of hind tibia .. 27 Fore tibia with a median posterior bristle; hind femur with an almost complete series of anteroventral bristles; thorax with a pair of short presutural acrostichal bristles; hind tibia without posterior setulae ....................+-eeeeees 28 Fore tibia without a posterior median bristle ....................--e2+ceeceues 29 BY J. R. MALLOCH. 43 28, Ices eairaiby APNONe 6 condo eb odio cgocd 66 ood cla nlusio cme nm DU dlamomo versicolor Stein. LOSS walla Usha IQEC so te5c5cgcouepoaccb ovo oD dHHoeoHOdC spilariformis Malloch 29. Thorax without a strong pair of presutural acrostichal bristles; abdomen without a brassy or violaceous tinge; posterior setulae on hind tibia absent ............ ENE Sod eV sttes FarM ae ew ee SR a arr sc Shelyeerer ae ce ere temet svat Stee ereT ear Daserapee Suction etiote ian eran achaeta Malloch. Thorax with a long pair of presutural acrostichal bristles; abdomen with brassy or violacecus reflections; hind tibia with one or more weak setulae on posterior S[oariesKersy VOKEENO 1ggNKOKON Veen ero erat ced oO SOO CANO Ole Gas Goro AtrD aeneiventris Malloch. N.B.—The descriptions of most of the species ascribed to the writer appeared in the series of papers published in the Annals and Magazine of Natural History, under the title “Exotic Muscaridae” from 1921 to date. HELINA HYPOPLEURALIS, N. Sp. 9.—Head black, orbits, face and cheeks with slight whitish pruinescence. Thorax black, slightly gray pruinescent, dorsum with four dark vittae, most distinct in front of suture. Abdomen blue-black, with whitish pruinescent checkerings. Legs black. Wings slightly yellowish. Calyptra white, the margin of lower one fuscous. Knobs of halteres brownish-black. Frons at anterior margin about one-third of the head width, narrowed behind, triangle shining, narrow, extending to anterior margin of frons, each orbit with two upper bristles directed backward; third antennal segment about three times as long as second; arista plumose; parafacials very narrow; cheek about twice as high as width of third antennal segment; eyes short haired. Thorax with 2 + 3 pairs of dorsocentral bristles, two postsutural intra-alars, prealar moderately long, and no outstanding presutural acrostichal bristles; prescutellar pair of acros- tichals long; sternopleurals 1:2; hypopleura haired below spiracle; hairs not descending on sides of scutellum. Basal abdominal sternite bare; genitalia normal. Fore tibia without a median posterior bristle; mid femur without ventral bristles; mid tibia with three posterior bristles; hind femur with two or three preapical anteroventral bristles; hind tibia with only two bristles, on anterodorsal surface. Fourth vein slightly curved forward at apex. Length, 7 mm. Type, E. Dorrigo, N.S.W., 30 January, 1923. HELINA NIGROHALTERATA, DD. SD. 6-—Similar to addita Walker in colour, chaetotaxy, and general habitus. Entirely black, thorax quadrivittate, abdomen with a pair of dark spots on second, and another on third visible tergite, legs black, both cross-veins of wings clouded, the outer one least noticeably so. Calyptra dirty yellow. Knobs of halteres blackish. Eyes hairy; frons at narrowest part distinctly wider than third antennal segment; arista with its longest hairs a little longer than its basal diameter; cheek about three times as high as width of third antennal segment. Thorax 2—3-+4 pairs of dorsocentral bristles, two pairs of long presutural acrostichals, and the prealar minute; sternopleurals 2:2. Abdomen as in addita, the bristles quite long and strong. Fore tibia without a median posterior bristle; fore femur normal; mid femur with three or four long posteroventral bristles on basal half; mid tibia with three or four posterior bristles; hind femur with an almost complete series of anteroventral bristles; hind tibia usually with three anteroventral and two anterodorsal bristles, and a few posterior setulae. Fourth wing-vein not bent forward at apex. Length, 7.5 mm. Type, Gisborne, Victoria, 2 April, 1922 (G. Lyell). 44 NOTES ON AUSTRALIAN DIPTERA, V, HELINA CASTIGATA, 0. Sp. fg. 9.—Male similar to preceding species, differing in having the frons narrower, the postsutural dorsocentrals three pairs, the abdominal dorsal spots much larger, subtriangular, and rather poorly defined, with a central dark line between them, and the hind femur with the anteroventral bristles present only on apical half. The fore femur in female is normal. Length, 5-6 mm. Type, male, allotype, two male and two female paratypes, Eungella Ra., 45 miles west of Mackay, Queensland, 1,400-2,400 feet, 13 to 25 Sept., 1923 (Goldfinch). HELINA ADVERSA, 0. SD. °.—Head black, antennae and palpi orange-yellow, apical half of third segment of former fuscous. Thorax and abdomen black, with gray pruinescence, the former faintly vittate, the abdomen with a dotted appearance. Legs orange-yellow, tarsi black. Wings clear, yellow at bases. Calyptra and halteres yellow. Eyes pubescent; arista plumose. Thorax with 2+ 4 pairs of dorsocentral bristles, and one pair of long, strong, presutural acrostichals; prealar short; sterno- pleurals 1:2; hypopleura with some hairs below spiracle; hairs not descending on sides of scutellum. Abdomen normal. Fore tibia without a median posterior bristle, femur normal; mid femur without ventral bristles; mid tibia with three posterior bristles; hind femur with two or three preapical anteroventral bristles; hind tibia with two anterodorsal and one anteroventral bristle, and a short posterodorsal bristle beyond middle. Venation of wings normal. Length, 8 mm. Type, Fish River, N.S.W., 25 March, 1923. There are one or two more species of this genus in my possession, but I consider that, owing to the very large number of species that have been already described, it is advisable to put forward, at this time, the foregoing key so that anyone who is interested in the group may be enabled to identify the more common species and, at the same time, get an idea of the various characters used in the differentiation of the species. Later on, when I have received all possible material in the genus, I may give a complete key to the species, but there is always the danger that, by putting off the presentation of a key, one may do so until such a presentation becomes impossible for some reason over which one may have no control. I shall always be glad to examine and report upon species of the genus should such be sent me. Practically nothing is known of the larval habits of the genus elsewhere and of the Australian species we know nothing at present, so that here is a good field for biological investigation. i Genus ANTIPODOMYIA Malloch. But one species of this genus is known, bancrofti Malloch. Genus NEOHELINA Malloch. The two species of this genus were diagnosed in my last paper on Australian Diptera. Genus ATHERIGONA Rondani. There are several species of this genus known to me from Australia, one of which, excisa Thomson, is very widely distributed, occurring in the Americas, BY J. R. MALLOCH. 45 Africa, and from Southern Asia to Australia. The larva lives in decaying fruits. Another species, tibiseta Malloch, is known at present only from Australia. Lack of males of other species prevents me from dealing with the whole genus at this time. ; Genus LispocrEPHALA Mik. Stein described tinctipennis from Australia. This species is unknown to me. He also places intacta Walker in this genus, with a doubt. I intend to revise the genus shortly, its limits at present being but imperfectly understood. Genus PycorHorA Schiner. There are two species definitely known to me from Australia, apicalis Schiner, the genotype, and minuta Malloch. I believe that there is a third species, but have so far seen only female specimens of it. Genus RHyNcOMYDAEA Malloch. There are two species of this genus in Australia, carinata ‘Stein, and australis Malloch. Genera LimnopHorA R.-D., Mytospina R.-D., and DicHAarTomy1a Malloch. I have already dealt with the species of these genera known to me from Australia in this series of papers. Genus MetoromytiA Malloch. There is but one species of this genus known, atropunctipes Malloch. Subfamily MUSCINAE. This subfamily, as here limited, contains comparatively few genera, but some of these are of great economic importance as they transmit certain diseases or in other respects injure man and domestic animals. Unfortunately I have seen but few specimens of the subfamily from Australia and cannot give a list of the native species at this time. I give here a key for the separation of those genera which I believe must occur, or which I have definite records of as having been found in Australia. Key to genera. 1. Propieura hairy in centre, i.e. below the humerus and above the propleural bristles; fourth wing-vein angularly bent forward at middle of its apical section; suprasquamal ridge not setulose posteriorly ...................... J Musca Linné Propleuras not. Haired ini-Centrewrs sia ss ais foes coe lensaale a eee aakes smerec ameirecstar aperer Avekstis 2 2 suprasquamal ridge with setulose hairs posteriorly ............-.-+..+.-.-:s:.+- 3 SUpLasduamaleeridseenOEemsSetllosempOSterlonivaleoe merece ecient 4 38. Fourth wing-vein subangularly bent forward at middle of its apical section; species not metallic blue or green in colour .................. Viviparomusca Townsend. Fourth wing-vein with a rounded forward curvature at middle of its apical section; species metallic blue or green in colour ................+...-+.--- Orthellia R.-D. 4. Prosternum, pteropleura, and hypopleura bare ............... Balioglutum Aldrich. At least part of the hypopleura or pteropleura haired .....................---- 5 5. Pteropleura with ome or two bristles and some hairs above ..................... 6 ACER ODIE WTA WO ATC! wegeues ce saener cre eee eae ER een once REE SLE A ae sso to) 0) cower anes Gretened ap seiensuetieneleuens ce 8 6. Fourth wing-vein subangularly bent forward close to middle of its apical section; species not metallic in colour on any part of body; frontal orbits without strong forwardly directed bristles at middle; prosternum haired ........ Biomyia R.-D. Fourth wing-vein with a rounded forward curvature near middle of its apical section ; species blue-black, or metallic blue or green in colour; frontal orbits of female each with a strong forwardly directed bristle near middle; prosternum hairy or bare 46 NOTES ON AUSTRALIAN DIPTERA, V- 7. Species metallic blue or green in colour; mid tibia with a strong median ventral [MISES 2 THRORECIMNIAN [DERE sosascoddndevsoosanscedacnegdbanso00C Pyrellia R.-D. Species blue-black in colour; mid tibia without a strong ventral median bristle, some- times one present on middle of posteroventral surface; prosternum haired on a0 (ota Tee ne A PONE, EN oh: RRM Parole SiC! Cap Ots GO: aIG.O Gi oet OeNORAES O-c-0 Morelia R.-D. 8. Arista pubescent: prosternum hairy ....................... Synthesiomyia B. & B. ASHEN), TOETRORES S HAORKEIMUNTA [NBIKS soadaoccooncdcbdousocoD DDO C Goo NODS DONS ODUONGOD 9 9. Presutural acrostichals well developed; hairs on hypopleura below spiracle ......... bets cle eT hs hs Ae SEEMS fe, SEIS RA Seer Passeromyia Villeneuve. Presutural acrostichals not developed; hairs on hypopleura on upper margin in front of spiracle and also below spiracle .................... Graphomyia R.-D. Genus Musca Linné. I have insufficient material in this genus to justify me in dealing with the Australian species, but I suspect that only one species, domestica Linné, belongs here, though there may be another which I have not yet seen. Genus VIVIPAROMUSCA Townsend. This genus was erected for the reception of Musca vivipara Port., but I believe most, if not all, of the viviparous species belong in it. The exact distribu- tion of vivipara has yet to be determined, and it may be very widely spread. Genus OrRTHELLIA R.-D. This genus is known also as Cryptolucilia and Pseudopyrellia. My material does not permit an opinion as to the number of Australian species at this time. I have, however, seen Jawta Wiedemann from Australia. Genus Bromyta R.-D. One or two of the Australian species recorded as Musca belong here. Genus BariocLtutum Aldrich. Only one species, illingworthi Aldrich, belongs in this genus. It occurs in Queensland. Genus SYNTHESIOMyIA Brauer and Bergenstamm. One species of this genus is known, nudiseta van der Wulp. Genus PASssEROMyYIA Villeneuve. But one species of this nestling infesting genus, Jongicornis Macquart (heterochaeta Villeneuve), occurs in Australia. Townsend described it under the name victoria. Genus GRAPHOMYIA R.-D. One Australian species seen by me belongs to this genus. Genus PYyRELLIA R.-D. I have no species of this genus from Australia, but serena Meigen is known to occur. Genus Morerita R.-D. I have not yet seen any species of this genus from Australia. The species occur usually in the higher altitudes and it may be turned up in this continent yet. Subfamily SromoxyDINAE. I have seen only the genus Stomoxys from Australia, but present a diagnosis for the separation of this genus and Lyperosia. A. Palpi very much shorter than proboscis; centre of DEOpLem awh alte see POE D DODO OHO HAAS Hono ebIgd Spoon SdeE OCD ONoo aes OOS yCOOuCSIS Stomoxys Geoffroy. AA. Palpi as long as proboscis or almost so; centre of propleura bare ................ susneBeneielisdaiteHepe rene sone) © hojrerGuswemsuie cor sue fei anctohn CIRCE ee Meo Ree Lyperosia Rondani. A FEW OBSERVATIONS ON THE GEOLOGY AND GEOGRAPHY OF NORTH-WEST AND DESERT BASINS, WHSTERN AUSTRALIA. By FREDERICK G. CLAPP.* (Communicated by Professor Sir T. W. Edgeworth David.) (Plates xvi-xix and five Text-figures. ) [Read 29th April, 1925.] Introduction. During seven months of the year 1924 I had an opportunity of making nearly 6,000 miles of geological traverses in northwest Western Australia, and in so doing I visited some localities that had never been studied geologically and others that had been but little studied. In the absence of sufficient time to prepare a detailed account of geological and geographical conditions in the country traversed, some outstanding features are presented herewith for the information of future investigators. The study does not pretend to be a finished one in any sense, except insofar as it was carried to a logical conclusion in connection with certain areas under option to my client for the purpose of determining certain economic conditions. Time was so much a factor that I even found it necessary to pass entirely, or to make extremely hasty visits to important geological localities in the vicinity of traverses, for business reasons, and to omit entirely certain other localities which, from the viewpoint of pure geology, would have been considered essential to a solution of the problems outlined in this paper. If my discussions appear one- sided, if they are devoted to certain areas in contradistinction to others of greater geological importance, or if some hypotheses advanced are later found untenable, I hope, at any rate, that the information may prove helpful. The country is a vast one and little known, the Government Geological Survey is handicapped by insufficient funds, and the duty of a geologist is obviously to help the cause of science by placing on record any information at his disposal. The country investigated covers in a general way the greater part of what are known as Desert Basin and Northwest Basin (Fig. 1),+ both of which consist predominantly of Permo-Carboniferous strata (Figs. 2 and 3). My routes of necessity traversed rocks of this age and also overlying sediments as well as older systems on the borders of the two Basins. Vast areas were never visited, but * Published with the permission of Mr. A. E. Broué, of 11 Castlereagh Street, Sydney, N.S.W., for whom the investigations in Western Australia were undertaken. 7 Fig. 1, representing the positions of the basins, was taken from an old map and has not been corrected to include new discoveries in outline of Desert Basin, such as are described in this paper. GEOLOGY AND GEOGRAPHY OF NORTHWEST AND DESERT BASINS, MAP OF WESTERN AUSTRALIA OHOWING EXTENT OF KNOWN ARTESIAN BASINS Scale of Miles 3° 100 150 WO 250 300 330 400 5Oo 33 0 Text-fig. 1—General map of Western Australia, showing relative positions cf Desert Basin and Northwest Basin. BY F. G. CLAPP. 49 available information supplied by other geologists has been freely used. Desert Basin covers a large portion of Kimberley, Northwest and Hastern Divisions of Western Australia, whereas Northwest Basin is situated entirely in Northwest Division. This paper should be read together with frequent reference to the geological map of Western Australia, published as Plate I in Chapter I of The Mining Handbook, Memoir No. 1 (1919), by Mr. A. Gibb Maitland. Outline of physical features. The topographical features of Western Australia have been previously classified into (1) Coastal Plain, (2) Hill Ranges and (3) Interior Plateau. In the country traversed, the width of Coastal Plain is nil in places between Port Hedland and Onslow, but expands to 100 miles or more in Gascoyne District of Northwest Division. Between Broome and Wollal in Desert Basin, this plain merges with the Interior Plateau so gradually that no boundary can be pointed out, and the Hill Ranges are missing. In order to maintain a consistent basis of discussion, the Coastal Plain is here considered as being practically synonymous with the area of Tertiary age; nevertheless such anomalies exist as “Cape Range,’ west of Exmouth Gulf (Fig. 4 and Plate xvii, fig. 2), which is of Tertiary age and over 1,000 feet high, although no other land east of Hill Ranges has an equal height. Another incon- sistency in the above mentioned synonym is that, in Gascoyne and Lyndon Districts of Northwest Division, much of the country that is topographically Coastal Plain is of Jurassic age. Such Ranges as Kennedy, Carrandibby and others (Fig. 3) in Northwest Basin, and Dampier and Edgar Ranges in Kimberley are topographically part of the Interior Plateau and are of Permo-Carboniferous age. My observations lead to the opinion that the term “Great Plateau” or “Interior Plateau” should be used with discrimination, for, in both basins, it merges on the west with the Coastal Plain; and even over 100 miles in the interior, midway between Dampier Range and De Grey River, I have found the so-called “plateau” to be only 300 feet above sea-level (barometrically). General remarks on the geology. Although geological work in this portion of the State is in its infancy, many valuable reconnaissances have been made, and all explorers have contributed to present knowledge. Mr. H#. T. Hardman, pioneer observer and one-time Govern- ment Geologist, conducted explorations in 1883 and 1884,* on which subsequent mapping has been based; and his mapping is accepted in areas in which my observations overlapped his. The work of Messrs. Gibb Maitland and Talbot of the present geological survey has been of tremendous value, as have reports made by other men; in the employ of the Government; and certain private investigators, whose work has not been published and. whose names I am not at * Hardman, Edward T.: Report on the geology of the Kimberley District, Western Australia; published by the Government Printer, Perth, 1884, pp. 1-22, with 16 plates; also geological map published separately. 7 Woodward, H. P.: Geological sketch map showing the possible artesian water area between the Minilya and Ashburton Rivs., 1907 (in two sheets) ; and others. E 50 GEOLOGY AND GEOGRAPHY OF NORTHWEST AND DESERT BASINS, liberty in all cases to mention, have given valuable information”* as to conditions in portions not personally visited. Other publications, by Talbot, Jack and others, a valuable paper on the physiography of Australia by Prof. Griffith Taylor and a report by Gibb Maitland in the Proceedings of the Third Interstate Conference on Artesian Waters have been considered, but the references are not quoted, as the records are not now accessible to me. Nearly all of my work lay in the Northwest and Desert Basins, and the economic problems were limited to those basins; therefore I will refrain from discussion of observations outside them, except such as bear on the Basin problems. Explorations were limited to areas of Recent, Quaternary, Tertiary and Permo-Carboniferous sediments, overlapping in a few instances into granitic areas of unknown age, Nullagine and Pre-Cambrian sediments and metamorphics. The essential characteristics of these formations have appeared in an excellent summary by Gibb Maitland.+ General Similarity of Northwest and Desert Basins. One might suppose that Northwest and Desert Basins have few characteristics in common. For instance, they are separated by a belt of Pre-Cambrian rocks 350 miles wide where narrowest and which rise in places to over 4,000 feet above sea-level. Desert Basin is roughly quadrilateral in shape, with its longer diameter approximately at right angles to the coast; whereas Northwest Basin, like the “Coastal Plain Basin” farther south, constitutes a belt nowhere more than 150 miles wide, parallel to the coast from Onslow to Murchison River. Desert Basin is roughly 325 miles in a northeast-southwest direction by over 450 miles northwest- southeast. On the other hand, Northwest Basin extends 400 miles north and south by 75 to 150 miles east and west. A popular conception of Desert Basin is that it constitutes a desert, as the name implies. As a matter of fact, large areas of Desert Basin are so barren and sandy as to have practically a desert chareter, whereas the greater part of North- west Basin is prosperous pastoral country, in which two score artesian wells have been sunk, yielding some of the greatest flows in the Commonwealth and proving an area of perhaps 15,000 square miles to have artesian possibilities. It must be remarked by anyone who glances at the geological map of the State, that the principal stratigraphical divisions (Tertiary, Jurassic, and Carboniferous) in the two Basins are identical. Furthermore, the details of stratigraphy through- out portions of the basins examined by me are so closely identical as to leave no doubt that the basins were once connected, either across the intervening 350- mile wide Pre-Cambrian and Nullagine Plateau or through Indian Ocean around the north end of this Plateau. In both Basins the Permo-Carboniferous system has been sub-divided into (1) “an Upper or Sandstone group,’ made up almost entirely of sandstones and analogous porous sediments, and (2) “a Lower or Marine group” (in which limestones appear ta predominate). The limestones are interlaminated with sandstones, shaly sandstones, and thin shales. * Among these are unpublished reports by T. A. Blatchford, dated 20 Dec., 1923, by H. W. Talbot dated 19 Nov., 1928, by D. J. Mahony (undated), by Leo J. Jones and HK. De Villa. 7 Gibb Maitland, A.: A summary of the geology of Western Australia. The Mining Handbook, Geol. Survey Memoir No. 1, Chapter I, 1919, pp. 7-55, 80 figs. and geological map. BY FE. G. CLAPP: 51 North escarpment of the Interior Plateau in Desert Basin. The north escarpment of the Interior Plateau constitutes a vast geographical feature. Commencing 77 miles southeast of Broome, or 35 miles southeast of Hamilton’s well, the escarpment first appears merely as a breaking away of the coastal type of country, known as “pindan sands,’ which extend to that point from the inner edge of the “sea-level plains” near Broome. The pindan sands, covered with a thick growth of many varieties of trees and shrubs, rise from sea- level to a height of perhaps 200 feet at Hamilton’s well and to about 700 feet at the “breakaways” mentioned. A few low hills in the vicinity rise a few rods higher; hence the name “ranges” by which they are generally known. Some person more or less familiar with that part of the country has suggested the name “Dampier Ranges” to distinguish them from Edgar Ranges, situated farther east in the vicinity of Jurgurra (popularly called “Jeegully”) Creek, which [ did not have an opportunity of visiting. The claim of the height of land along the north escarpment of the plateau to the designation “ranges” consists, firstly, ina slight steepening of the northeastward rise of the Plateau along the line where the Desert sandstones rise from beneath the plains of the west and south, and secondly, in eresional dissection of the sandstone itself, where it drops abruptly 100 to 300 feet into “breakaways’” tributary to a broad valley, which thence descends slightly northward and con- stitutes part of the Fitzroy lowland. A generalized section across Desert Basin is shown in Fig. 2, from which the formations are seen to dip southwest at an angle of 1 to 2 degrees; hence the escarpment is typical of those the world over on up-dip flanks of structures exposed in high-lying sandstones. Calculations based on observed dips in the escarpment and those on the coast and in Mt. Phire (see below) lead me to the opinion that the strata of the escarpment and higher beds may have a total thickness of as much as 10,000 feet. Whereas the valley north-east of the escarpment, at the place, 77 miles south- east from Broome, where touched by my motor road from Broome to Camp 12, can be reached by an easy grade of not over 100 feet per mile, and whereas the crest of the Plateau is accessible on its west side by an even more imperceptible grade, the valley descends eastward and the “ranges” or Plateau crest rises somewhat southeastward, so that within a mile of the northwest termination of the escarpment it becomes nearly 100 feet high and a few miles farther southeast it is precipitous to a height of 200 feet or more. Many isolated tablelands with flat tops, a few acres up to several hundred acres in extent, stand in this valley, and are only higher than the main Plateau by an amount equal to the normal rise in dip toward the northeast (a few rods per mile). Maps of the Lands Department show several of these outliers—named Goorda Tower, Babrongan Tower, Mt. Alexander, Mt. Jarlemai, ete.—far to the northeast, and some of these were visible in the distance from points visited on the north edge of the escarpment. The rocks of the north Plateau escarpment are fine-grained white sandstones, somewhat aluminous, but never calcareous or even approaching the consistency of shale. In places they contain numerous ferruginous concretions. The dip is southwest and ranges from nothing up to 2 degrees. In a few localities, as at the head of a deep “‘breakaway” at Camp 9, 130 miles southeast of Broome, areas were seen up to 3 miles in length and a few hundred feet wide composed of a hard, light gray to white quartzite capping the Plateau a short distance back from the escarpment. It is only a few feet thick and takes the place of limonitic laterite which is an almost constant feature of any sandstone outcrop in the semi-desert 52 GEOLOGY AND GEOGRAPHY OF NORTHWEST AND DESERT BASINS, areas. Quartzite is frequently only a vitrified phase of the surface-hardened sand- stones and is common throughout Australia in rocks of all ages. From aerial observers and other sources I have received reports of a great ledge of gray quartzite that extends from a point on the Broome-Derby road some 60 miles northeast of Broome north to near Cape Leveque, and this last-mentioned quartzite may constitute part of the same series of rocks. SOUTHWEST NORTHEAST r 8 ) Cj o = N . 8 qT = Q feet yy ~ g 3 RR S Feat x > Ss S 1000 Ss u y 5S ~S fe sc s ¥ (000 Ry > a eee 3 Soe XY Ss SCALE OF MILES 50) 50 i00 150 SECTION ACROSS THE DESERT BASIN HYPOTHETICAL WHERE DOTTED Text-fig. 2.—Generalized cross section. of Desert Basin from Jurgurra Creek to Oakover River. SOUTHWEST NORTHEAST { 2 pas SN Qz : 2 x SLTASSG iL Seo Leve/ SCALE OF MILES Le) 50 190 150 a SECTION ACROSS NORTHWEST BASIN Text-fig. 3—Generalized cross section of Northwest Basin from Williambury Station to Carnarvon. In contour the escarpment is not simple, but extends south-east for many miles from its north extremity, and then bends east, forming an irregular line of cliffs throughout—vertical and inaccessible to a vehicle at any point, rarely scalable by man, and the crest of which is accessible to horses only at points dozens of miles apart. From information at hand, the line of cliffs appears to extend east over 200 miles, and it may merge with certain tablelands described by Talbot and others in the vicinity of Canning Stock Route. The “breakaways” in the vicinity of the 123rd meridian indent the Plateau in places for many miles back BY F. G. CLAPP. 53 of the escarpment proper. Curiously enough, all rock holes that contained water at the time of my visit are situated at the head of “‘breakaways,”’ except in a few instances where found in canyon-like gorges near the base of cliffs. No water holes, and only a few ‘native soaks,’ were found south of the escarpment, and very few natives live in that part of the Plateau. East escarpment of Kennedy Range. A comparison of Figs. 2 and 3 will show how similar are the formations and conditions in the escarpment described above to those existing in that of the east edge of Kennedy Range in Gascoyne District of Northwest Division. As in the case of Kimberley, so in that of Gascoyne, the escarpment rises vertically 200 to 300 feet, and apparently can not be ascended by vehicle or animal at any place from its south end, near Gascoyne River, to a point on Merlingleigh Station, a distance of about 50 miles. Even farther north, the cliffs are intermittently abrupt and precipitous for many miles. North and south from Kennedy Range its topographical features are repeated at intervals, represented by a smaller range directly south of Gascoyne River, by Carrandibby Range, 50 miles south, and probably also by Moogooloo Range, north of Minilya River; so that, for over a distance of 150 miles, the escarpment feature is prominent in some degree for intermittent, but long, distances. LANOVTH GULF UTA tg >2e2 as BULLERA STA G/RAL/A STA ANTICLINE Jand Kk ages evel SCALE OF MILES ° 5 10 15 —e— ee ———_——_—_ _ _ SSS OECTION NEAR EXMOUTH GULF (FROM INDIAN OCEAN ACROSS CAPE RANGE SOUTH 75° EAST TOWARDS WINNING POOL) Text-fig. 4.—Cross section in vicinity of Exmouth Gulf, showing Tertiary folding. As with the north Plateau escarpment, so with that of Kennedy Range, the strata consist of fine-grained, white, non-calcareous sandstones, which dip west at angles of less than 5 degrees with only slight reversals (no reverse dip of more than a fraction of a degree for a short distance being visible in the face of Kennedy Range). As with the north Plateau, so with Kennedy Range, the land descends gradually westward from an altitude of approximately 1,000 feet (Plate xix, fig. 1), directly on the escarpment, to the Coastal Plain several hundred feet lower. Somewhere between the west base of Kennedy Range and Indian Ocean, the Permo-Carboniferous rocks pass beneath the Jurassics, which in turn pass beneath Tertiary sediments to the west. Significance of certain fossil collections. A point that needs further field attention is the abundant occurrence of large pectenoid valves and others resembling Productus in ferruginous conglomeratic 54 GEOLOGY AND GEOGRAPHY OF NORTHWEST AND DESERT BASINS, sandstones of hematitic hardness in the semi-desert area about 185 miles, by my tractor road, southeast of Broome (Fig. 2), estimated only a few miles west of McLarty Hills as mapped. This fossil locality lies between 83rd and 87th sand- ridges, counting from the north, and not far from my Camp 12. No outcrops exist between here and the north Plateau escarpment, but those mentioned are plentiful throughout an area of several square miles at an elevation of about 300 feet above sea-level. The fossils are reported by Mr. W. S. Dun to be not definitely determin- able, but of probable Permo-Carboniferous age. The outcrops are covered with dark red glaze that takes the place of ordinary laterite in dry arid regions in which the rocks contain much ferruginous matter, and here.they are as hard as many regionally metamorphic rocks, although vitrification is due to atmospheric and not to tectonic agencies. Similar glazing to that which is so prominent near Camp 12 is observable in King’s Peaks (two ferruginous sandstone hills about 100 feet high) on the coast roughly 60 miles north of Broome, between King’s Peaks and Pender Bay, and in the country between Lagrange and Anna Plains Station back of Ninety Mile Beach. In the last-mentioned locality, one fossil species was found by Mr. B. BE. Bardwell and presented to me, but its identity has not been determined. Glazing of the same type is reported by Mr. Gibb Maitland in similar rocks, from an altitude of 1,200 feet above sea-level at Trig. Station K37, on top of the south end of Kennedy Range, just north of Gascoyne River, where he collected unnamed species of Spirifera, Productus, Athyris (?) and Strophalosia. Since outcrops of similar character overlie the Desert Sandstone escarpment in the north and occupy a similar position on Kennedy Range, the results of the determination of the species collected near Camp 12 (if determinable at all) are awaited with interest, to learn whether any further evidence is afforded of the continuity of sedimentation from one Basin to the other, either across the inter- vening Plateau or through a submarine connection. Although not necessary for a consideration of any particular problem discussed herein, a publication of the names of identified Permo-Carboniferous species collected on the recent expedition, and determined by Mr. Dun, may be of value. The determinations were made through the agency of Prof. Sir T. W. Edgeworth David and the courtesy of Mr. BE. C. Andrews, Government Geologist of New South Wales. The localities are believed to be arranged in stratigraphical order from below upwards (from i to 5), but this order is by no means certain. All of the species are believed to be from the so-called “Lower or Limestone series” of the Permo-Carboniferous. List of identified species of Permo-Carboniferous fossils collected on the expedition to Western Australia. (Determinations by Mr. W. S. Dun.) 1. From a persistent earthy crinoidal limestone outcropping 6 miles northwest of Lyons River Homestead, between Lyons River and Kennedy Range on road from Lyons River Homestead to Mt. Sandiman Station shearing-shed. Amplexus pustulosus Hudl. Crinoid stem ossicles. Fenestella, sp. undescribed. Phyllopora, sp. undescribed. Stenopora, 2 species. Spirifera marcoui Waagen (sp. by Etheridge, Jr.). Spirifera musakheylensis (HMth., Jr., not Davidson). BYe ho Ga GmAPe 55 List of identified species of Permo-Carboniferous fossils collected on the expedition to Western Australia—continued. Spirifera cf. lata McCoy. Sptriferella australasica Etheridge, Jr. Cleiothyris macleayana Etheridge, Jr. Chonetes pratti Davidson. Productus spines. Pleurophyllum sp. Monilopora nicholsoni, Etheridge, Jr. 2. From limestones lying a few feet above Lyons conglomerate, 28 miles southeast of Gascoyne Junction, on road to Dairy Creek Station. Chonetes pratti Davidson. Productus, sp. nov. (will be named P. clarkei). Spirifera, sp. undetermined (musakheylensis ?). Spirifera, cf. marcowi Waagen. 3. Believed to be from same locality as No. 2 above. Stenopora, dendroid. Fenestella. 2 species. Aulosteges baracoodlensis. Productus, sp. nov. (will be named P. clarkei). Athyrid (Cleiothyris macleayana ?). 4. From flaggy sandstones, 5 miles west of Gascoyne Junction, on road to Carnarvon. Productus pellus. Spirifera cf. lata (Etheridge and Foord, not McCoy; will be renamed maitlandi). Pleurophyllum australe Hinde. af Chonetes sp. (pratti ?). 5. From well, 5 miles northeast of Merlingleigh Station (temporary camp) and east of Kennedy Range. Spirifera lata Foord, not McCoy. Spirifera cf. musakheylensis. Chonetes pratti Davidson. 6. From Merlingleigh Station, well, at temporary camp at foot of east side of Kennedy Range. Spirifera cf. musakheylensis. Spirifera marcout. A comparison of the strata in the east escarpment of Kennedy Range with those in the north escarpment of Desert Basin is interesting. In both instances the rock is essentially a fine-grained white or very light gray, frequently soft, porous rock, mainly siliceous to chalk-like, with no ascertainable calcareous admixture. The rock is finely stratified and contains numerous ferruginous and cherty concretions parallel to the bedding. Both the Desert Basin formations and those of Northwest Basin are overlain sea-ward by strata of supposedly Jurassic age, often having similar characteristics to the Permo-Carboniferous and not to be differentiated on lithologic evidence alone. Structure of beds underlying Plateau beds. The Permo-Carboniferous age of the two Plateaus, as well as that of outcrops for many miles east and north respectively, has been proven, by Gibb Maitland, on fossil evidence alone. No unconformity has been reported in the series. Never- theless, the discordance in character and structure of the sub-plateau beds from those constituting the Plateaus is so striking, both in Northwest and in Desert Basins, that, in my opinion, the possibility of the existence of a mid-Permian unconformity should be suggested and tested by exhaustive field examination. In numerous outcrops examined from Gascoyne Junction, east of Kennedy Range, north to Williambury Station (roughly 85 miles), and southeast to Dairy 56 GEOLOGY AND GEOGRAPHY OF NORTHWEST AND DESERT BASINS, Creek Station (50 miles), the rocks dip occasionally in directions opposed to normal, and in places they dip as much as 8 degrees toward the east. In travelling over the motor road from Gascoyne Junction to Dairy Creek Homestead an anti- clinal axis is crossed where, a few miles west of that homestead, abnormal and reverse dips cover a breadth of several miles. A similar fold was seen on north side of Gascoyne River between Arthur and Wyndham Rivers, and may or may not be identical with the axis mentioned. Slight east dips were seen at several points between Kennedy Range and Lyons River, between Lyons and Minilya Rivers and they are reported by Mr. E. De Villa and others on Minilya River (in unpublished manuscript). The best dome was seen in the southwest part of Williambury Station, 14 miles west of Moogoorie Homestead (Fig. 5 and Plate xix, fig. 3). This locality is situated 9 miles from the normal outcrop of Lyons conglomerate; yet this is domed upward, together with its overlying sandstones, projecting above the normal surface in a central valley almost entirely surrounded by outcrops, dipping outwards and consisting largely of cliffs up to 50 feet in height. & 3s Ss oy Sg a) ‘o yO v vy & g¢ ee g8 » 8 & x > SS ae SS Sms gy NQ v/ S WEST EAST DES | ee iN 9 -< - Ons” Conglomerate G OCALE ) INCH = 3° MILES, HORIZONTAL | INCH = 300 FEET, VERTICAL SECTION NORTH OF KENNEDY RANGE, WEST OF MOOGOORIE STATION Text-fig. 5—Cross section of doming at point on Williambury Station, at north end of Kennedy Range. Probable Permo-Carboniferous of Dampier Land. Although the western edge of the peninsula (once known as Dampier Land) west of King Sound is Jurassic, as mapped, I am under the impression that the balance of the peninsula may be largely of similar age to the Plateau escarpment, which is believed to be Permo-Carboniferous. One geological observer is convinced he saw folds in a sandstone north of Wanganut well, on the road from Broome to Beagle Bay Mission. My own impression is that the beds at the particular point are of duller shades than the Plateau beds of the coast and west of Wanganut well, and their elevation of several hundred feet gives the impression that they may constitute a fold against which Jurassic sediments were deposited to the west and south-west. The coarse ferruginous conglomeratic sandstones of King’s Peaks (some 60 miles north of Broome on the coast), between Carnot and Beagle Bays, BY F. G. CLAPP. 57 with their peculiar glaze and hematitic veining, suggest the beds that have been called Permo-Carboniferous near Camp 12 and back of Ninety Mile Beach. Similar hard coarse ferruginous rocks extend near the coast from King’s Peaks to Beagle and Pender Bays, but no fossils could be found in them. Remarks on Lyons conglomerate. The Lyons conglomerate (Plate xix, fig. 2) is one of the most interesting formations in Northwest Basin. It has been so fully described (Gibb Maitland, op. cit., pp. 34-35) that I need only remark that it is an immense boulder bed— an actual tillite—composed of innumerable ice-scratched pebbles and boulders and having other essential phenomena, extending from Lyndon River, with some interruptions, south for a distance of 400 miles; and its actual outcrop can be traced in Lyndon and Gascoyne Districts for about 200 miles. In considering the similarity of stratigraphic conditions in the two Basins it may be well to remark that Talbot and others have found a boulder bed in the Fitzroy Valley in the north part of Desert Basin at about the same stratigraphical position, and that a previously unknown tillite locality was recently discovered by myself at Braeside Station, south of Desert Basin. An interesting feature of the Lyons conglomerate is its presence in the centre of an almost perfect dome, some 9 miles west of its normal outcrop, at a point on Williambury Station, 14 miles west of Moogoorie Homestead and a short distance south of Minilya River. The east side of this dome is shown in Plate xix, fig. 3 and a cross section of the locality appears in Fig. 5. Other Conglomerates of probable Permo-Carboniferous age. On Moogoorie Station, intermediate between Minilya and Lyons Rivers, certain conspicuous outcrops of hard, coarse, quartzitic conglomerate occur below the “limestone series” and are almost certainly of Carboniferous or Permo- Carboniferous age. Similar conglomerate, resting on granite, is described by Gibb Maitland (loc. cit., p. 36) at Warrie well, on Towera Station on Yannerie River. Similar outcrops were also seen by me on Pardu Station, near the south- west corner of Desert Basin, and there seems no reason for supposing their age to be other than Permo-Carboniferous, although they might conceivably be Jurassic. The Pardu conglomerates are very ferruginous over many square miles, but an occasional white, unoxidized exposure classes them with a vast succession of white siliceous rocks of varying texture. They bring the known rocks of Desert Basin to within 50 miles of De Grey River or 40 miles farther southwest than reported on the published geological map. The contiguous area of granite is found to be narrower than mapped near the coast, but 100 to 150 miles southeast it is much wider than mapped, even including some of the area mapped as Carboniferous east of Paterson Range. Features in common to both Basins. Summarizing the evidence given, it appears that the north escarpment of the interior Plateau in southern Kimberley Division is similar, geologically and geographically, to that bounding the east face of Kennedy and Carrandibby Ranges in Gascoyne and Lyndon Districts of Northwest Division in the following respects: 1. In its height of 100 to 300 feet, either continuously (as in Desert Basin) or throughout intermittent stretches of 20 to 50 miles each, separated by low-lying areas (as in Northwest Basin). 58 GEOLOGY AND GEOGRAPHY OF NORTHWEST AND DESERT BASINS, 2. In a characteristic whiteness of the escarpment, i.e., in its texture of generally fine (but locally coarse) non-caleareous sandstone. 3. In the fact that both Plateaus are surmounted by sand-dunes, situated on the Plateau edge in part (in Kennedy Range), but 30 miles or more back from the escarpment elsewhere (in Desert Basin). 4. In a characteristic vegetation of spinifex, interspersed with shrubs and low trees, interrupted by sandy patches, but commonly having suitable feed for cattle or sheep. 5. In a continuous dip of the strata from the escarpment into the Plateau, until the latter merges with the low plain to the south or east respectively. 6. In the occurrence of small springs of fresh water on the down-dip side of the “ranges.”” The only genuine springs of which I know, in the western part of Desert Basin, are situated 45 miles east of Anna Plains Homestead and have been rarely visited; and in Northwest Basin many springs emerge along the west side of Kennedy Range. An interesting fact is that only saline waters are commonly found in bore holes 50 to 100 feet deep on the low plain directly east of the Kennedy and Carrandibby Range escarpments and that saline springs flow con- tinuously from a few spots on that side of the ranges and directly north of Gascoyne River, That is, fresh water exists in outcropping strata on the down- dip side of Kennedy Range and saline water on its up-dip side. A “salt marsh” is reported by the Lands Department over a practically sea-level plain, situated, roughly, 35 miles southeast of Anna Plains Homestead in Desert Basin; and wells sunk on sea-level plains between Edgar and Pardu Stations (back of Cape Villaret and Ninety Mile Beach respectively) are frequently saline, but this salinity is attributed to another cause than those pertaining to wells in Permo-Carboniferous strata. 7. The beds of both escarpments are flat or have a dip of less than 5 degrees, and they generally dip only 1 or 2 degrees in the direction of normal dip; in contrast to lower rocks of the country east of Kennedy Range escarpment and north of Desert Basin escarpment, where dips of 5 degrees and upwards are common and reverse (east and north, respectively) dips are not unknown. Only in one or two instances are reverse dips (those contrary to the direction of normal dip) found in either escarpment; but reverse dips of as much as 10 degrees have been seen in some instances, and those of over 1 degree in many instances, in the lower beds east of the escarpment in Gascoyne and Lyndon Districts of Northwest Basin. Reverse dips of greater angle are reported in Fitzroy Valley by Messrs. Talbot and Blatchford in unpublished private reports, and by other geologists in informal verbal communications. 8. The characteristic colour of the beds in the escarpments is white (except where ferruginous beds have been reddened owing to oxidation), whereas the lower series of sandstones and limestones appears to be almost any shade of gray, but seldom white throughout any considerable thickness. 9. No limestones or shales are known in either escarpment, but thin shales and great thicknesses of limestone, as well as of sandstone, abound in the lower land to the east (in Northwest Basin) and north (in Desert Basin). 10. Instead of being uniformly devoid of fossils, like the sandy escarpment beds throughout hundreds of feet vertically, the lower calcareous strata abound with fossils. BY F. G. CLAPP. 59 Distribution of Jurassic strata. On the published geological map, an area of Jurassic rocks is represented in western Desert Basin. It includes a small patch near Derby, nearly the whole of the peninsula west of King Sound, and a belt along the coast that has a diminish- ing breadth of from 40 miles near Broome to nothing at Wollal. In Northwest Division the Jurassic system consists of a belt of variable width extending south from Onslow, across Lyndon and Gascoyne Districts to beyond Wooramel River— the approximate southern limit of the area studied. I have no definite suggestions for adaitions or corrections to the map, so far as Northwest Basin is concerned; although I have an impression that a narrow belt of Permo-Carboniferous rocks may possibly be traceable from Yannerie River (the farthest point north at which it has been reported and mapped) to Ashburton River and thence somewhat east of that river to the coast. In Desert Basin some suggestions may be of value. As stated by Gibb Maitland (op. cit., p. 41), Jurassic rocks “‘may extend some distance into the Interior” east of Ninety Mile Beach. The Jurassics are not continuous, evidenced by the dis- covery of probable Permo-Carboniferous fossils in the characteristic glazed red conglomeratic sandstone on Anna Plains Station, the frequent occurrence of this rock north to beyond Lagrange, and the discovery of abundant Permo-Carboniferous (?) fossils in similar rock near Camp 12, west of McLarty Hills, as described in this paper. The identification of Jurassic strata in western Desert Basin is largely dependent on the discovery by Gibb Maitland (loc. cit.) of Belemnites 1,30uU treet below the surface in bore No. 2 at Broome... The mapped extensions of Jurassic north and south of Broome are doubtless based largely on the occurrence of certain outcrops—similar to those at Pt. Gantheaume and Hntrance Point near Broome— along the coast at widely separated localities, viz., at Cape Borda (according to a verbal report made to me by a geologist who visited it) and Cape Villaret, 25 miles southwest of Broome. The predominant rocks at these localities are fine-grained white sandstones in which small flakes of mica are prominent. At Pt. Gantheaume, 44 miles west of Broome (Plate xviii, fig. 2), the outcrops beneath the laterite covering, in cliffs 60 feet high, consist of fine to medium grained, somewhat micaceous sandstone, cross-bedded to regularly bedded, in which the dip averages 1 to 2 degrees in a direction S. 60° W. At Entrance Point, somewhat nearer Broome, in cliffs 30 feet high, the strata are similar, light gray to white, hard to soft, regularly bedded sandstones. At the south end of Cable Beach, 3 miles north of Pt. Gantheaume, underneath laterite deposits, are a few outcrops of white, micaceous, fine to medium grained, regularly bedded sandstone, fiat or dipping very slightly southwest. Mica flakes are also abundant in white sandstone thrown out of Hamilton’s well, 42 miles southeast of Broome, but the rock in other respects resembles that of the Plateau escarpment. At Cape Viilaret, 25 miles southwest of Broome, the 70 foot cliffs consist in part of Recent consolidated sand-dunes (Plate xvi, fig. 2), but in other portions they are fine-grained, white or oxidized, hard sandstone, with some coarse sand- stone layers, some thin bands of which contain mica flakes and some hematitic beds, weathered into fantastic forms and surface hardened as usual (Plate xviii, fig. 1). Coarse, white, unoxidized sandstones are interbedded with the fine-grained beds. The dip averages 14 degrees in a direction N. 15° W. Cape Villaret is the last coastal outcrop to the south at which I saw charac- teristic Jurassic material in Desert Basin; but Mt. Phire, 15 miles east of Anna 60 GEOLOGY AND GEOGRAPHY OF NORTHWEST AND DESERT BASINS, Plains Homestead, and 50 miles south of Lagrange, I found to be an unmapped table-top group of five hills (Plate xvi, fig. 3), distributed over 2 square miles, supplemented by an isolated hill 2 miles farther east. These hills are interesting physiographically, being the only known summits north of Callawa Hills (see herein below) that stand above the regular level of the “pindan sands,’ except an isolated tower-like hill with steep sides visible from Mt. Phire in the far distance and estimated 30 miles N. 35° EH. The height of the Mt. Phire group is estimated at 100 to 130 feet above the low plain. The hills are capped by fine-grained white micaceous sandstone and hard semi-porcellanitic material alternating with, or underlain by softer chalky beds. The dip is 14 degrees in a direction N. 65° W. The upper 10 to 30 feet consists in places of red massive sandstone, resting with very slight unconformity on finer grained beds (Plate xvii, fig. 1). Owing to the rough regularity in height of table-tops (Pt. Gantheaume, Capes Borda and Villaret, Mt. Phire and the unvisited peak to the northeast), all 70 to 130 feet above other outcrops, and the presence of these rocks lying practically flat within a few miles of the Broome bore in which Jurassic fossils were found, they all appear to be of probable Jurassic age. Other outliers, similar to Mt. Phire and the unvisited peak, may be dotted over the southwest part of Desert Basin, but they are as yet undiscovered. Probable Jurassic extension south of Desert Basin. Near the 121st meridian west of Oakover River, stands a persistent tableland that has been traced from near the junction of Oakover and Nullagine Rivers south to Braeside Station and thence seen to extend out of sight far to the southward. Although only a few miles in breadth, it rises from the valleys abruptly 100 to 120 feet, is capped by hard, white, frequently opalescent quartzite, often over 30 feet thick, and generally lies nearly flat. This tableland is doubtless the one mentioned by Gibb Maitland (op. cit., p. 47) as occurring near Carawine Pool and suspected by him of being Tertiary; but in conversation he admits that its age may be Jurassic, as the evidence seems to be absent. On my traverse south from near the junction of Oakover and Nullagine Rivers, this tableland was paralleled almost continuously, and glimpses of it were also obtained far to the north, where it enters Desert Basin and is apparently continuous with Callawa Hills. The last-mentioned have not been visited by me, but their near continuity and similar topography are worthy of note. A’ reason- able assumption is that Callawa Hills may be of identical age with Mt. Phire, which is similar topographically; and if this correlation be true, the age of the tableland of the Nullagine-Oakover divide is also probably Jurassic. Whether or not this be so, the tableland of the Nullagine-Oakover divide constitutes an outlier of Desert Basin rocks resting on Nullagine and older rocks south of the Basin proper. Braeside (tillite) formation. Less than a mile east of the tableland above described, between it and Oakover River, and situated 2 miles north of Braeside Homestead, more than 100 feet below the tableland and intermediate between it and the valley bottom, I found an unrecorded tillite deposit. It forms gentle slopes throughout an area of perhaps 100 acres and consists of soft, light gray, granular, siliceous and aluminous material capped by a thin bed of harder coarse conglomerate. The white material is filled with erratic pebbles of granite, gneiss, quartz, sandstone, shale, schist, etc., BY F. G. CLAPP. 61 but none of the peculiar opalescent quartzite from the tableland to the west was found in it. The pebbles are frequently finely striated, and are in size up to a foot in diameter. Some are rotten, owing to weathering, although the surface of an individual pebble appears very fresh. My impression is that this tillite is of pre-Jurassic age, reasoning from its topographic position with reference to the tableland, and also from the fact that no pebbles of Jurassic (?) quartzite were found in it. Otherwise it might be supposed equivalent to the Wilkinson Range beds of tillite found by Talbot in Princess Range, near Lake Carnegie, 350 miles south, and provisionally assigned by him to the Cretaceous system.* If the tableland of the Oakover-Nullagine divide be of Tertiary age, as was originally suspected by Gibb Maitland, the Braeside tillite may be of Cretaceous age; otherwise it is probably Permo- Carboniferous, constituting the first discovery of Permo-Carboniferous tillite on the south flank of Desert Basin. The difficulty of accepting a Permo-Carboniferous age lies in its quite different character from the very coarse, relatively dark Lyons conglomerate of Northwest Basin; but, as has been well said, “‘lithologic character is no criterion for the age of a tillite.” The tablelands and their significance. The common designation “Great Plateau of Western Australia’? seems, to a stranger in the country, to be too generalized and often used loosely. Of course, jn ascending the north escarpment in Kimberley Division at any one of many points, and travelling along it or across it for hundreds of miles at never less than 700 feet above sea-level, as has frequently been done farther east, one feels uncon- sciously that the Plateau is a unit. But, as was learned on the recent trip in travelling south from the escarpment, the plain slopes south from this elevation to roughly 300 feet above sea-level in inter-ridge valleys near the Fiftieth sand-ridge, 40 miles or so south of the escarpment and 170 miles more or less southeast of Broome. The occurrence of supposed Permo-Carboniferous fossils in these valleys 185 miles southeast of Broome and estimated 130 miles eastsoutheast of Anna Plains Station is some index of the age of the beds forming the Plateau; but the fact that Cape Villaret and Mt. Phire rise 70 and 130 feet respectively above the coastal Permo-Carboniferous (?) plain (Plate xvi, fig. 3) establishes them as remnants of a higher Plateau, the beds of which are probably of Jurassic age. It is this last-mentioned Plateau that may once have been continuous with that of Callawa Hills and the tableland of the Oakover-Nullagine divide. The two plateaus—a widespread one composed of Permo-Carboniferous strata and probably Tertiary peneplanation, and a more fragmentary one composed of Jurassic beds— should be carefully differentiated in any consideration of the physiography of northwest Western Australia. Some facts may seem to warrant the assumption that the sandstones of the north Plateau escarpment, and hence those of the great central portion of Desert Basin, are of Jurassic age, instead of Permo-Carboniferous; but Mr. Dun’s opinion that the fossils from Camp 12, 185 miles southeast from Broome, are probably Permo-Carboniferous, does not substantiate this view; hence we can continue to consider the north Plateau escarpment and the centre of Desert Basin to be Permo-Carboniferous as mapped. * Talbot, H. W. B.: Geol. Survey W. Aust., Bull. 83, 1920, pp. 59-60, fig. 16 and frontispiece. 62 GEOLOGY AND GEOGRAPHY OF NORTHWEST AND DESERT BASINS, Turning now to Northwest Basin, there is positive evidence (in the form of fossils collected by Gibb Maitland from Trig. Station K37, on top of the south end of Kennedy Range) that this range, together with Carrandibby and Moogooloo Ranges, is of Permo-Carboniferous age; and, since these hills have roughly an altitude of 1,000 feet above sea-level, they should be considered contemporaneous with the north Plateau escarpment. Near Winning Pool, southeast of Exmouth Gulf, some white, chalk-like hills, 50 to 100 feet high, have an east-facing escarpment and are mapped as Jurassic. These perhaps may be correlated with the outliers of supposed Jurassic age in western and southern Desert Basin; although in character they resemble the north Plateau escarpment and some beds in a ravine west of Wanganut well between Broome and Beagle Bay (southwest of the outcrops described hereinabove as Permo-Carboniferous). Higher outliers a few miles southeast of Winning Pool have not been visited. Far to the west of Winning Pool, on Giralia and Bullera and west of Exmouth Gulf, widespread folding of probable Tertiary age has occurred, so that Tertiary sediments are much higher in elevation than are any of the tablelands mentioned. Distribution of Tertiary and Post-Tertiary strata. The Tertiary strata of Northwest and Kimberley Divisions have received little attention. Some of the outcrops of low-lying limestones and calcareous sandstones along the coast are undoubtedly Quaternary and Recent. Careful study of the narrow coastal belt mapped as Tertiary and Post-Tertiary between Roeburne and Broome might lead to interesting differentiations. My own observations include the following: 1. Unconsolidated sand-dunes and sand-ridges.—These are the most recent formation in Western Australia and are still being deposited, not only on the coast (Plate xvi, fig. 1) but also over large areas in the interior. In my traverse south from the north Plateau escarpment, eighty-five of these sand-ridges were crossed, each one constituting an entity, very persistent, and few of them having any ascertained termination. Their trend varies with few exceptions between S. 75° W. and N. 75° W.; they range from a few feet to 60 feet in height, measured from the base of sharp ascent, but are much higher if referred to the centre of the inter- ridge valleys. The width of the ridges varies from 150 to 1,500 feet, and they are often composite, rough and hilly, with several crests. Most of them are covered with a growth of hummocky spinifex, but they have a greater proportion of sand than spinifex surface, and they support a few trees, as do also the valleys between them. The most interesting features of these sand-ridges are their great persistence, regularity of trend, and the fact that the valleys between them are only slightly encumbered with deep sand, so that an ordinary motor car, once arrived between any two of the ridges, could travel for long distances between them, by jolting over the spinifex clumps that are rarely absent from any square rod of surface. Little possibility exists of a motor car crossing these sand-ridges without special appliances for the wheels, which sink deep into the sand, but caterpillar tractors can cross them if well supplied with petrol, water and spare parts. The sand-ridges are not only present in the interior of the Plateau, but they strike the coast throughout a belt roughly 70 miles wide between Wollal and Pardu Homesteads and render motor travel along the main north-south road very difficult. In Northwest Basin similar sand-ridges rest on top of Kennedy Range (Plate xix, BYE GaE CMAP: 63 fig. 1) and they form high barriers to the good grazing country west of Exmouth Gulf. 2. Sea-level plains, including Roebuck, Anna and other Plains, at sea-level or within 5 feet of high water mark.—These are formed of silt and are saliferous and full of Recent marine shells. In the country north of Broome, dead cajiput trees, on and behind these plains, signify that the land is now sinking. The plains extend at times as much as 30 miles inland from the coast, fringed and backed by “pindan sands.” The sea-level plains appear to be, next to unconsolidated sand- dunes, the most recent formation of southern Kimberley Division. The following Post-Tertiary species have been identified by Mr. W. S. Dun from collections by me near Coolmakop well, 10 miles southwest of Male Station, on the southeast edge of Roebuck Plains, 30 miles southeast of Broome: Post-Tertiary species collected on Roebuck Plains. Arenlaria venuste Dunke. Cassidula angulifera Petit. Cerithoidea obtusa Wood. Cerithoidea kieneri. Ellobium auris-judie Linn. Melampus flexuosus (?). Plectotrema sp. (?). Arenlaria optima. Melaraphe scabra Linn. Nerita albicilla Linn. Pyrazus fluviatilis. Corbula crassa Hinds. Arca granosa Linn., var. cuniata Reeve. Pupoides pacificus Pfeiffer. Succinea operta Cox. Succinea scalarina Pfeiffer. 3. Consolidated sand-dunes.—Not necessarily older in every instance than the unconsolidated sand-dunes, yet consolidated to the consistency of sandstone; some- times cut by the sea waves and forming vertical sections 10 to 70 feet high on the coast from Broome southward. The cross-bedding and typical dune formation are illustrated in Plate xvi, fig. 2. 4. “Coquina” or shell rock.—Deposits of shells, 20 feet in thickness, extending co that height above high water mark near Kamelin Pool, and lower deposits of shell rock on the coast at Port Hedland and Broome. 5. Raised sea-beaches.—These consist of sandstone, limestone or comminuted masses of shells and sand-grains. Raised beaches have been seen at frequent intervals from Port Hedland to Lagrange, and they sometimes rise 50 feet above tide level. In topography they take the form of beach-like belts sometimes paralleling the coast for many miles. The surface of these beds is frequently considerably hardened and the outcrop has a concentric pseudo-folded appearance. The best exposures were seen 2 miles southwest of the southwest end of Ninety Mile Beach on Pardu Station, where a ragged limestone outcrop, 50 feet high, composed of shelly calcareous sandstone, forms the coast for a mile or two. At Port Hedland similar outcrops rise 10 to 15 feet above high tide. If the dead trees surrounding the sea-level plains north of Broome indicate a sinking of that part of the coast at the present time, the raised beaches indicate a recent rising of the coasts in the area from Lagrange southward. The pivot appears to be not far from Broome. 64 GEOLOGY AND GEOGRAPHY OF NORTHWEST AND DESERT BASINS, 6. A zone of red soil—This has been seen in a few localities near Broome and is exposed a few feet in thickness underneath the “‘petrified”’ coastal sand-dunes at Cape Villaret. The red soils are Recent or Quaternary in age. 7. “Pindan sands.”—Practically all of the country round Broome, with the exception of the sea-level plains and the small areas of rock, is known as the “pindan sands.’’ Such areas consist of level to gently rising unconsolidated sands, often rising to several hundred feet above sea-level at distances from the coast, and in general sloping upwards against older formations. The “pindan sands” are generally covered with a thick growth of shrubs or low trees of many varieties, known collectively as “pindan.” These sands are probably of various ages from Permo-Carboniferous to Recent, and can not be differentiated until enough wells are sunk in them so that fossils shall have been found and determined at wide- spread localities. The “pindan sands” cover nearly the whole of the peninsula west of King Sound, most of the country east of Broome, north and south of the Derby road, and extend southeast from Broome from the inner edge of Roebuck Plains to a distance of over 100 miles and beyond that until they merge with the sand-ridge country of the semi-desert; and they also occupy the inter-sand-ridge spaces. Most of the country from Broome south to Port Hedland that is not a sea-level plain is a part of the “pindan sands’; and in fact the term is doubtless applicable much farther south, except that the word “pindan” is apparently limited to the North. The “Cape Range” formation. The most unexpected discovery of any relative to the Tertiary system was in the “Cape Range,” extending south from Northwest Cape, where white limestones and interstratified chalky beds form a great anticline rising from below sea-level on the west side of Exmouth Gulf to a height of over 1,000 feet in the centre of the Range (Fig. 4), intersected by deep gorges extending back miles into it. Some beds of the chalky material are full of foraminifera, as yet unidentified, and only surmised to be of Tertiary age.* The east dips vary from nil on top of the Range to 8 degrees on the lower east flank (Plate xvii, fig. 2). Far up a gorge in the Range, at a point 15 miles south of Northwest Cape, I saw the dips flatten out and then dip toward the west at an angle of 2 degrees; but the gorge was not followed farther west. Rocks are also reported to dip sea-wards at Pt. Cloates, 75 miles south of Northwest Cape, on the west side of the Range. I infer that the Range is an anticline, because, firstly, west dips of 2 degrees were actually observed along a line not far from the centre of the mountain range; and secondly, if the Range were a monocline it would imply a shore line to the west, which has not been proved or even suspected. Hast of the anticline of Cape Range other anticlines were found, one of which, on Giralia Station, 20 miles east of Cape Range, has a height of at least 300 feet and a breadth of 10 miles. For the strata that comprise these anticlines and form the surface between them, I suggest the name “Cape Range” formation. The foraminifera have not yet been determined, but it seems probable that they will show a Tertiary age. * Note, 29th March, 1925.—Word has just been received from Professor Sir T. W. EH. David that Mr. F. Chapman states emphatically that these foraminifera are Oligocene types of Lepidocyclina and Cycloclypeus. The above is therefore an important discovery of raised and flexed Oligocene limestones in Western Australia.—EbD. BY F. G. CLAPP. 65 Large areas along the coast and for some miles back from it between Northwest Cape and Hamelin Pool may be part of the same formation. Country east of Onslow. An interesting section, 7 miles in breadth, of some of the very old rocks, exists in the interval from about 35 to about 42 miles east from Onslow. More or less continuous outcrops were passed on a northeast-southwest traverse, all dipping at an average angle of 20 degrees in a direction S. 60° E. Thus, the strata, if not duplicated by unseen faults, appear to be perhaps 13,000 feet in total thickness. They consist of rather uniform, hard, light gray, thick-bedded quartzites, with some medium-grained sandstone and traces of softer white material below the hard surface. Although this belt of outcrops appears to lie on the strike of the Mosquito Creek beds (pre-Cambrian) mapped less than 100 miles to the south, the rocks are perhaps similar in character to some of the Permo-Carboniferous rocks when surface-hardened. The locality was not observed to be cut by quartz veins or by ‘granite, like the Mosquito Creek beds seen elsewhere; and the only intrusion noticed was a large dyke of basalt, apparently occupying the centre of an anticlinal fold just west of Peedamullah Homestead. Similar gray quartzites of similarly great thickness, seen in the area between Nanutarra and Uaroo Homesteads, south of Ashburton River, are underlain by at least 2,000 feet of schists. The formation south of Nanutarra is folded into several great anticlines, in the centre of which valleys have been eroded; but in that area the quartzites and schists are intruded by innumerable quartz veins, granites and gneisses, so that here, as in the De Grey River country, certain granites can be proved younger than certain sedimentaries. Conclusion. The matters discussed in this paper constitute merely a few of the problems existing in the northwest of Western Australia. Other problems are (1) Climatic factors in geological history, (2) Number of glacial periods, (3) Unity or multi- plicity of the northwestern laterites, (4) Question of faulting on the east side of Northwest Basin, (5) Ages of folding in Kimberley Division, east side of North- west Basin and west edge of Northwest Basin, (6) Water problem in Desert Basin, (7) Origin of certain saline waters, (8) Climatic factors in human history, and many others that apply to that part of the Commonwealth. A vast field exists for geological work; the importance of the work to the future welfare of the State and the Commonwealth is very great, in addition to the benefits to be derived by science from such investigations. In conclusion I want to express my obligation to Mr. A. EH. Broué for permis- sion to make the facts and theories in this paper available to geologists. I also wish to thank Prof. Sir T. W. Edgeworth David for valuable advice on strati- graphical matters, Mr. W. S. Dun for final determination of fossil species, and Mr. A. Gibb Maitland for counsel and general advice as to methods of approaching the field and problems. EXPLANATION OF PLATES XVI—XIX. Plate xvi. 1. Typical sand-dunes near Broome. 2. Wave cut cliff of consolidated sand-dunes at Cape Villaret. 3. Mt. Phire—an outlying table-land of Jurassic age. F 66 a = Las GEOLOGY AND GEOGRAPILY OF NORTHWEST AND DESERT BASINS. Plate xvii. . Formations constituting Mt. Phire. East dips in Cape Range, west of Exmouth Gulf. Plate xviii. Surface hardening in sandstone of Jurassic age at Cape Villaret. . Sandstone of probable Jurassic age, Gantheaume Pt. Plate xix. View of top of Kennedy Range, site of Merlingleigh Homestead. Lyons Conglomerate (tillite) east of Gascoyne Junction. Dips on east flank of dome on southwest Williambury Station, Northwest Basin. THE GEOLOGY AND PETROGRAPHY OF THE CLARENCETOWN- PATERSON DISTRICT. Part iii. A Srupy or THE MAIN GLACIAL BEDS AT SEAHAM. By G. D. OsBorne, B.Sc., Dept. of Geology, University of Sydney. (Plate xiii; 1 Text-figure.) [Read 25th March, 1925.] Introductory. In the account of the stratigraphy of the Kuttung Series in the Clarencetown- Paterson district (Osborne, 1922) it was pointed out that, although difficult to classify satisfactorily, the Kuttung Series, in this, the type, area, showed a more or less well-marked threefold division into the Basal, Volcanic and Glacial Stages. The last of these was subdivided into a Lower Portion and the Main Glacial Beds, these two being separated by the dellenite-toscanite lava (Paterson type), which was found to occupy a constant stratigraphical horizon throughout the area. In the Lower Portion there occurs a definite band of varve rock, as well as a great thickness of arenaceous sediments, but in the Main Glacial Beds we have an assemblage of rocks which show unmistakable evidence of the existence of a pronounced glaciation during the period of their accumulation. The best locality in the Lower Hunter Valley to study the Main Glacial Beds is at the village of Seaham and in its immediate neighbourhood. This was recog- nised by C. A. Sussmilch who described some features of them in 1919 (Sussmilch and David, 1919), giving the following section :— Feet Corey CUMIEROCOUS MNUCISUOMES 6. co oo ob ov 05 06 264 TIMILe SB eadeGNioy DMBEGNMA.. Bais appearance. ) “9 ‘ a 2 f (30/, = # Beginning at the main road just ‘ley ay, (es Se = beyond (to the north of) its crossing Pu ‘ “y oy Qa a, of Felspar Creek, the Paterson tos- YY), hy? 3 s canite dips to the south. This is x B50 s © overlain by a notable fluvio-glacial Sy) ae z z = conglomerate, the salient features of “15, og Ss 5 tien lee lsen Given ly Wie 78, 009 Sal) 7 Sussmilch. In this unit there is a Si one 3 < band of gravel differing markedly von a4 ipo Wh ¢ from the rest of the deposit, in that Ty, ae = .| there is a greater continuity of bed- an ° ding, and an approach to uniformity ae ea tie in the size of the pebbles. It also bo loo 2 possesses a characteristic property con- 2 Ny, g = ° ‘ * 150 00. §0 Q ~$0 -100. - 150. Text-fig. 1.—Section along Pipe nected with the cementing material, the pebbles being coated with a film, purplish in colour, and probably com- posed of some compound containing manganese. These features made this horizon of distinct stratigraphical value in helping to determine the position of a fault in the area. The fluvio-glacial conglomerates occur sporadically through the allu- vium of Felspar Creek, but are stopped by a fault which seems to throw them against the tuffaceous sandstones under the Paterson tos- canite. These sandstones can be found forming the base of the hills immedi- ately to the south of Felspar Creek, on the west side of the main road. The BY G. D. OSBORNE. 69 Paterson flow, overlying the tuffaceous sandstones, is difficult to find on account of the heavy mantle of talus material and also because of its small thickness here- abouts. It does not outcrop in the Pipe Line cutting, which here is shallow, but, as will be seen on the map, it occurs both to the west and east of the section. Continuing along the section the conglomerate, with its distinctive pebble band, follows the toscanite. A little to the west of the Pipe Line the outcrop of these fluvio-glacial conglomerates comprises a number of large boulders derived from the underlying igneous rock, which on disintegrating, are apt to be mistaken for out- crops of the lava in situ. On these hills to the west of the main road there is an abundance of pebbles showing faint glacial striae. Following the conglomerate comes an extremely hard rock consisting of angular fragments set in a matrix. This combines some of the features of a tuff with those of a tillite, and there is no doubt in the writer’s mind that it is glacial in origin, at least in part. Continuing south, a small creek is crossed, and there then follows a coarse tuff which varies in texture from place to place. The dip on these beds is S 20° W at 8°. Overlying the tuffs comes the second definite bed of fluvio-glacial conglomerate. This unit contains many faintly scratched pebbles and outcrops well to the east. Following it are some interesting rocks which seem to be coarse glacial sediments mixed with tuff. The lower portion of these beds is composed of pebbly material, and this passes into finer sediment and then into a zone of transition to the overlying varve-rock. This zone of transition possesses some beautiful contortions in which the fine material is intimately mixed with the arenaceous sediment. The varve-rocks now succeeding constitute the main horizon of these in the district. They show evidence of contemporaneous contortion and of local minor dislocations. They are overlain by sandy sediments showing, in places, evidences of seasonal deposition and may be called “varve sandstones’. The textures in these rocks comprise three distinct degrees, the grainsize increasing towards the top, where the rocks are undoubtedly mixed with some tuff. Summarising the section thus far we have (see Text-fig. 1): Thickness in Feet. Fluvio-glacial Conglomerate (Felspar Creek horizon) 420 iteiisiogne Creeeke MMe co 95a Go 65 co of Go go 10 JNonGl “bhp “()) 66 oe Sto 50 Fluvio-glacial COrNGHACTARS ith tocol tillitic masses 40 Coarse Tuff with pebbles on Saul aus 50 White sandy sediment with eon. io ONCE varve rock .. .. aye | Rap a Late eet Wows 100 Main Fine-grained Tare. Rock bes ag Ce aE airs gts 100 Fine Sandy Sediment (Varve SEAR ERORO) BSI Ware 80 Fine-grained Sediments ROMS ore ores Meter: doe 100 Coarse Tuff SERA pe clone URC ECUTEE Latics Wt Il, Dice Wr ohm dole | Pack 70 Mor no oo oo IpOAd) Overlying the coarse tuff comes a series of layers composed of both tillitic material and glacial mudstones, some showing varve structures and indications of having been strongly contorted before consolidation. These layers are present in the following order, ascending stratigraphically: Feet. Limonitic contorted varve rock .. .. .. .. .. .- 2 Hard calcareous tillite .. 13 Tillite and purple varve rock 3 Grey mudstones BON bw ool seh 0 eee 5 Motales ast tein de 113 The above rocks include the most definite tillite occurring in the whole district. 70 GEOLOGY AND PETROGRAPHY OF THE CLARENCETOWN-PATERSON DISTRICT, ili, The sequence after this point is not clearly defined, chiefly owing to the inter- mittent nature of the outcrops and the lack of persistence of the units. But the general features are as follows: The tillite material is followed by glacial mud- stone in which one notable erratic of granite occurs. This underlies a definite fluvio-glacial conglomerate which has been referred to by Mr. Sussmilch as No. 3 Bed. Overlying this, just to the south of the Maitland road, are a number of units comprising tillite, agglomerate and mudstone, while in a similar position to the east is the second unit of fine varve rock. Tuffs now succeed the rocks just mentioned and also a fifth band of tillite, this being followed by a slate-blue rock, fine-grained and calcareous, which is extremely tough and resistant to weathering. In the centre of this unit there is a less competent portion, not unlike the varve rocks in many respects. Succeeding this horizon comes a drab-coloured mudstone, sometimes banded and locally con- glomeratic, which ends under a distinctive conglomerate with tuff bands, which has been regarded by Sussmilch as being the basal unit in the Permian or Permo- Carboniferous System for this district. Certainly this unit seems to differ in general facies from the conglomerates of the Kuttung Series, and it is clear that its horizon is very close to the junction line of the two systems, for, at a very short distance to the south (i.e. in the direction of dip), there occurs decomposed basalt which the writer thinks belongs to the Lower Marine Stage of the Permian System. It would appear that this conglomerate, if Permo- Carboniferous in age, has overlapped some of the older members of the Lower Marine Series, such as the Lochinvar shales, etc. Thus a generalised section of the rocks following the end point of the Pipe Line section is as follows: Feet Mudstones with erratics sgh eH OSS. EH hit 100 Fluvio-glacial conglomerate may tes fasitate| cone. ao 200 Tuffs and agglomerate .. .. BO Phe meine sarconn 6.5, muerG 50 TALLIES Te som Teese acest) eis. U Peet cede, cya NSCS) ROReen Pea CR aT 60 Varve rock (second horizon) “Ne wie) MER DIS 14a ee eee 100 Tuffs A Ath © Oe Ge RCS Aree 5am eet isn cet eal atone Css 80 Tillitie rock cee ae Dona Beth ME ee ME Eee Ne Sek e lant 4 olen |) bord 50 Fine-grained calcareous rock bcc a RR PURSE oe 50 Drab-coloured mudstone A cshal§ PS ost te. eae hea op 180 MOE, Ff Weisveh te Tes 870 The total thickness, therefore, of the glacial beds indicated in the sections described above is 1,890 feet. CONCISE ACCOUNT OF THE DISTRIBUTION OF THE MAIN GLACIAL BEDS. The most complete sections have been measured along the Pipe Line and to the south of the turn in the Maitland road near the cemetery, but a consideration of the distribution of the various beds gives some interesting information. The present description can only be brief. The basal member of the series, the Felspar Creek conglomerate, is well developed on the left bank of Felspar Creek near its confluence with the Williams River and again, as a result of the fault, on the hills to the west of the main road a little to the north of the hotel. Also, in the paddocks of the Porphyry Estate, the boulders from the tillitic mass are very conspicuous. To the west- north-west of the hotel, in the upper portion of a small creek, there is a large BY G. D. OSBORNE. 71 mass of tuff developed on the top of the Felspar Creek conglomerate. This tuff is pebbly in places. The angular-fragmental rock occurring on the Pipe Line appears to be restricted, but the succeeding tillitic conglomerate, nn which the Post Office stands, is fairly persistent. The main varve-rock horizon extends uninterruptedly from the Raymond Terrace road to the western margin of the region mapped, and the rocks show the usual varve structures and contortion zones until one comes to a spot, marked with a cross on the map, where a zone of brecciation or disturb- ance is found at the top of the rocks; a somewhat similar set of features occurs near the main road a little above the hotel. These structures will be described in detail later. The interesting arenaceous rocks and coarse tuffs following the varve rock go across the village in good development. The zone of tillite and contorted shale which is found just at the end of the Pipe Line section overlying the coarse tuffs, contains a band of purple material which can be traced as shown on the map. South from the main road there is a lack of persistence of units and irregular _masses of tillite, tuff, conglomerate, and mudstone are associated in haphazard fashion. Some of these tillitic units seem to be in much the same position as they were originally, when dumped by glaciers, not having suffered subsequent sorting by water action as did much of the sediment in the district. The restricted outcrop of the second horizon of varve rock, on the Raymond Terrace road and in the cemetery, suggests that they were developed in isolated basins, between which were dumped masses of till, ete. One or two of the horizons succeeding the second unit of varve rock are well developed and show out quite well physiographically. STRUCTURE OF THE ARRA. The structure of the area under consideration is quite simple. The rocks show a general dip towards the south, the values of reliable dips showing that there is slight rolling in the strata. Near Felspar Creek the dip has been measured, approximately only, at 15° in a southerly direction, while the lower yarve rock gives reliable dips of S. 12° W. at 30°, and S. 14° W. at 12° at the main quarry and near the Pipe Line respectively. It might be mentioned here that dips upon varve rock can only be regarded as reliable when measured upon exposures such as are afforded by fairly large quarries or recently eroded creek banks where creep and minor dislocations can be proved absent. The east and west alignment of the rocks forms the nose or front of the flatly- arched plunging anticline which, in a former paper, was described as the Williams River anticline. In addition to minor dislocations there is one important fault. It is a strike fault, and its probable position is shown on the map. The duplica- tion of some of the beds, bringing about the adjacency of rocks which are separated by an appreciable stratigraphical interval and, to a small extent, the physiographic features comprise the evidence for the existence of the fault. Thus the Paterson toscanite has been found on both sides of Felspar Creek which flows parallel to its strike. In fact, the repetition of the lava to the south of the creek, in the first place, formed the key to the structural features, and it was because of the difficulty of locating the southern outcrop of the toscanite (as has been indicated above) that the fault was not apparent during the former, less detailed, survey of the Seaham district. The fault is normal, and has an estimated throw of 450 feet, assuming, as is probable, that it dips steeply. It is interesting to note that there is no clear sign 72 GEOLOGY AND PETROGRAPHY OF THE CLARENCETOWN-PATERSON DISTRICT, iii, of the fault crossing the Williams River, since the toscanite and associated lavas do not appear to be duplicated to the east of Porphyry Point. However, the locality where one would expect to find the faulted outcrop of the Paterson lavas is given over to swamps and heavy alluvium, which would obscure the structural features. If the fault does not cross the river it is possible either that it dies out towards the east or, turning in its trend, joins up with the large fault which runs along the Williams River north from Raymond Terrace towards the east of Mt. Gilmore. LirHOLOGY AND PETROGRAPHY. A considerable amount of time has been spent in the petrological examination of these glacial beds. The majority of the horizons have been studied with the microscope, but in some cases the megascopic features only have been noted. In the following discussion the rocks will be taken in ascending stratigraphical order as far as is practicable. It should be pointed out that the truly igneous rocks, e.g., the tuffs and patches of agglomerate, will not be considered in much detail, as they will be included in the chapter on the petrography of the Kuttung Series which is now in preparation. The numbers cited refer to the specimens in the writer’s collection. The lithology of the Felspar Creek conglomerates is interesting. The boulders consist of a large variety of acidic types of igneous rocks with subordinate numbers of fine metamorphic and sedimentary rocks, quartzites being predominant among these. C. A. Sussmilch has described certain features of this horizon: while much of it is fluvio-glacial, there are portions which consist of almost true tillite. Specimen 463 of this rock, collected from the Pipe Line, shows, under the microscope, the presence of volcanic fragments, altered orthoclase and a little plagioclase set in fine matrix. Among the volcanic fragments there are some almost holohyaline rocks, in which banding and flow structure are conspicuous. Along some of the bands, haematite has separated by a process of devitrification. Microlitic felspar and some secondary silica are also present. Quite a number of volcanic inclusions have features which suggest cognate origin. They do not remind one of the underlying lavas in the Kuttung Series. In one or two fragments complete devitrification has occurred, while carbonation has developed throughout many of the rocks. The felspar fragments are not very abundant, the orthoclase being either kaolinised or replaced by calcite. Plagioclase is rare and its composition is a matter of doubt. The quartz is extremely angular in most cases, but some of it shows evidence of having been corroded. There seems to be an absence of pieces of quartz showing pseudo-inclusions of groundmass, such as is characteristic of the quartz in the acid lavas of the Volcanic Stage. The matrix is composed of numerous patches of a dull brown material, whitish by reflected light, but under high magnification seen to be composed of calcite or, at all events, of secondary carbonates. These patches, which are very feathery in outline, probably are the result of infiltration. Associated with the carbonate material there is chlorite, and it is possible that this has developed from felspar which has been replaced. The rest of the matrix is composed of innumerable specks of chlorite and other indeterminate material, together with a notable amount of pyrites, present in sections nearly rectangular in outline. Patches of recrystallised quartz occur in places. BY G. D. OSBORNE. 73 . Specimen 464. Locality, Pipe Line, west of the hotel. Megascopically seen to be composed of angular and subangular fragments of felsite, porphyry and chert cemented by a hard fawn-coloured material through which are scattered numerous quartz crystals. Microscopical characters: As in 463, there is a prevalence of vitrophyric rock types, many of which are in process of devitrification or otherwise altered. Thus haematite may be developing along lines in a fluidal rock, or silicification or carbonation is in evidence. Some of the fragments are pieces of large banded spherulites which are kaolinised. One specimen recalls a recrystallised shale or siliceous sediment. Secondary quartz and tiny wisps of muscovite were seen. The muscovite had evidently been developed from the cement which had formerly held the quartz grains together. Felspar chips are present, many being of fairly acid plagioclase. There is notable absence of albitised rock fragments. Similar patches of secondary carbonate, as seen in 463, are to be seen associated with silica in the matrix. The rock is a contaminated tillite, some tuffaceous material being present. Specimen 468. Locality, east-west road running from the main road towards the varve rock quarry. This rock is like a sandstone at first sight, but under the microscope shows many features which are directly connected with its glacial origin. In the first place the quartz grains, which constitute the greater part of the fragments, are angular or subangular, and there is evidence of little transport. The size of the grains is fairly regular, being about 0.15 mm. in average diameter. There is quite a large amount of felspar, both orthoclase and plagioclase. The orthoclase is not strongly kaolinised, and much of the plagioclase is albite, which does not look spongy like the secondary albite often seen in albitised rocks. There is also a little altered biotite, and many grains of dull brownish material which shows up like haematite at times, and kaolin at others. Under the high magnification the matrix presents some interesting features. There is an extremely fine-grained mass of material which is certainly partly Silica. This is besprinkled with points of calcite, muscovite, some haematite and kaolin. In addition to the silica mentioned above, there is some of secondary nature, which appears just like the quartz of silicified lavas. The silica in the former category does not look quite like quartz which has been introduced and yet hardly iooks as if accumulated in its present state. Many of the plagioclase pieces are perceptibly bent, the significance of this not being Known. The rock is a glacial sandstone. This material, when showing evidence of seasonal deposition, has been called “‘varve sandstone.” Specimen 469. Locality, Pipe Line just above the main varve rock horizon. Megascopically, the rock looks like a tuffaceous sandstone, possessing a white colour, weathering to fawn. Microscopical characters: The rock shows the presence of fragments of minerals, and a few pieces of rock. The grainsize of the rock chips varies from 0.6 mm. to 1.5 mm. The plagioclase is albite and there is more of it than orthoclase, which is somewhat kaolinised. Many of the quartz fragments are very slender in form while the majority are more equidimensional, all showing quite sharp extinction. Calcite, in ragged patches, is present in notable amount, but subsidiary pieces, more regularly bounded, are present, probably representing the replacement of angular chips of felspar. One or two pieces of the calcite show cleavage rather well. 74 GEOLOGY AND PETROGRAPHY OF THE CLARENCETOWN-PATERSON DISTRICT, ili, Haematite is also abundant. Among the rock fragments, devitrified and silicified rhyolites (?) occur, some of which have been originally spherulitic. In the matrix there is much secondary silica, often found wrapping around the rock fragments, and possessing features similar to that in specimen 468. Some chlorite is present, and also tiny pieces of felspar spherulites which have not suffered by decomposition. Specimen 472. Locality, Pipe Line, just north of the main road, near the cemetery. This is from the horizon of the tillite originally discovered by Professor David in 1914. The rock is grey and hard, and is composed of angular fragments, and some rounded pebbles, set in a very compact matrix which is fine-grained, and possesses many quartz grains. The rock weathers to either a yellow or purple colour, being of the latter colour at the place of original discovery at the turn in the Maitland Road. Under the microscope, apart from the large fragments which are composed of compacted varve rock or of foreign rock types, igneous and sedimentary, there is a lot of consolidated till, too fine at times to be resolved by the microscope, but, where a little coarser, seen to be composed of angular quartz and felspar, chlorite, and secondary material, together with indeterminate constituents. Specimen 476. Locality, Raymond Terrace road, just below the second horizon of varve rock. In hand-specimen it is seen to be essentially a quartzose tuff. The size of the fragments varies from place to place to a small degree, the average grainsize being 2 mm. Microscopical characters: In thin section the tuffaceous nature of the rock is apparent, although there is also a little glacial sediment. Thus one sees frag- ments of rocks, many of which are cognate in origin, also angular quartz grains which may have been blown out from a volcanic centre, and other subsidiary constituents. Mingled with all this material are strings of fine glacial sediment. The rock fragments comprise devitrified rhyolites, also some silicified rocks with phenocrysts of plagioclase and altered orthoclase. In some of the rocks, structures of peculiar pattern have resulted owing to the selective replacement, by haematite, of certain constituents in various rocks. Secondary calcite is not abundant. In the matrix there is much secondary silica, some secondary mus- covite, a little chlorite as well as other indeterminate constituents. The mus- covite and silica, which are intimately associated, occur in finely divided state and are often found mantling the large rock fragments. All these features suggest the possibility of the mica being autogenic, due to reactions which have gone on in the clayey sediment. Some greenish silicate, probably one of the chlorite group, is present, and also in the matrix some pieces of fairly fresh spherulitic felspar. Specimen 479.—This is the tuff associated with the tillite on the Raymond Terrace road just south of the park and overlying the second unit of varve rock. In hand-specimen it suggests a glacial origin, being similar to the highest tillite on the Pipe Line section. Microscopical characters: Under the microscope it possesses features which stamp it definitely as glacial. Some of the material is similar to the coarser layers of the varve rock. Quartz chips, fresh acid plagioclase, and frag- ments of volcanic rocks are set in an exceedingly fine matrix. There are abrupt variations in the size of the fragments of rock. Some of these are haematitised. BY G. D. OSBORNE. 75 Most of the rock types are acidic, volcanic in nature. The extremely angular nature of the quartz chips is noticeable. Orthoclase is being replaced by both albite and calcite and some pretty patterns have resulted by this dual replacement of the one grain. Elsewhere the orthoclase shows slight kaolinisation. The plagioclase is fresh albite and oligoclase. The matrix is exceedingly fine and much of it cannot be resolved by the microscope. It appears to contain a fair amount of finely divided silica. Specimen 480. Locality, Raymond Terrace road, just at its junction with Whyten Street. An extremely finely-textured rock which effervesces readily with acid. It possesses little variation in grainsize and is very massive. It is a sedimentary rock and shows swirl and current structures. Microscopically, it is seen to consist of numerous chips of quartz which vary little in grainsize, and fairly fresh acid plagioclase, together with some very ragged iron-ore set in an exceedingly fine groundmass. In this matrix there occurs some silica and possibly some chlorite, but the most important con- stituent is calcite, and in some cases this may be replacing tiny mineral chips. The derivation of this rock is a matter of doubt, but the general facies strongly suggests a fine glacial sediment. Specimen 492.—This is from the zone of complication between the main varve rock and the underlying tuffs just west of the road near the hotel. The rock is very pretty and shows strings of tuff, with a fairly uniform texture, seaming through the varve rock. Microscopical characters: Under the microscope the varve-shale material is seen to be very fine, sometimes possessing banding. In it are some detached rock fragments which are distinct in nature from the pieces in the tuffs. The material of the tuff stringers is much altered by deposition of haematite, while quartz crystals are quite abundant. A number of slides of the varve rocks have been examined. The lamination is, of course, very evident under the microscope, but often, in addition to the layers visible megascopically, there are some small bands, which can only be picked up in thin section. The coarser layers are bounded by sharp lines down- ward, but grade up into the following fine or winter layers. The coarser layers contain angular quartz and felspar and tiny rock fragments and much irresol- vable material, while the constituents of the fine layers are almost beyond identi- fication. Haematite is abundant in some of the coarser layers. INTERESTING STRUCTURES IN CERTAIN ROCKS. Peculiar relationships between some of the varve rocks and either sandy or tuffaceous rock are to be found in many parts of the area. Near the hotel, for example, just to the west of the road, and also at a number of places further to the west, one finds specimens showing intimate co-mingling of the two units, shale and tuff or sandy sediment. At times there is a suggestion that the tuffaceous material has been injected into the varve rock, while at other times the reverse action seems to be indicated. On account of the degree to which these features are sometimes developed, one is apt to mistake some of the rocks for agglomerates. It seems best to explain some of the associations as having been brought about during the accumulation of the glacial mud and tuffaceous material. Other structures similar to those that may be explained as just indicated, seem, as they are studied carefully, to have been produced as a result of the 76 GEOLOGY AND PETROGRAPHY OF THE CLARENCETOWN-PATERSON DISTRICT, iii, dragging force of moving ice. At the place marked with a cross on the map, west of the Pipe Line, one finds brecciation in the zone of transition from the under- lying varve rock to the tuffs. Layers of twisted varve rock pass into breccia bands, followed by more shale and finally massive tuff. In the breccia bands some extremely complicated relationships exist. Generally the varve material is found threading its way through the fairly sharply broken tuff, but at times a confused mass is found. It would appear that the tuffs developed upon the fine sediments, alternations occurring at the close of the shale accumulation period. The interference by the ice movement shattered some of the tuff, which had evidently become more coherent than the underlying varve rock which, though supporting the overlying tuffs (of no great thickness) was still plastic enough to be squeezed into the fractured tuff. Thus the varve rock gives the appearance of having intruded the tuff. CHEMICAL CONSIDERATIONS. The author carried out an analysis of a fresh portion of varve rock in order to see what bearing the results would have upon the question of the origin of the sediment. The sample was taken from a thick winter layer of compacted rock- flour, occurring in the more easterly of the two quarries at Seaham. The results are given in the table below. Along with the analysis are placed four other analyses of glacial sediments. il 2 3 4 5 Si@oi sav. 78.41 53.40 56.38 61.38 52.00 Al,Og ....- 9.32 15.70 17.81 15.58 16.11 EO! esatectuss ~25 3.22 5.04 2.58 WIEO oeaoco Le C2 2.62 2.80 3). 12 4.10 CAOU Mee. .O1 5.65 3.26 2.52 8.26 INGO oocoo 83 1.95 23 2.82 26 KO eee. 4.56 3.14 1.92 DW) 1.74 IE LOU" See so 1.74 3.14 IOS 30000 85 Ale f ae i cee ures AUKOrH cineca less than .01 B93 -40 .48 — IPEOn ts Ves .10 oY 5 al .14 WENO) ooo66 -01 .05 — —— — COs eee absent 6.24 3.40 78 — Ne S5i hee hertunts — 3.64 _- a — Ootantrencit — — — —- .09 SiKOw sin aioe — .09 -- — — ah 99.70 100.22 100.21 99.70 99.39 Specific gravity of No. 1=2.494 at 21°C. Varve Rock, Seaham, N.S.W. Anal. G. D. Osborne. Laminated Slate, near Adelaide, S.A. (Mawson, 1912). Anal. W. S. Chapman. Dwyka Tillite, S. Africa. Quoted by T. L. Walker (1921). Ojibway Clay, North-western Quebec. Anal. T. L. Walker. Stratified Glacial Clay. Quoted by M. #. Wilson (19138). oF WwW DH Examining the figures for the Seaham rock one is struck with the high percentage of silica. This, no doubt, is due to the fact that the glaciers worked over areas of acidic rocks, confirmation of this view being forthcoming in the results of the microscopical examination of the boulder beds associated with the varve rock. Alumina and magnesia are low, and lime is conspicuously so. But the most interesting feature of the analysis is in the values for the alkalies. These total 6.69% in a rock with nearly 80% silica. This without doubt indicates that much of the material that formed the rock, which under conditions of normal BY G. D. OSBORNE. nat ] weathering would have been leached away, has been retained as a result of the protective action of the glaciers which transported the rock-flour, which later was deposited in a lake. After allotting the requisite amount of alumina to soda and potash to form felspar molecules there is very little left. This latter would be present as kaolin, but its scarcity emphasises the undecomposed nature of the material. Consequent upon the almost complete absence of lime, one must postu- late all the soda as being present in the form of albite. The microscopical evidence of the glacial beds as a whole shows the prevalence of albite among the plagioclase felspar, and thus lends support to this view. The petrographical and chemical study of glacial sediments has not been pursued to any great extent in Australia, although the lithology of many occur- rences has been written up, and one or two analyses have been made. But else- where a fair amount of attention has from time to time been given to this subject. It will be interesting to notice certain features in connection with the analyses quoted above. Thus it may be pointed out that in all cases there is not a great excess of the molecular proportions of alumina over those of lime, soda and potash combined (some of the lime having been allotted to CO,). Of course one must not expect to find an absence of kaolin in all glacial clays, since the glaciers which produced the clays may have worked over areas given over to aluminous rocks, such as shales. T. L. Walker, of Toronto University, has written upon the chemical study of conglomerates (Walker, 1921), and has examined a number of analyses of undoubted glacial rocks and also of some sediments for which a glacial origin might be suspected. He has generalised that the value of the Na,.O/K.O ratio can be used to test the origin of a sediment, a high value indicating a probable deri- vation of the material by glacial action, and a low value pointing to the material having been well leached, and thus probably the result of decomposition under conditions other than glacial. He also says: “In case that material were assembled from a wide area of igneous rocks the glacial rock flour should have a composition similar to that given by Clarke as the average composition of igneous rocks. Where gathered from a field where both igneous and sedimentary types are asso- ciated the cement should lie between the average igneous and the average sedimentary rock in composition.” The present writer considers that the last paragraph could not be made of general application in connection with glacial areas, because it is highly improb- able that any glacial clays possess disintegrated material from such a wide area of rocks, that the analysis of the clays could come under any of the categories mentioned, except, of course, by accident. Then again, it seems that the validity of relying on the ratio of soda to potash is open to question, since there are many exceptions to the rule that during rock decomposition soda is always leached more readily than potash, and further, there is always the possibility of derivation of rock material from areas where potash is so dominant as to give rise to a value of less than unity for the ratio in question, even though no leaching had occurred. STRUCTURES IN VARVE ROCKS WITH REFERENCE TO THE ESTIMATION OF GEOLOGICAL TIME. Seasonally-banded glacial sediments—varve rock and varve sandstones—have, on numerous occasions, formed the subject of attempts to estimate the time taken for their accumulation. Baron De Geer and his students have done classical work 78 GEOLOGY AND PETROGRAPHY OF THE CLARENCETOWN-PATERSON DISTRICT, iii, in this direction (De Geer, 1912). R. W. Sayles has contributed a very important memoir upon the subject of seasonal deposition (Sayles, 1919), and has written much that deals with the matter now being considered. As a result of his studies, it is clear that much caution is necessary before one can speak with any degree of certainty upon the duration of the time taken for a glacial deposit to accumulate. Thus there is evidence, in deposits such as varve rocks, that the complete record is not always preserved, except in those deposits which have originated under deep-water conditions, far removed from disturbing agencies such as con- temporaneous erosion or ice movement, the former of which might remove quite a big proportion of material, sweeping it away to be deposited further from the ice front. At Seaham the varve rocks show evidence of contemporaneous erosion in many places. In some cases the erosion has been extensive enough to remove inches of sediment at the place of maximum scour. Another fact, that must be realised in connection with the estimation of time by the counting of layers, is the possibility of rhythms other than the annual one making their presence felt. It is probable that the effects of daily rhythms may be strong enough to be perceptible, at least under the microscope. Then again it has been found that material which may appear massive in hand specimen shows, under the microscope, alternations which are due to seasonal deposition. So also, some cases were encountered where material which was massive both megascopically and microscopically, on weathering showed the presence of seasonal bands. Thus there are at least four factors which may enter into the question and tend to make difficult any attempt to estimate, by counting layers in a large deposit, the time which was taken for its accumulation, and the more one considers these factors and their significance, the more one sees the need for caution in the investigation of this phase of the estimation of geological time. Thus the statement by C. A. Sussmilch that “the 200 feet of these varve shales which occur at Seaham must have taken about 3,000 years to deposit’ (because the average width of a pair of seasonal layers is about two-thirds of an inch) is hardly satis- factory in view of the points raised above. SUMMARY INDICATING THE SIGNIFICANCE OF SOME OF THE FOREGOING RESULTS. In the preceding paragraphs a detailed account has been given of the Main Glacial Beds of the Kuttung Series, which outcrop in the Seaham district. A map of the area shows that the beds possess quite simple tectonic features, there being one important strike fault and minor dislocations. Some of the units, in contrast to certain other beds, are discontinuous along their strike; among such units are some tillitic masses which probably represent compacted morainic material which has not been affected by fluvial action. The examination of the sequence in detail shows a total thickness of 1,890 feet of sediment. The petrographical investigation of the beds gives much interesting informa- tion. Apart from the horizon of the Felspar Creek conglomerate, there is a distinct absence of much bouldery or fragmental material derived from the under- lying Kuttung lavas. Much of the rock associated with the varve units shows evidence of having been transported by glaciers, in the presence of fresh plagio- clase and orthoclase, the latter sometimes being present as spherulites. A chemical analysis of a portion of varve rock is given, and it is of interest to find nearly 7% of alkalies present, along with high silica and low alumina. BY G. D. OSBORNE. 79 These figures confirm the glacial origin of the rock, by indicating the absence of much kaolinic material, and the presence of silicates which would be decomposed under ordinary conditions of weathering and transport. Descriptions of brecciated structures and examples of contemporaneous erosion have been given and it is to these features and the associated contem- poraneous contortions which the fine varve rock and varve sandstone exhibit, that some attention may be given here. The significance of all these characteristics is that in the Seaham area there must have been much land ice in late Kuttung times. This land ice at times overrode the sediments and produced the interesting structures already described. In contrast with the sediments developed close to the ice front (like the Seaham rocks), one finds that glacial deposits formed in deep water (often under marine conditions) show an absence of the structures mentioned above. The Tapley’s Hill slates or shales occurring near Adelaide, in South Australia, are regarded by Sir Douglas Mawson as being glacial in origin. They are associated with the tillites of the Adelaide Series, and possess laminations due to alternations in the material deposited, and these laminations are somewhat suggestive of varve structures. But contemporaneous contortions, current bedding and erosion features are absent, and the general facies of the rocks strongly suggests accumu- lation in deep water, with the resulting absence of disturbing agencies. If these Tapley’s Hill rocks are aqueo-glacial in origin, their distinction from the Seaham type of deposit is very marked, and, in a general way, one may be justified in concluding, in the absence of any contradictory evidence, that glacial deposits showing contortions, contemporaneous erosion and kindred phenomena, may have been developed close to the ice-front, while evenly-banded rocks, free from con- tortions, etc. (like the Tapley’s Hill shale), indicate deposition at a distance from the ice-front, where disturbances do not occur. ‘ In conclusion, the writer wishes to express his appreciation of the kindness and hospitality extended to him by Mr. L. B. Fisher and family of “Brandon” during his visits to the area in 1922-24; and in connection with the preparation of the paper he is indebted to Professor David and Assistant-Professor Browne for helpful discussion. List of Works referred to. DE GEER, G., 1912.—Compte Rendu Congrés Géol. International (Stockholm, 1910), pp. 241-253. Mawson, D., 1912.—Geological Investigations in the Broken Hill Area. Mem. Roy. Soe. South Awstralia, ii, Pt. 4, 245. OSBORNE, G. D., 1922.—Geology and Petrography of the Clarencetown-Paterson District, Pt. i. THESE PROCEEDINGS xlvii, 161-198. SAYLES, R. W., 1919.—Seasonal Deposition in Aqueo-giacial Sediments. Mem. Mus. Comp. Zool. Harvard xlvii, No. i. SUSSMILCH, C. A., and Davin, T. W. E., 1919.—Sequence, Glaciation, and Correlation of the Carboniferous Rocks in the Hunter River District, N.S.W. Journ. Roy. Soc. N.S.W., lii, 247-338. WALKER, T. L., 1921.—A Chemical Study of Conglomerates. Univ. of Toronto, Geol. Studies, No. 12. WILSON, M. H., 1913.—Kewagma Lake Map-Area, Quebec. Mem. Geol. Surv. Canada, No. 39, 96. EXPLANATION OF PLATE XIII. Geological Map of the village of Seaham, showing the outcrops of the Main Glacial Beds of the Kuttung Series. NOTES ON AUSTRALIAN DIPTERA, No. Vi. By J. R. MALLocH. (Communicated by Dr. E. W. Ferguson.) (Twelve Text-figures. ) [Read 27th May, 1925.] In this paper I present some notes on and descriptions of Australian acalyptrate Diptera. As in the case of material dealt with in preceding papers on these insects, the types will be returned to Dr. Eustace W. Ferguson to deposit in some Aus- tralian museum where they may be available to subsequent workers. Family Sciomyzidae. This family is very closely allied to the Sapromyzidae and much more difficult to distinguish from it than one would expect if he judged relationships by existing catalogues and books on classification. The Huropean fauna contains, for the most part, species which are readily assigned to one or other of these families by the use of characters listed for that purpose by Schiner and the earlier authors., But when one has before him species from all over the world, it is a much more difficult matter to separate the families. General habitus is a good index in Europe, but this fails in the Orient and Australasia. In fact the comparative size of the basal and anal cells of the wings, small in Sapro- myzidae and large in Sciomyzidae, is of no value in many cases, and new characters must be discovered if we are to maintain both families. That the families ought to be kept separate is my personal opinion, based upon a knowledge of the habits and habitats of the adults. So far as I have seen, the larvae of the Sciomyzidae are found in water, and in some cases in molluscs as parasites, and the adults in marshy situations almost exclusively. On the other hand I have found no aquatic nor semiaquatic larvae of Sapromyzidae, and the adults are not found in marshy spots, but in woodlands. It is possible that my attitude in the matter of separating these families is influenced by these biological factors, but in any case I have decided that until more data are available the families ought to remain separate. With this end in view, I have examined carefully the structural details of the species before me and find that they differ in the extent of the sixth wing vein more noticeably than in any other obvious character. This vein in Sciomyzidae is traceable to the margin, while in Sapromyzidae it is not. Ordinarily the prosternal plate in the former is small, while in the latter it almost fills the space between the fore coxae. The mesopleura in Sapromyzidae usually has a strong bristle on its hind margin which is rarely present in Sciomyzidae. Both families lack vibrissae, have the auxiliary vein of wing complete and distinctly separated from first on its entire length, the discal and basal cells separated by a cross-vein, the BY J. R. MALLOCH. 81 anal cell and sixth wing-vein present, and the tibiae with distinct preapical dorsal bristle. I have seen but two genera from Australia, Helosciomyza Hendel and Melina R.-D. Genus HrELosciomyzaA Hendel. This is an aberrant genus possessing long bristles on the costa, a character common to genera of Helomyzidae, but rare outside that family. In other respects it is a typical sciomyzid. Hendel described the genus in 1917 (Deut. Ent. Zeit., p. 33). I have described two species from New Zealand and placed rara Hutton in the genus (N.Z. Journ. Sci. and Tech. 5, 1922, 228). HELOSCIOMYZA FERRUGINEA Hendel. ¢.—Brownish-yellow. Frons orange coloured, orbits and triangle whitish; face testaceous yellow. Thorax with four rufous-brown vittae, the broad lateral pair narrowly divided by a grey line behind the suture; scutellum grey on sides. Fore and hind femora and tibiae dark at apices, the fore femora indistinctly so; hind tibia with a brown band at base; apical segment of tarsi dark. Both cross- veins narrowly clouded. Each orbit with two bristles; arista pubescent. Sternopleura with two or three bristles; mesopleura bare; mesonotum with two pairs of dorsocentral bristles and a pair of long prescutellar acrostichals. Hind tibia with two preapical dorsal bristles. Length, 5.5 mm. Originally described from Victoria. I have one specimen from Como, N.S.W. (H. Petersen). Family Sapromyzidae. I have received, from various sources, material belonging to this family and present data upon three previously described and one new species. Genus STEGANOPSIS de Meijere. This genus was founded for the reception of a Javanese species, pupicola de Meijere, which feeds in the larval stages in the chrysalides of a butterfly. Later Hendel described a species from Formosa, and Kertesz one, vittithorax, from New South Wales, while Frey erected a new genus, Steganolauxania, for the reception of the North American species Lauzania latipennis Coquillett, which was included in Steganopsis by some authors. This genus, I think, is unnecessary, latipennis agreeing very closely with the species I have seen from Australia and the Orient. I also consider it as very probable that the species described by Hendel is a synonym of the species redescribed below; Kertesz’s species certainly is. STEGANOPSIS MELANOGASTER (Thomson). Head yellow; frons opaque, ocellar spot black; face glossy, with a black spot on each side, and a brown mark below eye; antennae yellow, third segment largely brownish-fuscous; apices of palpi black. Thorax tawny-yellow, dorsum with about 6 brownish-red vittae, the sides of which are usually dark, sometimes black. Abdomen pitchy coloured, paler in male, glossy. Legs tawny-yellow, fore tarsi more or less blackened. Wings brownish-yellow, darkest along costa, becoming hyaline behind, apices narrowly hyaline. G 82 NOTES ON AUSTRALIAN DIPTERA, Vi, Anterior orbitals incurved, posterior pair curved backward; first and second antennal segments longer than wide, both haired below, third almost twice as long as first and second combined and nearly six times as long as thick; arista short haired; face convex; eye higher than long; labrum high and distinct. Thorax with three pairs of dorsocentral bristles; prescutellar acrostichals distinct; intra-alar absent; sternopleural 1; scutellum flattened, elongate, rounded apically. Fore femur without a comb. Wing venation as in Figure 1. Length, 3-4 mm. Locality, Botany Bay, N.S.W. (H. Petersen). Originally described from Sydney. Fig. 1. Steganopsis melanogaster, wing. I have before me a male specimen from the Philippine Islands which is, except for being paler in colour, almost identical with the male of melanogaster, and which is apparently convergens Hendel. As Thomson’s species is the oldest in the genus I do not attempt to treat at this time with the synonymy. It is strange that in Thomson’s description nothing is mentioned of the head characters of his type; the head may be missing in this. Genus TRIGONOMETOPSIS novum. Generic characters.—Similar to Trigonometopus Macquart, the face retreating below, eyes elongate oval, each orbit with two bristles, antennae short, arista subnude, thorax with three pairs of dorsocentrals, outer cross-vein beyond middle of wing, third vein not curved, costa to apex of fourth vein, and hind tibia with a preapical bristle. The general habitus of the two genera is similar and the black spot between antennae and eyes is present in both, though in the new genus it is not noticeably hairy. The presence of a well developed presutural bristle in the new genus distinguishes it from Trigonometopus. Genotype, Oxyrhina binotata Thomson. TRIGONOMELOPSIS BINOTATUS (Thomson). A yellowish testaceous species, with the abdomen more shining than thorax. A longitudinal line of brown colour extends from vertex to anterior margin of frons; a black spot between each antenna and eye; and a pair of small round black spots on middle of face. Dorsum of thorax brownish-yellow, paler on lateral margins, darker on lines of dorsocentral bristles and bordering the pale lateral margins, and with a blackish central vitta which is obscured by greyish dusting; pleura with a faint brownish vitta; scutellum pale in centre. Legs yellow. Wings clear, yellowish on anterior portion, veins brown. MHalteres yellow. Frons about 1.5 as long as wide; cheek about one-third of eye height; proboscis thick. Acrostichais in two series, prescutellar pair of moderate size; scutellar BY J. R. MALLOCH. 83 bristles subequal. Inner cross-vein almost at middle of discal cell; last section of fourth vein only about one-fifth longer than preceding section; outer cross- vein at a little more than half its own length from apex of fifth vein. Length, 3.5 mm. Originally described from Sydney, from whence I have a female specimen. The species described from the island of Guam as Trigonometopus setosus by Knab does not belong to that genus. It agrees fairly well with Trigonometopsis, but has the first pair of dorsocentral bristles on thorax in front of suture instead of just at it, and the cheeks with some long hairs along margins which are quite conspicuous. The frons is less protruded beyond eyes and the face is unspotted. This species should apparently be located in Panurgopsis Kertesz and not in Trigonometopsis, with the genotype of which it has much in common. Genus SAPROMYzOSOMA Malloch. This genus is distinguished from its allies by the lack of an intra-alar thoracic bristle, and the continuation of the short stubby black costal setulae to the apex of third vein. It is possible that this genus is the same as Homoneura van der Wulp, but I have not seen an authentic specimen of picea v. d. Wulp, the genotype, so cannot give a definite opinion. There are a number of Australian species of the genus, which appears to occur over most, if not all, of the faunal regions of the world. The species listed below appears to belong here, but the wing-tips are imperfect in the only specimen available to me so that I cannot be absolutely certain of this. SAPROMYZOSOMA MACULIFRONS (Macquart). Testaceous yellow, shining. Ocellar spot, a spot on each side of interfrontalia at middle of frons, basal two antennal segments, apices of palpi, and apices of tibiae black; abdomen pitchy; apices of mid and hind femora narrowly dark. Wings clear. Halteres yellow. Arista bare. Thorax with three pairs of dorsocentral bristles, the anterior pair short; prescutellar acrostichals strong; disc of thorax with short black hairs; scutellum subconvex. Abdomen broad, hypopygium not conspicuous. Fore femur without an anteroventral apical comb, the posteroventral bristles on apical half only; hind femur without a preapical anteroventral bristle; all tibiae with distinct preapical bristle. Inner cross-vein at middle of discal cell and almost below apex of first vein; outer cross-vein at a little more than half its own length from apex of fifth vein. Length, 3 mm. Originally described from Tasmania. One male, Seaford, Victoria (W. F.. Hill). Genus SAproMyzA Fallen. In this genus the intra-alar bristle of thorax is lacking, and the short black setulae of the costa are discontinued about midway between apex of second and third veins, gradually becoming evanescent. I give the description of one species. SAPROMYZA ATRIVENTRIS, N. SD. go 9.—Deep black, shining, thorax, fore coxae, and all trochanters orange yellow, mid and hind tibiae and tarsi more or less distinctly yellowish at bases. Wings yellowish, halteres brownish. 84 NOTES ON AUSTRALIAN DIPTERA, Vi, Frons entirely shining, orbits but little differentiated, both pairs of bristles long, sloping backwards; ocellars small; postverticals quite large; antennae inserted much above middle of profile, large, basal segment quite distinct, about as long as second, bare below, third about 1.75 as long as first and second combined and about twice as long as broad, rounded at apex; arista slender and long, distinctly pubescent; face slightly receding in profile, convex, sides a little paler than centre and with dense silvery pruinescence on entire length of parafacials, which extends along cheeks for a short distance; cheek not as high as width of third antennal segment; eye narrowed below, higher than long. Thorax with three pairs of dorsocentral bristles and a pair of long setulae in front of anterior pair, the prescutellar acrostichal pair long; scutellum convex, the four bristles subequal in length; mesopleura with one strong bristle; anterior sternopleural bristle short. Abdomen ovate, male hypopygium of moderate size. Fore femur without anteroventral comb; mid femur with anterior bristles distinct; hind femur without preapical anteroventral bristle; preapical tibial bristle distinct on all legs. Inner cross-vein a little before middle of discal cell; outer cross-vein at about half its own length from apex of fifth vein. Length, 5-5.5 mm. Type, female, and two paratype females, Illawarra, N.S.W.; allotype, Como, N.S.W. (H. Petersen). No species known to me from any part of the world has the same colours of body and legs as this one. Possibly the nearest related species may prove to be nigriceps Macquart, also an Australian species, but it has the antennae yellow at bases and the legs and abdomen partly yellow, and in other respects does not fit the species above described. Genus POECILOHETAERUS Hendel. This genus was erected for the reception of a single species, decora Schiner, which name, being preoccupied, was replaced by schineri Hendel. The genus is distinguished from Sapromyza, to which it is closely allied, by the anterior pair of orbital bristles being incurved instead of backwardly curved. In other respects the genera are very similar. POECILOHETAERUS SCHINERI Hendel. A deep black species with two moderately broad and very conspicuous white pruinescent vittae extending from anterior margin of frons over disc of thorax to apex of scutellum, a less distinct vitta on sides of mesonotum and three similar on pleura; parafacials and postocular orbits white, centre of cheek black; mesonotum with a pale brown median line. Legs black, mid and hind tarsi and bases and middle of mid and hind tibiae yellowish. Wings yellowish. MHalteres blackish. Arista pubescent; third antennal segment about twice as long as broad. Thorax with four pairs of strong dorsocentrals, the anterior pair in front of suture; one moderately strong pair of prescutellar acrostichals and in front of these a few weak paired hairs; sternopleura with two bristles; scutellum with dise slightly flattened, the apex a little transverse, bristles four in number. Fore femur without anteroventral comb; all tibiae with preapical bristle distinct. Costal black setulae discontinued before attaining apex of third vein. Length, 3-4 mm. BY J. R. MALLOCH. 85 Known only from Eastern Australia. I have a number of specimens from Botany Bay and Illawarra, N.S.W. (Petersen). Family Ortalidae. I take this opportunity to record the occurrence of the following European species amongst some material sent to me by Dr. C. F. Baker from Australia. , CHRYSOMYZA AENEA Fabricius. I believe that melanopsis Walker may he a synonym of this species. The species occurs also in North America, having been introduced in commerce no doubt. Family Borboridae. I have seen very few species of this family from Australia and pending the receipt of more material record but two species, one of them new. Genus SPHAEROCERA Latreille. This genus is distinguished from its allies by the lack of bristles on the scutellum; the position of the cross-vein at base of discal cell, which is basad of the one closing anal cell, causing the second basal cell to be shorter than the anal. There are other distinctions, but the above should suffice for the recognition of the genus. Leptocera has the fifth vein incomplete, the anal cell and cross-vein dividing second basal and discal cells absent. This genus occurs also in Australia, but I have not yet worked up the species I have on hand. SPHAEROCERA CURVIPES Latreille. This species has generally been referred to as subsultans Fabricius, due to- an early error in identification. It is the largest known species of the genus, averaging about 4 mm. in length, and is very widely distributed. I have seen one specimen from Botany Bay, N.S.W. (H. Petersen). The larvae and adults live on manure and carrion. Genus BoORBOROIDES novum. Generic characters.—Wach orbit with two outwardly curved bristles, inner vertical bristie incurved, outer outwardly curved; ocellars long; postverticals convergent; face concave, genal bristle absent; vibrissae present; proboscis stout. Thorax with 1 humeral, 1 presutural, 2 notopleurals, and 2 pairs of dorsocentrals; scutellum with 4 bristles; sternopleura with one or two bristles. Mid tibia with 2 long dorsal bristles near apex; hind metatarsus longer than second segment. Wing venation as in Figure 2. Genotype, the following species. BORBOROIDES ATRA, Nl. Sp. °.—Black, shining, lower part of parafacials, fore coxae, and bases of fore femora obscurely yellowish. Wings hyaline. Halteres black. Frons smooth and shining, with some short hairs; third antennal segment higher than long, rounded in front; arista pubescent. Dorsum of thorax regularly but not densely haired, that of scutellum bare. Apical genital organs a pair of slender lamellae. Mid femur with a series of short bristles on apical half of anterior surface. 86 NOTES ON AUSTRALIAN DIPTERA, Vi, Length, 2 mm. Type, Sydney, N.S.W., 1.11.24. This genus has as its closest relative Borborus Meigen, but differs materially from it in venation of wing, in the presence of two dorsal bristles on the mid tibia such as are found in some species of Leptocera, and in the elongate basal segment of hind tarsus. From Leptocera it may be readily distinguished by the presence of the basal and anal cells of the wing. The genus Leptocera occurs in Australia, the species having been referred to under the generic name Limosina. Family Ephydridae. Subfamily ErpHypDRINAE. Genus HECAMEDE Meigen. This genus has the frontal triangle with two or three pairs of short proclinate bristles in front of the anterior ocellus, each orbit with usually two short bristles, second antennal segment with a short forwardly directed apical bristle, and some finer hairs basally, arista short, with about six long hairs above, face concave on upper half, with a nose-like process in centre, the tip of which is invariably glossy black, three or four bristles on each side of lower half of face which are situated on slightly elevated black bases, cheek nearly as high as eye, genal bristle weak, clypeus slightly projecting. Thoracic bristles short, only the prescutellar dorso- central bristles distinct. Costa to apex of fourth vein. Mid tibia without long dorsal bristles. HECAMEDE ALBICANS Meigen. Black, densely pale grey pruinescent, almost whitish on abdomen, more yellowish on dorsum of thorax. Anterior margin of frons slightly rufous; antennae rufous. Bases of tibiae, and the tarsi yellowish, apices of latter darker. Wings white, veins yellowish. Halteres white. Length, 2.5 mm. One female, Encounter Bay, S.A., on beach. Hendel has described a species under the name persimilis from Formosa and de Meijere has described two others from Java. I consider it very probable that persimilis is merely a form of albicans. I have compared the Australian specimen with European examples and can find absolutely no distinguishing characters either of colour or structure. Subfamily CANACEINAE. In a previous contribution on the Diptera of Australia, I presented a key for the recognition of the genera of this subfamily under the impression that subse- BY J. R. MALLOCH. 87 quent collecting would disclose the presence of other genera than the one I then had from this continent. I now present the description of a new species of Canace recently received from Dr. Ferguson. The dark facial stripe appears distinctive, though the structure of the head is also characteristic. CANACE ALBICEPS, Nn. SD. ?.—Frons opaque, olive brown, the orbits greyish anteriorly; face densely white pruinescent, with a narrow dark vertical stripe, cheeks silvery white; clypeus whitish grey; palpi tawny-yellow; antennae black. Dorsum of thorax olive-brown, with very faint brown vittae, the most obvious one in centre; pleura grey. Abdomen dark greyish on dorsum, paler below. Legs greyish fuscous, extreme apices of femora, bases and apices of tibiae, and most of tarsi yellowish. Wings slightly greyish, veins fuscous. Halteres pale yellow. Eye about 1.75 times as long as high, narrowest part of cheek (at anterior margin of eye) about as high as eye; a long upcurved bristle on cheek below middle of eye and three along its lower margin on anterior half, the hindmost one upcurved; each orbit with four long bristles directed over eye; about four pairs of incurved interfrontal bristles present; ocellar bristles long, divergent, in line with anterior ocellus; postverticals divergent, small; centre of lower margin of face slightly angularly produced, not transverse. Thorax with four pairs of dorso- central bristles, the acrostichals sparse, paired; scutellum with two long apical, and two shorter basal bristles, and two discal setulae; pleura too much damaged to determine characters in type, but the sternopleura has no strong bristles. First visible abdominal tergite elongate; genital thorns two in number. Fore femur without anteroventral comb of bristles; fore tarsus dilated apically, basal segment longer than next three combined, each of latter broader than long, neither of them as long as fifth, the claws long and curved. Last section of fifth vein subequal in length to penultimate section of fourth; veins 3 and 4 parallel apically; penultimate and ultimate sections of costa subequal. Length, 2 mm. Type, Sydney, 10.9.21. Family Drosophilidae. DROSOPHILA POECILITHORAX, ND. SDP. °.—Head fuscous, frontal orbits, triangle, face and cheeks densely grey pruin- escent, palest on frons, anterior half of frons pale orange yellow, posterior half of interfrontalia brownish-red; second antennal segment brownish, third black, mouth-parts fuscous; a faintly indicated dark spot surrounding bases of orbital bristles. Thorax black, densely bluish-grey pruinescent on dorsum and with rather large dark brown dots at bases of hairs and bristles; humeri and scutellum con- colorous with disc, the scutellum with a curved dark brown streak on each side of median line extending the entire length. Abdomen shining fuscous, without distinct markings in type. Legs fuscous yellow, darkest on middle of femora. Wings clear. Facial keel rounded, becoming broad and disappearing before reaching mouth; labrum quite prominent; cheek fully one-third of the eye height, slightly produced at vibrissal angle, with a series of marginal bristles; arista with three hairs above and one below; anterior two orbital bristles at almost same point; postvertical bristles long. Only two series of acrostichal setulae between dorsocentrals, the prescutellar pair quite pronounced; basal pair of scutellar bristles much smaller 88 NOTES ON AUSTRALIAN DIPTERA, Vi, than apical pair. Legs normal. Second costal division twice as long as first and but little longer than third, the latter fully three times as long as fourth; pen- ultimate section of fourth vein not more than one-third as long as ultimate and nearly as long as ultimate section of fifth; outer cross-vein not more than one- third as long as apical section of fifth vein. Length, 1.25 mm. Type, Sydney, 26.2.24. This species belongs to the same section as obsoleta Malloch, repleta Wollaston, and hydei Sturtevant. It differs from the first-named species in having a distinct facial keel, only two series of intradorsocentral setulae, and entirely different wing venation. From the other two species it differs in its smaller size, less robust form, unspotted abdomen, venation of wing, and several other respects. Dr. Ferguson has suggested to me that australis Duda is probably the same as obsoleta Malloch. The descriptions agree very well, the venation given by Duda being in accord with that of my species and the thoracic characters similar. There are, however, a few points of difference and without an examination of the type of australis it were better to leave the matter in abeyance, though there is a very great probability that the species are the same. Family Agromyzidae. Subfamily MiricHrin ak. Genus Sromosis Melander. Related to Desmometopa, differing in having the frons without an M-shaped black mark; hind tibiae not dilated; mesopleura bare. I figure the head showing the long geniculated proboscis and other features (Fig. 3). Fig. 3. Stomosis flavoscutellata, head from side. The genus Stomosis was erected for the reception of luteola Coquillett which had been previously placed in Desmometopa. I have compared the following species with the type of Coquillett’s species and find that they agree in generic characters though specifically distinct. Lwteola is American. STOMOSIS FLAVOSCUTELLATA, Nn. Sp. 9.—Head orange-yellow, upper half of orbits, frontal triangle, and occiput black; third antennal segment brown above; arista fuscous; proboscis brown. Thorax glossy black, broadly yellow along lateral margins of mesonotum from humeri to base of scutellum; pleura black; scutellum yellow. Abdomen glossy black. Legs yellow, all femora black except at bases and apices. Wings clear. Halteres yellow. BY J. R. MALLOCH. 89 Ocellar triangle narrow, extending well over midway to anterior margin of frons; postvertical and ocellar bristles long, the former cruciate; upper three orbital bristles curved outward over eye, lower three or four incurved; inter- frontal hairs very weak; profile as in Figure 3. Thorax with two pairs of dorso- centrals; prescutellar pair of acrostichals distinct; four series of setulae between dorsocentrals; basal pair of scutellar bristles much shorter than apical pair which are divergent. Hind femur with a preapical anteroventral bristle. Last section of fifth vein as long as penultimate section of fourth, the latter about one-fourth as long as ultimate section of fourth. Length, 2 mm. Type, Melbourne, Victoria, 14 November, 1923. STOMOSIS VITTATA, N. SD. ?.—Differs from the preceding species in having the upper orbits yellow; dorsum of thorax with three broad black vittae on a yellow ground, the median one extending over base of scutellum; the pleura instead of being almost entirely black have a black vitta along upper margin, the sternopleura black and two other marks, one on pteropleura and the other on hypopleura. The abdomen is not entirely black above, but has the base yellow, the pale coiour extending almost to hind margin of first visible tergite in middle, and the bases of the other tergites yellow. Apices of tibiae rather broadly blackened; tarsi darkened apically. Structurally similar to preceding species. Length, 2.25 mm. Type and paratype, Sydney, 3.10.24, and 26.10.24. Subfamily AGROMYZINAE. Genus CrRopoNTA Rondani. The members of this genus are distinguished from those of Agromyza by the very pronounced spike-like apex of third antennal segment, and the presence of but one pair of scutellar bristles. The postscutellum is generally quite distinctly developed, sometimes extending well beyond apex of scutellum. I have seen two species from Australia, both of them apparently new. CERODONTA AUSTRALIS, nN. Sp. °.—Head yellow, interfrontalia generally darker than orbits, ocellar triangle, occiput, third antennal segment, and arista black. Thorax black, densely grey Se so 4 Fig. 4. Cerodonta australis, antenna. Fig. 5. Cerodonta robusta, antenna. pruinescent, with indications of two darker vittae along the lines of dorsocentrals. Abdomen black, apices of tergites with a yellow line. Legs yellow, apices of tarsi infuscated. Wings greyish, radius yellow basally. Halteres yellow. 90 NOTES ON AUSTRALIAN DIPTERA, Vi, Third antennal segment as in Figure 4. Thorax with four pairs of dorso- centrals, and one pair of acrostichal setulae behind suture and in line with the second pair of dorsocentrals. Venation normal. Length, 2.5 mm. Type and three paratypes, Sydney, 28 October, and 21 September, 1923, and 14 September, 1924. This species is very similar to denticornis Meigen, but I have examined many specimens of the latter and have found none in which the acrostichal setulae are present as in australis. In Hurope and North America denticornis, in the larval stages, does consider- able damage to wheat and hay by mining in the stems. CERODONTA ROBUSTA, Nl. Sp. 9.—Head lemon-yellow, ocellar spot and upper occiput black, antennal foveae, labrum, and apex of third antennal segment darkened; arista fuscous. Thorax shining black, broadly yellow on lateral margins to lateral angles of scutellum; upper half of pleura yellow; scutellum black. Abdomen black, the hind margins of tergites yellow. Legs dirty yellow, tibiae and tarsi subfuscous. Wings greyish, veins conspicuous. Halteres yellow. A robust species, much stouter than the others in the genus, and more resembling some species of Agromyza. Antennae shorter than usual in the genus (Fig. 5); head large; arista pubescent. Thorax stout, with the usual four pairs of dorsocentrals; about six series of setulose hairs between the dorsocentrals; scutellum convex; postscutellum low and small. Abdomen stout. Legs stouter than in australis. Wings broader than in that species; cross-veins separated by about length of outer cross-vein; last section of fifth vein subequal to preceding section; last section of fourth vein about seven times as long as preceding section. Length, 3 mm. ’ Type, Sydney, N.S.W., 20.11.24. Paratype, same locality, 14.9.24. Distinguished from any species of the genus by the short antennae, and the number of series of setulose hairs between the dorsocentral bristles on thorax. AGROMYZA PUSILLA Meigen. This cosmopolitan species is represented by a number of specimens taken at Sydney, N.S.W. One of these is mounted along with a specimen of Coenosia acuticornis Stein of which it was evidently the prey. The species, in the larval stages, mines in the leaves of a large number of plants both cultivated and wild. Amongst those recorded are Nasturtium, cotton, peas, cabbage, various clovers, Solanum, Valeriana, Hupatorium, Huphorbia, etc. Genus FERGUSONINA Malloch. In a lot of Diptera recently received from Dr. Ferguson I find four specimens of this genus, three of them males, which sex I did not have before me when I described the genus. I note that the venation of some of the specimens shows a departure from that of the genotype in having the outer cross-vein distinct. 'There are, however, no important changes necessary in the generic definition previously given, the remarkable flattened head, with the much enlarged frontal lunule, being ‘ characteristic of all. I give below a key for the identification of the species now on hand. BY J. R. MALLOCH. 91 Key to Species. Head and thorax yellow in both sexes, only a blackish tinge round ocelli; venation of LARGEST ISH I ay MEM AAD NeW e Che hao eRe ere A aenTe te one eet oe et eee microcera Malloch. Head with at least the ocellar spot, and thorax with at least partial vittae, black .. 2 Third antennal segment deep black; venation of wing as in Figure 7 .. atricornis, n. sp. PNTILENT AC wWeNtIRe VwiGLEaT syelll OW) a. o.srauaestAreue ote ca ator Meee el titer telienemolLenG er ae ceneueicn sven oreliene ese 3 Dise of thorax with six black vittae, median pair extending from anterior margin to beyond suture, lateral pairs not extending to anterior margin, but reaching beyond hind margins of median pair; scutellum entirely yellow; venation of wing as in TEMA DINE) stssh MeO Rae PAE BOLO MSEC Cee ORTON Came ee CCE CORE MG MeCR Ee Tek Citra ICR eee ey ce flavicornis, n. sp. Dise of thorax black, lateral margins yellow; scutellum black at base; venation of THAAD KeRh US Woah debi bb Ge) maaya tas Baio CRT cho: Geant. oir od io Cuero iCrero cece iene ar oneal scutellata, n. sp Fig. 6. Fergusonina microcera, wing. Fig 7. Fergusonina atricornis, wing. Fig. 8. Fergusonina flavicornis, wing. Fig. 9. Fergusonina scutellata, wing. Fig. 10. Pseudoleucopis magnicornis, wing. Fig. 11. Aphaniosoma nigridorsum, head from front. Fig. 12. Huhippelates pallidiseta, wing. 92 NOTES ON AUSTRALIAN DIPTERA, Vi, FERGUSONINA MICROCERA Malloch. The male of this species is all yellow, except dark rings round the ocelli. The abdominal and femoral bristles are mostly luteous. Fifth visible abdominal tergite about as long as the two preceding tergites combined; hypopygium with base subglobose, the forceps long, slender, heavily chitinized, directed forward below venter. Hind femur stout, with a long preapical bristle and one or two shorter bristles basad of it on anteroventral surface. Length, 1.5-1.75 mm. Male, Sydney, N.S.W., 29.10.24. FERGUSONINA ATRICORNIS, N. SD. ¢o.—Ocellar spot and third antennal segment black. Thorax with six faint rufous vittae, the anterior margins of the fused median pair and the posterior extremities of the lateral pairs black; postnotum black. Abdominal tergites blackish at bases, especially laterally. Otherwise the species is similar to the preceding. Wings smoky hyaline. Orbital setulae not so large as in microcera; only one vertical bristle on each side of frons, very long; ocellars short. Anterior pair of prescutellar dorso- central bristles very short. In other respects as microcera except in venation. Length, 2 mm. Type, Sydney, N.S.W., 29.10.24. FERGUSONINA FLAVICORNIS, 0. Sp. 9.—Lemon-yellow, shining. Head with a black spot over ocelli. Dorsum of thorax with six black vittae, the lateral pairs fused in front of suture and extending farther backwards than median pair, all of them brownish just in front of suture; pleura with a few dark patches; postnotum black. Abdomen with the tergites all broadly black, only the narrow hind margins and their sides yellow; ovipositor glossy black. Legs yellow. Wings greyish hyaline. Halteres yellow. All hairs and bristles black. Head as in microcera. Anterior of the two pairs of dorsocentral bristles of moderate length. Length, 1.75 mm. Type, Sydney, N.S.W., 30.11.24. FERGUSONINA SCUTELLATA, 0. SD. ¢6.—A much darker species and more strongly bristled than the preceding one. Arista black, sides of occiput as well as ocellar spot black. Only the lateral margins of mesonotum, sutures of pleura, and apical margins of scutellum yellow. Dorsum of abdomen except narrow margins of tergites 1 to 4 and apical half of 5 black. Legs yellow, a fuscous streak on anterior side of hind femur. Wings greyish hyaline, veins conspicuous. All bristles black. Orbital bristles quite distinct. Similar to other species in chaetotaxy, etc. Length, 2 mm. Type, Sydney, N.S.W., 1.1.25. Subfamily OcHTHIPHILINAR. This subfamily lacks the bristles on vibrissal angles of cheeks, has the auxiliary vein of wing complete and separate from first on its entire length, and the postvertical bristles parallel or convergent. BY J. R. MALLOCH. 93 Genus PSEUDOLEUCOPIS noyvum. Generic characters.—Wing as in Figure 10, the auxiliary vein quite distinct and more widely separated from first than usual in this subfamily. Frons with two pairs of orbital bristles; ocellars distinct; postverticals rather small, con- vergent; antennae large; arista subnude; vibrissae lacking. Thorax with two pairs of dorsocentrals, the prescutellar acrostichals distinct; scutellum with four bristles; mesopleura bare; sternopleura with one long and one short bristle. Legs normal. Genotype, the first described species herein. PSEUDOLEUCOPIS MAGNICORNIS, Nl. Sp. °.—Head black, frons opaque, orbits, lunule and cheeks whitish pruinescent; face glossy in centre. Thorax slightly shining, black, with dense lead-grey pruines- cence. Abdomen glossy black, with faint brownish dusting. KFemora fuscous, fore coxae, tibiae, and tarsi brownish-yellow. Wings clear. Halteres white. Frons fully one-third of head width, a little longer than wide; orbits distinct to below upper bristle, the anterior bristle on an isolated differentiated spot; third antennal segment about three times as long as height of cheek and about one- fourth longer than high; face glossy, with a sharp vertical keel or carina in centre which slopes off to sides. Thorax with more than eight series of intradorsocentral setulae. Legs rather stout. Wing as in Figure 10. Length, 2.25 mm. Type, North Harbour, Sydney, N.S.W., 30.3.23. PSEUDOLEUCOPIS FLAVITARSIS, 0. Sp. °.—Differs from the preceding species in having the orbits grey and differen- tiated on their entire length, the face does not have a sharp central keel or carina, but is slightly convex in centre and entirely grey pruinescent, third antennal segment smaller, and the legs black, with only the tarsi yellow. Otherwise as magnicornis. Length, 2.5 mm. Type, Sydney, N.S.W., 12.10.24. Family Geomyzidae. This so-called family is difficult to define and may yet have to be linked with Agromyzidae, but tentatively I allow it to stand, in the hope that when I get round to placing before students of the order my key to the families of acalyptrates, I may be able definitely to assign to the different groups the status which they appear to be entitled to. Later I will revert to this subject; meanwhile, I merely describe a new species of one genus which has been located in the family by authors. Genus APHANIOSOMA Becker. This genus was erected for the reception of an Egyptian species described by Becker. Later I described a North American species which I placed in the genus. I have not seen the genotype but, except for the narrower cheeks of the present species and the presence of a mesopleural bristle, which is not mentioned by Becker, the species appears to belong here. I figure the head of the new species (Fig. 11). 94 NOTES ON AUSTRALIAN DIPTERA, Vi, APHANIOSOMA NIGRIDORSUM, N. sp. d6.—F rons yellow, face and cheeks yellowish-white, occiput black, its lower margin yellowish, ocellar triangle black, antennae reddish-yellow, darkened at the insertion of arista, palpi yellow, arista fuscous. Thorax stramineous, disc broadly black, shining, with slight grey pruinescence, the hind margin of the black mark with an angulate notch in centre leaving a yellow triangle in front of scutellum, sternopleura black, except above, mesopleura black on lower margin; postnotum black; scutellum yellow, black on sides at base. Abdomen shining black, yellow at extreme base. Legs yellow. Wings hyaline. MHalteres yellow. Head as in Figure 11. Thorax with two pairs of dorsocentral bristles, the anterior pair short; prescutellar acrostichals distinct; scutellum slightly flattened; basal pair of bristles much shorter than apical pair, which are convergent. Veins 3 and 4 slightly and gradually convergent apically; inner cross-vein at two-fifths from apex of discal cell; last section of fifth vein distinctly longer than pen- ultimate section of fourth, the latter not more than one-fourth as long as ultimate section of fourth. Length, 1.5 mm. Type, Sydney, N.S.W., 16.12.23. Paratypes, same locality, October and November, 1923-24, 12 specimens. This species is much darker in colour than either of the already described forms and the thoracic markings are distinctive. Family Chloropidae. Subfamily CHLOROPIN AE. Genus CHLOROPELLA novum. Generic characters.—Ocellar bristles distinct, divergent and directed forward and upward; orbits each with three or four short but distinct bristles; vertex with two bristles on each side, the inner one situated in front of the outer, incurved, the outer outcurved, both rather distant from eye; triangle present; interfrontalia haired; arista normal, dorsal on the rounded third antennal seg- ment; eyes very short haired; face subvertical, not carinate; cheek normal; proboscis stout. Thorax with a short humeral, one short anterior and a long posterior notopleural, one pair of long prescutellar dorsocentral bristles and a pair of long presuturals; scutellum with the basal pair of bristles much shorter than apical pair. Legs long and rather slender, hind tibia without a sensory area. Costa to a little beyond apex of third vein, the latter ending in front of, the fourth behind, apex of wing, the veins divergent. Genotype, the following species. CHLOROPELLA BIPARTITA, Nl. Sp. 9.—Shining pale yellow. A black spot over ocelli and another on middle of occiput. Thoracic dorsum witk three rufous vittae, the median one extending from anterior margin to beyond suture and black on both extremities, the lateral pair not extending so far forward but carried farther backward, and only their posterior extremities black; pleura without black marks, sternopleura orange coloured below; scutellum with a fuscous mark on each side at base; postnotum black. Abdomen with a small round fuscous spot on each side of first tergite. Legs yellow. Wings clear. Frontal triangle extending to anterior margin of frons; arista pubescent; cheek about one-sixth of the eye height. Dorsum of thorax with numerous pale BY J. R. MALLOCH. 95 decumbent hairs; disc of scutellum slightly flattened, with a few short dark hairs. Abdomen tapered apically. First costal division a little shorter than third, the latter fully two-thirds as long as second; penultimate section of fourth vein fully twice as long as outer cross-vein, about half as long as last section of fifth, one-fifth as long as last section of fourth, and about 1.5 times as long as penultimate section of third vein. Length, 1.5 mm. Type, Sydney, N.S.W., October 29, 1924. This is the only genus of the subfamily known to me which has distinct presutural bristles on thorax. Genus PacHyLorpHusS Loew. An Old World genus, most abundantly represented in Africa, and absent in the New World, so far as our records go. Distinguished from Chloromerus, which it resembles in having the hind femora much swollen and armed with spines below, by the much thickened subapical antennal arista which is densely black haired and appears about half as thick as the third antennal segment. The fourth wing-vein is very faint beyond the outer cross-vein, though traceable to margin behind wing tip. This is the first record of the genus from Australia. PACHYLOPHUS LUTEA, N. SD. ?.—Testaceous yellow, shining. Ocellar spot, a portion of each antennal fovea, and a small spot on upper inner mouth-margin blackish. Thoracic vittae rufous, not very conspicuous; a spot on each humerus, a streak on lower margin of mesopleura, a spot on hypopleura, and the postnotum blackish; sternopleura red below. Abdominal tergites darkened basally. Wings clear. Halteres yellow. Frons projecting as far as one-third of eye length; third antennal segment about 1.5 times as long as high; cheek fully one-fourth as high as eye. Thorax as in normal species of the genus Chlorops. Penultimate section of veins 3 and 4 equal, either about .75 times as long as last section of fifth vein. Length, 3 mm. Type, Hidsvold, Queensland (Bancroft). The yellow colour preponderates more than in any other species of the genus. Subfamily BoraNopiinak. Genus THYRIDULA Becker. I described Thyridula atroapicata in a previous paper on Australian Diptera and now add another species. These two species are very similar in structure and colour so that I consider it is unnecessary to describe fully the new one and merely present a synopsis of the distinguishing characters as follows: A. Face entirely yellow; scutellum black at apex only; hind legs yellow, the tibia slightly darker in middle; pleural spots not very large, deep black and contrasting sharply with the yellow ground colour of pleura ........... SIT CEG IG. PAC ECR nE RCM Or CHOROIR oO REe Ce ER EONAR n CRON ee Meee ease ie atroapicata Malloch. AA. Face glossy brownish-black except on parafacials; scutellum with a broad black central stripe; hind legs yellow, femur and tibia broadly black, pleural spots very large, not very sharply contrasting with the rufous aro aiel Cyollorene Cit oe bE, Sooo oodonoouoeoo ocd Boo ouoe bud 6 centralis, n. sp. 96 NOTES ON AUSTRALIAN DIPTERA, Vi, THYRIDULA CENTRALIS, nN. Sp. °.—The tarsal claws of the hind legs are, as in the male of atroapicata, much larger than those of the other legs and very much curved, sickle-shaped. Length, 4 mm. Type, Sydney, N.S.W., 9.11.24. It is possible, but hardly probable, that this is the female of atroapicata. PARAHIPPELATES ANOMALA, N. Sp. ?.—Head including antennae rufous yellow, upper half of frons and the occiput fuscous, shining, but with greyish pruinescence, cheeks and face whitish pruinescent; arista fuscous. Mesonotum and scutellum glossy olivaceous black, with a quite noticeable purplish tinge, two very faint submedian lines and iateral margins broadly grey pruinescent; pleura grey pruinescent. Abdomen varying from brown to fuscous, with grey pruinescence, apices of tergites yellowish. Legs tawny yellow, mid and hind coxae, at least mid and hind femora, and some- times fore pair also, largely blackish; same tibiae blackish except basally, apical two segments of all tarsi fuscous. Wings greyish, veins yellowish basally. Calyptra white. MHalteres pale. Ocellar bristles of moderate length; arista entirely nude; vibrissal setulae pale and weak; eye distinctly higher than long, more than twice as high as cheek. Dorsocentral thoracic bristles except the hind pair short but distinct; scutellum with basal marginal bristles shorter than apical pair, the discal hairs short. Spur of hind tibia almost indistinguishable from the surrounding hairs. The three principal costal divisions subequal; outer cross-vein about its own length from apex of fifth vein. Length, 3-4 mm. Type, Blue Mts., N.S.W., 15 Jan., 1922. Paratypes, Mt. Eba, S.A., north of east and west line (Campbell). This species differs from all the others already described in having the hind tibial spur very minute, in fact in some specimens practically absent. However, it is unmistakably a Parahippelates. The glossy dorsum of thorax with its purplish tinge is quite distinct from the thoracic colour of any other Australian species. It appears probable to me that pruinosa Thomson and possibly also ornatifrons de Meijere belong here. Both have four pairs of dorsocentral bristles, but neither the original describers nor Becker give sufficient details to permit of a definite opinion. Genus EUHIPPELATES novum. Generic characters.