THE ANNALS AND MAGAZINE OF NATURAL HISTORY, INCLUDING ZOOLOGY, BOTANY, and GEOLOGY. (UEINO A CONTINUATION OF THE 'ANNALS ' COMBINKD WITH LOUDON AND CHARLESWORTH's ' MAGAZINE OP NATURAL HISTORY.') CONDUCTED BY ALBERT C. L. G. GUNTHER, M.A., M.D., Ph.D., F.R.S., WILLIAM S. DALLAS, F.L.S., WILLIAM CARRUTHERS, F.R.S , F.L.S., F.G-.S., AND WILLIAM FRANCIS, Ph.D., F.L.S. VOL. XIV.— FIFTH SERIES^ '^\-(Sonian LONDON: PRINTED AND PUBLISHED BY TAYLOR AND FRANCIS SOLD BY LONGMANS, GREEN, AND CO.; SIMPKIN, MARSHALL. AND CO.; KENT AND CO.; WHITTAKER AND CO. : BAILLIERE, PARIS : MACLACHLAN AND STEWART, EDINBURGH : HODGES, FOSTER, AND CO., DUBLIN : AND ASUER, BERLIN. 1884. " Omncs res creatas sunt divinae sapientite et potentiie testes, divitiae felicitatis humance : — ex barum usu bonitas Creatoris ; ex pulchritudine sapientia Domini ; ex oeconomia in conservatione, proportione, renovatione, potcntia majestatis elucet. Earum itaque indagatio ab bominibus sibi relictis semper sestimata ; a rere eruditis et sapientibus semper exeulta ; male doctis et barbaris semper inimica fuit." — Linx/EUS. "Quel que soit le principe de la vie animale, il ne faut qu'ouvrir les yeux pour voir qu'elle est le cbef-d'eeuvre de la Toute-puissance, et le but auquel se rappor- tent toutes aes operations." — Bruckner, Tkeorie du Systeme Animal, Leyden, 1767. Tlie sylvan powers Obey our summons ; from tbeir deepest dells The Dryads come, and throw their garlands wild And odorous branches at our feet ; the Nymphs That press with nimble .step the mountain-thyme And purple heath-flower come not empty-handed, But scatter round ten thousand forms minute Of relvet moss or lichen, torn from rock Or rifted oak or cavern deep : the Naiads too Quit their loved native stream, from whose smooth face They crop the lily, and each sedge and rush That drinks the rippling tide : the frozen poles, 'WTiere peril waits the bold adventurer's tread, The burning sands of Borneo and Cayenne, All, all to us unlock their .secret stores And pay their cheerful tribute. J. T.VYLOR, Norwich, 1818. ^^^ CONTENTS OF VOL. XIV. [FIFTH SERIES.] NUMBER LXXIX. Page I. On th:3 Cri/ptoniscidce. By Prof. R. Kossmanx 1 II. On the Spongia coriacea of Moutagu, = ie«co.s-o/e/i!'a coriacea, Bk., together with a new Variety of Leucosdenia lacunosa, Bk., elucidating the Spicular Structure of some of the Fossil (Jalcispougis; followed by Illustrations of the Pin-like Spicules on Verticillites helvetica, De hoviol. By II. J. Carteu, F.R.S. &c. (Plitel. ) .... 17 III. Some Remarks upon the VarialDiiity of Form in Luhomirskia baicalensis, and upon the Distribution of the Baikal Sponges in general. By Dr. W. Bybowski. With P.SS. by H. J. Carter, F.R.S. &e. (Plate II.) ' 29 IV. Descriptions of three new Species of Moths from the Island of Nias. By Arthur Gr. Butler, F.L.S., F.Z.S., &c 84 V. On the Systematic Position of the rulicidce. By Dr. Karl Kkapelin. (Plate III.) 36 VI. New Investigations on the Development of the Viviparous Aphides. By Dr. Otto Zacharias 54 VII. Notes on the South-Russian Spongillidce. By Dr. W. Dy- BOWSKI 58 VIII. On the Synonymy of some Heterocerous Lepidoptera. By Rudolph Rosenstock, B.A 63 The System of the Munactine.llidce, by Dr. R. von Lendeufeld ; On Orbulina universa, by M. C. Schlumberger ; On the Ascidian Genus Rhopalea, by M. L. Roule ; On the Process of Diges- tion in Salpa, by Dr. Ch. S. Dolley ; On a Species of Tachina occurring on the Tracheal System of Carahns, by M. N. Cliolod- kowskv 65 — 74 {y CONTENTS. NUMBER LXXX. Page IX. Notes on Species of Ascodictyon and Hhopahmaria from the Wenlock Shales, By Geoege Robert Vine 77 X. Descriptions of two new Species of WaJchenaera, Blackw. Bv the Rev. O. P. Cambeidge, M.A.,C.M.Z.S., &c. (Plate IV.) . . .'. 89 XI. A necoMdilsoiQon Pentastvmumj'iolijzomcm. By F. Jeffeey Bell, M.A " 92 XII. The Causes of Variation. By Romyn Hitchcock 93 XIII. Additions to the present Knowledge of the Vertebrate Zoology of Persia. By Jawes A. Mveeay 97 XIV. Additions to the Reptilian Fauna of Sind. By James A. Mtjeeay 106 XV. On a new Species of Ljieopodites, Goldenberg (L. Stockii), from the Calcifevous Sandstone Series of Scotland. Bv R. Kidston, F.G.S. (Plate V.) .' Ill XVI. Synopsis of the Families of existing Lacertilia. By G. A. BOULENGEE 117 XVI]. Bescription of a new Species of Pfeudacrcea from Natal. By Aethur G. Butlee, F.L.S., F.Z.S., &c 123 XVIII. Moas and Moa-hunters. By A. de Qfateefages 124 XIX. On the Presence of Eyes and other Sense-Organs in the Shells of the Chitonid(P. By II. N. Moseeey, M.A., FR.S., Linacre Professor of Human and Comparative Anatomy in the University of Oxford . . 141 On the Submaxillary in Masticating Insects, by M. J. Chatin ; On a new Type of the Class Hirudince, by MM. Poirier and A. T. de Rochbriine ; On a new Type of Elastic Tissue observed in the Larva of Erhtalis, by M. H. Viallanes 147 151 NUMBER LXXXI. XX. On some Peculiarities in the Geographical Distribution and in the Habits of certain Mammals inhabiting Continental and Oceanic Islands. By G. E. Dobson, M.A., F.R.S 153 XXI. Moas and Moa-hunters. By A. de Quateefages 159 XXII. On two Species of Alveolites and one ci Amplexopora from the Devonian Rocks of Northern Queensland. Bv Robert Ethe- RiDGE, Jun.. and AsTHtTB H. Foord, F.G.S. (Plate VI.) 175 CONTENTS. XXIII. Crustacea of the ' Albatross ' Dredgiiig-s in 1883. By Sidney I. Smith ". 179 XXIV. Notes on Sponges, with Description of a new Species. By Stuart 0. Ridley, M.A., F.L.S., &c " 183 XXV. On the Hard Structures of some Species of Madrepora. By Prof. P. Martin Duncan, F.R.S., V.P.L.S., &c 183 XXVI. Contributions to a Knowledge of Malayan Entomology. Part III. By W. L. Distant '. 198 XXVII. On the Rate of Development of the Common Shore-Crab (Carcimis mcenas). By George Brook, F.Ij.S. (Plate VII.). . . . 202 Proceedings of the Dublin Microscopical Club 207 — 212 New Books : — An Elementary Course of Botany, Structui-al, Phy- siological, and Systematic. By the late Prof. Arthur IIen- • FREY, F.R.S. F.L.S., &c. Fourth Edition. By Maxwell T. Masters, M.D., F.R.S. , F.L.S., assisted by A*. W. Bennett, M.A., B.Sc, F.L.S. — Second Annual Report of the United- States Geological Suryey to the Secretary of the Interior, 1S80- 81. By J. W. Powell, Director \ 213, 21 7 On Floral Polymorphism in Narcissus reflexus, by M. L. Crie ; Anatomy of Epelra, by M. Vladimir Schimkewitsch ; On the Physiology of a Green Planariau [Convohila Schultzii), by M. A. Barthelemy 220 222 NUMBER LXXXII. XXVIII. The Classificatory Position of Honiaster elongatus, Dune. & Sladen : a Reply to a Criticism by Prof. Sven Loyen. By Prof. P. Martin Duncan, F.R.S. , and W. Percy Sladen, F.L.S 225 XXIX. Aspects of the Body in Vertebrates and Arthropods. By A. S. Packard 243 XXX. A Contribution to the Knowledge of the Freshwater Sponge Dosilia Stepanoicii. By Dr. M. Dybowski 249 XXXI. Triassic Insects from the Rocky Mountains. By Samuel H. SCUDDER '. 234 XXXII. On the Affinities of the Onchidia. By Dr. R. Bergh . . 259 XXXIII. On a new Species of the Theclid Genus Thenfas from Colombia. By Arthur G. Butler, F.L.S., F.Z.S., &c 267 XXXIV. Some Phenomena in the Life-history of CiathruUnn eJeyans. By Miss Sara Gwendolen Foulke 2^ VI CONTENTS. Page XXXV. On Astylospongidce and Anomocladina. By Karl A. ZiTTEL 271 XXXVI. Contributions towards a General History of the Marine Polvzoa. Bv the Rev. Thomas Hincks, B.A., F.R.S. (Phxtes Vlil. & IX.) 276 New Books: — Memoirs of the Geological Survey of India. Paheon- tologia Indica, bein.o: Figures and Descriptions of the Organic Remains procured during- the progress of the Geological Survey of India. Series x. Indian Tertiary and post-Tertiary Verte- brata. Vol. II. Part G. Siwalik and Narbada Caruivora. By R. Lydekker, B.A., F.G.S., F.Z.S.— Report on the Zoological Collections made in the Indo-Pacific Ocean during the Voyage of H.M.S. ' Alert,' 1881-82 285, 291 Proceedings of the Geological Society 294 — 296 On the Copulation of Diffiitgia glohulosa, Duj., by Dr. Carl F. Jickeli ; How Lycosa fabricates her Round Cocoon, by Dr. H. C. McCook ../. '. . . 297, 298 NUMBER LXXXIII. XXXVII. Ophryocj/stis Biitschlii, a Sporozoau of a new Type. By AiME Schneider. (Plate X.) . . . •. \ 301 XXXVIII. Descriptions of Palaeozoic Corals in the Collections of the British Museum (Nat. Hist.). — No. II. By Robert Etheridge, Jun., and Arthur II. Foord, F.G.S. (Plate XL) 314 XXXIX. Diagnoses of new Species of Plcurotomidce in the British Museum. By Edgar A. Smith 317 XL. The Auditurv and Olfactory Organs of Spiders. By Fried- rich Dahl. (Plate XII.) 329 XLI. De.scription of a new Species of Microijule. By Oldfield Thomas, F.Z.S., Natural History Museum .....' 337 XLII. Notes on the Palaeozoic Bivalved Entomostraca. — No. XVII. Some North-American Leperditice and allied Forms. By Prof. T. Rupert Jones, F.R.S., F.G.S 339 XLIII. Description of a new Species of the Coleopterous Fiimily CetoniidcB from Madagascar. By Charles 0. Waterhouse 343 XLIV. On a Polythalamian from the Salt-pools near D^vh, in Transylvania. By Dr. Eugen von Daday oJ9 CONTENTS. VII Page Neiv Books: — Vergleichende Morphologie und Biologie der Pilze, Mvcetozoen und Bacteriea. Von A, de Baby. — Our Insect Allies. By Theodokk Wood 363, 3G6 Note on the Occurrence of some rare Foraminifera in the Irish Sea, by Charles Elcock ; On the Occurrence of a Process resembling Copulation in Comatida mediferranea, by Dr. C. F. Jickeli ; On the Organization of Anchinia, by M. N. Wagner ; On the Anatomy of the Tiiroghipld, by Dr. Alfred Naleper ; On the Luminosity of the Glow-worm {Lanipyris splendidula), by M. Wilhelm Kaiser 366—372 NUMBER LXXXIV. • XLV. Description of an Impregnated Uterus and of the Uterine Ova of Echidna hystrix. By Sir Bichard Owen, K.C.B., F.R.S., &c. (Plate XIII.) ' 373 XLVI. On the Coleopterous Genus Macrotoma. By Charles O. Waterhouse ' 376 XLVII. Notes on Batrachians. By G. A. Boulengeb 387 XLVIII. Notes on the Palseozoic Bivalved Entomostraca. — No. XVIII. Some Species of the Entomididce. By Prof. T. Rupert Jones, F.R.S., F.G.S. (Plate XV.) 391 XLIX. On new Species of l^epidoptera recently added to the Collection of the British Museum. By Arthur G. Butler, F.L.S., F.Z.S., .*tc : 403 L. Descriptionsof two new Moths from Madagascar. By Arthur G. Butler, F.L.S., F.Z.S., &c 407 LI. Local Colour-yarieties of Scyphomedusce : a new Species pro- duced in Forty Years. By R. von Lendexfeld, Ph.D 409 LII. Notes on Hawaiian Neuroptera, with Descriptions of new Species. By the Rev. Thomas Bi.ACKBURN, M.A 412 Lilt. Contributiuus to our Knowledge of Hydromedusa, a Genus of South-Americnn Freshwater Turtles. By Dr. A. Gunther, F.R S. (Plate XIV.) .' 421 LIV. Note on some East- African Antelopes supposed to be new. By Dr. A. Gunther, F.R.S , 425 LV. Description of a new Species of the Carabideous Genus Callistomimus. By Charles O. Waterhouse 429* LVI. Description of a new Species of Julodis (Coleoptera, Bupres- tidae). By Charles O. Waterhouse 429 vm CONTENTS. Pago New Books : — Journal and Proceedinirs of the Roval Society of New Soutb Wales for 1833. Vol. XVII. Edited by Prof. A. Liver- siDGE, F.R.S. — Internationale Zeitschrift fiir allgemeine Spracli- wissenscbaft. Edited by F. Techmer 430, 431 Proceedings of the Geological Society 432 — 437 On PalucUcella erecta, by Mr. E. Potts ; On a new Insect of the Genus Phylloxera (Phi/Uoxem scdicis, Licht.), by M. J. Lichtenstein. 437, 439 Index 44-0 PL.\TES IN VOL. XTV. Platk I. Leucosolenia lacunnsa. — Pin-like spicules on Verticillitos. II. Luboniirskia baicalensis. III. Systematic position of the Pulicidas. IV. New species of Walckenaera. V. Lycopodites Stockii. — Lepidodendron rimosura. VI. New species of Alveolites and Amplexopora. VII. Development of the Common Shore-Crab. VUI 1 . jY > Marine Polyzoa. X. Ophryocystis Biitschlii. XI. New Palfeozoic Corals. XII. Auditijry and olfactory organs of Spiders. XIII. Female organs and uterine ova of FJchidna hystrix. XIV. Hydromedusa platanensis. XV. Ne-w species of Entomididse. THE ANiNAI.S AND MAGAZINE OF NATURAL HISTORY. [FIFTH SERIES.] " per litora spargite museum, Naiades, et cireiun vitreos coiisidite fontes : PoUiee virgineo teneros h'lc carpite flores : Floribus et pictum, divse, replete canistrum. At vos, o N3aiiphffi Craterides, ite sub undas ; Ite, recurvato variata corallia trunco Vellite muscosis e rupibus, et mihi conchas Perte, De£e pelagi, et pingui conchylia sucoo." N.PartheniiGiaHiieftasii Eel. I. No. 79. JULY 1884. I, — On the Cryptoniscidaj. By Prof. R. KOSSMANN *. For tlie completion of my investigations upon the parasitic Isopods living Tipon Crustacea, the Epicaridia, the Royal Academy of Sciences granted me last autumn a considerable travelling stipend. The completion of my labour of many years has been brought into the immediate future by this liberality, for which 1 here with pleasure express my gratitude ; but its publication will nevertheless occupy so much time that a preliminary communication of the more important results of this last journey would seem to be justified. This journey was devoted to the investigation of the Cryp- toniscidffi, and therefore of that subdivision of the Epicaridia which has been least studied, and which, on account of its extensive retrogression, presents the greatest difficulties to investigation. In the year 1 843 Rathke f, in his " Beitriige zur Fauna Norwegens," described, under the name of Liriope pygmcea, * Translated by W. S. Dallas, F.L.S., from the ' Sitzungsbericlite der k. preuss. Akademie der Wissenschaften,' April 24, 1884, pp. 457-47o. ■j- Ratbke, " Beitrage zur Fauna Norwegens," in Nova Acta Acad Leop. Carol, xx. p. 60, tab. i. tigs. 8-12. Ann. & Mag. N. Hist. Ser. 5. Vol. xiv. 1 2 Prof. R. Kossmann on the Ciyptoniscidje. a minute creature which he had found in the cavity of the mantle of Peltogaster, which was discovered by him at the same time, a cirripede parasitic upon hermit-crabs; he reo-arded it as an Amphipod which had been swallowed by the Peltogaster. In the same year, in an article on the sexual characters of the Cirripcdia *, Goodsir published his discovery of a similar animal which occurred in Balanus halanoides^ and which the author regarded as tlie male of that cirripede, the herma- phrodite nature of which was not then placed beyond doubt. In the course of a dissertation upon Peltogaster, Steenstrupf found occasion to speak of Lirioije, and indicated in a some- what doubtful manner that it might possibly be a Bopyrid living upon Peltogaster. In the same year Darwin \ corrected Goodsir's statement, recognizing the latter's supposed male Balanid as a parasite, and, indeed, as belonging to the " loniens," i. e. to the Bopyridge. About the same time (text 1852, figure 1S55) Dana§ described under the name of Cryptothir minutus, as the male, a similar parasite from Creusia, and referred it, with Ltriope, to the Tanaid^e. Lilljeborg || then (1861) came back upon this Ltriope, and also showed that Liriope was an Isopod, and, indeed, a Bo- pyrid. He succeeded particularly in this, that besides the larviform and excessively minute creatures that Rathke and Dana had found, he also discovered the mature female form of Liriope, and he came to the conclusion that the former were the young males. Buchholz^ arrived at a somewhat different conclusion in some respects. Without knowledge of the statements of Goodsir and Darwin, he carefully described anatomically the animal observed by them. He regards the forms found by himself, so far as they look like Liriope, as older larva3, and * Goodsir, " On the Sexes, Organs of Reproduction, and Mode of Deve- lopment of the Cirripeds," in Edinb. New Phil. Journ. xxxv. p. SS, pis. iii. and iv. t Steenstrnp, " Bemarkuinger om slaegteme Pachyhdella og Peltogaster" in Oversigt af Danske Vidensk. Selsk. Forhandl. 1854. X Darwin, ' MouogTaph on the Subclass Cirripedia ' (Ray Society), vol. ii. p. 271. § Dana, ' Crustacea in U. S. Exploring Expedition under Commander Wilkes,' vol. ii. p. 801, Atlas, pi. liii. fig. G. II Lilljeborg, '■^Liriope et Peltoc/ader^' in Nova Acta Soc. Sci. Upsal. ser. 3, vol. iii. p. 1, and Suppl. p. 73 (see * Annals,' ser. 3, vol. vii. p. 47). ^ Buchholz, " Ueber Hemioniscus balani," in Zeitschr. fiir wiss. Zool. Ed. x\'i. p. 303. Prof. R. Kossinann on tlie Ciyptoniscidag. 3 the male as still unknown. Nevcrtlieless he was the first to give us an idea of the adult parasite, as Goodsir had only figured the head and the first four thoracic segments, but neglected all the remaining deformed part of the animal. A further enlargement of our knowledge upon this group was furnished in the year 1871 by a memoir of Fritz Miiller's*. Under the name of Cryptoniscus iDlanarioides he described a parasite resembling Lirio'pe^ but which displaces its host, a Peltogaster^ from the hermit-crab ; only the roots of the Pelto- gaster are preserved, and are apparently made use of by the Cryptoniscus for its own nourishment. Soon after the appearance of Miiller's memoir, and without any acquaintance therewith, I myself published notes f ujDon some forms belonging to this group from Semper's Philippine collections, namely an internal parasite from Sacculina jnsi- formis^ which I named Eumefor liriopides^ an external para- site on the abdomen of a rorcdlana {Zeuxo porcellancv) ; another on the head of an Alpheus [Zeuxo alphei) ; and, lastly, a parasite from the brood-cavity of a Bopyrus [Cahira lerna'odiscoides) . Any exact anatomical investigation of these forms was impossible, as I had only single spirit-speci- mens of them. Passing over Hesse's unscientific and useless memoirs in the ' Annales des Sciences Naturelles,' I come to the most recent and important work upon this subject, namely Fraisse's memoir " Die Gattung Crypto^iiscuSy Fr. Miiller " |. To its contents I shall have to refer repeatedly. There has long, as we have seen, been a general conviction of a near relationship between the Bopyrida3 and Cryptonis- cida3. When, therefore, I formed the plan of undertaking a monographic description of the Bopyridffi, I could not think of passing over the Cryptoniscidee. Moreover, there were points of diilerence between the statements of Buchholz and Fraisse and the results of my own occasional investigations, which made a thorough testing of the investigations of these two meritorious observers appear indispensable. I have therefore twice, in 1882 and 1883, thoroughly studied Hemioniscus halani in Christiansand itself, and in the autumn of 1883, with the aid of the Royal Academy which * Fritz Miiller, " Bruchstiicke zur Naturgeschichte der Bopyriden," in Jenaisclie Zeitschrift fiir Medicin und Naturwissenschaft, Bd. vi. p 01 Taf. iv. figs. 12-19. ■ ' t Kossmann. " Beitrjige zur Anatomie der schmarotzenden Ranken- fiissler, Anhang," iu Arbeiten aiis dem zool.-zootom. Institut der Uuiv. Wiirzbiirg, Bd. i. X Arbeiten aus dem zool.-zootom. Inst, der Univ. Wiirzburg, Bd. iv, 1* 4 Prof. R. Kossmann on the Cryptoniscida3. has been already gratefully mentioned, Ci-yptoniscus paguri at Malion, in the island of Minorca ; at other opportunities I have also examined single specimens of nearly all the forms pertaining here. Upon these investigations the following brief expositions are founded. But before I enter upon the actual subject, a formal neces- sity presses itself upon me. As in my monograph I must, of course^ employ a critically sifted and correctly established nomenclature, and yet must not lay too great difficulties in the way between the preliminary and the detailed work, I think I ought here to give a small table relating to this matter. I. Parasites on Cirripedes. a. Upon non-parasitic Cirripedes {Balanus balanoides and Creusid). 1. Cryptotiiir, Dana, 185:3 (U. S. Expl. Exped., Crustacea, p. 801). Synonymy: — Hemionisctis, Buchbolz, 1806. Cryptothiria (p. parte), Spence Bate & Westwood, 1868. b. Upon parasitic Cirripedes. a. Free in the mantle-cavity. 2. EuMETOR, Kossmann, 1872 (^Beitr. z. Anat. d. sclimar. Rankenfussler, Anhangj. /3. Perforating the mantle from within. 3. LrRiopsis, Max Schultze, 1859 ('' Anm. zu einer Aufsatz von Fritz Miiller," in Wiegmann's Archiv, Bd. xxv. p. 310). Synonymy: — Lirinpe, llathke, 18-13, nom. prseocciip. (Lesson, 1837, Trachy medusffi) . (This form is referred by F. Miiller and Fraisse to O'yp- toniscus, by Spence Bate to Cryptothiria). y. Attacking the peduncle from without, and displacing the cirri- pede down to the roots. aa. Upon Peltogaster, with short cephalon. 4. Cryptoniscus, Fritz INIiiller, 1870 (Jena. Zeitschr. Bd. vi. p. 01). /3^. Upon Sareulina, with long cephalon penetrating deeply into the Brachyuran. 5. Zeuxo, Kossmann, 1872 (Beitrage z. Anat. d. schm. Rankenf.), nom. prseoccup. by Templeton, but suppressed by Dana in favour of Tanms. II. Parasitic on Ostracoda. 0. Cyproniscus, Kossmann, gen. nov. Referred by Sars (" Oversigt af Norges Crustaceer," in Christ. Vidensk. Selsk. Forhaudl. 1882, no. 18, p. 73) to Cryptothiria. "III. Parasitic on Isopoda. 7. Cabirops, Kossmann, gen. nov. Synonym : — Cahira, Kossmann, 1872 (Beitrage &c., Anhang), nom. preeoccup. (Treitschke, 1825 Cahera, Jodoffjsky, 1837 Cabira, Lepidoptera). Referred by Sars (Oversigt &c. p. 74) to Cryptothiria. Prof. R. Kossmann on the Cryptoniscidai. 5 What from the first caused the greatest difficulty was the ascertainment of the sexual relations. The form first described by Rathke with eyes and natatory feet has been interpreted sometimes as the male, sometimes as the larva. Rathke himself evidently regarded his animal as adult, but without deciding any thing as to its sex. Dana took the corresponding form, which he found in Greusia^ for a male, but made no remark about its age. Lilljeborg's discovery proved that the animals described by Rathke and Dana were young forms ; and it seemed to Lilljeborg impossible to regard them as young female forms, because, although already settled upon the host, they showed no commencement of the trans- formation into the adherent female form. He further com- pared them with the youngest male Bopyrides found by Kroyer, and came to the conclusion that they were immature males. Buchholz regarded this same form (from Balanus balano'ides) as an old larva without recognizable sex ; he found no males, but at the same time declared that the sexually mature animals [Gryptothir balani) found by him were deci- dedly not hermaphrodite. Spence Bate, who had already seen and named the young animal *, thought, in 1868 f, it might perhaps be a male, and adds to the word " immature " a note of interrogation within brackets. He therefore doubted as to the immaturity of the animal, but without in any way ])roving that it was a male, still less a mature one. Lastly, Fraisse J asserts with almost perfect certainty that the copulation must take place " in the stage preceding- attachment," and accordingly describes both male and female animals of this stage (representing Rathke's Liriope)^ of which, however, he regards only the former as sexually mature. But his proof would not satisfy most readers. Thus, as regards the males, the testes are scarcely indicated in Fraisse's figure ; their form, aperture, or even structure he has not described at all ; the semen, which was squeezed out by crushing the animal, is not removed beyond the reach of doubt, on account of the mode in which it was obtained and the statements as to the form of its elements, and indeed it is rendered absolutely suspicious by the fact that Fraisse sup- posed he saw it also in the body-cavity of the female attached to the ovaries, where its presence may be pronounced to be impossible. Fraisse found these supposed males swimming * Speuce Bate, ' Report of the British Association,' 1860, p. 225. t Speiice Bate and Westwood, ' History of the British Sessile-ejed (hiistacea,' vol. ii. p. 267. \ Fraisse, " Die Gatlung- CryplvnisvusP he. cit. p. 23, Taf. xv. fig's. -SO aud 32. 6 Prof. R. Kossmann on the Ciyptoniscidse. freely ; but he observed exactly similar animals creeping or swimming about close to the SaccuUna, and these he regarded as females. He saw no genital apertures, or, indeed, genital organs of any kind in them, but observed their conversion into the adherent animal, which undoubtedly is of the female sex. As, however, in this, in a stage which is already strongly metamorphosed, the ovaries are still quite immature, the female larva, according to Fraisse's own opinion, can only be quite immature. Nevertheless, although he had " no oppor- tunity " of observing the copulation, and did not even himself see " the spermatophores adhering to the female," he is of opinion that this immature larviforra female must already be fecundated. " Thus, therefore," he writes, " the fecundated female attaches herself, while the male retains its form and probably perishes after the act of copulation." Evidently all that he has seen (or not seen) is opposed to Fraisse's own opinion, and tends to show that the male is indeed sexually mature and copulates in the above-mentioned larviform stage, but that the female becoming sexually mature at a much more advanced stage is also sought out and impregnated by the male only in this sessile condition. That the male is really larviform I can positively assert from careful examination of such stages. We can very dis- tinctly recognize the genital apertures at the base of the last pair of pereiopoda ; we find the mature testes in the trans- verse section, and witness the brisk movement of the sperma- tozoids. In the GryptoniscidcB, therefore^ the mature male is larviform and still famished with natatory feet upon the pleon. That the female is coupled before its sexual maturity there are no observations to show, and nothing warrants any such supposition. But that on arriving at sexual maturity it is sought out and copulated by the male is supported by the observation of Buchholz, who writes with regard to Crypto- ihir [Hemioniscus] : — " Nearly in every Balanus which con- tained one of the sacceiform animals there occurred one or more small, elongated, brownish animalcules" (here follows the description of Rathke's Liriope-ioxm) ; as also by Fraisse's own figure (Taf. xii. fig. 1), in which such a larva is shown clinging to the metamorphosed female ; and, finally, by the fact that, in 1872, 1 found such a male animal, as I then supposed, in the Philippine Eumetor Uriopides, and in 1883 in a Neapo- litan species of the same genus, three of them, and, indeed, clinging fast to the female. The natatory power of these males makes it very easy to understand that we do not always meet with them with the females as among the Bopyridai ; they probably often spontaneously quit the female, and are Prof. K. Kossmaim on the Cryptoniscidce. 7 certainly still more frequently missed by the observer. In short I acce]}t it as proved that the female is copulated only in the metamorphosed state. We have therefore free-swimming larviform males, and adherent, strongly retrograded females, which copulate with each other. But this is not all. My investigations furnish the most convincing indications that the two forms are only different stages of development of the same individual ; in other words, that among the Cryptoniscidai we have to do with di iJTotandrous hermaphrodism. According to the inves- tigations of Bullar and Paul Mayer such a thing is not unpre- cedented among Isopoda, but rather undoubtedly recognized among Cymothoida^ ; there, however, the sexual maturity of the male occurs much later, after the biramose natatory feet upon the plcon, characteristic of the larva, have already become transformed into branchial feet. Protandry with larval sexual maturity was previously entirely unknown. Of course it has not been possible for me to trace one and the same individual tlirough its whole course of development; to observe how, as male, it performed the duties of its sex, and how, after the metamorphosis had taken place, it was copulated as a female, and produced eggs. I can therefore, as already stated, only bring forward evidence in favour of my assertions. First of all negative : — T have never found a free-swimming Cryptoniscid larva of the last stage that had not male-deve- loped sexual glands. I have found and examined of Grypto- thir halani a considerable number, and of Eumetor three — all males. Fraisse, indeed, describes female larvffi, but these were already fixed, and not truly female, but immature, neuter ; the free-swimming examples of which he found some, were males. It is clear that the protandry is proved, unless we succeed in discovering females or neuters in the same stage of development as the males, hitherto exclusively found. A second piece of negative evidence is the following : — While in all Bopyridaj s. str., and in the Entoniscidtc the male becomes sedentary and remains with the female, it is in all Cryptoniscidse free-swimming and exceedingly active, and is often no longer to be met with near the fecundated female. How should this difference between two so nearly related groups be explicable if the male had not still another task to fulfil elsewhere after the fecundation of the female ? and what other can it be except that of itself growing into a female upon another host ? Finally, we have a piece of positive evidence in the presence in the mature female of a gland which is to be regarded almost 8 Prof. R. Kossmann on the CrjptoniscidiB. with absolute certainty as a retrograded testis. This gland was discovered by Buclihok* in Cryptothir halani, but is wanting in none of the Cryptoniscidge examined by me, while no female Bopyrid or Entoniscid possesses even the smallest trace of it. 13uchholz thought that this gland was to be re- garded as an accessory organ of the sexual parts ; notwith- standing repeated endeavours he could discover no efferent ductj but he found the granulated contents to be like those of the terminal section of the female genital duct. His figure and description represent the organ as a thin cord inflated in three places, situated on three sides above and outside of the ovaries, and filled with a finely granular substance. The inflated parts form cellular diverticula. Now this organ extends through the last three segments of the pereion, therefore the same in which the testis is situated in the male. These three segments of the larva are widely separated immediately before tlie last change of skin, and afterwards enormously enlarged by colossal reception of nutriment. At the same time, however, the triple division continues recognizable by transverse constrictions. If there- fore a testis no longer in function be present, what is more natural than that this should retain its original thickness in the three segments, but become stretched into a thin cord in the intervals — in short, acquire the form represented by Buch- holz ? and is not also the deflciency of the efferent duct, ascer- tained by me, in favour of the view that we have to do with an organ no longer in function ? Lastly, as regards the con- tents. In my transverse sections these appear as a finely granular, strongly refractive, and very strongly colouring- detritus, which in these three properties is absolutely similar to the contents of the testes in the males of the Bopyridae (provided there is no semen in them). From all this we may regard it as almost proved to demon- stration that the CryptoniscidjB are really protandrous herm- aphrodites, in which the testis attains its maturity in the final larval stage, and is then visible in the mature female as a rudimentary organ without an efferent duct. The question now arises, what circumstances have operated for the production of this kind of hermaphrodism, seeing that the Bopyridge, which are so nearly allied, are certainly not hermaphrodites ? We generally find hermaphrodism especially in slow-moving or adherent animals, and to these it gives an essential advan- tage in the struggle for existence. In the first place, it renders * Bucbholz, loc. cit. p. 316, Taf. xvi. figs. 2 & 3, G. Prof. R. Kossmann on the Ciyptonisciclre. 9 self-fertilization }30ssible in cases where a meeting of two individuals can occur only with difhcultj or not at all. Now- self- fertilization is certainly in other respects injurious ; as intensified incestuous breeding it has an exceedingly corrupt- ing influence upon the organization of the race. It is no doubt for this reason that in sonic hermaphrodite groups of animals the habit of mutual fecundation has been brouglit about. In this certainly the above-mentioned advantage disappears ; the contact of two individuals is, as in other cases, necessary. On the other hand, another advantage results from it, namely, when the comparatively obstructed chance of contact does occur, then at least two individuals are fecundated, and consequently there exists twice as much probability of the preservation of the species as if the animals were not herm- aphrodite. But if in this way hcrmaphrodism with mutual fecundation may in certain species replace that with self-fertilization with favourable results, this applies only to such as move with difficulty and not to sedentary species. In the latter, contact, and consequently mutual fertilization, is impossible. Here therefore we must rest satisfied with self-fertilization, if the animal were really sedentary during its whole existence. But this is the case in no animal proceeding from an c^g j and all animals proceed from an Qgg^ if not in every generation, nevertheless in generations recurring regularly after a certain time. From this it follows thivt even in animals which become sessile contact may take place, always supposing that one of the two individuals is not yet sessile. During this contact they might fertilize each other, if both kinds of sexual organs were already developed in both. This, however, for econo- mical reasons is usually impossible ; for parasites, at least, it is attachment that usually secures that quantity of nutriment which is necessary for the egg-production ; and, on the other hand, parasites generally require for the maintenance of their species such a colossal fertility, that the egg-formation of itself deforms the body and compels it to become attached. Hence, with special exceptions, it is not well possible that the free-moving individual should already possess ovaries ; consequently neither a mutual fertilization nor the one-sided one supposed by Fraisse can be accepted as taking place between two free-swimming animals. On the other hand, the animal may well be capable of the production of the semen, of which no great quantity is necessary, even before the commencement of adhesion, and thus the protaudry above described would be brought about. An advantage over the 10 Prof. R. Kossmann on the Cryptoniscidse. simple dimorphism of the sexes is, however, also obtained. If I assume (quite arbitrarily) that the individual occupies a week from hatching to male maturity, and then three weeks from hatching to female maturity, then to obtain 10 (or n) broods in the case of dimorphism, 10 (or n) individuals must escape all the dangers that threaten them each for a week, and 10 (or «) individuals each for three weeks (40 or 4n weeks), while in the case of protandry only 10 (or n) indivi- duals need to exist each for three weeks and 1 more for one week ( = 31 or 3n + l weeks). And the advantage is even still greater than these numbers show, as it is precisely the first week in which the animal (in our case) swims freely about, and consequently is much more exposed to dangers than during the next two weeks, when it is already adherent. Those 9 (or n — \) weeks which in our example are saved by protandry are, as one may easily convince one's self, all first weeks of life. It is the first week, that of the free-swimming stage, of nine female individuals that is saved. The notions as to the nature of the brood-cavity in which the ova in the Cryptoniscidte are sheltered until hatching have hitherto been very defective. Buchholz* found that in Ci^yptothir halani the deposited eggs " float to and fro . . . apparently free in the body-cavity ;" but in reality are " en- closed in a special, extremely delicate-walled, and perfectly transparent vesicle." His further statements upon this subject do not seem to me very clear. He finds this vesicle attached to the outer wall of the body at the spot *• at which the four genital apertures are situated If we separate the pedicle of the vesicle from this spot we obtain it in connexion with the four oviducts, which remain attached to it uninjured, and the outer extremities of which seem to pass directly into the vesicle." Thus, while Buchholz originally saw the ovi- ducts open outwards, he sees them afterwards open into the vesicle ; both observations which I can confirm as correct. In spite of this the original four sexual apertures are said to persist on the outer surface. " Nevertheless," he says, " the presence of external sexual apertures at this spot, simultane- ously with the opening of the oviducts into the egg-reservoir, is difiicult to understand." In my opinion, it is not to be understood ; and his attempted explanation, which I shall not reprint here, is quite unintelligible to me. The true condi- tion of things in Gryptotliir^ as I have observed with certainty, is, that there is a sinking in of the region of the genital aper- tures, at first in the form of a transverse groove. In this way * Bucl)linlz,/oc. cit. p. 31o. Prof. E,. Kossmann on the Cryptoniscidse. 11 a brood-space with a transverse entrance-fissure is formed by invagination, and as its inner surface is the former surface of the animal in the neighbourhood of the oviducal aperture, the oviducts of course no longer open outwards, but into this newly-formed brood-space. Somewhat more complicated, but perhaps more primitive, is the arrangement in the other Cryptoniscidaj, at least in Cryptoniscus and Liriopsis. Fraisse* falls into an error, which, however, his predecessor had escaped ; he says, " This brood-cavity was previously present, for it is simply the body- cavity," and adds, " How the ova get into it I cannot say ;" and, in fact, it would be hardly possible to establish a con- ceivable hypothesis upon this view : deposition of the ova in the body- cavity would be something unheard of among Crus- tacea. Into this brood- or, according to Fraisse, body-cavity lead two " respiratory " apertures, already very fully de- scribed by him, one of them in the neighbourhood of the mouth, the other further back, and the two united with each other by a longitudinal groove (upon the ventral surface). Of this groove Fraisse says (p. V6) : — " When the larvae are ready for a free existence, a fissure bursts which formed between the tsvo respiratory apertures during the third stage, and was previously covered and closed by a thin cuticular layer." Through this the larvae are set free. The actual course of events, as I have ascertained by the study of numerous transversely-sectioned specimens, is as follows. Here also, first of all, the two oviducts (which Fraisse, /. c. p. 9, was unable to detect) open on the ventral outer surface of the body. Their originally circular aperture becomes elliptical and is soon drawn out before and behind into a shallow groove, so that we may easily recognize two such parallel longitudinal grooves in the female when not yet quite mature. Now the wall between the two grooves begins to sink in, and thus we obtain, instead of the two shallow grooves, an elongated depression, the two side walls of which curve into one another before and behind. These side-margins now, however, grow towards one another until they touch throughout nearly their whole length ; there remains conse- quently only a hair-like slit, which passes at its posterior and anterior ends into a rounded hole, and these two holes lead into a cavity formed by the sinking in of the wall lying be- tween the two genital furrows. This cavity, into which of course the oviducts open, is the brood-cavity, but it has abso- lutely nothing to do with the body-cavity. It becomes tilled * Fraist('. cif. \\. 12. 