; i i ay? Re Nes } 3p OG aRN i 1 ia ie 4) Fy aren & INH ARVn tp tA! \ AAPARA Mf f= { i z=) } ~'/ lg ON YW W - VN \/ \ - ie A VA AANaa a AAA PARA AA lpuy, AAS VA A PAPA NY 1 A a é a ( ( CC CC CG re CC (@ ~\iF— WN as sy if ‘ @n\ AN AaaaAAArr AAV Na\a\a'a A RAR aN AN \ aan aa) all 4 RAAR | 2 A Ar | = \ \ lg\if — ii ms \i| oN A L=~ AN b | } - a Boas VAN} Ae a AiAaAR AY e.\, : lanl Y. a\. \ rN y } WN ANY \ VF \ENIAN FAA aa | Ga) { | / PAAR AAeeeArARRARAAAaanerannRanrnnccnaana i ALAAAA A\f aha mr f = F\{ oN N\f AEN ANANNONON A At AAAAY ae AY a aA y\ VAN \ AA iA AA’ AR ANA AY j | i | =~ | gan 4 nl) \ ! 1 ‘ \ | 1 | | A f [ f - / i i \ \ an| \ i aa -\ 1 ‘f \ = Ve y Y= V¥— Vi & f Nf { / law A ry a a) aN = =|/- DN i—\\/ aN es g \ SWAIATA = Si N\ a FN I~ || \ —~ ANIA I i NT ; 4 y W >, é i \/ YSN y NAN = FONE MW V Y yy \A\ WENN owl VN f\ \ | | i —Ni = i — \ mE \ | \ . Ni}. y - —\ \ A em = Vr | \ \ ae UN gal \ PAF | AVA AN : =i WA (a\an\ 2 } - \\ AIT AY | AL \ S WAWA APY | / \ PA\f Wi lA \A y ENN =\\/ { y ’ | { \ al lala | |I~aABAAAAA AA AAARARAAARARS | | a A) WV A\AVA'a| i NTN APA AN) WA PAY a & | ; PN AI NAV a ; FN } 4 j | f - _¥— AA Va aes if \ | ! ) aan ae Y | S | la r 7 \ ! y | i- if - ‘a 3 i AW A = | \ | g Wey f & é > ' PN ! | If lr f Z } la lan\ | ~\\l- =I { - ~ 4 iN | | \ - if ! = AF | } A) Aig | » \ ENN FAIA A-NIANIANIANIA! WAVES / i WA VW WY WWW \ i a ANY NAA \} Si AB Ran! a VEN NE A ae eay (a BAaRF-RY \ | LAY : . \ A -~ i an | q «CG w (CE | C& CGS Ce — Cm (@ a « LAG) y i by Tee Tey AY AONE tite Sato: 4 mY 5 ‘ Moth ge at ie a) cs Y ana aii THE aoe Ws JD ie es Or LINNEAN SOCIETY. ZOOLOGY. VOL. OROVILLE. Vi Instityg SS / i. 1) \ | = \} ( LS (SSS NW: Lf SStional Musew LONDON: SOLD AT THE SOCIETY'S APARTMENTS, BURLINGTON HOUSE, PICCADILLY, W., AND BY LONGMANS, GREEN, AND CO., AND WILLIAMS AND NORGATE. 1900-1903. Dates of Publication of the several Numbers included in this Volume. WO, 7) pp. 1-105, published July 31, 1900. oy let0), ,, 106-160, is November 1, 1900. ay US, » 161-260, 4 April 11, 1901. a loz: », 261-307, Fe July 15, 1901. Pelicans », 908-365, sf November 1, 1901. pelisee », 366-453, - April 1, 1902. 185, » 404-454, 3 July 1, 1902. 186, Ao, s May 1, 1905. PRINTED BY TAYLOR AND FRANCIS, RED LION COURT, FLERT STREET, LIST OF, PAPERS. Pare ta) Broom, Rosert, M.D., C.M., B.Se. On the Early Condition of the Shoulder-Girdle in the Polypro- todont Marsupials Dasywrus and Perameles. (Communicated by Prof. G. B. Howes, F.R.S., Sec.L.S.) (Plate 41) .. 449-454 _CHAPMAN, FrepeErick, A.L.S., F.R.M.S. On some New and Interesting Foraminifera from the Funafuti Atoll, Ellice Islands. (Plates 1-4 and 2 illustrations) .... 1-27 On some Foraminifera of Tithonian Age from the Stramberg Limestone of Nesselsdorf. (Plate 5).........0..-..0-- 28-32 Foraminifera from the Lagoon of Funafuti. (Plates 19 & ZA0)) icky» BEIGE obec StULS SOR aes Ce cara OI neeCrDNE 161-210 On the Foraminifera collected round the Funafuti Atoll from Shallow and Moderately Deep Water. (Plates 35 & SAS VUNE GLA Bir iinc, dence ect eae ao 1A tA Ao ae clavate 379-417 On some Ostracoda from Funafuti. (Plate 37) ...... 417-453 Cuitron, Cuartes, M.A., D.Sc., F.L.S., Research Fellow, Uni- versity of Edinburgh. The Subterranean Amphipoda of the British Isles. (Plates 16— te) Cdeido baa ben coe o0b dbU.an0s Gomel ono o mo blots 140-161 Diesy, LETTICcE. On the Structure and Affinities of the Tanganyika Gastropods Chytra and Limnotrochus. (From the Biological Laboratory, Royal College of Science.) (Communicated by Prof. G. B. Howes, F.R.S., Sec.L.S.) (Plates 38-40) .......... 454-442 IV Page EmsBietron, Artce L., B.Sc., 1851 Exhibition Science Research Scholar, Associate of the University of Wales (Cardiff College). Goidelia japonica—a New Entozoic Copepod from Japan, asso- ciated with an Infusorian (T7richodina). (Communicated by Prof. G. B. Howes, F.R.S., Sec.L.S.) (Plates 21 & 22) 211-229 The Life-History of the Black-Currant Gall-Mite, Hriophyes (Phytoptus) ribis, Westwood. (See WARBURTON, CECIL) 866-3879 GoopricH, Epwin §., M.A., F.L.S. Observations on Syllis vivipara, Krohn. (Plate 13) .... 105-108 PACH S.9 E.L.9- On the Corallum of Turdinaria. (Communicated by H. M. Bernard, M.A., F.L.S:) (With 14 illustrations)...... 358-365 On the Anatomy of the Prosobranch Genus Pontiothauma, E. A. Smith. (Communicated by Prof. G. B. Howes, LL.D., DSc HERS: Secs.) | (elateE2) 2c erate eee 455-462 PyorarFt, W. P., M.A., A.LS., F.Z.S. Some Points in the Morphology of the Palate of the Neognathe. (lates SESSA) eres cra aleumira dekere sien) >-jsgebelolefelcl = foie areks 343-357 RipEwoop, W. G., D.Sc., F.L.S. On Obesiella lyonsielle, a new Genus of Copepod Crustacean. (With Zinllastrations)) er lnee ee aeee ‘ciate Ghetasalles 463-465 Scuarrr, Ropert F., Ph.D., B.Sc., Keeper Nat. Hist. Coll., National Museum, Dublin. Rhynchodemus Howest: a New European Species of Terrestrial Planarian Worm. (Communicated by Prof. G. B. Howes, F.R.S., Sec.L.8.) (Plate 6 and 2 illustrations) ........ 33-42 Smitu, G. Exutot, M.D., Fellow of St. John’s College, Cambridge ; Professor of Anatomy, Egyptian Government School of Medicine, Cairo. Notes on the Brain of Macroscelides and other Insectivora. (Communicated by Prof. G. B. Howes, F.R.S., Sec.L.5.) CWrth2valiby stra pions) yee (cise cusevanerseleps seamen a tego nists 445-448 Sweet, Groraina, M.Sc., Melbourne University. On the Structure of the Spermiducal Glands and Associated Parts in Australian Karthworms. (Communicated by Prof. G. Pag B. Howes, F.R.S., Sec.L.8.) (Plates14&15) ...... 109-139 Tims, H. W. Marner, B.A., M.D., F.L.S., F.Z.S., Lecturer on Biology and Comparative Anatomy, Charing Cross Hospital Medical School ; Lecturer on Zoology, Bedford College, London. Tooth-Genesis in the Cavide. (Plate 26 and 7 illustra- EL OMS) Uraepeesy eed tes Molnar sherauceen Malian tases tsratnccye saiarevein 6 261-290 WALKER, ALFRED O., F.L.S. Contributions to the Malacostracan Fauna of the Mediterranean. (late) ee ne s/s OR ee SD 290- Warsurton, Cxcit, M.A., F.Z.S., Zoologist to the Royal Agri- cultural Society of England, and Anicr L. Emsierron, B.Sc., 1851 Exhibition Science Research Scholar, Associate of the University of Wales (Cardiff College). The Life-History of the Black-Currant Gall-Mite, Eriophyes (Phytoptus) ribis, Westwood. (Communicated by Prof. G. 307 B. Howes, F.R.S., Sec.L.8.) (Plates 33 & 34) ..,... 366-379 Waters, ARTHUR WitiiAM, F.L.S. Bryozoa from Franz-Josef Land, collected by the Jackson- Harmsworth Expedition, 1896-1897.. (Plates 7-12) .. 43-105 Watson, ARNOLD T., F.L.S. On the Structure and Habits of the Polycheta of the Family Am mochariao n(Platey23—25) Wetec. eee si ciate < 930-260 West, G. S., B.A, F.LS., A.R.C.S., Hutchinson Student of St. John’s College, Cambridge; Professor of Natural History at the Royal Agricultural College, Cirencester. On some British Freshwater Rhizopods and Heliozoa. (Plates ASSO) Medeor eet a evake i dian deta ioinh = scvelate cia ab e-oceteisvacerni 2 308- WitH, C. New and Old Phalangiide from the Indian Region. (Com- 342 munucated! by vel yelancenb HMO S.\ as saa. se os) 466-508 EXPLANATION OF THE PLATES. PLATE ForamInrrera from Funafuti. Sagenrna, CarrenTerta, Potyrrema, and Cycioctyrevs from Funafuti. Hererostecina and CyctoctyPeus from Funafuti. Foraminirera from Nesselsdorf (Tithonian). CarpENTERIA and Potyrrema from Funafuti. Ruyycuopremus Howest, Scharff. OoIDAHRWON HE — as 10 \ Bryozoa from Franz-Josef Land. 11. 12. ) 13. Repropvuction of Syilis vivipara. 14. 15: 16. SrpERMIDUCAL GuANDs in Earthworms. NIPHARGUS SUBTERRANEUS, Leach. ae 1. NipHarcus suBTERRANEUS, Leach. 2. Nipwarcus Kocutanus, Spence Bate. ig,/1& 2. Nirwarevus Kocutanus, Spence Bate. 3. CRANGONYX SUBTERRANEUS, Spence Bate. 19. 20 \ Foraminirera from the Lagoon, Funafuti. 22. 3 A new Enrozorc Corzrop ( Goidelia japonica). aal =a Srructure of the Ammocharide. 25. Ammocnarzs and Development of Ownnta. 26. Tootu-Genzsis in the Caviide. 27. MupirErRRANHAN MaLAcosTRACA. 28. 29. | British FresuwaTer Ruizoropa. 30. 31, I Pawate of the NEOGNATHE. 32. 5 33. | ki 34. ° Erioruyes (PHYTorTus) RIBIS. svalll PLATE 35. Reer-Buitpine ForaAnintegra [rom Funafuti. 36. Forawrsirera from Funafuti. 37. New Osrracopa from Funafuti. 38. Limyorrocuus and Ciytra. 39. CnyTRa. Limnorrocuus and Cnyrrea; with Aporrhais, Nassopsis, Plewrotomaria, and T’rochus. 41. Snovnper-Girpuus of Foetal Dasyurus and Perameles. 42. Anarvony of the Prosobranch Genus Pontiothauma, EH. A. Smith. 40. t BR Ry AeA. Page 146, line 9 from top, for Hrzopsis read Eriopis, Brugz. 285, line 5 from bottom, for Huromyide read Eryomyide. , line 19 from bottom, for Hromyide read Eryomyide. 322, line 13 from top, for Philontis fontana read Philonotis fontana, Jas, Brid. 366, line 2 from top, for Phytoptis rébis read Phytoptus ribis, Westw. 383, line 14 from bottom, for Nubecularia Bradleyi, Millett, read Nubecularia Bradyi, Mallett. 406, line 14 from top, for Anomalina grosserrugosa read Anomalina grosse- rugosa, Gimbel. 413, line 7 from top, for Verneuilina Davidiana read Valvulina David- dana, Chapm. 415, line 22 from top, for Caudenia nitida read Candeina nitida, V Orb. THE JOURNAL OF THE LINNEAN SOCIETY. On some New and Interesting Foraminifera from the Funafuti Atoll, Ellice Islands. By Freprrick Capmany, A.L.S., F.R.MS. [Read 21st December, 1899.] * (Puates 1-4.) THE expeditions dispatched by the Royal Society of London from New South Wales under the direction of Professcr W. J. Sollas, and of Professor T. Edgeworth David, for the purpose of making a boring in a typical atoll in the Pacific Ocean, have furnished zoologists and others with some very interesting material for detailed study. In the examination of this material and that of other coral-reefs, one cannot fail to be impressed by the importance of organisms other than corals in forming the great mass of the reef. Among the groups of organisms which are as active as coral, and even more so, in building up the enormous banks, mounds, and reefs of limestone in coral areas, we may mention the cal- careous alge Halimeda and Lithothamnion as prominent rock- formers ; the first-named often growing in the greatest profusion to the exclusion of almost all else, especially in the lagoon, whilst the latter grows in branching masses which entangle smaller organisms and loose sand, or encrust corals and millepores, often at a considerable depth. A laminated alga allied to Litho- thamnion is also found growing in nodular form in alternate concentric layers with the foraminifer Polytrema; and the condition favourable for this peculiar intergrowth seems to be a sandy area influenced by strong currents. At the surface on the Funafuti Atoll, these nodular intergrowths were found only on the seaward face of the reef. [ * This paper has been unavoidably delayed in publication.—Ep.] LINN. JOURN.—ZOOLOGY, VOL. XXVIII. 1 2 MR. Fe CHAPMAN ON FORAMINIFERA The Foraminifera, however, constitute the greater proportion of the enormous deposits of sand associated with the reef forma- tion, and which speedily become consolidated into limestone-rock through the chemical changes which so readily take place therein. The most important genera of the Foraminifera which were found in these coral deposits, taken in the relative order of their abundance, are Amphistegina, Heterostegina, Polytrema, Tinoporus, Orbitolites, Carpenteria, Gypsina, Calearina, and Miliolina. The three first-named are found commonly throughout the material of the reef-boring, in sand and rock alike, always retaining some part of their original structure; by this they can easily be re- cognized, although the matrix of the rock may have undergone the most marked changes. In the deeper parts of the Atoll- boring, other genera may have been present, for fragments of Tinoporus in various stages of decay were seen down to 245 feet, and Calcarina to 274 feet from the surface. Among other active agents in the building of the limestone- reefs may be mentioned the Aleyonarians, Echinoderms, Serpule, and the Mollusca. Through the kindness of Professor Judd, C.B., and the Coral- Reef Committee of the Royal Society, I have been favoured with the opportunity for thoroughly examining, in respect to the eroup of the Foraminifera, the whole of the materials obtaimed by the expeditions to Funafuti, which had been sent to the Royal College of Science, London. The numerous samples from Funafuti have in most cases now been examined, but to do justice to so large a quantity of material some further work is requisite, and it is therefore considered advisable to publish this paper as a first instalment, which deals with the more prominent or larger forms, especially those found adherent to the coral or encrusting other organisms, and inter- growing with them. These larger forms of the Foraminifera do not seem to have received the notice they deserve, for they often constitute a considerable bulk of the consolidated reef, and are rock-builders in the truest sense of the term. That they are often overlooked is due to the fact that they more frequently occur on the rough and encrusted pieces of reef-rock, which are not so systematically collected, probably owing to their present- ing little in the way of attractive specimens of corals and other more or less prominent organic remains. Whilst describing these species of Foraminifera the occasion FROM THE FUNAFUTI ATOLL. 3 is taken to include an account of several new forms; but this does not exhaust the collection in hand, which it is hoped will fnornish interesting material for further papers on the same group of animals. The localities at Funafuti are given in their relative positions around the atoll from N. EB. S. to W. FORAMINIFERA. Family MILIOLIDA. Subfamily PENEROPLIDIN®S. PENEROPLIS, Montfort [1808]. Subgenus novum Monatysiprum. Remarks on the Subgenus.—Test porcellanous, shell-wall very thin ; surface usually covered with minute tubercles, sometimes smooth or highly polished and with vertical rows of puncta (not perforations). Segments sub-globose, flattened or elongate, at first arranged in a spiral, afterwards rectilinear. Aperture simple, either a circular inverted orifice or an everted phialine termination. This subgenus is intended to comprise the long delicate crosier- shaped specimens of which “Nautilus litwus,” Gmelin, is the type. These particular forms were represented in the ‘ Challenger’ collections by imperfect specimens, described by Dr. H. B. Brady under the generic name Peneroplis. The Funafuti material has yielded some good examples of these forms; and in the light of these it now appears convenient to distinguish the specimens subgenerically at least, on account of their dimorphic character *, and especially since their apertures differ from the typical cribrate or dendriform orifice of Peneroplis proper, or Spirolina, and are similar to Nodosaria or Sagrina of the hyaline group. It is possible that d’Orbigny’s Nodosaria punctata f is a frag- ment of one of the tuberculate forms of Wonalysidiwm. PreneRoPiis (Monatysipium) Sonnasi, subgen. et sp. nov. (PL. 1. fig. 6.) Test consisting of a flat coil of sub-oval segments, which afterwards become a straight series with subglobular or slightly compressed chambers. Terminal orifice with a neat everted * In the sense employed by Prof. Kitchen Parker. + See ‘ Foraminiféres’ in Sagra’s. Hist. Cuba, 1839, p. 14, pl. 1. figs. 4, 5. 1 * 4. MR. F. CHAPMAN ON FORAMINIFERA margin. Surface of test covered with minute tubercles which under a low power resemble perforations. Length ,4 inch (52 mm.). The fragmentary specimens recorded and figured by H. B. Brady * are tuberculate similarly to the above species, but differ considerably in the shape of the chambers and in showing a strong partitional septum on the surface between each segment. Found in sand from the beach, Avalau Id. (South Id.), Funafuti. Coll. by Prof. Sollas. Prneropiis (MoNALYSIDIUM) PoLITA, sp. nov. (PI. 1. fig. 5.) Test imperfect but probably similar in general outline to the foregoing species, with the exception that the segments, espe- cially of the last part of the series, are more irregular in form. The rectilinear portion of the shell (the only part discovered at present) consists of six segments, subcylindrical and irregular ; shell-surface highly polished, with vertical rows of minute puncta on each segment, and we Aperture with an everted margin. Length of portion found +. inch (714 mm.). From the beach-sand, Avalau Id. ‘ South Id.), Funafuti. Coll. by Prof. Sollas. Family ASTRORHIZIDA. Subfamily RHABDAMMININE SaGENINA (nom. emend.)*+. (Sagenella, Brady [1879].) SAGENINA FRONDESCENS (Brady). (Pl. 1. figs. 1,2; Pl. 2. figs. 1, 2.) Sagenella frondescens, Brady, 1879, Quart. Journ. Micr. Sci. vol. xix. p. 41, pl. v. fig. 1; Butschli, 1880, in Bronn, Klassen etc. Thier-Reichs, p- 195, pl. v. ae 16; Brady, 1884, Report ‘Challenger,’ vol. ix. p. 278, pl. xxviii. figs. 14 & 15. Nothing has been done towards extending the somewhat limited range of distribution for this species since Brady described “ this singular little organism from the South Pacific.” The original localities were the Admiralty Islands, where it was found on calcareous plants and fragments of shells at 16-35 fathoms; and * Report ‘Challenger,’ 1884, vol. ix. pl. xiii. figs. 24 & 25. + Sagenina is here suggested for Sagenedla, which name has been previously bestowed on a genus of the Pelyzoa. See Hall’s Nat. Hist. N. York, 1852, vol.ii. p. 172, pl. 40 EH. figs. 6a, d. FROM THE FUNAFUTI ATOLL. 5 off Tongatabu, Friendly Islands, at 18 fathoms. Brady states that the specimens from the latter locality lack distinctive characters ; those around Funafuti, not so greatly distant, being in first-rate preservation, may serve to strengthen his identifica- tion. Some of the specimens from the Funafuti lagoon found attached to the surface of Halimeda-joints are extremely fine and characteristic. Seeing Dr. Heckel’s statement with regard to the apparently dubious character of some Astrorhizide, especially Rhabdammina, Rhizammina, Sagenella, ete., that they “ may also belong to the arenaceous Keratosa (Ammoconide),” I have carefully examined the composition of the test of Sagenina (Sagenella), with the result that there appears sufficient evidence in support of the view taken by Dr. Brady as to its rhizopodal nature. Sections of a Halimeda-joint with Sagenina attached. have been made in a vertical direction; also the test itself has been isolated. These were examined under a high power, and showed this organism to have a finely arenaceous to subchitinous test (that is to say, where finely divided material 1s wanting, a thin chitinous tube is the result) traversed by minute inosculating canals such as are seen in the other arenaceous rhizopods. Heckel describes a form of the Keratosa, namely Ammoconia sagenella *, with regard to which he says: “ Very similar to this latter, or even identical with it, may be that form which Brady has figured as Sagenella frondescens.” In the description of the species, however, the diameter of the separate branchlets is given as 1-2mm. The average diameter (and it varies but little) of the branchlets of Sagenina frondescens is *5 mm. Further, the example of Ammoconia figured in the Report is quite unlike Sagenina in consisting of a meshwork of tubes standing erect, instead of being adherent at every point. The general finish of the tubes of Sagenina, especially at the terminal or aperturai points, enables one to see the affinity of this form with other well-known rhizopods, both arenaceous and hyaline, as Webbina and the adherent Ramulina cervicornis. In one part of the transparent test which I have examined, there is a curious stel- late structure reminding one of the fused spiculose base seen in the tissues of Alcyonarians. There is not sufficient evidence, however, to lead one to form any definite conclusion from this peculiarity of structure. * Report ‘ Challenger,’ vol, xxxii. pp. 30 & 31, pl. viii. figs. 5 A, 5 B. 6 MR. F. CHAPMAN ON FORAMINIFERRA Sagenina frondescens is a well-distributed organism around Funafuti and in the lagoon. In places it is quite abundant. In the dredgings made by Prof. David and Mr. Finckh this species: was found as follows :— N. of Pava, 36 fathoms. Funamanu (Beacon Id.), 150 fathoms ; very fine examples. Tutanga, 200 fathoms (Halligan and Finckh Coll.), occasional. 8. of Funafuti, 30-120 fathoms. “To” S. of Funafuti, 60 fathoms, common on Cycloclypeus and Halimeda. In the lagoon boring it was found at the surface of the lagoon- floor; very rare. At 213 feet below floor ; very common, and fine examples. At 50 feet; frequent. At 62 feet; common. At 814 feet; very rare. Family LITUOLIDA. Subfamily Lirvo0oLIN a. Happonta, Chapman {1898}. Happonia TorREstensis, Chapman. Haddonia Torresiensis, Chapman, 1898, Journ. (Zool.) Linn. Soe. Lond. yol. xxvi. p. 452, pl. xxviii. figs. 1-5 and woodcut p. 453, This interesting genus, although but lately discovered in material from Torres Strait, has already been found in great abundance both around Funafuti and in the lagoon. The large series of specimens now before us shows to some extent the great variety of forms exhibited by this organism. In the first stages of its growth Haddonia is depressed and repent, having a coiled series at the commencement or a simple linear- shaped arrangement of segments; in the latter case it somewhat resembles Bdelloidina, to which it probably bears some relation- ship. The test afterwards grows off the surface of attachment, still retaining its point d’ apput, often continuing in a very erratic line of growth, and sometimes forming several bends and twists nearly at right angles, in one instance curiously resembling a larva of the Geometrine Lepidoptera when feeling the air for a fresh foothold. The largest specimen met with measures over 3 inch (about 16 mm.) in length, and was only slightly attached by the aboral end, coming away in the dredge. The organism has FROM THE FUNAFUTI ATOLL. 7 secondarily attached itself a little beyond the middle of the test to anechinoderm spine. This specimen came from Funamanu Id. (Beacon Id.) at 150 fathoms. H. Torresiensis was also found in the material from the boring in the lagoon at Funafuti. The specimens are of average size. In the dredgings made by Prof. David and Mr. Finckh, I have found H. Torresiensis as follows :— N. of Pava, 35 fathoms ; a young individual. Funamanu Id. (Beacon Id.), 150 fathoms; the large specimen previously mentioned. Tutanga, 200 fathoms (dredged by Messrs. Halligan and Finckh); rather small examples. S. of Fuafatu, 25 fathoms; a young individual. S. of Fuafatu, 60 fathoms; a young example attached to Cycloclypeus Carpentert (form B). S. of Fuafatu, 30-120 fathoms; a much contorted example. S. of Fuafatu, 119 fathoms; very common on a fragment of Millepore. “To” S. of Fuafatu, 40 fathoms, bottom fairly hard; a neat specimen torn off the rock. “To” S. of Fuafatu, 60 fathoms ; a neat specimen. S.S.W. of Fuafutu, 60 fathoms; common on coral rocks. Two specimens were also obtained from the lagoon-boring at 354 feet and at 814 feet below the lagoon-floor. BDELLOIDINA, Carter [1877]. BDELLOIDINA AGGREGATA, Carter. (Pl. 1. fig. 3, and fig. in text.) Bdelloidina aggregata, Carter, 1877, Ann, Mag. Nat. Hist. ser. 4, vol. xix. p. 201, pl. xii. figs. 1-8; Brady, 1884, Report ‘ Challenger,’ vol. ix. p. 319, pl. xxxvi. figs. 4-6. “Test adherent, depressed ; consisting of a number of closely- approximated chambers, arranged more or less regularly in a single, simple or branched, linear series, and intercommunicating by a row of pools on each septal face. Segments very short (on the axis of growth) and broad ; curved or irregular in outline; subdivided more or less completely by numerous secondary septa. Walls rough externally, interior surface smooth. Aperture porous. Diameter of the adherent patches 1 inch (4 mm.) or more.” * The above species was well and minutely described by * Brady, op. cit. pp. 319, 320. 8 MR. F. CHAPMAN ON FORAMINIFERA Mr. Carter, who drew attention to the porous nature of the shell- wall in this and some other Rhizopods.* With regard to the porosity of the shell-wall, it appears to me that Carter conclusively demonstrated the presence of pores in the wallof the test, and although these pores are not regular enough to be denominated tubules, yet they probably serve the same purpose for the sarcodic body. Dr. Brady has suggested that the puncta on the internal surface of Bdelloidina may be only superficial depressions. In corroboration of Carter’s original idea, it is worth noting a precisely similar structure in the shell-wall of the allied genus Haddonia, in which the test is stronger and more inflated, and the structure of the shell-wall greatly emphasized. There the porosity of the shell-wall seems to exist beyond question. The apertural pores at the distal end in Bdelloidina are evi- dently the terminations of the labyrinthular canals seen in the interior of the shell. The width of the test is usually 4 inch (about 4 mm.) and the Jength slightly more. One example from Fuafatu, however, measures as much as 3 inch (12°5 mm.). Bdelloidina is a genus likely to be overlooked, since its habit is to seek shelter in hollows of shells and coral-rock, creeping over the interior and filling up irregularities in the surface. Its usual colour, moreover, is that of a calcareous sand, for it is composed of little granules (fragments of organisms) or even minute shells, together with occasional sponge-spicules. This fact probably explains its rarity in records of foraminiferal faunas from coral areas. The only localized specimens of Bdelloidina aggregata previ- ously obtained were from the ‘ Challenger’ dredgings in shallow water near the Admiralty Islands on the north coast of Papua, generally adherent to molluscan shells. A fossil specimen has been recorded as adherent to a Creta- ceous Ammonite from Mr. Matthew Wright’s cabinet fT ; a drawing of this was made many years ago which is in the possession of Prof. Rupert Jones. I am enabled by the kindness of Prof. Jones to give a reproduction of this drawing for the first time. (See fig. 1, p. 9.) In the neighbourhood of Funafuti Bdelloidina aggregata is notably restricted to shallow water. * Carter, op. cit. pp. 202-208. + See Brady, Rep. Chall. 1884, vol. ix. p. 320. FROM THE FUNAFUTI ATOLL. 9 Near Pava Id., fairly numerous on fragments of Mullepere at 63 fathoms. Fig. 1—Bdelloidina aggregata, Carter from the Chalk of Kent. (From a drawing lent by Prof Rupert Jones of a specimen in the Collection of the late Matthew Wright.) Funamanu Id. (Beacon Id.), on Turbinaria, 50 fathoms (dredged by Prof. David and Mr. Finckh). 8S. of Fuafatu Id., in the interior of molluscan shells, 25 fathoms. “To” S. of Fuafatu Id., 60 fathoms. Family TEXTULARITD. Subfamily TEXTULARIINS. Vaxtvuina, d’ Orbigny [1526]. Vautvuina Daviprana, sp. noy. (PI. 1. fig. 4.) Test triangular, chambers triserially arranged and forming a more or less equilateral cone; aperture situated in a sub- erescentic depression, with a tooth-like valve projecting across, and sometimes a secondary cne nearly meeting from the opposite side. Texture somewhat coarsely arenaceous, composed of caleareous particles; colour white to pale cream. Length zs inch (1 mm.). The above species belongs to the same type as d’Orbigny’s Valwulina triangularis from the Tertiary beds of the Paris Basin*. The specimens from Funafuti differ, however, in having * Aun. Sci, Nat., vol. vii. 1826, p. 270, No. 1; Modéles, No. 25. 10 MR. F. CHAPMAN ON FORAMINIFERA a more decidedly triangular and elongate test, and the oral end less inflated. The valve is more usually tooth-like, rather than a simple flap as in d’Orbigny’s specimens. The elongated examples of V. Davidiana, which by the way are rare, may be compared with V. Oviedoiana of d’Orbigny * ; the latter, however, differs in the marked concavity of the face of each segment, as seen in the figures of d’Orbigny’s Cuban specimens. V. Davidiana is by no means common, being represented in all the samples mentioned below by one or two examples at most. It was found in the sand of the beach at Avalau Id. (South Id.) and at Fualopa Id. (Gold Id.) ; in the sand pumped up from the 1st boring (Sollas) “at no great depth”; and it also occurred in the lagoon boring at 213 feet below the lagoon- floor. Family ROTALITD A. Subfamily SPIRILLINING. Sprriniina, Hhrenberg [1841]. SPIRTLLINA SPINIGERA, sp.nov. (Pl. 1. figs. 7 a-c.) Test free, but perhaps at one time adherent or resting generally on one face, discoidal, consisting of a single tube in four con- volutions. The two sides unequal, the broader or basal face concave, having the primordial or central portion globulose and armed with a short blunt spine which does not project beyond the general surface; around the peripbery other blunt spies are arranged, to the number of fourteen or more, and the salient edges of the coils are more or less spinose. The upper (smaller) face is slightly convex and tolerably smooth. The terminal portion of the coil is separated from the rest of the shell and is slightly reverted. Diameter of test 7; inch (325 mm.). In having the lateral faces of unequal size this form resembles Spirillina inequalis, Brady v. It also shows some affinity towards S. tuberculata, Brady {, in its separated termination. The form may have been derived directly from S. lémbata var. * Foram. Cuba, 1839, p. 103, pl. ii. figs. 21, 22 (V. Ovtedoana on plate). + Quart. Journ. Mier. Sci. vol. xix. n. s. (1879), p. 278, pl. viii. figs. 25 a, b. t Ibid. p. 279, pl. viii. figs. 28a, >. FROM THE FUNAFUTI ATOLL. ial denticulata, Brady *, by the redundant outgrowth of the transverse bars or denticule, but this is merely a suggestion. From beach-sand, Avalau Id. (South Island), Funafuti; very rare. Collected by Prof. Sollas. Also from the boring in the lagoon at Funafuti at 62 feet below the floor of the lagoon (163 feet below low-water spring tides) ; very rare. SPIRILLINA TUBERCULATO-LIMBATA, Sp. nov. (PI. 1. figs. 8 a-c.) Test discoidal, consisting of a single tube with about six conyolutions. The larger and flat surface is limbate and has the peripheral edge of the coil sharp on that side; the smaller face is slightly rounded and strongly tuberculate. The upper part of the peripheral edge rounded. Diameter = inch (‘5 mm.). The dissimilar faces remind one of S. énequalis, Brady, but in that species the peripheral edge is blunt or square and the smaller face is not so characteristically tuberculate. S. twberculato-limbata, as its name implies, combines in one form the characters of two of Brady’s species, S. tuberculata + and 8. limbata ¢. This species was found at the depth of 40 feet in the 2nd boring at Funafuti (Sollas Coll. 1896), where it was common. It has occurred sparingly at the depths of 65 and 70 feet in the deep boring at Funafuti (David Coll. 1897). Also in dredgings N. of Pava, 36 fathoms (David Coll. 1898). Subfamily Roraniin 2. Discorzsina, Parker &§ Jones {1862 |. DiscoRBINA TUBEROCAPITATA, sp.nov. (Pl. 1. figs. 9 ace.) The superior aspect roughly conical, the apex swollen into a ball-like prominence ; two convolutions of the shell are visible on the apex. The inferior surface is flat or slightly concave, and exhibits five chambers ; in the area round the umbilical depression the surface of the shell is granulate. The peripheral edge has a thin and narrow flange. Shell translucent and of a pale opales- cent blue colour. Diameter “5 inch (°347 mm.). * Rep. Chall. vol. ix. 1884, p. 632, pl. Ixxxv. fig. 17. + Quart. Journ. Mier. Sci. vol. xix. n. s. 1879, p. 279, pl. viii. ,figs. 28a, 0: Rep. Chall. vol. ix. 1884, p. 631, pl. Ixxxv. figs. 12-16. t Quart. Journ. Micr. Sci. vol. xix. n. s. 1879, p. 278, pl. viii. figs. 26a, ); Rep. Chall. vol. ix. 1884, p. 632, pl. Ixxxv. figs. 18-21. 12 MR. F. CHAPMAN on FORAMINIFERA This curious little species is quite distinct from the other forms of Discorbina found at Funafuti. It appears to have had its origin in a form similar to D. pileolus, ’Orbigny *, whieh it somewhat resembles in the conical portion—a form common in the shallow waters of the Pacific. ; D. tuberocapitata is rare, and has occurred in two samples as far as they have been at present examined. In the deep boring at Funafuti (David, 1897) at a depth of 65 feet; and in the lagoon-boring at 50 feet below the floor of the lagoon. CaARPENTERIA, Gray [1858]. CaARPENTERIA UTRICULARIS, Carter. (Pl. 2. fig. 4; Pl. 4. figs. 3, 4.) Polytrema utriculare, Carter, 1876, Ann. Mag. Nat. Hist. ser. 4, vol. xvii. p. 210, pl. xiii. figs. 11-16. Carpenteria utricularis, Carter, 1877, Ann. Mag. Nat. Hist. ser. 4, vol. xx. p.176; Brady, 1884, Report ‘Challenger,’ vol. ix. p. 678, pl. xcix. figs. 6, 7, pl. c. figs. 1-4; Egger, 1893, Abhandl. bayer. Akad. Wiss., math.-phys. Cl. ii. vol. xviii., Abth. ii. p. 246, pl. xxi. fig. 18. The chief characteristics of this species are the strongly inflated segments, the occasional plurality of apertures (basal as well as terminal), and the coarsely reticulated structure of the walls of the test. In sections the shell-wall is seen to have a papillate (rugose) external surface. The colour of C. utricularis in the Funafuti specimens is sometimes brown or ash-grey, but generally speaking the specimens are of a vivid apple-green tint, probably due to the presence of a symbiotic alga in the coarse structure of the test. The apertural portion of the shell is often prolonged into a slender salient tube. C.utricularis is apparently the commonest form of Carpenteria dredged at Funafuti. This species has been recorded from the Pacific, Cape Verde Islands, the West African coast, the Mauritius, and the West Indies. Brady records it at depths ranging from 16 to 350 fathoms. In the samples collected round Funafuti by Prof. David and Mr. Finckh, I have found C. utricularis in the following :— North of Pava, 57 fathoms, on a gorgonid and a millepore. * Valvulina pileolus, Voyage Amér. Mérid. 1839, vol. v. pt. 5: ‘ Foramini- féres,” p. 47, pl. i. figs. 15-17. Discorbina pileolus (d’Orbigny), Brady, Rep. Chall. vol. ix. 1884, p. 649, pl. Ixxxix. figs. 2-4. FROM THE FUNAFUTI ATOLL. 13 Funamanu Id. (Beacon Id.), 50 fathoms. Funamanu Id. (Beacon Id.), 80 fathoms ; a specimen of bright green colour attached to a gorgonid. Off Funamanu Id. (Beacon Id.), 150 fathoms; a specimen attached to Halimeda. Tutanga, 94 fathoms; the largest specimen found, measuring nearly half an inch at the base. This is attached to a cylindrical millepore. Near Tutanga, 136 fathoms. S. of Fuafatu, 60 fathoms; on Oycloclypeus Carpenteri. S.S.W. of Fuafatu, 60 fathoms; very abundant on a large coral fragment. S. of Fuafatu, 25 fathoms. (A depressed form very near one of Gray’s figures of C. balaniformis.) A single specimen of C. uwtricularis was also found in the sand pumped up (at no great depth) from the 1st boring (Sollas Coll.). In the deep boring (David Coll.) this species was found at 65 feet and at 70 feet. CaRPENTERIA BALANIFORMIS, Gray. (Pl. 4. figs. 1, 2.) Carpenteria balaniformis, Gray, 1858, Proc. Zool. Soc. Lond. vol. xxvi. p- 269, tigs. 1-4. Carpenteria, Carpenter, 1862, Introd. Foram. pl. xxi. figs. 6-14. C. balanifornis, Gray, Rupert Jones, 1875, in Griffith and Henfrey’s Micrographic Dictionary, 3rd ed., vol. 11. pl. 42. fig. 28. C. balaniformis, Gray, Carter, 1877, Ann. Mag. Nat. Hist. ser. 4 vol. xix. pl. xii. fig. 18. “ Carpenteria young specimens, C. balaniformis?,” Brady, 1884, Report ‘Challenger,’ vol. ix. p. 677, pl. xeviii. figs. 14, 17. C. balaniformis, Gray, Agassiz, 1888, Three Cruises ‘ Blake,’ vol. ii. p. 168, fig. 514. C. balaniformis, Gray [non Brady |, Egger, 1893, Abhandl. bayer. Akad. Wiss., math.-phys. Cl. ii. vol. xvili., Abth. ii. p. 246, pl. xxi. figs. 13-15. The specific characters of C. balaniformis are perhaps difficult to define, since the original specimens of Gray appear to possess some of the characters of both C. monticularis and C. utricu- laris, Carter. However, the name may be retained for the small, balaniform, depressed, conical and usually smooth Carpenteri« often found clustering upon the stems of gorgonids and other objects of attachment, where the currents have more or less access to them. HKegger’s specimens were dredged off the coast of West Africa. Examples, which may be taken as typical of the species, were 7 14 MR. F. CHAPMAN ON FORAMINIFERA found round Funafuti in a dredging made between 115 and 200 fathoms. They are attached to corals as Synhelia? and Caryo- phyllia?, and to a mass of Serpula-tubes; on the latter they occur in great abundance (David Coll.). Tt also occurs in dredgiugs made by Messrs. Halligan and Finckh at Tutanga, 200 fathoms. C. balaniformis was also found in sand from the bore-hole, Funafuti, at 75 feet; rare (David Coll.). CARPENTERIA MONTICULARIS, Carter. (Pl. 2. fig. 5; Pl. 4. figs. 5, 6.) Carpenteria, Carpenter, 1862, Introd. Foram. pl. xxi. fig. 7. Carpenteria monticularis, Carter, 1877, Ann, Mag. Nat. Hist. ser. 4, vol, xix. p. 211, pl. xiii. figs. 9-12; idem, ibid., ser. 4, vol. xx. p. 68, and woodcut ; Brady, 1884, Report ‘Challenger,’ vol. ix. p. 677, pl. xcix. figs. 1-5; Egger, 1893, Abhandl. bayer. Akad. Wiss., math.-phys. Cl. ii. vol. xviii, Abth. ii. p. 246, pl. xxi. fig. 12. This species, although so well-known, does not appear to have been found in any great abundance in the localities from which it has been previously recorded. The exceptionally fine speci- mens figured by Brady came from Zamboanga, Philippine Islands, at 102 fathoms; and he also records it from nine other localities, chiefly in the Pacific and Indian Oceans, at depths varying from 40 to 1035 fathoms, and at the same time notes that these specimens are ‘for the most part of small size and not so well characterized.” Egger obtained this species from the coast of West Africa and the Mauritius. The test of this species is of considerable thickness, with smooth surfaces, unlike C. utricularis, and the fine tubuli are here and there relieved by coarser tubules, running parallel with them. The Funafuti specimens are in nearly all examples of a delicate yellowish- or salmon-pink tint. One specimen is coloured a delicate mauve. In the dredgings made by Prof. David and Mr. Finckh around Funafuti, I have found C. monticularis in the following :— Funamanu Id. (Beacon Id.), 80 fathoms ; one specimen. Tutanga, 86 fathoms ; one specimen. Tutanga, 135 fathoms; one specimen attached to a mollusean shell. FROM THE FUNAFUTI ATOLL. 1 i) | Tutanga, 200 fathoms. S.S.W. of Fuafatu, 60 fathoms ; one specimen. In the collections made by Prof. Sollas at Funafuti O. monti- cularis has occurred in :-— Beach sand, Gold Island; one specimen. Beach of Creek, Rocky Island; one specimen. In coral-rock of the 2nd shaft, trial pit, ocean side; frequent. Under solid platform, seaward face, 3rd shaft, trial pit; common. In sand at 90 feet in the Ist (Sollas’s) boring ; one specimen. The fragmentary tests of this species also occur in great abundance in some of the thin slices of the core from the deep boring at Funafuti carried out by Prof. Edgeworth David. The remains of this species sometimes constitute quite an appre- ciable amount of the rock. Cancarina, d’ Orbiyny | 1826 |. CaLCARINA HISPIDA, Brady, var. PULCHELLA, nov. (Pl. 1. fig. 10.) “Calcarma distinguished by unusual exuberance of spinous outgrowths,” Carpenter, 1862, Introd. Foram. p. 217, figs. xxxiii A & B, fies. xxxiv D, E, F, G. Many of the specimens of C. hispida from Funafuti differ considerably from Brady’s type-specimens*. ‘The shell-wall is more delicate, and the spurs are slenderer, with exuberant out- growths at the extremities, whilst they are decorated with longer spines than usual. This variation gives to the test a light plumose appearance. In point of size the variety equals the type. The specimens which Dr. Carpenter figured were from the Philippines. C. hispida var. pulchella occurs with frequency both in the Beach Sand, “‘ South Island” (Avalau Islet), and in the Ist boring at Funafuti, “at no great depth ” (Sollas, 1896). Subfamily TrNopoRiIn 2. PotytremMa, isso [1826]. [The members of this genus are often so like minute branching corals and polyzoa, and in one instance resembling externally certain calcareous alge, that even at the present time they may * Calcarina hispida, Brady, 1876, Proc. R. Irish Acad. ser. 2, vol. ii. p. 590. C. calcar, VOrb., var. hispida, Brady, Carter, 1880, Ann, Mag. Nat. Hist. ser. 5, vol. vi. p. 4598. C. hispida, Brady, 1884, Rep. Chall. vol. ix. p. 715, pl. eviil. figs. 8, 9. 16 MR. F. CHAPMAN ON FORAMINIFERA at first sight be referred to the above groups by those who are not intimately acquainted with the minute structure of these Rhizopods. Polytrema miniaceum and its varieties are extremely ubiquitous at Funafuti. ] PoLYTREMA MINIACEUM (Pallas). (Pl. 4. fig. 7.) Millepora miniacea, Pallas, 1766, Elenchus Zoophytorum, p. 251 Linné, 1788, Systema Nat., 13th (Gmelin’s) ed. vol. i. pt. 6, p. 3784, no. 6; Esper, 1791, Die Pflanzenthiere, pt. i. p. 225, pl. xvii. figs. 1-4. M. rubra, Lamarck, 1816, Anim. sans Vertébres, vol. ii. p. 202. Polytrema corallina, Risso, 1826, Hist. Nat. Europ. Mérid. vol. v. p- 340, no. 19. P. miniaceum (Esper), Defrance, 1816-1830, Dict. Sci. Nat. Atlas, Zooph. pl. xliv. figs. 4,4; Blainville, 1826, Dict. Sci. Nat. vol. xlii., Atlas Zooph. vol. i. p.17; idem, 1834, Actinologie, pp. 410, 673, pl. Lxix. figs. 4, 4a. P. rubra (Lam.), Dujardin, 1841, Hist. Nat. Zooph. Infus. p. 259. Pustularia rosea, Gray, 1858, Proc. Zool. Soc. Lond. vol. xxvi. p. 271. Polytrema rubra (Lam.), Carpenter, 1862, Introd. Foram. p. 235, pl. xiii. figs. 18-20. P. miniaceum (Linné), Schultze, 1863, Wiegmann’s Archiv, p. 81, pl. viii.; Allman, 1870, Ann. Mag. Nat. Hist. ser. 4, vol. v. p. 372. P. miniaceum, Blainville, Carter, 1876, Ann. Mag. Nat. Hist. ser. 4, vol. xvii. p. 185, pl. xi. figs. 1-6. P. rubra (Lam.), Schwager, 1877, Boll. R. Com. Geol. Ital. vol. viii. p. 26, pl. 72. P. miniaceum, Blainville, Mobius, 1880, Beitr. Meeresfauna Insel Mauritius ete. p. 85, pl. vii. figs. 1-17; Butschli, 1880, in Bronn, Klassen etc. Thier-Reichs, p. 208, pl. ix. fig. 11. P. miniaceum (Linné), Brady, 1884, Rep. Chall. vol. ix. p. 721, pl. ¢. figs. 5-9, pl. ci. fig. 1; Agassiz, 1888, Three Cruises ‘ Blake,’ pt. ii. p- 169, fig. 519; Egger, 1893, Abhandl. bayer. Akad. Wiss., math.- phys. Cl. ii. vol. xviii., Abth. i. p. 437, pl. xxi. figs. 1, 2, 16, 17; Goés, 1896, Bull. Mus. Comp. Zool. vol. xxix. no, 1, Memoir xx, p. 75 P. miniaceum occurs in the samples both from the lagoon and the outer part of the reef at Funafuti, as well as from the deeper soundings round about, down to 200 fms. The species is not so common in the lagoon as it is outside the reef, but nevertheless some very fine specimens have been noticed from the lagoon- deposits. In all cases the specimens were originally attached to some object for basal support; and it is noteworthy that the kind of support has greatly determined the mode of habit of the form, and: hence its variations. Where the organism has sufficient space and freedom around its base, the delicate arborescent form FROM THE FUNAFUTI ATOLL. 17 is the result. Should it become cramped in its surroundings, the rhizopod makes a thick base with short and strong prolongations. When the Polytrema affixes itself to a detached fragment affected by currents which roll it over, the organism forms a thin en- crustation, which in many cases is in turn encrusted by a Litho- éhamnion, then again by Polytrema, and so on, until the nodules often measure as much as 5 centimetres in diameter. This latter mode of growth in Polyérema has hitherto been unnoticed, and seems to merit a distinct varietal name*. An intermediate stage between this encrusting Polytrema and the arborescent typical form is represented by Carter’s P. mesentericum +. The red colour of Polytrema is apparently very little, if at all, dependent upon the influence of light. Some of the finest coloured specimens occur at considerable depths, and it is remarkable that off Funamanu Id. (Beacon Island), at a depth of 150 fathoms, Polytremata of the deepest and richest colour were found. Although commonly found in shallower water, P. miniaceum has been recorded in the ‘ Challenger’ soundings from a depth as great as 1000 fathoms. Egger found it ranging down to a depth of 411 metres (224 fathoms) near Mauritius. Specimens recorded from the Caribbean Sea by Goés were found at 115 fathoms. PoLyTREMA MINIACEUM (Pallas), var. INVOLVA, nov. (Pl. 2. fig. 3, & text-fig. 2.) Test consisting of a rudely parallel series of acervuline chamber- lets disposed round a nucleus, or partly encrusting an adventitious particle. The organism in this stage shows little tendency to send forth prolongations, but seems content with forming mere rounded prominences or papille at intervals on the surface of growth. ‘This variety often shows a remarkable tendency to alternate in its growth with layers of Zithothamnion, thus building up nodular masses of a composite nature {. The nodules sometimes reach the dimension of 5 centimetres. As previously noticed, the form named Polytrema mesentericum, Carter, may be in part comparable with this form. It is probable, however, that this was the basal portion of a large * See p. 1. t+ Ann. Mag. Nat. Hist. ser. 5, vol. v. (1880) p. 444, pl. xviii. figs. 3 a-h. ¢ Another species has been noticed at Funafuti which occurs in the lime- stones alternating in its growth with Lithothamnion, namely Gypsina inherens (Schultze); but it does not form anything like so complete an, investment as P. miniaceum var. involva. LINN. JOURN.—ZOOLOGY, VOL, XXVIII. 2 18 MR. F. CHAPMAN ON FORAMINIFERA specimen of Polytrema often seen adherent to coral-rock, where the test bas formed a meandering surface-growth rather than an arborescent structure. Fig. 2.—Core from boring, Funafuti Atoll, at 766 ft. Containing nodules of Polytrema miniaceum yar. involva nov. Nat. size. P. miniaceum var. involva occurs in the coarse beach-sand of the outer reef-slope at Funafuti, and also at various depths in the deep boring. Its occurrence in the latter is of very great interest. These nodules of intergrowth are scattered throughout the more sandy parts of the cores of limestone down to a depth of 640 feet, when they are more frequent and of larger development. The largest and most typical specimens come from about 660 feet, where they are embedded in an Amphistegina-sand. They are common in the cores down to about 790 feet, after that occurring occasionally, and usually forming smaller nodules. P. miniaceum var. involva has also recently been noticed by the author in the Tertiary limestones of Christmas Island, associated with Orbitoides. Family NUMMULINIDZ. Subfamily NUMMULITING. Heterostecina, d Orbigny [1826]. HETEROSTEGINA DEPRESSA, d’Orbigny. (Pl. 3. figs. 6 & 7.) Heterostegina depressa, d’Orbigny, 1826, Ann. Sci. Nat. vol. vii. p. 805, _ pl. xvii. figs. 5-7 ; Modéle, no. 99. HI. antillarum, donna, 1859, Foram. Cuba, p. 211, pl. vii. figs. 4,9 25. FROM THE FUNAFUTI ATOLL. 19 H. simplex, VOrbigny, 1846, Foram. foss. Vienne, p. 211, pl. xii. figs. 12-14. H. helvetica, Kaufmann, 1867, Geol. Beschreib. des Pilatus, p. 153, pl. ix. figs. 6-10. _ HI. depressa, var. simplex, VOrbigny, Goés, 1882, Retic. Rhizop. Caribb. Sea, K. Svenska Vet.-Akad. Handl. vol. xix. no. 4, p. 117, pl. viii. | fig. 303. H. depressa, VOrbigny ; Brady, 1884, Rep. ‘Chall.’ vol. ix. p. 746, pl. exii. figs. 14-20; Egger, 1893, Abhandl. bayer. Akad. Wiss., math.- phys. Cl. ii. Bd. xviii. Abth. ii. p. 241, pl. xx. figs. 34, 35; Goés, 1896, Bull. Mus. Comp. Zool. Harvard, vol. xxix. no. 2, Memoir xx. p. 79. As Dr. H. B. Brady has already pointed out in his monograph on the ‘ Challenger’ Foraminifera *, H. depressa can be arranged in two series according to their form and external characters. After a careful study of the Funafuti specimens, I have been able to identify these two series with the dimorphic forms A and B respectively. The megalospheric form (A) embraces the involute, bi-convex type, with the asymmetric (spiral) flange ; whilst the microspheric form (B) is represented by the com- pressed and explanate specimens. The difference between the primordial spheres of the two forms is not very great, but it is constant, their diameters being in the ratio of 3: 2. The megalospheric form is far more common than the micro- spheric, for in most samples it is in the proportion of about 500 to 1, or even greater. The megalospheric form in H. depressa attains a larger diameter in full-grown specimens than the microspheric, which is unusual amongst the Foraminifera yet examined with regard to dimorphism, but nevertheless instances are known: for example, in Adelosina polygonia, Schlumberger, the megalospheric shell is slightly larger than the microspheric (1°5 to 1-4 mm.). From so large a collection of specimens as we are now dealing with, possibly one of the most extensive from any particular locality, one may obtain a very complete idea of the extent of variation in a species from a given area. H. depressa im its megalospheric furm varies greatly as to size and thickness, ai.d the microspheric form varies curiously in its internal septal arrangement, to be noticed below. The two forms, however, are in themselves so distinct as to present little difficulty in relegating them to either series. Form B has a thin and transparent, or translucent, shell-wall, and thus the septation can be clearly seen. The earlier portion shows no secondary septation dividing the chambers into chamber- * Rep. Chall, vol. ix. p. 746, pl. cxii. figs 17, 18 19, 20. 2% 22 MR. F. CHAPMAN ON FORAMINIFERA lets; but after two whorls of the shell have been completed, and while the test is still small, rudimentary secondary septation begins at the peripheral angles of the chambers, after the manner of d’Orbigny’s Heterostegina simplex *. The next stage in the septal division is seen after the completion of about four whorls, when straight, bar-like, but mncomplete cross-septa are seen projecting from the septum proper, but which do not extend for more than 3 across the chamber-cavity. In many specimens this incomplete septation is apparent to within the last few chambers, when it takes on the usual characteristic secondary septation of the Heterostegine type. One specimen of the complanate type (B) from Funafuti ex- hibits a remarkable variation in having the secondary chamberlets themselves divided at right angles by a third series of septal bars. The surface of the shell in this form (B) is usually granulate, or covered more or less with numerous hyaline papille, which are the rounded bases of cones of non-tubulous shell-substance, their points being directed inwards to the median plane. When the granules are arranged, as is sometimes the case, along the septal lines, we have the variation named by d’Orbigny 4H. costata f. Form A, on the contrary, always possesses a smooth-surfaced shell. Form B is found most frequently at a depth of 60 fathoms round Funafuti. In the sarecode of a megalospheric form, which was stained with picrocarmine, a nucleus with a nucleolus was observed, measuring 70 x 60 p. A microspheric specimen contained, in the middle of the sarcode body, a group of about a dozen rounded particles, pro- bably nutritive; whilst scattered through the body-substance there were seen six or more rounded algal cells (? symbiotic). H. depressa is not confined to the outer side of Funafuti. It is equally common in the lagoon, where it often attains as large dimensions as specimens from the outside. Very large specimens have also been noticed in some of the deeper cores of the boring in the atoll. It is especially noteworthy that the largest speci- mens of H. depressa found at Funafuti, measuring nearly 3 inch in diameter, were from dredgings at 32 and 386 fathoms outside Funafuti Atoll, N. of Pava (coll. by Prof. David). * D’Orbigny’s figures, however, appear to belong to form A. t Foram. Foss. Vienne, 1846, p. 212, pl. xi. figs. 15-17. FROM THE FUNAFUTI ATOLL. 21 Subfamily CYCLOCLYPEIN. CycLocLyPeEvs, Carpenter [1856]. This genus is very restricted in its distribution. It appears to find its most congenial habitat in tropical areas where the water is practically free from suspended terrigenous material. From the investigations made round Funafuti we have learned of the remarkable and perhaps unique abundance of the various stages of this hitherto rare type. In point of fact only two other localities seem to have produced the larger form of C. Car- penteri, Brady, which often attains a diameter of 23 inches and merits the distinction of being the largest living foraminifer. The study of the Funatuti specimens at first presented no slight difficulty on account of the occurrence of so many appa- rently distinct varieties in the material examined. This, however, was cleared away by the discovery of the interesting fact that the whole of the series illustrates different phases or stages of one species, which is dimorphic *. In the course of obtaining additional knowledge regarding the coupling of species hitherto considered distinct, it seems most desirable to adhere to the principle of priority in the nomenclature of such dual forms, retaining the earlier of the two names for the pair. For example, in the following pages under the descrip- tion of the Funafuti specimens of Cycloclypeus, two already described species are included in one specific type as the two stages of that species; namely, C. Guwembelianus and C. Carpenteri, the former representing the young of form A, the latter appearing to include forms A and B. The latter species was not named by Carpenter, who figured, however, a large megalospheric specimen (A), whilst he described the exceptionally large examples from Borneo which are now seen to be the adults of form B. With regard to the name which the recent Cycloclypeus should now bear, the rule of priority in any case must be considered ; but in this it curiously happens that both species date from 1851, and since Carpenter figured both forms A and B, we may there- fore keep the name by which Brady designated Carpenter's original specimens, namely C. Carpenteri. The recent Cycloclypei are limited to the station off Borneo, where they were dredged from a considerable depth by Capt. Sir * In the sense employed by Munier-Chalmas and Schlumberger. 22, MR. F. CHAPMAN ON FORAMINIFERA Edward Belcher; also from Kandavu, Fiji, 210 fathoms, form B (in the British Museum, Natural History, labelled C. Carpenteri in Dr. H. B. Brady’s handwriting) ; from Mauritius (specimens of form A found by the Author, through the kind offices of Prof. Jeffrey Bell, on a Turbinaria in the British Museum); and from the Solomon Islands, amongst specimens which Capt. Guppy had sent Sir John Murray * (a specimen of form A). There is also an irregular, annulate variety in the British Museum, probably of O. Oarpenteri (form A), from the Macclesfield Bank, China Sea, 30 fathoms. The genus Oycloclypeus is also represented as fossils in the Tertiary limestones of early date on the S.E. coast of Arabia (Carter) ; from the Tertiaries of Java (Martin) ; and from similar rocks in Borneo (Newton & Holland). CycLocLyPeus CarPENTERI, Brady. (Pl. 2. figs. 6,7; Pl. 3. figs. 1-5.) Form A. Cycloclypeus, Carpenter, 1856, Phil. Trans. vol. exlvi. p. 155, pl. xxx. fies. 1 & 8; idem, 1862, Introd. Study Foram. p. 292, pl. xix. fig. 2. C. Guembelianus, Brady, 1881, Quart. Journ. Micr. Sci., n. s., vol. xxi. p- 66 (young form); idem, 1884, Rep. Chall. vol. ix. p, 751, pl. cxi. fies. 8a, b (young form). C. Carpenteri, Brady, Lister, 1895, Phil. Trans. vol. 1868. pp. 487, 438, pl. ix. figs. 52-54. C. Guembelianus, Brady, Verbeek and Fennema, Descr. géol. Java et Madoura, vol. i. pl. ix. fig. 127. Form B. Cycloclypeus, Carpenter, 1852, Phil. Trans. vol. cxlvi. p. 555 ; idem, 1862, Introd. Study Foram. p. 292. C. Carpentert, Brady, 1881, Quart. Journ. Micr. Sci., n. s., vol. xxi. p. 67 ; idem, 1884, Rep. Chall. vol. ix. p. 751; Martin, 1896, Jaarb. Mijn. Ned. Oost-Ind. pp. 53, 54. The series of specimens, which, from a biological standpoint, is referable to one species, consists of two dimorphic types, each represented by immature and adult specimens, with here and there some gradational links. It is a significant fact, which asserted itself during the examination of the specimens from Funafuti, and is borne out by material from other localities, that the young and the adult stages of the shell in each form, A and * To whom the author is indebted for the opportunity of examining these and other specimens at the ‘Chalienger’ Office. FROM THE FUNAFUTI ATOLL. 23 B, are much more numerous than the specimens which link them together in the intermediate stages of growth; and this seems to be further corroborated by the internal shell-structure, which shows an apparent resting-stage in the shell-development, bounded by an irregular periphery, which differs from the broken and subsequently repaired rings of chamberlets. Description * :— Form A (megalospheric form).—Test discoidal, outline circular, or nearly so; the central area umbonate or biconvex, in full- grown specimens occupying about one-third the entire diameter of the shell. The initial series of chamberlets are enclosed in a secondary growth of shell-substance which gives rise to the um- bonate shape of the central area of the disc. Beyond the central thickened area the annuli of the chamberlets are thin and com- pressed, forming a broad flange around the centre. For the first two or three annuli the chamberlets are approximately square, or even laterally elongated, but they subsequently become more radially elongated, not to so great an extent, however, as in the microspheric form next described. Diameter of test 7, inch to 75 inch. The megalosphere of a typical specimen of form A from Funafuti measured 260 x 240. It is ovate, with one end pointed and the other rounded, and it lies in the concave sides of two crescent-shaped chambers, following upon which are the annuli of small chamberlets f. The nucleus observed by Lister was seen in one case to occupy the megalosphere, and in another it was found in the second chamber; that in the megalosphere was of an oval shape, measuring 60x40, whilst the nucleus seen in the second chamber was nearly spherical, measuring about 85 » in diameter. In the specimens from Funafuti which I decalcified and stained with picro-carmine no nucleus was observed in the central area of the disc, but irregular and ovoid nuclear bodies were seen in the last series of annuli, in many cases extending through an adjacent chamber. The structure of the nucleus was somewhat granulate with darker enclosures, like those observed by Lister. * ‘Dr. Brady’s description of the young of form A (= C. Guembelianus, Brady) now applies to the central’ area of the disc of full-grown specimens of the megalospheric type. See p. 751, Rep. Chall., vol. ix. 1884. ; t See also Lister, op. cit. p. 438. 24 MR. F. CHAPMAN ON FORAMINIFERA Form B (microspheric form).—Test discoidal, outline circular ; in young individuals very thin, and with the surface granulate. The central area in young and middle-sized specimens is occupied by a small boss, about one-half the diameter of that in the megalospheric form ; after the test has attained the diameter of about 15 mm., this central prominence is soon levelled over, since at that period the animal thickens its superficial shell-layer, and covers over any irregularity of surface, including the papillate ends of the non-tubulous cones of the shell-wall. Diameter of test =, to 23 inches. In C. Carpenteri, especially in the microspheric form (B), the concentric rings of chamberlets are frequently incomplete owing to the fracture of the periphery during the growth of the animal, which repairs the edge at first by a series of chamberlets parallel with the broken edge, the length of each chamberlet being in succeeding annuli of increasing proportion in the parts requiring more material to bring the contour of the disc back again to its circular form. In this ability to repair the fracture, to which it is so liable on account of its form and thinness, it resembles Orbi- tolites in the porcellanous group, and especially O. tenuissima, Carpenter, who has already pointed out * that this phenomenon is commoner in the genus Cycloclypeus than in Orbitolites. The microsphere of a typical specimen of form B from Funafuti measured 140 p in diameter. It is difficult to obtain specimens of the full-grown micro- spheric form which show the central chamber intact, for in nearly all cases the median area, containing the sarcode, has been cleared away by a species of Cliona. Distribution around Funafuti, and in other localities in. the Pacifie.—The general geological and geographical distribution of Cycloclypeus has been already given (see anted, pp. 21, 22). Cycloclypeus Carpenteri has never yet occurred in the lagoon dredgings, although nearly all the other genera of Foraminifera found at Funafuti have been found irrespectively inside the lagoon and on the outer side of the reef. The depths at which Cycloclypeus has been found in the dredg- ings made by Messrs. Halligan and Finckh are as follows :— The young of both forms (A and B) from 30-200 fathoms. The fully-developed form of A from 30-200 fathoms. The fully-grown microspheric form of B from 46—200 fathoms. * Phil. Trans. vol. cxlvi. (1856) p. 556. FROM THE FUNAFUTI ATOLL, 25 At a depth of from 50-60 fathoms large quantities of both A and B were dredged in equal proportion. The immature stage of form A is common in some dredgings around Funafuti, where the fourm B is sometimes entirely absent. This sma!l form, the original of Dr. Brady’s C. Guembelianus, is seen to graduate into the larger forms of the same type, which when fully-grown attain a diameter of 4 inch (11 mm.). These fully-grown specimens had not been observed by Brady in the ‘Challenger’ collection, for he remarked * :—‘* Notwitnstanding their minute dimensions in comparison with the only other recent species hitherto obtained, they are to all appearances fully grown.” Brady, however, appears to have seen the adult specimens later, since there are some examples in the Brady collection at the British Museum (Natural History) of this type, probably obtained by him on his visit to Fiji. The larger of those specimens examined by Lister appears to be a fully-grown individual of the form A; these were obtained from the 8.E. of Nomuka in the Tonga Islands, at a depth of 20- 40 fathoms. The young of form A (=C. Guembelianus) was recorded in vol. ix. of the ‘Challenger’ Reports} from Station 174 C, off Kandavu, Fiji Islands, at 210 fathoms. in the ‘Summary of Results,’ however, the record reads “ Station 172 A, off Tongatabu, at 240 fathoms” f. The various records for Cycloclypeus Carpenter: around Funafuti are as follows, the localities being taken in rotation N. E. 8. & W.:— North of Pava Id.—At depths varying from 36 to 63. fms. At 36 fathoms examples of Cycloclypeus consisted chietly of the immature forms of the megalospheric type. A specimen (adult form A) from 63 fathoms is partially enclosed in an encrustation of Lithothamnion—an illustration of the consolidation of reefs by the intergrowth.and overgrowth of organisms. Off Funamanu Id. (Beacon Id.).—Form A occurs at 50 fathoms, very common, especially the young furm; at 8U fathoms there were very large examples of form B, encrusted with Polyzoa, and with Carpenteria and other adherent Foraminifera ; and at 150 fathoms good typical examples, fully developed, of form A. * Rep. ‘ Challenger,’ vol, ix. p. 751. t Pp. 92, 93, & 752. ¢~ Summary of Results, 1st part, p. 631. 26 MR. F. CHAPMAN ON FORAMINIFERA S.W. end of Funafuti—At 45-51 fathoms, “a bottom of hard rock ” (label on specimen). The tests are here stained a dark rich green in the central area, due to the presence of alg in the chamberlets. Off Tutanga Id.—Form A occurred at 35, 41, and 46 fathoms (at 41 fathoms numerous specimens of both forms were found resting on pink Lithothamnion in their places of actual growth) ; at 50-60 fathoms in association with form B; and at 200 fathoms. Form B was obtained at 46 fathoms; at 50-60 fathoms im asso- ciation with form A, both very abundant; at 135 fathoms, common; and at 200 fathoms in a pteropod ooze, in association with the form A and at all stages of growth. Off Fuafatu—Form A, 80-120 fathoms, common; at 40 fathoms the specimens are stained, especially towards the central area, witha dark green (? symbiotic) alga; at 60 fathoms, common. Form B, at 60 fathoms, common; at 119 fathoms, a fragment. The work of describing these specimens has been mainly earried out in the Geol. Division of the Royal College of Science, to which Institution the collections have been sent. All recent specimens described in this paper will be eventually placed in the British Museum (Natural History). EXPLANATION OF THE PLATHS. Puate 1. Fig. 1. Sagenella frondescens, Brady. A fragment of the branching test, detached, and viewed by transmitted light. The wall of the test over the sarcode-cavity is perforated by thread-like boring algw. 50. 2. S. frondescens, Brady. A fragment of the central area of the wall more highly magnified, exhibiting a reticulose character, and with a boring alga at #7. x 300. 3. Bdelloidina aggregata, Carter. X 4. 4. Valvulina Davidiana, sp. nov. X 25. 5. Peneroplis (Monalysidium) polita, subgen. et sp. nov. X 50. 6. dj * Sollast, re 50 x 50. 7 a-c. Spirillina spinigera, sp. nov. X 50. 8 a-c. i tuberculato-limbata, sp.nov. X 50, 9 a-c. Discorbina tuberocapitata, sp. nov. X 50. 10. Culearina hispida, Brady, var. pulchella, nov. X 18. PuatTE 2. Fig. 1. Sagenella frondescens, Brady, attached to test of Cycloclypeus Carpentert Brady. x 5. 2. A prolific growth of Sagenella frondescens attached to a large specimen of the microspheric form of C. Carpenteri. X 2. Chapman. Linn Soc. Journ. Zoor. Vou.XXYIIT Pr, I FP, Chapman ad nat.del, MP Parker lith. Geo. West & Sons imp. FORAMINIFERA FROM FUNAFUTI. Linn. Soc. JouRN. ZOoL. VoL. XXVIII. PL. 2: C CHAPMAN. Collo. is Morgan & Kidd, Richmond, S.W. F. C. Pho tomicr. SAL\YENINA, CARPENTERIA, POLYTREMA AND GY CLOCEVESUs From FUNAFUTI. \ PL, B. LINN. Soc. JOURN. ZOOL. VoL. XXYIII. HAPMAN. C.G & Kidd, Richmond, S.W. Collo. by Morgan F. C. Photomicr. HETEROSTEGINA anp CYCLOCLYPEUS FROM FUNAFUTI. Chapman. Lixw Soc. Journ. Zoo. Vor. XXVIII Pr. 4. % F Shapman ad nat.del. MP Parker lith, Geo Weet & Sons imp SCARPENTERIA AND POLYTREMA FROM FUNAPUTT. Fig. Fig. 1, ne o> aT SD ON FROM THE FUNAFUTI ATOLL. 27 . Polytrema miniaceum (Pallas), var. involva, nov. A section through part of a laminated mass, similar to that sometimes found intergrown with Lithothamnion. From trial pit, seaward face, Funafuti. x 36. . Carpenteria utricularis, Carter. A vertical section of the test, showing the papillate exterior and chambers partly filled with sponge-spicules. x 36. . Carpenteria monticularis, Carter. A vertical section taken through the base of the test, showing the smooth external surface and coarse and fine tubuli. xX 36. . Cycloclypeus Carpenteri, Brady. A full-grown specimen of the megalo- spheric form (A). X 6. . C. Carpenteri, Brady. A vertical section of the test, showing the extraordinary thickness of the secondary layer of shell. The thin flange-like periphery has broken away. x 10. PuatE 3. Two series of Cycloclypeus Carpenteri, Brady, showing the various stages of growth in forms A (megalospheric) and B (microspheric) respectively. 3 nat. size. . The central portion of a microspheric form of C. Carpenteri, Brady, cut a little out of the median plane, showing an interesting resemblance in its commencement to Heterostegina, to which genus it bears some relationship. xX 36. . A half-grown specimen of the microspheriec form of Cycloclypeus Carpentert. X 5. . A megalospheric form of C. Carpenteri in median section. The shell has been badly broken at various stages of growth and repaired again. x 16. . The sarcode of C. Carpentert (form A). xX 10. . A microspheric form of Heterostegina depressa, d’Orbigny. X 10. . A megalospheric form of H. depressa. X 10. Purate 4. All magnified 5 diameters. Carpenteria balaniformis, Gray. Lateral aspect. Off Funafuti, 115- 200 fathoms, . C. balaniformis, Gray. Oral aspect. Off Funafuti, 115-200 fathoms. . C. utricularis, Carter. Coloured by an alga. Lateral aspect. 8 fathoms. Off Funamanu (Beacon Id.). . C. utricularis, Carter. Coloured by an alga. Oral aspect. Off Funamanu. 80 fathoms. . C. monticularis, Carter. Oral aspect. Off Funamanu. 80 fathoms. . C. monticwlaris, Carter. Oral aspect. Off Funamanu. 80 fathoms. . Polytrema miniaceum (Pallas), Off Funamanu. 150 fathoms. 28 MR. F. CHAPMAN ON FORAMINIFERA FROM On some Foraminifera of Tithonian Age from the Stramberg Limestone of Nesselsdorf. By FrepErick Cuapmay, A.L.S., F.R.M.S. [Read 1st March, 1900.] (Patz 5.) Dunrine the past year or so I have been favoured by Dr. M. Reme=, of Olmiitz, with some specimens of Foraminifera and a sample of foraminiferal material from the marl formed by the weathering of the Red Limestone ot Nesselsdorf. Dr. Perner has already published a paper on Tithonian Foraminifera from the Stramberg Limestone*, in which he describes three species from the red marly limestone, namely; Bulimina variabilis, dOrbigny, Cristellaria varians, Bornemann, and C. rotulata (Lam.). Beyond this record we have no infor- mation of the Foraminifera trom beds of this particular age. Consequently the following small series is of exceptional interest. FORAMINIFERA. Family LITUOLIDA. Saobfamily LiItrvoLina. Haptorpnragemium, Reuss [1860]. HaAPLOPHRAGMIUM AGGLUTINANS (d’Orbigny). (PI. 5. fig. 1.) Spirolina agglutinans, d’Orbigny, 1846, Foram. Foss. Vienne, p. 187, pl. vii. figs. 10-12. The specimen before us is somewhat irregular in growth, but the chief characters of the test resemble those of the above species. Dating from the Lower Carboniferous rocks, this species ranges upwards in fossiliferous strata to the present day. It has also occurred in the mixed faunas from the Aptian beds of Surrey (Chapman). One specimen from the Red Limestone of Nesselsdorf. * Bull. Internat. Acad..Sc. Bohéme, 1898, pp. (1-8), pl. i. THE STRAMBERG LIMESTONE OF NESSELSDORE. 29 HAPLOPHRAGMIUM NEOCOMIANUM, Chapman. (PI. 5. fig. 2.) Haplophragmium neocomianum, Chapman, 1894, Quart. Journ, Geol. Soe. vol. 1. p. 695, pl. xxxiv. figs. 2a, 6. This species was originally described from the Aptian beds of Surrey, where it was fairly numerous. The present specimen very closely resembles the original figure. One specimen from the Red Limestone of Nesselsdorf. Subfamily TROCHAMMININA. AmMoptscus, Reuss [1861]. AM™MODISCUS INCERTUS (d’Orbigny). (PI. 5. figs.3 & 4.) Operculina incertus, VOrbigny, 1839, Foram. Cuba, p. 71, pl. vi. fis. 16 & 17. Ammodiscus incertus (d’Orb.), Wisniowski, 1890, Pamietnik Akad. Umiejet Krakow, vol. xvii. p. 190, pl. viii. (i.) figs. 11a, b. This well-known and variable species is numerous in the present series. The two specimens here figured apparently belong to the megalospheric and microspheric forms respectively. Common in the Red Limestone of Nesselsdorf. Subfamily ENDOTHYRING. Invotutina, Terquem [1862]. Invoturina Remesrana, sp.nov. (PI. 5. figs. 5 a-c.) Test conical, depressed, consisting of a simple coiled tube of about five whorls, the sutures distinct. The inferior surface slightly convex and covered with papille of exogenous shell- growth, excepting the last whorl, which is marked on the periphery with sharp furrows at right angles to the edge. The test is finely perforate on the inner parts of the tubes and calcareo-arenaceous on the outer. The aperture opens on the inferior side of the test. Diameter 5; inch (‘93 mm.); height ;1, inch (-208 mm.). The species Involutina Jonesi, Terquem & Piette*, is in some respects similar to this form, for example in the character of the granulations, but the outline of the test is very dissimilar. I have named this species in honour of Dr. Remes, of Olmiitz, to wh 1 we are indebted for this interesting series of specimens. Common in the Red Limestone of Nesselsdorf. * Mém, Acad, Imp. Metz, vol. xlii. 1862, p. 461, pl. vi. figs. 22 ac. 30 MR. F. CHAPMAN ON FORAMINIFERA FROM Invotutina contca, Schlumberger. (PI. 5. figs. 6a, 0.) Involutina conica, Schlumberger, 1898, Feuille Jeunes Naturalistes, ser. 3, Ann. 28, No. 382, pp. (1, 2), figs. 1-3. This species has been lately described from the Great Oolite between Villers-sur-Mer and Caen by M. Schlumberger. The specimens now before us are, if anything, larger than those from France. Frequent in the Red Limestone of Nesselsdorf. Family TEXTULARIIDA. Subfamily TEXTULARIING. Vanvuttna, d’Orbigny [1826]. VALVULINA CUNEIFORMIS, sp. nov. (PI. 5. figs. 7a, 0.) Test conical, flattened on opposite sides; septal face convex, with a valve-like aperture. Chambers numerous, narrow, and alternate in three series. Length of test <4, inch (‘373 mm.). Measurement across the long and short axes of oral face of test #. inch (463 mm.) and yz inch ('357 mm.) respectively. This species is somewhat like Textularia conica, d’Orbigny *, in its general shape. It differs, however, in the numerical arrangement of the chambers; the aperture also is characteristic of Valvulina. Another form which V. cuneiformis resembles is Valvulina paleotrochus, var. compressa, Brady 7, from the Carboniferous formation, but the latter is strongly concave on the apertural face. Frequent from the Red Limestone of Nesselsdorf. Family LAGENID 2. Subfamily NoDOSARIING. Lineuina, d’ Orbigny [1826]. LINGULINA NoDoSARIA, Reuss. (Pl. 5. figs. 8a, 6.) Lingulina nodosaria, Reuss, 1862, Sitzungsb. Ak. Wiss. Wien, vol. xlvi. p- 59, pl. v. figs. 12a, 0. he specimen here figured consists of two chambers only, and * Foram. Cuba, 1839, p. 143, pl. i. figs. 19 & 20. + Carboniferous Foram. (Pal. Soc.), vol. xxx. 1876, p. 89, pl. iv. figs. 5a, 0. THE STRAMBERG LIMESTONE OF NESSELSDORF. 3l is probably an immature example. This species has been found in the Aptian and Albian of Germany, France, and England. One specimen from the Red Limestone of Nesselsdorf. Linevrina ovaris, Schwager. (PI. 5. figs. 9a, b.) Lingulina ovalis, Schwager, 1865, Jahresh. Ver. vaterl. Nat. Wiirtt. vol. xxi. p. 116, pl. iv. figs. 21-24. This is here represented by a fragmentary example. It con- sists of three chambers, flattened and ovate, and more nearly resembles Schwager’s figures than others described from similar deposits. Schwager obtained his specimens from the Lower Oxfordian of Gruibingen and Weissenberg. One specimen from the Red Limestone of Nesselsdorf. Vaeinutina, d’Orbigny [1826]. VAGINULINA tRUNCATA, Reuss. (PI. 5. fig. 10.) Vaginulina truncata, Reuss, 1862, Sitzungsb. Ak. Wiss. Wien, vol. xlvi. p- 47, pl. i. fig. 9. This species now has a geological range from the Tithonian to the Ceuomanian. One specimen from the Red Limestone of Nesselsdorf. CristeLLaria, Lamarck [1816]. CRISTELLARIA Bronnt (Romer). (Pl. 5. fig. 11.) Planularia Bronni, Romer, 1841, Verstein. nordd. Kreidegeb. p. 97, pl. xv. fig. 14. This is a well-known Cretaceous species. One specimen from the Red Limestone of Nesselsdorf. CRISTELLARIA CALVA, Wisniowski. (PI. 5. fig. 12.) Cristellaria calva, Wisniowski, 1890, Pamietnik Akad. Umiejet Krakow, vol. xvii. p. 223, pl. x. (ili.) figs. 4a, 6. The above species was originally described from the Ornatus- zone of the Upper Jurassic of Poland. One specimen from the Red Limestone of Nesselsdorf. CRISTELLARIA GIBBA, @’ Orbigny. (Pl. 5. fig. 18.) Cristellaria gibba, d’Orbigny, 1839, Foram. Cuba, p. 63, pl. vii. figs. 20, 21. Several more or less distorted varieties of this species occur in the series before us. The typical, elongated form is also present, one of which is figured. Common in the Red Limestone of Nesselsdorf. 32 ON FORAMINIFERA FROM THE STRAMBERG LIMESTONE. CrISTELLARIA ROTULATA (Lamarck). (PI. 5. fig. 14.) Lenticulites rotulata, Lamarck, 1804, Annales du Muséum, vol. v. p. 188. no. 3; Tabl. Encycl. et Méthod. pl. cecclxvi. fig. 5. This widely distributed and very common form has already been recorded from the Stramberg limestone by Dr. Perner. Common in the Red Limestone of Nesselsdorf. CRISTELLARIA CULTRATA (Montfort). (Pl. 5. fig. 15.) Robulus cultratus, Montfort, 1808, Conchyl. Systém. vol i. p. 214, 54° genre. This species ranges from the Lias upwards. Frequent in the Red Limestone of Nesselsdorf. EXPLANATION OF PLATE 5. Fig. 1. Haplophragmium agglutinans (dOrb.). xX 30. 2. ‘i neocomianum, Chapman. X 30. 3. Ammodiscus incertus (d’Orb.). Form A. xX 380. 4. oh 3 ¥5 Form B. x 30. 5. Involutina Remesiana, sp. noy.: a, superior aspect ; 0, inferior aspect ; ¢, peripheral aspect. X 80. 6. Involutina conica, Schlumberger: a, superior aspect; 6, lateral aspect. x 30. 7. Valvulina cuneiformis, sp. noy.: a, superior aspect; 0, oral aspect. x 30. 8. Lingulina nodosaria, Reuss: a, lateral aspect ; 0, oral aspect. X 30. 9. 55 ovalis, Schwager: a, lateral aspect ; 0, oral aspect. X 30. 10. Vaginulina truncata, Reuss. X 30. 11. Cristellaria Bronni (Romer). X 30. 12. hi calva, Wisniowski. x 15. 13. gibba, VOrb, xX 380. 14. Ms rotulata (Lam.). X 80. 15. fs cultrata (Mont.). X 30. Chapman . Linn. Soc, Journ. Zoon. Vor. XXVILLPLS. 19 oe ad.nat.del. _ West,Newman imp. ARHammond lith. FORAMINIFERA FROM NESSELDORF (TITHONIAN). Pits we ty ON A NEW SPECIES OF PLANARIAN WORM. 3B Ruyncuopemus Howest: a new European Species of Terrestrial Planarian Worm. By R. F. Scuarrr, Ph.D., B.Se., Keeper Nat. Hist. Coll., National Museum, Dublin.* [Read 15th February, 1900.] (PLATE 6.) I rounp this worm during the autumn of 1899 in the Pyrenees, close to the village of Haux Chaudes. This village, famed for its medicinal waters, lies at a height of about 2000 feet above sea- level, in a most picturesque valley surrounded by dense forests. The worm was discovered under a stone coiled round the shell of a Helix nemoralis. The snail had apparently been overcome by the worm previous to being devoured, as it had retired far up into its shell and exhibited little sign of life. When I released the snail, the worm slowly crawled away, leaving an abundant track of mucus like that of a slug. I had hoped to be able to find out something more about it after a closer examination, but owing to the very dark pigment in the skin, no eyes were visible, nor were any openings to be seen on the underside. However, I had no doubt of the true nature of the worm, the general shape of the body indicating a terrestrial planarian. I believed that I had before me a gigantic member of the genus Rhyncho- demus, and this surmise has now proved to be correct. The width of the worm was 5 millim., and the length no less than 130 millim. (about 5 inches)—just twice as long as the largest Fehynchodemus hitherto known to science. The colour of the upper side was of a uniform greyish black (Pl. 6. fig. 1). The under side (PI. 6. fig. 2) is made up of the median sole, aptly called “ Kriechleiste ’’ by Prof. von Graff, since it is a somewhat raised plate on which the animal creeps. It occupies one third of the ventral side of the body (Pl. 6. fig. 2, s) and is coloured light grey; while on each side we find a dark grey stripe bordered by a brownish-grey marginal one. The latter, which will be referred to again later on, is von Graff's ‘“ Driisenkante” or glandular ridge. The dorsal and ventral surfaces are sharply distinct in contour, the former being convexly rounded, the latter almost flat. The anterior end of the body is not directly con- cerned with the creeping movements, and the sole does not extend * Communicated by Prof. G. B. Howes, Sec. L.S. LINN. JOURN.—ZOOLOGY, VOL. XXVIII. 3 34 DR. R. F. SCHARFF ON A NEW into it. It is finely pointed and somewhat cylindrical, and is the seat of the principal sensory organs of the body. As the animal moves about, it is kept off the ground and acts as a tactile organ much like the tentacles of a snail. In some of the Planarian worms, such as Placocephalus kewensis (Bipalium kewense), which has been taken in a few English and Irish greenhouses, this portion of the body assumes a characteristic cheese-cutter shape. The posterior end of the body of our worm is but slightly attenuated, and the sole is continued to the extreme tip. These were all the observations that I could make during the life of the single specimen in my possession *. I decided there- fore to kill it without further delay, and placed it in a 4-per- cent. solution of commercial formalin. The immediate effect was a very considerable shrinkage of the specimen, which became reduced to a length of 75 millim. and a width of 43 millim.t One advantage this treatment had on the worm was to reveal two openings on the underside. From the first, situated 40. millim. from the anterior end, a triangular body, viz., the pharynx (Pl. 6. fig. 2, ph.), had been partially forced out by the. con- traction. This larger opening is therefore the mouth, and the much more minute one (g), 13 millim. behind it, the genital pore. Being unable to perceive any trace of eyes or even of a sensory sroove at the anterior end of the body, I felt it would have been impossible to determine the specimen with any degree of cer- tainty without cutting sections. In this dilemma my friend Prof. Howes offered me help. He kindly commissioned one of his students, Mr. H. H. Swinnerton, to cut me a number of microscopic sections, and to their generous assistance I owe the pleasure of being able to describe this interesting specimen, and add another species to the known European Land-planarians. The name Rhynchodemus Howesi is therefore appropriate. The recent publication of Prof. von Graff's magnificent mono- graph on the Land-planarians (2), has rendered the identification of my specimen a comparatively easy task. To give a complete account of its anatomy would be quite impossible, as the single * It has now been deposited in the Dublin Museum. + Unfortunately, the vessel containing my specimen broke on the journey from the Pyrenees to Bordeaux, but the officials of the Museum in the latter town most liberally offered me a new jar with alcohol. Owing to these adverse circumstances the preservation of the worm is not altogether satisfactory. SPECIES OF PLANARIAN WORM. 35 specimen could not be entirely sacrificed to the microtome. I content myself therefore with giving a short sketch of some of the more important anatomical features which could be gathered from a careful examination of the sections. The head and a portion of the body containing the pharynx were cut into trans- verse sections, whilst a small part bearing the genital pore was utilized for longitudinal ones. From these the following obser- vations were noted :— One of the most striking objects in the series of sections near the anterior end of the body is the pair of eyes. If we examine such-a section more closely (Pl. 6. fig. 3), we notice externally the epzdermis (ep.). Some previous writers, especially Prof. S. Moseley (4) and Dendy (1), have referred to the great difficulty in obtaining a clear idea of the structure of this layer from ordinary cross sections. My single specimen, as already stated, had not been fixed in a satisfactory manner for histo- logical purposes. But whatever method is employed for fixing, the chief difficulty in recognizing the structure of the epithelium _ ‘lies in the fact that the greater part of it is filled with rod-like bodies, while numerous glands open between the cells to carry - their secretions to the exterior. The epithelium consists of a single layer of cells which, in this part of the body, are about equally high on the dorsal and ventral surface. The sole (s)— von Graff's “ Kriechleiste ”—possesses a ciliated epithelium, but cilia appear to be confined to this part of the under surface, The most noteworthy features in connection with this sole are the sensory grooves (sq.), which I failed to make out from a macro- scopic examination of the worm. They are probably united in front in the form of a horseshoe, the two branches being con- tinued backward for a little distance beyond the eyes on each ‘side of the sole. These sensory grooves have been demonstrated in the great majority of Land-planarians by Prof. von Graff. As a rule there are also sensory pits in connection with the grooves, but according to the same author (2. p. 42) these are absent in Rhynchodemus bilineatus, Rh. Scharffi, Rh. nematoides, and Rh. ochroleucus. No trace of such pits could be seen in the cross- sections of Rh. Howesi, so that it agrees in this respect with some of the other species of the genus. Almost all previous writers on the Land-planarians have referred to the peculiar rod-like “ Rhabdites,” which have becu compared to the stinging-organs of Ceelenterates, but whose a 36 DR. R. F. SCHARFF ON A NEW precise function is still unknown. It has been suggested that they serve for the capture of the prey; and this explanation is supported bv the observations of Dendy (1) and Woodworth (6). Prof. von Graff distinguishes three varieties of rod-like bodies (2. p. 55), viz.:—“ Rhabditen, Rhamniten, and Chondrocysten,” but observes that there is no real difference in kind—that the Rhamnites and Chondrocysts in fact only indicate the extremes in both directions of the true Rhabdites. The Rhabdites are short and stout, the Rhamnites longer and much more slender, while the Chondrocysts are oval or club-shaped masses. I could clearly trace the latter form in many of the sections and also the Rhabdites, but failed to perceive any rods that could be brought under the second denomination. As regards the glands which lie in the connective-tissue and open on the surface of the epidermis, only one kind is noticeable in the section referred to. They convey mucus to the skin, and become deeply stained by hematoxylin and less so by borax- carmine, which were the reagents used. Portions of mucus in the ducts between the epithelial cells are frequently seen, and lead one to mistake them for Rhabdites, but the latter are pointed at each end, and by that means can be distinguished. In the more posterior sections very large glands are noticed near the margins of the lower body-surface. These—the “' Kanten- driisen ” of Graff—are confined to that part of the body, and will be referred to later on. The muscular and nervous systems are dealt with so exhaus- tively by Prof. von Graff, Prof. Dendy, and others, and there appears to be so little difference in the various species, that a detailed description of the necessarily incomplete observations I have been able to make, appears undesirable. Tsere are a deep and a superficial muscular system, while the nervous system consists of two longitudinal cords, with an anterior bilobed gang- lionic mass or brain. The section (PI. 6. fig. 3) being slightly oblique, only one half of the brain (0) is visible. Close to it on the dorsal portion of the head lie the eyes, one on each side. One of them (e) is represented in the section. Prof. von Graft has drawn attention to the fact that two very distinct types of eyes occur among the Land-planarians. These two types he called ‘‘ Invertirte Kolbenaugen” and “ Retina-augen.” In the first type, the retina is placed in front of the eye, and the sensitive cells are turned away from the source of light, hence the term SPECIES OF PLANARIAN WORM. 37 “inverted club-shaped eyes.” The eyes of Rh. Howesi belong to the other or “retina” type. It consists simply of a cup-shaped organ internally coated with small pigment-cells and surrounded by a retina, so that the latter is placed facing the light. The open part of the cup is covered by a non-pigmented layer of cells—a kind of cornea. The latter has a thickness of 0°012 millim.; while the whole eye is 0°16 millim. long, and at its widest part 0°L08 millim. broad. The interior of the eye is partly filled with a peculiar cellular substance, whose true nature could not be ascertained from any of the sections available. I had no means of comparing the sections with longitudinal ones, and must presume that this structure is similar to that figured by Prof. von Graff (2. pl. 50. fig. 1). He explains (p. 144) that it is composed of a bundle of elongated columns, one end of which converges towards the centre of the eye, whilst the others are connected with the retina-cells by means of fine prolonga- tions which pass outward between the pigment-cells. Connective-tissue (Pl. 6. fig. 3, ¢.¢.) binds together the various organs and structures lying within the epidermis and maintains them in position. It consists of a mass of fibres containing nuclei, and bearing also the generaliy small pigment granules which pro- duce the black colour of the skin. The digestive organs do not extend to the most anterior part of the body, and are therefore not visible in the section (PI. 6. fig. 3), but a little farther back we find a considerable part of the transverse sections occupied by the alimentary canal (text-fig. 1, a). All Land-planarians and also the marine and freshwater Triclads agree in the possession of an alimentary canal consisting of three main branches and a muscular pharynx. One of the main branches runs anteriorly, the two others (text-fig. 1, r.al., l.al.) occupy the posterior portion of the body, and they all join in a common opening in front of the pharyngeal sac. I have already mentioned that the pharynx had become pro- truded through the mouth during the violent contractions when the worm was preserved. The mouth, as stated, is situated 40 millim. behind the anterior end of the body, and is indicated by a minute transverse slit scarcely visible in the living animal. The large pharynx, as seen in text-fig. 1 (ph), is somewhat cylindrical: and conically pointed at the free end. ‘Che transverse section figured is close to the external pharyngeal opening, and shows peculiar folding of the pharyngeal epithelium. Cilia were 38 DR. R. F. SCHARFF ON A NEW noticed only on the conical end portion of the pharynx. All the three branches of the alimentary canal give off lateral diverticula. None of them appear to be ciliated. There are a few other points in the section referred to (cf. text-figs.) which deserve to Fig. |.— Rhynchodemus Howesi. Transverse section through mouth (somewhat diagrammatic). x 70. glep, glandular epithelium; gi, glands; v.d, vas deferens; ov, oviduct; 2.¢,. nerve-cord ; ph, pharynx; 7.a/, right branch of alimentary canal; J.a/, left branch of alimentary canal. be mentioned. I have already briefly alluded to the glands which are so very abundant in the section near the anterior end. But farther back, where the underside is divided into three distinct areas, a number of large glands are seen to open on the surface towards the sides of the body. These are roughly indi- cated on the transverse section (text-fig. 1, g/). They are what von Graff described as ‘“‘ Kantendrisen,” and are strictly localized to that particular portion of the body. They are deeply stained by hematoxylin like the other glands, and their ducts have de- formed the epithelial cells between which the glands conduct their secretions to the exterior. In one important respect these epithelial cells differ from all those described by von Graff, namely, in size. He remarks SPECIES OF PLANARIAN WORM. 39 (2. p. 48) that the height of the “ Driisenkanten-epithel ” is less than that of the dorsal epithelium, while I find (g/.ep) that it greatly exceeds that in height. Another interesting fact worth noting is that Rh. Howesi is the only member of the genus Rhynchodemus in which these peculiar marginal glands occur. The cilia so well seen towards the middle of the underside do not extend to this glandular area. All Land-planarians are hermaphrodite, a pair of ovaries being situated near the anterior end of the body. These give rise to an oviduct on each side of the body (cf: text-figs., ov.), which generally unite before opening posteriorly into the genital atrium. The testes lie farther back than the ovaries, and the two vasa defer- entia (v.d.), carrying the spermatozoa, open as a rule separately into the vesicula seminalis (v.s.), from which a duct leads into the genital atrium. The genital opening is always situated behind the mouth. In R. Howesi it is only very faintly indicated by a minute white spot (PI. 6. fig. 2,7) on the underside of the body, 13 millim. behind the oral aperture. Fig. 2.— Rhynchodemus Howesi.—Longitudinal section through reproductive organs (somewhat diagrammatic). 50. ut, uterus; /.0v., left oviduct; 7.0v., right oviduct ; va, vagina; gl, glands ; p, penis, v.s., vesicula seminalis; g, genital pore; gl.c., glandular canal sp, spermatozoa ; 7p, duct from vesicula seminalis; 7.vd., right vas deferens ; l.ud., left vas deferens. Instead of giving.an accurate drawing of a longitudinal section of the reproductive organs, it seemed to me preferable tio show the disposition of the various parts in Rh. Howesi by means of a somewhat diagrammativ sketch (text-fig. 2). The repro- ductive organs in this species agree in the main points with those of R. terrestris and R. Scharff, but they are more nearly related 40 DR. R. F. SCHARFE ON A NEW to those of the former. The oviducts (text-fig. 1, ov) lie close to the nerve-cords (7.c), one on each side, in which position they run from the anterior ovaries backward beyond the genital pore and open independently into a wide glandular canal (text-fig. 2, gl.c.), as in Fh. terrestris. This canal is characterized by the possession of very elongated and ciliated epithelial cells between which open numerous glands (g/)—von Graff’s shell-glands. During their course the oviducts receive the contents of many yolk-glands. Every now and then, in transverse sections, short eylindrical ducts conveying food-material for the eggs are noticed opening into the oviducts. The latter, it may be mentioned, are ciliated. The glandular canal (g/.c.) opens anteriorly mto another, the vagina (va), and it is joined there by a duct from a large glandular organ—the uterus (wf). The numerous testes are arranged in a row on each side of the body close to the upper side of the nerve-cords. Two vasa deferentia (text-fig. 2, v.d.) convey the spermatozoa towards the muscular mass of the penis (p). But before they reach it they become dilated to form thin-walled convoluted ducts, which in the specimen examined were filled with masses of agglutinated spermatozoa (sp). Each of the vasa deferentia opens inde- pendently into a sac—the vesicula seminalis (v.s.)—which is surrounded by an enormous muscular mass constituting the penis (p). A narrow duct (d.p.) leads from the vesicula scminalis to the genital atrium, into which also opens the vagina. The atrium finally communicates with the exterior (g) by a short duct. The vesicula seminalis (v.s.), under a low power of the microscope, appears to be coated internally with a thick glandular epithelinm projecting far into the lumen of the organ. But a high power reveals the fact that the epithelium is thrown into short folds which are closely invested by a thick granular mass. Whether the latter is secreted by the epithelium or by glands lying on the exterior of the muscular penis, I was unable to ascertain. In the closely allied Rh. Scharffi (2. fig. 53, p. 203), the penis lies freely in the surrounding tissue, and is probably eversible to 1 considerable extent, while in that of the present species the movements would seem to be much more limited. This concludes my observations on the general structure of Rh. Howesi; and it still remains for me to make a few remarks SPECIES OF PLANABRIAN WORM. AL on its systematic position and relationship. The 348 species of Land-planarians are now divided into five great families, viz. :— Limacopside. Geoplanide. Bipaliide. Cotyloplanide. Rhynchodemide. Of which the latter is defined by Prof. von Graff as possessing two eyes near the anterior end of body, whilst tentacles, sucking- dises, and head-plates are absent. There can be no doubt, there- fore, from the description given above, that the new Pyrenean species belongs to that family. The latter includes the seven genera Rhynchodemus, Microplana, Amblyplana, Nematodemus, Platydemus, Dolichoplana, and Othelosoma. The genera Micro- plana and Amblyplana are devoid of sensory grooves ; and Nema- todemus has no sole, whilst both sensory grooves and a gole are present in the species described. The genera Platydemus and Dolichoplana have sensory pits which are absent in our species; and finally Othelosoma is distinguished from all other members of the family by the possession of a peculiarly folded and attenuated head-region. In fact the new European land-planarian undoubtedly belongs to the genus Rhynchodemus, and its nearest relation appears to be Rh. terrestris. It differs, however, from the latter in being more than 12 times as long, in the possession of rhabdites, of a marginal glandular area on the underside of the body, and in many minor structural points in the reproductive organs. As regards the geographical distribution of the family Rhyn- chodemid@, it is the only cosmopolitan one, and two of its genera, viz. Rhynchodemus and Microplana, have been found in Europe. The last genus in fact, with its single species MM. humicola (first described by Prof. Vejdovsky), is confined to our Continent. Thirty-four species of Rhynchodemus are known to science, of which five, viz., Rh. terrestris, Rh. bilineatus, Rh. pyrenaicus: Rh. albicollis, and Rh. Scharffi, occur in Europe. To these must now be added Rh. Howest. Some of them are apparently very local forms, Rh. bilineatus and Rh. albicollis having only been found in Germany, Rh. pyrenaicus and Rh. Howest in the Pyrenees, and Rk. Scharff in Ireland; but Rh. terrestris has a much wider range. It is known from Denmark, England, Treland, France, Germany, Austria, and the Balearic Islands, 42 ON A NEW SPECIES OF PLANARIAN WORM. whilst I took it myself some years ago on the Briinig Pass, in Switzerland. Though some of the European species have been discovered under somewhat suspicious circumstances, which render artificial importation probable, 1 quite agree with Prof. von Graff in the opinion that all the European Rhynchodemide are truly indigenous to our Continent. REFERENCES. 1. Denpy, A.—The Anatomy of an Australian Land-planarian- Trans. R. Soc. Victoria, vol. i. (part 2), 1889. 2. Grarr, L. von.—Monographie der Turbellarien: II. (Land- planarien), 1899. . 3. Kennet, J. von.—Die in Deutschland gefundenen Land- planarien Lthynchodemus terrestris und Geodesmus bi- lineatus. Arbeiten d. zool. Instituts Wiirzburg, vol. v., p- 120, 1879. 4. Mosenny, H. N.—On the Anatomy and Histology of the Land-planarians of Ceylon, &c. Phil. Trans. Roy. Soc. vol. 164, 1875. 5. Vespovsky, Fr.— Note sur une nouvelle Planaire terrestre (Microplana humicola). Revue Biologique du Nord de la France, vol. ii., 1890. 6. Woopwortr, W. M.—Contributions to the Morphology of the Turbellaria: I. Bull. Mus. Comp. Zool. Harvard Coll., vol. xxi., 1891. DESCRIPTION OF PLATE 6. Fig. 1. Rhynchodemus Howesi, natural size, dorsal aspect. 