—Similar to Hippelates Loew, differing essentially in the form of the scutellum which is nearly as long as the mesonotum, about twice as long as its basal width, gradually tapered to tip, where it is about one-fourth as wide as at base, centre with a broad shallow sulcus narrower posteriorly, the sides with rather dense microscopic spinules, the sulcate portion with sparser spinules, apex with a pair of pale divergent bristles, lateral margins with much shorter pale bristles. Mesonotum with three linear sulci. Genotype, the following species. EUHIPPELATES PALLIDISETA, N. SD. °.—Head yellow, ocellar region broadly blackened and grey pruinescent, occiput similarly coloured except lower margin; antennae and palpi yellow. BY J. R. MALLOCH. 97 Thorax black, dorsum and upper part of pleura densely grey pruinescent, lower half of pleura mostly glossy black; scutellum black, paler in centre, yellow at apex. Abdomen shining fuscous, basal and apical segments yellow. Legs yellow, all femora broadly infuscated in middle. Wings hyaline, veins yellow, the fol- lowing portions blackish: costa just in front of apex of first vein, first vein on apical fourth, second vein at base below dark part of first vein, third vein from fork to just beyond inner cross-vein, fourth vein from before inner to beyond outer cross-vein, fifth vein on either side of outer cross-vein, and both cross-veins. Halteres yellow. All hairs and bristles yellow. Frons with regularly arranged sparse short hairs, the triangle not defined; eyes almost bare; face sunken, with a sharp median carina, the antennae half hidden when face is viewed from side; arista slender, almost bare; cheek narrow. Mesonotum with four or five series of short decumbent hairs between the sulci, and three long bristles on each side on posterior margin, the inner one just laterad of the outer sulci; humeri quite prominent, well distinguished from mesonotum, each with two short bristles, both backwardly directed. Hind tibial spur fully as long as greatest diameter of tibia, situated well in front of apex. A peculiarity of the wing is that the fifth vein has no noticeable flexure near middle of discal cell (Figure 12, p. 91). Length, 2 mm. Type, Sydney, N.S.W., 24 June, 1924. Allotype, male, same locality, 15.10.24. EKUHIPPELATES PALLIDISETA Var. PALLIPES nov. Similar to the typical form, but the legs are entirely yellow. Length, 1.5-1.75 mm. Type, female, Sydney, N.S.W., 30.11.24. Allotype and one male paratype, same locality, 30.11.24 and 29.11.24. HIPPELATES BANCROFTI, fh. Sp. °.—Head yellow; frontal triangle shining black except on its anterior and lateral posterior angles; antennae darkened only at insertion of arista, the latter and its hairs black; cheeks silvery white; palpi yellow; occiput black on upper half. Thorax shining yellow, with a black spot on middle of anterior margin well concealed by head, a rather large black spot on dorsum behind each humerus, a fuscous mark above each wing base, a black line along notopleural suture, a large subquadrate black mark in middle of hind margin which is indented on anterior side and extends narrowly over base of scutellum; pleura black along lower margin of mesopleura and on lower half of pteropleura; postnotum black. Abdomen black, basal segment yellow. Legs yellow. Wings hyaline. Halteres yellow. Ocellar bristles cruciate; orbital setulae quite distinct; arista distinctly pubescent; cheek as high as width of palpi. Scutellum convex, almost rounded in outline, with four marginal bristles and some very fine black discal hairs; sternopleura with some fine pale hairs and a black setula above. Hind legs rather stout, spur of tibia much curved, situated before apex, about three-fourths as long as basal segment of tarsus. Veins 3 and 4 divergent at apices. Length, 2 mm. Type, Hidsvold, Queensland, 26 April, 1924 (Bancroft). Distinguished by its thoracic markings from any species of the genus. The specimen was reared from cotton bolls. H CONTRIBUTIONS TO THE CYTOLOGY AND PHYLOGENY OF THE SIPHONACEOUS ALGAE. I. THe CyToLoGy oF THE GAMETANGIA OF CODIUM TOMENTOSUM (Stack.). By May M. WILLIAMS, B.Sc., Linnean Macleay Fellow of the Society in Botany. (Forty-two Text-figures. ) [Read 27th May, 1925.] Introduction. The Siphonales are usually described as consisting of those Algae in which the plant body is a coenocyte or a mass of coenocytic branches. Cell walls or septa rarely appear except in connection with the reproductive organs. The plant body may assume various forms, from simple filaments, to some of the most complex forms found amongst the Algae. Hitherto, algologists have devoted their attention to systematic work on these algae, very few records appearing of cytological investigation. Davis (1904) described oogenesis in Vawcheria, and in the same year Ethel S. Gepp described the sporangia of Halimeda. Stephens (1899) and Davis (1900, 1903, 1904) discuss the Siphonaceous Algae as a probable ancestral stock from which the Phycomycetous Fungi have been derived. Before such an hypothesis can be definitely accepted, it is necessary that further cytological details amongst the Siphonales themselves should be investigated. Further, a study of the morphology and life cycle of the chromosomes, in connection with the theoretical problem of alternation of generations, is becoming of vast importance at the present day. Yamanouchi, in 1909, drew attention to this vast field of research and the problems with which it abounds. The. classification of the Siphonaceous Algae which seems to be generally accepted at the present day is that of Oltmann. He recognizes two classes. 1. Siphonocladiales in which the structure is not wholly coenocytic, division walls appearing at frequent intervals along the filament. In this order are included the Cladophoraceae, Siphonocladiaceae, Valoniaceae, Dasycladiaceae and the Sphaeropleaceae. 2. Siphonales in which the plant body is a true coenocyte. This order includes the Codiaceae, Bryopsidaceae, Caulerpaceae and Vaucheriaceae. This classification recognizes the Codiaceae as the most primitive member of the Siphonales. The reproduction here is the simplest found in the order and takes place either by biciliated zoospores produced in spherical branches, or by fusion of biciliated heterogametes, although zoospores and gametes are known in but few genera. The evolution of a more and more complex plant body is characteristic of the Codiaceae. The Bryopsidaceae are little more advanced. Here asexual reproduction is unknown; sexual reproduction takes place by means of biciliated heterogametes produced in lateral branches. BY MAY M. WILLIAMS. 99 The Caulerpaceae produce a very complex plant body, the entire plant being coenocytic. The reproduction is so far unknown. The Vaucheriaceae are recognized as the climax family of the Siphonales. They are freshwater forms. The asexual reproduction takes place by means of a large compound multiciliated' zoospore. Sexual reproduction takes place by the union of minute biciliated spermatozoids with a non-motile egg. The genus Codium was first established by Stackhouse in 1795, although the exact nature of the reproductive organs was not described till many years later. Thuret (1850) described the reproductive organs to be of the nature of zoospores produced in sporangia. Later, Berthold described two kinds of zoospores, and maintained that fusion of these was necessary before the young thallus could be formed. They were produced in the same kind of sporangia on different plants, both were of the same general form, only one was larger than the other and of a dark green colour. These he termed the macrozoospores. The others, the microzoospores, were smaller and of a yellowish colour. Since fusion occurred before germination could take place, Berthold established that the macrozoospores were the female gametes and the microzoospores the male. Went (1889) maintained that the plants of Codium tomentosum were dioecious, and further, he described the experiments which, he considered, proved conclusively that the macrozoospores germinated parthenogenetically without any fusion with the microzoospores. It is generally agreed that this statement requires further confirmation. Oltmann (1904) observed and figured the fusion of gametes and the subsequent development of the zygote which germinates immediately. Codium tomentosum is a cosmopolitan species. It consists of an elongated cylindrical, spongy thallus, branching dichotomously and fixed to the substratum by means of rhizoids. The thallus consists of a densely interwoven mass of medullary filaments which give rise to numerous club-shaped palisade branches. The filaments are composed of a parietal layer of protoplasm which lines the mucilaginous cell wall and surrounds a large vacuole. In this parietal layer of protoplasm, numerous small nuclei and chromatophores are scattered. The latter are irregularly distributed through the protoplasm; usually at the base of the branches they are few, while at the apices they are usually so closely aggregated as to give a green colour of an intensity almost equal to black. This is so much so that it completely obliterates the colourless hyaline protoplasm. Usually a good deal of starch is present, associated with the chromatophores. Division walls are absent from the filaments and normally only arise in connection with the reproductive organs. It has been claimed that the coenocytic structure has been gradually attained by the formation of fewer and fewer partitions in a succession of filamentous plants. The palisade branches are usually several times the diameter of the medullary branches which produce them. The palisade branches in their turn produce the gametangia, sometimes borne singly, but often produced in a row on either side of the palisade branch and are cut off from the latter by a plug-like wall (Text- fig. 1). As the gametangia mature, they drop off and a scar of attachment is left on the palisade branch. The writer’s observations confirmed the state- ment that male and female gametangia are produced on different plants, i.e. the plants are monoecious. The material for the present investigation was collected at Bondi, N.S.W. Here it grows profusely just outside the limit reached by the low tide. The material was fixed in the field in a weak chromo-acetic solution: Chromic acid 100 CYTOLOGY AND PHYLOGENY OF THE SIPHONACEOUS ALGAE, i, 0.4 g., Glacial acetic acid 1 c.c., Sea water 400 c.c. As the plant is of such a spongy nature, an air pump was used to remove as much of the air as possible to facilitate the infiltration of the fluid. The fixative was allowed to act for about 24 hours. Material was then washed thoroughly in sea water and passed in the usual method into paraffin. Sections were made varying from 3u to 5y in thickness and stained, Flemming’s triple stain and Haidenhain’s Iron-alum Haematoxylin being employed. The latter staining process gave the better results. Investigation. The gametangia are of two kinds, male and female, and are produced on different plants. The male gametangia are very much smaller than the female. Both contain very dense, coarse, granular cytoplasm, in which are embedded numerous nuclei and chromatophores. This is enclosed by a wall, distinctly two- layered, the outer layer being firm and thin, the inner thick and capable of a great degree of swelling. The gametangia are produced from the palisade branch in the following manner. A small protuberance is formed from the wall of the branch and into this a certain amount of protoplasm together with nuclei and chromatophores flows. The protuberance enlarges and the protoplasmic flow continues until a certain limit is reached, when the gametangium is cut off from the palisade branch by a thick septum. Definite evidence of this flow is recorded in the gametangium itself by the arrangement of the chromatophores, all of which have their long axis arranged almost parallel to the long axis of the gametangium, giving it a striated appearance. The nuclei, also, become elongated with their long axis arranged in the same direction. The protoplasm is evenly distributed, no vacuoles being visible from the beginning of growth. When the cross partition is formed, the gametangium is filled with colourless granular cytoplasm, nuclei, and very numerous chromatophores (Text-fig. 2). When development has proceeded thus far, certain changes occur in the game- tangia which are essentially similar in both male and female. The gametangium at first contains numerous very small, elongated nuclei, each containing a nucleolus, a very fine linin reticulum and surrounded by a thin nuclear membrane. The nuclei once more assume a rounded form. Very soon certain nuclei commence to disintegrate; the reticulum and membrane disappear, leaving the nucleolus which stains very brightly, but finally it also disintegrates. Co-ordinate with this nuclear degeneration, is the enlargement of certain other nuclei which are destined to undergo a heterotypic division. A. Mitosis in Female Gametangium. The nuclei as they enter the gametangium are small and elongated in the direction of flow of the protoplasm. In this condition they are more numerous at the apex of the gametangium than at its base. As soon as the dividing septum is formed, however, they lose their elongated appearance and become rounded once more, at the same time becoming more evenly distributed throughout the protoplasm. So far as the writer was able to observe, no definite number of nuclei enters the gametangium. The nuclei at this period have a thin nuclear membrane, a fine chromatin net work and a small nucleolus (Text-fig. 3). Some of the nuclei now commence to disintegrate, the nucleolus persisting till the last and then finally disappearing (Text-fig. 4). These supernumerary nuclei BY MAY M. WILLIAMS. 101 have completely disappeared before the mitoses which take place in the game- tangium are completed. Certain other nuclei enlarge considerably; at the same time the chromatin thread becomes very much more distinct. The nucleolus does not increase in size (Text-fig. 5). The selection of functional nuclei seems to be related to their orientation with regard to certain dynamic centres present within the gametangium. It is the nuclei in the vicinity of these centres which survive, and those further away perish. The nature of these bodies is indicated in Text-fig. 8. They are not very large, but absorb the stain with avidity so that they are easily recognizable in preparations. These bodies appear to be of the nature of dense protein centres surrounded by clear, colourless zones. Several such bodies are present in each gametangium. Observations indicated that they nourish several nuclei in their vicinity. They seem to be of a chemotactic and nutritive nature. A peculiar feature is that they are small bodies, smaller than the nuclei themselves. The bodies present to the writer very much the same structure and serve the same function as the coenocentra described by Davis (1903) in Saprolegnia and by Stevens (1899) in Albugo. The term coenocentrum, in these. fungi, is applied to the centre of a coenocytic body, such a centre being connected with definite processes in oogenesis. The coenocentrum, also, is a very large structure, very much larger than the nuclei. This is precisely the reverse of the writer’s observations in Codiwm. Hence, it would not seem justifiable to term these bodies, present in Codiwm, coenocentra. As the nuclei in the vicinity of these bodies enlarge they are seen to contain a thin, ragged and much branched chromatin thread. There is no manifestation of polarity of the nucleus. Soon it becomes evident that the network is composed of very small chromatin knots connected together by irregular fibrils (Text-fig. 5). The chromatin knots become very much larger, especially those near the nuclear membrane. Co-ordinate with this, the nuclei increase in size still further; ultimately the chromatin thread is located at one side of the nucleus. This growth period of the nucleus eventually results in its becoming two or three times its original size, but as the chromatin mass does not increase at the time, it becomes left at one side, giving the typical appearance of the so-called synapsis, as indicated in Text-fig. 6. This stage forms a striking feature of the cytology of the reduction division, but appears nowhere else in the whole life history of the plant. Owing to the minute size of the nuclei it is difficult to determine the factors responsible for synapsis, but certain it is that associated with it is a considerable increase in the size of the nucleus. At the same time, the threads of the chromatin undergo considerable modification, they become very much thicker, and the general appearance seems to indicate a shortening and thickening of the threads. The chromatin thread thus becomes more sharply defined, but still has the appearance of being irregularly interwoven even in synapsis. At this stage the nucleolus may be recognized associated with the chromatin threads. Soon the chromatin threads commence to loosen, and certain of them project into the nuclear cavity (Text-fig. 7). At the same time the chromatin threads have continued their processes of shortening and thickening. The exact mechanism of the separation of the individual chromosomes could not be clearly determined. Whatever the method employed, ten pairs of bivalent chromosomes are to be observed in the nucleus in early prophase (Text-fig. 9). These are usually distributed through the nuclear cavity. In late prophase, the nucleolus is sometimes to be observed in the nucleus in a fragmentary condition. The spindle fibres begin to make their appearance, 102 CYTOLOGY AND PHYLOGENY OF THE SIPHONACEOUS ALGAE, i, and attach themselves to the chromosomes, thus the achromatic spindle, with its equatorial plate, is established. The spindle is long and narrow, and its long. axis is invariably parallel to that of the gametangium (Text-figs. 10, 12). A distinct membrane surrounds the spindle at this period, but on account of the minute size of the nuclei, it is difficult to determine whether this is the original nuclear membrane. If the membrane is of this nature, it would follow that the spindle would be intranuclear, but that is by no means certain. Centrosomes with radiations were not observed. Davis (1903, p. 230) notes the absence of these structures from the intranuclear spindles of Saprolegnia. Centrosomes may be present in Codium tomentosum, but the granular nature of the cytoplasm, the large chromatophores, and the avidity with which they stain, would make such minute structures difficult to determine. The chromosomes are long rod-shaped structures, hence their arrangement on the equatorial plate is not always clear. In polar view of the various stages their number may be fairly accurately determined. The result of making a count of the number of chromosomes in the polar view of the metaphase of several hundred median sections proved them to be ten in number, but these are bivalent (Text-fig. 11). In metaphase the two halves of the bivalent chromosomes separate and pass towards the poles. Anaphase follows; the chromosomes begin to straighten out and are arranged near the poles (Text-fig. 13). After telophase, the two daughter nuclei are organized, the chromosomes lose their individuality Text-fig. 1—A palisade branch bearing a gametangium and showing the scars of attach- ment of older gametangia (x 60). Text-fig. 2—Young gametangium (female) indicating the distribution of nuclei and chromatophores (x 150). Text-figs. 3-21.—Mitosis in Female Gametangium. 3-14. First Division. 3.—This indicates the condition of the nucleus at the period of its entrance into the gametangium (x 2000). 4.—A degenerating nucleus (x 2000). 5.—A functional nucleus indicating increase in size and the nature of the chromatin reticulum which consists of granules connected together by delicate fibrils (x 2000). 6.—Nucleus in synapsis (x 2000). 7.—This indicates the loosening of the chromatin mass in synapsis, the chromatin threads projecting into the nuclear cavity in the form of loops (x 2000). 8=—This indicates the nature of the bodies which appear to select the functional nuclei. In the vicinity are three nuclei in synapsis (x 2000). 9.—Prophase: 20 chromosomes are developed and are distributed throughout the nuclear cavity, and are grouped in pairs (x 2000). 10.— Metaphase: the spindle has developed and the chromosomes are arranged on the equatorial plate (x 2000). 11.—Polar view of nucleus in metaphase indicating the presence of ten chromosomes (x 2000). 12.—A gametangium indicating the simultaneous division of the nuclei which are in metaphase. The long axes of the spindles are parallel to that of the oogonium (x150). 13.—Anaphase: the two sets of daughter chromosomes are passing towards the poles (x 2000). 14.—The two daughter nuclei are organized and the chromosomes are losing their identity. The spindle fibres are still faintly visible connecting the two nuclei (x 2000). 15-21. Second Division. 15.—Late prophase: a number of chromosomes have emerged from the ragged reticulum. The nucleolus is still present (x 2000). 16.—Metaphase: the two sets of daughter chromosomes are proceeding to the poles (x 2000). 17.—Polar view of prophase, indicating that there are 10 chromosomes associated with this division (x 2000). 18.— Anaphase: the two sets of daughter chromosomes are proceeding to the poles (x 2000). 19.—Longitudinal section of a mature gametangium indicating the nature of the female gamete primordia (x 270). 20.—The female gamete has a rounded contour. Nucleus is in a central position (x 610). 21.—The nucleus of the female gamete has taken up a peripheral position (x 610). BY MAY M. WILLIAMS. 103 104 CYTOLOGY AND PHYLOGENY OF THE SIPHONACEOUS ALGAB, i, and the ragged reticulum is once more established. The daughter nuclei are much smaller than the parent nucleus from which they were formed (Text-fig. 14). After a short period of rest the daughter nuclei undergo another division, which is essentially similar, except in the number of chromosomes, to the ordinary mitotic divisions throughout the plant. In early prophase, ten univalent chromosomes are differentiated out from the ragged reticulum (Text-fig. 15). In late prophase the achromatic spindle, with its equatorial plate, is established. A thin membrane surrounds the spindle, but soon undergoes rapid degeneration. No centrosomes or radiations were observed associated with the spindle. The long axis of the spindle of the second division is also parallel to that of the gametangium. In metaphase the chromosomes divide and pass to the poles (Text-fig. 16). Where the splitting of the chromosomes occurs which provides for the second division could not be accurately determined. Anaphase (Text- fig. 18) and telophase follow, and the daughter nuclei are organized, each with its well-defined reticulum and nuclear membrane. In prophase, ten chromosomes are differentiated out in polar view, and this number remains constant throughout the mitosis (Text-fig. 17). The daughter nuclei are very small compared with the parent nucleus of the first division. These are the only mitoses which occur in the organization of the female gametangium. They are simultaneous in all the nuclei concerned. Text-figs. 22-35.—Mitosis in the Male Gametangium. 22-29. First Division. 22.—A functional nucleus indicating the well defined chromatin reticulum which is differentiated into granules connected together by delicate fibrils. A degenerating nucleus is also indicated here (x 2000). 3.—Nucleus in synapsis (x 2000). 24.—This indicates the loosening of the chromatin mass in synapsis; certain of the chromatin threads are seen projecting into the nuclear cavity in the form of loops (x 2000). 25.—Prophase: the chromosomes are approximately 20 in number and are grouped together in pairs (x 2000). 26.—a. Metaphase: chromosomes are just separating to pass towards the poles (x 2000). b. Metaphase: in the vicinity of the spindle is one of the structures which select the functional nuclei (x 2000). 27.—Longitudinal section of the gametan-- gium indicating the simultaneous division of nuclei which are in this instance in metaphase (x 150). 28.—Polar view of chromosomes at metaphase indicating that they are ten in number (x 2000). 29.—Late anaphase: the two sets of daughter chromosomes are nearing the poles of the spindle (x 2000). 30-35. Second Division. 30.—The nucleus at the beginning of division. The reticulum is being differentiated into numerous chromatin knots (x 2000). 31.—Metaphase: the chromosomes have separated (x 2000). 32.—Polar view of late prophase indicating that ten chromosomes are present (x 2000). 33.—Longitudinal section of the male gametangium indicating the nature of the male gamete primordia (x 270). 34.—The male gamete before liberation ; it has a rounded contour and the nucleus is in a central position (x 610). 35.—The nucleus of the male gamete has migrated and taken up a veripheral position (x 610). Text-figs. 36-42.—Mitosis in Vegetative Coenocyte. 36.—A small nucleus of the central coenocytic threads of the plant indicating scarcity of chromatin material (x 2000). 387.—A nucleus of a palisade branch preparing for division; the chromatin material is more distinct, chromatin knots are being organized (x 2000). 38.—This indicates the nature of the spireme which is beginning to segment (x 2000). 39.—lLate prophase (polar view); chromosomes are differentiated and are 20 in number (x 2000). 40.—Metaphase: the chromosomes are arranged on the equatorial plate (x 2000). 41.—Same as the above (Text-fig. 40) viewed from the poles, indicating that 20 chromosomes are present (x 2000). 42.—Telophase: the two sets of daughter chromosomes have reached the poles and are losing their identity (x 2000). BY MAY M. WILLIAMS. 105 As soon as the daughter nuclei are formed, they separate and become evenly distributed throughout the cytoplasm of the gametangium. The protoplasm surrounding each nucleus, taking the latter as a centre of formation, becomes marked out by cleavage planes. Cell membranes make their appearance for the first time within the gametangium; each nucleus with its attendant cytoplasm is a female gamete primordium (Text-fig. 19), and becomes transformed into a single female gamete. At first, these are hexagonal in outline and fit closely to one another, but later, when ready for liberation, they separate and become rounded (Text-fig. 20). The nucleus then migrates from the centre of the gamete to a peripheral position (Text-fig. 21) and the gamete becomes pear- shaped. Cilia appear at the pointed end of the gamete in proximity to the nucleus. The gametangium opens by a pore at its apex through which the gametes escape. The liberation of the gametes is due to the shrinking of the wall of the gametangium, brought about by exposure and relaxation of pressure 106 CYTOLOGY AND PHYLOGENY OF THE SIPHONACEOUS ALGAE, i, at low tide. There are several hundred gametes produced in each gametangium. As there are countless numbers of gametangia produced in each branch of the plant, it follows that the gamete production is enormous. B. Mitosis in the Male Gametangium. The male gametangium is very much smaller than the female, but the pro- cesses connected with the organization of the gamete are similar in both. When the septum is formed, separating the male gametangium from the palisade branch which bears it, certain of the nuclei degenerate, while others enlarge considerably. There is an apparent marked increase in the amount of chromatin in the surviving nucleus, but no increase in the size of the nucleolus (Text-fig. 22). The nuclei which survive are those in the vicinity of certain bodies of the same nature as those described as occurring in the female gametangium. Such a structure is indicated lying close to the spindle in Text-fig. 26, b. Co-ordinate with the increase in size of the surviving nuclei, small granules make their appear- ance on the ragged reticulum, these being connected by delicate fibrils. With further increase in the size of the nucleus, the chromatin threads become eccentrically placed in the nuclear cavity, presenting the appearance of typical synapsis (Text-fig. 23). The nuclei are too minute to observe the detail of forma- tion or structure in synapsis. The chromatin threads undergo shortening and thickening so that they are very much more distinct in synapsis than in the pre-synaptic stages. The chromatin in synapsis presents the appearance of a mass of closely interwoven threads (Text-fig. 23). Later the chromatin threads commence to loosen, and certain of them may project into the nuclear cavity (Text-fig. 24). The chromatin thread arranged in synapsis must undergo further shortening and thickening, since the chromosomes when they are formed are short, thick structures. In prophase, 20 chromosomes were counted; these were grouped in pairs, each representing a pair of bivalent chromosomes (Text-fig. 25). These chromosomes become distributed in the region of the centre of the nucleus. In late prophase the spindle fibres appear; they attach themselves to the chromosomes, and the achromatic spindle, with its equatorial plate, is established. The spindle is long and narrow; it is surrounded by a thin membrane, well defined in metaphase, but during anaphase it breaks down, disintegration com- mencing at the poles. No centrosomes or radiations were observed in connection with the formation of the spindle. The long axis of the spindle is parallel to that of the gametangium (Text-figs. 26, 27). In metaphase the chromosomes separate and pass towards the poles; ten chromosomes were counted in polar view of this stage (Text-fig. 28). Anaphase follows (Text-fig. 29); at telophase, the two sets of daughter chromosomes are crowded closely together and nuclear membranes are formed. The chromosomes lose their identity, and a ragged reticulum is formed; the nucleoli also reappear. After a short period of rest, the daughter nuclei undergo another simultaneous division, similar to that which occurs in the female gametangium. There are ten chromosomes associated with the various stages of this division, as indicated in Text-figure 32, which represents a polar view of late prophase; with this reduced number the nucleus passes into telophase. The daughter nuclei are very small. The nature of this division is indicated in Text-figures 30-32. When this mitosis is complete, cleavage planes are formed in the cytoplasm, cell membranes make their appearance surrounding each nucleus and its attendant cytoplasm. This is the first appearance of cell walls within the gametangium. BY MAY M. WILLIAMS. 107 The nucleus with its attendant cytoplasm becomes a male gamete primordium, each giving rise to a single male gamete. These are at first hexagonal in outline and lie close together (Text-fig. 33), but later they separate and become rounded in outline (Text-fig. 34). Co-ordinate with these changes in the shape of the gamete, the nucleus which is originally centrally placed, moves, and takes up a peripheral position (Text-fig. 35). Cilia are developed on the side of the gamete near the nucleus. Liberation of the gamete is effected by the same means as has been already described in the female gametangium. The male gametes are very much smaller than the female and of an orange colour. Countless hundreds of gametes are produced in each gametangium. C. Division of the Nuclei of Vegetative Coenocyte. The nuclei of the coenocytic threads of the central region of the thallus are very smail, in fact they are little larger than the chromatophores themselves (Text-fig. 36). They contain a small nucleolus and a scarcely visible chromatin reticulum. These nuclei are scarcely satisfactory for a study of nuclear conditions in the coenocyte. The nuclei occurring in the outer branches, especially those of the palisade branches, however, are somewhat larger. A thorough study of a typical mitosis was made in order to make an estimate of the number of chromo- somes present in the vegetative body grown under normal conditions. Vege- tative mitosis agrees in essentials in both male and female plants. The nuclei occurring at the apices of the palisade branches contain a small nucleolus and a faint chromatin reticulum. In early prophase the nucleolus commences to degenerate, while the chromatin reticulum becomes very much more prominent and dispersed over the nuclear cavity. Chromatin knots become > visible (Text-fig. 37), resulting in the reticulum gradually becoming transformed into a convoluted spireme (Text-fig. 38). The thin spireme thickens and later splits to form chromosomes. These become regular in size and shape (Text-fig. 39). In late prophase spindle fibres make their appearance and numerous chromo- somes are arranged on the equatorial plate. In metaphase the chromosomes split and each half proceeds towards the poles of the spindle (Text-fig. 40). The nuclear membrane is only to be observed in the equatorial region of the nucleus; in anaphase it has completely lost its identity. In late prophase and in polar view of metaphase, 20 chromosomes were recognized as being present (Text-fig. 41). They are often difficult to count on account of their number in comparison with the small size of the nucleus. It seems probable that the splitting of the parent chromosome occurs when they have become arranged on the equatorial plate. In anaphase, the two sets of daughter chromosomes proceed to the poles; telophase follows (Text-fig. 42). When they reach the poles, nuclear membranes are formed, the chromosomes lose their identity and two typical resting nuclei result. General Considerations. The present communication is of interest in that it indicates that alternation of generations is present associated with a fairly primitive form of sexual reproduction; the gametes are isogamous and but little removed from the asexual zoospore from which they are supposed to have been derived. As the history of the discovery and establishment of alternation of genera- tions among the Thallophytes has been so admirably set forth by Yamanouchi (1912), it would be superfluous to relate it here, except to quote a few 108 CYTOLOGY AND PHYLOGENY OF THE SIPHONACEOUS ALGAE, i, of that author’s words which briefly summarize the present extent of the investi- gations in this direction (p. 485): “The existence of an alternation of generations among the Thallophytes, though somewhat obscure in the green algae on account of insufficient investigations, has been clearly proved by cytological study on Fucus (Yamanouchi, 1909), Cutleria (Yamanouchi, 1912), and Dictyota (Williams, 1904) among the brown algae, and in Nemalion (Wolfe, 1904), Polysiphonia (Yamanouchi, 1906), Griffithsia (Lewis, 1909) and Delesseria (Svedelius, 1911) among the red algae.” In addition to these, Williams established the presence of an alternation of generations in Laminaria in 1921. The investigation recorded in the present communication indicates that alternation of generations occurs in the life history of Codiwm tomentosum, although it is to be regarded as either a cytological alternation or as an aberrant type. The nuclei of the coenocytic threads of the vegetative portions of the plant contain 20 chromosomes. The nuclei of the gametangia, both male and female, undergo two divisions, in the first of which 10 bivalent chromosomes are present; the second division has the same number of univalent chromosomes. There are two possibilities as to the nature of the alternation of generations in Codium. In the first instance, since the nuclei of vegetative coenocytic branches contain the diploid number of chromosomes, they may be regarded as sporophytic in character. The nuclei of the young gametangium (before division occurs) may be compared with the spore mother cells of higher plants (in Codium, cell walls do not appear separating the spore mother cells, but they are enclosed by a common membrane). Certain of the nuclei degenerate; the surviving nuclei undergo two divisions. The daughter nuclei which result “from the second division may be compared with the spores of higher plants, that is, they are the beginning of the haploid generation. Still enclosed, they germinate to form the gametophytic generation. Hach nucleus of this generation, with its attendant cytoplasm, by cell formation becomes a gamete primordium, each eventually becoming a single uninucleate gamete. These gametes are liberated, fuse and germinate; the sporophytic generation is once more produced. This implies that alternation of generation in this type is purely cytological with probably a primitive type of gametophyte. The main objection to this explanation is that it is impossible to determine in the haploid generation where the haploid spore ends and the gamete primordium begins. Cytological alternation has been recorded in Fucus’ (Yamanouchi, 1909), but here the gametophytic generation, when initiated, undergoes one mitosis in the oogonium and four in the antheridium. The second possibility is that Codium may be regarded as an aberrant type so far as the generally accepted ideas of alternation of generations are concerned. This may be due to the fact that the reproductive organs are primitive in character; the gametes are but little removed from the asexual zoospore from which they are supposed to have been derived. The term “sporophyte” is usually applied to that portion of the life history which contains the diploid number of chromosomes and produces the spores; these in their turn produce the “gametophyte” or the haploid generation which produces the gametes. Mor- phologically, spores are not produced in the life cycle of Codium, hence the suggestion that, since the plant produces haploid gametes directly, but is itself diploid in nature, it may be regarded as a diploid gametophyte. The results of the investigations of many cytologists in other parts of the plant kingdom have led to the establishment of many definite facts with regard BY MAY M. WILLIAMS. 109 to the phenomenon of alternation of generations. Taking these into consideration, the best explanation of its occurrence in Codium tomentosum seems to be that it is cytological in character, with probably a primitive type of gametophyte. The various details accompanying gametogenesis in Codium tomentosum also present considerations of great interest. All the surviving nuclei in the game- tangium, both male and female, are to be found at the same stage of mitosis at precisely the same time, that is, simultaneous division of nuclei occurs. This has been recorded from various parts of the plant kingdom. Stevens (1899) and Davis (1903) emphasize its occurrence in Albugo and Saprolegnia respectively amongst the Phycomycetous Fungi. Yamanouchi has recorded it in the oogonia and antheridia of Fucus (1909), and in oogenesis, spermatogenesis and zoosporo- genesis in Cutleria (1912). In Codium this simultaneous division is of the nature of a reduction division necessary in the organization of the gametes. Nuclei of all ages enter the gametangium, so that this division not only maintains constancy in the number of chromosomes, but also serves to bring the nuclei to the one age once more. Previous to these mitoses, degeneration of certain nuclei occurs, these nuclei representing supernumerary sperm mother nuclei. The factors responsible for the selection of the functional nuclei are very interesting. Smali granular bodies which stain deeply, and of a different nature from the ordinary cytoplasm, make their appearance, embedded in the cytoplasm. They first appear as visible structures at the period when certain of the nuclei are degenerating and others are preparing for division. It is the nuclei in the vicinity of these bodies which enlarge and undergo a heterotypic division, that is, the surviving nuclei owe their preservation to their orientation with regard to these structures. They not only nourish one nucleus in their vicinity, but several. They are not permanent structures of the gametangium, but are of a transitory nature. They appear associated with the most active periods of gametogenesis and disappear when the first nuclear division is well advanced. They represent centres of dynamic activity, their function resembling that of the coenocentra described by Davis (1903) in Saprolegnia, and the single coenocentrum by Albugo (Stevens, 1899). Their earlier appearance in Codium tomentosum is probably related to the differences in the processes of formation of the structures with which they are associated in these types. A rather remarkable feature in Codium is their small size as compared with the nuclei. Whether they are vestigial or primitive structures amongst these algae is impossible to determine. Much interest: has become centred around this region of the Siphonales (Codium) of recent years on account of the fact that it is these plants with simple forms of gametangia, producing uninucleate swarming spores, which have been looked to as the possible ancestors of the Phycomycetous Fungi such as the Mucorales, Saprolegniales, and Peronosporales. It is generally agreed that the coenogamete is homologous with the gametan- gium. In the light of the present communication, it would seem that a further study of the nucleus of the Phycomycetous Fungi with regard to the significance of the mitoses which occur during oogenesis and spermatogenesis of these types is very desirable. Both Stevens (1901, p. 242) and Davis (1903, p. 251) consider the nuclear divisions of the oogonia and antheridia of Albugo and Saprolegnia respectively are not of the nature of reduction divisions, but are probably phylogenetic reminiscences, essential for the increase in the number of gametes in ancestral forms. Should this remark be proved to be correct, it follows that 110 CYTOLOGY AND PHYLOGENY OF THE SIPHONACEOUS ALGAE, i, the gametangia of Codium, at all events, are not homologous with the coenogametes of these Fungi, since such an homology would demand an agreement even in the details of cytology. Summary. 1. The filaments which go to make up the plant body of Codiuwm are coenocytic in structure; the nuclei of these coenocytic threads contain 20 chromosomes. 2. Gametangia are borne on the palisade branches; the plants are monoecious. 3. Certain nuclei of the gametangium (both male and female) undergo degenera- tion; others enlarge considerably and undergo two nuclear divisions. 4. The nuclear divisions of the gametangium are simultaneous in all nuclei concerned. 5. Selection of the functional nuclei is closely associated with the presence of a varying number of bodies, of the nature of coenocentra, within the gametangium. §. The first of the nuclear divisions which takes place within the gametangium is heterotypic in character. Ten chromosomes are present during this division, but these are bivalent. 7. The same number of univalent chromosomes are to be found associated with the second division; hence, the daughter nuclei associated with this division contain the reduced number of chromosomes. 8. The achromatic spindles are in all instances probably intranuclear and with- out centrosomes or radiations. 9. The period during which ten chromosomes are present may be regarded as the haploid stage; gametophyte generation is probably a primitive type. 10. Alternation of generations in Codiuwm tomentosum is purely cytological in character. For helpful advice and criticism during the course of this investigation, the author desires to express sincere thanks to Professor A. A. Lawson, in whose laboratory the investigation was carried out, also to Dr. J. McLuckie and Mr. P. Brough. Bibliography. ALLEN, EH. C., 1905.—Nuclear division in the pollen mother cells of Liliwm canadense. Ann. Bot. xix, 189-285. BEssEy, C. H., 1905.—The structure and Classification of the Siphonales. Trans. Amer. Micro. Soc. xxvii, 47-62. BLACKMANN, F. F., and TANsuLEY, A. G., 1902.—A revision of the Classification of the Green Algae. New Phytologist i, 17-23, 67-72, 84-96, 114-120, 133-144, 163-168, 189-192, 213-220, 238-241. Bower, F. O., 1890.—On Antithetic as distinct from Homologous Alternation of Genera- tions in plants. Ann. Bot. iv, 347-370. Carter, Nellie, 1919.—The Cytology of the Cladophoraceae. Ann. Bot. xXxxili, 467-478. Couuins, F. S., 1909.—The Green Algae of North America. Tufts College Studies ii, No. 3, Sci. Ser., 80-462. Davis, B. M., 1900.—The Fertilization of Albugo candida. Bot. Gaz. xxix, 297. 1903.—Oogenesis in Saprolegnia. Bot. Gaz. xxxv, 223-320. —— 1904.—Oogenesis in Vaucheria. Bot. Gaz. xxxviii, 81-96. 1908.—Spore formation in Derbesia. Ann. Bot. xxii, 1-20. DeERBES and Souigr, 1856.—Mémoire sur quelques points de la physiologie des algues. Compt. Rend. Acad. Sci. i, 59-72. Dixon, H. H., 1897.—Structure of Codiwm. Ann. Bot. xi, 588-590. Gupp, Ethel S., 1904.—The Sporangia of Halimeda. Journ. Bot. xlii, 193-197. Lawson, A. A., 1911.—The Phase of Nuclear Division known as Synapsis. Trans. Roy. Soc. Hdin. xlvii, 591-604. BY MAY M. WILLIAMS. mili Murray, G., and Boonie, L. A., 1889.—A systematic and structural account of the Genus Avrainvillea. Journ. Bot. xxviii, 97-101. STEVENS, F. L., 1899.—The compound Oosphere of Albugo Blitii. Bot. Gaz. xxviii, 149-243. —— 1901.—Gametogenesis and Fertilization in Albugo. Bot. Gaz. xxxii, 77-98, 157-169, 238-261. STRASBURGER, E., 1894.—The periodic Reduction of the number of Chromosomes in the life history of living organisms. Ann. Bot. viii, 281-316. WENT, F. A. F. C., 1889.—Les Modes de Reproduction du Codiuwm tomentosum. Nederlandsch Kruidkundig Archief. 5e Deel, 3 Suite, 440-444. West, G. S.—Algae. Vol. 1. Camb. Bot. Handbooks. WILLIAMS, J. L., 1904.—Studies in the Dictyotaceae. i. Cytology of the tetraspore and germinating tetraspore. Ann. Bot. xviii, 141-160. ii. Cytology of the Gametophyte. Ann. Bot. xviii, 183-204. —— 1921.—The Gametophyte and Fertilization of Laminaria and Chorda. Ann. Bot. Xxxv, 603-607. WOLFE, J. J., 1904.—Cytological Studies on Nemalion. Ann. Bot. xviii, 607-630. YAMANOUCHI, S., 1906.—The Life History of Polysiphonia. Bot. Gaz. xlii, 401-449. 1909.—Mitosis in Fucus. Bot. Gaz. xlvii, 173-197. — 1909.—Cytology of Cutleria and Aglaozonia. Bot. Gaz. xlviii, 380-386. —— 1912.—The Life History of Cutleria. Bot. Gaz. liv, 441-502. THE GEOLOGY AND PETROGRAPHY OF THE CLARENCHETOWN- PATERSON DISTRICT.* Part iv. PETROGRAPHY. By G. D. Osporne, B.Sc., Linnean Macleay Fellow of the Society in Geology. (Plate xxiii.) [Read 27th May, 1925.] Introduction. The present paper completes the series on the Clarencetown-Paterson district. Former papers have dealt with a discussion of the stratigraphy, physiography and tectonic geology of the whole region, and also with a detailed study of the Main Glacial Beds at Seaham (Osborne, 1922, 1925). In the following pages a detailed petrographical account of the Kuttung Series is given, and some little mention is made of certain Cainozoic basic rocks which occur in small masses throughout the area. The rocks of the Burindi Series are not described, because the area of their outcrops is small and also because these outcrops are the southward continua- tions of large areas to the north, where important massive igneous rocks occur. It is hoped that an examination of these will be possible in the future. The greater part of the material in the present paper is of a descriptive nature. The author feels that the time has not yet come when the broader problems of petrogenesis in connection with the Carboniferous igneous rocks can be solved, nor can much be done regarding the correlation of the Carboniferous areas to the north near Tamworth and Werris Creek with those of the Lower Hunter district. These things will only be possible after some examination has .- been made of the Carboniferous country between Singleton and Murrurundi— work which the writer hopes to pursue in the immediate future. Therefore in the following discussion a systematic account will be given of the rocks, taken in groups according to their identity, and some general remarks made upon certain interesting features exhibited by the rocks when considered as a whole. It should be pointed out that in 1919 a short report on the petrology of some of the rocks from parts of the district now being considered, was made by W. R. Browne (Sussmilch and David, 1919, Appendix 2). This report, though brief, proved of considerable help to the writer in connection with his introduction to the problems of the Carboniferous igneous rocks. In fact the author wishes here to emphasize his indebtedness to Assistant-Professor Browne for his help and advice during the whole of his work on the petrology of the Carboniferous rocks of New South Wales. It must also be mentioned, that, as a result of his studies in the Gosforth-Helah region, near Maitland, Dr. Browne has observed many facts about * About three-fourths of the work entailed in the preparation of this paper was carried out while the writer held the position of Demonstrator in Geology, University of Sydney. BY G. D. OSBORNE. 113 the Kuttung petrology which have not yet been placed on record by him, but which have been arrived at independently by the present author in connection with his work in the Paterson-Dungog area. In connection with the chemical investigation of the rocks, the writer is very grateful to Mr. H. P. White, F.C.S., through whose good offices an analysis was kindly performed by Mr. W. A. Greig. Thanks are also due to Mr. G. W. Card and Mr. C. A. Sussmilch for the loan of certain slides of the Kuttung igneous rocks. GENERAL DISCUSSION OF CERTAIN FEATURES OF THE ROCKS. Sequence of the lavas and correlation of various sections. The igneous rocks of the Clarencetown-Paterson district comprise a wide variety, embracing representatives of the following types: andesites and andesitic pitchstones, dacites, toscanites, dellenites, quartz-keratophyres, rhyolites and felsites. Associated with these are tuffs and flow-breccias of rhyolitic and kerato- phyric composition. The question of the order of extrusion of the lavas was briefly discussed in an earlier paper (1922) and it was pointed out that a general sequence, only, held throughout the area, but modifications were found in many localities, so that the order obtaining in one district was appreciably different from that occurring in another, and it was clear that only by extremely detailed field and chemical study would one be likely to solve the problems presented by the association of calcie and alkaline rocks. The difficulties involved in examining the facts of the sequence from a petrogenetic point of view are, in great measure, due to the presence of phenomena resulting from the albitization of some or all of the original felspathic content of certain rocks, whereby their magmatic relations with other rocks, and with each other, have become obscure. However, there is present an order of extrusion in which the following are the salient features: rocks of intermediate composition—glassy and lithoidal mica- amphibole-andesites (a@)—are the first flows to appear; these are followed by the pyroxene-andesites (0) which are of greater basicity, and then comes a series of quartz-keratophyres (c), dacites (d@), toscanites and dellenites and rhyolites (e), all more acidic than group (a). In this series there is considerable variation in the order of the types. If, as Browne and Walkom have suggested (1911, p. 294) the basalts (f) of the Permian (some representatives of which occur near Seaham) belong to, and represent the last expression of, the magma which was so active in Kuttung times, then the sequence through Kuttung and Lower Permian time would appear in a general way to represent one which was characterized, to quote Harker, by “a divergence from the initial type.’”’ And as Harker (1909) has shown for certain volcanic districts in Nevada, Mexico, and Germany, there is, in the case of the rocks now being considered, increasing acidity in the order (a), (c), (d), (e€) and increasing basicity in the order (a), (bd), (f). It must not be forgotten, however, that there are lavas in the Burindi Series, and these should really be considered in connection with any inquiry into the petrological evolution of the igneous rocks of the Upper Palaeozoic of New South Wales, but there seems to be evidence of a substantial time break between the Burindi extrusions and those of Kuttung times, and it is quite reasonable (at all events in the Paterson-Dungog district) to look upon the extrusion of the andesites which occur towards the base of the Kuttung, as being the beginning of a subse- i 114 GEOLOGY AND PETROGRAPHY OF CLARENCETOWN-PATERSON DISTRICT, iv, quent, but more or less distinct, period in the larger volcanic cycle, which began in Burindi time. In any case, the Burindi lavas, nearest stratigraphically to the Kuttung flows, are hornblende-andesites, almost identical with those described here. With regard to the correlation of flows over the whole area, the reader is referred to the account of the Regional Geology of the district given in the first paper of this series. There details of the relationship between certain lavas in the western areas and those in the east are given, and it is pointed out that some of the flows are characterized by the fact that they change along the strike. In his presidential address to the Royal Society of New South Wales (1923), Mr. C. A. Sussmilch attempted to correlate certain sections at Martin’s Creek and Helah with two which had been previously described by the writer (see Sussmilch, 1923, Plate I), but certain units were linked together, which are very different as regards both mineralogical and chemical composition, and also stratigraphical position. Thus, in the Glenoak section, a quartz-keratophyre was placed as being equivalent to the Mt. Gilmore type of toscanite, and a rock which, in the Martin’s Creek section, has been referred to as a dacite, but has later been described by the present author as a keratophyre, was shown as the equivalent of certain rhyolites in the eastern areas. In both cases these correlations are incorrect, and indeed it is only by the closest scrutiny of the microscopical details and field evidence that any correlation can be made wisely. On account of the peculiar features of order of succession, which the rocks under review exhibit, the present writer has been able to correlate the dominant units only. Hvidence of the Operation of late-magmatic Processes. Albitization.—Phenomena produced in igneous rocks as a result of the activity of cognate solutions and gases acting during the last stages of, or immediately after, the consolidation of the main masses, have frequently been the subject of investigation by petrographers. One of the most important of such processes is that of albitization, whereby particularly the basic plagioclase of a rock, and at times other constituents too, become replaced by albite. Bailey and Grabham (1909) described in some detail the changes of this nature in certain Upper Palaeozoic igneous rocks of Scotland, citing other instances where albitization appears to have occurred. Many spilitic rocks have evidently been developed as a result of albitization, and Benson (1918) mentioning how Neithammer (1908) has suggested and Bailey (1911) has confirmed that many keratophyres are albitized porphyrites, summarized his own views generally that albitization by late-magmatic solutions is, at least, of quite frequent occurrence. The article by Bailey and Grabham is perhaps the most detailed and con- vincing in showing how the albitization occurs and how its origin may well be connected with the magma which produced the rocks affected. Recently (1923) W.R. Browne drew attention, in a brief paper, to the occurrence of the phenomenon in many of the Upper Palaeozoic igneous rocks of New South Wales, and in the present investigation the writer has encountered numerous examples of albitiza- tion. It is thought here that the change has been one of magmatic origin, produced by solutions connected with the extrusions of the lavas, and after a description of the phenomenon, the reasons for this view will be given. In the rocks examined the replacement is found in non-vitrophyric types. Thus the well known pitchstones occurring towards the base of the Kuttung Series are immune from albitization. The more basic of the lithoidal types are also free from the change. In those lithoidal types affected, both plagioclase and ortho- BY G. D. OSBORNE. 115 clase may suffer replacement, whether they be phenocrysts or in the groundmass, although in the case of the latter it is difficult to make out the extent of the change satisfactorily. In general the albite attacks the phenocrysts by working along the cleavage eracks and around the margins. At times the plagioclase phenocrysts may be undergoing change without reference to any optical directions, and in such a case, had conditions permitted, a complete pseudomorph would have resulted. At other times, however, a particular zone or zones in the plagioclase may be albitized while the rest of the mineral is unaltered; in such cases, the more basic zones suffer. Frequently a little calcite is found associated with the albite, and in almost all cases there are scales of sericitic mica, which may be the result of normal weathering of the albite itself. How far the calcite may represent the residual lime left after replacement is not clear. One characteristic of the secondary albite is its spongy appearance, due possibly to the aggregation of tiny patches of albite which, though probably in optical continuity, are mixed with a certain amount of impurity and possibly with ultramicroscopic areas of residual basic plagioclase. In the change of orthoclase to albite, the ragged outline of the albite working from the margin across the crystal is very characteristic, and often the albite has a dirty yellow colour in ordinary light, which may be due to the presence of some limonitic material. The groundmass of certain rocks, when cryptocrystalline, shows a certain amount of replacement by albite which occurs in short prisms, and when pumiceous the ground may show a kind of selective albitization, the cusp-shaped bodies being partially or wholly replaced. In some cases one can find sufficient unreplaced material to permit the determination of the original composition of the lime-soda felspar, but at times the albitization has, as far as can be seen, been complete, and the consequences of such a circumstance are very important, especially in connection with the question of the relations of the various rocks. That the process of albitization in these rocks has been of magmatic origin, as specified above, seems to be established from the following criteria: The change has been selective, affecting those rocks which are at least as acid as the Martin’s Creek andesite. The lithoidal pyroxene-andesites, which are the most basic in the series, have not been altered in this manner. In regard to the immunity of the glassy andesites from albitization, it is not thought that this point has any significance in connection with the origin of the phenomenon, because, whatever the nature of the solutions involved, it is probable that these glasses offer little means of access for the waters, but in the case of the lithoidal pyroxene- andesites, they, being in direct association with albitized rocks, should likewise show evidence of having been affected, if the solutions were meteoric. Then again, the extent of the phenomenon does not in any way correspond with the extent of atmospheric weathering seen in any of the albitized rock masses, nor is the albitization to be found more complete near cracks and fissures. For example, in very fresh specimens taken from the deepest portions of the quarries in the andesite at Martin’s Creek and the keratophyre on the Williams River below Clarencetown, the albitization is quite well-developed, being in the latter case such as to have replaced completely the original lime-soda felspar. Also, in rocks which are strongly albitized there is often an absence of the effects of weathering in the minerals. 116 GEOLOGY AND PETROGRAPHY OF CLARENCETOWN-PATERSON DISTRICT, iv, Further, there is an almost constant association of chlorite with the albite, and it is difficult to imagine how chloritization of plagioclase, and even of orthoclase, as in one or two instances, would be brought about by meteoric solutions. Thus the process of albitization in these rocks seems to be of magmatic origin, occurring after the consolidation of the lavas, but being the effect of the operation of solutions directly connected with the lava magmas. That is to say, the process is a “deuteric” one as defined by Sederholm (1906) and its origin is indicated by the term “late-magmatic” as used recently by Browne (1925). Holmes’s definition (1920) of Sargent’s term “autometamorphism” (1917) would also include the process of albitization. Chloritization.—The occurrence of chlorite in certain of the Kuttung rocks is such as to place doubt upon its origin being simply connected with weathering. Thus in certain andesites and toscanites, a replacement of plagioclase by chlorite occurs, the latter confining itself to one or more zones, and rarely completely replacing the former. Elsewhere in the groundmass of these rocks, patches and veinlets of chlorite are found undoubtedly indicating selective replacement. In all cases of the presence of chlorite in the manner indicated here, albitization was found to have occurred. Bearing in mind the chemical composition of the chlorite minerals, and having regard to the constant association of albitization and chloritization, it seems reasonable to regard the former as being, like the latter, of late-magmatic origin. Kaolinization.—In various rocks throughout the Kuttung Series, kaolin occurs in notable amounts replacing felspar phenocrysts and also groundmasses which originally have been pumiceous and glassy. In many cases the field evidence is somewhat against the kaolin having been produced by the action of surface waters, although it must be understood that there are numerous cases where it is clear that the kaolin has developed by rock decomposition. Of the former cases, two can be cited. In a quarry near Seaham, specimens taken from about fifteen feet below the ground level show much more kaolinization than those close to the surface, and separating the kaolinized rocks over an appreciable extent is a band higher up, which has a groundmass in which kaolinization is almost absent. In the case of the Williams River keratophyre quarry, one finds a greater proportion of kaolinized felspar as one proceeds downwards, and the kaolinization is sometimes sporadic in occurrence, and found not close to joints. This field evidence, then, seems to suggest that the kaolin has been produced by the action of magmatic solutions, and in support of this one finds that albitization and chloritization are often associated features. That kaolin may be of deuteric origin seems to be demonstrated by certain features exhibited by the Prospect intrusion (see Browne, 1925), and it occurs in some rocks on the South Coast of New South Wales in such a way as again to indicate origin by the action of late-magmatic waters. Devitrification, and the Relation between the Glassy and Lithoidal Andesites. From the petrographical examination of the rocks it is clear that devitri- fication has occurred in a large number of cases. Thus, pumiceous material, originally glassy, has been affected and altered into material with cryptocrystalline texture. One can frequently see unaltered patches of pumiceous matter in a rock which has not undergone complete devitrification. This residual glassy content is different from that occurring as streaks in some rocks, such as the BY G. D. OSBORNE. iL 7/ Martin’s Creek andesite, where it appears that the glass is a fraction that did not have an opportunity to crystallize like the bulk of the rock. Then, in certain instances, it was noticed that some lithoidal rocks showed a sort of perlitic cracking, which almost certainly occurred during the rapid solidification of a glassy rock. The question of the time of devitrification is one of interest and of some difficulty. Judd (1889) described processes which had operated in rocks with vitrophyric groundmasses whereby phenocrysts actually enlarged themselves at the expense of the glass, the process being a kind of devitrification, regarded as secondary, but occurring at no great time after the congealing of the lava. In describing the origin of the pitchstones of Mull, Anderson and Radley (1917) showed by chemical evidence that the lithoidal rocks appeared to be the devitrified products of the pitchstones. These authors distinguished between primary and secondary devitrification, the former occurring just after the consolidation of the rock and the latter being a long process working for ages after solidification. Bonney and Parkinson (1903) described two types of devitrification, one primary and one secondary, but in giving detailed descriptions of each of these, did not explain clearly their time relations, although Bonney elsewhere (1885) has indicated his belief that the primary devitrification occurred during cooling and the secondary type took place long afterward. Harker (1909, p. 226) did not agree with those who postulated primary and secondary devitrification, and inclined to the view that it was a process of long duration. In the case of the Kuttung lavas there is little evidence available to help one to decide when the change occurred, but in view of the fact that certain authors have maintained that so-called primary devitrification occurs immediately following the consolida- tion of a glassy rock, it might be worthy of consideration whether the change has, in the Kuttung lavas at all events, had some relation to the activity of late- magmatic solutions. The possibility of a lava being devitrified would then depend to a great extent on the chemical composition and the physical conditions attending its cooling, and thus it would be reasonable to expect to find certain glassy rocks practically wholly unaffected, in association with lavas evidencing devitrification, an association which occurs in the Kuttung terrains. This discussion of devitrification leads us to enquire into the question of whether the lithoidal varieties of amphibole- and pyroxene-andesites are the devitrified equivalents of the glassy andesites, which are found in close association with them, and which, mineralogically, are in each case similar. Anderson and Radley demonstrated that the felsites of Mull were chemically very similar to the pitchstones, except as regards the content of combined water, and were convinced that the former had arisen from the latter by the escape of some of this water. The chemical evidence in the case of the horn- blende-biotite-andesites now being considered does not support the view that they have been derived from glassy andesites by devitrification. This may be seen from the chemical data given below: ie II. SiO. .. 64.88 O06 66.38 CHO oc 3.00 O10 4.11 Na.O .. 5.41 oN 3.55 K20 2% 2.79 O60 lo WH EEO) oa 1.63 4.56 I. Biotite-hornblende-andesite, Martin’s Creek. Il. Hypersthene-bearing biotite hornblende andesitic glass, Tumbledown Creek, east of Martin’s Creek. Analyst G.D.O. 118 GEOLOGY AND PETROGRAPHY OF CLARENCETOWN-PATERSON DISTRICT, iv, Of course, the wide variation in the case of the values for soda is to some extent due to the fact that albitization occurs in No. I, but the difference in the values of potash is too marked to allow of No. I being the devitrified product of No. II. Chemical examination of the two phases of the pyroxene-andesite has not been made, but the association of very similar rocks at Currabubula has been commented upon by W. R. Browne (1920, pp. 415-6). Concerning the field evidence it is to be pointed out that the two varieties of the biotite-hornblende-andesite were always found to be sharply bounded. Thus in a railway cutting to the east of Wallarobba, there is the following succession: lithoidal andesite 6 feet, glassy andesite 24 feet, lithoidal andesite 12 feet; all the bounding surfaces being sharply defined. In the case of the hypersthene-augite-andesite one was not able to delimit in the field the extent of the two phases, but it appeared as if there were separate flows of the two varieties. Microscopically there does not seem to be much in favour of the lithoidal andesites having been derived from the glassy types. Thus in the case of the pyroxene group, the groundmass of the stony phase is spongy in appearance, and under crossed nicols with high magnification there are no signs of residual glass. Also in the lithoidal phases of the mica-amphibole-andesites there are present streaks, rather than patches of glass, strung out in the direction of flow which, in the writer’s opinion, indicate original heterogeneity of the magma material at the close of the intratelluric crystallization. Mineralogical and Magmatic Relationships. The effect of late-magmatic processes upon many of the rocks has made a consideration of their relationships a matter of difficulty, but in spite of this, one can see to some extent how the various lavas are genetically connected from a mineralogical point of view, while a satisfactory understanding of their magmatic relations is not possible with the data at present available. Mineralogical Relationships.—Although quartz has been noticed in the Martin’s Creek type of andesite, it is doubtful whether it really belongs to the rock, probably being xenocrystic. This mineral makes its appearance in the quartz-keratophyres and dacites, increasing in importance in the toscanites, till in the dellenites and rhyolites it is conspicuous. In some cases it is present in the groundmass of certain acidic rocks. Plagioclase, where unaltered, shows a gradation from labradorite in the pyroxene-andesites to andesine in the mica-amphibole-andesites and toscanites, becoming acid andesine and oligoclase in the dellenites. It is not clear whether any primary albite occurs in the soda rhyolites. Orthoclase is very subordinate in the dacites, but becomes an important constituent in the toscanites and dellenites, and its further increase is seen in the sodipotassic and potassic rhyolites. Of the ferro-magnesian minerals, biotite is the most widespread, being found in all groups of rocks except the pyroxene-andesites, and hornblende is also fairly widespread, being present in the Martin’s Creek type of andesite, the Williams River keratophyre, in some of the dacites, and in both the Mt. Gilmore and Paterson types of toscanite and andesite, as well as in certain rhyolites. Augite and hypersthene have their best development in the pyroxene-andesites, but both occur along with biotite and hornblende in a glassy andesite from BY G. D. OSBORNE. 119 Portion 99, Parish of Barford; this rock being a connecting link between the two main groups of andesite. The iron ores in the rocks comprise magnetite and ilmenite. Often both are present, whereas one or other may occur alone. In this connection ilmenite is especially characteristic of the dacites. Apatite is of almost universal occurrence, but it is of greater abundance in those rocks which contain considerable ilmenite. Zircon is of quite frequent appearance in the hornblende-andesites, but only occasionally met with in some of the other types. Chemical Relationships.—With the data to hand at present, and in view of the presence of albitization, one cannot achieve much in the way of examining the magmatic relations of the rocks under discussion. i The salient features of the chemical evidence are as follow: The rocks which have been analysed show fairly high silica percentages, even in the case of the rocks with andesitic characters. Thus the andesites contain from 64.88% to 66.38% of silica, the quartz-keratophyres and dacites about 67%, and the dellenites, which are closely allied to sodipotassic rhyolites, have as much as 72% or 73%. Concerning the alkalis, potash has been found as an important constituent in all rocks analysed. Indeed it would appear that, were it not for the presence of albitization, it is probable that the Kuttung rocks in this district would show a general possession of a sodipotassic or in some cases a “dopotassic” character. Magnesia is low in most rocks, although it would probably be found .to be fairly important in an analysis of the pyroxene-andesite. A fair amount of titanium and also appreciable amounts of manganese are always present, while phosphorus varies considerably, sometimes being only present as a trace. What was the composition of the original magma, from which the Kuttung lavas were derived, is a question which it is difficult to answer. Chiefly from field evidence, strengthened by a consideration of the chemical and microscopical data, one must say that the suggestion put forward by Dr. Browne that the horn- blende-andesites may represent the composition of the original magma which later became differentiated into divergent types, does seem to be likely to prove correct. Spherulitic and Allied Structures. In the groundmasses of some rocks examples of spherulitic crystallization are found. The spherulites are often very minute and in general have never been found to attain any great size as is the case in certain volcanic rocks in various parts of the world. Sometimes there are present structures of the same nature as the spherulites, but these are not always circular in cross-section, representing rather growth from a series of points linearly disposed. Among the most frequent of such structures are some which may be called axiolites, being very similar in structure to those figured by Iddings (1910, p. 240). Often the spherulites are aggregated together so as to form an integral part of the rock, but at other times these aggregations may be present as inclusions. Examples of the latter cases are to be found in a felsite occurring to the east of the Railway Line about one mile south of Martin’s Creek. Specimen 26* from this locality shows an abundance of inclusions, many being entirely composed of aggregations of spherulites and axiolites, the spherulites averaging 4 mm. in radius. They are * Numbers refer to specimens in the writer’s collection. 120 GEOLOGY AND PETROGRAPHY OF CLARENCETOWN-PATERSON DISTRICT, iv, set in a groundmass which is somewhat kaolinized and haematitized. Under the microscope, the material composing the spherulitic structures is, for the most part, felspar which has a refractive index less than that of Canada balsam, but there is also some silica in association with the felspar. With regard to the time of development of these structures, it is to be pointed out that in some of the rhyolites the spherulites are found across the flow lines. This may mean either that the spherulitic crystallization occurred after consolidation of the material showing the flow structure, or, at all events, after this material became too viscous to move. Although the latter seems the more probable, Harker mentions (1909, p. 275) cases of the former. Judd also (1889), described certain pseudo-spherulitic structures, which he considered developed after the solidification of the rock, while Parkinson (1903), in discussing Iddings’s work on the Yellowstone Area (1885-86 and 1899), contends that there may be more than one period of spherulitic crystailization in a glassy rock, some spherulites developing before the perlitic cracking which is often characteristic of obsidians, and others appearing after the consolidation of the glass when its perlitic cracking was complete. Autobrecciation. A feature of many of the Kuttung lavas of acid alkaline composition is the presence, in the groundmass, of brecciated structures which have originated during the solidification of the mass. This type of structure has been termed auto- brecciation. The effects of its operation on a fairly large scale are to be seen in the features exhibited by certain flow-breccias in the Langlands section which have been described in a former paper (1922, p. 173). The autobrecciation has been observed only in the acid lavas and especially those which are sodic. It is clearly related to the high viscosity of such magmas, and has been commented upon by a number of authors. Woolridge has described the petrography of an interesting set of rocks from Llanwrytd, Wales (Stamp and Woolridge, 1923), which are dominantly sodic, and he has emphasized the prevalence of pyroclastic material and the existence of autobrecciation. These rocks are of terrestrial origin and it would seem that subaerial conditions should be favourable to the production of brecciated structures in acid sodic lavas, by the breaking up of a magma crust and the subsequent corrosion of the fragments in the residual molten material. But J. F. N. Green (1919) argues that autobrecciation is a feature of submarine flows of certain chemical composition. Associated with the Kuttung lavas showing evidence of autobrecciation is the occurrence of flows in which there is an abundance of vesicular glassy material, often now devitrified. PETROGRAPHICAL DESCRIPTIONS OF THE KUTTUNG SERIES. Nomenclature of the Rocks. Before proceeding with the petrographical details, it is pertinent to say something about the nomenclature of the rocks. As has been pointed out above, the process of albitization, whereby felspars and other material become replaced by albite of late-magmatic origin, has affected many of the rocks and in some cases obscured their relationships. In the cases where the albitization has been partial, the original identity of the rocks is often determinable, but when the change has been complete, with reference to the primary felspars, there is BY G. D. OSBORNE. AL generally no clue to the original nature of the rocks and they have to be referred to as soda rhyolites or keratophyres. It has been decided, therefore, in the case of the altered rocks to name the rock according to its original identity where possible, and to mention the possibility of a rock having had its felspathic constituents completely albitized, where such is suspected, the rock in such case being named according to its present constitution. Attention has been drawn by W. R. Browne to the difficulties arising out of the albitization of igneous rocks (Browne, 1923). Then, concerning the acid and sub-acid lavas, one has had to make use of the terms dellenite and toscanite for the following reasons: There are a large number of massive igneous rocks in the Kuttung series which possess con- siderable variety and which might all come under the two broad categories, rhyolite and dacite. To apply either of such terms to numerous rocks, however, would be impracticable in connection with the present petrographical discussion, Since fine distinctions between important flows would not be recognized, and con- fusion would result. The use of the terms toscanite and dellenite (both very much the same as quartz-trachy-andesite, but each much shorter and more service- able) permits the recognition of very slight gradations in chemical composition: thus the term toscanite has been applied to rocks where there is a fair amount of orthoclase, as well as considerable plagioclase which is basic andesine or acid labradorite, and the term dellenite applied to types where there is orthoclase, and plagioclase of the composition oligoclase or very acid andesine, quartz in both instances being dominant but not very excessive. The increase of quartz and the absence of orthoclase, always accompanied by an acidification of the plagioclase, produces a soda rhyolite, and the greater abundance of orthoclase places the rocks in the category of potash rhyolite. In connection with those rocks dominantly sodic, the name keratophyre has been applied to rocks which show much phenocrystic albite with stout prismatic habit, and a sort of coarse orthophyric fabric, or an approach thereto. Soda rhyolite has been reserved for acidic albitic rocks which have extensive cryptocrystalline groundmasses. In the following descriptions the dominant types in each section will be considered in detail, the less important representatives being treated only with reference to their variation from the main types. The distribution of the various types described below has already been indicated in the former paper and will not be repeated here. Andesites. The andesites comprise two fairly distinct groups, characterized by biotite and hornblende on the one hand, and augite and hypersthene on the other. There are also some rocks which act as connecting links between the two series, in that they contain both amphibole and pyroxene. In each of the two main groups there are glassy and lithoidal varieties. Mica-amphibole Group (Martin's Creek Type). Glassy Varieties. Megascopical characters: These rocks are black in colour, with a resinous lustre, possessing phenocrysts of felspar and biotite and a number of yellow streaks. : Microscopical characters: Specimen 80, from the Railway Cutting three-quarters of a mile east of Wallarobba Station, shows the presence of hornblende, plagio- clase and biotite and very subsidiary pyroxene, apparently hypersthene, also 122 GEOLOGY AND PETROGRAPHY OF CLARENCETOWN-PATERSON DISTRICT, Iv, apatite and magnetite set in a groundmass which is almost entirely glassy. The felspar individuals are fresh but cracked noticeably, and possess a tabular habit and a composition close to Ab,An,. The hornblende is quite fresh and shows the (110) cleavage rather well, while biotite, although fairly well-formed, evidences bleaching and some alteration. One or two sections of quartz may be of xenocrystic origin. The vitrophyric groundmass shows fluidal fabric and is noticeably cracked as a result of pressure following solidification. There are present a number of microlites of felspar and tiny magnetite crystals as well as small patches, which under high magnification appear to be greenish decomposition products. Lithoidal Varieties. Megascopical characters: These rocks vary in colour from light blue and grey to dark blue. Phenocrysts of hornblende and plagioclase, the latter often showing a black core, occur in varying amount in a dense dull-lustred groundmass which possesses some reddish blotches. The rocks fracture in various ways, according to the nature of the texture and mineralogical arrangement. Microscopical characters: Specimen 1, from the quarry formerly owned by the State Metal Quarries, Ltd., shows the following primary minerals present: plagio- clase, hornblende, biotite, magnetite, ilmenite (?), apatite and zircon. These are set in a groundmass which makes up two-thirds of the rocks. The plagioclase has the composition Ab,An, (andesine) and possesses a tabular prismatic habit. It is occasionally zoned and the grainsize is variable and up to 2 mm. in diameter. Albite, carlsbad and pericline twinning are all present, the last of these being rare. Albitization and chloritization (see Plate xxiii, No. 2) of the andesine have occurred and scales of sericite have developed by decom- position, but these features are not present to the extent of obliterating any one grain. The albite has used the imperfect cleavage cracks to facilitate its attack upon the more basic material, and in some cases has replaced the central portions of the andesine first of all. The chlorite is found pseudo- morphing the core of the plagioclase in most cases, while at other times it has selected some zone or zones which are completely replaced. Hornblende occurs subidiomorphically and shows a fair amount of resorption. The cleavage parallel to (110) is well developed and the pleochroism is strong. Biotite, which is present in slender pieces, is very much altered by chloritization and has been resorbed with the production of magnetite. Primary magnetite certainly occurs and there seems to be some ilmenite also. The apatite and zircon call for no special remarks, except to point out that the latter is subordinate to the former. Quartz occasionally occurs in this andesite, but under the microscope appears to be of a xenocrystic nature. The field evidence, however, shows that towards the base of the andesites at Martin’s Creek the number of quartz individuals increases. The groundmass shows lines of flow very well and consists of streaks of brownish glass and patches of cryptocrystalline material. It would appear that this glassy matter is the expression of original heterogeneity of the magma, rather than that partial devitrification has occurred. The red patches referred to above are due to staining by haematite derived from nuclei of iron oxides. Specimen 2 is somewhat lighter in colour and shows more phenocrystic horn- blende than does No. i. In thin section it is similar to No. 1, but there are some completely albitized phenocrysts. BY G. D. OSBORNE. 123 Specimen 3 differs from No. 7 in that the ratio of groundmass to phenocrysts is smaller, and there is a coarser grainsize in the microfelsitic portions of the groundmass. Specimen 4 is of a dark blue colour and possesses a very irregular fracture. The plagioclase individuals are ill-formed, and there is an almost entire absence of hornblende. Microscopically, it is seen to differ widely from the other varieties. Glassy material is almost absent from the groundmass and the spongy felspathic content is of a coarser texture than has been the case hitherto. The phenocrysts of andesine are somewhat replaced by albite and chlorite and are ragged in outline. Apatite and zircon are quite distinct. It may be mentioned here that in his account of the Martin’s Creek andesite, M. Aurousseau (1915) recorded the presence of hypersthene. The present writer has not observed this mineral in any of the many slides of the andesite which he has examined. Pyrozene Group. Glassy Varieties. Megascopical characters: These, in hand specimen, are black rocks possessing a subdued resinous lustre and a fair amount of yellowish-coloured plagioclase as phenocrysts. A little ferro-magnesian material is also discernible. Microscopical characters: Phenocrysts of plagioclase, hypersthene, augite, magnetite and apatite are set in a brownish-grey glass. The plagioclase is tabular in habit and shows twinning according to the albite, carlsbad and pericline schemes. There is a little zoning present, and many inclusions are seen. Some of the inclusions are arranged zonally and may consist of portions of the ground- mass, yet the edges of the felspar are not corroded to any extent. The composition of the mineral is that of basic labradorite. The (001) cleavage is only poorly developed in some individuals. Very noticeable is the absence of albitization and chloritization. Hypersthene is more abundant than augite and shows fairly distinct pleochroism, a fairly good cleavage and the presence of numerous cracks. Magnetite is generally found as inclusions in the hypersthene. Felspar inclusions are rare and apatite only occurs occasionally. The hypersthene is only slightly altered and then to a brownish-green silicate, possibly anthophyllite. The augite is subidiomorphic, up to 13 mm. in diameter, and possesses twinning, being of a brownish-grey colour. The groundmass is almost entirely glassy, but under the high magnification is seen to include a number of tiny patches of a greenish substance, which is probably derivative from a ferro-magnesian mineral, and some tiny felspar crystals, the remainder being brown glass. In one slide there was present what appeared to be an inclusion of a rock, essentially similar in composition to the host, that is to say, composed of augite, hypersthene and plagioclase, possess- ing an even grainsize and showing no decomposition. This probably represents a cognate xenolith which has been torn by the magma from some mass which erystallized under intratelluric conditions. Lithoidal Varieties. Megascopical characters: These varieties are dark blue to black in colour, the felspar phenocrysts being very readily seen and the pyroxene only with difficulty. The groundmass possesses a dull lustre. : Microscopical characters: In thin section one sees the same phenocrystic constituents as in the glassy varieties. The felspars are tabular in habit and 124 GEOLOGY AND PETROGRAPHY OF CLARENCETOWN-PATERSON DISTRICT, iv, variable in grainsize, the maximum being about 3 mm. in diameter. Zoning, and twinning according to the albite and carlsbad schemes, are present and the com- position of the plagioclase is Ab,An,, labradorite. Hypersthene is noticeably changed, the alteration developing around the margin of the grains in the first place and then along the cracks. The product seems to be something related to uralitic hornblende. Augite is not very conspicuous and is quite fresh. Apatite is present in acicular crystals. The groundmass is distinctive in its appearance and consists of a spongy mass that extinguishes in patches on rotation of the microscope stage. There is not the felsitic appearance which characterizes so many of the groundmasses of the other Kuttung rocks. Examination of this groundmass under the high power shows the presence of tiny green patches which clearly are pseudomorphs, together with abundant iron ore individuals all set in a felspathic mass which shows undulose extinction. There is no glass present, and in a general sense the groundmass is not heterogeneous. Amongst the pyroxene-andesites there is very little variation within the two groups and foregoing descriptions hold for almost all the occurrences in the area. Transitional Types of Andesites. Intermediate in character between the amphibole- and pyroxene-andesites come certain rocks of limited distribution; Specimen 327, from the northern side of a ridge.in Portion 99, Parish of Barford, is typical of these rocks. Megascopical characters: This is a greenish-black rock with a greasy rather than a pitchy lustre. Phenocrysts of hexagonal biotite and another ferro-mag- nesian mineral are clearly seen, also some tiny laths of felspar. Microscopical characters: In thin section one sees phenocrysts of plagioclase, biotite, hornblende, augite, hypersthene, and quartz set in a glassy groundmass, which contains a few felsitic inclusions. The rock shows a certain amount of cracking and there is also marked flow structure. Some of the phenocrysts are shattered as a result of pressure acting after the rock consolidated. The plagioclase, which is acid labradorite, shows a tabular habit, and albite and carlsbad twinning. Hypersthene is not abundant and is non-pleochroic, but shows alteration to a greenish substance. The augite is present in small equidimensional crystals and is fairly fresh, being of a grey colour. Hornblende is a fairly important constituent and shows the prismatic cleavage fairly well developed. Pleochroism is noticeable and decomposition is practically absent. The biotite is both bent and bleached at times, but it has not been resorbed. The quartz appears to be xenocrystic in origin, in view of the fact that it generally shows an association with a certain amount of cryptocrystalline material, which is quite similar to the groundmass of the included fragments, but unlike the glassy groundmass of the main rock. It is to be noted, also, that in one slide of these rocks quartz was almost absent and so also were the rock inclusions. The groundmass of 327 is brownish in colour and almost completely glassy. Chemical composition: A chemical analysis of Specimen No. i, a lithoidal andesite from Martin’s Creek and also a partial analysis of Specimen 327, a transitional type of glassy andesite, were made. These are tabulated below together with an analysis of another specimen of andesite from Martin’s Creek, quoted by Mr. Sussmilch (1923, p. 24). BY G. D. OSBORNE. 125 It will be seen immediately that in analysis No. I there is evidence of the rock having been albitized. The composition of the plagioclase calculated from the chemical data comes out at about oligoclase of the formula Ab,An,, while the unaltered felspar seen in parts of the rock under the microscope was andesine of the composition Ab,An,. In the case of analysis No. III there is evidence that the albitization has been less extensive. The figures for the rock from Tumble- down are of interest, showing the high content of combined water, and values for lime and the alkalis which are more normal for an andesitic rock. These circumstances are due to the fact that late magmatic processes have not affected the rock. From a determination made upon the alumina of this pitch- stone it was quite clear that there was sufficient present to allow all the lime to be calculated as anorthite, and with this information the plagioclase felspar in the rock has been found to be of the composition Ab,An,, this being about the same as that of the unaltered plagioclase in the lithoidal andesite (Specimen 1). I. II. Iil. Si@sm cree bts. tes 64.88 66.38 64.20 A1l;03 16.18 16.88 FesO3 1.52 1.90 FeO 2.43 2.52 MgO iL, mall 66 CaO 3.00 4.11 3.14 Na2,O 5.41 3.55 4.41 KO 2.79 1.72 3.52 H.O+ 1.63 4.56 i 78) H,O0- -50 1.04 .31 TiO. 89 65 P.O; trace 13 COs, trace 03 MnoO 05 —_ 100.49 100.14 Specific gravity of I= 2.642 at 16.5° C. Magmatic name for I = Lassenose. I. Andesite, lithoidal variety, Specimen 1, Martin’s Creek, N.S.W. Analyst, G. D. Osborne. II. Andesite, glassy variety, Specimen 327. Tumbledown Creek, near Martin’s Creek, N.S.W. Analyst, G. D. Osborne. Ill. Andesite, Martin’s Creek. Analyst, W. G. Stone. Dacites. Megascopical characters: The dacites are in general of a light colour, chiefly pink and cream, although darker varieties occur in the neighbourhood of Glenoak. The fracture of the rocks is generally very even, but becomes quite irregular in those types which contain a number of inclusions and spherulitic structures. The 126 GEOLOGY AND PETROGRAPHY OF CLARENCETOWN-PATERSON DISTRICT, iv, minerals visible in hand specimen comprise plagioclase, quartz, biotite, and sometimes orthoclase. At times the former two are very conspicuous and at other times they, as is generally the case with the latter two, are quite inconspicuous. The lustre of the rocks is dull, the appearance of the groundmass quite felsitic. The inclusion-bearing types are, like the massive rocks, of fairly wide distribution, but a notable occurrence of the former is near the foot of Mt. Johnstone, about 12 miles from Paterson. Here one finds a great variety in the appearance of the dacites, although the parent rock is fairly constant in composition. Microscopical characters: The minerals discernible under the microscope com- prise quartz, plagioclase, biotite, orthoclase, ilmenite, magnetite, hornblende, zircon, and apatite. The plagioclase varies considerably in grainsize, being up to 14 mm. in diameter. The twinning is chiefly according to the albite law, but sometimes after the carlsbad law as well. The habit is generally tabular, but at times almost prismatic. The chemical composition of the felspar varies from basic oligoclase to very acid labradorite. Decomposition to kaolin has progressed to a fair degree in many cases, and fresh felspars are rarely to be seen. Some replacement by albite, almost certainly of deuteric origin, is common and may have occurred to the extent of half the original crystal being replaced. Quartz is generally cracked and corroded, but always shows sharp extinction. It varies considerably in grainsize and when in the groundmass is very small indeed. Orthoclase, of course, is only of infrequent occurrence and then in a very subordinate amount. There is generally a good (001) cleavage present, and almost complete kaolinization is always to be seen. Biotite is fairly common in one or two slides, but sparingly present in the majority. Weak pleochroism and evidences of having been strained are generally seen, while resorption, with the production of magnetite, is only a rare feature. Hornblende is not a constant constituent, but is always fresh and ragged in outline. The ilmenite is quite a common mineral in the dacites, while magnetite only occurs rarely. The former shows typical alteration to leucoxene, and haematite is invariably found around the margins of the magnetite units. In two of the rocks tiny zircons have been found, and a very little apatite occurs in the majority of the dacites. The groundmasses of the rocks vary considerably. Sometimes the texture is quite even and entirely cryptocrystalline, consisting most probably of felspar and quartz. At other times there is a variation in the nature of the groundmass from place to place. Thus there may be spherulitic patches and ecryptocrystalline areas as well as some residual glass. No entirely glassy groundmasses have been found, although some rocks possessed a considerable amount of vitreous material. Devitrification has occurred in some specimens, especially those con- taining a pumiceous groundmass. In these cases the cuspate bodies left after the collapse of pumice bubbles have been changed to felspar and then kaolinized. Hxamination of Specimen 5 and certain slides kindly lent by Mr. Sussmilch, all from the little quarry at the foot of Mt. Johnstone, shows that one can at times obtain dacites free from enclosures and at other times it is difficult to do this. The rocks all have a typically felsitic appearance and show con- siderable amounts of phenocrystic biotite and quartz as well as plagioclase. The plagioclase varies from andesine, Ab,,An,, to labradorite, Ab.An,, and shows a tabular prismatic habit. There is a little albitization in the case of one rock. The biotite is bent and considerably altered while the quartz shows corrosion and inclusions of the groundmass. The groundmass is always pumiceous and in BY G. D. OSBORNE. 127 some rocks extensive silicification has occurred while devitrification is evidenced in certain cases. Haematite, which is the result of infiltration, is often present, causing a pink colour to be imparted to the rocks. In all cases the rocks are dacites, except in one case where there is a tendency to become a rhyolite. The rock inclusions in the dacites of this locality comprise obsidian, rhyolite and soda felsite. These are not cognate with the parent rock. In the groundmass - of some of the dacites, however, there are present spherulites, which occur in separate little masses surrounded by a border of finely divided silica. These composite inclusions do appear to be more of the nature of cognate inclusions. Specimens 171 and 495 are from Mt. Gilmore and show a large number of inclusions. In these rocks the felspar becomes as acid as basic oligoclase. Specimen 415 from the southern part of the Glenoak section is distinctive on account of its blotchy appearance. This, under the microscope, is seen to be due to the prevalence of haematite in certain parts of the groundmass. This rock ‘also evidences a little albitization of the plagioclase. Chemical composition: A partial analysis, by the writer, of a representative of the dacitic rocks, described above, and occurring at the little quarry near the foot of Mt. Johnstone is as follows: SiO, O60 0 67.76 AlsOz a6 60 12.89 CaO 26 5:6 1.16 Naz,O O.6 50 2.32 K.0 ae 2-0 1.88 The composition of the plagioclase, calculated from the above data is andesine of the formula Ab,,An,,. This rock is low in alkalis and the specimen was free from albitization. The silica percentage is of the same order as the values found in the andesites given above, and a complete analysis would almost certainly find its place in the same class and order of the quantitative classification as that in which the andesites are placed. Quartz-Keratophyres. Albitization has been detected in the rocks described under this heading, and it is probable that all the albite which causes the rocks in their present condition to be called keratophyres, came along during the late-magmatic period. The most important examples of these occur right throughout the area, and constitute the Williams River type. The best exposure of this type is in a quarry on the left bank of the Williams River below Clarencetown. Megascopical characters: The quartz-keratophyres are very constant in their appearance. When fresh the colour is a dark blue, but fresh specimens are rarely obtained and the prevalent colours of the somewhat weathered outcrops are fawn and light green. Flow structure is visible in the arrangement of the pheno- erysts which include quartz, biotite and plagioclase. The abundance of biotite often imparts to the rocks a spangled appearance. The groundmass is dense and dull lustred. Microscopical characters: The rocks show a uniformity in microscopical features, there being in the case of some flows a slight change to a finer texture in the upper portions. Specimens 131 and 262 from the Mt. Gilmore district are of the Williams River type, and may be taken as typical of the rocks. 128 GEOLOGY AND PETROGRAPHY OF CLARENCETOWN-PATERSON DISTRICT, iv, The phenocrysts make up 60% of the rock and are themselves present in the following proportions: felspar 60%, quartz 20%, biotite 13%, hornblende 4%, iron ore 2%, apatite less than 1%. In No. 131 the felspar is all albite and this slide was used in 1922 when the rock was named a biotite-quartz-keratophyre. A slide of No. 262, since examined, shows the presence of some residual more basic felspar in the central portions of some of the albite phenocrysts, so that it is very likely that the whole of the albite is of late-magmatic origin. The habit of the felspar is stout prismatic and the lamellar twinning is imperfectly developed. The (001) cleavage which has never been well developed, is obscured by the later albitization. Along some of the cracks sericitic mica has developed as a result of weathering. Associated with the albite is chlorite, which occurs as patches in the felspar and in the groundmass. The biotite is of a brown variety and shows streaky bleaching. Many pieces have been bent, and along the cleavage cracks are found veinlets of haematite. The quartz is present in equidimensional sections and shows many pseudo- inclusions of the groundmass. Hornblende is not a very prominent constituent, although its presence is significant. It is of a pale green colour and shows pleochroism of medium intensity. The prismatic cleavage is poorly developed, as also is the twinning which has its plane of composition parallel to (100). i itt Si@s-im. cae tas ane 67.06 67.71 DATS Oatuer okies) lets 15.95 15.24 Tae O econ) sore) pho 1678 1.48 NC Og sersic cree ace 2.3 1.89 MgO 1.87 .46 CaO 1.98 3.00 Na.O 4.62 5.87 K,0 2.01 il, Sil H.O+ -65 1.89 EGO en) dee ob 44 39 PSOE 43.20 Mee Gis 28 trace TOs eet.e eee -40 247 COSsr can weer tet absent absent SO3 oh Me Oe tor absent —— Cl sed fegsk pada trace — SEO! cht ie cee absent — IMT) eee eR, .39 — 99.80 100.21 Specific gravity of I = 2.62 at 16° C. I. Quartz keratophyre, Williams River, near Clarencetown, N.S.W. Analyst, G. D. Osborne. II. Quartz keratophyre, Church Hill, Currabubula, N.S.W. Analyst, H. Yates. BY G. D. OSBORNE. 129 The iron ore is mostly decomposed and appears originally to have been of both varieties, magnetite and ilmenite. Apatite with acicular habit and a stray section of zircon are present. The groundmass contains a fair amount of glassy material, but for the most part is cryptocrystalline. The arrangement of the felspar phenocrysts, especially on account of their stout prismatic habit, is such, when the groundmass is very subordinate, as to give rise to an approach to a coarse orthophyric fabric. Similar rocks to those just described occur higher up in the stratigraphical succession, examples of such being Specimen 263 in the Mt. Gilmore section, and Nos. 367 and 402 occurring near Glenoak. Chemical composition: Chemical analysis has been made of Specimen 131 of the quartz-keratophyre from the Williams River Quarry and is tabulated above, and with it is placed an analysis of a quartz-keratophyre from Currabubula. These two rocks have different textures and different modes of occurrence in the field, and it is premature to enquire into the question of whether they can be correlated one with the other, but the chemical data give evidence of a certain amount of general similarity, in that the rocks are both sodic and of about the same acidity. Toscanites and Dellenites. The rocks coming under this heading are fairly acidic in nature, but they differ entirely in texture and to some extent in the nature of the phenocrysts, from many of the rhyolites. It has been found that a certain flow will in one place be a toscanite and in another place a dellenite due to a gradual variation; but cases of a unit maintaining its toscanitic or dellenitic nature throughout the whole of its occurrence have been seen. The two most important sets of rocks in this group are those known respec- tively as the Paterson and Mt. Gilmore types of toscanite and dellenite. Micro- scopically the two types are somewhat similar, but there are certain differences and it will be well to describe each type separately. Mt. Gilmore Type. Megascopical characters: The general body colour of the rock is a brown, the two varieties of felspar, plagioclase and orthoclase, being quite conspicuous, as also is quartz, while subordinate ferromagnesian individuals occur at times, all set in a dense lithoidal groundmass. The texture is so coarse at times and the porphyritic constituents so dominant that one might easily mistake the rocks for specimens of granite-porphyry. Microscopical characters: The minerals present in Specimen 261 from Mt. Gilmore comprise quartz, orthoclase, plagioclase, biotite, magnetite and apatite. The plagioclase is of the composition Ab,An,. The habit is tabular and the grainsize goes up to a maximum of .6 mm. in diameter. Decomposition is slight, with the development of kaolin, but albitization seems to be absent. The orthoclase is strongly kaolinized and the (001) cleavage is hard to discern. The individuals of quartz are large and equidimensional, showing a lot of corrosion and cracking. Biotite is almost completely resorbed with the production of strings of magnetite. Where unaltered, the mineral is of a pale green colour and feebly pleochroic. The magnetite is allotriomorphic and unaltered. Tiny crystals of apatite are seen as inclusions in the felspar and in the groundmass. The examination of the latter is difficult because kaolin and haematite have effected quite a lot of replacement. There is undoubtedly a little glassy residuum J 130 GEOLOGY AND PETROGRAPHY OF CLARENCETOWN-PATERSON DISTRICT, iV, present, and the main part of the groundmass is cryptocrystalline. Spherulitic structures are absent. This rock is a biotite-dellenite. Elsewhere on Mt. Gilmore the unit becomes a toscanite on account of the plagioclase becoming Ab,,An, in composition and the orthoclase less abundant. Specimen 391 from the Glenoak extension of the Mt. Gilmore type is a dellenite, in which the quartz is of smaller grainsize than is usual in these rocks, and possesses little in the way of corrosion effects. At Martin’s Creek the horizon of the Mt. Gilmore flow is occupied by a toscanite which varies somewhat abruptly along the strike, becoming a dacite. The toscanite consists of basic andesine dominating over orthoclase and quartz, together with biotite and magnetite, all set in a cryptocrystalline groundmass, which has a little glassy base and shows kaolinization. Paterson Type. Megascopical characters: The rocks coming under this heading form a very distinctive unit which occurs throughout the whole of the area in a constant stratigraphical position. Since writing in 1922 the writer has not seen much to decide definitely the question of whether the occurrence of these rocks is as a composite sill or as a series of flows, but the microscopical evidence seems to strengthen the tentative view taken then that the latter case was the more likely. Thus, in some rocks, there appears to be a certain amount of tuffaceous material and some pumiceous structures. In hand specimen the rocks are of a general grey colour where exposed to the atmosphere, but, when broken open, the colour of the comparatively fresh rock varies from brown to grey and blue. Phenocrysts of quartz, orthoclase, plagio- clase and biotite are visible as in Specimens 69 and 71 from Hungry Hill, Paterson. At other times felspar is well shown, with quartz quite inconspicuous. The groundmass is always of felsitic appearance. Microscopical characters: The minerals present comprise plagioclase, quartz, orthoclase, biotite, hornblende, magnetite and apatite. The quartz is generally present in large equidimensional crystals showing corrosion and inclusions of the groundmass. Orthoclase is sometimes fairly plentiful, at other times it becomes quite subordinate and the rock tends to become a dacite. The orthoclase is kaolinized and shows incipient albitization. Plagioclase exhibits a greater susceptibility to be replaced by albite, and along with the albitization is found the presence of chlorite replacing certain portions of the lime-soda felspar. The composition of the plagioclase varies from basic oligoclase to basic andesine. Biotite occurs in strips showing resorption, bending, bleaching and partial altera- tion to chlorite. The hornblende is quite fresh, although not at all abundant, and it is not of universal occurrence throughout the specimens. Its presence, however, is significant. Magnetite shows peripheral alteration to haematite, while apatite, which is only occasionally seen, is of acicular habit. The groundmass is very dense and of a dark brown colour. The grainsize is cryptocrystalline in nature for the most part, although there are some patches of a glassy base. In two rocks from the south-west of Mt. Johnstone there are some pumiceous areas in the groundmass which have been devitrified, and this evidence points to these rocks having been extrusive. Specimen 6, from the top of Mt. Johnstone, shows the presence of a little green hornblende as well as the other constant constituents. The plagioclase is of the composition Ab,An, and the rock is a dellenite. BY G. D. OSBORNE. 131 Specimen 7, from the railway cutting near Paterson Station, shows less phenocrystic quartz than is usual, and there is more albitization. The felspar is acid andesine and the rock a dellenite. Specimen 70, from Hungry Hill, Paterson, is sandwiched in between two toscanites, Nos. 69 and 71. It is more felspathic in appearance than the type rocks, and the rock is almost an andesine-dacite allied to toscanite. I. II. II. IV. Vv. SiOz 73.04 72.98 74.60 72.79 68.36 Al,O3 13.86 12.58 13.41 | 28.02 13.24 Fe.03 1.60 28 1.28 1.69 1.29 FeO 45 1.66 30 n.d 3.39 MgO -48 49 26 28 1.15 CaO 1.44 2.81 1.08 1.24 2.51 Na2,O 8.40 4.80 3.38 39 2.05 K,0 4.39 2.99 4.50 4.38 5.34 H2O+ 008) 79 85 2.41 2.63 H.O- 225 43 — = — COz absent absent = _— = TiO; 22 .40 .16 = = ZrOz absent — 2 fA as P2Os .04 .08 .03 == uss V205 absent — a wa ma SO3 absent = ass a) ise Cl trace —- as —_ = S (FeS2) absent — a = tik F -- = = = ues Ni, Co. O absent = — = = CuO absent — == au’ Med MnO .07 .03 .06 — 27 BaO 04 — ell IL — = SrO absent — absent — aa *Li,O present — trace —— — TOTAL 100.07 100.32 100.02 99.95 100.23 eee 2.642 2.591 ain 2.416 ea I. Dellenite, Mt. Gilmore, N.S.W. Analyst, W. A. Greig. II. Dellenite, Paterson, N.S.W. Analyst, G. D. Osborne. III. Rhyolite, California. Analyst, H. F. Hillebrand (Washington, 1917, p. 184). IV. Obsidian, Hungary. Analyst, A. Logorio (Washington, 1917, p. 203). V. Dellenite, Dellen, Sweden. Analyst, H. Santesson (Washington, 1917, p. 165). * Spectroscopic. 132 GEOLOGY AND PETROGRAPHY OF CLARENCETOWN-PATERSON DISTRICT, iv, Specimen 489, from the Porphyry Hstate, Seaham, is a handsome blue rock with phenocrysts of quartz, orthoclase and plagioclase, and under the microscope is seen to be a toscanite. Specimen 493, from just south of Felspar Creek, is of interest in containing a little hornblende, and a slightly pumiceous groundmass. These features give it a great similarity to the rock from near Mt. Johnstone. It is necessary to mention here that, while throughout the area one finds these toscanite and dellenite lavas comprising a group which is isolated from the main Volcanic Stage, still, just near Seaham, in two places, at Felspar Creek and Porphyry Point, respectively, small flows of rhyolite are to be found associated with the other rocks. In the Volcanic Stage, apart from the Mt. Gilmore toscanite-dellenite suite, there are two examples of tuffaceous dellenites in which there are some fragments of felsite and soda rhyolite. Autobrecciation, induced during the solidification of the rocks, is a rare feature, occurring in one instance only. Chemical composition: An analysis of the Mt. Gilmore type of dellenite, which was kindly made for the author by Mr. W. A. Greig, is given above, together with one of the Paterson type of dellenite, Specimen 7, described above, which was carried out by the writer. In addition three analyses of similar rocks are tabulated. As pointed out in the brief description above, Specimen 7 shows albitization, the unaltered plagioclase being acid andesine. Calculated from the analysis, the lime-soda felspar is Ab,An,, so that the chemical evidence confirms the data derived from microscopical examination. In the case of the Mt. Gilmore rock it will be noted that the soda dominates over the lime, and this causes the calculated composition of the plagioclase to be about Ab,An,, practically the same as that determined for the Paterson rock. In both of these rocks there is an appreciable amount of potash, and this feature emphasizes the close relation these rocks have with the sodipotassic rhyolites. On account of the value for soda in the Paterson rock exceeding that of potash, the sub-rang to which it belongs is Lassenose, while the Mt. Gilmore rock fits in the sub-rang Toscanose. Comparing the first four of the analyses tabulated, it will be seen that the two Kuttung rocks are quite similar in some respects, while the other two rocks, from California and Hungary respectively, show striking similarities to the Mt. Gilmore rock, especially as regards the values for the alkalis. The analysis, No. V, of the rock which was described originally as a hypersthene- andesite and later called a dellenite by Brogger, differs from the analyses of the Kuttung rocks particularly in having higher potash and lower silica, and indeed it is clear that the two Kuttung dellenites are very close to becoming sodi- potassic rhyolites. Rhyolites. The rhyolites are very abundant, especially in the Volcanic Stage, and present a variety as regards both chemical composition and texture. On account of the fact that many of these rocks are not strongly porphyritic, it is difficult, in the absence of many analyses, to classify them properly. It seems best, however, to consider them in two main sections, viz., those which contain soda felspar in abundance with the absence of visible orthoclase, and those which are sodi-potassic and in some cases strongly potassic. Rhyolites characterized by Albite. These varieties are quite distinct from alkaline lavas whose richness in soda is reflected in the composition of the ferro-magnesian minerals as well BY G. D. OSBORNE. 133 as the felspars, that is to say, the comendite-pantellerite suite. The sodic character -of the rocks under consideration is due to the presence of albite, which might reasonably be regarded as of late-magmatic origin. Megascopical characters: The rocks are light-coloured, and in the case of one important flow, almost emerald green. In the porphyritic varieties the constituents visible are dominant felspar, some quartz and often a little biotite. Frequently, however, there are practically no phenocrysts which can be determined megascopically. The texture of the groundmasses is always felsitic Microscopical characters: The minerals present are quartz, plagioclase, biotite, magnetite, apatite, and rarely hornblende. Quartz may be quite abundant, while at times it is almost missing from the phenocrysts and occurs in the groundmass. It is generally corroded strongly and shows inclusions of the groundmass. The plagioclase is almost always albite, but in one slide some oligoclase was seen. In this case it was being replaced by albite and, indeed, it is probable that much of the phenocrystic albite in the rhyolites is not of primary crystallization, as it exhibits many of the features of the albite in albitized rocks. The felspar is well formed in most cases and possesses a stout prismatic habit. Biotite is present in variable amount, absent in certain rocks, and it is generally strongly resorbed. The pleochroism is always weak. Hornblende was not observed to be a regular constituent, but its presence was detected in a rock from near the top of Mt. Gilmore. The iron ore is mostly ilmenite, but both ilmenite and magnetite occur, the latter being slightly haematitized. Apatite is found in minute quantities in the majority of the rocks. The textures exhibited by these sodic rocks are interesting. Cryptocrystalline groundmasses are the rule and these may be so fine that little can be done in the way of resolving them. Where slightly coarser, one can detect the presence of quartz and acid plagioclase. Spherulitic structures are not common, being generally seen in those rocks which show the presence of secondary quartz. The presence of much pumiceous material, considerably altered, is however a distinctive feature. Thus there may be patches of material which consists of remnants of bubbles which have been re-welded, giving the bogen-structur of Mugge (1893). Secondary silicification has in most cases filled the interstices between some of the disrupted bubbles, while devitrification, followed by kaoliniza- tion, has brought about a selective replacement of the material composing the pumice remnants. Flow structure is to be found in the rocks which are not pumiceous and sometimes spherulites are found in the lines of flow, apparently having developed after the congealing of the material exhibiting the flow lines. The most important occurrence of the soda rhyolites is to the west of Martin’s Creek near the road at the foot of Mt. Johnstone. Specimen 258, from here, is a green rock with an even fracture, and is studded with tabular idio- morphic white felspars. Under the microscope some phenocrysts of corroded quartz are seen, but the felspar is much more abundant, having a spongy appearance suggestive of having developed by deposition during replacement. The grainsize of the felspar is fairly even, and averages about .6 mm. The groundmass is pumiceous and contains some tiny fragments of quartz and rock material. There has been some devitrification and kaolinization. The rock is best called a tuffaceous soda rhyolite. It is almost identical with some specimens from a Kuttung flow at Currabubula, the petrographical features of which have been given by W. R. Browne (1920, p. 408). 134 GEOLOGY AND PETROGRAPHY OF CLARENCETOWN-PATERSON DISTRICT, iv, While the soda rhyolites are being considered it can be pointed out that there are some flow breccias, e.g., Specimen 375, occurring in the Langlands section and elsewhere, which are very sodic and present structures which have developed as a result of autobrecciation during the cooling of an acidic alkaline magma. Rhyolites in which Orthoclase is an Important Constituent. These varieties are of more frequent occurrence than the sodic rocks. Using the criterion of the nature of the phenocrysts one can distinguish between sodi- potassic and those which are dominantly potassic, but it is to be remembered that in all cases it is likely that both the soda and potash molecules exist in the material of the groundmass so that the distinction is one of expediency only, and to be used in the absence of analyses. Megascopical characters: The rocks vary a great deal as regards colour, being often of a dark brown and other times light-coloured. The fracture is quite irregular in those containing large phenocrysts and even in the felsitic types. Quartz is conspicuous in certain specimens and almost absent from others, while the two varieties of felspar are generally distinguishable in hand specimen. Microscopical characters: Under the microscope the following minerals are present: quartz, orthoclase, plagioclase, biotite, ilmenite, apatite and magnetite. The quartz is always corroded and sometimes considerably cracked, though there are no strain shadows under crossed nicols. The orthoclase may be very dominant in those rocks which are taken as being strongly potassic, while it equals, roughly, the amount of plagioclase in the sodi-potassic varieties. The potash felspar is tabular in habit, is always kaolinized, and invariably shows the (001) cleavage. A little albitization is to be seen in several instances. The plagio- clase is generally idiomorphic, and of a tabular prismatic habit. The composition varies from albite to medium oligoclase. Biotite shows a slender habit and generally is strongly resorbed. Ilmenite, indicated by a titaniferous decomposition product, is of more frequent occurrence than magnetite, and apatite is fairly constant in minute prisms. The groundmass of the rocks is mostly cryptocrystalline and residual glass is not a constant feature, although there is some in certain rocks. Flow structure occurs in one or two cases, but it is the presence of pumiceous patches which forms the commonest feature. There has always been devitrification of this pumiceous glass and sometimes secondary silica has entered and affected these pumiceous areas. There are many of the sodi-potassic rhyolites in the Mt. Gilmore section as, for example, Specimens 354, 168, 198, 200 and 497. Of these 168 is distinctive on account of its reddish colour, and amygdaloidal appearance. The colour is due to the selective replacement, by haematite, of a pumiceous groundmass. Specimen 497 is of some interest in being the matrix of a volcanic con- glomerate at the top of the Mt. Gilmore section. It possesses abundant phenocrysts of quartz with subordinate orthoclase and biotite set in a cryptocrystalline groundmass. Associated with the toscanite of the Paterson type, there occur at Porphyry Point and Felspar Creek, Seaham, varieties of potash rhyolite, e.g., Specimen 486, which conform to the general description given above. BY G. D. OSBORNE. 135 Felsites. In addition to the rocks already described there are quite a number which are non-porphyritic and of a typical felsitic appearance. They are generally light in colour, although some dark brown types are known. Under the micro- scope one is unable to assign any definite identity to these rocks since phenocrysts are extremely rare and when present are generally of quartz or, if of felspar, are too small or too altered to admit of accurate determination. The groundmass in these rocks, which in places constitutes almost 95% of the whole, may be eryptocrystalline as in the case of Specimen 374, from Oakendale, or it may be an altered pumice as in Specimen 17, from near the Gostwyck Bridge, Paterson. In this rock the attenuated bubbles have been replaced by kaolin and filled with secondary silica. In the case of a group of felsitic rocks occurring to the east of the Railway on the southern side of the valley of Martin’s Creek (Specimens 25 and 26), the groundmass contains many spherulites and axiolites as already described (see above). In connection with the felsites one can point out that although the pheno- erystic felspar was very scarce and always somewhat decomposed, measure- ments that were made upon the extinction angles suggested that in all cases the felspars were strongly sodic, although the determination was never definite. Tuffs and Certain Detrital Rocks. Tuffs. A salient feature of the Kuttung Series is the abundance of tuff throughout. Much of this tuffaceous material is mixed with detrital sediment and it is impossible to arrive at the composition of the material of truly igneous origin in these cases. There are, however, some tuffs which appear to be of primary deposition and therefore of importance almost equal to that of many flows. These tuffs do not vary much in hand specimen, being generally of a rusty red colour, due to the presence of haematite. At other times the tuffs are greenish, due to the presence of a silicate of iron. The grainsize is generally even and medium, cases of coarse breccia and agglomerate being relatively rare. Microscopical characters: Microscopically there are seen to be certain. differ- ences in the various specimens and the leading types will be considered in turn. Specimen 147 is from the Mt. Gilmore section and contains fragments of acidic rocks, some of which are felsitic and some of the nature of keratophyres. Albite occurs along with quartz. Some patches of secondary silica are seen and haematite is the cementing material. It is interesting to notice again the prevalence of sodic constituents. Specimen 10, from the little quarry on the Paterson-Dungog road near Mt. Johnstone, contains fragments of quartz and andesine together with a lot of pumice cemented by a greenish matrix. The pumice has been replaced in part by haematite and shows up well in contrast to the rest of the rock. Secondary silica occurs in patches and there are some vitrophyric rock inclusions. Specimen 294 which is dark in colour is from the Tillimby paddock, Paterson. It is very finely textured and consists of quartz and indeterminate felspar set in a dark brown matrix. High magnification shows this to consist of tiny pieces of siliceous rock, haematite and chloritic (?) material. 136 GEOLOGY AND PETROGRAPHY OF CLARENCETOWN-PATERSON DISTRICT, iv, Specimen 298, from about three-quarters of a mile south of Martin’s Creek Station and east of the Railway Line, is of interest in that there is a considerable amount of biotite present. Specimens 373, 381, 383 and 449 are tuffs occurring in the Langlands section and are all of the same general facies, being siliceous, the fragments comprising abundant quartz, acidic rocks (many of which are quite glassy), spherulites, and chips of acid plagioclase. It will thus be noticed that nearly all the tuffs have constituents which give the rocks alkaline affinities and that in some cases biotite is an important constituent. Tuffaceous Rocks in the Main Glacial Beds. The lithology and petrography of the Main Glacial Beds have already been written (Osborne, 1925) and it has been pointed out there that many rocks which would be taken at first sight to be tuffs or breccias are, in part at least, of glacial origin. As a matter of fact there are only local masses of true tuff and breccia, but extensive deposits of tuffaceous sandstones and coarser rocks occur. With regard to the composition of the few true tuffs, microscopical examina- tion has shown the fairly constant features of a siliceous character and a content of acid plagioclase. Basal Stage Conglomerates. Conglomerates form such an important part of the Basal Stage in the vicinity of Wallarobba, Mt. Douglas and Clarencetown, and it is desirable that a petro- logical description of these be given. The pebbles in the conglomerates average about 4 inches in diameter, while some are about one foot in diameter. The nature of these old gravels has not been determined in any detail, but the pebbles occur in the following general order of decreasing abundance: quartzites, aplites, cherts, porphyries, felsitic rocks and granitic types. There is an almost entire absence of rocks belonging to the diorite-andesite suite. The nature of the matrix of the conglomerates no doubt varies from place to place. Some samples show, in hand specimen, the presence of fairly fresh chips of felspar along with quartz and tiny rock fragments. A fairly representative sample is Specimen 152 from Clarencetown, near the Williams River Bridge. Megascopically it is of a dark grey colour, the constituents being of a fine grainsize cemented by argillaceous material. Microscopically, it is seen to be composed of fragments of siliceous rocks, quartz, and plagioclase, with patches of chlorite and haematite, the cement being kaolinic. Many of the rock pieces are devitrified obsidians possessing spherulites. Other rock types are felsitic and may always have been lithoidal. There is evidence that the fragments have been somewhat rounded by transport. PETROGRAPHY OF CERTAIN PosT-CARBONIFEROUS Rocxs. As pointed out in another paper, there occur in the area under discussion remnants of widespread flows of Cainozoic basalt and several dykes of basic material. The material of the dykes is in all cases too much weathered to permit microscopical examination, with one exception, this being the dyke on Mr. A. J. C. Vogele’s property at Mt. Douglas. BY G. D. OSBORNE. 137 Specimen 82, from the margin of the dyke, is of finer grainsize than Specimen 106 which is from the central portion of the mass. Both are somewhat porphyritic in plagioclase and show a typical doleritic appearance. Microscopical characters: Specimen 82 is seen to be a good deal decomposed. There are present plagioclase, augite, magnetite, biotite, apatite and a little free quartz. The plagioclase occurs as phenocrysts at times, but the bulk of it belongs to the second generation, having a prismatic habit. It is very much saussuritized and its composition is almost indeterminate, although there is no doubt of it being calcic. Augite is grey and ophitic relations exist between it and felspar. There is very little biotite which, like the magnetite, is fresh. Chlorite and haematite are seen to be filling cracks rather than representing replacement patches due to decomposition. In addition there is a considerable quantity of a secondary greenish silicate. The quartz is interstitial and undoubtedly was the last mineral to crystallize. Specimen 106 is coarser in grainsize and there is a fair amount of pheno- erystic plagioclase which possesses a tabular habit and an average grainsize of 2 mm. The composition of the phenocrysts is about Ab,An,, while that of the felspar in the groundmass is slightly more acid. The latter is prismatic in habit and is associated with quartz, augite, titaniferous iron ore, biotite, quartz and apatite, the mass showing ophitic fabric. Chlorite is the chief decomposition product. Both the above rocks may have the name quartz-bearing gabbro porphyrite or porphyritic dolerite. Specimen 85 is from the basalt on Mt. Douglas. In hand specimen it has the normal basaltic appearance, and under the microscope is seen to be an olivine- basalt. The olivine phenocrysts are completely pseudomorphed by the material which is generally described as iddingsite, and limonitic stains which have originated from the alteration of the olivine show also that this mineral was originally distinctly ferriferous. - In the groundmass, which is subophitic in texture, there are laths of basic plagioclase, considerable iron ore, augite and apatite, all remarkably fresh in contrast with the olivine. A significant feature in connection with the order of consolidation is that the bulk of the iron ore crystallized in ragged masses after the felspar. List of Works referred to. ANDERSON, E. M., and RapDuLey, E. G., 1917.—The Pitchstones of Mull and Their Genasis. Quart. Journ. Geol. Soc. \xxi, pt. 2, 205-217. AUROUSSEAU, M., 1915.—Petrological Notes. No. 1. Igneous Rocks and Tuff from the Carboniferous of New South Wales. THESE PROCEEDINGS xl, p. 294. BaiLey, E. B., 1911.—The Geology of the Glasgow District. Mem. Geol. Surv. Scotland, p. 127. BaiLey, E. B., and GRABHAM, G. W., 1909.—The Albitisation of Basic Plagioclase Felspar. Geol. Mag., p. 250. BENSON, W. N., 1918.—The Geology and Petrology of the Great Serpentine Belt of New South Wales. THESE PROCEEDINGS xliii, pt. 2, p. 371. BonnNeEY, T. G., 1885.—Anniversary Address. Quart. Journ. Geol. Soc. xli. BONNEY, T. G., and PARKINSON, J., 1903.—On Primary and Secondary Devitrification in Glassy Igneous Rocks. @Q.J.G.S. 1903, p. 429. BROWNE, W. R., 1920.—The Geology and Petrology of the Great Serpentine Belt of New South Wales. Part ix, Section C. Petrography. THESE PROCEEDINGS xlv, pt. 3, 405. —— 1523.—Albitisation and Kindred Phenomena in certain Carboniferous and Permian lavas of New South Wales (abstract). Rept. A.A.A.Sc., Wellington Meeting, p. 345. K 138 GEOLOGY AND PETROGRAPHY OF CLARENCETOWN-PATERSON DISTRICT, iv. — 1925.—Notes on the Petrology of the Prospect Intrusion with Special Reference to the So-called Secondary Minerals. Journ. Roy. Soc. N.S.W. lviii, p. 240. and Wa.ukom, A. B., 1911.—The Geology of the Eruptive and Associated Rocks of Pokolbin, N.S.W. Journ. Roy. Soc. N.S.W. xlv, pp. 379-408. GREEN, J. F. N., 1919.—The Vulcanicity of the Lake District. Proc. Geol. Assoc. xxx, pt. iv, p. 153. Harker, A., 1909.—The Natural History of Igneous Rocks. Houmes, A., 1920.—The Nomenclature of Petrology. IDDINGS, J. P., 1885-86.—Seventh Ann. Rept. U.S.G.8. 249-295. 1899.—U.S8. Geol. Surv., Monograph 32, pt. 2. 1910.—Igneous Rocks, Vol. I. Jupp, J. W., 1889.—The Growth of Crystals in Igneous Rocks after their Consolidation. Q.J.G.S. xlv, p. 175. NBITHAMMER, G., 1908.—Die Hruptivgesteine von Loh Oleo auf Java. Tschermak’s Min. Petr. Mitth. p. 218. OSBORNE, G. D., 1922.—The Geology and Petrography of the Clarencetown-Paterson District. Part I. THESE PROCEEDINGS xlvii, pt. 2, 161-198. 1925.—The same. Part 3. THESE PROCEEDINGS, l, 67-79. SARGENT, H. C., 1917.—Quart. Journ. Geol. Soc. \xxili, 1917-18, p. 19. (Quoted by A. Holmes, 1920). SEDERHOLM, J. J., 1916.—Bull. Comm. Geol. Finland, No. 48, p. 142. Stamp, L. D., and Woo.LripGE, 1923.—The Igneous and Associated Rocks of Llanwrytd, Brecon. Quart. Journ. Geol. Soc. 1xxix, p. 17. SussmitcH, C. A., 1923.—Presidential Address. Journ. Roy. Soc. N.S.W. lvii. SussmitcH, C. A., and Davip, T. W. E., 1919.—The Sequence, Glaciation and Correlation of the Carboniferous of the Hunter River District, N.S.W. Journ. Roy. Soc. N.S.W. lii, 247-338, with two appendices. WASHINGTON, H. S., 1917.—Chemical Analyses of Igneous Rocks. United States Geol. Survey, Prof. Paper 99. EXPLANATION OF PLATE XXIII. Fig. 1.—Example of albitization of a phenocryst of andesine in a rhyolite from Mt. Gilmore. The remnants of andesine are white and stand out in contrast with the rest of the grain which is now albite. Ordinary light. Magnification, 28 diameters. Fig. 2.—Composite zoned crystal of plagioclase in the Martin’s Creek andesite showing selective replacement by albite and chlorite. The latter has replaced some of the central zones, while the former is found towards the periphery, and is separated from the chlorite by a zone of unaltered andesine (white). Crossed nicols. Magnification, 12 diameters. Fig. 3.—Williams River type of quartz-keratophyre. Clear grains of quartz and subidio- morphic grains of spongy albite are set in a dense groundmass. Polarized light. Magnification, 12 diameters. Fig. 4.—Lithoidal phase of pyroxene-andesite showing the presence of augite (oval grain near centre) and prismatic hypersthene. Polarized Light. Magnification, i2 diameters. Fig. 5.—Glassy variety of transitional type of andesite. Portion of a resorbed phenocryst of hornblende is to be seen towards the top, while augite and hypersthene are in association as indicated, and small pieces of biotite occur scattered near the centre. Ordinary light. Magnification, 14 diameters. AUSTRALIAN MYDAIDAE (DIPTERA). By G. H. Harpy, Walter and Eliza Hall Fellow in Economic Biology, University of Queensland, Brisbane. (Two Text-figures. ) [Read 29th April, 1925.] This paper, revising the described Mydaidae of Australia, was prepared several years ago and, at the time, no less than five species were found to be new, but all the specimens available were old and dilapidated in condition. No further speci- mens of these, or of any other undescribed Mydaids, appear to have been collected in recent years, and this paper, therefore, does not include new forms. No less than four Australian Entomologists have been seeking knowledge of the family, either for purposes of publication or of identification of material in collections, and so, to supply the growing demand for some knowledge concerning these hand- some flies, it is considered advisable to issue the paper in its present form, rather than to await the accumulation of sufficient material for description of specific details, as was originally intended. The Mydaidae have been revised twice, by Westwood in 1841, and by Gerstaecker in 1868, and both these authors dealt with the species of the World. Westwood’s papers are enhanced in value by the excellent coloured figures, whilst Gerstaecker’s work marks a big advance in the taxonomy of the family. In the present revision, various names have been placed as synonyms, one being the generic name Triclonus, the two specific names of which are now associated with the genus Diochlistus for the first time. Family MYDAIDAE. This family is recognizable by the complex venation of the wing described below. It contains more or less large elongate flies that have the antennae, usually long, containing five joints, of which the second is small, the third usually the longest and the fifth much wider than the others, inflated on live, but compressed on dried, specimens. Wings.—The costal and subcostal veins end before the apex of the wing; the mediastinal vein between them reaches beyend half the length of the costa; the radial vein runs into the subcostal considerably before the wing margin. The upper branch of the cubital vein runs into the subcostal, and the lower branch may run to the wing margin, to the subcostal vein, or into the upper branch. In the latter case, the second submarginal cell becomes entirely enclosed by the cubital fork. The first postical vein may be present or absent; it is suggested, in the latter case, that the first postical may be concurrent with the second as one undescribed species retains the two veins, still separated at the base, enclosing a triangular area. The second postical vein runs to about the apex of the wing. 140 AUSTRALIAN MYDAIDAE, The third and fourth posticals unite before reaching the wing margin, thus closing the fourth posterior cell. The fifth postical runs to the wing margin and joins the anal vein before reaching there. Distribution.—The family is not known from Tasmania and, judging from the collections examined, there appears to be one species limited in distribution to Queensland, whilst another extends into that State. The majority of the species occur from New South Wales to Western Australia. The correctness of some of the localities given with original descriptions is doubtful, and it is significant to note that Westwoad gives “Adelaidam, Australias occidentalis,” apparently confusing Adelaide with Western Australia. Diochlistus gracilis Macquart is only known by recent specimens collected in Eastern Australia, and yet, under a synonym, Westwood gives it as from Western Australia. Under the same species, Schiner adds “New Zealand,” an undoubted error which this author also makes under some of his other references to. Australian Diptera. Notes.—There are nine apparently valid species and two genera described from Australia, and of these, eight are definitely distinct; the other species is not represented in the Australian collections examined. Five undescribed Mydaids are very different in appearance from those already described, but they belong to the same two genera. The species are liable to vary in size, shape, and colour, and there are too few structural characters yet found to allow a reliable key to the species to be formed. To minimize the search necessary for the determination of species, a search that requires access to certain works that are rarely repre- sented in the various scientific libraries, the leading colour characteristics of each species are given here; although this scheme may help considerably in any attempt to identify material, reference to the original descriptions and figures needs to be made, as specimens of unusual colours and markings are otherwise liable to be misidentified. Key to the Genera of the Mydaidae. First postical vein present, not amalgamated with the second; there are three cells (marked 1, 2 and 3 in figure) between the cell enclosed by the branches of the cubital vein and the hind border of the wing .......................... Diochlistus. First postical vein completely coincident (or mostly so) with the second; there are therefore only two cells (marked 1 and 2 in the figure) between the cell enclosed by the branches of the cubital vein and the hind border of the wing ...... Miltinus. Genus DiocHListus Gerstaecker. Gerstaecker, Stett. Hnt. Zeit. xxix, 1868, 73—Triclonus Gerstaecker, ibid., 75. Type, D. mitis Gerstaecker, from Swan River, W. Australia. Synonymy.—The differences between the genera Diochlistus and Triclonus, as given by Gerstaecker, include the structure of the face; the area at the sides of the oral opening are produced and stand out from the eyes in the latter genus, whilst they lie level with, or recede from, the eyes in the former, but this cannot be accepted as of generic value as gradations are found. A second character is the position in which the proboscis is held, namely, horizontally in Diochlistus and vertically in Triclonus; both positions are found to occur on at least one species placed by Gerstaecker in the latter genus. The wing characters also grade between the species placed under these genera. As Gerstaecker has not offered any reliable character for dividing the genus into two, nor are there any apparent, it is advisable to place the second name as a synonym of the first. BY G. H. HARDY. 141 DIOCHLISTUS MITIS Gerstaecker. Gerstaecker, Stett. Ent. Zeit. xxix, 1868, 73, Pl. i, fig. 1; Osten-Sacken, Ber. Ent. Zeit. xxxvi, 1891, 313. A dull species, black or grey in colour. Hab.—Western Australia (Gerstaecker). South Australia; 4 $, 1 9, in the Macleay Museum. i wn 88-45 Wy > Text-figs. 1-2.—Diagram of the wing venation. 1, Diochlistus; 2, Miltinus. h., humeral vein; m., mediastinal vein. Se., subcostal vein; R#., radial vein; Cw., cubital vein with two branches; P,, first postical vein; second and subsequent postical veins denoted by figures Ps, etc.; and when two of these run together they are denoted as P24, ete.; An., anal vein; d.c., discal cell; c., second submarginal, also known as the cubital cell, which is enclosed by the two branches of the cubital vein. The figures 1, 2 and 3 are to draw attention to the number of cells between the cubital cell and the posterior border of the wing; the number of these cells constitutes the most readily perceived distinction between the two genera. DIOCHLISTUS AURIPENNIS Westwood. Mydas auripennis Westwood, Lond. Edin. Phil. Mag. vi, 1835, 281; Isis ii, 1838, 84; Arc. Hnt. i, 1841, 51, Pl. xiv, fig. 1; Walker, List Dipt. Brit. Mus. vi, suppl. 2, 1854, 368.—Triclonus auripennis Gerstaecker, Stett. Ent. Zeit. xxix, 1868, 75.— Mydas fulvipennis Macquart, Dipt. Exot. suppl. 4, 1849, 58, Pl. iv, fig. 3. A yellow species with black thorax. It is peculiar in having the two branches of the cubital vein uniting before they reach the subcostal vein, thus completely enclosing the cubital (second submarginal) cell. Hab.—Queensland: Brisbane, 1920; Jandowie (Darling Downs), 3 dg, 1 Q, December, 1920 (R. Illidge). 142 AUSTRALIAN MYDAIDAE, DIOCHLISTUS GRACILIS Macquart. Cephalocera gracilis Macquart, Dipt. Exot. suppl. 2, 1847, 32, Pl. i, fig. 5; Walker, List Dipt. Brit. Mus. vi, suppl. 2, 1854, 376.—Triclonus gracilis Kertesz, Cat. Dipt. iv, 1909, 34.—Mydas bispinifer Westwood, Trans. Ent. Soc. Lond. v, 1848, 88, Pl. xiii, fig. 2; Walker, loc. cit., 1854, 370.—Triclonus bispinifer Gerstaecker, Stett. Ent. Zeit. xxix, 1868, 75; Williston, Kansas Univ. Quart. i, 1893, 154, Pl. x, fig. 10—Mydas melleipennis Westwood, loc. cit., 1848, 87, Pl. xiii, fig. 1; Walker, loc. cit., 1854, 369.—Triclonus melleipennis Gerstaecker, Stett. Hnt. Zeit. xxix, 1868, 75.—Mydas clavata Macquart, Dipt. Havot. suppl. 4, 1849, 59, PI. iv, fig. 5— Harmophana clavata Thomson, Hug. Resa, Dipt., 1869, 463, Pl. ix, fig. 5—Mydas gracilis Jaennicke, Abhand. Senckenb. Naturf. Gesell. vi, 1867, 353, Pl. xliii, fig. 12—Mydas effracta Walker, Trans. Ent. Soc. Lond., iv, 1857, 126.—Triclonus effractus Gerstaecker, loc. cit., 76.—Mydas macquarti Schiner, Nov. Reise, Dipt., 1868, 153; Hutton, New Zealand Dipt., 1881, 31—Harmophana flavipes Thomson, Hug. Resa, Dipt., 1869, 463. Synonymy.—For this species, Gerstaecker used the specific name bispinifer, as he considered gracilis to be preoccupied by Mydas gracilis Macquart. Schiner proposed the name macquarti to take the place of clavatus, as he placed the species under the genus Mydas where the latter name had already appeared. Mydas melleipennis Westwood, Mydas effracta Walker and Harmophana flavipes Thomson appear to belong here. The other names were placed as synonyms by Gerstaecker. Note.—A black species with a pair of yellowish or reddish spots on each of the three or four basal segments of the abdomen. The legs vary in colour. Hab.—New South Wales: Sydney, December, a series of both sexes. Queens- land: Stanthorpe and Tambourine Mountain. Genus Mictinus Gerstaecker. Gerstaecker, Stett. Ent. Zeit. xxix, 1868, 88. Type, M. cardinalis Gerstaecker, South Australia. MILTINUS CARDINALIS GerstaecKer. Gerstaecker, Stett. Hnt. Zeit. xxix, 1868, 90, Pl. i, fig. 5. A reddish species; the abdomen with black lateral spots on the segments. Hab.—South Australia. In the Macleay Museum there are thirty-one specimens without further data. MILTINUS BICOLOR Westwood. Mydas bicolor Westwood, Arc. Ent. i, 1841, 53, Pl. xiv, fig. 2; Walker, List Dipt. Brit. Mus. i, 1848, 228; also vi suppl. 2, 1854, 368.—Miltinus bicolor Ger- staecker, Stett. Ent. Zeit. xxix, 1868, 90. Miltinus haemorrhous Gerstaecker, loc. cit., 89. Synonymy.—The colour characters of Mydas bicolor Westwood and of Miltinus haemorrhous Gerstaecker are the same. Both are recorded from the same State, and a series before me agrees with both descriptions and with Westwood’s figure. Apparently there are no published characters whereby haemorrhous can be distinguished from bicolor, even should they prove to be distinct, and on this account they are regarded as one species. BY G. H. HARDY. 143 Note.—A black species with the legs and the posterior segments of the abdomen red. Hab.—Western Australia: Perth, December, 1911, 6 ¢, 1 9, all of which have been attacked by Anthrenus. The species is very common around Perth, where it can be found hovering over bare sandy patches amongst low vegetation. MILTINUS MACULIPENNIS Westwood. Cephalocera maculipennis Westwood, Arc. Ent. i, 1841, 55, Pl. xiv, fig. 5; Walker, List Dipt. Brit. Mus. vi, suppl. 2, 1854, p. 375.—Miltinus maculipennis Gerstaecker, Stett. Ent. Zeit. xxix, 1868, 90. This reddish species with the apex of the abdomen black is rendered very conspicuous by having a large black blotch in the centre of the wing. Hab.—Western Australia: Cunderdin, 1 4, in the Queensland Museum. MILTINUS SORDIDUS Westwood. Mydas sordidus Westwood, Trans. Ent. Soc. Lond. v, 1848, 89, Pl. xiii, fig. 3; Walker, List Dipt. Brit. Mus. vi, suppl. 2, 1854, 370.—WMiltinus sordidus Gerstaecker, Stett. Hnt. Zeit. xxix, 1868, 90.—Mydas limpidipennis Westwood, I.c., 90; Walker, loc. cit., 369.—Miltinus limbipennis Gerstaecker, loc. cit., 90.—Mydas claviger Walker, List Dipt. Brit. Mus. i, 1848, 229; vi, suppl. 2, 1854, 367.—-Miltinus claviger Gerstaecker, loc. cit., 90. Synonymy.—The type of Mydas sordidus is from Adelaide; Gerstaecker gives Western Australia as a locality. The type of M. limpidipennis is from Western Australia, but as Westwood credited Adelaide to the same State (Australiae occidentalis), this recorded habitat cannot be considered satisfactory. Westwood’s two names appear to be referable to different forms of the same species. The type of Mydas claviger Walker is stated to be near M. sordidus, but it will probably prove identical with that species. Note.—A black species with triangular red lateral spots on the segments of the abdomen. Hab.—South Australia. MILTINUS STENOGASTER Westwood. Mydas stencgaster Westwood, Arc. Hnt. i, 1841, 53, Pl. xiv, fig. 3; Walker, List Dipt. Brit. Mus. i, 1848, 228; vi, suppl. 2, 1854, 368.—Miltinus stenogaster Gerstaecker, Stett. Ent. Zeit. xxix, 1868, 90. A black species with a red abdomen. Hab.—Western Australia. MILTINUS VIDUATUS Westwood. Mydas viduatus Westwood, Lond. Edin. Phil. Mag. iv, 1835, 281; Isis ii, 1838, 85; Arc. Hnt. i, 1841, 52, Pl. xiv, fig. 2; Walker, List Dipt. Brit. Mus. i, 1848, 299; and vi, suppl. 2, 1854, 369.—Miltinus viduatus Gerstaecker, Stett. Hnt. Zeit, xxix, ‘1868, 89.—Mydas concinnus Macquart, Dipt. Hxzot. suppl. 1, 1846, 58, Pl. vi, fig. 5; suppl. 3, 1848, 17, Pl. ii, fig. 1; Walker, List Dipt. Brit. Mus. vi, suppl. 2, 1854, 369; V.d. Wulp, Tijd. v. Ent. xix, 1876, 171—Mydas varipes Macquart, Dipt. Hzot. suppl. 4, 1849, 58, Pl. iv, fig. 4.—Miltinus varipes Gerstaecker, Stett. Hnt. Zeit. xxix, 1868, 89.—Mydas signatus Walker, Trans. Ent. Soc. Lond. iv, 1857, 126. 144 AUSTRALIAN MYDAIDAE. Synonymy.—Gerstaecker included Mydas concinnus Macquart and Mydas signatus Walker in the synonymy of this species. Mydas varipes Macquart also appears to be the same species. Note.—Normal specimens of this species are entirely black. In the Macleay Museum there is a variety, from South Australia, that has the fourth and subse- quent segments of the abdomen red; this variety must not be confused with M. bicolor, and it does not conform to the description of M. haemorrhous Gerstaecker. Hab.—New South Wales: Sydney and Blue Mountains. South Australia: two specimens in the Macleay Museum. AN ECOLOGICAL STUDY OF THE FLORA OF MOUNT WILSON. Part ii. THE EUCALYPTUS FORESDIS. By ArtHur H. K. PETRIE, Demonstrator in Botany in the University of Sydney. (Plates xx-xxii; and four Text-figures. ) [Read 29th April, 1925.] Contents. Introduction. General Physiognomy of Eucalyptus Forest. Classification of Communities. The Associations. The Hucalyptus-Doryphora Heotone. The Hucalyptus goniocalyx-E. Blaxlandi Association. The Hucalyptus piperita-E. haemastoma var. micrantha Association. 8. The Stratum-Societies of the Junction Flora. 9. Summary. AMMAN Pwd re Introduction. In the first of these Memoirs on the plant-covering of Mount Wilson, a study was made of the vegetation of the basalt soil, which comprises the Malayan Rain- Forest and certain communities on the outskirts dominated by the endemic Hucalyptus. In this second Part it is proposed to give a brief account of some of the salient features of the Eucalyptus Forests, which constitute the vegetation or the sandstone plateau, supplementing also the previous observations on the Eucalyptus communities of the basalt. A third publication will deal with the plant communities of the valleys and their distribution, correlating them with those of the plateau. A EKucalyptus Forest, the unique Hndemic Flora of this Continent, is perhaps one of the most extensive plant formations in the world, and its ecology is prac- tically an unopened book: little more has, therefore, been attempted in the present paper than to outline the main communities in the small area under consideration; to record some observations on their distribution and inter-relationships; and to hint at a few of the many important problems connected with their adaptations, development, and such other features as are urgently awaiting intensive study. I have to record my indebtedness to Assistant-Professor McLuckie for his help in the field and advice in the preparation of the paper; to Professor Lawson for the interest he has always shown in the Mount Wilson work, and for suggestions on a number of occasions; to Mr. M. B. Welch, B.Sc., A.I.C., for kind permission to reproduce one of his photographs of Mount Wilson; and to Mr. O. D. Evans for his generous assistance in the identification of a number of the types referred to in this paper. L 146 AN ECOLOGICAL STUDY OF THE FLORA OF MOUNT WILSON, ii, GENERAL PHYSIOGNOMY OF EHUCALYPTUS FOREST. Eucalyptus Forest, the chief expression of the Australian endemic flora, as has been said, is perhaps one of the most extensive plant formations in the world, and occupies the greater part of Australia. It was stated in Part I that the Australian flora possessed a number of outstanding peculiarities; and, indeed, many of the features of the Eucalyptus Forests are unique. With a few exceptions, the association dominants are species of Hucalyptus, and the fact, that there are about three hundred species in Aus- tralia, gives some idea of the peculiar complexity of the formation; despite this, however, the physiognomy is characteristically uniform, the components, almost entirely xerophytes, being typically sclerophyllous in life-form. All the species of Eucalyptus possess the same xerophytic facies imparted by their gnarled and spreading boughs, and broad crown of drooping isobilateral leaves. The foliage is small in amount for the size of the trees, being chiefly distal, a xerophytic adapta- tion which is not least among the distinctive physiognomic features of these trees. The bark is also characteristic, being either stringy, rough, often fissured, and dark grey in colour; or else smooth, pale bluish-grey in colour, containing chlorophyll in the phelloderm, and being cast annually. The accompanying plates (Pl. xx-xxii) help to bring out a number of these features. The trees are seldom close enough for their crowns to meet, and in any case the open nature of their foliage results in little obstruction to the penetration of light to the ground; there is thus considerable scope for the development of shrubs, which are a characteristic feature of most Hucalyptus Forests, and are extremely numerous in species and life-form (see Text-fig. 1), although the majority are markedly xerophytic, sclerophyllous and often coriaceous. The ground between the shrubs is occupied by smaller shrubby plants, grasses and annuals, also highly xerophytic. Thus, on the whole, Eucalyptus Forest is a very xerophilous type of vegetation; but some associations, adapted to damp or sheltered habitats, as has been seen on the basalt at Mount Wilson, contain more mesophytic types, although it is note- worthy that many of the latter are Malayan rather than endemic in origin. The contrast between Eucalyptus Forest and the Rain-Forest at Mount Wilson is profound: the former is lowly integrated and a characteristic expression of xerophily; the latter has the high integration characteristic of luxuriant vegeta- tion and is typically mesophilous. This feature of integration is clearly shown by the result of the cutting out of the dominants in the two communities: in the Eucalyptus Forest no great change as a rule ensues, except that the shrubs form a closer community; in the Rain-Forest, on the contrary, there is little doubt but that the majority of the subordinates, which are present only as a result of the powerful reaction of the dominants upon the rest of the community, would disappear entirely; even in the Eucalyptus-Doryphora association great changes have taken place. Thus it is that the interdependence characteristic of the Rain- Forest is distinctly lacking in Eucalyptus Forest: epiphytes are scarce; ombrophytes are absent. Another feature of contrast, so far as Mount Wilson is concerned, is the great floristic richness which often characterises Eucalyptus Forests and increases the possibility of variation; while the Rain-Forest is com- paratively poor in number of species at Mount Wilson on account of the severity of the climate. BY A. H. K. PETRIE. 147 CLASSIFICATION OF COMMUNITIES. With those who are carrying out the pioneering investigations in synecology in Australia, lies the responsibility of co-ordinating upon a scientific basis of classification the communities which they are studying; and the fact, that little has so far been done in this direction, added to the great complexity of Hucalyptus Forests, is such as to render their ecological classification a task often beset with difficulty. In this study the concepts of the British workers have been followed as closely as possible, these being based largely on Clements’s scheme (Clements, 1916) with the adoption, however, of Tansley’s conception of climax communities (Tansley, 1920) The Eucalyptus Forests have been divided into associations, regarding all communities resembling one another in floristic composition as belonging to one association, these being divided into consociations according to the distribution of the dominants. In certain cases the number of consociations in an association will probably be found to be extraordinarily large when all the Eucalyptus Forests have been surveyed, an inevitable consequence with lowly integrated communities and with a genus so versatile as Hucalyptus. This is a feature which leads to a certain difficulty in nomenclature: for instance, what is assuredly .one association, having consociations of Eucalyptus piperita and #H. haemastoma var. micrantha at Mount Wilson, is dominated in other localities on the Blue Mountains by HL. Sieberiana and other species, while in the Sydney districts consociations of #. haemastoma and E. pilularis have been observed. In designating such an association, as Clements has pointed out, only two names can be used; yet it is desirable that the two selected should be the dominants most widely distributed and most characteristic of the association. As this information has not yet been obtained in the case in question, the association has been named for the present from the two dominants occurring at Mount Wilson. It must be understood, however, that this term will possibly require alteration later on; indeed, the whole concept of the status of the communities of the Eucalyptus Forests and their nomenclature, given in this paper, must be regarded as more or less tentative, being only a first attempt based upon observations made over a small area. The low integration of Eucalyptus Forests is likely to be responsible for many problems connected with the definition of status. It is no uncommon occurrence for one society of shrubs to occur in two different associations; and the typical shrub-stratum of a Eucalyptus association may occur where the development of trees is inhibited, with the result that the shrub-community changes from the rank of a stratum-society to that of a definite association. Examples of these phenomena will be discussed in the subsequent pages. Having divided the Eucalyptus Forest into associations, it is desirable to have a name for Eucalyptus Forest as a whole, and for this the term “formation” seems to suggest itself. Unfortunately this word has had a somewhat chequered career, being used in different senses by different workers: yet this reversion to its use in the sense indicated by Warming (1909) seems perhaps more useful in the present instance than the most recent definition of Tansley (1920); this, however, only time and the work of subsequent investigators of Australian ecology will show. THE ASSOCIATIONS. Following upon the method of ciassification adopted above, two associations have been recognised for the present in the Eucalyptus Forest at Mount Wilson. 