AF ed te aka ; " ee AY Ebi yan bari . PUP Ot NUS HAE Lb 112 20 ge)? 4 adhere ua toe ; veda Hh aL pealehal very 4 the Leet thd Prd ainaas Ti ; erage here (a) git wth ; i} dA Yagi et eid rH . poate eaaipcanh LAST Ostet i mit Ree e Fab Lthrie rae Petar TCH ick Tyee ss ¥ Mah Pbeeabstanegi seh OL Tits ory sens eects ogee 7 tee bat gt sy Ee eset viteasmey et hh ae, ppnrahse a) Niles bobbed Mish eid het te Ne! Eide ok ates 4 att at hid ibe ttt aaitae Sahat piste oy t ygaks abreat bisa seid st i hd Hod Abie ssid a ta dagaga pao el tie my pha ae ty Tits As +4 Piast a Habid 1 4 L. aean opt ra HA Ma pane bhp ay 0 ‘ fut ys Mek ea? wh aw ; "7 ited ee aSsirebedet Ae {gieeieny fpreernr it ‘ b te ; ti ath ott fede sage § : ‘eins Nentitran’ SREY eee at Tent ih ; a Habana es { sty | 4 ure: Wh ant? Mh AF Hine) Harahetle Bi ies : D4 AW a8 iy eat t ae Ho * area asenton A re Awan satnee Haber, te rh nieu anni yee Hi cl rpeiarever nase! ty spre beter t A Monircaiiea * : ie a Wereratomay i toutt. beers ; ; pnt ‘ ; ryt ena oF * Mate cote " i! i 4 i 7 rr ior 4 baat. 6 2 : i ' Teieanayens Tasere He aie tet i Ne siaey Mei { ‘mi fateh adae tait Vad ahd } sie / Pel : 4 AR > vite) t leit Fir yee iaapy son way baht Meetaby rere Syriinsisels CL Viabee Tet! Lee ie i eh + fly Fal eat ee uaus ty Man tet es nt dete CHG Ree at at) ere p43 a risen Fatt pinta Mevebveid v4 hos i) ae The aby ais 7404) Hi it pe APSR M He aa 7 q ry) it stain 4 fet vii Talatel hase ih otha teh ihe tarp tet ah nee dey He sane a eh A AH ; ‘Volts 97a A , is pad ity Bia Hie tf pareetares ie et) AUT ot erie th 4 sau) Wik He Peet AER am athe ely fy ae rope i be aeleyt sh pareena iH Ther 43 3b bear ship fig? tte VAR, inh f unbeiel ie Prceatna TEA y ase malay : pala ¥ berdol =P hata Matty tert : 99 Wi ined y any Hid Fat! “ait qe a 33 ‘ au anee pias) f pated gle bArariebe ees 5 Hert SA teh i Tahteb ie AS \ A A A RAR ARAARAAARSSAASA a \ W/W \ g A) TA \ Ny WANA IANA pry WW a \ ARAARR C (CCE Kage C C C C C C => \ (ar —\ —V— AAAARARA An AARAAA\a\ SAAAR ARR A fal FA | - WAN \\ a iA Al i } | AVN ‘\/ Cate Vy j=x\ AAin A } VW ii \; V/ aaapain,iaie\ | if AAAAAAR AA NP NES Be v ‘i V = lam, \ : BARBARA AA RAAR AAAAY_V_\A\ ry AA RAAAAA Sj \ =~} } OW | | alan! ] | { | \ \ AA DAAAA AA f mip \ i q \ | y \ \, : ‘ \ es = mm \ ie \ \ | 4 | ) Vim } | } | ik \ \ } \ \ ' Aa? V | } ham | \ Pas Vi "2 fa Ip A; al) na W/ Wi = AVNAY al . ax\\ ill ‘= ENVY W/ fy’ i t if i i ) y Y vv \ i NAA AAAAA\ ala! NA \_\ AAA AANA f RAAAAA \ aN A l \ AR | NEN f } \ i J y | We | | AAAnn ANAAA aanaAss ‘| \i/ x\ = Ni ail 1 \\/ i} ’ gal l\aiapiA \ \ W \ | | ala y) (NY la; | \ I \ \FA i é I-A ‘a Y | a | ‘| Vy ¥ \' | Y Y 1 | i Wy, = Ara RA AA Alala|, PArW WAN \ pal PARIAIAAIAIAAY VV NAY.Y a Ar a\an\, (MN 1 \ | lr Naas (P| } | ] \ in Hv i ay beh nad » Sy BENS } Te THE een a, JO Uo IN A OF THE LINNEAN SOCIETY. ZOOLOGY. WOT XO: G 165990 LONDON: SOLD AT THE SOCIETY'S APARTMENTS, BURLINGTON HOUSE, PICCADILLY, W., AND BY LONGMANS, GREEN, AND CO., AND WILLIAMS AND NORGATE. 1898. Dates of Publication of the several Numbers included in this Volume. No. 167, 168, 169, 170, 171, 172, 166, pp. 1-178, », L79-232, » 233-357, », 308-452, » 453-516, », J17-557, » 008-719, published April 1, 1897. July 1, 1897. November 1, 1897. April 1, 1898. April 1, 1898. July 1, 1898. December 15, 1898. PRINTED BY TAYLOR AND FRANCIS, RED LION COURT, FLEET STREET. LIST OF PAPERS. Page Brenoam, Wm. Buaxvanp, D.Sc.(Lond.), M.A.(Oxon.), Aldrichian Demonstrator in Comparative Anatomy, Oxford. New Species of Pericheta from New Britain and elsewhere; with some Remarks on certain Diagnostic Characters of the Grenistan (Plate sth Grice l Ger myceccstieie anche, 4 hey eeeiousesuetcnen oes 198-225 Bernard, Henry Mryners, M.A.Cantab., F.LS. On the Affinities of the Madreporarian Genus Alveopora with the Paleozoic Favositide, together with a brief Sketch of some of the Evolutionary, Stages of the Madreporarian Skeleton. (Plate 33) ......... Seen catty aber te rendathonarels 495-516 Burne, R. H., B.A., Anatomical Assistant in the Museum of the Royal College of Surgeons of England. The “ Porus genitalis” in the Myaxinide. (Communicated by Prof. G. B. Howes, Sec. Linn. Soc.) (Plate 32) ...... 487-495 CAMBRIDGE, Rev. Octavius Pickarp, M.A., F.R.S. On some Arctic Spiders collected during the Jackson-Harms- worth Polar Expedition to the Franz Josef Archipelago. (Communicated by Prof. G. B. Howes, F.R.S., Sec.L.8.) (TPIS E39) Ase cr eens at Si eten coche ecialotaciaastelny eran ial BI°e 613-615 CaRPENTER, GEorGE H., B.Sc.Lond., of the Science and Art Museum, Dublin. On Pantopoda collected by Mr. W. 5. Bruce in the neighbour- hood of Franz-Josef Land, 1896-97. (Communicated by WWaatehaete Clarke) FIGiS:). (Pilate 46) sat...) 626-654 iv Page . CHAPMAN, FrEpERICK, A.L.S., F.R.MLS. On Haddonia, a new Genus of the Foraminifera, from Torres Straits... (Plate*28))i ia ieseuae eter weit peintcisciemes erie 452-456 CuapmMaNn, Freperick, A.L.S., F.R.M.S., and T. Rupert Jonus, JAGR Shy LAE S. On the Fistulose Polymorphine, andon the Genus Ramulina, 334-354 CuMMINS, Surg.-Capt. Hunry A., M.D., F.L.S. OnithesHoodiot (Wjopoda Next... tiee on aha) cle cee 623-625 Denny, Anrruur, D.Sc, F.L.S., Professor of Biology in the Canterbury College, University of New Zealand. Observations on the Holothurians of New Zealand; with Descriptions of four New Species, and an Appendix on the Development of the Wheels in Chtrodota. (Plates 3-7) .. 22-52 On Pontobolbos, a remarkable Marine Organism from the Gulf OH Milewoeeim, (Cees AS Ce Zl) Sosaceccancaeduodccoos 443-452 On some Points in the Anatomy of Caudina coriacea, Hutton. (Plate 29) is waves ke eies “ivolisiaites ete 5 RLG1S Se 456-464 DurrpsEn, J. E., A.R.C.Sc. (Loxd.). On the Relations of certain Stichodactyline to the Madreporaria. (Communicated by Prof. G. B. Howes, Sec.L.8.) ...... 635-653 Farquuar, H. A Contribution to the History of New Zealand Echinoderms. (Corumunicated by T. W. Kirk, F.L.S., Government Biologist, Department of Agriculture, New Zealand.) (Plates 13 CO US Orn art fr ie AORN ean NPI IPE AIS G46 o o-¢ 186-198 Preliminary Account of some New Zealand Actiniaria. (Communicated by T. W. Kirk, F.L.8., F.R.MS., &c.) (GeICHE S15) aceite in OS en MA reson cd ns bm ooo. 0c 527-537 GitcuristT, J., M.A., B.Se., Ph.D., Marine Biologist to the Govern- ment of the Colony of the Cape of Good Hope. Notes on the Minute Structure of the Nervous System of the Mollusca. (Communicated by B. B. Woodward, F.L.S.) (Plate Laie se oe Rava Slee sadestorersiotstetehe token te eee eee 179-186 Havinranp, Grorcr Darsy, M.A., M.B., F.L.8. Observations on Termites; with Descriptions of New Species. (Plates 22-25)» .. i. Geaccavay arse erelebendtelapeiey stances cana ements Page Howarp, L. O., Ph.D., Entomologist to the U.S. Department of Agriculture. On the Chalcidide of the Island of Grenada, B.W.I. (Communi- ‘cated by F. DuCane Godman, F.R.S., F.L.S., on behalf of the Committee for Investigating the Flora and Fauna of the Wiest=Inditan@liclam da.) iv dcpee acih ac task sea cesnt = ohn tohaue elem 129-178 Hurron, Captain FrepericKk Wo.uaston, F.R:S. The Problem of Utility. (Communicated by Alfred Newton, AILS) ecteatonyears wets adores at deat accherne, ole sagtel tion ar eathalens 330-334 Jamrson, H. LysteEr, B.A. On a probable Case of Protective Coloration in the House- Mouse (Mus musculus, Linn.), (From the Biological Labo- ratory, Royal College of Science, London. Communicated by; Prot. G. BY Howes, \See:u:S:) (Plate 30) 720... ... 465-473 JOHNSTONE, JAMES, Fisheries Assistant, University College, Liver- pool. The Thymus in the Marsupials. (From the Zoological Labora- tory, R. Coll. Sci. Lond. Communicated by Prof. G. B. Howes, ARMS eC ils!) mu dalatesesu=39)) se yinniy 2h snare 587-557 Jones, T. Rupert, F.R.S., F.G.8., and F. Cuapman, A.LS., F.R.MS. Onthe Fistulose Polymorphine, and on the Genus Ramulina. 334-354 Lussock, The Rt. Hon. Sir Joun, Bart., M.P., F.R.S., D.C.L., &e. On some Spitzbergen Collembola ..................+ 616-619 Micuasr., ALBERT D., F.LS., F.Z.S., F.R.M.S., &e. Report on the Acari collected by Mr. H. Fisher, Naturalist of the Jackson-Harmsworth Polar Expedition, at Cape Flora, Northbrook Island, Franz-Josef Archipelago,in 1896. (Plate ZL) ee MED Al Nat Sey oat ee el NL SETI Ao ee 800-001 Mivart, Sr. Grorer, F.R.S., F.L.S. INOWas On own LOWES Wespoesesssecnaboscbucontoc nur 620-622 PrIcKARD-CAMBRIDGE. (See CAMBRIDGE, Rey. O, Pickarp.) vi Page Poviton, Enwarp B., M.A., F.R.S., Hope Professor of Zoology in the University of Oxford. Natural Selection the Cause of Mimetic Resemblance and Common Warning Colours. (Plates 40-44) ........ 558-612 RipEwoop, W. G., B.Sc., F.L.S., F.Z.8., Lecturer on Biology at St. Mary’s Hospital Medical School, London. On the Structure and Development of the Hyobranchial Skeleton and Larynx in Xenopus and Pipa; with Remarks on the Affinities of the Aglossa. (Plates 8-11) .............. 58-128 On the Larval Hyobranchial Skeleton of the Anurous Batra- chians, with Special Reference to the Axial Parts. (Plate SUD ep Gee RM e ak tia che me te a nat React MT UME, Cae 474-487 Waker, ALFRED O., F.L.S. On some new Species of Edriophthalma from the Irish Seas, (Geb Relea by ea? U3) eee ene HOS Mideeemctclen Ns Ao 4 6 6 226-232 Waters, AnTHUR W1LLIAM, F.L.S. Notes on Bryozoa from Rapallo and other Mediterranean Localities—chiefly Cellulariide. (Plates1&2) ........ 1-21 Observations on Membraniporide. (Plates 47-49) .... 654-€93 Watson, Rev. Roprert Booe, LL.D., F.R.S.E., F.L:S. On the Marine Mollusca of Madeira; with Descriptions of Thirty-five new Species, and an Index-List of all the known Sea-dwelling Species of that Island. (Plates 19 & 20). 233-329 West, G.S., A.R.C.S.Lond., Scholar of St. John’s Coll., Cambridge. On the Histology of the Salivary, Buccal, and Harderian Glands of the Colubride, with Notes on their Tooth-succession and the Relationships of the Poison-duct. (Communicated by Prof. G. B. Howes, F.R.S., Sec.L.8.) (Plates 34 & 35.) 517-526 EXPLANATION OF THEH PLATES. ie} is) > ie! Ei MEDITERRANEAN Bryozoa. Hotornurians of New Zealand. XENOPUS LA&VIS. PIPA AMERICANA. SOaNoaORPoNe ae ee ae d 11. | ENOPUS and Pypa. 12. Nervous Sysrem or Moruusea. 13. ASTEROPSIS IMPERIALIS. a New Zearanp HcuinopErms. New Species or Pertcuxtra—New Britain. = 20. New Mo.uuvsca from Madeira. 21. EHEryrurzus Harmsworrut, n. sp., A. D. Michael. 22 Nest or Trrmires, species allied to 7. nemorosus, n.sp., Sarawak, Santubong. 16} aes 3 18. \ New Species or EprrorutaaALMA—lIrish Seas. 0. | |) TALAYAN and South African Termites. 7) PonTOBOLBOS MANAARENSIS. HApDONIA TORRESIENSIS, n, gen. & sp., Chapman. a CaAUDINA CoRIACEA, Hutton. 30. Protrctive Cotoration in the House-Mouse, 31. Hyon of Larval Anura. 32. Porus GeniTauis in the Myxinide. 33. Auveopora and the Madreporarian Skeleton. 34. 35. 36. New Zresauanp ACTINIARIA, | GLANDS OF COLUBRID. Wu PLATE 37. 38. | Neck Guanps of Marsupialia. 39. 40. Mimeric Resevsiances in Lepidoptera. 41. Mrseric Resevprancss in Insects other than Lepidoytera. 42. ResmmBiancss in Tropical American Lepidoptera. 43. ResemBiances in Jthomiine and Danaine. 44, ResemMBiances in Piering and Moths. 45. Arctic Sprpers. 46. Payropopa from Franz-Josef Land. 47. | 48. | MempBranrvoripD2. 49, J ERRATA. Page 328, line 6 from bottom, Hulima Sta/oi, read HL. Stalioi, Brus. 654, line 20 from bottom, Zremapora, read Tremopora, Ortm. 665, line 6 from bottom, Membranipora hydasi, Jullien, read M. Hyadest, Jullien. 681, line 20 from bottom, “Tremopora dendracantha, Ortmann, Die Japanische Bryozoenfauna,’ read Tremopora dendracantha, Ortmann, Die Japanische Bryozoenfauna, Arch. f. Naturgesch. Jahrg. 56, Band i. p. 29, Taf. ii. fig. 6 (1890). 3) ” ” THE JOURNAL OF THE LINNEAN SOCIETY. Notes on Bryozoa from Rapallo and other Mediterranean Localities.—Chiefly Cellulariide. By ArrHurR Wrtrtas Waters, F.L.S. [Read 5th November, 1896.] (Puatzs 1 & 2.) AtrHoueH now at work upon material collected in Rapallo, near Genoa, it has seemed better to give this paper a wider range, so as to be able to include any observations resulting from com- parison with my Mediterranean and other collections. In fact, it may be said to include a revision of my paper * on the Naples Bryozoa. General questions of structure and classification have received considerable attention, but are only dealt with where there seemed sufficient reason for doing so. When I wrote upon the Naples Bryozoa, a very large number of the descriptions, even by the leading authorities, were most unsatisfactory ; nor did I possess my present large collection, including preparations made in various ways}, so that only limited comparisons could be made. Since then the student has had for reference, besides a large amount of general literature, Hincks’s well-illustrated work on the British Marine Polyzoa, and * Ann. & Mag. Nat. Hist. ser. 5, vol. iii., 1879. + In all cases where I have had suitable material, stained sections have been cut, either for minute structure or, when the state of preservation has not been suitable for anything more, for the interzocecial connexions; nor can any series be considered complet until there are stained specimens mounted whole, as well as preparations of the opercula and mandible, besides calcareous. sections. LINN. JOURN.—ZOOLOGY, VOL. XXVI. 1 2 MR. A. W. WATERS ON BRYOZOA FROM RAPALLO also two papers by the same author on the “ Polyzoa of the Adriatic” *. In the paper above-mentioned I showed the value of the oper- culum, and afterwards of the avicularian mandible, for classifi- catory purposes. These are now being examined by all workers ; and in various papers, and especially in one recently read before the Microscopical Society, I have also shown that the rosette- plates are another important character; and now in this paper stress is laid upon noting the position of the radical fibre. Therefore, with much extended means of study, it is not surprising that, upon re-examining my Naples material and collection, points should turn up which were overlooked or not recorded. The greater part of the present communication relates to the Cellulariide. Most members of this family throw out corneous tubular radicals; and the position from which these grow is, as a rule, fairly constant in each species, so that it may be used as a specific character, or even in some may have generic importance. This, however, has been neglected, and has seldom been fully and correctly described. Also in other families more use may be made of the posztzon of the origin of these tubes, by which the colony becomes firmly attached. In the description of the species, it is now shown that in the Cellulariide there are two distinct kinds of articulation. In the larger number the new branch is given off from a smail chamber formed for the purpose. As the type of this section, Menipea Buskii is figured (Pl. 1. fig. 10); and I propose to restrict Menipea to those forms having this kind of articulation ; and it will then include IZ. Buskii, MacG., MI. erystallina, Gray, M. cy- - athus, Thompson, I. cervicornis, MacG., MW. compacta, MacG. On the other hand, probably *M. cirrata, Lamx., M. gracilis, Busk, I. patagonica, Busk, Wl. funiculata, MacG., WV. triseriata, Busk, WM. flabellum, L., M. ternata, Ell. & Sol., must, on this account, be elsewhere located ; and in fact, before noting this distinction, it had been felt that several species should be removed from the genus. Tn another section, including Serupocellaria, the jointing con- sists of nothing more than a partial breaking through or thinning * Anrv. & Mag. Nat. Hist. ser. 5, vol. xvii. p. 254, and vol. xix. p. 302. + M. cirrata and M, flabellum are Lamouroux’s types; but these, I think, will {all into other established genera. AND OTHER MEDITERRANEAN LOCALITIES. 3 of the walls of the zoccia near the commencement of the branch. In the zocecia in which this breaking through of the wall of the zocecial chamber has commenced, the polypide is seen unaffected, partly above and partly below this incipient division (see Pl. 1. figs. 11, 12). Menipea and Bugula are evidently very closely related ; for a specimen from Florida, sent to me as “Bugula?” by Miss Jelly, has zocecia with an area, just as in Bugula, and a sessile avicularium similar in shape and position to that occurring in B. avicularia, but the branches are articulated, with three zocecia to an internode, as in Menipea. The jointing kere occurs by breaking through or thinning of the walls without any chamber being formed ; the younger parts of the branch, however, show no trace of articulation. The Cellulariide are still in a state of confusion, and the genera are based upon very unsatisfactory characters. Smitt, recognizing this, united Cellularia, Scrupocellaria, and Menipea as Cellularia; but Busk and Hincks have not followed him in this. The retention of the name Cellularia has long seemed most undesirable, as it has been used in most various senses, and has included forms which are now placed under widely separated genera. Pallas, who gave the name Cellularia, had under it Tubucellaria, Cellaria, Notamia, Bugula, Scrupocellaria, Eucratea, Aetea, and others. With the exception of Cellularia cuspidata and C. Peachit (both of which should be placed with Scrupocellaria), the only other remaining species of Cellularia are from the ‘ Challenger ;’ but I think if we turn to Mr. Busk’s definition of the genus in his Report we must consider it as an admission, on his part, of inability to find characters upon which to base it ; for ““7carium bi- or tri-serial with more than four cells in each inter- node; with or without a sessile avicularium behind the upper and outer angle; with or without a pedunculate fornix,” does not contain a single definite statement. Then Hincks in his description says, ‘“ avicularia and vibracula usually wanting ;’’ but of six ‘ Challenger’ species, five have avicularia. However, since Hincks wrote this the presence or absence of avicularia has been shown in many families to be without value in generic classification. MacGillivray writes, “ zocecia biserial . . . avicularia usually absent.” As Busk in his ‘ Challenger’ Report neither figures nor men- tions the articulation when describing his species, I do not feel 1* 4 MR. A. W. WATERS ON BRYOZOA FROM RAPALLO able to discuss the position of his species without further exami- nation, but think most will fall under Serwpocellaria, though Cellularia cirrata and C. qguadrata may have to be placed in another family. If a name which has been used in such various ways can now be buried for ever, it will be a great gain, and it now seems quite superfluous. Among other points mentioned in the present communication is the existence of ovicells on the erect tube of Aetea, a genus: which has always been described as without ovicells. The stalk of Chlidonia Cordier exhibits a very curious struc- ture ; for the central parenchym-cord gives off a branch to a disk. about the middle of each internode. In Bugula plumosa there are cases of long tubes starting from the lower part of the colony, having much the same appearance as the radicals, but at the end they produce a polypide, and in this way a new colony may arise. Although sections have been cut wherever there was suitable material, the gland-like bodies have not been found in any of the groups under consideration; and I may now mention that nearly all the truly calcareous species have these glands, whereas none have been found in the corneous ones. Before proceeding to the special descriptions, | must express: my warm thanks to the owners of the small zoological laboratory in Rapallo (Professors L. Camerano, M. G. Paracca, and D. Rosa of Turin), for kindly placing it at my disposal during my three weeks’ stay in Rapallo in March 18938; and specially thank Dr. Paracca for the trouble he took in seeing that arrangements were made. This was rather too early for satisfactory work, as. a large number of the specimens brought in were not living, or were in an unsatisfactory condition for showing minute structure.. This, however, was not the case with all; and I was enabled to see the gland-like body in a large number of species, and also- to study the interzoccial connexion injmany. The list of species collected some years before from the sea-- weeds thrown up on the beach was almost doubled, as the dredge brought up specimens which would easily be overlooked among the seaweed. The ground is mostly mud, and the neighbourhood cannot be called favourable for Bryozoa. The best places for dredging are where the mud ceases, about two or three miles out;. and probably a better locality for the larger coral-like forms such as Porella cervicornis, Hornera, would be near Camoglli,. AND OTHER MEDITERRANEAN LOCALITIES. 5 where the depth is much greater but, not being as protected, the sea will often be more agitated. AETEA ANGUINA, forma REcTA, Hincks. (Pl. 1. figs. 1-5.) In Rapallo I saw, in March, a large number of specimens with a small transparent ovicell, at the top of the tubular prolon- gation which has been called a peristome. In most cases there were four divisions in the ovum, but in a few there were eight. ‘The ovicellular wall is very thin and delicate ; and unfortunately the material at command has not enabled me to study the ovicell and ova as fully as I hoped. The discovery of an ovicell is extremely interesting, as the genus has always been described as without one; in fact, Jullien * created a ‘“ tribus”’ Inovicellata for the Aeteide; but classification based upon the absence of a character is always rather risky. What Mr. Hincks calls the zocecium in this family, Jullien (loc. cit. p. 25) would call the peristome. Now, while Jullien is correct in refusing to call this tubular prolongation a zocecium, it does not seem that this tube, which has a hinged opening at the termination, should be compared with a peristome, which is a prolongation beyond the opercular opening. Jullien seems to fancy that he was the first to notice that the polypide did not live entirely in this part, but was also in the portion which Mr. Hincks speaks of as the “ elongate subfusiform body.” On reference to Smitt’s figures +, and to my figure and the text of the same paper ¢, he would have seen that it had been appreciated that the polypide is not only lodged in the upright portion, as Busk erroneously believed, but is also in the lower part. I think that both Hincks and Jullien must have overlooked my paper. In one case in which the ovicell contained a large unfissured ovum, the tentacles and stomach were mostly in the tubular prolongation, and the fusiform body was nearly filled up with the ovarium in which there were at least four young ova. Aetea truncata also occurs near Rapallo §. * “Mission du Cap Horn,’ p. 23. + Hafs-Bry. Utv. 1865, pl. ii. figs. 11, 12. + Ann. & Mag. Nat. Hist. ser. 5, vol. iii. p. 114, pl. xv. fig. 7. § Ihave to add to the list given in my paper on the “ Bryozoa of Naples,’ -and a supplementary one given in a note to a paper in the ‘ Transactions of the Microscopical Society,’ ser. 2, vol. v. p. 6, the following Neapolitan species :— Aetea truncata, Landsborough; Bugula spicata, Hincks; B. ditrupe, Busk; B. neritina, L.; Scrupocellaria Bertholleti, Aud.; S. Delilit, Aud.; S. in- 6 MR. A. W. WATERS ON BRYOZOA FROM RAPALLO ScRUPOCELLARIA REPTANS, Z., var. BERTHOLLETII, Awd. (Pl. 1. figs. 18 & 19.) : Acamarchis Bertholletii, Audouin in Savigny’s Egypte, pl. xi. fig. 3. Serupocellaria Bertholleti, Hincks, Ann. Mag. Nat. Hist. ser. 5, vol. xvii- p: 258, pl. ix. figs. 1-2. This only differs from S. reptans in having a less-developed. scutum, for the zocecial characters and the shape is the same in both, and in the type and the variety there are three or four external spines ; the chambers for the avicularia and the radical fibres are exactly similar, as is also the ovicell and the terminal disks of the radical. At the bifurcation, in both, there is a. single vibraculum. To return to the scutum: this is very variable, and I have specimens where some zocecia have no scutum; in others it is small bifurcate, then again some are larger bifurcate or cervi- corne, while in S. reptans occasionally the scutum is only bifureate, other zocecia showing gradations to the fully-developed characteristic seutum. When previously working on the Naples Bryozoa I came upon this, and made a note of it as a variety of S. reptans, with but slightly-developed scutum, but do not seem to have considered it worth describing. Both forms occur at Naples, Capri, Rapallo, and Trieste, and the variety probably came from Egypt. Jullien * says that he has obtained Scrupocellaria reptans with lateral avicularia; but in the specimens of both the type and the variety in my collection, from various places, this seems to be common; and Hincks, in his ‘ British Marine Polyzoa,’ had previously mentioned this character, which had been overlooked by older writers. On account of these lateral avicularia Hincks had removed it from Canda, where it had been placed by Busk. SS. reptans has two lateral rosette-plates, and Levinsen says that scruposa and scabra have also two. S. Delilii and obtecta have the same number, while in several species of Bugula there are four lateral plates. In the Cellu- lariide this character would seem to be of generic rather than of specific value; whereas we have seen that in Flustra the specific curvata, Waters; Beania hirtissima, var. robusta, Hincks, and var. cylindrica, H.; Schizoporella armata, Hincks; 8S. magnifica, Hincks ; S. marsupifera, B. ; S. ambita, Waters; Hornera lichenoides, Pontop. ; Entalophora clavata, Busk ; Buskia socialis, Hincks; Retepora complanata, Waters; and Palmicellaria parallelata, Waters. * Mission du Cap Horn, p. 69. AND OTHER MEDITERRANEAN LOCALITIES. 7 differences in the number and form of the rosette-plates are very considerable *. Of the seven Mediterranean species of Scrupocellaria, all except 8. incwrvata and the variety Bertholletii occur elsewhere. S. reptans, seruposa, scrupea, and inermis are Northern, while S. reptans and scrupea are also found in the Australian seas. ScRUPOCELLARIA Deitit, Awd. (PI. 1. figs. 14, 15.) Crisia Delilii, Audouin in Savigny’s ‘ Description de Egypte,’ p. 242, pl. xu. fig. 3. Scrupocellaria Delilii, Busk (non Alder), Q. J. Micro. Sc. vol. vi. p. 65, pl. xxu. figs. 1-3. Specimens from Naples, Rapallo, and the Gulf of Taranto belong most undoubtedly to the species figured by Savigny. The two converging spines are characteristic. Most colonies have no frontal avicularium, and in those that have it is only found on some zoccia. In one good colony there is only one, and that on a zocecium just below a bifurcation. The zocecium at each bifurcation has a central spine, a character which Mr. Hincks mentions in S. scabra, but which seems to be fairly common in the genus, as it occurs in NS. serupea, S. Bertholleti, and S. reptans. On the dorsal surface at the bifurcation there are two vibracula, whereas in S. scabra there are none; but the most important distinction between the two species is in the vibracular chamber, which is much narrower and longer than that of S. scabra, and the rooting attachment is more directly under the vibraculum. It will thus be seen that 8. scabra remains a Northern form not yet found south of the British Isles, while S. Dedcliz is Southern, ovceurring in the Mediterranean and Madeira. The internodes are much longer than those of the Mediterranean S. scrupea. This is closely related to 8. Macandrei, but from the figures in the ‘ Challenger’ Report they would seem to be distinct. SCRUPOCELLARIA SCRUPEA, Busk. There are two vibracula at a bifurcation similar to those which Levinsen has described as occurring in 8. seruposa. The only other species in which I have found two are S. ornithorhynchus, Delilit, and incurvata. The small avicularia, by the ovicells, are similar to those which Levinsen describes on 8. scruposa. These * “Tnterzocecial Connection in Flustra and other Bryozoa,” Journ. Roy. Micro. Soc. p. 279, pls. vii. & viii. (1896). 8 : MR. A. We WATERS ON BRYOZOA FROM RAPALLO had previously been overlooked, as they are very minute, and being on the inner side of the ovicell are in most positions of the zoarium concealed ; however, when looked for, they can generally be found upon Mediterranean specimens. The shape of the vibracular chamber and of the chamber for the insertion of the rooting-fibre is the same in S. serupea and scruposa, and these species are closely allied. It may be men- tioned that, while the presence of the scutum is a distinguishing character, it does not occur in all the zoecia. ScRUPOCELLARIA INERMIS, Vorman. (PI. 1. figs. 11 & 12.) Scrupocellaria inermis, Norman, Quart. Journ. Meer. Sc. n. s. vol. viii. p- 215 (4) pl. v. figs. 1-3; Kérchenpawer, Jahresb. Comm. wissensch. Unters. d. deutschen Meere, Jahrg. ii. & iii., 1875, p. 180. A fine stained and decalcified specimen, collected in Trieste, is figured, as it shows so well the way in which the articulation is formed, for in the zocecium at the bifurcation the polypide is seen on both sides of the articulation, proving that this is a break formed in mature zocecia. Although the object of the figure is to explain this structure, it will be as well to also consider the species systematically. This specimen has no median avicularia, and the lateral one is much smaller than in Norman’s figures. There are no spines and no scutum; whereas, as I have shown *, S. elliptica, Reuss, has both, and therefore the two specific names must be retained. The vibracular chamber is small, and appears to be narrower than in the specimens described by Norman, but being decalcified a complete comparison cannot be made. This chamber is much shorter and wider than that of 8. seruposa, and the chamber for the insertion of the radical is distinctly separated, both chambers being about the same width. The vibraculum is very delicate, but I should scarcely call it short. At a bifurcation there is only one vibraculum ; thus it is distinguished from S. scruposa, in which there are two. Pergens says t that S. scruposa form elliptica is found at Palavas (Dépt. de la Hérault) and at Banyuls. Presumably it is this species ; but as there has been confusion between SN. elliptica, Reuss, and S. inermis, Norm., we cannot be certain without a re-examination. * “North Italian Bryozoa,” Quart. Journ. Geol. Soc. vol. xlvii. p. 6, pl. i. figs. 16, 17. + “Notes sur les Bryozoaires,” Ann. Soc. Roy. Malac. Belgique, vol. xxiv. p. 7. AND OTHER MEDITERRANEAN LOCALITIES. ; 9 ScRUPOCELLARIA INCURVATA, sp. nov. (Pl. 1. figs. 16, 17.) Specimens from Naples in most respects resemble 8. Delilit, but the branches of the zoarium are stouter, with a much larger scutum extending beyond the opesium. There is usually a spine at each upper corner converging as in S. Delilit, Aud. Also the median avicularia are similar to those of S. Delilii and the ovicell is entire; but the two can be distinguished by the vibracular chambers, for in the present species they are somewhat oval, extending half across the zocecium, with the groove turned inwards, and the pair of vibracular chambers at a bifurcation have similar curved grooves. It will be seen that this curved groove, though not as large, is similar to that of Caberea, and, except on account of the difference in the oral aperture, it is difficult to see any reason for separating the two genera. S. scabra, Deliliz, and incurvata seem to be a natural group, distinguished, however, by striking differences in the vibracular chambers. In the whole of the Cellulariide one of the most useful specific characters is the way in which the radical fibre is attached, for in each species there seems nearly always to be some definite position from which it starts in all zoccia. In Scrupocellaria and Caberea it grows from a chamber at the base or side of the vibracular chamber. In Menipea aculeata, M. Buskit, M. cirrata, and WM. funiculata the radical arises from the base of the internode ; whereas in I. cervicornis, M. compacta, and M. ternata var. gracilis, it arises from the top. It is, however, not only in the Cellulariide that the position of the radical is a useful character, but in the description of all rooting forms it should be mentioned. CaBEREA Bory, Aud. The oval body to which I refer in my paper* on gland-like bodies (p. 277) occurs constantly in the Rapallo specimens. It is rather smailer than in those from Trieste, and stains deeply and uniformly. There is one very large avicularium below the bifurcation. This is apparent in the Rapallo, Naples, and Roscoff specimens, and also in one from Adelaide which I consider to be C. Boryi. It appears to have been overlooked in this species, but somewhat similar avicularia in the same position have been figured in C. rostrata, B. * Journ. Linn. Soc., Zool. vol. xxiv. p. 272. 10 MR. A. W. WATERS ON BRYOZOA FROM RAPALLO C. Darwinti, Busk, dredged by the ‘Challenger’ from Nightingale Island, is very similar to the European C. Boryz, as the vibracular chambers are identical in shape, so are the avicularia both lateral and median. The character by which the two species are distinguished is that in C. Bory: there is a cal- careous bar below the proximal border of the oral aperture, and to this the upper part of the scutum is “ soldered,” while in C. Dar- winid (Pl. 1. figs. 18, 21-25) the scutum is free and there is no bar right across the front, though sometimes it has commenced at both sides, thus showing us the early stage of C. Borys. The ‘Challenger’ specimens do not show the helicine mark on the scutum to which MacGillivray refers. The differences are so slight that I should have eailed Darwinii a variety of C. Boryz. Caberea Boryt, however, also occurs in the Southern hemi- sphere, from New South Wales, Adelaide (Victoria), and New Zealand; but in all these the lateral avicularia are somewhat larger than in the European specimens. In C. Boryi there is one vibracular chamber at the bifurcation. There are 18 tentacles. In ©. Darwinii there are two spines at the lower edge of the oral aperture and two at the distal. Bueuta pLumosa (Pallas). (Pl. 2. figs. 5-7.) See Hincks, Brit. Mar. Polyzoa, p. 84, for synonyms. Bugula plumosa, var. aperta, Hincks, Ann. § Mag. Nat. Hist. ser. 5, vol. xvii. p. 261, pl. ix. fig. 6. Bugula simplex, Hincks, loc. cit. p. 262, pl. ix. fig. 7. This seems to be subject to a certain amount of variation in the matter of spines. In some they are scarcely distinguishable, in others the outer one attains a moderate size. Again, on the inner corner of most specimens no spine can be found, in others only a protuberance, while, again, in some it is distinct. In Trieste I noticed that those collected in the sea had a fairly prominent outer spine, while those growing in the aquarium had only a protuberance; and there seems to be no difference in the colonial growth of the B. simplex form and the B. plumosa form from this locality. I have from Trieste typical specimens of both forms, but upon examining a series I am unable to separate them *. * In an undoubted B. plumosa from Roscoff, occasionally a slight prominence on the inner border may be detected. AND OTHER MEDITERRANEAN LOCALITIES. 11 From Rapallo I have. the plwmosa form with the small cap-like ovicell of the variety aperta of Hincks, and it occurs on both the forms from Trieste; but from analogy with other Bryozoa it would seem possible that the remarkably small ovicell may be a stage of growth; that is to say, the ovicells are only in the initial stage. We may in colonies of Bryozoa find the ovicells in every stage of growth; on the other hand, this is not always the case, for I have specimens of Retepora in which most zocecia show the small semicircular plate, indicating the commencement of the ovicell, though none have complete ovicells. If more abundant material should show that I am wrong in the interpretation of these points, then we have B. spicata var. aperta, B. spicata, B. plumosa var. aperta, and B. plumosa—all living side by side; but if these are to be specifically separated, it must be upon other distinctions than those at present indicated. The primary zocecium has a central spine at the base of the area *, and on each side two spines near the distal end. Itis not only the first zocecium which may possess ancestral characters, but two, three, or even more uniserially arranged may have the same shape before the normal zowcia are formed; and in this species there are sometimes near the base a number of zoccia longer than the normal ones, but with the area very short. The repetition of the ancestral characters in the few first zocecia we may see in B. avicularia, B. calathus, and B. neritina; and further, in B. plumosa from a single zocecium zocecia may spring out at each side and form new branches, but this does not seem to be a common method of increase. Another mode of increase is of great interest, and was found in colonies growing in the aquarium of the Zoological Station in Trieste—some basal zocecia producing either direct (Pl. 2. fig. 7), or from lateral zoecia (fig. 6), long processes which seem scarcely to differ from the radicals, or attachment fibres, except in being rather stouter. These processes may equal ten zocecia in length, and at the end a young polypide buds, and in some cases from * I term this “area,” for “‘ aperture ” seems eminently unsatisfactory seeing that there is no aperture, but only a thinner portion of the front wall. It might perhaps be better to term it “ opesium,” but at present a general term may be safer. ‘“ Aperture” is probably a relic from times when only dried specimens were examined and low powers used, and a real aperture was thought to exist. 12 MR. A. W. WATERS ON BRYOZOA FROM RAPALLO this there is a second bud (PI. 2. fig. 7), showing that anew colony may arise from this elongated process. This budding may occur in a colony in which the polypides are all dead, and the zoccia only contain brown bodies, as well as in colonies in full vigour. Bugula is without an operculum, but there is in the membrane of the upper part of the area a diaphragm, similar to the “‘irisoide”’ of Jullien, and through this the polypide is extruded. In not having an operculum attached by muscles, this genus would seem to be widely separated from Membranipora, but, on the other hand, the primary zoccia remind us of Membranipora. When the polypide is protruded, the front wall of the zocecium is carried with it, so that the way in which the polypide comes out cannot be compared with most other Chilostomatous genera: in fact, ib seems in this respect to have most analogy with the Ctenostomata. — On the dorsal surface, the distal wall ends lower down than on the anterior surface, and forms a sharp curve. In the above and most other species this is the only divisional mark on the dorsal surface, but in B. dentata there is another at the height of the distal end of the area. The space thus enclosed has a large opening and can therefore hardly be called a chamber, and is often filled with protoplasmic threads as well as the testes and spermatozoa. Round the edge of the distal walls close to the upper border, and near the dorsal surface, there is a row of numerous small rosette-plates. These rosette-plates and the four lateral ones seem to occur in the same position throughout the genus. They have been figured by Levinsen in ‘ Danske Dyr,’ pl. 11. figs. 5, 6. The radical tubes are thrown off from near the base of a zoccium. The fact may be mentioned for what it is worth, that as a rule the Bugule in the Northern hemisphere have the avicularia placed high up on the zoccia, whereas those in the Southern hemisphere have them near the base. B. neritina has them low down, and we may ask whether this form has been introduced from the Southern hemisphere. Bueuta pirrurm, Busk. (Pl. 2. figs. 2 & 3.) Bugula ditrupe, Busk, Q. J. Micro. Soc. vol. vi. p. 261, pl. xx. figs. 7, 8; Hincks, Ann. & Mag. Nat. Hist. ser. 5, vol. xvii. p. 260, pl. ix. figs. 3, 4. Specimens from Naples show that the “additional spine” about midway on one side of the area is not a constant character, AND OTHER MEDITERRANEAN LOCALITIES. 13: some specimens being entirely without while others have it to: some zoccia. This “additional spine” is found quite as fre- quently in the Mediterranean Bugula calathus as in B. ditrupe. Some specimens have to some zocecia three outside spines and two- inside, while other zocecia have three outside and three inside. The primary zocecium has a central spine at the base of the area, flanked by a spine on each side, and at the top of the area. a pair of spines at each corner, and in more than one specimen this primary zovecium has an avicularium; and this is the only species in which I have seen avicularia to the primary cells. The beginning of the colony grows up straight at first, then dichotomizing from a centre forms a cup. This is in most respects very similar to B. spicata, H., which 1 have biserial from Naples and Capri, while from Naples and Trieste it is 4-serial, and in these the outer avicularia are larger than the inner. The avicularia of B. spicata and B. turbinata are almost identical in shape. The B. ditrupe in my collection from Naples and Capri are: biserial, while those from Trieste are 4-serial, and in these again the outer avicularia are larger than the inner. A specimen from Rapallo, having been decalcified, does not admit of complete comparison. Bueura catatuus, Norman. (Pl. 2. figs. 4 & 10.) Bugula calathus, Norman, Q. J. Micro. Se. (n. s.) vol. viii. p. 218, pl. vi. figs. 3-8; Hincks, Brit. Mar. Polyzoa, p. 82, pl. xi. figs. 4-6; id. Ann. & Mag. Nat. Hist. ser. 5, vol. xvii. p. 260. Bugula avicularia, forma flabellata, Waters, Ann. & Mag. Nat. Hist. ser. 5, vol. i. p. 117. It bas been considered by Pergens and others that the Mediterranean form should be called Bugula calathus; and although I should not have been prepared myself to separate it from B. flabellata, yet in deference to the opinion of others I leave it as B. calathus, instead of considering it the Mediter- ranean form of flabellata. The cells, ovicells, and avicularia are smaller than those of B. flabellata from Roscoff ; whereas Hincks. says of the British B. calathus that these organs are on a larger scale than in B. flabellata. The mandible in B. flabellata is 0:16 mm. long, and in the Mediterranean B. calathus 0:12 mm. The character chiefly relied upon for the separation of the two species is the contrast in colour; but Vigelius (“ Ontogenie der Marinen Bryozoen,” Mitth.a.d. Zool. Stat. z. Neapel, vol. vi. 14 MR. A. W. WATERS ON BRYOZOA FROM RAPALLO p. 505) points out that the colour varies considerably in the Gulf of Naples according to the time when they are taken. The primary zocecium has a central spine at the base of the area, and one on each side halfway up as well as three spines at each upper corner. The three lower spines are often reduced to mere protuberances. An “additional spine” on the margin of the aperture about halfway up occurs in many zoecia of all ages ; in fact I have found it more frequent than in B. ditrupe, where it was first noticed by Hincks. The number of spines is undoubtedly a character of great diagnostic value in the Bugule, but until every character has been compared we are likely to go astray sometimes. SYNNOTUM AVICULARE, Pieper. (PI. 1. figs. 6, 7.) Gemellaria avicularis, Pieper, Jahres)... !estfilischen Provinzial- Vereins, vol. ix. p. 43, pl. il. figs. 5-9. Notamia avicularis, Waters, “ On the Use of the Av:cuiarian Mandible ”’ §c., Journ. R. Micro. Soc. ser. 2, vol. v. p. 6 (name only). Synnotum aviculare, Hincks, Ann. & Mag. Nat. Hist. ser. 5, vol. xvii. p. 257. oe Gemellaria egyptiaca, Savigny (name on plate), pl. xui. fig. 4, but Loricaria egyptiaca, Audouin, * Description de V Egypte’ (in text). Mr. Hincks has given a full description, in which he allows that the structure of the zoarium is essentially the same as in Notamia (now Epistomia), and agrees with Pieper that it may perhaps be regarded “als Verbindungs-Glied zwischen Gemellaria und Notamia.’” For my own part I should not have removed it from Epistomia (NM otamia), and still doubt whether a new genus is required, but, as that is not an important point, have put it under the genus proposed by Pieper and Hincks. There has been great confusion concerning the nomenclature of the genera Gemellaria and Notamia; and while recognizing the correctness of Gregory’s remarks * on the use of the name Gemellaria t, I feel much * “ British Paleogene Bryozoa,” Trans. Zool. Soc. vol. xiii. p. 227. + In reference to the date of publication of Savigny’s plates, on which the name “Gemellaire” first appeared, there seems a probability that they were placed in the hands of various naturalists before the text was published. I purchased a volume of plates from Fr riedlander, without text, but there was a manuscript list of plates and figures, headed ‘ Zoologie de Egypte: Iconographie des Echinodermes, des Polypes, et des Zoophytes; par Jules César de Savigny, 1806-1812.” This may have been copied from an older list! The names of all the genera, whether established or new, appear at the foot of the plates in the French form. AND OTHER MEDITERRANEAN LOCALITIES. 15 hesitation in making a change, seeing that the two names have been used by Hincks, Busk, Smitt, Vigelius, Lorenz, Freese, Winther, Levinsen, and others. As Hincks did not publish a figure, and as Pieper’s paper is not very accessible, one is now given. This is a very delicate species and may be easily overlooked. One specimen from Rapallo has the long tubular rooting-process given off from the back of the zocecium near the middle. Diploecium simplex, Kirkpatrick, has similar rooting-fibres from the centre of an internode, but this is not mentioned in the description; and as some specimens of Synnotum aviculare have few or no radicals*, I would again call attention to the frequency in the attached Bryozoa of radicular disks being formed without a chitinous tube growing from them. This occurs in Palmicel- laria parallelata and Alysidium Lafontii. The first internode in a branch has one zocecium. This is also the case in Epistomia bursaria; and in Didymia simplex similar internodes occur in the earlier branches, but not through- out the colony. Calwellia bicornis, however, has the first internode of a new branch double; whereas a specimen from Port Phillip sent to me, named +Calwellia gracilis, MacG., is Synnotum aviculare. The shape of the colony, and the zowcia, as well as the position of the radical, leaves no doubt in my mind that this is the Gemellaria egyptiaca of Savigny, and that the avicularia were overlooked, but this was often the case when their importance was less understood. Loc. Trieste; Naples; Rapallo; Portland, Victoria (MacG.); Port Phillip; S. Africa, a small specimen submitted to me in Miss Jelly’s collection. EpistomMia BURSARIA, LZ. (PL 2. figs. 8 & 9.) Notamia bursaria, Hincks, Brit. Mar. Polyzoa, p. 100, pl. iv. figs. 1-5; which see for synonyms. This, so far as I am aware, has not before been found in the Mediterranean, and I only obtained one specimen from Rapallo. The supporting stalk carries an avicularium just below the first * See “Mediterranean and New Zealand Retepore,” Linn. Soc. Journ., Zool. vol. xxv. p. 267. + MacGillivray, ‘‘ New or little-known Polyzoa,” pt. ix., Trans. Roy. Soe. Vict. vol. xxii. p. (1). 16 MR. A. W. WATERS ON BRYOZOA FROM RAPALLO internode, and this upper portion does not seem to have contained any polypide. It would thus seem that the upper part of the stem may be cut off as a barren internode, carrying a sessile avicularium identical in structure with those on the mature zocecia. The erect stem does not start direct from the creeping tube, but evidently there is a diaphragm, or rosette-plate, in the extended stolon by which stem and stolon are in connexion. BEANIA MAGELLANICA, Busk. (Pl. 2. figs. 11-14.) The small projections on the distal end of the zoccium,. described from my Naples specimens, are also seen in those from Rapallo. These Jullien did not find in the specimens from Cape Verd or from Tierra del Fuego, but my observation has been confirmed by others. The mandible of Beania magellanica and B. bilaminata, Hincks,. has a double “columella.” This I have figured (Journ. Roy. Micro. Soe. ser. 2, vol. v. pl. xiv. fig. 4). Ortman states that in the Japanese specimens the border of the avicularium is dentate, but this is not the case with those from the Mediterranean. As the large figure, given by Jullien, does not correctly show the muscles of the avicularium, I have added figures showing that besides the large adductors there is a semicircular row of short muscles, which no doubt contracts the integument behind the mandible and thus helps in the slow opening of the beak- The “cellular body” of the avicularium, as elsewhere mentioned, occurs in a sheath, the equivalent of the sheath of the polypide. The “eggs” described by Jullien occur one on each side of the zocecium, in the same position as the ovarium, contain refractive cells, and are similar in structure to what I term the “median body ” * in Schizoporella sanguinea. In my specimens of Beania magellanica they are usually globular, but sometimes. become sausage-shaped. The zocecia are often nearly filled by the testes which seem to grow from the two sides, lower down than the ovarium. The remains of the polypide become encysted, forming brown bodies.. The budding polypides originate from close to the connecting tubes, in fact sometimes seem to be almost within them. * “Observations on Gland-like Bodies in the Bryozoa,” Linn. Soc. Journ.,. Zool. vol. xxiv., 1892. AND OTHER MEDITERRANEAN LOCALITIES. 17 These connecting tubes have usually but one rosette septum, but occasionally there are two, with the parenchym threads attached to both and passing through the junction (see figs. 11 and 12.) Loe. Adriatic, Naples, Marseilles, Rapallo, Riou and Podesta (Marion), Bonifacio; Straits of Magellan, New Zealand, Port Jackson, Portland (Victoria), Mauritius, Cape Verd Is., Kerguelen Island, Japan. BEANIA MIRABILIS, Johnst. (Pl. 2. fig. 1.) In the Rapallo specimens there are normally two radicals to each zocecium, and this would seem to be generally the case. The position of the septum a short distance from the parent zocecilum appears to be constant. ‘There is no septum in the neighbourhood of the new zocecium. The septal division takes place in the stolon very soon after its formation. Although there is no ‘“‘ Kapsel,” the position of the septum may be compared with that in the stolon of Hypophorella expansa, Ehlers; namely, immediately beyond each new zoccium. BEANIA HIRTISSIMA, Heller, var. RoBUSTA, Hincks. Diachoris hirtissima, H., form robusta, Hincks, Ann. & Mag. Nat. Hist.. ser. 5, vol. vii. p. 74, pl. v. figs. 9, 9a; Waters, Journ. Micro. Soc. ser. 2, vol. v. pl. xiv. fig. 5 (mandible). This form occurs fairly abundantly at Naples, Capri, and Rapallo. It often has the avicularium showing on the dorsal surface instead of infront. There is no definite radical as in B. hirtissima var. typica, but there are often numerous fibres on the dorsal surface, which in some cases look like spines, in others more resemble the rooting-fibres. Ina specimen from New Zealand the distal spines are rather smaller than in the Mediterranean specimens, but both these and the marginal spines correspond in structure and position. In the New Zealand specimen the ovicell is larger than in those from Naples, with the mandible nearly twice as wide and with the distal end round. Hincks described this variety from Algiers. Beant Hirtissima, Heller, typica. Diachoris hirtissimia, Heller, Bry. Adriat. Meeres, p. 94, pl. i. figs. 6, 7. Lhave characteristic specimens from Rapallo. The usual arma- ture may be taken as 7 stout spines round the oral aperture close LINN. JOURN.— ZOOLOGY, VOL. XXVI. 2 i8 MR. A. W. WATERS ON BRYOZOA FROM RAPALLO up to the border, and six others of the same size farther removed, making an irregular second row. On the lateral margins of the zocecia there are about 10 delicate spines, directed inwards towards the median line, and 4-6 directed outwards. Dorsal surface with numerous spines, and sometimes, but not always, a radical tube starting from close to the proximal end. In Beania magel- lanica this tube is almost central. The Chaunosia hirtissima of Busk would seem to be the var. cylindrica, and I am not sure, therefore, which variety was dredged by the ‘ Challenger’ at Cape Verd Islands. Cuiiponta Corpieri, 4ud. (Pl. 1. figs. 8 & 9.) Eucratea Cordieri, Aud. in Savigny’s ‘ Egypte,’ p. 74; Waters, Bry. of the Bay of Naples, Ann. Mag. Nat. Hist. ser. 5, vol. iii. p. 116, pl. xv. figs. 9, 11, pl. xxiii. fig. 12. Cothurnicella deedala, W. Thomson, Proc. Zool. Bot. Assoc. Dublin, vol. i. p. 85, pl. vin. figs. 3-5, & Nat. Hist. Rev. vol. v. p. 146. Chlidonia deedala, MacGilliv. Zool. Vict. dec. xi. p. 35, pl. eviil. fig. 2. Chlidonia Cordieri, Savigny’s‘ Egypte,’ pl. xiii. fig. 3 (in tert Eucratea) ; D?Orbigny, Paléont. France. p. 40; Busk, ‘ Challenger’ Report, p. 8, pl. xxvii. fig. 11; Hincks, Ann. Mag. Nat. Hist. ser. 5, vol. xvii. p. 258; MacGillivray, Cat. Mar. Polyz. Vict. p. 10. The stalk, consisting of calcareous cylinders connected by corneous tubes, has a parenchym running through it, which in each cylinder sends out a branch to an external disk (see fig. 8). There are three other stalked genera, namely Rhabdopleura, Kinetoskia, and Stirparia, but in none of them, so far as can be judged from the descriptions, do the stalks resemble those of Chlidonia. A fuller knowledge of the earlier stages of these stalked species is much wanted, for of course these stalks are quite different from the radicals thrown out by many species of Bryozoa. In none of the species with stolons and stalks do we know the form of the primary zoccium. The stalk of Stirparia glabra (see opposite) has two interior chitinous cords (woodcut fig. 3), which are formed by an internal thickening of the chitinous wall, and the structure of the stalk in no way resembles that of Chlidonia. Rhabdozoum can scarcely be compared with the other stalked genera, as from the front of the zoarium a plain pedicular tube is given off, from which a fresh branch grows. In the stalked specimens of Rhabdozowm that I have examined many stalks grow from one stolon. AND OTHER MEDITERRANEAN LOCALITIES. 19 Epicaulidium has somewhat similar internodes. Busk in the ‘Challenger’ Report makes the zocecium of Chlidonia bicamerate with no connexion between the chambers, but sections show that there is no division. A very considerable thickening of the shell in front (PI. 1. fig. 9) contracts the zowcial chamber, but nowhere does there seem to be a second chamber. At the distal end there is certainly a small portion cut off, into which the parenchym passes, through pores of communication, then reaching the next zoccium after traversing the horny internode. Alysidium parasiticum, Busk, is rooted by corneous tubes, which sometimes have calcareous nodes, from which the stalk of a colony grows. Loc. Naples, Trieste, Rapallo, Algiers, Nice, Tunis, Egypt, Tyre, Calvados, Victoria (Australia), Cape York, New Zealand; Atlantic (jide Carus). On the stalk of Stirparia glabra there are capsules look- ing like the gonophores of Hydrozoa. These were a great puzzle; and I was unable to elucidate them until Mr. Kirk- patrick afforded me the opportunity of examining the British Museum specimens of Stirparia. However, Hincks mentions (Ann. Mag. Nat. Hist. ser. 5, vol. xi. p. 105) the radical tubes and the way in which they pass from one internode to another; and these radicals in the ‘ Challenger’ 8. glabra, from Bahia, are just below the internode, though sometimes passing from one inter- node to another, forming a knot of chitinous tubes, much like those occurring in Cellaria. But in an unnamed species from Kurrachee there are similar structures, which sometimes become very wide or club-shaped at the end ofa short or long stalk-like tube (woodcut fig. 1). In one piece.they are abundant, and one joint has as many as 17 radical tubes, some of which form capsules. Though some are torn or burst, there seem no definite opening and no contained organs, but at the apex there is an accumulation of protoplasm, such as is constantly seen in the tips of growing radicals. It is thus seen that the radicals occur in specimens from Bahia, West Australia, and Kurrachee, and that in those from the last two localities they may take the form of a capsule ; but the object of this modification is not clear. 9% 20 MR. A. W. WATERS ON BRYOZOA FROM RAPALLO Fig. 1. Stalk of Stirparia from Kurrachee, showing one branch and numerous- radicals, some of which are club-shaped. Fig. 2. Section of stalk (0) of Stirparia glabra, Hincks, from West Australia,. showing origin of a radical (a). Fig. 3. Transverse section through the stalk of Stirparia glabra, cutting through the chitinous cords formed by an internal thickening of the chitinous wall. EXPLANATION OF THE PLATES. Puate 1. Fig. 1. Aetea recta, Hincks ; showing ovicells. 25. 2, 3. Do.; ovicells. x 89. 4, Do. The ovarium fills up most of the lower portion of the zocecium,. and the tubular prolongation which is bent back carries an ovicell.. x 85. 5. Do.; showing the position of the polypide in the zoecium. X 85. 6. Synnotum aviculare, Pieper ; showing the lateral sessile avicularia. xX 25. 7. Do.; showing the large pedunculate avicularium. x 85. 8. Chlidonia Cordieri, Aud. Stalk showing disk, to which the parenchym: is attached. x 25. 9. Do. Zoccium, showing the position occupied by the polypide, and also the small separate chamber from which the next zocecium starts. x 85. an imp. # Foor Vou AVL 1 West, Newn2 Bt ve WN.50c. Jour iy a mad hah. SEUID) seoly ER e EVANINUEE CAINS 1 Ee Oi OA M nen te ne AW Waters del. _ AT. Hollie ith. 2. , oc. JOURN Maser Vou Scone co : 8 ia & es Temp. West, Newmnaat / ME DITHMRRANBAN BRYOZOA. Fig. Fig. 10. ql, 12. 13. 14. 15. 16. 17. 18, 20, 21. 22. 23, 25. I or 9 bo AND OTHER MEDITERRANEAN LOCALITIES. of Menipea Buskii, W. Thomson. Dorsal surface showing the two small chambers a and 6, from which the next zoecium starts. 25. Scrupocellaria inermis, Norman. Dorsal surface; transparent pre- paration showing the method of articulation, and the position of the polypide in the articulating joint. The vibracular chambers, and the chambers from which the radical starts are also shown. Do. Anterior surface, showing the position of the polypide at the articulation. Caberea Darwinii, Busk; Station 135, ‘Challenger.’ Lateral surface. X 25. Scrupocellaria Delilit, Aud. Dorsal surface. x 25. Do. Anterior surface. x 25. Scrupocellaria incurvata, Waters. Anterior surface. X 25. Do. Dorsal surface. x 25. 19. Serupocellaria Bertholletii, Aud. Fornices from one colony, X25. Scrupocellaria reptans, L. Fornix. x 50. Caberea Darwinii, Busk. Operculum. xX 85. Do. Avicularian mandible. x 250. 24. Do. Ova out of the ovicells. x 85. Do. Anterior surface, 265. Puate 2. Beania mirabilis, Johnston; showing the two radical processes, and also the diaphragm in the stolon near to the zoecium from which it has arisen. X 25, . Bugula ditrupe, Busk. Primary zoecium. x 85. Do. The same, showing the commencement of the colony. x 25. . Bugula calathus, Norm. ‘ Primary” and second zoecium. x 25. . Buguia plumosa, Pallas. Young zocecium from the extremity of a long basal process. X85. Do. Samecolony. x 12. Do. Process showing first and second zoccium. . Epistomia bursaria, L. Stalk and lower part of colony. x 25, Do. Origin of stalk from the creeping stolon. x 85. Bugula calathus, Norman. “Primary” zocecium. X 25. . Beania magellanica, Busk ; showing position of egg-masses. X 25. . Do. Connecting-tube with two septa. x 150. . Do. Egg-mass. x 250. . Do. Avicularia. x 85. . Bugula spicata, Hincks. Mandible. x 250, . Bugula ditrupe, Busk. Mandible. x 250. 22 PROF. A. DENDY ON THE Observations on the Holothurians of New Zealand; with Descriptions of four New Species, and an Appendix on the Development of the Wheels in Chirodota. By ArruuR Denpy, D.Se., F.L.S., Professor of Biology in the Canter- bury College, University of New Zealand. [Read 5th November, 1896.] (Puatus 3-7.) 1. IntRoDUcTORY REMARKS. My attention was first directed to the subject of this communi- cation shortly after my arrival in New Zealand, by the receipt of a remarkable specimen collected on the New Brighton (Christ- church) beach by Mr. R. M. Laing, M.A. I shortly afterwards received a second specimen of the same species from Mr. J. P. Grossmann, M.A., of which I made a detailed anatomical examina- tion. The species proved to be a new Colochirus, covered with overlapping scales, and is described in the context as Colochirus ocnotdes. Early in the course of this investigation, I was greatly struck with the inadequacy of our knowledge of the New Zealand Holothurians, and especially with the remarks to that effect made by Théel in his ‘ Challenger’ Report. I therefore determined to attempt a revision of the group, and proceeded to Wellington with a view to examining the type specimens in the Colonial Museum. Thanks to the kindness of Sir James Hector, I was enabled to do this, so far as the present condition of the specimens allowed. I found there specimens, apparently the original types, of “ Synapta uncinata,” “ Synapta inequalis,” “‘ Molpadia coriacea,’ “ Chirodota? alba,” “Thyone brevidentis,’ and “ Thyone longidentis,”’ described by Hutton in 1872 in his ‘ Catalogue of the Echinodermata of New Zealand.’ I unfortunately found these specimens in a very bad state of preservation, owing to the short-handedness of the Museum staff, but they were still of the greatest service in determining questions of synonymy. Captain Hutton himself very kindly placed at my disposal type-slides of the spicules of his “‘ Molpadia coriacea,” “* Chirodota alba,” “ Thyone caudata,” and “ Holothuria mollis,” all marked Wellington.” I also applied to Professor T. J. Parker for material from the HOLOTHURIANS OF NEW ZEALAND. 23 Dunedin Museum, and am greatly indebted to him for named specimens of “ Thyone caudata,” ‘ Hchinocucumis alba,” and especially his own “ Chirodota dunedinensis,” as well as for a new species of Psolws collected at the Macquarie Islands. Unfortunately the type specimens of Hutton’s “ Cucumaria Thomsont,” “Labidodesmus turbinatus,” and “Holothuria Robsoni’” had been sent from Dunedin to the Colonial and Indian Exhibition in London, and thence presumably to the British Museum ; so that I. am unable to add anything to our knowledge of these species, and can only hope that they will be re-examined by the Museum authorities. I am also indebted both to Captain Hutton and Professor Parker for several additional specimens of that remarkable Holothurian Caudina (Molpadia) coriacea, and to Captain Hutton again for the loan of specimens of Colochirus ocnoides, n. sp., and Cucumaria Huttoni, n. sp., from the Canterbury Museum. Mr. H. Farquhar, of Wellington, very kindly gave me another specimen of “ Hchinocucumis alba,’ dredged in Wellington Harbour, and allowed me to make use of a manuscript Catalogue ot New Zealand Hchinodermata compiled by himself. While I was staying at Wellington, my friend Mr. H. B. Kirk took me to some of his favourite collecting-grounds in Cook Straits, where we secured living specimens of Stichopus mollis, Hutton sp., and Colochirus calcarea, n. sp. Mr. Kirk also sent me a further supply of these species to Christchurch. To all these gentlemen I wish to express my very sincere thanks. The only other Holothurians known from New Zealand are those collected by the ‘ Challenger’ Expedition and described by Théel. I have of course been unable to examine the types of these, but the full descriptions and figures given in the ‘ Chal- lenger’ Report render this unnecessary. Indeed I ought also to express my indebtedness to the admirable monographic account of the group given in the ‘ Challenger’ Report, as but for it I could not have completed this investigation, for I need hardly point out that in the absence of zoological libraries such a piece of work becomes almost impossible. 2. SUMMARY OF RESULTS. The majority of the already known species of New Zealand Holothurians have been re-examined, especially with regard to 24 PROF. A. DENDY ON THE the hitherto almost entirely unknown internal anatomy and spiculation, and the results are given in the body of this paper, illustrated by figures of the characteristic spicules and other parts. Four new species are described and illustrated, viz., Cucwmaria Huttoni, Colochirus ocnoides, ,Colochirus calearea, and Psolus macquartensis. The ‘ Challenger’ specimens of Chirodota australiana are shown to be probably identical with C. dunedinensis, the sigmoid spicules of the latter having been hitherto unrecorded ; so that it is very doubtful if the Australian species has been found in New Zealand waters at all, unless indeed the two be identical. Echinocucumis alba, Hutton, is shown to be a Colochirus. Thyone brevidentis, Hutton, is shown to be almost certainly a Colochirus also. | Thyone caudata, Hutton, and Thyonidium rugosum, Théel, are shown to be identical with Thyone (Pentadactyla) longidentis, Hutton. Stichopus sordidus, Théel, is shown to be identical with Holo- thuria mollis, Hutton, which is really a Stichopus. Allowing for these determinations in synonymy, the number of known species of New Zealand Holothurians now becomes seventeen, but four of these, viz., Synapta inequalis, Cucumaria (?) Thomsoni, Cucumaria(?) turbinata, and Holothuria Robsoni, are still very imperfectly known and may prove to be synonymous; while two of them, viz., Ankyroderma Marenzelleri and Holothuria lactea, were obtained by the ‘Challenger’ in deep water, and perhaps ought scarcely to be included in the New Zealand fauna. It is remarkable that no fewer than three of our New Zealand species, viz., Cucumaria Huttoni, Colochirus alba, and Colochirus ocnotdes, are provided with overlapping dermal plates, a condition which elsewhere does not appear to be at all common in the group. Théel has already pointed out that the genus Colochirus is especially characteristic of the Indian and Pacific Oceans, so that it is not surprising to find that no fewer than four of our species are referable thereto. I have not thought it necessary to give figures of the general anatomy because no strikingly new anatomical facts have been discovered, and descriptions of the internal organs appear generally to be sufficient for systematic purposes. Perhaps the most interesting anatomical result is that Chirodota HOLOTHURIANS OF NEW ZEALAND. 25 dunedinensis is shown to be unisexual, as opposed to the herma- phrodite condition supposed to be characteristic of the genus. A little-known type of spicule is described in a species of Stichopus (figs. 88-87) ; and some apparently new facts concerning the structure and development of the wheels of Chirodota are given in the Appendix. The arrangement of the genera is that followed in the ‘ Chal- lenger’ Report. 3. Tot New Zeatann SPEcrIEs. SyNAPTA UNCINATA, Hutton. 1872. Synapta uncinata, Hutton, Cat. Echinoderm. N. Z. p. 16. 1886. Synapta uncinata, Théel, ‘ Challenger’? Holothurioidea, p. 27. The original description runs :—‘‘ Body curved, ? tapering behind, broadest at the mouth; tentacula thirteen, very short, merely tubercles, with two incurved hooks at the end of each. “ Pale reddish-brown. “ Length 6; breadth at anterior end °2. “ Dermal plates anchor-shaped; the flukes equal.” A small jar in the Wellington Museum, labelled “31. Synapta uncinata,” contained what is evidently the type and only known specimen of this species. The specimen is merely the anterior end of the animal, about half-an-inch in length, bearing the crown of thirteen tentacles. The individual to which it belonged may have been of considerable size, probably at least four or five inches in length. The thirteen tentacles are of equal size and arranged in single series. In their present contracted state they are short and stumpy, and each bears four incurved digits at its extremity, the two inner ones being larger than the two outer. They are evidently the two larger digits to which Hutton refers under the name of “incurved hooks.’ Each tentacle also bears along each margin, and facing inwards, a double row of small pro- jectious which are indistinguishable from sucker-bearing tube-feet. On cutting the specimen open, I found the madreporic ring and its appendages still in a fair state of preservation. There were nine long, straight, slender Polian vesicles and three madreporic canals. Of the latter, two were branched and slightly convoluted ; the other much convoluted, but apparently not branched. The madreporic canals were situate dorsally, on the same side as the mesentery. The longitudinal muscles were very strongly developed, and there were no retractors. 26 PROF. A. DENDY ON THE Although I examined preparations of the integument, both simply cleared with absolute alcohol and oil of cloves and after boiling with caustic potash (2 per cent. solution), I could find no spicules, nor could I find any pharyngeal skeleton. Possibly the spirit in which the specimen has lain for more than twenty years may have contained some acid which has effec- tually removed all calcareous deposits. SYNAPTA INEQUALIS, Hutton. 1872. Synapta ineequalis, Hutton, Cat. Echinoderm. N. Z. p. 17. 1886. Synapta ineequalis, Théel, ‘ Challenger ’ Holothurioidea, p. 32. The original description runs :—‘ Conical ; inflated anteriorly and tapering behind; soft, minutely papillose; tentacles none? “ Brownish grey. “ Length 1:0; breadth at the anterior end ‘33. ‘Dermal plates anchor-shaped, with one fluke much longer than the other.” The type specimen in the Wellington Museum is in a perfectly useless condition. It was nearly dry when [I first saw it and has apparently at some time been completely desiccated, so that it now resembles a small fragment of brown, shrivelled leather. As in the type of S. wneinata, the spicules have apparently all been dissolved out. CHIRODOTA DUNEDINENSIS, Parker. (Pl. 3. figs. 1-8.) 1881. Chirodota dunedinensis, Parker, Trans. N. Z. Inst. vol. xiii. p- 418. 1886. Chirodéth australiana, Théel, ‘ Challenger’ Holothurioidea, p. 16. 1886. Chirodota dunedinensis, Théel, ‘ Challenger’ Holothurioidea, p. 34. The original description of the species runs :—‘ Tentacles ten, each with about ten processes, which increase in size pro- gressively from the proximal to the distal end. Integument quite smooth, there being no tentacles or papilla. Colour yellowish (owing to the bright yellow viscera shining through the translucent skin) with small crimson spots which disappear in spirit ; tentacles whitish, with dark spots on the inner side at the base ; these spots are unaffected by spirit. Length, in the extended condition, about 4em. Otago harbour : littoral.” This is supplemented by the following generic diagnosis :— ‘‘Worm-like ; calcareous spicules in the form of wheels imbedded. HOLOTHURIANS OF NEW ZEALAND. Pai in the skin; tentacles shield-shaped, produced at the edges into finger-like processes (Tentacula peltato-digitata).” Professor Parker has very kindly sent me six specimens of this beautiful little Holothurian, three preserved after treatment with osmic acid and three after treatment with picric. Although it is ten or twelve years since they were bottled, the specimens are in an excellent state of preservation, especially those treated with picric acid. J am thus able to add the following particulars to the above description. The spicules (which have not been perceptibly injured by the method of preservation) are of two kinds :—(1) Wheels (figs. 1- 6), about 0-16 mm. in diameter, and each with six spokes (vide Appendix). (2) Contorted sigmoid spicules (fig. 7), resembling the siliceous spicules known as “contort sigmata’”’ in sponges ; measuring about 0°11 mm. from bend to bend. Both kinds of spicule are abundant and both are loosely and irregularly scattered through the interambulacral integument, not aggre- gated in papille ; but whereas the sigmata occur abundantly in all the interambulacra, the wheels appear to be entirely absent from the two interambulacra of the ventral surface. There is a pharyngeal skeleton in the form of a slender calcareous ring of ten simple pieces, of which the ends touch (fig. 8). The integument is thin and translucent, but the five longi- tudinal bands of muscle are well developed. There are, of course, no retractor muscles. The alimentary canal is not thrown into loops but,runs straight from mouth to anus, and is slightly convolutéd by the con- traction of the body. There is a single elongated, slender Polian vesicle, placed ventrally ; and what appears to be the madreporic canal is very small and situate dorsally. The sexes are distinct, and the females may be readily dis- tinguished by the relatively large eggs in the ovary, which give that organ a coarsely granular appearance very different from that ot the smooth-looking testis. The reproductive organs have the same form and relations in both sexes, consisting of a few long slender tubes hanging down from the end of the short genital duct, which lies dorsally at the anterior end of the body. The longer tubes extend back nearly to the hinder end of the body. The condition of the reproductive organs is especially inter- 28 PROF. A. DENDY ON THE esting, because the genus Chirodota is usually considered to be characteristically hermaphrodite ; and this is, I believe, the first species known to depart from the typical condition in this respect. The species is evidently very closely related to C. australiana, Stimpson*. Théel, however, being unaware of the existence of sigmoid spicules in C. dunedinensis, placed it in a different section of the genus. Asa matter of fact the Australian species seems to be distinguished by little if anything more than the aggregation of the spicules into two kinds of papille. A comparison of the two may even show them to be identical, but I do not think this very likely. Théel observes that in the ‘Challenger’ specimens identified by him with C. australiana, “the wheels and sigmoid bodies seem to be present all over the body,” and that the aggregation of the spicules into papille was not distinct. These specimens at any rate are probably identical with the Dunedin species. The record of locality in the ‘ Challenger’ Report seems to be some- what doubtful, viz. “ Habitat. Port William (New Zealand, Falkland Islands?) ; depth 5-10 fathoms; two specimens.” ANKYRODERMA ManrEnze_urRt, Théel. Ankyroderma Marenzelleri, Théel, ‘ Challenger’ Holothurioidea, p. 41, pl. ui. fig. 1. This appears to be a deep-water species, and only a single incom- plete specimen was obtained by the ‘ Challenger,’ from a depth of 700 fathoms (Station 169). It has not since been met with. Cavuprina corracEA, Hutton. (PI. 3. figs. 9-18.) 1872. Molpadia coriacea, Hutton, Cat. N. Z. Echinoderm. p. 17. 1879. Caudina(?), Echinosoma (?), Hutton, Trans. N. Z. Inst. vol. xi. p. 307. 1883. Caudina meridionalis, Bell, Proc. Zool. Soc. Lond. p. 58, pl. xv. fig. 1. 886. Caudina coriacea, Théel, ‘ Challenger’ Holothurioidea, pp. 47, 54, pl. iii. fig. 4. The original description runs as follows :—“ Body cylindrical, tapering rather suddenly into an attenuated and tapering tail, which is half the length of the body; skin rough, coriaceous ; body transversely wrinkled; anterior end smooth; mouth with from ten to twenty short, simple papille ; teeth five, long, bifid ; longitudinal muscles with a deep central groove.” * Vide Théel, loc. cit. p. 16 &e. HOLOTHURIANS OF NEW ZEALAND. 29 “ Pale brown. “ Length 4:0; breadth °65.” This description is preceded by the following diagnosis of the genus Molpadia :—‘ Body attenuated posteriorly ; tentacula simple, short, cylindrical.” I found in the Wellington Museum a single specimen in a jar labelled “34. Molpadia coriacea,” which is doubtless one of the specimens examined by Hutton, though as it had not been cut open it can hardly be the one from which the above description was taken. The specimen was in a very bad state of preserva- tion, having evidently been desiceated ; in fact it was nearly dry when I first saw it. The spicules, however, were still present. I have also received from Captain Hutton a type-slide of the spicules of this species, from Wellington. I have also had for examination a fine specimen from Akaroa, belonging to the Canterbury Museum; three fresh specimens east up on the shore at Oamaru, from Captain Hutton; and three spirit-preserved specimens from the same locality from Professor Parker. I have also seen a specimen cast up on the New Brighton beach; so that the species would appear to be not uncommon on the East coast of New Zealand. The original description was apparently based on a shrivelled specimen, and will be seen to require considerable modification. The ‘Challenger’ specimens were fragmentary, and the speci- mens described by Bell also seem to have been in a very unsatis- factory state. Théel states that he saw two specimens presented by Hutton to the State Museum of Stockholm, but he was evidently unable to make a complete examination of them. Under these circumstances the following particulars as to this very remarkable-looking Holothurian may be of interest. The animal consists of an inflated, subcylindrical or ovoid body, broadly rounded in front and contracting suddenly behind to form the stout cylindrical “‘ tail,”” which tapers gradually to the terminal anus. None of the specimens examined by me show any definite anal papille or teeth, as described by Théel for the ‘Challenger’ specimens. In the largest example which I have seen, the body and tail are each somewhat over 3 inches in length, the body about 17 inches in diameter in the middle, and the tail about 7 inch in diameter in the middle. The tentacles are fifteen in number, short and stumpy, each with four conical digits, two large and two small. Thev are 30 PROF. A. DENDY ON THE evidently the contracted tentacles which Hutton describes as “papille”’ in the neighbourhood of the mouth. In fresh speci- mens the body is of a dull purple colour, and the tail greyish yellow. The tentacles are orange, sometimes streaked longitu- dinally with pink. The body-cavity is filled with a large quantity of thin, reddish-brown liquid, resembling blood, to the presence of which the dull purple colour of the animal is largely due. The integument in fresh specimens is smooth and subglabrous all over, transversely wrinkled only at the root of the tail. It is thin and soft in the region of the body, though tough. It appears firmer in the tail because of the absence of any large, fluid- containing body-cavity therein. The alimentary canal appears to be quite typical. It consists in the swollen “body” of three limbs, descending, ascending and, descending, forming the usual S-shaped loop, the last limb ending in the long, slender rectum, which occupies the whole of the tail and is attached to the body-wall by radial muscle-fibres. The two respiratory trees, attached to the commencement of the rectum, are large and very copiously branched. The longitudinal muscles are powerfully developed, each con- sisting of two distinct bands with a deep groove between them. There are no properly developed retractor muscles, but these are probably represented by certain small slips of muscle which arise one from each half of each longitudinal muscle near its anterior end, and unite together in pairs to become attached to the radial plates of the calcareous ring. The ambulacral ring has a single, sausage-shaped Polian vesicle, placed ventrally, and a single small madreporic canal, placed dorsally close to the genital duct. There are fifteen elongated tentacular ampulle. The sexes are distinct. The reproductive organs are bulky and consist of a great number of copiously branched, slender, moniliform tubes, attached near the anterior end of the body to the single short dorsal genital canal. The pharyngeal skeleton (Pl. 3. fig. 18) consists of a complete calcareous ring of ten pieces. The five radials are rather large, and each ends posteriorly in a bifid prolongation. The interradials are small and simple, wedged in between the anterior ends of the radials. The anterior margin of the ring is provided with fifteen small sharp teeth, two belonging to each radial and one to each interradial. HOLOTHURIANS OF NEW ZEALAND. 31 The typical spicules (figs. 9-11) are described by Théel as “wheel- or cup-shaped deposits,” and his figures leave no doubt as to their identity with those examined by me. I should prefer to describe them, however, as rings, which enclose a cross on the one side and a square on the other. The margin is undulating, curving in at eight points, where the cross and square are attached. The diameter of the ring is about 0°055 mm. These spicules are very abundantly scattered through the integument, forming an almost continuous crust. As Théel has already pointed out, the spicules around the anus differ from those elsewhere, being irregularly branched or reticulate (Pl. 3. figs. 12-14). The tentacles contain only a few small reticulate plates lying near their bases. This Holothurian is very remarkable for the brownish-red celomie fluid, which bears many points of resemblance to human blood. Under the microscope it is seen to consist of a colourless liquid in which numerous corpuscles float. The most conspicuous of these are spherical bodies (Pl. 3. fig. 15) of a brownish-yellow colour, and about a quarter as large again as human red blood- corpuscles. These are extremely numerous, and sometimes appear polygonal from mutual pressure (PI. 3. fig. 15a). They are mostly of about the same size, though some are a good deal smaller than the majority. Each contains, as a rule, one ora few highly refractive particles. A 5 per cent. solution of common salt causes the cell-contour to become wrinkled, but they never appear flattened like human red corpuscles. Colourless corpuscles also occur in the cclomie fluid; these appear to be of two kinds, some being only slightly smaller than the brown corpuscles and coarsely granular, while others are only of about half the diameter of the brown corpuscles, and finely granular. The coelomic fluid does not coagulate on exposure to the air, but does so in alcohol. The above observations were made on fresh specimens which had been out of the water for some little time. The only kindred observations known to me are those of Howell, who describes* hzmoglobin-bearing corpuscles in the water-vascular and perivisceral fluids of Thyonella gemmata. Howell has also referred to the existence in this species of two kinds of colourless corpuscles; and the structural differences which * Howszbr, W. H. Studies from the Biol. Lab. of Johns Hopkins Univ., Baltimore, vol. iii. p. 284. 