12 Prof. R. Kossmann on the CryptoniscidfB. with deposited ova and thereby enlarges in proportion as the ovaries (and at the same time also the alimentary organ) diminish in volume. That the ova obtain the change of water necessary for respi- ration, is provided for by brisk pumping movements, which, as Fraisse correctly describes, are effected in this later stage by the musculature at the two brood-space or respiratory apertures ; and that the current of water sliall not wash out the ova, by a system of villiform valves which close the apertures eelpot fashion, as Fraisse has also shown. This whole ar- rangement is produced, moreover, before the last moult of the animal, and therefore, in a certain stage, the fissure of the brood-chamber still appears, as Fraisse says, " covered and closed by a thin cuticular layer." Subsequently, however, the fissure still holds together, only because the two margins are to a certain extent interlocked ; with a little effort they can be readily forced asunder without tearing anything. But spon- taneously the fissure certainly only opens when the brood- cavity is overfilled, and the parent animal performs the most violent contractions. As Fraisse correctly describes, these often still continue when all the ova are already expelled ; and as, at this time, the ovary and alimentary apparatus, tlie only coloured organs of the animal, are completely retrograded, the animal, which now resembles a torn and perfectly trans- parent rag, and yet contracts violently, presents a very remarkable ajjpearance. Passing over many details of less general interest I cannot .abstain from stating something with regard to the mode of taking nourishment and the alimentary organs. In Cryptothir [Hemioniscus] ^ as Buchholz has already shown, the cephalon and the pereion as far as the antepenulti- mate segment remain larviform throughout life ; in accord- ance with this the small boring and sucking apparatus, consisting of the labrum and labium, between which two styliform mandibles are placed, serves through life as the organ for the inception of nourishment. This anterior divi- sion of the body is, however, also deformed in the other genera. The genus Zeuxo^ which I discovered in 1872, lives upon parasitic Cirripedes, especially upon Sacculina, a parasite of the Brachyura. It perforates this animal, the pedicle of which ramifies like a root in the body of a common crab, at the point where the pedicle enters into the body of the crab, and draws nourishment from it, after the fashion of a plant ; it consequently intercepts the nourishment of the Sacculina, and often causes it to die away altogether with only the exception of the roots. These roots, singularly Prof. R. Kossmann on the Cryptoniscidaj. lo enough, remain alive, and arc made use of by the Zeuxo. The head of the latter, which is finally inserted deeply into the body of the crab, although always still in a great lacuna of the radiciform pedicle of the Sacculina, presents, besides the buccal aperture, only four cylindrical processes, of which one pair is usually longer than the other. By their form and position they give rise to the supposition that they are the antennge of the larva which have lost their articulations. They evidently effect the fixation of the animal. In some species the fore part of the body, from the place where it enters into the body of the host to the mouth, is drawn out into a long peduncle ; in others it is rather short. Upon another parasitic Cirripede {Peltogaster) lives the genus Cryj)t07tiscus^ F. Miiller, under exactly similar vital conditions. Its head docs not form a peduncle, but within the aperture which it has perforated we find four pad-like swellings surrounding the mouth, which, from analogy, we may also regard as modified antennaj. That this genus almost always brings about the destruction of the Peltogaster itself has been already indicated by Fritz Miiller and con- firmed byFraisse; now and then, indeed, we find a specimen which seems not to be seated directly upon the hermit-crab, but has bored somewhere into the mantle of the Peltogaster ] but if such stray examples, on the one hand, do not quite cause the destruction of the Peltogaster ^ on the other hand they do not seem themselves to arrive at female sexual maturity. It is otherwise with the genus Liriopsis^ Max Schultze {Liri'ope, Rathke), which also lives upon a Peltogaster. This animal (the anterior and posterior ends of which have been hitherto mistaken) does not perforate the pedicle of the Peltogaster, but slips into the cavity of its mantle, and per- forates the mantle from within. Thus the anterior half of the body is inserted into the blood-lacunas of the mantle, while the posterior half lies free in the mantle-cavity, and the perforation which the parasite has made causes a median constriction of the animal. But in this case, not only the head, but at least five segments of the middle-body are inserted into the host ; and as the aperture through which the brood of the Liinope swarms out is formed upon these seg- ments; the parasite, when the brood is mature, or perhaps a little earlier, must also break through the outer wall of the mantle. This is probably effected less by boring than because its own growth exerts such a pressure upon the tissues before it that the latter become atrophied and finally burst. The parasite then remains with the abdomen in the mantle-cavity 14 Prof. R. Kossmann on the Ciyptoniscicloe. and the fore part of the body outside in the open. I have been unable any longer to detect antennas or buccal organs in the stage of female maturity, and my experience was similar in some other genera. Among these are Eumetor^ which was likewise discovered by me in 1872, and has now been more accurately studied, and Cahircyps^ which inhabits the brood-cavity of the Bopyridse, its nearest allies. All these three genera, in the adult condition, have the head free, and consequently need then no apparatus for fixation or boring. It is true also that they can then take no more nourishment. But this is no longer necessary to them ; G ryptoniscus and Zeuxo also at this time take no more nourishment, although they have the head still inserted into the blood of the host. This follows with certainty from the retrogression of the digestive apparatus, already ascertained by Fraisse in Gnjpto- niscus. The knowledge of the nature of this digestive apparatus has advanced by various roundabout ways ; 1 will here only briefly refer to its course hitherto. Rathke* ascribed to the Bopyridas a liver consisting of seven pairs of follicles opening separately into the intestine ; Cornalia and Panceri f, certainly in a difierent genus, describe, instead of these fourteen folli- cles, two tubes running parallel to the intestine, their opening into which they did not see. Buchholzf found in Crypto- niscus halani, intercalated between the oesophagus and the rectum, a comparatively enormous vesicular reservoir, drawn out posteriorly into two caica, which, from its appearance, is " really to be regarded as the intestinal canal ;" and a similar condition of things in Entoniscus was described first by Fritz Miiller § and afterwards by Fraisse ||, the former characterizing the organ as the liver, the latter as rectum. This latter inter- pretation, as a section of the intestine, is also maintained by Fraisse for a corresponding organ in the Cryptoniscidas, which certainly, by its exceedingly vigorous growth, soon loses all trace of a division into parallel tubes. I have already else- where, with regard to the Bopyrida? and Entoniscidas, adopted the opinion of those who regard this organ as a homologue of the so-called liver of the Crustacea. I do so also uncondi- tionally with respect to the digestive organ of the Crypto- * Rathke, ' De Bopyro et Nereide,' 1837, p. 9, tab. i. fif?. 7 b. t Cornalia e Pauceri, " Osservazioni sopra im iiuovo geuere di cro- stacei isopodi sedentarii, Gyye hranchialis,''' 1858, p. IG, tab. ii. fig. 6e. t Bucliliolz, Jlemioniscus balani, loc. cit. p. olO. § Fritz Miiller, JEntoniscus purcellance, iu Ai'chiv f iir Naturg. Bd. xxviii. p. 11. II Fi-aisse, EtUmiscus Cavolinii, 1878, p. 17 Prof. R. Kossmann on the Ciyptoniscidaj. 15 niscidge. Both the development of the organ from a pair of cylindrical csecal tubes opening into the anterior part of the intestine and its histological nature prove the homology with the so-called liver. But just as I have already said with regard to the Bopyrida3 that this so-called liver does not function exclusively as such, but evidently performs '"' a func- tion as a section of the intestine," so must I also assert deci- dedly that the lumen of this so-called liver receives the food of the parasite, which is identical with the blood of the host, in immense quantities; that in this place this nutriment is digested and absorbed, during which the organ gradually shrivels up ; and that consequently the name of liver is by no means physiologically applicable to the organ. But at the same time it has been sufhciently demonstrated that the so-called Crustacean liver is no liver at all. The name reposes on an error called forth by the most insignificant superficial character, namely the colour of the organ. Hoppe- Seyler * and Krukenberg f found in the secretion of the so-called liver of the higher Crustacea a diastatic, a peptic, a tryptic, and a fat-decomposing enzyma. Max Weber \ believed that he found in the epithelium of the liver, besides the true hepatic cells, a second kind of cells, which he supposed to fill the above more pancreatic function, and he named the organ heixitopancreas. But Iloppe-Seyler showed in the freshwater crayfish, and Frenzel § has lately done so in many marine Crustaceans, that no biliary constituents at all are present in the secretion. There are no biliary acids or their soda and potash salts, no bilifuscin or the allied pigments • and bilirubin was sought in vain. To this must be added that Frenzel has also ascertained that Weber's assertion that there are two different kinds of epithelial cells is erroneous. In my Epicaridia nothing of the kind can have existed. In short this organ, Avherever it has the form of a gland is clearly the digestive gland of the Crustaceans, a glandula intestinalis. But in the Epicaridia, and especially in the Cryptoniscida3, the lumen of the intestine does not suffice for the reception of the food, and then this organ takes part in it in a very remarkable manner. Thus from a glandula intes- tinalis it becomes an intestinum glandidare^ a reservoir * Hoppe-Seyler, ' Physiologische Chemie,' p. 276. t Krukenberg, " Vergleich.-physiol. Beitr. zur Kenntniss der Verdaii- ungsvorgiiuge," and " Zur Verdauung bei den Krebsen," in Uutersucliun- geu aus den physiol. lustitut in Heidelberg, 13d. ii. X M. Weber, " Ueber den Ban und die Thatigkeit der sog. Leber der Crustaceen," in Arch. f. mikr. Anat. Bd. xvii. p. 385. § J. Frenzel, " Ueber die Mitteldarmdriise der Crustaceen," in Mittlieil. a. d. zool. Station zu Neapel, Bd. v. p. 50. 16 Prof. R. Kossmann on the Ciyptoniscldaj. functioning as an intestine, with a secreting and at the same time absorbent epithelium. The sucking stomacli so characteristic of the Bopyridaj and Entoniscidffi, with the large papillte projecting into its lumen, is deficient not only in Crjfptothir^ in which, indeed, the larvi- forra fore body would afford no space for it, but also in the other Cryptoniscid^e ; in them, evidently, the intestinum glan- duJare^ the walls of which perform lively movements, assumes its function. Finally, as regards the rectum. Its connexion with the anterior part of the intestine is interrupted in the more strongly deformed animals, and we find it only in larvae and young females. As Buchholz, F. Miiller, and Fraisse have already shown, it is inflated into a pear-shape not far from the anus. (In opposition to Fraisse's statements, I find an anus even in the older animals.) Of the bacilliform elements of the con- tents described by Buchholz I have seen nothing any more than Fraisse ; in my sections the lumen of this part is quite empty, as the large cells of the wall project into it like papillae, and leave only a small stelliform space free in each transverse section. I cannot confirm the statements of Buch- holz and Fraisse that pigment cells surround the rectum ; the brown pigment deposited in the neighbourhood of the intes- tine is extracellular, and is probably to be regarded as urinary concretion. Whether we have to do here, as Blanc * sup- poses, with a separation-product of the fatty body, may be regarded in this case as doubtful, as no distinct fatty-body elements could be demonstrated even in tlie youngest stages investigated by me. This, however, is no absolute impedi- ment. In the Bopyridce and Entoniscida; the fatty body, which is at first very large, is reduced in proportion as the ovary enlarges by the maturation of the ova ; and it is com- prehensible that in the Cryptoniscida?, in which sexual products, i. e. male products, are so very early developed, the fatty body will also be reduced very early ; and in a certain sense it is rendered unnecessary by the intestinum glandidare^ which, indeed, does not serve tor the accumulation of already assimilated nutritive material like the fatty body, but never- theless accumulates unassimilated food in enormous quantities ; and this, as it is assimilated, goes directly to the advantage of the ova. It may, however, be supposed that the urinary masses are separated from a fatty body vvliich existed during the larval period ; they are actually present in the greatest * Blanc, " Observations faites sur la Tanaits Oerstedii^' in Zool. Anzeiger, 1883, p. 687. On the Spongia coriacea of Montagu. 17 quantity during the male sexual maturity, and diminish sub- sequently in amount, not only relatively but absolutely. Nevertheless, I only wish to indicate a possibility ; it seems a more probable supposition that these urinary concretions originate in the blood and are deposited in the wall-less blood- lacunre, the most important of which indeed run along the intestine. The decrease of these pigment-secretions coincides with the commencement of the sedentary mode of life, and therefore also with the complete change of nourishment, and may consequently be caused thereby instead of by the disap- pearance of the fatty body. That the rectal vesicle produces a strongly smelling sub- stance, as Fraisse asserts, may be correct ; but I cannot contirra it, as unfortunately (or shall I, as a zoologist, say fortunately ?) I possess a very feeble sense of smell. The other internal organs of the Cryptoniscidte show no great differences from those of the Bopyrid^ ; what there is to be said about them and about the details of the external organization I reserve for my monographic publication. II. — On the Spongia coriacea of ilfow^a(7M, = Leucosolenia coriacea, Bh.^ together with a neio Variety of Leucosolenia lacunosa, Bk., elucidating the Sjncular Structure of some of the Fossil Calcispongiw ; followed hy Illustrations of the Pi7i-like Sjncidcs on Verticillites helvetica^ De Loriol. By H. J. Carter, F.R.S. &c. [Plate I.] In 1871 (' Annals,' vol. vii. p. 278) I gave a nomenclatural account of Montagu's Spongia coriacea =Grantia clathrus, Sdt.,= Leucosolenia coriacea^ Bk.,= Clathrina clathrus^ Grray, under the last name, which was subsequently changed by Hiickel into Ascetta clathrus (' Die Kalkschwamme,' vol. ii. p. 30), and now I propose to add the result of a structural examination, chiefly on living specimens, of this calcisponge from this place, viz. Budleigh-Salterton, South Devon. In limine, however, it is necessary to clear up the confusion that has arisen from Hackel having made a separate species of vSchmidt's Grantia clathrus under the name of Ascetta cla- thrus, with a different form of spicule from that which Schmidt has given as characteristic of it (Spong. Adriatisch. Meeres, Suppl. p. 24, Taf. iii. fig. 3 a), and which accords with that An7i. (& Mag. N. Hist. Ser. 5. Vol. xiv. 2 18 Mr. H. J. Carter on the of all the other authors mentioned but Hackel, who, instead of an equilateral triradiate with straight arms, obtusely pointed, has given one with undulating rays, which are inflated at the ends (o;;. cit. vol. ii. p. 36, Atlas, Taf. v. figs. Sd-f). Now had Hackel offered any explanation of this with re- ference to Schmidt's original announcement, one could have understood the discrepancy ; but he neither refers to Schmidt's characteristic figure of the spicule of Grantia clathrus {I. c), nor does he use the specimens which Schmidt gave hinii 1868 {ojy. cit. voh ii. p. 32), but gives the result of his exami- nation of those which he himself found in the spring of 1871 on the coast of the island of Lesina, in the Adriatic Sea, as typical of Ascetta clathrus [ib. p. 33). Thus we have to choose between Schmidt's published figure of his oicn speci- men in 1864 and Htickel's of his own in 1872, in which dilemma it is evident that the latter could not have been Grantia clathrus, and therefore that Hackel had no reasonable grounds for making it so. Hence the species at Budleigh- Salterton, being the original Spongia coriacea of Montagu, must be viewed as Schmidt's Grantia clathrus of 1864, and not as Hackel's Ascetta clathrus oi 1872; while as Gray's genus ^'■Clathrina^^ is founded on Clathrina sulphurea^^ Schmidt's Grantia clathrus, which, again, is equal to Leuco- solenia coriacea, Bk., and occurs here, as elsewhere, occasion- ally under sulphur-yellow and scarlet colours respectively, we must adopt Bower bank's or Gray's names for the whole ; and as that of the latter is most expressive of the anastomosing tubular structure of Spongia coriacea, Mont., while it does not indicate any particular colour or contain any other diffe- rent structure, such as Leucosolenia hotryoides, Bk., which is simply branched ('Die Kalkschwiimme,' Atlas, Taf. ix. fig. 10), it is evident that " 67a?/i!yi'?ia " is the most preferable, as the original generic name of Montagu, viz. ^^tSpongia^^ must necessarily be changed by some one. Of all the sponges growing on the rocks here that I have seen, no species is more strikingly beautiful than Montagu's Sr>ongia coriacea, which, when fresh and extending over an area of about four square inches, presents itself under the form of a reticulated structure of an icy-white colour, in which the reticulation is only just visible to the naked eye, but, when magnified, contrasts favourably, as it veils the dark rock beneath with the most chaste and exquisite network that could be produced artificially. ISIo representations of it liitherto do it full justice in this respect, least of all those of Johnston and Bowerbank. Spongia coriacea (t/* J/on^a(7M. 19 This network is composed of a hollow anastomosing thread or tibre of variable diameter, whose interstices or meshes (for a parallel illustration of this meshwork see the body of Leucosolenia lacu7iosa, var. Hillieri^ PI. I. fig. 2, a, f^ g) are subject to infinite variety both in size and shape throughout the specimen, which, when large, thins out from a massive variable amount of thickness in the centre, seldom more than 2 or ij-12ths of an inch, to a single reticulated layer at the circumference. These thickened centres are numerous in a specimen about the size mentioned, and form several little monticular elevations scattered over the surface, whose sum- mits respectively end in a little short open tube with naked margin, which is continuous with the neighbouring branches of the hollow thread-work, and thus forms an osculum or vent to this part of the sponge (see also PI. I. fig. 3), while the other ends of the thread are blind and attached to the rock over which the sponge may be growing. So that the hollow of the reticulated thread is homologous with the cloacal cavity of the other forms of Calcisponges. If we follow the development of this structure from the youngest form that can be seen, viz. that which has just come from the embryo, it will be found to consist of a simple erect sac, whose upper end is open and whose lower one is attached to the rock on which it may be growing. This is the com- mencement of the hollow thread out of which the largest specimen is finally constructed. It may now be about l-30th inch long by l-120th inch broad, narrowing to a point at the fixed end. Such are the measurements of the smallest forms which are just now (April 10th) to be seen of all sizes and all stages of development on the rocks at " Straight Point " here, where the species grows abundantly. In the next stage the sac sends out a tubular bud, which may also descend to the rock, and then, as the individual grows upwards and out- wards, other similar buds are put torth which either descend to the rocks for fixation or otherw ise anastomose with each other, until at last the reticulated structure first described is attained j but how the anastomosis is effected, that is, how the loops are formed, I have not been able to perceive. Thus the development is very simple, although the adult form appears to be so complicated, and the reticulated struc- ture not by any means confined to Grantia clathrus alone, but common to a great many different species of Calcisponges, of which the beautiful little Leucosolenia lacunosa, Bk., of the British shores is a stipitate form and also of different spicula- tion ; still, whatever the spiculation may be, the soft parts 2* 20 Mr. H. J. Carter on tJie appear to be similar in all. Grantia clathrus consists simply of the spicular structure, the spongozoa (Geisselzellen), and the granuliferous sarcocle (syncytium, Hiickel). The form of the spicules or skeletal support, which are chiefly situated in the outer part of the wall of the hollow thread, is of one kind only, viz. equiangular and equiradiate, with simple straight rays, obtusely pointed, as already men- tioned. The spongozoa do not appear to differ in form from those of the other Calcispongiaj, but instead of being arranged in juxtaposition around the interior of globular or sacciform cavities, with their respective cilia projecting into the in- terior, they appear to form in juxtaposition a continuous layer throughout the inner surface of the tubular thread after this manner. The " granules " of the sarcode, however, are very remark- able from their size and dominant presence ; and although they accompany the transparent sarcode everywhere, they a])pear when in situ among the spicules and spongozoa to be loosely grouped around a delicate nucleated cell respectively, the " Kern " of Hackel. It is now thirty-five years ago that I gave an illustrated description of this granuliferous sarcode in Spongilla (' An- nals,' 1849, vol. iv. p. 91, pi. iv. fig. 2, a--f) in the living state ; and as this seems to apply very nearly to that of Grantia datJirus (so far as the dead state of tlie latter goes in a prepa- ration to be presently mentioned), I will here quote the original paragraph, viz. : — " If a seed-like body [statoblast] which has arrived at maturity be placed in water, a white substance will, after a few days, be observed to have issued from its interior through the infundibular depression on its surface, and to have glued it to the glass ; and if this be examined with a microscope, its circumference will be found to consist of a semitransparent substance, the extreme edge of which is irregularly notched or extended into digital or tentacular prolongations precisely similar to those of the protean \cimoehci\j which, in progression or in polymorphism, throws out parts of its cell in this way (pi. iv. fig. 2, dd). In the semitransparent substance maybe observed hyaline vesicles of different sizes contracting and dilating themselves as in the protean (fig. 2, ee), and a little wnthin it, the green granules so grouped together (fig. 2,//) as almost to enable the practised eye to distinguish in situ the passing form [polymorphism] of the cells to which they belong ; we may also see in the latter their hyaline vesicles with their contained molecules in great commotion, and Spongia coriacea of Montagu. 21 between the cells themselves the intercellular mucilao-e " [syncytium of Hiickel] . One cannot help observing here that, as the illustration to the paragraph represents, the " granules " appear to have been dragged off their cells (Kerne), to become scattered in the pseudopodial sarcode, which thus also appears to be as homo- geneous as that of an Amcpha. Yet it seems questionable whether the cells from which this apparently homogeneous sarcode has been derived do not still retain their individuality, seeing that, in the conjugation (zygosis) of two Ehizopods, they with their granules appear to flow together as intimately as two drops of water, that is, their individuality becomes lost ; they jDut forth their pseudopods afterwards as if thoroughly amalgamated ; and yet, after a little while, they separate and appear to be the same in every respect as they were before the conjugation. Or, these cells and their accompaniments can unmake and remake themselves as the occasion may require and with the materials that are nearest, — so inexplicable are the j^henomena manifested by poly morpliiG sarcode ! Such facts would lead us to infer tliat the syncytium is composed of a congeries of polymorphic cells, which thus simulate a homogeneous substance, just as Rostafinski, and previously to him his teacher, A. de Bary, I think, has stated respecting that wonderful moving fungus ^thalium^ viz. : — That '' the contents of the spores at the time of germi- nation, give rise at first either to a naked zoospore provided with a nucleus, a contractile vacuole and long cilia [?tvvo], or to an amoeboid. These zoospores or amoebae flowing to- gether in masses give rise to mobile plasmodia " (Rostafinski, l)r. J., 'Monografia Sluzowce,' p. 83, in Polish, 1875; ap. Cooke, ' Myxomycetes of Great Britain,' p. 1) ; while in my observations on Ai^thalium at Bombay in 18G1 these appa- rently homogeneous masses or plasmodia evinced, during the restless unceasing changes in form of the fungus, the power of moving about and running together like so much v/ater, of constricting themselves isthmus-like almost to separation, of flowing back together again, of spreading themselves out den- dritically, and finally of ending in a motionless, circular, con- vex mass, which soon became a heap of black-brown spores ! Returning to the syncytium of Grantia clathrus one finds the granules so much more strikingly developed relatively here than in the other forms of Calcisponges, that one cannot help questioning their nature and import. Taking the granule singly, it is spherical, translucent, and glairy, glistening from refraction of light, of a faint yellow 22 Mr. H. J. Carter on the tinge, and varying under l-6000th of an inch in diamster, although rarely attaining tliis size in this state. They are, when in situ, congregated round a nucleated cell (the " Kern ") which is often so indistinct here as to be very difficult to see, owing to its delicate (? polymorphic) structure and the opaque mass which the granules form when closely applied to it in juxtaposition ; or they are scattered throughout the syn- cytium in the same way as in the Foraminifera, as the *' preparation," to which I have before alluded, which was made after Schultze's method, described by him in his exami- nation of EiqjlecteUa aspergillum (' Challenger ' Reports, separate copy, p. 5), plainly shows, where the granuliferous protoplasm or syncytium can be seen in a reticulated branched form extending across the cavity of the tubular thread, very much like that of Gromia oviformis, represented by Max Schultze in his " Organismus d. Polythalamien " (1854, tab. i. fig. 1). So that one feels inclined to infer that, ex- cepting for its spicules and the spongozoa, the sponge would be very nearly allied to a Foraminifer in this respect. Iodine does not turn them purple, nor does liquor potassas dissolve them ; but strong nitric acid appears to destroy their sphericity, which may be brought back again by the addition of liquor potass^. This glairy refractive appearance gives them the aspect of fat or albumen ; while, like the green granules in Hpoyigilla, they appear in the sulphur-yellow and scarlet varieties of Grantia clathriis to be the seat of these colours respectively, when they might be termed " pigmental." It is possible that they grow into the larger cells of the pro- toplasm (the "Kerne"), from which they appear to be derived, when they may fulfil other offices ; for Lieber- kiihn has long since shown that the " Korperparenchym," = syncytium, can enclose and extract nourishment from Infu- soria in the same manner as '"'' Actinoplirys soP^ (Muller's Archiv, 1857, Heft iv. p. 388). So the particles of wood taken into the plasmodia of j^thaUum indicate the same consequence. But whatever the office of the granules may be no one as yet lias demonstrated beyond conjecture what they are or what purpose they may subserve either in the sponges or in the Khizopoda, — so they are still called '^ the granules." The canal-system consists of the usual inhalant and excretory divisions : the former of minute pores which can only be seen by the microscope on the outside of a dried Avell-preserved specimen, where they are bordered by the granuliferous sarcode or syncytium ; here, too, probably, in the living state, composed of a congeries of distinct nu- Spongia coriacea of Montagu. 23 deated graniiliferous cells or bodies like the "investing mem- brane" of Spongilla, -wVxch. in combination there appeared to me to have the power of opening and closing a pore wherever they liked (see " Ultimate Structure of Sjjongilla,''^ ' Annals,' 1857, vol. xx. pp. 24, 25, pi. i. figs. 6, 7) ; the " Wanderzellen " or migrating, amoeboid cells of Schulze (Zeit. f. w. Zool. 1878, Bd. xxx. S. 409 &c.) ;— and the latter or excretory division, consisting of the general tubulation of tlie reticulated thread, opening in the way and at the vents mentioned ; homologous with the cloaca in the other forms of the CalcispongiiB, as before mentioned. The process of reproduction by ova &c. is probably tlie same as that of the other Calcispongige ; but in the hope of determining this as well as when the elements of reproduction begin to appear, I have gathered living specimens of Glathrina coriacea, Grantia coinpressa, and Grantia ciliata, var. spini- sjnculum (being together), from the " Rocks " here every lull moon {i. e. the " springs "), since the 11th of March last in- clusive, at which time I could see no trace of these elements in either of these species. They began to appear in the two latter on the 10th of April, and were strikingly developed, especially in Grantia compressa, on the 12th of May, but not advanced then beyond the ««;?segmented stage. In Glathrina coriacea no trace had then a]:)peared ; nor is there any now on h e 9th of June, altliough the fragments (taken from diffe- rent localities) were placed in pure spirit directly they were taken off the "Hocks," which preserv^es the collar and the cilium of the spongozoa in their extended state. It is there- fore plain that Glathrina coriacea, in point of time, does not develop its ova and spermatic cells so soon as Grantia com- pressa and Grantia ciliata, var. spinispiculum. My observa- tions on the former in this respect for the other summer months will be communicated hereafter. As before stated, the general structure of Glathrina coriacea = Grantia clathrus &c., that is, the reticulation formed by the continuous anastomosing of a hollow thread-like fibre or tube, is common to many Calcispongise, which Hackel has divided according to their spiculations respectively, so that they appear in several genera of the first two families of his "natural system," viz. the "Ascones" and "Leucones;" but of themselves they equally form a natural group in general structure as distinct as it is totally different from that of the other Calcispongiffi, that is, the anastomosing reticulation. Moreover, the little Australian calcisponge which I have de- scribed and illustrated under the name of Leucetta clathrata ('Annals,' 1883, vol. xi. pi. i. figs. 13-17) must form the 24 Mr. H. J. Carter on the type of a division of this family, in wliicli the anastomosing thread-like tibre is solid instead of hollow — a form entirely absent in the " Kalkschwamme " of Hiickel, but one of mucli interest, as I have heretofore shown, in elucidating the structure of some of the fossil Calcispongije, which I hope to still further advance by a description of the following new variety of Leucosolema lacunosa. Leucosolenia lacunosa ^ Bk., var. Hillierij Crtr. (PI. I. figs. 1-5.) Small, stipitate, erect ; body globular, obconic, rather compressed and turned to one side ; stem cylindrical and long, rather bent upon itself and compressed in its upper part. Body not hollow, but composed throughout of massive clathrous structure; stem solid (PI. I. figs. 1, 2). Colour pale yellowish white. Consistence firm, resilient in the head, hard and unyielding in the stem. Clathrous struc- ture or network (fig. 2, g) consisting of a mass of reticu- lated anastomosing thread-like tube (fig. 2,/) issuing from the stem in several divisions (fig, 2, t/), and terminating in the summit by a central dilatation into which the neigh- bouring branches of the reticulated structure are gathered together centripetally (fig. 3, Z»), finally opening by a single naked aperture more or less protruded, which is the osculuni (figs. 2 b and 3 a) . Pores minute, in the wall of the tubular structure. Stem consisting of a compact solid mass of spicules, compressed in its upper part, which is expanded scopiformly into the " divisions " mentioned (fig. 2, d), which, being solid like the stem at first, pass respectively by transition into the tubular form which characterizes the structure of the body (fig. 2, a), terminating below in a root-like expansion which is fixed to the object on which the sponge may be growing (fig. 2, e). Structure and composition of the wall of the reticulated fibre the same as that of Clathrina coriacea just mentioned, only the meshes of the network are elongated vertically, which of course is followed in direction by the branches of the tubular thread, i. e. from the stem to the summit. Spicules of three forms, viz.: — 1. Triradiate, equi- angular, inequiradiate, rays straight, smooth, rather obtusely pointed, the longest, which in the largest of these spicules is three or four times longer than either of the other two, directed backwards (fig. 4, a) ; the rest infinitely variable in size generally and in the unequal length of their rays, some nearly cquiradiate (fig. 4, h) ; longest ray of the larger tri- radiates (fig. 4, a) about 48-6000ths inch long by 2-t)000ths inch broad at the base. 2. Linear, acerate in appearance, but Spongia coriacca of Montagu. 25 consisting of two unequal portions divided by a slightly in- flated node which belongs to the longest part, and is therefore a little excentric (fig. 6, c) ; divisions smooth, but more or less varying in thickness here and there, especially towards the ends, which are obtusely pointed ; one division straight and the other a little curved so as to form a very slight angle with the straight one ; largest average size about 91-6000ths inch long by 2-6000ths inch broad (fig. 5, a). 3. Linear- vermiculate, smooth, attenuated towards the extremities, which are pointed ; divided like the foregoing by an excentric node (fig. 5jCc) ; amount of vermiculation and total lengtli very variable, the smallest perhaps about IG-GOOOths inch long by l-6000th inch broad, but immeasurable generally, from the amount of contortion (fig. 5,bb) • increasing in size and de- creasing in vermiculation so as at last to reach an interme- diate form (fig. 5, c/). The triradiates are equally present in the body and stem ; but the linear and vermiculate spicules are exclusively confined to the latter, wliere they form the outer layer and the triradiates the axial or internal structure ; they do not begin to appear before the stem begins to divide into the branches leading to the head (fig. 2, J), and then go on increasing in number and robustness, although not in length, down to the root-like expansion or oldest jjart, as is usual in most sponges. Size of largest specimen (fig. 1) 9-12ths inch in total length, of which the body is o^-12ths, and the stem the rest, viz. 5^-12ths inch ; greatest diameter of ■ the body about 3-12ths inch, that of the stem close to the body l-12th, and that towards its base l-24th inch. Hah. Marine ; growing on hard substances. Log. Rarasgate pier, liamsgate. Ohs. Independently of the interest attaching to this sponge as nyAYiQiyoiLeucosolenia lacunosa^ Bk., it is still more inter- esting as presenting a spiculation and structure which reveal the nature of tlie " filiform spicules " and structure of the fibres in some of the fossil Oalcispongiaj from the " Coral Hag" of Faringdon in Berkshire, These spicules, although first represented by Zittel, in 1878, in Feronella multidigitata (Abh. k. bayer. Akad. d. W. ii. CI, Bd. xiii. 2 Abth. Taf, xii. fig. 3), were just afterwards, that is in the same year, more particularly described and illus- trated under the above name, viz. " filiform spicules," by Sollas (' Annals,' 1878, voh ii. p. 356, pi. xiv. figs. 1-5) ; and subsequently described by myself (' Annals,' 1883, vol, xi. P-22).. Until Ziltel had kmdly convinced me by a microscopic preparation of Feronella multidigitata from the CretaceoUs 26 Mr. II. J. Carter on the of Le Mans, that Calcisponges existed in a fossilized state, I was inclined to discredit the fact, as my actual experience of the delicate structure and perishable i^ture of the spicules of the Calcispongiai then seemed to point out that their structure and spiculation was such that thej must inevitably go to pieces immediately after death, and therefore that the proba- bility of a Calcisponge becoming fossilized was very doubtful. When, however, convinced of the error I fully expected that recent specimens would be discovered which would ex- plain all the then anomalous structure and spiculations in the fossil ones ; and the first that tended chiefly towards this was the discovery, by Dr. Hinde, that the fibre of his Verticilliles d'Orhignyi from the Upper Greensand of Warminster Avas composed of three- and four-rayed calcisponge-spicules, which were so far loosened by disintegration that they could be easily extricated entire, and thus viewed under the microscope, mounted in balsam or otherwise, indeed a simple lens is sufficient (^Annals,' 1882, vol. x. p. 192 et seq. pi. xi.). At the same time Dr. Hinde discovered in his Sestrostomella rngosa from the Cretaceous of Vaches Noires, near Havre (ibid. pi. X. fig. 4, and pi. xii. fig. 12, &c.), the two-pronged " tuning-fork "-sha])cd minute spicule first represented by Dr. Bowerbank (Mon. Brit. Spong. vol. i. p. 268, pi. x. tig. 2o7) from a recent calcisponge at Freemantle, in S.W. Australia, and subsequently by Hiickel in his Leucetta pandora from the Gulf of St. Vincent &c. in S. Australia (' Die Kalk- schwamme,' Atlas, Taf. xxiii. fig. A). I was myself able also to confirm these observations respectively in Verticillites anasiomans from the Coral Bag of Faringdon and in a speci- men of Sestrostomella irova the Jura, kindly sent me by Zittel, when I also published the illustrated description of the little calcareous sponge from Freemantle, in which the clathrous structure was shown to be formed by the reticulated union of a thread-like element similar to that of Clathrina coriacea and Lencosolenia lacunosa^ but, as before stated, solid like the stem of the latter, and not holloic like the tubular thread of the head, being composed of a layer of small triradiates externally with a much larger and different triradiate-form axially or within (' Annals,' 1883, vol. xi. p. 33, pi. i. figs. 13-15) ; and now I have had the opportunity of describing one from the coast of England, in which the " filiform spicules," together with the solid fibre in the fossil species, also receive an explanation from a recent species. The specimens of this sponge, which are in spirit, were gathered on the pier at liamsgate by Mr. Hillier, after whom 1 have designated the variety, and presented to me by Mr. Spongia coriacea of Montagu. 27 B. W. Priest in September 1882, when I thought, from their resistance and apparent durability on being handled, that had I been acquainted with them earlier I should never have discredited the fact that a calcisponge could be fossilized. Thinking, however, from their resemblance that thej were specimens of Leucosolenia lacunosa, I put them aside under this belief; but lately I have had to examine them in con- nexion with the foregoing species, viz. Clathrina coriacea^ and then I perceived that the solidity of the stem and its spicular composition were like the fibre of Peronella muUi- digitata^ Zittel, of ? Scyphia perplexcij Quenstedt (tab. 125. fig. 63), and of Manon peziza, also Quenstedt (t. 182. fig. 30), respectively ; that is, that it was composed of triradiates in the centre faced by a layer of linear and vermiform spicules, each of which indicated by the kind of 7iode mentioned (PL I. fig. 5, ccc) near the centre, which slightly projects, that it represented the aborted state of a third ray, and thus a modi- fication of the triradiate. Now, when we consider that the stem, as it approaches the body (PI. I. fig. 2, «rZ), divides into a multitude of branches, each of which, although solid in the first instance, becomes transformed into a tahe to form the tubular thread of the body (fig. 2, /) , which by branching and anastomosing produces the clathrous structure in which the linear and vermicular spicules are entirely abseiit, and that the linear and vermicular spicules thus cease to appear where the transformation takes place^ it fol- lows that had the branches continued so/iVHike the thread of the clathrous structure in Leucetla dathrata, Gxiw {pp. et loc.cit.), they would have been identical in spicular composition and arrangement with the fibre of the fossils mentioned, where, on account of their contortion being perhaps more generally greater than that in Leucosolenia lacunosa, var. IliUieri^ the extreme thinness of the microscopic slice cutting oif the bends above and below, seldom allows one to be seen entire. Indeed the more contort ones in Leucosolenia Hillieri during the boiling out in liquor potassa3, from this together with the brittleness of the material, for the most part, come out broken. Thus the " filiform spicules " of the fossil Calcispongige seem to be elucidated. Pin-like Spicules (? parasitic) on Verticillites anastomans and A. helvetica. (PI. I. figs. 6-10.) At the conclusion of my paper on the Fossil Calcispongiaa of Faringdon ('Annals,' 1883, vol. xi. p. 33) I had only just time to mention Dr. Harvey B. Roll's discovery of pin- like spicules in that variety of Verticillites designated " helve- 28 On the Spongia coriacea of Montagu. tica " by De Loriol ; so I returned to the subject [ih. vol. xii. p. 26), when by having made and mounted microscopic sections myself, I was enabled to give a more detailed de- scription of the fact, and to announce that such spicules also existed in the same position in my specimens of Verticillites anastomans from Faringdon ; but on neither occasion had I time to illustrate this interesting discovery, which is so likely to pass unnoticed without representations, that I have availed myself of the present opportunity to fill up a vacant space with these, taken from my own preparations (PI. I. figs. 6-10). EXPLANATION OF PLATE I. Fig. 1. Leucosolenia laciinosa, var. Hillieri, u. Tar. Natural size. Fi(j.2. The same, inagniiied 3 diameters, to show : — a, body, composed throughout of massive dathrous or reticuh\ted structure. Body not hollow, h, vent or osculum ; c, stem ; d, scopiform expansion of stem into reticulated structure of body ; e, root-like attach- ment ; /, tubular thread-like fibre j g, meshes of clathrate structure. Fig. 3. The same. Summit more magnified, to show the continuity of the vent with the reticulated tubular structure of the body. a, vent ; b, tubulated thread or fibre ; c c, meshes or interstices of the reticulated structure. Fig. 4. The same, Triradiate spicides of the body and centre of the stem respectively, a, largest form ; h, smallest ; c, dotted line, illustrative of the iuequiradiate forms. Fig. 5. Tlie same. Substraight linear and linear contort spicules of the stem, a, substraight lijiear ; b b, linear contort ; c cc, nodes j dj intermediate form. N.B. — All these spicules are drawn to the scale of l-24th to 1-GOOOth inch. Illustrations of the Pin-like Spicules {? jmrasitic) ^-c. on Verticillites. Fig. 6. Verticillites Itelvetica, De Loriol. Horizontal section of the wall of a cylinder at the inflation, magnitied 4 diameters, to show the structure of the wall and its hourglass-shaped openings or canals, a, cavity of inflated chamber ; b, wall, apparently com- posed of little oval and quadrangular elements, because, in some instances, the section has passed through the hourglass-shaped canals, and in otliers not. Diagrammatic. Fig. 7. The same. Vertical section of part of the wall of a cylinder at the inflation, viewed from the inside ; magnified on the same scale, to show the structure of the wall and its hourglass- ehaped openings in this view, o, wall, noio seen to be continuous and not formed of separate elements, as the foregoing figure apparently represents ; b, inner opening of the hourglass-shaped cauals, with a dot in the centre, to represent the narrow part. Diagrammatic. Fij. 8. The same. Horizontal section of two of the so-called " oval elements " Avith the hourglass-canal between them filled with sand ; magnified on a scale of l-24th to l-1800th inch, a, out- side of inflation ; b, inside ; cc, so-called " oval elements," com- posed of homogeneous crystalline calcite, with minute fibrous Dr. W. Djbowski 07i Lubomirskia baicalensis. 29 gtructure (? product of fossilization) ; d, hourglass-shaped canal tilled with grains of quartz-sand; e e e, piu-Uke spicules in the " calcite," arranged arouud the funnel-shaped openings of the hourglass-canals outside respectively, as will be better understood by the next tigure, but here only seen in the section, where they may be ob.