2. The same, natural size, ventral aspect. ph., pharynx; g, genital aperture ; s, sole. 3. The same. ‘Transverse section through anterior region, x 435. e,eye; 6, brain; ep., epidermis; m, longitudinal?muscular bundles ; 8, sole; sg., sensory groove; c¢., connective tissue. ocharitf. Linn Soc. Journ Zoon. Vou XXVIIPL6. RE S.del. i Hei MP Parker lith. Sy. Ss Geo, West & Sons imp RHYNCHODE MUS HOWEHSI, Scharff. ede hi wo: ry ON BRYOZOA FROM FRANZ-JOSEF LAND. 43: Bryozoa from Franz-Josef Land, collected by the Jackson- Harmsworth Expedition, 1896-1897. By Arraur Witt Waters, F.L.S. [Read 15th March, 1900. | (PLatEs 7-12.) CHILOSTOMATA. THE Arctic Bryozoa from various localities have been well studied, having received attention from several leading workers, including Smitt, Busk, Hincks, Lorenz, Vigelius, &. That they should have been so thoroughly examined is no doubt accounted for by the fact that, when any important expedition returns, specialists are found to work up all the material collected under trying circumstances, while collections from less accessible places are put aside without description. The question arises what should be considered Arctic, for the Arctic circle forms but an artificial zoological division, and it would be simpler to include in “ Arctic” everything within the isothermal line of 32° F. (0° C.). This brings in practically the whole of Greenland, Davis Straits, Labrador, and Iceland. The Gulf of St. Lawrence is not included ; but as many Bryozoa have been described from there, and seeing that many of them occur in Arctic regions, they must be compared. If we were to take the winter isotherm, the Gulf of St. Lawrence would be included in a line passing through the south of Greenland, and by the island of Jan Mayen. A small collection of 19 species, brought by Leigh Smith from Franz-Josef Land, was described by Ridley °. The most important works for determining the northern Bryozoa are Smitt’s series © of valuable papers published in the Proceedings of the Swedish Academy; and when examining Arctic forms we appreciate the thoroughness of his labours, although now attention is paid to several characters which were not then used, and material progress has been made in classification. Besides his papers on the Scandinavian forms, he published a descriptive list ® of 58 species collected by 1 « Polyzoa, Coelenterata, and Sponges of Franz-Josef Land,’ Ann. Mag. Nat. Hist. ser. 5, vol. vii. p. 442, pl. xxi. 2 F. A. Smitt. “ Krit. Forteck. ofver Skand. Hafs-Bryozoer,” Mifversigt af K, Vet.-Akad. Forhandl. 1865-1871 (referred to as “ Krit. Fért.”). 3 Gifversigt K. Vet.-Ak. Forh. 1878, p. 11. AA, MR. A. W. WATERS ON Nordenskiéld, Stuxberg, and Théel from Novaya Zemlya and the mouth of the Jenisei ; and another list* of the collection of 74 species made by Sandberg and Trybom from the Kola peninsula. Further *, Smitt’s “Bryozoa marina in regionibus arcticis et borealibus viventia recensuit” gives a valuable réswmé of what was then known. The late Mr. Hincks® published a paper on “Tceland and Labrador Polyzoa;” and the same author’s papers‘ on the Bryozoa from the Gulf of St. Lawrence give exact descriptions and figures of various Arctic species. Mr. Hincks furnished the list of species from Barents Sea in W. 8S. M. D’Urban’s paper’ “ Zoology of Barents Sea,’ and described some new species in the same volume of the ‘ Annals,’ p. 272. Busk ° described some Bryozoa collected from Greenland, &c. by Dr. Sutherland and 8. E. Belcher ; and the next year had a paper in the ‘ Annals’* ona collection made by Mr. McAndrew on the coasts of Norway and Finland. The same author pub- lished * in 1881 a description of Arctic Bryozoa collected by Captain Feilden. An Austrian expedition, sent out by Count Wilczek to the island of Jan Mayen, north of Iceland, collected 75 species, which were ably described by L. Lorenz®. G. Levinsen” has written a valuable paper on 51 species of Bryozoa from the Kara Sea. Vigeiius” has a list of 27 species collected by the ‘ Willem Bareuts.’ ' Gfversigt K. Vet.-Ak. Forh. 1878, p. 19. ? Ibid. 1867, p. 443. * Ann. & Mag. Nat. Hist. ser. 4, vol. xix. p. 97. Mr. Hincks afterwards found that instead of coming from Iceland the species so named were from Davis Straits. 4 Ibid. ser. 6, vol. i. p. 214, & vol. ix. p. 149. > Ibid. ser. 5, vol. vi. p. 257. ® Quart. Journ. Mier. Se. vol. iii. 1855, p. 253. 7 Ann. & Mag. Nat. Hist. ser. 2, vol. xviii. p. 32. 8 Journ. Linn. Soe., Zool. vol. xv. p. 231. ° “ Bryozoen von Jan Mayen.” O6esterreichische Polarstation Jan Mayen. Beobachtungs-Ergebnisse, vol. iii. p. 83, 1886: Internat. Polarforschung. 10 « Bryozoer fra Kara-Havet,” Dijmphna-Togtets Zool.-Bot. Udbytte, 1886, p. 807. 1 “Catalogue of the Polyzoa coll. during the Dutch N. Polar Cruise of the ‘ Willem Barents’ in the year 1878-1879,” Nied. Arch. fiir Zool., Suppl. vol. iii, BRYOZOA FROM FRANZ-JOSEF LAND. 45 Kirchenpauer ' has described Bryozoa from East Greenland: Andeer Hennig * 27 species from West Greenland; and Nord- gaard* has published a useful paper on the Bryozoa from the coasts of Norway, in which he gives a list of Arctic literature, to which I would specially call attention, as there are several papers by M. Sars, G. O. Sars, Kirchenpauer, Danielssen, and others to which the worker may have to refer, but which need not be enumerated here. Stuxberg * has given a valuable list from the district of Novaya Zemlya; but when the older nomenclature is used we cannot always be certain what was intended, and may have two names for the same form: for instance, Lezeschara subgracilis = Myriozoum subgracile. Recently Olaf Bidenkap’ has written an important paper on the Bryozoa from East Spitzbergen. HK. von Marenzeller* named twelve Bryozoa dredged by the Austrian North Pole Expedition, and a few of these are from localities a degree or two farther east than any of the Jackson- Harmsworth dredgings, though not so far north, and most are from a depth of about 200 metres. In order to better under- stand the manner in which distribution has taken place, we now want collections from the neighbourhood of New Siberia and the Behring Straits.’ Smitt, who described so many Arctic forms, undoubtedly often united under one name quite distinct species, and much time is lost in considering which are correct specific names. The difficulty is increased by so many species being based, by many authors, origimally upon the form of growth, without any + “System. Verzeich. der in Ostgrénland gesammelten Hydroiden u. Bryo- zoen” in Koldewey, ‘Die zweite deutsche Nordpolfahrt, Bd. II. Abth. 1, 1874, p. 411. 2 « Bryozoer fran Westgrénland samlade af Dr. Ohlin ‘under the Peary Auxiliary Expedition,’ ar 1894.” ifvers. K. Vetensk.-Akad. Forh. 1896. 3 “Syst. forteg. ov. de i Norge, hidtil observ. arter af Marine Polyzoa,’ Bergens Museums Aarbog 1894-5, No. 2. 4 « Faunan pa och kring Novaja Semlja, Vega-Expeditionens,” Vetenskaplige Arbeten, vol. v. 1887, pp. 100, 117, & 179. ® « Bryozoen von Ost-Spitzbergen,” Zool. Jahrb. vol. x. 1897, p. 609, pl. 25. 5 «Die Coelent., Echin. u. Wirmer der k.-k. Oest.-Ung. Nordp.-Exp.,” Denksch. Akad. Wissensch. Wien, xxxv. p. 388. 7 Since my paper was read an important contribution has been published by O. Nordgaard on “ Polyzoa,” collected in the North Atlantic, in the Reports of the Norske Nordhays-Expedition, 1867-1878, pt. xxvii. 46 MR. A. W. WATERS ON adequate descriptions of the zocecia or of the minute characters. The name is accepted, and. subsequently more detailed description has been given, though there is the possibility of this only relating to a similar form. We have an example in Céllepora incrassata of Lamarck, the name being first given to a Mediter- ranean species: but under this name Smitt described more thoroughly a northern form, which has been thus generally known, while a Mediterranean one has been described under another name. In such a case it seems better to consider this as the species of Smitt, who more fully described it; and to replace the present recognized name by another would only add confusion, and this is often the result of too pedantically endeavouring to follow rules. In many cases, purposely, only one or two references are given, even where the name has been employed by various authors ; but, as far as possible, only descriptions about which there was a reasonable certainty have been referred to, leaving out probabilities. Since Smitt published his papers, the details of the structure of the zocecia have been more carefully examined, and it has been necessary to give attention to many points which are now recognized as furnishing specific characters, for it is known that the general appearance is subject to great variation. The use of the opercula and mandibles in determining species, which I introduced in 1878 *, has been most valuable, and these characters are receiving increasing attention. The ovicells are more carefully examined than they were at the time Smitt began his work. The position and form of the rosette-plates should always be examined when circumstances permit; as also whether the connection is direct from zocecium to zocecium, or through a pore-chamber 7. The position of the rooting process is another character of considerable value. In the Cyclostomata the position and nature of the “ closures’’ are characters which should not be overlooked. The glands at the side of the oral aperture and in the avicularian chamber differ considerably in position and size according to the species, and there are many species which I should recognize from sections of the soft parts; and no doubt, as * «The Use of the Opercula in the Determination of the Cheilostomous Bryozoa,” Proc. Manch. Lit. & Phil. Soc. 1878, vol. xviii. p. 8, pl. i. + See Waters, ‘ Observations on Membraniporidx,” Linn, Soc. Journ., Zool. vol, xxvi. p. 654. BRYOZOA FROM FRANZ-JOSHF LAND. AT time advances, such sections will be more largely employed, and then internal organs may prove as useful in classification as the outside shell. The number of tentacles *, although subject to a small range, should always be given; while a more general acquaintance with the primary cell may cause a modification of views. The collection under review, which is one of considerable importance, was sent to me in 1898, but until other work was finished it was impossible to devote myself seriously to the task. Since then sections have been cut in all cases where the material was at all suitable, slides of the opercula and mandibles of most species have been prepared, and calcareous sections of a few have been made. The collection was sent to me by Mr. Bruce, the naturalist on the Expedition, by whom a large part of the material was collected in 1897; and I take this opportunity of thanking Mr. Bruce for his kindness in entrusting me with the examination of the Bryozoa. It will be well to separate the Bryozoa into four divisions, namely :— No. 1. Those from the neighbourhood of Northbrook Island, where the Expedition wintered. The localities are Giinther Sound, Cape Flora, off Klmwood, off glacier between Cape Flora and Cape Gertrude. No. 2. Near Wilezek Land, 127 fathoms. No. 3. Lat. 77° 55’ N., long. 55° 25’ E., 115 fathoms ; lat. 77° 55’ N., long. 58° 20’ E., and 53° 16' E., 180 fathoms. These, and also those collected from No. 2, were dredged by the ‘ Windward’ on its third journey to Franz-Josef Land. These last are interesting as extending our knowledge of the Bryozoa farther to the east; for, with the exception of a few specimens described by Marenzeller, no Bryozoa have been mentioned from any locality so far eastward. A further interest attaches to this part of the collection, as Mr. Bruce informs me that they were collected by Captain Brown and Mr. Edward Else, the steward, to whom Mr. Bruce had, on a previous occasion, shown the method of collecting with a “swab” and had instructed them in the preservation of the material. This shows how much useful * The tentacles can be readily counted in transverse sections, and figures obtained from sections are more reliable than those from living specimens. 48 MR. A. W. WATERS ON collecting may be done in out of the way regions by men who are not trained zoologists. No. 4. Off Cape Mary Harmsworth, 53-93 fathoms, and 234 fathoms. This is the most northerly locality whence Bryozoa have been described. In this paper special consideration has been given to the question of geographical distribution; and we find that most of the species are widely, in fact generally, distributed in the Arctic regions, forming a very distinct Arctic fauna, with a large proportion only found in the northern seas. Out of the 59 species of Chilostomata, Hippothoa hyalina is cosmopolitan, being found in the Southern hemisphere. Porella concinna may have been found in the Southern hemisphere, but of this I have some doubt. Then we have the Kscharoides Sarsii, Sm., and Cribrilina punctata, Hass., said to have been dredged in the Antarctic by the ‘Erebus’ and ‘ Terror’; and to these I have alluded at some length, and expressed my opinion as to the probability of a mistake in labelling having been made. There are only 20 species common to the British coasts, including some from the Shetland Islands ; and the strikingly small number of 5 also known from the Mediterranean. When the Cyclostomataand Ctenostomata have been described, the total number will be over 70; but this must not be looked upon as a complete list of Bryozoa from Franz-Josef Land, and I am convinced that many more encrusting species will be found upon dead shells. In this paper reference is made to work previously done in the Arctic seas, and some species not found by this Expedition are given. A point which is now attracting attention is, what species occur in the colder regions of the two hemispheres, and before long the expeditions being sent to the Antarctic will help to settle some uncertain points. Si John Murray, in his* “ Marine Fauna of the Kerguelen Region of the Great Southern Ocean,” gives a list of 16 species of Bryozoa from the Northern and Southern Hemispheres, unrecorded within the tropics; but on a critical examination this list, compiled from Busk’s ‘ Challenger ’ Report, is much reduced. * Trans. Roy. Soc. Hdin. vol. xxxyiii. p. 454. BRYOZOA FROM FRANZ-JOSEF LAND. 49 CRIBRILINA MONOCEROS, Busk, is given from Station 253, from a depth of 3125 fathoms; but my attention has been called to the fact that at this depth the calcareous shell would have been dissolved, and probably a mistake has been made in labelling. BEANIA MAGELLANICA, Busk, has a very wide distribution, and has not yet been found near the Arctic regions; but, on the other hand, occurs in the Tropics, off Cape Verd, thus removing it from the list. ESCHAROIDES VERRUCULATA, Busk, collected by the ‘ Chal- lenger’ from Heard Island, I have again examined, and find it is the Rhynchopora longirostris, Hincks, and not the same as the Cellepora verruculata, Smitt, from Florida. In my paper on Membraniporide * I showed that Mempranr- PORA CRASSIMARGINATA, Var. INCRUSTANS, Busk, so far as con- cerned one specimen, was J. lineata, L.; another specimen was what I have named MW. incrustans ; while one poor specimen probably is MW. crassemarginata, H.; and if this is the case, then the distribution thereof is Naples, Capri, Villefranche-sur-Mer, _ Madeira, Tizard (China Sea) (Kirkpatrick), Tristan da Cunha (Chall.), and a variety from Japan, so that it is confined to temperate seas. _ Membranipora perfragilis, H., to which some of the < Chal- lenger’ specimens named erassimarginata really belong, has a distribution from South to North Australia, on to Japan, and also Heard Island (Indian Ocean), thus occurring in both hemi- ‘spheres and the tropics. With regard to MEMBRANIPORA GALATEA, var. multifida, Busk, from off the Azores, I came to the conclusion, on an examina- tion some time ago, that this was not correctly placed, but did not made a note of this at the last re-examination of ‘Challenger’ specimens. J have not critically examined the KIneEroskKias cyatHus, W. Thoms., but the two localities of the ‘ Challenger’ are in temperate seas, not far from the tropics. The less we say about the Cyclostomata the better, as there is much uncertainty about their determination, and without good specimens with well-preserved ovicells we may go astray, and many of the ‘Challenger’ determinations had to be made from single and not very satisfactory specimens. The “ Hornera LICHENOIDES,” Pont., from the South Atlantic, off the River Plate, has very pronounced ridges with ribs across, and it * Journ. Linn. Soc., Zool. vol. xxvi. p. 686, LINN. JOURN.— ZOOLOGY, VOL. XXVIII. 4, 50 MR. A. W. WATERS ON would be advisable to examine and make sections of some well- preserved specimens. The ‘“Ipmonga aTuantica,’ Forbes, from off Simon’s Bay, S. Africa, does not seem to me to be this species, and the small piece from Kerguelen is also open to doubt. With regard to EnratopHora, there is so much uncertainty about their determination, that attaching a name sometimes merely means that no characters are found by which separation can be made; also no doubt these simpler forms are older, and have a wider distribution than some of the more highly differ- entiated. In the present paper the genus Scrupocellaria is enlarged to include one or two species previously placed under Menipea, a genus which has incorporated some rather divergent forms. The Schizoporella hyalina, L., is considered to belong to Hippothoa, on account of the reproductive characters. The genus Porella is well represented in the Arctic Seas; and in this genus the opercula and mandibles are found of great use in separating the species, and the large ayicularian and oral glands may be found similarly useful. The Eschara elegantula of d’Orbigny is found, upon a com- parison of d’Orbigny’s specimen, not to be identical with Smitt’s Eschara elegantula, which is in consequence left as Porella saccata, Busk. The Arctic genus Rhamphostomella, Lorenz, has a more or less triangular or oval avicularium in the peristomial elevation, and usually has a denticle in the oral aperture. It seems more nearly related to Smttia than to Cellepora. The Cellepore all belong to the group separated off as Osthimosia by Jullien, and Schismopora by MacGillivray. Of the Retepore, one is purely Arctic and the other is thought to be the same as a Mediterranean species. 1. Gemetiarta tortcata (Z.). (PI. 7. fig. 4.) A specimen from Ginther Sound, 10 fath., shows that there are creeping stolons, which at. short intervals have adnate zocecia somewhat resembling those of Pyripora catenularia, Jameson. The erect branches may for a time be simple tubes, or they may at once take the usual biserial form. I cannot agree with Mr. Hincks when he speaks of the shoots rising from bundles of fibres, as this is rather reversing the case, for from individual mature zocecia radicle fibres are produced which unite BRYOZOA FROM FRANZ-JOSEF LAND. 51 into a bundle. I find 18 tentacles, while Van Beneden and Farre speak of 10, Sars 12, Dalyell 12-14, and Vigelius 12. Loc. This is a northern form, ubiquitous in the Arctic, and occurring off the British Isles and the coast of France. Jackson-Harmsworth Exp.: Giinther Sound, 10 fath.; off Cape Gertrude, 30 fath.; nr. Wilczek Land, 127 fath.; lat. 77° 55’ N., long. 53° 20' E., 180 fath.; off Cape Mary Harms- worth, 53-93 fath., and 50 miles N.W. Cape Mary Harmsworth, 234 fath.; off glacier between Cape Flora and Cape Gertrude, 30 fath.; Cape Flora off West Bay, 8 fath. 2. BREDTIA FRIGIDA, sp. nov. (PI. 7. figs. 1-3.) There are two species of Brettia collected by the Expedition, and this one may be what Smitt figured, pl. xviii. fig. 27, and which is referred to on page 292, Gifvers. Vetensk.-Ak. Foérhandl. 1867, under Bugula quadridentata. In the explanation of the plates it is called an’Hueratea-growth of B. quadridentata from Spitzbergen, 20 fathoms, but the growth is different from that of Bugula. The zoarium grows to about three-quarters of an inch high. The zoccia are unilateral, growing from the dorsal surface, with one distal or central rosette-plate aud two lateral (fig. 3). The fresh branches arise at intervals of one, two, or three zoccia ; and the two new zoecia do not spread out much, but at first are usually nearly parallel, the one growing from slightly lower down than the other. The area is surrounded by a raised border, and occupies more than half of the length of the zowcium. Sometimes a very minute spine can be seen at the oral end on one side, but more frequently no trace is visible. The radicle tubes grow from the distal end. There are 22 tentacles. The area is quite similar to that of Bugula, but in most Bugule there is a row of distal rosette-plates, though B. Murrayana, Johnst., has connections like those of Brettia frigida. This latter resembles Maplestonia simplex, MacG., in some particulars, but there are no annulated joints. In Breftia, as first defined, there are marginal spines; but Busk, in the ‘Challenger’ Report, describes from Station 196 (North of Australia) B. australs, B., without any spines. The B. cornigera, Busk, is similar in shape to the present form, but has 4 submarginal spines, Busk undoubtedly is incorrect when he says ‘‘ springing from a common stem of radicle tubes,” for ia the Bryozoa the radicle tubes always grow from the mature zowcia. Ovicells are unknown Ay 52 MR. A. W. WATERS ON in Brettia. Hincks made a genus Corynoporella for a species somewhat similar to the present, but with an articulated avicularium at the side of the aperture. Probably the genus is superfluous. Loc. Off Cape Mary Harmsworth, 53-93 fathoms; Cape Flora off West Bay, 8 fath.; lat. 77° 55’ N., long. 53° 20' B., 130 fath. 8. BRETTIA MINIMA, sp.nov. (PI. 7. figs. 5-7.) The zoccia are longer and more delicate than in B. frigida, W., and the area occupies about one-third of the length of the zoccium. The branching takes places frequently, with an interval of one or two zocecia, and spread out at a considerable angle, nearly at a right angle, much in the same way as in Alysidium Lafontii, Aud., so that B. minima can be distinguished from B. frigida withoué a lens. The radicles start from the dorsal surface at the distal end. Loc. Jackson-Harmsworth Exp.: Cape Mary Harmsworth, 53-93 fathoms. 4. Buauta Murrayana, Johnst. From Franz-Josef Land there are specimens of characteristic B. Murrayana, and others of characteristic B. fruticosa, Pack. ; but there are many intermediate forms which make separation difficult, though perhaps there are other characters not yet noticed by which they can be distinguished. A typical specimen from off Cape Mary Harmsworth has about six marginal spines on each side, larger avicularia to the marginal zocecia; the central avicularia are shorter than in the jruticosa variety. The branches are composed of 6 or more zoccia to arow. Another specimen from the same locality has 3-4 zovecia to a row, marginal avicularia and a few marginal spines, some zocecia having none. A small fragment also from the same place bas 8-9 marginal spines. This is in a uni-bilateral condition, but is probably torn from a larger piece. From near Wilezek Land, 127 fathoms, there is a Bugula with marginal avicularia, no bordering spines, two pairs of distal spines. The branches are not broad, and the zoccia are often but Joosely attached, and in several cases a zocecium grows out laterally from the radicular disk. In one or two cases a tubular growth springs from the distal end of the zocecium, and from the end of this elongated tube grows an ordinary zoecium. In BRYOZOA FROM FRANZ-JOSEF LAND, 53 most respects this resembles B. Murrayana, but in others it is like var. fruticosa. Sections of a specimen from off Cape Mary Harmsworth showed 21 tentacles, whereas a specimen of var. fruticosa from off West Point gave 17 tentacles; specimens from Greenland of B. Murrayana typica had 18-20 tentacles. In this and the variety there are two lateral rosette-plates, and one distal plate with numerous pores. In Bugula the rule is a row of small rosette-plates * near the distal border, and from the entirely different mode of connection it is possible that this species will be ultimately removed from Bugula. In Bugula it is not usual for the avicularia to be median. The Bugula hexacantha, Ortmann, from Japan would seem to be B. Murrayana. Loc. British seas, Scandinavia, Finland, Spitzbergen, Greenland, Barents Sea, Novaya Zemlya, Jan Mayen, Franklin Pierce Bay, lat. 79° 29' N. Japan? Jackson-Harmsworth HExp.: Cape Mary Harmsworth, 53-93 fathoms. 5. Bueuta MuRRAYANA, var. FRUTICOSA, Packard. In this variety the branches of the zoarium are usually narrower than in typical B. Murrayana, the marginal spines are absent, and the oral ones are often very slightly developed. There are no marginal avicularia, and the central ones are usually much less numerous than in typical B. Murrayana. A specimen from off West Point, Northbrook Island, has two oral spines on each side, avicularia placed centrally below the area of the zoccia, but no marginal avicularia. There are 17 tentacles. Another specimen from the same place, also from 2-3 fathoms, consists of four or more rows of zowcia. The outer zoccia have one outer spine, while the inner zocecia have two pairs of oral spines. A specimen from off Cape Gertrude, 30 fathoms, has two small spines on the outer side of the zocecia, and one on the inner. There are no avicularia in this specimen. Loc. Spitzbergen, Finland, Kara Sea, Davis Straits, Labrador, Gulf of St. Lawrence; Franz-Josef Land, lat. 79° 55’ N., long. 51° O' HE. (Ridley). * See Levinsen, Danske Dyr (Mosdyr), pl. i. f. 40. 54 MR. A. W. WATERS ON Jackson-Harmsworth Exp.: off Northbrook Island, 2-3 fath.; off Cape Gertrude, 30 fath. 6. Bueuta HarMsworruit, sp. nov. (PI. 7. fig. 12, Pl. 8. fig. 1.) The three lower zocecia retain the primary character and are elongate, tubular, slightly trumpet-shaped, expanding upwards ; aperture terminal with ten spines round the margin, two radicles arising from near to the base. The mature zoccia are biserial, have two spines at each corner, an area covering about two-thirds of the front of the zoccium, and a large avicularium placed a little more than a quarter of the length of the zocecium from the top. Ovicell unknown. This was at first taken for B. avicularia, L., but differs in having four spines, the area is shorter, the avicularia are somewhat longer, and the primary zoccia are not the same. Unfortunately there was only one piece, so that it was not advisable to cut any sections. Lorenz gives B. avicularia, L., from Jan Mayen. Loc. From off Cape Mary Harmsworth, 53-93 fath. 7. SCRUPOCELLARIA scABRA (Van Ben.). (Pl. 7. figs. 14-16.) Cellarina seabra, Van Beneden, Bull. Acad. Roy. de Belgique, vol. xv. pt. i. p. 73, figs. 8-6 (1848). Scrupocellaria Delilit, Alder, On New Brit. Polyzoa, Micr. Journ. n. s. vol, iv. p. 107 (18), pl. iv. figs. 4-8. Cellularta scabra, Smitt, “ Krit. Fort.” 1867, pp. 283 & 314, pl. xvii. figs. 27-34. Scrupocellaria scabra, Norman, Q. Journ. Micr. Se. n. s. vol. viii. p. 214 (3); Hincks, Brit. Mar. Polyz. p. 48, pl. vi. figs. 7-11; id. Ann. Mag. Nat. Hist. ser. 6, vol. iii. p. 427, pl. xxi. fig. 1. All the Arctic specimens now examined have a very large scutum, resembling in this respect the form described by Hincks from Gaspé Bay (St. Lawrence), and which he says also occurs in Greenland, and it might be an advantage to separate them as a variety from the British forms. There are usually two spines to each zocecium, but occasionally three, and three spines to the zocecium at the bifurcation, and there sometimes is one vibra- culum on the dorsal surface at the bifurcation. This is, however, often wanting; so that both Levinsen and I have referred to S. scabra as without any vibraculum at the bifurcation. In the same way, the large dorsal vibracula are frequently altogether wanting, so that whole colonies may be described as without BRYOZOA FROM FRANZ-JOSEF LAND. 55 vibracula; but when a radicle occurs, and they are not frequent, there is usually a vibracular chamber nearly at right angles to the axis of the zoarium. The seta is broad at the base and is symmetrical. There are 16 tentacles. In my paper on the Cellulariide, I have shown that S. scabra is a northern form and §. Delilii, Aud., Mediterranean and from Madeira; and I have been unable to accept some of the synonyms given by Mr. Hincks and Miss Jelly, and there have also been some mistakes in giving the range of localities. Loc. Spitzbergen, 6-150 fath.; Greenland, Finland, Novaya Zemlya, Kara Sea, Jan Mayen, Murman Sea; Franz-Josef Land, lat. 77° 55’ N., long. 51° O' HE. (Ridley); Kola, Iceland, St. Law- rence (Hincks); Norway, North Sea, British seas, Davis Straits. Jackson-Harmsworth Exp.: nr. Wilczek Land, 127 fath.; off Cape Mary Harmsworth, 53-93 fath.; off Cape Gertrude 30 fath. > 8. SCRUPOCELLARIA TERNATA, var. GRACILIS (Smitt). (Pl. 7. fig. 12.) Cellularia ternata, var. gracilis, Smitt,“ Krit, Fort.” 1867, pp. 283 & 805, pl. xvi. figs. 14-24. Cellarina ternata, var. gracilis, Verrill, Cont. Zool. Mus. Yale College, no. 43 (fide Jelly). Menipea gracilis, Levinsen, Bryoz, Kara-Havet, p. 307 (8); Busk, Journ. Linn. Soe. vol. xv. p. 282. Menipea arctica, Busk, Quart. Journ. Micr. Se. vol. iii. p. 254, pl. i. figs. 4, 5, 6; Ridley, Ann. Mag. Nat. Hist. ser. 5, vol. vii. p. 444; Lorenz, Bry. von Jan Mayen, p. 83. Menipea ternata, Hincks, Ann. Mag. Nat. Hist. ser. 5, vol. x. p. 3; Bidenkap, Bry. von Ost-Spitzbergen, p. 613. There has been considerable doubt as to the position of this form, for some authors, as Bidenkap, think, after the result of the examination of considerable material, that there should be no separation between S. ¢ernata and 8. ternata var. gracilis ; others, as Levinsen and Busk, would make two species. Lorenz gave a table showing how his S. arctica varied with respect to spines, avicularia, and scutum. The specimens from Franz-Josef Land are subject to very great variation: sometimes on the same colony the zocecia may be with or without a scutum, and this may be very minute or full-sized ; the spines may be absent, or there may be one, two, or three. The median cell may be mucronate or plain, but often 56 MR. A. W. WATERS ON the mucro only occurs on the younger zoccia. In some positions the lateral spines are readily mistaken for a mucro. There is one small specimen from off Cape Mary Harmsworth which has no lateral or suboral avicularia, no spines, no scutum, and the median cell is not mucronate. There are two moderate- sized specimens from off Giinther Sound, on each of which I have found one lateral avicularium, while the other zocecia are unprovided with them. These specimens have no scutum, and one lateral spine. On the other hand, there are specimens from Cape Mary Harmsworth in which both the scutum and lateral avicularia are fairly constant, and there is usually one lateral spine. The absence of suboral avicularia seems to be general in the Arctic forms which I have had the opportunity of examining ; aud no authors except Busk and Smitt mention anterior avicularia on S. gracilis, Busk says: “anterior avicularium small, rare, and only (?) on the median zocecium at a bifurcation.” In specimens sent me from Varanger Fjord as J. gracilis var., there are a few anterior avicularia to the zoccia at the bifur- cation; but in these specimens there is to each zocecium an extremely large scutum and very large elongate ovicells besides very long spines, and probably it should be separated specifically. In no other case have I seen anterior avicularia. The radicle is attached to a raised chamber at the base of the zocecium, and in the specimens examined there are many zocecia to an internode. Specimens from lat. 77° 55’ N., long. 53° 20’ E., and those from Varanger Fjord have 16 tentacles. Busk says 12 tentacles to his specimen from Franklin Pierce Bay. S. elongata, Sm., has 18 tentacles; S. Smittiz, Norm., 15 tentacles. There are two lateral rosette-plates to the entire side, and the distal wall has a number of small rosette-plates near the base. The mode of connection is similar to that obtaining in Serupo- cellaria, and that genus must be extended to receive this portion of the Menipea (auct.). The only difference is that there are no vibracule, but this is also the case in some typical Scrupocellaria, and in others they are rare. Loc. Spitzbergen, Greenland, Scandinavia, Kara Sea, Novaya Zemlya, Jan Mayen, Queen Charlotte Islands (H.), Labrador, Barents Sea, Franklin Pierce Bay ; lat. 79° 55’ N., long. 51° 0’ EH. (Ridley) ; lat. 79° 31' N., long. 68° 21' E., 230 met. (Marenzeller). BRYOZOA FROM FRANZ-JOSEF LAND. 57 Jackson-Harmsworth Expedition: Giinther Sound, 10 fath.; mr. Wilezek Land, 127 fath.; lat. 77° 55' N., long. 53° 20’ E., 130 fath.; off Cape Mary Harmsworth, 53-93 fathoms. 9. ScrupoceLtaRia Smirtit (Norm.). (Pl. 7. figs. 8-11.) Menipea Smittii, Norman, Q. Journ. Micr. Se. n. s, vol. viii. p. 214 (3). Menipea ternata, forma duplex, Smitt, “ Krit. Fort.” 1867, pp. 283 & 312, pl. xvi. figs. 25, 26. Menipea duplex, Vevinsen, Bry. fra Kara-Havet, Dijmphna-Togtets zool.-bot. Udbytte, 1886, p. 309, pl. xxvi. figs. 1-2. Menipea Smuttu, Wirchenpauer, Hydr. & Bryoz. p. 418, Die zweite deutsche Nordpolfahrt. There are specimens from 50 miles N.W. of Cape Mary Harmsworth which I think must be put to this species, although there are a variable number of spines. In the younger part of the zoarium there is usually but one spine, while in the older part there are sometimes as many as six. ‘The central spine in the zocecinm at a bifurcation may be wanting, though in other cases there are, besides the central one, two pairs of spines. There is no scutum. The lateral and suboral avicularia are both fairly constant. In some specimens there are generally six zocecia in an internode, in others ten 1s about the usual number. The most important character seems to be the position of the radicle chamber, which is inside the zocecium; whereas in S. gracilis, and in nearly all the other species, it is formed by a conical chamber outside the zocecial wall. These chambers are formed in all the zoccia, whether there is a radicle or not. Another very important character is that the articulation occurs beyond the distal end of the outer zocecium (fig. 8). Usually in Scrupocellaria, as in 8. scabra, S. gracitis, &c., the articulation is below the area of the outer zocecium (fig. 14). There are 15 tentacles. 1t is unfortunate that Norman did not take the varietal name duplex, as was subsequently done by Levinsen, but as Norman first gave a soecific name it seems the correct thing to retain that name. This species is closely related to WU. Jeffreysii, Norm. Loc. Spitzbergen, 50 fath. (Smitt) ; Hast Greenland (Azrch.) ; Kara Sea, 50-76 fath. (Levinsen). Jackson-Harmsworth Exp.: 50 miles N.W. Cape Mary Harmsworth, 234 fath. 58 MR. A. W. WATERS ON 10. ScRUPOCELLARIA ELONGATA (Smitt). Cellularia scabra, forma elongata, Smitt, “ Krit. Fort.” 1867, p. 284, pl. xvii. figs. 35, 36. Scrupocellaria scabra, forma elongata, Bidenkap, Bry. von Ost-Spitz- bergen, p. 614. There are two specimens from off Cape Mary Harmsworth, in one of which there is a distinct spinous process at each upper corner, and it wasat first marked “ bispinata ;” in the other the spines are only found on some zocecia. There is a small sub- oral avicularium to some zocecia and there are small lateral avicularia, no vibracula; in the specimen in which the spines are most developed there are a few scuta as figured by Smitt; the ovicell is elongate with a line down the middle and an area on each side, the radicle starts from near the base of the zocecium, the radicle-chamber being external; zocecium at the bifurcation acuminate. The jointing is low down as in Serupocellaria ternata var. gracilis (Pl. 7. fig. 12), and in this respect differs entirely from S. Smittii, Norm. (tigs. 8, 9), and is also much stouter than S. scabra, van Ben. There are 18 tentacles. Loc. Spitzbergen (Sm. & Biden.) ; Greenland, Finland (Si.). Jackson-Harmsworth Expedition: off Cape Mary Harmsworth, 53-93 fathoms. 7 11. Scruprocetrarta PEacuit (Bush). Cellularia Peachit, Busk, Ann. Mag. Nat. Hist. ser. 2, vol. vii. p. 82, pl. viii. figs. 1-4. Ina paper dealing with Cellulariide* I have stated my opinion that the genus Cellularia must be dropped, and that some species so placed belong to Serupocellaria. The differences between this and Serupocellaria ternata, var. gracilis, Sm., seem but slight, but as yet I have not had the opportunity of making sections of this species. The radicle arises from a chamber at the base of the zocecium. Loc. British, Spitzbergen, Novaya Zemlya, Kara Sea, Murman Sea, Barents Sea, Labrador. Jackson-Harmsworth Expedition: small specimen off Cape Mary Harmsworth, 53-93 fath. Besides the species of Cellulariide already mentioned, Smitt gives Caberea Kilisit from Greenland, and Levinsen mentions Kinetoskias arborescens, Kor. & Dan., from Kara Sea [and Nordgaard gives Bicellaria Alderi, B., from Spitzbergen; also Menipea Normani, Nordg., from off the coast of Norway]. * Journ. Linn. Soc., Zool, yol. xxvi. p. 1. BRYOZOA FROM FRANZ-JOSEF LAND. 59 12. Frusrra carsasea, Ell. & Sol. Smitt considered that this occurs both unilaminate and bilaminate, but the latter form has been named Flustra spitz- bergensis by Bidenkap*. The specimens brought back by the Jackson-Harmsworth Expedition are all unilaminate, and have 23-24 tentacles. Loc. Spitzbergen, Norway, Greenland ; Jan Mayen, Jugor-and Matotschkin-Schaar; Gulf of St. Lawrence; British seas: Iceland; Rattegat v. Anholl. Jackson-Harmsworth Expedition: Giinther Sound, 10 fath. ; off Kast Glacier on floe ; Cape Flora, nr. Wilczek Land, 127 fath. 13. FLUSTRA MEMBRANACEO-TRUNCATA, Siitt. This throws out many radicle processes, and undoubtedly what Nitsche and Hincks called “Thurmzocecia”’ are only these processes, and the appearance is often quite similar to the figures of these structures, though of course they only occur on the dorsal surface, whereas they have been figured as on the anterior surface. There are 16 tentacles, and 3 distal rosette-plates and 6 lateral. Loe. Spitzbergen, Greenland, Kara Sea, Kola, Jugor- and Matotschkin-Schaar, Norway. [Jan Mayen (Wordgaard).| _ Jackson-Harmsworth Expedition: Cape Mary Harmsworth, 53-93 fathoms. Besides the above two species of Flustra, the F’, securifrons, Pall., has been found in the Kara Sea; the F. serrulata, Busk, in the Kara Sea, from West Greenland, and from Franklin Pierce Bay ; F. abyssicola t, Sars, has been found off Novaya Zemlya. 14. MemBranipora mMonostacuys, Busk. (PI. 8. fig. 3.) Membranipora monostachys, Busk, Brit. Mus. Cat. pt. 2, p. 61, pl. Ixx. figs. 1-4. Hippothoa rugosa, Stimpson, Invert. of Grand Manan, 1863, Smith- sonian Cont. p. 18, fig. 9. A specimen from near Wilczek Land has the blind cells which were mistaken by Busk for avicularia, and they seem to be formed when growth is arrested by any cause, such as meeting a neighbouring branch. ‘There are no spines, though there are indications of a central spine in some zocecia ; and it is difficult to * Bryozoen von Ost-Spitzbergen, p. 617, pl. xxv. figs. 1, 2. + I regret that when describing F. separata, Waters, from off Nova Scotia in the Suppl. Rep. of the ‘ Challenger’ Polyzoa, Chall. Zool. vol. xxxi., I.did not recognize that it was Ff. abyssicola, Sars. 60 MR. A. W. WATERS ON know whether it should be placed with JL. monostachys, B., or MM. catenularia, Jam. The lower part of the zocecia is not con- tracted to the extent usual in I. catenularia; the margins, however, cau scarcely be called thin and slope inwards. The surface of the zocecium is nearly smooth with but slight tendency to form ridges. The opercular flap is large and very noticeable in balsam specimens, as it is less transparent than the membrane covering the rest of the opesium. There are 10 tentacles, while there are 12 in IZ. pilosa. Loc. Jackson-Harmsworth Expedition: Ginther Sound, 10 fath.; off Elmwood, 18 fath.; nr. Wilczek Land, 127 fath. 15. MemBrantpora LInEATA (L.). Membranipora lineata, Hincks, Brit. Mar. Polyzoa, p. 143, pl. xix. figs. 3-6; Waters, Linn. Soc. Journ. vol. xxvi. p. 678, pl. xlvii. fig. 11, pl. xlviii. fig. 9. For synonyms compare Miss Jelly’s Catalogue. Although this is widely distributed im the Arctic seas, I have only seen small pieces, which could not be used for study. There are 13 tentacles. Loc. Britain, Florida, Azores, Scandinavia, Spitzbergen, Fin- land, Greenland, Iceland, Davis Straits, Labrador, Kara Sea, Novaya Zemlya, Barents Sea, Jan Mayen. Jackson-Harmsworth Expedition: off Elmwood, 18 fath.; mr. Wilezek Land, 127 fath.; off glacier between Cape Gertrude and Cape Flora, about 127 fath.; off Cape Mary Harmsworth, 53-98 fath.; lat. 77° 55' N., lone. 55° 25’ H., 115 fath. 16. MEMBRANIPORA ARCTICA (d’O7b.). Reptoflustrina arctica, d’Orb. Pal. Fr. vol. v. p. 582. Membranipora Sophie, Busk, Q. J. Micr. Sc. vol. ii. p. 255, pl. i. fig. 7 (1855) ; Ridley, Ann. Mag. Nat. Hist. ser. 5, vol. vil. p. 446, pl. xxi. fig. 2. Membranipora lineata, f. Sophie, Smitt, “ Krit. Fort.” 1867, pp. 865 & 394, pl. xx. figs. 24, 25. Membranipora arctica, Lorenz, Bry. von Jan Mayen, p. 85, pl. vii. fig. 1; Nordgaard, Bergens Museums Aarbog, 1894-5, p. 17; Bidenkap, Bry. Ost-Spitzbergen, p. 620. In my paper on Membraniporide in this Journal, vol. xxvi. p- 680, I suggested that it would be better to retain the name M. Sophie and keep arctica for the Semiflustrellaria arctica of d’Orbigny ; but as Bidenkap had, about the time my paper was written, made a species IZ. spitzbergensis for the Semiflustrellaria arctica, we must follow him. There are three Membranipore which received the name BRYOZOA FROM FRANZ-JOSEF LAND. 61 arctica from d’Orbigny: first the above Reptoflustrina arctica, then in manuscript the Semiflustrellaria arctica (= M. spitzbergensis, Bidenkap). A specimen from Newfoundland (13705) so named in the Musée d’Hist. Naturelle, Paris, has a small, round, widely open avicularium like that in MW. flustroides, H. There is also the Reptoflustrella arctica, d’Orb., which has a few stout spines, a bar to the ovicell, and a triangular avicularium above the ovicell. This may be IZ. wnicornis, Flem. In a specimen from the Gulf of St. Lawrence sent to me by Canon Norman as JZ. Sophie, many of the zocecia, especially the younger, are entirely without any trace of spines, and the same condition is found in some zoccia in the Franz-Josef Land specimen. Hincks has described from the Houston Stewart Channel (Queen Charlotte Islands) JZ. Sophie form matura*. Loc. Spitzbergen, 10-30 fath.; Greenland, Jan Mayen, Kola, Norway, Assistance Bay (Busk); Gulf of St. Lawrence (in my coll.) ; Franz-Josef Land, lat. 79° 55’ N., long. 51°0' E. (Ridley) ; Norway. Jackson-Harmsworth Expedition : off Elmwood # mile, 18 fath.; nr. Wilczek Land, 127 fathoms. 17. MEMBRANIPORA SPINIFERA, Johnst. Some small specimens from Giinther Sound are placed under M. spinifera, though in some respects they seem to be inter- mediate between M. spinifera, Johnst., and I. cymbeformis, Hincks. The spines, about 8-10, are more delicate than in my specimen of J. cymbeformis from the Gulf of St. Lawrence. The lamina to which Mr. Hincks referred does not exist in the specimen of cymbeformis in my collection, from Gulf of St, Lawrence, as can be easily seen in a piece boiled in caustic potash nor is it found in the present form. M. spinifera, Joknst., is mentioned from the Barents Sea by Vigelius, and from the description this agrees with the specimens from Giinther Sound, 10 fathoms. J. cymbeformis has been recorded from Jan Mayen (Lorenz); Kara Sea (Levinsen); Davis Straits, Labrador, and Gulf of St. Lawrence (Hincks) ; Spitzbergen (Bidenkap). 18. MreMBRANIPORA MACILENTA, Jullien. (PI. 8. fig. 10.) Membranipora macilenta, Jullien, Bull. Soc. Géol. France, yol. vii. p. 25, pl. xvii. fig. 62, 1882. Specimens growing on Mytilus-shell fromnr. Wilczek Land, * Ann, & Mag. Nat. Hist. ser. 5, vol. x. p. 9, pl. xx. fig. 2. 62 MR. A. W. WATERS ON 127 fathoms, have large zocecia with very thin walls and an avicularium with triangular mandible at each upper corner. There is no ovicell. As far as can be seen, it agrees with Jullien’s species from the North of Spain, and is much like Membranipora membranacea with the spinous process developed into avicularia. There are 18 tentacles. Besides the Membranipore referred to in this paper, I. craticula, Alder, is said to have been found off Jan Mayen (Lorenz), and Franz-Josef Land (Ridley), Kara Sea, Novaya Zemlya, West Greenland, and Davis Straits. J. Plemingit, Busk, from Jan Mayen and West Greenland. M. armifera from the Gulf of St. Lawrence. JZ. trifolium, 8S. Wood, from Jan Mayen. [MM cornigera, Busk, from Bear Island (Nordgaard). | 19. Cripritina PuNcTATA (Hassall). (Pl. 8. fig. 22.) Escharipora punctata, Smitt, “Krit. Fort.” 1867, p. 4, pl. xxiv. figs. 4-7. . The species brought back by the Jackson-Harmsworth Expedition is no doubt the species figured by Smitt, but it seems exceedingly doubtful if this is the same as the British species, and therefore I only refer to Smitt; and not having had the opportunity of examining many specimens of C. punctata I hesi- tate to give a new specific or varietal name, especially as the Franz-Josef Land specimens are very fragmentary and not sufficient for thorough examination. The difference consists in the very wide ovicell, which is often almost concealed; and although Hincks speaks of the ovicell being subject to considerable variation, the difference between this very wide ovicell and the rather narrow ovicell of the ordinary British form is so very great, that until the lmks have been carefully studied we may doubt the identity. It is mentioned as occurring off Jan Mayen, but as Lorenz does not give any particulars we cannot judge which form he had before him. This species is of considerable interest from the fact that it occurs of the same size and with the same characterson a specimen of Hornera lichenoides, Pout., said to be from lat. 74° §. and long. 172° E., 380 fath., and to have been dredged by Sir John Ross in his Antarctic expedition. Mr. Busk mentions and had named the Hornera and also a piece of Escharoides Sarsii, Smitt, from the same dredge; but the Cribrilina punctata, Tass., had not been determined or mentioned when Mr. Kirkpatrick kindly BRYOZOA FROM FRANZ-JOSEF LAND. 63 allowed me to examine the specimens in the British Museum collections. Mr. Busk, however, in the Journal of this Society (vol. xv. p- 237), when describing Arctie Escharoides Sarsii, Sm., says it was collected in the Antarctic by the ‘ Erebus’ and ‘Terror,’ “ and was accompanied in the same collection by two other Arctic species.” That a single haul so far south—in fact the farthest south from which any Bryozoa have been found—should furnish three of the commonest Arctic species, two of which have prob- ably not been found outside the Arctic regions, seems so strange that we must pause and consider if there is no possibility of a mistake. This would be stronger proof of bipolarity of species than has yet been brought forward, and as much as the most ardent believers in the theory could expect. Unless the present Antarctic expeditions should bring back some of these species, we shall be justified in thinking that there has been some change of label or exchange of box. As Sir John Ross had been in the Arctic regions, there is the possibility of this exchange having occurred on board the ship, or it may have happened subsequently in Mr. Busk’s hands. There are several specimens of mites in the Escharoides, and although the probability was so great that their origin was Lordon, I asked my friend Mr. A. D. Michael if they could throw any light upon whether the Bryozoa came from the Arctic or Antarctic. He informed me that the specimens were immature Glyciphagus domesticus, de Geer, which is cosmopolitan, and therefore it does not help to settle the pomt. Mr. Michael informs me that this species had been found by the Jackson- Harmsworth expedition on the rocks some distance from the station, aud that a drawing had been made of it at the time. Often specimens of Bryozoa are so full of diatoms that a question of doubtful origin could be settled by an examination of sections, but unfortunately there do not seem to be either foraminifera or diatoms to give us the geographical origin. The occurrence of this Cribrilina identical with the Arctic form is of more importance than that of either H. lichenoides or E. Sarsii, being more highly differentiated, or at least having more distinctive characters. - Loc. Bohus Bay (Smiét) ; littoral, Norway (Smztt); Gullmaren, 7-10 fathoms. Jackson-Harmsworth Exp.: Lat. 77°55 N., long. 58°16’ E., 130 fathoms. 64 . MR. A. W. WATERS ON 20. CRIBRILINA ANNULATA (fab.). (PI. 8. fig. 21.) Round the distal border there are, in the specimen from Franz-Josef Land, calcareous processes in place of the spines. Loc. Spitzbergen, Jan Mayen, Kara Sea (Levinsen), Finland, New Brunswick, Gulf of St. Lawrence, Labrador, Scandinavia, British Isles, Denmark, Jackson-Harmsworth Exp.: off glacier between Cape Flora and Cape Gertrude, about 30 fathoms. I have already shown the geographical distribution of Flustra* and Membraniporat, and we now seem to be sufficiently well acquainted with Cribrilina to give a similar list for this genus. Ase Get ee. sees C. annutata ft, Fab., punctata, Gray, nitido- punctata, Sm., scutulata, Busk (? Cribrilina). British pees es eee ANNULATA, PUNCTATA, RADIATA, Aud., FIGU- LARIS, Johnst., Gatrym, Busk. Mediterranean. RADIATA, FIGULARIS, Gartym, Banzact, Aud., SETOSA, Waters. North Atlantic. PUNCTATA, RADIATA, BALZACI, SETOSA. South Atlantic. RapiaTa, MONOCEROS, Busk, Jlatimarginata, Busk. South African.. Jdabiosa, Busk. Australasian ... RADIATA, MONOCEROS, PHILOMELA, B., acantho- ceros, MacG., setirostris, MacG., speciosa, Hincks, tubulifera, H., clithridata, Waters. South Indian Raprava, philomela var. adnata, Busk. Ocean. North Pacific... moNnocEROS, hippocrepis, Hincks, furcata, H. South Pacific... MONOCEROS. SAAN fey eens bece PHILOMELA, Busk, reniformis, Ortmann. 21. ScHIZOPORELLA Crustacea (Smitt). -(P1. 8. figs. 11-13.) Myriozoum crustaceum, Smitt, “Krit. Fort.” 1867, p. 18, pl. xxv. figs. 88-91; Ridley, Ann. Mag. Nat. Hist. ser. 5, vol. vii. p. 448; Biden- kap, Bry. Ost-Spitzbergen, Zool. Jahrb. vol. x. p. 622. Leieschara crustacea, Levinsen, “ Bry. f. Kara-Havet,” Dijmphna Togtets zool.-bot. Udbytte, 1886, p. 317 (18). Schizoporella crustacea, Lorenz, Bry. von Jan Mayen, p. 87, pl. vii. fig, 2. * Journ. Roy. Mier. Soc. 1896, p. 286. + Journ. Linn. Soc., Zool. vol. xxvi. p. 664. + Species printed in small capitals occur from a second region. BRYOZOA FROM FRANZ-JOSEF LAND. 65 Lepralia plana, Dawson, Polyzoa of the Gulf of St. Lawrence, Rep. Canadian Geol. Sury. 1858, p. 256. Myriozoum planum, Hincks, Ann. Mag. Nat. Hist. ser. 6, vol. ix. p- 157. Myriozoum is the genus of Donati of which the type is I. truncatum, and it, together with WZ. coarctum and subgracile and the genus Haswellia, has very long pore-tubes, and the centre of the zoarium of the erect species has a spongy structure. Schizo- porella crustacea has longer pores than most of the Bryozoa, and they open diagonally, causing the open ornamentation; but the value of a character like this is difficult to estimate, as it is only one of degree, short pores occurring in a great part of the Chilostomata. The shape of the aperture corresponds with that of Myriozoum coarctum and M. subgracile. I do not consider this species the same as S. inezsa, M.-Ed. ; and therefore probably the fossil determined by Neviani as S. crustacea should be placed elsewhere. The description of L. plana, Dawson, was too insufficient to justify the adoption of that name. There are 15 tentacles. Loc. Greenland, 14-35 fath. (Busk); Spitzbergen, 10-80 fath. (Bidenkap); Kara Sea, 30-85 fath.; Kola; Jan Mayen, 15-180 metres (Lorenz); Gulf of St. Lawrence; lat. 79° 55’ N., long. 51° 0' E. (Ridley); Finland. Jackson-Harmsworth Exp.: off Elmwood, 18 fath.; Gunther Sound, 10 fath. 22. ScHIZOPORELLA HarMsworTHiI, nom. nov. (Pl. 9.. figs. 10-12.) Eschara Legentilii, Aud., forma prototypa, Smitt, “ Krit. Fort.” 1867, pp. 10 & 81, pl. xxiv. figs. 47, 48. Schizoporella cincta, Hincks, var., Hincks, Ann, Mag. Nat. Hist. ser. 6, vol. ix. p. 154, pl. viii. fig. 2. The imperforate area immediately below the oral aperture is often depressed, giving a characteristic appearance to the zocecium; in other cases it is raised and carries an avicularium. The pits on the surface are very deep and are perforated at the bottom. The suboral glands of this species are very interesting, as they arise from the upper part of the tentacular sheath, and the gland is attached to the sheath for some little distance. The glandular cells are distinctly nucleated, and the interior of the gland contains a nearly homogeneous mass which does not stain. From LINN. JOURN.—ZOOLOGY, VOL. XXVIII. 5 66 MR. A. W. WATERS ON this I think we may conclude that the suboral and avicularian glands are homologous. In this species, as well as in those alluded to in the subsequent part of this paper, it appears that the gland is emptied close by the wall of the tentacular sheath when the polypide is extruded; and we have seen that the avicularian gland of Lepralia foliacea*, Ell. & Sol., is attached in a somewhat similar manner to the sheath of the avicularium. The contents of the avicularian glands of Porella acutirostris, Smitt, are also a homogeneous mass; and we can scarcely doubt any longer that these glands are excretory organs. Karchenpauer has given the name Lepralia Smittii to Escharella Legentilu, forma prototypa of Smitt, but as Smitt figures distinct forms under his prototypa the name cannot be retained. Possibly Smitt had Lepralia borealis, Waters, before him as well as the present form. Loc. St. Lawrence (H.); Spitzbergen (Sm.); Greenland (Sm.). Jackson-Harmsworth Exp.: Lat. 77°55"N., long. 53° 16' E., 180 fath.; off glacier between Cape Flora and Cape Gertrude, about 30 fath. 23. SCHIZOPORELLA ELMWOODTA, sp. nov. (Pl. 9. figs. 1 & 18.) A dry specimen encrusting a piece of basalt from off Elmwood, 18 fathoms, seems to be new. : The zocecia are broadly ovate, considerably raised, and distinctly separated, with the surface finely granular, and with pores scattered over the surface. The aperture is close to the distal edge, and has a distinct sinus. The operculum is granular, and instead of having two muscular dots, as in most Schizoporelle, has a muscular ridge some distance from the border, and must perhaps ultimately be put in a separate division of the Schizo- porelle on that account. The ovicell is much raised, very broad and granular; and is not closed by the operculum. In a few zocecia there is a triangular avicularium on one side lower than the aperture. There are sometimes irregular elevations on the two sides of the zoccia. Besides the Schizoporelle brought back by this Expedition, S. candida, Stimpson; S. linearis, Hass.; 8. biaperta, Mich. ; S. sinuosa, B.; 8S. limbata, Lorenz; S. stylifera, Levinsen; S. auriculata, Hass., are said to have been found in the Arctic regions. * Waters, “Observations on Gland-like Bodies in the Bryozoa,” Linn. Soe. Journ., Zool. vol. xxiv. p. 272. BRYOZOA FROM FRANZ-JOSEF LAND. 67 Myriozoum. Whether this genus should be retained is very questionable, but for the present it is more convenient to keep the name for M. coarctwm, Sars, and IL. subgracile. d@ Orb., remembering that in descriptive papers we may often have to use generic divisions which we are aware may ultimately be altered. The genus was created by Donati with I. truncatum, Pall., as the type, and it has been based to a large extent upon the zoarial shape of growth; and the definitions given by Donati, Sars (Leieschara), d’Orbigny, and Smitt would not justify its retention. The aperture of JZ. truncatum is nearly round, while those of IZ. coarctum and subgracile are schizoporellidan, with a well-marked sinus. There is, however, one character which may be of great elassificatory value, and that is the long tubes, which I have called pore-tubes, from the surface to the interior, and from one zoecium to another, causing the spongy structure of the zoarium. These tubes havea disk near the junction with the zocecium ; and in the middle of the axial tubes, which are often very long, there are similar disks (Pl. 9. fig. 6d; fig.7d). The shell-structure is terminated by a membrane (fig. 86) perforated by these pores ; but over this, as in the greater part of the Chilostumatous Bryozoa, there is another fairly thick and somewhat chitinous membrane (fig. 7 a) continuously covering the zoarium. The eell-contents of the tubes attach themselves to this by delicate threads (fig. 8); and thus by this connection there may be an interchange between the sea-water and the contents of the tubes, which again are in indirect communication with the contents of the zoecia. Although in other genera these tubes are not developed to the same length, the structure is very similar. In figures 6,7 (Pl. 9), drawn to explain these tubes, the way in which the diaphragm is folded when the polypide is retracted is also shown. This folding, which at first I called an opercular fold, occurs in a similar position in most of the Chilostomata, but the appearance of this diaphragm varies much in different species* and at different times. It is attached to the tentacular sheath, _ and is, as described by Nitsche *, a strong sphincter which can completely close the tentacular sheath-cavity. The central * Nitsche, “ Ueb. Anat. u. Entw. von Hustra membranacea,” Zeits. f. wiss. Zool. vol. xxi. p. 17. Sys 68 MR. A. W. WATERS ON opening seems to be a chitinous ring in Lepralia sincera, Sm., from which there are rays of muscular tissue, and there are concentric rings, also no doubt contractile. Pergens * speaks of the tentacular sheath having four close- able openings: (1) the opercular opening, (2) Nitsche’s diaphragm, (3) oral opening, (4) anal opening; but this seems to me more complicated than is the case, and though the diaphragm may in some cases be a little distance from the external opening, yet it seems enough to speak of the opening through the diaphragm and the anal opening ; and I cannot follow Pergens in considering that any opening has been found from the tentacle-sheath into the body-cavity. Myriozoum ovum, Smitt, will be Stichoporina or Fedora; and M. marionensis, Busk, is the only species determined as Myriozoum from the Southern hemisphere. 24. Myriozoum coarctum (Sars). (Pl. 9. figs. 2, 3.) Cellepora coarcta, M. Sars, Nyt Mag. f. Natury. vol. vi. p. 148 (28), 1850? Leteschara coarcta, M. Sars, Beskr. N. Poly. 1862, p. 155 (17). For synonyms of Smitt, Busk, Hincks, Lorenz, see Miss Jelly’s Catalogue. Myriozoum coarctum, Bidenkap, Bryozoen von Ost-Spitzbergen, Zool. Jahrb. vol, x. p. 621; Hennig, Cifvers. af K. Vetensk.-Ak. Forh. 1896, p- 397. This species does not appear to have been figured, saving the section given by Smitt. The semicircular avicularia are placed in a row between the zoccia and vary considerably in size, so that the avicularian aperture may be as large as the oral aperture. There are 16 tentacles, as in MZ. subgracile, d’Orb., while in M. truncatum, Pall., there are about 26 or 27 tentacles. Pergens speaks of 33, but I have not seen so many, and sections he sent me correspond with those I made from material brought back from Naples. One of the pieces is about two inches long, and has clearly been broken off a much larger one. The branches anastomose at an angle of about 60 degrees and more, and some branches grow out nearly at right angles from the main branches. Loc. Spitzbergen, 19-80 fath. (Smztt), 40-95 fath. (Bidenkap) ;. Kola, Norway; Jan Mayen, 140-400 metres (Lorenz); Davis * “Untersuch. an Seebryozoen,” Zool. Anzeiger, 1889, p. 4. BRYOZOA FROM FRANZ-JOSEF LAND. 69 Straits, 100 fath. (Hineks) ; Murchison Sound, 45 fath. (Hennig) ; Finland ? (Sars). Jackson-Harmsworth Expedition: about 40 and 50 miles off Cape Mary Harmsworth, 234 fathoms. 25. Myriozoum susBGRaciLE, d’Orb. (Pl. 9. figs. 4-8.) Myriozoum subgracile, V’Orb. Pal. Frang. vol. v. p. 622; Smitt, “ Krit. Fort.” 1867, pp. 18 & 119; Hincks, Ann. Mag. Nat. Hist. ser. 4, vol. xix. p- 106; D’Urban, Ann. Mag. Nat. Hist. ser. 5, vol. vi. p. 274; Ridley, Ann. Mag. Nat. Hist. ser. 5, vol. vii. p. 448; Waters, Journ. Roy. Micr. Soe. vol. ii. p. 390, pl. xv. fig. 5; Bidenkap, Bry. von Ost-Spitzbergen Zool. Jahrb. vol. x. p. 622. Myriozoum pulchrum, Ortmann, Die Japanische Bryozoen-fauna, Arch. f. Naturgesch. 1889, p. 53, pl. iv. fig. 8. In the Franz-Josef Land specimens the subcircular avicularium is usually slightly above the oral aperture to one side, and placed diagonally. In the Japanese specimen in my collection the avicularium, though slightly lower, is similarly placed, but it is only found to a few zocecia. There are 16 tentacles, as in J. coarctum, Sars; and the two species are very similar, though it seems that they should be separated on account of the difference in size and position of the avicularium. We cannot be absolutely. sure which of the two species Sars and other writers previous to Smitt had before them. Some of d’Orbigny’s specimens from Newfoundland in the Musée d’Hist. Nat. belong to this species, but I had not time, when in Paris, to examine whether any of the specimens might be WZ. coarctum. Loe. Spitzbergen, 19-80 fath. (Smitt), 55-65 fath. (Bidenkap) ; Kara Sea, Greenland, Barents Sea, Murman Sea (D’Urban); Davis Straits, 100 fath. (Hineks); Franklin Pierce Bay (Busk) ; Newfoundland (d’Orb.); lat. 75° 36’ N., long. 57° 6' E. (Maren- zeller); Japan. Jackson-Harmsworth Exp.: off Elmwood, 5.W. 1} mile, 26 fath. ; off glacier between Cape Flora and Cape Gertrude, about 30 fath. ; Cape Gertrude, 30 fath.; nr. Wilezek Land, 127 fath. ; off Cape Mary Harmsworth, 53-93 fath. 26. HipporHoa Expansa, Dawson. (Pl. 8. fig. 19.) Hippothoa expansa, Hincks, Brit. Mar. Polyzoa, p. 291, pl. i. fig. 1, which see for synonyms. The ovicelligerous zocecia are usually shorter than the others, 70 t MR. A. W. WATERS ON and the aperture has the lower edge nearly straight, so that the operculum is similar to that of the ovicelligerous zocecia of Hippothoa hyalina, L. All the specimens from Elmwood were dry, and there was no material for cutting sections. The H. di- varicata has the aperture of the ovicelligerous and ordinary zocecia similar, both having a broad sinus. Loc. Gulf of St. Lawrence, Labrador, Maine, Queen Char- lotte Islands ; Greenland, 57 fath. ; Shetland, 100 fath. Jackson-Harmsworth Exp.: off Elmwood, 18 fath. 27. Hippotuoa HyatiIna, LZ. (PI. 8. figs. 16-18.) Mollia hyalina, forma hyalina, Smitt, “ Krit. Fort.” 1867, p. 16, pl. xxv- fies. 84,85. Compare list of synonyms in Hincks’s Brit. Mar. Poly. p. 271, and Miss Jelly’s Catalogue. Norman * considers this should be placed in the genus Celle- porella of Gray, and states that it was the only species in the genus as described by Gray, and that therefore the genus Dia- zeuxta of Jullien is merely a synonym. It would be unfortunate if the laws of priority obliged us to put the latter under Celle- porella, as it has been used in another sense by both Hincks and Norman. However, the reasons for separating Diazeuawia from Schizo- porella are based upon the reproductive characters mentioned by Jullien. The female zoccia carrying the ovicell are usually very short, and according to Jullien possess no polypide. I have, however, some specimens in which these zocecia are about the same size as the others. There are also very minute zoecial cells with very small opercula (fig. 16), which Jullien says are male: zoccia without polypides; and these may occur on the front of an ordinary zoccium, as in specimens from California, or there may be one at each side of the aperture, or they may occur scattered between other zocecia. Hippothoa has usually short ovicelligerous zocecia ; and if we are to attach most weight to the reproductive organs, then there seems little doubt that the right place for this species is Hippothoa. The shape of the aperture of the ordinary zocecia is similar throughout the genus. Jullien considers that larval peculiarities and other characters justify this being placed in a family Dza- zeuxide, and probably family distinction may have to be made.. The small specimens available have not been sufficient for cutting satisfactory sections. * Ann. & Mag. Nat. Hist. ser. 6, vol. xiii. p. 129. BRYOZOA FROM FRANZ-JOSEF LAND. 71 Barrois shows that the larva, although of the “ Escharina-”’ type, differs from others, and is in fact simpler, resembling rather closely that of his Mollia grantfera(Microporella impressa, Aud.). It has four pairs of oculiform points, and apparently from his figures two single ones; that is to say, these pigment-cells are more numerous than is usually the case. There is no ciliated crown. In the Franz-Josef Land specimens there are 14 tentacles ; but Jullien and Hincks mention 12 tentacles. Loc. Cosmopolitan. In the Arctic regions—Spitzbergen, Greenland, Jan Mayen, Kara Sea, Barents Sea, Scandinavia, It also occurs in South America, Australia, and the Kerguelen region. Jackson-Harmsworth Exp.: Giinther Sound, 10 fath.; off Cape Mary Harmsworth, 53-93 fath., 7/8/97, with ova and ovicells. 28. PSEUDOFLUSTRA PALMATA (Sars). (Pl. 8. figs. 7-9.) Eschara palmata, Sars, Forh. i Vidensk. Selsk. 1863, p. 146 (8). Escharella palmata, Smitt, “ Krit. Fort.” 1867, pp. 10 & 77, pl. xxiv. figs. 42-46; Whiteaves, Rep. on Gulf of St. Lawrence, 1874, p. 12: Levinsen, Bry. fra Kara-Havet, Dijmphna-Togtets zool.-bot. Udb. p. 318 (14), pl. xxvii. fig. 3; Lorenz, Bry. von Jan Mayen, p. 91; Marenzeller, Denks. K. Akad. Wissensch. Wien. vol. xxxv. p. 388. Flustra solida, Stimpson, Mar. Inv. Grand Manan, 1853, Smithsonian Cont. p. 19, fig. 12; Hincks, Ann. & Mag. Nat. Hist. ser. vol. vi. p. 282, pl. xv. figs. 2, 3, and ser. 6, vol. ix. p. 149, pl. viii. fig. 1. Eschara solida Vergelius, Niederl. Arch. f. Zool., Suppl. B. p. 15, figs. 2, 3. Pseudoflustra solida, Bidenkap, Bry. yon Ost-Spitzbergen, Zool. Jahrb. vol. x. p. 618. It is difficult to know where this should be placed. Hschara has been discarded, Hscharella of Smitt is not the same as Escharella of d’Orbigny, and. Smitt’s Hscharella corresponds now for the greater part with Smittia (Hincks). The name Pseudo- flustra is unfortunate, the foliaceous growth having suggested Flustra; but the entirely different aperture, the characters of avicularium and ovicell, besides the rosette-plates near the basal wall, all suggest the removal from Flustra,and the relitionship is in the direction of Lepralia. The genus Oyclicopora, Hincks, has a nearly round aperture, though the operculum has the lower edge more or less straight, in some respects resembling the operculum 72 MR. A. W. WATERS ON of Membranipora; but the genus is described as without avicu- laria, though we have often seen that it is dangerous to make the absence of avicularia a generic character: in fact negative cha- racters are not often to be relied upon. At present there does not seem sufficient reason for placing the palmata of Sars under Cyclicopora, and therefore the name given by Bidenkap is retained. Stimpson may have had this species before him ; but as the description appears quite insufficient for recognition, and might almost as well be applied to some other species, it seems only right to retain the name given by Sars, as he clearly described it. The form which Hincks described (loc. cit. pl. viii. fig. 1) with a distinct sinus in the aperture, and with semicircular avicularia, can hardly remain with this species. There are two large rosette-plates near the basal wall and several (6-8) on the lateral wall. I have not found any oral or avicularian glands. There are 18 tentacles. A specimen from lat. 77° 55’ N., long. 53° 20' E., has the peristome raised at the side, but this does not occur in any of the other specimens. The zoaria, according to Vigelius, may be uni- or bilaminate, but I have only seen them bilaminate. Loe. Spitzbergen (Smitt g Biden.), Greenland, Kola (Sm.), Barents Sea (Hincks), Matotschkin Schaar (Sm. & Stuaxb.), Kara Sea, Jan Mayen (Lorenz); lat. 79° 13' N., long. 68° 21’ E., 230 met. (Marenzeller); Norway (Nordgaard); St. Lawrence (Hineks) ; New Brunswick (Stimpson). Jackson-Harmsworth Exp.: off glacier between Cape Flora and Cape Gertrude, 30 fath.; off Cape Gertrude, 30 fath.; lat. 77° 55’ N., long. 53° 20’ E., and long. 53° 16! E., 130 fath. 29. LEPRALIA SINCERA (Sm7tt). (PI. 8. fig. 2.) Discopora sincera, Smitt, “ Krit. Fort.” 1867, p. 28, pl. xxvii. figs. 178- 130. Lepralia sincera, Hincks, Ann. Mag. Nat. Hist. ser. 4, vol. xix. p. 102, pl. xi. fig. 2; Lorenz, Bry. von Jan Mayen, p. 88; Hennig, (fvers. Vetensk.-Ak. Foérh. 1896, p. 357. Discopora reticularis, Van Beneden, Recherches sur les Bry. de la Mer du Nord, Bull. Acad. Roy. Belg. vol. xvi. p. 652, pl. ii. figs. 15-18. Hemeschara sincera, Busk, Journ. Linn. Soc., Zool. vol. xv. p. 237. Mucronella sincera, Nordgaard, Bergens Mus. Aarbog, 1894-5, p. 29, pl. i. fig. 6; Bidenkap, Bry. von Ost-Spitzbergen, Zool. Jahrb. vol. x. p. 625. BRYOZOA FROM FRANZ-JOSEF LAND. 73 There are 22 tentacles and two distal rosette-plates. I am unable to find any oral glands, and do not see any ovaria or testes in the sections prepared. The aperture and operculum vary somewhat from typical Lepralia; but there does not seem sufficient reason for removing it from Lepralia at present. Loc. Spitzbergen, 19-60 fathoms (Smztt); Greenland (Hennig); Kara Sea; Jan Mayen, 160-270 metres (Lorenz); Finland, Davis Straits (Hincks); Baffin’s Bay, 175 fath., Franklin- Pierce Bay, Smith’s Sound, 13 fath. (Busk); Inglefield Gulf, 25 fath., and Northumberland Island, 20 fath.(N.W. Greenland) (Hennig). Jackson-Harmsworth Exp.: off Cape Mary Harmsworth, 53- 93 fathoms. 30. LEPRALIA BOREALIS, sp. nov. (PI. 8. figs. 4-6.) Zoecia encrusting, rhombic, very distinctly separated by divisional lines, row of pores round the border, surface granular ; median suboral avicularium with small round mandible, large raised ovicell with numerous pores; fairly thick operculum nearly straight on the lower edge ; oral aperture subeircular, with a small lateral denticle on each side, four lateral rosette-plates near the basal wall. There are 18 tentacles and two oral glands, but there do not appear to be any avicularian glands. This in many respects resembles some of Smitt’s figures of Escharella Legentilit var. prototypa, but no doubt Smitt placed several distinct species under Legentilii. This differs from Smittia reticulata, MacG., in the shape of the operculum, and in having no central denticle (lyrula). The mandible has a large median lucida, and two large clear spots, and is somewhat of the Porella-type. The avicularian chamber extends to each border. Loc. Lat. 77° 55’ N., long. 53° 16’ E., 130 fathoms. 31. Lepratia crumnta, Norm. Lepralia cruenta, Norman, Ann. Mag. Nat. Hist. ser. 3, vol. xiii. 1864, p- 7 (88). Discopora cruenta, Smitt, Gifvers. Vetensk.-Ak. Férh. 1871, p. 1127, pl. xxi., and 1878, p. 23. Schizoporella cruenta, Hincks, Brit. Mar. Polyzoa, p. 270, pl. xxx. fig 5; id. Ann. Mag. Nat. Hist. ser. 5, vol. vii. p. 449, pl. xxi. fig. 4; id. ibid. ser, 5, vol. xiii. p. 211; id. ibid. ser. 6, vol. ix. p. 153, pl. viii. fig. 5; Ridley, Ann. Mag. Nat. Hist. ser. 5, vol. vii. p. 449, pl. xxi. fig. 4. 74 MR. A. W. WATERS ON Mucronella cruenta, Nordgaard, Bergens Mus. Aarbog, 1894-5, p. 30. Porina ciliata, forma dura, Smitt, “ Krit. Fort.” 1867, pp. 6 & 58, Plcxiv ties 17. Mucronella spinulifera, Hincks, Ann. Mag. Nat. Hist. ser. 6, vol. ii. p. 431, pl. xxi. fig. 3. Monoporella spinulifera, Hincks, Ann. Mag. Nat. Hist. ser. 6, vol. ix. p. 152. The dry specimens on Balanus, from off Elmwood, are reddish black with fairly large zocecia irregularly undulated, and more nearly resemble Smitt’s than any of the other figures; and Hincks may be right in considering that the M. spinulifera is the species described by Smitt, and that it is not the same as the Lepralia cruenta of Norman. The peristomial opening in the older zocecia is subtriangular, having a kind of sinus, and re- sembles in appearance that of Escharoides Sarsii, Sm., and may easily have been taken for adnate #. Sarsit. The younger zoccia show the straight proximal edge to the aperture. The oral aperture in older zocecia is very much depressed, and is not visible from the front; and when Hincks and Ridley speak of a sinus they could only have seen the peristomial opening. The operculum has a straight lower edge, and is fairly characteristic Lepralian (woodcut, p. 78, fig. 1). There is a row of pores round the border. The ovicell has been figured by Smitt in a paper apparently overlooked by Hincks when writing his ‘Brit. Marine Polyzoa,’ but to which he has subsequently re- ferred. In the Elmwood specimens the ovicells show the front zoccial wall extending over the lower corners of the ovicell, which is but little raised. Hincks placed this species under Mucronella, on account of the small projection below the aperture, which is not constant, and is sometimes entirely wanting. Ina subsequent paper he said perhaps it ought to go to Monoporella, and when Leprala and its allies have been thoroughly worked up a group may be separated off as Monoporella, though it is doubtful whether there will be any reason for removing cruenta from Lepralia as we now understand it. This is, however, one cf the numerous instances which shows that IZuweronella ought to be dropped. Loc. Greenland, Spitzbergen, East and West (Sm.) ; Finland, Matotschkin Schaar; St. Lawrence (Hincks); Franz-Josef Land, lat. 79° 55’ N., long. 51° O' EH. (Ridley) ; Hammerfest (Nordgaard). British seas. Jackson-Harmsworth Exp.: off Elmwood, 18 fathoms. BRYOZOA FROM FRANZ-JOSEF LAND. 15 32. Lmpratta Hipporus, Smitt. (PI. 8. fig. 20.) Lepralia hippopus, Smitt, “Krit. Fort.” 1867, pp. 20 & 127, pl. xxvi. fies. 99-105; Hincks, Brit. Mar. Polyzoa, p. 309, pl. xxxiii. figs. 8, 9. This is common on pieces of basalt from off Elmwood, at a depth of 18 fathoms. There are but very few avicularia, and it is often only after careful search that two or three are found on a large colony. The specimens entirely correspond with the description given by Hincks of the specimen from off North- umberland. The operculum does not close the ovicell. There is a Lepralia allied to this in the Southern hemisphere which has been named by MacGillivray Schizoporella pulcherrima. L. pulcherrima has a row of large pores round the border, and a semicircular avicularium at each side of the aperture, and the operculum, which is truly Lepralioid, is shorter than that of L. hippopus. The latter species is very similar to the Lepralia incisa, Busk, from Inaccessible Island (Challenger). Loc. Spitzbergen, Greenland, and Finland (Sm.); Gulf of St. Lawrence (Dawson, and in my coll.). Coast of Northumber- land. Jackson-Harmsworth Exp.: off Elmwood, 18 fathoms. 33. LEPRALIA PORIFERA (Smit). (PI. 8. figs. 14, 15.) Escharella porifera, forma typica, Smitt, Gifvers. Vetensk.-Ak. Forh. 1867, p. 9, pl. xxiv. figs. 30-32. Lepralia porifera, Hincks, Ann. Mag. Nat. Hist. ser. 4, vol. xix. p. 102, pl. x. figs. 1, 2 (1877). PSmittia Landsborovii, var. porifera, Hincks, Brit. Mar. Polyzoa, p- 343, pl. xxxvi. fig. 1 (non Ann. Mag. Nat. Hist. ser. 6, vol. i. p. 225, pl. xiv. fig. 2). Smittia porifera, Lorenz, Bry. Jan Mayen, p. 92; Nordgaard, Syst. fort. over dei Norge, hidt. observ. Art. af Mar. Polyz., Bergens Mus. Aarbog, 1894-5, No. ii. p. 26, pl. ii. fig. 1. The central denticle is wanting, as has been already mentioned by Hincks and others, and in this respect it differs from Smtéza ; also the operculum is fairly thick, and of the Lepralia-type, on which account it is placed under Lepralia. Loc. Spitzbergen and Greenland (Sm.); Norway (Nordg.); Davis Straits (Hincks), given as Iceland in mistake ; Jan Mayen (Lorenz). 8. Devonshire ? Jackzon-Harmsworth Exp.: Lat. 77° 55’ N., long. 53° 16’ E., 130 fath. Besides the species already mentioned, Lepralia megastoma, B., and Z. vitrea, Lorenz, are mentioned as Arctic species. 76 MR. A. W. WATERS ON PORELLA. The genus Porella is very well represented in the Arctic, and therefore some comparisons have been made with Porella from various localities, and the opercula and mandibles of some species not occurring in the present collection, as P. cervicornis, M.-Edw., P. levis, Flem., P. rostrata, Hincks, P. lorea, Alder, have been figured. As the result of these examinations, two eroups have been made, though itis felt probable that the second may be ultimately separated from Porella. The typical Porella may be adnate or erect, and has below or within the aperture an avicularium with a semicircular mandible, and this mandible has well-marked thickenings forming diagonal bars as mentioned by Busk *. The operculum is nearly straight on the proximal edge, with the corners cut off, and a muscular ridge on each side a little distance from the border. The ovicells are always or usually imperforate. In the group there is not usually any lyrula {denticle), though in P. concinna, Busk, and P. minuta, Norm., it can be found. Perhaps with more material the synonyms will be more reduced than I have been able to do in this paper. The second group has a very wide avicularian chamber, in which there is a pair of large glands. Where there has been suitable material these glands have been found. The operculum has muscular ridges farther from the edge than in the first group, and the mandibles have no marked cross-bars, but a lucida in the middle. ‘The mandibles are usually semicircular, but in P. acutirostris they are triangular. In this group are P. saccata, Busk, P. perpusillat, Busk (elegantula, d’Orb.); inflata, sp. nov.; P. acutirostris, Smitt, and apparently Perella marsupium, MacG.; perhaps Lepralia foliacea, B. & S., and P. margaritifera, Quoy & Gaim. In P. rostrata, Hincks, and P. malleolus, Hincks, the mandible spreads out at the distal end, aud this seems to be the case in some other Australian species, and occurs also in P. nitidissima, H.; but as P. rostrata has a distinct denticle, perhaps it should be put under Smittia. We may here remark that although the well-marked diagonal bars to the mandible are a noticeable character, yet the structure is not confined to Poredla, * Zool. Chall. Exp. pt. xxx. p. 147. + P. perpusilla, Busk, from Newfoundland, has a denticle. BRYOZOA FROM FRANZ-JOSEF LAND. TT and in some allies the bars are represented by lines, as in one or two species of Smitéza and Cellepora; and in the mandible of Schizoporella auriculata, Hass., the markings seem to indicate the same structure. Besides the species referred to, P. struma, Norman, is men- tioned from Barents Sea and Finland; and P. patens, Smitt, from Spitzbergen. 34. PoreLLa concinna, Busk. (PI. 11. figs. 9, 10.) From lat. 77° 55’ N., long. 53° 16’ E., there is a small specimen of the typical concinna with pores round the edge, and the peristome raised slightly at each side of the aperture. This also: occurs off Elmwood, and there is one specimen of the variety fig. 13 of Hincks, Brit. Mar. Polyzoa, also from off Elmwood. Loc. Greenland, Spitzbergen, Finland, Jan Mayen; lat. 79° 55’ N., long. 51° O' H. (fidley); Gulf of St. Lawrence, British Columbia (H.) ; Japan (Ortmann). British seas. Jackson-Harmsworth Exp.: off Elmwood, 18 fathoms; lat. 77° 55' N., long. 58° 16’ E., 130 fath.; off Cape Mary Harms- worth, 53-93 fath., and 50 miles off do., 234 fathoms. 35. PoRELLA CoMPRESSA (Sow.). (PI. 11. figs. 3, 4, 5.) Millepora compressa, Sowerby, Brit. Miscell. 1. (1806), p. 83, pl. xli. Eschara cervicorns, Busk, Brit. Mus. Cat. p. 92, pl. cix. fig. 7, pl. exix. fig. 1. Se cervicornis, forma Eschare, Smitt, “ Krit. Fort.” 1867, pp. 23 & 149, pl. xxvi. fig. 138, 139. Porella compressa, Hincks, Brit. Mar. Polyzoa, p. 330, pl. xlv. figs. 4— 7, and woodcut; Lorenz, Bryozoen von Jan Mayen, p. 90; Bidenkap, Zool, Jahrb. vol. x. p. 627. There has been great confusion between this species and the Arctic Cellepora incrassata, Sm., and also between these and the Mediterranean Porella cervicornis, M.-Ed. ; and even Busk, in his List of Polyzoa collected by Captain H. W. Feilden in the North Polar Expedition, gives as Cellepora cervicornis, Busk, a form which is probably C. inerassata, Sm. Miss Jelly in her Catalogue gives Busk’s species under both C. inerassata and P. compressa, and it seems hopeless to attempt to reduce the synonyms to order. Between Porella compressa and C. surcu- laris it is rather a question of names, as the two, although some- times similar in appearance, are quite distinct in the minute characters. On the other hand, P. compressa and P. cervicornis 78 MR. A. W. WATERS ON from the Mediterranean are closely allied, but the branches of P. compressa are much the broader, and in the British specimens there are several layers of zocecia in the older parts. In the Arctic specimens the branches are compressed, and about 4 mil- limetres broad, which is about the same sizeas those of P. saccata and Escharoides Sarsit. In P. cervicornis there is a bifid denticle on the bar of the avicularium. There are also differences between the mandibles of Porella cervicornis, M. Kd., and those of P. compressa, which, though not very great, would alone indicate the probability of these being two distinct species. The Franz-Josef Land specimens have 18 tentacles, and oral glands, but no avicularian glands were seen. There are two distal, and four lateral rosette-plates. Loe. Spitzbergen, Greenland, Jan Mayen (Lorenz), Novaya Zemlya, 30-60 fathoms ; Kara Sea (Smitt) ; Finland ; Norway ; Bohus Bay ; British seas and north coast of France. Jackson-Harmsworth Exp.: off Cape Mary Harmsworth, 53— 93 fathoms. 36. PoRELLA GLACIATA, sp. nov. (Figs. 2, 3.) ? Eschara cervicornis, forma Lepralie, Smitt, “ Krit. Fort.” 1867, pp. 23 & 149, pl. xxvi. figs. 136, 137. A specimen “off the glacier between Cape Flora and Cape Gertrude” is in most respects like P. inflata, sp. nov., but. the entirely different operculum and mandible show that they are quite distinct (see woodcut, figs. 2 & 3). The operculum is very broad, and has a very muscular ridge, and the mandible corre- sponds with that of a Porella sent to meas P. propinqua, Sm. (Pl. 11. fig. 15). Fig. 1. Fig. 2. Fig. 3. : monem i} ee AX!) xas x 250 Fig. 1. Lepralia cruenta. Figs. 2, 3. Porella glaciata. There is a row of pores by the border with ridges between the pores as in P. acutzrostris, Sm. (Pl. 10. fig. 10). The surface of the zoccia and of the globular ovicells is granular, the peristome is raised at the side, the avicularian chamber is wide and distinct with the mandible within the peri- stome, but on the top of a more or less tubular projection. BRYOZOA FROM FRANZ-JOSEF LAND. 79 This may be the Hschara propingua, Smitt (pars), but his figures 181, 132, 134 can hardly represent the same species as figs. 126-128. Jackson-Harmsworth Exp.: off glacier between Cape Flora and Cape Gertrude, about 30 fathoms. 37. Pornnua PLANA, Hincks. (Pl. 11. figs 11-13.) Porella Skenet, Ell. & Sol., form plana, Hincks, Ann. Mag. Nat. Hist. ser. 6, vol. i. p. 221, pl. xiv. fig. 6 (1888). Porella concinna, granular var., Hincks, Ann. Mag. Nat. Hist. ser, 6, vol. ix. p. 156, pl. vill. fig. 6 (1892). The zoarium is bilaminate, foliaceous, though in one small specimen from lat. 77° 55’ N., long. 53° 20’ E., 180 fath., there is only the creeping portion from which the colony arises. Hincks does not say whether his granular variety of P. concinna is foliaceous or encrusting. The shape of the zocecium and also of the secondary orifice is subject to great variation, and numerous round avicularia are scattered over the surface. In the younger _ zocecia there is an avicularium at the proximal edge of the aperture, and usually one on each side, more or less raised and turning inwards, with a few avicularia on the surface of the zoecium. In older zocecia these avicularia are within the aperture and there may be more than three; the surface is finely eranular with large pores round the border of the zocecium; the ovicells, which are also finely granular, are in some cases almost entirely concealed, in others are raised and globular. There are about 20 tentacles, avicularian glands, and also oral glands. I cannot agree in considering this a variety of P. Skene, and think it is entitled to specific distinction, though no doubt closely allied to P. concinna, Busk. Loc. Gulf of St. Lawrence, Trinity Bay, 96 fath. (Hincks). Jackson-Harmsworth Hxp.: Lat. 77° 55' N., long. 53° 20' E., 130 fath., and also lat. 77° 55’ N., long. 53° 16’ E., 130 fath. 38. PoRELLA SKENEI, var. PROBOSCIDEA, Hincks. (PI. 11. figs. 17, 18.) Porella proboscidea, Hincks, Ann. Mag. Nat. Hist. ser. 6, vol. i. p. 223, pl. xiv. fig. 4; Nordgaard, Bergens Mus. Aarbog, 1894-5, p. 25, pl. i. fig. 4, ? Eschara verrucosa, Smit, “ Krit. Fort.” 1867, pp. 22 & 142, pl. xxvi. fig. 185. From off Cape Mary Harmsworth there are some specimens growing adnate upon Céellepora incrassata and Serupocellaria 80 MR. A. W. WATERS ON scabra; but only in one or two specimens are the lateral avicularia found, and then only to some of the zomcia. The suboral avicularium is much raised, the rostrum being continued beyond the avicularium; there are a row of pores round the border of the zocecium, and the ovicell is wide and not punctured. There are many points of resemblance to P. concinna, Busk. Loc. Gulf of St. Lawrence, Orphan Bay, and off Cape Rozier, 38 fath. (H.). If this is Smitt’s species then also Spitz- bergen, Novaya Zemlya and Kola; Finland and North Cape (Nordgaard). Jackson-Harmsworth Exp.: off Cape Mary Harmsworth, 53-93 fathoms. 39. Porras SKENEI, Ell. § Sol., var. tRIDENS, Kirchenpauer- (Gee ab, amie, 57) Cellepora tridens, Kirchenpauer, Untersuchungs-fahrt der Pommerania, p. 188, figs. a & 0. ? Eschara Skenei, var. tridens, Busk, Ann. Mag. Nat. Hist. ser. 2, vol. xviii. p. 33, pl. i. fig. 3 (1856). Cellepora ramulosa, Manzoni, “ Bry. foss. Ital.” Sitzungsber. K. Akad. Wissensch., math.-nat. Cl. vol. lxi. pt. 1, 1870, p. 12, pl. v. fig. 29, pl. vi. fig. 30; Waters (part), Ann. Mag. Nat. Hist. ser. 5, vol. iii. p. 196. Palmiceliaria Skenet, var. tridens, Hincks, Brit. Mar. Polyzoa, p. 380 ; Waters, Zool. Chall. Exp. pt. Ixxix. p. 36 (vol. xxxi.). The Porella Skenet group does not furnish sharp lines by which the ordinary rather stout P. Skenei, HE. & S8., can be readily distinguished from the less stout form with flattened branches known as Porella lorea, Alder, which is somewhat stouter than the present cylindrical form with oral avicularia, usually on well-marked cylindrical processes ; the number of these processes in the specimens examined is usually 3, but Kirch- enpauer says + are frequent. The Porella elegans, Alder, is still more delicate, but so far has only been found where a stouter form algo occurs. Smitt united Porella lorea (Alder) and Porella elegans (Alder) under Discopora Skenei; and it is doubtful whether they will permanently be separated, though at present it may be advisable to indicate in which form the growth has taken place, and it must be remembered that the operculum is much wider and shorter in P. lorea, Alder, so that perhaps on this account it should be separated. P. bicornis, Busk, of the Crag is also closely allied, but it has not been proved to be the same as the erect species. ~~) LINN. JOURN.—ZOOLOGY, VOL. XXVIII. 6 BRYOZOA FROM FRANZ-JOSEF LAND. 81 The operculum of P. Skenez is longer in proportion to the width than in most of the typical Porella such as P. levis, Flen. The mandibles have the characteristic diagonal bars and a raised line below. Loc. Of this variety, Bukenfjord (Kirchenpauer) ; off Nova Scotia (Waters); Naplesand Capri( Waters) ; Norway. [Spitz- bergen, Nordg.| Jackson-Harmsworth Exp.: Lat. 77° 55’ N., loag. 58° 20' E., 130 fathoms. Other varicties are stated to have been found in Norway, Finland, Spitzbergen, Kara Sea, Jan Mayen, Greenland, Shetland, &c., and the coast of France. 40, PoreLua saccata, Busk. (PI. 10. figs. 8-12, 14-17.) Eschara saccata, Busk, Ann. Mag. Nat. Hist. ser. 2, vol. xviii. p. 33, pl. i. fig. 5. Eschara elegantula, Smitt, “ Krit. Fort.” 1867, pp. 24 & 154, pl. xxvi. figs. 140-146; Busk, Zool. ‘Challenger’ Exp., pt. xxx. p. 141, pl. xx. fig. 6; id. Journ. Linn. Soc., Zool. vol. xv. p. 235. Porella elegantula, Levinsen, Bry. fra Kara-Havet, p. 318 (14) ; Hincks, Ann, Mag. Nat. Hist. ser. 6, vol. i. p. 222, pl. xv. fig. 5; Bidenkap, Zool. Jahrb. vol. x. p. 627. Lepralia elegantula, Lorenz, Bry. von Jan Mayen, p. 89. For other synonyms see Miss Jelly’s Catalogue. As it seemed doubtful whether the Eschara elegantula of Smitt is the same as the L. elegantula* of d’Orbigny, I stayed in Paris, en route, in order to examine the specimen described by d’Orbigny, and that I found is what Busk has since described as Porella perpusilla. It should not be overlooked that the specimen with which Smitt made the comparison was, according to Smitt, marked “ Eschara fascialis.’ The specimen from which no doubt d’Orbigny’s description was taken is not so marked, so that I do not gather what Smitt examined. D’Orbigny’s original specimen No. 138622 of Hschara elegan- tula from Newfoundland, now in the Musée d’Histoire Naturelle in Paris, is composed of cylindrical branches between one and two millimetres in diameter, with the ordinary avicularia much smaller than in the more common P. saccata, but the mandibles of both are similar (compare Pl. 10. figs. 11, 12,13). In the Paris specimen already alluded to the avicularia are sometimes * Eschara elegantuia, & Orbigny, Pal. Frang. vol. v. p. 102. 82 MR. A. W. WATERS ON gigantically developed, to which d’Orbigny referred when he wrote “‘pourvus latéralement de cellules plus grandes formant une partie dentée” (PI. 10. fig. 19). These gigantic avicularia have not been found in P. saccata, B. A sketch of the colony (which I should call a memorandum sketch) to give an idea of the growth is reproduced, but this is not an exact drawing (PI. 11. fig. 1). D’Orbigny’s species either must stand for the cylindrical form, or, not having been recognized, becomes P. perpusilla, B., being of course closely allied to P. saccata, B.; but there seems sufficient reason for separation. Unfortunately Busk’s figures and descriptions of P. saccata are far from satisfactory. Another specimen marked in pencil “13622 Terre Neuve, Eschara elegantula” is probably a mistake, and was not so marked by d’Orbigny. Fischer unites 2. pavonina, VOrb., with Porella elegantula, d’Orb., but the specimen 13621 from the Ie de Ré is flabelliform (fig. 2, Pl. 11), with avicularian chambers smaller than those in P. saccata, and also has the ovicells imperforate and not much raised, The Arctic specimens of P. saccata which I have examined start from a broad base, and the colonies grow to 3-4 inches in height, with compressed branches about 4 millimetres broad ; the branches often gradually curve round, so that they are not all in one planeas described by Busk in his ‘Challenger’ specimens. On the front of the zocecium, near the proximal end, there are two pores which open into the avicularian chamber (fig. 14, Pl. 10). These are shown by Smitt in his figure 143, which seems to be from a worn specimen. The pores are not always seen when the zocecium is covered by the exterior membrane, but in specimens boiled in caustic potash and in sections these pores can be made out, and they canalso be distinguished in d’Orbigny’s specimen of P. elegantula from Newfoundland, and readily in the ‘ Challenger’ specimens. Sections show that there are very large oral glands which are placed nearer the basal (neural) wall than are the tentacles, that is they are underneath the tentacles, which is not a usual position for the glands. The cells forming these glands often have large vacuoles, and the contents of the glands are the same homogeneous unstainable material which we have seen in other glands. No avicularian glands were found. Specimens from near Wilezek Land, collected 16th or 17th July, 1897, have well-developed ovaria and testes. BRYOZOA FROM FRANZ-JOSEF LAND. 83 There are 16 to 17 tentacles, and one lateral and two distal rosette-plates. Loc. Spitzbergen, 30-65 fath.; Greenland; Finland, 30-60 fath.; Kara Sea, 46-75 fath. (Levinsen); Jan Mayen, 140-180 metres (Lorenz); Barents Sea (H.); Novaya Zemlya; Labrador; Baffin’s Bay; off Nova Scotia, 51 fath. (Challenger); Murchison Sound, 45 fath. (Hennig). Jackson-Harmsworth Exp.: off glacier between Cape Ger- trude and Cape Flora, about 30 fath.; off Northumberland Island; “off L. S. H.,” about 15 fath.; near Wilczek Land, 127 fath. 41. Porenta acuTrrostRis, Smitt. (PI. 10. figs. 1-5.) Porella acutirostris, Smitt, “ Krit. Fort.” 1867, p. 21, pl. xxvi. figs. 106— 108; Lorenz, Bryozoen von Jan Mayen, p. 90 (8); Hincks, Ann. Mag. Nat. Hist. ser. 6, vol. ii. p. 429, pl. xxi. fig. 5. Porella major, Hincks, Ann. Mag. Nat. Hist. ser. 5, vol. xiii. p. 51 (25), pl. iv. fig. 5. _ Specimens from off Northumberland Island, about 15 fath., have the surface of the zocecium and of the ovicell granulated. The shell when mounted in balsam is very transparent, so that the contents of the avicularian chamber can be readily examined, and the pair of large avicularian glands are seen to extend across a great part of the chamber. This tumid avicularian chamber spreads quite across the zoccium and contracts at each side. These glands are similar to those of Lepralia margaritifera, Quoy & Gaim., and Lepralia foliacea, BH. & S., though somewhat larger, and, judging from the shape of the avicularian chamber of P. inflata, sp. nov., it also doubtless has large glands. The wall of the gland is formed of a layer of nucleated cells, and the contents of the glands are a yellowish homogeneous substance which does not stain. There are 17 tentacles. Loe. Spitzbergen, 16-40 fath. ; Greenland (Sm.); Jan Mayen, 20-180 metres; St. Lawrence (H.); Cumshewa and Houston- Stewart Channel (Queen Charlotte Islands) (Z.). Jackson-Harmsworth Exp.: off Northumberland Island, about 15 fath. 42, PORELLA INFLATA, nom. nov. (Pl. 10. figs. 6, 7.) Porella levis, Smitt, pars (non Flem.), “ Krit. Fort.” 1867, p. 21, pl. xxvi. figs. 112-114, and probably 109-119. In the Hemeschara stage from “ between Cape Flora and 6 * 84 MR. A. W. WATERS ON Cape Gertrude, 30 fathoms,” the zowcia are hexagonal, granu- lated, having a raised avicularian chamber extending across the zocecium, with the avicularian opening somewhat raised and just within the oral aperture, but directed upwards and with a small round mandible. The mandible has a central lucida, but has no diagonal bars as in P. levis from Shetland (Pl. 11. fig. 14). Through the semitransparent walls of the avicularian chamber no avicularian glands can be seen, though no doubt, from the shape of the chamber, they occur at some seasons at any rate. The peristome is raised at each side, but is not continuous round the edge of the aperture. The ovicell is globular, narrow, much raised, with the granulations more distinct than over_the surface of the zocecium. The P. levis, var. subcompressa, Busk, may be related to P. levis of Fleming, but differs considerably from the Arctic P. inflata. The var. subcompressa is given by Kirkpatrick as found off Port Phillip, Australia. Although Smitt placed some adnate specimens from various localities under P. levis, forma lepralia, he does not seem in any way to have shown that there was any identity with the erect P. levis, Flem., and further seems to have placed two or three species together as P. levis. Loc. Spitzbergen, 20-30 fath. (Sm.); Bohus Sea; Norway; Finland. Jackson-Harmsworth Exp.: between Cape Flora and Cape Gertrude, 30 fath. 43. PoRELLA? OBESA, sp. nov. (PI. 12. figs. 22-24.) A specimen from off Elmwood, encrusting a Balanus-shell, has a thick shell with finely granulated surface, and pores round the border which can only be seen in the younger zocecia. The avicularian chamber is wide, extending to the edge of the zocecium, but the outline can only be seen in the younger zoccia; in the same way the ovicell is raised in the younger zoecia, but in the older ones is quite depressed; and the operculum does not close the aperture of the ovicell. There are pore-chambers, 6-8 in all. The lower edge of the operculum makes a large curve, indicating that there is a broad sinus to the oral aperture, and there are two muscular dots a short distance from the border of the operculum. The characters of the operculum would suggest that the species should be placed under Schzzoporella, and near what we may cail the 8. biaperta group. The small avicularian ~ BRYOZOA FROM FRANZ-JOSEF LAND. 85 mandible also has not the cross thickenings characteristic of typical Porella, but has a central lucida. There is, however, a small group, left for the present doubtfully under Perella, though probabiy a genus will have to be made for it when it is fully worked up. In this group the avicularian chamber is very broad, and in the species which have been studied contains a pair of large avicularian glands. The opercula, instead of the long muscular ridges of typical Porella, has muscular dots which may be a little elongate. The mandibles have no cross bars, but have a central lucida. The group at present consists of P. saccata, Busk, P. acutirostris, Sm., P. obesa, sp. uov., P. inflata, sp. nov., and probably P. (Lepralia) margaritifera, Quoy & G., and P. marsupium, MacG. 44. EscuarorpeEs Sarsit, Smitt. (Pl. 11. figs. 21-23.) Escharovdes Sarsii, Smitt, “ Krit. Fort.” 1867, pp. 24 & 158, pl. xxvi. figs. 147-154. I do not think that either Eschara grandipora of Blainville or E. lobata of Lamarck are synonyms. Smitt gives the avicularium as triangular, and on this ground it is separated from L. rosacea, Busk ; but in the specimens now examined, and also in a specimen from the Gulf of St. Lawrence, the mandible is rounded at the distal end, though Hincks speaks of the pointed mandible. Looking down on the avicularium, it may appear triangular, but none of the mandibles that I have been able to examine, either in my own collection or in the British Museum, including the one described by Ridley from Franz-Josef Land and the ‘ Challenger’ specimens, have I seen a triangular mandible or one that could be called “acute.” With these very small avicularia a mistake is easily made unless the mandible is prepared out. The mandibles are not sym- metrical and the amount of irregularity varies, as also the size ; but in the Chilostomata the avicularia generally are subject to considerable variation in size, whereas as a rule the opercula are in a species all of one size, often in a specimen all the opercula ‘seem to correspond in size most exactly. The operculum is thin, transparent, membranous, with a thickened bar across. The ovicell is but little raised, and sometimes there is an elongate area in front. The small avicularium occurs within the peristomial aperture. Some branching pieces of Z. Sarsii, Sm., and Porella compressa. 