148 AN ECOLOGICAL STUDY OF THE FLORA OF MOUNT WILSON, ii, They are: (1) An association dominated by Eucalyptus goniocalyx, E. Blazxlandi and E. viminalis, which will be termed for convenience the Eucalyptus goniocalyz- E. Blazlandi association. This includes what was previously called the Hucalyptus- Alsophila association and part of the Hucalyptus-Pteridium association. (2) The Eucalyptus piperita-E. haemastoma var. micrantha association, which clothes the greater part of the sandstone plateau and the dry slopes. In addition to these is the Eucalyptus-Doryphora association, which is perhaps better regarded as an ecotone. THE HuCALYPTUS-DORYPHORA HEICOTONE. This community has been described already in Part I (p. 485), and represents the ecotone between the Ceratopetalum-Doryphora and the Eucalyptus goniocalyz- E. Blaxlandi associations. In such places as it forms a thin, fragmentary fringe at the edge of the Ceratopetalum-Doryphora association, Hucalyptus goniocalyx and #. Blaxlandi are co-dominants, while H. viminalis is occasionally present; the optimum expression of the association, however, is found on the lower slopes of the mountain, on the first basalt residual reached by the road from Bell (see Map, Part I, Plate lvii). Here Eucalyptus viminalis occurs as a dominant in certain areas, while Hucalyptus oreades is locally dominant, forming societies in which the other species are only occasional. THE EUCALYPTUS GONIOCALYX-H. BLAXLANDI ASSOCIATION. Nomenclature. What were termed the Hucalyptus-Alsophila and Eucalyptus-Pteridium associa- tions have been described at some Jength in Part I. The system of nomenclature adopted therein for these communities, however, was purely a tentative one; for, at the time of writing, the observations on the distribution of the Eucalyptus trees were not verified, and therefore were not adopted as a basis of classification. Meanwhile, a special study has been made of the distribution of Hucalyptus at Mount Wilson, and it is now possible to place the nomenclature of these com- munities upon a more logical foundation. The co-ordination of the communities of the Junction Flora, however, is found to be by no means simple, as is indicated by the following diagrammatic representa- tion of the manner in which they overlap in the various strata: EK. goniocalyx-E. Blaxlandi association [RecN CR PNA IE LEASES SET ot RP LAE UNIS ae EES EE Ee ea ONE AY Ceratopetalum-Doryphora E. piperita consociation Pars i = WSS AR WS SETI Sere Bee Cs DE SESS) \ Dicksonia c AUS Oypin Mel Ordinary Sandstone Shrubs Polystichum-Blechnum Blechnum Pteridium = [= Na? SL ae It is here seen that the Alsophila zone does not delimit an association, since the dominant trees extend into the Pteridiwm zone; and also the Pteridium zone extends into the H. piperita Forest, an example of the low integration of EHucalyptus communities. These facts have led to the establishment of the following system of nomenclature: Ceratopetalum-Doryphora association Eucalyptus-Doryphora ecotone Alsophila stratum-society Pteridium stratum-society E. piperita consociation of the L. piperita- (Pteridium stratum-society E. haemastoma var. micrantha asso- J'Typical sandstones shrub ciation stratum-society E. goniocalyz-E. Blaxlandi association BY A. H. K. PETRIE. 149 Distribution and Structure. The Alsophila and Pteridiuwm stratum-societies have been discussed to some extent in Part I; it remains, however, in this and a subsequent Part, to record some observations on the distribution of the components of the tree-stratum. Eucalyptus goniocalyz, E. Blaxzlandi and E. viminalis are seldom if ever co-dominant in one place, more usually forming consociations in which the others are only subordinate; H. oreades occurs also as a locally frequent type in this association. The H. viminalis consociation, with H. Blaxlandi abundant, was observed only in one place (see Plate xxi, fig. 5), on a southern basalt slope near the road immedi- ately east of the Section line (see map, Part I, Plate lvii); elsewhere the E. goniocalyx consociation, with £. Blazlandi (f-a), E. viminalis (lo) and EH. oreades (if), is more usually found (Plate xx, fig. 1). The H. Blaxlandi consociation, with #. goniocalyx (0), occurs on the summit of basalt hills which are fully exposed to the west (e.g. summit of hill on left of Text-fig. 3, Part I, p. 489); -here the Pteridium society usually replaces Alsophila. Further reference will be made to the distribution of these consociations in Part III. The tree-fern barrier, and the junction between the Blechnum and Pteridium societies, are correlated with no difference in the structure and composition of the tree-stratum. At the periphery of the basalt caps the volcanic-soil stratum grows very thin, and finally disappears. It may be supposed that the shallow-rooting Alsophilas cease when the basalt soil is no longer deep enough for them to root in. The Hucalyptus trees, however, are not restricted to the basalt in this way, and, indeed, for some distance on to the periphery of the basalt they probably send their roots partly into the underlying sandstone: they appear to be controlled in their distribution by high moisture requirements, which are satisfied not only on the basalt but also on the adjacent sandstone. Thus in one locality the same consocia- tion spreads over the Alsophila and Pteridium societies up to the junction with the Hucalyptus piperita Forest. The association is plainly suited to a moister or richer soil than the typical dry-sandstone Eucalyptus Forests, being confined to the basalt and the adjacent sandstone which is probably enriched by leaching and percolation from the former, and also to the damp soil of the valleys. The dominant trees are of great stature, averaging 150 to 200 feet in height, and with a spread often 40 feet in diameter; but they are never near enough to one another to form a closed canopy, a point well illustrated in Part I, Text-fig. 4 (p. 488). The Stratum-Societies. Some additional observations have to be recorded upon certain of the stratum- societies of the Junction Flora, but for various reasons it is feltsadvisable to with- hold these until after the discussion of the Hucalyptus piperita Forest. THE EUCALYPTUS PIPERITA-H. HAEMASTOMA VAR. MICRANTHA ASSOCIATION. STRUCTURE AND PHYSIOGNOMY. This association is the main expression of Hucalyptus Forest as it occurs on the Hawkesbury Sandstone in New South Wales, and possesses the characteristic structure and physiognomy to which reference has already been made. Although the total number of dominants of the whole association is large, at Mount Wilsor there are only two, namely Hucalyptus piperita and H. haemastoma var. micrantha 150 AN ECOLOGICAL STUDY OF THE FLORA OF MOUNT WILSON, ii, oy soe Co ors aoe ae oe QO Ie @ ©) @ OO) Q, @ © (3) ® 4 @ © (2) ae. @ @) OVAK. @ D Text-fig. 1—Chart of a portion of the Hucalyptus piperita consociation, showing the complex composition of the shrub strata. The tall shrubs are indicated with heavy outlines. The structure of the lower strata is seen to be slightly more open beneath the canopy of the trees. (Scale, 1 in. = 6 ft.) BY A. H. K. PETRIE. 151 These trees occur in consociations, and attain a height of from fifty to one hundred feet according to the degree of exposure and soil-moisture. Hucalyptus Sieberiana and EH. eugenioides are subordinates. The shrub-stratum has a more or less open structure and ranges from four to six feet in height (see Plate xxi, fig. 3), although an occasional individual attains the stature of a small tree. The low-shrub stratum is also of open structure, reference to the chart (Text-fig. 1) indicating how much of the ground is bare; it may be supposed, however, that the habitat is covered to the limit of the com- paratively meagre supply of water and nutrient materials. The shrubs are all of the evergreen scicrophyllous type, but otherwise they tend to show unlike reactions to the environmental conditions, although many species belonging to different genera and even different families may have reacted along the same lines and hence may bear close similarities. There is a general tendency towards reduction in size, the most characteristic feature of xerophily. This is expressed particularly in the chlorophyll-bearing organs, but that the vegetation in question is not an extremely arid type is clearly evinced by the fact that, although the leaves are generally small, in only a few of the species are they minute or even acicular; and, while the majority of the species of Acacia bear phyllodes, A. discolor has pinnate leaves and yet finds a favourable habitat in this association. Quite another line of reaction is shown in a small number of types with comparatively large leaves, such as Banksia serrata and Persoonia salicina, which have broad leaves up to six inches in length; in the majority of such types, however, the leaves are isobilateral, as in the case of Persoonia. The structural adaptations of these sandstone xerophytes have not yet been submitted to detailed study, but it is hoped that such an inquiry will form the subject of a later publication. THE RELATION TO FIRES. Bush-fires are a fundamental factor controlling the development of Hucalyptus Forests in New South Wales, since their occurrence throughout the summer is so frequent; and the greater part of the Eucalyptus Forest of the Mount Wilson plateau has at some period been burnt, to which charred wood and blackened tree- trunks bear abundant evidence. These fires pass through the Forest with considerable ease and rapidity, a fact made possible by the low water-content of the plants and the inflammable essential oils present in the leaves of the myrtaceous and rutaceous types. AS a KEY TO SYMBOLS (Tex-fig. 1). Tree Stratum. Tall-shrub Stratum (contd.). Low-shrub Stratum (contd.). EK, Hucalyptus piperita. Ph, Pomaderris ledifolia. Li, Lissanthe sapida. Tall-shrub Stratum. Ps, Persoonia salicina. Ls, Lomatia silaifolia. Ad, Acacia discolor. Pu, Pultenaea scabra. Mg, Mirbelia grandiflora. Aj, Acacia juniperina Tl, Trachymene linearis. Pp, Phyllota phylicoides. Bm, Banksia marginata. Ts, Telopea speciosissima. Pt, Pteridium aquilinum. Bs, Banksia spinulosa. Low-shrub Stratum. Xa, Xanthorrhoea hastilis. Ce, Choretrum Candollei. As, Acacia linifolia var. X1, Xerotes longifolia. Hd, Hakea dactyloides. Bo, Bossiaea scolopendria. Xp, Xanthosia pilosa. Ia, Isopogon anemonifolius. Cf, Caustis flexuosa. Le, Leptomeria acida. D, Dianella sp. A line through a symbol Lp, Leptospermum stella- Ds, Dampiera stricta. indicates a young plant or tum Gb, Goodenia bellidifolia. seedling. 152 AN ECOLOGICAL STUDY OF THE FLORA OF MOUNT WILSON, ii, consequence of the latter, the fires frequently run up the Hucalyptus trees and burn off their foliage; but it is evident that the low water-content of the leaves is the chief factor accelerating the fires, since they travel mainly through the shrub-stratum, where the percentage of oil-containing types is not high. Moreover, the essential oils themselves, unless in leaves of low moisture-content, do not make burning possible; for fires have not been observed to have passed through the Rain- Forest, although Doryphora contains as much essential oil of an inflammable nature (approximately 1%) as many species of Eucalyptus (Penfold, 1921). The recovery of Eucalyptus trees after fire has been referred to in Part I: if the foliage is burnt, it is renewed by the production of adventitious shoots from the stem (see Plate xxi, fig. 4); if the trunk is burned, new shoots arise from the base: indeed it is only occasionally, except at very high altitudes (e.g. on the Snowy Mountains), that a tree does not recuperate after fire. Similarly many of the shrubs possess subterranean rootstocks from which they send up new shoots after subaerial destruction by fire. Thus after pyric denudation it is possible for regeneration to take place with- out the initiation of a sere: the previous community may reproduce itself without any alteration in structure, every plant reappearing in its previous place. At all events, since the area is a secondary one, and since it is improbable that the habitat factors are changed very much, if succession does take place, the sere is generally short and simple: there is as a rule nothing to prevent the migration and germination of seeds of the types previously existing in the habitat, which usually seize the opportunity of pyric denudation to establish their seedlings. Such aberrations aS appear in the early period of regeneration may be due to differences in readiness of renascence, in aptitude of establishment, or in aggressiveness, of different types. As an example of the first-mentioned factor may be instanced the rapid spread of Pteridium before the rootstocks of the shrubs have time to produce their shoots. In the association under discussion, however, on account of the unfavourableness of the habitat, this type is as a rule not sufficiently aggressive to prevent the developing shrubs from subsequently extinguishing it. Along with. Pteridium in the early stages we have noticed Dianella spp., Xerotes longifolia, Lomatia silaifolia, and Panax sambucifolius, which appear to possess the ability of rapid renascence. Many of the endemic types have also most unique adaptations for recolonisa- tion after fire. In a poor siliceous soil like that produced by the Hawkesbury Sandstone there is keen competition between the different components of the vegetation for the small quantity of available nutriment and moisture. It is, therefore, to be expected that a species will always endeavour to seize any oppor- tunity which may be afforded for gaining an ascendancy over the other members of the community; and it is obvious that, if seeds could be preserved during fire, such an opportunity would be gained, inasmuch as recolonisation would not have to await the comparatively slow process of migration from an unburnt adjacent area. Hence many types, notably most of the Myrtaceae and Proteaceae (especially Hakea), have developed an enormous amount of wood in the fruit, with the result that even if that structure is appreciably charred, the seed is uninjured; others, such as Persoonia, have succulent fruits, but hard seeds within; while hard seeds are found also in Actinotus, Acacia, etc. Thus protected from destruction during the fire, it is advantageous for these seeds to be able to germinate immediately afterwards: it is interesting to observe, therefore, that, while on account of their BY A. H. K. PETRIE. 153 massive structure these fruits often do not open for a number of years, the heat of a fire causes immediate dehiscence; and many seeds appear also to profit by the fire inasmuch as it chars the testa and so allows penetration of moisture for germination. These plants thus seem to have adapted themselves to benefit by fires, since they gain an opportunity of establishing seedlings before the regeneration of the previous components of the community; these seedlings are thereby given a start in competition, and can at all events take the place of any individuals which have not survived, and of such species as possess no means of renascence, thus gaining considerably in competition with the latter. In any case it is unlikely that much would be gained by scattering seeds through a more or less mature community growing in an unfavourable habitat, since all available space permitted for plants by the limiting soil-moisture factor would already be occupied. These plants, therefore, seem to find it best to withhold their seeds until a fire occurs; and fruits of Hakea sometimes remain on the tree for several years without dehiscing, the seeds still retaining their vitality. , The point to be gathered, then, is that the regeneration of Eucalyptus Forests following pyric denudation is not dependent on the comparatively slow process of migration, but is effected mainly by the renascence of the previous components, and also by aggregation, which is a most distinctive feature of this Australian endemic flora. Not only is the greater part of Eucalyptus Forest secondary in nature, but many areas are not even mature. The ultimate composition is no doubt established, but some of the shrubs take a number of years to attain their maximum height, since their growth has the slowness characteristic of xerophytes. The majority of the taller shrubs reach a height of from four to eight feet; but certain of them, notably Banksia serrata, Hakea dactyloides and Persoonia salicina, when given the opportunity, develop into small trees having a height of twelve to twenty feet. THE EUCALYPTUS PIPERITA CONSOCIATION. Distribution. This is the most characteristic community of the sandstone plateau, and forms the greater bulk of the vegetation of the Mount Wilson region (see Plate xx, fig. 2). Hucalyptus haemastoma var. micrantha grows frequently on the more exposed westerly ridges. H. Sieberiana occurs generally throughout the con- sociation as a subordinate, becoming abundant at the apex of exposed bluffs on the western side, sometimes to the exclusion of Hucalyptus piperita, so that it forms a society; such societies are no doubt fragmentary representations of a Hucalyptus Sieberiana consociation which occurs on other parts of the Blue Mountains. Hucalyptus eugenioides was observed in one small area on a westerly headland. Floristic Composition. The following list includes the more typical components of the consociation as it occurs on the sandstone plateau in the immediate neighbourhood of Mount Wilson. Yree Stratum. Hucalyptus haemastoma Sm. Hucalyptus piperita Sm. d var. micrantha DC. lo-lf Hucalyptus Sieberiana F.v.M. f-o* Hucalyptus eugenioides Sieb. lo * In these lists, where two symbols are given, the first represents the more typical frequency. ; 154 AN ECOLOGICAL STUDY OF THE FLORA OF MOUNT WILSON, ii, Floristic Composition—continued. Low-Tree Stratum. Banksia serrata L. Persoonia salicina Pers. Hakea dactyloides Cav. Callicoma serratifolia Andr. Acacia elata Cunn. Tali-Shrub Stratum. Trachymene linearis Spreng. Persoonia salicina Pers. Leptospermum flavescens Sm. Leptospermum stellatum Cav. Banksia spinulosa Sm. Hakea dactyloides Cav. Isopogon anemonifolius R.Br. Telopea speciosissima R.Br. Dillwynia ericifolia Sm. Petrophila pulchella R.Br. Acacia discolor Willd. Acacia longifolia Willd. Pultenaea scabra R.Br. Pomaderris ledifolia Cunn. Persoonia mollis R.Br. Persoonia acerosa Sieb. Choretrum Candollei F.v.M. Persoonia ferruginea Sm. Xanthorrhoea hastilis R.Br. Persoonia pinifolia R.Br. Banksia marginata Cay. Leptomeria acida R.Br. Leucopogon lanceolatus R.Br. Banksia collina R.Br. Hakea gibbosa Cav. Haemodorum planifolium R.Br. Darwinia taxifolia Cunn. Leptospermum stellatum Cav. var. grandiflorum Benth. Cassinia denticulata R.Br. Exocarpus stricta R.Br. Leptospermum scoparium Forst. Oxzylobium trilobatum F.v.M. Callistemon lanceolatus DC. Low-Shrub Stratum. Phyllota phylicoides Benth. Tonidium filiforme F.v.M. Dianella revoluta R.Br. Lissanthe sapida R.Br. Tetratheca ericifolia Sm. Tetratheca thymifolia Sm. Pteridium aquilinum Kuhn. Acacia linifolia Willd. var. Lomatia silaifolia R.Br. Dampiera stricta R.Br. o-r T-O Te la in gullies le in gullies a-f f-o f-o o-a O-a o-la o-la o-¢c o-¢c Low-Shrub Stratum (contd.). Goodenia bellidifolia Sm. Mirbelia grandiflora Ait. Amperea spartioides Brongn. Dracophyllum secundum R.Br. Calochilus campestris R.Br. Cryptostylis longifolia R.Br. Grevillea laurifolia Sieb. Dampiera Brown F.v.M. Dianella longifolia R.Br. Marianthus procumbens Benth. Goodenia heterophylla Sm. Stylidium graminifolium Swartz. Bossiaea heterophylla Vent. Bossiaea scolopendria Sm. Monotoca scoparia R.Br. Actinotus Helianthi Labill. Boronia pinnata Sm. Mitrasacme polymorpha R.Br. Helichrysum rutidolepis DC. societies Xerotes flexifolia R.Br. Xerotes longifolia R.Br. Patersonia sericea R.Br. Stylidium lineare Swartz. Persoonia chamaepitys Cunn. Lycopodium densum Labill. SymphyonemamontanumR.Br. Hibbertia serpyllifolia R.Br. Xanthosia pilosa Rudge Daviesia ulicina Sm. Marsdenia suaveolens R.Br. Lindsaya linearis Swartz. Panax sambucifolius Sieb. Comesperma ericinum DC. Helichrysum lucidum Henck. Orthoceras stricta R.Br. Smilax glyciphylla Sm. Billardiera scandens Sm. Boronia microphylla Sieh. Trachymene Billardieri F.v.M. Goodenia decurrens R.Br. Todea barbara Moore Gastrodia sesamoides R.Br. Creeper. Kennedya rubicunda Vent. Parasites. Loranthus celastroides Sieb. Loranthus pendulus Sieb. Cassytha glabella R.Br. Leptospermum lanigerum Society. eli ted Ve Any deb tab teh te} a) ©-E hemastoma var muccontho c= Eucalyptus pipentar > - with E Sieberiona: ~~~ ~~~ > (@) III} © -Eucaly plus piper iCa - - = seer Text-fig. 2.—Diagrammatic section (not to scale) of westerly headland at Mount Wilson, showing the distribution of the plant communities. <--Fucalyptus strvcla- =~ = ..---> ©- Casuarina ¢---Casuarina fana --- e 4. styla--~?| Beyond the limit of the trees, the shrub communities might be held to have the rank of associations, but since no sharp line can be drawn between those appearing as independent associations and those occurring as a subordinate stratum in the Forest, it is perhaps better to class all the communities as stratum- societies. It must be conceded, however, that the greatest alteration in the nature of the shrubs as they occur in the Eucalyptus piperita consociation takes place coincident with the dwindling away of the trees: within the Forest the change is rather that of the elimination of all but those types specially adapted to the more severe environment, the consequent greater abundance of the remainder, and the invasion of some of the new types from societies beyond the trees. The latter are absolutely distinctive and contain only a few types present in the shrub-stratum of the Eucalyptus piperita consociation. In these societies the majority of the characteristics are exclusive, being confined to exposed headlands of the highest altitudes of the Blue Mountains: such a restricted range points to a considerable degree of specialisation. Petrophila pulchella Society. The change in the composition of the shrub-layer as one passes from the Eucalyptus piperita consociation leads to the formation of a society dominated by Petrophila pulchella. A great many of the former shrub-components of the Forest are here absent, and also certain new types appear as invaders from the com- munities nearer the edge of the cliff. This society ceases coincidently with the tree-zone. It has a low and heath-like character, no doubt induced by the exposure of the environment (see Plate xxii, fig. 7). The Eucalyptus trees are stunted in this zone and are sparingly scattered, so much so that they are inclined to yield the dominance to the shrub-stratum. Since this habitat is differentiated by its increased aridity, consequent upon the extreme degree of exposure to the west, and by the closeness of the rock to the surface of the soil, the composition of the Petrophila society suggests an interesting clue to the components of the Hucalyptus Forest which are most xerophilous in their adaptations. The majority of the shrubs are excluded; but Petrophila pulchella seems well adapted to the habitat, and so has come to dominate the society. Certain other components, such as Hakea dactyloides, Isopogon anemonifolius, Banksia ericifolia, Trachymene linearis, Acacia linifolia var., etc., likewise can tolerate the conditions, but evidently not so well as Petrophila, which has consequently gained a marked ascendancy in the competition among the BY A. H. K. PETRIE. 17 species. Amongst the invaders from the adjacent societies at the apex of the headland may be mentioned Casuarina nana, Hakea pugioniformis and Leptos- permum lanigerum var. macrocarpum. Floristic Composition. Shrub Stratum. Low-Shrub Stratum. Petrophila pulchella R.Br. d Baeckea brevifolia DC. f Hakea pugioniformis Cav. f-c Goodenia bellidifolia Sm. f Hakea dactyloides Cav. f Dampiera stricta R.Br. it Banksia ericifolia L. o-f Hibbertia serpyllifolia R.Br. f Casuarina nana Sieb. (0) Xerotes longifolia R.Br. vr Trachymene linearis Spreng. oO Bossiaea sp. Vr Acacia linifolia Willd., var. to) Leptospermum lanigerum Sm. Xanthorrhoea hastilis R.Br. fo) var. macrocarpum Maiden Pomaderris ledifolia Cunn. la and Betche vr Acacia discolor Willd. 4 Hucalyptus stricta Society. Immediately below the Petrophila society, as the headland begins to slope towards the escarpment, occurs a low, stunted, closed community of Hucalyptus stricta, with an occasional tree of H. haemastoma var. micrantha still present (see Plate xxii, fig. 9). Hucalyptus stricta is modified in its life-form by the early pro- liferation of the stem apex, so that several erect shoots are formed, resulting in a shrubby growth: such trees are locally known as “‘mallees” in Australia. This character is well marked in the individuals occurring in this community, and, together with the nanism resulting from the exposure of the habitat, causes the plants to attain a height of about only four or five feet. The significance of nanism in the habitat is shown by the fact that Hucalyptus stricta in some localities grows to a height of twenty feet. Under the shelter of the canopy of Eucalyptus stricta occur a number of low shrubs, attaining about two feet in height. Floristic Composition. Tall-Shrub Stratum. Low-Shrub Stratum (contd.). Eucalyptus stricta Sieb. d Banksia spinulosa Sm. (9) Hakea dactyloides Cav. f Isopogon anemonifolius R.Br. fo) Persoonia acerosa Sieb. 9 Caustis flexuosa R.Br. (o) Persoonia salicina Pers. ip Petrophila pulchella R.Br. (0) Leptospermum stellatum Cay. r Pomaderris ledifolia Cunn. (0) Hibbertia serpyllifolia R.Br. te) Low-Shrub Stratum. Conospermum ericifolium Sm. 0 Phyllota phylicoides Benth. f Symphyonema montanum R.Br. r Trachymene linearis Spreng. f Xerotes longifolia R.Br. r Casuarina nana Society. Below the Hucalyptus stricta society occurs a dwarf-shrub heath composed chiefly of Casuarina nana, a stunted, heath-like bush about two feet high. As in a number of these societies, the dominant is not the tallest plant occurring in the community: here the monotony is relieved by an occasional plant of Casuarina distyla, a shrub about five feet in height. A few smaller shrubs also stand above the general level, especially Petrophila, which is frequent, and which, by its rich green, stands out conspicuously against the brown Casuarina. Eucalyptus is entirely absent from this wind-swept habitat. 158 AN ECOLOGICAL STUDY OF THE FLORA OF MOUNT WILSON, ii, Floristic Composition. Tall-Shrub Stratum. Low-Shrub Stratum (contd.). Casuarina distyla Vent. r-o Xerotes longifolia R.Br. ie Hakea pugioniformis Cav. Te Shrub-Stratum. Phyllanthus thymoides Sieb. r Petrophila pulchella R.Br. f Trachymene linearis Spreng. (0) Ground Stratum. Leptospermum stellatum Cav. oO Goodenia bellidifolia Sm. a Hakea dactyloides Cav. (0) Dampiera stricta R.Br. (o) Sowerbaea juncea Sm. (0) Low-Shrub Stratum. Hibbertia serpyllifoia R.Br. r-f Casuarina nana Sieb. d Patersonia sericea R.Br. r Tsopogon anemonifolius R.Br. r Hakea pugioniformis Society. This is another dwarf-shrub heath of a structure similar to that of the Casuarina nana society, and occupying erosion channels intersecting the habitat of that community (see Plate xxii, fig. 8). It is contrasted by its deep-green colour and the terete, pungent-pointed leaves at right angles to the stem, and the angular branching. A few taller types interrupt the closed community of the dominant in places. Floristic Composition. Tall-Shrub Stratum. Shrub Stratum (contd.). Hucalyptus stricta Sieb. r Hakea dactyloides Cav. fo) Leptospermum stellatum Cay. r Cassinia arcuata R.Br. r Low-Shrub Stratum. Shrub Stratum. Casuarina nana Sieb. r Hakea pugioniformis Cav. d Ground Stratum. Banksia ericifolia lL. (0) Goodenia bellidifolia Sm. a Petrophila pulchella R.Br. co) Hibbertia serpyllifolia R.Br. r Casuarina distyla Society. Finally, at the very apex of the headland, on a flat area of sandstone rock which has only a shallow layer of soil occurring in pockets and crevices, is found a more open society dominated by Casuarina distyla. The majority of the shrubs in this society are only about three feet high, as is the case with the dominants of the two societies just described; Casuarina distyla, however, attains a height of about five feet, and thus lends the community a distinctive appearance, a curious fact considering that this appears to be the most unfavourable habitat in the area under consideration. Floristic Composition. Tall-Shrub Stratum. Shrub Stratum—continued. Casuarina distyla Vent. d Lasiopetalum ferrugineum Hucalyptus stricta Sieb. to) Sm. var. cordatum Benth. r Hakea gibbosa Cav. r Low-Shrub Stratum. Shrub Stratum. Xanthorrhoea hastilis R.Br. (0) Petrophila pulchella R.Br. Leptospermum lanigerum Sm. Banksia ericifolia L. var. macrocarpum Maiden Cassinia arcuata R.Br. and Betche (0) Hakea dactyloides Cav. Callitris Muelleri Benth. and Brachyloma daphnoides Benth. Hook. seedlings Oo Leptospermum stellatum Cav. ‘Casuarina nana Sieb. (0) Isopogon anemonifolius R.Br. Acacia suaveolens Willd. (0) Baiksia spinulosa Sm. Dillwynia floribunda Sm. (0) Banksia marginata Cav. Trachymene Billardieri F.v.M. o Hakea pugioniformis Cav. Persoonia acerosa Sieb. vr ©0000 0 0 hHmMhO Trachymene linearis Spreng. BY A. H. K. PETRIE. 159 Floristic Composition—continued. Ground Stratum. Hibbertia serpyllifolia R.Br. f Chloanthes stoechadis R.Br. r Dampiera stricta R.Br. (0) Boronia pinata Sm. r Symphyonema montanum Pimelia linifolia Sm. ip R.Br. fo) Xerotes longifolia R.Br. r Baeckea brevifolia Society. This is a community of very small types, averaging about a foot in height, and occupying areas on the rocky headland where moisture collects in the soil as a result of drainage from the higher parts of the ridge. It is difficult to realise why this area is not occupied by more of the types belonging to the other communities ‘described, but indeed the sharp lines of demarcation between the societies on this headland have not yet been interpreted apart from the fact that zonation results from increasing exposure and shallowness of soil. The habitat of this society appears to be a favourable one for orchids, three species being recorded in the following list, although others might probably be found in the proper season. Floristic Composition. Low-Shrub Stratum. Ground Stratum. Baeckea brevifolia DC. d Eriochilus autumnalis R.Br. a Isopogon anemonifolius R.Br. . Prasophyllum densum Fitzg. a seedlings to) Chiloglottis diphylla R.Br. f Cassinia arcuata R.Br. fo) Selaginella uliginosa Spreng. (0) Hriostemon obovalis Cunn. fo) Mitrasacme polymorpha R.Br. o Dillwynia floribunda Sm. r Drosera spathulata Labill. Oo Leptospermum lanigerum Sm. Drosera binata Labill. (o) var. macrocarpum Maiden and Betche r Brachyloma daphnoides Benth. r THE STRATUM-SOCIETIES OF THE JUNCTION FLORA. Some additional observations have to be recorded upon certain of the stratum- societies of the Junction Flora. On the whole the components of these societies are less markedly sclerophyllous than those of the Hucalyptus piperita- E. haemastoma var. micrantha association of the dry sandstone, which are more extreme xerophytes; this shows how the H. goniocalyz-H. Blaxlandi association is an unusually mesophilous Eucalyptus Forest, a fact manifested also by the conspicuous pteridophyte element. The societies are being dealt with separately here because certain of them are found in the Hucalyptus piperita consociation as well as in the EH. goniocalyz- EH. Blaxlandi association. The term society is being used in every case rather than socies, for, although it is improbable that all the communities are climaxes, the data have not yet been obtained for a clear knowledge of their stability and stitus. Pteridium aquilinum Society. We have already seen how the Eucalyptus goniocalyz-E. Blaxlandi association occurs as a rule at the periphery of the basalt except on southerly slopes, usually soon giving place, however, to the Hucalyptus piperita consociation as it passes on to the edge of the sandstone. The abutting margins of both these communities are occupied by the Pteridium stratum-society, as was made clear earlier. A close examination, however, has revealed a distinct difference in the composition of the 160 AN ECOLOGICAL STUDY OF THE FLORA OF MOUNT WILSON, Iii, Pteridium aquilinum. Amperea spartioides. Acacia longifolia. Dianella sp. Dampiera stricta. Hucalyptus piperita. Leucopogon lanceolatus. Lomatia silaifolia. Poranthera microphylla. Persoenia salicina. Panax sambucifolius. Solanum xsanthocarpum. Tetratheca thymifolia. Trachymene linearis. Telopea speciosissima. Xerotes Brownii. @@Q@G20@@ 2 G2 OQ} o© > Xerotes flexifolia. {betonicifoyja- Text-fig. 3.—Belt-transect in the Pteridiwm society of the Hucalyptus piperita consociation near the basalt. Regeneration of typical sandstone shrubs is in evidence. (Scale 1 in. = 4 ft.) BY A. H. K. PETRIE. 161 subordinates of this society in the two consociations, as will be evident from the appended list and chart. This observation is of great significance, since it constitutes an important step towards the solution of the problem of the status of the Pteridium society which was discussed at some length in Part I. The Eucalyptus goniocalyz-E. Blaxclandi association is confined to the damper soil; and it is natural, therefore, that the Pteridium society therein should include comparatively mesophilous types. It will be remembered that in this association, beneath the Alsophila society, occurs a Blechnum stratum-society, which merges into the Pteridium society near the outer boundary of the association. These two societies, excluding the Pteridium society in the Eucalyptus piperita consociation, bear certain resemblances in composition; and there is every reason to believe, not only that the two societies are similar in status, being differentiated owing to different degrees of environmental favourableness, but also that both are climactic. A critical examination of the Belt-Transect in Part I (Text-fig. 5, p. 491), which includes both the Blechnum and the Pteridium societies, shows that there is no suggestion in either of development to a community of higher status: there is not even a seedling of any type to which Blechnum or Pteridium could yield dominance. This chart brings out also the general similarity of structure and composition between these two societies. Reference has been made to the fact that the Pteridium society shows unmistakable signs of having been burnt out at no remote period; but fires have been observed to have passed through the Blechnum society over considerable areas of the Hucalyptus-Alsophila Forest; and just as the latter unquestionably regenerates without change in structure, so also no doubt has the Pteridium society regenerated, and will regenerate if burnt again. Let us turn now to a study of the Pteridium society in the more typical sand- stone habitat of the Eucalyptus piperita consociation. Here a remarkable change has taken place in the structure of the society, which is the culmination of a slow transition as we pass from the one consociation to the other: the more mesophilous subordinates, such as Geranium pilosum, Tylophora barbata and Stellaria pungens, have been replaced by xerophytes, notably Lomatia silaifolia, Tetratheca spp., Trachymene linearis, and even an occasional Persoonia salicina and Acacia longifolia. These features are clearly evinced in the accompanying belt-transect (Text-fig. 3), which forms a most instructive comparison with that to which reference was previously made. The society in the Eucalyptus piperita consocia- tion is seen to include quite a number of species of shrubs characteristic of the consociation in its typical expression; and many of these shrubs are found to be in an early stage of renascence from old subterranean rootstocks or else to be in the seedling state. The importance of this comparison cannot be overestimated, for it has now given the clue to the interpretation of the Pteridium society. Hucalyptus piperita has been able to establish itself in this habitat close to the basalt: it will be shown in Part III that this species can extend even on to the basalt if freed from competition. The typical shrubs associated with it, however, find greater difficulty in growing here, perhaps because their shallower roots are more exposed to the basalt soil; and, moreover, they are subjected to a severe struggle with Pteridium, which finds the habitat not unsuited to its requirements, and which is able to migrate from the Eucalyptus goniocalyz-E. Blazlandi association by the slow but effective agency of its subterranean rhizome. Now when the Forest is burnt out, Pteridium spreads rapidly, and forms the first stage of the subsere. If, as seems to be the case, this is not an ideal habitat M 162 AN BRCOLOGICAL STUDY OF THE FLORA OF MOUNT WILSON, ii, for the sandstone shrubs, renascence and establishment of seedlings will no doubt, in any case, be delayed; and when these processes have to be effected in the face of heavy competition with Pteridium, it is easily seen that they will not be performed rapidly. So slow are they that the period between the last two fires in any spot has apparently never been long enough for the next stage of the sere definitely to establish itself, and Pteridium is at present the dominant throughout the whole of the Junction Flora. Yet renascence and migration are slowly but surely being accomplished, as is seen in the accompanying transect; although we are unable to state to what this may lead, or what may be the fate of the Pteridium. But we are almost safe in concluding that the Pteridiuwm community here is neither a climax nor a sub-climax, and perhaps for the sake of distinction we might term it a Pteridiwm socies; and we have undoubtedly made a consider- able advance in solving the problem of this somewhat enigmatic tract of vegetation. Floristic Composition. These lists represent typical examples of the two extremes of the Pteridium society, but it must be understood that there is an extensive ecotone region in which migration of sandstone types is very sparse. E. goniocalyx-B. Blaxlandi., BE. piperita. Shrub Stratum, Acacia penninervis Sieb. lo lo Acacia longifolia Willd. — Astrotricha floccosa DC. — r-oO Daviesia ulicina Sm, oO fo} Hucalyptus spp. saplings lf If Leucopogon lanceolatus R.Br. — oO Lomatia longifolia R.Br. — r Panax sambucifolius Sieb. — fo} Persoonia salicina Pers. —- fo) Telopea speciosissima R.Br. — (0) Trachymene linearis Spreng. —— f-c Fern Stratum. Blechnwm cartilagineum Swartz. a Blechnum discolor Keys. la Davallia dubia R.Br. fe) Dianella longifolia R.Br. f Dianella revoluta R.Br. ? Gleichenia dichotoma Hook. oO Lomatia silaifolia R.Br. — Phyllanthus thymoides Sieb. — Polystichum aculeatum Schott. le = Pteridium aquilinum Kuhn. d Xerotes longifolia R.Br. (e) sround Stratum, Acaena sanguisorba Vahl. fo) — Amperea spartioides Brongn. -o f Brachycome sp. fe) —_— Cardamine hirsuta L. r-oO — Chiloglottis formicifera Fitze. — vr Clematis aristata R.Br. seedlings r-o r Dampiera stricta R.Br. — fs Dipodiwn punctatum R.Br. = Tr Doodia aspera R.Br. f Galium umbrosum Sol. o-la Geranium pilosum Sol. a Hydrocotyle asiatica L. o-c BY A. H. K. PETRIE. 163 Floristic Composition—continued. E. goniocalyz-EH. Blazlandi. EF. piperita. Lobelia dentata Cav. ip Lycopodium densum Uabill. vr Mirbelia grandiflora Ait. vr Oxalis corniculata L. fo) Poranthera microphylla Brongn. r Solanum santhocarpum Schrad. — Sonchus sp. = c iF 79 | je} Stackhousia viminea Sm. Stellaria pungens Brongn. Stellaria flaccida Hook. Tetratheca thymifolia Sm. Tetratheca ericifolia Sm. [fe slieeataas hh 1 © Trifolium sp. in Tylophora barbata R.Br. fe Viola betonicifolia Sm. fa) Wahlenbergia gracilis DC. —— Xerotes flexifolia R.Br. — Unidentified grasses f t ue ails sence hails ‘Climbers. Hardenbergia monophylla Benth. —- Kennedya rubicunda Vent. — Smilaz australis R.Br. fo) = ho Acacia penninervis Society. Reference has already been made to the dense societies of this plant occurring in the Junction Flora. It grows as a tall shrub or sapling in closed societies, such as are characteristic of succession by aggregation following fire. Eventually the plant reaches the height of 30 feet, and in some places groups of such trees are to be seen. As an interpretation of the presence of these societies we may suppose that an occasional plant of Acacia penninervis, originally present in the association, might have preserved its seeds during fire; subsequent colonisation by aggregation from these, unhindered by competition with other species, would result in the local displacement of Pieridium and the formation of a dense and almost pure society of the new colonist. The plants, moreover, seem to multiply by the production of shoots from extensive lateral roots, and it may be that in some instances the societies have arisen in this way rather than by aggregation. Pultenaea flexilis Society. Although practically the whole of the Junction Flora bears the appearance of having been burnt within the last few years, in one place on a gentle south-west slope, a portion of the basalt junction was found inclosed in private land which has evidently not been burnt for a long period. The structure and composition of the vegetation here is of considerable interest, since it shows what the Junction Flora is capable of developing into when unburnt. The community occurs in the Eucalyptus viminalis consociation, with Pultenaea flezilis subdominant at the sandstone edge, becoming less abundant nearer the tree-ferns. This tall-shrub society extends from the foot of the Alsophila society a considerable distance down on to the sandstone; outside the fence flanking a road which has evidently acted as a fire-break, the characteristic Pteridium society occurs, as a result of regenera- tion following a recent fire (see Text-fig. 4). 164 AN ECOLOGICAL STUDY OF THE FLORA OF MOUNT WILSON, ii, It cannot be said definitely that the Pteridiwm society would develop to that dominated by Pultenaea flexilis; but the comparison of the vegetation on each side of the road shows that it may possibly do so in some parts, and the close resemblance in floristic composition adds support to the view. Cattle have heavily trampled the ground between the shrubs, which has probably resulted in modifications in the fern stratum. SW é----- Eucalyptus viminalis W PP. < Eucalyptus piperita 7 é Consociation---> REISS ERT INS SI) <-Pultenea flexilis Souety --- -? Y) ose In THE E. Viminatis CONSOCLATION — 7 ENS PCeeidium Society) “S355 “- Qd-e Blaxlandi Qd-e viminalig 77777 = Basalt Sol €-Preridium SOCtes | Text-fig. 4.—Diagrammatic section (not to scale) showing distribution of Pultenaea flexilis society. Floristic Composition. Tree Stratum. Hucalyptus viminalis Labill. d Hucalyptus Blaxlandi J.H.M. and R.H.C. a Acacia melanoxylon R.Br. fo) Eucalyptus goniocalyx F.v.M. r Tall-Shrub Stratum. Pultenaea flexilis Sm. sd-c Persoonia mollis R.Br. f (a near tree-ferns) Daviesia ulicina Sm. if Acacia longifolia. Willd. Citriobatus multiflorus Cunn. f-o (0) Acacia elata Cunn. young plants r 1© r (a at tree-fern boundary) Leptomeria acida R.Br. Acacia penninervis Sieb. Low-Shrub Stratum. Pteridium aquilinum Kuhn. Helichrysum riutidolepis DC. Xerotes longifolia R.Br. Lomatia silaifolia R.Br. Blechnum cartilagineum Swartz. Davallia dubia R.Br. Polystichum aculeatum Schott. Blechnum discolor Keys. Leucopogon lanceolatus R.Br. Ground Stratum. Similar to that given in Part JI, p. 492 Dawsonia sp. r Acianthus fornicatus R.Br. Tr Creepers. Geitonoplesium cymosum Cunn. Billardiera longiflora Labill. Clematis aristata R.Br. Clematis glycinoides DC.° Eustrephus Brown F.v.M. at tree-fern boundary Sitges Qo near tree-ferns near tree-ferns near tree-ferns -O BOM OmO ROMO “40000 BY A. H. K. PETRIE. 165 Pultenaea flexilis Society on Northern and Western Slopes. The Eucalyptus. goniocalyx-H. Blaxlandi association on the edge of the basalt on the westerly slopes, is generally occupied by the Pteridium society which was described in Part I (p. 493); societies of Pultenaea flexilis occur in addition, how- ever, aS was also mentioned therein. On northerly slopes the typical Pteridium society supplants the Alsophila society in the region of sandstone intermixed with basalt talus, where also are found societies of Acacia penninervis. A little further down the slope the Hucalyptus goniocalyx consociation, which includes the above societies, suddenly gives place to the H. piperita consociation. This community possesses a modification in floristic composition characteristic of its occurrence on sandstone slopes below basalt caps, as will be described in Part III. Along the junction of these two consociations is frequently found a narrow belt of the Pultenaea flexilis society, similar in composition to the societies of the western slopes. The composition of these societies differs somewhat from that of the society on the southwest slope, containing more xerophytic types on account of the greater exposure of the habitat. The oceurrence of these societies of Pultenaea flexilis is not easy of explana- tion. It may be that lack of further disturbance by fire will cause the society to develop in certain areas as a further stage in the sere initiated by the Pteridium society; from the conclusions to which we have come regarding the latter com- munity, however, it seems unlikely that this could be at all general. The societies on the northern and western slopes may have arisen by aggregation in the manner suggested for Acacia penninervis; but the structure of the society on the southwestern slope shows that this, at any rate, did not originate in this way. Floristic Composition. Tall-Shrub Stratum. Low-Shrub Stratum (contd.). Pultenaea flexilis Sm. d Pteridium aquilinum Kuhn. c Callicoma serratifolia Andr. la Xerotes longifolia R.Br. fy Telopea speciosissima R.Br. f Trachymene linearis Spreng. f Persoonia salicina Pers. f Acacia juniperina Willd. fo) Banksia spinulosa Sm. (a) Dianella sp. fo) Choretrum Candollei F.v.M. fo) Xerotes flexifolia R.Br. fo} Persoonia mollis R.Br. r Amperea spartioides Brongn. fo) Davallia dubia R.Br. fo) Low-Shrub Stratum. Lomatia silaifolia R.Br. a-c Climber. Daviesia ulicina Sm. c Bustrephus Browni F.v.M. r SUMMARY. é 1. The paper comprises an account of the Eucalyptus Forests of the plateau at Mount Wilson. 2. A general account of the main features of Hucalyptus Forest is given, and the classification of those occurring at Mount Wilson is placed upon a more permanent basis. 3. Two associations are described, viz., the Hucalyptus goniocalyz-E. Blarlandi association, which includes what were previously called tentatively the Hucalyptus- Alsophila association and the Hucalyptus-Pteridium association (in part); and the Eucalyptus piperita-H. haemastoma var. micrantha association, which occupies the dry sandstone plateau. The dominants of both these associations occur always in consociations. 166 AN ECOLOGICAL STUDY OF THE FLORA OF MOUNT WILSON, ii. 4. The nomenclature and the distribution of the consociations of the Eucalyptus goniocalyx-E. Blaxlandi association are discussed; and the structure and physiognomy of the Lucalyptus piperita-H. haemastoma var. micrantha associa- tion is described, special consideration being given to its relation to fires. 5. A description follows of a number of stratum-societies occurring on an exposed westerly headland in the Hucalyptus haemastoma var. micrantha consociation. 6. In conclusion, some further observations are recorded on the stratum- societies of the Junction Flora, with a special discussion of the status of the Pteridium society. References. Broueu, P., McLuckiz, J.. and Perrin, A. H. K., 1924.—An Ecological Study of the Vegetation of Mount Wilson, Part I: The Vegetation of the Basalt. Proc. Linn. Soc. N.S.W. xlix. CLEMENTS, F. E., 1916.—Plant Succession. Carneg. Inst. Wash. Publ. 242. PENFOLD, A. R., 1921—The Essential Oil of the leaves of Doryphora sassafras. Journ. Roy. Soc. N.S.W. lv. TANSLEY, A. G., 1920.—The Classification of Vegetation and the Concept of Development. Journ. Ecol. viii. WARMING, H., 1909.—CG cology of Plants. Oxford. EXPLANATION OF PLATES XX—XXII. Plate 3.0.8 1.—The Hucalyptus goniocalyx consociation, showing the dominant in the centre and a tall Alsophila on the left. 2.—A westerly slope of the sandstone plateau, clothed with the Hucalyptus piperita consociation. Plate xxi. 3.—A portion of the Hucalyptus piperita consociation, showing profuse development of shrubs. Pultenaea scabra is in the left foreground. 4.—The Pteridiwm society in the Hucalyptus piperita consociation near the basalt junction. The trees are seen to be recovering from fire by the production of adventitious shoots on the stems. 5.—A view in the Hucalyptus viminalis consociation. The dominant can be distin- guished by its white stem, the black trunks being those of EH. Blaxlandi. 6.—A view on the exposed westerly headland described in the text. The low trees in the background are Hucalyptus haemastoma var. micrantha, the taller trees being E. Sieberiana. The H. stricta society is immediately in front of these trees, while the middle distance and foreground are occupied by the Casuarina nana society. In the immediate foreground is a plant of Leptospermum lanigeruwm var. macrocarpum. Plate xxii. 7.—The Petrophila pulchella society in the Eucalyptus haemastoma var. micrantha consociation. 8.—Another view on the exposed westerly headland, showing the Casuarina nana society in the foreground, and the Hakea pugioniformis society in an erosion channel in the background. ; 9.—The Eucalyptus stricta society. A dead tree of E. haemastoma var. micrantha is seen in the middle distance. BURROWING HABITS OF ORNITHORHYNCHUS. By Harry BuRRELL, C.M.Z.S. [Read 27th May, 1925.] The only published observations of the burrowing of Ornithorhynchus are those of the French naturalist-voyager, Jules Verreaux (Observations sur l’Ornithorhynque, Rev. Zool. xi, 1848, 127), who spent some time in Tasmania, and claimed to have studied the habits of the animal closely. Unfortunately, Verreaux’s account contains so many statements which are not borne out by my subsequent observations that I cannot but regard the whole of it with suspicion. The interest attaching to precise observations of the actual burrowing method is considerable, the chief point of interest being to determine the relative extent to which the highly sensitive muzzle and the powerful claws of the pes are used in the process. The present observations were made during March of this year upon a single female animal which I had in captivity at my residence in Kensing- ton. The creature was placed in a specially contrived enclosure, the essential features of which were a plate glass front and a wooden back, which were four inches apart at the bottom, widening to six inches at the top, filled with sifted soil, which was put in a bucketful at a time, each bucket of soil being followed by a bucket of water, until the frame was full, when the whole was tamped and thoroughly flooded with water. The object of the taper was to cause the soil to wedge, and so prevent its collapsing on the burrowing animal. Water was used to consolidate the soil to something like the consistency of the banks in which the platypus burrows naturally in the wild state. The enclosure thus prepared was allowed to stand for twenty-four hours, and the animal was then introduced at one side, where a portion of the earth was removed to make room for it, at 2 p.m. No attempt was made to burrow until about 5.30 p.m., but I am not certain whether this time bears any relation to what normally happens. It may be, however, that the animal normally commences its burrowing in the late afternoon. The observa- tions which follow were made during a period of about an hour, by means of the plate glass front, and during this time I kept myself concealed as far as possible from the animal, so as to leave her undisturbed. The Platypus, in order to obtain the greatest purchase before commencing to burrow, tucks the tapering end of its pliable tail between its hind legs, and, simultaneously, hooks its out-turned hind claws into the earth at either side. While in this crouching attitude, with stiffened top lip and splayed fore-claws, it proceeds smartly to break away the earth, selecting the softest spot available. After burrowing for several inches the animal rests awhile; then it energetically contorts its neck and body systematically, so as to tamp the freshly loosened earth tightly into the otherwise over-large hollow. While stationary, the creature occasionally beats the surrounding walls with its trowel-like tail. But whether that action is really intended as an auxiliary to 168 BURROWING HABITS OF ORNITHORHYNCHUS, the tamping process, or is simply due to the strenuous exertion of the entire muscular system, or both, I do not profess to know. Nevertheless, it certainly does not distort or tend to break away the true design of a finished burrow by its spasmodic action. In fact, according to the structure and shape of the tail, which, in cross-section, is a miniature replica of that of the tunnel, it could not possibly do so by normal means. Therefore, I suggest that, designedly or other- wise, the tail must be regarded as an actual modelling tool. After a considerable spell, accompanied by laboured breathing and an occa- sional gulp in the throat, the operator again deliberately shovels the earth over its head with the end of its sensitive snout. In this way it creates a crude cavity in which to loll its head to one side, while, with the neck fore-shortened, it reaches to the utmost extent of its web-palmed paw, and scratches a hemispherical hollow to one side of the tunnel, contorting its shoulders the while. The energetic digger places its head in the recess so made, and, without any hesitation, performs a similar operation on the opposite side with the other splayed paw. Then, to dislodge the partition separating the two hemispherical recesses which is as yet untouched by the laterally working paws, it probes its muzzle vigorously into that section, thereby completing the circle, and furthermore, creating, as far as the out-stretched neck will permit, another “loll-hole’”’ to one side with its muzzle, in readiness for its head in the shift to follow. Meanwhile, the well-worked earth trickles around the wriggling creature’s body, principally about the powerful shoulders, whence a portion eventually filters down as far as the hips. While the fore-paws and rooting muzzle are working at high tension, the hind legs are, alternately, keeping the specialized fore-parts well up to their work by clinging tenaciously to the solid earth. In their struggle to do this, the hind claws actually aid the process of excavation by cutting the lateral angles which complete the arched design of the burrow. When the soil so worked has been reduced to tamping consistency, in order to distribute the load, the creature backs down the tunnel to the section previously completed. In so doing, it contorts its entire body, and if hard pressed, spirally so, until the superfluous soil adheres firmly to the hitherto somewhat irregular, roughly cut arch and runway. Another well-earned rest follows, after which the Platypus again forges ahead, to repeat the burrowing manoeuvrings in their every particular, section after section, until the subway winding from porch to terminal cavity becomes a perfectly-modelled, accomplished fact. When the animal is confronted with temporary obstacles, such as impoverished, caked earth between roots of trees, etc., in order to avoid retracing its steps to branch off in another direction it will, miner-like, follow the line, not of least resistance, but of favourable soil, even though it be compelled to work on its back to do so. The reason it occasionally adopts this upside-down attitude, or any other position, is so that it may dislodge the hitherto unforeseen obstruction by the most serviceable and powerful means, its versatile paws. Such actions I have critically observed while the creature was manoeuvring in a tight position—this even in the absence of greater obstructions than are occasionally met with, and usually overcome, when a pregnant Platypus is riveting her whole attention on river bank burrowing. Although it is customary for Ornithorhynchus to burrow in a normal position, lying on its back or on either side does not impede its progress in the least. In fact, it seems, at times, to prefer to work in spiral fashion, both while excavat- BY HARRY BURRELL. 169 ing and tamping its tunnel; but eventually it reverts to the normal attitude to model, accurately, its handicraft. The backward and forward progression of a burrowing Platypus while engaged in tamping is accomplished in separate sections of, approximately, six inches at a time—similar to the sections of pug tamped into position by the tail, prior to the brooding session. But it is not always accom- plished in the aforesaid two “mobile” movements. Sometimes, according to the consistency of the soil, the creature will, again and again, repeat the process before relinquishing its arduous duties. Platypus work with a system, entailing detail in every particular. The reversible action of the animal merits some remarks, especially the back- ward progression. Students of anatomy will have observed, no doubt, that the hind feet of a Platypus have a tendency to turn outwards from the flanks, and that the grooved, curved claws continue in that symmetrical trend towards the tail. This is as it should be for the purpose of gripping and scratching back surplus soil while the creature is tunnelling ahead. It is also as it should be when the creature reverses. Especially is this so when the animal is burdened with a packing of adhesive pug, and, incidentally, has the auxiliary agency of the front paws practically out of action. Admittedly, the muscular movements of the body at that time afford great assistance. Nevertheless, the hind quarters and cumber- some tail would become useless, if not a burden to the Platypus, during one of its essential functions but for the natural provision of a “two-way” action of the hind limbs. Preparatory to pulling backwards, the trend of the hind feet con- tinues until the sets of claws are facing one another beneath the tail. This stance enables the contortionist to take a firm grip of the flattened earth-floor, so that it can pull the body backwards step by step, while the fore-paws, when freed, do the shoving. Just as the front limbs act alternately in advancing, so do the hind limbs advance in the reverse, and vice versa. It is rather amusing to witness this act, for, at the outset, the fore-parts are usually obliterated with pug, and the tail, which in contour and elevation somewhat resembles the head, puts one at a loss to guess whether the creature is really coming or going. Apart from the claws, the hind foot of Ornithorhynchus, especially that of a male, is just as versatile in its actions as that of a chimpanzee, even to the power of deliberately gripping. And this apart from the versatility of the combined actions of the entire limb, which during a backward march resembles, in action and general appearance, the fore-limb of a grizzly bear, or, better still, that of a long-clawed sioth. The transformation of the pes to manus, and vice versa, is rather deceptive, but readily acquired. Is this not a reptilian trait? What other mammal in the world is able thus to interchange the functions of its hind and front legs to such advantage? It may be characteristic of moles, etc., but it is certainly a remark- able transformation. So much for the hind limb at present. Now I wish to record further data in connection with the shovelling snout of Ornithorhynchus, including the sensitive upper lip. At the outset of this paper, I unhesitatingly referred to the latter functioning organ as a stiffened upper lip, but not without reason and due consideration. When one handles the foremost part of that lip, it would appear to be almost as limp as its counter-parts. But when the animal is putting it to practical use, viz., rooting into the earth, the position is different. Quite apart from the resiliency of its nature, or whether the creature has the power of temporarily stiffening that N 170 BURROWING HABITS OF ORNITHORHYNCHUS. section of its lip at will or not, the solid earth into which it is thrust prevents the lip from falling. At the same time, the lift of the snout arrests any other diver- gence until the bony prong of the bill is sufficiently embedded in the grooved earth to lever or root the load away. The groove referred to has previously been created, or “lipped in,” by a very vigorous oscillation of the animal’s head. During the excavation of a complete burrow of any length, including all cavities, no superfluous earth whatever is ejected. Even the slight disturbance at the initial opening is tramped beyond all recognition as the animal continues to tunnel. Given favourable conditions, a Platypus can excavate a cavity, approximately 5 &< 5 X 6 inches, in five minutes. It can tamp loosened earth, completely, into a six inch section of a tunnel in fifteen minutes. This makes twenty minutes in all to a shift. Therefore, if the tunneller continued unceasingly in this manner, in sixteen hours it would have completed a tunnel twenty-four feet in length. This is the average length of a breeding burrow. The normal breathing of a Platypus, while asleep, I found to be fifteen respira- tions to the minute. That of the same healthy specimen, while tunnelling, was thirty to the minute, varied by conspicuous spasmodic gulps in the throat at approximate intervals of twenty-five seconds. The facial furrows (embracing the orifices of the eyes and ears) are closed throughout the entire proceedings, as is the case when the Platypus is under water. ATE I. - Proc. Linn. Soc. N.S.W., 1925. thus Murrayi F.v.M. et Tate. Proc. Linn. Soc. N.S.W., 1925. PLATE If. 83. Loranthus Mitchellianus (Hook.) Blakely. PLATE It. Proc. Linn. Soc. N.S.W., 1925. Se ieee ee) tag Loranthus Casuarinae Miq. 34. ee Siw ie i Proc. Linn. Soc. N.S.W., 1925. PLATE Ivy. 35. Loranthus Exocarpi Behr. oh ae A 1 , 1 i | 1 Ty ay ‘ Rees ie t 1 i ‘Lh : 0) 4 ‘ . : ; ; a mr J : i ! Pay. r i af iS a Y #i\ 2 ’ uh i } i} i i 1 ‘ i} © t een: A 1 7 i * 1 = ’ on ee 0 t OT Ms ¥ ; au i { : z i ut : 7 ‘ i Proc. Linn. Soc. N.S.W., 1925. PLATE VI. 36. Loranthus dictyophlebus F.v.M. Proc. Linn. Soc. N.S.W., 1925. parm seni o es ion SA x 37. Loranthus acacioides A. Cunn. PLATE vu. Proc. Linn. Soc. N.S.W., 1925. PLATE VIII. 38. Loranthus homoplasticus, n. sp. PLATE Ix. N.S.W., 1925. Proc. Linn. Soc. Loranthus vitellinus FK.v.M. 39. Proc. Linn. Soc. N.S.W., 1925. PLATE X. 39a. A. Loranthus vitellinus var. glabrescens, n. var. B. Adventitious roots and mode of attachment of L. vitellinus. IPED) Saul, Proc. Linn. Soc. N.S.W., 1925. Loranthus alyxifolius F.v.M. 41, PLATE XII. To t CcaRENceTowNn Proc. Linn. Soc. N.S.W., 1925. INTUTE LY ie GC for most part obscured by soul _ (Far wm BAaOSA A al Surface Drift and Talus Paterson Toscaniteand A A|/A ABKBDAAAA ted L A A AE op ® Associated Lavas A EGG wi - a some Tuffs a a4 SS SS LS ———— Zone of pe ae Erratic Brecciation etc. B Decomposed 4 To Maitiano Porpuyry Point “7 Wz ‘BURNBRAE” a Lp we iF he MAP OF THE VILLAGE OF SEAHAM Showing Che Outcrors of the Mai Gracia Beos of the Kurruns Series OOAFGCA Tuff be Is Va rv ; Ss SESS NOTES 4 On most of the geological horizons, rock-names have laced, but those without such names see been P interpreted by means of the following legen - Mudstones, sometimes banded ty tine calcareous sediment A (iLL) aFGC} BA Al i IN IN A Shales a US AX x Tuff etc.* A IN E xceedi Sa [4 44 glacial tn origin Toffs ete. Le Te | Hosea ate TRE Uae See ee 2 FG.C. stands - for Fluvio- glacial Cong lomoete stands for Tillite ( a Thee ¢ aA A SS Deconeesed CE o Son SS ee _L N o 0° , ai “os o soo S36 SCALE ——— G.0.0 dele. | BRANDON“ Conglomerate, probably Permian in Age , F.G.C. 2 ak AN XK AW a o->5- 3 Toto Se - Va To Paymono Tenaace, l SKETCH MAP OF THE aa COAST NEAR ULLADULLA SHOWING GLENDONITE HORIZONS d { { Collers Beach LEGEND Beach Sand RECENT TERTIARY ? Basalt Dyke E = nae eras AY | PERMO- | Upper Glendonite Bed ‘— _ CARBONIFEROUS! UPPER MARINE SERIES Fossiliferous Mudstone , (> | | Lower Glendonite Beds | \ 7 ULLADULLA HARBOUR {-Mudstane wth Concretions x Faults shewn thus IN] LOCALITY MAP , Soap hi - ie vb yoastig7 Peet eee a SU | & oS Se Ulf x S = aS rortlatheon pe ET ] > =H) df SS = jij & ~ Ve nS Walonggng & = (i 4 x Is > Ss wy JUUCADULLA “MSN ‘00G “NNI'TT ‘908g “St6T “AIX GLVIG wea bite tree od Lee oe oo te enn ne nme aed fe Gy ree nani WE = Pa OE “ « ty eee hyp ¥ ee Senex emg nal ATW eae Oats ee 25 lec Lgl err Pe Pa ‘8 EN TEN eR PLATE XV. Proc. Linn. Soc. N.S.W., 1925. IG; t -gsneiss erra te itw. 3. Grani wm st Glendonites 2. Group of glendonites. il. , > ' ’ — Proc. Linn. Soc. N.S.W., 1925. PLATE XVIII. 1. Caps caused by surface hardening in finegrained white sandstone of probable Jurassic age, Cape Villaret, Desert Basin. 2. Finegrained white sandstone of probable Jurassic age, dipping gently scawards, Gantheaume Pt., near Broome, Desert Basin. rs “s Proc. Linn. Soc. N.S.W., 1925. IPL AGID SI 1. Vegetation and sandhills at site of Merlingleigh Homestead ; summit of Kennedy Range, near north end. Common phase of the Lyons Conglomerate, east of Gascoyne Junction: 3. East (abnormal) dips on east flank of dome 14 miles west of Moogoorie Homestead. The central valley at right of escarpment is occupied by an up-fold of Lyons Conglomerate, 9 miles west of its normal cutcrop. bo a elie ee cary OPaR EO pre stninnten ghee J Q - ‘ 7 ae: a « a 7 r = 1 my Niaetat 1 r eps ts See, * ti aN a Bet Ssh fasta be = fa Sade ee ve fey 3 Sadie ees a } oy: Aah aan - ¥ Coe : ee i ae ' ut ’ i e nlp aa pe naan Se gn PSN ee meme pene ity J , : aa P : “ = é rm Aw Sa a + | | Mra | ler hes } ; a F x: Tm ; iP ik . : ety ‘ . : (* het " 7 de + ‘ > 7 : ‘ ; a a : +: ys s -* : " : ; . Ps wi . i , ; ' < ¥ Rak ' . a p 2 : z! si ie zt i . ~ . | , : ; ° S ; A a a . - ’ 4 - i « d ea : - ¥ ; ¥ a 7 « ; G ; : de F | ; ‘ A = i: > It ¥ ‘ ; | 4 ay i 5 . * : | . pe : ¥ - : ‘ > i a 2 a —~ . * y. = = fi . : ‘ - led . . S h cs : cH x F j ee epee hee ee a Proc. Linn. Soc. N.S.W., 1925. PLATE XX. 1 2 1. The Eucalyptus goniocalyx consociation, showing the dominant in the centre and a tall Alsophila on the left. 2. A westerly slope of the sandstone plateau, clothed with the Hucalyptus piperita consociation. eho he Proc. Linn. Soc. N.S.W., 1925. : PLATE XXII. 7. The Petrophila pulchella society. 8. View on exposed westerly headland. 9. The Hucalyptus stricta society. PLATE XXIII. Proc. Linn. Soc. N.S.W., 1925. NEW GENERA AND SPECIES (MOSTLY AUSTRALASIAN) OF BLATTIDAE, WITH NOTES, AND SOME REMARKS ON TEPPER’S TYPES. By ELANpD SHaAw, M.R.C.S., F.E.S. (Thirty-three Text-figures. ) [Read 24th June, 1925.] Recently the good fortune was mine to have in my hands for determination at the one time an unusually large amount of Blattid material, which included the collections of the old Macleay Museum in Sydney, the balance of the Aus- tralian Museum collection, some from the South Australian Museum, the whole of the Queensland Museum collection, the Blattidae collected in Australia in 1910-1913 by Dr. Mjéberg and sent to me by Dr. Sjéstedt, the Director of the Stockholm Museum, together with some fresh material of my own. It soon became evident that an examination of Mr. J. G. O. Tepper’s Types was most advisable, and, through the courtesy of the Director of the South Australian Museum, I was able to devote a brief holiday to this end, although the time at my disposal was too short to permit of a complete study of them. The present paper contains proposals for the erection of three new genera, Elfridaia, Tryonicus, and Ancaudellia; with a suggestion for a fourth, Cutiloidea; and descriptions of twenty-five species considered to be new to science. With reference to several of Tepper’s types notes have been added, which, it is hoped, will be of service in clearing up some doubtful questions; and an attempt has been made to explain some of the peculiarities of structure of the Panesthiinae, most strongly evidenced in the earth-digging group. There are in Australia many undescribed species, chiefly of the subfamily Pseudomopinae, and a great deal of work on Australian cockroaches remains to be done. This I had hoped to accomplish, but on account of a breakdown in health, and the changes entailed, it was mecessary to cease all regular entomological work, with little hope of taking it up again systematically. My collection was purchased by the Queensland Government, and is now in the Queensland Museum; this paper was practically completed before the transaction took place, but I have altered the numbers of specimens referred to as in my private register to those of the Queensland Museum Register, and all references through the paper to “Coll. auct.” should now read Coll. Q. Mus., and to those who entrusted specimens to me I wish to express my regret at my inability to complete the determination of them as I had hoped. For the drawings for Text-figures Nos. 6 to 11, 13 to 15, 29 to 31, and 33 I have to acknowledge the generous aid of Dr. C. Anderson, of the Australian Museum, and his artist assistant, Miss Joyce K. Allan; and for the drawings for all the others my grateful thanks are due to Mr. Henry Tryon, the distinguished Government Entomologist of Queensland, and his artist assistant, Mr. Helmsing, for without the kindly assistance of these two gentlemen the illustrating of this paper would have had to be abandoned. A 172 NEW GENERA AND SPECIES OF BLATTIDAE, The figures, unless the magnification is specified, are much enlarged, but not to a uniform scale. Subfamily HcroBiinag. Genus Eiscaua Shelford. ESCALA CIRCUMDUCTA Walk. Blatta circumducta Walk., Cat. Blatt. Brit. Mus., 1869, Suppl. p. 142.—Loboptera circumcincta Tepp., Trans. Roy. Soc. 8S. Aust., 1893, p. 37.—Ischnoptera annulata Tepp., loc. cit., p. 51. In a paper of mine (1914) in the Victorian Naturalist will be found some notes on this species, but only recently have I had an opportunity to examine Tepper’s types, and this examination confirms the opinion then expressed, that Tepper’s species was the 9 of Walker’s. Blatta circumducta. Tepper’s types are on one pin, labelled by himself as “Types ¢ and 9” and “Gilbert River, S. Austr., Sept., 87.” Both specimens are 9, and the vestigial tegmina have disappeared from the lower specimen, the upper one retaining that on the right side, and as Gilbert River is not his type locality this appears to be another instance (Shaw, 1916) where Tepper failed in 1915 to recognize the types from which he wrote his description in 1893. This species is common in Australia, ranging from Western Australia (Shel- ford, as Loboptera circumcincta Tepp., in Faun. Sudwest. Austr., 1909), through South Australia, Victoria and New South Wales to Queensland. The colour varies somewhat, some specimens in the author’s collection from New South Wales and Victoria being pale and showing the brown pronotal marking as a ring enclosing a pale disc, whilst others from Queensland have the general colour much darker and the pronotum almost entirely brown. The ootheca is carried by the female with the suture lateral, the sutural margin being convex and considerably longer than the ginglymoid margin, the divisions for the eggs to the number of about eleven on each side being plainly marked. The eggs do not lie opposite one another, but those of one side opposite the intervals between those of the other side, and each egg is indicated by one of the serrations of the sutural margin. There is little doubt that identical with this species is Tepper’s Ischnoptera annulata, under which further notes will be found. ESCALA LONGIUSCULA Walk. Blatta longiuscula Walk., Cat. Blatt. Brit. Mus., 1869, Suppl. p. 143.—Ischnop- tera obscura Tepp., Trans. Roy. Soc. S. Aust., 1893, p. 54. Tepper described obscura from a 92, but the three specimens indicated by him in 1915 as type material are ¢ of H. longiuscula Walk. Subfamily PHYLLODROMIINAE. Genus ISCHNOPTERA Burmeister. ISCHNOPTERA PARALELLA Tepp. Ischnoptera paralella Tepp., Trans. Roy. Soc. 8S. Aust., 1893, p. 53. Tepper described this species from a 9 from Kangaroo Island, and had before him at the same time a ¢ from Beverley, W.A., but I think he must have mistaken it for another 2 as, when indicating his types in 1915, he labelled it “Cotype’’ and at the same time labelled as “Type 4” a specimen obtained by Zeitz from a house in Adelaide in 1914. One of these § should be selected as allotype and described. BY ELAND SHAW. 173 ISCHNOPTERA ANNULATA Tepp. Tepper (1893) described this as from a 2 only, but in the South Australian Museum there are two ¢ specimens marked by Tepper as his Types. Both of these should be referred to the genus Hscala Shelf. and, as one of these is labelled “Victoria 3.95,” it cannot be his type of 1893, but is probably one of the specimens sent to him by the National Museum, Melbourne, and referred to as No. 4 in his paper of 1895. The other, which I believe to be his type of annulata, is really a 6 of Escala circumducta Walk. ISCHNOPTERA OBSCURA Tepp. = Escala longiuscula Walk. (q.v.). Genus LOBOPTERA Brunner von Wattenwyl. LOBOPTERA CIRCUMCINCTA Tepper (1893). Tepper’s types, ¢ and 9, in the South Australian Museum are both on one pin and, as was presumed by me (1914), are both 2 of Escala circumducta Walk. A careful examination in December, 1922, showed that the upper specimen, which Tepper may have taken for the ¢, has a lobiform tegmen-left on one side; and the lower specimen, while showing clear traces of having originally possessed lobiform tegmina, had then lost them both. LOBOPTERA DUODECEMSIGNATA Tepp. (1893). I have examined the types in Adelaide, and, although marked ¢ and 9, they are both female, and further material will probably show that they are the @ of an Escala Shelf. Tepper described the 9 only, and his designating a ¢ type in 1915 was probably an error, particularly as it is the example carrying a partly extruded ootheca. : LOBOPTERA HALMATURINA Tepp. (1893). Tepper’s types, ¢ and 9, are at Adelaide, also some “Cotypes.” There is con- siderable variation in the colour and markings of the abdominal tergites, but the distinctive marks on the thoracic tergites seem to be very constant. New Locality.—Victoria: Healesville, 1914 (auct). LOBOPTERA TRICOLOR Tepp. Loboptera tricolor Tepp., Horn Exped. Centr. Aust., Vol. 2, 1896, 357. In the South Australian Museum are two specimens labelled in 1915 by Tepper “Type of 0” and “Type of 9.” The vestigial tegmina of the “Type of g” are com- pletely separated, there are no wing vestiges, and the subgenital lamina is large and bears one style on the left side. The specimen labelled “Type of 9” is a young male larva, obviously an oversight of Tepper’s when selecting his types in 1915. Subfamily EPILnAMPRINAE. Genus ELFRIDAIA nov. The vertex of the head in both sexes exposed. Pronotum anteriorly truncate, posteriorly straight. Sexes similar. Tegmina vestigial, lateral, squamiform. Wings absent. A spiracular tube projecting on either side of the base of the supra- anal lamina. Subgenital lamina of the ¢ with two styles. Posterior metatarsus about as long as the remaining segments combined, biseriately spined beneath 174 NEW GENERA AND SPECIES OF BLATTIDAE, at the base, pulvillus extending towards the base of the segment; remaining pulvilli large with a prominent apical spine on each side; arolia moderately 1 1 1 large. Genicular spines 0.1.1. Apical spines a : cs ‘ aa Note.—Near Opisthoplatia Brunner v.W., but the vertex is exposed, and there are no vestigial wings. ELFRIDAIA EBOMAER, n. sp. Text-figs. 1 and 2. Above rufo-castaneous. Head nigro-castaneous; ocelliform spots small, testaceous; cheeks and mouth parts rufo-testaceous; antennae at least as long as the body. Pronotum finely punctate, the anterior margin truncate, exposing the vertex, posterior margin straight. Tegminal vestiges paler, punctate, squami- form, apex extending to about half-way across the metanotum. Metanotum with the posterior margin medially slightly produced, the postero-lateral angles more produced. Posterior borders of the meso- and metanotum and of abdominal tergites 1 to 7 with a row of indistinct longitudinal raised vittae. Postero-lateral angles of abdominal tergites 1 to 6 slightly backwardly produced, that of tergite 7 not produced. A tube terminating in a spiracle projecting from beneath it at either side. Text-fig. 1. EHlfridaia ebomae Shaw. co. Apex of abdomen, dorsal aspect. Drawn from the holotype. Text-fig. 2. Elfridaia ebomae Shaw. @. Apex of abdomen, dorsal aspect. Drawn from the allotype. Text-fig. 3. Calolampra candidula Shaw. ¢. Pronotum, showing outline of macula of disc. Drawn from the holotype. Supra-anal lamina rounded, slightly notched; that of the 6 (Text-fig. 1) is membranous, whitish, and not so ample as that of the © (Text-fig. 2), which is almost semicircular and castaneous. Subgenital lamina of the 6 small, rounded, styles situate in a well defined notch: of the © ample, rounded. Beneath rufo- testaceous, darker laterally and distally. Legs with the coxal process (Shaw, 1922) separated from the extremity of the coxal ridge (Shaw, 1922) by a distinct notch; posterior tibia with the spines on the outer aspect triseriately arranged; posterior metatarsus about as long as the remaining segments combined, pulvillus produced towards the base for about three-fourths of the length of the segment in the g, not so far in the 9, biseriately spined beneath, the series being longer in the 9. Arolia moderately large. Ootheca.—Membranous, flattened, suture carried laterally, with about fourteen divisions on each side. BY ELAND SHAW. 175 Length.—, 21 mm.; 92, 22 mm. Type material.—Holotype 0, No. 0/2878. Coll. Q. Mus. Allotype 9, No. 0/2878a. Coll. Q. Mus. Paratypes, 7 ¢, 20 9, and 3 oothecae. Coll. Q. Mus. Habitat.—Papua: Samarai (Holotype, Nov., 1914), Eboma Island (remainder of type material, 1914-15) (Auct.). Genus CALOLAMPRA Saussure. CALOLAMPRA CANDIDULA, Nn. Sp. Text-fig. 3. 6. Creamy-white with the pronotum, face and legs a little yellower. Antennae, of rather more than half the body length, with about the proximal one-fourth stramineous, the remainder dark greyish; ocelliform spots white, depressed, larger than the antennary sockets. Pronotum (Text-fig. 3) with the disc occupied by a prominent black macula, bilaterally symmetrical, and in form suggestive of a swallow-tail butterfly; posterior margin bluntly produced and in two paratypes with a few dark longi- tudinal striae. Tegmina and wings extending considerably beyond the apex of the abdomen; tegmina with a few small brown maculae situate on the veins. Supra-anal lamina subquadrate, with obtuse angles; cerci long, slender. Sub- genital lamina roundly produced, slightly emarginate, bearing two styles. Legs with the coxal processes pearly white; femora not strongly armed; genicular spines 0.1.1; posterior metatarsus longer than the remaining segments combined, biseriately spined beneath for its whole length, and with a few spines on the distal half of each lateral border; pulvillus apical; pulvilli of the remaining segments large, with one or two spines at each side; arolia present. Length.—27.5-32.0 mm.; tegmen 23.5-28.0 mm.; body 22.0-23.0 mm. Type material.—Holotype ¢, No. 9/2880. Coll. Q. Mus. Four paratypes, Austr. Mus., S. Austr. Mus., and Q. Mus. Habitat.—Queensland: Bellevue, H. C. Sturtridge, 1917 (Holotype); Aramac, F. Bradshaw, Aug., 1920. South Australia: N.E. Corner, F. Parsons. Note.—This species is near C. aspera Tepp., but is much larger, and the black macula of the pronotum is much more prominent. This macula is seen in other species of the genus, and is very constant in form. In (C. irrorata Fabr. it is well marked in the paler, while obscure in the darker examples. C. aspera Tepp., though pale, has the macula obscure, and in the present species it is dense and much more strikingly prominent. One of the paratypes is rather darker than the holotype, and beneath is stramineous with brownish mottling. CALOLAMPRA ATRA Tepp. Epilampra atra Tepp., Trans. Roy. Soc. S. Aust., 1893, p. 65. Tepper (1893) described both sexes, and in 1915 selected two specimens as his types and so labelled them. Both these specimens are 9, as also is a “Cotype.” Subfamily BLATTINAE. Genus EuzostTerta Shelford. EUZOSTERIA PURPURASCENS, 0. SD. ? closely allied to H. patula Walk., but rather larger, much smoother, and of quite different colouring. The rugosity of patula is replaced by minute punctures, with a coarse shagreening laterally; and the postero-lateral angles of the thoracic tergites are more produced. The postero-lateral angles of the fifth abdominal tergite are not backwardly produced, those of the sixth tergite scarcely, and those 176 NEW GENERA AND SPECIES OF BLATTIDAE, of the seventh tergite slightly and bluntly produced, not acuminate. Dorsal colour purple-bronze, except the discs of the thoracie tergites which show a green sheen, most marked laterally; the flavid margins of patula Walk. are replaced by the purple-bronze; all the tergites to the seventh abdominal bordered posteriorly with yellow, the discs of the thoracic tergites being also bordered laterally by a narrower yellow line; pronotum with the prominent triangular disc completely surrounded by a yellow line. Ventral colour generally rufo- castaneous; thoracic sternites yellow, with numerous small brown maculae; abdominal sternites distally yellowish, and laterally with a green sheen. Legs yellowish beneath, purple-bronze above, spines black; coxal ridges wide; tibiae on their outer aspect flattened, biseriately spined. Subgenital valves of the usual Blattine form. Length—38.0 mm. Type material.—Holotype 9, S. Aust. Mus. Habitat——wNorthern Territory: Roper River, N. B. Tindale. Note.—This very beautiful species I place in the genus Huzosteria Shelford, as it is closely allied to H. patula Walk., but, if Shelford’s character of the biseriately spined outer aspect of the tibia is distinctive of Polyzosteria, then both species should be in this genus. In both, the tibiae are more quadrangular than rounded, and, except for a single median spine at each end of the outer aspect, the spines are in two rows only. These two median spines are to be seen in P. limbata Burm. and in other species of Polyzosteria. 'The more definitely reflected lateral margins of the pronotum appear to be correlated with the more clearly defined three rows of spines, and a genus intermediate between the two may be necessary when further study has been given to the group. HEUZOSTERIA SORDIDA, N. Sp. ? dull black above and below, except the discs of the abdominal sternites which are castaneous and shining; the dorsal surface covered with prominent yellow folds or tubercles, being rugose between the tubercles. Head brownish- black, punctate; antennae brown, basal segments darker; palpi yellow. Margins of the thoracic tergites strongly reflected, yellowish with numerous blackish dots and maculae, the latter appear on the inner aspect of the reflected portion. Postero-lateral angles of the fifth abdominal tergite slightly, of sixth more, and of seventh strongly produced backwards. Supra-anal lamina rounded, slightly emarginate, rugose and tuberculate; subgenital lamina with valves of usual Blattine form. Coxae dull black, coxal borders and coxal processes yellow; femora castaneous; tibiae and tarsi yellow with black spines, which on the outer side of the tibiae are triseriately arranged. Claws and arolia brown. Length.—26.0 mm. Type material.—Holotype ?, No. 0/2876. Coll. Q. Mus. Habitat.—W. Australia: Beverley, F. H. du Boulay. Note.—This is quite distinct from swbverrucosa White, of which species the Hope Museum kindly gave me a specimen in 1914, and one may assume that Shelford was familiar with the type which is in the British Museum. Little or none of the metallic sheen usual in the genus is shown by this species. There are specimens in the South Australian Museum and in my collection, from South Australia. BY ELAND SHAW. 177 EVUZOSTERIA TUBERCULATA, n. Sp. Text-fig. 4. Testaceous, prominently and coarsely tuberculate, with dark fuscous or bronze-green in the hollows between the tubercles of the thoracic tergites and some splashes of the same on the abdominal tergites. Head with the vertex and frons brown or almost black, punctate, the line between the epicranial plates and the clypeus being pale and well marked; eyes very prominent; genae testaceous; palpi pale. Thoracic tergites with the lateral margins strongly reflected, pale outside; the dorsal margin of the reflected portion with a row of small brown tubercles, almost serrate; pronotum with prominent symmetrical Text-fig. 4. Huzosteria tuberculata Shaw. 2. Whole insect (x 3/2). Drawn from the holotype. tubercles (Text-fig. 4), the anterior one shaped like a fleur-de-lys; tubercles of the meso- and metanotum mostly sausage-shaped; of the abdominal tergites, rounded and disposed in almost regular rows. Supra-anal lamina in ? rounded, slightly emarginate, subtectiform, darker basally, with several coarse brown tubercles; subgenital lamina in 9 of the usual Blattine form; cerci short, blunt, flattened, yellow. Beneath smooth, shining; visible parts of the thoracic tergites testaceous; abdominal sternites testaceous, paler at the posterior margins, and fuscous laterally. Legs fuscous; distal portions of the coxae, the knees and the whole of the tibiae and tarsi testaceous, spines black; arolia small. Length.—Holotype 9, 29.0 mm.; allotype, larval 3, 20.0 mm. Type material.—Holotype 9, No. 0/2877; and allotype ¢, No. 0/2877a, Coll. Q. Mus. Paratypes 3. Habitat.—Victoria: Mallee District, 1918 (Holotype); Lake Hattah (Allotype). South Australia: Ooldea and Murray River. Note.—The projecting eyes, which are very marked in this species, bulging well beyond the general contour of the vertex and genae, are to be found also in some other species of this genus; the smoother species, with but slightly reflected margins of the thoracic tergites, having the eyes not or scarcely projecting. The subgenital lamina of the holotype only is described, the allotype 6 being a larva. In some examples the ground colour is considerably paler than in the holotype, which is a dark specimen, the larval allotype being very pale in the ground colour, while the dark parts of the legs are almost black. The fleur-de-lys shaped tubercle on the pronotum seems quite distinctive of this species. 178 NEW GENERA AND SPECIES OF BLATTIDAE, Genus LEPTOZOSTERIA Tepper. Tepper (1893) provisionally proposed this genus for L. prima, distinguishing it chiefly by the form of the supra-anal lamina of the ¢ (the only sex he knew), and by the unusual distribution of the colouring. In the following year he added another species, L. secunda, from a single 9, placing it in the genus chiefly on account of the colouring, in which the two have a strong resemblance. Since then, to the genus Platyzosteria have been added several other species in which the supra-anal lamina departed as much or more from the usual form, and Tepper himself must have ceased to regard the colouring as diagnostic, or he would have included in the genus zebra and coolgardiensis, both pale forms with dark bands, especially as the supra-anal lamina of coolgardiensis ¢ is back- wardly produced. I have examined Tepper’s Type of L. secunda in the South Australian Museum, and the posterior tarsal structure is that of Cutilia Stal, and I have no doubt of the correctness of Shelford’s opinion that the species is identical with Cutilia triangulata Br. v. W. (q.v.). The posterior tarsal structure of prima Tepp. is that of Platyzosteria, to which genus I think this species should be referred; and I see no reason for the retention of the genus Leptozosteria. Genus PLATYZOSTERIA Brunner y. Wattenwyl. PLATYZOSTERIA ARMATA Tepp. (1893). Tepper’s types in the South Australian Museum were examined by me, and his labels were reversed as to sex. This species was originally described from Western Australia, but the South Australian Museum has now more recent specimens from several localities in South Australia. It belongs to the spinose group, of which the most striking example is ferox Shelf., but is of a brownish castaneous colour, and Shelford’s (1909) redescription was probably made from a very dark specimen. The posterior metatarsus is of the Platyzosteria type, and the points of difference from P. rufofusca Tepp. will be found under that species. PLATYZOSTERIA PSEUDATRATA Tepper. Platyzosteria pseudatrata Tepp., Trans. Roy. Soc. S. Aust., 18938, p. 86. There is only a single ¢, the type, in the South Australian Museum. Shelford (1909) only knew it by description and was unable to distinguish it from P. analis Saugs. or P. melanaria Erichs.; but the type does not agree with Shelford’s (1909) figures of the latter, nor is it conspecific with what I recognize as analis Sauss., the common species in Victoria. It came from Central Australia, and appears to be allied rather to the group whose spininess finds its extreme example in ferox Shelf. P. pseudatrata is, I believe, a,good species; it is apterous, but with some crumpling of the mesonotum laterally, and is very broad in pro- portion to its length; the 6th and 7th abdominal tergites are laterally serrate, the 7th strongly so and there are prominent spines of the supra-anal lamina; the posterior metatarsus is not long and not spined beneath. PLATYZOSTERIA RUFOFUSCA Tepper. The type, a 9, and unique, is in the South Australian Museum. Compared with P. armata Tepp., it is smaller, not scabrous laterally, and the lateral margins of the 6th and 7th abdominal tergites are not denticulate, but only very finely serrate. The supra-anal lamina, which appears to be slightly deformed, differs in shape, and is not strongly denticulate. The posterior metatarsus is of the Platyzosteria type. BY ELAND SHAW. 179 PLATYZOSTERIA SCABRELLA Tepper. The types of ¢ and @ are in the South Australian Museum. The posterior metatarsi are short and unspined beneath; the 5th tarsal segment is brownish- yellow; the pale antennae show up very distinctly; and the degree of scabrousness varies in different examples, especially on the pronotum. P. scabrella may be taken as representing a group referred to previously by me (1922, p. 224), and the doubtful species then mentioned is here described as P. anceps. The little Western Australian species P. scabrosa, though not included in the key, is perhaps another member of this group, as may also be P. variolosa Bol., a species of which I have no personal knowledge. The following key may be of some service in separating the species of this group. Key for P. scabrella Tepp. group. (2) paeeVestiqials teominay PreSem Ge epee ocr cyen eel cre cieteienoveuee oilWeliel clfsuiciaronene bicolor Kirby. (1) 2. Entirely apterous. (6) 8. Thoracic tergites with shallow pits, not scabrous. (5) 4. Orange macula at antero-lateral angle of 7th abdominal tergite. babindae Shaw. (4b) eee NIOMOLANEe) Ma CUl ay erie pane sous eyo ties) ap ceehiciay en sicieeuch elec aeiiel sucpalvehagareey cus anceps Shaw. (3) 6. Thoracic tergites scabrous. (8) 7. Smaller species, antennae ochreous, lateral margins of 6th abdominal letercitemnotEserEatemenneercrie comin scabrella Tepp. (7) 8. Larger species, antennae fuscous, lateral margins of 6th abdominal tergite serrate ................... as Sees Behe Sy scabra Br. v. W. PLATYZOSTERIA ANCEPS, nN. sp. Text-fig. 5. Closely allied to P. babindae mihi (1922), but has no orange red macula at the antero-lateral angles of the 7th abdominal tergite; is larger, and the subgenital lamina of the ¢ is crenulate (Text-fig. 5). This species is also apterous, but the meso- and metanotum have shallow grooves indicating the position of vestigial tegmina and wings; the lateral margin of the 7th tergite is serrate, as is also that of the 8th abdominal sternite (Text-fig. 5), the last-named character being absent in both babindae mihi and scabrella Tepp. Differs from scabrella Tepp. in the presence of shallow pits of the thoracic tergites, the darker antennae, the presence of lateral grooves of the meso- and metanotum, the shorter subgenital lamina of the male with styles which, though lateral, are more caudally inserted and extend considerably beyond the lamina. The coxal borders are narrowly ochreous, and the tarsi are of the Platyzosteria type. Length.—, 21.0-24.0 mm.; 9, 19.0-22.0 mm. Type material.—Holotype 3g, No. 0/2892; allotype 9, No. 0/2892a; and 13 paratypes. Coll. Q. Mus. Habitat.—Queensland: Holotype, Upper Burnett District (C. Hogg); allotype, Mt. Forbes, Rosewood (C. Dreveson); Laidley (auct.); Jandowae (R. Illidge); Thulimbah (S. Perriman); Stanthorpe District (H. Jarvis); Childers (J. F. Illingworth); caves near Rockhampton (R. L. Higgins). Note.—This species was mentioned by me (1922) and examination of a larger amount of material shows that it can readily be separated from babindae mihi, and from scabrella Tepp., which lie on either side of it. PLATYZOSTERIA BICOLOR Kirby. Melanozosteria bicolor Kirby, Ann. Mag. Nat. Hist. (7), xii, 1903, p. 373. New Localities—Torres Straits: Darnley Island; New Guinea: Hall Sound. Macleay Mus., Sydney. 180 NEW GENERA AND SPECIES OF BLATTIDAE, PLATYZOSTERIA SCABROSA, n. Sp. Text-fig. 6. Small, black, apterous, densely scabrous. Head with the vertex and frons showing numerous shallow depressions, black, with the margins of the clypeus and labrum rufo-fuscous; antennae yellow, except the three proximal segments, ie a) Gael. Pgh lohagll eaifireneer ee ne) . OG Was v ; a 6 Q \. & \ i Text-fig. 5. Platyzosteria anceps Shaw. co. Apex of abdomen, ventral aspect. Drawn from the holotype. Text-fig. 6. Platyzosteria scabrosa Shaw. co. Apex of abdomen, dorsal aspect. Drawn from the holotype. which are fuscous. Thoracic tergites moderately, and abdominal tergites densely scabrous; lateral margins of the 7th abdominal tergite not serrate, but bearing a few hairs, each springing from a slight elevation; abdominal sternites black, finely scabrous, the distal margins rufo-castaneous. Supra-anal lamina of the ¢ (Text-fig. 6) subquadrate, emarginate, fimbriate, not so coarsely scabrous as the preceding tergites, tips and ventral surfaces of the cerci dark orange colour; of ? similar, more produced and more deeply emarginate. Subgenital lamina of the ¢ transverse, posterior margin faintly concave; styles slender, acuminate, slightly incurved, longer than the lamina and of about the same length as the cerci; of the 9, the subgenital lamina is of the usual bivalvular form. Legs rufo-castaneous, margins of the coxae narrowly rufous. Metatarsi of the Platyzosteria type. Length.—d, 10.0 mm.; 9, 11.5 mm. Type material—Holotype ¢ and allotype 9, Macleay Museum, Sydney. Habitat—Western Australia: King George’s Sound. Note.—This little cockroach probably belongs to the scabrella. Tepp. group, with no vestigial tegmina or wings, and a scabrous dorsal surface. Its small size and more scabrous surface distinguish it from scabra Br.v.W. and scabrella Tepp. Bolivar’s species variolosa, from New Caledonia, I do not know the size of, but Shelford’s (1909) note and figure of it could hardly refer to the present species. PLATYZOSTERIA JUNGII Tepper. Periplaneta jungii Tepp., Trans. Roy. Soc. S. Aust., 1895, p. 162. Tepper described the ¢ and, after examining the type material in the South Australian Museum, I have no doubt of this being quite a distinct species. There were 3 specimens, but Tepper, in indicating his types in 1915, unfortunately selected as the “Type of @” an adult 9 bearing on its label the date 2.1.96, so this obviously could not be the type. Of the other 2 specimens, one was a larval ¢ labelled “Yorktown 30.12.94 A: Jung,’ and must be selected as the type of the species; the third being another adult 9, bearing no label, and this BY ELAND SHAW. 181 specimen the Museum kindly presented to me, and it is now in my collection. This species belongs to the group which includes biglumis Sauss., but to that portion of it without ochreous coxal borders, and is most nearly allied to P. melanosa, a species from South-Western Australia, here described as new. P. jungii Tepp. is very broad, looks like a small Cutilia nitidia Br. v. W., but distinguished, not only by its tarsal structure, but by its vestigial tegmina which are not completely separated from the mesonotum, narrow and acuminate, and are produced a little beyond the posterior tergital margin. The ocelliform spots are not seen; the antennae are longer than the body, black for about their distal 2, then fuscous; mouth parts black. The distal abdominal tergites have their postero-lateral angles strongly produced backwards and the lateral margins of the 7th not serrate. The supra-anal lamina of the 9 is triangular, deeply emarginate, with apical serrations. Legs wholly black, posterior metatarsus with pulvillus extending upwards, with strong hairs, some of which appear to be disposed in basal biseriate rows beneath, but they are not spines as in Cuwtilia; ungues black. PLATYZOSTERIA MELANOSA, Dn. Sp. Text-figs. 7 and 8. Shining, black, with the exception of the ocelliform spots, the castaneous or fuscous margins of the clypeus and labrum, the castaneous tips of the cerci and the testaceous 5th tarsal segments. Vertex exposed; antennae with the basal segment black, the remainder missing. Vestigial tegmina not completely separated, narrow, acuminate, produced a little beyond the margin of the mesonotum; no wing vestiges. Text-fig. 7. Platyzosteria melanosa Shaw. co. Apex of abdomen, dorsal aspect. Drawn from the holotype. Text-fig. 8. Platyzosteria melanosa Shaw. @. Apex of abdomen, dorsal aspect. Drawn from the allotype. Postero-lateral angles of the distal abdominal tergites strongly backwardly produced, lateral margins of 7th tergite not serrate; supra-anal lamina of the 6 (Text-fig. 7) subquadrate, widely emarginate, fimbriate, lateral margins everted, three strong spines at each postero-lateral angle, exceeded by the cerci; of the © (Text-fig. 8) triangularly produced, strongly spined all round except towards the base, where there is some eversion of the margin; subgenital valves of the usual Blattine form. Terminal abdominal sternites in both sexes somewhat scabrous. Coxal borders black, posterior metatarsus about as long as the remain- ing tarsal segments combined, pulvillus produced for about half the length of the segment, with a few short hairs basally. Length.—, 22.0 mm.; 9, 24.0 mm. Type material.—Holotype 3, No. K45547, and allotype 9, No. K51764. Aus- tralian Museum, Sydney. 182 NEW GENERA AND SPECIES OF BLATTIDAE, Habitat—Western Australia: Michaelmas Is., King George’s Sound (Troughton, Grant and Wright, 21.11.21). Note.—This species belongs to the group of P. biglumis Sauss. with only partly separated tegminal vestiges, and to that section of this group without pale coxal borders. It is larger than biloba Sauss. and perplexa Shelf., and has not the peculiar structure nor the long cerci of curiosa Shelf., but the last species, as Shelford thought possible, may have been described from a larva. P. melanosa is most nearly allied to P. jungii Tepp., being of about the same length, but of a markedly narrower shape, has a few small departures from the uniform blackness of that species, and the supra-anal lamina of the 9 is of a quite different form. PLATYZOSTERIA ARDROSSANENSIS Tepp. (1893). A synonym of Zonioploca alutacea Stal (q.v.). Tepper’s “Type of 9,” as indicated by him in 1915, is labelled “Goolwa.” This is evidently an error, as Ardrossan is the only locality mentioned in the original description; but a 9 amongst the “Cotypes” bears an old label ‘Ardrossan 26-2-85’’; and was no doubt his type of @. PLATYZOSTERIA AVOCAENSIS Tepp. Tepper (1893) described the 9 only, but in 1915 he selected, in addition to his “Type of 9,” a specimen which he labelled “Type of ¢.” This, while probably conspecific with the type of 9, is not adult, but really a larval 9. The type of 9 is distinct from castanea Br.v.W., being larger, apterous, the lateral margins of the 7th abdominal tergite are serrate, and the ootheca is smooth, while that of castanea is prominently longitudinally keeled as shown by specimen No. 272 in my collection. PLATYZOSTERIA EXASPERA Tepper (1894). Tepper (1894) described the ¢ only, but there is confusion here also with regard to his types. In 1915, he selected, in addition, a “Type of 9,” using a specimen which reached him after the date of the publication of his paper. I am unable to separate either of these from castanea Br.v.W., and although Shelford appeared to consider both avocaensis Tepp. and ezaspera Tepp. identical with castanea Br.v.W., examination of the types convinces me that avocaensis at least is a separate species. PLATYZOSTERIA PROVISIONALIS Tepper. Periplaneta provisionalis Tepp., Trans. Roy. Soc. S. Aust., 1893, p. 108; 1895, p. 161.—Syntomaptera provisionalis Tepp., Kirby Syn. Cat. Orth. Brit. Mus., 1, 1904, p. 180.—Platyzosteria provisionalis Tepp., Shelford, Trans. Ent. Soc. Lond., 1909, p. 282. With regard to the type of this species, there are two specimens in the Type Collection of the South Australian Museum selected and marked as type material by Tepper in 1915, more than twenty years after he had described the species, and after he had left the Museum. As he was then a very old man there iy little wonder that numerous mistakes occurred in the selection of the types, and the matter has already been alluded to by me (1916) when dealing with the types of Ischnoptera brunneonigra Tepp. The two specimens referred to are not conspecific, that which he selected and labelled as “Type of 9?” also bearing old labels, giving Torrens Creek as locality, the date 1894, and the designation BY ELAND SHAW. 183 “% nymph.” It is really a 6, apterous, and has the posterior metatarsi spined beneath. The second specimen, labelled in 1915 as “Cotype,” bears also a label “Mt. Bryan EH. 4.6.87. T. Best.” It is an adult 9, 12 mm. long, with long narrow tegminal vestiges, and the posterior metatarsi missing. I consider this should be indicated as the holotype of provisionalis Tepp., the first specimen being an undetermined Cutilia. PLATYZOSTERIA ALBOMARGINATA Br. v. W. This species will probably be found widely spread from Western Australia through South and Central Australia to New South Wales, but I have no records from Victoria. Brunner (1865) gave a meagre description of the larval g, but Shelford (1909) added more details and figured the apex of the abdomen of the 6. The supra-anal lamina of the 9 is of a form almost identical with that of the ¢, but I think Shelford’s figure errs in representing the lateral margins of the 7th tergite as so coarsely serrate. The posterior metatarsus is shorter than the remaining segments combined, unspined beneath, and with a pulvillus extending proximally. The diagnostic points between this and the following species, brunnea Tepp., will be found under the latter. In some specimens the prominent white lateral border of the thoracic tergites is almost obsolete, only showing as a short narrow streak on the pronotum, and as a small dash or spot on the meso- and metanotum. New Localities—South Australia: Pungonda (A. Dubbe), Kangaroo Island (A. M. Lea), Murray River (H. S. Cope), Ooldea (A. M. Lea). S. Aust. Mus. Key for P. albomarginata Br. v. W. group. (8) 1. Pale border not extending round the abdomen. (3). 2% Sorell gases (GO sammy) shoosodcnacnocdog0upeGes inclusa Walker. (2) 38. Larger species. : (5) 4. Posterior margin of seventh tergite sinuate ........ variegata Shelf. (4) 5. Posterior margin of seventh tergite not sinuate. Cy Gs Abdomen) tapering) caudallys s745.. 2500 nee eee eee albomarginata Br. v. W. CC) eee Abdomen) broadmcaudalilya seers. eee cele clei eerie brunnea Tepp. (1) 8. Pale border extending round the abdomen. GO) Poe Stronelyaspinedmcaudallya essa on eee ee one spenceri Shelf. SD ELOROINO ER SO! oscars on crtaneron Sree eliald ran cc auencscet nceauensne ear Snee asacteate Bas obscuripes Tepp. PLATYZOSTERIA BRUNNEA Tepper (1893). An examination of Tepper’s type material in the South Australian Museum showed that is consisted of four specimens marked by Tepper in 1915, “Type of 6, Type of 9. Cotypes § and 9.” The “Type of ¢” was a 9 of P. communis Tepp. with a locality label “Brighton” attached, and was removed at my instance by the Museum entomologist. The three other specimens were larval 9°, the one selected as “Type of 2” being one of the original Gilbert River examples, dated “LL. Molineaux 30.5.87.”, and should be recognized as the holotype of the species. Capt. S. A. White has since brought in several specimens, including males, from Central Australia, and one of these should be selected as the allotype 4, unless Tepper’s real type of ¢ can be found. This species differs considerably from albomarginata Br. v. W., although it belongs to the same group of Platyzosteria, that is, those with a flavid lateral border but no tegminal vestiges. The abdomen of albomarginata Br. v. W. tapers gradually towards the apex, whilst in brunnea Tepp. the 6th tergite is as wide as the 5th, and only a little wider than the 7th, the postero-lateral angles of which are very strongly backwardly produced, 184 NEW GENERA AND SPECIES OF BLATTIDAE, and relatively wide apart. The lateral margins of the 7th abdominal tergite are strongly serrate in brunnea Tepp. and but faintly in albomarginata Br. v. W., and there is a definite thickening of the lateral borders of the abdominal tergites in the former, which the latter does not show. The cerci of albomarginata Br. v. W. are relatively much longer than those of brunnea Tepp. In colouring also the two differ, for whilst albomarginata Br. v. W. is almost black, brunnea Tepp. is brown, and has a broader flavid margin.* PLATYZOSTERIA SUBMARGINATA Tepper. Drymaplaneta submarginata Tepp., Trans. Roy. Soc. S. Aust., xvii, 1893, p. 11. Shelford (1909) regarded this species as identical with P. circumducta Walk., but while that species will probably prove to be a Cuwtilia Stal, an examination of Tepper’s types of submarginata show it to be a true Platyzosteria. The larval 9? which Tepper describes, and gives the length of as 21 mm., I was unable to trace in the South Australian Museum. PLATYZOSTERIA RUGOSA, n. sp. Text-figs. 9, 10 and 11. 6 apterous, nigro-castaneous. Dorsal surface densely and coarsely punctate. Head black; ocelli orange, minute; margins of the clypeus, and of the emarginate labrum pale; antennae longer than the body, black. Pronotum almost concealing the vertex, posterior margin straight, broadly bordered with orange-yellow except in the middle third where the castaneous disc reaches it; meso- and metanotum with the postero-lateral angles slightly produced, the outer fourth of the former and the outer fifth of the latter being orange-yellow. Abdomen slightly wider than the thorax; tergites 5, 6 and 7 with the postero-lateral angles produced; tergite 8 with the postero-lateral angles bearing spiracles; tergite 9 with the angles serrate posteriorly; tergites 6 and 7 with the posterior margins crenulate and serrate respectively; supra-anal lamina (Text-figs. 9 and 11) triangular, the lateral margins Text-fig. 9. Platyzosteria rugosa Shaw. co. Apex of abdomen, dorsal aspect. Drawn from the holotype. Text-fig. 10. Platyzosteria rugosa Shaw. co. Apex of abdomen, ventral aspect. Drawn from the holotype. . Text-fig. 11. Platyzosteria rugosa Shaw. co. Apex of abdomen, lateral aspect. Drawn from the holotype. serrate at the base, apex produced into a long, sharp, deflected spine, medially carinate at the base; subgenital lamina (Text-fig. 10) produced into a long, flattened, slightly reflected process whose apex, about 0.75 mm. wide, is crenulate; * Since writing the above I find that Mr. N. B. Tindale (1923) has published a figure of the holotype 2; and a figure, with some descriptive notes, of a ¢ from Pearson Island. BY ELAND SHAW. 185 styles about the length of the lamina, acuminate, with a short spine within the base; cerci long. Abdomen beneath smooth, nitid, nearly black, sternites 7 and 8 with the posterior angles and the outer portions of the posterior margins serrate. Legs black; coxae very narrowly margined yellow; posterior metatarsus long, unspined beneath, pulvillus occupying the whole segment. Other tarsal segmrents unspined, with large pulvilli. Arolia present. Length.—25.0 mm. Type material.—Holotype ¢, South Australian Museum. Habitat—wNorthern Territory: Mary River (W. D. Dodd). Note——The spinose distal tergites and sternites and the extended supra-anal lamina resemble those of Leptozosteria prima Tepper, but the texture of the dorsal surface and the form of the subgenital lamina are very different, and I leave this species in the genus Platyzosteria for the present. This unique specimen is a striking and handsome cockroach, and it will be interesting to see further material. PLATYZOSTERIA CINGULATA Shaw. Platyzosteria cingulata Shaw. Proc. Linn. Soc. N. 8. WALES, xlvii, 1922, p. 226. New Locality. Queensland: Toowoomba (H. Longman). Coll. Q. Mus. PLATYZOSTERIA CURIOSA Shelford. Platyzosteria curiosa Shelf., Faun. Sudwest. Aust., Bd. ii, Lief 9, 1909, p. 135, Plate xiii, figs. 11, 12. Described from a single 9 from Northam, Western Australia, and Shelford mentions as doubtful a second specimen in the Hope Museum. In 1920 the late F. P. Spry gave me a 9 (No. 230 Coll. Shaw), with ootheca attached, which I refer to this species. No locality was obtained, but doubtless it was Australian. The insect is much depressed, as is also the ootheca, which is carried suture dorsally; the suture as far as visible showing no serrations. In my specimen the supra-anal lamina is not dentate all round as shown in Shelford’s figure, but only about the apex, the remainder being spined. PLATYZOSTERIA PRIMA Tepper. Text-fig. 12. Leptozosteria prima Tepp., Trans. Roy. Soc. S. Aust., 1893, p. 96.—Platyzosteria aposematica Shelf., Trans. Ent. Soc. Lond., 1909, p. 288. Text-fig. 12. Platyzosteria prima Tepper. co. Apex of abdomen, dorsal aspect. Drawn from No. 0/2902, Coll. Q. Mus. Text-fig. 13. Zonioploca bicolor Shaw. oc. Apex of abdomen, dorsal aspect. Drawn from the holotype. Text-fig. 14. Zonioploca bicolor Shaw. o. Apex of abdomen, ventral aspect. Drawn : from the holotype. 186 NEW GENERA AND SPECIES OF BLATTIDAE, I have examined all the specimens enumerated below, and I think they will be found to be examples of a species having a considerable range of variation in both size and colouring. The dark markings on the thoracic tergites do not always display a pale discal portion; but all the examples have a very depressed form, a spinosity of the terminal abdominal segments, and a similar form of the terminal laminae in each sex. In all, also, the form of the posterior tarsus is the same, and is that of Platyzosteria. Tepper described prima from a single J, and Shelford aposematica from a single 9, giving a good figure. Both are recorded from Central Australia, as are also White’s specimen, and the remaining two from the fringe of that area taken in the faunistic sense. As the @ has never been figured the apex of the abdomen is shown in Text-fig. 12. Habitat—Central Australia: 8, Type of L. prima Tepp.; Spencer-Gillen Exped. 1889. 9, Type of aposematica Shelf. (Nat. Mus. Melb.); Capt. S. A. White (Adelaide Mus.). Queensland: Winton, 1912, ¢, Higgins (Q. Mus.); Aramac, 1920, larval 2 (Austr. Mus., Sydney, No. K43406). SYNTOMAPTERA GLABRA Tepper. Syntomaptera (Periplaneta) glabra Tepp., Trans. Roy. Soc. S. Aust., 1898, p. 107. There are five specimens under this name in the South Australian Museum. All are 2 and of these, three labelled “N. Territory, Pulleine’” (not Magarey) are of an apterous Cutilia, with typically spined posterior metatarsus; certainly not Walker’s Periplaneta glabra, nor Platyzosteria aterrima Erichs. (teste Shelford, 1909); but probably either Tepper’s parva or an allied undescribed species. The other two specimens are of a true Platyzosteria, with vestigial tegmina and yellow margined coxae, not aterrima Hrichs., which is apterous, and not parva Tepp., which is a Cutilia; but may be the 2 of glabra Walker, of which the ¢ only was described. All these specimens are 9, and it is unwise to be definite about them, but in any case neither of the two specimens of Platyzosteria can be the type of glabra Tepper, as they did not reach him until months after his description of that species was published. Genus Zonroptoca Stal. ZONIOPLOCA ALUTACEA Stal. Zonioploca alutacea Stal, Bih. Svensk. Akad. ii (13), 1874, p. 13.—Platyzosteria ardrossanensis Tepp., Trans. Roy. Soc. S. Aust., 1893, p. 92. Shelford (1909) saw Stal’s type and found that Tepper’s description of ardrossanensis “corresponded perfectly” with it. New Localities—South Australia: Goolwa, Tepper (South Australian Mus.), Myponga, A. H. Elston (Coll. auct.). Western Australia: Mullewa, 1914; Cunder- din, 1913-14 (S.A. Mus. and Coll. auct.). ZONIOPLOCA BICOLOR, n. sp. Text-figs. 13 and 14. Ochraceous. So densely covered with reddish-brown granules that, except at the thickened margins of the pronotum, the dorsal ground colour is scarcely visible. Head ochraceous, granulate, antennae reddish-brown, paler proximally. A short thick yellow streak extending backwards and outwards about the middle of each lateral half of the meso- and metanotum; these streaks appear on the abdominal tergites as irregularly shaped maculae; beneath sordid testaceous, not granulate. Legs with most of the dorsal aspect of the coxae and femora shining BY ELAND SHAW. ( 187 black, and that of the tibiae rufo-castaneous, and of the tarsi partly black, ventral aspect of the legs concolorous. Coxal borders pale yellow on both aspects. Pos- terior metatarsus about as long as the remaining segments combined, unspined beneath, pulvillus apical, arolia large. Supra-anal lamina ¢ (Text-fig. 13) large, rounded, very faintly emarginate, fimbriate, cerci broken off. Subgenital lamina ¢ (Text-fig. 14) subquadrate, angles obtuse; styles lateral, long, acuminate. Length.—, 17.0 mm.; 9, 22.0 mm. Type material.—Holotype 3g, No. K45509; and allotype 9, No. K45510, Aus- tralian Mus., Sydney. Habitat.—Western Australia: Cranbrook (Troughton and Wright). Note.—This species is much like Z. medilinea Tepp., without the black stripe, or granulose sternites; but the legs are more like those of alutacea Stal and pallida Shelf. They differ, however, in their striking colouring, whilst the ventral aspect of the legs is inconspicuous, the dorsal aspect, with the black and yellow coxae and femora, castaneous tibiae, and parti-coloured tarsi, would afford a startling spectacle should these cockroaches assume for defence the tilted forward attitude noted by me (1914) as used by Platyzosteria castanea Br. v. W. (but in that note I wrote “‘ventral’’ for dorsal). The leg colouring of this and other species supports the idea that the attitude may be employed by many cockroaches, and may account for the frequently found pale coxal borders. ZONIOPLOCA DIXONI Shaw. Zonioploca dixoni Shaw, Proc. Linn. Soc. N. S. WALES, xlvii, Pt. 3, 1922, p. 231. This species was described by me as new, and placed with some reservation in Zonioploca Stal; but since examining Tepper’s type of his Lepidophora furcata in the South Australian Museum, I have no doubt but that the two are identical, and my name therefore becomes a synonym of his (vide Eppertia furcata Tepp.). ZONIOPLOCA LATIZONA Tepper. Platyzosteria latizona Tepp., Trans. Roy. Soc. S. Aust., 1893, p. 92. Shelford (1909) was uncertain as to the position of this species, but an examination of Tepper’s types shows it to be a Zonioploca, probably near robusta Shelf., but smaller. The types, J and 9, are in the South Australian Museum, and a “Cotype” kindly presented to me by the Museum, is in the Queensland Museum collection, No. 0.2856. Genus EPPERTIA, nov. EXPPERTIA FURCATA Tepper. Lepidophora furcata Tepp., Trans. Roy. Soc. 8S. Aust., 1895, p. 21; Report Horn Exped. Cent. Aust., 1896, p. 363—Tepperia furcata Kirby, Syn. Cat. Orth. Brit. Mus. (Additions and Corrections to Vol. i), Vol. iii, 1910, p. 565.—Zonioploca dixoni Shaw, Proc. Linn. Soc. N. S. WALEs, xlvii, Pt. 3, 1922, p. 231. Tepper described this species from a ¢ taken at Lake Callabonna in South Australia. His type is in the South Australian Museum, marked as such by him- self, and the Museum possesses a second ¢ captured at Oodnadatta, Central Aus- tralia, and mentioned by Tepper in the Report of the Horn Expedition. This specimen is also marked “Type” in Tepper’s writing, evidently an error, as it was B 188 NEW GENERA AND SPECIES OF BLATTIDAE, not captured until some years later, and the Horn Expedition Report was not published until more than a year later than the original description. The next known of the insect was when the late F. P. Spry sent me specimens from the Victorian Mallee and from South and Central Australia, and not recognizing it as Tepper’s Lepidophora furcata, I (1922) described it as Zonioploca dizoni; but an examination of Tepper’s type showed me my error. Tepper thought it was a Perisphaeriine, never having seen the 9, but my @ (the allotype) and a 9 returned to Mr. Spry have the subgenital lamina of the usual Blattine form, and I think the genus should come after Zonioploca Stal. Kirby followed Tepper in placing it in the Perisphaeriinae. The “scale like appendage’ to which Tepper (1895) refers, and from which he took his generic name, is not, as he stated, parts of the abdominal tergites, but unusually large pleural plates bearing at their postero- lateral angles the hard ring-like tracheal openings. These, of course, are found in all cockroaches, and may be plainly seen in Scabina, Anamesia, and Zonioploca, and probably in many others. The largest known to me are those in the present species, and those nearer the thorax are larger than those more caudally situate. In the type the subgenital lamina is shorter than in the Oodnadatta specimen, or in the ¢ in my collection. It may be deformed, as the subgenital lamina frequently is, notably in Cosmozosteria, and is evidently variable, as my specimen bears an internal tooth on the divergent spines which is absent in the others. Tepper’s generic name Lepidophora was used in Diptera by Westwood in 1835, so Kirby, in 1910, renamed the genus Tepperia; but Tepperia had already been used by A. M. Lea in 1903 for a genus of Curculionidae (Proc. Linn. Soc. N. S. WALES xxviii, 1903, p. 660), and a new genus is therefore needed to replace Kirby’s. I would propose Eppertia, hoping thereby to preserve Kirby’s compliment to Tepper. Genus CurTinia Stal. Key to Cutilia. (34) 1. Not transversely banded. (29) 2. Possessing vestigial tegmina. (26) 3. Vestigial tegmina lobiform, subgenital lamina of the ¢& symmetrical. (19) 4. Not margined pale laterally. (18) 5. Legs not testaceous. (9) 6. Tegminal vestiges obliquely truncate. (8) 7. Large, relatively broad, legs black, ootheca smooth .... nitida Br. v. W. (7) 8. Smaller, not relatively broad, legs castaneous and rufous, oothéca: “fluted: ....:c..F ee a re ee ee eee nitidella Shaw. (6) 9. Tegminal vestiges more or less tapering, not obliquely truncate. CH) LO "Dorsume® tubercuilaiter Via sic cceis cme eeeeeeave ee tcuetccawadeke ete eas eee obscura Tepp. (10) 11. Dorsum not tuberculate. (15) 12. Spine at base of genital styles. (14) 18. Tegminal vestiges narrow, tapering sharply ............ Kellyi Shaw. (13) 14. Tegminal vestiges broad, tapering roundly .............. brevitarsis Shaw. (12) 15. No spine at base of genital styles. (17) 16. Small size (about 17.0 mm.), subgenital lamina of &% with QOS aay “GHEE so bbouogocouboudeGa0bo0nDDS scabriuscula Tepp. (16) 17. Large size (about 27.0 mm.), subgenital lamina of & with IQOLEINOVE oN SINN COMES coscnocondooanobebooabooDOSS imsularis Shaw, MS. G53) 91'S WRersi nteStaceOUuSin his was chacnhtsat at oesaioeeret ine ere eee melanesiae Shelf. (4) 19. Margined pale laterally. BY ELAND SHAW. 189 (23) 20. Testaceous or ferruginous. (22) 21. Postero-lateral angles of the distal abdominal tergites of ¢ DEOGUICEGN Wire tsth tis ita o cs evens olweW alan cs SNAP pthc avoRer wileh s Mee, cae veka totter heydeniana Sauss. (ZL) BRo INNOE KO sccocsessioceccrsscac0c0nccnoccusnvogGooUDuO CaO sedilloti Bol. (20) 23. Castaneous. : (25) 24. Spine at the base of genital styles ...................... philpotti Shaw. (24) 25. No spine at the base of genital styles .................. communis Tepp. (3) 26. Vestigial tegmina not free at the tips, but represented by a sulcus; subgenital lamina of o& asymmetrical. Cv vMarcainedspaleplaterallliya race -ir-rienlel clement olels oielier ater eraten ate uncinata Shaw. (27) BB, INGE wares! TRIS IeneemeVhy 5oscceaceocancccccn0dn500K0 illingworthi Shaw. (2) 29. Apterous. Gi) P30, Marrined wpalewlateralliys esq 5 caches cee er aketetre le otis ols) cls sina tepperi Shaw. (30) 31. Not margined pale laterally. CSP oe ee SLE U21T) OU rey seen sare ea ok oe eter ureitreseice ue tessa levis tocogbetic vakeatietishvosiel ala) Gy eureueatys spryi Shaw. S20) ee aa EAL COOUS 25 a reyrcclaye holersie eR Gus Suave AUN Oe Gi Gila, SESS wea eis eedeeen Gere aetts parva Tepp. (1) 34. Transversely banded. (38) 35. Nigro-castaneous, banded pale. (37) 36. Large species (about 22 mm.), with broad interrupted PASCIASMOL NOTA Ve. hevaniinesae eaeta Sho aneele ane aun easter eae subbifasciata Tepp. (36) 37. Small species (about 12 mm.), abdominal tergites 8 and 9 only thandedeepacareh Ah Soe hele es eas eee eee feriarum Shaw. (35) 38. Testaceous, banded dark. (40) 39. Dise of pronotum with 3 fuscous vittae arranged triangu- larly, tegminal vestiges completely separated ...... secunda Tepp. (39) 40. Dise of pronotum with large black central macula, tegminal vestiges not completely separated .................. nigrofasciata Shaw. SDECieS MNGCERtAe: SCdIS) craacisarire cust WAGs Reber Lae brunni Alfk. CUTILIA OBSCURA Tepper. Periplaneta obscura Tepp., Trans. Roy. Soc. 8S. Aust. xvii, 1893, p. 107.—Not Platyzosteria obscura Tepp., Shelford, Trans. Ent. Soc. London, 1909, p. 280, Pl. viii, Fig. 19; Faun. Sudwest. Austr. ii, Lief 9, 1909, p. 135. I have examined Tepper’s type in the South Australian Museum, and it is certainly a Cutilia Stal, the caudal metatarsi being typical; the tergites, except the pronotum, are tuberculate; the coxal borders are concolorous; the vestigial tegmina are complete; and the tips of the valves of the subgenital lamina are visible dorsally. Shelford’s description of the Western Australian ¢ which he supposed to be of this species, was probably written from quite a different cock- roach. He would hardly have overlooked the tarsal structure, or the tuberculate tergites; and while Tepper’s species is but 15.0 mm., Shelford’s is over 20.0 mm. long, and has pale bordered coxae. Tepper’s species appears to be a good one, and differs from any other Cutilia known to me. CUTILIA SCABRIUSCULA Tepper. Periplaneta scabriuscula Tepp., Trans. Roy. Soc. 8S. Aust., 1893, p. 108.— Syntomaptera scabriuscula Tepp., Kirby, Syn. Cat. Orth. Brit. Mus., i, 1904, p. 130.— Platyzosteria scabriuscula Tepp., Shelford, Trans. Ent. Soc. London, 1909, p. 280. A good species, widely distributed from Western Australia to Victoria; but an examination of Tepper’s types in the South Australian Museum shows that the posterior metatarsus is longer than the remaining segments combined, is biseriately spined beneath, and has an apical pulvillus; and that the species should go into Stal’s genus Cutilia. Mr. A. M. Lea has taken the species from ants’ nests at Port Lincoln and Goudie, South Australia. 190 NEW GENERA AND SPECIES OF BLATTIDAE, CUTILIA TEPPERI Shaw. Cutilia tepperi Shaw, Mem. Q. Mus. vi, 1918, p. 157.—Drymaplaneta circum- cincta Tepp. MS. I examined Tepper’s types of circumcincta in the South Australian Museum. There are two specimens labelled by him “Type of @’’, and “Type of 9”. These are both 2, and were relabelled by me. I could find no records, and I think it is clear that he never published any description under his MS. name. CUTILIA COMMUNIS Tepp. Drymaplaneta communis Tepp., Trans. Roy. Soc. S. Aust. xvii, 1893, p. 110.— Methana antipodum Brancsik, Jahr. ver. Trencsin. Com. xix-xx, 1897, p. 58, Pl. i, fig. 4 (teste Shelford). I have examined Tepper’s types in Adelaide, and these agree with his descrip- tion and type localities; but the posterior metatarsus is biseriately spined beneath, the pulvillus is not quite apical but extends a short distance upwards, and the whole segment, though long, is not longer than the other segments combined. This species is typical of a group lying between Platyzosteria and those undoubted examples of Cutilia with the long posterior metatarsus and apical pulvillus. There are several species which would fall into this group, and although time and oppor- tunity do not permit of their being indicated in the present paper, I would venture to suggest the name Cutiloidea as being appropriate for an intermediate genus distinguished from Platyzosteria by the presence of a biseriate row of spines beneath the posterior metatarsus, and distinguished from Cuwtilia on the other hand by the fact that this row of spines is not the whole length of the segment, the distal half of which, approximately, is occupied by a backward production of the pulvillus; the genotype being communis Tepp. There is very little variability in the form of the supra-anal lamina of the 9 in Cutilia and this, if found correlated to a special tarsal structure, may prove a useful character. The ¢ communis has a gland opening on the Ist abdominal tergite, situate in the middle line, close to the anterior margin, and concealed by the metanotum. Shelford (1909) has pointed out that an allied species semivitta Walk. possesses a similar structure, and this species is also a Cutilia. In colour communis varies from nigro-castaneous to fuscous, and in the larva the disc of the tergites is often much paler. In them also the pale thoracic border is found all round the abdomen, and in some adults this persists in the form of lateral yellow spots on tergites 1 to 4. There is a striking superficial resemblance between some larvae of this species, larvae of Methana marginalis, and the adult form of Cuwtilia sedilloti Bol., but sedilloti has tegminal vestiges, and paler legs; and the prolonged postero- lateral angles of the meso- and metanotum of marginalis (the rudiments of the flying organs) readily separate them. Tepper’s use of the word “Cotype” is indefinite; he certainly did not use it in the restricted sense of a Primary Type, nor does he mean what is usually understood by “Paratype,” for I have in my collection specimens marked “Cotype” by Tepper himself, captured at localities not mentioned in the original descrip- tion and on dates long subsequent. What I think he meant was specimens com- pared by him and agreeing with his type; but then we know that most of his Blattid types were not so designated by him until 1915, many years after he had published his descriptions. BY ELAND SHAW. 191 CUTILIA CIRCUMDUcTA Walker. Periplaneta circumducta Walk., Cat. Blatt. Brit. Mus., 1868, p. 143. Specimens taken by Dr. Mjoberg at Broome and in Kimberley district, Western Australia, and by Mr. N. B. Tindale at Groote Eylandt in the Northern Territory are perhaps this species. The vestigial tegmina are completely articulated, there are no wing vestiges, and the posterior metatarsi are of the Cutilia type. Larvae of this species closely resemble those of C. communis Tepp., but the tarsal structure of the latter is of the Cutiloidea type, the yellow border disappears from the abdomen either completely or partly at maturity, and the face is immaculate. Whether these specimens should be referred to Walker’s circumducta or not, they certainly are not Tepper’s submarginata (q.v.) which has posterior metatarsi of the Platyzosteria type. CUTILIA ILLINGWORTHI Shaw. Cutilia illingworthi Shaw, Proc. Linn. Soc. N.S.W. xlvii, 1922, p. 227, Text-fig. 4. New Localities—Victoria: Mallee district. A small 2 (No. 347, Coll. Shaw) 18.0 mm. long, sent to me in 1920 by the late F. P. Spry, is probably this species from an unexpected locality. Lord Howe Island: One small 92, Macleay Mus., Sydney. CUTILIA PARVA Tepper. Periplaneta parva Tepp., Trans. Roy. Soc. S. Aust., 1895, p. 162.—Platyzosteria parva Tepp., Shelford, Gen. Ins., Fasc. 109, 1910, p. 7. The type of this species is in the South Australian Museum and agrees well with the description. The posterior metatarsus is typically that of Cutilia Stal, the insect is completely apterous, the pale segments of the antennae can be plainly seen, and the ootheca is not longitudinally fluted. CUTILIA PULCHRA, n.sp. Text-fig. 15. 6. Head yellow, a broad castaneous streak extending from the vertex to the margin of the clypeus; labrum castaneous, slightly emarginate; antennae longer than the body, fuscous. Pronotum exposing the vertex, nigro-castaneous, almost black, with two broad yellow transverse bands; mesonotum and metanotum similarly coloured, but with only one yellow band; margins of all the bands Text-fig. 15. Cwutilia pulchra Shaw. co. Apex of abdomen, ventral aspect. Drawn from the holotype. irregular; lateral margins of all the thoracic tergites narrowly castaneous; ves- tigial tegmina completely separated, apex obliquely truncate, coarsely punctate, with a dark macula occupying the anal portion; abdominal tergites rich rufo- castaneous, except laterally, where on tergites 2 to 7 a large yellow macula is situate, the area surrounding this macula being darker than the disc; postero- 192 NEW GENERA AND SPECIES OF BLATTIDAE, lateral angles of tergites 5, 6 and 7 backwardly produced, lateral margins entire. Legs with the coxae sordidly greyish, all the coxal ridges with their surfaces spinose; coxal borders paler, and a dark macula occupying the upper portion of the middle and posterior coxal grooves; tibiae castaneous; tarsi testaceous, posterior metatarsus longer than the remaining tarsal segments combined, bi- seriately spined beneath, pulvillus apical, remaining segments unspined, with large pulvilli; ungues missing, Supra-anal lamina subquadrate, widely emarginate, fimbriate, lateral margins serrate; subgenital lamina (Text-fig. 15) subquadrate, posterior margins straight; styles lateral, castaneous, a large strong spine at the base of each, nearly as long as the style, distal half of spines yellow; cerci longer than the lamina. Length.—19.0 mm. Type material—Holotype ¢, Macleay Mus., Sydney. Habitat—New South Wales. Note.—This unique specimen, which is not included in the above key, is in the Macleay Museum, Sydney, with no other label than N.S.W.; and when fresh its rich colouring must have made it a very striking species. CUTILIA SECUNDA Tepper. Leptozosteria secunda Tepp., Trans. Roy. Soc. S. Aust., 1894, p. 183.—Polyzos- teria triangulata Br. v. Wattenwyl, Ann. Mus. Civ. Genova, Ser. 2a, xiii, 1893, p. 33; Krauss., Denkschr. med-nat. Ges. Jena, viii, 1903, p. 750, Pl. Ixvii, fig. 1 (fide Shelf.) —Cutilia triangulata Shelford, Trans. Ent. Soc. London, 1909, p. 291. Brunner vy. Wattenwyl’s few words, referring to an undescribed species, cannot be regarded as a sufficient diagnosis; but Tepper’s description is good and well detailed. That Krauss gives a good description and figure under a wrong name does not make that name right. The inappropriateness of name which Shelford (1909) condemns was quite appropriate to Tepper when he placed the species as a second one in his genus Leptozosteria; and anyhow a consideration of this kind should not weigh in a question of priority. I think the explanation is that Shelford showed, here and there, evidence of a not unnatural irritation at what he considered to be Tepper’s want of courtesy. I have examined Tepper’s type of secunda and the posterior tarsal structure is typically that of Cutilia. This species sometimes occurs in houses. New Localities —Queensland: Laura (T. G. Sloane, 1916), Townsville (F. H. Taylor, ex house), Gordonvale and Babinda (J. F. Illingworth, 1917), Clerment (Miss B. Carne, 1920), Cairns (J. F. Illingworth, ex house). New Guinea and Darnley Island (Macleay Mus., Sydney). CUTILIA SUBBIFASCIATA Tepper. Drymaplaneta subbifasciata Tepp., Trans. Roy. Soc. 8S. Aust. xvii, 1893, p. 112.— Platyzosteria subbifasciata Shelf., Trans. Ent. Soc. Lond., 1909, p. 286.—Cutilia subbifasciata Shaw, Vict. Nat. xxxi, 1914, p. 106. I have examined the specimens from the Northern Territory marked by Tepper as his types. They are a larval ¢ and a larval 9; both belong to the Genus Cutilia — Stal, and both have the dorsal pale lateral border continued around the abdomen. This species varies in colour, and in some specimens, as in Mr. Tindale’s from the Roper River, most of the pale colour is absent; it also affords another example of the occasional persistence in the adult of the pale border of the abdominal tergites usually lost at the last ecdysis. In the Macleay Museum in Sydney the BY ELAND SHAW. 193 locality “N.S.W.” on two old specimens must be regarded as doubtful, and a third, a larval 2, is labelled ‘“Polyzosteria macleayi Sauss., Port Darwin,’ a name I am unable to trace, and one not mentioned by either Shelford or Kirby. New Localities——Port Darwin and Melville Isd. (Baldwin Spencer, Aug., 1912. Nat. Mus. Melb.); N.S.W. (Macleay Mus.); Roper River, N. Terr. (N. B. Tindale, S. Aust. Mus.). Genus CosMOZOSTERIA Stal. The species of this genus are difficult to separate as, on the whole, there is little difference in structure between them. The supra-anal lamina of the ¢ in all the species is straight and fimbriate posteriorly, with deflected spines at the postero-lateral angles, except in C. picta Tepp. (= gloriosa Shelf.) where it is slightly convex and unspined; the subgenital lamina of the ¢ is in all widely emarginate with strong divergent spines (less marked in picta Tepp.). All the species are castaneous, and all, with the exception of C. froggatti Shelf., have prominent flavid markings. The flavid markings vary much in some species, the larvae showing them more than the adults. In C. zonata Walk., for example, while the adult has simply 3 or 4 transverse bands, the larva may have a lateral flavid border practically all round. I had hoped to give a satisfactory key to the species, but the more I examine them the greater the uncertainty, and now I have only time to offer a few remarks about some of the species. I have examined Tepper’s types and his (. maculimarginata is probably a distinct species, though his “Type of g” and “Type of 9” are both female larvae, and in this genus it is not wise to describe new species unless from an adult. His C. picta is a good species and Shelford’s gloriosa, also described from a 9, is identical with it. His “Cotype,” also an adult 9, is not conspecific with the type, but I believe it to be identical with a species of which I have several specimens from North Queensland, and which I here describe as C. sloanei. Tepper’s subzonata is also, I think, a distinct species, certainly not bicolor Sauss. (as Shelford thought), nor identical with the species occurring around Brisbane which I (1918) mistook for it. This last species does not appear to agree with any of the others, though nearer to C. subzonata Tepp., and I here describe it as C. brisbanensis. Another curious point about this genus is that in a very large proportion of the ¢ the subgenital lamina is found to be deformed, one postero-lateral angle (commonly the right one) being rounded instead of being produced into the usual divergent spine; or one spine may be much longer than the other. OC. froggatti Shelf. and ©. lateralis Walk. differ structurally from the remainder in the abdominal tergites not being scabrous, but closely covered with shallow pits. As previously pointed out, the pale marginal colour of the larva is to a large extent lost in the adult. In C. zonata Walk. all the larvae in my collection show a yellow anterior pronotal border, and in several the yellow border is carried practically all round the insect. When maturity is reached, only the three bands on the posterior borders of the thoracic tergites are left, this being Walker’s (1868) trifasciata; but sometimes the anterior pronotal band persists and his (1868) quadrifascia appears. Then again, in other species such as C. sloanei mihi, the pronotal yellow margin in the adult varies in extent, sometimes becoming obsolescent, leaving only a couple of small antero-lateral bands. Amongst some specimens from Northwest Australia of a species near C. zonata Walk. are a § and ? in the Stockholm and Sydney Museums respectively, with 194 NEW GENERA AND SPECIES OF BLATTIDAE, the castaneous dise of the pronotum completely surrounded by a broad yellow band, the remaining specimens having a quadrifasciate form. These latter are probably the normal form of an undescribed species, and those with the complete band a variety. Several specimens of the normal form were taken by Mr. N. B. Tindale at Maria Island and the Roper River in the Gulf of Carpentaria, and one by Dr. Mjoberg at Broome in Western Australia. COSMOZOSTERIA BRISBANENSIS, DN. Sp. Cosmozosteria subzonata Shaw, Mem. Q. Mus., 1918, p. 162. Near subzonata Tepp., but is smaller, the line of demarcation between the dark dise of the thoracic tergites and their yello6w posterior border is clean cut, the lateral yellow spots of the abdominal tergites are smaller, only occupying the cephalic half of the tergital width, and the postero-lateral angles of the meso- and metanotum are not so much produced caudally. The proportion of the two species as to length and width is about the same, and both are relatively broader than bicolor Sauss. The laminae and tarsi are of the Cosmozosteria type. Length.—@, 20.0-22.5 mm.; 9, 22.5-27.0 mm. Type material.—Holotype ¢, No. 0/2893; allotype 9, No. 0/2893a; and para- types, 32. Coll. Q. Mus. Habitat—Queensland: Brisbane, Ipswich, Rosewood, Mt. Tambourine, Christmas Creek, Jandowae, Upper Burnett district. COSMOZOSTERIA SLOANEI, N. SD. Rich castaneous. Pronotum relatively wider than that of C. bicolor Sauss.; thoracic tergites with a few scattered shallow pits; abdominal tergites scabrous; angles of the 9th lobiform, yellow; supra-anal and subgenital laminae and tarsi of the Cosmozosteria type. A broad yellow border surrounding the front and sides of the pronotum, except at the postero-lateral angles which are castaneous; this border is continued on the meso- and metanotum as large irregular maculae, not reaching the postero-lateral angles, and similarly on abdominal tergites 2 to 7. Coxal borders narrowly yellow. Length.—Z, 20.0-22.5 mm.; 9, 22.5-27.9 mm. Type material.—Holotype g, No. 0/2894; and allotype 9, No. 0/2894a. Coll. Q. Mus. Habitat.—North Queensland: (Holotype) Laura near Cooktown (T. G. Sloane, July, 1916); (Allotype) Kuranda; Townsville (G. F. Hill); Herbert River, ex beach; Cairns and Gordonvale (J. F. Illingworth). Note.—This species varies considerably in the extent of the yellow margin. In the types it is an almost continuous band, though never so extensive as in C. bicolor Sauss., where it is quite continuous round the thoracic tergites at least; whilst in some specimens of C. sloanei the posterior one-third to two-thirds of the lateral borders of the thoracic tergites is castaneous, and in others the posterior end of the yellow margin turns slightly inwards upon the pronotum. ~ Where the yellow margin of the thoracic tergites is diminished the yellow maculae on the abdominal tergites are replaced by small dots. Those specimens with less yellow are usually larger, and I may be confusing two distinct species, but I regret that the short time at my disposal compels me to leave the genus in a rather chaotic state. : BY ELAND SHAW. 195 Genus ANAMESIA Tepper. ANAMESIA ORNATA Tepp. Pseudolampra ornata Tepp., Trans. Roy. Soc. 8. Aust., 1892, p. 98. There is no doubt but that this beautiful species should be placed in Anamesia. None of the abdominal tergites is produced, and none of the thoracic tergites has a thickened margin. Habitat.—S. Australia: Murray Riv. (H. S. Cope, S. Aust. Mus.). Note.—With regard to the non-production of the postero-lateral angles of the distal abdominal tergites in Anamesia Tepp. and Desmozosteria Shelf., this char- acter should not be taken as absolute, but only as compared with the angles in allied genera. Even within the limits of one species such as A. lambii Tepp., specimens in my collection show a variation from an evident though slight production, to an entire absence of any. ANAMESIA PUNCTATA Tepper. Pseudolampra punctata Tepp., Trans. Roy. Soc. 8S. Aust., 1893, p. 97. Described by Tepper from specimens obtained near the Fraser Range, Western Australia, by the Elder Expedition in 1891, and Shelford mentions it as from Tennant’s Creek, South Australia. There are specimens in the South Australian Museum from other localities in South Australia, and Mjoberg in 1910-13 took it in Kimberley district, and at Derby in Northwest Australia. The species seems to vary considerably in colour, some specimens being much paler than others, and in some the punctures are almost obsolete. It is possible that A. walkeri Shelf. may be a very pale example. Genus DEesmozosTErRIA Shelford. DESMOZOSTERIA CINCTA Shelf. Text-figs. 16 and 17. Desmoczosteria cincta Shelf., Trans. Hnt. Soc. Lond., 1909, p. 303. Shelford described the @ from Central Australia, and the type is in the Senckenberg Museum. A ¢ in the Queensland Museum from Charleville in the far west of Queensland is, I believe, conspecific, and descriptive notes supplementary to Shelford’s description are added. Coe ee \ Ne Se Sart ee | 16 A 17 Text-fig. 16. Desmozosteria cincta Shelford. o. Apex of abdomen, dorsal aspect. Drawn from the allotype. Text-fig. 17. Desmozosteria cincta Shelford. o. Apex of abdomen, ventral aspect. Drawn from the allotype. Allotype ¢. lateral margins of the abdominal tergites also ochreous, tips of the angles of the 9th tergite ochreous; 6th abdominal sternite sinuate posteriorly; supra-anal lamina (Text-fig. 16) subquadrate, angles obtuse, widely emarginate, fimbriate, extending to about ? the length of the cerci, rufo-castaneous, with the posterior border widely ochreous; subgenital lamina (Text-fig. 17) 196 NEW GENERA AND SPECIES OF BLATTIDAE, quadrate, nearly black, lateral margins concave, posteriorly widely emarginate, styles lateral. Posterior metatarsus not so long as the remaining segments combined, not spined beneath, with the pulvillus extending upwards the whole length of the segment in the form of a compressed ridge. Type material.—Allotype 3, No. 0/2897. Coll. Q. Mus. Habitat Queensland: Charleville (L. Franzen). DESMOZOSTERIA GROSSE-PUNCTATA Shelf (loc. cit.). Shelford did not know the locality of his type in the Oxford Museum, but presumed it to be Australia. Basedow’s specimen, however, agrees almost perfectly with Shelford’s description, and his locality brings this species into line with the rest of the genus. Habitat—Western Australia, Dr. H. Basedow, one larval 9, No. 40060, Aust. Mus., Sydney. DESMOZOSTERIA ZEBRA Tepper. Polyzosteria zebra Tepp., Horn Exped. Cent. Aust., Vol. 2, 1896, p. 362.—Platy- zosteria zebra Shelf., Trans. Ent. Soc. Lond., 1909, p. 288—Cosmozosteria zebra Kirby, Syn. Cat. Orth. Brit. Mus., Vol i, 1904, p. 133. Shelford was in doubt as to the generic position of this species, and, although I have seen the type, I have some uncertainty in placing it in Desmozosteria Shelf. Tepper’s description is good as far as it goes, but the margins of all the thoracic tergites are definitely thickened, the postero-lateral angles of the 5th abdominal tergite are not produced, of the 6th scarcely, and of the 7th not much. In D. zebra the cerci are not short nor flattened, the spines of the tibiae are triseriately arranged on the outer aspect, and the caudal metatarsus is as long as the remaining segments combined, is unspined beneath, and bears a long pulvillus; the other segments have large pulvilli, and the supra-anal lamina is rounded, but damaged. The angulation of the caudal tergites, as a character, does not appear to be quite reliable, for in some species of Cosmozosteria the angles of the 5th tergite are a little produced, and those of the 6th usually more so; and in Anamesia and Desmozosteria the 7th tergite sometimes has a slight backward production. Genus TEMNELYTRA Tepper. This quite distinct genus includes four species, of which one is added in the present paper. In all, the posterior metatarsus is shorter than the remaining tarsal segments combined, and in two of the species, truncata Br. v. W. and subtruncata Tepp., there is a remarkable modification of the tarsal claws; the external claw being abortive and not so long as the somewhat asymmetrical arolium, to the outer side of which it appears to be attached. It is only with considerable magnification that the abortive claw can be distinguished, the tarsus appearing to terminate in a single claw and large arolium. In specimens in which only four tarsal segments are found, the first (metatarsus) and second are longer than normal, and the metatarsal pulvillus extends some distance upwards. The first abdominal tergite of the ¢ bears an organ (“gland opening’’) which varies in each species, but always takes the form of a symmetrical ‘tumefaction bearing a tuft of bristles on its cephalic aspect. Shelford (1909) figures the tergite of 7. truncata. BY ELAND SHAW. 197 Key to the species of Temnelytra. (4) 1. Metanotum not bearing vestigial wings, claws asymmetrical. (3) 2. Larger size, testaceous, abdominal tergites with lateral WOO? XC bocobsesoceuooaoood oMdbo Ooo abobsoMoOS truncata Br. v. W. (2) 38. Smaller size, fuscous, abdominal tergites concolorous ..... subtruncata Tepp. (1) 4. Metanotum bearing vestigial wings, claws symmetrical. (6) 5. Pronotum with two prominent fuscous vittae ............. undulivitta Walk. (5) 6. Pronotum fuscous, with pale lateral borders ............. tryoni Shaw. TEMNELYTRA TRUNCATA Brunner v. W. Polyzosteria truncata Br.v. W., Nouv. Syst. Blatt., 1865, p. 217—Temnelytra harpuri Tepp., Trans. Roy. Soc. 8S. Aust., 1893, p. 39.—Zonioploca harpuri Kirby, Syn. Cat. Orth. Brit. Mus., Vol. i, 1904, p. 137.—Temnelytra truncata Shelf., Trans. Ent. Soc. Lond., 1909, p. 305, Pl. ix, figs. 37a and 37b. There is some question as to Tepper’s types of T. harpuri as the specimen labelled “Type of g” is really a 9, captured after the publication of his description, and although he described both sexes, there is no @ in the South Australian Museum. However, this species is almost certainly identical with Brunner’s T. truncata of which he described the 2 in 1865, and Shelford (1909) figured parts of the ¢. ' The tarsal claws of this species are asymmetrical, the outer claw being modified. Brunner v. Wattenwyl (1865), in addition to New South Wales, gives New Zealand as a locality for this species. This is possibly an error, Brunner not differentiating the New Zealand species 7. undulivitta subsequently described by Walker (1868). TEMNELYTRA SUBTRUNCATA Tepper. The types of ¢ and 9 are in the National Museum, Melbourne, and this is quite a distinct species. The first abdominal tergite of the ¢ bears a medial gland opening surmounted by bristles, resembling that of truncata Br.v. W., but quite different from that of uwndulivitta Walk. (q.v.). This species and truncata share several characters in common, such as the absence of wing vestiges, similarity of structure of the first abdominal tergite, and asymmetry of the tarsal claws. TEMNELYTRA TRYONI, n. sp. Text-figs. 18 and 19. ¢. Head testaceous, with the vertex dark fuscous, and a dark longitudinal band, narrower in the middle, occupying most of the frons; antennae fuscous, hirsute. Pronotum with the disc dark fuscous laterally, bordered yellow, the extreme margin fuscous. Tegmina pale fuscous bordered yellow, extending to the posterior margin of the first abdominal tergite. Venation, except the vena Text-fig. 18. Temnelytra tryoni Shaw. oc. Apex of abdomen, dorsal aspect. Drawn from the holotype. Text-fig. 19. Temnelytra tryoni Shaw. oc. Apex of abdomen, ventral aspect. Drawn from the holotype. Text-fig. 20. Scabina antipoda Shelford. oc. Left tegmen. 198 NEW GENERA AND SPECIES OF BLATTIDAE, dividens, obscure. Wing vestiges lateral, lobiform, separated from the metanotum. Abdominal tergites fuscous with pale borders; 1st abdominal tergite with a medial, broad-based triangular tumefaction furnished cephalically with bristles; 7th tergite sinuate caudally. Supra-anal lamina (Text-fig. 18) transverse, fimbriate, widely emarginate, lateral margins slightly concave. Subgenital lamina (Text- fig. 19) transverse, caudal margin concave; styles smooth, long, not sharply pointed. Abdominal sternites fuscous. Legs testaceous; coxae with some fuscous markings; posterior metatarsus nearly as long as the remaining segments com- bined, biseriately spined beneath, pulvillus short, claws symmetrical. 9. Larva. In general colouring similar to the adult. Postero-lateral angles of the mesonotum scarcely, and of the metanotum well produced backwardly. Subgenital lamina of the usual Blattine larval form. Length—20 mm. (approximately). Type material.—Holotype ¢, No. 0/2896 (in two parts); and paratype, larval 9, No. 0/2896a. Coll. Q. Mus. Habitat—Queensland: National Park, Lamington Plateau (3,000 ft., H. Tryon, Jan., 1917). Note.—The holotype is rather damaged, and the apex of the abdomen is on a separate card; but the species is quite distinct and the paratype is in more perfect condition. TEMNELYTRA UNDULIVITTA Walker. Periplaneta undulivitta Walk., Cat. Blatt. Brit. Mus., 1868, p. 144. This species much resembles another New Zealand one, Cutilia sedilloti Bol. (1882), from which its tegminal structure at once distinguishes it. In common with the new species 7. tryoni mihi it has symmetrical tarsal claws, a fairly similar form of ist abdominal tergite, and wing vestiges separated from the metanotum; although in the only ¢ in my collection the postero-lateral angle of the metanotum presents a well marked notch, and the wing vestige, instead of being separate, is only evidenced by a lateral curved crumpling. The subgenital lamina of the larval 9 is of the usual Blattine form. New Localities—Chatham Is. (1887, J. J. Fougére, teste F. W. Hutton). New Zealand: Invereargill, The Hump, 3,500 ft., 1912; Longwoods, 1913; and Hunter Mts., near Green Lake (under stones, 1919, A. Philpott). Genus Scasina Shelford. SCABINA ANTIPODA Shelford. Text-fig. 20. Pelmatosilpha (?) antipoda Kirby, Ann. Mag. Nat. Hist. Ser. 7, xii, 1903, p. 376.—Scabina antipoda Shelf., Trans. Ent. Soc. Lond., 1909, p. 306. Kirby described this species from Queensland, but neither he nor any subse- quent writer appears to have noticed a strong spine which extends forwards and outwards from the antero-external angle of the quadrate tegmen (Text-fig. 20). This is present in the ¢ only, the ? having no trace of it. The object of this spine is to me obscure, and I can only suggest that it acts as a support to the angles of the pronotum, which is very wide, and which in the ¢ can get no support from the humeral angle of the tegmen, since in that sex the costal portion of the tegmen is deflected. This deflection is not found in the 92, nor is the spine. This species occurs commonly under the bark of trees in the Rain Forest areas of South Queensland, at an elevation of about 2,000 ft. BY ELAND SHAW. 199 Genus Meruana Stal. Shelford (1909) includes eight species in this genus; Hanitsch (1915) added semimarginalis from Borneo, and to these, three are added in the present paper, making twelve in all, including pallipalpis Serv..of which the type is lost and the position uncertain. Of two of these species, semimarginalis Hanitsch and hosei Shelford, only the 2 is known; but in the remaining nine (excluding pallipalpis Serv.) the supra-anal lamina of the ¢ does not show any great variation in form. It is usually subquadrate, fimbriate, and with some emargination; in curvigera Walk. it is medially carinate, in marginalis Sauss. it is entire or has a small medially-situate backward production, in sjéstedti mihi it has a deep angular emargination, and in convexa Walk. it is almost bilobate. The subgenital lamina of ¢ is subquadrate and its posterior margin may have a wide, shallow, rounded emargination as in marginalis Sauss., soror Sauss. and mjdbergi mihi; may be bilobate as in sjéstedti mihi; deeply and angularly emarginate with divergent postero-lateral angles (resembling Cosmozosteria) as in curvigera Walk. and parva mihi; or with blunt projections at the base of the styles as in convera Walk. In all the species known to me the styles are long, slender, slightly tapering, and not very sharply pointed, with the exception of sjdstedti mihi whose styles are very different from those of the other species, being broad, flattened and acuminate, somewhat resembling a Malay “kris.” Ootheca. In the Queensland Museum there is a 9 M. marginalis Sauss. with an attached ootheca carried suture downwards, and this is the only ootheca of the genus:I have seen. I cannot say whether this position is usual, but the position of the ootheca, as a character, is not quite reliable, a 9 Blattella germanica L. in my collection, for instance, carrying it uppermost, although the lateral position is given as a generic character. In the key to the species below, the form of the laminae has not been used, as this is not constant enough, and relates to the one sex only. The presence or absence of a flavid humeral streak, the colour and markings of the pronotum, and particularly the dark markings of the vertex and frons afford valuable characters, P and apply to both sexes. aN x Key to the species of Methana. (4) 1. Uniform castaneous. (S)) Ba Ibayescre SAE OniGe BO rb Fooaaduaoncutoo aso UGd000D magna Shelf. (2) 8 Srmenlicie grz, wacko BO whe SGooasucagacndoueoueuLod convexa Walk. (1) 4. Not uniform castaneous. (6) 5. Pronotum testaceous, with fuscous vittae ......... curvigera Walk. (5) 6. Pronotum castaneous, with flavid margins. (14) 7. Pronotum not bordered flavid posteriorly. (13) 8. Castaneous dise of the pronotum immaculate. (10) 9. Tegmina without a flavid humeral streak ........ semimarginalis Hanitsch. (9). 10. Tegmina with a flavid humeral streak. (12) 11. Flavid humeral streak longer than the anal area, dark band of the vertex not continuous with that OLSUINS. LT OMS | Pe castes seas aur) cusiisy: duaitel ancoicocaiersiat ened ek een Galena y's marginalis Sauss. (11) 12. Flavid humeral streak not longer than the anal area, dark band of the vertex continuous with thatiofmthe;: trons! (2. fee ee oe bleoiseicios Seen parva Shaw. (8) 18. Castaneous disc of the pronotum with two flavid dae HOO) Ee1(\5 aR erctcio tomar als Cache OI na a oh ota ara ee eors soror Sauss. (7) 14. Pronotum bordered flavid posteriorly. (16) 15. Tegmina uniformly castaneous .................- hosei Shelf. (15) 16. Tegmina with a flavid humeral streak. 200 NEW GENERA AND SPECIES OF BLATTIDAE, (AB RAT.OWertexiblack tne n/accin 2 lei eictorto erie rape terror icie * mjébergi Shaw. (17) 18. Vertex testaceous. (AD) WOs IMeoMs TV OHENI® csococacocsc0000ccocouD DO S000K0 sjostedti Shaw. (19) 20. Frons with a broad castaneous band ............... papua Shelf. Species incertae (Sedisit sc] pepsi sienete eletele tel ene! terion ile pallipalpis Serv. METHANA CONVEXA Walker. Periplaneta convexa Walk., Cat. Derm. Salt. Brit. Mus., Suppl. Blatt., 1869, p. 152.—Methana convexa Kirby, Syn. Cat. Orth. Brit. Mus., Vol. i, 1904, p. 136.— Methana rufescens Kirby, Ann. Mag. Nat. Hist. (7) xii, 1903, p. 374 (teste Shel- ford).—Paraphoraspis (?) castanea Tepp., Trans. Roy. Soc. 8. Aust., 1894, p. 173. This species is distributed over a long range on the eastern side of Australia from Victoria to North Queensland, and occurs not only near sea level, but at high elevations such as Uralla, N.S.W., 3,500 ft. (W. W. Froggatt) and Tambourine Mountain, 2,000 ft. (auct.). Tepper’s type of Paraphoraspis (?) castanea is in the South Australian Museum, and is a 9 of this species, and came from the most southerly recorded locality, Howbulan, Victoria. METHANA PARVA, n. sp. Text-fig. 21. Closely resembling M. marginalis Sauss., but differing from it in the following points: (1) Size much smaller than marginalis Sauss.; (2) flavid humeral streak not extending beyond the anal area; (3) subgenital lamina of the ¢ longer, deeply emarginate; (4) castaneous band of the frons joining that of the vertex; (5) tegmina not extending beyond the body. The flavid humeral streak is in this species confined to the costal area, not extending, as in marginalis Sauss., to the tips of the 1st or 2nd branches of the radius; and in marginalis Sauss. the costal area is longer than the anal area. The subgenital lamina of the ¢ is in marginalis Sauss. transverse with a straight posterior margin, whilst that of this species (Text-fig. 21) is prolonged back- wards, deeply and widely emarginate, and ends in divergent points. In all my 6 specimens, one of these points is longer than the other, but this appears to be a deformity, the lamina of a larval ¢ being symmetrical. This shortening of one of the postero-lateral angles of the subgenital lamina commonly occurs in the ¢ of the genus Cosmozosteria Stal. In this species the dark band of the frons is always joined to the dark vertex; in marginalis Sauss. this is never so, and a large number of specimens have been examined. This character also serves to separate the larvae. The anal veins all terminate on the vena dividens, none of them reaching the margin of the tegmen. This character is common to all the species of Methana in my collection. Length.—Q, 20.0-21.0 mm.; 9, 20.0-22.0 mm. Type material.