32 PROF. A. DENDY ON THE I have observed in these in Caudina are suggestive of those dis- covered and more fully investigated by Hardy * in the Crustacea. Cucumaria Hurtoni,n. sp. (Pl. 3. figs. 19, 20.) Body cucumiform, bluntly rounded in front, more tapering behind; posterior extremity slightly turned up dorsalwards. Integument smooth, but hard and thick, with minutely but distinctly scaly appearance, owing to the presence of over- lapping calcareous plates, which are distinctly visible to the naked eye. No distinct teeth or papille around mouth or anus. Length of body a little more than 3 inches, diameter in the middle about two-thirds of an inch. Tentacles copiously branched, dendriform, ten innumber. The mid-ventral tentacle is longer, and that on each side of it shorter, than any of the others; the mid-dorsal is also perhaps rather shorter than the lateral tentacles, but all are well-developed. The body when fresh is of a pale greyish-yellow colour. The tentacles are abundantly speckled with brown. The ambulacral appendages are very feebly developed. Three irregular, multiple, but rather scanty rows of small pedicels (? any fully developed tube-feet) are visible on the ventral ambu- lacra, in the middle part of the body. None are visible on the dorsal surface. Dissection failed to reveal the presence of any ampullee. The five longitudinal muscles are undivided and powerfully developed. The retractors of the pharynx are remarkably strong, and spring from the corresponding longitudinal muscles at about one-third of the distance from the anterior to the posterior extremity of the contracted animal. The pharynx (PI. 3. fig. 19) is very large and consists of two dis- tinct parts. The anterior part is about half an inch in length, eversible, muscular and soft-walled. The posterior part is about an inch in length and is supported by tie very strongly developed calcareous ring, to the radial pieces of which are attached the retractor muscles. The tentacles are attached at the junction of the anterior and posterior portions of the pharynx, a little in front of the retractor muscles. In the retracted state they point straight backwards, and are thus lodged in the posterior part of the pharynx, which they completely fill. The calcareous ring (Pl. 3. fig. 19) is very complex, and consists- of ten compound pieces. The radial pieces are rather narrow, but * Harpy, W. B. Journ. of Physiol. vol. xiii. p. 160. HOLOTHURIANS OF NEW ZEALAND. 33 they are prolonged backwards each into a pair of long slender many-jointed processes, the hinder extremities of which curl inwards at the bottom of the pharynx. These bifurcate prolon- gations are about an inch in length. The interradials are of about the same width as the radials in front, and are continued backwards between the bifurcate processes of the latter, for about three quarters of an inch, each in the form of a long flat rod made up of many joints. Owing to their much greater width, the backward prolongations of the interradials are much more conspicuous than those of the radials. The entire ring closely resembles that of Cucwmaria conjungens, as figured by Semper*; but in our species the interradials are prolonged much farther backwards, and the radials and inter- radials are separated by considerable intervals, instead of touch- ing one another as in Semper’s figure. This last feature is, however, probably dependent upon whether or not the pharynx is distended by the enclosed tentacles, as it is in our specimen. There is a single dorsal madreporic canal running forwards for a short distance along the pharynx by the side of the genital duct. There are two Polian vesicles, situate ventrally, and remarkable for their great length and slenderness (fig. 19). Each igs rather more than two inches long, so that when pulled out they reach backwards from their attachment at the hinder end of the pharynx to beyond the hinder end of the body. Following immediately on the pharynx, the alimentary canal is for a short distance thick-walled and tubular. The thin-walled narrow intestine is very long indeed, and very greatly convo- luted, the posterior portion, for some distance before the rectum, being twisted into a close spiral. The rectum is about an inch long, and attached to the body-wall by numerous radial muscles. The respiratory trees are strongly developed, copiously branched, with four main branches, one branch lying in each interam- bulacrum except the mid-dorsal. The single specimen is a female ; and the ovaries consist of two remarkably small bunches of short, slender tubes, attached right and left to the dorsal mesentery a little in front of the middle of the body. The single oviduct runs up along the dorsal mesentery as usual. The spicules are flat reticulate scales or plates (Pl. 3. fig. 20), of * © Reisen im Archipel der Philippinen—Holothurien, pl. xiv. fig. 4. LINN. JOURN.—ZOOLOGY, VOL. XXVI. 3 34 PROF. A. DENDY ON THE more or less oval shape, very abundant, large and small, measuring up to about 0°92 mm. by 0°77 mm. in the two diameters. The specimen was found at Oamaru, and is the property of the Canterbury Museum. Cucumartia (?) THomsont, Hutton. 1879. Cucumaria Thomsoni, Hutton, Trans. N. Z. Inst. vol. xi. p. 307. 1886. Cucumaria Thomsoni, Théel, ‘ Challenger’ Holothurioidea, p. 116. The original description runs as follows :—‘‘ Body fusiform, scarcely subpentagonal. Skin rough, wrinkled. Ambulacra with the tubercles densely crowded in about 5 or 6 rows. No feet on the interambulacral areas. Tentacles — ? “Rich brown, the white tips to the feet giving the ambulacral areas a spotted appearance. Length 1? inches. “Stewart Island.—Presented to the Museum by G. M. Thom= son, Esq., after whom I name it. A single specimen in spirit.” This is all the information we have concerning the species, Théel simply mentioning it. Professor Parker, who is now in charge of the Dunedin Museum, informs me that the type specimen was sent to the Colonial and Indian Exhibition in London, and thence to the British Museum, where it should now be, so that we must look in that direction for more light on the subject. CucumaRia (?) TURBINATA, Hutton. 1879. Labidodesmus turbinatus, Hutton, Trans. N. Z. Inst. vol. xi.. p- 307. 1886. ? Cucumaria turbinata, Théel, ‘ Challenger’ Holothurioidea,. Pellife The original description runs as follows :—“ Body rounded, suddenly contracted posteriorly into a short-pointed tail, and anteriorly into a rather long, cylindrical neck; skin smooth, slightly transversely wrinkled; the two dorsal ambulacra with two rows each of rather distant feet; the three ventral ambu- lacra either like the dorsal or with more crowded feet in several. rows. Tentacles —? “Body white, covered: with a brown epidermis, which easily peels off, except round the ambulacral feet. Length 23 or 3 inches. ‘“‘ Stewart Island.—Presented to the Museum by G. M. Thom- son, Esq. Two specimens in spirits.” In this case again, Professor Parker informs me that the type: : ; : HOLOTHURIANS OF NEW ZEALAND. 35 or types were sent to the Colonial and Indian Exhibition, and thence to the British Museum. Conocurrus aLBa, Hutton, sp. (Pl. 4. figs. 21-32.) 1872. Chirodota (2?) alba, Hutton, Cat. Echinoderm. N. Z. p. 17. 1879. Echinocucumis alba, Hutton, Trans. N. Z. Inst. vol. xi. p. 307. 1886. ? Echinocucumis alba, Théel, ‘Challenger’ Holothurioidea, pe Ld. The original description runs as follows :—“ Cylindrical, taper- ing behind; skin reticulated, and with longish papille on the back and sides ; tentacula ten, large, branched and plumose. “White ; skin translucent; tentacles brownish white, spotted with violet on the inside near the base. ** Length 1 inch. “ Wellington Harbour (H. Travers).” To this was added in 1879 the following :—“ The receipt of another specimen of the Chirodota (?) alba of my Catalogue has enabled me to dissect it, and I find that it has five well-marked ambulacra, and should be placed in the genus Echinocucumis.” Théel observes that the species certainly does not belong to the genus Hehinocucumis, and I find that it is really a Colo- chirus. My own observations have been based upon :—(a) Two speci- mens from the Wellington Museun, in the same bottle, labelled “33. Chirodota ? alba. Wellington Harbour.” (One of these is probably Hutton’s type.) (4) One specimen from the Dunedin Museum, labelled ‘‘ Hehinocucumis alba,’ and probably the specimen dissected by Hutton. (c¢c) One specimen dredged in Wellington Harbour (12 fathoms, mud), and given to me by Mr. H. Farquhar, who had already identified it with Hutton’s species. The external appearance is very characteristic. The body, which is only about an inch in length, is bluntly rounded in front and tapers rather suddenly behind to a slender tail, which occupies about 7 to 3 of the total length of the animal. In three of the specimens the posterior half or more of the body is strongly bent dorsalwards, the middle of the body being some- what swollen or bellied. I counted the tentacles in the larger of the Wellington Museum specimens, in which they were well extended, and found them to be dendriform, ten in number, with the two ventral very much smaller than the remainder. Five ambulacral bands of rather conspicuous tube-feet (or 3% 36 PROF. A. DENDY ON THE papille) extend from end to end of the body, being in approxt- mately single series in the anterior and posterior portions, but more numerous in the middle, especially on the ventral surface, where there are also a few scattered on the interambulacra. Dissection revealed ampulle only on the ventral side, in the middle part of the body. The entire body is covered with overlapping calcareous scales, which give rise to the reticulate appearance of the skin observed by Hutton. The anus is surrounded by a slight fringe of slender spicules. The internal anatomy agrees very closely with that of Colo- chirus ocnoides, described below. ‘There are two genital bundles, right and left, attached to the dorsal mesentery a Jittle in front of the middle of the body. The Polian vesicle is single, dorsal, much inflated, lying to the left of the single dorsal madreporie canal. The respiratory trees are, as usual, attached to the anterior end of a long rectum, and the alimentary canal resembles that of C. ocnotdes. The calcareous ring is composed of ten y-shaped pieces, with their forks directed backwards (PI. 4. fig. 32). The radial pieces are much longer and project much further backwards than the interradials, so that the ends of the limbs of the interradials articulate with about the middle of the stems of the radials, which widen out somewhat at this point. The notch between the two limbs of the radials is also much smaller than in the interradials. The most abundant spicules are the large and small reticulate plates or scales (PI. 4. figs. 21-24), measuring up to about 1 mm. in diameter. In addition to these there are small, deep reticulate cups (Pl. 4. figs. 27-31), the margins of which are beset with a variable number of very conspicuous, rather slender, short pro- jections. These cups, with their crown-like margins, are highly characteristic ; they measure about 0°02 mm. across the margin, and about the same in depth. Perforated rods (figs. 25, 26) also occur in boiled-out preparations. They doubtless belong to the tube-feet. CoLOCHIRUS OCNOIDES, n. sp. (PI. 4. figs. 33-43.) Body cucumiform, cylindrical, with the posterior third or thereabouts strongly turned up dorsalwards and tapering to the anus. Anterior end much more bluntly rounded (when the HOLOTHURIANS OF NEW ZEALAND. 37 tentacles are retracted) than the posterior. Tentacles ten, tufted with short branches; the two ventral much smaller than the eight others. Integument hard, covered with a close armour of imbricating scales. In the anterior and posterior portions of the body the scales overlap very regularly, the free edges pointing towards the extremity in each case. In the middle third the arrange- ment is less regular. The colour of the living animal is pink, and the colouring-matter appears to be located in the central portion of each scale. The oral and anal openings are guarded by a few irregular nodules. The fully developed tube-feet appear to be confined to the middle third of the ventral surface of the body. They are very numerous and arranged in three crowded ambuiacral bands. They have sucking-disks and ampulle. Small papille are irre- gularly scattered over the dorsal surface, but chiefly on the ambulacral areas. Both tube-feet and papille die away towards the extremities, leaving the terminal portions of the body smooth but scaly. Both are far more numerous in the larger specimens. The internal anatomy is typical. The five retractor muscles are well developed. The alimentary canal is very long and much convoluted in the middle part of its length, but still showing the usual arrangement in descending and ascending loops. The rectum is long, its actual dimensions varying with the state of contraction of the animal. The two respiratory trees are well developed and copiously branched, and extend forwards to the anterior end of the body. The genital organs consist of two voluminous tufts of slender ceca, situated in the middle third of the body, from which the genital duct runs forward dorsally along the mesentery to open close to the crown of tentacles (if not actually between two of them). There is a single Polian vesicle, attached to the ambulacral ring a little to the left of the mid-dorsal line, and a single madreporic canal attached dorsally (Pl. 4. fig. 33). The calcareous ring (PI. 4. fig. 34) consists of ten Y-shaped pieces with their forks directed posteriorly. (In the radial pieces, to which the retractor muscles are attached, the single arm of the Y appears to be jointed on to the fork, but this is probably due to accidental fracture, caused by the excessive contraction of the muscles.) The radials and interradials are more nearly equal and similar than in C. alba. The principal spicules are large and small, flat, reticulate 38 PROF. A. DENDY ON THE plates or scales (Pl. 4. figs. 35, 36), measuring up to about 1 mm. in diameter, and varying from oval to roundedly triangular in outline. In addition to these, there are small reticulate cups (Pl. 4. figs. 41-43), of the same fundamental form as in C. alba, but much larger and with the marginal projections represented merely by blunt warts. These cups measure about 0-054 mm. in diameter. Perforated rods (Pl. 4. figs. 37-40) occur. The type specimen, from which the anatomical details were taken, measured about 53 mm. in length (exclusive of tentacles). Another specimen from the same locality (New Brighton) was of about the same size, but the tail portion was longer and narrower, apparently less contracted. Two more specimens, belonging to the Christchurch Museum, and of unknown locality, were respectively rather more and rather less than twice as long as the type, and thick in proportion. One of the two larger specimens was also dissected and exhibited a peculiar abnormality in the tentacles, two dorsal being the smallest instead of the two ventral, which latter were well developed. The left dorsal retractor muscle was also found te be disconnected from the calcareous ring, so that its anterior end projected freely into the body-cavity. It looked as though it had been ruptured and the ruptured end had healed up. Possibly the degeneration of the dorsal tentacles is to be connected with the injury to this muscle. All the tentacles, however, were completely retracted. It is not easy to convey in writing an adequate idea of the distinction between this species and the preceding. Both species are remarkable for their scaly armature. It is very interesting to find in two closely allied species from such near localities a slight difference in external characters accompanied by a slight but well-marked difference in spiculation. CoLOCHIRUS CALCAREA, n.sp. (PI, 5. figs. 44-53.) Animal minute. Body short, sausage-shaped, bluntly rounded at both ends and without dorsal flexure. Colour chalky white in life and in spirit. Tentacles dendriform, ten in number, the two ventral much smaller than the eight others; pale yellow in colour. Tube-feet in almost or quite simple series on the three ventral ambulacra. Papille on the dorsal surface, chiefly in two rows, sparse, moveable and retractile, each sometimes HOLOTHURIANS OF NEW ZEALAND. 39 with a pink tip. Anus surrounded by five teeth, each overhung {in life) by a slender cylindrical process *. The largest specimen which I have seen measured, after pre- servation in spirit, about half an inch long by rather less than + inch in greatest diameter. In spirit specimens the tube-feet and papille appear to be very irregularly arranged, and no distinct teeth are visible around the mouth or anus. The integument is thick, friable and chalky white, owing to the very numerous minute spicules which it contains. The pharynx is provided with a slender caleareous ring of ten simple triradiate pieces, each with the odd ray pointing for- wards (fig. 53). The radial pieces are rather larger than the interradials, and differ from them in being deeply notched pos- teriorly between the paired arms and in having the apex of the odd arm slightly bifid. Following on the pharynx the short slender cesophagus passes directly into the wide intestine, the first part of which seems to have thicker walls than the remainder, but does not appear to be muscular. The intestine is rather wide throughout, looped in descending, ascending, and descending portions as usual, but not convoluted. The rectum is very short, the respiratory trees springing from close to the anus. The respiratory trees are short, digitate, with very short wide branches, very short even in proportion to the size of the body. The reproductive organs consist of two small bunches of short ceca, placed one on each side of the dorsal mesentery rather far forward, just behind the cesophagus. Retractor muscles as usual, but rather long and slender, spring- ing from about the middle of the body. Madreporic canal single and dorsal. Polian vesicle single, rather short, inflated, situated on the left side. The most abundant spicules are small, oval, reticulate plates or buttons, each with four perforations, and studded on both surfaces with rounded, pearl-like knots, which give them a highly characteristic appearance (Pl. 5. figs. 44-47). These spicules measure about 0°:062 mm. in longer diameter. There are typically fourteen knobs on each side, arranged as shown in fig. 47, but often * IT could not make certain whether or not similar processes occur also between the teeth. The processes look like ambulacral appendages. 40 PROF. A. DENDY ON THE one or two more or less, as in figs. 44, 45. There are also much smaller spicules (Pl. 5. figs. 48, 49) of similar general form, but concave on one side and with the knobs very small and irregularly scattered. These doubtless represent the cups of the other species of Colochirus ; they measure about 0:°029 mm. in longer diameter. Curved perforated rods (Pl. 5. figs. 50-52) are also abundant, especially at the bases of the tentacles and in the tube- feet. The tube-feet and papille are also provided with beautiful reticulate end-plates. This interesting little Holothurian occurs not uncommonly on seaweed in Cook Straits, near Wellington, where it may be collected at low tide. CoLOCHIRUS BREVIDENTIS, Hutton, sp. (Pl. 5. figs. 54-61.) 1872. Thyone brevidentis, Hutton, Cat. Echinoderm. N. Z. p. 16. 1886. Thyone brevidentis, Théel, ‘ Challenger ’ Holothurioidea, p. 141. The original description runs as follows :—‘ Cylindrical, rather flatter below, and rounded at both ends ; papillate all over with suckers, those on the back larger and farther apart; tentacula ten, very short, of unequal length, triangular, frondose ; teeth very short; longitudinal muscles narrow, attached to the dental apparatus. “Brown ; tentacles pink, with yellowish tips. “ Rather more than an inch in length.” Théel merely mentions the species; and the only specimen which I have seen is one belonging to the Wellington Museum, doubtless Hutton’s type. Unfortunately it is in a very bad state of preservation. The tentacles and the whole of the viscera had been removed, leaving nothing but the integument, in the form of a dry and empty shell, shaped lke a short, thick sausage. The integument is thick and hard, owing to the very abundant spicules, and it is not corrugated. The smaller “suckers’’ observed by Hutton are doubtless tube-feet ; they occur thickly and irregularly scattered over the ventral surface, except at the anterior and posterior ends. The larger papille, perhaps twice the size of the retracted tube-feet, are abundantly and pretty uniformly scattered over the remainder of the body ; they appear to have retractile apices, and both tube-feet and papille are provided with well-developed reticulate end-plates. The spicules are very numerous, forming a dense crust in the integument. The most abundant and characteristic are the HOLOTHURIANS OF NEW ZEALAND. 4} knobbed reticulate plates or buttons (PI. 5. figs. 54-57), distin- guished from those of C. calcarea by their somewhat larger sizeand less regular shape, measuring about 0:09 mm. in longer diameter. The characteristic reticulate cups of the genus are represented by small perforated plates beset with minute knobs. Some of these are closely similar to the cups of C. calcarea; some are almost, if not quite, flat and of irregular shape (PI. 5. figs. 58-60) ; they are very numerous and vary a good deal in size. In the integument of the dorsal surface large reticulate nodules are also abundant; these are of rounded or oval form, and up to 0°5 mm. in longer diameter. Curved perforated rods (PI. 5. fig. 61) and reticulate end-plates also occur. Thus the spiculation makes a near approach to that of my Coluchirus calcarea, but differs especially in the presence of the large nodules. The arrangement of the ambulacral appendages and the spicu- lation leave no doubt in my mind that the species is really a Colochirus. The shortness of the “teeth,” referred to by Hutton, and the unequal length of the tentacles also favour this view. It seems to be nearly related to C. calcarea, but differs from that species in its much larger size, in the much more numerous and more prominent papille, and in the spiculation as already pointed out. PsoLUS MACQUARIENSIS, n. sp. (PI. 7. figs. 70-72.) Body in spirit slug-like, somewhat flattened ventrally to form a creeping sole, which extends from end to end and is not sharply defined. Broadly rounded in front and behind, about ¢ inch long and 4 inch broad in the middle. Integument soft and smooth, but corrugated, chiefly transversely. Mouth anterior ; anus at posterior end but slightly dorsal. There are three sharply defined ventral ambulacra, each with two or three rows of irregularly alternating tube-feet, possibly forming a single series when the body is extended. The remainder of the integu- ment is quite devoid of ambulacral appendages. Tentacles ten, bushily branched and of about equal size. Pharynx subglobular, thin-walled, swollen out with the indrawn tentacles. No complete calcareous ring, the pharyngeal skeleton being reduced to a few white patches composed of ageregations of granules. The retractor muscles, however, are well developed. Cisophagus short, marked off by a sharp 42 PROF. A. DENDY ON THE constriction from the globular, thick-walled stomach, which is followed by a long and rather wide thin-walled intestine. Rectum very short, attached to the body-wall by radiating muscle-fibres. The two respiratory trees are well developed and copiously branched as usual. There are four long, narrow Polian vesicles, and a single dorsal madreporic canal running in the mesentery close to the genital duct as usual. The Polian vesicles and madreporic canal are arranged at approximately equal intervals round the ambu- lacral ring. The ampulle of the tube-feet are few and small. The reproductive organs consist of two bunches of short ceca attached one on each side to the dorsal mesentery far forward, alongside the oesophagus. The spicules are small, oval, reticulate plates (Pl. 7. figs. 70-72), very sparingly scattered through the integument. Diameter about 0.1 mm. Some of them have small warts on the surface. The type specimen from which this description was taken was collected at Macquarie Island, and given to me by Professor Parker, and I have since received through Mr. H. Suter four smaller specimens collected at the same place by Mr. A. Hamilton, and evidently belonging to the same species. ‘T'wo of these still show in spirit a distinct purple tint on the dorsal surface, which appears to be normally quite smooth. THYONIDIUM LONGIDENTIS, Hutton, sp. (PI. 6. figs. 62-69.) 1872. Thyone longidentis, Hutton, Cat. Echinoderm. N. Z. p. 16. 1872. Thyone caudata, Hutton, Cat. Echinoderm. N. Z. p. 16. 1879. Pentadactyla longidentis, Hutton, Trans. N. Z. Inst. vol. xi. p. 307. 1886. Thyonidium rugosum, Théel, ‘ Challenger ’ Holothurioidea, p. 95, pl. v. fig. 5. 1886. Thyone longidentis, Théel, ‘ Challenger’ Holothurioidea, p. 141. 1886. Thyonidium caudatum, Théel, ‘ Challenger ’ Holothurioidea, p.147. 1891. Phyllophorus caudatus, Ludwig, Bronn’s Klassen und Ordnungen, Echinodermen, p. 347. 1891. Phyllophorus rugosus, Ludwig, Bronn’s Klassen und Ordnungen, Echinodermen, p. 347. It is only after much trouble that I have arrived at the conclu- sions expressed in the above synonymy, from which it will be seen that I regard three “species” hitherto kept separate as identical. My chief reasons for this conclusion are—the identity in structure of the extremely complex calcareous ring in all three, the identity HOLOTHURIANS OF NEW ZEALAND. 43 in spiculation, and the identity in locality ; the condition of the type specimen of 7. longidentis being such as to preclude detailed comparison in other respects. I regret to have to differ from the opinion of so great an authority as Théel, who, while redescribing T. caudatum from specimens furnished by Captain Hutton, also describes his own T. rugosum as distinct; but I cannot help thinking that there has been some oversight here, and that an impartial observer would admit that Théel’s own descriptions are almost sufficient to establish the identity of the two. In order to justify my conclusions I must now discuss the evidence in some detail. The original description of Hutton’s Thyone longidentis runs as follows :— “Cylindrical; body rugose, with numerous small suckers ; head smooth, transversely striated; tentacula five, short, thick, clavate, pedunculated, frondose; dental apparatus very long, nearly half the length of the body, tubular for half its length, the rest cut into five teeth. “ Brown, tenacula pale brown, “ About an inch in length. “The dental apparatus is composed of five plates, each bifid for half its length, joined to one another, and the lateral process of the two adjacent plates together form a tooth.” In 1879 Captain Hutton changed the name to “ Pentadactyla longidentis,” erecting the new genus Pentadactyla especially for the reception of this species, with the generic diagnosis—“ Feet evenly spread over the greater part of the body. Tentacles five, pedunculated, frondose; dental apparatus very large.” He also observes of the species, “It is, however, evident that it is not a Thyone, but belongs to the family Aspidochirote.” Why this conclusion should have been arrived at is difficult to see, for the author describes the tentacles as “frondose,” and I find from examination of the type specimen that well-developed retractor muscles are present. Théel, no doubt correctly, sup- poses the species to be a Dendrochirote, though possibly not a Thyone. Unfortunately, the type specimen (from the Wellington Museum) was represented merely by the empty skin, from which all the viscera have been removed, and by the separated pharynx (in the same bottle) with the calcareous ring intact (PI. 6. fig. 62) 44, PROF. A. DENDY ON THE and two retractor muscles still attached. The tentacles had appa- rently been removed. The calcareous ring (PI. 6. fig. 62) is very complex, made up of many small pieces, and agreeing minutely in structure with that of Théel’s Thyonidium rugosum, as figured in the ‘ Challenger’ Report. The spicules (PI. 6. figs. 63-65), abundantly scattered through the integument, are perforated plates, or, as Théel calls them, “tables,” each with a conical central spine or “spire”? on one side. The Spire is made up of two rods united apically. These spicules differ from those figured by Théel for his Lhyonidium rugosum only in that a great many of them are irregularly quadriradiate in outline. They measure about 0°38 mm. in maximum diameter. The spires appear foreshortened in the figures ; they may attain a length of at least 0°18 mm. If Captain Hutton’s observations on the number of tentacles were correct, it would seem inevitable that this specimen should form the type of a distinct genus, as he proposed. On the other hand, the detailed agreement of the very complicated calcareous ring with that of Théel’s Zhyonidium rugosum, which was obtained by the ‘Chalienger’ from near Wellington, together with the close agreement in spiculation and in external charaeters, including the pedials, which in both cases are wart- like with retractile ends only, convince me that the two are really identical. The tentacles may easily have been wrongly counted in the first instance, for it is by no means easy to make sure of the number when they are bunched together in spirit specimens. Only the apparent slight difference in spicula- tion then remains to be accounted for, and this is probably more apparent than real and does not amount to a specific distinction. Turning now to Hutton’s Thyone caudata, we find the original description as follows :— “ Oylindrical, tapering rather suddenly to the tail; body rough, covered with papille, except the posterior end, which is transversely ridged; tentacula ten, moderate, peduncled and frondose ; dental system large, with five bifid teeth. ‘“‘ Reddish brown, paler at the extremities. “ Length 2°5; breadth °5.” This species was redescribed by Théel from specimens furnished by Captain Hutton, and in the same work the author HOLOTHURIANS OF NEW ZEALAND. 45 describes his Thyonidium rugosum, also from New Zealand. Of both species he remarks that they are nearly allied to Thyonidium japonicum, and yet he does not seem to have been struck with their resemblance to one another. His descriptions alone, as already pointed out, are almost sufficient to establish the identity of the two, there being only two apparent points of difference worthy of notice, the first concerning the number and arrangement of the tentacles, and the second the structure of the calcareous ring. Before proceeding to discuss these points, I should say that I have examined a specimen labelled Thyone caudata, received from the Dunedin Museum, and a type-slide of spicules given to me by Captain Hutton and bearing the same name. Now as regards the number and arrangement of the tentacles, I find that in the Dunedin specimen there are actually only nineteen tentacles, five pairs of small alternating with four and a half pairs of very much larger ones, and the smaller ones lying slightly inside the larger. Thus one of the larger tentacles has apparently been removed, accidentally or otherwise; and I have no doubt the arrangement was originally typical, there being twenty tentacles arranged in an outer circle of five pairs of larger ones and an inner circle of five alternating pairs of smaller ones. I venture to think that in Théel’s specimen the arrange- ment of the tentacles was really the same, but that two of them have been lost. His description strongly confirms this view of the case. He says:—‘ The tentacles are retracted, their true position being difficult to determine. In conformity with the general condition in the genus Thyonidium, the tentacles are unequal and arranged in pairs, five pairs being several times smaller than the eight remaining tentacles, which are distributed as three pairs and two odd tentacles. Thus, these species deviate from the typical forms with twenty tentacles by the circumstance that two of the tentacles are unpaired.” As regards the calcareous ring, we find that in the ease of * Thyonidium rugosum”’ Théel figures but does not really describe it, while in the case of “ 7. caudatum” he describes but does not figure it. It is evident, however, that the ring is practically identical in-the two. In“ 7 caudatum”’ it is described as ending posteriorly in “five slender bifurcate prolongations.” Hutton also speaks of ‘five bifid teeth.” I must confess that these descriptions alone do not convey a very clear idea to my mind. 46 PROF. A DENDY ON THE On referring to the Dunedin specimen of “ 7. caudatum,” how- ever, I found the calcareous ring (fig. 66) to be practically identical in structure with that figured by Théel for “ 7. rugosum,” and also with that of the type specimen of T. long:- dentis (compare figs. 62 and 66). Now neither the actual specimens nor the figure referred to show what I should term “ five slender bifurcate prolongations ” or “ bifid teeth ;” and I am inclined to think that there has been a good deal of confusion in the terminology and description of the parts of this very complex structure. Again, in describing 7’. longidentis, Hutton says—‘‘ The dental aparatus is composed of five plates, each bifid for half its length, joined to one another, and the lateral process of the two adjacent plates together form a tooth.” As a matter of fact each interradial plate bifurcates posteriorly and the divisions unite in pairs to form the five slender radial pro- longations or teeth, exactly as shown in Théel’s figure for “ 7. rugoswm.” T found in the Dunedin specimen a single madreporic canal and a single much elongated Polian vesicle placed nearly opposite to it. The spicules (Pl. 6. figs. 67-69) are like those figured by Théel for “7. rugosum;” but both in the Dunedin specimen and in Hutton’s type-slide of “ Z. caudata” there are plenty of irre- gularly quadriradiate ones like those of the type of 7. longidentis. Comparison of Théel’s figures of the spicules and calcareous. ring of “ 7. rugosum” and my figs. 62-69, representing the corre- sponding parts in “ 7. caudata ” and T. longidentis, will, I think,. leave little doubt in the mind of the reader that the three are specifically identical. SticHoPus MOLLIS, Hutton, sp. (PI. 7. figs. 73-82.) 1872. Holothuria mollis, Hutton, Cat. Echinoderm. N. Z. p. 15. 1879. Holothuria mollis, Hutton, Trans. N. Z. Inst. vol. xi. p. 308. 1886. Holothuria mollis, Théel, ‘ Challenger’ Holothurioidea, p. 239. 1886. Stichopus sordidus, Théel, ‘ Challenger’ Holothurioidea, p. 167,. pl. viii. fig. 3. The original description runs as follows :—“ Body soft, taper- ing slightly posteriorly ; a row of large tubercles like suckers’ on each side, and another row of fewer suckers on each side of the back; lower surface with many small suckers irregularly placed; tentacles twenty, shortly peduncled, on hollow cylinders, and ending in frondose appendages, which are a ee HOLOTHURIANS OF NEW ZEALAND. AT longer on the outside; dental apparatus short; respiratory organs moderately branched; longitudinal muscles very broad. “ Yellowish, largely mottled with brown above, and in a lesser degree below; tentacles yellowish brown. “ About 6 inches in length, and 13 in breadth. “The suckers are arranged in five rows, but the two upper ones are nearly obsolete ; the three other rows are near together, and often run one into the other; from two to four abreast in a row; these five rows can be well seen near the posterior end, when the intestine and muscular tissue has been removed.” In 1879 (doe. cit.) Hutton observes, “This species in many respects approaches Stichopus. I have had no specimens for dissection, and cannot therefore say whether the reproductive organs are in one or two bunches. A. knowledge of this will settle to which genus it should be referred.” My own dissections show that the reproductive organs are arranged in two bunches, and the species must therefore be referred to the genus Stichopus. Théel quotes Hutton’s species in a list of imperfectly known forms, but I think there can be no doubt that his Séchopus sordidus, which the ‘Challenger’ obtained plentifully in Cook Straits, is identical therewith. Thanks to the kindness of Mr. H. B. Kirk, I was able myself to collect specimens in Cook Straits, in the neighbourhood of Wellington, where the species is abundant in rock-pools, and also to make observations on the living animal. Hutton’s type specimen appears to be no longer in existence, but I have examined carefully a type-slide of spicules prepared by him, and I do not think there can be any doubt of the identification. Probably the living animal varies considerably in colour. Hutton described it as yellowish, mottled with brown, but the usual colour appears to be nearly black, and it is possible, as we shall see presently, that some of the light-coloured specimens may belong to a distinct species. One specimen, however, which undoubtedly belongs to Hutton’s species, was during life “an even white, not striped, on the ventral surface,” as I am informed by Mr. Kirk. The dorsal processes in life are very conspicuous, large and sharply conical, some of them forming a crown around the ventrally situated mouth, as described by Théel. The tube-feet are numerous, and in life appear to be irregularly scattered over As PROF. A. DENDY ON THE the ventral surface, but in spirit specimens an arrangement in three broad, ill-defined bands is recognizable. The living animal attains a considerable size, but varies much according to its state of contraction. In spirit it becomes very much contracted and deeply corrugated, the conical projections being withdrawn. It is very difficult to prevent the animal from discharging its viscera when one attempts to preserve it. The internal anatomy is typical. There is a bunch of long, slender genital tubes on each side of the dorsal mesentery, situated very near the anterior end. There is a pair of well- developed respiratory trees. The calcareous ring (Pl. 7. fig. 82) consists of ten short aud simple pieces, the radials being larger than the interradials, and both notched posteriorly. The tentacular ampulle are finger-shaped and well developed. There is a single large, inflated Polian vesicle placed ventrally, and a single unbranched madreporic canal running forwards in the dorsal mesentery. The most abundant spicules are the usual “ tables ”’ charac- teristic of the genus (Pl. 7. figs. 73-77), with basal plates about 0:06 mm. in diameter. In addition to these occur thin, reticulate, bilateral plates (Pl. 7. figs. 78, 79), as mentioned by Théel, and curved rods (PI. 7. figs. 80, 81). One specimen, also found near Wellington, contains in addition large numbers of another and very peculiar type of spicule in the integument. These spicules are represented in figs. 83-87. They might be described as dichotomously foliaceous, and are apparently in fundamental constitution akin to those described by Bell for Cucumaria inconspicua* from Port Phillip Heads. They measure about 0°03 mm. in greatest diameter. The specimen in which they occur is of a much lighter colour than usual, in spirit being of a mottled yellowish brown. I can, however, detect no other points of distinction. External cha- racters, the number of tentacles (20), the calcareous ring, the single Polian vesicle and madreporic canal, the tentacular am- pulle, all appear to agree with the corresponding structures in S. mollis. Under these circumstances I prefer to await further information before erecting a new species. Hotoruuria Rossont, Hutton. 1879. Holothuria Robsoni, Hutton, Trans. N. Z. Inst. vol. vi. p. 308. 1886. Holothuria Robsoni, Théel, ‘ Challenger’ Holothurioidea, p. 239. * Beit, F.J. Proc. Zool. Soc. Lond. 1887, p. 532. " } | HOLOTHURIANS OF NEW ZEALAND. 49 The original description runs as follows :—“ Elongated, rather slender. Skin, smooth. Feet, scattered sporadically over the ventral surface, apparently none on the back. Tentacles, 20. Anus, round. Back, pale purplish brown. Ventral surface, dirty white, with scattered brown spots. Length, 44 inches. “Cape Campbell. Presented to the Museum by Mr. C. H. Robson, to whom I have much pleasure in dedicating it.” I am informed by Professor Parker that the type specimen was sent to the Colonial and Indian Exhibition and thence to the British Museum. Nothing further is known of the species. HotorHvurta tactsa, Théel. Holothuria lactea, Théel, ‘ Challenger’ Holothurioidea, p. 183, pl. x. figs. 9, 15. This is a deep-water species, obtained by the ‘ Challenger’ at Station 169, Hast of New Zealand, at a depth of 700 fathoms, and also at Station 78, in the North Atlantic, at a depth of 1000 fathoms. APPENDIX. On the Structure and Growth Changes of the Wheels in Chirodota dunedinensis. The fully-developed wheel lies in the integument with its two faces parallel to the surface of the body. The two faces differ very markedly in structure, and, as the position appears to be constant, we may conveniently distinguish them as the inner and outer faces respectively. The outer face is shown in Pl. 3. fig. 1; from which it will be seen that there is a small round hole in the centre, from which the six flattened spokes radiate to the margin of the wheel. At the margin the ends of the spokes are connected together by the rim, which is strongly incurved on this face and provided with a finely serrated edge. The inner face of the wheel is shown in fig. 2. Here we see in the middle a six-rayed cross, covering over a central depression or cavity in which lies the smal] round hole already seen on the outer face. The rays of the cross are much narrower than the corresponding spokes, into which they merge at their outer ends. Fig. 2a represents an optical vertical section of the wheel; passing through two spokes, through the centre of the cross and through the hole in the outer face, which is here placed lower- LINN. JOURN.—ZOOLOGY, VOL. XXVI. 4 50 ’ PROF. A. DENDY ON THE most. The strong curvature of the spokes is well seen, and also the central cavity, with the cross on one side and the small round hole on the other. We may now pass on to describe the growth changes of the wheel. It has long been known that it first appears as a six- rayed cross, and I find that this cross persists as the cross on the inner face of the fully-formed spicule. In the earliest stage which I have met with (PI. 3. fig. 3) there is already a thickening at the end of each ray of the primary cross and on its outer face. These thickenings increase in size and extend both outwards (towards the outer surface of the integu- ment), and centripetally (towards the centre of the cross), till they meet together (figs. 3a, 4), leaving a hexagonal aperture in the middle. The lines of junction between the adjacent thickenings are at first clearly visible (figs. 4, 5), but they presently disappear, and at the same time the margin of the aperture becomes rounded off. Thus is formed the small round hole seen in the middle of the outer face of the fully-formed spicule. Meanwhile the thickening of each primary ray has also been extending centrifugally to form one of the six spokes of the wheel, each of which is for some time irregularly bifid at its extremity (figs. 4, 5). When the spokes have attained their full length their ends thicken laterally and on the outer aspect, and these thickenings, meeting and fusing, form the rim of the wheel, in the manner well known for other species.* It is the formation of the first, centripetal thickenings and the consequent development of the small round hole on the outer face of the spicule which have not, so far as I am able to ascertain, been hitherto described. EXPLANATION OF THE PLATES. Puate 3. Figs. 1-8. Chirodota dunedinensis. Figs. 1-2. Wheels, viewed from the two surfaces and represented as opaque objects. (Zeiss D, Oc. 2, Camera.) Fig. 2a. Optical vertical section of wheel. 3. Early stage in the development of the wheel, the ends of the primary rays just commencing to thicken. (Zeiss D, Oc. 2, Camera.) * Cf. Ludwig in Bronn’s Klass, u. Ordnung. d. Thier-Reichs, Bd. ii. Abth. 3 (Echinod.), p. 58, and also Kishinouye, Zool. Anz. Bd. xvii. p. 146. it ge gt Ae ae oe Gy. he Linn. Soc .Jounn.Zoon Vou XXVl EL. Arthur Dendy del. FH Michael Lith. Mintern Bros .imp. HOLOTHURIANS OF NEW Zn AT AND) Linn. Soc. Journ. Zoom Vor. ©xVi. Pr Ay Arthur Dendy del. FH.Maichael bth. Mantern Bros.irne. HOLOTHURIANS OF NEW ZEALAND. Linn. Soc.Journn.Zoon.Vor.XXVI.Pr.5. D endy ; Arthur Dendy del. EH.Michael hth . VMantern Bros.imp. HOLOTAHURIANS OF NEW ZEALAND . HOLOTHURIANS OF NEW ZEALAND. 51 Fig. 8a. Optical vertical section of slightly later stage, with the thickening greatly increased. Figs, 4-5. Later stages; the thickenings have met together so as to enclose a hexagonal aperture. (Zeiss D, Oc. 2, Camera.) (In figs. 3, 4, 5 the primary six-rayed cross is shaded and the spicule is represented as transparent.) Fig.6. Later stage, represented as an opaque object and showing the commencement of the rim of the wheel. (Zeiss D, Oc. 2, Camera.) 7. Sigma-shaped spicules. (Zeiss D, Oc. 2, Camera.) 8. Part of calcareous ring, x 12. Figs. 9-18. Caudina coriacea. Figs. 9-11. Characteristic spicules of the integument. (Zeiss D, Oc. 2, Camera.) 12-14. Irregular spicules from tip of tail. (Zeiss D, Oc. 2, Camera.) Fig. 15. Coloured corpuscles of the ccelomic fluid. (Zeiss D, Oc. 3.) 15a. Group of similar corpuscles, polygonal from mutual pressure. (Zeiss D, Oc. 3.) 16. Large, coarsely granular, colourless corpuscle from the ccelomic fluid. (Zeiss D, Oc. 3.) 17. Small, finely granular, colourless corpuscles from the ccelomic fluid. (Zeiss D, Oc. 3.) 18. Part of calcareous ring, x 3. (A few transverse cracks, appar- ently accidental, have been omitted.) Figs. 19, 20. Cucumaria Huttont. Fig. 19. Pharynx, skowing calcareous ring, &c., X 2. 4%, interradial plate ; v., radial plate ; 7.7., retractor muscles ; 7.c., madreporic canal; p.v., Polian vesicles; a.c., alimentary canal. 20. Small reticulate plate from integument. (Zeiss D, Oc. 2, Camera.) Puate 4. Figs. 21-32. Colochirus alba, (Specimen from the Wellington Museum.) Figs. 21-24. Small reticulate plates from integument. (Zeiss D, Oc. 2, Camera.) 25-26. Perforated rods. (Zeiss D, Oc. 2, Camera.) 27-31. Reticulate cups from different points of view. (Zeiss D, Oc. 2, Camera.) Fig. 32. Part of calcareous ring, X 13. Figs. 33-43. Colochirus ocnoides. (Specimen from New Brighton.) Fig. 33. Dissection, showing pharynx, crown of tentacles, &.,. xX 3. 4, interradial plate; 7.m., retractor muscles; m.c., madreporic canal; p.v., Polian vesicle; a.c., alimentary canal; g.d., upper end of genital duct ; znz., integument around crown of tentacles. 34. Part of calcareous ring, x 15. (The crack across the radial plate is probably accidental.) 35. Part of large reticulate plate from integument. (Zeiss D, Oc. 2, Camera.) 36. Small reticulate plate from integument. (Zeiss D, Oc. 2, Camera.) 4* 52 ON THE HOLOTHURIANS OF NEW ZEALAND. Figs. 37-39. Perforated rods. (Zeiss D, Oc. 2, Camera.) Fig. 40. Side view of rod. (Zeiss D, Oc. 2, Camera.) Figs. 41-43. Reticulate cups from different points of view. (Zeiss D, Oc. 2, Camera. ) Puate 5. Figs. 44-53. Colochirus calcarea. Figs. 44-47. Knobbed reticulate plates from integument. (Zeiss D, Oc. 2, Camera. 46 in profile.) 48-49. Knobbed reticulate cups. (Zeiss D, Oc. 2, Camera.) 50-52. Perforated rods. (Zeiss D, Oc. 2, Camera.) Fig. 53. Part of caleareous ring, drawn from microscopical preparation and slightly restored. x 30. Figs. 54-61. Colochirus brevidentis. (Specimen from Wellington Museum.) Figs. 54-57. Knobbed reticulate plates from integument. (Zeiss D, Oc. 2, Camera.) 58-60. Small knobbed reticulate plates. (Zeiss D, Oc. 2, Camera.) Fig. 61. Perforated rod. (Zeiss D, Oc, 2, Camera.) PuateE 6. Figs. 62-65. Thyonidium longidentis. (Type specimen from Wellington Museum.) Fig. 62. Calcareous ring, in a much contracted condition, x 8. Figs. 63-65. Reticulate plates from integument. (Zeiss C, Oc. 2, Camera.) Figs. 66-69. Thyonidiwm longidentis. (Specimen from Dunedin Museum, labelled Zhyone caudata.) Fig. 66. Sketch of portion of calcareous ring, X 53. (The outlines between the polygonal plates are in parts very difficult to make out, and the drawing does not pretend to absolute accuracy in this respect.) Figs. 67-69. Reticulate plates from integument. (Zeiss C, Oc. 2, Camera.) Prats 7. Figs. 70-72. Psolus macquariensis. Reticulate plates from integument. (Zeiss D, Oc. 2, Camera.) Figs. 73-82. Stichopus mollis. Figs. 73-77. Characteristic spicules (“Tables”) from various points of view. 78-79. Thin reticulate bilateral plates. 80-81. Curved rods. (Figs. 73-81 drawn from Capt. Hutton’s type slide; Zeiss D, Oc. 2, Camera.) Fig. 82. Part of calcareous ring, X 23. Figs. 83-87. Stichopus mollis ? Additional spicules from a yellow specimen collected near Wellington (Zeiss D, Oc. 2, Camera.) Lins. Soc. Journ. Zoot .Vor.XXVI.PL.G. Arthur Dendy del. Mintern Bros.imp. Jal OIL, OAM ENUIRUVAIN| SOUP ING WP 771 /NIL/AINSID) - FH Michael ith. Dendy. ; Linn .Soc.Jounw. Zoou Vor XXVI.Pu.7. Arthur Dendy del. FH Michael lith. ~Mintern Bros.imp. -HOLOTHURIANS OF NEW ZEALAND SKELETON AND LARYNX OF XENOPUS AND PIPA. 53 \ On the Structure and Development of the Hyobranchial Skeleton and Larynx in Xenopus and Pipa; with Remarks on the Affinities of the Aglossa. By W- G. Ripewoop, B.Sce., F.L.S., Lecturer on Biology at St. M ry’s Ho=pital Medical School, London. [Read 19th November, 1896. | (Puatzs 8-11.) ContTEnts. Page LEN TRODUCTION (eaccecenceticceceseccseseeceee ae seceeescceeen sees 53 Hyobranchial Skeleton of Xenopus levis, female............ 56 Laryngeal Skeleton of Xenopus levis. (Mem aleycscsccssetrcsescsceessctione nce sesusijidenestlsetsestss 60 Mia GO Pesctracireclacidce'cabelerisleeeis iewieieiseeessiaiaacetisensmese ste 64 Hyobranchial Skeleton of Pipa americana, female......... 67 Laryngeal Skeleton of Pipa americana. TET S):ga se shcaa been Sadan sabinanbeceenosdansnpebanoncseciéocde 69 IMIRHID iceaadaeadoueexauocusnadodancocnucoscernseaaacooodooachs 71 Laryngeal Muscles of Xenopus levis. IBermalll @ swiise fst Seleis'ec craic els naltctts Sabeealro ate Gove naioea wate 77 IN IE Fe CBee eonsapoc oso docsboboneTcrucnadescrooCbocnEcoUROnten 83 Laryngeal Muscles of Pipa americana. Memale ns vcanciciwceccstnutstoesccaenscsdwerteraesaressscoes 85 Ila eure cratic csenseacse sc socccnecatessdenetencsaacsseeacats 89 Development of the Hyobranchial and Laryngeal Skeleton. 91 ACTOS UTES. opconsccnopodeescqodesc0526006068000200006000 93 JEG (OGFIBTIUG, “Gasconebeospoopoonvceougeoenacescbeouncebs 102 General Conclusions, and Discussion of the Affinities of CHIT) UE Edo EE Ma Gaaceon gue cpoaduconue donee taLaEpoacen soouaosodados 111 List of Authorities referred to ...........-..eseseeeeeevoneereee 122 Hxplanation of the Plates 2.00. ...00.....cccsscorenecsccceacenss 126 INTRODUCTION. CONSIDERABLE interest has for many years centred around the little suborder of the Anura Aglossa and the question of the affinities of the two sole genera, inter se, and with other Anura. Pipa and Xenopus, so remote from one another geographically and so essentially different in general configuration, size, denti- tion, spawning-habits, and many peculiarities of internal struc- ture, yet exhibit numerous anatomical resemblances which appear to point conclusively to a common ancestry. Standing out pro- minently from among the latter characteristics is the remarkable complexity of the larynx and the extraordinary sexual dimorphism which it exhibits. It is with this organ that the present paper essentially deals. It both Pipa and Xenopus the relation of the larynx to the hyobranchial skeleton is so intimate, that a minute 54 . MR. W. G. RIDEWOOD ON THE HYOBRANCHIAL study of the latter is necessarily demanded; and, in order to insure the correctness of the morphological value attributed to the various parts, a chapter on the development of the hyo- branchial skeleton is added. I have studiously avoided all refer- ence to the columella auris and stapedial cartilages, since these are not related anatomically with the larynx. The expression ‘‘hyobranchial skeleton,” therefore, is here used in a limited and strained sense ; but I do not think that any confusion is likely to be caused by this departure from the strict application of the words. Passing over the works of Fermin (13) and Schneider (37), who mention only incidentally the hyobranchial and laryngeal skeleton of Pipa—Xenopus was not then known,—the first de- scription of these parts occurs ina short monograph by Breyer on Pipa (4. 1811). Mayer (25) in his treatise on the same animal, published fourteen years later, describes for the first time the muscular system. He furnishes a detailed account of the larynx of both sexes; but the description suffers considerably from the lack of pictorial illustration. In his second contribution (26. 1835) he describes the muscles of the hyoid apparatus of Xenopus, and treats of the laryngeal skeleton of both sexes. To his figures of the Xenopus larynx (Taf. 3. figs. 6 & 7) he adds, for purposes of comparison, one of the larynx of the male Pipa (Taf. 3. fig. 5). By far the greater part of our present know- ledge of the subject we owe to the patient and exhaustive investi- gations of Henle (19). Although his book refers to the vertebrate larynx generally, he has devoted quite an exceptional amount of space to the aglossal toads, and has allotted to them the whole of one of the five plates by which his work is so admirably illustrated. He describes with great care the hyobranchial skeleton of Xenopus and Pipa, the laryngeal skeleton of both sexes, and the laryngeal muscular system of the female Pipa and the male Xenopus. Henle was the first to suggest (p. 15) that the hyoglossal foramen, so characteristic of the Aglossa, is formed by a union of the anterior or hyoidean cornua, and that the hind edge of the foramen corresponds with the notch at the front of the body of the hyoid in other Anura. He agrees (p. 16) with Meckel (27. p. 229) and Mayer (26. p. 81), that the anterior cornua have been reduced in Pipa and are only represented by the delicate pointed cartilage at the front of the hyobranchial apparatus. Henle also (p. 16) propounded the view, already hinted at by Mayer (25. p. 541, and 26. p. 31), that the anterior SKELETON AND LARYNX OF XENOPUS AND PIPA. 55 portion of the ventral wall of the larynx of both Xenopus and Pipa is in reality the hinder part of the median hyobranchial plate, and that the rod-like ossifications in the laryngeal skeleton are the columelle or thyrohyal bones of other Anura. The account given by Hoffmann (20) is culled almost entirely from Henle’s treatise, many paragraphs being verbatim extracts. Hoffmann, however, does not seem to have had a clear idea of the nature of the coalescence of the skeletal parts described by Henle ; for in two places he speaks of a fusion of the arytenoid cartilage with the body of the hyoid, unjustifiably substituting “ cartilago-arytenoidea”’ in the one place (p. 518) for “ Stimm- ladenknorpel”’ of Henle (p. 16), and in the other (p. 524) for “ Ringformiger Knorpel” of the same author (p. 26). Parker (33), in his exhaustive work on the skull of Batrachia, gives good figures of the hyobranchial skeleton of larval and adult stages of both Xenopus and Pipa. He ignores Henle’s view of the mode of formation of the hyoglossal foramen, and by labelling the car- tilage which bounds it in front the first basibranchial, tacitly suggests its origin as a secondary fenestration in the body of the hyoid. He considers the great lateral expansion of the hyo- branchial skeleton of both Xenopus and Pipa as a derivative of the first and second branchial arches, and the thyrohyal as formed from the third and fourth. Cope (7) copies Parker’s figure of the hyobranchial skeleton of Xenopus, and gives a new interpretation to the parts, the lateral wing being considered as derived from the first, and the thyrohyal from the second branchial arch. The paper is, however, marred by contradictory statements, and the views expounded in the text do not accord with those expressed by the figure. Gronberg (18) furnishes outline text-figures of the hyobranchial skeleton of Xenopus and Pipa (p. 636), and several excellent illustrations of the larynx of both sexes of Pipa (Taf. 38. figs. 10-20). He pays most attention to the larynx of the male, and treats’ of the intrinsic laryngeal muscles of this sex alone. He regards the whole of the floor of the larynx of Pipa as derived from the branchial skeleton, and compares it (p. 635) with the cartilago thyreoidea of mammals. Wilder, in his admirable paper on the Amphibian larynx, devotes a couple of pages (42. pp. 291-292) to the hyobranchial and laryngeal skeleton of the Aglossa, and gives four figures (Taf. 21. figs. 58-61) of these structures in Xenopus. He refers also to the laryngeal musculature of this genus. Unfortunately his observations on the Aglossa 56 MR. W. G. RIDEWOOD ON THE HYOBRANCHIAL are not of the same standard of excellence as the rest of the work. Such, briefly stated, is the extent of our present knowledge on the subject. With the various conflicting views and contra- dictory statements before me, I set myself to ascertain to what extent, if at all, the hyobranchial skeleton enters into the form- ation of the laryngeal box in the Aglossa; whether the lateral laryngeal ossifications of Xenopus are really the thyrohyals, and whether the ossifications in the floor of the larynx of Pipa are homologous with these. A furtker object of the investigation was the determination of the morphological value of the great wings of the hyobranchial skeleton in Xenopus and Pipa; the experimental verification of the assumed absence of the hyoidean cornua in Pipa; and an inquiry into the functional value of the great arytenoid rods which lie in the interior of the larynx in the male of this genus. The results of the investigation are arranged in the following order :—First, a comparative treatment of the adult hyobranchial skeleton of Xenopus and Pipa, in- cluding a detailed account of the laryngeal skeleton of both sexes ; then a description of the laryngeal muscles of the two genera; the details of the development of the hyobranchial and laryngeal skeleton ; and, in conclusion, afew observations on the bearing of the facts on the problem of the affinities of Xenopus and Pzpa to one another and to the tongued Anura. Before proceeding tarther, I take the opportunity of expressing my grateful acknowledgments to Sir William Flower, K.C.B., for placing in my way every facility for the execution of this research. In addition to the adults of Xenopus and Pipa, which by the kindness of Sir William Flower I was permitted to examine at the Natural History Museum, I have received numerous specimens, both larval and adult, from Prof. G. B. Howes, of the Royal College of Science, Mr. F. E. Beddard, F.R.S., and Mr. G. A. Boulenger, F.R.S.; and to them my thanks are hereby gratefully tendered. Hyosrancuiat SKELETON oF Xenopus levis, FEMALE. In describing the hyobranchial and laryngeal skeleton of the Aglossa it is advisable to commence with that of the female Xenopus, firstly, because in Xenopus this portion of the skeleton is less aberrant than in Pipa, and, secondly, because the modifi- cation is less in the female than in the male. By choosing the SKELETON AND LARYNX OF XENOPUS AND PIPA. 57 simplest of the four, we experience less difficulty in instituting a comparison with the more familiar tongued Anura; and, after having definitely settled the morphological value of each skeletal part, we can readily proceed to ascertain how the more modified parts in the other three are related to those of the first. As in other Anura, so in the Aglossa, it is impossible to draw, in the hyoid apparatus of the adult, a sharp line of demarcation between those structures which are hyoidean and those which are branchial in origin. The anterior cornua representing the main portion of the hyoid arch are present in Xenopus, but disappear during metamorphosis in Pipa (see p. 108). In both genera, however, there is arelic of hyoidean cartilage in the median cartilage situated in front of the hyoglossal foramen. The first figure published of the hyobranchial skeleton of Xenopus is that of Mayer (26. Taf. 3. fig. 7). This is fairly correct in detail, but fails to show the hyoglossal foramen. Henle’s figure (19. Taf. 2. fig. 1), also of the female sex, is far superior. The figure which Parker gives (33. Pl. 58. fig. 5) is, on the whole, less useful than Henle’s. It shows that the larynx -lies between the thyrohyals, but it furnishes no details of the relationship. A similar figure open to the same objection is given by Gronberg (18. p. 636, fig. A, ii.), and two others, with the said details introduced though incorrectly, by Wilder (42. Pl. 21. figs. 58 and 59). In the female Xenopus the anterior or hyoidean cornu (Pl. 8. fig. 1, ch.) is broad and flat, and consists of a rod-like axis, ex- panded on its inner and outer edges into very thin and delicate plates of cartilage which have hitherto escaped notice. At the upper posterior end of the arch the axis is unaccompanied by its lamellar expansions ; it takes a sharp curve upwards and out- wards, and is attached by ligament to the skull about 3 mm. from the articular head of the quadrate. At their anterior or median extremities the cartilaginous rods of the right and left cornua unite at a very obtuse angle. The external lamelle are continuous in front of this union; but the internal or posterior expansions stop at some distance from the median line. From the point of union of the cornua there extends back a median rod of cartilage (Pl. 8. fig. 1, ch.'), very slightly, if at all, broader than the axes of the cornua. The basal plate or body of the hyoid is produced outwards into two smaller anterior processes (Pl. 8. fig. 1, ap.) fitting closely behind the anterior cornua, and 58 MR. W. G@. RIDEWOOD ON THE HYOBRANCHIAL a pair of very large postero-lateral plates, usually described as the “ale” or “wings” (Pl. 8. fig. 1, a)—a most convenient term, as it merely designates the part without expressing its morphological value. The true value of these processes is dis- cussed later, chiefly upon the evidence furnished by a study of the development. The ale are perfectly flat plates of cartilage, with an even margin, slightly pointed posteriorly, and with a blunt notch on the mesial edge at about one-fourth of the length from the posterior extremity. The greatest diameter, at about one-third from the posterior end, is rather more than one-third of the length. Mayer (26. p. 33) says that in the female Xenopus the wings are smaller and thinner than in the male, a statement with which I cannot agree. But since he says (p. 29) the male “ist grésser und schlanker gebaut als das Weibchen,” it is evident that his female specimens were not up to the average size, or that his males were exceptionally large. The anterior processes and the ale are of the same thickness of cartilage as the basal plate itself; but there is a much thinner extension of cartilage, recalling the lamellar expansions of the anterior cornua, which forms the lateral margin. This thins away behind along the external margin of the ala, and is continued forward around the external and anterior edge of the anterior process. Itis indicated in Plate 8. fig. 1 by fainter shading. The hyoglossal foramen (PI. 8. fig. 1, h), which is such a con- spicuous feature of the hyobranchial apparatus, is mainly closed by membrane: the hyoglossal muscles, which rise up to be dis- tributed to the mucous membrane of the floor of the mouth, pass between the posterior edge of the membrane and the cartilage which bounds the foramen posteriorly (Pl. 8. fig. 1, y). The shape of the fontanelle varies ; but it is always oval and elongated in an antero-posterior direction, its greatest diameter lying behind the middle of its length. Henle (19. p. 15), with remarkable introspection, considered that the hinder border of this hole in the basal plate represents the anterior edge of the basal plate of the hyobranchial skeleton of the frog, and that the presence of the foramen is to be explained by the union of the two anterior cornua in front. He based his conclusions almost entirely on the relations of the hyoglossal muscles ; but, as 1 hope to prove in the sequel, he was fully justified in his determination. Behind the foramen the basal plate is sharply constricted, but it almost immediately widens out behind, and becomes indistinguishably SKELETON AND LARYNX OF XENOPUS AND PIPA. 59 fused with the anterior edge of the cricoid cartilage of the floor or ventral wall of the larynx, in the position marked @ in fig. 1. The reasons for fixing this as the region of coalescence are based upon a comparison of the cricoid skeleton with that of the frog (see p. 62) and on a study of the development of the parts (see p- 100). The distance yx, therefore, in fig. 1 represents the very greatly diminished antero-posterior diameter of the large basal plate of the frog’s hyobranchial skeleton. Henle has already pointed out in the middle of the basal cartilage of Xenopus an incipient ossification, which he compares (19. p. 17 footnote) with that larger Y-shaped ossification which occurs in the middle of the body of the hyoid of Alytes. The ossification occurs only late in life, so that it 1s quite possible for the larynx of a sexually mature Xenopus to show no trace of it. Externally to the region of coalescence between the laryngeal cartilage and the basal plate, and separated from it by two small membranous areas, are the anterior ends of two rods of bone which project backwards and, diverging slightly, run along the right and left sides of the larynx proper (Pl. 8. fig.1, 4). These rods are circular in section, are thicker behind than in front, and are narrowest at a short distance from the anterior end. The posterior epiphyses are cartilaginous and exhibit, in fully grown specimens, an incipient endochondral ossification (PI. 8. fig. 8, ep.). The epiphysis is fused mesially with the hind end of the lateral walls of the cricoid cartilage. The rods are the only fully ossified parts of the whole hyobranchial and laryngeal skeleton ; and in this respect, as also in form, they resemble the columelle or thyrohyals of the tongued Anura*, with which they have been considered homologous by all writers who have given attention to the subject, from Henle onwards. The attachment of the rods to the hind part of the basal plate, their lateral position with regard to the larynx proper, their divergence even, all point to the same conclusion ; and, since I hope to prove in the following pages that all the salient features of the cricoid and arytenoid cartilages of the larynx of the frog or other anuran can be dis- tinctly recognized in the cartilaginous apparatus that lies between the two bony rods in the female Xenopus, there would appear to be here sufficient evidence, without having recourse to that afforded in addition by a study of the relations of the muscles, * Ossifications in the body or basal plate occur in Bombinator, Alytes, and. other forms; but even in these the columellz are the most completely ossified. 60 MR. W. G. RIDEWOOD ON THE HYOBRANCHIAL and of the development of the rods themselves, to firmly establish the homology. The posterior ends of the thyrohyals are wider apart than the inner edges of the ale, and lie below or ventral to the latter. The posterior epiphysial cartilages are bound by a very short and tough ligament to the ventral surface of the ale, just external and slightly posterior to the notch on their inner edge already mentioned. LarynGEaL SKELETON oF Xenopus levis. The first account of the larynx of Xenopus is that of Mayer (26. 1835). He describes both sexes and gives figures of each (Taf. 3. figs. 6 and 7). Henle (19), like Mayer, discusses both sexes, and his figures of the dorsal and ventral surfaces of the female larynx (Taf. 2. figs. 1 and 2) are exceedingly accurate. The larynx described by Wilder (41. p. 573, and 42. pp. 291 and 292) must have been in asad state indeed if his figures (42. Taf. 21. figs. 58-61) represent it at all truthfully. These show a number of dismembered cartilages in the roof, and no cartilage in the floor. It is, also, hardly just and fair to condemn Henie’s excellent work and to charge him with overlooking the true arytenoids, on the strength of observations made, as Wilder himself admits (42. p- 290), on a single specimen. This most recent contribution of Wilder’s is so good in the main that it is a great pity it has been marred by the incorporation of the results of “‘ hasty study.” Female. The larynx proper of Xenopus differs from that of the more normal anuran, not only in its superior size and the extensive spreading of the cricoid cartilage so as to form an almost com- plete laryngeal box, but more especially in the union of the lateral] edges of its floor with the inner surfaces of the ossified thyrohyals, the confluence of the anterior region of the floor with the back of the basal plate of the hyoid, and in the coa- lescence which has taken place between the posterior epiphysis of the thyrohyal and the postero-lateral corners of the cricoid. Although the lateral edges of the floor of the larynx are so closely applied to the columelle, the actual sides of the larynx are free, and the narrow space between them and the thyrohyals is filled with a fatty tissue similar to that found on the under surface of various parts of the hyobrancbial skeleton. There is a pear-shaped membranous area (Pl. 8. fig. 2, f) in the cartila- SKELETON AND LARYNX OF XENOPUS AND PIPA. 61 ginous side-wall of the larynx, which is of considerable value as ‘a guide to the recognition of homologous parts in Xenopus and tana (see p. 62). The floor or ventral wall of the larynx has the form of an extensive plate of cartilage, incomplete only in a pair of elongated membranous areas (PI. 8. fig. 1, f') towards its anterior end, where the anterior fifth of the thyrohyal remains free. The floor of the larynx is produced posteriorly into a pair of cartilaginous bands (6r.) with very irregular edges, which run back along the mesial surface of the bronchi, in the possession of which latter the Aglossa depart again from the anuran type. The irregular outgrowths from the sides of the bands extend only to the middle of the upper and under surfaces of each bronchus, so that the outer half of the bronchial tube is entirely membranous. These bronchial prolongations of the cricoid cartilage are not unrepresented in other Anura, for in the Discoglosside (e. g. Discoglossus) two thin cartilaginous bars run back from the floor of the larynx to the under surface of the root of the lung, where they end in a cross-piece, T-shaped, which may be regarded as equivalent to one of the bronchial half-rings of Xenopus. In Alytes these rods are even more elongated than in Discoglossus, but the terminal T-piece is very little developed. In Bombinator the processes are quite short and stout (Pl. 11. fig. 18, d7.). Even in Rana the opening of the lung into the larynx is supported by a ring of cartilage which sends off irregular processes on to the lung. The strongest of these (PI. 9. fig. 7, 6r.), running along the ventro-internal aspect of the lung, would appear to correspond with the process already referred to in the Discoglossidx, and with the bronchial cartila- ginous band of Xenopus. The roof of the larynx of the female Xenopus is deeply notched behind (PI. 8. fig. 3), so that a mem- branous area intervenes between the cricoid and bronchial cartilages. The anterior edge of the roofing cartilage is delicately rounded, and is rather thicker than the other parts. The antero-posterior diameter of the roof is not more than twice the transverse diameter. The arytenoid, like the cricoid, is exceptionally shaped. The sides of the glottis are supported by a pair of delicate cartila- ginous flaps (Pl. 8. fig. 3, av.) which, from their relations to the glottis and to the laryngeal muscles, evidently correspond to the arytenoid cartilages as we find them in the common frog. They are continued backwards into relatively massive processes (ar.’) 62 MR. W. G. RIDEWOOD ON THE HYOBRANCHIAL which, behind the glottis, meet one another in the median plane by perfectly flat and smooth articular surfaces *. These masses of cartilage lie immediately in front of the thickened anterior edge of the roofing cartilage of the larynx and usually contain a small centre of ossification each. These ossifications, like those of the posterior epiphyses of the thyrohyals and the anterior part of the floor of the laryngeal complex, only appear with advancing age. A female may be fully grown and sexualiy mature without exhibiting a trace of ossification in these regions. The lateral motion of the arytenoids is limited by a pair of concavo-convex upgrowths of the side-walls of the larynx (PI. 8. fig. 3, 0/.), called by Henle (p. 17) the “Scheuklappenartige Fortsitze” or blinker- shaped processes. There are no vocal cords, and their absence may account for the exceptional shape and mutual relations of the arytenoids. The failure of the arytenoids to touch one another in front of the glottis is also a matter of no little importance. The relations that obtain between the cricoid cartilage of Xenopus and that of other Anura can best be made out by the study of a median section of the laryngeal skeleton. By this means it is seen that the roofing part of the larynx of Xenopus (Pl. 8. fig. 2, ¢) is nothing more than the median dorsal rod, greatly expanded, of Rana esculenta (Pl. 9. fig. 7, c), while the floor (Pl. 8. fig. 2, c'c’) is made up of the coalesced enlarged transverse bars (PI. 9. fig. 7, ec’ & c’). Even the blinker- shaped processes of Xenopus (Pl. 8. fig. 3, bl.) can be recognized in the frog’s larynx in the swellings (PI. 9. fig. 7, 67.) at the sides of the anterior ventral bar of the cricoid. The mem- branous area (PI. 8. fig. 2, f) in the side-wall of the larynx of Xenopus is a great guide in the determination of the parts, since it fixes the position of the space (Pl. 9. fig. 7, f) at the sides of the ring of cartilage on the ventral surface of the frog’s larynx. Owing to the great expanse of cricoid cartilage in Xenopus, the connexion with the posterior epiphysis of the thyrohyal, situated in the frog close behind the arytenoid (PI. 9. fig. 7, 7’), is thrown * Seeing that in the larval larynx (Pl. 11. fig. 10) the arytenoids are in contact with the anterior edge of the dorsal part of the cricoid, and are even overlapped by it, it becomes doubtful whether, after all, the massive part of the arytenoid of the adult is a process of the true arytenoid—it may be simply an enlargement of the posterior end. In the male of Pipa it is tolerably certain - that the bulk of what is known as the arytenoid is really a process of the true arytenoid ; but the relations are simpler in Xenopus. SKELETON AND LARYNX OF XENOPUS AND PIPA. 63 very far back (see Pl. 8. fig. 3), and the union is brought about in Xenopus by the confluence of the cartilages and not by ligament as in the frog. Henle states (p. 14) that in Alytes obstetricans, Bufo cinereus, Rana esculenta, and Rana temporaria the larynx is connected with the epiphysis of the thyrohyal by continuous cartilage. J have examined these four species, and have found the connexion to be ligamentous in all. Wilder refers (42. p. 286) to a cartilaginous union in Bufo lentiginosus, but the difficulty of obtaining this species in England has pre- vented me from checking the observation. Relatively to the size of the whole body, the arytenoids are much smaller in Xenopus than in the frog, although the larynx as a whole is very much larger. JI have elected to institute this comparison with the larynx of the edible frog, not merely because the figures of the latter by Ecker (12) and others are so familiar to the majority of anatomists, but also because of the fact that the laryngeal skeleton of the female Xenopus will bear a closer comparison with that of Rana esculenta than with that of any other species of Rana. It is true that Wilder (42. p. 287) has shown that the larynx of Rana esculenta is not typical of Anura, nor even of the genus Rana; but by the very possession of the posterior ventral bar of the cricoid (Pl. 9. fig. 7, c’), in which it departs from the other species of the genus, the larynx of the edible frog approaches the nearer to that of Xenopus. In spite of this parallelism existing between the larynx of Xenopus and that of the frog, an extensive study of the skeleton of many different genera of Anura has convinced me that that amphibian (Pipa excepted) the hyobranchial and laryngeal skeleton of which most nearly approximates to that of Xenopus is undoubtedly Bombinator—a determination which I had arrived at prior to the publication of the hint to the same effect by Wilder (42. p. 291). The notch at the front of the hyobranchial basal plate in this toad is excessive, so great in fact as to leave only a narrow transverse isthmus connecting the right and left halves, as in some Hylide*. The paired ossifications in the * In spite of the depth of the notch at the front of the basal plate, the point at which the hyoglossal muscles rise above its level is not in any way thrown back. The notch is closed by a membrane, the front edge of which lies nearly straight and is situated a considerable distance in front of the incised cartilage. This holds good also for Alytes, Discoglossus, Rana, Bufo, and most Anura where the notch is at all deep. In Xenopus the hyoglossal muscles rise immediately in front of the cartilage that forms the posterior border of the oval foramen, while the membrane spreads out in front over the rest of the hole. 64 MR. W. G@. RIDEWOOD ON THE HYOBRANCHIAL body of the hyoid of Bombinator are exceptional, and these are not represented in Xenopus; but the large size of the lateral processes of the hyobranchial skeleton, more especially the posterior pair, is significant, and the great breadth of the plate- like hyoidean cornua renders them remarkably similar to those of Xenopus. The laryngeal skeleton of Bombinator is not only expansive and dome-shaped, and larger in the male than in the female, but is also elongated in an antero-posterior direction, especially in the male sex. These features are well illustrated by Wilder (42. Taf. 21. figs. 45-47), but in his figures he unfortunately omits the conspicuous bronchial processes that project from the posterior edge of the cricoid cartilage (PI. 11. fig. 13, br.). The broadening of the cricoid ‘cartilage in Bom- binator is, as in Discoglossus (where, owing to the larger size of the larynx, the fact can be better made out), confined to the roof and floor. The sides consist of mere bands or rods of cartilage extending vertically from roof to floor, and the ligament that runs out to the thyrohyal epiphysis is attached to the middle of the external surface of the bar. In the Discoglosside, as in the Aglossa, there are no traces of what Henle (p. 11) calls the cartilages of Santorini, small abstricted portions of the arytenoid cartilages occurring in a limited number of Anura, and first figured by Saint-Ange (36. Pl. 26. figs. 4 & 5). They are the “‘ pre-arytenoid” cartilages of Ecker (12) and the “apical” cartilages of Wilder (42). The absence of these cartilages, however, cannot be allowed to carry much weight, since they are not constant in closely allied genera, and may even be sometimes missing in the female of a species normaily possessing them (12. p. 313). Male. The sexual differences in the laryngeal skeleton, so striking in the aglossal Anura, are nevertheless not confined to them. Wilder (42) has already pointed out that in Bufo there are suggestions of sexual dimorphism in the larynx, and that im- portant sexual differences occur in the Hylide and in Bombinator. But the increased size and complexity of the larynx in the male is even more widely spread among Anura than he indicates. Ecker (12. p. 313) has shown that in Rana esculenta the arytenoid cartilages are thick, strong and large in the male, and. thin, smaller and more hollowed in the female. In Alyées and cas SKELETON AND LARYNX OF XENOPUS AND PIPA. 65 Discoglossus I find that the sexual differences are quite as strongly marked as in Bombinator, while in Pelobates, Cerato- phrys, and Pelodytes, although the differences are less pro- nounced, they are clearly recognizable. It is highly probable that more extended observations will amplify the list still further. In the male Xenopus the larynx, relatively to the size of the byobranchial skeleton and to that of the whole body, is con- siderably greater than in the female, but yet, owing to the smaller size of the male as compared with the female, the actual cubic content of the larynx is about the same in both. The larynx of an average male is shorter than that of the average female, but it is broader and more inflated. The roof, also, extends proportionately farther forward. The narrow bar of *eartilage (Pl. 8. figs. 1 and 3, z), which forms the hinder edge of the hyoglossal foramen in the female, is in the male partially replaced by ligament (PI. 8. figs. 4, 5, and 6, 7). This is a matter of the highest importance from a morphological point of view, for the substitution has the effect of severing the true basal plate of the hyobranchial skeleton from its lateral out- growths in such a way as to make itappear part of the laryngeal skeleton. The basal plate, already so reduced in size, is now separated from all the rest of the hyobranchial skeleton, except the thyrohyals. In it Henle (19. p. 17) has described a hexagonal ossified plate, evidently a further development of the small centre of ossification occurring in the same position in the female. I must confess that in the specimens which I have had the opportunity of examining I have found no trace of ossifica- tion in this region. Whether my specimens were too young, or whether Henle’s specimen was abnormal I cannot say,—the bony plate which he figures is so large in comparison with the remainder of the larynx that it is hardly likely that his observa- tion is at fault. The thyrohyals are slightly curved, although straight in the female; and, owing to the lateral expansion of the laryngeal skeleton proper, they appear to lie more on the ventral side of the larynx than in the female (see PI. 8. figs. 1,3,4and6). It thus happens that they are not visible ina dorsal view of the skeleton, whereas in the female very little of the thyrohyal is overlapped by the cricoid cartilage (see Pl. 8. fig. 3). The floor of the LINN. JOURN.