86 MR. A. W. WATERS ON Sow., from the same locality are undistinguishable from the shape of the colony ; ard when dealing with the earlier descriptions, where the form of growth was considered the most important character, it is impossible to judge which species is referred to. Tt is also said to occur encrusting shells and stones. There are 16 tentacles. There are small oral glands which seem to be attached to the sheath. The contents of the pore- tubes at a slight distance from the external surface of the zocecium are large elongate cells, with a central nucleus, but near the surface the cells are smaller. A figure of the operculum (fig. 25) and mandible (fig. 24) of Escharoides rosacea, Busk, from Loch Tyne is given for com- parison. The operculum of rosacea has a thin muscular ridge at each side and the lower edge of the operculum is nearly straight. The mandible is semicircular, and has a small median lucida and a thinner part at each side, both opercula and mandible being of the Porella type. The mandible in rosacea is however lateral instead of being median. I have referred at some length under Cribrilina punctata, Hass. (p. 62), to the three species of common Arctic Bryozva said to have been found in the Antarctic by Hooker, in the ‘ Erebus’ and ‘Terror,’ and have suggested the probability of an error in labellmg having occurred. The mandibles of the ‘Erebus’ and ‘Terror’ specimen are characteristic of L. Sarsi, and fall within the range of the Arctic specimens examined. Loc. Spitzbergen, 20-60 fath.; Greenland; Finland (Sm.) ; Kara Sea, 49-65 fath. (Levinsen); Jan Mayen, 160-180 metres (encrusting, no avicularia) ; Franklin Pierce Bay, Smith’s Sound, 18 fath. (Busk); Gulf of St. Lawrence (Hincks); Northumber- land Island (Hennig); off Franz-Josef Land, lat. 79° 55’ N., long. 51° 0’ E. (Ridley). According to Hooker and Busk, lat. 74° 0'S., long. 172° 0’ E., 330 fath. Jackson-Harmsworth Exp.: off glacier between Cape Flora and Cape Gertrude, about 30 fath. ; off Cape Mary Harmsworth, 53-93 fath. 45. Porina THBULOSA, Norman. Lepralia tubulosa, Norman, Rep. Brit. Assoc. 1868, p. 308. Anarthropora monodon, forma minuscula, Smitt, “ Krit. Fort.” 1867, pp. 7 & 65, pl. xxiv. figs. 20 & 22. J*or other synonyms see Miss Jelly’s Catalogue. The suboral pore enters the peristome above the operculum. BRYOZOA FROM FRANZ-JOSEF LAND. 87 The oral aperture has a small tooth on each side, forming a Sebizoporellidan aperture. Loc. Spitzbergen, 30-50 fath.; Barents Sea (D’Urban); Jan Mayen, 15-180 metres; Davis Straits, 100 fath. (H.); Gulf of St. Lawrence (4.); Shetland and Wick. Jackson-Harmsworth Exp.: off Elmwood 3 mile, 18 fath.; Giinther Sound, 10 fath.; nr. Wilezek Land, 127 fath. 46. MICROPORELLA SPATULIFERA (Smtt). (Pl. 12. fig. 6.) Lepralia spatulifera, Smitt, “ Krit. Fort.” 1867, pp. 20 & 124, pl. xxvi. figs. 94-98 ; Lorenz, “ Oest. Polarstation Jan Mayen,” p. 89. There is one specimen from near Wilczek Land, and I have one in my collection trom the Gulf of St. Lawrence, sent to me by Canon Norman, and in both cases there is in many zowcia below the club-shaped process a tubular pore or perhaps avicularium, and it would seem that the species should be removed from Lepralia to Microporella. The shape of the aperture is that of Microporeila, and not of Porina. The zoccia and the ovicells are punctate, with one pair of stout spines and one pair of thinner ones above the oral aperture, and one or two small oval avicularia at the side of the aperture. The operculum is thin and does not show the subtriangular shape of the aperture, the distal end is round and the sides are nearly straight. Tne ovicell is not much raised, in fact sometimes not at all. ‘he wide club-shaped process is articulated. 4 The specimen was so loaded with grains of sand, that it was hopeless to attempt to cut sections. Hincks in his papers on the St. Lawrence Bryozoa does not mention this species. Loc. Spitzbergen (Sm.); Finland (Sm.); Jan Mayen, 100- 180 metres (Lorenz); Gulf of St. Lawrence (A. W. W. coll.). Jackson-Harmsworth Exped.: ur. Wilczek Land, 127 fathoms, on shell. 47. ?Smirrra JacKsonit, nom. nov. (PI. 12. fig. 18.) Mucronella coccinea, Bidenkap, Bry. von Ost-Spitzbergen, Zool. Jahrb. vol. x. p. 624, pl. xxv. figs. 5, 6. Wherever it may ultimately be placed, it seems advisable that this Arctic form should be recorded either as a species or variety. In some points it differs decidedly from the British and Mediterranean Mucronella coccinea, Abild.; but it is waste of time trying to decide what Abildgaard meant, in consequence of his meagre description and figures. 88 MR. A. W. WATERS ON As pointed out by Bidenkap, there are 4 oral spines, whereas in the British and Mediterranean UW. coccinea there are usually 6, though in a beautiful specimen of the tesselated variety from Htrétat there are only 4. I bkave been unable to find any lyrula (central denticle); there is a sinus in the much raised secondary aperture, and internal thickenings on both sides of the sinus. There are about 8 pore-chambers to each zocecium, and 20-21 tentacles. The ovicell is smaller than in the Naples IL. coccinea, but the shape of the zocecia and avicularia is similar, and the surface is coarsely granulated. Growing on Diastopora intri- caria there are the primary and the second zocecium of a Smittia, which no doubt is the 8S. Jacksonit. The primary has 13 spines, and the second zocwecium with the mucro much raised has 8 spines. The Naples and British coccinea has 12-13 spines to the primary zocecium. In the sections cut no glands are recognized. This in many respects is similar to Smittia prestans, Hincks, from Australia, especially in the shape of the peristomial sinus (see my figure, Ann. Mag. Nat. Hist. ser. 6, vol.iv. pl. i. fig. 7). Loc. Spitzbergen. Jackson-Harmsworth Exp.: Lat. 77° 55’ N., long. 55° 25’ E., 115 fath. ; lat. 77°55’ N., long. 53° 20' E., 180 fath. 48. SMITTIA TRISPINOSA, Var. LAMELLOSA (Smitt). (Pl. 12. figs. 19-21.) Lischarella Jacotint, forma lamellosa, Smitt, “ Krit. Fort.” 1867, pp. 11 & 86, pl. xxiv. figs. 53-57. Lepraha Jeffreysii, Norman, Proc. Roy. Soe. vol. elxxiii. p. 208. Lepralia trispinosa, Johnst., var., Hincks, Ann. Mag. Nat. Hist. ser. 4, vol, xix. p. 100, pl. xi. fig. 1. Smittia trispinosa, var. arborea, Levinsen, Bry. fra Kara-Havet, p. 320 (16), pl. xxvii. figs. 7, 8. This form should at present be kept distinct, whether we call it 8. Jamellosa, or only a variety. In the Arctic region it seems usually to form erect tubular zoaria. The zocecia are nearly flat and the secondary aperture is not raised, the aperture having a quadrate appearance. ‘The avicularia are triangular, and are directed upwards, usually by the side of the aperture. The ovicells have sometimes a few large perforations as in 8. trispinosa. There are 17 tentacles and about 14 lateral rosette-plates. There are oral glands; and in the section cut it looks as though there were two pairs, one close up to the diaphragm and one connected with the tentacular sheath lower down. Loc. Spitzbergen (Smit); Greenland, 100 fath. (Worman & Sa eee BRYOZOA FROM FRANZ-JOSEF LAND. 89 Hennig); Kara Sea (Levinsen); Davis Straits, 100 fath. (Hincks); Dogger Bank (Hincks) ; Reykjavik Harbour, 15-20 fath. (Wallich). Jackson-Harmsworth Exp.: Lat. 77° 55’ N., long. 53° 16’ E., 130 fath. Lorenz mentions S. trispinosa, Johunst., from Jan Mayen, 160-180 metres. 49, Smirrra Pracutt, Johnston. Discopora coccinea, forma Peachi, Smitt, “ Krit. Fort.” pp. 26 & 170, pl. xxvi. fig. 164; & fig. 167 (as ventricosa). From off Elmwood there are specimens of what may be called very typical S. Peachii, Johnst. There are six spines, the sur- face is granular, and the mucro is distinct. The determination, or rather separation, of S. Peachit, Johnst., 8. ventricosa, Hass., and S. variolosa, Johust., is always a great difficulty, although I have typical specimens of all three. Hincks says that Lorenz has united S. Peachii and S. ventricosa. This does not seem to have been done by Lorenz, though it has been indicated by others. The genus Mucronella is based upon the presence of a mucro ; but various authors place Discopora emucronatu, Sm., with S. Peachii, showing the artificiality of the distinction based upon the presence of a mucro. Loc. Various Arctic localities; Kola, Jan Mayen, Gulf of St. Lawrence (Dawson). Abundant on the British, French, Danish, and Scandinavian coasts, and is given by Pergens from the Mediterranean. Specimens in my collection from Capri are the var. octodentata, Hincks. Jackson-Harmsworth Exp.: off Elmwood, on basalt and on Balanus, 18 fath. 50. Smirrra vENTRICOSA (Hass.), var, (Pl. 12. fig. 17.) Mucronella ventricosa, Hincks, Brit. Mar. Polyzoa, p. 368, pl. 1. figs. 6-8. From off the glacier between Cape Flora and Cape Gertrude there is a Smittia which I think must be considered as a variety of S. ventricosa, Hass. It is growing on Hornera, and the shape of the large, finely granulated zoccia is variable, often being elongate. The border at the upper part of the aperture is raised, sometimes forming a kind of cap, and near the distal border there are four spines; whereas in S. Peachii there are usually six and in S. ventricosa four or six, but in both cases the spines are usually lower down. The mucro is not very wide. There are 18 tentacles. I must however confess to never feeling sure in 90 MR. A. W. WATERS ON the determination of S. Peachii and S. ventricosa, and do not know of any characters by which they can be saticfactorily distinguished. An examination of the specimen from Franz-Josef Land which Ridley called Mlucronella ventricosa var. connectens, shows the pore-chambers distinctly, and these Ridley termed fenestra, but they do not furnish any ground for separation as a variety. The S. ventricosa has been found from various Arctic localities, and is common from the British and French coasts, and Manzoni records it from the Mediterranean. 51. Smrrrra Lanpsgorovit, Johnst., var. (PI. 12. fig. 7.) There is a small encrusting specimen from off Elmwood, 18 fathoms, which seeems to be a variety of S. Landsborovii. Above the distal border of the oral aperture there is a projecting hood to most zocecia, and in the ovicelligerous zocecia this is attached to and extends below the ovicell. The surface of the zovecium 1s punctured with large pores. There has been much confusion concerning S. Landsborovit, especially in the Arctic regions, as no doubt several distinct species were alluded to by Smitt under his Escharella Landsborovit, and in Busk’s British Museum catalogue a mistake was made between S. reticulata, MacG., and S. Landsborovii. The type occurs off the British seas, in the Mediterranean, Australia, New Zealand; and Ortmann gives it as from Japan. Smittia reticulata, MacG., S. reticulo-punctata, Hincks, 8. trispinosa, Johnst., S. rigida, Lorenz, S. abyssicola, Norm., have been mentioned from Arctic localities. 52. P PHYLACTELLA LABIATA (Smitt), (Pl. 12. figs. 3 & 4.) Discopora coccinea, forma labiata, Smitt, “ Krit. Fort.” 1867, pp. 27 & 175, pl. xxvii. fie. 176. Discopora labiata, Smitt, op. cit. 1878, p. 23. Mucronella labiata, Levinsen, Bry. Kara-Havet, p. 325; Bidenkap, Bry. von Ost-Spitzbergen, p. 610. Phylactella grandis, Hincks, Ann. Mag. Nat. Hist. ser. 5, vol. vi. p. 280, pl. xv. figs. 4, 5. Although probably Phylactella, which is so largely based upon peristomial characters, will not be found to be a satisfactory genus as now understood, it has seemed better to leave P. labiata here rather than place it under Smttia. The peristome in front is much raised and entire, but behind it-is nut raised, and there there are four long spines. ‘There is a BRYOZOA FROM FRANZ-JOSEF LAND. 91 row of small pores round the zocecium, and the surface of the zocecia and ovicells is finely granulated. The larve are large and are more fully developed within the ovicell than is usual ; in fact, I have not come upon larve as largely developed in any other species, the corona and cilia being very distinct. A sketch (Pl. 12. fig. 5) of a free swimming larva of Schizoporella unicornis, Johnst., is given for comparison of the size. There are 21 tentacles; while in a Mediterranean Phylactella, which is probably a variety of collaris, there are 15 tentacles. In the aperture there is a wide lyrula. Loc. Spitzbergen (Sm.) ; Kara Sea (Sm. df Lev.), 40-125 fath. ; Barents Sea, 160 fath. (H#.); Finland (Sm.). Jackson-Harmsworth Exp.: off glacier between Cape Flora and Cape Gertrude, about 30 fath.; lat. 77° 55’ N., long. 55° 25’ E., iietasae-) late 772 55 INey lone. 55. 20° EK.) and 55> 16 “E., 130 fath. RHAMPHOSTOMELILA. Has a very narrow lyrula, and the fact of its presence indicates the relationship with Smttia, and probably it must be placed in the family Smittiide. The operculum is thin and delicate, as in Smittia, but there is a raised circular ridge reminding us some- what of the operculum of Membranipora hians, Hincks. ‘The mandibles in some characters approach to those cf some Cellepore, as C. pertusa, Smitt. 53. RuaMPHOSTOMELLA CosTaTa, Lorenz. (P1.11. figs. 26, 27; 2 tos. 1625) Cellepora scabra, Smitt, “Krit. Fort.” 1867, pp. 30 & 181, pl. xxviii. figs. 186 & 188. Rhamphostomella costata, Lorenz, Bry. von Jan Mayen, p. 94, pl. vil. fig. 11 (in error 12 in text); Hincks, Ann. Mag. Nat. Hist. ser. 6, vol. iil. p- 426, pl. xxi. figs. 6, 7, 8. In the Franz-Josef Land specimens, there is a small lyrula, which is slightly bifid, and the suboral rostrum sometimes spreads out at the top as described by Hincks. The gigantic avicularia described by Hincks are very abundant, but the rostral avicularia are usually smaller than in my specimen from the Gulf of St, Lawrence, while the gigantic avicularia are more recumbent. There are no oral spines in the specimens examined, There are 18 tentacles ; and the oral glands are well-developed, consisting of two parts: the first, the pendulous part arising from close to the diaphragm (figs. 1, 2), shows no distinct structure, though there 92 MR. A. W. WATERS ON are indistinct traces of a lobular structure and of large cells; to the base of this is attached the second or globular part, in which there are distinct nucleated cells. This division into two separate parts is indicated in the oral glands of many species, and is very distinct in a few, as in Lepralia eliminata, Waters. Tam very doubtful whether 2. costata should be separated from &. scabra, Smitt, and think Lorenz has made more species than will stand, but I have not the material for a complete study of this Arctic genus. Loc. Gulf of St. Lawrence (H.); Jan Mayen (Lor.); Finland. Jackson-Harmsworth Expedition: off Cape Gertrude, 30 fath.; off Cape Mary Harmsworth, 53-93 fath. 54. RuaaMPHOSTOMELLA PLICATA (Smitt). (Pl. 11. figs. 28, 29.) Cellepora plicata, Smitt, “ Krit. Fort.” 1867, pp. 380 & 184, pl. xxviii. figs. 189-191; Hincks, Ann. Mag. Nat, Hist. ser. 4, vol. xix. p. 106, pl. xi. figs. 3, 4. Smittia plicuta, Hincks, Ann. Mag. Nat. Hist. ser. 5, vol. xiii. p. 52. Ehamphostomella plicata, Lorenz, Bry. yon Jan Mayen, p. 94; Hincks, Ann. Mag. Nat. Hist. ser. 6, vol. iii. p. 426. ? Cellepora bilaminata, Hincks, Ann, Mag. Nat. Hist. ser. 4, vol. xix. Os I salle So ties. Gy C5 ? Rhamphostomella bilaminata, Lorenz, Bry. von Jan Mayen, p. 95, pl. vii. fig. 10. It is difficult to decide upon the range of variation in Rhampho- stomella, but it seems as if the species plicata must stand, and R. bilaminata appears like a vigorous growth of the same, though the absence of spines may be a ground for separation as a variety. At first I named a piece from off Cape Mary Harmsworth R. bilaminata, but afterwards put it under FR. plicata.. The zocrium of FR. plicata is usually nearly smooth, the peristome is much raised, forming a triangular peristomial aperture with a spine on each side, and a miuute lyrula within the aperture.’ There are oral glands close to the tentacular sheath, and there are 17 tentacles. Loe. Spitzbergen, Greenland, and Finland (Sm.); Jan Mayen (Lorenz); Davis Straits (H.); Gulf of St. Lawrence (4.) ; Cumshewa, 20 fath. (Queen Charlotte Islands) (4.). Jackson-Harmsworth Exp.: off glacier between Cape Ger- trude and Cape Flora, about 30 fath.; near Wilczek Land, 127 fath.; off Cape Mary Harmsworth, 53-93 fath. BRYOZOA FROM FRANZ-JOSEF LAND. 93 CELLEPORA. From the genus Céllepora of some years ago, say the time of Busk’s British Museum Catalogue, some groups have been separated : first the Rhynchopora, Hincks; then I have * shown that others should be removed to Lagenipora of Hincks; Lorenz has made the genus Rhamphostomella for an Arctic group. Jullien ¢ created the genus Osthimosia for species having a fairly wide sinus in the oral aperture; and later in the same year+ MacGillivray divided the Céellepore into holostomatous and schizostomous, leaving those with a nearly straight proximal edge in the aperture as Cedlepora, calling the rest Schismopora. The Cellepora of MacGillivray is not represented in the Arctic regions, and only by one or two species in the Northern hemi- sphere, while they are abundant in the Southern. As I have more than once pointed out, itis unfortunate that MacGillivray did not give a generic name to the holostomatous group, leaving the rest as Cellepora to be again reduced as fresh generic characters are found. The Céllepora of all the older authors does not remain in the genus of MacGillivray, and as the genus Osthimosia was earlier than Schismopora, we ought perhaps to accept it, and make a genus Holostomata for the Cellepora of MacGillivray. All the Cellepore with nearly orbicular apertures, or with a distinct sinus in the aperture, have two muscular dots some distance from the border. This is a character which does not occur in the genus Lepralia, but is general in Schizoporella ; though until the genus Schizoporella has been thoroughly worked out, the last word cannot be said about the classification of the Celleporide. A few so-called Cellepora have been removed to Porella. Hoping to be able before long to revise the family Celleporide, it has seemed better to leave the three species at present under Cellepora, though all may have to be ultimately called Osthimosia, Jull. 55. CELLEPORA INCRASSATA, Smitt. (PI. 12. figs. 11-14.) Celleporaria incrassata, Smitt, “ Krit. Fort.” 1867, pp. 33 & 198, pl. xxviii. figs. 212, 216. Cellepora incrassata, Smitt, op. cit. 1878, No. 3, p. 20; op. cit. 1878, No. 7, p. 29; Hincks, Ann. Mag. Nat. Hist. ser. 4, vol. xix. p. 105; Levinsen, Bry. Kara-Havet, p. 324 (20); Nordgaard, “System. fort. over de i Norge, hidtil. observ. art. af Mar. Polyzoa,” Bergens Mus. Aarbog, * Journ. R. Micr. Soc. 1899, p. 8. t+ Mission du Cap Horn, Bryozoaires, p. 64, +t Zool. of Victoria, decade xvii. p. 241. 94. MR. A. W. WATERS ON 1894-5, No. 2, p. 33; Stuxberg “ Faunan p& och kring Novaja Semlja,” Vega-Exp. vol. v. 1887, p. 179; Bidenkap, Zool. Jahrb. vol. x. p. 629. Cellepora cervicornis, Busk, Ann. Mag. Nat. Hist. ser. 2, vol. xviii. p. 32, pl. i. fig. 1; id. Journ. Linn. Soc., Zool. vol. xv. p. 238; Lorenz, Bry. von Jan Mayen, p. 95, pl. vii. fig. 12. Celleporaria surcularis, Packard, “ List of Labrador Marine Animals,” Canadian Naturalist, vol. viii. p. 410. In both the Mediterranean and Arctic region there is a common species of solid cylindrical Cellepora composed of several layers, branching dichotomously, the ends of which are frequently tapering. By the naked eye, species from these two localities cannot be distinguished, and in consequence there has been much confusion, but they are by no means identical and the differences must be recorded. The Cellepora incrassata, Lamk., is stated by Lamarck to be a Mediterranean species, and this is probably what I have considered to be the C. coronopus, S. Woods; and as the minute characters enabling it to be distinguished were first given by Woods and Busk, I consider that this name should stand for the recent Mediterranean form, as Lamarck’s description was quite insufficient. Busk, with a meagre description, named the Arctic form Cellepora cervicornis, and in his paper in the Journal of this Society, 1881, considers that it is not the same as the Mediterranean C. incrassafa, as figured by Marsigli. Un- fortunately there has been so much confusion between various forms that have been called Cellepora cervicornis and Eschara cervicornis, that it would only add to it to retain the name, as Busk himself indicated. The Arctic species has been considered to be the Celleporaria of Packard, and this probably is the case, though he says two or three lines in thickness, and does not give the characters which we now look for. In C. coronopus, W., from the Mediterranean there are large vicarious avicularia with spatulate mandibles and other avicularia with triangular mandibles of varying sizes*. In C. cncrassata there are very large vicarious spatulate avicularia, and also very large vicarious semicircular avicularia, besides the small oral avicularia with semicircular mandibles. In C. inerassata there is a small avicularium at each side of the oral aperture, whereas in C. coronopus there is a small avicularium to oue side of the sinus. The ovicells of C. éncrassata are imperforate with an area in * Waters, “On the Use of the Avicularian Mandibles in the determination of Chilostomatous Bryozoa,” Journ. R. Mier. Soe. ser. 2, vol. v. pl. xiv. figs, 29, 30 ees —s — BRYOZOA FROM FRANZ-JOSEF LAND. 95 front, whereas the prominent small ovicells of C. coronopus from the Mediterranean have a number of large pores over the surface. C. incrassata has 17 tentacles. In the Southern hemisphere there are two solid cylindrical Cellepore. The C. conica, Busk, which does not differ much from O. avicularis, H., has two oral avicularia, the mandibles of which are semicircular, whereas in C. avicularis and C. coronopus they are triangular. There are spatulate vicarious avicularia, and perforate globular ovicells. The C. cylindriformis, Busk, has perforated ovicells, a large oral avicularium with triangular mandible rather to the side of the aperture, and also vicarious spatulate avicularia. Loc. Spitzbergen, 16-160 fath. (Sm.); Greenland (Sm.) ; Novaya Zemlya, 30-80 fath. (Sm.) ; Matotschkin Schaar, 30-50 fath.; Kara Sea (Levinsen); Kola, 28--100 fath. (Sm.); Jan Mayen (Lorenz); Finland (Sm.); Norway (Nordgaard); Labrador, Davis Straits (H.) ; ? Newfoundland (d’Orb.). Jackson-Harmsworth Exp.: off glacier between Cape Gertrude and Cape Flora, about 30 fath.; Wilczek Land, 127 fath.; off Cape Mary Harmsworth, 53-93 fathoms. 56. CELLEPORA PumiIcosa, Busk. (PI. 12. figs. 15, 16.) Cellepora pumicosa, Busk, Brit. Mus. Cat. p. 86, pl. cx. figs. 5, 6; Hincks, Brit. Mar. Polyzoa, p. 398, pl. liv. figs. 1-3; Waters, Ann. Mag. Nat. Hist. ser. 5, vol. iii. p. 198. As it is extremely doubtful whether C. pumicosa, Linn., is referable to this species, it will be best to accept the first reliable description and figure, and call it Busk’s species. The genus does not seem to be a genus of Fabricius, as it was previously employed by Linné. The specimens from Franz-Josef Land have larger zocecia than those from the Mediterranean and British seas, the avicularium is also larger, as are the opercula and mandibles. The ovicells are widely open, imperforate or with one, two, or three large pores. It seems almost impossible to find any satisfactory characters for separating C. pumicosa, B., C. armata, H., aud C. ramulosa, L.; but in what I have considered C. pumicosa the proximal border of the operculum is the arc of a circle nearly as large as that of the distal edge, whereas in OC. armata and C. ramulosa the distal edge fits into what may be called a wide sinus. In all, the oral rostrum may be much longer than is shown in any of the figures with which I am acquainted ; and in a specimen 96 MR. A. W. WATERS ON sent to me by Joliet, from Roscoff, as C. ramulosa there is a long spinous process beyond the mandible. This seems to be represented in the Southern hemisphere by the Cellepora Eatoniensis of Busk, since described by Julhen as Osthimosia eveza, Jul). Loc. British seas; Mediterranean. Jackson-Harmsworth Exp.: Lat. 77° 55' N., long. 55° 25’ E., 115 fath.; lat. 77° 55’ N., long. 53° 16’ E., 180 fath.; 50 miles N.W. Cape Mary Harmsworth, 234 fath. 57. CELLEPORA VENTRICOSA, Lorenz. (PI. 12. fig. 10.) Cellepora ventricosa, Lorenz, Bry. von Jan Mayen, p. 96, pl. vii. fir. 13. The zocecia of this species are much larger than those of C. incrassata, Smitt, and the two species can be distinguished by the naked eye. ‘The operculum is also longer, and the ovicell is narrow and imperforate. There are 21 tentacles. In the Cape Mary Harmsworth specimens I do not find any vicarious avicu- laria, nor does Lorenz mention any. Loc. Jan Mayen (Lorenz). Jackson-Harmsworth Exp.: off Cape Mary Harmsworth, 53-93 fathoms. 58. ReTEPoRA CELLULOSA (Z.). (PI. 12. figs. 8, 9.) Retepora cellulusa, Smitt, “ Krit. Fort.” 1867, pp. 35 & 2038, pl. xxviii, figs. 222-225; Waters, Medit. & New Zealand Retepore, Linn. Soc. Journ., Zool. vol. xxv. p. 259, pl. vi. figs. 17 & 20, pl. vii. fig. 12; Bidenkap, Bry. von Ost-Spitzbergen, Zool. Jahrb. vol. x. p. 630, From off Cape Mary Harmsworth there is a piece about 22 inches in diameter spreading out and forming a rather flat expansion. I am unable to find any oral spines, nor did Bidenkap find any in his Spitzbergen specimens, whereas they are readily found in the Mediterranean RP. cellulosa. Also the operculum does not widen towards the proximal border so much as in the Medi- terranean specimens. These differences are interesting, but are not thought to furnish sufficient ground for separation ; and it is curious that this species, common in the Arctic and the Medi- terranean, is not found off the British coasts. Loe. Spitzbergen, 19-35 fath., Greenland, Kara Sea, Norway, Adriatic, Naples, Capri, Rapallo, Villefranche-sur-Mer. Jackson-Harmsworth Exp.: off Cape Mary Harmsworth, 53-93 fath. BRYOZOA FROM FRANZ-JOSEF LAND. O7 59. RETEPORA ELONGATA, Siniét. Retepora cellulosa, forma notopachys, var. elongata, Smitt, “ Krit. Fort.” 1867, pp. 36 & 204, pl. xxviii. figs. 226-282. Ketepora elongata, Levinsen, Bry. fra Kara-Havet, p. 328 (19), pl. xxvii. fig. 11; Bidenkap, Bry. von Ost-Spitzbergen, Zool. Jahrb, vol. x. p. 629; Waters, Journ. Linn. Soe., Zool. vol. xxv. p. 256, pl. vi. figs. 9,10; Hennig, Cifv. af K. Vetensk.-Ak. Férh. 1896, p. 361. Discopora elongata, Smitt, CAfv. af K. Vetensk.-Ak. Forh. 1878, pp. 25 & 32. Retepora Wallichiana, Hincks, Ann. Mag. Nat. Hist. ser. 4, vol. xix. p- 107, pl. xi. figs. 9-13, 1877 ; id. op. cit. ser. 5, pp. 29 & 42; Nordgaard, Mar. Polyzoa, Bergens Museums Aarbog, 1894-5, No. 2, p. 31. Retepora tenella, Ortmann, Die Japanische Bryozoen-fauna, Arch. f, Naturgesch. 1890, p. 34, pl. ii. fig. 21. There are 15 tentacles, and large oral glands hanging down from the diaphragm a long way into the zowcium. A specimen from off Cape Mary Harmsworth (Aug. 7th) has the ovaria and testes well-developed. fetepora Beaniana, King, has been found off Jan Mayen. Loc. Spitabergen, 20-80 fath. (Sw.); Greenland; Finland, 50 fath.; Kara Sea (Z.); Kola (Sm.); Novaya Zemlya, 30-50 fath. (Sm.); Matotschkin Schaar (Sm., Stuxb.); Davis Straits (H.) ; Godthaab, 150 fath.; Murchison Sound (Hennig in litt.); Queen Charlotte Islands, 15-20 fath. (H.); Vancouver Island (#.) ; Inglefeld Gulf, 25 fath. (Hennig); Sagamibai, Japan, 200 fath. (Ortmann). Jackson-Harmsworth Exp.: off Cape Mary Harmsworth, 53-93 fath. LINN. JOURN.— ZOOLOGY, VOL. XXVIII. 7 MR. A. W. WATERS ON 98 uvdep “UvdUBAIeqpayy pue copeaqey “eOURIMBT “4G Jo J[NyH "wag syueleg pue vljmeg efeaony "Bag BIT * KK OK “uesaaqzyidg > O% “pueluserp * * “sur P&G * * * “SUI 66-&9 —-+, —— “EON “YAO MS UA FL Aavyy odeg, NoLL YET § ON * AISZO[E AG ON "YOST ‘ON “sopoequoy, “nemo p ‘uafeurds gs “HY “* (‘qdQ,p) 79220.00 i 9] se ecoesee (Tq) ppnaur) “ce "GT ‘esses yengy ‘skyonjsouow nuodruniquayy “FL “ggg “pynounj-oaopupaquiad “ “eT "TOS 9 “]|GL ‘vasnguva Dupsnjy “GT ‘(ysng_) Zyavaq é TT (ung) vyphuoja ue “OT (“mao ny) 22pp0eegy aS 6 (ung) szqavub “ABA *090U.102 (uag avs) nsqnas p21 7)0000n.LIgy ‘Aou'ds “22y240MSULOFT = youd “vsoa -2jntf “IBA - qsugoe ‘vuvivunpy oynbing vs aod “ds ‘mun “ see sao ds ‘opibef Dipzaug we * (KY) 77100240) Lu] paUna 4) “ Ro Se 0) “6 tc aA tis 99 BRYOZOA FROM FRANZ-JOSEF LAND. *VIJOOG GAO NT “81099 BAON, AAGTERUA YO) qsitg 2. uvdue “udjonsio yy } ‘avptjodousog ‘uedn suredg ‘yr * OK OK * * tok ok OR Xe OK eee * oH * KK eee > % «2 68 €8 GL “* Aou “mou ‘ngnpfiL2 o mreses UT) *S2.0980U12NID SS oe cy ysng ‘NID990S “ce steeeeaeeeeeees SUIDAT yy ‘suapui “leA “lauagg es Leese “syOUL_, “apa -sogoud "wer ‘aways “ eee cee ces syoulyy ‘ound «e Tee oA OTT ‘ds “pypravb ra “595 (MOQ) vssaiduoo ENE “MUULIUOD DIPILOT seererees ug “‘p.1a {rod “ce panoooOnOC THR ‘sndodday “c see TOLLO NT “209A a “*,0u "ds ‘s2jvau0g = sess (Mig) puaaws pyppiday “(saeg) vypupod vugsnpopnasd sec ce eae T “nuyolhy “t * mosmucy ‘vsundia noyjoddayy *q10,P ‘ajvonubqns % “-(S1eg) win200000 wWnozormuh jy copbagaD aOR OGTR ‘ds ‘@zpoomutja “ “Aou “WOU “VY IhOMSUMO FT we “("mg) vaongsn1a DyjaLodozwmagy * Cqeq) vgppnuun a ““(ssep_) vpngound 1u27r192.4) [Ne ‘wzuaponu viodrun.squa yy “9G “GG “PG apne ee - — aa =~ a f | | | sit | "T9940 “189 5) x» urdup atte Bas oxi eee * % # % * eee % ane wee see ; cl 16 seen ee “Wg “‘nywhuoja « “GC ce eae |e ete ve * ee ieee ase * Stele a Seo ope [ee ral OG. (es (TY) psognppaa podajayy “RG | ie : . wee free | owe | oes oon * ioe lec ede Te | 96 | 'Zteaory “soaruguan = “LG | & * eee nee tee Aci Sc an Ann * see * eee see OZ-S1 C6 trees ysng “psoowund 4 “gg x * x % * | x | x Sacral Gtk SE ae Has LT (GR errr “Ug “azpsspwour p.codapja) “GG i e}O]tey,) 1) aod ane ie % sae eee % % % ee | x | see * x LT | oR (wg )opnaud “a “EC + P| imooo ape |) 300 ||, : 95% If gm * |] QT | 16 Nor] ‘M20)800 mpjarwozsuydiumyy “EG feet Rod) Se ees pats eat * || Te | 06 | Cwg) vyerqn7 nppozonpiyg EE | | | } | | * i nae | 06 . BAP “WULO.LOQS PUD'T or TG, x, | | | * eat GQ ava ‘CSsVzT) vsoars2ue0 ‘i OE © * * om om oA || ib a eae | * Si 68 | “ysUyOR yo © OF BZ | | lene | on * ano * Sou BHO Li aren jaaencoo (‘wig Josoy Fs | | | | -jauln) “tea “psourdsrip “SE 5 | fs 4p 980 | one fea, = Se eee + © E 2 2a ee Se se a | Si =) St e > P| QE | me a ; Pao a | et | 5 is} OF oly oO 5 AS =] | | =I eal Bw | -~I S . 5 5 © “Ytomsutepy | —T | ee | | : : : z | Awe eden A f for | 8 | | 2 | | 100 15, BRYOZOA FROM FRANZ-JOSEF LAND, 101 EXPLANATION OF THE PLATES. PLATE 7. Brettia frigida, sp. nov., < 12. Off Cape Mary Harmsworth. . Do.; natural size. é . Do. Dorsal surface showing the distal rosette-plate, through the transparent wall, x 25. . Gemellaria loricata (L.); Gunther Sound. This shows the creeping zocecia from which the branches arise. x 12. . Brettia minima, sp. nov.; X 12. Off Cape Mary Harmsworth. . Do. ; natural size. . Do. Dorsal surface, x 25. 9. Scrupocellaria Smittii (Norman); X 25. 50 miles N.W. Cape Mary Harmsworth. Do. Avicularium with mandible, x 85. . Do. Avicularian opening, x 85. . Scrupocellaria ternata, var. gracilis, Sm.; X 25. Gimther Sound. Showing the articulation. In this specimen some zocecia have two outer spines, and sometimes the median zocecium is acuminate. Lateral avicularia rare. . Bugula Harmsworthii, sp. nov.; x 8. Off Cape Mary Harmsworth. . Serupocellaria scabra (van Ben.); < 25. Off Cape Mary Harmsworth. . Do. Scutum, x 50. . Do. Dorsal surface showing a vibraculum at the bifurcation, and one to a lower zocecium. X 25, Puate 8. Bugula Harmsworthii, sp. nov.; xX 25. Off Cape Mary Harmsworth. . Lepralia sincera (Sm.). Operculum, x 85. Off Cape Mary Harms- worth. Membranipora monostachys, Busk; X 15. Near Wilezek Land. Lepralia borealis, sp. nov.; X 25. Specimen mounted in balsam, showing through the semitransparent shell the position of the avicularian chamber. From lat. 77° 55’ N., long. 53° 16' E. . Do. Operculum, x 85. . Do. Mandible, x 250. . Pseudoflustra palmata (Sars). Operculum, X 85. Lat. 77° 5d’ N., long. 53° 16' B. . Do. Mandible, x 250. . Do. Transverse section, showing the two distal rosette-plates. . Membranipora macilenta, Jullien; x 25. Near Wilczek Land. . Schizoporella crustacea (Sm.). Mandible, x 250. Gtnther Sound. Do. Operculum, x 85. Do.; x 50. Showing primary zowcium and the three zowecia growing from it. The development of the zoccium on the left has heen arrected. Giinther Sound. . Lepralia porifera (Sm.), Mandible, x 250, From lat. 77° 55’ N., long. 53° 16’ E. Do. Operculum, x 89. 102 Fig. 16. 1 . Do; do. Operculum of ovicelligerous zocecium, x 85. . Hippothoa expansa, Dawson. Operculum, x 85. Off Elmwood. . Lepralia hippopus (Sm.). Opereulum, x 85. Off Elmwood. . Cribrilina annulata (Fab.); Xx 25. Off glacier between Cape Flora Fig. 1. — CO cr MR. A. W. WATERS ON Hippothoa hyalina, 1.. Operculum of male zocecium, x 85. From California. Do.; do. Operculum of female zocecium, x 85. and Cape Gertrude. . Cribrilina punctata (Hass.); x 25. Lat. 77° 55’ N., long. 53° 16’ E. Puate 9. Schizoporella elmwoodie, sp. nov.; X 25. . Myriozoum coarctum (Sars); X 25. Off Cape Mary Harmsworth, 234 fathoms. . Do. (a) Operculum, X 85. (6) Mandible, x 85. . Myriozoum subgracile, VOrb.; x 25. . Do. (4) Mandible, x 85. (6) Mandible, x 250. (c) Operculum, x 85. . M. subgracile, VOrb. Section of tissue showing the polypides in position, the covering integument (a), over the inner one (0); also the long pore-tubes which in various places have a disk separating the contents on the two sides of the disk. The oral diaphragm is seen as withdrawn. X 25. . M. subgracile, @Orb. Section through polypide showing dm, the diaphragm; op, the operculum, with dotted lines to show the position when partly open; d, disks in the pore-tubes. 89. . M. subgracile, d’Orb. End of pore-tube showing the covering integument (a), and the inner one (4), X 500. Examined with ,5- immersion. . M. subgracile, Orb. Covering integument (a). X 85. . Schizoporella Harmsworthii, sp. nov.; X 25. Lat. 77° 55'N., long. 53° 16’ E. . Do. Tentacular sheath with oral glands attached near the oral aper- ture, and connected to the lateral walls; x 85. (July 13th, 1897.) . Do. Oral glands, x 250. . Schizoporella elmwoodie, sp. nov. Operculum, x 85. Puatz 10. . Porella acutirostris, Sm.; xX 25. From near Northbrook Island (“LS.H.”). . Do. Mandible, x 250. . Do. Operculum, x 85. . Do. Piece mounted in balsam, showing the avicularian glands through the semitransparent shell, x 50. . Do. Section of avicularian gland, x 250. Fig. 6. 14. 15. 16. 17. 18. I), or pe fer) -1 BRYOZOA FROM FRANZ-JOSEF LAND. 103 Porella inflata, nom. noy.; X 25. From between Cape Fiora and Cape Gertrude, 30 fath. . Do. Mandible, x 250. . Porella saccata, Busk; X 25. From near Wilczek Land, 127 fath. Section showing the position of the oral glands (g/), which are behind the tentacles so that they are not seen in the zoccia cut through near the frontal surface ; 0v, ovum. X 25. . Do. Transverse section showing the large oral glands (g/), the avicu- arian sheath (av), the tentacles (¢), in their sheath, cesophagus (@), cecum (¢), testes (fe). x 50. . Do. Oral gland, x 250. In the upper part the cells are partly absorbed. . Porella saccata, B. (elegantula, B.). Mandible, x 85. Off Nova Scotia (‘ Challenger ’). . Do. Mandible, x 85. Near Wilezek Land. 13. Porella elegantula, VOrb. (perpusilla, Busk). Mandible, x 85. Newfoundland. From d’Orbigny's type-specimen No. 13622, Mus. Nat. Paris. Porella saccata, Busk; * 25. Specimen boiled in caustic potash, showing the large avicularian chambers, that of the left zocecium has been broken away, also showing the pores at the base of the avicularian chambers. From near Wilczek Land, 127 fathoms. Do. Operculum, x 85. Do. Mandible, x 250. Do. Diagrammatic section, showing the frontal pore connected through a rosette-plate with the avicularian chamber, which farther down is similarly connected with the zoccium. (ov.) ovicell, (av. ch.) avicularian chamber, (z. ch.) zocecial chamber, (mm.) man- dible, (op.) operculum. Porella elegantula, VOrb. Operculum, x 85. Do. From the original specimen No. 18622, showing the gigantic avicularia on the left, and pores on the front similar to those of P. saccata, Busk. Puate LI. . Porella elegantula, Orb. Specimen from Newfoundland named Eschara elegantula by d’Orbigny, No. 13622 in the Musée d’Histoire Naturelle. This is only a reproduction of a rough sketch made as a memorandum of the way in which the specimen grew. Nat. size. . The Eschara pavonina of dOrbigny, Ile de Ré (Pal. Fr. vol. v. p. 101. No. 18621, Mus. d’Hist. Nat.). . Porelia compressa, Sowerby. Operculum, x 85. British specimen. . Do. Mandible, x 250. British specimen. Identical with Arctic. . Do. Calcareous section showing the position of the pore-tubes, x 25. Off Cape Mary Harmsworth. . Porella Skenei, var. tridens, Kirch. Operculum, x 85, Lat, 77° 55’ N., long. 53° 20’ E. . Do. Mandibles, x 250. Fig. Figs. Fig. . Rhamphostomella costata, Lorenz. Section showing the oral glands . Schizoporella unicornis, Johnst., x85. Free swimming larya from . Microporelia spatulifera (Sm.), X 85. . Smittia Landsborovii, Johnst., var., x 25. . Retepora cellulosa (L.). Operculum, x 85. Off Cape Mary Harma- . Do. Mandible, x 85.. . Cellepora ventricosa, Lorenz. Operculum, xX 85. . Cellepora incrassata,Sm. Operculum, xX 85. From Greenland. 12. MR. A. W. WATERS ON . Porella cervicornis, Pallas. Mandible, x 250. Naples. . Porella concinna, Busk. Mandible, x 250. Durham. . Do. Operculum, x 85. Durham. . Porella plana, Hincks. Mandible, x 250. . Do. Operculum, x 85. . Do. Operculum, x 50. From lat. 77° 55’ N., long. 53° 16’ E. : . Porella levis, Flem. Mandible, x 250. Shetland. d . Porella from Gulf of St. Lawrence (sent as P. propingua, Sm.). Man- dible, x 250. . Porella rostrata, Hincks. Mandible, x 250. Port Phillip Heads. . Porella Skenei, var. proboscidea, Hincks. Mandible, x 250. Off Cape Mary Harmsworth. . Do. Do. Operculum, x 85. . Porella lorea, Alder. Mandibles, x 250. Shetland. . Do. Operculum, x 89. . Escharoides Sarsii, Sm. Mandible, x 250. Off Cape Gertrude and Cape Flora. . Do. Operculum, x 85. Do. . Do. Mandible, x 250. Said to be from the Antarctic, lat. 74° 0’ S., long. 170° H.; 330 fathoms. . Escharoides rosacea, Busk. Mandible, x 250. Loch Tyne. 25. Do. Operculum, x 85. 26. Rhamphostomella costata, Lorenz. Operculum, x 85. Off Cape Gertrude and Cape Flora. . Do. Mandible, x 8&5. . Rhamphostomella plicata, Sm. Operculum, xX &5. . Do. Mandible, x 85. . Rhamphostomella bilaminata, Hincks. Mandible, x 85. Gulf of St. | Lawrence. PLATE 12. terminating near the diaphragm, x 85. From off Cape Mary Harmsworth. Do. Gland, x 250. Examined with ,';-immersion. & 4. Phylactella labiata, Sm. Larva in ovicell, x 85. From lat. 77° 55’ N., long. 53° 16’ E. July 13th. Trieste, for comparison of the size. worth. Do. Vicarious mandible, x 85. BRYOZOA FROM FRANZ-JOSEF LAND. 105 Fig. 13. Cellepora incrassata, Sm. Vicarious mandible, x 85. From same specimens as figs. ]] & 12. 14. Do. Lateral oral mandible, x 85. From near Wilezek Land. 15. Cellepora pumicosa, Busk. Mandible, x 85. Lat. TUS tafsy IN, long. D5?) DY’ 1. 16. Do. Opereulum, x 85. 17. Smittia ventricosa (Hass.), var. x 2b, Off glacier between Cape Flora and Cape Gertrude. 18. Smettia Jacksonii, nom. nov. Mandible, x 85. Lat. UT° Bey! I. long. 55° 25! KE. Figs. 19, 20. Smittia lamellosa (Sm.). Operculum and mandible. « 85. Fig.21. Do. Do. x 25. 22. Poreila obesa, sp. nov. Mandible, x 250. 23. Do. Operculaum, x 835, 24. Do., x 25. Off Elmwood. LINN. JOURN.—ZOOLOGY, VoL. XXVIII. § ra at bcs BHR a \ Wh Linn. Soc. Journ. Zoor. Vou. XXVITPL 7. J oo fs Te (/\ | 1968) eae aa Ss alll (8 5) ei Seg = West,Newmsn,imp. ~ Pek vOZOA FROM FRANZ -JOSEF LAND Waters. Linn. Soc. Journ. Zoon. Vou. XXVIILP1.8 2 (85) Se en 7 (85) i a SS 14 (250) jp ee ery, pom Ses Be * an! ree Be rep) | a Bis. AW Waters del. 21 (x25) West, Newman imp. _A'T.Hollick lith. Ei wOKAON EROM PRANZ-JOOKF LAND hes): 19) Linn, Soc. Journ. Zoou. Vou. XXVILL eeraeicaye: ) West,Newman imp. 12 (*250) AT. Hollick lith. - 10 (25) AW Waeters del. BiEMWwOZAO HAO YRANZ=JOSEE LAND. Linn. Soc.Journ. Zoou. Vor XXVILL P1.10 10 (250) 19 (25) 16 (250) West,Newman imp AW Waters del. ] AT Hollick lith. BYOZOA FROM FRANZ-JOSEF LAND. iS 15 Law. Soc.Jourw. Zoor, Vou. XXVITPI11. 22 (x85) 28 (1250) 24:(*250) ae a 29 ang) 30(«85) West Newman imp. Linn, Soe Jovan. Zoot, Vou. XXVIII PI 12. a ir fe 21 (25 24 (25) AW Waters del. Be. Bees) West,Newman imp. A.T Hollick lth. BRYOZOA FROM FRANZ-JOSEF LAND. BRYOZOA FROM FRANZ-JOSEF LAND. 105 Fig. 13. Cellepora inerassata, Sm. Vicarious mandible, x 85. From same specimen as figs. 11 & 12. NN 14. Do. Lateral oral mandible, x 85. From near Wilezek Land. 15. Cellepora pumicosa, Busk. Mandible, x 85. Lat. 77° 55!’ N., long. / 55° 25! Hi. ? at 16. Do. Operculum, x 85. 17. Smittia ventricosa (Hass.), var., X 25. Off clacier between Cape Flora and Cape Gertrude. 18. Smittia Jacksonii, nom. nov. Mandible, x 85. Lat. 772 55' N long. 55° 25' E. Figs. 19, 20. Smittia lamellosa (Sm.). Operculum and mandible. x 83. ieee Don Dox xX 25! 22. Porella obesa, sp. nov. Mandible, x 250. 23. Do. Operculum, x 85. 24. Do., x 25. Off Elmwood. OF) Observations on Syllis vivipara, Krohn. By Epwin 8S. Goopricn, M.A., F.L.S. [Read 7th June, 1900.] (Pxate 13.) Wattst working last winter at the Zoological Station in Naples, I found in the tank of the large laboratory an interesting small Syllid, which I believe to be the Syllis vivipara originally named by Krohn in 1869 (2). Since Krohn gave but a very brief description of his worm, without figures, merely stating that in general structure it closely resembles Claparéde’s Syllis Armand: (probably 8S. prolifera, Krohn), it is by no means easy to make certain whether we are really dealing with his species. The worm redches to a length of 2 cm., and appears pale yellow in colour owing to the intestine, which is seen through the transparent aud almost colourless body-wall (Pl. 13. fig. 1). The palps are joined together below the prostomium for about half their length (figs. 1&2). The dorsal cirri are of considerable length, especially in the anterior region (fig. 1). As Krohn mentions, the terminal joint of the chet are provided with a single hook (fig. 6). Internally, the alimentary canal is of quite normal structure. The pharynx possesses a single tooth, aud nine papille at its anterior end (figs. 1 & 3). LINN. JOURN.—ZOOLOGY, VOL. XXVIII. 8 106 MR. E. 8S. GOODRICH ON SYLLIS VIVIPARA. Tt is with regard to its method of reproduction that this Syllid is of peculiar interest. In all the specimens I examined (about twenty) either ova, or embryos in various stages of development briefly described this viviparous reproduction, and stated that the embryos grow within the body-cavity of the parent to an advanced — stage, when they completely resemble the adult excepting in their smaller size and lesser number of segments. According to this observer, they then escape by the breaking-off of the hinder region of the parent’s body. Ehlers, in his large work on the Polycheta (1), gives an account of Krohn’s observations, and suggests that fertilization must take place by way of the nephridia. Since Krohn’s note, published more than 30 years ago, no confirmatory evidence has been given of viviparity in any Syllid, excepting for the mere mention of a case (Syllis incisa) by Levinsen (3). Malaquin, in fact, in his monograph of these Polychxta (4), throws doubt on the accuracy of the observations recorded by these authors. The figures herewith given will, I hope, place beyond doubt that we have here a truly viviparous Syllid. In Syllis vivipara the ova are produced only in the posterior third of the body, where they are shed into the eelom in com- paratively small numbers. Here they undergo development from before backwards, and as many as eighteen embryos may be found in one parent. The course of development seems to differ in no marked way from that pursued by other Syllids outside the mother. Malaquin’s figures of the stages in the ontogeny of usyllis monilicornis (4) would apply almost equally well to Syllis vivipara. As the embryos grow larger they have some difficulty in accommodating themselves within the parental body-cavity. They usually take up a longitudinal position, with their head directed backward, and extend along many segments through the incomplete septa. In fig. 1 (Pl. 13) is shown a worm the young of which must be almost ready to escape. They have fully acquired the parent’s form, differing from it only in size. The ciliain the rectum can be seen to work from behind forwards, as in the adult. We are naturally led to enquire how these embryos are nourished, and how the ova from which they have developed were fertilized. With regard to the first question, there can be no doubt what- y| were found in the ccelomic cavity of the posterior region. Krohn \ ? 4 MR. E. S. GOODRICH ON SYLLIS VIVIPARA. 107 ever that the yolk stored in the egg is not enough to provide nourishment sufficient for the growth of the embryos to the large size they attain in the body-cavity of the parent. Further food- material the young must derive from the mother during develop- ment; and, since there 13 no special connection of tissues between the two, this food would appear to be taken in either by mere diffusion through the general surface of the body, or by the mouth or anus. In some such way the embryos must obtain nourishment from the ccelomic fluid of the mother. The second question is less easy to answer. In the first place, since all the specimens I found contained ova, it is just possible that we have here a case of parthenogenetic development *. Should, however, spermatozoa actually at some time penetrate to the ceelom and fertilize the ova, it is by no means easy to see how such a process couid take place. The surface of the body is covered with a cuticle, and it is not likely that the spermatozoa reach the celom by entering the alimentary canal and boring their way through its wall. The only other means of access would appear to be the nephridia. Now these organs, so far as T have seen, do not undergo any of those changes in size and structure at maturity, such as are known to take place in many allied forms. The nephridia are delicate tubes, with a very narrow lumen, opening internally by a small nephrostome, and externally by a minute pore. The cilia of the nephrostome and of the canal act in such a way as to force substances down the canal to the exterior. That spermatozoa struggle up this narrow and difficult path is hard to believe. One possibility remains to be considered, though it must be admitted to be a somewhat improbable one—namely, that it is a case of self-impregnating hermaphroditism. A few viviparous hermaphrodite Polychztes are known, such as Neres diversicolor, Salmacina Dysteri, and Pomatoceros triqueter ; and it has been suggested by Schroder for the first species (7), and by de Saint- Joseph for the two last (6), that in these worms self-impreguation occurs. The evidence of hermaphroditism in Syllis vivipara is by ao means convincing. In sections of the hinder segments, whilst * MM. Mesnil and Caullery have already suggested that in one viviparous form of the Cirratulid Dodecaceria concharum the development of ihe ova is parthenogenetic (5). §* 108 MR. E. S. GOODRICH ON SYLLIS VIVIPARA. ova can be seen loose in the body-cavity, the segmental gonads have in places very much the appearance of testes, the cells at the periphery being numerous and small; but I have never found ripe spermatozoa, and the point remains doubtful. It is in the hope of inducing some observer to settle this interesting question, that I publish the foregoing observations. BIBLIOGRAPHY. (1) Enters, E—Die Borstenwiirmer. Leipzig, 1864-8. (2) Krony, A.—‘ Ueber ein lebendiggebirend Syllisart.” Arch. fiir Naturg. voi. xxxv., 1869. (3) Levinsen.—“ Syst. overs. over de Nordiske Annulata.” Vidensk. Meddelelser, Copenhagen, 1883. (4) Matraguin, A.—Les Syllidiens. Lille, 1893. (5) Mesuit, F., & M. Cauttery.—“ Les formes épitoques des Annélides.’’ International Congress of Zoology, 1898. (6) Satna-JosEpH, Baron de.—‘‘ Sur les Annélides poly chétes des 7 cotes de Dinard.” Ann. Se. Nat. vol. xvu., 1894. (7) Scuréprr. — Anatomisch-histologische Untersuchung von Nereis diversicolor, O. Fr. Mill. Kiel, 1886. EXPLANATION OF PLATE 13. (All the figures are of Syllis vivipara, Krohn.) Fig. 1. Enlarged dorsal view of an adult specimen containing several weil- developed young. d.c., dorsal cirri; ph., pharynx; 7., intestine ; Y-) young: 2. Ventral view of the head, showing the mouth (m.) and palps (p.). 3. Sketch of the pharynx in the extended condition. 4, Ventral view of a portion of three posterior segments containing ova and developing embryos. pp., parapodium ; s., septum; 7., intestine ; e., embryo ; d.c., dorsal cirri; 0., ovum. 5. Transverse section of an adult worm, showing two embryos in the celom. w., body-wall; 7., intestine ; pp., parapodium; y. y., young. 6. Distal end of a cheta. ae ee | a = — on : = : Pee aS g 5 | | : a | : = Wu (e) je) N i Ke) (e) me Oo (e) 2 | 2 So pete ac] 7 S 7 E Q a) - PAP PODVCIiION OF -SYLEIS VIVIPARA., MISS G. SWEET ON AUSTRALIAN EARTHWORMS. 109 On the Structure of the Spermiducal Glands and Associated Parts in Australian Earthworms. By Grorerna Sweet, M.Se., Melbourne University. (Communicated by Prof. G. B. Howzs, Sec. Linn. Soe.) [Read 21st June, 1900.] (Puates 14 & 15.) In the present somewhat unsatisfactory condition of the classifi- cation of the Australian earthworms, additional knowledge on important points in their anatomy is much needed. At the suggestion of Prof. Spencer, the following investigation has been earried out on what is one of the most constant and conspicuous organs in our Australian earthworms. This is the spermiducal gland, which is characteristic without exception of the indigenous forms in Australia. In addition to this, the study has included the histology of the various accessory structures and ducts associated therewith, for while the general macroscopic structure in these forms has been well described by Prof. Baldwin Spencer and Mr. J. J. Fletcher, the microscopic details have been comparatively unknown, except in the one form, Megascolides australis, also described by Prof. Spencer *. And bere I would acknowledge my indebtedness to him for allowing me the use not only of his laboratory in the Melbourne University, but also of his large collection of Aus- tralian earthworms f and of literature thereon; also for much valued assistance on many ocvasions. I have throughout made use of the generic divisions given by Beddard in his monograph of this group, this being the most easy means of reference. The characteristic genera of earthworms in Australia are Megascolides, Cryptodrilus, Diplotrema, Fletcherodrilus, Diporo- cheta, Megascolex, Digaster Gncluding Didymogaster and Perisso- gaster (Fletcher)), and Acanthodrilus. The 32 species examined include a number from Queensland, New South Wales, Victoria, and Tasmania. The male reproductive organs in these forms, as is well known, consist typically of three parts :—(1) two pairs of testes, generally * Trans. Royal Society of Victoria, vol. i. pt. 1, 1888. + The specimens of earthworms described by Mr. Fletcher were kindly sent by him to Prof. Spencer. 110 MISS G. SWEET ON THE SPERMIDUCAL somewhat flat finger-like bodies attached to the anterior septum in segments 10 and 11. They are enclosed in a very delicate membrane connected with the peritoneal lining of the general body-cavity. The sperm mother-cells, which originate by multi- plication of the cells of this peritoneal lining, pass out into (2) the seminal vesicles or sperm-sacs. These are generally found in pairs in segments 9 to 12, where they arise as out- growths of the septa, enclosing a small portion of the body-cavity. Here the sperm mother-cells develop so as to form by division the spermatozoa. These pass to the exterior by means of (3) two pairs of vasa deferentia. The latter open by means of two pairs of much-folded and ciliated funnel-shaped structures, from the segments in which the testes lie, and lead back through the following segments to open externally by a single pair of openings on segment 18, in the typical Australian Oligocheta. Associated with the external openings of these vasa deferentia are a pair of glands, the so-called ‘‘atria”’ or “ prostates.” The name “spermiducal glands,’ proposed for these glands by Beddard, is here retained as being the most appropriate. In addition, there are often present near the male pore other glands internally, or one or more “ penial setz.” These organs are in general constant throughout the group in many respects, but in detail they often vary widely in different species, both in structure and to a certain extent in position. A. Macroscopic CHARACTERS OF THE SPECIES. I. MEGASCOLIDES. 1. M. wwrermentvs * (Spencer). Only one pair of testes and of rosettes is present in this species. Sperm-sacs are found in segments 12 and 14. The spermiducal glands (sp.g., Pl. 14. fig. 3) are tubular in form and slightly coiled: they lie transversely to the body in segment 18, their ducts (sp.d.) opening on its external surface near the mid-ventral line. The duct of the spermiducal gland is almost straight and very muscular. ‘ The vas deferens on each side (v.d.) enters the gland-duct just where the latter leaves the gland. Occupying the space internally, between and covering the two spermiducal ducts, and below the nerve-cord, is a white mass of * Proc. Royal Society of Victoria, 1892, p. 182: Cryptodrilus intermedius (Spencer). Beddard, ‘ Oligocheeta,’ p. 492. GLANDS OF AUSTRALIAN EARTHWORMS. IDL glandular ceils, which may extend through one, two, or even four segments (a.g., fig. 3), or, on the other hand, may rarely be absent altogether. The surface of this glandular mass has a mottled brown appearance, due to small patches of pigment. Its presence in these segments probably depends on the state of maturity of the individual. 2. M. vicrorim * (Spencer). Two pairs of testes and rosettes are present in segments 10 and 11; the sperm-sacs are situated on the posterior wall of segment 9 and on the anterior wall of segment 12. The spermi- ducal glands (sp.g., fig. 4) are long, regularly-coiled tubes occupying segments 18 to 26, or even 29, one coil to each segment. The spermiducal duct is long and muscular, running transversely in a double loop towards the ventral surface in segment 18, where it opens. The two vasa deferentia of each side remain quite dis- tinct for their whole length. They run alongside the gland-duct for some distance, and pass straight through the substance of the lower part of the gland, entering the gland-duct just before it leaves the gland. 6 3. M. InsuLARis + (Spencer). Testes and rosettes in segments 10 and 11. Sperm-sacs in segment 9 (posteriorly) and segment 12 (an- teriorly). Spermiducal glands very long coiled tubes in segments 18 to 27; duct long, muscular, and but slightly curved. The two vasa deferentia on each side join before reaching segment 18, and the single vessel so formed runs in the muscular wall of the gland- duct, and so enters the duct about one-third of its length from the gland. 4. M. HOBARTENSIS { (Spencer). Two pairs of testes and rosettes in segments 10 and 11. Sperm-sacs in segments 9 (posteriorly) and 12 (anteriorly). The spermiducal glands are long and coiled, extending through segments 18 to 24 or 25. The muscular duct from each re- sembles closely that of WZ. insularis, being long and slightly curved. The vasa deferentia also unite on each side before they * Proc. Roy. Soc. Vict. 1892, p. 189: Cryptodrilus victorig, Spencer. Bed- dard, doc. cit. p. 488. t Ibid. 1894, p. 41: Cryptodrilus insularis, Spencer. t 4bid. 1894, p. 37: Cryptodrilus hobartensis, Spencer. OLE. MISS G. SWEET ON THE SPERMIDUCAL reach the level of the gland, the lower part of which they traverse, entering the spermiducal duct just as it leaves the gland. 5. M. arrenvatus * (Spencer). Testes and rosettes, two pairs in segments 10 and 11. Sperm-sacs in segments 9 (posteriorly) and 12 (anteriorly). Spermiducal glands are two tubular, slightly-coiled glands lying transversely in segments 18 and 19, each with a short straight muscular duct. The two vasa deferentia of each side remain separate, running up within the muscular wall of the spermiducal duct, and entering the duct as it leaves the gland. On each side, just behind the opening of the spermiducal duct, is a slightly muscular sac enclosing a long curved penial seta. As in M. intermedius, there is on the ventral surface internally, below the nerve-cord and between the two spermiducal glands, a great mass of unicellular glands, which may occupy one to three segments. 6. M. austraris tT (McCoy). In this form, which has been so completely described by Prof. Spencer, there are two, or occasionally three, pairs of testes and two pairs of rosettes. The sperm-sacs are found in segments 11, 12, 13, and 14, on the posterior wall in each case. The spermiducal glands are very much coiled, tubular struc- tures in segment 18; the duct of each is much coiled above and straight as it passes through the body-wall, receiving as it does so the two vasa deferentia. 7. M. roseus f (Spencer). Testes two pairs in segments 10 and 11 with rosettes. Sperm-sacs in segment 12 on the anterior wall. Spermidueal glands tubular and closely coiled, either confined to segment 18 or extending into the segments before and behind, 1.e. 17 and 19 (cf. fig. 5). The duct of the gland has strongly muscular walls, and is slightly curved. The vasa deferentia on each side do not unite, although they lie close side by side, deep down in the body-wall. They enter the spermiducal duct about halfway along its length. * Proc. Roy. Soc. Vict. 1892, p. 155: Megascolides attenuatus, Spent Beddard, Joc. cit. p. 491. +t McCoy, Prodromus Zoology of Victoria, Decade 1. Spencer, Trans. Roy. Soc. Vict. vol. i. pt. 1,.1888. Beddard, loc. cit. p. 495. ¢ Proc. Roy. Soc. Vict. 1892, p. 154. Beddard, oc. cit. p. 491. GLANDS OF AUSTRALIAN EARTHWORMS. 113 8. M. tupErcuLatus* (Fletcher). Two pairs of testes and rosettes in segments 10 and 11. Sperm-sacs on posterior wall of segment 9 and on the anterior wall of segment 12. Spermiducal glands tubular and very much coiled, lying trans- versely in segment 18. The gland-duct is long, and curved proximally into an S-shape ; it is embedded chiefly in the body- wall. The two vasa deferentia enter each gland-duct at the same point at about half the length of the latter. On each side, lying between the spermiducal duct and the mid-ventral line, are a pair of small, slightly muscular saes, each enclosing a long, somewhat robust penial seta, which pierces the body-wall close to and on the inner side of the male opening. IL. CryYpropRILts. 9. C. ILLAWARRE ft (Fletcher). One pair of testes and of rosettes are seen in segment 11: one pair of sperm-sacs also on the anterior wall of segment 12. The spermiducal glands are apparently somewhat long and straight flattened structures, lying transversely in segment 18. The gland-duct is short and straight, almost completely enclosed in the ventral body-wall, through which it runs vertically down- ward to open on the surface. The vasa deferentia are two in number on each side; not one, as might be inferred from the presence of only one pair of testes and rosettes. These run at varying levels, the anterior one in the muscle of the body-wall, the posterior one at its edge, until they reach the level of the posterior half of segment 13, when the upper one descends and runs alongside the lower one in the muscle-layer back to segment 18. They then join just outside the wall of the spermi- ducal duct, and the single one thus formed runs down in the substance of the wall to open into the duct close to its external opening. At the inner end of each gland-duct is a pair of short nurrow sacs, nacreous in appearance, from the muscle-fibres in their walls, and each containing a long curved penial seta (s., fig. 24). These sete are curiously hooked near their tips, and each opens independently of the others, through the body-wall, * Proc. Linn. Soc. N.S. W. 1887, p. 611: Notoscolex tuberculatus (Fletcher). Beddard, loc. ct. p. 494. + Ibid, vol. iii. 1888, p. 1528. Beddard, doe. cit. p. 503. 114 MISS G. SWEET ON THE SPERMIDUCAL on the inner side of, and sometimes slightly behind the male pores, and not through them as Mr. Fletcher has suggested. 10. CRYPTODRILUS COORANIENSIS * (Spencer). Two pairs of testes and of rosettes in segments 10 and 11. Sperm-sacs in segments 7, 8,9, and 12. Those in segments 7 and 8 appear to be sometimes absent in specimens otherwise similar. In segments 10 and 11 the testes are not visible as distinct structures from the so-called sperm-sacs, but are seen in section to be simply much denser parts of the ventral portion of these sacs, immediately in front of each of the rosettes, which are also enclosed within the sac, and surrounded by developing sperm (cf. fig. 15). These sacs in segments 10 and 11 thus appear to correspond to what Beddard has termed ‘“ sperm-reser- voirs.’ In segments 10 and 11 the sperm-reservoirs, and in segment 12 the sperm-sacs, form complete rings around the alimentary canal, those in segment 10 being much larger and less definite in shape than the others. The spermiducal glands are bilobed or flat, both varieties occurring in a single individual. Hach gland-duct is short and thick, and receives at half its length the two vasa deferentia which have distinct openings into it, on opposite sides of the duct. Ill. DreLotrema. 11. D. rragruis t (Spencer). Two pairs of testes and rosettes in segments 10 and 11. One pair of sperm-sacs on the anterior wall of segment 12. The spermiducal glands are long, coiled tubes extending through segments 15 to 25, and opening as usual in segment 18 (cf. fig.6). The posterior end of the gland is turned back through four or five segments. The gland-duct is somewhat less mus- cular than usual, and runs straight down to the surface, where it opens. The two vasa deferentia on each side remain distinct until they reach the level of segment 17, when they join and open to the exterior in segment 18 in front of and quite distinct from the spermiducal duct. On the internal surface of the ventral body-wall in segments 16 to 22, between the spermiducal * “Descriptions of Australian Harthworms,” Part 1. Proc. Roy. Soe. Vict. vol. xiii. pt. 1. t bid, vol. xiii. pt. 1. GLANDS OF AUSTRALIAN EARTHWORMS. 115 ducts, is a glandular mass divided off into groups transversely by the septa, and supporting the nerve-cord. On the inner side of each spermiducal duct there are also a pair of slightly muscular sacs, each enclosing a curved penial seta. These open quite independently of each other and of the gland-duct, on the inner or ventral side of the latter. LV. FLEercHERODRILUS. 12. F. untcus* (Fletcher). Two pairs of testes and rosettes in segments 10 and 11. Sperm-sacs in segments 9 and 10 (posteriorly) and 11, 12, and 13 (anteriorly). he gonads and rosettes appear distinct from the sperm-sacs in segments 10 and 11. The spermiducal glands (sp.q., fig.2) are long tubular structures, lying transversely in segment 18, with the distal end somewhat coiled. Towards the mid-ventral line the tube becames smaller and nacreous, marking the region of the duct, which then suddenly swells out, forming a well-marked muscular bulbus, straight through the centre of which runs the duct. Into this duct, at the level of the nerve-cord, there opens on each side a single vas deferens, which arises anterior by the union of the two ducts which leave the two rosettes. The ventral end of the sac passes beneath the nerve-cord and the longitudinal muscle- band, and meets its fellow of the opposite side in the mid-ventral line. Here they unite and form a strongly muscular duct, which runs vertically downward and slightly forward to open on a papilla in the mid-ventral line of segment 18. It is possible that this common part may be partially protrusible. V. DrrorocH#TA. 13. D. Copenannt + (Spencer). Two pairs of testes and rosettes in segments 10 and 11. Sperm-sacs in segments 9 (posteriorly) and 11 and 12 (an- teriorly). Spermiducal glands are long, coiled tubular structures, and may extend through segments 17 to 23, though sometimes less. * Proc. Linn. Soc. N.S. W. iii. (2) 1889, p. 1540: Cryptodrilus unicus, Fletcher. bid. iv. (2) 1890, p. 990: Cryptodrilus purpureus (Fletcher). Beddard, Joc. cit. p. 481. + Proce. Roy. Soc. Vict. 1892, p.3: Pericheta Copelandi (Spencer). Beddard, loc, cit. p. 442. 116 MISS G. SWEET ON THE SPERMIDUCAL The two vasa deferentia of each side remain quite distinct, though close together after segment 13. They enter the spermi- ducal duct, which is curved into a U-shape, near its external opening on segment 18. Close by, and on the inner side of each duct, and only visible under the microscope, are a pair of small and thin sacs, each containing two (or sometimes three) very fine, smooth, and slightly curved penial setz, which open into the termination of the spermiducal duct by definite tubes lined with columuar cells. 14. D. Baxurt * (Fletcher). Two pairs of testes and rosettes in segments 10 and 11. Sperm-sacs in segments 9 (posteriorly) and 12 (anteriorly). Spermiducal glands are tubular, somewhat coiled, and extend through four segments, 18 to 21 (fig. 1). The gland-duct is long and slightly curved, and very muscular. On the inner side of each duct is a pair of small sacs, each containing a penial seta surrounded by muscle-fibres. The division between the gland and the duct is more sharply marked than usual. The two vasa deferentia of each side join at the level of the prostate just before they enter the muscular wall of the duct, and so open into the duct itself at about half its length. Into the last portion of the spermiducal duct open by a pair of well-defined tubes the two penial sete, which are stouter than in D. Cope- land. VI. MrcaAscouex. 15. M. Frencutt (Spencer). Two pairs of testes and rosettes in segments 10 and 11. Sperm-sacs in segment 9 (posteriorly) and segment 12 (ante- riorly). The spermiducal glands are flattened and bilobed, and are situated in segments 18 and 19, one lobe in each (fig. 8). The whole surface of the gland is divided up into irregular or polyhedral areas, which, as will be seen later, correspond with certain points of internal structure. The septum separating segments 18 and 19 is continued between the two lobes of the gland, and is pierced by the branch of the duct on its way to the main gland-duct. The main gland-duct, which is in segment 18, * Proc. Linn. Soc. N. S. W. ii. 1887, p. 616: Pericheta Bakeri (Fletcher). Beddard, Joc. cit. p. 440. _ + Proc. RB. 8. Vict. 1892, p. 9: Pericheta Frenchi (Spencer). Beddard, /oc. cit. p. 379. GLANDS OF AUSTRALIAN EARTHWORMS. aly/ is formed by the union of two ducts, one from each lobe. The two vasa deferentia of each side unite in segment 13 to form one, which is larger in proportion than usual. This enters the spermiducal duct just at the union of its two main branches. 16. M. Firipert * (Spencer). A pair of rosettes,in segment 11 are seen to be enclosed within a large membranous sac, in which, however, there was no indica- tion of a testis. Sperm-sacs in segment 12 are mushroom-shaped, and attached to the anterior wall by a definite stalk. In two specimens examined each sac was found to be filled with Gregarines. The spermiducal glands are situated in segments 17 and 18; they are bilobed, one lobe in each segment (fig. 14). Occasionally a tendency to division into three lobes was seen in the larger and bifid character of the posterior lobe. The anterior is usually the smaller, and is often only united to the posterior one at the base by the branches of the gland-duct; while in other individuals again the gland is merely bifid, and confined to one segment. The surface of this gland, as in J. Frenchi, is divided up into numerous polygonal areas. The gland-duct is well marked, almost straight, and strongly muscular. As it passes down towards the ventral surface it curves slightly and enters a large hyaline muscular bulbus (fig. 14), within which the duct makes a double U-loop. The single vas deferens enters the muscular sac, and runs backwards and then alongside and behind the spermiducal duct, curving with it. The vas deferens joins the duct just after the latter has finally curved, to pass vertically downward to the external opening on segment 18. From its appearance, this last part of the spermiducal duct is probably protrusible. In one specimen, in which there was no prostate on one side, the vas deferens of that side simply ran straight to the surface in segment 18 without any increase of muscular fibres. 17. M. rasmantcoa f (Spencer). Two pairs of testes and rosettes in segments 10 and 11. The testes are large and finger-like. Sperm-sacs in segments 9 (posteriorly) and 12 (anteriorly). Spermiducal glands distinctly bilobed in segments 17 and 18; one lobe in each segment. In some individuals polygonal areas * Proc. R.S. Vict. 1892, p.19: Pericheta Fielderi (Spencer). Beddard, doc. + Ibid. 1894, p. 47: Pericheta tasmanica (Spencer). 118 MISS G. SWEET ON THE SPERMIDUCAL were visible on the surface. The spermiducal duct is large though not always very long, and is muscular. It receives one main branch from each lobe, the main duct lying in and opening on segment 18. The vasa deferentia, two on each side, are very large in proportion to the size of the animal. They run close side by side just embedded in the body-wall until, in segment 12, those on the right side unite, while those on the left do not unite till seement 14. They enter the spermiducal duct close to the union of the branches from the two lobes. 18. M. Hoeerr* (Spencer). Two pairs of testes and rosettes in segments 10 and 11. Sperm-sacs in usual positions in segments 9 and 12. Spermiducal gland in segment 18, bilobed and small ; the surface appears smooth, without any markings. The duct is straight, or may be slightly curved, and is strongly muscular. The vasa deferentia, two on each side, enter the spermiducal duct close side by side, at about half its length. 19. M. Denpyi t (Spencer). Two pairs of testes and rosettes in segments 10 and 11. Sperm-sacs in segments 9 (posteriorly) and 12 (anteriorly). Spermiducal glands in segments 18 and 19, may be flattened with but the smallest indication of a second lobe, or they may be clearly divided into two lobes (fig. 9). The surface of the gland is sometimes very much grooved, though notin any regular fashion, and having no relation to internal structure. The duct is single, i. e. it does not branch outside the gland-substance; it is long and curved into almost an S-shape in some cases, and at others forming three or four loops. The two vasa deferentia of each side remain quite distinct and enter the spermiducal duct side by side, just where it leaves the gland. 20. M. porsatis t (Fletcher). Two pairs of testes and rosettes in segments 10 and 11, both testes and rosettes being very large. * Proc. B.S. Vict. 1892, p.6: Pericheta Hoggii (Spencer). Beddard, loc. cit, p- 379. + Ibid. p.12: Pericheta Dendyi (Spencer). \ Beddard, Joc. cit. p. 380. + Proc. Linn, Soc. N.S. W. vol. ii, 1887, p. 618: Pericheta dorsalis (Fletcher). Beddard, doc. ci¢. p. 376. GLANDS OF AUSTRALIAN EARTIWORMS. 119 Sperm-sacs are conspicuous globular bodies in segments 9 (posteriorly) and 11 and 12 (anteriorly). The spermiducal glands vary in position, and to a certain extent in shape. In one individual each was distinctly tubular, occupying segments 18 and 19, the wider portion in segment 19 being coiled. In this also the duct was very short, and com- pletely embedded in the body-wall. In a second individual, the gland was much wider in proportion to its length, and appeared at first as a single flat elongated body in segments 18, 19, and 20, the duct forming a double-U curve. In the third speci- men (fig. 10) the gland appears as a wide flat tube in segments 18, 19, 20 and 21; slightly curved in the first and last segments. The duct in this case was very short. The surfaces of the glands in the first and second individuals were quite smooth, apart from erooves incidental to curving. But in the third individual, the gland clearly showed an irregular division of the surface into areas exactly similar to those on the gland of MW. Frenchi. The two vasa deferentia of each side enter the gland-duct on - opposite sides at half its length from the exterior. The external opening is situated on a very stronyly-marked papilla. In this papilla are situated two distinct sets of glands connected with the external openings. 21. M. nusra* (Spencer). Testes and rosettes, two pairs in segments 10 and 11. Sperm-sacs in segments 9 (posteriorly) and 12 and 14 (anteriorly). Spermiducal glands bilobed, lying entirely in segment 18, the posterior wall of which is pushed back. The spermiducal duct is long and coils slightly as it leaves the gland. The two vasa deferentia of each side unite about segment 13 to form one on each side, which runs up the wall of the spermiducal duct and enters it just inside the gland. 22. M. minor fT (Spencer). Segments 10 and 11 are occupied by what appear to be large sperm-sacs filled with masses of developing sperm. At the base of those near the nerve-cord on each side is a small spherical, * Proc. R. 8. Vict. 1892, p. 8: Pericheta rubra (Spencer). Beddard, /oc. cit. p. 379. t “Descriptions of Australian Earthworms,” Pt.1. Proc. R. 8. Vict. vol. xiii. pt. 1. 120 MISS G. SWEET ON THE SPERMIDUCAL denser and more compact portion, which probably represents the true testis. The rosettes, though covered with sperm, were not seen to be included in the membrane enclosing the testes—though in other respects these agree with the sperm-reservoirs of Beddard. The true sperm-sacs are very compact saccular structures in segments 9 (posteriorly) and 12 (anteriorly), and sometimes in 13 (anteriorly), those in segment 12 being the least saccular of the three. The spermiducal glands are bilobed, or sometimes merely flat, and bifid structures in segment 18, the anterior wall of which becomes somewhat pushed out of position. The duct is smail and S-shaped, and less muscular than usual. The vasa deferentia unite in segment 12 to form one on each side, which enters the spermiducal duct as it leaves the gland. 23. M. iInrErRMeEptIus * (Spencer). As in JZ minor, though even to a more marked extent, the sperm-reservoirs are developed one on each side in segments 10 and 11, enclosing masses of developing sperm. The testes and rosettes, the latter of which are distinctly ‘‘ Maltese-cross ” in shape, are enclosed in the same membrane. In both segments each reservoir (cf. fig. 15)is divided off into two parts, the larger ventral part enclosing testis and rosette. The septum between segments 8 and 9 is lost, being merely represented by a few fibrous cords. Occasionally also that between segments 9 and 10 is absent, in which case the sperm-reservoir of segment 10 is lost, the rosette alone being present. The sperm-sac usually found on the posterior wall of segment 9 is absent; a pair is present in segment 12 (anteriorly). The spermiducal gland appears to be generally bilobed, with a strongly-marked U-shaped duct, the whole lying in segment 18. In one individual the gland on one side was flattened and irregular in shape, showing only a slight division into two parts. The duct was coiled, and the two vasa deferentia of that side opened into the duct a little way from the gland. On the other side the gland was absent, and the two vasa deferentia united in segment 18 and formed, with the addition of a thick wall of muscle-fibres, a duct corresponding in size and position with that on the other side. * “ Descriptions of Australian Harthworms,” Pt.1. Proc. R. S. Vict. yol. xiii. pt. 1. GLANDS OF AUSTRALIAN EARTHWORMS. 121 24. M. Inurpe@nr* (Spencer). Two pairs of testes and rosettes as usual in segments 10 and 11. Sperm-sacs in segments 9 (posteriorly) and 12 (anteriorly). Spermiducal glands are bilobed and flattened, and wrap round the alimentary canal; their ducts are short but S-shaped. The two vasa deferentia of each side unite only as they reach the level of the spermiducal duct. The single vessel thus formed enters the wall of the duct, and runs up in it, uniting with the gland-duct just as it emerges from the gland. VII. DieastEer. 25. D. armirera fT (Fletcher). Two pairs of testes and rosettes in segments 10 and 11. Sperm-sacs in segments 9 (posteriorly) and 12 (anteriorly). In one individual the whole of the segments 10 and 11 was filled with masses of developing sperm in which the rosettes were embedded. No definite membrane enclosed the mays, nor could any special representative of a testis be found, even in section ; but it may probably be regarded as a sperm-reservoir. The size of the true sperm-sac in segment 12 is quite abnormal, and has almost completely closed up segment 13 by the pushing back of its anterior wall. The blood-vessels in this segment were also unusually large. In one specimen sperm-sacs were also present in segments 14 and 15 (anteriorly). The spermiducal glands are present in segments 17, 18, and 19, or sometimes only in segment 18, which is then much extended. The giands are lobate, usually with three lobes more or less divided off. The whole surface is distinctly marked off into similar irre- gular areas to those seen in Megascolex Frenchi. The gland-duct is comparatively large and long, and is slightly curved (fig. 13). The two vasa deferentia of each side run very close together in the body-wall, and enter the spermiducal duct just after it leaves the gland. Lying posterior to the gland-duct are two narrow sacs with slightly muscular walls. Each of these contains two well-marked penial setze in a well-marked sheath. Lach is curved and long, and is ornamented near the tip with numerous little short spines, as described by Mr. Fletcher, ending in a chelate point. The * “Descriptions of Australian Harthworms.” Proc. R. 8. Vic. vol. xii. pt. 1. + Proc. Linn. Soc. N.S. W. i. (2) 1887, p. 947. Beddard, (oe. ci. p. 486. LINN. JOURN.—ZOOLOGY, VOL. XXVIII. 9 122 MISS G. SWEET ON THE SPERMIDUCAL sete reach the surface through a small duct opening into the spermiducal duct just before that duct reaches the external opening. 26. D. GAYNDAHENSIS * (Spencer). Two pairs of testes and rosettes in segments 10 and 11. Sperm-sacs large in segments 9 (posteriorly) and 11 and 12 (anteriorly). Spermiducal glands in segment 18 are small and flat. The duct is small and short, running straight down to the external opening on 18. The two vasa deferentia on one side join to form one duct near the gland. Those on the other side apparently remain distinct. In each case they traverse the substance of the spermiducal gland and run down inside the wall of the duct, with which they unite halfway to the exterior. 27. D. minor tT (Spencer). Two pairs of testes and rosettes in segments 10 and 11. Sperm-sacs in segments 9 (posteriorly) and 12 (anteriorly)— in one specimen in segments 9 and 11. In segments 10 and 11 in the first specimen, the testes and rosettes were enclosed in true sperm-reservoirs. " The spermiducal gland is completely bilobed and not flattened : ’ the duct is formed by the union of a branch from each half. The vasa deferentia, two on each side, having remained separate through their whole length, pass straight through the substance of the lower part of the anterior lobe of the spermiducal gland before entering the duct of that branch just before it unites with its fellow from the posterior lobe. 28. D. BrunNEUS { (Spencer). One pair of testes in segment 10, and two pairs of rosettes in segments 10 and 11. The anterior rosettes are very small indeed. Sperm-sacs in segment 12 (anteriorly). Spermiducal glands larger than in D. gayndahensis, flattened and slightly bilobed. The duct is small and less muscular than usual, and runs straight into the body-wall. The two vasa * ‘Descriptions of Australian Harthworms,” Proc. R. 8. Vict. vol. xiii. pt. 1. + Ibid. pt. 1, Proc. R. 8. Vict. vol. xiii. pt. 1. } Ibid. pt. 1, Proc. R. §, Vict. vol. xiii, pt. 1. GLANDS OF AUSTRALIAN EARTHWORMS. 123 deferentia of each side remain quite distinct, and enter the spermiducal duct as it leaves the gland. 29. D. syivarica * (Fletcher), Two pairs of testes and rosettes in segments 10 and 1i. Sperm-sacs are racemose structures in segments 9 (posteriorly ) and 12 (anteriorly). This differs considerably from the original description, where the structures in segments 9 and 12 are called testes and described as “smooth and white bodies drawn out above into digitate processes tapering into fine points,’ while no sperm-sacs were mentioned whatever. The organs in segments 9 and 12, here called sperm-sacs, and which have both the position and appearance of those structures in other forms, are conspicuous, white, and greatly mammillated. Also there are distinct digitate though small testes in their normal positions on the anterior wall of segments 10 and 11. Spermiducal glands are bilobed and large. The duct is short and muscular, and embedded in the body-wall. The two vasa deferentia of each side enter the spermiducal duct as it leaves the gland. 30. D. excavata } (Fletcher). In one specimen, testes and rosettes were present in segments 10 and 11. The sperm-sacs were grape-like, in segments 9 (posteriorly) and 12 (anteriorly). In this the spermiducal gland consisted of a long tubular structure, the anterior part lying in segment 18, considerably larger and flatter than the posterior part which extended back in segments 18 and 19. The duct leaves the gland at the beginning of the tube and is short, running down obliquely to open on the surtace. The duct is less muscular than usual, and into it at the base of the gland opens the single vas deferens formed by the union on each side of the two vasa deferentia. On the inner side of the duct and opening into its lower portion are two slightly muscular seta-sacs, each containing one long seta. In a second specimen there was no appearance of testes as distinct from the sperm-reservoirs, which occupied segments 10 * Proc. Linn. Soc. N.S.W. vol. i. (2) p. 558: Didymogaster sylvatica (Fletcher). Beddard, Joc. ct. p. 487. + Ibid. vol. ii. (2) p. 383: Perissogaster excavata (Fletcher), Beddard, loc. cit. p. 486. Q* 124 MISS G. SWEET ON THE SPERMIDUCAL and 11 and enclosed the rosettes. The spermiducal gland was of a constant width throughout. A third individual resembled the second in every respect, except that on one side the spermiducal gland was turned for- ward, lying in segment 18 entirely. As will be seen later, this differs somewhat from the original description. Around the spermiducal duct the body-wall is much thickened owing to the presence in it of groups of long unicellular glands, whose ducts open in groups on the surface around the male opening. 31. D. qurEnstanpica * (Fletcher). Two pairs of testes and rosettes in segments 10 and 11. Sperm-sacs very small in segments 11 and 12 (anteriorly). The spermiducal gland is flat in some specimens and well-developed (Pl. 14. fig. 11), while in others it appears to be either very small and tubular (fig. 12), or absent altogether, as in six specimens of thoseexamined. In one of these specimens neither sperm-sacs nor testes could be seen, probably pointing to its immature condition. The two vasa deferentia, which along their length are separated by a short distance, enter the gland-duct quite separately at about half its length. VIII. AcanrHopRILus. 32. A. SYDNEYENSIS (Fletcher). One pair of rosettes in segment 11. Masses of sperm in segments 11 and 12, the first probably representing a sperm- reservoir, and the second sperm-sacs. Nothing comparable to a testis could be seen. The single vas deferens of each side runs back along the edge of the body-wall, and opens mT of the spermiducal glands on segment 18. The spermiducal glands (sp.g., Pl. 14. fig. 7) are two pairs of long, coiled tubular structures, one pair opening to the surface in segment 17, and the other in segment 19. Both extend back as far as segment 26. The muscular duct of each gland extends from its: external opening back through about 13 segments, and then passes into the second region which extends through another 33 segments, while the most distal glandular region is found in the last three of these segments. The glands lie near the nerve-cord on each side, passing through the ventral opening in the septa with the nerve-cord and ventral blood-vessel. * Proc. Linn. Soc. N.S. W. vol. iii. (2) p. 1529: Perissogaster queenslandica (Fletcher). Beddard, /oc. cit. p. 485. GLANDS OF AUSTRALIAN EARTHWORMS. 125 B. Generat Macroscortc Fraturss. I. Testes and Sperm-sacs. The position of the testes in segments 10 and 11 appears mearly constant throughout the Australian genera. Occasionally we find that there is but one pair of testes, in which case it iy situated in segment 11, e.g. Cryptodrilus illawarre and Megasco- lides intermedius, or rarely in segment 10 as in Digaster brunneus. Rarely less than two pairs of rosettes are present, e. g. Megascolex Fieldert, Cryptodrilus illawarre, and Megascolides intermedius ; though the anterior pair may be very small (e.g. in Digaster brunneus). Very often, as in Megascolex intermedius, I. minor, Digaster minor, and Oryptodrilus cooraniensis, the testes or rosettes, or both, are enclosed within sperm-reservoirs. There is, when present, one of these structures on each side in contrast to the median position of the typical “ sperm-reservoirs”’ to which the name was originally given by Beddard. Hach reservoir may or may not be divided by a connective-tissue septum into two or more chambers (cf.s.7. Pl. 14. fig. 15), the ventral of which is larger and contains the rosette (7.), and also the denser, more compact, and more darkly-staining portion, which represents the testis in such cases except in Digaster armifera, where it is absent. The rosette is, rarely, not enclosed in this sac, when it is present, as in Megascolex minor; but since it nevertheless contains the “testis, it may be regarded as answering to Beddard’s definition of a ‘“‘sperm-reservoir” as a sac enclosing the gonads and the rosettes. The true sperm-sacs are usually found on the posterior wall of segment 9, and on the anterior wall of segment 12. In addition to this, there is sometimes a pair of true sperm-sacs in segment 11, e.9., in Digaster gayndahensis, Diporocheta Cope- landi, and Megascolex dorsalis. Sometimes that in segment 12 only may be present, as in Digaster brunneus, Megascolex Fieldert, Cryptodrilus illawarre, Diplotrema fragilis, and Megas- colides roseus. In other forms, that in segment 9 may be absent, and present in segments 11 and 12, e. g. Digaster queenslandica. In Megascolex rubra they are found in segments 9,10, and 14; in Megascolides australis in 11, 12, 18, and 14; in Megascolides intermedius in 12 and 14; in Digaster armifera in 9, 12, 14, and 15; and in Fletcherodrilus unicus in segments 9,10, 11,12, and 13. Those occurring in segments 9 and 10 126 MISS G. SWEET ON THE SPERMIDUCAL are attached to the posterior wall of the segment, and those im segments behind 10 to the anterior wall of the segment. These sperm-sacs vary in form as well as in position. Sometimes they form definitely shaped masses enclosed in a firm membrane. In this case, they are simply lobed or mammillated, or form a ring around the alimentary canal. At other times they are much looser, racemose structures, when the investing membrane is easily broken. Again, the whole cavity of the segment may be completely filled by a mass of seminal cells, obliterating the cavity of the succeeding segments by backward pushing of the septum. In several instances Beddard has cast doubt on the character of the sacs in segments 13 and 14 when so existing, preferring to regard them as egg-sacs rather than sperm-sacs as: described. In no species which I have examined has there been any sign of ovain such a position, these sacs invariably containing developing sperm. II. Spermiducal Glands. These are usually found in segment 18 and sometimes in contiguous segments, only very rarely extending in front of segment 18. The two main types of external form are tubular and lobate (including flat and undivided forms). (1) Tubular Glands.—These lie either transversely to the body, when they are generally confined to segment 18; but in some species they may, when mature, extend as far back as. segment 26. They are usually slightly coiled, but in some cases, e. g., Megascolides roseus and M. australis, they are twisted ixto: an almost inextricable mass, lying in segment 18. They are characteristic of the following genera:—Megascolides, Orypto- drilus (partly), Fletcherodrilus, Diporocheta, and Acanthodrilus.. With regard to Digaster, it will be noted that the two species. D. excavata and D. queenslandica are somewhat different from the remainder of the genus, more especially the former species. They have been described as bilobed, with a flattened lobe: anteriorly and a long tubular portion posteriorly. In the specimens of D. excavata examined by me only tubular “ pro- states” have been seen, though in one individual a band of connective-tissue in the position described for the gland-duct might possibly be mistaken for a duct. It, however, does not contain any duct whatever, the true gland-duct leaving the GLANDS OF AUSTRALIAN EARTHWORMS, 127 front end of the tubular gland. In D. queenslandica, as before stated, the majority of specimens have no spermidueal gland ; in others it is a small, shortened, flattened tube (fig. 12), while yet another has a large well-developed flattened single lobe (fig. 11). In Cryptodrilus illawarre the gland forms, as will be seen later, a transitional stage between such a tubular gland as that of Dipio- trema fragilis and the second group of forms. In Acanthodrilus sydneyensis, as is typical of the genus, there are two pairs of tubular prostates lying longitudinally. The first pair open on segment 17, and the second pair on segment 19, and both run backwards, coiling somewhat as they go, as far as segment 26. (2) The Lobate Glands, including single-lobed, bilobed, and trilobed forms, are characteristic of the remaining genera, viz., Megascolex Digaster (excluding D. queenslandica and D. exea- vata), and Cryptodrilus. Among these, the external form of the gland in Megascolex dorsalis would suggest the possibility of its being an inter- mediate condition between the tubular and lobate forms, as also that of Cryptodrilus illawarre (vide infra). These glands usually occupy but one segment, that in which they open, 18; but often the posterior lobe is found in segment 19 separated from the anterior lobe by the septum between these two segments. III. Ducts (Vasa Deferentia and Spermiducal Ducts). There is a certain amount of variation in the arrangement of each of these ducts, and in their relations one to the other. (1) Lhe Spermiducal Duct.—The main part of the duct, which is usually well supplied with circular muscles, both in the lobate and tubular forms, may be—(1) short and straight or but slightly curved: e. g., Megascolex tasmanica, M. Hoggit, Diporo- cheta Copeland, Cryptodrilus cooraniensis, Megascolides inter- medius, MI. attenuatus, Digaster minor; occasionally being confined entirely within the body-wall, e. g., Digaster brunneus, D. gayndahensis from Queensland, and Cryptodrilus illawarre. (2) It may be long and more or less curved, sometimes but slightly, as in Megascolides roseus, M. victorie, M. insularis, M. hobartensis, Megascolex rubra, and Diporocheta Bakeri ; at other times it is much longer, forming a U-shaped loop, e. g. Megascolex intermedius, or, again, forming a double loop as in Megascolex dorsalis and M. Dendyi, &c. (8) Inthe lobate forms 128 MISS G. SWEET ON THE SPERMIDUCAL the main duct may either (a) leave the substance of the gland before the branches from the separate lobes have united, as in Megascolex Frenchi; or (b) it may leave the gland as a single duct, as in the great majority of cases, e. g., Megascolexr Dendyt, M. tasmanica, or Digaster armifera. Another variation connected with the spermiducal duct is in the presence of a muscular swelling or bulbus on each side in Megascolex Fielderi and Fletcherodrilus unicus. In the latter it partakes more of the nature of a swelling caused by an enormous and somewhat sudden thickening of the muscular layer of the duct-wall, straight through which the duct runs on its way to join its fellow of the opposite side, and open to the surface. In Megascolex Fielder, however, it is different. The bulbus is almost spherical (m.0., Pl. 14. fig. 14), and similar in size to that in F. unicus. The duct is sharply bent just before it enters the side of the bulbus, which it does obliquely. Within the bulbus the duct makes a complete coil, the vas deferens accompanying it until it opens into the duct a short distance from the external opening. (2) Vasa deferentia.—In accordance with the usual presence of two pairs of rosettes, there are generally two pairs of vasa deferentia. Even when the second pair of rosettes is absent, the two vasa deferentia are usually present on each side. Rarely there may be only one pair of vasa deferentia, e. g., Megascolides intermedius and DMegascolex Fieldert. They may join when there are two on each side as soon as possible after the 11th segment, z.€. in the 12th, e. g., Megascolex tasmanica, M. minor; or they may unite in the 13th or 14th segments, not always at the same level on the two sides, e.g., Fletcherodrilus unicus, Megascolex Frenchi, M. rubra. They may join just before they reach the spermiducal duct, as in Megascolides insularis, M. hobartensis, Cryptodrilus illawarre, Diplotrema fragilis, Diporocheta Bakert, Megascolex Illidget, Digaster gayndahensis, D. excavata; or they may join as they enter the duct, e. g., Megascolides attenuatus, MM. australis, M. tuberculatus, Dipcrocheta Copelandi, Megascolex Hoggit, M. Dendyt, I. intermedius, Digaster armifera, D. syl- vatica, D. minor, D. brunneus ; or they may open into the duct quite independently of each other, e. g., on opposite sides of the duct—such are Megascolides victorie, M. roseus, Cryptodrilus cooraniensis, Megascolex dorsalis, Digaster queenslandica. Variations in the relations of the spermiducal duct to the vasa GLANDS OF AUSTRALIAN EARTHWORMS. 129 deferentia fall into four groups. (1) Those forms in which they unite just where the gland-duct leaves the gland, e. g., Megascolew tasmanica, M. Dendyi, M. rubra, M. minor, M. intermedius, M. Illidgei, Megascolides intermedius, M. attenuatus, Digaster armi- Sera, D. excavata, D. minor, D. brunneus, and D. sylvatica. Sometimes the vasa deferentia traverse the substance of the lower part of the spermiducal gland before they join the gland-tube, e.g., in Digaster minor, Megascolides victorie, MW. hobartensis, and Megascolew rubra. (2) Those in which the vasa deferentia enter the gland-duct at some point along the length of the latter: e. g., Megascolex Frenchi, M. Fielderi (3), M. dorsalis (3), WM. tasmanica, M. Hoggii, M. Dendyi (4), Diporo- cheta Copelandi (=), D. Bakeri (), Fletcherodrilus unicus (2), Megascolides insularis (3), M. australis, M. roseus (3), M. tuber- culatus, and Digaster queenslandica, D. gayndahensis, and Cryptodrilus cooraniensis. (8) Those forms in which they join just before the external opening, or they may even open together on the surface, e. g., Cryptodrilus iliawarre, and sometimes Diporocheta Copelandi. (4) Those in which the vasa deferentia open quite independently of the spermiducal openings, e. g. Diplotrema fragilis—in front, but on the same segment ; and in Acanthodrilus sydneyensis, on different segments. It will be seen from the examples given in the second group, that there are several exceptions among these Australian forms to the statement made by Beddard that the vasa deferentia enter the gland-duct at its commencement in the Perichetide. Associated with these ducts are to be seen, in certain genera and species, penial sete. These are usually situated two on the inner side of each spermiducal duct. They are enclosed in a sheath within slightly muscular sacs—usually each sac only contains one seta, ¢.g., in Megascolides attenuwatus (cne on each side of the body), WZ. tuberculatus, Cryptodrilus illawarre, Di- porocheta Bakeri; but sometimes there may be two or even more in each sac, ¢.g., in Diporocheta Copelandi and Digaster armifera. These sete may open by definite ducts into the cavity of the spermiducal duct and so reach the surface, e.g. in Dipo- rocheta Bakeri, D. Copelandi, and Digaster armifera. At other times they simply protrude straight through the body-wall, in- dependently of the duct, e. g., Diplotrema fragilis, Megascolides tuberculatus, Oryptodrilus illawarre. They are larger than usual in Megascolides tuberculatus, Diporocheta Bakeri, and Digaster 130 MISS G@ SWEET ON THE SPERMIDUCAL armifera, and thinner than usual in Diporocheta Copelandi. In all, except Cryptodrilus illawarre and Digaster armifera, they are smooth and slightly curved. In C. dlawarre (P1.15. fig. 24) they are curiously bent back or recurved near the tip; while in Digaster armifera (as already described by Mr. Fletcher) the tip is beset with five spines, and ends in a double chelate joint C. Histotnoey. I. Spermiducal Glands. Typically each gland is a hollow structure, the body-wall being composed of long glandular cells, with a columnar cell-lining and covered by peritoneal membrane. Though there are considerable differences in the microscopic structure of the spermiducal gland in various forms, yet there is a certain amount of constancy in the relations of the external and internal structure of the gland. Thus the variations in its histology fall chiefly into two main groups, corresponding (1) to the tubular forms, and (2) to the lobate forms. 1. The Tubular Forms (P1. 14. fig. 16).—Here there is a central duct running the whole length of the gland, and directly continuous to the exterior. This duct is lined throughout by epithelium generally of ordinary columnar cells (¢.e.), with a nucleus at the inner end of each (fig. 17), or sometimes of cubical cells (fig. 16). The imner ends of the cells are usually more or less horny, while the nuclei are usually clearly visible even when the boundaries of the cells are not so (fig. 17). In Megascolides australis, however, as previously described by Professor Spencer, these lining-cells are more granular in appearance, staining deeply and rarely showing any nuclei. Outside this lining is often a small amount of connective-tissue with small blood-vessels. The outermost layer of the wall of the duct making up the bulk of the gland is composed chiefly of coarsely granular gland-ceils (g.c., Pl. 14. fig. 16 & Pl. 15. fig. 17). Each of these has a pear-shaped outer end, containing a nucleus often pushed to one side: it has a long narrow duct-like inner end which opens into the lumen of the gland, between the cells of the columnar lining—sometimes singly, e.g., in Megascolides roseus, M. australis, M. attenuatus, M. insularis, M. hobartensis, and Diplotrema fragilis; and sometimes in groups, especially where the lining is at all horny, e.g. in Megascolides intermedius (fig. 17), GLANDS OF AUSTRALIAN EARTHWORMS. 131 M. victoria, M. tuberculatus, Digaster excavata, Fletcherodrilus: unicus, Diporocheta Copelandi, and D. Bakeri (occasionally in the latter they appear singly). In some forms there are a limited number of cells which are generally scattered and are similar to the gland-cells around them, except that they are more granular and stain very much more readily with methylene-blue than the others, and are there- fore very conspicuous, e. g. in Megascolides victorie and Fletchero- drilus unicus, in both of which cases they are very numerous; also in Digaster armifera and Diporocheta Bakert. At other times similarly staiming cells form a more or less distinct layer at about half the thickness of the wall of the gland. These cells have a more rounded club-shaped end than the ordinary eland-cells, and are very well-marked in Megascolides insularis (P1. 