—Holotype g, No. 0/2873; allotype 9, No. 0/2873a; 8 paratypes. Coll. Q. Mus. Habitat. Queensland: Green Is., Moreton Bay; Tambourine Mt., 2,000 ft. (auct.); National Park, Lamington Plateau, 3,000 ft. (H. Hacker). Note.—The species which I take to be Saussure’s marginalis is considerably larger than this, averaging 28 or 29 mm. in length, and is common in the Brisbane district. Saussure’s description gives 29 mm. as the length of the 9, and as the two are evidently not conspecific, I am regarding the smaller species as the new one. It is a curious fact, however, that, when in 1914, Prof. E. B. Poulton kindly brought me some duplicates from the Hope Museum, the single specimen labelled BY ELAND SHAW. 201 Methana marginalis Sauss. agreed entirely with my parva, and it seems probable that the two species are mixed up in the older collections. Saussure’s type is in the Paris Museum. Variety HACKERI (with abbreviated tegmina).—In the Queensland Museum there is a 2 (No. 0/2904) taken in Dec., 1921, by Mr. H. Hacker at the Queensland National Park, in which the tegmina are truncate, not extending beyond the 2nd abdominal tergite, and the wings are lobiform. The anal is shorter than the costal area. Text-fig. 21. Methana parva Shaw. co. Apex of abdomen, ventral aspect. Drawn from the holotype. Text-fig. 22. Methana mjébergi Shaw. ¢. Apex of abdomen, dorsal aspect. Drawn from paratype No. 0/2874. Coll. Q. Mus. Text-fig. 23. Methana mjobergi Shaw. o. Apex of abdomen, ventral aspect. Drawn from paratype No. 0/2874. Coll. Q. Mus. METHANA MJOBERGI, n. Sp. Text-figs. 22 and 23. Vertex testaceous, frons occupied by a broad, dark, castaneous streak with irregular edges, and broadening below into a curved, transverse marking; antennae fuscous, labrum bilobate. Pronotum with wide, flavid, anterior and lateral margins, widest at the postero-lateral angles, posterior margin narrowly flavid, disc dark castaneous, paler centrally, and extending in a small medial angular projection towards the vertex. Tegmina castaneous, about the length of the body, with a flavid humeral streak. Legs testaceous, spines brown, tibiae of the ¢ brownish. Posterior tarsi typically Methana. Arolia large. Abdomen fusco- castaneous, lateral margins of the tergites paler. Supra-anal lamina (Text-fig. 22) of the ¢ quadrate, fimbriate, apex widely and slightly concave; of the 9 more pro- duced and more deeply emarginate. Subgenital lamina (Text-fig. 23) of the ¢ subquadrate, widely and roundly emarginate; styles long, round, slender. Length.—@, 20.0-22.0 mm.; 2, 20.0 mm. Type material—hHolotype ¢ and allotype 9, Stockholm Mus.; two paratypes. No. 0/2874. Coll. Q. Mus. Habitat —N. Queensland: Malanda (2,400 ft.); Cedar Creek (E. Mjoberg, 1910-13). METHANA SJOSTEDTI, N. Sp. Head testaceous, immaculate; antennae pale fuscous; labrum bilobate. Pro- notum with a broad flavid margin all round, narrowly edged with fuscous, widest at the postero-lateral angles, disc occupied by a bilobate castaneous macula. Tegmina castaneous, shorter than the body, with a flavid humeral streak extend- ing to more than half their length. Legs dark testaceous, spines brown, tibiae of the ¢ darker, except at their upper and outer aspect; posterior tarsi typical of the genus, arolia large. Abdomen dark castaneous above and below. Supra- anal lamina of the § quadrate, deeply and angularly emarginate, angles sharp; 202 NEW GENERA AND SPECIES OF BLATTIDAE, of the @ produced, widely and roundly emarginate; subgenital lamina of the ¢ subquadrate, with a shallow emargination and rounded angles, being almost bilobate; styles long, wide, flattened, acuminate. Length.—2, 25.0 mm.; 9, 25.5 mm. Type material.—Holotype ¢ and allotype 9, Stockholm Mus.; 2 paratypes ¢, Queensland Mus., No. 0/2899. Habitat.—N. Queensland: Atherton (2,466 ft., EH. Mjoberg 1910-13), Stockholm Mus.; Kuranda (F. P. Dodd and J. F. Illingworth), Queensland Mus. Note.—This species is near soror Sauss. and mjébergi mihi; the flavid humeral streak is intermediate in length between the two, the flattened styles and shorter tegmina distinguish it from either, and the vertex and frons are immaculate, whilst soror Sauss. has a dark macula on the vertex, and mjobergi mihi one on the frons. M. parva mihi is smaller; the flavid humeral streak is very short, and both vertex and frons are maculate. Genus TRYONICUS nov. Distance between the eyes on the vertex considerably greater than that between the antennary sockets. Posterior tibiae triseriately spined on their outer aspect. Posterior metatarsi at least as long as the remaining segments combined, biseriately spined beneath; remaining segments spined beneath; all the posterior tarsal pulvilli small, apical; arolia absent. Tegmina of the ¢ vestigial, lanceolate, extending beyond the metanotum; wings vestigial, lateral, membraneous, folded; © entirely apterous. Note.—It would appear that a new genus is required for the reception of the curious cockroach described below, with its wide apart eyes, long lanceolate tegminal vestiges, spined tarsi and absence of arolia. It should perhaps come into the Blatta group of the Blattinae, but, as it cannot be placed in any of the existing genera, I propose to erect for it the new genus Tryonicus, named after its discoverer, Mr. H. Tryon, the distinguished Government Entomologist of Queensland. TRYONICUS MONTANUS, n. sp. Text-figs. 24, 25 and 26. Black, shining. Vertex and frons black, mouth parts brownish; antennae not so long as the body, hirsute, proximal segments cylindrical, and distal segments pyriform; eyes wider apart on the vertex than the antennary sockets (Text-fig. 25). Pronotum parabolic, truncate anteriorly, exposing the vertex, posterior margin slightly concave, lateral margins narrowly thickened and everted, as are also the lateral margins of the meso- and metanotum, and of the abdominal tergites. Vestigial tegmina of the ¢ long, narrow, completely separated from the mesonotum (Text- fig. 24), thickened margin extending to the apex, in length extending about half- way across the second abdominal tergite; wing vestiges small, lateral, mem- braneous, testaceous, well separated from the metanotum, with the anal portion once folded longitudinally underneath the anterior portion, not folded again, but projecting beyond the outer third of the costal margin; 2 completely apterous. Postero-lateral angles of the abdominal tergites backwardly produced, apices blunt; posterior margin of the 7th abdominal tergite sinuate. Supra-anal lamina of the ¢ trigonal, apex truncate, lateral margins slightly thickened and everted, caudal margin pale, sparsely fimbriate, extending to about half the length of the broad-based cerci, which on the dorsal aspect are flattened, with well defined edges, and on the ventral aspect are convex, rough, and densely hirsute; of the © trigonal, extending to more than three-fourths the length of the cerci (Text- BY ELAND SHAW. 203 fig. 26). Subgenital lamina of the ¢ triangular, styles issuing from well defined notches, the lamina and the styles being sparsely furnished with long hairs; of the 9, of the usual Blattine form, but sharply compressed towards the apex. Dise of the abdominal sternites castaneous. Legs dark castaneous, trochanters and tarsi paler, posterior tibiae of the ¢ with a constriction below the knee, and Text-fig. 24. Tryonicus montanus Shaw. co. Whole insect (x 4). Drawn from the holotype. Text-fig. 25. Tryonicus montanus Shaw. Vertex of head, showing eyes and antennae. ~ Text-fig. 26. Tryonicus montanus Shaw. @. Apex of abdomen, dorsal aspect. Drawn from the allotype. the segment somewhat dilated distally; those of the 9 cylindrical. Inner aspect of the tihiae furnished with a long brush; all the tarsal segments spined beneath, hirsute, and with small pulvilli; posterior metatarsus of the 3 longer, and of the 9 about as long as the remaining segments combined; arolia absent. Length.—d, 11.0-12.0 mm.; 9, 13.0-16.0 mm.; of tegminal vestiges 3.5 mm. Type material—Holotype 4, No. 0/2872; allotype °, No. 0/2872a and 8 para- types, 2 g and 6 9, Coll. Q. Mus.; 1 paratype ¢, Aust. Mus., No. K47293. Habitat— Queensland: Lamington Plateau, National Park (3,000 ft. H. Tryon, 1917-18; A. J. Turner, 1921). New South Wales: Dorrigo (3,500 ft. A. Musgrave, Dec., 1922). Genus SryLtopyGa Fischer von Waldheim. STYLOPYGA SHELFORDI, N. Sp. Black, nitid, entirely apterous. Head with the margin of the clypeus, the labrum, and palpi rufo-fuscous; also the antennae, except the proximal and one or two following segments which are darker. Dorsal surface furnished with hairs set in shallow pits, denser caudally; lateral margins of all the tergites, and posterior margins of the distal tergites with a thin fringe of stiff hairs. Posterior margin of the 6th tergite sinuate; supra-anal lamina of the ¢ sub- triangular, apex emarginate, cerci long and flattened; subgenital lamina of the 6 transverse, setose, posterior margin straight; styles small, slender; a broad fuscous streak on either side of the subgenital lamina. Legs rufo-fuscous; posterior tibiae darker, and somewhat dilated in the ¢; coxal ridges punctate- (0) 204 NEW GENERA AND SPECIES OF BLATTIDAE, coxal borders narrowly pale brownish-yellow; posterior metatarsus long, biseriately spined for its whole length beneath, pulvillus apical; second tarsal segment also spined beneath, arolia moderate in size. Ootheca short, deep, smooth, with eight divisions on each side. Length.—Z, 16-17 mm. 92, 19-20 mm. Type material.—Holotype 4, No. 0/2870; and allotype 9, No. 0/2870a; and several paratypes, Coll. Q. Mus. Paratype with ootheca attached, Coll. Stockholm Mus. Habitat.—S. Queensland: Samford, Capalaba, Green Is., Tambourine Mt. (auct.), Christmas Creek (H. Mjoberg). WN. Queensland: Malanda (H. Mjoberg). Note——This species has been known tc me for several years from various localities in South Queensland, and I welcome the opportunity of paying a tribute to Shelford’s memory by naming it after him. STYLOPYGA PAPUAE, DN. SDP. Closely allied to shelfordi mihi, but is castaneous rather than piceous, the antennae are darker, the coxae are yellowish-brown darkening proximally, the coxal borders are broadly yellow, the tibiae of the ¢ are not dilated, and the pronotum is relatively wider. Length.—, 17.0 mm.; 9, 18.0 mm. Type material.—Holotype ¢, No. 0/2871; allotype 9, No. 0/2871a; and para- type 3g, No. 0/2871b, Coll. Q. Mus. Habitat.—_S. E. Papua: Kui-ara (on the coast; auct. 1914-15). STYLOPYGA IMMUNDA Shelford. Shelford described this species as from Queensland, and says the type is in the Stockholm Museum. My collection contains specimens from the northern part of that State, and amongst the Mjoberg material collected in 1910-13 are others from the same area. S. immunda appears to vary considerably in size, and the colour is nigro-castaneous rather than “piceous.” A large proportion of cock- roaches described as black are seen to be really dark castaneous when examined in a good light and under magnification. To Shelford’s description may be added: Pronotum subtruncate anteriorly, exposing the vertex; no wing vestiges, but postero-lateral angles of the metanotum slightly produced. Sixth and seventh tergites with sparse hairs. Coxal ridges (Shaw, 1922) punctate; posterior metatarsi long, biseriately spined beneath. Ootheca large, smooth, deep, with about 8 divisions on each side. Length.—¢, 21.0 mm.; 9, 22.5-26.0 mm. Habitat.—N. Queensland: Cooktown (T. G. Sloane); Malanda and Bellenden Ker (EH. Mjoberg); Cairns “ex log, ex bark” (J. F. Illingworth). Genus PERIPLANETA Burmeister. PERIPLANETA AMERICANA Linné. This cosmopolitan species is widely spread in Australia, and I also have it from New Guinea, New Hebrides, and Lord Howe and Norfolk Islands. Its habits appear to vary somewhat. In New Guinea I found it common in houses, but australasiae Fabr. was much commoner in ships; whilst in Australia the latter is the common species breeding about dwellings, and, although americana L. is frequently found in houses, it usually flies in from outside and one does not see its larvae about. When both species live together in the same place, australasiae BY ELAND SHAW. 205 Fabr. will probably be found gradually to displace americana L., just as when Supella supellectilium Serv. invades places already occupied by Blattella germanica L., it tends to oust the latter. That P. americana L. can harbour the vibrios of Asiatic cholera has been demonstrated by M. A. Barber (1918). PERIPLANETA AUSTRALASIAE Fabricius. Shelford (“A Naturalist in Borneo,’ 1916, p. 115) states that australasiae as a specific name refers to Australia, whence the older naturalists thought it came; and, on the following page, that the species is only a rare immigrant to Australia. Whatever may have been the case when Shelford wrote, this cockroach is now well established in Australia, and, as to the name, I think Shelford must have been in error in supposing that Fabricius can have used it in any sense other than “of Southern Asia.” The word Australasia as applied to Australia and New Zealand was not in use in 1775 when Fabricius described the species, nor was it commonly so until about 60 years later. Burmeister (1838) queried Fabricius’s locality, and thought the species American, also stating that Fabricius’s records were not infrequently unreliable. In Australia this species is quite common about dwellings, and together with the following new species usually conceals its ootheca by covering it over with mud or some convenient loose debris, and I have seen them numerously placed in chinks of a rough outhouse door, and so covered as to be difficult to discern. The ootheca is frequently attacked by one of the Chalcididae, Geniocerus (Tetrastichus) hagenowi Ratz., and about fifty individuals may emerge from one ootheca. Unlike americana L., the larvae of australasiae Fabr. may be found in all instars in dwelling houses, but I have never seen one which would agree with Shelford’s (1911) figure of Specimen 1.13767 from the British Museum collection of Blattidae, although a large number were examined with this in view. PERIPLANETA IGNOTA, n.sp. Text-figs. 27 and 28. This cockroach has been known to me for some considerable time, having occurred in my own house together with the two preceding species, but it was not Text-fig. 27. Periplaneta ignota Shaw. Pronotum showing outline of the pale macula. Text-fig. 28. Periplaneta ignota Shaw. co. Apex of abdomen, dorsal aspect. Drawn from the holotype. recognized as separate until a good series was obtained from the caretaker’s isolated room, perched at the end of a jetty at Wynnum, to the south of the mouth of the Brisbane River. Since then I have seen it in other collections from localities in Queensland, but it is not to be found in the older collections of the Australian Museum, Sydney, or the Macleay Museum. 206 NEW GENERA AND SPECIES OF BLATTIDAE, The following table gives the differences between the three species: P. australasiae P. americana Linné. P. ignota Shaw. Fabr., Yellow humeral streak of tegmen| present. absent. absent. Pale marking like an inverted mushroom on dark pronotum|] absent. absent. present, ist abdl. tergite o&| ample, with gland | narrow, with no | ample, with gland opening. gland opening. opening. 7th abdl. tergite o| sinuate, angles | subsinuate, angles | sinuate, angles backwardly pro- not backwardly backwardly pro- duced. produced. duced. ; Supra-anal lamina d'| short, lateral mar- | produced, mem- |} short, rounded gins concave, braneous distally, posteriorly, no wide, shallow lateral margins emargination. emargination, or convex, deep with posterior acute angular margin straight. emargination. Supra-anal lamina ?| emargination right | emargination acute | emargination right angled. angled. angled. This table shows that ignota mihi is nearer to australasiae Fabr., but beyond the pale pronotal marking which is sometimes obscure, it can readily be separated " from that species by the absence of the yellow humeral streak, and by the different form of the supra-anal lamina (Text-fig. 28) of the g; and this lamina is quite distinct in americana L. from that of either of the other two species; besides which, the postero-lateral angles of the abdominal tergites are not backwardly produced in americana L. The two portions of the pale pronotal marking, as shown in Text-fig. 27, are usually joined together into one macula. Length and 9, 25.0-30.0 mm. Tegmina 6, 24.0-28.0 mm.; 9, 23.0-25.0 mm. Type material.—Holotype g, No. 0/2868, and allotype 9, No. 0/2868a, Coll. Q. Mus. Several paratypes, Colls. auct., Q@. Mus., Stockholm Mus. Habitat.—Queensland: Wynnum, Cleveland, Cannon Hill, Goodna (auct.), Brisbane distr. (J. C. Bridwell, H. Hacker), Noosa (W. R. Colledge), Burdekin distr. (L. Kelly), Toogoolowah (R. L. Higgins), Colosseum (EH. Mjoberg). Genus Doryara Stal. DORYLAEA FLAVICINCTA de Haan. New locality.—Panaeati Is., Louisiade Archipelago. Coll. auct.). (Revd. F. J. Barnes, 1915. DORYLAEA FLAVIFRONS, n. sp. Text-figs. 29 and 30. Uniformly castaneous dorsally. Antennae dark fuscous; vertex and frons castaneous, frons occupied by three orange-yellow maculae arranged in a triangle with its base towards the mouth, and an orange-yellow line separating the castaneous vertex from the eyes and antennary fossae; margin of the clypeus BY ELAND SHAW. 207 and labrum rufous, maxillary palpi pale. 'Tegmina semi-corneous extending to the 3rd abdominal tergite, venation obscure, anal vein terminating at the junction of the middle and outer thirds of the sutural margin. Posterior margin of the 7th abdominal tergite sinuate. Supra-anal lamina (Text-fig. 29) of the ¢ rounded, faintly emarginate, with a median longitudinal carina, extending to about half the length of the cerci; of the 9 longer and more deeply emarginate. Subgenital lamina (Text-fig. 30) of the ¢ produced, widely emarginate, terminating in divergent points. Styles long, slightly incurved, a little longer than the lamina. Coxae pale, with large dark maculae, remainder of the legs castaneous, strongly spined; posterior tibiae, on the outer side, triseriately spined; posterior metatarsus long, biseriately spined beneath, 2nd segment spined beneath, 3rd and 4th seg- ments unspined, pulvilli, ist and 2nd apical, 3rd and 4th large. Genicular spines 0,1, 1. Arolia large. Length.—, 26.0 mm.; 9, 25.0 mm. Type material.—Holotype ¢, and allotype 9, New Ireland, Coll. Macleay Mus., Sydney. Paratypes 3, Coll. auct. Habitat.—New Ireland (Types); New Hebrides (W. W. Froggatt); ? New Zealand (A. Philpott). Note.—The record from New Zealand is probably one of an introduced species, reaching that country in imported fruit (vide Shaw, 1922, p. 230). Text-fig. 29. Dorylaea flavifrons Shaw. o. Apex of abdomen, dorsal aspect. Drawn from the holotype. Text-fig. 30. Dorylaea flavifrons Shaw. o¢. Apex of abdomen, ventral aspect. Drawn from the holotype. Text-fig. 31. Ancaudellia serratissima Brunner v. Wattenwyl. Apex of abdomen, Drawn from a specimen in the Macleay Museum. Subfamily PANCHLORINAE. Genus ONIscosomMA Brunner v. Wattenwyl. ONISCOSOMA MINIMA, N. Sp. 6 fully winged. Head nearly black, mouth parts testaceous, a yellow longitudinal streak on the vertex; antennae with the sockets pale, 1st segment large and brown proximally, the two or three succeeding segments testaceous, and the remainder brown, with, in the larva, a couple of white segments towards the apex. Pronotum bearing numerous coarse tubercles, nearly semicircular anteriorly, concealing the head, convex posteriorly; a fine median longitudinal line is visible, and the widest part of the pronotum is about two-thirds along this 208 NEW GENERA AND SPECIES OF BLATTIDAE, line from the anterior margin; anterior half testaceous with some brown markings, posterior half occupied by a fusco-castaneous macula with irregular edges and diverticula, a few fuscous spots around the margin. Tegmina hyaline with brown mottling, chiefly along the veins, but sometimes on the membrane between them. Supra-anal lamina triangular, rounded at the apex; cerci broad, blunt, somewhat incurved; subgenital lamina large, extending well beyond the supra-anal, testaceous, with a large dark macula occupying the disc, the lateral margin on the right side being more deeply emarginate; two slender acuminate styles. Abdominal sternites testaceous, with some opaque white markings laterally; stigmata well marked, fuscous. Legs testaceous. Larvae darker generally than the adult, with numerous fuscous markings, dots and tubercles. Length.—14.0 mm.; tegmen 11.0 mm.; body 9.0 mm. Type material.—Holotype g, No. 0/2867, Coll. Q. Mus.; paratypes, 2 larvae. Habitat.— Queensland: Bunya Mts. (H. Tryon, Oct., 1919). Subfamily PANESTHIINAE. Genus PANESTHIA Serville. Some years ago it was suggested by me (1914) that species of the genus Panesthia bit off their flying organs for one another, and recently Mr. A. M. Lea, of Adelaide, told me he had observed this taking place. The reason for this is apparently to enable the insect to move more readily backwards and forwards amongst the loose soil or dessicated wood in which it lives. The two common Australian species are australis Br. v. W. and laevicollis Sauss. (unless indeed I am mixing up the latter with cribrata Sauss.); australis Br. v. W. occurs in Victoria and New South Wales, Jaevicollis Sauss. in Queensland and the north part of New South Wales, where, in the New England district, the two species overlap. Both species appear to live in families, and one usually finds associated together a pair of adults and from about 12 to 20 larvae in different ecdyses, from the penultimate down to quite early ones, and it is only where the insects are very abundant that one loses sight of this familial habit. Probably both species are viviparous. Genus ANCAUDELLIA, noy. Text-fig. 31. Mr. A. N. Caudell (1924) suggests the need for removing Panesthia serratissima Br. v. W. to another genus on account of the structure of the postero-lateral angles of the 7th abdominal tergite. With this I quite agree, and, as I have before me several specimens of J, 9, and larva from various localities, I would propose Ancaudellia as the name of a new genus, of which the type would be serratissima Br.v. W. It differs from Panesthia Serville as follows, and the table given below may be of some use in separating the genus from its immediate neighbours: Lateral margin of the 7th abdominal tergite deeply notched anterior to the postero- lateral angle which is somewhat laterally divergent, though produced backwards at the apex (Text-fig. 31). (4) 1. Sixth abdominal tergite with the postero-lateral angles not, or but slightly produced. (3) 2. Seventh abdominal tergite with the lateral margin PentlySinuatey wets te: satvous ee auoreuepetevene nes cackerene npeacheleze Panesthia Serville. (2) 8. Seventh abdominal tergite with the lateral margin deeply notched anterior to the postero-lateral angle Ancaudellia Shaw. (1) 4. Sixth abdominal tergite with the postero-lateral angles backwardlys produced! A. cmsn cies clokic ceeke ce cleieneoeie Miopanesthia Saussure. BY ELAND SHAW. 209 ANCAUDELLIA SERRATISSIMA Brunner v. Wattenwyl. Text-fig. 31. Panesthia serratissima Br. v. W., Nouv. Syst. Blatt., 1865, p. 394. Brunner described this species from Ternate. He states that the larvae are completely black, but this is not so, as I have taken them in Papua of the brownish-yellow colour usual to Pamesthiine larvae shortly after an ecdysis, and the young larva shows plainly the distinctive character of the genus. The ¢ varies in the depth of the excavated portion of the anterior pronotal margin, and in the prominence of the tubercles bordering it; it also bears on the vertex Saussure’s (1895) “dépression en fossette.” New Localities —S.H. Papua: Kui-ara (Coll. auct.). New Guinea: Katow; Woodlark Is. and New Ireland (Macleay Mus., Sydney). Genus PLANA Brunner y. Wattenwyl. Key to the species of Plana. (4) 1. Supra-anal lamina entire. (3) 2. Larger, more densely punctate, pronotum more transverse, SUbSenitalelaminamcusleSsss.visiblemeraanen seis cee eiene dilatata Sauss. (2) 3. Smaller, smoother, pronotum less transverse, subgenital AMaMin ace MOLE MVisibley as. «Gaps clase orks ee ee ... robusta Tepper. Geta supEa-analelaminaycrenulatemen aes oe oie aoe: crenulata Shaw. PLANA DILATATA Sauss. This species and P. robusta Tepp. may have been confused by Tepper himself, but Saussure’s (1895) notes (Rev. Suisse Zool.) help to clear the matter up. It is a question whether Tepper’s robusta is not Saussure’s dilatata, and the smaller and smoother species with the more visible subgenital lamina which Saussure calls robusta Tepp., and which I have separated as such, may never have been known by Tepper, and should really be Saussure’s species also. Saussure (1895) says of robusta Tepp. “Mus. Genavense, specimina typica Tepperi.” Tepper’s types ¢ and 9 in the South Australian Museum are both 6, and were not indicated by him as his types until 22 years after he had described the species. Until a more critical examination is made of all the material in the South Australian Museum, it is better to leave the matter open. Var. MAJoR Sauss.—The ¢ is easy to distinguish by the pronotal structure, but the 9 difficult. PLANA CRENULATA, n. Sp. Text-fig. 32. Large. General colour nigro-castaneous dorsally, rufo-castaneous ventrally. Head smooth, nigro-castaneous, covered by the pronotum; margin of the clypeus tawny; antennae of ¢ (9 damaged) castaneous, lst segment longer than the two succeeding segments. Pronotum (Text-fig. 32) nearly semicircular, postero- lateral angles rounded, and posterior margin convex; behind the anterior margin in both sexes is a tumefaction forming two prominent tubercles in the dg, but only slightly raised and rounded in the 9, further caudally is a depression succeeded by ridges meeting in the middle line, rough and broken up in the 6, but smooth in the 9, and bounded by two deep sulci converging caudally; the posterior half of the pronotum is smooth in the 9, but occupied by a large median diverticulum extending to within 2 mm. of the posterior margin in the ¢. Mesonotum and metanotum smooth, rufo-castaneous in the disc, with the postero-lateral angles somewhat backwardly produced. Abdominal tergites 1 to 5 smooth in the disc, the outer 4 on each side being sparsely but coarsely punctate; 6th tergite punctate along the caudal margin; 7th tergite and supra- 210 NEW GENERA AND SPECIES OF BLATTIDAE, anal lamina densely and coarsely punctate; the postero-lateral angles of tergites 2 to 7 backwardly produced, that of the 6th being produced into a strong prominent recurved spine, and that of the 7th tergite being divergent and but slightly recurved; the outer portion of the caudal margin of tergites 4, 5, and 6 is thickened, and crenulate, and the plane of tergite 7 is bent at a considerable angle to that of the tergites anterior to it. Supra-anal lamina transverse, caudal margin rounded, crenulate. Abdominal sternites with the disc smooth, rufo- castaneous, margins punctate and darker. Legs rufo-castaneous, tibiae darker, anterior tibiae fossorial, anterior femora bispinose. The holotype has a group of three spines—one of them small—on the right anterior femur; the coxal process (Shaw, 1922) appears to be completely separated from the coxa, as in all the Panesthiinae. Length.—, 50.0 mm.; 2 45.5 mm. Type material.—Holotype ¢, No. 0/2861; and allotype 9, No. 0/2861a, Coll. Q. Mus. Habitat.—Queensland: Noosa Head (H. A. Longman, R. L. Higgins). Note.—This fine species I at first mistook for P. gigantea Tepp. (q.v.), but an examination of his types undeceived me. The peculiar structure of the caudal margins of the distal tergites is of a like character, but more marked in Macro- panesthia gigantea Tepp. (= M. muelleri Sauss.) and under that species an attempt is made to explain it. In addition to my types, I know of only a single ¢ in the Queensland Museum. Text-fig. 32. Plana crenulata Shaw. co. Outline of whole insect (nat. size). Drawn from the holotype. Text-fig. 33. Macropanesthia hirsuta Shaw. co. Outline (x 3/2). Drawn from the holotype. Genus HEMIPANESTHIA Saussure. Hemipanesthia Sauss., Revision des Panesthiens, Rev. Suisse Zool. iii, 2, 1895, p. 327. BY ELAND SHAW. 211 HEMIPANESTHIA KRAUSSIANA Saussure. H. kraussiana Sauss., loc. cit., p. 328, Pl. ix, fig. 6.—Panesthia kraussiana Suask, Mem. Soc. Sc. Phys. Nat. Genéve, xxiii, 1872, p. 150. Saussure described this species from the neighbourhood of Melbourne, and the type is in the Geneva Museum. The Macleay Museum collection contains 1 4, 1 9 (Rockhampton) and 2 larval 2 (N. S. Wales). Of these the ¢ measures 32 mm., the 2? 37 mm., and the larvae 24 mm., and 26 mm. in length. The 9, except for being larger, agrees well with Saussure’s description, in which the 9 only is mentioned, although his figure and the explanation of the plate are marked 6. If Saussure knew the 9 only, the designation of his figure is an error, the ¢ in the Macleay Museum would be the allotype, and I tentatively regard it as such. A few notes of the ¢ are added. The ¢ closely resembles the 9, but the anterior margin of the pronotum is not emarginate, nor tuberculate, and is more definitely reflexed; the discal irregularities are considerably more marked, a well defined trifoliate excavation with a tubercle on each side occupying the middle third of the disc. The postero- lateral angles of the 7th abdominal tergite are of the same form as in the Q. The supra-anal lamina has its posterior margin slightly crenulate in the middle third (in the 9 this is more marked, agreeing with Saussure’s figure). Anterior femora 3-spinose. The larva closely resembles the adult, and also has the anterior femora 3-spinose. : Type material.—Allotype ¢, Macleay Mus., Sydney. Habitat—Queensland: Rockhampton; N. 8S. Wales. Genus MACROPANESTHIA Saussure. MACROPANESTHIA GIGANTEA Tepper (¢ only). Geoscapheus giganteus Tepp. (¢ only), Trans. Roy. Soc. S. Aust., 1894, p. 176. Macropanesthia muelleri Sauss., Rev. Trib. Panesth., Rev. Swisse Zool., 1895, p. 329, 12, Tbe, wie, By, I have examined Tepper’s types of G. giganteus in the South Australian Museum. These are labelled “Type of ¢” and “Type of 9”; they agree in the main with his description and measurements, but both are of the male sex, and they are not of the same species. His “Type of ¢’’ must stand as the holotype of the species, and Saussure’s Macropanesthia muelleri, which is identical with it, becomes a synonym. Tepper’s “Type of 9” is really a ¢ of Macropanesthia rhinocerus Sauss., and Tepper’s name for it, already occupied by another species, must go down as a synonym of Saussure’s. The confusion may be due to Tepper’s late indication of his types, and it is very improbable that when Saussure described muelleri, also from a ¢, he had seen Tepper’s excellent descrip- tion of it published a year previously. In many of the Panesthiinae the 7th abdominal tergite and supra-anal lamina are bent downwards more or less sharply, their plane joining that of the proximal tergites at an angle, and presenting an extensive surface posteriorly. This surface is usually coarsely and densely punctate, and is frequently furnished around its borders with strong spines and crenulations. On account of this bending, the posterior margin of the sixth tergite becomes more prominent, is sometimes simply everted, but is more frequently also armed with crenulations or denta- tions, or, as in the present species, with a row of strong tubercles, in some examples almost spines, repeated, but less prominently, on the 5th and 4th tergites. Plana crenulata mihi shows these crenulations very well. P. dilatata Sauss. and his var. major also have them, but not robusta Tepp. In the genus Panesthia D 212 NEW GENERA AND SPECIES OF BLATTIDAE, several species such as lata Walk. and sloanei mihi show this armature in some degree; whilst Hemipanesthia kraussiana Sauss., which does not possess it, is of more depressed form, and may not be an earth digger. Salganea morio Burm. has the posterior margin of the 6th tergite entire, but the 7th makes up for this by being crumpled and having dentate lateral margins. I have kept specimens of the present species alive for several months in glass jars of sand, watched them going underground, and the whole apparatus seems to be most effective as an aid to the insects in burying; the abdominal somites are retracted, the smooth anterior portion of each sliding up beneath the preceding one, and the shortened abdomen with its posterior end, shaped and armed for the purpose, takes a good purchase against the collapsed back and sides of the burrow, thus enabling the fossorial tibiae and scoop-like anterior part of the pronotum to do their work, and to be thrust forward when the way is prepared. As the ¢ only has been described, and I have several 2 in my collection, I have selected one as an allotype, and add a few notes upon it. - 9. Much resembling the 3, but the pronotum is relatively wider, the anterior margin is gently recurved medially, and slightly tumefied, but not tuberculate; the excavation of the disc is not so pronounced, and the portion of the disc behind the converging sulci is not excavated. The armature of the distal ab- dominal tergites is similar to that of the ¢. Supra-anal lamina with definite crenulations in the middle of the posterior margin. Anterior femora 2-spinose. Length.—Z, 35.0-39.0 mm. (Allotype, 38.5 mm.). Type material.—Allotype 9, No. 0/2862, Coll. Q. Mus. MACROPANESTHIA RHINOCERUS Saussure. Macropanesthia rhinocerus Sauss., Revis. Trib. Panesth., Rev. Suisse Zool., iii, 1895, p. 329, Pl. ix, fig. 4—Geoscapheus giganteus Tepp. (9 only), Trans. Roy. Soc. S. Aust., 1894, p. 176. Saussure described the ¢ from Bowen, Queensland, and Tepper’s “Type of 9?” of his Geoscapheus giganteus is really a ¢ of M. rhinocerus Sauss., sent to him from North Queensland by Mr. C. French. The National Museum, Melbourne, has two damaged specimens with no locality labels, and the Macleay Museum, Sydney, has a larval 2 from Cleveland Bay, and this, the neighbourhood of Townsville, is the district from which rhinocerus may be expected. MACROPANESTHIA HIRSUTA, Nn. Sp. Text-fig. 33. 6 nigro-castaneous, except the disc of the abdominal sternites and portions of the legs, where the colouring is of a rich castaneous. Head finely punctate, margin of the clypeus and base of the labrum brownish. Pronotum anteriorly produced, covering the vertex, finely punctate and crumpled, two prominent tubercles with their apices recurved situate within the anterior margin, disc excavated, the excavation extending backwards to within 1 mm. of the posterior border. Meso- and metanotum laterally finely punctate. Abdominal tergites 1 to 6 laterally sparsely coarsely punctate; tergite 7 and supra-anal lamina coarsely punctate; postero-lateral angles of the 6th tergite not produced, but a large, erect, blunt tubercle is situate 2 mm. along the posterior margin; postero-lateral angles of the 7th tergite produced into strong divergent recurved spines; abdominal sternites coarsely punctate laterally. Anterior femora 2- or 3-spinose. Legs densely hirsute where they may come into contact with the body or with one another. BY ELAND SHAW. 213 Length.—42 mm. Type material—Holotype g, S. Aust. Mus. Habitat.—Queensland: Dalby (Mrs. F. H. Hobler). Note.—The hairiness from which this species is named, is more abundant than in any other of the Panesthiinae known to me. It is more abundant in the group which live underground entirely (Tepper’s Geoscapheusidae) than in those which live in rotten wood, and probably acts as a triturating agent to break up sticky soil, leaving the leg movements unimpeded. List of references. Barper, M. A., 1913.—Cockroaches and Ants as carriers of the Vibrios of Asiatic Cholera. Far Eastern Assn. of Tropical Medicine, 3rd Congress, Saigon. Borivar, 1882.—Ann. Soc. Ent. France (6), ii. BRUNNER V. WATTENWYL, 1865.—Nouveau Systéme des Blattaires Vienne, 1865. BURMEISTER, 1838.—Handb. der Entomologie, Vol. ii, Berlin. CAuUDELL, A. N., 1924.—Malayan and Hast Indian Blattidae. I. Introduction and Subfam. Panesthinae. Philippine Journ. Sci. xxiv, No. 6. Hanitscu, R., 1915—Malayan Blattidae. Journ. Roy. Asiatic Soc. (Straits Branch), No. 69. SaussurE, H. pr, 1872.—Mém. Soc. Sc. Phys. Nat. Genéve. Mélanges, iv. , 1895.—Rev. Trib. Panesth. Rev. Swisse Zool. iii, 2. SHAW, Hland, 1914.—Australian Blattidae. Part i. Notes and preliminary descrip- tions of New Species. Vict. Nat., xxxi, No. 7. , 1916.—Australian Blattidae. Part ii. On the Type of Ischnoptera brun- neonigra Tepper, with a description of the Male Insect. Vict. Neat. xxxiii, No. 6. , 1918.—Australian Blattidae, with descriptions of eleven New Species. Mem. Queensland Mus. vi. , 1922.—Descriptions of New Australasian Blattidae, with a note on the Blattid Coxa. Proc. Linn. Soc. N.S. WALES, xlvii, Part 3. SHELFORD, R., 1907-1910.—Genera Insectorum. Fasc. 55, 73, 74, 101, and 109. Bruxelles. 1907-1910. , 1909.—Studies of the Blattidae. A Revision of the Old-World Blattinae belonging to the Polyzosteria group. Trans. Ent. Soc. Lond. Part ii. ; , 1911—The British Museum Collection of Blattidae enclosed in Amber. Journ. Linn. Soc. Zool. Vol. xxxii, No. 212. , 1912.—New Blattidae from New Guinea. Collected by Prof. Dr. Schultze. Entomologischen Rundschau, 29. Jahrgang No. 7. , 1916.—A Naturalist in Borneo. Edited by Edward B. Poulton, London. 1916. TrpperR, J. G. O., 1893.—The Blattariae of Australia and Polynesia. Trans. Roy. Soc. S. Aust. Vol. 17. , 1894.—The Blattariae of Australia and Polynesia. Supplementary and Additional Descriptions and Notes. Trans. Roy. Soc. S. Aust. Vol. 18. , 1895.—Notes on Victorian and other Blattariae, and Descriptions of New Species. Trans. Roy. Soc. 8. Aust. Vol. 19. TINDALE, N. B., 1923.—The Flora and Fauna of Nuyts Archipelago, and the Navigator Group. Trans. Roy. Soc. S. Aust. Vol. xlvii. : WALKER, F., 1868.—Catalogue of the Specimens of Blattariae in the Collection of the British Museum. London. 1868. FOSSIL PLANTS FROM THE NARRABEEN STAGH OF THE HAWKESBURY SERIES. By A. B. WALKoM, D.Sc. (Plates xxiv-xxxi, and one Text-figure.) [Read 24th June, 1925.] Introduction. This paper contains the results of the examination of a large number of fossil plants from the Narrabeen Beds, in the collection of the Geological Survey of New South Wales, and a few specimens from the Australian Museum Collection, and I have to express my appreciation of the opportunity afforded me of studying these interesting collections. The figured specimens, with the exception of Taeniopteris triassica (Text-fig. 1), are in the collection of the Geological Survey of New South Wales. The specimens examined were in the great majority collected by Mr. B. Dunstan, F.G.S., Chief Government Geologist of Queensland, while Lecturer in Geology at the Technical College, Sydney. Mr. Dunstan collected extensively from the beds at Turrimetta Head, a horizon approximately 150 feet below the base of the Hawkesbury Stage. Valuable additions to the collection in the Mining Museum were made by Mr. C. A. Sussmilch, F.G.S., Principal, Technical College, Newcastle. The Upper Coal Measures (Permian). in New South Wales are followed, usually without apparent stratigraphical break, by the Hawkesbury Series which consists of three stages, Narrabeen, Hawkesbury Sandstone and Wianamatta, the Narrabeen Stage being the lowest. The Narrabeen Stage has a maximum thickness of about 1,750 feet and consists mainly of fine-grained sandstones and sandy shales. The upper portion of the stage is formed by a thickness of about 130 feet of red and purple shales which serve to mark a conspicuous horizon dividing the Narrabeen Stage from the Hawkesbury Sandstone Stage. The roof of the Bulli Coal Seam has been regarded as the upper limit of the Upper Coal Measures, and from the strata within a few feet above this seam a small flora has been described (Dun, 1910, 1911), including the following species :— Glossopteris Browniana. Schizoneura gondwanensis (= S. australis). Cladophlebis cf. Roylei. Rhipidopsis ginkgoides var. Sussmilchi. Taeniopteris cf. McClellandi. This small collection of species is obviously more closely related to the Permian flora than to the typical Mesozoic flora which appears higher up in the Narrabeen Stage, and is to be regarded as a remnant of the Glossopteris flora. Although fossil plants are very abundant at certain horizons in the Narrabeen Stage and have been collected for many years, no description of any of the BY A. B. WALKOM. 215 collections has been published. Mr. W. S. Dun prepared lists of the plants deter- mined from the different Stages of the Hawkesbury Series, and these lists were published in the account of the Western Coalfield (Carne, 1908, p. 42). Mr. Dun very kindly handed over to me, for examination, the collections and also the drawings which had been prepared for him by Mr. F. R. Leggatt and I have to express my gratitude to him for his generosity in this respect. The results of my examination confirm the determinations made by Mr. Dun as contained in the lists prepared by him. The late R. EHtheridge, Jr., described examples of Schizoneura australis (= S. gondwanensis Feistmantel) from olive-green shales of the Narrabeen Stage at 1,274 feet 6 inches in the Cremorne Bore (Rec. Geol. Surv. N.S.W., iv (1), p. 32), and a further specimen with ftructification from 2,870 feet in the Sydney Harbour Colliery Shaft at Balmain (ibid., vii (3), p. 284). He also described a specimen from 1,410 feet to 1,417 feet 6 inches in the Cremorne Bore as Sagenopteris salisburioides R. M. Johnston (ibid., iv (1), p. 34). The following is a list of the species described from the collection under examination :— Phyllotheca australis Brongn. Coniopteris sp., cf. lobata (Oldham). Fern stems (? Osmundaceae). Cladophlebis sp. (fertile). ? Cladophlebis sp. (sterile). Thinnfeldia Feistmanteli Johnston. Thinnfeldia lancifolia (Morris). Thinnfeldia narrabeenensis Dun. ? Sphenopteris sp. Taeniopteris Tenison-Woodsi (?) Etheridge Jr. Taeniopteris crassinervis Feistmantel. Taeniopteris triassica, n. sp. Taeniopteris wianamattae (Feistmantel). Williamsonia sp. (flowers). Williamsonia sp. (stems). Ginkgoites sp. ? Rhipidopsis narrabeenensis, n. sp. Brachyphyllum angustum, n. sp. Araucarites sydneyensis, n. Sp. Carpolithus sp. The specimens have mostly been obtained from the shales about 6-8 feet above high-water mark at Turrimetta Head, just to the north of Narrabeen. The horizon of these shales is about 150 feet below the top of the Narrabeen Stage. The result of the examination of this flora of the Narrabeen Stage is some- what disappointing to one who knows the abundance of plants on some horizons. There is, however, some satisfaction in the knowledge that the descriptions herein will form a basis for future work as well as for future collecting, for it must be remembered that past collecting has all been without any guide as to which forms were known and which were unknown. The flora of the Narrabeen Stage should be one of the most interesting of the Australian fossil floras when it is better known, as it is almost certainly of Lower Triassic age and floras of this age are not well known in the Southern Hemisphere. The occurrence of Glossopteris and Schizoneura in the ‘basal beds ‘shows that the upper Palaeozoic flora had not entirely disappeared at the commencement of the Narrabeen Stage, but these forms did not persist for long and are not known in association with the typical Lower Mesozoic flora in Australia. 216 FOSSIL PLANTS FROM THE NARRABEEN STAGE, N.S.W., Equisetales. PHYLLOTHECA AUSTRALIS Brongn. Plate xxiv, figs. 1-3. Phyllotheca australis Brongniart, Prodr. Hist. Veget. foss., 1828, p. 152; Etheridge, Geol. Pal. Qland, 1892, p. 189, Pl. 17, fig. 13—(For full synonymy see Arber, 1905, p. 17.) Fragments of equisetaceous stems are abundantly represented in the collec- tions, but in only one specimen of those examined are the leaves preserved (Pl. xxiv, fig. 2). This specimen can be referred to Phyllotheca australis. The remaining examples are either stem impressions or pith casts, and while, strictly, they should only be determined as Phyllotheca sp. or even “Equisetaceous stems,” there is every probability that they all represent the one species. Associated with the stems there are some examples of nodal diaphragms, one of them being figured on Plate xxiv, fig. 3. The question might be raised as to whether any of the stems should be referred to the genus Schizoneura, but, in the absence, from the collections examined, of any foliage referable to this genus, and on account of the pith casts and stems being quite unlike any of the stems referred to Schizoneura from the Ipswich Series of Queensland, it seems probable that the examples from the Narrabeen Series represent stems of Phyllotheca. Filicales. Cyatheaceae. ? CoNIOPTERIS sp. cf. LOBATA (Oldham): Plate xxix, figs. 4, 6. Frond bi- (? tri-) pinnate. Ultimate pinnae long, narrow, 4-5 mm. broad, opposite or alternate. Pinnules small, ovate, attached by whole base; median vein present, not persistent to apex, together with several secondary veins which make an acute angle with the median vein, and divide once in their course to the margin. These specimens appear to be close to those described as Coniopteris lobata from the Jurassic rocks of Graham Land by Halle (1913, Pl. 1, fig. 27; Pl. 3, fig. 13; and text-fig. 5). This species has been recorded also from Upper Gondwana rocks of India and from the Rhaetic of Poland. Osmundaceae. CLADOPHLEBIS sp. Plate xxiv, fig. 4. The specimen figured appears to be an example of portion of a fertile frond of a species of Oladophlebis, possibly C. australis. It is very similar in general appearance to a fertile example of ©. denticulata figured by Seward (1910, p. 345, fig. 258) from Yorkshire. It is, however, the only specimen of its kind among the collections and, in the absence of preservation of any details of structure, it is not easy to determine its true nature satisfactorily. It is quite different from the fertile specimens referred to Cladophlebis australis from the Ipswich Series of Queensland (see Walkom, 1917, p. 3, Pl. 7, 8). ? CLADOPHLEBIS sp. Plate xxiv, fig. 5. Several specimens, very imperfectly preserved, probably represent isolated pinnae of a species of Cladophlebis. The pinnae are long, parallel-sided, about 2.6 cm. wide, with pinnules closely-set, opposite or alternate, and making a wide angle with the rachis. The pinnules have a distinct midrib and secondary veins (observed with difficulty in parts of the specimen) which make an acute angle with the midrib, and which divide once (?) in their course. BY A. B. WALKOM. 217 Thinnfeldieae. THINNFELDIA FEISTMANTELI Johnston. Plate xxiv, figs. 6-9; Pl. xxv, figs. 1, 2. (For synonymy see Walkom, 1917, p. 17.) The description of this common Mesozoic species has already been given ina previous paper (Walkom, 1921, p. 9). The specimens from the Narrabeen Series are typical examples and exhibit considerable variation in size. In some of the smaller examples the pinnules are more pointed than in the larger ones and may even be somewhat falcate. Some of the large specimens show very well the character of the pinnules seated directly on the rachis; those below the point at which the rachis divides are well separated from one another and are contracted at their base. This species of Thinnfeldia seems very close to that figured by Zeiller (1903, Pl. vi-viii) as Ctenopteris Sarrani. Seward and Holttum (1921, p. 41) express the ‘opinion that the two are probably specifically distinct; they also suggest that T. Feistmanteli would be more appropriately referred to the genus Ctenopteris in view of the fact “that the habit of the frond as a whole is recognized as an important character in distinguishing between such genera as Ctenopteris, Thinn- feldia and Ptilozamites.” They do not, however, suggest that the other Australian species of Thinnfeldia, T. lancifolia and T. odontopteroides, should also be referred to Ctenopteris. The species recently described by me as Thinnfeldia talbragarensis (Walkom, 1921, p. 9) forms a connecting link between 7. lancifolia and T. Feistmanteli and, as all four species of Thinnfeldia above-mentioned are charac- terized by their frond habit, in particular by the dichotomous forking of the rachis, it seems to me impossible to separate them generically and I am unable to agree with the suggestion that any one of them should be transferred to Ctenopteris. In the collections there are several specimens like that figured on Plate xxiv, fig. 9. These may be fragments of fertile fronds of 7. Feistmanteli. They do not show any detail of structure, exhibiting only the distribution of the sori and the shape of the pinna. They agree in these general characters with the specimen figured from the Ipswich Series of Queensland (see Walkom, 1917, Pl. 1, fig. 3). THINNFELDIA LANCIFOLIA (Morris). Plate xxv, fig. 3; Pl. xxvi, figs. 1-3; Pl. xxvii, figs. 1, 2, 4, 5. (For synonymy see Walkom, 1917, p. 21.) “Frond divides dichotomously into two linear pinnae which are inclined to one another at an acute angle. Pinnules vary in form with their position; the majority are elongate, tapering, with a rather acute tip, and have a distinct midrib which does not usually persist to the tip of the pinnule. The pinnules on the inner sides of the branches become smaller as the point of branching is approached, and gradually change from elongate acute to a rather ovate semi- circular or more or less rhomboidal shape, without a midrib, the veins arising directly from the rachis. The pinnules on the rachis below the junction may be of either type. The venation is alethopteroid, the secondary veins being given off from the midrib at a rather acute angle. In the basal portion of the pinnules some of the veins come direct from the rachis; the pinnules are decurrent and are connected by a narrow lamina along the rachis’ (Walkom, 1917). This species is of common occurrence in the Lower Mesozoic rocks of Australia, and also occurs in the Stormberg Beds of South Africa. 218 FOSSIL PLANTS FROM THE NARRABEEN STAGE, N.S.W., THINNFELDIA NARRABEENENSIS Dun, MS. Plate xxvi, fig. 4; Pl. xxvii, figs. 3, 6; Pl. xxviii, figs. 1-4. Frond large, pinnate, dichotomous. Rachis strong, pinnules large, elongate, obtusely pointed, opposite or subopposite, with margins usually entire, occasionally broadly lobed. In the basal portion of the frond the pinnules are modified, being reniform to triangular and somewhat contracted at the base. The elongate pin- nules have a prominent midrib, with secondary veins close, slightly curved, and branching. The basal pinnules have venation of the odontopteroid type. This is a much larger and more robust type than any of the other Australian species of Thinnfeldia. The individual pinnules attain a size of 7.5 cm. long by 2 em. broad; their form varies somewhat; they may be gradually rounded with blunt apices, or tapering to acute apices. The species seems to be most closely related to those specimens described as Danaeopsis Hughesi from rocks of Rhaetic age in India, South Africa, China, Tonkin and Queensland. From this species it differs in its somewhat smaller size and in the secondary veins being less numerous and not branching as frequently. Whether the two species should be referred to the same genus is a question about which there is some doubt. No fertile examples of D. Hughesi have been - found and there does not appear to be strong evidence for referring the species to the Marattiales. The species 7. narrabeenensis is closely associated with other species of Thinnfeldia and is linked with them by its general characters, and there would seem to be no justification for separating it from this genus, unless indeed, all the other Australian species at present referred to this genus are also moved. It seems possible that the specimens referred to Danaeopsis Hughesi may have to be, at some future date, removed to the genus Thinnfeldia. This species differs markedly from Neuropteridium in the venation and in the attachment of the pinnules to the rachis. Filicales incertae sedis. ? SPHENOPTERIS sp. Plate xxix, fig. 5. The fragment figured on Plate xxix, fig. 5 may be referred, provisionally, to Sphenopteris. It is portion of a bipinnate frond, with small pinnules, but no trace of the venation is preserved. In habit it may be compared with such species as Callipteridium stormbergense Seward (1903, Pl. viii, fig. 2) from the Stormberg beds of South Africa, and Scleropteris crassa Halle (1913, Pl. 4, figs. 4-9) from the Jurassic flora of Graham Land. ? Fern Stems. Plate xxix, figs. 10, 11. Several specimens represent plant stems covered with a series of spirally arranged, elongate oval structures which probably represent a crowded mass of leaf-bases. The leaf-bases are roughly 8 mm. in diameter and in one of the specimens they bear small U-shaped scars, which suggest the impression left by the trace of the vascular strand of the petiole in Osmundites. It seems probable that these stems are those of a fern, possibly belonging to the Osmundaceae. TAENIOPTERIS TENISON-WoopsI (?) Etheridge Jr. Plate xxix, fig. 1. (See Walkom, 1917, p. 32.) The two leaves figured on Plate xxix, fig. 1 resemble Taeniopteris Tenison- Woodsi in the size and form of the leaf and also in the venation. The only BY A. B. WALKOM. 219 difference, a slight one, appears to be that the secondary veins in our examples are slightly curved in their course from midrib to margin whereas in examples of the species previously examined they have been generally straight. The species has only been recorded in Queensland in beds belonging to the Ipswich Series or its equivalents. A fragment which probably represents the same species has been figured as T. Carruthersi from the Stormberg beds of South Africa (Seward, 1903, Pl. viii, fig. 5). TAENIOPTERIS CRASSINERVIS (?) Feistmantel. Plate xxix, fig. 2; Pl. xxxi, fig. 12. Portion of a large Taeniopteris is figured on Plate xxix. It agrees with T. crassinervis except that the veins are somewhat closer in our specimen. The fragment figured on Plate xxxi, fig. 12 is probably portion of a specimen of the same species. J’. crassinervis occurs in Queensland Lower Mesozoic rocks and in the Rajmahal Group in India. TAENIOPTERIS TRIASSICA, n. Sp. Text-fig. 1. In a small collection of plants from the Narrabeen Stage, belonging to the Australian Museum, there are two examples of a Taeniopteris different from any Text-fig. 1. Taeniopteris triassica, n. sp. (x 4). The veins shown here and there indicate the inclination of the secondary veins to the midrib. A portion is also shown (x 2) to indicate the nature of the venation. 220 FOSSIL PLANTS FROM THE NARRABEEN STAGE, N.S.W., hitherto described from Australia. The most complete of these (No. F 17820) is figured in Text-fig. 1. The frond is simple, very long (32 cm.) and relatively narrow (5.8 cm.). The midrib is very prominent and persists to the apex which is obtusely rounded. The secondary veins make a wide angle (about 70°) with the midrib; they branch occasionally and there are about 11 of them per cm. This very large frond is comparable in size with J. crassinervis which has been recorded from Mesozoic rocks in Hastern Australia but it is not so wide in proportion to its length, nor are the veins at right angles to the rachis. The other large Australian species, 7. wianamattae, has veins closer together and is very different in shape from the present species. In dimensions, 7. triassica may be compared with T. Jourdyi Zeiller from the Rhaetic of Tonkin, but in this the secondary veins are very much more numerous (35 to 50 per cm.), and also with the European species 7. vittata. Another comparison may be suggested with Nilssonia taeniopteroides Halle from the Mesozoic flora of Graham Land (Halle, 1913, p. 47). With regard to this latter comparison, however, further specimens would need to be obtained before the identity of the two could be proved, as it would be necessary to have some evidence of the mode of attachment of the lamina. TAENIOPTERIS WIANAMATYAE (Feistmantel). Plate xxix, fig. 3. (For synonymy see Walkom, 1917, p. 38). The specimen figured on Plate xxix, fig. 3 appears to be the apical portion of a leaf of T. wianamattae. The secondary veins are numerous and make an angle of 60-70° with the midrib. Cycadophyta. ? WILLIAMSONIA sp. (Flowers). Plate xxix, figs. 7-9. Specimens which are not unlike some of the flowers referred to Williamsonia are not uncommon in the Narrabeen Series. As preserved they consist of a series of bracts (5 to 8 in number), petaloid in shape and radially arranged, borne on a stout peduncle. The bracts are 1.5 to 2.5 cm. long, and up to 8 mm. greatest breadth. In the most complete specimen the peduncle is 4 cm. long and 5 mm. wide. All the specimens are merely impressions and do not show any further detail of the structure. Two of them are figured on Plate xxix, figs. 7, 8. They may be compared with examples described by Wieland from the Jurassic of Mexico (1913, Pl. 27, fig. 6) as Williamsonia Tlazolteotl. Associated with the specimens described are some (e.g. Pl. xxix, fig. 9) which show the expanded bracts viewed from above. Attention may be drawn to the apparent absence of leaves which might be referred to the genus Williamsonia. It is probable that such will be found later, in view of the presence of these flowers (?) and of the stems described below. ? WILLIAMSONIA sp. (Stems). Plate xxx, figs. 1, 2. Impressions of stems covered with leaf bases or petioles in spirally arranged series are referred tentatively to Williamsonia. The impressions of the leaf bases are rhomboidal in shape, 6-7 mm. by 4 mm. One specimen has portions of the petioles attached. BY A. B. WALKOM. 221 These stems may be compared with numerous examples which have been figured by various workers, e.g. Williamsonia sp. figured by Berry (after Wieland) from the Liassic of Mexico (Berry, 1918, p. 347, fig. 6c); stems of Williamsonia, also from Mexico, figured by Wieland (1913, Pl. 34, 35); fern or cycad stems from the Liassic of Sweden, figured by Antevs (1919, Pl. 6, figs. 36, 41). They may also be compared with such stems as Bucklandia indica from the Lower Jurassic of India (Seward, 1917, p. 488, fig. 579). Williamsonias range from Upper Palaeozoic to Cretaceous. Ginkgoales. GINKGOITES sp. Plate xxxi, fig. 1. The specimen figured on Plate xxxi, fig. 1 represents a fragment of a leaf which may be referred to Ginkgoites. There does not appear to be sufficient of the leaf preserved for correct determination, and it is even possible that the fragment represents portion of a large Baiera such as B. Simmondsi which occurs in the Upper Triassic rocks of the Ipswich Series of Queensland (cf. Walkom, 1917a, Pl. 2). ? RHIPIDOPSIS NARRABEENENSIS, n.sp. Plate xxx, figs. 3, 4. The two examples figured on Plate xxx, figs. 3 and 4 occur on the same specimen. Fig. 3 shows the leaf divided into a number (7) of wedge-shaped segments, and borne on a petiole. The veins are fine and numerous, about .5 mm. apart. The habit of the leaf seems to remove it from Psygmophyllum, though attention may be drawn to a certain degree of resemblance to P. Haydeni which occurs in Permian rocks of India and Russia (see Seward, 1919, pp. 86, 90). Dun (1910) described leaves from the roof of the coal seam in the Sydney Harbour Colliery (at the very base of the Narrabeen Beds) as Rhipidopsis ginkgoides var. Sussmilchi, and Seward (1919, p. 85) has suggested that they are probably referable to Psygmophyllum, which seems doubtful. Coniferales. BRACHYPHYLLUM ANGUSTUM, n. sp. Plate xxx, figs. 5, 6. Examples of portions of sterile coniferous branches with short narrow obtusely pointed leaves arranged in spirals are referred to Brachyphyllum, this generic name being regarded as purely provisional. The branches are thin, the impressions being not more than 3 mm. wide, and the leaves would appear to be up to about 7 mm. long and only 1-2 mm. wide. The specimens may also be compared with sterile branches of Voltzia (Seward, 1919, p. 289) which is typically Permian to Lower Triassic and is very similar to Ullmannia (Permian) and Walchia (Upper Carboniferous-Permian). ? ARAUCARITES SYDNEYENSIS, nh. sp. Plate xxxi, fig. 2. Small cones, up to 3 cm. long and 2 cm. diameter, with cone scales arranged spirally, suggest the reproductive structures of some species of the present day Araucaria. They appear to occur abundantly on certain horizons and they resemble the male flowers of A. excelsa (Seward and Ford, 1906, p. 326, fig. 10) in appearance and in size. They are quite distinct from any other fossil so far found in Australian Mesozoic rocks. 222 FOSSIL PLANTS FROM THE NARRABEEN STAGE, N.S.W., Seeds. CARPOLITHUS sp. Plate xxxi, figs. 3-5. There are a number of oval seeds in the collection. These are up to 1.5 cm. long by 1 cm. broad, but none of them show details of their structure and it does not seem advisable to do more than refer them to Carpolithus sp. Plantae incertae sedis. A. The three specimens figured (Plate xxxi, figs. 7-9), resemble one another in having a series of lines radiating from a central area. The three, however, apparently represent distinct types. Reference may be made to their general similarity to specimens which have been figured as fruits of Williamsonia. Some of them may be compared with the examples of Williamsonia pecten figured by Nathorst (1909, Pl. 2, figs. 16, 17) from the Oolite of Scarborough and with fructifications described by Wieland (1913) as Williamsonia Netzahualcoyotl from Mexico. B. Figure 6, on Plate xxxi is a drawing made by Mr. Leggatt at the same time as most of the others illustrating this paper. I have not seen the original specimen, but it is apparently one which might belong to Stenorachis or perhaps to Beania. C. Plate xxxi, figures 10 and 11 represent other fragments which I am unable to determine but which seem worth figuring with the object of drawing attention to them in the hope that further collecting may reveal additional and more complete examples. List of References. ANTEYVS, E., 1919.—Die Liassische Flora des Horsandsteins. KA. Sv. Vetenskap- sakad., Handl. 59, No. 8. ARBER, H. A. N., 1905.—The Glossopteris Flora. Brit. Mus. Catalogue. Berry, E. W., 1918.—Palaeobotany: A sketch of the origin and evolution of floras. Ann. Rept. Smithsonian Inst., 1918, 289-407. Carne, J. H., 1908.—Geology and Mineral Resources of the Western Coalfield. Mem. Geol. Surv. N.S.W., Geology No. 6. Dun, W. S., 1910.—Notes on some fossil plants from the roof of the coal seam in the Sydney Harbour Colliery. Journ. Roy. Soc. N.S.W., xliv, 615-619. , 1911.—Note on the occurrence of Jaeniopteris in the roof of the coal seam in the Sydney Harbour Colliery. Journ. Roy. Soc. N.S.W., xlv, 554-5. Hate, T. G., 19183.—The Mesozoic Flora of Graham Land. Wiss. Ergebnisse der Schwed. Sudpolar-Exped. 1901-1903. Bd. iii, Lief. 14. NatuHorst, A. G., 1909.—Paldobotanische Mitteilungen. 8. tber Williamsonia, Wielandia, Cycadocephalus und Weltrichia. K. Sv. Vet. Akad., Handl., Bd. 45, No. 4. SEWARD, A. C., 1903.—Fossil Floras of Cape Colony. Ann. S. Afr. Mus., iv, pt. 1. , 1910.—Fossil Plants, Vol. ii. , 1917.—Fossil Plants, Vol. iii. , 1919.—Fossil Plants, Vol. iv. and R. EH. Hortrrum, 1921.—On a collection of fossil plants from Southern Rhodesia. Geological Survey S. Rhodesia, Bull. 8, pp. 39-45. and SIBILLE O. Forp, 1906.—The Araucarieae, recent and extinct. Phil. Trans. Roy. Soc. B198, pp. 305-411. BY A. B. WALKOM. 223 Watxkom, A. B., 1917.—Mesozoic Floras of Queensland. Pt. i (contd.). Queens- land Geol. Surv. Pub. 257. , 1917a.—Mesozoic Floras of Queensland. Pt. i (concluded). Qwueens- land Geol. Surv. Pub. 259. , 1921.—Mesozoic Floras of New South Wales. Pt. i. Mem. Geol. Surv. N.S.W. Pal. No. 12. WIELAND, G. R., 1913—La Flora Liasica de la Mixteca Alta. Inst. Geol. Mecico, Bol. 31. ZEILLER, R., 1903.—Flore Fossile des Gites de Charbon du Tonkin. Etudes Gites Min. France, Paris. EXPLANATION OF PLATES XXIV-XXXI. The figures, with the exceptions noted below, were originally drawn, for Mr. W. &. Dun, by Mr. F. R. Leggatt. Some of them have been slightly modified, particularly as regards the venation, under my direction by Mr. P. T. Hammond, of the Geological Survey, who has also prepared the drawings of the following figures :—Plate xxiv, fig. 2; Plate xxix, figs. 10, 11; Plate xxx, figs. 1, 2; Plate xxxi, figs. 8, 9. Except where otherwise indicated the figures are two-thirds natural size. Plate xxiv. Fig. 1:—Phyllotheca australis Brongn. Pith cast. Fig. 2.—Phyllotheca australis Brongn. Two nodes showing the character of the leaf-whorl at the upper one. Fig. 3.—Phyllotheca australis Brongn. A nodal diaphragm. Fig. 4.—Cladophlebis sp. Possibly fragment of a fertile frond. Fig. 5.—? Cladophlebis sp. Figs. 6-8.—Thinnfeldia Feistmanteli Johnston. These figures indicate the variation in the type of pinnule. Fig. 9.—Thinnfeldia Feistmanteli Johnston. Portion of a fertile pinna showing the arrangement of the sori. Plate xxv. Figs. 1-2.—Thinnfeldia Feistmanteli Johnston. Portions of large fronds. The venation is somewhat diagrammatic in figure 1. Fig. 3.—-Thinnfeldia lancifolia (Morris). Portion of a typical frond with some seeds (?) superposed. Plate xxvi. Figs. 1-3.—Thinnfeldia lancifolia (Morris). Fig. 4.—Thinnfeldia narrabeenensis Dun. The venation is somewhat diagrammatic in this figure. Plate xxvii. Figs. 1-2.—Thinnfeldia lancifolia (Morris). : Fig. 3.—Thinnfeldia narrabeenensis Dun. The venation on the lower half of the middle pinnule in this figure is drawn accurately. Figs. 4-5.—Thinnfeldia lancifolia (Morris). The venation in figure 4 is rather diagrammatic. Fig. 6.—Thinnfeldia narrabeenensis Dun. Plate xxviii. Figs. 1-4.—Thinnfeldia narrabeenensis Dun. Figure 2 shows the basal part of a frond, figures 1 and 3 the apical part. Plate xxix. Fig. 1.—Taeniopteris Tenison-Woodsi Etheridge Jr. (7). Fig. 2.—Taeniopteris crassinervis Feistmantel. Fig. 3.—Taeniopteris wianamattae (Feistmantel). Fig. 4.—? Coniopteris sp. cf. lobata (Oldham). Fig. 5.—? Sphenopteris sp. Fig. 6.—? Coniopteris sp. cf. lobata (Oldham). Figs. 7-9.—? Williamsonia sp. Flowers? Figs. 10-11.—Fern stems (? Osmundaceae). Figure 11 shows one of the leaf bases enlarged. 224 Figs. Figs. Figs. Fig. Fig. Figs. Fig. Figs. FOSSIL PLANTS FROM THE NARRABEEN STAGE, N.S.W. Plate xxx. 1-2.—? Williamsonia sp. Stems. 3-4.—? Rhipidopsis narrabeenensis, n. sp. 5-6.—Brachyphyllum angustum, n. sp. Plate xxxi. 1.—Ginkgoites sp. 2.—? Araucarites sydneyensis, n. sp. (x 4/3). 3-5.—Carpolithus sp. 6.—Planta incertae sedis. (Possibly Stenorachis or Beania). 7-9.—Plantae incertae sedis. j Figs. 10-11.—Indeterminate plant fragments. Fig. 12.—Taeniopteris crassinervis Feistmantel. REVISION OF THE AUSTRALIAN SPECIES OF CHRYSOBOTHRIS (FAM. BUPRESTIDAE), TOGETHER WITH NOTES, AND DESCRIPTIONS OF NEW SPECIES OF COLEOPTERA. By H. J. Carter, B.A., F.E.S. (Nine Text-figures. ) (Read 29th July, 1925.] Sixteen names are recorded for Australian species of the genus, which are (as with other genera of Buprestidae) greatly confused in our collections. This is not surprising, since the general facies of different species is very similar so far as the upper surface is concerned. Of the sixteen, I think I have clearly identified eleven. Of the remaining five I would note as follows:— 1. CO. peroni L. & G. from Kangaroo Island is impossible to determine from the incomplete description (applicable to most of the species), while its figure is evidently incorrectly drawn, the narrow width compared with length being inconsistent with the dimensions as stated in description. An example from Mulwala (Murray River, Victoria) labelled peroni in the South Australian Museum is identical with C. mastersi Macl. 2. C. auropunctata Deyr. is described from New Guinea. Kerremans (Gen. Ins.) gives Australia also for this species, but its imperfect description prevents determination. I think it should be omitted from the Australian list until the type can be examined and compared with other species. 3. C. regina Kerr.—The curious wording of the description of the apex of abdomen apparently indicates the trispinose apex that is characteristic of C. incana Macl., but the elytra are said to have “cing fossettes, deux a la base, une au milieu du disque .. . . deux au tiers posterieur.” If this “one” is placed on the suture I have never seen an example. If, however, it is in the usual position it surely has its “fellow” on the other elytron. The wording may be a “lapsus calami” or the type may be a “freak’”’ example. 4. C. carteri Obenb., from S. Queensland, described as having a non-carinate abdomen would point to the ¢ of my species (C. octomaculata, infra) but for two characters that are inconsistent with this—viz. (1) “circular coppery foveae superficial” on elytra, (2) underside “green in the middle, coppery on the sides.” In some examples of C. viridis Macl. the carina is subobsolete, so that the very variable viridis may include carteri Obenb. 5. ©. blackburni Obenb.—I cannot distinguish this from some examples of mastersi Macl. The only species with which Obenberger compares his three Australian species is C. australasiae Hope, and this is done so vaguely as to give little help, besides giving no indication of the species determined by him as australasiae. The species so determined by Kerremans (of which I have an example before me labelled australasiae by him from the British Museum) is clearly mastersi Macl. 226 AUSIRALIAN COLEOPTERA, The following synonymy seems certain, the doubtful cases being indicated by a (?):— 1. GC. incana Macl. = interioris Blackb. = ? regina Kerr. Macleay’s type exactly corresponds with Blackburn’s detailed description. 2. CO. saundersi Macl. = hopei Obenb. The description of the latter exactly corresponds with Macleay’s type. 3. C. viridis Macl. = frenchi Kerr. = ? simplicifrons Kerr. Kerremans’s descriptions do not indicate any clear distinction from the variable C. viridis Macl. C. frenchi Kerr.—The type has lost the head and thorax. Its inadequate description had led me to consider it as a synonym of the variable and widely distributed C. viridis Maci., while the drawing and notes sent me of its abdomen are quite in conformity with this synonymy. C. simplicifrons Kerr., is, I think, doubtfully distinct from C. viridis 92, though at present I cannot definitely call it a synonym. I have an example from the Dorrigo, N.S.W., which corresponds with Mr. Blair’s notes on the type. The apex of abdomen is rather squarely emarginate, with sharply defined medial carina (as is the case with the 9 of C. viridis, the ¢ often having this carina vaguely defined), the underside duller aeneous instead of brilliant green, eyes slightly wider apart, etc. Distinctive Characters.—The most distinctive characters that separate the species are found on the underside, especially in (a) the structure of the apical segment of the abdomen, (0) the colour and sculpture of the abdomen. Further differentiating characters lie in (c) the form and sculpture of the prothorax, (d) the micro-sculpture of the elytra and the form of the elytral foveae (fossettes). Finally, the size and colour may be helpful as secondary considerations, but are often fallacious guides. The occurrence of dwarfs—probably due to malnutrition of the larvae—is well known to collectors, while the ordinary variations of size in the species of wide distribution are considerable. In the long series (24 examples) of C. mastersi Macl. before me, the dimensions vary from 16 X 6 to 10.5 X 4 mm., and of C. incana Macl. from 20 X 7 to 12 X 4.5 mm., the females in general being larger than the males. Colour.—The colour of the upper surface is singularly uniform, compared with species of other genera. In general, more or less violet bronze, sometimes greenish or olivaceous, the colours, in old specimens, become nearly or quite black. In all the Australian species examined by me—with the slight but constant modification noted in C. octomaculata infra—the elytra have six foveate depressions, three on each elytron, of a brilliant metallic copper or green, similarly placed; though an obliteration of this metallic colouration by age or chemical action would seem to account for the different number of these recorded in descriptions. The size of these foveae seems constant in the same species, but varies in the genus sufficiently to serve as a useful aid in diagnosis. The first pair are placed within the angles formed by the prominent basal lobes, the second pair are medial on the 2nd costae, the third pair on the apical third, on the 3rd costae. Elytral sculpture.—In general there are four well-marked costae on each elytron, besides a short scutellary costa; but in some species only the first costa (nearest the suture) is clearly defined, parts of the others being indicated. The costae are prominent in mastersi, incana, subsimilis and amplicollis; less so in saundersi and australasiae, and vague in viridis and (?) frenchi. An excellent character—requiring a good lens or (better) a good binocular microscope—is the micro or ground sculpture of the upper surface, always BY H. J. CARTER. 227 constant for the species, but with considerable variation in the genus. Thus australasiae Hope is distinguished by more transverse rugae (very strong on pronotum); arcana, mastersi and saundersi have round punctures distinctly separated, with different degrees of density and size; viridis is scarcely punctate, but scalose, as in many Cisseis; saundersi and amplicollis are distinguished in having the pronotum almost entirely punctate, little strigose. Form of Prothorax.—The sides of prothorax are, in general, not very variable; the majority of species having what Thomson called “subhexagonal” outlines, the sides contracting obliquely in front and behind, with a larger medial space, sometimes concave. The following notable exceptions to the above are as follows:—in australasiae Hope, the sides widen from apex to near base, where they are rather abruptly and bluntly rounded (vide Saunders’ figure, Trans. Hnt. Soc. Lond., 1868); in amplicollis Thoms. the sides have a distinct bulge and are widest at the apical third; in viridis Macl. the sides are nearly straight, or feebly undulate. Abdominal characters.—The apical segment is strongly carinate in mastersi, saundersi, arcana; feebly so in australasiae, viridis ¢ and in the 2 of amplicollis, and non-carinate in subsimilis, caelatus, n. sp., and carteri Obenb., while the carina appears only in the female of octomaculata. The abdomen is varicoloured, chiefly golden with blue or violet apical margins to segments in australasiae, mastersi, amplicollis and octomaculata, concolorous in arcana, saundersi and subsimilis—the second of these almost black—and chiefly bright green with purple patches in viridis. The structure of the apices varies sexually, the ¢ generally having the wider excision, and often bispinose; arcana is remarkable for its trispinose apex, the carina being produced into a medial spine; suwbsimilis is exceptional in having a forked tooth on each side of a wide triangular excision and may thus be said to be 4-spined. The apical excision is small in saundersi: and in the 9 of australasiae. Distribution and Habit.—As with many other Buprestidae, some of the species have a very wide distribution. Thus, of the longer series before me, 33 of arcana, 19 of viridis, 24 of mastersi, 10 of subsimilis, I find that the first occurs widely in N. W. Australia and Queensland (1 ex. from Narrabri, N.S.W.); mastersi occurs in Queensland, Western Australia, South Australia, N. Victoria; saundersi in Queens- land and Western Australia; vividis in Queensland, New South Wales, and Western Australia; amplicollis in Western Australia and South Australia, subsimilis in Queensland, Western Australia and Victoria. Mr. J. Clark tells me that they frequent acacias, like Melobasis, and fly rapidly at the least disturbance. They are comparatively rare, few collections having them in any number. I have, however, been able, with the courteous assistance of my colleague entomologists of the British Museum and of all the Australian Museums, to examine a large number of species—many examples compared with types—as well as the Macleay types. The following are undescribed :— CHRYSOBOTHRIS CAELATUS, D. Sp. 6. Robust, widely oblong-ovate, above and below obscure violet-bronze, the depressed areas of pronotum and the six elytral foveae coppery, middle of abdomen obscure green. Head with embossed surface, rather strongly pilose, a curved “horse-shoe”’ impression between the eyes, with three elongate, shining nodules in front of this ‘and smaller irregular nodules near epistoma, the last feeviy sinuate, interocular F 228 AUSTRALIAN COLEOPTERA, area wide and sulcate-carinate (a short linear sulcus between carinate margins), frontal area regularly punctate, antennae wanting. Prothoraxr: Apex subtruncate, base very sinuous, medial lobe rather strongly produced and truncate at scutellum, widest in front of middle, sides angulately jagged, with a strong triangular emargination at apical third, thence irregularly narrowed to base, all angles produced and acute. Disc with wide medial sulcus, its basal third smooth, surface irregularly embossed; rugose-punctate near middle, with larger nitid knobs towards sides, with an underlying system of coarse punctures. Scutellum trian- gular, smooth. Elytra rather convex, thrice as long as prothorax and one-third wider than it, with six large, rather deep, coppery foveae in the usual positions, the basal and post-medial round, the medial rather square; the usual costae evident but not strongly raised, the surface punctures stronger than usual— larger and more widely separated than in CO. mastersi—with some rugae apparent near costae and at sides; margins behind finely serrate, underside coarsely punctate, the abdomen longitudinally rugose-punctate; apical segment of abdomen subcarinate, a short, rather flat, smooth, longitudinal rugosity forming a sort of carina; apex with wide concave excision, dentate at extremities. Dimensions: 17 X 7.2 mm. Habitat.—New South Wales: Mossgiel (in Australian Museum). Type, No. K.32692 in Australian Museum. © has the elytra more sharply attenuated to apex, the apical segment scarcely excised, the carina less evident (sub-obsolete). Paratype, No. K.32691 in Australian Museum. The female example is labelled Mitchell River, and, though differing as above, is clearly conspecific with the Mossgiel example. It is the widest and most -convex of the Australian species, and it may be differentiated by its roughly embossed pronotum with its dentate sides. Name from caelare, to emboss. CHRYSOBOTHRIS OCTOMACULATA, 0. Sp. Ovate, dark purple-bronze above, elytra with six large circular coppery foveae, together with an oblique elongate spot adjacent and exterior to the medial foveae; underside brilliantly varicoloured, metasternum and segments of abdomen golden- green and purple, the latter violaceous at anterior margins and sides, underside of femora golden-green and purple, tarsi blue. Head sparsely pilose, rugose-punctate, depressed parts coppery, epistoma with arcuate triangular excision, eyes moderately converging behind, interspace slightly wider than in viridis Macl., longitudinally wrinkled on vertex. Prothoraxr short and transverse, sides subhexagonal, sometimes undulate, with a small con- cavity in middle, more strongly and subsinuately narrowed behind than in front, equally wide at anterior and basal third, all angles produced and acute. Disc with medial line generally indicated by short depression, a wide transverse depression near apex, two shallow foveae on posterior half, surface finely, transversely rugose-striolate. Scutellum triangular. Hlytra more than thrice as long and nearly one and a half times as wide as prothorax, shoulders unusually squarely rounded, thence to apical third sub-parallel, margins on apical half rather strongly serrated, usual costae vague, often sub-obsolete on basal half—except the presutural—a wide convexity (not carinate) running obliquely from the shoulder separates the Medial fovea from the adjacent coppery mark, a wide depression following its course within this; surface minutely, densely and clearly punctate. Metasternum sparsely punctate, abdomen longitudinally rugose-punctate, apical segment non- BY H. J. CARTER. 229 carinate, slightly depressed in middle, with wide square excision limited by two prominent teeth; front femora with wide obtuse tooth. Type dg, No. K.32689 in Australian Museum. 2 differs in having the apical segment of abdomen clearly but not prominently ecarinate, with a narrower, semicircular excision limited by two wider teeth. Type 9, in Macleay Museum. Dimensions: 9-12 X 4-5 mm. Habitat—Queensland (Duaringa, Port Denison, Charters Towers, Brisbane). Ten examples are before me, five of each sex; the sexes only distinguishable as above, being alike in form, size, and in the unusual ornamentation of the additional coppery spot, forming an adjunct to the medial fovea but always separated from it. Three examples from the Macleay Museum include both sexes labelled Port Denison in Mr. Masters’s handwriting. The female is near C. black- burni Obenb., but besides coming from Western Australia this species is said to have the foveae as in australasiae Hope, which is not the case with octo-maculata. Var.—Two 92 examples (one from the British Museum) are only to be distinguished from the above by the absence of a clearly shown additional mark exterior to the medial spot. Table of Australian Chrysobothvris. il, ANDES Ghe Fajovsloponverny esos) GoscasoacauoosugecsunoooboooGKD500 subsimilis Thoms. ANGOERS Ge Aoyelornsynr VekyomNHK) Suosocaccanscgcsuceconvnoo0envaDD oon GOODS incana Macl. BANE Ot Howley ZomOs® soocaoécacnosuensgnsvsoouoUbadNooDODDHUUOU aC OOOO DD 2 2. Form wide (7 mm.), sides of pronotum angulately toothed ......... caelatus, n. sp. Form normal (less than 6 mm. wide), sides of pronotum not so toothed .......... 3 Seeeronotuml’ widest Near baSellatses « sclatacds «cls c.elelelolsiens « o) atsle ates australasiae L. & G. IBronotumenol widestsbehind smiddle! ieee cuss « clels cues © «oracles « sl cnenenelcieie ys) ciehelauenss 4 4. Sides of pronotum with obvious enlargement in front of middle .. amplicollis Thoms. Sidesmotspronotumiunearlyastraionit sey se aise cit ca cictecheichess sicielcieeuenec viridis Macl. Sides of pronotum subhexagonal (subequally attenuate each way) .............. 5 5. Apical segment of abdomen carinate in both sexes .............0 cece eee eceees 6 Apical segment of abdomen carinate only in @ ................ octomaculata, n. sp. 6. Underside concolorous (nearly black) .............e eee e cece eeaee saundersi Macl. Mindersidembrillianthy wwaricoloroush secede deca oer crenata mastersi Macl. The following captures are chiefly the outcome of two very interesting expeditions:—(1) With the Royal Australasian Ornithologists’ Union camp at Byfield, near Yeppoon (Rockhampton district, Queensland) in October, 1924; (2) with the University Expedition, organized by Professor Launcelot Harrison, of Sydney University, to Barrington Tops, New South Wales, in January, 1925. ASTRAEUS INTRICATUS, n.sp. Text-fig. 1. Nitid, head metallic greenish at apex, purple at base, pronotum metallic purple, brighter at sides, elytra cyaneous, with the following yellow markings on each; an elongate oval basal spot, an elongate lateral mark extending from shoulder to about the middle, having an oblique inward extension over four lateral intervals, an irregular postmedian, subfasciate mark connected with a subsutural club- shaped spot extending forward behind the basal spot, and also with a triangular mark extending backward to the base of apical spine; underside and femora brassy bronze, tibiae and tarsi testaceous, antennae purple coppery. Head closely and strongly punctate, longitudinally carinate in middle, and, like the pronotum, clothed with fine yellow hair. Prothoraz obliquely narrowing * Australasiae L. & G. has the apex of abdomen arcuately excised in ¢, with a small triangular excision in 9, non-spinose in both sexes; but has been included under 2-spinose species from the difficulty of clear distinction between this structure and some of the 2-spinose species. 230 AUSTRALIAN COLEOPTERA, from base to apex, sides scarcely arcuate, apex nearly straight, base strongly bisinuate, medial line indicated on basal half by narrow smooth space, on apical half by wide depression; whole surface coarsely punctate, the sculpture close and subrugose at sides, more sparse on disc. EHlytra striate-punctate, the striae outlined by finely cut costae, intervals between costae flat, containing shallow punctures, elytra divergent and bispinose at apex, the sutural spine robust; prosternum coarsely, abdomen finely and rather closely, punctate; underside rather densely tomentose. Dimensions: 10 X 4 mm. Habitat.—New South Wales: Monaro district. A single male example in the Macleay Museum appears to be nearest to A. vittatus VY. d. Poll. from N. W. Australia; but the pattern of the elytral colours is distinct, while there are other colour differences, e.g. vittatus has the pronotum bronzy green in middle, and the head black with purplish reflections. Type in the Macleay Museum. STIGMODERA (Subgen. CASTIARINA) HARRISONI, n.sp. Text-fig. 2. 9. Oblong-ovate; head, pronotum, underside and appendages brilliant brassy green, the basal area of pronotum showing slight purplish gleams. Elytra blue- green with the following light markings on each—a sub-basal triangular spot (having two of its sides parallel to base and suture respectively) connected with a blood-red lateral mark behind the shoulder; a narrow, lunate, medial spot reaching neither side nor suture, a short narrow preapical fascia extending obliquely backward from the 4th interval, connected with a blood-red lateral macula extending over three lateral intervals backwards, half-way between the fascia and apex; the discal markings yellow. Head excavated and channelled between eyes; irregularly punctate, the punctures small and close near base. Prothorax widest at base, arcuately narrowed to, and subsinuate near, apex; moderately bisinuate at base and apex, posterior angles rectangular, anterior acute; disc without medial sulcus, save for a large basal fovea; two punctate excisions at base (half-way between scutellum and side); disc irregularly punctate, the punctures nowhere dense, but closer and finer on apical half. Scutellum large, triangular, depressed in middle and finely punctate. EHlytra rather flat, of same width as prothorax at base and thrice as long as it, sides slightly widened at shoulder, very feebly compressed behind this, slightly widest behind middle; apices widely bilunate, with a defined but short external tooth, the sutural angle at apex separately rounded and feebly produced; margins entire; striate-punctate, intervals quite smooth and nearly flat, under- side glabrous and finely punctate, sternum sparsely, abdomen densely so; last segment rounded behind. 3 latet. Dimensions: 17 X 6.4 mm. Habitat—New South Wales: Barrington Tops. A beautiful species belonging to the producta Saund.-spectabilis Kerr. group, that would stand next to the latter in my tabulation, from which it differs as follows:—(1) Larger size and flatter form; (2) apices wider with less pro- nounced teeth; (3) quite differently shaped yellow maculae (the sub-basal trian- gular, the four hinder being narrowly fasciate), the same in spectabilis being larger and round. vs BY H. J. CARTER. 231 A single specimen taken by the University expedition is dedicated to Professor Launcelot Harrison, who so ably organized and conducted the research on Barrington Tops in January, 1925. Type in the Macleay Museum. STIGMODERA SUBVERSICOLOR, 2D. Sp. 6. Hlongate, cylindric; head, underside, and appendages coppery, pronotum dark blue with undefined violet-coppery margins; elytra blue (in one example violaceous) with the following markings yellow:—two large post-basal spots* in general produced to basal margin, thence again backward to beyond the humeral swelling; four rounded spots midway (two discal spots in advance of two marginal), a preapical undulate fascia, widely interrupted at the suture, widening in each direction on margins. Head deeply excavated and channelled, closely uniformly punctate. Prothoraz rather tumid, apex subtruncate, base moderately bisinuate, sides well rounded, more strongly converging to apex than to base, feebly sinuate near the produced, acute hind angles; disc with medial channel distinct, not. deep, closely punctate, the lateral sculpture becoming coarse and subrugose. Hlytra at humeral swelling slightly wider than prothorax, feebly constricted behind this, apices slightly divergent and finely bispinose; margins entire; striate-punctate, the intervals clearly punctate and evenly convex on apical, flat on basal half; underside rather closely clad with white recumbent hair. %. Pronotum more or less concolorous (in one example greenish-blue), underside bronze-black. Dimensions: 10-11 X 3.5 mm. Habitat—wWestern Australia: Tammin (Mr. J. Clark). Four examples, two of each sex, sent by my friend, Mr. Clark, show a species more robust in form, but close in pattern to S. versicolor L. & G. In all examples ‘of versicolor that I have seen, the pronotal margins and underside are a brilliant metallic green in both sexes; while in the above described the sexual colouration is marked, while the pronotal metallic gleams are only noticeable in one example under observation and in this case with ill-defined limits. In three examples of versicolor before me, the prothorax is more parallel and less tumid than in my species, while the preapical fascia does not extend to the sides in two examples, and just reaches them in the third. The variable post-basal markings occur in both species. Type in Coll. Carter. Archaeozodes strandi Obenb. = Stigmodera versicolor L. & G. Dr. Obenberger’s description exactly tallies with this well known species. The genus Archaeozodes is thus redundant. Dr. Obenberger persistently describes Australian Buprestidae as new species on very finely drawn differences. Thus in Archiv fiir Naturgeschichte, 1924, p. 69, one of the most widely distributed of Australian insects known as the fire beetle,; Merimua atrata, is redescribed as M. corporaali from New Guinea. I have specimens from an island in Torres Strait which correspond with his description, but which I consider but slight variation from the normal. Such variations may be found by examining long series of most common insects. Again Briseis >In one example the post-basal spot is isolated from the marginal yellow area. ¥+ From its curious habit of settling on hot ashes from camp and bush fires, slag heaps and its attraction to strong light. 232 AUSIRALIAN COLEOPTERA, sagitta Obenb. (p. 68 of the same publication) cannot, I think, be differentiated from the well known B. conica L. & G. Neotorresita achardi subsp. occidentis Obenb.—I have already stated (THESE Proc. 1924, p. 526) that Neotorresita achardi Obenb. = Pseudanilara cupripes Macl., and have in the same place described as new, Ps. occidentalis, with clear distinc- tion from its allies. This is apparently not the subspecies noted by Obenberger. The names occidentis and occidentalis are unfortunately close. Paracephala impressicollis Obenb. must be very near P. trans-secta Cart., but in his description, the author does mt indicate the relation of species to P. trans-secta Cart., nor indeed to any other member of the genus. This omission constitutes a serious deficiency in descriptions, especially in the case of genera in which the species are difficult to identify, as in Cisseis. Thus Cisseis carteri Obenb. (p. 109 of the above) is described without even a reference to the groups into which I divided the species in the ‘interessante Bearbeitung”’ politely mentioned by Dr. Obenberger. The same remark applies to the descrip- tions of five species of Cisseoides—a genus which Obenberger considers that I wrongly merged with Hypocisseis. Lucanidae. CERATOGNATHUS BITUMULATUS, n.sp. Text-fig. 3. 6. Robust, brownish-black, sparsely clothed with yellowish-grey squamules, mandibles rather long, curved, without lateral or vertical auriculations, apices bidentate, with a third internal tooth, triangular, a little behind apex, the basal two-thirds with a laminate thickening having a sub-crenulate or granulose upper edge and terminating abruptly towards apex; clothed sparsely with silky hair and punctate externally. Head punctate, with two round tubercles on front, eyes large and moderately prominent, the flabellae as long as the rest of the antennae. Prothoraz feebly bisinuate at apex, more strongly so at base; sides rather widely and evenly rounded, all angles obtuse; lateral margins uniformly, horizontally explanate, extreme border crenulate, disc regularly and rather closely punctate, medial line smooth and slightly convex on basal half, an oval, slightly flattened depression Text-fig. 1. Astraeus intricatus, n. sp. Text-fig. 2. Stigmodera (Castiarina) harrisoni, n. sp. Text-fig. 3. Ceratognathus bitumulatus, n. sp. Text-fig. 4. Ceratognathus ocularis, n.sp. <6. Text-fig. 5. Ceratognathus ocularis, n.sp. 9. BY H. J. CARTER. 233 on apical half. Scutellum semicircular, punctate. Hlytra sub-parallel, wider than prothorax at base, somewhat irregularly punctate, the punctures tending to run in ill-defined longitudinal series, some sub-obsolete costae just visible; each elytron with a blunt, round tumulus on the apical declivity, more or less covered with squamules. Sternal area punctate, abdomen longitudinally rugulose. Fore tibiae more spinose than usual in the genus, with three strong apical spines and about six others of larger size than the usual marginal denticulations; hind tibiae a little swollen at middle on interior margin. Dimensions (including mandibles): 14 X 6 mm. Habitat.—New South Wales: Mt. Kosciusko (H. J. Carter, Jan., 1906) and Barrington Tops (H. J. Carter, Jan., 1916). 1 ©. Rather wider and more oval, the mandibles short and stout, the antennal flabellae much shorter. Dimensions: 13 X 6 mm. Three examples examined, ¢ and 9? from Kosciusko, and a ? from Barrington Tops, show a species clearly separated from its nearest relative C. gilesi Blkb., by the presence of the elytral tumuli and the differently formed mandibles of the 6. C. gilesi also has a single large tubercle on the head, the pronotum has smooth spaces, the margins are not explanate, and there is a deep medial sulcus. From C. flabellatus Boil., and OC. macrognathus Boil. it differs inter alia by the differently shaped prothorax—in both these species the sides being “presque droits.” Types in the National Museum, Melbourne. CERATOGNATHUS OCULARIS, n.sp. Text-figs. 4, 5. 6. Oblong, convex, brown, whole upper surface with sparse golden squamules. Head, as with the whole upper surface, cellulose-punctate, frontal area without tubercles, clypeus trilobate, the antennal orbits forming acute convex lobes pointing obliquely forward, a third smaller subconical process jutting horizontally from the middle of the clypeal area, mandibles short, wide seen sideways, non-auriculate, trifid at tips, the middle tooth longest; eyes round and very prominent, antennae slender, the leaflets nearly as long as rest of antennae. Pro- thorax: Sides rather strongly and evenly rounded, anterior angles subacute, posterior obtuse, margins a little explanate, extreme border minutely serrulate in places; medial sulcus wide throughout and emphasized by a line of golden squamules on each side, more uniformly placed than elsewhere. Scutellum large, oval. Elytra parallel, convex, wider than prothorax at base, suture on apical half widely costate and with three other, fairly well defined, wide, subcostate impressions continuous from base to apical declivity. Legs slender, fore tibiae finely serrate on exterior edge, having one larger spine on apical third besides the longer but fine apical spines; tarsi very slender, hind tarsi with claw joint as long as the rest combined. Underside coarsely punctate, the punctures close on abdomen, more distant on sternal area, prosternum carinate. %. Differs in having the head much narrower, without the defined clypeal lobes, eyes not prominent, mandibles very short, the antennae with leaflets about half as long as in g¢; prothorax having anterior angles close to head, followed behind by a wide sinuation; size. generally larger. Dimensions: 3g, mandibles included, 7.5 X 3 (+), 98 X 3.5 mm. Habitat.—New South Wales: Barrington Tops (Sydney University Expedi- tion); Mount Wilson (H. J. Carter). Twelve examples (4 g, 8 ¢) show the smallest of the described Australian species, distinguished by unusual sexual characters, e.g., clypeal structure, promin- 234 AUSTRALIAN COLEOPTERA, ent eyes of 4, differently shaped prothorax, besides the usual differences found in the mandibles and antennae. A single 2 only from Mount Wilson. Type series in Macleay Museum, University of Sydney. Table of Australian Species of Ceratognathus Westw. ieMandiblesioticy withverectulateral horny ryaecrcrsedeioen-nertelev stele ete eiel tote ieuere le kole done enelionare 2 Mandgdiblesro£ iouawithouteerect lacena le DOr) segs) o retin isle nei cuisine i-iicl taal ent wemens 3 Oe GALEFAly MOLI (ACUCE as cus ec aie exetayoserenehe feels on hecene renelreysacieieree oacteceeereseeuetecusite niger Westw. Lateral horn laminate (subquadrate) ...............2200+e eee euee froggatti Blkb. 3. Mandibles of &@ with careniform enlargement on upper margin terminating in a GeEnti£orm: PKOCESSs 5, sie Misys: & anccasvaeuchtwe teks 8 Gio o odors, bus RSs edomte tte robewe Wien emeniawe came 4 Manadibles: Of CG MOC: SO iia sus acces cave ods spas Cees deel © ep si ohn) Seu ske Ons BPORMAT Renee LOE Ren ieke 5 45 "Colour ‘Chestnut=PrOwI cree se sie erase shee lacie tava veoorevap cualfeisncaprenanie cre onaiarrn euaen sae enchedei entre ear 6 Colour sfuscoussblack G22 Hess. BTR a 5 ALE OES. TR ee ee 7 5. Mandibles of @ with round medial lobe on inside margins ....... mentifer Westw. Mandibles! without lobesvormlaterale toothy secre ocr cilchensieliensieieicl nme cnei-neaemens 8 6. Mandibles curved, sides of pronotum rounded ................. westwoodi Thoms. Mandibles and sides of pronotum nearly straight .......... *macrognathus Boil. 7. Pronotum strongly sulcate, with smooth areas ..............+-+2e00 gilesi Blkb. Pronotum not (or scarcely) suleate without smooth areas .... bitwmulosus, n. sp. 8. Size large, 15 mm. (or more) long, sides of pronotum nearly straight .......... 9 Size small, less than 9 mm. long, sides of pronotuin rounded .................. 10 Se shrect tooth) on clypeusrabove each (Cye ae ae eer coe elec oienteeienelas frenchi Blkb. This) clypeail ttoothvabsenty 6 tyac.. Oa ie sees Soe tetas oe eiape cheseeele caches *flabellatus Boil. 10.. Head with frontal tubercle, transversely bipartite .............. rufipennis Westw. Head without such tubercle, clypeus tricristate .................. ocularis, n. sp. Although the species described by M. Boileau are unknown to me in nature, they are described in such detail—one is also well figured—that I have ventured to include them in the above table. This table only takes into account the male characters; the females would be difficult, in some cases impossible, to tabulate. C. abdominalis Parry from Moreton Bay, Queensland, was described from a unique 9. Monsieur Boileau has suggested its probable synonymy with C. froggatti Blkb. Tenebrionidae. SARAGUS WILSONI, DN. Sp. 6. Widely ovate, rather depressed, glabrous, the greater part of surface reddish-brown, foliate margins, antennae and tarsi castaneous, pronotum, pro- sternum, greater part of head and medial area of abdomen black. Head minutely and irregularly punctate, epistoma straight in front, oblique with reflexed margin at sides, eyes separated by a space of the width of one eye, antennae with joint 3 nearly as long as 4-5 combined, 4-8 successively shorter, 9-10 nearly round, 11 oval. Prothorax: Disc convex, foliate margins wide and concave, total width : width of disc as 11:8. Anterior angles advanced and bluntly rounded, posterior acute and falcate, lateral border fine and reflexed; disc micro- scopically punctate, medial line very lightly impressed, without evident basal foveae. Scutellum transversely oval. Hlytra ovate, of same width as prothorax at base, and rather more than twice as long; foliate margins as wide as the pronotal, continuous to, but narrowed at, apex; the red colour dotted through with round, black, subdermal spots, seen alike from above and below; border narrowly reflexed, disc nearly smooth, with a few subobsolete costae faintly indicated and (with a lens) line of minute punctures just visible, except near sides, where, besides the usual lateral row of large punctures, the adjacent two or three rows of small punctures clearly shown; prosternum finely transversely * Unknown to author. BY H. J. CARTER. 235 striolate and sparsely pustulate, episterna more strongly rugose, abdomen minutely striolate; fore legs having two basal tarsi considerably widened, hind legs having basal tarsus as long as apical, the intermediate short. Dimensions: 17-18 X 12 mm. @ wanting. Habitat.—? Western Australia (C. French, per F.. EH. Wilson). Two male examples sent by Mr. Wilson who obtained them from Mr. French with other Western species, unlabelled. The species is not much smaller than S. magister Pasc., from which it differs in colour, flatter elytra and smoother surface, almost impunctate to the naked eye, except at sides. It is much wider than perlaevis Cart. The curious subdermal spots on the elytral margins are characteristic, as are the wider basal tarsi of the fore feet, the latter being probably sexual. Both examples are similar in colour which may, or may not, be due to immaturity. Type in Coll. Carter. Paratype in Coll. Wilson. NyYCTOZOILUS LATERALIS, n.sp. Text-fig. 6. Opaque brown-black, ovate, convex; tarsi and antennae reddish. Head: Surface scabrous and subpunctate, rather abruptly widened at antennal orbits, forehead with deep, medial, longitudinal sulcus extending from base to clypeal suture, antennae having 3rd segment longer than 4th-5th combined, the 1ith longer and narrower than 10th. Prothorar arcuate-emarginate at apex, anterior angles acute and prominent, the tips slightly blunted; base nearly straight in middle, sinuate near sides, posterior angles acutely produced back- wards; sides evenly rounded, widest in middle, subsinuate near front and hind angles; the lateral border thick and convex, a deep and fairly wide sulcus within; disc sparsely rugose, without distinct punctures, a medial sulcus clearly shown on greater part of length, obsolescent near apex and base, a large irregular foveate depression on each side of this. Scutellum widely transversely triangular. Hlytra wider than prothorax at base, widely rounded in humeral region; each with three finely crenulate costae, the exterior of these shortest and sometimes obscure, besides a geminate sutural costa bifurcating behind scutellum to join the first of above costae at base, the costae becoming obliterated on apical declivity; interstices irregularly rugose and subfoveate without reticulation or transverse ridges; the usual lateral row of large punctures irregular and often undefined. Beneath subnitid and almost impunctate, apical segments with very minute punctures; tibiae feebly tomentose; posterior tarsi with basal joint as long as the rest combined. Dimensions: 16-17 X 8-9 mm. Habitat—Queensland: Byfield near Yeppoon (T. G. Sloane and H. J. Carter). Three examples (2 ¢, 1 9) were taken under dead Eucalyptus branches. The species belongs to the first group of my table, where it would stand between N. ruficornis Cart., and mastersi Macl., having somewhat similar elytral sculpture to the former and the thick pronotal border of the latter. The strong frontal sulcus is a notable character. Type in Coll. Carter. As eight species have been added since my tabulation of Nyctozoilus was published (Ann. Q. Mus., 1911, p. 9), I append an emended table for the help of students. Table of Nyctozoilus Guér. 1, Pronotum rugose or squamose; not, or very vaguely, punctate ................ 2 PronGtums. clearly» PUNCLALS, cxa-cyscswe tone spay ar swwavsorerco rants Weoreiroiey etrsonaire Vowervow'a attayier enrayteyronen folate ve\larrevesetre's 8 2. Lateral border of pronotum crenulate ....................--e+e0- deyrollei Bates wateral border Of pronotumy entire) jaca cere oe eieenoe een oeiereciecrerciers 3 236 AUSTRALIAN COLEOPTERA, 3. Elytra 6-costate, intervals not, or very vaguely, reticulate ...................... 4 HlvtramG-costatewintenvalscleanivane ti Clllake mewn itrticasiien meine kneels icn-ieleaaeniels hel welts 6 47 slwateral border of Pronotum, thine 6, .rc-c-opens cise eueueese) © Samson erie dee vai obs levee seus (oe coronas earache teres 5 ILBVIGIEN InoCkasP CE jOROROWOI WAG senoosasooaccovadsbooucoolbobuNE lateralis, n. sp. 5. Lateral border of pronotum refiexed, hind angles produced ...... carbonarius Cart. Lateral border of pronotum sub-horizontal, hind angles not produced ruficornis Cart. G2leateral border Of PRONOULIUMI EIN erence ecee cherer er neheeon cn ell Beier ciel ok wen evened oiePebonen one nen ets te LEK! loyopraleye ie joperorooynsbon WYOAY Wake Go cccaddcodgacnucdounuauUucG mastersi Macl. 7. Pronotum coarsely rugose, anterior angles subrectangular ........ obesus Guér. Pronotum smooth (felt-like), anterior angles acutely produced .... denticollis Cart. Seely trasrneticulate, notticostacem asm ected aeienelelet = cheers eerie iene hotels irregularis Blackb. IN tras COSEAEC ry aen- 5 epegene enousle amed ou eg sivels Wotenopron ch taarieye nel sy shies oRepc hement pepe etel ave ctaleicok te coh NeR welt Renonetiots 9 9 OLYtral “8 =COSCAEC. 5. Aa. 5 ora cap shronssy cyai nya ergor'eyray ohiey aiatieigai aecatecnsecoptetiatioatietisiieneliomege) aush euch epekewehelspencieier ole 10 LY UPA C=COSCATC) ra ncare o tehena canoe. al) uy reg: a1 sieves: ower syrehelfov suai 1 eharlevowotietall seewalexeyrona\conehe evenemeuensyeneiene 15 HOE Size Nar ceyalo=2Oherviin.: LOM Leth Ae ei hae tahareltes be etaeteton gle biterute eis) onero Us Neem sieve De Rete Leo teee 11 Sizexsmaill, wi2)-mmsglone tonrdléssiras nate hie ao Se es oh Se eee nee 13 Lele Salty traliCOstae Mund wate wee. phereieterct sn otctienetelectoneuctov cherie) ellellolieWiel Mei levehenedchonemensicne tohcaekeme Mer Reece 12 Blytraliicostae’ Straight ys oe chags « ceetotae 2 ol eee cts Gite iodehsyeuisncl cuayaisl ences puncto-costatus Cart. 12. Elytral intervals irregularly reticulate (20 x 10 mm.) ....... marginatus Cart. HKlytral intervals foveate-punctate (1/6 x 9 mm.) .........-..::.... crassus Cart. 18, IDA! ieavwewis Cleeyihy eee oosoocancscunadd0gucB0bebODDONC pusillus Cart. IDIKAURUL MIA Woe (Oe Ay warbowoehy)) eNO) ooaganodusuododuconeubononoos 14 14. Blytral intervals with transverse ridges -.........-........++:--+-. *daemeli Haag. Blytral intervals subobsoletely vermiculate .....................+0- parvus Cart. 15a by tral aintvenvalsmclearly, sre blicula emacs cis rewoneuefoie ep cutensvene ferelenemeusdeichenheuerener: kemeeoisioneions 16 Elytral intervals obsoletely or indefinitely reticulate .......................0-. 20 1G, Chae lepese, WSO samen, OMNES cocododonsdedsgncbodncondD Dd ObU doo Oooo dO OOS GOODS el Sine: Saoeyll, WD ieee, Wome Or WAS ooosoocousgcccndcg0nGDOO EDO DOOD DSOOBODDGD OOS 18 17. Pronotal border thin, hind angles directed outwards ........... reticulatus Bates Pronotal border thick, hind angles directed backward ........ carlovillensis Cart. UD, I2iOMOUohoN Wry ThAChiGubOORy GINO oaonsuccousouducnscc Doo uauueo Uo uO OCOOONOOE 19 IPyeoavowbion OAK, THUONOHNE sdcosccenobodosc0osoccnddacucGgGuG j inaequalis Blackb. 19. Pronotal margins reflexed, extreme border thickened ...... erassicornis Blackb. Pronotal margins horizontal, extreme border thin .......... approximatus Blackb. 20. Sides of pronotum sinuate anteriorly and posteriorly .......... hardcastlei Cart. SiGes OF rome Siw GmMk; pwoOseModhy scocacacdacagoso5c0bC ob DDOUnOOODHNOS 21 21. Elytral intervals indefinitely reticulate ...................... vermiculatus Cart. Elytralsintervals feeblys, wrinkled! sees ie eee aie sexcostatus Champ. BYALLIUS ANGUSTATUS, n.sp. Text-fig. 7. Elongate, ovate, convex, subopaque black; apical joints of antennae pale brown, tarsi beneath and apices of tibiae clothed with red tomentum. Head finely punctate, epistoma truncate, forming an angle with antennal orbit; antennae enlarging to apex, joint 3 as long as 4-5 combined, 8-10 transverse, 11 oval. Prothorax considerably narrower at apex than at base, widest near base; apex arcuate-emarginate, anterior angles acutely produced forward, base sub- truncate, sides very little widened, lightly widening to one-third of length, thence subangulately but feebly widening to middle third, thence roundly contracting behind, posterior angles obtuse; margins very thick and subvertically revolute, sulcate within; disc nearly smooth, punctures only to be seen with the aid of lens; without foveae or medial line. Scutellum widely triangular and shallow, with triangular depression behind. Elytra at extreme base narrower than pro- thorax, shoulders obsolete; narrowly ovate, apical declivity steep; disc vermicu- lately rugose-punctate, with three well-raised undulate costae besides the geminate sutural costa (bifurcating behind the scutellum), intervals irregularly rugose and punctate, a row of large punctures at sides. Underside minutely punctured, the *I am a little doubtful of the distinction between parvus Cart. and daemeli Haag. 7 Species unknown to the author. Wi BY H. J. CARTER. 237 first three abdominal segments with fine longitudinal wrinkles; post tarsi with basal segment nearly as long as the rest combined. Dimensions: 15 X 7 mm. ; Habitat—New South Wales: Nerriga, between Goulburn and Braidwood (H. G. Carter). A single example (? 4) was lately taken by my son. It is very distinct from the six described species by the combination of smaller, especially narrower, form (of both prothorax and elytra), the strongly thickened and elevated border of pronotum with its minutely punctured disc. It is perhaps nearest to B. ovensensis Cart., in which, however, the pronotal border is narrower and less raised while the anterior angles are outwardly directed. Type in Coll. Carter. Byallius (Styrus) revolutus Cart.—When describing Styrus revolutus (Trans. Roy. Soc. S. Aust., 1914, p. 381) I commented on its marked distinction from other species of Styrus in the “entire, thickened, strongly revolute margins to prothorax—its truncate base” and its similarity to Byallius in the form of the prothorax. I now find that these characters point to its inclusion in Byallius, to which genus I would now refer it. CHARIOTHES VARIPENNIS, N. Sp. Shortly oblong, glabrous and nitid. Head, pronotum and underside black, elytra varicoloured, legs red-brown, antennae, palpi and tarsi pale red. Head: Epistoma semicircular, separating sulcus nearly straight, much more finely punctate than frontal area, the punctures on the latter rather coarse; antennal joints 1-2 short, 1-6 cylindric, 5-11 gradually increasing in width, 11th largest, oval. Prothorax subquadrate, wider than long, apex nearly as wide as base, the base produced a little backwards in middle, anterior angles widely obtuse, posterior rectangular, lateral border narrow, without foliation, an elongate depres- sion near border, a transverse depression near basal margin, disc sparsely and finely punctate without a sign of medial line. Scutellum triangular. Elytra of same width as prothorax at base, subparallel for the greater part, rather bluntly rounded at apex, lateral border very narrow; striate-punctate, each with eight striae, besides a short scutellary one containing close, moderately large punctures, intervals convex and impunctate; the suture bright green, the two adjacent intervals purple, the next three dark green, the lateral three purple, coppery and bright green respectively. Underside minutely and faintly punctate. Dimensions: 6-6.5 < 2-2.3 mm. Habitat.—Queensland: Cairns district (Dr. J. F. Illingworth). Two examples sent by Dr. Illingworth to the Imperial Bureau of Entomology were presented to the British Museum. The species is near (. cupripennis Pasc., and may be only a geographical race of that species. Mr. K. G. Blair has kindly compared it with Pascoe’s species and notes the following differences :— C. cupripennis Pasce. C. varipennis Cart. Antennae and tarsi dark. Bright red. Elytra feebly striate, interstices nearly flat. Strongly striate, punctures much deeper and stronger. Colours not longitudinally arranged, humeral and Colours in marked longitudinal apical areas reddish-coppery, suture green, bands. median area mainly brassy. Type in British Museum. 238 AUSTRALIAN COLEOPTERA, Chariothes (Menephilus) brevis Cart.—I concur with Mr. Blair’s opinion that this species should be transferred to Chariothes. It differs from C. varipennis in having more brightly coloured elytra, produced anterior angles to prothorax, the margins narrowly but clearly foliate, disc densely and strongly punctate, elytral intervals flat inter alia. Text-fig. 6. Nyctozoilus lateralis, n. sp. Text-fig. 7. Byallius angustatus, n. sp. Text-fig. 8. Cardiothorax hopsoni, n. sp. Text-fig. 9. Cardiothorax harrisoni, n. sp. CARDIOTHORAX HOPSONI, n. sp. Text-fig. 8. 6. Elongate-ovate, nitid dark bronze above, black beneath, antennae and tarsi reddish. Head: Epistoma arcuate (concave) in front, subsinuate at sides, antennal orbits rather tumid, frontal impression deep and scutate, its central area raised, antennae stout, gradually widening outwards. Prothoraz elongate, widest in front cf middle, apex arcuate, anterior angles prominent and rather acute, sides moderately rounded, converging rather strongly and sinuate before the posterior angles, these subrectangular, with a short blunt tooth produced backward, base slightly arcuate, this emphasized by the produced angles; foliate margins moderately wide on apical half, strongly narrowing to base, without separating sulcus or setae; disc with a clean-cut medial channel, a light transverse depression near base and posterior angles. Scutellum smooth, oval. EHlytra widely ovate, wider than prothorax at base, epipleural fold evident but not prominent, humeri obliquely rounded off, elytra striate (or sulcate) punctate, fine and rather close punctures clearly visible in the deep striae, intervals 3 and 5 wider than rest and moderately convex on disc, becoming more strongly so near apex; epipleurae and underside impunctate; the hind femora with a strong blunt tooth directed inwards; mid-femora swollen and sub-dentate, hind tibiae moderately enlarged and flattened. ®. Without these special femoral and tibial characters. Dimensions: g, 21 X 7.3 mm.; 9,19 X 7 (+) mm. BY H. J. CARTER. 239 Habitat.—New South Wales: Barrington Tops (Mr. John Hopson). Three examples were taken by Mr. Hopson, the well known pioneer and naturalist of the district, in Fagus brush on the nine-mile spur of the plateau, of which two (the sexes) are under observation, the third being snapped up by a tame magpie. It is a large species that combines to some extent the sexual characters of femoratus Bates and the walckenaeri Hope group, with a posterior emargination to the prothorax that is unlike any other, though approaching Hope’s species. The distinct seriate punctures on the elytra is a very rare feature in the genus. Type in Coll. Carter. CARDIOTHORAX HARRISONI, n. Sp. Text-fig. 9. EHlongate-ovate, convex, very nitid bronze above, black beneath, antennae and tarsi piceous. Head: Epistoma rather bluntly pointed and reflexed in front, its sides oblique, frontal impression rather small, foveate within (in some examples impressed as if by a cloven hoof); antennae less stout than usual, lightly enlarged outwards. Prothoraz 4 X 5 mm., widest near middle, apex arcuate, anterior angles rather sharply subrectangular, sides moderately widely and evenly rounded, a feeble sinuation near the very small posterior tooth (this little more than a slight pro- duction of the border, downward and obliquely outward), base subtruncate; foliate margins horizontal and moderately wide, bearing two to three distinct setae, with a short suleus separating it from disc, extreme border finely reflexed. Scutellum large, oval, convex and. smooth. Elytra: Shoulders little defined and obliquely rounded; sulcate, with six well marked sulci on disc, the seventh and sometimes the eighth visible on sides; intervals sub-depressed; epipleurae and underside smooth and impunctate. Dimensions: 15-16.5 X 5.5-6 mm. Habitat—New South Wales: Barrington Tops. Six examples taken by the University expedition of January, 1925, in the Eucalyptus country near the Bull’s Head Crossing. The sexual characters are not well defined, shown chiefly by the enlarged tarsi and widened apices of front tibiae and the slightly curved under margin of hind tibiae of the male. The species is nearest to C. macleayensis Cart., from which it differs in larger size, more convex form and different elytral sculpture, macleayensis having only five clearly : marked sulci on each elytron, with more convex intervals, also with larger and more out- wardly directed hind angles to prothorax. Type in the Macleay Museum. ADELIUM SUB-LAEVIGATUM, ND. Sp. Widely ovate; sub-opaque black above, nitid black beneath; antennae and tarsi reddish-brown. Head sub-laevigate (clypeus sparsely and minutely. punctate), the clypeal suture strongly impressed, a frontal triangular impression based thereon, antennal segments sub-conic, 3 as long as 4-5 combined. Prothoraz rather flat, widest behind middle; sub-laevigate, the only evident sculpture consisting of (1) a feebly impressed medial line near basal half, not extending to base, (2) some undefined depressions separating foliation from disc, (3) a small deep fovea on each side placed on this depression behind the middle, (4) a few minute, shallow punctures on disc. All angles obtuse, the posterior the more widely so; sides widely rounded, with a thin reflexed edge; feebly sinuate before the hind angles; base sub- 240 AUSTRALIAN COLEOPTERA, truncate, the hind angles produced a little backward. Scutellum semicircular. Elytra: Shoulders widely rounded; striate-foveate-reticulate, the striae near suture almost entire (without transverse septa), the elongate reticulations more defined and short towards sides. Underside glabrous and laevigate, except for lateral foveae of abdomen, and the epipleurae with shallow punctures. Dimensions: 15 * 7 mm. Habitat—Queensland: Byfield, near Yeppoon (H. J. Carter). A single example taken under dead boughs, belongs by its elytral sculpture to the geminatum-reticulatum group, but differs from all described species thereof by the opaque black, sub-laevigate surface of its pronotum and the non-dentate posterior angles; in the latter character resembling A. plicigerum Pasc. Type in Coll. Carter. SEIROTRANA ANOMALA, D0. SD. . Oblong-ovate, depressed, bright bronze above, darker bronze beneath, apices of femora testaceous, tarsi reddish, antennae piceous. Head densely rugose-punctate; antennae submoniliform, gradually widening to apex, 3 not much longer than 4, 11 elongate-oval. Prothoraz 3 X 4 mm., arcuate- emarginate at apex, anterior angles rather widely acute (apex slightly blunted), base subtruncate, sides moderately rounded on anterior three-quarters, then rather strongly sinuately incurved to the rectangular posterior angles, extreme margins finely, irregularly crenulate (in two out of nine examples this crenulation sub-obsolete); disc densely and rather coarsely punctate with larger punctures irregularly and sparsely placed; without distinct medial line, but foveate depression generally indicating its position near base. Scutellum transversely punctate. Elytra wider than prothorax at base, humeri rather squarely rounded; seriate- punctate (the seriate punctures smaller and more distant than in S. parallela Germ.), the 5th, 7th and 9th intervals having rows of elongate, shining nodules, other intervals containing minute round nodules, all intervals impunctate, suture flat throughout. Episterna and epipleurae coarsely punctate, rest of underside very finely so. Dimensions: 10-12 X 4.5-5 mm. Habitat—New South Wales: Barrington Tops. Var.—Three examples of the nine before me (of which one is from Eccleston, Allyn River) have the femora concolorous with the general surface; but are without any other distinction. I can find no sexual distinction in the structure. Six examples of the typical form and two of the variety were taken by the University expedition, generally under dead Eucalyptus boughs lying on the ground. The species is exceptional in having the crenulate sides of prothorax of my Group i, with the punctured disc with larger punctures interspersed of Group ii. It is more naturally allied to geniculata Haag. and femoralis Macl., from which it can be distinguished by the pronotal sculpture alone. Type in Macleay Museum. SEIROTRANA CARBO, N. Sp. Elongate-ovate, depressed, dull black above, nitid black beneath, tarsi and apical segments of antennae piceous. Head coarsely rugose-punctate on forehead, clypeal area punctate only; antennae rather stout, 3rd segment about 1% times as long as 4th; 8th-11th oval and successively widening, 11th longer than 10th. Prothorar 3 X 4 mm., circularly emarginate at apex, anterior angles acute, base subtruncate (with a wide, shallow BY H. J. CARTER. 241 “hay” in the middle), sides lightly rounded, widest at middle, extreme border subcrenulate, with some feeble irregularities on posterior half; a wide shallow concavity preceding the obtuse hind angles; disc closely and finely longitudinally strigose-punctate; medial channel well defined throughout in one example (2); less defined in the second (? 2). Scutellwm transverse, oval. Hlytra wider than the prothorax at base, humeral angles a little rounded but defined. Seriate- punctate, with rows of large round punctures, intervals finely punctate, the 3rd, 5th, 7th and 9th having rows of elongate, shining studs, becoming sharply raised towards the sides; the alternate intervals with small, feebly indicated nodules; the suture itself irregularly raised. Episterna and epipleurae with large, sparse punctures, the first three segments of abdomen finely but deeply, longitudinally strigose. Dimensions: 12 X 5 mm. Habitat.—Queensland: Byfield (H. J. Carter). Two examples (I think the sexes) were taken under dead Hucalyptus boughs. The species belongs to my first group, having the pronotum laterally crenulate, disc regularly vermiculate-punctate, without larger punctures interspersed. In this group it would come near catenulata by colour; but is readily separated from that species by the smaller size, the pronotum with less widened sides and much more finely sculptured disc inter multa alia. Type in Coll. Carter. Since tabulating the genus Seirotrana in 1908, new species have been added, while other modifications of the table are desirable, though I find it difficult to separate some of the species in the few words convenient for such tabulation. The following table may now replace the former one. S. mastersi Pasc. and S. monticola Blackb. are now transferred to Group ii. With more material I find that these can scarcely be said to have crenulate margins to the pronotum, though a slight irregularity can be seen in some examples, while both have the dual system of puncturation noted for Group ii. S. orphana Pasc., as with S. repanda Pasc., was described as an Adelium but clearly belongs to this genus. S. foliata Carter = Adelodemus squalidus Macl. and must be placed as a synonym. I was deceived by the small size and abraded clothing of the unique example, which exposed the elytral sculpture to an unusual extent, while the post tarsi were wanting. Table of Seirotrana. GROUP i. Sides of pronotum crenulate. 1. Dise of pronotum vermiculate or strigose-punctate without larger punctures inter- STONSOG! Teer eareah in icdeneyce s= Saatseseecd ica Btawal caratiob ney uiceey eee ce Sey on bre ro ray taV oe ha gre pediicabs cone epee raps 2 Disc of pronotum densely punctate with larger punctures interspersed anomala Cart. IDISCMOL PLONOLUIM NOAUIOSE). cxyor oni enas osek= eisie aes cele sin cu stensuennay eer e eva © crenicollis Pasc. var. denticollis Cart. Dise of pronotum with wide longitudinal sulcus ............ nosodermoides Pasc. 2. Elytra with all intervals costate and crenulate ................ strigipennis Bates Elytra with alternate intervals catenulate ............. 2.2 see e cee e eee eee eee 3 DLC MONA VCllO Wr tails DOK o cians ates 5) 5 6 esis ue.i67-a) sc di0" ey eapewaayomaiietlatare el Gha) sietyene) « louesgees i 3 lepeust suensuens ase 4 LETION AAU TI COLOTOUSM rete Stee se ee seer ee oe eee Cec eettap amen aot ok oes ore tas ie! gre onerered aescerseerieuee cee 5 AC olour. black*(On) Nearly: NSO. celoats stale oles Oil cuebte | I oe see geniculata Haag. Colour bright copper, size much smaller .....................00- femoralis Macl. 5. Size large (more than 15 mm. long) ...............2ceeeceececees major Blackb. Sizersmalleri@essathanwlbpmim=aslOneyl casera ecient ie ie oun eichoei clei: 6 B COlOwP MAO coocgucoapccsxcpacKbdanKdo DOC DOODO DS OOOOH DO ODDHOONO OOH USO ODN OS 7 WOTOUR PROMZE® Far ereee ths ccesiertoke ear erenare aoe. hive orlej loners ae oan en aos epee ie BSUS, ooetioue tee sioun cue lewe 8 7. Pronotum coarsely rugose-punctate, sides strongly crenulate .... catenulata Boisd. Pronotum finely strigose-punctate, sides feebly crenulate ............ carbo, Nn. sp. 242 AUSTRALIAN COLEOPTERA, 8. Non-catenulate intervals of elytra bearing small granules .....................- 9 Non-catenulate intervals without such granules ...................- vicina Cart. 9. Elytral intervals, at least on basal half, with transverse septa .. swb-cancellata Cart. Elytral intervals without transverse septa ..................-----++--+-se-++ eee 10 10. Colour dark bronze, suture not, or scarcely nodulose ................... POSS APES ilil Colour coppery-bronze, suture strongly nodulose ................ vertebralis Cart. Nil, Selene popsncwwyeas Ge Ghyses, Wey soos docodonoaccuscv nc uDd odd DONDOO proxima Pasc. SSeS DONORS OF Ghee, GoeVMle oosoccacodcunvdoconan do MNDDDDNS minor Cart. Group ii. Sides of pronotum entire; disc generally finely punctate with larger punctures interspersed. 1. Elytra having alternate intervals catenulate .............. sce eee e eter tenes 2 Elytra without raised, nitid, elongate nodules ..............-.eeeeeeeeeeeeeee 10 % Dkr mochwiles Comsoeuenghy ma ooococcc0scno gc KK KD OHO OC ON ODDO DNDS OO OOTS 3 IDIGAITCAU! TROCUNIGS TWAOECINEOWOEUOLHIE) scondcadcccosns000 bn aeons OOo OOD DDOOUDDDO EN NOO GOGO 6 2 Size iewse Gor Wann iG) sao, Wee) soaoncconc5cnoduen nous UoGK0KS mastersi Pasc. Sima goallier Ces einen WH sobs, WOME) scacasnocncdcno bond D DOO Ob EMOGDOOOUONGARD 4 A, Tokamall iaoelwules losesianchboeyl 5s onccaccoodn deco snoDC en ocoD Ud SOnGeDUeSOOOCOOOOS 5 Elytral nodules irregular and oblique ...............0s0eeeeeeeee nodicauda Cart. 5. Narrow and parallel, seriate punctures large .................. parallela Germ. OVO, GEREN INOS Genel soacdacccandconnnn eo dpooUds000NDE bimetallica Cart. 6. Colour bronze, nodules not confined to apical area .............2... ese eeeeeee U Colour black, nodules only apparent near apex ................... simplex Blackb. 7. BElongate sub-parallel, seriate punctures large ..................02+e+-eseeeeeees 8 Ovatersseriates punctuness small eee cr ciclenenelsieleinaicy ey cn Aedeaienen nee ai rer eenea nena 9 8. Ground punctures of pronotum very fine ....................2.- elongata Hrichs. Ground punctures of pronotum not very fine .................. integricollis Haag. GQ), Idtoya00 Moos Ayal yA, Colloyhe GoyaeyAy Gonccaccogcdg00uG0d5UGuS osOoUd repanda Pasc. Form more elongate, colour dark bronze ................-.+0+- monticola Blackb. iQue intenvalsrotvelytratlat eo ramisislecuccstsiensichehovel-Nelensnc] or cpen ef oli-liene ee) ile ken iY a ROIS Re ROMS 11 IDAATANUS Oe GUIYATEE, COMES coacrcbasccodoocoddH doo coo DO DODO ODOU COO O DDH DOboOOUOOS 13 iil, Size leiese. Gaines rein I wari, IOs) soscdocnaoccocadccocncoucude punctifera Macl. SizeismalliGuo-12 tina. VOM) POs Sistewers kectst okcostocatene. arene) eee heitey ates ose aeulaes otal ence a nite ere 12 12. Blytral intervals with minute pustules ......................-.«.. simsoni Cart. ID lytraleintenval Shim pus tulla ce meer cteretleiel i ekalinii iil) ieee ier ie aerate orphana Pasc. 13. Colour bronze, alternate intervals interrupted by punctures ........ tumulosa Cart. Coloursblack-velytraleintenvals uniform. seein nies uniformis Cart. BRYCOPIA QUADRATICOLLIS, N. Sp. Elongate, narrowly ovate, bright bronze, glabrous and very nitid; antennae and tarsi pale red, legs darker red. Head very minutely punctate, eyes prominent, antennae moniliform, except the subcylindric 3rd segment, 8th-11th successively wider, 11th oval, much larger than 10th. Prothorax nearly square (2 mm. long), slightly wider than long, apex lightly arcuate, anterior angles feebly advanced and obtusely rounded; sides subparallel for the greater part, feebly narrowed at apex; base truncate, posterior angles rectangular, without foliate margins, a very narrow horizontal . border visible; disc minutely and sparsely punctate, without medial line or evident foveae. Scutellum equiangular triangular. Hlytra of same width as prothorax at base, shoulders obsolete, sides subparallel; striate-punctate, the punctures in striae round and regular, closer and larger near suture, smaller and more distant towards sides; intervals flat with some minute sparse punctures. Underside minutely and sparsely punctate.: Dimensions: 6 X 2 (+) mm. Habitat—New South Wales: Barrington Tops (H. J. Carter). I took a single example in rotten Eucalyptus wood at an altitude of 5,000 ft. It is very distinct in its cylindric form and nitid surface from any of its congeners. BY H. J. CARTER. 243 In my tabulation (Proc. Linn. Soc. N.S.W., 1920, p. 247) it would come next to minor Cart., thus:— 30a. Sides of prothorax subparallel ...................20008- B. quadraticollis, n. sp. Type in Macleay Museum. Cistelidae. CHROMOMOEA ELEANORA, DN. SDP. Elongate, subparallel, metallic-black above, nitid black beneath, oral organs, tibiae and tarsi (sometimes femora also) testaceous (becoming red in older specimens), three apical joints of antennae, also tarsal lamellae, infuscate. Head densely striolate-punctate; antennae lineate, 3-10 successively shorter and slightly wider, 11 shorter and narrower than 10. Prothorax of same width as head, about as long as wide, base and apex truncate, sides slightly obliquely widened from apex to base, and a little rounded at apex, disc closely and finely punctate, a medial line indicated near base, bisecting a large basal depression, a transverse sulcus parallel and close to base. Scutellum transversely oval. Elytra wider than prothorax at base and three and a half times as long, shoulders squarely rounded, sides nearly parallel for the greater part; striate-punctate, the punctures largely hidden in the deep sulci, intervals widely convex, densely trans- versely wrinkled or strigose-punctate, a sparse tomentum visible on apical area. Underside finely punctate and glabrous; legs unarmed in both sexes. Dimensions: 10-11 x 3 mm. Habitat—New South Wales: Barrington Tops. Four examples examined, one taken by myself, December, 1915, the others taken by the University Expedition, January, 1925. Except ©. violacea Cart., from the same district and the following species this is the largest of the genus, and differs from C. rufescens Bates in colour and sculpture, the latter being much finer in Bates’s species (of which I possess a cotype). From violacea it differs in size, colour and in the shorter and wider prothorax, which is unusually wide for the genus, but I am unwilling to erect a new genus on this character alone, since in other respects it is a typical Chromomoea. I name it in honour of Miss Eleanor Chase, lecturer in Zoology in the University of Sydney, an active collector of the expedition. Type in the Macleay Museum. CHROMOMOEA OCULATA, DN. Sp. Elongate; parallel, red or castaneous, underside and appendages testaceous, mandibles infuscate. Head and pronotum very finely and sparsely punctate, eyes unusually large and prominent, invading the frontal area, the interspace between eyes less than the transverse diameter of one eye, as seen from above; antennae narrowly lineate, segments less enlarged than in rufescens Bates, violacea or eleanora, in length intermediate between the shorter segments of rufescens and the longer of eleanora, the 11th equal in length to the 10th. Prothorax subcylindric, slightly longer than wide, truncate at apex and base, anterior angles rounded, posterior sharply rectangular, medial channel shown for the greater part, but not continued to apex and lying in two longitudinal depressions, one basal, the other pre-apical; a transverse fovea near posterior angles. Scutellum triangular. EHlytra consider- ably wider than prothorax, humeri prominent and squarely rounded, sides parallel for the greater part, apices separately and sharply rounded; striate- G 244 AUSTRALIAN COLEOPTERA. punctate, punctures in striae very close; intervals equal and regularly convex, sharply so on apical third, finely and sparsely punctate, not wrinkled. Prosternur punctate, elsewhere laevigate and nitid. Legs unarmed in both sexes. Dimensions: 10-12 X 24-3 (+) mm. Habitat.—South Queensland: Tambourine Mountain (H. J. Carter). I took four examples, including the sexes, by beating foliage in November, 1924. One of these, apparently more mature than the others, is a darker red in colour. It is distinguished from C. rufescens Bates by its much larger eyes as well as by its much finer and sparser surface punctures, the elytra without transverse striolation. Type in Coll. Carter. GASTEROMYCETES OF AUSTRALASIA. ii. A REVISION OF THE GENUS TULOSTOMA. By G. H. CunnNINGHAM, Mycologist, Dept. of Agriculture, Wellington, N.Z. (Plates xxxiii-xxxv. ) [Read 29th July, 1925.] This paper is the second of a series dealing with the genera and species of the Australian and New Zealand Gasteromycetes. It is the intention of the writer to deal with all genera and species of this sub-class, revising the descriptions and presenting photographs when available, together with micro-photographs of salient microscopic characters. In addition it is intended to work out the develop- ment of one or more species of each genus (when material is available), and finally to rearrange the whole according to a modern system of classification. When completed, all these papers will be issued in book form. The author will gladly receive material for description and determination, especially any showing develop- mental stages of different species and genera of Gasteromycetes. Dr. J. B. Cleland, The University, Adelaide, has very generously lent the whole of his extensive collections, comprising more material than exists in the whole of the other herbaria of Australasia. Members of the genus Julostoma are found in practically every country throughout the world, but are most abundant in hot and sandy regions. They are characterized by the stipitate, two-layered peridium, which dehisces by an apical orifice, the long, frequently branched, septate capillitium and the coloured, continuous spores. They are usually found with the peridium alone projecting above the surface of the sand or other substratum in which they grow. This makes them difficult to observe, as owing to their inconspicuous colour they are readily overlooked. Immature plants are buried 5-6 cm. or more beneath the surface, and may be obtained only by careful examination of the soil in the vicinity of mature plants. Structure of the Mature Plant. Peridium.—This consists of two layers, an exoperidium and an endo- peridium. The exoperidium is usually fugacious, soon falling away, save at the base, where it persists as a thick toughened membrane enclosing the endo- peridium in a cupulate structure. The former is perforated by the stipe, to which it may occasionally be attached, but usually it is free. In structure the exoperidium consists of somewhat loosely woven hyphae in which are embedded numerous particles of sand, earth or vegetable matter, according to the sub- stratum in which the plant may happen to be growing. Occasionally it does not fall away, but persists as an investment surrounding the endoperidium; this was the principal character upon which was erected the species 7’. adhaerens Lloyd. In other species it is partly adherent, then giving a shaggy appearance 246 GASTEROMYCETES OF AUSTRALASIA, li, to the specimen by reason of the irregular appearance of the adhering particles, or it may consist of hyphae woven into little projections, which persist and give to the whole plant a verrucose appearance. The endoperidium is in the nature of a thin, tough, papyraceous membrane enclosing the gleba; it is apically perforated by the so-called mouth, and basally attached to the stipe but separable from it. It consists of an outer intricately woven layer of hyphae of a diameter about half that of the capillitial threads. These hyphae are freely septate, hyaline, and not swollen at the septa. They may be somewhat even, when the endoperidium appears smooth, or may extend for some little distance as free hyphae, giving to the whole peridium a strongly sulcate or fibrillose appearance. Enclosed within this is a second thin layer of. pseudoparenchymatous tissue, the cells of which are more or less isodiametric, and it is this layer which gives the endoperidium its toughness. Inside this second layer is a third arising directly from it, and from the free ends of the hyphae of which it is composed arise the threads of the capillitium. The gleba consists of copious capillitial threads immixed with the abundant spores. The capillitium consists of very numerous long, branched, scantily septate threads. They are usually colourless, but may appear tinted, especially at the septa. The septa may be swollen to about twice the normal thickness, or not, and swollen and non-swollen septa are not infrequent in the same specimen. The ultimate branchlets are thinner than the main branches, and are rounded at their free ends. The spores are usually globose or subglobose, and average about 5 w in diameter. They are all some shade of brown, and the epispore may be rough or smooth. The spores are borne on clavate or cylindrical basidia, but instead of being produced apically, as is usual in the Gasteromycetes, they are borne laterally on short sterigmata. Frequently the sterigmata are detached with the spores and remain attached to them, when the spores are said to be “pedicellate’’ or “apiculate.” It is extremely difficult to obtain basidia in other than very immature plants, for spore production is well advanced ere the plants appear on the surface. Mouth.—This differs in certain species so that it serves as a good group character. Three distinct types may be recognized :— (1) A circular, definite, erumpent-tubular mouth, characteristic of the type of the genus, 7. brumale. Sometimes in this group the mouth is not circular, but elliptical and almost plane; this condition may arise as a result of weather- ing, for it is most frequently met with in old and weathered specimens. (2) A definite, circular, umbonate mouth surrounded by a layer of hyphae arranged radially around it. This type is said to be fibrillose. In old specimens the fibrils frequently disappear, when the mouth is seen to resemble closely the first type. (3) An indefinite plane or erumpent aperture. The orifice in this group appears simply as a rupture in the apex of the endoperidium. (4) Other types of mouths are known. For example a species found in Brazil, T. exasperatum, has a raised mouth so resembling those termed fimbriate in the genus Geaster, etc., as to have this name applied to it. Finally Lloyd (1906) mentions a group which may have several mouths in the same endoperidium. The writer has not seen any specimens he would consider to belong to this group. Stipe.—This varies greatly in length and breadth, even in specimens of the same collection. It consists of an outer cortex, which is usually deciduous, and BY G. H. CUNNINGHAM. 247 an inner: stipe proper, composed of pseudoparenchymatous tissue. The stipe is hollow and is invariably filled with a fibrous tissue, hence it is said to be “stuffed”. The cortex usually falls away more or less completely, but may persist in the form of coarse scales, as in 7’. australianum. When it falls completely away the stipe is usually smooth, in which condition it often exhibits longitudinal striae. The colour varies according to age, but is by no means constant, even in the same collection. At the base is usually present a bulbous structure which may be hard and woody like the stem, or may consist solely of loosely interwoven mycelial tissue. This structure varies greatly in size and shape, and so is of slight importance to the systematist. Systematic Position of the Genus. In the most recent work (Fischer, 1900) the genus is placed in the family Tulostomataceae of the order Plectobasidiineae, containing the four genera Tulostoma, Queletia, Battarrea and Sphaericeps. Discussion on its true phylo- genetic position will be deferred until such time as other related genera have been dealt with, and until certain details of development have been worked out. Owing to the difficulty of procuring developmental stages, the life-cycle and development of any member of the genus has yet to be worked out. Bessey (1887) mentions the fact that the gleba matures underground, elongation of the stipe commencing only after this has been effected. Schroeter (1887) has studied the development of spores and basidia in T. brumale, and was the first to point out that the spores were’ borne laterally on a basidium which, in this particular, resembles that of certain of the Uredinales, save that it is continuous, and not septate. No other work appears to heave been undertaken with reference to development of members of the genus. Specific Characters. Some sixty species have been described; of these twenty have been recorded from Australia, but it is doubtful whether many of these are species, or merely forms of already described species. Considerable confusion exists in the delimitation of species, and by certain workers the genus is considered to be the most difficult of any included in the Gasteromycetes. The writer believes this confusion to have arisen through the authors of various species fixing upon variable characters, such as shape, size and colour of the peridium, length and colour of the stipe, and degree of septation of the capillitium. He has recently made a critical examination of some 200 plants of the species 7. poculatum (the only species found in abundance in New Zealand), and has found that the peridium varies in size from 2 to 12 mm.; it may be perfectly smooth, or more or less covered with the adhering exoperidium; or may be pitted as in the so-called species 7. punctatun. Peck; it may be dingy white or dark brown in colour; and this colour may be spread over the whole peridium, or confined to a portion only. Each of these variations, supposing extremes only were examined, is sufficient to erect a “species” upon. That numerous species have been erected on such flimsy characters is very evident when the literature dealing with the genus is consulted. Owing to this variability it would be unwise to continue to erect species upon colour, roughness or otherwise of the peridium, or even upon size (unless very marked), for this too is by no means constant. 248 GASTEROMYCETES OF AUSTRALASIA, ii, The difficulty is to decide what are fixed characters in such a variable genus; about half of the species which have been described to date cannot be separated on any recognizable character. Recently the writer has worked over various collections of this genus collected in Australia, Tasmania and New Zealand, endeavouring to arrange and describe the collections in such a manner that they would be readily recognized. When arranged according to the older ideas of classification, hopeless chaos resulted; mouth characters were sufficient to split the whole of the collections into three or four groups; drawings and photomicrographs of the capillitium of some ten species were compared, but no one species showed characters not present in others to a greater or less extent. Finally all collections were treated as unknowns, and certain spore characters alone considered, when it was found possible to separate all into definite groups, as is shown in the artificial key. It became, then, a relatively simple matter to subdivide these groups by mouth characters, colour of the peridium, ete. Spore markings are the most constant of all characters present in the genus, being the least variable; they may readily be defined, and may equaiiy readily be illustrated. The writer finds that possession of a good equipment is a necessity if reliable results are to be obtained. Personally he uses apochromatic objectives, and ‘does not hesitate to use the oil immersion when in doubt as to the markings on any spores. When apochromats are used good results cannot be expecsed unless a good condenser is used—the Abbé pattern is worthless for critical work. In addition, attention must be paid to proper correction for cover slip thickness and refractive index of the mountant, for these objectives are hypersensitive in this direction. For temporary mounts the most suitable medium the writer has used is lactic acid solution, 50% acid in water. Spores are placed in this and heated to nearly boiling point of the solution, when all markings are readily seen, and pseudo-markings (such as folds due to shrinkage) disappear. For critical obser- vations under the oil immersion, canada balsam mounts are desirable. * TULOSTOMA Persoon. Syn. Fungi, 1801, p. 1389.—Tylostoma de Toni, Sacc. Syll. Fung., vii, 1888, p. 60. Peridium stipitate, globose to depressed-globose; consisting of an outer thin, usually fugacious exoperidium, and a thin, membranous, coloured or colourless, smooth or rough endoperidium; dehiscing by an apical orifice or mouth, which may be definite or indefinite, naked or fibrillose, erumpent-tubular, umbonate or plane. Stipe inserted in a “socket” at base of endoperidium to which it is attached; woody, smooth or scaly, striate or not, concolorous throughout, stuffed; usually with a mycelial bulb at base. Gleba of capillitium and spores, without sterile veins. Capillitium copious, consisting of numerous tinted or hyaline, very long, usually branched threads attached to the endoperidium, septate or not, septa inflated or not. Spores coloured, globose or subglobose, seldom angular, smooth or variously roughened. j Habitat.—Solitary, gregarious or caespitose on ground or more rarely on decaying wood. Distribution.—World-wide; most plentiful in warm and sandy regions. * Tulostoma: This was the original spelling used by Persoon, and later by Fries. In Sylloge Fungorum, it was by de Toni arbitrarily changed to Tylostoma. This is contrary to the International Rules of Nomenclature, so the original spelling of the word must be retained. BY G. H. CUNNINGHAM. 249 Species may be divided into groups according to the mouth characters, but the writer has found it more convenient to make the initial division on spore characters, separating on mouth and peridial characters later. The spores may be divided into three groups, as given in the key; these divisions are relative, but once seen the differences are readily recognized (see photomicrographs of spores). All descriptions are original, save where otherwise stated, and have been drawn up from material examined by the writer. All photographs are original and have been taken by my friend Mr. H. Drake, of this laboratory. Their uniform excellence assists greatly in the illumination of characters imperfectly described in the specific descriptions. The writer is indebted to Dr. J. B. Cleland, The University, Adelaide, for the loan of all collections of this genus in his herbarium; to Mr. L. Rodway, Government Botanist, Hobart, and Mr. C. C. Brittlebank, Plant Pathologist, Dept. of Agriculture, Melbourne, for the loan of specimens from their respective herbaria. Finally he would like to record his appreciation of the work under- taken by Mr. C. G. Lloyd, Cincinnati, Ohio. Mr. Lloyd has examined practically every historical specimen in the herbaria of Europe and America, and by the aid of critical notes and copious illustrations, published in his Mycological Notes, has succeeded in making clear the characters upon which most of the older species have been based. Artificial Key to Species. SUOGRES JOSITSCWA KTAOOUWN sossacsssoccccogcosuodovoDUDUe UGE UOOE 8. T. poculatum. Spores finely but distinctly verruculose. Mouth definite, tubular, entire. Peraiyi Sao, Oe MORE SOQ scoonccuacdcoa0oc00dKn 2. TT. albicans. TESFPGLILACN {OMIOGNOSIME 6 bn og oldies Guw.0-Simte oS blo-eo-od d Ulowible.d ulate o 1. T. pubescens. Mouth, definite, fibrilloses. ii. ..<- sande tA ote tere ord axcien as 9. T. minutum. Spores strongly verrucose or verrucose-echinulate. Mouth definite, tubular, entire. Peridium uncoloured (dingy white or pallid tan). Peridium smooth. SOs weer WO Mh sooscocccscaccccacu0gcooNES 3. T. McAlpinianum. Soares Over ID fh occocoacecnccoxn svg uvvo0ecace 4. T. macrosporum. Peridium rough with adhering exoperidium ........ 5. TT. adhaerens. Peridium coloured. PENI CGEM ORO sococoebovndcendccocboaoee 6. T. brumale. Peridium chocolate-brown ..........-............ 7. T. Purpusii. Mouth definite, fibrillose. Spores closely and finely verrucose ................... 10. T. subfuscum. SOOKOS SUAS cncacoosccovocsovnapooeunouspocaooDoADY 11. T. striatum. Wikartey Thavaksovoutes: ts 6 oben clears Ganieore ors odio b ia-6 crotera cot nediowe Slocd orate 12. T. australianum. Srorion I: Mouth definite; circular or elliptical, tubular or plane. a. Spores finely and distinctly verruculose. 1. TULOSTOMA PUBESCENS, n.sp. Plate xxxiii, fig. 1. Peridium depressed-globose, up to 10 mm. high, 20 mm. diam.; exoperidium persistent, dingy brown, almost black, in the nature of coarse mycelial fibres mixed with vegetable debris; endoperidium ferruginous, pubescent with closely appressed silky threads. Mouth definite, 2.5-3 mm. diam., circular, plane. Stipe up to 4 cm. X 6 mm., equal, densely pubescent, colour of the peridium, stuffed, rugulose. Gleba reddish-brown; capillitium hyaline, threads sparingly septate, slightly thicker than the spores, branched, septa plane. Spores globose or subglobose, 250 GASTEROMYCETES OF AUSTRALASIA, ii, 4-5.5 uw, apiculate; epispore finely and delicately verruculose, pallid ferruginous, 0.75 « thick. Habitat.—Solitary on the ground. Distribution.—South Australia: Pt. Gawler (Apr., 1923, J. B. Cleland). Type in Herb. Cleland. It is with some hesitation that the writer has erected this species, for it is based on a single specimen. Nevertheless, the plant is so characteristic that it may readily be recognized. The plane definite mouth, adhering fibrous exoperidium, pubescent endoperidium and stipe are very distinctive, for certain of these characters have been met with only in this specimen. The plant is some- what damaged, hence the shape of the peridium, as shown in the photograph, is not characteristic. 2. TULOSTOMA ALBICANS White. Plate xxxiii, fig. 2; Pl. xxxv, fig. 19. Bull. Torrey Bot. Club, vol. 28, 1901, p. 428. Peridium depressed-globose, up to 10 mm. high, 12 mm. diam.; exoperidium soon falling away from the upper portion but persisting at the base of the endoperidium, which is smooth, thin, papyraceous, dingy-white or pallid-tan. Mouth small, 1 mm. diam., circular, short-tubular, entire. Stipe 2-4 cm. X 3-6 mm., equal, bay-brown, finely striate, fibrillose, stuffed, with a small mycelial pad at the base. Gleba reddish-brown; capillitium hyaline or tinted, threads branched, sparsely septate, septa moderately swollen. Spores globose or subglobose, 4-6 » diam., frequently apiculate; epispore pallid ferruginous, finely and moderately verru- culose, 1 yu thick. Habitat.—Solitary or gregarious on the ground. Distribution.—North America; Australia. New South Wales: Dubbo (Oct., 1913. +J.B.C.), *Forbes (Aug., 1915. J.B.C.), Manildra (Oct., 1916. J.B.C.), *Near Barellan (Aug., 1918. J.B.C.), Narrabri (May, UGG), dlB(Cs)). South Australia: *Reynella (July, 1914. J.B.C.; Id., Lloyd, No. 80, as T. McAlpinianum), **Monarto South (J.B.C.), ***Beaumont, Adelaide (June, 1917. J.B.C.), between sand dunes and Milton Railway Station (Jan., 1920. Prof. Howchin), Berri (Jan., 1921. J.B.C.; Id., Lloyd, No. 750, as 7. simulans), Beau- mont (July, 1921. J.B.C.), 10 miles west Overland Corner (Feb., 1922. J.B.C.), Big Swamp, 12 miles west Port Lincoln (Feb., 1922. J.B.C.), Ooldea (Aug., 1922. J.B.C.), Flinders Range, Pt. Augusta (Aug., 1922. J.B.C.), Mr. Zeitz Coll. Victoria: Melbourne (June, 1885. F. M. Reader), near Dimboola (July, 1890. F. M. Reader). Tasmania: Domain, Hobart (Mar., 1920. L. Rodway). Var. NIGROSTIUM, 0. var. Peridiue similar to T. albicans, but differing in that the mouth is strongly coloured, usually brown, frequently lead colour. Other characters as above. Distribution—New South Wales: Bumberry (Sept., 1916. J.B.C.), The Rock (July, 1917. J.B.C.; Id., Lloyd, No. 382, as T. albicans). * These forms possess spores more strongly marked than the usual. ** This is a larger and more globose form. *** Doubtful determination as the mouth is gone. 7 J.B.C. = J. B. Cleland. BY G. H. CUNNINGHAM. 251 T. albicans is characterized by the pallid peridium, definite tubular mouth, and especially by the finely verruculose spores. Lloyd has placed one of these collections under T. McAlpinianum, another under 7. simulans Lloyd, and a third under 7. australianum. It is difficult to separate 7. McAlpinianum from this species, other than by the spore markings (see note under this species). In his paper Lloyd (1906) considers 7. simulans to be a form of T. mammosum (T. brumale). The characters of this latter are the coloured peridium, darker mouth, freely septate capillitium and strongly verrucose spores. His TJ. brevipes he claims to be an uncoloured form, possessing all other characters; but his 7. simulans is claimed to have a differently coloured peridium and different capillitium. If so, then it has no relationships with T. brumale whatsoever. The specimen so named in the collection from Berri, South Australia, is typically T. albicans, and does not differ in any particular, so far as the writer is able to judge. T. albicans is the most abundant species in Australasia, judging from the collections that have passed through the ,Writer’s hands. b. Spores strongly verrucose or verrucose-echinulate. 3. TuLostoma McALPINIANUM Lloyd. Plate xxxiii, figs. 3, 4. Tylostomeae, 1906, p. 15. (Hmended). Peridium depressed-globose or globose, up to 12 mm. high, 15 mm. diam.; exoperidium soon falling away from the upper portion, but remaining at the base of the endoperidium as a thickened, closely adherent disc; endoperidium smooth, partly covered with adhering particles of the exoperidium, or slightly pitted, papyraceous, dingy-white or pallid-tan. Mouth small, 1-1. mm. diam., circular or elliptical, short-tubular, entire. Stipe 2-8 cm. X 3-5 mm., equal, slightly or not thickened basally, coloured bay- or chestnut-brown, fibrillose, striate, woody, stuffed. Gleba reddish-brown; capillitium hyaline or tinted, threads sparingly branched, sparsely septate, septa slightly swollen. Spores globose or subglobose, 5-8 u diam.; epispore pallid ferruginous, coarsely, bluntly and sparsely verrucose, 1 » thick. Habitat.—Solitary or gregarious on the ground. Distribution—New South Wales: Narrabri (June, 1918. J.B.C.). South Australia: River Murray, near Morgan (Nov., 1913. J.B.C.), Adelaide (July, 1920. J.B.C.; Id., Licyd, No. 710, as above), Kinchina (July, 1922. J.B.C.). Victoria: Canterbury (March, 1909. GC. French, Jr.), Melton (Sept., 1899. C. French, Jr.). The writer has found it impossible to separate this species from the preceding, save by the spore markings. It is evident that Lloyd himself is not clear as to the characters of this species, for he states (1906): “While it is very difficult to draw up a diagnosis of the differences between this and TJ. albicans, the plants are not the same and our photograph will show it better than our description can. The plant in general appearance very much resembles 7. mammosum, but is uncoloured and has almost smooth spores. J. albicans belongs to a different type of plants in shape and cortex from 7. mammosum.’ Further evidence as to the difficulty Lloyd has had in placing his species, is shown from his determinations of some of Dr. Cleland’s specimens. For example, one collection (No. 80) which he has determined as being 7. McAlpinianum, the writer places under TJ. albicans, for the spores are quite different from those of the collection (No. 710) Lloyd placed under 7. McAlpinianum. A third collection (No. 79) Lloyd claimed to be this species has spores which are coarsely echinulate and fully twice as large. 252 GASTEROMYCETES OF AUSTRALASIA, ii, In his original description of 7. McAlpinianum, Lloyd stated that the spores were almost smooth, yet three collections named by him as this species have verrucose spores. Such confusion can occur only when species are erected on such variable characters as colour, shape and markings of the peridium. As three collections with verrucose spores have been determined by Lloyd as being T. McAlpinianum, the writer has arbitrarily fixed upon this as being the only character of specific value upon which it may be separated from TJ. albicans, for it is obvious that until some such stand is taken no means exists of determining this species accurately. See also remarks under 7’. albicans. 4. TULOSTOMA MACROSPORUM, nh. sp. Plate xxxiii, fig. 5; Pl. xxxv, fig. 20. Peridium depressed-globose, up to 8 mm. high, 12 mm. diam.; exoperidium completely falling away from the endoperidium, leaving this smooth save at the base; endoperidium white, smooth, thin, papyraceous. Mouth entire, short-tubular. circular, 1 mm. diam. Stipe 2-3 em. X 1.5-2 mm., equal, gmooth, sparsely striate, woody, stuffed, slightly dilated at the base. Gleba pallid-violaceous; capillitium tinted or hyaline, threads _ sparsely branched, septa slightly swollen. Spores globose or subglobose, 9-13 mu diam.; epispore ferruginous, densely and coarsely echinulate, spines about 1 w long. epispore 1.5-2 uw thick. Habitat.