—ZOOLOGY, VOL. XXVI. 9) 66 MR. W. G. RIDEWOOD ON THE HYOBRANCHIAL larynx of the male is complete *, the membranous spaces which occur between the anterior ends of the thyrohyals in the female being filled up by cartilage. The portion of the side-wall of the larynx situated externally to the arytenoids is much more massive in the male; and, whereas in the female the blinker- shaped process terminates anteriorly in an abrupt edge, it fuses in the male with the true basal plate of the hyobranchial skeleton (Pl. 8. fig. 6, 61.). The ericoid cartilage is very much thicker than in the female, and the thickening has proceeded irregularly into the interior of the larynx (PI. 8. fig. 5) in a manner recalling the formation of the turbinal ingrowths of the nasal capsule in higher vertebrates. The general impression is one of solidity and massiveness, in striking contrast with the delicacy and elegance of the female larynx. A comparison of median sections of the male and female larynx (Pl. 8. figs. 5 and 2) shows that most of the differences are those of degree and not of kind; but there is one most important structural dissimilarity to be noted. The posterior processes of the arytenoid cartilages of the male are confluent with one another (Pl. 8. fig. 6, ar.'). The union is brought about by a thin horizontal sheet of cartilage, situated immedi- ately behind the glottis, and dorsal to the broad, flat, articular surfaces by which, as in the female, the two massive parts are applied to one another in the median plane (see fig. 5). The relations of the connecting cartilage are such as to warrant the assumption that there is here a procricoid cartilage, not free as in the female Pipa(Pl. 9. figs. 2 and 3, pe.), but united on either side with the arytenoids. This is certainly the most plausible explanation of the facts, but it mvolves the curious anomaly, that a procricoid cartilage is present in the male Xenopus and in the female Pipa, while no trace of it can be detected in the female Xenopus and in the male Pipa. The arytenoids proper (Pl. 8. figs. 5 and 6, ar.) stand nearly vertically, and do not slope downwards and outwards as in the female. Henle (p. 18) states that the arytenoids are ossified in the male, but this I am not in a position to confirm. The bronchial cartilages are disposed just as in the female. The bronchi have long been known to be shorter in the male than in the female, but undue stress seems to have been laid on this fact in consequence of the failure to take into account the * Wilder erroneously states it to be entirely membranous (42. p. 292, and Taf. 21. fig. 59). SKELETON AND LARYNX OF XENOPUS AND PIPA. 67 smaller size of the body of the male. An extensive series of measurements shows that the proportion between the length of bronchus and the distance between the snout and cloaca is the same—one-seventh—in both sexes. HYoBRANCHIAL SKELETON OF Pipa americana, FEMALE. The first figures published of the hyobranchial skeleton of Pipa are those of Breyer (4). These figures, correct as far as they go, show, however, none of the details of the skeletal parts lying towards the median line. Mayer (25), in his monograph on Pipa, says very little about the hyobranchial skeleton, and gives no figures of it. Jt is in his paper on Xenopus (26. p. 31) that he mentions the absence in Pipa of the anterior or hyoidean cornua—a fact previously noticed by Meckel (27. p. 229). Henle’s description and figure (19. Taf. 2. fig. 11) of the female are the first upon which any reliance can be placed. Henle agrees with Meckel and Mayer that the anterior cornua are reduced, and states (p. 16) that they are only represented by the pointed cartilage at the front of the hyobranchial skeleton. Parker (33. p. 651) was the first to show that the hyoidean cornua are present in the embryo. He does not figure the hyo- branchial skeleton of the adult, but in the ripe young, just about to quit the maternal pouches (PI. 60. fig. 7), this portion of the skeleton has already acquired most of the characters of the adult. A criticism of this figure will be found on page 110. The figure furnished by Grénberg (18. p. 636, fig. A. iii.), although only an outline text-figure, shows more accurately than any published previously the relations obtaining between the hyobranchial and the laryngeal skeleton in the female. On comparing the hyobranchial skeleton of the female Pipa with that of Xenopus, one recognizes at once the large wings (Pl. 9. fig. 1, a), the greatly reduced basal plate (y#.), and in front of it the large foramen (4) through which the hyoglogsal muscles rise to be distributed to the mucous membrane of the floor of the tongueless mouth. The basal plate and the thyro- hyals (¢) have entered into the formation of the floor of the larynx ; the bronchi are elongated and are supported by bronchial cartilages (67.). The hyobranchial skeleton terminates in front in a tapering rod of cartilage (ch.'), evidently the homologue of the rod-like cartilage (Pl. 8. fig. 1. ch.’) which extends forwards in front of the hyoglossal foramen in Xenopus. The foramen is Be 68 MR. We G- RIDEWOOD ON THE HYOBRANCHIAL more circular in Pipa than in Xenopus, but tends to become elliptical with increasing age. Even in very old specimens the antero-posterior diameter is not vastly in excess of the trans- verse, so that the hole is never so drawn out as in the middle- aged Xenopus. As in this latter genus, the anterior three-fourths of the foramen are closed by a fatty membranous tissue. The antero-lateral processes of Xenopus (Pl. 8. fig. 1, ap.) are missing, and the wings are of an entirely different shape. The stalk of attachment to the axial parts of the hyobranchial skeleton is very narrow. The basal plate is much constricted a short distance behind the hyoglossal foramen so as to be divided into two parts connected by an isthmus (PI. 9. fig. 1,2). The anterior portion forms the hinder border of the foramen, the posterior enters into the constitution of the floor of the larynx, It is from the posterior edge of the latter portion that the thyrohyals arise. Henle states (19. p. 16) that the isthmus is ligamentous as it is in the male; but Grénberg (18. p. 635) has already corrected this error. The hyobranchial skeleton of the old female differs in numerous particulars from that of the female just sexually mature. The foramen, as already pointed out, is elliptical instead of circular, the pointed cartilage in front is longer and thinner, the wings are more incised on their inner margin, while the stalks are narrower and more elongated. Breyer states (4. p. 14) that the “lamina” or wing of the hyo- branchial skeleton of Pipa is longer and narrower in the male than in the female. Meckel (28. Theil iv. p. 338) agrees with Breyer that the wing of the female Pipa is broader than that of the male, but says it is longer also. Mayer (26. p. 32) affirms that the wing is especially broad and strong in the male. The difference in proportion is, however, in my opinion, one of age,. not of sex. LARYNGEAL SKELETON OF Pipa americana. The first mention of the larynx of Pipa was made in 1764, by Fermin, who describes it in the male as the lambdoid bone *; * Fermin, 13. p. 150. ‘x x x deux cavités distinctes et séparées ’une de Yautre par un diaphragme, attaché 4 un Os triangulaire, qui a la forme d’un Q grec, que l’on pourroit appeler l’os Lambdotde. Il est situé au dedans de la: cavité générale, ou sa base se trouve fixée par un fort ligament 4 la partie supérieure du Sternwm; et duquel il déborde un peu. De la base du méme Os- sortent deux ligamens assez forts qui simplantent dans la partie moyenne de- la machoire inférieure.” SKELETON AND LARYNX OF XENOPUS AND PIPA. 69 and the first figures are those of Schneider (37. Tab. 2. figs. 4, 7-11). The specimen figured,a male, was in a bad state of preservation, and this it was in all probability which led Schneider into the mistake of regarding the larynx as a con- stituent of the sternal skeleton (37. pp. 261 and 263). The larynx of Pipa was first described in any detail by Breyer (4), who gives figures, rather poor, of both sexes. A much better description, referring principally to the male larynx, was furnished in 1825 by Mayer (25), but without figures. In his paper on Xenopus, ten years later, he gives a figure (26. Taf. 3. fig. 5) of the larynx of the male Pipa, broken open to show the elongated arytenoids lying within. The first figures show- ing the detailed structure of the larynx we owe therefore to Schneider and Mayer, and not to Henle as stated by Grénberg (18. p. 634), who appears not to have seen the ‘ Historia am- phibiorum’ nor Mayer’s second contribution. Much valuable information is to be gathered from Henle’s careful description and illustrations of the larynx of both sexes. The study of the larynx of Pipa was again undertaken by Groénberg in 1894, and the eight figures that he gives (18. Taf. 38. figs. 10-17) are excellent. He treats of both sexes, but chiefly of the male. Female. The larynx of the female Pipa is smaller in proportion to the size of the whole hyobranchial skeleton than in the Xenopus female. It is also relatively shorter in an antero-posterior direction, but the length increases with age. The only ossified parts are the thyrohyals (PI. 9. fig. 1, ¢), which are very much shorter, broader, and flatter than in Xenopus, and are situated in the floor of the larynx and not at the sides. Their anterior ends are nearly in contact, and the posterior epiphyses are indis- tinguishably fused with the postero-ventral band of the cricoid cartilage. The space between the bones themselves is closed by membrane, a fact which Gronberg in his excellent paper fails to notice, although Meckel (28. Theil vi. p. 451) and Henle (19. p- 19, and Taf. 2. fig. 11. *.) had already pointed it out. Breyer states that the ossified bars are absent in the young female. His figure (4. Tab. 2. fig. 5) of the adult female larynx he subscribes “ Larynx femina a parte inferiore cum laminis OSsels,” but he most unfortunately locates the bones on the roof of the larynx in his text (p. 15), “‘ lamina inferior mere cartilaginea est; 70 MR. W. G. RIDEWOOD ON THE HYOBRANCHIAL lamina superior pariter cartilaginea duo tamen exhibet ossicula plana,” etc. This error was exposed by Meckel in 1833 (28. Theil vi. p. 451). It may be well to point out here that, while Henle describes the orientation of parts with the toad standing on its hind legs in an erect, human position, Breyer and most writers place the animal belly downwards. What, therefore, in this paper is called the roof of the larynx, is the ‘lamina superior’ of Breyer, the ‘dorsale Wand’ of Grénberg, and the ‘hintere Wand’ of Henle. Wilder (42) avoids the difficulty by speaking of the ‘ pharyngeal’ and ‘ cardiac’ surfaces. The roofing cartilage of the larynx has a deep angular notch posteriorly and an approximately semicircular one anteriorly. The incisions are so deep that a median longitudinal section of the larynx shows only a very small extent of cartilage in the roof (Pl. 9. fig. 3, c). Into the anterior space there fits a thin unpaired plate of cartilage, which overlaps the hinder portions of the arytenoids. This plate (figs. 2 and 3, pc.) is probably to be regarded as a dismembered part of the ericoid cartilage. Its nearest representative, Henle points out (19. p. 43), is the small cartilage similarly placed in some Chelonia, (see Henle, 19. Taf. 5. figs. 28 and 29, d). Dubois (9. p. 181, and fig. 5) calls this latter the procricoid, and the name may safely be extended to the cartilage under consideration in Pipa. The arytenoid cartilages are rather larger in proportion to the size of the larynx than in the female Xenopus, and the process (Pl. 9. figs. 1 and 2, d) to which the tendon of the dilator muscle is attached extends outwards considerably beyond the outer margin of the thyrohyal; whereas in Xenopus the whole of the arytenoid is confined to the space between the two thyrohyals (cf. Pl. 9. figs. 1 and 2, and Pl. 8. fig. 3). The anterior parts. supporting the sides of the glottis are thin; the posterior parts underlying the procricoid, although touching one another in the median line, are not massive as they are in Xenopus, nor do they articulate by broad flat surfaces. A median section of the larynx shows the existence of a firm fibrous mass (fig. 3, ve.) covered by a thin mucous membrane, and attached to the mesial surface of the arytenoid cartilage. Its posterior edge is free, and although the tissue is not as elastic as the vocal cord of tongued Anura, there can be little doubt that the two structures are homologous. It may be said of the male Pipa and of both sexes of Xenopus, that vocal cords are absent, but this statement should be SKELETON AND LARYNX OF XENOPUS AND PIPA. (pl qualified in respect of the female Pipa. The median section also demonstrates the small extent of cartilage in the median line of the roof (fig. 3, ¢) and floor (c’) of the larynx. Owing to the close approximation of the anterior ends of the thyrohyals and the presence of the membranous space in the floor of the larynx there is less difficulty than in Xenopus in recognizing the mor- phological posterior edge of the basal plate of the hyobranchial skeleton (figs. 1 and 3, wv). The fenestra seen in the side-wall of the larynx of Xenopus does not occur in Pipa. The bronchi are elongated and are supported by bronchial car- tilages. These cartilages have the form of independent bronchial rings, and are not, as they are in Xenopus, united into a band running lengthwise along the bronchus. The bronchial skeleton is also more complete in Pipa, the rings extending three quarters of the way round, while in Xenopus the lateral projections of the longitudinal bronchial cartilage do not extend more than halfway round. And, what is still more important, the bronchial rings of Pipa are incomplete on the mesial side, whereas in Xenopus it is the external wall of the bronchus that is not supported by cartilage. Male. The larynx of the male Pipa is unique and unparalleled not only among the Anura but in the whole animal kingdom. It is almost entirely ossified, and thus contrasts forcibly with the larynx of the male Xenopus which, although more massive and inflated than that of the female, is certainly not more bony. The larynx of the male Pipa is about twice as long as that of the female—five or six times so according to Mayer (25. p. 540), but this computation is excessive. It is proportionately broader in front, so that in a dorsal or ventral view it has roughly the form of a rectangular cushion or pillow. The floor is continued forwards into a pointed process of unossified cartilage, which is attached by ligament (Pl. 9. fig. 5, 2’) to the middle of the transverse bar forming the posterior boundary of the hyoglossal foramen. The cartilaginous process at the front of the larynx und the bar bounding the back of the foramen are the posterior and anterior portions respectively of the basal plate of the hyo- branchial skeleton. The ligament connecting them has the same morphological value as the cartilaginous isthmus of the female (Pl. 9. fig. 1, 7). It must not be confounded with the paired ie MR. W. G@. RIDEWOOD ON THE HYOBRANCHIAL ligament (Pl. 8. fig. 6,7) of the larynx of the male Xenopus, although the physiological significance is probably the same. The ligament in Pipa connects the anterior and posterior portions of the basal plate, but in Xenopus the basal plate is not subdivided, but is incorporated bodily into the constitution of the larynx, while the ligaments arise from its anterior end. Except in the ligamentous nature of the isthmus the byobranchial skeleton of the male Pipa does not differ materially from that of the female. The laryngeal walls are remarkably complete; the roof (Pl. 9. fig. 4) extends nearly as far forwards as the floor, and the notch in its posterior border is but slight. The floor has a small but deep notch behind (PI. 9. fig. 5) and, a short distance in front of the notch, a pair of small fenestrae; but beyond these it is one expanse of bone. The greater part of the floor is formed by the flattened thyrohyals, which can be distinguished from the rest of the bone by their yellower colour and closer texture. The two thyrobyals actually touch one another anteriorly, and the mem- branous area in the floor of the female larynx is here closed by a narrow strip of ossified cartilage. Along the sides of the larynx the junction of the cricoid and thyrohyal is marked by a very conspicuous groove (fig. 6). The roof of the larynx is marked by a cruciform area in which the ossification is less complete than elsewhere. Grénberg (18. Taf. 38. fig. 11) represents this in his figure as actually cartilaginous. Although Gronberg was unable to consult Henle’s treatise on the larynx, and had to remain content with the copies of the figures of that work published in Bronn’s ‘ Klassen und Ord- nungen’ (20), he detected that Henle had confounded the dorsal and ventral surfaces of the larynx of the male Pipa. Curiously enough, Henle (19. p. 20 footnote) charges Breyer with doing exactly the same thing. ‘“ Rudotphi hat die hintere und vordere Fliche verwechselt.”” Asa matter of fact, Breyer’s description is correct and Henle himself was wrong. He had probably observed that Breyer had described the thyrohyals on the wrong surface of the female larynx, and became afterwards confused as to the sex for which the description was erroneous. The two bronchi open together close to the median plane, by a pair of apertures which occupy only a small proportion of the posterior laryngeal wall. The bronchial cartilages are arranged as in the female. The bronchi of Pipa have since the SKELETON AND LARYNX OF XENOPUS AND PIPA. 73 time of Breyer (4. p. 15) been known to be longer in the female than in the male, but the difference seems to have been much exaggerated. Grénberg (18. p. 638) puts the length of bronchus of the male at 15 to 20 mm., and that of the female at 25 mm., and my own measurements accord tolerably well with these. The arytenoids are even more remarkable than the rest of the laryngeal skeleton. The glottis, which is small in proportion, is situated quite at the front of the pillow-shaped larynx. It is bounded by small arytenoid cartilages (PI. 9. fig. 5, ar.) which are continued outwards and backwards as relatively huge bones (ar.'), each provided with four articular surfaces and two pro- cesses for muscle-attachment. The bulk of each bony mass has the form of a rod with a very irregular warty surface. The rods extend the full length of the laryngeal cavity: their posterior ends are bluntly poited and lie over the bronchial apertures. They are composed of a very hard and brittle bone, and when broken across exhibit a central whiter and softer part. Their extreme length is about 22 or 23 millim. What renders these rods the more remarkable is the fact that there is not the least suggestion, anticipation, or foreshadowing of them in the female. The posterior parts of the arytenoids of the female Ppa are insignificant even when compared with those of the female Xenopus. Henle has already figured and described * with great minuteness the configuration of these bones, so that it will here suffice to state that they articulate with one another by two pairs of processes, one towards the dorsal and the other towards the ventral side, and that each articulates by a large flat surface with the roof (figs. 4 and 5, v), and by a ball-and-socket joint with the floor of the larynx. As a consequence of these peculi- arities of articulation the rods lie close up to the roof, and are raised somewhat above the floor. The fibrous tissue which occupies the position of the vocal cords is less conspicuous than in the female, but the arytenoid cartilages are distinctly hollowed out, and are thickened along those edges that bound the glottis. Although there is a very sharp line of demarcation between the cartilage and bone, the two are perfectly continuous and move * Tn referring to Henle’s description it must not be forgotten that he had confounded the dorsal and ventral surfaces, and that in his figures (19. Taf. 2. figs. 16-20) the words ‘vorn’ and ‘hinten’ must be transposed throughout. In Mayer’s earlier work (25. p. 541), where a less detailed description is given, the orientation of parts is correct. 74 MR. W. G. RIDEWOOD ON THE HYOBRANCHIAL together. I fail to see what advantage is to be gained by giving distinct names to the cartilaginous and ossified portions as Gronberg does (18. p. 636). The hinging of the rods is such that they can swing in the horizontal plane, and the free posterior ends are capable of moving through a distance of about 3°5 millim. each. The hinder parts of the rods can touch one another in the median plane, as may be seen by manipulation, and more conclusively by the flatness of the more projecting parts of their mesial surfaces, where the impact, of which more anon, normally takes place. They are, however, prevented from crossing the middle line by a forwardly directed process of cartilage (Pl. 9. fig. 5, er.) standing up from the inner surface of the posterior part of the floor of the larynx. A similar crest, of much smaller size, hangs down from the roof. The two rods are notched internally, so that they can strike one another without encountering these crests. The rods are actuated by special dilator muscles attached to outstanding processes (Pl. 9. fig. 5, d) of the parts situated anteriorly to the centre of oscillation (v). On the contraction of the muscles the anterior parts of the rods are divaricated, and the longer posterior portions are brought into contact. Although Miiller (30. p. 222) has compared the rods with the two prongs of a tuning-fork, most authorities liken them to the clapper of a bell. Mayer, for instance, states (25. p. 542) that they are disposed in such a manner, “dass sie an die innere Wandung der Hohle desselben, wie der Schwengel in der Glocke, anschlagen kdénnen.” Henle erroneously regarded the bones as fixed in position and unable to swing bodily to and fro, and concluded (19. p. 31) that the sound uttered must be due to molecular vibration. Grénberg (18. p. 637) exposed the fallacy of this statement and showed that, although the rods do not swing about fortuitously at every movement of the animal, they are nevertheless capable of motion, and that they are controlled by special muscles attached to them. A study of the muscular mechanism has convinced me that the rods do not beat against the laryngeal walls as a clapper strikes a bell, but that they strike one another in the manner indicated above. It is evident that the larynx is thus curiously modified for the production of sound, and the sound to be expected on a priori reasoning is a series of short, sharp taps caused by the impact of the rods, following one another as quickly as successive contractions of SKELETON AND LARYNX OF XENOPUS AND PIPA. 75 the muscles can be effected,—a sound similar to, though probably more rapidly repeated than, the taps given by certain wood- boring beetles. Having arrived at this conclusion, it was most gratifying to me to hear Mr. Arthur Thompson, when recounting at a recent meeting of the Zoological Society his observations on the habits of the Surinam toads then breeding in the Society’s menagerie, mention, quite incidentally, the rapping noise which they made. All those anatomists who have paid any attention to the subject have remarked on the failure of anyone to hear the sound, which they were convinced must emanate from such a complicated laryngeal apparatus. Gronberg, in his recent contribution, says (18. p. 637) “es wiirde von grossem Interesse sein, zu erfabren, ob Pépa wirklich eine diesem Apparat entsprechende Stimme hat. Mein Freund, Freiherr A. von Klinckowstrom, der lebende Wabenkréten in Surinam oftmals beobachtet hat, will niemals einen Laut von ihnen gehort haben.” And even in the current year Wilder writes (42. p. 291) that the matter “is deserving of careful investigation from the side of the physicist, as well as that of the naturalist.’ Mr. Thompson’s remarks were there- fore most opportune. On further inquiry I gathered that the sounds were heard by many visitors tc the Society’s Gardens, who variously described them as resembling the tapping of a key on the glass of the tank in which the animals lived, the striking of two chisels together, and the distant sound of a bricklayer’s trowei. All were agreed that there is a metallic ring about the sound, and that two, three, or four taps follow one another in quick succession,—then, after a pause there comes a repetition, and so on throughout the day and night. There is at present insufficient evidence that the noise is an accompaniment of the amatory overtures, but it is worthy of remark that these animals were mute before the breeding period, and have since relapsed into silence. Mr. Thompson stated it as his firm belief that both sexes shared in the clamour and that their voices were indistinguishable; but, personally, I fail to see how such a tapping sound could possibly emanate from the female larynx, where there is no backward growth, enlargement nor ossification of the arytenoids. At this same meeting of the Zoological Society Mr. G. A. Boulenger, F.R.S., stated that the Xenopus during the breeding period utters a sound which he compared with that produced by 76 MR. W. G. RIDEWOOD ON THE HYOBRANCHIAL pushing the wet finger across a pane of glass. On referring to the structure of the larynx, it is evident that the sound must be produced by the vibration of the arytenoid cartilages or the edges of the glottis, since there are no vocal cords. Judging by the similarity in the shape of the arytenoids in the two sexes, it is probable that there is but little difference in the voices of the male and female. LaryncEAL Muscues or Xenopus levis. In view of the fact that the laryngeal skeleton is so dissimilar in the male and female of the two genera of the Anura Aglossa, it is not surprising that the muscles in relation with it should also differ in a remarkable degree. The following pages contain an account of the laryngeal muscles of both sexes of Xenopus and of the female Pipa. The specimens of male Pipa at my disposal were all unsuited for an examination of the muscular system, but I regret this the less as, thanks to Grénberg, our information on the laryngeal muscles of the male Pipa is not deficient. The object of this portion of the investigation was not only one of correlation of the various muscle-masses occurring in the two sexes of the same species, and of comparison of the muscles of Pipa with those of Xenopus and the tongued Anura, but of the search after additional evidence of the homology existing between the parts of the laryngeal complex of the two aglossal forms and those of the hyobranchial and laryngeal skeleton of more normal batrachians. As has already been pointed out by Henle (19. p- 24), muscles running normally from the hyobranchial skeleton to the larynx become in Xenopus and Pipa intrinsic laryngeal muscles, owing to the annexation by the larynx of a part of the - hyobranchial skeleton. And some of the trunk muscles attached in other Anura to the hyobranchial skeleton here become extrinsic laryngeals. The first account of the laryngeal muscles of Xenopus we owe to Mayer (26. p. 80). His description, which refers to the male, is unaccompanied by illustrations. Henle followed, four years later, with observations on the female, giving figures of the intrinsic muscles of the larynx. Beddard has recently described the muscular anatomy of both Pipa (1) and Xenopus (2) in greater detail than had been done previously; but he does not allude to the laryngeal muscles of either genus. The information furnished recently by Wilder concerning the laryngeal muscles of Xenopus SKELETON AND LARYNX OF XENOPUS AND PIPA. 77 is of a very fragmentary and unreliable character, but he excuses himself (42. p. 309) on the ground that his only material con- sisted of a series of sketches drawn from a specimen dissected several years before. His figures (42. Taf. 21. figs. 58 and 61) are presumably reproductions of these sketches. Female. Dilator laryngis (P1. 10. fig. 2, d).—This dilator muscle of the glottis, called by Henle (19. p. 27 and Taf. 2. fig. 5, m) the ‘unterer Hrweiterer,” arises from the external and dorsal surfaces of the posterior three-fourths of the thyrohyal, and partly also from the roof of the larynx. None of the fibres arise from the floor; a point of some importance when considering sexual differences. The anterior end of the muscle runs into an ex- tensive aponeurosis, the fibres of which pass directly inwards, in a direction transverse to the long axis of the larynx, and, after passing over the ‘ blinker-shaped’ process of the laryngeal car- tilage (Pl. 8. fig. 3, 0/.), are inserted into the dorso-external edge of the massive posterior enlargements (ar.’) of the arytenoid cartilage, and into the posterior part of the arytenoid (ar). guarding the glottis. The anterior part of the insertion of this muscle is not tendinous. No great difficulty besets the determi- nation of this muscle, for, since it rungs from the hinder part of the thyrohyal to the outer surface of the arytenoid, it evidently corresponds with the dilator laryngis * of other Anura. The part of the muscle that arises from the roof of the larynx (Pl. 10. fig. 2, d') also spreads on to the thyrohyal, beneath the greater belly of the dilator. It is separable from the latter, and its tendon, which is about as long as the muscle itself, passes along the dorso-internal border of the thyrohyal, and, running in a groove between the roof and side of the larynx proper as around a pulley, is attached to the postero-externo-dorsal corner of the arytenoid enlargement. The relations are not such as to warrant the recognition of this as a distinct muscle. That the dilator muscle should spread from the thyrohyal inwards on to the roof of the larynx is not exceptional, for in the frog a few of the fibres of this muscle arise from the cricoid cartilage * Dilatateur, Saint-Ange, 36. p. 421, and Pl. 26. fig. 3’, d’. Hyo-ex-glottique, Dugés, 10. p. 126, and Pl. vii. (bis), fig. 47, no. 26. Oeffner des Stimmladeneingangs, Henle, 19. p. 23. Dilatator laryngis, Ecker, 12. p. 315, and Goppert, 16. p. 63. 78 MR. W. G. RIDEWOOD ON THE HYOBRANCHIAL (see 12. p. 313). Wilder figures in Rana clamitans (42. Taf. 20. fiz. 32, ¢’) an accessory slip of the dilator arising from the laryngeal cartilage instead of the thyrohyal,an exact counterpart of this slip in Xenopus. Dilator laryngis anterior (Pl. 10. fig. 2, Aphis muscle, the ‘‘oberer Erweiterer” of Henle (19. p. 27, and Taf. 2. fig. 5, m’), arises from the external surface of the anterior part of the thyro- hyal and extends over rather less than one half its length. It is inserted into the dorso-external surface of the arytenoid car- tilage, in a line passing obliquely across the compressor muscle (c), which is peculiarly modified in consequence. No sharp line of separation can be distinguished between this muscle and the dilator just described, with which its fibres form a continuous series. So far as can be made out, the junction is oblique, the hinder part of the dilator anterior overlapping the front part of the dilator. Both muscles run from the thyrohyal to the aryte- noid, and the insertion of the dilator anterior is in a continuous line with that of the larger muscle (see the dotted line on the left side of fig. 2, Pl. 10). The fibres of the two muscles com- mingle externally as they arise from the thyrohyal, so that were it not for the greater distinctness and more marked individuality of the muscle under consideration in the male Xenopus and in the female Pipa, there would not be sufficient evidence to justify the conception of it as a separate muscle worthy of a distinctive name. It is a muscle which has arisen in all probability by the separation of the anterior part of the normal dilator laryngis, and may therefore be designated the “ dilator laryngis anterior.” The only other determination which is at all plausible is that it represents the “ constrictor” muscle* of other Anura, which has shifted back along the side of the glottis instead of uniting with its fellow of the opposite side in front. The constrictor normally arises from the thyrohyal and runs on the anterior side * Hyo-pré-glottique, Dugés, 10. p. 126, and Pl. vii. (bis), fig. 47, no. 25. Verengerer des Aditus laryngis, Henle, 19. p. 24 and Constrictor, 19. Taf. i. fig. 42, 2. Constrictor aditus laryngis, Ecker, 12. p. 315. Peri-arytenoideus ventralis, Wilder, 41. Hyo-laryngeus, Goppert, 16. p. 63. Constrictor laryngis, Wilder, 42. Taf. 20. fig. 31, es. Saint-Ange (36) does not distinguish between the constrictor and the dilator. He calls them both “dilatateurs” (p. 421), and marks them d and d' in fig. 3', Pl. 26. SKELETON AND LARYNX OF XENOPUS AND PIPA. 79 of the dilator. If this suggestion be rejected the constrictor muscle is not represented in Yenopus*. Compressor glottidis (Pl. 10. fig. 2, ¢).—Lying on the right and left sides of the glottis and extending a short distance behind it is a muscle which I take to represent the compressor glottidis + of the frog, combined in all probability with the outer or posterior compressor {. Henle calls it the “ Compressor der Stimmlade” (19. p. 27, and Taf. 2. fig. 5, p). In front it is attached, close to the median line, to the upper surface of the basal plate of the hyobranchial skeleton at a short distance behind its anterior border (Pl. 8. fig. 1, 7). The muscle passes directly backwards, external to the arytenoid, and unites, behind the glottis, with its fellow of the opposite side in a linea alba, some of the fibres being inserted directly into the glottidean portion of the arytenoid cartilage just at its junction with the massive posterior enlargement. A broad sheet of fibrous tissue extends back from the hind end of the muscles and linea alba to be inserted into the front edge of roofing cartilage of the larynx. It is probably to be regarded as an aponeurotic extension of the muscle itself. The indirect connexion thus established between the muscle and the cricoid cartilage is not without parallel, for Goppert states (16. p. 63) that in the common frog some of the fibres of the external * Henle definitely states (19. p. 26) that the constrictor is absent in Pipa: he makes no mention of it in Xenopus. The constrictor is absent in Bom- binator (Henle, 19. p. 24, and Wilder, 42. Taf. 21. figs. 46 and 47), a fact of some importance considering how closely the hyobranchial and laryngeal skeleton of this toad resembles that of Xenopus. Wilder has remarked its presence in Adyzes, and notes that it occupies an exceptionally ventral (cardiac) position. Gronberg has recorded its occurrence in the male Pipa, but con- cerning this see pp. 90 and 91. t Compressor, Henle, 19, p. 24, and Taf. i. fig. 42, p’. Hoher, medianwarts liegender Constrictor, Ecker, 11. fig. 14, Co./. Hyo-arytenoideus anterior, Ecker, 12. p. 315, and fig. 205, Co./. Sphincter anterior, Goppert, 16. p. 63. Compressor laryngis, Wilder, 42. Taf. 20. fig. 31, cp. t Compressor, Henle, 19. p. 24, and Taf. i. fig. 42, p'. Tiefer constrictor laryngis, Ecker, 11. fig. 14, Co./’. Hyo-arytenoideus posterior, Ecker, 12. p. 315, and fig. 205, Co./’. Sphincter posterior, Goppert, 16. p. 63. Sphincter dorsalis, Wilder, 42. Taf. 20. fig. 31, sfd. Dugés does not discriminate between the two muscles, but applies the term hyo-post-glottique to both, 10. p. 126, and PI. vii. (bis), fig. 47, no. 27. Saint-Ange calls them both constricteurs, 36. p. 421, and Pl. 26. figs. 3’, 4’ and 5’, ¢ and ¢’. 80 MR. W. G. RIDEWOOD ON THE HYOBRANCHIAL division of the compressor (his “ sphincter posterior ’’) arise from the cricoid cartilage, a fact which Henle had previously recorded (19. p. 24) but ina somewhat unsatisfactory manner. The middle of the compressor muscle is tendinous and closely bound to the external face of the arytenoid over a narrow area extending backwards and outwards from the middle of the length of the glottis. While the dilator anterior muscle is to all intents and purposes connected with the arytenoid cartilage, it would be more correct to say that it is inserted into the tendinous middle tract of the compressor, which in its turn is bound to the ary- tenoid. For, when the compressor is removed bodily from the skeletal parts, the dilator anterior muscle is found to be still very tightly bound to it, showing that the fibres do not simply decus- sate. The same treatment also shows that the two bellies, anterior and posterior, of the compressor cannot well be regarded as distinct muscles. Under any circumstances, they cannot be individually homologous with the two muscles running along- side the glottis in the frog, although the digastric compressor of Xenopus is probably equivalent to the two muscles of the frog taken collectively. The muscles in the immediate vicinity of the glottis in Anura. are subject to extreme variation, even in genera which are ad- mitted to be closely allied. Henle made a comparative study of these muscles, and Wilder has latterly reconsidered the subject. Wilder acknowledges that the study is a most confusing one;, but he has arrived at the conclusion (42. p. 308 ef seq.) that the primitive periglottidean muscle in Anura is a sphincter surround- ing the arytenoid cartilages and unconnected with the thyrohyals.. The muscle may remain as such unchanged (Bombinator*), or part of it may become differentiated into a compressor and a constrictor, while part remaius as a sphincter (Lana); or the entire muscle may differentiate into compressor and con- strictor (Bufo). The compressor and constrictor both acquire a secondary connexion with the thyrohyal. Of Xenopus he says (42. p. 318), “The sphincter in Dactylethra appears very simple, divided into four quarters by median and lateral raphés,” and he notices that the “lateral raphés appear to be connected with the * Henle is in error in describing (19. p. 24) a connexion between the com- pressor muscle and the hyobranchial skeleton in Bombinator and Discoglossus. I have examined the muscle in these genera and find it to be an intrinsic laryngeal in both. SKELETON AND LARYNX OF XENOPUS AND PIPA. Sl method of insertion of the Dilatator.’”’ He fails, however, to note that the anterior ends of the sphincter muscles are attached, not to the arytenoid cartilages or one to another, but to the portion of the hyobranchial cartilage between the anterior ends of the thyrohyal bones. This fact leads me to think that the periglotti- dean muscles of Xenopus are not so simple as he makes out, and that they do not represent an undifferentiated sphincter such as occurs in Bombinator. Wilder says (42. p. 307) that in a late larval stage of the tadpole of Rana the as yet undivided sphincter attaches itself to the inner edges of the two thyroid processes. This occurs before the constrictor and compressor have been differentiated, and it is just this stage of development which, I take it, persists in Xenopus. The periglottidean muscle of Xenopus may therefore be regarded as including an undiffer- entiated constrictor; and the absence of any muscle which can with certainty be regarded as the constrictor laryngis lends support to the view. Hyoglossus.—Turning now to the ventral aspect of the larynx, three important muscles are to be noted :—the hyoglossus, the geniohyoideus internus, and the petrohyoideus. The hyoglossus (Hyoglossus, Henle, 19. p. 26) consists of three separate bellies, the external of which (hyoglossus externus, Pl. 10. fig. 1, h.e.) arises from the ventral surface of the thyrohyal at about the middle of its length; while the middle division,—which Henle (19. p. 26) erroneously declares to be absent in Xenopus, although he mentions it as occurring in Pipa,—takes its origin from the floor of the larynx at about one-third of the length from the anterior end (fig. 1, h..). Mayer (26. p. 30) calls the hyoglossa muscle of Anura the ceratoglossus, and states that it appears to be wanting in Xenopus,—clearly an error of observation. He considers it equivalent to the muscle which he had previously called the laryngoglossus in Pipa (25. p. 537, and fig. 2, e). The middle division of the muscle, the hyoglossus internus, is really double, and the two halves are at their origin perfectly distinct. But anteriorly they unite in the median plane, and at the front of the larynx they are inseparable. The external divisions arising from the thyrohyals, although closely applied to the internal division, do not fuse with it. The three muscle- trunks, after running forward ventral to the laryngeal cartilage and dorsal to the m. geniohyoideus internus (Pl. 10. fig, 1, 9.2.), are still readily separable as they pass upward, immediately in LINN. JOURN.—ZOOLOGY, VOL. XXVI. 6 82 MR. W. G. RIDEWOOD ON THE HYOBRANCHIAL front of the basal plate, to be distributed to the mucous membrane occupying the position of the lost tongue. It is on the relations of its anterior end that we have to rely for an identification of this muscle as the hyoglossal, for the origin posteriorly is quite exceptional. The hyoglossus in tongued Anura arises from the posterior extremity of the thyrohyal. In Xenopus, therefore, the muscle has undergone considerable diminution in length, arising as it does from the middle of the length of the bone. The hyoglossus internus is not represented in other Anura, and it is all the more curious to note that in the male Xenopus the normal external trunks have disappeared, leaving only this ex- ceptional median portion. Petrohyoideus (PI. 10. fig. 1, p).—Arising from the posterior part of the ventral surface of the larynx close to the median line, and running out sideways, is a peculiar muscle which repre- sents the petrohyoideus of more familiar Anura*. It occurs in both sexes of Xenopus, in the female Pipa, and probably also in the male. Mayer and Henle are silent concerning it: in fact, throughout the whole of the literature on Pipa and Xenopus that I have examined, I have met with no reference to such a muscle. The muscle spreads beneath the ventral surface of the larynx in a fan-like manner, breaking up into small bundles of fibres which alternate more or less regularly on the right and left sides. The number of radiating bundles is subject to considerable variation, in some specimens being as small as three, while in one larynx examined as many as seven distinct divisions could be recognized on one side and six on the other. The muscle passes out sideways, ventral to the geniohyoideus internus, and diminishes considerably in width. Then, remaining of the same diameter for the rest of its length, it takes a circuitous course and is ultimately attached to the back of the auditory region of the skull. There is, I think, sufficient evidence here for identifying the muscle with one of the posterior petrohyoids of the frog, most probabiy the third or fourth of those described by Ecker (12). Geniohyoideus.—The geniohyoideus has already been described by Henle (19. p. 26) as double in Xenopus. The outer division, geniohyoideus externus, arises from the mandible a short distance from the symphysis, and is inserted into the ventral surface of * Masto-hyoidien, Dugés, 10. p. 125, and PI. vii bis, fig. 44, nos. 20, 21, 22. Stylohyoideus, Walter, 40, p. 36. Petrohyoideus, Ecker, 12, p. 66, and figs. 60, 61, and 205, Poles SKELETON AND LARYNX OF XENOPUS AND PIPA. 83 the ala of the hyobranchial skeleton, at about the level of the anterior ends of the thyrohyals, in the position marked g.e. in fig. 1, Pl. 8. This external geniohyoid of Xenopus is called by Mayer (26. p. 30) the genioceratoideus, but he considers it homologous with the muscle which he had previously named the geniohyoideus in Pipa (25. p. 538). The geniohyoideus internus (PI. 10. fig. 1, 7.7.) (geniohyoideus of Mayer, 26. p. 30) is attached by its anterior extremity to the mandible, nearer to the symphysis than the externus, runs parallel with the latter and extends backwards below the larynx to be inserted into the ventral surface of the posterior epiphysis of the thyrohyal. It runs ventral to the hyoglossus, but dorsal to the petrohyoideus. The two internal geniohyoid muscles are in close contact with one another for the greater part of their length, but they diverge somewhat pos- teriorly. In the frog, the geniohyoideus has a single belly in front and divides posteriorly into two parts. The internal of these (geniohyoideus medialis of Walter, 40. p. 32) is attached to the inner border of the anterior end of the thyrohyal, and the outer (¢. lateralis of Walter, 40. p. 33) to the hinder of the two lateral cartilaginous processes of the body of the hyoid. If we admit that these two divisions of the muscle correspond with the geniohyoideus internus and externus respectively of Xenopus, a perfectly legitimate supposition, we must not fail to notice that the internal division is attached in Xenopus to the posterior epiphysis of the thyrohyal, but in the frog to the anterior end of that bone. Obliquus internus.—The anterior end of the internal oblique muscle, though inserted mainly into the ventral surface of the great wing of the hyobranchial skeleton, sends a small slip with tendinous extremity to the posterior end of the thyrohyal (Pl. 10. fig. 2,0.2.). The incorporation of the thyrohyal into the laryngeal complex of Xenopus demands that this slip should be regarded as alaryngeal muscle, and a passing notice is therefore here accorded to it. Male. In considering the muscles of the male Xenopus, it will be sufficient to note only those points in which they differ from the muscles of the female. The same seven muscles are in relation with the larynx, although they are very much modified in form. The trend of such modification is not in one uniform direction, for, while the dilator muscles are excessively enlarged as com- pared with those of the female, the geniohyoideus internus, the 6* 84 MR. W. G. RIDEWOOD ON THE HYOBRANCHIAL petrohyoideus, and the hyoglossus are considerably reduced. The most remarkable feature is, without doubt, the extensive development of the dilator (PI. 10. figs. 3 & 4, d). This muscle covers the whole of the sides of the laryngeal complex, nearly the whole of the ventral surface, and a considerable portion of the dorsal. It is distinctly differentiated into a superficial and a deeper layer (d and d’), and the tendon of the latter passes for- ward and inward in the pulley-like groove at the side of the roof of the larynx, and is inserted into the dorso-external border of the posterior massive part of the arytenoid. The aponeurosis that passes over the blinker-shaped process of the laryngeal cartilage is very delicate, far more so than in the female. The dilator muscle in the female extends round the outer border of the thyrohyal and very slightly, if at all, on to its ventral surface ; but in the male the belly of the muscle has spread so as to cover the whole of the ossified part of the thyrohyal and to meet its: fellow of the opposite side in the ventral median line (see PI. 10.. fig. 8). The only parts of the ventral surface of the larynx over which the muscle does not extend are, first, a small triangular area in front occupied by the hyoglossus, and, secondly, a larger space, also triangular, at the posterior end between the petro- hyoids. The dilator anterior (P1. 10. fig. 4, d.a.) is better differentiated from the dilator than in the female, although relatively shorter and smaller in bulk. Its fibres are inserted into the arytenoid cartilage over the anterior two-thirds of the side of the glottis. They pass outward and downward, but very slightly backward, and spread out in the form of an aponeurosis over the ventral. surface of the anterior part of the dilator. Wilder (42. Taf. 21.. fig. 61, ¢) identifies this muscle as the “ dilatator laryngis.” He describes it, however, as arising from the thyrohyals, an origin peculiar to the female ; and I cannot help thinking that he was mistaken in the sex of the animal he examined. The hyoglossus: (Pl. 10. fig. 8, 4.7.) is much more feebly developed than in the female, and although the relations of the anterior end are the same, the posterior extremity does not reach nearly so far back. In the female the outer fibres arise from the thyrohyals, but in the male: the whole of the muscle arises from the cartilage of the laryngeal’ floor. The reduced hyoglossus thus probably corresponds with the median belly only (hyoglossus internus) of the muscle in the female. The geniohyoideus internus (Pl. 10. fig. 3, g.t.) 18 more SKELETON AND LARYNX OF XENOPUS AND PIPA. 85 slender than in the female, but exhibits, even in a more marked degree, the same diminution in width in that part of its length which underlies the anterior end of the larynx. Itruns dorsally to the petrohyoideus as in the female, but is separated from the cartilaginous floor of the larynx by the intruded sheet of the dilator. The petrohyoideus (P|. 10. fig. 3, p)is also more feebly developed than in the female. Only the anterior fibres reach the median line, so that a triangular space of bare cartilage is left between the hinder parts of the expanded laryngeal extremities of the muscle. The radiation of the fibres is much more uniform and regular than in the female. The compressor glottidis (Pl. 10. fig. 4, c) does not differ materially from that of the female. Many of the fibres are distinctly attached to the arytenoid car- tilage posteriorly, and the connexion of the muscle with the fascia-sheet covering the dorsal surface of the arytenoid plate (Pl. 8. fig. 6, ar.') is comparatively unimportant. Owing to the feebleness of the anterior dilator, the anterior and posterior bellies of the compressor are less distinct than in the female, and, when the muscle is isolated and examined under the microscope, muscle-fibres can be seen passing from one belly to the other. In the female, the intervening part is wholly tendinous. Laryneeat Muscrzs oF Pipa americana. The first description of the hyobranchial and laryngeal muscles of Pipa we owe to Mayer (25), who includes them in his chapter on the general muscular anatomy of the female. Meckel (28) paid some attention to the subject, as also did Henle (19). Henle’s description applies to the female sex *, and is well illus- trated. More recently, Grénberg (18. pp. 637 and 638) has resumed the study of the subject, but his remarks refer ex- clusively to the intrinsic muscles of the male larynx. His figures (Taf. 38. figs. 18-20) are, however, most valuable, being the only illustrations extant of the laryngeal muscles of that sex. Female. The muscles in relation with the larynx in the female Pipa resemble in the main those of Xenopus. The same arguments, therefore, which have previously been adduced for deteriining * Not male, as stated by Gronberg (18, p. 637). 86 MR. W. G. RIDEWOOD ON THE HYOBRANCHIAL the homologies of the muscles of Xenopus with those of other Anura will apply in the present case, and need not be repeated. The most important feature in which the laryngeal musculature of Pipa differs from that of Xenopus is in the absence of the geniohyoideus internus. Arising from the roof of the larynx in the elliptical area marked a in fig. 5, P].10,is a very loose tissue, mainly connective tissue, but containing numerous scattered muscle-fibres which are inserted into the mucous membrane of the pharynx behind the glottis. No particular direction can be ascribed to the fibres, for they cross one another in a most irregular manner. Radiating outwards and backwards from the postero- external border of the same area are a few muscle-fibres arranged in the form of a thin sheet, attached distally to the internal coneave edge of the great wing of the hyobranchial skeleton. Neither of these tracts appears to be of sufficient importance to warrant the application of a distinctive name. The dilator laryngis (Pl. 10. figs. 5 & 6, d)[Laryngeus proprius, Mayer, 25. p. 538 ; Dilatator aditus laryngis, or Erweiterer, Henle, 19. p. 26, and Taf. 2. figs. 14, 15, & 23, m] arises from the postero-external border of the laryngeal complex, and, considering the large size of the muscle, its posterior attachment is remark- ably small. It occurs partly on the dorsal and partly on the ventral surface of the larynx. Its belly is large and well defined, and narrows considerably anteriorly, and it is inserted by a small tendinous extremity into the most external part of the lateral promontory of the arytenoid cartilage (Pl. 9. fig. 2,d). The dilator laryngis anterior (Pl. 10. fig. 5, d.a.) is very clearly dis- tinguishable from the dilatcr. It arises from the fascia of the external surface of the anterior part of the latter muscle, passes upwards, forwards, and inwards, and gradually diminishes in width. It runs dorsal to the compressor (c) and is inserted into the edge of the arytenoid cartilage guarding the side of the glottis. A few of the fibres, also, are attached indirectly to the arytenoid cartilage by being inserted into the aponeurosis between the anterior and posterior bellies of the compressor. Compressor glottidis (Pl. 10. fig. 5, ¢) [Compressor laryngis, Henle, 19. Taf. 2. figs. 14, 15, p].—The anterior and posterior bellies of this muscle are very sharply differentiated, and the aponeurosis between the two extends obliquely backwards and outwards, exactly as in Xenopus. The aponeurosis is attached by its ventral surface to the anterior edge of the lateral pro- SKELETON AND LARYNX OF XENOPUS AND PIPA. 87 montory of the arytenoid, and by its dorsal surface to some of the fibres of the dilator anterior. Posteriorly, the compressor muscles of the right and left sides unite, immediately behind the glottis, in a linea alba which extends backwards for a distance equal to the width of each muscle. The anterior extremity of the com- pressor is attached to the lateral edges of the median isthmus of cartilage (Pl. 9. fig. 1,2) which connects the part of the hyo- branchial cartilage bounding the hyoglossal foramen with that which forms the anterior end of the laryngeal complex. In this respect the compressor differs from that of Xenopus, for in this latter genus the attachment of the anterior end of the muscle is to the dorsal surface of the equivalent cartilage. The hyoglossus | Laryngoglossus, Mayer, 25. p. 537 and fig. 2, e; Riickwartszieher der Zunge, Meckel, 28. Theil iv. p. 340° Hyoglossus, Henle, 19. Taf. 2. figs. 22 and 23, 4] is a much longer muscle than that of Xenopus, and extends nearly the full length of the larynx. The external and internal divisions also are better differentiated. The hyoglossus internus (Pl. 10. fig. 6, 2.2.) arises trom the ventral surface of the larynx near the posterior end, aud the area of attachment slopes obliquely backwards and out- wards. The muscle crosses dorsally to the petrohyoideus and runs forwards beneath the larynx, diminishing at the same time in width. The internal hyoglossals of the right and left sides converge anteriorly ; they fuse together, pass through the median foramen in the hyobranchial skeleton, and spread out as a single muscle (Pl. 10. fig. 5, 4.7.). This is inserted into that part of the mucous membrane of the floor of the mouth which lies over the beak or point of the hyobranchial skeleton (Pl. 9. fig. 1, ch.’). The external hyoglossus (Pl. 10. fig. 6, h.e.) is in the main slightly broader than the internal, but its posterior area of attachment is smaller than, and lies posteriorly to, that of the latter. The two muscles, internal and external, thus overlap one another at their posterior ends, but they separate out anteriorly. The external trunks occupy a lateral position with regard to the internal as they pass through the hyoglossal foramen. Then, instead of continuing forwards, they turn sharply outwards at right angles to their former course (P1.10. fig. 5, .e.), the left to the left and the right to the right. They spread out and become inserted into the mucous membrane of the lateral parts of the floor of the mouth. Henle (19. p. 25) mentions the three muscles spreading out in the floor of the mouth, but he fails to note the 88 - MR. W. G. RIDEWOOD ON THE HYOBRANCHIAL quadruple nature of the hyoglossal muscle posteriorly. He also states (p. 77) with Mayer (25, p. 537) that the external hyoglossus arises from the thyrobyal or columella. Careful dissection, how- ever, shows that the origin is internal to the cartilage which is to be regarded as the epiphysis of the thyrohyal. The petrohyoideus (Pl. 10. fig. 6, p) arises from the membranous ventral wall of the larynx, close to the median line, at about one- third of the length from the posteriorend. It extends outwards and backwards, coils round the external edge of the great wing of the hyobranchial skeleton, and, passing forwards and upwards, is attached to the dorso-external crest of the auditory capsule, close beside the inner end of the columella auris. It runs ventral to the hyoglossus and the geniohyoideus externus, but dorsal to the obliquus internus. It is a very long and meandering muscle, and is of uniform width all along, except at its laryngeal extremity. It is relatively smaller than in the female Xenopus. Geniohyoideus (Mayer, 25. p. 538; Henle, 19. Taf. 2. figs. 22 and 23, 3).—Henle (p. 26) has already noted that the inner divi- sion of the geniohyoideus, attached to the laryngeal skeleton in Xenopus, is absent in Pipa. The external division is a long, slender muscle, of even diameter, extending from the symphysial region of the mandible to the postero-external region of the ventral surface of the great hyobranchial ala. The position of its insertion is shown at ge. in fig. 1, Pl. 9. The recognition of this muscle in the embryo is a matter of no great difficulty, and as soon as the wing of the hyobranchial skeleton is clearly differ- entiated from the neighbouring parts, the insertion of the muscle is seen to be related to it exactly as in the adult (see P1.11. fig. 8, ge.). Seeing that the wing is purely a derivative of the hypo- branchial plate of the larval skeleton, it would be more satis- factory to substitute for geniohyoideus the word geniohypo- branchialis, the name by which Schulze (88) designates this muscle *n his treatise on the development of Pelobates. It is worthy of remark that, while in Pipa the geniohyoideus externus is attached to the ala near its posterior edge, the position of the insertion of this muscle in Xenopus lies in front of the line joining the anterior extremities of the thyrohyals (see ge. Pl. 8. fig. 1 and Pl. 9. fig. 1). Obliquus internus (Pl. 10. fig. 5, 0.7.).—In the larynx of the young but sexually mature female depicted in fig. 5, Pl. 10, the laryngeal division of the internal oblique muscle is of consider- SKELETON AND LARYNX OF XENOPUS AND PIPA. 89 able size. It partially overlaps the dilator muscle and is inserted into the dorsal surface of the larynx. In older specimens, how- ever, the greater part of the muscle becomes transformed into a fatty substance of a dark brown colour, similar to that which occurs in abundance around the pointed cartilage at the front of the hyobranchial skeleton and on the underside of the ale. Such part of the muscle as remains unchanged is attached to the postero-lateral corners of the laryngeal skeleton, in the position already recorded for it by Mayer (25. p. 536) and Henle (19. p- 26 and Taf. 2. figs. 14 and 23,7’). At no age does the muscle develop a tendinous extremity as it does in Xenopus. Male. For details of the laryngeal muscular system of the male Pipa» I have been obliged to rely entirely on the account given by Gronberg, for both Mayer’s and Henle’s descriptions of these muscles refer to the female, and although I have had access to three specimens of the male, they were all unsuitable for a care- ful examination of their muscular anatomy. Gronberg describes (18. p. 638) four intrinsic muscles of the larynx. For the muscle around the glottis he accepts Henle’s name Compressor: it appears to correspond exactly with that muscle which in the female I have called the compressor, so that it is unnecessary to discuss it further. The other three, however, open up debatable points. They arise from the sides, roof and floor * of the larynx, and extend nearly its full length ; and their bellies are so united that Gronberg confesses his inability to decide whether they are three originally distinct muscles, or whether they are divisions of the same muscle, its tendon having split into three. He elects to adopt the former view. The tendon of the most dorsally-lying muscle (18. Taf. 38. figs. 19 and 20, 7) is attached to the arytenoid cartilage guarding the glottis, and this muscle he identifies as the “dilatator laryngis”’ of Ftana. The tendon of the ventrally- disposed muscle (18. figs. 18 and 20, S 3) runs to the external pro- montory of the ossified part of the arytenoid, and the contraction of the muscle serves to approximate the bony arytenoid rods lying in the laryngeal cavity. Since Gronberg regards these ossified parts as external arytenoids not represented in other Anura, he is * Henle (19. p. 26) states that in the male Pipa the whole of the ventral surface of the larynx is covered by muscle, an observation which is not borne out by Gronberg’s figure (18, Taf. 38. fig. 18). 90 MR. W. G. RIDEWOOD ON THE HYOBRANCHIAL forced to conclude that the muscle also is unique. He does not suggest any name for it. The third muscle overlaps the other two on the external or lateral surface of the larynx, and its tendon (18. fig. 20, S2) unites antero-ventrally to the glottis with its fellow of the opposite side. From these relations he considers it to represent the “ hoher, lateralwirts liegender Constrictor des Larynx” of Ecker (11. p. 31, fig. 14, ea./.) (called in the English edition the “ Constrictor aditus laryngis,” 12. p. 314, fig. 205, c.a.l.). On comparing Grénberg’s figures of the male with mine of the female, it will, I think, be evident at a glance that the dorsal muscle, the tendon of which he marks SZ, is that which I have called the dilator anterior (Pl. 10. fig. 5, d.a.). The tendon in both cases is inserted into the edge of the arytenoid cartilage at the side of the glottis, and it is quite possible to conceive that the muscle has assumed a more antero-posterior direction in the male, in consequence of the glottis opening directly forwards, towards the aperture of the mouth, instead of upwards as in the female. The other two, S 2 and § 3, I take to correspond with my dilator laryngis. Although in both sexes of Xenopus the insertion of the dilator lies entirely behind that of the dilator anterior (Pl. 10. figs. 2 and 4), yet in the female of Pipa the attachment of the tendon of the dilator to the promontory of the arytenoid is situated farther forward than the insertion of the anterior dilator into the side-wail of the glottis (Pl. 10. fig. 5). Assuming, as I think we are justified in doing, that the promon- tory of the female arytenoid (Pl. 9. fig. 2, d) is equivalent to the simular promontory of the male which Gronberg marks a in his figures 10, 11, 13, and 14, it is but logical to regard his ventral muscle, S 3, and my dilator laryngis as identical. This leaves only his muscle § 2 unaccounted for. I have not been able to recognize in the female Pipa any third dilator, nor any union of tendons anterior to the glottis; but in both sexes of Xenopus there is an imperfectly differentiated slip of the dilator (Pl. 10. figs. 2 and 4, d'), the tendon of which is elongated and inserted into the massive part of the arytenoid. It is true that in Xenopus the insertion of the slip is far behind that of the dilator anterior and the glottis, but an examination of fig. 5, Pl. 10, will show that a similar slip differentiated from the more dorsal part of the dilator, on the rotation of the glottis forwards, would come to occupy the position of Grénberg’s 8 2. SKELETON AND LARYNX OF XENOPUS AND PIPA. SE It now only requires an extension of the tendons to meet below the glottis to complete the resemblance. The changes in the relative positions of the bellies of the muscles would follow as a consequence of the rotation of the glottis. The dilator anterior would straighten out and occupy a dorsal position, leaving the greater dilator on the ventral surface of the larynx and its acces- sory slip laterally. As an alternative view, it may be suggested that the ventral muscle, 83, which Gronberg says is absent in the frog, is the equivalent of that curious muscle which in Disco- glossus and Alytes runs from the ventral surface of the posterior epiphysis of the thyrohyal to the ventro-anterior point of the arytenoid cartilage. Wilder figures the muscle in Alytes (42. Taf. 20. figs. 86 and 37) and regards it as the displaced constrictor laryngis, which otherwise must be admitted to be absent. Whether Wilder’s interpretation is correct must remain for future investigation to decide: the two muscles do not unite in front of the glottis as do the constrictors of the frog. These conclusions savour, no doubt, of conjecture, being based on the study of figures only, drawn by two different authors, each of whom has examined the muscular system of one sex only. In spite of the interesting union of the tendons S 2 antero-ventrally to the glottis, | am not disposed towards an unqualified accept- ance of Gronberg’s determination of the muscles as the con- strictors of the common frog. An exhaustive study of the muscles, extrinsic as well as intrinsic, of the larynx of the male and female Pipa would most assuredly repay any investigator fortunate enough to have the necessary material at his command. DEVELOPMENT OF THE HYOBRANCHIAL AND LARYNGEAL SKELETON. The development of the hyobranchial and laryngeal skeleton of the Aglossa is practically unknown. Parker’s treatise on the development of the skull of Xenopus and Pipa (83) is the only work to which we can turn for information; and this does not furnish much. On comparing the hyobranchial skeleton of the tadpoles of Rana, Alytes, Pelobates, and Bufo, one is struck by the uniformity of structure that exists in the different genera during the earlier stages of development; and the observations now to be recorded show that the skeleton of Xenopus and Pipa conforms to the general ground-plan. The various modifications 92 MR. W. G. RIDEWOOD ON THE HYOBRANCHIAL that render it so aberrant appear only during and after meta- morphosis. This is of course the conclusion one would expect to arrive at; but hypotheses, conjectures, and arguments are of infinitesimal value compared with ocular demonstration of fact. Concerning the methods adopted in this portion of the research, extensive preliminary trials were made on the tadpoles of the common frog and of Alytes, with a view to discovering the method which would demonstrate the facts in the most intelligible manner, and yet leave the skeleton in a condition in which it may be preserved for subsequent examination. The easiest methods of preparation are those which fail to satisfy the second condition. The steeping of the body in water at a temperature of about 80° C., a method largely used for the preparation of the skeleton of Elasmobranch fishes, was rejected at once on account of the shrinkage and distortion which it occasions. Putrefactive maceration, which may sometimes be employed with advantage in the preparation of cartilaginous skeletons, was found unsatis- factory on account of the tendency which the delicate skeleton showed to fall to pieces, and the possibility of loose cartilages becoming lost or overlooked—objections from which the hot- water method is not free. As I had previously found the method of staining with borax-carmine and clarifying with clove-oil of the greatest value when studying the carpus and tarsus of Anura, larval as well as adult (23), I applied the process to the branchial skeleton; but had to discard it on account of the resulting brittleness of the tissues. Clarifying by glycerine and caustic potash I also tried; but the preparations made in this way cannot be kept permanently in the clarifying fluid, since the potash brings about the ultimate disintegration of the cartilages. Parker’s method, of staining with an ammoniacal solution of carmine and then steeping in glycerine, is open to less objection than the preceding methods; but it resembles them all in being inapplicable at the critical period when the cartilages are becoming absorbed,—the most important stage of all. All methods involy- ing clarification and the use of transmitted light fail utterly here. I was therefore forced to fall back on the very simple expedient of dissecting the parts under a simple microscope by powerful reflected light: a tedious and delicate operation, it is true; but with fine instruments, combined with a knowledge, to be gained only by experience, of the relative toughness of muscle, peri- chondrium, and cartilage, the difficulties can be overcome. SKELETON AND LARYNX OF XENOPUS AND PIPA. §3 A short trial was given of that method of investigation which is now finding such favour with embryologists, and which has been employed with such excellent results by Gaupp (14) in his mvestigations on the hyobranchial skeleton of Rana Jusca. The head or other part is cut into microscope-sections of known thickness, and the organs to be studied are reconstructed there- from in wax plates, the thickness of which is the same multiple of the original as the linear magnification. The wax sheets are then pressed together, and an enlarged model of the organs is thus obtained. Without in the least wishing to depreciate the results of Gaupp’s investigations, for which I have the greatest admiration, I would point out that at the critical periods, when the cartilages are forming or are becoming absorbed, their outline is extremely difficult to make out in sections, and that in fashioning the wax plates it is very largely a matter of personal opinion to decide where the line shall be drawn between the cartilage and the surrounding tissues; and thus an error of a centimetre or more, according to the magnification, may very © easily creep into the model. Ifthe sections are cut thin, this difficulty of discrimination increases ; while if the sections are thick, it is almost impossible to represent truthfully in the model narrow rod-like structures which make a small angle with the planes of section. The method is even more tedious than that of actual dissection by reflected light, and far less gatis- factory and conclusive. I had already completed this portion of the investigation by the time that Wilder’s paper (42) reached me; and go I did not include a trial of the method he adopted in his recent work on the adult laryngeal skeleton of Amphibia, viz., that of mounting in turpentine and Canada balsam after slow staining with a weak alcoholic solution of methyl-blue, washing, and dehydrating. I have since made a few experiments in this direction, and can testify to the differentiating action which the dye has upon hyaline car- tilage. The method, however, is useless for the demonstration of cartilage which is just forming or is undergoing absorption, and could not therefore be applied successfully to such purposes as the present. Xenopus levis. Of the large number of tadpoles of Xenopus examined, it must suffice to choose only three stages for minute description and delineation, because in the earlier stages of development the 94 MR. W. G. RIDEWOOD ON THE HYOBRANCHIAL hyobranchial skeleton undergoes but little change, and that very slowly ; and because the later stages, at which the metamorphosis is actually taking place, I have been unable, in spite of strenuous endeavours, to procure. I have already delayed the publication of this paper several months in the hope of being able to obtain the latter stages; but as there seems to be no prospect of achieving this in the near future, I submit such results as I have already obtained. For most of the specimens examined I am indebted to the generosity of Mr. G. A. Boulenger, F.R.S. In making comparison of different tadpoles of the same species, I find that actual size is no criterion of the degree of develop- ment, since so much depends on the favourable or adverse con- ditions of life. The carefully recorded measurements of Parker’s Xenopus larve were useless to me for correlating his tadpoles with mine. The only reliable data are afforded by the extent of development of the paired limbs. The oldest of the three chosen stages (Stage IIT.) is characterized by the presence of all four limbs, the fore as well as the hind being well developed, and by the possession of a large tail which shows no signs of absorption. The larve of the second stage have the hind limbs powerfully developed; but the fore limbs are only half as large as in Stace III. The first stage is a most comprehensive one, in- cluding tadpoles with hind limbs well developed and with the fore limbs recently extruded; tadpoles with hind legs not yet markedly flexed and with no trace of fore limbs ; tadpoles with the hind legs just appearing on the surface of the body; and tad- poles with no traces of paired limbs at all. There is no appre- ciable difference beyond one of size in the hyobranchial skeleton of any of these tadpoles of Stage I.; but a trifling variation can be discerned in the relative size of the larynx. In the figures of the developing hyobranchial skeleton (Plate 11) the mandible is included, because it serves to give an idea of the size of the mouth at each stage—a most desirable item of information,—and because it demonstrates the relative huge- ness of the hyoid arch, and shows that while the hyoid and mandibular arches are closely approximated in Xenopus, they are widely separated in the corresponding stages of Pipa. The ossifications in the mandible are not indicated in the figures. Parker shows (33. Pl. 56. fig. 6, and Pl. 58. fig. 4) a pair of ‘ mento- meckelian”’ or “inferior labial” cartilages in the symphysial region of the mandible of the early larva of Xenopus. In this SKELETON AND LARYNX OF XENOPUS AND PIPA. 95 position, however, I find not a pair of cartilages, but a single median piec? which shows signs of division into two only in the third stage of development. This cartilage I take to be an integral part of the mandibular arch—in fact, the median inferior element or basimandibular. An examination of the first stage especially shows that, although the inferior portions of the man- dibular and hyoid arches are so dissimilar in size, they are built essentially upon the same plan, and consist each of a pair of ceratal elements and a median basal piece (see Plate 11. fig. 1). In this respect the mandibular arch of Xenopus exhibits the retention of a most primitive character. The above interpretation of the symphysial cartilage opens up the question of the morphological value of mento-meckelians in general ; and [ am inclined to think that, in Anura at least, the symphysial cartilages or bones are not labials developed, as the upper labials undoubtedly are, in special relation with the suctorial lips, but the modified right and left divisions of a median basimandibular. The mouth of the young Pipa and Xenopus is not suctorial, but has the form of a wide slit from the very first; and this fact it is which gives the young of the Aglossa such a fish-like appearance. I have not been able to satisfy myself concerning the presence of mento-meckelian cartilages in Pipa. The mandible exhibits a symphysial segmen- tation as early as Stage IT. (fig. 5); and at Stage VI. (fig. 9) the two rami are as distinct as in the adult. As already shown by Parker (33. pp. 638, 651, and 655), no symphysial elements are to be recognized in the adults of either Xenopus or Pipa. Stace I. (Plate 11. fig. 1.) Tadpoles ranging from those in which the hind legs are just appearing, to those with the fore legs recently extruded. The branchial skeleton has attained its maximum larval deve- lopment ; but it as yet shows no signs of absorption. On com- paring the hyobranchial skeleton of Xenopus at this stage with that of a more familiar anurous batrachian, e. g. Rana or Alytes, one cannot fail to notice how small are the three branchial clefts in proportion to the large expanse of the branchial cartilage. Instead of four elongated bars of cartilage on each side, con- nected distally by an epibranchial marginal bar, and confluent proximally with a cartilaginous hypobranchial plate, we have rather a pair of greatly inflated cartilages of considerable size, 96 MR. W. G. RIDEWOOD ON THE HYOBRANCHIAL each approximately resembling in shape the half of a longitu- dinally divided egg-shell. The cartilage is thin and fairly uniform throughout, and is perforated ventrally by the three branchial slits sloping obliquely outwards, backwards, and slightly upwards. The two cartilages are closely applied one to another in the median plane for about one-half of their length. Their applied faces are nearly flat, and are readily separable, no fusion having yet taken place. The two cartilaginous basket-works look upwards and slightly inwards, and the inner and anterior edges are involute. Of the three branchial clefts the first is situated about halfway from the anterior end, so that in front of it there is quite a large extent of unbroken cartilage. This latter represents the first branchial arch of the tadpole of the frog. The length of the most anterior cleft is about one-half of the transverse diameter of the basket. The second and third clefts are nearly parallel with the first: they are successively shorter, and lie rather closer to the median line. The two bars of cartilage separating the three clefts are the second and third branchial arches, while the cartilage bounding the last cleft posteriorly represents the fourth branchial arch of the frog- tadpole. Seeing that in fishes the cleft is situated in front of the arch of the same denomination (7. e. that the first branchial cleft is bounded behind by the first branchial arch, and so on), the three clefts here present represent the second, third, and fourth branchial clefts respectively. It may seem unnecessary to insist on this point ; but when three clefts occur, as here, in a continuous cartilage, there is a temptation to speak of them loosely as the first, second, and third, and then arises a danger of losing sight of their morphological enumeration. The inner or pharyngeal surface of the branchial cartilage is not smooth, but is covered with a multitude of minute, blunt, arborescent growths. These are not shown in the figures. There are also sharply defined ridges running along the upper surface of the second and third branchial arches, and continued over the undivided cartilage both outwards, backwards, and upwards, to the external edge, and inwards, forwards, and upwards, to the internal overhanging border. These latter processes are evidently equivalent to those outgrowths in the branchial apparatus of the Rana tadpole which Gaupp calls spicula ii. and iii. (14. Taf. xvii. fig. 1), and to which Parker (84) attaches undue importance by calling them the true branchial arches. The flap-like outgrowth SKELETON AND LARYNX OF XENOPUS AND PIPA. 97 of the antero-external edge (Plate 11. fig. 1, &) is very constant inits occurrence ; and this also is not unrepresented in the genus Rana. The hyoid arch consists of three elements—two large massive lateral cartilages or ceratohyals (fig. 1, ch.), which later will form the anterior cornua of the hyoid apparatus, and a small median cartilage or basihyal (62.). There are thus in the hyo- branchial skeleton at this period five distinct and separable cartilages, as has been held to be the case in the tadpole of the common frog from the time of Cuvier (8) onwards. Gaupp states (14. p. 433) that in Rana fusca tadpoles the whole cartilage is continuous, without fibrous tissue intervening. I have not been able to procure tadpoles of this species; but I have made a careful examination of tadpoles of suitable age of Rana esculenta and &. temporaria, and in both of these the five cartilages are very sharply defined. I am inclined to think that the imper- fections of the method adopted by Gaupp are largely responsible for his departure from the generally accepted view. As men- tioned above, it is very difficult to discriminate between embryonic cartilage and fibrous tissue by their histological characters. The ceratohyal or lateral hyoid bar is a stout massive cartilage closely wedged in between the branchial cartilage behind and the meckelian cartilage in front. The posterior edge is hollowed out superiorly so as to allow the anterior pointed extremity of the branchial basket to overlap it. The anterior edge of the cerato- hyal is thin, and runs parallel with the attenuated mandible, almost in contact with it. The distance between the outer extremities of the hyoid arch is considerably less than the maximum diameter across the branchial skeleton. The two ceratohyals touch one another in front; but the posteriorly directed processes of their median ends are separated by the basihyal, and are wedged in between the divaricated ends of the branchial skeleton. The basihyal is cuneiform and pointed anteriorly. Its posterior surface is applied to the branchial cartilage immediately in front of the thyroid bodies*. I regard this median cartilage as the basihyal because of its intimate relation to the ceratohyals; but Iam quite prepared to agree with Parker (34) that when, as in Alytes, six elements instead of five are present in the larval hyobranchial skeleton, the extra * Tam at a loss to understand why Parker (33. Pl. 58. fig. 3, 7.9.) figures the “‘ thyroid glands” close up under the auditory capsule. LINN. JOURN.