14. fig. 16) and M. hobartensis. The blood-vessels in these forms are some- times found in the connective-tissue immediately around the columnar-cell lining; but often the main blood-vessel runs down the side of the gland, branching there and sending in fine capillaries amongst the glandular cells (0.v., Pl. 15. fig. 17). The forms Cryptodrilus illawarre, Megascolex dorsalis, and Digaster excavata will be described with the lobate forms. In Acanthodrilus sydneyensis (Pl. 15. fig. 18) the structure of the three parts of each tube is as follows :—The proximal part is the gland-duct (fig. 18 a), which as usual is lined by columnar cells, the cell-outlines being rarely seen while the nucleus is large and distinct. The muscle-layer forming the thickness of the wall (¢.m.’) is composed, as in other forms, chiefly of circular fibres. The next part forms an intermediate region between the duct and the glandular part proper. The lining of non-glandular epithelial cells is continued back from the duct (fig. 186). The glandular cells (.c.) are quite distinct in appearance from those previously described. They are almost columnar in shape, with an inner slightly rounded end and a nucleus near their outer end, surrounded by the granular cell-contents. This is succeeded by the third and most distal part (fig. 18 c)—the most glandular. This is most striking in appearance, owing to the rapidity with which the cell-contents stain in methylene-blue, as may be seen on reference to the figure, where it is indicated by the depth of the shading. The facts show that the deeply- staining material (mucous or other secretion) is situated variously at the inner or outer end of the cell; and it may be 132 MISS G. SWEET ON THE SPERMIDUCAL inferred that in the latter case it has not yet made its way to the cavity of the gland; while in the former it is just about to be passed out into that cavity, as the outer part of the cell in these cases is very finely granular, and does not take the methylene- blue at allreadily. The cells are much larger than those seen in other forms, and, by the irregularity in shape of their inner ends, they make the cavity of the gland in this region very uneven, though of greater calibre than the intermediate portion. The blood-vessel runs up alongside the wall of the gland and branches on the surface. It will be found that this differs considerably from the description given by Beddard of the spermiducal gland of Acanthodrilus among other genera. Acanthodrilus is described by him as being similar to other genera of Megasco- licide, such as those having tubular spermiducal glands, z. e. the Cryptodrilide and Perichetide, and all but one of the Acanthodrilide. That this is not so in this species is evident. In the first place, the cells of the columnar lining, even in the intermediate portion, are not granular, the nucleus only being generally visible, and indeed this is by no means a constant character in many of the genera, as before described. There is further no columnar-cell lining at all in the glandular part proper of this Acanthodrilid. Again, the glandular cells here are quite distinct in shape and appearance from the pear-shaped cells with long duct-like inner ends, found so constantly in such genera as Megascolides, Diporocheta, and the ordinary Acantho- drilids. They are here moreover only one layer thick. Even in the intermediate portion the gland-cells are more like those in the glandular part proper, than like those found in the genera above named. 2. The Lobate Forms.—These show in microscopic section a distinct and characteristic lobular arrangement, typically without any central tube. The lobules are more or less well-defined accordingly as the whole gland is less or more compact, since in the more closely-packed glands the cells of contiguous lobules have a tendency to overlap, thus masking to a certain extent the radiate appearance so conspicuous in the looser forms. The lobules, especially in the latter case, are separated by a small amount of connective-tissue (P1].15. fig. 19). In this there run the blood-vessels of the gland (6.v.), which are often very numerous and definite, e.g. in Digaster brunneus, and at other times have more the nature of lacune; the very fine terminal branches of the GLANDS OF AUSTRALIAN EARTHWORMS. 133 gland-duct are also to be found in this connective-tissue. The latter are lined by a well-marked cubical epithelium surrounded by a few muscle-fibres. Hach lobule has typically a radiate appearance, with sometimes a small lacuna or space (J.1.) in its centre, e.g. in Megascolex Dendyi, M. tasmanica, and M. dorsalis, which is however generally obliterated by the approximation of the inner ends of the cells. In very rare cases, a communi- cation can be traced between it and one of the terminal branches of the gland-duct (as in fig. 20). In no instance was an epithelium of any kind to be seen around the central lacuna. The cells forming the substance of the lobules are of three kinds :—(i) The chief are those having the appearance of ordinary long narrow gland-cells with swollen outer ends, which contain large darkly- staining granules with generally a nucleus. (ii) The second group of cells are a number which appear to be quite empty and rarely showing a nucleus: they are probably simply cells which have discharged their contents into the gland-duct. (iii) Here again in some forms, e.g. Megascolex Hrenchi and Digaster queenslandica, we find the cells seen in the tubular forms which stain readily with methylene-blue, and which serve here to accentuate the radiate appearance of the lobules, especially in D. queenslandica. The inner ends of the first and second sets, and of the third when present, come together in the centre of the lobule and evidently pour their contents into the gland- duct by means of the lacuna and duct before mentioned. As before stated, the three species Cryptodrilus illawarre, Megascolex dorsalis, and Digaster excavata, are intermediate forms between the tubular and lobate glands. In the two former there is a well-marked central duct running the whole length of the gland (fig. 24), and lined by columnar cells with a cuticular edge as usual in tubular forms. The main mass of the gland, however, is made up of lobules sometimes indistinctly separated from one another. In the centre of each lobule is a lacuna, which is connected with the central duct by a complex series of branches. Blood-sinuses run between the lobules. In Digaster excavata the branching is very much less complicated. There is but one series of branches from the central tube to form lobules, which are sometimes but seven or eight in number in the circumference of the gland. The central duct with its columnar lining is very conspicuous. It will be noted that this imter- mediate character agrees entirely in the first two cases with 134 MISS G. SWEET ON THE SPERMIDUCAL what might have been expected from the external form, for it is often difficult to know to which group to refer the glands in these species by merely examining the outside of the gland. Il. Accessory Glands. (1) Glands found within the cceelom, as in Megascolides inter- medius, M. attenuatus, and Diplotrema fragilis. These include two distinct types. In Megascolides intermedius and M. attenu- atus the glandular masses occupy the internal surface of the ventral body-wall between the spermiducal ducts. In the former, M. intermedius (Pi. 14. fig. 3), these structures (a.g.) vary in the number of segments they occupy, e. g. from two to four segments, or they may even be absent altogether. In IL. attenuatus they were found to occupy three segments—17, 18,19. They consist (Pl. 15. fig. 21) of a mass of long unicellular glands covered by the thin peritoneal membrane ; the gland-cells have pear-shaped outer ends, with granular contents and distinct nuclei. The long narrow duct-like inner ends of these cells (g."¢.") run down through the longitudinal and circular layers, and open on the ventral surface between the columnar cells of the epidermis. In MM. attenuatus the gland-cells at the sides of the mass are much (sometimes two or three times) longer than those in the centre, otherwise they resemble those in M. intermedius. In each case the nerve-cord runs above, and is supported by the glandular mass. In M. intermedius (fig. 3) the surface is seen to be mottled, owing to the presence of small patches of brown pigment. Diplotrema fragilis (P1.15. fig. 22).—Here the gland-mass takes up the ventral surface of segments 16 to 20 or 21. It is divided up transversely by the septa which traverse it. In segment 18 the spermiducal duct runs down alongside this accessory gland- mass to open on the surface. In microscopic structure the gland-mass is very different from that in Megascolides inter- medius. It is composed of a number of groups, a central one on which rests the nerve-cord, and two others on either side of this (cf. fig.). Hach contains a central cavity which is lined by w single row of very granular cells varying somewhat in size, each with a nucleus. The cavity of the central group is some- what irregular, while its floor is formed by the layer of longi- tudinal muscles on which are situated gland-cells. Hach of the lateral groups has a definite opening to the exterior through the GLANDS OF AUSTRALIAN EARTHWORMS. 135 body-wall (fig. 22, a’), but no opening could be seen in relation to the central gronp. It is probable, however, that this opens by means of a communication with the duct leading from the inner of each set of lateral groups. Each group is surrounded by a distinct layer containing muscle-fibres. This doubtless assists the propulsion of the secretion outwards. Between each of the two lateral groups on each side and between them and the central group, in the region of the spermiducal-duct opening, is a muscular sac (s.s.) containing a large curved but smooth penial seta (s‘). There are thus two of these on each side, and the gland-groups open, not around the opening of the spermiducal duct, but around the penial sete. This will be clear on a reference to the figure. (2) Glands embedded in the body-wall, e.g. Megascolex dor- salis and Digaster excavata. In M. dorsalis (Pl. 15. fig. 23) there are two completely different sets of glands separated by the sper- midueal duct. The first and upper of these (a.’g.'), which lies to the outer side of the spermiducal duct, is composed of a number of alveoli. Hach alveolus contains a central cavity surrounded by large finely granular gland-cells, each with a nucleus pushed to its outer edge. The alveoli are separated from each other by a number of connective-tissue septa. No connection could be traced between these and the gland-duct, though from their position itis probable that they do, at times, open into the spermi- ducal duct. The second set are those (a.g.) to be found on the inner side of the gland-duct. They are much lower down in the papilla. The cells of these groups are much smaller and clearer, though arranged as before around a central space. Hach group opens on the surface not by means of a single duct, but by a number of fine ducts from the gland-cells forming the group. These run down and open to the surface between the columnar cells of the epidermis. It is the latter set only which are found in Digaster excavata. In this form the groups are more nume- rous, but more scattered and looser in structure. The groups of unicellular ducts, which are longer than in Megascolex dorsalis, are to be seen very clearly opening on the surface. One or two groups open into the spermiducal duct, close by the opening of the sete into the latter. 136 MISS G. SWEET ON THE SPERMIDUCAL Summary. It will be seen from the above description that there is, in these species, a continuous series of forms of increasing com- plexity in internal structure. Commencing with the tubular forms, we have first those in which the gland-cells open into the lumen between the columnar cells of the lining (PI. 14. fig. 16). There is here obviously a very close resemblance in structure to the clitellum or to the special accessory gland in Megascolides intermedius and M. attenuatus, if we neglect the occurrence of the muscle-layers there. Next we have those forms in which perhaps, on account of the increase in cornifica- tion of the lining epithelium, the gland-cells open into the lumen in groups (PI. 15. fig. 17). The next stage is seen in such a form as Digaster excavata, where the lumen gives out a single series of branches, each of which bears at its end a group of gland-cells, which is thus situated at about half the thickness of the gland-wall. Here, however, the columnar lining of the main duct is not continued up these branches (cf. Digaster excavata). Further complication is met with in such forms as Megasco- lides illawarre and Megascolex dorsalis, where the main duct all along its length branches considerably more than in Digaster excavata, and here to a great extent the branches are lined by cubical cells continuous with the columnar lining of the main duct (Pl. 15. fig. 24); but here also there is no cell-lning to the fine terminations of the branches. From these we pass to such highly complex forms as, for example, Megascolex tasmanica, M- Frenchi, in which the main duct breaks up immediately it enters. the gland, so that there is no central tube in these lobate forms (Pl. 15. figs. 19 & 20). Corresponding with this, the branching is still further increased, and as a result the lobules become more closely compacted together and fill up the entire interior of the: gland. Consequent on this increase in branching, the groups. are much more numerous and are pushed out towards the peri- phery, so that the mass of the gland is made up of these groups, the cells in which become more closely pressed together and their ducts much shortened. The blood-vessels which originally ran along the surface of the gland, or just outside the lining of the duct, have become much branched corresponding with the branching of the duct, and serve very often, with the small GLANDS OF AUSTRALIAN EARTHWORMS. 137 amount of connective-tissue in which they run, to separate lobule from lobule, at the same time sending branches into them between the cells, to which they supply materials for secretion. The result of this investigation into the structure of the spermi- ducal glands of an entirely different series of species from those studied by Beddard, clearly corroborates the statement that the more complicated lobate forms are derived in a compara- tively simple way from the tubular. Acanthodrilus sydneyensis, however, differing as it does so markedly from the Acanthodrilide generally, cannot be referred to the consecutive series at all. Conclusions. The following are the more important conclusions arising out of this paper :— 1. That for the glands in question there is a correspondence between internal structure and external form. 2. That there exists a complete series of conditions connecting the simplest tubular form with the most complex lobular. 3. That a cell-lining is absent from the finest terminations of the duct. 4, That the structure of the accessory glands in Diplotrema fragilis is peculiar, and that of the spermiducal gland in Acanthodrilus sydneyensis unique. EXPLANATION OF THE PLATES. Reference Letters. a.g., accessory glands. 1.'m.', longitudinal muscle-band. a.'g.', Upper accessory glands. 1./., lacuna in centre of lebule. b.v., blood-vessel. m.b., muscular bulbus. ce. columnar epithelium. 2.€., nerve-cord. ¢l., cuticular layer. p., peritoneum. c.m., circular muscle-layer of body- r., Ciliated rosette. wall. s., penial seta. ¢.'m.', circular musele-layer of gland- s.s., setal sac. duct. s7r., sperm-reservoir. c.t., connective-tissue. sp.d., spermiducal duct. gc., gland-cells. sp.'d.', fine branches of spermiducal gc.', special club-shaped gland-cells. duct. g."'c."', duct of gland-cell. sp.g., spermiducal gland. i.m., longitudinal muscle-layer. v.d., vas deferens. The outlines of all the figures have been drawn beneath the camera lucida, and on pl. 14 the position of the yas deferens is in all cases indicated by dotted lines. LINN. JOURN.—ZOOLOGY, VOL. XXVIII. 10 MISS G. SWEET ON THE SPERMIDUCAL Puate 14. Fig. 1. Spermiducal gland on left side of Diporocheta Bakert, showing spermi- Or 14. 15. 16. Fig. 17 ducal duct and two setal sacs. 6. . Fletcherodrilus wnicus. Spermiducal gland of right side. The muscular bulbus (7.6.) is seen on each side to run under the ventral longitudinal muscle-band (¢.'7.') which supports the nerve-cord (7.¢.). X 6. . Megascolides intermedius. 'The spermiducal glands lie in segment 18, transversely to the body. On the ventral surface is seen the mass of unicellular glands, the surface being mottled with patches of pigment. The nerve-cord runs above this. xX 6. . Megascolides victorie. Spermiducal gland of right side lying in segments 18 to 29. The duct is much coiled. X 6. . Megascolides roseus. Spermiducal gland of left side. Gland very much coiled. xX 6. . Diplotrema fragilis. Spermiducal glands and ventral accessory glands. x 18. . Acanthodrilus sydneyensis. Two pairs of spermiducal glands. The position and opening of the vas deferens on segment 18 is shown by a dotted line. . Megascolex Frenchi. Lobate spermiducal gland of right side, in segments 18 and 19. The markings on the surface of the gland may be clearly seen. xX 6. . Megascolex Dendyi. Spermiducal gland of right side. Lobate in segments 18 and 19, with a curved duct in segment 18. x 6. . Megascolex dorsalis. Spermiducal gland of left side. Surface showing irregular markings. Duct very short. xX 6. . Digaster queenslandica, Gland of right side. Surface-markings slightly visible. No duct to be seen from above. x 6. . Digaster queenslandica. Spermiducal gland of another individual. Tubular in appearance. No duct visible. X 12. . Digaster armifera. 'Trilobed gland of right side with setal sacs and muscles belonging to them, attached to the terminal part of the gland- duct. x 6. Megascolex Fielderi. Bilobed gland of left side with lobular surface- markings. Duct curved and entering muscular bulbus. x 6. Megascolex intermedius. Portion of section showing sperm-reservoir of right side with its division into two parts, the lower containing the sperm-rosettes, and developing sperm in both. The vas deferens from the previous segment may also be seen, and the nerve-cord in position. Zeiss A.A. oc. (2). Megascolides insularis. Transverse section of spermiducal gland showing ordinary gland-cells, special club-shaped cells, and columnar lining. The whole enclosed in peritoneum. Zeiss D. oc. (2). PuatE 15. . Megascolides intermedius. ‘Transverse section through gland-wall, showing groups of gland-cells. The outlines of the columnar cells lining the duct are not visible. The nuclei are clearly seen, as also Sweet. > x XIX A R.Hammond lith. G Sweet del. West, Newman imp. Loar, Soc Journ. Zoor. Vor. XXVIIPL 14). so i RTIHWORMS ad XVI XIX XX XX XX VII XXVIII XX\X = SSS > = = Sys oe = ze ee > =< < ~~ x< Eas Scie ese £ XXVIII XX IX 4 ' \ wi ut . ' 3 ' 16 ‘ : \ t ; x i “5 ST os b ‘ ' i ; mt fe eae J i Gye! rad oat, Sweet. eee ond lith. G.Sweet del A RHamm Linn. Soc. Journ. Zoon Vou XXVIIIPL 15. —— ~~ West, Newman imp. Sey fe etree Piet opm Or nn Per sruslareen nara Maga pa erd (= GLANDS OF AUSTRALIAN EARTHWORMS. 139 the horny layer next the lumen. Two small blood-vessels are seen running in from the surface. Zeiss D. oc. (2). Wig. 18a. Acanthodrilus sydneyensis. Transverse section through duct of 18 8. 18 ¢. 19. 20. Al 22. 23. 24. gland showing nuclei of columnar lining and the circular muscle- layer. Acanthodrilus sydneyensis. Transverse section through the inter- mediate part of gland showing nuclei of columnar cells surrounded by short granular cells. Acanthodrilus sydneyensis. Oblique section through terminal glandular part of gland. A blood-vessel is seen running up one side. The irregular gland-cells are shown; the relative depth to which they are stained with methylen-blue is shown by the shading. Zeiss D. oc. (2). Digaster brunneus. Section across lobate gland showing the granular gland-cells and the apparently empty cells all with their ducts Opening into a verysmall central space without any lining membrane. Around the lobule are a number of blood-vessels separating it from its neighbours. Zeiss H. oc. (2). Megascolex tasmanica. Section across lobule in gland showing the gland-cells opening into a much larger lacuna, which is connected with one of the five terminations of the gland-duct by means of a branch with a single layer of much flattened epithelium cells. The cut ends of some of the glandular cell-ducts are seen in the lacuna. No blood-vessels were visible, and are probably present in the form of sinuses. Zeiss EH. oc. (2). Megascolides intermedius. Portion of transverse section through region of the spermiducal gland showing portion of wall of spermiducal duct cut through, and the mass of unicellular glands occupying the ventral surface here, with their ducts opening to the surface. The position of the nerve-cord is shown and the muscular layers of the body-wall. Zeiss C. oc. (2). Diplotrema fragilis. Portion of transverse section showing the rela- tions of the spermiducal duct, accessory glands, sete and setal sacs, and nerve-cord. The vas deferens has opened two sections anterior to this, while the spermiducal gland opens two or three posterior to this. The opening of one group of gland-cells is seen cut through ata’. Zeiss A. oc. (2). Megascolex dorsalis. Transverse section of the accessory glands (a.g. and a.'g.') surrounding the opening of the spermiducal gland which is seen cut through obliquely. The groups of unicellular ducts are clearly seen in the lower portion of the section. Zeiss C. oc. (2). Cryptodrilus illawarre. Portion of transverse section showing the lower part of the gland, the gland-duct, vas deferens, and the two setze in their sacs. Zeiss A. oc. (2). 10* 140 DR. C. CHILTON ON THE SUBTERRANEAN The Subterranean Amphipoda of the British Isles. By CaarnEs Cuitron, M.A., D.Sc., F.L.S., Research Fellow, University of Edinburgh. [Read 21st June, 1900.] (Puates 16-18.) AurnHouaH the first subterranean Amphipod that was definitely recognized and described as such was a British species, and although the different species known at the time were described and figured in some detail by Spence Bate and Westwood in 1863 [1. pp. 311 to 328] *, very little has been published on the group by subsequent British writers, though on the Continent several important papers dealing with the subject have appeared. These Amphipods are known to be widely distributed in the southern parts of England and they have been also recorded from Dublin, and, judging from experience in other countries, they are probably fairly abundant; but at the same time it is by no means easy to obtain specimens, for they are either alto- gether overlooked by the ordinary householder, or, if they are seen, their presence is, as Mr. Stebbing [2. p. 30] has pointed out, kept a secret from the fear that the well may otherwise be closed by the sanitary authorities. There are several points that are still uncertain with regard to the number and distribution of the British species, and in the course of a wider study of sub- terranean Crustacea I was anxious to solve these so far as possible,, and to attract fresh attention to the subject in the hope that further knowledge might be obtained upon it. This paper is the result of the work that I have been able to do on the subject; but though various friends have generously supplied me with all their available material, I regret that the results must seem somewhat meagre, and that I am not able to add very much to the information given many years ago by Spence Bate. How- ever, I give fuller details of some of the species and revise the — nomenclature, after comparison with some of the European species. T also give a very short sketch of the growth of our knowledge on the subject, and some brief notes on the distribution of the British species. * The numbers in square brackets refer to the list of references at the end of the paper. AMPHIPODA OF THE BRITISH ISLES. 141 For supplying me with material I desire to record my best thanks to the Rey. T. R. R. Stebbing, who has very kindly placed the whole of his specimens of the group at my disposal, to Pro- fessor D’Arcy W. Thompson, C.B., of University College, Dundee, to Dr. R. T. Scharff of Dublin, and to Dr. S. F. Harmer of Cam- bridge. Monsieur Edouard Chevreux of Bone, Algeria, and Professor Franz Vejdovsky of Prague have given me specimens from some localities on the Continent that have been most useful for the purpose of comparison. Historical. The first mention of subterranean Crustacea in the British Isles was made by Leach [3. p. 403] in the article ‘‘ Crustaceo- logy ” in the ‘ Edinburgh Encyclopedia,’ published probably in 18138 or 1814. His specimen came from a well in London, and was by him distinguished from Gammarus pulex under the name G. subterraneus. His description is very brief and insuf- ficient, but from what he says it is, I think, evident that the specimen must have belonged to the genus now known as Niphargus ; and from our present knowledge of the distribution and occurrence of the species of this genus, it is perhaps not unreasonable to suppose that it was the species which is most commonly found and which has long been known in England by the name Niphargus aquilex, Schiddte, and I have accordingly in this paper followed Della Valle in adopting Leach’s name for this species. Spence Bate and Westwood give Leach’s species as a synonym of WV. aquilex, though retaining the later name of the species [1. p. 314]. For many years nothing appeared on the subject in England, though on the Continent species of subterranean Amphipods were described by Gervais [4. p. 127], Koch [5], Schiddte [8], Caspary [6], Hosius [7], &c., and in 1851 the genus Wiphargus was established by Schiddte. In 18538 a subterranean Amphipod was found by J. O. West- wood at Maidenhead [9. p. 218]; this was at first referred to WV. stygius, Schiddte, but was shortly afterwards described as a separate species, V. aquilex, by Schiddte. This species was accepted as a good one by Spence Bate and Westwood, though on different grounds from those relied upon by Schiédte in defining the species. Adam White, in his ‘Popular History 142 DR. CHARLES CHILTON ON THE SUBTERRANEAN of the British Crustacea’ [10], gave the species under the name Niphargus aquilex, Schiddte, and suggested that it was the same as Gammarus subterraneus, Leach. Some account of it was given anonymously in the ‘ Natural History Review’ in 1857. In 1859 Spence Bate [11] described two new species of M- phargus, viz. N. fontanus and N. Kochianus, and also established the new genus Crangonyw for another species, C. subterraneus, found by the Rev. H. R. Hogan at Ringwood. In the same year Hogan gave some account of the habits, food-supply, &c. of these three species [12. pp. 166-169]. All the British subterranean species were given by Spence Bate in the “Catalogue of the Amphipoda in the British Museum’ in 1862 [13. pp. 174 &e.], and afterwards by Bate & Westwood in their ‘ History of the British Sessile-eyed Crustacea,’ where full descriptions are given and some general remarks are made on their affinities, habits, distribution, &c. [1. pp. 311 to 328]. I do not know of any further work in English specially dealing. with these Crustacea, though naturally they have frequently been referred to in some of the numerous works dealing with the similar forms found in other countries; some remarks upon them are made too by Stebbing in the course of the Biblio- graphical Introduction to his ‘‘ Report on the ‘ Challenger’ Amphipoda.” Quite recently a species of Miphargus has been recorded from Cringleford, near Norwich, by Dr. 8S. F. Harmer [14. pp. 489- 491]; and the Rev. T. R. R. Stebbing [2] has drawn attention to the whole subject of the British subterranean fauna in his address to the Associated Societies at the Dover Meeting of the British Association. Of the writers on European species it will be sufficient to mention the names of Valette St. George [15]; Joseph [16]; de Rougemont [17], whose papers led to so much controversy ; Humbert [18], who described a variety of Miphargus puteanus, Koch, with an accuracy and fullness of detail that has scarcely been equalled since; Moniez [19]; and WrzeSniowski [20], in whose paper a full historical account of the subject will be found. More recently Chevreux [22] has described a new species of Niphargus from France, Vejdovsky has published valuable papers: AMPHIPODA OF THE BRITISH ISLES. 143 on Crangonyx [21] and on the rudimentary eyes of Miphargus [34], and Armand Viré [23] deals with two species of the latter genus in his work ‘La Faune souterraine de France,’ issued during the present year. In North America several similar Amphipods have been dis- covered and have been described by Cope [24], Packard [25j, Forbes [27], Benedict [26], &c.; full information of the cave- fauna of North America will be found in Packard’s larger memoir published in 1888, and he has rediscussed some of the theoretical bearings of the facts in a subsequent paper [35]. The New-Zealand forms first recorded by me in 1881 and 1882 are fully described in my paper in the ‘ Transactions of the Linnean Society’ in 1894 [28. pp. 163-284]. Closely allied Amphipods, but with normal eyes, were described from Tasmania by G. M. Thomson in 1892 [29]; and within the last yeara blind species of Miphargus has been found by O. A. Sayce in Gippsland, Victoria, Australia [30. pp. 152-159]. Distribution of the British Species. Owing to the isolated localities in which they are found and to the fact that specimens have fallen into the hands of many different observers, a large number of species of Miphargus has already been described, some of them probably on insufficient grounds. In his revision of the genus in 1890 Wrzesniowski gives a list of 14 species, six, however, he marks as doubtful; and Stebbing [31. p. 425] has since suggested that one of these, Niphargus Moniezi, may perhaps more appropriately find a place under Neoniphargus. On the other hand, Chevreux has recently described a new species, V. Virez, from the grottos of the Jura and tells me by letter that he has other new species to describe. Other writers, such as de Rougemont and Della Valle, again, have supposed that the different species described are merely forms of one wide-spread species, and have caused great confusion by reducing them all to synonyms of Miphargus puteanus, Koch. The latest effort of this kind is that of Hamann [83. p. 234], who recognizes neither Miphargus nor Crangonyx, and classes all the subterranean Amphipods of Hurope under Gammarus puteanus, Koch. It is almost incredible that he should come to this con- clusion after avowedly studying Wrzesniowski’s work, and it will be evident to all that he has quite failed to appreciate 144. DR. CHARLES CHILTON ON THE SUBTERRANEAN the careful observations of the latter, and his view that all the recorded species are forms of one cannot be for a moment accepted. Fortunately there is no great difficulty in identifying our British species. Besides Niphargus fontanus, which I have not seen, we have three species. The first, N. subterraneus, Leach, is known from many districts in the Southern Counties of Eng- land and as far north as Norwich; it has not been recorded from Scotland nor Ireland; Mr. Thomas Scott tells me that he has frequently sought for subterranean crustaceans from wells near Edinburgh without success, and I have not been able to hear of them in any part of Scotland, though I have made frequent inquiries ; neither could it be found in the Irish caves investigated by Carpenter and others, though it had been specially looked for years before by Wright and Halliday [36. p. 26]. In Europe the species is very widely distributed and has been recorded from many localities in France, Austria, Bohemia, Germany, Poland, &c. The second species, V. Kochianus, Spence Bate, has been recorded from several localities in the South of England, and is also found in Ireland at Dublin. In Europe it is so far known from Munich only, though it is doubtless to be found at other places; and it seems probable that V. puteanus, described by Hosius from Bonn, is the same species. _ The third speeies, Crangonyx subterraneus, Spence Bate, is known in England only from Ringwood and Marlborough and appears to be rare, for in each case only a single specimen was obtained. In Europe a few specimens have been recorded from Radotin near Prague by Vejdovsky, and many years previously it was taken at Munich by de Rougemont. From this it appears that, with the exception of WV. fontanus, which is as yet known only from Spence Bate’s description, all the British forms belong to species widely distributed in HKurope. Genus Nipuarcus, Schiddte. 1851. Niphargus, Schiddte, Det danske Videnskabernes-Selskabs Skrifter, 5e Raekke. Naturvidenskabelig og Mathematisk Afdeling, Ba. ii. p. 26. The genus Niphargus was established by Schiddte in 1851 for the reception of some subterranean Amphipoda and was accepted ! AMPHIPODA OF THE BRITISH ISLES. 145 by Spence Bate & Westwood, who gave a fuller diagnosis in 1863 [1. p. 311]; in 1876 Humbert gave a still fuller description [18. p. 312]; and more recently Wrzesniowski [20. p. 620] has given a full critical history and account of the genus, followed by a Latin diagnosis, of which I give a translation here :— “* Hyes none (or rudimentary). “ Body compressed, not carinate. Fourth and fifth segments of the pleon bearing slender setules. Epimera small. “ Superior antenne longer than the inferior, with a secondary appendage of not more than two joints; primary flagellum bearing olfactory sete and hyaline bacilli. Flagellum of inferior antenne in both sexes bearing only hyaline bacilli. “ Gnathopoda similar, subchelate, with the penultimate joint (propodos) dilated, of nearly the same form in both sexes. “Terminal uropoda biramous, inner ramus very small, outer ramus elongate, two-jointed (or one-jointed ?). “Telson more or less deeply cleft. “Molar tubercle of the mandibles supplied with a long seta; palp narrow, three-jointed. “ First maxille furnished with forked spines ; palp large, two- jointed, of the same form in the right and left maxille, armed at the apex with spines and sete ; inner lobe narrow, bearing only two or three setz at the apex. “Second maxilla with the inner lobe bearing sete at the apex only. “ Outer lobe of the maxillipedes armed on its inner margin with teeth, at the apex with teeth and strong sete; inner lobe armed at the apex with three strong teeth and a very few setz, the inner margin destitute of sete ; palp elongate, the last joint very narrow towards the apex, unguiform.” This is rather a long and unwieldy generic description, and no doubt the diagnosis could be considerably curtailed if Wephargus were compared with allied genera and the characters common to several genera carefully eliminated ; and this will I hope be done by the Rev. T. R. R. Stebbing in his account of the Amphipoda prepared for ‘ Das Tierreich.’ In his paper on new genera of the Gammaride already published [31] he does not give revised diagnoses of the genera previously known, but among the new genera he gives one, Neoniphargus, which appears to come very close to Niphargus, the only important difference being that in 146 DR. CHARLES CHILTON ON THE SUBTERRANEAN it the third uropoda are not elongate and their outer branch consists of one joint only. The affinities of Miphargus were fully discussed by Wrzesni- owski in 1890, and in the present state of our knowledge it does not appear possible to add very much to what he then said. Its nearest allies seem to be the Neoniphargus already mentioned and Crangonyx, with the genera Paracrangonyx and Hucrangonyx lately established by Stebbing, while the little-known marine genus Eriopsis also seems to be nearly related. Naturally enough it has frequently been compared with Gammarus, since freshwater species of this genus are often present in the districts where the subterranean Wiphargi are found. Wrzesniowski has pointed out that most of the external characters of Miphargus are shared by various species of Gammarus, and that consequently these alone are not sufficient to distinguish the two genera; he has, how- ever, shown that there are important differences in the mouth- parts, and these he has described in great detail. Although there are considerable differences between the mouth-parts of a typical Gammarus and a typical Miphargus, there are already known some intermediate species, and no doubt others will here- after be described which will still further bridge over the gap between the two. Some of the numerous species from Lake Baikal referred to Gammarus by Dybowsky present external resemblances to Wiphargus; but although many of them have been assigned to new genera by Stebbing, no satisfactory account of their mouth-parts has as yet been published, and without this it would be useless to attempt any detailed comparison. In his account of the genus Crangonyw Professor Vejdovsky has drawn special attention to the sensory sete found on various parts of the body aud appendages; and he tells me that in these and also in some parts of the internal anatomy, especially in connection with the renal gland in the base of the lower antenne, he has found good points of difference not only between allied genera such as Gammarus, Niphargus, and Crangonyx, but in some cases even between different species of the same genus—a paper dealing with these points willshortly be published by him. It is unfortunate that most of these characters, important as they are, can be observed only in fresh specimens or require to be elucidated by the cutting of serial sections, and that consequently they have been and, toa large extent, probably will continue to be neglected by the systematist. AMPHIPODA OF THE BRITISH ISLES. 147 NIPHARGUS SUBTERRANEUS, Leach. (Plates 16 and 17. fig. 1.) 1814, Gammarus subterraneus, Leach, Edinburgh Encyclopeedia, vol. vii- p- 408. 1841-4. Gammarus puteanus, Koch, Deutschlands Crustaceen, Myria- poden und Arachniden, Heft 36, Taf. xxii. 18538. Niphargus stygius, Westwood, Proc. Linn. Soc., No. li. p. 218. a . Spence Bate, Ann. & Mag. N. H. ser. 2, xix: p. 146. 1855. Niphargus aquilex, Schiddte, Oversigt over det Ke. Danske Vidensk.- Selskabs Forhandlinger, 1855, pp. 349-351; also in Nat. Hist. Review, i. p. 41, fig. B. 1857. Miphargus aquilex, White, Popular History of British Crustacea, p. 187. 1857. Gammarus puteanus, La Valette St. George, “De Gammaro puteano,” Dissertatio Inauguralis, 1857. 1862. Miphargus stygius, Spence Bate, Cat. Amphip. Brit. Mus. p. 174, pl. xxxii. fig. 1. 1863. Niphargus aquilex, Spence Bate & Westwood, Brit. Sessile-eyed Crust. 1. p. 315. 1888. Miphargus aquiler, Stebbing, Report on the ‘Challenger’ Amphi- poda, p. 316. 1889. Gammarus puteanus (“ & main triangulaire ”), Moniez, Rev. Biol. du Nord de la France, i. pp. 41-46. 1890. Niphargus puteanus, Wrzesniowski, Zeitschrift fur wissenschaft- liche Zoologie, L. 4, p. 673. 1893. Niphargus subterraneus, Della Valle, Gammarini del Golfo di Napoli, p. 704 (in part). 1896. Gammarus puteanus, Hamann, Europiiische Hohlenfauna, Jena, p: 234 (an part). 1899, Mphargus aquilex, Harmer, Trans. Norfolk & Norwich Naturalists’ Society, vol. vi. pp. 489-491. 1900. Miphargus puteanus, Armand Viré, Faune souterraine de France, p. 34 &e. Specific diagnosis.—Body slender, first four side-plates not so deep as their respective segments ; postero-inferior angles of the first three segments of the pleon broadly rounded. Superior antennz half as long as the body ; flagellum of about 20 joints, rather more than one and a half times the length of the peduncle; secondary appendage of two joints, shorter than the first two joints of the main flagellum. Inferior antenna with the flagellum of 7 to 9 joints. Maxillipedes with the lobes small, inner lobe reaching only to the end of the inner margin of the meros, and outer lobe reaching only halfway along the inner margin of the carpus. \ \ 148 DR. CHARLES CHILTON ON THE SUBTERRANEAN First gnathopod with the basos very broad, the width at the distal end in the male being fully one-half the length: ischium and meros both short and subquadrate and of about equal size ; meros with its distal border fringed with a row of long sete, of which one is much longer than the others and is finely barbed, curved, and reaches about as far as the end of the propodos; greatest length of the carpus considerably less than that of the propodos; extremity very oblique and nearly all occupied by the articulation of the propodos, so that the posterior border is very short; a tuft of sete at the antero-distal angle, and a row along the outer edge of the short posterior border, on the inner surface is a row of about seven long sete along the half of the oblique extremity nearest to the posterior border: propodos large, subtriangular, broadest distally where the breadth is slightly greater than the length of the anterior border: palm transverse, straight or slightly convex; anterior border with a tuft of about six sete at base of dactylos, and a smaller tuft situated a little proximally to this; the posterior border bears six short transverse rows of setz ; palm defined by a stout spine, and near it two smaller spines coarsely serrated, along the palm is a row of short sete, with one or two longer ones at intervals: dactylos fitting closely on to the palm, terminal claw very acute, secondary claw distinct and sharp with a short seta at its base, outer border convex and bearing only one seta opposite the base of the secondary claw. Second gnathopod slightly larger than the first, but very similar in form: basos not so broad, the greatest breadth being rather less than half the length; ischium and meros similar to those of the first gnathopod, but the meros without the large curved seta; carpus as long as the propodos, rather more slender than in the first gnathopod, the posterior surface longer and slightly concave for the reception of the base of the pro- podos, both edges bearing a row of sete; propodos as in the first gnathopod, but with the breadth at distal end distinctly greater than the Jength of the anterior border; posterior border very convex and with nine rows of sete ; dactylos as in first gnathopod. First and second pereiopoda equal in length, third slightly longer, fourth and fifth each distinctly longer than the one immediately preceding. Basa of third, fourth, and fifth pereio- poda rather narrow, the greatest breadth being rather more than one-half of the length. AMPHIPODA OF THE BRITISH ISLES. 149 Terminal uropoda in the fully-grown male nearly one-half the length of the body ; outer ramus of two subequal joints, the first with four tufts of sete, the terminal one nearly free from sete, except a small tuft at the apex. In the female the terminal uropoda one-fourth the length of the body, and the last joint of outer ramus only about one-third of the first. Telson reaching to the end of the peduncle of the third uro- poda, split to more than half its length, each lobe narrowing towards the apex, which bears three stout sete; outer margin convex, and with two fine set a little anterior to the apex. Length of fully-grown specimens about 11 mm. Habitat. Southern Counties of England ; (Europe generally). The above description will, I hope, be sufficient fur the identi- fication of this species. A much more detailed description is given by Humbert of Mphargus puteanus var. Forelii, from Lake Geneva. Wrzesniowski indeed thinks this form sufficiently distinct to rank as a separate species, WV. Forelii; but though he is perhaps right, the differences are very slight. Wrzesniowski’s full description of Miphargus puteanus var. Vejdovskyi may also be consulted. The points chiefly relied upon by Wrzesniowski for the sepa- ration of the different species of Niphargus are the lengths of the antennz and terminal uropoda in proportion to the body, the depth of the side-plates, the relative lengths of the pereiopoda, the shape of the propoda of the gnathopoda and of the lower margins and angles of the first three segments of the pleon, and the relative lengths of the two joints of the outer ramus of the third uropod. These points he found to be fairly constant in specimens of different ages from the same locality ; and they no doubt are good specific characters, though it is a little difficult to estimate some of them with accuracy, and I am inclined to think that Wrzésniowski relied a little too much on the supposed accuracy of the descriptions and figures given by other writers. It must be remembered, too, that the relative lengths of the two joints of the outer ramus of third uropoda varies in the two SeXes. In addition to these points, I have found the shape of the maxillipedes and of the more proximal joints of the gnathopoda, especially the basos and the carpus, useful points for distinguishing the British species; while the number of sete arising from the outer margin of the dactylos is a character that may be useful for readily separating some species, such as J. tatrensis, 150 DR. CHARLES CHILTON ON THE SUBTERRANEAN Wrzegniowski. The slight differences of the two gnathopoda in the form of the carpus have hitherto received little attention. I do not propose to enter into any points of internal anatomy, but, as the question of the degree of the degeneration of the eyes is of special interest, I may call attention to a recent paper by Vejdovsky, in which he criticises the accounts pre- viously given, and, after a careful examination by means of serial sections, comes to the conclusion that m none of the species specially dealt with is there any organ present corresponding to a normal Arthropodan eye; and that only in WV. puteanus, Koch, is it possible to speak of a true eye-rudiment, and that even here the degeneration has gone so far that the cells of the “ Augen- anlage’? have been modified into a tendon-like bundle that serves to attach the large brain-ganglia to the walls of the head 34. p. 10}. The chief points by which the present species differs from other species of the genus appear to be :— The body is slender. The four anterior side-plates are not so deep as their segments. The upper antenne are not more than half the body-length. The gnathopoda have the basos broad, the carpus not longer than the propodos, the propodos rather broader than long, the dactylos with only one seta on its convex margin and only three or four very small ones on the concave margin. The first three segments of the pleon have the postero-inferior angle broadly rounded. The outer branch of the third uropod has the two joints of nearly the same length in the male, and in fully-grown specimens the last joint bears only few very short sete. (In younger males the terminal joint, though of about the same length as the first, may be more abundantly supplied with sete, and these may be longer than is shown in Plate 16. ur. 3.) Niepareus Kocuranus, Spence Bate. (Plates 16 and 17. figs. 2 & 3.) 1859. Niphargus Kochianus, Spence Bate, Proc. Dublin Univ. Zool. & Bot. Assoc. i. p. 239, figs. 1 and 1a; Nat. Hist. Rev. vol. vi. p. 165, fig. 1. 1859. Niphargus Kochianus, Hogan, Nat. Hist. Rev. vol. vi. pp. 166-169, 1861. Niphargus Kochianus, Hogan, Rep. Brit. Assoc. (1860) pp. 116-117. . AMPHIPODA OF THE BRITISH ISLES. 151 1862. Miphargus Kochianus, Spence Bate, Cat. Amphip. Brit. Mus. p. 177, pl. xxxii. fig. 3. 1863. Niphargus Kochianus, Spence Bate & Westwood, Brit. Sessile-eyed Crust. i. p. 528. 1889. Gammarus Kochianus, Moniez, Rev. Biol. du Nord de la France, 1. p. 48. 1890. Hip hoctou Kochianus, Wrzesniowski, Zeitschrift fiir wissenschaft- liche Zoologie, L. 4, p. 674. 1893. Miphargus Kochianus, Della Valle, Gammarini del Golfo di Napoli, p- 704 (2 part). 1896. Gammarus puteanus, Hamann, Kuropiische Hohlenfauna, p. 234 (in part). Specific diagnosis.—Body less slender than in NV. subterraneus, first four side-plates as deep as their segments. Second and third segments of pleon with posterior angles acute, that of the first rectangular. Superior antenna about two-thirds the length of the body, slender; third joint of peduncle more than half as long as the preceding joint; flagellum containing about 16 joints, secondary appendage of two slender joints, as long as the first two joints of the main flagellum. Inferior antenna with the last two joints of peduncle equal, rather slender; flagellum of about 8 joints, half as long again as the last joint of peduncle. Maxillipedes with the lobes large, inner lobe reaching con- siderably beyond the distal end of the inner margin of the meros, and the outer lobe reaching as far as the end of the carpus. First gnathopod shorter and rather stouter than the second ; basos not specially broadened ; carpus as long or longer than the propodos ; propodos subquadrate, widest distally where the width is rather greater than the length; palm transverse, produced anteriorly as it approaches the inferior angle, so that this is rather less than a right angle, defined by a stout spine with a smaller one at its base; dactylos fitting closely on to the palm, convex border with only one or two sete. Second gnathopod slender ; carpus much longer than propodos, its inferior margin not much shorter than the anterior, and supplied with five tufts of sets; propodos as in first gnathopod, but narrower, its articulation with the carpus very narrow. Terminal uropoda about one-fifth of the length of the body; outer branch with the first joint about four or five times as long as the peduncle, and supplied with three or four tufts of sete ; 152 DR. CHARLES CHILTON ON THE SUBTERRANEAN second joint slender, one-fourth as long as the first, with a single seta at the apex. Telson cleft for three-fourths of its length; lobes widely separated, their outer border nearly straight, with one long plumed seta, and a smaller one one-fourth from the apex, mner border slightly convex, extremity with four stout sete. Length about 5 mm. Habitat. Ringwood, Hants; Warminster and Marlborough, Wilts ; and Dublin. (In Europe recorded from Munich.) This species appears to be less common than the preceding, and has consequently received less attention. I have specimens from Marlborough and from Dublin, and have algo been able to examine specimens from Munich; and though there are a few small differences that could easily be pointed out, I think these all belong to the one species. I feel pretty certain that WV. puteanus, Hosius, from Bonn belongs to this species, aud it seems not unlikely that VV. puteanus, Koch, from Regens- burg, also belongs to the same species. However this may be, it appears clear that the present species is quite a different form from NV. subterraneus, and that it is widely distributed in Europe and the British Isles. The points by which it may be distin- guished have mostly been mentioned in the brief description already given. Perhaps the most characteristic points are to be found in the gnathopoda, which have the carpus much longer in proportion to the other joints, and have the propodos articulated to it only at the distal end and lying in the same straight line as the carpus; while in JV. subterraneus the propodos usually lies nearly at right angles to the carpus, and its articulation with it is consequently oblique and occupies part of the short inferior margin. The propodos itself is not so broad as in WV. subter- raneus, aud hag the inferior angle of the palm produced in a very characteristic manner, especially in the first gnathopod. The amount of this projection varies a little in different specimens, as may be seen from a comparison of figs. 2 gn.", gn.*,and 3 gn.', gn2, and it appears to be greater in small specimens. Hig. 2 gn.’ shows the propodos of the first gnathopod of a small specimen (4mm. long) from Dublin, and in this the projection is very marked so that the gnathopod might almost be called imper- fectly “chelate” instead of “‘subchelate,” the inferior margin bears only two distinct transverse rows of sete, though the third and most proximal one is represented by a single seta. The e AMPHIPODA OF THE BRITISH ISLES. 153 portion of the propodos bordering the palm and its inferior angle differs from the rest of the joint in presenting a peculiar stippled and striated appearance ; and it is evidently this that is represented by the shading in Spence Bate’s figure in the ‘ Hist. Brit. Sessile-eyed Crustacea,’ p. 323. This “ marginate” appear- ance is not so marked in the larger specimen from Marlborough (see figs. 3, gz.', gn.’), nor in the specimens from Munich, and the tufts of sete on the inferior margin of the propodos are more numerous. The first gnathopod is considerably shorter and rather stouter than the second; and may also be distinguished from it by the long curved seta on the meros, which is present just in the same position as in LV. subterraneus, in which species the two gnathopoda are not so easily distinguished. The large size of the inner and outer lobes of the maxillipedes is another point that readily separates this species from JV. sub- Lerraneus. The terminal uropoda do not appear to become so elongate as in NV. subterraneus, but the specimens at my disposal are not sufficient to enable me to say how far they differ in the two Sexes. NripHareus FonTaNus, Spence Bate. - 1859. Niphargus fontanus, Spence Bate, Proc. Dublin Univ. Zool. & Bot. Assoc. i. p. 288, figs. 2 & 2a; Nat. Hist. Rev. & Quarterly Journal of Science, vol. vi. p. 165, fig. 2. Niphargus fontanus, Hogan, Proc. Dublin Univ. Zool. & Bot. Assoc. i. p. 240; Nat. Hist. Review, vol. vi. p. 166. 1862. Niphargus fontanas, Spence Bate, Cat. Amphip. Brit. Mus. p. 175, pl. xxxii. fig. 2. 1863. Niphargus fontanus, Spence Bate & Westwood, Brit. Sessile-eyed Crust. i. p. 319. 1889. Gammarus fontanus, Moniez, Rey. Biol. du Nord de la France, i. p. 48. 1890. Miphargus fontanus, WrzeSniowski, Zeitschrift fiir wissenschaft- liche Zoologie, L. 4, p. 674. 1893. Niphargus puteanus, Della Valle, Gammarini del Golfo di Napoli, p. 704 (tn part). 1896. Gammarus puteanus, Hamann, Europaische Hohlenfauna, p. 254 (tn part). 1896. Niphargus fontanus, Walker & Hornell, Jour. Mar. Zool. & Micros. vol. ii. p. 54. Bate and Westwood describe this species as follows :— “Moderately robust. Coxe of the first four pairs of legs LINN. JOURN.—ZOOLOGY, VOL. XXVIII. iLL 154 DR. CHARLES CHILTON ON THE SUBTERRANEAN nearly as deep as the segments to which they are attached. Gnathopoda having the propoda pear-shaped, with the palm oblique, and as long as the superior margin, which is nearly straight; inferior margin convex and posteriorly produced ; palm defined by one or two movable spines. Anterior segments of the tail with the posterior angles pointed. Length 3 inch.” I have not yet been able to examine a specimen of this species, though Bate and Westwood record it from Ringwood in Hants, Corsham in Wiltshire, and High Elms in Kent. It appears to be readily distinguished from the other British species by the pear- shaped propoda of the gnathopods and by the oblique palm. It is given by Wrzesniowski as a good species. Specimens identified by Messrs. Hornell and Sinel as WV. fon- tanus have been found in two wells on the outskirts of St. Helier, Jersey, and are mentioned by Messrs. Walker and Hornell in their report on the Schizopoda &c. of the Channel Islands. T have endeavoured to trace these specimens or to obtain others from the same locality, but without success, although Mr. Hornell used his best efforts on my behalf. Genus Cranconrx, Spence Bate. Crangonyx, Spence Bate, Proceed. Dublin Univ. Zool. & Bot. Assoc. 1859, p. 240; Nat. Hist. Review & Quarterly Journal of Science, vi. p. 165; Cat. Amph. Crust. Brit. Mus. p. 178: Bate & Westwood, Hist. Brit. Sessile-eyed Crust. i. p. 326: Della Valle, Gammarini del Golfo di Napoli, p. 681: Vejdovsky, Sitzungsherichte der K6nigl-bohmischen Gesellschaft der Wissenschaften Prag Mathematisch-naturwissenschaft- liche Classe, 1896, x. p. 5. I have given above only the most important references bearing on this genus; a full historical account of it will be found in the one last given. Vejdovsky, however, has not given an amended diagnosis of the genus, and in order to avoid confusion, I shall not attempt to do so either, as that will, I presume, be done by Mr. Stebbing in his forthcoming account of the Amphipoda for “ Das Tierreich.’ It is closely allied to the genera Paracrangonyx and Eucrangonyx recently established by Mr. Stebbing. Boruta, Wrzesniowski, is also very nearly allied to Crangonyzx, and is indeed considered by Vejdovsky to be identical with it; while Goplana, Wrzes- niowski, is considered by him as merely asubgenus of Crangonyz, and the species Goplana polonica as a surface-species repre- AMPHIPODA OF THE BRITISH ISLES. 155 senting the form from which the subterranean species of Cran- gonyx in Europe is probably derived. Nearly allied species are found in the surface and underground waters of North America, and were formerly assigned to the genus Crangonyx, though most of them have now been placed in new genera by Mr. Stebbing. CRANGONYX SUBTERRANEUS, Spence Bate. (Plate 18. figs. 4.) 1859. Crangonyx subterraneus, Spence Bate, Proc. Dublin Univ. Zool. & Bot. Assoc. i. p. 240; Nat. Hist. Rev. vol. vi. p. 166, fig. 3. 1861. Crangonyx subterraneus, Hogan, Rep. British Assoc. (1860), pp. 166-169. 1862. Crangonyx subterraneus, Spence Bate, Cat. Amphip. Brit. Mus. p. 178, pl. xxxii. fig. 6. 1863. Crangonyx subterraneus, Spence Bate & Westwood, Brit. Sessile- eyed Crust. i. p. 327. 1890. Crangonyx subterraneus, Wrzesniowski, Zeitschrift fiir wissenschaft- liche Zoologie, L. 4, p. 697. 1893. Crangonyx subterraneus, Della Valle, Gammarini del Golfo di Napoli, p. 681. 1896. Crangonyx subterraneus, Vejdovsky, Sitz. kegl.-bdhm. Gesellschaft der Wissenschaften, 1896, x. pp. 3-32, pls. i. & ii. 1896. Gammarus puteanus, Hamann, Europaische Hohlenfauna, p. 234, (in part). 1899. Eucrangonyx Vejdovskyi, Stebbing, Trans. Linn. Soc., Zoology, ser. 2, vil. p. 423. Specific diagnosis.—First four side-plates nearly as deep as their respective segments, the fourth much the largest, being about twice as long as the third ; the lower margins of all convex and supplied with a few sete. Superior antenne about one-fourth the length of the body; the flagellum of about 12 joints; secondary appenaage of two slender joints, the first much longer than the second. Inferior antenne with the flagellum of 4 joints, the articula- tions between them slightly oblique. Mandible with the palp rather broad, its second joint half as broad as long, its inner margin being produced and convex. First gnathopod rather shorter than the second, carpus sub- triangular, much shorter than the propodos: the propodos subquadrate, length of anterior border one and a half times the breadth ; palm oblique, defined by a stout spine, and supplied along its length with peculiar sete split at the ends. The second gnathopod similar, but with the anterior border of propodos twice the breadth of the joint, and the palm rather more oblique. 156 DR. CHARLES CHILTON ON THE SUBTERRANEAN Basa of the last three pairs of pereiopoda broad, with the pos- terior margin expanded, convex, greatest breadth three-fourths of the length. Third uropoda reaching only slightly beyond the extremity of the preceding pair; the outer branch one-jointed, twice as long as the peduncle, inner branch rudimentary, minute. Telson reaching well beyond the end of the peduncle of the third uropod, hinder margin somewhat emarginate, lateral angles with 2 or 3 sete. Length about 4 mm. Habitat. Ringwood in Hampshire, and Marlborough in Wilt- shire. (In Europe recorded from Munich and Prague.) This species was first described by Spence Bate from a single specimen found at Ringwood. From the figures given by de Rougemont [17. pl. i. figs. 1 & 2, and pl. ii. fig. 1], it appears that he really had before him either the present species or a closely allied one of Crangonyx; but his account of it adds nothing of importance to what was already known, and his asser- tion that it was merely a young stage of Niphargus puteanus has been already criticised by Humbert and Wrzesniowski and shown to be erroneous. The species does not appear to have attracted further attention till 1896, when Vejdovsky published his important paper [21] and dealt exhaustively with its sense- organs and internal anatomy. In his revision of the Gammarida, Stebbing [31. p. 423] gave to the species described by Vejdovsky the new name Hucrangonyx Vejdovskyi, saying that it appeared to him to differ from Spence Bate’s species: the points of differ- ence to which he refers apparently being the possession of a small wmner branch to the terminal uropoda and of an emargination in the telson. I have received from Mr. Stebbing a small mounted specimen of Crangonyx subterraneus from Marlborough, and have been able to compare it with mounted specimens of Vej- dovsky’s species ; and after careful examination I have no doubt that both belong to the same species, and Professor Vejdovsky, who made a hasty comparison of the same specimens when I visited him in Prague in March last, was of the same opinion. In all points that can be observed the different specimens seem to be practically identical. The specimen from Marlborough is very small, and in its present mounted condition itis not possible to be quite certain whether it possesses a small inner branch to the terminal uropoda or not, though I think it has; but this AMPHIPODA OF THE BRITISH ISLES. 157 branch as figured by Vejdovsky is very small and delicate, and can no longer be seen in his specimens now that they are mounted in Canada balsam, but the rest of the uropod is so similar that I think the fine inner branch must be present in the Marlborough specimen also. In this specimen, again, the telson can be seen in side-view only, but it is evident that the two hinder angles project a little and bear set, as shown by Vejdovsky, and I think the posterior edge between them is slightly emarginate. In connection with this point it is perhaps worth while drawing attention to the fact that Jurinac has described considerable differences in the telson in the two sexes of his Wiphargus croaticus, the male having the telson roundly indented (rather than cleft) for a third of its length only, while in the female the telson is sharply cleft to two-thirds of its length [37. pp. 12, 15, & 16, and pl. i. figs. 3 & 12]. In Cran- gonyx mucronatus, Forbes, again, the differences of the telson in the two sexes are quite startling; and I have recorded the fact that in Calliopius subterraneus, Chilton, the telson is rather more deeply emarginate in the male than in the female. The lower antenna is shown in figure 4,a.2. It was wanting in Spence Bate’s specimen ; the articulations between the joints of the peduncle are slightly oblique, just as in Cragonyx com- pactus, Chilton. The mouth-parts, so far as I have been able to examine them, are in minute agreement with those of the Prague specimens, the mandibles, second maxille, and maxillipedes being practically identical; the first maxillze I have not seen. The first gnathopod was described by Spence Bate as being larger than the second; and this character was made use of by Della Valle as a specific character for separating Crangonyx subterraneus, Bate, from C. compactus, Chilton. Vejdovsky has rightly shown that really the two gnathopoda differ only slightly in size, and that the second is actually the longer, though from his figures it appears that it is scarcely so stout in proportion ; the same thing is true of the Marlborough specimen, as may be seen from a comparison of figs. 4, gn.' and 4, gn... For an account of the peculiar split sete on the propoda of the gnathopoda, reference must be made to Vejdovsky’s paper ; they are present in the Marlborough specimen, but the specimen is very small and the details of them cannot now be made out. The large size of the fourth side-plate is a peculiar character, 158 DR. CHARLES CHILTON ON THE SUBTERRANEAN and was first pointed out by Vejdovsky ; the lower margins of the first four side-plates are supplied with a few sete, each arising from a slight indentation of the margin; this has been rather exaggerated in his figure 1, so as to make it appear as if the margins were actually crenate. His fig. 3 is more correct, though even here there is rather more crenation than in the specimen. Accessory branchiew have been described by Vejdovsky on the last three segments of the pereion; similar appendages had previously been described by Wrzesniowski in Boruta and Goplana. BIBLIOGRAPHY. 1. Spence Batre & Westwoop.—British Sessile-eyed Crustacea, vol. i., London, 1863. 2. T. R. R. Svessrne.—Report of Corresponding Societies’ Committee, British Association Report, Dover, 1899. 3. W. EH. Leacn.— Crustaceology: Appendix.’ The Edin- burgh Encyclopedia, vol. vii. 1814, pp. 429-437. 4, P. Gervats.— Note sur deux Espéces de Crevettes qui vivent aux environs de Paris.” Annales des Sciences naturelles, Zool. 2 sér. iv. (1835). 5. K. L. Kocu.—Deutschlands Crustaceen, Myriapoden und Arachniden. Regensburg, 1835-1841. 6. Ropert Caspary.— Gammarus puteanus.” Verhandlungen des naturhist. Vereins der preussischen Rheinlande und Westfalen, 6 Jahrg., Bonn, 1849, pp. 39-48, Taf. 11. 7. Avaust Hostus.—‘ Uber die Gammarus-Arten der Gegend von Bonn.” Archiv fir Naturgeschichte, 15 Jahrg. 1850, Bd. i. pp. 283-248, Taf. i. u. il. 8. J. C. Scuioprx.— Bidrag til den underjordiske Fauna.” Det danske Videnskabernes-Selskabs Skrifter, 5e Raekke. Naturvidenskabelig og Mathematisk Afdeling, Bd. ii., Kjo- benhavn, 1851. 9. J. O. Westwoov.—Proe. Linn. Soc., No. li. p. 218 (1853). 10. Apam Wuitrt.—Popular History of British Crustacea. London, 1857. 11. C. Spnnce Batz.—‘ On the Genus Miphargus.” Dublin 12. 13. 14. 15. 16. 1%. 18. 19. 20. 21. 22. 23. 24. 25. 26. AMPHIPODA OF THE BRITISH ISLES. ‘159 University Review, vol. i. (1859) p. 237; also in Natural History Review, vol. vi. (1859) p. 163. A. R. Hoaan.— On the Habits of Niphargus fontanus (n. 8.), NV. Kochianus (n. s.), and Crangonyx subterraneus (n. g. & s.), Spence Bate.” Nat. Hist. Review, vi. 1859, pp. 166- 169. C. Spence Bare.—Cat. Amphipoda Brit. Mus. London, 1862. 8. F. Harmer.—“ On the Occurrence of the ‘ Well-shrimp,’ Niphargus, near Norwich.” Trans. Norfolk & Norwich Naturalists’ Society, vol. vi. pp. 489-491. VatertTe St. Grorer.—‘ DeGammaro puteano.” Dissertatio Inauguralis, 1857. Gustav JosEPpH.—‘‘ Arthropoden der Krainer Tropfstein- grotten.” Berliner entomologische Zeitschrift, Bd. xxv. 1881, Heft i. P. pe Rovgemont.—Etude de la Faune des Eaux privées de Lumiécre. Paris, 1876. Atois HumBert.—“ Description du Niphargus puteanus, var. Foreli.” Bull. de ia Soc. vaudoise des Sciences naturelles, iv. (1876) pp. 278-398, pls. 6 & 7. R. Montrz.—“ Faune des Haux souterraines du Département du Nord.” Rev. biolog. du Nord de la France, i. (1888- 1889). A. Wrzesntowsk1.—Uber drei unterirdische Gammariden.” Zeitschrift fiir wissenschaftliche Zoologie, L. 4, pp. 600- 725, pls. XxXVill.—-xxxil. F. Vespovsxy.— Ueber einige Siisswasser-Amphipoden.” Sitzungsberichte d. Kg.-bohm. Gesellschaft d. Wissen- schalten, Prag, 1896, x. K. CoEvREvx.—“ Sur un Amphipode d’Eau douce, Miphargus Virei, nu. sp., provenant des grottes du Jura.” Bull. Mus. Hist. nat. Paris, 1896, no. 4. Armand Vir&.—La Faune souterraine de France. Paris, 1900. E. D. Corr.—“ On the Wyandotte Cave and its Fauna.” American Naturalist, vi. July 1872, pp. 1-17. A.S. Packarp.—*‘ The Cave Fauna of North America.” Nat. Academy of Sciences, iv., First Memoir. J. E. Benepicr.—“ erento description of a new Genus 160 DR. CHARLES CHILTON ON THE SUBTERRANEAN and three new Species of Crustaceans from an Artesian Well at San Marcos, Texas.” Proc. U.S. Nat. Mus. xviii. No. 1087. 27. S. A. Forpes.— List of Illinois Crustacea.” Bull. Illinois Mus. Nat. Hist. i. 6 (1876). 28. C. Curtton.—The Subterranean Crustacea of New Zealand.” Trans. Linn. Soc., 2nd ser. Zoology, vi. part 2. 29. G. M. THomson.—“ Note on Tasmanian Crustacea.” Proc. Roy. Soc. Tasmania, 1892. 30. O. A. Saycr.—‘ On a new Species of Miphargus from Aus- tralia.” Proc. Roy. Soc. Victoria, xii. pt. 1. 31. T. R. R. Sressine.—‘ Amphipoda from the Copenhagen Museum and other sources,” Part ii. Trans. Linn. Soc., 2nd ser. Zool. vii. part 8 (1899). 32. Detia Vatte.—“ Gammarini del Golfo di Napoli.” Fauna & Flora, Monograph xx. (1898). 33. O. Hamann.—Europiische Hohlenfauna. Jena, 1896. 34. F. Vespovsky.—“ Ueber einige Siisswasser-Amphipoden, II. Zur Frage der Augenrudimente von Miphargus.” Sitzungs- berichte d. Kg.-béhm. Gesellschaft der Wissenschaften, Prag, 1900. 35. A. S. Packarp.—“ On the Origin of the Subterranean Fauna of North America.” American Naturalist, xxviii. pp. 727- 751. 36. G. H. Carpenter.—“ Animals found in the Mitchelstown Cave.” Irish Naturalist, iv. pp. 25-35. 37. A. E. Jurtinac.—“ Ein Beitrag zur Kenntniss der Fauna des Kroatischen Karstes und seiner unterirdischen Hohlen.” Inaugural Dissertation, Munchen, 1888. EXPLANATION OF THE PLATES. a.s. = superior antenna. ai. = inferior antenna. md. = mandible. mxp. = maxillipede. gn. = first gnathopod. gn.” = second gnathopod. prp.2 = first pereiopod. ur. = terminal portion of pleon with uropoda. T. = telson. Chilton. Linw Soc. Journ. Zoon. Vor XXVIII Pr 16. C, Chilton del. M.PParker lith. Geo, West & Sons imp. 1. NIPHARGUS SUBTERRANBDUS, Leach. Chilton. Linw Soc. Journ. Zoou. Vou. XXVIII Pr. 17, C. Chilton del. MP Parker lith, Geo, West & Sons imp. 1. NIPHARGUS SUBTHRRANEUS, Leach. Ge uy KOCHIANUS, Spence Bate. Chilton, Linn Soc. Journ. Zoon. Vor XXVIII Pr 18. C.Chilton del. MPFParker lith, Geo. West & Sons imp Gs, NPA CUS KOCHIANUS, Spence Bate. 4, CRANGONGA SUBTERRANEUS, Spence Bate. AMPHIPODA OF THE BRITISH ISLES. 161 Puate 16. Fig. 1. Miphargus subterraneus, Leach. The sex of the specimen from which each drawing was made is indicated by the usual sign. Puats 17, Fig. 1. Miphargus subterraneus, Leach. 2. Niphargus Kochianus, Spence Bate, drawn from a specimen from Dublin. Prare 18. Fig. 2. Niphargus Kochianus, Spence Bate, Dublin specimen. 3. Ditto, from a Marlborough specimen. 4, Crangonyx subterraneus, Spence Bate. (All the figures considerably magnified.) LINN. JOURN.—ZOOLOGY, VOL. XXVIII. 11 AMPHIPODA OF THE BRITISH ISLES. 161 Pvate 16. Fig. 1. Niphargus subterraneus, Leach. The sex of the specimen from which each drawing was made is indicated by the usual sign. Puate 17. Fig. 1. Mphargus sudterraneus, Leach. 2. Niphargus Kochianus, Spence Bate, drawn from a specimen from Dublin. Prats 18, Fig. 1. Miphargus Kochianus, Spence Bate, Dublin specimen. 2. Ditto, from a Marlborough specimen. 3. Crangonyx subterraneus, Spence Bate. (All the figures considerably magnified. ) ForaMInirera from the Lagoon at Funafuti. By Freprerick Cuapmay, A.L.S., F.R.M.S. [Read 6th December, 1900.] (Piates 19 & 20.) INTRODUCTION. By the study of Foraminifera under abnormal or adverse conditions one is able to learn much that is of biological interest, and more of the real value of the so-called species, than by the mere examination of samples which have been obtained from ordinary extensive littoral or deep-sea deposits, formed under conditions more or less favourable. So far as I know, we have never before had so good an oppor, tunity of examining a foraminiferal lagoon fauna in detail. The present paper is based on the material dredged by Messrs. G. Halligan and A. E. Finckh across the lagoon of Funafuti; and this was courteously sent to us by Prof. T. Edgeworth David, of Sydney University, under whose direction the dredgings were made. | The dredgings were taken along a line running due E. and W. starting from below the Mission Church on Fongafale Islet, at intervals of half-a-mile, until the opposite rim of the Atoll was LINN. JOURN.— ZOOLOGY, VOL. XXVIII. 12 162 MR. F. CHAPMAN ON FORAMINIFERA reached at Fuafatu Islet *. These dredgings were sent over in 6 oz. capsuled bottles and numbered consecutively 1 to 18. The starting point for the dredgings, Fongafale, presents an unbroken rim; but on the contrary Fuafatu, on the opposite side of the Atoll, is situated on the part of the rim which is broken up by channels. These diverse features afford most interesting comparisons of the foraminiferal faunas at the two localities; for at the former place the Foraminifera have lived under more or less tranquil conditions; whereas those of the latter locality were subjected to the action of the currents and surf, and consequently had an abundant food-supply, and this could not tail to make a decided difference in the assemblage found there. Although the lagoon of Funafuti is open to the sea by many broad channels, chiefly on the 8.E. and N.W., the area itself is so large, about 9 miles along the line of the soundings, that we might well expect to find a facies in the middle of the lagoon differing from that near the rim of the Atoll. The results obtained not only bear out this expectation, but they offer many new points with regard to the habits and pre- ference of conditions exhibited by such lowly forms as the Rhizopoda, far exceeding anything we had anticipated. The present collection is a remarkable object-lesson on the importance of environment in producing variations ; and provided these conditions remain fixed for lengthened periods, the varieties may converge to definite centres of form, and come to be regarded as species. The simplicity of the Rhizopod-structure and the almost infinite adaptability of these animals to their environ- ment enable us to point to innumerable connecting-links or intermediate forms, the despair of the systematist, which afford the biologist material for some important deductions and a general comparative study. The present work has been chiefly carried out in the Geological Laboratory of the Royal College of Science; and I am indebted to Prof. Judd, C.B., F.R.S., for the facilities he has there given me. * See Map of Funafuti Atoll, p. 167. FROM THE LAGOON AT FUNAFUTI. GENERAL ANALYSES OF THE SAMPLES. 163 The composition of the 18 samples of dredgings from the Lagoon are as follows :— Distance ample| from {Depth in Suda No. | Mission | fathoms. Ione Church. 47 per cent.—Chiefly Milio- lina, Orbitolites, Textularia, Ue mile 10 Calcarina, Amphistegina, and Heterostegina. 23 p. e.—Chiefly Orbitolites, c a Calcarina, Textularia, Am- mee eie. | 15s phistegina, and Hetero- | stegina. ‘5 p. e.—Chiefly Sagenina, a 13 miles. | 20 Amphistegina, and Hetero- stegina. 20 p. e.—Chiefly Orbitolites, Planorbulina, Gypsina, Cal- m-.| 2 miles. 23 carina, Aimphistegina, and Fleterostegina. 14 p. e.—Chiefly Sagenina, ie. 23 miles. 24 Gypsina, Amphisteqina, and Heterostegina. 9 p. e.—Chiefly Sagenina, : Calcarina, Amphistegina, ie:.| 3 miles. 21 anes TE, 14 p. e.—Chiefly Sagenina, Gx aa Bye Gypsina, Amphistegina, Sg Bules. ms and Heterostegina. 2 p. c.—Chiefly Sagenina, : Gypsina, Nonionina, Am- o.. : 2 ta oe p aes e phistegina, and Heteroste- gina. . 1 p. e.—Sagenina, Bolivina Devos a D5 Pp LY ’ 2 | 42 miles. Te Amphistegina, and Hetero- stegina. Other Organisms. Halimeda 3 per. cent. — Lamelli- branchs, Gastropods, Alcyonarian spicules, Hehinoderma, Serpulz, Ostracoda and other Crustacea. 2 p. e. —- Echimoderma, Polyzoa, Lamellibranchs, Gastropods, Serpulz, Ostracoda. ‘Dp. c.—Lamellibranchs, Gastropods, and Ser- pulee. 4 p. c. — Hchinoderma, Polyzoa, Lamellibranchs, Gastropods, Serpulz, Ostracoda and other Crustacea. 1p. ¢. — Hchinoderma, Polyzoa, Lamellibranchs, Gastropods, Serpulz, and | Crustacea. 1 p. ec. — Echinoderma, Polyzoa, Brachiopods, Gastropods, _Lamelli- branchs, Serpulz, Ostra- codaand other Crustacea. 1 p. e.—Pteropods, Poly- Z0a, Lamellibranchs, Gastropods, Serpule, and Ostracoda. 1 p. e.—Sponges, Gastro- pods, Lamellibranchs, Serpulz, and Crustacea. 1 p. c.—Serpulz, Lamelli- branchs, Gastropods. 12* 50 per cen Monee: 90 p. ¢. 85 p. ¢. 164 Distance ample| from No. Mission Church. Ores; 5 miles. die cces 5% miles. Ms sae 6 miles. Dr acsas 63 miles. Ae. i, 7 miles. Deisaisee 73 miles. (Giieees 8 miles. ieee 84 miles. iS} aes 9 wiles. ER Uc MR. F. CHAPMAN ON FORAMINIFERA Depth in fathoms. 16 19 20 ( |90 p.¢.—Chiefly Spirolocu- 12 toh TABLE (continued). Foraminifera. Other Organisms. 12:5 p. e.—Chiefly Sagenina, Haddonia, Amphistegina, and Heterostegina. 12°5 p. c—Chiefly Had- donia, Carterina, Gypsina, Polytrema, Nonionina, and Amphistegina. 1 p. e.—Sagenina, Gypsina, Polytrema, and Amphiste- gind. 9 p. ¢.—Chiefly Carterina, Discorbina, Gypsina, Poly- trema, and Amphistegina. 9 p. e.—Chiefly Orbitolites, Haddonia, Truncatulina, Calearina, Gypsina, Amphi- stegina, and Heterostegina. 9 p.c.—Chiefly Textularia, Calcarina, Gypsina, Amphi- stegina, and Heterostegina. 9 p. e.—Chiefly Sagenina, Placopsilina, Carterina, Calcarina, Gypsina, Poly- trema, and Amphistegina. lina, Miliolina, Hauerina, Peneroplis, Orbitolites, Textularia, Verneuitina, Globigerina, Cymbalopora, Discorbina, Calcarina, Gyp- sina, Polytrema, Amphiste- gina, and Heterostegina. 98 p. ¢e.—-Chiefly Spirolocu- lina, Miliolina, Hauerina, Peneroplis, Textularia, Ver- neuilina, Spirillina, Cym- balopora, Discorbina, Trun- catulina, Tinoporus, Poly- trema, Calcarina, Polysto- mella, Amphistegina, and Heterostegina. 2°5 p. e.—Serpule, Gas- tropods, Lamellibranchs, Polyzoa, Pteropods, Os- tracoda and other Crust- acea. 2°5 p. ¢e.—LHchinoderma, Polyzoa, Lamellibranchs, Gastropods, and Crust- acea. 1 p. e.—Polyzoa, Gas- tropods, and Serpulze. 1 p. c.—Polyzoa, Gastro- pods, and Serpule. 1 p. ec. — Hchinoderma, Lamellibranchs, Gastro- pods, Heteropods, Ser- pulz, and Ostracoda. 1 p. ¢.—Lamellibranchs, Gastropods, and Ser- pul. 1 p. e.—Polyzoa, Lamelli- branchs, Gastropods, and Serpule. 1 p. c.—Sponge spicules, Aleyonarian spicules, Echinoderma, Brachio- pods, Lamellibranchs, Gastropods, and Ostra- coda. 1 p. e.—Sponge spicules, Aleyonarian __ spicules, Echinoderma, Hetero- pods, and Ostracoda. Halimeda. 90 p. ¢. 90 p. ¢. 1 p.¢. FROM THE LAGOON AT FUNAFUTI. 165 By a reference to the foregoing Table, we find that the pre- vailing genus of the Foraminifera under all the varying conditions is Amphistegina. The specimens of this genus which are found in the middle of the lagoon are, as a rule, much smaller than those found within reach of the influence of marine currents from the outer side of the reef, whether through channels in the rim or through the submarine interstices of the reef-platform. The spurred forms Calcarina and Tinoporus (slightly resembling each other in general form but not related) are both common near the rim at Fuafatu Islet, and on the opposite side of the lagoon at Fongatale Islet ; but the latter genus rapidly dis- appears * on travelling across the lagoon, and Calcarina only is found in the various samples taken from the lagoon-floor. The number of genera of the Foraminifera found close to the lagoon-shore, near the Mission Church at Fongafale, is 21; but on the opposite side, at Fuafatu, the number is increased to 28. This is obviously owing to the greater influence of marine con- ditions and food-supply from the seaward face. At Fongafale this seems to be due to the passage of water through the reef- platform, and at Fuafatu to currents passing through channels in the rim of the Atoll or between the islets. — In the middle of the lagoon only three genera are present, namely, Sagenina, Amphistegina, and Heterostegina. Whilst the greater number of generic forms gradually die out towards the centre of the lagoon, a few, with some special varieties, make their appearance and in tolerable abundance, in consequence of the more tranquil conditions prevailing there. A noteworthy form in this respect is Carterina spiculotesta, whose tiny brown tests are usually found attached to Halimeda-fronds. The thin papery form Gypsina vesicularis var. squamiformis, and the adherent Planorbulina larvata are also found under similar conditions. PERCENTAGE COMPOSITION OF THE FoRAMINIFERAL SAMPLES. A Chart of the percentage composition of the samples with regard to the Foraminifera is given below. By this we see the almost equal proportion of Halimeda and the Foraminifera occurring near the rim of the Atoll at Fongafale Islet. There is * A solitary specimen of Tinoporus baculatus, much worn, and probably carried in by currents, was found in Sample 11, 3} miles from the rim of the Atoll. 166 ME. F. CHAPMAN ON FORAMINIFERA N° OF PERCENTAGE OF FORAMINIFERA DEPTH IN SAMPLE 10. 20 30 40 DOp OO mEaHO, 80 90 FATHOMS, Ss aS) yy i=) S (or SS co ey es NS i ~ XS ~ ~~ PERCENTAGE OF FoRAMINIFERA.—Readings to the left of the zigzag line give percentages of Foraminifera; to the right, percentages of Halimeda and other miscellaneous organisms. FROM THE LAGOON AT FUNAFUTI. 167 then a sudden drop 14 miles out to ‘5 per cent. of Foraminifera, followed by an oscillation along the floor of the lagoon, and in no wise dependent upon depth. The percentage of Foraminifera OIG Z, SWulite fala t 7 Fatato |. SFunangongo t. Funamanu |. 2H Falefatu 1. Avalau | _Afotutoa t. FunaFrutrt ATotL,—Showing the principal Islets and the positions whence the Lagoon samples were taken, here falls as low as 1 and never rises above 20. This low per- centage is persistent until within 1 mile of the opposite rim rear Fuafatu Islet, where there is a sudden rise to 90 per cent., whitch is further increased within 4 mile of the shore to 98 per cent. 168 MR. F. CHAPMAN ON FORAMINIFERA DEscRIPTION OF THE FORAMINIFERA *. Family MILIOLIDZ. Subfamily NUBECULARIING. Nousecuraria, Defrance [1825]. NvUBECULARIA DivaRIcaTA, Brady. (PI. 19. fig. 1.) Sagrina divaricata, Brady, 1879, Quart. Journ. Mier. Sci. n.s., vol. xix. p- 276, pl. viii. figs. 22-24. Nubecularia divaricata, Brady, 1884, Rep. Chall. vol. ix. p. 186, pl. Ixxvi. figs. 11-15. N. divaricata, Brady, Millett, 1898, Journ. R. Micr. Soc. p. 261, pl. v. fig. 4. This rare form has previously been recorded from Humboldt Bay, Papua; off Raine Island, Torres Strait; off Tongatabu, Friendly Islands ; and lately by Millett from three stations in the Malay Archipelago. Our specimen closely resembles that of the figure given by Millett, and it also shows the phialine termination of the aperture characteristic of this species. Lagoon, Funafuti; very rare. Sample 18 (73 fathoms). NUBECULARIA LUCIFUGA, Defrance. Nubecularia lucifuga, Defrance, 1825, Dict. Sci. Nat. vol. xxv. p. 210, Atlas Zooph. pl. xliv. fig. 3. N. lucifuga, Defrance, Egger, 1893, Abhandl. bayer. Akad. Wiss., math.-phys. Cl. ii. vol. xvii. p. 250, pl. xxi. figs. 4-7. N. lucifuga, Defrance, Millett, 1898, Journ. R. Micr. Soc. p, 261, pl. v. fig. 7. A rare form in tropical areas. Recorded from Tongatabu, Friendly Islands; Malay Archipelago, etc. It is represented at Funafuti chiefly by spirally-wound specimens, originally attached to other organisms. Lagoon, Funafuti; generally rare. Sample 1 (10 fathoms); sample 13 (26 fathoms) ; sample 17 (12 fathoms). * The synonymy of the species is here restricted chiefly to references subsequent to Brady’s ‘Challenger’ Report (1884). FROM THE LAGOON AT FUNAFUTI. 169 Nvusecunaria Brapyi, Millett. (Pl. 19. fig. 3.) Nubecularia inflata, Brady, 1884, Rep. Chall. p. 135, pl. i. figs. 5-8. N. Bradyi (N. inflata preoccupied), Millett, 1898, Journ. R. Micr. Soc. p. 261, pl. v. figs. 6 a, 6. This species usually occurs in quite shallow water. It was obtained by the ‘ Challenger’ off Honolulu, Sandwich Islands ; off Tongatabu ; off Tahiti; Nares Harbour, Admiralty Islands; and Balfour Bay, Kerguelen Island. It is common in the Greek Archipelago, and Millett found it in the Malay Archi- pelago. Our specimens do not show the fistulose or spouted tendeney which the aperture assumes, shown in the specimens figured by Brady and Millett. Lagoon, Funafuti; very rare. Sample 1 (10 fathoms); sample 18 (74 fathoms). NUBECULARIA LACUNENSIS, sp. nov. (PI. 19. fig. 2.) Test porcellanous, free, or (?) attached ; consisting of a series of compressed and concave segments somewhat cuspid at the peripheral margin, and arranged in an obscurely rotaline manner. Superior face conical, depressed ; inferior, flat. Diameter °55 millim. Lagoon, Funafuti; very rare. Sample 18 (73 fathoms). Subfamily MILIOLININ&#. Britocuttna, d’ Orbigny [1826]. BILOCULINA OBLONGA, @ Orbigny. Biloculina oblonga, d’Orbigny, 1839, in Sagra’s Cuba, Foram. p. 165, pl. viii. figs. 21-23. This species has the aboral end considerably swollen, and the inverted aperture of the penultimate chamber very apparent, thus differing from B. elongata, d’ Orbigny, which is more regularly oval. D’Orbigny’s specimens came from the neighbourhood of the Cuban coral-reefs. Lagoon, Funafuti; very rare. Sample 8 (26 fathoms) ; sample 18 (73 fathoms). BILOCULINA SUBSPH ERICA, @ Orbigny. Biloculina subspherica, @Orbigny, 1839, in Sagra’s Cuba, Foram. p. 162, pl. viii. figs. 25-27. 170 MR. F. CHAPMAN ON FORAMINIFERA This species was also described from the coral area around Cuba. Lagoon, Funafuti; very rare. Sample 18 (74 fathoms). BILOcULINA RINGENS (Lamarck). Miliohites ringens, Lamarck, 1804, Ann. du Muséum, vol. y. p. 351 ; vol. ix. pl. xvii. fig. 1. Liloculina ringens (Lam.), Egger, 1893, Abhandl. bayer. Akad. Wiss., math.-phys, Cl. ii. vol. xviii. p. 220, pl. i. figs. 7-9. B. ringens (Lam.), Millett, 1898, Journ. R. Micr. Soc. p. 262. B. ringens (Lam.), Flint, 1899, Rep. U.S. National Mus. (1897) p. 294, pl. 39. fig. 2. A widely distributed species both geographically and as regards depth of water. Millett found it generally distributed in the gatherings from the Malay Archipelago. Lagoon, Funafuti; very rare. Sample 18 (73 fathoms). Sprrotocurina, d’Orbiqny [1826]. SPIROLOCULINA ROBUSTA, Brady. Spiroloculina robusta, Brady, 1884, Rep. Chall. vol. ix. p. 150,fpl. ix. figs. 7, 8. S. robusta, Brady, Flint, 1899, Rep. U.S. Nat. Mus. (1897) p. 296, pl. 42. fig. 1. This species was found by Dr. Brady in the dredgings near Culebra Island, West Indies, at a depth of 390 fathoms. The specimen from Funafuti is characteristic but small. Lagoon, Funafuti; very rare. Sample 18 (74 fathoms). SPIROLOCULINA EXCAVATA, d’Orbigny. Sprroloculina excavata, V@Orbigny, 1846, Foram. Foss. Vienne, p. 271, pl. xvi. figs. 19-21. S. excavata, VOrb., Rupert Jones, 1895, Pal. Soc. Mon., Crag Foram. p. 106, pl. v. fig. 2; woodcuts, figs. 2 a, b. S. excavata, VOrb., Millett, 1898, Journ. R. Micr. Soc. p. 264. S. excavata, dOrb., Flint, 1899, Rep. U.S. Nat. Mus. (1897) p. 296, pl. 41. fig. 5. Our specimen is small and very thin-shelled, somewhat resembling S. impressa, Terquem, but quite regular in shape. Lagoon, Funafuti; very rare. Sample 14 (16 fathoms). FROM THE LAGOON AT FUNAFUTI. 171 SPIROLOCULINA NITIDA, d’Orbigny, var. FOvEOLATA, Egger. (Pl. 19. fig. 4.) Spiroloculina foveolata, Egger, 1893, Abhandl. bayer. Akad. Wiss., math.-phys. Cl. ii. vol. xviii. p. 224, pl. i. figs. 33, 34, Sprroloculina nitida, d’Orb. (reticulate variety), Millett, 1898, Journ. R. Mier. Soc. p. 266. This is a very interesting variety which has been recorded from the Mauritius, the Malay Archipelago, and the S. Pacific. It is well developed at Funafuti, but in the external form perhaps more nearly resembles S. planulata (Lam.). Lagoon, Funafuti; common. Sample 18 (74 fathoms). SPIROLOCULINA TORTUOSA, sp. nov. (PI. 19. fig. 5.) This curious form somewhat resembles S. lamella of Egger *, but is more regular in shape, and is of the S. ewcavata type. The sutural edges of the central series of chambers are strongly con- torted. The extremities of the test resemble those of S. impressa, Terquem. Length ‘6 millim. Lagoon, Funafuti; frequent. Sample 5 (24 fathoms); sample 15 (19 fathoms); sample 17 (12 fathoms). SPIROLOCULINA ANTILLARUM, d’ Orbigny. Spiroloculina antilarum, VOrbigny, 1839, in Sagra’s Cuba, Foram. p- 149, pl. xi. figs. 3, 4. The ‘Challenger’ collections yielded this species from only one station in the S. Atlantic, at 350 fathoms. Jt is rare at Funafuti, and appears to be nearly related tojthe form S. grata, Terquem. Lagoon, Funafuti; rare. Sample 18 (74 fathoms). SPIROLOCULINA G@RATA, Terquem. Spiroloculina grata, Terquem, 1878, Mém. Soc. Géol. France, sér. 3, vol. i. p. 55, pl. x. figs. 14, 15. S. grata, Terq., Egger, 1893, Abhandl. bayer. Akad. Wiss., math.- phys. Cl. ii. vol. xviii. p. 224, pl. 1. fig. 39. This is one of the most frequent species of the genus in shallow-water sands of the coral-reefs. The specimens found here are typical in form and size. Abhandl. bayer. Akad. Wiss., math.-phys. Cl. ii. vol. xviii. p. 223, pl. i. figs. 24, 25. 172 MR. F. CHAPMAN ON FORAMINIFERA Lagoon, Funafuti. Sample 1 (10 fathoms); sample 13 (26 fathoms) ; sample 17 (12 fathoms) ; sample 18 (74 fathoms). SPIROLOCULINA TENUISEPTATA, Brady. Spiroloculina tenuiseptata, Brady, 1884, Rep. Chall. vol. ix. p. 153, pl. x. figs. 5, 6. S. tenuiseptata, Brady, Egger, 1895, Abhandl. bayer. Akad. Wiss., math.-phys. Cl. ii. vol. xviii. p. 223, pl. i. figs. 48, 49. S. tenuiseptata, Brady, Millett, 1898, Journ. R. Micr. Soc. p. 265. The occurrence of the above species in the lagoon-material is further evidence that it occurs in shallow water as well as in deeper ; for Millett has lately recorded it from the Anchor Muds of the Malay Archipelago in depths of 12-14 fathoms. The ‘Challenger’ specimens were obtained from depths of more than 500 fathoms. The specimens from Funafuti are typical. Lagoon, Funafuti; very rare. Sample 1 (10 fathoms) ; sample 2 (153 fathoms). SPIROLOCULINA ACUTIMARGO, Brady. . Spiroloculina acutimargo, Brady, 1884, Rep. Chall. vol.ix. p. 154, pl. x. fies, 12-15. S. acutimargo, Brady, Balkwill & Wright, 1885, Trans. R. Irish Acad. vol, xxviii. p. 323, fig. 1. S. acutimargo, Brady, Egger, 1893, Abhandl. bayer. Akad. Wiss., math.-phys. Cl. ii. vol. xviii. p. 222, pl. i. figs. 26-28. S. acutimargo, Brady, Millett, 1898, Journ. R. Micr. Soc. p. 264. As a rule the Funafuti specimens are small, with exception of an exainple from sample 1. The species is well distributed over tropical areas, and occurs in shallow and deeper water alike. Lagoon, Funafuti; common in two places. Sample 1 (10 fathoms); sample 17 (12 fathoms); sample 18 (73 fathoms). Miniotina, Williamson [1858]. MILIOLINA VALVULARIS (feuss). Triloculina valvularis, Reuss, 1851, Zeitschr. deutsch. geol. Gesellsch. vol. iii. p. 85, pl. vil. fig. 56. Miliolina valwularis (Reuss), Brady, 1884, Rep. Chall. vol. ix. p. 161, pl. iv. figs. 4, 5. M. valvularis, Brady, Goés, 1894, Kong]. Svenska Vet.-Akad. Handl. vol. xxv. p. 115, pl. xxii. fig. 871. M. valvularis (Reuss), Millett, 1898, Journ. R. Micr. Soc. p. 501, pi. xi. figs. 5-7. FROM THE LAGOON AT FUNAFUTT. Wie M. valvularis (Reuss), Flint, 1899, Rep. U.S. Nat. Mus., Rep. for 1897, p. 299, pl. 44. fic. 5. The Funafuti specimens are very small and occur in the deeper part of the lagoon. Lagoon, Funafuti; very rare. Sample 10 (26 fathoms); sample 11 (25 fathoms). MILIOLINA CIRCULARIS (Bornemann). Triloculina circularis, Bornemann, 1855, Zeitschr. deutsch. geol. Gesellsch. vol. vii. p. 349, pl. xix. fig. 4. Miliohna circularis (Born.), Egger, 1893, Abhandl. bayer. Akad. Wiss., math.-phys. Cl. ii. vol. xviii, p. 235, pl. 11. figs. 61-63. M. circularis (Born.), Jones, 1895, Pal. Soc. Mon., Crag Foram. p. 121, pl. v. fig. 4. M. circularis (Born.), Millett, 1898, Journ. R. Micr. Soc. p. 499, pl. xi. figs. 1-3. OM. circularis (Born.), Flint, 1899, Rep. U.S. Nat. Mus., Rep. for 1897, p. 298, pl. 44. fig. 1. A species met with off Cuba and the coast of Australia, amongst other localities. Millett has lately recorded it from the Malay Archipelago. The biloculine form of the species occurs in the lagoon at Funafuti. Lagoon, Funafuti; specimens rather small. Sample 8 (26 fathoms) ; sample 18 (74 fathoms). Mixroina LaBiosa (d’ Orbiqny). Trilocuiina labiosa, d’Orbigny, 1839, in Sagra’s Cuba, Foram. p. 178, pl. x. figs. 12-14. Mihiohna labiosa (d’Orb.), Millett, 1898, Journ. R. Micr. Soc. p. 502, pl. xi. figs. 8, 9. M. labiosa (d’Orb.), Flint, 1899, Rep. U.S. Nat. Mus., Rep. for 1897, p- 299, pl. 45. fig. 3. This wild-growing form is not unfrequent in one of the samples from the Funafuti lagoon. It here appears to form a link between Miliolina valvularis and Nubecularia inflata. Lagoon, Funafuti. Sample 17 (12 fathoms); sample 18 (74 fathoms). MILIOLINA SUBROTUNDA (Jlontagqu). Vermiculum subrotundum, Montagu, 1803, Test. Brit. pt. 2, p. 521. Miliolina subrotunda (Mont.), Millett, 1898, Journ. R. Micr. Soc. p. 502. M. subrotunda (Mont.), Flint, 1899, Rep. U.S. Nat. Mus., Rep. for 1897, p. 299, pl. 44. fig. 6. 174, MR. F. CHAPMAN ON FORAMINIFERA The specimens found at Funafuti are characteristic. Lagoon, Funafuti. Sample 17 (12 fathoms); sample 18 (74 fathoms). MILIOLINA TRICARINATA (d’Orbigny). Triloculina tricarinata, VOrbigny, 1826, Ann. Sci. Nat. vol. vii. p. 299, No. 7. Miholina tricarinata (d’Orb.), Millett, 1898, Journ. R. Mier. Soe. p. 502. M. tricarinata (dOrb.), Flint, 1899, Rep. U.S. Nat. Mus., Rep for 1897, p. 298, pl. 44. fig. 4. The type form of this species is not common here, and varies somewhat in size. Lagoon, Funafuti. Sample 1 (10 fathoms); sample 4 (23 fathoms); sample 18 (73 fathoms). MILIoLINA TRICARINATA (d’ Orbd.), var. TERQUEMIANA, Brady. Miliolina Terquemiana, Brady, 1884, Rep. Chall. vol. ix. p. 166, pl. exiv. fis.. 1a, 0. M. tricarinata (d’Orb.), striate variety, Millett, 1898, Journ. R. Mier. Soe. p. 503, pl. xi. figs. 10, 11. This variety is distinguished by the numerous interrupted strie with which the surface of the test is covered. It was rare in the ‘Challenger’ collections, off Ceylon and the coast of Madagascar; but has been found not uncommon in the Malay Archipelago by Millett. Lagoon, Funafuti. Sample 17 (12 fathoms); sample 18 (74 fathoms). MILIoLina TRICARINATA (d Orb.), var. BERTHELINIANA, Brady. Miliolina Bertheliniana, Brady, 1884, Rep. Chall. vol. ix. p. 166, pl. exiv. fig, 2. M. tricarinata (d’Orb.), reticulated variety, Millett, 1898, Journ. R. Mier. Soe. p. 503, pl. xi. fig. 12. This shallow-water form is usually rare, and occurs in two samples only, near the rim of the atoll at Funafuti. Lagoon, Funafuti. Sample 1 (10 fathoms); sample 18 (7% fathoms). MILIoLINA RETICULATA (d’ Orbiqny). Triloculina reticulata, d’Orbigny, 1826, Ann. Sci. Nat. vol. vii. p. 299, No. 9. Miliolina (Triloculina) reticulata, d’Orbigny, Egger, 1893, Abhandl. bayer. Akad. Wiss., math.-phys, Cl. ii. vol. xviii. p. 289, pl. i1. figs. 83, 84. FROM THE LAGOON AT FUNAFUTI. 175 Quingqueloculina reticulata (dOrb.), Schlumberger, 1893, Mém. Soe. Zool. France, vol. vi. p. 214, fig. 25, & pl. ii. fig. 62. Miliolina reticulata (VOrb.), Flint, 1899, Rep. U.S. Nat. Mus., Rep. for 1897, p. 301, pl. 46. fig. 5. This species occurs not uncommonly at Funafuti. The speci- mens from the lagoon comprise both the triloculine and the quinqueloculine forms. Lagoon, Funafuti. Sample 1 (10 fathoms); sample 17 (12 fathoms) ; sample 18 (73 fathoms). Minionina UNDosA (Karrer). Quingueloculina undosa, Karrer, 1867, Sitzungsb. Ak. Wiss. Wien, vol. lv. p. 361, pl. iil. fig. 3. Miliolina undosa (Karrer), Egger, 1893, Abhandl. bayer. Ak. Wiss., math.-phys. Cl. ii. vol. xviii. p. 237, pl. il. figs. 41, 42. M. undosa (Karrer), Millett, 1898, Journ. R. Micr. Soe. p. 506, pl. xii. figs. 5 a-c. M. undosa (Karrer), Flint, 1899, Rep. U.S. Nat. Mus., Rep. for 1897, p. 300, pl. 45. fig. 4. This species varies considerably in size in different parts of the lagoon, and attains its largest dimensions in 12 fathoms near the lagoon-margin. Lagoon, Funafuti. Sample 1 (10 fathoms); sample 4 (283 fathoms); sample 17 (12 fathoms); sample 18 (73 fathoms). Mirrorina ParxKert, Brady. Miliolina Parkeri, Brady, 1884, Rep. Chall. vol. ix. p. 177, pl. vii. fig. 14. M. Parker?, Brady, Millett, 1898, Journ. R. Micr. Soe. p. 507, pl. xii. figs. 4a, 6. Although this form is widely distributed, it appears to be restricted to the neighbourhood of coral-reefs. Lagoon, Funafuti. Sample 1 (10 fathoms); sample 17 (12 fathoms); sample 18 (73 fathoms). Minrorina Frrussactt (d’ Orbigny). Quingueloculina Ferussacit, d’Orbigny, 1826, Ann. Sci. Nat. vol. Vii. p. 801, No. 18: Modéle No. 32. Miliolina Ferussacii (d’Orb.), Millett, 1898, Journ. R. Micr. Soe. p. 507, pl. xii. figs. 6a, 6, 7 a, 6, ¢. This species is very numerous in some samples from the lagoon, but is never found far away from the margin. The variations from the typical form are very many, and they often approach Spiroloculina in the expansion and thinning of the test along the 176 MR. F. CHAPMAN ON FORAMINIFERA median plane, by which the quinqueloculine form is reduced to a complanate series of chambers. Lagoon, Funafuti. Sample 1 (10 fathoms); sample 17 (12 fathoms) ; sample 18 (73 fathoms). MIiILioLiIna aG@LuTINaNs (d’ Orbigny). Quinqueloculina agglutinans, d’Orbigny, 1839, in Sagra’s Cuba, Foram. p. 168, pl. xii. figs. 11-18. Miliolina agglutinans (d’Orb.), Egger, 1898, Abhandl. bayer. Akad. Wiss., math.-phys. Cl. ii. vol. xviii. p. 239, pl. ii. fig. 55. M. agglutinans (d’Orb.), Goés, 1894, K. Svenska Vet.-Akad. Handl. vol. xxv. p. 110, pl. xix. fig. 848, pl. xx. fig. 849. M. agglutinans (d’Orb.), Millett, 1898, Journ. R. Micr. Soc. p. 508. M. agglutinans (d’Orb.), Flint, 1899, Rep. U.S. Nat. Mus., Rep. for 1897, p. 301, pl. 47. fig. 2. The above form occurs on both sides and at the extreme margins of the lagoon at Funafuti. It is somewhat curious that in Sample 1 the tests are of the WU. Ferussacit type, whilst on the opposite side of the atoll, im sample 18, they are all of the form of WZ. seminulum. Lagoon, Funafuti. Sample 1 (10 fathoms); sample 18 (74 fathoms). Mintonina Linnzana (d’Orbigny). Triloculina Linneiana, d’Orbigny, 1839, in Sagra’s Cuba, Foram. p. 172, pl. ix. figs. 11-18. Miliolina Linneana (dOrb.), Millett, 1898, Journ. R. Micr. Soe. p. 509. M. Linneana (d’Orb.), Flint, 1899, Rep. U.S. Nat. Mus., Rep. for 1897, p. 300, pl. 46. fig. 3. A single specimen only was found at Funafuti in the lagoon- dredgings, but it is of frequent occurrence in many of the beach- sands of the atoll. Lagoon, Funafuti. Sample 1 (10 fathoms). Minrotra Bicornts (Walker 5 Jacob). Serpula bicornis, Walker & Jacob, 1798, Adams’ Essays, Kanmacher’s ed. p. 633, pl. xiv. fig. 2. Miliolina bicornis (W. & J.), Egger, 1893, Abhandl. bayer. Akad. Wiss., math.-phys. Cl. ii. vol. xviii. p. 287, pl. ii. figs. 73, 74. M. bicornis (W. & J.), Flint, 1899, Rep. U.S. Nat. Mus., Rep. for 1897, p- 800, pl. 46. fig. 2. The specimens found here are not very typical. Lagoon, Funafuti. Sample 17 (12 fathoms); sample 18 (74 fathoms). FROM THE LAGOON AT FUNAFUTI, ilyire MILIOLINA ALVEOLINIFORMIS, Brady. Miliolina alveoliniformis, Brady, 1879, Quart. Journ. Mier. Sci., n. s. vol. xix. p. 54. M. alveoliniformis, Brady, Egger, 1893, Abhandl. bayer. Akad. Wiss., math.-phys. Cl. ii. vol. xviii. p. 232, pl. ii. figs. 17-19. This coral-reef species is restricted, in the lagoon-dredgings, to the margin. They are here small, some specimens being quite minute, measuring only ‘+5 mm., with the exception of a single individual from Sample 17 (12 fathoms), which is 2°3 mm. in length. Lagoon, Funafuti. Sample 1 (10 fathoms) ; sample 17 (12 fathoms ); sample 18 (74 fathoms). MILionina OBLONGA (Montagu). Vermiculum oblongum, Montagu, 1803, Test. Brit. p. 522, pl. xiv. fig. 9, Milholina oblonga (Montagu), Rupert Jones, 1895, Pal. Soc. Mon., Crag Foram. p. 120, pl. ill. figs. 31, 32, & pl. v. fig. 5. M. oblonga (Mont.), Millett, 1898, Journ. R. Micr. Soe. p. 267, pl. v. figs. 14a, 0. M. oblonga (Mont.), Flint, 1899, Rep. U.S. Nat. Mus., Rep. for 1897, p. 297, pl. 48. fig. 3. The specimen found is very elongate, but otherwise charac- teristic. Lagoon, Funafuti. Sample 4 (23 fathoms). Mittorina Boveana (d’ Orbigny). Quinqueloculina Boueana, VOrbigny, 1846, Foram. Foss. Vienne, p. 293, pl. xix. figs. 7-9. Miliolina Boueana (d’Orbigny), Goés, 1894, K. Svenska Vet.