^erved to slope inwards with the head ex- ternally ; ff, row of triradiate spicules in the " calcite " within the pin-like spicules. Diagrammatic, with the detail relatively magnified. Fig. 9. The same. Pin-like spicule, more magnified, to show its shape and relative proportions. Fiff. ^^- The same. Vertical section of a portion of the wall at the infla- tion, viewed from the inside, magnified to the same scale (see a less magnified portion, fig, 7). a a, wall composed of the homo- geneous crystalline calcite with minute fibrous structure before mentioned ; hhh, constricted parts of the hourglass-shaped canals, respectively filled with quartz-sand, also as before men- tioned \ c cc, position of the triradiate spicules in the wall around the hourglass-canals, shown by their truncate ends ; d, part of the slice where the layer of triradiates has been ground off, showing that ee e, the cross-sections of the pin-like spicules, are in circles, indicative of their infundibular mode of arrangement around the external openings of the hour- glass-shaped canals. Diagrammatic, with the detail relatively magnified. III. — Some EemarJcs upon the Variability of Form in Lubo- mirskia baicalensisj and ui^on the Distribution of the Baikal Sponges in general. By Dr. W. Dybowski *. With P.tSS. by H. J. Caktee, F.K.iS. &c. [Plate II.J During the printing of mj memoir on tlie sponges of Lake Baikal t, I received from Irkutsk, from my brother Dr. Bene- dict Dybowski, a photographic representation, prepared by him, of Lubomirskia baicalensis^ and also a communication upon the general occurrence and distribution of sponges in Lake Baikal. These notes possess no little scientific interest, and may therefore serve to complete my memoir, so that I regard it as advisable to publish them as a brief supplement to my work above cited. * Translated from a separate copy, communicated by Mr. Caiter, of the paper published in the ' Bulletin de I'Academie des Sciences de St. Petersbourg,' tome xxvii. pp. 45-50. t W. Dybowski, " Studien liber die Spongien des russischen Peiches ]nit besonderer Beriicksichtigung der Spongien-P'auna des Baikal-Sees," in Mem. de I'Acad. des feci, de St. Petersb. ser. 7, tome xxvii. no. 6 (1880). 30 Dr. W. Dybowski on Lubomirskia baicalensis. The variability of the sponges in regard to their mor- phology is a generally known fact. That Lubomirskia baicalensis follows this general rule we have a satisfactory proof in the material before ns. The morphological variability of Lubomirshia baicalensis has already attracted the attention of Miklucho-Maclay * ; but, as may be concluded from his words [l. c. p. 8), he was acquainted only with inconsiderable variations. In order to furnish as complete a description as possible of the form and structure of our sponge, I will here summarize all that is already known upon the subject, and enlarge it by my own observations. But, remarkable as are the differences in the form of the sponge under consideration, all these morpho- logical deviations may always, by careful investigation, be referred to one and the same type. The simplest and therefore typical form of our sponge is that of an arborescent stem with cylindrical erect branches f. The branches originate sometimes at different heights (see Miklucho, I. c), but sometimes at the same level (see Dy- bowski, I. c). This simplest form as just described is modified in various ways, and the most important modifications are the fol- lowing : — I. Forms lu which the type is distinctly recognizable. a. The erect cylindrical branches of the sponge are not free throughout, but unite -with one another by several transverse anastomoses of various thickness and length (see Dybowski, L c. p. 12, tab. i. fig. 1). _ b. The erect cylindrical branches of the sponge stand so close to each other or to the stem that at the surfaces of contact the_y coalesce either with one another or with the main stem. By this means there are formed very variously shaped, elongated, more or le.-s flattened bodies, from the top of which larger or smaller branches originate (see PI. II. tig. 1 b). II. Forms in which the type is almost entirely effaced. a. Arborescent sponges. a. The branches do not stand erect, but form a more or less acute angle with the main stem. The individual branches are not cylindrical, but thinner at the free extremities than at the base. (Specimen in my collection.) /3. The short nearly cylindrical branches are pinnately arranged ; but * Miklucho-Maclay, " Ueber eiuige Schwiimme des ncirdlichen Stillen Oceans und des Eismeeres," in Mem. de I'Acad. des 8ci. de St. P^tersb. ser. 7, tome xv. no. 3. t See Dybowski, /. c. tab. i. fig. 1 ; Miklucho-Maclay, /. c. tab. i. fig. 5 ; and Middeiidorff, ' Sibirische Reise,' Bd. iv. Theil ii. Lief. 1, p. iOOo. Dr. W. Dybowski on Lubomirskla baicalensls. 31 tbey orio-in.ite only from one side of the stem Csemipinnate') (see ri. II. fig. 1 c). b. Bush-liliB sponges. When numerous twigs grow forth in various directions from a short and thick base, a bush-like form is pro- duced (PI. II. fig. 1_ a). In such a bush we observe very differently formed twigs. Most of them are furcate and grow together in their lower part. The size of a bush is somefimcs enormous. With regard to the occurrence and distribution of tlie sponges in Lake Baikal, my brother sends me the following information * : — • In the south-western portion of Lake Baikal f, that is in the whole stretch from Listwiennischnaja and Possolsk on the one hand, to Kultuk on the other, the sponges occur wherever the necessary conditions are present \. As a rule the sponges occur wherever the bottom of the lake is stony and where large blocks of rock or wood are lying about on it ; further numerous sponge-stocks are found in those places where the steep rocky shore forms terraces projecting into the water in steps. On the other hand, if the bottom be sandy, muchly^ or covered with small easily movable stones, no sponges occur. At a depth of 100 metres there- fore, where the bottom is always covered with fine mud, a few small coating stocks occur only where large blocks of stone or logs of wood project out of the mud. On the western shore these conditions are abundantly realized, so that here sponges are almost everywhere met with. Close to the shore, and in only inconsiderable depths, turf-like or cushion- like sponge-stocks exclusively occur ; globular ones are rave, arborescent forms are never found. The sponges usually are firmly attached to the surface of large pieces of wood or rock ; they show, however, a special inclination for rotting wood, so that in certain places almost every log of wood bears sponges. Sometimes one finds large logs which are regularly coated Avith a crust of sponges. Li considerable depths, as, for example, at a depth of 3 fathoms ( = 6*3 metres), often quite close to the steep rocky shore, bush-like sponges exclusively occur. In fine still water, when the surface of the lake is as smooth as a * In the absence of a monographic treatise on the Baikal sponges the observer only distinguishes the arborescent from the turf-like or cushion- like stocks. The separate species or varieties are not specially treated, but only the conditions of the Baikal sponges in general. t See Dybowski, I. c. p. 7, fig. 1. X Extract from a letter dated " Irkutsk, 1878." 32 Dr. W. DybowskI ow Lubomlrskia balcalensis. mirror, if we go about in a boat we can delight ourselves with the sight of numerous beautiful colonies of sponge- bushes growing up from the bottom of the lake. The bushy- sponges sometimes attain an enormous size. One of them brought up by the dredge was so large that it by itself filled the whole bag of the dredge ■^. Such large stocks, however, are rare ; usually the size of the sponge-bushes does not exceed 60 centim. At a depth of 6-25 metres, arborescent or fruticose sponges occur ; at greater depths only cushion- or turf-like sponges. At a depth of 100 metres (the greatest depth at which sponges have as yet been taken) we find only occasionally a few small and flat stocks, and even these only when the above- mentioned conditions are realized. On the eastern shore the bottom of the lake is generally covered with sand or with small easily movable stones ; there- fore the s]3onges are here much scarcer. On the larger blocks the sponges oidy rarely occur, because the west winds, which are here prevalent, cau.se a continual succession of waves, which is evidently injurious to the growth of the sponges. On the rocky terraces of the shore there are a few cushion-like sponges, but not in such abundance as on the other side, on the western coast of the lake. In the Angara (see Dybowski, /. c. fig. 1) the sponges occur everywhere in great abundance from the mouth up to the Taltzinskaja manufactory. They are exclusively turf-like or lamelliform sponges, which adhere to large logs of wood and blocks of stone or to smaller pieces of wood ; arborescent forms never occur ; evidently the sponges require still water in order to become developed in the arborescent form. Between the Taltzinskaja manufactory and the city of Irkutsk sponges occur much more rarely, as here large stones and logs of wood are less numerous. Above Irkutsk the Angara has not yet been examined for sponges. If we briefly summarize all that is known with regard to the distribution of the Baikal sponges, the following may be said : — 1. Close to the shore of the lake, at a depth of 2-6 metres, only turf- like. 2. Ax a depth of 6-25 metres, arborescent or fruticose. 3. At a depth of 25-100 metres turf-like stocks again occur. At all these depths, of course, the sponges only exist under the above mentioned favourable conditions. * The Ijag holds about 40 pounds of mud. Dr. W. Djbowski on Lubomirskia baicalensls. 33 The colour of the sponges is for the most part more or less dark grass-green ; sometimes, however, they occur olive-green or brown ; only those sponges which come from considerable depths (60-100 metres), or have grown under stones, are almost colourless. The sponges of an indeterminate dingy greyish colour are probably dying or dead stocks. The following parasites have hitherto been observed upon the Baikal sponges — Qammarus parasiticus^ G. violaceus^ and G. violaceus var. virens^. [P.S.— On the 29th of May last I received from Dr. Dy- bowski small fragments of all the freshwater sponges from Lake Baikal in Central Asia and the Pachabica-See at its south-west extremity that he had described and illustrated in the ' Memoirs of the Imp. Acad, of St. Petersburg ' (tomes xxvii. and xxx. nos. 6 and 10, 1880 and 1882 respectively), together with one of his "Dosilia (?) Stepanowvi,^^ from the neighbourhood of Kharkow in Southern Russia, to which I alluded in the 'Annals' of April last (p. 272), by which I have been enabled, through microscopical examination, to confirm all that he has stated of the forms of their several spicules, especially those of the latter, of whose statoblast Dr. Dybowski kindly sent me a sketch on the 30th May last, to show that it possessed a tubular extension of the chitinous coat, accompanied by the cirrous appendages and armed with statoblast-birotules of different lengths, like those of Mr. H. Mills's Carterins (olim Carterella) tuhisperma from the Niagara River ; but on comparison with a mounted specimen of the latter, I find D. Stcimnoum sufficiently different to merit a distinct appellation. During my examination of the material above mentioned I have been struck with the ability and accuracy of Dr. Dybowski's observations, hence look forward with much pleasure to his description and illustrations of the statoblasts of his " Dosilia (?) Stepanoivii^^'' which he intends to publish on the earliest opportunity. — H. J. C. P.P.S. — In another letter from Dr. W. Dybowski, dated 1st June, he sends me sketches of the statoblast of Mr. Potts's Spovgilla friabilis, Leidy, var. segregata, which was also found near Kharkow, in Southern Russia, thus adding another locality to those already mentioned of this species. — H. J. C] * Dr. B. Dybowski, "Beitrjige zur iijilieren Keuntniss der in dem Baikal-See vorliommenden niederen Krebse aus der Gruppe der Gamma- riden/' in the Horee Soc. Entom. Boss. Beiheft zii Bd. x. p. 75, tab. x. fig. 3, p. 76, tab. xii. fig. 5, p. 147, tab. iii. fig. 3 (St. Petersbourg, 1874). Ann. <&) Mag. N. Hist. Ser. 5. Vol. xiv. 3 34 Mr. A. G. Butler on three neio Moths. EXPLANATION OF PLATE II. Fig. 1. Luhomirskia baicalensis,Vdl\^&, s^. About one third of the natural size. Three specimens, viz. a, b, and c. From a photograph. Fig. 2. The same. Skeletal spicule, lateral view and transverse section. Magnified 650 times, a, lateral view ; b, transverse section. (Mim. de I'Acad. Imp. d. Sc St. Petershourg, 7^ serie, t. xxvii. no. 6, Taf. ii. fig. 5, 6.) lY, — Descriptions of three new Species of Moths from the Island ofNias. By A. G. Butlek, F.L.S., F.Z.S., &c. The three following species were added last year to the National Collection ; two of them are especially interesting, as affording an admirable instance of protective assimilation. Agaristida. Ophthalmis decipiens^ sp. n. Allied to 0. mollis ; grey-blue, with a faint greenish tinge ; veins black ; the basal third of primaries crossed obliquely by irregular black stripes, the last two of which form an -shaped blue marking at basal third of costa, followed by a broad, oblique, snow-white, externally blue- edged belt from costa to external angle ; veins beyond the belt bluish ; a blue and white subbasal dot : secondaries reddish orange, with the base, costa, and a rather broad external border black ; a strongly dentated greenish-blue marginal stripe ; thorax black ; head dotted with blue and white ; abdomen dull black, regularly banded with creamy wdiite. Wings below somewhat as above, but the primaries with one or two subbasal blue-edged white spots, an oblique lunulated bluish stripe just before the middle, ajjical area greenish blue, with black longitudinal stripes between the veins ; secondaries with a pale blue basal spot ; the dentated blue margin covering the outer half of the external border. Pectus black, spotted with bluish white; venter broadly banded with white. Expanse of wings 68 millim. At once separable from L. ohliquaria by the great width and white colour of the belt across the primaries, and the absence of the black spot on the orange area of the secondaries. Euschemidae. PancetJiia simulans^ sp. n. Has the general aspect of Ophthalmis decipiensj but is of a paler blue-grey colour ; the wings are crossed by four oblique series of black markings, the first consisting of unequal oval spots, the third much the largest, but those on the secondaries subconfluent ; the second series consists of two reversed curved lines united by cross lines at the extremities upon the costal margin and the first median branch ; below this is a pyriforra spot on the internal border, followed by a squamose black line across the secondaries ; the third series consists of unequal oval spots, the second and fifth large and double on all the wings; lastly, a series of elongated black spots gradually lengthening and widening towards the costa of the primaries, where they unite so as to represent the black border on the • Ophthalmis ; body blue-grey, thorax banded with dull grey • abdomen with the last three segments bright ochreous. Wino-s below duller than above, the black markings badly defined ; pectus grey 5 venter bright ochreous. Expanse of wings 48 millim. A more elegantly formed species than P. georgiata and differently marked. 3* 36 Dr. Karl Krapelln on the Pulicidse. V. — On the Systematic Position of the Pulicidas. By Dr. Karl Krapelin*. [Plate ni.] After my investigations on the buccal organs of the Diptera and Rhynchota + had led me to the conclusion that in the former the true sucking-tube (not to be confounded with the labium, which serves only as its sheath) was formed by a dorsal and a ventral half-gutter (labrum and hypopharynx), and in the latter by two double half-gutters laterally inter- locked, it seemed natural to study also the aberrant members of the two series in the light of this criterion, which ap])lied to all typical forms, in order to arrive at greater clearness with regard to their relationships. In this respect no small interest undoubtedly attaches to the group Pulicidffi, which, notwithstanding much difference of form, presents such a uniformity of organization, and as to the systematic position of which for more than a century the most different opinions have been expressed, without any generally acceptable and well-established view having yet been arrived at. The history of these opinions has ah-eady been given pretty completely by Taschenberg in his Monograph on the Fleas J, so that here a short recapitulation may suffice. Linn^, as is well known, created an order Aptera for the wingless insects, Myriopods, Spiders, &c., and in this the flea found its place. A simihar position was assigned to it by Geoffroy, Cuvier, and Dumdril, as also by Gervais ; while, on the other hand, the order Aptera was by many rejected as unnatural, and the relationship of the Pulicidse with various winged insects was asserted. Thus Kircher referred them to the Orthoptera, Fabricius and Illiger to the E-hynchota, Rosel, Oken, Strauss-Durckheim, Newman, Burmeister, Walker, Von Siebold, and others to the Diptera. Lastly, there were also very early naturalists who would associate the flea with none of the existing orders of insects, but postulated a distinct order for it. The leader in this direction is De Geer. He was followed by Lamarck, Latreille, Kirby and Spence, MacLeay, Leach, Dug^s^ Bouchd, and Van der Hoeven, and, * ' Festschrift zum 50-jaliiigen Jiibiliium des Realgymnasiums des Jolianneiims,' Hamburg, 1884. Translated by AV, S. Dallas, F.L.S, + In part set forth in the preliminaiy communication "Ueber dio. Mund-werkzeuge der saugenden Insekten" (Zool. Anz. 1882, pp. 674-79) and in a memoir, "Zur Anatomie und Physiologie des Eiissels voii Musca " (Zeitschr. f. wiss. Zool. xxxix. pp. 683-719). X Taschenberg, 'Die Flohe ' (Halle, 1880). Dr. Karl Krapelln on the Pdiclclae. 37 among later investigators, by Landois and Tasclienberg. But although the last two authors especially pronounced most decidedly in favour of the independent position of the Fleas in the system, and although the most accepted special works upon the Diptera exclude the Fleas as not belonging to the series of forms in that order *, we find that even in the most recent manuals of zoology the group of insects in question is almost without exception cited as a suborder of the Diptera. This may pass in the first place as a proof that really stringent arguments have not yet been brought forward in favour of either view ; but we might also derive the hesitation felt by many zoologists to raise the rank of the Fleas (even under otherwise sufficient grounds) from the circumstance that they lead a parasitic existence, and by this means have possibly undergone profound and peculiar morphological changes by " adaptation," as is sufficiently established for other groups of parasitic forms. In opposition to this, however, it must be remembered that with only isolated exceptions (the females of the Sarcopsyllidffi) the Pulicidee are not stationary, but only temporary parasites, that their whole development is completed without parasitism, and that therefore we cannot well assume any considerable adaptation to a parasitic mode of life. But if this be so, if we succeed in proving that the Pulicidai possess a series of morphological characters which cannot be regarded as acquired by parasitism, we must necessarily, in judging of their position in the system, consider the same points of view to be prescriptive that have been generally adopted for the establishment of orders, suborders, and families in the class of insects. These general points of view, however, do not offer us a very brilliant prospect. The Linnean principium divisionisj the form, number, and texture of the wings, having proved to be untenable, we find on the one hand the kind of transforma- tion and its various stages, and on the othei the structure of the organs of the mouth, raised into the most important criteria of the nearer or more distant relationship of the groups of insects. But, as is always the case, when a single character is thrown too much into the foreground, and the general morphological relations of the two series of forms are not allowed to be prescriptive, difficulties make their appearance even with these apparently so thorough-going principles of division, which considerably diminish their value. The * It is interesting that the well-known work on the Diptera of the ' Fauna Austriaca ' by Schiuer certainly expresses itself decidedly enough in the above sense, but then gives a detinition of the true Diptera, which might very well embrace the Pulicidae. 38' Dr. Karl Krapelin on the Pulicidss. group of the Orthoptera, which is certainly not very natural, and their multifarious relations with the Neuroptera, the suc- torial ApidfB, the biting Mallophaga, and lastly the pupal rest of the male Coccidaj, may sufficiently establish this pro- position. It is still worse, however, as regards general avail- ability, with the distinctive characters of the orders generally cited — the segmentation of the thorax and tarsi, the structure of tlie wings, of the different buccal organs, antennae, &c. The mere fact of the agreement or difference of these organs individu- ally cannot give us certainty as to the systematic relationship of two series of forms, but only the examination whether the general organization of one group, as expressed in the deve- lopment of all morphological characters, shows or does not b\\ow jjhyhgenetic relations with those of another group; in other words, whether the observed ditterences in the structure of the parts may be referred equally v/ell to a different '' fun- damental plan " in their arrangement, as to simple changes of form and reductions, such as may be explained by altered function. Self-evident as this proposition appears in the light of modern zoology, the history of opinion as to the sys- tematic position of the flea nevertheless shows very plainly how little it has hitherto been taken into consideration by entomologists. One important aid in such investigations upon the true phylogenetie relationships of forms is unfortu- nately at present still almost wholly shut out from us. I refer to the anatomical structure of the organs. The knowledge of this, and especially that of the generative organs, is at present 60 imperfect that a detailed consideration of the internal organization seems to be of little use in the classification of insects. After these prefatory remarks upon the principles which are or should be of force in the grouping of insect-forms, the question as to the systematic position of the Pulicidse may be postulated as follows : — Do they or do they not, in the totality of their organs, show near relations of affinity with any of the other groups of insects ? In the former case we should have to arrange them in this group of insects ; in the latter we must establish an independent order for them. I naturally commence my examination with that order of insects which, in the judgment of zoologists, has the most right to receive the Pulicidse into it, namely the Diptera. The series of the Diptera must decidedly be called a unitary one ; but the two characters so often brought prominently forward (a perfect metamorphosis and suctorial buccal organs) do not alone establish this unity, seeing that we must also ascribe them to the Lepidoptera, the Apid^, and the male Dr. Karl Krapelin on the Pulicidae. 39 Coccidffi. Nay, even if we add the footless larvee and the fusion of the thoracic segments as further criteria, we might perfectly well unite the Bees with the Diptera. It is not the simple fact of the suctorial buccal organs that is of importance, hut their specific structure, the position and arrangement of the parts composing the suctorial apparatus. If we fix our attention upon this point we at once recognize that the fly's proboscis is constructed upon a perfectly different fundamental plan from that of the Apidee, that the two are not directly phylogenetically referable to each other, but that, on the other hand, the great variations in the buccal apparatus of the Diptera only represent modifications of one and the same type, distinctly demonstrable throughout. The characteristic of the bee's trunk consists in the development of the loivar parts of the mouth into the sucking organ, while the man- dibles retain their original function ; that of the fly's pro- boscis, on the contrary, in the employment of the labrum and hypopharynx for the formation of the sucking-tube, with which the mandibles and maxillae associate themselves as St} lets more or less developed as required, while at the same time the labium in all cases has to form a protective sheath for the comparatively delicate tube through which the fluids ascend. Tliis fundamental plan of the employment of the parts of the mouth occurs, as already pointed out in the introduction, in all the groups (except the Pulicidte) which have hitherto been placed in the group Diptera, in the piercing Culicidffi, Tabanidse, and Asilidas, the difterent families of honey-suckers, and the Pupipara, which are so depressed in position through parasitism ; nay, a bridge seems even to be thrown over towards the rudimentary buccal organs of the CEstridee, through the structures which occur in Gutereh^a. In figs. 1-3 (PI. III.) I have drawn transverse sections of the pro- boscides of those groups of flies which, upon one hand or the other, have been referred to as allied to the flea. While those of Tabanus and Gulex (figs. 1 and 3) agree not only in the position but also in the number of the pieces composing the proboscis, that of Melojyhagus (fig. 2, the representative of the Pupipara) shows a great reduction, which finds its expression in the entire absence of the mandibles and maxillas* ; but * The two valves embracing tlie proboscis of the Pupipara have been very erroneously interpreted as maxilla3, their palpi, or even as a bipartite epi pharynx (Meinert). From the whole arrangement of the proboscis, wliich is freely movable in a wide cavity of the head extending as far as the prothoracic ring, we can here have to do only with a conical prolongation of the head which has become paired, some- what such as we should obtain if we imagined the slight emargination at the apex of the frontal cone of Hhingia carried down to its base. The strongly projecting cheeks of many Conopidte might also perhaps bo regarded as analogous, 40 Dr. Karl Kriipelin on the Pulicidae. nevertheless it is easy even here to recognize the typical position of the pieces forming the sucking-tube (dorsallythe labrum and ventrally the hypopharynx) , and the labium which encloses these as a sheath. _ Further, the latter bears at the end that enlarged portion which is so characteristic of all Diptera, and which is probably to be interpreted as formed by uniarticulate labial palpi. The same unity in the Diptera appears also in the special structure of the thorax and its appendages. That this appears always separated from the head by a deep incision is cer- tainly not without significance ; but it can furnish no decisive datum for the collocation of the Diptera. Of more importance, no doubt, is the fusion of the thoracic segments into a compact thoracic mass, which occurs in all the forms referred to this group. It is indeed true that in orders of insects (I refer particularly to the llhynchota) the formation of the thorax as regards the separation or fusion of the segments composing it shows manifold differences, without its being necessary that we should separate forms which are united for other reasons, seeing that the fusion or separation of the thoracic segments has to do essentially with a function of the mechanism of flight, and the free segmentation of the thorax in a wingless form may very well be explained as a correlative phenomenon of adaptation. But the conditions are different if, on the contrary, a wingless form exhibits complete amalgamation of the tlioracic segments. In my judgment it thereby demon- strates most unmistakably its descent from winged insects, and in this sense the compact structure of the thorax, with the characteristic process of the mesothorax described as the *• scutellum," in Aklophagus^ the Nycteribiidag, and the Brau- lida3, decidedly acquires the significance of a still uneffaced relationship with the winged groups standing next to them. And just as on account of this character tlie assumption is justified that the forms just mentioned stand in close phylo- genetic r.-lationship with winged insects, so does the ex- amination of the dorsal a])pendages of the thorax lead to tlie same conclusion. All Diptera do not possess a pair of wings and a pair of halteres ; but the two organs which, because special, are certainly of such great importance in characterizing the Diptera, disappear so gradually in the continuous series of forms, that we may trace their progress to the rudimentary state, as it were, step by step. An Ornithohia pallida which, as Lipoptena cervi, follows a per- fectly different mode of life, enables us at once to understand the case, when we see Melophngus, which is never parasitic upon birds, entirely destitute of wings. But as regards the Dr. Karl Krapelin oyi the Pulicldse. 41 halteres, these, notwitlistancling Scliiner's assertion to the contrary, are quite recognizable in the sheep-tick, wliile in the Nycteribiida3 they show all gradations down to quite minute points, so that the complete absence of these apparently insig- nificant organs in the Braulldas need not give us any further disturbance. The ventral thoracic appendages, the legs, cer- tainly present but few differences in the group of the Diptera, nevertheless the five tarsal joints which are usually present are not always constant ; and further, other orders of insects sufficiently prove how little importance attaches in general to the number of tarsal joints and the development of the different sections of the legs. The developjnental stages of the Diptera do not show a community of type so distinctly as the structural characters just referred to. The larva? are certainly throughout distin- guished by the absence of jointed thoracic limbs, which is of special interest in the case of those forms which live free upon leaves by prey (many larvaa of Syrphidaj) ; but witli regard to the structure of the head, the armature of jaws, and the development of the traclieal system, there are, as is well known, such important differences, that they have been successfully employed for the systematic division of the order into several suborders and sections. Nevertheless even here intermediate grades are not wanting between the different structural characters (witness the variable development of the first cephalic segment) ; nay, in Brauer's* opinion, the family Lonchopteridas may possibly prove to be a perfect transitional group between the Orthorapha and Cyclorapha, so that the multifarious forms of the larvffi at least offer no veto against the unitariness of the stem of the Diptera. The same thing can also be said of the pupo3, which indeed likewise fall under two main types, but are so far brought together by Brauer's inves- tigations, that these furnish a proof that the so-called " tun- pupa?" (obtected pupee) show very different grades of structure, and in many of them the enveloping larva- skin bursts exactly as in the ordinary moulting, and consequently is to be referred simply to a delayed moulting at the close of the larval period. In the latter case, moreover, if the appendages of the segments of the body are not so closely attached to each other and to the body as in the naked and consequently less protected and more easily injured " mummy-pupa3," no important objection against the natural relationship of the two groups can be derived from this circumstance, which evidently results from * F. Brauer, ' Die Zweifliigler des Kais. Museums in Wien,' p. 9 (Vienna, 1883) ; also iu the Ueulischr. d, math.-naturwiss. lilasse d. k.-k. Akad. d. Wiss. Bd. xlvii. 42 Dr. Karl Krapelin on the Pulicidge. altered condition. The " mummy-pupEe," however, show- many differences amono- themselves with regard to the closer or looser appression of the appendages of the body, as may be demonstrated by a comparison of the pupee of the Asilidaj, which rest in the ground, and those of Tipulce which live in the water. Of anatomical peculiarities of the Diptera especial mention must be made of the " sucking-stomach," which is always present, as also of the large thoracic salivary glands, the eft'erent ducts of which, wherever the buccal organs perform any function, unite into an unpaired closed canal, which, running along in the cavity of the hypopharynx, opens at its extremity. The testes are almost always two ; the Malpi- ghian vessels almost as regularly four. As regards the tracheal system, the constant absence of the first thoracic stigma and the small number of abdominal stigmata are to be noticed ; while the nervous system, as is Avell known, shows all possible forms of development, from the most ex- treme concentration to a very considerable segmentation of the ganglionic chain. If Ave turn from this brief account of the Dipterous type to the characters of the Pulicida3, we must admit, in the first place, that in a whole series of points of comparison an agree- ment between the Diptera and the Fleas can be demonstrated. Like the Diptera, the Fleas have a suctorial buccal apparatus, a perfect metamorphosis, and footless larvee ; as in them also the tarsi are five-jointed, there are four Malpighian vessels, and one pair of testes. But, as has already been indicated at page 38, we could only ascribe decisive weight to this agree- ment if all these characters were peculiar to the Dipterous stem alone, and if at the same time, by more detailed comparison, real tenable parallels could be drawn between the different parts of the organs, as between the different stages of develop- ment. This, however, is by no means the case. The number of Malpighian vessels and of testes recurs in the same way in the Rhynchota, and therefore proves no more in favour of the relationship between the Fleas and the Diptera than the number of the tarsal joints or the annulation of the terminal knob of the autennce, which may be recognized in all possible groups of msects. At the first glance more importance seems to attach to the agreement of the two groups in the larval state, which in fact goes so far, that Brauer* has no hesita- tion about arranging the larva of the flea in his group of orthoraphal eucephalous Dipterous larva3. In opposition to * Brauei', "Km-ze Charakteristik der Diptereularven," in Verb, k.-k, zool.-bot. Ges. in VVien, 18G9, p. 846. Dr. Karl Kiapelin on the Pulicidss. 43 this, however, we must not forget that maggot-like larvae also occur in groups far removed from the Flj-type, in Hymeno- ptera and Beetles, and therefore cannot possibly be of decisive importance in judging of relations of affinity ; as also, on tlie other hand, that the pupa of the Fleas witli its quite separate Hmbs differs so much at least from the general type of the mummy-pupre, that from this very fact it has been attempted to set up a relationship of the Fleas to the Hymenoptera *. Hence the point of the question how far the analogous characters in Diptera and Pulicidgs depend upon true phylogenetic affinity would have to be sought in the investigation whether the construction of the sucking- apparatus is carried out in both cases on the same plan, i. e. with the same employment of homologous parts. That it is only from this discussion and from that as to the structure of the thorax and its appendages that a real decision of the question before us can be arrived at, may indeed be deduced from the consideration that in these organs we find the only characters which, on the one hand, are confined to the order Diptera, and, on the other, may be traced tlirougli- out their whole series of forms, and therefore must be regarded jcar' e^ox')*' ^S typical. The structure of the buccal apparatus of the Pulicidas has been very frequently discussed without the question of its relationship to the sucking-apparatus of other groups of insects having as yet been solved. Thus to cite only a few : — Dug^s f thinks that the proboscis of the fiea may be placed side by side with that of the Tabanidas, but also finds resemblances to the Hippoboscidae and Apidre. L. Landois| suggests a resemblance of the mouth-apparatus of the Puli- cidai to the rostrum of the Hemiptera ; while Taschenberg§, again, thinks he recognizes the Dipterous type, and espe- cially calls attention to the presence of a " tongue " as the most characteristic part of the moutli of a fiy. This ex- traordinary diversity of opinions is principally to be ascribed to the uncertainty of the interpretation of this very " tongue " of Taschenberg's. The mandibles, maxillas, and labium have long since been recognized with certainty ; but the un- paired piercer " (to express myself neutrally) has been referred to as the labrum (Westwood, HaJler, Bonnet), as the hypo- * As by Duges, hi his " Reclierches sur les characteres zoolog-iques du genre Pulex," in Ann. Sci. Nat. tome xxvii. p. 157. t Loc. cit, p. 151. i L. Landois, ''Anatomie des Ilimdeflolies," in JSTova Acta Acad. Leop.-Car. 1866, p. 56. § Loc, cit, p. ■11. 44 Dr. Karl Krapelin on the Pulicidge. pharjnx (Gerstfeldt), as the epipliarjnx (Karsten), and lastly, as already mentioned, as the '' tongue" (Savigny, Taschen- berg), and therefore all serious homologizing must have been prevented, the more, as even the real components of the suck- ing-tube were not made out with certainty. In figs. 10 and 13 I give two transverse sections through the anterior part of the Pulicid proboscis. Fig. 10 represents a section from Palex irntans ; fig. 13 a simihir section, but nearer the base of the proboscis, from Sarcopsylla penetrans'^ . The sections show at once that the structure of the sucking- tube in tlie two most distant groups of the Pulicidse is quite accordant. In both cases it is the mandibles hnd), which, in conjunction with the " unpaired piercer," form tlie true food- canal ; embracing the latter above and laterally, they join firmly together in the median line below. A glance of com- parison at figs. 1-3 shows that this " unpaired piercer " is hollowed into a groove on the underside exactly in the same way as the lahrum of the Diptera, and that to begin with there is no hypopharynx, but at the utmost perhaps an epipharynx. But if we trace the further course of this struc- ture by the aid of longitudinal and transverse sections it is easily seen that its upper covering immediately after its en- trance into the capsule of the head is in chitinous union with the upper margin of the arch of the head, while the inferior plate, i. e. the one which immediately forms half the sucking- channel, passes continuously into the chitinous covering-wall of the pharynx. Consequently we find in the organ in ques- tion precisely the same conditions as in the labrum of the Diptera, and there is no doubt at all that we have to do here with a true labrum. A connexion of this with the labium by means of a strongly chitinized, brown uniting piece, as asserted by Duges (/. c. p. 150), really has no existence at allf, and this removes the last possibility of regarding this struc- ture as a " tongue," i. e. as an extension or appendage of the labium. The interpretation of the " unpaired piercer," as labrum, being thus established beyond a doubt, the comparison of the proboscis of the flea with that of the Diptera can present no further difficulties. The employment of the labrum {Ir) as the unpaired covering lamella of the food-canal is apparently the same in both groups. But it is otherwise with the other components of the sucking-tube. In place of the horizontally- * The material was kindly sent to me from Assumption by my honoured colleague Dr. H. Toppen. t This chitinous piece rather forms the lever for moving the mandible, a.s will be shown elsewhere. Dr. Karl Krapelin on the Pulicidse. 45 placed mandibles of the Tabanidte and Culicidce, which, as is proved by those Diptera which do not pierce, are only secon- darily implicated in the closure of the sucking-canal, we see in the PulicidjB the vertically- placed mandibles, bent in towards each other laterally, appear as integral parts of that tube — a different inferior closure, such as exists in the hypo- pharynx throughout the whole group of the Diptera, being here entirely deficient. This absence of the hypopharynx, which, as is clear from what has been said, has as its conse- quence a totally different importance of the mandibles, and consequently a perfectly peculiar type of sucking-tube"^, proves in like manner of importance as regards the discharge of the salivary glands. The unpaired salivary duct in the lumen of the hypopharynx is replaced in the Pulicidaa by paired extremely fine half-tubes (fig. 13, s), each of which, running along the inner side of a mandible, may be traced from the basal part of the latter as a closed duct into the interior of the head, and, further, as far as the thoracic salivary gland !• Equally great differences in their arrangement and physio- logical importance may be demonstrated by a comparison of the other constituents of the proboscis of the flea with the homologous organs of the Diptera. A labium unpaired throughout its whole length, and at the utmost furnished at its apex with one-jointed terminal lobes, occurs nowhere among the Pulicidse, although something of the kind was formerly ascribed to Sarcoj^sijUa. The labium of Sarcojys^Ila at least presents (as fig. 8 may show) a biarticulation of the " palpi," even with an indication of further segmentation, so that in this point also the unity of the Pulicide group appears. This difference of the segmentation of the labium in Diptera and Fleas, with which a typical difference in the relative length of the unpaired basal part to the paired section to be regarded as palpi, goes hand in hand, can, however, hardly be so highly estimated in its phylogenetic significance as the further fact that the labium of the Diptera shows quite a different attach- ment to the head, and so has quite a different physiological value from that of the Pulicidse. In the former it generally attaches itself by its gradually widening base to a more or * Particular attention may here be directed to the two peculiar lateral lamellae of the labium, which apparent!}', by their elasticity, force the upper parts of the mandibles asunder, and thus bring about a closer appo- sition of their lower parts. t Kraft and Landois believe that they have demonstrated an opening of the thoracic salivary glands into the oesophagus not far from the region of the neck (see Landois, I, c. p. 18). 