—Caespitose on the ground. Distribution.—New South Wales: Dubbo (July, 1915. J.B.C.). Type in Herb. Clel. (Jd., Lloyd, No. 79, as T. McAlpinianum). The caespitose habit, violaceous gleba, and especially the large and coarsely echinulate spores, characterize this species. If 7. McAlpinianum is to be main- tained as a species—separable only upon spore markings—then the above also must be considered a distinct and clearly defined species. 5. TULOSTOMA ADHAERENS Lloyd. Plate xxxili, fig. 6; Pl. xxxiv, fig. 7. Mycological Notes, vol. 7, 1923, p. 1199. Peridium depressed-globose, up to 10 mm. high, 15 mm. diam.; exoperidium a firm, closely adhering sand case, bay-brown, rough; endoperidium dark-brown, tough, membranous. Mouth definite, elliptical, about 3 X 1 mm., entire, not or slightly protruding. Stipe 2-3 ecm. X 3-5 mm., fibrillose, ochraceous or bay-brown, woody, stuffed, equal, geniculate, with a slight mycelial bulb at base. * Gleba ochraceous or pallid-ferruginous; capillitium subhyaline, threads branched, slightly swollen at the septa. Spores globose or subglobose, 5-6 u diam.; epispore pallid-ferruginous, closely and finely verrucose, 0.75 uw thick. Habitat—On the ground. Distribution.—New South Wales: Narrabeen (July, 1916. J.B.C.). Type Coll. in Herb. Clel. (Lloyd, No. 794). The above description has been drawn up from type collection material kindly donated by Dr. Cleland, and now in my herbarium, No. 2192. Lloyd states that this is the first collection made in Australia possessing this type of mouth. He places it (1906) in Section 6 defined as “Mouth definite, naked, elongated, sometimes several on the same peridium.”’ The elongate plane mouth and rough adhering exoperidium characterize the species. As to whether this last feature is constant, or merely a condition due to immaturity of the plants, it is impossible to determine from the material at hand. “BY G. H. CUNNINGHAM. 253 6. TULOSTOMA BRUMALE Persoon. Plate xxxiv, fig. 8; Pl. xxxv, fig. 21. Syn. Meth. Fungi, 1801, p. 139.—Tulostoma mammosum Fr., Syst. Myc., vol. 3, 1829, p. 42—Tylostoma pedunculatum Schroet., Cohn, Beitr. Biol. Pflan., vol. 2, 1877, p. 65. Peridium globose or slightly depressed-globose, up to 12 mm. diam.; exoperidium usually completely falling away from all but the base of the endo- peridium, occasionally persisting as irregular, scattered patches; endoperidium papyraceous, chestnut- or bay-brown, smooth. Mouth small, 1-15 mm. diam., circular, short-tubular, entire, protruding, darker than the endoperidium. Stipe 2-4 em. XK 2-4 mm., bay- or reddish-brown, equal, smooth or slightly scaly, sparingly striate, stuffed, with a small mycelial bulb at the base. Gleba ferruginous; capillitium hyaline, threads branched, somewhat freely septate, septa moderately swollen; spores globose, 4-5 u diam., frequently apiculate; epispore pallid-ferruginous, moderately and sparsely verrucose, 1 u thick. Habitat.—Solitary on the ground. Distribution—England; Northern and Central HEurope; North America; Australia; New Zealand. Victoria: Frankston (July, 1905. D. McAlpine; JId., Lloyd, No. 23, as T. albicans). New Zealand: Sandhills, Levin, Wellington (Dec., 1919. E. H. Atkinson). Lloyd states (1905) that there is one collection of this species from Australia at Kew, colleeted at White River in 1870, which he believes to be the typical form. On the other hand, he claims that the true form does not occur in North America, and lists (1906) under 7. simile and T. rufum Lloyd, all forms found in that country. The New Zealand and Australian specimens listed above agree in all particulars with recent descriptions, although some forms approach T. Purpusii in the colour of the endoperidium, lending strength to the contention that the latter is better considered as a variety of T. brwmale. One recent worker (Rea, 1922) describes the spores as being “pinkish”. Surely this is an unusual colour for the spores of any species belonging to this genus. If correct, then our form is not the same as the European, as the spores are decidedly ferruginous, and under the microscope, pallid-ferruginous, when viewed with apochromats, which render colour tones correctly. 7. TuLostomMA Purpusit P. Hennings. Plate xxxiv, fig. 9. Hedwigia, vol. 37, 1898, p. 274. Peridium depressed-globose, up to 15 mm. high, 2 cm. diam.; exoperidium imperfectly falling away from the endoperidium, remaining as small, irregular patches upon the upper surface and firmly attached basally; endoperidium dark- chestnut, chocolate or sepia-coloured, thin, firm, membranous, frequently furfuraceous. Mouth small, 1-2 mm. diam., circular or elliptical, short-tubular, entire, frequently darker in colour. Stipe 1.5-2 cm. X 3-5 mm., equal, ochraceous or ferruginous, floccose but not scaly, sparsely striate, stuffed, often slightly inflated basally. Gleba ferruginous; capillitium tinted or hyaline, threads sparingly branched, scantily septate, septa slightly swollen. Spores globose or subglobose, 5-7 » diam.; epispore pallid-ferruginous, moderately and somewhat closely verrucose, 1 uw thick. Habitat.—Solitary or caespitose on the ground. Distribution.—North America; Australia. 254 GASTEROMYCETES OF AUSTRALASIA, ii, New South Wales: Under pine tree, Botanic Gardens, Sydney (June, 1908. EH. Cheel; Jd. by Lloyd as above); specimens in Herb. Cleland. Garden, Neutral Bay, Sydney (Dec., 1919. J.B.C.). The writer is of the opinion that this is best considered as a large, more deeply coloured form of 7. brumale, for it is difficult to separate it from this species other than upon these two characters. Lloyd states (1906): ‘This species can easily be taken for a giant form of 7. mammosum. It is a rare plant and I have seen from America only the type specimens at Berlin, which were collected in Colorado by a Mr. Purpus. We refer here (for the time at least, rather than make a new species) plants from Australia with the same general characters, but which differ in the more persistent cortex and the spores, which in the Australian species vary from 4 to 7 up.” Section II.—Mouth definite, fibrillose. a. Spores perfectly smooth. 8. TULOSTOMA POCULATUM White. Plate xxxiv, figs. 10, 11; Pl. xxxv, fig. 22. Bull. Torrey Bot. Club, vol. 28, 1901, p. 431. Peridium depressed-globose, up to 10 mm. high, 12 mm. diam.; exoperidium breaking away completely save the persistent basal portion; endoperidium fawn- coloured or dingy-white, smooth, papyraceous. Mouth raised, surrounded by a circular fibrillose zone which may attain a diameter of 3 mm. Stipe 2-3 cm. X 3-5 mm., tan-coloured, sulcate, striate, equal, stuffed, slightly bulbous at the base. Gleba ferruginous; capillitium tinted or hyaline, threads sparingly branched, slightly swollen at septa. Spores globose or subglobose, frequently subangular, 4-6 wu diam., apiculate; epispore pallid-ferruginous, perfectly smooth, 0.75 uw thick. Habitat.—Solitary or gregarious on the ground. Distribution.—North America; Australia; New Zealand. New South Wales: Belmore, Sydney (July, 1909. A. A. Hamilton), The Rock (July, 1917. J.B.C.; Id., Lloyd, No. 381, as above), Coolamon (May, 1918. J.B.C.), near Barellan (Aug., 1918. J.B.C.), * Wombeyan Caves (Nov., 1919. J.B.C.). South Australia: Beaumont, Adelaide (May, 1920. J.B.C.), * Kinchina (July, 1922. J.B.C.). New Zealand: Sandhills, Levin, Wellington (Novy., 1919. S. A. Cunningham, E. H. Atkinson, G.H.C.; Id., Lloyd, No. 67, as above). This species is characterized by the pallid, smooth peridium, fibrillose, raised mouth and smooth, apiculate spores. The smooth spores especially are a constant character of all the specimens listed above. b. Spores finely and distinctly verruculose. 9. TULOSTOMA MINUTUM White. Bull. Torrey Bot. Club, vol. 28, 1901, p. 430. Peridium depressed-globose, 0.5-1 em. high, 1-1.2 cm. diam.; exoperidium dingy- brown, imperfectly breaking away from the upper part of the endoperidium, but remaining at base; endoperidium pallid-chestnut-brown, membranous. Mouth slightly raised, fibrillose, small, 2 mm. diam. Stipe 1-2 cm. X 2-4 mm., slender, stuffed, brown, striate, frequently with small mycelial pad at base. * Doubtful determinations; mouths gone, but spores J. poculatwm type. BY G. H. CUNNINGHAM. 255 Gleba ferruginous; capillitium tinted, threads sparingly branched, septa slightly swollen. Spores globose or subglobose, 4-6 u diam., apiculate; epispore ferruginous, minutely and closely verruculose, 1 u thick. Habitat.—Solitary on the ground. Distribution.—North America; Australia. South Australia: Berri (Jan., 1921. J.B.C.), Beaumont, Adelaide (July, 1921. J.B.C.). These specimens are doubtfully referred to 7. minutum White. They are too small to be referred to other species possessing fibrillose mouths, with the exception of 7. poculatum, and differ from this species in the spores being closely and finely verruculose and not smooth. c. Spores verrucose. 10. TULOSTOMA SUBFUSCUM White. Plate xxxiv, fig. 12; Pl. xxxv, fig. 18. Bull. Torrey Bot. Club, vol. 28, 1901, p. 433. Peridium depressed-globose, up to 12 mm. high, 15 mm. diam.; exoperidium dingy-brown, imperfectly breaking away from apical portion, but remaining as a firm membrane at the base of the endoperidium; the latter smooth, varying in colour from bay-brown to dingy-ferruginous, tough, membranous. Mouth raised, minute, surrounded by a scanty fibrillose zone 1-2 mm. diam. Stipe 2-3 cm. X 2-3 mm., fibrillose or scaly, leathery, dingy-brown, striate, equal, stuffed, with small mycelial bulb at base. Gleba ferruginous; capillitium hyaline or tinted, threads branched, sparsely septate, slightly thickened at the septa. Spores globose to subglobose, 4-6 » diam., sometimes apiculate; epispore pallid-ferruginous, finely and moderately verrucose, 0.75 w thick. Habitat—Solitary or gregarious on the ground. Distribution.—North America; Australia. New South Wales: Mummulgum (Dec., 1916. J.B.C.; Id., Lloyd, No. 267, as T. poculatum). This species differs from T. poculatum in the somewhat larger size, darker colour of the peridium and stipe, smaller and less fibrillose mouth and finely verrucose spores. Lloyd (1906) considers it to be a form only, but it is quite a distinct plant. The colour of the peridium varies from dark-brown (almost chocolate) to bay-brown, but as colour has little specific value, this variation is of little moment. 11. TULOSTOMA STRIATUM, n.sp. Plate xxxiv, fig. 13; Pl. xxxv, fig. 17. Peridium depressed-globose, up to 15 mm. high, 20 mm. diam.; exoperidium pallid-tan colour, soon falling away save where persistent at the base, in some specimens persisting as irregular roughened patches; endoperidium pallid-tan or dingy-white, smooth, papyraceous. Mouth raised, irregularly circular, surrounded by a fibrillose zone, up to 3 mm. diam. Stipe 2-6 cm. X 2-4 mm., equal, pallid-tan, stuffed, striate, slightly enlarged towards the base. Gleba ferruginous; capillitium hyaline, threads somewhat flattened, branched, sparingly septate, slightly swollen at the septa. Spores globose or subglobose, 4-6 » diam.; epispore coarsely and sparsely verrucose, verrucae arranged in striae, 1 uw thick. Habitat.—Solitary or caespitose on the ground. Distribution.—New South Wales: Forbes (Aug., 1915. J.B.C.; JId., Lloyd, No. 81, as J. poculatum). South Australia: Berri (Jan., 1921. J.B.C.; Id., LIBRARY = te % 256 GASTEROMYCETES OF AUSTRALASIA, ii, Lloyd, No. 751, as ZT. poculatum. Type); Ooldea (Aug., 1922. J.B.C., two collections), Unknown locality (J. G. O. Tepper; Jd., Lloyd, No. 14, as T. aus- tralianum . Specimens in Herb. Victorian Dept. Agr.). This species resembles 7’. poculatum in mouth characters, but is a much larger plant, and in addition possesses verrucose spores. The verrucae are arranged in striae, giving the spores a very characteristic appearance, and it is upon this character that the species is erected. It somewhat resembles the description of J. tuberculatum White, but as there is no mention of these peculiar and characteristic spore striae in the description, it is considered to be distinct and so kept separate. Section III. Mouth indefinite. 12. TULOSTOMA AUSTRALIANUM Lloyd. Plate xxxv, figs. 14-16. Tylostomeae, 1906, p. 20 (EHmended). Peridium strongly depressed-globose, almost pulvinate, up to 15 mm. high, 24 mm. diam.; exoperidium falling completely away save at the base; endoperidium smooth, dingy-white, tough, thick, membranous. Mouth indefinite, plane, a simple, irregularly torn aperture. Stipe up to 15 cm. X 6 mm., equal, covered with coarse deciduous scales, markedly striate, woody, stuffed, bay-brown, with a strongly developed woody bulbous base. Gleba ferruginous; capillitium hyaline, threads branched, moderately swollen at the somewhat sparse septa. Spores globose or subglobose, 4-6 u diam.; epispore finely and sparsely verrucose, pallid-ferruginous, 0.75 uw thick. Habitat.—Solitary on the ground. Distribution.—South Australia: Monarto South (May, 1921. J.B.C. Two collections, both Jd., Lloyd, Nos. 792, 781, as above), same locality (Sept., 1921. J.B.C.; Id., Lloyd, No. 793, as above). This species is characterized by the indefinite plane mouth, scaly stipe and verrucose spores. In his original description Lloyd states that the spores are smooth, and the stipe is not scaly. He further mentions that the species is the most abundant in Australia. “It has a large head and a short stipe, and in general appearance is the same as 7’. album (type specimen at Kew), but it has smooth spores (slightly granular in 7. album). Thus his original conception of the species was a plant with a non-scaly stipe and smooth spores. Nevertheless, he has determined no less than three collections of the plant described above as being T. australianum. Rather than erect another species, the writer has emended Lloyd’s description to include these collections. Lloyd has also determined as JT. australianum one collection (Dept. Agr. Vic. Coll., No. 14) which the writer places under 7’. striatum. Another the writer places under 7. albicans. Lloyd (1906) lists two additional species under this section. As the writer has not seen specimens, he cannot pass an opinion as to their validity, so gives Lloyd’s descriptions to enable the student to place them. TULOSTOMA READERI Lloyd. Tylostomeae, 1906, p. 21. Peridium uncoloured, firm, white. Cortex of the nature of a sand case, peeling off imperfectly and persistent at the base. Mouth an enlarged torn opening. Stipe long, dark, rough but not scaly, frequently with mycelial fibres. Capillitium hyaline, broad threads. Septa rare with rounded ends and not swollen. Spores 5-6 w, granular. BY G. H. CUNNINGHAM. 257 This plant has a general resemblance to J. granulosum, but has not the same mouth. Species collected by F. M. Reader at Casterton, Australia. TULOSTOMA EGRANULOSUM Lloyd. Tylostomeae, 1906, p. 21. Peridium uncoloured, with an irregular, torn aperture. Cortex as a sand case imperfectly separating, thickened and persistent at the base. The cortex does not separate as freely as most species, but adheres with a “pitted” effect on the peridium. Stipe dark, rigid. Capillitium subhyaline, with rare but swollen nodes. Spores 5-6 uw, granular. This plant is very close to 7. granulosum of Europe, but the mouths of these specimens are not furnished with “granular fibrils.” Australia: D. McAlpine, F. M. Reader. In Cooke’s Handbook of Australian Fungi, 1894, pp. 224-225, the following species are recorded as being found in Australia. Judging from the descriptions, one is led to believe that the whole were compiled from Saccardo’s Sylloge Fungorum, vol. 7. a. TULOSTOMA ALBUM Massee, Grevillea, vol. 19, 1891, p. 95. Lloyd states (1805) that: “The specimen on which this is based is old, weathered and bleached out. It has a large head and a short thick stalk, but I could not make out its mouth characters.” It is therefore impossible to place this species, for in the original description no mention is made of its mouth characters. b. TULOSTOMA GRANULOSUM Lev., Demid. Voy.. Vol. 2, 1842, p. 120, t. iv, fig. 1— Tulostoma brachypus Czern., Bull. Mosc., 1845, p. 144.—The description is too inadequate to allow of this species being placed. c. TULOSTOMA FIMBRIATUM Fr., Syst. Myc. vol. 3, 1829, p. 48.—It is probable that this is a misdetermination of 7. poculatum, for no species with ‘‘fimbriate’”’ mouths is known to have been collected in Australia. Lloyd (1905) states that there are specimens at Kew from Swan River so labelled, but he was unable to examine their mouths. d. TULOSTOMA LEPROSUM Kalchbr., Grev., vol. 4, 1876, p. 72.—This is referred to T. mammosum by Kalchbrenner, but is separated on account of the peridium being covered with a “lurid umber mealy scurf.” The description is too imperfect to allow of its being recognized. ; €. TULOSTOMA MAMMOSUM Fr.—This is a synonym of 7. brumale Pers. (a.v.). f. TULOSTOMA MAXIMUM Cooke and Massee, Grev., vol. 15, 1887, p. 94.—Lloyd, who has examined the type at Kew, states that this is a synonym of Chlamydopus Meyenianus (K1.) Lloyd, Letter, No. 67, 1918, p. 12. g. TULOSTOMA PULCHELLUM Saccardo, Bull. Soc. Myc. Fr., vol. 5, 1889, p. 118.— “This species . .. rests on a single little specimen ... made in Australia by Tepper. While it is not fair to form an opinion of a ‘species’ on one little speci- men, this must form a section by itself. The very short stem is itself a departure from all others and the stem does not appear to be inserted in a socket as in all other true Tylostomas. The colour is now chocolate brown, the surface, under a lens, scurfy. The mouth is simply an indefinite opening. Gleba, rust colour. Spores, unusually large for a Tylostoma, 8-9 uw, globose, pale coloured, smooth [5-6 uw, according to Saccardo].” Lloyd, Myc. Notes, vol. 7, 1923, p. 1233.—The species is said to grow upon branches. h. TULOSTOoMA WiGHTII Berkeley, Hook. Jour. Bot., 1842, p. 157.—The species was described from specimens collected in India. No specimens from Australia are at Kew. 258 GASTEROMYCETES OF AUSTRALASIA, li. Literature Cited. Besspy, C. E., 1887.—Growth of Tulostoma mammosum. American Naturalist, vol. 21, p. 665. CLELAND, J. B., and CHEEL, E., 1916.—Notes on Australian Fungi. III. Nidulariaceae and Lycoperdaceae. Journ. Roy. Soc. N.S.W., vol. 50, pp. 105-129. CooxKE, M. C., 1892.—Handbook of Australian Fungi, 457 pp. London. FISCHER, Ep., 1900.—Plectobasidiineae, in Engler and Prantl, Naturlichen Pflanzen- familien, 1'**, pp. 329-346. Luioyp, C. G., 1905.—The Lycoperdaceae of Australia, New Zealand and Neighboring Islands. 44 pp. Cincinnati. , 1906.—The Tylostomeae. 28 pp. Cincinnati. REA, CARLETON, 1922.—British Basidiomycetae. 799 pp. Cambridge. SCHROETER, J., 1887.—Ueber die Entwickelung und die systematische Stellung von Tulostoma Pers., Cohn’s Beitrage zur Biologie der Pflanzen, Bd. 2, p. 65. ; Toni, J. B. DE, 1888.—Lycoperdaceae, in Saccardo’s Sylloge Fungorum, vol. 7, pp. 48-154. WHITE, Miss V. S., 1901.—The Tylostomaceae of North America, Bull. Torrey Bot. Club, vol. 28, pp. 421-444. EXPLANATION OF PLATES XXXIII-XXXV. Plate xxxiii. 1. Tulostoma pubescens G. H. Cunn. x 5/4. This plant is much broken and so is not accurately represented in the photograph. 2. Tulostoma albicans White. Natural size. 3. Tulostoma McAlpinianum Lloyd. Natural size. 4. Tulostoma McAlpinianum Lloyd. Se 83/7. Peridium, showing structure of the definite, tubular mouth. 5. Tulostoma macrosporum G. H. Cunn. Natural size. Note the caespitose habit. Three peridia are joined together in the specimen on the right; the stipes have been broken away. 6. Tulostoma adhaerens Lioyd. Natural size. Plate xxxiv. 7. Tulostoma adhaerens ‘Lloyd. x 2. Peridium showing definite, elliptical, almost plane mouth. 8. Tulostoma brumale Pers. Natural size. 9. Tulostoma Purpusii P. Henn. Natural size. Specimen on the right has lost its stipe. 10. Tulostoma poculatum White. Natural size. Peridium in centre (lower) is inverted to show stipe socket. 11. Tulostoma poculatum White. x 2. Peridia showing fibrillose mouths; weathered specimen on the left. 12. Tulostoma subfuscum White. Natural size. 13. Tulostoma striatum G. H. Cunn. Natural size. Plate xxxv. 14. Tulostoma australianum Lloyd. x 3/4. 15. Tulostoma australianum Lloyd. Natural size. 16. Tulostoma australianum Lloyd. x 3/2. Peridium showing the indefinite mouth. 17. Tulostoma striatum G. H. Cunn. Spores. Note striae on the surface of the epispore. 18. Tulostoma subfuscum White. Spores. Note rounded sparse verrucae on the epispore. 19. Tulostoma albicans White. Spores. Note verruculose surface of the epispore. 20. Tulostoma macrosporum G. H. Cunn. Spores. Note large size of the spores and their rough surfaces. 1. Tulostoma brumale Pers. Spores. Note verrucose nature of the epispore. 22. Tulostoma poculatum White. Spores. Note that spores are smooth and irregular in shape. All photographs were taken by H. Drake. Photomicrographs magnified 540 diam. The photomicrographs were taken with a 4 mm. apochromatic objective, N.A. 0.7, comp. ocular x 10, 200 mm, tube length; Watson’s parachromatic condenser, yellow-green screen, filament lamp, x 540, exposure 2 minutes. TWO NEW SPECIES OF CALLISTEMON, WITH NOTES ON CERTAIN OTHER SPECIES. By EpwIn CHEEL, Curator of the National Herbarium, Sydney. [Read 29th July, 1925.] CALLISTEMON SUBULATUS, NSD. Low-spreading shrub with numerous stems or branches arising from the base, which, when crowded together, form compact bushy shrubs more than 4 feet high, the individual stems more or less virgate or of whip-stick nature, rarely exceeding 2 to 3 cm. in thickness. Leaves more or less subulate-awl-shaped, or occasionally semi-terete, sharp- pointed, with a somewhat obscure midvein slightly channelled on the underside, which is only visible under a lens; 34 to 4 cm. long, usually 3 to 4 mm. broad and comparatively thick, so much so that the venation and oil-glands are invisible when held up to the light, but a few oil-glands may be seen on the upper and lower surfaces as minute raised dots in dried specimens; slightly hairy in the juvenile stage, but soon becoming glabrous with age or when properly matured. Spike 5 to 8 cm. long, with a few scattered hairs along the rachis when young, but becoming glabrous when fully matured. Bracts from broadly-ovate to narrow-linear, more or less convolute striate, and covered with soft silky hairs, varying in length from 5 to 12 mm. Calyx-tube usually glabrous, or a few scattered hairs at the base, except in some specimens from Dorrigo and Gippsland which are rather more hairy than the type specimens, but otherwise the characters appear to be the same. Sepals ovate or semi-orbicular, somewhat brownish or slightly tinged with purple, rarely exceeding 2 mm. in diameter. Petals green, orbicular, 2 to 3 mm. diameter, glabrous or the margin very slightly ciliate. Stamens 14 to 2 cm. long, the filaments as well as the anthers of a rich crimson colour. Style greenish, 24 cm. long, with a comparatively conspicuous globose stigma. Fruits 3-celled, crowded together in the spike, the individuals truncate at the orifice, usually about 3 to 4 mm. diameter or rarely exceeding 5 mm. I have chosen, for the type, specimens collected from plants growing along the bed of the Nattai River in October, 1912, which were exhibited at the October, 1915, meeting of this Society (vide THESE PRocEEDINGS, x], 1915, 626). Several seedlings have been raised from the Nattai River plants and have been cultivated at Ashfield and at the Botanic Gardens, Sydney, as well as at Hill Top, at an elevation of 2,000 feet, and the offspring have flowered (during the past three years) and in no way differ from the parent plants. In comparing them with other known species, it is interesting to note that the foliage character and the habit of the plant is somewhat like C. Sieberi, but the filaments and anthers of the latter are creamy-white and much shorter than those of the new species, and the fruits are also different. H 260 TWO NEW SPECIES OF CALLISTEMON, Its nearest affinity is with OC. lanceolatus, but the smaller and differently shaped leaves, as well as the fruits, are quite distinct. In the National Herbarium, Sydney, there are specimens from the following localities: New South Wales: Nattai River, via Hill Top (EH. Cheel, January and October, 1912, and October, 1913); Conjola (W. Heron, January, 1900), determined by the late Mr. E. Betche as a “Narrow-leaved form of C. lanceolatus approaching C. rigidus and almost intermediate between the two. Fruiting specimens wanted.” As mentioned above, it is quite distinct from C. lanceolatus, both as regards the habit of the plant, as well as the structural characters of the bracts and fruits. Mr. Betche afterwards altered his decision and erased the name lanceolatus and substituted rigidus, but it is more distinct from OC. rigidus than C. lanceolatus; Nowra (J. L. Boorman, February, 1910); Moruya (J. L. Boorman, November, 1911); Dorrigo (W. Heron, November, 1912); Wyndham (J. L. Boorman, August, 1915); Cotter River (C. Weston, July, 1917); Heathcote (E. Cheel, in bed of the river). Victoria: Near Cann River, Gippsland (H. B. Williamson, No. 1619, January, 1918). CALLISTEMON CHISHOLMI, N.SDp. Plants of a shrubby habit, or occasionally growing into small trees, the lower part of the trunk and branches with bark similar to that of the black tea-tree, which is of a fibrous nature, the upper branches and branchlets comparatively smooth. . Leaves 5 to 8.5 cm. long, 4 to 7 mm. broad, linear-lanceolate, acuminate coriaceous, the lateral veins rather oblique, just visible to the naked eye, intra- marginal veins quite close to the margin of the leaf. Oil glands numerous, visible when held up to the light, raised into minute tubercles on both sides of the leaves when dried and examined with an ordinary lens. Spike up to 8 em. long, the rachis slightly woolly, bracts glabrous, faintly striate. Calyx-tube glabrous, lobes glabrous, greenish coloured with tinge of purple, petals greenish, tinged with red or purple, glabrous, or the margin minutely ciliate. Stamens about ? inch long, the filaments as well as the anthers, of a crimson or blood-red colour. Style slightly exceeding the stamens. Fruits sSub-cylindrical, 4 mm. long, 3 mm. diameter, with a thin rim and truncate orifice, the valve rather deeply sunk. The plant has some resemblance to C. linearifolius and C. rigidus, but is distinguished by the leaves being less prominently veined, and the paler filaments and differently shaped fruits. It is named in honour of Mr. J. R. Chisholm, who has interested himself in the flora of the remote parts of Northern Queensland. Habitat.—Western watershed, Thompson River Fall, North Queensland (J. R. Chisholm, August, 1921). CALLISTEMON LINEARIFOLIUS DC. Metrosideros linearifolia Link, Enumer. Pl. Hort. Berol. ii, 1822, 26.—C. linearifolius DC., Prodromus, iii, 1828, 223. Bentham (1866) includes this under C. rigidus R.Br., as a synonym, but the specimens in the National Herbarium, Melbourne, labelled C. linearifolium, were collected in the neighbourhood of Parramatta by Woolls. Specimens collected by the writer quite close to Granville Station are almost identical with those collected by Woolls. Other localities are Clyde, Macquarie Fields, Cabramatta, BY E. CHEEL. ‘ 261 Hornsby, Lane Cove River, Cowan and Berowra. Plants cultivated at Ashfield, raised from seeds collected at Hornsby, flowered in October, 1921, and proved to be quite distinct from C. acuminatus Cheel (see THESE PROCEEDINGS, xl, 1915, 626), which it very closely resembles and for which it had been mistaken; it is also quite distinct from C. rigidus R. Br. CALLISTEMON PACHYPHYLLUS Cheel. The type specimens were collected at Bulahdelah and a description, together with an illustration, has been published (Cheel, 1911). Plants raised from seeds of the type have been cultivated in the Botanic Gardens and the colours matched that given as crimson-red (in bud) (see Dauth., Repert. des Couwl., 114, 2), tending to strawberry-red (Dauth., 110, 1) and finally dull-dark crimson (auth, 168, 3). Additional localities are Byron Bay and Wallis Island. In addition to the normal form or species, some distinctive varieties have been collected which may be described as follows: C. PACHYPHYLLUS var. ANGUSTIFOLIUS. Flowers similar to the type, leaves smaller, narrow-linear, 34 to 6 cm. long, 3 to 4 mm. ‘broad. Habitat.—Wardell (W. Bauerlen), Coff’s Harbour (J. L. Boorman). C. PACHYPHYLLUS Var. VIRIDIS. Leaves similar to var. angustifolius, but filaments yellowish-apple-green (Dauth., 266, 1 and 2). Habitat—New South Wales: Coff’s Harbour (J. L. Boorman, June, 1911, and March, 1912; also A. Webber, June, 1920). Queensland: Caloundra (Miss E. Taylor, September, 1920—comm. by C. T. White, Government Botanist, Queens- land, and marked C). C. PACHYPHYLLUS Var. RUBRO-LILACINUS. Leaves and other characters similar to the type, but the filaments are of a reddish-lilac tint (Dauth., 179). Habitat—Caloundra (Miss E. Taylor, comm. by C. T. White). CALLISTEMON RUGULOSUS DC. . Prodromus, iii, 1828, 223. The description given by Decandolle is in latin and may be translated ag follows: ‘Leaves linear-lanceolate, rigid, mucronate, plain, oil-glands raised on one surface and on the margins into minute tubercles, somewhat scabrous, 3-nerved, the lateral nerves parallel and close to the margin. Fruits glabrous.” The following are given as synonyms: Metrosideros rugulosus Willd., M. scabra Coll., M. giandulosa Desf. and M. macropunctata Dum. Cours. It is also described by Link (1822) who quotes Willdenow’s work, and from the brief latin description given, there can be no doubt it is the same species. Then we have Miquel (1856) who gives the habitat as “arid parts of New South Wales, together with Encounter Bay, Lake Alexandriae, Gawler Town, Lyndock Valley and Crystal Brook.” Dr. Behr is given as the collector of specimens from the South Aus- tralian localities. Mueller (1858-59) seems to have been under the impression that Miquel’s species was distinct from that of Decandolle and describes it under the name C. coccineus, giving as localities: St. Vincent’s Gulf, Murray River, Spencer’s Gulf, Flinders Range and Kangaroo Island. The latter are in the Melbourne Herbarium, and are in no way different from those which I have also 262 TWO NEW SPECIES OF CALLISTEMON, examined from Gawler Town and Lyndock, both of which localities are quoted by Mueller for his C. coccinews. Bentham (1866) evidently follows Mueller in regarding ©. coccineus F. v. M. as distinct from C. rugulosus DC., and includes ag synonyms: Metrosideros rugulosa Desf. together with Callistemon scaber Loddiges (Bot. Cab., t. 1288). From a careful analysis of Decandolle’s descrip- tion and a close study of the figure of Loddiges, it seems clear to me that Miquel’s and Mueller’s species in no way differ from that of Decandolle. We may also include C. coccineus F. v. M. of Moore and Betche (1895), Bailey (1900) and J. E. Brown (1882). This species is frequently cultivated in gardens and there are several specimens from garden plants represented in the herbaria of Sydney and Melbourne. There are also specimens from Cygnet Bay, Wirraba Forest, Laura, Port Lincoln and Hog Bay. Victorian localities are as follows: Lake Albacatya, N.W.; Mount Arapilis; and without specific locality specimens col- lected by Hooker and Mueller labelled “CO. coccineus, C. rugulosus Miq. vix DC.” CALLISTEMON LAEVIFOLIUS. C. rugulosum var. laevifolia F. v. M., in Miquel, Ned. Kruidk. Arch. iv, 1856, 141.—C. coccineus var. laevifolius F. v. M., Fragm. Phytog. Aust. i, 1858-59, 13. The original plants were collected in Port Lincoln by Wilhelmi in 1852. Miquel described it as a variety of CO. rugulosum DC., and Mueller regarded it as a variety of C. coccineus F. v. M., but, although the filaments are similar in colour to those of C. rugulosus, the habit of plant and differences in foliage and other characters are so marked that it is proposed to raise it to specific rank. Plants of a spreading shrubby habit. Leaves 1 to 1% inches long, 3 to 5 lines broad, rather broadly-lanceolate, obtuse, but the mucro at the apex very prominent. Lateral veins obscure, but the central and marginal nerves much more prominent. Oil-glands entirely obscured owing to the thickness of the cuticle, quite smooth. Bracts, calyces, petals, filaments and fruits similar to those of C. rugulosus. Distribution.—In addition to the Port Lincoln specimens, there are col- lections made at Coffins Bay (J. H. Maiden, January, 1907), Venus Bay (Major Warburton in 1859), and near Spencer’s Gulf (without the collector or date being given). In the Melbourne Botanic Gardens there is a very fine plant cultivated by the late Mr. W. R. Guilfoyle labelled C. coccineus var. splendens. CALLISTEMON LILACINUS. Plants of this were raised from seeds originally obtained from the Botanic Gardens, Berlin, in 1913, labelled C. lanceolatus var. lilacina. The habit is quite distinct from C. lanceolatus, commonly found in the Port Jackson district, as will be seen from the following description: Filaments of a purplish-mauve-lilac colour (Dauth., Pl. 186). The leaves are similar to those of C. rugulosus in shape, but are much thinner in texture and the oil-glands are not raised on the surface. Specimens identical with those raised from Berlin seeds sown under No. 856 on 29th May, 1913, have also been collected near Como by the late Mr. EK. Betche in November, 1894, and recorded (THESE PROCEEDINGS, xxx, 1903, 884) under the name C. coccineus F. v. M., on account of the filaments and anthers being of the same colour. A closer study, however, of plants under cultivation in the Botanic Gardens, Sydney, and at Ashfield, has convinced me that they are quite distinct from both C. rugulosus and C. lanceolatus with which they have been previously confused. This view has also been confirmed, through the discovery of plants BY E. CHEEL. 263 growing wild at Gosford by Mr. T. Johnston in November, 1921. The late Mr. J. H. Camfield also collected specimens in the Port Jackson district without giving the specific locality in November, 1894. It is a most beautiful plant when in flower, some of the forms flowering nearly all the year round. Variants have also been produced from the typical C. lilacina, which suggest that its origin is probably as a result of hybridism. The variants may be described as follows: C. LILACINA—CARMINA. Similar in habit to the type, but the filaments are of a rich reddish-plum colour or deep carmine-violet (Dauth., Pl. 174, 4). Attention was drawn to this form by an exhibit of fresh flowering specimens at a meeting of this Society (THESE PROCEEDINGS, xlvii, 1922, xxviii). C. LILACINA—ALBA. Filaments creamy-white with very faint flush of rose. It is interesting to note that two plants have been discovered in nature at Long Bay, by Mr. H. Burrell, of the true C. lanceolata, but with the filaments pure white. One of the latter is now cultivated in the Botanic Gardens, Sydney, and the other at Taronga Park. The habit of the plant and character of foliage of the latter form are quite distinct from the lilacina forms. ‘ CALLISTEMON FLAVO-VIRENS. C. rugulosus var. flavo-virens Cheel, in Maiden, Jllust. N.S. Wales Plants, iii, 1911, p. iv. Leaves 24 to 4 inches long, 3 to 6 lines broad, linear-lanceolate, acute, some- what attenuated at the base, and thickly studded on both sides with comparatively prominent raised oil-glands. Inflorescence and young shoots densely clothed with silky-pubescence. Flower-spikes 2 to 34 inches long. Bracts very variable, 4 to 7 lines long, minutely pilose or even villous in some spikes. Calyx-tube villous, the lobes very short, rounded, deciduous. Petals rather larger than those of C. rugulosus, greenish, sprinkled with a few resinoid oil-glands. Filaments about 7 to 9 lines long, yellowish-green in colour, anthers yellow. Fruits somewhat more contracted at the orifice than those of C. rugulosus and more closely resembling those of C. phoeniceus from Western Australia. In the National Herbarium, Sydney, there is a fine series of specimens collected by Mr. J. L. Boorman, who in his notes states “it is a yellow form of C. lanceolatus, 2 to 4 feet high, in the beds of creeks at Boonoo-Boonoo and Stan- thorpe.” Some fine seedling plants of this were raised from seeds collected at Stanthorpe and cultivated in the old native flower border in the Botanic Gardens, Sydney, and also in the Native Flora Plantation in the Centennial Park. The characters of the seedling plants were in no way different from those of the parent plants, and although included as a variety of C. rugulosus in the descrip- tive key drawn up by the writer (1911) I now find the characters sufficiently distinct from those of C. rugulosus in that the filaments are of a yellowish-green without any trace of rosy-pink, and the leaves are longer, thinner in texture, and the oil-glands are more numerous and, although distinctly raised as minute tubercles on both surfaces, the margins are not in-rolled nor scabrous as in OC. rugulosus. It is therefore proposed to raise this to specific rank in view of seedlings breeding true. There are also specimens collected from a plant cul- tivated in the Domain Gardens, Christchurch, New Zealand, by the writer in March, 1909, which appear to belong to this species. 264 TWO NEW SPECIES OF CALLISTEMON, CALLISTEMON PALLIDUS DC. The original description of Decandolle (1828) is in latin and may be freely translated as follows: “Leaves glaucous oboval-oblong, mucronate, glabrous when fully matured, lateral nerves barely visible, calyx-tube glabrous.” Decandolle’s description is evidently based on specimens collected in Tasmania during Baudin’s Expedition and figured by Bonpland (1813) under the name Metrosideros pallida. Hooker (1860) included this species under C. salignus as a synonym and states that it is abundant on river banks in all parts of the colony. Hooker further states that at first sight this does not appear to differ much from C. viridiflorum, but it is quite a distinct species, having much longer and less rigid leaves, with shorter stamens, and that he can find no difference between them and C. salignus DC. as found in New South Wales, and the Tas- manian plant, except that the calyces of the former (C. pallidus) are sometimes, but not always, hairy and the leaves are hardly as long. Hooker’s latter remarks were in reply to Planchon, who, having examined Hooker’s herbarium, had pointed out that the Tasmanian plants under the name C. salignus were distinct from the New South Wales plants under the same name (vide Hook., Fl. Tasm., i, p. 131). Bentham evidently recognized that the Tasmanian plants were distinct from the New South Wales plants and describes them as varieties under the names C. salignus var. hebestachyus and var. viridiflorum. The following is Bentham’s description of C. salignus var. hebestachyus: “Leaves rather small. Calyx and rhachis pubescent or villous.” The speci- mens examined by Bentham are in the National Herbarium, Melbourne, and as types we may take those from South Port and Mersey River, collected by C. Stuart in January, 1850, under No. 1637. In the National Herbarium of Sydney there is also a fine series of specimens, identical with those examined by Bentham and which appear to me, to have scarcely any resemblance, except in the colour of the flowers, to the common coastal C. salignus DC. of New South Wales, and, besides, the oil-glands and venation afford sufficiently distinctive characters as shown in the key to the species (Cheel, 1911). In order to separate them, I propose to adopt Decandolle’s name C. pallidus to include the larger broad-leaved form which is common in Tasmania and also the name C. viridiflora DC. for the smaller stiff-leaved, greenish flowering form. The C. lophanthum Sweet (1827-28) quoted by Bentham as a synonym in connection with his C. salignus var. hebestachyus seems to be somewhat different from the Tasmanian plants, but may be only garden forms of the same species, as they are stated by Sweet to have been introduced into English gardens from France. As there are no mature fruits figured in connection with either of the above- mentioned illustrations it is very difficult to judge as to whether they are seedlings from the New South Wales plants or from those of Tasmania. It will be noticed from the figures that the lateral venation is not very prominent and as the lateral venation in all the New South Wales plants (of C. salignus) is very prominent, they can scarcely belong to this species and are best referred to C. pallidus DC. as synonyms for the present, especially seeing that several seedlings raised from seed of C. salignus, which are now under cultivation under varying conditions, all show the same characteristic features of the parent plants and have no resemblance to those from Tasmania. BY E. CHEEL. 265 Description.—Usually upright bushy shrubs which often develop into a small tree somewhat resembling C. salignus DC. in general appearance, but may be distinguished by the less flexile and less drooping branches so characteristic of the latter species. Leaves clothed with appressed silky silvery-coloured hairs when young, not rufous as in C. salignus, which drop off with age; the leaves then are of a pallid-green or semewhat glaucous colour. Oblong or oboval in general out- line, slightly tapering at the base and terminating at the apex with a sharp mucro, thicker in texture than those of C. salignus and as a consequence the venation is not so prominent, varying in size from about 1 to rarely more than 24 inches long and 38 to 6 inches broad. Oil-glands appearing on both sides of the leaves as minute tubercles quite visible to the naked eye. Flower-spikes from 2 to 3% inches long, the rachis pubescent or villous or rarely almost glabrous. Bracts not seen, apparently very deciduous. Calyx-tube cylindrical or campanulate, the lobes very small and generally villous. Petals greenish, slightly pubescent, the margins sometimes fringed with soft hairs, and the underside usually dotted with resinoid oil-glands. Stamens creamy-coloured or yellowish, tinged with a greenish-tint, 4 to # inch long, anthers slightly darker in colour than the filaments. Fruits ovate or semi- ovate, with a truncate orifice and thick rims, the valves nearly level with the rim. Synonyms.—Metrosideros pallida Bonpland (1813); Callistemon salignus Hook. (1860) (not DC.); C. salignus var. hebestachyus Benth. (1866); OC. lanceo- latus Ewart (1909) (not DC.). What appear to be only forms of the same species are also found on mountain peaks in this State (New South Wales) and Victoria at altitudes above 2,000 feet. The localities are as follows: Victoria: Buffalo Range (F. v. Mueller, 1856); Granite Hill, Wilson’s Promon- tory (J. W. Audas, November, 1908). New South Wales: Mount Warning (W. Forsyth, November, 1898, and R. H. Cambage, January, 1913), Jenolan Caves (W. F. Blakely, July, 1899), Clyde Mountains, near Nelligen (J. lL. Boorman, March, 1909), Top of Tidbinbilla (5,000 feet western side, R. H. Cambage, No. 3052, Nov., 1911), Yerranderie (R. H. Cambage, October, 1909), Barren Mountain, near Meldrum (J. L. Boorman. December, 1915), Mount Jellore (2,000 feet, E. Cheel, May, 1916), Barrington Tops (J. L. Boorman, December, 1915, L. Harrison, January, 1925). There are also specimens in the National Herbarium, Sydney, collected at Yarrowitch near Walcha by J. H. Maiden in November, 1897, and at Stanthorpe, Queensland, by J. L. Boorman in July, 1904, which vary slightly from the Tasmanian forms, but may be safely included under this species. Some Stan- thorpe specimens with coriaceous leaves referred to by Bailey (1900) seem to belong to this species, rather than to C. lanceolatus. List of References. BaiLzy, F. M., 1900.—Queensland Flora, ii, 595. BENTHAM, G., 1866.—Flora Australiensis iii. Brown, J. H., 1882.—Forest Flora of S. Aust., p. 38, Pl. 38. Brown, R., 1812.—In Aiton, Hortus Kewensis, ed. 2, iv, 415. BoNnPLAND, A., 1813.—Description des plantes rares cultivées &@ Malmaison et a Navarre. Paris, t. 4, p. 86 et p. 101, t. 41. (Quoted by some authors as “Jard. Malm.”’). 266 TWO NEW SPECIES OF CALLISTEMON, CHEEL, E., 1911.—In Maiden, Illustrations of N. S. Wales Plants, iii. p. iv. , 1915.—Proc. Linn. Soc. N.S.W., xl, 626. , 1922.—Proc. Linn. Soc. N.S.W., xlvii, p. xxix. Cortuia, Luicit.—Hort. Ripul., p. 91 (quoted by Decandolle under C. rugulosus). DECANDOLLE, A. P., 1828.—Prodromus, iii, 223. DESFONTAINES, R. L.—Cat. Pl. Hort. DuMONT DE CouRsET, G. L. M., 1802-1811.—Botanist Culliviv Hort., ed. 2, 379; 5, 178; et 7, 477. Ewart, A. J., 1909.—Victorian Naturalist, xxv, 145. GUILFOYLE, W. R.—Australian Plants Suitable for Gardens, Parks, etc., p. 90. Hooker, J., 1860.—Flora Tasmaniae, i. Link, H. F., 1822.—Hnumer. Pl. Hort. Berol. ii, 26. MIQUEL, F. A. G., 1856.— Ned. Kruidk. Arch. iv, 141. Moore, C. and BercHe, E., 1895.—Handbook Flora of N.S.W., 195. MUELLER, F. v., 1858-59.—Fragm. Phytog. Aust. i, 13. , 1889.—Second Census. SWEET, R., 1827-8.—Flora Australasica, Pl. 29. WILLDENOW, C. L.—Enumeratio, Suppl. p. 31. A COMPARISON OF THE MALE GENITALIA OF THE PALAEKOSETIDAE WITH THOSE OF OTHER LEPIDOPTERA HOMONEURA. By JoHn R. Eyrer, Pennsylvania, U.S.A. (Communicated by A. J. Turner, M.D., F.E.S.) (Twelve Text-figures. ) [Read 26th August, 1925.] The lepidopterous family Palaeosetidae was described by Turner (Proc. Ent. Soc. Lond., 1921, 599). Based on a study of the type, Palaeoses scholastica Turn., the family was assigned a position in the Hepialoidea of the Lepidoptera Homoneura, differing from the families Hepialidae and Prototheoridae in certain structural details of the wing venation, mouth parts and tibial spurs. Through the courtesy of Dr. Turner, the author has been able to examine the male genitalia of this interesting family, and the following is a comparison of these structures with those of other Lepidoptera Homoneura. An attempt is also made to express the characters offered by these structures in a numerical manner, much as Tillyard (Proc. Linn. Soc. N.S.W. 1919, 275-328) has done with the wing venation, and thus calculate the percentage of archaic characters preserved in the genitalia of each family. In comparison with the genitalia of other Lepidoptera Homoneura the genitalia of Palaeoses (Text-figure 11) exhibit the following distinctive features. The eighth sternite is normal in size and shape, neither heavily chitinized, as it is in the Hepialidae, nor rudimentary or absent, as in the Micropterygidae. The ninth sternite or vinculum is U-shaped and of moderate size, as in certain Hepialidae, e.g. Hepialus lupulinus L. (Text-figure 7). The tegumen, or fused ninth and tenth terga, is narrow, emarginate on the meson of the dorsal surface and bears two pairs of anal processes, one pair caudad of the anus and the other slightly laterad. These processes closely resemble similar ones in the Prototheoridae (Text-figures 9 and 12), but are longer and more attenuate. Similar processes also occur in certain Hepialidae (Text-figures 8 and 10), but usually in this family the lateral pair are fused with the aedeagus and become a part of the intromittent organ. The pleural region of the tegumen in Palaeoses is expanded on each side and forms a pair of lateral lobes similar to those in certain species of Micropteryx (Text-figure 6), but directed inward rather than caudad as in Micropteryz. The juxta consists of a quadrate chitinized plate situated on the meson directly caudad of the vinculum. It is similar to the same structure in the Hepialidae and Prototheoridae. The aedeagus, a recurved chitinized process, is articulated to the caudal margin of the juxta and forms the ventral support for the membranous penis. A similar type of aedeagus is present in many Hepialidae, H. lupulinus (Text-figure 7) et al. As will be seen in Text-figure 9, the aedeagus of the Prototheoridae is more complicated, having a large semi-mem- branous terminal portion armed with a pair of recurved processes and lateral lobes. The harpes or valves of Palaeoses are articulated to the lateral margins of I 268 COMPARISON OF MALE GENITALIA OF LEPIDOPTERA HOMONEURA, the juxta and each is divided into two areas, a narrow finger-like sacculus and a broad spoon-shaped cucullus. The harpes of the Prototheoridae and most Hepialidae are undivided, but a type somewhat similar to that of Palaeoses is to be found in H. hectoides Boisd. (Text-figure 8). The male genitalia of Palaeoses have little in common with those of the Micropterygoidea which depart quite markedly from those of the Hepialoidea and approach the type of genitalia found in the primitive Tineoidea. Only in the Mnesarchaeidae (Text-figures 1, 4, and 5), are any distinct MHepialoid tendencies preserved. A U-shaped vinculum, similar to that of Palaeoses and many Hepialidae, occurs in this family, and Mnesarchaea fusca Philp. (Text-figure 5) possesses a pair of anal processes probably homologous with the more caudal pair of Palaeoses and Prototheora. The penis is membranous in the Mnesarchaeidae and is occasionally supported on the ventral side by chitinized spines, possibly the vestige of the aedeagus. In the Micropterygidae (Text-figures 3 and 6) and the Eriocraniidae (Text-figure 2) the aedeagus forms a tubular chitinized sheath for the penis, the juxta is greatly reduced in size and the harpes are often partially articulated to the vinculum. As previously mentioned a comparison of the archaic characters retained in the wings of the Lepidoptera Homoneura has been made by Tillyard. He used as the basis for this comparison the wing of Belmontia, the fossil ancestor of the Lepidoptera and Trichoptera, in which the maximum number of archaic characters were preserved. Unfortunately no such ancestral type of genitalia is at present known. From a study of the types of genitalia in the Lepidoptera and those insect orders closely related to the Lepidoptera, i.e. Neuroptera, Mecoptera and Trichoptera, some idea is obtained of the relative value of each structure comprising the genitalia complex and a hypothetical ancestral type may be postulated. The characters of such a type, as explained by the author in a previous publication (Ann. Ent. Soc. Amer., 1924, 275-343), are as follows: Tegumen hood-like, composed of the fused ninth and tenth terga, often armed with one or more anal processes which seldom become definite enough in shape and position to be homologized with the uncus and gnathos of the Lepidoptera Heteroneura; vinculum composed of the ninth sternite which retains its normal shape and size; eighth sternite unmodified; harpes or valves undivided, at least in part articulated to a membranous or lightly chitinized basal plate or juxta; penis membranous and encased in a tubular, lightly chitinized aedeagus. Text-figs. 1-12. . Mnesarchaea hamadelpha Meyr. Ventral aspect . Hriocrania semipurpurella Stph. Lateral aspect. . Sabatincea chrysargyra Meyr. Ventral aspect. . Mnesarchaea loxoscia Meyr. Ventral aspect. . Mnesarchaea fusca Philp. Ventral aspect. . Micropteryx calthella L. Lateral aspect. . Hepialus lupulinus L. Ventral aspect. . Hepialus hectoides Boisd. Ventral aspect. 9. Prototheora petrosema Meyr. Ventral aspect. 10. Perrissectis australasiae Don. Ventral aspect. 11. Palaeoses scholastica Turn. Ventral aspect. 12. Metatheora corvifera Meyr. Ventral aspect. OAD OF Ww DH Abbreviations. a—anus, ae—aedeagus, al—anal lobe, ap—anal process, cuc—cucullus, h—harpe or valve, j—juxta, /—lateral lobe of tegumen, mp—median process of tegumen (possibly the homologue of the uncus), p—penis, sac—sacculus, Sg—eighth sternite, T,—eighth tergite, t—tegumen or fused ninth and tenth terga, v—vinculum or ninth sternite. BY JOHN R. EYER. 269 jo.australasiae Be ae . ey: 37 < 12. Corvifera 270 COMPARISON OF MALE GENITALIA OF LEPIDOPTERA HOMONEURA, In the following table the characters of the male genitalia for the families of the Lepidoptera Homoneura, with the exception of the Anomosetidae, are sum- marized, together with calculations of the percentage of archaic characters retained by each. The hypothetical archaic type just described is regarded as possessing 100% archaic characters. Six characters are tabulated, each of which is regarded as a separate unit, and the extent to which each is retained in its primitive state in the different families is expressed in fractions of 100. TABLE OF ARCHAIC GENITALIA CHARACTERS RETAINED IN FAMILIES OF LEPIDOPTERA HOMONEURA. Character. Palaeosetidae. Hepialidae. Mnesarchaeidae. Eriocraniidae. Micropterygidae. Prototheoridae. Highth Sternite: 1. Normal, same shape and degree of chitinization as other abdominal sterna—100 2. Hind margin heavily chitinized—66% 3. Entirely chitinized, plate-like—33% 0 100 668 334 100 100 4. Dechitinized, vestigial or wanting—0 | Ninth Sternite: (Vinculum) Shape: 1. U-shaped or quadrate, same size as other sterna—100 2. U-shaped or quadrate, somewhat elon- gated—75 3. Irregularly U- or V-shaped, elongated —5b0 4, Emarginate, outer angles of front margin produced—25 5. Narrowly transverse, often with saccus —(I) Position : 87.5] 100 75 25 100 BT ott 1. Normal—100 2. Front margin retracted beneath 8th sternite—50 3. Entire sternum retracted beneath 8th sternite—0 1 | Tegumen: (Ninth and tenth terga) 1. Hood-like—100 2. Transverse with at least one pair of | anal lobes—50 3. Emarginate or completely divided, with] 100 0 | O jj 0 50 50 one or two pairs of anal processes—0 Valves or Harpes: (Size and complexity) Large and simple—100 | Medium size, occasionally divided—75 | Medium size, invariably divided—50 | | Usually small—25 100 50 7 i 2s | 25 (| 0 Vestigial or absent—0 | | | gm go be BY JOHN R. EYER. — Del TABLE OF ARCHAIC GENITALIA CHARACTERS RETAINED IN FAMILIES OF LEPIDOPTERA HOMONEURA.— (Continued. ) o g 3 o Ss | 3 s s 3 2 | e 3 a ic fo) = Character. 5 R 2 & 3 is rf @ 3 = K 3 S | 3 iS ce a 5 3) o © a e ac & G = ) SI a a Ay ss a < | | Juxta and Basal Attachment of Valves: 1. Valves entirely attached to juxta, juxta large—100 2. Valves one-half attached to juxta, juxta medium—66# 3. Valves one-third attached to juxta, | juxta small—334 6632 100 100 100 | 334 33% 4. Valves entirely attached to vinculum, juxta absent—0* | | Aedeagus: 1. Tubular, basal portion at least chit- | inized—100 2. Semi-tubular, dorsal and ventral sur- faces chitinized—75 3. Plate or spine like—50 4. Plate like with elaborate terminal armature—25 100 50 25 | 50 0 75 5. Vestigial or absent—0 Average percentage preserved archaic] 76 67 BM LA wile 45 characters | | * Rarely found in Lepidoptera Homoneura but characteristic of certain Tineoidea and Macrolepidoptera.. Conclusions. From the foregoing comparison it was seen that the male genitalia of Palaeoses possessed certain structures quite similar to or identical with those in the Prototheoridae and Hepialidae, i.e. the emarginate tegumen with two pairs of anal processes, the large juxta to which the harpes are entirely attached, and a non-tubular type of aedeagus. The vinculum, although shaped much like that of the Prototheoridae and certain Hepialidae is not retracted beneath the eighth sternite. In the retaining of archaic characters, however, the genitalia surpass those of the Hepialidae and Prototheoridae and stand intermediate between the latter family and the Micropterygidae. As would be expected from the evidence offered by other structures, the Micropterygidae retain the greatest number of archaic genitalia characters in the families of the Lepidoptera Homoneura. In this respect the Eriocraniidae are the most specialized and the Mnesarchaeidae, which are less generalized than the Eriocraniidae in the matter of wing venation and mouth parts, retain more archaic characters than either the Hriocraniidae or the Hepialidae. THE AUSTRALIAN SPECIES OF ONCOPERA (HEPIALIDAE, LEPIDOPTERA). By JoHn R. Eyer, Pennsylvania, and A. J. TuRNER, Brisbane. (Plate xxxii.) [Read 24th June, 1925.] Introduction (by A. J. Turner). Some two years ago, in response to a request from Mr. Eyer for specimens of Hepialidae for the examination of their genitalia, I sent him examples of Oncopera mitocera Turn. taken in the National Park, Macpherson Range, Queensland. I was surprised to hear that I had, in his opinion, included another species, and it was only after a considerable time that I recognized that I had a second example of this previously unrecognized species, to which I have given the name epargyra. Meanwhile Mr. Eyer returned my first specimen with a microscope slide of the genitalia. These circumstances led me to make a critical examination of all my material, and I came to the conclusion that I had probably a third closely allied species from Herberton, for which I proposed the name brachyphylla. Mr. Eyer confirmed this from an examination of the genitalia, and sent me a slide of these, together with microphotographs of the genitalia of mitocera and brachyphylia, and a drawing of those of intricata. He also sent a key to the four species with the exception of epargyra, of which he had previously returned the only preparation he had made. Mr. A. Philpott, of the Cawthron Institute, Nelson, New Zealand, has very kindly completed the key, and Mr. W. C. Davies of the same Institute has sent me an excellent photograph of the genitalia of epargyra. Key to the species of Oncopera based on the characters of the male genitalia (by John R. Eyer; epargyra by A. Philpott). Group I. Sacculus absent. Valves without sacculus, cucullus expanded and spoon-like; vinculum wide-margined and possessing a prominent median process to which the posterior margin of the eighth sternite is attached; eighth sternite triangular, longer than wide, and with the anterior margin emarginate; tegumen without anal spines; scale sacks on conjunctiva between eighth tergite and tegumen present; aedeagus a small transverse plate; opening of penis bounded laterally by two heavy processes from the meson of Le gKePhnts} =a) boa \ ch Cas ean enee ot reneNeney contre cua cstretD exo ot OrciOro/ CIRC mERS mnie ehakal Oroiora.c O°OrO c mitocera Turn. Valves without sacculus, cucullus not expanded distally; vinculum narrow-margined and with the median process only slightly developed; eighth sternite rectangular, almost as broad as long; tegumen with a pair of short sharp anal spines; scale sacks absent; other characters as in O. mitocera ..............++-. intricata Walk. Group II. Sacculus present. Valves with short sacculus, cucullus long and finger-like; vinculum narrow-margined and without median process; eighth sternite keystone-shaped; tegumen with two anal spines which are continuous with two long inner processes which cover the opening of the penis; scale sacks present; aedeagus a rectangular plate much longer 6) gE 0 01011 6 = Saye mee tect ean tt aie MeL Oa whee Serena clones letested MPM EL UREA Sarto brachyphylla Turn. Valves with sacculus narrow and acutely pointed, almost as long as cucullus, cucullus less broad than in brachyphylla, not dilated apically; eighth sternite shuttlecock- shaped, upper angles projecting; vinculum broad, not emarginate, without median process; aedeagus a somewhat ovate plate ...................2.08- epargyra Turn. BY JOHN R. EYER AND A. J. TURNER. 273 Key to superficial characters and descriptions of new species (by A. J. Turner). 1. Antennae slightly clubbed towards apex. Femoral hair-tufts of o&@ very large and Teaching PEVOnd tarsi, cc: oreyecvesste ane G. srsie ene vey wrote cane asye Sues Ge se, vusuouanes ereeuaul eveneus intricata Antennae short, thread-like. Femoral hair-tufts of & moderate and not reaching [ovensCoy aKa leq ey ko} LAO Ane rane ol cence n Ree RCR En SRO tee ateES ia ChchencT heer uCrenieOR tLe IO aera Sicko a emaraT ce 2 2. Expanse, 0 42-46 mm. Forewings elongate and rather narrow, without silvery-white MArkKIne som swaithy GOES) “OMY ss sree ees wow sajseveesasceast op suey susdekep oueacenepoie ene sapere elope ce mitocera Expanse, ¢0 35-40 mm. Forewings shorter and proportionately broader, frequently with oblique or longitudinal silvery-white streak .....................00.0. 3 3. Forewings rather sharply pointed, with much fuscous suffusion. Hindwings with APICES LLC Ve sega aven once ka roey a serene 2 Sy eth Sistah cy Gy qa cng sthay ht ap ayes creme yw pha clreles epargyra Forewings with apex more bluntly rounded, little or no fuscous suffusion. Hindwings Valid CHEE KCC a TVOTRA TED con em pipe eee tOGaiNEnG cheweroEted S’atO Cis oeralo cla/oiolotdo me aio einer B brachyphylla ONCOPERA INTRICATA WIlk. Plate xxxii, fig. 1. New South Wales: Mt. Kosciusko. Victoria: Melbourne, Warragul, Gisborne. Tasmania: Hobart, Maria Island, Swansea, Bothwell, Launceston, Deloraine, Moina (2,000 ft.). ONCOPERA MITOCERA Turner. Plate xxxii, fig. 2. Annals Queensland Museum, x, 1911, p. 132. North Queensland: Cairns, Atherton, Herberton. Queensland: National Park (3,000 ft.). ONCOPERA EPARGYRA, n.Sp. Plate xxxii, fig. 4. émapyvpos, overlaid with silver. ¢. 36-41 mm. Head and thorax fuscous-brown. Antennae fuscous. Abdomen fuscous, becoming whitish-grey towards apex. Legs fuscous; posterior pair and tuft pale-ochreous. Forewings elongate-triangular, costa straight, slightly arched near base and apex, apex sharp-pointed, termen nearly straight, strongly oblique; dark-fuscous mixed with brown, paler towards base and costa; four or five fuscous spots on terminal half of costa; a variable, silvery-white, discal spot at two-thirds connected by a silvery-white oblique streak to dorsum before tornus, but this is not always present; cilia brownish barred with fuscous. Hindwings with apex round-pointed, termen slightly rounded; dark-fuscous; a suffused whitish-grey apical blotch, containing four costal fuscous dots; cilia as forewings. Queensland: National Park (3,000 ft.) in December, two specimens. ONCOPERA BRACHYPHYLLA, n.sp. Plate xxxii, fig. 3. BpaxvduAdos, short-winged. dg. 35-40 mm. Head and thorax brown. Antennae fuscous. Abdomen fuscous- brown. Legs brown. Forewings triangular, costa straight, arched towards apex, apex round-pointed, termen bowed, oblique; brown with little or no fuscous suf- fusion; costal area paler with some darker strigulae; a variable white discal mark at three-fourths, sometimes connected by an oblique white streak with dorsum before tornus; sometimes a short, broadly-suffused, white streak from base below middle; cilia brown. Hindwings with apex rounded, termen slightly rounded; fuscous-brown, cilia fuscous-brown. North Queensland: Evelyn Scrub near Herberton, in January and February; four specimens received from Mr. F. P. Dodd. 274 THE AUSTRALIAN SPECIES OF ONCOPERA. Explanation of Plate xrxaxii (by John R. Hyer). 1. Oncopera intricata Wk. 2. Oncopera mitocera Turn. 3. Oncopera brachyphylla, n. sp. 4. Oncopera epargyra, Nn. Sp. a—anus. A membraneous tube at the very caudal end of the genitalia usually opening just behind and on the meson of the tegumen. ap—anal process. A pair of processes from the caudal margin of the tegumen which in the Hepialidae often surround the anus. ss—scale sacks. A pair of eversible sacks covered with scales which occur on the conjunctiva between the 8th tergite and the tegumen. The Hepialidae is the only family in which I have found them to occur in this position. They are usually ventral and between the sternites. t—tegumen. The ninth tergite which is highly modified to form the dorsal hood of the genitalia and surround the anus. p—penis. A membraneous tube usually opening near the centre of the genitalia and in the Hepialidae often supported by a pair of processes from the meson of the tegumen. pr—mesal processes of the tegumen just referred to. ae—the aedeagus, i.e., the chitinized armature supporting the penis. In the Hepialidae it consists of a plate on the ventral wall of the penis; in the higher Lepidoptera it is usually a chitinized tube which completely surrounds the penis. v—valve. A pair of finger-like claspers articulated to a median plate, the “juxta’’, and serving as claspers to hold the female during copulation. In species such as brachphylla the valves are divided into a posterior finger-like lobe, ‘“‘cucullus’’, and an anterior acutely pointed lobe, the “‘sacculus’. These lobes take many forms in various species of Lepidoptera but the sacculus is usually the more heavily chitinized and the cucullus the more hairy of the two. j—juxta, just described above. ee—cucullus, just described above. se—sacculus, just described above. vm—vinculum. The ninth sternite usually modified to form a v- or u-shaped structure with a pair of lateral arms which extend dorsad to meet the tegumen. mp—median process of the vinculum. s8—eighth sternite. In the Hepialidae this forms a chitinized plate underlying the vinculum. Its shape varies in different species. ON THE TACHINID GENUS HUTHERA (DIPTERA), WITH DESCRIPTION OF NEW SPECIES FROM AUSTRALIA, AFRICA AND SOUTH AMERICA. By Pror. Mario Bezzi, Turin, Italy. (Communicated by Dr. E. W. Ferguson.) (Three Text-figures. ) [Read 26th August, 1925.] To the Dipterous genus Huthera belong the most striking and beautiful forms of the family Tachinidae, and, as is the case with many other genera of this interesting family, the species seem to be very rare, only isolated specimens being found; perhaps the flies are met with only occasionally because the bionomics are quite unknown. The genus has a wide distribution, being recorded from North America and from the Mediterranean subregion. I know it, however, from the whole Ethiopian region, as well as from Australia and from South America. It is therefore of world-wide distribution, ranging from about latitude 45° North to 36° South. The genus Huthera was founded by Loew (1866, pp. 46-47) with the single species tentatriz, based on a solitary specimen collected at New York by Baron Osten-Sacken. It is one of the eleven species of Tachinidae selected by Loew for description from one thousand species of North American Diptera; they are all striking forms (Loew, 1872, p. 122), but Huthera and Himantostoma (now Imitomyia) are the most striking of all. Osten-Sacken (1876, p. 154) has the genus Huthera near Scopolia, quoting the species from New York and Texas. The next record of the genus Huthera is to be found in an important paper by Prof. Mik (1889, pp. 129-134), in which the author has presented notes on the characters, as well as on the systematic position, together with the description of the new species, HL. mannii from Lesina, Dalmatia (one male, collected by Novak) and from Brussa, Asia Minor (one couple, collected by Mann). The paper is illustrated with two figures drawn by Prof. Brauer, and contains the information that the specimens from Brussa were already recognized by Osten-Sacken as belonging to Huthera. Prof. Mik is of the opinion that the genus belongs to the Phasiinae, sensu Rondani. Brauer and Bergenstamm (1889, p. 140) have the genus in the group Schineriidae, between Ocypteridae and Gymnosomatidae, with a figure (Plate x, fig. 278) of the head of the female mannii; subsequently (1891, p. 411 and p. 433; 1893, p. 110 and p. 152) they have repeated the characters, assuming them always from the species mannii, the true genotype tentatrix being quite unknown to them. Tyler Townsend (1892, p. 141 and p. 144) has placed Huthera with the Tachinidae, having excluded it from the preceding paper of 1891, dealing with Phasiidae, Ocypteridae, etc. J 276 ON THE TACHINID GENUS EUTHERA, Prof. Brauer (1893, p. 474 and p. 493) again has the genus near Schineria, giving mannii as type. Coquillett (1897, p. 37 and p. 120) has placed the genus near Demoticus, Tecording the species from Pennsylvania and Georgia. Williston (1896, p. 148) records only the name of the genus. i Prof. Strobl in his works on the Diptera from Bosnia and Dalmatia (1898, p. 443 and 1900, p. 597) quotes the species mannii from Island Lesina, but without having seen it. Coquillett (1902, p, 114) described a second North American species, EHuthera bicolor, from Texas, only from the female. Aldrich (1905, p. 469) records the genus near Tachina, with the two North American species, without adding new localities. I myself (1907, p. 566) have the genus near Selimeria, with the species mannii and the usual citations. Prof. Williston (1908, p. 374) has the genus near Acemyia, giving (p. 368, fig. 80 and 81) figures of head and of wing of tentatriz. Coquillett (1910, p. 543) gives tentatrix as genotype of Huthera, and Johnson (1910, p. 781) records the species tentatrix from New Jersey. Tyler Townsend (1912, p. 49) places the genus Imitomyia in the subfamily Phasiidae, tribe Hutherini. Villeneuve (1913, p. 43) has recorded H. mannii from Nyasaland, but this record probably refers to one of the Ethiopian species described below. Tyler Townsend (1916, p. 178) erected the new genus Hutheropsis for mannii, defining it (1916a, p. 322) and locating it in the fam. Rhadogynidae. Dr. Villeneuve (1924, p. 32) has considered the genus Huthera as forming a special tribe in the Phasiinae, with the genera Hermyia, Clara, Baraclara, etc.; while in the same year Prof. Stein (1924, p. 178) has the genus in the group Trixa, near Redtenbacheria. Finally Johnson (1925, p. 204) has the genus in the Tachininae, near Acemyia, recording the species tentatrix from various localities. From the above quoted works is to be seen how different are the opinions of authors about the systematic position of Huthera. In my collection I have placed it with Schineria near the Phaniinae, but Dr. Villeneuve (1924, p. 33) has placed the latter genus near HEchinomyia, excluding it from the Phaniinae. The seven known species, five of which are before me, can be distinguished | as follows:— : 1 (12) The small apical scutellar bristles diverging; female with orbital bristles, frontal bristles curved inwardly and all decussate; three pairs of dorsocentral and two pairs of acrostichal bristles before the suture; female with the frontal stripe considerably narrower than one of the parafrontalia, these latter very broad, shining black, not or slightly dusted; submarginal cell with hyaline spot. 2 (7) New World species; head entirely red, with quite bare parafacialia, with prominent mouth border and with rounded, swollen facial keel; front tarsi of female quite simple, coxae and femora reddish, second abdominal segment without strong macrochaetae in middle of hind border; wing-base and costal cells of a bright yellowish colour; calypters likewise yellowish and opaque (Huthera, s. str.). 3 (6) All the coxae reddish, tibiae and tarsi entirely black; terminal half of second costal cell and stigma blackish; third posterior cell quite hyaline; alula broadly yellowish at base; scutellum with two pairs of lateral bristles. 4 (5) Abdomen entirely shining black; the dark middle band of wings interrupted by broad hyaline streaks into the submarginal and the discoidal cell prCAD ICCC EecuCr eecHCerieci mca oere cuenta Eom oTanicic Ging GlmarGiaadre crtioraroictEO o.oo tentatrix Loew. 5 (4) Abdomen reddish-yellow with brownish dorsal spots; dark middle band of wings much broader and not interrupted by hyaline streaks .... bicolor Coquill. BY M. BEZZI. 277 6 (3) Coxae in greatest part black, tibiae and tarsi reddish; second costal cell and stigma entirely yellowish; third posterior cell blackened; alula entirely black; no hyaline streaks into the middle band of wings; scutellum with three pairs of lateral bristles, the middle one smaller .................. barbiellinit, n. sp. 7 (2) Old World species; head in greatest part black; female with front tarsi flattened ; coxae always black, second abdominal segment with strong macro- chaetae; wing base whitish hyaline and costal cells blackish, third posterior cell partly infuscated; alula entirely black, calypters whitish hyaline (Hutheropsis T. T.). 8 (9) Mouth border prominent; facial keel swollen and rounded in profile, parafacialia distinctly pilose in their whole length; frontal stripe dark reddish .. mannii Mik. 9 (8) Mouth border not at all prominent; facial keel sharp, straight in profile, low; frontal stripe deep black. 10 (11) Palpi red; coxae and femora black; parafacialia thinly pilose in the middle; scutellum with a pair of diverging apical bristles and with a pair of discal ones; bend of fourth vein angular, the apical crossvein straight and less CO) SPA ONG YETI Ss SPAN BSN eee kere al ot Sta PRES CSP ROR IS MM Sin: Sed AP oy a ERE AD Lr) Ae tuckeri, n. sp. 11 (10) Palpi black; coxae and femora red; parafacialia quite bare in middle, with only a few hairs above; scutellum with only the two pairs of lateral bristles, without apical and discal ones; bend of fourth vein more rounded, the apical CROSSVEINIMOTE DIGUES .Lhy cc eucichels c OGRE econ ie be ethers tas péringueyi, n. sp. 12 (1) Small apical bristles of scutellum decussate; female without orbital bristles; frontal bristles erected, not decussate; five presutural dorsocentral and three presutural acrostichal bristles, mouth border prominent, frontal stripe red, in the female much broader than one of the parafrontalia, these latter very narrow and densely dusted, not shining; face yellow in the middle, palpi red, wing base whitish hyaline, costal cells blackish; submarginal cell without hyaline spot; calypters white (Macreuthera, n. subgen) .......... skusei, n. sp. Genus EHuTHERA Loew. For generic characters, sexual differences, etc., see Mik, Brauer and Coquillett. It may be added: facial ridges with 3-4 thin, bristly hairs just above the vibrissae, hind coxae destitute of bristles, all the wing-veins bare, even at common stem of radius and at base of third vein. Subgenus EUTHERA, S. str. The species of this subgenus are well characterized by the rich wing-pattern, having yellow base of wings and yellow calypters; they are exclusively American. 1. EUTHERA TENTATRIX Loew. Text-fig. 1, A. This is the genotype known only from the United States (Massachussets, Pennsylvania, New Jersey, New York, Georgia and Texas). I have in my collec- tion one female from Tifton, Ga., October, 1896, and another female from Great Notch, N.Y., 30th October, 1906. Seems to be an aestivo-autumnal species, being not found before August. The male of the present species is not yet described; the possibility is not excluded that it may have an infuscate third posterior cell, as in barbiellinit. 2. HUTHERA BICOLOR Coquillett. Described from Texas, without precise locality, and not found subsequently. The wing pattern seems to be like that of the following new species in the shape of the middle dark band, but the third posterior cell is quite hyaline. 3. EUTHERA BARBIELLINII, n. sp. Text-fig. 1, B. Closely allied with the two preceding species, but distinct by the richer wing- pattern. 278 ON THE TACHINID GENUS EUTHERA, Type dg, a single, badly preserved specimen in the writer’s collection, caught near Sao Paulo, Brazil, by the Count A. A. Barbiellini, in whose honour this fine insect is named. 6. Length of body 6 mm.; of wing 5 mm. Head entirely reddish, blackened only towards the middle of the occiput. Frons narrowed towards the vertex and there only one-half as broad as at antennae; parafrontalia black, densely dusted with grey; middle stripe red, at vertex as broad as one of the parafrontalia, but at antennae twice as broad; frontal bristles black, rather numerous, but not long, never strong, directed inward and decussate, extending below to the end of first antennal joint. Face entirely shining, yellowish with a narrow transverse black stripe in the middle; parafacialia as broad as the third antennal joint and rather opaque, being clothed with whitish dust, peristomialia shining, unspotted, twice as broad as the parafacialia, with whitish hairs behind. Antennae linear, elongate, longer than the face, with the two basal joints reddish and the third black; second joint three times as long as the first, with black setulae, third joint one and a half times as long as the second, attenuated at end; arista basal, entirely reddish, only a little longer than the third antennal joint, considerably thickened at base. Proboscis short and thick, dirty yellowish, palpi club-shaped, Text-fig. 1. Wings of Huthera, enlarged. A. tentatrix Loew; B. barbiellinii, n. sp. reddish. Thorax and scutellum entirely black, but too badly preserved for descrip- tion; scutellum with three pairs of lateral bristles, the apical ones small and diverging, no discal. Calypters greatly developed, bare above, entirely of a bright yellowish colour. Halteres with yellowish stalk and blackish knob. Abdomen entirely shining black with black setulae and black bristles, these latter rather thin and short, present only at hind border of the segments, on the first and second segment only at sides; venter with all the sternites concealed; genitalia rounded, black, of middle size, well visible but not prominent. Coxae and trochanters black, only those of the front pair in part reddish, femora, tibiae and basal joints of the tarsi reddish, claws and pulvilli short, but distinctly, even if little, more developed than in the female of tentatriz. Wings with typical nervation, but the apical crossvein straighter, the terminal stalk of the first posterior cell being therefore longer. Wing base bright yellowish to the basal and anal crossveins, but along the fore border the yellow colour is extended into BY M. BEZZI. 279 the costal and subcostal cells to the end of stigma, which is likewise yellow; along the first basal cell the yellow colour is extended to a little before the small crossvein. The broad hyaline stripes into the base of submarginal and of dis- coidal cell are whitish. The middle dark band is broad and entire, extending from the costa to the fifth vein; the whitish-hyaline band beyond it is as broad as the dark preapical band; the marginal cell is entirely black to the end; the submarginal cell is narrowly infuscated at the extreme end. The third posterior cell is almost entirely dark brown, with a whitish hyaline streak along the fifth vein. Alula entirely blackish to the base. Subgenus HuTHERoOpsSIS T.T. The characters on which this subgenus was founded are not of sufficient value to be considered as generic, inasmuch as the Ethiopian species here described differ from the type-species in degree of ciliation of the parafacialia. The latter species has no prominent mouth border, while in mannii the mouth border is as prominent as in the American species. All the species show the infuscation of the third posterior cell, as in barbiellinii, but less intensive and equally present in both sexes. 4. HUTHERA MANNII Mik. 4, 9. See descriptions and figures of Mik, Brauer and Stein. Known only from Dalmatia and Asia Minor; but Dr. Villeneuve tells me that he knows the species from British East Africa and from Formosa. 5. HUTHERA TUCKERI, n.sp. ¢. Text-fig. 2, A. Distinct from the preceding on account of its mouth border not being prominent and of the lower facial keel. Type d, a single, badly preserved specimen in the South African Museum, Cape Town, from Transvaal, Koopmuiden, 30th October, 1918, collected by Rev. W. HE. Tucker and named in honour of its discoverer. 6. Length of body 7 mm.; of wing 5.5 mm. Head entirely black, shining only on lower part of occiput and on peristomialia. Frons rather narrow at vertex, being there twice as broad as the small ocellar triangle, gradually broadened forwardly, being at antennae three times broader than at vertex; frontal stripe deep black, broader than one of the parafrontalia, these latter white dusted, very narrow on basal half, broadened on apical half, and at antennae about as broad as the stripe; bristles black, decussate; no orbitals, ocellar very thin. Face entirely black, with slight whitish dust, not or very little shining; middle keel sharp, straight in profile, mouth border not at all prominent, parafacialia white dusted, with scattered thin hairs, as broad as the anterior part of the parafrontalia. Antennae wanting. Palpi rather thin, entirely yellowish, proboscis dirty blackish. Thorax entirely black; the baek is rather shining with whitish dust, but the pattern is not distinguishable; scutellum like back. All the bristles black; three presutural dorsocentral and two acrostichal, scutellum with two strong lateral, one discal before end, and seemingly with a pair of very thin diverging apical. Calypters bare, whitish hyaline, with a narrow white border; halteres with whitish stalk and with blackish knob. Abdomen cylindrical, thick, entirely black, rather shining, with slight whitish dust, which is denser at sides below; setulae and bristles black, the latter at sides of all segments, and in middle of hind border from the second segment to 280 ON THE TACHINID GENUS EUTHERA, the end; genitalia rounded, shining black, not prominent but open; sternites concealed. All the coxae and the legs of first pair entirely black; other legs wanting. Wings normal, bend of fourth vein angular but obtuse, the apical crossvein straight and less oblique, parallel with the hind crossvein, the stalk of the first posterior cell being about one-half the length of the apical crossvein. Wing base whitish-hyaline. Costal, subcostal and marginal cell entirely black; Text-fig. 2. Wings of Huthera, enlarged. A. tuckeri, n.sp.; B. péringueyi, n. sp. first dark band broad and entire, extended equally from the second to the fifth vein, including the two crossveins and going.a little beyond them. Preapical band a little narrower than the whitish-hyaline band before it, end of sub- marginal cell entirely filled with brown; infuscation of third posterior cell less intensive, extending across the terminal half of this cell and of the axillary cell, but evanescent in middle at wing border. Alula entirely black. 6. HUTHERA PERINGUEYI, n.sp. 9. Text-fig. 2, B. Closely allied with the preceding species, but distinct by the characters of the above key; it seems impossible that it may be the female of preceding species. Type 9, a solitary, well preserved specimen in my collection from Chabra, Congo, dedicated to the memory of the late Doctor L. Péringuey, whose contribu- tions to South African Entomology are universally esteemed. ©. Length of body 7 mm.; of wing 5.5 mm. Head black, but the sides of the ocellar plate, the facial ridges and the antennal grooves are distinctly yellowish. Occiput shining black, considerably swollen, white dusted below at sides. Frons about as broad as one eye; middle stripe deep black, narrower than one of the parafrontalia, ocellar plate and parafrontalia very shining, the latter white dusted -in front. Frontal bristles decussate; two orbital, the first pair much smaller than the second; ocellar bristles well developed. Parafacialia narrowed below, white dusted, with only a few hairs on the upper part. Face yellowish above, remainder black, but clothed with slight whitish dust, middle keel straight and low, mouth border not at all prominent, peristomialia shining black. . Antennae longer than the face, first joint reddish; second joint reddish-brown, twice as long as the first, with black hairs; third joint black, linear, distinctly broadened and BY M. BEZ2ZI. 281 upcecurved at end, two and a half times as long as the second joint; arista as long as the third antennal joint, entirely reddish, bare, thickened at base. Palpi deep black, rather short and clavate; proboscis short, dirty reddish. Thorax black, shining, the whitish dust forming three broad stripes on the presutural part of back, pleurae white dusted. Scutellum very characteristic, triangular, flat, obtuse at end, black with slight whitish dust, there are only two pairs of lateral bristles, no discal and no apical, only a few short setulae on dise. Calypters whitish, opaque, halteres with blackish knob. Abdomen as in preceding species, but distinctly, even if little, compressed; bristles more developed, chiefly the middle pair at hind border of second segment; sides of third segment somewhat reddish. Coxae blackish, all the femora reddish, those of hind pair with blackish stripes below; tibiae and tarsi entirely deep black, front tarsi distinctly thickened. Bend of fourth vein rounded, the apical crossvein rather long and more oblique than the hind crossvein, the terminal stalk of first posterior cell much shorter than the apical crossvein. Pattern as in the preceding species, but with the preapical dark band broader and more curved. Subgenus MACREUTHERA, NOV. This new subgenus is characterized by the different form of head, which is narrower and higher than in the other species; by the very different frons of the female, lacking the orbital bristles; and by the decussate apical scutellar bristles. The mouth border is as prominent as in mannii, the female front tarsi are not distinctly thickened. The type species is more robust and of greater size than the others; the abdominal macrochaetae are strong; in wing pattern it agrees with the Old World species, while in the quite bare parafacialia it agrees with the American ones. Type: The following new species, EH. skusei. EUTHERA SKUSEI, n.sp. 9. Text-fig. 3. A robust species with reddish frontal stripe and with yellowish black-margined face. Type 9, a single specimen in the Health Department (New South Wales). Collection from Hidsvold, December, 1922. Dedicated to the memory of the great Australian Dipterologist, Frederick A. A. Skuse. g. Length of body 10 mm.; of antennae 2.8 mm.; of wing 8 mm. Head in front view as broad as high. Occiput black, white dusted at sides and below, and there with a reddish stripe along the hind border of the eyes; just above the neck there is a deep black stripe, extending to the reddish postvertical spot and narrowly margined with white dust; a row of black postocular bristles, and black hairs below. Frons broad and gradually widening forward; middle stripe parallel sided, bare, much broader than one of the parafrontalia, entirely reddish, except for a black spot on the ocelli, parafrontalia narrow, near the vertex half as broad as the middle stripe, broadened in front, black, white dusted. Frontal bristles short and thin, erect, not decussate, ending at base of antennae; orbital bristles quite absent; one pair of ocellar; inner vertical long and strong. In profile the frons is much more prominent than in the allied species; eyes narrow, with the vertical diameter twice as long as the horizontal one. Face long and narrower than in the allied species, with the middle keel rounded and very prominent, and with the antennal grooves almost parallel; it is yellowish in middle, margined with black at sides and below; parafacialia quite bare, narrow, 282 ON THE TACHINID GENUS EUTHERA, not broader than the third antennal joint, reddish, but densely clothed with whitish dust; mouth border very prominent and black. Peristomialia much broader than in the other species, as broad as one-third of vertical diameter of eye, dark reddish, with a blackish triangular spot just below the eye. Antennae long, porrect, longer than the face; first joint short, dark reddish, second joint more than twice as long as the first, dark reddish, with black setulae and with a long erect bristle above near the base; third joint black, three times as long as the second, linear in the basal half, dilated in the terminal half and ending wee es as ai ee Text-fig. 3. Wings of Huthera skusei, n.sp., enlarged (same scale as preceding figures). obtuse; arista basal, not longer than the third antennal joint, quite bare, thickened and yellowish at base, whitish and very thin in the rest. Palpi clavate, dark reddish, black haired; proboscis short, dirty brownish. Thorax entirely black, rather shining; on the back the whitish dust forming two broad stripes separated by a black one, continued in front of suture by two narrow lines, the sides are broadly dusted forming three broad and two narrow stripes on whole back; pleurae white dusted. All hairs and bristles black, dorsocentral numerous, but rather small, three to four behind and five in front of suture; acrostical two behind and three in front of suture; two sternopleural, four to five hypopleural, a tuft on upper border of pteropleura. Scutellum like back, rather convex, with two pairs of long lateral bristles, one pair discoidal, apical very thin and small, decussate. Calypters greatly developed, bare, whitish and with a white border; halteres with black knob. Abdomen entirely shining black, with whitish dust; hairs black; second segment with two strong macrochaetae in middle of hind border, third segment with four, fourth with a complete row. Sternites concealed. Coxae and legs entirely black; front tarsi not distinctly thickened; claws and pulvilli short. Wings normal, bend of fourth vein V-shaped, the apical crossvein concave; stalk of first posterior cell one-third the length of the apical crossvein; small crossvein a little beyond middle of discal cell; hind crossvein a little S-shaped. Base of wing whitish-hyaline to the small crossvein, and even beyond this vein into the base of submarginal cell; costal, subcostal and marginal cells entirely black; submarginal cell entirely black in its terminal half, quite destitute of the hyaline spot of all the other species. The broad middle band filling up the whole space between the two crossveins with irregular sides, ending below at fifth vein. The preapical band is formed only by a black margin of the apical crossvein, prolonged only a little beyond the band and there ending rounded. The infuscation of the third posterior cell is less intensive, slightly developed. Alula entirely black. BY M. BEZZI. 283. Literature Cited. AwpricH, J. M., 1905.—A Catalogue of North American Diptera. Washington. Brzzi, M., and Stein, P., 1907—Katalog der Palaarktischen Dipteren, Band iii- Budapest. BRAUER, Fr., 1893.—Vorarbeiten zu einer Monographie der Muscaria schizometopa. (exclusive Anthomyidae). Verh. Zool-bot. ges. Wien, xliii, 447-525. BRAUER, Fr., and BERGENSTAMM, J. von, 1889.—Die Zweifltigler des K. Museums zu Wien. iv. Vorarbeiten zu einer Monographie der Muscaria schizometopa (exclusive Anthomyidae). Denkschr. Math.-naturwiss. Cl. Akad. Wiss. Wien, lvi, 69-180. , 1891.—Id. v, Pars. ii, Op. cit., lviii, 305-446. , 1893.—Id. vi, Pars. iii, Op. cit., Ix, 89-240. CoquILLeTT, D. W., 1897.—Revision of the Tachinidae of America North of Mexico. Washington. : , 1902.—New Diptera from North America. Proc. U.S. Nat. Mus. xxv, 83-126. , 1910.—The type-species of the North American genera of Diptera. Op cit., xxxvii, 499-647. JoHNsoN, Ch. W., 1910.—In J. B. Smith, Insects of New Jersey, Diptera. Ann. Rep. N.J. St. Mus., 1909, 703-814. , 1925.—Fauna of New England. 15. List of the Diptera or two-winged flies. Occ. Pap. Boston Soc. Nat. Hist., vii. Lorw, H., 1866.—Diptera Americae septentrionalis indigens. Centuria septima. Berl. Entom. Zeitschr., x, 1-54. , 1872.—Id., Centuria decima. Op. cit., xvi, 49-124. Mix, J., 1889.—Ueber die Dipterengattung Huthera Lw. Wien. Entom. Zeit., viii, 129-134. : OSTEN-SACKEN, C. R., 1878.—Catalogue of the described Diptera of North America. Second Edition. Washington. STEIN, P., 1924.—Die verbreitesten Tachiniden Mitteleuropas nach ihren Gattungen und Arten. Archiv. f. Naturgesch., 90, 1-271. Srrosx, G., 1898.—Dipterous-Fauna of Bosnia, Hercegovina and Dalmatia (serbish) Serajewo, x, 135 pp. (sep.). , 1900.—Dipterenfauna von Bosnien, Hercegovina und Dalmation. Wiss. Mittheil aus. Bosn. u. d. Herceg., vii, 552-670. TOWNSEND, Ch. H. T., 1892.—The North American Genera of Calyptrata Muscidae. Paper 11. Trans. Am. Ent. Soc., xix, 133-144. 1912—A readjustment of Muscoid Names. Proc. Ent. Soc. Wash., xiv, 45-53. , 1916.—New Muscoid Genera (Dip.). Hnt. News, xxvii, 178. , 1916a.—New Genera and Species of Muscoid Flies. Proc. U.S. Nat. Mus., li, 299-323. VILLENEUVE, J., 1913—Myodaires supérieurs de 1l’Afrique tropicale (Ire liste). Revue Zool. Afric., iii, 24-26. , 1924—Contribution a la classification des “Tachinidae” paléarctiques. Ann. Sci. Nat. Zool. (10), vii, 5-39. Wiuiston, S. W., 1896.—Manual of the Families and genera of North American Diptera. Second Edition, New Haven. , 1908.—Id. Third Edition, illustrated. New Haven. THE COMBOYNE PLATHAU. Its GENERAL CONFORMATION AND FLORA. By E. C. CuisHotm, M.B., Ch.M. (One Text-figure. ) e [Read 26th August, 1925.] The Comboyne Plateau is situated in the County of Macquarie, about 170 miles north and a little east of Sydney, though 250 miles by rail and road, and about 20 miles by air line from the coast. Its average height above sea-level is 2,200 feet, though its highest point, Mount Gibraltar, rises to about 3,200 feet. There are two other mountains or high elevations, Mount Bulli, about 2,700 feet, and The Kopje, about 2,600 feet above sea-level. These three points occupy the south-east quadrant, and are situated near the edge, which falls away very abruptly in the case of Mount Gibraltar, less so in the other two cases. The rest of the Plateau is hilly, well intersected with perennial streams forming three watersheds. That on the east, taking its origin at the eastern base of Mount Bulli, is the watershed of the Camden Haven River, which, running through Kendall, reaches the sea at Laurieton; that on the south and south-west is the watershed of the Manning, which, lower down, runs through Wingham and Taree; that on the north, north-west and west forms the watershed of the Hastings, running out at Port Macquarie. The area of the Plateau is estimated at about 70 square miles, the larger portion of it having been covered with Softwood Brush or Rain Forest. The formation is basaltic, the soil being of a rich red colour. This red soil reaches in places to a considerable depth, getting shallower towards the edges, especially to the north and west, where the formation is of sandstone or sedimentary deposit, and here, as one would expect, the hardwood timber, especially Hucalyptus, with a sprinkling of Casuarina torulosa, predominates. At the junction of the two formations it is usual to find Tristania conferta (the Brush Box). At isolated spots on the area, especially elevated ones, the sandstone—in using the term “sandstone” I am not confining myself to pure sandstone, of which there is very little, though it does occur in isolated patches—or sedimentary formation out- crops, and here again the hardwoods prevail. At the extreme summit of Mount Bulli one comes across Hucalyptus quadrangulata, this being the only place on the Comboyne where it is found, and the farthest north that it has been recorded, the Hunter River being its northern previously recorded limit (see Maiden, Critical Revision, Genus Hucalyptus, ili, p. 76). The country on top is now mostly cleared for dairying purposes, so that it is difficult to imagine this plateau in its virgin state. There is now only a comparatively small area of virgin brush forest left in the centre, though the Government has set apart 200 acres at the western edge, on Mumford’s Creek, as a reserve, which area includes two fine waterfalls, the Allen Falls, not very high, but over which a large body of water is pre- BY E. C. CHISHOLM. 285 cipitated, and the Rawson Falls, the finest on the plateau, with a vertical drop in the first face of 270 feet, comprising a large body of water. This brush forest reserve has remained practically in its virgin beauty, except that the Red Cedar was removed years ago, and here there is a splendid assortment of softwoods with N Sy ee Sa Ann ) woe “Fenoos, SF % 7, ois Vn, \ by VS xe Can tet ‘Aye We, See te ee’ PES ranse Rough Map of the Comboyne Plateau. E. C. Chisholm, del. Seale about 3 inch to 1 mile. Elkhorns, Bird’s Nest ferns and a host of climbing ferns, tree-ferns and palms. Hardwoods in the form of Brush Box, Tristania conferta, and Tallow Wood, Eucalyptus microcorys also are found here, with a sprinkling of Hucalyptus saligna (the Sydney Blue Gum). The average annual rainfall recorded for the Plateau for the last 16 years is about 60 inches; the maximum in this period was 98.87 inches in 1921, and the minimum 42 inches in 1915. Last year (1924) 57 inches were recorded. The history of the early settlement is interesting, but as this aspect is not the purpose of this paper, I will only allude to it briefly in passing. The earliest Pioneers were two brothers O’Shaunessey, who arrived on the Plateau about 30 years ago. After remaining here for a time one of them died and the other, failing to get sympathetic encouragement from the Government, abandoned his selection and left the Plateau, and for some years no one ventured to take up land. About 25 years ago the present pioneer settlers arrived, mostly from the 286 THE COMBOYNE PLATEAU, South Coast of New South Wales—Robertson, Kangaroo Valley and neighbouring places—in several families, who, with Government encouragement, began clearing for permanent settlement. When the brush forests were cleared and planted with Paspalum the growth was phenomenal, due to the rich humus of decaying leaves having enriched the soil from time immemorial to a depth of several feet, having been replenished year by year while the forest was under natural conditions, but, once cleared, each year of grass cultivation gradually depleted this rich humus, which was not being replaced, until at the present time the soil has lost to a great extent its former value and requires the help of fertilizers to produce good crops or grass. Below is an analysis of the soil of a part that had 20 years or so ago been part of a brush forest, but since that time has been under Paspalum or other grasses. There is no doubt that the Earthworm plays a large part in fertilizing the soil by bringing up earth from the deeper layers and depositing it at the surface, so replenishing the top soil. But for this factor the fertility of the soil of this plateau would be much less than it is, as the Harthworm here is very numerous. Cockchafer larvae, which are abundant, probably do a little in this direction also. The following analysis of a sample of Comboyne soil was kindly carried out for me by Mr. Ronald White, B.Sc.:— Specific Gravity (HzO = 1) 4.7. (Hther used to determine Sp. Gr. and then calculated to water). Very slightly soluble in hot or cold water. Chocolate brown in colour. Probably of voleanic origin, igneous rock broken down by weathering. Composition. Iron as ferric oxide and carbonate .. .. 4.32% Lime, Calcium Carbonate and Calcium Sulphate: CAnnvarite 0.78% Maenesium™= asichloride s.) «ce cioy acid mers ee ten ele OSc%6 Potassiumieas: chloride 28 mise | ue ace | he en eerie OO Sodium chloride and ysulphates) sce) en foe eee oeOSSS Organic matter (Humus) .. .. . 2.11% Clayey and siliceous matter, with Sane of Alaninous earths and phosphates A Oe Gir ects (Gow. Weeou Uekas” clase baie 100.00% Examined bacteriologically, cultures on Agar and Gelatine gave colonies containing Saccharomyces (yeasts), Streptococcus aureus and an unidentified anaerobic bacillus. There was no appearance of the presence of any organism capable of nitrogen-fixation, to which fact any soil deficiencies might be due. The Flora of the Plateau. The Filicales are well represented. Of the Tree-ferns both Dicksonia and Alsophila are numerous. Asplenium nidus (Bird’s Nest Fern) and Platycerium bifurcatum, especially the former, are prominent features of the Brush Forest, almost covering the trunks and branches of some of the forest trees. One of the favourite hosts of these is Tarrietia actinophylla (Stave Wood), trees of which are also generally covered with many climbing ferns and orchids. Adiantum aethiopicum (the common Maiden-hair) is quite rare, though other species of the genus are fairly plentiful, among them A. formosum, which seems to prefer rocky situations. Ground and climbing ferns are numerous. The Pines are only represented on the Plateau by one species, Callitris Macleayana (Stringybark Pine or Port Macquarie Pine) which grows to fine proportions. Araucaria Cunningham (Moreton Bay Pine) is found between the Plateau and Kendall to the east. BY E. C. CHISHOLM. 287 There are only two Palms, Archontophoenix Cunninghamiana (Bangalow) and LIinospadiz monostachyus (Walking-stick Palm). Livistona australis (Cabbage Paim) is entirely absent, curiously enough, although represented well on the coast and also on the next plateau to the south-west, viz., the Bulga, about 10 miles away. All the Araceae grow here, the most interesting species of which is Typhonium Brownii, an Arum, whose flower is large and striking, growing close to the ground and of a liver colour. It appears to favour the western end of the plateau. The Liliaceae and Iridaceae are poorly represented, while the Orchidaceae are mostly of the genus Dendrobium, of which D. speciosum (Rock Lily) and others are found as epiphytes on the trunks and branches of numerous brush trees. I have never seen the former growing on rocks here, though smaller forms, e.g., D. Kingianum, are commonly found in rocky situations. The remaining Monocotyledons, viz., Grasses, etc., I am not including amongst the flora recorded. The only Casuarina is C. torulosa (Forest Oak), which grows on sandstone Tidges, or sedimentary rock, mostly at the edges of the plateau. The family Fagaceae is represented by Fagus Moorei (Negrohead Beech), which is extremely rare, although many trees were undoubtedly destroyed during clearing. This appears to be near its extreme limit south, though it has lately been recorded from Barrington Tops. Of the genus Ficus I have recorded three species; F. stephanocarpa is a rare form here; the other two, F. rubiginosa and F. macrophylla, Port Jackson and Moreton Bay Figs respectively, are fairly common in the brushes and grow to a good height. The Proteaceae are very poorly represented. Orites excelsa is called here “Silky Oak,’ from the resemblance of the wood to that of Grevillea robusta, which does not occur so far south, nor, so far as I have been able to determine, does any species of the genus. This timber, O. excelsa, which is plentiful, is in fair demand for milling and used for much the same purposes as Grevillea robusta, being much in demand for cabinet work. Stenocarpus salignus (Red Silky Oak or Beef Wood) is uncommon on the plateau, but is in great demand for cabinet work, the timber being a rich red, beautifully figured. It is used about the coast for gun stocks. No Banksias occur here. Hakea, Persoonia and Lomatia occur, but in one or two species only. Lambertia is absent. Santalaceae are represented by EHxocarpus cupressiformis (Native Cherry) growing on sandstone or sedimentary rock, though I know of one very fine tree growing in the midst of brush timber and this is the largest tree of its kind I have seen. I have so far only recorded two species of the family Loranthaceae, viz., Phrygilanthus celastroides growing on Cryptocarya australis (Moreton Bay Laurel) and Daphnandra tenuipes (Yellow Wood), and Loranthus dictyophlebus growing on Cryptocarya australis and Doryphora sassafras. Codonocarpus attenuatus, of the family Phytolaccaceae, is a handsome tree, bearing bell-shaped fruit in clusters at the top of the tree and with a light delicate foliage. Clematis, with two species, is fairly common, but Ranunculus lappaceus (Buttercup) is decidedly uncommon. Legnephora Moorei, a climber with large, ivy-shaped leaves, is fairly numerous, festooning the trees. Its fruit is a black 288 THE COMBOYNE PLATEAU, berry forming clusters resembling small bunches of grapes, and very like them in outward appearance. Eupomatia laurina occurs in the brushes. The foliage is intensely shiny and of a bright green. Of the Monimiaceae, Wilkiea macrophylla is one of the “Native Plums,” bearing black oval fruit in clusters. Daphnandra tenwipes and Doryphora sassafras are plentiful everywhere. The Lauraceae are fairly well represented, and among them are some valuable timbers and handsome trees, Cinnamomum Oliveri and C. virens, “Black Sassafras” or “Cinnamon Laurels.” The former especially is worthy of cultivation, having a fine canopy and good shade, with a nice appearance. Litsea dealbata is one of the so-called ‘““‘Wild Plums,” bearing oval black fruit. It is a shapely shrub with large leaves with well marked veins and silvery underside. This would be worth cullivating. Litsea reticulata (Scaly Beech) is used a good deal for cabinet work, as it is easily worked and the timber has a pretty figure, and for building pur- poses. COryptocarya glaucescens (Nutmeg Laurel) is very abundant and Crypto- carya australis (Moreton Bay Laurel) is also plentiful and marked by its beautiful symmetry and thick foliage; both would make fine ornamental shrubs. Passing to the Saxifragaceae, Quintinia Siebert (Opossum Tree) grows in the brushes, especially mear creeks, and attains fair proportions. Anopteris Macleayanus has a superficial resemblance to the Waratah. The Pittosporaceae include Pittosporum undulatum, which is not plentiful and mostly found in the neighbourhood of creeks. Hymenosporum flavum is a plant well worthy of cultivation, bearing large yellow jasmine-shaped flowers with a fine perfume and large leaves, though of rather straggling habit. The family Cunoniaceae is represented by several valuable species. Schizo- meria ovata (White Ash) is a handsome tree with large leaf and bearing fruit to outward appearance like that of the Lilly Pilly. It is very little, if at all, used here as a timber. Ceratopetalum apetalum (Coach Wood), a very valuable timber, used a good deal in coach work and railway carriage building, grows plentifully in the brushes and to fine proportions. C. gummiferwm (Christmas Bush) does not occur on the Plateau, though between here and the coast, on the road to Kendall, I have seen fine trees a foot in diameter at the butt and very tall. They occupied a narrow belt, but for some unknown reason the large trees seemed to be dying. Geissois Benthami (Leather Jacket) is used a good deal for milling purposes in much the same capacity as Coach Wood, though inferior to it. A peculiarity of this timber is that in certain patches the wood is of stony hardness and gritty, so that the expert axemen avoid it if possible, as they complain that it takes the edge off the axe more than any other timber. On a superficial inspec- tion the timber resembles Coach Wood rather closely. Ackama Muelleri (Brush Corkwood) is plentiful in the brushes. Callicoma serratifolia is common along the course of creeks, but grows to no great size. The Rosaceae are well represented by four species of Rubus, R. parvifolius alone growing on sandstone, the other three species in the brushes. The Leguminosae, including the Acacias, are poorly represented, only seven of the latter being found, of which A. juniperina, A. floribun@a and A. decurrens var. mollis are decidedly rare. The prevailing species are A. melanoxylon, A. intertezta and A. binervata. The two former, with A. elongata, do well on the basalt, A. binervata on sandstone or sedimentary rock. I have only come across one Cassia, viz., C. sophera. The only representative of Daviesia found here is D. corymbosa var. arborea, a very handsome tree, especially when in bloom, BY E. C. CHISHOLM. 289 flowering in great profusion and growing on the sandstone at the edge of the Plateau, especially on the northern side. Gastrolobium Boormani is fairly plen- tiful on the southern side. Indigofera australis is not plentiful and found mostly just below the edge. It is reputed poisonous to stock. Kennedya is represented by three species, only found on sandstone. There is a total absence of Hardenbergia. The family Rutaceae is fairly represented. Boronia and Eriostemon are not found. Bosistoa euodiformis (Bone Wood), the vernacular name from the very light colour of the timber, is used a good deal in cabinet and ornamental work. It grows to a large-sized tree and is an inhabitant of the brushes. Geijera salicifolia is a handsome tree with dark green leaves and would make a fine ornamental tree under cultivation. Zieria Smithii is present as a small bush. Acronychia laevis is a very common species, bearing a profusion of white berries, superficially resembling the fruit of the Lilly Pilly, which are greatly sought after by the Top Knot fruit pigeon, Lopholaimus antarcticus. A. Baueri is not so plentiful. The family Meliaceae is we]! represented, some of the species being very valuable for their timber. Cedrela australis (Red Cedar) is the most valuable of all, and has been cut out here for many years, with the exception of a few trees in inaccessible places. A few seedlings and small trees are still to be seen at the western end of the Plateau. Melia Azedarach (White Cedar) is decidedly rare; for a long time after my arrival here I was told there was no White Cedar growing on top, but after careful searching I located three or four trees. It seems to have no particular value as a timber tree. Dysoxylum Fraseranum (Rosewood) is a valuable timber, used in the mills for building purposes, mostly for lining walls and ceilings. It has a pretty figure and for this reason is used in ornamental work. It is rich red in colour and has a very pleasant prefume. D. rufum is plentiful, but of no value as a timber. Synoum glandulosum (Pencil Cedar) grows plentifully, but is more of a shrub than a tree, and as such would make a handsome ornamental tree under cultivation on account of its symmetrical shape and pretty foliage. It does not seem to have any particular uses as a timber. All of this family grow on the basalt, and are inhabitants of the brush forests. Comesperma is the only genus of the family Polygalaceae found, and it grows on sandstone country at the edge. Of the family Euphorbiaceae, Breynia oblongifolia is a handsome shrub bearing small fruit hanging under the leaves. Baloghia lucida (Brush Bloodwood) grows to a fair sized tree, and is chiefly interesting from the amount of reddish sap that runs out when an incision is made into the trunk. Homalanthus popu- lifolius (Bleeding Heart Tree), the popular name being due to its heart-shaped leaves which at one stage assume a bright-red hue, is a handsome tree with large Izaves and should be worth cultivating. It is fairly plentiful. The family Celastraceae is represented by Celastrus australis, a climbing vine with bright green shining leaves, which when in fruit is covered with small orange-coloured berries, and Denhamia pittosporoides, a tree resembling EHlaeéo- carpus reticulatus to some extent, though its fruit resembles that of Pittosporum. Of the Sapindaceae, Sarcopteryz stipiteta is chiefly remarkable for its bright red berries, which arrest the eye. Dodonaea triquetra grows on the sandstone or sedimentary rock. Akania Hillii is the only representative of the Akanaceae. This has sharp thorny edges to the long narrow stiff leaf and bears a red berry. 290 THE COMBOYNE PLATEAU, Of the Rhamnaceae, Emmenospermum alphitonioides is a fine shapely tree with shiny green leaves and bearing a small white flower and small orange-coloured berries. It would make a good ornamental tree. The family Vitaceae is represented by Vitis hypoglauca, which is very plen- tiful, festooning even the highest trees in the brushes, and V. Baudiniana, much less plentiful and occurring mostly at the southern and western edges of the Plateau. This does not seem to grow to the same proportions as V. hypoglauca, the trunks of whose vines attain a diameter of 7 or 8 inches, which, when cut into, exude, in great quantities, a watery juice made use of as a drink by bushmen when thirsty. The family HElaeocarpaceae is represented by Hlaeocarpus and Sloanea. Of the former genus I have only found #. reticulatus and it is not very plentiful, seeming to prefer the edges of brushes at the junction of basalt and sandstone or sedimentary strata. Sloanea Woollsii (Yellow Carrabeen) is a brushwood tree of some importance, used a good deal at the mill. One use to which it is put is the manufacture of broom handles. It is a tall tree with a curious buttress formation of the trunk as it rises from the ground, and generally triangular. The family Malvaceae has two genera represented in Sida and Hibiscus. H. heterophyllus, a tall shrub, bears a large conspicuous white flower with a purple centre having a superficial resemblance to a Magnolia, and with a prickly stem, and is found only on the southern slopes of the mountain side. The family Sterculiaceae is represented by three genera and four species. Brachychiton acerifolius (Flame Tree) is one of the glories of the brushes when in flower, which only happens once in three or four years, and at this time it sheds its leaves and the whole tree becomes one mass of bright red cup-shaped blossoms, forming a beautiful contrast to the dark green of the brush trees in the background. Brachychiton populneus (Kurrajong) is not seen on top, and I only know of one tree on the western side some distance down on junction strata. Tarrietia actinophylla (Stave Wood) is used for milling purposes. It is but- tressed as it rises from the ground and is the host of numerous ferns, orchids and mosses, and is usually one of the most heavily laden of all. Commerconia Fraseri, called here “Kurrajong,”’ has large angular leaves, somewhat resembling those of the Plane Tree, and medium-sized white flowers. It prefers the edge of the basalt. There are two species of Hibbertia, H. volubilis, which is a twining and trailing vine with a large handsome yellow flower like a dog-rose, and H. dentata, like a smaller edition of the former with smaller flowers and also of climbing habit. These, especially the former, are numerous, though the first-named seems to prefer the basalt, while H. dentata grows more on the edge. These two are the only representatives of the Dilleniaceae. Of the family Violaceae, the only species of Viola I have found is V. hederacea, and that mostly in moist situations, especially near streams. One species only of Flacourtiaceae, Streptothamnus Beckleri, has been observed. It is a vine with ivy-shaped leaves and the flower in bud superficially resembles the Fuchsia, hanging on a long stalk. I have only met with one species of Passiflora, viz., P. alba, with a white flower; I have cultivated this and it has made a handsome ornamental vine, bearing flowers and fruit freely. It is an introduced species, and is poisonous to stock. Pimelea ligustrina is the only representative of the Thymeleaceae, and is found growing on sandstone formation. BY E. C. CHISHOLM. 291 The family Myrtaceae is mostly represented by Eucalyptus on the sandstone or sedimentary rock, Hugenia and Tristania on the basalt, or at the junction of the two. Myrtus Beckleri, a tree of small growth or shrub, would do well in a shrubbery, as it is a shapely species mostly found on basalt. Hugenia Smithii (Lilly Pilly), fairly plentiful on basalt, bears pinkish-white fruit in profusion; it is a tree well suited for ornamental purposes on account of its thick foliage and symmetry. Hugenia corynantha is called here “White Cherry,” on account of its light-coloured wood; the fruit is pear-shaped, the smaller end distal and of a red colour and a fair size. Hugenia cyanocarpa, a Lilly Pilly with purplish fruit, resembles H. Smithii. Syncarpia laurifolia (Turpentine) only grows in a small area on the southern side ef the Plateau, at the edge of a brush forest; it is used principally as a timber for wharf piles, being one of the few timbers which the Teredo will not attack. Backhousia myrtifolia grows along courses of creeks, often at the edge of the basalt. Tristania conferta (Brush Box) is a fine tree, growing to high dimensions and greatly valued as a hardwood, a large quantity passing through the mills. Its favourite situation seems to be at the junction of basalt and other formation at the edge of brush forests. As an ornamental tree it is well ’known about the suburbs of Sydney. Tristania lawrina, a Water Gum, called erroneously here by many “Flooded Gum’, is generally found in damp situations, especially along the courses of creeks, and the nearer it is to water the larger the tree grows, and the redder the timber becomes. It is valued for many purposes, one of which is the making of axe handles, but to get the most benefit from it by making the wood tougher, it is usual to let the timber lie under water for a week or so before fashioning it. J. neriifolia I have failed to find here, though I have often searched for it. Most of the species of Hucalyptus grow on the sandstone or sedimentary rock at the fringe of the Plateau, though several species grow in the brushes on the basalt, but probably in these cases the depth of the basalt is not great. They are never very numerous in the brushes, and this rather points to the fact that although many seeds must be shed, only a comparatively few survive; it is quite uncommon to find seedlings under trees. Along the courses of creeks, where one would expect the basalt capping to be thinner, there are magnificent specimens of EH. saligna. I have recorded thirteen species for the Plateau; some are rare, and two are represented by single specimens. Around the base of the Plateau, within a radius of ten miles, I have recorded eight more. These occur on the lower lands approximating sea-level. Many of these are species found in the Picton district, which the surrounding country somewhat resembles. #H. Andrewsi, called here ‘“Messmate” and “White Top,’ is a species with a useful timber, a large quantity of which passes through the mills and is used for much the same purposes as Tallow Wood (E. microcorys), though inferior to it, the two timbers being much alike. The former is inclined to possess gum veins, while #. microcorys is free from them, and will stand a heavier strain. HE. Andrewsi is only found on the southern edge of the Plateau, on the higher elevations, on sedimentary formation; it grows to a fair height with good girth, the trunk and main branches being covered with a rough bark resembling that of #. piperita, while the upper branches are smooth. LE. pilularis (Blackbutt) is only found on the northern and eastern edges of the Plateau, growing on junction strata or sedimentary rock, and continues easterly to the coast; some very fine trees are to be seen along the road to Kendall. It does not seem to be so plentiful between the Plateau and Wauchope in a north-easterly direction, though nearer the coast and about Port Macquarie it is plentiful. This timber passes through the mills’ in large quantities, but being full of gum veins it is of not the same 292 THE COMBOYNE PLATEAU, value as Tallow Wood. EH. acmenioides (White Mahogany) grows on sandstone all around the fringe of the Plateau and is very plentiful. It is not much used as a timber, and is considered worthless by the mills; it is good for fencing purposes. E. microcorys (Tallow Wood) is about the most useful hardwood timber of all, and in great demand at the mills. Being free of gum veins and durable, it has many uses. One peculiarity about the timber is its greasy nature, and for this reason it is largely used as flooring for dancing halls, and also to a great extent in bridge building and decking. It grows to a fine tall tree with a straight trunk, many trees having a diameter, a few feet from the butt, of 5 feet. This tree is found growing on basalt in the brush forests, and frequently at the junction of basalt and other strata, and also on the latter alone. It is fairly well distributed throughout the Plateau, though it has been very much thinned out in late years for milling purposes. H. quadrangulata (Black Box) is only found in a very limited area on the eastern side at a high elevation on the summit of Mount Bulli; a few seedlings have found their way to the lower levels, though these make poor growth on the red soil compared to those on higher ground on black soil. This tree has not been recorded previously further north than the Hunter River. Its timber is not considered of any particular value and is not used at the mills. It does not grow to any great size. EH. saligna (Sydney Blue Gum or Flooded Gum) is a valuable timber and passes in fair quantity through the mills. The timber is of a red colour and used for wood-blocks and by wheel- wrights and for shipbuilding purposes. In the neighbourhood of creeks at the edges of brush forests and on the basalt itself, especially if near running water, it grows to immense proportions. It is a smooth gum with a varying amount of rough bark at the butt, and is to be found growing all over the Plateau, though considerably thinned out in the centre, and found well distributed on the junction strata at the edge of the Plateau. EH. grandis shows a considerable resemblance to H. saligna; the timber of the two species is much alike in colour, and the two forms are often confused, both going by the name of “Flooded Gum.” I have gathered from timber men that one timber is more interlocked than the other, but the real: distinction lies in the fruit, that of H. saligna having invariably three valves which are well exserted and straight, or slightly everted, while in that of H. grandis the valves are four, or frequently five, in number, less exserted and inverted and of a paler colour. The timber is used for much the same purposes as that of H. saligna. At the junction of basalt with other formations I have seen splendid specimens of this species, though it is not nearly as plentiful as H. saligna. The trunks of the two trees are almost identical in appearance, as is also the general shape of the tree and the foliage. E. propinqua, a Grey Gum, and the common Grey Gum here, grows only on the edges of the Plateau outside of the basalt, and is found on the lower levels in fair quantity, where it seems to do better, growing to larger proportions than on top; near sea-level about Kendall there are some fine trees. The timber is red, but little used here, probably on account of it. being too far from the mills. The bark is shed in patches, the denuded surface acquiring a salmon tint, giving the trunk a remarkable appear- ance. The fruit of this species is small. H. punctata, a larger-fruited Grey Gum, is not found right on top, but below the edge of the Plateau on the mountain side, especially to the north and west. It has a red timber, very hard and remarkably like that of Ironbark, and has been passed as such on many occasions. However, on striking the timber with the blunt end of the axe the impact gives a dull sound and the axe does not rebound, as in the case of Ironbark, which gives a ringing note and a larger rebound. This is one of the many tests employed to distinguish BY E. C. CHISHOLM. 293 between the two. It is inferior to Ironbark, though looked upon as a very yaluable timber and as a fairly efficient substitute; it has not the same tensile strength, and, moreover, contains many gum veins. It is used largely for railway sleepers and wood blocking. JH. canaliculata is a Grey Gum with a large fruit and angulated pedicel. The fruit is larger and coarser than that of H. punctata, and the timber, instead of being red, as is the case with the latter, is yellow. There is one tree only known to me that answers to these characters, and it is growing on top of a hill which has been partially cleared and on basalt formation close to an existing brush forest. In appearance it is quite different from H. punctata, especially in regard to its bark, which is rougher and has not the same clean appearance. Moreover, a peculiarity that struck me was its manner of flowering. This takes place on the medium-sized stems close to the main trunk, about half way up the tree, and directly off these stems which are 8 inches or so in diameter. The size of the tree in question is not large, the height would be about 40 feet and the diameter of the trunk at the butt ten inches to a foot. FE. tereticornis (Forest Red Gum) is found only at the edge of the Plateau and is not very plentiful, though much more so off the mountain in the surrounding low level land. This species is only found on the sandstone or sedimentary soil, frequently in the company of FE. propinqua and, on the lower levels, of H. amplifolia and EH. paniculata. It appears to have no particular value as a timber, the saw mills here not using it; the timber is red. I know of only one specimen of E. amplifolia (Broad-leaved Forest Red Gumf growing at the edge of brush at the junction of basalt and other strata, though in the surrounding country around the foot of the mountain it is very plentiful. The timber does not seem to be of any particular value, and it does not attain large proportions. H. corymbosa (Bloodwood) is plentiful outside the basalt all round the edge of the Plateau. As a timber it is not thought very much of on account of its numerous gum veins; it is used for posts and culverts, but is not used by the mills here. Leptospermum flavescens is the only species of the genus recorded, and is not at all plentiful, being found mostly off the basalt. Callistemon is represented by a crimson-flowered form, C. lanceolatus var., growing on rugged country at the edge of pure basalt. Melaleuca Leucadendron grows towards the southern side of the Plateau, where there is a mixture of soils. Two forms of the Araliaceae are represented: Tieghemopanax Murrayi, which grows in great profusion in the brushes, has a palm-like appearance with a leaf somewhat resembling that of the Red Cedar. There is a good deal of confusion as to the identity of this species; it commonly goes by the name of “Aralia.” It belongs to the family Araliaceae, but is not the true Aralia, which does not grow here. This tree is interesting on account of the occurrence of two or three isolated individuals in the Jamieson Valley on the Blue Mountains, close to the track leading down the ‘Valley of the Waters” at Wentworth Falls. As it is essentially a North Coast form, one wonders how it found its way there. The other species is 7. sambucifolius, which is fairly plentiful. The family Epacridaceae is represented only by Trochocarpa laurina, a tree sometimes called here “Wild Cherry,” from the fact that it bears small fruit, though with little resemblance to the Cherry, and a Leucopogon. Sideroxylon australe (Black Apple), of the family Sapotaceae, is a tree inhabiting the brush forests, the fruit of which is large and of a coarse texture. It is plentiful. 294 THE COMBOYNE PLATEAU, Diospyros cargillia (Hbenaceae), one of the so-called “Black Plums,” bearing a small egg-shaped black fruit with the calyx of the flower well represented at its base, is not plentiful. Of the Apocynaceae, most are climbing forms. Chilocarpus australis is a vine which, growing up into the trees, festoons them and is interesting on account of its long oval salmon-coloured fruit of the same tint as a ripe persimmon. Alyzia ruscifolius is a low shrub with prickly leaves growing in whorls of three. Lyonsia straminea and L. largiflorens are creepers festooning the brush trees. The family Asclepiadaceae is represented by Marsdenia, of which M. rostrata is the only one noted. This is interesting from the fact that in some districts it is under suspicion of being poisonous to stock, though not here as yet. I have found it in partially cleared paddocks used as grazing land for dairy cows. Ehretia acuminata (Borraginaceae) is interesting from the fact of its being one of our few deciduous trees. On superficial inspection it bears a good deal of resemblance to the cultivated Cherry in its general shape and form of leaf. It bears a profusion of small yellow fruits. It is a brush tree, and plentiful. Clerodendron tomentosum (Verbenaceae), an inhabitant of the brushes bearing a purple fruit surrounded by a fleshy collar, is not very plentiful. Gmelina Leichhardtii (White Beech), a very valuable timber, is much sought after as a softwood for the interior of houses, a large quantity passing through the mills. It grows to a fine tall tree, the trunk having a whitish appearance and the timber being light coloured. It is 4 typical brush tree. The Labiatae are poorly represented. Plectranthus parviflorus is a small plant liking moist situations, bearing a spiked purple inflorescence. Prostanthera ovalifolia var. latifolia is a shrub bearing handsome purple flowers growing on mixed strata. The Solanaceae are well represented in the genus Solanum, of which S. opacum, bearing small white flowers, is of small growth; S. pungentium is thorny and grows close to the ground with a large purple flower. The others, S. simile, S. aviculare and S. verbascifolium var. auriculatum, especially the last-named, are of much higher growth. They all bear fruit, mostly yellow or blackish in colour, and purple flowers. S. pseudo-capsicum is an introduced plant growing in great profusion here, having a white flower and orange-red berries. S. verbascifolium var. auriculatum, commonly called “Tobacco Plant” from the nature of its large leaf, has taken possession of many acres of the top after the areas have been cleared of the original brush. Duboisia myoporoides, known as “Cork Wood” on account of its rough bark, is also a feature of the cleared land, and grows into a medium-sized tree, bearing small white flowers. The leaves contain duboisine, consisting of a mixture of the alkaloids hyoscine and hyoscyamine, which is used as a mydriatic in ophthalmic practice. The only representative of Bignoniaceae is Tecoma australis, which is a climber festooning the brush trees, and when bearing flowers is very handsome. Hranthemum variabile, of the family Acanthaceae, is a small plant with a mauve-coloured flower. Myoporum acuminatum is the only representative of the Myoporaceae, and is very rare. I only know of one tree on the Plateau. It would make a fine ornamental tree, being shapely, with dense foliage. Turning to the Rubiaceae, Morinda jasminoides is a climber bearing small whitish flowers in profusion, the fruit being a small orange-coloured berry with a mosaic pattern dividing it into segments. Psychotria loniceroides is a small tree growing fairly plentifully in the brushes, having a leaf shaped somewhat after that of the Loquat, but softer in consistency. BY E. C. CHISHOLM. 295 Sambucus xanthocarpa (Native Elderberry) is fairly plentiful and is the only representative of the family Caprifoliaceae. Melothria Cunninghamii is a climbing form of the Cucurbitaceae. Members of the family Compositae are not numerous, being mostly of the genera Helichrysum, Gnaphalium, Erechites, Senecio and Olearia. On reviewing this paper several interesting facts obtrude themselves. The entire absence of Livistona australis (the Cabbage Palm) which one would expect to occur here, since it inhabits basalt formation and is to be found 20 miles east on the coast and 10 miles south-west on the Bulga Plateau at an elevation as high or higher than this. One cannot help wondering why it has not gained a foothold on this plateau, and this is the more remarkable since the fruit forms a part of the food of several species of Fruit Pigeon which are abundant on both plateaux: and migrate from one to the other. The absence of Ceratopetalum gummiferum is more easily explained, seeing that it grows almost entirely on sandstone. The small area of sandstone, comparatively speaking, on the plateau also accounts for the paucity of genera of the Proteaceae and Epacridaceae. I have been on the look out for Alphitonia excelsa (Red Ash), which I have failed to find, and can get no information about it, so I am almost forced to the conclusion that it does not occur here. The genus Flindersia (The Teak and its allies) does not occur, nor does Castanospermum australe (Black Bean), their habitat being further north. In conclusion I wish to record my grateful thanks to Mr. J. H. Maiden, late Director, and Dr. Darnell-Smith, the present Director of the Botanic Gardens, for affording me facilities for having the plants identified, and also to the staff of the National Herbarium, especially to Mr. W. F. Blakely, who has done most of the work and given me much information. List of the Plants of the Comboyne Plateau. Hymenophyllaceae: Trichomanes venosum R. Br. Cyatheaceae: Dicksonia Youngiae C. Moore, Alsophila australis R. Br., A. Cooperi F.v.M., A. Leichhardtiana F.v.M. Polypodiaceae: Dryopteris decomposita R. Br., D. punctata Thunb., D. parasitica (L.) O. Kuntze, D. acuminata Lowe, Arthropteris Beckleri, A. tenella Forst., Davallia dubia R. Br., Athyrium umbrosum Ait., Asplenium nidus L., A. adiantoides L., Blechnum cartilagineum Sw., B. serrulatum Rich., B. Patersoni R. Br., B. discolor Forst., B. capense (L.) Schlecht, Doodia aspera R. Br., Pellaea falcata R. Br., Cheilanthes tenuifolia Sw. var. Sieberi Benth., Adiantum aethio- picum L., A. formosum R. Br., A. affine Willd., Pteris tremula R. Br., Histiopteris incisa Thunb., Pteridium aquilinum L., Polypodium Brownii Wickstr., P. diver- sifolium Willd., Cyclophorus confluens R. Br., Platyceriuny bifurcatum Cav., Pellaea paradoxica (R. Br.). Gleicheniaceae: Gleichenia circinata Sw., G. flabellata R. Br. Osmundaceae: T'odea barbara (L.) Moore. Pinaceae: Callitris Macleayana F.v.M. Potamogetonaceae: Potamogeton tricarinatus F.v.M. Cyperaceae: Gahnia aspera Spreng., G. psittacorum Labill., Lepidospermum concavum R. Br. Palmae: Linospadix monostachyus Wendl. and Drude, Archontophoeniz Cunninghamiana Wendl. and Drude. Gymnostachys anceps R. Br., Pothos longipes Schott. eos F _Araceae:. Typhonium Brownii Schott., Colocasia macrorrhiza Schott 296 THE COMBOYNE PLATEAU, Liliaceae: Stypandra glauca R. Br., Xerotes longifolia R. Br., Xanthorrhoea resinosa Pers., Geitonoplesium cymosum A. Cunn., Hustrephus latifolius R. Br., Rhipogonum album R. Br., Smilax glycyphylla Sm., 8. australis R. Br. Iridaceae: Libertia paniculata Spreng. Orchidaceae: Dendrobium speciosum Sm., D. Kingianum Bidw., D. pugioni- forme A. Cunn., D. gracilicaule F.v.M., D. teretifolium R. Br., Bulbophyllum Shepherdi F.v.M., Spiranthes australis Lindl. Casuarineae: Casuarina torulosa Ait. Fagaceae: Fagus Moorei F.v.M. Moraceae: Cudrania javanensis Trec., Ficus rubiginosa Desf., F. macrophylla Desf., F. stephanocarpa Warb. Urticaceae: Urtica incisa Poir., Laportea gigas Wedd. Proteaceae: Persoonia media R. Br., Helicia glabrifiora F.v.M., Orites excelsa R. Br., Hakea saligna R. Br., Lomatia ilicifolia R. Br., Stenocarpus salignus R. Br. Santalaceae: Hxocarpus cupressiformis Labill. Loranthaceae: Loranthus dictyophlebus F.v.M., Phrygilanthus celastroides (Sieb.) Hichl. Polygonaceae: Polygonum hydropiper L. Phytolaccaceae: Codonocarpus attenuatus Hook. Ranunculaceae: Clematis aristata R. Br., OC. glycinoides DC., Ranunculus lappaceus Sm. Menispermaceae: Legnephora Moorei Miers. Magnoliaceae: Drimys dipetala F.v.M. Anonaceae: Hupomatia laurina R. Br. Monimiaceae: Piptocalyx Moorei Oliv., Wilkiea macrophylla A. DC., Palmeria scandens F.v.M., Daphnandra tenuipes Perk., Doryphora sassafras Endl. Lauraceae: Cinnamomum Oliveri Bailey, C. virens R. T. Baker, Litsea dealbata Nees, L. reticulata Benth., Cryptocarya glaucescens R. Br., C. australis Benth. Saxifragaceae: Quintinia Sieberi A. DC., Anopteris Macleayanus F.v.M. ~ Pittosporaceae: Pittosporum undulatum Andr., Hymenosporum flavum F.v.M., Bursaria spinosa Cav., Billardiera scandens Sm., Citriobatus multifiorus A. Cunn. Cunoniaceae: Aphanopetalum resinosum Endl., Geissois Benthami F.v.M., Ackama Muelleri Benth., Schizomeria ovata D. Don, Ceratopetalum apetalum D. Don, Weinmannia rubifolia Benth., Callicoma serratifolia Andr. Rosaceae: Rubus moluccanus L., R. parvifolius L., R. rosaefolius Sm., R. Moorei F.v.M., Acaena ovina A. Cunn. Leguminosae: Acacia elongata DC., A. melanoxylon R. Br., A. intertexta, A. binervata DC., A. longifolia Willd., A. floribunda F.v.M., A. decurrens Willd. var. mollis Lindl., A. juniperina Willd., Cassia sophera L., Daviesia corymbosa Sm. var. arborea Maiden, Gastrolobium Boormani Maiden and Betche, Indigofera australis Willd., Kennedya rubicunda Vent., K. prostrata R. Br. Geraniaceae: Geranium dissectum L., Pelargonium inodorum Willd. Oxalidaceae: Ozalis corniculata L. Rutaceae: Bosistoa euodiformis F.v.M., Geijera salicifolia Schott., Hvodia micrococca F.v.M., Zieria Smithii Andr., Acronychia laevis R. and G. Forst., A. Baueri Schott. Meliaceae: Cedrela australis F.v.M., Melia Azedarach L., Dysoxylum Fraser- anum Benth., D. rufum Benth., Synoum glandulosum A. Juss. Polygalaceae: Comesperma ericinum DC. Euphorbiaceae: Breynia oblongifolia J. Muell., Claoxylon australe Baill., Baloghia lucida Endl., Homalanthus populifolius Grah. BY E. C. CHISHOLM. 297 Celastraceae: Celastrus australis Harv., Denhamia pittosporoides F.v.M. Sapindaceae: Diploglottis Cunninghamii Hook., Sarcopteryx stipitata Radlk., Nephelium leiocarpum F.v.M., Dodonaea triquetra Wendl. Akaniaceae: Akania Hillii Hook. Rhamnaceae: Hmmenospermum alphitonioides F.v.M. Vitaceae: Vitis Baudiniana F.v.M., V. hypoglauca F.v.M. Hlaeocarpaceae: Hlaeocarpus reticulatus Sm., Sloanea Woollsii F.v.M. Malvaceae: Sida rhombifolia L., Hibiscus heterophyllus Vent. Sterculiaceae: Brachychiton acerifolius F.v.M., B. populneus R. Br., Tarrietia actinophylla Bailey, Commerconia Fraseri J. Gay. Dilleniaceae: Hibbertia volubilis Andr., H. dentata R. Br. Violaceae: Viola hederacea Labill. Flacourtiaceae: Streptothamnus Beckleri F.v.M. Passifloraceae: Passiflora alba Link and Otto. Thymeleaceae: Pimelea ligustrina Labill. Myrtaceae: Myrtus Beckleri F.v.M., Hugenia Smithii Poir., EH. corynantha F.v.M., H#. cyanocarpa F.v.M., Syncarpia laurifolia Ten., Backhousia myrtifolia Hook and Harv., Tristania conferta R. Br., T. laurina R. Br., Eucalyptus Andrewsi Maiden, #. pilularis Sm., HE. acmenioides Schau., H. microcorys F.v.M., E. quad- rangulata Deane and Maiden, H. saligna Sm., H. grandis Maiden, H. propinqua Deane and Maiden, #. punctata DC., E. canaliculata Maiden, EH. tereticornis Sm., E. amplifolia Naudin, HL. corymbosa Sm., Leptospermum flavescens Sm., Callistemon lanceolatus DC. var., Melaleuca Leucadendron L. Oenotheraceae: Hpilobium glabellum G. Forst. Araliaceae: Tieghemopanax Murrayi R. Viguier, 7. sambucifolius R. Viguier. Umbelliferae: Hydrocotyle asiatica L. Epacridaceae: Trochocarpa laurina R. Br., Leucopogon sp. Sapotaceae: Sideroxylon australe Benth. and Hook. Ebenaceae: Diospyros cargillia F.v.M. Gentianaceae: Hrythraea australis R. Br. Apocynaceae: Chilocarpus australis F.v.M., Alyxia ruscifolius, R. Br., Lyonsia straminea R. Br., L. largiflorens F.v.M. Asclepiadaceae: Marsdenia rostrata R. Br. Borraginaceae: Hhretia acuminata R. Br. Verbenaceae: Clerodendron tomentosum R. Br., Gmelina Leichhardtii F.v.M. Labiatae: Plectranthus parviflorus Henck., Prostanthera ovalifolia R. Br. var. latifolia Benth. Solanaceae: Solanum opacum A. Br., 8S. aviculare G. Forst., S. simile F.v.M., S. verbascifolium L., var. auriculatum Ait., S. pseudo-capsicum L., 8S. pungentium R. Br., Duboisia myoporoides R. Br. Bignoniaceae: Tecoma australis R. Br. Acanthaceae: Hranthemum variabile R. Br. Myoporaceae: Myoporum acuminatum R. Br. Plantaginaceae: Plantago varia R. Br. Rubiaceae: Morinda jasminoides A. Cunn., Psychotria loniceroides Sieb. Caprifoliaceae: Sambucus xanthocarpa F.v.M. Cucurbitaceae: Melothria Cunninghamii Benth. Compositae: Olearia dentata Moench., Helichrysum elatum A. Cunn., Z. Beckleri F.v.M., H. diosmifolium Don, H. bracteatum Willd., H. ferrugineum Less., Gnaphalium japonicum Thunb., Hrechites prenanthoides DC., Senecio dryadeus Sieb., Olearia ramulosa Benth. 298 THE COMBOYNE PLATEAU. Additional Determinations. (Added 26th August, 1925.) Cycadaceae: Macrozamia Perowskiana Mig. Only one plant found; stem about eight inches in diameter. Taxaceae: Podocarpus elata R. Br. Brown or She Pine; very rare. NOTES ON SPECIES OF PTEROSTYLIS. By Rev. H. M. R. Rupp, B.A. (Communicated by E. Cheel.) (Ten Text-figures. ) [Read 26th August, 1925.] Including an undetermined Tasmanian form, and a doubtful New South Wales P. striata, my herbarium contains 32 species of these interesting terrestrial orchids. Dr. R. S. Rogers in his paper on the Distribution of Australian Orchids (Trans. Roy. Soc. S.A., xlvii, 1923) gives 43 as the total number of determined Australian species. Maiden and Betche in their Census of the Plants of N.S.W. (1916) credit this State with 26, and, if specific rank be allowed to P. squamata R. Br. and P. Mitchelli Lindl., this number will be 28. I do not know whether other species have been added to New South Wales records since 1916, but of those in the Census, my collection lacks only Fitzgerald’s P. pedoglossa, P. clavigera, P. coccinea, and P. Woollsii, and of these I am acquainted with the first- and last- named. Until very recently, I had no Queensland specimens, but through the kindness of Miss H. Geissmann, of Tambourine Mountain, this defect is being remedied. There are six species (plus the doubtful P. striata) from New South Wales localities only, two from Tasmania only, three from Western Australia only, and one, or possibly two (see note on P. decurva) from Victoria only. The remainder are from various localities in the above States and South Australia. For all my South Australian specimens I am indebted to Dr. R. S. Rogers, of Adelaide, and for Western Australian plants to Mrs. G. E. Perrin, of Launceston, and Dr. Rogers. In the locality records given below, the initials M.R. denote specimens collected by myself; in all other cases the collector’s name is given. To Miss H. Geissman I am indebted for beautiful photographic studies of Tambourine Mountain plants which I regard as P. curta R. Br. (two forms), P. Baptistii Fitzg. (see note under this species and P. acuminata), P. nutans R. Br., P. grandiflora R. Br. (see noté), P. obtusa R. Br., P. decurva Rogers (doubtful), P. pedunculata, R. Br., and P. longifolia R. Br. 1. P. opHiIoGLossa R. Br.—This species shares with P. concinna the distinction of a “forked” labellum. In P. concinna the lobes or prongs are short and blunt, in P. ophioglossa they are long and fine-pointed, hence the name. I have recorded this species from Copmanhurst, N.S.W., and recently received a specimen from Mrs. G. Annand, of Lismore, Richmond River. The flower is usually much larger than that of P. concinna, which it resembles, but the fine points of the lower lip are more curved. Herbarium localities: Near Smith’s Lake, Bungwahl (Myall Lakes), July, 1924 (A. Rupp and M.R.), Paterson, June, 1925 -(M.R.)—a form with beautiful crimson striae on a translucent-greyish flower. 6 300 NOTES ON SPECIES OF PTEROSTYLIS, Range of species: Hastern Victoria (?) through N.S.W. into Queensland, at least as far north as Port Curtis. Chiefly in the coastal areas. 2. P. concInNA R. Br.—The flowers are a duller green than those of P. ophioglossa, with darker striae. The long points of the lateral sepals are typically very straight. Sandringham-Black Rock, Vic., Aug., 1897, and Sept., 1922 (M.R.), Cheltenham, Vic., June, 1915 (Dr. and Mrs. R. S. Rogers), Lane Cove, N.S.W., July, 1919 (M.R.), Lindisfarne, Tas., July, 1920 (M.R.), Bungwahl, N.S.W., June, 1924 (M.R.). Range: All States of the Commonwealth, except Western Australia, chiefly coastal and often littoral. 3. P. curTA,R. Br.—Maiden and Betche record two forms, one as var. grandi- flora (Bentham). I have collected what I take to be two distinct forms both in New South Wales and Tasmania, but have always regarded the larger and handsomer as the type, and the other—far less common—as a variety. Miss Geissmann has sent photographs, with the vernacular name “Swallow-tail’, of flowers apparently identical with the small form. She states that it blooms much earlier than the other; in the Bulladelah district of New South Wales it is rather more than a month earlier. Miss Geissmann informs me that Dr. Rogers was of opinion that it was a hybrid P. curta * P. pedunculata. I have not observed any indications of the latter as a possible parent, and should have expected a darker flower, whereas it is always a lighter green than the typical P. curta, with which it seems to agree in essential characteristics. In Victoria I have only seen the larger form. In Bailey’s Queensland Flora the author notes a doubtful var. grandiflora with flowers more than 2 inches long, the lobes of the lower lip ending in long points. I rather suspect that this is really P. Baptistii Fitzg. A true P. curta, however, should be distinguishable by the peculiar twist of the brown labellum in the mature flower, for observation of which fresh specimens are necessary. It is present in the small form above referred to. Wannon River Falls, Vic., Aug., 1893 (M.R.), National Park, S.A., Aug., 1909 (Dr. and Mrs. R. S. Rogers), Bellerive, Tas., Sept., 1920 (M.R.), Low Head, Tas., Oct., 1921 (Mrs. G. E. Perrin and M.R.), Launceston, Tas., Oct., 1922 (M.R.), Bulladelah District, N.S.W., early July (small form) and Aug. (large form), 1923 (M.R.). Range: All States except Western Australia, extending to the interior slopes of the Dividing Range in Victoria and New South Wales. 4. P. ACUMINATA R. Br.—Moore and Betche (Handbook of the Flora of N.S.W.) describe the dorsal sepal or galea as much incurved. I do not think this is a constant characteristic, as I have not infrequently found it comparatively straight from the main curve of the flower. Miss Geissmann sends photographs under this name from Tambourine Mountain, Queensland, but her plant appears to me beyond doubt identical with the much larger and handsomer P. Baptistii Fitzg. I sent Miss Geissmann dried specimens of the latter, and she concurs in the above opinion. Moreover, her description of the habitat of the Tambourine flower agrees with that usually characteristic of P. Baptistii rather than with that of P. acuminata, which in my experience (New South Wales and Victoria) occurs chiefly in sandy scrubs and heathland, not in rich mountain gullies or about sheltered forest creeks. In general appearance P. acuminata resembles P. curta, but is easily distinguished by the acuminate sepal-points and the green, non-twisted labellum. BY H. M. R. RUPP. 301 Parramatta River, N.S.W., May, 1916 (M.R.), Lane Cove, N.S.W., April, 1918 (M.R.), Bungwahl, N.S.W., June, 1924 (M.R.), coastal moors, near south end of Moreton Bay, Q. (Miss H. Geissmann). (The Tambourine Mountain plant flowers in September.) Range: Victoria, New South Wales, and Queensland. Bailey (Queensland Flora) says the lobes of the lower lips are produced into “long fine points’. I have invariably observed them to be comparatively short, though acuminate. 5. P. Bapristir Fitzg.—Probably the largest of all species in this genus. I have measured flowers 3 inches from base to tip of galea. The plant may reach as much as 2 feet 6 inches in height, but is also frequently quite dwarf. Until I resided in the Bulladelah district, I had only seen this plant in the dried state in the National Herbarium at Sydney. I found in great abundance in that district in 1923 a very large Pterostylis which appeared to be identical with Fitzgerald’s plant, except that only in rare cases were the stem-leaves strongly developed. The National Herbarium authorities, however, confirmed my determination, and Dr. R. S. Rogers expressed the opinion that the plant was undoubtedly P. Baptistii. The first appearance of this orchid above ground gives no hint of its subsequent development. The basal rosette is comparatively small, and in early stages the plant is not easily distinguished from P. pedunculata, the rosette-leaves of both being rather dark green, and much crisped along the margins. Yet one developes the largest, and the other one of the smallest, of the flowers in the genus. Herbarium specimens give no conception of the grace and beauty of the mature P. Baptistii, with its stately poise, and its harmonies of translucent-white, brown, and green. As mentioned above, Miss H. Geissmann has sent photographs of a Tambourine orchid which I have no hesitation in identifying with the plant so abundant at Bulladelah. Dr. H. L. Kesteven of the latter place found a specimen with two flowers perfectly developed, but rather under the average in size; it is now in my herbarium. The Bulladelah-Myall Lakes district is my only herbarium locality at present, and numerous specimens were collected in Sept., 1923, and Sept., 1924. Range of species: Imperfectly known; probably from about the N.S.W. National Park, northward along the coastal forests well into Queensland. 6. P. FurcAaTA R. Br. (Text-fig. 1)—Many years ago I found a colony of Pterostyles with very large flowers, green with darker longitudinal bands, in a gully on the slopes of Mount Buninyong, Vic. At the time they appeared to me to agree fairly with the description of P. cucullata R. Br., and for years they were so labelled. Subsequently I was disposed to doubt this determination; and when I first saw dried specimens of P. Baptistii, I wondered whether I had not discovered a form of that species a very long way beyond its supposed limits. During a residence in Tasmania from 1920-1923, I found two Pterostyles which appeared to me obviously distinct species, but which both passed under the name P. cucullata; the larger of the two much resembled the Buninyong orchid, except that it was more slender and had narrower and more stalked leaves. Mr. J. H. Maiden agreed that they certainly seemed distinct species. The larger he suggested as P. cucullata, and he thought that the other might prove to be what Robert Brown had called P. dubia. But in 1923, I sent specimens of the two Tasmanian plants and the Buninyong plant to Dr. Rogers for examination. He replied that in his opinion the Buninyong orchid was a form of P. furcata, though not typical; the smaller Tasmanian was a typical P. cucullata, and the larger one, P. falcata Rogers. This was satisfactory save for one point. I had Tasmanian specimens which I had labelled P. furcata, but they were 302 NOTES ON SPECIES OF PTEROSTYLIS, clearly not identical with the Buninyong specimens. I sent one to Dr. Rogers, but he was unable to determine it. This latter is described below as an undeter- mined species. Mount Buninyong, Vic., Nov., 1896 (M.R.). Range: Victoria, Tasmania, and South Australia. I have a single flower, without leaves, etc., from Mt. Kosciusko, N.S.W., which resembles Dr. Rogers’s figure of P. furcata. 1. Pterostylis furcata R.Br. Mt. Buninyong, Vic. x 3. 2. Pterostylis falcata Rogers. Mt. Barrow, Tas. x 4 approx. 7. P. FALCATA Rogers (Text-fig. 2).—See above, under P. furcata. Dr. Rogers has described and figured this species (Proc. Roy. Soc. Vict. xxviii (N.S.), Part I). It appears to be common in the colder parts of Tasmania. Mrs. G. HE. Perrin collected numerous specimens in the western highlands (Mt. Pelion, Mt. Ossa, etce.), Jan., 1922. In bogs among the foothills of Mt. Barrow I found many with larger flowers than Dr. Rogers has shown. It is a very graceful and handsome species, but lacks the rich brown markings of P. Baptistii, which it approaches in size. Mount Barrow, Tas., Jan., 1922 (M.R.). Range: Tasmania, Victoria, and probably New South Wales. Mr. R. H. Dowe, of Tamworth, has described to me an orchid which he found on the higher slopes of the New England tableland, which I think is very likely this species. 8. P. sp. ? (Text-fig. 3).—This is the plant referred to under P. furcata as an undetermined Tasmanian species. Dr. Rogers suggested that it might be a hybrid. Its abundance and wide distribution seem to me against its being so BY H. M. R. RUPP. 303 regarded. .I found it in abundance in the gullies of Mount Nelson and Mount Wellington in the south, and at Cataract Gorge in the north; on Mount Barrow I found numerous plants in fruit which seemed identical. The only other Pterostyles associated with it were P. nutans and P. obtusa (or decurva, see below), and beyond the generic characters it has no resemblance to either. It seems to have affinities with P. cucullata, but lacks the robust leaf development of that species, and the point of the galea is not similar. Comparing it with Dr. Rogers’s figure of P. furcata, it is typically taller (the Cataract Gorge plants were small, but were growing on an exposed ledge of rock under a she-oak), with a smaller flower, and a longer space between the latter and the first bract. The length of the lateral sepals is about the same as in P. acuminata. The flowe1' is almost wholly green, but turns rather a dark-brown when dried. Mount Nelson, Tas., Oct., 1920 (M.R.), Cascades, near Hobart, Oct., 1920 (M.R.), Cataract Gorge, Launceston, Oct., 1922 (M.R.). Range: Tasmania. 9. P. cUCULLATA R. Br.—This species also is described and figured in Dr. Rogers’s paper alluded to above. For many years I had specimens collected on the banks of Merri Merri Creek, near North Coburg, Vic., labelled P. Mackibboni F.v.M., but Dr. Rogers has shown that this is really a form of P. cucullata. The typical form I did not see until I found it in Tasmania. The species is an interesting one, owing to its variations. It may reach a foot in height, with a remarkable development of stem-leaves; on the other hand the leaves may be all basal, with the flower on a stalk not more than an inch long. The “hood” of the flower is most frequently greatly incurved like a parrot’s beak, but the flower itself is upright and not bent over like P. nutans, the “Parrot’s Beak” of popular designation. Rich dark-brown markings often make it a very attractive flower, but they are not a constant feature. Merri Merri Creek, Vic., Sept., 1897 (M.R.), Cherry Gardens, S.A., Sept., 1915 (Dr. and Mrs. R. S. Rogers), Mount Wellington, Tas. (at about 3,500 feet), Nov., 1920 (M.R.). Range: Recorded from all States except Queensland and Western Australia, but the confusion with P. falcata, mentioned above under P. furcata, should be remembered. 10. P. NuTANS R. Br.—Probably the commonest of all our Pterostyles, and widely familiar under the name of ‘‘Parrot’s Beak.” Near Launceston I found a colony of this species which in two successive years came into bloom a month after the general flowering time was over; but beyond unusual slenderness I could find no variation from the type. In the Bulladelah district a distinct variety occurs, flowering at least a month earlier than the type. It has a diminutive flower with a very short galea, and the labellum is more hispid than usual. It is probably the var. hispidula. Miss Geissmann reports what appears to be an identical form from Tambourine Mountain, Q. At Bulladelah I found a few weird teratological specimens which are now in the possession of Dr. R. S. Rogers. Bream Creek, Geelong, Vic., Oct., 1891 (M.R.), Kuitpo Reserve, S.A., Oct., 1906 (Dr. and Mrs. R. S. Rogers), Ryde, N.S.W., July, 1916 (M.R.), Hobart, Tas., Aug., 1920 (M.R.), Bulladelah, N.S.W., May, 1924 (var. hispidula), and July, 1924 (M.R.). Also other localities. Range: All States except Western Australia. 11. P. pEDUNCULATA R. Br.—The resemblance of the basal rosette to that of P. Baptistii in the young plants has been noted above. In the southern States, including Tasmania, there appears to be but one form of this species, with a M 304 NOTES ON SPECIES OF PTEROSTYLIS, flower slightly bulging below, translucent-white marked with red-brown. In New South Wales this type occurs, but there is another, usually taller, with a very small narrow flower almost wholly red-brown. The two are very distinct when . Pterostylis sp, Mt. Nelson, Tas. x 3. . Pterostylis pyramidalis Lindl. Jarnadup, W.A. x 3. . Pterostylis decurva Rogers. Lower Fern Tree Gully, Vic. x 4 approx. . Pterostylis grandiflora R.Br. Bulladelah, N.S.W. x #3. Oop ow seen together; both grew in the Bulladelah district. Miss Geissmann’s Queensland photograph looks like the smaller-flowered form. In one of the Mount Wellington gullies in Tasmania I obtained a specimen bearing two flowers. Sandringham, Vic., Oct., 1896 (M.R.), Balhannah, S.A., Sept., 1918 (Dr: and Mrs. R. S. Rogers), Mt. Wellington, Tas., Oct., 1920 (M.R.), Launceston, Tas., Oct., 1921 (M.R.), Bulladelah, N.S.W., Sept. and Oct., 1923 (M.R.), Paterson, N.S.W., Oct., 1924 (M.R.). Range: All States except Western Australia. 12: P. nana R. Br.—This dainty and diminutive species is sometimes found in colonies so dense that the plants might be plucked up by the handful. Chiefly littoral, but not exclusively so. BY H. M. R. RUPP. 305 Sandringham, Vic., Oct., 1896 (M.R.), National Park, 8.A., July, 1912 (Dr. and Mrs. R. S. Rogers), Westbury Road, Launceston, Tas., Sept., 1921 (M.R.), George Town, Tas., Oct., 1921 (Mrs. G. EH. Perrin and M.R.). Range: All States except Queensland. 13. P. pyramMimpALis Lindley (Text-fig. 4).—Flower resembling that of P. nana, but larger. It differs markedly from that species in its strongly developed, ovate-lanceolate stem leaves, and in the absence of a basal rosette. Jarnadup, W.A., Sept., 1919 (Dr. and Mrs. R. S. Rogers), Margaret River, W.A., Sept., 1922 (Mrs. G. HE. Perrin). Range: Western Australia only. 14. P. auaTA Reichb. f. (= P. praecox Lindl.).—It might occasionally be mistaken for P. obtusa, but is quite distinct. The flower is greyish, with green or red-brown bands. Lower lip with a very broad sinus, the long, finely-pointed lobes either embracing the galea or sometimes extending out in front of it. Stem from 4 to 8 inches, the stem-leaves increasingly bract-like from above downwards. Kangaroo Island, S.A., July, 1909 (Dr. and Mrs. R. S. Rogers), Bellerive, Tas., Sept., 1920 (M.R.), Launceston, Tas., Aug., 1922 (M.R.), Cape Barren I., Bass Straits, July, 1923 (Archdeacon Atkinson). Range: All States except Queensland and Western Australia. 15. P. srriataA Fitzg—I have provisionally so labelled three specimens collected on the Nandewar Range, west of Barraba, N.S.W. They were growing in association with P. reflera and P. truncata. They have a general resemblance to P. obtusa, but the labellum is narrow-linear and rather acute, and there is no projection at the sinus of the lower lip, which is a pronounced feature in P. obtusa. The galea is acuminate, the upper petals are rather blunt. Flower about 1 inch long, banded with red-brown and narrower green striae. Stem-leaves all approximately similar. Nandewar Range west of Barraba, N.S.W., April, 1914 (M.R.). Range: New South Wales. 16. P. oprusa R. Br. and 17. P. pEcurvA Rogers (Text-fig. 5).