-Akad. Handl. vol. xxv. p. 114, pl. xxi. fig. 865. M. Boueana (VOrb.), Millett, 1898, Journ. R. Micr. Soe. p. 510. A solitary specimen, not quite typical, but nearest to the above species, was found in the lagoon-dredgings. Lagoon, Funafuti. Sample 1 (10 fathoms). Mitionina Bosciana (d’ Orbigny). (Pl. 1. fig. 7.) Quinqueloculina Bosciana, d’Orbigny, 1839, in Sagra’s Cuba, Foram. p. 191, pl. xi. figs. 22-24. Miliolina Bosciana (d’Orb.), Millett, 1898, Journ. R. Micr. Soe. p. 267, pl. vi. fig. 1. The specimens from Funafuti are of the smooth-shelled type, and their sutures are not so deeply impressed as in the specimens figured by Millett. LINN. JOURN.—ZOOLOGY, VOL. XXVIII. 13 178 MR. F. CHAPMAN ON FORAMINIFERA Lagoon, Funafuti. Sample 17 (12 fathoms); sample 18 (74 fathoms). MILIOLINA FUNAFUTIENSIS, sp. nov. (Pl. 19. fig. 6.) Test elongate, distinctly triangular, sometimes with a well- marked keel; aperture slightly prolonged. Surface with delicate, slightly oblique, vertical striations. Length ‘5 mm. The above form is of the type of JZ. oblonga and near M. Bosciana; it is distinguished chiefly by the® triangular contour in cross-section. The striations are finer and more regular than those seen in the striated variety of M. Bosciana*. Lagoon, Funafuti. Sample 1 (10 fathoms); sample 17 (12 fathoms); sample 18 (73 fathoms). MInrioLiIna SCROBICULATA, Brady. Miliolina scrobiculata, Brady, 1884, Rep. Chall. vol. ix. p. 173, pl. exiii. figs. 15 a-c. This rare form is probably a depauperated variety of IZ. bicornis. It was found originally im shore-sand from Madagascar, and at Nares Harbour at 17 fathoms (Brady). The individual found at Funafuti is characteristic but small. Lagoon, Funafuti. Sample 17 (12 fathoms). Subfamily HavUERINING|. Haverina, d@ Orbigny [1846]. HAUERINA ORNATISSIMA (Karrer). Quinqueloculina ornatissima, Karrer, 1868, Sitzungsb. Ak. Wiss. Wien, vol. lviii. p. 151, pl. iii. fig. 2. The specimens found in the lagoon are confined to the samples from near the margin. They are all very small, with one exception from Sample 17. Lagoon, Funafuti Sample 17 (12 fathoms); sample 18 (73 fathoms). Subfamily PENEROPLIDING. Cornuspira, Schultze [1854]. CoRNUSPIRA INVOLVENS, Reuss. Operculina involvens, Reuss, 1850, Denkschr. Akad. Wiss. Wien, vol. i. p. 370, pl. xlvi. fig. 20. * See Millett, Journ. R. Micr. Soc. 1898, p. 268, pl. vi. fig. 3 (Miliolina Bosciana, costate variety). FROM THE LAGOON AT FUNAFUTT. 179 Cornuspira wnvoluens, Reuss, 1863, Sitzungsh. Akad. Wiss. Wien vol. xlviii. p. 39, pl. i. fig. 2. C. mvolvens, Reuss, Flint, 1899, Rep. U.S. Nat. Mus., Rep. for 1897, p. 303, pl. 48. fig. 3. The above species is represented here by some rather dwarfed or immature examples. Lagoon, Funafuti. Sample 17 (12 fathoms) ; sample 18 (73 fathoms). PENEROPLIS, Montfort [1808]. PENEROPLIS PERTUSUS (Forskdl). Nautilus pertusus, Forskal, 1775, Descr. Anim. p. 125, no. 65. Peneroplis pertusus (Forskal), Mobius, 1880, Foram. yon Mauritius, p. 78, pl. iii. figs. 9-12, P. pertusus (Forskal), Flint, 1899, Rep. U.S. Nat. Mus., Rep. for 1897; p- 304, pl. 48. fig. 4. The forms of Peneroplis, usually regarded as varieties of a single species, are clearly related, but possibly do not show a more complete series of passage-forms than many other so-called species of Foraminifera. The specimens found at Funafuti fall into three groups-—namely, the nautiloid form with dendritine apertures, the compressed nautiloid and prolengthened form, and the delicate crosier-shaped type (subgenus Monalysidium). The first-named is by far the commonest form found at Funafuti. Lagoon, Funafuti. Sample 1 (10 fathoms); sample 17 (12 fathoms) ; sample 18 (73 fathoms). PENEROPLIS ARIETINUS (Batsch). Nautilus (Lituus) arietinus (pars), Batsch, 1791, Conch. See-sandes, p- 4, pl. vi. fig. 15, e. Peneroplis arietinus, Parker, Jones & Brady, 1865, Ann. & Mag. Nat. Hist. ser. 3, vol. xvi. p. 26, pl. i. fig. 18. P. pertusus (Forsk.), type c, Brady, 1884, Rep. Chall. vol. ix. p. 204, pl. xiii. figs. 18, 19, 22. The rectilinear portion of the series in our specimens some- times shows a tendency to curve either inward or outward. This type is evolved from an emaciated P. pertusus by the unrolling of the last coil or so of chambers. Lagoon, Funafuti. Sample 17 (12 fathoms); sample 18 (74 fathoms). igh 180 MR. F. CHAPMAN ON FORAMINIFERA Peneroriis (Monatystpium) (?) ponrrus, Chapman. (PI. 19. fig. 8.) Peneroplis (Monalysidium) politus, Chapman, 1900, Journ. Linn. Soc., Zool. vol. xxviii. p. 4, pl. i. fig. 5. The specimen now figured, from the Funafuti lagoon, is a weak, thin-shelled peneroplid, and would probably end in a lengthened series such as the specimen previously figured, which however had lost its initial series. Lagoon, Funafuti. Sample 18 (72 fathoms). PENEROPLIS (MONALYSIDIUM) CYLINDRACEUS (Lamarck). Spirolina (Spirolinites) cylindracea, Lamarck, 1804, Ann. du Muséum, vol. v. p. 245, no. 2. Peneroplis pertusus (Forsk.), type d, Brady, 1884, Rep. Chall. vol. ix. p. 305, pl. xiii. figs. 20, 21. The slender tests composed of a small spiral and a long chain of subglobular chambers are frequent in one sample from the lagoon. The surfaces of the chambers are relieved with delicate vertical striz, which under a high power can be resolved into rows of minute tubercles. Lagoon, Funafuti. Sample 17 (12 fathoms). Preneropiis (Monatysipium) Sounasi, Chapman. Penerophis (Monalysidium) Sollasi, Chapman, 1900, Journ. Linn. Soc., Zool. vol. xxviii. p. 3, pl. i. fig. 6. Fragments of the slender test of this species were found in the lagoon; they do not, however, exhibit the initial portion of the shell. Lagoon, Funafuti. Sample 17 (12 fathoms); sample 18 (74 fathoms). Orprrouites, Lamarck [1801]. ORBITOLITES MARGINALIS (Lamarck). (Pl. 20. figs. 1-8.) Orbulites marginalis, Lamarck, 1816, Hist. Nat. Anim. sans Vert. vol. ii. p- 196, no. 1. Orbitolites marginalis (Lam.), Carpenter, 1885, Phil. Trans. vol. clxxiv. p. 599, fig. 1. Q, marginalis (Lam.), Flint, 1899, Rep. U.S. Nat. Mus., Rep. for 1897, p- 304, pl. 50. fig. 2; pl. 51. fig. 1. This species is very characteristic of the shallow-water dredgings FROM THE LAGOON AT FUNAFPUTT. 181 of the Pacific Islands. Out of thirteen ‘Challenger’ Stations, eleven were from the Pacific. It has been found also in the West Indies, off Madagascar, in the Malay Archipelago, the Mediterranean, and the Red Sea. More than half the specimens from the lagoon at Funafuti appear to have been at one time attached by a face of the shell to various living organisms, but chiefly to joints of Halimeda. Some examples were actually found in position on these calcareous plants. This habit of Orbitolites attaching itself to foreign objects is not unknown, but it appears to have only been noticed hitherto in the species Orbitolites complanata. The specimens thus found upon other organisms are invariably irregularly discoidal, and are generally concavo-convex in shape. They also have patches of redundant shell-growth scattered over the surfaces, sometimes arranged in radial strigille around the peripheral area. Lagoon, Funafuti. Sample 1 (10 fathoms), some of the specimens beautifully smooth and regular, others concave or inflexed and with redundant patches of shell-growth on the surfaces (attached varieties); sample 2 (154 fathoms), very irregular and inflexed, specimens large ; sample 3 (20 fathoms), irregular specimens; sample 4 (23 fathoms), irregular speci- mens; sample 5 (24 fathoms), irregular specimens; sample 7 (24 fathoms), very irregular specimens ; sample 8 (26 fathoms), a regular specimen; sample 14 (16 fathoms), irregular speci- mens; sample 15 (19 fathoms), an irregular specimen ; sample 17 (12 fathoms), both regular and irregular specimens; sample 18 (74 fathoms), regular and irregular specimens. ORBITOLITHS CoMPLANATA, Lamarck. Orbitolites complanata, Lamarck, 1801, Syst. Anim. sans Vert. p. 576. O. complanata, Lam., Egger, 1893, Abhandl. bayer. Akad. Wiss., math.-phys. Cl. ii. vol. xviii. p. 249, pl. iii. fig. 40. Although rarer than the foregoing species, O. complanata is common and fairly typical in two of the lagoon samples. The variety plicata of Dana* is found in some abundance in the sand from the lagoon beach at Funafuti, but it does not occur * Marginopora vertebralis, Blainville, var. plicata, J. D. Dana, 1848, Wilkes’ U.S. Expl. Exped., Rep. Zooph. p. 706, pl. 60. figs. 9, 9.@, 4. 182 MR. F. CHAPMAN ON FORAMINIFERA in the dredgings. It is somewhat surprising that O. duplea should be altogether wanting in these foraminiferal deposits. Lagoon, Funafuti. Sample 1 (10 fathoms); sample 2 (153 fathoms). Family ASTRORHIZIDA. Subfamily RHaABDAMMININ4|. Hyprramuina, Brady [1878]. HYPERAMMINA RAMOSA, Brady. Hyperammina ramosa, Brady, 1879, Quart. Journ. Micr. Sci., n. s. vol. xix. p. 33, pl. iii. figs. 14, 15. Hf, ramosa, Brady, Goés, 1893, Arctic and Scand. Foram., Sy. Vet.-Ak. Handl, vol. xxv. p. 18, pl. iv. figs. 61, 62. H. ramosa, Brady, Flint, 1899, Rep. U.S. Nat. Mus., Rep. for 1897, PecdOnpl Wi tie. 1. It is unusual to meet with this species in low latitudes, but it has already been recorded by Goés from the Pacific and the Caribbean Sea; and by Flint from Cape Hatteras and the Gulf of Mexico. The specimens from Funafuti lagoon exhibit the primordial segment, and the test is yellowish brown and composed of very slender and minute calcisponge-spicules. Lagoon, Funafuti. Sample 15 (19 fathoms). Sagenina, Chapman [1900].* SaGENINA FRONDESCENS (Brady). Sagenella frondescens, Brady, 1879, Quart. Journ. Micr. Sci, n. s. vol. xix. p. 41, pl. v. fig. 1. S. frondescens, Brady, 1884, Rep. Chall. vol. ix. p. 278, pl. xxviii. fios. 14, 15. Sagenina frondescens (Brady), Chapman, 1900, Journ. Linn. Soc., Zool, vol. xxviii. p. 4, pl. i. figs. 1, 2; pl. ii. figs. 1, 2. This is one of the most characteristic and widely spread organisms both of the lagoon and the outer reef. It is an attached form, and makes its growth usually upon the fronds of Halimeda, but is also found growing upon other objects which present a suitable surface for the purpose. * Sagenella, Brady, 1879, name preoccupied for a genus of Polyzoa, see Journ. Linn. Soc., Zool. vol. xxviii. p. 4. \ FROM THE hues AT FUNAFUTI. 183 Unlike the other Foraminifera, this form is conspicuously absent from dredgings taken immediately inside the rim of the atoll on the opposite sides of the lagoon, and increases in numbers and character towards the middle of the lagoon. S. frondescens is apparently restricted to the S. Pacific. Lagoon, Funafuti. Sample 3 (20 fathoms); sample 4 (23 fathoms); sample 5 (24 fathoms); sample 6 (21 fathoms) ; sample 7 (24 fathoms); sample 8 (26 fathoms); sample 9 (25 fathoms); sample 10 (26 fathoms); sample 12 (23 fathoms) ; sample 14 (16 fathoms); sample 15 (19 fathoms); sample 16 (20 fathoms). Family LITUOLIDA. Subfamily Lirvoiin as. PLAcopPsILINA, d’Orbigny [1850]. PLACOPSILINA CENOMANA, d Orbigny. Placopsilina cenomana, d’Orbigny, 1850, Prodr. Paléont. vol. ii. p. 185, No. 758. This species is not a common form, but it is generally distri- buted in tropical and subtropical areas. It was found by the ‘Challenger’ at five or six stations amongst the coral-reefs of the Pacific. The specimens found in the lagoon at Funafuti are typical. The test sometimes branches towards the end. Lagoon, Funafuti. Sample 5 (24 fathoms); sample 7 (24 fathoms); sample 16 (20 fathoms); sample 17 (12 fathoms). Happonta, Chapman [1898]. Happonta TorrEstENnsis, Chapman. Haddonia Torresiensis, Chapman, 1898, Journ. Linn. Soc., Zool. vol, xxvi. p. 452, pl. xxviii. figs. 1-5 and woodeut p. 453. H. Torresiensis, Chapman, 1900, Journ. Linn. Soc., Zool. vol. xxviii. p. 6. The particular part of the lagoon where this species appears to thrive best is near the middle, from sample 10, and four miles from the nearest reef; at this place the Haddonie have a peculiar habit of sheltering between the fronds of Halimeda, and, by adhering at various points to the joints, cement them into a more or less coherent mass. 184 MR. F. CHAPMAN ON FORAMINIFERA Lagoon, Funafuti. Sample 6 (21 fathoms); sample 7 (24 fathoms); sample 19 (26 fathoms); sample 11 (25 fathoms); sample 14 (16 fathoms); sample 15 (19 fathoms); sample 16 (20 fathoms); sample 18 (73 fathoms). Subfamily TROCHAMMINING. Carrerina, Brady [1884]. CARTERINA SPICULOTESTA (Carter). Rotalia spiculotesta, Carter, 1877, Ann. Mag. Nat. Hist. ser. 4, vol. xx. p- 470, pl. xvi.; ibid. 1879, ser. 5, vol. ui. p. 414; ibid. 1880, ser. 5, vol. v. p. 452. Carterina spiculotesta (Carter), Brady, 1884, Rep. Chall. vol. ix. p. 346, pl. xli. figs. 7-10. C. spiculotesta (Carter), Millett, 1899, Journ. R. Micr. Soe. p. 365. This curious little organism, although resembling Trochammina in general form and outline, constructs its test of minute calcareous spicules of a fusiform shape. The primary whorls of the test are usually strongly coloured a deep reddish brown, presumably by the sarcode; the last whorl or so of the shell is more often white, although in one example (a fully grown specimen from Funafuti) is strongly coloured to the periphery. The little spicular bodies appear to be somewhat felted together in the central part of the test in each segment, but lie more or less parallel along the outer borders. C. spiculotesta has been found in the Gulf of Suez, the Red Sea, the Gulf of Manaar, the Malay Archipelago, and the S. Pacific. It is normally adherent, and is found attached to Halimeda and Lithothamnion at Funafuti. Lagoon, Funafuti. Sample 10 (26 fathoms); sample 11 (25 fathoms) ; sample 13 (26 fathoms); sample 15 (19 fathoms) ; sample 16 (20 fathoms). Family TEXTULARIIDA. Subfamily TEXTULARIING. TextuLarta, Defrance [1824]. TEXTULARIA FoLIUM, Parker J Jones. Textularia folium, Parker & Jones, 1865, Phil. Trans. vol. clv. pp. 870, 420, pl. xviii. fig. 19. FROM THE LAGOON AT FUNAFUTI. 185 L. fohum, P. & J., Moebius, 1880, Foram, Mauritius, p. 92, pl. viii. figs. 16, 17. The principal localities for this delicate little species are situated in the Pacific, where it is found in the shallow water of coral-reefs. It has also occurred at Mauritius, Bass Strait, Torres Strait, and Melbourne. Lagoon, Funafuti. Sample 18 (73 fathoms). TEXTULARIA Contca, @’ Orbigny. Textularia conica, dOrbigny, 1889, in Sagra’s Cuba, Foram. p. 143, pl. i. figs. 19, 20. T. conica, d’Orb., Egger, 1893, Abhandl. bayer, Ak. Wiss., math.- phys. Cl. ii. vol. xviii. p. 273, pl. vi. figs. 34-36, T. conica, d’Orb., Flint, 1899, Rep. U.S. Nat. Mus., Rep. for 1897, p- 285, pl. 29. fig. 6. This common and widely distributed species is best represented in the lagoon in the samples close to the rim of the atoll. In all cases they are rather undersized. Lagoon, Funafuti. Sample 1 (0 fathoms); sample 4 (23 fathoms) ; sample 14 (16 fathoms) ; sample 17 (12 fathoms) ; sample 18 (73 fathoms). TEXTULARIA RUGOSA (feuss). Plecanium rugosum, Reuss, 1869, Sitzungsh. Ak. Wiss. Wien, vol. lix. p. 453, pl. 1. figs. 3 a, 6. Textularia rugosa (Reuss), Flint, 1899, Rep. U.S. Nat. Mus., Rep. for 1897, p. 284, pl. 29. fig. 2. This striking coral-reef species is well represented in the lagoon dredgings. It is one of the exceptions to the rule that the best developed specimens occur near the margin of the lagoon, for extraordinarily large examples were found in samples 6 and 15, at 3 and 14 miles respectively from the edge. One of these measures 5 mm. in length; whereas the usual length of large specimens is 2 mm. Lagoon, Funafuti. Sample 1 (10 fathoms); sample 2 (153 fathoms); sample 6 (21 fathoms); sample 7 (24 fathoms) ; sample 15 (19 fathoms); sample 17 (12 fathoms); sample 18 (73 fathoms). Vernevuiuina, d’Orbigny [1840]. VERNEUILINA SPINULOSA, Reuss. Verneuilina spinulosa, Reuss, 1849, Denkschr. Ak. Wiss. Wien, vol. i. p. 347, pl. xlvii. figs. 12 a-c, 186 MR. F. CHAPMAN ON FORAMINIFERA V. spinulosa, Reuss, Egger, 1895, Abhandl. bayer. Ak. Wiss., Cl. ii. vol. xvii. p. 281, pl. vil. figs. 11, 14-16. The lagoon specimens are below the average size. The examples from dredgings farthest from the margin of the lagoon have lost their delicate spines, which seems to prove that they have been carried in, in common with many other species, by the action of currents. Lagoon, Funafuti. Sample 1 (10 fathoms); sample 4 (23 fathoms) ; sample 17 (12 fathoms) ; sample 18 (74 fathoms). Vauvurina, @’Orbigny. Vatvunina Davipiana, Chapman. Valvulina Davidiana, Chapman, 1900, Journ. Linn. Soe., Zool. vol, xxviii. p. 9, pl. i. fig. 4. A fully developed example of this interesting species was found near the margin of the lagoon. Lagoon, Funafuti. Sample 18 (73 fathoms). Cravuuina, d’Orbigny [1826]. CLaVULINA PaRIsrENsIs, d’Orbigny. Clavulina Parisiensis, d’Orbigny, 1826, Ann. Sci. Nat. vol. vii. p. 268, No. 3: Modéle No. 66. C. Parisiensis, d’Orb., Flint, 1899, Rep. U.S. Nat. Mus., Rep. for 1897, p. 289, pl. 35. figs. 2, 3. The specimen found in the lagoon is a very neat example with the sutures of the chambers perfectly distinct. Lagoon, Funafuti. Sample 1 (10 fathoms). Subfamily BULIMININ 4. Boxivina, @’Orbigny [1839]. Borivina Punctata, a’ Orbigny. Bolivina punctata, d’Orbigny, 1839, Foram. Amér. Mérid. p. 61, pl. viii. fies. 10-12. B. punctata, VOrb., Flint, 1899, Rep. U.S. Nat. Mus., Rep. for 1897, p. 292, pl. 38. fig. 1. A single specimen of this delicate form was found in the central part of the lagoon. Lagoon, Funafuti. Sample 9 (25 fathoms). FROM THE LAGOON AT FUNAFUTI. 187 Botrvina timpata, Brady. Bolkvina hmbata, Brady, 1881, Quart. Journ. Micr. Sci., n. s. vol. xxi. p. 57. A typical example of this species was found near the margin of the lagoon. Lagoon, Funafuti. Sample 18 (74 fathoms). Bouivina TortTuosa, Brady. Bolivina tortuosa, Brady, 1881, Quart. Journ. Micr. Sci., n. s. vol. xxi. 57. : B. tortuosa, Brady, Egger, 18983, Abhandl. bayer. Ak. Wiss., math.- phys. Cl. i. vol. xviii. p. 298, pl. viii. figs. 43, 44. A widely distributed form in warm areas. The specimen from the lagoon is rather below the average size. Lagoon, Funafuti. Sample 18 (73 fathoms). Family LAGHNIDA. Subfamily PoLYMORPHININ |. Sagrina (d@’Orbigny, 1839) emend. Parker & Jones {1865}. SaGRina RapHANUS, Parker J Jones. Uvigerma (Sagrina) raphanus, Parker & Jones, 1865, Phil, Trans. vol. clv. p. 364, pl. xviii. fies. 16, 17. Siphogenerina (Sagrina) raphanus, P. & J., Egger, 1893, Abhandi. bayer. Ak. Wiss., math.-phys. Cl. ii. vol. xviii. p. 317, pl. ix. fig. 36. In common with the other; true reef-loving Foraminifera, this species is confined to samples from the margin of the lagoon. Lagoon, Funafuti. Sample 1 (10 fathoms); sample 17 (12 fathoms); sample 18 (74 fathoms). Family GLOBIGERINID. GiopreERina, d’Orbigny [1826]. GLOBIGERINA BULLOIDES, d@’Orbigny. Globigerina bulloides, d’Orbigny, 1826, Ann. Sci. Nat. vol. vii. p. 277, No. 1: Modéles, Nos. 17 & 76. G. bulloides, d’Orb., Flint, 1899, Rep. U.S. Nat. Mus., Rep. for 1897, p. 321, pl. 69, fig. 2. The conditions under which these specimens lived are not 188 MR. F. CHAPMAN ON FORAMINIFERA favourable to the growth of this pelagic foraminifer, for the specimens found are al) below the usual size and are restricted to the lagoon-margin. Lagoon, Funafuti. Sample 2 (153 fathoms); sample 17 (12 fathoms); sample 18 (74 fathoms). GLOBIGERINA BULLOIDES, d’Orbigny, Var. TRILOBA, Reuss.’ Globigerina triloba, Reuss, 1849, Denkschr. Ak. Wiss. Wien, vol. i. p. 374, pl. xlvii. figs. 11 a-e. G. bulloides, d’Orb., var. triloba, Reuss, Brady, 1884, Rep. Chall. vol. ix. p- 595, pl. lxxix. figs. 1, 2; pl. Ixxxi. figs. 2, 3. This variety accompanies the type form in two out of the three occurrences in the lagoon. Similarly these are small starved specimens. Lagoon, Funafuti. Sample 17 (12 fathoms); sample 18 (74 fathoms). Family ROTALIIDA. Subfamily SPIRILLINING. SprriuLina, Hhrenberg [1841]. SPIRILLINA TUBERCULO-LIMBATA, Chapman. Spirillina tuberculo-limbata, Chapman, 1900, Journ. Linn. Soc., Zool. vol. xxviii. p. 11, pl. 1. figs. 8 a—c. Although in general form this is like Spirillina inequalis, Brady, the two faces of the test are differently decorated to that species, there being no tubercles on the inferior face. It is confined to one point near the lagoon-margin. Lagoon, Funafuti. Sample 18 (74 fathoms). SPIRILLINA SPINIGERA, Chapman. (PI. 19. figs. 9 & 10.) Spirillina spinigera, Chapman, 1900, Journ. Linn. Soc., Zool. vol. xxviii. p- 10, pl. i. figs. 7 a-c. The material from the lagoon affords us many additional specimens of this interesting form. The specimen (fig. 9) showing the long spinous decoration appears to belong to the megalospherie type of shell; whilst the larger forms, which by the way are more numerous, and which have a thinner peripheral — edge, serrate rather than spinous (fig. 10), are presumably microspherie. Lagoon, Funafuti. Sample 18 (73 fathoms). FROM THE LAGOON AT FUNAFUTI. 189 Subfamily RoraLiins. CyrmBatopora, Hagenow [1850]. Cympatopora Poeryt (d’ Orbigny). Rosalina Poeyrx, d’Orbigny, 1839, Foram. Cuba, p. 100, pl. iii. figs. 18-20. Cymbalopora Poeyt (d’Orb.), Egger, 1895, Abhandl. bayer, Ak. Wiss., math.-phys. Cl. ii. vol. xviii. p. 381, pl. xvili. figs. 51, 52. C. Poeyi (d’Orb.), Flint, 1899, Rep. U.S. Nat. Mus., Rep. for 1897, p. 326, pl. 72. fig. 1. When living, the central part of the test of this species is usually of a rose or red-brown tint. The specimens from the lagoon are invariably white, and those from near the middle of the lagoon are discoloured and yellowish-brown. This leads one to conclude that these and many other forms are not actually living in the lagoon, but are carried in by ocean currents. The depressed variety sometimes found accompanying the typical forms is here very rare. Lagoon, Funafuti. Sample 1 (10 fathoms); sample 14 (26 fathoms); sample 16 (20 fathoms); sample 17 (12 fathoms) ; sample 18 (73 fathoms). CYMBALOPORA TABELLHFORMIS, Brady. Cymbalopora tabelleformis, Brady, 1884, Rep. Chall. vol. ix. p. 687, pl. cii. figs. 15-18. C. tabelleformis, Brady, Egger, 1895, Abhandl. bayer. Ak. Wiss., math.-phys. Cl. ii. vol. xviil. p. 382, pl. xviil. figs. 54, 55. The specimens from the lagoon are typical but rather small. The species is chiefly known from the coral-reefs of the islands of the Pacific, and it also occurs off Mauritius. Lagoon, Funafuti. Sample 18 (73 fathoms). CymBaLopora (TRETOMPHALUS) BULLOIDES (d’Orbigny). Rosalina bulloides, VOrbigny, 1839, Foram. Cuba, p. 104, pl. ili. figs. 2-5. Tretomphalus bulloides (d’Orbigny), Mobius, 1880, Foram. Mauritius, p. 98, pl. x. figs. 6-9. Cymbalopora (Tretomphalus) bulloides (d’Orb.), Brady, 1884, Rep. Chall. vol. ix. p. 638, pl. cil. figs. 7-12. C. bulloides (d’Orb.), Egger, 1893, Abhandl. bayer. Ak. Wiss., math.- phys. Cl. ii. vol. xviii. p. 381, pl. xviii. fig. 53. It is very probable that the examples found in the lagoon have floated in by the action of currents, especially since this form is 190 MR. F. CHAPMAN ON FORAMINIFERA usually found near the surface of the ocean. Our specimens are of the normal size. Lagoon, Funafuti. Sample 17 (12 fathoms); sample 18 (73 fathoms). DiscorBina, Parker b Jones [1862]. DIsScORBINA ARAUCANA (d’ Orbigny). Rosalina araucana, d’Orbigny, 1839, Foram. Amér. Mérid. p. 44, pl. vi. figs. 16-18. Discorbina araucana (d’Orb.), Egger, 1893, Abhandl. bayer. Ak. Wiss., math.-phys. Cl. 11. vol. xviii. p. 386, pl. xiv. figs. 4-6. A single specimen, rather small, was found in the lagoon. Lagoon, Funafuti. Sample 16 (20 fathoms). DiscorBina RUGOSA (d’Orbigny). Rosalina rugosa, VOrbigny, 1839, Foram. Amér, Mérid. p. 42, pl. ii. figs. 12-14. Discorbina rugosa (d’Orb.), Egger, 1893, Abhandl. bayer. Ak. Wiss., math.-phys. Cl. ii. vol. xviii. p. 383, pl. xv. figs. 1-3. D. rugosa is not uncommon in the lagoon dredgings. One of the specimens is coloured a warm-brown. Lagoon, Funafuti. Sample 10 (26 fathoms); sample 14 (16 fathoms) ; sample 17 (12 fathoms); sample 18 (74 fathoms). DiscorBINA GLOBULARIS (d’ Orbigny). Rosalina globularis, VOrbigny, 1826, Ann. Sci. Nat. vol. vii. p. 271, pl. xiii. figs. 1-4: Modéle No. 69. Discorbina globularis (d’Orb.), Flint, 1899, Rep. U.S. Nat. Mus., Rep. for 1897, p. 327, pl. 72. fig. 2. Our specimens are somewhat variable in size, and the species is not uncommon in the lagoon. Lagoon, Funafuti. Sample 10 (26 fathoms); sample 15 (19 fathoms); sample 16 (20 fathoms); sample 17 (12 fathoms) ; sample 18 (74 fathoms). DiscorBina Saver (d’Orbigny). Rosalina Saulen, .d’Orbigny, 1839, Foram. Amér. Mérid. p. 42, pl. ii. figs. 9-11. Discorbina Saulcii (d’Orb.), Parker & Jones, 1872, Quart. Journ. Geol. Soe. vol. xxviii. p. 156. D. Saulen (VOrb.), Egger, 1893, Abhandl. bayer. Ak. Wiss., math.- phys. Cl. ii. vol. xviii. p. 392, pl. xv. figs. 51-53. The specimen from the lagoon is characteristically flattened FROM THE LAGOON AT FUNAFUTI. 191 on the superior face, and the colour of the shell is not unlike that of D. rugosa. Lagoon, Funafuti. Sample 17 (12 fathoms). DiscorBina concinna, Brady. Discorbina concinna, Brady, 1884, Rep. Chall. vol. ix. p. 646, pl. xe. fios. 7, 8. D. concinna, Brady, Egger, 1893, Abhandl. bayer. Ak. Wiss., math.- phys. Cl. ii. vol. xviii. p. 388, pl. xv. figs. 22-24. In addition to the localities for this species given by Brady, the ‘Gazell-’ collected it from Cape Verde, Mauritius, and West of Australia. Lagoon, Funafuti. Sample 17 (12 fathoms). Discorpina VILARDEBOANA (d’Orbigny). Rosalina Vilardeboana, d’Orbigny, 1839, Foram. Amér, Mérid. p. 44, pl. vi. figs. 13-15. Discorbina Vilardeboana (dOrb.), Egger, 1893, Abhandl. bayer. Ak. Wiss., math.-phys. Cl. ii. vol. xviii. p. 387, pl. xv. figs. 13-15. This form in common with the other Discorbine was found at the lagoon-margin. Lagoon, Funafuti. Sample 18 (74 fathoms). DiscoRBINA PILEOLUS (d’Orbigny). Valwulina pileolus, dOrbigny, 1839, Foram. Amér. Mérid. p. 47, pl. i. figs. 15-17. Discorbina pileolus (VOrb.), Hgger, 1895, Abhandl. bayer. Ak. Wiss., math.-phys. Cl. ii. vol. xviii. p. 390, pl. xv. figs. 61-63. This species is a well-known form from the Pacific; and is frequent in shallow-water dredgings in tropical and subtropical areas. The Funafuti specimens are typical but not common. Lagoon, Funafuti. Sample 18 (73 fathoms). DiIscoRBINA ORBICULARIS (Terquem). Rosalina orbicularis, Terquem, 1876, Anim. sur la Plage de Dunkerque, p- 75, pl. ix. figs. 4a, b. Discorbina orbicularis (Terq.), Egger, 1893, Abhandl. bayer. Ak. Wiss., math.-phys. Cl. ii. vol. xviii. p. 389, pl. xv. figs. 16-18, 76-78. This is well represented in point of numbers, and the specimens are fairly typical. A solitary specimen occurred near the middle of the lagoon. Lagoon, Funafuti. Sample 8 (26 fathoms); sample 17 (12 fathoms); sample 18 (73 fathoms). 192 MR. F. CHAPMAN ON FORAMINIFERA DiscorBINa TUBEROCAPITATA, Chapman. Discorbina tuberocapitata, Chapman, 1900, Journ. Linn. Soc., Zool. vol. xxviii. p. 11, pl. i. figs. 9 a-c. This striking form is not so well developed in the lagoon samples, as in the shallow-water dredgings on the outer part of the reef. Lagoon, Funafuti. Sample 11 (25 fathoms); sample -13 (26 fathoms); sample 16 (20 fathoms). DISCORBINA RARESCENS, Brady. Discorbina rarescens, Brady, 1884, Rep. Chall. vol. ix. p. 651, pl. xe. figs. 2, 3, and 4? D. rarescens, Brady, Egger, 1893, Abhandl. bayer. Ak. Wiss., math.- phys. Cl. ii. vol. xviii. p. 388, pl. xv. figs. 45-47. The species which have undoubtedly lived in the lagoon are not numerous, and apparently the present form is an example which has adapted itself to those conditions. It appears to grow attached to the surfaces of other organisms such as Halimeda. Brady records this species from the Philippines and Torres Strait, and Egger from West Africa and Western Australia. Lagoon, Funafuti. Sample 4 (3 fathoms); sample 6 (21 fathoms) ; sample 18 (26 fathoms); sample 14 (16 fathoms), sample 16 (20 fathoms); sample 17 (12 fathoms). PuLanoRBULINA, @’ Orbigny [1826]. PLANORBULINA MEDITERRANENSIS, d’Orbiguy. Planorbulina Mediterranensis, d’Orbigny, 1826, Ann. Sci. Nat. vol. vii. p. 280, pl. xiv. figs. 4-6: Modéle No. 79. P. Mediterranensis, VOrb., Egger, 1893, Abhandl. bayer. Ak. Wiss., math.-phys. Cl. 11. vol. xviii. p. 380, pl. xiv. figs. 24-26. P. Mediterranensis, d’Orb., Goés, 1894, Kongl. Svenska Vet.-Ak. Handl., vol. xxv. No. 9, p. 91, pl. xv. fig. 786. P. Mediterranensis, d’Orb., Flint, 1899, Rep. U.S. Nat. Mus., Rep. for 1897, p. 828, pl. 72. fig. 6. This species has a wide geographical distribution. It has already been recorded from the Pacific Ocean. The lagoon specimens are small and arrested in their growth, and are very like the specimen figured by Goés to which reference is given above. Lagoon, Funafuti. Sample 1 (10 fathoms); sample 18 (26 fathoms); sample 18 (73 fathoms). FROM THE LAGOON AT FUNAFUTI. 193 PLANORBULINA LARVATA, Parker 3° Jones. Planorbulina vulgaris, var. larvata, Parker & Jones, 1860, Ann. Mae. Nat. Hist. ser. 3, vol. v. p. 294. P. larvata, Parker & Jones, 1865, Phil. Trans. vol. cly. p. 379, pl. xix. fig. 3a, b. P. larvata, P. & J., Egger, 1893, Abhandl. bayer. Ak. Wiss., math.- phys. Cl. ii. vol. xviii. p. 189, pl. xiv. fig. 31. Amongst the several planorbuline forms this is one which possesses most individuality, the specimens being nearly all characteristic. Lagoon, Funafuti. Sample 1 (10 fathoms); sample 4 (23 fathoms); sample 6 (21 fathoms); sample 7 (24 fathoms) ; sample 8 (26 fathoms). PLANORBULINA LARVATA, Parker J Jones, var. CRISPATA, Nov. (Pl. 20. fig. 5.) This is a curious aberrant form in which the several series of chambers have been added to in a semicircular manner, instead of a discoidal, giving rise to a curled or semilunar-shaped test.. Greatest length 2°2 mm. Lagoon, Funafuti. Sample 5 (24 fathoms). PLANORBULINA ACERVALIS, Brady, var. FIMBRIATA, Nov. (Pl. 20. fig. 4.) This example differs from the usual form in having the peripheral series of chambers separated and slightly flattened and squared at the extremities, producing a kind of fringe around the discoidal test. The central area is smoother than usual, but shows a certain amount of acervuline segmentation forming a superficial series of chamberlets. Diameter 1-4 mm. Lagoon, Funafuti. Sample 8 (26 fathoms). TruncatTuLina, d’Orbigny [1826]. TRUNCATULINA VARIABILIS, d Orbigny. Truncatulina variabilis, d’Orbigny, 1826, Ann. Sci. Nat. vol. vii. p- 279, No. 8. T. variabilis, VOrb., Egger, 1893, Abhandl. bayer. Ak. Wiss., math.- phys. Cl. ii. vol. xviii. p. 404, pl. xvi. figs. 57-59, 68, 64. Our specimen was found at some distance from the edge of the lagoon ; and the test being of a brown tint in the central portion appears to point to the conclusion that the organism was living 72 svtu. Lagoon, Funafuti. Sample 13 (26 fathoms). LINN. JOURN.—ZOOLOGY, VOL. XXVIII. 14 194. MR. F. CHAPMAN ON FORAMINIFERA TRUNCATULINA PYamMmaA, Hantken. Truncatulina pygmea, Hantken, 1875, Mittheil. Jahrb. ung. geol. Anstalt, vol. iv. p. 78, pl. x. fig. 8. T. pygmea, Hantken, Egger, 1893, Abhandl. bayer. Ak. Wiss., math.- phys. Cl. ii. vol. xviii. p. 400, pl. xvi. figs. 60-62. T. pygmea, Hantken, Flint, 1899, Rep. U.S. Nat. Mus., Rep. for 1897, p. 334, pl. 77. fig. 6. This is a species of Truncatulina usually confined to deep water. Lagoon, Funafuti. Sample 17 (12 fathoms). TruncatuLtina Uneertana (d’ Orbigny). itotalina Ungeriana, d’Orbigny, 1846, Foram. Foss. Vienne, p. 157, pl. viii. figs. 16-18. Truncatulina Ungeriana (d’Orb.), Flint, 1899, Rep. U.S. Nat. Mus., Rep. for 1897, p. 338, pl. 77. fig. 2. This species is represented by a very small specimen from the lagoon. Lagoon, Funafuti. Sample 16 (20 fathoms). TRUNCATULINA ROSTRATA, Brady. Truncatulina rostrata, Brady, 1881, Quart. Journ. Micr. Sci. vol. xxi. Nn. 8. p. 65. T. rostrata, Brady, 1884, Rep. Chall. vol. ix. p. 668, pl. xiv. fig. 6 a-c. The test of this species differs in having a peripheral slit-like aperture in addition to the normal truncatuline orifice. This form is restricted almost without exception to the coral islands of the Pacific. Lagoon, Funafuti. Sample 1 (10 fathoms); sample 4 (23 fathoms) ; sample 14 (16 fathoms) ; apa 17 (12 fathoms) ; sample 18 (73 fathoms). TRUNCATULINA RETICULATA (Czjzeh). Rotalina reticulata, Czjzek, 1848, Haidinger’s Naturw. Abhandl. vol. ii. p. 145, pl. xiii. figs. 7-9. Truncatulina (Rotalina) reticulata, Czjzek, Egger, 1893, Abhandl. bayer. Ak. Wiss., math.-phys. Cl. ii. vol. xviii. p. 402, pl. xvi. figs. 42-44, Planorbulina reticulata (Czjzek), Goés, 1896, Bull. Mus. Comp. Zool. Harvard, vol. xxix. No. 1, p. 72. Truncatulina reticulata (Czjzek), Flint, 1899, Rep. U.S. Nat. Mus., Rep. for 1897, p. 384, pl. 78. fig. 3. FROM THE LAGOON AT FUNAFUTI. 195 This pretty little species is found in one sample only from the lagoon. The specimens are of normal size. Lagoon, Funafuti. Sample 18 (74 fathoms). Anomatina, d’ Orbigny [1826]. ANOMALINA AMMONOIDES (Rewss). Rosalina ammonoides, Reuss, 1845, Verstein. bohm. Kreide, pt. i. p- 36, pl. xii. fig. 66, pl. viii. fig. 53. Anomalina (Rosalina) ammonoides, Reuss, Egger, 1893, Abhandl. bayer. Akad. Wiss., math.-phys. Cl. ii. vol. xviii. p. 378, pl. xiv. figs. 35-37. A, ammonoides (Reuss), Flint, 1899, Rep. U.S. Nat. Mus., Rep. for 1897, p. 335, pl. 78. fig. 4. This widely-distributed species is here represented by a small solitary specimen. Lagoon, Funafuti. Sample 17 (12 fathoms). CaRPENTERIA, Gray [1858]. CARPENTERIA PROTEIFORMIS, Goés. (Pl. 19. fig. 11.) Carpenteria balaniformis, var. proteiformis, Goés, 1882, K. Svenska Vet.-Akad. Handl. vol. xix. No. 4, p. 94, pl. vi. figs. 208-214, pl. vil. figs. 215-219. C. proteiformis, Goés, Brady, 1884, Rep. Chall. vol. ix. p. 679, pl. xevil. figs. 8-14. C. proteiformis, Goés, 1896, Bull. Mus. Comp. Zool. Harvard, vol. xxix. No. 1, p. 74, pl. vi. figs, 8-17. Both examples found in the lagoon are attached to joints of Halimeda. Lagoon, Funafuti. Sample 6 (21 fathoms); sample 7 (24 fathoms). CARPENTERIA UTRICULARIS, Carter. Polytrema utriculare, Carter, 1876, Ann. Mag. Nat. Hist. ser. 4, vol. xvii. p. 210, pl. xiii. figs. 11-16. Carpenteria utricularis, Carter, 1877, Ann. Mag. Nat. Hist. ser. 4, vol. xx. p. 176. C. utricularis, Carter, Egger, 1893, Abhandl. bayer. Akad. Wiss., math.-phys. Cl. ii. vol. xviii. p. 488, pl. xxi. fig. 18. C. utricularis, Carter, Chapman, 1900, Journ. Linn. Soc., Zool. vol. Xxvill. p. 12, pl. 2. fig. 4, pl. 4. figs. 3, 4. This species, like the preceding, appears to prefer the quieter areas of the lagoon. It was found attached to Halimeda. Lagoon, Funafuti. Sample 10 (26 fathoms). 14* 196 MR. F. CHAPMAN ON FORAMINIFERA Putvinvutina, Parker & Jones [1862]. PULVINULINA REPANDA (Fichtel § Moll). Nautilus repandus, Fichtel & Moll, 1803, Test. Micr. p. 35, pl. iii. figs. a-d. Pulvinulina repanda (¥F. & M.), Egger, 1893, Abhandl. bayer. Ak. Wiss., math.-phys. Cl. ii. vol. xviii. p. 405, pl. xviii. figs. 84-86. P. repanda (F. & M.), Goés, 1896, Bull. Mus. Comp. Zool. Harvard, vol. xxix. no. 1, p. 75. P. repanda (F. & M.), Flint, 1899, Rep. U.S. Nat. Mus., Rep. for 1897, p. 828, pl. 72. fig. 8. A full-sized specimen of the above occurred in the lagoon dredgings. Lagoon, Funafuti. Sample 3 (20 fathoms), PULVINULINA OBLONGA ( Williamson). Rotalina oblonga, Williamson, 1858, Rec. Foram. Gt. Brit. p. 51, pl. iv. figs. 98-100. Pulvinulina oblonga (Williamson), Egger, 1893, Abhandl. bayer. Ak. Wiss., math.-phys. Cl. ii. vol. xviii. p. 415, pl. xvii. figs. 28-25. The specimen found in the lagoon is a very delicate thin- shelled example. The ‘Challenger’ obtained it from the South Pacific in depths varying from 17 to 275 fathoms. It has a wide geographical distribution. Lagoon, Funafuti. Sample 17 (12 fathoms). Roraria, Lamarck | 1804. Roratia Beccarti (Linné). Nautilus Beccarni, Linné, 1767, Syst. Nat. 12th ed. p. 1162; 1788, ibid. 15th (Gmelin’s) ed. p. 3570, no. 4. Rotaha Beccart (L.), Egger, 1893, Abhandl. bayer. Ak. Wiss., math.- phys. Cl. ii. vol. xviii. p. 420, pl. xix. figs. 25-27. R. Beccaru (L.), Flint, 1899, Rep. U.S. Nat. Mus., Rep. for 1897, p. 331, pl. 75. fig. 2. This species is commonly found in temperate seas, but it is not confined to them. The specimens from the lagoon have the characteristic brown stain of the sarcode remaining with them, and have therefore probably lived where they were found. Lagoon, Funafuti. Sample 1 (10 fathoms). Catcarina, d’Orbigny (1826). CaLcaRina Hispipa, Brady. Calcarina hispida, Brady, 1876, Proc. R. Irish Acad. ser. 2, vol. i. p. 590. FROM THE LAGOON AT FUNAFUTI. 197 C. calear, var. hispida, Carter, 1880, Ann. Mag. Nat. Hist. ser. 5, vol. v. p. 453. C. hispida, Brady, 1884, Rep. Chall. vol. ix. p. 713, pl. eviii. figs. 8, 9. This is one of the commonest of the smaller Foraminifera in the lagoon. From its low specific gravity and form it seems to be carried in large numbers to quite the centre of the lagoon. That it does not live there is evident from the fact that the specimens from samples taken at considerable distance from the lagoon shore are invariably stained of a muddy brown and are partially weathered or decomposed. Lagoon, Funafuti. Sample 1 (10 fathoms); sample 2 (154 fathoms); sample 4 (23 fathoms); sample 5 (24 fathoms); sample 6 (21 fathoms); sample 7 (24 fathoms); sampie 8 (26 fathoms); sample 14 (16 fathoms); sample 15 (19 fathoms) ; sample 16 (20 fathoms); sample 17 (12 fathoms); sample 18 (74 fathoms). Cancartna Derranctt, d’Orbigny. Calearina Defrancit, dOrbigny, 1826, Ann. Sci. Nat. vol. vii. p. 276, pl. xiii. figs. 5-7. C. Defrancit, VOrb., Brady, 1884, Rep. Chall. vol. ix. p. 714, pl. eviii. fig. 6 a—-c. This generally rare form is found in one sample only in the lagoon. Lagoon, Funafuti. Sample 1 (10 fathoms). Subfamily TrINOPORIN:. Trnororus, Montfort [1808], emend. Carpenter [1860]. Trvoporvs BACcULATUS, Montfort. Tinoporus baculatus, Montfort, 1808, Conchyl. Syst. vol. i. p. 146, 37e genre. T. baculatus, Montf., Carpenter, 1860, Phil. Trans. p. 557, pls. xviii., xix. T. baculatus [Montfort ?], Carpenter, Brady, 1884, Rep. Chall. vol. ix. p. 716, pl. ci. figs. 4-7. This species undoubtedly attains to its fullest development in the shallow waters of the South Pacific. At Funafuti it is re- markably abundant, and the beach-sand of the lagoon is largely composed of this species. It is rare, however, in the waters of the lagoon, and occurs in three samples only. Lagoon, Funafuti. Sample 1 (10 fathoms); sample 11 (25 fathoms) ; sample 18 (74 fathoms). 198 MR. F. CHAPMAN ON FORAMINIFERA Gyestna, Carter [1877]. GYPSINA GLOBULUS (Reuss). Ceriopora globulus, Reuss, 1847, Haidinger’s Naturw. Abhandl. vol. i. p. 33, pl. v. fig. 7. Gypsina globulus (Reuss), Uhlig, 1886, Jahrb. geol. Reichsanst. vol. xxxvi. p. 197, figs. 7-9. G. globulus (Reuss), Brady, Parker & Jones, 1888, Trans. Zool. Soc. vol. xii. p. 229, pl. xlvi. fig. 18. Some very fine examples of this species occur in the lagoon dredgings. They vary very much in size, the diameter of the test ranging from ‘5 mm. to 3°5 mm. The smallest specimens came from the middle of the lagoon. Lagoon, Funafuti. Sample 1 (10 fathoms); sample 2 (15 fathoms); sample 4 (23 fathoms); sample 6 (21 fathoms); sample 7 (24 fathoms); sample 11 (25 fathoms); sample 12 (25 fathoms) ; sample 13 (26 fathoms); sample 14 (16 fathoms) ; sample 18 (73 fathoms). GYPSINA INHHRENS (Schultze). Acervulina inherens, Schultze, 1854, Organ. der Polythal. p. 68, pl. vi. fig, 12. “Gots inherens (Schultze), Brady, 1884, Rep. Chall. vol. ix. p. 718, pl. cii. figs, 1-6, G. inherens (Schultze), Goés, 1894, K. Svenska Vetensk.-Akad. Handl. vol. xxv. No. 9, p. 91, pl. xv. fig. 787. G. inherens (Schultze), Flint, 1899, Rep. U.S. Nat. Mus., Rep. for 1897, p. 336, pl. 79. fig. 6, The adherent forms of Gypsina are phenomenally numerous at Funafuti. To reduce the gradational series found there into something like order, and keeping in view some central type of structure, it seems convenient to group the more irregular and acervuline structured organisms under the name of G. inherens, reserving the regularly discoidal forms as varieties of the mound- like G. vesicularis, from which they graduate down to thin scale-like forms. Lagoon, Funafuti. Sample 1 (10 fathoms); sample 7 (24 fathoms); sample 18 (26 fathoms); sample 14 (16 fathoms); sample 15 (19 fathoms); sample 16 (20 fathoms); sample 17 (12 fathoms). GYPSINA VESICULARIS (Parker & Jones). (PI. 19. fig. 12.) Orbitolina vesicularis, Parker & Jones, 1860, Ann. Mag. Nat. Hist. ser. 3, vol. vi. p. 31, no. 5. FROM THE LAGOON AT FUNAFUTI. 199 Gypsina vesicularis (P. & J.), Carter, 1877, Ann, Mae. Nat. Hist. ser. 4, vol. xx. p. 173. G. vesicularis (P. & J.), Egger, 1893, Abhandl. bayer. Ak, Wiss., math.- phys. Cl. ii. vol. xviii. p. 382, pl. xiv. figs. 20-25. This species is usually found accompanying the spherical form G. globulus, into which it appears to graduate, but very rarely. On the other hand, it frequently shows affinities with the thinner varieties, by having a peripheral flange around the mound-like test. Lagoon, Funafuti. Sample 1 (10 fathoms); sample 2 (153 fathoms); sample 4 (23 fathoms): sample 5 (24 fathoms) ; sample 6 (21 fathoms); sample 8 (26 fathoms); sample 11 (25 fathoms); sample 13 (26 fathoms); sample 15 (19 fathoms) ; sample 16 (20 fathoms) ; sample 17 (12 fathoms); sample 18 (73 fathoms). GYPSINA VESICULARIS (P. g& J.), var. Discus, Goés. (Pl. 19. fig. 13.) Tinoporus vesicularis (P. & J.), Goés, 1882, Sv. Vet.-Akad. Handl. vol. xix. No. 4, p. 104, pl. vii. figs. 245-247. Gypsina vesicularis (P. & J.), var. discus, Goés, 1896, Bull. Mus. Comp. Zool. Harvard, vol. xxix. No. 1, p. 74, pl. vil. figs. 4-6. Upon this remarkable little variety Dr. Goés wrote as follows * :— “A variety closely allied to the type, from which it differs only in its lenticular shape and its more plainly differentiated set of the central cycle of chambers. It is not found affixed. Caribbean Sea, 400 fathoms ; scarce.” This is one of the most distinct varieties of the G. vesicularis type. The little button-shaped tests are usually depressed in the central region on both sides. It is noteworthy that this form was absent from the samples near the margin of the lagoon, appearing to prefer the quieter waters and at greater depths. This fact is further corroborated by Goés’ record of its occurrence at 4.00 fathoms. Lagoon, Funafuti. Sample 4 (23 fathoms); sample 5 (24 fathoms); sample 6 (21 fathoms); sample 7 (24 fathoms) ; sample 8 (26 fathoms); sample 10 (26 fathoms); sample 14 (16 fathoms); sample 15 (19 fathoms). * Op, cit. p. 74. 200 MR. F. CHAPMAN ON FORAMINIFERA Gypsrva vestcutaris (P. & J.), var. MONTICULUS, nov. (PI. 19. figs. 14 a, 6.) Test attached, roughly discoidal; the attached surface flat or undulate, and showing the more regular central series of chambers, sometimes spiral, followed by the acervuline chambers ; the free, superior surface sharply conical in the centre, rapidly sloping to the level of a thin peripheral flange. Distinguished from the thin wild-growing Planorbuline by the absence of peripheral orifices and the entire absence of regularity in the peripheral chambers. Diameter ranging from 2 to 4 mm. Lagoon, Funafuti. Sample 2 (153 fathoms); sample 4 (23 fathoms); sample 5 (24 fathoms); sample 7 (24 fathoms); sample 8 (26 fathoms); sample 11 (25 fathoms); sample 138 (26 fathoms); sample 15 (19 fathoms) ; sample 16 (20 fathoms) ; sample 17 (12 fathoms). Gypstna vEsSIcULARIS (P. & J.), var. SQUAMIFORMIS, NOV. (Pl. 19. figs. 15 a, 6.) Test thin, normally attached, encrusting surfaces of organisms. Full-sized specimens measure about 3 or 4 mm. in diameter. Shell consists of a single layer of acervuline chambers. The paper-like fragments of the test of this variety are sometimes very numerous in the shallow-water sands, both within and outside the lagoon. Lagoon, Funafuti. Sample 3 (20 fathoms); sample 4 (23 fathoms); sample 5 (24 fathoms); sample 7 (24 fathoms); sample 8 (26 fathoms); sample 13 (26 fathoms); sample 16 (20 fathoms); sample 17 (12 fathoms). Potytrema, isso [1826]. POLYTREMA MINIACEUM (Pallas). Millepora miniacea, Pallas, 1766, Elenchus Zoophytorum, p. 251. P. miniaceum (l.), Goés, 1896, Bull. Mus. Comp. Zool. Harvard, vol, xxix. No. 1, p. 75. P. miniaceum (Pallas), Chapman, 1900, Journ. Linn. Soc., Zool. vol. xxvill. p. 16, pl. 4. fig. 7. This species is found living in the lagoon in some abundance, but it never attains to the size which it does outside the atoll, usually measuring only 2 mm. across. It is generally found attached to joints of Halimeda. Lagoon, Funafuti. Sample 1 (10 fathoms); sample 6 (21 FROM THE LAGOON AT FUNAFUTI. 201 fathoms); sample 7 (24 fathoms); sample 10 (26 fathoms) ; sample 11 (25 fathoms); sample 13 (26 fathoms); sample 14 (16 fathoms) ; sample 16 (20 fathoms) ; sample 17 (12 fathoms) ; sample 18 (74 fathoms). POLYTREMA MINIACEUM (Pallas), var. ALBA, Carter. Polytrema miniaceum, var. album, Carter, 1877, Ann. Mag. Nat. Hist. ser, 4, vol. xix. p. 215, pl. xiii. fig. 14. This variety is generally rare at Funafuti. Lagoon, Funafuti. Sample 1 (10 fathoms); sample 2 (153 fathoms); sample 7 (24 fathoms); sample 13 (26 fathoms). PoLyTREMA PLANUM, Carter. (PI. 20. figs. 6, 7.) Polytrema planum, Carter, 1876, Ann. Mac. Nat. Hist. ser. 4, vol. xvii. p- 211, pl. xiii. figs. 18, 19. [Immature growth.] Gypsina melobesioides, Carter, 1877, ibid. ser. 4, vol. xx. p.172. [Full- grown form, description only. | G. melobesioides, Carter, 1880, ibid. ser. 5, vol. v. p. 445. Polytrema miniaceum (Pallas), var. znvelva, Chapman, 1900, Journ. Linn. Soc., Zool. vol. xxviii. p. 17, pl. 2. fig. 3, and text-figure 2. P. planum, Carter, Chapman, 1901, Ann. Mag. Nat. Hist. ser. 7, vol. vil. pp. 82, 83. In my earlier description of the Funafuti specimens, this particular form was regarded as a variety of the common branching type of P. miniacewm, which is distinguished by its red colour. Since then I have found abundant specimens of P. planum of very extensive growth on rough reef-specimens from the same collection, which show that the affinities of P. planwm are rather towards the white variety of P. miniacewm, var. alba, Carter; for in the fresh specimens its test is often of snowy whiteness, not unlike, as Carter remarks, the sugar- coating on a bride-cake. The cemented reef-specimens are often discoloured or yellowish in their tint, but never pink or rose- coloured. It is exceedingly difficult in some cases to distinguish P. planum from certain encrusting plants such as Lithothammon and Lithophyllum; and indeed the only distinctive character between Lithophyllum onkodes and P. planum, when viewed externally, is the presence of the little granular conceptacles often, but not always, to be seen studding the surface of the forms. Carter’s original description and figure dealt only with a young or immature growth, but a comparison between the type- specimen of the so-called Gypsina melobesioides at the Natural 202 MR. F. CHAPMAN ON FORAMINIFERA History’ Museum, South Kensington, confirms the identity of the two forms. This organism, as I have already shown, is of great importance as a rock-builder, by forming concentric encrustations round fragments of coral and alge. It also plays an important part in cementing loose fragments together, by filling crevices and eventually enwrapping several fragments, until they form a solid conglomerate. In the lagoon P. planum is found binding together the separate fronds of Halimeda as they lie on the lagoon-foor. It is impossible to distinguish P. planwm in the lagoon material without the aid of thin sections made from the accreted fragments for microscopic examination. Lagoon, Funafuti. Sample 6 (21 fathoms); sample 10 (26 fathoms); sample 11 (25 fathoms); sample 12 (23 fathoms) ; sample 14 (16 fathoms). Family NUMMULINID. Subfamily PoLYSTOMELLING. Nonronina, d'Orbigny [18261. NONIONINA DEPRESSULA (Walker & Jacob). Nautilus depressulus, Walker & Jacob, 1798, Adams’ Essays, Kan- macher’s ed., p. 641, pl. xiv. fig. 33. Nonionina depressula (W. & J.), Egger, 1893, Abhandl. bayer. Akad. Wiss., math.-phys. Cl. ii. vol. xvii. p. 427, pl. xix. figs. 38, 39. This species is rare in the lagoon. Lagoon, Funafuti. Sample 13 (26 fathoms): sample 18 (73 fathoms). NonrIONINA UMBILICATULA (Montagu). Nautilus umbihcatulus, Montagu, 1803, Test. Brit. p. 191; Suppl. p. 78, pl. xviii. fig. 1. Nonionina umbilicatula (Montagu), Egger, 1893, Abhandl. bayer. Akad. Wiss., math.-phys. Cl. ii. vol. xvii. p. 426, pl. xix. figs. 36, 37. This species is more numerous than the preceding, and the specimens are as a rule typical. Lagoon, Funafuti. Sample 1 (10 fathoms); sample 4 (28 fathoms); sample 7 (24 fathoms); sample 8 (26 fathoms) ; sample 11 (25 fathoms). FROM THE LAGOON AT FUNAFUTI. 2038 PotystoMELLA, Lamarck [1822]. PoLysTOMELLA STRIATOPUNCTATA (Hichtel 5 Moll). Nautilus striatopunctatus, Fichtel & Moll, 1798, Test. Micr. p. 61, pl. ix. figs. ae. Polystomella striatopunctata (F. & M.), Egger, 1893, Abhandl. bayer. Akad. Wiss., math.-phys. Cl. ii. vol. xviii. p. 433, pl. xix. figs. 49, 50. P. striatopunctata (F. & M.), Goés, 1896, Bull. Mus. Comp. Zool. Harvard, vol. xxix. No. 1, p. 78. P. striatopunctata (F. & M.), Flint, 1899, Rep. U.S. Nat. Mus., Rep. for 1897, p. 337, pl. 80. fig. 2. The lagoon-specimens are neat and small. Lagoon, Funafuti. Sample 1 (10 fathoms); sample 17 (12 fathoms). PoLYSTOMELLA MACELLA (ichtel g Moll). Nautilus macellus, var. a, Fichtel & Moll, 1798, Test. Micr. p. 66, pl. x. figs. e-g. Polystomella macella (F. & M.), Egger, 1893, Abhandl. bayer. Akad. Wiss., math.-phys. Cl. ii. vol. xviii. p. 482, pl. xx. figs. 22, 23. The lagoon-specimens are rather small. Lagoon, Funafuti. Sample 1 (10 fathoms); sample 5 (24 fathoms); sample 16 (20 fathoms); sample 17 (12 fathoms) ; sample 18 (7% fathoms). PoLYSTOMELLA CRISPA (Linné). Nautilus crispus, Linné, 1758, Syst. Nat. 10th ed. p. 709. Polystomella crispa (L.), Egger, 1893, Abhandl. bayer. Akad. Wiss., math.-phys. Cl. ii. vol. xviii. p. 452, pl. xx. figs. 20, 21. P. crispa (1.), Flint, 1899, Rep. U.S. Nat. Mus., Rep. for 1897, p. 338, pl. 80. fig. 3. This occurs very sparingly in the lagoon. Lagoon, Funafuti. Sample 2 (154 fathoms); sample 4 (23 fathoms); sample 17 (12 fathoms). PoLyYSTOMELLA SUBNODOSA (Munster). Robulina subnodosa, Miinster, 1838, Neues Jahrb. fiir Min. p. 391, pl. iii. fig. 61. Polystomella subnodosa (Minster), Reuss, 1855, Sitzungsb. Ak. Wiss. Wien, vol. xviii. p. 240, pl. iv. fig. 51 a, 6. P. subnodosa (Minster), Brady, 1884, Rep. Chall. vol. ix. p. 734, pl. cx. fig. 1 a, 6. 204 MR. F. CHAPMAN ON FORAMINIFERA P. subnodosa (Minster), Goés, 1894, Kong]. Sv. Vet.-Akad. Handl. vol. xxv. No. 9, p. 102, pl. xvii. figs. 817-819. The specimens from Funafuti agree in most respects with those figured by Goés from Novaya Zemlaia and off Newfound- land. By a comparison of the chief characters of the tests of this species and P. striatopunctata, there seems to be a direct affinity between those forms; and it is possible that they represent a single species, of which P. subnodosa is the megalo- spheric type, with an umbonate centre, and P. striatopunctata the microspheric type, with a depressed centre. P. subnodosa, as a recent form, has been but rarely found up to the present, for, besides the localities mentioned above, it has only been found amongst the islands south-west of Papua, at depths of from 6 to 28 fathoms. Lagoon, Funafuti. Sample 18 (73 fathoms). Subfamily NUMMULITING. AmpuistEeina, @’ Orbigny [1826]. AMPHISTEGINA Lussonul, d’ Orbigny. Amphistegina Lessonii, VOrbigny, 1826, Ann. Sci. Nat. vol. vii. p. 304, No. 3, pl. xvii. figs. 1-4: Modéle No. 98. A. Lessonit, V’Orb., Mobius, Beitr. Meeresfauna Insel Mauritius, p. 99, pl. x. figs. 10-14, pl. xi. figs. 1-3. A. Lessonit, VOrb., Flint, 1899, Rep. U.S. Nat. Mus., Rep. for 1897, p. 338, pl. 80. fig. 4. This is the only species of Foraminifera which was obtained from every sample of the lagoon dredgings. It varies in size from a diameter of *75 to 3°5 mm. The test is often flexuose on the peripheral edge; and towards the middle of the lagoon it shows a tendency to make abnormal outgrowths, in beak-like processes near the last series of chambers in adult shells, or as buds around the oral aperture. The test sometimes shows a tendency to become partially attached to alge, at the oral region, especially in the deeper parts of the lagoon. Lagoon, Funafuti. In samples 1 to 18. Samples 4, 6, and 15 yielded the finest specimens. FROM THE LAGOON AT FUNAFUTI. 205 Hererrosrreina, d Orbigny [1826]. HETEROSTEGINA DEPRESSA, d’Orbigny. ; Heterostegina depressa, Wd’ Orbigny, 1826, Ann. Sci. Nat. vol. vii. p. 305, pl. xvii. figs. 5-7: Modéle No. 99. Hf. curva, Mobius, 1880, Beitr. Meeresfauna Insel Mauritius, p. 105, pl. xiii. figs. 1-6. HI, depressa, d’Orb., Egger, 1893, Abhand]. bayer. Akad. Wiss., math.- phys. Cl. ii. vol. xviii. p. 433, pl. xx. figs. 34, 35, H. depressa, VOrb., Goes, 1896, Bull. Mus. Comp. Zool. Harvard, TO soars INGS Il jes 7) HH, depressa, d’Orb., Chapman, 1900, Journ. Linn. Soc., Zool. vol. xxviii. p- 18, pl. 3. figs. 6, 7. This species is well represented in the lagoon samples, but it does not occur in a few from near the middle of the lagoon. Hi. depressa here ranges from °75 mm. to 9 mm. in diameter. No specimens of the form B (microspheric) occur in the samples. The largest specimens occur in sample 8 (26 fathoms). Lagoon, Funafuti. Sample 1 (10 fathoms); sample 2 (154 fathoms); sample 3 (20 fathoms); sample 4 (23 fathoms); sample 5 (24 fathoms); sample 6 (21 fathoms); sample 7 (24 fathoms); sample 8 (26 fathoms): sample 9 (25 fathoms) ; sample 10 (26 fathoms); sample 14 (16 fathoms); sample 15 (19 fathoms); sample 17 (12 fathoms); sample 18 (73 fathoms). In the annexed synopsis of species from the Funafuti Lagoon the following significations are used :—v.r.=very rare, r.=rare, f.=frequent, c.=common, v.c.=very common. These indications are printed in three kinds of type: capitals (V.C.) signifying the specimens are well developed or extra large; ordinary type (v.c.) that the specimens are of normal size; and italics (v.c.) that the specimens are starved or of diminutive size. MR. F. CHAPMAN ON FORAMINIFERA 206 "UA ’ se dee wee nee eee . on oe eee o soe eee oe se coe 5 eee cee eee eos oe eee eee . . 9 coo On cee . oe . - . oe . . see eee on . eee oe ee eee . . oe "VA ‘odureg Jo ‘oN se ecececes eee eeesees eee eeeceee "tess (lodaey) vsopun e spicanio( qo, p) DIDINMAIIL “ seeeee Ape Atay ‘nunuuyay Mag “ABA aS i seevce Ape A ‘nuoruanbiay, 1A a gs “* CqtO,p) vpoupunonig ** (NseyUOP) vpungougns ss Tope acaoucionnoy (a1 9))0)) M8019) OY BOODOOOOO OOOO (‘utog) $2.00) WILD 6G srreeees (Ssnaqy) s7vm7na7Dr Ue Apeag, ‘obunweynav * Apeag, “wgngdasonuag ie * tuenbisey, ‘2006 sf “GIO, P “wncoyyryun Be ** -aou “ds ‘nsonj.1o7 Ue see eecoores RGYasis |‘ngnjoanof “IBA “GIO, p “wprj1u0 e "Q10,P ‘npnrvox0 ie srereeeee Kowlg “wgsnqos puynoojouds GnnBP (joaeuery) swabues & Hoe OP “‘paUloyasqns e me “440.P ‘phuojgo DU MoOj ET “- -aou “ds ‘seswawnan) eesne STL “hpoug 66 * goueajecy ‘nbnfron) si ' Kpeagr ‘opno.umarp vriunynoaynyy “OUR NT ‘NOODV] GDHL NI VUATININVUOT AHE FO NOIAAIUISIG, 10 ATaAV], 207 FROM THE LAGOON AT FUNAFUTI. OtA “TA > “TA “TA ‘OA coe ane see eee | eee eee see see oe eee eee eee eee ere soe Oo cee see coe ae nee eae ove cee see eee ee eee aoe aoe “I A w wn a J z@) tereereeescseesrss “CTO .p ‘nynjound nunuog alefatetelalalajelate)sislete *q1Q,P ‘SISUAIS IMD DUYNUD)OD) setrreseerosees eu ‘punmping nurnapy nerseessesceses genoy ‘Dsopniids nuymnaula 4 etelcrelatsts eee (ssneyy) psohid as BOODUDOUOOUGUOOUON *q1O,p “pauuoo ee weet eee eseenssses f bY oat ‘WnyOf DLM0INILAT, seeeeeseceee (roqaeay) mgsazopnards 12.1a)40/) sreeeesenses eng) ‘saswarsaulo J, DUWOppVET "q1O,p ‘vunmouss nuyrsdoon) gt sees ces ceseenes (Apeag) suaasapuous pruruahogy srreeresereonee Kpere ‘nsowne nurmuUnsadhzy BOO OCOOODOOOUDO “Ter, ‘pypunjduoo G6 trreeesseceesss Comer) semurbunue saguopgqucQ deyg “smog 2 - nee Giurteig 9) snaonapunha ys ‘deyg ‘snzujod 3 (wnrpishjouopy ) secre ences sccce (qosqyeq) snungarwp 66 rereeeenee sees (TRYSIOY) SUSNIUad SrdolaUag Teeeeeeeeseesooresssnayy ‘swanjoaur virdsnutop sreeeerseees (TOTIVY) DULISSLIDULO DUMANDET ee ercee Apeag. “09101 NILQO.LO8 x ress caour “dy ‘srswargnfpuny 5 sresseeeesesereeee(QiQ p) mumlosog” # Hence eieiseisieeS sie (2G UG) HPD) DUDINOT ce centres sccceee (usequ07) nbhu0)go 6b ee nergy ‘sausofuuyoann as seers ereesssces CL YM) $2U40920 “6 ser ar oeereeeres (‘q10,P) DUBDUUYT 6c sreteeserses (CTO) p) sumuynphbo the third has on its outer surface two sete, one being feathered and much longer than the other. All the complex hooking apparatus is confined to the fourth and smallest factor of the appendage. This apparatus is built up of five prehensile claws or processes: the first, and smallest, is a short stout organ with its inner face toothed near the extremity (figs. 6,7); the next process is similar to this, only slightly longer; the third differs only in the fact that it is more slender; the fcurth has a long, slightly curved, smooth portion which abruptly passes over into a 218 ALICE L. EMBLETON ON A NEW smaller part ending in three teeth on its inner surface; the fifth is the same, being however longer and stouter (Pl. 22. figs. 6, 7). In none of the descriptions of allied forms can I find an account of such an apparatus as this on the second antenna, though the number of parts going to make up the limb agrees with that in other genera, but this hooking arrangement is peculiar to Gotdelia japonica. Doubtless these hooks serve to fix the parasite to the walls of the rectum of its host. Mouth-parts.—These consist of an upper lip, mandibles, and maxille, all grouped on a raised dome-shaped area, occupying a relatively small space, owing to the reduction and degeneration of the various parts (Pl. 21. figs. 2, 3, mv.). Upper Lip—Overhanging the mouth-opening is a flattened lip; its lower edge is not pointed into a “ beak”’ as Kossmann says of Clausidium:—“ die Oberlippe léuft nach hinten za in linen spitzen Schnabel aus, und erinnert insofern an die Riissel- bildung der verwandten Copepoden.” Claus, however, speaking of the same creature says:—‘ Der breite Oberlippe bleibt von der zweilappigen mit feinen spitzen itiberkleideten Unterlippe vollkommen getrennt, kann die letztere aber beim Weibchen bedecken. Hine dem Riissel von Parasiten vergleichbare Schnabelbildung vernussen wir durchaus.” In Claus’ fig. 3, the lip appears very similar to that in my specimens. The double- lobed lower lip he mentions is present, fringed with fine sete. The upper lip forms a sort of flap, alike in female and male, its lower edge being slightly curved (Pl. 22. fig. 10, /.w), with a line of fine setz just above the free edge. In the male this lip bears stronger and more numerous sete, being particularly well- developed at the side angles of the lip. Beneath it, and just above the lower lip, is the small, almost circular, mouth-opening. The upper lip almost completely covers the mandibles, which are very small and simple. ‘ Mandibles.—These are a pair of very small chitinous structures of simple organization. Hach mandible is a slightly curved claw-like organ, ending in a serrated hook (Pl. 22. figs. 8, 9, 10, md.), the lower edge of which is toothed; this claw articulates on to the main stem of the mandible. The mandible of each side approaches the other in the middle line, not sloping to each other in an accurately transverse direction, but both pointing a little posteriorly. There are no “ accessory pieces” or palps on this simple organ, though Canu, in all the three known genera, ENTOZOIC COPEPOD FROM JAPAN. 219 gives two or three additional plates or processes. The form of the mandible in Goidelia japonica is constant for the adult and immature stages of both sexes. Canu also finds that the mouth- parts of the Clawsidizd@ are not altered by the various moults. Claus, in Clausidium, depicts the mandible with this claw-shaped end-piece, but in this case there is, in addition, a fringed process beneath it, entirely absent in Gozdelia japonica. Therefore, by taking the mouth-parts as the basis for classification, it becomes necessary (on this point alone) to separate this new form from the other three genera, though it is undoubtedly related to the group, in spite of its entozoic mode of life, which has made its structure much more modified than these less parasitic forms. These three known genera, Canu asserts, form, in respect to degrees of parasitism, a series: in such a series Goidelia japonica would have to be placed at the one end as being the most parasitic. Maxille—Alike in all stages and both sexes. They are greatly reduced, being merely small oval organs (fig. 10, mw.), unsegmented, and perfectly simple except for the fact that they possess three feathered sete. They are situated behind the mandibles, but are not covered by the lips; their insertion is on a level with the base of the mandibles, and, so far as it is possible to judge from preserved specimens, they seem incapable of movement as a whole, though doubtless the setose processes function instead. The maxille here agree very closely with the descriptions given by Canu in Hersiliodes and Giardella, though they are smaller and more simple in Gozdelia and have fewer processes. Canu says that in the above two genera traces can be seen of a division into a masticatory internal lobe, aud a palp-like external lobe—quite absent in my specimens, however. Koss. mann describes the maxilla of his Clausidiwn as “eime fast rehgeweihformige Maxille,” and from his fig. 6 the organ has indeed almost the appearance of a stag’s horns; he figures only three sete, as also does Claus for the maxilla of the same form: these two figures agree in all other respects with mine. First Maxillipede 2 .—This appendage differs strikingly in the various stages of development; I have been able to distinguish four very sharply defined stages in the female :— A.—The youngest and least differentiated condition, in which the appendage is scarcely more than a flattened disc or plate, bearing a fringe of short sete. A rudimentary palp can just be 220 ALICE L. EMBLETON ON A NEW discerned (Pl. 22. fig. 11, p), though in this stage it has not yet developed its two sete. B.—The organ at this stage is larger and more modified, it ends in a strong recurved hook (figs. 12, 13); its lower edge is fringed with sete, and about a third of the way along this border is another spine, almost a strong as the terminal one. Laterally, on the inner surface, there is a palp (figs. 12, 18, p), the rudiment of which was seen in stage A, but now it bears two long terminal sete; this palp is apparently a moveable organ and serves as a brush. C.—This is a very interesting stage, for one can make out within the chitin of the maxillipede two folded hooks; the terminal spine and sete of stage B remain, but the palp is not always present at this stage. It is obvious that the two immature enclosed hooks belong to the new appendage, and that they will unfold at the next moult when the old chitinous coat is shed (fig. 14). D.—This is the largest and final stage in the development of this appendage. All sign of setose fringing is gone along with the palp (fig. 15). It is now a clearly-outlined claw-like organ, ending in a powerful recurved tip, directed backwards; along the posterior surface is a second, stout, hooked process. It is probable that this organ is prehensile, and helps to fix the parasite securely to its host. First Maxillipede 3 —I find this limb is unaltered in all the stages in the male, except that in its very early stages it has sete on its surface as well as on its edge. It is a flattened petaloid plate of chitin, having its free border fringed with fine sete (Pl. 22. fig. 16); and is evidently greatly reduced, for it resembles the first and very undifferentiated form which this appendage has in the stage A of the female. The’ first is applied closely to the highly-specialized second maxillipede, from which, however, it can be detached by careful dissection of a specimen boiled in potash. Judging from the three known genera, I had expected to find this appendage highly developed in the male, as it isin all the three allied forms, where, however, it is alike in the two sexes, being a short solid organ with two joints and armed. with hooks and sete. That it should be so degenerate in Goidelia in the male is a very remarkable fact, and serves to distinguish this from any other form. Second Mawillipede 2 .—This is very feebly developed in all ENTOZOIC COPEPOD FROM JAPAN. 22M the stages, beg merely a papilla with but slight elevation above the surface. Relatively to the mouth-parts it is situated some- what postero-laterally (Pl. 21. fig. 2, mp.?), and is less con- spicuous in specimens boiled in potash than in those which have been treated with borax-carmine (or alum-carmine), and cleared in oil of cloves. According to other authors, the second maxillipede is a very important appendage in Clausidiwm, Hersiliodes, and Giardella, and, unlike the first maxillipede, it differs in these genera in the female and male. In Clausidium female, Canu says it has three parts (Claus describes four), bearing sete; in the male it has two well-marked basal joints and a terminal factor in the form of a short prehensile hook. In the female of Giardella it is as in Clausidium ; but in the male of Giardella and Hersiliodes there is one basal joint with two feathered sete, a median long joint with teeth on its mner ridge and two sete on its internal face; finally there isa terminal segment in the form of a spine recurved towards the end and almost as long as the rest of the maxillipede. Speaking of the family as a whole, Canu remarks :— Maxilli- pedes bien développés: les internes fournissant d’importantes différences sexuelles.” I cannot help thinking that Canu has named the two pairs of maxillipedes erroneously. Those which Claus calls ‘“‘vordere” and “ hintere,” I call respectively “ first ” and “‘second.” Canu uses the terms “interne” and “externe,” yet his description and figure of the pair designated “ interne ”’ coincide with the ‘“‘ hintere” of Claus and my “second”; by “externe,”’ therefore, we must conclude he refers to the “‘vordere” (or “‘ obere’’) or first maxillipede, which he says is alike in both sexes. In Goidelia japonica, however, neither the first nor the second maxillipedes are alike in the female and male: the first maxillipede is highly specialized in the female and small and unimportant in the male; the second maxillipede being, con- versely, important in the male and inconspicuous in the female. These appendages seem to offer peculiar difficulties, for Koss- mann has mistaken the first maxillipede for a second maxilla, concerning which Claus observes :—“‘ In Wahrheit aber ist Kossmann’s zweite Maxille der obere Maxillarfuss, und die als Maxillarfuss beschriebene Gliedmasse der Kieferfuss des zweites Paares. Der obere Jieferfuss sehliesst sich nach Form, Lager und Gebrauch am nichsten an die gleichwerthigen Gliedmassen der Coryceiden (Copilia, Sapphirina) an, und fihrt in weitere 222 ALICE L. EMBLETON ON A NEW Reduction zu den entsprechenden Maxillarfiissen der Ergasiliden. Ich vermag nur zwei Abschnitte zu unterscheiden ein mit mehreren Fiederborsten besetztes, in Stiletborsten wie gablig gespaltenes Endglied, und ein umfangreiches Staumglied, dessen Innenseite noch am oberen Ende einen breiten mit starker Borste bewaffneten Fortsatz entsendet. Der untere Maxillarfuss bewahrt eine voll- stindigere Gliederung, die sich auch an den gleichwerthigen Gliedmassen der Ooryceiden erhilt, nahert sich aber in weiblichen Geschlechte mehr den bei Oyclopiden (Cyclops, Oithina) beob- achten Formverhiltnissen. Auf zwei langgestreckten je zwel Fiederborsten tragenden Staumgliedern, welche knieformig gegeneinander gebent werden, folgen zwei kurze Hndglieder, von denen das obere sehr starke wenn auch kaum gekrummte Borsten triigt. Die sexueile Umformung der miinnlichen Kiefer- fiisse, die schon bei Coryceiden, ferner bei Bomolchus und Verwandten sehr augenfiillig ist, fuhrt in unserem Falle zur Ausbildung eines sehr eigenthiimlichen Klammer apparates, welches mit der bereits besprochenen Umformung des weiblichen Abdomens in Correlation steht und die Fixirung des Mannchens am weiblichen Kérper als andauernde Verbindung unterhilt. Die beiden unteren Glieder erscheinen aufgetrieben und bes- onders das zweite stark verbreitet; an Stelle der Endglieder finden wir einen beweglichen mit einer Fiederborste besetzten Haken, der dem beweglichen Arme einer Scheere ahnlich auf- wirts gegen der vorausgehenden Abschnitt eingeschlagen werden kann. Am oberen Winkel des letzteren erheben sich noch drei weitere Anhiinge, ein oberes Schaufelf6rmiges verbreitertes und gezahntes Sttick und zwei demselben dicht anliegende Zabn- stiicke welche offenbar in die vorgebildeten Gruben des weiblichen Abdomens eingreifen.” Second Masxillipede 3 .—This is a very speeialized organ in the adult, subserving as it does the function of prehension—this, in the female, is performed in all probability by the hooked first maxillipede ; in both sexes, of course, the second antenna is also modified as a fixing organ. In the immature male the second maxillipede is merely a blunt cone made up of three segments; the basal portion is swollen and rounded (Pl. 22. fig. 18), with a small spine on the line of union between this and the next joint; this second segment is an undifferentiated ring, which in its turn bears a spine where it connects with the last segment—but this spine is on the opposite ENTOZOIC COPEPOD FROM JAPAN. 223 side to the previous one. It is difficult to decide if the terminal papilla-like part is a distinct segment or only a portion of the second ; it is, however, quite free of spines or hooks, and it ends bluntly. There seems to be no stage between this and the adult form, nor can I find any stage earlier than this ; apparently there are but these two stages in the male. The adult organ possesses a very characteristic chelate organi- zation. There is a basal joint which, compared with the next seoment, is rather slender, and is directed forwards and outwards (Pl. 22. fig. 18); articulating with this is the main joint, large and strong, on the distal end of which the chele work. The chelz are composed of two parts—a pad (pd.) and a hinged beak (6%.); the protruding pad on which the beak bites is on the inner surface of the limb; at its base itis surrounded by a circlet of fine set; the moveable beak is hinged on in such a way as to close down on this pad, it being worked by powerful muscles. There are no set or spines on this appendage in the adult; it is essentially characteristic of the male, there being nothing comparable to it in the female. The two maxillipedes are so arranged that they slope forwards and inwards, almost meeting in the middle line (Pl. 21. fig. 3, mp.”), the chele thus working in front of the mouth-parts. The first and second joints are bent on each other, forming a very pro- nounced elbow, which points outwards, for the first segment slopes forwards and outwards, while the second slopes forwards and inwards, the angle between being about 60°. Close to the basal joint is the first maxillipede, which is con- siderably flatter in the adult than in the immature stage; in the adult it has the form of a very flat plate of chitin, with its rounded antero-lateral border fringed with sete. This appendage in Goidelia japonica differs fundamentally from the corresponding organ in Clausidium, as described by Claus and Canu. These authors show that in these cases (Clausidium, Hersiliodes, and Gardella) it is armed at the end with a powerful recurved spine in the male (Gardella), rather more reduced in Hersiliodes, yet in no instance is it a chelate structure as is here shown to be the case. According to Claus, there is a slight resemblance in Clausediwm, though even there nothing occurs of the nature of chelew, and tke basal joint has a stout seta, while there are numerous defensive weapons at the extremity: “an 224 ALICE L. EMBLETON ON A NEW Stella der Endglieder finden wir einen beweglichen mit einer Fiederborste besetzten Haken der den beweglicher Arme einer Scheere iihnlich, aufwiirts gegen den vorausgehenden Abschnitt eingeschlagen werden kann.” However, in this EHchiuroid parasite there are no accessory parts to this appendage, only the clear-cut chela closing down to its simple pad. The whole organ forms a powerful prehensile apparatus, and seems to be in no way connected with the function of alimentation. Since Kossmann calls the first maxillipede the “second maxilla,” the male maxilli- pede is, according to his figure, an unjointed upwardly-curved claw, while in the female it is a three-jointed apparatus bearing sete of various kinds, that of the male being devoid of such processes and a much smaller organ than in the female. In Goidelia japonica, as has been shown, this appendage is a mere vestige in the female, and in the male is a powerful chela (Pl. 21. fig. 3, mp.’; Pl. 22. fig. 18). Thoracic Legs 1-4.—In the genus “ Hersilia” Canu says: “ Premiére paire des pattes thoraciques trés différentes des trois suivantes semblables entre elles et transformée en organes d’adhesion. Pattes des deuxicme, troisicme et quatrieme somites thoraciques formées de deux articles et de deux rames 3-articulées. Pattes thoraciques de la cinquiéme paire simples, 2-articulées et aplaties.” In the genus Giardella, according to Canu: “ Pattes thoraciques des quatre premicres paires de forme normale, biramées, 4 rames 3-articulées; de la cinquiéme paire simple, 2-articulées et aplaties.” Canu looks upon the head with the first thoracic segment as constituting the first metamere ; the abdomen thus has five segments, the furca forming a sixth. “ C’est la premicre somite qui porte les orifices génitaux dans les deux sexes. Chez le male, ces orifices sont au nombre de deux situés symétriquement sur la face ventrale 4 Ja limite postérieure du segment; ils sont re- couverts par une paire de pléopodes lamelleux terminés par une longue soie. Chez la femelle, les deux ouvertures génitales sont placées symétriquement dans une situation latéro-dorsale au tiers antérieur du segment.” In the genus Hersiliodes the first pair of legs, according to Canu, “sont biramées et triarticulées, sans déformation et parfaitement pourvues de soies natatoires, il en est de méme dans les trois paires suivantes, toutefois la rame interne de la quatriéme paire est plus allongée et peu riche en soies natatoires.” ENTOZOIC COPEPOD FROM JAPAN. 225 According to Claus in “ Hersilia” apodiformis: ‘von den vier zweliistigen Beinpaaren zeigt das vordere dem Cephalothorax angehorige Paar eine se bedeutende Umformung, dass Philippi diese Gliedmasse als Kaufuss in Anspruch nehmen konnte.” Kossmann says of the same species:—‘‘ Die Spaltfiisse des ersten Paares sind zu einem kriftigen Bewegung und Haft- apparat umgewandelt welche im Verein mit den ibrigen Schwimmfiissen unseren Copepoden ein rasches Umbhergleiten auf dem glatten Panzer seines Wirthes moglich macht. Die beiden Grundglieder scheinen mit den zwei ersten Gliedern des inneren Spaltastes zu einer Masse verwachsen zu sein, ohne dass die Contouren der einzelnen Gleder verwischt sind, das eine Grundglieder triigt beim Mannchen eine lange Borste, welche beim Weibchen zu einem kolossalen riickwirts gericht- eten Dom oder Zahn entwickelt ist. Das andere Grundeglied trigt einen kiirzeren, etwas nach aussen gekrummeten Zahn, der dicht neben jenem nach aussen zu liegt. Die kriimmig dieses Zahnes fiillt nun eine Saugscheibe aus, welche umgefahr auf der Mitte des inneren Spaltastes angebracht ist; an der Spitze des letzteren befindet sich noch eie zweite kleinere, nebst drei Zahnuchen. Der iiussere Spaltast ist swach entwickelt, viel kiirzer als der innere, beim Mannchen mit lingeren und kurzeren Borsten bedeckt, die beim Weibchen durch Dornen ersetzt sind.” Goidelia japonica differs very strikingly from all these various descriptions. The first four pairs of thoracic appendages are very highly specialized, and all alike. Each limb is biramous, and possesses a basal part of two segments, which is attached to a median sternum; this sternum is nothing more than a flat plate of chitin, with a transverse notch in the centre making it appear almost bi-lobed. Posteriorly its free edge is fringed with sete ; laterally (Pl. 22. fig. 19, s) there articulates with it the first and smallest segment of the basal part of the limb (6°) ; this segment is small and irregularly shaped, bearing on its inner surface one long feathered seta which runs parallel with the setose edge of the sternal plate (fig. 19). The other basal segment is three or four times as large as this joint and rounded, its anterior (or outer) surface being smooth; on its inner surface (Pl. 22. fig. 19, 6°), near the articulation with the first basal joint, there is a very powerful short spine recurved and feathered, its insertion being apparently sunk in a small pit. The inner ramus of the limb is attached on this surface near the LINN. JOURN.— ZOOLOGY, VOL. XXVIII. 16 226 ALICE L. EMBLETON ON A NEW spine; both rami have three articulations. The inner branch bears long feathered setze on its inner face, one on each of the first two joints, and six on the terminal joint, three of which arise laterally and three at the extreme tip (Pl. 22. fig. 19). The opposite surface is clothed with soft fine sete. The outer ramus is also adapted for locomotion, the first joint exhibiting distally on its outer face a short, sharp spine; a corresponding spine occurs on the second joint, which has in addition a long bearded seta on its inner side. The terminal joint has the sharp stout spine on its outer face, but on its inner surface and at the tip there are four long feathered sete (fig. 19), with a fifth one, much smaller, pointing outwards. The whole outer surface of these three joints is covered with soft fine sete. The thoracic appendages of Goidelia japonica possess no sucking-dises like those occurring in the previously-described forms. Thoracie Leg, 5.—In both sexes this is a uniramous organ, very feebly developed. It is practically the same in male and female, though in the latter (adult) it is broader and shorter relatively than in the male; in both cases it consists of three short equal segments, the first two each armed laterally with one long seta, the terminal segment carrying two (PI. 21. figs. 2, 3). Abdomen.—The first abdominal segment of the mature female is conspicuously larger than any of those following it. In reality it is not one segment, but is formed by the fusion of three; on its ventral surface it is pierced by a pair of genital pores (fig. 2, 7) placed obliquely. The abdomen in the adult male is marked with a lateral spine, borne on the first segment (fig. 3, 7) ; near the base of each of these the minute genital openings are placed. The ventral border of this segment is fringed posteriorly with sete (fig. 3) ; this edge is not parallel with the other rings, but slopes forwards to the centre, where there is a clearly-defined square area (#)— the significance of this is obscure, but it may have some sexual function to perform. Furce.—The furce are short and offer no evidence of dimorph- ism between the sexes; each possesses a long inner seta and a shorter external one. The groove between the furez runs far forwards on the dorsal surface, and the anus opens into it on the last undivided abdominal segment. Ventrally the groove is less conspicuous, and does not continue so far forwards as on the dorsal surface. ENTOZOIC COPEPOD FROM JAPAN. 227 GOIDELIA JAPONICA, Nov. gen. et sp. Diagnosis of Generic Characters.—Body flattened, segmented, and cyclopoid ; sexes equal in size, dimorphism only being apparent on closer examination of the appendages; the male not attached to the female ; no eyes. 1st antenna alike in female and male ; composed of 5 segments. 2nd antenna alike in female and male; composed of 4 segments, forming a characteristic elbow ; the three terminal segments con- stitute a powerful hooking apparatus. Mandibles alike in female and male; very small and claw- like; with xo accessory parts (such as occur in the other known genera). lst maxillipede Q differs much in the immature stages; in the adult it is a claw-like organ with two recurved hooks (in the other genera this appendage is alike in both sexes, being very highly developed). lst mazillipede 3 alike in all stages; very degenerate, being only a fringed plate. 2nd maxillipede Q alike in all stages; very degenerate, being only a papilla (Gn the other genera this appendage shows sexual dimorphism, but in both sexes it is very highly developed). 2nd maxillipede 3, different in the various stages; very specialized as a powerful chela (in the other genera there is no suggestion of a chelate apparatus). Thoracic legs 1-4 all alike; biramous; no sucking-discs (such as occur in the other genera, where also the Ist leg is unlike the others). Thoracic leg 5 uniramous in female and male. Abdomen of 6 segments ; in the female the first three are fused. Furce show no difference in male and female; each furca bears a long and a short seta. TRICHODINA (n. sp.). (Pl. 22. fig. 20.) The rectum of Hchiurus unicinctus gives shelter not only to the parasitic Copepod Groidelia japonica, but also to a minute Infusorian: being so excessively small, this creature would in all probability have escaped my notice had it not occurred in such abundance. I first saw it in a series of transverse sections of the whole Echiurus, where it was very conspicuous in all sections of the rectum, for the methylene-blue which I was using as a double stain settled far darker in these little parasites than in any 16* 225 ALICE L. EMBLETON ON A NEW of the surrounding tissue. On examining the contents of the rectum, I found that it contained immense numbers of these Infusorians, but having only preserved material at my disposal I was unable to investigate more than the most salient features; in fact, to make out these points many different reagents and stains had to be used. On the whole, I found it best to first clear the material of corrosive sublimate by immersion in iodine solution, then stain with alum-carmine, finally clearing in oil of cloves. As is seen in fig. 20, the form of the body is quite spherical when looked at from above (as in i.) or from below; iu. shows a lateral view. It is not easy to say without watching living speci- mens, or at least examining fresh material, which is the anterior and which the posterior end. Above and below there is a circlet of cilia; at one end they encircle a depressed pit-like area around which there seemed to be signs of a skeletal ring of spines or hooks (ii., 2). At the opposite end the cilia surround a dome-like protuberance—which may, however, have been thus forced out as a result of post-mortem contraction. In the rim in which the set of cilia are inserted, I saw, in many specimeus, indications of a gullet, or esophagus ; this would point to the fact that this is the anterior end, which would go first in swimming. Situated just beneath this crown of cilia there is an enormous darkly- staining nucleus; it is very long and constricted at intervals so: as to look like a string of beads (zc). A clear spot was present near the nucleus, whick may be the contractile vacuole, or a reservoir (ve). So far as [ can form an opinion from the material at my disposal, this is a new species of Zrichodina, a peretrichous Infusorian related to Vorticella. ‘The common species is found creeping about the surface of Hydre; but I find that a repre- sentative of the genus has been met with (by Rosseter *), living an endoparasitic life in the viscera of the newt; very few were found in the testes, but myriads occurred in the renal organs— it is stated that long dearth of water in no way hurts them.. This Japanese species may be identical with the form occurring in the newt, for it also leads a truly entozoic life within the rectum of its host ; but as this alone is not sufficient to justity the application of a new specific name, I prefer to leave the question of species open, pending a fuller knowledge of the creature and of the habits of the accepted species of the genus. Cambridge, Nov. 1900. * Journ. R. Mier. Soc. vi, (2) pp. 929-938, 1 pl. Embleton. Laur. Soc. Journ. Zoou. Vou XXVIUL Pr, 2]. Parker & West ump. ALE del. M.P Parker lith. pian te eta tea Miri Embleton. Liusn. Soc. Journ. Zoor. Vou. XXVIII PriGae ALE. del MP. Parker lith. Parker & West amp. A NEW ENTOZOIC COPEPOD. ENTOZOIC COPEPOD FROM JAPAN. 229 EXPLANATION OF PLATHS 21 & 22. Reference Letters. A=adult 9. B=adult §. C=immature 2. a'=iirst antenna. a?=second antenna. /'=first basal sezment. 6?=second basal segment. 6/.=beak. f=furea. g=genital pore. h=hooks, /./.=lower lip. /.v.=upper lip. md.= mandible. mp.'=first maxillipede. mp.?=second maxillipede. ma.= maxilla. #¢e.=nucleus. e@=cesophagus. pd.=pad. p=palp. s=sternal plate. 7¢h.'-°=thoracie legs 1-5. vc.=vacuole. «=curious glandular patch between Ist and 2nd abdominal segments in the adult male. Goidelia japonica. | Figs. 1-19 drawn with the camera lucida.| Fig. 1. Outline-drawings of :— A. Adult Q. B. Adult ¢. C. Immature 9. Superposed on a correspondingly magnified millimetre-scale, the divisions of which are to 1/10 mm. . Stage B of 1st maxillipede 2, drawn from the ventral face, and magnified considerably more than figs. 5 & 4. 3. Adult male from the ventral side, showing the peculiar modification of the two basal abdominal segments. 4. Dorsal view of young female. [Figs. 5, 6, 7, 8 all drawn on the same scale to show relative sizes. | 5, First antenna. 6. Second antenna, @, Liceul ” 3. 8. Mandible showing terminal claw. 9 Pe on an enlarged scale. 10. Mouth-parts iv situ, showing the upper and lower lips, mandibles» and maxilla. Figs. 11, 12, 15, 14, 15.—Stages in the development of the first maxillipede 2 —all drawn on the same scale. 11. Stage A, showing the undifferentiated condition. 12. ,, 5B (ventral view). 13. ,, B (dorsal view). 14. ,, C, showing the new hooks folded up within. 15. ,, OD, adult organ. 16. First maxillipede ¢. 17. Second maxillipede 4, immature stage with chelx undifferentiated : the first maxillipede is shown én situ. 18. Second maxillipede 4, adult chelate form. 19. One of the first four thoracic legs, showing its relative position to the sternal plate. Trichodina ? sp. Fig. 20., Zrichodina, showing the curious beaded nucleus in various positions, with the two circlets of cilia and the clear vacuole (contractile vacuole ?). 230 MR. A. T. WATSON ON THE STRUCTURE On the Structure and Habits of the Polycheta of the Family Ammocharide. By Anyoup T. Watson, F.L.S. {Read 20th December, 1900. | (PLatEs 23-25.) Ir is a somewhat curious fact that, although the members of this interesting family of tubicolous Polycheta are widely distributed, and some of them abundant, upon the British shores, they have thus far received comparatively slight notice at the hands of British naturalists. Yet almost sixty years have elapsed since Delle Chiaje first brought these worms to light, and figured a specimen in some detail, to which he assigned the name Owenia. During those sixty years, contributions towards a full descrip- tion have, from time to time, been made by Grube, Koller, Claparéde, von Drasche, McIntosh, Cunningham and Ramage, Gilson, Lo Bianco, and Ogneff, all of which have been ably summarized by de Saint-Joseph in his ‘ Annélides Polychetes des Cétes de France.’ Nevertheless, as much remains to be learnt of their structure and habits, the following notes may perhaps be regarded in part as supplementary to Prof. Gilson’s paper, since it was on his suggestion (27. p. 381) that I have paid special attention to certain of the points hereinafter men- tioned. They are the result of observations extending over a period of about three years, some of which were made at the Marine Laboratory at Port Erin, whose Director, Prof. Herd- man, has rendered me most valuable assistance ; the larger part at Sheffield, on specimens obtained from the coasts of Lancashire and North Wales; and also on foreign specimens, both living and preserved, for which I am indebted to Doctors Dohrn and Lo Bianco, of the Zoological Station at Naples. I also acknowledge, most gratefully, my indebtedness to my friend, Dr. Fauvel, for valuable advice on points of technique and bibliography. The methods of research described in his admirable work, ‘Les Ampharétiens, have frequently been useful to me. In the absence of an English description of the Ammocharide (a family created by Grube), it will perhaps be well, in the first. place, to give a general account of their characteristics, embodying my own observations with those of other naturalists, AND HABITS OF THE AMMOCHARIDZ. 231 and thereby making the later portion of my paper more intel- ligible. The body of the animal is cylindrical, and especially rigid in the anterior portion. It consists of from twenty to twenty-seven segments (Pl. 25. fig. 20), and varies in length from about 30 mm. in the English to 90 mm. in the Italian species. The diameter at the anterior end measures from 1 mm. to 3 mm. The cephalic segment is slightly swollen and bears a crown of flat, laciniated, tentacular branchie, six in number (0.¢., Pl. 23. figs. 1, 2 & 3) (three on either side of the body), which are ciliated on the inner surface. ‘These are absent in the youngest speci- mens, the mouth being then surrounded only by a funnel-shaped lip, which is quite entire except for a ventral notch. The processes first appear short and squarish in form, with a slight indent on the upper edge, which gives the margin a castellated appearance. In fully-grown living specimens their height somewhat exceeds the diameter of the cephalic segment, but in preserved animals it is slightly less. Each process has a wide, very flat base, giving rise to three or four branches, which are subdivided into seven or more twigs, each terminated by two small rounded mucus- secreting lobes. The crown and branchial processes are well supplied with coelomic fluid; and in the living specimens the latter have, under favourable illumination, a beautiful appear- ance, due to the branches from the dorsal blood-vessel and to the network of capillaries with which they are furnished. In British specimens the processes, viewed by transmitted light, are semitransparent, pale greenish blue, or yellow, and tinted with red; whilst those from Naples are blood-red and in parts coloured with a reddish-brown pigment. In cross-section the branchiz are somewhat horseshoe-shaped, the concave side directed towards the mouth. They are divided into two lateral groups dorsally by a crescent-shaped lobe (c./., Pl. 23. fig. 2), which is figured and described by von Drasche (12. p. 13) as the head containing the brain. This organ possesses another inter- esting feature, which seems to have escaped his notice. From his admirable drawings (12. pl. i. fig. 3), it would appear that the inner wall of the lobe is in unbroken continuity with the ectoderm. Favourable sections, however, clearly show that this is not the case, and that for a short distance at the apex of the erescent the inner wall is entirely separated from the outer body-wall and forms a kind of ciliated upper lip, which, by 232 MR. A. T. WATSON ON THE STRUCTURE means of a network of muscular fibres, can be opened or tightly closed at will (c.p., Pl. 24. fig. 8). Free communication can be thus established between the coelomic fluid-containing cavity and the surrounding sea-water. The ventral separation of the branchial processes is, doubtless, a development of what ap- peared originally as a “‘ notch” in the funnel-shaped lip. It is interesting to note that one member of this family, Myriochele (Mgrn. 4. p. 211, and 13. p. 410) is distinguished by the absence of cephalic branchie, development of these organs having apparently been checked at an early stage. During life the interior of the base of the crown is occupied by three lobes (Pl. 28. fig. 2), namely, the dorsal (c.l.), already mentioned, and two lateral ones (J./.) formed by swellings at the base of the ventro-lateral branchial processes, which, distended by the ccelomic fluid, and meeting in the centre, may form more or less completely a kind of vestibule—an arrangement which appears to have hitherto escaped notice. Beneath these lobes, which almost disappear owing to contraction in preserved speci- mens, is the mouth (m., Pl. 23. figs. 1 & 3), a transverse opening situate a little distance from the inner dorsal surfaces, the space between the mouth and the ventral boundary being occupied by the peculiar “‘ Lippen-organ” (J.0., figs. 1,3 & 7), with its massive bilobed lips, first figured by von Drasche (12. p. 6, pls. 1. & i1.), the funetions of which will be hereafter described. The branchial crown is marked off from the “ thoracic” region by a reddish-brown fold, which commences laterally and becomes very prominent dorsally (c., Pl. 24. figs. 7 & 8). There is a single black or dark-coloured ‘“‘eye-spot”’ on the ventral face of the crown (e., Pl. 25. fig. 20), situate at the base of the branchiz on either side of the opening. The segments are neither divided by distinct constrictions, as in Arenicoia and Clymene, nor are they sharply divided from each other; only here and there does the body contour show a slight incurving, indicative of their demarcation (Pl. 25. fig. 20). The “ thoracic region” comprises the buccal segment and three other short segments fused together, the only external indica- tions of which are three bundles of simple yellow, slightly pinnate bristles. Of these, the first two ouly are visible in the ventral view (Pl. 25. fig. 20); as the third bundle, consisting of much shorter and less numerous bristles, is situate quite dorsally. Internally there is a septum between the buccal AND HABITS OF THE AMMOCHARIDS. 233 segment and those which follow it, but the next three segments are without septa and enclose a single chamber. The succeeding three “abdominal” segments increase in length, the seventh being sometimes four or five times as long as the buccal and “thoracic” segments taken together. From the eighth segment their length is rapidly reduced, the penultimate segment being very minute. The “thoracic” region and each of the “abdominal” segments is internally cut off from its neighbour by a septum; and with the exception of the anal, the penulti- mate, and antepenultimate segments, the anterior extremity of each is indicated externally by a bundle of dorsal or dorso- lateral sete. These sete are slightly pinnate at the extremity. Each bundle of sete is followed by a torus, which, in the abdominal region, is so long that the tori (t¢., Pl. 25. fig. 20) on the two sides almost unite at their extremities, and, at this part, form approximately a complete girdle; though in the last segments the tori will not extend over more than about one- sixth of the circumference. The tori of the first four segments are red owing to the presence of blood. From each torus proceed a considerable number of minute bifid uncini (PI. 28. fig. 4), which are disposed in very regular, close-set, longitudinal rows (w., Pl. 23. fig. 6); and it has been estimated by de Saint- Joseph (29. p. 401) that there will be about 7600 uncini in a single turus in the third “abdominal” segment. The total equipment of an entire animal is estimated at from 150,000 to 450,000 hooks. This immense provision of uncini is necessitated by the bur- rowing habits of the animal, and renders it most difficult to expel the living worm, uninjured, from its tube. The anterior part of each uncinus is preceded by an enlargement, which tends to retain it in the epidermis in which its shaft is buried, though this provision is not always sufficient, as uncini may occasionally be found torn from the tori and embedded between the layers of the inner lining of the tube. Sickly worms, too, occasionally expel their uncini from the mouth of the tube. The shaft of the uncinus is flat and straight, and terminated proximally (Pl. 23. fig. 4, c’) by a thread-like elongation, which curves sharply backwards and penetrates the basal membrane. The uncini are actuated by appropriate muscles, and are used for seizing-purposes at will. There are no ventral uncini in the “thoracic ” segments, neither are there any dorsal sete 234) MR. Ae T. WATSON ON THE STRUCTURE in the penultimate and antepenultimate. The anal segment has no sete of either kind, and is truncated so that the ventral face, which is terminated by two small lobes (a.p., Pl. 25. fig. 20), extends slightly further than the dorsal. These lobes, as will be seen hereafter, are of special interest. The margin of the anus is ciliated and crenated. In a dorsal, external view of the animal, longitudinal, paired, milk-coloured bands are seen extending as far as the fifth torus (12. p. 10). These are due to a thickening of the epidermis. The first pair take their rise laterally, and, curving towards the back, terminate at points against the bundies of sete of the first torus. From these points there spring a wider pair of bands, which, in the form of bows with their convex sides facing each other, tie together the sete of the first torus to those of the second. Similar bows bind together the sete of the three following segments. An olive-green zigzag canal runs almost from end to end of each of the bands of the second “abdominal’’ segment (e.c¢., Pl. 23. fig. 6). Nephridia have not yet been found in the Ammocharide ; but it seems probable that these canals, with their internal funnel-shaped mouths and external slit-like openings, represent those organs, though Gilson is mistaken in supposing that they play a part in the emission of the genital products (27. p. 379). These, as I shall show, are discharged through characteristic pores at the anal extremity. A lateral view of the “ thoracic” segment shows an interest- ing point of adaptation of the animal to its mode of life. Von Drasche (12. p. 9) has noted the fact that, although the markings are not quite constant, there is, in the dorsal part of the collar, a transverse patch of brown colour, often divided into two portions by the dorsal line. This coloration extends as a line on either side round the fore-edge of the thorax, and swells into the round pigment-spots (eye-spots ?) already mentioned. From these spots the lines descend, and, still going forward, meet at an acute angle on the ventral line, about the level of the first bundle of bristles. He also observes that, in the anterior part of the animal, the course of the nervous system is practically indicated by this coloration ; the dorsal patch marking the brain, whilst the lateral lines indicate the cesophageal commissures uniting in the large ganglion, from which springs the ventral cord. This consists of a dotted band, without ganglia or large nerve-fibres, which runs the whole length of the animal. AND HABITS OF THE AMMOCHARIDA. 