46 Dr. Karl Krapelin on the PulicidEe. less developed cephalic cone, with the upper lateral parts of which it is connected, and so is enabled from the base onwards to form that sheath of the delicate piercing apparatus (the two pairs of jaws as well as the labrum) which often arches together above so as to constitute almost a closed canal. In the Fleas, on the contrary, there is no such union of the labium with the lateral or upper parts of the head ; it simply articulates with a firm brown chitinous piece (fig. 9, cJi) in the median line of the lower surface of the head, and this union, as is well known, is frequently so loose that it is difficult to obtain Sarcopsyllce^ for example, with the labium preserved *. Hence, in its basal part, it does not form the sheath for the piercing-apparatus, but shows only a compara- tively shallow groove (fig. 15), which only in the ante- rior section of the proboscis, when the stem of the labium has become cleft into the paired palpi, becomes developed, at least in Pulcoc^ into two flaps, embracing the piercing-organ at the sides (fig. 10, Ip). But to make up for the deficient protec- tion of the basal part of the sucking-tube (and in this we have a fundamental deviation from the type of the Diptera) the maxillaj hava come in, originating as two broad plates from the whole length of the side of the head, and taking here, not only the constituents of the piercing-apparatus, but also the base of the labium, under their protection, as shown by fig. 15 in Pulex. We seek in vain for analogies to all these characters among the Diptera, and we may therefore be justified in asserting that all the parts of the Pulicide proboscis (with the sole exception perhaps of the labrum) differ so much in position and employment from the homologous parts in the Diptera, that we cannot well speak of direct phylogenetic relations between the two types of proboscis. We arrive at precisely similar conclusions as to the rela- tionship of the Pulicidaj and Diptera when we take into consideration the second group of characters peculiar to the Diptera, Avhich appear in tlie structure of the thorax and its dorsal ojypendages. Instead of the always freely movable head of the Diptera, we find a broad union of it with the pro- thorax in the Pulicida3 j instead of the compact thorax with its scutellum, Avhich is so characteristic even of the wingless Pupipara, we have three sharply separated thoracic segments, without a trace of any such dorsal mesothoracic process ; and instead of the pair of wings and the halteres, the latter of which are aborted only in the most extreme cases of parasitism, * Even in recent handbooks we may find the statement that the labium of Sarcopsylla is indistinct. Dr. Karl Krapelin on the Pulicldae. 47 there is nothing, absolutely nothing, that could lead us to conclude that the Fleas were formerly in possession of any such organs. Even tlie marked tripartite condition of the thorax ought a priori to have banished the idea of rudi- mentary wings ; nevertheless the older authors (Kirby, Dug^s, &c.) have fallen into the serious error of regarding separated lateral margins of the thoracic segments as such. But these *' processes of the pleura3," as Taschenberg * among others has conclusively proved, have nothing at all to do with wing- rudiments, and are to be regarded as characteristic structures sui generis. When Taschenberg therefore for this reason declares the generally-employed denomination of '' Aphani- ptera," founded upon this erroneous conception, to be inad- missible, we can only agree with him. It is only by giving up this name that we can seriously hope that the deeply rooted notion of the " Diptbres sans ailes," as Strauss-Durck- heim called the Fleas, will be completely suppressed. The wide gap which exists precisely in the most important characters between the Pulicida3 and the Diptera must have been made sufficiently evident by the preceding remarks. That it is also expressed in other systems of organs than those hitherto considered may therefore only be briefly indicated. The sucking- stomach, which apparently is met with in all groups of Diptera, is entirely wanting in the Pulicidfe j while, on the other hand, the proventriculus beset with nume- rous chitinous spines of the latter has no analogy among the Diptera. The sucking-mechanism of the pharynx or of the so-called " fulcrum " of the Diptera is formed by a single powerful pair of muscles 5 in the Fleas, on the contrary (as in the Rhynchota), a whole series of separate pairs of muscles (which, however, are interpreted by Landois as flexors and retractors of the labrum) are present for this function. Lastly, the presence of a stigma in the prothorax of the Fleas indi- cates more profound differences in the tracheal system 5 while as regards the simple ocelli of the Pulicidai and the deep lateral pits of the head, we may find analogous phenomena among the Rhynchota, but not among the Diptera. After all this the Fleas cannot well remain in the order Diptera. There remains then the investigation of the ques- tion whether they show near relations to any of the other groups of insects. Hymenoptera and Orthoptera, of which earlier authors have thought in this connexion, cannot well come into the question in the present state of our knowledge, as it would be opposed to all rational system to assert a reia- * Loc. cit, p. 21. 48 Dr. Karl Kriipelin on the PulicidEe. tionship of the Fleas to the Hymenoptera upon the sole accordance of the pupse, or to the Orthoptera upon the segmentation of the thorax. The order Lepidoptera also cannot agree in a single one of the more important characters with the Pulicidas, and thus there remains only the group Ehynchota for serious comparison. As a matter of course, considering the fundamental difference of development between Pulicidje and Rhynchota, we can hardly expect to find real intimate relations between the two groups, at least not so close as we must postulate for forms of one and the same order ; nevertheless 1 think I may indicate some points of view which deserve to be well considered in judging of the phylogenetic connexion between Fleas and Rhynchota. In the first place there can be no doubt that the order Rhyncliota does not even ap])roximately present a unitary type in the same degree as that of the Diptera. We find united in it animals with suctorial and masticating buccal apparatus, with perfect, imperfect, and without metamorphosis. The head is sometimes freely movable, sometimes attached by a broad surface to the prothorax. The thorax, so very uni- formly constructed in the Diptera, shows all possible stages of structure, from the enormous development of the sepa- rated prothorax in Scutata and Membracina, to the compact thorax sliowing scarcely an indication of segmentation of the Pediculina, or tliat of many Mallophaga more or less sharply divided into three distinct segments ; and like the thorax itself, its dorsal appendages also present no unity of type. With such polymorphism of almost all organs it is easily intel- ligible that we should be able to find in this Protean group ana- logies for a whole series of characters of the Pulicida\ Thus the segmentation and winglessness of the thorax in the Fleas may be without difficulty placed side by side with the similar con- ditions among the Mallophaga, which, at the same time, present examples of the antcnnary pits of the head already mentioned. The absence of facetted eyes in Pulicidffi agrees with what occurs in Coccida^, Pediculina?, and Mallophaga, the pupa enclosed in a cocoon unites them with the Coccida? ; the absence of sucking-stomach and the number of the Malpighian vessels and testes are even common to them and to all forms of Rhynchota. For the reasons above given, however, we must not ascribe a serious significance to all these agreements unless the Rhynchotan type sought for finds expression at least in the last of the cliaracters to be discussed, those of the buccal apparatus, and shows near relations to the homologous organs of the Fleas. According to the present state of our know- Dr. Karl Krapelin on the Pulicidae. 49 ledge it cannot well be maintained that there is a clear imi- tariness of structure in the buccal organs of the Rhynchota, as, at any rate among the Aptera (the Pediculina and Mallo- phaga), conditions occur which depart widely from those of tlie more highly organized groups. But as the arguments upon this point are not yet closed and I have made no investigations upon these lower forms, we must content ourselves with examining at least the sucking-apparatus of the Hemiptera and Cicadae in search of any agreement with the proboscis of the Fleas that may exist. With regard to the arrangement of the parts of the mouth in these higher groups of the Rhyn- chota, I have already published some statements in a previous note *, and these observations have since been confirmed and extended by Geise f- According to these the true sucking- tube of the proboscis is formed by the two maxillee closing laterally against each other into a double tube, while the mandibles are placed alongside of this tube as lateral piercing- setfe. From more recent investigations I do not hesitate to declare this view J so far erroneous that it is not the maxillne but rather the mandibles that interlock in the median line to form the sucking-tube (see figs. 11, 14). I am led to this changed interpretation of the two pairs of jaws in the first place by the fact that in transverse sections through the head the lateral setse finally come to be the lower ones, as, indeed, Geise correctly shows in his figs. 25 and 31. Secondly, I think that in the Cicadae I have found distinct traces of basal joints of the maxillae connected with the outer setffi. Fig. 12 shows the lower part of the face of a large tropical Cicada. On each side of the broad labrum (Ir) there is here an oblong plate ipl), which terminates almost in the middle line beneath the labrum in a blunt hairy tubercle and a peculiar whip-like appendage (fig. 6,/) . If this structure be prepared out of the head, a connexion, certainly only by articulation, with the lateral piercing sette may be easily demonstrated, for protrusion and retraction of which not only the chitinous sinews (fig. 6, sp and sr), but also the corresponding muscles (tig. 6, pm and rm) are attached to this chitinous piece. If this interpretation of the chitinous piece occurring in all Cicada3, I ulgorinse, &c., as the basal part of a jaw, perhaps even with palpiform appendages, be correct, this must, of * Zool. Anzeiger, 1882, p. 574. t G-eise, ' Die Mundtheile der Illiynchoten ' (Bonu, 1883). X On my part this resulted merely from what I now believe to be a wholly unjustified homologiziiig with the buccal organs of the Lepido- ptera, the sucking-tube of which is undoubtedly formed of the maxilla (see also Kirbach, Zool. Anz. 1883, p. 553). Ann. iic Mag. N. Hist. Ser. 5. Vol. xiv. 4 50 Dr. Karl Krapelin on the Pulicidse. course, be a maxilla, and thus the composition of the sucking tube out of the two mandibles would be finally decided. Bu then we should at once be enabled, in one point, to carry ou a corresponding comparison between the buccal organs of the Pulicida3 and Rhynchota, inasmuch as we need only suppose the labrum of the latter, which is indeed often enough developed into along, slender, stylet-like organ, to sink from above be- tween the mandibles*, in order to arrive at conditions which might perfectly well be placed side by side with those occur- ring in Pulicidse (compare fig. 11 with fig. 15). It appears further that upon the basis of my conception a connexion might be established between the modes of discharge of the saliva in the Pulicidge and Rhynchota, if we assume that the paired half-grooves along the inner side of the mandibles of the Pulicidge (fig. 13, s) have coalesced in consequence of the changed adhesion of these jaws, caused by the emergence of the labrum, as a constituent of the sucking-tube, into an unpaired efferent canal (figs. 11 and 14, s). The variable part taken by the two mandibles in Hemiptera and Cicadge (see fig. 14) in the formation of this salivary tube would come in support of this hypothesis. Among the lihyn- chota, as is well known, a hypopharynx is not developed as a separate organ, or only as a rudiment (in Cicadaj), so that in this circumstance also a parallelism between Pulicidge and Bugs may be found. The labium of the Rhynchota consists of four consecutive cylindrical joints furnished with a deep longitudinal groove along the upper surface. It has been said that it is destitute of pgilpi, but I think that this mode of expression is not correct. A labium divided into four or five successive rings is in complete contradiction to the plan of the organ deduced from the con- sideration of the masticating mouth. But notwithstanding Geise's assertion to the contrary {I. c. p. 11), there is nothing to prevent our regarding the cylindrical and often much more voluminous basal part of the labium as the submentura and raentum, as a direct continuation of which arise the multiarticulate palpi fused together in the median line. That there is really an amalgamation in the terminal joint of the labium is rendered probable by the circum- stance that both in the Hemiptera and in Cicadge a pretty * Geise asserts sometliing of the kind when he represents the labrum in C'oriva and Sigara as taking part with the constituents of the sucking- tuhe {I. c. p. 53, fig. 29) ; unf(jrtunately I must reject this assertion — welcome as it would be to me tor the homology attempted above — as positively erroneous. Dr. Karl Krapelin on the Pulicidse. 51 considerable notch appears at the apex*, although the side lobes thus produced are not jointed off from tlie unpaired piece in the same way as is usually the case, with the labella of the Diptera for example. But if this conception of the structure of the labium of the Rhynchota be correct, a comparison of it with that of the Pulicidse presents no difficulties. A fusion of the longitudinal fissure of the labium of Sarcopsylla (fig. 8), for example^ nearly to the apex, would essentially realize for us the conditions existing in Rhynchota (compare the labium of Cicada in fig. 5). And with this apparent equivalence of the parts an approximately similar physiological application of them would be associated. It has already been pointed out that the labium of the Pulicidie has undertaken the guidance of the sucking-tube only in its distal and not in its proximal part. But exactly the same thing may be asserted of the labium of the E,hyn- chota, which in the basal section of the rostrum shows an effacement of the dorsal furrow and decidedly turns down- wards, and thus devolves the guidance of the sucking-canal and of the piercing setge entirely upon the labrum. In the latter circumstance, indeed, there is an essential difference between the proboscis of the Fleas and that of the Rhyn- chota, as in the former the labrum, which has become one of the constituents of the sucking-tube, cannot possibly be employed to envelop the whole apparatus. But precisely this different application of the labrum renders intelligible a further fundamental difference between the two types of proboscis, which must be found in the physiological appli- cation of the maxilla3. In the sucking-tube of the Rhyn- chota, which^ under the double guidance of the labium and labrum, is sufficiently enveloped and protected throughout its whole length, the maxillte might, without damage, be brought in to complete the true piercing-apparatus ; they have become long thin structures, destitute of palpi. Hanking the sucking- tube. In the Pulicidse, on the contrary, in which the basal section of the sucking-tube, in consequence of the peculiar employment of the labrum, was destitute of an envelope, the maxillai^ developed into broad plates (figs. 4 and 7 and \b^m)^ had this important function of protection transferred to them. That under such a change of function the palpi * The section across the tip of the rostrum of Notonecta (fig. 14) shows the labium as consisting of two perfectly separate parts. Geise's state- ment that in Corn a the third and fourth joints of the labium are com- pletely cleft, depends, according to my investigations, upon an erroneous interpretation of the conditions coming into view at the tip of the rostrum. 4* 52 Dr. Karl Krapelin on the Pulicidge. also came to full development and importance, can hardly be regarded as a serious obstacle to the homology here attempted. The preceding indications will suffice to prove that in fact, without any great violence to the data given, a certain parallel may be drawn between the buccal organs of the Fleas and those of the higher Rhynchota, and that this comparison is at least far easier to carry out than that between the Pulicidge and the Diptera. If we bring the other agreements and diffe- rences of the three groups in question into the account, the result must be a phylogenetic alliance, although a distant one, of the Fleas with the Rhynchota rather than with the Diptera. But I repeat that the demonstrated relations certainly by no means justify a union of the two groups. The only possi- bility that presents itself is therefore to place the Pulicidse as an equivalent order Siphonaptera "^ side by side with the two most nearly allied orders. The entire series of insects with suctorial mouth-organs would consequently have to be divided in the first place into two groups, one of which (Hymenoptera, Lepidoptera) is characterized by having the lower parts of the mouth, maxillse, and labium employed in the formation of a sucking-apparatus, while in the other, on the contrary, it is almost exclusively the upper parts (labrum and mandibles) that are implicated in the formation of the true food-canal. This latter group would include the three orders Diptera, Siphonaptera, and Rhynchota, which 1 may, in conclusion, briefly characterize as follows : — 1. Diptera. Insects with perfect metamorphosis. Head free, with facetted eyes. Sucking-tube formed by a dorsal and a venti'al half-channel (labrum and hypopharynx), more or less enclosed throughout its length by the labium, which is bent up like a sheath and furnished with uniarticulate apical palpi. Mandibles deficient or styletiform, pushing in between the labrum and hypopharynx. Maxillge, when present, with palpi. Salivary efferent duct an unpaired closed canal in the interior of the hypopharynx. A " sucking-stomach." Tho- racic segments amalgamated, usually with a pair of wings and a pair of halteres. 2. Siphonaptera. Insects with perfect metamorphosis. Head attached to the thorax by a wide surface, without facetted eyes. Buccal organs suctorial. Sucking-tube formed * As the name " Aphaniptera " is inadmissible for reasons already given, and that adopted by Taschenberg-, " Suctoria," has already been employed twice, for a group of Oirripedes and for the Acinetse, I think it best to fall back upon Latreille's name " Siphonaptera." Dr. Karl Krapelin on the Pulicidge. 53 by a dorsal and two lateral channels (labrum and mandibles) . its anterior section only more or less enclosed laterally by the multiarticulate terminal palpi of the labium, and at the base, besides the latter, by the lamelliform palpigerous raaxillge. Salivary efferent ducts paired, developed as a channel along the inner surface of the mandibles. No "sucking-stomach." Thoracic segments free, without wings and halteres, with pleural processes upon the last two segments. 3. Rhynchota. Insects usually with imperfect metamor- phosis. Head free or broadly united to the thorax, with or without facetted eyes. Buccal organs usually suctorial. Sucking-tube (in the higher groups) composed off two lateral half-channels (the mandibles), only in the anterior portion enclosed by the labium and its apical multiarticulate palpi, which are united nearly to the apex ; at the base by the labrum. Maxilla styliform, without palpi, applied laterallv to the mandibles in the channel of the labium or the labrum. Salivary efferent duct unpaired, formed by two half-channels of the mandibles closing too-^^ther from the sides. No " sucking-stomach." Thoracic segments free or amalga- mated. Four, two, or no wings ; no halteres. EXPLANATION OF PLATE UI. The letters in all the figures refer to the same parts: — Ir, labrum; m d, mandibles ; m, maxillae ; m t, maxillary palpi ; /, labium ; I p, labial palpi ; /*, hypopharynx ; n, food-canal ; s, salivary duct. Fiy. 1. Transverse section through the proboscis of Tabanus, sp., anterior third. Fiff. 2. Transverse section through the proboscis of Melophaffus ooinus, middle. Fiy. 3. Transverse section through the proboscis of Culex pipiens $> middle. Fi(j. 4. Maxilla of Sarcopsylla penetrans, side view. Fiy. 5. Labium of Cicada, sp., side view. Fig. 6. Lower part of the maxilla of Cicada sp., and its union with a lamelliform appendage {pi) of the fore part of the head. /', whip-like process of the plate ; p m, protrusor ; /■ m, retractor of the maxilla ; sp and s r, the sinews belonging to them. At .v the sinew of the protrusor articulates with a chitinous rod which is perpendicular to the surface of the plate, and therefore does not appear distinctly in the figure. Fiy. 7. Maxilla of Pidex irritans. Fiy. 8. Labium of Sarcopsylla j^enetrans from above. Fiy. 9. Tjabium of Ptdex irritans, side view, ch, basal chitinous piece. Fiy. 10. Transverse section through the proboscis of Ptdex irritans, an- terior third. Fiy. 11. Transverse section through the rostrum of Notonecta ylauca, basal third. 54 Dr. O. Zacharias on the Development Fig. 12. Front view of tlie head of Cicada sp. i^l, plates with which the maxillae articulate. Fuj. 13. Transverse section through the proboscis oi Sarcopsylla penetrans, middle. Fig. 14. Transverse section through the rostrum of Notonecta glauca, apex. Fig. 15. Transverse section through the proboscis of Pulex irritans, base. VI. — New Investigations on the Development of the Viviparous Aphides. Bj Dr. Otto Zacharias *. Since the appearance of Metschnikoff's ' Embryologische Studien an Insecten ' (1866) the development of the embryo of the viviparous Aphides has not again been made the sub- ject of a monographic investigation. What the Russian author established with regard to the mode of development of the "pseudova" of Aphis Rosce and A. Pelargonii -passes pretty generally for all that is observable at present. Metschnikoff's description of the development of Aphides (at least in its fundamental features) is regarded as a " rocher de bronze," which presents no point of attack for an incisive criticism. This, however, is not the case, and I will, in a memoir that will appear very shortly, furnish the proof that Metschnikoff's description of the Jirst developmental stages (as far as the formation of the S-shaped germinal streak, and even some- what later) by no means agrees with the facts. For the subsequent stages I have also obtained quite different results of investigation, which I shall venture to summarize at the conclusion of this preliminary note. The observation of the embryonic development of the vivi- parous Aphides is for many reasons a difficult matter. Besides the minuteness and delicacy of the objects with which we have to do, there is a third condition which causes many obstacles to the investigation, namely the clearness and strong- refractive power of the protoplasmic contents of the egg. If in the case of the eggs of many other insects we have to contend with the obscurity of their yelk, it is in the Aphides the crystal clearness of the latter which frequently acts very prejudicially : prejudicially, inasmuch as under the circum- stances indicated the upper half of the egg constantly acts upon the lower half (or vice versa) ^ like a lens with a very short focus, and not only enlarges but also distorts those * Translated from the ' Zoologischer Anzeiger,' no. 168, May 26, 1884, pp. 292-296. of the Vivi'parous Aphides. 55 parts of the einbiyo which lie in a plane passing through the middle of the egg parallel to the object-slide. This opens up a rich source of illusions for those who come uncritically to the investigation ; but for those who are aware of the optical behaviour of the pseudovitellus there arises the unconditional necessity of correcting every surface-picture observed by the side view corresponding to it, and, if possible, by the other surface-picture (that directly opposite to the first). In the former case the embryo must be turned 90°, in the latter 180°. Without this method of rolling, already mentioned by Harting in his well-known work on the microscope, it is not possible to make out the earliest development of the ^p/^w-embryo. Of course it is not easy to practise the method referred to, and many a fine preparation is sacrificed by clumsy handling of the wire which is employed in producing the rotation. In my memoir I will describe in detail the rolling method as it should be constantly employed in the more delicate investi- gations in insect embryology. I will now briefly indicate in what principal points the results of my investigations differ from those obtained by the distinguished Russian naturalist. The pseudovum possesses no chorion, but only a vitelline membrane, Huxley's '^pseudo-vitelline membrane." This encloses the whole contents of the %gg^ which, at a certain early period (as Leuckart first remarked *), shows a distinction between peripheral and central cells. We read also in Huxley as follows : — " They [the pseudova] exhibit a central darkish matter surrounded by a clear cortex." Upon this point the sim'plest observation gives clear information. The deve- lopment of the embryo now starts from the '' clear cortex," the blastoderm^ which forms a multilamellar vesicle, and, indeed, in this way, that at its lower pole {i. e. that turned towards the vagina) a thickening is formed, from which the germinal streak grows forth laterally (and near to the inner wall of the blastoderm) in the form of a small thick tongue. The yelk at this time contracts strongly, and places itself, as a rounded mass, also at the inferior pole of the germinal vesicle. This is the profile view, so to speak. If we now roll the pseud- ovum through 90° we obtain a view en face ; and Metschni- koff appears to liave this alone in his eye when he speaks of a germinal and a vitelline " hill," the appearance of which characterizes the earliest embryonic stage of the viviparous Aphides. In the surface-view our glance of course falls first upon the broad side of the tongue, which now looks like a * ' Zur Kenntniss des Generatijusweclisela und der Parthenogenese bei den Insekteu ' (1868), p. 20, 56 Dr. O. Zacharias on the Development " hill," and behind it rises the contracted yelk, and also appears like a hill. This therefore explains how Metschnikoff came to the notion of a germinal and vitelline hill. But such a notion is not justified by the facts, and still less that of a special genital hill from which the reproductive organs are to originate. It is moreover quite incomprehensible that so practised an observer as Metschnikoff even then was could overlook the fact that the tongue of the germinal streak growing freely into the cavity of the blastoderm immediately shows a deep groove in its median line, becomes rounded off on both sides throughout its whole length, and thus produces two distinctly marked germinal pads. Metschnikoff, on the contrary, repeatedly remarks that the embryo in its early stage shows 710 trace of such pads *. This negative judgment I can only explain by the fact that Metschnikoff apparently does not practise the " method of rolling," and therefore did not get to see the different aspects of the embryo. In this opinion I am only strengthened by the examination of his figures 16, 17, 18, 19, and 20 (pis. xxviii. and xxix.), as also by the reading of the text relating to them (pp. 444-448). Every one who comes fresh and unprejudiced to the investigation of the development of the Aphides will make the surprising observation that the deve- lopmental processes which finally lead to the formation of the S-shaped germinal streak are not performed in the plane in which Metschnikoff places them in his figures, but in one standing directly perpendicular to it. In my memoir I shall produce the exact proof of this, and also furnish the requisite figures which I have found to be verified in hundreds of preparations. As regards the S-shaped germinal streak, which is well known to all investigators of Aphides, the inferior curve of this letter (which is turned to the left) represents the cephalic hood, the same structure, it may be said in passing, which Huxley, entirely mistaking the relative position of the Aphis- germ, characterized as the " abdominal hood." The upper curve (turned to the right) represents the rudiment of the abdomen, and the intermediate part contains the material for the forma- tion of the head and thorax. 'J'he limhs take their origin from a special superficial layer, the so-called limh-plate (" Extremitatenplatte "), as to the origin of which Metschnikoff has not got at the truth eitlier in Simulia and Corixa or in Aphis Rosa; (see his paper, loc. cit. pp. 400, 427, and 448). For the Aphides I have * Zeitschr. f. wiss. Zool. Bd, xvi. (1806), pp. 44b aud 450. of the Viviparous Aphides. 57 succeeded in proving that the so-called " Extremitatenschicht " is a part of the blastoderm which at a very early period enters into an intimate fusion with the true germinal streak. The details of this are treated of in my memoir. Besides the two vertex-plates (Huxley's " procephalic lobes ") which originate from the primitive lateral plates, we have also to distinguish in Aphis-embryos a median plate^ which is produced from the ventral part of the cephalic hood. I would name it the mandibular plate^ as the two mandibles are formed from it. The first and second maxillce originate from that part of the limb-plate which overlies the procephalic portion of the germinal streak, or is amalgamated with the latter. The arrangement of the three pairs of buccal organs is of such a kind that a hexagon is formed, the outer angles of which are constituted by the first maxilla. Later on the rudiments of the mandibles and first maxillse are enclosed in the depth of the head, and from them originate the " retort-shaped bodies " (of Metschnikoff), which secrete the chitinous stylets of the rostrum. In the mature embryo we perceive two such bodies on each side^ not one only, as Metschnikoff' s figures show. By the demonstration that it really is by the transformation of the mandibles and maxillce that the retort- shaped bodies are produced^ the parts of the mouth of the Aphides are first brought into homology with the corresponding organs in other insects. According to Metschnikoff, as is well known, the mandibles and first maxillae are completely retrograded, and the ^' bodies " secreting the piercing seta3 originate quite newly. This was a priori very improbable, and observation gives quite another result. By carefully crushing half- mature embryos the actual conditions may often be very beautifully brought into view. Witlaczil, who has treated of the anatomy of the Aphides in much detail in a recent memoir *, gives a very full descrip- tion of the structure of the retort-shaped bodies as studied by him by means of sagittal and transverse sections through fully- developed animals. For orientation in many difficult points of the development of Aphides, e. g. as to the question whether or not the Mal- pighian vessels are present in these animals at any time, I have turned to the allied group of the Coccidge, and not without success. Thus I distinctly saw in Coccus hesperidum that the brown masses of substance corresponding to the secondary vitellus in Aphis Rosce arrange themselves very * ' Zur Anatomic der Aphiden ' (Wien, 1882). 58 Dr. W. Djbowski on the early in the form of two long cords, which open close together into the intestine in the region of the rectal section. Accord- ing to this observation T do not hesitate to adopt Witlaczil's view, according to which the green cell-mass in the abdomen of the viviparous Aphides (which is likewise arranged into two cords) represents the Malpighian vessels of other insects. In a very pale Aphis-embryo I was able clearly to detect, besides the very distinctly marked dorsal vessel, the point of convergence of the two cords, but without distinctly seeing the point of discharge (as in Coccus hesperidum) . The &g^ of the viviparous Aphides, for which I have here and there, for convenience' sake, employed the antiquated term " pseudovum," therefore presents exceedingly interesting and very distinctly observable developmental processes, which cannot be sufficiently studied. In recent times Dr, Arnold Brass* (of Leipzig) and Dr. Ludwig Willf (of Kostock) have occupied themselves with the earliest stages of develop- ment. The last-mentioned gentleman has also already an- nounced the publication of a work upon the later stages of the Aphis-embryo. In course of time many other workers will certainly have to be registered for this highly interesting subject. VII. — Notes on the South-Russian Spongillidge. By Dr. W. Dybowski \. Dr. p. T. Stepanow, Professor of Zoology in the University of Charkow, has had the kindness to send me some specimens of the freshwater sponges for scientific investigation. The sponges preserved in the Museum of the above University and those collected by Prof. Stepanow himself are from the following localities : — 1. From the river Udy (a right affluent of the Siewiernyj Daniec), Gov. Charkow. 2. From Lake Lebiedin (Circle Lebiedin), Gov. Charkow. 3. From the river Kolomak (left affluent of the Worska, a branch of the Dnieper), Gov. Poltawa. * " Das Ovarium uud die ersten Eutwicklungsstadien des Eies der vivi- pareii Aphiden," in Zeitscbr. f. Naturwis*. Bd. iv. (1882). + " Zur JBildung des Eies iind des Blastoderms bei den viviparen Apliiden,'' in Arbeiten des Zool. Instit. zu Wiirzburg, 1883, Heft :'>. X Translated by W. S. Dallas, F.L.S., from the ' Sitzungsbericlite der Naturforscher-Gesellscbaft bei der Univeisitat Dorpat,' Bd. vi. pp. SOT- SIS (1884). South-Russian Spongillidae. 59 4. From the river Siewieruyj Daniec, Gov. Cliarkow. 5. From Lake Wielikoje (Circle Lebieclin), Gov. Gharkovv. Offering- my best thanks to the sender of these sponges, I now communicate the results of my investigations upon them. These results are as follows. The Spongillse of these five localities represent three species, namely : — A. Spongilla lacustris. B. Meyenia [Ephydatia) fluviatilis. C. Dosilia (?) Stepanowiij n. sp. The first two of these * I have only briefly described ; but the third, as a form hitherto unknown to science, has received as accurate and detailed a description as possible. 8pong{lla lacustris^ Carter, occurs in very thin lamellte, coating in spots or cones the leaves of Acorus Calamus and Que7x.us s]). ; they are evidently quite young and undeveloped specimens, in which I have consequently found no gemraules. The skeleton-spicules are 0'224-0"130 millira. long, O'OIO- 0*002 milHm. thick f. The size of the parenchyma-spicules varies between 0*050 and 0*060 millim. in length, and 0*002 and 0-004 millim. in thickness. These spicules perfectly agree with those described by me {loc. cit.). Among the spicules such deformities frequently occur as I have already described and figured {I. c.) ; but among the most peculiar are the clavate and pin-shaped spicules, the heads of which are sometimes quite smooth and rounded, but sometimes variously misshaped, or furnished with small spines |. Localities. Daniec, Kolomak. The Meyenia Jluviatilis, auct., agrees perfectly with that described by me {l. c. p. 13). I have before me four frag- ments of a large, fully developed sponge and a couple of smaller ones. In all numerous gemmules are present. The fragments belong to cushion-like sponges ; the smaller sponges, on the contrary, are growing round a stem, 8 millim. thick, of Arundo sp. Localities. Lake Lebiedin, river Udy. * See Dybowski, " Studien iiber die Slisswasser-Schwamme des Russi- schen Reiches," in Mem. Acad. Sci. St. Petersb. ser. 7, tome xxx. no. 10, p. 6, tab. i. figs. 4, 6, 7, and p. 13, tab. i. iig. 3, and tab. ii. fig. 9. t In my memoir (I. c.) p. 10, column 1, line 4 of the measurements, 0-009 is erroneously printed instead of 0-002, and in column 3, line 6, 0-028 instead of 0'002, X The figures necessary for the more ready intelligence of all the_ de- scriptions here given, I have prepared with the aid of a Hartnack's prism, and will publish them when opportunity serves. 60 Dr. W. Dybowski on the Dosilia (?) Stepanoivu, n. sp. This Spongilla, from Lake Wielikoje, is of very peculiar interest. So far as I know, no form similar to our sponge has hitherto been described among the European Spongillidae, at least, I have been unable to find any notice of it in the literature accessible to me ; on the other hand, among the exotic (American and Asiatic) Spongillidse, I find analogous, and, it seems to me, nearly allied forms. The genus DosiHa, Gray*, which Carter f places in his " Meyenia,^^ possesses two species — Dosilia plumosa (from Bombay) and D. Baileyi (from New York). The latter appears to me the form most nearly allied to our sponge J. To justify and support my opinion I will here give as accurate a description as possible of our sponge, and then place side by side the most prominent characters of the two (Kussian and American) sponges, so as to facilitate for other authors the comparison of the two sponges with one another. Description. — Of the sponge under consideration I have before me six small spirit-specimens, all ol which are defec- tive and do not enable us to form any definite notion of their form. They are chiefly shapeless masses, growing round various foreign bodies (such as leaves and stalks of grasses, fragments of bast, very thin twigs, and even quills of a small wing-feather). The specimens in spirit are pale tawny, and look not unlike soaked bread. The skeleton-spicules are long and slender acerates with acute ends, that is to say, they have the form of the spicules proper to the Spongillge with smooth spicules in general ; but they are somewhat smaller, their length being 0-200-0-104, and their thickness 0-065-0-004 millim. The surface of the skeleton-spicules is, however, not smooth, but furnished with very short, acute, and exceed- ingly scattered spines §. * J. E. Gray, " Notes on llie Arrangement of Sponges, Avith the De- scription of some new Genera," in Proc. Zool. Soc. Loncl. 18G7, pp. 550- 553, pis. xxvii. & xxviii. t Carter, " Historj- and Classification of the known Species of Sjmn- gilla," in Ann. &- Mag. Nat. Hist. ser. 5, vol. vii. (1881), pp. 78-107, pis. V. & vi. \ Neither of the two species mentioned is known to me by autopsy. I have draAvii my conclusions as to the affinities of our sponge only from the statements in the literature, and must therefore for the present abstain from a certain and final decision. § The above-mentioned spines are so small and inconspicuous that they may very easily be overlooked. They are most conveniently observed with the light of an oil-lamp and with the objective no. 8. When they have once been observed they may quite easily be recognized with objec- tive no. 4. The spines appear most distinctly at the periphery of the spicules. South-Russian Spongillidse. 61 The skeleton-spicules of the sponge under notice constitute, as it were, a transitional form between the smooth and the spiny Spongillid spicules*. The parenchyma-spicules are also remarkably peculiar and characteristic. They are small acerates, measuring 0*040-0'050 millim. in length and 0"02o- 0"010 millim. in thickness, and furnished with spines ; in form they generally resemble those of Spongilla lacustris^ but the form and arrangement of the spines are quite different from those in the spicules of the latter Spongilla. Here the spines occur in three different forms : the middle portion of the spicule is beset with long, obtuse, vertically projecting spines, but the two ends present small, pointed booklets, while towards the middle of the spicule there exist pointed erect spines. The long straight spines of the middle section are frequently also covered with small acute spines ; the free end of the spine is sometimes rounded off, sometimes truncated, and sometimes furnished with a knob or a T-shaped rod. The gemmules I have not been able to find, but I found in the parenchyma of the sponges numerous amphidisci. The spindle-shaped amphidisci are very long ; their dimen- sions are as follows : — millim. Total length 0-040-0-028 Thickness of the shaft 0-002-0-004 Diameter of the disk 0-012-0-010 The shaft is furnished on its surface with large perpen- dicular spines. The two terminal disks {disci) are furnished at the mar- gins with deep notches ; the teeth thus formed have a per- pendicular position. The actual form of the disk, as well as the number of teeth on each disk, I have been unable to ascertain. If we summarize the most important characters of our sponge and compare them with those of Dosilia Baileyi^ it appears that the two sponges must belong to the same genus, but are specifically different. These characters are as follows : — Dosilia Baileyi^ Carter (/. c. p. 95). " Coating, surface smooth. Structure friable, crumbling. Skeleton-spicule curved, subfusiform, gradually sharp-pointed, smooth. Flesh-spicule minute, curved, fusiform, gradually sharp-pointed, covered with erect obtuse spines throughout, extremely small towards the extremities, and extremely long and perpendicular about the centre of the shaft. Stato blasts * See Dybowski, /. c. tab. i. tigs. 3 & 5. 62 On the South-Russian Spongillidse. globular ; aperture infundibular 5 crust, which is thick and composed of granular cell-substance, charged with birotulate spicules consisting of a long, straight, sparsely spiniferous shaft whose spines are large, irregular in length, conical and perpendicular, terminated at each end by an umbonate disk of equal size, deeply but regularly denticulated, whose processes are claw^-like and turned inwards, arranged perpendicularly, with one disk resting on the chitinous coat and the other form- ing part of the surface of the statoblast. " Locality. New York. In a stream on the Canterbury Road, West Point." Dosilia Stepanowii^ n. sp. Surrounding ; surface smooth. Skeleton-spicules long, pointed, and covered with small, acute, but scantily distributed spines. Parenchyma-spicules small, pointed, and covered with spines. Spines in the middle section long, obtuse, and per- pendicular ; at the two ends small, acute, hook-like ; towards the middle small, acute, perpendicular. Amphidisci spindle- shaped, their shaft long and furnished with a few large per- pendicular spines. The two terminal disks are toothed at the margins. The teeth have a perpendicular position. Locality. Lake Wielikoje. In conclusion, it may be mentioned that I found in the parenchyma of the Spongillidaj just investigated by me some siliceous corpuscles, which it seems to me may be small paren- chyma, or coating spicules of still unknow^n Spongillidse. From this I conclude that other unknown Spongillidte must certainly occur in the waters of the Government of Charkow and of South E-ussia generally. Would it not be advisable on the part of the University of Charkow to make a prize-problem of the investigation of the South-liussian Spongillidge ? It would be a very grateful theme for a pupil of that University, the solution of which might advance the knowledge of a group of animals which is still but little known not only in Russia but also in the rest of Europe. Further, I may call attention to the fact that very numerous Diatoms and Algee are present in the parenchyma of the Spongillid^, so that these j)lants, otherwise so difficult to discover, are to be sought in the interior of these sponges. Postscript. — The Spongilla from Lake Hertha (in the island of Kiigen), kindly communicated to me by Dr. Braun of Dorpat, pi-oves, from my investigation of it, to be a Spon- Mr. R. Rosenstock on Heterocerous Lepidoptera. 