—ZOOLOGY, VOL. XXVI. vi 98 MR. W. G RIDEWOOD ON THE HYOBRANCHIAL small cartilage between the ceratohyals is undoubtedly the basi- hyal, and that the larger, and more posterior, unpaired cartilage, which is of more constant occurrence, has rather the value of a basibranchial. Schulze (38) still holds to the original view of Dugés (10), that this latter is the basihyal (basihyoid). The larynx at this stage is quite diminutive in size, and lies above the level of the internal reflected margins of the branchial cartilages. The width of the larynx is hardly more than that of one of the bronchi, and the length of the glottis is about half the width of the larynx. The bronchus is not sharply marked off from the lung, but appears rather as a tubular non-sacculated con- tinuation of it. The bronchus and lung are about equal in length. As might be expected at this early stage, the lungs are very small, and the total length of bronchus and lung together is not more than the maximum transverse diameter of the branchial skeleton. Arising from the dorsal surface of the anterior end of each bronchus is a curious thin-walled sac (Plate 11. fig. 1, p) which runs up laterally to the cesophagus, and is lodged beneath the ribs (see p. 120) of the anterior vertebre. The significance of these sacs it is difficult to estimate: they are probably to be regarded as accessory lobes of the lung, since their walls so closely resemble the lung-tissue. They are present in all the larve of Xenopus that I have examined, but are altogether absent in the adult. I have found nothing to correspond with them in either the young or adult of Pipa. Srace II. (Plate 11, fig. 2.) Tadpoles with fore limbs mode- rately well developed, pigmented, but not markedly angulate. By comparison with Stage I., the most important difference to be noted in Stage II. is the reduction in size of the branchial skeleton. The hyoid arch and mandible show no great change. The curvature of the outline in the dorsal view of the branchial skeleton is no longer seen, but each half is roughly five-sided. The inflation is less conspicuous, and the basket is considerably shallower. The overhanging fold at the anterior end is missing, although the ledge projecting outwards from the applied mesial surfaces still remains. The branchial clefts have not altered ; and they serve to show that the absorption has been greatest anteriorly. The first branchial arch (PI. 11. fig. 2, cb. 1), which before fitted so closely into the concavity at the back of the ceratohyal, now barely overlaps the ventral ledge of that cartilage SKELETON AND LARYNX OF XENOPUS AND PIPA. 99 The ceratohyal hag not shared in this modification, and so a vacuity now exists between its outer end and the branchial skeleton. The anterior portion of the basihyal has been ab- sorbed, and a cleft is beginning to appear between the median extremities of the ceratohyals. The distance between the outer ends of the ceratohyals is now slightly less than the maximum transverse diameter of the branchial skeleton. The larynx, bronchi, accessory pulmonary lobes, and the lungs are all larger than before, but are not otherwise noteworthy. Stace III. (Plate 11. fig. 3.) Tadpoles with well-developed fore limbs, sharply bent. Absorption of the tail not yet commenced. Considerable changes have taken place in the hyobranchial skeleton since Stage II., far more than would be suspected from a superficial examination of the tadpoles. The five cartilages to be seen in Stages I. and II are no longer distinguishable. The basihyal has either been absorbed, or has fused up with the neighbouring cartilages in such a way as to leave a deep cleft, enlarging posteriorly between the median ends of the ceratohyals. The posterior end of this cleft marks the position of the future hyoglossal foramen; and it only requires a further enlargement of the cleft, accompanied by a secondary union of the ceratohyal cartilages in front, to bring about the adult relations. It is a matter of the keenest regret that I am unable, from lack of material, to add a circumstantial account of these later changes. The mandible and ceratohyal have changed but little; and the concavity at the back of the latter, vacated by the first branchial arch, still remains. The branchial skeleton, on the other hand, has been considerably reduced. The first branchial arch is now no thicker than the second, and a wide space intervenes between it and the ceratohyal. All trace of inflation has disappeared, and the flattened branchial apparatus bears a much closer resemblance to that of the common frog than it did before. The three branchial clefts have not altered except in size, and, judging by analogy with tadpoles of other species of Anura, they will not do so. The arches will be absorbed, and the clefts will disappear in consequence; but the latter will play only a passive part in the change. The distance across the branchial skeleton is now considerably less than the extreme width of the hyoid arch. That the absorption of the branchial skeleton has been a marginal one is seen by the fact that in the first stage the external edge of the 7% 100 MR. W. G. RIDEWOOD ON THE HYOBRANCHIAL branchial basket-work fits underneath the auditory capsule so closely that considerable care is required to separate the two ; but in this third stage the whole of the branchial skeleton lies in the floor of the pharynx. The most interesting feature of this stage is the formation of the thyrohyals, which may be seen arising from the hind edge of the median basal plate of cartilage, as a paired outgrowth per- fectly independent of the four branchial arches (PI. 11. fig. 3, ¢). This is universally accepted as the mode of origin of the thyro- hyals in the tongued Anura. It was clearly set forth in the works of Cuvier (8), St.-Ange (36), and Dugés (10), was con- firmed by Parker (32) in his exhaustive treatise on the develop- ment of the skull of the common frog, and has recently been corroborated by Schulze (38) and Gaupp (14). These thyrohyals as yet show no trace of ossification, and are quite free from the laryngeal skeleton. Here, then, is conclusive evidence of the secondary nature of that connexion between the hyobranchial and the laryngeal skeleton, which distinguishes the Aglossa from all other Amphibia. The fourth branchial arch has shrunk considerably since Stage II., especially towards its median end, so that not only is it now no thicker than the third arch; but it no longer underlies the larynx. In fact, although the larynx has not shifted in position relatively to the branchial clefts, there has appeared a space between the fourth branchial arch and the larynx. Into this space the thyrohyals grow. Owing to paucity of material I have been unable to determine the mode of origin of the great ale of the adult hyobranchial skeleton ; but it is highly probable that, as in Pipa, these are the result of secondary outgrowths from the undivided hypo- branchial part of the branchial skeleton (PI. 11. fig. 3, 2b), and not, as might be suggested by a comparison of fig. 3, Pl. 11. and fig. 1, Pl. 8, of the persistence of such part of the branchial skeleton as still remains, under a condition of closure of the clefts. The ceratohyals, still massive, are fated to undergo a certain amount of absorption. The absorption, however, is not excessive, and the Jamellar expansions of cartilage on the internal and external edges of the anterior cornua of the adult clearly owe their presence to the fact that the absorption of cartilage is less complete than in most Anura. The laryngeal skeleton has altered but little since Stage I]. The arytenoid. cartilages (PI. 11. fig. 10, av.) are very large in proportion. They SKELETON AND LARYNX OF XENOPUS AND PIPA. 101 resemble in shape the arytenoids of most Anura (cf. Bombinator, fig. 13), and differ in a marked degree from the arytenoids of the adult. They are closely applied to one another in the median line, as in the typical Anuran larynx, and not as in the adult Xenopus. The cricoid is an annular cartilage, more extensive on the ventral than the dorsal surface, and quite free from the thyro- hyals. The ventral portion terminates anteriorly in a blunt point a short distance behind the anterior ends of the arytenoids, and is produced behind into a pair of narrow bands (fig. 10, 67.) which support the ventro-internal walls of the bronchial tubes. This early appearance of the bronchial cartilages is well worthy of note. The roof of the larynx is expanded in front, and the dorso-lateral corners are produced into a pair of blunt processes which are doubtless to be identified with the blinker-shaped processes of the adult (PI. 8. fig. 3, 67.). The posterior edge of the roof is concave. The sides of the cricoid do not extend forward, so that the whole of the arytenoid can be seen in a side view. Parker neither discusses nor figures the branchial skeleton of his youngest Xenopus larva—first stage, with no traces of paired limbs, total length 13 inch—but the figure he gives (33. Pl. 56. fig. 6) of the hyoid arch and mandible of this stage corresponds almost exactly with that of my Stage I. The broadened posterior extremity of the basihyal he regards as the first basibranchial, but beyond that we do not differ. Parker’s second stage—tad- pole, 13 inch long—of the limbs of which we have no means of judging, would also appear to correspond with my first stage ; for in his side view of the hyobranchial skeleton (which, curiously enough, he draws upside down), the branchial cartilage is still considerably inflated and fits up closely behind the ceratohyal. He is silent concerning the hyobranchial skeleton of his third stage,—tadpoles at their largest size— ; but in his fourth stage— young, with large legs and diminishing tail—which, judging from the second qualification, should be older than my Stage III., he figures (33. Pl. 58. fig. 4) the two ceratohyals as still distinct, and the basihyal as still recognizable. The cartilage, which in this figure he indicates by outline only and marks 6r. 1, is, I am convinced, the whole of the hypobranchial plate, shown in my fig. 3 at hd. 102 MR. W. G. RIDEWOOD ON THE HYOBRANCHIAL Pipa americana. Here, as in Xenopus, the mode of development of the hyo- branchial skeleton is practically unknown. Parker has con- tributed a certain amount of information*, but beyond this the subject has not been touched. Gronberg (18. p. 635) speaks of having discovered in the larva of Pipa that, as in the adult female, the isthmus between the anterior and the posterior parts of the basal plate is a continuous cartilage, but he gives no further information, nor any figures of larval structures. The Specimens available for this part of the investigation were suffi- ciently plentiful to render it possible to obtain much more complete and satisfactory results than was the case with Xenopus. For many of the specimens I am indebted to the kindness of Prof. G. B. Howes; the remainder were obtained by purchase from V. Frit of Prague. It will be found convenient to choose six stages of develop- ment for description here. There is more uniformity in the size of embryos of the same age than in Xenopus, since the young of Embryos of Pipa. Measurements in millimetres. From snout iene Length of | Length of to root of Rasa extended extended tail. ‘ hind limb. fore limb. Site) IE, Gagoaoboo00: 10 12 4 0 SHEVE®) 1D esco6000005 12 ll 6 3 Stage mii eee 115 6:5 75 5 13 85 ral 5 12 10 a 5 StalpenliVien aneeccnes 11 9 75 4:5 11/5 10 7 4:5 9 10 8 45 12 6 75 4:5 Blase Vacs esce: 11 9 9 5 11 9°5 10 5 StalcenVilgrececnees 14 0 10 55 13°5 0) 11 5:5 15 0 11 8 13 3 10 6 14 0) 10°5 5d * For a criticism of which, see p. 110. SKELETON AND LARYNX OF XENOPUS AND PIPA. 103 Pipa are less dependent on external conditions, being provided at the outset with a large quantity of food-yolk, and being lodged during their development in the integumentary pits on the back of the mother. But it happens unfortunately that very little idea of the extent of development of the hyobranchial skeleton can be obtained from an examination of the external features, and it is consequently necessary in most cases to dissect out this portion of its body before deciding what stage any particular embryo has reached. The table on p. 102, com- piled from the Pipa embryos examined, will show how useless actual measurements are for discriminating the various stages. Stage I. is distinguished by the fact that the fore limbs are not yet extruded, and Stage VI. is marked by the complete absorp- tion of the tail ; but there is no safe guide for distinguishing the intermediate stages. Srace I. (Pl. 11. fig. 4.) Embryo with abdomen much distended by unabsorbed yolk. Fore limbs not yet extruded. Length of body 10 mm., tail 12 mm. T have called this Stage I. because it is the youngest that I have been able to examine; it is, however, highly probable that the hyobranchial skeleton of younger embryos would show many features of interest. Owing to the fact that the larve of Xenopus are freely-swimming tadpoles, while the young of Pipa only quit the integumentary pits on the back of the mother after their metamorphosis is complete, it is no easy matter to correlate the stages of development in the two genera; but, judging from the extent of development of the hyobranchial skeleton alone, this first stage of Pipa would seem to be equivalent to a stage inter- mediate between those numbered II. and III. in Xenopus. The branchial arches are rods of cartilage and not sheets, as in the early stages of Xenopus, and the lines of demarcation of the five elements of the larval hyobranchial skeleton are not to be distinguished. The hyobranchial skeleton is one continuous cartilage, considerably broader across the branchial than across the hyoid region. The branchial skeleton has the appearance of having already suffered considerable absorption. The larynx lies in the deep indentation in the middle of the posterior border, and does not overlie the branchial cartilage. The three branchial clefts are approximately equal in length, but in width the pos- terior has the advantage. The first branchial arch is separated 104 MR. W. G. RIDEWOOD ON THE HYOBRANCHIAL from the hyoid by a distance equal to its own width; and the breadth of the cartilage connecting the hyoid and branchial divisions of the hyobranchial skeleton is nearly equal to the antero-posterior diameter of the median basal plate, from which latter the four great lobes, two hyoid and two branchial, radiate. The hyoid cornua are large, and each in shape resembles an axe-head. The antero-external convex edge is thin and shows signs of absorption. It is important to note that, although the external or lateral extremities of the mandible and ceratohyal are nearly in contact, a considerable space is enclosed between their more median parts, and these relations are maintained as long as the ceratohyal is recognizable. In Xenopus, it will be remembered, the median ends of the ceratohyals lie close up behind the mandible. Although the hyobranchial skeleton of the early Pipa embryo exhibits the aforementioned peculiarities, it is nevertheless far less aberrant than that of the Xenopus larva. Parker states (33. p. 649) that there are neither external nor internal gills in the embryos of Pipa, but it is evident that he had never seen Wyman’s paper describing (44. p. 371) the existence of three pairs of external gills, internal gills, and a pair of branchial clefts. It would be difficult to believe that a hyo- branchial skeleton, conforming so closely to the normal anuran type as does that of the present stage, should not be associated with branchiz of some sort. Stage II. (Pl. 11. fig. 5.) Embryo with abdomen not distended by yolk, although on dissection much is found still unabsorbed. Hind limbs flexed, 6 mm. in length when straightened out. Fore limbs 3 mm. Length of body 12 mm., tail 11 mm. This stage does not differ to any great extent from the first, and is not so strictly intermediate between Stages I. and III. as could be wished; but here again the deficiency is to be accounted for by the difficulty of obtaining embryos of any particular age. The chief difference between this and the preceding stage is to be noted in the hyoid arch. Considerable absorption has taken place—a significant foreshadowing of the later complete disappearance of this arch. The erosion is greatest anteriorly, so that the antero-external edge, before convex, is now concave, while the anterior indentation between the two cornua has broadened out. Since no absorption yet SKELETON AND LARYNX OF XENOPUS AND PIPA. 105 takes place at the free extremities of the cornua, there is no diminution in the total width of the hyoid skeleton. The most important change which has taken place in the branchial skeleton is the severing of the commissural or epibranchial cartilage between the lateral extremities of the first and second branchial arches. An indication of this is to be seen in Parker’s figure (33. Pl. 63. fig. 4) of a stage much earlier than this. It is interesting as being the first step in the total disintegration of the branchial arches which so soon supervenes. The larynx has advanced in position so that it is now in contact with the branchial skeleton. Stage III. (Pl. 11. fig. 6.) It is impossible to give a diagnosis of embryos of this and the two following stages, as a glance at the table of measurements will show. The changes in the hyobranchial skeleton at this period appear to be unaccompanied by any marked alteration in the external appearance of the embryo. The general aspect of the hyobranchial skeleton is now entirely changed, and the absorption of cartilage is proceeding rapidly. The hyoid cornua are reduced to rods of cartilage, while the branchial arches are breaking up, chiefly at their peripheral ends. Before proceeding to a detailed description of the skeleton at this stage, it may be well to point out that in Xenopus the hyoid arch remains practically unchanged during the time that the branchial arches are undergoing their gradual reduction, but that in the early stages of Pipa it becomes rapidly diminished and the branchial skeleton follows but slowly. At Stages IIT. and IV. of Pipa the rapid absorption of the branchial arches begins, and then it is a race, as it were, between the hyoid and branchial arches for first disappearance. The rate of absorption at this period is excessive, so much so that embryos the hyobranchial skeleton of which has reached Stages IIT. and 1V., and even V., are externally indistinguishable: in other words, that the rate of change in the hyobranchial apparatus exceeds by a good deal that of the general metamorphic changes of the body. In this Stage III. the diameter across the branchial skeleton is not greater than that across the hyoid arch. This may be partly due to an elongation of the hyoid cornua; but the more important factor in this change of proportion is undoubtedly a shrinkage of the branchial arches. This is indicated to some 106 MR. W. G. RIDEWOOD ON THE HYOBRANCHIAL extent by the shortening of the branchial clefts before they become broken open peripherally ; but it is more conclusively shown by the fact that, while in Stages I. and II. the aortic vessels made up of the four efferent branchial arteries run alongside the commissural or epibranchial cartilages that connect the outer ends of the branchial arches, there exists in Stage III. a con- siderable interval between the paired aorta and such parts of the commissural cartilages as yet remain. This is not to be explained by a widening or separation of the aorte, for the actual length of the branchial vessels has not increased. The commissure between the first two arches was already ruptured in Stage II.: it has now completely disappeared. The second commissurai cartilage has severed its connexion with the second branchial arch, but remains attached to the third. The third commissure is still intact. The first branchial arch is undergoing reduction at its proximal as well as at its distal extremity. The cartilage at the junction of the first ceratobranchial and the hypo- branchial plate (Pl. 11. fig. 6, +) is very thin, and showed in two of the specimens examined a distinct indentation in the edge,— a foreshadowing of the dismemberment of the arch which is completed in Stage TV. Bg SSE IE Diagram of a section across the glandulo-muscular sac of the spermiducal gland of P. Arturt (cf. Pl. 16. fig. 4¢). The penis at rest. a. Penial duct. 6. Portion of the duct passing through the wall of the penial sac. c¢. Por- tion of the duct lying free in the “penial sac.” d. The penis itself. e. “ Atrium,” or cavity of the muscular part of the glandulo-muscular sae. Fig. 2. SE See Oe Similar diagram, showing penis protruded. Letters as in Fig. 1. The part of the penial duct (c) has been pulled out by the protrusion of the penis. the recurrent limb of the penial duct (as in fig. 4c) did not extend more than halfway along this diverticulum, and this 210 DR. W. B. BENHAM ON was the case in the specimen the “penis” of which was drawn out on pealing off the cuticle. Further investigation, as well as a series of sections, shows that the penis is a cylindrical str ueture, with strongly muscular wall, traversed by two narrow canals—the sperm-duct and duct of the spermiducal gland (text-fig. 3). It lies, when at rest, in the “ penial sac” (text-fig. 1), and the penial duct, which passes to it (6, c), is bent upon itself; when in use, the free end is foreed out of the male pore by the contraction of the muscular wall of the sac, and the tube (¢) simply unbends (text-fig. 2). It is, in fact, a protrusible organ and not an introvert. Hitherto our ideas upon the “ penis ” of the genus Pericheta, as of other forms with stout ‘ penial ducts,” like the Acantho- drilide, have been somewhat vague; but it has been presumed that this duct is capable of eversion, z.e¢. it is a “ pleurecbolic introvert.” In the present instance—as sections of the penis in a condition of retraction and protrusion demonstrate—there is no unfolding of the wall of the penial duct: its free end is merely pushed outwards, in the same manner as the penis of the Leech. Beddard makes the following remarks in his general account of the “penis” in the Oligocheta, p. 124 of his ‘ Monograph.’ “T have found specimens of P. Houlleti, killed in alcohol, with the terminal part of the muscular duct of the spermiducal gland everted. I have not noticed the occurrence of this in allied forms, but it very possibly takes place.” . . .‘‘ In some species of Pericheta the muscular duct of the spermiducal gland opens into a wide and rather thin-walled terminal chamber which opens directly to the exterior: it is here, again, possible that this terminal chamber is protrusible ; but I have no facts at hand to prove or disprove the possibility.” The “glandulo-muscular sac” in the present species does not present the firm, compact appearance figured for the ‘“ mus- cular bulb” of other species, which is usually hemispherical and smooth. But the muscular tissue is loose, the fibres not beng so definitely arranged in circular and longitudinal series; and further, this sac contains two great oval glands, one in front and the other behind the entrance of the penis (text-fig. 3, a, a’). Each gland is a somewhat pear-shaped organ lined by a single layer of tall gland-cells. Rosa has deseribed, in P. glandulosa, a group of conspicuous NEW SPECIES OF PERICHETA. Pail glands at the side of the “spermiducal gland,” but these are isolated and open by several independent pores around the male Diagram of a longitudinal section of the glandulo-muscular sac of P. Arturi. a,a', The anterior and posterior glands, lined by glandular epithelium. 6. Duct of gland leading to the ‘‘atrium” (¢). d. The end of the protruded penis, cut longitudinally to show the two ducts. jf. The duct of the spermiducal gland. g. The sperm-duct. c. External pore, common to fand g. pore. In several other species, e. g. P. aspergillum and P. ber- mudensis, glands of smaller size similarly open by small isolated pores in this region; but I do not recall any other species of the genus with a “ glandulo-muscular” apparatus like that of P. Arturi. The copulatory apparatus amongst Earthworms is very varied, and among those organs which serve as a “penis,” that of Eudrilus appears to bear most resemblance to that of the present worm ; in it a small tongue-shaped papilla, traversed by the sperm-duct, lies, when at rest, in a subspherical muscular sac, as described and figured by Perrier and Beddard and Horst. Beddard has suggested that this papilla or “ penis” is capable of protrusion when the muscular sac is everted. Perrier* has figured it protruded, and it is so in a specimen in my possession. It is not only very much smaller than the penis in P. Arturi, but its mode of protrusion is evidently quite different, being due to the contraction of the wall of the atrium, and not to the unfolding or straightening of a part of itself. And in other * Nouv. Arch. Mus. Paris, 1872, pl. 2. fig. 28. LINN. JOURN.—ZOOLOGY, VOL. XXVI. 15 212 DR. W. B. BENHAM ON cases the copulatory organ is formed by an eversion of a mus- eular sac carrying with it the pore of the sperm-duct which traverses its wall, as again in Moniligaster *. I describe another species, P. malamaniensis with a penis similar to that of P. Arturi. Amongst the slight variations in structure of this spermiducal gland and apparatus in P. Arturi may be mentioned the fact that, in one instance, the gland on one side was very much reduced. In another case it was three-lobed on one side (see Pl. 16. fig. 406), and the penial duct passed below instead of above the gland, possibly as the result of a sudden contraction or struggle at the death of the worm. The worm may be characterized as follows :— PrEricHzTA ARTURI, 0. sp. Violet anteriorly, greyish brown posteriorly. Measures about 100 x 4mm. (?), with 90 segments. Male pores slit-like, on circular porophores; a long thread-like penis protrudes from each on peeling off the cuticle. One pair of spermathece in vIlI., pore anterior; globular sac, short duct; diverticulum longer than sac, slightly undulating, and scarcely dilated terminally. Dorsal pore x1./x11. A dorsal gap interrupts the ring of chet, which are 50 before and behind the clitellum; those on rv. to vii. larger. Simple, paired intestinal ceca inxxvir. Spermiducal gland cleft into two quadrate lobes, each much incised; penial duct long, between the lobes, sharply bent in U-shape ; large glandulo-muscular sac with a prolongation between the lobes of the gland, receiving the penial duct. Hab. Blanche Bay, Gazelle Peninsula, New Britain. Affinities.—Of the four species of Pericheta enumerated by Beddard as showing a single pair of spermathece in segment VIII., none show any close resemblance to the present one; though P. sangirensis, M., exhibits at first sight certain general resemblances. A detailed comparison, however, readily distin- guishes the two—in coloration and character of spermiducal apparatus; while the spermatheca has a well-marked duct into which the diverticulum opens. In a few general features, too, Rosa’s P. atheca bears a resemblance. * Bourne, Qu. Jour. Micr. Sci. xxxvi. pl. 28. fig. 5d. NEW SPECIES OF PERICHETA. 213 Pericheta malamaniensis. In 1885, in my earliest contribution to the literature of Earthworms, I referred * (p. 256) to the fact that a species of Pericheta from the Philippine Islands presented ‘ numerous nephridia’”’ in each segment. Later, in 1891, I gave the name P. malamaniensis to this species, and was guilty of an indefensible procedure in giving a name to a new worm without a diagnosis of it. This error I propose now to rectify. My attention was drawn to this worm, which I had so long neglected, by coming across some drawings of it, in which I recorded the fact (entirely forgotten) that the spermiducal gland is provided with an exceptionally large muscular sac. I at once concluded, in view of my discovery in P. Arturi, that here too I should find a ‘penis’ of the same character as in that worm. I consequently re-examined the series of sections which I cut some eleven years ago, with the result that my expectation was confirmed. PERICHATA MALAMANIENSIS, Benham, 1891. The material at my disposal consisted of three worms and a piece, collected during the ‘ Challenger’ expedition at Malamani. The length of the worm is 90 to 100 mm. with a diameter of about 5 mm.; there are 104 segments. The worm is dirty yellowish brown, nearly uniform {, without cheetal bands. The clitellum is distinctly brown. The anterior end is rather obtuse ; the prostomium is small. The male pores are slit-like, on rather conspicuous and slightly prominent, but not extensive, “ porophores,” separated by 10 cheete ; there are no copulatory papille. There is a single pair of spermathecal pores between segments Vit./v1i1. in a line with the male pores. About 15 chete measured in segment vit. lie between the lines joining the pores. The dorsal pores commence between segments X11./x11I., and are visible on the clitellum, which is quite normal and presents no chete. * Qu. Jour. Mier. Sci. xxvi. t Qu. Jour. Mier. Sci. xxxii. p. 316. { The worms were in glass-stoppered bottles, so that this colour has nothing to do with the “ cork,” as Michaelsen has suggested might be the case. 15* 214 DR. W. B. BENHAM ON The chet are set in a prominent ridge—perhaps due to the excellent preservation of the worm. There is a small dorsal gap, but no ventral one. The chetz are more closely set ventrally than dorsally. There are 38 chetz on segment 1. 99 40 99 99 Vv. e 54 as exelalva aA 62 bs Me KOXGVE Internal Anatomy.—There is nothing very striking in regard to the septa; septal glands occur in segments Iv., V., v1.; the gizzard is relatively large, bell-like in shape, and appears to occupy segments VIII., IX., and x. The intestine is distinctly sacculated after the 14th or 15th segments, but narrows in the 26th and 27th. In the latter segment a pair of large ceca arise, and extend forwards into segment xx.; their lower faces are slightly notched. Behind the 27th segment the intestine again enlarges. Above the intestine, on each side of the dorsal vessel, after the 27th segment, is a pair of racemose (“ glyco- genic”) Phagocytar organs * in each segment, containing abundant pseudonavicule. There is a conspicuously large “heart’’ in segment x1i1., and there are two smaller “hearts” in the preceding segments. The genital organs present two interesting features, viz.: the very large size of the spermatheca, and the great development of the “‘ muscular bulb ” of the spermiducal gland. The sperm-saes lie in segments x1. and x11., and the testes in the usual segments. The spermiducal gland (Pl. 16. fig. 6 2) isin two distinct lobes, as in P. Arturi, and from each a delicate duct passes away to unite together to form a larger “penial duct,” which, after passing some little distance backwards, bends upon itself and becomes much thicker and more muscular (d): it then runs forwards to open into the middle of the outer margin of a great glandulo-muscular sac (0) ; this extends from segment xv. to seg- ment xrx. and presents three slightly marked subdivisions, namely, two terminal which contain each a great gland, and a smaller middle one, the “atrium.” Sections through this structure show the same general arrangement as in P. Arturi—i.e., the penial duct penetrates its wall, and projects freely as a “‘ penis” * Schneider, Zeitschr. f. wiss. Zool. lxi. 1896, p. 363, describes these organs for P. indica. SS NEW SPECIES OF PERICHATA. 215 into the cavity of the ‘‘ atrium” (text-fig. 4); but, so far as I have been able to make out, the protrusion of the penis is effected in a manner different from that in P. Arturi. I find no narrow duct, capable of being unwound in eversion, and I conclude that the sac itself must be everted, as in Hudrilus, so that the penis is carried outwards with it. P. malamaniensis.—A diagram of about a third of a transverse section through the body, at the level of the male pore. a. Sperm-duct: on the right it is cut through as it passes from the body-wall to the penis, which it traverses to the tip (ef. text-fig. 3). 6. The spermiducal gland. c. Its duct. d. Muscular penial duct, outside the muscular sac. d@'. The continuation of this as a narrow, tubular, protrusible penis. e, Atrium, or cavity of the muscular part (middle region) of the glandulo-muscular sac (cf. Pl. 16. fig. 6a). f. Itsmuscular wall. g. Peculiar muscular pad projecting into the atrium. %. Body-wall. Mr. Beddard, in a recent paper (P. Z.S. 1896), has described an arrangement of the sperm-ducts in relation to the penial duct of P. Perkinsi, which he rightly poimts out (p. 200) has not hitherto been observed. From the observations which I have made, I would suggest that the arrangement here is not so unusual as our lack of information on the microscopical structure of the worms inclines him to consider it to be; for in both the worms which I have recently examined I find essentially 216 DR. W. B. BENHAM ON the same thing. In P. malamaniensis the sperm-duct crosses the great glandulo-muscular bulb, as ordinary dissection shows, and passes apparently into the narrow duct of the gland (a, fig.6 a); but in reality the two sperm-ducts together run along- side this penial duct for some distance, then unite together, and the single sperm-duct accompanies the duct of the gland throughout the whole length of the penis, only opening into the gland-duct close to the tip of the latter, where it undergoes a slight dilatation. The conjoined duct now opens by two pores, just below the tip of the penis. The epithelium of the sperm-duct and gland-duct are quite different, as can readily be distinguished in section. In P. Arturi precisely the same thing occurs (text-fig. 3); but, owing to poor condition of preservation, the difference in the lining is not so readily recognized, and if it were not for the observation in P. malamaniensis might have been overlooked. The spermatheca (Pl. 16. fig. 66) is of considerable size and lies in segment virt. The sac is large and globular, with a thick muscular duct, nearly as long as itself; into it there opens a small diverticulum, consisting of a slightly undulating muscular duct, terminally dilated to form a small oval sac. The present species may be diagnosed thus :— PERICHETA MALAMANIENSIS, Benham. Brown ; 100 mm. X 5 mm., with 104 segments. Male pores slit-like, separated by 10 chete; no copulatory papille; a single pair of spermathece opening between segments V1I./VIII., consisting of an enormous globular sac, with long stout duct, receiving a small diverticulnm, terminally dilated. Dorsal pore xrr./xti. Cheetal ring with dorsal gap; 50 chet in front of and 60 behind the clitellum. Spermiducal gland in two distinct lobes; a glandulo-muscular duct, containing a distinct tubular penis. Hab. Malamani, Philippine Islands. Pericheta Flowert and P. Madeline. I take this opportunity of describing two other species of Pericheta which appear to be new to science.- Both were received through the courtesy of Prof. ¥. J. Bell, for identi- fication, and have been returned tothe National Collection. My NEW SPECIES OF PERICH ETA. 2 thanks are due to the Director of the British Museum of Natural History for the opportunity of examining them. PrricnmTa FLOWERI, n. sp. Three specimens of this worm were collected by a son of Sir W. Flower, at Bukit Timah, Singapore, in some rotting timber. They were all ina rather poor state of preservation, and indeed one looked as if it had been swallowed by a frog. Of the two which were of use for identification, one measures 120 mm. x 4 mm., and consists of 116 segments; the other is slightly smaller. In colour they are rich brown, though not very dark, with a violet iridescence and nearly white chetal rings. The pre- clitellian region is more distinctly purple than the rest; the clitellum is purplish brown; the ventral surface of the worm is lighter than the dorsal, as is usually the case. The male pores are slit-like, with crenated lips, each on a flat rounded papilla, of a light (in fact nearly white) colour; there are 10 cheetz between these porophores. No copulatory papille are present, but one ofthe “ penial ducts” is partially protruded.* There are four pairs of spermathecal pores between segments v./VI., Wi./VII., Vil./VIII., Vitt./1x.; having about 12 chete between, as counted in the 8th segment. The first dorsal pore is between segments XII./XIII. The clitellum does not entirely cover the usual segments, as it does not reach the intersegmental groove at either end; there are about a dozen cheetz on the ventral surface of the 16th segment ; there are none on the other clitellar segments. The chetal ring is not quite complete, being interrupted by a small dorsal gap ; but there is no perceptible ventral gap. I counted 18 chet on segment 1. 35 5p a VI. 40 .; o MITE. 45 a 56 XXVI. The chet are rather closer together over the ventral surface than dorsally ; but none are larger than the rest. Internal Anatomy.—N one of the septaare noticeably thickened; the gizzard and pair of ceca are normal. * JT had returned the specimen to the British Museum before I had examined P. Arturi, but from the absence of a muscular bulb I conjecture that there is no “ penis.” e 218 DR. W. B. BENHAM ON There are two pairs of large and lobulated sperm-sacs in segments XI. and XII. The ovisac in segment X1Vv. is rather large. There are four pairs of spermathece (Pl. 15. fig. 36), in seg- ments VI. to Ix.; the sac is oval, with a very short duct, indeed scarcely recognizable; the diverticulum is characteristic, having a long, narrow, nearly straight duct, terminating in a globular dilatation. The spermiducal gland (fig. 3 a) is somewhat kidney-shaped in outline, with three deep notches ; the penial duct is moderately long, curved, and without a muscular bulb. The characters of this worm may be summarized as follows :— P. FLOWERI, n. sp. Rich brown, with light chetal bands; anterior end purple. Measures 120 x4 mm., with 116 segments. Male pores slit-like, with crenate lips, on lighter flat porophores separated by 10 chetz. No copulatory papille. Four pairs of spermathece, vi. to Ix., opening anteriorly. The sac is ovate, with very short duct; diverticulum long, narrow duct, with terminal globular swelling. Dorsal pore xt./xim. Chetal ring with dorsal gap; 40 chete in front of, and 45 behind clitellum. Spermi- ducal gland oval, with three deep clefts ; duct curved, short ; no bulb. Hab. Bukit Timah, Singapore. Affinities.—There are a considerable number of species with four pairs of spermathece ; some of these are manifestly different from the present worm. From others it differs in smaller and fewer points. It resembles in several points P. Perkins?, Bedd., from the Sandwich Isles, which, however, has no chete on the 16th segment. The spermathece appear to be different, if IT understand Beddard’s description; and though the spermi- ducal gland bears some likeness, I have not observed the peculiar arrangement of the sperm-ducts described by him*. But in P. Perkinsi the chetz differ in size in various segments, and are larger ventrally than dorsally. From P. padasensis, Bedd., the present worm differs in its much smaller size; in the possession of a very short duct to the spermatheca ; in having only two, instead of four pairs of sperm- * J did not make sections of the worm, and had returned it before I had examined P. Arturi and P. malamaniensis, NEW SPECIES OF PERICHATA. 219 sacs; and in the absence of a muscular bulb to the duct of the spermiducal gland. P. trinitatis, Bedd., has a clitellum lke that of P. Floweri; but the “loose” character of the spermiducal gland and the shape of the spermatheca, as well as other characters, serve to distinguish the two. P. coa, Rosa, agrees in many points, but the spermiducal gland has a muscular bulb, and is otherwise different; so too is the spermatheca. P. Floweri agrees in many points with P. enganensis, but differs in the cheta-formula, character of cecum, spermatheca. P. bossche reminds one of the present worm in its general external characters, but there are three pairs of sperm-sacs; the spermiducal gland is different, and the spermathecal diverticulum is only half as long as the main sac. The other species with four pairs of spermathece are mostly characterized by more definite peculiar features, serving to mark them off from the present worm. Prericua£ta MapELina, n. sp. This handsome worm was collected by Mr. Everett on Mt. Kina Balu, Borneo, but it does not appear to be identical with any of those described by Beddard * from the same locality. The worm (of which I only received a single specimen, which was not fully mature) measures 185 x6 mm., and consists of 110 segments. The plan of colouring is similar to that of P. pulchra, Mich., P. zebra, Benham, and many other species from tbis part of the world, in having alternate rings of dark and light colour. In the present instance the darker bands, on removal of the cuticle are rich chocolate-brown ; the light chetal bands are dead bluish white, without any yellowish tone. The dark bands, widest dorsally, extend right across the ventral surface as narrow streaks of lighter brown. At the hinder 18 segments of the worm, as is again very frequently the case, the dark bands are so wide as to meet, sending narrow outgrowths between the individual chete and almost obliterating the white bands; on the ventral surface of this region the dark bands are wide, so that the “ tail” is darker than the anterior end. The prostomiam is margined by white, * Ann. & Mag. Nat. Hist. (6) xvi. 1895, p. 69. 220 DR. W. B. BENHAM ON and the mouth is surrounded by a white ring at the anterior part of segment I. A dark streak traverses the mid-dorsal line, behind the cli- tellum, indicating a dorsal gap in the ring of chete. There is nothing to note in respect of the male pores, which are separated by 16 chete. There are no special copulatory structures; but the worm.was not fully mature, although prob- ably fully grown. There are four pairs of spermathecal pores at the anterior — margins of segments VI. to 1x., with about 20 chete between. The first dorsal pore is between segments X11./xtrr., but is small. The ring of chete is not complete, there being a dorsal gap equal to about twice the normal interchetal space ; but there is no perceptible ventral gap. The chete, though showing no difference in size, are closer together ventrally than dorsally ; they number 36 in segment II. 56 93 v. 56 ‘3 x 66 a Xx 60 s XXV. Internal Anatomy.—There are no remarkably stout septa, though the four following the gizzard are thicker than the rest. The cylindrical gizzard lies in segments ViI. and Ix., @. e., there is a portion of cesophagus between it and the septum x./Xxr. The cxcum is slightly notched on its ventral surface. ‘The sperm-sacs present a slight peculiarity, which I do not remember to have noticed before; there is a pair in each of the segments XI., x11.; the “ Samenblasen,” as usual, being in seg- ments x., x1. Hach sperm-sac (PI. 16. fig. 5 c) is constricted into a larger ventral and medial portion, and a smaller, round, outer portion, which is provided dorsally with a narrow, pointed, finger- shaped prolongation (f), quite suddenly and distinctly separated from the rounder sac. The four pairs of spermathece, which are all alike, lie in segments VI. to Ix.: each is a pyriform sac (fig. 5 6), with a short wide duct, into which opens the duct of the diverticulum: this terminates in a globular swelling. The spermiducal gland (Pl. 16. fig. 5a) is nearly square and occupies segments XVI., XVII., and xvi. Its margin is greatly NEW SPECIES OF PERICHETA. 221 incised, so as to be divided up into a number of variously sized lobules. The muscular penial duct (d) is §-shaped and opens into a muscular sac or bulb (6). P. MavELina, n. sp. Alternate bands of purplish chocolate and white. Measures 135 x6 mm.; 110 segments. Male pores separated by 16 chete ; no (?) copulatory papilla. Four pairs of spermathece in vr. to IX., opening at the anterior margin. A pyriform sac, with small diverticulum. Dorsal pore x11./x1m1. Chetal ring with dorsal gap; 56 chet in front of, 60 behind clitellum. Spermiducal gland rectangular, occupying three segments ; much and deeply incised: penial duct S-shaped, opening into round muscular bulb. Hab. Mt. Kina Balu, North Borneo. A ffinities—Two worms from Kina Balu, viz. P. Hveretti and P. kinabaluensis, are naturally recalled to mind as being, one or the other, possibly identical with the present worm ; but in both there are several small spermathece in each of the segments VI., vil.—much as in the genera Aynotus and Microcheta. Further, the normal intestinal czecum is absent in both ; and there are thus fundamental differences. The absence of copulatory papille in P. Madeline must not be taken into account in referring to other worms, for it is not mature, and possibly they may develop late. Only one of Beddard’s Bornean species has four pairs of spermathecex, viz. P. padasensis; but in many particulars the two worms differ. I have been unable to find any species which is very near to the present one in the totality of its characters. Remarks on Michaelsen’s Criticism of the value of certain Specific Characters of the Genus Pericheta. In a short paper dealing with new species of Percheta from Java * I insisted on the necessity for the careful description of certain characters which all lumbricologists recognize as of value in the discrimination of species of this genus, and on the need of careful figures in illustration of these specific characters. Ina recent contribution Dr. Michaelsen } discusses my remarks, and * Ann. & Mag. Nat. Hist. xvi. 1895, p. 40. + “Oligochzten aus Kikenthal. Ergebn. zool. Forschungsreise in d. Molukken und in Borneo,” Abhandl. Senckenb. Gesell. xxiii. Hft. 2, 1896. 222 DR. W. B. BENHAM ON to a certain degree appears to have misunderstood the relative value which I set u pon these “ specific characters.” In writing of P. capensis (p. 229) he refers to two new “species,” P. sexta and P. Willeyi, which I founded, but which I pointed out differed but very slightly from P. capensis and P. operculata. Michaelsen, rightly I think, now unites these and certain other “species” with P. capensis, but I see little need for the formation of “ subspecies.” If they are sufficiently distinct, it seems to me that the term species is as good as a subspecies; and seeing that P. sexta and P. Willeyi were each founded on one specimen, I should go further than Michaelsen and leave out “subspecies.” I suggested (p. 46) that possibly P. sexta was a “hybrid” or an abnormal specimen. As Mi- chaelsen has had a greater abundance of material he is in a better position than I was to determine the matter, and the names I used for these two worms must be dropped. Now, with regard to Michaelsen’s criticism of the ‘‘ specific characters.” He takes them, one by one, and shows that each may vary. This, no doubt, is true; but I never intended that any one of these characters by ztself is sufficient to diagnose a species. I merely wished to insist upon special attention being paid to all these points, because they do vary; and until more trouble is expended on an accurate description of each presumed “ new species” we shall not be in a position to decide whether it is or is not a mere variation. I agree with him fally that colour, or size, or number of segments, and so on, must be taken with certain anatomical characters. He points out that colour is no safe guide since it is variable; it may fade on being affected by the discoloration of the spirit by the cork. Quite so. On p.42 I wrote “no doubt the colours will soon fade in spirit,” in referring to the fact that we did not know the colour of the living worm. But I do not believe that the colour-pattern—the characteristic alternate banding of many species—will fade rapidly: the worms we have to deal with now have not usually been in spirit very long; and one can generally see whether fading has or has not taken place to any extent. Again, with regard to size, 1 expressly wrote “ within certain limits” (p. 47). I meant that, for example, a worm of the size of P. indica is not likely to belong to a species which in size is about that of P. musica. And there is no doubt that it is some NEW SPECIES OF PERICH ATA. 223 guide to the identification of a worm to know about what size other presumed similar forms may be. The spermatheca is another organ that is of general diagnostic value. Michaelsen remarks, no doubt with truth, that it contracts in spirit. So does the whole body and every organ of the worm; but, nevertheless, the general shape and proportions of sac to diverticulum would not greatly alter. Of much greater importance is the fact that the size depends on the presence of spermatozoa; for we know that in Lumbricus herculeus the worm may have fully-developed sperm-sacs and chtellum, but the spermathece are frequently extremely minute, even in the freshly-killed worm. Very probably they would be invisible if that worm had been preserved in alcohol; and it is quite within the bounds of probability that P. atheca, P. acystis, and others, which have been described as being without spermathece, are in the condition of the Lumbricus just referred to, ¢. e. that the spermathece are not yet functional. Michaelsen also objects to the value of the position of the spermathece, whether they lie, for example, in the 7th or the Sth segment, and open to the exterior between these segments. Here, again, variability may occur: a worm with its spermatheca normally in segment 8 may sometimes have it in 7. Several authors have described such variations. Nevertheless, out of the hundreds, or more probably thousands, of specimens of L. hereu- leus that have been opened under my supervision, and have been drawn by students, so that I have been able to note any departures from the normal, I have only noted such variation in the position of the spermathecz some two or three times. Until we know more of the variability of the animals we are justified in regarding a given position as fixed, if any considerable number of specimens reveal it ; and it appears to me that all lumbricologists describing new species should state explicitly the number of specimens they have examined, and should be in a position to state how far such and such an organ is subject to any alteration in size and position. Too frequently no mention is made of the number of speci- mens upon which a species is founded; still more frequently a new species is founded for one single specimen. I have done this myself; and it serves to call attention to some new permutation of characters which, later on, may be proved to be mere variations from some well-known “type”’ species. Michaelsen, 224: DR. W. B. BENHAM ON in the paper under review, does note these variations, and makes “subspecies” for what other zoologists, if they had not had a series of specimens to examine, would have termed “ new species.” With regard, again, to the position of the spermiducal gland and the gizzard. These organs usually appear to occupy more than one segment, but how many segments—if more than one—are morphologically. occupied by them is unknown. It appears to me, from my own observations, that each belongs to a single segment—the gizzard to the 8th (as Rosa has suggested), the spermiducal gland to the 18th,—but that each may push the septa limiting the segment backwards and forwards or in both directions. And when we state in a description that the spermi- ducal gland “occupies three segments,” we do so with an idea of conveying the relative size of the organ, and do not intend— as Michaelsen appears to think—to convey the idea that it belongs morphologically to these segments. Certain things are postulates of the genus: this is one. This organ, too, is liable to variation, as I myself have pointed out (J. c. p. 43). Again it becomes most necessary to have really careful description and figures of the organ ; for there is probably a very fair constancy in the general characteristic appearance of the gland, which is more readily expressed by a figure than in words. The minuter lobulation of course varies, but in some species it is very much more extensive than in others; and this is too often expressed by vague terms, such as “ compact” or “ loosish,” without even a statement as to the general outline of the gland, which I still believe, in spite of slight variations, to retain its character. It appears to me, in view of the great number of species in the genus Pericheta, that the time has come to insist on these matters. Comparison of ‘‘ new species” with old ones, and of individuals of the same species with one another, will, I doubt not, reduce the aggregate number of species as more specimens are carefully examined. EXPLANATION OF THE PLATES. Puate 15. Fig. 1. Pericheta nove britannice, n. sp. Fig. 1a. View of the male pores and copulatory structures. s. Anterior sucker- like organ. ss’. Posterior sucker-like organ limiting the “ porophore,” which bears the male pore(¢). p. One of the copulatory papille, of which there is one each side of each pore. Fig. 10. Spermiducal gland. x4. d, Penial duct. a a Ts Benham. Linn. Soc Journ. Zoor Vou. XXVI.P1.15. AVAIL Zita Zod WB Bdel.C Berjeau lth. West, Newman imp. NEW SPECIES OF PERICHATA. Lixn. Soc.Journ Zoor, Vor. XXVI. F116 . WB.Bdel.C Berjeau lith : West, Newman. imp NEW SPECIES OF PH RICHATA, NEW SPECIES OF PERICHATA. — 225 Fig. 1¢. Spermatheca, sp. Spermatozoa in diverticulum. 2. Coagulated mass in main sac. Fig. 1d. Cecum: side view of the intestine (¢) in segment 27, showing origin and lobulation of the cecum (c). s; Septum. Fig. 2. Pericheta Sedqwickii, n. sp. Fig. 2a. View of male pores. po. Porophore surrounded by grooves: the actual pore of sperm-duct is very small. Fig.20. Ventral view of the clitellum. 3. The body-wall has been slit up and spread out; the “limited” nature of the organ is shown, as well as the existence of complete rings of chet# in each of its constituent segments. Fig. 2¢. Spermatheca. x 6. div. Small, sessile diverticulum. Fig. 2d. Spermiducal gland. x 4. d. Penial duct. Fig. 2e. Normal cheta from segment xu. xX 110. Fig. 2f. Large cheta from seventh segment. x 110. Fig. 3. Pericheta Floweri, n. sp. Fig. 3a. Spermiducal gland. Fig. 30. Spermatheca. PuateE 16. Fig. 4. Pericheta Arturi, n. sp. Fig. 4a. View of the 18th segment, showing the male pores, with the penis (p) portruded from one pore. po. Porophore. x 6. Fig. 40, The pair of spermiducal glands of one specimen in which the penis isat rest. 4. On the right side is shown the normal shape of the gland: that on the left is abnormal. Fig.4¢. The spermiducal gland (x 6) of a specimen in which the penis is protruded. 6. Glandulo-muscular sac. c. Its externally directed finger-shaped prolongation, or “ penial sac.” d. The U-shaped penial duct (cf. @ in text-figs. 1, 2) penetrating the wall of this prolon- gation. g/. The gland. Fig.4d. A spermatheca. x 4. div. Diverticulum. Fig.4e. The gizzard and vascular system of its neighbourhood. A thin but complete septum between segments vill./Ix. is inserted close to the hinder border of the gizzard. a, b, ¢. Commissural vessels. d. Dorsal vessel. Fig. 5. Pericheta Madeline, n. sp. Fig. 5a. The spermiducal gland. x 4. gl. Gland. d. Penial duct. 5. Mus- cular bulb (or sac). Fig. 5d. Spermatheca. Fig. 5c. The sperm-sacs—to show the peculiar filamentous prolongation (/). s*. The second sperm-sac. sep. Septa. 7, r'. Ciliated rosettes in the “capsules” (Samenblasen). Fig. 6. Pericheta malamaniensis. Fig.6a. Spermiducal gland. a. Sperm-duct. 0. Glandulo-muscular sae or “atrium.” d. Penial duct. gl. The gland. Fig. 60. Spermatheca. 226 MR. A. O. WALKER ON SOME On some new Species of Edriophthalma from the Irish Seas. By Atrrep O. Watxer, F.L.S. [Read 15th April, 1897.] (Puates 17 & 18.) Or the four species described in this paper, two, viz. Leuco- nopsis ensifer and Stenothoé crassicornis, were taken during the dredging and trawling operations of the Liverpool Marine Biological Committee in April 1896, in the Lancashire Sea Fisheries’ steamer ‘John Fell.’ The steamer was at the time engaged in taking fish for the purpose of obtaining ova for the experimental fish-hatchery at Port Erin, and it was found that by attaching a tow-net with a light cane ring to the back of the trawl-net a short distance behind the “‘ foot rope” many small Crustacea were taken, including the above. The same mode of fishing has also been successfully employed by Mr. R. L. Ascroft, in a trip made by him, in the steam-trawler ‘ Britannia,’ to the northern part of the Bay of Biscay ; and it resulted in the capture of some new and rare species of Cumacea and Amphipoda. Of the other two species, Apseudes hibernicus was taken during a week’s collecting at Valentia by Mr. F. W. Gamble; while Parapleustes megacheir was obtained in the expedition of the Royal Irish Academy, in 1888, and came under my notice while naming the collection of Amphipoda in the Dublin Museum of Science and Art. Until the publication of the Report of the Committee on the Marine Zoology of the Irish Sea, in the British Association Report, 1896, very little had been done in investi- gating the Edriophthalma of the Irish Seas, except in the neighbourhood of the Isle of Man, where the Liverpool Marine Biological Committee’s station at Port Erin had enabled me to collect a large number of species during the last 10 years, all of which (including those taken on the English and Welsh coasts of Liverpool Bay) are published in the above Report. I must not omit to mention an important paper by the Rev. A. M. Norman, F.R.S., on the Cumacea of the ‘ Lightning,’ ‘ Porcupine,’ and ‘Valorous’ Expeditions, in the Ann. & Mag. Nat. Hist. ser. 5, vol. ili. (1879) p. 54, in which several species from the Irish seas are described or mentioned. JI may add that I hope shortly to publish a list of Crustacea from Valentia, with the Isopoda and Amphipoda in the Dublin Museum of Science and Art. NEW SPECIES OF EDRIOPHTHALMA. 227 CUMACEA. Fam. LEUCONID2. LEUCONOPSIS, n. gen. A. O. Walker, Report on the Marine Zool. of the Irish Sea, Brit. Assoc. Report for 1896, p. 419 (1897). Female with a distinct two-jointed appendage to the fourth pair of feet, not furnished with natatory sete. Lower antenne short, with the third joint conical, with three minute one-jointed rudimentary flagella. Rami of uropoda subequal. Male with the third pair of feet each provided on the second joint with a pair of curved blade-like processes. Remaining characters as in Leucon. LEvcoNopSIS ENSIFER, A. O. Walker. (PI. 17. figs. 1-1 h.) Female. Carapace about as long as the free thoracic segments ; dorsal crest of fourteen teeth beginning about the middle of the upper margin and curving down to the base of the rostrum; a small tooth on the upper and near the posterior margin; lower margin with the anterior half coarsely toothed and forming with the anterior margin an acute angle, the upper portion of which is finely toothed. Rostrum about a quarter the length of the carapace, obliquely truncate, almost horizontal; lower margin with two or three teeth near the extremity and two or three near the base. Fourth pair of legs with an exopodite or imperfect natatory appendage, two-jointed, reaching nearly to the end of the first joint, which is as long as the remaining four. Telson triangular, as in Leucon. Uropoda with peduncle and both rami subequal in length ; peduncle almost spineless, inner ramus with six unequal spines on the inner and two on the outer side of the first joint ; second joint with two very short spines on the inside; outer ramus obliquely truncate, with five plumose sete on the inner side and four at the end. Length 53 mm. Male. Upper margin of carapace as long as the free segments ; lower margin with five or six teeth on the anterior half, increasing in size anteriorly, forming a right angle with the anterior margin, LINN. JOURN.—ZOOLOGY, VOL. XXVI. 16 228 MR. A. O. WALKER ON SOME which has five teeth just below the rostrum, the second from the rostrum being the largest ; rostrum horizontal, blunt, about one- sixth the length of the carapace, with five small teeth on the lower margin. First pair of legs with seven teeth on the lower margin of the first jomt. Second pair with a large spine at the distal end of the second, and two unequally long spines at the end of the third joint. Third pair with an appendage on the second joint, consisting of two parallel curved -blades, twice as long as the succeeding three joints. Length 83 mm. The above interesting species has a general resemblance to Leucon, from which genus, however, it may be at once distin- guished by the appendages on the fourth pair of legs in the female and the third pair in the male. It was taken in the tow- net, attached to the back of the trawl-net, on April 22, 12 miles S.W. of Chicken Rock, 33 fathoms. Also noticed in the Tenth Annual Report of the Liverpool Marine Biology Committee, and their Biological Station at Port Erin (Isle of Man), p. 14 (1897). ISOPODA. Tribe CHELIFERA. Fam. APSEUDID. APSEUDES HIBERNICUS, n. sp. (Pl. 17. figs. 2-2d; Pl. 18. figs. 2e, 2 f.) Female. Rostrum triangular, blunt, and depressed at the point, suleate on the upper surface, without a spine at the point; margins nearly straight. Ocular lobes well developed; eyes rather indistinct. Epistome provided with a strong spine. Person (mesosome) having the 1st segment rather narrower than the cephalosome, with a strong tooth directed forwards on the antero-lateral corner; the dorsal plates of next four seg- ments widely separated ; no spines on the ventral surface. Pleon (metasome) less than } the length of the whole animal, about equal to that of the three preceding perseon-segments ; epi- meral projections obliquely truncate at the ends. Telson about equal in length to the four preceding pleon- segments, widest in front. The last three perreon-segments, pleon, and limbs are densely clothed with plumose hairs. NEW SPECIES OF EDRIOPHTHALMA. 229 Upper antenne.—The first joint more than twice as long as the two next, finely granulated on the inner side of the proximal half ; flagellum rather longer than the peduncle, 15-jointed; accessory appendage 11-jointed. Lower antenne.—First joint more than half as long as the second, 3rd very small, 4th and 5th equal, three times as long as the 8rd; flagellum 11-jointed; antennal scale with 14 sete. First legs (chelipedes) as in A. spinosus (M. Sars), except that the tooth on the lower margin of the 1st joint is much smaller. The remaining legs closely resemble those of A. talpa (Mont.) and A. spinosus, as figured by G. O. Sars respectively in “ Mid- delhavets Saxisopoder” (Archiv for Mathematik og Naturv., B. xi. 1886, p. 263, pl. 2), and ‘ Crustacea of Norway,’ vol. ii. Isopoda, p. 7, pls. i. & ii. The number of spines on the fossorial legs is apt to vary—in the specimen described they differed on each leg. Uropoda much as in A. spinosus ; the first three joints of the appendages are much longer than the succeeding ones. Length 10 mm. This species is nearest A. talpa (Mont.), from which it differs in the form of the rostrum (which wants the spiniform tip figured by G. O. Sars, 7. ¢.), in the absence of ventral spines, by the comparative shortness and width of the pleon and telson, and in the form of the latter, which is wider in front than behind, while in A. ¢alpa and A. spinosus it is the reverse. A single female was taken by Mr. F. W. Gamble between tide-marks on Church Island, Valentia Harbour, in August 1896. There are also two specimens in the Dublin Museum of Science and Art from the West Coast of Ireland, the exact locality of capture not being known. AMPHIPODA. Fam. STENOTHOIDA. STENOTHOE CRASSICORNIS (PI. 18. figs. 3-3 e), A. O. Walker, Brit. Assoc. Rep. for 1896, p. 420 (1897). Also in the Tenth Annual Report of the Liverpool Marine Biology Committee, and their Biological Station at Port Erin (Isle of Man), p. 16 (1897). Mandibles without a palp. 230 MR. A. O. WALKER ON SOME Maxillipedes with the basal lobe very small, divided to its base. Antenne stout, the flagellum of the lower but little longer than the last joint of the peduncle ; its first joint almost as long as the remaining four together. First gnathopods as in S. marina. Second gnathopods with the palm of the propodos defined near the base by a triangular tooth, the distal extremity expanded and cut into four blunt lobes, of which the proximal is much the largest ; dactylus with a prominence on the inner margin, coin- ciding with the palmar lobus. Pereopods short and strong, the third (meros) joint in the last three pairs much produced backwards, as in Proboliwm calea- ratum, G. O. Sars. Third uropods with four spines on the upper surface of the peduncle, which is twice as long as the first jomt of the ramus. Telson with three pairs of dorsal spines on its proximal half, the first pair the smallest. Length 2 mm. In the form of the hand of the second gnathopods this species approaches S. tenella, G. O.8., and S. Dollfusi, Chevreux ; but both these (perhaps identical) species are remarkable for the length and slenderness of their antenne and pereopods. Three males taken in the bottom tow-net 6 miles W.S.W. of Calf of Man, in 23 fath., April 22, 1896. Fam. PARAMPHITHOIDAG. PARAPLEUSTES MEGACHEIR, n. sp. (PI. 18. figs. 4-4:c.) Body smooth; second segment of pleon (metasome) having a small dorsal tooth, the next three segments dorsally depressed. Head nearly as long as the first two segments of the perwon (mesosome), lateral and post-antennal angles slight and obtuse. Eyes wanting. Coxal plates of the first two segments of the perewon small and rhomboidal, with the anterior angle acute, especially in the first ; 4th coxal plate wide and as deep as the segment. Third pleon- segment with the lower margin strongly and the hind margin slightly curved; hinder angle obtuse as in P. pulchellus, G. O. Sars. Upper antenne about two-thirds the length of the body; 1st NEW SPECIES OF EDRIGPHTHALMA. 231 joint thick and as long as the next two, 3rd joimt much more slender and about one-third the length of the second. Lower antenne about three-quarters the length of the upper ; last joint of the peduncle rather shorter than the preceding joint. Maxillipedes strong; dactylus of the palp longer than the preceding joint, very strong, and with a fringe of minute setules on the upper margin; 5th joint much shorter than the 4th. Gnathopoda very unequal, the propodos of the 1st being about one-fourth of the size of that of the 2nd pair. The first pair have the anterior margin of the 5th joint (propodos) 3 longer than that of the 4th; dactylus about $ the length of the propodos, the distal half serrate ; posterior margin curved, the distal third being minutely crenate, with a spine about one-third of the length from the carpus. The Ist joint is as long as the 5th, the 2nd and 3rd very short; the carpus has the posterior margin convex, setose, with a minute tubercle in the middle. The second pair have the propodos ovate, longer than the other joints together, having the palm defined by a small tooth two- fifths of the distance from the carpus to the base of the dactylus ; the distal portion of the palm is divided into three lobes with crenate edges by two deep sinuses. The sete on the lobesare of a peculiar form, being divided at some distance from their base into two unequal branches. The first joint is about half as long again as the next three, the carpus being very short, hollowed to receive the base of the propodos, and produced posteriorly. The perzeopoda and third uropoda resemble those of P. pul- chellus, Sars. The telson is spoon-shaped and notched at the tip to about one-fourth of its length. This last feature does not agree with Sars’s definition of the family (Paramphithoide) as having the telson ‘with the tip not incised.” Since, however, the telson itself in other respects, as well as the entire animal, agrees with Sars’s description of Parapleustes, I am not disposed to make a new genus of it. Length 8 mm. This species may be at once distinguished from all others of the genus by the shape and large relative size of the 2nd gnathopods, the disproportion between these and the first pair being much greater even than in P. latipes (M. Sars). Four specimens in the collection of the Dublin Museum of LINN. JOURN.—ZOOLOGY, VOL. XXVI. iL7/ 232 ON NEW SPECIES OF EDRIOPHTHALMA. Science and Art, taken during the expedition of the Royal Irish Academy, 1888, in 750 fath. off the S.W. of Ireland. EXPLANATION OF THE PLATES. Puats 17. Fig. 1. Leuconopsis ensifer, n. sp., male (telson vichaleia) 2in. objective. (Only one exopodite is shown.) la, Anterior margin of cephalothorax, male. 1 in. obj. 16. Third perzopod without the exopodite, male. 1 in. obj. lc. Upper antenna of female (4 in.), with the extremity highly magnified (4 in.). ld. Lower antenna of female. 4 in. 1 ¢. First pereeopod 50 J in. 1f. Second _,, 43 1 in. lg. Fourth ,, a 2 in. 1%. Telsonanduropoda ,, 1 in. Hig. 2. Apseudes hibernicus, n. sp., female. 2 in. 2a, Chelipede of the same. 1 in. 2b. Second (fossorial) leg. Po 2c. Third leg. i 2d. Fitth leg. 5; Puatse 18. Hig. 2 e. Seventh leg of Apseudes hibernicus (1 in.), with propodos (3 in.). 27. Last pleon-segment and telson from below. 1 in. Fig. 3. Stenothoé crassicornis, nu. sp. Upper and lower antenna. 3a. Maxillipedes, with basal portion. } in. 36. First gnathopod. 3c. Second gnathopod. 3d. Last pereeopod. 3e. Urus. All with 3 in. objective. Hig. 4. Parapleustes megacheir, n. sp. (pereopeds and pleopods omitted). 2 in. 4a. First gnathopod, 2 in. 4b. Second gnathopod. 2in. 400, Distal part of palm. } in. 4c. Telson and 3rd uropod. 1 in. Figs. 2,2 /, 3-3 ¢, 4, and 400 are drawn on a reduced seale. West, Newman imp. Soe Tounn Zon Vor OEM AR Hammond hth INMEEAN S IDSEH (GME S) Oley AED ANOS MatLab eI E Walker. Linw. Soc. Jounw. Zoon Vor. XXVI. Pb aye A.R.Hammond hth. West, Newman imp. NEW SPECIES OF EDRIOPHTHALMA. ON THE MARINE MOLLUSCA OF MADEIRA. 233 On the Marine Mollusca of Madeira; with Descriptions of Thirty-five new Species, and an Index-List of ail the known Sea-dwelling Species of that Island. By the Rev. Roserr Booe Warson, LL.D., F.R.S.E., F.L.S. [Read 6th May, 1897.] (Prates 19 & 20.) : THE species here described were chiefly obtained in dredgings which I carried on in Madeira between 1864 and 1874. Much additional material was sent me by my lamented friend the late T. Vernon Wollaston, as executor of the Rey. R. T. Lowe, who dredged and collected from 1826 to 1872, when, with all his treasures of botanical research, he was lost in the Bay of Biscay. Some wreckage on the French coast from the s.s. ‘ Liberia,’ in which Mr. Lowe had sailed for Madeira, was the only record of all on board. In addition to these gatherings, I have carefully gone over the Mollusca secured by Mr. J. Yate Johnson in his 30 years’ study of the various forms of life both in the deeper and shallower waters of the Madeiran sea. It is to his collection that I owe the Coralliophaga, which is by far the most important of all the species presented here. The entire list, including the 35 new species which follow, extends to 382 species. When not otherwise mentioned, the species here given have been collected by myself; I have been careful to note the fact when this has not been the case. List OF NEW SPECIES FROM MADEIRA. 1. Cylichna spreta. 19. Scalaria Fischert. 2. Amphisphyra flava. 20. —— Smithit. 3. Philine complanata. 21, Aclis vitrea. _ 4, trachyostraca. 22. trilineata. 5. desmotis. 23. tricarinata. 6. Doridium laurentianum. 24, Eulima fulva. OG maderense, 25. sordida. 8. Pleurobranchus Dautzenbergi. 26. badia. 9: Lowei. 27, —— rhaphium. 10. Nassa antiquata. 28. trunca. 11. Murex (Ocinebra) medicago. 29. —— inconspicua. 12. Trophon Lowet. 30. Odostomia omphaloessa. 13. Bittium depauperatum. 31. (Turbonilia) undata. 14. incile. 32. Schismope depressa. 15. Cecum atlantidis. 33. Montacuta triangularis. 16. Natica (Nacca) furva, 34. Coralliophaga Johnsoni. 17. Scalaria rhips. 35. Teredo Dallit. 18. aspera. LINN. JOURN.—ZOOLOGY, VOL. XXVI. 18 234 REV. R. B. WATSON ON THE Fam. ScAPHANDRIDA. Gen. CyticHna, Lov. 1.* CYLICHNA SPRETA, 0. sp. Shell smallish, oval, somewhat broad just below the middle and contracted upwards, truncated above, where the edge is carinated and thickened but not furrowed. Longitudinals: the lines of growth are very faint, delicate, close-set, and hair-like. Spirals: there are very slight and superficial microscopic scratches which extend to the whole surface. Colour dull pellucid white. Mouth club-shaped, bent, narrow above, widened but drawn out and lop-sided in front. Outer lip rising from the outer edge of the apex, bends back, and with a slightly patulous front makes a minute semicircular sweep round the top, whence with a slight divergence in its direction to the right it runs straight forward, with a barely retreating edge, till it nears the point of the shell, where scarcely patulous it sweeps round and merges with a very faint twist into the point of the pillar, continued up the body as a mere film to its junction with the outer lip. Ape, which is contracted and small, consists of the round edge which is thickened by a pad within the shell, in the centre is a somewhat abrupt minute pore.—L. 0:09 in. B. 0:05. This species a good deal resembles the young shell of C. ovata, Jeffr., but has the mouth more bent, the swell of the body- whorl lies higher and is not so large, while the apex and the front of the shell are smaller; the apical pore is also more contracted. The few specimens of this species I got from about 30 to 40 fms. off Porto Santo; they are nearly all quite young shells. Gen. AmpHisPHyrA, Lov. 2. AMPHISPHYRA FLAVA, 0. Sp. Shell globose, thin, glossy, brownish yellow, with large open mouth and a hidden spire. Sculpture: there are faint soft longi- tudinal undulations, and, like these, following the lines of growth are very fine silky striations with a mere suggestion of spirals in the structure of the shell. Colowr yellowish with a faint tinge of ruddyish brown. Spire very small, and sunk in a minute pore-like depression. Whorls: that of the body, which is alone visible, is shortly globose above, but, below this, on the left is hard twisted in round the pillar, while, in forming the * These numbers correspond with those of the figures on Plates 19 and 20. EE MARINE MOLLUSCA OF MADEIRA. 235 mouth, it sweeps widely and freely away to the right. Outer lip: it rises in the young shell barely perceptibly (in the older shell, not at all) above the apex, with very little of an advancing edge, and forming no sinus it turns at an obtuse angle and advances a little obliquely but in a straight direction to the periphery, from which, neither contracted nor expanded but retaining its forward edge, it slopes slightly backwards, and then makes a great patulous sweep round the base out to the left of the pillar, from the tip of which, with a free circular sweep, its sharp edge bends round to juin the body, leaving behind it a narrow groove and an open-mouthed, but minute, umbilical depression, above which to the apex the mouth-edge is defined by a broadish-spread, thin, whity film.—H. 0:133. B. 0°13. From Funchal Bay along the S. coast eastward to Punta Sao Louren¢go down to 50 fms.; rare. Mr. Johnson got one speci- men; I several, but chiefly young. It resembles A. globosa, Lov., but differs in colour, in smaller body, in larger opener mouth, in flatter top, and in sunken apex. Fam. PHILINID &. Gen. Puinine, Ase. 3. PHILINE COMPLANATA, N. sp. Shell thin, tumid, smooth, a little oblique in form, and with a flattened top. Sculpture: rather feeble, close, irregular lines of growth. Colour pure white, translucent, not glossy. Spire tabulated, nucleus slightly impressed, but the rest of the spire becomes gradually a very little prominent. Whorls two and a half, with a small mammillary nucleus. Mouth pear-shaped. Outer lip open: it springs from slightly below the top edge, advancing a very little, it curves at the top freely round and then runs downward straight, but in its direction a little obliquely towards the right; across the base it sweeps with a free curve, is a little patulous, and on the left rises to join the point of the pillar, which is barely if at all prominent. Inner lip more or less curved; it runs a little obliquely but in a line of increasing straightness, carrying the thin glaze spread on the body across to - the sharp-edged curved pillar, leaving behind it a minute chink and a mere suggestion of an umbilicus.—L. 0-1. B. 0-067. Funchal Bay, 50 fms. This species resembles P. finmarchica, M. Sars, but is smaller, Ss 236 REV. R. B. WATSON ON THE less tumid, and the outer lip does not, as in that species, rise above the crown except, and but occasionally, in the very young shell: the sculpture too is quite different. 4, PHILINE TRACHYOSTRACA, 1. Sp. (rpaxudorpaxos, rough- shelled.) Shell oval, but slightly truncated across the top, flattened, thin, translucent, very rough and fretted on the outside surface. Sculpture: the whole surface is very harshly roughened by a network of coarse, sharp, narrow, unequal projecting bars, which give a crusted appearance to the shell; where this network is rubbed off the surface of the shell appears like frosted glass — fretted in squares: the longitudinal bars run on the lines of growth, those which cross them are spirals. Colour: the surface made up of the bars is dead dirty white. Spire: alittle sunken; in the bottom of a small, open, funnel-shaped depression is the mammillary apex, round which coil two whorls. Suture barely impressed. Mouth pear-shaped, rather small. Outer lip rises shortly, makes a short and narrow curve at the top, runs down with a hardly prominent edge and but slight curve to the base, where it sweeps round freely (but is barely patulous) to join the point of the pillar. Inner lip: down the body from well inside the mouth this lip projects as a narrow square-set shelf, which dies off on the straight longish sharp-edged pillar, which is slightly truncate at its point.—L. 0°11. B. 0:07. Of this curiously marked Philine I got only two specimens, one of them young—both from about 50 fms. Funchal Bay. 5. PHILINE DESMOTIS, 0. sp. (decpwres, enchained.) Shell rhomboidal, but with the upper left-hand corner rounded off; flattened, thin, but not fragile, horny, hardly glossy. Sculp- ture: there are many unequal and somewhat irregular lines of growth, coincident with whose curves are very fine microscopic lines which seem to pervade the substance of the shell; crossing these nearly at a right angle are impressed chain-like lines whose links are 3 to 4 times as long as they are broad, towards the lower edge of the shell these links are more and more elon- gated, the raised surfaces between the chains are slightly wider than the chain lines, and project on the edge of the shell which they crenellate strongly above, more feebly but still more or less traceably, especially in the young shell, all along its whole edge; a slight translucent pad encircles the top of the shell. Colour MARINE MOLLUSCA OF MADEIRA. D3 white, somewhat opaque, and with a yellowish tinge. Spire not in the least prominent, the apex being small, mammiform, and slightly sunken; round it rapidly enlarging, and defined by a small slightly impressed suture, coils a single whorl whose upper narrow, flattened but rounded edge slightly overtops the apex. Mouth very large, being about 4 times as broad as the body, narrowing above, it is very wide and scoop-like in front. Outer lip perfectly straight and parallel to the axis, at the top and at the base it turns quickly but roundly and a little patulously to join the spire. Inner lip has a narrow thin appressed glaze across the upper part of the very slightly oblique body ; after the glaze dies out, the thin sharp lip runs on with a slight twist to join the outer edge of the shell.—L. 0-07 ; do. of the body-whorl 0:05. 3B. 0-056. Punta de Sio Lourengo, Santa Cruz, Porto Santo, to 50 fms. Not rare. In shape this species is very like the young of P. aperta (L.), but is slightly narrower; the outer lip does not rise so high as in that species; the body-whorl is slightly narrower, longer, and more parallel to the axis: texture and sculpture are of course quite different, and the full-grown shells are utterly unlike. Than P. scabra, Miull., this is broader, the top does not slope obliquely down to the right ; the body-whorl is narrower and not so much swoln to the right, and its direction being more oblique the shell is more narrowed upwards; the whole front of the shell below is wider, opener, more scoop-shaped. The chain-link sculpture is in a single row, and is only doubled when a fresh row first begins to appear: the links are fully 3 times as long as they are broad. Than P. catena, Mont., this is much wider, opener, and straighter . the sculpture-chains are more distant from each other, and the links are smaller and longer. P. Loveni, Malm, is much narrower, has a larger body-whorl; the outer lip falls much short of the apex, the chain-sculpture is double, and the links are much shorter. P. cingulata, Sars, and P. guadrata, Wood, differ much from this both in form and sculpture. Fam. DoRIDIID&. Gen. Dortptum, Weckel. 6. Dortp1um (?) LAURENTIANUM, n. sp. (Not figured.) Body unknown. Shell small, but relatively to its size strong, somewhat flattened 238 REV. R. B. WATSON ON THE on both faces; outline trapezoidal, the front and back lines being fairly parallel, though somewhat converging upward (i. e. where the head of the animal would be); the lower edge is rounded; the top of the shell is produced into a short oblique wing-like extension, from which four to six short small bluntish points project. This wing-like projection slopes downward to the left towards the somewhat unformed nuclear top of the pillar, from which it is separated by a small rounded sinus with a reverted edge; below the nuclear pad the lip-edge is broadly but shortly reverted, leaving behind it on the left a well-marked very oblique umbilical furrow ; the whole mouth-edge is strong ; the front face _ of the shell is slightly concave, the back is rather flatly convex. The surface of the shell is somewhat rudely marked by lines of growth, and microscopic, close-set, faint, regular longitudinal markings can just be traced; the colour is translucent white, only the embryonic knob is brownish and of a limy texture. The living animal I never found, and only four minute shells presented themselves in dredgings of 50 fms. from Punta de Sao Loureng¢o, whence the specific name is taken.— L. 0:06. B. 0-036. They are probably Doridiums. Of the four specimens, two — were unfortunately lost by a friend. 7. DorIDIUM MADERENSE, n. sp. (Figs. 7, 7 a, 76.) Body about half an inch in length and a third of an inch in breadth ; it is well arched; the foot is oval in contour, with the mantle-covered square-fronted body just perceptibly projecting in front; while the tail, which is short, square, abruptly truncated, hardly bifid, but with a slight tubercular prominence at either corner, is more prominent behind; there is no flagelliform appendix; the entire head is covered by the unbroken mantle which extends over the body, either side of which is inlapped by the mantle-flaps, as in Philine ; in the substance of the tail the shell can be distinctly seen with the spire directed towards the animal’s head, while in the body, rather on the left side, can be traced (what is apparently the stomach and entrails) a slightly opaque, elongately oval substance defined by a brownish outline, but varying slightly in form and position. I failed to detect either the branchial plume or the vent, though there is no mem- branous lobe to hide these asin D. carnosum, Cuv. (see Veyssiere, Tectibranches, p. 48, pl. ii. 42). Colour: the whole animal is opaque (or rather not quite translucent) white, dotted, especially MARINE MOLLUSCA OF MADEIRA, 239 on the foot, with a very few and extremely minute specks of erimson, and each of the two tail-tips has a bright but minute erimson fleck. Preserved in spirit, the animal’s colour changed to a uniform brownish-black. Shell: in shape like one outspread wing and tail of a bird which has been cut longitudinally through the middle; in the centre there is a small thickened nucleus, with a short pillar and an open coil of one and a half whorls. In substance, colour, and ornamentation it is much like the shell of Philine scabra, Mill., or P. punctata, Clerk. A fine but distinct marginal bar of somewhat varying strength, beaded on the outer face and pro- jecting sharply on the inner face of the shell, runs round nearly the whole edge, and three or four others occur at somewhat irregular intervals between the upper and lower edges of the shell. Besides these there are many spiral and longitudinal undulations with much finer rounded spirals.