—For the purpose of these notes these two species are dealt with together. For many years I had a specimen labelled P. obtusa? which I obtained near Lower Fern Tree Gully, Vic., in November, 1897. The doubt was due partly to the unusual elongation of the galea, and partly to the time of flowering. I collected numerous specimens of the Tasmanian P. obtusa which appeared to agree with the Victorian plant in both these points. In 1923, I found P. obtusa growing freely on the Alum Mountain, ~ Bulladelah, flowering in the autumn. In every case the galea was merely acuminate, with no long decurved point. Dr. R. S. Rogers (Trans. Roy. Soc. ‘8. Aust. xlvii, 1923) described a new species, P. decurva, from Lower Fern Tree Gully, Vic., and sent me a specimen collected by Mr. EH. EH: Pescott in that locality in November, 1923. Comparing my own plant from the same locality, collected twenty-six years before, there could be no doubt of their identity; and this raised the further question whether the Tasmanian plant supposed to be P. obtusa might not prove to be really P. decurva. Dr. Rogers sent me South Australian specimens of P. obtusa which were autumn-flowering, and agreed precisely with the Bulladelah plants. More recently Miss Geissmann has sent Queensland specimens, collected in May, 1925, which are identical with the Bulladelah and South Australian plants, except that the flowers are larger and the stem-leaves are more strongly developed. But with them Miss Geissmann sent two smaller specimens which she regards as distinct, and which (except that here also the stem-leaves are very prominent) appear identical with the Victorian 306 NOTES ON SPECIES OF PTEROSTYLIS, P. decurva and the Tasmanian supposed P. obtusa. Dr. Rogers writes to say that he is disposed to agree that the Tasmanian plant is really P. decurva, not P. obtusa; but further investigation is desirable. And if Miss Geissmann’s smaller Queensland plants are P. decurva, we have the two forms flowering simultaneously on Tambourine Mountain in the autumn. It might be suggested that P. decurva is a variation from P. obtusa originating in the north, gradually altering its flowering-time as it spread southward, until in Tasmania it has become distinctly a. summer flower (Nov.-Jan.). I do not know whether the typical P. obtusa, with a short galea, occurs in Tasmania at all. My numerous specimens were collected on Mounts Nelson and Wellington in the south, and Mount Barrow in the north, and in every instance they have a long decurved galea. In this form the fine points of the lower sepals are curved outwards after embracing the galea (Miss Geissmann calls attention to this in the Tambourine plants); in the typical P. obtusa they are usually approximately straight. Until fuller investigation has settled the distinction between these forms all my specimens are included with the long decurved galea under P. decurva, confining the name P. obtusa to those with a merely acuminate galea. Dr. Rogers does not think that Miss Geissmann’s smaller plant is identical with P. decurva, but here also further investigation is required. Localities —P. obtusa: Hindmarsh Valley, S.A., April, 1912 (Dr. and Mrs. R. S. Rogers), Bulladelah, N.S.W., May, 1923 (M.R.), Tambourine Mt., Q., May, 1925 (Miss H. Geissmann). P. decurva: Lower Fern Tree Gully, Vic., Nov., 1897 (M.R.), and Nov., 1923 (EH. EH. Pescott, per Dr. R. S. Rogers), Mt. Nelson, Tas., Nov., 1920 (M.R.), Mt. Wellington, Tas., Dec., 1920 (M.R.), Mt. Barrow, Tas., Jan., 1922 (M.R.), Tambourine Mt., Q., May, 1925 (Miss H. Geissmann). Range: P. obtusa, all States except Western Australia (Tasmania doubtful) ; P. decurva, all States except Western Australia (?). 18. P. REcURVA Benth.—A most remarkable looking flower, almost having the appearance of being “turned inside out.” According to Dr. Rogers, the plant may attain a height of 20 inches. Stem-leaves numerous; lying rather close to the stem but not appressed. Flower about an inch long, banded longitudinally “dark greenish-grey” (Rogers). Lower lip with an acute sinus, the lobes not touching the galea, but recurved right away from it. Galea considerably shorter than lower lip, abruptly recurved, the petals very distinct from it, short, their points just meeting round the galea. Woogenellup, Stirling Ranges, W.A., Sept., 1919 (Dr. and Mrs. R. S. Rogers), Mount Barker, W.A., Sept., 1922 (Mrs. G. BE. Perrin). Range: Western Australia only. 19. P. REFLExA R. Br.—Somewhat similar in habit to P. obtusa, but the flower usually considerably larger. In New South Wales the highland forms seem to have consistently larger flowers than those on the coast. There appears to be a good deal of difference between forms included under this name. In some, the lower portion of the flower is extremely broad in proportion, and the points of the sepals, though fine, are comparatively short. In others, the flower is slender and the sepal-points are very long. Nandewar Range, N.S.W., April, 1914 (M.R.), National Park, S.A., May, 1918 (Dr. and Mrs. R. S. Rogers), Warrabah, near Barraba, N.S.W., May, 1924 (R. H. Dowe), Bungwahl, N.S.W., June, 1924 (M.R.). Range: Dr. Rogers reports it from all States. It is not in the Flora of Tasamnia, but of course may have been found there subsequently. BY H. M. R. RUPP. 307 20. P. trRuNcATA Fitzg.—I found this in association with P. reflexa and the doubtful P. striata at about 3,000 feet on the Nandewar Range, N.S.W. It was at first determined for me as P. praecox, but subsequent familiarity with that species (= P. alata) showed it to be a very different plant. The Nandewar orchid seemed to me to agree with descriptions of Fitzgerald’s P. truncata, and as Dr. Rogers concurred in this opinion I have retained it. The flower is almost twice as large as that of P. alata, on a very short stem of 1 to 2% inches. Stem-leaves several, expanded or almost appressed. Flower banded with dark-green or reddish-brown striae; galea acuminate, upper petals bluntly truncate. Points of the lateral sepals extending far above the galea, hooked at the tips in all my specimens. Nandewar Range, N.S.W., April, 1914 (M.R.). Range: New South Wales and Victoria. 21. P. GRANDIFLORA R. Br. (Text-fig. 6).—The graceful lines of the flower crowning the prominent stem-leaves are very attractive, and the upper petals and galea are broadly expanded to form a rich red-brown hood, the rest of the flower being banded with the same colour against translucent-white. The basal rosette is usually withered before flowering time. The stem-leaves are developed with the longest one about the middle. Miss H. Geissmann sends a photograph of P. grandiflora from Tambourine Mountain, Q. Lane Cove, N.S.W., June, 1917 (M.R.), Bulladelah, N.S.W., May and July, 1924 (M.R.). Range: New South Wales and Southern Queensland. 22. P. PARVIFLORA R. Br.—In contrast with the last, this is probably the least attractive of all the Pterostyles (P. aphylla Lindl. included). The basal rosette may sometimes be present during flowering, but is not attached to the flowering stem. (This is true of some other species also—P. grandiflora, P. obtusa, P. longifolia—but in these cases it is unusual for the rosette to be present at flowering time. This seems to be the only real distinction between the typical P. parviflora and the form aphylla; the rosette of the latter disappears before flowering. ) Buninyong, Vic., May, 1896 (M.R.), Mt. Compass, S.A., April, 1915 (Dr. and Mrs. R. S. Rogers), Capertee, N.S.W., April, 1917 (M.R.), Lane Cove, N.S.W., June, 1917 (M.R.). Range: All States except Western Australia. 23. P. BARBATA Lindl.—A peculiar and striking flower; from a few inches in height, and slender, to a robust plant of nearly 18 inches. Leaves crowded at the base but not rosulate, extending up the stem gradually as appressed bracts. Flower very upright, pale green, except the labellum; upper petals and galea appearing as if one segment. Labellum much exserted, often curved down, apparently by the weight of the brown knob at the end; filiform with many bright yellow hairs. Lower lip sharply decurved away from the upper; points of its lobes hardly acute. Ringwood, Vic., Oct., 1896 (M.R.), The Hermitage, S.A., Sept., 1913 (Dr. and Mrs. R. S. Rogers), Kangaroo Valley, Hobart, Oct., 1920 (M.R.), Launceston and Low Head, Tas., Oct., 1921 (Mrs. G. E. Perrin and M.R.), Grampians, Vic., Oct., 1922 (Mrs. G. H. Perrin). Range: All States except Queensland. I do not think it comes far north in New South Wales. 308 NOTES ON SPECIES OF PTEROSTYLIS, 24. P. rurrosa Lindl.—Obviously very closely allied to P. barbata, but more slender, less leafy, and the galea and lobes of the lower lip are produced into very long fine points—the latter not decurved sharply as in P. barbata. Labellum similar. Upper King River, W.A., Sept., 1919 (Dr. and Mrs. R. S. Rogers), Albany, W.A., Sept., 1922 (Mrs. G. E. Perrin). Range: Western Australia only. 25. P. CYCNOCEPHALA Fitzg—A small plant, probably often confused with P. mutica, from which it may be distinguished by a more robust habit, by the flowers being massed together, and by the “swan-headed” labellum-appendage from which its name is derived. The labellum is irritable, springing back towards the column when shaken or touched. I have observed this characteristic in all the remaining species included below except P. Daintreyana. Merri Merri Creek, Vic., Sept., 1897 (M.R.), Monarto South, S.A., Sept., 1906 (Dr. and Mrs. R. S. Rogers), Bellerive, Tas., Oct., 1920 (M.R.), Bulladelah, N.S.W., July, 1924 (A. Rupp). Range: New South Wales, Victoria, South Australia and Tasmania, on grassland. 26. P. muticA R. Br.—Similar to P. cycnocephala and of similar habit, but more slender, reaching, I think, a greater height, and flowers fewer. Whole pliant, as in P. cycnocephala, pale green. Warialda, N.S.W., Sept., 1905 (M.R.), Monarto South, S.A., Sept., 1913 (Dr. and Mrs. R. S. Rogers). Range: All States except Western Australia. 27. P. sqguaMaTA R. Br. (Text-fig. 7).—Maiden and Betche (Census) include . P. squamata and P. Mitchelli as varieties of P. rufa R. Br. Dr. Rogers (South Australian Orchids) restores them to specific rank, and I prefer to follow him, because, though P. rufa seems to be a very variable species and to approach both these forms, the latter typically show scarcely any variation over a very wide range of habitat. I have P. squamata from Launceston, Tas., and from Bulladelah, N.S.W.—900 miles distant—and they are alike in all respects. So also I have P. Mitchelli from Grangeville, S.A., and from Warialda, in the extreme north of New South Wales, and if the specimens got mixed I could not distinguish them. I think that any confusion between P. squamata, P. rufa, and P. Mitchelli is due to the variability of P. rufa. P. squamata is named from the unusual number of appressed stem-bracts. Flowers several, small, sepal-points acuminate, galea prominently down-curved and then suddenly straightening at the tip. Flowers rather prominently marked with red. Basal rosette withered at flowering time. Westbury Road, Launceston, Nov., 1921 (M.R.), Markwell, near Bulladelah, N.S.W., Sept., 1924 (Dr. H. L. Kesteven). I had found a budding plant at Bulladelah early in Sept., 1924, which I considered to be either P. rufa or P. squamata. Dr. Kesteven sent me the specimens now in my herbarium. Range: New South Wales, South Australia, and Tasmania; probably Victoria also. 28. P. RuFA R. Br. (Text-fig. 8) —See remarks under P. squamata. Dr. Rogers has sent me two beautiful South Australian specimens which have no red about them at all. The only other specimen I have, and all specimens I have seen in New South Wales, are rather heavily marked with red like P. squamata, but the BY H. M. R. RUPP. 309 flowers are twice the size of that species, and the sepals have long fine points. In Dr. Rogers’s flowers they have still longer points. Basal rosette withered at flowering time. Blackwood, S.A., Nov., 1906 (Dr. and Mrs. R. S. Rogers), Barraba, N.S.W., Sept., 1913 (M.R.). Range: All States. 7. Pterostylis squamata R.Br. Launceston, Tas. x 4. 8. Pterostylis rufa R.Br. Barraba, N.S.W. (complete specimen), Blackwood, S.A. (flower). x 3. 9. Pterostylis Mitchelli Lindl. Warialda, N.S.W. x 3. 10. Pterostylis vittata Lindl. National Park, S.A. x 3. 29. P. MircHeELii Lindl. (Text-fig. 9) —Basal rosette not withered at flowering time. Otherwise considerably resembling Nos. 27 and 28, but always with green flowers. Dr. Rogers very carefully describes these three closely-allied species (South Australian Orchids, p. 42). Grangeville, S.A., Aug., 1907 (Dr. and Mrs. R. S. Rogers), Gunyerwarildi, Warialda District, N.S.W., Sept., 1916 (M.R.). Range: All States except Western Australia, according to Dr. Rogers. I should be surprised to find such a characteristically “interior” species in Tasmania. 30. P. DAINTREYANA F.v. Muell—An interesting little plant, apparently restricted to the Port Jackson district or not extending far from it. In mossy places on sandstone outcrops. Plant up to 1 foot high, very slender; basal rosette 310 NOTES ON SPECIES OF PTEROSTYLIS. present at flowering time. Stem-leaves bract-like below, expanding towards the ~ very small green flowers, which have long points to the sepals. Gordon, N.S.W., June, 1917 (M.R.). 31. P. virratTa Lindl. (Text-fig. 10).—Chiefly in tea-tree scrub on the edge of sea beaches. Stem from a few inches to well over a foot. No basal rosette observed. Stem-leaves very strongly developed, lanceolate, the largest about the middle. Flowers several, 4 inch to about 1 inch long from the roof of the galea to the down-turned tips of the lateral sepals; sometimes greenish, but more usually heavily marked with reddish or purplish-brown. An attractive species. Sandringham, Vic., July, 1892 (M.R.), National Park, S.A., May, 1909 (Dr. and Mrs. R. S. Rogers), Cannington, W.A., Sept., 1922 (Mrs. G. HE. Perrin), Cape Barren I., Bass Straits, July, 1923 (Archdeacon Atkinson). Range: All States except New South Wales and Queensland. 32. P. LONGIFOLIA R. Br.—Somewhat resembling No. 31, but more slender, leaves longer and narrower, and flowers always quite green. Basal rosette withering before flowering time, or, if present, on a separate stem. Plant occasionally very tall. National Park and Kangaroo Island, S.A., July, 1909 (Dr. and Mrs. R. S. Rogers), Lane Cove, N.S.W., July, 1919 (M.R.), Hobart, Tas., Aug., 1920 (M.R.), Grampians, Vic., Oct., 1922 (Mrs. G. HE. Perrin), Bulladelah, N.S.W., July, 1923 (M.R.). Range: All States except Western Australia. (The accompanying pen-and-ink sketches are all from herbarium specimens, though an effort has been made to avoid the straightening effect of pressing upon the curve of the galea.) PLATE XXIV. Proc. Linn. Soc. N.S.W., 1925. 4, 5. Cladophlebis sp. 6-9. Thinnfeldia Feistmanteli. 1-3. Phyllotheca australis. Lie Le ale GT Cw age ik Proc. Linn. Soc. N.S.W., 1925. PLATE XXV. 1, 2. Thinnfeldia Feistmanteli. 3. 7. lancifolia. PLATE XXVI. Proc. Linn. Soc. N.S.W., 1925. narrabeenensis. AL, aN, 1-3. Thinnfeldia lancifelia. Proc. Linn. Soc. N.S.W., 1925. PLATE XXVIII. Thinnfeldia narrabeenensis. Proc. Linn. Soc. N.S.W., 1925. PLATE XXIX. 1. Taeniopteris Tenison-W oodsi. 2. T. crassinervis. 3. T. wianamattae. 4, 6. ? Coniopteris cf. lobata. 5. ? Sphenopteris sp. 7-9. ? Williamsonia sp. 10,11. Fern stem. Proc. Linn. Soc. N.S.W., 1925. PLATE XXX. 1,2. ? Williamsonia sp. 3,4. ? Rhipidopsis narrabeenensis. 5,6. Brachyphyllum angustum. Proc. Linn. Soc. N.S.W., 1925. PLATE XXXI. 1. Ginkgoites sp. 2..? Araucarites sydneyensis. 3-5. Carpolithus sp. 6-11. Plantae incertae sedis. 12. Taeniopteris crassinervis. Proc. Linn. Soc. N.S.W., 1925. PLATE XXXII. 1. Oxvcopera intricata. 2. Oncopera mitocera. 9 3. Oncopera brachyphylla. 4. Oncopera epargyra. Proc. Linn. Soc. N.S.W., 1925. PLATE XXXIII. Species of Tulostoma. 1. pubescens. 2. albicans. 3, 4. McAlpinianum. 5. macrosporum, 6. adhaerens. Proc. Linn. Soc. N.S.W.. 1925. PLATE XXXIV. Species of Tulostoma. 7. adhaerens. . 8. brumale. 9. Purpusii. 10, 11. poculatwm. 12. subfuscum. 13. striatum. fe Proc. Linn. Soc. N.S.W., 1925. IZA SOOO Species of Tulostoma. 14-16. australianum. 17. striatum. 18. subfuscum. 19. albicans. 20. macrosporum. 21. brumale. 22. poculatum. » NOTES ON AUSTRALIAN DIPTERA. No. vii. NS By J. R. MALLocH. (Communicated by Dr. HE. W. Ferguson.) (Twenty-three Text-figures. ) [Read 30th September, 1925.] In this paper I present notes on some previously described species and descriptions of some others that appear to be new to science. With the exception of one species which belongs to the British Museum, all the types will be sent to Dr. E. W. Ferguson, from whom most of them were received for identification. Family Sepsidae. The members of this family have a characteristic ant-like appearance that readily distinguishes them from most of their allies. Hendel’s most recent papers dealing with the acalyptrate Diptera give prominence to the presence or absence, divergence or convergence, of the postvertical pair of bristles as characters for the separation of the families. While one must admit that these bristles are quite important as criteria in grouping the insects, a careful scrutiny of the families discloses the fact that, just as is the case with other characters, there are some departures from the general rule here. Hendel ascribes to Sepsidae divergent postvertical bristles, but in one of his papers he parenthetically refers Eurychoromyia Hendel to the family, despite the fact that it lacks these bristles. I rather incline to doubt the propriety of assigning this genus here, but have not seen it, so cannot give a definite opinion on the point. However, there is one undoubted sepsid amongst the Australian material before me, which lacks the postverticals, and for which I herein propose a new genus. The preapical tibial bristle is stated to be absent by Hendel, but I am confident that in some species it is present, though small and weak, as I can detect a setula or short bristle in practically the normal position in these. Strictly speaking the family is distinguished from its nearest allies by the distinct auxiliary vein, which is complete and well separated from the first vein; the presence of vibrissae; bare, or almost bare, arista; absence of presutural dorsocentral bristles; presence of one or more long setulose hairs on lower margin of metathoracic spiracle; incomplete sixth wing vein; lack of pteropleural and sternopleural bristles; and the vestigial palpi. The postvertical bristles are absent or present; when present they are divergent; orbit with, at most, one distinct bristle. In 1906 de Meijere published a revision of the Indo-Australian species of Sepsis in which he recorded two species from Australia (Ann. Mus. Nat. Hungar., vol. 4, p. 165). I have both of these species before me, as well as one he described from Singapore, but did not record from Australia, and in addition have three related forms as yet undescribed, for two of the latter having to propose new genera in this paper. A 312 NOTES ON AUSTRALIAN DIPTERA, Vii, All the known species of the family occur in the adult stage upon garbage, carrion, or vegetation, some being very abundant on flowers; the larvae feed in manure and carrion. The adults are very rarely found in houses, so do not come into contact with human food as a rule, and thus may be considered as innocuous. Key to known Australian Genera. 1. Postvertical bristles lacking ; each orbit with one strong bristle .... Xenosepsis, n. gen. Postvertical bristles present, divergent; no distinct orbital present .............. 2 2. Basal cells of wings separated by a strong vein ....................:. Sepsis Fallén Basal cells of wings not separated (Fig. 1); mesopleura with a strong bristle on hina sniawe in all Owe eases soos SU Oe ee le es cate eee Australosepsis, n. gen. Genus Sepsis Fallén. This cosmopolitan genus is represented by four species in the material now before me. One of these, hirsuta de Meijere, has, like violacea Meigen, but one distinct pair of thoracic dorsocentral bristles, and has also a strong mesopleural bristle as in Australosepsis, characters which might be utilized to separate them subgenerically from the other species, but there is nothing to be gained by adopting this course unless in a consideration of the species from all over the world, so I leave the matter to some future worker to decide. I give below a key for the identification of the available Australian species. Key to Species. 1. Thorax with one pair of strong dorsocentral bristles; frons quite conspicuously haired; dorsum of thorax evenly haired, most noticeably so in male; fore femur of female with anteroventral and posteroventral series of short black spines on apical half, of male with similar series, which are confined more to median third and are not situated upon raised bases; wing without a black spot at apex of second vein; legs more noticeably hairy than in other species; mesopleura with a distinctabristlesonshind smanreinga DOVenaneiaeiorcieieie cienierniar hirsuta de Meijere Thorax with two pairs of strong dorsocentral bristles; frons bare except on orbits; disc of mesonotum with very inconspicuous hairs, which are practically all confined to acrostichal area and in line with dorsocentrals; fore femur in female without well developed spines, of male with distinct spines, some of which are on distinctly elevated bases; mesopleura without a distinct bristle ............ 2 2. Wing normally with a black spot at apex of second vein; no acrostichal setulae present behind level of anterior pair of dorsocentrals; veins 3 and 4 of wings notatall-orxonilyishehtly. conversent ra picallly, saree deieieieorerieieoiene renters 3 Wing without a black spot at apex of second vein; acrostichal setulae continued sparsely well behind transverse line of anterior dorsocentrals; scutellum not opaque velvety-black, but black, coloured as disc of thorax; veins 3 and 4 quite distinctly convergent apically ..................... coprophila de Meijere 3. Black species, the scutellum velvety opaque-black, with apex white dusted; fore femur in both sexes without prominent hairs on anteroventral surface basally, Wwaithed-3 short) black setulae qeicsmisisc eicrs ciieicna cnaketchela sietiens plebeia de Meijere Fulvous yellow species, scutellum yellow; fore femur of male with numerous black hairs on basal half of anteroventral surface .................. hirtifemur, n. sp, SEPSIS HIRSUTA de Meijere. Text-figures 2, 3. This species is similar to violacea Meigen in having but one pair of dorso- central bristles and in possessing a mesopleural bristle. It differs, however, in the spinose fore femur of the female, in which respect it is like Nemopoda species. The preponderating colour in most specimens is fulvous yellow, with the dorsum of thorax and of abdomen more frequently dark; some specimens are almost entirely black. The fore femur of male is as in Figure 2, that of female as in Figure 3. I have not seen a specimen with a dark subapical spot on wing. BY J. R. MALLOCH. 313 Length, 4-6 mm. Originally described from Parramatta and Botany Bay, N.S.W. I have before me many specimens of both sexes from Sydney and Como, N.S.W. SEPSIS COPROPHILA de Meijere. I have before me one female which I identify as this species. The characters cited in the key, coupled with de Meijere’s description in the paper already referred to, should enable anyone to identify the species. Originally described from Singapore. I have it from Hidsvold, Queensland. SEPSIS PLEBEIA de Meijere. Text-figure 4. A small black species closely resembling the Huropean cynipsea Linné. The fore femur lacks the rather dense hairs on the basal half of anteroventral surface so evident in hirtifemur, and the armature as seen from behind is as shown in Figure 4. The female has no fore femoral spines. Length, 3-4 mm. Originally described from Sydney. I have it from Sydney, Como, Cronulla, and Botany Bay, N.S.W. r . Wn, eat Mma gaya enn’ Text-figure 1.—Australosepsis fulvescens, n. gen. et sp. Wing of male. Text-figure 2.—Sepsis hirsuta de Meijere. Fore femur of male. Text-figure 3.—Sepsis hirsuta de Meijere. Fore femur of female. Text-figure 4.—Sepsis plebeia de Meijere. Text-figure 5.—Sepsis hirtifemur, n. sp. Fore femur and tibia of male. Text-figure 6.—Australosepsis fulvescens, n. gen. et sp. Fore femur and tibia of male. Text-figure 7.—Xenosepsis sydneyensis, n. sp. Fore femur of male. Text-figure 8.—Piophila —————_. Wing. 314 NOTES ON AUSTRALIAN DIPTERA, Vii, SEPSIS HIRTIFEMUR, n. Sp. Text-figure 5. Male——Fulvous yellow, shining; dorsum of thorax with a central dark suf- fusion; abdomen polished, blackened above and with a purplish tinge. Wing with a blackish spot at apex of second vein, the costal vein white beyond the black spot. Sternopleura white dusted above. Vertical, postvertical, and ocellar bristles strong; a few microscopic hairs on orbits; a long setula on second antennal segment; vibrissae duplicated; cheek very narrow; eye facets enlarged in front. Thorax with two pairs of widely separated dorsocentrals, a series of hairs in line with these, and paired acrostichal hairs, anteriorly; basal scutellars minute. Abdomen with distinct tergal bristles. Fore femur and tibia seen from behind as in Figure 5, the anteroventral surface of femur with numerous black hairs on basal half; mid and hind femora each with an anterior bristle; mid tibia with one anteroventral and two posterior bristles; hind tibia attenuated basally, with one anteroventral, one anterodorsal, and one posterodorsal bristle; mid metatarsus with some short black bristles. Inner crossvein of wing at less than one-third from apex of discal cell; first posterior cell not narrowed at apex. Length, 3 mm. Type, Mosman, N.S.W. Genus AUSTRALOSEPSIS novum. Text-figure 1. Generic characters.—Venation of wing as in Pandora Haliday, the vein separating first and second basal cells lacking (Fig. 1). Differs from that genus, the only other of the family with this venation, in having the vibrissal angle but slightly produced, the frontal orbits without strong median bristle, fore femora as in Sepsis, not simple, in male, and the scutellum short and convex on disc. An offshoot from Sepsis apparently. Genotype, Australosepsis fulvescens, n. sp. There appears to be but one species in the material available, but there is a remarkable difference between the paler and darker specimens, so that it appears justifiable to give to the dark specimens a varietal name, as indicated here. A. Thorax, abdomen and legs entirely or almost entirely fulvous yellow ............ suis atiatle! Va\ie\ fousa fel i rotairen duvena ctor se ste AceMerss as sn eR eet eS Re SO Te eee naarre erect ance memes fulvescens, n. sp. AA. Thorax and abdomen entirely black, the former shining, the latter glossy and with a purplish tinge; legs largely black. ...................- var. atratula n. AUSTRALOSEPSIS FULVESCENS, n. sp. Text-figures 1, 6. Male.—Shining fulvous yellow, sometimes a little darkened on dorsum of thorax and abdomen. A black spot at apex of second wing vein, and a brown suffusion along costa to a little beyond humeral cressvein, costal vein from a little beyond apex of second vein to apex of fourth white. Postvertical bristles moderately strong, divergent, all four verticals and one pair of ocellars present; a very short setula at middle of each orbit; frons bare; eye longer than high; cheek not as high as width of third antennal segment; vibrissae duplicated, the angle very slightly produced. Thorax with one humeral, one mesopleural, two notopleural, two pairs of dorsocentral, and two long and two very minute scutellar bristles. Abdomen constricted at apex of second tergite as in Sepsis, the tergites with distinct apical bristles. Fore femur and tibia of male as in Figure 6; fore tarsus of same sex normal, mid femur with BY J. R. MALLOCH. 315 38 or 4 anterior bristles, mid tibia with one anteroventral and 3 or 4 posterior bristles, mid tarsi with a number of black bristles on basal two segments; hind femur usually with an anterior bristle beyond middle; hind tibia slender on basal half, thickened beyond, and with three bristles at middle, one anteroventral, one anterodorsal and one posterodorsal. Wing as in Figure 1, the crossveins sometimes a little more approximated. Length, 3-4 mm. Type and three paratypes, Sydney, N.S.W., 17.2.24; allotype. One paratype, Cronulla, N.S.W. AUSTRALOSEPSIS FULVESCENS var. ATRATULA, N. var. Male and female.—Differ from the type form in having the thorax and abdomen black; the femora and tibiae of mid and hind legs are largely black and apices of tarsi fuscous. Structurally as in typical form. Length, 3-4 mm. Type, male, allotype, 2 male and 3 female paratypes, Sydney, 17.2.24; one male, Cronulla, N.S.W., Dec., 1924 (H. Petersen). Genus XENOSEPSIS novum. Generic characters.—Differs from all Sepsidae known to me in lacking the postvertical bristles. Hach frontal orbit with one prominent bristle which is directed outward over eye; face carinate; one outstanding vibrissa present; head otherwise as in Sepsis. Thorax as in Sepsis; mesopleura with one bristle; dorso- centrals one pair. Legs as in Sepsis, but the preapical tibial bristle is quite evident, and there is a short longitudinal slit on dorsal surface of hind tibia near base in male. This last character may be sexual and not of generic import. Venation of wings as in Sepsis. Genotype, Xenosepsis sydneyensis, n. sp. XENOSEPSIS SYDNEYENSIS, n. sp. Text-figure 7. Male.—Shining black, dorsum of thorax shagreened, that of abdomen polished and slightly purplish. Antennae and centre of face yellowish. Legs black, coxae, trochanters, fore femora and tibiae, bases of other femora, and tarsi, except their apices, yellowish. Wings without dark preapical spot. Halteres yellow. Cheek almost linear; frons bare. Thorax with a pair of short bristles on acrostichal lines about midway from suture to dorsocentrals; upper part of sterno- pleura white dusted; scutellum short, apical bristles long, basal pair microscopic. Abdomen not noticeably constricted at second tergite, bristles evident only on the apical tergites, two at apex quite prominent. Fore femur as in Figure 7; mid femur with some short anterior setulae; mid tibia with an anteroventral bristle beyond middle; basal segment of mid and hind tarsi setulose; hind femora and tibiae without median bristles. Inner crossvein at about one-third from apex of discal cell; veins 3 and 4 slightly convergent apically. Length, 4 mm. Type and two paratypes, Sydney, N.S.W., 8 and 14.1.23. Family Piophilidae. This family has much the same appearance as Sepsidae, but they are not so slender and ant-like. They differ in venation of the wing, as shown in Figure 8, the auxiliary vein being practically fused with first at its apex; the costa is almost = 316 NOTES ON AUSTRALIAN DIPTERA, Vii, broken just in front of apex of this vein. Palpi large; vibrissae strong; no setulae on hypopleura; preapical tibial bristle undeveloped; postvertical bristles divergent. I have before me two species referable to this family, both of which appear to be European in origin. I give below a diagnosis for their identity. It appears probable that the chaetotactic characters cited below will serve some taxonomist as criteria for the division of this genus into at least subgenera. A. Frons without orbital bristles; humeral and sternopleural bristles undeveloped ; dorsum of thorax very sparsely haired, distinctly shagreened; fore tarsi not (OHO O(cY0 UaRUne ee oie BONE Pern Ua ty Ere ret rare Sea Reem hata sae ARR mr at eo eR eA hens br casei Linné. AA. Frons with a pair of distinct orbital bristles; humeri and sternopleura each with two bristles; dorsum of thorax with numerous erect hairs, not shagreened; fore (WANS Gonos! aincl ChieyaCl ssosneceoscodscechousasdsavguued latipes Meigen. PIOPHILA CASEI Linné. A greenish or bronzy black species. Anterior half of frons, the face, cheeks, and palpi yellow. Legs variable in colour, generally yellow, with most of fore femora, all of fore tibiae, and apices of hind femora and hind tibiae broadly, black, the fore tarsi sometimes all black, sometimes yellowish in middle. The shagreened thorax with its comparative absence of hairs, transversely rugose scutellum, and undilated fore tarsi characterizes this species. Length, 4-5 mm. Localities—Sydney, Blue Mts., Illawarra and Waterfall, N.S.W., and Hidsvold, Queensland. This is the species known the world over in economic literature as the “Cheese Skipper’’, from the lively actions of the larvae, which feed in cheese and preserved meats. Sometimes the flies occur in great numbers in provision houses. -PIOPHILA LATIPES Meigen. I identify as this species a female which agrees with Meigen’s description so well that there is no doubt in my mind that I am correct. It is very closely related to nigriceps Meigen. The latter, however, has no strong orbital bristles and is different in colour and structure. Legs yellow, fore pair black except the coxae, trochanters, bases of femora, and knees. Length, 3.5 mm. Locality—Sydney, N.S.W., 14.4.25. Family Sapromyzidae. Genus SApromMyzaA Fallén. I have already defined this genus and Sapromyzosoma in one of my papers on Australian Diptera. I now present descriptions of a few species of each genus, some of them previously undescribed. One species herein included, aberrans, might be placed in a separate subgenus from the others because of the presence of a distinct bristle behind the supra-alar bristle. I have noted the presence of this bristle in many Oriental and South American species of this genus and Minettia, but I have not yet decided the true significance of the character and therefore leave this species in Sapromyza in the meantime. The bristle referred to must not be confused with the posterior intra-alar bristle, the presence of which distinguishes Minettia from Sapromyeza, both bristles being present in certain species of the former genus which I have seen. BY J. R. MALLOCH. 317 SAPROMYZA ABERRANS, Nl. Sp. Female and male.—Head opaque clay-yellow, frons with a transverse dark grey mark extending between inner vertical bristles, and an irregular brown mark on each orbit, the two meeting on middle of frons; a dark mark between each antenna and eye; centre of face fuscous; apex of third antennal segment darkened; palpi black. Thorax clay-coloured, dorsum with four broad brownish vittae, the laterals irregular, subdivided by a grey vitta behind suture, the inter- vening areas, and especially the median one, greyish pruinescent; pleura with a broad blackish vitta over upper half, divided into two anteriorly, one branch above, the other below the spiracle, and a narrower vitta on upper part of sternopleura; scutellum fuscous, sides appearing darker than disc; postnotum clay-yellow. Abdomen concolorous with thorax, bases of tergites dark brown. Legs clay-yellow, fore and mid femora almost entirely, hind femora and all tibiae at apices, blackened. Wings brownish hyaline. Halteres whitish. All frontal bristles strong and long; arista rather long haired; thorax with three pairs of strong postsutural dorsocentral and one pair of prescutellar acros- tichal bristles; scutellum flattened, bristles subequal. Genitalia of female not spined. Fore femur without anteroventral comb; hind femur without preapical anteroventral bristle; preapical tibial bristle present. Inner crossvein at or slightly before middle of discal cell; penultimate section of fourth vein fully three- fourths as long as ultimate section. Length, 5-6 mm. Type, female, and allotype, Hungella Ra., 45 miles west of Mackay, Queens- land, 1400-2400 feet, 25.9.23 (Goldfinch). The specimen listed as allotype is in poor condition and teneral, but evidently belongs to this species. SAPROMYZA PUNCTISETA, Nl. Sp. Male.—Head testaceous; orbits, frontal triangle, and the lines on which the frontal orbital bristles are situated yellowish-white dusted; ocellar spot fuscous; antennae and palpi orange-yellow; arista fuscous, face slightly darker in centre than on sides. Thorax darker than head, and densely lead-grey pruinescent, apex of scutellum and sutures of pleura paler, a smail dark brown dot at base of each hair and bristle on dorsum. Abdomen brown, becoming greyish apically, apices of tergites yellowish. Legs testaceous yellow, fore femora, fore tibiae, apices of mid and hind tibiae, and of all tarsi brown or fuscous, the hind femora at apices and mid and hind tibiae near bases sometimes brownish. Wings hyaline. Halteres yellow. Head as in victoriae. Thorax with three pairs of postsutural dorsocentrals and four series of intradorsocentral setulae, the median pairs quite strong from behind suture to scutellum; scutellar bristles equal. Legs and wings as in victoriae. Length, 4 mm. Type and two paratypes mounted on same card, Victoria (C. French). Brit. Mus. SAPROMYZA VICTORIAE, N. SD. Male.—Yellow testaceous, abdomen slightly shining. Head with whitish pruinescence on the narrow frontal orbits and on the lines of orbital bristles, the two stripes separated by a line of the reddish-yellow frontal colour, the frontal triangle forming a narrow whitish streak to anterior margin; face paler than 318 NOTES ON AUSTRALIAN DIPTERA, Vii, frons, whitish pruinescent on sides, and with a faint dark vertical central line; antennae reddish-yellow; arista fuscous; palpi yellow. Thoracic dorsum with four broad brownish vittae, the outer one on each side between submedian one and lateral margin, the median pair continued over scutellum; pleura not vittate. Apices of abdominal tergites a little paler than bases. Legs yellowish testaceous, fore femora, apices of all tibiae and a ring near middle of at least mid and hind tibiae brownish. Wings hyaline. Halteres yellow. Anterior fronto-orbital bristles rather far removed from eye; ocellars small; arista subnude; eye higher than long, narrowed below; cheek not as high as width of third antennal segment; palpi slender. 'Thorax with three pairs of postsutural dorsocentral, and one pair of prescutellar acrostichal, bristles, six series of intradorsocentral setulae present anteriorly. Fore femur without antero- ventral comb; preapical tibial bristle present on all legs, long on fore pair; hind femur without preapical anteroventral bristle. Inner crossvein at middle of discal cell; last section of fourth vein nearly twice as long as preceding section; outer crossvein at about its own length from apex of fifth vein. Length, 4 mm. ‘ Type, Melbourne, Victoria (G. F. Hill). SAPROMYZA SPINIGERA, N. SD. Female.—Reddish testaceous, slightly shining, the surface obscured by yellowish-grey dusting, thoracic dorsum with two faint, slender, brownish, sub- median vittae; abdomen more greyish than thorax; orbits and frontal triangle greyish, much paler than other parts of frons; palpi blackened apically. Wings yellowish-hyaline. All frontal bristles long and strong, surface hairs numerous but short on sides and anteriorly; arista plumose; face with a convexity in centre above mouth; cheek about as high as width of third antennal segment. Thorax with four pairs of strong dorsocentral, and four pairs of quite conspicuous acrostichal, bristles, the anterior pair of both series in front of suture; a short bristle mesad of the presutural intra-alar bristle; an irregular series on short hairs between the acrostichals and dorsocentrals; scutellum flattened above, bare on disc, bristles subequal; both sternopleurals strong. Abdomen stout, third visible tergite with the bristles on lateral portions of hind margin denser and longer than usual; fourth tergite with quite dense long black setulose hairs or bristles on entire length of exposed portion at the lateral curves which are directed upward and out- ward; genital segment not spinose. Fore femur with an anteroventral comb; hind femur without an anteroventral preapical bristle; preapical tibial bristle present. Inner crossvein at middle of discal cell; sixth vein almost complete, faint apically. Length, 5 mm. Type, Ararat, Victoria (G. F. Hill). An aberrant species, possessing as it does a bristle mesad of the presutural intra-alar, and a very long sixth vein. SAPROMYZA HIRTIVENTRIS, 0D. Sp. Male and female.—Similar to preceding species in colour, differing in having the stripe between the faint submedian vittae grey or fuscous, the palpi yellow, frontal orbits and triangle less definitely greyish, and the abdomen more greyish fuscous. Tarsi dark at apices. Arista sparsely pubescent; frons without fine surface hairs; face not convex. Thorax with the same dorsocentrals as in last species, but the acrostichals, except BY J. R. MALLOCH. 319 the prescutellar pair, very short, the intradorsocentral area with four series of hairs, short in front; extra bristle mesad of the presutural intra-alar present. Female with a large section in middle of second and third visible tergites furnished with rather dense, fine, moderately long, erect, black hairs, the other hairing and bristling normal. Internal processes of male genitalia (3) long, heavily chitinized, glossy black. Legs and wings as in spinigera. Length, 7-8 mm. Type, female, and allotype, Ararat, Victoria (G. F. Hill). Paratypes, four, Bamawm, Victoria (G. F. Hill). SAPROMYZA FUSCOCOSTATA, Nl. SD. Female.—Shining fulvous yellow. Third antennal segment, arista, and a small ocellar spot, fuscous. Thoracic dorsum with two fuscous vittae along the lines of dorsocentrals. Abdomen with a series of dark central spots apically, one to each tergite. Legs fulvous yellow, fore femora, all tibiae, entire fore tarsi, and apices of mid and hind tarsi, black. Wings yellowish, costa from base to apex of fourth vein, and outer crossvein, narrowly fuscous brown. Halteres yellow. Frontal orbits shining, well separated from eyes anteriorly, the anterior bristle smaller than the posterior one; postvertical bristles long; ocellars minute; face convex; parafacials white pruinose; antennae normal; arista pubescent; cheek about one-fourth as high as eye. Thorax with two pairs of strong and one pair of weak anterior postsutural dorsocentrals, one pair of strong prescutellar acrostichals, and six or eight series of intradorsocentral setulae; scutellum convex, the bristles equal; propleural bristle strong. Abdomen normal. Fore tibia with very poorly developed anteroventral preapical comb; fore tarsi not noticeably thickened; preapical tibial bristle present. Inner crossvein at middle of discal cell; ultimate section of fourth vein about equal to penultimate one. Length, 6-7 mm. Type and one paratype, Mittagong, N.S.W. SAPROMYZA SCIOMYZINA Schiner. This species was recorded from New Zealand by Schiner in his original description. This is an error similar to that which he made in his locality for Poecilohetaerus schineri Hendel (= decora Schiner), both of which were really from Australia. I unfortunately did not take sciomyzina into consideration when I described atriventris mihi in a recent paper on Australian Diptera, as no species that I have seen from the two regions are identical, and thus created a synonym, as the two names apply to the same species. Genus Sapromyzosoma Malloch. This generic name had been used by Lioy many years ago, but I made use of it intentionally a year or two ago because it may possibly be the same as Lioy’s and I am in doubt as to whether van der Wulp’s name Homoneura is applicable to the same group. If my concept and Lioy’s are identical the name Sapromyzosoma will remain, and if they are different, and my concept is the same as Homoneura, then the latter will replace Sapromyzosoma, a rather cumbersome name at best. I believe that I have correctly identified one of Kertesz’s species, described from New Guinea, amongst those sent to me and redescribe it briefly herein, using characters in large part not mentioned in the original description. Black species 320 NOTES ON AUSTRALIAN DIPTERA, Vil, are very rare in this genus and the one now listed, signatifrons, has very much the appearance of a Minettia, mimicking, rather closely, longipennis Fallén, the genotype of that genus, except that the wings are not blackened at bases. SAPROMYZOSOMA PROXIMELLA, Ni. SD. Male.—Testaceous yellow, subopaque. Apical halves of palpi, ocellar spot, arista, and a mark on middle of occiput at neck, black; orbits not shining; antennae yellow. Thorax greyish pruinescent, with a pair of faint brownish vittae along the lines of dorsocentral bristles, and a brownish mark alongside the upper margin of each humeral callosity. Abdomen unmarked. Legs clay-yellow. Wings hyaline, marked as in horvathi Kertesz, both the crossveins and apices of second, third, and fourth veins with brown clouds, the one on outer crossvein most distinct on upper extremity, spotlike, those on apices of second and fourth veins distinctly proximad of the tips, that on third at tip. Halteres yellow. Frons fully one-third of the head width; orbits not differentiated, the bristles long; ocellar pair long, parallel; arista plumose, longest hairs nearly as long as width of third antennal segment. Thorax with three pairs of strong postsutural dersocentral bristles, and one pair of prescutellar acrostichals; intradorsocentral setulae in about six series; scutellum flat, bristles equal, apical pair cruciate. Fore femur with an anteroventral preapical comb. Inner crossvein of wing a little beyond middle of discal cell; last section of fourth vein very little longer than preceding section; outer crossvein at about half its own length from apex of fifth vein. Length, 3 mm. Type, Townsville, N. Queensland (G. F. Hill). This species is very similar to horvathi Kertesz and brevicornis Kertesz, the wings being almost identical in venation and markings in all three (as Fig. 9). S. horvathi has the frons with a central blackish line, the third antennal segment blackened at apex, thorax more obviously vittate, and the abdomen with blackish fasciae at hind margins of the tergites. I have many specimens of brevicornis from Formosa before me; they all have the arista much shorter haired, the longest hairs being no more than one-third as long as the width of third antennal segment, and the thorax is even less noticeably vittate than in proxvimella. Kertesz described horvathi from New Guinea, brevicornis from Formosa. SAPROMYZOSOMA PREAPICALIS, nN. Sp. Text-figure 9. Male and female.—Similar to proximella, but the palpi are yellow, the thorax is less evidently dusted and without any traces of dark vittae, and the insect is larger. Wing as in Figure 9. Length, 4-4.5 mm. Type, male, allotype, and five paratypes, Waterfall, N.S.W., January, 1925 (H. Petersen). SAPROMYZOSOMA ARMATA, Nl. SD. Female.—Testaceous yellow, shining. Third antennal segment, arista, and apices of palpi, black; abdomen sometimes more or less fuscous in part, but this colour is evidently a discoloration. Wings hyaline, inner crossvein sometimes slightly darker than the other veins. Halteres yellow. All frontal bristles strong, no evident surface hairs present; arista distinctly pubescent; face concave in profile. Thorax with three pairs of strong postsutural dorsocentral, and one pair of prescutellar acrostichal, bristles, the intradorso- BY J. R. MALLOCH. 321 central setulae in about eight series; scutellum fiat on disc, bristles equal. Abdomen stout, genital segment armed with about eight short stout bristles. Fore femur with preapical anteroventral comb; preapical tibial bristle present; inner crossvein distinctly beyond middle of discal cell; outer crossvein at not more than half its own length from apex of fifth vein; penultimate section of fourth vein about two-thirds as long as ultimate section. Length, 5-7 mm. Type, Ararat, Victoria (G. F. Hill). Paratypes, Seaford, Victoria (W. F. Hill); East Australia (Dr. T. P. Lucas; Brit. Mus.). A male specimen which appears to belong to this species is from Hidsvold, Queensland. In this specimen the last section of fourth vein is barely longer than the preceding section. This species agrees very well with the description of fuscicornis Macquart, but the antennae are given as black in that species and there is no mention of the colour of the palpi, which leads one to assume that they are yellow in fuscicornis. SAPROMYZOSOMA SIGNATIFRONS (Kertesz). Male—Glossy black. Frons subopaque, brownish-black, the anterior margin narrowly testaceous yellow, antennae brownish-yellow, arista fuscous; face slightly shining, with white pruinescence; palpi brownish. Thorax not vittate, almost without pruinescence. Legs fuscous, tibiae and tarsi testaceous yellow. Wings hyaline. Knobs of halteres black. All frontal bristles strong; antennae normal; arista plumose; face slightly convex in middle vertically; cheek about as high as width of third antennal segment. Thorax with three pairs of strong postsutural dorsocentral, and one pair of prescutellar acrostichal bristles; intradorsocentral setulae in 10-12 series; scutellum flattened on disc, not transverse at apex, bristles equal; anterior sterno- pleural short. Abdomen stout. Fore femur with preapical anteroventral comb; preapical tibial bristle present on all legs, weak on hind pair. Inner crossvein a little before middle of discal cell; outer at about half its own length from apex of fifth vein; penultimate section of fourth vein more than three-fourths as long as ultimate. Length, 3-4.5 mm. One male, Melville Is., N.T. (G. F. Hill). SAPROMYZOSOMA EIDSVOLDENSIS, 0. Sp. Female.—Head yellowish testaceous. ocellar spot, palpi, and third antennal segment broadly, black; sides of interfrontalia darker than centre; arista dark. Thorax darker than head, largely dark greyish, and with grey pruinescence, centre of dorsum broadly grey, the edges of the grey stripe each with a dark line, a dark brown mark along the inner side of each humeral callosity, faint indications of sublateral vittae posteriorly, and dark brown dots at bases of all bristles and hairs, largest at bases of the bristles; humeri and scutellum yellow testaceous, the scutellum dark on sides, and with two faint discal marks. Abdomen testaceous yellow, with a central spot on each of the apical two or three tergites, and a fascia on the hind margin of each, black, the fasciae roundly widened on each side on dorsum and narrowly connected with the black lateral margin of each tergite below; sternites clay-yellow. Legs testaceous yellow, femora sometimes darkened basally, each femur with a black spot near apex below, opposed to a similar spot 322 NOTES ON AUSTRALIAN DIPTERA, Vii, on each tibia near base. Wings hyaline, crossveins very faintly clouded. Halteres yellow. Frons one-third of the head width, surface with a few short hairs, all bristles long and strong; antennae normal; arista distinctly pubescent; cheek about as high as width of third antennal segment. Thorax with 1 + 3 dorsocentrals, one pair of strong prescutellar acrostichals, and six series of intradorsocentral setulae; both sternopleurals strong, rather closely placed; scutellum flattened, bristles equal. Abdomen normal. Fore femur with weak preapical anteroventral comb; all tibiae with preapical bristle. Inner crossvein almost at middle of discal cell; outer crossvein about half its own length from apex of fifth vein. Length, 3.75 mm. Type and two paratypes, Hidsvold, Queensland, 20.4.24 (Bancroft). SAPROMYZOSOMA FUMIFRONS, 0. Sp. Male and female.—Tawny-yellow, shining. Frons infuscated except on anterior margin, face similarly coloured on sides, and above in centre; antennae fuscous, a little paler sometimes at base of third segment; arista black; palpi yellow. Thorax without markings. Abdomen with the hind margin of each tergite black, all except the first, and sometimes second, in male, entirely infuscated, general colour paler in female. Legs yellow. Wings hyaline, both crossveins and apex of second vein conspicuously clouded with fuscous, the mark on outer crossvein broadest on its upper extremity; the third vein with an almost indistinguishable cloud at apex. Halteres yellow. All frontal bristles long and strong; orbits poorly differentiated; surface hairs microscopic; antennae normal; arista with its longest hairs not as long as width of third antennal segment; cheek not as high as width of that segment, the hairs fine and short. Thorax with three pairs of strong postsutural dorso- centrals, one strong pair of prescutellar acrostichals, and 6-8 series of intradorso- central setulae; scutellum flattened, bristles equal; both sternopleurals distinct. Abdomen ovate, bristles weak. Fore femur with weak anteroventral comb; hind femur without distinct anteroventral bristles; tibiae all with preapical bristle. Inner crossvein beyond middle of discal cell; first posterior cell not narrowed at apex. Length, 4-4.5 mm. Type and two paratypes, Como, N.S.W., Dec., 1923; allotype and many para- types, Waterfall, N.S.W., Jan., 1924 (H. Petersen). Distinguished from its allies by the dark frons, face, and antennae, and the tripunctate wing. Se a ns, Se TE rym maT TINN 1 1 Text-figure 9.—Sapromyzosoma preapicalis, n. sp. Wing. Text-figure 10.—Australina geniseta, n. sp. Profile of head. Text-figure 11.—Australina geniseta, n. sp. Wing. BY J. R. MALLOCH. 323 Genus AUSTRALINA novum. Text-figures 10, 11. Generic characters.—Runs to Prosopomyia Loew in Hendel’s key to the genera of this family in “Genera Insectorum”. It differs from this, and in fact from any related genus in possessing two very conspicuous bristles on a slightly tumid area on lower part of occiput, in having a prominent elevation in centre of face, and the frons very narrow for this family, about three times as long as wide, slightly wider in front than at middle, and not more than one-fourth of the head width. The postvertical bristles are small but distinct, ocellars microscopic, verticals and orbitals long, the latter both curved backward; profile of head as in Figure 10; the arista distinctly pubescent. Thorax as in Sapromyza; the pre- sutural dorsocentrals absent; prescutellar acrostichals present. Venation of wing as in Figure 11, the same as in Sapromyzosoma. Tibial preapical bristle present. Genotype, the following species. AUSTRALINA GENISETA, n. sp. Text-figures 10, 11. Male and female.—Frons yellowish, densely white dusted except on two narrow, submedian longitudinal vittae, ocellar spot fuscous; face, cheeks, and lower half of occiput, black, a grey pruinescent mark on each parafacial below base of antennae, and one in centre of face above the protuberance, tip of the latter shining; upper part of cheek behind, including the part upon which the strong genal bristles are situated, yellow; antennae and palpi dusky yellow. Thorax blackish-brown, pale on dorsum, where it is densely whitish-grey dusted and has four dark vittae, the submedian pair paler and narrower than the sub- laterals and becoming subobsolete posteriorly; lateral margins of mesonotum dark brown; scutellum yellowish, brown on sides. Abdomen brownish testaceous. Legs black, tibiae and tarsi stramineous. Wings conspicuously marked with dark brown (Fig. 11). Halteres dull-yellow. Frons with short sparse stiff hairs. Thorax with three pairs of postsutural dorsocentrals, and six series of intradorsocentral setulae; scutellum slightly flattened on disc, the four bristles equal; propleural, mesopleural, and both sterno- pleural bristles strong, the latter rather closely placed. Hypopygium of male large. Fore femur without an anteroventral comb; hind femur with two or three preapical anteroventral bristles. Length, 4-4.5 mm. Type, male, and allotype, Darwin, N.T., the male labelled “on Pandanus”, the female has an empty puparium mounted with it, and bears a label as follows: “pupated 6.2.14, emerged 17.2.14, associated with coccids on Pandanus” (G. F. Hill). As extremely little is known of the habits of the larvae of this family, despite the frequency of the occurrence of the adults, this record is of considerable interest. The puparium is rufous-brown in colour, and opaque. The anterior extremity is as wide as the average of the whole, distinctly flattened dorsoventrally as in many Drosophilidae, the small subsessile anterior respiratory organs each have a large number of pale papillae irregularly arranged at their apices. The body surface appears to be devoid of protuberances, at least on dorsum and sides, and the apical three or four segments are much constricted, forming a stout tapered tail, the divisions between the apical slender segments quite pronounced. Posterior spiracles not elevated. 324 NOTES ON AUSTRALIAN DIPTERA, Vii, Genus INCURVISETA novum. Generic characters.—This genus has the anterior pair of fronto-orbital bristles incurved, much as in Camptoprosopella Hendel and Poecilohetaerus Hendel, but the general habitus and several characters of the chaetotaxy distinguish it from either of these genera. Poecilohetaerus has four equally strong dorsocentrals, the anterior pair in front of the suture, and the head is differently shaped. In the other genus there are but two pairs of postsutural dorsocentrals and the arista is plumose, the hairs on the upper side being much longer than those on the under side, while the anterior pair of incurved orbital bristles are very close to the posterior pair, which is not the case in Incurviseta. The wing in the latter is the same as in Sapromyza, and the arista is pubescent. Sternopleurals two in number. Genotype, Sapromyza maculifrons Macquart. INCURVISETA MACULIFRONS (Macquart). I have already recorded this species from Australia. The specimen upon which I based my identification was in rather poor condition, the frontal bristles being partially destroyed, and the wings lacking their tips. I provisionally placed it in the genus Sapromyzosoma, but I have seen other species with the same characters as the genotype and have decided that it is entitled to distinct generic status. Family Ephydridae. I defer publishing a generic synopsis of the Australian Ephydridae until I have seen a better representation of the species. Meantime I present descriptions of some of the genera and species that have been in my possession for some time. Subfamily NorirHiIninae. This subfamily as treated by most authors contains a large number of genera which are rather unsatisfactorily distinguished from the other groups by the presence of a bristle at apex of second antennal segment. I consider that the presence of strong bristles on the dorsal surface of mid tibia is a more reliable character for distinguishing the group, and if this is accepted the number of known genera will be reduced to less than half a dozen, two of which are represented in material now before me, viz. Paralimna and Notiphila. I present data on the former below. Genus PARALIMNA Loew. Mid tibia with three or four long strong bristles on dorsal surface extending from near base to near apex; thorax in the Australian species with 1 + 3 dorso- central bristles, and a pair of long prescutellar acrostichals which are nearly in line with the anterior pairs of dorsocentrals and appear to belong to that series; mesopleural bristles 1 to 3; sternopleural 1; frons with a pair of long strong bristles proximad of, or in transverse line with, anterior ocellus; arista plumose; face with bristles on sides which are rather far removed from eyes, central eleva- tion quite prominent just below antennal insertions, the antennal foveae quite pronounced. I present a diagnostic key for the identification of the three species now known to me, none of which I have identified as already described. BY J. R.. MALLOCH. 325 Key to Species. 1. Bach frontal orbit with one strong backwardly curved bristle at middle, no short bristles between it and eye; the pair of long frontal bristles a little behind anterior ocellus transversely; ocellar and postvertical bristles indistinguishable ; each humerus with but one bristle; abdomen without a dark dorsocentral RUA GLE 1 aes raat teach MENG RCH DO SRS Eine rhe Din VORE RRA TCE CEG tt OOK HOR OROTD REE co DERM SE Die uniseta, N. Sp. Hach frontal orbit with a long backwardly curved bristle about middle and between it and eye a pair of much shorter bristles which are forwardly directed; the pair of long frontal bristles distinctly proximad of anterior ocellus; ocellar bristles minute; abdomen with a distinct dorsocentral dark vitta connecting up the GARE BLAS CLAY hae iced ohare ere a ee anariee duck e nae taie Holand ome avepeaeeebtratcusahs recent fave tebe kate cacddteky coisierawe 2 Thoracic dorsum with a pair of subcontiguous dark central vittae, the large bristles set in dark dots, the fine hairs without dark dots at bases; each humerus with a setula above the strong bristle which is directed mesially; fore tarsi OVC Kaiti ae ai enahescciravte see faaerask strani acei at Yosken aye bane ra verre cleelae rearep a eel ve won Avoway sisuabeiet ae eae oe atrimana, Nn. sp. Thoracic dorsum without dark dorsocentral vittae, but with very distinct dark dots at bases of all hairs and bristles; each humerus with but one bristle; fore tarsus WMO TORE OH lekeyl Seem WONOW sacoccacaccocanoccdnsede millepuncta, n. sp. to PARALIMNA UNISETA, N. Sp. Male.—Head black, densely yelllowish-brown dusted; interfrontalia lavender- grey pruinescent, the triangle concolorous with rest of head; antennae and palpi black. Thorax densely tawny-brown pruinescent, paler on sides of mesonotum behind suture. Abdominal tergites greyish pruinescent, bases and apices narrowly brown, the apices very narrowly so. Legs black, femora slightly grey pruinescent. Wings hyaline, veins fuscous. Halteres whitish, stems yellow. A pair of long parallel forwardly projecting bristles almost in line with the posterior margin of anterior ocellus; but one strong bristle on each orbit proximad of middle, some microscopic hairs in front of bristle and on the anterior margin ef interfrontalia; frons wider than long and nearly half the head width; face rather abruptly protuberant below bases of antennae, almost vertical from that point to lower margin; about three bristles and some hairs on each parafacial; arista with about 10 rays above; eye higher than long; cheek not more than one- third of the eye height, genal bristle distinct. Thoracic chaetotaxy normal; intra- dorsocentral hairs very short and fine, in about 10 series; scutellum flattened on disc. Posteroventral bristles on fore femur mostly as long as femoral diameter; anterodorsal setulae on hind tibia distinct. Base of section of costa proximad of apex of first vein with about three outstanding bristles, and one at apex; last section of fourth vein about four-fifths as long as preceding section. Length, 3 mm. Type, Fish River, N.S.W., 25.3.1923. PARALIMNA MILLEPUNCTA, DN. SD. Female.—Black, opaque, densely pale-grey pruinescent. Frons largely dark brown, not spotted, but irregularly suffused; face slightly suffused with paler brown; antennae and palpi black. Thoracic dorsum with dark brown dots at bases of the hairs and bristles, the dots more or less aggregated, those at bases of the strong bristles on mesonotum and scutellum larger and darker; mesopleura with an irregular brown suffusion in middle. Abdomen with a chocolate-brown dorsocentral vitta which is connected with the anterior and posterior fasciae on each tergite, the anterior fasciae narrow in centre and broader on sides; a blackish spot over each spiracle. Legs black, grey dusted, basal segment of fore tarsi and basal two segments of mid and hind pairs yellow. Wings hyaline. MHalteres yellowish. 326 NOTES ON AUSTRALIAN DIPTERA, Vii, Frons rather noticeably short haired, a pair of bristles almost in line with hind margin of posterior ocelli; interfrontal bristles well in front of anterior ocellus; the two bristles between orbitals and eye short and quite close together; face much as in wniseta, but with one moderately long, and one shorter, bristle on each side below tips of antennae, and below and laterad of these some fine hairs; cheek about one-third of the eye height, genal bristle stout; arista with 9-10 rays above. Thorax as in uniseta, but the fine hairs less regularly arranged and much more sparse, especially on pleura and sides of mesonotum. Fore femoral bristles sparser and finer, and distinctly shorter than in wniseta; hind tibia with a few noticeable posterodorsal setulae. Costa lacking the three or four bristles beyond the basal break which are present in last species. Length, 5-6 mm. Type and 8 paratypes, Eidsvold, Queensland; 1 paratype, Narrandera, N.S.W. PARALIMNA ATRIMANA, Nl. SD. Male.—Differs from preceding species in being more lead-grey in colour of the pruinescence; in having the thoracic dorsum with faint dark dots at bases of the large bristles only, a distinct dark dorsocentral vitta; the abdominal markings narrower, no spot over the spiracles; and the fore tarsi entirely black. The face has three larger bristles on each side, the orbital bristles are longer, there is a fine upcurved bristle above the humeral bristle, and there are one or two costal bristles beyond the basal break as in wniseta. Length, 3-4 mm. Type and one paratype, Belaringar, N.S.W., 1.6.23; one paratype, Collaroy, Sydney, N.S.W., 29.1.24; three paratypes, Narrandera, N.S.W.; one paratype, Hidsvold, Queensland. Genus NoTirHiLa Fallén. This genus is distinguished from Paralimna by the shorter costal vein, which ends at, or just beyond, apex of third vein; and the presence of 1 + 2 dorsocentral bristles on thorax. I present below the description of one species. NovirHiILAa FUSCIMANA, 0. Sp. Male and female.—Head black, with brownish-yellow pollen, darker on inter- frontalia and triangle in front of anterior ocellus; antennae yellow, third segment brownish above; arista fuscous; palpi yellow. Thorax black, brownish pollinose on dorsum, more greyish and with a slight olive tinge on pleura, mesonotum with a pair of faint, darker, median vittae anteriorly, narrowly separated by a pale line; the bases of the large bristles inserted in small black spots; mesopleura with a brown central mark. Abdomen greyish pruinescent, each tergite with four large, irregular, brownish, or fuscous, spots, the bristles and hairs inserted in dark dots. Legs tawny-yellow, coxae, femora except apices, and the fore tarsi fuscous, fore tibiae of male dark apically. Wings hyaline, veins dark. MHalteres yellow. One or two fine hairs in front of the anterior orbital bristle; face with 2 or 3 bristles on lower half, and many fine hairs which extend above middle, of the sides; genal bristle strong. Hind tibia with one or two setulae on basal half of posterodorsal surface. Costa with one long and one short bristle at apex of first vein. Length, 3-4 mm. BY J. R. MALLOCH, 327 Type, female, Fish River, N.S.W., 25.3.23. Allotype and one paratype, Eids- vold, Queensland; three paratypes, Narrandera, N.S.W. A pair of paratypes from Murray Bridge, S.A., have the dorsum of thorax almost uniformly olive-brown, but do not differ in any other respect from the type material. There is at least one other species of the genus in my hands yet. Subfamily ErpHypRINAR. Genus HYDRELLIA Rob.-Desv. HYDRELLIA TRITICIT Coquillett. Black, distinctly shining. Frontal triangle and lunule connected, both glossy; orbits shining, a stripe between triangle and lunule, and the orbits opaque velvety- black; face and cheeks silvery-white pruinescent; antennae and palpi black. Thorax with a slight olive tinge on dorsum, a large velvety-black spot on each side in front of wing base; pleura white dusted. Abdomen with a greenish or bronzy tinge, very faintly dusted. Legs fulvous, variably marked with black, the fore tibiae almost or entirely, and fore tarsi entirely black, and mid and hind tarsi dark at apices, sometimes apices of fore femora, bases of mid tibiae and middle of hind tibiae blackish. Wings greyish hyaline. Halteres lemon-yellow. Frons about half of the head width, broader than long, declivous from middle forward; ocellar bristles minute; postverticals long; upper orbital bristle directed over eye, twice as long as the lower forwardly directed one; face narrowest about midway from antennae to lower margin, with about three bristles on each side; genal bristle fine; cheeks narrow; arista rayed above, bare below. Thorax with two pairs of dorsocentrals, the anterior pair close to suture; mesopleura and sternopleura with one bristle each; scutellum flat above, with four marginal bristles and two short fine lateral marginal hairs. Abdomen slender. Legs slender, fore tarsi flattened a little apically. Inner crossvein about one-fourth to one-third from base of discal cell and a little before or beyond apex of first vein; penultimate section of fourth vein about five-sixths as long as ultimate section. Length, 2-2.5 mm. Originally described from Australia, this slender species appears to be very common in New South Wales, as I have many specimens from Botany Bay, Como, and Sydney. The type specimen is in the United States National Museum. Genus ItyrHEesA Haliday. Generic characters.—Mid tibia without outstanding dorsal bristles; costa to apex of fourth vein; a pair of quite long divergent bristles situated within the ocellar triangle, none on interfrontalia proximad of anterior ocellus; each orbit with two bristles, the upper one sloping backward and outward, the lower one forward; second antennal segment with a short apical bristle; arista long rayed above; face convex, with distinct pruinescence, a short but pronounced pro- tuberance above middle, and a number of long hairs or setulae on sides which stand clear of the narrow parafacials; mouth large; labrum not exposed; genal bristle small. Thoracic chaetotaxy as in Paralimna. Venation as in Figure 12, but the outer crossvein usually present. ILYTHEA DEFECTA, Nn. sp. Text-figure 12. Female.—Shining black, with sparse brownish dusting, that on face more golden-brown and dense. Antennae below, mouth-parts, legs, and halteres dusky yellow. Wings with fuscous markings as in Figure 12. B 328 NOTES ON AUSTRALIAN DIPTERA, Vii, Each side of face with three or four long setulose hairs, the upper one of each series curved upward, and between them and eyes some short hairs; frons about half the head width, much shorter than wide, concave in front. A few hairs between dorsocentral bristles; basal pair of scutellar bristles shorter than apical pair. Mid femur with two or three anterior bristles on apical half. Outer crossvein absent in type specimen. Length, 1.75 mm. Type, EHidsvold, Queensland, 2.4.24 (Bancroft). This species is very similar to Ilythea flavipes Cresson described from Costa Rica, but in it the outer crossvein is present. 14 Text-figure 12.—Ilythea defecta, n. sp. Wing. Text-figure 13.—EHphydra breviseta, n. sp. Male hypopygium. Text-figure 14.—Hphydra acrostichalis, n. sp. Male hypopygium. Text-figure 15.—Scatella australiae, n. sp. Male hypopygium. Text-figure 16.—Scatella alticeps, n. sp. Head. Genus EpHypra Fallén. I have before me a number of specimens referable to this genus in the widest sense. They differ from the genotype and other Huropean species in having the posthumeral bristle lacking, and the thoracic dorsocentrals, except the posterior two pairs, almost indistinguishable from the hairs. In the former character they are similar to the New Zealand forms which I have seen, but the latter have five pairs of dorsocentrals which are all strong. It may be necessary to erect a new subgenus for the Australian species, of which there appear to be two in my material. The genus Ephydra may be distinguished from its allies by the protuberant bristly face; large mouth-opening; concealed labrum; almost bare arista; two pairs of long outwardly curved orbital bristles; lack of spines on mid tibiae and second antennal segment; costa extending to fourth vein; normal abdomen; almost straight, long, tarsal claws; and minute pulvilli. There are five pairs of dorso- centrals on thorax, and at least one mesopleural and one sternopleural present. The two species now available may be distinguished as below. BY J. R. MALLOCH. 329 Key to Species. il, IMI IGG Sw dteedittate'c ale Siees ric Slate eich c soa raie aici CHG ER ORONS erorb ae seec re mate IE SRE MCE OSG IC nee ere 2 TENSTIO EN KES oi orateuchona Oh Onch aie OnGncgERO Kaa CREM GicuP aay URERON tai Cunha WAM CRU boa tet MRC Mit MM Ab 3 Zecescutellarvacrostichals not) developed’ 7. ..5-+ 45 -Hes see ee eee eee. breviseta, n. sp. Prescutellar acrostichals well developed .....................-. acrostichalis, n. sp. 3. Prescutellar acrostichal region quite obviously tumid .............. breviseta, n. sp. Prescutellar acrostichal region not abnormal, regularly convex .... acrostichalis, n. sp. EXPHYDRA BREVISETA, nN. Sp. Text-figure 13. Male and female.—Black, frontal triangle, disc of thorax, and most of dorsum of abdomen, shining, with a more or less metallic-greenish lustre, the abdomen sometimes with purple or violet reflections. Frontal and facial orbits and cheeks greyish, centre of face brownish dusted, interfrontalia opaque, blackish; antennae and palpi fuscous. Thorax with lateral margins of mesonotum and four faint discal vittae greyish; pleura greenish-grey dusted. Abdomen except dorsum, and most of legs largely greenish-grey dusted; knees, and mid and hind tarsi, brownish. Wings greyish hyaline. Halteres yellow. Frontal triangle with all the hairs short. Thorax in male without prescutellar acrostichals, the surface normal, in female with well developed acrostichals which are situated upon an obvious rounded elevation; only the posterior dorsocentrals strong; scutellum short, rounded, convex, the basal bristles shorter than the apical pair. Male hypopygium as in Figure 13. Length, 5-6 mm. Type, male, allotype, four male and one female paratypes. Woy Woy, N.S.W., 2.9.23 (Mackerras); one male paratype, Mosman, N.S.W., 24.9.22. EFPHYDRA ACROSTICHALIS, n. sp. Text-figure 14. Male and female.—Similar to the preceding species, but the dorsum of abdomen less shining, the frontal triangle comparatively narrower, and the male hypopygium as in Figure 14. Length, 5-6 mm. Type and allotype, on same mount, Middleton, S.A. The abdomen of the type male of this species and those of two paratypes of the other were boiled in a 10% solution of caustic potash to facilitate delineation, and afterwards placed on the same mounts with the specimens. Genus SCATELLA Rob.-Desv. This genus has the head similar to that of Hphydra, the orbitals being directed outward over eyes, the face protuberant and setulose, second antennal segment without a spine, arista pubescent, mouth-opening large, labrum not exposed, and genal bristle present. The humeri have no bristles, the dorsocentrals are in two or three pairs, in the latter case the anterior pair are almost in transverse line with a pair of long strong acrostichals which are situated close to suture, this last feature characteristic of the genus. The tarsal claws are not very long, and are distinctly curved, while the pulvilli are well developed. Costa to apex of the fourth vein, wings usually with five or more hyaline spots between the veins, which are most readily seen when the wing is viewed from the base looking obliquely along the surface to apex. None of the Australian species possess any outstanding structural peculiarities. The hypopygia of the males possess a peculiar feature in this genus, the claspers being connected about middle by a chitinous band which arches below as shown in Figure 15. I present a key for the separation of the Australian species known to me. 330 NOTES ON AUSTRALIAN DIPTERA, Vii, Key to Species. 1. Only two strong pairs of dorsocentral bristles present, the one in line with the strong acrostichals close to suture very weak, represented by fine short hairs .. 2 Three pairs of strong dorsocentrals present, the anterior pair almost in line with the strong acrostichals close to suture, and as long as or longer than these ...... 4 2. Face silvery-white; cheek about half as high as eye, the genal bristle small and weak; second visible abdominal tergite in male not half as long as third, fourth about twice as long as latter; wings faintly spotted ............ alticeps, n. sp. Face yellowish-brown; cheek much less than half as high as eye, the genal bristle rather conspicuous; second visible abdominal tergite in male not noticeably shorter than third): wings squitesdistinctlya spotted ame sites eeieieneisioer steiner 3 3. All hyaline spots on wings very distinct, the one beyond outer crossvein in first posterior cell almost or entirely divided in middle, showing usually as a small spot against third and another against fourth vein, both these veins distinctly arcuated at that point; scutellum convex; setulae in front of anterior dorso- CENLTAISHVerR yews opere ois ene eee asta Gotha enero race ee eine ce nitidithorax, n. sp. All hyaline spots on wings faint, the one beyond crossvein in first posterior cell not touching veins, entire, the veins not at all or very faintly arcuated at this point; scutellum flattened on disc; setulae in front of anterior dorsocentrals strong, but much weaker than anterior acrostichals .................. australiae, n. sp. 4. Frontal triangle shining, with slight dusting; thoracic dorsum with distinct vittae on anterior half; legs black; wings quite conspicuously spotted .... vittithorax, n. sp. Frontal triangle not at all shining, densely yellowish-grey pruinescent like mesonotum, the latter without vittae; tarsi largely yellowish, darker at apices; wings faintly SPOELCA) — eeshewavswwrsrensmeuste eters naveue Sto a, eS sou SEIT Ne eee reat ee eae CS IE immaculata, n. sp. SCATELLA ALTICEPS, n. sp. Text-figure 16. Male.—Frons fuscous, triangle almost glossy, ocellar spot and the narrow orbits opaque-brown; face and cheeks densely white dusted, almost silvery; antennae and palpi black. Thorax black, disc shining, darker on each side, lateral margins and pleura grey dusted; scutellum concolorous with mesonotum, apex slightly white dusted below. Abdomen black, with a slight greenish tinge, only slightly shining owing to the presence of quite dense brown-grey dusting. Legs black, tarsi brownish, femora grey dusted. Wings brownish-grey, the clear spots very faint. Halteres yellow. Head as in Figure 16; a pair of divergent hairs behind posterior ocelli. Thorax with 3-4 fine hairs before anterior dorsocentrals, three pairs behind and two behind the acrostichals, the hindmost pair closer together than the preceding pair; scutellum slightly flattened on disc, basal hairs short. Fore femoral bristles short and fine; two or three long hairs at base of basal segment of fore tarsus below. Third vein slightly arcuate near middle of apical section. Type, Collaroy, Sydney, N.S.W., 10.9.21. SCATELLA NITIDITHORAX, Nl. Sp. Male and female-——A much darker species than the preceding one, and very like stagnalis Fallén of Europe both in structure and wing markings. Face and cheeks yellowish-brown dusted, the thorax more distinctly shining on disc than in alticeps and with two rather evident grey dusted vittae anteriorly, the scutellum without grey dust at apex below, pleura brown dusted, abdomen much less dusted and more shining on dorsum, the legs black, and the spots on wings distinct. The cheeks are not more than one-sixth of the eye height, with a stronger and longer genal bristle than in alticeps. Thorax with similar setulae, but there are no postsutural acrostichals in any specimen before me, and the scutellum is not flat, but is convex, on disc. The division of the pale spot in first posterior cell of BY J. R. MALLOCH. 331 wing is coupled with a rather distinct arcuation of third and fourth veins at position of the spot. Fore tarsi of male without fine hairs at base below. Length, 2-3 mm. Type, male, allotype, and two paratypes, Sydney, N.S.W. Two specimens which may belong here or to a closely related species have the spot in first posterior cell complete, and the veins very slightly arcuated. Localities.—Tarro, Hunter River, and Belaringar, N.S.W. SCATELLA AUSTRALIAE, n. sp. Text-figure 15. Male.—A less shining and more brownish species than the last, distinguished as noted in the key. The triangle is very slightly shining; the thoracic dorsum is rather noticeably vittate, and less shining; the postsutural acrostichals are absent; and the third and fourth veins are quite regularly, though slightly, diver- gent on apical sections. Legs black, fore tarsi without fine basal hairs. Hypopygium of male as in Figure 15. Length, 2 mm. Type, Woy Woy, N.S.W., 22.9.23 (Mackerras). SCATELLA VITTITHORAX, 0. SD. Female.—Somewhat similar to nitidithorax in colour and structure, but the frontal triangle is less shining, the thorax more distinctly marked with pale grey lines and dark brown vittae, and less shining. The wings are marked as in that species, but the hyaline spot in first posterior cell is complete, the one in the cell above it is larger, and the veins on either side of both spots are arcuate. Except in the possession of three pairs of strong dorsocentrals the species is similar to nitidithorac. Length, 3-3.5 mm. Type and three paratypes, Sydney, N.S.W. SCATELLA IMMACULATA, DN. Sp. Male and female.—Black, entirely covered with grey dusting, more yellowish- grey on mesopleura and parts of dorsum of thorax, even the frontal triangle with- out any gloss. Legs black, grey pruinescent, tarsi yellowish, darker at apices. Wings very faintly spotted. The fore femur has rather longer posteroventral bristles than in the other species, there are no fine outstanding hairs at base of fore tarsus in male, there are sometimes one or two setulae behind the anterior acrostichals, and there is no arcuation of the third vein, nor of the fourth. Length, 2-2.5 mm. Type, female, allotype (greasy), and one female paratype, Belaringar, N.S.W., 8) 7a Genus LIMNELLIA novum. Generic characters.—Similar to Scatella R.-D., differing in having no out- standing pair of acrostichals near suture, these hairs being almost uniform in length on entire extent of the series. The anterior orbital bristle is very much smaller than the posterior outwardly directed one, and is directed more forward than outward; the face is not so much produced and has about six bristles in an evenly spaced lower marginal series, and two or three on each side of the convexity, the central part having only a few short hairs. Otherwise as Scatella. Genotype, the following species. 332 NOTES ON AUSTRALIAN DIPTERA, Vii, LIMNELLIA MACULIPENNIS, n. sp. Text-figure 17. Female.—Head black; frons brown, with a grey pruinescent spot on each side of ocelli at vertex, and three on anterior margin; face white dusted, most noticeably so in antennal grooves and on sides; antennae and palpi fuscous. Thorax and abdomen black, the former opaque-brown dusted, densely so on dorsum which has a pair of submedian vittae and the sides grey pruinescent; pleura largely grey pruinescent; abdomen shining, conspicuously so apically. Legs black, tarsi yellowish. Wings fuscous, with numerous hyaline spots (Fig. 17). Ocellar, vertical, and upper orbital bristles strong; postverticals absent; arista pubescent; cheek narrow; genal bristle weak. Thorax with two pairs of post- sutural dorsocentral bristles, the acrostichals in two complete series; basal scutellar bristles minute; disc of scutellum flattened. Legs normal. Length, 1.75 mm. Type, Sydney, N.S.W., 29.5.21. Family Phoridae. In the present paper I present a synopsis of the four species of the genus Dohrniphora Dahl known to me at present from Australia. Subsequently I hope to furnish a generic synopsis of the family and keys to the species of each genus I have seen from Australia. The family is a very widely distributed one and contains some species with very remarkable life histories. The Ant Decapitating Fly of North America (Apocephalus pergandei Coquillett), the larvae of which affix themselves to the neck of a certain species of ant, feeding there until they decapitate their host, is among the most interesting biologically, but some of the apterous and semiapterous forms that occur in the nests of ants and termites are amongst the most peculiar in the order in so far as structure is concerned. Genus DowrnipHorA Dahl. This genus is distinguished by the venation of the wings, lack of a series of bristles on third vein, there being normally only one at base of the vein; presence of two strong bristles at base of mid tibia, and lack of long bristles on hind tibia, those that are sometimes present being very short. The genus belongs in the group which has the postantennal bristles reclinate and the tibiae of at least one pair of legs with strong bristles. ‘ The species known to me are found in the larval stages in carrion, dead molluses, decaying fungi, and in garbage. Key to Species. 1. Hind tibiae with several short stout isolated black bristles on anterodorsal and anteroventral surfaces, without short laminate series of bristles on the flattened dorsal stripe; fore tibiae with one long bristle near middle on anterodorsal Sunfacersscutellum swith toursbrnistles i armreieiciicienciciiereienreneienel setitibia, n. sp. Hind tibiae with at most one or two very short bristles on anterodorsal surface near base, the dorsal surface of same tibiae either not flattened, or if so, then with laminate series of short bristles or setulae ................-0--eseeeeeees 2 2. Hind tibiae with a short bristle near base on anterodorsal surface; thorax and abdomen black, the abdomen with pale hind margins to tergites; scutellum with Four bristles! res Pets ONS © EUSA s LES cael ccialie tba ial a) elf redettamneearete) ohemelte nigrita, N. sp. Hind tibiae without any distinct bristles except at apices ...................... 3 3. Scutellum with only two strong bristles; hind tibiae without distinct flattened dorsal stripe, with a rather pronounced dorsal ridge which is evenly setulose; fore tibia with two or three short isolated bristles on anterodorsal surface; thorax and Abdomen’ DIAC cece sia swiss wie Siete are tele teoe eiele Sette eT eT te etna ocr at atratula, n. sp. BY J. R. MALLOCH. 333 Seutellum with four well developed bristles; hind tibiae with very evident dorsal flattened stripe which has numerous laminate series of short black setulae; fore tibia with one long bristle, and beyond it a series of very short setulae, on anterodorsal surface; thorax fulvous-yellow, scutellum black; abdomen fulvous- yellow, conspicuously marked with black .................. nigroscutellata, n. sp. DOHRNIPHORA SETITIBIA, Nn. Sp. Text-figure 18. Male.—Yellowish testaceous, the head more yellow. Abdomen with a small spot on each side of first tergite, a large transverse spot on each side of tergites 2 to 5 inclusive, and a transverse spot on sixth tergite, black; hypopygium largely dark, the apical process yellow. Mid coxae with a large black mark behind; hind femora hardly darkened apically. Wings brownish hyaline. Halteres dusky- yellow. Frons longer than wide, subopaque, with numerous piliferous punctures, the postantennal pair of bristles close together, anterior transverse series of four bristles straight, posterior series with the outer one on each side a little in front of median pair; ocellar region black; antennae normal; arista subnude; clypeus produced as far as anterior margin of frons; one bristle close to lower anterior margin of eyes, and two much longer bristles on posterior portion of each cheek; palpi rather large and wide, setulose at apices; proboscis short. Thorax normal; one bristle above and behind, and two below, prothoracie spiracle, and two on lower margin of propleura; scutellum short, slightly rounded apically, the median pair of bristles a little shorter than the lateral pair. Hypopygium smaller than usual, without long bristles on sides. Fore tibia with a stout median bristle, and beyond it a series of very short setulae, on anterodorsal surface; mid tibia normal except in having one or two very short stout setulae near middle on anterodorsal surface, the dorsal surface flattened and with a few microscopic hairs; hind femur normal; hind tibia flattened on dorsum, with some mieroscopic hairs on flat part, and three or four anterodorsal and anteroventral bristles which are not longer than the tibial diameter. Wing as in Figure 18. Length, 3.5 mm. Type, Sydney, N.S.W., June, 1921. 20 Text-figure 17.—Limnellia maculipennis, n. sp. Wing. Text-figure 18.—Dohrniphora setitibia, n. sp. Wing. Text-figure 19.—Dohrniphora nigrita, n. sp. Wing. Text-figure 20.—Dohrniphora atratula, n. sp. Wing. 334 NOTES ON AUSTRALIAN DIPTERA, Vii, DoHRNIPHORA NIGRITA, n. Sp. Text-figure 19. Male.—Shining black. Antennae except apex of third segment, the palpi, and proboscis, fulvous-yellow. Abdomen opaque-black, the apices of tergites narrowly clay-coloured. Legs obscure tawny-yellow. Wings greyish hyaline, slightly infus- cated on apical half of costal margin. Halteres yellow. Frons wider than long, the setigerous punctures sparse in centre, postantennal bristles strong and divergent, the frontal bristles as in previous species; head otherwise as in that species. Thorax differing from last in having about four bristles on lower margin of propleura, and the median pair of scutellar bristles longest. Hypopygium large, with a few long fine lateral bristles. Fore tibia as in setitibia; mid tibia without the short anterodorsal setulae; hind tibia with but one subbasal short bristle, and with many series of diagonal laminate setulae on the flattened dorsal stripe. Wing as in Figure 19. Length, 3.5 mm. Type, Sydney, N.S.W., 28.10.23. DOHRNIPHORA ATRATULA, n. Sp. Text-figure 20. Male.—Black, shining. Antennae fuscous, palpi yellow. Abdomen with very faint pale hind margins to tergites. Legs fuscous, more yellowish on fore coxae and tibiae and tarsi. Wings greyish hyaline. Halteres yellow. Frons subquadrate, slightly, evenly convex, almost impunctate, and sparsely haired, bristles as in previous species, but the upper series practically straight, head otherwise much as in nigrita. Thorax with two bristles on lower margin of propleura, the scutellum with a strong bristle on each side, and basad of each of these a weak short hair. Fore tibia with two or three short anterodorsal bristles; hind tibia without a flattened dorsal stripe; these characters link the species with mordax Brues, described from Formosa, to which it is closely allied, but it differs very strikingly in colour from that species. Wing as in Figure 20. Length, 1.75 mm. Type, Sydney, N.S.W., 8.1.23. DOHRNIPHORA NIGROSCUTELLATA, N. Sp. Male and female.—Fulvous-yellow, shining. Head black, antennae, face, palpi, and sides of mouth, yellow. Scutellum subopaque-black, the dark colour at times suffusing hind part of mesonotum and postnotum. Abdomen in male with paired black dorsal spots, which sometimes unite centrally leaving only the hind margins of the tergites yellow or testaceous; hypopygium black, apex of apical process yellow; in female the basal two or three tergites are fulvous-yellow, the remainder opaque-black, the black part without distinct chitinous plates on dorsum. Legs yellow, hind femora sometimes dark at apices posteriorly. Wings yellowish, their apices more or less infuscated. Head similar to that of nigrita; the clypeus of female produced, and proboscis of same sex chitinous and elongated as is usual in females of this genus. Other characters much as in nigrita. Length, 2.5-3.5 mm. Type, male, allotype, 3 male and 1 female paratypes, Sydney, N.S.W. Family Drosophilidae. LEUCOPHENGA FLAVOHALTERATA, Nl. SD. Male.—Head testaceous-yellow, ocellar spot, upper half of frontal orbits, occi- put except its lower third, and palpi, fuscous; frons reddish-brown posteriorly. Thoracic dorsum reddish-brown, darker behind; pleura stramineous; scutellum BY J. R. MALLOCH. 335 fuscous, the apex narrowly yellow margined; postnotum fuscous. Abdomen shining black, base of first complete tergite yellow, that of second with a large transverse yellow spot on each side of anterior margin, broad at lateral curvature, not extend- ing to lateral margins, and almost obsolete, or connected by a mere line, centrally; fifth tergite with a yellow central mark. Legs stramineous, knees of mid and hind pairs inconspicuously brownish. Wings brownish hyaline, costa quite obviously brown, most distinctly so at apex of first vein and along costa for some distance before apex of second vein. Halteres pale yellow. Frons about one-third of the head width, narrower anteriorly; palpi not widened; facial carina subobsolete. Thorax normal. Wing normal. Length, 2.5 mm. Type, Cronulla, N.S.W., December, 1924; paratype, Waterfall, N.S.W., January, 1925 (H. Petersen). This is the only Australian species known to me in which the wings are marked and the halteres unicoloured yeilow. Family Agromyzidae. Subfamily OCHTHIPHILINAE. PSEUDOLEUCOPIS FASCIVENTRIS, n. sp. Text-tigure 21. Male.—Black, densely whitish-grey pruinescent. Antennae and palpi black; frons entirely pruinescent. Thorax not vittate. Abdomen with a broad deep black fascia on basal half of each tergite which does not extend over the lateral curve, these fasciae very conspicuous when the abdomen is viewed from directly above, and from in front they appear greyish-brown ana but little darker than the grey portions of tergites. Legs black, basal segment of fore tarsi and basal two segments of mid and hind tarsi yellow. Wings hyaline. Halteres yellow. In structure and chaetotaxy similar to the two known species of the genus, but the third antennal segment is quite sharply angulate on upper apical extremity, and evenly rounded below. Length, 2.5 mm. Type and two paratypes, Waterfall, N.S.W., January, 1925 (H. Petersen). A slightly larger male from Cronulla, N.S.W., taken in December, 1924, by the same collector, has the abdominal fasciae continued over the lateral curves of tergites to, or almost to, the extreme lateral margins of tergites. In other respects it agrees very well with the type, though it may represent a distinct species. Both the previously described species have the abdomen shining-black, with only faint brownish dusting when seen from behind. The subfamily Ochthiphilinae is very similar to Sapromyzidae, but in no species of the former are there distinct preapical tibial bristles, the arista is never plumose, and the mesopleura is normally bare. Family Chloropidae. Subfamily BoraNoBiiNak. Genus BatracHomMytaA Skuse. Generic characters.—Differs from the other Australian genera in the sub-family in having the mesopleura with numerous long soft hairs on upper posterior part. Two new species of this genus are in the material before me from Australia, and I have beside me two others from Tasmania. The species are very much the same in structure, but differ markedly in colour. I give below a diagnosis of the characters for distinguishing the new species from Australia. 336 NOTES ON AUSTRALIAN DIPTERA, Vii, Synopsis of Species. A. Third antennal segment entirely, or almost entirely, deep black; hairs on femora avery bam WIEN Gils Ke) chm, amen ene eee ee eee EhEmS ft hor ncanEs Tero sero! chao et Grok ers atricornis, n. sp. AA. Third antennal segment, palpi, and all femoral hairs yellow ...... flavicornis, n. sp. BATRACHOMYIA ATRICORNIS, 0. SD. Male and female.—Head orange-yellow, with slight whitish or yellowish dusting, opaque, a large spot over ocelli and the third antennal segment black; palpi black or yellowish. Thorax brownish testaceous, shining, with slight traces of darker dorsal vittae, pleurae and postnotum nowhere black. Abdomen con- colourous with thorax, paler basally, dark brown apically. Legs a little paler than the thorax. Wings greyish hyaline, slightly tinged with yellowish-brown basally. Hairs mixed, black and yellowish, those on femora mostly black, dorsum of thorax usually with a central stripe of pale hairs. Halteres fulvous. Vertex projecting behind posterior level of eyes when seen from above, the posterior ocelli a little behind level of eyes, the declivitous part of occiput as far behind these as they are from anterior ocellus; entire frons with rather dense erect black hairs; eyes higher than long, hairs pale; cheek fully as high as third antennal segment, pale-haired below, vibrissal angle black-haired; antennae small, third segment higher than long; arista thick basally, densely clothed with micro- scopic appressed black pubescence. Dorsum of thorax rather densely haired, the surface appearing slightly punctate, notopleural region devoid of fine hairs but with four or more fine black bristles posteriorly; scutellum thick and convex, almost bulbous, haired as mesonotum and without differentiated bristles. Abdomen ovate, quite densely haired. Legs stout, hairy; sensory area on hind tibia distinct. Wings large, venation as in Figure 21. Length, 5.5-7 mm. Type, allotype, and two paratypes, Sydney, N.S.W., 14.9.24, and 6.10.24. The two taken on latter date have the palpi yellowish, but I can distinguish no other differences between them and the others. The type has the palpi black. BATRACHOMYIA FLAVICORNIS, 0. Sp. Male.—A paler species than atricornis. In addition to the characters listed in the key, it differs in having the ocellar dark spot smaller, not extending beyond the ocelli, and less intensely black; the hairs on cheeks are all pale; the vertex is not so much extended backward; the ocelli are slightly proximad of the hind margins of eyes; there is a large black mark on the sternopleura; and the apical hairs on scutellum are more differentiated. Length, 6 mm. Type, Kosciusko, N.S.W., 7.12.22 (Nicholson). The two Tasmanian species of the genus will be described in a paper dealing with the Chloropidae of that island which is now ready to send to the press. Genus BENJAMINELLA novum. Generic characters.—Most closely related to Cestoplectus Lamb, having the frons above antennae and in front of frontal triangle depressed or flattened, the front of head being almost vertical from anterior portion of triangle to mouth in the male, less obviously so in female, with a slight convexity in profile only at antennae in the former sex. The distinctions between this genus and Cestoplectus are as follows: each orbit with three or four short black bristles, none in Cestoplectus, notopleura with one anterior and one posterior bristle, not BY J. R. MALLOCH. 337 two posterior; thorax with one pair of prescutellar dorsocentrals, not none; second and third wing veins bare, haired either above or below in Cestoplectus. Genotype, the following species. BENJAMINELLA ALBIFACIES, n. sp. Text-figure 22. Male and female.—Frontal triangle, the section of frons between same and eye on its hind section, and occiput except lower third black, declivitous part of frons, and the antennae lemon-yellow, face and palpi white, the colours not so strikingly contrasted in females; proboscis and cheeks yellow, the latter with a large reddish mark on anterior half; arista black; cephalic hairs yellow. Thorax glossy-black, humeri mostly yellow, notopleural margin whitish-yellow, postalar callosity and tip of scutellum yellow. Abdomen black, yellowish at base and below. Legs black, coxae yellowish in part, tibiae apically, and tarsi basally, brownish-yellow. Wings hyaline. Knobs of halteres yellowish-white. Vertical bristles rather pronounced, ocellar pair minute; triangle broad, ocellar region elevated; declivitous part of frons with a depressed central line; lunule transverse; antennae spread apart sideways in male, third segment orbicu- lar; arista very slightly pubescent, fully as long as declivitous part of frons; profile as'in Figure 22; palpi quite large, slender. Mesonotum with quite dense short decumbent black hairs; scutellum semicircular, disc haired as mesonotum, with two long and two short marginal bristles. Abdomen tapered apically, male hypopygium quite large, the left clasper long and slender, curved across ventral surface. Legs slender, normal. Wings normal for the group, inner crossvein a little before apex of first vein; second section of costa fully twice as long as third; last section of fifth vein distinctly longer than penultimate section of fourth, and twice as long as outer crossvein. Length, 2 mm. Type, male, and allotype, Sydney, N.S.W., 29.10.24, and 28.9.24. Paratype male, Wilmot, Tasmania, 8.1.23 (A. Tonnoir). The genus is named in honour of Dr. Marcus Benjamin, editor of the Proceedings of the United States National Museum. Text-figure 21.—Pseudoleucopis fasciventris, n. sp. Wing. Text-figure 22.—Benjaminella albifacies, n. gen. et sp. Head of male, in profile. Text-figure 23.—Tricimba scutellata, n. sp. Scutellum from above. Genus TRICIMBA Lioy. TRICIMBA SCUTELLATA, n. Sp. Text-figure 23. Male.—Shining-black; frons, except the triangle, opaque, its front margin, ' face, cheeks, palpi, and antennae except upper margin of third segment, testaceous- yellow; lateral margins of mesonotum and upper half of pleura in middle, slightly grey pruinescent; dorsum of abdomen brownish; legs yellow; wings hyaline; halteres yellow, knobs whitish. 338 NOTES ON AUSTRALIAN DIPTERA, Vii, Frons fully half of the head width, with short stubbly bristles, triangle broad, smooth, extending fully to middle of frons; eyes hairy; antennae short, third segment higher than long; arista slender, microscopically pubescent, second seg- ment more than four times as long as thick and about one-third as long as third; face not carinate in middle; vibrissal angle very little produced, cheek about one- seventh of the eye height; proboscis geniculated, but not slender. Thorax with the three impressed longitudinal lines consisting of single series of punctures, the bristles and hairs luteous; scutellum as in Figure 23. Legs normal. Second costal division about 1.5 as long as first, third about as long as first; penultimate section of third vein equal in length to penultimate section of fourth and about half as long as ultimate section of fifth. Length, 1 mm. Type, Sydney, N.S.W., 9.8.24. Distinguished from carinata Malloch by the smaller size, noncarinate face, shape of scutellum, ete. Genus BoTanopia Lioy. This generic name has been used by several writers on the family to supplant Oscinis of authors, not Latreille, the latter’s genus belonging to the subfamily Chloropinae. Becker proposed for the same purpose the new generic name Oscinella. European authors generally disregard Lioy’s work and Becker does not accept Botanobia. Which name is really valid has not been definitely decided, and this is not the place to discuss the matter, but the concepts are the same. The genus has the costa continued to apex of fourth vein; scutellum normal in form and armature; hind tibia without a curved apical spine, and with distinct sensory area; only one or two pairs of posterior dorsocentrals on mesonotum; arista pubescent or bare; proboscis not slender or conspicuously geniculated. The most widely distributed and common genus of the family. Some of the species, such as frit Linn., are of considerable economic importance, feeding in the larval stages in the stems of cultivated grains and grasses. I present descriptions of two outstanding species in this paper, intending to supplement this with descriptions of all the Australian species and a full specific synopsis in a future paper. BOTANOBIA NIGROANNULATA, Nl. Sp. Female.—Head yellowish testaceous, frons and antennae more yellow, centre of occiput broadly black, with a black central stripe extending upward to the large black ocellar spot; inner mouth margin black on sides. Thorax yellow testaceous, with five dorsal black vittae, the median one broad, tapered posteriorly, not extending to hind margin, and connected broadly with the narrower sub- median vittae in front of suture, these also not extending to hind margin; lateral vittae not extending in front of suture; pleura with a spot at anterior spiracle, a stripe over lower margin of mesopleura and pteropleura, the lower half of sternopleura, and a mark on hypopleura, black; scutellum black on disc; postnotum black. Abdomen pale testaceous-yellow, each tergite black on apical half or more. Legs testaceous-yellow, a spot on base of fore coxa, a broad ring on middle of each femur, a band near apices of fore and mid tibiae, and apical half of hind tibia, apical three segments of hind and apical two segments of mid tarsi black. Wings clear. Halteres yellow. Frontal triangle not extending beyond middle of frons; surface hairs on frons numerous and black; eyes hairy; arista with very short pubescence; cheek not as high as width of third antennal segment. Thorax quite densely short-haired; BY J. R. MALLOCH. 339 prescutellar acrostichals lacking; scutellum sparsely haired, with four distinct bristles. Legs rather slender, apices of hind tarsi slightly broadened. Outer cross- vein about twice its own length from inner, and more than that from apex of fifth vein. Length, 2.75 mm. Type, Sydney, N.S.W., 21.9.24. Paratype, Sydney, 6.11.21. The paratype has the median and submedian thoracic vittae extending to hind margin, and the fore and mid tarsi more infuscated than in type. BoOTANOBIA DILATA, Nn. Sp. Male and temale.—Similar to the preceding species. Differ as follows: Disc of thorax normally with the black vittae fused, the scutellum less broadly black; pleura with a greater amount of black; the annuli on legs very much reduced, usually mere spots on one surface only, often absent on tibiae, the apical black part of hind tarsi more intense, and not extending to base of third segment; and the apices of palpi and third antennal segment are infuscated. Structurally similar to preceding species, but the apices of hind tarsi are more conspicuously dilated. Length, 2-2.75 mm. Type, allotype male, one male and two female paratypes, Sydney, N.S.W., October and November, 1924; one female, 24.2.25. Family Muscidae. Subfamily ANTHOMYIINAE. Genus FUCELLIA Rob.-Desv. In a recent lot of material sent to me by Dr. Ferguson there is a male specimen belonging to a species of this genus. I did not include Fucellia in my key to the genera of this subfamily in a recent paper because its occurrence in Australia was unknown to me. The genus may be distinguished from all three previously listed Australian genera of the subfamily by the equally wide frons in both sexes, which occupies one-third of the head width, and has a pair of cruciate interfrontal bristles in both. There is a series of short, but distinct, rather widely spaced bristles on the underside of the costal vein from apex of first vein to about apex of second, a character which is very rare in Muscidae; there are no distinguishable fine hairs on the under surface of the scutellum, a very rare character in the subfamily Anthomyiinae. The hind tibiae have at least three posterodorsal bristles, the lower calyptra is not produced as far as the upper, and the arista is subnude. The larvae usually feed upon seaweed, but one species destroys the eggs of a small marine fish, the Grunion, occurring on the Pacific coast of North America, which are deposited in the sand in a peculiar manner at high tide so that they are well above the normal high water mark. Adults of many species are very common on the seashore in most parts of the world, though they also occur on the shores of lakes and rivers many miles from the sea. FUCELLIA MARITIMA Haliday. This species is distinguished from its allies by the presence of a setigerous process at base of hind femur below in male, and the yellowish palpi and tibiae in both sexes. 340 NOTES ON AUSTRALIAN DIPTERA, Vii, Length, 5-6.5 mm. A very widely distributed species in temperate regions. The specimen before me is from Sydney, N.S.W., January, 1925. Subfamily PHAONTINAE. LIMNOPHORA NARRANDERAE, 0. SDP. Male.—Black, densely lead-grey pruinescent. Frons when seen from behind with the triangle and orbits grey pruinescent, the interfrontalia opaque black. Thoracic dorsum in type with a partial dark brown central vitta, the lateral vittae absent; scutellum with a small brown spot at base in centre, this mark probably variable. Abdomen with a pair of irregularly subtriangular, fuscous spots on each of the first three visible tergites, which extend over their entire length, but are linear anteriorly on each segment, these spots hardly visible on fourth tergite. Legs black. Wings hyaline. Halteres yellow. Calyptra white. Frons at vertex about one-third of the head width, narrowed anteriorly, orbits narrow, with an inner series of strong bristles and numerous long erect lateral hairs; parafacials linear in middle; vibrissal angle hardly produced; cheeks with numerous long hairs, some hairs above vibrissae; antennae extending about three- fourths of the distance to mouth; arista with some distinct hairs basally which are about as long as its basal width; cheek almost linear; eye about twice as high as long. Thorax with 2 + 3 dorsocentrals; numerous erect hairs on disc, about six series between the dorsocentrals. Abdomen narrowly ovate. Fore tibia with- out median posterior bristle; mid tibia with one posterior median bristle; hind femur without distinct ventral bristles; hind tibia with one anteroventral and one anterodorsal bristle. Penultimate section of fourth vein barely longer than outer crossvein and about half as long as ultimate section; first vein bare; first posterior cell slightly narrowed apically. Length, 4.5 mm. Type, Narrandera, N.S.W., 24.3.25. This species will run down to divergens Malloch in my recently published key to the species of this genus. It is readily distinguished from that species by the more conspicuously haired orbits; distinct hairs on the arista; less produced vibrissal angle; much more conspicuously haired and less evidently marked dorsum of thorax; uniformly grey scutellum, which in divergens is largely velvety- black; less clearly defined abdominal spots; and unclouded crossveins of wings. The anterior intraalar in divergens is strong, whereas in the new species it is absent. Subfamily LIsprn ae. LISPA UNISETA Malloch. I described this species from the female only. In a recent lot of specimens Dr. Ferguson sent two specimens of each sex thus enabling me to characterize the male. This sex will run to caption 7 in my recently published key to the species of the genus. It is readily distinguished from both species included there by the presence of but three strong pairs of dorsocentrals on thorax (1 + 2); the armature of hind femur, which consists of two or three short fine bristles on basal half of anteroventral and posteroventral surfaces; and the colour of the apices of mid tibiae, which are whitish. The females have this last feature less distinct. Length, 5-5.5 mm. Localities, Narrandera, N.S.W., 24-25.3.25; and Gunnedah, N.S.W., 1.5.25. THE INFLUENCE OF CERTAIN COLLOIDS UPON FERMENTATION. Parr ii. By R. Greic-SmitH, D.Se., Macleay Bacteriologist to the Society. [Read 30th September, 1925.] It was shown in Part i of this series that certain mineral and other colloids accelerated the fermentation of citrate and of the slowly fermentable carbohydrate, lactose, by the high temperature organism. The work in the main was in agree- ment with results obtained by investigators who had tested other bacteria and the case seems clear so far as the bacteria are concerned. It, however, seems peculiar that other living organisms such as yeasts are not affected in the same way. Soéhngen found that while colloids generally had no influence upon the activity of yeast, the bio-colloids, such as peat, paper, charcoal and soil, hastened the alcoholic fermentation by assisting the elimination of the carbon dioxide from the fermenting fluid. He used five grams of pressed yeast with 100 c.c. of 10 per cent. dextrose and it is perhaps to be expected that the dispersal of the carbon dioxide would be advantageous in a rapid fermentation such as this would be. As it did not seem reasonable to expect that the mineral colloids would really differ in their behaviour towards bacteria and yeasts, a few tests were made to see how far Sohngen’s conclusions were justified. The apparatus used for determining the production of carbon dioxide by the high temperature organism was still in commission and it was used in some tests for ascertaining the influence of a colloid upon the fermentation of dextrose by yeast. The results, however, were not sufficiently consistent to enable any deduction to be made. If we assume that the colloid acts by adsorbing the cellular enzymes and also the fermentable substrate, it is easy to understand that they would have no influence upon the endo-cellular enzyme, zymase. Yeasts, however, possess some extra-cellular enzymes and one of them is invertase. It is possible that if yeast Were grown with sucrose, the colloid might hasten the activity of the invertase and the larger amount of the resulting invert sugar would give rise to an increased production of carbon dioxide. The matter was tested and two small experiments, one with silica, the other with asbestos, were in favour of the idea. A third test may be given in extenso. It was made with 0.5 gram of saccharose and 0.2 gram of activated silica in 30 c.c. of dilute yeast water containing 0.2% of potassium phosphate and 0.1% of magnesium sulphate. The pH value was brought to 6.2. A few drops of a well distributed suspension of distillery yeast were used, the same number of drops being added to each flask. The experiment showed that silica brought about an increased production of carbon dioxide upon the second and third days. It had apparently assisted the yeast invertase to invert the sucrose. In its way, this result with yeast is in agreement with the previous experiments with the high temperature organism. 342 THE INFLUENCE OF CERTAIN COLLOIDS UPON FERMENTATION, ii, Table i—The Influence of Silica upon the Inversion of Saccharose and subsequent Fermentation of the Invert-sugar by Yeast. Daily yields of carbon dioxide in mg. at 24°. Days AE SOL mae meee | || 2 3 4 5 Shien, 4) Ee 55 oso oc 54 131 67 12 % 54 | 129 fall mul 2 Deen el ea 60 18 2 Average eS spake 54 129 66 14 2 | Control (no Colloid) Ae 58 i3lat 57 i? 9 54 110 51 31 14 HH |] IAG 57 31 1 Average are ART 15 56 WILL 55 26 10 The effect of a colloid upon the inversion of saccharose was then tested directly by noting the disappearance of saccharose. Two flasks each with 450 c.c. of nutritive fluid containing 20% of saccharose were seeded each with three drops of a well distributed suspension of a distillery yeast. One flask received 0.5% of asbestos as the colloid. By a mistake the flasks were incubated at 38° during the first day instead of 30° as was intended, but as the inversion appeared to be progressing favourably, the flasks were maintained at the higher temperature. The sugars were estimated according to the routine followed by Lane and EHKynon (Journ. Soc. Chem. Ind. 1923, 32T.), the only modification being in the use of a flask provided with a cork pierced with three holes, one for the nozzle of the burette, one for introducing the methylene blue, and the third holding an angular tube as a steam get-away. Table ii—The Influence of Asbestos on the Inversion of Saccharose by Yeast at 38°. Days Sec) yp aces LEG Ylowe’ cheers Start iL 2 3 Asbestos, saccharose inverted .. .. = 8.3 13.8 14.2 saccharose not inverted .. 20 Wil a} 3.0 0.5 Control, saccharose inverted .. .. — 6.2 1033 10.3 saccharose not inverted .. 20 13} 583 6.5 3.8 The rate of disappearance of the saccharose shows that the colloid had a decided influence in accelerating the inversion activity of the yeast. A confirmative test was made with silica as the colloid and the temperature was maintained at 30° instead of at 38° as in the last case. The influence of the colloid is not so clearly shown in this as in the last experiment. It may be the difference of the colloid but it is more likely the effect of the difference in the temperature. Invertase has its optimum working temperature between 55° and 60°, and it is to be expected that the nearer the fermentation temperature is to this the greater will be the inversion. BY R. GREIG-SMITH. 343 Table iiiimThe Influence of Silica on the Inversion of Saccharose by Yeast at 30°. Days Ae rhe Meret eect) IStart 1 2 3 4 5 6 | | Silica, saccharose inverted are — ibe) 4.0 So¢ | wt@o4) } alab.% 10.3 saccharose not inverted ..| 20.2 | 18.3 — 8.8 5.2 3.4 Lo saccharose unaccounted for| — , — —- et |) 6G 5.6 Seti Control, saccharose inverted .. — | 1.2 3.8 8.2 10.0 10.9 10.5 saccharose not inverted ..| 20.2 | 18.9 | — ORD 6.1 4.0 1.6 saccharose unaccounted for| — | — — 2D 4.1 i) 6 8 8.1 The last two experiments confirm the previous finding that colloids such as asbestos and silica, accelerate the activity of organisms which secrete extra- cellular enzymes such as invertase. In summing up the last table and noting the saccharose unaccounted for, presumably to form alcohol, carbon dioxide, glycerin and fixed acids, it would seem that the colloid did to a certain extent increase the production of alcohol and carbon dioxide. As the point was of some importance, it was decided to test the matter by the disappearance of dextrose during its fermentation in the presence and absence of a colloid. A test with the same nutritive fluid blended with glucose-syrup instead of saccharose showed a slightly greater disappearance of the dextrose from the third to the fifth day. Although less sugar disappeared from the control fluid, it seemed to be fermenting more freely. The foam on its surface was more persistent and when filtered to eliminate the carbon dioxide it was always more cloudy than the test fluid. Possibly the yeast cells had agglomerated on the particles of silica, but such could not be demonstrated microscopically nor was there any pronounced agglutination of the yeast cells in either of the fluids. It seemed that the method of conducting the last experiment could be improved upon. The fermentation of the control fluid was apparently different from that of the test with the colloid and it seemed possible, in spite of a vigorous shaking just before drawing off the portions, that more yeast cells might have been abstracted from the control than from the test. In the next experiment, the medium was divided into 50 c.c. portions con- tained in 100 c.c. conical flasks and whole portions were taken for the dextrose determination. The medium used was the same as the last, with the addition of 0.25% peptone, that is to say it contained starch-glucose syrup 360 grams, yeast water 100 c.c., monopotassium phosphate 2, anhydrous magnesium sulphate 1 and peptone 2.5 grams per litre. The pH value was brought to 6.0. The progeny from a single cell of a recently obtained pressed yeast was distributed in Hansen’s fluid and six drops were added to each flask. Table iv.m—The Influence of Silica upon the Fermentation of Dextrose. Percentage of dextrose consumed. Days ee Me MA DeNEA MEU ests 2 | 3 4 5 7 | SMe, OsBDYso 56 10 00 47 53 55 58 68 Controlge sss. ween =} akeds 43 | 50 | 51 53 | 58 344 THE INFLUENCE OF CERTAIN COLLOIDS UPON FERMENTATION, ii, There was an increased consumption of the dextrose under the influence of the silica and it was maintained throughout the experiment. The controls always had more foam on the surfaces of the fluids and this led to testing the viscosity of a fermenting solution containing starch-glucose. There was a slightly increased viscosity in the test containing silica but in view of the fact that the starch-glucose contained much dextrin, the experiment was repeated at a later date when fuller’s earth was used as the colloid and commercial dextrose, containing 75% of dextrose, as the sugar. The medium was the same as that used in experiment iv, with the substitution of 16% of commercial dextrose for the starch glucose. Two flasks were used, each with 500 c.c. of medium; one received in addition 2.5 grams of fuller’s earth. The seeding was a five days’ culture at 28° of a distillery yeast in Hansen’s fluid. The supernatant liquid had been decanted and replaced with diluted meat extract; the cells were broken apart and uniformly distributed by repeated passage through a capillary nozzle. Hight drops (about 0.25 c.c.) were added to each flask. During fermentation, the control fluid carried more froth than the test and foamed strongly on being shaken; the test carried little froth and foamed slightly. On the second day, the test contained 7.08% of dextrose and the experiment was concluded on the third day. The fluids were filtered through paper, then through paper pulp in order to obtain brilliant solutions. Table v.—The Influence of Fuller’s Earth upon the Viscosity and Alcoholic Fermentation of Dextrose. Three days at 28°. | Fuller’s | Control | earth ( | | At start— | ‘ Dextrose 7 ial 7/5 iL aD At end— IDSxtrOsen” wis iis). NSAI ee: OF... Tee Oe 6.01 2.83 = Dextrose consumed % Tee Seed ote 49 76 INGOACG IW WIETEINE 65° bo oo oo oo | 2.64 4.13 Wilson (Cee = 1l)) 65 of 56 99 1.246 1.257 The viscosity of the test fluid was greater than that of the control, but as there was a difference of only ten seconds in the run of 184 minutes by the 100 c.c. of liquid through the capillary nozzle, the difference is too small to explain any retention of carbon dioxide during fermentation. The difference in the dextrose consumed is the greatest that has been noted in these experiments with yeast. It was possibly due to the seeding cells being five days old and in consequence perhaps slightly enfeebled. The stimulating influence of the colloids upon enfeebled bacterial cells has been already noted. It has been seen that silica hastened the fermentation of dextrose and it remained to show that the other insoluble colloids did the same. The medium used was much the same as in the previous experiments and consisted of commercial glucose 100 grams, yeast water 100 c¢.c., potassium dihydrogen phosphate 5 grams, magnesium sulphate anhyd. 1 gram, and peptone 2 grams per litre; the reaction was brought to pH 6.0. The fluid when prepared contained 7.5% of dextrose. Three hundred c.c. portions were pipetted into a number of 500 c.c. flasks and with the exceptions of the agar and the controls each received 1.5 grams of colloid. Ten drops of a uniformly distributed suspension BY R. GREIG-SMITH. 345 of a distillery yeast were added to each flask. Set 1 was incubated at 37° but, as the fermentation seemed to have been too quick and it was hoped to obtain wider differences with a slower fermentation, the flasks of set 2 were incubated at 28°. Table vi—The Fermentation of Dextrose by Yeast. Percentage of dextrose consumed. Set 1d aie se | Set 2 at 28° Days rae os 2a ieee | iL | 2 IDayiStt hrs meen eheee Sieeee LEE 2 3 | Tale, 0.5% Ben Mis ce og 23.9 80.7 Kieselguhr, 0.5% . se 67.9 O)ab oh iK@iollon, WHS ca 66. oc 20.9 80.6 French chalk, 0.5% Ae Bye) 5 2 90.5 Asbestos, 0.5% : ee U9) 6 76.3 Agar, 0.2% es 47.0 90.0 Fuller’s earth, 0.5% a 22.9 85.1 Control (no colloid) Ete D3ieo 87.9 Control (no colloid) ae PAIL 6 5) 20) All the colloids increased the consumption of dextrose, that is, they had hastened the fermentation by one or the other day of observation. Set 2 incubated at 28° certainly showed greater differences than set 1 at 37°, probably because the fermentation was caught when from 50 to 60% of the dextrose had been fermented. It was noted that the control of set 1 at the end of the first day contained more carbon dioxide in solution than the others as was evidenced by a more vigorous effervescence when the liquid was discharged upon a filter to eliminate the dissolved gas previous to measuring off the portion for analysis. This would make it appear that Sodhngen’s idea of the colloids increasing the fermentation by assisting the elimination of the carbon dioxide was correct. The asbestos test with the smaller consumption: of dextrose was anomalous. It had a lessened effervescence like the other colloids, but the asbestos fibres had formed a felt- like film on the surface of the fermenting fluid and this had possibly prevented the aeration of the liquid. In set 2 the control test and the agar test, which was viscous, showed most effervescence upon filtration. A few more colloidal substances were tested in a further experiment in which the peptone was replaced by marmite. The colloids undoubtedly accelerated, although the differences were not so pronounced as in the last table. This might have been caused by the substitution of marmite for the peptone. | Table via.—The Fermentation of Dextrose by Yeast. Percentage of dextrose consumed. Set 1 at 28° | | | Set 2 at 37° | | IDE WS) Ges Ca OE ee 1 2 3 | 4 |Days ote oF fa = | 3 | | | | Silicic acid, 0.05%| 31.9*| 50.5*/ 71.0*| 91.7 |Wood charcoal, 0.25%| 40.3 | 70.3 | 91.0 Ferric hydroxide, Kieselguhr, 0.25% coll Bos 70.4 OIL GV 0.015% . | 2809 | S200 || 7H. || O06 jaele, O28 o5 65 call BBo |) Oot! 91.1 ASE, OIG so coll BlsFe] S206 || 7.8 1 OB.z Gelatin, 0.25% 30.8*| 63.4*| 84.4* Fuller’s earth, Control (no colloid) . 38.1*| 65.1*| 86.2 0.25% s coll 2008) | Sil | W8o2 | OB! Control (no colloid) .. ..| 31.1*| 50.5*| 69.5*] 90.0 co SS ee OO The colloidal silicic acid and the colloidal ferric hydroxide were sterilized separately and added to the flasks previous to seeding. The hydroxide was 346 THE INFLUENCE OF CERTAIN COLLOIDS UPON FERMENTATION, li, coagulated at once and the silicic acid separated out during the first day; they acted, therefore, as insoluble colloids. The tests marked with an asterisk foamed when pipetted on to the filter, the others were still. The foaming does not seem to have much significance, for the first day’s results are just the opposite of what should have happened if an excessive amount of carbon dioxide dissolved in the fluid had hindered the fermen- tation. On the other hand, in set 2 the gelatin test always effervesced when filtered and the amount of dextrose consumed was always lower than the control. So far the evidence is rather in favour of Sdéhngen’s idea. ' Agar which tended to hinder the fermentation when present to the extent of 0.2%, accelerated when a smaller quantity was added; the fluid was not so viscous. It is, however, irregular in its action. In another experiment in which 0.05% was used, it acted as a depressant. Its behaviour is under investigation. Taken as a whole it is clear that certain mineral and other suspended colloids accelerate the activity of yeast. The substances that have been tested with positive results include silica, asbestos, ferric hydrate, talc, kaolin, fuller’s earth, kieselguhr, French chalk, agar and charcoal. The main point that has been brought out is that yeast, like the high tempera- ture bacterium, ferments more actively in the presence of certain suspended colloids. It is true that the effects were not so great, but they were sufficiently pronounced to show that living bacteria and yeasts do not materially differ in this respect. ; In my previous work with colloids, use had been made of the high temperature organism and the complete dissolution of a sugar to carbon dioxide had been taken as evidence of the bacterial activity. The work in this paper has shown that a partial dissolution, the inversion of saccharose by yeast, was increased by colloids. Some work that had been done with invertase itself necessitated a confirmation of the accelerating influence of colloids upon inversion by another living micro-organism. The Inversion of Saccharose by Bac. levaniformans. The organism chosen was Bac. levaniformans, an ally of Bac. vulgatus, occur- ring in sugar crystals in an almost pure condition. There are several races of this organism. One of them, previously described as 68 (Proc. Linn. Soc. N. S. WALES, 1901, 613), was isolated from brewers’ crystals. At the time of their isolation, small crystals of household sugar were examined and found to contain the normal white race. The race 68 grew as a yellow expansion on ordinary nutrient agar and gave no film on bouillon. A heaped loop of an agar-culture was added to a bouillon-culture and the cells distributed by repeated passage through the capillary nozzle of a pipette. Four drops of the suspension were added to flasks containing 50 c.c. portions of a saccharose medium which consisted of saccharose 80, sodium phosphate 2, potassium chloride 5 and peptone 2 grams in a litre of tap-water. It had been brought to a pH value of 6.8 with dilute hydrochloric acid and had been steamed on six successive days. After the flasks had been incubated at 37° for a day, it was noted that two of the controls were clear, while the others were turbid. This was again noted in a second experiment and also that some of the controls which had begun to ferment, had more gum than the others indicating that, although they had been seeded with the same number of cells of the same vitality, the onset of fermen- BY R. GREIG-SMITH. 347 tation in the control flasks had been irregular. This destroyed the scheme of the experiments. A similar irregularity had been noted in the case of the high temperature organism and had led to the examination of the influence of colloids upon fermentation. The experience with the tests indicated that the seeding cells should have previously been grown in the same fluid, and also that the gum in the cultures should be eliminated by precipitation with alcohol when preparing the solution for the determination of the sugar. The method adopted was to pipette 25 c.c. of the fermented fluid into a 100 c.c. flask, neutralize it to phenolphthalein and make to volume with strong spirit. After standing overnight, the solution was decanted or filtered off and the alcohol eliminated by evaporation from an aliquot portion. The residual fluid was transferred to a 100 cc. flask, slightly acidified to methyl-red with acetic acid, treated with lead subacetate, made to volume and filtered into a 100 cc. graduated jar. After measuring, the filtrate was acidified with dilute acetic acid, treated with an excess of potassium oxalate and filtered. The sugar was determined in the filtrate. Two experiments were made and the following results were obtained on the second day in each case. By the fourth day the amounts were very much the same at about 80% and are not recorded. Table vii.—The Inversion of Saccharose by Bac. levaniformans in presence of some Colloids. Percentage of saccharose inverted in two days at 37°. SES AEE es eae BE ae 5 Set 1. Set 2. PUULeM SS eCarthsase: acy cee de ows 69.0 Tale Bit, OS al Bets eeces Mein ake 58.4 Asbestos A MME MIN ii taesoLS) emcee. Boke 64.7 IGA OHNE ee ahah eh hc, wearer 47.5 Kieselguhr ae am dete ghee } Mae. 56.6 Animal charcoal .. .. .. ..| - 40.0 French chalk PIM ae Mie Vo ® pee 54.5 Waillowarchancoaliaeusars anienens 21.2 SUNT A aye Meate, intel, Auer bee | deus lens 52.4 | Control (mo colloid) .. .. .. Sil G2 Control (no colloid) i eee 48.2 EE All these colloids with the exception of willow charcoal accelerated the inversion of sugar. In a future paper I shall show that this form of charcoal does hasten the inversion and is no exception to the fact which seems to be well demonstrated that the mineral and other colloids such as have been tested, accelerate the inversion activity of Bac. levaniformans. Bac. fluorescens liquefaciens is supposed to be able to secrete invertase, but when it was grown in a fluid containing sugar, no inversion of the saccharose occurred. So far it has been shown that the mineral and other insoluble colloids accelerate the fermentation of dextrose by yeast, the inversion of saccharose by yeast and the inversion of saccharose by Bac. levaniformans. It seems to be a stimulation of the cell activity rather than an adsorption of ferment and ferment- able substrate upon the colloidal surfaces. To test the matter it was decided to test the activity of a naked enzyme when in contact with the same colloids. The enzyme chosen was the same that had already been under investigation, viz., invertase. The Action of Certain Colloids upon Invertase. Invertase was prepared from autolysed yeast in the manner usually recom- mended and its activity was tested. 348 THE INFLUENCE OF CERTAIN COLLOIDS UPON FERMENTATION, ii, One-tenth of a gram of the dry powder was stirred with 10 c.c. of water which was acidified with oxalic acid to a pH value of 4.4. After half an hour the suspension was centrifugalized and the opalescent supernatant fluid was pipetted off. Two c.c. portions were pipetted into flasks containing 25 c.c. of 8% saccharose, 5 ce. of “/,, potassium dihydrogen phosphate and 0.2 gram of colloid. In the cases of gelatinous aluminium hydrate and ferric phosphate 3 c.c. of a 3% suspension were used. Previous to the addition of the invertase, the saccharine fluids with their colloids had been brought to a pH value of 4.4 (pale green with brom-cresol green) with decinormal hydrochloric acid. Some of the colloids absorbed the acid, but the absorption had slowed down by the end of half an hour. The flasks were then placed in a thermostat at 40° for 45 minutes, treated with 2 c.c. of basic lead acetate, cooled, made to 100 c.c. and filtered; 50 c.c. of the filtrate were treated with 5 c.c. of 10% potassium oxalate, acidified with dilute acetic acid, made to 200 c.c. and filtered. The invert sugar was determined in the filtrate by means of Fehling’s solution. About 50% of the sugar was inverted under the conditions. Preliminary tests showed some variation, but when the method as described was followed it was shown that the colloids had no influence upon the inversion of saccharose. The tests were made in three lots in order to avoid delay between the time of inversion and the determination of the invert sugar. Table viili—The Infiuence of Colloids upon the Inversion Oe Saccharose by Invertase. Percentage inverted. Set 1. Set 2. Set 3. No colloid | 533 5 I No colloid .. 47.4 No colloid ah latte: 49.0 Kaolin | 52.7 French chalk 47.9 Silica ater Abe © pairs 48.7 Kieselguhr ree Done Charcoal , 47.2 Tale ‘ 5 48.7 Diatomaceous earth DB oI Asbestos 46.0 Ferric phosphate . svell| AWia4 Gelatin .. 5: 53.4 i Aluminium hydrate| 44.9 Agar powder ae =| 53.4 | | If the individual tests are taken to whole numbers, it will be clearly seen that the colloids were inactive. The low number for the gelatinous aluminium hydrate was probably due to the solution of the hydrate in the hydrochloric acid of which a considerable amount had to be added to bring the solution to pH 4.4. The indifference of charcoal is in agreement with the work of Griffen and Nelson (Journ. Amer. Chem. Soc., 1916, 722), who found that charcoal, alumina and albumen had no influence upon the activity of invertase. Since the colloids do not accelerate the inversion of saccharose by invertase itself, it follows that the adsorption of the enzyme, if it does occur, is not the explanation of the efficiency of the colloids with living cells. The matter being of some importance, it was decided to test another enzymic fermentation, that of the decomposition of urea by bacteria and by the enzyme, urease. The Ammoniacal Fermentation of Urea. A flask with urea had undergone fermentation and from it a culture of B. fluorescens liquefaciens was obtained. It was seeded into flasks containing urea and buffer salts with and without kieselguhr. Six flasks were used, one for each determination of the ammonia by the method which will be described later. BY R. GREIG-SMITH. 349 Table ix.—The Fermentation of Urea by B. fluorescens liquefaciens. Percentage of urea fermented. Days MR eee eh, Giniayiee iaiey aye il | 2 | 4 Comtroluiry timed leeks 2d dec Ps 16.0 Us || 20.6 Eieselpuih ice 94). Tisch Betoldae sah OS 16.3 28.6 43.6 The kieselguhr undoubtedly accelerated the fermentation. A urea-fermenting organism was isolated from soil. Morphologically it was like Urobacillus duclauxii and culturally like U. pasteurii.* It grew well at 37° on nutrient agar containing urea. The bacterium was distributed in dilute meat-extract and an equal number of drops were seeded into twelve flasks, each containing 50 c.c. of a fluid consisting of 1% peptone, 1% urea and 0.5% sodium chloride. The urea was sterilized apart from the other components of the medium. The flasks were in groups of three, each group containing 0.2 gram of suspended colloid, with the exception of one group, which acted as control. The results of the experiment show that these suspended colloids undoubtedly accelerated the activity of the bacterium. Using the same organism, tests were made with other colloidal substances. The progress of the fermentation was watched both by the rise in the pH values and by the amount of ammonia found in a special control flask. This was found to be necessary when a progressive determination similar to the previous table was not made; one could not guess just when to stop the fermentation in order to show the effects of the colloids. In the two tests about to be tabulated all the flasks, with the exception of the sterile controls, contained bacteria. The infected Table x.—The Influence of Colloids upon the Bacterial Fermentation of Urea. Percentage of urea fermented. Days tLe SoM Ee DUR hte Baraca OF kere, 1 2 3 Control 9.6 8 .5F 14.5 Asbestos CAR Ree ae 15.5 HI” | 95.6 TEGO IND AUT Bae) TORMENT OR es Ch ee coat em 115) 9 31.2 96.0 Charcoal yee, Tum gt cent Wenn 11.4 20.2 39.0 * A motile rod with rounded ends. 0.6: 2-34 with longer and thread forms; spores were terminal and the Gram stain was positive. No growth was obtained on agar or gelatin devoid of urea, although, in bouillon without urea, a slight turbidity developed. In urea bouillon, a strong turbidity was produced. Milk without urea was unaffected. On nutrient agar with urea there developed at 37° a moist glistening translucent expansion. The colonies in urea gelatin appeared as a collection of pale brown lobules radiating from a central mass of small brown granules. These granules also appeared in the medium around the colony. In time the surface colonies became a mass of brown granules (not crystals) in a crateriform apparently liquefied area. The stab in urea gelatin was filiform surrounded by a white haze with little or no surface growth; at a later date (li days) the gelatin was consumed, leaving a gas bubble, the upper part of the canal was ciliate and the lower part was studded with crystals, a haze was spread through the upper two-thirds of the gelatin. According to Geilinger (through Abst. in Bact. iv, p. 93) the identification of the urea-fermenting bacteria is a matter of some difficulty. 7 A large loop of the control of the second day was found to be sterile. 350 THE INFLUENCE OF CERTAIN COLLOIDS UPON FERMENTATION, ii, controls of set a@ were bright but they yielded growths of bacteria when sown upon urea-agar slopes and they had the faecal smell characteristic of the bacillus. Set a was stopped on the fourth day and set 0 on the third day. As in all previous cases, each flask in each set received the same number of drops of a well distributed culture of the bacillus. A sterile flask was included in each test in order to control the urea decomposed during the sterilization of the media. In set a this was equivalent to 4.1 and in set b to 4.6% of urea. These percentages have been subtracted from those obtained with the others. The colloidal solutions of silicic acid and of ferric hydroxide coagulated upon being added to the saline medium; they acted, therefore, as insoluble colloids. The fluids were coloured with phenol-red and all were brought to the same pH value (7.2) before adding the sterile solution of urea. Table xiim—The Influence of Colloids upon the Bacterial Fermentation of Urea. Percentage of urea fermented. Set. a. b. | Silicic acid 0.1gram 63.2 45.9 Silica Beets og ead 0. a 34.5 44.4 Aluminium hydrate Oral a 5.4 23.0 Aluminium phosphate .. Op rs 48.6 36.2 Ferric hydroxide .. O50 56 ila}, al 30.2 Kieselguhr Oar ke 16.2 30.9 Fuller’s earth OSA5 Bi) 58) 36.8 Asbestos oc Do% py 44.8 12.4 French chalk OS2i 7.6 Til 5 7 Agar Sale ROM ieee 6.9 5.4 Gelatin: sie of sale Ole! fees 3.6 11.6 Control (no colloid) {0.6 8.5 10.4 | 6.5 As a rule the colloids accelerated the fermentation of urea by the bacillus, and although they do not run parallel in each set their action is quite clear. The finely divided precipitate of silica formed on the addition of the silicic acid sol to the culture fluid was the most efficient accelerator. During the sterilization of the fluids, the agar had softened and had set as a cake at the bottom of the flasks. Before infection the cake was chopped up with a platinum chisel; it acted as lumps of solid agar. After noting the activity of the finely divided precipitate of silica in the silicic acid test it was decided to try the effect of having the agar as a finely divided flocculent solid. Seven flasks were prepared as in the previous tests and in two of them the agar (= 0.2 gram) was shaken up after each sterilization. On cooling it separated in fine floccules. The same weight of agar-fibre was sterilized in test tubes without water and added to their respective flasks just before inoculation. One of the flasks served as a sterile control and gave an amount of ammonia equal to 5.0% of urea; this has been deducted from the other totals. Table xii—The Action of Agar in the Bacterial Fermentation of Urea. Percentage of urea fermented in five days at 37°. | Agar-fibre Sie Seine dk | 16.4 16.9 16.6 Finely divided agar .. See chr Mocs bolus wa Ob to, Pyles 74 21.0 Control (no agar) Meshes Bete oil 8.3 8.7 BY R. GREIG-SMITH. 351 The effect of the subdivision of the colloid is decided, but the difference is not so great as one might have expected. Similarly in set b of the previous table, the difference between the finely divided silica and the precipitated silica of commerce is not very pronounced. The case having been proved for the fermentation of urea by bacteria, the influence of colloids upon urease was tested. Soy beans were used as the source of the enzyme. They were ground to a coarse meal and extracted with petrol. The defatted meal was ground to a fine flour and bottled for use. Two grams of the flour were shaken with 100 c.c. of water in a bottle at intervals during half an hour and filtered. Half a gram of urea was dissolved in 250 c.c. of water and 25 c.c. portions were pipetted into 100 e.c. conical flasks. These were fitted with corks, through each of which passed a rather wide tube holding a small plug of cotton-wool. The plug was subsequently moistened with dilute sulphuric acid and served to trap any ammonia attempting to escape from the flask. Hach flask then received 10 c.c. of a buffer mixture of salts of pH value 5.6 and enough methyl-red to colour the fluid. The colloid, 0.2 gram, was added and the flasks were inserted in the thermostat at 40° where they remained for 15 minutes. They were examined and brought to a.pH value of 5.6, that is, to a pale brick-red colour by the addition of a few drops of acid or alkali. Two c.c. of the soy bean extract (= 0.04 gram bean-flour) were added to each flask, the plugs were moistened with dilute acid and the flasks returned to the thermostat* where they remained for half an hour.+ The flasks were withdrawn from the thermostat, 5 c.c. of ‘/, sulphuric acid were added to each, the plugs were pushed in and the tubes were washed into the flasks. The fluids were transferred to 500 c.c. flasks and boiled for 5 minutes to expel the carbon dioxide. After cooling, 1 gram of magnesium oxide was added and the flasks were connected up with condensers and boiled.t The distillates Table xiiiimThe Influence of some Colloids upon the Fermentation of Urea by Urease. Percentage of urea fermented. Set. ii nee | 3 | 4 Control (no colloid) Rue xe¥ 56 Bal 55 57 Asbestos we BL ae 58 51 — 53 Silica Bek CRM Bes) ge ral Whe. 56 — 52 - Kieselguhr Parnes — 51 — 59 Tale ’ pri ie sian ote — | 46 51 — Fuller’s earth BA A, OS — | — 52 | 50 Charcoal 3 5 — — 56 50 Ferric phosphate — — —— 56 * The thermostat consisted of a copper vessel with wires soldered vertically round the top. Each wire slid through the helical spring of an American clothes-peg which gripped the tube passing through cork of the flask. Thus the flasks were safely held while immersed in the water. 7 The fluid lost its acid reaction as the ammonium carbonate was formed. A repeated adjustment to pH 5.6 had the effect of slowing the fermentation and was not adopted. This is a departure from the usual method which consists in adding a definite number of e.c. of standard acid to the fermented urea solution, eliminating the carbon dioxide by passing a current of air through the acidified solution and determining the free acid in the flask. It was found that the colloids absorbed more or less acid, some- times as much as 1 ce. (= 6% of urea fermented) and this rendered the method untrustworthy. 352 THE INFLUENCE OF CERTAIN COLLOIDS UPON FERMENTATION, 1i, were caught in 20 c.c. portions of deci-normal acid and titrated, using Congo-red as the indicator. Controls showed that an allowance had to be made for the magnesium oxide. The results showed that the controls were generally better fermented than the colloid tests and enough seemed to have been done to indicate that these substances do not accelerate the enzymic fermentation. In the hope that some further information might be given by a slower fermentation, three tests were made in which the ammonia was determined at half-hourly intervals and in which a variable amount of bean extract was used. Table xiv.—Varying Amounts of Urease. Percentage of urea fermented. Ratio of Soy- bean meal to urea. 3 hour. | 1 hour. |1% hours at 40°. AO. 5) Control 56 100 100 Kieselguhr Byil | 91 100 MS ti) Control 33 61 465) Kieselguhr 27 56 71 3 3 § Control 47 | 77 Oat Asbestos 46 76 S)il The graphs of these numbers did not give any further information and the tests simply confirmed the previous work in showing that these colloids, rather than accelerate the fermentation, tended to retard it. : A very weak solution of urease was used in the following sequel to the last experiment. Table xv.—The Slow Fermentation of Urea by Urease. Percentage fermented. Days PaEe abs CaS il | 2 3 4 Gh COimirOl 56. so 18.6 27.9 66.4 66.4 Kieselguhr ae 2.0 5) of) 1 o%) 10.2 b. Control oe 23.5 40.9 55.9 5b 9) Kaolin ane ae 17.4 28.6 Bis) ol 40.6 Ratio of Soy-bean meal to urea, 1: 25. The slow fermentation exhibits in a more marked manner the inhibiting action of the colloid upon the urease. We have thus come to the conclusion that the mineral colloids and suspensoids, such as have been used in the tests, assist the fermentation by living organisms such as bacteria and yeasts and depress or tend to depress the fermentation by the naked enzymes, invertase and urease. Their depressing influence upon the enzymes is doubtless bound up with their faculty of being adsorbed upon the colloid and thus being removed from the mass of the substrate in the supernatant fluid.* The reason for their * Waksman (Abstracts of Bacteriology, vi, 269) in a review of the literature upon the “Enzymes of Microorganisms,” writes that enzymes ‘are readily taken up by finely divided substances such as kaolin, charcoal, infusorial earth, alumina, protein. Adsorbed enzymes are inactive.” BY R. GREIG-SMITH. 353 accelerating action upon the living cell cannot be explained and it seems reason- able to suppose that since the action is the opposite the modus operandi will not be on account of adsorption. The Lactic Fermentation of Dextrose. The lactic fermentation of dextrose promised to give further information regarding the influence of the colloids and a set of tests was prepared. The organism used was an acid-former that had been isolated from a liquid suspension of yeasts and bacteria sold by some local firms to bakers for raising their doughs. The preparations are known as spontaneous or “spon” yeasts. The bacterial flora is mixed, but by far the most numerous organism is one which grows feebly on the ordinary solid media, better on media containing dextrose and better still in dextrose nutritive fluids. Several of what were probably races of one organism were isolated; one* of them produced most acid in a modified Hansen’s fluid, attaining a pH value of 3.4 as against 3.9 and 4.6 with the others. The fluid in which the bacterium was grown contained starch glucose 5%, peptone 1%, potassium phosphate 0.35%, sodium phosphate 0.15% and sodium chloride 0.15%. The pH value was 6.6. Fifty c.c. were put into flasks together with 0.25: gram of the colloid ‘except in the case of agar, of which 0.1 gram was taken. The agar liquefied during the sterilization of the flasks, and when the liquid cooled the agar settled out as a lumpy kind of clot. During the experiment, 5 c.c. of fluid were extracted daily from each flask, coloured with a drop of phenolphthalein and titrated with ‘/., potassium hydrate until a faint pink colour was obtained. The pink colour disappeared and after an interval of one minute the colour was adjusted with a further addition of alkali. In the experiments with urea, it had been found that the colloids absorbed hydro- chloric acid, but in this experiment there appeared to be no absorption of the lactic acid formed by the microbe. Two experiments were made, the first with a medium containing 0.5% of peptone but as the results were relatively the same as with a medium containing Table xvi.—The Influence of Colloids upon the Lactic Fermentation of Dextrose. Acidity as c.c. of normal acid per litre of medium. Time in days Fae a 2 3 4 5 6 French chalk S70 hans 38 58 64 65 68 Tale Rai tie! oly I Pb US 39 58 62 3 65 Animal charcoal ah aes 38 56 62 63 64 \Wivilllony Goegyeooe! 55 oo’ ao 36 56 3 65 66 Asbestos Bit Laalbnas! pee 36 53 61 63 63 Fuller’s earth yee Rete 3. 34 Gil 59 64 65 Keseleuhrine-r. | sa .h es dex 38 53 59 61 | 64 Silica We er eS) Ls: 37 51 57 Gil) 8 HAO hee. fk. ake Pbk 36 51 55 58 63 Control (no colloid) .. .. 37 49 56 60 63 Control (mo colloid) .. .. 36 50 He || wy 61 Gelatin) "Si. Beteae iets Bees. 4 Ss 36 48 55 59 63 Agar Fae Re sikle eres Mest ss 32 43 53 59 62 *Tt is a short rod, almost a coccobacterium, forming small translucent white colonies on dextrose agar. After some subcultivation, it grew rather better, especially on agar containing saccharose and glycerin. Dextrose-agar coloured with brom-cresol purple was rapidly turned yellow, even with the feebly growing cells. Milk was in some instances slightly acidified but never curdled. The acid formed in fluid dextrose media was almost entirely lactic. It grew better at 28° than at 37°. 354 THE INFLUENCE OF CERTAIN COLLOIDS UPON FERMENTATION, ii, 1%, the second experiment is alone recorded. The figures for the fourth day of the 0.5% peptone test were practically the same as those of the third day with the 1%; the smaller amount of peptone reduced the acidity by about 8%. The experiment shows that many of the colloids have accelerated the acidifi- cation of the fluid. French chalk, tale, the charcoals, asbestos and fuller’s earth were the most- efficient, while gelatin and agar had no action or depressed the fermentation. The general outcome is that the lactic fermentation falls into line with the other bacterial fermentations which are assisted by the presence of these colloids. The Influence of Colloids on Diastatic Activity. The action of malt-diastase upon starch was tested to see if it differed from urease and invertase in its behaviour in the presence of mineral colloids. Pre- liminary tests with iodine as an indicator showed that the qualitative method was not sufficiently regular in its action to bring forward the results as evidence of the activity or otherwise of the colloids. The actual determination of the sugar resulting from the fermentation promised to be more acceptable. Two grams of soluble starch were put into a.100 c.c. conical flask along with 50 ¢.c. of water. The starch was liquefied and upon cooling 5 c.c. of a buffer mix- ture of salts of pH value 4.9 were added. Two grams of ground malt were digested overnight at 22° in 100 c.c. of water. The clear filtrate was diluted four- fold and 5 c.c. were pipetted into each flask together with 0.2 c.c. of chloroform. The flasks were corked and incubated overnight at 37°. The slow fermentation with the small quantity of malt extract (= 0.025 gram of malt) converted about 60% of the starch and this was considered to be about the extent of change best suited to show the effect of the added colloid, of which 0.2 gram was present in each flask. Some tests with a larger proportion of malt had given from 75% to 78% of altered starch, but as this is so near the point where starch shows a resting stage or rather a slowing up in the fermentation (Trans. Guinness Research Lab. i (1), 87), the smaller quantity was used. Table xvii.—The Influence of Certain Colloids upon the Saccharification of Starch by Diastase. Maltose from 100 grams of dry starch. Test Fie LS Pe ee, TN ter eee il 2 3 Control St) Foe eld oa Gb kh cela ons 63.2 61.5 64.1 Kieselguhr Aen amet at LS + ae 25.3 36.9 — TCA me tede sis aes pie Unie a es 42.6 50.0 — Tale Ve pps aks We eee ee eee 48.1 Ho 7% |) BO.Y Charcoal week AR ee HGR ee 55.4 56.0 — Asbestos es a5 ae ae bra vie 57.9 49.9 BT 58) French chalk Sn Pe a se ILO o a) ol 31.5 Kieselguhr (local) .. a a ab 9.4 15.6 — Kaolin SNE esd. Ua ee nese ere 8.0 BS} 5 8 5) Mullerispeeanth ys a) eae) Ve Bee 9.2 8.8 —_— Agar powder we oes Bie olomey atn P Gnaionaet 53.1 — The results show that the mineral colloids with charcoal and agar not only did not assist the diastase to saccharify the starch, but also that they varied in their activity in depressing the saccharification. This point has been noted, to some extent, by other investigators in regard to certain mineral and organic BY R. GREIG-SMITH. 355 colloids. Pincussen (Bot. Abstracts, 1924, 692) reported that a number of enzymes including the diastases were depressed by colloidal ferric hydroxide, and Hagihara noted that cholesterol depressed diastatic action. It is possible that if the enzyme had been allowed to act more slowly, the colloids might have had time to exert some influence. To test this point, a weaker extract of malt was taken and the tests were incubated at a lower temperature, viz. 37°. The 2% solution of malt was diluted ten times, so that the ratio of malt to starch was 0.5 to 100 as against 1.25 to 100 in the previous table. Twelve flasks were used in the experiment, four controls and four each with asbestos and animal charcoal. Table xviii—The Slower Fermentation of Starch by Diastase. Maltose from 100 grams of dry starch. | | ARV S wearers Toate OB a Ureere nko: mn es 3 | 4 pole eee DARiay SG aNd DRI TAINS | Control Doe | 78} | 62.2 61.0 Charcoal 41.6 | 48.6 58.9 HO 08} Asbestos 37.5 | 50.2 51.6 56.1 The results are similar to those already obtained and show that, whether by a slow or by a quick fermentation, the colloids such as charcoal and asbestos lessen the activity of the enzyme. The Influence of Colloids upon the Diastatic Activity of Starch-dissolving Bacteria. Raw starch contains a number of starch-dissolving bacteria among which are Bac. mycoides* and Bac. vulgatus. The latter is by far the most numerous and a colony was selected for testing the diastatic activity. It turned out to be Bac. levaniformans, an ally or race of Bac. vulgatus, that produces an opalescent solution of gum-levan when grown in fluids containing saccharose, peptone, sodium phosphate and potassium chloride. The bacillus was grown in a nutrient fluid consisting of 1% of meat extract with 0.15% of potassium phosphate and brought to a pH value of 6.8. Kaolin and asbestos were chosen for testing the hydrolytic effect. As in most of the previous experiments, each test had a flask to itself; there were twelve flasks each containing 50 c.c. of fluid, 2 grams of soluble. starch and 0.25 gram of colloid except in the case of the controls. Although the organism is an active starch dissolver, it did not promise to show well in the experiment for the reason that in bouillon and in meat extract Table xix.—The Action of Bac. levaniformans upon Soluble Starch. Maltose from 100 parts of dry starch. | | LD) DiS iu aster emus ctee Mcteencs Reman Sere 2 | 3 4 5 | — | Control Bee ata wien heen chet 48.5 | 59.5 62.6 69.0 Kaolin are rep tir eis 50.1 64.1 67.2 69.3 Asbestos Re cree ae 48.0 59.1) 64.9 68.5 * Bac. mycoides and another starch-dissolving bacterium, both isolated from starch, were examined, but they did not seem to be capable of converting soluble starch to maltose. The products of fermentation, after clarification, did not reduce Fehling’s solution in a normal manner like maltose. During the boiling the solution became blackish and suddenly turned to a yellow-green colour. 356 THE INFLUENCE OF CERTAIN COLLOIDS UPON FERMENTATION, ii, it forms a strong cohesive film on the surface of the fluid, with very little sub- surface growth. If contact between the colloid and the bacterium is a desideratum, it would not occur to any extent with this bacterium. The results were according to promise and the colloids tested had little or no influence. Kaolin did accelerate a little but asbestos did not. At a later date the matter was again tested with another culture of the same organism isolated from sugar. Each test flask contained 50 ¢c.c. of a medium consisting of meat extract 0.5%, monopotassium phosphate 0.15% and sodium chloride 0.5% brought to a pH value of 6.5. Each flask also contained 2 grams of soluble starch and, excepting the controls, 0.25 gram of colloid. The fermented fluid was transferred to a 200 c.c. graduated flask with the aid of 50 c.c. of water, 6 c.c. of basic lead acetate were added and then strong spirit to volume. After shaking, cooling and adjusting to volume, the fluid was filtered into a graduated jar and a known volume of the filtrate was transferred to a basin, treated with 10 c.c. of 10% potassium oxalate and 0.5 c.c. of dilute acetic acid and evaporated until the spirit had been expelled. The residual fluid was made to 200 c.c. and filtered. The reducing sugar in the filtrate was determined as maltose.* The treatment of the fermented starch with basic lead acetate and spirit was distinctly advantageous in obtaining a clear filtrate. Table xixa.—The action of Bac. levaniformans upon Soluble Starch. Maltose from 100 parts of dry starch. TD ASViSie We NER Ola beck cy Mukce een eyed 2 3 4 5 Control Aa SRA R Be 16.5 56.4 58.5 55.4 Animal charcoal Ae ato 19.5 60.2 59.2 59.4 Fuller’s earth .. as LPR R Oh Seg 60.8 62.9 59.6 Silica ae 2iky tein ero uy iets 19.8 AG.6 || B3.8 55.6 Asbestos AMIS wheel eve 17.9 56.9 55.9 Ba} 5 I . It is evident that the mineral colloids and charcoal have a distinct influence in accelerating the diastatic activity of the bacterium more especially in the early stages of the fermentation. The most indifferent was asbestos which, as in Table xix, had little or no action. Animal charcoal and fuller’s earth were the most efficient accelerators. A test was made to see the action of talc upon the diastatic activity of Bac. levaniformans. Two flasks, each containing a litre of medium similar to that Table xixb.—The Influence of Tale in accelerating the diastatic Activity of Bac. levaniformans. Maltose from 100 grams of dry starch. | | BIS) EVN S Wats foe stg 4 tases wad A props 3 | 4 | 7 11 | Control SAA bleeds eng oy 42.3 55.6 | 50.5 BS 57 ADEN, 5 Oo ena eae Seema 8 46.9 55.8 51.3 40.7 * Bac. vulgatus is known to produce diastase. Vignal found that it also produced invertase, protease, cytase and rennin. Effront noted that its diastatic enzyme had strong amylolytic and weak saccharifying powers (Waksman, Enzymes of Micro- organisms, Abstr. of Bact. vi (1922), 281). Estienne (through Journ. Soc. Chem. Ind., 1925, B184) notes that it also secretes maltase. Judged by the reducing powers of a solution before and after boiling with dilute hydrochloric acid, the sugar formed by the action of Bac. levaniformans upon starch was maltose. BY R. GREIG-SMITH. 