63 gilla lacustris, auct. It agrees perfectly with the specimeus of that species obtained from the Ludwinow estate (see Siissw.-Schw. d. Russ. Reiches, p. 6). This i'act is of interest as furnishing a small contribution to the zoogeography of the Spongillas, especially as, so far as I know, no Spongillse were previously known from that locality. VIII. — On the Synonymy of some Heterocerous Lepidontera. By Rudolph Rosenstock, B.A. I INCIDENTALLY discovered and noted the following synonyms while systematically studying the collection of Lepidoptera in the British Museum. They are for the most part redescrip- tions by the late Mr. Walker of species previously described either by himself or other authors. 1. NOCTUITES. Poaphila congesta, Walk. Vene- = zuela. Remigia triangularis, Walk. N. = India. Anthophila erecta, Walk. San Domingo. Toxocampa costimacula, Walk. Sylhet. 2. Pyealites. s. = Hypena discliisalis, Walk Africa. Marimatlia coiifisinalis, Walk. = Loc. ? Pyralis dispansalis, Walk. San = Domingo. Lepyrodes lepidalis, Walk, Cey-^ Ion, N. ludia. Stenia pipleisalis, Walk. Sierra i^_ Leone. '' ~ Hymeuia meridionalis, Walk. S India. Botys hortalis, Walk, Bogota, = Santarem, strictalis, Walk. N. Ame-" rica. olliusalis, Walk, U. S America. ofellusalis, Walk. Loc philealis, Walk. Venezuela. = senippialis, Walk. codrusalis, Walk. Bogota. I _ Bogota, j ~ Hypena seiiialis, Gum. Central Africa. Anthophila semipiirpurea, Walk. Loc. ? Carcha hersilialis, Walk. Sau Domingo. Samea (^Ouen.) sidealis, Walk. Sierra Leone. ( This is evi- dently an Old- World species of wide range.) Botys marialis, Walk. San Do- mingo. flavidalis, Walk. N. Ame- rica. - lycialis, Walk. San Do- mingo. - dorisalis, Walk, Villa Nova. 64 Mr. R. Rosenstock on Heterocerous Lepidoptera. Botys semizebralis, Walk. S. In- = Botys amyntusalis, Walk. Cey- dia. Ion. convectalis, Walk. S. In- dia. \ = neoclesalis, Walk. Cape. suspicalis, Walk. Ceylon. niemmialis, Walk. Loc. = campalis, Walk. Jamaica, ? San Domiugo. ogmiusalis, Walk. San = gastralis, Guen. San Do- Domingo, mingo. cinctipedalis, Walk. Geor- = oxydalis, Guen. U. S. gia. America. Ebulea beronalis, Walk. Hon- = acastalis, Gtien. Honduras. diu-as. Spilodes helvialis, Walk. U. S. = apertalis, Guen. N. Ame- Anierica. rica. Botys gnomalis, Walk. San = Omiodes humeralis $, Guen. Domingo. San Domingo. peleusalis, Walk. San Do- = acta that it is '•' valde similis et affinis Tachince cinerece." Schiner is of the same opinion. As regards the mode of penetration of the larva into the body of the Carahus, we must, in all probability, consider that the fly de- l)Osits its egg in the stigma, and the larva, escaping from the egg, bores through the wall of the trachea and gradually extends its body into the body-cavity of the beetle. During this time there are formed around the larva, on the part of the hypodermal layer of the trachea, chitinous deposits, which are strongest in the vicinity of the abdominal wall at the hinder extremity of the larva, and here form a brown cup, the margin of which, however, passes without any sharp boundary into the translucent chitinous lobes which sur- round the rest of the body of the larva. ^ A little while ago, Jules Xiinckel d'Herculais described a parasitic fly {Gi/rnnosoma rotumlatuin), the larva of which lives in the body of Pentatoma §. In this case also the larva has its hinder extremity turned towards the stigma, aad this end is embraced by a chitinous cup, called by Kiinckel " le siphon." Kiiuekel, however, thinks that " le siphon " is a secretion of the larva itself, and by no means a pro- duct of the hypoderm of the infested insect. Kiinckel also describes the mode of penetration into the body in a different fashion, namely, that the fly sticks its eggs to the ventral segments of the Pentatoma, and the escaping larva penetrates between the ventral segments into the abdominal cavity and only by degrees becomes connected with the stigma. However this may be with regard to the Penta- toma and Gijmnosoma rotundatum, in Carabus the case is most probably as I have suggested. The penetration of the larva through the stigma is in this instance evidenced by the fact that even the very smallest larvie are attached to the stigma, and that they are onlv met with on the stigma. — Zooloyischer Anzeiyer, no. 1G9, June 9. 1884, p. 316. * Ann. Sci. Nat. tome x. p. 248. t 'Ivroyer's Natm-historisk Tidsskrift,' 1838. See also Erichson's ' Bericht iiber die wiss. Leistuugen im Gebiete der Eutomologie im Jahre 183s; BerUn, 1840, p. 93. X ' Diptera Scaudinavite,' tome iii. pp. 1038-1039. § Aim. Soc. Ent. Fr. ser. 5, tome ix. (1879). THE ANNALS MAGAZINE OF NATUEAL HISTOEY, [FIFTH SERIES.] No. 80. AUGUST 1884, IX. — Notes on Species o/'Ascodictyon and Rliopalonariayv-oyu the Wenlock Shales. By George Robert Vine. In the ' Annals and Magazine of Natural History ' for June 1877, Prof. H. Alleyne Nicholson and Robert Etheridge, jun., published their joint paper on ^^ Ascodictyon, a uqw Provi- sional and Anomalous Genus of Paleozoic Fossils." In that paper (pp. 463-468) the authors describe forms from the Devonian (Middle, of Ontario) and Carboniferous strata (of Scotland). In a paper read before the Geological Society and ultimately published in their journal (" Notes on the Polyzoa of Wenlock Shales," Feb. 1882), I placed upon record (p. 54) the name Ascodictyon Jiliforme, Vine, as a provi- sional one. Since the publication of that paper I have been closely engaged on the study of Silurian and Carboniferous Ascodictya^ and I find that the forms that I originally placed under the above name may conveniently remain. The or- ganism, however, is such a peculiar one, and my opportunities of studying its varied aspects so singularly fortunate from the possession of a large series of specimens, that I make no apology for adding further details of the species to the brief notice already referred to. The genus Ascodictyon is Palaeozoic, though not peculiarly Ann. & May. N. Hist. Ser. 5. Vol xiv. 7 78 Mr. G. R. Vine on Species of so. In examining a series of Cretaceous Polyzoa in the possession of Miss E, C. Jelly, one specimen appeared to me to belong to the genus. In describing an American Silurian form — in some respects similar to forms found in the Wenlock Shales — Mr. E. O. Ulrich ( Journ. of Cincinnati Soc. Nat. Hist. April 1879, pp. 18, 19) has established a genus under the name oi Rkopalonaria. The species of the genus, B. venosa, Ulr., the author places with the Crisiidffi, remarldng, " that the form has only been observed incrusting Streptelasma corniculum. On account of the great delicacy of the fossil, the fronds themselves are rarely found ; but instead we find a series of impressions on the exterior coat of the Streptelasma, which very well represents the fronds and cells of the same." A specimen of the species described by Mr. Ulrich is before me. It is from the same Cincinnati rocks ; as I shall have to refer to the genus again, I have thought that it might be more satisfactory to make reference to an actual specimen than to the mere description of the same. ASCODICTYON, Nicholson & Etheridge, Jun. Ascodidyon, Nich. & Etli. jim., Ann. & Mag. Nat. Hist., June 1877, pi. xix. ; " Notes on Polyzoa of tlie Wenlock Shales," Vine, Quart. Journ. Geol. Soc. Feb. 1882, pp. 52, 68. 1. Ascodictyon filiforme^ Vine. Ascodidyonjiliforme, Vine, " Notes on Polyzoa of the Wenlock Shales," o/j. cit. pp. 54, .56 (merely referred to in the above). Organism filamentous, forming linear, contorted or clustered threads, adherent to shells, stems of crinoids, fragments of trilobites, but rarely to corals. Filamentous threads hollow, but surrounded by delicate calcareous walls ; the hollows filled with a dark brown granular mass. Lagena-Wko. divergences developed on the sides of the thread, sometimes as single vesicles, otherwise as groups of vesicles. Peculiarly clustered stellate fibres are also formed at unequal distances. Locality. Buildwas beds, generally distributed throughout the whole of the washings, but more abundant in nos. 36 and 38. This curious organism begins its existence as a mere speck upon stone or shale or stem, which forms the nucleus of a colony. From this delicate filaments are developed (fig. I. 1 a, a^), sometimes in two or three, at other times in four different directions ; these vary in thickness, but the average size, both in breadth and depth, may be taken as measuring between y^ and 2^0^ inch. The threads are sometimes, but rarely, white, more generally of a dark brown tint. Viewed as Ascoclictyon and Rhopalonaria. 79 __opaque objects on shell-fragments they appear like fine hairs laid in lines across the surface, or contorted, crossing- each other at different angles, or running in parallel lines ; but it would be useless to direct attention to any special feature in their modes of growth, for they vary considerably ; but one Fig'. I. 1. ^5co^«rfyo??^fo/orm<', Viue, y|-o inch thick. Two filiform threads run- ning almost parallel. The point of origination is a^ ; the portion marked a is nearly similar, but wanting the nucleus ; «'- single vesicle. 2. A. radiciforme, Vine, -fig inch thick. One of the filiform contracted threads, placed here for comparison. 3. Thread of A.filiforme, drawn from a transparent specimen, showing dark brown pulp. feature which appears to me to have something to do with the development of the colony must not be lost sight of. Occa- sionally some of the filaments bifurcate, and before bifurcation takes place the organism contracts at intervals, and out of this contracted portion a new thread originates. It must not be supposed, however, that this simple explanation is, on the whole, a plainly satisfactory one. The two different sorts of threads, though apparently allied, give rise to two distinctly separable colonial growths in the after stages of their exis- tence. Yet in the earliest, or initial stages, it is not easy to distinguish the difference between tliem. It will be better, however, to keep the two forms distinct. In a few cases I have been able to reduce the thickness of the shell to which a typical A. Jiliforme is attached, and have mounted the specimen in balsam on glass. It is then seen that the filament is hollow, and the central parts filled with a dark brown granular matter. This granular matter, which I 7* 80 Mi'c G. R. Vine 07i Species of shall call the pulp, is sometimes continuous, at other times slightly separated from the adjoining mass, when the grains appear like a row of beads dotting the centre of the threads (figs. I. & II. 3 to 5). Surrounding the pulp are delicate and transparent walls, which are distinct and clearly defined. In a few cases the filament bifurcates and the pulp in the undivided portion separates at the node and passes into the divisions thus formed. Heal development takes place at irregular inter- vals. On the sides of the thread there is a slight protuberance of the wall, which increases in size until a roundish knob or a lagena-like vesicle is formed. Into these vesicles, which may be either single or in groups, the pulpy mass passes ; but to what extent the colony increases beyond the A.-siluriense stage, I am unable to say ; although I believe we may safely regard this species, at least, as the ultimate outcome of the colonial development of A. jiliforme, Vine. Fiff. II. 4, 5. Ascodictyon filiforme (transparent and semi transparent), showing vesicles and pulp. 6. Ditto, sliowino- clusters of vesicles, passing into 7. A. silunense, Vine. At this stage of inquiry it may be well to ask, whether the name A. silurtense, Vine (Wenlock Polyzoa, oj). cit. p. 52), should be suppressed. If the Silurian name is suppressed the Devonian name [A. stellatum, Nich. & Eth.) must be suppressed also, unless the one name may be allowed to embrace both the type and the varieties. In opposition to this view, I think the wiser course would be to allow the forms to retain their present names as given below, for the simple reason that it would be perhaps impossible, or almost impos- sible, to make another collection similar to my own unless the Ascodlctjon «nJ Ilho[)alonana. 81 sliales were searclied with the same minute care that I have bestowed upon those furnished to me by Mr. George Maw. Then, again, ^. siluriense and^. radicifonne were the tirstfruits of my hibours; and A. jiliforme was the result of closer exa- mination. As with me so with others, because the matured forms will, I fancy so at least, be considered by the student of micropalajontology as by far the most important as initial stages in an inquiry like the present one. Another conside- ration is with me of much greater weight than any previously given. It may be, after all the care that I have exercised in thus tracing the origin and development of a colonial growth, that some few facts or fossil illustrations may have been over- looked, and it may be necessary at some future time to limit the type now characterized as A. filifomie. If this should happen, the suppressed names would have to be restored. In his observations on A. stellatum Prof. Nicholson remarks {op. cit. p. 465) , that in its youngest stage the organism " pre- sents itself simply in the form of scattered oviform or pyriform calcareous vesicles attached to the exterior of foreign bodies. When mature it consists of similar vesicles combined into clusters.*' I do not doubt the accuracy of Prof. Nicholson's observations, though I cannot, on the whole, endorse them from my own labours : all the vesicles are united to the filiform thread, though in a few isolated instances " apparently " they are not so. I have examined a large series of A. filiforme for the purpose of putting the observations of Prof. Nicholson to the test, and I am consequently unable to confirm his views. In his description of fig. (3, pi. xix. {op. cit.), the author says, " Four young vesicles (?) of the same {A. stellatum) &c. ;" the vesicles are not foraminated, as in the other figures (2 to 5^ pi. xix.) , and in this special feature the Ontario vesicles are allied to, though not identical with, those shown in fig. II. of the present paper as gradational stages in the development of>4, Jiliforme, Vine. 2. Ascodictyon stellatum, Nich. & Etli., jun., var. siluriense, Vine. =Ascoclictyon stellatum, Vine, Quart. Journ. Geol. Sue. Nov. 1831, p. 618. = Ascodictt/071 stellatum, var. siluriense, Vine, Q. J. G. Soc. Feb. 1882, p. 52. The details of this species have been given in the works referred to above. 82 Mr. G. E. Vine on ^jpecies of 3. Ascodictyon radiciforme^ Vine. = Ascodicf 1/071 radians?, Vine, Quart. Journ. Geol. Soc. Nov. 1881, p. 619. = Ascodictyon radiciforine, Vine, Q. J. Geol. Soc. Feb. 1882, p. 53. When I wrote the details already given in the papers re- ferred to above I did not possess the fine suite of specimens which has enabled me to extend my remarks on the origin, growth, and apparently final (?) development of this peculiar group of organisms, found as yet only adherent to other organisms in the Wenlock shales. In speaking oi A.filiforme^ I incidentally referred to a single form apparently related to the S])ecies, but which I said it would be best to keep separate. I have now to consider the relationship of that form, but which, not being the earliest in the development of the species, I will defer till later on. In fig. III. 5 I have sketched the fragment of acrinoid stem, magnified about 2 diameters, on which one of the finest of my colonies of the earliest stages of this species [A. radiciforme) is attached. I do not, however, found the whole of the evi- dence which I am about to bring forward on a single specimen. I have corrected both the positive and the negative evidence by appeals to between sixty and seventy other specimens ; but as this one affords me evidence of continuous growth, I have built up my description mainly on it. And here I must be allowed to say that, contrary to my general plan in drawing, the whole of my sketches are drawn by the eye, and not by the aid of the camera lucida j but every figure is a faithful delinea- tion of the original. In fig. III. 1, 2, and 3 I have drawn a fibre which is conti- nuous on the fragment of crinoid stem already referred to. At 2, a and a', we have two nuclei, either of which may be referred to as the originating nucleus of the colony ; but it will be best to speak of them separately. In fig. III. 2, a, a kind of false stellate cluster is formed ; but this I regard as only an offshoot ; the true nucleus is at a', and at this the central part only. It appears to me that the central nucleus originates by the combination of minute spores, which up to the present have escaped my observation ; but immediately after the combination, delicate prolongations of the central mass or radii are sent off which do not in every case produce fibrous threads. At a one of these minute combinations is formed, but only one of the rays is apparently developmental *. * It will be understood by the palaeontologist that in restricting myself thus I only take the evidence presented to me. I have not the least doubt but that -what I sa}^ of one might be said of all the rays ; but I have only the one evidence to rely upon — the positive. Ascodictyon and Rhopalonaria. 83 This extends for a short distance, when another, or pseudo- combination, takes place, the result of which is shown, so far as I am able to trace it ; but in that portion which recedes towards a" the fibre is slightly contracted at certain distances. Ym. Ill, 1 to G. Varied forms of Ascodictyon radiciforme, Vine (already described in text). In fig. III. 1 the fibre is white, undoubtedly hollow, with here and there a broken surface, showing the dark brown pulp (or matrix) below. In fig. III. 3, which is a continuation of fig. III. 2, at a" we find the continuation of the contracted fibre and 84 Mr. G. U. Vine on Species of other features which it may be well for the reader to refer to^ the stellate cluster being a continuation of the fibre at 3*. In following the above remarks it will be easily understood that I attach very great importance to the method of formation and the character of the tongue-like vesicles which form the radii of this anomalous species, A. radiciforme. Sometimes after single vesicles are formed, at other times after a combination of vesicles, the organic matter of the thread or fibre undergoes other changes, to which reference has not been made. A group of vesicles combined as in fig. III. 3*, some of the cells, and occasionally the whole of them, will contract towards the centre or separate from the nucleus, forming club-like cells. In nearly all the instances where I have seen this mode of con- traction the cells have a very delicate covering of calcareous matter, and are foraminated either in a single line along the centre or indifferently over the whole surface ; the latter^ however, is a very rare occurrence. I have given figures of the club-like cells (fig. III. 4 a. Ah), showing the direction and positions of the foraminated surfaces ; but in fig. III. 4 I have shown the basal attachment of one of the fibres, which shows that the under surface was more densely foraminated than the upper. The irregular fibre shown in fig. III. 6, though not strictly speaking a portion of a stellate cluster, is likewise foraminated. In the ' Journal of the Cincinnati Society of Natural History; 1879, voh ii. pi. vii. figs. 24 and 24 a, Mr. E. O. Ulrich describes and figures a very peculiar fossil adherent to the exterior coat of StrepteJasma cornieulum. For the species a new genus is founded — BJiopaJonarlaj from rhopalony a club — which Mr. Ulrich places in the family Crisiidai ; and he says that " the genus is related to Hi'p'pothoa, but in the form and arrangement of the cells they differ widely." There is not, however, any relationship to Hi'ppotlioa in the species described ; but as the cells are somewhat club-shaped, it may be well to accept the genus if more fully defined. In the species, li. venosa, Ulr., the " cells are uniserial^ long, acutely elliptical, and joined together at their contracted ends. . . . Cell-mouths not clearly determined, hut appear to he situated near (he oniddle of the celV f. In my own speci- men of this species I cannot detect any cell-mouths, but the other characters are well defined by the author. In the Ludlow rocks and also in the Wenlock shales there are several fossils that might have been conveniently placed in this genus if my specimens did not throw soiije little light t Italics mine. Ascodictyon and Ehopalonaria, 85 upon their origin and development. In fig. IV. I Lave given sketches of four distinct types of very common forms fomid in the shales. In fig. IV. 1 we have the rarest of these, showing elongated cells very similar to some of the cells of Stomato- pora elongata, Vine (fig. IV. 2), only that the cells are in an opposite direction to what (apparently?) is the case in >S'. elongata. I cannot, however, detect any orifices in these cells such as we have in 8. elongata. Another form, very common indeed, is shown in fig. IV. e3. Only that the cells are not club-shaped, I suggested for this type R. boteUus, as descriptive of its peculiar sausage-like character. But even of this type better evidence is afforded by Fi^. IV. 6^- 1, 3, 4, 5, RJiopalonaria, ITlricli : species described in text, 2. Stomatopora elongata, Vine. fig. IV. 4, in which both the boielloid and the rhopahid charac- ters are shown in one colony. Some of the cells in the last figure liave, one would suppose at first sight, cell-mouths. This is not so : those cells that are shown thus are a little more cal- careous than the others ; the walls are broken at this part, and the dark brown matter is shown below. I have some few specimens of the last two types, and the whole of the walls or outer covering is destroyed, and the hotella-V\ke matrix is still adherent to the fragments to which the original organism was fixed. In fig. IV. 5 we have a fourth and 86 Mr. G. R. Vine on Species of last type, which is simihir in some respects to R. venosa, Uh-ich *. In addition to my own observations I have the evidence of a most careful observer as to the existence of RhopaJonaria in the Ludlow rocks. In 1881, J. D. Longe, Esq., of Chel- tenham, sent me a series of sketches (fig. V,), with the Fio'. V. Rhopulonaria, Ulrich, from tlie Ludlow rocks. Figures supplied by J. D. Longe, F.G.S., Cheltenliam, following remarks : — " I also enclose a sketch of a very abundant, encrusting, creeping * stoloniferous ' form, which I have on shells (Sjnrifer) from some Upper Silurian bed .... probably Ludlow." These beautiful forms are different from any known to me in the Wenlock shales, and their publica- * Not similar to his figures, but similar to some cells seen on the spe- cimen of Ulrich's species in my o-wii cabinet. Ascocllctyon and Rhopalonaria. 87 tion with the above details may be the means of lielping others in their researches among the micaceous and other shales of the Ludlow series of rocks. My labours are entirely confined to the Wenlock series and some only of the shales over the Wenlock Limestone. I do not think it necessary to speak more fully now of the genera Ascodictyon and Rhopalonaria, It is important, how- ever, in vindication of my remarks, to make some reference to the various opinions on the organisms which Prof. Nicholson gives in the section of his paper entitled " Systematic Position and Affinities " {I. c. p. 466). He says, in the first place, that Dr. Strethill Wright (to whom Scotch specimens of A. radians, Nich. & Eth., were submitted for examination) '' was unable to throw any light upon their nature." Prof. Huxley, to whom the same specimens were submitted, after considerable hesitation, suggested that they might be Protozoa. Mr. H. B. Brady, after a protracted examination of both the Scotch and the American forms, has arrived at the conclusion that they cannot be referred to the Foraminifera. Some of the American specimens [A. fusiforme, N. & E., and A. sfeUatuin, N. & E.) were kindly submitted by Mr. H. B. Brady to the Rev. Thomas Hincks, who suggested that they were possibly allied to the recent Anguinarioi. Neither Prof. Nicholson nor Mr. Etheridge expresses any positive opinion as to their systematic position or affinity. The difficulties encountered by these various authorities when speaking of this remarkable group are valuable so far in helping to establish the uniqueness of the types submitted to them ; but none of the suggestions help to throw light upon their nature and affinity. Yet I have respected the whole of the remarks, and have compared specimens of the fossil species with specimens of every known living type suggested as " probable " by these authors, but without any definite results. Perhaps it would be wise to pause here, for it is not for me to suggest possible affinities when so many experts have failed. Yet 1 cannot allow the paper to pass out of my hands without making a suggestion, which may possibly share the same fate as the others. There are not, so far as I am aware, any Cyclostomatous Polyzoa which may be considered as truly stoloniferous. Some of the Hydrozoa are ; but I know of none whose stolons are adherent to stone or shell, such as are found in these ancient rocks, neither am I aware that the stoloniferous Ctenostomatous Polyzoa are adherent to stone and shell, like Ascodictyon or Illiopalonaria. Yet it seems to me that we have, in Ascodictyon Jiliforme ai least, primitive representatives 88 On SpecAes o/"Ascodictyon rtwc? Rliopalonavia. of the stoloniferous Vesiculariidse, such as Vesicularia and BowerhanMa^ or, possibly, some member of the more humble race of the Entoprocta. Barrois has already, in his ])aper " On the Embryogeny of the Cyclostomatous Polyzoa " (Ann. & Mag. Nat. Hist. Nov. 1882, p. 402), spoken of a pro-Bryozoan race, composed of " free swimming organisms." May Asco- dictyon be the attached, or larval tbrm, of some of the as yet unknown pre-Upper- Silurian types of organic life, polyzoan or otherwise ? There is, however, another suggestion which may help to throw some little light on the development of A. radiciforme^ Yine, though we cannot hope by the comparison to explain away all the difficulties which surround the subject. I refer now especially to some remarks contained in a paper by Mr. George Busk, F.E.S., entitled "Notes on a peculiar Form of Poly- zoa closely allied to Bugida, =Ki?ietoskias, Kor. & Danielsf e i " (Quart. Journ. Microsc. Soc. vol. xxi. new ser.). After speaking of the development of the various species of Kineto- skias, Mr. Busk says, " I have yet scarcely adverted to the most rem.arkable feature of Kinetoskias, viz. the peduncle or stem, which appears to exist in all species. . . . The mode of formation of this part of the zoarium, which is undoubtedly the homologue of the bundle of separate radical tubes so commonly met with among the Polyzoa, is extremely curious and interesting, and, at the same time, in some points as yet more or less obscure, as, in fact, 7nay he said resjjecting the mode of formation and development of the more ordinary form of radical tubes'^. " In the more common forms they are cylindrical, jointed, chitinous tubes, with rather thick walls and with very scanty contents, beyond a few granular particles and irregular threads, representing, as it would seem, the remains of an endosarc, with which, in order that their progressive increase in length, and occasionally complicated branching &c,, may be effected, we must suppose the tube to be furnished. In fact it is otherwise impossible, without assuming the presence of a germinal material, to account for the fact that even after the tubes have attained a considerable length the extremity, or a considerable part of the tube, may undergo great changes in form, as is seen in the production of hooks and other means of ensuring adhesion to foreign bodies, changes showing a most extraordinary adaptability to circumstances. Not the least remarkable of these adaptations is the division of the extremity of the tube into a multitude of yqtj minute tubular * Italics mine. On two new Species of Walckeiiaera, Blachw. 89 filaments, each of which may be traced into independent con- nexion with small foreign bodies," Had Mr. Busk been writing" of Ascodictijon instead of species of Kinetoskias^ he could not have given more faithful descriptions of some of the specimens found in the Palaeozoic rocks. The suggestive inference to be drawn from these remarks, and others that might have been given, is that the dark brown masses (pulp &c.) in the various species of Ascodictyon are probably the remains of endosarc in these once living filaments and semi- tubular and bulbous tubes. 'J'here is just one other point in Mr. Busk's paper to which I will direct attention in conclusion, because it will help us to understand and appreciate at its proper value Ascodictyon and the abortive or " blind cells " of Rhopalonaria : — " That the radical and connecting tubes, like the avicularia and vibra- cula, represent modified zooids, is, I believe, generally admit- ted ; nor can it be denied in this case [Bugula &c,) that each successive joint or internode is a distinct zooid." And in a note the author says, " In Bicellaria and in Notamia it may almost be said that the inhabited part of the zooecia is simply a dilatation at one part of the internode of a radical tube, which is continued to the ultimate extremity of the branch." X. — Descriptions of two new /S^ec2*e5 0/ Walckenaera, Blachw. By the Rev. O. P. Cambeidge, M.A., C.M.Z.S., &c. [Plate IV.] Among a large number of Spiders sent to me for identification, dming the last three or four years, by Major-General A. W. M. van Hasselt, from Holland, are two of the curious genus Walckenaera, Bl., which I believe to be undescribed. Their discoverer having kindly permitted me to do so, I now sub- join descriptions and figures of these novelties. Family TlieridiidaB. Genus Walckenaera, Bl. Walckenaera Hasseltii, sp. n. Adult male, length ^V of an inch, or f of a line. Gephalothorax rich black-brown. Legs yellow, tinged with orange-brown. Perhaps i 1 some examples they would be bright orange-yellow. 90 Rev. 0. P. Cambridge 07i two Abdomen jet-black, tliinlj clothed with short fine yellowish hairs. The upper part of the caput is slightly raised above the ordinary level, and presents a flattish summit, the middle of which forms an oval distinctly marked by a surrounding groove ; the hinder slope or occipital portion looked at in profile is short and abrupt. The profile resembles in this respect pretty nearly that of W. Beckii^ Cambr., and a not very strong, small, narrow-oval, slightly curved indentation or fovea runs backwards from just above each lateral pair of eyes a little below and parallel w^ith the upper margin of the caput ; in the middle of the ocular area are a few short strong upturned hairs. The cephalothorax is short, almost round, and the thoracic region is somewhat flat and its sur- face rugose, while that of the caput is smooth and glossy. Eyes small and very indistinct, placed on the fore part of the caput, one pair (the hind centrals) on the anterior edge of the raised portion in a transverse line about a diameter apart from each other ; immediately beneath them is a slight trans- verse indentation, and just below each extremity of it is a lateral pair placed obliquely, and midway between the two lateral pairs (and contiguous to each other) is the fore central pair. The height of the clypeus is about two thirds of that of the facial space. Legs slender, not very long nor very unequal in length, 1, 4, 2, 3, furnished with very short fine hairs only. Palpi short, similar in colour to the legs, excepting the radial and digital joints, which are dark yellow-brown. The radial joint is similar in length to the cubital, but much stronger ; it is of a rounded spreading form and has a short bifid apophysis bent abruptly inwards at its fore extremity on the outer side. The digital joint is rather large, oval. Palpal organs highly developed, prominent, with several spines and corneous processes ; a very long, slender, filiform spine issues from the middle of their outer side and curves round beneath with a long, free, sinuous, exceedingly slender, hair-like point, and another short black spine is curved in a circular form at their extremity. Falces small, conical, directed strongly backwards, and of a deep yellow-brown colour. Maxillce and lahium of the ordinary form common to the genus, and similar in colour to the falces. Sternum large, short, heart-shaped, convex, and of a glossy •deep black-brown colour. An example of this distinct spider, which I have great neio Species of Walckenaeva, Blachio. 91 pleasure in naming after its discoverer, was sent to me in 1880 by Major-General A. W, M. van Hasselt from the neighbourhood of the Hague, Holland. This species appears to be nearly allied to W. sordidata^ Thor. ( W. atra, Bl.), a spider I have never seen ; but the description of it is not sufficiently close to the spider now described to justify the conclusion that the two are identical. The profile also is much like that of W. erythropus, Westr. ; but the caput is less elevated and the palpi totally unlike. It is also allied to, but I think quite distinct from, fV. elegans^ Cambr., a Bava- rian species (P. Z. S. 1872, p. 7G6, pi. Ixvi. fig. 23). Walchenaera nemoraUo'ideSj sp. n. Adult male, length ^V of an inch ; adult female yV of an inch. In size and form, as well as in the coriaceous punctured upper surface of the abdomen, this little spider closely resembles W. nemoralisj Bl. ; but the colour of the cephalothorax and abdomen is blacker and that of the legs is a clearer yellow than in that species. The two may also be more readily distinguished by the form of the radial joint of the palpi. The apophyses of this joint are in a similar position, but the outer (tapering) one is much larger, longer, and more prominent, being double as long as the joint, slightly curved, projecting outwards at right angles to it, and very slightly hooked at the point. The other apophyses are very similar to those of W. nemoralis^ but proportionately larger and of a different form, that in front being constricted near the middle. In W. nemoralis the outer apophysis is not only shorter and less strong, but consists apparently of two parts, a basal por- tion, prolonged and ending Avith a very fine, sharp, somewhat thorn-like addition. Examples of this spider have been sent to me at different times during the last two years by Maj.-Gen. A. W. M. van Hasselt from Holland. Mons. Simon, on examining one of these which I forwarded to him, considered it to be only an example of W. nemoralis, Bl. ; but the differences above noted (as well as some other minor ones) are so constant that I do not feel the smallest doubt of its being specifically distinct. I have received also one example of the typical W. nemoralis, B\., ^, from Holland; but the species now described has not yet been found in Great Britain. The female resembles the male in colours, but the occiput is simply a very little gibbous when seen in profile, and the height of the clypeus is rather less than half that of the facial space. 92 Prof. F. J. Bell on Pentastomum polyzonura. EXPLANATION OF PLATE IV. Fig. 1. Walckena'era Hasse.Uii. a, spider, magnified ; b, ditto, in profile, withont legs or palpi ; c, slio^litly perspective view of caput ; d, caput, from in front, showing tlie position of the eyes ; e, left palpus, from in front and rather inside ; /, natural length of spider. Fig. 2. Walckena'era nemoralio'ides. a, spider, magnified, S\ b, ditto, in profile, with legs and palpi removed ; c, caput, from in fi-ont, showing the eyes ; d, left palpus, from in front and rather on the inner side ; e, radial joint of palpus ; g, 2 in profile, without legs or palpi ; h, natural length of spider ; A:, genital aperture, $ . Fiff. 3. Walckena'era nemoralis, Bl. Part of palpus of c? . XI. — A second Note on Pentastomum poljzonum. By F. Jeffeey Bell, M.A. In the sixth volume of the current series of the ' Annals ' (pp. 173-176) I published a short note on the rediscovery of t\\Q. Pentastomum poly zonum of Harley, two female specimens of which had been acquired by the British Museum in 1880. Lately we have received other specimens which formed part of the collection of the late Dr. Edwards Crisp, but are without any indication of origin* and not in first-rate condition. A short time since an interesting essay on the structure of Pentastomum was published by Mr. W. E. Hoyle in the ^Transactions of the Hoyal Society of Edinburgh' (vol. xxxii. pp. 165-191), in which he describes a new species (P. ^;ra- telis) , and gives an account of its anatomy. Mr. Hoyle was fortunate enough to have examples of both sexes of the parasite, and he describes the male as being 13-17 millim. in length, and as having sixteen or seventeen annuli. Of the two specimens which formed the basis of my former note neither was male ; of the seven specimens now received one is a male, and I have been able to observe that it, while measuring 36 millim. in length, has only seventeen rings, and that the most anterior of these are much less pro- minent than they are in the female. In addition, therefore, to the numerous points of similarity indicated by Mr. Hoyle, we have another in the smaller number of annuli in the male than in the female. Another point is to be observed in the * Although a careful search has been made in Dr. Crisp's collections, there are no indications of the Pentastomum annulatum of Baird, which did, I believe, on the dispersal of the Zoological Society's museum collec- tions, pass into the hands of Dr. Crisp. It is greatly to be wished that this type should be found. Mr. R. Hitchcock on the Causes of Variation. 93 fact that of the six female specimens now before me two have twenty, while the others have only nineteen annuli ; in other Avords, the result to which I was led {torn. cit. p. 176), a good deal to ray surprise, as to the great value of the number of rings in the body, is a little shaken, although it falls in rather with one's general experience as to the specific value of num- bers such as these. It is to be noted, further, that the two females with twenty annuli measured respectively 75 and 80 millim., or less than three with nineteen rings, which measured 90, 95, and 105 millim. ; a specimen of 46 millim. in length had nineteen rings. The fact that the male has seventeen annuli, while that of P. protelis has sixteen or seventeen, and the discovery of the fact that the female of P. poJyzonum is not absolutely limited to nineteen rings, diminishes the gap that separated the two species, Mr. Iloyle being apparently inclined to give as much importance as I did to the seeming constancy of the number of rings in the female. While these considerations, then, tend to the union of the species P, protelis with P. 2)oltjzonum, the fact that Iho two animals, the small carnivore and the voracious snake, do live in the same area gives a clenching force which, to my mind, is almost irresistible. XII. — The Causes of Variation. By ROMYN HiTCHCOCK*. The recent studies of Dr. W. B. Carpenter upon Orhitolites^ are of special interest, owing to the remarkable manner in which the stages of variation and development have been traced. The monograph by Dr. Carpenter, published in the Reports of tlie ' Challenger ' Expedition, was the subject of some remarks recently made by the writer before the Biological Society of Washington, in which an effort was made to explain how such a simple sarcode organism as the animal Orhitolites has been led to produce a shell of complex form. Dr. Carpenter regards it as the expression of a ]iot understood " progressive tendency along a definite line towards a higher specialized type of stnicture in the calcareous fabric." This, however, is merely a statement of the facts observed, and in no wise assists in their explanation. Elsewhere it may be gathered from the author's words that he regards the * From the * American Journal of Science ' for July 1884, pp. 49-52. t Phil. Trans, part ii. (1883). Ann. & Mag. N. Hist. Ser. 5. Vol. xiv. 8 94 Mr. R. Hitchcock on the complex shell as significant of a "7;/an so definite and obvious as to exclude the notion of ' casuaV or ' aimless ' variation." The facts seem capable of a somewhat different interpreta- tion, which seems more in accord with our present knowledge of simple organisms, and quite sustaining the views of Darwin that " plan," in the sense used by Dr. Carpenter, should be superfluous. For if there be an inherent tendency to variation among these organisms, as Dr. Carpenter seems to believe, how do we explain the persistence of the original Orbitoline type, 0. tejiuissima? Biologists seek to discover the causes of variations which they observe ; but it seems not less important that the persistence of types should also be explained. 0. temiissima is a very ancient species, and surely any inherent tendency to change would have mani- fested itself during the long period of its existence, even under unfavourable conditions. The observations I have to offer may be said to relate entirely to change of environment ; but their tendency is to demonstrate that the changes observed in the shells of this family are not due to any inherent tendency resulting in a definite plan, but that they are due to causes easily understood. It is far from my intention to deny a definite plan of growth to these organisms. But plan of growth does not imply that there have been causes acting within the organism — special tendencies of the protoplasm toward higher structure. It seems to be such an assumption that has led Dr. Carpenter to speak of a "not understood" progressive tendency, &c. In my opinion the causes of such progression as can be observed are easily understood ; and the i)lan of growth becomes a natural consequence of these causes, which are purely physio- logical, and independent of any supposed tendency to varia- tion. While Dr. Carpenter, on the one hand, seems to regard variation as due to an inheient tendency of the protoplasmic body, the writer, on the other hand, attributes it entirely to the more or less favourable conditions of life of the different species. Moreover, I am quite unable to understand how any inherent tendency to variation impressed upon the sarcode could fail to find expression in some differentiation of the sarcode, which in the cases in question has not been ob- served. The same view seems to be held by 0. Sclnnidt, wlio, in his ' Grundziige einer Spongien-Fauna des Atlantischen Gebietes,' alludes to Dr. Car})enter's previous studies, and compares the changes observed in the Sponges and Forami- nifera. lie says the changes in the latter are found in the general habit of the form and the variable grouping of Causes of Variation. 