—L. (body) half an inch, B. 0°3. Shell, L. 0:09, B. 0-065. Funchal, Punta de Saio Lourengo, Canical. From shallow water to 50 fms. The shells of this species I found not unfrequently, but only one living animal presented itself. It lived with me for some days. I never saw it swim—it always crawled, and then never on the surface of the sand of the bottom; as soon as it could it passed in just below the surface, burrowing not deeply, but forcing the sand to rise and pass over its back in a way very curious. Cf. Acera marmorata, Cantraine, Malac. Médit. p. 73, 1. 2. Doridium carnosum, Cuvier, and D. membranaceum, Meckel, Veyssiére, Moll. Opisthobranches, pp. 44-49, pl. i1. 42-47. Fam. PLEUROBRANCHIDG. Gen. PLEUROBRANCHUS, Cuv. 8. PLEUROBRANCHUS DaUTZENBERGI, 0. sp. Shell oval, flattened, fairly strong, white, dull outside but glossy within. Sculpture: strongish, irregular and unequal lines of growth, with longitudinal striz, similar to but keener than those of P. plumula; the oblique furrow of that species is entirely absent. Colowr semitransparent to opaque yellowish white. Spire: a small rounded knob lying well over to the right and bent down on the back of the shell so as to be quite 240 REV. R. B. WATSON ON THE dominated by the further growth of the first whorl, which rises up on the left and forms the most prominent part of the shell at this end as it turns over with a sharp twist and encloses a small pear-shaped body-space within the ianer lip, which, with a sharp finely-projecting edge, sweeps freely round and, slightly expand- ing, runs straight down to the base, where, scoop-like and with a semicircular curve, it passes round to form the sharp, barely curved owter lip, which, parallel to the other, rises to or nearly to the full height of the apex, leaving, however, a small nick at its Junction with the apical whorl.—L. 018. B. 011. A well-marked species, to which I have attached the name of M. Ph. Dautzenberg, who, in conjunction with M. Henri Fischer, is so ably working out the Prince of Monaco’s dredgings in the ‘ Hirondelle’ and the ‘ Princesse Alice’ at the Azores. The species is not very common. I have the two specimens of it got by the Rev. R. T. Lowe in 1829 at Magdalena. Another of his specimens and a few of mine came from near Funchal, but most of mine and one of his came from the east end of the south shore towards Punta de Sao Lourengo. Mr. Lowe has a note on the living specimen which he got at the Gorgulho, near Funchal, that the ‘“‘animal was white, subpellucid, about 3 in. long,” that “ the edges of the mantle were loose and floating, as in Stgaretus’’; the branchiz he says he had not seen. 9. PLEUROBRANCHUS LOWEI, 0. sp. Shell strong, shallow, long, narrow-pointed and gathered in at the caudal end, broad, flat and open at the cephalic end, with nearly straight sides. Colowr white and translucent. Sculpture fine sharpish regular lines of growth; of longitudimals, a sug- gestion is traceable under the microscope in exceptionally good light—~a very faint trace of such a diagonal impression as forms so marked a feature in P. plumula, Mont., is visible on the outside and shows more distinctly in the interior. Spire consists of 33 well-rounded whorls parted by a distinct impressed, slightly marginated suture ; the apex lies flat on the back of the shell and a little behind the extreme point; the tip is minute, rounded, and glossy, of a pale ruddy-brownish colour; from this point the two sides slope away as a cloak hangs on a peg, and the back of the shell is here well rounded. Mouth spreads fully open but for the small, long, narrow curved chamber which lies in under the spire and the edge of the inner lip. Outer lip rises MARINE MOLLUSCA OF MADEIRA. 241 almost to the point of the shell, runs slightly obliquely towards the right, then, with scarcely an angulation, it runs straight forward, sweeps quite patulously in a semicircle round the front, and so with a very gentle curve the inner lip, sharp-edged and a little prominent, rises to the top, where it makes a quicker curve, and then turns almost at a right angle in the line of axis to join the base of the spire, where it leaves but half conceals a small shallow umbilicus, round which the lines of growth appear as strongish wrinkles, and the traces of longitudinal strie in exceptionally good light seemed faintly more definite.—L. 0:35. B. 0°17. Of this species Mr. Lowe, in 1827, got one good specimen at Labra, near Punta de Sio Lourengo, to the east of which I got a quite unmistakable fragment. These, so far as known to me, are the only representatives of this well-marked species. Fam. NassipD@. Gen. Nassa, Lam. 10. Nassa ANTIQUATA, N. sp. Shell small, strongish, pale brownish white flecked with chocolate-brown, compact, with a rather tall conical spire, a small closely-coiled turbinated apex, a short conical rounded. base, avery short butstrong pillar. Sculptwre—Longitudinals: on the first three normal whorls are about 9 rather feeble ribs, helped to prominence by a row of nearly adjacent small flattened tubercles which run down them; on the lower whorls these riblets disappear and only close sharp lines of growth appear, which come to strength on the base but especially behind the outer lip, where there is a flat strongish rather remote callus. Spirals: below the suture there is a small, flat, slightly raised ribbon defined on its lower side by a distinct stippled furrow; this ribbon and furrow begin to show on the third regular whorl, but are barely traceable on the external callus of the outer lip ; on each of the upper regular whorls there are three pronounced rounded threads, rising where they cross the ribs into tubercles and separated from one another by narrow furrows; these spiral threads are barely traceable on the intermediate whorls, but to the number of 9 or 10 become strong and crowded on the base; at the point is a strong furrow, and below it a thread twisting round the pillar and defining the canal; below it the pillar is scored by about six threads and weak furrows. Colour a very pale ruddy brownish 242, REV. R. B. WATSON ON THE white, which is porcellanous on the pillar and inthe mouth. Spire somewhat tall, conical, with hardly convex profile-lines. Apex small, pellucid white, consisting of three compact, turbinated, slightly swoln, smooth whorls, the extreme tip of which is very small. Whorls 9, conical, almost straight-sided, not constricted below, with a short rounded base. Sutwre linear, scarcely im- pressed, but distinct from a slight prominence of the subjacent whorl. Mouth oval, short, not expanding, channelled above but not in the lip; the canal at the pillar is narrow and deep, it is inclined to the left and widens as it advances. Outer lip nearly semicircular, but its curve is slightly flattened above and in- tensified in passing round to the base; it has a little way within the mouth about 10 teeth, of which one at the top, two in the middle, and one at the edge of the canal are strongish; the others are small, but all run some way into the mouth. Inner lip well reverted and thickened, with an abrupt edge across the body; on the base and down the pillar it leaves a small umbilical chink behind it ; just within the mouth it bears a few flattened tubercles, which vary in number and size; the tip of the pillar is well flanged along the canal-edge.—L. 0°45. B. 0°22. Rare—Funchal and Cabo Girao (Lowe, Watson). None of the Wassas of our seas resemble this species, differing as it does from them in its narrow compact form and in its sculpture. Judging from figures, IV. glabrata, A. Ad., from the Pacific is a good deal like in many ways, in some of its forms especially. Amidst the inconceivable multitude of Ligurian Tertiary forms figured by Bellardi it seemed probable that this species might be found, but nothing satisfactory presents itself there, nor elsewhere has the search among Tertiary fossils been successful, though diligently prosecuted. I do not attempt a comparison of it with NV. semistriata, Broc., nor with N. labiosa, J. Sow., nor with the N. corniculum, Olivi, nor with the UN. tri- fasciata, A. Ad., nor with the NV. Gallandiana, Fischer, for the relation of these species has become more and more impossible for off-hand treatment. Fam. MuRicipD2. Gen. Murex, L. 11. Murex (OcrnEBRA) MEDICAGO, 0. sp. Shell strongish, pale brown, rather stumpily biconical, with a MARINE MOLLUSCA OF MADEIRA. 243 tallish scalar spire, spinous whorls, a small but coarse tip, a shortish base, and a flattened snout. Sculptwre—Longitudinals : there are no varices, but on each whorl 7 to 8 quite distinct rounded ribs, which are made prominent by the small, open, internally imbricated, upturned prickles, of which a prominent series crowns each whorl at its shoulder a little below the suture ; others similar, but smaller, crest each spiral as it crosses the successive longitudinals ; the whole surface is covered with sharp, thin, continuous forward-facing lamine, not close-set but some- what crowded and imbricated on the crest of the ribs. Spirals: there are 5 to 6 well-marked spiral threads on the last whorl, about 3 on the penultimate, and 2 on the earlier whorls; the first, which forms the shoulder, lies a little remote from the suture ; the twisted snout is scored by the strong-vaulted old- canal scars. Colour pale slightly ruddy brown, paler on the tip and on the spines. Spire rather high, conical, subscalar. Apex is small, but not fine nor sharp; it consists of 14 whorls, which in very young and perfect specimens appear faintly ribbed ; the extreme tip is semi-immersed. Whorls 7 to 8 in all; they are flattened and very slightly declining from the suture to the shoulder, from below which they are rounded and contracted on the base to the somewhat small flattened and triangularly-shaped snout, the point of which is slightly reverted and emarginated. Suture indistinct. Mouth narrowly oval, rounded and patulous above; in front there is a rather narrow, oblique, open canal. Outer lip barely patulous, well arched, thin on the edge ; within it has a feeble varix near the edge with 5 to 6 rounded but rather feeble teeth. Inner lip with a slightly projecting thin edge; it forms a continuous curve across the body and then straight down the pillar, the point of which is cut off abruptly and obliquely at the canal, over which the edge projects sharply but without closing it in; between this edge and the old canai-swelling is a very slight and open furrow.—H. 0°6. B. 0°3. Notcommon. Madeira, Punta de Lourenco to 50 fms.; Magda- lena (dredged) ; Selvagem Grande, shore (Lowe, Watson). This species somewhat resembles Zrophon Lowei, Wats., but the shape of the whorls, of the apex, and of the mouth, as well as the sculpture and colour, are different. Murex cristatus, Broc., resembles it most, but is narrower and differs in shape of whorls, in sculpture, and in colour. 244. REY. BR. B. WATSON ON THE Gen. Tropuon, Montfort. 12. TropHon Low4t, n. sp. Shell strong, biconical, with a straight-sided spire and a very small glassy turbinated tip; the snout is shortish but small. Sculpture—Longitudinals : there are 7 strong, rounded, slightly shouldered ribs which run continuously down the spire; they rather tend to multiply towards the tip, where they pass into being round tubercles; they are parted by slightly narrower, shallow, rounded and open furrows; these, as well as the-ribs, are closely scored with small superficial harsh lines. Spirals : there are on the ‘body-whorl 12 to 15 narrow, slightly raised threads, roughened but not squamated by the crossing of the longitudinal threads; they are chestnut-tinted, with flattened interspaces of about twice their breadth; in each of these le more or less distinct smaller threads which are occasionally double ; a fine rounded thread encircles the top of each embryonic whorl, but tends to die out in the later whorls. Oolowr yellowish grey, with chestnut tint on the spirals. Spire straight-sided, conical, slightly higher than its breadth above the body-whorl ; in its upper part slightly scalar ; the tip is sharp, very small and prominent, and consists of the 4-whorled glassy turbinated embryo-shell. Whorls 7 besides those of the embryo-shell ; they are slightly shouldered below the suture, between which shoulder and the suture they are slightly hunchy; the last is rather small and slightly flattened. Suture not sunken, linear, undulated. Mouth small, narrow, oval, pointed at both ends, not expanded, exactly half of the shell in length. Canal rather narrow, short, turned back and to the left behind the pillar. Outer lip very slightly curved, not patulous, with a thin but strong straight edge, strengthened externally by a strong varix-like, somewhat remote rib, and internally by a strongish pad on which project 6 stumpy, somewhat elongated teeth. Inner lip flat, sub- patulous, extremely thin, but with a very slight straight edge, which becomes a little stronger towards the point of the pillar ; there a very small furrow appears behind the lip just where it takes a slight twist; at its upper end, just before it is joined by the outer lip, a feeble tubercle occurs, helping to define an indistinct gutter-lke sinus, which does not cut the outer lip-edge. Operculum small, narrow, slightly curved, ruddy chocolate in colour; the apex is small and lies, when extruded, towards the A en IR a MARINE MOLLUSCA OF MADEIRA. 245 animal’s head, when retracted it points towards the canal.— L.0°8. B. 0°34. Not common. Madeira, 50 fms.; Labra; Punta de Sao Lou- renco (Lowe, Johnson, Watson). This is the species which McAndrew gives at p. 40 of his Report as living on the “ rocks” of the ‘‘shore;” “ rare, species obtained in Canaries ;” see also J. c. p. 82, where he says of it “rare—white.” There are unnamed specimens of the species in his collection both at Cambridge and in the British Museum. On the shore-rocks I certainly never found it. Mr. Jobnson’s one young specimen came from a deepish-water coral, and Mr. Lowe’s were dredged in Labra from a depth of some fathoms : but one or two of them are occupied by hermit-crabs, whose presence seems to indicate shallow water. This species most nearly resembles JT. fusulus, Broc., but it is narrower, flatter, with a narrower mouth; the sculpture is very markedly different. It has some likeness to Murex cristatus, Broe., and very worn specimens may be easily confounded, but the two species are unmistakably different. I feel by no means sure that this is not IL productus, Bellardi, Moll. Terr. terz. Piemonte, pt. i. p. 99, vil. 6; but bis description is very vague, his measurements suggest a larger and narrower shell, and no specimen is available. Fam. CERITHIID ®. Gen. Brrrium, Leach. 13. BITTIUM DEPAUPERATUM, 0. sp. Shell: a tall narrow cone with a contracted and produced base, barely translucent, slightly glossy, of a uniform pale yellow colour, occasionally palely brown banded. Sculpture: there are weak spiral threads, of which, on the body-whorl, five lie above the periphery and four on the base; these last are not cross- hatched but simple; the furrows which part these ridges are shallow and narrow; the number of the ridges and furrows is fewer on the earlier whorls, on the 2nd whorl they are only two; the threads are studded with rather blunt round tubercles which run in continuous and somewhat diagonal lines across the whorls —about 20 of these lines are on the last whorl; parallel to these cross-lines the whole shell-surface is feebly and microscopically undulated. Colour yellowish white, with occasionally a palish 246 REV. BR. B. WATSON ON THE ruddy-brown band on the lower two spiral threads. Spire fine and regular with very straight outlines—the extreme tip though small ends somewhat abruptly and flatly. Whorls 10, flat on the side, of very regular increase. Suture neither broad nor deep, but well marked—from the distinct though small con- traction of the lower part of the superior whorl. Mouth small, oval, angulated above, and having a small shallow gutter at the point of the pillar. Outer lip flatly and regularly arched; the tubercles of the exterior sculpture very slightly and indeed rarely affect the inner surface. Inner lip thin but distinct, turned back and appressed on the pillar, where it is perpendicular with a slight twist resulting in a small tuberosity in front at the upper edge of the gutter.—L. 0°24. B. 0-08. This is a much slimmer form than B. reticulatum, da Costa, with also a blunter apex; slimmer, too, than Bzttiwm incile, Wats., and with much straighter contours and narrower base, and with smaller tubercles on the more numerous spirals. The species has a far-off but yet distinct suggestion of a Cerithiella. Very rarely a feeble varix appears on the last whorl. This species I found abundantly on the south coast of Madeira and at Porto Santo, but it does not seem to have presented itself to other collectors. 14. Brrrium Incite, n. sp. (Figs. 14 & 144.) Shell a tall narrow cone, but somewhat coarse both in sculpture and in its proportions; its contour-lines are very slightly curved, and from the periphery the base contracts slowly ; its whorls are glossy, strongly defined, and almost turreted. Sculpture: the bluntly rounded apical whorl is microscopically and very faintly spiralled and longitudinally marked; the three following whorls have 2 and the succeeding four whorls 3 strong spiral ridges; on the base below the periphery there are 3 of these ridges, the last forming the bulge of the column ; none are varicose, they are all strong but not very projecting, the appear- ance of prominence which on the spire they present being largely due to their being set with coarse rounded tubercles some 16 on each ridge, arranged in very regular longitudinal lines across the whorls; the longitudinal grooves which part them are broader but hardly so deep as those which separate the spiral ridges ; the whole surface of the shell is microscopically but roughly fretted with sharpish longitudinal and ruder spiral lines; on the base MARINE MOLLUSCA OF MADEIRA. 247 below the tubercles lies a small plain encircling thread, with a small narrow groove, within which lies a stronger ridge, these two ridges and the furrow between them are pale in colour, sometimes speckled; within the last of these ridges is a strong furrow, and then the strong ridge forming the twisted columella ; this ridge and the furrow beyond it are stained deep chestnut. Colour varies from dark brown to ruddy chestnut, with a whitish band round the top of each whorl occupying the highest and extending sometimes to the 2nd spiral ridge, with an occasional intrusion to the ridge-tubercles here and there. Rarely the shell, though quite fresh, is pure dead white. Spire rather stumpy for the genus; the apex, though small, is not drawn out, and ends in a small, rounded, half-immersed tip. Whorls 9 to 93, rarely 10, nearly flat on the side, of slow and very regular in- crease ; relatively to the axis of the spire the longitudinal ridges run a little transversely. Suture very strongly marked, but its in-girdling appearance is due, not so much to its depth and breadth, as to the way in which the succeeding whorl projects below it. Mouth irregularly rhomboidal, with a small gutter rather than a notch at the point of the pillar. Outer lip straight and sharp, very slightly indented on the base, where it sweeps round with a semicircular curve to the point of the pillar, which leans away from it diverging slightly from the line of the axis. Inner lip has on the pillar a thin but well-marked projecting edge, which thins across the body but recovers its strength at the upper corner near the outer lip. Operculwm small, elliptical, thinnish, paucispiral, with a central nucleus; the outer surface is closely scored with fine, curved, radiating lines densely crossed by a minutely microscopic tissue whose lines show the curves of growth.—L. 0:22. B. 0-075. Madeira, Porto Santo, Selvagens, Grand Canary. This species is found very abundantly. Mr. McAndrew, however, does not refer to it, nor did I find it in other collections. It is the same as a species sent to me from the Mediterranean as B. lactewm, Phil., but which is, I think, distinct from that species ; the longitudinal spiral and basal threads are the same in number, but in B. incile the apex is smaller and more sunken, being neither so much produced nor so scalar as in that other; the contour-lines of the shell, too, are distinctly convex, not straight ; the last whorl is more contracted, while the base is attenuated and rounded, not square. 248 REY. R. B. WATSON ON THE Fam. C #c1pD &. Gen. Czcum, Flem. 15. CHCUM ATLANTIDIS, 0. sp. Shell very small, thin, transparent, of a dull glossy white, a good deal bent, subannulated. Sculpture: there are towards the tip a few sparse, somewhat feeble, but rather sharply topped, encompassing rings, which die out on the concave curve of the shell; one or two similar but feebler rings appear near the mouth; on the intermediate space undulations rather than rings are traceable; the whole surface is marked by faint lines of growth, and excessively minute, sharp, densely crowded microscopic longitudinal strie. The apex is closed by a flat somewhat impressed plate, from the forward side of which pro- jects a short, blunt, laterally compressed triangular plug. The mouth, which is circular, is neither contracted nor patulous. Operculum thin, impressed, brown, with circular lines of growth. —L. 0085. 5B. 0:022. ©. vitreum, Carp., is larger than this species, and lacks its sculpture and circular rings, and has contraction of the lip. The plug in C. atlantidis is like that of C. trachea, Mont., but the shell is much smaller than in that species; the circular rings are not close-set but are well parted; their tips are not flattened but sharp or rounded. C.atlantidis, without being rare, is certainly not common, and was not found in the collections of Lowe, McAndrew, or Johnson. I got it only at Porto Santo. For the beautiful figure of this species I am indebted to the Marquis de Folin, who was kind enough to draw it for me. Fam. NaTicip a”. Gen. Natica, Adans. 16. Narica (Nacca) FURVA, n. sp. Shell white, with two dark smoky bands which are some- times absent but sometimes occupy nearly the whole shell, which is pretty strong, depressedly conical, with a small but a Jittle raised spire and rounded whorls; the last of these is large, with an open mouth; a continuous white porcellanous pad fills the upper part of the mouth and nearly chokes the umbilicus. Sculpture—Longitudinals none, but lines of growth which, MARINE MOLLUSCA OF MADEIRA. 249 however, from the suture a good way downwards present them- selves as well-defined rounded curved riblets parted by narrower but open furrows; these riblets cease rather suddenly and pass into very faint lines of growth. Spirals none. Colour porcel- lanous white, somewhat translucent except round the whorls below the suture; below this dead-white band lies a broadish smoky-brown band, which in the upper whorls encircles their base ; in the body-whorl below this dark band a white band of about the same breadth occupies the periphery ; below it on the base is a slightly narrower dark band ; the whole base round the umbilicus is white, but there is often a rusty tinge in the umbilicus and on the edge of the umbilical pad; there are colour- variations from uniform pure white to dark brown, with a pale base, but the spire has always a dark tinge. Kpidermis: there are traces of a hard, corneous, yellowish-brown integument. Spire unusually small, but well-exserted and with a minute prominent dark tip. Whorls 43, not in the least angulated or gibbous; those of the spire are unusually small; the body-whorl is large, and gives breadth to the shell in spite of being longitudi- nally and obliquely drawn out. Swtwre scarcely oblique, linear. Mouth semi-oval, open, long rather than large, the whole plane of its edge retreats extremely from above to the base ; its height is nearly four-fifths of the whole height of the shell. Outer lip thin, well arched, retreating to the base, but from that point advancing slightly to the pillar. nner lip oblique, nearly straight, thickened by a broad white porcellanous pad which fills the whole upper corner of the mouth, projecting there in a blunt, low, rounded prominence; the face of this pad projects bluntly all the way to the point of the pillar, it crosses the body with a straight well-defined edge whose direction is oblique; where it quits the body to join the rust-stained umbilical pillar- pad it is more or less deeply cut by the umbilical furrow which sometimes feebly, sometimes very strongly, twists out round the pillar, circumscribing it markedly but failmg to cut in on the edge of the inner lip, which here to the point of the pillar is shortly reverted and slightly thickened. Umbzlicus is sometimes a mere depression, but normally is a strong but rather narrow furrow coiling round the pillar and deeply penetrating the middle of the shell. Operculum calcareous, pure white, pretty strong, lustrous, fairly flat, but slightly padded in the nuclear region, from which a very slight rounded swelling curves with LINN. JOURN.—ZOOLOGY, VOL. XXVI. 19 250 » REV. R. B. WATSON ON THE the progress of growth ; outside of this swelling lies a small shallow, open, but well-defined furrow, beyond which on the extreme edge of the operculum rises a single small, simple, narrow flange.—H. 0°47. B. 0°42. The marked features of this species are the curved radiating riblets below the suture, the two smoky bands, the small dark prominent spire, the sharp apex, and the entire absence of the slightest trace of the reddish-brown flammulations some trace of which is always visible in J. variabilis, Récl., and N. Dill- wynit, Payr. Itis very common from the whole coast of Madeira and of Porto Santo, from 10 to 50 fms. I found it sparingly represented in the collections of Mr. Lowe and of Mr. Johnson. Fam. SCALARIIDS. Gen. Scatanta, Lam. 17. SCALARIA RHIPS, n. sp. (pi, wicker-work.) Shell small, conical, rather strongly but sharply ribbed and spiralled ; carinated round the smooth and flattened base, a high narrow spire, depressed whorls which are rounded at their profile, an impressed suture; it is dull, somewhat ruddy, and has a small, conical, fine-tipped, pale chestnut-coloured, smooth em- bryonic apex. Sculpture—Longitudinals: there are on the 5th (i. e. last) whorl 22 prominent but narrow palish ribs cut off at the basal carina, separated by smooth surfaces of double their breadth ; they follow closely the plane of the axis, and slowly diminish in number all the way up the spire. Spirals: there are on the body-whorl 5 prominent spirals very similar to the longitudinals, beneath which they pass without forming knots but throwing out these others somewhat sharply ; of these spirals the two at the periphery are stronger than the others and slightly more distant from each other; the wide flattened base (whose edge projects slightly beyond the curve of the last whorl, and thus forms a carina) is microscopically scored with curved radiating lines and with much ruder but fainter spiral threads. Colowr somewhat ruddy, with rather paler ribs and base, the pillar being almost white, while the apex is glossy chestnut. Spire high and narrow, with barely convex profile-lines. Whorls 52 exclusive of the apex; they are rather short, with a rounded profile. Suture strong and deep, very little oblique. Apex a very beautiful glossy pale chestnut little cone of 4 whorls, which F MARINE MOLLUSCA OF MADEIRA. 251 are microscopically marked with spiral and longitudinal striz; the extreme tip is shghtly immersed, small, bluntly rounded ; the basal whorl of these four is slightly tumid, with a minute, flatly spreading edge, from within which the regular growth of the shell begins abruptly. Mouth slightly elliptical. Zip not fully developed.—H. 0°14. B. 0-06. This is a singularly beautiful little shell; among the Scalarias of the Italian tertiaries there are some species that faintly recall the Madeiran form; among living species 8. decussata, Kien. (not Pease), alone has some slight features of resemblance, but they are very slight. The sculpture is like wicker-work, and hence the name of the species. I got only one young specimen from 50 fms. in Funchal Bay. 18. ScALARIA ASPERA, 0. sp. Shell ruddy, small, high and narrowly conical, roughly fretted on the whole surface, strongly ribbed, tuberculately carinated within the rounded base, and with an oblique, impressed, and crenulated suture. Sceuwlptwre—Longitudinals: there are on each whorl up to the very top 10 or 11 strong, narrowish, rounded not outspread ribs, running continuously with a slight trend to the right from whorl to whorl down the spire, on either side of which is one stronger than the rest, that on the outer lip being peculiarly prominent ; the interspaces are rounded and slope up the sides of the ribs ; towards the top of the spire they are nearly crowded out by the ribs. Spirals: the whole surface of the shell is covered by irregular, unequal, flattish rounded threads, of which about 30 of the larger can be counted on the body-whorl; but there are others smaller past counting; all these spirals, both larger and smaller, are sharply densely crenellated by tooling produced by minute tubercles which often run over into fine longitudinal threads; well within the base round the pillar runs a very coarse string-keel which towards the lip is strongly con- tinuous and rises on the ribs in low swoln rounded tubercles. Colour a dull ruddy hue as if dusted over with a palish powder. Spire high and narrow, with very straight profile-lines. Whorls 7, exclusive of the embryonic apex which is broken off; they are small but not short, slightly hunchy beneath the suture, but very slightly convex below. Suture oblique, impressed, and rather strongly crenulated. Mouth circular, but a little flattened on the upper inner side. Zip a minute sharp, not expanded 1Q* 252 REY. B. B. WATSON ON THE flange projects round the edge of the mouth, attached as a shelf across the body ; it projects minutely down the pillar, but leaves no chink behind it.—L. 0:2. B. 0:08. From 50 fms., Funchal Bay. This species has so much character, that though I found only one possibly young and not quite perfect specimen, I do not think it will ever be difficult to recognize. It is more like S. fortzlis, Wats., ‘Challenger’ Gaster. p. 139, ix. 1, than any other I know; but that species has a shallower and less oblique suture with flatter-sided whorls, and the whole sculpture, but especially the basal keel, is totally different. WS. funiculata, Wats. (J. ¢. p- 141, ix. 4), is still more divergent. It is certainly not the S. crenulata, Linn., of the Canaries, nor the S. Hotessieriana, dOrb., of Cuba. It belongs to the very marked and curious group which, besides those referred to here, includes the S. longis- sima, Seguenza, the S. torwlosa, Broc., and several others given by Sacco from the Italian tertiaries, and by Deshayes from the Paris basin. 19. ScaLaRta FISCHERI, n. sp. Shell small, delicate, translucent, with short, rounded, de- pressed whorls, a scalar spire whose whorls are each rather more contracted at the bottom than at the top, fine close- set spurred ribs and very distinct spirals, a strongly impressed, not very oblique suture, and asmall base. Sculptwre—Longitu- dinals: 23 or 24 thin projecting riblets, which crowd the surface, and from each of which close to the suture rises a small tooth, often broken off. Spzrals: of these some 17 to 20 can be counted on the 2nd last whorl just above the corner of the mouth; they are rounded, well raised, can be traced as they cross the riblets, and are, like the whole surface of the shell, fretted with minute longitudinal scratches. Colowr translucent white. Spire high and narrow. Whorls 7, exclusive of the embryonic tip; they are short, with a rounded profile, but lie like somewhat oblique slabs from the depth and straightness of the suture and the ex- pansion of each whorl below the suture in consequence of the minute projection there of the tooth which crowns each rib. Suture deep, strongly marked by the flat shelf below it on the top of the succeeding whorl ; it runs somewhat obliquely but in a very straight line across the shell. Apex has a peculiar almost metallic sheen, is a very perfect, rather high, fine-pointed cone, > MARINE MOLLUSCA OF MADEIRA. 253 which, as set on the spire, slightly diverges from the axis of the shell ; it consists of 4 full whorls, flat-sided, glossy, faintly striated longitudinally, which are parted by a barely impressed, slightly chestnut-tinged suture. Mouth very slightly elliptical in the axial line, the pillar being little curved. Zp patulous all round and flat-edged, except just where it crosses the body-whorl, and is there attached so as to leave no umbilical chink whatever. —H. 0:25. B. 0:09. My specimens are nearly all young, but the size was measured from the juxtaposition of 2 or 3 larger, well-preserved fragments. The apex, the suture, the shape of the whorls, the stumpier form, the crowded longitudinals bespurred close to the suture, the erisper spirals, and the absence of any approach to an umbilicus, markedly differentiate this species from S. Smithii, which super- ficially it seems very like. Not rare, from deepish water along the S. coast from Funchal to Punta de Sao Lourenco; other collectors did not meet with it. I have named the species in remembrance of my much lamented friend Dr. Fischer, of the Jardin des Plantes, Paris. 20. ScaLarita SMITHTI, n. sp. Shell small, with short, rounded, slightly depressed whorls, a spire which is a little scalar, fine translucent mucronate ribs, very delicate spiral threads, a slightly impressed little-oblique suture, and a rounded umbilicated base. Sculptwre—Longi- tudinals: on the lower whorls there are about 20 fine sharply projecting translucent riblets, which run in a slightly oblique discontinuous line from whorl to whorl down the spire, on the earlier whorls they are somewhat fewer in number; on each riblet, slightly above the periphery, there projects a small nearly right-angled tooth—very often broken off. Spirals: there are about 15 to 20 very fine rounded little raised threads, which tend to become fainter about the periphery of each whorl. Oolour saccharine white. Spire high and narrow. Whorls 6 (exclusive of those of the apex); they are rather short and tumid, markedly broader below than above, and are of very regular increase. Suture fairly impressed and rather oblique. Apex: 3 complete rounded whorls form a small high regular blunt cone on a small base set on a little to one side of the axis of the spire; these whorls are microscopically barred longitudinally. Mouth almost quite round, not small. Lip sharp-edged, not patulous, 254 REV. BR. B. WATSON ON THE level-fronted, slightly detached (in the full-grown shell) from the body, and leaving behind it a small distinct funnel-shaped umbilicus.—H. 0:17. B. 0:07. This species very much resembles 8. Fischeri, but is distinet. It is a larger form, the spire is in proportion broader, the whorls are rounder and less depressed; the apical whorls are 3 not 4, they are smaller and the way in which they are set on the top of the spire is peculiar; the spiral threads are sparser, flatter, finer, and more rounded, and at the periphery somewhat fainter ; the longitudinal ribs are fewer and sparser ; the minute tooth on these, when present, occurs lower on the whorl (that is, nearer the periphery), while in S. Hschert it comes very near the suture ; the base of the embryonic apex is much smaller. J have named the species after Mr. H. A. Smith, of the British Museum, whom often, and always profitably, I have had occasion to consult. Specimens, several in number, were mixed up with shells of S. Fischeri I had got from Funchal and Punta de Sio Lourengo and also from Porto Santo, Madeira. Gen. Actis, Lov. 21. ACLIS VITREA, 0. sp. Shell thin, hyaline, glossy, tall and narrow. Sculpture: there are very faint unequal lines of growth ; a very feeble spiral angulation is sometimes traceable round the base of the whorl. Colour glossy, transparent white. Spire high, with very regular narrow outline and a bluntish rounded half-immersed tip. Whorls 7; their curve is a very regular flattened arch. Suture shallow, oblique. Mouth a full round oval, small. Outer lip thin, prominent, but toward the body it is drawn back almost into a sinus, somewhat expanding on the base. nner lip thin, sharp, and patulous; its connection across the base with the outer lip is long and very filmy. Umbilicus a small funnel-shaped shelley depression.—L. 0°08. B. 0:029. This species is a good deal like A. Walleri, Jeffr., but is much smaller, with 7 instead of 10 to 11 whorls; the spire is much broader, the tip much larger and coarser in proportion to size, the whorls are longer, more regularly rounded, much less ex- panded round the axis, and the contour of each is more com pressed, the suture is more oblique, the mouth smaller, lip open rounder. This species is abundant in deep water from Funchal to Punta MARINE MOLLUSCA OF MADEIRA. 255 de Sao Lourengo, and from Porto Santo. The animal is dark green in colour. I have met with it nowhere but in my own dredgings. The shell somewhat resembles that of Aclis Gulsone, Jeffr., but is very much smaller, the spire is much more contracted and ends in a minute knob of a point, the outer lip is inflected instead of being prominent and expanding, the suture is much deeper, and the whorls are more prominent and rounded. 22. ACLIS TRILINEATA, 0. sp. Shell thinnish, semitransparent, rather glossy, longish, narrow with squarish outlines. Sculpture: there are on each whorl 3 very strong rounded but slightly crested threads, absent on the embryonic tip, faint on the two succeeding whorls but well- marked on all the others—no others appear on the base; the Ist, which is the strongest, lies a little remote from the suture and forms for the whorl a well-marked shoulder; the 2nd, though helped by the bulge of the whorl, is barely more prominent than the first; ‘the 3rd is slightly feebler than the others, and les near but quite clear of the suture: besides these there are faint close-set microscopic striations and vaguer traces of longitudinal markings; the furrows between the threads are shallow and rounded. Colour white and semitransparent. Spire high and narrow, ending a little abruptly in a small rounded, not pro- minent tip. Whorls 63. From the suture there is a downward sloping shoulder to the first thread, from which the very straight contour-line runs down parallel to the axis, and with scarcely any contraction into the suture below. Suture broad, open, and shallow, rather oblique. JMowth oval, rather large. Outer lig thin, with a free convex sweep. Inner lip thin, shar p, and prominent, with a well-rounded curve it spreads very thinly across the body. Umbilicus a shallow, small, funnel-shaped depression.—L. 0:08. 3B. 0°03. Of this species I got only 5 specimens (and of these but one full-grown) in deep water from the east end of the island. The shell slightly resembles A. ascaris, Mont., but is smaller, less delicate in the whorls, spire, and tip, and is different in sculpture. 23. ACLIS TRICARINATA, 0. Sp. Shell strong, rather stumpy, semitransparent, and somewhat glossy. Sculpture: excessively strong spiral keels project from 256 REV. BR. B. WATSON ON THE each whorl, from below the embryonic whorl two of these appear on each succeeding whorl, weaker than these a third one lies just above the suture and runs out on the last whorl at the corner of the mouth nearly as strong as the other two: on the base a similar but weaker keel encircles the pillar. These keels are parted by a flat shallow furrow, which is rather over *001 in. in width, and which is scored across by pretty close-set distinct but not strong round-topped longitudinal threads, which again are scored by micro- scopic spiralstrie; these striz, but hardly the longitudinal threads, are visible on the top of the spiral keels. Colour white, glossy, but hardly brilliant. Spare high and narrow, with a blunt, globular, regular, one-sided, hyaline, glossy tip, consisting of one whorl on which some very faint longitudinal strie are doubtfully traceable. Whorls 6, with a contour very much angulated by the spiral keels and the broad sunken suture. Swture is wide, deep, and rather oblique. Mouth obovate, rather large. Outer lip has a sweep which in itself regular is much disturbed, especially on the base, by the spiral keels and furrows, these give the thin lip- edge an unfinished appearance. Inner lip somewhat irregular, from the forward tip to the umbilicus it is patulous and slightly curved; where it strikes the base the curve is obtusely and roundly angulated; across the body the lip runs in an oblique straight line with a projecting strongish edge till past the um- bilicus, when it lies close back on the body and though thinner continues till it joins the outer lip. Unmbilicus is a deep narrow chink.—L. 0°08. B. 0:0385. This species I found sparingly (some 25 specimens) at the Gorgulho shore, and in deeper water at Punta de Sao Lourengo. It is the unnamed “rare” species which McAndrew gives (Geog. Distrib. p. 32) as dredged by him in 12 fms. at Orotava, Tenerife, and specimens of which are in his collection. At first sight the shell is startlingly like the Hydrobia bicarinata, des Moulins, from the South of France, but the differences are very marked. Fam. EULIMID&. Gen. Euztima, Psso. 24. HULIMA FULVA, 0. sp. Shell shining but not brilliant, very small, straight, with a blunt rounded tip, slightly convex whorls, and a slightly produced somewhat truncate base. Sculpture: lines of growth are just recog- > MARINE MOLLUSCA OF MADEIRA. 257 nizable under the microscope, but the surface is not very glossy. Colour a rich deep chestnut, with a pale yellowish narrow band round the base of each whorl] and occupying the outer lip and extreme base of the shell. Spire narrow, with straight outlines, the whorls being barely rounded. Apex small, but very bluntly rounded, and its two sides are scarcely unequal. Whorls 6; of regular and slow increase, barely rounded. Suture slightly oblique and a little impressed. Mouth rather short and somewhat broad. Outer lip has the edge slightly sinused near the body and convex to the base—in direction it runs straight; on the broadish base it is fairly patulous and in its curve regular. Inner lip a longish but very little convex curve across the body to the point of the short pillar; it is a little thickened, but not in the least expanded.—H. 0°067. B. 0-027. In deep water at the east end of the island—not rare. This species is very like H. Jeffreysiana, Brus., but is smaller and narrower, with fewer, less rounded whorls, a much blunter spire, a shorter base, and a much shorter and rounder mouth ; its colour also is a much deeper richer chestnut. Tt is absent in all the other collections I have seen. 25. EULIMA SORDIDA, 0. sp. Shell shining, very small, straight, with a very narrow spire, very blunt tip, barely convex whorls, very blunt base, and oval mouth. Sculpture very doubtful, if any. Colour a sordid yellow. Spire narrow and straight. Apex very small, but blunt and round; its two sides are very equal. Whorls 7, of very regular and slow increase, almost perfectly flat on the sides. Suture scarcely impressed, but visible from the transparency of the shell—-hardly oblique. Mouth very small, oval, but narrow pointed at the top. Outer lip: its edge advances a little from where it leaves the body, but the whole curve of the mouth is very regular. Inner lip is very slightly patulous on the pillar with a minute chink behind it, and it is very feebly angulated where it joins the body.—H. 0-066. _B. 0:024. Punta de Sio Lourengo—rare. Found only by myself. This species is very much less like H. Jeffreysiana, Brus., than is E. fulva. From the latter it differs in being smaller with nearly a whole whorl more, it is narrower, the mouth is smaller and rounder, the base more truncate, the whorls are less rounded, the suture less impressed, the apex is smaller and rounder, and 258 REV. R. B. WATSON ON THE even in bleached specimens the paler and whiter colour is well marked. 26. HULIMA BADIA, 0. sp. Shell small, brilliant, dark brown, conical, with straight contour-lines, a smallish mouth, a rather attenuated base, and a very small tip. Sculpture: there are minute hair-like lines of growth with some very faint superficial and irregular spirals. Colour a deep rich chestnut, darker than in EL. fulva; the extreme point of the base is pale and transparent. Spire high and narrow, conical, with straight contour-lines which do not perfectly correspond on the two sides. Apes exceedingly small and sharp, but round withal. Whorls 9, of exceedingly slow and regular increase and just perceptibly rounded in outline, the first three being markedly more so than the others; the last one is small, but that is from being short rather than narrow. Mouth oval, rather small, and not much pointed above. Outer lip thin, well arched, with an edge retreating well above and advancing not very much at the periphery. Inner lip very faint on the body, thickened and somewhat expanded on the pillar, behind which is a slight umbilicus ; at the base the pillar is very slightly truncated—H.0O1. B. 0:044. Very rare. The few known specimens I got at Punta de Sao Lourengo. The sharp apex and larger size differentiate this species markedly from £. fulva, which it resembles in colour. In general appearance it is like H. Jeffreysiana, Brus., but the apex is distinctly sharper, the contour-lines of spire are straighter, the whorls are shorter, of slower increase, and of more convex out- line; in shells of the same length, #. Jeffreysiana has one whorl fewer. Ihave no doubt some one will say they are the same, but that will none the less be a mistake. Than the young of LL. stenostoma, Jeffr., this is much finer in the apex and is broader in its proportions; like H. gracilis, F. & H., in apex, it is of a stumpier form. Than #. Philippi, Weink., this is slimmer, the spire is not bent, the base is more elongated, the periphery is not carinated. With the young of larger species it is needless to compare so slim and small a torm; a careful comparison fully confirms one’s first impression of diversity. 27. EULIMA RHAPHIUM, 0. sp. (pdaguoy, a small needle.) Shell very small, very narrow throughout its whole length MARINE MOLLUSCA OF MADEIRA. 259 with a very small tip, an attenuated base, and a small narrow oval mouth. Sculpture none, the surface being glassy. Colour hyaline white. Spire exceedingly high and narrow ; the contour- line on the right is quite straight, that on the left slightly curved, but both are perfectly uninterrupted by any swell of the whorls or contraction of the suture. Apex minute and symmetrically rounded. Whorls 11, of very slow and regular increase, the last is both short and narrow but not contracted. Mouth small, narrow, oval, pointed above. Outer lip thin, straight, its edge hardly retreats above or advances at the periphery. Inner lip has a slightly thickened edge and curves very regularly across the body and down the pillar, with hardly the slightest furrow behind it; it has no truncation at the base.—H. 0°11. B. 0:03. One specimen I got from deep water in Funchal Bay. This ig a very much slimmer form with a finer apex and narrower spire than any of the other Madeiran species, none of which in their youngest state approach it in these respects ; even the small form of #. Philippit, Weink., of the same length is much broader, has fewer whorls, a much larger base, rounder mouth, and more projecting outer lip. 2. psila, Wats., from the W. Indies, resembles it most, but is a larger shell with a coarser apex. 28. EULIMA TRUNCA, n. sp. Shell conic-oblong, ivory-white, strong, straight in all its lines, with a broad short base and sharp tip. Seulpture none, the whole surface being brilliantly glossy ; but a feeble flattened varix appears on each whorl and runs interruptedly up the spire. Colour ivory-white, but very slightly pellucid. Spire high and conical, with very straight contour-lines, the whorls being just barely rounded in outline. Apex very small and sharp, but very slightly impressed. Whorls 9, of very slow and regular increase, the upper ones just barely, the lower not at all rounded in outline, the last is short with a truncated well- rounded base. Sutwre hardly oblique, slightly abruptly but distinctly impressed. Mouth fairly oval, but with its curve on the left-hand side a little constricted, shortly pointed above. Outer lip thick, with a rounded edge which has a feeble and wide sinus above and a slight forward curve about the periphery of the shell. Inner lip a good deal thickened and well defined 260 REV. R. B. WATSON ON THE across the body, but still more on the pillar, at the point of which is a very shallow sinus.—H. 02. B. 0-1. Found by Mr. Lowe in dredging near Punta de Sao Lourengo, near which I also got it as well as on the Gorgulho shore to the west of Funchal. This species slightly resembles E. paivana, W., but is larger, with a sharper apex, a more abrupt base, straighter contour-lines, flatter whorls, a shallower but more sharply impressed suture. I have examined H. Stalioi, Brus., with which Jeffreys (P.Z.S8. 1884, p. 368, xxviii. 3, 3a), simply followed by Tryon (Manual, vill. p. 275, lxix. 53), on examination of my specimens, identified this shell. The Madeiran species is certainly not that figured by Jeffreys and copied by Tryon; but Brusina, I believe, questions the correctness of Jeffreys’s identification, and the figure with a strangely bent spire as given in the P.Z.S. (loc. sup. cit.) is obviously not the same as Brusina’s figure in Journ. de Conch. 1877, and described in the same journal for 1869, p- 242. The #. glabella of Searles. Wood is a much (three times) larger shell, and has, like Brusina’s species, a more obtuse apex than H. trunca of Madeira. LH. microstoma, Brus., and LZ. inter- media, Cantr., are both much slimmer forms, and especially in the spire narrower. The varix in the Madeiran species resembles that of H. polita, L., a very much larger species with a much narrower spire. 29. EULIMA INCONSPICUA, N. Sp. Shell small, narrow, but less contracted in the spire than most species, thin, translucent, straight, with a produced base, a shortish narrowly oval mouth, and a small tip. Sculpture none, but the surface is dull. Colour translucent white. Spire high, not bent, but the contour-lines are slightly convex. Apex very small and sharp. Whoris 10, barely convex, of regular but not very slow increase. Suture very slight and scarcely at all oblique. Mouth smallish, oval, narrow, pointed above. Outer lip thin, sharp, rather straight in its direction, well rounded and patulous on the base; near the body it retreats, forming a marked sinus. Inner lip slightly concave, on the body almost imperceptible but a little thickened and reverted on the pillar, where its well-defined edge is marked by a slight chink; at the point of the pillar there is a slight truncation.—L. 0:13. B. 0°04. MARINE MOLLUSCA OF MADEIRA. 261 Very rare. The only specimens of which I know were got at Punta de Sao Lourengo. Compared with EL. intermedia, Cantr., this, besides being smaller, is narrower, has a longer base, a more slowly contracted spire, and a slightly larger apex. EH. microstoma, Brus., has a shorter rounder mouth with a more truncate and slightly broader base. The E. glabra, Jeffr. Lightn. & Pore., P. Z.S8. 1884, p. 367, xxviii. 2), is slightly like this species, but is a much larger and stronger shell, with a broader coarser apex and a larger mouth. Fam. PYRAMIDELLID2. Gen. Opostromtia, Flem. 30. ODOSTOMIA OMPHALOESSA,N. Sp. (Oupaddecoa, umbilicated.) Shell small, conical, translucent, dullish, with a somewhat swoln umbilicated body-whorl, a shortish spire, and a small abruptly truncated tip, across which the sinistral embryonic shell lies on its side. Sculpture none but very faint lines of growth : there is no keel at the periphery even in young shells. Oolour translucent, almost transparent, with a slight ruddy tinge, so that the general appearance isa little horny. Spire short, small, of few whorls; conical, with a small truncated tip crowned by the sinistral embryonic shell which lies on its side across it. Whorls 43 exclusive of the embryonic tip, fairly curved in contour, and not compressed; the last in particular is large and a little tumid, with freely curved outlines: it makes up 2 of the shell’s height. Swtwre is strongly marked, being neither shallow nor narrow : its line is very little oblique. Mouth short and round, the length being to the breadth very nearly as 5 to 4, very little pointed above. Outer lip sweeps with a very free curve all round, is not inflected above, and is barely patulous below. Inner lip is very thin and barely convex on the body, which it quits rather early (breaking scon but not immediately into a strong but short tooth), very slightly reflected, not straightened down the pillar, at the point of which it is barely patulous and isnotangulated. Umbilicus wide and open, but soon contracted. —H.0°07. B. 0:036. Rather common, but absent in other collections than my own. This species is rather like a small O. acuta, Jeffr., but is less conical, has the whorls, especially the last, more tumid and rounded, with no keel round the base nor on the periphery even 262 REV. R. B. WATSON ON THE in the younger shells, has the suture less oblique and deeper, the mouth is rounder, the pillar is not straightened nor the lip flattened and produced at the base, the umbilicus is freer, even a little larger, and the pillar-tooth lies a little higher on the body- whorl. 31. Opostomta (TURBONILLA) UNDATA, D. sp. Shell cylindrical, strong, dull, pure white. Scuwlptwre—Lon- gitudinals strong, rounded, slightly oblique ribs, of which there are 16 or 17 on the penultimate whorl, but they are dispropor- tionately numerous on the last whorl, where they become narrower and more crowded near the lip-edge ; on the base they are feebly present, they are separated by strongish rounded furrows; they are traceable up to the tip, but not on the embryonic whorl; on the later whorls one or two are varicose. Spirals: there are no definite threads, but there is a very faint suggestion of close superficial microscopic puckering extending to the whole surface, but very obscure. Colour a dull glossy deadish white. Spire high, less conical than usual, that is broader in the upper and narrower in the later whorls than most of the group. Apew abruptly and somewhat squarely cut off, the extreme tip being slightly immersed, leaving in profile a minute rounded dome. Whorls 6, very equally rounded and loosely twisted, the last is just 4 the total length of the shell. Swtwre rather deep, but small and very little oblique. Mouth widely oval, bluntly pointed above, slightly patulous on the base, rather more than a fourth of the whole length. Outer lip not sharp, bent in but not sharply at its junction with the body-whorl just below the periphery ; its whole curve is very regular. Inner lip as a mere glaze and very little obliquely it crosses the body, is scarcely angulated at the pillar, down which it runs with a very sharp but barely pro- jecting edge, and only at the extreme point of the pillar (where it becomes faintly patulous and forms a slight angulation in curving to the right) does the lip palpably project. There is no umbilicus and no pillar-tooth H. 0°15. _B. 0°05. Extremely rare. Two specimens are all I found; they came from the Gorgulho shore. At first sight this species is very like O. clathrata, Jeffr., with which Dr. Gwyn Jetireys at first classed it, but afterwards revised his opinion. That species is more transparent, less of a dead white, is more conical, has smaller ribs with spiral striz, MARINE MOLLUSCA OF MADEIRA. 268 a smaller and more narrowed apex, and a more turned-over tip, the mouth is narrower, the pillar-edge more prominent, and there is an umbilical chink. Fam. PLEUROTOMARIID2. Gen. Scuismors, Jeffr. 32. SCHISMOPE DEPRESSA, n. sp. (Figs. 32a, 326, 32 c.) Shell very small, hyaline white with a tinge of pale brown, obliquely depressed, flattened above, with a minute apex and linear suture, a large sub-circular mouth, and a large half- funnel shaped fissure for an umbilicus; its last whorl is cari- nated on its upper surface by the raised edges of the scar of the respiratory orifice. Sculpture—Longitudinals : there are on the earlier whorls close-set radiating riblets which are some- times obsolete, and in all cases degenerate on the base and on the later whorls into mere threadlets widely parted; the whole surface is further scored by sharp radiating curved seratches. Spirals: nearly half of the last whorl is keeled (but not angulated) by the canal-ridge which rises well within the peri- phery, and leads to the narrow oblong respiratory orifice which is formed in the shell as it grows, and is plugged up at its pos- terior end as the shell-edge is pushed forward in growth ; along the interior of the shell the edges of the disused canal project in a minute sharp-edged flange; on the outside from the front of the orifice on to the mouth-edge an irregular depression (visible within the mouth also) sears the shell-surface ; the shell-substance shows no trace of this interruption in the earlier whorls, and the strong oblique downward bend of the last whorl only arises with the expansion of the last whorl and very shortly before the respiratory orifice claims a place. Above the keel the flat surface is scored by about ten fine threads and furrows which become fainter and sparser near the suture; below the carina the whole surface is similarly but more strongly and irregularly scored. Spire very much flattened, apex hardly raised. Whorls barely three, of sudden increase. Suture scarcely impressed. Mouth large, gibbous like a nearly-full moon, excessively oblique. Outer lip greatly descending, flat above, well rounded but scarcely patulous below, where at the pillar-pomt it forks and the outer edge of it sweeps round a little within the edge of the 264 REV. RB. B. WATSON ON THE umbilicus, and rejoins the pillar-lip at the corner of the mouth. Inner lip projects very slightly, is a little patulous, and leaves behind it only a broad umbilical fissure. Across the body it is hardly curved, and joins the outer lip almost at a right angle. —H.0:025. B. 0-032. Mouth: H. 0021; B. 0-017. Common from Funchal eastward and at Porto Santo, deep water. Tt does not seem to have been observed by any one but myself. If the shell be really nacreous, the layer of nacre must be so transparently thin as to give no opalescent reflection. The species is very like S. tabulata, Wats. (‘ Challenger’ Gasterop. p. 117, viii. 7), but is very much smaller in all its dimensions, especially in height of spire; in spite, too, of superficial resem- blances, it differs in sculpture, the riblets being here much fewer and feebler and more curved, the canal-keel and respiratory hole lie much nearer the suture, the spirals are feebler, the last whorl too is larger and not so much contabulated. It should be. noted how deceptively different the adolescent is from the full- erown shell. This is the shell a hasty identification of which from my speci- mens led Dr. Gwyn Jetireys to quote Scisswrella costata, d’Orb., for Madeira. Fam. ERYCINIDA. Gen. Monracura, Turt. 33. MonvactuTa TRIANGULARIS, D. sp. Shell triangularly round but not at all rhomboidal, some- what tumid, strongish but almost transparent, bright but not brilhant. Sculpture fine, close-set, somewhat unequal concentric lines of growth. Colour clear white. Hpidermis, none visible. Margin rotundly oval but for the upward and backward promi- nence of the beaks ; the edges of the valves meet directly with very little expansion at coming together. Beaks small, rounded, sufficiently prominent to give a triangular aspect to the shell; they are somewhat nearer the back end of the shell, towards which but still more upward they slightly turn. Hinge-line broken into a right angle by the beak; the edge is long and very narrow in front, behind it is comparatively very short and broad. Hvnge- plate has a deep triangular cleft from the interior of the shell to the beak ; in the right valve on either side of this cleft there rises a solid little rounded tooth; in the left valve the cleft is MARINE MOLLUSCA OF MADEIRA. 265 bordered by a remote longish narrow lamina which rises into a small rounded prominence. Inside glossy, with a suggestion of faint radiating lines in the inner substance of the shell.—L. 0°18. H. 0:09. Rare; but found by Mr. Lowe and Mr. Johnson as well as myself. I dredged it both at Funchal and at Punta de Sao Lourengo. This species is much more triangular than IL. biden- tata, Mont., and much more oval than I. striata, Mont. Fam. CYPRINID2. Gen. CorattiopHaaa, de Blainv. 34. CORALLIOPHAGA JOHNSONI, N. sp. Shell very irregularly rhomboidally oval, the straightish hinge- margin which runs out in front to a small bluntly rounded corner being vaguely parallel to the rounded lower margin, while the oblique straight but slightly incurved front line is in a way parallel to the easy sweep of the convex, almost semicircular curve of the posterior margin: the valves are tumid above and in front, but are a little compressed behind and below; they are in substance thin, subpellucid and fragile, with fine sharpish but unequal concentric lines of growth. Colour dirty yellowish white, with a chestnut tinge towards the beaks. Hpidermis very thin, worn off except towards the hind margin, where it shows a tendency to slight irregular puckering. Beaks small, rounded, polished, pointing forwards, and bent in on the hinge-line so as to meet one another; they lie a little in front of the middle of the dorsal margin. Margins: they fit quite closely, and, where the lips are not inverted, meet one another in the face abruptly, except below where the shell is slightly flattened out. Ligament ruddy brown, thin but strong, prominent, short, ceasing abruptly at the beaks. Hinge-line straight, narrow. Teeth: there are three small rounded laminz in each valve, forming a little shelf on the inner side of the hinge-margin; they are nearly parallel with the hinge-margin ; the front one, which is a little hummocky, is the shortest ; the second, which is directly under and behind. the beak of the shell, is longer and more compressed; while the posterior one is somewhat vaguely spread out and cut up; in the left valve it is a little more developed than in the right, other- wise in the two valves the teeth are very much alike. Inside not nacreous, but doubtfully opalescent, only vaguely fretted, quite LINN. JOURN.—ZOOLOGY, VOL. XXVI. 20 266 REV. R. B. WATSON ON THE smooth and shining, but not polished except on the muscular scars, of which the anterior, semilunar in form, is pushed up into the extreme front rounded corner of the shell, while the hinder one, also placed very high, is oval ; the two are connected by astrongly marked but narrow ragged-edged pallial line, which below and behind retreats into a broad, very shallow sinus.— H. 0:55. B. 0°6. Mr. J. Yate Johnson, whose name I have attached to this very interesting species, got the few specimens known of it in a mass of oysters and corals dredged up from over 30 fathoms off Funchal. On superficial examination I took it for Modiolarea trapezina, Lam.; but Mr. Edgar A. Smith kindly examined it for me, and he assures me it is a Ooralliophaga, and distinct from any in the British Museum. The young shell scarcely shows the compressed posterior and the expansion below which characterize the adult. Fam. TEREDINIDS. TEREDO, L. 35. TerEpo Dati, n. sp. Shell small, convex, solid, translucent, glossy internally and externally, scored by a very slight ridge and minute furrow from the beak to the point of the shell, and by a fine sharp line which curves across the surface from the beak to the front marginal angle, answering to a fine raised white rib in the interior. Sculpture: the front area is covered with low, rounded, fine (sometimes, but rarely, strongish) threads, the excessively minute but somewhat irregular microscopic serration of which is almost wholly confined to the upper edge, from which, however, it faintly extends to the lower side of the flat furrow above; where these threads abut upon the ends of the answering mid-area ridges their termination is sharply defined by a small but strongly marked furrow, which curves down with posteriorly convex sweep from the beak to the apex of the right angle at the margin where the front area and the central area meet. On the central area the ridges are stronger, are nearly contiguous, and are much more strongly and obliquely serrated, almost tubercled. In number they are usually about 20; but even in specimens of equal growth the number sometimes amounts to 80. Behind this mid-area is a shallow flat with a slight depression, sometimes a MARINE MOLLUSCA OF MADEIRA. 267 furrow, across which the tails of the mid-area ridges spread somewhat rudely in concentric upward-facing curves; the back- edge of this is a very regular curve from the apex to the point, and which drops rather abruptly into a little shallow furrow, beyond which the thin flat concentrically-undulated posterior ear projects (somewhat as in 7. megotara, Hanl.), placed rather low but still quite on the shoulder of the shell; a very shallow obtuse angle lies between this ear and the back margin of the posterior area. In the young shell this ear is generally somewhat ob- solete; the ear-edge is not reverted. Dorsal line not much irregular. Beaks are like smallish round knobs from which, somewhat posteriorly, a minute cup-shaped process projects per- pendicularly, while the apophysis springs from the same knob but a little farther back: it is like that of 7. megotara, Hanl.; it is a curved, very slightly twisted, shortish narrow ribbon projecting rather directly towards the lower point of the valve; its front edge is slightly roughened but not thinned out; the inside is glossy; above the beak it is shortly reverted along the hinge- line, which is only slightly hollowed out and furrowed. In the belly of the shell two very marked ribs run divergingly from behind the beak to the margin: one, narrow but prominent and well defined, corresponds exactly to the curved sharp impressed line on the outside formed at the junction between the fine horizontal threads of the front area and the stronger perpen- dicular ones of the middle area; the other has more, though im- perfectly, the character of a projecting shelf to strengthen the attachment of the ear. At the narrowed and somewhat inturned point of the shell is the usual rounded tooth or knob.—H. 0°15. B. 0:12. This species (whose name I have borrowed from my eminent friend of the Washington Smithsonian Institution) is from the south-eastern coast of Madeira, but the precise locality I failed to note. It varies very greatly in the number of the threads on the entire area of its outer surface. On specimens of much the same size I counted of these from 12 to 37. At first sight it seems very like 7. Stutchburii, de Blainv.; but on closer examination the differences stand strongly out, and the two species may be dis- tinguished at once by the sharp line of demarcation (like that of Xylophaga dorsalis, Turt.) between the threads of the front area and those which are their prolongation at right angles down the middle area, and the distinction is even more visible in the fine 20* 268 REV. BR. B. WATSON ON THE but very distinct white riblet which on the inner face of the shell follows this externalline. A similar feature seems to exist in the T. chlorotica, Gould, a Pacific Ocean species which unfortunately I know only from description, and that (both of Gould* and Tryon ft) is too vague and too divergent from the Madeiran shells to allow of guesses as to its relation to the species from Madeira, of which, moreover, the pallets are wanting. INDEX-LIST OF THE SEA-DWELLING Monuusca or MADEIRA. I have called this an Index-List in justification of the alpha- betical arrangement adopted. Geographical details will be found in the Presidential Address to the Conchological Society for 1890, published in the ‘ Concho- logical Journal,’ vol. vi. no. 11, July 1891; and it will suffice to note here that a special interest attaches to Madeira from its position as a point where the Mollusca of Western Europe, of the Mediterranean, of West Africa, and even of the North Atlantic and Eastern North America, find a centre of con- vergence. The list of collecting stations is given below. To the specific names which follow are added for each the author’s name, and the title and date of the publication in which it appeared, with the addition of at least one reference to some easily accessible illustrated work. Somewhat more is added where difficulty more or less obvious exists. Novelties in nomenclature have been shunned. Except when actually wrong, authors have been left to their own mode of spelling, with a general impression that, as in the instance of the word Gastero- pod, it is a little presumptuous to correct Cuvier when backed by Homer in declining yaorfp. Subgenera have been very sparingly introduced, and the writer has abstained as much as possible from manufacturing species out of those trifling varia- tions to which every living creature is subject. Two lists of the Mollusca of Madeira have been already published: the one of McAndrew, presenting in his Report to the British Association of 1850 the results of his collecting and * Gould, Inverteb. of Massachusetts, 2nd. ed. p. 33, fig. 360. t Tryon, Amer. Mar. Conch. p. 131, figs. 292 to 294. MARINE MOLLUSCA OF MADEIRA. 269 dredging during “‘a few days spent in the Madeira islands ;” the other, published in 1889, that of Prof. Nobre, enumerating the species collected by Mr. Ernesto Schmitz of the Seminario de Funchal. McAndrew individualizes 156 (I follow his reckoning) species, but 29 are unnamed, and of the remaining 127, three (Dentaliwmn dentalis, Marginella guancha, and Neritina viridis) have crept in by mistake, and with almost equal certainty the same may be said of four others (viz. Poromya granulata, Pectunculus siculus, Murex cristatus, and Amphisphyra hyalina). Besides these, four were wrongly identified, viz. Bulla ampulla, Chiton fascicularis, Pecten maximus, and P. opercularis; finally, Risso« purpurea, probably a slip for &. violacea. Thus 12 more have to be deducted, leaving 115. The other list, “ Contribucides para a Fauna Malacologica de Madeira,” was published in the ‘ Instituto,’ no. 8, 1889 (Porto), by Senhor Augusto Nobre. It contains the names, but barely more, of 93 species: one of these, Litorina canariensis, V Orb., is merely the young form of L. striata, which also occurs in the list; another, Trochus conuloides, is a re-duplication of 2. zizy- phinus. Five more, viz. Mytilus edulis, Tellina serrata, Natica flammulata, N. Alderi, N. Josephina, “ dredged at Funchal,” require confirmation. Marginella Philippi calls for further examination. Hight, therefore, of Nobre’s 93 species must for the present be, I think, excluded*. Of the 85 which remain, 59 were already given by McAndrew, so that 26—not an inconsiderable addition —swell McAndrew’s list of 115 to a total of 141 known species in all. To these I now add 35 new species and 206 previously determined species—that is, 241; bringing up the entire number of observed Mollusca from the Madeiran sea to 382. This result has come about through a combination of circum- stances not very usual. During ten years’ residence in Madeira I had opportunity to collect and dredge. In1874 the collections of the Rev. R. T. Lowe, continued from 1827 to 1872, were sent to me by his literary executor, my lamented friend Mr. T. V. Wollaston. In 1896 Mr. J. Yate Johnson (instead of him- self publishing, as I had long hoped he would do) sent to me his * Fuller criticism of this List will be found in the Presidential Address to the Conchological Society, 1890, J. c. supra. Six more of the species given in the List should, I fear, have been excluded: see, at the end here, a strange list of dredging products. 270 REV. BR. B. WATSON ON THE very valuable collection, the accumulation of very many years. The mutual relations of these different collections are supplied in the following list, where, for the sake of shortness, M. repre- sents McAndrew; L., Lowe; N., Nobre; Jn., Johnson; and W., Watson. If my reference to dredging-depths is somewhat indefinite, it is so left on purpose. Unable personally to superintend this dredging, I have only the boatmen’s testimony to the obedience rendered to my order that dredging should go down to 50 fms. One, indeed, of my four boatmen was honest, but truthfulness is not a notable characteristic of the Funchal boatmen ; and it is not in human nature, uncoerced, to do more troublesome work at 50 fms. than the easier at 20 or 830—especially when to those engaged the work appears silly, aud their employer is regarded as a madman. I have gratefully to acknowledge counsel and help from the late Dr. Gwyn Jeffreys, from Mr. Edgar A. Smith of the British Museum, and from the Marquis of Monterosato. To the last especially I am indebted for determining (and that with Medi- terranean specimens sent me for comparison) the very hopeless wave-beaten fragments of Vermetide with which I have had to deal. Names and Depths of Dredging or Collecting Stations. Canical, towards east end of the Island. 10 +015 fms. Cruz, Santa; 7 to 8 miles east of Funchal. 10 to 50 fms. Cruz, Porto da. North coast to deep water. Cruz, Punta da. Shore to deep water. Two miles west of Funchal. Desertas shore. Islands 20 to 30 miles S.E. of Funchal. Funchal. Shore to 50 fms. Gorgulho, Fort. Shore west of Funchal, and some shallow dredging. Labra*; bay east of Canigal. Depth uncertain. Lourengo, Punta Sao, extreme east end. 15 miles from Fun- chal ; to 50 fathoms. * Abra is Arabic for a bay or mooring-place ; the letter Z of the Madeiran name is probably the article. The Moorish “reivers” for long found the island a happy hunting-ground. MARINE MOLLUSCA OF MADEIRA. Dri Machico ; east of Santa Cruz. 10 to 15 fms. Magdalena; 13 miles west of Funchal. Shallow. A jar, how- ever, with some shells within it was brought up entangled in a fisherman’s line from 100 fms. Moniz, Porto; the furthest north, and almost the furthest west, point of the north coast. Shore. Piedade; south coast near the east end. 25 to 35 fms. Ribeiro Secco ; north coast off Fayal. 10 fms. Santo, Porto. Large island 35 miles north-east of Madeira. Shore to 50 fms. Seixal; shore, north-west coast. Selvagens. Uninhabited islands, some 200 miles south of Madeira. Shore. | Inpex-Lisr oF SPECIEs. . Acera bullata, 1776, Miller, Prod. Zool. Dan. p. 242. no. 2921, and Zool. Dan. 11. Ixxi. 1-5: Gwyn Jeffreys, Brit. Conch. Iv. 430, vill. 3; & v. xev.1. Hab. From the Lofotens to the Mediterranean. (W.)S.W. coast from Funchal to Punta de Sao Lourenco, 10 to 50 fms. My specimens are many but small. Acirsa, see Scalaria. . Aclis (Hemiaclis) ascaris, 1819, Turton (as Turbo), Conch. Dict. p. 217: Gwyn Jeffreys, Brit. Conch. rv. 102; & v. 210, Ixxii.2. Hab. From Shetland to the Adriatic. (W.) Canical and Funchal, 10 to 50 fms. Not common. 3. Aclis supranitida, 1842, S. Wood (as Alvania), Ann. & Mag. 1x. 584, pl. v.: Gwyn Jeffreys, Brit. Conch. tv. i. 4; & v. 210, ixxi. 3. Hab. From Norway to the Adriatic and Aigean. (W.) Canigal and Funchal, 10 to 50 fms. Not common.—McAndrew’s collection in the Brit. Mus. contains this species from the Canaries, but it is not named in bis Report. . Aclis tricarinata, 1897, Wats. antea, p. 255. . Aclis trilineata, 1897, Wats. antea, p. 255. . Aclis (Cioniscus) unica, 1803, Montagu (as Turbo), Test. Brit. 11.299: Gwyn Jeffreys, Brit. Conch. rv. 100; & v. 210, Ixxii. 1. Hab. Norway to the Adriatic and to St. Helena. (W.) Canical, Funchal, &c. Not common. 7. Aclis (Cioniscus) vitrea, 1897, Wats. antea, p. 254.—Dr. Gwyn Jeffreys (Lightn. & Poreup. Moll., P. Z.S. May 20, 1884, bo oO Oe 272 REV. R. B. WATSON ON THE p- 344) gives Aclis(Pherusa) Gulsone, Clerk,as Madeiran on my authority. Iam not aware that I ever thought so; if I did Iwas mistaken. A. vitreasent by me was probably the species which led to the mistake on Dr. Gwyn Jeffreys’s part. 8. Acmea virginea, 1776, Miller (as Patella), Zool. Dan. Prod. 1. pl. xii. 4, 5: Gwyn Jeffreys, Brit. Conch. rir. 248 ; & v. 200, lviii. 4. Hab. Doubtfully Arctic or Mediterranean. It extends from Iceland (Zorell) to Mogador (McAndrew). (L., Jn.,W.) From Funchal along the whole south-east coast and also to Porto Santo. Very common. 9. Acteéon pusillus, 1843, Forbes (as Tornatella), B. Assoc. Rept. /Egean Inverteb. p. 191; Watson, Chall. Rep. p. 627; Dall, ‘Blake’ Rep., Harv. Coll. Bull. xvi. 39. Hab. North Atlantic from Cuba to Mediterranean, 40 to 1456 fms. (W.) Magdalena, 100 fms., in an old jar entangled in a fishing-line ; Punta de Sao Lourengo, 50 fms. 3 specimens. 10. Acteon tornatilis, 1767, Linné (as Voluta), Syst. Nat. p. 1187: Gwyn Jeffreys, Brit. Conch. rv. 433, vii.4; & v. 224, xev. 2. Hab. From the Lofotens to the Mediterranean and Moga- dor. (W.) Funchal.—I have three specimens given me as from Funchal; other specimens, too, I have seen said to be from the same locality. I never found it, but (hesitatingly) accept it as Madeiran on the strength of these specimens and the fact that both McAndrew and Lowe got it at Mogador. 11. Adeorbis subcarinatus, 1803, Montagu (as Helix), Test. Brit. 11. 438, vii. 9: Gwyn Jeffreys, Brit. Conch. ry. 231, i. 5; & v. 216, lxxix. 1. Hab. From the English Channel to the Adriatic and the Aigean. (W.) From deep water. One specimen. Alvania, see Rissoa. 12. Amphidesma castanea, 1803, Montagu (as Donasz), Test. Brit. p- 573: Gwyn Jeffreys, Brit. Conch. 11. 418, viii. 1; & v. 188, xliii. 1. From the English Channel to the Canaries. (M., L., N., Jn., W.) Everywhere; very common. 13. Amphisphyra flava, 1897, Watson, antea, p. 234. 14. Anomia ephippium, 1767, Linné, Syst. Nat. p. 1150: Gwyn Jeffreys, Brit. Conch. 11. 80,1. 4; & v. 165, xx.1. Hab. From Iceland to Madeira, and from N. America to the Black Sea. (M., L., N., Jn., W.) Funchal. Not common. 15. Aplysia punctata, 18038, Cuvier, Ann. Mus. 11. 295, i. 2-5: 16. 17. 18. 19. 20. MARINE MOLLUSCA OF MADEIRA. 273 Gwyn Jeffreys, Brit. Conch. v.5,1.1 & xevii.1. Hab. From Norway (see Sars, Moll. Norv. p. 863) to the Mediterranean and Canaries. (M., W.) From Funchal eastwards, 10 to 50 fms. Not uncommon. Aplysia ocellata, 1839, @Orbigny, Moll. Canar. p. 44, v. 1-4. Hab. Canaries. (W.) From west of Funchal to the extreme E. point of the island, from the shore to 50 fms. Many young shells.—Mr. Pilsbry (Manual, 1st ser. xvi. 76) adopts A. dactylomela, Rang, for this species, but gives no expla- nation of his passing-by d’Orbigny’s express statement that they cannot be united—a statement he would not have made hastily. Arca diluvit, 1802, Lamarck, Ann. Mus. vr. 219; An.s. Vert. 1819, vi. (1) 45, 2nd ed. (1835) v1. 476, note of Deshayes ; Philippi, Enumeratio, 1. 59 (as A. antiquata), 1. 43, pl. v. 2. Hab. Mediterranean to Canaries. (W.) Funchal, 50 fms. 3 separate valves.—-On the strength of the habitat I admit this species, but with the gravest doubt. Arca nodulosa, 1776, Miller, Zool. Dan. Prod. p.247: Gwyn Jeffreys, Brit. Conch. 1. 180; & v. 176, c.2. Hab. From Norway to the Mediterranean and Adriatic. (L., Jn.) A few living specimens from 10 to 30 fms., dredged in Labra near the extreme EH. point of the island. Arca pectunculotdes, 1834, Scacchi, Ann. Sicil. vr. 82; Phi- lippi, Enum. 11. 44, xv.38; Forbes & Hanley, Brit. Moll. (as A. raridentata) 11. 241, xlv.8. Hab. From Greenland to Mediterranean. (Jn.) One specimen, with the valves united. Arca plicata, 1795, Chemnitz, Conch.-Cab. x1. 244, ceiv. 2008; Beechey (as A. gradata), Voy. ‘Blossom,’ p. 152, xhi.1; E. A. Smith (as A. domingensis, Lam.), ‘Challenger’ Lamellib. p. 265.—Lischke traces this species (as 4A. domin- gensis, Lam.) under most of its synonyms from Japan to Australia, to Natal, to the 8S. Pacific, to Panama, to the West Indies, the Red Sea, and the E. coast of Africa. In the face of such a record I have not the courage to refuse right of citizenship in Madeira to the two rather undergrown, much weathered, but still quite recognizable _ specimens which I got from deepish water at Punta de Sao Lourengo, the eastern point of Madeira, and which alone represent the species in the island. 274 21. 22. 23. 24. 25. 26. REY. R. B. WATSON ON THE Arca scabra, 1795, Poli, Test. Sicil. 11. 145, xxv. 22; Kobelt, Conch.-Cab. 2nd ed. vir. pt. 2, p. 141, xxxvi. 5,6. Hab. Mediterranean.—Mr. Lowe got, in Labra near Punta de Sao Lourengo, in 1829, one valve of an Arca in crevices of a stone with Gorgonia. It differs from all the other Madeiran species, and I feel constrained to accept his determination of it. Arca tetragona, 1795, Poli, Test. Utr. Sicil. 1. 137, xxv. 12, 13: Gwyn Jeffreys, Brit. Conch. 11. 180, iv. 5*; & v. 176, xxx. 6. Hab. From Finmark to the Mediterranean and the Canaries. (M.,L., N., Jn., W.) Everywhere, very com- mon, but scarcely ever full-grown. Argiope decollata, 1784, Chemnitz (as Anomia), Conch.-Cab. vill. 96, xxvii. 705: Gwyn Jeffreys, Brit. Conch. 11.18; & v. 164, xix.3. Hab. From Guernsey to the Augean and the Canaries. (M., L., Jn., W.) Very common. Argonauta argo, 1758, Linné, Syst Nat. p. 708; Woodward, Manual Moll. p. 66, fig. 32. (W.) Porto Santo. One per- fect specimen. Assiminea litorina, 1825-380, d. Chiaje (as Helix), Mém. An. s. Vert. m1. 215, xlix. 836-38: Gwyn Jeffreys, Brit. Conch. v. 101, xevi.6. Hab. 8.W. England to Mediterranean and Tenerife. (L., Jn., W.) Madeira and Selvagens. Not un- common. Atlanta Peronwi, 1817, Lesueur, Jour. de Phys. txxxv. 390, il. 1; Woodward, Man. Moll. p. 200, pl. xiv. 21-23.—A pelagic species. Everywhere; Madeira and Porto Santo. Common. . Atys Jeffreysi, 1868, Weinkauff (as Cylichna), Conch. Mit- telm. 11. 199; Monterosato, Nomenclatura, p. 145 (as Roxaniella).—No good figure exists of this species. Brocchi, who (¢este Philippi) erroneously identified it with Bulla ovulata, Lam., gives a figure which resembles B. redacta, Desh., much more than it resembles either the B. ovulata, Lam., or the Mediterranean or Madeiran species. Jeffreys’s figure (Ann. & Mag. 1856, vol. xviz. 188, 11. 18, 19), though better, is not characteristic. The B. semistriata, Desh. Coq. foss. Paris, 11. 