357 used in Experiment xix, were seeded with the bacillus. One flask received 5 grams of gently ignited tale. Portions measuring 100 c.c. were abstracted from each flask from time to time and treated as in Experiment xixa. As with the colloids of Experiment xixa, tale shows an accelerating action in the early stages of the fermentation. In addition, it shows that in course of time the maltose becomes consumed and the colloid tends to inhibit this consumption. It is possible that there may be a balance between the formation and the destruction of the sugar and that the impetus given to the formation by the talc makes it appear as if it inhibited the destruction. The experiments tend to show that the diastatic activity of bacteria, as exemplified by this form of Bac. vulgatus, is accelerated by the presence of some mineral colloids and of charcoal. The evidence is not so pronounced as in other cases, but this may possibly be due to the unsuitability of the organism. It may also be due to the fact that starch itself is a colloid and the effect of the additional colloid cannot be so pronounced as when a non-colloid substrate is employed. Conclusion. Certain mineral and other colloids were tested to see if they accelerated the fermentation of carbohydrates by yeast after the manner that they hastened the fermentation of certain compounds of carbon by the high temperature organism. Silica induced a quicker production of carbon dioxide in the fermentation of saccharose by a distillery yeast. Asbestos and silica hastened the inversion of saccharose by yeast and silica accelerated the fermentation of dextrose. Wood- charcoal, talc, kaolin, fuller’s earth, asbestos, kieselguhr, French chalk and ferric hydrate also hastened the slow fermentation of dextrose. The differences in the relative amounts of fermented material were not so great as when dealing with bacteria, but they were sufficiently marked to show that the yeasts and the bacteria obeyed a rule and were favourably influenced by the presence of these colloidal substances. The inversion of saccharose by Bac. levaniformans, an organism closely related to and perhaps identical with Bac. vulgatus, was hastened by the presence of mineral and other colloids, as silica, fuller’s earth, French chalk, kieselguhr, asbestos, kaolin, tale and charcoal. The inversion of the sugar by diastase, prepared from yeast, was not influenced by these colloids. The ammoniacal fermentation of urea by Bac. fluorescens liquefaciens was assisted by the presence of kieselguhr. The fermentation of urea by a urobacillus was accelerated by most of the previously named colloids and by agar, gelatin, silicic acid, aluminium hydrate and aluminium phosphate. Agar in floccules was more effective than in the fibrous condition. These colloids had no influence upon the fermentation of urea by the enzyme urease contained in an extract from the soy bean, and it did not matter whether the fermentation was quick or slow. The lactic fermentation of dextrose by a lactic bacterium was hastened by most of these colloids; kaolin, gelatin and agar did not act. The diastatic fermentation of starch by Bac. levaniformans was accelerated by kaolin, animal charcoal, fuller’s earth and tale. The acceleration was not pronounced, possibly because starch is itself a colloid, but it was sufficient to show that the bacterial fermentation of starch falls into line with other fermentations. The colloids had an inhibiting influence upon the action of malt diastase on starch. 358 THE INFLUENCE OF CERTAIN COLLOIDS UPON FERMENTATION, ii, It is evident that the influence of the insoluble colloids in accelerating the fermentative activities of bacteria and yeasts is not due to the adsorption of their enzymes. All the colloids were insoluble. When the sols of silicic acid and ferric hydroxide were used, they were coagulated by the salts of the media necessary for the growth of the microorganisms, and therefore acted as insoluble colloids. The soluble colloid, gelatin, feebly stimulated the urobacillus, depressed the alcoholic and had no action upon the acid fermentation of dextrose. The research has shown that yeasts and bacteria have their fermentative activities accelerated by the presence of certain mineral and other colloids such as tale, kieselguhr, silica, fuller’s earth, charcoal and agar and that the isolated enzymes are not influenced by the same colloids. I am indebted to Mr. W. W. L’Estrange for much kind assistance. _- THE HAEMATOZOA OF AUSTRALIAN MARINE TELEOSTEI. By I. M. Macxerras, B.Sc., M.B., Ch.M., Linnean Macleay Fellow of the Society in Zoology, and M. J. MackrrrRas, B.Sc., M.B. (Six Text-figures. ) [Read 30th September, 1925.] Introduction. No blood parasites have been recorded from Australian marine fish, which is surprising in view of their common occurrence in other parts of the world. The “ring bodies’ described by Johnston and Cleland (1909) from the red blood cells of Monocanthus sp. may possibly be protozoa, but their interpretation, that they are centrosomes, is more probably correct. The same authors later (1910) listed 46 marine Teleostei examined with negative results. The findings here recorded serve to bring the Australian haematozoal fauna into line with that of other regions. While many species of Trypanosomes and Haemogregarines are known from marine fish, blood inhabiting Trypanoplasms have, so far as we can determine, only been recorded hitherto from freshwater fish. It is a pleasure to acknowledge our indebtedness to the late Mr. A. R. McCulloch of the Australian Museum, Sydney, who has identified most of our host species. The nomenclature followed here is that given in his “Check List’ (McCulloch, 1922). Part of this work was carried out while one of the authors (I. M. Mackerras) was the holder of the John Coutts and Science Research Scholarships in the University of Sydney. Methods. Smears were made from the heart blood as soon as the fish was caught, dried as rapidly as possible and stained on our return home either with Leishman’s Stain or with a solution of “Soloid’’ Hosin-Azur in pure methyl alcohol. Both methods were used without previous fixation and the latter gave particularly good results, especially if the staining time was extended (3-5 minutes in undiluted, and 15-45 minutes in diluted stain). Conditions were such as to preclude the use of wet-fixation methods, which is to be regretted, as many important details of morphology are not seen satis- factorily in dried films. Having in mind the use which may be made of parasites in determining host relationships, we have consequently refrained from discussing the affinities of the parasites here described. Fish blood is remarkable for the extreme rapidity with which it clots, either in the body after death, or on being drawn from the living animal. Films must therefore be made as soon as the fish is caught, and haste is necessary once the heart is opened in order to obtain even three or four satisfactory preparations. D 360 THE HAEMATOZOA OF AUSTRALIAN MARINE TELEOSTEI, Material. The material was collected at Sydney and at Broken Bay, near Sydney, . N.S.W., during the past four years at such times as opportunity offered. Fish were examined at all seasons, the majority, however, during the winter, and parasites were found in the months of January, May, June and November. Two hundred fish belonging to 31 species and 27 genera have been examined. The following list gives those species whieh have been searched with negative results, together with the number of individuals of each species examined. It is to be noted that the infections found have always been scanty, consequently we may have missed an occasional parasite in those smears classified as negative, although each was submitted to a prolonged and careful search. ANOTAKSIGUS SGARTEUS (Olie% eyacl Wee WARP soobsonocccngoucgeoudsooUunOKS ICO WOOHMIS Semcowiros (Enon Wie seosnancescnc co oso seus oboabodUonONOOdES Cantherines ayraudi (Quoy and Gaimard). Leather-jacket .................... 1 Cantherines hippocrepis (Quoy and Gaimard). Variable Leather-jacket ........ Cantherninvesmsp a @VOUune) -ueather=iACke aera niece niente rere aenene Ciidoglanus Megastomus mGRICh) py eo Stilarye Cab HS hierererereetel enero cine nero Dactylopagrus morwong (Ramsay and Ogilby). Morwong .................+5. Dicotylichthys punctulatws Kaup) Porcupine) Mish) ). 22.1. ees eeia. a cis selec ae TPO OSUS CTO CINE) (OlGl Withifey 55 eccasacnonobonuonbodcnoobausand00ODS Happocampusmiovuacnollandiagemsteind= Seas Horses nein oeriaeieieieiaenien 1 Le noo IONS lcinocings (CAst@login)), Conese IDS ssssdscaccactousounnooocnons Lotetlacatlarias: Gunth; | Beardies ss of iho os ae ng tote oeiete eee tie tela operas ontueete Neoplatycephalus macrodon Ogilby. Tiger Flathead ...................-.----- Ooi ainoOlgois helms (Cir, eral Web. WHO s5acés005ncccsouncocooenuEs POOR OSOMUSNOUL AUS a GOLSLCl) ES naDDeLE eee 2 Blatycephatusyfwscus Guy. and val) Duskyveliatneadmrresi-icieiecieieniaieiceneieleiereneons Pseudolabrus gymnogenis (Gunth.). White-spotted Parrot Fish .............. LARC OCI TOM OOM UCKHtD (RMON), ILEUORR sosccagsuosonsuddecodonsuedsouec ScrenamantarcticasCastelnall ae wils hee eee ee eee ae eee 1 Scomber australasicus Cuy. and Val. Mackerel Scorpaena cardinalis Rich. Red Rockeod Scorpis lineolatus Kner. Sweep Spheroides hamiltoni (Rich.). Common Toado Trachichthodes affinis (Gunth.). Nannygai Trachwurusiadeclivus (senyns)-) Yellow-tailue seen ae ocean eee ene Upeneichthys porosus (Cuv. and Val.). Blue-striped Red Mullet Verreo oxycephalus (Bleeker). Pig-Fish CC Ce PNP MMP OANMNRrP HE AAT PRP HEH PrP EOP HE NODS TAH Parasites have been found in the following:— Aulopus purpurissatus Rich. Sergeant Baker. (Fam. Aulopidae, Ord. Iniomi.). 8 examined, Trypanosomes in 1, Haemogregarines in 1. Gilbertia annulata (Gunth.). Banded Sea _ Perch. (Fam. Serranidae, Sec. Perciformes, Subord. Percoidea, Ord. Percomorphi.). 3 examined, Haemo- gregarines in 1. Gilberlia semicincta (Cuv. and Val.). Half-banded Sea Perch. 38 examined, Trypanosomes in 8, Haemogregarines in 4. jf Parma microlepis Gunth. White-ear. (Fam. Pomacentridae, Sec. Pomacentriformes, Subord. Percoidea, Ord. Percomorphi.). 9 examined, Trypanosomes in 1, Try- panoplasms in 8, Haemogregarines in 7. The total number of individuals infected was 15, or 7.5% of all fish examined. BY I. M. MACKERRAS AND M. J. MACKERRAS. 361 Text-figure 1. Trypanosoma pulchra, n. sp.; a, b and c from Gilbertia semicincta (Cuv. and Val.); d, e and f from Parma microlepis Gunth.; g, small form from Gilbertia semicincta (Cuv. and Val.). x 1180. Haemoflagellates. TRYPANOSOMA PULCHRA, Nl. SD. Hosts: Gilbertia semicincta (Cuv. and Val.) and Parma microlepis Gunth. Present in 3 out of 38 of the former and 1 out of 9 of the latter. Sydney and Broken Bay, N.S.W. January, May and June. An average individual from the type slide is shown in Fig. la and has the following measurements: Length of body 48.34, diameter of body at level of trophonucleus 3.54, kinetonucleus is 1.74 from aflagellar end and measures 1.6u by 0.8u, trophonucleus is 22.54 from aflagellar end and measures 4.0u by 2.94, diameter of undulating membrane 1.0u to 1.5u, free part of flagellum 7.34. The body length varies from 57.1u to 40.8%, but the relative proportions of the parts and the general appearance of the individuals remain constant within fairly narrow limits, except for two small specimens to be described later. The cytoplasm of the body is rather dense and stains a deep blue with Eosin-Azur. It shows irregular lighter and darker areas and a number of small scattered vacuoles. The aflagellar end is beak-shaped and stains a very pale blue. The undulating membrane is clear and takes a pale pink stain. The flagellum appears to arise directly from the deep red kinetonucleus, there being no separate basal granule visible. The trophonucleus stains uniformly pink and shows no sign of a karyosome or of chromatinic granules. In one infection (G. semicincta No. 26), two very small individuals were seen along with the ordinary form (Fig. 1g). Their lengths are 27.24 and 26.54 and they differ from the common form in that the flagellum is relatively longer, the undulating membrane is relatively wider (being in both cases about the same as in the larger type), and in the condition of the trophonucleus which shows a number of dark granules scattered throughout, but rather more numerous at the ends. What relation these bear to the common type, or whether they are possibly a distinct species, we are at present unable to say. Similarly, we cannot decide positively whether the infections from the two hosts represent one or two species. Though the hosts are not very closely related, they are found together in the same situation on rocky bottoms and the parasites are morphologically indistinguish- able. Consequently, we see no valid reason for separating them at present. 362 THE HAEMATOZOA OF AUSTRALIAN MARINE TELEOSTEI, Type slide (G. semicincta No. 35g in our series) in the Australian Museum, Sydney. Text-figure 2. Trypanosoma aulopi, n. sp. x 1180. TRYPANOSOMA AULOPI, N. SD. Host: Aulopus purpurissatus Rich. Present in 1 out of 8. Sydney, N.S.W. May. This species, of which we have but one scanty infection, is separated from T. pulchra, nov., because of its very marked pleomorphism. In general body form, proportions, and staining reactions it is similar to the common larger form of T. pulehra, nov., but individuals of widely different sizes are encountered with equal frequency. The following are the body lengths (in mw) of nine consecutive individuals arranged in order of size:—57.1, 53.0, 50.3, 47.5, 44.8, 36.3, 33.5, 33.5, 29.1. Nothing was found quite as small as the small specimens from G. semicincta. The host belongs to the Iniomi, an order widely separated from the Percomorphi, to which the hosts of 7. pulchra, nov., belong. It is, however, also a rock fish living in a not very different habitat and there are definite possibilities of cross infection occurring. Taking everything into consideration, and having in mind the usual practice amongst haematozoologists, it seems most reasonable in the present state of our knowledge to propose a separate name for this species, leaving it for future work with much more extensive series from all three hosts to decide the identity or otherwise of the various forms. Type slide (A. purpurissatus No. 1 of our series) in the Australian Museum, Sydney. Text-figure 3. Trypanoplasma parmae, n. sp. x 1180. TRYPANOPLASMA PARMAE, Nl. SD. Host: Parma microlepis Gunth. Present in 3 out of 9. Sydney, N.S.W. November. BY I. M. MACKERRAS AND M. J. MACKERRAS. 363 Owing to the soft and mobile nature of the body most of the parasites become somewhat distorted in dried films and the following description is based on those specimens which had obviously retained their natural shape more or less com- pletely. This species is monomorphic and of a somewhat broad type, varying from 12.5u long by 3.8u wide to 14.74% long by 5.0u wide. The cytoplasm stains a faint blue and is finely and irregularly granular. Anteriorly it is pale and indistinctly vacuolated, whilst behind the trophonucleus it is denser and contains numerous round or irregular chromatoid bodies, 0.254 to 1.254 or occasionally more in diameter; these sometimes extend round the anterior margin of the tropho- nucleus. Along the full length of the dorsal edge of the body, and marking the origin of the undulating membrane, is a narrow zone of densely packed fine granules which stain a reddish colour with Hosin-Azur. The trophonucleus, which is situated behind the middle of the body, is reniform or occasionally oval in shape, measures 3.7yu to 3.0u by 1.5u to 2.34 and is fairly densely packed with large irregular chromatinic masses. The kinetonucleus is a large oblong or oval mass, from 1.7u to 2.0u long by 1.0u to 1.34 wide, and is situated close to the ventral surface about 2u from the anterior end of the body; it stains an intense reddish-black and rarely gives any indication of internal organization. The blepharoplast is a minute darkly stained granule, which usually appears single but is occasionally clearly double; it lies close to the anterior extremity of the body. No definite rhizoplast could be made out in our preparations. The anterior free flagellum measures from 18.04 to 25.0u in length. The posterior flagellum is attached to the body by a very narrow undulating membrane which is little more than 0.25u wide; its free terminal part measures 18.0u to 26.8u. Many of the specimens in our slides were rounded and some showed irregular bulgings as if about to burst. Such distorted examples are from 9.5 to 11.14 long by 6.9u to 8.54 wide, the measurements of other parts being similar to those given above. In one or two specimens a number of clear unstained vacuoles, lu to 2u in diameter, were seen; these had very sharply marked outlines and gave the impression of being signs of impending dissolution rather than of any normal process. The special interest attaching to this parasite is that, while Trypanoplasma (or Oryptobia) has long been known as a parasite of invertebrates, of the alimentary tract of frogs and marine fish, and of the blood of freshwater fish, this is, so far as we are aware, the first record of its occurrence in the blood of a marine fish. Type slide (P. microlepis No. 4d of our series) in the Australian Museum, Sydney. Haemogregarines. : HAEMOGREGARINA AULOPI, Nl. Sp. Hosts: Aulopus purpurissatus Rich. and ? Parma microlepis Gunth. Present in 1 out of 7 of the former and ? 2 out of 9 of the latter. Sydney, N.S.W. May. This parasite was present in the same slide as Trypanosoma aulopi, nov., and was not nearly so scanty. It is small and is not so uniform in shape as is the case with most Haemogregarines. In certain respects it resembles the little known genus Pirhaemocyton Chatton et Blanc, which is parasitic in Geckos. Three distinct types of parasite are present in this infection, large oval forms, smali rounded forms and slender forms. The large oval forms (Fig. 4a) are least common, measure 4.54 by 2.94, and occur singly. The outline is definite _-~ ts 364 THE HAEMATOZOA OF AUSTRALIAN MARINE TELEOSTEI, and the cytoplasm clear, except for a peripheral zone which is denser and stains blue; sometimes delicate branching strands of denser blue cytoplasm invade the clear central area. The chromatin is collected for the most part in blocks at the periphery, but strands of linked granules extend in many cases into the central area. The rounded forms (Fig. 4, 6 and d) most frequently occur two in a cell, but single infections are often met with, and once four were seen in one cell. The size seems to bear some relation to the number present in the cell, single infec- tions measuring about 3.3u by 2.54, double infections are slightly smaller, not more than 3.0u long, usually less, while in the quadruple infection the parasites measured 2.3u by 1.8u. The cytoplasm is similar to the larger form, but the chromatin is more densely aggregated into peripheral blocks or bands; one specimen was intermediate between the two forms in size and chromatinic arrange- ment. It seems likely that the smaller forms are the products of division of the larger, fission being into two, and these may divide again, giving rise to the four very small parasites described above. The slender form (Fig. 4, c and d@) always occurred in pairs. In one case a pair of the slender form and a pair of the rounded form occupied the same cell (d@). They were fairly uniform in size, measuring on an average 4.74 by 1.5u. Their outline is always more delicate than in the other forms and the chromatin is collected into a block at one end with an occasional granule at the periphery further along the body. Nothing was seen linking these with the other forms, and what stage they represent we cannot say at present, though they may be the gametocytes. In Parma microlepis No. 6a, a single cell was found containing four parasites rather closely resembling those shown in Fig. 4d, and in Parma microlepis No. 7a, one cell was also seen containing two short rather shrunken parasites. These are quite possibly H. awlopi, nov., but, as no other parasites of this type were found in all our material from Parma microlepis, one cannot be sure of the identification. Type slide (A. purpurissatus No. 1 of our series) in the Australian Museum, Sydney. Text-figure 4. Haemogregarina aulopi, n. sp. a, b, ¢ and d, different types of infections; e, normal host cell. x 1400. Text-figure 5. Haemogregarina gilbertiae, n. sp. a, b and c, intracellular forms; d@ and e, vermicules; f, normal host cell. x 1400. BY I. M. MACKERRAS AND M. J. MACKERRAS. 365 HAEMOGREGARINA GILBERTIAE, Nl. Sp. Hosts: Gilbertia semicincta (Cuv. and Val.) and Gilbertia annulata (Gunth.). Present in 4 out of 38 of the former and 1 out of 3 of the latter. Sydney and Broken Bay, N.S.W. January and November. This species is larger and more uniform in shape than H. aulopi, nov. It is usually single, but double infections frequently occur, and once three parasites were found in one cell (Fig. 5c). The shape is broadly oval and the size is in general very uniform, being about 4.2u by 2.54. Smaller or larger forms are rare; the smallest individual seen measured 3.5u by 2.0u, while a few longer, somewhat slenderer individuals were found which measured up to 5.6u by 2.2u. The parasites in single infections were slightly larger than those in double infec- tions in which the members of a pair were almost always of exactly the same size. The cytoplasm is clear and only stains at the periphery. The chromatin is arranged in a loose skein towards one end of the parasite and is sometimes more concentrated in blocks at the sides (Fig. 5b); occasionally it is widely spread out in the form of scattered granules. Two free parasites were seen, the first (Fig. 5d) apparently being an early stage in the development of a free vermicule. It measures 4.34 by 3.4u and consists of two parts separated by a band of chromatinic granules. The larger part resembles the ordinary intracellular parasite, while the smaller tapering part stains like the fully developed vermicule. This second form (Fig. 5e) measures 10.4u by 1.94; its outline is delicate and the cytoplasm stains a pale greyish-blue; the chromatin is collected into two masses of granules towards one end. Both ends are rounded and the body has not the streamline form usual amongst vermicules as exemplified by the parasites shown in Fig. 6g. The intra- cellular parasites could not be differentiated into gametocytes and schizonts. Type slide (G. semicincta No. 27d of our series) in the Australian Museum, Sydney. Text-figure 6. Haemogregarina parmae, n. sp. a, normal host cell; 6. c and d, schizonts; e and f, intracellular gametocytes; g, free gametocytes. x 1400. HAEMOGREGARINA PARMAE, Nl. SD. Host: Parma microlepis Gunth. Present in 7 out of 9 examined. Sydney, N.S.W. November and June. The more numerous and more abundant infections available have enabled us to study this species in greater detail than was possible in the case of the other 366 THE HAEMATOZOA OF AUSTRALIAN MARINE TELEOSTEI. species described above. Two distinct stages are present in our preparations; they can be allotted with little doubt to the schizogenous and gametogenous generations respectively. The schizonts occur most frequently three or four in a cell; occasionally single or double infections are to be met with, and very rarely as many as six or seven parasites occupy a single cell, these being probably the result of a double initial infection. Since our infections are never so abundant as to make it probable that double infections are more than a very occasional occurrence, the presumption is that the three or four parasites usually found have originated from a single initial infection. There is usually definite enlargement of the infected cell, the amount of enlargement and the size of the contained parasites probably being an indication of the duration of the infection of that cell. The individual parasites (Fig. 6, b and c) vary from 6.2u to 9.6u long, about 8u being the commonest, by from 1.24 to 1.74 wide. The body is delicate in outline, tapering more towards one end than the other and is. usually somewhat curved. The cytoplasm stains a delicate blue, paler towards the centre which, occasionally, has the appearance of a definite oval vacuole. The chromatinic material is aggregated towards the middle with strands and granules extending into the more attenuated end of the body. The gametocytes are more definite in outline and stouter bodied than the schizonts. When contained within a cell (Fig. 6e) both ends are usually rounded, but in forms about to leave the cell (f/f), and in free forms (g) the posterior end tapers to a point, while the anterior end is broadly rounded; several transition stages between the type shown in (e) and those shown in (g) are to be seen in our slides. The intracellular gametocytes averaged 7.1lu by 1.94, and the free vermicules varied according to their development, the largest being 13.4u by 2.94 and the usual size 11.54 by 2.34. The cytoplasm stains much darker than is the case with the schizonts, and always has a pinkish tint which they never show; it is somewhat blotchy in appearance and the anterior end is always darkly stained. The chromatin may extend throughout the posterior half of the body or be collected into a mass of blocks and granules near the middle. It is not possible to differentiate between male and female gametocytes in our material, nor was there any indication of the occurrence of a process of maturation. The gametocytes must leave the cells very rapidly on the blood being drawn, because a considerable number of free forms were seen, which is remarkable in view of the haste with which smears were made. Of course, they may exist free in the blood, but this is not the case with other species of Haemogregarine which we have been able to observe in the living state. The proportion of schizonts to gametocytes varies in the different infections, those showing the most gametocytes being probably the oldest. | Type slide (P. microlepis No. 9c of our series) in the Australian Museum, Sydney. References. JOHNSTON, T. H., and CLELAND, J. B., 1909.—Notes on some Parasitic Protozoa. Proc. LINN. Soc. N.S.W. xxxiv, pt. 3, p. 501. JOHNSTON, T. H., and CLELAND, J. B., 1910.—The Haematozoa of Australian Fish. Journ. and Proc. Roy. Soc. N.S.W. xliv, p. 406. McCuutocnH, A. R., 1922.—Check List of the Fish and Fish-like Animals of New South Wales. Australian Zoological Handbook No. 1. Published by the Royal Zoological Society of N.S.W. THE GASTEROMYCETES OF AUSTRALASIA. iii. THE GENERA BOVISTA AND BOVISTELLA. _ By G. H. CunnineHam, Mycologist, Dept. of Agriculture, Wellington, N.Z. (Plate xxxvii.) [Read 30th September, 1925.] Both Bovista and Bovistella are included in the family Lycoperdaceae. Although species of each genus are not numerous, they are fairly widely distributed, save the Australasian species which, with one exception, are limited to this biologic region. The name Bovista was first used by Dillenius (1719) and later in a generic sense by Persoon (1801); but it was not until the appearance of Morgan’s paper (1892) that members of the genus came to be recognized in their present sense. Prior to this the name appears to have been loosely applied to any lycoperdaceous plant without a sterile base, as reference to the Gasteromycetes in Saccardo’s Sylloge Fungorum (1888) will show, for herein are included under Bovista members of the genera Lycoperdon, Bovista, Bovistella, Catastoma and Calvatia. Cooke (1892) likewise appears to have had but a vague idea as to what should be included in Bovista for, according to Lloyd (1905), of the seven Bovistas included in this work, five are Catastomas and one is a Calvatia. On account of the habit of members of the genus, Lloyd (1902, a) places it in the tribe Bovisteae, or “tumblers.” Bovistella was erected by Morgan (1892) as a monotypic genus. His con- ception of the genus was one possessing the capillitium of Bovista with, in addition, a sterile base. In his earlier papers Lloyd (1902, a, 6b) claimed that the presence of a sterile base was not a good character on which to separate these two genera (basing his argument on the fact that it has not been considered of sufficient moment to separate Globaria from Lycoperdon) and proposed to separate the two upon habit, Bovista being a genus in which plants break away at maturity from the point of attachment, Bovistella persisting as does Lycoperdon. The writer would follow Lloyd’s emendation in so far as the habit is concerned, for it is the only character upon which separation is possible. In a later paper Lloyd (1905) further emends Bovistella to include all plants possessing a rooting base together with either pedicellate spores or capillitium of the Bovista type or both. Thus many species included in Lycoperdon (possessing the capillitium of Lycoperdon) are by Lloyd placed in Bovistella. He claims (1906, a) it would simplify matters to take from the genus Lycoperdon (which is a large and unwieldy genus) a natural section and place it in another genus with which the section agrees in a prominent character and in which it differs from the remainder of the genus Lycoperdon. The writer cannot agree with Lloyd in this grouping, for capillitium characters are of greater importance than spore characters, especially when they are so well defined as in the different genera of E 368 THE GASTEROMYCETES OF AUSTRALASIA, the Lycoperdaceae. It would simplify matters to erect a section under Lycoperdon for those species possessing pedicellate spores, and thus save the confusion that would otherwise be caused. Structure of the Mature Plant. A mature plant consists of two groups of tissues, (a) peridium, (0) gleba. (a) Peridium.—This is invariably globose, elliptical or depressed-globose and consists of two distinct tissues, (1) the exoperidium, (2) the endoperidium. The exoperidium in the mature plant usually falls away, leaving the endoperidium exposed, or may be partially persistent as in Bovista purpurea, here appearing in the nature of a much broken, thin and fragile outer covering (Plate xxxvii, fig. 2). It consists either of loosely interwoven non-gelatinized hyphae, or else pseudo- parenchymatous tissue. Frequently it persists on the endoperidium as minute scales or blunt points, giving to this structure a roughened appearance. The endoperidium is in the nature of a very thin, papyraceous membrane enclosing the gleba and perforated by a single apical orifice; it varies in colour from bay-brown to purple, the colour being fairly constant with the species. Microscopic examination shows the endoperidium to consist of intricately inter- woven, gelatinized hyphae and in addition (in Bovista brunnea) are present numerous ramifying hyphae, several times the diameter of the normal hyphae, thick-walled and filled with coarse granular matter. The apical orifice (mouth) through which the spores escape at maturity is usually circular or elliptical in outline and may be plane or surrounded by a slightly elevated, toothed margin. (b) Gleba.—The whole of the tissue enclosed within the endoperidium is termed the gleba. At maturity it consists of a mass of short branched capillitium threads immixed with very numerous spores. Hach thread is of a peculiar structure, consisting of a thick stout stem from which arise numerous dicho- tomously branched acuminate lateral branches (Plate xxxvii, fig. 1). These structures are characteristic of the genera Bovista and Bovistella. The wall of the thread varies in thickness and colour and is usually perforated with numerous pits or depressions, which penetrate to the lumen. Their function is unknown. The threads vary according to the species in the degree of branching, thickness of the wall of the main stem, length of the branches and presence or absence of pits. In all the species examined no septa are present. In the mature plant these threads appear aggregated into small balls, giving to the ripe gleba a granular appearance. The spores are obovate or globose, thick-walled, 5-64 in diameter, and possess either a rough or a smooth epispore. All are some shade of brown. To each is attached the sterigma (absent in Bovistella pusilla) an appendage which is hyaline, varies in length from 10 to 25u, and is usually acuminate. Dehiscence of the spores is effected through the mouth, but the exact manner of emergence is unknown. The capillitium threads are slightly hygroscopic and possibly owing to this and to the expansion and contraction of the endoperidium brought about by changes in atmospheric humidity or temperature, spores are forced through the opening and dispersed by the wind. In Bovista plants break away at maturity from the point of attachment and are rolled over the ground by the wind, often for considerable distances. Doubtless numerous spores are forced out through repeated blows upon the endoperidium during the rotation of the plants when being carried along in this manner. As Bovistella remains attached to the substratum at maturity, this probable method of spore discharge BY G. H. CUNNINGHAM. 369 cannot occur. The development of one of the species of the genus Bovista is being published elsewhere. Artificial Key to the Genera. Plants breaking away from the point of attachment at maturity ...... Bovista Plants remaining attached to place of origin, not breaking away at maturity SSRN Oth OS ee Orne OOO CTOR aD DECRORO OIE ICRC IOUS OFF ON erin En Bahu Pirie Setar Bovistella Bovista Dillenius ex Persoon. Syn. Meth. Fung., 1801, p. 136. Plants breaking away from the point of attachment at maturity. Peridium globose, subglobose or shortly pyriform; consisting of an outer, usually fugacious exoperidium, and a membraneous, tough, firm endoperidium, which dehisces by an apical, definite or indefinite mouth. Gleba without a sterile base; capillitium of free threads, each consisting of a thick stem and dichotomous, tapering, acuminate branches. Spores coloured, continuous, rough or smooth, globose, obovate or elliptical, pedicellate or apedicellate; basidia tetrasporous. Habitat.—Solitary on the ground. Distribution.—HKurope; North and South America; Australia; New Zealand. Two species only are known in Australia and/or New Zealand, both being confined to this biologic area. 1. BovistA BRUNNEA Berkeley. Flora N.Z., vol. 2, 1855, p. 119. Peridium depressed-globose, up to 2.5 cm. diam., with a minute rooting base which usually falls away at maturity; exoperidium white, evanescent; endo- peridium chestnut-brown or pallid umber-brown, firm, smooth, shining. Mouth up to 2 mm. diam., irregularly circular or indefinite and irregularly torn, slightly erumpent, toothed or entire, frequently almost plane. Gleba pallid ferruginous-brown; capillitium of the usual type but more scantily branched, walls thin and pitted. Spores globose or obovate, 4-6u, pedicel tinted or hyaline, acuminate, 10-134 long; epispore pallid-ferruginous, closely and finely verruculose, 1.54 thick. Habitat.—Solitary on the ground. Distribution.—Australia; New Zealand: Shore of Lake Te Anau, Otago, Jan., 1920 (EH. H. Atkinson), Methven, Canterbury, Feb., 1925, Immature (J. C. Neill). This species is characterized by the firm, dark-brown, smooth and shining endoperidium, stout, sparingly branched, thin-walled capillitium, and pallid, distinctly verruculose spores. Two collections from Lake Te Anau were forwarded to Lloyd for deter- mination (Nos. 73, 572). He determined one as Bovista brunnea, the other as Bovistella bovistoides. The writer has since ascertained that both collections consisted of these two species. Lloyd (1906, 0) believes this species to be an old weathered form of Bovista tomentosa Vitt., for he finds, in certain Australian collections, the immature specimens possess a finely spinous exoperidium. As there is no indication in Vittadini’s work that he described immature plants, the writer fails to see where the connection between the two lies, for this reasoning may equally well be applied to Bovistella bovistoides, as the immature plants of this species also possess a finely tomentose exoperidium. 370 THE GASTEROMYCETES OF AUSTRALASTA, The type was collected by Colenso on the banks of the Manawatu River, Wellington, N.Z., and is now preserved in the Kew Herbarium. The writer has seen no specimens of this species from Australia. 2. BOVISTA PURPUREA Lloyd. Mycological Notes, 19238, p. 1201, Figs. 1, 2, 5, 9. Peridium globose or depressed-globose, up to 2.5 cm. diam., with a small pulvinate rooting base; exoperidium thin, dingy-white or brown, evanescent, falling away in irregular flakes or scales, but partly persisting towards base; endoperidium usually lead-coloured, often purplish, smooth, firm, shining, dehiscing by an apical, irregularly torn, circular or elliptical mouth, which is erumpent, irregularly toothed and up to 5 mm. diam. Gleba purple-brown; capillitium of the usual type, but stout, freely branched and not pitted. Spores obovate or globose, 5-6u diam., pedicels hyaline, acuminate, up to 124 long; epispore chestnut-brown, verruculose, lu thick. Habitat.—Solitary on the ground. Distribution—New Zealand: Mapua, Nelson, May, 1922 (G.H.C.; Type coll., Lloyd No. 700), Ashburton, Canterbury, Jan., 1925 (J. C. Neill), Blenheim, Marl- borough, May, 1925 (J. C. Neill). This species is characterized by the dark colour, flaking endoperidium, distinctly verruculose spores and stout, strongly branched, non-pitted capillitium. Excluded Species. Of the seven species recorded in Cooke’s Handbook (1892) one, B. brunnea, is valid; B. Muelleri, B. hyalothrix, B. hypogea, B. anomala and B. cervina should be placed in Catastoma, and B. olivacea in Calvatia. BovIsTELLA Morgan. Journ. Cincinnati Soc. Nat. Hist., vol. 14, 1892, p. 145. Plants remaining attached to place of origin, not breaking away at maturity; with a well developed rooting base. : Peridium globose or pyriform, of two layers; an external, thin, usually fugacious exoperidium, and an inner thin, flaccid, membraneous endoperidium which dehisces by an apical definite or indefinite mouth. Gleba with or without a well defined sterile base; capillitium of free threads, each consisting of a thick stem and dichotomous, tapering, acuminate branches. Spores coloured, continuous, rough or smooth, globose, obovate or elliptical, pedicellate or apedicellate. Habitat.—Solitary on the ground. Distribution.—HEurope; North America; India; Australia; New Zealand. Artificial Key to Species. Spores pedicellate— Peridium pallid-tan, finely, tomentose <............-.-..¢- 1. B. verrucosa. 1Peroobinten GClayelke joo, BIeKONENS oo oc00scns0d00d050000GGns 2. B. bovistoides. SPORES! ADE GI CET AEC mac aiciec. as che ccuahineme RCM OER eae once ee ee 3. B. pusilla. 1. BovVISTELLA VERRUCOSA, n. sp. Plate xxxvii, figs. 3, 7. Peridium globose or shortly pyriform, up to 15 mm. diam.; with a strong, well-marked rooting base; exoperidium in the nature of a very delicate layer, soon more or less completely flaking away; endoperidium dingy-white or pallid-tan, minutely and delicately tomentose, appearing almost smooth, very thin and fragile, flaccid, opening by an apical, irregular, indefinite plane mouth. BY G. H. CUNNINGHAM. 371 Gleba bay-brown, sterile base absent; capillitium threads of the usual type, pitted. Spores globose, 4-6, pedicels hyaline, acuminate, up to 12u long; epispore tinted, finely and closely verrucose, 1.5u thick. Habitat—Solitary on the ground. Distribution—South Australia: Monarto South, Sept., 1922, J. B. Cleland. Type, in Herb. Cleland. The small size, pallid colour and tomentose nature of the endoperidium and rough pallid spores characterize this species. 2. BoVISTELLA BOVISTOIDES (Cooke and Massee) Lloyd. Plate xxxvii, figs. 4, 6, 8. Myc. Notes, 1906, p. 247.—Mycenastrum bovistoides Cke. et Mass., Grev., vol. 16, 1888, p. 26—Scleroderma bovistoides (Cke. et Mass.) De Toni, in Sacc. Syll. Fung., vol. 7, 1888, p. 489. Peridium globose, depressed-globose or shortly pyriform, up to 2 cm. diam., with a strong rooting base which frequently attains a length of 1.5 cm.; exoperidium thin, white, persisting as small areolate areas over the upper part of the endoperidium, but scanty or absent from lower, frequently falling away completely when the endoperidium appears marked with lines arranged in an areolate manner; endoperidium flaccid, dull bay- or chestnut-brown, darker basally; mouth apical, usually elliptical, frequently indefinite, slightly erumpent and toothed, sometimes almost plane. Gleba olivaceous, becoming umber, sterile base absent; capillitium of the usual type, threads much branched, thick-walled, pitted, dark chestnut. Spores globose, seldom obovate, 4-64 diam., pedicels tinted, attenuate, up to 15y long; epispore chestnut-brown, minutely and delicately verruculose, almost smooth, lp thick. Habitat.—Solitary on the ground. Distribution.—India; Australia; New Zealand. South Australia: Flinders Range, near Pt. Augusta, Aug., 1922 (J.B.C.), Mr. Zeitz coll. (Herb. Clel.); New South Wales: Wagga, July, 1914 (J.B.C.); New Zealand: Banks’s Peninsula, Dunedin, Otago, July, 1922 (Miss H. K. Dalrymple), Tapanui, Otago, March, 1922 (GeH.C:)). Different collections of this species vary in the size and extent of the areolate markings. As it is impossible from the material at hand to separate any one collection from another, it is thought advisable to keep all together. The colour of the peridium varies much according to the age of the plant, so that it cannot be used as a specific character. The rooting base separates this from Bovista brunnea, which otherwise it resembles closely. As this structure may frequently be absent from old weathered specimens, and as the areolate markings also disappear, it is extremely difficult to place old specimens of either species. The spore markings and capillitium in such cases serve as the only characters by which separation can be made, for the spores of Bovista brunnea are more coarsely verruculose than those of Bovistella bovistoides, and the capillitium is more scantily branched, more pallid in colour and the walls are thinner. The collection made at Tapanui is so well marked, and has such a small rooting base that the writer was at first inclined to separate it and place it under the genus Bovista, but critical examination of the capillitium and spores shows it to be but a strongly marked form of Bovistella bovistoides with a weak development of the rooting base. 372 THE GASTEROMYCETES OF AUSTRALASIA, The species is characterized by the areolate markings and dark colour of the endoperidium, chestnut, verruculose spores and stout, strongly branched, thick- walled, pitted capillitium. 3. BOVISTELLA PUSILLA Lloyd. Mycological Notes, 1910, p. 457. “Plant globose, 1-1.5 cm. diam.; cortex smooth, peridium thin, flaccid, sterile base none; gleba dark brown; capillitium deeply coloured, with numerous tapering branches. Spores compressed-globose or slightly ovoid, 4 X 4u; deeply coloured, smooth, with short but distinct apiculus. Brisbane, EK. Jarvis”’. The writer has not seen specimens. The description given is that of the original. The species is characterized by the apedicellate spores. Excluded Species. (a) Bovistella aspera (Lev.) Lloyd, Lyc. Aus., 1905, p. 28 = Lycoperdon asperum (Lev.) De Toni, in Sace. Syll. Fung., vol. 7, 1888, p. 119. (6) Bovistella australiana Lloyd, Lyc. Aus., 1905, p. 28. (c) Bovistella cuprica Lloyd, Letter No. 60, 1915, p. 9. This the writer believes to be a Lycoperdon, but the description is so poor as to make the placing in any genus uncertain. (ad) Bovistella glabrescens (Berk.) Lloyd, Lyc. Aus. 1905, p. 28 = Lycoperdon glabrescens Berk., Fl. Tas., vol. 2, 1860, p. 264. (e) Bovistella Gunnii (Berk.) Lloyd, Lyc. Aus., 1905, p. 29 = Lycoperdon Gunnii Berk., Fl. Tas., vol. 2, 1860, p. 265. (f) Bovistella rosea Lloyd, Myc. Notes, 1906, p. 248. (g) Bovistella nigrica Lloyd, Myc. Notes, 1922, p. 1115. All are excluded from Bovistella as they possess the typical Lycoperdon capillitium. They will be dealt with under the genus Lycoperdon in a subsequent paper. Literature Cited. CookE, M. C., 1892.—Handbook of Australian Fungi, 458 pp. London. DILLENIuS, J., 1719.—Catalogus plantarum sponte circa Giessam nascentium, 240 pp. Frankfurt. Luoyp, C. G., 1902, a.—Genera of Gastromycetes, 12 pp. Cincinnati. , 1902, b—Mycological Notes, No. 9, p. 85. ————., 1902, e—The Bovisteae, Mycological Notes, No. 12, p. 113. , 1905.—Lycoperdaceae of Australia, New Zealand... . ., 44 pp. Cincinnati. , 1906, a—The genus Bovistella. Mycological Notes, No. 23, p. 277. , 1906, b.—Mycological Notes, No. 30, p. 392. Morean, A. P., 1892.—North American Fungi. V. Journal of the Cincinnati Society of Natural History, vol. 14, pp. 141-148. PERSOON, C. H., 1797.—Tentamen disp. meth. fungorum . . . ., 76 pp. Leipzig. == , 1801.—Synopsis methodica fungorum . . ., 706 pp. Gottingen. Saccarbo, P. A., 1888.—Sylloge fungorum . . . ., vol. 7, 882 pp. Patavia. EXPLANATION OF PLATE XXXVII. 1. Capillitium of Bovista purpurea Lloyd. x 110. This shows the main stem and dichotomous, tapering branches. 2. Bovista purpurea Lloyd. x 3. The specimen in the centre shows the partly adhering exoperidium. 3. Bovistella verrucosa G. H. Cunn. Natural size. Note the roughened surface of the endoperidium. 8. cernibl 9. BY G. H. CUNNINGHAM. 373 Bovistella bovistoides (Cke. et Mass.) Lloyd. x %. These specimens (from Tapanui) show unusual development of areolate markings. Capillitium of Bovista purpurea. x 55. Capillitium of Bovistella bovistoides. x 55. Capillitium of Bovistella verrucosa. x 55. The opaque areas in the lumen are air bubbles. Spores of Bovistella bovistoides. x 540. The hyaline pedicels are scarcely dis- e. Note the thick-walled capillitium. Spores of Bovista purpurea. x 540. The photographs of the mature plants were taken by Mr. H. Drake, of this laboratory. The photomicrographs were taken by the writer, but the plates were developed and printed by Mr. Drake, to whom the writer’s thanks are due. A NEW FOSSIL INSECT WING FROM TRIASSIC BEDS NEAR DEEWHRY, N.S.W. By R. J. Tittyarp, M.A., Sc.D. (Cantab.), D.Se. (Sydney), F.R.S., F.N.Z. Inst., F.L.S., F.G.S., F.E.S., C.M.Z.S., Entomologist and Chief of the Biological Department, Cawthron Institute, Nelson, N.Z. (Plate xxxvi and one Text-figure.) [Read 30th September, 1925.] Through the kindness of Professor Sir T. W. EH. David, F.R.S., Emeritus Professor of Geology in the University of Sydney, I have been enabled to study a remarkable new insect wing which has recently been discovered in a magnificent state of preservation in a quarry at Beacon Hill, near Deewhy (not far from Manly), N.S.W. The bed from which this wing comes is a shale bed, about the middle of the Hawkesbury Sandstone (Middle Triassic), i.e., about 500 feet above the base of this sandstone; thus it will be seen that the wing is older than those found at Ipswich, Queensland, and at St. Peter’s, Sydney. Professor David writes as follows concerning the discovery of this fossil: “There was more of the fossil when the quarrymen broke open the shale, but unfortunately Mr. Scully (a most careful and enthusiastic collector, one of the quarrymen) accidentally broke it in trying to get it out, and the body flew into a thousand fragments. We must be thankful, however, for the beautiful wing. But for Mr. Scully it would have been pressed into bricks, probably in a few minutes’ time.” Before describing this fossil, which consists, unfortunately, only of about the distal two-thirds of an enormous wing, I would like to say what a great loss has been sustained just by the failure to retrieve the rest of the specimen. This fossil belongs to the rarest and least known of all orders of fossil insects, and it is so wonderfully preserved that its head and body, particularly if, as I suspect, the mouth-parts were displayed in situ, would have afforded one of the most wonderful discoveries ever made in fossil Insecta. It is greatly to be hoped that some way may be found to get all the quarrymen in the Triassic shale-beds around Sydney interested in these wonderful fossils, a large number of which have probably gone already into the making of bricks for Sydney suburban houses. In 1916, when I first started studying fossil insects, I described as Mesotitan giganteus Till.* a very large but badly preserved fossil insect from the Wianamatta Shale (Upper Triassic) of St. Peter’s, Sydney, placing it in the Order Protorthop- tera. The condition of this fossil was not such as to allow of any accurate placing or diagnosis of it, but considerable portions of three wings were preserved, showing the main courses of the basal parts of the veins, and also the remarkable * “Mesozoic and Tertiary Insects of Queensland and N.S.W.” Queensland Geol. Survey, Publ. No. 253, 1916, p. 40, Pl. 7, fig. 2. BY R. J. TILLYARD. 375 archedictyon or meshwork of fine veinlets and crossveins between them. A con- siderable part of the body was indistinctly visible; but, most unfortunately, a large triangular crack in the rock had removed the head entirely. Having given this fossil some further study from time to time, I have become more and more inclined to the view that it is not a member of the Order Protorthoptera at all, but belongs to the very rare and problematical Order Protohemiptera, founded by Handlirsch for the wonderful Lower Permian fossil Eugereon boeckingi Dohrn from Germany. This insect was of large size (Handlirsch’s estimate of its expanse of wing is 160 mm., or over six inches). Its most remarkable charac- teristics are the broadly flanged pronotum, resembling that of certain Car- boniferous forms, and the very small head provided with mouth-parts forming a very long and slender sucking-beak, which projects straight out in front of the head. Probably no type of mouth-parts has given rise to more discussion than this, and its relationships, if any, to the mouth-parts of recent Hemiptera are still not satisfactorily determined. Hence the discovery of the head and mouth-parts of a related fossil form, in the wonderful state of preservation of this new fossil, would be a find of the utmost importance, and it is to be hoped that it will one day be, made. The head and mouth-parts of the new fossil not being available, one has to fall back on what is left of the wing for evidence of its relationships. The parts preserved are just those, viz. the distal two-thirds of the wing, which are absent from Hugereon and also to a great extent from Mesotitan giganteus. Nevertheless the general scheme of venation agrees well enough in all three cases, and the strongly-marked character of the archedictyon, in conjunction with the immense size (no other fossils approaching these except only the gigantic Meganeuridae, which are of a very different type), make it certain to my mind that the new fossil must be placed in the genus Mesotitan, and that that genus must now find its proper place within the Order Protohemiptera. Order Protohemiptera. Restricting our study of this Order to venation, we may state that the following combination of characters will define it:— Apart from the anals, there are only two convex veins in the wing, viz. R, and Cu,, and both of these are unbranched; no anterior median (MA) or interpolated convex sectors are present. (This character enables us at once to distinguish this Order from Protodonata). Cuz, which in nearly all Orders is a simple vein, is strongly developed, with many branches; this isa striking and most unusual character. A complete set of costal veinlets is present, arising from the well- developed Sc. Throughout the rest of the wing, between all the main veins and their branches, the membrane of the wing is covered with a very charac- teristic archedictyon, consisting for the most part of very closely set, transverse, delicate crossveins, the distance between two such consecutive crossveins being very small compared with their length; in the wider spaces, and towards the margin of the wing, these tend to become interrupted by a weak and irregular line dividing them into upper and lower sets, or even to become split up into very irregular polygonal areas. This archedictyon is so characteristic that I am not able to name a single wing outside of this Order which possesses it, and I think a careful study of Plate xxxvi will convince anybody that i#t must be regarded as one of the diagnostic characters of the Order. F 376 A NEW FOSSIL INSECT WING. Using venational characters only, the two families represented in this Order may be defined as follows:— Costal veinlets excessively numerous and close, either transverse or slanting towards the base; Cu, only reaching to about half-way along SIAL Y= See caer Seen Gea Okorcepione, chalga.) a\c.4\o cla ane Fam. Eugereonidae Handl. Costal veinlets moderately numerous, not so close together, slanting very obliquely towards apex; Cu. ending up far beyond half-way .............. renee Ae. Sheets bf atheed Bate oe ices Rae Fam. Mesotitanidae, fam. nov. The Eugereonidae are only known from the Lower Permian of Germany, the Mesotitanidae from the Triassic of Australia. Family Mesotitanidae, nov. This family is here proposed for the reception of Mesotitan giganteus Till. from the Upper Triassic of St. Peter’s, Sydney, and the new species described below. Its characters have already been defined in the Key given above. Genus MesoTiran Till. Characters as for the family; in addition, Rs pectinate, four-branched, M two- branched, Cu, with pectinate series of descending branches. Genotype, Mesotitan giganteus Till. Text-fig. 1.—WMesotitan scullyi, n. sp., preserved portion of forewing from Beacon Hill, near Deewhy, N.S.W. Order Protohemiptera, fam. Mesotitanidae. Length 86.5 mm. Archedictyon omitted. MESOTITAN SCULLYI, n. sp. Plate xxxvi; text-fig. 1. Total length of wing preserved, 86.5 mm., indicating a full wing-length of at least 130 mm. and an expanse of wing of 275 mm. or about 11 inches. Greatest width (at break), 34.5 mm. Apex of wing strongly pointed, slightly nodding. The preserved part of the wing consists of about the distal two-thirds, and does not include the origin of Rs or the forkings of M and Cu, nor is any of the anal area present. Sc markedly concave, very strongly developed, with slender but clearly marked costal veinlets arising along its whole length; end of Sc about 12 mm. before apex. R, markedly convex, but not quite as stout as Sc, running below and subparallel to it and ending just beyond it above apex. Rs a slender, concave vein, unbranched BY R. J. TILLYARD. 377 until well past half-way along wing, when it gives off an obliquely descending pectinate series of four veins, the first being R,.,;, the second R,, and the third and fourth two short terminal branches from R., the last of which ends at the actual apex of the wing. M with two branches only, M,.., and M,.,,, running subparallel to one another and also to Rs and R,.,,; above and Cu, below. Both these branches being concave, there is clearly no MA present. Cu, not very strongly developed, but markedly convex, running subparallel to and a little above the well-developed, concave Cu., from which a descending pectinate series of five branches is visible. Archedictyon as shown in Plate xxxvi. Pigmentation evidently present, the wing being mottled in alternating areas of light and dark pigment, these areas being mostly squarish and tending to form irregular transverse fasciae; eight or nine such dark brown fasciae can be made out crossing the lighter parts of the wing from the costal margin on the part preserved. The wing is almost certainly a forewing. Type, Holotype, Specimen No. 20270, labelled ‘“‘Beacon Hill, Deewhy,”’ on under- side of rock; in Coll. Mining Museum, Sydney, N.S.W. The species is dedicated to its discoverer, Mr. Scully, to whom the thanks of all palaeontologists and entomologists are due for preserving it from destruction. The photograph in Plate xxxvi was taken by Mr. W. C. Davies, Curator of the Cawthron Institute, to whom my thanks are due for this excellent result. In order to see the convexity and concavity of the veins correctly in this photo- graph, it should be looked at either from the broken basal end, or else with the posterior margin uppermost. EXPLANATION OF PLATE XXXVI. Mesotitan scullyi, n. sp., preserved portion of forewing. Order Protohemiptera, ‘fam. Mesotitanidae. Length 86.5 mm. [W. C. Davies photo. Postscript, added 2nd September, 1925.—Since the above paper was written, an artist friend of Mr. Scully has made a sketch of the insect, from Mr. Scully’s recollections of it as it appeared when first discovered complete. This sketch shows some interesting points. Chief amongst these is the presence, on the some- what large head, of a strongly projecting beak, such as is characteristic of the Order Protohemiptera. The bases of the forewings are greatly expanded, as in Eugereon, leaving no room for large hindwings. Towards the end of the abdomen are a pair of leaf-like processes, which I think are really the partially expanded apical portions of the partly folded hindwings, the rest of which probably lie close to or under the abdomen and would not therefore be noticed by Mr. Scully unless specially looked for. NOTES ON AUSTRALIAN COCCIDAE WITH DESCRIPTIONS OF NEW SPECIES. By WALTER W. FrRocGatt, F.L.S. [Read 28th October, 1925.] During the time I was engaged in writing the Descriptive Catalogue of the Australian Coccidae, published as Science Bulletins by the Department of Agricul- ture of New South Wales, I received and collected a large number of undetermined species of Coccids from all parts of Australia. Among them I have some remark- able forms that I propose to work out and publish under the heading of Notes on Australian Coccidae. In 1898 the late W. M. Maskell described a new species of Lecanium from specimens discovered in a cavity in the stem of Casuarina in Victoria under the name of Lecanium casuarinae. In my investigation into the habits of Coccids, I have on several occasions found Lecaniums or the remains of Lecaniums in the cavities in branches or tree-trunks of our forest trees, usually in those occupied by ants. As the inner surfaces of these cavities are hard, dry and sapless, the larvae can apparently develop into adult females with very little moisture or apparent food. It is difficult to give adequate descriptions of these curious forms, because usually the adult females have become rounded, dry structures with all the appendages aborted and hidden in the brittle sack or bubble, which represents the final development of the adult coccid. Still there are characteristic differences in the three species dealt with that are sufficient to separate them. In two the larvae were found sheltering under the adult coccid shell. LECANIUM EUCALYPTI, 0. Sp. This coccid in all stages of development was found firmly attached to the hard corrugated surface of a cavity in the stem of a small eucalypt growing in the forest at Brooklana, Dorrigo, 10.2.1925. The surface of the cavity was coated with fine brown kino, and covered with a resting colony of small black ants (Crematogaster sp.). Adult 2 dark reddish-brown, almost black turning to rich reddish-brown when treated with potash, and showing a fine granulated structure. Form variable, but usually sub-globular, with the apical margin having the anal cleft showing in an oval process in a depression raised well up from the outer margin of the body, which fits closely against the wood. The whole might be likened to a round brown bubble contracted round the edges, about one-quarter of an inch in diameter and one-twelfth of an inch in height. BY W. W. FROGGATT. 379 Immature 2? oval-or shield-shaped, flattened down on the wood; reddish-brown, with a little floury secretion on both the upper and under surface. Larva shield-shaped, dorsal surface slightly convex and segmented, with the anal segment rounded on either side of a deep anal cleft; outer margins finely striated, with slight lateral clefts. The antennae composed of nine irregularly shaped segments, the basal one forming a wedge-like process on either side of the rostrum, the second turning upward, third longer, 4th-5th shorter, 6th-7th swollen and rounded, 9th produced into a lance-shaped tip fringed with fine hairs. Legs well developed, with the femur thickened and the turn pointed. General colour light reddish-brown with black eyes. LECANIUM DIXONI, n: sp. I am indebted to Mr. C. French, Jr., for this interesting species. He received several specimens from Mr. J. E. Dixon, who found them in July, 1923, attached to the inner surface of a cavity in the stem of a Loranthus, parasitic upon a gum tree (Hucalyptus sp.) growing at Eltham, Victoria. Adult 2 about a quarter of an inch in diameter and of somewhat similar bubble-like form to that of the previous species. The derm lighter reddish-brown, with a narrow dark brown dorsal stripe from the front margin to the base of the anal cleft. The anal portion of the derm depressed above the hind margin, covered with fine irregular fovea. The anal cleft ridged on either side, forming a raised ridge, which in conjunction with a short ridge above, and short crossbar across the centre forms a cross-like pattern above the anal cleft. Larva yellowish-brown, changing into rich pink when placed in potash. Dorsal surface shield-shaped, the central portion enclosed in a circular ring with a parallel ridge to the base of the anal cleft. The outer margin is flattened, forming a flange. The lateral marginal clefts well defined, anal cleft small. Legs long, femora thickened, tibiae long, fringed with fine hairs, tarsal claw small, and turned inward. Antennae long cylindrical, composed of six irregular joints, 1st short broad, 2nd very long, 3rd oval, about a third of the length of the second; 4th of a similar irregular oval shape, smaller and shorter, 5th not quite half the length of the 4th, 6th very small pointed and fringed with fine hairs. LECANIUM GALLITRIS, 0. Sp. This species occurs in cavities under the bark of the Desert Cypress (Callitris calcarata) usually where wood ants have formed a nest. The species of ant, in whose nest this type was found is Podomyrma bimaculata. I have on several occasions found the mature dead female coccids in such cavities as noted on page 30 of my “Descriptive Catalogue of the Scale Insects (Coccidae) of Australia.” During a recent visit to Forbes when examining a pile of cypress logs in a sawmill yard I found a cavity in one, containing a small colony of these wood ants, tenanted by several specimens of this Lecanium attached to the surface of the wood hidden under the dead bark. Adult dark brown to almost black, with the margins of the derm crenulated and closely attached to the surface of the wood, the dorsal surface shagreened; the front margin arcuate giving it a somewhat heart-shaped form; the anal cleft very small. General form irregular, often varied from the shape of the cavity in which they have developed, broadly oval, and could be easily mistaken for a bubble-shaped mass of brown resin. Brittle and thin, forming an oval cavity beneath, all appendages aborted. 380 NOTES ON AUSTRALIAN COCCIDAE. One specimen (not so mature) one-third of an inch in diameter, broader and more flattened than dead females. Front margin arcuate, outer edge finely crenulated, anal cleft small, lateral clefts distinct. Outer margin, central portion of the dorsal surface bright reddish-brown. Under surface convex, mottled with dull yellow, clouded with floury secretion. Antennae and legs aborted. THE LABIAL PALPI OF TRICHOPHYSETIS CRETACEA AND ARGYRIA AMOENALIS. By ALFrrep Puitpotr, Hon. Research Student in Lepidoptera, Cawthren Institute, Nelson, N.Z. (Communicated by Dr. A. J. Turner.) (Two Text-figures. ) [Read 25th November, 1925.] While engaged in a study of the maxillae in the Lepidoptera, the writer came across two instances of very unusual structure of the labial palpi. In doth cases the species belonged to the Pyralidae, one, Argyria amoenalis Butl., being placed in the subfamily Crambinae and the other, Trichophysetis cretacea Butl., in the Pyralinae. I cannot learn that these interesting instances have already been referred to; it will therefore be useful to have them described and figured. The labial palpi of the Lepidoptera are, on the whole, very uniform in struc- ture. Such differences as there are arise, firstly, from their covering of hairs or scales, the extent and direction of which often result in greatly modifying the apparent shape of the segments; secondly, from the direction of the segments themselves, as drooping, porrect or curved upwards; and, thirdly, as the result of atrophy, the three original segments being sometimes reduced to two, or to one only; in a few instances the palpi have entirely disappeared. But in the vast majority of Lepidoptera the labial palpi consist of three more or less cylindrical segments, the first, or basal, being almost always a short one. A detailed account of the two interesting instances referred to above will now be undertaken. Trichophysetis cretacea Butl. In this species the basal segment of the palp in the male is of normal structure; it is cylindrical and slightly curved. The third, or terminal, also shows nothing unusual. It is long, cylindrical and blunt-pointed, and exhibits a promin- ent Johnston’s organ at the apex. The second segment is greatly modified. It articulates in the usual manner with the first segment, but almost immediately gives rise to the third segment and extends inwards as a slender tentacle. This tentacle is a little less than half as long as the terminal segment and has its apical half slightly swollen and clothed with minute sensory hairs. I am not aware of any other instance of a process being found on the second segment of the labial palp, though an apical process of the first segment occurs in some species of Porina, and in Sabatinca. The process in Trichophysetis is, however, perhaps more correctly to be regarded as the actual segment rather than an outgrowth, the articulation of the terminal segment having worked down from the apex to near the base of the second segment, and the portion of that segment left projecting having become modified into a tentacle-like structure. The female of the species has quite ordinary palpi. 382 LABIAL PALPI OF TRICHOPHYSETIS AND ARGYRIA. The labial palpi of this species are, I believe, the most strangely modified in the whole of the Lepidoptera. The basal segment is stout, being much swollen ventrally. The second segment is also thick, but is broadened apically. On the upper surface of the broadened portion the organ is widely and deeply excavated. On the apex of the distal area of this excavation, and bending over it, is a dense patch of short, thick, slightly curved hairs. On the proximal portion is a similar, but more extensive, patch, directed towards the distal area. The third segment is reduced to a small subtriangular piece, articulating with the apex of the proximal side of the excavation. The whole of the second and third segments is covered with very long fine hair. Text-fig. 1.—Argyria amoenalis Snel. Labial palp. a, male; b, female. Text-fig. 2.—Trichophysetis cretacea Butl. Labial palp. a, male; b, female. Argyria amoenalis Snel. This curious apparatus is strongly suggestive of a strigil. The opposing tufts of hair would form the brushing apparatus, while the modified third segment would act as a spring to keep the organ undergoing the cleaning process pressed against these brushes. But the usual tibial strigils are present, so that it is hardly likely to be for the benefit of the antennae that the labial strigils have been developed. Is it possible that it may be the cleaning of the haustellum for which the structures are designed? But, if so, how is it that the female of amoenalis possesses quite ordinary palpi? Also, it seems probable that the apparatus is confined to this one species of the genus. A. plumbolinealis Hmpsn., A. pentadactyla Z. and A. strophaea Meyr. have been examined and found to possess normal labial palpi. A close study of the habits of both sexes of Tricho- physetis cretacea and Argyria amoenalis might explain much in relation to the very interesting modification of their labial palps. Perhaps some Australian lepidopterist will, sooner or later, be able to throw light on the matter. For the material on which this paper is based I have to offer my best thanks to Dr. A. Jefferis Turner who very kindly came to my assistance with various specimens. CONTRIBUTIONS TO OUR KNOWLEDGE OF THE FLORA OF NEW SOUTH WALHBS. By W. F.. BLAKELY, National Herbarium, Botanic Gardens, Sydney. [Read 28th October, 1925.] Proteaceae. GREVILLEA SHIRESSII, N. Sp. Frutex gracilis, 2-4 m. altus, ramulis novellis hirsutis; ramis angulatis; foliis anguste-lanceolatis, acuminatis, plus minusve trinervis, 8-16 cm. longis, 2-4 cm. latis; racemis axillaribus, floribus 5-8 pallide-violaceis; pedicellis gracilibus 12-15 mm. longis; perianthio lato plus minusve gibberoso, lobis linearibus apicibus connatis; stylo perianthium excedente, 10-15 mm. longo; ovario glabro, stipitato; fructo oblique-cylindrico, costato, vel sulcato, cum stylo pedicelloque 5 cm. longo, 7-9 mm. lato; seminibus lineari-ovatis marginibus revolutis 10-12 longis, 3 mm. latis. Slender, graceful shrubs, 6-25 feet high, glabrous, except the minute shoots which are infested with pale, rufous, silky hairs; branchlets angular or compressed. Bark slightly rough, reddish-brown throughout. Leaves narrow, lanceolate, terminating in a rather long curved point, somewhat triplinerved, venulose on both surfaces, undulate, shining above, dull and much paler underneath, the midrib prominent on both sides, the intramarginal vein distant from the edge, giving the leaf a triplinerved appearance, lateral veins distinct, forming an angle of about 45° with the midrib; the nerve-like margins diminishing into the moderately short, compressed petiole, 8-16 cm. long, 2-4 cm. broad. Infloresc- ence axillary, forming short racemes of 5-8 flowers, glabrous, except the common peduncle, which is infested with minute, deciduous, silky hairs. Flowers pale violet to greenish, tinged with pale purple-brown; pedicels slender, terminating in a linear, lanceolate, dilated green, persistent bract-like expansion which extends to the fissure of the corolla, 12-15 mm. long, the dilated portion 5-8 mm. long. Buds obliquely clavate, inflated in the lower half. Corolla broad, somewhat gibbose, the segments linear, their apices connate, orbicular, the broad basal portion minutely glandular, hispid inside. Style exceeding the corolla, 10-15 mm. long, pale green, somewhat obscurely striate, the striae extending into the stipes. Stigmatic disc oval, apiculate, 2.5 mm. long, 1.5 mm. broad. Ovary glabrous, stipitate, the stipes purple-brown, and nearly as long as the style. Fruit obliquely cylindrical, pale brown, with 8-10 faint longitudinal lines or ridges, including the style and stipes, 5 cm. long, 7-9 mm. broad in the middle. Seeds narrow-ovate, rather acute at both ends, the margins revolute, forming ventral surface, the back convex, marked with three or more faint lines, 10-12 mm. long, about 3 mm. broad. 384 CONTRIBUTIONS TO OUR KNOWLEDGE OF TIIJE FLORA OF N.S.W., Description of Seedlings.—Hypocotyl slender, slightly compressed, 3-5 cm. long. Cotyledons obovate or oblong, hastate, thick and veinless, dark green, 10-13 mm. long, 4-5 mm. broad. First pair of leaves sessile, narrow lanceolate, with a slightly curved mucro, attenuated at the base, the lamina reaching to the stem, obscurely penninerved, light green. Stem slightly angular or compressed; the growing point tipped with rufous, silky hairs. Out of three seedlings examined, the first leaves were alternate on two, and opposite on the other. Range.—So far the species has only been found at Mullet Creek, Hawkesbury River, about 1 mile northwest of Wondabyne railway platform, 40 miles by rail north of Sydney. It extends from the head of the salt water of Mullet Creek, as far as the northern arm, a distance of about a mile and a quarter. (D. W. C. Shiress and W. F. Blakely, 24/12/1922.) Named in honour of David William Campbell Shiress, my friend and companion on many botanical excursions during the last fourteen years. G. Shiressii is an attractive looking shrub, with light green, undulating leaves, and it seems to belong to section Lissostylis, but it does not appear to have any very close affinities amongst the Hastern species, and for the present it is placed next to G. oleoides Sieb., because of its long leaves. Tremandraceae. TETRATHECA SHIRESSII, nN. Sp. Caulis debilis, teres, glaber, vage ascendens ad 3-9 dm. ramis paucis; folia rara, sessilia vel brevi-petiolata, lineari-lanceolata, quasi minutis glandulis denticulata, inferiora rara alternata, 10-25 mm. longa, 3-4 mm. lata, superiora 3-4 in verticillo; flores axillares et terminales, pedicillis curvatis, 10-20 mm. longis, sepalis 4, lanceolatis, 4 X 2 mm.; petalis 4, erubescentibus angusto-lanceolatis, 10 xX 4.5 mm.; staminibus 8, antheris quadrangularibus, ovario glanduloso- pubescente, stylo simplice, terete, capsula cuneata, minute glanduloso-pubescente, 6-8 X 5 mm. A weak scrambling undershrub, 1-4 feet high, or frequentiy prostrate, glabrous except for a minute glandular vestiture on the nascent parts; internodes 3-10 cm. long; leaves few, very variable, rather distant, sessile or shortly petiolate; the lower ones opposite, and usually in whorls of three, linear lanceolate to elliptical- lanceolate, paler underneath, the margins recurved, and minutely glandular- denticulate, 10-20 mm. long, 3-4 mm. broad; upper leaves usually in pairs, linear, acute, straight or curved, 10-25 mm. long, 1-4 mm. broad; flowers axillary and terminal, solitary or in pairs, on slender, curved, slightly dilated pedicels, 10-20 mm. long; sepals 4, dark purple-brown, lanceolate, one-nerved, 4-5 mm. long, 2 mm. broad; petals 4, light pink, narrow-lanceolate, 10-15 mm. long, 4-5 mm. broad, or one or two slightly broader than the others; stamens 8; filaments about 1 mm. long; anthers somewhat quadrangular, linear-oblong, minutely hispid along the angles, dark coloured or nearly black, except the yellow top; ovary glandular-pubescent, shorter than the anthers; style simple, terete, nearly 3 mm. long, the lower half pink, the upper half white; stigma very small, scarcely perceptible; capsule cuneate- ovate, minutely glandular-pubescent, and somewhat viscid, slightly venulose, 6-8 mm. long, 5 mm. broad; seeds pubescent, almost triangular, 3 mm. long, with a white caruncle about 1 mm. long. Range.—Heathcote, J. Foster; Kariong Trig., three miles northwest of Wonda- byne, Hawkesbury River district; very common over a wide area. It grows in all kinds of situations, sometimes from the crevices of bare rocks, in hard, dry BY W. F. BLAKELY. 385 sandstone rubble, in swampy and semi-swampy land, and in thickets of Angophora cordifolia. In the latter situation it scrambles up through the Angophora to a height of five feet. D. W. C. Shiress and W. F. Blakely, the Type; on the east side of Sugarloaf Trig., and at the head of Patonga Creek; Gosford district, J. Foster. Among the New South Wales species, 7. juncea Sm., appears to be its nearest affinity, from which it differs greatly in size and habit, in the long terete branches, more variable leaves, especially the upper ones and larger flowers with their longer, filiform pedicels. Araliaceae. ASTROTRICHA CRASSIFOLIA, 0. SD. Fruticulus virgatus 6-12 dm. altus, ramis juvenilibus floribusque tomento brevo incano vel ferrugineo vestitis, ramis glabrescentibus; foliis alternis, breviter petiolatis, erectis, rigidis, lineari-oblongis valde obtusis, crassis, canaliculatis, supra laevibus nitidisque, subtus tomentosis marginibus valde revolutis, 2-5 cm. longis, 3 mm. latis; paniculis angustis, terminalibus vel in axilibus superioribus; floribus breviter pedicellatis, 4-14 in umbellis; alabastris clavatis, pedicellibus, erassis, quam flores brevioribus; calycis lobis prominentibus, petalis reflexis, mucronatis, extra tomentosis, intra glabris; antheris oblongis, versatilibus, quam filamentos brevioribus; fructis maturis ovati-vel lati-oblongis, glabris praeter annulum hirsutum densum ad apicem. A small virgate shrub 2-4 feet high, the young branches and inflorescence covered with a close, hoary, ferruginous, floccose tomentum, the old branches quite glabrous. Leaves alternate, petiolate, erect and rigid, linear-oblong, very obtuse, thick, canaliculate, smooth and shining above, flocculent beneath, the midrib obscure, the margins strongly revolute, 2-5 cm. long, 3 mm. broad; petioles compressed, 3 mm. long. Inflorescence terminal and in the upper axils, the whole rarely exceeding 6 cm. long; umbels of 3-14 shortly pedicellate flowers, buds shortly clavate, and like the thick pedicels, densely floccose. Calyx teeth prominent. Petals 5, reflexed, densely floccose, glabrous within, one-nerved, mucronulate; stamens 5; filaments filiform, pilose; anthers oblong, versatile, shorter than the filaments; styles slightly protruding, rather thick, with more or less incurved stigmas. Ripe fruit compressed, ovate to broadly oblong, somewhat thick and venulose, glabrous, except for a dense floccose ring at the top. Range.—Only known at present between Warrah Trig. and Woy Woy, on the northern portion of Broken Bay. D. W. C. Shiress and W. F. Blakely. Affinities.—A. crassifolia is closely allied to A. ledifolia DC., and the characters which separate it from that species are the thick, rigid, linear, strongly revolute, very obtuse leaves, shorter inflorescence, longer points to the calyx, and differently shaped fruits. From A. linearis A. Cunn., it is distinguished by its broader and shorter, smooth, obtuse leaves, longer calyx points, and longer and more oblong fruit. Compositae. OLEARIA STILWELLAE, Nn. Sp. Fruticulus 3-6 dm. altus, radicibus carnosis crassisque; caulibus erectis, pedunculis et caulibus juvenilibus capillorum bifurcatorum vel simplicium tomento incano vel rufo vestitis; foliis alternis, petiolatis, angusti-vel lati-oblongis, nonnumquam obtusi-lanceolatis, minute denticulatis, supra fusci-viridibus 386 CONTRIBUTIONS TO OUR KNOWLEDGE OF THE FLORA OF N.S.W. glabrisque, subtus tomentosis, 4-11 cm. longis, 2-3.5 cm. latis; capitulis mediocriter magnis, solitariis, axillaribus, pedunculis 10-19 cm. longis; involucris turbinatis 2 cm. longis; bractis imbricatis acuminatis extra semiglabris; bractis interioribus scariosis discum excendentibus; radiis ligulis albis vel coerulis 15-18 mm. longis; floribus disci numerosis; corolla glabra 10 mm. longa; acheniis linearibus sericeo- tomentosis, 5 mm. longis; setis pappi numerosis, barbellatis, inaequalibus. A dwarf herbaceous shrub, with thick, fleshy roots and erect stems 1-2 feet high, the young stems and pedicels, and under side of the leaves clothed with a dense, hoary or reddish, silky tomentum, consisting of simple and bifurcate, silky hairs. Leaves alternate, petiolate, narrow to broad oblong, or obtusely lanceolate, minutely denticulate, the denticulations erect and asperate, dark green and glabrous above, silky-tomentose beneath, 4-11 cm. long, 2-3.5 cm. broad; venation distinct, channelled above, prominent beneath; petiole terete or nearly so, canalicu- late, 5-20 mm. long. Flower heads moderately large, on axillary peduncles 10-19 cm. long, with two or three small floral leaves along them. Involucres turbinate, 2 cm. broad at the top; bracts imbricate, acuminate, semiglabrous on the back, the inner ones scarious, exceeding the disc; the outer ones thick and of the same colour and vestiture as the peduncles. Ray florets 10-12, the ligular pale blue or white, 15-18 mm. long; disc florets more numerous, the corolla glabrous, 10 mm. long; anthers of the disc florets acute at the base, terminating in long, protruding points; style lobes long, minutely glandular, compressed-acuminate, flexuose. Achenes linear, silky-hairy and slightly striate, 5 mm. long. Pappus bristles barbellate, unequal, the inner ones exceeding the outer ones. Named in honour of Miss Sylvia Stilwell, who with Mr. D. W. C. Shiress discovered this beautiful species at Nana Glen, 35 miles south of South Grafton. O. Stilwellae belongs to section Dicerotriche, and its thick, fleshy roots seem to distinguish it from all the other species. It is closely allied to O. pannosa Hook., from which it differs in the differently shaped, denticulate leaves, much smaller turbinate involucres, pointed anthers, more highly coloured tomentum, and in the fleshy roots. From 0. grandiflora Hook., it is distinguished by its narrower and minutely denticulate leaves, different venation, smaller involucres, and acuminate bracts. TWO NEW SPECIES OF SILKY LACEWINGS (FAMILY PSYCHOPSIDAE, ORDER NEUROPTERA PLANIPENNIA) FROM AUSTRALIA. By R. J. Tirtyarp, M.A., Sc.D. (Cantab.), D.Sc. (Sydney), F.R.S., F.N.Z. Inst., F.L.S., F.G.S., F.E.S., C.M.Z.S., Entomologist and Chief of the Biological Department, Cawthron Institute, Nelson, N.Z. (Plate xxxviii.) [Read 28th October, 1925.] The rare and beautiful insects belonging to the family Psychopsidae, or Silky Lacewings, have their headquarters in Australia, where ten species are already known; in addition to these, there are only three or four other known species, found in South Africa, Burma and China. The family is a very ancient one and originally spread all over the world; fossil remains of its ancestors have been found in the Upper Triassic of Ipswich and in the Liassic and Upper Jurassie beds of Europe. Two fine new species are here described, bringing the Australian total up to twelve. One of these comes from Central Queensland, and was discovered by Mr. W. B. Barnard, of Toowoomba, to whom it is dedicated. His specimens were given to Mr. R. Illidge, of Brisbane, who very kindly presented them to me; one of them is being presented by me to the Queensland Museum. The other is a unique specimen from Roebourne, W.A., sent for description by Mr. Glauert, Biologist of the Western Australian Museum, Perth. The photographs from which Plate xxxviii has been prepared were taken by Mr. W. C. Davies, Curator of the Cawthron Institute, to whom my best thanks are due. These insects are evidently attracted to light on warm, still nights, and I think a considerable number of new species might yet be found in out-of-the-way parts of Australia, if they were looked for by this means between October and March. As each species is only out on the wing for a short time, collectors should note the date of capture; the Roebourne species has only the year (1920) on the label, and this will make it less easy to obtain more of it. Undoubtedly the easiest way to obtain specimens of this family is to obtain the larvae from under the bark of Eucalypts. Only rough-barked trees should be selected, with fresh, healthy bark on them, and the best places to search are round the edges of a gum-flow or near large cracks in the bark. The bark should be levered up with a chisel or wooden wedge, and the characteristic grey pubescent larvae, with their large calliper-like jaws, will be found either on the raised bark or on the wood beneath it. As these larvae are cannibals, only one should be kept in each tube; the tubes must be kept dry and well aérated, the open ends being stopped with cotton-wool. 388 TWO NEW SPECIES OF SILKY LACEWINGS, Family Psychopsidae. Genus PsycHorsis Newman. PsyYCHOPSIS BARNARDI, nD. sp. Plate xxxviii, fig. 1. $. Total length 10 mm.; forewing 19 mm. Head small, 1.6 mm. wide; eyes dark greyish-brown, mottled; antennae pale testaceous basally, shading to dark brown distally (apices missing). Epicranium pale testaceous, varied with livid greyish; vertex and frons dark brown, except for pale testaceous areas surrounding the antennae, the brown thus forming a kind of X-mark with the four arms reaching to the eyes; surmounting this mark, on the epicranium, is a smaller, diamond-shaped, median area of brown, not quite as dark. Clypeus, labrum, labium and palpi all pale testaceous. Thorax: Prothorax pale testaceous, pronotum covered with dense hairs of same colour. Mesothorax shading from medium testaceous anteriorly to dull brownish posteriorly. Metathorax brownish with livid greyish markings on pronotum. Legs short, pale testaceous, apices of tibiae and tarsi dark brown. Wings: Forewing 11 mm. wide at tornus, hindwing 9 mm. wide at same place. Rs with 13-15 descending branches before end of vena triplica in forewing, 11 such in hindwing. M,., in forewing fusing with Cu, about half-way. Posterior margin of both wings concave distally, that of hindwing only slightly so; tornus of forewing very prominent. Both wings with a complete series of costal cross- bars and three complete series of gradate crossveins; forewing also with from four to seven crossveins forming the upper half of an additional gradate series on disc, between basal and middle series. Forewing subhyaline, the veins pale testaceous, the membrane slightly tinted. On the vena triplica, at one-fourth from base, is a rounded, reddish-brown spot 1.2 mm. in greatest diameter, with darker border; from it two blackish lines diverge posteriorly to M, where they join with the irregular brownish markings along the posterior margin. In the apical half of the wing, three fasciae are present, each irregular in form, narrow, and bordered on either side by a fine blackish line; the most basal of these fasciae is not darkened between the lines, but the other two are shaded brownish; the three fasciae converge together strongly towards the tornus, but the most distal one, lying close to the termen, ends up before reaching the other two; these latter, however, join together to form a brown blotch which runs into an irregular patch of bright reddish-brown above the tornus, bordered by three small black dots. About half-way between costa and tornus are two larger black spots, one lying on the distal side of each of the first two fasciae before they merge; there is also a very small black spot in the brown blotch below them. Along Cu and between it and posterior margin are some irregular brownish blotches, from one of which, at about three-fifths from base, an oblique black line runs upwards for nearly 2 mm., suggesting the beginnings of the formation of an extra fascia. Both costa and termen are lightly mottled with markings of pale brownish testaceous. Hindwing with dull creamy veins, and hyaline, iridescent membrane. A large blackish, oval spot, 2 mm. in greatest diameter, lies below end of the vena triplica, and a slight brown cloud around tornus, with indications of one or more dark spots on the tornus itself. Abdomen dull medium brownish above, with indistinctly outlined darker markings mid-dorsally and laterally on segments 2-8. Segments 9-10 pale testaceous, forming a large bulb carrying a pair of large and very hairy surgono- pods, valve-shaped, depressed, with apices convergent and well rounded; basally, touching the lateral raised margin of the bulb on either side, is a small nodular BY R. J. TILLYARD. 389 swelling lying in a slight hollow of the outer face of each surgonopod. Tergite of segment 9 projecting between these two surgonopods as a sharp process nearly as long; viewed laterally, it is bifid, the upper branch being short, sharp and tooth-like, the lower longer, lobed in the vertical plane and well rounded. Habitat.—Central Queensland. Types.—Holotype male in Cawthron Institute Collection and paratype male in Queensland Museum, Brisbane, both taken at Blackwater, Central Queensland railway, on 9th Nov., 1922, by Mr. W. B. Barnard, to whom this species is dedicated. The paratype is larger than the holotype (forewing 22 mm.) but has a small piece missing from above the apex of the left forewing, and does not show the structure of the ninth tergite, owing to the surgonopods being tightly ‘closed. Two other males in Cawthron Institute Collection, one taken by Mr. Barnard at Huntley Station, Central Queensland railway, on 10th Nov., 1921, and one by Mr. E. J. Dumigan at Clermont, Central Queensland, on 15th November, 1924. This species comes nearest to Ps. gracilis Till., from which it can be at once distinguished by its much more robust build, broader and hairier wings, less angulated tornus in both wings, larger black spot on hindwings, and marked differences in the fasciation and other markings of the forewing; in particular, the dark patch in basal part of forewing touches the base in Ps. gracilis, but is one-fourth of the length of the wing from base in Ps. barnardi, n. sp. The new species is also related, but less closely, to Ps. mimica Newm., Ps. margarita Till. and Ps. elegans Guér. PSYCHOPSIS MACULIPENNIS, nh. sp. Plate xxxviii, fig. 2. 6. Total length 10 mm.; forewing 19 mm. Head pale ferruginous; eyes dark grey; antennae pale testaceous basally, shading to dark brown distally. Thorax pale dull testaceous tinged with ferruginous at bases of wings. Legs pale testaceous tinged with ferruginous on femora. Wings dull creamy all over, the costa slightly darkened. Forewing only 10 mm. wide at tornus, hindwing 9 mm. wide at same place. Rs with 15-16 branches before end of vena triplica in forewing, 12-13 such in hind. Posterior margin not concave distally in either wing, tornus not prominent. M;,, in forewing fusing with Cu, only for a short distance, well beyond middle, then separating again. Both wings with a complete series of costal crossbars and three complete series of gradate crossveins; of these the discal one in forewing tends to become doubled, and there are also a few extra crossveins in the disc. Forewing with three dark spots along vena triplica, each being blackish on the membrane but bright ferruginous when crossing a vein; of these, the basal one’ is the largest, reaching down into the disc below Rs; the middle one is smaller, only extending from Sc to Rs; the distal one is very small; there is a suspicion of a fourth spot towards end of vena triplica. Just beyond end of vena triplica is a larger, dark brown, semicircular spot, and below this a small dark blotch on the gradate series. The costa carries four small dark spots, one obliquely above each of the four already mentioned. On the posterior margin are five small but well-defined black spots, three in the basal half and two in the distal; there is also a very small black spot on the termen, a little beyond tornus. Hindwing with an oval blackish spot, 1.6 mm. in greatest diameter, just below end of vena triplica. 390 TWO NEW SPFCIES OF SILKY LACEWINGS. Abdomen dark brownish above, pale testaceous on sides and beneath. Sur- gonopods pale testaceous, hairy, in the form of broad valves; tergite of segment $ projecting slightly between their bases as a short, button-like process. Habitat—Roebourne, W.A. Type.—Holotype male (unique) in Western Australian Museum, Perth, W.A.; label “1920—130/ Roebourne’”’. Cawthron Institute, Nelson, N.Z. 10th Sept., 1925. EXPLANATION OF PLATE XXXVIII. Fig. 1 (below) :—Psychopsis barnardi, n. sp., holotype male. Expanse 38 mm. Fig. 2 (above) :—Psychopsis maculipennis, n. sp., holotype male. Expanse 41 mm. THE ANATOMY OF LINDSAYA LINEARIS AND LINDSAYA MICROPHYLLA. By May M. Wirrirams, M.Sc. (Thirty-four Text-figures. ) [Read 28th October, 1925. ] Introduction. On account of recent anatomical investigation on the rhizome of the genus Lindsaya, many new and interesting facts have been revealed. These facts have led to a great deal of discussion, but at present it is generally accepted that the type of stele characteristic of the genus is one which is intermediate between the protostele and the solenostele. The primitive nature of the stelar structure is of further interest since the plant presents other features which are relatively advanced. This form of anatomy, generally known as the “Lindsaya-type’’, was first observed in Davallia repens, a form closely allied to the Lindsayas, by Trécul in 1885. Unfortunately this investigator overlooked the most interesting features. In 1902, Tansley and Lulham described this type of stele for five species of Lindsaya. It was these investigators who gave it a perfectly correct interpreta- tion. Gwynne-Vaughan, in 1903, described further interesting forms in the sequence of events leading from the Lindsaya-type of stele to the solenostele. The most recent work is that of McLean Thompson, published in 1920, who describes the stele in terms of the ontogeny. The species described in the present communication are two which are of relatively common occurrence in New South Wales, namely, Lindsaya linearis (Swartz.) and Lindsaya microphylla (Swartz.). In some instances they occur in fairly dry sandy soils associated with such types as Schizaea bifida. Under these conditions they show features typically associated with xerophily. In other instances they occur in fairly moist positions, when the leaves show a filmy nature. JL. linearis particularly is often found in very swampy country. The ferns are quite small, with creeping rhizomes which often show dichotomous branching. In L. linearis the fronds are from a few inches to a foot in height, the sporophylls usually taller than the ordinary leaves. The rachis is dark and shining. The pinnules are entire or nearly so, and usually distant. In L. microphylla the fronds are thin, with a slender, usually flexuous, rachis. The barren pinnules are lobed. In both species the sori are continuous. The various parts of the plant were fixed in the field in a 1% chromo-acetic solution and passed in the usual manner into paraffin. Sections were then made of the various parts varying from 3u to 12u in thickness and stained in Flemming’s triple stain, this proving very satisfactory for all parts of the plant. G 392 ANATOMY OF LINDSAYA LINEARIS AND L. MICROPHYLLA, Investigation. A. Rhizomes. The rhizomes of both types are covered by a fairly extensive development of scales which become very profuse around the apex. The rhizomes show a fairly regular dichotomous branching which, although uncommon among the Poly- podiaceae as a whole, has been observed in numerous species of Lindsaya, and seems to be a fairly constant feature of the genus. These branches bear the same structure as the main axis from which they arise. The apices of the stems are alike in both forms and are of the general nature described for Polypodiaceous types. This is indicated in Text-fig. 1, where it will be observed that the apex consists of a small cone, slightly raised above the general level of the rhizome, and terminated by a large initial cell, tetrahedral in shape, which is deeper than broad. Segmentation takes place in the usual manner, segments being cut off parallel to the cutting faces of this cell. These segments are further divided into one inner cell and two outer cells. The latter give rise to the cortex and epidermis, the former to the vascular system of the rhizome. Behind the apex, the cortex becomes sclerized and from the outer limits scales arise. Siliceous nodules, often irregularly shaped, occur in the cortical cells. The vascular systems of the rhizomes of both types have been mentioned as conforming with the “Lindsaya-type’ by Gwynne-Vaughan and Bower. This is further indicated in Text-figs. 2-4. These figures represent sections made from the internodes of the respective types. It will be observed that the greater part of the bundle is occupied by xylem, accompanied by xylem-parenchyma and surrounded by phloem, pericycle and endodermis. Towards the upper or dorsal surface of this solid core of xylem, a pocket of internal phloem associated with internal parenchyma occurs. The dorsal mass of xylem is very thin, forming a bridge over the internal phloem, while the ventral mass of xylem is very much thicker. The xylem tracheids are of fairly uniform size and are composed solely of scalariform elements; spiral or annular tracheids corresponding to the Text-figures 1-8. 1. A longitudinal section of the stem apex of Lindsaya microphylla showing the tetrahedral apical cell (X) and the typical segmentation. x 180. 2. Transverse sections of the vascular bundles of L. microphylla (2a) and L. linearis (2b), showing the central core of xylem (zy) with its internal phloem (i. ph.) surrounded by external phloem (ex. ph.), pericycle (p) and endodermis (en). x 40. 3. A transverse section of the dorsal portion of the vascular bundle of L. microphylla. “zy. p., xylem parenchyma; wy., xylem; i. ph., internal phloem; »m., internal parenchyma ; ex. ph., external phloem; p., pericycle; en., endodermis. x 180. 4. A transverse section of the dorsal portion of the vascular bundle of rhizome of L. linearis. x 180. The same as the above except that a leaf trace is about to depart, forming a small dorsal projection. The break in the continuity of the dorsal are of xylem will occur in the vicinity of w x 100. 6. A stage further advanced in the departure of the leaf trace. The dorsal arc of xylem has broken and rotated laterally outwards. Internal and external phloem are in communication. p. xy., protoxylem. x 100. 7. The gap caused in the xylem by the departure of the leaf trace has again closed. x 100. 8. A transverse section of the internal phloem showing xylem tracheid incorporated within it. x 180. ou BY MAY M. WILLIAMS. eo. SES: CAS SE OK eR A VEBCZOS\ 5 OCG Y yy Qs AX MOVANT ep SIG) Z ete oes Ma ORO Iw SBOU 393 394 ANATOMY OF LINDSAYA LINEARIS AND L. MICROPHYLLA, protoxylem were not observed, although a careful search was made for them. Xylem parenchyma is more abundant in the rhizome of L. microphylla than in that of L. linearis. The “internal pocket’ is composed of phloem and parenchyma. In L. linearis, this parenchyma consists solely of a layer of cells separating the xylem and phloem; in ZL. microphylla, the internal phloem itself has a few intermingled parenchymatous cells as well. The sieve tubes of both the internal and external phloem are of fairly uniform size, and show no differentiation into protophloem and metaphloem, protophloem elements not being present in the stem. It is interesting to note in this connection that Gwynne-Vaughan describes protophloem as being present on the outer margin of the external phloem of Davallia repens. The external phloem consists of one or two rows of sieve tubes surrounding the xylem and separated from it by a layer of parenchymatous cells. The pericycle is composed of large cells, two to four cell layers deep and surrounded by the characteristic endodermis. The root trace departs from the outer margin of the ventral portion of xylem directly opposite the dorsal arch and passes through the cortex with the minimum of disturbance. The leaf traee comes off as a single curved strand, in departing breaking the dorsal arch of xylem, thus placing the internal and external phloem in communication (Text-figs. 5-7). Text-figure 5 represents a section approaching the node, and it will be observed that here the internal phloem has formed a projection resulting in a thrusting up of the dorsal are of xylem in its vicinity. This is brought about by an increase in the amount of phloem present in the stem, causing this dorsal projection in order to accommodate the added bulk. The phloem now commences to extend laterally outwards; the strain thus set up causes a breaking at the dorsal are at a point X (Text-fig. 5), where this dorsal projection meets the unaltered part. This break in the continuity of the dorsal xylem places the internal and external phloem in communication. There is no involution of the endodermis. The free end of the xylem then commences to rotate laterally out- wards, and is closely followed by the internal phloem. Thus the internal phloem of the leaf trace is derived from the internal phloem of the bay. This condition is indicated in Text-fig. 6. A wide gap is formed in the stelar tissues as a result of the movement of the arc, and the free communication of-the internal and external phloem is very pronounced at this point. It is here that protoxylem elements begin to make their appearance in the trace which had hitherto been without them. The arc of xylem now commences to raise itself from the dorsal surface of the broken stele, and deepens its concavity so as to take on a gutter- Text-figures 9-15. 9. A longitudinal section of the apex of the petiole of ZL. linearis showing the wedge- shaped apical cell forming two rows of segments. x 200. 10. A transverse section of the petiole of L. linearis. x 45. 11. A transverse section of the petiole of L. microphylla. .x 25. 12. A transverse section of the petiolar bundle of LZ. microphylla showing three pro- toxylem groups (p. zy.) and M shaped mass of metaxylem surrounded by phloem pericycle and endodermis. x 200. 13, A transverse section of the petiolar bundle of L. linearis. x 200. 14. A transverse section of the pinnule of JL. linearis showing upper and lower epidermis with thick cuticle, stomata confined to the lower surface, and spongy « mesophyll. x 110. 15. A longitudinal section of the root apex of L. linearis showing the Pyramidal apical cell with its segmentation, and the root cap. x 200. 395 BY MAY M. WILLIAMS. C.-8% JES | CL aman} Tb UI S, Te t ) XS A wee: War 4 aes xX oi sega 396 ANATOMY OF LINDSAYA LINEARIS AND L. MICROPHYLLA, shaped form. The broken dorsal vault again becomes complete before the leaf trace is properly detached (Text-fig. 7). The arc of xylem continues to raise itself until it becomes detached from the xylem of the stem. At the same time the external phloem, pericycle and endodermis become folded in until, finally, complete separation of the two structures is brought about and the meristele of the petiole is formed, its plane of Symmetry being parallel to the dorsal surface of the stele. These facts are in very close agreement with those described by Tansley and Lulham (1902, pp. 158-9) in L. orbiculata, with one important difference, namely, that no internal endodermal cells appear in the internal phloem in connection with the departure of the leaf trace. These investigators have stated that “as the free end of the arc moves away from the stele the internal endodermis approaches the outer one, they touch and open, thus placing’ the inner strand of parenchyma which had dilated considerably into connection with the cortex’. In L. microphylla and L. linearis the leaf trace, as it separates, is always limited externally by unbroken endodermis. Although not directly stated, it seems that the underlying idea of these authors is that the internal pocket arose by flow of tissue, the parenchymatous cells being derived from the cortex. McLean Thompson (1920, p. 727), who describes the structure in terms of the ontogeny, considers that “the formation of medulla and inner phloem involves neither cortical intrusion nor tissue flow, but is the result of static intrastelar change by which also the pockets in the adult stele arise.” During the course of the investigation, the writer observed the presence of isolated tracheids in the internal phloem pockets, further indication of change of procambial destination of cells in situ (Text-fig. 8). This is the idea put © forward by Bower. It is generally recognized that the great interest of the “Lindsaya-type” of stele is that it appears to furnish a phylogenetic link in the sequence from the protostele to the solenostele. Gwynne-Vaughan (1903, p. 717) expresses very concisely the nature of this change: “As the leaf and the trace increased in importance relative to the stem, the phloem lying on the adaxial side of the leaf trace became extended downwards into the substance of the xylem of the protostele. Gradually reaching further down through the internode, this internally decurrent phloem at length comes into contact with that decurrent from the leaf trace below, and a continuous solid core of phloem was thus formed within the stele. Then the ground tissue lying in the adaxial concavity of the leaf trace also began to extend downwards into the stele forming at first a prolongation that ended blindly in the core of phloem but eventually it reached down from one leaf trace until it met with that decurrent from the next leaf trace above. In this manner an internal strand of ground tissue was formed which is continuous throughout the stem and the stele became a solenostele’’. This mode of formation of the solenostele implies that there was present a continuous core of internal phloem before the ground tissue became decurrent at all. But, taking into consideration Boodle’s (1901) description of the structure of the node of Gleichenia flabellata, it appears that phloem and ground tissue were decurrent together into the xylem of the protostele. It follows that the soleno- stele may have originated by either of these two methods, and does not necessarily come along one line of descent. Another feature of interest with regard to this ‘“Lindsaya-type” of stele is that it has frequently been described in the ontogeny of higher types of stele. Since BY MAY M. WILLIAMS. 397 Lindsaya presents other features which are relatively more advanced, Bower (1918, p. 61) considers that Lindsaya may be held to have stood still anatomically at an early stage of the ontogeny, a fact possibly related to its restricted leaf development, while others with more ample leaf development have progressed to solenostely and other complications. B. Leaves. A two-sided wedge-shaped apical cell forming two rows of segments is the type for the leaf of both forms. This is shown in Text-fig. 9 (L. linearis). The outer divisions of the segments are the marginal cells which give rise to the pinnae of the leaf; the other segments are mainly formative of the petiole. As each pinna corresponds to a segment of the apical cell, it follows that they alternate on opposite sides of the rachis. The leaf first makes its appearance as a blunt conical emergence, but as the leaf grows it assumes the form of a flattened cone with a broad base, more convex on the outer side and very soon showing circinate vernation. 1. Petiole—The minute anatomy of the petiole of L. microphylla and L. linearis is practically identical. The transverse section is circular to oval in outline (Text-figs. 10-11). The cortex is composed of very large, thin-walled cells, particularly large in L. linearis, in which intercellular spaces are absent. Two or three layers of sclerenchyma occur on the outer limits. a The centre is occupied by a single vascular strand, subrotund in outline. The petiolar bundle is concentric and contains a xylem strand, f-shaped in L. microphylla (Text-fig. 12), the arms of the being very thick, and /\-shaped in L. linearis (Text-fig. 13), the bases being directed towards the apex of the stem. The metaxylem elements in L. linearis are few in number but very large, in some instances the arms of the / being occupied by a single row of tracheids. These large tracheids have scalariform thickening laid down on their walls. Immediately one end of the leaf trace becomes separated from the stem stele, protoxylem groups make their appearance in the bundle which had hitherto been without them; these protoxylem groups being three in number. ‘Two are placed on the ends of the arms of the xylem, appearing when the ends become separated from the stem bundle; consequently one appears before the other. The third group is placed at the apex and its position is variable in the two species. In L. microphylla it was observed in all positions intermediate between endarchy and exarchy, primarily endarch but finally it forms the con- nection between the two arms of the . In L. linearis, on the other hand, this third protoxylem group is always exarch, appearing in that position from the first in the leaf trace. It often, in section, has the appearance of being squeezed out by the large metaxylem elements. The three protoxylem groups in the Polypodiaceae are considered as being primarily endarch. A somewhat similar transition of the median protoxylem group from an endarch to an exarch position was described by Marsh and confirmed by the writer (1924) in various species of Cheilanthes. Here the movement of this protoxylem group may be traced at different levels of the petiole till it becomes exarch in position, and finally, in Cheilanthes fendleri and Ch. tenuifolia, becomes completely detached from the main mass of metaxylem elements, and is broken up into three or four smaller groups. Bertrand and Cornaille (1902, p. 99) considered that the trace with three protoxylem groups resulted from the fusion of two binary chains (the latter 398 ANATOMY OF LINDSAYA LINEARIS AND L. MICROPHYLLA, term being employed for the trace with two protoxylem groups) and the reduction of the median bipolar group thus formed to a minimum. Both of these forms have been derived in their turn from more complicated types. Gwynne-Vaughan (1903, p. 722) has objected, however, that hardly any evidence can be brought forward to support this theory of reduction. The facts rather seem to the writer to present a series in the progression of the leaf trace. The movement of this median protoxylem from the endarch to the exarch position causes a break in the continuity of the metaxylem elements which appear to die out at this level. The protoxylem comes to form the only connection between two distinct metaxylem masses, and by the splitting of this protoxylem group in this position into two, there would be formed two xylem masses, each with two endarch protoxylem groups. Later, the phloem, pericycle and endodermis might follow this; thus two bundles would be present in the petiole. These in their turn, by further division, would give rise to more complicated types. The phloem of the bundle in the forms under consideration is separated from the xylem by a single layer of parenchyma. It consists of two rows of sieve tubes which surround the xylem without any interruption in their continuity. The sieve tubes are of uniform size and represent the metaphloem. Protophloem is not represented in either type. This is surrounded by the pericycle which is several layers deep. The whole is encircled by the endodermis. When the Text-figs. 16-34. 16. A longitudinal section of the root apex in Lindsaya microphylla. x 200. 17. A transverse section of the root apex of L. linearis showing the apical cell and its segmentations. x 200. 18. A transverse section of the root of ZL. linearis. x 45. F 19. A transverse section of the root bundle of ZL. linearis showing its diarch nature. The xylem is surrounded by phloem, pericycle and endodermis. x 200. 20. A transverse section of the root bundle of Lindsaya microphylla. x 200. 21. A cortical cell of the root of L. linearis, showing the septate branching and multi- nucleate nature of the fungus infecting it. The nucleus of the host is well developed. x 450. 22. A longitudinal section of the pinnule of ZL. microphylla in the region of the sorus showing the nature of the indusium and the receptacle deflected towards the lower surface. x 45. 23. A longitudinal section of the young sporangium of ZL. microphylla showing the basal cell and the sporangial cell proper. x 200. 24. A longitudinal section of a later stage in the development showing the central cell which has cut off the wall cells and the three rows of cells in the stalk. x 200. 25. The central cell is here cutting off the tapetum. x 200. 26. At this stage of development the archeosporium has divided intag four cells, the spore mother cells. x 200. 27. A longitudinal section of the sporangium showing the disintegrated tapetal cells. The spore mother cells have enlarged, separated and become rounded. x 200. 28. A transverse section of a spore mother cell showing the chromosomes after synapsis. x 450. 29. A section of a spindle formed during the division of the spore mother nucleus, showing three of the four poles which are sometimes formed. x 450. 30. A mitotic spindle formed during the nuclear division of a cell of the petiole tip of L. microphylla, showing the large number of chromosomes present. x 450. 31. A mitotic spindle formed during the nuclear division of the spore mother cell of L. microphylla showing the reduced number of chromosomes. x 450. This indicates the typical arrangement of the spores formed from the division of the spore mother cells of LZ. microphylla. x 200. 33. The spores of L. linearis. x 200. 34. The spores of L. microphylla. x 200. BY MAY M. WILLIAMS. 399 400 ANATOMY OF LINDSAYA LINEARIS AND L. MICROPHYLLA, petiole reaches the level of the pinnae, branches are given off from the petiolar bundle, which constitute the vascular supply to the rachis. It is perhaps worthy of note that very occasionally the writer has observed, in L. microphylla, the petiole itself branching dichotomously. 2. Pinnules—In both types the barren pinnules are frequently lobed. The veins show a very regular dichotomous branching. The pinnules of L. linearis are of a much deeper green colour than those of L. microphylla. ' The cells of the upper and lower epidermis are large and have a very thick cuticle. The stomata are confined to the lower surface of the leaf and open into large air cavities (Text-fig. 14). The pinnule in both types is charac- terized by the absence of palisade tissue, the whole of the tissue between the two epidermes being composed of parenchyma with large intercellular spaces. The cells of this mesophyll contain very large chloroplasts which are distributed fairly evenly around the margins of their walls. The absence of palisade tissue and the arrangement of the chloroplasts probably bear relation to the light intensity. The vascular bundle of the pinnule is usually collateral. It is surrounded by endodermis, composed of large thin-walled cells. This encloses a mass of tracheids representing the xylem surrounded by parenchyma with a few sieve tubes. C. Roots. The roots arise in connection with the bases of the leaves. The primary roots branch profusely; the branches arise in two lines, corresponding to pro- toxylem groups in the vascular bundle of the primary root. Text-figures 15, 16, 17 indicate that the apex of the root of both ferns is characterized by the presence of a three-sided pyramidal apical cell, from the sides of which segments are cut off in regular succession. The formation of the body of the root and the root cap may be traced to definite segments of the apical cell. The cortex is two or three cell layers deep and very quickly becomes sclerized (Text-fig. 18). The vascular bundle of the root is surrounded by the endodermis which is not always to be observed in the older portions (Text-figs. 19-20). The root is diarch; xylem is composed of protoxylem and metaxylem; the elements of the latter are large and have scalariform thickening laid down on their walls. The phloem is well developed and is separated from the xylem by a layer of parenchyma. It does not, however, extend round beyond the protoxylem groups where xylem and pericycle are in contact. There is a multilayered pericycle, the cells-of which are large and often have a great deal of starch stored within them. A curious feature of the root in both forms is that fungal hyphae appear in the cortical cells a little behind the apex (Text-fig. 21), indicating the presence of mycorhiza in the roots. The cortical cells in which the fungus is present are large and have thin walls. Fungal hyphae were observed passing through the walls from one cell to another. The fungus is a branching, septate, multinucleate form. The ends of the branches are often swollen. Fungal sporangia were also observed in the cells. No root nodules are visible externally, so that the presence of the fungus evidently does not stimulate the plant to form new tissues. Symbiosis was observed to be a constant feature in the root, no matter from what situation the plant was collected. Mycorhiza have been recorded from other parts of the Pteridophyta. Laing (1899) describes an endophytic fungus present in the prothallus of Lycopodium clavatum. In 1917, Lawson found the same in the prothalli of Tmesipteris and Psilotum. BY MAY M. WILLIAMS. 401 D. Sorus. The sporophylls are usually not lobed, but are often rolled inwards, affording protection for the sori. The latter are borne on the ends of the veins which are linked together distally to form a continuous vascular commissure. The sporangia have followed this extension of the vascular supply and thus the fusion sorus has arisen. This is characteristic of the genus. The tracheids composing the vascular commissure to the sporangia are of a different nature from those present in the ordinary veins of the leaf. Those of the vascular commissure are shorter, broader and of the nature of storage tracheids. This has also been described by Bower (1914) as occurring in connection with the fusion sorus of the Blechnum type. The nature of the sorus, indusium and receptacle, and their early development in Lindsaya linearis have been carefully investigated and described by Bower (1918, pp. 14-15). The marginal cells, as in Dicksonia, become the apex of the receptacle by segmentation; the two indusial flaps originate intramarginally, as in Dicksonia, differing in bulk, the inner (abaxial) being thinner while the receptacle is deflected towards the lower surface. The sporangia at first show a gradate sequence, but later this is not strictly maintained, and young sporangia become interpolated between the older, leading to the typical mixed condition. These facts are also true for L. microphylla, as indicated in Text-fig. 22. The sporangia develop in the manner typical for Leptosporangiate Ferns. One of the cells forming the wall of the receptacle enlarges and is cut off by a trans- verse wall into a basal cell which undergoes no further division, and an upper cell which forms the sporangium proper (Text-fig. 23). The upper cell undergoes divisions which result in the formation of a pyramidal apical cell from which various segments are cut off, giving rise to the three rows of cells seen in the stalk. These undergo no further longitudinal divisions. The upper part of the sporangium enlarges and a periclinal wall is formed in the apical cell. Thus the wall cells are cut off, and remain permanently one layer thick (Text-fig. 24). The central cell of the sporangium cuts off a single layered tapetum by further division (Text-fig. 25), and this, later by a further periclinal division, becomes double. The archeosporium develops up to the quadrant stage (Text-fig. 26), but often undergoes no further development in L. microphylla, so that there are usually only four spore mother cells, although occasionally six are present. In L. linearis, however, there are frequently eight spore mother cells developed. As the subsequent development is the same in both types, in order to avoid unnecessary duplication, the following description and text-figures will be taken from L. microphylla. As soon as the spore mother cells are formed, the tapetum begins to disintegrate, the spore mother cells separate, and become rounded and very much larger (Text-fig. 27). The nuclei of these cells enlarge considerably until they become almost as large as the cell itself, leaving a very thin selvage of protoplasm around the wall. The linin network of the nucleus at this time becomes very distinct. After the complete disintegration of the tapetal cells, the nucleus prepares for division. It undergoes the typical stages associated with reduction division, chromosomes are formed (Text-fig. 28) and finally a spindle is produced. The chromosomes pass to the poles of the spindle, and two nuclei are produced. These nuclei undergo another division differing from the former in that it is homotypic 402 ANATOMY OF LINDSAYA LINEARIS AND L. MICROPHYLLA, in nature. Thus four nuclei, and finally four spores are produced from each spore mother cell, the tetrad having the typical arrangement indicated in Text-figure 32. It frequently happens that the second division occurs so rapidly after the first that a spindle with four poles is produced (Text-fig. 29). The number of chromosomes present is the same as that found associated with the second division. This is frequently found in the division of the spore mother cell of the Filicales. Text-figure 30 represents a mitotic spindle observed in a petiole tip. Here the number of chromosomes present was counted as being from 32 to 40. In the second division of the spore mother nucleus only 16 chromosomes are present (Text-fig. 31). The wall becomes organized into various parts (annulus, etc.). This has been described and figured by Bower for L. linearis (1918, p. 16, fig. 14). After the division of the spore mother cell, the four spores are enclosed within the mother membrane. This disintegrates and the spores become separated, their walls undergoing various modifications. The outer wall becomes very thick, and spiny. The form of the spore in both cases is tetrahedral (Text-figs. 33-34). The number of spores present in L. microphylla, by actual count, was 16-24, in L. linearis 24-32. The latter confirms the estimate made by Bower. Bower (1918) considers that the marginal origin of the receptacle in such types as Lindsaya and Dicksonia, etc., where it becomes downward-directed as development proceeds, illustrates how the initial steps in the modification to a superficial sorus may be actually observed in the ontogeny of the typical Marginales. Conclusion. The view which is universally supported at the present day, is that Lindsaya is a genus anatomically primitive, but possessing other characters which are more advanced. Its affinities, as determined by anatomical structure and soral character, are with the Davallieae. The soral characters also indicate relationship with the Dicksonieae, except that as the sorus becomes older it assumes a “mixed” character. Thus the sorus is more advanced than that of the Dicksonieae, although the anatomical structure is very much more primitive. In respect to its anatomical structure, then, Lindsaya takes a lower place than any of the Dicksonoid series. The mature stelar structure is regarded as furnishing a phylogenetic link between the protostele and the solenostele. These are regarded as primitive types of vascular arrangement in the Polypodiaceae. Gwynne-Vaughan (1903, p. 732) remarks that these primitive characters do not run parallel with Bower’s division of the order into Gradatae and Mixtae. Bower, however, supports the view that there are several lines of descent among the Polypodiaceae alone, and a primitive type of vascular arrangement might occur in the primitive members of each line of descent. He considers that in Lindsaya no structural advance was made to complete solenostely, owing to the restricted size of the species of the genus which made distension of the stele unnecessary. BY MAY M. WILLIAMS. 403 It seems that the convergence of evidence from many forms, indicates that anatomical structure and soral characters have not necessarily advanced together in the progressive development of the Filicales, that one or other might have advanced while the other remained stationary; that is, these characters have taken quite independent courses of evolution. Factors influencing one have not necessarily made themselves felt upon the other. Since the structure of the rhizome of Lindsaya is constant in so many species, it has been suggested that it forms a good generic character. Summary. 1. Apex of the stem in both types is occupied by a tetrahedral apical cell. 2. The stem anatomy of Lindsaya linearis-and L. microphylla coincides with the “TLindsaya-type”’. The “internal pocket” of phloem is due to change in procambial destination of cells. i 3. The leaf trace departs as a single strand from the dorsal arch in much the same manner as described by Tansley and Lulham, except that no internal endodermis appears. 4. Apices of the leaves are occupied by a two-sided wedge-shaped apical cell forming two rows of segments. 5. Petiolar bundle forms a single strand containing three protoxylem groups, two of which are endarch in position, the third (medium) group in L. linearis exarch, but in ZL. microphylla occupying positions intermediate between endarchy and exarchy. 6. Stomata are confined to the lower surface of the pinnule. Cuticle thick. No palisade tissue present. 7. Apices of root occupied by a three-sided apical cell. 8. Roots are diarch. 9. Symbiosis occurs in the roots of both forms, fungal hyphae causing the infection being septate and branching and multinucleate. 10. Sorus is marginal in origin; true indusium is present; sorus is mixed in character. 11. Tracheids composing the vascular commissures are of a different nature from those of the ordinary veins of the pinnule. 12. In L. microphylla, 4-6 spore mother cells are formed, in L. linearis, usually 8. 18. Sorus of Lindsaya indicates how the initial steps in the ontogeny of the superfilicales may have occurred. 14. Nearest affinities of the Lindsayas are the Davallieae. In conclusion I wish to thank Professor Lawson, in whose laboratory the investigation was carried out, for advice and guidance; also Dr. McLuckie and Mr. Brough for helpful criticism. Literature Cited. BERTRAND and CORNAILLE, 1902.—ftude sur quelques caractéristiques de la structure de Filicinées actuelles. Mémoires de L’Université de Lille, tome 29. Boopue, L. A., 1901.—On the Anatomy of Gleicheniaceae. Ann. Bot. xv, 703-745. Bower, FE. O., 1908.—Origin of a Land Flora. , 1914.—Studies in the Phylogeny of the Filicales. iv. Blechnwn. and allied Genera. Ann. Bot. xxviii, 363-431. , 1918.—Td., vii. Pteroideae. Ann. Bot. xxxii, 1-68. , 1923.—The Ferns. Camb. Bot. Handbooks. CAMPBELL, D. H., 1905.—Structure and development of Mosses and Ferns. 404 ANATOMY OF LINDSAYA LINEARIS AND-L. MICROPHYLLA. GwWYNNE-VAUGHAN, D. T., 1901.—Observations on the Anatomy of Solenostelic Ferns. i. Loxsoma. Ann. Bot. xv, 71-92. , 1903.—Id., ii. Ann. Bot. xvii, 689-742. TANSLEY, A. G., 1908.—Lectures on the Evolution of Filicinean Vascular System. New Phytologist, 2. TANSLEY, A. G., and LuLHAM, R. B., 1902.—On a new type of Fern Stele and its probable phylogenetic Relations. Ann. Bot. xvi, 157-164. THOMPSON, J. MCLEAN, 1920.—New Stelar Facts and their Bearings on the Stelar Theories of Ferns. Trans. Roy. Soc. Edin. Vii, Pt. 4, 715-735. WILLIAMS, MAy M., 1924.—Anatomy of Cheilanthes tenuifolia. Bot. Gaz. 1xxviii, No. 4, 378-396. THE DEVELOPMENT OF TWO AUSTRALIAN SPONGE-CRABS. By Hersert M. Hate, Zoologist (Crustacea), South Australian Museum. (Contribution from the South Australian Museum.) (Plates xxxix-xl, and five Text-figures. ) [Read 25th November, 1925.] It has been often stated that all marine Brachyura hatch as Zoeae, and pass through two or more pelagic larval stages before reaching the adult form. In 1914, however, Miss Rathbun recorded a female of Paranazia serpulifera from the Monte Bello Islands, Western Australia, with “young crabs in the adult state” beneath the pleon; she remarks that juveniles in two stages thus accompany the mother “but whether the first is hatched directly from the egg or not, it is impossible to tell”; both young stages are figured by Rathbun. The complete life-history of the majority of marine crabs is unknown, and Rathbun (1918), bearing in mind the case of the Spider-crab from Australia, later pertinently remarks, “It is unwise to draw sweeping conclusions from a few cases”’. In 1922, Montgomery described some well developed young taken from beneath the pleon of an Australian Dromiid crab (Paradromia lateralis), and states that, apart from Rathbun’s Oxyrhynch, “this is the only other available record of such a case among marine Brachyura’”’. Indication that some Australian Sponge-crabs do not have pelagic larval stages is to be found, however, in a paper by Haacke (1885), who suggested that the free-swimming larvae of sponges and Tunicates lodge on the carapaces of the young crabs sheltering under the abdomen of their parents, and that each member of a brood thus acquires its cloak from the mantle of its mother. The title of Haacke’s paper. does not indicate that these juvenile crabs are mentioned by him, and he does not remark upon their presence beneath the pleon as unusual; further, the names of the Dromiids concerned are not given. Haacke states that he examined pea-sized Dromiids which had the sponge- cloak firmly fastened to the carapace (and not merely held in position with the posterior chelipeds) so that the crabs could not be robbed of their covering with- out injury. He then concludes that, as stated above, the cloak is possibly acquired very early in life, and that as a result of this probable “erblicher Symbioses’, there is a tendency for different Dromiid species to have special species of sponges and Ascidians. There are before me an adult female of Cryptodromia octodentata with young beneath the abdomen, and several specimens of Paradromia lateralis like- wise burdened. The ova of these two species, as is usual with crustaceans which have a “direct” development, are relatively large; this is also the case with a third Australian Sponge-crab—Platydromia thomsoni—and leads one to suspect that this species likewise hatches at an advanced stage. The metamorphosis is apparently almost completely suppressed in Paradromia lateralis and Cryptodromia 406 e DEVELOPMENT OF TWO AUSTRALIAN SPONGE-CRABS, octodentata, and possibly in Paranazia serpulifer also; it must be remarked, how- ever, that the pleopods of the juveniles of the two Sponge-crabs are very similar to the natatory abdominal appendages of the Megalopas of some other crabs. CRYPTODROMIA OCTODENTATA Haswell. Cryptodromia octodentata is one of the largest of the Australian Sponge-crabs, the carapace of some specimens attaining a width of 85 mm. The cloak is usually a sponge or, more rarely, a mass of Ascidians, but sometimes other objects are held over the carapace. A specimen recently observed was using a piece of kelp weed as a shield and the ovigerous female illustrated on Plate xl is sheltered by the valve of a Fan-shell (Pecten medius); this shell admirably fits the back of the crab, and portion of it is so conveniently broken as to enable the crustacean to grip the inner edges of the chipped part with the hind chelae and so hold the mask firmly in position. An old example, 74 mm. in width, holds no shield on the back, but the carapace is considerably eroded and attached to it is an assortment of marine growths, including two species of plants. The pleon of the adult female, with the aid of the exopods of the pleopods, forms a veritable pouch for the retention of the ova and later, as it proves, for the little crabs. In large examples the abdomen is half as wide as the carapace and, if it be closely adpressed, its tip overlaps the basal fourth of the outer maxillipeds; thus, when the tip is placed in its normal position between the anterior edges of the coxae of the large chelipeds, the greater part of the pleon stands away 10 mm. or so from the body. In specimens which bear a large mass of eggs, the abdomen is naturally forced far out of its usual position, but the contents of the apron are prevented from falling out by the large, flattened exopods of the second to fifth pleopods (Text-fig. 5b); the outer rami of these appendages are densely fringed with outstanding hairs, and are so shaped that, when pushed outwards by the eggs or young, they overlap and form perfect side walls to the brood pouch. The ova of some dried examples are more than 1mm. in diameter but were probably larger in life, for Rathbun (1923) states that “The eggs are very large and numerous, being 2 mm. in breadth”; the ova are attached to the long hairs of the slender, two-jointed endopods of the pleopods. As previously mentioned, one female (preserved in alcohol) has a number of juveniles under the pleon; in this case the capacity of the brood pouch is strained to the utmost, for the well developed family consists of five hundred and thirty individuals (Pl. xxxix). The pleon and the exopods of the pleopods are forced far out from the body to accommodate the young crabs, some of which have the legs clasped around the hairs of the endopods of the pleopods. A description of these juveniles, and notes concerning a few of the succeeding stages, are given below. Brood Young. Text-figs. la to If. Colour completely bleached. Exoskeleton thin. Carapace very convex, about one-fourth longer than greatest width, the surface smooth and clothed with Text-figs. 1-5.—Cryptodromia octodentata Haswell. 1. Brood young; la and 1b, dorsal and ventral views (x 12 diams.); le and 1d, first and second antennae (x 83 diams.); le, pleon (x 213 diams.); 1f, pleopod of fifth segment (x 53 diams.). Carapace of young example 6.9 mm. in width (x 5 diams.). Carapace of young example 15 mm. in width (x 234 diams.)._ Carapace of half-grown example 29 mm. in width (x 1.4 diams). . Adult female; 5a, carapace, (x 0.9); 5b, pleopod of fifth segment (x 1.9 diams.). ote & bo BY H. M. HALE. 408 DEVELOPMENT OF TWO AUSTRALIAN SPONGE-CRABS, moderately long hairs, which are plumose for their whole length; the hairs are sparse on the posterior two-thirds of the dorsum. Branchial (or cervical) groove not very distinct. Front cut into three teeth, the median projection small, down- bent and apically bifid; each of the outer teeth (Text-figs. la and 1b, A) is capped with three or four short, stout spines and has plumose hairs and a few spines on the margins. A strongly marked angle (the supraorbital tooth) at about the middle of the orbit bears three spines (Text-figs. la and 1b, B). The outer, or posterior, angle of the orbit is slightly prominent and bears two or three spines (Text-figs. la and 1b, C). The anterolateral borders are provided with five to seven spines and with plumose hairs, and are a little incised at about two- thirds of their length (Text-figs. la and 1b, D), the incision followed by a slight protuberance which bears the last spine or two of the anterolateral series. The posterolateral margins are slightly convex and convergent, furnished anteriorly with a few spines and plumose hairs; there is a projection of the margin imme- diately behind the branchial groove (Text-figs. la and 1b, BF). Suborbital lobe armed with five or six spines; no notch in the orbital margin below the outer angle. Subhepatic region very slightly swollen. First antennae (Text-fig. 1c) with peduncle stout, three-jointed; with two flagella, the inner of which is short, slender and composed of three articles, while the outer lash consists of six articles (the basal three stout) and is furnished with a dense brush of long hairs. Peduncle of second antennae four-jointed (Text-fig. 1d); second article furnished with spines and a few plumose setae; with inner angle slightly produced and with outer apical part strongly forwardly produced and lobular; flagellum rather long, composed of nine articles. Eyes large. Margins of ischium and merus of third maxillipeds furnished with short, stout spines. Chelipeds and legs much as in adult, armed with some marginal spines, with slender, plumose setae and (near apices) with a few simple setae; serrations of fingers of large chelae coarse. Pleon bent under body, narrow, seven-segmentate, furnished with plumose hairs and a few short spines; apex of telson slightly incised; third to fifth segments with one spine, and sixth with two spines, at each posterolateral angle (Text-fig. le). Pleopods of second to fifth segments stout, long, each with protopod large and more than half as long as exopod, which is flattened and provided with sixteen long, plumose hairs on the margins; endopod rudimentary, unjointed, not much more than one-third as long as exopod and with apex furnished with four minute hooks (Text-fig. 1f); the apex of the exopod of the last pair of pleopods reaches to level of three-fourths of length of telson, and the fringing hairs extend well beyond apex of pleon. Length of carapace, 2.8 mm.; width of carapace, 2.2 mm. Growth Changes. Text-figs. 2-5. There is also before me a series of examples of various sizes, most of the small specimens being preserved in a dry state. The smallest member of this series is 4.8 mm. in length and 4.1 mm. in width, and differs from the brood young in the following particulars: The carapace is relatively slightly wider at about the middle of its length and the front is relatively a little narrower. The median tooth is very obscurely subbifid, is bent downwards and slightly forwards, and bears a few small lateral spines. Many of the marginal spines persist, but are very tiny. The posterior (outer) angle of BY H. M. HALE. 409 the orbit is more prominent, and below it is a small V-shaped notch. The supra- orbital tooth is more acute. The suborbital lobe bears tiny spines and the inner angle is acute. The anteroexternal angle of the buccal cavern is somewhat acute and behind it is a slight projection. Four tiny spiniform projections have developed on the anterolateral margins; the first has two spines on its anterior slope, the third is minute, and the fourth is just behind the incision (D in the figures) which is very much less distinct than in the brood young. The clothing (which is mostly denuded) is of the same character, but the hairs are relatively smaller. The apex of the pleon is narrowly rounded. It has been determined that in some other Decapods the general increase in size after a moult amounts to roughly one-sixth, so that if we venture here to apply an approximate rate of growth, this specimen, since attaining the size of the brood young, has passed four, or at most five, ecdyses. The next example is 7.8 mm. in length and 6.9 mm. in width (Text-fig. 2); thus, in proportion, the carapace is again wider and the front narrower. The median projection is simple and is still more forwardly directed. The suborbital lobe bears a few tiny spines and the inner angle is dentiform. The first of the small anterolateral teeth has three minute spines on its anterior slope. The clothing of the dorsum is denuded, but that of the subbranchial regions consists of slender hairs, sparsely plumose for their whole length, and that of the anterior parts of the under surface of short and densely plumose hairs. A specimen 15.5 mm. in length is 15 mm. in width. The median frontal pro- jection is nearly horizontal. The suborbital lobe bears no tiny marginal spines. The anteroexternal angle of the buccal cavern is spiniform and the projection near it is conical. The first of the four teeth of the anterolateral margins is larger than the others, and immediately in advance of the fourth is a very slight protuberance. The dorsum is densely clothed with erect setae, each of which is provided with tiny lateral spikelets, the lastnamed being largest on the apical half. The subbranchial regions bear thin and very sparsely plumose hairs. The carapace of individuals 29-30 mm. in length (Text-fig. 4) is as wide as long. The projection or tubercle just in advance of the last tooth of the anterolateral series is usually moderately prominent, and near it there is sometimes an additional low tumidity on the dorsum of the carapace. The clothing consists of setae which have a brush of lateral setae on their distal half and others (more particularly on the legs) which have lateral spikelets for almost their whole length. The female with brood is 50 mm. in length and 53 mm. in width (Text-fig. 5a); comparing this individual with the juveniles it is now evident that the following changes occur during growth. At each moult the carapace grows relatively wider vehind the level of the orbits and relatively smaller in front of this level, so that, while the distance between the hinder angles of the orbits is more than three- - fourths of the width of the carapace in the young, it is not much more than one-third of the width in the mature female. Thus that part of the anterolateral margins in front of the incision (D) gradually becomes more oblique and the portions posterior to the notch (which at first slope outwards) eventually con- verge. At an early period the tiny marginal spines disappear and four dentiform projections appear on the anterolateral margins while, later, a fifth tubercle (more er less distinct) is developed, very close to the incision (D) which then resembles the interspaces between the other teeth. The projection of the margin below the branchial groove assumes a more or less dentiform or lobular character. The 410 DEVELOPMENT OF TWO AUSTRALIAN SPONGE-CRABS, ’ supraorbital tooth gradually becomes relatively smaller and the outer angle of the orbit more prominent, while a notch appears in the orbit below this angle; the inner angle of the orbit becomes spiniform. A conical tooth develops below the anteroexternal angle of the buccal cavern, which also becomes spiniform. The tumidity on the subhepatic region is accentuated. The clothing changes in character and is finally very dense. The serrations of the fingers of the large chelipeds increase in number, and small tubercles take the place of tiny marginal spines. Additional articles are gradually developed in the flagella of the antennae. The basal article of the peduncle of the first pair is now relatively stouter and the third thinner than in the young; the slender inner lash consists of fourteen articles and the tapering outer flagellum of thirty-two articles, with a dense brush of hairs. Both inner and outer distal angles of the second article of the second antennae are produced and the flagellum contains forty-two articles. In the female the abdomen increases in width. The pleopods change entirely; it is probable that the four pairs of immature abdominal appendages atrophy at an early stage, and that the permanent pleopods of both sexes then develop. Hyman (1920) shows that this happens in Uca pugilator. It has been noted above that the species attains a width of 85 mm. It is evident that the relative widening of the carapace at each moult obtains throughout the life of the animal, for in very large examples the carapace is approximately one-sixth wider than long. In aged specimens the chelipeds, particularly in the males, are very strong, with the serrations of the fingers blunt, and the marginal tubercles are often spiniform. The fourth projection of the anterolateral margins of the carapace may be acutely spiniform, as are the others. The projection behind the branchial groove is prominent, and a row of more or less conical teeth may develop on the posterolateral margin; traces of these posterolateral teeth are sometimes found in small examples. The branchial groove is usually deep and in two examples the cardiac region is well delineated. All the specimens dealt with above were taken in St. Vincent Gulf, South Australia. PARADROMIA LATERALIS Gray. Plate xl, B. The brood pouch of this species is much as in Oryptodromia octodentata; the tleon of an adult female, when closely folded against the sternum, reaches to the middle of the length of the outer maxillipeds. The ova are large, an advanced egg taken from a female 17 mm. in width being 1.14 mm. in length and 1 mm. in width; Henderson (1888) records an ovigerous female from Bass Strait. I am much indebted to Mr. Melbourne Ward for a female of Paradromia lateralis, 14 mm. in width, with young in the apron; this specimen was recently secured “On reef at Shelly Beach, Manly, New South Wales; on sand under stones below low tide.” Several dried examples from St. Vincent Gulf also have young in the brood pouch and in one case juveniles in two stages are present. Brood Young, 1. The little crabs in the earlier stage are very soft, and when relaxed have the appearance of being ready to moult. They are only one-third larger than the ripe ova, being approximately 1.5 mm. to 1.6 mm. in length and 1.15 mm. to 1.25 mm. in width. The pleon is somewhat loosely flexed; the pleopods of the second to fifth segments are much as in the next stage, each having a long protopod, a long fiattened and hair-fringed exopod and a short endopod with minute epical hooks. BY H. M. HALE. 411 These examples, judging from their dimensions, are representatives of the stage described by Montgomery (1922), but their condition is too poor to warrant further description here. The peraeopods are moderately stout and the lastnamed author treats this stage as “definitely post-larval’, although the pleopods are certainly not “similar to those of the adult” as he states. It is strange that the sedentary young of this species and of Cryptodromia octodentata should retain pleopods well developed for swimming. Brood Young, 2. Plate xl, B, figs. la to 1f. The young of the next stage are about one-fourth larger than those of the preceding (the proportionate increase is greater in Rathbun’s Oxyrhynch, namely, “more than one-half’). About twenty examples of this instar are contained in the pleon of the female collected by Mr. Ward, and the following deseription is based upon these fresh specimens :— Colour olivaceous-brown. Exoskeleton thin. Carapace very convex, not much longer than greatest width; dorsum furnished with short spines intermixed with setae which are shortly and sparsely plumose; regions moderately well defined and branchial groove distinct. Front a little downbent, with three teeth, the median of which is armed with three spines on each lateral margin; lateral margins of outer teeth with stout spines and setae similar to those of dorsum, and apices capped with three to four spines (Pl. xl, fig. la, A). A large supra- orbital tooth (Pl. xl, fig. la, B) at the middle of the length of the orbit bears three apical spines, and between it and the posterior angle of the orbit is an inconspicuous projection armed with two spines. Posterior margin of orbit bent abruptly outwards and furnished with a spine; hindermost point of orbit (PI. xl, fig. la, C) not prominent, but margin of carapace immediately beyond this point curved backwards, thus forming a somewhat rounded projection (PI. xl, fig. la, D) which is furnished with about five spines. Behind this is another prominent protuberance of the lateral margins (PI. xl, fig. la, H). Posterolateral margins a little sinuate and slightly convergent, provided with a few spines and setae; there is a projection immediately behind the branchial groove (PI. xl, fig. la, F). Suborbital lobe armed with five or six spines. A short, stout spine and several smaller spines on subhepatic region, very close to posterior angle of orbit. First antennae (PI. xl, fig. 1b) with two flagella; peduncle stout, three-jointed, furnished with a few spines; inner lash short, composed of three rather robust articles, and outer flagellum four-jointed (the three basal articles stout) and with a brush of long hairs. Peduncle of second antennae (PI. xl, fig. 1c) four-jointed, armed with some short spines and a few setae; second article with inner apical angle a little produced, and outer apical part prominently forwardly produced and lobular; flagellum slender, composed of eight articles. Eyes large. Margins of ischium and merus of outer maxillipeds with short spines. Chelipeds and legs stout, armed with short spines, simple setae, and sparsely plumose setae. Pleon bent under thorax, narrow, seven-segmentate, surface furnished with short spines and setae (PI. xl, fig. 1d); lateral margins of telson sinuate and posterior margin incised; tumid posterolateral margins of telson with short spines; second to fifth segments with a spine at each posterolateral angle. Pleopods of second to fifth segments long, eagh with protopod stout and nearly half as long as the exopod, which is flattened and bears fourteen plumose marginal hairs 412 DEVELOPMENT OF TWO AUSTRALIAN SPONGE-CRABS, on the apical half; endopod small, unjointed, scarcely more than one-third as long as exopod, and with four tiny hooks near apex (PI. xl, fig. 1f). Uropods represented by a tiny subcordate rudiment on each side (Pl. xl, fig. le, U). Length of carapace, 1.86 mm.; width of carapace, 1.7 mm. Growth Changes. Plate xl, B, figs. 2 and 3. In a specimen 5.6 mm. in width (Pl. xl, fig. 2) the carapace is a little wider than long and its margins still bear spines, which are, however, very tiny. The supraorbital tooth (B) is relatively smaller than in the brood young and the small projection of the supraorbital border has disappeared. The posterior angles of the orbit are prominent (C), there is a little notch below the angle, and the inner angle of the suborbital border is spiniform. The two projections of the anterolateral margins each end anteriorly in a tubercle (D and #). The regions of the carapace are not defined but the branchial groove is distinct. A small subhepatic tooth (S) is developed. As in Oryptodromia octodentata, the carapace becomes relatively wider behind the level of the orbits, and relatively narrower from this level forwards, at each moult; thus the distance between the hinder angle of the orbit (C) and the apex of the first tubercle of the anterolateral margins (D) gradually increases. Montgomery suggests that either the supraorbital tooth of the young or, alternatively, the tiny projection just posterior to it, becomes the posterior angle of the orbit in the adult; it is not probable that the orbit thus changes its position, nor is such conclusion strengthened by a comparison with the stages of Cryptodromia. The subhepatic tooth is not apparent in the brood young, but evidently appears at an early stage. Other mature characters develop as in the preceding species. An adult female (Pl. xl, fig. 3a) has the carapace distinctly wider than long; the supraorbital tooth is much reduced and no more prominent than the outer angle of the orbit. The inner flagellum of the first antennae consists of nine articles, and the outer of twenty-three articles, while the lash of the second antennae is twenty-jointed. As in Cryptodromia octodentata, the pleo- pods change utterly (Pl. xl, fig. 1f and 3b). PLATYDROMIA THOMSONI Fulton and Grant. A female of this small species, dredged by Sir Joseph Verco in Investigator Strait, South Australia, is 11 mm. in width and bears a score of eggs each more than 1 mm. in diameter. The ova are thus considerably larger than those of some crabs which are known to hatch as Zoeae; for instance, the numerous eggs of a large female of Carcinides maenas are only 0.29 mm. to 0.8 mm. in diameter. It is therefore likely that the young of Platydromia do not emerge from the eggs as Zoeae, but hatch at a later stage of development. References. BAuxss, H., 1921.—Zool. Anzeiger, lii, p. 178. HAACKE, W., 1885.—Wollkrabben und ihre Mdntel. Der Zoologische Garten, xxvi, 1885, p. 203. HENDERSON, J. R., 1888.—Challenger Rep., Zool., xxvii, 1888, p. 5. HyMAn, O. W., 1920.—Journ. Morphology, xxiii, 1920, p. 485-524, Pl. i-xii. 1922.—Smithsonian Report for 1920, 1922, p. 443-460, Pl. i-vi. MONTGOMERY, S. K., 1922.—Proc. Zool. Soc., 1922, p. 193-196, fig. 1-3. RATHBUN, Mary J., 1914.—Proc. Zool. Soc., 1914, pp. 653, 661 and 662, PI. ii, fig. 9-10. — 1918.—Bull. U. S. Nat. Mus., xevii, 1918, p. 12. 1923.—Hndeavour Biol. Res., v, 1923, p. 153. 1924.—Arkiv for Zoologi, Band 16, No. 23, 1924, pp. 6-7. BY H. M. HALF. 413 EXPLANATION OF PLATES XXXIX-XL. Plate xxxix. Cryptodromia octodentata.—Female with young beneath the pleon; the sponge cloak of the mother is shown above (nat. size). Plate xl. A. Ovigerous female of Cryptodromia octodentata sheltering beneath the valve of a Pecten (2? nat. size). B. Paradromia lateralis. Fig. 1. Brood young; la dorsal view (x 17 diams.); 1b and le, first and second antennae (x 57 diams.); 1d, pleon (x 22 diams.); le, lateral view of sixth and seventh segments of pleon (x 57 diams.) ; 1f, pleopod of second segment (x 57 diams.). Fig. 2. Carapace of young example 5.6 mm. in width (x 6% diams.). Fig. 3. Adult female; 3a, carapace (x 3% diams.); 3b, pleopod of fifth segment (x 7 diams.). DESCRIPTIONS OF NEW SPECIES OF AUSTRALIAN COLEOPTERA. Parr xvim. By ArtHuR M. Lea. [Read 28th October, 1925.] SCAPHIDIIDAHE. ScAPHISOMA POLITUM Macl. S. queenslandicum Blackb. The types of S. politwm agree well with some specimens that were compared ‘and agreed with the types of S. queenslandicum. HISTERIDAHE. PLATYSOMA INCONGRUUM, 2D. SD. Black, legs and antennae obscure reddish-brown. Head with a transverse oblong, defined by impressed lines, and containing sharply defined but rather small punctures; clypeus with minute punctures. Prothorax about twice as wide as the median length, sides feebly rounded, but in front strongly rounded to the apical emargination, near each side with a wide and rather deep channel, each side in consequence appearing to have a fairly wide elevated margin, on the inner edge of which, part of the submarginal stria may be faintly traced; punctures small and irregular, becoming fairly large in parts of the sublateral channels and almost absent from the margins. Scutellum small and triangular. Elytra about one-fourth longer than wide, each with six well- defined dorsal striae, of which the second is not quite as long as the others; epipleurae each with three striae and some fairly distinct punctures. Propygidium and pygidium opaque, finely punctate and shagreened. Prosternum with two short striae between front coxae; chin piece wide and bilobed. Metasternum with a narrow median line; middle impunctate, but side-parts with distinct punctures of varying size and density. Abdomen with sides punctate and shagreened. Front tibiae with tarsal groove well defined, with three external teeth, of which one near the apex is separated from the others (one of which is very small) by a wide incurvature; middle and hind tibiae narrower and longer than the front ones, and with three or four small teeth. Length, 5.5 mm. Hab.—Queensland: Cairns (H. Allen). Although by Leconte and Horn’s table this species would be referred to Platysoma, it certainly does not appear at first glance to belong to that genus, and probably a subgenus will be proposed for it, as was done for P. extrarium, an equally aberrant species. The type is probably a female; a second specimen (probably a male) agrees perfectly with it on the upper surface, but its prosternum has the striae conjoined at the base, and well defined to the base of the chin- piece (this unfortunately is broken in front), the median line of the metasternum is deeper, and is within a large, shallow, elliptic depression; its head was detached from the prothorax and is seen to have a deep notch on each side of the base on the upper surface, and a large round fovea on each side of the under surface, these being normally concealed. BY A. M. LEA. 415 NITIDULIDAE. CIRCOPES PICTUS, N. sp. Reddish-brown, parts of upper surface and of pygidium black, prosternum, legs and antennae paler than metasternum and abdomen. Densely clothed with depressed pubescence, on part of elytra black, on rest of upper surface white and stramineous or ochreous, on under surface and legs whitish. Head wide; labrum thin and shining. Antennae short, club briefly ovate. Prothorax rather strongly and evenly convex, greatest width (near the basal third) about thrice the median length, sides strongly rounded, hind angles embracing shoulders, apex less than half the width of base, emarginate for reception of head to about middle of eyes. EHlytra at base narrower than base of prothorax, sides obliquely narrowed to apex, not covering pygidium. Inter- coxal process of prosternum subtriangularly continued beyond coxae. Legs short and stout. Length, 2.25 mm. Hab.—Queensland: Cairns (A. M. Lea). Type (unique), I. 12044. Readily distinguished from all previously named Australian species by its beautiful clothing; on the upper surface this is so dense that the derm is normally concealed, but where it has been abraded or disarranged rather dense minute punctures may be seen. The derm of the prothorax has a large medio-basal dark blotch, the outlines of which are not traceable on the type; the greater portion of the elytra is black, the parts with pale pubescence being evidently paler than the other parts. On the elytra there is a large spot of ochreous pubescence, variegated with whitish, on the suture from the base to near the middle, and there are thin lines, spots, or individual white hairs on other parts of the elytra; the pygidium has black pubescence on the middle, whitish on the sides. BRACHYPEPLUS XANTHORRHOEAE, Nl. SD. Black, or blackish-piceous, elytra obscurely paler, muzzle, antennae, palpi and legs reddish. Upper surface (except of abdomen, which is minutely pubescent) glabrous. Head subtriangular, a shallow depression on each side in front; punctures dense and sharply defined but rather small, becoming sparser and smaller in front. Prothorax flat except at sides, apex shallowly emarginate and slightly narrower than base, front angles rounded, hind ones rectangular; punctures as on head. Scutellum large, with small punctures. Elytra scarcely longer than wide; with series of punctures in narrow striae, interstices each with a distinct row of small punctures, but the sutural interstice with two rows of smaller ones towards base. Abdomen with punctures much as on head. Length, 3.25-4.25 mm. Hab.—South Australia: Kangaroo Island (J. G. O. Tepper); New South Wales: Sydney (A. M. Lea). A highly polished species, structurally fairly close to B. basalis, B. Koebelei, and several others of the genus, but upper surface almost entirely dark; from B. planus it is distinguished by its high polish and very different elytral sculpture. Numerous specimens were taken from closely compacted young leaves of species of Xanthorrhoea. BRACHYPEPLUS INSIGNIS, 0. Sp. Reddish-castaneous, apical half of elytra black. Clothed with very short, yellowish pubescence, longer on tip of abdomen than elsewhere; each side with a fringe of very short pubescence. I 416 NEW SPECIES OF AUSTRALIAN COLEOPTERA, XViil, Upper surface with dense and fine punctures, causing the derm to appear shagreened. Head subtriangular, a shallow depression each side in front, with a few very minute granules. Antennae short, club almost circular. Prothorax gently concave in middle, base not twice as wide as the median length and finely margined, sides evenly rounded in front, feebly incurved near base, apex rather deeply emarginate; with small setiferous granules. Elytra slightly longer than wide; with fine, scarcely punctate striae, the odd interstices each with a row of minute setiferous granules. Abdomen with three segments exposed dorsally, with small setiferous granules, becoming numerous on sides of two first exposed segments; under surface with dense and sharply defined punctures, first segment along middle about once and one-half the length of second, the two combined slightly longer than third, the length of fourth and distinctly shorter than fifth. Mesosternum with somewhat stronger punctures than on abdomen. Prosternum with irregular punctures, middle of apex transversely strigose. Length, 5.5-6.75 mm. Hab.