95 tlie chfimber-systcms, while among the Sponges the variation is in tlie microscopic detail. " One may speak of the micro- scopic form of Foraminifera, but not of microscopic elements." The complexity of the shell is merely in the multiplicity of chambers and the manner of their intercommunication. The process of growth, even in the complex 0. complanata, is in all respects identical with that in other species, and in no essential feature differs from that of PeneropUs. What Dr. Carpenter designates as a " higher specialized type of struc- ture " does not represent an advanced degree of specialization in any part ; nor can we discover any advantage to the organ- ism arising therefrom. It is true there is an advance in com- plexity ; but unless accompanying this there is an evolution in function, or unless it results from some effort of adaptation which confers some benefit upon the organism, it seems not proper to regard complexity of shell-structure as a proof of biological advancement. Seeking for an explanation of the cause of the increased complexity of shell-structure, so beautifully illustrated in the Milioline family, the writer was led to the conclusion that it is entirely due to the favourable conditions of life and the abun- dance of food available. It is true, as already said, this may be regarded as a mere statement of the influence of environ- ment causing variation ; but a careful consideration of the subject will show that there is a broad distinction between environment as a cause of variation and adaptation to environ- ment ; for in this case we are unable to perceive any benefit to the organisms arising from their adaptation to changed conditions. If it be said we can seldom discover the benefits supposed to be derived from adaptation, it may be answered that it is usually possible to infer how the changes observed may prove beneficial. In the case under consideration, however, an examination of the changes that have taken place does not indicate any possible benefit to the organism. The multipli- cation of chamberlets necessitates very intimate intercommu- nication for the transference of food and the continuation of the processes of life. The organism is not thereby better adapted to its surroundings, but is made more dependent for its existence upon the continuance of the favourable conditions under which it has developed. The advance in complexity — the multiplication of chamberlets — would only be possible under the most favourable conditions, for all the nutriment received by the interior segments must be col- lected by the sarcode at the margin of the shell, and the necessary food could only be obtained where the supply was 8* 96 Mr. E. Hitchcock on the Causes of Variation. abundant. It may be conceived tliat if 0. complanata were placed in situations less favourable as regards food it would die of starvation, owing to the quantity of inner sarcode requiring nourishment, while 0. tenuissima needs only more favourable conditions as regards food and, perhaps, tempera- ture to become as highly complex in structure as the last- mentioned species. As a further proof of the influence of environment leading to changes which cannot be regarded as special adaptations, in the usual meaning of the word, the forms of Of compla7iata found on Fiji reef are especially characterized by thick plicated margins, as though growth proceeded with too great rapidity to produce symmetrical disks, and these forms are associated with the largest repre- sentatives of the species. The distinction above referred to seems an important one, which, if it lias already been recognized, has not been promi- nently brought forward in the writings with which I am familiar. Before the Biological Society the subject was briefly considered in the following Avords : — " liegarding the subject from this point of view, we are led to examine more closely the relations between the spiral and the cyclical methods of growth. Their intimate relation is only noticeable when we observe how one has been derived from the other. When the spiral growth of Orhiculina pro- duces a complete circular disk, further spiral growth becomes impossible ; and if we concede that the extrusion of the sarcode to form successive chambcrlets is due to nutrition and growth, the cyclical plan then becomes a necessity. In this way it may be supposed cyclical growth originated, purely a result of nutrition, not by adaptation to environment, but as a result of it ; not because such growth is or ever was better adapted to the conditions of life. " We find here a steady course of variation a result of physiological ])rocesses, independent of those external causes to which Ave are accustomed to attribute such changes. These variations, as successively produced, have been perpetuated through inheritance, until the plan of growth has, in some species, totally changed. Herein, therefore, we may find an indication of how the plan of growth originated, and a sugges- tion that the inscrutable laws wliich govern the progress of evolution may each have beginnings equally simple, and not beyond the range of human insight to discover. Evolution in this case seems not to be a result of a definite plan of growth, but the plan of growth is the result of physiological pro- cesses. However great and important the influences of environment and selection may have been in the production On the Vertebrate Zoology of Persia. 97 of genera and species, perhaps the attractiveness of the idea and tlie ease with which it enables us to dimly uniL'rstand many biogenetic problems permits us to lose sight of other influences more obscure, but of equal importance in the history of life." This view of the subject seems to derive still farther support from the geographical and bathymetrical distribution of the species. Without entering into a lengthy discussion of this part of the subject, it may be said that as a rule the more complex species are found in the warmer waters under conditions most favourable to the activity of nutritive pro- cesses. As an example, the very large specimens of O. coinplanata from Fiji reef may be taken. On the other hand, the ancestral form 0. tenuissima still inhabits the colder and deeper waters, retaining the simple characters of its earliest known condition. XIII. — Additions to the present Knowledge of the Vertebrate Zoology of Persia. By James A. Mukuay. Since the publication of Mr. Blanford's valuable work on the Zoology of Persia (1876), giving a complete list of the animals inhabiting that country^ nothing, I believe, has been published as an additional contribution, except a single paper in the Proc. Zool. Soc. for 1881, which added five species to the already large list of reptiles ; these are Agama jiersica^ Scincns conirostris, Hydrophis temporalis, Gatacldcena dia- dema, and Hydrophis cyanocincta, the first three being newly described species. The Kurrachee Museum, having now rather an extensive collection of Mammals, Birds, and Reptiles from Eastern Persia — very kindly made for the institution by Mr. W. D. Gumming, of the Persian Telegraph, during the past three years — and having also acquired a collection, comprising thirty- six species of Reptiles and seven Mammals (also from Persia) — made, it is said, by a member of some foreign exploring commission in 1876-77 — I am enabled, after careful examina- tion of these materials, to add a few more species to the existing knowledge of the Vertebrate fauna of the country. For the collection said to be made by a member of some foreign commission, the institution is indebted to Mr. Poss- man, also of the Persian Tclegraj)h. Although this collection dates as far back as 1876-77 the specimens are in an excel- 98 Mr. J. A. Murray on the lent state of preservation ; each specimen has been carefully labelled, giving the date, year, and localities of capture, the latter being chiefly Bushire, Tanjistan, and Charbagh, near Bushire. Among Mammals there is nothing new, but a few species, of the occurrence of which Mr. Blanford seemed to doubt and which are comprised in the collection, are noted below. 1. RMnolophus ferrum-equinum, Schreb. Rhinolophusferrum-equinum, Schreb., Blf. E. Pers. ii. p. 19. \S'- Bushire, 26. 11. 83. 2. Ursus tliihetanus^ F. Cuv. Ursus, sp. ?, Blf, E. Pers. p. 47. Vrsus yedrosianm, Blf. J. A. S. B. xlvi. pt. 2, p. 317 ; P. A. S. B. 1870, p. 4. Mr. Blanford records this from Beloochistan, on the assu- rance of the natives of the country. Major Mockler very kindly procured for me three skins with skulls of the animal inhabiting the Beloochistan hills. These with two other skins from the Sind hills were those of Ursus thibetanus. There is now a live specimen in the Kurrachee Zoological Gardens, from the Sind hills. 3. Deljihinus plumheus, Duss. Not recorded in Zool. E. Pers. The Kurrachee Museum has two skulls from Lingah. 4. Dipus Blanfordi^ sp. nov. Dipus macvotarsuH ?, Wag-uer, Blf. E. Pers. p. 74. 1 (^ , 1 ? adult (pregnant) , and two adolescent : Bushire, July 1882. Four juv. : Tangak, May 1877. One adolescent, Tanjistan, June 1877. These agree in every particular with Mr. Blanford's de- scription. The long black tuft of hair beneath the hind feet is very characteristic, also the broad whitish band across the upper part of the thigh and the rufescent fawn thigh-patch. These ditlcrcnces being constant, the Persian form must be considered distinct ; and Mr. Blanford having first character- ized the species, I have much pleasure in associating his name with it. I must, however, add the following particulars to the de- scription given by Mr. Blanford : — Mammae 8 — one pair under the throat in front of the fore legs, one pair behind Vertebrate Zoology of Persia. 90 the fore legs, and two inguinal pairs. In adults the tail is unicolorous up to the pencil of hairs and of a pale isabelline colour, while the black portion of the pencil in adolescents is a dark brown, tinged slightly witli rufescent ; the back is much darker in colour in adults, owing to the bases of the hairs, which are of a dark ash-colour, showing through ,• claws horny. Molars of upper jaw all biplicate, inside and out : of lower jaw, 1st biplicate on both sides, 2nd triplicate without and biplicate within, 3rd hij)h'cate outside and rounded witliin. The following are dimensions of an adult pregnant female and an adolescent male: — Adoles- Adiilt 2 . cent J • in. in. Length from tip of nose to root of tail 5-25 3-75 Ditto of tail 7-2 G-0 Ditto of pencil at end of tail 0*75 0'56 Total length 13-2 10-31 Height of ear from upper margin of outer conch 075 0*o Ditto from base of skull 0*82 IJreadth of ear laid flat 0-02 0-44 Length of tarsus, foot, and claws 2'62 2-12 Longest whisker (reaching to beyond the axil of the thigh) 3-25 3-0 Skull, from upper edge oi foramen mar/nuni to end of nasal bones 1-25 Skull, from lower edge of ditto to front of upper incisors 1'12 Breadth across hinder part of zygomatic arches 0-87 Ditto, between orbits 0'5 Length of lower jaw from condyle to inner base of hicisors 0*72 Teeth-line, upper jaw 0-18 Teeth-line, loAver jaw 0-18 Space between inner edge of 1st molars 0-18 Across tympanic bones U"'J3 5. Dipus Loftusi. Dipus Loftusif Blf. East Pers. p. 76, 1 (J: Bushire, 6. 12. 83. 1 ? juv. : Tanjistan, Nov. 1876. 1 ? and four foetal young : Nov. 1876. The foetal young of this species have the whiskers fairly well developed, and the tail is less than half the length of the head and body. Length of largest specimen (spirit) : — in. Head and body 5-37 Tail G-0 Tarsus, foot, and claws 1-95 100 Mr. J. A. Murray on the 6. Lagomys rufescens, Gray. Lagomys rufescens, Blf. East Pera. p. 83. 1 S' Bushire, 4. 1. 84. This agrees well with the description of it by Mr. Blan- ford, except that the chin, throat, and underparts are a silky yellowish white, as are also the fore and hind feet and the soles of the feet. The longest whisker is white at its extre- mity, and the lower series of 4-5 white throughout. Tliis is recorded from Afghanistan, Northern Persia, and Mesopotamia, but not from S.E. Persia. Length 7'5 inches. Among Birds I have to add : — 1. Faico peregrinator, Sund. Falco peregrinator, Sund., Blf. East Pers, p. 103. Four live birds were netted at Bushire and sent to me for the Kev. Mr. Watson, who trained them for the quarry. All belonged to the atriceps type. 2. Circus mao'uruSf Gmel. 2 J: Bushire. 1 J : Fao, Sliat-el-Arab. 3. Hypocolius ampelinus, Bp. 1 J , 1 $ : Bushire, 13. 9. 83. According to Mr. Gumming this species passes through Bushire in November. Its range extends N.E. to Fao, on the !Shat-el-Avab, as far as at present known, southward to Sind. It breeds in the country. Mr. Gumming and Mr. Betts have taken the eggs at Fao. This is not recorded from E. Persia by Mr. Blanford. 4. Ardeola leucqptera. Several specimens from Bushire. Breeds in June and July. 5. Sterna fuliginosa, Gmel. 2 ? , Bushire, in breeding-plumage, received with ten eggs ; and 1 ? taken on the Astola Island, S. Persia. 6. Pelecanus onocrotalus and P. crispus. Both common in the Persian Gulf. The first has been found breeding at Fao, or rather 50 miles west, on a mud island surrounded by a large marsh, whence Mr. Gumming obtained five eggs. Vertebrate Zoology of Persia. 101 7. Phcemcopterus minor, Geoffr. A male sent to me from Lingah by a Mr. Belclier, a pas- senger to Bussorali, with a note to the effect that it was shot out of a flock of larger ones, evidently P. antiqiiorum. Among Reptihs the collection contains several species not recorded from Persia. 1. Stellio nuptus, \?ix. fiiscus. Stellio nuptus, var. fuscus, Blf , East Pers. p. •'>20. 2 J, 1 ? : Bnshire, July 1876. Mr. Blanford records this from Jalk and Kalagan, in Beloochistan, and not from Persia. 2. Centrotrachelus Asmussi *. Centrotrachelus As^nussi, Strauch, Blf. East Pers. p. 337. 3 ^ : Bushire, July 188,^. Largest specimeii 23'5 inches in length. I have had a live specimen since August 1882, still in excellent condition. The animal is extremely lively during the hot hours of the day up to 4 o'clock ; after this hour it sleeps soundly, curling itself in a corner of the box in which it is kept. Its means of defence is the spiny tail it possesses, which it lashes like a whip when disturbed. It is extremely fond of having cold water thrown on its body j it then appears much pleased, standing high on its fore legs, with head erect, turning it round, upwards, and looking with each eye, and extending the loose skin of its body to double the- usual size. Whether the animal burrows for itself or occupies the burrows of field- rats &c. is a question, as, although it has a foot of soft sand in the box (without a bottom), on the bare ground, it has never yet attempted to burrow. 3. Hemidactylus Cocteaui, D. et B, Three : Charbar, Beloochistan, June 1880. Two : Charbagh, near Bushire, August 1876. One : Bushire, August 1883. Four: Tanjistan, July 1877. The non-entry of this species by Mr. Blanford is evidently an omission. * ? C loricatus, Blf. * Eastern Persia/ ii. p. 340, described from the neighbourhood of Bushire. 102 Mr. J. A. Murray on the 4. Gymnodactylus hrevipes. Gymnodactylus brevipes, Blf. East Pers. p. 344, Three : Bush ire, Oct. 1883. One : Tanjistan, Aug. 1876. Tliis has been recorded from Aptar, near Bampur, In Be- loochistan. Mr. Curaming's collection from Bushire contains well-marked specimens of this species. 5. Gymnodactylus scaber, Riipp. Thirteen : Bushire, June, July, August, 1883. Seven: Tanjistan, September 1876. In form the counterpart of Gymnodactylus petrensis^ Murray (Vert. Zool. Sind, p. 362). Kostral broader than high and cleft above. Upper labials 10-12 ; lower labials 8-10. Pupil vertical ; first labial and three small shields behind rather smaller than those covering the muzzle ; interorbital space and occiput with large conical tubercles interspersed, a few also on the muzzle and a line of 3-4 in front of each eye. Two pairs of chin- shields, the first largest and in contact. Back covered with granular scales and sharply-keeled trihedral tubercles, the latter as large as or slightly larger than the vertical ear-o])ening, and arranged across the middle of the trunk in 14 longitudinal rows ; between the hind limbs the number is six. The tubercles on the sides of the body are rather smaller and subcarinate. Scales across the middle of the abdomen in 18-20 rows. Preanal pores 5-6. Outer surface of limbs wnth large trihedral tubercles. A pair of tubercles on each side of the sacral region. The fore limb laid forward reaches the end of the snout ; laid back it reaches the axil of the hind limb. The hind limb laid forward extends beyond the axil of the fore limb. Tail verticillate, with three rows of sharply- keeled trihedral tubercles on each side to within an inch of the tip, beyond which it is covered with irregularly-arranged imbricate scales. Subcaudals distinct, single, about 44-54 ; a few of the anterior ones bifid. Length 4"5 to 5 inches, of which the tail is 2'5 to 2*75. Colour greyish brown, with three longitudinal rows of dusky subquadrate spots on the back ; in some specimens one more row of rather indistinct spots on each side. Tail with 10-12 dark bands above. Ilab. Bushire and Tanjistan, in Persia ; Fao, in Southern Mesopotamia, at the head of the Persian Gulf, on the banks of the Shat-el-Arab, and Charbar, in Beloochistan. Vertebrate Zoology of Persia. 103 Collected by Mr. W. D. Gumming, to whom I am in- debted for a large collection of reptiles, iisli, &c. from Bushire and Fao, in Southern Mesopotamia. The synoptical table below will show the differences between this species and the other allied forms of Gymnodactylus. I Dorsal tubercles, Gyninodactylus caspieus, Eich- wald brcvipes, W. Blf. lietcrocercus, W. Blf. pctrensis, Murray scaber, Eiipp kachensis, Siol. frajnatus, Gilnther Oldhami, Thcob 18-20 10 12 12 14 12-14 6-8 30 Abdo- minal scutes. ? 22 25-30 34-35 18-20 28-30 34 ? Pores. Femoral on each thigh. Pre- anal. *32-34 ... I 4 none 4 5-6 *48 .... I 4 •40 Labials. Upper. Lower 11 9 8-10 10-12 10-12 11-12 11 11 8-9 7 7-8 8-10 8-10 8-9 9 10 6. Pristurus rupestris. Pristurus rupestris, Blf. East Pers. p. .')50. Seventeen specimens from Bushire and Tanjistan, 1876-77. 7. Ceramodactylus Dor ice. Ce7-amodadyhis Dorice, Blf. East Pers. p. 353, I have seven specimens of this lizard from Tanjistan, nearly 300 miles further north-east of Bunder Abbas, where Marquis Doria's single specimen was obtained. 8. Ceramodactylus affinis. General form of Ceramodactylus Dorice, but of a more robust habit. The nostril is placed rather behind the outer hind angle of the rostral, instead of immediately above and between the suture of the rostral and first labial. The three shields behind the nostril are flat, and not distinctly sivollen, as in C. Doria. The mental is rather of a different shape, having slightly concave instead of straight sides, and a very * In a coutinuous line on hoth thii^hs- 104 Mr. J. A. Murray on the convex hind margin. Upper labials 11, with a number of smaller ones behind, scarcely larger than the granular scales of the tympanic region. Tail not attenuate, as in C. Dorice, but ends ratiier abruptly ; it is three fifths the length of the head and body, while that of C. Dorice is nearly the length of the head and body. The head is short and thick, and presents a very different aspect from that of C. Dorice) its height is about a fifth less than its greatest breadth, or a fifth less than that of C. Dorice, both in height and breadth. The fore and hind limbs are very much stouter and shorter than those of C. Dorice. The fore limb laid forward, the tips of the fingers reach only to the end of the snout, while in C. Dorice the tips of the fingers extend beyond the snout by the length of the foot. The hind limb laid forward does not extend as far as the axil of the fore limb, while that of C. Doric reaches nearly to the ear-opening. Fingers and toes much shorter and more robust. Length, head and body 2'62 inches, tail 1'37. Colour : instead of the spotted character of the markings of C. Dorice, this species is banded, the bands being rather curved, with the concavity in front. There are four bands, one on the occiput, one behind the shoulder, another on mid- trunk, and the fourth on the loins. The tail has 7-8 dark bands. The ground-colour is apparently ochraceous. There are 2-^3 dark oblique streaks below the eye along the labials. Sides of the body slightly darker than the back. Hah. Tanjistan, in Persia. Two specimens among seven of C. Dorice, collected in 1877. 9. Ehagerrhis productus, Ptrs. Head moderate, distinct from neck. Snout rather de- pressed and produced beyond the mental. Tail short ; anal bifid. Eyes rather large; pupil circular; rostral produced, obtuse in front, reverted on to the upper surface of the head, Vertebrate Zoology of Persia, 105 and forming a triangular suture with the prefrontals ; below, the rostral is deeply concave. Nasals 2. Nostrils situated in the anterior portion of the postnasal, which is received in the hind, nearly quadrangular, cavity of the prenasal. Loreal 1, square. Preocular 1, reverted on to the upper surface of the head, but not touching the vertical. Postoculars 2. Upper labials 8 ; lower labials 11, 6 in increasing series from the mental, and 4-5 other smaller ones about the size of the adjoining scales. Sixth upper labial the largest ; the fifth under the eye. Mental narrow, angular and pointed behind, slightly convex in front. Mental groove distinct. Two pairs of subequal chin-shields. Two pairs of frontals ; ver- tical elongate, obtusely triangular in front and triangular behind, where it is received in the triangular concavity of the occipitals. Occipitals rounded behind, convex laterally, and longer than their greatest width ; temporals variable, gene- rally 2-3 alongside the occipitals, and in some split into small plates. Scales in 17 rows, all, even the head-shields, minutely puuctulated with brown (only seen through a Coddington lens). Ventrals slightly angulated. Colour (spirit- specimens) ochraceous, with equally distri- buted dusky spots in five rather oblique series. A small dark brown spot under the eye ; another less distinct above it (not present in all specimens), a third obliquely behind the eye, followed by a large temporal spot, also oblique, and extending beyond the last labial. Lower parts unspotted, pale yellow. Ilah. Bushire, I c? , 2 ? ; Tanjistan, 3 c? , 2 ? . Length of largest specimen 46 inches, of which the tail is 7 inches ; length of smallest specimen 15 inches, of which the tail is 2'5 inches. 10. Bufo viridiSj Laur. The following is a description of specimens from Bushire: — Crown of the head flat, smooth, and devoid of osseous ridges ; orbitals elevated and covered with horny- tipped tubercles ; interorbital space equal to the width of the eyelid. Hind crown, back, and the upper surface of fore and hind limbs covered with rather close-set horny-tipped tubercles. Parotoids oblong, flattened above about their middle, and separated from each eye by a deep groove. Length of each parotoid equal to the distance between its front edge and the nostril. Tympanum distinct, about one half the size of the eye. Under surface of the body smooth, without any trace of tubercles. Fore feet with a large mesial palmar pad and a smaller one at base of first finger. Laid side by side the 106 Mr. J. A. Murray o?i the first finger is very slightly longer than the second and as long as the fourth ; third finger longest. Tarsus witli a mesial and lateral longitudinal row of distant tubercles, and a cuta- neous fold on the inner side extending to the inner metatarsal tubercle. The tubercle on the outer edge scarcely so promi- nent as the inner one, which is elongate. Soles of the hind feet faintly tuberculate. Toes half-webbcd ; the tip of the first reaches tlie second joint of the second toe. Hind limb long ; laid forward alongside the body the metatarsal tubercle reaches the eye, and one half of the foot extends beyond the snout. Length 3 inches ; hind limb from anus to tip of second finger 4*2 inches. Colour yellowish, a dark spot ou each eyelid ; anotlier oblique one from the liind edge of the eye to the tympanum, and a third very small one on eacli nostril ; fore and hind legs with 2-4 transverse blotches on their upper surface. Sides of the first and upper surface of first and second fingers black. Under surface pale yellowish. Tips of toes slightly swollen and of a brownish colour. These specimens come near to Bofo olivaceus^ Blanf. (East Pers. p. 434, pi. xxviii. fig. 3) , but differ from it by having the dorsal surface closely set with horny-tipped tubercles, a flat instead of a concave crown, by its under surface being smooth and not tuberculate, and by its shorter hind limbs. From B. vulgaris it is distinguished by its longer hind limbs, having a cutaneous tarsal fold, a distinct tympanum, and no dark band below the parotoid. XIV. — Additions to the Reptilian Fauna of Sind. By James A. Murray. Since the publication of my work on the ' Vertebrate Zoology of Sind,' a collated descriptive account of all the species of mammals, birds, and reptiles (including several new species) known to inhabit the province, some little interest appears to have been aroused in zoological inquiries, which has resulted in the Kurrachee Museum acquiring several collec- tions of reptiles from hitherto unknown localities in Upper Sind. Among these are four species from the barren sandy wastes of the frontier districts, collected by my indefatigable corre- Reptilian Fauna of Sind. 107 spondcnt Mr. F. Gleadow, of the Forest Department, three of which I believe are undescribed forms. These are :— 1. Melanochelys pictus . Head two thirds as broad as long, its greatest length 3 inches. It is covered with skin, divided into plates ; a long central one above a single broad frontal ; a superciliarj on each side, and a small subtriangular plate behind in suture with the sides of the central plate. Temples covered witli numerous irregular-sliaped plates. Upper jaw witli a small festoon on each side^ the groove in the middle of the jaw ratlier deep. A plate in front of the eje in suture with the sides of the frontal ; another nearly as large under the orbit, and a third about twice the size of the latter behind the eje. Anterior half of neck covered with small subimbricate plates in transverse series. Shell oblong-ovate, elevated, mucli arched, nearly half as high as long, nodosely tricarinate, the costal carina being much nearer tlie vertebral carina than the marginal plates. Length of shell over curves 14 inches; breadth over vertebrals 11*75 inches. The sternum is bent uj)wards from the suture of the pectorals with the postgulars ; greatest length of sternum to point of furcate projection of anal plates 12 inches. Anals deeply notched posteriorly, the distance between the projecting ends being 1'5 inch and the depth of notch 1"37 inch. Width of sternum at axillary incision 6 inches, at inguinal 4' 75. Gulars together broader than long, their hind margin received into the subtriangular concavity in front of the postgulars, which are as broad as long. Pectorals very narrow, each 3 x 1'5 inches, the suture between them about equal in length to that of the postgulars and slightly more than half of that of the abdominals. Abdo- minals nearly rectangular, winged beyond the inguinal incisions,, and forming a suture on each side with the inguinal plate and sixth marginal. Postabdominals longer than broad, the length of their suture together slightly less than that of the abdominals ; transverse sutures of postgulars with abdo- minals and abdominals with postabdominals straight ; suture of postabdominals with anals concave ; the suture together of the anal plates is shorter than the suture of a single one with the postabdominal. Nuchal plate oblong (0*75 x 0*5 inch). Vertebrals hexagonal j first somewhat bell-shaped, convex in front, straight behind (except a concavity raesially to receive an apophysis of the second vertebral), sinuately concave on each side in its anterior half and convex lower down. Second and third vertebrals hexagonal, as broad as long ; fourth 108 Mr. J. A. Murray on the similar, but concave beliind in suture with the convex front of the fifth, which is about twice the size of the other verte- brals, broader than long (2*37x3"37 inches), with the sides sloping outward to the last marginal ; its suture with the caudals is straight. Caudal notched, the suture of both plates a little more than half the length of the fifth vertebral. Tail short, 1*5 inch in length. Costals large, four in number, nodosely carinate on their upper margin, but not extending, or scarcely seen on the fifth vertebral. The first costal is subtriangular and largest, convex in front, in suture with one half of the first and the whole of the upper margin of the second, third, and fourth marginals ; second costal in suture with the fifth and sixth marginals, its greatest breadth about two thirds its greatest length (4 x 2*75 inches); third costal in suture with the seventh, eighth, and anterior thii-d of the ninth ; it is nearly the size of the second in length and breadth ; fourth costal smallest and forming sutures with the posterior two thirds of the ninth marginal, entire tenth, mesially the apex of the eleventh, and the sides of the fifth vertebral. Marginals variable in size, not serrated poste- riorly, but slightly dilated at the eighth, ninth, and tenth. Feet anteriorly covered with imbricate scutes, posteriorly, or higher up, subimbricate, the scutes much smaller. Toes short, strongs and webbed to the claws ; they are covered with annular scute-like plates. Claws strong and hooked. Sides of the legs fringed with large scales. Colours : all the scales on the tarsi and feet with a yellow spot. Head black, with large spots, blotches, and streaks of yellow ; a patch on each side of the snout, also on each side of the nostrils ; one under each eye, another at the maxilla on the labial margin, and two behind each eye. There are also some .large blotches on the tympanic and temporal regions, and three on each side of the lower jaw. Shell olive or greenish brown, the marginals, lower part of costals, and vertebrals with pale yellowish blotches and streaks of irregular shape. Sternum pale yellow, wnth linear transverse lines, very close together on the abdominal plates, and forming a large patch. The gulars and postgulars are not marked. Mr. Gleadow, obtained this species in the Sind " Doro,^^ in the Kushmore Talooka, Upper Sind. It difiers from all the described forms of Melanochelys, first by its greater size, next by the size and shape of the vertebrals and costals, and lastly by the markings of the shell and the spotted character of the head and feet. Reptilian Fauna of Sind. 109 2. Hemidactylus kushmorensis. Head rather depressed. Rostral grooved above, slightly wider than high. Upper labials 10; lower labials 8. Two pairs of chin-shields, the first only in contact. Muzzle covered with granular scales. Nostrils between the rostral, first labial and three small shields behind about equal in size to those covering the muzzle. Crown of the head interspersed with numerous rounded tubercles. Back with rounded tubercles arranged in twenty-two longitudinal series across the middle of the body; a few tubercles between the hind limbs are subtrihedral. Tail verticillate, each verticil armed laterally with three rows of rather elongate subtrihedral tubercles, except on its posterior third, where they are re- placed by imbricate scales. Fore and hind limbs on their upper surftice studded with round tubercles. Toes covered with imbricate scales. Claw on thumb well developed. Scales on the throat about one third the size of those on the abdomen, across the middle of which they are arranged in 32-3G longitudinal series ; the anterior half irregularly and minutely 1-3 crenulate, less conspicuous on the posterior half. Femoral pores 10-12 on each thigh. Under surface of tarsi covered with large imbricate scales. Subcaudals single, 44-46. Middle toe with six pairs of plates and an odd one at each end. Colours neutral grey or brown, with three rows of squarish dark blotches, forming either longitudinal or obliquely trans- verse interrupted bands ; a few smaller spots on the sides. A dark streak through the eye with a pale line above it. Scales on the under surface of the body freckled with 1-3 dark spots ; many, especially on forward part of body, without them. Tail with 14-15 dark bands. Pupil vertical. Ilab. Upper Sind, Kushmore and Thool Talookas. Two only of six specimens with unreproduced tails. Length 4 to 4"25 inches. Type from Bhaner, Upper Sind frontier. Difiers from all the other species of the genus in having a greater number of dorsal tubercles, also femoral pores, except H. Gleadowi, and fewer abdominal, scutes, except H. triedrusj and in having rounded tubercles. The following Table will sufficiently show the differences between it and the other allied species of the genus : — Ann. i£' Ma(j. N. Ilist. Ser. 5. Vol. xiv. 110 On the Reptilian Fauna of Sind. Species. Dorsal tubercles. Hemidactvlus Gleaclowi, Murray (V. Z. sind) H. karacbiensis, Murraii (V. Z. Sind) ::... H., sp., Blanford (E. Persia) ... H . persicus, Blf. H. kusbmorensis, sp. uov H. subtriedus, Jerd.^ H. triedrus, Baud. * H. maciilatus, B.ct B.* H. Pieresi, i^eZaari; * H. gracilis, Blanford * 15-16 16 14 16 20-22 18-20 Nuine- roiis. do. do. do. Abdo- minal scutes. 38-39 .38-40 40 42-44 32-36 ? 30 37-4] 40-42 ? Pores. Femoral on each thigh. 13 10-12 8 7-8 10-14 32- Preanal. 36+ Labials, Upper. 10-12 9-10 10 11-12 9-10 7 9-10 11 11-12 ? Lower. 8-10 8-10 8-9 9 8 10 ? 4-25 4-25 3. Gymnodactylus scaher^ Riipp. Two specimens collected at Sukkur and one at Laid, be- tween Shikarpoor and Sukkur, by Mr, F. Gleadow (see my preceding paper, " On Additions to the Fauna of Persia," for description of this species). It replaces Gymnodactylus petrensis in Upper Sind. 4. Acontiophis paradoxa. Acontiojjhis paradoxa, Giiiith. Proc. Zool, See. 1875, p. 232. This snake was known from a single specimen only, the locality of which has only now been ascertained. I count ventrals 180, and subcaudals 52, My specimens are greyish brown ; a dark line from behind the eye to the nape, a subovate dark patch on the occiput and a border along the margins of the occipitals from the anterior half of the super- ciliaries. A dorsal series of quadrangular dark spots with white interspaces, nearly of the same width to within an inch of the end of the tail, where they become smaller and more faint, and gradually disappear. Total length of larger of two specimens 14*25 inches, of which the tail is 2*10 inches. * From Giintlier and Theobald's works on Reptiles of Br. Ind. t In a nearly continuous line. On a new Species o/'Ljcopodites, Goldenherg. Hi Hob. Upper Sind, Thool Talooka, at Zungipoor, frontier districts. Mr. F. Gleadow states that the two specimens of this snake were dug up from depressions a quarter-mile apart in a con- siderable area of blown sand forming hillocks 20 to 30 feet hiffh. XV. — On a new Species of Lycopodites, Goldenherg {L. Stockii) , from the Galciferous Sandstone Series of Scotland. By R. KiDSTON, F.G.S. [Plate v.] The genus Lycopodites, as originally employed by Brong- niart* and most of the older writers, did not contain any plant which was really entitled to the name, in so far as it was used to infer their closer affinity to the recent Lycopodium than that held by the genus Lepidodendron j and Brongniart, in his later writings, discarded his genus Lycopodites^ as sub- sequent investigations had shown him that his original view of the plants he included in it was founded on an erroneous notion of their true nature f. Hence, when Goldenberg resuscitated the genus Lycopo- dites^ it was used by him in an entirely different sense from that to which it had been applied by previous writers, and in fact was a new genus though under an old name. To enable us more clearly to appreciate the light in which Goldenberg regarded his genus Lycopodites^ I quote his intro- ductory remarks regarding it. Lycopodites, Goldenberg. 1855 f. " Branches with leaves placed spirally or in verticils. Sporangia placed in the axils of the leaves or forming ter- minal cones." " In the genus Lycopodites we place the true herbaceous Lycopods of former ages, which agree in all essential points so exactly with recent Lycopods that, at the most, they can only be regarded as a subdivision of the genus Lycopodium. " The fossil plants included by Brongniart and others under this name are probably only young twigs of Lepidodendron * 'Prodrome,' p. 83 (1828), and ' Classification desvegetaux fossiles,' p. 4G (1822). t Tableau d. genres de Veget. foss. p. 40 (1849). X * Flora Saritpontaua Fotsilis,' Heft i. pp. 0, 10 (ISSS). 9* 112 Mr. R. Kidston on a new Species of or ConifeiEe. This view, which we here accept, has been pointed out by Brongniart in his last work (' Tableau des genres de vegeteaux Ibssiles, consideres sous le point de vue de leur classiiication botauique et de leur distribution geolo- gique '), and is thus stated by him : — " ' The plants really analogous to the recent Lycopods are very few in number in the fossil state. " ' I do not know even one which, by its dhnensions and the disposition of its leaves, may be compared with certainty to the species of the genus Lycopodium properly so called ; the greater part of the plants which I have designated or which have been indicated as Lycopodites, are probably either the upper portions of young branches of Lepidodendron or the branches of Conifers. " ' Thus the greater part of the Lycopodites with dichotomous branches from the Carboniferous formation appear to belong- to the first class ; those species with distichous pinnate branches evidently belong to Conifers of the genus Walchia. The greater part of the species from more recent forma- tions, as the Lias and Oolite, belong to this latter group ; such are in particular Lycopodites Williamsonis and patens. " ' Among these there is, however, one species, which has all the characters of a Lycopod, or perhaps more the character of the genus Selaginella. This, the Lycopodites J'alcatuSy L. & H.*, has lately been rightly separated, and from its delicate and dichotomous branches, apparently distichous leaves (but which are probably opposite and unequal), has all the appear- ance and essential characters of the numerous species of the genus Selaginella. " ' 1 know no species which resembles the true Lycopods, as at present defined, nor the genus Tmesipteris.'' " We have therefore in Lycopodites the addition of a new genus of fossil plants to the Carboniferous flora, which, according to Brongniart, is at present only known by one species from the Oolitic formation of Englandf. For the plant-remains which hitherto have been described and figured from the Carboniferous formation under the name of Lycopodites Bronniiy longifolius^ &c., belong, as we have seen, to quite other genera of plants, as they do not exhibit any of those points which form the principal characteristics of club-mosses. " Many years ago 1 found plant-impressions in the Carbo- niferous rocks of this neighbourhood (Saarbriick) which, in the character of their growth, showed a great similarity to our herbaceous Lycopods. At the Meeting of the Natur- * ' Fossil Flora,' pi. Ixi. t Lycopodites falcatus, L. & H. Ljcopodites, Ooldenberg (L. Stockii). 113 historisclien Vereins der preussischen Rlieinlande at Kreuz- nach I exliibited a pretty complete example, and also later, in the Transactions of this Society, made some preliminary remarks on the occurrence of such plants in the Carboniferous formation. 1 succeeded later in discovering several other species, of which some even bore distinctly their organs of fructification. It then appeared that, as regards the position and form of the fruit, these fossil remains agree completely in all essential points with our living Lycopods. " The Lycopodites which we are about to describe may, like our recent club-mosses, be placed in two subdivisions, accord- ing to whether the sporangia are seated in tlie axils of the leaves or form terminal cones." The six species described by Goldenberg are classed under these two heads : — A. Sporangia placed in the leaf-axils. 1. Lycopodites denticulatus^ Goldenberg. 2. elongatuSj Gold. B. Sporangia forming terminal cones. '6. Lycopodites primcevus, Gold. 4. Icptostachyus^ Gold. 5. ■ mac tH)phy tilts, Gold. Q, taxinuSj L. & H., sp. In regard to his last-mentioned species^ which he identities as Knorria taxina, L. & H., its claim to belong to the genus Lycopodites rests on very slender grounds. His figure only shows a small portion of a stem 1^ inch long and about \ inch wide, with spirally arranged leaf- scars, in general form very like those of L. Stoclcii (PL V. fig. 1) ; but this single character, as shown in Goldenberg's figure, appears of too little importance to be of generic value. But apart from the question as to the systematic position of Goldenberg's L. taxinus, it is clearly not the same fossil as that named Knorria taxina by Lindley and Hutton. The specimen, from which the last-mentioned author's plate is taken, is preserved in the " Hutton collection," New- castle-on-Tyne. Tiiis I have compared with their figure, which, I am sorry to say, is not a very correct representation of the fossil. I believe Lindley and Hutton's plant is merely a small stem of Cordaites, certainly a quite different plant from Goldenberg's Ljycopodites taxinus. In the same year (1855) in which Goldenberg described 114 Mr. E. Kidston on a neio Sjiecies of his specimens, Geinitz figured and described another species, Lycojjodites Guthieri^ . All these plants Schimper has placed in the recent genus Lycopodhim^. Both Goldenberg and Geinitz considered them, if not identical with Lycopodium^ extremely closely related to that genus. Schimper's opinion of the systematic position of these plants has been adopted by Renault:}:, who has described a small Lycopodiaceous stem under tlie name of Lycopodlum puncta- tum^ ; another, in the same communication, is described by Ad. Brongniart as Lycopodium Renaultn. Both these small stems, though perhaps closely related to Lycopodmm ^ possess some structural diii^'erences which we are inclined to regard as of sufficient importance to exclude them from the recent genus in which they are placed. Renault, in his very interesting paper, has pointed out the presence of areolated vessels in his fossils, which he admits do not occur in recent Lycopodium. It is quite possible that the specimens described by Renault and Brongniart are stems of plants similar to Goldenberg' s Lycojiodites ; but as they are not identical in structure with any known Lycopodium^ I think it better to place them in Lycopodites than in Lyco- podium. The same train of reasoning inclines me to reject the genus Lycopodium for Goldenberg's species, notwithstanding their undoubted close relationship to it. If we place them in Lycopodium, we commit ourselves to the opinion that the fossils are identical in all essential structural peculiarities with their recent representatives — an opinion not at present satis- factorily substantiated by proof. I therefore propose to rein- state the genus Lycopodites^ Goldenberg, for the following species : — A. Sporangia placed in the axils of the leaves. 1. Lycopodites denticuIatuSj Gold. 2. elonyatuSj Gold. B. Sporangia forming terminal cones. 