44, pl.v. 27, 28), offers as good a representation of this species as any | know.— Hab. Mediterranean. (N., Jn., W.) Everywhere ; excessively common. . Auricula equalis, 1832-84, Lowe (as Melampus; 1854, Auri- cula), Zool. Jour. v. 288, xiii. 1-5; Gray, Moll. Anim. 111. 29. 30. dl. 32. 33. 34. MARINE MOLLUSCA OF MADEIRA. 275 eccv. 8, copied from Lowe as Cassidula. Hab. Under stones below high-water mark at the East end and along the whole North coast of the island; also at the Selvagens. Unknown elsewhere. (L., Jn., W.) Very common.—Pfeiffer, Zool. Blatter, x11. (1866), gives this species on the authority of de Paiva as from the South coast as well; but de Paiva’s collectors are not to be trusted. Auricula gracilis, 1882-4, Lowe (as Melampus; 1854, Auri- _cula), Zool. Jour. v. 288; Morelet (as A. vespertina), Hist. Nat. Acores, p. 210, pl. v. 9. Hab. Under stones below high-water mark on North shore in fissures of the rocks ; also at the Selvagens. Always rare. Unknown elsewhere. (L., W.) Auricula Paivana, 1866, Pfeiffer (as Alexia), Mal. Blatt. xu. 146; Wollaston, Test. Atlantic. p. 295. Unfigured. Hab. Under stones below high-water mark in the Selvagem Grande, and there common: unknown elsewhere. (W.) I follow Wollaston in giving this on Pfeiffer’s authority as a distinct species, though, as Wollaston mentions, Dr. Fischer shared my opinion that it is a mere variety of A. bidentata, Mont. Auricula Watsoni, 1878, Wollaston, Test. Atlant. pp. 269, 295. Unfigured. Hab. North coast, under high-water mark. Selvagens: rare. Unknown elsewhere. (W.) Avicula hirundo, 1769, Linné (as Mytilus), Syst. Nat. p. 1159: Gwyn Jeffreys, Brit. Conch. 11. 95, 1.3; & v. 178, xxv. 6. Hab. From Channel Islands to Mediterranean and Canaries. (M., L., Jn., W.) From Funchal along the coast east- wards; also Porto Santo. 10 to 50 fms. Axinus croulinensis, 1847, Jeffreys, Ann. & Mag. xx. 19, ser. 3, 11. 122,v. 2: Brit. Conch. 1. 250; & v. 180, xxxiii. 2. Hab. From the Lofotens to the Mediterranean. (W.) Rather common. Bifrontia zanclea, 1844, Philippi, Enum. 11. 225, xxviii. 11; Sowerby, Manual, 4th ed. p. 84, xvi. 354-6; Woodward, Manual, p. 135; Kobelt, Prodromus, p. 218 (as Omalawis). Hab. Mediterranean: in a few localities. Everywhere ; very common. (M., N., Jn., W.)—Of course every one knows that a good deal may be said against the generic name here adopted, but even more perhaps can be urged in its favour, and especially while so little is known of the animal OD. 36. 37. 38. REV. R. B. WATSON ON THE it may be permissible to retain an old friend, whom if we forsake to whom shall we turn. Omalaxis of Deshayes is like one of Adangon’s facetious names. Homalalaxis of Herrmansen is fatuous. Homalazxis is a hybrid, neither Greek nor Latin. Torinia is a mere guess, and thereafter the mazes of Ilaira, Huomphalus, Evomphalus, Omalazxon, Pseudomalaxvis, &c. may well send us back thankfully to Bifrontia, where if we are ignorant we are at least humble. Monterosato (Conch. d. profundita &c. Palermo, p. 16), who adopts the last of the above-quoted generic names, rejects the identification of this Madeiran species with that of Philippi. His opinion deserves the utmost respect, but I am not able to adopt it in this case. Bittium depauperatum, 1897, Watson, antea, p. 245. Bittium incile, 1897, Watson, antea, p. 246. Bittium reticulatum, 1778, da Costa (as Strombiformis), Brit. Conch. p. 117, viii. 13: Gwyn Jeffreys, Brit. Conch. tv. 258, iv. 4; & v. 217, Ixxx. 4. Hab. From the Lofotens (Lovén) to Mediterranean and Canaries. (M.,L.,N., Jn., W.) Every- where ; very common. Bulla (Haminea) hydatis, 1767, Linné, Syst. Nat. p. 1183: Gwyn Jeffreys, rv. 487, vill. 5; & v. 224, xev. 3. Hab. Gt. Britain to Mediterranean, Canaries, and St. Helena. (W.) Not common and small. . Bulla punctata, 1868, A. Adams, Cuming’s Collection, teste Sowerby in Reeve’s Conch. Icon. xv1., Bulla sp. 15. Not Bulla punctulata (so in Explanation of Plates and Index, pp- 604 & 607, corrected from B. punctata of the text) of Sowerby’s Thesaurus ; not Bullea punctata of John Adams, Trans. Linn. Soe. v. 1, pl. i. figs. 1, 2, where there is a Philine, and which in Sowerby’s Thesaurus, 11. 600, exxv. 161, is given by A. Adams as B. punctata; nor is it the Bulla punctata or punctulata of C. B. Adams, which (teste Carpenter) is the B. Adamsi, Mke.; nor the Atys (Roxania) punctulata of A. Adams. Whether Pilsbry’s Manual of Conch. contains the species is, I think, very doubtful; the figure pl. xxvil. 40, 41, taken from Reeve, not good in the original, is not mended in the copy.—Hab. Madeira and Canaries (f. McAndrew, who gives it as Bulla ampulla). (L., Jn.,W.) Everywhere; pretty frequent, especially young shells. 40. 41. 42. 45. 44. 45. 46. 47. 48. MARINE MOLLUSCA OF MADEIRA. 277 Cadulus Jeffreysi, 1875, Monterosato (as Helonyx), Nuova Revista, p. 20. no. 293; Gwyn Jeffreys, Brit. Conch. v. 196, ci. 3 (as Siphonodentalium subfusiforme, Sars). Hab. Gt. Britain to the Mediterranean and to St. Helena. (Jn., W.) Rare. Cecum atlantidis, 1897, Watson, antea, p. 248. Caecum glabrum, 1803, Montagu (as Dentalium), Test. Brit. p- 497: Gwyn Jeffreys, Brit. Conch. tv. 77; & v. 209, lxx. 5. Hab. From Norway to Mediterranean and Canaries. (W.) Punta de Sio Lourencgo. One specimen. Cecum trachea, 1808, Montagu (as Dentalium), Test. Brit. 11. 497, xiv. 10: Gwyn Jeffreys, B.C. rv. 75, i. 6; & v. 209, Ixx. 4. Hab. From Scotland to Mediterranean, Canaries, and Cuba. (W.) P. deSao Lourengo. One broken specimen. Caecum vitreum, 1858, Carpenter, P. Z. 8S. p. 432. no. 29; Tryon, Man. Conch. vitt. 215, lxvi. 54 (see McAndrew, Canarian List, p. 31, as C. glabrum). Japan (see A. Adams, Ann. & Mag., Nov. 1868). (W.) Everywhere very common, in Funchal Bay especially.—The name alone of this species, but without description, bears an earlier date than that given above. Calyptrea chinensis, 1767, Linné (as Patella), Syst. Nat. p- 1257: Gwyn Jeffreys, Brit. Conch. 111. 273, vi. 6; & V. 201, lx. 1. Hab. From English Channel to Black Sea and Canaries. (M., L., N. Jn., W.) Everywhere; very common. Cancellaria minima, 1856, Reeve, Conch. Icon. sp. 77; Kobelt, Mart. & Chemn. Conch.-Cab. 2nd ed. trv. pt. 4, p. 81, pl. xxi. 3,4. Hab. Straits of Gibraltar, Madeira. (L., Jn., W.) Everywhere; very common. Capulus hungaricus, Linné (as Patella), Syst. Nat. p. 1259: Gwyn Jetfreys, Brit. Conch. 111. 269, vi. 5; & v. 201, lix. 6. Hab. From North Finmark to Mediterranean. (L., Jn., W.) Funchal and Punta de Sao Lourenco, 10 to 40 fms. The pullus shells very common; only one full-grown shell. Cardita calyculata, 1767, Linné (as Chama), Syst. Nat. p- 11388; Born, Mus. Ces. Vindob. pl. v. 10, 11; Poli, Test. Sicil. 11. xxii. 7-9. Hab. Iberian Peninsula, Mediterranean, Canaries. Everywhere—Madeira, Porto Santo; the Selva- gens. Semi-fossil, Canigal beds. Very common. . Cardium aculeatum, 1767, Linné, Syst. Nat. p. 1122: Gwyn Jeffreys, B. C. 11. 268; & v. 180, xxxiv. 1, 1°. Hab. From 278 REV. R. B. WATSON ON THE 50. ol. 52. 58. 54. 56. S. England to Mediterranean. (L., N., Jn., W.) Every- where very common, but small. Cardium exiguum, 1789, Gmelin, Syst. Nat. p. 3255: Gwyn Jeffreys, B. C. 11.278; & v. 181, xxxv. 2. Hab. Extant from the later Tertiaries it extends from North Norway to the Black Sea. (N.) Dredged at Canical it must, on Senr. Nobre’s responsibility, be admitted here. Cardium norvegicum, 1792, Spengler, Scrivt. Nat. Selskab. v. (1) p. 42: Gwyn Jeffreys, B. OC. 11. 294; & v. 182, xxxv. 7. Hab. From North Norway to Canaries. (M., L., N., Jn., W.) Everywhere ; common, but small. Cardium papillosum, 1791, Poli, Test. Sicil. 1. 56, xvi. 2-4; Gwyn Jeffreys, 1. 275; & v. 181, xxxv. 1. Hab. From the Channel Islands to Mediterranean and Canaries. (M.,L., N., Jn., W.) Everywhere ; very common. Cardium transversale, 1854, Deshayes, P. Z. S. p. 333 ; Smith, ‘Challenger’ Report, p. 162, viii. 3. Hab. Alboran Islands, Mediterranean; Acores to Canaries. (Jn., W.) Everywhere; extremely abundant. Cardium tuberculatum, 1767, Linné, Syst. Nat. p. 1122: Gwyn Jeffreys, B.C. 11. 273; & v. 181, xxxiv. 3. ‘Hab. Great Britain to Mediterranean and Canaries. (M., L., N., Jn., W.) Everywhere; very common. . Carinaria Lamarckii, 1810, Péron & Lesueur, Ann. Mus. xv. 69, 11.15; Gray, Fig. Moll. 11. pl. clxi. 1 (as C. mediterranea) ; Woodward, Man. Moll. p. 200, fig. 105, pl. xiv. 19 (as C. eym- bium). Hab. Mediterranean. (L., W.) Floating alive off the Desertas. Fry rare in dredgings. Cassidula, see Auricula. Cassis saburon, 1757, Adanson, Coq. Sénégal, p. 112. no. 8, pl. vii. 8; Bruguiére, Dict. Encycl. 1. 420. no. 4; Lamarck, An. s. Vert. vir. 227, 2nd ed. (Deshayes) x. 36. no. 21. Hab. From the Bay of Biscay to the Mediterranean, Mogador, and Senegal; but it has not been recorded from the Canaries. I never found it nor have seen it from Madeira, and the shells labelled as this species in Lowe’s collection belong to C. sulcosa. Having it therefore as Madeiran on Senr. Nobre’s authority alone, with no information beyond the name, and in absence from his list of the common C. sulcosa, I may, I trust, without discourtesy, hold the species as more than doubtfully Madeiran. Cassis sulcosa, 1780, Born (as Buccinum), Mus. Ces. Vind. p. 241; Lamarck, An. s. Vert. vir. 226, 2nd ed. vol. x. 34. no. 19; Wood, Ind. Test. pl. xx. 25. Hab. Mediterranean 57. 58. 59. 60. 61. 62. 63. MARINE MOLLUSCA OF MADEIRA. 279 to Madeira. (M.,L., Jn., W.) Nearly everywhere; very abundant. Cavolina* gibbosa, 1828?, Rang (as Hyalea) in d’Orb. Voy. Amér. Mérid., Moll., 1836-48, p. 97, pl. v. figs. 21-25; Souleyet, Voy. ‘Bonite,’ Zool. 11. 144, iv. 18-19; and Ptérop. p.38, x. 3,4; Pelseneer, ‘ Challenger’ Report, pt. 65, p- 82. Hab. Pelagic. (L., Jn., W.) Rare. Cavolina inflexa, 1813, Lesueur (as Hyalea), Nouv. Bull. Soc. Phil. rr. 285. no. 69, pl. v. 4. 4, B,c, D; d’Orbigny, Voy. Amér. Mérid., Moll. v. 103, vi. 16-20; Lamarck, An. s. Vert. 2nd ed. vir. 422. no. 16; Souleyet, Ptérop. p. 41, ml. 9-12; Pelseneer, ‘Challenger’ Report, pt. 65, p. 85. Hab. Pelagic. (M.,Jn., W.) Common. Note—McAndrew gives this as Hyalea vaginella. Cavolina quadridentata, 1821, Lesueur (as Hyalea) in Blain- ville, Dict. Se. Nat. vol. xxii. p. 81; d’Orbigny (as H. quadri- spinosa), Voy. Amér. Mérid., Moll. v. 85, vi. 1-5; Pelseneer, ‘Challenger’ Report, pt. 65, p. 78. Hab. Pelagic. (Jn., W.) Not uncommon. Cavolina tridentata, 1775, Forskal (as Anomia), Descrip. Anim. Itin. orient. p. 124, and Icon. pl. xl. fig. 6; Wood- ward, Manual, p. 204, fig. 107, & xiv. 32; Pelseneer, ‘ Chal- lenger’ Report, pt. 65, p. 838. Hab. Lands End to Eastern America ; Mediterranean to Canaries. (M., L., W.) Rare. Cavolina trispinosa, 1821, Lesueur (as Hyalea) in Blainville, Dict. Se. Nat. xxir. 82; Souleyet, Ptérop. p. 45, pl. it. 1-7; Voy. ‘ Bonite,’ 11. 161, vi. 1-10. Of general distri- bution. (Jn., W.) Common. Cerithiopsis atalaya, 1885, Watson, Cerithiopsides from N.E. Atlantic, Journ. Linn. Soc., Zool. xrx. p. 94, pl. iv. 9, 9a. Hab. Madeira. (Jn., W.) From Funchal to Punta de Sao Lourengo. Not uncommon. Cerithiopsis diadema, 1885, Watson, Cerithiopsides from N.E. Atlantic, Jour. Linn. Soe., Zool. x1x. p. 98, pl. iv. 8; Gwyn Jeffreys, Moll. Lightn. & Porcup., P. Z. 8. 1885, p. 60, vi.9, 9a. Hab. Madeira and Mediterranean. (W.) From Funchal eastwards to Punta de Sao Lourengo. Frequent. * Note—So Abildgaard wrote his proposed genus. It would be curious to learn by what authority names, except when wrong, are changed. If Cavolinia and all the host of emendations so dear to the doctrinaire mind be adopted, where shall change be checked? Are we seriously enamoured of a fresh Babel ? 280 64. 65. 66. 67. 68. 69. 70. 71. REV. R. B. WATSON ON THE Cerithiopsis fayalensis, 1885, Watson, Cerithiopsides N.E. Atlantic, Jour. Linn. Soc., Zool. xix. p. 92, iv. 5-5a; ‘Challenger’ Report, Moll. p. 527, pl. xxx. 2. Hab. Mediterranean, Portugal, Agores to Madeira. (W.) Frequent. Cerithiopsis Jeffreysi, 1885, Watson, Cerithiopsides N.E. Atlantic, Journ. Linn. Soc., Zool. xix. 90, iv. 2; Gwyn Jeffreys (as O. pulchella but not of C. B. Adams), Ann. & Mag. 3rd ser. 11. 129, v. 8; Brit. Conch. rv. 269, & v. 217. Hab. English Channel to Mediterranean. (W.) Rare. Monterosato (in litt.) suggests that Conti’s name of “ con- catenata”’ should have precedence : see Conti, Foss. di Monte Mario; but there are other questions, besides, regarding the species which had need of settlement before further disturbance of the name. Cerithiopsis Metawa, 1829, d. Chiaje (as Murea), Mém. An. s. Vert. 111. 222, pl. xlix. 29-31; Tiberi (as Cerithiwm Crosse- anum), Jour. de Conch. 1863, p. 160, vi. 2: Gwyn Jeffreys, B. C. 1v. 271; & v. 217, lxxxi. 4. Hab. Shetland to Medi- terranean and Canaries. (M., L., N., Jn., W.) Everywhere ; common. Cerithiopsis minima, 1864, Brusina (as Cerithium), Conch. Dal. p. 17; Bucquoy, Dautzenberg, & Dollfuss, Moll. Rouss. p. 207, xxvii. 5-9. Hab. Mediterranean. (W.) Very rare. Cerithiopsis tiara, 1885, Watson, Cerithiopsides N.E. Atlantic, Jour. Linn. Soec., Zool. xix. p. 92, iv. 6, 6a. Hab. Madeira and Mediterranean. (W.) From Funchal eastwards to Punta de Sao Lourenco. Not uncommon. Cerithiopsis tubercularis, 1803, Montagu (as Murex), Test. Brit. 1. 270: Gwyn Jeffreys, B.C. 1v. 266, iv. 5; & v. 217, Ixxxi. 1. Hab. Norway to Mediterranean. (W.) From Funchal eastwards to the Point of the Island and to Porto Santo. Not uncommon. Cerithium rupestre, 1826, Risso, Hist. Nat. Europ. rv. 154; Philippi, Enum. 1. 194, xi. 7. Hab. Mediterranean. (L., Jn.) Selvagens, fide Bardo de Paiva. Chama gryphoides, 1767, Linné, Syst. Nat. p. 1189; Wood- ward, Man. p. 276, xviii. 8, 9 (as C. macrophylla, Chemn.). Hab. Bay of Biscay to Mediterranean and Canaries. (L., N.?, Jn., W.) Everywhere; very common.—It is curious se MARINE MOLLUSCA OF MADEIRA. 281 McAndrew, whose collection has 10 specimens from the Mediterranean and Canaries, did not get it. Chascax, see Fasciolaria armata. 72. Chiton cancellatus, 1839, G. B. Sowerby, Conch. Illustr. no. 5, figs. 104-5: Gwyn Jeffreys, B. C. 111. 217; & v. 198, Ivi. 1. From Sweden and Norway to Mediterranean. (L., Jn., W.) From west of Funchal to East point and Porto Santo. Common. _ 73. Chiton (Acanthochiton) discrepans, 1827, Brown, Ill. Conch. p. 65, xxi. 20: Gwyn Jeffreys, B. C. 111. 214; & v. 198, lv. 4. Channel Islands to Mediterranean and Mogador. (M., as C. fascicularis; L., Jn., W.) Everywhere ; very common. 74. Ohiton (Callochiton) levis, 1767, Pennant, Brit. Zool. tv. 72, xxxvi. 3; Philippi, Enum. 1. 107, vii. 4, 1. 883: Gwyn Jeffreys, B.C. mr. 226; & v. 199, lvi.6. Hab. From Northern Norway to Mediterranean and Canaries. (W.) From Fun- chal eastwards and Porto Santo. Specimens not rare, but all small. 75. Ohiton (Ischnochiton) marginatus, 1767, Pennant, Brit. Zool. tv. 71, xxxvi. 2: Gwyn Jeffreys, B. C. 111. 221; & v. 199, lvi.5. Hab. From Lofotens to Sicily and Mogador. (M.,as OC. cinereus; L., W.) Canical, near Punta de Sao Lourengo. Very rare. Chiton (Ischnochiton) sp.—Dr. Gwyn Jeffreys considered this to be C. Rissot, Payr.; but out of 30 or 40 very small and separate valves which I collected, about half are those of the head alone, not one is of the tail, and on none is the girdle preserved; so that the species cannot be deter- mined with any confidence. Cioniscus, see Aclis. 76. Circe minima, 1803, Montagu (as Venus), Test. Brit. p. 121, 1.3: Gwyn Jeffreys, 1. 822, vi.4; & v. 183, xxxvil.6. Hab. From Finmark to Mediterranean and Canaries. (M., L., Jn., N., W.) Everywhere; very common. 77. Clio (Creseis) acicula, 1828, Rang (as Creseis acicula and CO. clava), Ann. Se. Nat. x111. 317, xvii. 5,6; Souleyet, Hist. Ptérop. p- 56, vi. 5,7; Woodward, Manual, p. 205, xiv. 34. Hab. Pelagic. (W.) Everywhere, but not common. 78. Clio pyramidata, 1767, Linné, Syst. Nat. p. 1094; Souleyet, Hist. Ptérop. p. 50, v. 7-10 ; Woodward (as Cleodora), Man. p. 205, xiv. 38. Hab. Pelagic. (Jn., W.) Everywhere. 79. Olio striata, 1828, Rang (as Creseis), Ann. Sc. Nat. ximt. LINN. JOURN.—ZOOLOGY, VOL. XXVI. 21 282 REV. R. Be WATSON ON THE 315, xvii. 8; Souleyet, Hist. Ptérop. p. 55, vi. 8. Hab. Pelagic. (W.) Funchal Bay; very rare. 80. Clio (Styliola) subula, 1827, Quoy & Gaimard (as Cleodora), Ann. Se. Nat. x. 238, vil. 1-3; Souleyet, Hist. Ptérop. p- 55, vi. 2,6. Hab. Pelagic. (Jn., W.) Everywhere; very common. 81. Columbella (Mitrella) cribraria, Adanson, ‘ Le Barnet,’ p. 146, pl. x. 1; 1822. Lamarck (as Buceinum), An. s. Vert. vit. 274. no. 43; Quoy & Gaimard, ‘Astrolabe’ Moll. 1. 421, pl. xxx. . 21, 22; Kiener (Buccinum), p. 22. no. 23, xvi. 58. Hab. Guinea Coast, St. Helena, Ascension, Canaries, West Indies, &e. (M.,L., Jn., W.) Funchal to East endof Island. Very common. 82. OColumbella Kraussit, 1844, Sowerby, P. Z. 8. p. 53; Thes. 83. Conch. 1. xl. 180; Krauss, Stid-Afrik. Moll. (as C. cerealis, Mke., MS.); Sowerby (George), Mar. Shells 8. Africa (as C. cerealis, Krauss), p. 21. Hab. Port Elizabeth, the Cape, and probably the Mediterranean. (L., Jn., W.) Every- where; common. Lalso found it at the Canaries.—It is very like C. catenata, Sow., in colour; but that species is larger, stouter, and more coarsely ribbed. It is even liker C. Bro- deripii, Sow.; but the Madeiran species is so variable as to suggest much caution. Itis the C. albuginosa, Reeve, Conch. ‘Icon. xt. no. 223; and probably C. pediculus, Monterosato, and some other Mediterranean forms which he indi- vidualizes. Oolumbella minor, 1836, Scacchi, Cat. Reg. Neap. p. 10, f.11; Philippi, Enum. Moll. Sic. 1.190, xxvu. 12. (N., W.) Hab. Mediterranean.—This as a species is painfully sug- gestive of some depauperated forms of C. scripta, L., from which, however, since some are willingly responsible for separating it, it may be observed to differ in a slightly finer apex and more delicate spiral threads round the snout ; but, as distinctive features, neither colour nor size count for much. 84. Columbella rustica, 1767, Linné (as Voluta), Syst. Nat. p- 1190; Adanson, ‘Le Siger,’ p. 185, ix. 28; Philippi, Enumeratio, 1. 228, xii. 11. Hab. From Bay of Biscay to Mogador and Mediterranean. (M., L.,N., Jn., W.) Every- where ; very common. 85. Columbella scripta, 1767, Linné (as Murex), Syst. Nat. p. 1225; MARINE MOLLUSCA OF MADEIRA. 283 Kiener, xiii. 43 (as Buccinum Gervillei), and xvi. 56 (as B. corniculum). Hab. Mediterranean. (L., Jn., W.) Magdalena; Funchal; Punta de Sao Lourengo and Porto Santo. 40 to 50 fms. Not common. 86. Conus mediterraneus, 1790, Hwass* in Encycl. Méthod. 1. 701, pl. ceexxx. 4; Wood, Index Test. p. 83, pl. xv. 101 ; Reeve, Conch. Icon. xvi. 89.—I give this species here on the authority of Senhor Nobre. Specimens were occasionally brought to me by Funchal boatmen, but I never considered them Madeiran. A very worn specimen in Mr. Johnson’s collection seems rather to confirm my doubts. Still its presence on the coast of Portugal, throughout the Medi- terranean, and at the Canaries makes its existence at Madeira anything but unlikely. 87. Coralliophaga Johnsoni, 1897, Watson, antea, p. 265. Coralliophila, see Murex. 88. Orania anomala, 1776, Miller, Prod. Zool. Dan. p. 237 ; Zool. Dan. 1. 4, pl. v. 1-8 ; Woodward, Man. pp. 235-6, ff. 157-8 : Gwyn Jeffreys, B. C. 11. 24, 1.3; & v.165, xix.6; Davidson, Rec. Brach., Trans. Linn. Soc., Zool. tv. 183, xxvii. 1-90. Hab. Greenland and Spitzbergen to Mediterranean. (L., Jn., W.) Funchal, 50 fms. A few valves. 89. Oraspedotus Tinet, 1830, Calcara (as Monodonta), Ricerche Mal. p. 14, t. 8; Philippi, Enum. wu. 157, xxv. 19 (as Mo- nodonta limbata, but not of Quoy & Gaim., and not a Monodonta, being umbilicated and non-nacreous). Hab. Mediterranean. (Jn., W.) Funchal, 50 fms. 2 specimens. 90. Crenella rhombea, 1827, Berkeley (as Modiola), Zool. Jour. 111. 229, xvili. 1: Gwyn Jeffreys, B. C. 11.131; & v.172, xxviii. 5. Hab. English Channel to Mediterranean and Canaries. (W.) Funchal, 50 fms. Punta de Sao Lourengo, 50 fms, Not common. Orossea, see Rissoa. 91. Ouspidaria costellata, 1836, Deshayes (as Corbula), Morée, p- 86, xxiv. 1-3: Gwyn Jeffreys, B.C, 111.49; & v. 191, xlix. 3 (as Neera). Hab. Norway to Mediterranean and Canaries. (M., L., Jn., W.) Everywhere ; very common. 92. Ouspidaria cuspidata, 1792, Olivi (as Tellina), Zool. Adr. p. 101, iv. 3: Gwyn Jeffreys, B. C. 111. 58, u. 4; & v. 191, * Hwass, teste Bruguiére, Encycl. Méthod. i. p. 598, wrote the whole articie on Conus. 21* 284 93. 94. 96. ie 98. 99. 100. REV. R. B. WATSON ON THE xlix.5. Hab. From Spitzbergen and Greenland to Medi- terranean and Canaries. (M., L., Jn., W.) Everywhere ; common. Cuvieria columella, 1827, Rang, Ann. Se. Nat. x11. 328, xlv. 1-8 ; Souleyet, Hist. Ptérop. p. 59, pl.iv. & xiv. 1-6; Wood- ward, Man. p. 205, pl. xiv. 35. Hab. Pelagic. (Jn., W.) Not common. Cyclostrema serpuloides, 1808, Montagu (as Helix), Test. Brit. Suppl. p. 147, xxi.3: Gwyn Jeffreys, B. C. 11. 290, vil. 3; & v. 201, lxi. 8. Hab. From Great Britain to Medi- terranean. (W.) From east of Funchal to Punta de Sao Lourengo, and Porto Santo. Common. . Cylichna cylindracea, 1767, Pennant (as Bulla), Brit. Zool. tv. 117, Ixxvii. 85: Gwyn Jeffreys, B. C. 1v. 415, viii. 1; & vy. xcii.4. Hab. From Finmark to Mediterranean and Canaries and St. Helena. (M.,Jn., W.) Everywhere; very common. Cylichna spreta, 1897, Watson, anfea, p. 234. Cylichna umbilicata, 1803, Montagu (as Bulla), Test. Brit. I. 222, vil. 4: Gwyn Jeffreys, B. C. 1v. 413; & v. 228, xciii. 3. Hab. From the Lofotens to the Mediterranean. (W.) Santa Cruz, Machico, Canical; down to 50 fms. Rare.—These specimens from Madeira were identified as above by Dr. Gwyn Jeffreys: after much hesitation I at length, despairing to mend the instruction, have accepted his determination, which the Marquis of Monterosato still rejects. Cylindrobulla fragilis, 1856, Gwyn Jeffreys (as Cylichna), Mar. Test. Pied., Ann. & Mag. Feb. 1856, p. 188, ii. 16, 17, & ‘Washington’ Ital. Exped. Notes, Ann. & Mag. July 1882, p. 84 (as Cylindrobulla). Hab. Mediterranean, 10 to 1536 fms. (W.) Porto Santo, 50 fms. One specimen. Cyprea (Trivia) candidula, 1835, Gaskoin, P. Z. 8. p. 200 ; Kiener, Iconog. p. 150. no. 136, li. 1; Reeve, Conch. Icon., Cyprea, sp. 151, 154; Sowerby, Thesaurus, rv. p. 50, pl. xxxvi. ff. 508-9. Hab. From Portugal to W. Mediterranean ; Canaries and Bight of Benin. (M., L., Jn., W.) Everywhere; very common. Cyprea (Trivia) europea, 1803, Montagu, Test. Brit. Suppl. p- 88: Gwyn Jeffreys, B. C. 1v. 408, vil. 4; & v. 222, xci. 2. Hab. From Norway to Mediterranean. (W.) Puntade Sao Lourenco. 2 specimens. 101. 102. 103. 104. 105. 106. 107. MARINE MOLLUSCA OF MADEIRA. 285 Cyprea lurida, 1767, Linné, Syst. Nat. p. 1175; Kiener, Iconog. p. 82. no. 71, pl. xxiii. 1; Sowerby, Thes. rv. 6, sp. 7, x. 64-5; Reeve, Conch. Icon., Cyprea, sp. 32, pl. ix. 32. Hab. Mediterranean to the Cape Verd Is., Guinea Coast, and St. Helena. (W.) Porto Santo. 2 specimens. Cyprea (Trivia) pulex, 1827, Solander, Zool. Jour. 111. 368 ; Kiener, Iconog. p. 142, liii.1; Reeve, Conch. Icon. sp. 144, pl. xxv.; Sowerby, Thes. tv. 44, pl. xxxvi. 492-4. Hab. From Portugal to Mediterranean. (M., L., N., Jn., W.) Everywhere ; common. Cyprea pyrum, 1790, Gmelin, Syst. Nat. p. 3411; Reeve, Conch. Icon. sp. 26, pl. vili.; Sowerby, Thes. iv. p. 25, pl. xxiv. 202-5. Hab. From Portugal to the Canaries. (1.,Jn., W.) Porto Santo. Not very common. Cyprea spurca, 1767, Linné, Syst. Nat. p. 1179; Kiener, Teonog. p. 61, xxx. 1, la; Reeve, Conch. Icon., Cyprea, sp. 68, pl. xiv.; Sowerby, Thes. rv. p. 88. no. 131, pl. xviii. 118-122 & xxxvi.516. Hab. From West Indies to Portugal, Mediterranean, Canaries, Mogador, and St. Helena. (L., N., Jn., W.) Punta de Sao Lourengo and Porto Santo. Plentiful. Cytherea, see Venus. Defrancia, see Pleurotoma. Diplodonta rotundata, 1803, Montagu (as Zellina), Test. Brit. p. 71, u.3: Gwyn Jeffreys, B. C. 11. 254, v.7; & Vv. 180, xxxiii.4, Hab. From English Channel to Mediterranean and Canaries. (M., L., N., Jn., W.) Everywhere; very common. Diplodonta trigonula, 1831; Bronn, Ital. Tertiar-Geb. p. 96, iii. 2; Philippi, Enumeratio, 1. 31, iv. 6, & 11. 24 (where be gives both D. trigonula, Bronn, and also his own D. apicalis as if different); Wood, Crag Moll. 11. 146, xii. 2 a, b; Hornes, Foss. Moll. Wiener, p. 218, xxxii. 4 a, b. Hab. Mediterranean to Canaries. (M., L., Jn., W.) Everywhere ; very common. Dolium galea, 1767, Linné (as Buccinum), Syst. Nat. p- 1197; Woodward, Man. p. 115, vi. 12. Hab. From Portugal to Mediterranean and Canaries. (L., W.) Fun- chal, Punta de Sao Lourengo, Porto Santo ; 10 to 50 fms. Not uncommon, but the specimens mostly young or in fragments. 286 108. 109. 110. IESE 112. 113. 114. 115. 116. Wy. 118. REV. R. B. WATSON ON THE Donax trunculus, 1767, Linné, Syst. Nat. p. 1127: Gwyn Jeffreys, B.C. 11. 407; & v.188, xl. 7. Hab. From Channel Islands to Mediterranean, aud (me ipso teste) both Canaries and Mogador. (Jn., W.) Porto Santo, &c. Not unfrequent as single valves. Donax venustus, 1791, Poli, Test. Sicil. 11. xix. 28, 24; Romer in Mart. & Chemn. Conch.-Cab. 2nd ed. x. pt. 3, p- 31, vi. 10-20. Hab. Mediterranean. (Jn., W.) From Funchal along South coast, also Porto Santo. Common. Doridium laurentianum, 1897, Watson, antea, p. 237. Doridium maderense, 1897, Watson, antea, p. 238. Emarginula fissura, 1767, Linné (as Patella), Syst. Nat. p. 1261: Gwyn Jeffreys, B. C. 111. 259, vi. 3; & v. 200, lix. 2. Hab. From Finmark to Mediterranean and Canaries. (M., L.,Jn., W.) Everywhere; very common, in every form of transition to Semperia Paivana, Crosse, Jour. de Conch. 1867, p. 76, i. 2. Emarginula Huzardi, 1826, Payraudeau, Moll. Cors. p. 92, pl. v. 1, 2; Lamarck ed. Desh. vir. 587. no. 11; Reeve, Conch. Icon. sp. 4; Gwyn Jeffreys, Lightn. & Poreup. Moll., P. Z.S. 1882, p. 679. no. 5. Hab. Mediterranean. (L., W.) Funchal, Punta de Sao Lourengo, Porto Santo. Not common. Emarginula tenera, 1878, Monterosato, Enum. & Synon. p- 19 (the name alone) ; id. Nomenclatura, p. 36; Locard, Malac. Frane. p. 337; Watson MS., #. alba.—This species stands much in need of description and figure. It is small, translucent, white, high, well-rounded, with about 30 strong ribs and occasional weaker intermediate ones, with small sharp cross-bars and deep lattice-like depressions, having a long, narrow, square-cut, sharply bordered slit; the top is ineurved and a little depressed, and directly overhangs the extreme front edge. L.O11. B.0065. H. 005. Hab. Bay of Biscay, Mediterranean. (W.) Almost everywhere ; not rare. Eulima badia, 1897, Watson, antea, p. 258. Eulima fulva, 1897, Watson, antea, p. 256. Eulima inconspicua, 1897, Watson, antea, p. 260. Eulina intermedia, 1842, Cantraine, Malac. Méd. Suppl. (p. 40 ?), Bull. Bruxelles, p. 14(?): Gwyn Jeffreys, B. C. tv. 208; & v.214,]lxxvii.5. (Note.—Sars’s shell, Moll. Norv. 119. 120. 121. 2. 123. 124. 125. 126. 127. MARINE MOLLUSCA OF MADEIRA. 287 p- 210, pl. xi. 20, is not Cantraine’s species, though very near.) Hab. Atlantic, Hast and West, and Mediterranean. (M. [as #. nitida], W.) Shore at Gorgulho rock E. of Funchal, Punta de Sio Lourenco. Not common. Hulima Jefveysiana, 1869, Brusina, Jour. de Conch. 1869, xvi. 245 (as Letostraca); Gwyn Jeftreys, Lightn. & Pore. dredg., P. Z. S. 1884, p. 366, xxviii. 1. Hab. Medi- terranean.— Gwyn Jeffreys says that McAndrew got it at the Canaries, Rochebrune from the Cape Verd Is.; and that Verrill identified it from New England. (Jn., W.) From Funchal to the island’s east extremity. Not rare. Hulima microstoma, 1869, Brusina, Jour. de Conch. xvii. 244; Monterosato, Enum. p. 385; Locard, Malac. Frane. p- 208; Kobelt, Prodromus, p.115. Hab. Mediterranean. (W.) Punta de Sao Lourencgo. 2 specimens. Eulima paivensis, 1873, Watson, P. Z. 8. p. 364, pl. xxxvi. 29; Tryon, Manual, viit. 277, lix.58. Hab. Madeira and the Selvagens. Not common; but both Mr. Lowe and I got a good many specimens. Eulima Philippii, 1868, Weinkauff, Conch. Mittheil. 11. 228; Philippi, Enumeratio, 1. 158, & 11. 130, ix. 10 (as #. dis- torta, but not of Defrance): Gwyn Jeffreys, B. C. Iv. 205 ; & y. 214, Ixxvu. 5. Hab. From the Lofotens to the Medi- terranean and Canaries. (M., L., W.) Extremely common every where.—Locard, Prodrome, p. 205, gives H. incurva, Renier’s name, priority, but that author’s name-list confers no priority, and the changes rung on the name of this species create mere confusion in the midst of which no private rights of nomenclature avail. Names are for public profit, not for individual glorification. Hulima rhaphium, 1897, Watson, antea, p. 258. Eulima sordida, 1897, Watson, antea, p. 257. Eulima subulata, 1802, Donovan, Brit. Shells, v. pl. elxxi.: Gwyn Jeffreys, B.C. 1v. 208; & v. 215, lxxvu.7. Had. From Britain to the Mediterranean. (M.,Jn., W.) Everywhere ; common. Eulima trunca, 1897, Watson, antea, p. 259. Eulimella, see Odostomia. Fasciolaria armata, 1854, A. Adams (as Latirus), P.Z.S. p- 314. no. 11; Watson, ‘ Challenger’ Moll. p. 243, xiii. 1. 288 REV. R. B. WATSON ON THE Hab. Madeira; Tenerife; N.W. Africa. (L., W.) Punta de S. Lourengo and Porto Santo. Not common. 128. Hissurella greca, 1767, Linné (as Patella), Syst. Nat. p. 1261: Gwyn Jeffreys, B. C. 111. 266, vi. 4; & v. 200, lix. 5. Hab. From the Frroes to Mediterranean and Canaries. (M., L., N., Jnu., W.) Everywhere ; common. 129. Fossarus ambiguus, 1767, Linné (as Helix), Syst. Nat. p- 1251; Adanson, Sénégal, p. 173, xiii. 1; Philippi, Enu- meratio, 11. 147, xxv. 1. Hab. From Bay of Biscay to Mediterranean, Canaries, and St. Helena. (L., Jn., W.) Everywhere ; common. 130. Gastrochena dubia, 1777, Pennant (as Mya), Brit. Moll. Iv. 82, xliv. 19: Gwyn Jeffreys, B. C. 111. 91, 11.5; & v. 193, li. 6. Hab. From the Brit. Channel to Mediterranean, Canaries, Mogador, and St. Helena. (M.,L., Jn., W.) Punta de So Lourengo to 50 fms., Magdalena shore. Very rare. 131. Haliotis tuberculata, 1767, Linné, Syst. Nat. p. 1256: Gwyn Jeffreys, B. C. 11. 279, vii. 1; & v. lx. 2. Hab. From the Channel to Mediterranean, Canaries, and Mogador. (M., L., N., W.) All along the coast ; common. Hemiaclis, see Aclis. 132. Homalogyra atomus, 1841, Philippi (as Zruneatella) in Wiegm. Archiv, vit. pt. 1, p. 54, v. 4; Enum. nm. 134, xxiv. 5: Gwyn Jeffreys, B. C. tv. 69, 1.5; & v. lxx. 2. Hab. From North Norway to Mediterranean. (Jn., W.) Everywhere ; common. 133. Homalogyra rota, 1853, Forbes & Hanley (as Skenea), Brit. Moll. 111. 160, Ixxiii. 10 & Ixxviii. 1, 2: Gwyn Jeffreys, B. C. rv. 71; & v. 209, Ixx. 8. Hab. From North Scotland to Mediterranean. (W.) Everywhere; common. Hydrobia confusa, v. Frauenfeld, erroneously identified as Bythinia similis, Drap., lives entirely in the freshwater runnels, quite remote: from the sea. 134. Lanthina communis, 1822, Lamarck, An. s. Vert. vt. (2) 206: Gwyn Jeffreys, B. C. (as LI. rotundata, Leach) tv. 186, 11. 1 & frontispiece ; & v. 214, Ixxvii. 1. Hab. Oceanic.. (M., L., Jn., W.) From Funchal to East point aud Porto Santo. Common. 135. Janthina exiqua, 1822, Lamarck, An. s. Vert. vi. (2) 206 ; Forbes & Hanley, Brit. Moll. 11. 555, lxix. 8, 9; Sowerby, 136. 137. 138. 139. 140. 141. 142. MARINE MOLLUSCA OF MADEIRA. 289. Ill. Index, xii. 4. Hab. Ocean. (M., W.) From Funchal to East point. Rare. Lanthina pallida, 1841 (?), Harvey in Thompson’s Brit. Mar. Conch. p. 152, & Ann. & Mag. v. 96, 11.2; Philippi, Enumeratio, 11. 224, xxviii. 14; Forbes & Hanley, Brit. Moll. 11. 553, xix. 10,11. Hab. Ocean.—This species is given by McAndrew, and by him alone, as from Madeira, and the specimens in his collection at the British Museum are correctly identified; but his collection at Cambridge, which is more to be trusted, has no specimens, unless, as the Rev. A. H. Cook suggests, 7 fine specimens of “ J. prolon- gata, Madeira,” in the McAndrew collection be so regarded. Jeffreysia glabra, 1844, Alder (as Rissoa ?), Ann. & Mag. XI1I. 325, vii. 1-4: Gwyn Jeffreys, B. C. (as Jeffreysia diaphana) tv. 59, 1.3; & v. lxix.5. Hab. Shetland to the Mediterranean. (W.) Funchal to East point and Porto Santo. Common. Jeffreysia globularis, 1853, Gwyn Jeffreys in Forb. & Hanl. Brit. Moll. rv. 268, exxxui.5: Gwyn Jeffreys, B. C. Iv. 62; & v.lxix.7. Hab. Scotland. (W.) Funchal to Hast point of Island. Rare. Jeffreysia opalina, 1849, Gwyn Jeffreys, Ann. & Mag. 2nd ser tr. ool; Horb. & Hani. Bs Moll, iv. 267, exxxui- 10 & mm. 2: Gwyn Jeffreys, B. C. rv. 60; & V. 209, lxix. 6. Hab. From Shetland to Mediterranean. (W.) From Funchal to Hast point and Porto Santo. Lachesis minima, 1803, Montagu (as Buecinum), Test. Brit. p. 247, viii. 2: Gwyn Jeffreys, B. C. 1v. 318, vi. 1; & v. 218, Ixxxiv. 3. Hab. From Channel Islands to Medi- terranean and Canaries. (M.,L.,Jn.,W.) Everywhere; very common. Lamellaria perspicua, 1767, Linné (as Helix), Syst. Nat. p: 1250: Gwyn Jeffreys, B. C. 1v. 235, 11.6; & v. 216, Ixxix. 2. Hab. From Labrador to United States, and from Nor- way to Mediterranean and Canaries. (M., L., Jn., W.) Everywhere, but not common. Lasea rubra, 1803, Montagu (as Cardium), Test. Brit. p- 83, xxvu. 4: Gwyn Jeffreys, B. C. 11. 219, v.2; & v.179, xxx. 1. Hab. From Iceland to Mediterranean and Canaries. (M., L., Jn., W.) Everywhere; common. Leda pygmea, Minster.—I noticed this species in Mr. Yate Johnson’s collection, but its presence there seemed to me accidental. 290 143 144. 145. 146. 147. 148. 150. 151. REY. R. B. WATSON OF THE Lima hians, Gmelin (as Ostrea), Syst. Nat. p. 3832: Gwyn Jeffreys, B. C. 11. 87, ii. 2°; & v.170, xxv. 5. Hab. From Northern Norway to the Mediterranean and Canaries. (M., N., Jn., W.) Everywhere ; common. Lima squamosa, 1818, Lamarck, An. s. Vert. vi. (1) 156; Sowerby, Thes. 1. 84, xxi.1 & 18(?). Hab. Japan (f. Dun- ker); N. Zealand (f.v. Martens); Mediterranean. (M., L., N., Jn., W.) Everywhere ; common. Lima subauriculata, 1808, Montagu (as Pecten), Test. Brit. Suppl. p. 63, xxiv. 2: Gwyn Jeffreys, B. C. 11. 82; & v. 169, xxv. 3. Hab. From N. America and Greenland to Mediterranean and Canaries. (L., Jn., W.) Everywhere ; very common. Limacina bulimoides, 1836, d’Orbigny (as Atlanta), Voy. Amér. Mérid. v. 179, xii. 86, 38; Souleyet, Rev. Zool. p- 188; Voy. ‘ Bonite,’ 11. 224, xii. 35-42; Ptéropodes (as Spirialis), p. 65, xv. 3, 4; Woodward, Manual, p. 207, pl. xiv. 42; Pelseneer, ‘Challenger’ Report, p. 30. Hab. Pelagic. (W.) Everywhere; abundant. Limacina inflata, 1836, dOrbigny (as Atlanta), Voy. Am. Meérid. v. 174, xii. 16-19 ; Souleyet (as Spirialis rostralis), Ptérop. p. 62, xiv. 7-12; Pelseneer, ‘ Challenger’ Report, p- 17. Hab. Pelagic. (Jn. “from 30 fms.,” W.) Every- where ; very common. Limacina Lesueurii, 1836, d’Orbigny (as Atlanta), Voy. Amér. Mérid. v. 177, xx. 12-15; Souleyet, Rev. Zool. 1840, p. 286, & Voy. ‘ Bonite,’ 11. 216, xi. 11-16 (as Spirialis ventricosa) ; Pelseneer, ‘ Challenger’ Report, p. 24. Hab. Pelagic. (Ju., W.) Everywhere; abundant. . Limacina (Peracle) reticulata, 1840, @Orbigny, Voy. Amer. Mérid. v. 178, xii. 32-85, 39; Souleyet, Rev. Zool. 1840, p. 188, & Voy. ‘ Bonite,’ 11. 220, xii. 17-19 (as Spzrialis clathrata) ; Pelseneer, ‘ Challenger’ Report, p. 34, pl. 1. 7,8. Hab. Pelagic. (W.) Funchal and Canical. 4 spe- cimens. Limacina trochiformis, 1840, d’Orbigny (as Atlanta), Voy. Amér. Mérid. v. 177, xii. 29-31; Souleyet, Rev. Zool. 1840, p. 237, & Voy. ‘ Bonite,’ 11. 223, xiii. 27-34 (as Spiria- lis); Pelseneer, ‘ Challenger’ Report, p.29. Had. Pelagic. (W.) From Funchal, east coast. Not common. Liriola, see Siphonaria. Litorina neritoides 1767, Linné (as Turbo), Syst. Nat. 152. 153. 154. 155. 156. 157. 158. 159. 160. MARINE MOLLUSCA OF MADEIRA. 291 p- 1232: Gwyn Jeffreys, B. C. 11. 861; & Vv. 206, Ixv. 2. Hab. From Norway to Mediterranean and Canaries. (M., L., Jn., W.) All along the coast. Very common. Intorina punctata, 1789, Gmelin, Syst. Nat. p. 38597; Adanson (“Le Marnat”’), Sénégal, p. 168, pl. xi1.1; Philippi, Abb. & Beschr. 11. 198, iv. 11. Hab. From the Mediter- ranean to Cape of Good Hope. (N.,Jn., W.) I accept the species as Madeiran on the authority of Sen. Nobre. The specimens of Mr. Johnson and myself from the East end of the Island and from Porto Santo are too young and in too bad condition for independent recognition. Litorina striata, 1831, King, Zool. Jour. v. 345. no. 51; d’Orbigny, Moll. Canar. (as L. canariensis) p. 78, vi. 8-10, & (L. affinis) p. 79, vi. 11-13.—the young tubercled shell. Hab. From Acores to St. Helena. (M., L., N., W.) Every- where ; very common. LIncina borealis, 1767, Linné (as Venus), Syst. Nat. p. 1184: Gwyn Jeffreys, B. C. 11. 242, v. 2; & v. 179, xxxi. 7. Hab. From America and Iceland to Mediterranean and Mogador. (L.,Jn., W.) Funchal. are. Lucina divaricata, 1767, Linné (as Tellina), Syst. Nat. p- 1120: Gwyn Jeffreys, B. C. (as Loripes) 1. 235, v. 43 & v. 179, xxxii. 5. Hab. English Channel to Mediterranean and Canaries. (M., L., Jn., W.) Everywhere; very common. Lucina lactea, 1767, Linné (as Tellina), Syst. Nat. p. 1119: Gwyn Jeffreys, B. C. (as Loripes) 11. 233, v. 4; & v. 179, xxxil. 4. Hab. Southern Britain to Mediterranean and Canaries. (Jn., W.) Funchal. Very rare. Lucina reticulata, 1791, Poli (as Tellina), Test. Sic. 1. 48, xx. 12; Philippi, Enumeratio, 1. 31, 11.14. Everywhere ; very common. Lucina spinifera, 1803, Montagu (as Venus), Test. Brit. p- 577, xviii. 1: Gwyn Jeffreys, B. C. 11. 240; & v. 179, xxxii. 6. Hab. From Norway to Mediterranean and Canaries. (M., L., N., Jn., W.) Everywhere; very common. Lyonsia norvegica, 1788, Chemnitz (as Mya), Conch.-Cab. x. 345, clxx. 1647-8: Gwyn Jeffreys, B. C. m. 29,11. 1; & v. 190, xlviii. 2. Hab. From Lofotens to Mediterranean. (M., L., Jn., W.) Everywhere; common. Mactra subtruncata, 1778, da Costa, Brit. Conch. p. 198: Gwyn Jeffreys, B. C. 1.419; & v. 188, xli.3. Hab. From North Norway to Black Sea and Mogador. (Jn., W.) Porto 292 REV. R. B. WATSON ON THE Santo, 40 to 50 fms. One stained valve.—The locality whence my specimen came is not a place where ballast is thrown out, and the presence of the species in Mr. John- son’s collection, as also at the Canaries and Mogador, compels me unwillingly to admit itas Madeiran. I do not believe, but cannot disprove. Mangelia, see Plewrotoma. 161. Marginella guancha, 1839, VOrbigny, Moll. Canareaen 88, vi. 382-34. Hub. Canaries. (Jn.,W.) From the North coast.—McAndrew gives it as “ frequent” in Madeira, but his specimens, both in his best collection at Cambridge and in that at the British Museum as well as those he sent me, are all marked as from “ Canaries.” The solitary specimen I got from Madeira was in sand from deep water on the north coast; probably, therefore, not a wanderer, but it remains solitary. In Mr. Johngon’s collection also only one specimen presented itself. 162. Marginella miliaria, 1767, Linné (as Voluta), Syst. Nat. p- 1189; Sowerby, Thes. Conch. 1. 399, Ixxviii. 227-280. Hab. From Portugal to Mediterranean and Mogador. (M., L., Jn., W.) Everywhere; common. M. Philippi (sic), a species given by Sen. Nobre as of Monterosato, who says that it is the M. minuta, Phil. Enum. u. 197, xxvii. 23 (but not of Pfeiffer), is one which I am quite unable to identify among any I know as Madeiran. MM. celata, which Monterosato gives in his ‘Nomenclatura,’ p. 139, as Madeiran from me, is a species I have been accustomed to reckon as M. miliaria, L., differing only in colour. Some confusion seems to exist here. 163. Marginella secalina, 1844, Philippi, Enumeratio, 1. 197, xi. 15 & xxvii. 19; Weinkauff, Conch.-Cab. ed. 2, v. pt. 4, p. 29, iv. 21-28. Hab. Mediterranean to Canaries . (teste McAndrew’s List corrected propria manw and his specimens in the Cambridge Museum). (L., Jn., W.) From Funchal to Punta de Sao Lourengo. Plentiful. 164. Mathilda quadricarinata, 1814*, Broechi (as Turbo), Conch. Foss. p. 375, vil. 6; Kobelt, Jahrb. 1. 226, xi. 2. Hab. Western Mediterranean and St. Helena. (W.) Funchal, Punta de Sao Lourengo, and Porto Santo. Not rare. Megathyris, see Argiope. * A five years’ priority over the Mathilda of Semper has been claimed for the name Cingulina of Adams, but no sinistral-tipped shell can pass under Adams’s definition. See ‘Challenger’ Gasteropoda, p. 499. 165. 166. 167. 168. 169. 170. itil 172. MARINE MOLLUSCA OF MADEIRA. 293 Melampus exiguus, 1832-4, Lowe, Zool. Journ. v. 291, xiii. 6, 7; Pfeiffer, Monog. Auric. Viv. p. 56; id. ibid. Suppl. p- 826; id. Auric. Mad., Malak. Blatter, 1866, p. 142; Paiva, Mal. Mad. p.150. Had. Madeira. (L., W.) North shore, Punta de Sio Lourenco; Selvagens. Rare. Mesodesma cornea, 1791, Poli, Test. Sic. 1. 73, & I. xix. Stoll. Hab. Mediterranean.—lI give this species on the authority of Sen. Nobre, who got it in dredgings from Canical. I have never met with it. Mesalia, see Scalaria. Mitra cornicula, 1767, Linné (as Voluta), Syst. Nat. p.1191; Kiener (as JL lutescens, Lam.), Iconog. Coq. Viv. itt. 31, xi. 32, xi. 36, &c. Hab. Portugal and Mediterranean to Moga- dor. (M., L., N., Jn., W.) Everywhere; excessively common. Mitra Zebrina, 1839, dOrbigny, Moll. Canar. p. 86, vi. 29-31 (not Mf. Zebra, Leach); Sowerby, Thes. rv. 23, sp. 300, pl. xxui.481. Hab. Canaries. (M., L., N., Jn., W.) Everywhere. Alive on the shore-rocks, and there dark green; but in deep water dead and discoloured brown by passage through fish’s stomach. Extremely variable, with proclivities towards IZ. tricolor, Gd. Modiolaria discors, 1767, Linné (as Mytilus), Syst. Nat. p- 1159: Gwyn Jeffreys, B. C. 1. 126; & v. 171, xxviii. 3. Hab. From the U.S, coast to N. Greenland, Canada, Nor- way, and Mediterranean. (Jn., W.) One young valve alone in each collection. The range of the species alone justifies its introduction into this list. Modiolaria subclavata, 1859, Libassi, Atti Panorm. rrr. 18. fig.7; Gwyn Jeffreys, Moll. Lightn. & Pore., P. Z. 8. 1879, p- 568. Hab. From Brittany to Canaries. (L., Jn., W.) From Magdalena, Cabo Giréo, Funchal to East end of Island. Very abundant. Monodonta, see Trochus. Montacuta bidentata, 1803, Montagu (as Mya), Test. Brit. p-44, xxvi. 5: Gwyn Jeffreys, B. C. 11. 208; & v.177, xxxi. 8. Hab. Norway to Mediterranean. (Jn., W.) Funchal. 15 to 20 valves. Montacuta ferruginosa(?), 1808, Montagu (as Mya), Test. Brit. Suppl. 1. pp. 22 & 166, xxvi. 2: Gwyn Jeffreys, B. C. 11. 210; & v. 178, xxxi. 9. Hab. From New England to Greenland, North Norway, Mediterranean, Canaries, and Mogador. 3fms.to 733 fms. (l., W.) Coast from 294. REV. R. B. WATSON ON THE Funchal eastward, and Porto Santo. Very abundant.— McAndrew’s List does not give the species, but in his own copy at the Cambridge Natural History Museum it is entered and initialled by himself for Madeira, Canaries, and — Mogador. The query I have suffixed to the name expresses the grave doubts I feel, more than shared by Dr. Gwyn Jeffreys, of the correctness of this identification; but as he has taken the responsibility of publishing it as Madeiran on the strength of my specimens, I am content to follow him. With more reason than in many like cases elsewhere the shell might pass for new ; but the species is a variable one. 173. Montacuta triangularis, 1897, Watson, antea, p. 264. 174. Murex (Ocinebra) aciculatus, 1822, Lamarck, An. s. Vert. vil. 176: Gwyn Jeffreys, B. C. rv. 310; & v. 218, Ixxxiv. 2. Hab. From the English Channel to Mediterranean and Mogador. (M.,L.,Jn., W.) From Magdalena to island’s East point and Porto Santo. Very abundant. 175. Murex (Coralliophila) brevis, 1832, de Blainville (as Pur- pura), Ann. Se. Nat. 1. 2338, xi. 10; Philippi (as Pyrula squamulata), Enumeratio, 1. 207, xi. 21 (not 31), 1. 180. Hab. Mediterranean. (lL, Jn., W.) Porto Santo. One specimen each.—Mr. Lowe has left no record of locality for his one very young specimen. In the face of confusion in the identification of this species I have unwillingly gone behind Philippi, whose figure and description are un- mistakable. Murex cristatus, Broc.—McAndrew, Report, p. 40, gives this as “ rare ;” for the Canaries he does not mention it: neither at Cambridge nor in the British Museum does his collection contain specimens, but Bellardi (vol. i. p. 87) gives it from Upper Miocene. 176. Murex (Ocinebra) Kdwardsii, 1826, Payraudeau (asPurpura), Moll. Cors. p. 155, vii. 17, 18; de Blainville, Faune frang. p- 129, v. (8) 5. It exists from the Upper Miocene on- wards. Hab. From Bay of Biscay to Mediterranean and Canaries. (M., L., N., Jn., W.) From Funchal westwards. Very abundant. 177. Murex (Ocinebra) erinaceus, 1767, Linné, Syst. Nat. p. 1216: Gwyn Jeffreys, B. C. rv. 806, v. 5; & v. 218, lxxxiv.1. In existence from Middle Miocene onwards. Hab. From Cattegat to Black Sea and Mogador. (M., L., N.,Jn., W.) From Funchal to East point and Porto Santo. Abundant. Murex fusulus, see Trophon. MARINE MOLLUSCA OF MADEIRA. 295 178. Murex (Ocinebra) medicago, 1897, Watson, antea, p. 242. 179. Murex (Pseudomurex) Meyendorfii, 1845, Calcara, Cenno, 180. 181. 182. 183. 184. 185. p- 38, iv. 832; Kobelt, Jahrb. Mal. Ges. 1. 1874, p. 222, ix. 1. Hab. Mediterranean. (L., N., Jn., W.) From Funchal to East point and Porto Santo. Abundant. Mytilus, see Avicula. Mytilus edulis, L._—This is a species which Senhor Nobre gives under var. M. galloprovincialis. Johnson’s collection contains two valves. I have several, chiefly young, but all got in circumstances and with the appear- ances of haying been brought in ballast. Belonging to the North Pacific, probably, and under diverse forms and names from Baffin’s Bay to the Mediterranean, its presence in Madeira is not improbable, but I am not sure that Senhor Nobre, on review of its claims, still supports them. Nassa antiquata, 1897, Watson, antea, p. 241. Nassa costulata, 1804, Renieri (as Buccinum), Tay. Conch. Adr.; Philippi (as B. variadile), Enum. 1. 221, xii. 6; Payraudeau (as B. Cuviert), Moll. Corse, p. 168, viii. 17, 18. Hab. From Portugal to Mediterranean. (L., Jn., W.) Everywhere ; very common. Nassa incrassata, 1777, Stro6m (as Buccinum), Norsk. Selskab. Vid. 1v. 869, xvi. 25: Gwyn Jeffreys, B. C. rv. 351; & v. 219, Ixxxviii. 1. Hab. From Iceland to Mediterranean and Mogador. (M.,L., N., Jn., W.) Everywhere; very common. Nassa limata, 1808, Chemnitz (as Buceinum), Conch.-Cab. x1. 87, clxxxviii. 1808-9 ; Philippi, Enumeratio (as B. pris- maticum), 1. 219; Brocchi (as B. prismaticum), Subappen. 1. 837, v. 5-7. Dates from Upper Pliocene onwards. Hab. Mediterranean to Canaries. (M., L., N., Jn., W.) Everywhere ; very common. Nassa reticulata, 1767, Linné (as Buecinum), Syst. Nat. p- 1204: Gwyn Jeffreys, B. C. rv. 346, vi. 4; &v. Ixxvii. 3. Dates from Lower Pliocene. Hab. From North Norway to Black Sea and Mogador. (L., Jn., W.) Porto Santo, Porto Moniz. Not common. Nassa Watsoni, 1877, Kobelt, Iconog. europ. Meeres-conch. Heft 7, p. 151, xxvi. 5,6. “ Madeira,” Canaries. Hab. The Selvagens (not in Madeira proper). At the Canaries itis common. (L., W.) 186. WNatica Dillwynii, 1826, Payraudeau, Cat. Moll. Corse, p. 120, pl. v. ff. 27, 28 ; Philippi, Conch.-Cab. 2nd ed. 11. pt. 1, p. 69, 296 187. 188. 189. 190. 191. 192. 193. 194. REY. R. B. WATSON ON THE pli. £4. (L., Jn., W.) Punta de Sao Lourenco, Porto Santo, Punta da Cruz, Funchal, Gorgulho, &e. Abundant. Natica funel, 1843, Récluz, P. ZS. p. 207; Philippi in Mart. & Chem. Conch.-Cab. 2nd ed. 11. pt. 1, p. 16, 11. 7, 8 (not 6); Adanson, Sénégal, 1757,p.174, xii.3 (“le Fanel’’). Hab. Senegal. (L., Jn., W.) Punta de Sao Lourengo and Labra, also Porto Santo. Not common. Natica furva, 1897, Watson, antea, p. 248. Natica porcellana, 1839, d’Orbigny, Moll. Canar. p. 84, Vi. 27, 28; Philippi in Mart. & Chem. Conch.-Cab. 2nd ed. II. pt. 1, p- 62, x. 4. Hab. Canaries to St. Helena. (M., L., Jn., W.) Everywhere, but nearly always dead, and lacking the operculum, which is thin, horny, and reddish. Natica variabilis, Recluz, MS.; Reeve, Conch. Icon. 1x. xxiii. 104; Philippi m Mart. & Chem. Conch.-Cab. 2nd ed. 11. pl. xi. f. 3, as WV. labrella, but not that of Lamarck, not the description on p. 68 nor fig. pl. xi. 17, nor pl. xix. f.1; H. Adams, P. Z.S. 1869, p. 274, xix. 8 (as WV. mar- morata). See Watson, ‘ Challenger’ Report, p. 435, for discussion of the species. Had. Canaries. (L., Jn., W.) From the Gorgulho bay to Punta de Sao Lourengo, at the East point of the Island. Common. Neera, see Cuspidaria. Ocinebra, see Murex. Octopus vulgaris, 1798, Lamarck, Mém. Soc. Hist. Nat. Paris, vol. 1. p. 18: Gwyn Jeffreys, B. C. v. 144, frontispiece & pl. vii. 1. Hab. West Indies, and from Scotland to Medi- terranean and Canaries, and Indian Ocean, &c. (Jn., W.) Not rare. Odostomia acuta, 1848, Gwyn Jeffreys, Ann. & Mag. 2nd ger. 11. 338; id., B. C. rv. 130, & v. 211, Ixxii. 8. Had. From Lofotens to Mediterranean. (Jn., W.) Along 8. shore from Funchal eastwards and Porto Santo. Common. Odostomia albella, 1846, Lovén, Ind. Moll. Scand. p. 19: Gwyn Jeffreys, B.C. 1v.121; & v. 211, lxxi. 3. Hab. From Norway to Mediterranean. (L., Jn., W.) Funchal east- wards to East point. Very abundant. Odostomia (Turbonilla) bulinea, 1840, Lowe, P. Z.S. p. 40; Brocchi, Conch. Foss. Subap. p. 388, vi. 7 (as Turbo stri- atus, but not that of Mont.). Hab. Mediterranean. (L. Jn., W.) Funchal to Punta deSao Lourengo and Porto Santo. 195. | IY 198. 199. 200. 202. 203. MARINE MOLLUSCA OF MADEIRA. 297° Very abundant but always dead, the shell bored and the animal eaten. Odostomia (Lurbonilla) clathrata, 1848, Gwyn Jeffreys, Ann. & Mag. 2nd ser. 11. 345; id. B. C. rv. 148, & v. 212, Ixxiv. 9. Hab. From Portugal to Mediterranean; Ireland (?). (W.)Puntade Sao Lourengo. Not rare.—Dr. Gwyn Jeffreys says that McAndrew got it at Orotava in the Canaries; and if so it is probably the unnamed one in his Madeiran list, but I have no note of having seen it in his collection. . Odostomia conoidea, 1814, Brocchi (as Turbo), Conch. Foss. Subap. 659, 2nd ed. vol. 11. p. 495, xvi. 2: Gwyn Jeffreys, B. C. rv. 127; & v. 211, lxxiii. 6. Hab. From North Norway to Mediterranean. (Jn., W.) Funchal, Porto Santo. Not rare. Odostonia conspicua, 1850, Alder, Trans. Tyne Club, 1. 359: Gwyn Jeffreys, B. C. tv. 182; & v. 211, Ixxiii. 9. Had. From Lofotens to Mediterranean. (Jn., W.) Funchal to Punta de Sao Lourengo. Rare. Odostomia (Turbonilla) indistincta, 1803, Montagu (as Turbo), Test. Brit. Suppl. p. 129: Gwyn Jeffreys, B. C. rv. 149; & v. 218, xxv. 1. Hab. From Norway to Mediter- ranean and Canaries. (Jn., W.) Everywhere ; common. Odostomia (Turbonilla) interstincta, 1803, Montagu (as Turbo), Test. Brit. 11. 324, xii. 10: Gwyn Jeffreys, B. C. tv- 151; & v. 219, lxxv. 12. Hab. From Lofotens to Medi- terranean and Canaries. (M., L., Jn., W.) Everywhere ; very common. Odostomia (Turbonilla) lactea, 1767, Linné (as Turbo), Syst. Nat. p. 1238; Gwyn Jeffreys, B. C. rv. 164; & v. 218, lxxvi. 8. Hab. From Finmark to Mediterranean and Canaries. (M., L., W.) Everywhere; common. . Odostomia (Turbonilla) nitidissima, 1803, Montagu (as Turbo), Test. Brit. 11. 299, xu. 1: Gwyn Jeffreys, B. C. ry. 173; & v. 214, Ixxvi. 8. Hab. From Finmark to Mediter- ranean. (W.) Canigal and Punta de Sao Lourenco and Porto Santo. Not rare. Odostomia omphaloessa, 1897, Watson, antea, p. 261. Odostomia (Turbonilla) Pointeli, 1867, de Folin, Fonds de Mer, p. 100, xi. 4; Monterosato, Nomenclatura (as Aniso- cycla), p. 99. Hab. Mediterranean. (W.) Canical, Punta de Sao Lourengo, Porto Santo. Not rare. LINN. JOURN.—ZOOLOGY, VOL. XXVI. 22 * 298 204. 205. 206. 208. 209. Zi 212. 214. REV. R. B. WATSON ON THE Odostomia (Turbonilla) pusilla, 1844, Philippi, Enumeratio, 11. 224, xxviii. 21: Gwyn Jeffreys, B. C. tv. 167; & v. 213, Ixxvi. 4. Hab. From English Channel to Mediterranean. (W.) Two specimens. Odostomia rissoides, 1844, Hanley, P.Z.S. p. 18: Gwyn Jeffreys, B. C.1v. 122; & v. 211, lxxii. 4. Had. From Norway to Mediterranean. (W.) From the Gorgulho and Funchal to Punta de Sao Lourengo and Porto Santo. Very rare. Odostomia (Turbonilla) rufa, 1836, Philippi (as Melania), Enumeratio, 1. 156, ix.7: Gwyn Jeffreys, B. C. 1v. 162; & v. 212, Ixxvi.1. Hab. From Norway to Mediterranean and Canaries. (M., W.) Funchal to East poimt and Porto Santo. Not uncommon. . Odostomia (Turbonilla) scalaris, Philippi (as Melania), Enum. Moll. Sic. 1. 157, ix. 9; & 11. 187: (as Chemnitzia) Gwyn Jeffreys, B.C. rv. 160; & v. 213, Ixxv.7. Hab. From Massachusetts and New England to Finmark and the Mediterranean. (W.) Funchal, two broken specimens. Odostomia (Hulimella) scill@, 1836, Scacchi (as Melania), Notiz. Conchyl. p. 51: Gwyn Jeffreys, B. C. 1v. 169; &v. 213, Ixxvi. 5. Hab. From Finmark to Mediterranean and Canaries. (M., Jn., W.) One specimen alone fell to Mr. Johnson and one to myself. Odostomia tricincta, 1856, Gwyn Jeffreys, Ann. & Mag. p- 185,11. 12,18. Hab. North-west of France to Mediter- ranean and Madeira. (W.) From Funchal to East point. Frequent. . Odostomia turrita, 1844, Hanley, P. Z. 8S. p. 18: Gwyn Jeffreys, B. C. tv. 185; & v. 211, lxxiv.2. Hab. From the Cattegat to Mediterranean. (Jn., W.) Funchal to Hast point and Porto Sauto. Common. Odostomia (Turbonilla) wndata, 1897, Watson, antea, p. 262. Odostomia unidentata, 1803, Montagu (as Turbo), Test. Brit. 111. 324, xxi. 2: Gwyn Jeffreys B.C. 1v. 184; & v. 211, Ixxiv. 1. Hab. From North Norway to Mediterranean. (W.) From Funchal to East point and Porto Santo. Common. . Odostomia (Eulimella) ventricosa, 1843, Forbes (as Par- thenia), Brit. Assoc. AXgean Rep. p. 188; G. O. Sars, Moll. Arct. Norv. p. 209, xi. 19, & xxii.16. Hab. From Norway to Mediterranean. (W.) One specimen. Oliva (Olivella) leucozonias, 1839, Gray, Zool. of ‘ Blossom,’ 215. 216. 217. 218. 219. MARINE MOLLUSCA OF MADEIRA. 299 p- 130, xxxvi. 24; Marrat, in Sowerby’s Thesaurus, rv. sp. 213, pl. cecl. (bis) fig. 446. Neither of these authorities gives any habitat for the species. Weinkauff, who identifies it with O. pulchella, Duclos (see his Monog. pl. v. f. 11, 12), gives Senegal as its habitat. Not knowing it from Madeira, I quote it here simply as given “ from Funchal” in Senhor Nobre’s list, and with extreme hesitation. Omalaxis, see Bifrontia. Ostrea cochlear, 1795, Poli, Test. Sic. 1. 179, xxviii. 28. Hab. From Bay of Biscay to Mediterranean. (Jn., W.) Funehal, &e. Common. Ostrea cristata, 1780, Born, Mus. Ces. p. 112, vii. 3; Poli, Test. Sic. 11. 177, xxviii. 25-27. Hab. Mediterranean. (L., Jn., W.) Magdalena, Funchal, &e. Common. Ovula carnea, 1789, J. L. M. Poiret (as Bulla), Voyage, 1. p- 21; Bruguiére, Enecycl. 111. 685, sp. 5, pl. ccelxvii. 2; Kiener, Icon. Coq. Viv. 1. 10, vi. 2. Hab. Mediterranean. (W.) Magdalena. One specimen brought up from 100 fathoms in a jar accidentally entangled in a fisherman’s line. Oxygyrus Keraudren, 1817, Lesueur (as Atlanta), Jour. de Phys. uxxxv. 391, ii.; Woodward, Man. Moll. pl. xiv. 24,25. Hab. Pelagic. (Jn., W.) From Funchal to East point, &e. Panopea, see Saxicava. Parzetta.—The confusion both in the classification and no- menclature of this genus is very great. I touch on no disputed points when I state that no one at work on the shore of Madeira can doubt that there are four, and only four, very distinct and very easily differentiated species, and under the law of priority their names fall to them without difficulty. Alphabetically arranged they are :— Patella aspera, 1819, Lamarck, An. s. Vert. v1. (1) 328; 2nd ed. by Deshayes, vit. 529; d’Orbigny, Moll. Canar. p- 98, vu. 9,10 (as P. Lowez). [Hanley, see Wood, Ind. Test. p. 185, xxxvii. 19, says that this is Dilwyn’s P. re- panda of Gmel. Syst. Nat. ed. Linn. Syst. Nat. 13+; but both the locality “freto Magellanico,’ and the description “testa tenut .... margine repando,” remove Gmelin’s species very far from that found in Madeira. Dillwyn, indeed, ascribes Gmelin’s P. repanda to “ Island of Cerigo,” but gives no authority for this assertion of Fayanne.] 22* 300 220. 221. 222. 223. 224. REY. R. B. WATSON ON THE Hab. From the Bay of Biscay to Mediterranean and Canaries. (M.,1L., N., Jn.,W.) Very common everywhere. Patella cerulea, 1767, Linné, Syst. Nat. p. 1259 (see Han- ley, Ips. Linn. Conch. p. 421); Philippi, Enum. 1. 110, m1. 84, vu. 5; Hidalgo, Moll. Mar. Esp. pl. 1. 1-8, h. 1, 2; d’Orbigny (as P. crenata), Moll. Canar. p. 97, pl. vii. 1-8. Hab. Bay of Biscay to Mediterranean and Canaries. (M., L., N., Jn., W.) Very common everywhere. Patella rustica, 1767, Linné, Syst. Nat. p. 1261 ; Wood (as P. lusitanica, Gmel., see Syst. Nat. p. 3715), Index Test. p- 188, pl. xxxvi. 66: Lamarck (as P. punctata), An. s. Vert. vr. (1) 338; ed. 2% Deshayes, vir. 537. no. 34; Delessert, pl. xxii. 4; Reeve (as P. nigropunctata), Conch. Icon. pl. xxxiii. 57; d’Orbigny, Moll. Canar. p. 97, pl. vii B. 13-15. Hab. Bay of Biscay to Canaries. (M.,L.,N., Jn.; W.) Very common everywhere. Patella vulgata, 1767, Linné, Syst. Nat. p. 1258; d’Orbigny (as P. Candez), Moll. Canar. p. 98, pl. vii B. 11,12: Gwyn Jeffreys (as P. vulgata), B. C. 111. 236, v. 3; & v. lvii. 1-4. Hab. From north of Norway to Mediterranean and Canaries. (M., L., N., Jn., W.) Very common everywhere. Patella Gussonii, see Siphonaria. Pecten corallinoides, 1839, d Orbigny, Moll. Canar. p. 102, pl. viis. 20-22; Sowerby, Thes. Conch. 1. 65, xii. 3, 4; Kiister, in Mart. & Chemn. Conch.-Cab. 2nd ed. vit. pt. 2, p- 58, pl. xv. 7-9 (who denies the correctness of Sowerby’s identification, which may however pass). Hab. Canaries and Madeira. (M., L., N., Jn., W.) Everywhere; very common. Pecten fenestratus, 1843, Forbes, Brit. Assoc. Report, p. 192; Gwyn Jeffreys (as Plewronectia), Med. Moll., Ann. & Mag. July 1870, p. 40, and (2s Amussiwm) in Report on the Lightn. & Pore. Exped., P. Z. 8. 1879, p. 561 (identify- ing Forbes’ sp., however, with Pecten Philippii, Acton, Ricerche Conch. fig. 1 =P. Actoni, v. Martens, Malak. Blitter, 1857, p. 194, iii. 1-8, but mentioning at the same time some remarkable features of the species which made him hesitate in the identification). Kobelt, Prodromus, p. 440, extends this synonymy, and in his Monograph in the Conch.-Cab. 2nd ed. p. 268, pl. lxx. ff. 7, 8 (a copy of Acton’s figure), quotes Verrill, Proc. U.S. Nat. Mus. 1880, p- 403, and also Trans. Connect. Acad. 1882, v. p. 582, for 225. 226. 227. 228. 229. 230. 231. MARINE MOLLUSCA OF MADEIRA. 30L this species as found on the Hast coast of America; but | this identification Verrill (1884, Proc. Conn. Acad. vr. 261) now rejects. Hab. North Atlantic and Mediterranean. (Jn., W.) Punta de Sao Lourengo. Very rare. Pecten flexuwosus, 1795, Poli (as Ostrea), Test. Sicil. 11. 159- 161, xxviii. 1-8, and (as O. plicata) fig. 11; Philippi (as P. polymorphus), Enumeratio, 1. 79, & 11. 57, pl. v. 18-21. Hab. Mediterranean. (M., L., Jn., W.) Everywhere; common. Pecten Jacobeus, 1767, Linné (as Ostrea), Syst. Nat. p. 1149 ; Pennant, Brit. Zool. tv. 100, xl. 1. Hab. Mediterranean to Canaries. (M., L., N., Jn., W.) Everywhere; common. —The P. maximus of McAndrew’s collection in Cambridge, from the Canaries is a young specimen of P. Jacobeus. The same is probably the case in other instances. Pecten Loveni, 1853, Dunker, Index Moll. Guin. p. 44. no. 115, ix. 81. Hab. The Guinea coast. (N.) Dunker got two separate valves from Loanda. Nobre, besides adding Sierra Leone, gives it as dredgedin Funchal Bay.— The species is one I have never seen. Dunker’s authority is of course of great weight; his figure, but less so his description, suggests the young form of P. fenestratus, Forbes. Pecten pes felis, 1758, Linné (as Ostrea), Syst. Nat. ed. 10. vol. 1. 697, & 1767, do. do. ed. 12. p. 1146; Sowerby, Thes. Conch. t. 67, xvii. 162, xx. 184. Hab. From Bay of Biscay to Mediterranean and Mogador. (M., L., N., Jn., W.) From Magdalena to Punta de Sao Lourengo. Not common, and the valves never joined. Pecten pusio, 1767, Linné (as Ostrea), Syst. Nat. p. 1146: Gwyn Jeffreys, B. C. 11. 51; & v. 166, xxii. 1. Hab. From Norway to Mediterranean and Cape of Good Hope. (M., L., N., Jn., W.) Everywhere ; very abundant. Pecten similis, 1811, Laskey, Mem. Werner. Soc. 1. 387, vii. 8: Gwyn Jeffreys, B.C. 11. 71; & v. 168, xxii. 5. Had. From Finmark to Mediterranean (and Jamaica, f Barrett). (M., L., Jn., W.) Everywhere, from Magdalena to Punta de Sao Lourengo and Porto Santo. Very abundant. Pecten solidulus, 1853, Reeve, Conch. Icon. vii. pl. xxxiil. 155; Philippi (as P. gibbus, Lam.), Enumeratio, 1. 83; Récluz (as P. Philippi, Michel.), Jour. de Conch. tv. 52, u. 15; Monterosato, Coq. Maroc., Jour. de Conch. 1889, 302 232. REV. R. B. WATSON ON THE p. 20. Hab. Mediterranean to Canaries. (M., L.,N., Jn., W.) Everywhere ; very abundant.—The Marquis of Monte- rosato has done a great service in showing that this species is not the P. gibbus, Lam., nor the fossil P. Philippi, Michelotti of 1839, and that his own name of P. commu- tatus for the species is later than Reeve’s. After following him “ per tot discrimina” nominum, we may hope that this unfortunate species will rest under Reeve’s shelter. Pecten varius, 1767, Linné (as Ostrea), Syst. Nat. p. 1146. Hab. From Norway to Mediterranean.—This species enters here as one given in Senhor Nobre’s list, supported, how- ever, by an entry in Mr. Lowe’s List (Journ. Linn. Soe. 1860, p. 173) of shells observed at Mogador in which Pecten varius occurs. Mr. Lowe’s own copy, however, of that paper, along with the three separate valves sent him from Mogador, is in my possession, and a note appended by himself states that he had not found the species, but received it along with fourteen others (and probably several more, see l. ¢. p. 172) “from Mrs. Elton.” 233. Pectunculus glycimeris, 1767, Linné (as Arca), Syst. Nat. p. 1148: Gwyn Jeffreys, B. C. 11. 166, iv. 4; & V. 175, xxx. 2. Hab. From the Lofotens to Mediterranean, Mogador, and Canaries. (M., L., N., Ju., W.) Funchal, Labra, Punta de Sao Loureng¢o, Porto Santo. Abundant. Pectunculus siculus, 1843, Reeve, Conch. Icon. vol.i. Monog. 5, pl. vii, fig. 41.—McAndrew alone gives this species (a syn. of P. bimaculatus, Poli), and he has added to it ‘‘ frequent,” but no one else has found it. In the British Museum shells ticketed “ P. siculus, presented by Mr. McAndrew,” are marked ‘‘Canarian.” The locality whence the Cambridge Museum specimens belonging to his own more special collection were derived is not noted, but the name attached indicates that he held the species to be distinct from Pecten glycimeris. 234. Pedipes afra, 1790, Gmelin (as Helix), Syst. Nat. p. 3795. no. 194; Lowe, Zool. Jour. v. 296, xiii. 8-12, and P.Z.S. 1854, p. 217 (note under Truncatella), & p. 218 ; Deshayes, Lamarck, An. s. Vert. 2nd ed. 1x. 42; Pfeiffer, Monog. Auric. 1.68; H. & A. Adams, Genera, 111. pl. Ixxxii. 4, 4°. The whole credit of this species really belongs to Adanson (Hist. Sénégal, p. 11, pl. i. 4), who carefully, minutely, and ~ most accurately described and figured the animal and the shell under the name of “ Le piéton—Pedipes.” Hab. From Lisbon (see Wollaston, Test. Atlant. pp. 50, 265, 293) to Senegal and St. Helena. (Jn., W.) Very common MARINE MOLLUSCA OF MADEIRA. 3803 under stones and large shingle between high- and low-water mark. Peplidea, see Plocamophorus. 235. Phasianella pulla, 1767, Linné (as Turbo), p. 1238: Gwyn Jeffreys, B. C. 111. 338, vili.1; & v. 204, Ixiv.1. Hab. From Great Britain to Mediterranean, Mogador and Canaries. (M., L., N., Jn., W.) Everywhere ; very common. 236. Philine aperta, 1767, Linné (as Bulla), Syst. Nat. p. 1183: Gwyn Jeffreys, B. C. 1v. 457, viii. 7; & v. xcvi. 1. Hab. . From North Norway to Mediterranean and Cape of Good Hope. (M., L., Jn., W.) Everywhere ; common. 237. Philine complanata, 1897, Watson, antea, p. 285. 238. Philine desmotis, 1897, Watson, antea, p. 236. 239. Philine scabra, 1776, Miller (as Bulla), Zool. Dan. Prod. 1. 41, Ixxi. 11,12; Wood, Crag Moll., Gasterop. p. 181, xxi. 12 (as Bulla); Philippi, Enum. 1. 121, vii. 17 (as Bulla anqgustata); Forbes & Hanley, Brit. Moll. 111. 543, exiv E. 4,5: Gwyn Jeffreys, B. C.1v. 447; & v. 224, xevi. 1; G. Sars, Moll. Norv. p. 294, xviii. 13. Hab. From Greenland to Mediterranean. Fossil in English Coralline Crag and the older glacial clay beds of Norway. (W.) Funchal, 50 fms. —I think it just possible that my specimens may belong to Miiller’s species (which finds representation under some- what diverse figures), and cannot take the responsibility of describing the Madeira species as new on the strength of merely two specimens, obviously of one species, but of which one is little more than embryonic and the other is slightly chipped. At the same time I greatly doubt whether they can be left under this name: their spire is too high, the corner of the outer lip droops too far below the plane of the apex, and the sculpture is too fine, too uniform, and too sparse to suit Miiller’s species. 240. Philine trachyostraca, 1897, Watson, antea, p. 236. 241. Pinna rudis, 1767, Linné, Syst. Nat. p. 1159: Gwyn Jeffreys, B. C. 1v. 99, frontispiece & iii. 1; & Vv. 170, xxvi. Hab. From Great Britain to Mediterranean and Canaries. (M., L., Jn., W.) Everywhere; common. 242. Plewrobranchus Dautzenbergi, 1897, Watson, antea, p. 239. 243. Pleurobranchus Lowei, 1897, Watson, antea, p. 240. 244, Pleurobranchus plumula, 1803, Montagu, Test. Brit. 1. 214, xv. 9, vignette ii. 5; Gwyn Jeffreys, B. C. v. 11, xevil. 4. Hab. From Bohusliin, Sweden, tothe Aigean. (Jn., W.)— 304 245. 246. 247. 248. 249. 251. REV. RB. B. WATSON ON THE This species is poorly represented by two small rather damaged shells and a somewhat broken but well-grown specimen whose dorsal margin-line is straighter than usual, and in that respect resembles P. brevifrons, Phil., which, however (f. Monterosato), cannot reckon as a species. , Pleurotoma (Mangelia) anceps, 1830, Hichwald, Naturhist. Skiz. Lithauen &c. p. 225: Gwyn Jeffreys, B. C. (as Defrancia teres) tv. 362, & v. 219, lxxxvili.5. Hab. From Lofotens to Mediterranean and Canaries. (M., Jn., W.) Funchal, Punta de Sao Lourengo, 30 to 40 fms. Not common. Pleurotoma (Mangelia) gracilis, 1803, Montagu (as Murex), Test. Brit. p. 267, xv. 5; Gwyn Jeffreys, B. C. 1v. 363, vy. 219, Ixxxviii.6. Hab. Gt. Britain to Mediterranean and Canaries. (M., L., N., Jn., W.) From Magdalena to Punta de Sao Lourengo. Common. Pleurotoma (Clathurella) histriv, 1832, Cristofori & Jan, Catalogus Conch. Foss. p.10; Gwyn Jeffreys, Ann. & Mag. Nat. Hist. ser. 4, vr. 82; Brugnone, Pleur. Foss. Palerm. p- 28, i. 21; Bellardi, Moll. Terz. Piemonte, vol. 11. 266; Monterosato, Enumeratio, p. 46. Hab. Mediterranean. (Jn., W.) Funchal, Punta de Sao Lourengo. Very abundant. Pleurotoma (Mangelia) incrassata, 1837, Dujardin, Mém. Tour. p. 292, xx. 28; Philippi, Enumeratio, 11. 168, xxvi. 6. Hab. Mediterranean. (Jn., W.) Funchal, Punta de Sao Lourengo. Rare. Pleurotoma (Clathurella) Leufroyi,1828, Michaud, Bull. Soe. Linn. Bord. 11. 121, 1.5, 6; Gwyn Jeffreys, B. C. Iv. 366 ; & v. 219, lxxxix.1. Hab. From Norway to Mediterranean and Canaries. (Jn., W.) Funchal, Punta de Sao Lourengo, Porto Santo. Very abundant. . Pleurotoma (Clathurella) linearis, 1803, Montagu (as Murex), Test. Brit. 1. 261, ix. 4: Gwyn Jeffreys, B. C. rv. 368, vil. 1; & v. 220, Ixxxix. 2. Hab. From Iceland and Northern Norway to Mediterranean and the Canaries, (M., L., W.) From Funchal to Punta de Sao Lourengo and Porto Santo. Very abundant. Pleurotoma (Mangelia) nebula, 1803, Montagu (as Mures), Test. Brit. p. 267, xv. 6: Jeffreys, B. C. rv. 884; & v. 220, xci. 1. Hab. From North Norway to Mediterranean and 252. 253. 254. 255. 256. 257. 258. 259. MARINE MOLLUSCA OF MADETRA. 305 Canaries. (M., L., N., Jn., W.) Everywhere; very abun- dant. Mr. Johnson mentions having got it from 30 fms. Pleurotoma (Clathurella) purpurea, 1803, Montagu (as Murex), Test. Brit. p. 260, ix. 3: Gwyn Jeffreys, B. C. 1v. 873; & Vv. 220, Ixxxix. 5, 6. Hab. From Gt. Britain to Mediterranean and Canaries. (M., L., W.) Everywhere; abundant. Pleurotoma (Clathurella) reticulata, 1804, Renieri (as Murex), Tay. Alfab. p. 2: Gwyn Jeffreys, B. C. 1v. 370, & v. 220, Ixxxix.3,4. Hab. From Norway to Mediterranean. Everywhere ; common. Pleurotoma (Mangelia) rugulosa,1844, Philippi, Enumeratio, 11. 169, xxvi. 8: Gwyn Jeffreys, B. C. Iv. 881; & v. 220, xe. 4. Hab. S.W. England to Mediterranean. (L., Jn., W.) Very abundant.—This is the species which Monterosato (Nomenclatura, p. 1380) identifies as Mangelia Stosiciana, Brus., Jour. de Conch. 1869, p. 235. Pleurotoma (Mangelia) septangularis, 1803, Montagu (as Murex), Test. Brit. p. 260, ix.5: Gwyn Jeffreys, B. C. rv. 390; & v. 222, cxi. 5. Hab. From South Norway to Madeira and Canaries. (M., L., Jn., W.) Everywhere; very abundant. Pleurotoma (Mangelia) striolata, 1836, Scacchi, Catalogus, p- 12; Philippi, Enumeratio, 1. 168, xxvi. 7: Gwyn Jeffreys, B.C. 1v.376; & v.220,xc.1. Hab. From Norway to Mediterranean and Canaries. (M., L., N., Jn., W.) Everywhere ; extremely abundant. Johnson mentions having got it in 380 fms. Pleurotoma (Mangelia) Vauquelini, 1826, Payraudeau, Moll. Corse, p. 145, vii. 14, 15; Kiener, Iconog. vol. v. Pleurot. p. 76, xxvi. 2; Weinkauff, Conch. Mittelm. mm. 166; Kobelt, Prodrom. Moll. Mar. Europ. p. 141. Hab. Medi- terranean. (M., L., Jn., W.) Everywhere; abundant. Poromya granulata.—This species, as represented by a single valve, appears in McAndrew’s List, p. 37, but it is not to be found in his col- lection either at the British Museum or at Cambridge, and no one else has met with it in Madeira. Plocamophorus madere, 1842, Lowe (as Peplidea), P. Z. 8. p- 51. (L.) Gorgulho Bay. Psammobia costulata, 1822, Turton (as Tellina), Conch. Die. p- 87, vi. 8: Gwyn Jeffreys, B. C. 11. 394; & v. 187, xl. 2. 3806 REV. R. B. WATSON ON THE Hab. From Norway to Mediterranean and Canaries. (M., L., W.) Everywhere; very common. 260. Psammobia ferroensis, 1782, Chemnitz (as Tellina), Conch.- Cab. vi. 99, x. 91: Gwyn Jeffreys, B. C. 11. 396; & v. xlu. 38. Hab. From Iceland to Mediterranean and Canaries, and fossil from the later Italian Tertiaries. (N.) Senhor Nobre alone has been fortunate enough to secure this species: ‘‘ Dredged at Canical.”’ Pseudomurex, see Murex. 261. Purpura hemastoma, 1767, Linné (as Buccinum), Syst. Nat. p- 1202; Philippi, Enumeratio, 1. 218, & 11. 187, xxvii. 2. Mediterranean to Senegal (see Adanson, as Purpura Sakem, p- 100, vii. 1), Cape de Verdes (Bowditch, p. 242), Lagos (E. A. Smith, P2Z. 8: 1871) p. 732). «Gi, Ne dare) Everywhere ; very common. Pyrula, see Murex. 262. Ranella (Aspa) marginata, 1788, Gmelin (as Buccinum), Syst. Nat. p. 3486. no. 63; Martini (as Utriculus fimbriatus- dentatus), Conch.-Cab. 111. 425, pl. exx. 1101-2; Sowerby, Genera, 11. pl. ccili. f. 2; Kiener (as R. levigata), Iconog. vil. p. 34, xii. 2; Deshayes (2. levigata), Encycel. 111. 882 ; Lamarck (2. levigata), An. s. Vert. vit. 154, 2nd edit. (Deshayes) rx. 550; Kuster (2. levigata), Conch.-Cab. 2nd edit., 111. pt. 2, p. 154, xxxixa. 8; Hornes (R. levigata), Foss. Moll. Wien. Bass. 1. 214, xxi. 7-11; Seguenza (ft. marginata), Formazioni Terz. p. 108. Hab. Canaries and N.W. Africa, but best known asa fossil. (L., W.) Funchal, Porto Santo. Living, but rather rare. 263. Ranella reticularis, 1780, Born (as Murex), Mus. Vind. p- 300, pl. xi, 5; Lamarck (as &. gigantea), An. 8. Vert. vir. 150, 2nd ed. (Deshayes) 1x. 540; Reeve, Conch. Icon. vol. 1. sp. 38; Philippi, Enumeratio, 1. 211, & m. 183. Hab. From Bay of Biscay to Mediterranean. (L., N., Jn., W.) Porto Santo, Funchal. Rare.—This is Murex olearium of Linné’s 10th, but not of his 12th edition. 264. Ranella scrobiculator, 1767, Linné (as Murex), Syst. Nat. p- 1218; Adanson, Sénégal (as “ Jabik”’), p. 121, pl. viii. 18 ; Deshayes (as Zriton), Encycl. 11. 1056, pl. eecexiv. 3; d’Orbigny (as &. abbreviata), Moll. Canar. p. 94. Hab. Mediterranean to Senegal. (L., N., Jn., W.) From Gor- gulho Bay to Punta de Sao Lourengo. Abundant. 265, 266. 267. 268. 269. 270. 271. MARINE MOLLUSCA OF MADEIRA. 307 Ranella Thome, 1853, d’Orbigny, Moll. Cuba, 11. 164, xxiii. 23. Hab. Island of St. Thomas, W. Indies. (Jn., W.) Punta de Sao Lourengo and Porto Santo; 5 full-crown (8 living) and 4 young shells. There are 2 specimens in the Brit. Mus. from the Cape Verd Is.—I have failed to trace this species to 1846, to which year d’Orbigny ascribes it. I+ does not occur in his Amér. Méridion. For note on the species, see Watson, ‘Challenger’ Report, p. 400. Ringicula auriculata, 1811, Ménard (as Warginella), Ann. Mus. xvi. 331; Philippi, Enumeratio, 1. 231, 1. 198, xxvil. 13; Morelet, Jour. de Conch. 1878, p. 275, v. 14; Watson, Jour. de Conch. 1878, p. 312, x. 4. Hab. Bay of Biscay to Mediterranean. (M., L., N., Jn., W.) Every- where ; extremely abundant. Ringicula Somert, 1867-71, de Folin, Fonds d. 1. Mer, 1. pt. 1, p. 14, 1. 7; Morelet, Monog. Ringicula, Jour. de Conch. 1878, p. 128, v. 8. Hab. Cape Verd Is. (N.)— I give this species entirely on the authority of Senhor Nobre. It is one I have not seen, and of which T cannot judge. Among the very great number of specimens of Ringicula collected in Madeira I have never seen any variety on which to found a species, and indeed the peculiarities obvious in the figures of R. Someri are exactly those of the immature shell of R. awricuiata. Rissoa (Onoba) abjecta, 1873, Watson, Mad. Moll., P. Z.S. p. 385, xxxvi.23; Weinkauff, Conch.-Cab. 2nd ed.1. pt. 22, p- 164, xx. 15. Hab. Madeira. (W.) Funchal. One specimen, now lost. Rissoa (Cingula) albugo, 1873, Watson, P. Z. 8. p. 379, xxxv. 17; Wienkauff, Conch.-Cab. 2nd ed. 1. pt. 22, p. 150, xviii. 9. (Jn., W.) Everywhere ; abundant. Rissoa (Alvania) aurantiaca, 18738, Watson, P. Z.8. p. 367, xxxiv. 8; Weinkauff, Conch.-Cab. 2nd ed. 1. pt. 22, p. 145, xvili.l. (Jn., W.) Everywhere abundant.—The R. auran- tiaca of Brusina is (f- Monterosato) Barleeia rubra, Mont. Rissoa (Cingula) callosa, 1868, Manzoni, Jour. de Conch. pp. 166 & 214, x. 3; Weinkauff, Conch.-Cab, 2nd ed. 1. pt. 22, p. 153, xvii. 14,15. (Jn., W.) Funchal eastwards to Punta de Sao Lourenco and at Porto Santo. From 80 fms. upwards to the shore.—The identification of this species is 308 REV. R. B. WATSON ON THE that of Mr. McAndrew. He was kind enough to send me specimens ‘of Manzoni’s species, but they were too much rubbed to render independent estimate possible. The figures of the species in the ‘Journal de Conchylioiogie ’ (J. c.) are not good, and as reproduced in the ‘ Conchy- lien-Cabinet’ (J. ¢c.) and in Tryon’s ‘ Manual’ (rx. 350, lxv. 6) have still less character. 272. Rissoa (Alvania) canariensis, 1839, d’Orbigny, Moll. Canar. p: 78, vi. 5-7; Watson, P.Z.8. 1878, p. 376, xxxv. 13; Weinkauff, Conch.-Cab. 2nd ed. 1. pt. 22, p. 118, ii. 19, 20. Hab. Mediterranean to Canaries, (L., Jn., W.) Everywhere; very abundant. Semifossil in the Canigal beds. 273. Rissoa (Alvania) cancellata, 1778, da Costa (as Turbo), Brit. Conch. p. 104, viii. 6-9: Gwyn Jeffreys, B. C. rv. 8; & v. 207, lxiv.3; Watson, P.Z.S. 1873, p. 367, xxxiv. 2. Hab. From Gt. Britain to Mediterranean, Mogador, and Canaries. (M., L., N., Jn., W.) Everywhere ; very abundant. 274. Rissoa coriacea, 1868, Manzoni, Jour. de Conch. pp. 166 & 242, x.6; Watson, P. Z.8. 1873, p. 369, xxxvi. 27; Wein- kauff, Conch.-Cab. 2nd ed. 1. pt. 22, p. 141, xvi. 11. Hab. Mediterranean. (Jn., W.) From Funchal to Punta de Sio Lourenco, in 30 to 50 fms. Not common. 275. Rissoa (Alvania) costata, 1796, John Adams (as Turbo), Trans. Linn. Soc. m1. 65, xiii. 18, 14: Gwyn Jeffreys, B. C. tv. 22; & v. 207, Ixviti. 2; Watson, P.Z.S. 1878, p- 3869, xxxiv. 5. Hab. From Norway to Mediterranean and Canaries. (Jn., W.) Everywhere ; abundant. 276. Rissoa (Alwania) costulata, 1844, Alder, Ann. & Mag. vol. XIII. p. 324, viii. 8, 9; Gwyn Jeffreys, B. C.1v.35; & v. 208, lxvili. 1; Watson, P.Z.S. 1873, p. 378, xxxv. 15. Hab. From Gt. Britain to Mediterranean. (L., Jn., W.) Everywhere; very abundant. 277. Rissoa (Alvania) crispa, 1873, Watson, P. Z. S. p. 369, xxxiv. 6; Weinkauff, Conch.-Cab. 2nd ed. 1. pt. 22, p. 157, xix. 9, 10. (Jn., W.) From Funchal to Punta de Sao Lourenco and Porto Santo. Abundant. 278. Rissoa (Cingilla) cristallinula, 1868, Manzoni, Jour. de Conch. p. 5, x. 2; Weinkauff, Conch.-Cab. 2nd ed.1. pt. 22, p- 153, xviii. 18, 16. Hab. Canaries. (W.) Porto Santo. Two specimens. 279. 280. 281. 282. 283. MARINE MOLLUSCA OF MADEIRA. 3809 Rissoa (Cingula) depicta, 1868, Manzoni, Jour. de Conch. p. 168, x. 4; Watson, P.Z.S. 1873, p. 382, xxxv. 20; Weinkauff, Conch.-Cab. 2nd ed. 1. pt. 22, p. 152, xvii. 12. Hab. Canaries and Madeira. (Jn., W.) From Funchal to Punta de Sao Lourengo. Abundant.—This Madeiran species is quite obviously distinct from &. semistriata, Mont., and from &. Galvagne, Arad.; probably it is distinct from Manzoni’s species and would be better called 2. punctifera, as the Monog. of 1873 (l. c.) suggests, but satisfactory specimens of Manzoni’s Canarian Ft. depicta are wanting for present revision. I leave it therefore as given in the P.Z._S. of 1873. Rissoa (Alvania) euchila, 1886, Watson, P. Z. S. 1878, p- 377, pl. xxxv. 13 (as R. novarensis, but not that of von Frauenfeld as stated on the authority of Baron Schwarz vy. Mohrenstern; the error was corrected in the ‘ Chal- lenger’ Report, and frequently since); Weinkauff, Conch.- Cab. 2nd ed..1. pt. 22, p. 147, xvii. 14. (Jn., W.) From Funchal to Punta de Sio Lourengo, at various depths. Abundant. Rissoa (Orossea) gibbera, 1873, Watson, P. Z.8. p. 371, xxxiv. 7; Weinkauff, Conch.-Cab. 2nd ed, 1. pt. 22, p. 158, xix. 11,12. (Jn., W.) From Funchal to Punta de Sio Lou- renco and Porto Santo. Abundant.—I do not believe this species to be a Rissoa, and doubt the claim of Crossea to be there admitted ; but with entire ignorance o the animal, it is better guieta non movere. Rissoa (Pisinna) glabrata, 1824, von Mihlfeldt (as Helix), Verhand. naturhist. Gesellsch. 1. 218, i. 10; Philippi (as R. punctulum), Enumeratio, 1. 154, x. 11, & 1. 180; Watson, P. Z.S. 1873, p. 386, xxxvi. 24. Hab. From Bay of Biscay to Mediterranean and Canaries. (L., Jn., W.) From Funchal to Punta de Sio Lourenco. Abundant. Rissoa (Cingula) imnominata, 1897, Watson (1873 as R. concinna, Monterosato), P.Z. 8. 1873, p. 381, xxxv. 19. (Jn., W.) Everywhere; abundant.—There is difficulty over both the differentiation and the name of this species. In 1869 the Marquis of Monterosato published a Medi- terranean species as Lt. concinna (see Test. Nuov. Sicil. p. 8. no. 2). In 1875 (Nuov. Revista, p. 26) he ranked it as a var. of Rk. Galvagni, Arad.; but in 1878 (Enumerazione &c. p- 26) restored it to specific rank, but marked that the name 310 284. 285. 286. 287. REY. R. B. WATSON ON THE of F. concinna he had given it was preoccupied by Searles Wood for a different species; and in that belief in his ‘ Nomenclatura,’ 1884, p. 66, he changed the name of his Mediterranean species to R. beniamina. But &. concinna was a mere catalogue-name which Searles Wood, when he came for the first time to describe the species, replaced by a true specific name, viz. &. punctura. R. concinna thus remains as the name of Monterosato’s Mediterranean species. That name therefore I should have adopted here for the Madeiran species now in view but for the fact that Monterosato rejects my identification of the Madeiran species with his from the Mediterranean. The material for a thorough independent opinion I do not possess, and I am content to accept the judgment of an authority in every way so trustworthy—compelled, however, in these circumstances to propose for the Madeiran species the new name I have given above. Rissoa (Alvania) Leacocki, 1873, Watson, P. Z. 8. p. 367, xxxiv. 1; Weinkauff, Conch.-Cab. 2nd ed. 1. pt. 22, p. 157, xix. 8. (L., Jdn., W.) Gorgulho shore, Punta de Sao Lourenc¢o down to 45 fms., Porto Santo to 50 fms., Selvagens. Abundant, Rissoa (Pisinna) lincta, 1873, Watson, P. Z. 8. p. 387, xxxvi. 26; Tryon, Manual of Conch. 1x. 341, Ixix. 47. Hab. Madeira, Canaries. (W.) Funchal and Santa Cruz. Five specimens from Madeira, one from Tenerife. Rissoa (Alwania) Macandrewi, 1868, Manzoni, Jour. de Conch. pp. 164 & 237, x. 1; Watson, P. Z.S. 1873, p. 372, xxxiv. 8; Weinkauff, Conch.-Cab. 2nd ed. 1. pt. 22, p. 159, xix. 18-15. (Jn., W.) Funchal, Cruz Point. Very abundant. Rissoa (Onoba) Moniziana, 1873, Watson, P.Z.S. p. 378, xxxiv.10; Weinkauff, Conch.-Cab. 2nd ed. 1. pt. 22, p. 146, xvi. 2, 8. (Jn., W.) From Funchal to Punta de Sao Lourenco and Porto Santo. Abundant. Rissoa (Alvania) Montagui, 1826, Payraudeau, Moll. Corse.—One speci- men which I got in Funchal Bay, 50 fms., the ordinary anchorage and discharge for ballast, is, so far as I know, the only representative of this species found in Madeira ; and I stated in P. Z. 8. 1878, p. 390, that I did not on such evidence regard it as Madeiran. Weinkauff, however (Conch.-Cab. 2nd ed. 1. pt. 22, p. 114), gives it as Madeiran, but without quoting any authority. It ought not without further evidence to reckon as Madeiran. MARINE MOLLUSCA OF MADEIRA. 311 Rissoa parva, 1778, da Costa (as Turbo), Brit. Conch. (W.) Funchal, 50 fms. Two rubbed and broken specimens, doubtless not Madeiran, but brought in ballast. 288. Hissoa (Cingilla) picta, 1867, Gwyn Jeffreys, Ann. & Mag. p: 485; Watson, P.Z.S. 1873, p. 881, xxxv. 18; Wein- kauff, Conch.-Cab. 2nd ed. 1. pt. 22, p. 151, xviii. 10. (L., Jn., W.) Everywhere in shallow water. Very abundant. Rissoa (Setia) pulcherrima, Gwyn Jeffreys; Watson, P. Z. 8. 1873, p. 383, is now suppressed. 289. Rissoa (Pisinna) sabulum, 1842, Cantraine, Bull. Acad. Brux. 1x. pt. 2, p. 848; Watson, P. Z.S. 1873, p. 387, xxxvl. 25. Hab. Mediterranean. (L., Jn., W.) In shallow water from Funchal to Punta de Sao Lourenco, Porto Santo, and the Selvagens. Abundant, but local. —It seems strange that both this species and R. glabrata, v. Muhl., should be identified with R. punctulum, Phil. 290. Rissoa similis, 1836, Scacchi, Cat. Reg. Neap. p. 14; Wat- son, P. Z.S. 1873, p. 379, xxxiv. 16, 16a (given then, as now, in utter disbelief in the whole group into which R. parva has been split, but given here because work such as Baron Schwarz v. Mohrenstern’s great Monograph on the group cannot without impertinence be gainsaid except by work of equal fulness and care. The species, as well as the others of the group, is given here from his identification of specimens from Madeira). Mediterranean. (L., Jn., W.) Everywhere, including the Selvagens. Very abundant. 291. Rissoa (Setia) spadix, Watson, P. Z. S. 1878, p. 383, xxxvi. 22 (as RB. perminima, Manzoni).— Weinkauff, Conch.-Cab. 2nd ed. 1. pt. 22, p. 149, quotes my paper &c. under R. perminima, Manzoni, whose description he follows, but his figure 8, pl. xviii., is a copy of the Madeiran species now published (see Reeve, Conch. Icon, as, on my sug- gestion (l. ¢.), distinct) ; that is, the description refers to a different species from that of the figure. I adopted in the P.Z.S. paper of 1873 Dr. Gwyn Jeffreys’s identification of my Madeiran species. I am still, as then, unfortunately without any specimens of Manzoni’s Canarian species; but on fresh and very careful review I am satisfied that Manzoni’s description will not cover the Madeiran species, and Mr. Edgar A. Smith’s opinion confirmatory of my own is adverse to Dr. Jeffreys’s identification.—(Jn., W.) From Funchal to Punta de Sao Lourengo and Porto Santo. Not abundant. 312 292 293. 294. 295. 296. 297. 298. 299. REV. BR. B. WATSON ON THE . Lissoa (Alvania) spreta, 1873, Watson, P.Z.S. p. 378, xxxlv. 9. (Jn., W.) Santa Cruz, Machico, Labra, Punta de Sao Lourengo, Porto Santo; 20 to 50fms. Excessively abundant.