235 The entire nervous system is fully described by von Drasche (12. p. 13), who observes that it is very rudimentary in character, and lies outside the muscular layer near the surface of the epidermis. It may be asked whether the localization of the coloration over the nervous system may not be connected with the animal’s sensitiveness to light? Some of my sections seem to suggest the possibility, but at present this must remain undetermined. I am able to show that a special function is performed by the triangular ventral area (v.a., Pl. 25. fig. 20) bounded by the two lines which descend from the “ eye-spot.” The body-wall is at this point very much thinner than that of adjoiming parts, and modified, as I shall show more fully, in relation to the building- habits of the animal. The structure of the creature is remarkable in many other respects, and for minute anatomical descriptions reference should be made to the papers of Claparéde, von Drasche, Gilson, and Ogneff. Speaking generally, I may point out that a cuticular covering to the epidermis is absent, except on the anterior portion of the body; that the underlying circular muscles are limited to the “thoracic” region; and that the longitudinal muscles, which are exceptionally powerful, form throughout its entire length an inner lining for the body-cavity — a layer which is continuous except where it is interrupted by the dorsal and ventral mesenteries of the alimentary canal. These powerful muscles impart considerable rigidity to the fore part of the body ; a feature which gradually disappears posteriorly, the thickness of the muscular layer in that part being greatly reduced. According to Gilson (30. p. 95) there are no peritoneal parietes distinct from the longitudinal muscular layer, his view being that these two layers are replaced by one musculo- glandular layer which, amongst other things, is excretory in function. This view is, however, opposed by the observations of von Drasche (12. p. 20) and Ogneff (32.). A striking and characteristic feature of the Ammocharida, which has from the first attracted the attention of naturalists, is the existence in the anterior segments of peculiar, very long, cylindrical, rigid glands (¢.g., Pl. 28. fig. 6) of considerable calibre which hang loosely in the body-cavity. Hach is attached to the body-wall by one extremity which, greatly reduced in diameter, forms a minute duct running through the body-wall, opening 236 MR. A. T. WATSON ON THE STRUCTURE between the end of the torus and bundle of sete of that side of the segment to which the gland belongs. There are sometimes seven pairs of these glands, one pair to each of the first seven segments, but those of the third segment may be either rudimentary or non-existent. These are the thread-glands (‘ glandes filiéres’’) of Claparéde (6), the struc- ture of which has been fully worked out by Gilson (20). Their function is to secrete—by exudation from the epithelial cells of the wall of the gland (20. p. 317, & 29. p. 402)—a very thick viscous liquid containing bundles of exceedingly fine colourless threads, which are used by the animal in the formation of the membranous lining of its tube, but not, as some writers have suggested, for the attachment of its external stony coveriny. The secretion for this latter purpose is supplied by the “Lippen- organ ”’ (/.0., figs. 1, 3, 7 & 9). The esophagus, which is almost as long as the “ thoracic” segment, descends directly from the mouth, and is followed by the intestine, which, with a series of swellings and constrictions, runs straight to the anus. De Saint-Joseph notes (29. p. 403) that in the third and fourth “abdominal” segments the intestine is green owing to the glands which cover it. The alimentary canal passes through the segmental septa, and in addition thereto is sup- ported throughout by a dorsal and ventral mesentery. For the greater part of its length it is surrounded by the dorsal vessel, which forms a peri-intestinal sinus. This sinus bifurcates and leaves the esophagus on entering the thoracic region. Hach branch then goes forward, and, after skirting the nearest group of branchial processes, descends, and, reuniting with its fellow, gives rise to the ventral vessel. Numerous short ceca arise from each side of this vessel, forming round ampulle, of which de Saint-Joseph states (29. p. 403) that he has counted as many as forty in the third abdominal seement. The blood is red. The septa, which are membranous and traversed by fine muscular fibres, have been specially studied by Gilson (27). They may completely separate each segment from its neighbour, but by means of valves in the interseemental membranes interecommuni- cation may be established. These valves (Pl. 23. fiy. 5) are of two kinds, namely, a simple slit, making a “flap” valve (f-v.), opening like a door towards the anterior, which is always situate dorsally; and a sphincter valve (s.v.), which is placed ventrally. I have been fortunate AND HABITS OF THE AMMOCHARID. 237 enough to see both kinds in action, and note that the coelomic fluid (in which sometimes ova were floating) almost invariably passed steadily through the sphincter valves when travelling towards the anus, and rapidly past the large slit or “ flap” valve when going in the opposite direction. In the absence of assistance from circular muscles, it will be seen that the latter form of valve is specially well adapted for rapid release of the imprisoned coelomic fluid, and that the animal is enabled speedily to retract the posterior portion of its body, in case of danger, by means of its longitudinal muscles. My observations show that,. with the exception of the buccal and “ thoracic” segments, and possibly the smaller ones towards the anal extremity, each side of each segment is provided with two valves, one of each kind. Those attached to the septum at the base of the “thoracic ” segment are described by Gilson (27. p. 384), and are excep- tionally powerful. As will be seen later, this is apparently another adaptation connected with the habits of the animal. The foregoing will probably suffice as a general description of the structure of the worms, and we may next consider their habits. They are exceptionally bardy animals, whose welfare is apparently more dependent upon a good supply of rather muddy sand than upon a liberal allowance of oxygen, since they thrive under conditions so various as those which prevail respectively on the British shores at low-water mark, and at a depth of 2975 fathoms near the Caribbean Islands (18. p. 410); whilst both Dr. Fauvel and I have found them to live for years in small’ aquaria destitute of alga and without change of either sea-water or sand, the only attention they received being an occasional raking of the surface and the addition of river-water to eompen- sate for evaporation. They are gregarious, and dwell embedded near the surface of the sand in flexible sand-covered tubes. The tube is unique in structure, and, when fully extended, frequently two or three times as long as its inmate. It is well described by Grube (2), in 1846, as a transparent gelatinous tube, the middie part of which is supported with great firmness by a crust of grains of sand and fragments of shell which, when flat, are ““olued ” on by their highest edges so as to overlie one another like the tiles on a roof. Gilson (20. p. 822) has given a most careful description of the minute structure of the tube, and although I am unable entirely to confirm his speculations on the mode of its formation, it is 238 MR. A. T. WASTON ON THE STRUCTURE satisfactory to find that his observations on the appearance of the finished tube are quite in accordance with my own, and con- sistent with what we might expect as the result of the process of building, which I have had the good fortune to witness on several occasions, and shall hereafter describe. From the fact that the sand in the Bay of Naples is singularly devoid of minute flat particles of stone or shell, his specimens would probably not show that most interesting characteristic, “ imbrica- tion,’ mentioned by Grube (2). Gilson consequently does not -refer to it; but I know from experience that when suitable material is supplied, these worms gladly employ it in the manner described, and the process by which the desired result is attained is wonderfully ingenious. De Saint-Joseph remarks (29. p. 898) that the fragments of shell are sometimes fixed at a right angle by the edge ; but the difference between his deserip- tion and that of Grube is due to a difference in the state of contraction of the tube at the time of observation. I have frequently seen a tube present both appearances, separated by an interval of, say, half an hour. The tube (PI. 24. fig. 12) consists of two distinct parts: (1) an external covering of foreign matter (s.c.) which may be very minute fragments of shells of molluscs, calcareous tubes of annelids, grains of quartz, or flat particles of other rocks ; and (2) an internal tube (¢.¢.) which is translucent, colourless, flexible, elastic, and exceedingly tough—this part being secreted by the worm. The tube as a whole differs, I believe, from all others in that the particles of foreign matter, instead of being more or less attached to one another, are designedly kept entirely apart, firmly affixed only to the inner membranous tube, with the attachment limited to one side or edge (Pl. 24. fig. 12). The structure thus pro- duced is thoroughly protective, and, at the same time, like a coat of mail, it accommodates itself easily to the violent contor- tions of the body, which are involved in the habits of the worm. Carrying its tube with it, the animal is capable of burrowing through the sandy sea-bottom, and consequently the position of the tube in relation to the surface is by no means constant, though it is usually more or less vertical. It is interesting to note that the tube itself has an anterior and a posterior end, and is always buried in a definite direction. The anterior, being the growing end (at all events as regards the stony covering), is invariably nearer the surface ; whilst the b) AND HABITS OF THE AMMOCHARID®. 239 posterior, which never receives additions except to its internal sheath, may be embedded deep in the sand. The membranous tube terminates at each end with a conical, very elastic tip (c.t., Pl. 24. figs. 12 & 13), which projects more or less beyond the stony covering, and has at its extremity a very minute perforation. At the posterior or buried end, this tip is often preceded by a length of from 10 to 15 mm. of naked tube (Pl. 24. fig. 13), which Gilson (20. p. 322) has mistaken for a secretion at the outset of the animal’s existence. It is simply an extension (possibly late in origin) underground, where the worm is unable to affix its stony covering. The animal is capabie of reversing in its tube, and thus exploring the sand at either end, and, like the Terebellide (17), the Ammocharide invariably adopt the sanitary method of ejecting the excreta, at the anterior end of the tube, into the open sea. When the anterior end of the tube of an English specimen is seen under a low power (PI. 24. fig. 12), the imbricated arrange- ment and transparent conical extremity of the internal sheath are very noticeable, the free edge of the bits of stone being directed upwards, whilst the tip, in which is the minute per- foration, is drawn in, as is common when the animal is just about to emerge. Gnulson (20. figs. 21, 22 & 23) gives excellent figures illustrating the structure of the tube. His longi- tudinal section of the posterior naked end shows that this membranous sheath consists of a number of layers, and that the lengthening of it is produced by internal additions, which advance by stages beyond the extremity of the external ones, each advance being marked externally by a stepped appear- ance. This section also shows that the tube is occasionaily strengthened by new external additions on the face of the layers previously formed. Both these points are clearly visible in the drawing of this portion of the tube of Owenia (Pl. 24. fiz. 13). A cross section shows the internal layers to be concentric. By removing the stony covering from the body of the tube, Gilson demonstrated that the inner sheath is imperforate, and constructed as de- scribed from end to end. By means of tangential sections he found (20. p. 3238), in the wall of the inner tube, a some- what irregular system of coarse striz which, under a high power, he was able to resolve into a large number of very fine, irregular, longitudinal strie, and to recognize as the filaments 240 MR. A. T. WATSON ON THE STRUCTURE produced by the thread-glands. Gilson’s figure, 23, is specially interesting as delineating a section through the complete tube not stripped of its earthy covering. The section was made by a method specially devised by him for the purpose, and shows both the inner sheath of concentric layers and the outer irregular zone. The former is the part first laid down, as shown in the sections of the tip; whilst the latter is formed of masses of foreign materials subsequently affixed to the former by secretion, in which the various fragments are seen to be more or less completely embedded. My own examination of the structure of the inner tube was made by the ruder method of scraping off the sandy covering, and, after subjecting the inner sheath to maceration in distilled water, tearing the layers asunder. In this way I found it easy to separate the external rugged layer (which represents the cement by which the stony covering is attached) from the tube proper, aud, by means of stains, to establish the fact that the fibres in the former (which are embedded in a homogeneous secretion) may assume either a radial or any other direction; whilst those of the sheath take a course which is usually more or less longitudinal or transverse in relation to the tube. This somewhat lengthy preliminary description will render intelligible the questions to which I have successfully directed my attention. They are as follows :— (1) The method of construction of the imbricated tube and the elastic conical tips. (2) The reason for the adoption of this particular arrangement. (3) The function of the “ Lippen-organ ” or “‘ Metastomium ” (Gilson). (4) The raison d’étre of the immense number of uncini and of the double form of septal valves. (5) The existence of a cephalic opening, and of anal pores. (6) The method of reproduction. (7) The larval form (Ahéraria). T do not now propose to deal with these questions seriatim, but rather with such points as have not already been dealt with, in my account of the habits of the worm. As already explained, the tube has a definite anterior and posterior extremity, and is usually embedded more or less vertically in the sand, the anterior end being always nearest the surface, and also that to which AND HABITS OF THE AMMOCHARID®. 241 external stony additions are made. During the day the tube, into which the inmate retreats, is frequently entirely buried ; for the Ammocharide, like many other annelids, appear to be largely nocturnal in their habits. It is not usually until towards evening that the animal becomes active, and, after exposing ten or more millimetres of its tube above the surface, ventures to protrude its branchie. In order to appreciate the builder’s difficulties we must realize the fact that the tube of an annelid is necessarily constructed From within; that while it is, so to say, secreted, or built by the worm round its own body, it 1s so arranged that the inmate shall all the time be entirely detached from it, and free to change its position within it at will. If we imagine the head of the worm protruding through the growing conical tip (c.z., Pl. 24. fig. 12), it will then be seen that, in order to produce the imbricated tube, the base of each new fragment added must be placed carefully underneath the fragments which are already attached, and must be there affixed to the membranous tube; in other words, it must be inserted and fixed between the inner sheath and the last row of fragments added to the outer covering. Were the process reversed and additions made at the other end of the tube, as by a tiler covering a roof, there would seem to be but little difficulty. The matter would be almost as simple as the placing of one stone above another; but the problem is as I have stated, and its solution (which absorbed considerable time) took me completely by surprise. The beautifully trans- parent, conical, elastic part of the tube (c.t., Pl. 24. fig. 12), which on retreat of the animal is drawn inwards (the worm thereby frequently pulling together the latest shelly additions and closing the tube with them), is figured as almost fully dis- tended by the water, which is being driven before it by the advancing worm. At first the animal is bidden by the stony sheath, but in another moment the branchial tentacles, con- tracted and twisted together in the form of a living cone, appear beyond it and fill the elastic transparent tip, which I might perhaps compare to an india-rubber teat. Then, cautiously thrusting the tip of a single tentacle through the terminal pore, the animal carefully feels round to make sure that all is safe. Satisfied on this point, a second tentacle follows the first, then several more; finally the whole branchial crown is forced through the minute opening, and the tentacles, spreading themseives out, LINN. JOURN.—ZOOLOGY, VOL. XXVIII. 17 242 MR. A. T. WATSON ON THE STRUCTURE cause the elastic sheath, stretched around them like a sphincter, to slip from their outer surface, and settle as a folded ring about the front edge of the “ thorax.” For a short distance the “thorax” is also visible through, and tightly invested by, the basal part of the membranous cone. The branchial crown (0.2., Pl. 23. figs. 1 & 2) now appears like a lovely red or golden flower, and, basking in the light, may for a time continue quietly to enjoy the water, which it sets in motion with its tentacular cilia. The expanded tentacles are dorsally widely separated by the cephalic crescent (c.l.) into two lateral groups, and just below the outer margin of the crescent a groove, bordered with cilia, is visible. The ventral division is less apparent, but is marked by the outermost tentacles of each group being, like sentinels, usually slightly advanced within the crown; and whilst the swollen tricuspid lobes (¢./. & /./.), with ciliated surfaces, occupy its base (PI). 23. fig. 2), the dorsal lobe gently rises and falls as though in the act of breathing. Butif the call of hunger claims attention, or a desire for work arises, all ischanged. The tentacles become lively, the boughs bend towards each other across the open crown, or perhaps throw themselves violently outwards; opposite twigs incline and meet, whilst the bilobed tips, like miniature fingers, twitch and move as though striving to grasp something. Then, as the water does not supply the need, the worm, in its tube, curves over on one side until the surface of the sand is reached, and that which is desired can be obtained. When the animal is thus seeking sand, or fine particles of shell, these at once adhere slightly to the mucus of the bilobed tips of the tentacles, or are grasped by them and worked into the horseshoe- shaped internal hollow, which, as a ciliated channel, conveys them to the base of the crown. When the sandy particle has in its progress reached the three lobes, one of two things happens: it may either be kept in motion for a short time by the surface- cilia, and then be rejected without apparently having fulfilled any purpose whatever (the two ventral tentacles usually acting as “‘ejectors”), or, conveyed within the grasp of the lobes, it may be carried down, by a kind of peristaltic action, through the opening into the vestibule beneath. Powerful transmitted light is required to observe what follows, and in this way, with the aid of a suitable arrangement of the microscope and a low power, the fragment can be seen to be now revolving with a jerky AND HABITS OF THE AMMOCHARIDA. 243 motion. It is beg manipulated by the “ Lippen-organ” (J.0., Pl. 23. figs. 1 & 3, Pl. 24. fig. 7) which, protracted and retracted at each movement, is apparently licking or rasping it, as would the radula of a mollusk. After a few minutes’ treatment, the fragment may be ejected by the “ Lippen-organ,” which, pro- truding between the lobes, thrusts it upwards, within reach of the ejecting tentacles, which complete the expulsion from the crown. One cannot doubt that, although these worms also swallow fine grains of sand, the process just described is an act of feeding. We may assume that the material last secured, although bearmg food upon its surface, was unsuitable for building purposes; but when, by chance, a tempting flat grain of sand, or preferably a minute fragment of shell, is obtained, the “ Lippen-organ ” assumes a different réle, namely that of a builder; a function the probability of which was suggested by Gilson (27. p. 381).* The selected fragment is then passed down into the vestibule as before, and there subjected to the treatment just described, but for a much longer period. By this process any available nourishment is doubtless at once removed from the surface, and the material is most carefully cleaned. Then, all being ready, the “ Lippen-organ ” (J.0., Pl. 23. fig. 1), having turned the fragment into such a position that the thin edge is uppermost, begins to rise steadily, carrying the fragment in front of it. The two lateral branchial lobes separate themselves widely apart to allow of the passage, and the margin of the crown, between the ventral branchiz, becomes simul- taneously stretched. The upper surface of the “ Lippen-organ,” with its two bilobed ends, is now visible, carrying the shell, which it intends to fix to the outer side of the tube. In another moment, the position of the fragment is so changed that the flat side, instead of the edge, shall rest upon the “ Lippen-organ ”; and this centinues to rise, until its bilobed extremity is made to project between the lateral ventral tentacles, and to touch the exterior base of the elastic conical tip, which, by retraction of the cephalic segment, * In this connection it is curious to note that, at times, the worm appears greatly to prefer building with white materials, and even specimens from Naples (where, to judge by their tubes, only black materials are available) have been noticed to select white when they had the opportunity, and to reject darker ones supplied to them. What means of colour perception such an organism can possess is an interesting question. 17* 244, MR. A. T. WATSON ON THE STRUCTURE combined with a wonderful special arrangement, next to be described, has been brought within reach. I have called atten- tion to the fact (antea, p. 235) that a small portion of the body- wall, represented by a ventral triangular area (v.a., Pl. 25. fig. 20) just above the cesophageal ganglion, is very considerably reduced in thickness. The object of this is now apparent. Simulta- neously with the protrusion of the “ Lippen-organ,” the circular muscles (¢.m., Pl. 28. fig. 1) of the lower part of the ‘“ thoracic” segment contract, the powerful septal valves (¢.v., Pl. 23. fig. 1) are doubtless brought into action, and by means of the imprisoned coelomic fluid a very marked swelling of the anterior part of that segment occurs, especially in the area above referred to, with the result that, at the very edge of the “ thorax,” a projecting pouch (¢.p., Pl. 23. fig. 1) is produced. It will be borne in mind that the tube (m.t.) tightly invests the ‘“ thorax,” as with a skin, and as it consequently covers the pouch, it assumes its form; and frcm this it may be realized that the lobes of the “‘ Lippen-organ ” (Z.0., Pl. 23. fig. 1), on beg protruded, find themselves pretty nearly on a level with the curved upper covering of the pouch. By a muscular action of the upper surface of the “ Lippen- organ,” assisted by occasional little pushes from behind, the fragment (fs., Pl. 23. fig. 1) now commences to travel, sliding along smoothly towards the outer lobes of this, whilst the epidermis on the underside of this end of it is forced upwards and made to take an active part in transferring and affixing the fragment, at a tangent (Pl. 23. fig. 1), to the curved upper covering of the pouch. This done, the ‘‘ Lippen-organ” is retracted into the vestibule, the ‘“‘ thoracic”? muscles are relaxed, the pouch disappears, and the parts resume their ordinary aspect ; but the fragment just added will be found standing vertically, with its base in the required position, and the lower end of its inner side firmly attached to the elastic conical sheath. It is certain that the cement for this purpose is supplied by the ‘‘ Lippen-organ,” which must consequently be regarded as a glandular structure. From my observations of the habits of the animal, I have no doubt that the external thickening of the membranous sheath is also due to the secretion of this organ, and not, as has been supposed, of the thread-glands. The “ Lippen-organ,” which is seen in situ in the longitudinal section of the “ thorax” (/.0., Pl. 24. fig. 7), is beautifully figured by its discoverer, von AND HABITS OF THE AMMOCHARID &. 245 Drasche (12. pl. ii. fig. 3), who describes it (12. p. 6) as being composed of two dorso-ventral lips (facing each other), of which each is bilobed, “in a form similar to that of Ascaris” ; _the dorsal lip being generally rather the larger. In preserved specimens, the two lips are usually found separated from each other by about 180°; but in life they are inclined towards each other at various angles. Where the bases of the two lips touch there is a small deep fold, which, however, is cecal and does not lead into the body- eavity. The organ is retracted by means of a local folding of the centre of its cup-shaped membrane. Vou Drasche adds (12. p. 12) that the organ is a folded portion of the epidermis, consisting of high, wedged-shaped, badly stain- ing, ciliated cells, with small nuclei. As regards preserved specimens, I shall not attempt to improve upon this excellent description, except to question the existence of the ciliated nature of the cells, at all events as regards the upper surface of the organ. I have seen it at work many times, and although I _ have observed ciliary action on the dorsal lobe and other parts, I have failed to detect anything of the kind on its surface. On the contrary, the impression, conveyed by watching the living animal, is that the function of these tall, transparent cells is the secretion of the cement for attaching the outer stony covering and strengthening the membranous tube. In certain operations, which I hereafter describe, the ventral end of the organ is pro- truded between the ventral tentacles, so as to bend far over the edge and manipulate the outside of the membranous tube. On such occasions the organ presents a beautiful semitransparent appearance, with the two protruding terminal lobes, tensely distended as from pressure of the secretion with which they are charged. A transverse section (Pl. 24. fig. 9) of this end of the “ Lippen-organ ” shows that the clear, tall cells, which are so noticeable on the upper face, have disappeared from the uniler, and given place to much shorter nucleated cells and a folded arrangement of the epidermis (/.c.). From the way in which the latter structure is protruded, and takes part in the operations of the animal, I have no doubt that some of the secretion is discharged from its surface, probably near the centre line, below the ventral end of the organ, although a discharge may also take place feom the upper surface, as is clearly shown in the central fold of one of my sections. 246 MR. A. T. WATSON ON THE STRUCTURE The contents of the first two pairs of thread-glands, like those of the long cells (h.c., Pl. 24. fig. 9) of the ‘“‘ Lippen-organ,”’ strongly resist stains, and it seems very probable that their secretions are identical. The fact that the secretion of the “abdominal” thread-glands stains more readily, may possibly indicate some difference in composition. The membranous conical tip (c.¢., Pl. 24. fig. 12) is a character- istic and essential feature of the tube, and in the event of its accidental removal it is at once re-formed by the worm. By taking advantage of this fact, I have been enabled to make two very interesting observations on (1) the method by which the membranous tube is formed, and (2) the means by which the animal can easily, at will, cut off any desired portion. The latter is a habit which I had noticed to exist, from frequently finding fragments in the aquarium cut off evenly as though with a knife; showing that the worm possesses some special means of accomplishing the apparently difficult task of cutting through the tough membranous sheath. In order to make sure of these observations, a healthy vigorous worm was selected, carefully removed from the aquarium, and placed in sea-water in a glass dish. The tube, as usual, was considerably longer than the animal, and to wduce work I at once cut off enough from each end to reduce the tube to about the length of the tenant, taking care, of course, not to injure the worm. ‘This, in its shortened tube, was then transferred to a shallow glass dish containing fresh sea-water, and watched under the microscope. Very soon the animal became aware of the unusual “ draughtiness” of its dwelling, and proceeded to investigate the position by cautiously protruding ite tentacles and examining the edge of the tube. Having ascertamed the extent of the damage, and by re- peated search satisfied itself that no enemies were near, it retired just within the shelter of the tube, so that the tips only of the branchial tentacles were, now and then, visible at the fractured end. By transmitted light, it was just possible to see between the sand-grains, and through the membranous tube, that something was going on inside: the branchial tentacles were more or less completely twisted together ; the worm was revolving slowly within its tube, with the first bundle of sete in motion, and it occasionally advanced slightly and then retreated. This process continued for varying periods (apparently dependent upon the intensity of the light), which sometimes extended over AND HABITS OF THE AMMOCHARID. 247 an hour or more; and then a swelling of the tube suddenly commenced, at a point about a millimetre distant from the fractured end. This internal swelling gradually separated the sand-grains far apart, and, through the interstices, it could be seen that the branchial tentacles were tightly screwed together to form a solid living cone, and that the fore part of the “thoracic” segment was so fully distended by the coelomic fluid, that all trace of a division between it and the base of the branchial crown had disappeared. The worm kept up a constant revolving motion, the bristle-bundles being actively at work, whilst, between the ventral tentacles, both ends of the “ Lippen-organ ” (J.0., Pl. 23. figs. 1, 3, & 7) simultaneously rasped the membranous tube, their ventral lobes striking downwards, and their dorsal lobe vigorously pulling in the opposite direction. In a few minutes the tube was in this way burst and cut through, the severed part or ring being gradually thrown otf by the tentacles, which, quietly unfolding themselves one by one, with- drew from the interior and passed the ring over their tips. Freed from the stony encumbrance, the branchial tentacles were seen to be naked and fully exposed to the water; but careful illumination showed that the “thoracic” segment was once more invested with a new membranous, tightly-fitting covering, so beautifully transparent as to be almost invisible, and this I found to be the foundation layer for the new conical tip. It had been formed while the animal was sheltering and revolving in the old tube; the necessary secretion being doubtless supplied by the first two pairs of thread-glands, and passed forward, by and over the tentacles, which, as I have stated, were screwed together so as to form a conical mould. On one occasion, a tentacle was seen to be extended beyond the rest, as though to form a pivot to keep the minute aperture at the tip of the tube open. Immediately the tentacles were freed from the stony ring, the animal proceeded to complete the formation of the membranous tip. It advanced so far in its tube that the whole of the branchial crown was exposed, and the new tip, as a trans- parent skin, was fully stretched over the “ thoracic” segment. Then, for the space of five minutes, the first two bundles otf bristles worked vigorously, like brushes, spreading the cement which, with accompanying violent contortions of the body-wall, was poured out by the thread-glands, the animal meanwhile keeping up a constant, steady revolution within its tube. Then 248 MR, A. T. WATSON ON THE STRUCTURE the edge of the new tip (and sometimes the interior) was manipulated by the underside of the ventral end of the “ Lippen- organ,”’ and the animal once more retired into the stony tube. It was found that a beautifully perfect, though apparently delicate tip had been produced. After a short interval the other end of the tube was similarly renewed. The part taken by the sete in this operation is, I think, noteworthy. It has hitherto been overlooked, and as I have witnessed it in the tube-forming operations of Panthalis (24) and Nerine, it is probably true of other annelids. Although it was impossible, owing to the stony covering of the tube, to see what was taking place low down, there can be no doubt that the action, as regards the “abdominal” thread-glands, would be a repetition of what I have described. Gilson’s interesting ex- periment shows that a marked exhaustion of the contents of the thread-glands results from the repairing and extending of a fractured tube. As already stated, the lengthening of the tube is due to constant additions, internally, of new membranous tips, the old layer always being forced outwards ; consequently, the description of the formation of the membranous tip is really a description of the formation of the outer layer of the com- pleted tube. As I have shown, it is due to the secretion of the thread-glands of the “thoracic” segment, supplemented by those of the “ Lippen-organ.” Its outer coating, during the life of the animal, is quite unaffected by the sea-water, and it is obviously impossible for it to be renewed. It is an interesting fact, however, that the tube as a whole, though exceedingly tough during the life of the animal, speedily decomposes after its death or expulsion ; and it would seem probable that during life a constant renewal of the internal layers is taking place, and that these (which are doubtless due to the secretions of the larger glands) differ, in composition, from those of the outer sheath— a probability which is also indicated by the difference in the reaction to stains already alluded to. The tube is essential to the existence of the animal, which, when in great straits, bending its body into the form of the letter S, will make a tube only one third of its own length suffice temporarily for its protection; but if, by chance, the worm should be expelled from the tube, death will follow after a short time, as the animal is then, like the Serpualide and some other worms (19. p. 75), evidently quite incapable of replacing it. AND HABITS OF THE AMMOCHARID#. 249 The Ammocharide are generally regarded as “ sedentary worms,” but, from what has been said in the earlier part of this paper, it will have been gathered that their habits are not entirely of that character. Members of this family are, in fact, capable of burrowing in their tubes through the sand; and although this is usually done without travelling far afield, still they are quite capable of working their way through a consider- able thickness to the surface, or of leaving the sand entirely, and reburying themselves in new sites. Experiments made with the object of learning something of these habits, have elicited interesting facts. When the worm is first captured and exposed, say by digging and rapidly washing away the sand by means of a sieve, it is generally found that the tube is of considerable length, firm and swollen at the part where the worm at the time happens to be, but diminishing to the diameter of thin twine in the unoccupied portion. In English specimens, the large propor- tion of fragments of shell in the outer sheath imparts a striking whiteness to the tube. A ball of sand, loosely held together by mucus, is commonly found covering the short length of other- wise naked tube at the posterior extremity. This is, doubtless, formed by the branchial tentacles. From the nature of the building-operations, it is obviously impossible for the “ Lippen- organ” to properly affix sand-grains underground, though there is ample evidence, in the form of imprints (d, Pl. 24. fig. 13), that this organ strengthens the membranous tube by external addi- tions of its secretion. If such a tube be left in sea-water, without sand, it will be found, after an interval of an hour or two, that it has entirely changed its appearance, and has become spindle-shaped, and only one half or even one-third of its original length. The worm, having drawn over itself the portion which was previously unoccupied, has thus caused the fragments of shell or sand to come more closely together and stand at right-angles to the length of the tube. If the specimen be now reburied beneath a few inches of sand, it will be found, after a short interval (varying of course with the thickness of the layer), that it has burrowed upwards, so as to bring the anterior end of its tube to the surface, a position which is apparently needful for its welfare ; and when it is remembered that, in doing this, the free edges of the external fragments have been so placed as to meet with the greatest resistance from the surround- ing sand, it will be realized that the operation involves great 250 MR. A. T. WATSON ON THE STRUCTURE muscular exertion. The necessity for the unusually ample equipment of uncini in the “abdominal” region, in order to carry the tube with the body, is thus explained. The rate of burrowing is not, of course, rapid when a con- siderable thickness of sand has to be penetrated, ten hours being required, under favourable circumstances, to bore through 150 mm., though 20 mm. would be run through in a few minutes. As the tube is very similar in colour to the sand in which the animal dwells, observation upon its displacements was at first im- possible ; but, by the admixture of a sufficiently large proportion of white (Calais) sand, I ultimately ascertained the burrowing act to consist of a combination of an undulatory with a screwing motion. When engaged in this operation, the animal occupies the anterior end of the tube, the conical tip of which is filled by the densely-packed tentacles; while the first row of shelly frag- ments, spread out, possibly forms a cutting edge. By twisting the anterior portion of the body and tube round in the sand, the animal practically converts them into a short hollow screw, about 10 to 15 mm. in diameter, which, working somewhat spas- modically at intervals of a minute or two, to afford time for rest and stretching the tube, gradually makes its way to the surface. On reaching this, the action of the worm confirms the observa- tion in a very curious manner, as, in order to get the desired length of tube exposed, it bends hard over, and pressing on the surface of the sand, describes another revolution with its anterior end, which results in a further stretching of the tube, accompanied by the formation of a saucer-like hollow in the sand. It will be noticed that I speak of stretching the tube; and in explanation I may state that, as a rule, the animal does not carry the whole tube bodily with it to the surface, but attains its end by causing it to stretch the required distance. In some of my experiments, a tube apparently only 65 mm. long stretched to a length of 200 mm., but when again uncovered contracted to its original dimensions. Occasionally, however, the animal, for the purpose of changing its habitat, quits the sand, and in this case it brings the whole tube with it. This is an act which probably sometimes costs the animal its life, for McIntosh states (10. p. 103) that Owenia filiformis, with its gravelly tube, is a favourite food of haddocks, cod, and flounders. In order to re-bury itself, the worm reverses its position, and, protruding the branchial ten- tacles through the opening at the posterior end of the tube, AND HABITS OF THE AMMOCHARIDA. 251 proceeds to force them into the sand. In this operation it is assisted by the “ Lippen-organ,” the ventral end of which, acting very much like the foot of a mollusk, strikes over the edge of the membranous tube, and so, with its underside, digs a burrow in the sand, enabling the animal gradually to screw itself down- wards. The “thoracic” setz also appear to take part in this work. This operation is much slower and more laborious than the upward burrowing, and the animal not unfrequently cuts off, and leaves on the surface, a portion of the tube. The tube is thus re-embedded in its natural position; and it is very un- usual indeed to find a tube in which a change in the direction of building has taken place, though I am bound to admit this occasionally occurs. With reference to the method of reproduction, Claparéde (6), von Drasche (12. p. 20), and Gilson (30. p. 99) have respec- tively described the formation of the sexual elements as taking place on the walls of the ventral blood-vessel, in the tissues which clothe the longitudinal muscles, mesenteries, and blood- vessels, and on the inner face of the musculo-glandular tissue, of which the last-named author considers the body-wall of Owenia to be composed. All agree that the elements are shed into the body-cavity, that they float about in the ccelomic fluid, and are frequently found, apparently straying, in the “‘ thoracic” and buccal segments. The means of their escape from the body-cavity is unknown, though Gilson (27. p. 370) confidently suggests that the “ epithelial canal” in the second “‘ abdominal ” segment is the natural outlet. Starting from this standpoint, I have directed my attention, firstly, to observing the movements of the ova when they appear in the cephalic segment; and, secondly, to ascertain in the living animal the mode of emission of the sexual elements, and the early life-history. By the first line of research I failed to secure the information I desired, namely, as to the fate of the ova in the cephalic seg- ment, though it resulted in the discovery of the interesting, and probably unique structure for a Polychete, the opening of the cephalic lobe (c.p., Pl. 24. fig. 8). Although this structure is exceptional in the Polycheta, a similar prostomial pore is met with in some Oligocheta (Enchytreide), a fact to which Mr. E. S. Goodrich has kindly called my attention. Upon this point Beddard makes the useful contribution that ‘“ the dorsal pores are missing without exception in those Oligocheta which live in 252 MR. A. T. WATSON ON THE STRUCTURE water ; but these latter worms have a pore upon the head, which appears to be wanting in the earthworms.” He adds that “ Dr. Michaelsen has thought that the head-pore serves to relieve the brain-pressure—to act in fact as a kind of safety-valve for the liberation of superfluous fluid” (26. p. 349). From what we know of the habits of the Ammocharide, it is clear that if ever such provision be needed, it would be in their case. But to continue my notes:—For several days a large Italian specimen in the desired condition was under observation, and it was noticed that the number of ova in the segment varied, increasing and diminishing, and extending even into the bases of the branchiz, and ultimately almost entirely disappeared, without any apparent cause. Occasionally a considerable number were congregated near the centre of the groove in the cephalic lobe, and it was expected that a discharge would take place at that point, but nothing of the kind occurred. The observations and sections, therefore, simply show that, at the point indicated, there is an opening into the ccelom, guarded by a network of muscular fibres (some of which may be free, protrusible processes), and the ova may be either aerated at that point, or make their escape there under compression or when disintegrated. As to which is the more likely, I will not venture to say, though the ova which reach this spot are, apparently, so few that it cannot be supposed that all in turn pass into this segment and return to the “abdominal” segment, though this is possible. The emission of ova from the “ epithelial canal’ was watched for in vain; in fact, only once did a worm emerge sufficiently for its tube to expose the opening of these ducts, though, when re- burying the tubes, it sometimes happened that a sufficient length lay naked in the sand. Fortunately, the solution came at last. On the evening of May 23rd, 1898, a large living specimen from Naples, which had settled in a small aquarium, protruded from the anterior end of its tube about 20 mm. of the posterior portion of its body, and for the space of about ten minutes discharged spermatozoa in two white, smoke-like streams, which issued from tubular open- ings on the ventral wall (right and left) of the anal extremity. During the discharge the animal waved its tail-end vigorously from side to side, thus assisting dispersion. The cloud of spermatozoa at first settled down, as a white layer, on the surface of the sand, but in the course of about half an hour AND HABITS OF THE AMMOCHARID®. 253 nearly all had disappeared. The spermatozoa (PI. 25. fig. 19) were found to be of the form figured by von Drasche (12. pl. ii. fig. 14). Three other specimens in the same aquarium, with branchie fully displayed, were apparently quite unaffected by what was taking place ; and this observation was one evening in the following July supplemented and confirmed by a colony of English specimens which had, a few months before, settled in another aquarium. About 7.30 p.m. several males discharged spermatozoa, the swarms of which were still close to their respective tubes, and readily recognized by the cloudy masses speedily diminishing. Shortly afterwards, several followed exactly the same course, and about the same time three females discharged ova. The latter procedure differs somewhat from the former, in that little more than the anal extremity is exposed from the anterior end of the tube. A quantity of fecal matter was first expelled, and then the ova were, as before, discharged through the two pores at the anal extremity, but much more slowly than the spermatozoa and enveloped in albumen, sc that while issuing the discharge had a thread-like appearance. When deposited, there was formed a small jelly- like mass near the tube; and the contrast between the energetic movements of the male and the quiet discharge of the female was very marked. In one case a male, whilst discharging sperma- tozoa, actually stirred-up with his tail a mass of ova which had been deposited by a neighbour. With very few exceptions, all the worms discharged their genital products within an hour or two, and as to the cause of this simultaneous action it may be that the swarm of spermatozoa from the first animal which discharges acts as a stimulus. Hornell has observed (22. p. 18) similar synchronous emission in the case of Branchiomma vesiculosa. A few days after spawning, nearly all the worms cut off about 10 mm. of the anterior end of their respective tubes. The anal pores (a.p., Pl. 25. fig. 20) are very striking and interesting structures, their lumina being lined with very large, regular, radiating, cylindrical, transparent cells (a.p., Pl. 24. figs. 10 & 11), which refuse stain when treated with hematoxylin and eosin. Possibly their function may be glandular and connected with the supply of albuminous matter, since they appear to be most fully developed in, if not entirely confined to, the females. I have been unable definitely to determine at what stage in the 954 MB. A. T. WATSON ON THE STRUCTURE animal’s development these cells first appear. I have failed to detect them in the youngest specimens (only 10 mm. long), which were obtained by the kindness of Dr. Herdman from Naples; but I found them fairly well developed in other specimens (only 16 mm. long) which, though probably only a few weeks older, were full of ova. In order to reach these pores, the genital products pass through the sphincter septal valves (s.v., Pl. 23. fig. 5), which can easily be traced in the sections right up to the lumina of the pores. On one occasion, whilst examining a preserved Ammochares from Naples (which in 90 per cent. alcohol had uot become hardened), I accidentally caused these pores to open and work by osmotic action, the specimen having been transferred to spirit of a different strength. The ova, deposited on the 17th July 1898, were examined every 24 hours and sketched from life from the 19th to the 21st inclusive (Pl. 25. figs. 14-18). On the second day, the eggs were in the morula stage, and a few specimens, which had been transferred from the aquarium to a small glass dish, had, on the 20th July, attained the two different forms represented by figs. 15 & 16, Pl. 25. In view of the subsequent development, it is interesting to note that, whilst watching the specimen fig. 16, I observed, now and then, a thread or two, with a cell at the extremity of each, shot forth from the point (m/), the threads being immediately retracted. Fig. 15 is also deserving of special attention, since it corresponds, most closely and suggestively, with a stage in the development of the Hydrozoa, as shown in Fewkes’ figure of Agalma (11. pl. ii. fig. 12) of somewhat similar age. It was found that development had been retarded by removal, as speci- mens taken the same evening from the aquarium had reached the stage with characteristic sete, fig. 17, which will be recog- nized as Mitraria. On the following evening, the specimens in the glass dish were found dead, and those in the aquarium, although vigorous, had apparently made little or no further progress in development. The larva attached itself by its sete to the bottom, and, using them as a pivot, rotated in a top-like fashion in the watch-glass, whilst four others, affixing themselves by the same means toa speck of floating organic matter, in the form of a cross, caused it to revolve like a miniature wheel. This rotation was doubtless due to cilia at the apex of the bell, as the motion of those at the margin was scarcely perceptible. AND HABITS OF THE AMMOCHARID®. 25 Or When making my sketches, I had neither seen Metschnikoft’s nor any other figures of Mitraria, and I had constantly to wait until the animal returned to its original position: to this fact is due the omission in fig. 18 of the two crescent-shaped thickenings of the epidermis figured by Metschnikoff (7. pl. xviii. fig. 8), as although they were seen distinctly when drawing other parts, no subsequent view enabled me to give their exact form and position. Upon comparing the figures (17 & 18) with those of Metschni- koff (7. pl. xvii. figs. 5 & 8), the chief difference relates to the outline of the ciliated margin of the hydrophyllium ; that of my specimen was perfectly plain, and, when seen from below, some- what oval in outline, whilst Metschnikoff’s figures show the margin to be deeply lobed. ‘hat Metschnikoft’s Mitraria is the larval form of a species of Ammochares there can, I think, be little doubt. Its agreement with that of the English form, Owenia, is exceedingly close. The lingual organ of the young worm, figured by Metsch- nikoff (7. pl. xviii. fig. 12), agrees closely, both in form and position, with that of Ammochares, whilst the position of the ‘“ schleimabsondernde Organe”’ corresponds with that of part of a “thread-gland.” The form of the uncinus (7. pl. xviii. fig. 12 4) is unmistakably that of the family, though Metsclinikoff’s figure of the worm (7. pl. xviii. fig. 12) shows only a single row of uncini to each of the “abdominal” segments. The anterior smoothly truncated end of the worm, shown in his figure, corresponds with what might be expected from Cunningham and Ramage’s observation (15. p. 656) that, “‘in small specimens the mouth is surrounded by a funnel-shaped lip, quite entire, except for a ventral notch,” and that “it is only in the larger speci- mens that the branched processes are seen.” Metschnikoft’s statements (7. p. 241) that while each of the first three segments of his young worm bore on either side a bundle of fine tapering sete only, all the other segments were equipped with both sete and transverse rows of uncini; that the first eleven sezments were extraordinarily long, and distinguishable only by the arrangement of the bristles, since the usual segmental cross- furrows were absent ; and that the last body-segment ended with two roundish little lobes, between which was the anal opening (points all in striking agreement with the structure of the Ammocharide), convey, in combination with his figure, what 256 MR. A. T. WATSON ON THE STRUCTURE apparently is an accurate description of a young Ammochares. It will doubtless be admitted that these facts completely refute the suggestion of Hacker (31. p. 15), based mainly upon the form of the bristles, that Itraria should be regarded as related to the Hermellide. Further study of the early stages of development will certainly yield an amplereward. The imperfections of the present account are largely due to the fact that I was striving to rear the larval forms, and consequently hesitated to kill any of the specimens. My observations were made in the summer of 1898, and, un- fortunately, the efforts which I made in the two following years proved abortive. . The Ammocharide have a very extended range, viz., the Arctic Ocean (Spitsbergen), North Sea, Mediterranean, Adriatic, Davis Straits, the Antilles, Brazilian seas, Japanese seas, and the Philippines. De Saint-Joseph (29. p. 404) has called attention to their connection with the Maldanide on the one hand, and the Ser- pulide on the other; whilst Gilson has emphasized many of the interesting points in their structure, and argues (30. p. 90) that, the absence of muscular rings from the greater part of the oll and the existence of the two longitudinal mesenteries are charac- teristics of inferiority, and recall the Archiannelids, a contention which, adopting Meyer’s view of “ Annelid Descent” (18 6. p- 1147), appears to be supported by the discovery of the anal outlets for the genital products, and also by the exceptional larval forms. List oF WoRKS CITED, arranged in order of appearance. 1. Dette Cutase, S8.—Descrizione e Notomia degli Animali invertebrati della Sicilia Citeriore, Tav. 175, tom. i. p. 35 (1841). 2. Grupe, E.—Beschreibung neuer oder weniger bekannter Anneliden. Arch. f. Naturgeschichte, 12. Jahrg., 1846. 3. Konirker, A. von.—Kiirzer Bericht iiber emige im Herbst 1864, an der Westkiiste von Schottland angestellte, verg.- anat. Untersuchungen. Wurzb. Naturw. Zeits. Bd. v., 1864. 10. te 18. HG): AND HABITS OF THE AMMOCHARID®. Dail . Marmerren, A. J.— Annulata Polychexta Speisbergie, Gronlandiz, Islandie et Scandinavie, hactenus cognita. (Kfvers. af K. Vet-Akad. Forh. 1867, No. 4. - McInrosH, W. C.—On the Structure of the British Nemerteans and some new British Annelids. Trans. R. Soc. Edinb. vol. xxv., 1868. . CrararkprE, EH.—Les Annélides chétopodes du Golfe de Naples. Mém. Soc. Phys. et Hist. Nat. Geneve, t. xx., avec suppl. (1870). . Metscunixorr, E.— Ueber die Metamorphose einiger Seethiere (Mitraria). Zeit. £. wiss. Zool. vol. xxi., 1871. . Craparrpr, E.—Recherches sur les Annélides sédentaires. Mem. Soc. Phys. et Hist. Nat. Geneve, t. xxii., 1873. . Marton, A. F.—Sur les Annélides de Marseille. Revue des Sc. Nat. Montpellier, t. iv., 1875. McIntosuo, W. C.— Marine Invertebrates and Fishes of St. Andrews. 1875. Fewxes, J. W.—On the. Development of Agalma. Bull. Mus. Comp. Zool. Harvard College, vol. xi. (1883-85), pl. iii. fig. 12. . Drascue, R. von.—Beitrige zur feimeren Anatomie der Polychaeten. Zweites Heft: Anatomie von Owenza /fili- formis. Wien, C. Gerolds Sohn, 1885. . McInrosn, W. C.—Report on the Annelida. ‘ Challenger’ Exp., vol. xu. pp. 410-418 (1885). . Erste, H.—Die Capitelliden. Fauna und Flora des Golfes. von Neapel, xvi., 1887, p. 336. . CunnincHam & RamaGge.—Polych. sedentaria of the Firth of Forth. Trans. R. Soc. Edinb. vol. xxxin., 1888. . Gruson, G.—On Secreting Cells. Report of the Brit. Assoc. 1890 (Cardiff). . Warson, A. T.—The Tube-building Habits of Zerebelia littoralis. Jour. Roy. Microsc. Soc. 1890. Mryrr, E.—(a) Die Abstammung der Anneliden. Biol. Centralblatt, Band x., 1890. (6) Translated in the ‘American Naturalist,’ vol. xxiv., 1890. Souter, A. Etudes sur quelques points de |’Anatomie des Annélides tubicoles de la région de Cette (Montpellier). 1891. LINN. JOUBN.—ZOOLOGY, VOL. XXVIII. 18 258 20. al. 22. 23. 24. 25. 26. 27. 28. 29. 30. dl. 32. MR. A. T. WATSON ON THE STRUCTURE Gitson, G.— Les Glandes filieres de l’Owenta fusiformis. La Cellule, t. x., fase. 2, 1893. Lo Branco, 8.—Gli Annel. tubic. del Golfo di Napoli. Atti dell’ Accad. delle Scienze di Napoli, 2 ser. t. v., 1893. Hornet, J.—Method of Ova-dispersion in Tube-building Worms. Jour. Marine Zool. & Microscopy, vol. i. p. 18 (1893). Gitson, G.—On the Septal Organs of Owenia fusiformis. Rep. Brit. Assoc., Ipswich, 1895. Warson, A. T.—Observations on the Tube-forming Habits of Panthalis @rstedi. Port Erin Biol. Station. 1895. Benuam, W. B.—“ Polycheta.” The Cambridge Natural History, 1896, p. 325. Bepparp, F. E.— Oligocheta.” The Cambridge Natural History, 1896. Gitson, G.—Les Valves septales de ?Owenia. La Cellule, t. xu., fase. 2, 1897. Fauver, P.— Recherches sur les Ampharétiens. Bull. Scientif. du Nord de la France et de la Belgique, tom. 0.005 USIP Sarnt-JosepH, de.— Annélides Polychetes des Cétes de France. Ann. Sci. Nat. sér. 8, tom. v. (Zool.), 1897, pp. 397-405. Giison, G.—Cellules musculo-glandulaires paroi du Corps et fonction exerétoire de l’Owenia. La Cellule, t. xiy., fase. 1, 1898. Hacker, V.—Die pelagischen Polychaeten und Achaeten- larven der Plankton-Expedition. Ergebnisse der Plankton- Expedition der Humboldt-Stiftung, Bd. ii. H.d. p. 15 (1898). OeneFF, J.—Prof. Gilson’s “Cellules musculo-glandu- Jaires.”’ Biol. Centralblatt, xix.(1899). Jour. Roy. Mier. Soc. 1899, Part 2 (April), p. 159. AND HABITS OF THE AMMOCHARIDA. 259 EXPLANATION OF THE PLATES. Reference Letters. @., anterior extremity. @.p., anal pores. 0., brain. 6.¢., branchial tentacles. b.w., body-walls. ¢., collar. ch., bristle-bearing process. c.l., cephalic lobe or crescent. ¢.m., circular muscles. ¢.p., cephalic pore. .#., conical tip of tube. d.c., intestine. ¢., eye-spot. @., esophagus. .c., epithelial canal. f.s., fragment of stone or shell. fv., flap-valve. h.b., epidermal band. h.c., high cells of Lippen-organ. 7, under-lip. J.c., shorter cells ot Lippen-organ. 1./., lateral lobe. ¢.m., longitudinal muscles. /.0., Lippen- organ. J.r., upper-lip. m., mouth. m.t., membranous tube. p., posterior extremity. s., sete. s.¢., stony covering of tube. sp., septum. s.v., sphincter valve. 7, torus. 7¢g., thread-gland. ¢.p., thoracic pouch. t.v., “thoracic” septal valve. w., uncini. v.a., thin ventral “ thoracic ” area. Prats 23. Fig. 1. Diagrammatic section of Ammochares, illustrating method of forming the imbricated tube. 2. Ammochares filiformis. Cephalic segment viewed from above, showing the internal lobes (¢./. & 7.) as they would appear during life. This is a restoration of fig. 3. 3. Ammochares filiformis. Cephalic segment viewed from above (von Drasche), showing the ‘“ Lippen-organ ” (/.0.) and mouth (m.), which are frequently visible after death owing to contraction of the three lobes. 4. Uncini of Ammochares :— a’. Ammochares assimilis, Sars (4. Pl. xii.). be 5 ottonis, Grube (5. Pl. xv.) che 5p Susiformis, D. Chiaje (29. Pl. xxii.). d’. Uncinus of Mitraria, Metschnikoff (7. Pl. xviii.). 5. Diagrammatic vertical section of a posterior segment of Ammochares, showing the action of the septal valves (fv. & s.v.). 6. Dorsal view of second “‘ abdominal” segment of Ammochares, showing the zigzag epithelial canal (¢.c.) in the lateral bows ; also (in dotted outline) the form and position of the thread-glands. PuatE 24. Fig. 7. Vertical longitudinal section of anterior extremity through centre line of Owenia filiformis (English), showing, ventrally the ‘‘ Lippen- organ” (/.0.) in sitw when at rest; and dorsally the cephalic lobe, with indication of cephalic (prostomial) pore (¢.p.). X45. (From a thick section.) 10. Fig. 14. 11535 16. ie 18. 19. 20. ON THE STRUCTURE OF THE AMMOCHARIDE. Vertical longitudinal section through centre of cephalic lobe, showing the brain (.) and cephalic (prostomial) pore (¢.p.) with its trans- verse muscular fibres. Ammochares filiformis (Naples). x 100. . Transverse section of ventral end of ‘‘ Lippen-organ,” showing the tall to} ? o 5 clear secreting-cells (%.c.) and the underlying folded structure (/.c.). Ammochares filifurmis (Naples). x 100. Horizontal section through the anal lobes and pores (a.p.) of Ammo- chares filiformis (Naples). A deep section, showing the internal radiate arrangement of cells lining the pores. x 100. . Horizontal section, showing glandular structure of cells and lumina of anal pores (@.p.). Ammochares filiformis (Naples), 2. x 100. Anterior or exposed end of tube, showing the clear elastic tip of inner sheath, and the imbricated arrangement of the external stony covering (s.c.). Owenia filiformis (English). - Posterior or buried membranous end of tube of Owenia filiformis (English), showing its telescopic construction and consequent external stepped appearance; also the strengthening of the wider part by external disc-like additions (d.). x 22. PuarEe 25. Figs. 14, 15 & 16.—Stages in the development of the larva of Owenia filiformis. The embryo 48 hours after discharge of ova. oa) 29 (a 99 ” me) 39 fa 99 ? Larva of Owenta. Lateral view 96 hours after discharge of ova = Mitraria. The same, viewed from beneath. A ripe spermatozoon of Owenia. Ammochares filiformis (Naples). Ventral view. x 2. (From a preserved specimen.) aiVevsome Imm. Soc. Journ. Zoor. Vou XXVIWU Pr. 23. Ee de LUC) WO cre TW. del 12) Me Parker lith. Parker & West imp. STRUCTURE OF THE AMMOCHARIDA.. .poc. Journ. Zoor. Vou. XXVIII Pr. MP Parker ith. Parker & West imyp.- STRUCTURE OF THE AMMOCHARIDA. “Watson. Limn. Soc. Journ. Zoor. Vou XXVIIL Pr.25. 1b A.T.W del. MP Parker lith. Parker & West mp. AMMOCHARES & DEVELOPMENT OF OWENIA. ON TOOTH-GENESIS IN THE CAVIID®. 261 -Toors-Genests IN THE Cavitpm. By H. W. Marzrr Try, B.A., M.D., F.LS., F.Z.S., Lecturer on Biology and Com- parative Anatomy, Charing Cross Hospital Medical School; Lecturer on Zoology, Bedford College, London.* [Read 17th January, 1901.] (PLATE 26.) THoveH much has been already written concerning the dentition of the Rodentia, no paper has yet appeared, so far as I am aware, dealing with the tooth-genesis of that Order. Previous research has been mainly, if not entirely, in the direction of attempts to discover the germs of teeth no longer functional in the adult, more particularly in the region of the diastema and premaxilla, and thereby to endeavour to homologize the existing incisors of the Rodents with those of other mammals. The morphological value of an enquiry into the tooth-genesis is due to the fact that several views have been advanced from time to time to account for the origin of the complex crown of the mammalian molars from the Haplodont type of tooth. In 1896, in a paper dealing with the tooth-genesis in the Canide [22), 1 adopted the suggestion of Forsyth Major that there was the possibility that the Rodentia may have been derived from the Multituberculata, even though the teeth of the Dogs could not be so derived. I was led to do this mainly for two reasons :—(1) the similarity between the dentitions of the Multi- tuberculata and the Rodentia; and (2) the valve of the Paleonto- logical evidence collected by Dr. Forsyth Major and published by him in numerous papers. It was therefore with the object of testing the validity of this suggestion from the embryological standpoint, that I commenced this research, the results of which are here set forth. The selection of Cavia as an animal upon which to work was chiefly due to the fact that my friend Mr. Martin F. Woodward had very kindly given me a series of foetal Guinea-pigs. Upon this material ready to hand, I commenced the investigation. I have also examined specimens of feetal rats, mice, and rabbits (with which I hope to deal more particularly in the future). These, though showing differences in detail, nevertheless appear to me * From the Morphological Laboratory, Cambridge. LINN. JOURN. —ZOOLOGY, VOL. XXVIII. Da 262 DR. H. W. MARETT TIMS ON to confirm the conclusions upon the broader issues with which I am here mainly concerned. If my identification of the individual teeth in Cavia be correct, it then follows that the identification usually accepted of the teeth of the Rodents in general may in many instances be incorrect. This being sc, 1 have deemed it advisable to limit myself, in the present instance, to the teeth of the Caviide, and not to deal with the Order as a whole as was originally my intention. The method I have followed has been that of cutting and examining serial transverse sections of the jaws, after staining in bulk. In some cases the stain used has been borax-carmine; in others Tomes’s ferric-perchloride and tannic-acid method was adopted. The jaws were decalcified in a 2 per cent. solution of each of the following, viz.:—hydrochloric acid, nitric acid, and ferric perchloride. The object in using a combination of the two acids being to neutralize the shrinkage of the tissues caused by the former by the swelling caused by the latter. After decalci- fication, the iron in the tissues was reduced by a 38 per cent. solution of tannic acid. Wax models of the younger stages of the teeth have been made, while in the older the teeth were examined in situ by clarifying the jaws in oil of cloves. I shall commence with a detailed description of each stage, comparing my results in each case with those of other writers, and will leave a general consideration of the problems involved to be dealt with in the latter part of the paper. CAVIA COBAYA. Stage 1. Circumferential head-length... .. 15 em. body-length ... 4cm. Fig. 1. ann A pe Reconstructed diagram, showing the number and position of teeth present in Stage 1. pi., permanent incisor; dm., deciduous molar ; m!, first molar. In this, the youngest stage examined, rudiments of three teeth were to be found in the upper jaw, none, however, showing any traces of calcification. The three teeth present are the incisor, TOOTH-GENESIS IN THE CAVIIDA. 263 the deciduous molar, and the molar immediately behind it, the two latter being usually interpreted as dpm.1 and m.1, but which, as will be seen subsequently, I believe must be regarded in a somewhat different manner. The enamel-germ of the upper incisor has assumed the characteristic bell-shape. There is a decided labial downgrowth of the dental lamina (PI. 26. fig. 5), which is, I believe, to be regarded as the vestigial remains of the deciduous incisor. That it cannot be looked upon as affording evidence of a pre-milk dentition is clear from an examination of the later stages, the bell-shaped germ of this stage growing continuously to form the permanent incisor of the adult. This labial downgrowth is also extremely well marked in the case of the lower incisors. I was not able to detect any trace of a tooth anteriorly or posteriorly to this in the incisor region in the upper jaw. Throughout the several succeeding sections the dental lamina can be traced definitely running through the diastema, but without any indications of enamel-germs. It then dips deeply into the substance of the jaws both upper and lower, and gives rise to a well-marked enamel-germ. This germ has slight traces of both labial and lingual downgrowths in the upper jaw; while in the lower jaw the latter only is present, but more dis- tinct than is the case with the corresponding process in the upper jaw. From a consideration of the reconstructed diagrams of this and the subsequent stages, I think it will be evident that this is the germ of the deciduous tooth. From the wax model it is seen to consist of a cone surrounded by a cingulum. The latter structure projects to form a very definite upwardly- directed and somewhat pointed process arising from the inner side of the main cone. There is a corresponding, though less pronounced projection on the outer side. Some little distance behind this tooth the dental lamina again becomes distinctly enlarged at its deeper extremity, forming a flask-shaped mass, as yet not invaginated by any appearance of the dental papilla. No further trace of any germ is visible at this stage. With regard to the labial downgrowth of the dental lamina in connection with the functional incisor, which I have represented (Pl. 26. fig. 5), it is interesting to compare it with thec ondition found in other Rodents. Huxley was the first [12] to note the existence of minute milk predecessors to the large functional incisors in the rabbit. This discovery has been verified by Pouchet & Chabry [14], Freund [10], and Woodward [27]. “19% 264. DR. H. W. MARETT TIMS ON Writing in ‘ Nature’ [11], referring to these vestigial teeth, Huxley pointed out that “it would be interesting to examine foetal guinea-pigs in relation to this point; at present they are known to possess only the hindermost deciduous molars, so far agreeing with the marsupials.” Freund [10] has also discovered the existence of vestigial milk-incisors in the squirrel in both jaws. Though he inclines to the belief that these vestiges are to be interpreted in this manner, yet he implies the existence of some doubt. From his description of the connection between the neck of the permanent teeth and that of the vestigial teeth, and from the fact that the rudimentary enamel-germ and the large incisor are included in the same thickened connective-tissue capsule, I think there can be little doubt but that they are vestigial milk-incisors, and that Freund, as Woodward says, is * over-cautious and inclined to underestimate the facts which suggest that condition.”” Again, Mr. M. F. Woodward (J. c.) has: described in the mouse “a pair of very minute calcified tooth- rudiments in connection with the two large upper incisors, one each side of the jaw.” These rudiments he considers to represent ‘€ the last stage in the reduction of a vanishing tooth, the earlier stages of which are to be seen in the rabbit and squirrel.” The guinea-pig will thus be seen to add a still further stage in the degeneration of this tooth, forming a very complete and inter- esting series. Two observers, Freund [10] and Adloff [1], have also examined specimens of Cavia at this stage; their measurements differ considerably from my own, and possibly the discrepancy may be due to the method of measuring: as I estimated the length from the tip of the snout to the root of the tail circumfer- entially, this would obviously give a greater length than if measured in a straight line. Freund has also examined two younger stages, the earlier one having a cephalic length of 45 mm., the measurement of the other not being given. This writer gives but few details as to the condition found in these three young stages, merely stating that he found no trace of a germ for the second incisor, and nothing in the diastema beyond a few thickened cones of epithelium, which he could not satisfy himself were to be regarded as tooth-rudiments. Adloff, on the other hand, gives more details of the examination of his speci- mens at this stage, and accompanies his description by three figures. His description of the tooth which he identifies as Pd.3 TOOTH-GENESIS IN THE CAVIIDA. 265 corresponds with the first cheek-tooth of my specimen, except that he mentions and figures a lingual downgrowth only, whereas in mine both lingual and labial downgrowths were present in the upper jaw, but only the former in the lower. This point is, I think, worthy of mention, for the reason that I have already [22] expressed my adherence to the current belief that the upper jaw retains the more primitive condition. In neither of these cases is the labial downgrowth present in the lower jaw, while in the upper it is present in the one case and not in the other. The conclusion appears to me to be that this labial downgrowth has already ceased to exist in the lower jaw, while in the upper jaw it is in the process of disappearance. The lingual downgrowth is present in both upper and lower jaws in Adloff’s specimen as well as in my own. In this I think one may find further con- firmation of the view that the dentitions present in the Mammalia tend to disappear from without inwards. Adloff makes no mention of any indication of the incisors, which are well marked in my specimen. He also says that m.1 and m.2 were not yet to be found; whereas m.1 was present in mine as a flask-shaped non-invaginated mass, m.2 not being indicated. Possibly mine may have been a slightly older speci- men, which would not invalidate, but rather strengthen, the deductions I have drawn. Stage 2. Circumferential head-length ...... 2°7 cm. body-length ....... 7°5 em. 99 Reconstructed diagram of Stage 2. pi., permanent incisor; ppm., permanent premolar; dm., deciduous molar; m!, first molar; m°, second molar ; m, third molar ; 0.c., ‘‘ concentric epithelial bodies.” The incisors are now teeth of considerable size and curve back- wards deeply in the substance of the jaw. They are well-calcified, 266 DR. H. We MARETT TIMS ON but the enamel appears to be deficient on the posterior surfaces. No trace of any downgrowth, lingual or labial, is visible. Some distance from the anterior end of the jaw, in a position posterior to where the permanent incisor cuts the gum, the dental lamina grows deeply into the substance of the jaw; it persists through some sections and then ceases, it is bilaterally symmetrical. Possibly this may mark the position of 7.2, which has dis- appeared. Throughout the diastema no trace of any tooth-germs is visible. The most anterior cheek-tooth is the so-called ppm.1, it is in a somewhat rudimentary condition and as yet quite uncalcified. It consists of a single cone with a blunted summit. The internal cingulum is present, and also indications of an external. I would here specially note the presence of a spherical body, composed of concentrically arranged cells, lying in the line of the connecting neck of dental lamina (Pl. 26. fig. 9). I have already figured [22] a precisely similar structure in con- nection with pm.4 of the dog, the only difference being a slight one of position. In the latter case, it lies at the free extremity of a labial downgrowth of the dental lamina; whereas in the present instance, it is directly in the line of the dental lamina running between the oral epithelium and the tooth. In the guinea-pig this structure is present on both sides of the upper jaw, and persists for some time. A similar structure is also to be found in connection with the so-called m.2. I have already referred (doc. cit.) to the fact that Mr. M. F. Woodward has found. a similar structure in Gymnura in the same position, viz., in connection with the posterior premolar. In discussing the question, I there stated that I was not able to give an explanation of the condition, “but from the fact of its connection with the dental lamina and its presence in precisely the same situation in these forms” (i.e. Canis and Gymnura) “ T do not think it is a chance structure, and it is possible that it may represent the remains of a predecessor to this tooth.” From a further consideration of this point in Cavia, I am still more inclined to this opinion. Mr. Woodward has figured a calcified vestigial incisor in the mouse [25] which, as he describes, appears in section “as a narrow loop forming about 2 of a circle, a few of the mesoblast-cells having flattened themselves against the outer surface of the dentine.” From a comparison TOOTH-GENESIS IN THE CAVIIDA. 267 of the two conditions, it is easy to imagine that a still further stage of degeneration would give the appearance seen in the dog, guinea-pig, and Gymnura. Just as the rudimentary germ of the premolar is on the eve of disappearing, the tooth immediately behind it, the so-called dpm., is commencing to appear, but placed much more deeply and lying altogether underneath the preceding tooth. Tracing it backwards, it is seen to be well-calcified, but has not yet reached the surface of the gum. No labial or lingual downgrowths are to be seen, though traces of both have been noted in Stage 1. There is an interval of 70 sections between the neck of this tooth and that of the preceding: consequently I think there can be no possible morphological connection between the two. If such be the case, then these two teeth can no longer be regarded as the morphological predecessor and successor, the one of the other. I believe the correct interpretation to be that the more anterior tooth is a premolar, probably pm.4, belonging to the successional series, and the so-called dpm. is the first true molar. The reasons which lead to this conclusion will be discussed below. IT shall therefore in what follows speak of the five teeth in the upper jaw of the guinea-pig as pm. 4, ms.1, 2, 3, 4. The deciduous tooth at this stage (Pl. 26. fig. 1) possesses two antero-posterior rows of cusps, and has the appearance of two similar portions one behind the other, the anterior being the larger. This remark applies to all the cheek-teeth both at this and later stages ; and for this reason I think there isa possibility of the correctness of the Concrescence theory. The external row has three distinct cusps, of which the centre one is the more pro- nounced and the posterior slightly smaller than the anterior. The internal row also consists of three cusps, the anterior being the largest. The middle cusps of the two rows are separated one from the other by a wide depression; while the first and third cusps of both rows are connected respectively by ridges which bound the depression anteriorly and posteriorly. Behind the posterior ridge is a second smaller depression separating the outer from the inner posterior portions of the tooth which do not possess any definite cusps. The anterior and central cusps form the anterior larger portion of the tooth; the third cusps with the posterior portion of the tooth together forming a miniature of the anterior part. The first and third cusps of the exterpal row are 268 DR. H. W. MARETT TIMS ON slightly undermined at their bases, representing the involution of enamel which is carried to such extremes in the teeth of some other Rodents. The second molar, or second cheek-tooth of the adult dentition, resembles the anterior part of the tooth just described. There are two external cones, the posterior being of considerable size, the anterior inconspicuous, and the same may be said of the two internal cusps. The antero-internal and antero-external cusps are partially fused with each other transversely, whereas the two posterior cusps are separated by a deep but narrow cleft (Pl. 26. fig. 8). The posterior part of the tooth is made up of a mass of considerable size with a rudiment of a cusp, both internally and externally, the latter being slightly the larger. At the base of the tooth, on both its outer and inner aspects, is a well- marked rounded prominence which I think must be regarded as the cingulum. The third molar is not calcified. It presents a broad, trans- versely elongated surface with an external and an internal cusp, the former being the larger. ‘here is a well-marked lingual downgrowth of the dental lamina. In connection with this tooth there is one of the concentric epithelial bodies to which I have already referred. If these bodies really do represent the last stage in the dis- appearance of a tooth, we have here in connection with an undoubted molar tooth evidence of three dentitions, from the central one of which the permanent tooth developes. Adopting the line of argument I have previously used when referring to the Marsupial dentition [21], it would seem to show that the molar teeth do belong to the successional series—a view which, though held by many, is not universaily accepted. The fourth molar is present in a very rudimentary condition. The second, and last stage of Cavia cobaya examined by Adloff had a head-length of 8 cm. This measurement corresponds exactly with my Stage 3; but from the description given, it is evident that Adloff’s was a much younger specimen, the difference being no doubt due to a difference in the method of measuring. From a comparison of the results, I am inclined to think that his specimen must have been slightly younger than my Stage 2. He finds that the first “Anlage” in the hinder portion of the jaw is that of the premolar of the first dentition, that is of the deciduous tooth. This is in agreement with what I have found TOOTH-GENESIS IN THE CAVIIDS. 269 namely, that this tooth commences to develope earlier than the tooth in front of it, which I believe to replace it. In connection with m.2 he describes and figures both lingual and labial downgrowths of the dental lamina; the latter he interprets as a pre-milk vestige. This interpretation I shall discuss subsequently. The posterior molar (m.3) he states is not at this stage developed. Stage 3. Circumferential head-leneth...... 3 cm. body-length...... 9°3 cm. 29 Reconstructed diagram of Stage 5. Lettering as before. The incisors are now large teeth which have just cut the gum. No trace of any vestigial tooth is to be seen, here or in the diastema. The first cheek-tooth is well-developed though not calcified. The “concentric epithelial body” is clear and distinct. It appears to occupy a similar position relative to the edge of the jaw as in the previous stage; but the tooth with which it is connected is now more deeply placed. Its connection with the surface being severed, the relative position of the ‘‘ concentric epithelial body ” to the neck of the enamel-germ can no longer be definitely ascertained. The tooth itself is transversely elon- gated, its axis in this direction being double that of the antero- posterior axis. Excluding the internal and external cingula, representatives of two antero-posterior rows of tubercles are present about the centre of this tooth, as is seen in PI. 26. fig. 2. Of these, the outer is more pronounced, and becomes the antero- external cone of the adult tooth. The second cusp from the outer side is the second largest; it attains its maximum at a point in a plane slightly posterior to that of the principal cone, where the latter is gradually shelving upwards. In the posterior 270 DR. H. W. MARETT TIMS ON part of the tooth these two cones are indistinguishable the one from the other, having fused to form a solid mass. The second cheek-tooth (Pl. 26. fig. 4), which is the deciduous tooth, is well calcified, and on the verge of cutting the gum. It consists of a large antero-external cone, the apex of which has an elongated cutting-edge, still showing indications of two tubercles. On comparison with the model of this tooth at the previous stage, this cone appears to be formed by a fusion of the anterior and central external cusps. The well-marked median internal cone of the earlier stage has now entirely disappeared. The postero- external and postero-internal cusps are relatively much smaller. Owing to the greater size of the tooth, the excavation of the posterior part of the tooth in a forward direction appears to be more pronounced. On the other hand, the undermining of the base of the antero-external cusp from within outwards, previously mentioned, is now scarcely perceptible. No marked trace of the external cingulum is present either in this or the preceding stage. Microscopically, the epithelial neck of the tooth-germ is no longer visible; and I have not been able to detect any lingual downgrowth such as would suggest that this tooth had any morphological successor. The condition of the third cheek-tooth has become complicated by the presence of infoldings of the enamel. Here again, there is a large external cone which has to the outer side of its base two minute cusps, which I consider as belonging to the external cingulum. The well-marked inner cone, described in the pre- ceding stage, appears to have fused by its apex with that of the external cone, the two being separated at their bases, giving in section the appearance of an elongated foramen. It might be thought that this foramen was due to the tunnelling forwards of the substance of the tooth from its posterior end, such as was found in the second cheek-tooth. Such, I believe, cannot be the cause, for two reasons: (1) in the previous stage no trace of any tunnelling is observable; and (2) the external and internal cones are separated by a deep fissure extending down to almost the root of the tooth. This foramen seems, therefore, to be due to a fusion of the apices of the two cones cutting off the deeper part of the fissure from the surface. This occurs towards the anterior part of the tooth. About the centre a communication takes place between this cavity and the internal surface of the TOOTH-GENESIS IN THE CAVIIDA. 271 tooth, separating the internal cone from its base; and as this detached portion is fused by its apex with the external cone, the tooth in section has the appearance of an inverted V, the external limb of which is considerably longer than the internal. How this communication is brought about, whether by the rupturing inwards of the central cavity, or by the extension into the latter of a channel running outwards from the internal surface, I am not in a position to say. If the condition of this tooth be traced still further backwards, the communication is still seen to be present ; and, in addition, the central cavity communicates with the surface, the apices of the external and internal cones being separated. This may have been brought about by the formation of a cleft from with- out inwards, or from within outwards ; or, what I think the more probable is, that the apices of the two cones have here remained separate, not having undergone fusion, as in every instance the posterior moiety of the tooth seems to be in a somewhat earlier stage of development than the anterior. Ina section through this region, the apex of the internal cone Fig. 4. sues Diagrammatic sections through Third Upper Cheek-tooth. A from Stage 2. B, C, D, H, from Stage 3. lies as an isolated mass to the inner side of the external, the latter retaming its connection with the fused bases of the two cones, appearing almost identical in section with that through the centre of the deciduous tooth. This I regard as being a 272 DR. H. W. MARETT TIMS ON fact of some importance, and to which I shall again have to refer. In quite the posterior part, the tooth forms a solid transverse mass with a blunted apex, slightly more prominent on the outer side. The accompanying illustrations will, perhaps, render this description more intelligible (fig. 4). I would here add that I have found a similar condition in m.1 of the rat, the only exception being that in this animal there is the further complication of a channel of communication between the central cavity and the external surface of the tooth, which cuts off the apex of the external cone from its basal attachment. As the apex of this tooth in the guinea-pig now nearly reaches the surface of the gum, its epithelial connection is broken up, no labial or lingual downgrowths of the dental lamina nor any appearance of a “concentric body” being visible. The fourth cheek-tooth is almost identical in pattern with that of the third as seen in Stage 2, and the description afore given would apply equally to the tooth under consideration; the only addition which it is necessary to record, is the presence of a relatively large tubercle to the outer side of the main cone in the posterior part of the tooth. It attains to such a size, that were the cusps not carefully followed throughout, it might easily be mistaken for the main external cone with a slightly more prominent internal cone. The fifth cheek-tooth is deeply situated, its enamel organ being in the bell-shaped stage. Between this and the tooth in front is another of the “ concentric bodies” to which reference has been made. Its exact relationship to the tooth bebind cannot be established, owing to the dental lamina being difficult to follow. It can, however, be seen to lie well to the labial side of the teeth and between them. Stage 4: (Fetal) (fig.5). Circumferential head-length.. 4 cm. p body-length.. 10cm. The jaws were examined by clarifying in oil of cloves. The condition is interesting as the deciduous tooth is about to be shed, having entirely disappeared in the subsequent stage ; it is therefore lost either just after birth, or, as is more generally stated, at quite the late period of intra-uterine life. All five teeth are well-calcified, the deciduous tooth lying between the crown of the anterior permanent tooth and the free margin of the TOOTH-GENESIS IN THE CAVIIDA. DATS gum, somewhat to its lingual side. The anterior cheek-tooth of the adult consists of two plates, an external and an internal, separated one from the other by a deep oblique cleft, so deep that the plates are but slightly connected by their bases and are easily separable. The position of the tooth is partially rotated through an angle of about 45°, so that the external and more prominent cone lies antero-externally, the internal being postero- internal. When examined from the internal surface, the apex of the outer cone appears to be folded inwards so as to reach the apex of the internal cone, but without being fused with it. The folding gives rise to a transverse groove crossing the apex of the external cone. Fig. 5. A. View of teeth in Left Upper Jaw (Stage 4), seen from the inner side. B. Crown surface of Deciduous Tooth. C. ss = Successional Tooth. 10), ss 5 Fourth Cheek-tooth. The inwardly-folded apex is supported by a vertical ridge upon the inner face of the external cone, so that this cone would on horizontal section have a triangular shape, the curved base being external, the apex internal with a slightly backward inclination, which is applied to, but not fused with, the outer face of the internal cone towards its posterior margin, as seen in fig. 5 C. It lies in a separate capsule of its own, quite distinct from that of the deciduous tooth, which is nearer the margin of the gum. The latter has not undergone any rotation, is very minute and its cusps complete, the enamel not having disappeared from their apices. The third and fourth cheek-teeth had already assumed the characters of adult teeth, the former being the larger. With regard to these teeth, I would note the absence of tubercles from the crown-surface, the enamel having partially 274: DR. H. W. MARETT TIMS ON disappeared even though the teeth had not cut the gum. This confirms the observation made by Saint-Loup [18], which led him to ask the question whether this is a case of the hereditary transmission of acquired characters, since it cannot be due to wear. The posterior tooth is much smaller; its characters I was not able definitely to ascertain owing to its position and the difficulty of dissecting it out from its osseous surroundings. The surface view of the deciduous tooth is seen in outline in fig. 5B; a comparison of this with the teeth shown in Pl. 26. fies. 7 & 8 is, I think, suggestive, the same general pattern being noticeable: more particularly is this the case on com- parison with the tooth of Stichomys of the Lower Eocene; the latter is, however, slightly more complicated. This fact may afford some additional argument in favour of the multituber- culate origin of the Rodent molars. Stage 5 (Post partum). Circumferential head-length 5 cm. 2 body-length 12 cm. Fig. 6. 19 OV ON A. View of teeth in Left Upper Jaw (Stage 5), seen from outer side. B. Crown-surface of permanent premolar and three molar teeth. Examination of the clarified jaw shows the presence of four cheek-teeth only, all traces of the deciduous molar having dis- appeared. Of these teeth the second is the largest, and is the only one in addition to the incisor which has actually cut the gum. AJ] have assumed the characteristics of the adult denti- tion. The enamel is absent from the crown-surfaces, all trace of definite cusps bemg wanting. A side view of the teeth situ is shown in fig. 6 A, while the crushing surfaces are repre- sented in fig. 6 B. TOOTH-GENESIS IN THE CAVIID®. 275 Having now given an account of the conditions found in these various stages, the question arises as to what interpretation may be placed upon them. In the first place, it will be noted that the deciduous tooth is the first to arise, or at least is the more advanced in the first stage examined. In tracing its subsequent development, it does not appear to be connected with any germ which could be interpreted as either a predecessor or a successor: - in other words, this tooth seems to be represented in one denti- tion only, and it might be either a milk-tooth the successor of which had become suppressed, or vice versa; and I know of no definite data upon which to form a decided opinion. The fact of its early appearance, and of its being shed zn utero, might seem to favour the former alternative. J am inclined, however, to regard it as a permanent tooth accelerated for the following reasons :—(1) its development is but little in advance of the incisor or of the molar immediately behind it, both of which are known to belong to the second dentition; (2) in the dog the carnassial tooth of both dentitions is developed in advance of the other teeth, which supports the view that teeth in this position have a tendency to become accelerated; and (3) I believe that the dentitions in the Mammalia tend to disappear from without inwards. This being so, I would regard all five tooth-germs found in the guinea-pig as belonging to the permanent series, the first and fourth having, in my opinion, vestigial remains of milk predeces- sors in the “ concentric bodies”’ afore described (Pl. 26. fig. 7). Though the pattern of the so-called molars and premolars in the adult Cavia are practically identical, still in their earlier stages they are somewhat different, the anterior cheek-tooth being transversely broader and more multitubercuiate than the posterior ones. The deciduous tooth appears to me to partake more of the nature of the latter; and I am inclined to believe that it is to be regarded as the anterior molar, thus agreeing with the suggestion first made by Woodward in relation to other mammals. The tooth which replaces the deciduous is the anterior cheek- tooth, which is not therefore its true morphological successor, but merely drops backwards and occupies its position. In this connection it is interesting to note what Forsyth Major says [8] in referring to Prolagus sardus :—“ The anterior of the three deciduous teeth is not situated directly above the 276 DR. H. W. MARETT TIMS ON anterior premolar, but slightly backward, closely appressed to the second deciduous, so that with its anterior moiety it covers only the posterior part of the premolar ; besides it could not possibly cover the latter completely, being much smaller.” And, as he points out, Fraas [9] states that the anterior pre- molar in Prolagus has no deciduous predecessor, but that it “comes into place through the same lacuna.’ Clearly, then, this is not a unique condition in the Rodents, and I have referred [22] to what I believe to be a somewhat similar condition in Canis. Regarding the deciduous tooth as the first molar, its successor in position would be pm.4. Connected with the latter is a “ concentric body.” Though this tends to support my con- tention, too much weight must not be attached to it, since a similar structure has been referred to in relation to the posterior teeth in Cavia. Another question opens up: If these “‘ concentric epithelial Lodies” are really tooth-vestiges, since they are found in the posterior part of the jaw, either the posterior cheek-teeth must be regarded as premolars, or else these bodies are the vestiges of deciduous molars, and therefore the usually accepted distinction between molars and premolars breaks down. I am inclined to the latter opinion, having always held that the molars belong to the permanent series. If this be a correct interpretation, then in this particular the Rodents retain an extremely primitive condition ; and the statements made by Fraas with regard to the tooth-change in Prolagus may not be far wrong, and certainly do not merit the unfavourable comments which have been made upon them. It will have been seen that the cheek-teeth above described arise as a single Primitive cone to be soon followed by the appearance of external and internal cingula. As the tooth elongates, two antero-posterior rows of cusps arise; the primary cone becoming the median-external in position and the largest in size. The anterior and posterior cusps of each row respectively become united forming transverse ridges, the median cones remaining separated by a cleft. The anterior cusp and median cone of the outer row together with the anterior transverse ridge form a crescentic edge, and this gives rise to the anterior moiety of the adult tooth. The median-internal cone disappears. The postero- internal cusp, together with a subsequent backward extension of the end of the tooth, forms the posterior part of the adult tooth, TOOTH-GENESIS IN THE CAVIIDR. © 2G while the narrow band connecting the anterior and posterior portions represents the posterior transverse ridge. In the younger stages the teeth are decidedly more multitubercular than in the adult. Compare fig. 2 (Pl. 26) with the adult teeth of Cavia as shown in fig. 7 B. In the posterior molars a further small postero-external extension of the tooth arises in the form of a rounded process. It is also present, though to a much less extent, in the anterior cheek-teeth. Itis very pronounced and plicated in the posterior molar of Dolichotis, and its size forms the principal difference between the molars of this fossil rodent and those of the existing Caviide (Pl. 26. fig. 7). It may here be noted that the root of the tooth is frequently seen to be lateral in position, as shown in Pl. 26. fig. 10. This is not usually so marked in other animals, in which the obliquity of the adult teeth is not present to the same extent as in the Rodents. Within the limits of the Mammalia comparative odontologists have referred to the existence of fowr distinct dentitions—a Pre-milk, a Milk or Deciduous, a Permanent or Successional, and a Post-permanent. The existence of all four at one and the same time has not, so far as I am aware, been shown to be present in the same animal. The Pre-milk dentition is said by Leche and others to be present in the Marsupials, but such an inter- pretation of the vestigial representatives, such as undoubtedly occur in Myrmecobius, Phascologale, Dasyurus, and others, depends upon the functional teeth of the Marsupials representing the true milk dentition. I have previously [21] expressed my belief that another and more probable explanation is forthcoming, and in this opinion I am supported by Wilson & Hill [24] and by Tomes [23]. I would regard therefore Leche’s vestiges as remains of a deciduous dentition. The evidence as to the existence of traces of a Post-permanent dentition in many mammals is, I think, undoubted: they have been described in Man, Seal, Hedgehog, and Dog; and if my interpretation be correct, it is also to be found in Kikenthal’s lingual down- growths of the dental lamina as described by him in Didelphys and in the Cetacea. In the Rodentia there are well-marked evidences of at least two dentitions—the milk and permanent, though the former seems tending to disappear. LINN. JOURN.