3. Lycojjodites primcevus, Gold. 4. leptostachyus, Gold. * ' Die Versteinenmgen der Steinkohleuformation iu Sacbsen,' p 32 pi. i. fig. 1 (1850). "^ ' t Schimper, Tiaite d. paleont. veget. vo]. ii. p. 8, pi. Ivii. (1870). X Cours d. botan. foss. p. 74, pi. xii. figs. 9, 10 (1882). § Ann. des Sciences Nalurelles, 5<~ ser. Bot., vol. xii. pp. 178-182 pip. xii.-xiv. ri869). Lycopodites, GolcUnberg (L. Stockii). 115 .5. Lycopodites macropliyllus^ Gold. 6. Guthieri^ Gopp. (=Z/. stadtygynandroides, Gutbier). 7. Stockii, Kidston, u. s. C. Sporangia unknown. 8. Lycopodites punctum, Renault, sp. 9. Renaultiij Ad. Brong. sp. 10. (?) taxinus, Gold, (not L. & II.). The above list contains all the plants, as far as I am aware, which are entitled to be placed in Lycopodites, Goldenberg; but since Brongniart and Goldenberg wrote on tliis subject, notwithstanding the warnings given by these two authors, many Lepidodendroid twigs have been figured and described as Lycopodites. Most of these belong, I believe, to Lepido- dendron Sterithei'gii and Lep. rimosum, and probably to other species of the same genus. I have seen specimens of these two L^epidodendra with small delicate branches springing from stems of considerable size, which could be specifically identified with described species of Lycopodites. These small lateral branchlets are merely the result of very unequal dichotomy*. Whatever view may be taken of these Lepidodendroid twigs, or so-called Lycopodites, it will, I think, be admitted on all hands that they do not find a suitable place in Golden- berg's genus Lycopiodites. Lycopodites Stockii, Kidston, n. s. Description. Cone terminal, composed of a number of oval sporangia ; leaves arranged in whorls, dimorphic (?), the larger leaves ovate-cordate, acuminate with a strong central niith'ib, the (?)smaller leaves transversely oval. L'enmrks. The specimen from which the above description is taken, is about 4 inches long; of this, the cone, which is imperfect at its apex, occupies I5 inch. The leaves are mostly dis])laced, but the form of many of them is well shown. Unfortunately the state of preservation of the fossil is not all that could be desired ; but it is sufficiently distinct to enable one to give a description by which it can easily be recognized. In PI. V. fig. 1 is given a careful sketch of the plant. The cone consists of a number of oval bracts : some of them appear reniform in shape, but the fossil is so much compressed that the individual contour of the sporangia * As au example, I give on PI. V. fig. 5, some small branchlets o^Lepid. rimosvm from the l^pper ( ■oal-measmes, Tim.^hury, Somorset. 116 On a neio Species o/Lycopodites, Goldenherg. cannot be well made out. The leaves shown on the sides of the stem are ovate acuminate, with a very distinct middle nerve (fig. 2). An interesting point in the fossil is the occurrence of a ver- tical row of curiously formed leaves (?), entirely different from those just described. One of these (a, fig. 1) is shown en- larged at fig. 3. This curious structure has very much the appearance of a sporangium ; but the occurrence of a terminal cone and of sporangia situated in the axils of the leaves of the same species, is altogether vmknown in any Lycopod, either fossil or recent. The most perfect of these curious structures (whether leaves or sporangia) appears to have three inflations (fig. 3) ; but I am rather inclined to think that this appearance has been caused by its being pressed against the stem, and that we have imder consideration a leaf and not a sporangium. In this case we have merely a dimorphic condition of leaves, such as occurs in those fossils already described by Goldenberg and Geinitz, and is common in the recent genus SeJaginella. In this example, in no case are they exhibited so clearly that one can positively afiirm they are leaves ; but I believe this to be their true nature notwithstanding their sporangium-like form. In the enlarged sketcli (fig. 3) the dark dentate margin has no connexion with the supposed leaf, but only a small broken piece of carbonaceous matter, which probably represents the cortex. The leaves appear to have been arranged in whorls of 6 or 8, as shown by the cicatrices on the enlarged portion of the stem (fig. 4). My thanks are due to Mr. T. Stock, Edinburgh, who has submitted this fossil to me for examination and description, and after whom I have pleasure in naming it. I believe the opinion generally current regards the genus Lepidodendron as the ancestor of our herbaceous Lycopods ; but I am rather inclined to believe that the genus Lepido- dendron has entirely disappeared, and that our recent Lyco- pods are the descendants of Goldenberg's Lycopodites. Horizon. Calcil'erous Sandstone series (Culm of Stur). All the species of Lycopodites^ Goldenberg, previously de- scribed have been derived from the Coal-measures; hence the discovery of the genus so far down in the Carboniferous formation is of considerable interest. Locality. Glencartholm, Eskdale, Dumfries. On the Families of existing Lacertilia. 117 EXPLANATION OF PLATE V. Fiy. \. Lycopodifes Stockii, Kidston, n. sp., nat. size-, a, h, c, d,f. Sporan- gium-like leaves (?) ; e. Reniform sporangia of terminal cone. Fig. 2. Leaf, enlarged, seen on fig. 1, g. Fig. 3. Sporangium-like leaf (?), enlarged, seen on fig. 1, a. Fig. 4. Smnll portion of stem, enlarged, showing leaf-cicatrices. Fig. 5. Li'pidvdendron rimosum, Sternberg. XVI. — Sr/nojjsis of the Families of existing Lacertilia. By G. A. BOULENGEE. Whilst engaged in a revision of the Lizard-collection in the British Museum, I have felt the necessity of a thorough syste- matic rearrangement of the order Lacertilia. The classitica- tions proposed by Dumeril and Bibron and Gray, and now still generally in use, with slight modifications, are, on the whole, as unnatural as can be, and founded to a great extent on characters of pholidosis and physiognomy. Physiognomy is worth nothing as a guide in the formation of higher groups ; as to the characters afforded by the scales I have convinced myself that they are very deceptive, and ought to be taken into consideration in the definition of families only when accompanied by other characters. Like Cope, whose lizard- families* I regard as the most natural hitherto proposed, I shall lay greater stress on osteological characters and on the structure of the tongue. Special importance must also be attached to the presence or absence, and the structure, of dermal ossifications on the head and body, and these will be found to correspond with many other characters. Bocourtf, to whom is due the merit of having pointed out their syste- matic importance, did not realize the very great progress made by means of that character, the modifications of which he so ably illustrated, for he still maintains the artificial group Scincoidiens, in spite of the objections of Cope, whose views are evidently confirmed by the researches of tlie French herpetologist. The order Lacertilia, as restricted by Giintlier %, may be divided into two primary groups only, the Chameleons on the one hand, and all the other Lizards on the other. The Amphisbffinians, which by nearly all recent authors are sepa- * Proc. Acad. Pliilad. 18G4, p. 224, and Proc. Am. Assoc. Adv. Sc. xix. 1871, p. 236. t Mission Scient. Mexique, Rept. p. 470 (1881). t Phil. Trans. Roy. Soc. clvii. 18(i7, p. 625. 118 Mr. G. A. Boulenger on the rated as a suborder, or even as an order, I include among the true lizards, and regard them as a degraded type of the Teiidffi, with which they are to some extent connected by the Chalcides and their allies. The principal characters which have been put forward in favour of their separation are : — (1) absence of interorbital septum ; (2) absence of columella cranii ; (3) very short mandible, causing the quadratum to be nearly horizontal ; (4) division of the occipital condyle; (5) ab- sence of postorbital and fronto-squamosal arches; (6) absence of scales. These characters, which are mostly negative, are not all constant throughout the group, and many will be found, to a greater or less degree, to be characteristic of all strongly degraded, burrowing forms, such as Aniella near the Anguidffi, AneJytrops [Typhline) and Dibamus near the Skinks, &c. The im])ortance of these characters justifies our placing the Amphisba^nas in a separate family ; but, in my opinion, not in a higher group, for the following reasons : — 1. The absence of interorbital septum also occurs in Ophiognomon among the Teiidse, and there is every grada- tion between the skull of that genus and that of higher members of the same family : besides Aniella and Dibamus^ which belong to totally different families, also possess the same negative character. 2. The columella disappears gradually witlithe interorbital septum ; it is hardly distinguishable in Ophiognomon and totally absent in Aniella and Dibamus. 3. The aberrant lower jaw, not in itself a very important character, is not even constant, the genus Blanus differing in that resjDCct as much from the typical AmpJiisbcena as from a typical Lizard. 4. The division of the occipital condyle, also a character the importance of which ought not to be exaggerated, is not even constant, the Acrodont Amphisbanians forming ex- ceptions. 5. The absence of postorbital and fronto-squamosal arches, which occurs in the most diverse groups of Lizards, cannot be regarded as more than a family character. 6. The naked integuments (if we may apply this term to the skin of the Amphisbgenians with its soft scales) are not special to the group, but occur also in Geckos ; and they are so closely approached by those of some Cercosaurine and Chalcidine Teiidas as to render any sharp distinction im- possible. On the other hand, characters such as are afforded by the tongue, which in all Amphisbffinians is in every respect similar to that of the Cercosaurine and Chalcidine Teiidiv, Families of existing Lacertilia. 119 the preanal pores of most Ampliisbaenians, and the anterior limbs of Chirotes are indicative of affinity to tlie Teiida?. Respecting the latter, it may be remarked that in the other Lacertilia which dispense with the limbs, the fore pair dis- appear before the hind pair, and this holds true for the Ophi- dians, the less modified type still showing rudiments of])elvis, whilst not one preserves any thing of the pectoral arch. A reverse process obtains in the Teiidte and Amphisba^nidas. I have already put forward my objections to recognizing the suborder Nyctisaura*. Having separated the Charaadeons, we are in presence of the large suborder of true Lizards. This I have divided into twenty families, which I regard as perfectly natural groups. But there is great difficulty in arranging these families in a line. Two characters seem to demand special attention — those of the lingual papilla? and the clavicle, as, excepting the Geckos and Eublepharidas, they exactly correspond, i. e. the forms with smooth or villose tongue have a slender, non dilated clavicle, whereas those with scaly tongue have the clavicle strongly dilated proximally and generally enclosing a fora- men. Order LACERTILIA. Suborder I. LACERTILIA VERA. A. Tongue smooth, or iciih villose papillce ; clavicle dilated, loop-shaped proximally ; no jwstorbital or fronio-squamosal arches. Fani. 1. Geckonid^. Vertebra) ampliiccslian ; parietal bones distinct. Fani. 2. Eublepharid.15. Vertebrce proccBlian ; parietal single. B. Tongne smooth or with villose papillcB ; clavicle not dilated proximally. Fam. 3. ITroplatid.^. Vertebrae amphiccelian ; iuterclavicle minute; no postorbital or postfi'onto-squamosal arches. Fam. 4. PyGOPODJB.T':. No postorbital or postfronto-squamosal arches ; pre- and postfrontal bones in contact, separating the frontal from the orbit. Fam. 5. Agamid.^. Postorbital and postfronto-squamosal arches pre- sent ; supratemporal fossa not roofed over by bone ; tongue thick ; acrodont. Fam. 0. Iguanid.^. Postorbital and postfronto-squamosal arches pre- sent; supratemporal fossa not roofed over by bone; tongue thick; pleurodont. Fam. 7. XENOSArsiD.^. Postorbital and postfronto-squamosal arches present ; supratemporal fossa not roofed over ; anterior portion of tongue retractile. Fam. 8. Zoxukidje. Postorbital and postfronto-squamosal arches com- plete ; supratemporal fossa roofed over ; tongue simnle. Fam. 9. Anguiu,^. Postorbital and posttronto-squaniosal arches present; supratemporal fossa roofed over ; body with osteodermal plates * Ann. .t Mag. Nat. Hist. (6) xii. 1883, p. 308. 120 Mr. G. A. Boulenger on the with irregular, arborescent, or radiating channels ; anterior portion of tongue retractile. Fam. 10. Aniellid^. No interorbital septum, no columella cranii, no arches. Fam. 11. HKLODERMATiDiE. Postorbital arch present, postfronto-squa- mosal arch absent ; pre- and postfrontals in contact, separating the frontal from the orbit. Fam. 12. Varanid^^. Postorbital arch incomplete ; postfronto-squa- mosal arch present ; supratemporal fossa not roofed over ; nasal bone single ; tongue deeply bihd, sheathed posteriorly. C. Tongue covered ipifh imhricate scale-like papillcc or with oblique plica ; clavicle dilated lyroximalhj ^ frequently loop-sliapcd. Fam. 13. Xantusiid^. Parietals distinct; postorbital and postfronto- squamosal arches present ; supratemporal fossa roofed over. Fam. 14, TEiiDiE. Postorbital and postfronto-squamosal arches present; supratemporal fossa not roofed over ; no osteodermal plates. Fam, 15. Amphisb^enidje. No interorbital septum ; no columella cranii ; no arches ; premaxillary single. Fam. 16. LACERXiDiE. Arches present ; supratemporal fossa roofed over ; premaxillary single ; no osteoderuuil plates on the body. Fam. 17. GEPEHOSArRiDiE. Arches present ; supratemporal fossa roofed over; premaxillary single; body with osteodermal plates with regular channels (a transverse one anastomosing with perpendicular ones). Fam. 18. SciNCiDiE. Arches present ; premaxillary double ; body with osteodermal plates as in the preceding. Fam. 19. Anelytropid^. Premaxillary single; no arches; no osteo- dermal plates. Fam. 20. Dibamid^e. Premaxillary double ; no interorbital septuui ; no columella cranii ; no arches ; no osteodermal plates. Suborder II. RIIIPTOGLOSSA. Fam. 21. Cham.^leontid.?5. The Geckonidce and ILublepharidce^ whicli dilFer from all other families in combining a dilated clavicle with a simply papillose tongue, are well distinguished from each other by the vertebree, which are amphicoelous in the former and pro- coelous in the latter. As characters of minor importance may- be mentioned the coossification of the parietal bones in the Eublepharid^, while they remain distinct in the Geckonida?, which are also distinguished, constantly I believe, by having one bone less in the mandible, the supra-angulare having coalesced with the angulare. Next come the Uroplatidce^ which are now for the first time separated from the Geckos. Although agreeing in most respects with the latter, their sternal apparatus differentiates them widely ; the clavicle is slender, not at all dilated, and the interclavicle is reduced to a minute bone. Except the chameleons, all other lizards in which the pectoral arch is not Families of existing Lacertilia. 121 rudimentary have a large interclavicle. To this very im- portant character is added another ; the nasals are united into a single bone, a peculiarity which is found elsewhere only in the VaranidiB among recent lizards. A single genus, Uro- ^jlates^ from Madagascar, is known. After the Uroplatida? I have placed the Pygopodidce (=Pygopid8e + AprasiadtB + Lialisidge of Gray), which family is now based on new characters. They were formerly arranged with or near the " Sciucoids," a view which cannot be main- tained, since that group was an assemblage of forms having totally diiferent affinities, and " Scincoids " will now be found scattered through the following families : — Anguidee {Anguisj Dijiloglossus, &c.), Aniellidge, Teiidaj {Gymnoplithalmus^ &c.), Scincida3, Anelytropidaj, and Dibamidee. The skull of the Pygopodida3 in its simplicity of structure approaches that of the Geckos, and the parietal bones remain distinct in all the genera except Lialis ; the bones of the lower jaw are still more reduced in number, the angular, supra-angular, and arti- cular having coalesced, a character by which they approach the snakes. The affinities of this little group are very obscure, and a complete investigation of their anatomy is highly desirable. The two closely allied families Agamidce and Iguanidce remain as before. The Xenosaivridce must be regarded as intermediate between the Iguanidte, with which Peters was inclined to associate them, and the Anguidffi, near which they are placed by Cope. The Zonuridoi correspond only in name with the Zonu- rida3 of Gray and most other authors. The members of Gray's Zonuridaj will be found in the following families : — Anguidai {GerrhonotuSy FseudojJus, &c,), Lacertidai [Tachy- dromus), Gerrhosaurida3. They have, like the Anguidai, a villose tongue, though not retractile at the end, a slender chavicle, and in some the body even presents bony plates, which arc destitute of symmetrical canals. As here under- stood, the Zonurida3 comprise the genera Zonurus, Flaty- saurus, and Cliavicesaura. The Anguidce correspond to Cope's Anguidas and Gerrho- notidte, the diiferential characters of which latter group seem to me insufficient for family separation. As Cope has shown, this group is perfectly natural, though containing " Chalci- doid" and " IScincoid " forms, and an excellent illustration of how misleading it is to trust only to external characters. The " Scincoid " forms correspond to Bocourt's Diplo- glossidce. The family Aniellidce was also established by Cope. I would regard it as a degraded form of the Anguidse. 122 On the Families of existing Lacertilia. The Helodermatidce, as already shown bj Cope, have tlie greatest affinity to tlie AnguidaB, from which they are, how- ever, well distinguished by the structure of the skull. The grooved teeth might be given provisionally as another family character. It would be highly irajjortant to have some infor- mation on the osteological characters of Steindachner's Lan- thanotidte, as tliere is reason to suspect they will enter this family. The Varanidce, which come last in the series of alepidote- tongued lizards, remain characterized aa before, and form a perfectly isolated group. We have next a series of families characterized by the peculiar scale-like lingual papillas and the proximally dilated clavicle. The XantusiidcB are closely allied to the Teiid^, but distinguished by the different skull and scarcely incised tongue. The Teiida form a very natural group, comprising the Cercosauridai, Chalcididaj, Chirocolida;, Anadiadaj, and part of the Gymnophthalmida3 of Gray and the Tretioscincidas of Bocourt. It thus contains " Lacertoid," " Chalcidoid," and *' Scincoid "" forms of the Dumerilian system, all passing into one another by insensible gradations and all agreeing in the structure of the skull, tongue, and pectoral arch. All are confined to the New World, whereas the analogous family Lacertid(B is restricted to the Old World. As mentioned above, I regard the Amphisb^nidee as strongly degraded forms of the Teiidfe. I establish a family Gerrhosaurida> for GerrliosaicruSy which was formerly associated with the Zonuridaj, but which agrees closely with the Scincidas, from which it is to be dis- tinguished by the coalesced premaxillaries. Although the arrangement of the scales of the body is different from what we see in the latter family, the underlying dermal bony plates are precisely similar in their symmetrical canals. The Scincidce correspond to Cope's Scincid^, Sepida?, and Acontiida3, and to Bocourt's group Aspidoscinciens, less the Diploglossidaj. The Aneli/tropidce, a small family so named by Cope and synonymous with the Typhlinidas of other authors, are a degraded type of the Scincida?, having completely lost the cranial arches — which, in some forms of the latter group, show a tendency to disappear — and also the osteodermal plates. The Dibamidce, characterized for the first time, and com- prising only the genus Dibamus, go still further in the direc- tion of degradation, and are exactly analogous in this series to the Aniellida^ in the other series. Mr. A. G. Butler on a new Species of Pseudacraja, 123 XVII. — Description of a new Species of Pseudacrtea from Natal By Akthur G. Butler, F.L.S., F.Z.S., &c. The genus Pseudacrcea is one of the most interesting- groups of butterflies, the species of which mimic the various forms of Acrcca, Planema, &c. The present species is in the collection of Mr. Walter de Rothschild, and was captured in Natal by Mr. Peter Colville, after whom I have much pleasure in naming it. The nearest allies of P. Golvillei are P. Trinienii and P. Boisdavaliij between which it is in some respects interme- diate ; it appears to me to resemble Acrcea horta rather than the groups copied by its two allies. Pseudacrcea ColviUeij sp. n. illa3 are much smaller and are distinctly separated from eacli other. * Loc, cit. p. 173. + Loc. cit. p. 114, pi. v., with five figures. X " Notes on Moa-caves in the Wakatipu District," by Taylor White (* Transactions ' &c. vol. viii. p. 97). § Note added to the preceding by F. W, Hutton (ibid. p. 101). 134 M. de Quatrefages on Taking into account these various data and the characters distinguishing the Brevipennes inhabiting the other regions of the globe, we may form a very precise idea of what these large species of Moas were like. They presented the general form of the emeu [ = Cassowary], but upon a much larger scale. Like this, they had the greater part of the neck naked ; but they were destitute of the characteristic crest, and in this respect resembled the emou \promcBus\. Very probably the legs were naked and the body was covered with silky plumes, in which darker or lighter and more or less reddish tints of brown predominated, variegated with black and white, at least in some species. Documents, to which I shall have to return further on, enable us to complete this picture, and make known to us the mode of life of these strange birds*. The Moas were slug- gish and stupid animals, as is shown by a proverb which is repeated at the present dayf. They were essentially seden- tary and went about in pairs accompanied by their young. No doubt they sometimes disputed the field on which they were seeking the same food, for the Maoris still, in speaking of a struggle between two pairs of combatants, say — " Two against two, like the Moas." Their nests were formed of various dried grasses and fragments of ferns simply brought together into a heap. They ate various species of plants growing upon the borders of the woods and marshes, the young shoots of certain shrubs, &c. ; but their principal food appears to have been the root of a species of fern which they dug up either with the beak or with the feet. To assist in the trituration of these articles of food, the Moas, like many other birds, swallowed small pebbles, which, when rounded and polished by friction in the stomach, acquire a peculiar aspect and are still called Moa-stones by the natives, who know them well \. But this very polish rendered them unfit for the service which the bird expected from them, and then he disgorged them just as do the ostrich and the emou§ * Letter from Mr. John White to Mr. Travers (' Transactions ' &c. vol. viii. p. 81). Mr. Travers informs us that his correspondent has occupied himself for more than thirty-five years in collecting all possible information upon the past of the Maoris, that he has been initiated by their priests into all the mysteries of indigenous knowledge, so that he knows the history of their race better than the natives themselves. f Extracts from a letter from F. L. Maning, Esq., relative to the extinc- tion of the Moas ('Transactions,' &c. vol. viii. p. 102). The author translates the Maori proverb by the words, *'as inert (ngoikae) as a Moa." \ ' Hochstetter,' p. 186. § " Note on Discovery of Moas and Moa-huuters' Remains at Patana River, near Wangarey," by J. Thorn, jun. (' Transactions.' &c. vol. viii. Moas and Moa-hunters. 135 [Promceus]. These stones were not always of the same nature, and varied with the localities*. III. The details that I have just given assume not only that man and the Moas were contemporaneous, but also that the disappearance of the latter is of recent date. Such, in fact, is the conclusion to which we are led by the results of a regular inquiry pursued in New Zealand for nearly forty years by a great number of investigators and distinguished naturalists. Nevertheless, until the last few years it was quite permissible to liarbour doubts on the subject. One of the most autho- ritative of New-Zealand geologists, Dr. Julius Haast, has pronounced most absolutely in a very different sense. While accepting as demonstrated the coexistence of man and the Moas at a very distant epoch answering to our prehistoric times, he denies that the Maoris themselves ever knew these great birds f. On the other hand, Mr. W. Mantell, to whom his nume- rous researches justly give an authority upon this point, has distinctly and repeatedly expressed the opposite opinion, and believes that these large Brevipenues were hunted and exter- minated at a comparatively recent epoch by these very Maoris J. Lastly, Mr. Stack, who is accepted by his confreres as a very competent judge, has adopted an intermediate opinion. He regards the belief iu the recent destruction of the Moas as inadmissible, but does not wish to throw it back into a very distant past§. p. 85). A certain uumber of these Moa-stones have been collected and appear in the museum at Auckland, and no doubt in other collections in New Zealand. * Haast, loc. cit. p. 73. t " Moas and Moa-hunters : Address to the Philosophical Institute of Canterbury, 1871," by Julius Haast (' Transactions,' &c. vol. viii. p. 66 1872). Dr. Haast has maintained his original opinion in other memoirs' and in the work that he has published under the title of ' Geoloo-y of the Provinces of Canterbury and Westland, New Zealand,' 187U. X " Oil the fossil Remains of Birds collected in various Parts of New Zealand by Mr. Walter Mantell of Wellington," by Gideon Algernon Mantell, LL.D., F.Pi-.S. (Quart. Journ. Geol. Soc. vol. iv. p. 22o^ 1848)- "Address on the Moas:" Extracts by W. B. Mantell (' Transactions ' &c! vol. i. p. 18, 1869). However, in this latter paper Mr. Mantell seems disposed to throAv further back the epoch of the destruction of the Moas in consequence of the obscurities iu the traditions which he has been able to collect upon the subject. Mr. White's letter already cited, and to which I shall reAert, fuU^- answers this objection. § " Some Observations on the Annual Address of the President " bv the Rev. J. W. Stack (' Transactions ' &c. vol. iv. p. 107). ' 136 M. de Quatrefages on To show how the question has become elucidated, and justify the point of view which I have adopted, it is neces- sary to enter into some details. The bones of Moas have been met with under the most different conditions of deposition. Sometimes they simply rest upon the suface of the ground, or are scarcely covered by a few centimetres of sand*. But usually they are found buried at various depths in the sands of the seashore, in the alluvia of the rivers, in the marshes, and also in caves. The quantity of these remains accumulated in restricted spaces is sometimes very remarkable. In digging canals for the drainage of a marsh at Glenmark there were obtained the remains of 144 adult and 27 young birds f. I might cite many other examples, but I confine myself to a summary of the details given by Mr. Booth of the discovery made by him at Hamilton, in a small half-dried lagoon I. Having been informed of the discovery of a few bones, lie opened a first pit of 4 feet square, and obtained from it fifty-six femora with a proportionate quantity of other bones. Regular diggings were then organized. It was ascertained that the deposit to be worked formed a sort of irregular crescent, measuring 40 feet from one point to the other, and 18 feet across at the middle, with a depth of from 2 to 4 feet. In this restricted space were collected about 3^ tons (more than 8500 kilogrammes) of bones, and those who took part in the work estimated the number of Moas accumulated in this estuary at more than 400. These bones were very unequally perserved. A great number fell into paste at the least contact. Hence they had not all been deposited at the same period. But in consequence of the conditions of interment the Hamilton swamp could not furnish certain data as to the relative age of the deposits. It is otherwise with the caves, which were scientifically excavated * Dr. Hector, loc. cit. p. 115; Dr. Haast, loc. cit. p. 103 ; Rev. J. W. Stack, loc. cit. p. 109 ; Rev. K. Taylor (' Transactions ' &c. vol. v. p. 97). Those bones which were seen in <^reat numbers scattered over the ground have rapidly disappeared. Mr. Stack endeavours to explain their per- sistence for centuries by saying that the Maoris carefully preserved the jungles, whicli, on the contrary, the European colonists have caused to disappear. The latter, by destroying this shelter, have facilitated the action of atmospheric agents and have thus brought about the disappearance of these bones, which had hitherto remained intact. I think it useless to point out what, in this interpretation of the facts, has but little foimda- tion and is opposed to daily experience. t Haast, loc. cit. p. 89. [ "Description of the Moa-swamp at Hamilton," bj B. S. Booth (' Transactions ' &c. vol. vii. p. 123, pi. v.). Moas and Moa-Jmnters. 137 bj Hochstetter. Here very distinct layers, separated by a bed of stalagmite, contained different species. At the top was Meionornis didiformis, at the bottom Palapteryx elephantopus. The bones of the former seemed to be still fresh, those of the latter were semi-fossilized. This diversity of aspect corre- sponded with differences of chemical composition, itself con- nected with more or less advanced alteration. The quantity of organic matter found in the bones of Moas which have been analyzed has proved very variable. It is sometimes only 10 per cent. ; but sometimes, also, it rises to 30 per cent. — a proportion almost exactly the same as is met with in fresh bones of the ostrich*. Hochstetter, arguing from his own personal observations and from some previously known facts, approximated to the opinions of Dr. and Mr. W. B. Mantell. He thought that the extinction of the Moas could not be thrown back several thousand years f- He regarded their existence as alone capable of explaining the development which the popu- lation of New Zealand had attained J, and attributed the origin of anthropophagism to the deficiency of animal food resulting from their extermination §. He consequently iden- tified the existing race of Maoris with the hunters of the Moas. To sustain a very different doctrine Dr. Haast especially appeals to geology. The boues of Moas, he says, occur principally in the beds which were formed during the glacial period or immediately after it|i. Having himself collected a certain number of these bones in situ, it seems to him to be demonstrated that these large birds represented, in New Zea- land, the gigantic quadrupeds which inhabited the northern hemisphere during the post-Pliocene period. Hence he does not hesitate to refer the existence of the Moas to an epoch as far from the present time as that of the mammoth, tlie rhino- ceros, cave-lion, and cave-bear, the remains of which are found in European Quaternary deposits ; and he asserts that if the Moas survived these times, geologically so different from ours, they were nevertheless speedily annihilated^. It will be seen that Dr. Haast seems to assume not only the analogy of the glacial phenomena which took place in New Zealand and in Europe, but also their contemporaneity. We have to do here with geology proper, and questions of this kind are out of my province. Nevertheless, even accept- ing these two propositions as true, and reasoning by analogy, * Hochstetter, he. cit. p. 190. f Ibid, \ Loc. cit. p. 194. § Loc. cit. p. 196. II Loc. cit. p. 68. ^ Loc. cit. p. 7o. 138 M. de Quatrefages on we might at once raise some grave objections to the conse- quences which Dr. Haast draws from them with regard to the antiquity of the extinction of the Moas. It is very true that the great Mammalia mentioned by Dr. Haast no longer exist and are known to us only by their remains. But with them lived other species which survived them, and are even still living. The monks of St. Gall still ate the urus in the fifteenth century ; the reindeer, in Pallas^s time, descended in winter to the shores of the Caspian Sea ; the aurochs and the elk still inhabit Poland ; the chamois, the ibex, and the marmot are close to us. Why should all these species of Moa have been condemned to perish with the geological period that witnessed their appearance? Dr. Haast no doubt would object to me that the European Mammalia of which I have mentioned the names, and others wliich it is useless to enumerate, have generally migrated either in longitude or in altitude. But, even without bringing the action of man into play, this change of habitat was im- posed upon them by the transformation of the nature of the climate. Tliis had become continental, instead of insular as it Avas in glacial times. In New Zealand this was not the case. Whatever may have been the movements of elevation or depression of its land*, it remained isolated in the middle of the sea, and its climate cannot have varied, at least in the lower regions, except within very narrow limits. Dr. Haast himself, although starting from other data than those indi- cated by me, insists upon considerations of the same kind, and shows very well that, in this great island, the extension of the glaciers by no means involves the existence of a climate much more rigorous than that of the present day f. The general conditions of existence remaining the same, what reason can the New-Zealand palseontologist give for regarding the extinction of all the Moas as necessary ? In all his writings, published to the present day, which have come to my knowledge, Dr. Haast maintains the general opinions indicated above}. It would seem that they have * Tlie ' Transactions of the New Zealand Institute ' contain several memoirs explanatory uf tlie glacial phenomena of which New Zealand was the theatre. I need not dwell upon these, and I shall only call attention to those of MM. Travers and Dobson, who, in expounding then* own views, ha^e summed up those of their confreres (see " Notes on Dr. Haast's supposed Pleistocene Crlaciation of New Zealand," by W. F. L. Travers, vol. vii. p. 409, and " On the Date of the Glacial Period," by A. Dudley Dobson, ibid. p. 440). But, upon this question, Dr. Haast's work upon the geology of the Provinces of Canterbury and Westland should especially be consulted, t Loc. cit. p. 72. X Besides the " Address," above cited. Dr. Haast has published, in the Moas and Moa-Jiunters. 139 for him the value of axioms capable of serving as a criterion — so much so that positive or negative facts have no value in his eyes, or rather cannot really have taken place, except they agree with his theory. If we speak io him of more or less complete skeletons found upon the ground side by side with a little heap of Moa-stones, which would seem ito indicate that the bird died upon the spot and has never been buried, he declares that he cannot believe that these bones could have resisted the influence of atmospheric agents for hundreds if not thousands of years *. If we speak to him of the recol- lections preserved by the natives with regard to the existence of the Moas, their external characters, their mode of life, and the means employed in killing them, he replies, that the most civilized Europeans have no traditions relating to the mam- moth and the rhinoceros, and that an inferior race which has attained only to a condition corresponding to that of our neolithic populations cannot have preserved any relating to an epoch separated from them by an immense number of years f. He adds that distinguished men have vainly inquired into the traditions in question J. He refers particu- larly, like Dr. Colenso, to the fables which, in New Zealand as everywhere else, have become mixed up with the recol- lection of actual facts in the memory of peoples §. He con- nects what is said of the Moas with vague reminiscences of Cassowaries brought by the Maoris from their original country ||, or with information furnished by occasional emi- grants^. The examination of the ovens, exactly like those of the present islanders, and of the remains of repasts contain- ing bones of Moas, furnishes him with a demonstration of the contemporaneity of certain men with those birds ** ; but the former, in his eyes, were an absolutely savage population, knowing only how to chip and not to polish stone. If some ' Tiausactions of the Nev/ Zealand luatitute/ tlie loliowiug iiieinoirs upon the same subject :— in vol. iv. 1872, " Additional Notes," p. 90; " Third Paper on Moas and Moa-hurters," p. 94, pi. vii. ; in vol. vii. 1875, " Researches and Excavations carried on in and near the Moa-bone Point Cave, Sumner Ptoad, in the Year 1872," p. 54; "Notes on an Ancient Native Burial-place near the Moa-bone Point Cave, Sumner," p. 54, pis. iii. & iv. ; " Notes on the Moa-himter Eiicampment at Shag- Point, Otago," p. 91 ; " Results of Excavations and Researches in and near the Moa-bone Point Cave, Sumner Road (Postscript)," p. 528. Dr. Haast has also maintained his theory and the consequences whicli he derives from it in his book entitled, ' Geology of the Provinces of Canterbury and Westland, New Zealand,' 1879. * "Address," p. 71. t Ibid. p. 75. t Ibid. p. 76 et xeqq. § Ibid. p. 70. II Ibid. p. 77. i, Ibid. p. 106. *' Ibid. p. 82. 140 M. de Quatrefages on Moas and Moa-hmters. polished haches have been found mingled with the ancient kitchen-middens, this, he declares, is because they were lost or intentionally hidden in modern times, long after the hunters of the Moas had disappeared*. The latter, he says repeat- edly, never had any thing in common with the Maoris who occupied New Zealand at the time of the arrival of the Euro- peans. I think I have sufficiently indicated the mode of reasoning and the nature of the arguments employed by Dr. Haast. I shall not follow him here into the discussion of a number of subjects upon which he touches, but which are only indirectly connected with the principal question. However, I think I ought to quote literally the conclusions with which he termi- nates his third memoir f : — "1. The different species of the Dinornis or Moas began to appear and flourish in the post-Pliocene period of New Zealand. "■ 2. They have been extinct for such a long time that no reliable tradition as to their existence has been handed down to us. " 3. A race of Autochthones^ probably of Polynesian origin|, was contemporary with the Moa, by whom the large wingless birds were hunted and exterminated. " 4. A species of wild dog was contemporaneous with them, which was killed and eaten by the Moa-hunters. " 5. They did not possess a domesticated dog. " 6. This branch of the Polynesian race possessed a very low standard of civilization, using only rudely chipped stone implements, whilst the Maoris, their direct descendants §, had, when the first Europeans arrived in New Zealand, already a high state of civilization in manufacturing fine polished stone implements and weapons. "7. The Moa-hunters, who cooked their food in the same manner as the Maoris of the present day do, were not can- nibals. "8. The Moa-hunters had means to reach the Northern Island, whence they procured obsidian ||. * Ibid. pp. 85, 104. t Third paper, * Trausactious ' &c. vol. iv. p. 106. X It is difHcult to understand the association of ideas which Dr. Haast here wishes to express. § Here, agaui, Dr. Haast's idea is not easy to understand. Throughout he carefully distinguishes the existing Maoris from the Moa-hunters. Here he seems to regard the former as being the grandsons of the latter. II Dr. Haast's investigations were made principally in the province of Canterbmy, which is in the South Island. On the Presence of Eyes in the Shells of the Cliitouid£e. 141 " 9. They also travelled far into the interior of this island to obtain flint for the manufacture of their primitive stone implements. " 10. They did not possess implements of nephrite (green- stone) ^■. " 11. The polishing process of stone implements is of con- siderable age in New Zealand, as more finished tools have been found in such positions that their great antiquity cannot be doubted, and which is an additional proof of the long ex- tinction of the Moas." Thus Dr. Haast here appears to be absolute in every thing, and it is with an appearance of absolute certainty that he asserts or denies facts. But we shall see that he has himself been obliged to go back over some of these propositions and to recognize that some of them are not well founded. Never- theless the general convictions of the learned geologist have not been shaken on this account, and we shall have to inquire whether this persistence is justified. [To be continued.] XIX. — On the Presence of Eyes and other Sense- Organs in the Shells of the Chitonida^. By H. N. Moseley, M.A., F.R.S., Linacre Professor of Human and Comparative Anatomy in the University of Oxford. On examining a specimen of Schizochtton incisus, preserved in spirit amongst a lutmber of other animals dredged by Captain W. Chimmo, R.N., in the Sulu Sea, in H.M.S. 