—This species was named, described, and figured as a possible variety of &. Macandrewi, but that opinion is really not tenable. ftissoa (Onoba) striata, 1797-1800, John Adams (as Turbo), Tr. Linn. Soe. 111. 66, xii. 25, 26: Gwyn Jeffreys, B.C. ry. 37; &v. 208, lxviii. 2; Watson, P.Z. 8. 1873, p. 868 (the var. lirata, Wats. l.c. pl. xxxiv. fig.3). (Jn.,W.) Every- where, but never abundant. ; Rissoa(?) tenuisculpta, 1873, Watson, P. Z. 8. p. 889, xxxv. 28; Weinkauff, Conch.-Cab. 2nd ed. 1. pt. 22, p. 148, xviii. 6,7. Hab. From Bay of Biscay to Mediterranean, in very deep water. (W.) Funchal, Punta de Sao Lourengo. Very rare. Rissoa violacea, 1814, Desmarest, Bull. Soc. Phil. p. 8, pl. i. 8; Watson, P.Z.S. 1873, p. 378, xxxv. 14; Weinkauff, Conch.-Cab. 2nd ed. 1. pt. 22, p. 110, ii. 17-19, xvii. 4-6. Hab. From Norway to Mediterranean and Canaries. (M., Jn., W.) Porto Santo. There alone, but in abundance. —By a slip of the pen McAndrew gives this species both for the Canaries and for Madeira as 2. purpurea. Rissoa Watsoni, 1873, Schwarz v. Mohrenstern in litt. ; Watson, P. Z. 8S. 1878, p. 375, pl. xxxv. fig. 11; Weinkauff, Conch.-Cab. 2nd ed. 1. pt. 22, p. 174, xxii. 10,11. (Jn, W.) Everywhere; abundant. Roxaniella, see Atys. Saaxicava (Saxicavella) carinata, 1811, Brocchi (as Mytilus), Conch. Foss. Subappen. Ist ed. p. 585, & 2nd ed. 11. 406, xiv. 16; Searles Wood, Crag Moll. 11. 289, xxix. 5 a-e. Hab. Mediterranean, but existing from the later Tertiary period. (Jn., W.) Funchal, Cruz Point. Very local, but not rare. Saxicava (Saxicavella) plicata, 1809, Montagu (as Mytilus), Test. Brit., Suppl. p. 70: Gwyn Jeffreys, B. C. rt. 75, iii. 2; & v. li. 1 (as Panopea). Hab. From the Shetlands to the Mediterranean. (L., Jn., W-) Everywhere; very abundant. Saxicava rugosa, 1767, Linné (as Mytilus), Syst. Nat. p- 1156: Gwyn Jeffreys, B. C. 111.81, iti. 38; & v. 192, li. 3, 300. 301. 302. 303. 304. 305. 306. MARINE MOLLUSCA OF MADEIRA. 313 Hab. Nearly all seas. (M., L., Jn., W.) Everywhere; very abundant. Scalaria aspera, 1897, Watson, antea, p. 251. Scalaria clathratula, 1798, G. Adams, Micr. xiv. 19: Gwyn Jeffreys, B. C. 1v.96; & v. 210, lxxi. 5.—McAndrew gives this species from the Canaries, but not as, Dr. Jeffreys (of. ref.) says, from Madeira, nor does any other collector seem to have found it. I got only ten young specimens from deepish water at five stations along the South-east coast from Funchal to Punta de S40 Lourengo and from Porto Santo. Scalaria cochlea, 1844, G. B. Sowerby, Thes. Conch. r. pt. 4, p- 103 bis, xxxvy. 142; Dunker, Moll. Guinea, p. 18, ii. 46-48. (M.,L., Jn., W.) Patildo Mar, Funchal; Punta de Sao Lourenco; Porto Santo. Rare. Scalaria communis, 1819, Lamarck, An. s. Vert. vr. 2nd pt. p- 225; Deshayes, Encycl. Méthod. 111. 951, pl. ceccli. 3; Gwyn Jeffreys, B. C.1v. 91,11.38; & v.1xxi.8. Had. From North Norway to the Mediterranean and the Canaries. (N., W.) Two bought specimens; having these, and con- sidering the distribution of the species, I give it here on the authority of Senhor Nobre. Scalaria commutata, 1877, Monterosato, Ann. Mus. Genov. 1x. 420; Kiener, Cog. Viv. 1x. pt. 2, pl. iii. 9 (as S. mono- cycla, but not that of Lamarck); Philippi, Enum. r. 167, x. 2, 1. 145 (as S. pseudoscalaris, but not that of Brocchi) ; Sowerby, Thes. 1. pt. 4, p. 101, xxxvi. 131, 182, 184 (as S. clathrus, but not of Linné). Hab. From Normandy to Mediterranean. (L., N., Jn., W.) Porto da Cruz, Porto Santo. Neither common nor abundant.—The separation of this species from S. pseudoscalaris, Broc., is hardly justified by the absence of spines at the top of the longitudinal ribs; there is an indication, though weak, of such a development in the Madeiran form. Scalaria Fischeri, 1897, Watson, antea, p. 252. Scalaria formosissima, 1884, Jettreys, ‘ Lightning’ and ‘Porcupine’ Moll., P.Z.S. p. 140, x. 10; Dautzenberg, Moll. Agores Dragages ‘ Hirondelle,’ p. 56; Tryon, Man. Conch. 1x. 62, xvi. 34. Hab. Porcupine Bank and Acores ; North Atlantic, 340 to 1514 fms. (L.) Two specimens, without record of locality. LINN. JOURN.—ZOOLOGY, VOL. XXVI. 23 314 REV. R. B. WATSON ON THE 307. Scalaria frondosa, 1829, J. Sow. Min. Conch. vi. 149, dixxvi. 1; Searles Wood, Crag Moll. 1. 92, vii. 15; Tiberi (as S. soluta), Jour. de Conch. 1863, p. 159, vi. 3, & 1868, p. 84, v. 2.—I got this species in from 10 to 50 fms. on the whole S.E. coast of Madeira, from Funchal to Punta de Sado Lourenco, and also from Porto Santo; but the specimens were few and were all young shells. It did not occur in any other collection. Shell pure ivory-white, glossy, hunchy, that is very broad in proportion to height, with well-rounded rather depressed whorls, a broad very fine pointed spire, distant strongish oblique ribs spiralled, without an umbilicus. Sculpture— Longitudinals: each whorl is obliquely crossed by about 14 strongish but not thick, projecting reverted ribs, which with a very marked sinistral twist run continuously from whorl to whorl down the spire; each rib runs out near the suture into a well-marked but generally bluntish tooth; besides the ribs the whole surface is finely scored, not roughened, by stric, strong at the base of each rib. Spirals: the whole polished surface of the shell, including, but more faintly, the ribs themselves, is marked somewhat feebly by flatly rounded spiral threads, which, somewhat crowded on the earlier whorls, are on the later whorls parted by interspaces about as broad as the threads. Colour ivory-white, but a little translucent. Whorls probably 6 or 63, but of full-grown shells only fragments presented themselves—the largest as measured below has 43, exclu- sive of those of the embryo. Suture somewhat oblique, deeply impressed. Apex a very perfect small sharp cone of 4 complete whorls which are barely convex, with a linear suture: they are polished but microscopically cancellated by longitudinal and spiral scratches. Mouth a very little elliptical, rather small. Outer lip broadly expanded into a thin lamina, which is continued all round, leaving a kind of furrow, but no chink between it and the body-whorl. L.0:21. B. 0-11. I have described this species because the existing de- scriptions are so vague that identification is extremely difficult, as is proved by the ten synonyms quoted by Jeffreys, P. Z. 8. 1884, p. 186. My own identification ig somewhat of a guess, but if wrong can be checked from 308. 309. 310. dll. 812. 313. 314. 315. MARINE MOLLUSCA OF MADEIRA. 3l5 the description and from the presence of the living form ; and the presence of the species not only im the Mediter- ranean but in 547 fms. out in the Atlantic “ off Madeira”’ (see Jeffreys, as quoted above) may justify what is at least better than the addition of still one other erroneous synonym. Scalaria (Cirsostrema) hellenica, 1843, Forbes, Agean Invert., Brit. Assoc. Rep. p. 189; Philippi (as Rissoa coronata), Enum. Moll. Sic. 11. 172, xxii. 7. Hab. Medi- terranean. (Jn., W.) Funchal, Punta de Sao Lourengo, Porto Santo. Rather rare—Dr. Gwyn Jeffreys quotes Hoérnes as authority for this asa fossil of the Vienna basin 3 but I doubt the identification. Scalaria pulchella, 1832, Bivona, Nuov. Moll. p. 11, i. 3; Philippi, Houmeratio, 1.158, x.1; &11.145. Hab. Medi- terranean. (W.) Funchal, Santa Cruz, Canical. Rare, but fragments and young shells common. Scalaria rhips, 1897, Watson, antea, p. 250. Scalaria Schulzit (1868, but then the name alone was given by Weinkauff). 1844, Philippi (Zeits. Malakol. p. 108) gave this species, in error, as S. multistriata, Say. This error of identification on the part of Philippi, Weinkauff (after further error in the Jour. de Conch. 1862, p. 348) corrected in his Conch. d. Mittelm. 1868, vol. 11.289. Hab. Mediterranean. (Jn., W.) Funchal, Punta de Sao Lourengo, Porto Santo. Not rare.—I here give this species on the authority of Dr. Gwyn Jeffreys. I feel doubtful of the identification, but have not material for an independent opinion on the Madeiran form. Scalaria Smithit, 1897, Watson, antea, p. 253. Scalaria (Acirsa) subdecussata, 1835, Cantraine, Bull. Acad. Brux. vol. m. 888; do. Malac. Médit. pl. vi. 24. Hab. Mediterranean to Madeira. (M., L., Jn., W.) From Magdalena to Punta de Sio Lourengo and Porto Santo. Not abundant. Scalaria Turtone, 1819, Turton (as Z. Purtonis), Conch. Dic. p- 208, xxvii. 97: Gwyn Jeffreys, B.C. 1v. 8,9; & v. lxxi. 2. Hab. From Scotland to Mediterranean (not Norway). (M., L., Jn., W.) From Funchal to Punta de Sao Lourengo and Porto Santo, but not abundant. Scaphander (Weinkauffia) diaphana, 1839, Aradas and 23* 316 316. 317. 318. 319. 320. 321. 322. REV. R. B. WATSON ON THE Maggiore (as Bulla), Cat. Conch. Sicil. p. 40; Forbes (as Bulla turgidula), Agean Invert., Brit. Assoc. Rept. 1843, p- 188; Gwyn Jeffreys (as Scaphander gibbulus), Ann. & Mag. Feb. 1856, Mar. Tert. Piedm. p. 188, pl. ii. 20, 21; Vayssicre (as Weinkauffia diaphana), Jour. de Conch. 1893, p. 90, iv. 1-8. Hab. Mediterranean. (W.) From Funchal to Punta de Sio Lourenco and Porto Santo. Rather rare. Schismope depressa, 1897, Watson, antea, p. 263. Sepia officinalis, 1767, Linné, Syst. Nat. p. 1095; Gwyn Jeffreys, B. C. v. 188, vi. 8. Hab. From Norway to Mediterranean. (Jn., W.) Rare. Stphonaria (Liriola) Gussonii, 1829, O. G. Costa (as Ancylus), Oss. Is. Pant. &c. p. 20. no. 25; Philippi (as Patella pellucida), t. p.111, vii. 7, & (as P. Gussoniz) p. 255 5 & vol. 11. p. 84. Of. Dall, Jour. de Conch. 1878, p. 68, & 1879, p. 285, with many other valuable references given by the Marquis of Monterosato in his ‘ Nomenclatura,’ p. 150. Hab. California, West Indies, and Mediterranean. (M., L., Jn., W.) Everywhere abundant. Skenea planorbis, 1780, Fabricius (as Turbo), Faun. Gronl. p- 384: Gwyn Jeffreys, B. C. 1v. 65, i.4; & v. 202, lxx. 1. Hab. From Spitzbergen to Greenland and Florida and to the Mediterranean. (Jn., W.) All along the S.E. shore and Porto Santo. Very abundant. Solarium Archite, 1880, O. G. Costa, Cat. Test. Taranto, Acad. Se. 111. 40, & Fauna Napoli, p. 5, i. 1; Monterosato, Notiz. Solar. Medit. p. 10, figs. 21-23. Hab. From Bay of Biscay to St. Helena and Mediterranean. (Jn., W.) Funchal. Very rare. Solarium fallaciosum, 1872, Tiberi, Bull. Mal. Ital. v. 35; Monterosato, Solarii Medit. p. 8, figs. 12-20. (= Sol. siculum, Cantr., probably, and to S. stramineum, auct. nec Lam.) Hab. Mediterranean. (Jn., W.) Very rare.— McAndrew’s List gives this sp. (as S. straminewm) for Vigo, but not for Madeira, nor for the Canaries ; and represented so poorly as are both this species and the preceding one, they must be held as a little doubtfully indigenous to Madeira. Solarium hybridum, 1767, Linné (as Trochus), Syst. Nat. p- 1228; Lamarck, An. s. Vert. vir. 4; & 2nd ed. 1x. 99; Philippi, Enumeratio (as S. Jutewm), 1. 74, x. 127 ; Montero- sato, Solarii Medit. p. 7, figs. 10,11. Hab. Mediterranean MARINE MOLLUSCA OF MADEIRA. 317 and St. Helena. (Jn.,W.) Funchal to Punta de Sao Lourengo and Porto Santo.—Hanley (Ips. Lin. Conch. p. 315), whose opinion I forsake with extreme reluctance, so uniformly is he wise and right, regards Linné’s 8. hybridwm as the “ New Holland” species, which, though like, is quite distinct; but when Linné says expressly for S. hybridum, “habitat in mare Mediterraneo,” the other identification, even though it were by Linné himself, must be dropped. 323. Solariwn mediterranewm, 1872, Monterosato, Notiz. Foss. Mte. Pellegrino &e. p. 31, but without description, which followed in 1873, in Solar. Med. p. 6, figs. 8,9. Hab. Mediterranean. (L., Jn., W.) Magdalena from 100 fms., living; Funchal; Labra. Rare.—This isthe S. simplew of Tiberi, but not of Brown, which is a Pliocene fossil not found living in the Mediterranean. Itis the S. pseudoperspectivum of Philippi but not of Brocchi, which also is a Pliocene fossil not now alive in theMediterranean. It is the S. sul- catum of Costa but not of Lamarck, which (¢este Defrance) is a Grignon fossil=S. patulum, Lam. It is the 8. pul- chellum of Tiberi but not of Michelotti (see Trans. Royal Soc. Edin. xv. pt. 1, p. 215), a Miocene fossil. When one so endowed as Hornes (Foss. Moll. Wien, 1. 464), corroborated by the opinion of one possessing the opportunities and abilities of Monterosato, and followed by so capable a judge as Sacco (Moll. terz. Piemonte &e. pt. 12, p. 48), pronounces a definite judgment on this Solariwm as being distinct from WS. pseuwdoperspectivum of Brocchi, it would be an impertinence for an ordinary observer to set up an opposite opinion ; but the two species have very much in common. 324. Solecurtus antiquatus, 1799, Pulteney (as Solen), Cat. Dors. p- 28, iv. 5: Gwyn Jeffreys, B. C. 111. 6,1.1; & v. 190, xlvi. 2. Hab. From Great Britain to the Mediterranean and Canaries. (M.,L.,Jn.,W.) Everywhere; abundant. John- son’s specimens got at Funchal came from 30 fms. 325. Solecurtus candidus, 1804, Renieri (as Solen), Tav. Alf. Conch. Adr. p.1: Gwyn Jeffreys, B. C. m1.3; & v. 190, xlvi. 1. Hab. From Shetland to the Mediterranean and Canaries. (M., L., Jn., W.) Not abundant, but found everywhere. 326. Solemya togata, 1793, Poli (as Tellina), Test. Sicil. 11. 42, xv. 20; Philippi, Enumeratio, 1. 15, 1.17; & 11. 12 (as 318 327. 328. 329. 330. dol. 302. REY. R. B. WATSON ON THE Solenomya mediterranea, Lam.). Hab. From the Lusitanian coast to Mediterranean and Canaries. (W.) A fragment of a full-grown living specimen and some very young shells ; but there are in the British Museum Madeiran specimens of McAndrew’s dredging not enumerated in his List. Spirialis, see Limacina and Peraclis. Spirula Peroni, 1822, Lamarck, An. s. Vert. vir. 601; & 2nd ed. by Deshayes, x1. 280; Woodward, Manual, pp.13&77, pl.i.9. Hab. Oceanic. (L., W.) Canical, Porto Santo.— Linné (Syst. Nat. p. 279) called the species Nautilus spirula. By a mistake on their respective plates, but corrected in the text by both authors, de Blainville (Manuel de Mal. p- 381, pl. iv. 1 a, b) and Deshayes (Encyclop. Méthod. 11. 975, pl. eeeclxy. 5) introduced a good deal of confusion by calling the species S. australis. Spondylus Powellii, 1892, H. A. Smith, Journal of Conchol. vol. vir. p. 70; McAndrew, Brit. Assoc. Rept. 1850 passim as S. gedaropus. Hab. From Madeirato Cape Verd. (M., L., N., W.) Funchal, Punta de Sao Lourenco.—I assume that McAndrew and Nobre refer to this species under the name S. g@daropus, as certainly did both Mr. Lowe and myself till Mr. Smith brought better counsel. Tectura, see Acmea. Tellina balaustina, 1767, Linné, Syst. Nat. p. 1119: Gwyn Jefireys, B. C. 11. 371; & v. 186, xl. 38. Hab. From. Shetland to Canaries. (M., L., Jn., W.) From Magdalena to Punta de Sao Lourengo. Very abundant. Tellina balthica, 1767, Linné, Syst. Nat. p. 1120: Gwyn Jeffreys, B. C. 11. 3875, vii. 3; & v. 186, x1. 5. Hab. From Japan, Behring Sts., and N.W. America to Massachusetts ; the Black Sea and Mogador. (W.) Porto Santo. Very rare and really admissible only on the ground of its distri- bution, my Madeiran acquaintance with it’ extending only to one valve. Tellina (Oudardia) compressa, 1814, Brocchi, Conch. Foss. Subappen. 11. 823, xii. 9; Hornes, Foss. Moll. Wien, 11. 89, xiii. 6; Cantraine, Bull. Acad. Brux. 1835, 11. 398 (as T. Brocchit). Hab. N.W. coast of Africa. (L., N., Jn., W.) Everywhere; very abundant. Tellina donacina, 1767, Linné, Syst. Nat. p. 1118: Gwyn Jeffreys, B. C. 11.886; & v. 187, xli. 4. Had. Shetland to 380. 334. 330. 336. 337. 338. 339. MARINE MOLLUSCA OF MADEIRA, 319 Mediterranean and Canaries. (M., L., N. Jn., W.) Very abundant everywhere. Johnson, from 380 fms., I from 100 fms. Tellina fabula, 1781, Gronovius, Zoophyl. 1:1. 263, xviii. 9: Gwyn Jeffreys, B. C. 11.382; & v. 186, xli. 2. Hab. From North Norway to Black Sea and Cape of Good Hope. (Jn., W.) Porto Santo. Rare. At Mogador it is very common. Tellina inearnata, 1767, Linné, Syst. Nat. p.1118; Hanley, Ipsa Linn. Conch. p. 39: Gwyn Jeffreys (as T. sqgualida, Pult.), B. C. 11. 3884, xl. 8. Hab. From Scotland to Medi- terranean and Canaries. (M., L., N., Jn., W.) Everywhere ; abundant.—This is the 7. zmcarnata of Linné, Poli, Forbes & Hanley, and Romer, but not of Born nor Chemnitz nor Schroter &e. Tellina serrata, 1814, Brocchi, Conch. Foss. Subappen. p. 510, xii. 1; Romer, Conch.-Cab. 2nd ed. x. pt. 4, p. 39, xii. 1-4. A species published. by Senr. Nobre as Madeiran, but got there by no one else, would be entitled on the score of its distribution (from Portugal to Mediterranean, Canaries, and Mogador) to a place here had Senr. Nobre either got it himself or supplied information regarding it ; but an exceptional species merely forwarded by a friend as “dredged at Funchal” really lacks certification.—I have just learned at the last moment from Canon Norman that he dredged this species in Madeira this spring. Tellina tenuis, 1778, da Costa, Brit. Conch. p. 210: Gwyn Jeffreys, B. C. 11. 379; & v. 186, xl. 1. Hab. From North Norway to Mediterranean, the Black Sea, and Mogador. (Jn., W.) Machico and Porto Santo. Not rare, but very local. Teredo bipennata, 1819, Turton, Conch. Die. p. 184, figs. 38- 40; Forbes and Hanley, Brit. Moll. 1. 80, 1. 9-11: Gwyn Jeffreys, B. C. 111. 182; & v.194. Hab. From the Feroes to Mediterranean, also Vancouver Island and California. (L., Jn., W.) From Punta Delgada to Punta de Sao Lourengo and met with in floating timber. Teredo Dallii, 1897, Watson, antea, p. 266. Teredo malleolus, 1819, Turton, Conch. Dic. p. 255; also Dithyr. Brit. pl. u. 19; Forbes & Hanley, Brit. Moll.1. 84, i. 12-14: Gwyn Jeffreys, B. C. m1. 181; & v.194. Hab. From W. Indies to Western Europe. (W.) Funchal, Porto Santo. Teredo megotara, 1853, Hanley, in Forb. & Hanl. Brit. Moll. 1. 77, 1. 6, xvii. 1, 2: Gwyn Jeffreys, B. C. 11.176; & Vv. 320 340. 341. 342. 3848. 344. 345. REV. R. Be WATSON ON THE 194, liv. 4. Hab. North Atlantic on both sides from Spitz- bergen and Greenland southward. (UL., Jn., W.) From the Gorgulho and the Desertas (Bugio) to Punta de Sio Lourengo and Porto Santo. Teredo Stutchburii, 1828, de Blainville, Dic. Scien. Nat. 111. 268; G. B. Sowerby, Thes. Conch. v. 124, sp. 18, pl. cece] xix. 8; Fischer, Jour. de Conch. 1856, p. 255. Hab. Sumatra (f- Fischer, but i¢ does not appear in v. Martens’s &c. List). (W.) From Funchal to Punta de Sao Lourengo. Not common. Thracia papyracea, 1791, Poli (as Tellina), Test. Sic. 1. 43, xv. 14 & 18: Gwyn Jeffreys, B. C. 111. 36, ii. 2; & v. 191, xlviii. 4, 4%. Hab. From Iceland to Mediterranean and Canaries. (M., L., Jn., W.) From Magdalena (100 fms.) to Punta de Sao Loureng¢o and Porto Santo. Not abundant. Thracia pubescens, 1799, Pulteney (as Mya), Cat. Dors. p. 27, iv. 6: Gwyn Jeffreys, B. C. 111.388; & v. 191, xlviii. 5. Hab. From English Channel to Mediterranean. (L., Jn., W.) From Funchalto Puntade Sao Lourengo. Not abundant. Tornatina, see Utriculus. Triforis perversa, 1767, Linné (as Zrochus), Syst. Nat. p- 1231; Gwyn Jeffreys, B. C. (as Cerithium), iv. 2615 & v. 217, lxxx. 5. Hab. From Norway to Mediterranean, the Canaries, Mogador, St. Helena, and California (teste E.A. Smith). (M., L., Jn., W.) Everywhere; enormously abundant. Triptera, see Cuvieria. Triton chlorostoma, 1822, Lamarck, An.s. Vert. vit. 185; & ed. 2, Desh. 1x. 636; Quoy & Gaim. Astrol. 11. 541, xl. 16, 17; Kiener, Iconog. vol. vir. p. 19, xii. 2; Reeve, Icon. vol. 11. vill. 25; Kobelt, Conch.-Cab. 2nd ed. 111. pt. 2, p- 161, xli. 1, 2,5,6; Tryon, Manual, mr. 138, vii. 47, 48. Hab. Red Sea(?); Isle de Bourbon (teste Deshayes and also Quoy & Gaimard) ; New Caledonia (?) ; Philippines (?) ; Ventral Pacific (?); Sandwich Islands (?); West Indies, Bermuda (teste Tristram). [Vote.—Of these localities, two at least besides Madeira are trustworthy and establish a very remarkable habitat.] (., Jn., W.) Paiil do Mar, Magdalena (100fms.), Punta deSao Lourengo. Not common. Triton corrugatus, 1822, Lamarck, An. s. Vert. viz. 181 & 2nd ed. Desh. 1x. 628; Deshayes, Encycl. Méthod. 111. MARINE MOLLUSCA OF MADEIRA. > {oot 1056, cccexvi. 3 a, b; Kiener, Iconog. vol. vit. p. 14, vill. 1; Reeve, Iconog. 1. pl. v. 15. Hab. From the Bay of Biseay to Mediterranean and Grand Canary. (L., N., jase W.) From Funchal to Punta de Sao Lourengo and Porto Santo. Not rare. 346. Triton cutaceus, 1767, Linné (as Murex), Syst. Nat. p. 1217: Gwyn Jeffreys, B. C. rv. 803, v. 4; & v. 218, Ixxxiti. 4. Hab. From English Channel to Mediterranean, Mogador, and Canaries. (W.) Funchal, Porto Santo. Rare. 847. Triton nodifer, 1822, Lamarck, An. s. Vert. vi. 178; & 2nd ed. Desh. 1x. 624: Gwyn Jeffreys, B. C. 1v. 301; &v. 218, Ixxxiii. 3. Hab. From English Channel to Mediterranean. (1, N., W.) Funchal and eastwards. Not uncommon. 348. Triton olearium, 1767, Linné (as Murex), Syst. Nat. ed. 12 (not ed. 10), p.1216; Adanson(Le “ Vojet’”’), Sénégal, p. 118, viii. 12; Chemnitz, Conch.-Cab. (as Buccinum quinquan- gulare), Iv. 96, 97, exxxi. 1052, 1053, 1054, 1056, and also XI. 115, exci. 1837-8; Deshayes (as 7. succinctus), Encycl. Méthod. m1. 1057, eccexvi. 2; also in Lamarck, An. s. Vert. 2nd ed. 1x. 628 ; Kiener, Iconog. (as F sweeinctus), vol. VIII. 33, vi. 1; Reeve (7. olearium), Conch. Icon. 11. Monog. pl. ix. 32. Hab. South America, Cuba, Mediterranean, Cape of Good Hope, Australia, New Zealand, Japan, and the Pacific ; but v. Martens does not give it in the Indian Ocean. (M., L., N., Jn., W.) Funchai to Punta de Sao Lou- renco. One specimen from Porto Santo presents the exact features of T. martinianum, d'Orb., from the West Indies. Not uncommon. [Note.—The 7. olearium (Murex) of Linné’s 10th ed.= Ranella gigantea, Lam., is Murex reticu- laris, Born. | 349. Triton reticulatus, 1826, de Blainville, Faun. frang. p. 118, iv. p.5; Kiener, Iconog. vii. 26, xviii. 3; Reeve, Conch. Icon. 11. xvii. 72; Kobelt, Conch.-Cab. 2nd ed. mt. pt. 2, p- 287, lxv. 6, 7: Philippi, Enumeratio (as Ranella lanceo- lata), 1. 211; & 11. 183, xi. 28. Hab. Mediterranean aud Senegal. (L.) One specimen dredged in Labra. Very doubtful. 350. Triton tritonis, 1767, Linné (as Murex), Syst. Nat. p. 1222; Born (as Muresw), Iudex Mus. p. 315; Chemnitz (as Bue- cinum), Conch.-Cab. tv. 112, exxxiv. 1277, exxxv. 1282-3 ; Wood (as Murex), Ind. Test. pl. xxvii. 95; Fab. Columna 322 REV. R. B. WATSON ON THE (as B. variegatum), Aquat. Obs. p. 53, f. 4; Lamarck (as T. variegatum), An. s. Vert. vit. 178; & 2nd ed. Desh. rx. 623; Deshayes (as 7. variegatum), Encyel. Méthod. m1. 1054, plates vol. 111. ceeexxi. 2 a,b (the nomenclature of the plates being by Bory de St. Vincent, Aug. 1, 1824, the text by Deshayes in 1832); Philippi (as Z ¢ritonis), Enumeratio, 1. 212, & 11.183; Kiener, Iconog. (do.) vol. viii. 28, 11.; Reeve, Conch. Icon. (do.) vol. 11.1.3 & ii. 8. (L., N., W.) Funchal, Piedade. Not very uncommon. 351. Frochus (Clanculus) Bertheloti, 1839, d’Orbigny (as Mono- 352. 358. donta), Moll. Canaries, p. 81, vi. 17-20; Philippi, Conch.- Cab. 2nd ed. vol. 11. pt. 3, p. 271, xxxix. 17; Fischer in Kiener, Iconog. vol. x1. p. 295, xcv. 1. Hab. Madeira. (M., L., N., Jn., W.) From the Gorgulho to Punta de Sao Lourengo and Porto Santo. Not abundant.—For the original definition of Clanculus by de Montfort see Conch. Syst. 1.191. Adams’s definition I am unable to understand. The umbilicus in the 7. Bertheloti is a true, not a * false ” one, and though turreted like a corkscrew is perforated to the very apex. The one most prominent feature by which de Montfort differentiates the group is the umbilicus, and to alter his definition to “ perforée ou non ombiliquée ” is impossible. Trochus (Gibbula) Candei, 18389, dOrbigny, Moll. Canar. p: 82, vi. 21-23 ; Philippi, Conch.-Cab. 2nd ed. vol. 11. pt. 8, p. 227, xxxiv. 15. Hab. Canaries. (L., N., Jn., W.) Everywhere ; extremely common both in its typical form and in that of the umbilicated var. of 7. Saulcyi, of which d’Orbigny made a separate species. Trochus (Trochocochlea) colubrinus, 1849, Gould, Bost. Soc. Nat. Hist. tr. 107; Exploring Exped., Shells, p. 183, fig. 223; Watson, ‘Challenger’ Report, p. 63. Hab. Canaries and Madeira. (L., N., Jn., W.) Everywhere ; excessively common.—The name 7’. Sauciatus, Koch, as of earlier date, has been suggested for this species, but for Gould’s species there is certainty, while for Koch’s the habitat is unknown and the description—‘“ centrum” of the base “vertieft und schwarz-rothgefleckt ’"—is quite inapplicable. Does any one know what Koch’s species was? My lamented friend Dr. Fischer, whose knowledge and judgment were equally trustworthy, regarded (see 354, 355. 356. 357. 358. 309. 360. MARINE MOLLUSCA OF MADEIRA. 328 Kiener, Iconog. p. 180) it as a variety of the variable and widespread 7. sagittiferus, Lam. That is where I would place 7. colubrinus ; but abandoning Gould’s species, of which we are certain, we drift into the mare magnum of mere opinion, and dealing here with the local fauna alone, I feel it best to rest on Gould’s nomenclature. Trochus (Ziziphinus) conulus, 1767, Linné, Syst. Nat. p- 1280; Philippi, Conch.-Cab. 2nd ed. 11. pt. 3, p. 64, xiii. 8, 9; Fischer in Kiener’s Iconog. vol. xr. p. 121, xl. 1, xlix. 1. Hab. From the Acores to Mediterranean and Canaries. (M., L., N., Jn., W.) Pretty common. Trochus (Ziziphinus) exasperatus, 1777, Pennant, Brit. Zool. Iv. 126: Gwyn Jeffreys, B. C. 111. 324; & v. 208, Ixii. 3. Hab. From Scotland to Black Sea and Canaries. (M., L., N. Jn., W.) Everywhere; very abundant. Trochus (Ziziphinus) granulatus, 1778, Born, Index Mus. p- 337, xii. 9,10: Gwyn Jeffreys, B. C. 111. 327; & v. 204, lxii. 5. Hab. From the extreme S.W. of Scotland to Mediterranean and Canaries. (M., L., Jn., W.) From Funchal to Labra. Not uncommon. Trochus (Gibbula) magus, 1767, Linné, Syst. Nat. p. 1228: Gwyn Jeffreys, B. C. 111. 805; & v. 203, Ix. 1. Hab. From S.W. Sweden to Mediterranean and Canaries. (M.,L.,N., Jn., W.) From Funchalto Punta de Sao Lourenco. Abundant. Trochus (Ziziphinus) striatus, 1767, Linné, Syst. Nat. p- 1280; Hanley, Ipsa Lin. Conch. p. 821, pl. v.7: Gwyn Jeffreys, B. C. 11. 322; & v. 203, Ixii. 2. Hab. From Scotland to Mediterranean and Canaries. (M., L., N., Jn., W.) Everywhere ; very abundant. Lrochus (Ziziphinus) zizyphinus, 1767, Linné, Syst. Nat. p- 1281: Gwyn Jeffreys, B. C. 111. 330, vu. 4; & Vv. 204, Ixili. 6. Hab. From South Norway to Mediterranean and Canaries. (M., L., N., Jn., W.) Everywhere; pretiy abundant. Trophon fusulus, 1814, Broechi (as Murex), Conch. Foss. Subap. Ist ed. p. 209, & 2nd ed. 11. 199, viii. 9; Libassi (as Murex Spade), 1859, Conch. Foss. Palermo, p. 43, i. 29; Bellardi (as Pollia), Moll. terz. Pied. 1.169, xii. 4; Watson (as Murex), ‘Challenger’ Rep. p. 160; Kobelt (as Zrophon), Jahrb. 1887, p. 120, v. 1. In existence from the Upper Miocene onwards. Hab. From the Bay of Biscay and 324 361. 362. 363. 364. REV. BR. B. WATSON ON THE Agores to Mediterranean. (U., Jnu., W.) Labra &e. Not rare.—The dentition of this molluse excludes it, as Kobelt (2. c.) shows, from the genus Pollia as well as from Murex. With Trophon it is further connected by its longish, nar- rowish, bluntly-pointed, oval operculum, which, when seen in its place within the upturned mouth of the shell, has its apex in front towards the point of the mouth, a little ineurved towards the pillar-lip, along which the later edge-layers of growth successively lie. In this connection it should be ob- served that in Adams’s ‘Genera’ the opercula, when diversely ended, are turned upside down, and the consequent reversal of right and left side is sometimes corrected by the figure presenting the internal, not the external surface. This mistake is avoided in Woodward’s most accurate ‘ Manual,’ but it is widespread and sometimes very perplexing, even Philippi has not escaped it. Dr. Brot is nearly always right, as are also Dr. Kobelt and Dr. Fischer, though in the ‘Manual’ of the latter the different position of shells and opercula adopted in the “gravures” and in the “planches” is somewhat confusing. Trophon Lowei, 1897, Watson, supra, p. 244. Truncatella subcylindrica, 1767, Linné (as Helix), Syst. Nat. p. 1248. no. 696 (see Hanley, Ips. Lin. Conch. p- 579); Draparnaud (as Cyclostoma truncatulum), Moll. p- 40, 1. 28-31; Risso (as J. truncatula), Hist. rv. 125, fig. 5; Lowe (as 7. truncatula), Zool. Journ. v. 280, xiii. 13-18; Philippi (as 2zssoa), Enumeratio, 1. 151. 1, & (as Truncatella) vol. 11. 183, xxiv. 3; Deshayes (as Truncatella truncatula) in Lamarck, An. s. Vert. 2nd ed. vir. 362 (note) ; Pfeiffer, Monog. Auric. pp. 186 & 188: Gwyn Jeffreys, B. C. 1v. 85, ii. 2; & v. 209, Ixxi.1. Hab. From Scotland to Mediterranean and Canaries. (L., Jn., W.) Wherever the shore is shingly, even at the Selvagens. Truncatella Lowet, 1852, Shuttleworth, Diag. New Moll. p. 12 ; Pfeiffer, Monog. Poeumonopomorum, Suppl. 1, p. 7, do. 2, p. 5, do. 3, p 11, no. 145 abs Menenne: (W.) Madeira. Turbo and Turbonilla, see Odostomia. Turbo (Bolina) rugosus, 1767, Linné, Syst. Nat. p. 1234; Fischer in Kiener, Iconog. vol. x1. p. 41, xv. 1. Had. From Bay of Biscay to Mediterranean, Mogador, and 365. 366. 367. MARINE MOLLUSCA OF MADEIRA. 825 Canaries. (M.,L., N., Jn., W.) Everywhere very abundant. The subgeneric name is given above as Rafinesque (Ann. Nat. p. 144) wrote it in 1815. Risso, following, also wrote it thus in 1826. The form Bolma was a mistake of Gray. Turritella terebra, 1767, Linné (as Turbo), Syst. Nat. p.1239: Gwyn Jeffreys, B. C. rv. 80, ii. 1; & v. 209, Ixx. 6-11. Hab. From the Lofotens to the Mediterranean. (W.) Funchal ; two specimens.—I give this species with a good deal of hesitation, but I cannot admit either 7. triplicata, Broc., or T. bieingulata, Lam., both of which have been brought to me as Madeiran: the former is very common at the Canaries. Tylodina citrina, 1833-4, Joannis in Guérin’s Mag. de Zool. 1. pl. xxxvi. 1; Vayssiére, Moll. Opisth. Marseille, Ann. Musée, 1885, p. 151, v. 130-136. Hab. Mediterranean to St. Helena. (Jn., W.) Gorgulho, Funchal, Punta de Sao Lourencgo. A good many young shells. One full-grown specimen I secured in a rock-pool near the Gorgulho fort, west of Funchal, after watching it for a considerable time. As Philippi remarked, its rapid movements are extremely unlike those of Patella. The broad membranaceous edge of the shell flaps about in the moving water like a light cloak in a breeze. The bright yellow colour of the animal and of the integument of the shell (which suggested its very graphic name) turns to deep brown in drying or even when preserved in spirit. Rubbed specimens are white with a yellow tint. Adams’s (Genera, 11. 42, Ixi. 4) description and figure are both inadequate and incorrect. Vayssicre’s. description in all its details, as well as his figure of the animal and of the egg-ribbon, are admirable ; although the last when I examined it, freshly deposited, seemed flatter than is suggested by his term “ demi-cylindrique.”’ Tylodina Rafinesquit, 1836, Philippi, Enumeratio, 1. 114, vii. 8,&11.89; Cantraine, Mal. Méd. p.94. Hab. Mediterranean. (W.) One specimen. —The form of the shell in this species is so distinct from that in Z. citrina,that I find it impossible to accept Vayssicre’s suggestion and unite the two. I do not gather from his remarks that he has ever seen Philippi’s species, whose figure, however, markedly shows the dif- ference. Tornatina, see Utriculus. Among the shells of Mr, Johnson’s collection there occurred a specimen 326 — 368. 369. 370. 371. 372. 373. 374. REV. R. B. WATSON ON THE of Ungulina oblonga, Lam. An. s. Vert. v. 487, & 2nd ed. vi. 122, also Sowerby’s Genera, pl. xlv. ; “from Porto Santo,” but as he had marked the specimen “ very doubtful,” I have not put it in the List. Utriculus mammillatus, Phil. (as Bulla).—I1 exclude this species, McAndrew does not give it from Madeira, and in his own copy of his Brit. Assoc. List for the Canaries he corrects and initials the correction from Cylichna mammillata to C. truncata (which = U. t¢runcatulus), Dr. Gwyn Jeffreys has indeed published it (B. C. v. 223) as Madeiran on the strength of shells I sent him, but the identification of these as U. mammillatus was not mine. Utriculus nitidulus, 1846, Lovén (as Cylichna), Index Moll. Scand. p.16: Gwyn Jeffreys (as Cylichna), B. C. rv. 412; & V. 222, xciii. 2; G.O. Sars, Moll. Norv. p. 286, xvii. 19 & xxvi. 38. Hab. From north of Norway to Mediterranean. (W.) Three specimens. Utriculus tornatus, 1883, Watson, Moll. ‘ Challenger,’ Linn. Journ., Zool. vol. xvi1. 835; ‘ Challenger’ Mollusca Report, xv. 651, xlviii. 10. Hab. Tenerife and Madeira. (W.) Everywhere in great numbers. Utriculus truncatulus, 1792, Bruguiere (as Bulla), Encycl. Méthodique, 1. 877. no. 10: Gwyn Jeffreys, B. C. 1v. 421; & V. 223, xciv. 2: G. O. Sars, Moll. Norv. p. 285, xvii. 18, xxvi. 2. Hab. From North Norway to Mediterranean and Canaries. (Jn., W.) Funchal, Porto da Cruz, Porto Santo, Rather abundant; the var. pellucida, Brown, Ill. p. 4, xix. 45, 46, mixed up with the type form. Umbrella mediterranea, 1819, Lamarck, An. s. Vert. vt. 1st part, p. 343, 2nd ed. Desh. vit. 574 ; Delle Chiaje, Mem. trv. 200. no. a. Ixix. 5 & 19; Delessert, pl. xxii. 12: Philippi, Enumeratio, 1. 118, vii. 11; & 1. 88: Vayssiére, Moll. Opisthobr. p. 183-4 &c. pl. vi. 187-150. Hab. Mediterranean, Canaries, and St. Helena (?). (L., W.) Gorgulho, Funchal, Punta de Sao Lourengo, 380 to 40 fims., and Canical shore. Rare. Venerupis irus, 1767, Linné (as Donazx), Syst. Nat. p. 1128: Gwyn Jeffreys, B. C. 111.86, ii. 4; & v. li. 5. Hab. From South England to Mediterranean, Mogador, and Canaries. (M., L. N., Jn., W.) Everywhere; abundant. Venus casina, 1767, Linné, Syst. Nat. p. 1180: Gwyn Jeffreys, B. C. 1. 337; & v. 184, xxxviil. 5. Hab. From Norway to Mediterranean and Canaries. (M., L., N., Jn., W.) Everywhere; very abundant. Venus (Cytherea) chione, 1767, Linné, Syst. Nat. pe 1131: 375. 376. 377. 378. 379. 380. 381. 382. MARINE MOLLUSCA OF MADEIRA. S2n Gwyn Jeffreys, B. C. 1. 332; & v. 184, xxxviii. 3. Hab. From South England and Ireland to Mediterranean. (M., L., N., Jn., W.) Everywhere; very abundant. Venus effossa, 1836, Bivona in Philippi’s Enumeratio, tr. 43, ii. 20; Pfeiffer, Conch.-Cab. 2nd ed. x1. 197, xxxii. 1-4; Weinkauff, Conch. Mittelm. 1.115. Had. Mediterranean and St. Helena. (L., Jn., W.) Magdalena, Punta de Sao Lourengo, Porto Santo. Not rare, but very local. Venus fasciata, 1778, da Costa (as Pectunculus), Brit. Conch. p- 188, xii. 3: Gwyn Jeffreys, B. C. 11.334, vi.5; & v. 184, xxxvil. 4. Hab. From the North Cape, Norway, to the Aigean, and existing since the later Tertiary period. (N.) For its right of citizenship in Madeira, see Watson, Journ. of Conch. 1890, pp. 374 & 376. Venus (Cythreea) rudis, 1791, Poli, Test. Sicil. 1. 94. 15, 16; Philippi (as Cytherea venetiana), Enum. t. 40, iv. 8; & 1.32. Hab. From Bay of Biscay and Mediterranean to Black Sea, Canaries, and St. Helena. (L., Jn., W.) Every- where; very abundant. Venus verrucosa, 1767, Linné, Syst. Nat. p. 1130: Gwyn Jeffreys, B. C. 11. 389; & v. 186, xxxviii. 6. Hab. From Scotland to the Mediterranean, the Cape of Good Hope, and the Indian Ocean. Hxtant since the later Tertiary period. (M., L., Jn., W.) Gorgulho to Punta de Sio Lou- renco, and Porto Santo. Abundant. Vermetus gigas, 1832, Bivona, Effem. Scient. & Lett. p. 5, ii. 1, 2; Philippi, Enumeratio, 1. 170, ix. 18 a & 6; Mon- terosato, Monog. Vermet. Medit. p. 30, iii. 1-8. Hab. Mediterranean. (L., Jn., W.) Madeira. Vermetus rugulosus, 1878, Monterosato, Enumeraz. & Synon. p. 29; id. Monogr. Vermet. Medit. p. 22, pl. i. 8 Hab. Mediterranean. (Jn., W.) Madeira. Vermetus triqueter, 1832, Bivona, Effem. Scient. & Lett. p- 6; Philippi, Enumeratio, 1. 170, ix. 21; Monterosato, Vermet. Medit. p. 26, 11.4-9. Hab. Mediterranean. (L., Jn., W.) Madeira. Weinkauffia, see Scaphander. Xylophaga dorsalis, 1819, Turton (as Pholas, but in 1822 amended to Xylophaga), Conch. Dic. p. 185: Gwyn Jeffreys, Brit. Conch. 111. 120, iv. 3; & v. 198, lili. 4. Had. From North Norway to Mediterranean. (W.) Funchal, Cruz Point, Punta de Sao Lourenco. Abundant. 828 REV. R. B. WATSON ON THE The following species, of which I dredged shells in Madeira, have no claim to be included in the above. list :-— Off Funchal, even to 50 fms. Madeiran Land-shells. Achatina acicula. Helix pulchella. Limneea truncatula. Planorbis glaber. Pupa anconostoma. Foreign Land-shells not found in Madeira. Assiminea Grayana. Bythinia tentaculata. Limnea peregra. Neritina fluviatilis —A species also obtained by Johnson. Pisidium sp. Planorbis carinatus. intermedius=submarginatus. sp. Valvata piscinalis. - Off Santa Cruz, 15 to 20 fms. Achatina acicula. Ancylus fluviatilis. Craspedopoma Lyonnettianum. Helix cellaria. cristallina. Helix madeirensis. pulchella. pusilla. vulgata. Hydrobia similis. Limnea truncatula. Pupa anconostoma, Off Machico, 15 to 20 fms. Ancylus fluviatilis. Helix vulgata. Hydrobia similis. Limnza truncatula, Pupa anconostoma. Of Canigal, 20 fms. Achatina acicula. Bulimus ventricosus, Helix compacta. paupercula. pisana. Pupa sp. Off Punta de Sto Lourengo. Clausilia exigua. Helix compacta. paupercula, Marine species dredged by me or brought to me as Madeiran, but which I reject :— Actzxon tornatilis. Arca incongrua. lactea. Bouchardia rosea. Bulla striata. Cardium edule. magnum. —— medium. Cerithium vulgatum. Conus mediterraneus. Corbula gibba. Crania anomala. Crenella discors. Crepidula gibbosa. Cypreea lynx. Eulima Staloi.—See Jeffreys, P. Z. 8. 1884, p. 368. Hydrobia ulvee. Litorina litorea. obtusata. Lucina Adansonii. Modiolaria discors, Murex brandaris. Nucula nucleus. Purpura lapillus. Rissoa Montagu. Terebra favat. Trochus Eltonz. umbilicatus. Turritella bicingulata. triplicata. Venus cancellata. exspinata. gallina. —— striatula. Such a list as is here presented suggests the caution which should be exercised in working out the Molluscan fauna of this island. _ Journ. Zoor Vor. XXVI Je, II) 20. J.Green del. ethth. Mantern Bros .imp. NEW MOLLUSCA FROM MADEIRA. Lisw.Soc.Journn.Zoon. Vou. XXV1.Pxn.20. Watson. 32Cc. Mintern Bros. imp. J.Green del.et lith. NEW MOLLUSCA FROM MADEIRA MARINE MOLLUSCA OF MADEIRA. 329 EXPLANATION OF THE PLATES. [The numbers beneath the figures correspond to those of the order of description in the text and of enumeration in the list on p. 233,] Prats 19. Fig. 1. Cylichna spreta. 2. Amphisphyra flava. 3. Philine complanata. 4. Philine trachyostraca. 4a. The same. Small portion of shell, highly mag- nified, showing sculpture. 5. Philine desmotis. 5a. The same. Small portion of shell, highly mag- nified, showing sculpture. 7, 7a, 76. Doridium maderense. 8. Pleurobranchus Dautzenberge. 9. Pleurobranchus Lowet. 10. Nassa antiquata. 11. Murex (Ocinebra) medicago. 12. Trophon Lowei. 13. Bittiwm depauperatum. 14, 14a. Bittium incite. 15, Cecum atlantidis. 16. Natica (Nacea) furva. 17. Scalaria rhips. 18. Scalaria aspera. 19. Scalaria Fischeri. 20. Scalaria Smithit. Puate 20. . Aclis vitrea. . Aclis trilineata. . Aclis tricarinata. . Lulima fulva. . Hulima sordida, . Kulima badia. Kulima rhaphium. . Hulima trunca. . Hulima inconspicua, 30. Odostomia omphaloessa. 31. Odostomia (Turbonilla) undata. 32!a, 326, 32.c, Schismope depressa. 33. Montacuta triangularis. 34. Coralliophaga Johnsoni. 35. Teredo Dallii. Fig. bo bo he) = bm wo wy wb SSISaALS LINN. JOURN.—ZOOLOGY, VOL. XXVI. 24: 2 330 CAPT. F. W. HUTTON ON THE PROBLEM OF UTILITY. The Problem of Utility. By Captain F. W. Hurroy, F.R.S. (Communicated by Atrrep Newron, F.L.S.) [Read 6th May, 1897. ] Dr. A. R. Wattacek has lately published in the Journal of this Society * a paper with the above title, in which he comes to the conclusion that, “ whether we can discover their use or no, there is an overwhelming probability in favour of the statement that every truly specific character is or has been useful; or, if not useful, is strictly correlated with such a character.’ I quite admit the force of his argument and the weight of his facts. I acknowledge that many specific characters have been shown to have had a utilitarian origin ; but it seems to me that Dr. Wallace’s conclusion will not be justified until it has been proved that there is an overwhelming probability that no agency exists capable of producing non-utilitarian characters. This, I venture to think, he has not done; and I should like to be allowed to give my reasons for so thinking. Dr. Wallace truly says it is very difficult to determine posi- tively that any one of the characters now presented by each organism “zs not, nor ever has been, useful to its possessor.” But also it is often very difficult to determine positively that a cha- racter zs, or has been, useful. These things are generally matters of inference only, and in studying them we should try to make the problem as simple as possible. Now what are the tests by which we recognize a character as of utilitarian origin? Dr. Wallace says that any new species formed under Natural Selection must exhibit either some difference of structure or function adapting it to new conditions, or some distinction of colour, form, or peculiar ornament serving as distinctive recognition-marks ; and they may, in addition, have other characters which are correlated with a useful one. The evidence for this last group of characters is very slight and uncertain, and they cannot be very numerous. At any rate each asserted case would require separate proof, and the principle of correlation should not be used asa cloak for our ignorance. The tests, therefore, are virtually reduced to two: adaptation and recognition-marks. Recognition-marks can be useful only among those animals * Journ. Linn. Soc., Zool. vol. xxv. p. 481. CAPT. F. W. HUTTON ON THE PROBLEM OF UTILITY. 3381 which are capable of recognizing them by the senses. Colour can only be recognized by sight: consequently colours in animals without eyes are not recognition-marks. But in some groups of blind animals colour constitutes a specific character, as in the shells of Lamellibranchs ; in these cases, therefore, it must be either adaptive or non-utilitarian. Sculpture might possibly be recognized by touch ; but we cannot suppose that in the Mollusca the sexes recognize each other in that way, although ornament, and slight differences in shape of the shell, constitute their chief specific characters. Even with animals possessing eyes there are some specific characters which cannot be regarded as recognition- marks, for they cannot be seen; as, for example, the teeth on the radula of Gastropods. The venation of the wings in Lepido- ptera and Trichoptera is obscured by scales or hairs, and yet it often furnishes good generic and sometimes specific characters ; occasionally even the venation differs in the two sexes. Some crabs are always covered with seaweeds, and the species cannot be ascertained until these seaweeds have been removed. And, generally, obscure characters cannot be explained as recognition- marks where there are conspicuous characters to answer that purpose. Many species of Orthopterous insects differ from each other in the number or position of the spines on the legs, and no one will suppose that the male of one of these insects stops to count the number of spines on the legs of a female before making love to her. These specific characters, therefore, are not recog- nition-marks. Are they adaptations ? Adaptations are of two kinds: those which are useful to their possessor, and those which have been useful to former ancestors. We can eliminate the last group by taking only the specific cha- racters of a species the habits of which agree with those of other species of the genus to which it belongs; for in these cases the habits must have remained the same during the whole of its specific life, and the specific characters must, ex hypothes?, have been developed by their present possessors. Now can we suppose that the colours which distinguish the shells of the different species of Tellina, which live in sand, have any adaptive value ? Can we suppose that a spine more or less, or a different arrangement of the tubercles, on the carapace of a crab has any adaptive value? Can it matter in the struggle for life whether a vein in the wing of an insect branches once or twice; or can slight differences in the number or position of the spines on the DAK a 332 CAPT. F. W. HUTTON ON THE PROBLEM OF UTILITY. legs give one insect an advantage over another? Again, can we suppose the slight differences in the number or shape of the teeth in the species of Helices, or whether they have ten or twenty ribs to the tenth of an inch on their shells, have any adaptive value? Yet they are often good and constant specific characters. Or take the shape of the spicules in the Holo- thurians or in the Sponges; or the skeleton in the Radiolarians, or the small differences in the leaves of ferns and mosses; or the various ornamentation on the frustules of Diatoms: can all or any of these characters—which are certainly as stable as specific characters which are acknowledged to be useful— can any of them be explained by the principle of utility? If such is the case, which, among two or more species living together, is the best adapted to the conditions, and which the worst? And why has not the worst died out? ‘Take, for example, the different colours and shapes of the shells of Mussels (Mytilus), two or more species of which often live together under exactly the same conditions: if one colour or shape is more advantageous than the others, why are the others there? We cannot plead want of time; for many of these species date back to the Pliocene period. Suggestions that it might be this, or it might be that, have no weight with me when T find that the effects which ought to have been produced, if these suggestions were correct, have not been produced: when, for instance, I find that the two species which are compared are equally numerous. T could give numerous cases in detail if I thought it necessary ; but, no doubt, every naturalist who has done systematic work on species could furnish them by the score. So I will restrict myself to one example drawn from a class of facts different from those which I have already mentioned. There is a genus of small Fruit-Pigeons, called Péilopus, found from the Malay Peninsula through the Malay Archipelago to Australia and Polynesia. Itis a large and dominant genus, divided into twelve sections. One of these sections— Ptilopus (restricted) —contains twenty-three species, of which no fewer than thirteen * * P. pelewensis, in the Pelew Islands; P. rosetcapillus, in the Ladrones ; P. ponapensis, in Ruck Island, Caroline Group; P. Hernsheimi, in Kushai, Caroline Group; P. Richardsi, in Ugi, Solomon Group; P. Greyi, in New Caledonia and the New Hebrides; P. rarotongensis, in Rarotonga, Cook’s Group ; P. purpuratus, in Tahiti, Society Islands; P. chrysogaster, in Huahine CAPT. F. W. HUTTON ON THE PROBLEM OF UTILITY. 333 are found isolated from other species of the section, each on its own island or small group of islands. Nowit is highly improbable that all these thirteen species were developed on other islands on which they are no longer found; or that other species of the section Ptilopus formerly lived on each of these thirteen islands and have become extinct on them. If it had been a single case only, we might have had some doubt; but when it comes to having to apply the same explanation to more than half of the whole number of species, it seems to me to be absurd. Conse- quently it appears almost certain that most of these species were developed each on its own island; and, this being allowed, we have the problem of the origin of their specific characters reduced to its simplest form. li these species originated in the islands in which they are found, the colours which distinguish them cannot be recognition- marks, because there is no other species in each island with which they could be confounded. The colours cannot be due to cor- relation, because they are the only characters which have changed. They cannot have been useful to ancestors, because they have only lately been developed. And we cannot suppose that they give any special advantage in each island, because all the islands have practically the same climate and the same flora and fauna. This exhausts the resources of the principle of utility, and we are driven to the conclusion that these specific characters have a non- utilitarian origin; and yet they are found “in every individual constituting the species, neither more nor less.” I therefore submit that, whether we can discover the cause or not, there is an overwhelming probability in favour of the state- ment that these truly specific characters have had a non-utilitarian origin; and if the specific colours have not had a utilitarian origin in these cases, it is quite probable that they may not have had a utilitarian origin in other cases where two or more species of the section are found in the same island; and it follows that recognition-marks do not necessarily arise through selection. Indeed, there appears to be no reason why even adaptations should not arise by the same non-utilitarian process ; although, and Raiatea, Society Islands; P. coralensis, in Carlsoff Island, Paumotu Group; P. Smithsonianus, some island, Paumotu Group; P. Tristrami, in Hivaoa, Marquesas Group; and P. Huttond, in Rapa, 8.E. of the Austral Islands. Further knowledge is more likely to increase than to diminish this list. 384 MESSRS. T. R. JONES AND F. CHAPMAN when they had attained a selection value, they would be rapidly increased by the principle of utility. This is exactly the opposite of the conclusion arrived at by Dr. Wallace, except that it applies only to a certain number of characters, while Dr. Wallace would make his conclusions apply to “ every truly specific character.” Ifhe would change “ every ” into “‘many a,’ I should no longer object to it. The discovery of the principle of Natural Selection by Mr. Darwin and Dr. Wallace marked a great advance in Biology; but it cannot be regarded as final. There is yet much to be learnt; and it seems to me far more probable that there are undiscovered causes at work, than that all the apparently useless characters which I have mentioned are due to the principle of utility. Just as gravitation does not express the whole of the relations between inorganic bodies, so, probably, selection does not express the whole of the relations between orgunic bodies. On the Fistulose Polymorphine, and on the Genus Ramulina. By T. Rurert Jonus, F.RS., F.G.S., and F. Caapman, A.L.S., F.R.M.S. Parr II *.—The Genus Ramulina. [Read 3rd June, 1897.] CONTENTS. Page 1, General Remarks on the genus Ramulina, Rupert Jones, 1875 ... 834 S2:sGenenic! Characters of Mamulinan cacseceecns is teenes see ce eee eEe Eee 307 Gb ap SPECLESIOL A 2NULING Leave concen eenersecnaanceencer east see the eee eReee 339-353 Is aio Ua OMNES: LEIS D4 4 e Setccnoscosaounndocsdboboododéusbooubs 339 Dh, Like (MOORE, VexeNahy, UMESL DWP} CA ooonbonodéapauoonsesacnosone 340 3. aculeata, Wiricht, | Bigs 23—42)) Sa decsn.cescee a teeeeeeeeeeee 345 4, R. Grimaldit, Schlumberger. Figs. 43, 44 ............00cceoees 350 5. Rvcervicorms, Chapman. Figs, 45-51 1.3... eseesceen eee 351 Sai, Ogawa OF JKCHOONIWOG 5.06cadacondcacdonngosadvascosbobocebSecodaboooeeee 353 § 5. Distribution and Range in Time of Ramulingd .........cccccceceeeeeee 353 §1. General Remarks on the Genus Ramulina. THis genus was instituted in 1875 T on the peculiar characters of some small tubular fossils having enlargements at the junction * For Part I. see the Linnean Society’s Journal, Zoology, vol. xxv. (1896) pp. 496-516. + Proc. Belfast Naturalists’ Field Club, vol. i. 1870-86, Appendix III.1875, p. 88, pl. iii, figs. 19, 20. ON THE GENUS RAMULINA. 330 of branches and at intervals along their length. The swollen portions resembled the somewhat loosely-set or distant segments of d’Orbigny’s Dentalina aculeata; and this latter was regarded as belonging to the new genus. Though there is room for doubt as to the generic standing of d’Orbigny’s species above men- tioned, yet some aculeate Dentaline so neariy approach it in form that it may be allowed to remain as such *. At all events, since 1875 numerous portions of so-called famuline have been figured and described; and others have been discovered, but not yet published; and both as to the characters of separate segments, and of tubes and parts of tubes, simple or branching, they cannot be regarded as parts of auy fistulose Polymorphine yet known f. Though the latter have more or less ramified outgrowths, these do not radiate from separate subglobular chambersas in Ramulina; and the tubules of the latter are very fragile. We may note that d’Orbigny alludes to the fragility of his Dentalina aculeata, which we formerly assigned to Ramulina. It is noticeable that Ramuline “are numerous in the St. Erth clay, although the tubulose Polymorphine are but sparingly represented ” (F. W. Millett, Trans. Roy. Soc. Cornwall, vol. xi. * Messrs. Balkwill and Millett in their paper on ‘The Foraminifera of Galway,” 1884, p. 15, state that a ‘‘ Ramuline” specimen from Chalk-marl of Charing, Kent, labelled “‘ Dentalina aculeata (d’Orb.),’ was obtained from a London dealer “more than twenty years ago”—about 1860? They also observe, “it is perhaps worthy of notice that many of the Ramuline, in form and texture, resemble the cervicorn outgrowths of certain Polymorphine.” t In the case of the Polymorphina proteus of Beissel, the initial series resembles Polymorphina externally, but its internal arrangement is on the plan exhibited by some of the bulbous portions of the ordinary forms of Ramulina. The figs. 9 & 10 in pl. xi1. represent the two sides of one specimen, which has numerous short tubules set irregularly about the surface. There are faint indications of rugosity. Fig. 11 is the outside, and fig. 12 the inside of a specimen haying projections mostly on one side, and these have a cavernous junction with the body. The latter is slightly, and the projections are distinctly, rugose or aculeate. Fig. 13 is a ramuline trifid tube, swollen at the junction of the three parts; aculeate with strong, short, scattered prickles. Fig. 14 shows the section of an irregularly surfaced specimen, with hollow bases in the projections. Fig. 15 is the section of a smooth subspherical specimen with irregular septa. Fig. 16 is subspherical and smooth, with a low rough growth at the apex. The ramuline tube, fig. 13, is grouped by the author as being of the same species as the others ; and is of similar formation to the projections in fig. 12, in which the cavernous junction of the outgrowth with the rest o¢ the test is characteristic.—“ Foram, Aachener Kreide,” 1891. 336 MESSRS. T. R. JONES AND F. CHAPMAN 1895, p. 658). Onthe contrary, these Polymorphine are plentiful in the Crag of Suffolk; but Ramuline are absent there. Further, more or less perfect and characteristic specimens, having a definite commencement and a great tubular extension, have been met with. These, though markedly distinct from Polymorphine, have nevertheless their relatively small initial chambers constructed on a Polymorphine plan, just as Articulina has a Milioline, Frondicularia a Nodosarian or Polymorphine, Flabellina a Cristellarian commencement, and so on. This feature, of course, indicates that Ramulina is allied to Polymorphina (perhaps subgenerically, as Articulina is to Milio- lina, &c.); and yet another feature, namely the presence of a somewhat Polymorphine and valvular septation in the swollen parts of the branching tubes, also shows that Ramulina has some- what of the nature of Polymorphina, but without identifying itself with the latter. This structure, found by us in the nodular or inflated parts of Ramulina, shows itself in Beissel’s figures of ‘* Polymorphina proteus”; and we refer Beissel’s form to Ramulina, although its exterior is strikingly like that of Poly- morphina gibba var. diffusa *. In some of Beissel’s figures of ‘“ Polymorphina proteus,” notably pl. xu. figs. 9 to 16, we have a Polymorphina-like test giving rise to a Ramulina, prickly on the surface, and in every way comparable with those already known from various Cretaceous deposits. At first sight the figures 9-11 and 16 might appear to belong to an ordinary fistulose Polymorphina; but examined in the manner shown in figs. 12, 14, & 15, which are half-sections of the test, showing the interior, the subdivision into chambers is seen to differ materially from that of a typical Polymorphina, and agrees in many striking particulars with the mode of septation shown in the bulbous or swollen portions of Ramulina, as ex- emplified in thin sections of the bulbs of the latter genus from the Chalk-marl, investigated during the writing of the present memoir. We note that the late Dr. H. J. Carter has described a sub- cylindrical “branched form of the apertural prolongation from the summit of Carpenteria monticularis,’ beginning with a fork of two branches, of which the remaining one divides afterwards into three (Aun. & Mag. Nat. Hist. ser. 4, vol. xx. 1877, p. 68, * See previous part of this Memoir, pp. 505 & 516. ON THE GENUS RAMULINA. Sol woodcut). He also described and figured the dendriform growth of another specimen (op. cit. vol. xix. 1877, p. 212, pl. xiii. fig. 9 ). The possible existence of delicate fistulose growths in other Carpenterie and in Polytrema was alluded to by this careful observer in the first-mentioned paper *. These may possibly come into the category of simple fistulose growths, like those of Polymorphina and Cristellaria already described; but the branching-tube of the Carpenteria differs, by its freedom, length, and special pattern, from those of Poly- morphina. If this be so, we may extend our foregoing proposi- tion that Ramulina has a definite initial structure analogous to, or comparable with, a certain Foraminiferal type, and suggest that, besides Polymorphina, some forms of Carpenteria and other genera may supply the structural commencement of distinct kinds of such ramifying growths. Leaving the Dentalina aculeata of dOrbigny as a Nodosarian (so regarded by Brady, Millett, and others), we have to'treat of true aculeate Ramuling. It has already been remarked by one of ust, that it appears desirable to associate the tubular and sometimes globular-chambered Ramuling, whether smooth or prickly, from Cretaceous strata, with certain recent Foraminifera found in moderately deep water, and figured by H. B. Brady. A careful consideration of the Cretaceous forms of Ramulina has since led us to believe that there are four types present in that formation. Que of these is a thin-shelled setose form, repre- sented by #. globulifera; the second, a coarser form as regards the test-structure, having stronger prickles, we propose to call Rk. aculeata (after J. Wright); the third type, having smooth- surfaced stolon-tubes and bulbs, is referred to as R. levis; the fourth is &. cervicornis of the Gault; and a good recent form is the fifth, namely &. Grimaldii, Schlumberger. §2. Generic Characters of Ramulina. Test free or attached ; branching; consisting of a calcareous tube, swollen at intervals, so as to form more or less definite, often irregular segments, from which lateral stolons or branches * Dr. A. Goés, in 1882, suggested that the Ramuline may be only parts of aulostomatous Nodosarine, Polymorphine, Planorbuline, Globigerine, and Carpenterie.—K. Svensk. Vet.-Ak. Handl. vol. xix. no. 4, p. 15. t+ Quart. Journ. Geol. Soc. vol. 1. 1894, p. 717. 338 MESSRS. T. R. JONES AND F. CHAPMAN are sometimes given off. The swollen segments are normally subdivided in the interior by septa, which in some cases extend across from wall to wall, but in others are not quite continuous, and so placed as to form a valvular opening. In the attached forms the swollen portion of the test is more, definitely Poiy- morphine both on the exterior and in the interior. The one or more apertures in the forms that are free are circular, being formed by the open end of the calcareous stolon-tube. In the attached forms, the apertures are usually formed by the pro- tracted terminations of the stolon-tubes, and are semicircular in outline where the entire test is adherent to the foreign body. In this point they closely resemble the terminal apertures of the Arenaceous genus Sagenella. The surface of the test in Ramulina is sometimes smooth, especially in the attached forms ; but more frequently the surface is either hirsute, strongly prickly, or tuberculate. An instance is known where Ramulina has been found within the chambers of another Foraminifer ; also on and in fossil Echinoderm-tests, traversing the ambulacral pores ; but this cannot be true parasitism. With regard to isomorphous forms resembling Ramulina in those other groups of the Foraminifera which are characterized by a difference of test-structure, it is instructive to note the comparable forms of Nubecularia tibia, Jones & Parker, WV. luci- fuga, Defrance, and WV. nodulosa, Chapman, in the Porcellanous group. In the Astrorhizide of the Arenaceous group we note the close resemblance shown in Aschemonella catenata (particu- larly fig. 8 of pl. xxvil. in Brady’s Monograph, about which Canon Norman remarks, in a letter, that Dr. Brady has here confused two types under one name). The singular form of Sagenella is also strongly suggestive of an attached cervicorn Ramulina, although there is no distinct initial series in the figured examples, pl. xxviii. figs. 14, 15. As the result of a systematic examination of all the published varieties of Ramulina, as well as many specimens from the Chalk- detritus (Chalk-marl) of Charing, Kent, the Gault of Folkestone, and other material of our own collecting, we have come to the conclusion that Ramulina, as at present known, can be specifically divided into five well-marked types, as described further on. The known specimens are here arranged under their specific heads according to date of publication. The following catalogue raisonné of the published and best ON THE GENUS RAMULINA. 339 known specimens supplies full evidence of their general agree- ment with accepted types and subtypes of Ramulina, and of the varietal or even specific differences among themselves. § 3. Enumeration of the various Type-species of Ramulina, and of the published Figures referable to them. 1. Ramulina levis, Rupert Jones, 1875. Figs. 1-4. With the original description of this species given in 1875 there was associated another, nearly similar, form of Ramulina, namely R. brachiata. Since this latter does not differ from the former in any essential particulars, and, in fact, helps to complete the diagnosis of the species, it is here referred to A. levis. Specific Characters.—Test free, straight or branching ; con- sisting of pyriform or subglobular segments united by more or less curved stolon-tubes. These latter sometimes proceed laterally from the segments, at others radially. The surface of the testis quite smooth. Texture thick, in this latter respect resembling ft. aculeata. References to published Figures and Descriptions of the JSiqured Specimens. 1. ‘ Polymorphum, Spherula vitrea levis,’ Soldani, 1791, Testaceo- graph. &e. vol.i. part 2, p. 115, pl. cxix. fig. a—Subglobose, smooth; two blunt projections, or short tubes, on one side, make it look like a skin-bag tied up at two places. Medi- terranean. 2. Ramulina levis, Jones, 1875, Rep. & Proc. Belfast Nat. F. Club, 1873-74; Appendix ITI, 1875, p. 88, pl. iii. fig. 19.— Three slightly inflated pyriform segments, united by short stolons, one of which projects laterally at right angles to the line of growth. Surface smooth. Chalk; Ireland.—Fig. 1. 3. Ramulina brachiata, Jones, 1875, Rep. & Proc. Belfast Nat. 340 MESSRS. T. RB. JONES AND F. CHAPMAN F. Club, 1878-74; Appendix III., 1875, p. 88, pl. ui. fig. 20. —A nearly spherical segment, with four more or less frag- mentary stolon-tubes. Surface smooth. Chalk; Ireland. —Fig. 2. 4, Ramulina levis, Balkwill and Millett, 1884, Journ. Micro- scopy, vol. ii. p. 84, pl. iv. fig. 7— A subtriangular, some- what inflated segment, giving off four straight or slightly curved tubes. Surface of test smooth. Recent; Galway, Ireland.—Fig. 3. 5. Ramulina Bradyt, Rzehak, 1895, Ann. k.-k. naturhist. Hofmus. vol. x. p. 228, pl. vi. fig. 5; and R. exigua, Rzehak, 1895, op. cit. p. 228, pl. vi. fig. 4.—Pyriform and subpyriform segments, with tubular extensions. 2. exigua exhibits the remnants of three tubes at one end. Surface of test smooth ; in R. Bradyi slightly grooved. Rk. Bradyi was briefly described by Rzehak in 1891, op. czt. vol. vi. part 1, p. 10. Lower Pliocene of Briiderndorf, Lower Austria.—Fig. 4. 2. Ramulina globulifera, Brady, 1879. Figs. 5-22. Specific Characters.—Test free, branching ; composed of seg- ments of different sizes, connected by stoloniferous tubes of diverse lengths. Segments from two to eight, or even more, in number ; globular or subglobular ; but sometimes greatly distorted in appearance (when they usually consist of a single segment with short straight or curved stolon-tubes projecting at various angles). Each segment provided with numerous stolon-tubes (sometimes radiating from different portions of the periphery), and which terminate in other chambers. The stoloniferous tubes, narrow in proportion to the bulk of the segments, are circular in section. Texture hyaline; shell-material usually very delicate and thin ; surface hispid or aculeate. 1. Ramulina globulifera, Brady, 1879, Quart. Journ. Micr. Sci. n. 8. vol. xix. p. 272, pl. vii. tigs. 32, 33.—Test free, con- sisting of from three to seven globular or subglobular chambers, with many tubulated apertures radiating from various parts of the surface. Chambers connected by some- what long stolon-tubes, usually straight, but sometimes twisted or otherwise distorted. Surface of test covered more or less thickly with fine prickles. The general struc- ON THE GENUS RAMULINA. 3841 ture of the test is very thin and delicate. Recent; Atlantic and Pacific Oceans.—Fig. 5. 2. Marginulina, cf. Dentalina aculeata, dOrb., Berthelin, 1880, Mém. Soc. Géol. France, sér. 3, vol. i. no. 5, p. 35, pl. ii. (xxv.) figs. 10-13 6.—Globular, subglobular, and ovoid seg- ments, with remnants of stolon-tubes at opposite ends in figs. 10, 11, and 12. Fig. 18 has one aperture only. The segments are more or less distorted in outline, and there- fore cannot be referred to such symmetrical or regular forms as Dentalina and Marginulina. The surfaces of the tests are thickly clothed with fine prickles. Gault ; Montcley (Doubs), France.—Fig. 6. 3. Moebiusispongia parasitica, Duncan, 1880, Journ. Roy. Micr. Soe. vol. i. p. 377, pl. x. (This has been shown by the recent researches of Mr. A. V. Jennings, Journ. Linn. Soc., Zoology, vol. xxv. (1895) pp. 317-8319, to be a true Ramulina.)\—Numerous thin-shelled, inflated, and irregular segments, with stolon-tubes uniting them; surface covered with fine prickles. The test is within a chamber of Car- penteria raphidodendron, Moebius, and a few of the stolons pass into the adjoiming chamber. Recent; Mauritius.— Bis. 7. 4. Ramulina globulifera, Brady, 1884, ‘Challenger’ Reports, vol. ix. p. 587, pl. lxxvi. figs. 22-28. (Figs. 25 and 27 were given by the same author in the Quart. Journ. Micr. Sci. n. s. vol. xix. 1879, pl. vii. figs. 32, 33.) See also Journ. Roy. Micros. Soc. 1887, p. 915.—For description see No.1. Fig. 28 shows the peculiar external develop- ment of a segment of Ramulina, an internal septation being 342 MESSRS. T. BR. JONES AND F. CHAPMAN probably indicated at the constricted junction. Atlantic and Pacific Oceans.—Fig. 8. 5. Tinoporus baculatus, Sherborn & Chapman, 1886, Journ. Roy. Micr. Soe. ser. 2, vol. vi.. p. 758, pl. xvi. fig. 24.—A com- pressed spheroidal segment, bearing about five short tubular processes. Surface of test granulate, or perhaps originally prickly. This specimen was suggested to be referable to Ramulina by A. Rzehak in 1891, Ann. k.-k. naturhist. Hofmus. vol. vi. part i. p. 5; and this was further noted by G. A. De Amicis, Boll. Soc. Geol. Ital. vol. xi. fase. 2 (1894), p. 108, in his list of known Ramuline and their, localities. London Clay, Piccadilly, London.—Fig. 9. 6. Lagena levis, Terrigi, 1889, Atti R. Accad. Lincei (Memorie), ser. 4, vol. vi. p. 112, pl. vi. figs. 2 & 3.—Subglobular segments, symmetrical ; but (viewed from above) laterally compressed. One of the figures shows a single remnant of a stolon-tube; the other has a tube at each of the opposite ends. Surfaces of the tests thickly clothed with fine prickles. Tertiary ; Palo, near Rome.—Fig. 10. 7. Nodosaria, Wisniowski, 1890, Mikrofauna [low Ornatowych Okolicy Krakowa, Part I. p. 15, no. 19, pl. i. (viii.) figs. 22 a, 6. —A subspherical segment with remnants of stolon-tubes at opposite ends. Surface of test covered with fine prickles. Ornatus-zone, Middle Jurassic ; Cracow, Poland. 8. Nodosaria hispida, Haeusler, 1890, Abhandl. Schweiz. Palaont. Gesellsch. vol. xvii. p. 103, pl. xv. fig. 40.—A simple subglobular segment, with stolon-tubes at the oppo- site ends (broken). Surface of segment bearing numerous well-developed prickles. Pholadomya-marls, Transversarius- zone, Middle Jurassic ; Saint Sulpice, Switzerland. 9. Lagena protea, Chaster, 1892, Rep. Southport Soe. Nat. Sci. for 1890-91, p. 62, pl. i. fig. 14.—An irregularly inflated 10. WL. 12. 13. 14. ON THE GENUS RAMULINA. 343 oblong segment, with two moderately long stolon-tubes, at nearly opposite ends. The surface of the segment clothed with fine hair-like prickles. Recent ; Coast of Lancashire. line, tk Ramulina globulifera, Egger, 1893, Abhandl. k. bayer. Ak. Wiss., Cl. II. vol. xviii. pt. ii. p. 310 (118), pl. ix. fig. 62.— An ovoid inflated segment, constricted across the shortest diameter, probably due to the conjunction of two spherical segments. Five tubules disposed peripherally (?) around the segment, three near one end and two at the other. Surface of the inflated segment bearing numerous fine prickles. Recent; Coast of Western Australia.—Fig. 12. Ramulina globulifera, De Amicis, 1895, ‘ Il Naturalista Sici- liano,’ Anno xiv. p. 48, pl. i. figs. 14a, 6, e—Three separate segments, all more or less spherical, with stout cylindrical stolon-tubes proceeding from them ; also with several smaller tubes disposed over the surface. The test is covered with fine prickles. Pliocene of Bonfornello, Sicily.—Figs. 13a, b, e. Specimen B-—-Chapman Collection.—A curved stolon-tube, with two tubules proceeding out of it at right angles, on the curved side. Gault; Folkestone. C— Chapman Collection.—A much compressed segment, with an irregular margin caused by the outgrowth of arrested tubules, one of which is longer than the rest. In one part of the segment there is a perforation, as though the margin was there completed by the coalescence of three tubules. Surface of test bearing a few fine prickles. Gault; Folke- stone. E—Chapman Collection.—See Journ. R. Micr. Soc. 1896, pl. xu. fig. 3.—An elongate pyriform segment, terminating at each end in a stolon-tube. The entire surface of the test is clothed with fine prickles. Gault ; Folkestone.—Very 344 16. 18. MESSRS. T. R. JONES AND F. CHAPMAN similar to fig. 6, but with a longer and more slender stolon- tube at each end. . L--Chapman Collection.—A subglobular segment having 4 long stolon-tube at each end. The test of the inflated portion is translucent, and exhibits evidence of internal septation. The surface both of segment and stolons bears a few fine prickles, sparsely scattered. Gault ; Folkestone. — Fig. 14. P--Chapman. Collection.—A double segment, closed at one end and terminating in an open and lipped (?) tube at the other. The commencing (?) segment is subglobular, and has a rather blunt ‘point. The second segment is elongate- pyriform. Gault; Folkestone.—Fig. 15. . Ramulina globulifera, Brady, Chapman, 1896, Journ. R. Micr. Soc. pp. 582-583, pl. xii. fig. 4.—A pyriform segment of Ramulina having remnants of stolon-tubes at the opposite ends. Surface finely prickly. Gault; Folkestone.—Fig. 16. Ramulina globulifera, Brady, Chapman, 1896, loc. cit. fiz. 6. —A much-compressed segment having a stumpy stolon- process and remnants of tubes on an inflated area of the surface. Gault; Folkestone.—Fig. 17. 19. R. globulifera. Rowe Collection.— Attached to fossil echino- derm tests, passing through the ambulacral pores and forming irregularly wandering tubes on the surface, more or less branching, and occasionally swollen; with prickly surface. Chalk *; Margate.-—Figs. 18-22. * Ramulina globulifera, Brady, is catalogued with a doubt in a list of the Cenomanian Forasinifera of Mecklenburg by G. Schacko, Arch. Freund. Nat. Mecklenburg, 1892, p. 157. ON THE GENUS RAMULINA. B45 3. Ramulina aculeata, Wright, 1886. (Non @ Orbigny’s Dentalina aculeata.) Figs 23-42. Specific Characters.—Test free ; composed of subangular, in- flated segments connected by stout, curved or rarely straight stolon-tubes. The segments almost invariably show a septate division of the interior; the septa, however, nearest the stolon- tubes are always interrupted in the centre, thus forming a valvular or constricted entrance to the tubes. The walls of the test much thicker than in #. globulifera. Surface of test bears numerous strong prickles or tubercles ; and these appear to be formed of non-tubular shell-materiai. 1. “Small flask-like objects, curiously echinulate,” Williamson, 1847, Mem. Lit. Phil. Soc. Manchester, vol. viii. p. 78, pl. iv. figs. 73, 74. Named Dentalina aculeata in the same work for 1872, ser. 8, vol. v. p. 186.—Test pyriform, and irregularly compressed. Oral apertures prominent, and probably once connected with stolon-tubes. Surface covered with strong prickles. Chalk-marl; Kent. 2. Lagena distoma-aculeata, Parker and Jones, 1865, Phil. Trans. vol. ely. pt. 1, p. 348, pl. xvii. fig. 5.—An irregular flask- shaped segment, with an attenuated neck. The surface of the test decorated with blunt prickles or rounded processes. Eocene; Grignon, Paris.—Fig. 23. 3. Lagena aspera, Reuss, Brady, 1884, ‘ Challenger’ Rep. vol. ix. p. 457, pl. lvii. fig. 12—This subglobose specimen differs from Z. aspera, Reuss, to which species Brady refers it, by being compressed and inequilateral, and by having a narrow oblique tubular aperture at one end and a blunt eccentric projection at the other. The “ento- and ectosolenian” characters are not apparent. The surface is finely tuber- culate. Locality? Recent.—Fig. 24. LINN. JOURN.—ZOOLOGY, VOL. XXVI. 25 346 MESSRS. T. R. JONES AND F. CHAPMAN 4. Ramulina aculeata, Wright, 1886, Proc. Belf. Nat. F. Club, 1885-1886, Appendix IX., 1886, p. 331, pl. xxvii. fig. 11.— Two asymmetrical subspherical segments, closely conjoined by a short stolon. Surface of test bearing scattered tubercles, the remnants of strong prickles. Chalk; Keady Hill, Ireland.—Fig. 25. 5. Lagena hispida, Haeusler, 1887, Neues Jahrb. f. Min. Bd. i. pp- 185 & 189, pl. v. fig. 12.—A subtrigonal and inflated segment *, with remnants of the stolon-tubes at the angles. Gault ; Ste. Croix, Switzerland.—Fig. 26. 6. Ramulina aculeata, Burrows, Sherborn, and Bailey, 1890, Journ. Roy. Micr. Soe. p. 561, pl. xi. fig. 16.—An elongate, subspherical segment, with remnants of stolon-tubes at the opposite ends. Surface of the test bearing many strong prickles. Red Chalk; Speeton, Yorkshire. 7. Nodosaria hispida, Haeusler, 1890, Mém. Soc. Pal. Suisse, vol. xvii. p. 103, pl. xiv. fig. 15—Two pyriform segments, with a constricted junction, and one extremity produced. Surface tuberculate, probably from the remnants of strong prickles. Pholadomya-mavrl, Transversarius-zone, Middle Jurassic ; Switzerland.—Fieg. 27. . Lagena hispida, Terrigi, 1891, Mem. Descriz. Carta Geol. Italia, vol. iv. p. 77, pl. ii. fig. 2.—An acute-oval segment, not quite symmetrical; sharp at the ends; one evidently stoloniferous. Surface furnished with numerous distinct prickles. Tertiary; Capo di Bove, Italy. 9. Polymorphina proteus +, Beissel, 1891, Abhandl. k. Preuss. geol. Landesanstalt, n. s. part ili., pp. 57-62, pl. xii. fig. 18. —A bulbous, subspherical segment, with three tubes of equal length, apparently on one plane and at nearly equal angles. The surface has a few strong prickles. The section co * A somewhat similar example is figured as Lagena bicornuta by Dr. G. Egger, from the South Atlantic (Abhandl. k. Bayer. Akad. vol. xviii. 1893, p. 3820, pl. x. figs. 92, 93). It is flattened flask-shape, pinched at the bottom into two short, horn-like, blunt angles; neck protruding ; surface roughly pitted and porous. t The “Polymorphina proteus” in pl. ix. figs. 1-56, and pl. xii. figs. 1-8, comprises various common species and varieties; but pl. xii. figs. 9-16 are different and include the portions of Ramulina here described. Possibly this should be regarded as a distinct species under the name of Ramulina proteus (Beissel). See Journ. Linn. Soc., Zoology, vol. xxv. p. 499, and p. 384 of this paper. ON THE GENUS RAMULINA. 347 (fig. 28 6) shows segmental cavities at the bases of the tubes. From Marl of the zone of Belemnitella mucronata in the Upper Chalk ; the Frederichsberg, Aix-la-Chapelle.—Figs. 28 a, b, c. 10. Lagena tuberculata, Perner, 1892, Cesk Ak. Cisare Fran- tiska-Josefa Prague (Paleontograpbica Bohemiz, vol. i. pp- 49-65), p. 56, pl. v. figs. 19 a, b—A single irregularly flask-shaped segment and stolon-tube. The surface bears a number of strong blunt prickles. The test is thick, as is shown by the orifice, fig. 296. Chalk; Bohemia.— Figs. 29 a, 6. 11. Ramulina globulifera, var. miocenica, Rzehak, 1895, Ann. k.- k. naturhist. Hofmus. vol. x. p. 222, pl. vi. fig. 6. FR. Kittli, Rzehak, 1895, op. cit. p. 221, pl. vi. figs. 7, 9, 11 (also men- tioned in a list, op. cit. 1891, vol. vi. pt.i. p. 2). &. ef. aculeata, Rzehak, 1895, op. cit. p. 222, pl. vi. figs. 8, 10.— In fig. 6 (our fig. 30) we have a strongly aculeated, elongate- oval segment, bearing four, short, divergent, lateral tubes, and terminated at each end by a stolon-tube (broken). An example of somewhat similar character is shown in fig. 9 (our fig. 31); but here the segment is triangulo-globose, with three large stolon-tubes proceeding from it, and with a fourth, central, smaller tube at right angles to the others. The surface in this specimen is sparsely, though strongly, aculeate. The figs. 7 and 8 are elongate-pyriform segments terminated by stolon-tubes more or less short. The re- maining figs. 10 and 11 represent sections of the test, and show. marked differences in their thicknesses. The first two 348 12. 138. 14. 15. 16. MESSRS. T. R. JONES AND F. CHAPMAN named are from the Bartonian of Briiderndorf, Lower Austria; and the last from the Miocene of Osluwan, Austria. —Fig. 30 (‘‘ miocenica”’); fig. 31 (“Kattli”’). Hormosina lateralis, Grzybowskiego, 1895, Rozpr. Ak. Krakow. Ser. II. vol. ix. p. 186, pl. ii. figs. 2 a, 6.— Irregular and subangulate segments ; in one specimen, two connected with a short stout stolon, and marked with irregular pro- jections. Surface decorated with beaded or small blunt processes, in more or less parallel lines along the length of the shell. Tertiary; Carpathians.—Figs. 32 a, 6. Ramulina aculeata, Chapman Collection. See Journ. R. Micr. Soc. 1896, p. 583, pl. xii. fig. 7.—A complete segment with fragment of another, and of subglobular form. Surface with strongly developed prickles. Gault; Folkestone.—Fig. 33. Ramulina aculeata, Chapman Coli., loc. cit. fig. 8—A ong stoloniferous fragment with lateral processes (broken). Surface of test strongly prickled. Gault; Folkestone.— Fig. 34. Ramulina aculeata, Chapman Coll., loc. cit. fig. 9.—A stolo- niferous portion, with stumpy processes, somewhat con- torted. Surface covered with strong prickles.—Gault ; Folkestone.—Fig. 35. H—Chapman Coll. — An attached initial segment of Rk. aculeata adherent to a minute pebble. It is irre- gularly subpyriform, bending along the edge of the object ON THE GENUS RAMULINA. 349 to which it is attached, and gives off a stolon-tube of some length, which was probably connected with other inflated segments. The surface of the test is studded with strong, sharp prickles. Gault; Folkestone.—Fig. 36. 17. O—Chapman Coll.—A subglobular segment, having three stolon-tubes radiating from it. The surface of the test is sparsely studded with prickles. Gault ; Folkestone. 18. Q—Chapman Coll.—A fragment of one of the stolon-tubes of Ramulina aculeata, showing the coarse blunt spines and the perforations of the test. Gault; Folkestone — Fig. 37. 19. R—Chapman Coll.—A. fragment of a stolon-tube of Ramu- lina aculeata. Gault; Folkestone.—Fig. 38. 20. Specimen M ; 21.8; 22. T—Chapman Coll.—Three examples of Ramulina aculeata (figs. 89, 40, 41) which have been sliced horizontally and medially to demonstrate the presence of septa within the segments. In fig. 39 (M) we have a well- inflated segment, of subtriangular shape; and from two of the corners there proceed stolon-tubes. The interior of the segment is divided by partial and complete septa, somewhat after the manner of Polymorphina, although not so regularly arranged. At the junction of the stolon-tube with the segment there exists a partial partition of the shell-cavity, the septum being interrupted either in the middle or at one side near the outer shell-wall. Fig. 40 (S) is similar in shape to the foregoing, but there are no apparent septa within the cavity of the segment. The junctions of the stolon-tubes with the segment are partially divided by imperfect, valve-like septa; one being thin, and with its free edges pointing up the tube; the other almost completely closing the tube, except a fine median canal, left for intercommunication. The common triangular arrangement of the stolon-tubes 350 MESSRS. T. R. JONES AND F. CHAPMAN in Ramulina aculeata is shown in section in fig. 41 (T), where there is no septation of the internodal cavity ; but the tubes are divided from the central portion by a valvular septum in one case; in another by a complete septum (this may, however, be accounted for if the section were not quite median); and at the third tube there is no septum. Gault; Folkestone. 23. Ramulina aculeata, Rowe Collection.—A._ flask-like segment, which has found a point @appui in an ambulacral pore of an echinodermal test. Chalk; Margate-——Fig. 42. 4. Ramulina Grimaldii, Schlumberger, 1891. Figs. 43, 44. Specific Characters.—Test attached. Having a distinct Poly- morphine commencement, and proceeding to a rectilinear out- growth of great extent, which branches off laterally at almost regular intervals, untilit becomes less regular and loses itself in large and small ramifications, sometimes anastomosed. The laterals have racemose endings. 1. Ramulina Grimaldii, Schlumberger, 1891, Mém. Soc. Zool. France, vol. iv. p. 509, pl. v.—Test attached to molluscan shells. The outgrowth, specified as above, is sometimes eleven times the length of the commencing Polymorphine series. The outgrowths are not entirely apical, but some- times start from the sides of the early part of the test. The initial form appears to be that of Polymorphina gibba, d’Orb. Recent; between Fayal and Pico in the Azores.— Figs. 43, 44. ON THE GENUS RAMULINA. 3oLl 5. Ramulina cervicornis (Chapman, Polymorphina Orbignii, var. cervicornis, 1892). Figs. 45-51. Specific Characters.