—Western Australia: Cue (H. W. Brown). Larger than any species before me, except B. auritus, and very distinct by its concave prothorax and colours. In some respects it appears allied to B. olliffi, but it differs from the description of that species in having all parts of the upper surface granulate, the granules on the elytra being quite distinct on the alternate interstices, and on the head they are smaller and sparser than elsewhere; the abdomen has only three segments exposed dorsally, and the size is smaller. The fringe on each side is composed of very short hairs, and could be easily over- looked, it is quite even from apex of prothorax to apex of elytra, but on the abdomen is less regular. BYRRHIDAE. ByYRRHINUS CONVEXUS Blackb. (formerly Notiocyphon). As three cotypes of Notiocyphon convexum Blackb. appeared to belong to the Byrrhidae, although the genus (as a new one) was referred to the Dascillidae, I asked Mr. Arrow to examine the type (now in the British Museum). In reply he stated that it “Belongs to genus Byrrhinus, and is near but distinct from (Trinodes) punctipennis Macl.” This coincides with my own opinion. CLERIDAE. TENERUS ABBREVIATUS White. T. ruficollis Macl. I cannot regard the type of TJ. ruficollis other than as a variety of T. abbreviatus; on a series of specimens the large pale apical patch with a con- spicuous purplish spot gradually changes, till only a vague space near the apex is pale, the dark spot being quite indistinguishable from its surroundings. TENEBRIONIDAE. SARAGUS STRIGIVENTRIS Lea. This species was described in a paper (Trans. Roy. Soc. S. Aust., 1915, 795) dealing with the beetles brought back by Captain S. A. White from an expedition into the interior of South Australia, and recorded from the Everard Range; but the only specimen from there was in somewhat damaged condition (its legs and antennae were missing) and so the type was made from a specimen from Eyre’s Sand Patch (although this was perhaps not made sufficiently clear in the descrip- BY A. M. LEA. 417 tion). Subsequently Carter stated (Proc. Linn. Soc. N. S. Wates, 1917, 717) that the name was synonymous with S. sphaeroides. He has recently kindly allowed me to see the type of that species and I can say with confidence that the two species are distinct. Comparing the types side by side, the following differences may be noted: S. sphaeroides Cart. Subopaque. Head finely shagreened and with moderately large punctures; each side conspicuously inecurved to the side of the clypeal suture. S. strigiventris Lea. Shining. Head not shagreened and with some- what sparser and distinctly smaller punctures; each side not at all, or very feebly incurved to the side of the clypeal Prothorax densely punctate* and finely suture. shagreened. Prothorax sparsely and finely punctate, Elytra more coarsely shagreened than and not at all shagreened. prothorax; punctures (owing to shag- Elytra not shagreened; punctures reening) not very sharply defined; each larger and sharply defined, even with three interstices slightly but dis- posteriorly ; without raised interstices. tinctly elevated above their fellows. Larger spine smaller and acute. Apex of front tibia with two spines, the larger of which is blunt-tipped, and almost as long as basal joint of tarsus. In addition to the type of strigiventris there are seven other specimens from Eyre’s Sand Patch now before me (they were overlooked when the species was named), and these all agree with it, and they all differ from the type of sphaeroides in the details noted. The Everard Range specimen is certainly closer in general appearance to the type of sphaeroides, and it has faintly elevated elytral interstices, but it differs from the type of that species in having the prothoracic punctures much smaller and sparser (very conspicuously so on the margins), elytra not really shagreened and with more sharply defined, although not larger punctures. On sphaeroides the elytral shagreening is distinctly coarser than on the prothorax, and on close examination the surface appears to be very densely granulate-punctate. The Everard Range specimen was picked up dead, and its opaque appearance is evidently due to weathering and not to shagreening. On passing through Adelaide recently Mr. Carter saw these notes, and commented upon them as follows:— “With a long series of examples before me, shown by Mr. Lea, from Eyre’s Sand Patch, S.A., I am convinced of the clear distinction of Saragus strigiventris Lea from S. sphaeroides Cart. from Condon, W.A.”’ MELANDRYIDAE. ORCHESIA PICTIPENNIS Lea. A specimen from Victoria appears to belong to this species, but differs from the types in having the prothorax of the same shade of castaneous as most of the elytra; these have very ill-defined markings, which consist on each side of a vaguely infuscate spot near the base, an obscurely flavous spot near the middle, then an infuscate spot, and then another obscurely flavous spot. * Carter, Proc. Linn. Soc. N. S. Waues, 1911, 197—‘‘Disc of pronotum coarsely punctate.” 418 NEW SPECIES OF AUSTRALIAN COLEOPTERA, XViii, ORCHESIA ALPHABETICA, Nl. SD. Piceous-brown, sometimes almost black; parts of prothorax, of under surface and of legs usually paler, elytra with sharply defined whitish markings, the suture narrowly eastaneous. Clothed with very short, ashen pubescence. Head evenly convex and with small dense punctures. Antennae rather long and thin, with a loose club. Prothorax widely transverse, base much wider than apex, with three depressions at base, the sublateral ones feeble, the median one still more feeble; punctures small, dense, and inconspicuous. Elytra long and thin; widest just before middle, with a feeble stria on each side of suture and a more sharply defined one near each side; punctures as on prothorax. Hind spurs slightly longer than their supporting tibiae. Length, 3.5-4.5 mm. Hab.—Tasmania: Cradle Mountain and Waratah (H. J. Carter and A. M. Lea). An elongate fusiform species, rather longer and thinner than O. eucalypti, and with more numerous elytral markings (they are much less numerous than in O. bryophila); structurally the species is nearest to O. medioflava. Specimens were beaten in abundance from a shrub at Waratah. The elytral markings are in three series; on each elytron they consist of: A, a spot at the basal fourth near the suture, the spot sometimes round, but often shaped like a reversed U or V; B, a series of narrow and usually conjoined markings on the derm near the middle, and commencing (sometimes as an isolated round spot) near the suture; and, C, narrow postmedian markings, commencing as a line running parallel with the suture for a short distance, then directed obliquely forwards, then backwards, and then outwards to the margin at about the apical third. On many specimens the antemedian markings resemble an irregularly placed M, and on many the postmedian ones an irregular W, but the latter are often enlarged, so that the posterior part is directed obliquely from near the suture to the side, but with two or three projections in front; occasionally the subbasal spot is narrowly connected along the suture with the antemedian markings. On one specimen the subbasal and antemedian markings are conjoined to form a large, irregularly angular blotch on each elytron. On some specimens the prothorax is of a dingy flavous or testaceous, with two more or less distinctly infuscated spots. TASMOSALPINGUS QUADRISPILOTUS Lea. A specimen from the Dividing Range (Blackburn’s collection) differs from the type of this species in being slightly larger and somewhat paler, with the elytra mostly flavous, but leaving the lateral and apical margins infuscated, the ‘suture narrowly infuscated, with, at the middle, the infuscation angularly dilated and connected with a large elongate spot on each elytron; these spots, however, not touching the margins as on one of the Tasmanian specimens; the seriate arrangement of punctures is less noticeable than on the type, but is fairly distinct towards the suture. TRICHOSALPINGUS MAJOR, DN. SP. Pale flavo-castaneous, legs and antennae somewhat paler. Clothed with very short, whitish pubescence. Head wide and feebly convex between eyes, with two shallow depressions in front; punctures small and crowded. Antennae thin, scarcely passing base of prothorax. Prothorax rather strongly transverse, sides moderately dilated near apex, a subangular depression on each side of base; punctures crowded but more sharply defined than on head. Elytra much wider than prothorax, slightly BY A. M. LEA. 419 dilated to beyond the middle; punctures more sharply defined than on prothorax, and rather less crowded; with feeble striation. Length, 5-5.75 mm. Hab.—South Australia: Mount Lofty Range (R. J. Burton). Larger than any previously named species and uniformly pale. From some directions parts of the elytra appear to be non-striated, and the striae are nowhere sharply defined and with larger special punctures. MORDELLIDAE. MORDELLA FELIX Waterh. Six specimens, from Tasmania (Launceston, Mole Creek and West Tamar), probably represent another variety of this species; they differ from the typical form in being smaller, 3-4 mm., with a large patch of pale pubescence on each shoulder, in addition to the other basal markings, the median V on each elytron longer and narrower, and the subapical fascia conspicuously concave on its inner edge. MorpELLA HUMERALIS Waterh., var. EXRUFA, Nl. Var. Five specimens from Tasmania (Waratah) have elytral clothing as on normal specimens, but the derm beneath the humeral markings is quite as dark as the rest of the elytra. On a few other Tasmanian specimens the red is scarcely perceptibly indicated. MORDELLISTENA RHIZOPHORAE, Nl. SD. Flavous; a slightly infuscated circle at basal third of elytra, tips of hind tibiae and of hind tarsal joints black. With moderately dense, pale pubescence. Thin and subparallel-sided. Scutellum,small and transverse. Pygidium moderately long and acute. Hind tibiae with three oblique black ridges, the first long, the third very short, spurs very unequal; basal joint of tarsi with two long black ridges, the second joint with one. Length, 3 mm. Hab.—Northern Territory: Darwin, reared in August from a mangrove gall (G. F. Hill). The slight but distinct circle on the elytra should readily distinguish the present from all previously described species; there is also a vague infuscation near the apex of each elytron. The front and middle legs and the antennae are missing from the type. TOMOXIA OBLIQUIALBA, DN. SD. Black, base of antennae and palpi obscurely reddish. Clothed with black and whitish pubescence. Rather wide. Scutellum with sides obliquely cutting into elytra. Pygidium rather short and thin, its tip truncated. Spurs to hind tibiae red and very unequal. Length, 5-5.25 mm. Hab.—Queensland: South Johnstone River (H. W. Brown). In some lights parts of the head and prothorax appear to be opalescent, and the dark pubescence to have a purplish gloss. The pale pubescence is dense on the head, margins the side and base of prothorax (from the latter there is a short spur on each side of the middle, as if, with the lateral pubescence, to mark off the three usual large dark spots), scutellum, three lines on elytra and on parts of under surface and legs. The species is structurally close to T. aterrima, but the pygidium is narrower and the elytral clothing different; on each of two 420 - NEW SPECIES OF AUSTRALIAN COLEOPTERA, XViii, specimens there are three distinct white lines on the elytra: a sutural one, and an oblique one from each shoulder to the suture near the apex, but before joining the suture the oblique lines become rather ill-defined. CURCULIONIDAE. MYLLOCERUS GRISEUS Lea. Two specimens of this species, from Wyndham, are in better preservation than the types; their elytral spots are more numerous, sharply defined and black, and the legs (except as to their clothing) are almost entirely black. MYLLOCERUS ARMIPECTUS Lea. Two specimens from Derby appear to belong to this species, but differ from the type in being slightly larger, and in having the clothing of the upper surface ef a pale greyish-blue, mottled with very pale brownish spots. MYLLOCERUS HILLI Lea. Numerous specimens of this species, from Darwin, have the scales more of an emerald-green than golden-green or golden; on several the median vitta of the pronotum is rather feeble, but the lateral ones are nearly always sharply defined. On many of them the elytral setae are longer than on the typical ones, so that they could fairly be regarded as belonging to C, of the 1914 table, where they would be associated with M. doddi and M. setistriatus; from the former they differ in being smaller, prothorax less convex in middle and trivittate and legs and antennae thinner; from the latter in the antennae being considerably longer and thinner, and front femora much less acutely dentate. MYLLOCERUS GNOPHOLOTUS, new name. M. griseus Lea, 1914, n. pr. The name griseus having been used by Roelofs for a species of Myllocerus, from Japan, in 1873, the above substitute is proposed for the Australian species also so named. MYLLOCERUS BIFASCIATIPENNIS, new name. M. fasciatus Blackb., 1889, n. pr. The name fasciatus having been used by Faust for a species of Myllocerus, from Suyfun, in 1887, the above substitute is proposed for the Australian species also so named. MYLLOCERUS SCITULUS, Nn. SD. Piceous-brown or reddish-brown, legs paler. Densely clothed with pale green scales, becoming still paler on head, under surface and legs; prothorax feebly trivirgate, elytra still more feebly maculate; with depressed or subdepressed setae, on the elytra confined to a single row on each interstice. Head rather wide, with a small median impression. Eyes rather distant and less convex than usual. Rostrum about as long as its basal width, sides almost parallel, gently concave along middle, and apparently without a median carina. Antennae moderately long and thin, bases rather distant; scape moderately curved; first joint of funicle not much, but distinctly, longer than second. Pro- thorax very little wider than the median length, apex very feebly incurved to BY A. M. LEA. 421 middle and scarcely narrower than base, the latter moderately bisinuate, sides feebly rounded; punctures concealed, but their positions indicated by setae. Elytra much wider than prothorax, parallel-sided to near apex; with rows of punctures in regular striae. Femora feebly dentate, tibiae apparently non-denticulate. Length, 4 mm. Hab.—Queensland: Dalby (Mrs. F. H. Hobler). As the prothorax is distinctly (although not deeply) trivirgate, this species could be referred to E. dd, of the 1914 table of the genus; from the species placed there (M. hilli and M. sulcicornis) it differs in being considerably smaller, eyes smaller and less prominent, antennae shorter and elytral setae pressed flat amongst the scales; M. anoplus, also from Dalby, is about the same length, but narrower, prothorax longer and not trivirgate and antennae shorter; M. gratus (some specimens of which have the prothorax trivirgate) has very different eyes, muzzle, and antennae. The infuscations of the upper surface are not very deep, the elytral spots in particular being very ill-defined; seen directly from above the elytra appear to be without setae, but they are quite distinct from some directions. The elytral punctures appear to be very small and in narrow striae, but on abrasion they are seen to be large and in rather wide striae. MYLLOCERUS CHAUNODERUS, N. SDP. Black: or piceous-brown, legs more or less reddish. Densely clothed with whitish-grey scales, on the upper surface with numerous sooty-brown spots; in addition with short decumbent setae, on the elytra placed in a single row on each interstice; muzzle with some long, whitish bristles. Head wide. Hyes very prominent. Rostrum scarcely as long as the basal width, gently concave and with a fine median carina. Antennae rather widely separated; scape rather strongly curved, thickened towards apex, under surface slightly grooved at apex; first joint of funicle slightly longer than second. Pro- thorax strongly dilated to base, the sides there flange-like, wider than elytra, and fully thrice the median length; apex straight, punctures rather sparse and normally concealed. EHlytra parallel-sided to near apex; seriate punctures large and in rather deep striae, but appearing small and narrow through clothing. Femora scarcely visibly dentate. Length, 4.5-6 mm. Hab.—Northern Territory: Groote Hylandt (N. B. Tindale). The base of the prothorax is usually noticeably wider than the elytra, but on some specimens it is scarcely perceptibly so. In the 1914 table of the genus this species would be associated with M. laticollis, from which it is at once dis- tinguished by the under surface of head, this being unarmed in both sexes; M. latibasis has the apex of prothorax distinctly incurved to middle. The male is usually smaller than the female, with longer legs and antennae, and less convex abdomen, but there appear to be no external features by which the sex of a single specimen may be determined. The derm is everywhere normally con- cealed, it is usually black, but occasionally is dark brown or even reddish-brown, the reddish cclour of the legs, however, is always evident. On the upper surface the dark spots vary considerably in number and extent, on some specimens they cover most of the elytra and on others less than half; on many specimens they appear more as obscure mottlings than distinct spots; on the pronotum there is usually an obscure vitta towards each side, and occasionally a faint median one; on an occasional specimen some of the scales on the under parts of the head 422 NEW SPECIES OF AUSTRALIAN COLEOPTERA, XViii, have a slight bluish gloss. The setae, although not conspicuous, even from the sides, cause the surface in parts to appear speckled (flea-bitten). From most directions the femora appear to be unarmed. Numerous specimens were obtained on wattles, Acacia spp. MYLLOCERUS OBSCURUS, Nl. SD. Black, parts of legs sometimes obscurely reddish. Densely clothed with greyish-white scales, becoming almost white on under parts; elytra with obscure brownish spots or mottlings, the prothorax with a wide, median brownish vitta. With stout, depressed setae. Byes less convex than usual, rather large and oblong-elliptic. Rostrum slightly longer than wide, sides gently incurved, feebly concave and with a fine carina along middle. Antennae long, rather close together at base; scape moderately curved and thickened at apex; first joint of funicle not quite twice the length of second. Prothorax with base strongly bisinuate and much wider than apex, the latter shorter than the median length; punctures fairly large but normally con- cealed (although indicated by setae). Elytra not much, but distinctly, wider than base of prothorax; with large punctures in deep striae, but appearing small and thin through clothing. Femora rather feebly dentate, tibiae with small and acute, but normally almost concealed denticulations. Length, 7-8 mm. Hab.—Northern Territory: Roper River (N. B. Tindale). In the 1914 table of the genus, this species would be associated with M. fugitivus and M. subrostralis; from the former it is distinguished by its larger size, shorter rostrum, less prominent eyes, and median vitta; from the latter by the shorter rostrum, sparser setae on upper surface and median vitta. There are five specimens before me but of these only two have the clothing in good condition; on two specimens some of the elytral scales have a slight bluish gloss. The elytral setae, although depressed, are not quite flattened amongst the scales, so that they are fairly distinct from the sides. The tooth on each front femur is small, but appears to be rather acute when the scales have been removed. MYLLOCERUS TRICARINIROSTRIS, N. Sp. Black, parts of legs obscurely reddish. Densely clothed with scales, mostly black or brownish on the upper surface, whitish or greyish-white on the sides and under parts. Elytra with numerous semi-erect setae, varying in colour with the scales amongst which they are set. Eyes fairly large but not very convex, oblong-elliptic. Rostrum about as long as its basal width, sides slightly narrowed to apex; with three thin carinae, of which the median one terminates in a narrow inter-ocular fovea, the others almost at the base. Antennae thin and moderately long, fairly distant at base; scape for the genus rather slightly curved, first joint of funicle half as long again as second. Prothorax with base moderately bisinuate and much wider than apex, the latter feebly incurved to middle, and distinctly wider than the median length. Elytra not much wider than base of prothorax, parallel-sided to near apex; with regular rows of apparently rather small punctures, in narrow striae. Femora very feebly dentate; tibiae not denticulate. Length, 6 mm. Hab.—South Australia: Kingoonya (A. M. Lea). The elytral setae are not quite as long as in the species referred to B d, of the 1914 table, but regarding the species as being placed there, it would be associated BY A. M. LEA. 423 with M. acutidens, which is a larger species, with paler scales, longer elytral setae, and base of prothorax narrower in proportion; regarding it as belonging to B s, it differs from the four species placed there, in its longer elytral setae; the obscure markings of its upper surface would associate it with M. castor, from which it differs in the prothorax wider at base, and with much smaller punctures, and elytra with longer setae. On the only specimen taken the elytral scales, as viewed from above, appear to be mostly black or blackish, with numerous small pale spots, but the scales on six interstices on each side are almost entirely pale; on a wide space towards each side of the pronotum the scales are almost entirely dark. The elytral setae, from the sides, appear to be dense and irregular, but, from directly in front or behind, they are seen to be placed in a single row on each interstice. The upper surface has not been abraded, but the punctures of the pronotum appear to be smaller and closer together than usual, the elytral punctures appear small through the clothing, and they probably are really smaller than on most species, as the elytral scales are unusually small. MYLLOCERUS HERBIVORUS, DN. SD: Black, parts of legs reddish, densely clothed with green, or greyish-green scales, in addition with numerous short semierect setae, on the elytra condensed to form a row on each interstice. Eyes prominent and almost round. Rostrum about as long as the basal width, sides feebly incurved, median carina feeble. Antennae long and thin, rather distant at base; scape rather strongly curved, first joint of funicle almost twice the length of second. Prothorax slightly wider than long, apex very feebly incurved to middle and about the width of base, the latter rather feebly bisinuate, sides moderately rounded, with dense and large punctures, normally concealed but indicated by setae. Elytra much wider than prothorax, shoulders rounded, sides almost parallel to near apex in male, slightly more dilated in female, with rows of large punctures in wide striae, but appearing small and narrow through clothing. Legs long, femora very feebly dentate. Length, 4-5 mm. Hab.—Northern Territory: Connexion Island (N. B. Tindale). On many specimens the prothorax has one or three faintly infuscated vittae, those with three could be referred to EH, of the 1914 table of the genus, where they would be associated with M. hilli, from which they differ in having considerably shorter rostrum, eyes smaller and more distant, and elytra with almost uniform scales. Those with the prothorax not trivirgate could be referred to F, or FF; the ones with the scales not green could be associated with M. foveiceps, from which they differ in the rostrum being longer and differently impressed, and the prothorax with sides more rounded, base narrower, and less deeply bisinuate; the ones with greenish scales have the sides of the prothorax less strongly rounded than in M. elegans and could be associated with M. mastersi, to which they are certainly close, but from which they differ in having the sides of the prothorax more strongly rounded, and the scape more evenly and less strongly arched. Some of the more brightly coloured specimens resemble M. darwini, but that species has the prothorax more strongly transverse. On many specimens the scales on the under surface and legs are almost white, on others they are as green as those on the elytra; probably on living specimens they are all green. The male is usually smaller than the female, has longer legs and antennae, less convex abdomen and larger punctures. Numerous specimens were taken in long grass by means of the sweep-net. 424 NEW SPECIES OF AUSTRALIAN COLEOPTERA, XViil, MYLLOCERUS DECIPIENS, N. Sp. Blackish, parts of legs reddish. Densely clothed with pale greenish-blue scales, on the upper surface with numerous sooty-brown spots, with more or less depressed setae on the elytra, confined to a single row on each interstice. Head with a narrow inter-ocular impression. Eyes moderately large and oblong-ovate. Rostrum about as long as the basal width, sides gently incurved. Antennae moderately long and thin, bases rather distant; scape moderately curved; first joint of funicle about one-third longer than second. Prothorax slightly narrower than the median length, apex straight and equal to base, which is strongly bisinuate, sides rounded, but arcuate near base; punctures normally concealed but indicated by setae. Elytra about one-third wider than widest part of prothorax, parallel-sided to near apex; with rows of punctures in regular striae. Femora feebly dentate; tibiae not visibly denticulate. Length, 4.5 mm. Hab.—Northern Territory: Groote Eylandt, unique (N. B. Tindale). In general appearance strikingly close to M. sulcicornis, but eyes larger and closer together, prothorax longer and without a median vitta, and under surface of scape not grooved. The scales are mostly of a pale bluish colour, but regarding them as partly green, the species, in the 1914 table of the genus, should be referred to H, and from each of the species placed there (M. gratus and M. anoplus) it is distinguished by its larger size, larger and somewhat different eyes, longer antennae and spotted elytra. Regarding the scales as not green, it could be associated with H. varius, which has very different eyes, etc. Towards each side of the pronotum there is an irregular dark vitta, on the elytra there are numerous small elongated spots, seldom crossing two interstices and absent from the sides. The median carina of the rostrum is distinct only at the apex, but probably extends to the base, although normally concealed. The elytral punctures appear to be narrow and in narrow striae, but on abrasion they will probably be found to be of considerable size. MYLLOCERUS AERUGINOSUS, Nn. Sp. Black or blackish, legs reddish. Densely clothed with golden-green, or bluish- green scales; in addition with numerous flavous setae, on the elytra fairly long and suberect. Head flatttened between eyes, and with a narrow median impression. Eyes moderately large and oblong-elliptic. Rostrum about as long as its basal width, sides incurved to middle, scarcely concave along middle; without a normally visible median carina. Antennae long and thin, bases distant; scape evenly curved; first joint of funicle about one-third longer than second. Prothorax distinctly transverse, apex slightly incurved to middle and the width of base, this rather strongly bisinuate; with vague transverse impressions near base and apex; punctures large and normally concealed, but indicated by setae. Elytra much wider than prothorax, parallel-sided to near apex; with large punctures in rather wide striae, but appearing small and in narrow striae on account of clothing. Legs rather long, femora slightly but acutely dentate, tibiae not denticulate. Length, 4.5-5 mm. Hab.—Northern Territory: Groote Eylandt (N. B. Tindale). The three specimens obtained on the island at first glance appear to belong to M. darwini, but they differ from some cotypes in being slightly narrower, the prothorax slightly longer, feebly incurved to middle of apex, and with feeble transverse impressions (Somewhat as on M. rugicollis but less pronounced) ; BY A. M. LEA. 425 from M. mastersi, to which they also appear to be very close, they differ in the prothorax being slightly shorter, and in the incurvature of apex and transverse impressions; the impressions, however, are less pronounced than on the species referred to F i, of the 1914 table, so the species should be referred to LL; M. carinatus (there associated with M. darwini), is a larger maculate species, with different elytral clothing, and M. usitatus (there associated with M. mastersi) has edentate femora. MYLLOCERUS PICTUS, Nn. Sp. Black, parts of legs obscurely reddish. Densely clothed with bright green scales, interspersed with erect, whitish setae, becoming rather long on elytra. Eyes rather large and prominent. Rostrum scarcely longer than the basal width, sides feebly incurved, moderately concave along middle, with median carina normally concealed. Antennae long and thin, bases distant; scape moderately curved; first joint of funicle almost twice the length of second. Pro- thorax moderately transverse, base and apex equal, the former straight, the latter moderately bisinuate; punctures normally concealed but indicated by setae. Elytra much wider than prothorax, feebly dilated to beyond the middle; with regular rows of rather large punctures, but appearing much smaller through clothing. Femora feebly dentate, tibiae not distinctly denticulate. Length, 4.5-5 mm. Hab.—North Western Australia: Wyndham, in February (J. Clark). In the 1914 table of the genus this species would be associated with M. doddi and M. setistriatus, from both of which it is distinguished by its smaller size, feeble dentition and uniformly green scales. Structurally it is fairly close to M. hilli, but the prothorax is not vittate, and the setae are longer; at first glance it resembles some specimens of M. mastersi, from which it is at once distinguished by the upright elytral setae. In general appearance it is rather close to M. aeruginosus, but the eyes are more convex, the basal joint of the funicle is longer, and the elytral setae are longer and more erect. MYLLOCERUS ARMIPES, Nn. SD. ¢. Piceous-brown, legs and antennae reddish. Densely clothed with pale slaty-grey scales, becoming almost white on under surface, elytra with numerous small brownish spots. Eyes rather large and oblong-elliptic. Rostrum slightly shorter than the basal width, sides feebly incurved, median carina normally visible towards apex. Antennae rather long and thin, bases distant; scape moderately curved; first joint of funicle about one-third longer than second. Prothorax moderately trans- verse, sides strongly rounded but close to base incurved, apex gently incurved to middle and, if anything, slightly wider than base, the latter almost straight; with a vague transverse impression near apex, and another near base; with large punctures, partly concealed, but their positions clearly indicated by setae. EHlytra much wider than base of prothorax, parallel-sided to near apex; with rather large punctures in wide striae, but appearing small and narrow through clothing. Basal segment of abdomen shallowly depressed in middle, the apical one rather flat. Femora very stout, strongly and acutely dentate; tibiae long with numerous small denticulations, but each of front ones with a rather large one in middle; front coxae armed. Length, 5.5-6 mm. Hab.—Queensland (Dr. E. W. Ferguson), Endeavour River (C. French). 426 NEW SPECIES OF AUSTRALIAN COLEOPTERA, XVili, Rather closely allied to M. rugicollis, but transverse impressions of pronotum less conspicuous and armature of tibiae very different. Regarding the prothorax as slightly wider at the apex than at the base, the species, in the 1914 table of the genus, would be associated with M. amblyrhinus, but that species has a wider and more convex head, with rounder and more prominent eyes, and very different armature of the legs; passing that species in the table, and also M. rugicollis, it would be associated with M. fuscomaculatus, from which it differs in having larger eyes, different prothorax and setae, and very different legs. The elytral setae are pressed flat amongst the scales, so that even from the sides they are scarcely traceable, a few apical ones only are fairly distinct. On the type each front coxa has a small, conical tooth, projecting downwards, and each middle coxa has a still smaller one; the larger tooth of its front tibia is almost as large as the femoral tooth; on a second male the coxal armature is much less distinct, and the larger tibial tooth is much smaller than the femoral one. BELUS MACULIPENNIS, Nn. Sp. Dark reddish-brown. Upper surface irregularly clothed with yellowish or stramineous pubescence, becoming white on under surface. Head rather convex, with crowded punctures. Rostrum long, thin, and slightly curved; with distinct punctures behind antennae, very small elsewhere. Antennae thin, extending beyond middle coxae. Prothorax slightly longer than wide, sides gently rounded and widest near base, with a moderately distinct median line; punctures as on head. Elytra long, thin and parallel-sided, except that the shoulders are rounded, and the apex triangular; with a faint longitudinal depression on apical half near suture and another half-way between suture and each side; with crowded punctures. Legs long; front femora moderately stout, and unarmed; front tibiae moderately curved, lower surface denticulate; front tarsi with basal joint large, about as long as second and third combined. Length (excluding rostrum), 7-8 mm. Hab.—Queensland: National Park (H. Hacker). Type in Queensland Museum. Structurally fairly close to B. filum and B. filiformis, but rostrum considerably longer and thinner, prothorax less dilated towards base, and clothing differently placed. Some parts are darker than others, but only the claws and eyes are black; the rostrum is sometimes paler than the rest of the head, but on three specimens has a dark line (inconspicuous from most directions) from the base to beyond the middle, the apical joint of the antennae is paler than the preceding ones. On the upper surface the clothing margins the eyes, forms three lines on prothorax (the median one narrower than the others), and has a feebly spotted appearance on the elytra, but on each of the six specimens taken by Mr. Hacker there is a large distinct spot on the middle of each elytron, and a streak near the apex, and by these markings it may be readily distinguished; on the metasternum and abdomen the pubescence is dense on the sides, and forms a feeble line on each side of the middle of abdomen, but these are sometimes obscured. From some directions both head and prothorax appear to be densely granulate. The narrow part of the elytra begins suddenly, and a line cutting through them at its commencement would cut off an equilaterally triangular apical portion. ORCHESTES AUSTRALIAE, Nl. SD. Flavo-castaneous; mesosternum, metasternum, and base of abdomen black. Moderately densely clothed with pale pubescence, and golden setae, becoming white on under surface and legs. BY A. M. LEA. 427 Head rather small, inter-ocular space narrow. EHyes large and prominent. Rostrum about the length of prothorax, slightly curved, shining and with small punctures. Prothorax with sides strongly rounded and increasing in width to base, this almost twice the width of apex; with dense, partially concealed punctures. Elytra oblong cordate, not covering pygidium; punctate-striate, inter- stices with rather large punctures. Hind femora each about as large as prothorax, finely serrated posteriorly. Length, 3 mm. Hab.—New South Wales: Orange, in January (H. J. Carter). On the two specimens taken by Mr. Carter, the upper surface appears slightly mottled, owing to the rather patchy distribution of the pubescence. This is the first true Orchestes to be recorded from Australia, as perpusillus, referred by Pascoe to the genus, has been transferred to Rhamphus (Lea, Proc. Roy. Soc. Vic., 26 (n.s.), 1914, p. 226); in the latter genus the claws are simple, in the former they are appendiculate or bifid. RHAMPHUS AUSTRALIS Blackb., and R. DISTINGUENDUS Blackb. In the original descriptions of these species no mention is made of any kind of clothing, and two cotypes of the former species in the South Australian Museum, and one of the latter, appear to have the upper surface glabrous, except that some minute setae may be traced on the posterior sides of the elytra. RHAMPHUS AMPLIPENNIS, 0. SD. Black, antennae (except club) and tarsi flavous, tip of rostrum obscurely reddish. Upper surface and parts of legs with very short, pale pubescence or setae, on the elytra in two or three almost regular rows ‘on each interstice. Head with crowded punctures; interocular space from above appearing as a narrow shining line. Rostrum flat, slightly curved, shining, with small punctures, its tip resting on mesosternum. Prothorax with sides strongly dilated from apex to near base where the width is more than twice the median length, with a shining, slightly elevated median line; punctures dense and rather coarse. Elytra dilated from base to beyond the middle, where the width is almost twice that of the prothorax; punctate-striate, punctures in the striae rather large but often obscured, interstices with sharply defined punctures about as large as those at base of head, but less crowded. Length, 2.5 mm. Hab.—South Australia (Macleay Museum), Mount Lofty Ranges (A. M. Lea). In several collections this species was standing as R. australis, but it may be at once distinguished by the elytral clothing, this consisting of almost uniform Tows (two or three on each interstice) of minute pale setae, in some lights causing the upper surface to appear greyish. Its elytra are wider in proportion than in any other species of the genus before me. The species is quite common on the “blackwood”, Acacia melanozylon, but its powerful hind femora frequently enable it to escape capture. RHAMPHUS MEGALOPS, DI. SD. Black; apex of rostrum and tarsi reddish, antennae paler, but club black. Clothed with fine whitish pubescence or setae, on the elytra forming two regular rows on each interstice. Head with crowded punctures behind the eyes. Eyes large, almost touching at base, the interocular space very narrow to between antennae. Rostrum some- what flattened, slightly curved, shining, with small punctures, its tip resting 428 NEW SPECIES OF AUSTRALIAN COLEOPTERA, XViii, on mesosternum. Prothorax with sides strongly rounded and widest at about basal third; punctures much as on base of head; median line faintly defined or absent. Elytra with sides dilated from base to about apical third, where the width is about one-third more than the greatest width of prothorax; punctate- striate, interstices with somewhat rugose punctures, about as large as those on prothorax. Length, 2.5 mm. Hab.—South Australia: Tarcoola and Ooldea (EH. H. Ising and A. M. Lea). In general appearance fairly close to the preceding species, and with similar clothing, except that on the elytra the pubescence never appears to be in more than two rows on any interstice; but less robust, and greatest width of prothorax distinctly before the base instead of almost at the extreme base; when the preceding species is viewed from behind, a point is soon reached where the out- lines of the prothorax and elytra appear to be continuous, and the basal angles of the former appear to be acute; from the same point the present species appears to be notched on each side, where the elytra and prothorax meet, and the basal angles of the latter to be rounded off. The eyes are even larger than on that species, and seem to touch at the base, instead of being slightly but distinctly separated there. Numerous specimens were obtained on some stunted wattles (Acacia sp.) growing between rocks at Tarcoola. RHAMPHUS MICROSCOPICUS, N. SD. Black, basal half of antennae and apical fifth of rostrum flavous or reddish. Upper surface with minute pubescence, seriate in arrangement on the elytra. Head with crowded punctures. Rostrum thin, shining, slightly curved, its tip resting on mesosternum. Prothorax with sides rather strongly dilated to near base, median length more than apical width, but conspicuously less than basal width; with dense and rather sharply defined punctures, but less crowded than on base of head. Elytra with sides dilated to beyond the middle, and sub- continuous with those of prothorax; striate-punctate, the interstices also punctate. Length, 0.9-1 mm. Hab.—Western Australia: Swan River, Bridgetown, Donnybrook (A. M. Lea). Decidedly smaller than normal specimens of R. acaciae, and consequently one of the smallest weevils in Australia, if not actually the smallest. A few Tasmanian specimens of R. acaciae are not much longer, but their elytra are decidedly narrower. COSSONIDEUS* PASCOEI Woll. Eleven specimens from Beverley (Western Australia) evidently belong to this species, which is distinct by its pale elytra with the sides (and to a less extent the suture) infuscated, by the large eyes and five-jointed funicle; the middle of the metasternum has also a large pale blotch. In dealing with the genus, as with others of the family, Wollaston used exaggerated expressions, the eyes were described as “excessively large and prominent”; they are certainly larger than is usual in the subfamily, but as their combined width is less than that of the interocular space (the full width of the base of the rostrum) “excessively” is misleading; the legs also are not “exceedingly elongate’, being in fact scarcely longer than in many species of Cossonus and Notiosomus. The male differs from the female in having the antennae inserted slightly nearer the base of rostrum, abdomen with a rather wide depression on basal segment, con- *In Wollaston’s monograph the name was spelt Cossonideus on pages 435, 517, and 653, but on page 448 Cossonidius; Scudder amended the name to Cossonoideus. BY A. M. LEA. , 429 tinued on to second and, less distinctly, on to metasternum (on the female it is smaller and confined to the basal segment); punctures especially on under surface, larger and closer together, and front femora more strongly curved on the outer side, and almost straight on the inner; the slight asperities of the femora, mentioned by Wollaston, appear to be remnants of femoral dentition, and are confined to the male. CHRYSOMELIDAE. GELOPTERA PARVONITENS, Nl. SD. o. Metallic green, under surface black, in parts with a metallic gloss, labrum, antennae and legs flavous. Head with dense but sharply defined punctures, becoming rather elongated towards base. Antennae long and thin, second joint shorter than third but decidedly wider. Prothorax about twice as wide as long, sides evenly rounded in middle, angles rather acutely armed; punctures rather dense and deep, slightly larger than on head, but not quite as dense. Elytra at base about one-fourth wider than prothorax; basal half with somewhat similar punctures (except the shoulders which are smooth and impunctate), but on apical slope mostly confined to distinct striae. Abdomen with fourth segment distinctly longer than third, and about twice the length of fifth. Femora unarmed. Length, 2.5-2.75 mm. ®. Differs in being slightly larger, upper surface more or less bronzy, antennae slightly shorter, elytra more dilated, abdomen larger and more convex, with the fourth segment smaller, and legs shorter, with the basal joint of front tarsi smaller. Hab.—Queensland: National Park (H. Hacker). Type in Queensland Museum. I refer this species to Geloptera as it is certainly close to G. microcalla, from which it differs in being smaller and the prothorax with sides quite evenly rounded in middle; it might, however, almost as well have been referred to the glabrous section of Hdusa, and to the vicinity of EH. glabra and E. chlorophana, from which it differs in being smaller, with consistently larger punctures. On the male only the extreme tip of antennae is infuscated, on one female the whole apical joint and parts of the four preceding ones are infuscated, on two others the infuscation is less pronounced. On one female the upper surface is of a beautiful golden-green, and the punctures near the base of its head are smaller and sparser than on other females. ENDOMYCHIDAE. - DAULOTYPUS MINOR, 0. Sp. Reddish-flavous, tarsi and palpi paler (almost white), head, prothorax and antennae (tips excepted) deeply infuscated. Moderately clothed with pale, sub- depressed pubescence, in addition with suberect setae. Head with irregularly distributed punctures. Clypeus with a depression on each side. Antennae rather long and thin, first joint moderately stout, slightly longer than third and about twice the length of second, ninth slightly longer than eighth but distinctly wider at apex, tenth slightly longer and distinctly wider than ninth, eleventh almost as long as ninth and tenth combined, its tip obtusely pointed. Prothorax more than twice as wide as long, sides finely Margined and somewhat uneven, widest near apex; with a narrow, deep, longitudinal impression on each side of base extending to about the middle; a narrow deep impression close to base; punctures sparse and irregular. Elytra much wider than prothorax, widest at about basal third; with somewhat irregular 430 NEW SPECIES OF AUSTRALIAN COLEOPTERA, XVili, rows of rather large punctures, becoming smaller and more confused posteriorly. Abdomen with basal segment, in middle, as long as three following combined; sixth distinct at sides, strongly incurved to and not traceable in middle. Legs rather long and thin. Length, 3-3.25 mm. Hab.—New South Wales: Forest Reefs (A. M. Lea). Differs from D. picticornis in being considerably smaller, prothorax more convex in middle, with the latero-basal impressions and sides different, basal segment of abdomen larger, femora thinner, labrum somewhat smaller, less conspicuously bilobed in front and antennae uniformly coloured, except for their pale tips. The sides of the prothorax are slightly dentate at the apical third, and again close to the apex, each slight tooth being rendered more distinct by a seta. COCCINELLIDAE. RHIZOBIUS DISCIPENNIS Blackb. A small, strongly convex, variable species, common in the Cairns district of Queensland; it is readily distinguished from most species of the genus by a conspicuous subsutural row of strong punctures on each elytron, marking the outer edge of a polished space on which the punctures are much smaller than elsewhere. Form 1.—Elytra with a more or less conspicuous metallic-green gloss, pro- thorax black. Several cotypes belong to this form. Form 2.—Like the preceding, but elytra with a violet gloss. Form 3.—Elytra black, or at least very dark brown, front and sides of prothorax obscurely diluted with red. In addition to many specimens of this form from Cairns, Cooktown, etc., there are two in the South Australian Museum from Melville Island. Form 4.—EHlytra blackish, with a conspicuous reddish vitta on each, the two conjoined near apex; prothorax more or less obscurely reddish. There are many specimens intermediate between this form and the preceding one, as the vittae are traceable, although not sharply defined. Form 5.—Upper surface of a uniform pale castaneous-brown, or at most the suture and sides slightly infuscated. Several cotypes belong to this form. RHIZOBIUS XANTHURUS Muls. R. tricolor Lea. The elytra of this species are of uniform colour throughout, although varying in individuals from a rather bright blue to deep purple. It occurs in South Australia, Victoria and Tasmania. RHIZOBIUS HIRTELLUS Crotch. R. ruficollis Blackb. This species occurs in Queensland, New South Wales, Victoria, Tasmania, South Australia and Western Australia. In Blackburn’s table of the genus (Trans. Roy. Soc. S. Aust., 1892, 258) R. hirtellus and R. ruficollis are bracketed together as having ‘‘Prothorax entirely: bright rufous, in strong contrast to the elytra,’ but on the preceding page he stated that he considered them to be distinct. Standing in his collection* was a single specimen from Queensland *In his note-book, under No, 5220, recorded also:as hirtellus. BY A. M. LEA. 431 labelled as hirtellus (and I have taken eight of the same species on Mount Tambourine), but this species has whitish hairs only on the elytra, decidedly long ones mingled with others of moderate length} whereas Crotch says ‘‘pubesc- ence suberect, ashy, intermingled with long black hairs.” The prothorax is occasionally uniformly red, but usually it has an infuscate medio-basal blotch (this blotch is faint but traceable on a cotype of ruficollis); on a few specimens the blotch is blackish and sharply defined. One South Australian specimen has the elytra partly purple, and many have the longer hairs dark brown but mingled with paler ones. RHIZOBIUS CARNIFEX Muls. R. calomeloides Lea. This species, which occurs in South Australia, Victoria, and Tasmania, varies considerably in size (4.5-6.5 mm.) and the dark prothoracic markings are more sharply limited on some specimens than on others; two, from Victoria, have the prothorax entirely red and the suture conspicuously red throughout. The speci- mens from Tasmania named R. calomeloides, are smaller and with less sharply defined markings than usual. RHIZOBIUS STRAGULATUS Er. R. blackburni Lea. This species was originally referred to Scymnus by Erichson; it was unknown to Mulsant; Crotch referred it to Midus, but queried it as possibly a Rhizobius; in Masters’s Catalogue it appears as a Pharus. Blackburn in 1892 commented on the species as unknown to him; in 1901 he commented on some of Erichson’s types sent to him by Prof. Kolbe, but without mentioning (either then or subse- quently) any Coccinellidae. In his collection, however, were two specimens marked “Lithophilus stragulatus Hr. compared with type’; and these specimens agree with the original description, and with the type of blackburni. I think I was right in referring the species to Rhizobius, its coarsely granulated eyes, abdomen, antennae and legs are all in agreement with many species of that genus, the form is certainly more depressed than usual, but this alone could not be considered as of generic importance. RHIZOBIUS PYGMAEUS Blackb. (formerly Midus). This species is certainly congeneric with the preceding one, and hence should also be referred to Rhizobius; it bears a strong resemblance to Scymnus corticalis Lea, but that species has the face more vertical, eyes much more finely faceted, prothorax much more transverse, under surface with much smaller punctures and abdomen different. CHILOCORUS BAILEYI Blackb. The type of this species had the elytra entirely dark, and belongs to a rare form; others in the Blackburn collection, from Cairns, and the majority I have taken there myself, have each shoulder flavyous, occasionally each flavous part is small and vaguely indicated, on others it is very conspicuous and occupies more than half the base of each elytron; on such specimens the prothorax is usually entirely flavous, or at most with a basal infuscation. pn eS er OU! EE ee ee ee Oe ee oe jy These specimens probably represent a variety of hirtellus. J THE SALINITY OF INSHORE OCHANIC WATERS OF AUSTRALASIA IN RELATION TO FISHES. By W. J. Puiciipes, F.L.S., Dominion Museum, N.Z., and F. J. T. Grice, M.Se., Dominion Laboratory, N.Z. (Communicated by Professor H. G. Chapman). [Read 25th November, 1925.] Introduction. According to the Challenger Handbook (Science of the Sea, 1912), seawater contains from 3.5% to 4% of salts, but if the percentage of salt falls considerably, true marine algae are scarce. Marine algae have an optimum, maximum and minimum limit with regard to salinity as well as with regard to light and temperature. The object of this paper is to introduce the study of the salinity of comparative inshore areas of Australasian waters to shed light on the distribution of plankton and on fish life. We believe that too many investigators in the past have neglected to balance the whole of the influences at work in the destruction of plankton, and too many theories with too little observation behind them have been made on the influence of minor currents and bacterial action in temperate zones of both hemispheres. The collection of a large number of samples on a given day at tide specified has enabled us to prepare a paper along different lines from other observers chiefly owing to the fact that samples have been simultaneously collected over so large an area of ocean. Analyses made by Dittmar show that the relative proportions of the minor constituent salts of seawater to the major sodium chloride content are remarkably constant, and it was therefore considered that complete analysis of the minor salts would serve no useful purpose. We therefore confined our attention to accurate determinations of chlorine present in chlorides, which may be taken as a satis- factory index of total salinity. The chlorine determinations were made volumet- rically by titration with silver nitrate, which had been carefully standardized against both British Drug House (Analytical Reagent) and Merck’s (Guaranteed Reagent) sodium chloride. An analysis was, however, made of a typical New Zealand seawater and was found to correspond closely to the mean analysis of Dittmar (both calculated to the same chlorine content). We have preferred to express our results in terms of percentage of chlorine (as chlorides) rather than as total solids calculated from these values by an arbitrary factor after the manner of Dittmar (in Challenger Reports) and others. In connection with assistance rendered in the collection of samples, we desire to make special mention of the following, to each of whom we are indebted for assistance:—Mr. E. B. Harkness, State Fisheries Dept., Sydney; Mr. F. Lewis, Chief Inspector of Fisheries, Melbourne; Mr. Bruce, Chief Inspector of Fisheries, Adelaide; Mr. D. Carter, N. Z. Union S. S. Co., Auckland; Capt. Hammond, Northern S. S. Co.; Mr. S. Reid, Portland Cement Co.; Miss A. I. Mackay, Urenui School; and Mr. E. D. Crawford, Tararu. BY W. J. PHILLIPPS AND F. J. T. GRIGG. 433 Salinity Determinations of Australian Seawater. Chlorine No. Date. Locality. | Collected by. per cent. fa | il 23rd Oct., 1923 Fremantle ..... | The Harbour Master 2.094 2 12th Dec., 1923 Fremantle ..... | The Harbour Master 2.094 3 6th Sept., 1923 Port Adelaide .. | Mr. Bruce .....:.. 2.112 4 6th Sept., 1923 Port Lincoln ... Mr. W. D. Randall 2.112 5 6th Sept., 1923 eters LAGAN ba ooe Mr. A. Frinsdorf .. 2.467 6 6th Sept., 1923 South Channel, Port Phillip .. Mr. W. J. McCulloch 2.041 ia 6th Sept., 1923 Westernport Bay | Mr. G. J. Ward 2.077 8 6th Sept., 1923 Bass Strait .... Wh, @e dip GOK Goose ZnO 9 6th Sept., 1923 Port Phillip Bay | Mr, UWewis -....... 2.059 10 6th Sept., 1923 1P@ME INAMAY ooo6c Mr. J. Northey .... 2.077 note 5th Sept., 1923 North of Cape | S.S. Maianbar (per Hawke) s.5 4-4 Mr. EB. B. Harkness) 2.094 Ue 5th Sept., 1923 East of Cape | S.S. Maianbar (per RBENWAE S oops Mr. E. B. Harkness) 2.094 13 5th Sept., 1923 South of Cape | 8.S. Maianbar (per Hawke sa. Mr. E. B. Harkness) 2.094 14 5th Sept., 1923 Off Camden Haven | S.S. Yulgilbar (per Mr. EB. B. Harkness) 2.094 15 5th Sept., 1923 Spitaie, POS Luo . S.S. Pulganbar (per E. Long. 153° 21%’ Mr. E. B. Harkness) 2.094 16 5th Sept., 1923 Sh dbeye, POO Bey. - S.S. Pulganbar (per HY Longe. 153° 237 Mr. EF. B. Harkness) 2.094 17 5th Sept., 1923 Korogoro Point | S.S. Wollongbar (per Mr. E. B. Harkness) 2.094 18 5th Sept., 1923 Korogoro Point | S.S. Wollongbar (per Mr. E. B. Harkness) 2.094 19 5th Sept., 1923 East of Tacking | S.S. Wollongbar (per Points Je-cescne Mr. E. B. Harkness) 2.094 20 5th Sept., 1923 SBilvatesi2c5 4a (joe be, |, Th BE. Long 151° 55’ Harkness) ...... 2.077 21 5th Sept., 1923 North of Smoky |.J. Anderson (per Capel naciscroce Mr. EB. B. Harkness) 2.077 22 12th Dec., 1923 Rockhampton .. Wie, IN, IDEN coon 2.147 Australian samples.—The average chlorine percentage in samples of water collected on the Australian coasts was 2.102. This estimate includes the abnormal sample secured at Port Pirie. Taking the Australian States separately, we find average salinities to give results very similar in most cases. Western Australia, with 2.094 on both the 23rd October and the 12th December, has apparently a fairly constant salinity at Port Fremantle during the early summer months. As might perhaps have been expected, samples from Spencer and St. Vincent Gulfs showed the highest recorded salinity. Evidently Kangaroo Island, off Point Victoria, effectually prevents the influence of the volume of water brought down by the river Murray lowering the salinity of Investigator Strait and the Great Australian Bight. The average of South Australian samples was 2.230% chlorine. It is interesting to note that Westernport Bay, Bass Strait, and Port Fairy had each 2.077% chlorine, which amount is also that of the highest recorded sample from New Zealand, taken by Captain Aspden in the Bay of Plenty. Tasman Sea.—In order to ascertain something of the nature of the water between New Zealand and Australia, we considered it advisable to have samples 434 SALINITY OF INSHORE OCEANIC WATERS OF AUSTRALASIA, collected for us at regular intervals across the Tasman Sea by the mail steamer between Wellington and Sydney. On being approached, the Chief Engineer on the S.S. Ulimaroa kindly consented to do this for us, and some ten samples were taken at regular intervals of from nine to ten hours, the first being collected nine hours after leaving Wellington Harbour and the last in Sydney Harbour. By a curious coincidence these two samples of water had the same salinity. It would appear at this season that, some three hundred miles off the New Zealand coast, the salinity rises slightly and remains higher than usual around New Zealand until just before the Australian coast when a further rise is noticeable. The total distance from Wellington to Sydney is 1,240 miles, and, from our graph below, we may determine that the influence of tropical waters in the Tasman Sea was at this season strongest on the eastern coast-line of Australia and also several hundred miles west of New Zealand. There is, of course, the possibility that the higher Intervals of 9 hours each. Graph to indicate changing salinity between Wellington, N.Z., and Sydney, August, 1928. Salinity Determinations: Tasman Sea, Wellington to Sydney. Chlorine No. Date. Locality. per cent. 1 24th Aug., 1923 ALO TL" Stee, BO ZO? ID, Ibonaes 2.023 2 25th Aug., 1923 39° 30’ S. Lat. 170° 20’ EK. Long 2.059 3 25th Aug., 1923 38° 50’ S. Lat. 167° 33’ BH. Long 2.059 4 25th Aug., 1923 38° 02’ S. Lat. 164° 47’ EH. Long Pye itale4 5 26th Aug., 1923 37° 122 S! Dat. 162° 08% ©. Long 2.094 6 26th Aug., 1923 36° S14 ISs Wat liao 334 Hs wong, 2.077 7 27th Aug., 1923 35° 45’ S. Lat. 156° 59’ EH. Long 2.077 8 27th Aug., 1923 35° 02’ S. Lat. 154° 28’ KH. Long 2.077 9 27th Aug., 1923 34° 10’ S. Lat. 152° 21’ H. Long 2.094 10 28th Aug., 1923 Sydney harbounme timer ercicreien 2.023 BY W. J. PHILLIPPS AND F. J. T. GRIGG. 435 salinity on the Australian coast-line may be due to the vicinity of the great land mass causing greater evaporation; but this does not account for the rise to the west of New Zealand. Further evidence of constancy in the chlorine constituent over short periods is afforded by comparison of the results from the Tasman Sea with those of New South Wales coasts. The sample taken just off Sydney Heads (No. 9) agreed exactly with other samples collected off various points of the New South Wales coast-line over a week later. New Zealand Samples.—A study of the comparative salinities of water samples taken on our coasts appears to prove that, given equable weather conditions, there is no difference in general salinity at this season between North Auckland and Stewart Island. Save for the influence of the Taieri and Clutha Rivers we are at a loss to account for the salinity on the east coast of Otago, which at Bluff and St. Clair was found to be lower than might have been expected, while at Port Chalmers and Warrington the average rose distinctly. Though taken some distance from the river mouth, the Oamaru sample showed distinctly the influence of the Waitaki River. There is a possibility that throughout the year the Canterbury Bight and the east coast of Otago have a lower average inshore salinity than most other parts of New Zealand. The rise in salinity at Stewart Island would appear to show that no influence of large bodies of fresh water ice melting at the Antarctic affected the most southern limits of the New Zealand coast-lines at this time. In general, salinity around New Zealand is distinctly lower than found in the case of Australian samples. Salinity in relation to Plankton.—We have found that marine diatoms cease movement in a few minutes when transferred to water of 1.5% salinity, and similarly that freshwater infusorians and diatoms cease movement in a like solution. In each case, as far as we could ascertain, the organisms ultimately died. As far as our examinations of river water in New Zealand have gone, living organic life has been small. The Clutha River in Otago is estimated to discharge into the sea 1,000,000 cubic feet of water per minute. The sea which it meets contains always quantities of diatoms which will ultimately die or are prevented from increasing by the influx of freshwater, while all organic material in the freshwater of the river is killed and finds a resting place in the bed of the ocean off the mouth of the river. Where a river discharges its contents into the sea, this process has been going on for ages. Eddies and minor currents carry the river water and its refuse along the coast in varying directions according to weather and tide conditions. Heavy rains must also be responsible for loss of considerable quantities of marine plankton. In the open seas, particularly in tropical zones where salinity and temperature are highest, denitrifying bacteria have been shown to cause enormous destruction of diatoms; but that denitrifying bacteria could increase to the same extent in areas of low salinity appears improbable. Latitude.—Latitude is an undoubted factor in dealing with comparative salinities of ocean areas, for latitude and trade winds determine approximately the amount of heat and hence the degree of evaporation likely to occur. As a possible solution for the high degree of salinity in water from South Australia, we may mention that there is a possibility that the great land areas adjoining, with warm winds from the interior, would cause an evaporation more intense than elsewhere in Australia or New Zealand. Furthermore, there is the possibility that water in the Spencer Gulf and probably in the Australian Bight is held in a kind of pocket outside the influence of minor currents and eddies. Actually Port 436 SALINITY OF INSHORE OCEANIC WATERS OF AUSTRALASIA, Salinity Determinations of Water Samples collected around New Zealand at Full Tide on 6th September, 1923. Chlorine No. | Locality. Collected by per cent. il | Mangonui (S.S. Clansman) ....... Sh Ve MVCSUOM abacoc 1.988 2 Whananaki (North Auckland) .... Wo INueENG) Soocuace 2.006 3 1 m. S. Tiri Tiri (S.S. Manaia) .. NG Labial Gog uo oc 6 1.935 4 Near Sail Rock (S8.S. Ronaki) .... (per S. Reid) ..... 2.041 5 LMG Ieyeeiere Tel godcanscscoac | Caretakers. eee 2.023 6 | Puhoi Wharf (S.S. Omana) ...... HW 185 INTOMNONIS) soénaooc 0.302 7 |5 mls. S.E. Slipper Isl. (SS. IN 2270 WD) bg ern wate nna test cu spare Tees AD, WEDGE 21/5 bo 2.059 g N.E. Kawau Isl. (S.S. Apanui) .... AS Wi PaliGe? ae acaecen 2.041 9 2 mls. Whangarei (S.S. Mahurangi) (per S. Reid) ..... 2.041 10 | Hauraki Gulf, 27 mls. Pt. Fitzroy : RM GSESmn Vaio talc) mew ny een ee Wh, RL Some .. ese 2.023 11 | Inner Harbour, Auckland (SS. Clawanone)) eis wise tla cosdala wiiscowe: sae TA ISUSUCr ea sna sons 1.935 12 | Auckland, Queen’s Wharf ........ Wr Ronsterds serie 1.952 13 | Tararu, Thames (S.S. Wakatere) .. EK. L. Baggertrom 1.899 14 MhamMes WWINAE », Lucas & Le Souef (1909) BY L. HARRISON AND HAZEL C. WEEKES. 471 Egernia striolata Peters Rec. by Lucas & Frost (1893) Lygosoma (Hinulia) quoyi D. & B. 3 $5 do. L. (Siaphos) maccoyi Lucas & Frost °F >A do. L. (Liolepisma) pretiosum O’Shaugh. eels; do. and Egernia whitei Lacep. } Now recorded from spéci- Lygosoma (Liolepisma) entrecasteauci D. & B.{ mens in our collection. It seems likely, when the large Scincid fauna of Australia is adequately examined, that a considerable number of species will prove to be viviparous, since this mode of reproduction offers obvious advantages. Definite placentation has been recorded only in the case of Tiliqua scincoides by Flynn (1923); although Haacke (l.c.) mentions certain relations between the uterine wall and that of the yolk-sac in Trachysaurus rugosus. Of the nine species mentioned above as being viviparous we have examined pregnant females of five, the first, second, fifth, eighth, and ninth. In L. entrecasteauxi alone have we found a definite placenta comparable with that of the classic Chalcides tridactylus described by Giacomini (1891). The remaining four exhibit highly vascularized external allantoic and “uterine” walls, with their respective circulations in close apposition, but no marked placental area such as we describe below. The con- ditions would appear to be much the same as obtain amongst the species of Chalcides (Graham Kerr, 1919, p. 480), of which one only has a definite placenta, while in others there is merely close apposition between the maternal and foetal circulations. Since Flynn (1923) has recorded the existence of an allantoplacenta in Tiliqua, and has promised a further account of it, we do not propose to do more than draw attention to the fact that in a pregnant 7. scincoides in our collection, containing young 140 mm. in length, we find neither the villous folding of the uterine wall, nor the great modification of the chorionic epithelium which occurs in the European Chalcides tridactylus, and in Lygosoma entrecasteauxi as described below. We have to thank Mr. J. R. Kinghorn, C.M.Z.S., of the Australian Museum, Sydney, for identification of our lizards; Mr. G. Burfield, of the Department of Physiology, University of Sydney, for making the photomicrographs reproduced in Plate xlviii; and Mr. Horace Weekes for the drawings which appear on Plate xlvii. ii. Material and Methods. The material at our disposal comprises nine pregnant females of Lygosoma entrecasteauxi collected at Barrington Tops, 150 miles north of Sydney, at a height of 4,500 to 5,000 feet, during the last fortnight of January and the first week of February, 1925, by various members of the Sydney University Party. The lizards were fixed and preserved entire, after opening up the ventral body- wall to expose the contained foetuses within the uteri, in a fluid recommended by Bles (1905) which has proved to be excellent for general field work, since animals may be preserved in it indefinitely. It has given a good general fixation, which is not, however, in the case of all kinds of cells, good enough for fine histological work. It has, moreover, one serious fault in that it renders tissues very brittle, and so makes difficult subsequent dissection of material fixed in it. The embryos examined are all approximately in the same stage of develop- ment, 30 to 40 mm. in length, indicating a fairly long gestation period, since % 472 PLACENTATION IN LYGOSOMA ENTRECASTEAUXI, comparatively small differences are shown amongst those collected at random over a period of three weeks. The breeding season would appear, moreover, to be a very regular one, as numerous males were vainly sacrificed in an endeavour to secure younger stages, there being no external differences between the sexes. All the obvious females showed by their bodily contours an advanced condition of pregnancy. The present communication, then, describes what is practically a single stage, and that much too advanced for the satisfactory elucidation of the earlier onto- genetic features, which should prove of considerable theoretical interest. The authors will endeavour to obtain a series of earlier stages during the coming summer, but have thought it advisable, in view of the interest attaching to the discovery of a second fairly complex allantoplacenta in the Scincidae, to publish the description which follows of the actual placental condition, hoping to be able on a future occasion to give some account of its development. Examination was made by means of whole mounts of both foetal and maternal placentae, and of the two together; and by means of paraffin sections of the conjoint placental area alone, and of the whole uterine sac with its contained ‘embryo, taken transversely to the long axis of both oviduct and foetus. All preparations were stained with HEhrlich’s haematoxylin, eosin being used as a counterstain for the sections. iii. Anatomical Relations. Of the nine pregnant females available, two contained seven young, five others five each, and the remaining two only three. It seems remarkable that the number should be odd in every one of nine examples, and we cannot find any explanation for this condition. It is obvious, when large young are produced in a small parent, that these must be somewhat finely adjusted to the space available for their development. One would therefore expect the row of embryos contained in one oviduct to alternate with those contained in the other, and this is found to be so. Moreover, those lizards which contain three young are the smallest of our specimens, but size does not appear to offer any explanation in regard to the five individuals which have five contained young, since, if size were a factor, there is sufficient variation amongst them to lead one to expect four young in one case and six in another. The number of embryos in either oviduct seems a matter of chance. From three of our specimens the embryos had been removed for dissection and sectioning without a note having been taken of their positions. Of the remaining six, four had more embryos in the right oviduct than in the left. The contained foetuses in our specimens are all well advanced, and approaching full term. We have selected for examination what appears to be the youngest stage present, in which we find the foetus to be 26 mm. in total length when extended; as well as several older stages, in which the average total length is 33 mm. There is, therefore, not much difference between our youngest and oldest stages. The embryos within their membranes cause the oviducts to exhibit a series of moniliform swellings (Pl. xlvii, fig. 1) very closely adpressed end to end, so that they adjoin by flat surfaces, and are connected one to another by a short, strap-like section of the oviduct which joins the centres of these apposed surfaces. These swellings are called by Giacomini for Chalcides “incubatory chambers’, but we use the term “uterus” for the whole uterine portion of the oviduct, which must have become greatly distended from its original condition, since the small ovaries lie on its mesial border at the middle of its length, and not anteriorly. Mutual pressure of BY L. HARRISON AND HAZEL C. WEEKES. 473 the embryos contained in the two uteri brings about a flattening of their mesial faces also. Hach swelling is thus roughly oblong in surface view, with an average length of 6 mm., by 9 mm. wide. Certain structures are more or less visible through the thin and much distended uterine wall. As the parent lies upon her back, with the ventral body- wall opened up, and the uteri exposed, the region of the yolk-sae for each embryo shows as a creamy area, ventral and lateral, i.e., anti-mesometrial in position. On turning back the uterus of either side, the allantoplacental region is seen to be dorsal and mesial, and the maternal placental area shows as an opaque patch (Pl. xvii, fig. 1) plainly visible to the unaided eye. Text-fig. 1.—Placental area of uterine wall viewed as a transparent object. Arteries white, veins stippled; R, area shown in Text-fig. 2a; V, villous ridges; x cire. 12 diameters. The series of prominent blood vessels, arteries and veins parallel and, in general, alternate, which runs out from this area round the uterine wall (Pl. xlvii and xlviii, fig. 1; Text-fig. 1) stops at the margin of the yolk-sac, and so helps to delimit the boundaries of the latter. The foetus lies with its long axis parallel to that of the parent, and with the yolk-sac upon its left, and the main portion of the allantoic cavity upon its right side. It is thus in an obliquely vertical position, lying upon its back, with its head curled up towards the dorsal side of the parent body, and directed anteriorly when in the left uterus, posteriorly when in the right. The tail is folded forwards over the ventral surface of the foetal body as far as the head, close to which it turns inwards and ends in a close spiral. The same general relations hold for all our specimens with the exception of one in which the yolk-sac area is distinctly more mesial in position. On dissecting away the uterine wall, and disclosing the foetus lying within its membranes, it is at once obvious that the allantoic sac does not completely surround the whole periphery, but turns inwards round the edge of the yolk-sac and passes internal to the latter structure. Externally, the yolk-sac is covered only by the serosa. It is, at the stage we discuss, apparently much reduced, and extends over only about one-fourth of the circumference of the “blastocyst”? in section, the remaining three-fourths being occupied by the allantois. The yolk- sac is mushroom-shaped, the stalk ascending to meet the allantoic stalk in a common umbilical stalk close to the mid-line of the foetal body at a little distance in front of its hind limbs. The relations may be seen at a glance by reference 474 PLACENTATION IN LYGOSOMA ENTRECASTEAUXI, to the stereogram reproduced as Figure 3 of Plate xlvii. By reference to the same figure, the allantoic stalk is seen to pass directly upwards and to open out, above the embryo, into a wide allantoic cavity which passes round the foetus in all directions to extend, at the lower pole, inwards between it and the yolk-sac, ending, in a ring-like manner, around, but quite free from, the yolk-stalk at its junction with the sac. To the amnion we have not given any particular attention. It is an extremely delicate membrane, which arises from round the base of the umbilical stalk, and closely invests the foetus through all its curvatures. We can find no trace, at this stage, of a sero-amniotic connection. There is no trace of egg-shell or of secondary membranes apparent in our stages, nor can we determine the origin of a coagulum which lies between the uterine wall and the serous membrane in the yolk-sac region (PI. xlviii, fig. 4; Pl. xlix, fig. 3), which may be the remains of albumen, or may be a uterine secretion. iv. The Placentae. Lygosoma entrecasteauxi possesses both an omphaloplacenta and an allanto- placenta, the latter being much more highly organized in the stages at our disposal. The arrangement of the blood-vessels in the uterine wall (PI. xlviii, fig. 1, and Text-fig. 1) suggests that possibly the yolk-sac placenta was more important at an earlier stage. These vessels are so very prominent, and end so suddenly at the periphery of the area of uterine wall overlying the yolk-sac, entering into small relations with the sparse vessels of this region, that the arrangement suggests a change-over from an earlier and more richly vascular omphaloplacenta to a later allantoplacenta. We find in Lygosoma quoyi that the maternal portion of the omphaloplacenta is very richly vascular. In this form, as far as our observa- tions have gone, there is no definite allantoplacenta, and the yolk-sac placenta is very prominent in stages seemingly equivalent to those of L. entrecasteaucxi which we have examined, so that the two are not strictly comparable. In Chalcides, according to Giacomini (1891, p. 356), the yolk-sac placenta ‘se developpe tardivement et reste rudimentaire”. (a) Maternal blood-circulation—We reserve a full statement of the details of maternal circulation for a future occasion, when we shall have earlier develop- mental stages for examination and comparison. Here we give merely a brief outline of the arrangement of vessels in the uterine wall. A large artery and vein run longitudinally on the dorsal side of the uterus, slung in a fold of the mesometrium, and standing out as a prominent ridge bisecting the allantoplacental areas (PI. xlvii, fig. 1, Ut. b. v.). The vein receives a single large branch vein from each placental area, to which latter the artery sends also a single branch. The placental artery runs dorsal to the vein, and branches in a manner best seen by reference to Text-fig. 1, which shows a series of paired branches running parallel with one another transversely round the uterus. The placental vein is branched in a somewhat similar fashion, the branch veins in general alternating with the branch arteries. Arteries and veins extend to the periphery of the yolk-sac area, as mentioned above. The arteries give off frequent minute branches to the capillary network of the uterine wall, which is extraordinarily rich, and ramifies as a close plexus over the whole of its area (Text-fig. 2, a). From this reticulum small vessels enter the branch veins, and are carried back to the main placental vein. Over the yolk-sac area there remain only a few small irregularly ramifying vessels. The maternal portion of the allantoplacenta appears as a fusiform to BY L. HARRISON AND HAZEL C. WEEKES. 475 elliptical opaque whitish area (PI. xlvii, fig. 1, Plac.), the opacity being due to the fact that the internal face is raised up into a series of villous ridges, covered by a cubical epithelium, and containing otherwise only capillary blood-vessels with their endothelial lining. At either end of the placenta these ridges run in a general longitudinal direction, but in the middle region, for about half its length, they are arranged transversely. Hach apparent major ridge is formed by the Text-fig. 2—(a) Vascular reticulum in uterine wall in optical section; (b) Uterine villous fold in optical section. closely sinuous plaiting of a much narrower fold, the whole having a very complex form. The vessels occupying these ridges are fed by short branches from the main branch arteries and veins. A prominent ridge runs round the periphery of the whole area. The nature of these ridges is shown in Text- figure 2, b; and some idea of their complexity may be gained by reference to the section reproduced as Figure 3 of Plate xlviii. (bo) Foetal blood-circulation.—A section of the umbilical stalk is shown in Plate xlix, figure 6, from which it will be seen that the endodermal lining surrounds a simple cavity, formed by the junction of the yolk-sac and the allantoic stalks, which passes into the body of the foetus, and runs posteriorly for some considerable distance before entering the gut. The splanchnopleural mesoderm is thrown into somewhat intricate folds, wrapping round the various blood-vessels. Close to the body, the dorsal space surrounded by these folds is filled with proliferated mesenchyme, surrounding the umbilical and vitelline veins, which are both single, and one of the paired vitelline arteries. An umbilical artery, formed by the junction of two distinct arteries in the allantoic stalk (see Pl. xlix, fig. 5), and the other vitelline artery lie on the ventral side of the stalk lumen. When these vessels are traced into the body of the embryo, the umbilical vein enters the liver, and there joins with the intra-hepatic vessels, and with the ductus venosus; the umbilical artery is found to branch from the aorta posteriorly, and not from the sciatic artery as in the chick; the vitelline vein enters the liver; and the two vitelline arteries are given off separately from the aorta. As the yolk-stalk is followed outwards, it is seen that a second and smaller vitelline vein enters the umbilical vein. As the allantoic stalk passes upwards, the umbilical artery divides into two branches, which come to lie on either side of the umbilical vein. When the 476 PLACENTATION IN LYGOSOMA ENTRECASTEAUXI, allantoic stalk opens out into the sac, the blood vessels are carried across the lumen of the latter in a remarkable and interesting manner direct to the placental region. In three of the four embryos of which we have serial sections, a pleated fold arises from the inner allantoic wall (Pl. xlvii, fig. 3; xlviii, fig. 2; xlix, fig. 5, Pil.), within the inner free edge of which the vessels are carried. Since this pleat is covered externally with endoderm continuous with that lining the allantoic cavity, there would appear to be no doubt that it has arisen originally as a fold of the allantoic wall. The base of this fold is of considerable extent, and can be traced round the inner wall of the allantois almost to the yolk-sac region. Figure 5 of Plate xlix shows clearly enough the way in which the vessels link up; and in the photomicrograph which appears as Figure 2 of Plate xlviii the umbilical vein is shown actually in transition to the placental region. In our fourth series of sections a somewhat different condition occurs. In place of the flat fold, a blunt finger-like process is pushed out from the inner wall, and passes across the lumen to bring about the same ultimate result. Since the fold occurred in three embryos which were dissected (and puzzled us consider- ably until sections disclosed its actual nature) as well as in three sectioned embryos, it would appear to form the normal method of transference. Flynn (1923, p. 77, and Fig. 1, p. 74) describes and figures for Tiliqua scincoides a somewhat similar arrangement. He writes:—“But the outer wall of the vesicle is also supplied by vessels which leave the allantoic stalk near the body of the embryo, and pass right across the vesicle to ramify through the mesen- chymal layer of the placental face of the allantois. The presence of such a method of transmission of allantoic vessels—by direct cellular bridges across the allantoic cavity—is interesting in that it has already been recorded by Hubrecht for a monodelphian mammal, viz., Erinaceus’”. Flynn figures the allantoic stalk as passing down posterior to and parallel with the yolk-stalk, but his “cellular bridge’ offers so close an analogy with the allantoic stalk of Lygosoma entrecasteauxi, and with its continuation as the edge of a pleated fold, that we suspect further examination will prove the relations to be the same in both forms. In any event, the condition shown by Lygosoma has nothing in common with that described by Hubrecht for the hedgehog. (c) The omphaloplacenta—This structure is obviously in a condition of regression, and it would serve no purpose to discuss in any detail the scarcely different stages which we possess. In Figures 4 and 5 of Plate xlviii, the general relations of our extreme stages are shown. The yolk-sac is more voluminous in the latter than in the former, and contains a homogeneous coagulum within portion of its cavity, which resembles generally that found lying between the serosa and the uterine wall in the former, which is shown in more detail in Figure 3 of Plate xlix. Since this coagulum stains deeply with haematoxylin, it differs altogether from that found in the allantoplacental region, which remains practically colourless. Possibly this coagulum represents the remains of an albuminous coat, but it seems more likely that it is some trophic secretion of the uterine wall, which has gravitated to the bottom of the cavity after fixation. The uterine epithelium is here cubical, the cells measuring on an average .028 mm. in height by .016 to .022 mm. in width, which is approximately the same size as those covering the villous ridges of the allantoplacenta, and greatly larger than those of the intermediate area. The epithelium is underlain by a network of small capillaries, which here and there appear to cause slight bulges of the overlying cells, but no folding into ridges can be seen. BY L. HARRISON AND HAZEL C. WEEKES. 477 The chorion consists of an ectodermal layer of cubical cells measuring .03 mm. in height and width, bounded internally by a narrow layer of mesoderm. The endodermal cells of the yolk-sac itself are relatively enormous, measuring .124 mm. in height by .05 mm. in breadth. Their general form is indicated in Figure 4 of Plate xlix. They show large, densely-staining nuclei, and the cytoplasm is extensively vacuolated, and is packed to a varying extent with large yolk sphaerules. (d) The allantoplacenta.—The general disposition of the maternal portion of this organ has already been dealt with in the description above of the maternal circulation. The thickened and plaited area which is clearly delimited by a ridge running round its margin, measures 10 mm. long by a little more than 3 mm. wide in the example figured (Text-fig. 1; Pl. xlviii, fig. 1). The thickened chorionic epithelium which characterizes the foetal portion of the placenta is of slightly greater extent, its margin passing that of the maternal area right round the periphery. The way in which the inner face of the uterine wall is folded leaves a series of crypts lying between the complicated villous ridges. There is, however, no interpenetration of chorionic ectoderm into these crypts, maternal and foetal layers being closely apposed, so that the general form of the villous ridges impresses itself upon the chorionic surface, but without any kind of fusion whatever. However carefully the placental area is dissected away, the maternal and foetal membranes at once become free from one another, and a certain degree of separation follows upon careful embedding and sectioning of the whole blasto- cyst within its uterine chamber. Reference to Figure 2, Plate xlviii will show this separation, as well. as the fact that the sectional outline of the maternal villi is complementary to that of the foetal chorionic ectoderm. Drawings made by means of the camera lucida, cut out in profile and then apposed, have shown that the two membranes fit closely together. Flynn (1923, p. 76) describes the connection between these epithelia in Tiliqua as very intimate, and suggests a certain degree of plasmodium formation, and of interpenetration of foetal cells between those of the uterine wall. Our photographs and drawings show conclusively that such things do not occur in Lygosoma. We have to do simply with two epithelial sheets with every cell clearly bounded, which are in a position of close apposition, but are in no way joined or fused, nor do they exhibit amoeboid processes or plas- modium formation. Certain portions of the photomicrograph which forms Figure 3 of Plate xlviii might perhaps be interpreted as suggesting this last, but examina- tion of the actual sections shows that these represent sections of villi cut tangentially. The maternal crypts thus form chinks or interstices between the two mem- branes, and are for the most part filled with a ropy, non-staining coagulum which, as already mentioned, differs from those found in the yolk-sac region. The villous ridges project on an average .1 mm. from the uterine wall. The cells of the epithelium covering them measure in section .025 mm. by .017 mm. At the edges of the placental area there is no transition of this thickened epithelium into the squamous type which lines the uterus, as occurs in the foetal ectoderm. The epithelium is strongly ciliated, and the last lateral thickened and ciliated cells of either side of any section adjoin a non-ciliated squamous cell of the general surface. The cells have a densely granular, dark-staining cytoplasm, but show no signs of vacuolization, or of glandular activity. No specialized glands occur over the area. The cells of the chorionic ectoderm are columnar and extremely elongated, being in all cases twice as long as broad, while cells occur quite frequently which 478 PLACENTATION IN LYGOSOMA ENTRECASTEAUXI, are three times as long as broad. The free surfaces are mammillate in form with a dark border in section and strongly ciliated. The whole of the cytoplasm is densely granular and shows little sign of vacuolization. The nuclei are large and vesicular, in many cases occupying almost the whole width of the cell; most of them show one large densely staining nucleolus, central or slightly eccentrie in position, but in some there are two nucleoli placed at opposite poles. No signs of active multiplication are apparent in the form of mitotic figures, but this is possibly due to the slowness of fixation since some cells show two small nuclei closely apposed to one another along the long axis of the cell while here and there two extremely narrow cells with their nuclei side by side suggest recent division, in which the cell has not yet grown to its maximum size. The outline of the free surface of this epithelium in section is very irregular and consists of a succession of bays and prominences which correspond to the prominences formed by the uterine villous ridges of the maternal placenta. The extent of thickened ectoderm slightly exceeds that of the maternal placental area, but peripherally there is a marked flattening out of the epithelium until at the margin of the foetal placenta the cells are absolutely cubical and only about one-quarter the height of those of the middle of the placental region. These cubical cells have densely staining nuclei right at their bases and exhibit a very definite border of denser material which stands out clearly from the remainder of the cytoplasm. Passing beyond the margin of the placental region the chorionic ectoderm becomes squamous, and the cytoplasm loses its granular appearance and the nuclei their vesicular form, the latter becoming ellipsoid in shape and staining very densely and uniformly (Text-fig. 3, b). Text-fig. 3—Chorionic ectoderm in surface view; (a) in middle of foetal placenta; (vb) transition zone at edge of same. The somatopleural mesoderm has throughout most of the placental region a fairly complex structure. The main vascular network lies close at the base of the chorionic cells, so much so that in transverse section the extremely numerous capillary vessels have the appearance of piercing the cell bases. These minute vessels appear in all cases to have endothelial lining, the lumen being surrounded by an extremely fine but quite definite membrane, which is frequently enlarged to surround sickle-shaped nuclei which are curved round the narrow lumen of the vessel. Internal to these capillaries lies a definite cellular membrane with BY L. HARRISON AND HAZEL C. WEEKES. 479 flattened nuclei lying parallel to the bases of the ectodermal cells. Internal to this is a sheet of mesenchymatous connective tissue, thin over the greater part of the placental region but forming a fairly thick cushion from the caudal end, commencing where the allantoic vessels crossing the cavity attach themselves to the placental region, and running round in the direction of the yolk-sac. This tissue consists of a loose reticulum containing scattered nuclei, and is bounded internally, by a second endothelial layer. Through this connective tissue, through- out most of its extent, there runs a thin sheet of muscle fibres. v. Comparison with Chalcides. The literature of placentation in Chalcides is not very voluminous, and is, apparently, confined to a period of about five years following upon Giacomini’s original announcement in 1891. We quote below (p. 485) a number of titles having reference to it, but most of these papers, unfortunately, do not appear to be available in Australia, and we have had access to two only of Giacomini’s papers, viz. what is apparently a reprint of his original account in the Monitore zoologico italiano which appears in the Archives Italiennes de Biologie, Vol. xvi, for 1891; and a brief summary of his results in the Anatomischer Anzeiger for the same year. Lesser references to his results occur in Strahl (1891), Kerr (1919), and Schauinsland (1906). In Chalcides, apparently in the vicinity of Sienna, although this is not definitely stated, ripe ova from 2.5 to 3 mm. in diameter descend into the oviduct about the middle of May, and are there fertilized. Neither an albuminous layer nor secondary membranes are formed about them. They differ from those of Oviparous reptiles in the small yolk content. The young are born, free from foetal envelopes, at a length of 95 mm., after a gestation period of three months. The number of ova varies between 5 and 15, the average being 8 to 10, but all do not necessarily develop. The ova become spaced out in the oviducts, producing ultimately moniliform swellings which persist for some time after birth. The ova lie with their animal pole mesometrial in position. The embryo lies trans- versely to the long axis of the blastocyst, upon its left side on the yolk-sac, with the allantois, when first developed, upon its right side. The latter grows down ventrally, but never passes round the yolk-sac externally, that is to say, between it and the serosa. The placental area is elliptical, the villous ridges of the uterine wall running in the general direction of the long axis of the ellipse. The chorio-allantoic membrane shows ridges corresponding to the “crypts” in the uterine wall, but there is no close interlocking between the two membranes, which are easily separable, and between which interstices filled with coagulum occur. The coagulum is stated to be the product of the thickened epithelium covering the uterine villi, and may be in part produced by the cytolysis of proliferated cells, but this is not certain. The blood circulation is not described in detail, but it is mentioned that the vascular network of the foetal is richer than that of the maternal placenta, so much so that in places on the peripheral region the meshes surround only one or two epithelial cells. An omphaloplacenta precedes the allantoplacenta, but is not very highly developed, and is marked chiefly by the persistence of thickened chorionic ectoderm in this region. The chorionic ectoderm of the allantoplacental region is uniformly and strongly thickened, much more so than in the yolk-sac region, the cells being narrow and very high, with granular contents. This epithelium is thrown into villosities, which correspond to the “crypts” in the uterine wall, and the 480 PLACENTATION IN LYGOSOMA ENTRECASTEAUXI, bases of these villi are underlain by a rich vascular network, as are those of the uterus. These last arise before those of the chorion and are, in general, narrower in form. They are occupied internally by a delicate reticular connective tissue, which supports the capillary vessels, and are covered by a simple epithelium of a secretory character, which is much thicker than that lining the non-placental portion of the wall. The cell contents are dark and granular at their bases, becoming light and transparent towards their free faces. Outside the area of placental modifications, both uterine wall and chorio- allantoic membranes show closely apposed and extremely rich capillary networks. No special description is given of the method by which the allantoic vessels reach the placenta, but from words used (1891, p. 344):—‘“. . . Jallantoide, en s’étendant dans le celoma blastodermique ou externe, sous forme d’une vésicule revétue, intérieurement, par l’entoblaste, et extérieurement par la lame viscérale du mésoblaste (feuillet connectif ou vasculaire de l’allantoide) dans laquelle courent les vaisseaux allantoidiens, ne se soit pas encore adossée, par sa face externe ou connective a la face interne de la vésicule séreuse, cependant, cette derniére commence déja a se pourvoir etc.’’, it would seem that the vessels pass round the allantoic wall, and are not carried across the cavity by any special contrivance. We are handicapped in making a comparison between Chalcides and Lygosoma by the fact that we have not access to the complete descriptions of Giacomini. The figures given in the paper on which we have chiefly to rely illustrate only the external features of the blastocyst. We do not even know whether the Italian embryologist has published figures illustrating the histological relations in Chalcides, but the fact that no such figures have been reproduced in any general work known to us seems to imply that none exists. Moreover, we have but one stage, and not a developmental series, for comparison. Despite these limitations, however, it is possible to make some comparison, and to indicate that the correspondence between the two forms is, in general, remarkably close. Chalcides is a larger lizard, producing a greater number of young at a birth. The general relations of the embryo to its membranes are identical, as well as the relations of the foetus to the maternal organs. Both have an omphaloplacenta of the same type, which in Chalcides is at no stage very much developed, but in Lygosoma seems to indicate a more actively functional condition at an earlier ontogenetic stage. In both, the allantois does not surround the yolk-sac externally; the epithelial lining of the uterine wall changes in character and becomes notably thickened over two placental areas at opposite poles; and the chorionic ectoderm is modified in an apparently identical manner at the same places. In both, the mesodermal elements of the chorion and allantois fuse to form a complex allantochorion, which carries a rich vascular network in correspondence with a similar plexus in the uterine wall, the two being closely apposed. In the allantoplacental region each form shows a series of interstices between foetal and maternal tissues, filled with a coagulum identical in character. Finally the foetal and maternal membranes are easily separable in both, and there is no actual fusion between them. Lygosoma entrecasteauxi differs from Chalcides tridactylus in the following points :— 1. The allantochorion is not thrown into villous folds corresponding to those of the uterine wall. BY L. HARRISON AND HAZEL C. WEEKES. 481 2. The villous ridges of the latter run chiefly in a transverse direction and not longitudinally. 3. Both uterine and chorionic epithelia are ciliated. 4. The allantoic vessels are carried across the cavity in a special fold of the inner allantoic wall, and do not pass round the periphery of the allantoic sac. vi. Theoretical Considerations. It is not proposed at this stage to enter upon a detailed discussion of the significance of the reptilian allantoplacenta, but certain outstanding features appear to demand brief comment. The parallelism between the occurrence of a fairly highly-organized placenta in two not very closely related Scincid genera, Chalcides and Lygosoma, accompanied in each case by indications of simpler placental conditions in related species, is remarkable. Brief references to the placental condition in Chalcides are fairly general in text-books, but Graham Kerr (1919) is the only embryologist known to us who has given expression to the importance of this phenomenon, and even he has not discussed its theoretical bearings. In the important discussions of placentation associated with the names of Van Beneden, Hubrecht, Assheton, Jenkinson, and Hill, no consideration has, so far aS we are aware, ever been given to the possible value of a reptilian placenta in elucidating the evolutionary history of that organ. Giacomini’s original papers are not available to us, but, so far as we are able to judge from the summaries and extracts from which we quote, he himself was fully seized with the importance of his discovery, originally announced much earlier by Studiati, and it is remarkable that his extensive work should have attracted so little attention. Placentation in Scincid lizards, a specialized family of a specialized group of reptiles, obviously can have no direct genetic relationship with the analagous process in mammals. To our thinking, however, this very fact makes the reptilian placenta so much the more important, for there is here displayed an independent placentation in an otherwise non-placental group, which in many respects affords a very close parallel to mammalian conditions. If the statement of Lucas and Frost quoted above (p. 470) be confirmed, namely that Lygosoma entrecasteauxzi is oviparous through part of its range, we should have the extraordinarily interesting condition of a species exhibiting in part normal sauropsidan development, and in part a placental mode; but it is probable that this statement rests upon a wrong conception of the contents of the oviducts, such as the same authors have put forward in the case of Tiliqua scincoides. Be that as it may, the conclusion seems to us inevitable that placentation is a functional adaptation which, given certain prerequisites, may have arisen independently on many occasions before the higher mammals settled down to the placental mode. These prerequisites comprise (1) an ectodermal layer which has entered into some metabolic relation with the external environment; (2) a yolk-sac; and (3) an allantois. Without committing ourselves in any way to the view advanced by Hubrecht that the external layer of ectoderm of the anamniote larva is homologous with the mammalian trophoblast, we may justifiably commence an argument from function with this layer. The three prime functions to be provided for in the development of an embryo are respiration, nutrition, and excretion. In the anamniote larva, nutrition is provided for by means of the contained yolk-mass; excretion by the early establishment of archinephric ducts opening into a cloaca; so that the one function remaining to the external cellular layer is that of 482 PLACENTATION IN LYGOSOMA ENTRECASTEAUXI, respiration. Since we hold the conventional view that the extra-embryonic phenomena exhibited by the amniota are the direct consequence of adaptation to a terrestrial environment, and to the evolution of a shelled egg, we view the extra-embryonic ectoderm primarily as a pneumatoblast, the yolk-sac remaining homologous with the internal yolk-mass of the anamniote larva, and the allantois being a new structure for the reception of poisonous excretory waste. The view we would take of amnion formation was developed before we had studied Hill’s work upon the early ontogenetic features of marsupials, but appears to be supported by his observations, which have heen confirmed by Hartmann, that there is an early division into formative and non-formative ectoderm, clearly differentiated morphologically, in the marsupial blastocyst. We would suggest that the formative ectoderm, possibly through the assumption of other functions, has lost its primarily pneumatoblastic character, and that it is due to the necessity for a maximum area of respiratory ectoderm that there is rapid proliferation not only in the direction of the yolk-sac, but also, in the form of amniotic folds, above the developing embryo. The extremely rapid proliferation noted by Wilson and Hill in Ornithorhynchus, in which the new-laid egg contains already a well-formed embryo, compared with the condition in birds, where the proliferation comes at a much later stage, since the egg at laying is only in an early stage of segmenta- tion, and moreover displays a condition of arrested development, seems to us highly significant in this regard. For us, then, amniotic folds are primarily due to the necessity for a maximum extension of pneumatoblast, and the water-jacket function is secondary. This view, if acceptable, would cut the ground from under Hubrecht’s argument that, if the prime purpose of the amniotic cavity be to sub- serve the function of a water-jacket, it should exist from the outset in a function- ally perfect condition. Selenka’s view of the mechanical origin of the amniotic folds owing to the downgrowth of the embryo into the yolk-sac is untenable on several grounds, such as the nonoccurrence of these folds in the Hlasmobranchs, the absence from them of yolk, the necessity for accounting for proliferation of the edges of the folds, and the backward production of the reptilian amnion far beyond the limits of the embryo. The hypothesis outlined here would have the effect of dividing the nonformative ectoderm into two categories, pneumatoblast and ectodermal lining of the amniotic cavity. This last would, by virtue of its position, lose most, if not all, of its functional significance, and might tend towards elimination in the course of evolution. It seems to us that this suggestion may have a bearing upon the problem of amnion formation amongst the Mammalia. The yolk-sac we look upon as homologous with the yolk-mass of the anamniote larva, increased in size to permit of the elimination of a larval period, and hence, owing to the mechanical difficulties involved in its enclosure, folded off from the body of the embryo. Since yolk has to be carried to all parts of the body of the embryo, and since blood is the carrier, blood is first formed in the yolk-mass of the anamniote, and in the splanchnopleural mesoderm covering the yolk-sac of the amniote. That is to say, a blood circulation comes into existence upon the yolk- sac to subserve the function of nutrition. Having come into existence, it is available for other purposes. An omphaloplacenta becomes possible by virtue of the existence of a trophic blood-circulation upon the surface of a yolk-sac. Similarly an allantois has come into existence for the reception of excretory waste, and its continually increasing bulk in the sauropsid egg, together with the gradual diminution of the yolk-sac as its contents are absorbed, brings a blood- circulation primarily concerned with the metabolism of its own cellular walls into relation with the external covering of the blastocyst, using the word in its ae BY L. HARRISON AND HAZEL C. WEEKES. 483 widest sense, and makes possible an allantoplacenta. These views have been taught, we should suppose, in most schools of zoology for more than a generation, and their reiteration here would be gratuitous were it not for the paradoxical standpoint assumed by so distinguished an embryologist as Hubrecht, obsessed apparently by his preponderant studies of ontogenetic phenomena in the Mammalia. It may seem presumptuous for workers with small experience in embryology to enter the lists against admitted champions, but we feel strongly that a con- sideration from the functional rather than from the morphological viewpoint must lead to a clearer understanding of the facts of early ontogeny; and we are of opinion that the independent occurrence of an allantoplacenta in two not very closely related reptiles justifies our functional point of view, and delimits the extent to which the existence of an allantoplacenta may justifiably be used as a phylogenetic argument. Thus the view of Hill (1897), still held by Flynn (1922), that the marsupials must be derived from placental ancestors, rendered inherently improbable by the persistence of the egg-shell in marsupials, becomes, to our thinking, untenable in the light of the fact that two reptiles have developed an allantoplacenta. We agree with the general statement of Giacomini, who writes (1891, p. 354) :—‘“‘Le placenta du Seps (Chalcides) confirme la loi générale établie par Ercolani, Turner, Romiti, et appuyée par Tafani, sur le mode de nutrition des embryons et des fetus dans l’utérus, puisque les aptitudes a sécréter, constatées dans les élements de la portion maternelle, manifestent que celle-ci, dans certaines, périodes du développement, acquiert la fonction d’élaborer et de fournir un supplément de matériaux nutritifs au nouvel étre. C’est pourquoi, ce n’est pas seulement pour les fonctions de la respiration que l’allantoide, chez le Seps, prend, dans une région donnée, aprés s’étre soudée avec la membrane séreuse de von Baer, la forme de placenta allantoidien, mais bien, spécialement, pour les fonctions de nutrition, ou mieux encore, d’absorption des substances assimilables dont l’ccuf fécondé, qui se développe, est dépourvu. Ici également, disais-je, nous trouvons lunité du processus physiologique qui préside a la nutrition de tous les cufs en voie de développement et qui, méme, sert & nous expliquer le pourquoi des dispositions particuliéres rencontrées dans les annexes fcetales du Seps’’. We suggest, then, that there are three stages in placentation:— (i) A chorioplacenta, primarily conditioned by the chorionic ectoderm (non-formative ectoderm, trophoblast), in accordance with the view of Minot. This we consider to be first respiratory, though later it may (and does) assume other functions. We have advanced a theoretical argument for its existence. In fact, thickening of the chorionic ectoderm over the embryonic region is well shown in sections of a 4 mm. embryo of Tiliqua scincoides in the collection of the Department of Zoology (Fig. 7, Pl. xlix) indicating a very early functional activity for this layer. We do not discriminate between a unilaminar and a bilaminar blastocyst wall, nor concern ourselves with the origin of the underlying lamina where this wall has a double character. The ectoderm is the important thing. Flynn (1922, p. 541) has suggested an initial stage of placenta- tion by means of the non-vascular wall of the bilaminar omphalo- pleure, to which he has applied the term “metrioplacenta”. The term seems meaningless, and we cannot see that any particular 484 PLACENTATION IN LYGOSOMA ENTRECASTEAUXI, virtue attaches to this region which is not shared by the ectoderm of the remaining portion of the blastocyst wall. The chorioplacental stage is never omitted, and the use of the word “placenta” in compounding the term is meant to imply merely that the membrane is concerned in metabolic exchange between the surrounding environment and the embryo. Such use may not be considered happy, since the term is obviously capable of application to the sauropsidan egg, in connection with which the word “placenta” seems out of place. But it falls into line with the accepted terms which follow, and is an expression of what we believe to be a real homology, namely that of the sauropsidan chorionic ectoderm with the mammalian trophoblast. (ii) An omphaloplacenta, which both ontogenetically and phylogenetically precedes the third stage, but which may be omitted in ontogeny. (iii) An allantoplacenta, the final stage phylogenetically, which may never be attained (some marsupials), and which may follow directly upon the chorioplacental condition (most Hutheria), the omphalo- placenta having been eliminated. Since it is our intention upon a future occasion to expand the thesis briefly outlined in this section, we have not burdened our list of references with the voluminous literature of placentation. References to the workers whom we mention will be found in the bibliographies to Hill’s papers, which we include. vii. Summary and Conclusions. The present communication consists of a description of the placentation, at one stage of the Scincid lizard Lygosoma (Liolepisma) entrecasteauxi, with what we consider to be adequate figures. We hope to obtain earlier ontogenetic stages during the coming summer which will render possible a more extended account. The chief facts observed and conclusions arrived at are as follows:— 1. The lizard is strictly viviparous and not ovo-viviparous, no trace of shell or albumen having been observed. 2. An omphaloplacenta is present, which is in a state of regression, but shows some evidence of having been more actively functional at an earlier stage. 38. A true allantoplacenta is present, comprising a uterine part of vascular villous ridges with modified epithelium, and a foetal part with greatly modified chorionic ectoderm and a complex allantochorion. 4. The main allantoic blood vessels passing to the placental area are carried across the cavity by means of a remarkable fold in the inner wall. 5. The occurrence of true placentation in two not very closely related Scincid lizards, Lygosoma and Chalcides, in:Australia and Europe respectively, indicates that the allantoplacenta is a functional adaptation, which may have arisen inde- pendently many times in evolution, and upon the mere occurrence of which phylogenetic statements cannot justifiably be based. 6. A consideration of reptilian placentation suggests to us three stages of placentation, chorioplacenta, omphaloplacenta, and allantoplacenta arising in that serial order both in ontogeny and in phylogeny. ‘ References. Buses, 1905.—The Life-history of Xenopus laevis Daud. Trans. Roy. Soc. EHdin., xli, pp. 789-821. BY L. HARRISON AND HAZEL C. WEEKES. 485 FLYNN, 1922.—The Phylogenetic Significance of the Marsupial Allantoplacenta. Proc. LINN. Soc. N.S.W., xlvii, 4, pp. 541-544. , 1923.—On the Occurrence of a true Allantoplacenta of the conjoint type in an Australian Lizard. Rec. Aust. Mus., xiv, pp. 72-77. GIACOMINI, 1891.—Matériaux pour l'étude du développement du Seps chalcides. Arch. Ital. de Biol., xvi, pp. 332-359. (Reprint from Monitore Zool. Ital., Ann. ii, Nos. 9-10.) , 1891a.—tiber die Entwickelung von Seps chalcides. Anat. Anz., vi, pp. 548-551. , 1893.—Contribution 4 la connaissance des annexes feetales chez les Reptiles, le et 2e. Note préventive. Arch. Ital. de Biol., xviii, pp. 336-349. (Abstract of paper in Monitore Zool. Ital., iii, Nos. 6-9.) , 1894.—Sur lVoviducte des Sauropsides. Arch. Ital. de Biol., xxi, pp. 147-151. (Abstract of paper in Monitore Zool. Ital., iv, Nos. 10-12, 1893.) , 1894a.—Nouvelle contribution 4 la connaissance plus parfaite des annexes foetales chez les reptiles; Réception du sac vitellin et de l’allantoide dans la cavité abdominale. Arch. Ital. de Biol., xxi, pp. 151-154. (Abstract of paper in Monitore Zool. Ital., iv, No. 7, 1893.) HaAackeE, 1885.—Ueber ein neue Art uteriner Brutflege bei Reptilien. Zool. Anz., viii, pp. 435-439. Hiuu, 1897.—The Placentation of Perameles. Quart. Jowrn. Micr. Sci., xl, pp. 385-442. , 1910.—The Harly Development of the Marsupialia, etc. Quart. Journ. Micr. Sci., lvi, 1, pp. 1-134. Kerr, 1919.—Text-book of Embryology: II, Vertebrata. London, Macmillan. Lucas and Frost, 1893.—The Lizards indigenous to Victoria. Proc. Roy. Soc. Vict., 1893, pp. 24-92. Lucas and LE Souerr, 1909.—The Animals of Australia. Melbourne, Lothian. SCHAUINSLAND, 1906.—Die Entwickelung der Hihaute der Reptilien und der Vogel, in Hertwig’s Handbuch, i, 2, pp. 177-234. STRAHL, 1891.—Placenta und Hihaute, in Merkel u. Bonnet, Ergebnisse der Anat. w. Entwick., i, pp. 557-8. STUDIATI, 1851.—Intorno alle connessione dell’uovo coll’ovidutto nel Seps tridactylus. Mem. Accad. Torino, 2, xv, pp. 101-113. EXPLANATION OF PLATES XLVII-XLIX. Lettering. All. allantois; All. Ch. allantochorion; All. St. allantoic stalk; Amn. amnion; Amn. C. amniotic cavity; Cap. capillary; C.b.v. chorionic blood-vessel; C.H. chorionic ectoderm; C.7. connective tissue; Ch. chorion; Co. coagulum; Foet. Plac. foetal placenta ; T.A.W. inner allantoic wall; Mus. muscle; Plac. placenta; Pl. pleated fold of inner allantoic wall; S.A.C. sero-amniotic connection; U.A. (1 & 2) umbilical arteries; U. St. umbilical stalk; U.T. uterus; U.V. umbilical vein; Ut. W. uterine wall; Ut. b. v. main uterine blood-vessels; Ut. Cap. uterine capillaries; Ut. plac. uterine placenta; Ut. V. uterine villous ridge; Vac. vacuole; V.A. (1 & 2) vitelline arteries; V.V. vitelline vein; Y.G. yolk-granules; Y.S. yolk-sac; Y.S. Plac. yolk-sac placenta; Y. St. yolk-stalk. Plate xlvii. Fig. 1.—Left uterus of L. entrecasteauxi from dorsal aspect with three embryos, showing placental areas. Fig. 2.—Foetus of same with uterine wall and foetal membranes partially dissected away to show general relations. Fig. 3.—Stereogram to show relations of allantois and yolk-sac, with their respective stalks and placental regions. Plate xlviii. Fig. 1.—Photomicrograph of total preparation of uterine placental villous ridges and uterine blood-vessels. Fig. 2.—Photomicrograph of section of the allantoplacental region showing general relations of foetal and maternal membranes. Fig. 3.—Photomicrograph of section of uterine wall showing villous ridges and their capillary vessels. . 4.—Photomicrograph of section of portion of embryo showing reduced yolk-sac passing round head region, with coagulum lying between serosa and uterine wall. - 5.—Photomicrograph showing yolk-sac at a somewhat earlier stage, with contained coagulum. area, showing 486 Fig. Fig. Fig. Fig. Fig. Fig. Fig. PLACENTATION IN LYGOSOMA ENTRECASTEAUXI. Plate xlix. 1.—Section of portion of foetal allantoplacenta showing transition of chorionic ectoderm from placental to extra-placental condition. 2.—Section of allantoplacental region showing apposition of chorionic ectoderm to maternal villous ridges. 3.—Section of omphaloplacental region showing details of foetal membranes and uterine wall. 4.—Section of portion of yolk-sac wall. 5.—Section of margin of allantoplacental region showing method of transference of allantoic blood-vessels across cavity. 6.—Section of umbilical stalk. 7.—Section through 4 mm. embryo of Tiliqua scincoides to show early thickening of chorionic ectoderm over embryonic area. DESCRIPTION OF A NEW SPECIES OF MYCETOPHILIDAEH (DIPTERA) WITH LUMINOUS LARVAE. By E. W. FEerauson, M.B., Ch.M. (One Text-figure.) [Read 25th November, 1925.] The occurrence of a species of Mycetophilid with luminous larvae in Tasmania does not appear to have been recorded hitherto. The species concerned, of which a pair in copula and a single male were taken by Mr. A. M. Lea in the Ida Bay Caves, appears to belong to the same genus as that occurring in New Zealand. This latter species was described by Skuse (Proc. Linn. Soc. N.S.W., (2) v, 1890, 678) under the name of Bolitophila luminosa. A new genus—Arachnocampa—was recently erected by Edwards (Ann. Mag. Nat. Hist., (9) xiv, 1924, 175) for this species. The Ida Bay specimens conform well with Edwards’s description of the genus, but the species appears distinct from A. luminosa. ARACHNOCAMPA TASMANIENSIS, 0. SDP. 6. Antennae very long, longer than head and thorax together, the two basal joints small, cup-shaped, the 15 flagellar joints long, cylindrical, the first flagellar ene me is — me So os ache cada Arachnocampa tasmaniensis. n. sp. a. Dorsal view of parts seen after removal of ninth tergite. b. Ninth tergite and anal lobes. joint 13 times as long as the second, the others decreasing progressively in length towards the apex, the 4 subterminal approximately equal in length, the terminal joint very small; scape light brown, flagellum lighter yellowish-brown at base, becoming darker after the second joint. Palpi long, incurved, the apical joint (0) 488 NEW SPECIES OF MYCETOPHILIDAK WITH LUMINOUS LARVAE. flagelliform, yellow. Head black, with some dark hairs on occiput and face, eyes hairy. Mesonotum dark brown with an indistinct darker dorsocentral stripe on each side, these stripes bearing a double row of short hairs; median line also slightly darkened; a tuft of dark hairs above each wing root. Prothoracic ring and lateral membraneous portion of thorax between mesonotum and pleurae testaceous in colour. Scutellum light brown. Pleurae dark brown. Abdomen elongate, cylindrical, the first and apical segments dark brown, the remaining segments dark brown with a broad basal yellow band. Coxae pale yellow or whitish, the anterior practically entirely so, the intermediate and hind coxae dark brown at apex; femora light yellowish-brown, the posterior pair darker, tibiae and tarsi dark brown; anterior metatarsi 14 times as long as tibiae; intermediate tibiae and metatarsi subequal; posterior tibiae not quite 14 times as long as metatarsi. Wings elongate, not as long as abdomen, brownish in colour; costa extending to apex; R, ending in costa, anterior to apex and about opposite distal end of M,; Rs arising anterior to end of subcostal, curved towards apex, ending in costa at extreme apex; subcostal ending in costa near middle and slightly before distal end of Cu. Halteres yellowish; clubs infuscated. %. Similar, thoracic lines and abdominal annulations less distinct; apical lamellae light brown. Length 11 mm.; wings 7 mm.; antennae 45 mm. Type and allotype female in South Australian Museum; paratopotype male in collection of Dept. of Public Health, Sydney. The specimens bear the following label :— “Ida Bay Caves, Tasmania. Arthur M. Lea, December, 1909. In total dark- ness fully 4 mile from entrance.” The species differs from the description of A. luminosa Skuse to some extent in the coloration, but chiefly in the relative lengths of the fore metatarsi and tibiae. 5 Concerning the larvae of this species, Mr. A. M. Lea writes as follows:—‘‘The Ida Bay glow-worm was in the caves in thousands; you saw them sparkling in crowds. They are eaten by a spider-like creature and by the Jdacarabus I named. The larvae construct a hanging affair for themselves, along which they pass; the glow is very bright, the area being about the size of the head of a London hill pin. I saw an apparently identical larva under logs in deep gullies and I am told they are often to be seen in abandoned mines.” It would appear from this description that the habits of the cave larvae are very similar to those of the Waitomo Caves in New Zealand. The identity of the luminous larvae found in gullies and old mines is more doubtful, and it will probably be found that these belong to some species of Ceroplatus. I am indebted to Dr. I. M. Mackerras for the figures of the male genitalia which were drawn irom the paratype; the hairs on the segments are not shown. THE NEMESTRINIDAE (DIPTERA) OF THE AUSTRALASIAN REGION. By I. M. Macxerras, B.Sc., M.B., Ch.M., Linnean Macleay Fellow of the Society in Zoology. (From the Department of Zoology, University of Sydney.) (Plate 1 and seventeen Text-figures. ) [Read 25th November, 1925.] Page. Introduction oa tee es ee ie Sr ae Xe “ic ae dc ae | 4bSS) The genus Trichophthalma ioe wee Pee ace Dee ee ay ee an to) 7 Text-figure 7.—Lamellae of ovipositor seen from the side of— a, T. bancrofti, nov. and b, T. rosea Macq. Text-figure 8.—Egg of JT. rosea Macq. These differences are shown in Text-fig. 6 and are small, but easily seen in mounts, and they are quite constant throughout quite good series of the various species and varieties. It will be seen from the figure that 7. punctata punctata Macq. and T. punctata orientalis, nov., differ in the shape of the distostyle, whereas 7’. harrisoni, nov. and the first named are closely similar. The first two are much alike in coloration and markings (at least there is a continuous series), while 7. harrisoni differs markedly from either. On the other hand, 7. harrisoni, nov., closely resembles 7’. bivitta bivitta Walk., but differs from it in genitalia. These differences would not be worthy of much consideration were it not for the fact that all the evidence strongly indicates that they are quite constant, and for the interesting distributional questions raised. The whole series is very difficult and much more extensive material must be investigated before definite conclusions can be reached. The arrangement here given represents a BY T. M. MACKERRAS. 501 distinct advance in our knowledge of this series, but it must be understood that it is only tentative and may require considerable modification before finality is reached. Female genitalia. The lamellae of the ovipositor not only offer interesting evidence of generic relationships, but also supply a character by which the rosea group may be separated from the rest of the genus. In the rosea group (Text-fig. 7, b) each lamella is flattened and leaf-like and is short and broad in side view. In the other species it is more elongate and parallel-sided and is curved so that the apposition of the two lamellae forms a longitudinal tube. The members of the costalis subgroup in general have much shorter lamellae than do those of the novae-hollandiae subgroup, but the difference is not sufficiently marked to form a useful character. Coloration and markings. For the separation of many species one must rely on small differences of colour and pattern, and to a small extent on differences in size and general body shape. This renders descriptive work rather difficult, because one frequently finds that two series are obviously distinct when placed side by side, and yet it is almost impossible to give an adequate word picture of the differences between them. A careful study of a long series of a number of species points definitely to two conclusions: firstly, that the coloration and pattern of specimens of a given species from the one locality are in general remarkably constant, and secondly, that specimens of a species from one locality often differ markedly from those from another area. An important fact bearing on the justifiability of using these characters is that several very closely related species may occur in one locality. For example, at Barrington Tops several closely related species of the punctata series were taken, all occurring at the same time in the same situation. In the case of three of these we have sufficient numbers to demon- strate their uniformity. In the absence of any linking forms, one feels bound to give them separate specific rank, and it seems reasonable to extend our conclusions, based on this and other experiences of the uniformity of various species, to those of which we have but few specimens. The sexes are often similar in appearance, but in many species there is more or less dimorphism, and in such cases it is frequent to find that the males are readily separable, while the females are difficult to distinguish (e.g., 7’. rosea Macq. and TJ. eques Schin.). There is rarely any very great variation in size, long series sometimes showing not more than a millimetre or two differ- ence, and in such cases the finding of one or two very small individuals may raise the suspicion that one is dealing with a distinct species. One has, however, refrained from using size alone as a character of value, although it is often useful for purposes of quick identification, as in the case of 7’. fusca, nov., which may be separated at a glance from its allies by its very small size. Distribution. Two distinct sets of facts are here discussed: firstly, the distribution of the various species throughout Australasia, and secondly, the seasonal distribution, about which there is some interesting information. These will be taken up separately. Apart altogether from its interest in relation to a possible Antarctic 502 NEMESTRINIDAE OF THE AUSTRALASIAN REGION, connection with South America, the genus Trichophthalma is one of the most suitable in the whole Diptera Brachycera for the study of local distributional problems. The very plasticity which makes systematic work so difficult, takes on an added interest when looked at from the point of view of the effect of isolation and of varied environment. Studies of the distribution of, and relationships between, species in different parts of Australia also lead to interesting con- clusions which may have a more general application. It is necessary, however, to emphasize that, although practically all the material in Australia has passed through my hands, the available evidence does not permit of more than tentative conclusions being expressed, and these are put forward here to interest others in their investigation rather than in their possible application. The evidence strongly indicates that the centre of recent development of the genus is in eastern New South Wales, whence it has spread to Banks Island in the north, to Tasmania in the south, and also to the south-western corner of Western Australia. The last mentioned area presents certain special features, which will be taken up later, the western limit for the present purpose being taken as the eastern half of South Australia. This dispersal from an eastern centre is best demonstrated by considering the fauna of each of the States separately. The part of New South Wales especially to be considered includes the Eastern Divide and the country to the east of it. Twenty of the thirty known species occur in this area and the number of individuals greatly exceeds two-thirds of the whole collection. One species, 7. fusca, nov. and one subspecies, 7. punctata punctata Macq. are only known from west of the Divide and really belong to other States, but the remainder are dominantly New South Wales species. This part of the country may be divided into two areas, the highlands and the coastal zone, presenting different conditions and showing marked differences in their faunas. The rosea group is practically restricted to the coast, only occasional specimens of TJ. albimacula Walk. and T. laetilinea Walk. being known from the mountains. The novae-hollandiae subgroup is well known both from the coast and from the mountains and also extends a little to the west of the Divide. These flies are, however, much more abundant on the mountains and 7. nigro- vittata, nov., is not known from the coast as yet, although it occurs on the coast of Queensland. 7. costalis Westw. is equally prevalent in both areas, but the series in general is a dominantly mountain one, as is also the punctata series, although two species, T. bivitta bivitta Walk. and T. confusa, nov., are purely coastal. The first is represented in the mountains by 7. bivitta nigricosta, nov., but nothing corresponding to J. confusa, nov., has yet been found there. An important fact bearing on this distribution and on the differences in prevalence of the different species at different times is that at the beginning of the season there is an abundance of attractive flowering plants on the coast and few or none on the mountains, while, later on, the highlands are rich in Leptospermum, the most attractive of all, while the coastal flowers are not so numerous or so attractive. This seems to be the most important factor underlying the differences in prevalence, but there is also evidence of a direct climatic influence, a fairly moist climate with mild temperatures being most favourable. In general, spring on the coast and midsummer on the mountains present fairly similar conditions and in both the Trichophthalmas abound, but the faunas of the two are quite distinct, the first having only the rosea group, while the second has all the rest. The coastal fauna in the middle of summer consists mainly of BY I. M. MACKERRAS. 503 the hardier of the species common on the mountains, together with a small special development. Queensland possesses ten of the New South Wales species, which gradually die out as one proceeds north, the rosea group reaching the Johnstone R. (T. eques Schin.), the costalis subgroup extending as far as the Endeavour R. (T7. punctata orientalis, nov.), while JT. novae-hollandiae Macq. extends as far as Banks Island. Three of the species in this State are not known from New South Wales; 7. bancrofti, nov., is nearly related to T. novae-hollandiae Macq. and is not yet known south of Brisbane. While it may turn up later in New South Wales, there is no room for doubt that it is typically a Queensland species, as is probably also 7. fusca. Of the other two species, fulva Walk. and intermedia, nov., our knowledge is too scanty to warrant their consideration. In general the Queensland fauna is a very impoverished replica of that of New South Wales, except that the novae-hollandiae subgroup reaches as great, or possibly a greater, degree of development. The Victorian fauna is also an impoverished one, there being only eight species known, all but one variety being typical New South Wales forms. The case of JT. punctata Macq. is interesting; subspecies orientalis, nov., is abundant in eastern New South Wales and rare in Victoria, while subspecies punctata Macq. is known only from one specimen from west of the Divide in New South Wales, but is not uncommon in Victoria. Tasmania possesses only 7’. punctata punctata Macq., this species being abundant there. This State has been sufficiently well collected to make it certain that the other species are exceptionally rare, if they occur at all. It is a matter of considerable interest that the marginal forms, T. novae-hollandiae Macq. in the north and 7. punctata Macq. in the south, are not primitive unsuccessful species pushed to the periphery, but amongst the most successful, numerous and wide-spread in the whole genus. There is every indication of a radial dispersal from eastern New South Wales of a type that is the reverse of that which holds good for the mammals, which is unexpected, for it is difficult to understand why this genus should not repeat in miniature what is said to hold good on a larger scale in other groups. South Australia in turn has only a very small number of the New South Wales species, only three, or possibly four, of these being known; in addition the Tasmanian 7’. punctata punctata Macq. also extends as far as South Australia. There is, however, a greater development of endemic species, which is what one would expect owing to the greater distance from the centre. T. subcostalis, nov., probably replaces the eastern J. rufonigra, nov., and TJ. griseolineata, nov., is the South Australian representative of the nicholsoni type, while 7. variolosa Licht. cannot be placed at present, but links up with JT. eques Schin. or T. albimacula Walk. The Trichophthalma fauna of Swanland, or south Western Australia, is very interesting, but at present very scanty, there being only six species recorded, two of which form a characteristic series, while the others are sharply differ- entiated from their eastern relatives. The most striking feature of these species is the great elongation of the proboscis and the reduction of the labellae. The leucophaea series is separated, in addition, by the abdominal markings and comes nearer to certain Chilian species than any of the other Australian species. Three species are represented in other parts of Australia, 7. costalis costalis Westw. being replaced in the west by a well marked subspecies which is almost worthy of separate specific rank, while 7. ruficosta, nov., and TJ. grisea, nov., are P 504 NEMESTRINIDAE OF THE. AUSTRALASIAN REGION, specifically distinct from, but clearly related to, the eastern 7. confusa, nov., and T. primitiva Walk. If one excludes JT. fulva Walk., the distribution of which is somewhat anomalous, the evidence points to a prolonged isolation from the eastern species in an area which has been considered to represent ancient Australia. There is no evidence, however, that the Western Australian Trichophthalmas are archaic; on the contrary, those that show affinities with the eastern species are more highly specialized than their eastern relatives in the length of the proboscis and in wing and body markings. The indications are that they are derived from the eastern species and not the reverse, and that they form an earlier and longer separated element of the radial dispersal from New South Wales. The conclusions expressed above are open to the serious objection that the different areas have been studied to very different extents, eastern New South Wales being especially well collected, and it may be suggested that the differences met with may well be explained on these grounds alone. I have had this possibility clearly in mind throughout this study and have endeavoured to check my conclusions by reference to separated localities which have been fairly well collected, Brisbane for example, and I am convinced that there is a good deal more in the differences met with than can be explained in this way, although I fully expect new records and species to turn up, not only in the other States, but in New South Wales also. There is no indication at present as to the mode of origin of the groups and subgroups, but within the series there is considerable evidence of the part played by isolation in the evolution of the species. There are three sets of cases indicating separation for varying lengths of time. There is in the first place the simple effect of long distance separation without any barrier. The modifica- tions at the extreme ends of a continuous distribution would be slow in develop- ment, because of the slight amount of mingling that would inevitably occur. The endemic South Australian species are the best instances of this type and the leucophaea series is probably in part the result of the same process, but of longer duration. Another factor, however, enters here, namely, the development of a barrier which has probably completely isolated the Swanland species and which has consequently allowed them to differentiate from the eastern forms more rapidly than when they were in actual contact. Some such additional factor is necessary in order to account for the unique character of the fauna of Swanland. This barrier is the desert, the country along the Great Australian Bight having become progressively more arid since the close of the Pleistocene. There is no evidence that any of the Australian species are adapted to even moderately arid conditions. There is, however, a good deal of evidence that during the Pleistocene this area was well watered and was probably well suited for the development of Nemestrinids, and the western species may have reached their present home then (or possibly earlier, but there is no indication of what the climate was like earlier than the Pieistocene). Another post-Pleistocene barrier which has led to an interesting development is the separation of Tasmania from the mainland. All, or nearly all, of the Trichophthalmas must have been driven out of Tasmania during the Pleistocene, when the temperature was some ten degrees, F., lower than at present and the climate fairly inhospitable. T. punctata Macq. either survived in Tasmania or invaded it just before its final separation and there developed into the particular subspecies found there. This form has reinvaded the mainland, but none of the mainland species have reached BY. I. M. MACKERRAS. 505 Tasmania since the Pleistocene. The explanation of this is found in a study of the wind-rose for Melbourne, when it is seen that the prevailing winds during the months when the Trichophthalmas are on the wing are from the south, an interesting correlation of distribution with metereological conditions. The third type of isolation is more recent and less complete than those described above. It consists in the splitting of a species into sections separated on different mountain tops, or part on the coast and part on the mountains, ‘the greater environmental differences in the latter case producing a more marked effect. In the first instance the differences found, though apparent enough when dealing with long series of specimens, are not sufficient to warrant even subspecific distinction; T. rufonigra, nov., is a good example of this type, the coloration of the abdomen in the males, and of the wings in both sexes, differing noticeably between specimens from Kosciusko and those from Barrington Tops. The two subspecies of 7. bivitta illustrate the second type. It will be observed that the evidence presented above indicates that many of the species are not earlier than the Pleistocene in age, and this conclusion is supported by the systematic studies, there being every indication that the species are still in a state of flux and have not reached the stable condition seen in many other groups. The genus is an old one, and the groups and subgroups are well stabilized, but the species have apparently been undergoing marked changes and development in fairly recent times, possibly owing to the stimulating climate of the southern half of eastern Australia during and immediately after the Pleistocene. One other question remains to be considered. In many animals distributed along the east coast of Australia, northern specimens differ markedly from the southern, and the same holds good for JT. nigripes Macq., the differences found being dealt with fully under the description of that species. There are indications that further material will reveal a similar state of affairs in some other species. While the effect of isolation has been stressed above, this is by no means the only factor involved in the evolution of the species, but it is the only one of which evidence can be produced at present... It is limited to the instances and conditions described and can hardly be made to apply, for example, to the bulk of the New South Wales species. The other factor or factors are, however, quite unknown. The differences in seasonal occurrence, as well as distribution, between the rosea and the costalis groups have been mentioned above. There is a certain amount of information available as to the seasonal limits of the different species, which it is worth while summarizing here. Most of our data comes from the New South Wales species and especially, in the case of the rosea group, from those species occurring at the Royal Zoological Society’s Biological Station at National Park, where a particular study of this question was made. The season in New South Wales may be divided into two parts with a definite gap between, the month of October being practically a blank as far as Trichophthalmas are concerned. August and September on the coast are entirely taken up by the rosea group, none of the other species appearing at this time. Each species has a season of about four or five weeks, the period of maximum prevalence being much shorter. The species appear and reach their maximum in a definite order. Thus, while three or four species may be taken together at almost any time in these months, the relative numbers of the species taken at different times vary markedly; for example, in the middle of August one sees a great number 506 NEMESTRINIDAE OF THE AUSTRALASIAN REGION, of T. rosea Macq., many T. primitiva Walk. and a few T. eques Schin. and T. albimacula Walk.; in the latter part of September, on the other hand, 7. rosea Macq. and T. primitiva Walk. are only represented by occasional ragged stragglers, while 7. eques Schin. is commoner and 7. albimacula occurs in fair numbers together with 7. ricardoae Licht., which was not seen earlier. The order of appearance is: 7. primitiva Walk., T. rosea Macq., T. albimacula Walk., T. eques Schin., 7. ricardoae Licht., with 7. laetilinea Walk. last, according to the records of this species which I have not taken myself. The season of this group at National Park coincides with that of Epacris microphylla, their favourite food plant in that locality. This is, however, not their only control, for at Woy Woy they feed on Leptospermum sp. and disappear long before the flowers are done. The climatic factor is probably the final control in this group. The costalis and novae-hollandiae subgroups are on the wing in coastal areas from the end of October to the end of January. There is much less information as to succession of species, but the order seems to be—punctata series, costalis series, novae-hollandiae series, with a certain amount of overlapping. In Queens- land, members of the novae-hollandiae subgroup have a much more extended season, the range being from October to April. It must be remembered that this subgroup has perhaps its greatest development in South Queensland and that it is the most adaptable of all the Trichopthalmas to different climatic conditions. For most species, however, the indications are that the time of appearance in Queensland is rather earlier than in New South Wales. On the mountains the season is distinctly later than on the coast, great numbers of Trichophthalmas being on the wing in February. A succession of species may also be found for the mountain forms, but the evidence is not yet sufficient to draw any definite conclusions. Habits. These flies are usually captured feeding at flowers. They are very selective, only one or two species of plant being attractive in any one locality. On the Hawkesbury Sandstone heath in early spring, when there is an abundance of flowering shrubs, practically the only one frequented is Epacris microphylla. Grass-trees (Xanthorrhoea sp.), Leuwcopogon sp. and Sprengelia sp. are occasionally visited for very short times. Crowea saligna was in one locality very attractive to T. eques Schin. In other types of country, such as the low Pleistocene sand- flats at Woy Woy, Leptospermum grows thickly and is very attractive. Members of this genus are also the most attractive plants on the mountains later in the season. On the coast in the middle of summer Kunzia, Angophora and Hucalyptus flowers are the ones visited, but they never attract such numbers as does Epacris microphylla or the Leptospermums. When feeding, the insect grasps the flower, or the stem just below it, with its legs and keeps its wings vibrating rapidly, but not so rapidly as when in flight, fhe note produced dropping markedly in pitch each time the fly touches a flower. This is particularly well illustrated in Pl. 1, Fig. 4. I have never seen Trichophthalmas actually hovering while feeding, the plant in all cases being grasped by the legs. It is very rare, however, to see the wings come to rest. The fly stays two or three seconds at each flower and then quickly flies to another, twenty to thirty flowers being visited in a minute. The feeding may be kept up for several minutes and almost all the attractive flowers in a patch visited before the insect leaves it. They are very easily disturbed by any sudden movement and disappear in a BY I. M. MACKERRAS. 507 flash. This description of the feeding habits indicates the great difficulties encountered by Mr. Nicholson in photographing these insects alive and the patience and dexterity necessary in order to get an exposure. Males are frequently seen playing and hovering in the air when not feeding, but one rarely takes a female at such times. Their loud, high-pitched, characteristic note usually attracts attention to them before they are seen. Their circling and twisting movements, indulged in at irregular intervals whether disturbed or not, are dazzling in their swiftness and their capture would be extremely difficult were it not for the fact that they almost invariably return immediately to the spot where they had previously been hovering. When feeding on flowers, however, if disturbed they disappear in a flash and never return. This hovering habit is not uncommon in other Diptera, but reaches its highest development in the Trichophthalmas. These habits are approached by Trichopsidea oestracea Westw., but are in marked contrast with those of HExeretoneura maculipennis Macq., as was well seen at Barrington Tops, where both Trichophthalma and Exeretoneura were abundant, the latter being a lethargic, slow-moving type and not feeding on the flowers. , Nothing is known of the life history of any of the Australian Nemestrinids. Attempts to find the larvae in situations where the adults are abundant have all failed, numerous Asilid and Therevid larvae being turned up, but nothing that could be assigned to the Nemestrinidae. There is little indication from the habits of the adults as to what to look for, but White’s (1914) record of the habit of the female 7. punctata Macq. of resting on the young wattles may possibly have a bearing on their egg-laying habits. We have seen females of T. costalis Westw. in a sandy locality resting on bracken or hovering and flying about close to the ground, but they did not give any other indication of possibly being about to oviposit. Studies of seasonal distribution strongly indicate that there is not more than one brood each season, no species being known to appear twice in one year. Several females of 7. rosea Macq. were brought to the laboratory alive and in one case eggs were laid, but to date (two months later) have not hatched. They were scattered in little irregular masses all over the lower part of the sides of the jar and on the sand at the bottom. The contrast with the neat egg-masses of Tabanids ovipositing under similar conditions is very marked; this, however, may not be the natural method. The egg (Text- fig. 8, p. 500) is pale creamy-white in colour, without any sculpturing or marking, torpedo-shaped and distinctly more bluntly rounded at one end than the other. The eggs appear to be sticky and adhere to one another, to the sand, and to the sides of the vessel. An average egg is 1.30 mm. long by 0.22 mm. wide. Systematics. The different groups and subgroups are easily recognized, but it is much more difficult to recognize some of the closely related species, especially in the punctata series of the costalis subgroup. While the primary subdivision is on antennal characters, there are sufficient differences in other respects to make the position of an insect clear even when the antennae are missing, and these are given in brackets in the following key, which (except for the genitalia, which are dealt with fully on p. 499) serves also as a definition of the groups and subgroups. It is interesting to note that species from different groups or sub- groups may resemble one another strongly in general appearance, for example, i(LIBRARY Be eee a Mass-C x KE 508 NEMESTRINIDAE OF THE AUSTRALASIAN REGION, T. intermedia, nov. and YT. bancrofti, nov., T. rufonigra, nov. and T. novae-hol- landiae Macq., while T. nigrovittata, nov. and 7. leucophaea Walk. have a super- ficial resemblance to species of the rosea group. The costalis subgroup is fairly uniform, but I have divided it into three series, partly for convenience and partly to indicate the relationships of the species. There are no very marked characters separating these series and a linking form is seen in TJ. intermedia, nov., which has been placed provisionally in the costalis series. The novae- hollandiae subgroup is very uniform, as is also the rosea group, with the exception of T. laetilinea Walk. and T. ricardoae Licht., which form a series apart from the rest and show interesting characters linking them with the costalis group. My costalis subgroup corresponds with Hardy’s Group 1, the novae-hollandiae subgroup with his Group 2, and the rosea group with his Groups 4, 5 and 6. Groups 4 and 5 of Hardy may be reasonably separated as series, but Group 6 (T. primitiva Walk.) should go in with the species of Group 4, to which it is closely related in all bodily characters. Key to groups and subgroups. A. Third antennal segment conical, never divided by a sulcus near its base. (Palpi always obvious, never obscured by hair; wing with M;..4 continuous, or nearly SOs withirest of Oblique: svVeli-)y mcrae cchecleloe eicierescsel koro or onCe: costalis group. a. Palpi slender, second segment short, acuminate, apical pit small ........ Lpabieice Toners: teldst eure letersi ite rcek site. raise, aurehaltantene tertetleneseh sie eM sik elaine Beemer eto e costalis subgroup. b. Palpi stout, second segment as long as first and stouter than it, apical joule lees) incl COWS sooudooscaccoese novae-hollandiae subgroup. B. Third antennal segment long, cylindrical, with well-marked sulcus near its base. (Palpi greatly obscured by hair and vein Mz3.,4 markedly displaced towards wing-root, or abdomen with shining pale transverse markings or spots)......... She OREO: CE eos OMe aor t,t one peter cicero neck Oe Oke o otra eco Gio cuneea Geos rosea group. Key to species of the costalis subgroup. 1.. Brown or greyish nearly concolorous species, abdominal markings irregular if contrasted, palpi moderately slender (costalis series) ................+.-- 3 Fulvous species, with strong contrast between markings and ground colour of abdomen ‘palpilisvery sSlemdier ca omy ters: cio wohnenciiciier tence soe ach sence ac hene uke oReee Role rereiere 2 2. Proboscis more than three times the head length, labellae small and inconspicuous (LEWCODRG@EG, ISCTIES)) cr sisceee acer are coors poe Nereus er oes cays) eaarene eveaetePakene cere Rtene en eae 12 Proboscis at most twice the head length, labellae at least moderately prominent (PUIVECEEARISERIES) yori i cos ere eNO ooh tte ev ohne che ote cee Mec heater phe abs aulteliat 13 3. Abdomen without transversely disposed spots, at most with longitudinal darker TORCH cae ciecetewanatea¥e tn ee eiove Pouetoue es .Wv ols eeaeieutene eilek oils ce Java waptep aU eRe e eRe eR eie Var So OTC ER TE eS 4 Abdominal segments with transversely disposed darker spots, on some segments linked) ‘biyabasak evr ke Pennine Geaireyeceylavrs Geeta ae ob os eee tetra tae tie cece ke eens wel aeeuel eet WIL 4, Thorax grey, abdomen yellowish with irregular median grey stripe .. intermedia, nov. Thorax and abdomen) ,darl= VO GO Wil sorg is eveier eres cuensciicr sie ie ie), cusuc lotions avqeuetencr heats ensicne tenes microns 5 Bs cE US SY Sa ae, al ssesrsiconcw ss vses eta ese rrenceiveuse seta icnccicm forceice) Sea erral elie: Gre encatrer esse Stes etic ite re CMON SCORCH SC eer tee a are SE 6 INEMATES ON Aer asics. eeu ocSisaiia se eenrieasn ierratcala, ate tence eo uae etre nebterte ems We 5 Hovia|ioneyich cleme pnew ohonsTielie enero ee aero nee 9 6. Abdomen with prominent basal band of white hairs .....................+.--- 0 BRIG SLUVC) gM OF AKG Py eecemnt arcntecnonetacoceD hc Oke OrOROR ard Glo coo Oro ceo otto ain pric Geaia Go ueotIGL Oreo: 0-6 ba a:0 Gone 8 7. Apical third of wing clear white, Western Australia ............. costalis apicalis. Brown coloration of wing fading gradually to apex, East Australia ...... phe eee Hela dices BERS Stale ccos terion alte elect ats alco aL geek acl STL teu Sa C PORE NERO RG OLE lene Eos costalis costalis Westw. 8. Very large dark species with grey pleurae. Eastern Highlands ...... rufonigra, Nov. Medium sized species, reddish-brown in colour, pleurae white, South Australia oratlettess gelsa Seiten stbains dat uesst ouch ohrebiowrentan erie eieeroia ar eivahtc Rar toed etre ameanc a os eee Eee re eR subcostalis, nov. 9. Proboscis more than twice the head length, labellae inconspicuous, Western Aistra tai v6 Rodis Silas eon ete ieee Sash Sma ale Gultocltew stoay Sasi atls dace cepa re costalis apicalis, nov. Proboscis less than twice the head length, labellae conspicuous, Eastern Australia... 10 lO Mearcexdarksspeclesmwithmenre van pleuicale ene eIeen eine rein neaene een nent rufonigra, Nov. Medium-sized dull brown species, pleural hairs white ........ costalis costalis Westw. 1 3te 12. 13. 14. 15. 16. ie 18. Ug, 20. BY I. M. MACKERRAS. 509 Very small reddish-brown species, proboscis long, ninth tergite of male as in MMEp a Sales (a9) Gioia la CeO CRON CTI RO MERETCNE CASS EG Oletaiy cca R REL MARA Bat tase a emia Reaper fusca, nov. Medium-sized greyish-brown species, proboscis short, ninth tergite not PDH H SSID A Aa sneto one cccono eiCERISIOINCRC A CECUCEA ET IIE e cic ean RCE Reames oa obscura Westw. All abdominal stripes about equal in breadth .................... leucophaea Walk. Brown side margins and black median stripe very narrow relative to the grey ITUCETINIE GIA LOMA CASA way stey ce-ch eich anc) Vicar or ohae chon Sik SR eet chee is longirostris, nov. Abdomen with narrow median black stripe or row of small spots, side margins and posterior edges of segments never with any dark markings ........... 14 Abdomen with dark median stripe, side margins always, and apical edges of SCAMEAMS Oe, wan Cercle woe OSS geoadasconsvcanevgnuHbuoUuEGoasoUeunHe 15 Thorax brown with two indistinct paler stripes, abdomen orange to drab with small ediainy DlaCK SpOtS.c..: isla 6 Professor H. G. Chapman, M.D., BS. _—- Professor L. Harrison, B.A., B.Se. *. Cheel. tA. F. Basset Hull, . EE Sir T. W. Edgeworth David, K.B.E., A. B. Walkom, D.Se. | en C.M.G.; 'D.8.0., BAZ, D.Se., F.R.S. H. S..H. Wardlaw, DiS, T. Storie Dixson, M.B., Ch.M. Py GRAY See teraenee D:Se.; BE, FES. | Auditor: F. H. Rayment, F.C.P.A. oe see Elected 22nd April, 1925, in place of Professor WES DAs Haswell. se : g t Elected 19th August, 1925, in place ‘of. Mr.: T. Steel. rae + ‘ pee eas 23rd September, rake in Lien of Mr. a R. MeCulloch.. ; NOTICE 47 at. hi With ‘is, eeecabian of Volumes LVI of ‘he PROCEEDINGS —of ahieh: the Seecie stock was totally destroyed in the Garden Palace Fire—the Publications of the Linnean Society of N.S.W.. may be obtained from the Society, ‘Macleay House, oS 16 College Street, Sydney, or from Dulau & Co., 34-36 ee Street, ‘Cavendish — ae: Square, London, W., at the following prrees: = eager « . . FIRST SERIES. - PaaS Proceedings for 1882, Vol. VII. Part i ts 608.3 Part (25) 108.5 Part 3, ce 5 Part. 4, ‘tos. ; ‘Proceedings for 1883; Vol. VIII.—Part 1, 10s.; Part 2, 5s.; > Part 5, Sgn Beunth ae ~88. ee ‘Proceedings for 1884, Vol. IX—Part 1, 8s.; ‘Part 2, 12s.; / Part Ces Se. Ho ‘Part 4s £1 5: Proceedings for 1885, Vol. X.—Part 1, 12s.; Part 2, 7s. 6d.; Part’ 3, OEP w eh iSeeOGe $ sy A ae f SECOND SERIES. ISP ore niet ae te Proceedings for 1886, Vol. 6d; Part 2, 125; Part 3, 13s.; 12s, 6d. : sett SE SMT ‘Proceedings for 1887, Vol. I1l.—Part 1, 7s.; Part 2, 8s.; Part), 128; : Part. Ae ERS ASS: Wig gus: Proceedings for 1888, Vol. IiI.—Part 1, fee Parts) 21 4s.; > Part. oh £1; Part 4, 18s.~ ‘Proceedings for 1889, Vol. IV.—Part 1, 11s:; Part 2, 168.2 Part. 3,- 19s. 7 Bare. 4, hs, Proceedings for 1890, Vol. V.—Part 1, lis.; Part 2, 9s. Part 3, 9s. 5 Pa 1 4) “9s. Si Proceedings for 1891, Vol. VI.—Part 1, 10s. ; Re Part 2; 9s. 6d.; were 3, 17s. i Part 4 Tsj6deuets ‘ Proceedings for 1892, Vol. VII:—Part 1; 6s. 6d.; Part 25 ven 6d.; ‘Part. a ase Op art. 4, = Proceedings for 1893, Vol. VIIL—Part 1, 5s.; 5 Part 2) Lis Part 3; 68.5 Part 4,98. | Proceedings for 1894, Vol. IX.—Part 1, 12s.; Part 2, 12s. ; > eee 3, 18s.3 5 Part 4 8s. rt 2, 8s. ba; Part 3) 308 anne as. 64.; Dart a 2, 6s, éd.; Part 3, 1s. ba; < a: Bee 20 9els Past & igs, con Part 3, a ; Part 2 Te ec — Pe ; Part 3, eee Sparh ‘ei tes: 98.5 Part. Med OSS. = Part * “3, 12s. 6d; 2, 18s. 6d; Part 3, 1s. 6433 Part ia, Lbs. oi Ls 6d. re Part 3, 15s. ; Part 4, “18s. oy Be ig Aa oe ee zs Part 4, ‘12s, bee Bites 6d.; Part — 1 Part ate 18. Ge i 4, os 6d, ‘Part re 18s. “64. on Cae: Rar we sioee ae “4 it neyal 4to. LI. “and 308. ay 5 AEA te Wik ane ae Oe “Abstract of Proceedings ‘ Sif te ‘ _ Donations and’ Exchanges ._ a i mtd ra ee aah i " s vl P tha, % ist of Members” ~~ tL OI LIBRARY MBL/WH ti D | Ih Mh LADW WH + ba | Ae Weis: Sih ' TEA Ae By Fore ate BA hae eens ay tot heey . Te é + i hay o neat Fpas " aS, a4. ‘ of hh r i oi yaaa . ¥ iy oenaeart Hiat e a, at iat care ape a Lah