'Nassau' in 1871, and by him presented to the Anatomical Department of the Oxford University Museum, I was asto- nished to remark on the shells certain minute, highly refracting, rounded bodies arranged in rows symmetrically ; they struck me at once as resembling eyes, and further examination proved that such is really their nature. On searching for eyes on the shells of other Chitonidaj I found them present in the majority of the genera, differing, however, in each genus more or less in structure and arrangement. The eyes in the Chitonida3 are entirely restricted to the outer surface of the shells on their exposed areas (tegmentum), not extending at all on to the laminas of insertion (articula- * It is with this stone, often called jade, that the Maoris fabricated their stone clubs, haches, and various ornaments. It was of great value in their eyes, and often plays a part in their legends. Upon this point I have given some details, borrowed from Sir George Grey, in a book en- titled ' Les Polynesiens et leurs migrations.' Anil. & Mag. N. Hist. Scr, 5. Vol, xiv. 11 1-12 Prof. H. N. Moseley on the mentum), and never being present on the girdle or zona, which is occupied, as is well known, by various calcareous structures, some of which have been carefully investigated by Reincke *. In the case of all the intermediate shells the eyes are con- fined to the areas laterales, or to the line of demarcation between the area? laterales and the area ventralis, which latter is usually entirely devoid of them. The eyes, which are mostly circular in outline as seen on the shell-surfaces, measure about yt7> of an inch in diameter in Scht'zocJnton incisus, -g}^ of an inch m Acantliopleura spini- gera^ and in CorepMum aculeatum (in which they are oval in outline) ^^ of an inch by about i^^. In Enoplocliiton they are smaller still and only with difficulty seen at all. The eyes appear, when viewed by reflected light with a low powSr of the microscope, as highly refracting, convex, circular spots, looking as if made of glass or crystal ; they are surrounded and set off by a narrow zone of dark pigment, which is the margin of the choroid seen through the superficial shell-substance. In the centre of each convex spot is a smaller circular area, somewhat darker, caused by the outline of the iris, but showing a brilliant speck of totally reflected light, due to the lens. The entire substance of the tegmentum in the Chitonidffi is traversed by a series of branching canals, which are occu- pied in the living condition of the animal by corresponding ramifications of soft tissues, accompanied by abundance of nerves. The nerves and strands of other soft tissue enter the substance of the tegmentum along the line of junction of its margin with the upper surface of the articulamentum. A narrow area, perforated all over by pores, so as to have a sieve-like appearance, here intervenes between the two com- ponents of the shells, and in some shells the actual margin of the tegmentum itself is perforated. In the case of the inter- mediate shells, in most genei'a there are a pair of slits (incisuras laterales) , one on either side, in the lateral lamina of insertion ; these slits lead to two narrow tracts in the deeper substance of the shell, which follow the line of separation between the area centralis and the areas laterales of the tegmentum. These narrow tracts are permeated by numerous longitudinal canals which lodge each a specially large stem of soft tissue and nerves, which ramifies in the substance of the tegmentum. Corresponding with this tract on the under surface of the shell are a series of minute openings leading into it, through which further strands of soft tissue, possibly mostly nervous, pass • " Beitrjige zur Bildimgsgescliiclite der Staclielu &c. im ^lantclrandc der Ohitoneii," Zeitschr. fiir wi.'^s. Zool. Bd. xvii. S. 305. Presence of Eyes in the Shells of the Chitonidic 1-13 from the surface of the sliell-bcd into the shell, to give the general network of soft tissue. In the anterior and posterior shells there are usually a considerable number of such marginal slits, each with a corresponding tubular tract and ramifying strands of soft tissues. The network of soft tissues contained in the canals within the tegmentum ramifies towards the shell-surface and termi- nates there either in eyes or in peculiar elongate bodies, which, apparently, are organs of touch. These latter are long, some- what sausage-shaped bodies, which terminate at their free extremity in dicebox-shaped plugs of transparent tissue, which show a somewhat complicated structure. The tegmenta of the shells of most Chitonidai are perfo- rated at tlie surface by circular apertures or pores of two sizes, arranged in more or less definite patterns with regard to one another and sometimes with regard to the eyes also. The end plugs of the sense-organs above described lie in these larger pores. From the sides of the sausage-shaped sense-organs are given off more or less numerous fine strings of soft tissue, which, diverging, pass to the smaller pores above described and there terminate in very small plugs, just like those of the larger similar organs, but less complex in structure. The eyes are evidently to be regarded as having arisen as modifications of some of the organs of touch above described. They are 'connected with the same network as terminal organs of its ramifications in the same manner, and have points of resemblance to them which are convincing as to the lioraogeny of the two. The soft structures of each eye lie in a more or less pear-shaped chamber excavated in the substance of the tegmentum. The stalk of the pear, which forms the canal for the passage of the optic nerve, is directed always towards the free margin of the tegmentum, whence the nerve reaches it. One side of the bulb of the pear is closely applied to the outer surface of the tegmentum, and here its wall is pierced by a circular aperture, which is covered by the cornea. The cornea is calcareous ; it resists the action of strong boiling caustic alkalies, but collapses at once when treated with acid. Probably some soft tissue is present in its sub- stance, but I have been unable as yet to find it. The cornea in sections shows itself to be formed of a series of concentric lamellai ; its substance is continuous with the general calcareous substance of the tegmentum at its margins. The pear-shaped cavity of the eye formed by the shell- substance is lined by a dark brown pigmented choroid mem- brane of a stiff and apparently somewJiat chitinous texture. This membrane exactly follows the shape of the cavity, but 144 Prof. H. N. Moselcy on the by projecting beyond the margin of the cornea all round forms an iris of less diameter than the latter. A perfectly transparent, hyaline, strongly biconvex lens is fitted in behind the iris- aperture. The lens is composed of soft tissue, and dissolves in strong acetic acid, gradually but completely, showing a fibrous structure in the process. The optic nerve at some distance from the eye is a com- pact strand ; but within the very long tube continuous with the choroid — the nan-ow part of the pear — its numerous fine fibres are much separated from one another and loose. The retina is formed on the type of that of Helix^ and not, as might have been expected, that of the dorsal eyes of Oncidium. It is not perforated by the optic nerve, but is composed of a single layer of very short but extremely distinct and well- defined rods, with their extremities directed towards the light. Beneath them is a layer or several layers of nuclei amongst the ultimate ramifications of the nerve. Not all the fibres of the nerve entering the eye-cavity proceed to the retina. A large number of the peripherally- placed fibres pass outside the retina all round, and, perforating the choroid at its outer margin, end at the surface of the shell, all round the area occupied by the cornea. They terminate in small plugs of tissue, corresponding to those minor organs of touch universally distributed over the shell in the smaller pores already described — being, in fact, exactly similar and identical structures with these. They apparently form a sen- sitive zone round each eye, and they arise from the optic nerve just as do the other minor sense-organs from the nerves of the larger organs of touch. The choroid sacs of the eye show a curious open fold or gutter leading from the bulb superficially along the stalk of the pear, recalling curiously the choroid fissure. Tn some genera of the Chitonidge eyes are entirely absent. This is the case with the genus Chiton. The shell in Chiton is perforated ; the usual small and large pores and the small and large touch-organs are present, but I have as yet found no trace of eyes. I have examined especially C. magnijicus and C. marmoratus. In Moljmlia, Maugina^ Lorica, and Isclinochiton there appear to be also no eyes so far as a cur- sory examination has yielded evidence to me. The arrangement and forms of the eyes vary much in diffe- rent genera, and will probably prove of great value in classi- fication, which has hitherto proved so difficult a problem. The genus ScMzocMton is distinguished by having the mantle deeply notched posteriorly in correspondence with a deep median notch in the posterior shell. In 8chizochiton mctsxis the eyes are restricted to single rows traversing the Presence of Eyes in the Shells of the ChitouidcC. 145 lines wliich in the intermediate shells separate the central from the lateral areas, and which correspond in position with the marginal slits and the courses of the principal nerves. There are six rows of eyes on the anterior shell, two on each of the intermediate shells, and six on the posterior shell — twenty- four rows altogether, with an average of about fifteen eyes in each, or in all 360 eyes. In the specimen examined all the rows except one have the eyes arranged in a single straight row at regular intervals, but at the base of one row there are, as an exception, two eyes side by side. There are also a very few irregularly scattered eyes on the lateral area, showing that the condition here existing is probably derived from one in which the eyes were more ancestrally diffused. In Acanthopleura sjnniger the eyes are irregularly scat- tered around the bases of the tubercles with wliich the surface of tlie tegmentum is covered, and are confined, in the specimens I have examined, to the region of the margins of the shell adjoining the mantle. The eyes in this species seem to be very liable to be broken or to flake off, in consequence of the decay of the surface-laminae of the tegmentum. Hence those remaining on old specimens are tliose probably most recently formed by the mantle at the margin of the tegmentum. In decalcified tegmenta of some species I have seen eyes thus apparently in process of formation and not yet completed. In some specimens to be referred apparently to this species I have been unable to find any eyes at all. It will be necessary to examine a series of specimens of various ages to discover whether the eyes are originally more widely extended over the shell-surface or always marginal only in this species. In a large GorepMum aculeatum, the exposed shells of which were densely covered by a green alga, immense numbers of eyes were found when the alga was scrubbed off, and at the newest margin of the shell not yet encroached upon by the plant. The eyes are very small and their corneas are oval in outline, the long axis of the oval being directed vertically parallel with the height of the shell. The two kinds of pores are arranged in vertical parallel lines with great regularity, the large pores occurring at intervals in the lines of smaller pores. The eyes are never placed on the tubercles, with rows of which the shell is covered, and which are possibly contrivances for protecting the eyes from being rubbed and destroyed. The eyes are present in enormous numbers. I estimate the numbers present on the anterior shell alone at 3000, counting only the younger ones, which are in good condition, near the free margin of the tegmentum, and not the older eyes, more or less destroyed by the boring of the shell by algte and animals on the rest of the. area. On the remaining shells, at 146 On the Presence of Eyes in the Shells of the Chitonidaj. a moderate estimate, reckoning, as before, only the eyes in tolerable condition, there must be at least 8500 eyes. In Tonicia marmorata the eyes have the peculiarity of being sunk in little pit-like depressions of the shell-surface. This, no doubt, is a contrivance for preventing them from being- worn off, and the result is that they are all retained complete in large old specimens. They are arranged in single, straight, radiating rows on the anterior and posterior shell, disposed with considerable symmetry. There are thirty-four such lines on the anterior shell in one specimen, containing about eighteen eyes each. On each lateral area of the intermediate shells there are from two to four similar rows of eyes, with a few eyes grouped irregularly also. In some forms placed in the British-Museum collection as species of Tonicia, there are no eyes present ; these possibly will be found to require to be placed in a separate genus. In Ornithochiton the eyes are not sunk so deeply in pits, but are disposed somewhat as in Tonicia, though the rows are not so regular. In Ghitonellus there are no eyes and but a scanty supply of organs of touch. I have been unable to trace the nerves supplying the shells and eyes directly to their source, although I have no doubt that they proceed from the parietal (branchial) nerve, from which I have traced numerous offsets proceeding in the required direction. I have searched in vain for any similar eyes in the shells of Patella and allied genera. The tegmentary part of the shell of the Chitonidai appears to be something siii generis, entirely unrepresented in other Mollusca. Its principal func- tion seems to be to act as a secure protection to a most exten- sive and complicated sensory apparatus, which in the ChitonidjB takes the place of the ordinary organs of vision and touch present in other Odontophora, and fully accounts physiologi- cally for the absence of these in the group. In some respects the arrangement of the hard and soft parts curiously resembles that existing in the Brachiopoda. It is most remarkable that these eyes should have been missed hitherto by all writers on the shells of the Chitonidas. The fact is due, no doubt, to their minuteness and to the fact that they are not very easily seen with a powerful lens in the dried condition of the shell in most instances. In order that they may be made most conspicuous the dried shell should be wetted with spirit, and a lens as powerful as Hartnack's no. 4 objective be used. Dr. W. B. Carpenter^ observed the perforate structure of the tegmentum in Chiton, but did not apparently investigate * ' Cyclopoedia of Anatoiuy and Physiology/ article " Shell," p. 565. Miscellaneous. • 147 the contained soft structures. He writes, "In C7«'ton the exter- nal layer, which seems to be of a delicate fibrous texture, but which is of extreme density, is perforated by large canals, wiiich pass down obliquely into its substance, without pene- trating, however, as far as the middle layer.' ' My father-in-law. Dr. Gwyn Jeffreys, has pointed out to me that Costa* figures what are evidently the eyes on one of the intermediate shells of a very small species of Chiton {To7iicia) ruhicundus. They are figured as mere black dots and referred to as fine punctuations, but their arrangement is correctly shown. The late Dr. Grayf, in his well-known paper on the struc- ture of the Chitons, wrote: — " The greater number of species have a part of the valve which is not covered by the mantle, but exposed. This exposed part consists of a perfectly dis- tinct external coat, peculiar, I believe, to the shells of this family. The outer coat of these valves is separated from the lower or normal portion by a small space, filled by a cellular calcareous deposit, which is easily seen in a section of the valves." I have prepared drawings illustrating the arrangement and structure of the eyes and other sense-organs in the shell in various genera of Chitonida?., and hope to publish them with a more complete account of my results in the coming winter. I beg to express my best thanks to Dr. Giinther for giving me every facility in making use of the British-Museum col- leetiou. Dr. Woodward kindly went over the fossil Chitons in the Paleeontological Department with me, but we could detect no traces of eyes in any of them. This is remarkable, since the ancient forms of the group appear to be allied to Schizochiton. MISCELLANEOUS. On the Submaxillary in Masticating Insects. By M. J. Chatin. The maxilla in masticating insects is supported by a basal piece the functional importance of which cannot be disputed, but which possesses a still greater interest from the point of view of the mor- phology of the parts of the mouth and even of the appendicular organs considered generally. Nevertheless it has hardly been even mentioned l)y a few writers, among whom we must cite Kirby and Spcnce, who gave it the name of the cardo (hinge), a term happily enough representing its mode of articulation ; Erulle gave it the name of submaxillary, which I here retain, so as not to introduce any neologism into an exposition already full of details. ' Fauna di Napoli : Animali molli, Chitone,' taf. iii. fig. l,e. J. E. Gray, " On the Structure of tlie Chitovs,'''' Phil. Trans. 1848. 148 Miscellaneous. In order to acquire a sufHciently exact knowledge of tho funda- mental characters of the submaxillary and of the variations which it may present, it is indispensable to multiply the objects of inves- tigation and to select them with care, not limiting observations to a few common species which have been almost exclusively studied. OUgotoma Saundersii may be taken as a starting-point for this series of analytical and comparative investigations. Its submaxillary in fact is very simple ; it has the appearance of a small piece trans- versely developed and rising slight!}^ on its internal surface, where a prominence, which will soon become more strongly marked in other types, is sketched out. In (Edipoda cinemscens the form is already considerably modified, chiefly as regards the configuration of its lower surface. This is not only destined to limit the submaxillary towards its base, but it has also to provide for the articulation of the maxilla considered as a whole ; the ginglymus, scarcely represented in Olujotoma by slight sinuosities, here gives rise to the formation of deep cavities which impress a peculiar physiognomy upon this region of the submaxillary. Entomologists have long since indicated the genus (Edijioda as one of those in which the maxilla is most firmly articulated with the head. It will be seen that this remark fully agrees with the results of anatomical analysis. In Decticus verrucivorus the general aspect undergoes further changes, the origin of which must be sought in the inner and outer surfaces, but no longer on the basal surface. Each of the lateral surfaces commences with an inferior tuberosity ; then comes an excavated middle part, surmounted by an upper portion, which is very prominent, especially at the outer surface. From this results a most singular form, which can only be correctly interpreted when we examine the submaxillary isolated and freed from the surrounding parts. This dissection, always delicate, is particularly difficult in Gn/lJus domesticus, the submaxillary of which presents an appearance which, more than in the preceding types, justifies the name selected by Kirby and Spence ; the depressions and articular facets of the iafe- rior and superior surfaces, the orientation of the piece and its rela- tions, all concur here to form a regular hinge. On the other hand, the articulation of the maxiUa is very feebly constructed in Phasma japetus, in which several of the characters proper to Gryllus domesticus are effaced. This tendency is still more strongly marked in Mantis religiosa ; the submaxillary, chiefly developed vertically, becomes in that species almost abnormal, and in its general conformation greatly resembles some maxiUaries, In the great green grasshopper (Locusta viridissima) it better displays the double part assigned to it, of securing the articulation of the maxilla and forming for it a sufficiently solid base to support the whole organ, and thus to second or even replace the maxillary. Thus the inferior surface is deeply excavated, while the transverse dimensions become more appreciable. The relative proportions of the different parts of the submaxillary Miscelhin eous, 1 4 J) •are so profoundly modified in Ilydrophilus piceus that we have some difficulty in recognizing them, especially in a rapid examination. The inferior surface is undulated and the outer surface rather short ; the inner surface presents a marked ohUquity and bears a tuberosity which claims our more particular attention, because this arrange- ment, indicated in Olv/otoma Saiindersii &c., tends to become general in many other masticating insects. The mandibles, as is well known, play the most active part in the division and mastication of food ; but the masilho also assist in the operation to a variable extent according to the species, and the inferior projection of the inner surface from this point of view acquires particular importance. It did not escape Latreillo, who sometimes mentions it under the name of molar. It is pretty con- stantly met with, but it presents frequent modifications. I confine myself to indicating the following : — In Garahus auratas this prominence occupies an intermediate position between the lower and the inner surface ; in Forjicula auricularia it becomes conical and represents a lacerating rather than a grinding tooth ; in Blaps producta it seems to bo wanting, but its absence is compensated by a peculiar arrangement : the submaxillary considerably exceeding the maxillary, especially within, the inner surface of the submaxillary comes to project at the base of the maxillary, and may thus in its entirety fulfil the function generally reserved for the " molar" above indicated. Although reduced to their essential points, the preceding descrip- tions suffice to show on the one hand all the interest that attaches to the morphological study of the submaxillary, and on the other the variations presented by this piece, which is too often overlooked, but the correct interpretation of which is indispensable in the com- parative investigation of the appendicular organs in the Arthropoda. — Comptes Rendus, July 7, 1884, p. 51. On a new Type of the Glass Hirudinese. By MM. PoiKiER and A. T. de Rochbrttne. As the crocodile lives in the water, says Herodotus, the interior of his mouth is covered with Bdellas (Lib. II. Chap. Ixviii. j). 94, cd. MiiUer). The translators of the Greek historian, down to Scaliger, understood the word fjhXXeiov to refer to leeches ; since then several have asserted that these animals were Diptera of the genus Gulex. The scientific researches of one of us during a pretty long sojourn in Seuegambia enable us definitely to settle a still controverted question, and to prove that the BdeUas of Herodotus must be referred to the class Hirudineaj. The remarkable type under consideration lives attached not only to the buccal mucous membrane of Crocodilus vidgaris, cataphrac- tus, and leptorhynchus., but also to the lingual papillae of Grjmyio- plax cBgyptiacus and to the interior of the pouch of Pelecanus crispus and onocrotalus. In its general form and the presence of branchial tufts on each Ann. di- Ma(j. N. Hist, fter, 5. Vol. xiv. 12 1 50 MisceUaneo us. siped hache and the polished hache are among the characteristic traits which, among us, distinguish two epochs. It is also well known that the populations of these two epochs belonged * Sixth proportion, p. 140. t Second proposition, p. UO. 160 M. de Quatrefages on to different races, and that the one wliicli was more advanced in civilization attacked and conquered that which preceded it. To. find in New Zealand our two Paleolithic and Neolithic ages, characterized in the sauieway by instruments indicating a difference in social condition, was to introduce an important argument in favour of the ethnological distinctness of the Moa-hunters and the Maoris. But, by excavations in the Sumner cave and the neighbouring dunes. Dr. Haast himself discovered, at various times, fragments of haches and other instruments perfectly polished ; and further, some unin- jured specimens, similar in every respect to those which are known to be the w^ork of the Maoris. Among these objects some were in nephrite. All of them were found under con- ditions which attested their contemporaneity with the men who hunted and ate the great brevipennate birds. I shall only cite a hache which was pL'iced immediately beneath the stones forming an oven which had served for cooking Moas*. In presence of these material proofs, furnished by himself, Dr. Haast, Avith the most honourable candour, did not hesi- tate to admit that the Moa-hunters had attained a degree of civilization equal to that presented by the Maoris when Europeans first visited New Zealandf- It is, I thinkj permissible to think that this equality of social development, manifested by similar characteristic indus- tries, ought to have inspired Dr. H«ast with some doubts as to the soundness of his theory. Nevertheless he has not given up any of his general ideas. He has persisted in denying the ethnical identity of the Moa-hunters and the Maoris, and in throwing back the epoch of the destruction of. the Moas into a past time which he seems to regard as geological |. I find no one except Mr. Colenso who has accepted this doctrine as absolute §. I have already stated that Mr. Stack • " Researclies in Sumner Moa-Cave " (' Transactions ' &c. vol. vii. P-77). t Ibid. p. 80. Before Dr. Haast had given up this particular point, numerous discoveries of instnuueuts and weapons in polished stone mixed with remains of Moas had been made in many places. I have already stated how Dr. Haast had endeavom-ed to explain or intei-pret facts of this nature, and I need not revert to this matter. The loyal and distinct declaration of the eminent geologist frees me from the necessity of enter- ing here into anv details, X Haast, ' Geology of the Provinces of Canterbury and Westland, New Zealand.' See especially the thirteen propositions formulated at p. 430, and chap. xvi. (c) p. 437. § '' An Account of some enormous Fossil Bones of an unlmown Species of the Class Aves, latelv discovered in New Zealand," in Ann. & " Nat. Hist, 1844. Moas and Moa-hunters, 161 refuses to admit a veiy great antiquity for the destruction of the Moas. He also recognizes that the Maori traditions contain some allusions to these birds. In his childhood he heard talk of Moa-feathers found upon a rock where the last of these Brevipennes had concealed itself. However, he also thinks that this may perhaps have referred to feathers of Cassowaries brought to New Zealand by the ancestors of the Maoris*. We see that Mr. Stack does not regard the latter as descendants of the autochthonous Moa-hunters supposed by Dr. Haast. Upon this latter point, moreover, the ideas of the New- Zealand geologist do not appear to be by any means fixed. I hav^e reproduced above the terms employed by him in the conclusion of his third memoir, and have cursorily indicated how vague and contradictory they are, notwithstanding their apparent precisionf. In another memoir he expresses a verv different idea, and regards the Melanesian negroes as havin'o- preceded the Maoris in New Zealand, and ascribes to them the extermination of the Moas J. Moreover, in support of his new opinion he invokes only those very traditions which we have seen him reject in the most formal manner. Still, he only knows them fi-om the Rev. Richard Taylor's book. It is from this that he borrows a quotation from Sir George Grey, whose classical work § he does not seem to have read. Lastly, in his geology of the province of Canterbury, he formally adopts Mr. Colenso's views, and repeatedly speaks of the predecessors of the Maoris as autochthonous inhabitants who lived in the Quaternary epoch. At the same time he supposes that these children of the soil of New Zealand had more or less close affinities with the Melanesians ||. I have too often contended against this old idea of autoch- thonism, to render it necessary for me to revert to it here. But this conception being got rid of, I am glad to agree with L)r. Haast. The opinions maintained by the learned geologist as to the existence of two races inhabiting New Zealand before the arrival of Europeans, and as to the nature of those * "Notes ou Moas and Moa-hunters'' ('Transactions' &c vol iv p. 108). ■ ■ ' t See the notes at foot of p. 140. \ "Notes on an Ancient Native Burial-phice "' ('Transactions' &.c. vol. yii. p. 91). Dr. Haast has subsequently insisted upon this idea and sought to show, by what takes place in AustraHa, that very inferior bhick tribes may very well know the processes of polishing stoiie (' Geoloo-y of the Provinces of Oantfrbury and Westlaud,' chap. xvi. p. 411). § Polynesian Mythology. II ' Geology,' &c., tirst proposition, p. 430. 162 , M. de Quatrefages on two races, are perfectly well founded. Melanesian negroes really occupied New Zealand before the Maoris. Upon this point craniological investigations have confirmed what I wrote eleven years before the publication of Dr. Haast's memoir*. But this ethnical duality of the New-Zealand populations by no means implies as its consequence the destruction of the Moas by the first occupants. In Europe the Palaeolithic men did not exterminate the reindeer and the chamois, nor even the urus. To support his views and to throw back the extinction of the Moas to a past which, he says, cannot be calculated even by centuries t, Di*> Haast no less invokes the results of his excavations in the Sumner cave. He describes it as con- taining two layers, which, according to him, were distinctly separated. In the lower one were found ovens and numerous Moa-bones ; this was formed of the remains of the repasts of the Melanesians. The upper layer, he states, presented only the shells of various Mollusca, formerly eaten by other natives who were the forefathers of the existing Maoris. Mr. MacKay, a member of the Geological Survey, who assisted Dr. Haast in his researches, has also published a note, in which he puts forward nearly the same opinions as his chief J. But the clearly marked distinction, upon which MM. Haast and MacKay insist, does not occur elsewhere. At several points a mixture of shells and Moa-bones has been met with. And, further, the locality first investigated by those geologists was afterwards explored by Capt. Hutton and Mr. Booth, both of them familiar by long practice with researches of this kind. Now the facts ascertained by them contradict formally and upon several points the statements of the first explorers. Among other things, MM. Hutton and Booth most frequently found the Moa-bones associated with beds of shells ] and they * A. de Quatrefages, " Les Polynesiens et leurs Migrations " {' Eevue des Deux-Mondes,' Februaiy 1864). These articles, enlarged and fur- nished with notes and with four maps, were afterwards collected into a volume, which appeared under the same title. A. de Quatrefages and E. Ilamy, ' Crassia ethnia,' p. 291. Among other evidences of the presence of two races in New Zealand, the Museum possesses a dried head of a Maori chief, the tattooing of which attests its origin, while the hair is purely Melanesian. I have had this engraved in a book, of which I have already spoken (' Hommes fossiles et hommes sauvages,' pp. 486, 487, figs. 171, 172). _ t Loc. cit. (' Transactions ' &c. vol. vii. p. 81). \ " On the Identity of the Moa-hunters with the present Maori Race " (' Transactions ' &c. vol. vii. p. 98). Moas and Moa-hunters. 163 have further ascertained that tlie beds with and without bones were often differently superimposed*. The increasing rarity of the Moas at a given point, the movements of the popuhxtion which must often have been the consequence of it, the accidental association of the two kinds of food in the same repast, and the necessity of having recourse to a diet previously disdained, explain in the simplest manner the difference in the results furnished by excavations made at very adjacent points by equally competent observers. But we see that in tiieir totality these results are irreconcilable with the interpretations of Dr. Haast. V. Among the propositions that Dr. Haast has sustained, those relating to the history of the dog must detain us for a time. We have seen that, in his third memoir, he admits the exis- tence of a wild dog contemporaneously with the Moas, and absolutely denies that the Moa-hunters had domestic dogsf. Upon this latter point the New-Zealand naturalist is far from being in accord with himself. In his first researches he had found only a few bones of the dog among the remains of feasts, and he explained this scarcity by saying that this animal was only exceptionally eaten when its owner was short of provisions J. Here, then, he accepted the notion that the domestication of the dog was practised by the Moa-hunters. It is true, he added, that perhaps also it was killed in the chase, which supposes that the animal lived in the wild state, and it is to this latter opinion that he seems to have finally come. But if this hypothesis were true we should have found, from time to time, the bones of the dog side by side with those of the Moas, its contemporaries. Now we have already stated that no fossil terrestrial mammal has yet been met with in New Zealand §. To this statement the dog forms no ex- * "Moa-bones were never found unassociated with beds of shells, and altbougli shell-beds did occur without Moa-bones, these just as often underlaid beds with Moa-bones as overlaid them" (" Notes on the Maori Cooking-places at the Mouth of the Shag River," by Capt. F. W. Hutton, in ' Transactions ' &c. vol. viii. p. 105). t Fourth and fifth propositions. X " Either when its owner was short of provisions, or perhaps . , . . " (Addiess, he. cit. p. 81)). § In my first article on the Moas, when speaking of the small number of Mammaha found in New Zealand and the absence of fossils of animals of that class, I forgot to add the epithet terrestrial (cieriens). Headers will, however, I fancy, have filled up this omission. Aquatic Ma)nmalia, on the coutiary, have repeatedly been found in the strata of New Zealand 164 M. de Quatrefages on ception *. In fact, the bones of that animal have only been found in the ancient ovens, or among the fragments scattered around primitive kitchens. But then, in opposition to what has been said by Dr. Haast, they occur in abundance. I hardly find an excavator who has not indicated their existencCj and they are always associated with bones of Moas. Here, however, we meet with a fact which may appear singular at the first glance, and upon which the New-Zealand naturalist has repeatedly insisted. The bones of all kinds scattered in the vicinity of the ovens are very rarely gnavvedf. From this Dr. Haast concludes that the Moa-hunters were not accompanied by dogs ; for these, he says, would not have failed to attack the remains of their masters' repasts. But, in speaking thus, he forgets that the canine race introduced into New Zealand Avas essentially destined to furnish food and clothing J. The Maori dog, coming from the Manaia Islands, belonged to that Polynesian race which all travellers represent as living only upon vegetables, and which must have retained its ancient habits in New Zealand §. Moreover, if some dogs took to eating meat their masters would soon have perceived that this food modified the taste (Haast, ' Geology of the Provinces of Canterbury and Westlaud,' chaps. X. & xii.). I have elsewhere referred to the fact that the Cetacea play a part in the traditions of the Maoris (' Les Polynesiens et leurs Migra- tions,' chap, iv.), and that every animal of this kind thrown upon the shore belonged of right to the Ariki, the chief of the territory (' Journal des Savants,' January 1873). * Capt. Rowan has ascertained the presence of a dog's slieleton in the hollow trunk of a tree bnried in the silt of a river near Wellinglon Har- bour. This tree was at a depth of G metres (about 20 feet) and beneath a layer of lignite. But beside and behind the bones there were found the hairs of the animal, with some fibres of hemp and a stalk of the same plant. It is evident that the carcass had been carried into this hole by some flood of the river, and that the event was quite recent. This has been w ell understood by Dr. Hector. That naturalist adds, that the burial of this dog is of earlier date than any other known (" On the Remains of a Dog found by Capt. Rowan near White Clifls, Taranaki," in * Transactions ' itc. vol. Lx. p. 243). t The only fact of this nature that I have seen mentioned in the various memoirs written by the New-Zealand naturalists has been by Capt. Hutton. Two Moa-bones collected by his co//abonifeiS'. septemcarinata from 25 to 150 fathoms. The most remark- able cases are among the deep-water genera. Galacmitha rostrata and G. Bairdii^ from between 1000 and 1500 fathoms, have eggs 3 millim. in diameter in alcoliolic specimens, while in the vastly larger lobster they are less than 2 millim. The largest Crustacean eggs known to me are those of Parapa- siphae sulcatifrons ^ a slender shrimp less than 3 inches long, taken between 1000 and 3000 fathoms. Alcoholic specimens of these eggs are fully 4 by 5 millim. in shorter and longer diameter, fully ten times the volume of the eggs of Pasiphae tarda from 100 to 200 fathoms, more than 350 times the volume of those of a much larger shallow-water Palcemon, and each one more than a hundredth of the volume of the largest individual of the species. From the peculiar environ- ment of deep-water species it seems probable that many of them pass through an abbreviated metamorphosis within the egg, like many freshwater and terrestrial species, and these large eggs are apparently adapted to produce young of large size, in an advanced stage of development, and specially fitted to live under conditions similar to those environing the adults. XXIV. — Notes on Sponges, toith Description of a new Species. By Stuart O. Ridley, M.A., F.L.S., &c. The following remarks are either based on specimens recently added to the collection in tlie British Museum, or suggested by the study of the collection. MONACTINELLIDA. Clialmidse. Cladochalina diffusa, n. sp. Cladochalina diffusa, Hidley, Report on the Zoological Collections made during the Voyage of H.M.S. ' Alert,' p. 672, pi. xli, fig. 1), d, d'. Suberect, branching subdichotomously in one or more parallel 184 Mr. S. 0. Ridley on Sponges. planes ; branches tortuous, more or less compressed, sometimes forming broad expansions terminated by subcylindrical pro- longations, simple or branched ; greatest diameter of terminal branches about 10 millim. Surface either approximately even, or echinated by few and sharp vertical projections, 1-2 millim. high. Vents numerous, opening flush with surface, and entered at a slight depth by the openings of the excre- tory canals ; diameter 1-1 '5 millim., scattered at intervals of 3-7 millim. over the anterior surface of the branches. Consistency in spirit firm, but compressible and elastic ; in dry state firm, but harsh to touch, and but slightly compressi- ble and elastic : colou]' in spirit bright ochreous brown, in dry state pale grey. Main skeleton approximately rectangular in arrangement ; primary fibres about *4 millim. apart at surface, where they terminate vertically in the dermal reticu- lation, diameter about '1 to "14 millim.; secondary fibres vertical to primaries, about '3 millim. apart, diameter '07 to •1 millim. ; fibres of both kinds consisting of a compact axial mass of spicules, and of a margin of transparent pale amber- yellow horny substance about '025 millim. broad. Dermal skeleton forming subquadrate meshes "18 to "SG millim. in diameter, formed of spiculo-fibres, which are usually devoid of any visible horny margin, and range in thickness from *025 to "1 millim. Sarcode pale amber-yellow, subtransparent. Spicules smooth acerate, slightly curved, tapering to sharp points from within about two diameters of ends, size '11 by •0063 millim. Hah. Singapore, between tide-marks. This species was obtained by H.M.S. ' Alert,' and figured in the 'Report ' &c. [I. c. supra), but not described. The largest specimen, which is dry, is very irregularly rooted, and near its base shows the palmate development of the stem and branches very strongly ; it measures 190 millim. (7| inches) greatest lateral, and 95 millim. (3| inches) in greatest present height ; its branches are almost smooth, but those of the younger spirit-specimen show the aculeation above described. Both specimens deviate from the erect habit in the turning of the branches outwards and to the sides soon after they are given ofi". The fibre is stronger and stouter than in any Cladochalina with which I am acquainted, and gives the species the firmness of a Pacliychalina ; in the tendency to become flattened it also resembles that genus. The variation in the character of surface aculeation exhi- bited by this Chalinid is important and significant in relation to its value in classification. Mr. S. 0. RiJlej oa Sponges. 185 Axinellidae. Ech inodicttj itm taesenterinum . Spangia mesenterina, Lamarck, Ana. Mus. Hist. Nat. xx. p. 444. ^chinonema vasiplicata, Carter, Auu. & Mag, Nat. Hist. 1882, ix. p. 114. This fine species has been described under the above two haines. I have ah-eady stated (Report on the Zoological Collections made during the Voyage of H. M.S. 'Alert:' London, 1884, p. 454) that Mr. Carter's species is referable to Echinodictijum^ mihi. Examination of a specimen in the Museum at the Jardin des Plantes, Paris, which agrees with Lamarck's description, has shown me that that species is iden- tical with the former. Suberitidae. For the view that the Suberitidse are really Monactinellid and Monaxonid, derived from a Diactinellid type, and not reduced Tetractinellids, evidence is afforded by the heads of the spicules of the species described below. In those heads which {as in many other species) exhibit a small terminal rounded process or knob, the central canal shows a small in- flation near the centre of the larger division of the head, and a fine undilated prolongation in the direction of (but scarcely extending into) the small process, apparently indicating that the spicule was originally prolonged on both sides of the present head, the small terminal knob of the head and the fine prolon- gation of the central canal being rudiments of the second ray. An additional argument in favour of this view is the fact that the projection and its corresponding section of central canal occur in young spicules, and tend to be lost in adult examples. Suherites niassa, Schmidt, var. As the original form of this species has (like, indeed, most known species of Suheriles) never been fully characterized, I think it well to describe some interesting specimens from Mauritius, which differ from the originals only in their external form. Sponge massive, consisting of vertical convolutions or sinuous laminae, about 45 millim. (If inches) high, and 10 millim. thick above, appressed towards each other, dividing and uniting with each other ; they are rounded above and rise to approximately the same height. General appearance that of the human cerebrum. Vents scattered, subcircular, about 1 millim. in diameter, placed low down on the sides of the convolutions at some distance below the top of the sponge; Ann. & Mag. N. Hi^t. Ser. 5. Vol. xiv. 15 186 Mr. S. O. Ridley on Sponges. Colour, in dry state, orange-brown. Skeleton — composed of long, imperfectly separated bands of the skeleton-spicule, massed closely together and parallel in direction, extending from the base towards the apex of the lobes composing the sponge, beneath the layer which forms the immediate sur- face. Dermal layer consisting of short spicular columns, '4- •6 millim. in height, arising vertically or obliquely to the surface, where the spicules spread out so as to form, by the lateral divergence of their apices, brushes, which are in contact with each other laterally ; between the bases of these brushes are placed the chones of the inhalent canal-system. Spicules — spinulate, smooth, head oval to globular, a small basal rounded prolongation in nearly all spicules, except those fully adult, where it appears usually to be wanting ; neck mode- rately distinct ; shaft normally straight, tapering to a sharp point from near middle : size — main skeleton "8 millim. by •019 millim. (both head and shaft), dermal skeleton '