—ZOOLOGY, VOL. XXVIII. 20 278 DR. H. W. MARETT TIMS ON The only evidence of the existence of prelacteal vestiges that I have as yet been able to meet with, is to be found in a paper by Adloff [1]. He describes and figures such vestiges in connection with Jd.3 and 7d.2 in Spermophilus| citillus (head-length 1-5 cm.), and Pd.2 and Pd.3 and 7.2, Pd.2, and Pd.3 in Spermophilus lepto- dactylus (head-length 2°71 cm.); also in Sciwrus Brooket in connection with Pd.2 and Pd. 3, though he does not mention their existence in two other species which he examined, viz., Securus Prevosti and 8. vulgaris. He further notes their presence in connection with Pd.3 and .2, m.2 in Cavia cobaya (1°5 em.). Thus there seems, at first sight, abundant evidence in favour of the existence of the Pre-milk dentition in these animals; but before implicitly accepting these statements, it is necessary to examine them somewhat more critically. Firstly, Adloff’s inter- pretation in connection with the molars of Cavza depends entirely upon his belief that the molars belong to the Deciduous or Milk dentition. In this opinion he follows Hoffman, Beauregard, Owen, Leche, and others ; on the other hand, Woodward, Lataste, and Magitot believe the molars to belong to the successional series, and in this opinion I concur. Consequently, according to the latter view the vestiges mentioned by Adloff in this position would be regarded as vestiges of the milk predecessors of the molar teeth ; and, indeed, I have already described the existence of such a vestige on the labial side of the molar teeth in a foetal pup of about the seventh week [22], as well as the presence of a “concentric body ” to the labial side of the molars in the guinea- pig; in both of which cases I have interpreted them as vestiges of a milk dentition, the lingual downgrowth representing a Post- permanent dentition. With regard to the existence of prelacteal vestiges in the premolar region in the afore-mentioned forms, I may point out that I believe them to be the only examples recorded, Leche’s and Woodward’s discoveries being confined to the outer incisor region ; the very region in which I found well-marked evidences of three dentitions in the dog, but in that instance the three undoubtedly being the Milk, Permanent, and Post-permanent. With regard to the presence of Pre-milk vestiges in connection with Pd.2 of Sciwrus Brookei, there are certain points to be borne in mind. As I understand Adloff, this tooth Pd.2 is the anterior of the two premolars so generally present in the Sciuride. He only TOOTH-GENESIS IN THE CAVIIDE. 279 mentions having examined one stage of this species with a head- length of 1°5 cm., and he mentions that neither 7. 2 or m.3 were developed. Now in Cavia we have seen that the anterior cheek- tooth present does not develope until after the appearance of the second and third. Similarly, in the dog I have shown (loc. cit.) that the smal] anterior premolars do not appear until some time after the larger posterior ones. According to Flower and Lydekker [5, p. 450] the first upper premolar is “small and deciduous.” It therefore seems possible, if not probable, that the tooth which Adloff identifies as Pd. 2 is in reality Pd. 3, that is the posterior deciduous tooth. Now the deciduous tooth in the guinea-pig I have shown to be developed in series with the molars, and in them I have shown the presence of “ concentric bodies” which I regard as milk vestiges. In like manner, there- fore, the so-called Pre-milk representative in connection with this tooth might be so interpreted. Whether this be the correct explanation or not, I do not consider it possible to identify the teeth from the examination of the condition found in a single specimen. With regard to Pd.3 in Cavia, I think the interpretation given by Adloff is erroneous, since he appears to have missed the peculiarity of the tooth-change; the “‘prilakteale Anlage” in this case being identical with the “ concentric body.” As to the incisors in Spermophilus I am unable to express a definite opinion, not having had an opportunity of examining a specimen; nevertheless I would point out that Iam not in accord with Adloff in his identification of dentitions, and the interpretation previously given with regard to the incisors in ‘Cavia probably applies equally in the case of Spermophilus. I claim to have shown reason for believing that the existence of pre-milk vestiges in the Rodents is still ‘““non proven” ; and I cannot refrain from quoting Woodward, who, though a believer in their existence, and having carried out extensive researches on the dentition of the Rodents, says [27]:—“ I do not think it is probable that we should find traces of such a vestigial structure persisting in a specialized group like the Rodentia; the ancestry of which are to be sought according to Cope in the generalized Tillodontia, who in all probability possessed a typical milk- dentition which has become gradually suppressed as their descendants became more and more specialized.” 20* 280 DR. H. W. MARETT TIMS ON Of the various theories which have been propounded to account for the origin of the complex crown of the Mammalian molars from the Haplodont type, it will be necessary to refer to three only. They are the Tritubercular, the Multitubercular, both of which are well known, and the theory of Cingulum-Cusp development suggested by me in my work on the Canide. Tritubercular Theory.—This view,so ably advocated by Cope and Osborn in America, and so widely accepted both in this: country and on the continent, is too well known to need any re-statement. One of the important arguments advanced against this theory is that the Paracone, and not the Protocone, is ontogenetically the first to appear. This has been found to be the case in Marsupials [15], Carnivores [22], Ungulates [20], some Insectivores [28], and Primates [16 and17]. To these may now be added Cavia, as representing the Rodents. The other cusps are secondarily added. The Multitubercular Theory, first put forward by Forsyth Major, suggests the primitive condition of the mammalian cheek-teeth to have been multituberculate, and that during the course of evolution a diminution in the number of cusps has taken place. This theory does not appear to me to be applicable to such Orders as those just mentioned with their full complement of teeth, and in which embryology has shown that the teeth develope by the addition of cusps to a single primitive cone.. It is evident that this theory presupposes the acceptance of the Concrescence theory as set forth by Dybowski, Gervais, Rose, and more particularly by Kikenthal. Though I am unable to accept these combined views as a whole to account for the origin of the Rodent molars, nevertheless they afford a certain amount of satisfactory evidence; the suggestion which I would offer will be discussed subsequently. Theory of Cingulum-Cusp development.—The uniformity of development of the antero-external cones in both jaws suggests the Paracone and Protoconid as being homologous, and as representing the primitive reptilian cones. The remaining cusps I believe to have been mainly derived from the Cingulum, a structure of great antiquity as shown both embryologically and paleontologically. The details of the subsequent development of the cusps I need not here repeat, as they have been already published [22]. This view is in harmony with the Tritubercular theory up to the TOOTH-GENESIS IN THE CAVITD®. 281 point at which the rotation of the cusps is presumed to have taken place. In tracing the course of the molar evolution in Cavia, it has been found that the tooth begins by the formation of a single cone, which subsequently becomes the antero-external cone of the adult tooth, so far agreeing with what has been noted in other mammalian orders. In the development of teeth from a multituberculate type as usually understood, one would not expect to find the development of a single cone taking place first, as is the case here and elsewhere, but of several. Ag I have already shown, this single cone of the guinea-pig has both external and internal cingula, the latter bemg the better marked. Both develope secondary cusps, which disappear in the course of the subsequent development of the tooth. There is thus a tendency to the suppression of cusps after a certain period, the adult tooth being less multituberculate than at an earlier stage, though more so than in its youngest condition. Consequently there is evidence in the later stages of development in favour of the Multitubercular theory. Though I have not personally met with any direct embryo- logical evidence in support of the Concrescence theory, yet upon general grounds I am disposed to accept it to a certain extent. This theory supposes a fusion to have taken place not only antero-posteriorly of teeth of the same dentition, but also transversely of teeth of different dentitions. The former would account for the diminution in the number of teeth of the same dentition in the transition from the Reptiles to the Mammals ; while the latter was suggested in order to explain the existence of the triple longitudinal rows of cusps as seen in the fossil Multituberculata. It is the latter part of this theory that I find myself at present unable to accept. The progressive shortening of the jaws would naturally tend to a crowding of the teeth, which may be conceived to have become fused antero-posteriorly as a result, and Ameghino [2] has adduced some presumptive evidence in support of this; and I have already referred to the fact that the cheek-teeth in the guinea-pig have similar anterior and posterior portions. It is possible to believe that there may be some close connection in the way of cause and effect between the two processes, though actual evidence is as yet wanting. On the other hand, it is 282 DR. H. W. MARETT TIMS ON dificult to imagine how shortening of the jaws could have had any effect in bringing about a fusion of teeth of different denti- tions; nor, indeed, can one perceive any other change which would produce such an effect. The suggestion I would offer is, that the three longitudinal rows of cusps are due to the primitive cones with cingulum-cusps developed to their inner and outer sides respectively. It may be objected, that these inner and outer secondary cusps are as pronounced and of equal size as the central primitive cones in the true Multituberculata. I do not, however, consider this to be any great difficulty, since the Multi- tuberculata must have been extremely specialized animals, as is shown by their dental formule ; and, moreover, a very similar condition of the cusps is to be seen in the molars of existing frugivorous bears. Hach molar tooth of the Plagiaulicide and Polymastodontide, in which there are three longitudinal rows of cusps arranged in numerous transverse rows, would consequently represent an antero-posterior fusion of several teeth with their external and internal cusps. In other members of these families, for example Bolodon, in which the molars bear only two antero- posterior rows of tubercles, one of the three rows is non-developed. From a comparison with the teeth of existing mammals, I am inclined to believe that the series in this form which is wanting is that of the external cingulum, it being quite exceptional to find this series well-developed, though it is to be found in some of the Insectivora and in Ofocyon among the Canide. This conclusion receives some confirmation from a comparison with the teeth of several species of the Polymastodonting. In a paper by Osborn and Earle [13] describing these, they state that in P. taoensis, “ although the lower molars typically exhibit but two rows, we occasionally observe a postero-external accessory row upon the first and second molars;” and again, “ the comparison with Meniscoessus shows an average addition of two cusps to the first molars in both jaws, and an apparent degeneration of the outer row in the second upper molar.” In the course of the development of the molars in the guinea- pig, the three longitudinal rows of tubercles are present as a transitory condition, the external cingulum disappearing giving rise to a tooth with but two antero-posterior rows of tubercles (Pl. 26. fig. 6). The next point to which I would refer is, the similarity of dentitions found in the Rodentia and Multituberculata. In both, TOOTH-GENESIS IN THE CAVIIDA. 283 the incisors are reduced in number, there is an absence of canines with the presence of a diastema. The number of cheek- teeth in some of the Multituberculata is in excess of that found in most Rodents, while in others it is not in excess of that present in the Lagomorpha. In any case, this is only what might be ex- pected, as itts well known that a progressive reduction in the number of cheek-teeth is, and has been, taking place throughout almost the whole mammalian series. There still remain other points to be considered in the same connection. Within the limits of the existing Hystricomorpha very different patterns of cheek-teeth are to be found. The crown-surfaces of the so-called anterior permanent premolar of Aystrix leucura (Camb. Zool. Mus. 861 D) is shown in fig. 7, Fig. 7. Crown of First Upper Permanent Premolar of Hystrix leucura before eruption. Enlarged. (Camb. Univ. Zool. Mus.) and of Cavia cobaya in Pl. 26. fig. 7B. The former is decidedly multituberculate, the latter is not. In all the Hystricide the tooth-change is now known to occur and comparatively late in life, whereas in Cavia the deciduous tooth is shed zm utero. It may, therefore, justly be inferred that the Hystricide are more primitive in this respect than are the Caviide. Though this may not altogether justify any conclusions as to the pattern of the molar crowns, nevertheless, taken in conjunction with what has been said above, I think it affords some additional evidence in favour of the conclusion that the multituberculate is the primi- tive pattern of the Rodent molars. Lastly, there is a large amount of evidence collected by Forsyth Major in favour of this view, which he was the first to set forth in his paper on the Miocene Squirrels [6]. This opinion, however, was not shared by Cope, and is not by Osborn. Cope derived the Rodentia from the Tuillodontia, a suborder of the Bunotheria, from a type closely allied to Esthonyx, 284. DR. H. W. MARETT TIMS ON Psittacotherium being not far from, if not on the direct, line of ancestry [4]. His arguments are based mainly on the presence or absence of the first and third incisors; and the condition found in Esthonyx, Psittacothertum, Calamodon, and Tillothercum are referred to as evidence in support. An elaborate theory is then drawn up to show how the Rodent molars may have been produced mechanically from the molars occurring in the above- mentioned fossil forms. This theory, though very ingeniously worked out, is but a theory, and cannot be admitted as evidence. As to the incisors, though I admit that these forms may be so arranged that different stages in their reduction may be made to appear, and the increase in the size of 7.2 to become evident, nevertheless it must not be forgotten that a similar condition is to be met with in the Multituberculata, two incisors only being characteristically present in the genus Polymastodon, of which one is very large and “‘rather slender, sharply grooved, restricted enamel-band, and a deep postero-external groove. The lateral incisor [2.8] is a very small conical tooth, compressed antero- posteriorly, with its enamel confined to the anterior surface.” Further, in a note (“ Note on the Marsupialia Multitubercu- lata”) appended to his paper (Joc. cit.), Cope stated that the incisors of the Plagiaulacide, Chirogide, and Polymastodontide are similar in structure and functions to those in the Rodentia. Osborn and Earle also say (loc. cit.) that the condyle of the lower jaw is “oval, and its long axis is placed obliquely, not antero-posteriorly as in the Rodents.” Cope refers to this latter fact as an objection; but it appears to me to be only an objection to his “ mechanical theory,” and not to the multi- tuberculate theory of descent of the Rodents, for, according to Osboru and Earle (Joc. cit.), the obliquity is ‘greater in some specimens than in others,” which shows that this is a character which is not stable but undergoing modification. A further objection may be cited from the joint paper of these authors, as they say that Polymastodon foliatus is the most primitive type of the genus, being “ distinguished by small size and very few tubercles.” With regard to size, I do not think it is necessarily any proof of primitiveness ; and as to the number of tubercles, surely the statement partakes somewhat of the nature of “ begging the question.” To sum up the matter, it appears to me that the balance of TOOTH-GENESIS IN THE CAVIID®. 285 evidence is distinctly in favour. of the multituberculate origin of the Rodents. This leads to a consideration of the fossil Rodents, for a know- ledge of which, especially of the South American forms, we are indebted very largely to the researches of Ameghino, whose work on the fossil mammals of the Argentine [3] forms the source from which other writers have largely drawn. Winge [25], Schlosser [19], and Forsyth Major [6, 7, 8] have also added much that is of value upon the same subject. Four genera of the Caviide are reputed to be found in the Tertiary and Pleistocene of Brazil and the Argentine. The members of this family are easily recognized by their high molars composed of two or several triangular prisms which generally form straight lamelle, and “determinent une aréte tranchante sur la face interne des dents 4 la machoire supérieure, et une _ aréte externe 4 la machoire inférieure.” From a consideration of the characters of the post-tympanic and jugular processes and of the masticatory muscles, Winge [25] regards Cavia, Dolichotis, and Hydrocherus as descendants of the American Capromyine and places them close to Dasyprocta and Oelogenys. Ameghino [8], dividing the Hystricomorpha into eight families, places the Capromyide, the Eromyide, and the Caviide in close affinity, and these three families together close to Octodontide, the points of difference being mainly dental. It will thus be seen that these two authorities agree in their general conclusions. A careful study of the tooth-pattern of the fossil Caviide, as figured by Ameghino, throws but little light upon the evolution of the molar crowns, there being apparently but little change of pattern from the Eocene, though a comparison of the posterior upper molar of Dolichotis from the Pliocene and of Cavia shows a reduction in the latter (Pl. 26. fig. 7). In the table of genealogical descent suggested by Ameghino (Joc. cit.), he would derive Cavia from Hedimys through Paleo- cavia, Hocardia, and Phanomys; there is, however, practically no difference between the molars of these forms such as would throw any light upon the tooth-genesis. If the Euromyide of the Inferior Eocene be compared with the later Pliocene forms, there is a simplification of the molar crowns though obviously of the same pattern. The more so is this the case on comparison with the Octodontide. > 286 DR. H. W- MARETT TIMS ON The same may be said with respect to the Hystricide. The diagram (Pl. 26. fig. 8) shows the outline pattern of Stichomys, Spaniomys, and the recent Hystria leweura (Camb. Zool. Mus.), fig. 7, p. 283. As the teeth are much worn, no information can be obtained as to the original disposition of the tubercles ; nevertheless, the outline-pattern, though simpler, is sufficiently similar as to suggest a possible line of descent. Though it is difficult to obtain any decided results from a comparison of the fossil teeth, owing to the wearing-down to which they have been subjected, still it seems evident that the complexity of the molars, which is undoubtedly more common in the existing forms than in the earlier ones, is due to the external and internal plications of the enamel rather than to the develop- ment of new tubercles. These animals, which are undoubtedly Rodents with the characteristic dentition and molar pattern, extend back to the Inferior Eocene. The Tillodontia are not found before the Lower and Middle Hocene, at which period, as we have seen, typical Rodents are present. It is difficult to conceive that the well-developed canines should have disappeared so rapidly and so suddenly together with at least two premolars, and that the incisors, which are only “becoming scalpriform” [26] in the Tillodontia, should have so quickly developed, if we are, with Cope, to regard them as the ancestors of the Rodents. And again, the Rodent molars have already assumed their charac- teristic pattern, whereas the molars and premolars of Tillotheriwm are “distinctly tritubercular, while those of Hsthonyx are quite unlike any Rodent molars” [4]. There is also the fact that the humerus in the Tillodontia possesses an entepicondylar foramen, which is not present in any existing Rodent. On the other hand, some of the Multituberculata are considerably older than the earliest known fossil Rodent, extending back into the Jurassic Period. In them the canines and several of the premolars have already disappeared, the incisors reduced in number, one being large and functional, and the pattern of the cheek-teeth in some instances approaching even in some degree to the unworn teeth of the existing Hystricide. TOOTH-GENESIS IN THE CAVIID®. 287 Summary and Conclusions. 1. That the deciduous tooth in Cavia is the first cheek-tooth to develope, the tooth immediately behind it being the next to appear. 2. That the deciduous tooth is replaced by the tooth which developes in front of it, which is its successor in position only, and is not its true morphological successor. 3. The general pattern of the deciduous tooth resembles more closely that of the posterior cheek-teeth than of the anterior tooth. Consequently, the deciduous tooth may possibly represent the first of the so-called molar series. 4. No trace of any representative of a true pre-milk dentition has been discovered. 5. The presence of “concentric epithelial bodies” has been noted in connection with the first and third cheek-teeth. It is suggested that these bodies represent the last traces of milk- teeth. If this be correct, then it would tend to confirm the view, which is not accepted by all, that true molar teeth belong to the permanent series. It would alyo lead to the conclusion that the usually accepted fundamental difference between premolars and molars did not always hold good, the molars having milk prede- cessors, of which these bodies are the vestiges. Moreover, the presence of a similar structure in connection with pm.4 of Gymnura and of Canis, tends to confirm the opinion expressed above that the deciduous tooth is the first tooth of the molar series. 6. That in the evolution of the cheek-teeth there is a tendency to the suppression of some cusps and a fusion of others. This conclusion, in conjunction with the evidences of Paleontology, is in favour of the multituberculate origin of the Rodentia. 7. That the first cusp to develope is the antero-external, the so-called Paracone, and not the Protocone as should be the case according to the Tritubercular theory. 8. That a rotation of the whole tooth takes place through an angle of about 45°, probably due to the peculiar conformation of the Rodent jaw, so that the anterior part of the adult tooth is represented chiefly by the external cone, the posterior part chiefly by the postero-internal cone. 9. The complexity of the Rodent molars is further increased by involutions of the enamel, the first to appear being at the lingual side of the tooth, and followed by another on the external surface in the teeth of the Rat. 288 DR. H. W. MARETT TIMS ON List of Authors cited in the text. 1. Avtorr (Paul).—* Zur Entwicklungsgeschichte des Nage- thiergebisses.” Jenaische Zeitschr. f. Naturwiss., Bd. xxxii. (1898) pp. 347-411. 2. Amecuino (Florentino).—“ Sur l’Evolution des Dents des Mammiféres.” Bol. Acad. Nac. Ciencias en Cordoba, 1894, pp. 381-517. 3. Amucutno (Florentino).—‘ Los Mamiferos Fosiles de la Republica Argentina.” Buenos Aires. 1889. 4. Corr (BH. D.).—“‘ The Mechanical Causes of the Origin of the Dentition of the Rodents.’’ Amer. Nat. vol. xxu. 1888, pp. 3-18. 5. Fiuowrr (W. H.) and Lypexxer (R.).—‘‘ Mammals, Living and Extinct.” 1891. 6. Forsyrn Masor (C. I.).—“‘ On ‘some Miocene Squirrels, with remarks on the Dentition and Classification of the Sciurine.” Proce. Zool. Soc. 1893, p. 179. 7. Forsyrn# Masor (C. I.).—“ On the Malagasy Rodent Genus Brachyuromys.” Proc. Zool. Soc. 1897, pp. 695-720. 8. ForsytH Masor (C. I.).—“ On Fossil and Recent Lago- morpha.’ Linn. Soc. Trans., Zool. ser. 2, vol. vii. pp. 433- 520. 9. Fraas (D.).—‘‘ Die Fauna von Steinheim.” Jahresh. des Vereins fiir Vaterland. Naturk. in Wirttemberg, 1870, pp: 145-306. 10. Frevnp (Paul). Beitrage z. Entwicklung d. Zahnanlagen bei Nagethieren.” Archiv f. mikr. Anat., Bd. xxxix. (1892) pp: 525-554. 11. Huxtey (T. H.).—“‘ On the Evolution of the Mammalia.” Nature, vol. xxiii. p. 228. 12. Huxtey (T. H.).—“‘On the Application of the Laws of Evolution to the Arrangement of the Vertebrata, and more particularly of the Mammalia.” Proc. Zool. Soc. 1880, pp- 649-662. 13. Osporn (H. F.) and Ear.z (Charles).—‘“ Fossil Mammals of the Puerco Beds.” Bull. Amer. Mus. Nat. Hist. vol. vii. (1895) pp. 1-70. 14. PoucuEr et Coapry.— Contributions 4 l’Odontologie des Mammiferes.” Journ. Anat. et Phys. 1884, p. 147. via Jie ia a eas ty ye ia Neogene i phi ian nw. Zoor. Vou. XXVIUL Pu. 26. =? . JOUR Lax. Soc Times. A su < ’ Parker & West imp. MP Parker hth. TOOTH GENESIS IN THE CAVIIDA. TOOTH-GENESIS IN THE CAVIIDA. 289 15. Rosz (C.).—‘‘ Ueber die Zahnentwickelung des Beutel- thiere.” Anat. Anz., Bd. vii. (1892) p. 693. 16. Rose (C.).—“ Ueber die Entwicklung der Ziihne des Menschen.” Archiv f. mikr. Anat., Bd. xxxviii. (1891) p- 447. 17. Rosx (C.).—“ Ueber die Entstehung und Formabinderungen der menschlichen Molaren.” Anat. Anz., Bd. vii. (1892) p. 392. 18. Saryt-Love (M. Remy).—“ Recherches sur l’Evolution des Dents chez les Rongeurs.” Bull. Mus. Hist. Nat., tome iii. (1897) pp. 315-317. 19. Scutosser (M.).—“< Die Nager des europaischen Tertiars.” Paleontographica, Bd. xxxi. (1884-85) pp. 19-323. 20. Tarker (J.).—‘* Zur Kenntniss der Odontogenese bei Ungulaten.” Inaugural-Dissertation. Dorpat, 1892. 21. Tims (H. W. Marett).—‘‘ Notes on the Dentition of the Dog.” Anat. Anz., Bd. xi. (1896) pp. 5387-546. 22. Tims (H. W. Marett).—‘‘On the Tooth-genesis in the Canide.” Linn. Soe. Journ. vol. xxv. 1896, pp. 445-480. 23. Tomes (C. S.).—‘* Manual of Dental Anatomy.” 5th ed., London, 1898. 24. Witson (J. T.) and Hut (J. P.).—*“ Observations upon the Development and Succession of the Teeth in Perameles, &ec.” Quart. Journ. Micr. Sci. n. s. vol. xxxix. pp. 427-588. _ 25. Winee (H.).—“ Jordfundne og nulevende Gnavere fra Lagoa Santa.” HE Museo Lundi, Forste Bind. Kjobenhavn, 1888. 26. Woopwarp (A. Smith).—“ Outlines of Vertebrate Palzon- tology.” Cambridge, 1898. 27. Woopwarp (M. F.).-—“ On the Dentition of the Rodentia, with a description of a Vestigial Milk-Incisor in the Mouse.” Anat. Anz., Bd. ix. pp. 619-631. 28. Woopwarp (M. F.).—“ On the Teeth of certain Insecti- vora.” Proc. Zool. Soc. 1896, pp. 557-594. EXPLANATION OF PLATE 26. Fig. 1. Drawing of wax model of Left Upper Deciduous Molar. Postero- internal view. Stage IT. 2. Drawing of wax model showing crown-surface of Left Upper Succes- sional Molar. Stage IIL. 3. Drawing of wax model of Left Third Upper Cheek-tooth. Posterior view. Stage II. 290 MR. A. O. WALKER ON THE Fig. 4. Drawing of wax model of Left Upper Deciduous Molar. Postero- internal view. Stage III. 5. Section through Left Upper Permanent Incisor, showing Labial down- growth of Dental Lamina. Stage I. 6. Section through Successional Molar. Stage ITI. 7. Orown-surfaces of the last two Upper Molars of A. Dolichotis platycephalica (after Ameghino). B. Cavia cobaya. Both much enlarged. 8. Crown-surface of Upper Molar of A. Stichomys constans. Inferior Eocene (after Ameghino). B. Spaniomys riparius. 5 ob 3: 55 C. Hystrix leucura. Recent. (Camb. Univ. Zool. Mus.) 9. Section through Deciduous and Successional Molars of Cavia, showing “concentric epithelial body ” in connection with the former. Stage IT. 10. Section through a Posterior Molar, showing lateral position of the root. Contributions to the Malacostracan Fauna of the Mediterranean. By AtFrep O. Waker, F.L.S. [Read 7th March, 1901.] (PLATE 27.) Tue following results of a short stay at Cannes and Hyéres are interesting as showing what may be done in a few hours’ dredging from an open boat, in depths never exceeding 35 fath., and with the simplest apparatus. This consisted of a tow-net of tiffany (such as is used by gardeners for shading greenhouses), strengthened at the bag end by cheese-cloth sewn over it for about 2 ft. in length, and attached to a cane rim 6 or 8 in. in diameter. The cane is important, as the net should be as light as possible so as not to scoop up the sand, in which case it fills up immediately. This net is attached to a stone heavy enough to remain on the bottom while the boat is rowed rather quickly ; the distance of the net from the stone varying from 3 feet on coarse sand to 6 feet or more on mud. The stone stirs up the Crustacea, which find their way into the net with a certain amount of sand, though far less than in the case of a dredge (however light), or metal-rimmed tow-net. This, with two small buckets such as are used by children at the sea-side, a small muslin-bag attached to a brass rim with a brass grating on the top, two or three glass jars (e.g. French-plum jars), 75 fathoms of line, and plenty of tubes large and small, constitute the MALACOSTRACAN FAUNA OF THE MEDITERRANEAN. 291 entire apparatus required. A basket 18 inches by 12 inches, by 10 inches deep, contained the whole. An open boat rowed by one man was used. I dredged five times at Cannes and once at Hyéres, for about 2 hours each time. The total numbers of species in each division of the Malaco- straca were as follows :— Rodoplatmallmatay tesss.ucsscca-s-coseeessceadcce cen votes 10 Schizopoda, including 1 new species .. ... ......... 8 Winmmacearernnrs rere ena Mi Mas OM woe Gan aber vces 9 Tsopoda exclusive of Chelifera, 2 new species ...... 9 Amphipoda, including 2 new species and 2 not previously recorded in the Mediterranean ... 41 AAO POM ee Mrr nets eeeeeeR aol cy lase ce ave don wales il 78 IT need hardly say that other classes of marine animals were also brought up, especially Nudibranchs, some of which were most beautifully coloured. One of these was ultramarine-blue with 2 or 3 longitudinal white stripes; another had the inside of the wavy mantle dark green, while the rest of the animal was white with scarlet spots. The following were the stations :— Cannes. 1. Feb. 4. Lan. Soc. Journ. Zoor.. Vou. XXVILPL 29. a nore oeaee — Govest di ae é ° West, Newman imp A Samnend Jith. . E BRITISH FRESHWATER RHIZOPODA Linn. Soc.Journ. Zoon. Vou. AXVIIL PL. 30. West,Newman imp. BRITISH FRESHWATER RHIZOPODA. ON THE PALATE OF THE NEOGNATH®. 343 Some Points in the Morphology of the Palate of the Neognathe. By W. P. Pycrart, A.L.S., F.Z.S. [Read 2nd May, 1901.] (Plates 31 & 32.) Iy a recent memoir on the Struthious birds (Trans. Zool. Soc 1900, vol. xv.), I dealt at some considerable length with the arrangement of the bones of the palate, and instituted some comparisons between this arrangement and that prevailing amongst the “ Carinate.” In this same memoir I proposed to adopt the characters of the palatal bones as a more convenient and more exact distinguishing feature than the characters of the sternum, which have done duty ever since their introduction by Merrem in 1813. In accordance with the characters of the palate, the Tinamous are to be divorced from the “ Carinate”’ with which they are generally associated, and placed with the Struthious birds. The Struthious birds and Tinamous constitute a group by themselves—the Paleognathe ; whilst the remaining forms made up a second group—the Neognathe. The contention that the Struthious (Paleognathine) palate is or a more ancient type than the Neognathine is admitted by all. The typical Paleognathine palate is that of Dromeus; and as this is to serve as a standard of comparison with the palatal bones now to be discussed, we may briefly enumerate its salient features. The vomer (Pl. 31. fig. 1) is of great size, flattened dorso- ventrally, extending far forwards as a broad, median, grooved plate to beyond the middle of the beak; and backwards, to terminate in a pair of rami beneath the pterygoids, with which they ultimately fuse. The palatines are comparatively short, flattened, and slightly twisted bones. They run forwards, tapering as they go, to terminate in a slender rod, closely approximated to the mesial border of the very large maxillo- palatine process: they extend backwards to the level of the orbital process of the quadrate. The proximal half of the palatine (pa.) is relatively broad, and tts mesial border 7s applied to the outer border of the fused vomero-pterygoid bar. To this vomero-pterygoid bar and its relation to the palatine, attention is now specially directed. At the same time, it would be well also to carefully examine the form of the palatines and 344 MR. W. P. PYCRAFT ON THE MORPHOLOGY their relation one to another, and to the maxillo-palatine process, in the accompanying figures. In the palate of Rhea (Pl. 31. fig. 2) the vomer is relatively smaller than in Dromeus. It is deeply cleft anteriorly, thereby exposing the parasphenoidal rostrum, which is concealed in Dromeus when the skull is viewed ventrally, and posteriorly it terminates in a pair of broad “feet” closely approximated. Furthermore, the vomer is peculiar in that it is trough-shaped, the parasphenoid being received into the trough. The feet of the vomer, if further examined from the ventral aspect of the skull, will be found to be partly concealed by the palatines. Thus the primitive position of the palatines, which obtains in Dromeus, has in Rhea undergone a change. They have moved, from their original position outside the vomero-pterygoid bar, towards the middle line, and in doing so have come to underlie this bar. This movement of the palatines inwards, and their connection with the distal end of the pterygoid, has played a very important part in the evolution of the avian palate. We may trace the early stages in this movement of the palatines in the skull of the Tinamous. The palate of the Tinamous, as I have already pointed out, closely resembles that of Rhea. This close resemblance being undeniable, we need only concern ourselves here with the differences between the two skulls. The most important of these differences for the present connection are concerned with the vomer, palatine, and pterygoid. The pterygoid, in the Tinamous, appears to be relatively longer than in Rhea. This appearance is deceptive, and is really due to the fact that the vomer and palatines have shifted considerably forwards, whereby the vomer terminates much further forward along the parasphenoidal rostrum—so much so, that the choane (which are narrower than in hea) le almost entirely in front of, imstead of almost entirely behind the ant- orbital plate (prefrontal), terminating immediately beneath the bony style (lachrymo-nasal pillar), taking the place of the maxillary process of the nasal. In other words, the choanze extend forwards as far as the anterior limit of the lachrymo- nasal fossa. The palatines of the Tinamous have become more rod-shaped. OF THE PALATE OF THE NEOGNATHS. 345 Seen from the ventral aspect of the skull, the proximal end of each, instead of expanding into a broad plate to underlie the base of the vomer, and extending backwards in a spike-shaped fashion along the pterygoid to within a short distance of the basipterygoid process, as in Rhea,—on the contrary, cross the pterygo-vomerine bar obliquely, as a narrow flattened lamina, attached merely by its mesial border. «che palatine of the Tinamou, then, is confined to the distal end of the pterygoid, instead of extending back to within a short distance of the basipterygoid process as in hea. The vomer, as we have already remarked, has shifted relatively further forwards; it also appears to have undergone a slight relative reduction in size. The most important features in all this are: (1) the forward shifting of the palatine to the distal end of the pterygoid; and (2) the inward shifting towards the middle line. Compare figs. 1-5, pl. 31. But it would be well to pause for a few moments to survey briefly one or two other features of the Tinamine palate before passing on. Compared with that of Rhea, it will be remarked at once that in the Tinamous the maxillo-palatine processes have undergone a great reduction, though they are still of considerable size. In Rhea, these processes extend inwards and backwards, in the form of a pair of broad, plate-like, more or less fenestrated lamine. The postero-internal angle of the plate is continued backwards in the form of a long and delicate rod, closely approximated to the outer border of the palatine, and terminating beyond the middle of that bone. The great size of the maxillo- palatine process restricts the forward extension of the quadrato- jugal fossa to the level of the antorbital plate. In the Tinamous the maxillo-palatine processes—though, as we have just remarked, of considerable size—are relatively much smaller than in Rhea. The backward extension (the palatine process) may be described as having the form of a narrow, concavo-convex band tapering to a point, and terminating at about the middle of the outer border of the palatine, which it supports as in Rhea. The furthest point of this maxillo- palatine process lies immediately behind and below the antorbital plate. This reduction of the maxillo-palatine process has extended the length of the quadrato-jugal fossa, which now 346 MR. We P. PYCRAFT ON THE MORPHOLOGY reaches as far forward as the anterior limit of the choanz, and terminates immediately below the level of the anterior limit of the lachrymo-nasal fossa. The anterior end of the palatine in 2hea is widely separated from the palatal process of the premaxilla, while in the Tinamous the anterior end of the palatine just touches the free end of the palatal process of this. This connection is brought about, partly by the forward shifting of the palatine, and partly by the narrowing of the beak, consequent upon the lateral reduction of the maxillo-palatine process. We are now ina position to summarize the facts herein set down, and to select therefrom such as directly illustrate the transition from the Paleognathine to the Neognathine palate, which is the object of this paper. The most primitive arrangement of the Avian palatal bones is to be found in the skull of Dromeus. The vomer is here of great size, terminating posteriorly in a pair of rami, continued directly backwards beneath, and fusing with, the pterygoids. The palatines are connected, caudad, with the outer border of this vomero-pterygoid bar, and are widely separated one from another. Anteriorly the palatines are connected solely w the maxillo- palatine processes. In Rhea the vomer is relatively smaller, and the paired extremities are closely approximated, but are continued backwards as in Dromeus, beneath the pterygoids. The palatines have shifted inwards, losing their original connection with the outer border of the vomero-pterygoid bar, and, taking up a new position beneath this bar, have formed therewith a squamous suture. The approximation of the rami of the vomer towards the middle line has brought the distal end of the pterygoid into relation with the parasphenoidal rostrum. The distal extremity of the palatine is far removed from the palatal process of the pre- maxilla, and is connected wholly with the mesial border of the maxillo-palatine process. In the Tinamous, the vomer is still relatively further reduced, and does not embrace the parasphenoidal rostrum so completely as in Rhea. The relations between vomer and pterygoid are much asin Rhea. The palatines bound the feet of the vomer externally, and will be found to be connected with the pterygoid by an oblique and scarcely perceptible suture. By the reduction in OF THE PALATE OF THE NEOGNATH SA. 347 the size of the maxillo-palatine processes, they begin to come into contact with the palatal process of the premaxilla: the one touching the other by the tip only. The shifting forward of the palatine and vomer is a feature of great importance, as thereby an approach is made towards the Neognathine palate. The approximation of the palatines to the palatal process of the premaxilla is another Neognathine feature. Some confusion seems to exist, even now, as to the nature of the palate in the Struthious birds. Thus, in so recent and authoritative a work as the‘ Dictionary of Birds’ (article Skull) the palates of Struthio, Apteryx, and the Crypturi are said to be Schizognathous, whilst the palate of Dromeus is described as Desmognathous. Now Huxley, who introduced these terms, added yet another—Dromeognathous, for the special purpose of expressing the fact that the palate of certain “ Carinate ” birds— the Crypturi—was Struthious in type, and could not therefore be included amongst his Schizognathous forms. The palate of the Paleognathz might be described as Desmo- gnathous; but certainly there are no members of this group in which it is Schizognathous. It would be better to adopt the term Huxley coined for the Crypturi—Dromzognathous. This form of palate is not Desmognathous in the sense in which Huxley used this term. In the Dromeognathous palate the palatines never meet one another caudad, in the middle line, and never overlap the palatal process of the premaxilla anteriorly. The pterygoid is never segmented, and consequently is never free, but is immovably united with the vomer, vomer and palatines, or palatines only—as in Struthio. We may now pass on to consider the peculiarities of the Neognathine palate, and the changes which it undergoes within this group. This, indeed, is the avowed purpose of the present contribution. At the same time, so far as is possible, we shall attempt to show how the Neognathine has arisen out of the older Paleognathine form. The Neognathine form may be briefly characterized as that in which the palatines meet one another in the middle line, caudad. If, for working purposes, we confine our description to adult skulls, we might define the Neognathine skull, in all but a few cases to be dealt with presently, as that in which the 348 MR. W. P. PYCRAFT ON THE MORPHOLOGY palatines meet one another in the middle line caudad, and support the vomer between them, whilst the pterygoids join the palatines not by suture, but by a true joint. The Schizognathous skull of any adult Gull or Plover will admirably illustrate the differences between the Paleo- and Neognathez. The changed form and relations of the palatines are here almost diagrammatically emphasized. Anteriorly, they are seen to be quite independent of the maxillo-palatine processes, passing below them, and forwards, to fuse with the palatine processes of the premaxille. Behind, they touch one another and join the long pterygoids by a joint. On both ventral and dorsal aspects strong keels have been developed. The inner ventral keels have grown downwards so as to enclose the vomer in a deep, cavern-like hollow. The vomer itself is seen to be held in position by the embrace of the mesial dorsal border of the palatines. In size it is now relatively greatly reduced, but has developed a strong blade-like keel * passing backwards into a pair of rami attached to the dorsal border of the palatines as we have just indicated. The maxillo-palatine processes are not unlike those of the Tinamous, being shell-like scrolls of bone. They do not, however, extend so far backwards, afford support to the palatines, nor embrace the vomer. The pterygoids are long, rod-shaped, and articulate with the palatines by a true joint. Basipterygoid processes for the support of the pterygoids, caudad, have been dispensed with. The pterygoid, in some Lari, is keeled dorsally ; and this keel increases in height from before backwards, so that immediately behind the pterygo-palatine articulation it has attained a con- siderable height, rising to embrace the parasphenoidal rostrum on either side. It is possible that modifications in the form of the pterygoid, to be discussed presently, may be traceable to the ex- cessive development of this terminal portion of the pterygoid keel. The adult skull in the Neognathe, it has just been remarked, differs from that of the Paleognathe, amongst other things, in that the pterygoid is a free bone, articulating at the one end with the quadrate, at the other with the palatine. This being so, it follows that the relations between the vomer * The yomer in many forms, e. g. Cariama, is represented only by a blade- like lamina; in others this is reduced to a mere spicule, ¢.g. some Galli whilst in many forms it is entirely wanting. OF THE PALATE OF THE NEOGNATHE. 349 and pterygoid must be quite other than those which obtain in the Paleognathe. A comparison of adult skulls will show that this is the case: that the palate in the two forms is quite different in this respect, the vomer in the Neognathous palate being supported by the palatines. But the skull of the young bird throws quite a different light upon the nature of the relations between pterygoid, vomer, and palatine. Care must be taken, however, to select favourable types for study, specialization having, in the skulls of many “forms, obliterated more or less completely the evidence for the facts which follow. If the nestling skull, preferably of some Schizognathous form, such as of the Lari, Charadrii, Otide, Sphenisci, or Colymbi, be examined, the pterygoid will be found to be continued forward into a sharp point, which either slightly overlaps or just touches the bifid end of the azygos vomer. That is to say, the right and left ptery goids are connected with the right and left limbs of an originally paired vomer—as in the Paleognathe. ‘The palatines, which, as we have already remarked, have moved inwards to meet one another in the middle line, underlie the distal ends of these pointed pterygoids. Immediately behind the palatines the pterygoids segment, the segmentation at first resembling a fracture, but later this fracture becomes transformed into a true joint. By this time the terminal ends of the segmented ptery- goids have become perfectly ossified, and simultaneously have begun to effect a union with the underlying palatines, the dis- tinction between pterygoid and palatine being marked by a fine suture. Eventually all trace of the suture disappears, and with it the evidence of the pterygo-vomerine connection. The exist- ence of the distal end of this segmented pterygoid is entirely obliterated, so that there is nothing to show that this joint is not a true articulation between two distinct bones—pterygoid and palatine. In other words, there is no indication of the fact that this joint is formed by segmentation of the pterygoid, and the fusion of its segmented portion with the palatine to form a palato-pterygoid articulation. In some skulls the palatine ex- tends backwards below the segmented portion of the pterygoid to join with it in forming the articulation (Pl. 32. fig. 2). This segmented portion of the pterygoid I have elsewhere called the hemipterygoid; it is the mesopterygoid of W. K. Parker. 350 MR. W. P. PYCRAFT ON THE MORPHOLOGY This hemipterygoid varies greatly in its relative size in dif- ferent groups. In some it has become so greatly reduced that it ceases to segment off from the main body, and remains as a kind of peg projecting from the antero-dorsal angle of the pterygoid trunk. ‘This is the case in the Galli and Anseres. In these groups the vomer, when present, is therefore supported entirely by the palatines. In the Falconide, amongst the Accipitres, again it has entirely disappeared, the support of the vomer being undertaken by the palatines. As a consequence, then, of the inward movement of the palatines, the hemipterygoid element is slowly undergoing sup- pression. In some cases, as we have just remarked, only the merest vestige remains. In the majority of cases its connection with the vomer is but of the slightest. The support of this element has practically been transferred to the palatines. The nature of this support is seen with almost diagrammatic clearness in the Penguins. Herein the palatines, caudad, are plate-shaped. The mesial border of each runs at first beneath the hemipterygoid, then beyond this upward and forward for a considerable distance. The whole of this region beyond the hemipterygoid is applied to the dorsal border of the vomer, which only just reaches back to the tip of the hemipterygoid. Numerous stages in the decay of the disappearing hemi- pterygoid are to be found. The skull of an immature Tetra- pteryx paradisea in the British Museum collection affords an admirable object-lesson in this degeneracy. Here (PI. 82. fig. 4) the free end of the bemipterygoid failsto reach the vomer, which is now entirely supported by the palatine in the manner just described. The vomer, like the hemipterygoid, is also in many cases completely suppressed. Attention must now be directed to the palatal bones of certain Coraciomorphe (Gadow). Lack of suitable material (in the shape of embryos or nestlings) has greatly hampered me in the investigation of these groups ; but enough has come to light to enable me to deal therewith in the present contribution. It is to be hoped that help in this matter will come to hand shortly. We should be very grateful at the Natural History Museum for ripe embryos and nestlings OF THE PALATE OF THE NEOGNATH A. 351 to make up the blanks, which are many, in the collections. When this has been done, one or two very interesting points can be definitely settled. The Cuculi and Psittaci may be dismissed in a few words. The former have the typical Neognathine palate; the pterygoid being segmented, and the bhemipterygoid fusing with the pala- tines, and forming a joint with the main body of the pterygoid. The pterygoid in the Psittaci also forms a joint at its distal end, but nothing is yet known concerning the presence or absence of the hemipterygoid—we do not know whether it is present and fuses with the palatine, or has been lost by atrophy as in the Galli or Anseres. Certain of the Coraciiformes and Passeriformes afford us some interesting modifications of the type. Briefly, these modifications seem to show that the forms in question differ from all the remaining Neognathe in that the pterygoid does not segment, but is continued forward directly on to the vomer when present, as in Paleognathz. That is to say, the hemipterygoid element has not been lost by atrophy, but remains permanently in connection with the main pterygoid body. Between this and the normal Neognathine type are many gradations. The most extreme forms of this modification are perhaps to be found in the Capitonidz and Bucconide. In Megalema marshallorum, one of the Capitonide (Pl. 32. fig. 7) the pterygoid is continued directly forward on to the vomer, terminating in a sharp point running obliquely over the dorsal border of its right and left limbs. This connection between the vomer and an unsegmented pterygoid is of course a Paleognathine character. Moreover, the palatine, as will be seen in the figure (pa.), is confined to the ventral border of the pterygoid, and in no way comes into relation with the vomer. It is significant, how- ever, that in this skull the region of the pterygoid shaft that corresponds to the hemipterygoid is not a direct continuation of the shaft, but a curved plate rising somewhat suddenly from the distal end of the pterygoid shaft iramediately above the free end of the palatine. This latter fits into the curved hemipterygoid border and abuts against the antero-ventral extremity of the pterygoid (fig. 7). On this account, from the ventral aspect of the skull, the pterygo-palatine connection appears to be by means of a joint, as in all the other Neognathe. This joint is 32 MR. W. P. PYCRAFT ON THE MORPHOLOGY continued upwards and forwards beneath the hemipterygoid for some distanee. All this seems to imply that this unsegmented pterygoid is really not a primitive but a secondary character— an approximation to the original type. The suggestion that the unsegmented pterygoid of the species described above is a secondary and not a primitive character, is confirmed by what obtains in Cyanops asiatica. Here the form of the pterygoid, its hemipterygoid plate, and its relations to the vomer, are precisely similar to what obtains in JZ. marshall- orum, but the hemipterygoid is cut off from the main shaft, and fuses by its distal end with the palatine. Except that the fusion of the hemipterygoid with the palatine is not so complete as usual, this is a perfectly normal Neognathine palate. There is one particular, however, in which the hemipterygoids of these two species differs from the normal type, and that is their relatively greater size and close approximation to the parasphe- noidal rostrum, which is held by them in close embrace. Another well-marked type of palatal modification is afforded by the Passeriformes. Foreshadowings of this occur in the Pici, and many modifications thereof occur amongst the Passeriformes. The evolution of these modifications I propose to deal with in a further contribution to this subject, wherein the morphology of the palate in the whole of the Coraciomorphe will be, as com- pletely as possible, set forth. In this Passerine type, which has perhaps reached the high- water mark of specialization in the Corvide, the hemipterygoid (P1. 32. fig. 6) is split off, not by transverse fracture, but by a very oblique segmentation extending from the ventral border of the distal end of the shaft forwards and upwards. ‘The vomer is in contact with the distal end of this reduced hemipterygoid. The main shaft of the pterygoid, immediately behind the hemiptery- goid, expands into a slipper-shaped plate, which is closely applied to the parasphenoidal rostrum on either side. The palatines run along beneath the hemipterygoid, but instead of terminating at the proximal end of this segment, run backwards to articulate with so much of the ventral border of the distal end of the shaft of the pterygoid as is applied to the parasphenoidal rostrum. From the ventral aspect of the skull the proximal ends of the palatines appear to ie in a groove hollowed out of the ventral border of the parasphenoidal pterygoid plate. OF THE PALATE OF THE NEOGNATH A. 353 The peculiarly modified distal extremity of the free pterygoid the adult skull and the connection of this extremity with the fused hemipterygoid and palatine are features of great interest. The resultant oblique joint appears to be one which allows of but little motion. A comparison of the figures will make the peculiarities of this palate easier to understand. Jt has been suggested (p. 348) that the dorsal keel of the pterygoid in the Lari might have some significance. We would remark here that it is possibly by the segmentation of a strongly keeled pterygoid that the large plate-like hemipterygoid of such forms as Megalema, for instance, ay have been derived. Later on in development the keel of the shaft of the pterygoid may have been lost. Before closing this paper, I would draw attention to the modi- fications of the pterygoid in certain of the Caprimulgi. Steatornis affords a most perfect illustration of the segmenta- tion of the pterygoid. In a nestling in the Museum Collection, this bone (PI. 32. fig. 5) is continued forward as an unbroken and completely ossified rod to terminate in a sharp point above a ves- tigial vomer. The palatines have met together mesially beneath these pointed pterygoid extremities, which as yet remain one with the main shaft. In the adult (fig. 5 a) segmentation has taken place, not immediately behind, but some distance distad of, the extreme posterior ends of the palatines. The hemipterygoid fusing with the palatine, an oblique palato-pterygoid joint is formed (cf. fig. 5 6). The other Caprimulgine forms to which allusion has been made are mentioned here, not on account of the hemipterygoid, about which I can at present say nothing, but because of the peculiar modification which the palato-pterygoid articulation undergoes. In Caprimulgus europeus the pterygo-palatine articulation is a perfectly normal (Neognathine) one; the pterygoid shaft articu- lating by a joint with the extremity of the palatine. In Eurostopus nigripennis the pterygoid articulation is as in Capri- mulgus, but the palatines send backwards on to the pterygoids two minute processes, one on either side. In Nyctibius this backward extension of the palatines has encroached still further upon the pterygoids so as to underfloor these for a considerable extent, thus entirely masking the nature of the pterygo-palatine articulation from the ventral aspect of the palate. a4 MR. W. P. PYCRAFT ON THE MORPHOLOGY In Podargus humeralis the pterygo-palatine articulation is so oblique, that at first the pterygoid appears to be an unsegmented bone as in the Paleognathe. The peculiar form of the pterygoid in Podargus at first much disconcerted me, seeming, as it did, to show that the unseg- mented, pointed, pterygoid was not alone peculiar to the Palszo- gnathe. A more careful study, however, has placed its real nature beyond all possibility of doubt—it has modified a cup- shaped articular surface into an elongated facet. The palatines in Podargus, as in Steatornis, are peculiar in that they send inwards a ventral keel to meet in the middle line, thus forming a tubular passage. In Steatornis the floor of this tube lies further forward than its roof, underlying the vommer. In Podargus the floor of the tube lies directly under its roof. Summary. Briefly, the result of this paper has been to show that the differences between the Palewo- and Neognathine palate are those of degree and not of kind. The Paleognathine is undoubtedly the older form. In it the vomer and pterygoid are uninterruptedly connected, one with another, throughout life; whilst the palatines remain per- manently separated one from another caudad, and are connected only with the maxillo-palatine processes distad. In the Neognathz the vomero-pterygoid relations are inter- rupted by the segmentation of the pterygoid distally ; whilst the palatines, caudad, have moved inwards to meet in the mid- ventral line beneath the distal ends of the pterygoids, with which they eventually fuse, and, distad, have lost their primitive con- nection with the maxillo-palatine processes, and have established a new connection with the palatine processes of the premaxilla. The fusion of the distal ends of the pterygoid with the under- lying palatine is accompanied by segmentation of the former and the formation of a pterygo-palatine joint. In my recent memoir on the Paleognathe I inadvertently described the inward movement of the palatines as having resulted in “thrusting the vomer forwards” (p. 206). This is inexplicable, since it is obvious, from the very next sentence, that T had not lost sight of the fact that this bone still retains its primitive connection with the pterygoid. That both pterygoid OF THE PALATE OF THE NEOGNATOR. 305 and vomer, however, have been affected by this movement there can be no doubt, since, as I have pointed out, in many cases it has brought about the suppression or atrophy of the hemiptery- goid element, and has assumed the functions thereof by taking up the support of the vomer, as in the Anseres and Falconide. The most primitive form of Avian palate is most certainly the Dromzognathous, not, as has been stated, the Schizognathous. The Schizognathous and A®githognathous palates’ are both specialized forms derived by modification of the Dromeognathous type. The Desmognathous palate is a highly specialized con- dition which appears to have arisen independently amongst both Schizo- and Aigithognathous forms. The Neognathine palate is undoubtedly undergoing a further change, a change resulting in the transference of the support of the vomer from the pterygoid to the palatine. This has followed upon the movement of the palatines from the original position outside the pterygo-vomerine bar to a position beneath this. But the disturbance does not end here, for it is in the most highly specialized forms accompanied by the degeneration of the distal end of the pterygoid and the suppression of the vomer. EXPLANATION OF THE PLATES. Puate 31. Fig. 1. Ventral view of the skull of Dromeus nove-hollandia, showing the most pvimitive arrangement of the palatal bones among living birds. The vomer is of great size, and extends backwards in the form of a pair of broad limbs beneath the laminate, pointed pterygoid. The palatines are connected by suture, posteriorly with the external lateral vomero- pterygoid border, anteriorly with the maxillo-palatine process. Fig. 2. The palate of Rhea americana. The vomer has relatively decreased in size. Its relations with the pterygoid are much the same as in Dromeus; but this fact is masked by the palatines, which have- moved inwards beneath the pterygo-vomerine articulation so as to approach one another in the middle line. The palatine, as in Dromeus, is connected by suture with the maxillo-palatine process. anteriorly. It is interesting to note that the posterior narial aperture of hea has been largely filled up by the inward and backward ex- tension of the maxillo-palatine precess. In Dromeus this aperture is very large. ‘The premaxillary processes of thea are also very large. Fig. 3. The palate of Nothoprocta perdicarius. Compared with Rhea, it will be seen that the vomer is, relatively, still further reduced, and that the palatines have moved still further inwards beneath the- LINN. JOURN.—ZOOLOGY, VOL. XXVIII. 25 306 MR. W. P. PYCRAFT ON THE MORPHOLOGY pterygo-vomerine articulation. Distally, the palatines are seen to have come into relation with the maxillo-palatine processes, inasmuch as they just touch their hindmost extremity. The quadrato-jugal (inferior temporal) fossa has greatly increased in length. Fig. 4. The palate of Rissa tridactyla showing the typical Neognathine palate. The inward movement of the palatines has reached its maxi- mum, meeting one another in the middle line, beneath the pterygoid and vomer. Following upon this, the distal end of the pterygoid has become divorced from the main body, to form the hemipterygoid (fig. 1, Pl. 32). Later, the latter fuses with the palatine; and at the point of fracture, immediately caudad of the palatine, a joint is formed. Thus, in the adult Neognathx by the disappearance of the hemi- pterygoid element the pterygoid appears to bea free bone, articu- lating with the palatine, instead of being connected therewith by squamous suture. The vomer in the adult skulls of this type appears now to be completely divorced from all association with the pterygoid. Fig. 5. The palate of the Common Fowl (Gallus bankiva var. domestica). The bones in this palate have undergone still further specialization | The hemipterygoid appears to be totally suppressed, so that the vomer is actually supported only by the palatine (see also Zetrapterye, fig. 4 a, P1.32). The slight groove indicated in the figure immediately caudad of the vomer was, in the freshly prepared skull, filled by two threads of cartilage running backwards from the vomer, and indi- cating its sometime further backward extension, wedged in between the palatines and articulating with the now suppressed hemi- pterygoid. Puate 32. Fi ee) .1. The pterygoid of a nestling Podiceps cristatus, lateral view ; showing the still distinct hemipterygoid element which extends forwards to the vomer. Later the hemipterygoid, losing itself by fusion with the palatine, gives the appearance, in the adult, of a true palato-pterygoid articulation, thereby making it appear that the vomer in the Neo- enathe is unconnected with the pterygoid, and thus, on this account, sharply distinguishing the Neo- from the Palzognathex. g. 2. The pterygoid of a nestling Oceanodroma leucorrhoa, lateral view. The palatine has extended backwards beneath the hemipterygoid to share in the articulation with the main shaft of the pterygoid. g. 3. The pterygoid of a nestling Pygoscelis teniata, lateral view. At this stage the hemipterygoid appears as if wedged into the distal end of the main shaft, as by fracture; later a perfect glenoid cavity is deve- loped between the distal end of main shaft and the hemipterygoid element. Fig. 3a. The dorsal aspect of fig. 3, showing an early stage in the decline of the hemipterygoid, which just fails to reach the vomer. BR uw = 9 a ee vss. West,Newman imp, ‘Lins. Soc. Journ. Zoor. Vou. XXVIILPIL. 31. Fyeraft. HGréoved th. wi EO EE NO GINAMT ELA, Imm. Soc. Jourw. Zoor, Vor XXVIIT Pl 32. West,Newman. imp. PALATE OF THE NEOCNATHE Fig. un) te g ig. 4a. &. 5d. OF THE PALATE OF THE NEOGNATH A. By5)// g. 4. Lateral view of the pterygoid of Tetrapteryx paradisea. The hemi- pterygoid and palatine bear the same relation to the main body of the pterygoid as in Oceanodroma (fig. 2). In this figure, which represents the external lateral view of fig. 4, the hemipterygoid element is seen to be more degenerate than in Pygoscelis (fig. 3a), and the vomer is in consequence supported entirely by the palatines. . 5. Lateral view of the pterygoid of a nestling of Steatornis caripensis, wherein the hemipterygoid element has not yet segmented off from the main shaft. . The lateral view of the pterygoid of an adult Steatornis caripensis , The distal end of the pterygoid has now segmented off to form the hemipterygoid. It is connected with the main shaft by a sigmoid articulation, but remains traceable throughout life by reason of the fact that its distal end projects above the palatine. Ventral aspect of fig. 5a, showing the transverse articulation with the palatine and fused hemipterygoid element. . 6. Lateral view of the pterygoid of the Rook (Corvus frugilegus). The hemipterygoid just reaches the vomer. Later, on its fusion with the palatine, the articulation with the main body of the pterygoid is oblique, not transverse as in the majority of Neognathe. 7. Lateraljview of the pterygoid of Mega/ema marshallorun. Compared with fig. 6 it will be seen that in Megalema the pterygoid has reverted to the original, Paleognathine, unsegmented condition, the hemipterygoid being continuous with the main shaft and extending forward to support the vomer, which is entirely free from the palatine. , . 8. Lateral view of the"pterygoid of Bucco Dysoni. The pterygoid, as in fig. 7, is unsegmented. The vomer is vestigial or wanting. By further specialization the palatine has almost completely fused with the pterygoid, only a slight cleft marking the distinction between the two. Fig."8 a. Ventral view of fig, 8, showing the last traces of an originally trans- yerse palato-pterygoid articulation. ExpnanatTion oF Lerrers. h.pt.=hemipterygoid. inf.t.foss.=inferior temporal fossa (quadrato-jugal fossa). m2x.p.=maxillo-palatine process. pa.=palatine. par,.=parasphenoidal rostrum. p.p mx.=palatine process of premaxillary. pt.=pterygoid. vo. =vomer. 308 MR. S. PACE ON THE On the Corallum of Turbinaria. By 8. Pacr, F.Z.S. (Com- municated by H. M. Brernarp, M.A., F.L.S.) [Read 18th April, 1901.] The Formation of the Cup. No actual observations upon the early stages in the growth of the remarkable cup-shaped corallum of Turbinaria appear to have been yet recorded. It has usually been assumed that the parent polyp becomes submerged in the common ccenenchyma of the coral, and that its calicle is not recognizable after the corallum has attained the cup form. Thus Mr. H. M. Bernard, an admitted authority on this group, writes *:—“ A ring of buds shoots up round and from the sides of the parent polyp, together forming a cup, the wall of each bud rising up as a distinct cone above the level of the fusion of their walls to form the common ceenenchyma. The parent polyp dies away, and its primitive protuberant cone is immersed under the conenchyma formed from the fusion of the walls of a ring of daughters. These daughters carry on the colony, the budding of the daughters being limited to their free or outer sides, z. e. to the sides turned away from the axis of the cup.” To illustrate his com- parison with what occurs in the case of Madrepora, Mr. Bernard gives the two diagramst which are copied in figs. 1&2. The supposed dying away of the parent polyp in Turbinaria was evidently assumed in order to explain the fact that it is so very unusual to find any trace of a calicle occupying a central position at the base of the cup. An examination of younger growth-stagest than are con- tamed in the British Museum collection, and the dissection of several small cups, have revealed the interesting fact that the parent polyp does not die away, but that it bends over to one side and takes part with its daughters in forming the rim of the cup. In a normal cup the parent calicle can always be traced as one, generally the largest, of the innermost ring of calicles. * Ann. Nat. Hist., ser. 6, vol. xx. (1897), pp. 131-2. + Catal. Madrep. Corals Brit. Mus., vol. ii. London, 1896. + The corals collected by me in Torres Straits I have presented to the British Museum. CORALLUM OF TURBINARIA. 359 The actual process is represented diagrammatically in figs. 3 to 10; and fig. 5 may be taken instead of Mr. Bernard’s diagram i, Ys 7 Ah, AY > <7 QC Ys Q Fig. 1. Diagram of Madrepora showing the relationship of the parent calicle to the corallum, after Bernard, Fig. 2. Erroneous diagram of Turbinaria, after Bernard. Figs. 3-5. Diagrams illustrating the formation of a Turbinarian colony and the fate of the parent calicle. '[Figs. 1-5 are similarly shaded.] Figs. 6-10. Diagrams of successive transverse sections of the stem of a young corallum to illustrate the mode of budding. The calicles are similarly shaded in each diagram. (fig. 2) for comparison with that of Madrepora (fig. 1). It will be seen that the agreement between the two types is even closer than, was suggested by Mr. Bernard; the only difference being that while in Madrepora the parent polyp retains its axial 360 MR. S. PACE ON THE position and grows up in advance of its daughters, so that it is at all stages the terminal polyp of the colony, in Turbinaria the parent polyp bends away from its first daughter-bud and then grows up together with and alongside its daughters, so that eventually it comes to form one of a ring of calicles of which its daughters are the other members. While the central area of the cup is typically wholly ecenen- chymatous, occasional examples are met with in which a calicle does occupy a central position within it; but, as already stated, this is a very unusual occurrence, and it would appear to be the result of a secondary torsion on the part of either the parent or of one of the daughter polyps. I have also noticed in a few specimens the existence of a slight central elevation which rather suggested that a calicle was buried at that point; but as no dissection was made, it may well be that the appearances observed were in reality due to the presence of some commensal or parasite. The Variation of the Corallum. A few words regarding some of the modifications which the Turbinarian cup undergoes with advancing age, and by the direct influence of its environment, may not be out of place, since I have had rather exceptional opportunities * for the ob- servation of corals and their habits, and since the so-called species of Turbinaria have been to so large an extent founded upon what are in reality but acquired characters. Bernard, in his Catalogue of the British Museum Turbinarie, found himself obliged to group them according to the forms ultimately assumed. by the cup; but, inasmuch as it was obvious that many of these might be adaptational or even accidental, he pointed out that his classification was purely morphological, and only to be regarded as a provisional one. It will now be my endeavour to show that the variations of a Turbinarian colony from the primitive cup- shape—the “ crateriform” type of Bernard—can be readily ex- plained by reference to the conditions under which the coral has * During a stay of nearly three years in Torres Straits, while engaged in the investigation of the commercial pearl-shell, some thousands of examples of Turbinaria in all growth-stages have passed through my hands or under my notice. Turbinarians are exceedingly plentiful on the reefs in this region, and! young individuals, as well as large cups, are very commonly found on the- backs of the pearl-shell collected by divers. CORALLUM OF TURBINARIA. 361 grown; though it by no means follows that heredity plays no part in determining the form of growth assumed by the corallum under any particular conditions, and it may well be that the tendency towards one type rather than another is inherited; this, however, can only be established by experiment. As might be expected, the largest and most perfect cups are those formed at depths below the tidal zone, in clear water, and where the growth of the corallum is unrestricted by neighbouring objects *. Above extreme low-water mark there is a greater tendency for the coral to lose its cup-shape, and to become irregular by _the folding and crumpling of its walls and by the adoption of an encrusting habit. Again, specimens are common on the reefs in which certain calicles have budded to form secondary, more or less independent, colonies: subsidiary cups may thus be formed within the parent cup, and some such individuals present a regularly “storied” appearance. In other cases the secondary colonies, instead of forming cups, take on an arbo- rescent growth like that of a Madrepora. This modification f, which I may term the “ madreporiform ” type, is a not uncommon one where the coral is growing at the bottom of a hole in the reef, and where growth in a vertical direction is of obvious advantage to the colony. When a Zurbinaria grows upon a shelf or ledge of rock it generally loses its cup-form; the side turned away from the free edge of the shelf ceases to grow, and the corallum thus becomes a more or less flattened, expanded plate§ overhanging the ledge. When, during growth, the lower surface of the cup comes into contact with the substratum, irregular root-like outgrowths will * At the time of publication of the British Museum Catalogue the largest cup in that collection was stated (with a certain amount of pride) to measure: as much as sixteen inches in diameter; such a specimen is, however, in reality quite a small one compared with the giants occurring on the shelling grounds in Torres Straits. + These daughter cups, the result of proliferation of individual polyps, must not be confounded with the cup-shaped folds of the wall of the parent cup, which are of much more common occurrence. + A very good example of this type of growth is figured by Ortmann as Ee, acai Foal. Jahrb., Syst. vol. iii. pl. vi. fig. 4. § The specimens of 7. reniformis and T. foliosa figured by Berned (Catalogue, pls. xvii. & xviii.) probably owe their form to this cause. 362 MR. S. PACE ON THE generally arise at the pointstof contact; and the specimen may thus acquire an appearance rather suggesting that of the Banyan tree with its numerous false stems. This type of growth has deen described as ZT. radicalis by Bernard*. : Not infrequently a cup becoming accidentally broken from its stalk continues to live in this detached condition ; and, if it has at the same time been inverted, subsidiary cups will often be formed upon its upturned lower surface. In addition to these modifications, due mainly to position, there are others which direct observation on the reef has enabled me to trace to another cause: namely, the danger of becoming silted up or clogged with the fine mud which is always a pro- minent feature on a coral-reef. This is apparently the greatest evil which a coral has to dread, and the structure of the corallum is frequently much modified in such a manner as to adapt it to life on muddy ground. Under such conditions the cup of a Turbinaria is often flattened out, the “ peltate ” type of Bernard, and the colony may even assume a convex form ; or else the cup may be cleft on one side, or perforated at its base, so as to render it impossible for any silt to lodge within it. Where the side of the original cup becomes cleft, one of the lobes thereby formed may extend round the outside of the cup, and, the growth of the other lobe being arrested, the corallum may take on a roughly spiral form, and silting will be obviated by the presence at the base of the corallum of a con- tinuous gutter by which any foreign matter will be carried off. Those forms which Bernard has termed the ‘“ Turbinarie JSrondentes” belong to this type of growtht. A very common method by which a Turbinarian defends itself against silt is that in which at an early stage the margin of the cup, or rather disc, becomes bent down at regular intervals or frilled{, so that with further growth, the details of which are susceptible of various modifications, a very perfect gutter system results. In what * Brit. Mus. Catalogue, pl. ix. | The specimens of 7. awrtcularis and T. calicularis figured by Mr. Bernard (Brit. Mus. Catal., pls. x. & xi.) are poor examples of this type. It is shown in its most perfect form by a specimen of Montipora (in which genus this modifi- Cation is much more common than in Turbinaria) figured by Mr. Saville Kent in his ‘ Naturalist in Australia,’ pl. xxiv. p. 146. { The specimen of 7. peltata figured by Mr. Bernard (Brit. Mus. Catal., pl. vi.) shows this very well. CORALLUM OF TURBINARIA. 363. Mr. Bernard has termed the “ bifrontal” type of growth, the elevated folds become greatly extended and their apposed lower surfaces fuse together all over: a corallum may thus arise which consists of numerous close-set vertical plates bearing polyps: upon each of their faces and connected with each other by but a shght attachment at their bases. The narrow, more or less radial, interspaces of this type are practically open all round and so afford no lodgement for silt*. In Bernard’s ‘“foliate”” and “ mesenteriform ” types the elevated folds, instead of fusing together back to back, persist either as open frills, or, meeting, fuse only along the lines of junction. In this way a corallum. consisting of a series of connected cylinders or cups, open at their bases, and bearing polyps alternately upon their inner and outer faces, may arise. By the suppression of the elevated folds a series of cups, each having, like the parent cup, an internal. polyp-bearing surface, may be formed; while an exceedingly interesting extension of this type of growth is afforded by those eases in which the growth of the depressed folds has been. arrested, so that the corallum has come to consist of a series of. Figs. 11-14. Diagrams to show the origin of various types of growth by the: folding of the margin of the primitive cup or disc. The polyp- bearing surfaces are dotted. * A series of specimens well illustrating this type is figured by Mr, Bernard as T. gracilis, Brit. Mus. Catal., pl. xxiii. 364 MR. S. PACE ON THE cups bearing polyps only upon their outer faces*; in such an example any mud settling within the cup will of course not affect the polyps. The derivation of the above-mentioned types from the primitive cup or disc is illustrated very diagrammatically in figs. 11 to 14 (p. 363). What Bernard has termed the “tabulate” type of growth is certainly, as he suggests, expressive of periodicity in the growth of the colony; and this periodicity appears to be often de- pendent upon the monsoons. The “set’’ of a current over a reef, and consequently also the “lay” of the silt, is in many places markedly different at these seasons; and, with every change in the direction of the drift, those parts of a coral which have been overwhelmed and killed by silt will tend to become again exposed, and may then take on a fresh Jease of life, while the opposite face of the colony may in turn be buried until another change takes place in the set of the current. It is always possible to find some evidence of this periodicity on any large block of Porites, Turbinaria, &e., as it occurs on a reef; and, though of course many other factors besides the monsoons are concerned in effecting those changes which are continually taking place in the set of a marine current, yet such extremely regular alternations, as are expressed by the typically “ tabu- late” type of Turbinarian growth, can only be due to their succession. The “clomerate’’ type, in which the corallum becomes enormously thickened to form the large hemispherical masses + so common on many reefs, would appear to be an adaptation mainly to withstand the battering of the surf and the rush of the tide over the reef. In many localities the strength of the current is so great that a corallum of any other form would most certainly be swept away; and on very exposed situations massive forms of Turbinaria, Porites, and such-like are the only corals met with, * A specimen referred to 7. magna, Bern., in the Saville Kent collection in the British Museum from Shark’s Bay, Western Australia, is a very perfect example of this type of growth, which is also to be seen in some parts of the specimens figured in the British Museum Catalogue, pls. xil., xiil., & xiv. |} Examples of these are contained in the Saville-Kent collection in the British Museum, and many such may ke recognized in the beautiful collotype plates of coral-reefs which illustrate Saville Kent's ‘Great Barrier-Reef.’ CORALLUM OF TURBINARIA. 365 We thus see that many of those features which have been relied upon for the discrimination of “species” in this group are in reality but of secondary value ; that,as with other corals, the characters of a Turbinaria, and more particularly the general form of the colony, are largely influenced by the conditions of its environment. On the other hand, al/ variation among binarians (and the same is equally true of other genera) is certainly not the mere expression of adaptive modification. This is proved by the fact that specimens living side by side, and consequently under exactly the same conditions, so frequently exhibit quite obvious differences of types such variation can only be genetic. I have myself observed, growing upon the same pearl-shell, three large Turbinarian cups which were quite typical examples of what must, in my opinion, be regarded as three distinct and well-marked species. The question of defining the limits of a “species” is in no eroup such an easy one as it appears to the student who works only at the inadequate material represented in our museums* ; and, in the case of the corals, it is a problem of the greatest difficulty. At every turn the zoologist who studies Nature, not merely in the museum or laboratory, but also in the field, is confronted by facts such as those to which I have alluded,—facts which bring him once more face to face with that ever-recurring question, “ What is a species ?”—a question to which no satis- factory answer is as yet forthcoming ; to which, indeed, no satis- factory answer can be looked for until such time as taxonomic research is placed upon a more truly scientific basis—until, in short, the zoological student has at his disposal large series of specimens and other data which have been collected with the express view of aiding the solution of those problems which are summed up in that familiar word “ species.” * See a recent note in ‘ Nature’ (vol. lxiii. pp. 490-1), in which the scientific collecting of zoological material is discussed. CORALUUM OF TURBINARIA. 365 We thus see that many of those features which have been welied upon for the discrimination of “‘ species” in this group are in reality but of secondary value; that, as with other corals, the characters of a Twrbinaria, and more particularly the general form of the colony, are largely influenced by the conditions of its environment. On the other hand, al/ variation among Turbinarians (and the same is equally true of other genera) is certainly uot the mere expression of adaptive modification. This is proved by the fact that specimens living side by side, and consequently under exactly the same conditions, so frequently -exhibit quite obvious differences of type: such variation can only be genetic. I have myself observed, growing upon the same pearl-shell, three large Turbinarian cups which were quite typical examples of what must, in my opinion, be regarded as three distinct and well-marked species. The question of defining the limits of a “species” is in no sroup such an easy one as it appears to the student who works only at the inadequate material represented in our museums* ; and, in the case of the corals, itis a problem of the greatest diftculty. At every turn the zoologist who studies Nature, not merely in the museum or laboratory, but also in the field, iy confronted by facts such as those to which I have alluded,—facts which bring him once more face to face with that ever-recurring question, “ What is a species?”——a question to which no satis- factory answer is as yet forthcoming ; to which, indeed, no satis- factory answer can be looked for until such time as taxonomic research is placed upon a more truly scientific basis—until, in short, the zoological student has at his disposal large series of specimens and other data which have been collected with the -express view of aiding the solution of those problems which are summed up in that familiar word “ species.” * See a recent note in ‘ Nature’ (vol. lxiii. pp. 490-1), in which the scientific -collecting of zoological material is discussed, LINN. JOURN.- ZOOLOGY, VOL. XXVIII. 26 366 MR. C. WARBURTON AND MISS A. L. EMBLETON ON The Life-history of the Black-Currant Gall-mite, Eriophyes (Phytoptis) ridis, Westwood. By Crcim Warsurton, M.A., F.Z.S., Zoologist to the Royal Agricultural Society of England, and Atice L. Empxerton, B.Sc., 1851 Exhibition Science Research Scholar, Associate of the University of Wales (Cardiff College). | [Read 7th November, 1901.] (Piatzss 33 & 34.) Tue Black-Currant plant disease due to Eriophyes ribis first attracted attention in England in the year 1869, though the cause was not at first clearly ascertained. Since that time it bas been the subject of frequent references by economic ento- mologists, who have generally recorded the fact that the pest was on the increase. To this day, however, our knowledge of the life-history of the mite is extremely limited, and the state- ments of yarious observers with regard to it are either too vague and general to be of much practical value, or are absolutely inconsistent and conflicting. No complete account of its life- cycle has yet been attempted, and its methods of distribution have remained a matter of conjecture. The difficulty of the research is, of course, largely due to the minute size of the mite, which rarely exceeds one hundredth of an inch in length*. To observe and record the condition of the mites inside the buds at various seasons of the year is a tolerably simple matter, but a thorough investigation of their habits in- volved watching the creatures throughout their wanderings, and here their small size proved a serious obstacle. In a paper on “ Insects affecting the Orange” T, H. G. Hulbard described certain phenomena with regard to an allied mite which suggested to him that it owed its distribution largely to the aid of various insects and arachnids, and it was the clue thus afforded by him that was immediately responsible for the observations on distribution in the present investigation. * The measurements are:— 9, length ‘23 mm., breadth ‘04 mm. Gy, » L6 » 08 t U.S. Dep. Agric. Ent. 1885. THE LIFE-HISTORY OF THE BLACK-CURRANT GALL-MITE, 367 The Eriophyide. The Eriophyidze (Phytoptide) or gall-mites are vermiform Acari, possessing only two pairs of legs, which have no claws, but are furnished with bristles and “ feather-hairs” (Pl. 33. fic. 8; Pl. 34. figs. 10,11). The elongated body is transversely striated, and terminates in a muscular disc-like organ (Pl. 33. fig. 9; Pl. 34. figs. 12, 13). It also presents certain bristles, of which the most important are a pair proceeding dorso-laterally from above the tail-disc. About two hundred species of gall-mites have been described. They are all of small size, and are vegetable feeders, usually causing the abnormal growths known as galls on the leaves or stems of the plants they infest. The Black-Currant Disease. The disease is easily recognized by the presence, on infested bushes, of swollen and distorted buds, which, if still green, are found on examination to contain large numbers of the parasite. Badly attacked buds are entirely abortive, and eventually remain on the stems as brown dry knobs from which no leaves have arisen (PJ. 33. fig. 1). In milder cases of attack sufficient vigour is retained to give forth an enfeebled shoot (Pl. 33. fig. 2). As the hold of the disease upon the plant increases the effect becomes very striking. The failure of a large number of the buds forces into premature development the buds which would normally burst forth the following year, making overdrafts, so to speak, on the plant’s vitality, and a stage is reached when it is no longer able to respond to the excessive calls made upon it. The provision for next year’s foliage is already exhausted and the plant dies. Life-history of the Mite. The observations here recorded began on May 20, 1901, and were carried on without intermission until the middle of October. On May 20, of last year’s wite-infested buds, ‘those of which the growth had been entirely arrested were in some cases cracked, mites being visible externally in the fissures. Individuals were also found wandering on the stems, Clearly the migration from the abortive buds had only recently com- menced. Mites have been recorded by Newstead and othersjas wandering on the plants much earlier in the spring. It is Die 368 MR. C. WARBURTON AND MISS A. L. EMBLETON ON probable that these early wanderers are nearly all doomed to destruction, but their presence is easily accounted for. Many of the infested buds are not too much injured to put forth leaves however weakly (PI. 33. fig. 2); and if they succeed in doing this, the mites are deprived of their shelter and rendered home- less before the formation of next year’s buds. This compulsory quitting of buds of which the mite has not succeeded in entirely destroying the germinating power may be regarded as more or less fortuitous, and is a very different matter from the definite migration which appears to take place from the wholly abortive buds. As we have seen, this commenced, in 1901, about the middle of May. The activity of the mites attained its maximum about May 30, and practically ceased about the middle of June, by which time the arrested buds were dried up and destitute of lving mites. A very careful investigation of the behaviour of the mites during this migration period revealed some interesting habits which have hitherto escaped observation. Three different methods of locomotion are employed by these creatures in their search after new feeding-grounds. These are: (1) crawling, (2) adherence to passing insects, and (3) leaping. Crawling. The extremely anterior situation of the four short legs would seem to be ill-adapted for locomotion. Nevertheless the mite can crawl along quite actively, at the rate of three or four milli- metres, or twelve to fifteen times its own length, a minute. The motion of the legs is very scrambling and haphazard in appearance, but by wild exertion they drag the unwieldy body forward—an inert mass, trailing in the rear. When, however, the mite desires to change its direction, or to surmount an ob- stacle, the tail apparatus, with its muscular dise and bristles, comes into play. By this apparatus a hold is obtained upon the surface over which the mite is crawling, and the body is swung round, or the anterior portion is reared up and the obstacle surmounted. The tail-bristles are stronger and less wavy than they are represented in most figures of the mite, and appear to be accessory motile organs of no slight importance. | Their position in crawling may be seen in figs. 7, 20. THE LIFE-HISTORY OF THE BLACK-CURRANT GALL-MITE. 369 Distribution by Insects. When the mites were first observed on the outside of the abortive buds, one curious point in their behaviour attracted attention, even under the slight magnification of a pocket-lens. Though some were actively crawling about, others appeared to be standing on end, and motionless, except for the waving of their legs. A series of observations and experiments were undertaken with a view to ascertaining the precise nature and purpose of this phenomenon, with the following results :— A migrating mite, after crawling for a short distance in the manner already described, would obtain a firmer hold upon the surface of the bud with its tail-dise and assume an upright attitude (Pl. 33. figs. 5,6, 9, Pl. 34. figs. 14,15). ‘The necessary hold was not always gained at the first attempt, the dise some- times slipping, and here again the tail-bristles came into play, serving to anchor the animal to the bud and to give a certain amount of prop-like support to its rigid body. The position was not necessarily vertical, but at right angles to the supporting surface, and frequently oblique or even horizontal, and it was. remarkable how the vermiform, soft-bodied mite would maintain for several minutes an attitude apparently so ill adapted to its structure. All the time its four short legs would be waving wildly in the air. A number of mites standing up in this way bore a remarkable resemblance to diminutive Hydras with greatly retracted tentacles. After indulging in this performance for a period varying from one to five minutes, the mite would generally relax its rigid attitude, bring down its feet to the surface again, and continue its progress by crawling, only to resume its upright position and erotesque waving of legs a little farther on. This behaviour was highly suggestive of a desire, on the part of the mite, to attach itself to any passing object, and its readi- ness to do so was easily proved in the most conclusive manner. Tf touched with a needle-point, it immediately let go its hold on the bud and was carried off on the needle. A camel’s-hair brush or a feather applied to an infested bud was found to be swarming with mites on subsequent examination. In nature, the most likely carriers of the mites would, of course, be insects or arachnids. Accordingly spiders and insects of various kinds were either induced to run over infested buds, or examined after having been observed to come into contact with them spon- 370 MR. GC. WARBURTON AND MISS A. L. EMBLETON ON taneously, and in almost every instance mites were found attached to their bodies or appendages. The fact was recorded of four different species of spider, of the larva of the currant-moth (Abraxas grossulariata), of the currant Aphis, of the larva of the two-spot ladybird (Coccinella bipunctata), of the black ant (Lasius niger), and of various other insects. So uniform was the result, that the investigation into the various creatures capable of distributing the mite was presently discontinued, as it was clear that almost any insect might perform that function, though those which wander widely and especially affect currant- bushes would necessarily be most efficient. It is probable that the currant Aphis is especially instrumental in spreading the disease. It crawls slowly along, feeling its way with its antennae, to which the mites readily attach themselves, and the winged individuals would be extremely likely to convey the pest direct to another currant-bush. Method of Attachment. The mites do not seem in any true sense to grasp the objects presented to them, nor, indeed, do they possess any prehensile organ, unless the tail-dise may be placed in that category. Moreover, ordinary hairs and bristles are much too thick, rela- tively to the mites, for seizure by means of the jaws or legs. Attachment always seems to take place, in the first instance, simply on account of some adhesive substance with which the bodies of the animals are coated. A mite may be removed by any portion of its body being touched by the antenna of an aphis, but it quickly coils itself round the appendage in a worm-like fashion and brings its tail- dise into play. Possibly the bristles and ‘‘ feather-hairs”’ (Pl. 33. fig. 8) on the legs of the mite may to some degree entangle themselves among the fine hairs on the bodies and legs of insects. When removed on acamel’s-hair brush the mites wriggle in and out among the hairs and soon secure a tolerably firm hold. Leaping. While the behaviour of the upright mites was under observa- tion under the microscope, it was noticed that individuals some- times disappeared from the field of view with a suddenness that made it impossible to see what precisely had happened. This THE LIFE-HISTORY OF THE BLACK-CURRANT GALL-MITE. 371 ‘occurred several times before it was realized that the dis- appearance was not accidental, but that the animals were, in fact, leaping. As soon as this was suspected, it was easy to verify it by con- centrating attention on one individual and using powers which allowed the whole of its flight to be followed. The conclusions arrived at were these:—After several vain attempts to attach itself to a passing insect, a mite would cease to wave its legs, remain rigid a moment, and then launch itself forth, torpedo-like, into space. The precise mechanism by which this was effected could not be determined, but the terminal muscular disc, which had been observed to be retractile, was evidently the propelling organ. The tail-bristles were at first suspected of taking some part in the action, but further observation showed that, by re- taining too firm a hold on the bud, they sometimes rendered the leap abortive, the mite simply falling backwards with considerable impetus instead of darting away. No great distance was covered by the leap, the longest measured being four millimetres, or about sixteen times the animal’s length. When the mites leaped from a bud placed on a microscope-slide they alighted on their heads and fell over with the tail-dise most distant from the point of departure. It was an interesting and suggestive fact that while the mites would remain upright with waving legs for several minutes in the still air of the laboratory, they could be induced to leap at once by blowing upon them with the breath or by means of a pipette. It would seem, then, that they first of all try to come in contact with a passing insect, and, failing this, take advantage of a puff of air to attain their object. In view of the extremely doubtful advantage of a blind leap into space, the conjecture may be hazarded that the mite thus sometimes attains a flying insect which hovers near enough to fan it by the beating of its wings. Destination of the Migrating Mites. The problem of the immediate object of the mites in leaving the old buds by crawling, leaping, and adhering to insects next demanded a solution. At the height of the migration the new buds were already visible and beginning to swell, and the manner in which and the extent to which they acquired the disease had still to be ascertained. Moreover, the leap into space would 372 MR. C. WARBURTON AND MISS A. L. EMBLETON ON necessarily land many of the creatures on the ground, and it was conceivable that they sought or made some kind of shelter there from which a new attack sprang at a later period, or even that they sought the roots and set up there another form of the in- festation. Finally, as it transpired that here and there a resting- bud which had begun to show after the conclusion of the migra- tory period contained the mites, it seemed possible that some might find a temporary sbelter under the loose bark of the stem in the neighbourhood of such buds while still undeveloped. All these points were subject to careful investigation, the results of which may now be stated. Shelter under Bark. Unless called upon to furnish shoots on account of the exten- sive destruction of the ordinary buds by disease or injury, the resting-buds remain as almost invisible knobs under the bark of the stem. There is usually some loose bark in their vicinity and this was carefully searched for the mite, but with uniformly negative results. Specimens of a Lyroglyphus were found, and also some empty and longitudinally split shells which might have: been hypopal casts, but of the gall-mite not a specimen. Behaviour on the Ground. To trace the actions of such minute creatures amongst the precipices and chasms into which ordinary soil is converted by the microscope is well-nigh impossible. By preparing a specially fine mould the difficulty was reduced, and it was koped that any tendency to burrow into the earth or to encyst in sheltered recesses would at all events be detected. Experiments were made with both wet and dry earth, but here again the results were entirely negative. The mites showed no indication of having attained a desired end, but crawled laboriously among the par- ticles of earth, rearing themselves at intervals and waving their legs as though in the hope of rescue at the eleventh hour by some passing insect No burrowing, no encystment, no deposition of eggs was noted. For hours, even for days, the mites wandered aimlessly, becoming less and less vigorous till at length they died. On the dry mould they were more active at first, as the wet soil seemed to have a paralyzing effect for the time being. In the latter case the mites, hewever, lived the longer, several showing signs of life after the third day. / THE LIFE-HISTORY OF THE BLACK-CURRANT GALL-MITE. 373) Further experiments were made with black-currant roots which were placed upon the soil and dusted with actively migrating mites. Again nothing definite happened, the mites seemed in no way contented with their new environment, and acted as though the object of their migration were as far from attainment as ever. Such negative results are, of course, inconclusive, but it seems likely that all the mites which fail to the ground are doomed to perish unless they should have the extreme good fortune to be carried by some passing insect to another bush. Entry into the new Buds. During the first week in June the mites were wandering actively about the stems, and some were found in the axils of the leaves, and close to and upon the new buds (PI. 33. figs. 3,4). On June7 a mite artificially placed upon a young bud was seen to work its way in between the sheathing-leaves. New buds were removed at frequent intervals and examined for mites, which were found inside for the first time on June 8. During the ten days of more or less active migration which succeeded, the search for mites was successful in a fair percentage of cases, but from the number of buds which afterwards proved to be diseased it is likely that the presence of one or two of the animals was frequently overlooked— a fact, perhaps, not greatly to be wondered at. The mites in any one bud were always extremely few in number. ‘Till June 12 the weather had been hot and dry, and on that date very few mites were wandering, and those still within the old diseased buds showed slight signs of life. The migration was apparently almost at an end. Rain then fell, and this seemed to revive many of the mites and to prolong the migration period for afew days. With rare exceptions, wandering mites were not seen after June 19, by which time the old abortive buds were entirely lifeless. The migration, therefore, was at an end, and of the hosts of mites in existence at the end of May an infinitesimal number had obtained a footing in the new buds, all the rest having presumably perished. Behaviour of the Mites within the young Buds. The migrating mites were for the most part adult, and eggs could be seen in the transparent bodies of the females (Pl. 33. fic. 7). The date of the deposition of these eggs seemed to bea matter of some importance, though by no means easy to ascertain. The plan of searching for them by the dissection of individual buds was abandoned, but new buds were daily removed from 314 MR. C. WARBURTON AND MISS A. L. EMBLETON ON infested bushes, cut into small fragments, and placed in tubes of spirit. These were shaken vigorously and then allowed to stand, and the sediment examined under the microscope. Eggs were first found in the buds removed on June 26. After that date they occurred in increasing numbers. By the end of the first week of July the new attack appeared to be firmly established. Mites in all stages of development, as well as eggs, ould easily be found by dissection of the new buds, some of which seemed to show sigus of abnormal development. In every case the mites were centrally situated in the buds—a remarkable fact in view of their particularly compact structure in this plant, which would render it by no means easy for the immigrant mites to reach the interior. As the brood increased, the mites worked centrifugaliy, ‘encroaching gradually upon the more external portions of the bud. By July 20 the removal of two outside leaves sufficed, in some cases, to reveal the mites, and by the end of the month they were almost external in the most advanced buds, covered only by the loosely applied outermost leaves, while occasionally one or two individuals were found actually on the outside of the buds. The multiplication of the mites in the new buds had been so rapid during July and August, that it appeared highly probable that a second migration period was approaching. None such, however, was observed. The buds continued to swell, but re- mained green, and did not burst, and no wandering mites were detected during the autumn. The creatures were now established in their winter-quarters and were reproducing less actively. No new fact was to be added to the record of their life-history. Some might succumb to the rigours of winter, but the survivors would be ready to recommence the life-cycle in the following spring. Animals associated with the Mite. Runnicg with great activity over the twigs of infested bushes, examples of a red mite of the genus Actineda were constantly found. No conclusive evidence of its habits was obtained, but one specimen had attached to its jaws what appeared to be empty Hriophyes-skins, and it is quite possible that it preys to some extent on the mite, for it belongs to a predaceous group of the Trombidiide. Within the diseased buds specimens of a Tyroglyphus mite were THE LIFE-HISTORY OF THE BLACK-CURRANT GALL-MITE. 375 very frequently observed, and on one occasion, in October, a species of Tarsonemus was found. These are vegetable feeders, and could not have been preying upon the Hriophyes. In a large proportion of the mite-infested buds dissection revealed the presence of a small dipterous larva, apparently that of a Cecidomyid fly (PI. 34. figs. 21,22). It was hoped that some examples would be reared and the imago obtained and identified, but its development is so slow that larve observed in October are little larger than those seen in July, and show no signs of pupa- tion. Probably the fly will not emerge until next spring, when it is quite likely to prove a new species. It is always found at, or near, the centre of the bud, and feeds upon the mites. Its slow growth probably implies a moderation of appetite which allows the mites, by their extreme fertility, to renew their numbers as fast as they are depleted, and thus to keep up the food-supply. The Red-Currant Plant and the Mite. As none of the characteristic swollen buds appeared on red- currant bushes, which, moreover, flourished in the immediate neighbourhood of failing black-currant plants, it was believed that the red-currant was practically immune. When the Cecidomyid larva above mentioned was first observed and some doubt was entertained as to its habits, red-currant buds were examined to see if they contained the grub, the inference being that the pre- sence of the grub would show that it fed upon the bud and not on the mite. The result of the examination was entirely unexpected, for the mites were found in considerable numbers. The attack differed remarkably from that on the black-currant, the infestation work- ing from the outside inwards. At first they were only found in the axils of the leaves at the base of the buds, and perhaps within the first brown sheathing-leaves. Later they penetrated more deeply and had almost reached the centre. That they were not merely sheltering there but were obtaining nourishment was proved by the presence of eggsandlarve. No ereat harm, however, seemed to be done by them, nor were they ever found except on bushes near to badly attacked black-currant plants. 