—Test attached ; commencing with a Poly- morphine series, and continuing (usually from the apex, though not always confined to that region) to grow ina strong, branching manner, often closely resembling the antlers of deer. Apertures at the terminal points of the branchlets. 1. Polymorphina Orbignii, var. cervicornis, Chapman, 1892, Geol. Mag., dec. 3, vol. ix. p. 54, pl. i. fig. 5.—Test attached to a fragment of Nuwcula. The initial series is that of Polymorphina fusiformis, Romer. The cervicorn outgrowth is five or six times the length of the initial series of chambers. ’ It is apical, and, after proceeding for a short distance in a vertical direction, branches off laterally, bearing the longest part of the outgrowth. Gault; Folkestone——Fig. 45. 2. Specimen A—Chapman Collection.-— Commencing series broken open, and revealing a Polymorphine septation; the outline is that of P. angusta, Egger. At the apex there isa slight constriction, upon which follows a tubular extension, with a lateral offshoot. The whole is attached to a frag- ment of molluscan shell. It is worthy of notice that the constriction above referred to, as occurring between the Polymorphine series and the outgrowth, probably facilitates the separation of the two kinds of growth, giving rise to Ramuline without distinct commencements, as in the aculeate examples. Gault; Folke- stone.—Fig. 46. 3. D—Chapman Collection.—An elliptical commencement, re- sembling P. fuszformis, Romer, is followed by an apical outgrowth, which, before extending far, sends out a lateral 352 MESSRS. T. R. JONES AND F. CHAPMAN cervicorn branch. Attached to a piece of molluscan shell. Gault ; Folkestone.—Fig. 47. 4, G—Chapman Collection.—This example exhibits an initial Polymorphine structure on the plan of P. fuszformis, Romer, but differs from the usual forms of Ramuline commencements in having outgrowths at both ends of the Polymorphine portion. At one end is a short cervicorn outgrowth, at the points of which the apertures are clearly seen ; at the other end are the rudiments of two tubes, one springing forth laterally. This specimen is incomplete, and what was possibly the most interesting portion has been removed by the fracture of the Frondicularia to which it was attached. Gault; Folkestone.—Fig. 48. 5. J—Chapman Collection.—The initial Polymorphine series, referable to P. fusiformis, Romer, is succeeded by an out- growth at each end. One is definitely cervicorn ; and the other (broken) was probably of the same character. Test attached to a fragment of molluscan shell. Gault; Folke- stone.—Fig. 49. 6. Chapman Collection. Journ. R. Micr. Soc. 1896, p. 584, pl. xii. fig. 10.—A Polymorphine commencement and a single process, widening rapidly, subtriangular in form, with an aperture at each of the anterior angles. Gault; Folke- stone.—Fig. 50. 7. Chapman Collection. oe. cit. fig. 11.—A dichotomous form, attached. The initial Polymorphine chambers (obscured by fracture) produce two divergent, lumpy, apical tubes; one having a doubly dichotomous, blunt, branching termination ; the other ends in a swollen, dichotomous, and, as it were, stunted branch. Gault; Folkestone.—Fig. 51. ON THE GENUS RAMULINA. 353 § 4. Occurrence of Ramulina. Ramulina, so far as at present ascertained, makes its first appearance in strata of Jurassic age, having been found in the Transversarius-zone of Switzerland and the Ornatus-zone of Poland. It occurs in the Neocomian (Bargate) Pebble-beds of Littleton and Chilworth, Surrey *, and in many succeeding formations. Thus we know of it as specimens from various other Cretaceous deposits, such as the Gault of Folkestone, Merstham, and Godstone; and as figures and descriptions of examples from the Gault of France and Switzerland. We have specimens also from the Chalk-mar] of Charing and Folkestone, and the Chalk of Kent, the Phosphatic Chalk of Taplow, Buckinghamshire, and the Chalk of Ireland. The Chalk of Aix-la-Chapelle aud that of Bohemia have also yielded definite forms which have been figured. The Ramuline above referred to are mostly of the types that have the test covered with short prickles: one with the shell thin (globulifera) ; and the other, which we recognize as R. acu- leata, with a thick test. Since the identity of any of these forms with d’Orbigny’s Dentalina aculeata is not regarded as satisfactory by some authorities, and as Mr. Joseph Wright has figured a specimen of the latter kind as “ R. aculeata (d’Orb.),”’ we use that specific name for these prickly Ramuline, without reference to d’Orbigny’s named specimen, thus, &. aculeata, Wright. Norz.—As regards the distribution of Ramuline in the Gault of Folkestone, it is interesting to note the gradual increase in their numbers from Zone III. (where they first make their appearance in this formation) to the top of the Gault at its junction with the Chloritic Marl; and so, increasingly, upwards to the Upper Chalk. § 5. Distribution and Range in Time of RAaMULINA. 1. Ramurina avis, Jones. Chalk; Ireland. Tertiary; Austria. Recent; Ireland. 2. R. auoBuLirerRa, Brady. Jurassic ; Poland and Switzerland. Lower Greensand ; Surrey. Gault; Folkestone and France. Tertiary ; England, Austria, Italy, and Sicily. Recent; Mauritius, West Australia, Lancashire, Atlantic, and Pacific. * Quart. Journ. Geol. Soe. vol. 1. 1895, pp. 717, 718. LINN. JOURN.—ZOOLOGY, VOL. XXVI. 26 ook OF THE GENUS RAMULINA. 3. R. acunEata, Wright. Jurassic; Switzerland. Gault ; Switzerland, Folkestone. Red Chalk; Speeton. Ceno- manian; Bohemia. Chalk-marl; Kent. Chalk; Ireland, Aix-la-Chapelle. Tertiary; France, Italy, Austria. Recent; South Atlantic (?). 4. R. Grimatpti, Schlumberger. Recent; Azores. 5. R. cERVICORNIS, Chapman. Gault; Folkestone. Table showing the Range in Time and the Relative Numbers of Forms noticed. EEG E e b s Entries of Total Bis) |) ee) 2 Weabhened |e 00s ras ml/O;}a | m& forms. eet Ramulina levis, Jones ......|...... Ay |) Ori) easttecee 5S Ves qn R. globulifera, Brady ...... 2) dSalesu eon 16 7 23 | ae 3H RB. aculeata, Wright ......... won| Imago. x6. ” ” Wing. X32. . Termes lacessitus. Soldier. x8. * 3 Side view of soldier’s head. x8. Ae A Nymph. x6. . Termes hospitalis. Soldier. x8. Bs i Side view of soldier's head. x8. | AA . Imago. x6. sy i Wing. x3. daviland,Photo. N Linn. Soc.Journ.Zoon Vou. XXVI_ PL? 2 E Wilson, Lith.Camoridge. Ole INS RIMMEL eh, SPEC WS) AIG I Ie 1D) EMOROSUS. SARAWAK. SANTUBONG. . | Haviland | Linn.Soc.Journ.Zoon Vou. XXVI.Pl.23. Westwood Bequest E Wilson, Lith.Cambridge. MALAYAN «5. AFRICAN TERMITES. Linn. Soc.Journ.Zoor Vou. XXVI.P] 24. Haviland. E.Wilson, Lith Cambridge. MALAYAN 2S. AFRICAN TERMITES. 1 Westwood Bequest Haviland. Linn. Soc.Journ.Zoou Vou. XXVI.P1.25. Cy By Westwood. Bequest E Wilson, Lith.Cambridge. MALAYAN 25S. AFRICAN TERMITES. PROF. A. DENDY ON PONTOBOLBOS. 443. On Pontobolbos, a Remarkable Marine Organism from the Gulf of Manaar. By Arruur Denpy, D.Sec., F.L.S., Professor of Biology in the Canterbury College, University of New Zealand. [Read 18th November, 1897.] (Puatses 26 & 27.) SoME years ago I received from Mr. Hdgar Thurston, Super- intendent of the Government Central Museum at Madras, a number of Sponges collected in shallow water off the shores of RAmésvaram Island, 7. e. the Madras coast of the Gulf of Manaar. Among these were fifteen specimens of an organism which at once struck me by its peculiar appearance, and which microscopic examination soon showed to be no sponge. I have lately made a careful investigation into the structure of this organism. At first I felt certain that it was animal in nature, and commenced my investigations from the zoological standpoint. I have, however, been gradually forced to the conclusion that it is, at any rate largely, of Bacterial origin. Not being a bacteriological specialist myself, I feel some diffidence in offering the present paper for publication; but the organism in question is so remarkable, and evidently forms such a conspicuous feature in the marine life ot the Gulf of Manaar, while at the same time it presents such a striking resemblance in some respects to certain obscure forms of life usually regarded as animal, that I venture to hope the following description and suggestions may be of some general biological interest. It seems desirable, for convenience of reference, to propose a new generic name for the organism under discussion, and I therefore name it Pontobolbos, from the Greek wovros, sea, and BoAPos, bulb, in allusion to its marine habitat and concentrically laminated structure. The question of generic diagnosis may be conveniently left to the future. Pontobolbos manaarensis, nu. sp., forms hemispherical or cushion- shaped masses, of a brown colour in the living condition, at- tached to rocks in shallow water (Pl. 26. figs. 1-4). The size of the spirit-preserved specimens varies from 13 mm. to as much as 36 mm. in diameter, and the colour is pale grey. The upper surface is more or less convex, smooth and even, but very finely granular in appearance. ‘The lower surface, which has been detached from the rock, is more or less flattened or even concave 44,4, PROF. A. DENDY ON PONTOBOLBOS, its exact form no doubt depending upon the nature of the surface on which it lay. One specimen (fig. 3) is irregular in shape, and appears to have been formed by the fusion of three growing side by side. The upper surface occasionally exhibits small round pits, which may be either deep, as in fig. 1 and at p in fig. 2, or shallow and pock-like, as at p in fig. 4. The latter are especially developed near the margin. It seems certain that these pits are not proper to the organism: there is no canal-system in connection with them, and they are probably due to some external agency, possibly to the action of parasites. Their presence certainly increases the superficial resemblance of the organism to a sponge. The texture, after preservation in spirit, is very compact, tough and like that of indiarubber, and there is a good deal of sand in the deeper layers. When cut in half vertically (P1. 26. fig. 2) the organism exhibits a very strongly marked, concentrically lamellated structure, although the lamelle do not show the least tendency to separate from one another. Even a very thick slice in spirit is translucent, and shows to the naked eye that the lamellated appearance is due to the occurrence of thin opaque layers in a transparent ground- substance. These opague layers evidently mark old surfaces of growth, and the distance between successive surfaces varies considerably, up to about 0°75 mm. Under a low power of the microscope a tolerably thin vertical section exhibits the appearance shown in PI. 26. fig.5. It will be seen that the opaque layers consist of a dense flocculent granular substance, and that they are connected together by a coarse network of similar but less dense material the strands of which ramify, chiefly in a radial direction, through the intervening layers of transparent ground-substance (cf. fig. 6). This floceu- lent material stains fairly readily with borax-carmine, and very intensely with aniline stains such as iodine-green, fuchsin, and methyl-violet. I have not, however, been able to detect any nuclei in it. It also stains faintly yellow with iodine, and more intensely yellow with chlor-zinc-iodide after soaking in iodine. It does not give the xanthoproteic reaction with nitric acid and ammonia, nor does it give the brick-red colour with Millon’s reagent, so characteristic of protoplasm, though both these reagents acted satisfactorily upon the protoplasm of fragments of Oscillarian alge included in the sections. Frequently irregular layers of much coarser granular material, A REMARKABLE MARINE ORGANISM. 445 which does not stain with any of the above reagents, are met with, as at 7,7 in fig. 5. On treatment with sulphuric acid these granular layers dissolve with brisk effervescence and forma- tion of needle-shaped crystals in the neighbourhood: we may therefore conclude with some degree of confidence that they consist of carbonate of lime deposited by the organism. They are not to be confounded with the foreign particles of sand which occur imbedded in the deeper layers as at g, g in fig. 6. Careful examination of the transparent ground-substance between the flocculent, deeply-staining material shows that the former is not by any means structureless as it appears at first sight, but contains, or perhaps one should even say consists of, innumerable exceedingly slender unbranched threads, lying close together, and generally, but by no means always, arranged in a radial manner, more or less at right angles to the successive surfaces of growth. These threads are most strikingly brought into view by mounting in chlor-zinc-iodide a thin section previously soaked in a solution of iodine in potassium iodide (Pl. 27. fig. 7). They then assume a beautiful blue or violet colour, which is also produced by treatment with iodine and sulphuric acid and indicates that they consist, at any rate largely, of cellulose. The threads are only about 0:0018 mm. in diameter, and further examination shows them to be the cellulose sheaths of chains of short, rod-like bacteria. In the deeper layers of | the organism nearly all the sheaths are empty, but a few may still contain the chains of bacteria to which they owe their origin (Pl. 27. fig. 8). The bacteria themselves are stained yellow or brown by the action of the iodine in the chlor-zinc-iodide method, and are then clearly seen to consist of short rods, somewhat less in diameter than the containing sheath, only about 0:00135 mm., and two or three times as long as they are broad. The size of the rods, however, varies a good deal, as will be seen by reference to fig. 8. As we approach the surface we find more and more of the rods, until in the youngest layer of the colony we have a -dense and almost solid mass of more or less felted chains of rods. Thus fig. 9 shows a small portion of a thin vertical section cut by the paraffin method and stained, after drying on the slide, with carbol-fuchsin, followed by iodine. It will be noticed that the chains of rods are very distinct, having taken the stain strongly, but the cellulose sheaths are not visible. This figure 446 PROF. A. DENDY ON PONTOBOLBOS, should be compared with fig. 7, showing a portion of one of the deeper layers where all the rods have disappeared and the empty sheaths are brought into view by the chlor-zinc-iodide method. It appears to me that there are two possible views as to the nature of -Pontobolbos: first, that the organism is entirely bacterial in origin, the flocculent layers and network being due to the formation of slime or jelly by the bacteria themselves ; second, that it is due to symbiosis between the bacteria and some gigantic rhizopod the protoplasm of which is seen in the floceulent layers and network. We may consider these two views separately. That the great bulk of the entire organism consists of filaments of some Schizophyte and their sheaths, I think has been already sufficiently demonstrated. The filaments appear to- be very closely related to Crenothrix. Crenothria Kiuhniana is described as consisting of “‘ cocci, rods and thread-forms. The threads are colourless, 1:5—5 p thick, and club-shaped at the extremity, reaching a diam. of 6-9 p. The threads form colonies with a brick-red, olive-green, or dark-brown to brown-black coloration, caused by impregnation with oxide of iron. The threads are distinctly articulated, and ensheathed. The segments are set free when the sheath bursts, and develop into new threads. In other cases the segments remain enclosed, and. subdivide into discs, which, by vertical fission, break up into globular forms (cocci). These again develop into new threads, either within the sheath, eventually penetrating it, or after they are set free.” “The micro-organism appears in little whitish or brownish tufts in wells and drain-pipes, and it not only renders drinking- water foul, but may stop up the narrower pipes.” The above description is taken from Crookshank’s ‘Manual of Bacteriology’ (2nd edition, p. 322). The figures showing colonies of threads growing out of a zoogleea of cocci are especially suggestive of a comparison with Pontobolbos. The chief differences appear to be that in Pontobolbos the threads. are more slender and of the same diameter throughout, and that they grow much more closely together to form compact layers. on the top of which the zoogloea is presumably passed out from the upper ends of the sheaths. This zoogloa, if such it be, may form the base from which another layer of rods grows- A REMARKABLE MARINE ORGANISM. 447 up, so that by repetition of the process a large stratified mass is formed in which the lower layers are composed almost entirely of empty cellulose sheaths and the flocculent staining-material (remains of zoogleea ?). I have not been able to demonstrate the existence of cocci m the flocculent layers or network of Pontobolbos, though I have detected occasional indications of their formation within the sheaths; and the fact that the outermost flocculent layer is already present before the threads in the layer below it have discharged their contents (Pl. 27. fig. 9) seems to indicate that it cannot be entirely due to zoogloea formation. According to Klein *, zoogloea masses always present themselves as uniformly eranular, the granules or micrococci being of the same size. The darkly staining layers and network of Pontobolbos, on the other hand, are by no means uniformly granular but rather floceulent. They contain granules of various sizes, but com- paratively few that are at all sharply defined. The Bacteria, such as Leptothrix and Crenothrix especially, are well known to be very closely related to the blue-green alge (Cyanophycee)*t, and there are certain species of the latter group which are also in some respects pretty closely comparable with Pontobolbos. Such especially appears to be the case with the Oscillarian Lyngbya (Phormidium) membranacewm, which forms firm, leathery layers of a greenish-black or olive colour; with threads 3-8 p thick, greenish or brownish, and delicate colourless sheaths. The terminal cell is narrowed and often provided with fine cilia at the apex. This species was placed by Kiitzing in the genus Phormidium, distinguished from Lyngbya by the fact that the sheaths are grown together to form a firm, often laminated structure. More recent authors, however, appear no longer to keep the two genera apart. This description also is certainly very suggestive of Pontobolbos, though the threads are very much thicker. I am not aware, however, that the darkly staining layers and network have ever been seen ina Lyngbya or indeed any Oscillarian. If the threads of Pontobolbos were really Lyngbya * *Micro-Organisms and Disease,’ p. 60. +t Thus Le Maout and Decaisne include Bacterium and Leptothrix, together with Oscillatoria, Lyngbya, and others, in the one family Oscillatoriee. t Vide Reinhold, “Die Cyanophyceen (Blautange) der Kieler Fohrde” (Schrift. d. Nat. Vereins. fiir Schleswig-Holstein, Bd. viii. Heft 2, p. 178). A4A8 PROF. A. DENDY ON PONTOBOLBOS, filaments, it would make it all the more difficult to account for the origin of these layers, for Lyngbya is supposed to reproduce by means of motile filaments and not by means of gonidia or cocci, and the zoogloea hypothesis of the origin of the granular layers would accordingly be untenable. The elongated form of the cells in the filaments, however, seems to exclude the idea that we are dealing with a true Oscillarian in the case of Pontobolbos. True Oscillarian filaments do occasionally occur as foreign bodies (Pl. 27. fig. 8 a), but they offer a striking contrast in appearance to the vastly more numerous filaments which I consider as bacterial (fig. 8). Another comparison has been suggested to me by Mr. R. M. Laing, M.A., B.Se., and that is with the calcareous pebbles formed by Schizothrix fasciculata as described and figured by Murray *. These pebbles were found at the bottom of a pond in Michigan, and I extract the following account of them from Mr. Murray’s interesting paper :—“‘The specimens vary in size from an inch to three inches and a half in diameter, are hollow in the interior, and show a stratified or concentric-zoned struc- ture. On decalcifying a portion of one, I found that it was composed of a densely interwoven mass of filaments evidently not all of the same nature, but the predominating kind was clearly a species of Schizothrix, while mixed with it there were other forms +, notably filaments of Stigonema and Dichothria. f examined portions here and there from a number of the pebbles, and in all cases found the strong stout sheaths and filaments of Schizothriv composing by far the greater part of the decalcified mass. At the surface the Schizothria filaments had been clearly alive when the pebbles were gathered, while nearer the centre older sheaths only were found. It was apparent that the Schzzothrix died off internally while fresh crops were produced on the surface adding to the growth of the pebble. .. . S. fasciculata is known from the countries of Central Europe, and forms small, stony, cushion-like calcareous masses. The filaments are always entangled where they are incrusted, but on the surface, where they are more or less free, there is very little entanglement and the filaments are almost straight and parallel. The ordinary incrustation formed by this * Phycological Memoirs, April 1895. + Compare the occasional occurrence of Oscillarians in Pontobolbos (vide Pl. 27. fig. 8a). A REMARKABLE MARINE ORGANISM. 449) species is said to be zoned in section, and its surface is generally mammillate.”’ The resemblance of these calcareous pebbles to Pontobolbos is certainly very striking, especially when we bear in mind that the latter also contains a considerable quantity of carbonate of lime. At the same time there appears to be no doubt that the filaments of Pontobolbos do not belong to a Schizothria. The ease of WSchizothrix, however, seems to show that somewhat similar filaments are capable by themselves of giving rise to a massive laminated structure without the co-operation of any symbiotic animal organism. ‘We come now to the consideration of the second hypothesis as to the nature of Pontobolbos, viz., that it is a symbiotic organism, consisting of bacterial filaments growing in association with a gigantic Rhizopod. This hypothesis would at once explain the existence of the darkly staining flocculent layers and network, which form by far the most conspicuous feature in sections stained by any of the ordinary methods applicable to animal tissues. Nor is it impossible to point to certain known Rhizopods the structure of which seems to support this view. I refer to some of the large Foraminifera, such as the extinct Nummulites, and especially to that remarkable fossil form Loftusia persica, described by Carpenter and Brady *, which attains a length of 34 inches and a breadth of 17 inches, and exhibits a laminated appearance in section extremely suggestive of Pontobolbos. In Loftusia, as in many Foraminifera, the typical calcareous skeleton is replaced by sand. In many siliceous and horny sponges a similar replacement takes place. Indeed, there seems to be a strong tendency, both amongst Protozoa and Sponges, to replace the proper skeleton with foreign material whenever such material is available. They do this in accordance with what might be termed “‘ The Law of Economy in Skeleton-formation.” It is also well known that symbiotic (or parasitic ?) Schizo- phytes are frequently met with in large numbers in Sponges, some of which, indeed, appear to be habitually infested with such forms. Thus Schulze has described* the occurrence of Oscillaria spongelie in Spongelia pallescens, where it occurs even * Phil. Trans. Royal Soc. 1869, p. 739 (Plates Ixxvii. to lxxx.), t Zeit. wiss., Zool. Bd. xxxii. p. 147. A450 FROF. Ae DENDY ON PONTOBOLBOS, in the ciliated embryo. Marshall has described * the occurrence of an Oscillaria in Dysidea (Psammoclema) ramosa, in which it is noteworthy that he was unable to detect chlorophyll. Trachycladus levispirulifer, again, a common Australian sponge, according to Carter t owes its characteristic red colour to an Oscillarian which has the form of short rods consisting of only four cells each. ~ Amongst the Protozoa, many Radiolarians are habitually infested with “ yellow cells,’ now generally believed to be sym- biotic Alge ; and, in short, the association of low forms of animal and vegetable life in what is presumably a symbiotic manner is a phenomenon of common occurrence, while the parallel case of the Lichens will occur to everyone. Assuming that Pontobolbos owes its origin to such symbiosis, it is not difficult to see what mutual advantages would be derived by the associated organisms. The Rhizopod gains an admirably firm and tough skeleton of cellulose, and is thus relieved from the necessity of secreting the usual calcareous skeleton. In accordance with the law of economy in skeleton-formation the latter may have disappeared, though possibly the irregular calcareous layers described above may represent the proper skeleton in a vestigial condition. With the proper skeleton, of course, all traces of the chamber-arrangement, so characteristic of the Foraminifera, may be supposed to have disappeared also, if indeed it ever existed. The Schizophyte, on the other hand, being in all probability bacterial in nature and devoid of chloro- phyll, may well be supposed to obtain its food-supply at the expense of the Rhizopod. The chief objections to the hypothesis of symbiosis in the case of Pontobolbos lie, first, in the failure, as noted above, to ébtain the characteristic protoplasmic reactions with Millon’s reagent and with nitric acid and ammonia; and, second, in the failure tc detect nuclei in the supposed protoplasmic layers and network. These objections appear to me at present to be very serious, although it must be borne in mind that micro-chemical reactions are not always conclusive, and that nuclei have not yet been detected in all the known lower organisms, while the staining with borax-carmine, aniline dyes, and iodine must also be accounted for. All I have been able to do has been to describe * Zeit. wiss. Zool., Bd. xxxv. p. 111. t Ann. & Mag. Nat. Hist. [5] xvi. p. 357. Dendy. Linn. Soc Journ. Zoot Vor. XXVI Ph.a6. AD.deladnat. Geo West & Sons imp Parker & Percy litn. PONTOBOLBOS MANAARENSIS. @ (eo West & Sons imp. so woes Remsen ares Spine enere NN Loar Soc Journ. Zoot Vou. XXVI.Pn.27,| - Ss a en a PS Ss ce a ae SS en oa ) AD .deladnat. Parker & Percy lith Dendy. TOBOLBOS MANAAREWNSIS. 9 A REMARKABLE MARINE ORGANISM. 451. the organism as I found it, to state what appear to me to be the only two possible explanations of its very remarkable structure, and to leave the future to decide which, if either, of these explanations is correct. In conclusion it may be worth while to refer for a moment to the enigmatical fossil Stromatoporide *, some of which bear a marvellously close resemblance to Pontobolbos, and may quite possibly have been of the same nature. Christchurch, N. Z., June 30, 1897. P.S.—From the report of the meeting at which this paper was read, I see that Professor Howes found some apparent vestiges of sponge-spicules in the specimen which I forwarded to him for examination. I have not observed any spicules in my own numerous preparations, and am quite sure that those found must have been present simply as foreign bodies. I note also that Mr. G. Murray suggested a comparison with the algal pseudomorphs of sponges described by various authors, but I have seen nothing in the structure of Pontobolbos to indicate any connection with a sponge. Christchurch, N. Z., January 1898. EXPLANATION OF THE PLATES. Puate 26. Pontobolbos manaarensis, gen. et sp. nov. Fig. 1. The largest specimen, side view. Natural size. Fig. 2. The largest specimen, vertical section. Natural size. , pit. Fig. 3. An irregular specimen, formed apparently by union of three growing side by side. Seen from above. Natural size. Fig. 4. A small specimen seen from above. Natural size. p, pits. Fig. 5. Part of a vertical section cut by hand, stained with iodine-green and mounted in glycerine. Drawn under Zeiss A. Oc. 1. Showing the network of darkly staining material (protoplasm ?), the successive surface-layers of the same (s), and the irregular layers of calcareous granules (/). s.1. the outermost layer of the specimen. Fig. 6. Part of a vertical section stained with iodine-green, cut by the paraffin method and mounted in Canada balsam. Drawn under Zeiss D. Oc. 1. Sbowing the darkly staining layers and network and a few sand-grains (9). * For a general account of these fossils, see Nicholson and Murie, Journ. Linn. Soe., Zool., vol. xiv. p. 187. 452 MR. F. CHAPMAN ON HADDONIA, PLATE 27. Fig. 7. Small portion of a very thin vertical section through one of the deeper layers, mounted in chlor-zinc-iodide after soaking in a solution of iodine in potassium iodide. Drawn under Zeiss D. Oc. 3. sh, the empty sheaths of the bacterial rods (stained blue) ; pr, protoplasm ? (stained yellow); 0, fragments of Algse (stained blue). Fig. 8. Small portion of a very thin vertical section through one of the deeper layers, mounted in a solution of iodine in potassium iodide. Drawn under Zeiss F. Oc. 1. ch, chains of bacterial rods (stained yellow); sh, empty sheaths (unstained) ; pr, protoplasm ? (stained yellow). Fig. 8a. Fragment of an Oscillarian found in the same section and drawn to: the same scale. Fig. 9. Small portion of a vertical section, including the outermost layer, cut by the paraffin method and stained with carbol-fuchsin after drying on the slide, followed by iodine and mounted, after again drying, in Canada balsam, Drawn under Zeiss F. Oc. 1. ch, chains of rods;. pr, outermost layer of protoplasm ? On Haddonia, a new Genus of the Foraminifera, from Torres Straits. By Freprrick Cuapmay, A.L.S., F.R.M.S. [Read 18th November, 1897.] (PLATE 28.) Amone the many varied types of the Foraminifera, those which are adherent upon foreign bodies are often of peculiar interest, chiefly by reason of the ability of the protoplasmic body of the animal to wander freely over the surface of the object of its support. Such genera, for example, as Vubecularia, Sagenella, Placopsilina, Bdelloidina, Webbina, Stacheia, Ramulina, Vitri- webbina, Carpenteria, Rupertia, Gypsina, and Polytrema, on first acquaintance, were each with more or less difficulty assigned a well- defined position in the Order to which they belong; but as they became more fully understood they were found to exhibit many points of interest in their plans of growth and the structure of their tests, while in their errant condition they sometimes furnished important data for estimating the inter-relationships of other and more exclusive groups. Many of these adherent Foraminifera find their habitat on coral-reefs and the coral débris derived from them; and the A NEW GENUS OF FORAMINIFERA. A453 specimens now to be described were in like manner found on ceoral-rock. The original block of rock, on which the new foraminiferal type herein described was found, was collected in 1889 by Prof. A. C. Haddon in the Torres Straits, and on this account I propose for it the generic name Haddonia. For the opportunity of, and facilities for, describing the speci- mens I am indebted in the first place to Prof. J. W. Judd, C.B., LL.D., F.R.S., who had previously received the coral-rock for the Geological Collection of the Royal College of Science, through Mr. M. F. Woodward of the Biological Laboratory, who had noted it among a collection of corals therein deposited by Prof. Haddon ; and subsequently, by the kindness of Messrs. R. Kirk- patrick and H. M. Bernard, I have been furnished with another specimen from the Haddon Collection, now in the British Museum (Natural History), which has helped to complete our knowledge of the chief characters pertaining to the test of the genus. Specimen of Coral-rock, showing the new Arenaceous Foraminiferon Haddonia torresiensis, occurring in association with a Polytrema, % nat. size. Happonta, gen. nov. Salient Characters of the Genus.—Test calcareo-arenaceous, adherent, and sinuous; the commencement sometimes straight, sometimes spiral. Chambers imperfectly septate, or openiy labyrinthic. Shell-wall coarsely porous. LINN. JOURN.— ZOOLOGY, VOL. XXVI. 33 454 MR, F. CHAPMAN ON HADDONIA, HADDONIA TORRESIENSIS, sp. nov. (Plate 28.) Test caleareo-arenaceous, surface rough, of a whitish to yellow or brown colour; consisting of imperfectly septate chambers, their breadth being about twice their height, which are here and there subdivided obliquely, somewhat in the manner of Teatularia, but very irregularly, the general plan being a moniliform series of segments. The test commences either with a straight or a sinuous series of chambers, or more rarely with a flat coil of a single whorl, after which the chambers are arranged in a more or less rectilinear manner. The test is adherent to coral- rock, fragments of which, with quartz-grains and organisms such as Discorbina, fragments of Polytremata, and pieces of mollusean shells, are used to construct the sheli-wall. The sinuous manner of growth in Haddonia 1s suggestive of a Serpula- tube, and the shell is often bent upon itself. The interior of the test is smooth or even polished, and partially subdivided by imperfect and curved septa (irregularly labyrinthic). The outer wall of the test is perforated by coarse pores such as are seen 1 Rupertia in the hyaline group; and the salient angles of the imperfect septa, which form flying buttresses on the interior of the proper wall, usually show in section a layer of transparent (hyaline) prismatic shell-structure which is non-perforate. Aperture usually crescent-shaped, sometimes gaping, but more often having a valvular flap, formed by a prolongation of the superior surface of the test, which nearly closes up the orifice as in Valvulina and some of the Milioling. This apertural flap or valve is in some specimens directed towards the distal end of the test (Pl. 28. fig. 1), but in others towards the proximal end (cf. fig. 8); in other words, the organism is sometimes adherent by one side of the flattened and asymmetrical test, and at other times by the relatively opposite side. In the one case the extruding sarcode will be directed towards the surface of attach- ment, in the other away from it. The last few chambers also communicate with one another by a curved slit in the middle of the transverse septal wall, when the septation is more than usually complete. The sarcode when dry is of a rich brown colour (as seen in the recesses of the test of Haddonia in a section), with cavernous or honeycomb structure, and bearing here and there fragmentary sponge-spicules, which are adventitious. Length 4 to nearly § of an inch. On coral-rock, Torres Straits. Luyn. Soc. Journ. Zoot. Vou.XXVI.Pr.a8. Geo.West & Sons ump. hapmian del. ad mat. Parker & Percy lth. we HADDONIA TORRESIENSIS. ' A NEW GENUS OF FORAMINIFERA. 455 Affinities of the Genus.—The specimens above described were discovered on two separate blocks of coral-rock. The organisms appear to have lived on the upper surface of the mass, where they would be exposed to the greater circulation and excitation of the water. The great abundance of specimens of Haddonia may be gathered from the fact of there being 381 distinct speci- mens attached to one of the pieces of rock measuring about 5 X 4 inches *. The genus, more than perhaps any other I have studied, exhibits points of form and structure in common with several other, and presumably distinct, genera. It falls naturally into the arenaceous group of Foraminifera, by reason of the nature of its test-structure, and appears referable to the labyrinthic group of the subfamily Litwoline. The valve-like aperture shows a certain resemblance, if not relation, to the Clavuline and Valvuline groups of the Trxtu- LARIID® ; but although this detail of structure is often of great use in determining the position of the organism, it must be borne in mind that the characters are not always a safe criterion of affinity. The freedom of growth and the undulating contour of Had- donia also call to mind the little-known genus Bdelloidina of Carter +, but that has nothing like so complicated a shell- structure as Haddonia. With regard to the coarse porous nature of the shell-wall, it has been already remarked that in this respect this genus shows a similarity to Rupertia. And, indeed, on examining the dried sarcode of Haddonia under a high power, sponge-spicules were detected in the substance of the protoplasm, which pointed to still further relationship with the group to which Rupertia belongs. The presence of sponge-spicules in the sarcode of Polytrema nd Carpenteria is well-known; and on this account the latter genus was at first believed by its discoverer Dr. J. E. Gray, and also by Dr. Carpenter, to typify an annectant animal form between the Rhizopods and the Sponges. Although these broken sponge-spicules are normally present in the garcode, they are sometimes found embedded in the * Associated with Haddonia upon the same block were other Foraminifera, -as Polytrema, Carpenteria, and Gypsina. Tt Ann. & Mag. Nat. Hist. ser. 4, vol. xix. (1877) p. 201, pl. xiii. figs. 1-8. 33* 456 PROF. A. DENDY ON SOME POINTS IN structure of the test, being presumably caught up in its growth. The probable use of these spicules in Haddonia, as in the other genera mentioned, would seem to be for the stiffening and support of the sarcode, whether within the chamber-cavities or at the points of extrusion, as in Polytrema. EXPLANATION OF PLATE 28. Fig. 1. Haddonia torresiensis, gen. et sp. nov. A typical specimen. Avpertural valve situated on the upper surface of the test. x6. Fig. 2. Haddonia torresiensis. A specimen with a spiral commencement. The end of the test is broken away, and shows the open labyrinthic structure. x6. Fig. 3. Haddonia torresiensis. Apertural end of test, showing valve-like opening; the valve in this instance is situated nearer the attached surface of the test. x9. Fig. 4. Section through the shell-wall of Haddonia torresiensis. 0, the external surface of the test; z, the inner surface; 0, the hyaline shell-layer ; p, the coarse pores in the shell-wall. x60. Fig.5. A fragment of the dried sarcode of Haddonia torresiensis, highly magnified. ¢, irregular cavities in the sarcode, probably due to drying ; s, a sponge-spicule embedded in the sarcode. X332. On some Points in the Anatomy of Caudina coriacea, Hutton.. By Artuur Denpy, D.Sc., F.L.S., Professor of Biology in the Canterbury College, University of New Zealand. [Read 2nd December, 1897. ] (Piate 29.) THE occurrence of small projections arranged, usually in groups, around the anus * has been known for a long time, but compara- tively little light has yet been thrown upon their exact nature. Tt has, indeed, been recognized that these projections are of two kinds, anal teeth and anal papille. The former occur as cal- careous plates or scales, radial or interradial in position}; the * IT use the term anus in accordance with the usual custom, though cloacal aperture would doubtless be more appropriate. + Vide Ludwig’s “ Echinodermen” in Bronn’s ‘Klassen und Ordnungen des Tr-hiereichs,’ p. 141. THE ANATOMY OF CAUDINA CORIACEA. 457 latter as more or less elongated, tentacle-like projections, with or without calcareous deposits in their walls. In the apodous family Molpadiide anal papille have been frequently recorded, and Ludwig appears to have been the first to suspect * that these structures represent modified tube-feet or ambulacral papilla. To Gerould, however, is due the credit of definitely proving + that the five minute finger-like papille, which surround the anus in Caudina arenata, are in direct connection with the radial ambulacral vessels and are indeed homologous with the terminal tentacles of Asteroids and other Echinoderms. ‘Gerould also found a pair of short, blind diverticula coming off from the radial ambulacral vessel or canal just in front of the terminal tentacle, and considered, no doubt correctly, that these indicate the former existence of a pair of tube-feet, which now, however, no longer project beyond the surface of the body. The remarkable genus Caudina is represented in New Zealand waters by C. coriacea, Hutton. In this species Théel described ¢ the existence of five groups of from five to seven papille sur- rounding the terminal anus; one papilla in each group having a very well-developed calcareous skeleton, and possibly representing an anal tooth. He gives, however, no details as to the structure or relations of these organs, and when I first examined the species myself I did not succeed in detecting them §. Since then, however, I have obtained a much more abundant supply of well-preserved material, having found hundreds of young speci- mens || thrown up on the beach at New Brighton, near Christ- church, N.Z., and having also received some interesting fragments of very young specimens dredged by Mr. H. Suter in Lyttelton Harbour. The study of this material has proved that the “anal papille ”? in Caudina coriacea are very much better developed than in C. arenata, and has induced me to offer for publication the following account of these interesting organs. * Loc. cit. p. 100. + Vide Gerould, “The Anatomy and Histology of Caudina arenata.” Bull. Mus, Comp. Zool. Harvard, vol. xxix. p. 164. { Vide ‘Challenger’ Holothurioidea, p. 47. § Vide “ Observations on the Holothurians of New Zealand.” Journ. Linn. Soc., Zoology, vol. xxvi. p. 29. || One of these specimens is represented of natural size in Pl. 29. fig. 13. It jis very small as compared with the adult. 458 PROF. A. DENDY ON SOME POINTS IN A general description of the external form and internal struc- ture of Caudina coriacea has been given by me in my paper on the Holothurians of New Zealand already referred to, while a much more detailed account of the anatomy and histology of the closely allied C. arenata has recently been vublished by Gerould *, so that it is unnecessary to describe the general anatomy in this place. As Caudina is still a comparatively little-known genus, however, I may be allowed to remind the reader that the body has a very peculiar form, consisting of an inflated, ovoid, anterior portion, bearing the mouth and tentacles in front, and a narrow, cylindrical, posterior portion, sometimes known as the “tail,” with the anus situated at its extremity. Fig. 13, Pl. 29, though taken from a young and consequently very small specimen, exhibits the characteristic form of the species. : The extreme tip of the tail of a very young specimen, dredged in Lyttelton Harbour, exhibited the appearance shown in fig. 1. It will be seen from this that the anus (a) is surrounded by five somewhat triangular groups of papille, which I shall pre- sently show to be radial in position. Each group consists of one larger projection, placed next to the anus and forming the apex of the triangle, and (usually) seven smaller papille, of which one is placed exactly in front of the larger projection, while the others are arranged in two diverging rows of three each on either’ side, the most anteriorly placed being the smallest. The five larger projections may have been those to which Théel referred as' perhaps representing anal teeth; and I shall show later on that they are indeed quite different in structure from the others, and that they are apparently present only in very young indi- viduals. As Théel observed, these five larger projections contain a specially well-developed skeleton, consisting of densely packed , irregularly branched spicules. As development goes on, however, they seem to disappear completely, while the other papille may even increase in number, for in one adult specimen I counted as many as nine of them ina group. Figure 2 represents a group of these finger-shaped papille, which I propose in future to speak of as anal tentacles, from a specimen still quite small, though apparently older than that from which fig. 1 is taken. There are only six tentacles in this group, one of the most * Loe. cit. THE ANATOMY OF CAUDINA CORIACEA. 459 anterior being on one side presumably as yet undeveloped. It will be seen that the largest tentacle is placed most posteriorly in the middle of the group. It is strictly radial in position and evidently corresponds to the terminal tentacle of Gerould, being homologous, as that author has pointed out, with the terminal tentacle of Asteroids. It will also be seen that there are no suctorial disks and no end-plates, but that the larger tentacles, in this respect unlike those of C. arenata, contain irregularly branched spicules scattered in their walls. In the specimen from which fig. 2 is taken no trace was visible of anal teeth, the five projections which are supposed to represent them having com- pletely disappeared, nor have I found them in any but the very youngest individuals. The specimens in which Théel discovered them appear to have been mere fragments of tails; and we must assume either that these also belonged to extremely young individuals or that he was referring to the five largest anal tentacles, unless it be that the teeth persist in some individuals and not in others. Having thus described the external appearance of the organs surrounding the anus, we may pass on to investigate their relations to the internal structures. This may be done by means of a series of transverse sections of the so-called tail of one of the youngest specimens. The general arrangement of the parts concerned is shown in Pl. 29. fig. 38, which represents a transverse section taken at some little distance in front of the anus (the 80th section from the posterior extremity). It will be seen that the integument is here very thick, and that the cloaca is attached to the body-wall by very numerous radi- ating bands of muscle. These deserve a brief notice. Hach consists of a hollow cylinder of fibres running lengthwise side by side, and covered externally by a thin layer of epithelium containing many conspicuous, darkly-staining nuclei. Gerould describes in Caudina arenata certain muscle-cylinders which run outwards through the connective-tissue layer of the integument and terminate immediately beneath the epithelium. He points out that the striking resemblance of these muscular tubules to the vessels which connect the ambulacra with the radial canal, led Semper to the erroneous conclusion that they are actually in connection with the canal. Gerould himself 460 PROF. A. DENDY ON SOME POINTS IN found no such connection, but found the muscle-tubules or cylinders directly continuous with the transverse muscles of the body-wall. He suggests, however, that they may be rudimentary ambulacral vessels, the central ends of which have lost their » primitive connection with the radial canal and have secondarily become united to the transverse muscles of the body-wall. He inclines towards this view on account of the arrangement of the fibres in a hollow cylinder, but this hypothesis appears to be now completely negatived by the discovery of a similar tubular structure in the radiating muscles which connect the cloaca with the body-wall in Caudina coriacea. The radiating bands of muscle to a large extent obliterate the body-cavity in the tail-region, but a portion of the coelom remains conspicuous as a distinct canal (figs. 7, 8, ¢.c.) surrounding the inner aspect of each of the five radially-placed longitudinal muscles. These longitudinal muscles (7.m.), as usual, form a very conspicuous feature in transverse sections. Hach is composed for the greater part of its length of two almost separate halves running side by side, separated from one another by a deep fissure continued outwards from the ccelomic canal (fig. 7, ¢.c.) just mentioned. The surface of the longitudinal muscle is covered by a distinct layer of epithelium which exhibits a very marked thickening on its inner aspect, as shown in fig. 7. This thicken- ing may be due to the contraction of the muscle having thrown the epithelium into folds, but if so, it is difficult to account for its occurrence on the inner aspect of the muscle only. I am inclined to think, however, that it is more likely the expression of a mass of proliferating epithelial cells from which the very numerous corpuscles of the coelomic fluid may be derived. Fig. 7 shows four of these corpuscles (corp.) in the ccelomic canal at the side of the longitudinal muscle. I have already described them in some detail in my “Observations on the Holothurians of New Zealand.” * Immediately on the outside of the longitudinal muscle-band lies the radial ambulacral canal or vessel (figs. 7-9, r.a.), lined by distinct epithelium, and with a layer of longitudinal muscle- fibres on its outer aspect. Outside this lies the radial blood-sinus or lacuna, which is extremely small and ill-defined, often not recognizable as a distinct cavity, but shown pretty clearly in some sections (figs. 8, 9, 7.s.). Outside this again lies the radial * Page 31. THE ANATOMY OF CAUDINA CORIACEA. 461 nerve-cord with its associated celomic canals *. The nerve-cord is as usual divided into inner and outer bands, separated from one another by a thin layer of connective tissue. In the inner band (fig. 7, 7.5.) the nerve-cells are aggregated on the inner aspect, and in the outer band (fig. 7, 0.b.) they are aggregated on the outer. Between the inner band and the blood-sinus lies the hyponeural ¢ ceelomic canal (figs. 7 & 9, hyp.). Outside the outer band lies the epineural ccelomic canal (ep.). If we trace these various radially-placed organs to the hinder extremity of the body, we find that they terminate respectively as follows:—The longitudinal muscle-band gradually diminishes in thickness, the fissure between its two halves disappears (fig. 8) and the muscle then dies out (figs. 5, 9, 10, 11), its place being taken by bands of transverse fibres (figs. 5, 10, 11, ¢.m.) placed in the radii around the anus, and continuous with the inter- radially-placed transverse or circular muscle-bands. Just posterior to each radially-placed transverse muscle-band the integument projects to form a supposed vestigial anal tooth, which consists of a solid mass of connective tissue containing spicules and surrounded by the epidermis (figs. 1, 6, 12, 7.). The radial ambulacral vessel, after the termination of the longitudinal muscle, increases greatly in diameter (fig. 9, 7.4.) and gives off two or three wide branches on each side (figs. 5, 10, 11, amp.). The main trunk and its branches terminate respec- tively in the central and lateral anal tentacles of each group (fig. 12). Just before it gives cff its lateral branches, the radial ambulacral vessel appears to lose the layer of muscle-fibres from its outer aspect (fig. 9), but these appear again after the division has taken place, as a very distinct layer of longitudinal fibres on the outer aspect of each branch (fig. 11, mf). At first the swollen branches of the ambulacral vessel appear to partake somewhat of the nature of ampulle, and they probably function as such in controlling the protraction and retraction of the anal tentacles, by forcing liquid into them or causing it to be with- drawn. It will be seen from a comparison of figs. 11 and 12, that in the specimen from which the sections were taken, these * Tt seems pretty safe to assume that these canals are ccelomic, although I believe that their derivation from the body-cavity has not been definitely established. + “ Pseudohemal canal” of some authors, ‘“ hyponeural canal” of Gerould. 462 PROF. A. DENDY ON SOME POINTS IN circumanal ampulle* were in a state of dilatation while the tentacles themselves were in a state of contraction. It will also be seen from figs. 5 and 11 that the cireumanal ampulle contain a large number of nucleated corpuscles (corp.), which shows that they must have been filled with liquid at the time of the animal’s death. In the contracted anal tentacles themselves the ambulacral vessels appear as narrow cavities surrounded by an inner epithelium, outside which lies a layer of very slender longitudinal muscle-fibres difficult to recognize, and then a very delicate outer epithelium (fig. 12). I have not been able to trace the radial blood-sinus or lacuna beyond the point where the ambulacral vessel gives off its lateral. branches. Gerould, however, in the case of Caudina arenata,,. states that the radial blood-lacunze communicate with a circular lacuna surrounding the anal opening. He also figures ~ what appears to be an immense dilatation of the radial blood-lacuna,,. occupying almost the entire cavity of the anal papilla or terminal tentacle. Up to the pomt where the radial ambulacral vessel gives off its branches to the anal tentacles, the radial nerve maintains its: typical structure, consisting of an inner and an outer band accompanied respectively by the hyponeural and epineural canals (fig. 12). Shortly after the ambulacral vessel divides, however, the inner nerve-band dies out, leaving only the outer band, which also divides into branches, one of which accompanies each branch of the ambulacral vessel into an anal tentacle (figs. 11, 12). The epineural canal (fig. 11, ep.) also divides, and its branches accompany the corresponding nerves for some distance, but afterwards disappear. At first also each branch of the ambulacral vessel, where it swells out to form an ampullary organ, is accom- panied on its outer aspect by a distinct canal, lying outside the layer of longitudinal muscle-fibres and inside the nerve (fig. 11). Whether these canals are derived by a branching of the hypo- neural canal or by sudden enlargement and branching of the radial blood-sinus, I have been unable to determine owing to * As I am not aware that ampullary organs have hitherto been described in this position, I have ventured to distinguish them by the term “circumanal ampullz” from the numerous other ampulle which occur in the body of a. Holothurian. t Loc. cit. plate iv. fig. 50. THE ANATOMY OF CAUDINA CORIACEA. 465 the minuteness and ill-defined character of the parts in question. Just before the inner nerve-band dies out, there appears to be only a single space between it and the circumanal ampulle (fig. 10). This may be a continuation of the hyponeural canal or a sudden enlargement of the radial blood-sinus. It is a noteworthy fact that it may contain numerous corpuscles (corp. fig. 10) similar to those in the adjacent ampulle. The chief conclusions arrived at in this paper may be summed up as follows :—Caudina coriacea possesses two kinds of papille at the hinder extremity of the body around the anus. (1) Five blunt radially-placed projections, which contain abundant spicules and are mere solid processes of the body-wall, without nerves or ambulacral vessels. These projections are apparently present only in extreme youth, and may represent anal teeth in a ves- tigial condition. (2) Five radially-situated groups of anal tentacles, containing branches of the radial nerves and of the radial ambulacral vessels, and with loosely seattered spicules in: their walls. The branches of the ambulacral vessels are swollen out to form circumanal ampulle, the function of which is evidently to assist in protraction and retraction of the anal tentacles. The anal tentacles are doubtless homologous with the tube- feet of typical Holothurians, which have undergone a change of function and now serve as tactile organs. Gerould has described how Caudina arenata often lives buried in the sand, with only the tip of the tail projecting. Under these circumstances one can well understand the development of sensory tentacles around the aperture, which is not only an anus but also the entrance to the respiratory organs. Caudina coriacea doubtless has a. similar habit, and this peculiar mode of life may also serve to: explain how it was that the dredgings of the ‘ Challenger’ in Cook Straits and of Mr. Suter in Lyttelton Harbour only brought up fragments of the “tail,” these having apparently been cut off by the dredge while the body remained buried in. the sand or mud. Christchurch, N.Z., August 27, 1897. AGA Fig. Fig. Fig. Fig. Fig. Tig. Fig. Fig. Fig. Fig. Fig. Fig. Fig. amp. C.C. cl. cm. corp. ep. hyp. 2.0. mf. 0.0. 1. ?.Me rn. TS. Sp. z. td. ten. tm. t.7. 1. 2. PROF. A. DENDY ON CAUDINA CORTACEA. DESCRIPTION OF PLATE 29. CaupIna corraces, Hutton. Posterior extremity of a very young specimen viewed as an opaque object, showing the anus surrounded by the five groups of “anal papilla.” (From a “tail”-fragment dredged in Lyttelton Harbour.) x 380. Group of anal tentacles of an older but still young specimen, viewed as a transparent object. Zeiss D. Oc. 1. (Figs. 8-12. Transverse sections of the “tail” of fig. 1, numbered from the posterior extremity forwards.) . The 80th section. Zeiss A. Oc. 1. . The 18th section. Zeiss A. Oc. 1. . The 8th section. Zeiss A. Oc. 1. . The 3rd section. Zeiss A. Oc. 1. (Reversed as compared with fig. 1.) . Radial portion of the 79th section. Zeiss D. Oc. 1. . Radial portion of the 18th section. Zeiss D. Oc. 1. . Radial portion of the 13th section. Zeiss F. Oc. 1. . Radial portion of the 10th section. Zeiss D. Oc. 1. . Radial portion of the 8th section. Zeiss D. Oc. 1. . Radial portion of the 3rd section, showing only a group of anal tentacles and a vestigial anal tooth. Zeiss D. Oc. 1. . Entire young specimen from New Brighton Beach. Natural size. Reference Letters. Ampullary dilatations of the branches of the radial ambulacral vessels (circumanal ampulle). Ceelomic canals partially surrounding the longitudinal muscles. Cloaca. Circular muscles. Corpuscles. Epineural canal. Hyponeural (pseudohzmal) canal. Inner band of radial nerve. Muscle-fibres cut across. Outer band of radial nerve. Radial ambulacral vessel. Radial longitudinal muscle. Radial nerve. Radial blood-sinus. Spicules. Supposed vestigial anal teeth. Ambulacral vessel of anal tentacle. Anal tentacle. Transverse muscles. Nerve of anal tentacle. Note.—The sections were cut from a specimen treated with borax-carmine and. acid alcohol, and the spicules are therefore not shown. ee ray A Dendy del. A.R.Hammond Iith. Tae Soc. Journ. Loon. Vou. XXVI. Pl. 28. 2 3385 aes eo 2904 Oe, No i x f West,Newmanimp. ei an = ys! i» Lusw. Soc. Journ. Loon. Vou. XXVI. Pl. 28. “\; Ai Nie =| 3 3 y ix e | A Dendy del. AR Hammond lith. PROTECTIVE COLORATION IN THE HOUSE-MOUSE. 465 On a probable Case of Protective Coloration in the House- Mouse (Mus musculus, Linn.). By H. Lysrer Jameson, B.A. (From the Biological Laboratory, R. College of Science, London.) * [Read 2nd December, 1897. ] (Puate 30.) Tue colony of Mice I am about to describe was found by me in October 1895, inhabiting the sandhills on the North Bull, Dublin Bay. This tract of sandhills runs along the coast for about three miles, from Clontarf to Sutton, on the north side of Dublin Bay. A tidal channel of considerable breadth, averaging, I should say, about a quarter of a mile across, separates the sandhills from the mainland at high water, giving place to extensive mud-flats at low tide. A narrow strip of water about 20 yards wide, however, always intervenes, fed by a couple of small streams on the Dollymount shore. A bridge has been built connecting the west end of the island with the mainland, and here also is the North Bull wall, a breakwater running out across the mouth of the Liffey for about a mile, built in 1823. A coastguard station, a cottage, and a golf club-house are the only buildings on the island. On the 6th of October, 1895, when walking on the sandhills, I observed some mice running about that harmonized strikingly in colour with the sand, appearing to be of an unusually pale tint. I had the good fortune to capture one of these, and found it to be a pale example of Mus musculus, Linn. En- couraged by this discovery, during the following week and on various subsequent occasions as opportunity allowed, I set “.Cyclone” traps and was rewarded by obtaining thirty-six specimens, a very large proportion of which were considerably paler than the typical Mus musculus as it occurs in houses and farmyards in Ireland and England. These specimens, as might be expected, varied somewhat in shade of fur among themselves, and my series shows every gradation from the typical Mus musculus to the individual which I have numbered 380 in the following list and in my collection, that being the most extreme case of pale coloration. Striking as is the light colour of the dorsal surface in this race, it is still more unlike Mus musculus typicus in the ventral fur, which in the majority of cases is pale. * Communicated by Prof. G. B. Howes, Sec. Linn. Soc. A66 MR. H. LYSTER JAMESON ON buff or yellowish white, resembling that of some of the wild subspecies that have been described from various localities (see Thomas 1, Barrett-Hamilton 2), rather than the smoky grey with which we are familiar in the domestic race. In dimensions this variety agrees with Mus musculus, as will be seen from a perusal of the following table, in which the speci- mens have been arranged according to colour, No. 1 being indistinguishable from the domestic form, and No. 30 the palest of all. Nos. 31-86 are also very pale, but have been separated from the rest as immature. Dimensions. Head and No. Sex body. Tail, Hind foot. Har. mm. mm. mm, mm. AL SOLE TES SON 90 76 18 QS Gees: sg dual 82 75 18 Saige OMB AB amen eee 78 75 19 A amr Sere auc 87 75 18 BAO AEE ts telantehs Measurements lost. GrergG napareeuidieeoum eye KG 74 17 AT aire eaten en atage 75 78 18 BEG resin cans arene arate Measurements lost. ORGS ONE HOT 80 82°5 185 Oats Secpershs, Sameteuslede ise 69 76 16 12 TN Leese eee ee 83 87 18 13 TDN annals ici 63 69 17 11 Sed hue estore wie eee role 80 74 17 13°5 A hus gttehshoae eke e mente Measurements lost. MO eteintccrrertatoieds 65 17°5 MGR GY SOR h 725 71 16 Lae tO wash ntre aremeieyels 94. 95 16:5 135 18. Label lost. AUG ist SOR ae titeetaae hola rs 80 80 ily 12 QO Mids Lee eae 80 78 17 12 GALA ORY AR eri eated Sey 75 80 16 12 Dee Oat san eens ralereen ss 88 84 17 13°5 2 ONO OMEN recent a 78 79 18 12 DAR UO: SCRUM pict 88 79 17 13 HAT Loi ESCO Naan ae 82 82 17 13 QO Gi sise Norrenin oe 88 76 165 18 DTP TOR CS Mech atateds els 84 86 18 12 QEi WEG: eerenleradiayeens 64 66 18 12 DOG pysuauier gem csaeireat: 80 75 17 12 SO Near humeuetetteeel aiectate 75 73 17°5 SLO ae tonier 45 50 16 7 SOE EG Bisuetoun syste teens 78 83 i/ 12 Soe pire euelensutels cr eveve 45 48 16 if BAY Gian eae en Measurements lost. Sha NSN OL ee 60 56 155 SGddinGug. tie ee cmt 52 515 155 18 Oolour.—Nos. 1-5 are quite as dark as average specimens of PROTECTIVE COLORATION IN THE HOUSE-MOUSE. 467 Mus musculus from the mainland of Ireland and from Great Britain. Some of the specimens I compared them with were trapped in corn-fields in the Hastern Counties of Ireland, and are certainly not ‘“smoke-stained ” as is so often the case with mice caught in houses. From No. 6 onwards they grow gradually lighter, both dorsally and ventrally. No. 9 shows a distinctly rufous colour dorsally, but has the grey belly of the House- mouse; No. 10, but slightly lighter than No. 9 dorsally, has a very white belly ; and from this point onwards through the series the mice are all various shades of rufous or fulvous grey dorsally, and with the exception of one or two, that retain traces of the orey belly of their ancestors, they are pale buff ventrally. In Nos. 20, 23, 24, 29, and 30 the grey bases to the hairs on the ventral surface have almost or completely disappeared. The specimens numbered 10 to 36, z. e. all the markedly pallescent individuals, can be broadly described in the following terms :— Dorsal fur much paler than in the domestic type; rufous grey or fulvous grey is the usual colour. Ventral surface pale buff, separated from the darker dorsal fur by a more or less clean-cut line. The slaty-grey baxes to the hairs which characterize Mus musculus typicus are usually, but not always, reduced both in shade and in extent. All over the body, on the ears and tail, the hairs are generally lighter and more fulvous than in Mus musculus typicus; the ears, nose, and tail are also lighter. The feet are white, or pale buff, in all cases ; in the ordinary House- mouse they vary from smoky-grey to white. The claws are flesh-coloured ; the eyes dark brown. Many of these mice, how- ever, do not fulfil all these requirements, intergrading in some one or more characters with the darker mainland form; but everybody will admit the difficulties that beset one in describing an unstable aberration that has not yet become stereotyped into a constant subspecies or species. Notwithstanding the perfect intergradation, the mean colour of this race is not only lighter than that of the type, but is lighter than most, if not all, of the light-coloured subspecies of Mus musculus which are preserved in the British Museum. During October 1875, I made the following observations on the habits of the mice. They live in burrows in the sand, often of a considerable depth. ‘The nest is made in a chamber of such a burrow; I dug out three of these nests, all of which were from two to three feet from the entrances, and had several A468 MR. H. LYSTER JAMESON ON passages leading from them. ‘Two of them were made entirely of blades of bent-grass (Ammophila [Psamma] arenaria), the third of tidal refuse, straw, paper, and feathers. I found one - mouse in a hole only a few inches deep, outside which was a heap of freshly thrown-up sand; there can be no donbt it was in the act of burrowing, which proves that they make their own holes ; indeed, on the North Bull there is no other mammal to make them. This fact is of interest, as, if MZws musculus in the wild state uses the burrows of other animals as Thomas. suggests (1), we have here a habit that has been acquired by these mice since they colonized the island, to enable them ine better to escape from their enemies. They are very numerous; the sandhills are in places riddled with their burrows, and the island is probably stocked with as: many mice as it can support, for I found in several cases the mice I trapped had been devoured more or less completely by the survivors. Presumably they live on the seeds of such plants as grow on the sandhills, and on the varied diet of city offal cast up on the sands from the mouth of the Liffey. At dusk I more than once observed mice among the refuse left at high-water mark. They are probably quite as prolific as their darker domestic ancestors; the uterus of one female contained nine feetuses. The numerical increase must be very much accelerated by the absence of terrestrial scent-hunting enemies. Running very rapidly over the sand, these mice are almost impossible to follow with the eye, so perfectly do they harmonize with their surroundings. When we come to enquire into the factors which have led to the evolution of this race, we have not far to look. The short- eared owls, which in autumn and winter always frequent the sandhills, as well as the numerous hawks that come over from the mainland, and can be seen any day in pursuit of prey on the North Bull, most readily capture those mice which contrast most strongly with the sand and the arid vegetation peculiar to such places; in short, they pick out the darker mice. The very scanty cover, and the probability that the mice have to travel some distance for their food over bare sand, make the dangers. to dark and conspicuous mice much greater. Under these circumstances, coupled wita the fact that the hawks and owls hunting by sight are the only enemies these PROTECTIVE COLORATION IN THE HOUSE-MOUSE. 469 mice have to cope with, it is very natural that a protectively coloured race should be gradually supplanting the dark one that ‘gave rise to it. Isolation, no doubt, has also been a very important factor in intensifying the effects of competition; the absence of direct communication with the mainland, and the consequent impossi- bility of frequent immigrations of dusky specimens from the houses on the adjoining shore, have allowed Natural Selection to carry on its weeding-out of unfavourable variations without disturbances of any kind. The immigrations (if any) from the mainland, by swimming or by the bridge (which is made of open woodwork and of considerable length), must be so very exceptional, that their effects on the colony need not be con- sidered. The alternative explanations which might be suggested do not seem to me very hopeful. The objection that there may be no causal relation between the factors I have enumerated and the paleness of the mice, can only be established or refuted by a study of the mice inhabiting similar sandy islands elsewhere. Should similar races be found on such islands, the relation would be at once established. That this colony may represent the descendants of some foreign subspecies, introduced in a ship, is unlikely when we remember the perfect intergradation between our variety and the type. It was suggested to me that in Ireland we might have a wild form of Mus musculus, as well as the introduced domestic race that alone occurs in Great Britain. The known subspecies have recently been enumerated by Barrett-Hamilton (2). First, comparing the Clontarf mice with Thomas’s wild race from Portugal, the dimensions of the former agree with Mus musculus, not with Thomas’s race (1). Specimens of us mus- culus bactrianus (Blyth), or a closely allied form, in the British Museum, differ from the Clontarf mice in that their paleness. is due to a predominance of white tints, not of yellow. Mus musculus flavescens (Fischer) and Mus musculus spretus (Lataste), which according to Barrett-Hamilton probably intergrade, seem to come nearest to our race. I have never seen a typical example of either of these subspecies, but Southern European mice in the British Museum, which are probably referable to one or other of them or to something intermediate between them, are not half so light as my extreme cases. In fact, the lightest mice in my LINN. JOURN.—ZOOLOGY, VOL. XXVI. 34 470 MR. H. LYSTER JAMESON ON series are as much lighter than any of the above subspecies as are the latter than the type! Then it was suggested to me that, as in certain districts in the Eastern counties of Ireland a buff-coloured hare (Lepus variabilis) occurs, these mice might be a parallel case. In the first place, I have never seen a buff-coloured mouse from the mainland of Ireland, and, in the second place, the variation in the hares is apparently a “sporadic sport.” Mr. Williams, the Dublin taxidermist, tells me that he has taken typical leverets and buff ones from the same uterus in hares sent to him for preservation. I have never seen any intergradation between the two colours in the hare. And even if the ultimate cause of the pale coloration in the mice was traced to the same circumstances (whatever they are) which cause the variation in the hare, we have still the fact before us, attested by a comparison of specimens from the mainland with those from the North Bull, that the mean colour of the North Bull specimens is infinitely paler than the mean on the mainland. I think, however, that the very great variability of this new race is the strongest evidence of its being a recently evolved variety of Mus musculus typicus, which has not yet settled down into the comparative stability which usually characterizes a species. When I came to enquire into the probable age of this colony, I was confronted with some very remarkable topographical evidence. Being convinced that these sandhills had a very recent origin, I referred to such old maps and charts of the Dublin Coast and County as were contained in the National Library of Ireland, and I owe a debt of gratitude to Mr. J. De W. Hinch for the valuable assistance he gave me in hunting up these records. A detailed account of the topography, past and present, of the North shore of Dublin Bay would be out of place here; and as my friend Mr. R. Lloyd Praeger is at present collecting evidence on this subject, with a view to investigating the origin of the flora of the North Bull, I will content myself with giving such facts as point to the date at which the island first appeared. I regret that I cannot give the exact dimensions of the North Bull, it has increased considerably since the last edition of the Six-inch scale Ordnance map was published (1867). Mr. Praeger PROTECTIVE COLORATION IN THE HOUSE-MOUSE. 471 hopes to have the island surveyed thoroughly before publishing his paper. I estimate its present length at about 22 miles. The 1848 edition of the Six-inch Ordnance map figures the North Bull as a bank one mile and fifty chains long, the main island being only seventy-five chains long and separated by a patch of sand, covered at high water, from several small islands entered as “ Islands in Raheny Parish.” These were known as the “ Bull Caives.” The tidal channel is now dry, and supports the usual sand-loving plants that characterize the rest of the island, the smaller islands having united with the larger one. The islands are figured similarly in the 1838-1868 sailing chart. In 1839 Beaumont, in his “ Report upon a Design for the Improvement and Deepening of Dublin Harbour,” speaks of the “Young Bull Calves”? forming seaward of the old one. Nimmo, in his 1823 map, figures a “ sand-bank”’ one mile and thirty chains long, which is probably the island. He gives no details. In Taylor’s two maps, 1816 and 1819, the island is figured about + mile long according to the scale. Dawson’s map, 1809, is a small map in which the island is figured as a small oval patch of land a little under 3 mile long. The disagreement between these early records perhaps speaks more for the inaccuracy of the maps than for the variability in relation of land and sea. In 1805, in the “Report on the Improvements of Dublin Harbour by the Directors General of Inland Navigation” (Tidal Harbours Commission), there occurs the statement, “ Since the building of the South Pier [about 1790]... . a considerable strip of the North Bull remains dry at high-water, and has on it a growth of marine plants.” The first map that figures this island is Cowans, 1800, where it is delineated as a small dry area on the submerged sand-bank. Bligh’s map, of the same date (in Warburton’s ‘ History of the City of Dublin,’ 1818), figures it as a quarter of a mile long. In the Dublin Harbour Improvement Commission (1800-1801), a letter is published from Richard Broughton to William Gregory, dated from the Ballast Office. Broughton refers to “the dry part of the North Bull;” and further on he says: “The North Bull has increased very much both in extent and in height during the progress of the works on the South side, and more especially since their completion.” (This refers to the building 34% —— 472, MR. H. LYSTER JAMESON ON of the South wall from the Pigeon-house to Poolbeg, which took place about 1790.) Rocque’s map, revised by Bernard Scale (1773), is a very careful and elaborate chart of the environs of Dublin. The- North Bull Island is not figured, and certainly would have been if it existed, as the map goes into very great detail. Though the South wall is not yet built, the Pigeon-house and Poolbeg are connected by a row of stakes, which Haliday informs us (‘ Sean- dinavian Kingdom of Dublin,’ pp. 235-236) were put there in 1717. It is needless to add that none of the many maps before this date that I have examined, going as far back as 1610, show any land, though the submerged bank is generally figured, and in some stated to be exposed at low tide. On the other hand, Dalkey Island and “Clontarf Island” (now, I believe, part of the mainland) are figured in most of them. McKenzie’s chart, 1775, does not bear a figure of the island. Evidently between 1775 and 1800 the island first arose, and unfortunately between these two dates I cannot find any maps. If the South wall was not built until 1790, it is probable that the island (if it existed before that date) was too small to support a colony of mice until the wall was finished; especially when we consider Broughton’s statement in 1810, when the island was only mile long. These data allow us about 100 years for the evolution of the pale race, with a maximum limit of 120 years ; and certainly, when one considers the exceptional circumstances, one cannot but think that a century or less would be more than sufficient to produce the form with which we are dealing. A very interesting observation, which emphasizes the import- ance of Isolation in intensifying the effect of Natural Selection, has. recently been recorded by Kane (3). He discovered, on a small island off the South-west coast of Ireland, a melanic race of the Geometer Camptogramma bilineata, which he names var. ¢solata ;. and he holds that this melanic form owes its existence to the weeding-out of the light ones by the Gulls, Pipits, and Bats that frequent the island, and that the darker ones, harmonizing with. the colour of the rocks, escape more readily. On the adjoining mainland, as well as upon other parts of the coast where the rocks are of a dark colour, more or less melanie forms occur, but along with the type; but on this particular island the type is not to be found. I do not myself know any case of the kind where a protective ‘ T Doc, JOURN SS son. Vol. XXNVI1. PI PROTECTIVE COLORATION IN THE HOUSE-MOUSE. 473 adaptation can be pointed to in which we have time data, in which we can fix the maximum possible age of the variety; but very probably the dark Lepidoptera of the great Factory districts of England present such a case, and owe their existence to the smoky surroundings. For an excellent summary of records of this kind in Insects, see Carpenter (4). The study of such cases is a most fascinating one, as, owing to the limited number of factors at work, we may be able to trace, step by step, the causes and their modifying effects upon the species, and so to clear up many points as to the details of the process of elimination known as Natural Selection. PAPERS REFERRED TO. (1) Tuomas, O.—‘ On a Wild living Mouse of the Mus musculus group in Portugal.” Zoologist, 38rd ser. vor, xx. p. 137, 1896. (2) Barrert-Hamitton, G. E. H.—‘On Wild Forms or Sub- species of Mus musculus.” Zoologist, 8rd ser. vol. xx. p- 178, 18986. (3) Kane, W. F. pr V.—“ Observations on the Development of Melanism in Camptogramma bilineata.” (Report to the Fauna and Flora Committee of the Royal Irish Academy.) Trish Naturalist, 1896, vol. v. p. 74. (4) Carpenter, G. H.—“ Colour Changes in Insects.” Natural Science, vol. 11. p. 287, 1892. EXPLANATION OF PLATE 30. Varieties of Mus musculus. The dark individual on the right-hand side is No. 4 in table (p. 466), tne individual on the left is No. 27; that in the centre No. 30. ATA DR. W. G. RIDEWOOD ON THE LARVAL HYOBRANCHIAL On the Larval Hyobranchial Skeleton of the Anurous Batrachians, with Special Reference to the Axial Parts. By W. G. Ripewoop, D.Sc., F.L.S., F.Z.S., Lecturer on Biology at St. Mary’s Hospital Medical School, London. [Read 20th January, 1898.] (Prate 31.) INTRODUCTION. THE observations recorded in the following pages were made with the especial object of ascertaining whether the peculiarities of the larval hyobranchial skeleton of Alytes are common to all the genera of the family Discoglosside, and whether they occur in any genera not belonging to this family. The peculiarities in question consist in the presence of an anterior copula, and the complete separation of the hypobranchial plates by the posterior copula, which thus extends back to the laryngeal sinus, as already fully described in a communication recently presented to the Zoological Society, and appearing shortly in the ‘ Proceedings.’ Tadpoles of twenty-one species were examined, belonging to nineteen genera; and for all of the specimens I am indebted to Mr. G. A. Boulenger, F.R.S., whom I have always found most ready and anxious to encourage morphological enquiries of this. kind, and to whom my thanks are hereby gratefully tendered. Considering how few of the species of Anura are known in their larval stages, the following list will be admitted to be very complete and representative. The three numbers of the formula appended to each name stand respectively for the distance in millimetres from the snout to the root of the tail, the length of the tail, and the length of the hind limb. No measurement is given of the fore limb, because, in order to make the comparison between the different species as perfect as possible, the tadpoles were all selected at that stage in which the larval hyobranchial skeleton is most com- plete and most characteristic. This stage is recognized externally by the small size of the hind limbs, and by the fact that the fore limbs are not yet extruded through the atrial wall. SKELETON OF THE ANUROUS BATRACHIANS. ATS Ranide. Rana esculenta. 18. 24. 6 Rana temporaria. 12. 22. 4 Rana Whitehead. LOS a2% Oxyglossus levis. ie 29S Rhacophorus leucomystax. 16265) 10 Phyllobates trinitatis. 5. 8& O Engystomatide. Microhyla ovnata. (oe a Cystignathide. Pseudis paradoxa. 41. 82. 28. Telmatobius marmoratus. 35. 45. 7. Chiroleptes platycephalus. 25. 34. 5. Calyptocephalus Gayt. 41. 54. 9. Bufonide. Bufo vulgaris. 12. 17. 3. Hylide. Hyla arborea. 1222) 18: Pelobatide. Pelodytes punctatus. 13. 162 U1. Pelobates fuscus. 35. 60. 16. Leptobrachium Hasselti. PA 272 9: Discoglosside. Discoglossus pictus. SU Maes alt: Alytes obstetricans. 20. 35. 4. Bombinator igneus. 16. 20. 6. Dactylethride. Xenopus levis. 25. 35. 6. Pipide. Pipa americana. OP 12S V4 GENERAL Part. The results of the investigation may be summarised in the following words:—The two fundamental peculiarities of the larval hyobranchial skeleton of Alytes above specified are found in the tadpoles of Bombinator and Discoglossus, but not in any of the other genera examined. This discovery is a lasting tribute to the perspicacity and intuition of the late Prof. Cope, who first brought together in taxonomic relationship the genera Alytes, Discoglossus, and Bombinator, which had previously been relegated 476 DR. W. G@. RIDEWOOD ON THE LARVAL HYOBRANCHIAL to three distinct families. Without entering into a discussion of the question whether the larval stages of Anura represent a true recapitulation of ancestral evolution, the similarity in the structure of the larval hyobranchial skeleton in these genera can, I think, be safely taken to indicate close genetic relationship. The position of the anterior copula of the Discoglosside is in almost all the other forms examined occupied by loose connective tissue, the middle or hinder part of which is firmer and has the form of a transverse ligamentous band (fig. 2, 7) connecting the ceratohyals. Gaupp, in his description of Rana tadpoles (4. p- 404), alludes to this band as being partially cartilaginous. Although I have failed to recognize anything approaching the structure of cartilage in the ligament, I fully concur with this author in regarding it as the morphological equivalent of the anterior copula of