376 MR. C. WARBURTON AND MISS A. L. EMBLETON ON Conclusions. The observations above recorded may be thus briefly sum- marized :— 1. Of the mites surviving the winter, those which have not succeeded in entirely arresting the growth of the buds, but are driven out by their development before May, probably perish. 2. There is a definite migration period, which takes place as the abortive buds dry up and become uninhabitable, the new buds being then ready fer the reception of the mites. In 1901 this period extended from the middle of May to the middle of June. Any mites found wandering in the autumn are probably of the nature of an overflow. 3. Distribution is effected by (1) crawling, (2) adhering to insects, (3) leaping. 4. There is a brief period when the total number of living mites is exceedingly few, the old buds being dead, while the emigrants which have attained the new buds have not yet increased to any considerable extent. This period in 190% coincided with the last week in June. ; 5. The mites are unable to maintain life in the ground, nor do. they attack the roots. 6. The red-currant plant can contract the disease, but does not appear to suffer greatly from it. The mites first appear on the ’ outside of the buds, penetrating inwards as they increase in number. 7. Infested buds very commonly contain a Cecidomyid larva. which feeds on the mites. If the results thus obtained are trustworthy, certain inferences follow with regard to the treatment of the disease. The most important are these :— 1. Any treatment of the ground under infested bushes is. unnecessary, as the mites do not live in the soil. 9, Spraying in the early spring is only calculated to destroy mites which would perish in any case. 3. The only time when spraying would be likely to prove: beneficial is at the end of May and the beginning of June, when it is undesirable on account of the blossom. 4. The removal of all the new shoots from infested bushes at. the end of June, if practicable, would apparently clear the plants. of the disease. It is at all events important to remember that at. this time the pest is reduced to a minimum. THE LIFE-HISTORY OF THE BLACK-CURRANT GALL-MITE. Wh BIBLioGRAPHY. ERIOPHYES RIBIS, Westwood. Phytoptus ribis, Westwood, Gardeners’ Chronicle, 1869. Phytoptus ribis, Nalepa, Anz. Ak. Wien, xxx. p. 105 (1893). Eriophyes ribis, Nalepa, Das Tierreich, 1898. Though references to the black-currant gall-mite, and sugges- tions for its extermination, are frequent in the writings of economic entomologists, the serious contributions towards our knowledge of its life-history are few in number, and may here be conveniently summarized. 1. Newsteap.—“ Recent Investigations on the Currant-bud Mite, Phytoptus ribis.” The British Naturalist, 1894. Eeg-laying commenced, Feb. 20. Thirty per cent. of the old infested buds dry by May 19. Mites found in the axils of the leaves on June 2. First found within the new buds on July 27, with eges. Noobservations made during August. Migra- tion takes place during March. 2. Warspurton.—“ The Black-currant Gall-mite, Phytoptus ribis, Westwood.” The Journal of the Royal Agricultural Society, 3rd ser. vol. vill. p. 754 (1897). Eggs found every month except December. Mites migrate during the spring, when slightly infested buds open and badly infested begin to dry up. No means of distribution except by crawling detected. Attempts to ascertain what became of the dislodged eggs unsuccessful. 3. Witson.—“ Disease of the Black Currant caused by the Gall-mite (Phytoptus ribis).” Pamphlet issued by the County Council of Fife, Dec. 1898. * A few eggs found in December, numerous in February. Infested buds are dried up in July, before which time the mites wander. Distribution probably largely by wind, and perhaps by insects and birds, but no evidence offered in favour of this suggestion. EXPLANATION OF THE PLATES. Prats 33. Fig. 1. Twig of black currant showing diseased buds. From the apex to the point marked A is the current year’s growth ; between A and B is last year’s wood with the buds which shrivelled at the end of June 1901 ; below B is older wood with dry brown buds which dried in the summer 1900. Fig. LIFE-HISTORY OF THE BLACK-CURBANT GALL-MITE. . Diseased buds which have managed to produce dwarfed leaves. . New bud at the time when the mites are establishing themselves in it, and are now to be seen on the outside. Fig. 3 magnified. A piece of diseased bud, showing the mites and eggs in situ. . Two mites in the erect position, prior to leaping. . Lateral view of mite as seen when crawling, drawn on a ‘5 mm, chart. (Ultimate divisions=,'; mm.) Leg of mite, showing the ‘‘ feather-hair ” and bristles. . Posterior extremity of mite, showing the tail-disc and two bristles. when in the erect attitude assumed before leaping. Pruats 34. Dorsal view of the anterior end of the mite, showing the characteristic sculpturing of the “‘ carapace.” . Ventral view of same, with the mouth-parts (m.) and external sexual organs (s.). Dorsal view of posterior end. . Ventral view of posterior end. Diagrammatic representation of the positions assumed by the mite while endeavouring to obtain a firm hold by its tail before standing erect. . Showing how the erect attitude is assumed. . An egg, drawn on a‘l mm. chart, (Ultimate divisions=;35 mm.) . An egg at a later stage, with the contained mite, ready for hatching. Larval form of the mite. . Mite about to undergo ecdysis. . Mite crawling, as seen from above. . Cecidomyid larva found in the diseased buds—lateral view :— m.= mouth, s,=spiracle, 7.=trachea. . Magnified drawing of anterior end of the Cecidomyid larva. [Figs. 7, 10, 11, 16, 18, 19 drawn by the “ camera lucida.” ] _ Linn. Soc.Journ. Zoon Vou. XXVII.P1. 33. t i i e Se ORES ace prumble oe ad nat. West,Newman imp. ERIOPHYES ‘(eaytor tus) RIBIS. .Soc.Journ.Zoon.Vou. XXVIII. Pl, 34. L Eniaton del.adnat. : West,Newman imp. ammond lith. _ ERIOPHYES (PHYTOPTUS ) RIBIS. ~ ON THE FORAMINIFERA OF THE FUNAFUTI ATOLL. 379 On the ForaminiFEra collected round the Funafuti Atoll from Shallow and Moderately Deep Water. By FrupErick Cuapman, A.I.S., F.R.M.S, [Read 5th December, 1901.] (Puatses 35 & 36.) ConTENTS. I. Introductory Remarks. Pp. 379-380, IT, Foraminifera from the Ocean-Beaches (Sollas Collection), With Notes and Description of New Species. Pp. 380-386. III. Foraminifera from the Lagoon-Beaches, Funafuti (Sollas and David Collection). Pp. 386-387. IV, A Description of the Reef-fragments, with adherent Foraminifera, from the Reef-face, Funafuti (David and Woolnough Collection), With a Table of Distribution and Notes and Description of a New Species. Pp. 387-396. VY. A Summary of the Foraminifera found at depths from 16-200 fathoms round Funafuti (Halligan and Finckh Collection). With Notes and Description of New Species. Pp. 397-413. VI. Notes on the Distribution of some of the Species of Foraminifera from. the Reef-slope at Funafuti. Pp. 413-415. I. InrRopucToRY REMARKS. The collections upon which the following) results are based are these :— (1) A series of shore-sands from the ocean and lagoon-beaches of the Funafuti Atoll. Obtained by Prof. W. J. Sollas,. F.R.S., during the first expedition to Funafuti in 1896. (2) Samples from the rocks forming the seaward slope of the reef at depths from 16-200 fathoms, obtained by means of heavy steel chisels and tangles; also samples of sand from various depths. Collected by Prof. Edgeworth David and Mr. Woolnough in 1897. (83) A series of sand and reef-rocks collected by Messrs. G. Halligan and A. E. Finckh round the Atoll at depths down to 240 fathoms. Also soundings and dredgings taken along a line due west trom Tutanga. Collected in August 1898. ‘380 MR. F. CHAPMAN ON THE During my microscopical examination of the thin slices of the cores brought up by the boring operations at Funafuti it was evident that, in order to arrive at some definite conclusions re- specting the meaning of the various changes in the facies of the foraminifera and other microzoa found at different levels, we ought to know more about the actual distribution of the microzoa living round the atoll and in the lagoon. By a careful study of the differences in the distribution of the smaller organisms it is possible to gain information of considerable value as to the depth at which they best developed, and also with regard to the ac- companying physical or hydrographical conditions. The present paper should therefore serve as a basis for the discussion of the significance of the contents of the core as far as the foraminifera ‘are concerned (and these organisms, by the way, constitute the ereater proportion of the Funafuti core), since the results are derived from samples obtained from fairly shallow-water deposits, speaking in the hydrographical sense. This interesting subject, in its bearing on the contents of the -core, may be discussed in a later paper of this series. The deep-sea soundings taken round Funafuti are also of considerable interest, and their description may be reserved for another paper on the subject. TI. Foramintrera from the OcEAn-BEacues, Funarvurtt, from material collected by Prof. Somuas, 1896. Norz.—The actual label-names attached to the samples of ands have the following signification, and the native name alone is here retained :—‘‘ Our Islet” = Fongafale Islet or Funafuti Island; “South Island” = Avalau Islet; “Gold Island”= Fualopa Islet. The following terms in the Table denote the relative abundance of the specimens :—v.r.=very rare 5; r.=rare; f.=frequent ; .= common; V.c.=very common; ex.c.=excessively common. FORAMINIFERA OF THE FUNAFUTI ATOLL. | N.end Name. | Fongafale Islet. i Wabecularia lwcopfuga, Deir. |. csje.ces-e-) enone 2. 55 Brady, Millett .....-......--- sy oe “a lacunensiss Chap: ...-22 5-40-0624) ee -ase 4. * divaricata, Brady ............ I iXag Bea 5. | Biloculina irregularis, @Orb. ............... et eaenaiaet 6. ¥ TUG AOS (LUEMA,)) | saenoocaapeneboedd!) | aoacae 7 | Spiroloculina muda, @Orb. ..2.2.-20.----| sae. 8. a » var. foveolata, Egger| eae 9. + Gannadpombara, COG, soocescos! Howoce 10. a acutimargo, Brady ......... eee es 11. 5 antillarum, VOrb............. ieee: 12. a Girth, WUEH 6 cooobonsounendosee rarer 13. CP AMGUC:, IEIAREP ooo gcabeccdaed| cebonc 14. | Miliolina circularis (Born) Bash eae enone Mee acne ay 4 x var. sublineata, Brady} ....... 16. rs subrotunda (Mont.)......--:------| ...... mee a UGloeas@e (CVKOTS,)) Scscaccseacesoenee| aeeace 18. a PETODUUA, (IVETE) conopnncoccceoenad) — gaoeee 19. Me Oallopg a (NUCH) ceocoascsscdoccend| | aaneoe 20. BS Jiaciaeg (CKO) scopanescassace|| — cortoce 21. 7 SCOUT, (Ue)) seeacccseeneansosd| conus 22. Re funafutiensis, Chap. .......2-0--| ..:--- 23. yy LAP (Ch OED.) \odaodeasesasansead| senbee 24. 5 Charan (CVONID)) coconanocbed| — aeaooe 25. Ay PUG OREO, (CKOWD,) oancconceooscen| | acaoae 26. : . var. Bertheliniana,| ...... Brady. Zilfe 55 3 var. Terquemiana,| ...... Brady. 28. nm HIG CTTS. (N/a C5 Uo) ceodaseseccacba| | decade *29. 5 agglutinams (VOrb.) ....2..066.| veeeee *30. “a Tigiponsseee (GODS) suascscesccocssl| boone Sule 5 Tepe (GM O1N0))) Sedecospececocd| 6) ehabe 32. # undosa (Karrer) FAP Rane AAR OS ace | MmenE ee 33. Fe reticulata (VOrb.) ...... aieateseantia: «cleale as 34. aA GrO (UEIENGKY) 445 coondscoccob0od| — aeceoe 30. Articulina JSunalis, vax, inornata, Brady...| —...... 36. | Hauerina compressa (d’'Orb.) .......2-.2-.--| ------ one S ornatissima (Karrer) ............| ..-+- 38. | Planispirina exigua, Brady..............065] «2... 39. | Sigmoilina celata (Costa) .........660000008| caves 40.| Cornuspira involvens, Reuss .............02| s..0- 41.| Peneroplis pertusus (Forskal) ...........2006) 0 veeeee 42, * Grictimus) (Batsch)) .--.c-n-cceses| | siasl- *43. (Monalysidium) cylindraceus| ...... (Lam.). 44, as (GW) Uaamigas) (Ere) ha) Bacseanbecedl| | Baecod 45, 35 (Us) Solas; 5 @ happy oaseaccanea<|ti | saanins 46.| Orbitolites marginalis (Lam.) .........+-- r. 47. es Gujpicas © anpenteryuer seers. eee saces 48, 5 complanata, Lam. ........+...++- v.c 381 Avalau | Fualopa Islet. Islet. e€x.¢. * See notes appended to this list. LINN. JOURN.—ZOOLOGY, VOL. XXVIII. 382 MR. F. CHAPMAN ON THE Name. . | Orbitolites complanata, var. plicata (Danz) . | Haddonia minor, sp. NOV. .....-.....-.-++--| Textularia rugosa (Reuss) .....-.-..-....++- | 52. “ Gaiuaden, GL ONE 0); coasogesceodgsceosae | %53.| Verneutlina spinulosa (Reuss) ..... caress | 54.) Valuulina Davidiana, Chap. .............-- 55. | Clavulina angularis, @Orb..........-..10+++- 56. | Bolivina limbata, Brady .................- ne BY. ‘5 ODUCT, VBSEEKG Wi clon osacmogbeanscoosse 58. | Sagrina raphanus, P. & J... 12. eeeeeeeeeees | 59.| Globigerina sacculifera, Brady ............ 60. | Spirillina vivipara, Eby...............00000-+ | 61. zu inequalis, Brady...............++- 62. Ne spinigera, Chap. ..........++.--++- x63. | Patellina corrugata, Will. ..............+6- 64. | Cymbalopora Poeyt (V’Orb.) .....- 2.2.00 x65. a » var. sguamosa (d’Orb.) 66. HA tabelleformis, Brady ...... 67. Bs (Tretomphalus) bulloides (d’Orb.) 68. | Discorbina araucana (VOrD.) ..........06+5- 69. 7» Vilardeboana (d’Orb.) ........- 70. 5 OSTHAD (CHO) coacanenasacoocons Tole 3 PUGOSEA(AAOT DS) nusnateeonare cd. 72. » globularis (d’Orb.)........ ....-- 73. bs tabernacularis, Brady ......... ¥T4. pe ACUMINATE, SP. NOV. ....0.0000+- 1( By. % CONGINNO- ADE ACY apse seetselsr 76. Re orbicularis (Terq.) ..-..:--..-- 77.| Planorbulina larvata, PB. & J.......,.2..4-- 78. * acervalis, Brady ...........- #19. i retinaculata, P. & J......-..- 80. | Truncatulina lobatula (W. & J.) .......-. 8l. & variabilis (d’Orb.).........+-- 82. a ORCHID VBICGN Soseoocnccoode: 83. M reticulata (Czajzek) ......... 84.| Anomalina coronata (R. &I.).......0.00ee \x85.| Calcarina Spengleri (L.) ....0.se.ceeeeee oes 86. Me Toasyonaap, WBN bencaceeoncee. oe 87. x i var. pulchella, Chap. . 88. | Tinoporus baculatus (Montf.) ............6++ x89. a 3 yar. florescens, nov. . 90. | Gypsina inherens (Schultze) ............... Dil. » globulus (Reuss) .......2.002.....- 92. | Polytrema miniacewm (Pallas) ........-... | 93. | Polystomella striatopunctata (F. & M.)... 94. 2 macella, (Wy ds ME) teecess. es: 95. - rasp (UB). deerqcacoasesconosons 96. | Amphistegina Lessonit, V’Orb..........--+.+- Heterostegina depressa, Orb. Fualopa Islet. * See notes appended to this list. FORAMINIFERA OF.THE FUNAFUTLI ATOLL. 383 Of the foregomg samples of foraminiferal sands from the seaward beaches of Funafuti, that from Avalau Islet is by far the richest in organisms. The Foraminifera are there in great profusion, and the species very numerous for a coral area; the specimens themselves are beautifully preserved, even down to the most delicate ornamentation of spines and the perfect contour of the shell in many of the fragile forms. Besides the Foraminifera we find in the sand of Avalau Islet fragments of calcareous Alge, spicules of a Calcisponge, Aleyo- narian spicules, a few Heteropods and numerous Ostracoda ; the valves of the last-named organisms are more than usually abundant and varied, and these, together with other Ostracoda from Funafuti, will be enumerated and described in a separate paper. ‘fhe sand here examined from Fongafale Islet is water-worn and perhaps wind-polished, so that the result—the occurrence of five species only of Foraminifera—is not surprising. We now proceed to the description of new species, and notes on the more remarkable forms of the Foraminifera occurring in the beach-sands of Funafuti. Notes on the Foraminifera of the Beach-Sands, Funafuti. Mirtonina taBrosa (d’ Orbigny). The specimens from Avalau Islet exhibit the same tendency to merge into Nubecularia Bradleyz, Millett, by growing irregularly in a lateral direction until the milioline character is entirely lost, which Millett remarks in his description of the Malay fora- minifera *. There are apparently no specimens from Funafuti, such as were found in the Malay soundings, which pass into Miliolina valvularis (Reuss). MILIOLINA AGGLUTINANS (d’Ord.). The form which is rather frequent at Avalau Islet partakes of the general characters of J. Bosciana (d’Orb.), and might perhaps with equal reason be assigned to that species. Millett figures a similar example from the Malay Archipelago 7. Minroniwa Ferussacit (d’Ord.). Probably more than half the number of specimens from Funafuti are represented by the flattened costate variety, formed * Journ. R. Mier. Soe. 1898, p. 502. + Tom. cit. p. 268, pl. iv. figs. 4 ac. 384 MR. F. CHAPMAN ON THE almost on a spiroloculine plan, similar to the figure given by Millett of his Malay specimens *. PENEROPLIS (MoONALYSIDIUM) CYLINDRACEUS (Lam.). A very delicate little form, which by its thin shell-structure and simple oral aperture seems to belong to the subgeneric type Monalysidium. ORBITOLITES COMPLANATA, Lam., var. PuIcaTA, J. D. Dana. Marginopora vertebralis, Blainville, var. plicata, Dana, 1848, in Wilkes’ United States Exploring Expedition Reports, ‘‘ Zoophytes,” p. 706, [in vol. of plates referred to as Marginopora vertebralis?] pl. 60. figs. 9 Oia;,0: Orbitolites laciniatus, Brady, 1881, Quart. Journ. Micr. Sci. vol. xxi. INES: (ps 4a. O. complanata, var. laciniata, Carpenter, 1885, Report on the Genus Orbitolites, Zool. Chall. Exp. part xxii. pl. vii. This is the well-known thick variety of Orbitolites with the plicated margin, and which Brady showed to be a stage of shell- growth dependent on a phase of reproduction, since the edge bears chamberlets with megalospheric young. J. D. Dana described this variety as plicata in 1848, and H. B. Brady appears to have overlooked this when he described his specimens from Fiji and elsewhere. HADDONTA MINOR, sp. nov. (PI. 36. figs. 1, 2.) Test attached by the earlier segments, which are frequently grouped in a triserial manner, as in Verneuilina, &e.; afterwards growing erect or in a vermiform fashion, similar to H. Torres- zensis, but is much smaller. Aperture horseshoe-shaped. Average length of test 2-4 mm. ; average diameter ‘7 mm. Avalau Islet; very rare. VERNEUILINA SPINULOSA (Heuwss). The specimens from Avalau Islet are in very fine condition, and the spinous processes are exceptionally long. PATELLINA CoRRUGATA, Williamson. It is very unusual to find this species in low latitudes, but it is not unknown from such localities ; it has, for instance, been recorded from Mauritius and elsewhere. It is, however, more abundant in temperate and colder areas. * Journ. R, Mier. Soc. 1898, p. 507, pl. xii. figs. 7 a—c. FORAMINIFERA OF THE FUNAFUTI ATOLL. 385 CyMBALOPORA PoryI, var. sqguAMOSA (d’Orb.). Rotaha squamosa, d’Orb., 1826, Ann. Sci. Nat. vol. vii. p. 272. no. 8. Rosalina squamosa, d’Orb., 1839, Foram. Cuba, p. 100, pl. iii. figs. 12-14, Cymbalopora Poeyi, d’Orb., depressed var., Brady, 1884, Rep. Chall. vol. ix. p. 637, pl. cii. figs. 14 a-d. This variety is a neat depressed form of the heavier sub- conical specific type; in its earlier stages it is sometimes found parasitic upon alge. DIscoRBINA ACUMINATA, sp.nov. (PI. 36. fig. 3.) Test conical, elongate; the apex terminating in a sharp point. The inferior face deeply sunken. Chambers arranged in about six whorls, the segments long and set obliquely. Surface of test ornamented with radiating striz centred in the apex and the umbilicus respectively. Height -3 mm.; diameter -2 mm. Although D. acuminata is related to D. tabernacularis, Brady, it differs in having straighter and longer sides to the cone and a pointed aboral extremity. Shore-sand, Avalau Islet, Funafuti; rare. PLANORBULINA RETINACULATA, Parker & Jones. Planorbulina retinaculata, P. & J., Phil. Trans. vol. cly. 1865, p. 380, pl. xix. fig. 2. A wild-growing modification of P. mediterranensis, d’Orbigny, parasitic on shells or alge, in which the chambers of the later whorls are partially separated, and bear numerous apertures especially around the periphery of the test. Parker and Jones’s specimens were found in the West Indies. This form is especially worthy of notice, as it does not appear to have occurred often, if at all, since the original description was published. P. retin- aculata occurs at Funafuti detached from their surfaces of support and mingled with the sand. CaLCaRINA SPENGLERI (Linné). This species appears to be almost exclusively confined to the East Indian Archipelago, and therefore its occurrence at Funafuti in at least one of the samples of beach-sands is interesting as adding to its geographical range. The Funafuti specimens are small but typical, 386 MR. F. CHAPMAN ON THE Tinoporus BacuLAtus (Montfort), var. FLORESCENS, nov. (Pl. 36. fig. 4.) This variety has its distinguishing feature in the curious dehiscent or florescent terminations of the spurs of the test. This is proved by thin sections to be formed by the redundant overgrowth of the acervuline or compressed outer layers of cham- berlets upon the intermediate shell-growth forming the spurs. This overgrowth is very thin, and covering the spurs forms a recurved edge around their extremities. Occurs at Avalau Islet. Ill. Foramtnirera from the Lacoon-Beacues, FUNAFUTI. Two samples of the foraminiferal sand of the lagoon-beaches are noticed here, with the species of foraminifera found therein. One is from Funafuti Island (Fongafale I.), collected by Prof. Sollas in 1896; the other from the S.E. of the Atoll at Funafala Islet, collected by Messrs. Halligan and Finckh in 1898. The dredgings taken across the lagoon commencing off Fonga- fale at a depth of 10 fathoms have been microscopically examined for foraminifera and already reported upon *. Lagoon-beach Name. pane oy S. end of 8 ‘| Funafala I. | 1.| Nubecularia divaricata, Brady ....... |e pi peuewoumee v.r. a2 = lucifuga, Defrance ......... | v.r. | 3% | Jacunensis, Chapman ....?. hee loeenBee ss Vas 4, | Spir -oloculina nitida, Ouse Aenea: etinieterrroc Te 5. | 5) mie ooraltns | © acossnade v.r. | Egger 6. | # ROR ONS Pep aranpeo-nciocnec|s sormaccbec : Fl Mi CHUM Whit OOM, execccacs| | ecoosacne vr. 8.| Miliolina seminulwim (Li.)...cecccesccessees| aeaneeree @: Oalaane oblonga (Mont.) .......0.-++.+ v.r. | 10. | H HOON (LENEDS) }agecarotecnecnd| > orgeacoce r. His | 3 tricarinata (d’Orb.), var. Ber-| ify v.r. | theliniana, Brady. | ie Hp tricarinata, var. Ter- (z quemiana, Brady. 135) 2 reticulata (d’Orb.) ........+-.. Ts (oe 14. Terai (CROMD,)) secoontecvbs| = axcooncce | v.r. Ney | Peneroplis Periusts:(orskall)\ccsescceejit) oeeseeerer f. 16. f arietinus (Batsch) ............ lieve 1 pace Speer us 17. | Orbitolites complanata, Lam. ............| v.c. V.c. 18. s marginalis (Liam.) .....6..0+6 ie c. * See Journ, Linn, Soc., Zool. yol, xxviii. pp. 161-210, FORAMINIFERA OF THE FUNAFUTI ATOLL. 387 | Lagoon-beach Lagoon-beach area te Name. | at eoneaeate I Sh eat al | Funafala I. 19) Lextulania gramen, Orb. ......-....-.-- eae v.r 20. < rugosa (Reuss) .......--...+- a GNA ed if Pile as siphonifera, Brady............ May nica eceee ces v.r. 22..| Clavulina angularis, VOrb................ I) Seeesaass v.r. 23.| Valvulina Davidiana, Chapman....Js..) .....-.5. Wee 24.| Sagrina raphanus, Parker & Jones ...) —...... ae ip 25.| Cymbalopora Poeyi (@Orb.) ..........-. | v.r. c. 26. | Discorbina globularis (VOrb.)............ [Paatl a goieats ee v.r. 27.| Truncatulina Akneriana (VOrb.) ...... ati Cede a v.r. 28.| Calcarina hispida, Brady..............-... | C: | V.C. 29.| Tinoporus baculatus (Montf.) ............ | eX.€, v.c. 30. | Gypsina inherens (Schultze) ............ v.r. ils on CCSHCOMEHES (2&5 U5) Gocsescsdoes v.r. 32.) Polytrema miniacewm (Pallas)...........- | C. G | 38. | Amphistegina Lessonii, V@Orb............. | ex.c. v.c. | 84. | Heterostegina depressa, VOrb............. v.r. | v.r. i IV. A Desorterion of the REenF-FRAGMENTS obtained from the ReEnF-FACE, FUNAFUTI, upon which ADHERENT FORAMINIFERA have been found. It seems desirable to keep the description of this series of specimens distinct from the foraminiferal sands, chiefly in order to show how important a part the larger encrusting and adherent foraminifera play in forming the growing reef, a fact which has been brought into prominence by the evidence of the Funafuti collections, both of the core and the samples dredged up from the living reef. These reef-samples are here arranged, firstly, in their order of position around the Atoll from N., E., 8., to W.., and, secondly, in the order of the depth from which they were dredged. The foraminiferal sands which are described subse- quently are arranged in order of depth; the bathymetrical distribution of the various organisms may thus be readily seen. N.W. of Pava I., 63 fathoms (1897). Two reef-specimens. (1) An encrusting mass of Polytrema planum measuring 3 cm. X 2°75 cm. This specimen was evidently torn off the reef at a weak point of attachment. It is smooth exteriorly, with a slightly undulate surface, and shows on the under, attached, surface a rudely concentric manner of growth, 388 MR. F. CHAPMAN ON THE At one side of this specimen a full-grown megalospheric test of Cycloclypeus Carpenteri has been partially encrusted and over- grown by the Polytrema. Pava I., 240 fathoms (1898). Specimen A 51. A deep-sea coral with numerous adherent tests of Carpenteria balaniformis, and a doubtful specimen of C. rhaphidodendron. Funamanu (Beacon Id.), 25 fathoms. An aleyonarian stem encrusted in places with a pale green Polytrema planum and a species of bryozoa, and bearing on its surface numerous specimens of Carpenteria monticularis, C. utri- cularis, and Polytrema miniaceum. Funamanu (Beacon Id.), 45 fathoms (coll. A). A somewhat thin and flexuose piece of reef-rock measuring 5 em.x3 cm., encrusted with alge, foraminifera, hydrozoa, and bryozoa. The foraminifera are Polytrema planum and P. miniaceum, both represented only by young growths. Funamanu (Beacon Id.), 50 fathoms (1897). Svecimen C 1. Woral-rock encrusted with Lithothamnion Philippii var. funa- futiensis, Carpenteria monticularis, Polytrema planum, P. mini- acewm and var. alba, Serpule, and bryozoa. Specimen © 2. A thin fragment of coral-rock with adherent organisms— Lithothamnion, foraminifera, Serpule, and bryozoa. The forami- nifera are Sagenina frondescens, Bdelloidina aggregata, Carpen- teria monticularis, C. utricularis, and Polytrema miniaceum. (2) Another specimen of P. planum growing on a base of hard ? coral-rock, measuring 3 cm.x2 cm. The Polytrema has grown irregularly, forming thin layers. On the rougher side of this specimen Carpenteria monticularis occurs, and here and there are little patches of the pink Polytrema miniacewm. FORAMINIFERA OF THE FUNAFUTI ATOLL. 389 Off Funamanu (Beacon Id.), 80 fathoms (1897). Specimen C 6. An axis of a Gorgoniid with an encrusting Lithothamnion and bryozoa, also some adherent foraminifera and Serpule. The foraminifera are Carpenteria monticularis, Polytrema miniaceum» and P. planwm. Another similar fragment (see Pl. 35. fig. 2) shows, in addition to the above species, a good example of Carpenteria utricularis and a large mass of Carpenteria rhaphi- dodendron. Off Funamanu (Beacon Id.), 80 fathoms (1898). Specimens A 22. Several fragments broken from the reef; some consisting almost entirely of successive layers of Polytrema planum having a snowy or frothy texture and appearance. ‘T'wo of the fragments have well-developed specimens of Carpenteria rhaphidodendron adhering to their surfaces, one of them measuring 3 em. in height. A fragment of Turbinaria perforated by Cliona carries several specimens of Haddonia torresiensis on one surface, and on the opposite face numerous bryozoa, a sponge, Halimeda, and the pink encrusting Lithothamnion Philippi var. funafutiensis, and the following foraminifera :— Carpenteria monticularis, C. utricularis, and Polytrema miniaceum iu a young stage of growth. Specimen A 24. Two fragments of reef-rock, the upper surfaces of which are entirely overgrown with pure white examples of Polytrema planum. On the under surface bryozoa, Serpule, and Polytrema miniaceum occur. Falefatu, 38 fathoms (1898). Specimen A 19. A piece of hard calcareous rock, measuring 18x11x4 em., overgrown on the upper surface with knobs and crusts of Litho- thamnion Philippi var. Sunafutiensis, Psammocora sp., bryozoa, and brachiopoda (Crania). Also the foraminifera Polytrema planum and P. miniaceum. The corals are chiefly adherent to 390 MR. F. CHAPMAN ON THE the upper (cleaner) surface, the bryozoa on the lower surface, whilst P. planum and P. miniaceum are on both surfaces. Off Tutanga, 60-100 fathoms (1898). Specimen A 35. A fragment of reef-rock measuring 7°5 x 4°5 cm., encrusted with Polytrema planum, which almost completely covers the specimen. There are also a few thin crusts of Lithothamnion Philippi var. funafutiensis associated with it, and an example of Cycloclypeus Carpenteri (form B), measuring 2°5 em. in diameter. The base of the rock is cavernous and drilled by boring organisms. Off Tutanga, 115-200 fathoms (1898). Carpenteria balaniformis very numerous on deep-sea corals (Oculinide). Off Tutanga, 117 fathoms (1897). Specimen A 32. A rough, irregular fragment of reef-rock, measuring about 10x9x6 em., consisting of an aggregate of organisms, as foraminifera, minute corals, hydrozoa, and Serpule, solidified by intergrowth and encrustation. By far the larger mass of the rock is formed of the encrusting and cementing organism Poly- trema planum, which here still retains the pale green colour so frequently seen in living specimens. This green coloration gives to the Polytrema au illusionary resemblance to an alga. The large form (B) of Cycloclypeus Carpenteri is represented in this block by a specimen measuring 5 cm. in diameter, whilst there are numerous examples of the smaller form (A) embedded between the other organisms. Off Tutanga, 135 fathoms (August 1898). Specimen B7. A collection of reef-fragments ; consisting of some large masses of Polytrema planum (see Pl. 35. fig. 4), one or two measuring about 5 cm. square; some lamellibranch shells overgrown inside and out with Serpule and Polytrema planum, fungoid corals accreted with growing organisms, chiefly Polytrema planum ; a fragment of an alcyonarian stem measuring 4°5 cm. in length FORAMINIFERA OF THE FUNAFUTI ATOLL. 391 and having a diameter of 1:8 cm., encrusted with Polytrema planum, which has ensured its preservation. Dredgings taken near Tutanga (bearing 102° to Tutanga, 155° to Tegasu), 136 fathoms (1897). Fragmentary rock-specimens and Halimeda-joints. One piece consisting of an encrusting mass of Polytrema planum measures 3X2 em. and is 4 mm. thick. The outer surface is smooth and of a very pale green colour. ‘lo the under surface a fine speci- men of Haddonia torresiensis is attached. The smaller fragments bear numerous dark-coloured speci- mens of Polytrema miniaceum. A fragment of Aleyonarian largely composed of the aggluti- nated spicules. Associated with these are Sagenina frondescens (on Halimeda) and Cycloclypeus Carpentert (form A). Dredgings west of Tutanga, 200 fathoms. Specimen A 2. Two rather massive pieces of organic calcareous rock. and fragments of a Gorgoniid stem. The largest piece of rock measures 12°5x6xX3°5 cm., and consists of large flaky masses of Polytrema planum built in tiers, rudely resembling the nest of the wasp (Vespa); with many adherent foraminifera, bryozoa, Serpule, brachiopoda (Cranza), and alge. Besides P. planum the other foraminifera are Poly- trema miniaceum, whose small pustular tests are scattered over a large portion of the rock, and Carpenteria serialis, sp. nov. (Pl. 35. fig. 3). There are apparently two kinds of alge—one a thin, pink, encrusting form, and the other a thread-like or filamentose and branching organism rather doubtful in its affinity. The smaller piece of rock measures 8°5 x 5 x 2°5 em., and has a double nodular shape, with a lumpy surface, overgrown with Polytrema planum, P. miniaceum, and several species of bryozoa. The alcyonarian stems are encrusted with a pink alga, and foraminifera (as Carpenteria utricularis, Polytrema miniaceum), a sponge, and numerous bryozoa, 392 MR. F. CHAPMAN ON THE South of Fuafatu, 25 fathoms (August 14th, 1897). Reef-fragments with broken shells and many foraminifera. The latter are :— Orbitolites complanata, rare. Sagenina frondescens, common and well-grown, on shells and Halimeda. Planorbulina acervalis, very rare, on Halimeda. Polytrema planum, on Halimeda, aud forming button - like masses. P. miniaceum, very common. Off Fuafatu, 60 fathoms (1897). Specimen C 4. A flat piece of coral covered with pink Lithothamnion, forami- nifera, Serpule, bryozoa, and brachiopoda (Cranza). The foraminifera are Sagenina frondescens, Haddonia torresi- ensis, and Polytrema miniaceum. Off Fuafatu, 60 fathoms (1897). Specimen C3. Coral-rock encrusted with Lithothamnion, Polytrema planum, and bryozoa. S.S.W. of Fuafatu, 60 fathoms (1897). Specimen C 5. Calcareous rock encrusted with Lithothamnion, foraminifera, a small coral, Serpule, and bryozoa. The foraminifera are Haddonia torresiensis, Carpenteria monticularis, 1nd Polytrema miniaceum. South of Fuafatu, 119 fathoms (1897). Reef-fragments, one of which measures 2°5X2 cm. It is_ encrusted with Polytrema planum to which are attached several specimens of Haddonia torresiensis. There are also specimens of Cycloclypeus Carpenteri ({orm A) in the accompanying sand. FORAMINIFERA OF THE FUNAFUTI ATOLL. 398 TABLE tllustrating the DistRIBUTION of ReEF-FoRMING ForRamI- NIFERA (adherent and encrusting species) round the Atoll of Funafuti *. Name. Localities. 1. | Sagenina frondescens| Pava, Funamanu, (Brady). Tutanga, and Fuafatu. 2. | Haddonia torresiensis,| Pava, Funamanu, Chapman. (Pl. 35.) Tutanga, and figs. 1 & 1a.) Fuafatu. 3. | Bdelloidina aggregata| Pava and Funa- Carter. manu. 4. | Carpenteriautricularis,| Pava, Funamanu, Carter. Tutanga, and Fuafatu. 5. | Carpenteria balani- | Pavaand Tutanga. formis, Gray. 6. | Carpenteria monticu-| Kunamanu, ‘Tu- laris, Carter. tanga, and Fua- fatu. 7. | Carpenteria rhaphi-\? Pava; Funamanu dodendron, Mobius.) and Fuafatu. (Pl. 35. fig. 2.) 8. | Carpenteria _ serialis,) Tutanga. Spaeov. (bl) 35: fig. 3.) 9. | Polytrema planum | Pava, Funamanu, (Carter). (Pl. 35.) Falefatu, Tu- figs. 2 & 4.) tanga, Fuafatu. (Pallas). 11. var. alba, Carter. * This table includes data given in my earlier paper on the same subject. Journ. Linn. Soc., Zool. vol. 10. | Polytrema miniacewum| Funamanu, Fale- fathoms. 30. 36, 50, 60, 136, 150, 200. 25, 35, 40. 60, 80, 119, 136. 25, 50, 60, 63. 25, 50, 57, 60, 80,94, 136, 150, 200. 115-240. 25, 50, 60, 80, 86, 135, 200. 60, 80, ? 240. 50, 60, 63, Depths in | Conditions of growth. Found chiefly on Halimeda joints. Adherent to reef-rock and reef-organisms. Found growing on reef- rock, millepores, corals, and molluscan shells. Grows attached to Aali- meda, Aleyonarian stems, millepores, or bare reef- rock. Growing on deep-sea corals and Serpule. Growing attached to Poly- trema planwm, Alcyo- narian stems, molluscan shells, or bare reef-rock. Found only at two or three depths. Grows in massive clusters, throwing out tubes which are often joined terminally by a platform-like growth of! Polytrema planum. Attached to reef-fragments. Found encrusting reef-rocks and rounding off sharp angular fragments by en- 119, 135, 136, 200. 25, 38, 45, fatu, and Fuafatu. Polytrema miniaceum,| Funamanu. xxviii. pp. 1-27. Tutanga,| 5 50, 60, 80, wrapping them in succes- sive layers of aceryuline cells until the original contour entirely disap- pears. It even grows over living organisms, such as Cycloclypeus, until they are quite covered up by the rapid growth of they foraminifera. Growing profusely on Poly- trema planum, Halimeda joints, or reef-rock. 136, 200. 0. Associated with P. minia- ceum but very rare. See 394 MR. F. CHAPMAN ON THE Besides the foregoing adherent foraminifera, the ordinary species of smaller dimensions which live freely on the reef play an important part by the enormous quantities of their tests which speedily become encrusted and consolidated by the growth of organisms. A cavernous rock is thus formed, the interstices of which do not become filled until a much later date in the history of the reef-formation. Such noteworthy species are Amphistegina Lessonit, Tinoporus baculatus, Heterostegina depressa, Calcarina hispida (which, however, is one of the first organisms to disappear by solution), Orbitolites complanata and O. marginalis, and Cyeloclypeus Carpenter. As regards the last named species, C. Carpenteri, it is interesting to note its occurrence at four localities round Funafuti, namely, Pava, Funamanu, Tutanga, and Fuafatu. It has a range in depth of 30-200 fathoms. At 50-60 fathoms both the megalospheric and the microspheric forms occur, form A greatly preponderating. At 80 fathoms form B (the microspheric or large discoid form) was most frequently dredged up alive by Prof. David. Notes on Foraminifera from the Reef-fragments, Funafuti. CARPENTERIA BALANIFORMIS, Gray. Carpenteria balaniformis, Gray, 1858, Proc. Zool. Soc. Lond. vol. xxvi. p. 269, figs. 1-4. C. balaniformis, Chapman, 1900, Journ. Linn. Soc., Zool. vol. xxviii. p. 18, pl. 4. figs. 1, 2. Other deep-sea corals with numerous attached specimens of C. balaniformis have been sent on in further collections since the first paper on Funafuti Foraminifera was written. One of the corals came from 240 fathoms off Pava I., and is the deepest sample obtained from the reef. This specimen bears no less than 31 individuals of the above species distributed over the surface. A curious example of fusion between two shells also occurs, which points to the ability which this genus may possess of forming colonies and large masses of almost indefinite size, providing the growing test is not broken up by predatory fishes or by mechanical means. FORAMINIFERA OF THE FUNAFUTI ATOLL. 395 CARPENTERIA RHAPHIDODENDRON, Mobius. (Pl. 35. fig. 2.) ? Polytrema brunnescens, J. D. Dana, 1849, U.S. Exploring Exped., Atlas Zoophytes, p. 707, pl. 61. fig. 3. Rhaphidodendron album, Mobius, 1876, Tageblatt der 49 Versammlung deutscher Naturforscher und Aerzte in Hamburg, p. 115. Carpenteria rhaphidodendron, Mobius, 1880, Beitrige zur Meeresfauna der Insel Mauritius und der Seychellen, Berlin, p. 81, pl. v. figs. 6-10, pl. vi. fig. 1-6. It is possible that the specimen which Dana described as “ pale brownish, thick incrusting, cavernous, surface gibbous, lacerate and very irregular,” is a young specimen of the above species. Since the full-grown form has been so admirably figured and described by Mobius from Mauritius, it is unnecessary to further disturb the nomenclature by substituting Dana’s name. This species is very much in evidence in certain parts of the core obtained from Funafuti, where it sometimes constitutes thick layers between the ordinary foraminiferal sand and reef- rock. More often, however, it is represented only by fragments broken down to a more or less uniform size, as though by the agency of browsing animals. C. rhaphidodendron appears to be most at home in depths between 63 and 80 fathoms. CARPENTERIA SERIALIS, sp. nov. (Pl. 35. fig. 3.) Test hyaline, somewhat glassy or polished in texture, consisting of more or less numerous chambers sometimes shaped like a calabash or water-pot with a distinctly spouted aperture placed a little eccentrically, at others of a combination of flask and long cylindrical spout. The apertures, in fresh specimen, armed with sponge-spicules. Chambers arranged in roughly linear fashion or in a meandering series. Adherent to reef-rock. Diameter of the chambers at their base 1—2°25 millim. Diameter of aperture about °3 millim. This organism appears at first sight to bear a deceptive resemblance to certain forms of Polyzoa. Found at Tutanga, 200 fathoms. Frequent. 396 MR. F. CHAPMAN ON THE PoLYTREMA PLANUM, Carter. (PI. 35. figs. 2 & 4.) Polytrema planum, Carter, 1876, Ann. & Mag. Nat. Hist. ser. 4, vol. xvii. pp. 211, 212, pl. xiii. figs. 18, 19. Gypsina melobesoides, Carter, 1877, Ann. & Mag. Nat. Hist. ser. 4, vol. xx. p. 172. Polytrema miniaceum, var. involva, Chapman, 1900, Journ. Linn. Soe., Zool. vol. xxviii. pp. 17, 18, pl. 2. fig. 5, and text-figure 2. P. planum, Carter, Chapman, 1901, Ann. & Mag. Nat. Hist. ser. 7, vol. vii. pp. 82, 83. P. planum, Chapman, 1901, Journ. Linn. Soc., Zool. vol. xxviii. pp- 201, 202, pl. 20. figs. 6, 7. This freely-growing foraminifer produces extraordinarily large masses of calcareous rock at Funafuti. At first sight they were thought to be calcareous alge, for their habit of growth is very similar to encrusting forms of Lithothamnion and Lithophyllum. A microscope-section of the organism, however, at once reveals its relationship with Polytrema and Gypsina. It is extremely difficult to decide from an examiuation of the external surface whether the specimens are foraminifera or alge, unless one has a special knowledge of the minute differences met with in the respective groups. It is not uncommon to find this organism forming an encrusting mass of about 5 centimetres square. One example from Tutanga (60-100 fathoms) measures about 7 x5 centimetres. Another piece of calcareous rock, composed almost entirely of laminar growths of P. planwm, has a measurement of 13 xX5 x4 centi- metres. As regards the depths at which this peculiarly interesting foraminifer is found, the limits of its best development are from about 80 to 200 fathoms. Chapman. Linn. Soc. Jounm. Zoor. Vor. XXVIII Prost Chapman ad nat, del. MP. Parker lith. Parker & West imp. REHEF-BUILDING FO RAMINIFERA FROM PUNARFUTI. FORAMINIFERA OF THE FUNAFUTI ATOLL. 397 V. A Summary of the Foraminirera found at pepras from 16-200 FraTHoms ROUND Funarumi, in dredgings made by Messrs. Hatniean and FIncku. In this table the results of the detailed examination of the sands obtaied during the latter part of the work of the Expedition, in 1898, are given. The columns are arranged in order of depth, the localities occupying a secondary place. The reason adduced for this is that in the case of Foraminifera from the loose sands found either in shallow or fairly deep water the temperature seems to be a more important factor than local surroundings; whereas in the ease of the reef samples the organisms are largely dependent upon environment. The dredgings include, besides Foraminifera, the following organisms—Calcareous Aloe ( Halimeda,Corallina, Lithothamnion, and other genera), Calcisponges and loose spicules, Aleyonarian spicules, Serpule and boring Annelides, Polyzoa, Ascidian spicules, Lamellibranchiate shells, Heteropods, Gasteropods, Pteropods, Echmodermal plates and spines, Ostracoda (chiefly Bairdia and Loxoconcha) and other Crustacea. In the annexed synopsis of species from the Funafuti Atol, for the sake of uniformity of treatment and in order to enable a comparison with the Synopsis given in my former paper (pp. 206- 209 of the present volume), the following significations are used :—v.r.= very rare, r.=rare, f.=frequent, c. = common, v.c. = very common. LINN. JOURN.—ZOOLOGY, VOL. XXVIII. 28 MR. F. CHAPMAN ON THE 398 o > So Oi nuorenng | : “ Q0Z-E11 “esueyny, | “09 ‘ngvyenyy Jo ‘g | Fe > ‘eSueqn y, 09-0G ce 0g ce nuemeun 7 | ‘ereyengy Jo's OF “e "TA > “ gg “eaeg jo “an | 3 e “esteyny, JO “MW | g ce ‘VA “ €@ “eas jo MN | oie ce 6 seesesessees Kyurey “MppaUIjQNS “ABA WLOT, ‘20070909 MURIONTTA” Cee twee wees ee at LOMIVY, “pyndadsp ec | Ce fr ee LOLAB IT NIDUWALI “ce sete e erie remo reer eeeees Apeag, ‘pussypbots 66 : Apeag ‘obuwajnon zs > Apeng ‘nangdasinuay 0 ‘tuonbaay, ‘gph a oe Ce * “qiO,P “WUNLDIIUYWY “cc ‘sou ‘ds ‘oynaind HY Peele cee came erey wmionbzo,f, “‘pssaudula se seesccrece Reyaboi(s ‘ognjoanof “IBA “c “ BOOEOCOOOCSnOONOUGROnOOTE IG 10.2 ‘ppuru c SOOO ROGUOUOON ODS SanOnoG Goon qt0.P ‘panqueay 66 “ ctosteseeeteeseseeeseees “GQ, p ‘Dypanoxa seem eee ener were eres ees enceee Apeag DISNGOL muygnoopo.ndsy Apeag, ‘nynjgnovjuap aed“ seen e en eee see ¢ wer] ) suabuid 1Q,P ‘vbuojgo nurgnooprg See eee eee wee tee “af OP “909U) ce so goueajary ‘wbnfiony ce ce Bebe eee e ewe ew ere eee wee eeee Apeag, ‘pynaDAUp DIMDINIOGNNT “@Aeg JO “NT Ty? Gil “OUTUG AT ‘SHONMIVE 00Z OL OT WOU AGA IWOAVNOY TH ACISHNO GNNAOA VAAFININVAOY AHL AO uTavy, MA 6d Hid Ss Xo ‘ONT 399 FORAMINIFERA OF THE FUNAFUTI ATOLL. "TA ‘TA ‘OA "TA ‘TA "*(B4S0K) »gn700 numroulbry Soran go waionfiog “SUUNULULOD 2 FOEROL Ce RB ARN CHE Caen * kpeag ‘onbrxa DUaUudswuny gy SOOO OOOO O nin iio iii ir (rere) DULASSIAWULO OG ae VOR “‘pssaLdauoa BULLIND ET “sou “ds ‘to i oS BO RAC Er ape aCnU Apu Ig, “Uuujnprnung CET TOMAELIEO) See pala ieee Apeag DpoUsour eA © COCO pee tee e wees penounn AVorRT ‘sypun OG . teen wees Aprig ‘DnaUN 3 SesieD) Dm ONLUOD DUIINIIIA fefeteCetsarelatelarateloereteteleverereteeaersretsterercts “Apmag * nywayronue srsesees KvIG ‘yynjna29G0Was te eee mee tte eee e ten eees Aperg ‘pyn.Lyyna ce ‘deyp ‘sisuann foun e erereretoiefeletelelstareterectene(eeistey-isteys (q10, p) DUDS T 7 SOOOOOIOO CCN initiry Apeag ‘saprojpphhum 6c SOOGOnOOOO OO On SCn nn cc nnn iii (-quoyt) nbuojgo OG 29000 Aperg, ‘sanwofrurjoanpn ye CEPA) syivong = pavoat SOOUOHDOUNDOUUODONOOUROG (q.10,p) DUDdWUUT GG sen eee eter e were ruses (q10,p) myjayoqnd GG ajatstareleVsioipiaisiululelaletelslelvistatetereyui (‘q10,p) supa bby (73 5 . on een e ewes (‘q410,~) (OTST ce Sew ee ete emer ewe eee toes eoressees Apeag ‘ WAY MDT ce i ii i a i ia] (LoaIV yy ) DsopUn ii5 ee oRb. sess COTO Pp) wynINaIIAL ep cod ERO iatelaleleidatehiteretiteseteyetersiatsye (-q.10,p) DUDUAAN) 6 so CUUIT) wagnunuas ie Apeag, ‘pupruandia J, “Ved ss of i IOP) vyoUrwvaIL) & (QUOT) vpunjosqns s jeeseees Cn ii era ir) (Gio) SOY) oe 28* | | | | “1 (6a alsie see “A aiere eas Y | . | ernie San At Siete } mars eae ariixrejeiavastoversnalel tielevakete(sip{ajeiece Pp 39 Sal Wnryof DIMDINALAT, ON tee ict woe Bory aj) oer eon con ev cwesceecnsa= nese tesece . oar CP@) “nysajopnouds DUIMIZLO) “OD, | | wee | see 500 oan Sree) Pevejeva!siuce)s) ajo iatalelel ute aeeuenee aye pnULOnDS DUMLUDYION 7, ‘Gy, BE 5 Fees. Mls | vee Sere dnb 4 ticedgoodooner phe rrintsteleteretreletale “= “sou ‘ds “oun ts eh HOS Wien reeae PMI Colaletelscarelnletoreibietarsieiatere efelalsvalaresule P ‘deyg *S18UalsallO} DYyLoppoeL ei) tae eis S00 SBzi J aeooooapSagaocaonG SDEOGOOEOOGK ehh Ge) ‘nap hawhbo DUYPLOPIAPT Gl, “aa [tata | ott eee foe | octe | see ory Ibori) 7° RN TA || -2e74 || eee Posen tereeerererees QQ.p “nupuowoon purpsdoo) ow, eee eee 2 eee see woe aieie) nee | ctovouuonoo CoOGecAoG on Apeg, ounUou 6c ‘OL, eee atols te and alee non cnn Ana looorodooocoo” od0n (-qt0.~) sunurjpbbo iG 69 nee woe aon coe [teers teens ste eeres “AOU ‘ds “Uungnjassay 3 R29 B60 || ot 4] ond. s|}-a5" || “ada leroobonen Cura ssssees (TOMAR) $2880 mnuubniydojdnzy | *L9 a eee oe . . wae wee eee oe OG nee wae eee ee een twee weer eens et eecee (4.10J}U0J ) snwnad.loas apy doay ‘99 . 1 ‘ . , boo Hon ‘9 Jhb |jpunmoedogs cmon eels aweneior se (AEE) suaosapuouf nupwabog "GQ aoe Os Pee (Cn ca sirerteseeseeeees (QOUBITO(T) NISOT pUyooy £9 OPO POO RNA f “nana ducoa se *E9 ‘UA oe wee ° atere oe BDo Eo no SO Oooo ooo aanta h (:1@) ‘xapdnp (iG 69 . | wae . . *, Pe peewee eeseeeses (‘uerq) sypurpumul SOO) “19 | | STAN see ate Taya) | Macatale wee re Cn ‘deyo sD] 105) 73 oe 09 one see wee siete 6qq jjocuoK aeeeeee WITOLD) ‘srngu) (unrprshpouojy ) 73 6G | see . . cee see wae Aon vee [ote ceneeesesecsscesssserseesee (qosyeq) snUrjarD GG "9G «ee eee sare [errs 600 wen [tener eerneene see eeeerseerseees TRysdoyy ‘snsnpuad syjdorauag LG Stave san. |lobodoomoaonds ateielsieisyare teres ssnoay ‘swanjonur plrdsmutoy ‘9G "CA “1 ae “I aes see see “TA see tz — oa a oy | | MR. F. CHAPMAN ON THE ray “ Q0z-G11 ‘vSueny, | : qe E ro) e cS ‘esurqny, | ‘nuevmeon yy | “JUNG AT “ON ‘NUvUeUN ‘eae JO"M'N | : ‘nyejyeny Jos | ‘vaed JO “N ‘nyeyeny jo *s | @ UP) “Guanes ® ‘3 | “ ge ‘esueqny, JO" | Po S Lk n= ! S| & ] ce | cs |) Ge] le | eo | oo | to] & Ship oO | S|°o © | ; oO EB |) Te Or ne | [is | Sh 3 $ | : S 3 3 3 3 3 a *(panuryjuod) SWLOILF OOS OF OT Woszy Jooy YNJVUNY Ot} opis}Mo pUNoF vaozrurcuvtoy, OY} JO STAAL 400 401 BORAMINIFERA OF THE FUNAFUTI ATOLL. “TA tebe arte et eee tenensastaesersecesares ‘deyg ‘nsoUuawn “ ener eee tet e ee Penne teens scererses Apragy “gsnqo.b «6 Se Sn Cn] ssnoyy yD qD)Up ce PavaleeTeeiolss telesteztaiuniseselserse oils see: “povaLiod “ee DS DHPO costal Tabs su BuRonOU AR USE ToOOOONGH (iN) 0) “mmjound “ G90 LODE DUGGEHSHASBo00 0000005 FST ‘sapro1wmjngva) mULanjog serveveeeresscusssreessereeecses (Knmter) pgnquiny Dummfug seer reer es eeerrve Be SZ OY DUDISHAQIALY IY 46 “ godaq, ‘msowpnbsqns nuynhur4 TIEM. ‘227700W00 ie SposmuDD AV OKEI a ‘s9.002QNS “ QtO,p ‘wuassrzunhaya < sreeeseseeee Kovtgy ‘s77ava “RA “QIO,p ‘sunbaja nuuyng see eee eee ener tones tees Voy MET ‘norspunha 73 qIO,P ‘srenwrwW09 a2 sivelucinnsiiesleio Dosey ues Ser @) iO ‘saupynbun DUynnD)D vet eeeeetteeentcteseesetasecnreeserie CTIEAA) vosnf'munnagn 4 tebe n eee een ener seereteeeees CA OTT ‘ds “DpUNjOd tee Eeeeeeeeesseeeseerseseee SA OTT ‘ds § pypnar CG qiO,p ‘wsobns viurhupny+ SPsnooridr sopcasasHBOeeSoBOOHEO appa) ‘psomauds DUIPNAULAA peters erernceeee (ssnoyy) psohns CG wae eneeseee Apeag. ‘puafrvoydis c seesesesesesces Kipter ‘nsogngsif ‘AVA us 8 eee meet e tere eet e et eeeeee oOUVAJO(T “nyngaubns (a3 sete erence reee Aprag “y9atLod 73 6 qto,P ‘sunurgnp bho shee eet eee e reece ete este erences “qIO.~P “wauUnsb 73 ete ew ee ee tee ee eee eee haere eee *q40,p “SnYyIowp 3 eee ee we eee eee tee Peele aaah aye “99VUW09 iG slofesirieleeseiy site sive Seite SMO TET ‘DUUDSHIQSUDLI ce Jase soccer ene O TS “pyndsi.a 13 Sa F Sioned TOT ‘mapaIwoan « eee eee rete ernst eeetnroe ‘OI ‘601 ‘S01 “LOT ‘901 “GOT TOL ‘SOL GOT “TOT ‘OOT ‘66 "86 16 ‘96 “G6 76 “66 CO 16 ‘06 68 ‘88 MR. F. CHAPMAN ON THE 402 om Se = > Be ‘esueqn J, 006 “ OSGi ‘nuvmeun | > ‘nueweun 7 | “ 0s-eT1 ‘eauemy, | “09 ‘ngejeny jo s | “ 09-0¢ ‘eSueqny, | 0g “ OF ‘epeyeny jo - | : “\9e “eae jo “aun | "(panurjuos) SULOYIL OOS] OF OL WoAZ Foor, NJvUN,T 4} OpIsyNO punoj viofrurueloy oy JO o[quy, a | ce ‘esueyny, fo oe “‘eAeg JO "N | cg a3 nee ‘eat 30 ANN | : ‘ G&S a3 GG “ce seahingine Dasioee ease sielis nese send (ORIG ly) 52.0708 Aree ee eee eee eter een etre esers oe (qosyeq ) To) DIMDSOPONT SOc er icici eee “f y oa “‘DIDUUD.LDUL-O DI Dil ve ‘nqejeny jo "¢ SUP HT Marg JON] Pi iiiiiiii: Sn ee ey Peer etree teem eee (‘qtQ,P) srutof yy ee tte eee Bnet e tweeter e tees “qlQ,P ‘SUMUULOD ce GiO,P VUtdqgosioo yg eae “QiO,P “DynQuir) a mens aTG p ‘agpuadnod sevsieisnsiyiessiriclsselss ss USOC Oa MONT “1.daUulaoey 3 Speieisisalaisis eilsitauae sells gece STO NT “‘praqajd ce [sence eens ceer ects eseseveseere LEJSOATIG ‘puxoud C6 sete escceeesees Apeagy ‘ounwodas “ABA ‘c ts G 3 “TITAN ‘sopoowabny |(alatstatelstafetatsistatetate Usleisirs seis ele leuieecee Ay ua of ‘seyp.ads 9 oe teeta teeter estes eset serseseeecce ssnayy ‘pjsoarjnav BE HROB oon coos sporinacenobratiiadesnoaub aGii¢cy ih} Sy pad) nuabory Peete eee eeeneeeresessseescerer ssney “yap10a0 DyaULo)sopray/y inte (eteluislsteisteieisisiateisiatsvatsiaicisVateiatnints ssneyy DIDLWS pirhioqualy sy, peiapaoBdaannaaRSPAADASAO. yyy (999) d0Y 710 ) ‘6 ee ee i et *q1Q,p “‘nyphia) it npbA000 poppesopect (<(a(1(@) 10) NSsmla sreseeeesseses- BZUOMdeg “lgnjna DULINPISS)D “OULB ‘ON 403 FORAMINIFLERA OF THE FUNAFUTE ATOLL. iz > GN . wee wee "pg ‘snunydns munabog Bo ine “pg aa4Lod sent eesonees “TEAN ‘nsopnhup “ ves Kye ‘npdnucaqua & sees “7 “onuadsp nursahro) * Apeag “‘nunzuanbayy s ‘G4O.P ‘vssaudiuoo ~ ‘tO, p ‘svwnewewoo ee =" qio.p ‘j7nb TELE Bi ‘sou ‘ds ‘ SELQv.Laue “es (tLtOg) swag 2 CW RA) ynprdouo s “ ssnayy ‘2yanquajyog s RandooonHH80000 * ssnoyy ‘pansnjgo 66 “ ssneqy ‘77077200).00 a * "QIO,P ‘srusofiuas e “"" ssnay ‘s22qnicnn is Theme reese esse ett reeseeroen *q19,P *S2.00) 1010.10 6 nYatololeratetsVatolelafehatetelele\ialelvisteiris}efe}s/aye (ju0 yy) DID] Ni? a PICCODDOXG s)s)e/s)»\s\e/s\e\0 +)=/e (0/151 10) Cuter) gD IN Jou GG Geceie epee eis HCO), p “nqqub 01.0) 794824) pacer "(qUOyT) s2emarap as (guury) wamnha, nuynuhn 4 Peete ees e reer e eres eoesseece “**(qosyeq ) D1D)809 ce sees vee ‘ssnoy “Dynjos 66 “"GO,p ‘vugnyh nuynnub.wopy (FL “Sy '9€ Td 22s) Apatet ‘ayanjnypods CARED U Os] ere TOMIURED “pigwas “VV ss “(ET ST "QE [Tq 908) “G10, p ‘vzpU2Wnd puynbuery COCCI ICI i iri ei ici ie ir ee beeee (mwa) pnbygo Henn peewee rere steer ene (SoGeodopcance wees teenere oComuiay | SLLDINI IIA IUL (73 Jee cercens seecceesessecses (qosyeq) s SYDGI)L00 ee ‘SnaN ‘vyojnayYUnoqns s “GOT POL “COT ‘COL TOT ‘O9L ‘6ST ‘891 LET ‘OGT 'GeT nasi! ‘691 CG) TSI ‘OST ‘GFL ial! ‘LPI wal “GET PET “SPL ‘CPI “TRI ‘OFT ‘6ET meteas “LET ‘9ET “Gel FEI Retea MR. F. CHAPMAN ON THE ‘esuvyny, | 006 jira anspor ornanbuquecacoDDonCoDbone Apeag: “‘ppquiy “A 1-CA lsowaansc where paoetins Seater ii ‘deyo “nuabuuds « “5 oa “i cA [fats sirtsane won ccUKit te) “ngoQUiay- ojna.waqny ss BAG abet a i Lens SOM COrOOL ROC OOD Sk noe Apeag. ‘sypnhaur 1G . ae | see eeenee see e ees e cares Sree sop oe ty “pundrare nurpetudgy aa styl ciel seibs s{Keecuire. itecepe soodiel poorinubod.cp.cy Hone onomatons eR “‘pynjnaopnbygo DYUITINT - “T ea | Gg. |p Gu |) “obo ||" om 2am itbooton etree tees et eee ween “fp Ny cal ‘suaosvyap purprouwwydy ares AWN noo theooo- 4] oom |) coc ooo dsp onopeooOCnoo Pe ee oe “qlO, p “‘Dpuvy DULvapuUny a sence [cee [ene Heer Mebae /lecee|peranc||h 00m || "ena! ||GasbrerRac. sononcosmenaoGG feos “40, D ‘vswaarwn munyngsO a5n O oO | ered Wee eee A907 “pigup | AnDODSCSOOS Apeage Ham “ “ A “I sete wee wane ei Apearg “‘D.dof YROIDS ce on TA "9 coo Wi oon tf o- | onn |} 90o. |lpognooconnoooo o vee eeenenes Apeg “nynqophwuoo 14 “7A ah “CA ere wee eee e sewer cetee ewes Apeag *s2) dag) INDO GG . see see A “A aielaiejnle/stotals(=|ain/sis}alalejaia\alernis “AOU ‘ds ‘MoI )0L0QNS GG oe os “ayy, ‘vudaphyood < *7 "Qi, p ‘2a0,0ajny = coalieg 00t TA me AO, P “.LQNM e tA 4 A Beir 4 ee canon stews ssnoyy gee "IRA oe “ fue alllars “7 “1 : “qIQ,p ‘saproqyng vurvahrqop5 Navas , ventral edge view. x 465. 3. Cythere phylloides, sp. noy.: a, left valve; 6, dorsal aspect ; c, posterior view. x 40. 4. Xestoleberis acuminalis, sp. noy.: a, right valye; 6, ventral aspect; ¢, posterior aspect. x 90. . Loxoconcha tumida, sp. nov.: a, right valve; 6, ventral aspect; c, pos- terior view. X 45. 6. Bythocythere armata, sp. nov.: a, left valve; 6, dorsal aspect. x 45. on LINN. JOURN.—ZOOLOGY, VOL. XXVIII. 3] 434 MISS LETTICE DIGBY ON THE STRUCTURE OF On the Structure and Affinities of the Tanganyika Gastropods Ohytra and Limnotrochus. By Lerrice Dresy. (From the Biological Laboratory, Royal College of Science.) (Communicated by Prof. G. B. Howss, F.R.S., Sec. Linn. Soc.) [Read 20th February, 1902. ] (PLates 38-40.) Turs paper is based on material which formed part of the collection brought by Mr. J. E.S. Moore from Lake Tanganyika. He has kindly placed at my disposal specimens of both Chytra Kirkii and Limnotrochus Thomsoni, and has greatly helped me in my work. CHYTRA. The genus Chytra was separated by Moore from Smith’s original Limnotrochus (4. p. 307), the older generic name. being reserved for the reception of the single species Limnotrochus Thomsoni. The genus is now represented by a single species, Ohytra Kirkii (=L. Kirkii, Sm.), and is one of the most characteristic forms among the group of peculiar operculate molluscs found only in Lake Tanganyika. It has been generally placed among the Hydrobiide (ef Fischer 2.), a family often used as a receptacle for curious and abnormal types which have not been satisfactorily classed. But in 1897, Moore (4. p. 807) regarded it as more probably belonging to the Xenophoridex, on account of its conchological similarity to numerous fossils which are referred to that group. Up to the present time, however, the anatomy of the animal has never been described. External Features.—The shell (Pl. 38. figs. 1 & 2) has already been described by Smith (8. p. 286). The horny operculum (fig. 3) is of the Littorinoid type, the inner surface being convex and the outer concave. The tentacles are long and filiform, and the eyes are situated on tubercles at their bases. Nervous System.—The nervous system seems to be transitional between the dialyneurous and the zygoneurous types (Pl. 39. fig. 14). The cerebral ganglia (g.c.) are closely approximated, and the pleural ganglia (g.p.) are separated from them by a THE GASTROPODS CHYTRA AND LIMNOTROCHUS. 435 shghbt{constriction. The supra-intestinal cord (.sp.) is consider- ably elongated ; from the supra-intestinal ganglion (g.sp.) arise two nerves, the left pallial nerve (n.p.’) and the visceral (2.v.’). A very fine nerve (¢.p.’) connects the left pleural ganglion with the pallial nerve, which it joins soon after its origin from the supra-intestinal. ganglion. The sub-intestinal cord (m.si.) is somewhat shortened. In like manner the sub-intestinal ganglion (g-st.) gives rise to the right pallial (x.p.”) and the right visceral nerve (z.v.'). A nerve connects the right pleural ganglion and the right pallial nerve (c¢.p.”). The visceral nerves (x.v.’, nv.) unite by a straight commissure at the base of the rmantle- cavity. The pleuro-pedal and cerebro-pedal connectives are long (fig. 15).. The pedal ganglia (g.pe.) each give off three nerves, two of which run anteriorly and one posteriorly. The otoeysts lie close behind the pedal ganglia; they are filled with many barrel-shaped otoliths (fig. 16) of various sizes. Radula.—The radula (PI. 39. fig. 20) has a feature which is also a characteristic distinctive of the radula of Hipponyx conicus (cf. 9. vol. i. p. 162), there being a strong pointed projection on the admedian tooth which overhangs the inner end of its serrated edge: .The lateral teeth are sharply pointed and serrated. Possibly this acute form of dentition indicates a carnivorous habit. Viscera—YVhe mouth is carried on a short snout, and opens ~ into the buccal mass (PI. 39. fig. 13) (6.m.), which leads into a long and very slightly coiled cesophagus (@.). The stomach (s¢.) is two-chamnbered (Pl. 39. fig. 138 & Pl. 40. fig. 25). The anterior stomachic chamber contains a crystalline style (s.c.); the. posterior bears a small, but complete spiral cecum (ce.). The intestine (¢.) arises from the lower portion of the anterior -chamber of the stomach, and coiling twice, bends sharply forward. The rectum (7.) attains a considerable dimension, since its walls are very glandular, but it narrows suddenly before opening by the anus (a.), which has a circular, thickened rim. The “liver” (/.) occupies a great portion of the visceral mass, anda bile-duct opens at 0. The kidney (#%.) is. large, and the renal aperture (a@.7.) is far back in the mantle-cavity (Pl. 39. figs. 12 & 13). The heart (v. & aur.) (Pl. 39. fig. 13) is of the ordinary Monotoeardian' type. . The gili (Pl. 39. fig. 13, 9.) extends througheut: the mantle-cavity, and consists (Pl. 38. fig. 10) ; 31* 436 MISS LETTICE DIGBY ON THE STRUCTURE OF of a single row of plumes (g.’), with blunt rounded ends: (Pl. 39. fig. 17), each of which is attached along its dorsal edge, and contains a skeletal support continued from the point of attachment. Hypobranchial glands are present (q.h., Pl. 39. fig. 12). . The reproductive aperture (a.g.) (Pl. 39. fig. 12) is a wide. oval slit. This, in the female (Pl. 38. fig. 11), leads into a large dilated uterus (2.), which is continued to the posterior end of the- mantle-cavity, where it is seen to arise from an oviduct (ov.), which ramifies in the genital tissue (g.g.). This tissue (Pl. 39. fig. 13)) is intimately bound up with the “liver”’ (/.), and with it composes. the greater part of the visceral hump. There is an accessory gland-like organ (9.g.', Pl. 38. fig. 11): closely adhering to the uterus (w.), which protrudes into the- mantle-cavity. Its function is unknown, but it may possibly: secrete an investment for the eggs. The reproductive organs: of the male have not been identified, and it is not known whether the accessory gland-like organ is present in the male, or- whether Chytra Kirkii is viviparous or not. Affinities.—In attempting to define the affinities of Chytra, by far the most important feature presented by this genus is the- obvious combination of characters distinctive of several well- known Prosobranchiate forms. Thus the nervous system is. strikingly like that of Capulus (1. pl. 8. fig. 35). So also, when viewed from the side, the position and relationships of the pedal, pleural, and cerebral ganglia, with their connectives, are very similar to those figured by Bouvier for Xenophorus. COhytra may be further compared to the Xenophoride in the obviously similar character of its shell (4. pp. 307 & 317, pl. 23. fig. 6a), and consequently the shape of the body; in the- situation of the eyes on the tentacles; in the general character of the gill and gill-plumes; in the filiform osphradium; in the position of the renal aperture; and in the presence of an accessory gland-like organ in relation to the genital apparatus. This apparent affinity of Chytra to the Xenophoride is perhaps strengthened by a detailed comparison of Chytra Kirkit with Aporrhais pes-pelecani (cf. Pl. 39. fig. 18 & Pl. 40. fig. 22). In both, there is the same arrangement of the buccal mass, a more or less comparable radula (Pl. 39. fig. 20), a very similar eill, and hypobranchial glands (9.h., Pl. 40. fig. 22). Further, the characters of the nervous system of Aporrhais pes=pelecani THE GASTROPODS CHYTRA AND LIMNOTROCHUS. 437 (Pl. 40. fig. 28) are generally similar to those of Chytra Kirki *, except that in Aporrhais the supra- and sub-intestinal cords are considerably elongated, and there is a direct zygoneurous connection on the right side. In Chytra Kirkii there is, however, also a practically zygoneurous condition, the connecting nerve, instead of joining the sub-intestinal ganglion (PI. 39. fig. 14), joins the pallial nerve immediately after its origin. Lastly, the whole stomachic apparatus is similar: there is a_ well-developed erystalline style in Aporrhais (st., Pl. 40. fig. 22), and the internal valvular portion of the spiral cecum is present, but not the complete structure. Further, the barrel-shaped otoliths which are found in Chytra are also found in Typhobia (6. p. 188) and in the Cerithiide. The presence of the slightly developed spiral eexeum in Chytra brings this interesting form into relationship with Trochus (7. & Pl. 40. fig. 27), Plewrotomaria (fig. 28), and Nassopsis (5. p. 189, & Pl. 40. fig. 26). The presence of a crystal- line style characteristic of Chytra is characteristic of widely different molluscan forms. It is present in all the halolimnic Gastropods of Tanganyika, in some Trochide, Pteroceras, and the Strombide. These facts suggest that in Chytra we are probably dealing with the direct ancestor of the families Hipponycide and Capulide, since the genus combines the nervous system of the one and the radula of the other; 2.e¢., two of the most salient features of Prosobranchiate anatomy. Limnorrocuus THomsonr. (PI. 38. figs. 1-9; Pl. 40. fig. 24.) Limnotrochus Thomsont is now the single species of the genus (1). Like Chytra, it is an inhabitant of Lake Tanganyika, living at considerable depths and apparently in solitude, for never more than a single specimen is dredged at a time. Con- sequently it is difficult to procure the living animal, but the empty shells are pretty common. The conchological characters (Pl. 38. figs. 1, 2) presented by the genus, however, need no comment here, as they have already been fully described and figured (8. p. 285). The operculum (PI. 38. fig. 3) is “ horny ” and somewhat oblong in shape, concave towards the inner side, and Littorinoid in character. * Haller’s figure (3. taf. xvili. fig. 6) of the nervous system of Chenopus (Aporrhais) pes-pelecani and his description of the same (p. 579) is absolutely incorrect. He must either have ficured an entirely different animal, or the specimen that he dissected was quite abnormal, 438 MISS LETTICE DIGBY ON THE STRUCTURE OF External Features.—The general features of the mantle-cavity (Pl. 38. fig. 9), the character of the tentacles, the position of the eyes and of the apertures, are all very like those already deseribed in Chytra, the most apparent differences being the smaller size of Limnotrochus Thomsoni, the flatter shape of the visceral hump, and the pigmented band down the centre of the snout and on the inner sides of the tentacles. Nervous System.—The nervous system is zygoneurous on the right (Pl. 38. figs. 5,6). The ganglia and nerves are protected by a closely adhering connective tissue which contains skeletal elements. The cerebral ganglia (¢.c.) are separated by a distinct cerebral commissure, and are situated decidedly further back than in Chytra. The right pleural ganglion (4.p.') gives off a fairly long supra-intestinal cord, which connects the right pleural ganglion with the supra-intestinal ganglion. A fine nerve comes off from this cord just before it unites with the supra-intestinal ganglion. This nerve appears to join with another arising from the left pleural ganglion, but I was not able, actually, to see the connection. The supra-intestinal ganglion (g.sp.) gives off the left visceral nerve. The sub-intestinal cord springs from the left pleural ganglion, and soon passes into the sub-intestinal ganglion (.s2.) which, in the usual way, gives off the right pallial and right visceral nerves. A very short nerve (z) connects the sub-intestinal ganglion directly with the right pleural. Viewed from the side (fig. 6), the relative position of the cerebral and pleural ganglia is rather curious, the pleural ganglia being posterior and ventral to the cerebrals and nearly fused with the pedal ganglia, from which they are separated by a very short pleuro-pedal commissure (c.p.pe.'), the cerebro-pedal commissure (c.c.pe’.) beg longer. Each pedal ganglion gives off two nerves to the foot, but I was not able to see the nerve leading to the otocysts. The otocysts contain many rectangular otoliths (Pl. 38. fig. 7), Radula.—The radula (PI. 38. fig. 4) is distinctive, its most striking feature being the blunt protuberance on the underside of each of the two lateral teeth. The admedian tooth is large, and has a broad upper portion, bluntly serrated along its anterior face, which overhangs a lower portion, whose sharply serrated eee) is turned towards the inedian tooth. Viscera.—The mouth leads into a short buccal mass, A pair of salivary glands, diminutive sacculated organs, open into the THE GASTROPODS CHYTRA AND LIMNOTROCHUS. 439 buccal cavity. The cesophagus is straight, except for a sharp bend before it opens into the stomach (PI. 40. fig. 24), which, like that of Chytra, has two chambers, the anterior of which contains a large crystalline style. The walls of the posterior chamber are very thin; those of the anterior chamber very thick and muscular. The most remarkable feature about the gastric apparatus of Limnotrochus lies, however, in the fact that it possesses an even better developed spiral cecum (ce., Pl. 40. tig. 24) than Chytra (fig. 25) itself. The base of the cecum is connected with a longitudinal fold which extends to the anterior chamber; the aperture of a bile-duct (d.) lies below the spiral cecum. The coils of the intestine, the rectum, and the anus are very similar to those of Ohytra. The kidney is fairly large, aud occupies the same position as in Chyéra, surrounding the ‘“‘monotocardian ”’ heart. The gill (g., Pl. 38. fig. 9) is short, the gill-plumes are triangular, and the osphradium (o.) is filiform. In both specimens which I dissected, the long slit-like aperture of the genital gland (PI. 38. fig. 9, a.g.) opened into a considerably dilated sac, slightly curved towards tie left side of the animal. In the case of one specimen this was undoubtedly the uterus, as it contained bundles of very long spermatozoa ; and in the genital tissue of the visceral coil there were well-developed ova. There is a large accessory gland-like organ (fig. 9, g.g.') like that of Chytra, which spreads posteriorly into a tongue-shaped body, partly underlying the uterus (cf fig. 8). Affinities.—It will have been gathered from the preceding description that Limnotrochus Thomsoni, in the general plan of its organization, and in the disposition of the viscera, is distinctly like the genus Chytra, but in minor details very different. The shell of Limnotrochus Thomsoni (Pl. 38. figs. 1, 2) may bé compared to some of the so-called Littorinas of the marine Jurassic deposits, and in particular to Littorina sulcata (4. p. 317). It is also dissimilar to that of Cancellaria. Curiously enough, the nervous system of Limnotrochus (Pl. 38. fig. 5) is also like that of Voluta (3) and Cancel- laria, there being the same condensation of the cerebral, pleural, and pedal ganglia, and the same zygoneurous condition of the right side. The radula (Pl. 38. fig. 4) cannot be associated with any known type, its nearest approach is to be found in the varieties of the Melanoid group (9. vol. i. p. 121). ‘440 MISS LETTICE DIGBY ON THE STRUCTURE OF Lastly, the pigmented snout, the triangular gill-plumes, so characteristic of the Rhachiglossa, the longitudinal fold and the other features of the stomach, and the simple pouch-like salivary glands, all bring Limnotrochus into close relationship with Typhobia. Like Ohytra, Limnotrochus has thus obviously scattered affinities, and it is more difficult to place than even Chytra itself. It is undoubtedly allied both to Chytra and to Typhobia, as weil as to the Stromboid group represented by the genera Strombus and Aporrhais; and it is perhaps more nearly allied to these genera than to any other living types. But, on the other hand, it is quite distinct from them all, and must, at any rate for the present, be regarded as unique. List of Papers cited in the Text. 1. Bouvier, E. L.—‘‘Systéme Nerveux, Morphologie Générale et Classification des Gastéropodes Prosobranches.’’ Annales des Sciences Naturelles, Zool. 7° série, pp. 1-510 (1887). 2. Fiscuer, P.—Manuel de Conchyliologie et de Paléontologie Conchyliologique. Paris, 1877. 3. Hatter, B.—“ Die Morphologie der Prosobranchier.” Mor- pholg. Jahrb. vol. xix. pp. 553-589. 4. Moors, J. E. S.—“ On the Hypothesis that Lake Tanganyika represents an old Jurassic Sea.” Qu. Journ. Micr. Sci. vol. xli. pp. 803-321. 5. Moorz, J. E. S.—‘‘ Molluses of the Great African Lakes. Part 1V. Wassopsis and Bythoceras.” Qu. Journ. Micr. Sci. vol. xli. pp. 187—200. 6. Moors, J. E. S.—“ Molluscs of the Great African Lakes. Part I. Typhobia and Bathanalia.” Qu. Journ. Micr. Sci. vol. xli. pp. 181-192. 7. Ropert, A.—‘‘ Gastéropodes: Le Troque.” ‘ Zoologie Descriptive,’ Paris, vol. ii. pp. 381-415. 8. Smirnu, E. A.—“On a Collection of Shells from Lakes Tanganyika and Nyassa, and other Localities in Hast Africa.” Proc. Zool. Soc. 1881, pp. 276-300. 9. TroscueL, F. H.—‘“ Das Gebiss der Schnecken.” Band i. Berlin, 1856-1863. THE GASTROPODS CHYTRA AND LIMNOTROCHUS. 441 EXPLANATION OF THE PLATES. Reference Letters. @., anus. &., aperture of the bile-duct. a.@., opening of the cesophagus into the stomach. ag., genital aperture. a.i., opening of intestine leading from the stomach. a7r., renal aperture. au., auricle. b.7., bueeal mass. ¢@., spiral cecum. c.¢.pe.', left cerebro-pedal connective. ¢.p., pericardial cavity. c.p.', left pleural connective. c.p.", right pleural connective. c.p.pe.', pleuro-pedal connective. Greve: g- gill. g.', gill-plume. g.c.', left cerebral ganglion. g.c."', right cerebral ganglion. g.e., cut edge of gill. g-G-, genital gland. g.g.', accessory genital gland. g-h., hypobranchial gland. g-p.', left pleural ganglion. g-p.'', vight pleural ganglion. g-pe., pedal ganglion. Fig. 1. Limnotrochus Thomson. 2. y » 3. ted 2»? 4h 29 29 D. s aA 6. 19 ? ce ” ” 8. 6 2 9. ” 9 10. Chytra Kirhkii. 11. cS i The gill, with mantle reflected to the left. Female reproductive organs. g.8., salivary glands. g.si.. Sub-intestinal ganglion. g-Sp., Supra-intestinal ganglion. 9-9-5 genital groove. 2., Intestine. k., kidney. 1., liver. nd., mantle. m.e., cut edge of mantle. n.p.', left pallial nerve. n.p.", right pallial nerve. ?.St., Sub-intestinal nerve. 2.Sp., Supra-intestinal nerve. | 2.v.', lett visceral nerve. | n.v.", right visceral nerve. | o., osphradium. @., esophagus. ot., otocyst. ov., oviduct. 7., rectum. ’ s., snout. s.c., crystalline style. S.S., Style-sac. st., stomach. é., tentacle. @, uterus. v., ventricle. 2., Zygoneurous connection. Prats 38. Front view of shell. Back view of shell. Both natural size. Operculum, external face. x 6. Radula, 2 laterals, admedian and median teeth ; from a drawing by Mr. J. E. 8. Moore. Magnified. Nervous system, dorsal aspect. ud lateral aspect. Both x10. Otolith. Highly maguified. Female reproductive organs. X 7. Mantle-cavity, with mantle reflected to the myelin, SK Pe x 12. x 6. 442 Fig. xFig. ON THE GASTROPODS CHYTRA AND LIMNOTROCHUS. Puare 39. 12. Chytra Kirkit. Mantle-cavity, with mantle reflected to the right side. x 10. Is! A Fy Dissection of the alimentary canal. x 10. A or ae se Nervous system, dorsal aspect. x 10. ib). in = 7 lateral aspect. > 10. 16. % * Otolith. Highly magnified. 17. 3 By Gill-leaves, isolated. Highly magnified. 18. s yi The shell, lateral aspect. 19. 3 Fs i from beneath. Both slightly enlarged. 20. a : Radula, 2 laterals, admedian and median,teeth ; from a drawing by Mr. J. E. S. Moore. Magnified. 21. 8 BA Operculum, external face. x 6. Prats 40. 22. Aporrhais pes-pelecani. General dissection. 2. 23. 4 Ps Nervous system from, \the} dorsal aspect. x about 23. 24. Limnotrochus Thomsoni. Dissection of the stomach. x 10. 25. Chytra Kirkii. Thesame. xX about 10. 26. Nassopsis nassa. Dissection of the stomach. x 10. 27. Trochus turbinatus. The same, after Robert (7). x 10. 28. Pleurotomaria Beryrichti, The same. From a drawing by the late Mr. F. Woodward. x10. * The figures marked thus are from drawings by Mr. J. E. 8. Moore. Ln. Soc. Journ. Zoor,. Vou. XXVITI Pr.38. L.D. del. MP. Parker lith. Parker & Wast imp. LIMNOTROCHUS & (GIBIN? ACIS. Linn, Soc. Journ. Zoov. Nou DOOV MIM roe: WY NG L.D. del. MP. Parker lith. Parker & West imp. CHY TRA. Digby. Li. Soc. Journ. Zoon. Vou. XXVIII Pr. 40. MP. Parker lith. Parker & West imp. MeiEOMmNOG LUIS dé CH YN LEAS WITH APORRHAIS, NASSOPSIS, PLEUROTOMARIA, & TROCHUS. ON THE BRAIN OF MACROSCELIDES. 445, Notes on the Brain of Macroscelides and other Lusectivora. By G. Exxtor Suiru, M.D., Fellow of St. John’s College, Cambridge; Professor of Anatomy, Egyptian Government School of, Medicine, Cairo. (Communicated by Prof. G. B. Howes, F.R.S., Sec. L.S.) [Read 1st May, 1902.] My friend Dr. Robert Broom has recently discovered that the organ of Jacobson and its cartilages in the Elephant-Shrew present a peculiarly close similarity to the corresponding parts in the Marsupialia, and has moreover found* metatheroid features in the skeleton of Jlacroscelides in addition to those previously recorded by Kitchen Parker. It seemed to him that it would be of considerable interest to submit to careful examination the other parts of the body which present distinctive features in the Marsupialia. Accordingly he has kindly sent me the heads of an adult and a foetal Jlacroscelides proboscideus, and asked me to make a report upon them. The heads had been simply placed in spirit, so that the brains were not in a condition altogether favourable for histological study. In fact the fetal brain was too soft to permit of auything more than a study of the configuration of its surface. The adult brain, however, was sufficiently firm to be cut in paraffin. The sections were stained with lithium-carmine. As the presence or absence of metatheroid features can in almost all mammalian brains be detected by the naked eye, I first submitted the brain to a thorough examination both by this means and with the help of a lens. I then split it in the mesial sagittal section and studied the mesial surface in the same way ; aud as many points still needed elucidation, I cut a series of coronal sections of one hemisphere and dissected the other. The brain of Macroscelides has been figured from the dorsal aspect by Peters tT; and its general features need not be described in detail, since they differ to no great extent from those of Talpa, which have been so thoroughly described in Ganser’s classic monograph ¢. * Proc. Zool. Soc. 1902, vol. i., March, 18th. t Peters, ‘Reise nach Mossambique,’ Zool. i. Saugethiere, pl. xxiv, fig. 13. Berlin, 1852. t Vergleich, Anat. Studien. ‘“ Ueber das Gehirn d. Maulwurfs,’’ Morph. Jahrb. Bd. vii. p. 591 (1881). AAA PROF. G. ELLIOT SMITH ON The large projecting olfactory bulbs recall those of the Marsupial Perameles. The features of the base of the brain— the large tubercula olfactoria, the tracti olfactorii and their large tubercles, the peculiar flattening of the nuclei amygdala, Tes 1s Macroscelides probossideus: the brain, X 2. (a) Dorsal aspect. (4) Ventral aspect. (¢) Left lateral aspect. b.0., bulbus olfactorius. cb., cerebellum. c.g., corpora quadrigemina, (7, fissura rhinalis. 7.f., lobus floceuli. J/.p., lobus pyriformis. .0., medulla oblongata. .a., nucleus amygdale. ¢.@., tuberculum acusticum. 7.0., tractus olfactorius. 7.¢., tuberculum tractus olfactorius, (7z., trapezinm. v., pons Varolii. the small pons Varolii, and the large trapezoid bodies—all closely resemble those which are found equally in the Insectivora, Polyprodont Marsupials, and the Dasypodide. The peculiar lateral extension of the pyriform lobes of the hemispheres into THE BRAIN OF MACROSCELIDES. 44.5. a distinct angle occurs also in Hrinaceus, Talpa, and Perameles. So far as the shape of the cerebral hemispheres, and especially also that of the floccular lobes of the cerebellum, is concerned, the brain more closely resembles that of Perameles than that of the Insectivora. At the same time, distinctions of this kind have little if any ordinal value. The rhinal fissure is incomplete, being intermediate in this respect between Hvrinaceus and Perameles. The caudal margin of each hemisphere (fig. 1@) consists of two limbs of equal length meeting at an angle of slightly more than 90°. A large lozenge-shaped space is left between the mesial half of each hemisphere and the cerebellum, and*in this the pineal body, the whole of the anterior and a considerable part of the posterior pair of corpora quadrigemina are exposed. Such a complete uncovering of the mid-brain is rare in mammals. A much slighter degree of uncovering occurs in the Marsupial Dasyurus and some of the smaller Dasyuride. . Diameter of the eyes twice as long as the distance between their lower edge and the base of the tubercle. Tibize of palps twice AS oN? OS |ORORNGL Be hoo No aone coerce tOB ta Suscus. 6°. Diameter of the eyes equal to the distance from their lower edge to the base. Tibize of palps three times as long as broad .......- Thorellit. 6!, Palps smooth....... 5 ito SS\0.0t, UL LORE cee ae aaa . minutus. Gmebemmoraot the legs Maly yes cia pele le) eels =e 666 hirsutus. ZALEPTUS SUBCUPREUS, Thor. 1889. Thorell, (10) 1889, p. 609. The row of tarsal granules is wanting in the males. Two specimens from Kollads Salween Hill and Hungdarow. ZALEPTUS FESTIVUS, Thor. 1889. Thorell, (10) 1889, p. 611. The tarsal row of granules is wanting in the males. ZALEPTUS SULPHUREUS, Thor. 1889. Thorell, (10) 1889, p. 614. Hungdarow. ZALEPTUS FUSCUS, Li. Sp. @. Cephalothorax almost semicircular. Last two segments of the cephalothorax well developed. Body raised towards the fourth abdominal segment, and the foremost third of the free ventral 450 MR. C. WITH ON PHALANGIID& segments is more raised than the rest, and the body is granular with exception of these. Grooves between the segments of the scutum well marked. The cephalothorax has a well-marked depressed triangle. Frontal eminence distinct ; procursus frontales well separated, obtuse, and dentate. The tubercle is rather low, and seen from the side a little longer than high. The crest is convex. In front it is much narrower towards its base and much broader than high above. The top is deeply grooved and bears in front one single obtuse tooth. Diameter of the eyes as large as the distance between them, and twice as large as the distance between their lower edge and the base of the tubercle. Basal joint of the antenna smooth. Procursus maxillarius internus almost square, as the usual lower branch is missing. The upper branch is pale and slender. ‘The lower edge is granular. Femora of the palps shorter than the tarsi and spinous below and above. Patelle a little broader towards their points and strongly toothed. Tibise twice as long as broad and inwardly spinous. Femora of the legs spinous. F. I=62 (leg=30), I1=114 (54), ITI=63 (283), LV (): Body 53 mm. The second joint of the antenna measures 14 mm. Colour of the body black. Space between the eyes and the ventral segments lighter. Articulations yellow. Antenne, the tarsi, and the points of the tibie of the palps yellow. Legs blackish, with the base of the femora and points of the tarsi lighter. Extremities of the second pair of tibize white. One female from Calcutta. ZALEPTUS THORELLII, n. sp. 3, 2. Cephalothorax semicircular. Body straight, the upper- side and the cox@ are finely granular. Ventral segments, some parts of the cephalothorax, and the lamina frontalis smooth. Cephalothorax bears a well-marked depressed triangle, open in front. Lamina frontalis undeveloped; the procursus frontales are large, broad, and strongly toothed. The tubercle is rather low, the upperside somewhat convex ; the front and back almost straight, and seen from the side it is as long as high. It is surmounted by a single tooth behind, and four in front. Seen from the frout it is broader than high and scarcely narrowed towards its base. The top is deeply channelled. Diameter of the eyes equals the distance between the lower edge of the eyes and the base of the tubercle. FROM THE INDIAN REGION. 481 Basal joint of the antenna smooth. Procursus max. internus broad and high, and has an upper process: the lower process is missing, and the lower edge is toothed. Femora of the palps toothed below; patelle toothed chiefly above, broader towards their edges. Tibiw three times as long as broad. ‘Tarsi at their lower and inner edges bear the usual row of granules. Femora of the legs and partly their patelle and tibiz spinous. F. I=6 (80), I[=104 (58), I1I=6 (80), IV=83 (?) mm. Body 4mm. Second joint of the antenna 1 mm. in length. Cephalothorax yellowish, richly spotted with brown. Abdomen brownish with a metallic sheen. Genital plate and coxe brown ; the free ventral segments are greyish-brown, indistinctly spotted with white and black. Femora and the patelle of the palps brownish, the terminal joints and the antenna yellow. Legs light brown, with the extremities of tibie 2 and 4 white. In the females the tarsi of the palps are smooth, the abdomen lighter and non-metallic ; the dorsal segments bear a row of indistinct yellow spots. Male and female from Berbhoom district. ZALEPTUS MINUTUS, 0. sp. @. Cephalothorax almost triangular. Abdomen flat, and. falls gradually behind. Cephalothorax has the ordinary depressed triangle; and there is a longitudinal groove along the margin. Body bears few hairs and bas large granules above; the coxe and genital plate are almost, and the ventral segments com- pletely smooth. Lamina frontalis well developed; the procursus are large, spinous, and separated. The tubercle is low and as high as long; the front edge higher than the back. Seen from the front it is broader than high and narrowed at its base. The top is convex and smooth, with the exception of a single anterior tooth. Diameter of the eyes equal to the distance between their lower edge and the base and less than the distance between them. Basal joint of the antenna granular. Labrum obtuse and pointed, with two short hairs. Proc. max. int. two-branched and low ; the upper branch slender, the lower strong and recurved. Both upper and lower edge are sometimes toothed. The palps are broken away. Femora of the legs well toothed and hairy; tibie of ail the legs bear false articulations. Legs brittle and femora beset with few bairs. F. I=6, I1=11, I1J=6, IV=9 mm. Body 3mm. Second joint of the antenna measures ? mm. 482 MR. C. WITH ON PHALANGIIDE Abdomen black above: cephalothorax and coxe dark brown; free ventral segments lighter. Antenne and legs reddish-brown. 3. Second dorsal segment bears a low tubercle as in Scoto- menia; except that in this species it is on the first segment. Palps very long and slender. Tibie and pateile much longer than the femora, which are longer than the tarsi. Patelle almost as long as the tibie, enlarged towards their ends. Tibie four times as long as broad. Palps, including the tarsi, smooth ; their length is 33 mm. Colour red-brown above, with a broad longitudinal black band running through the abdomen. Palps yellowish-brown. Colours of the two other specimens are as in the male. Four specimens from Darjeeling. ZALEPTUS HIRSUTUS, U. sp. 2. Cepbalothorax almost trapezoid. Abdomen oval. Body becomes higher towards the second abdominal segment and falls very gradually behind. Segments of the scutum marked by distinct transverse grooves. Free segments separated from each other by broad articulations. Depressed triangle of the cephalothorax not distinct. The upper- side, including the base of the tubercle, bears large flat granules. Coxe almost and the free ventral segments completely smooth. The whole body densely hairy. Lamina frontalis well developed, the procursus small, narrow, and toothed. Tubercle, seen from the side, much higher than long. The front face is the higher and slopes steeply forwards and downwards. The back is the lower and straight. Seen from the front, it is almost as high as broad, and narrowed at its base. Top provided with a low groove and crested with a row of strong teeth. Number of teeth different in each series and not placed in a regular row—but some nearer to, others farther from the middle line. Diameter of the eyes smaller than the distance between them, and one third of the distance between their lower edge and the base of the tubercle. Basal joint of the antenna smooth or almost smooth above and twice as long as broad. Proc. max. int. low and two-branched— the upper branch pale and slender, the lower curved back and strong, with its lower edge bearing a few large teeth. Femora of the palps bear a few small teeth below. The terminal seg- ments are broken away. Legs very brittle; femora almost smooth and hairy. All the tibie have false articulations. FROM THE INDIAN REGION. 483 F. 1=53, W=103, W1=53,1V=7mm. Body4mm. Second joint of the antenna measures 1 mm. Body red-brown, the top of the tubercle black. The underside darker. The articulations are yellow; the legs and antenne brownish. Four females from Darjeeling. GaGRELLA, Stol., 1869. T have already considered the difference between Gagrella and Melanopa on the one side, and Zaleptus on the other (Zaleptus, antea, p. 474). I have retamed the genera Hypsibunus, Zaleptus, and Ceratobunus, because they have a certain practical value in the synonymy of the group. This, however, is not the case with Melanopa. The differences between Gagrella and Melanopa according to Thorell (10. p. 659) are as follows :— (1) The first pair of femora are never more than four times as long as the second joint of the antenne; since, however, in the males of G. atrata, Stol., they exceed this proportion, but do not reach it in the females, the character has no value in the definition of the genera. (2) Femora of the legs in Melanopa not only enlarged towards the extremity from the middle, but from the base. This seems only a necessary consequence of the shorter legs, and is certainly of little value. (3) Fourth pair of coxe relatively broader towards their extremities than the third. I cannot confirm this character in Gagrella Hansenti and G. varians, which are Melanope according to Thorell’s definition. (4) Claw of the palps not dentate. In Gagrella imperator the teeth are almost wanting in some specimens but not in others. In Gagrella Hansenii they are well developed. Since all the above characters are variable and insufficient for diagnosis of new species, I prefer to reject the genus Melanopa and refer its species to Gagrella; and I hope that the following synopsis will be of value, although it has defects, partly due to myself, partly to the vagueness of the original description. Many of the species I do not know, and of most of them I have had but one or two specimens—a very unfortunate fact when they vary, and it becomes necessary to use characters which are partly inconstant. 484 MR. C. WITH ON PHALANGIIDA Synopsis of Species. a. Scutum bears one or two spines in a longitudinal row. a’, First pair of femora more than four times longer than the second joint of the antenna. a’. Tibiz of palps at least four times as long as broad. a’, Frontal eminence provided with a tooth 6°. Frontal eminence smooth. a*. Tubercle smooth or almost smooth above. a’. Tubercle higher than broad and lone . 6°. Tubercle more or less low. a°, Cephalothorax more or less yellow. a’, Abdomen with a _ longitudinal darker band. a*, Cephalothorax yellow, with the exception of an indistinct black band between the tubercle and the front. One brown spine .. 6°. Articulations of the cephalo- thorax black. Two black spines with yellow base....... Ha aon 6". Abdomen without loaeiandiine blacksband terrae shanty teres at 6°. Cephalothorax black or brown. a. Legsyellowish-brown. Twospines. The tubercle is low with a few teeth in front. The scutum is eranular. Joint 2 of antenna= l¢mm. F. [=10; B.=6-63 mm. o°. Legs black. Tubercle high and quite smooth. One spine. The scutum is pitted. Joint 2 of antenna=limm. F.I=83; B.= ABET SMT Cigna etre ore alter ame 6‘. Tubercle toothed. a'°, Tubercle has only afew teeth in front. ©, Tubercle is crested with a longi- tudinal row of teeth. “, Abdomen brown, with two spines and two yellow spots. Second joint of the antenna=13 mm. if = 6-es.— 45-62 minnie : b''. Abdomen brown or black, with one or twospines. The cephalothorax dentata. nobilis. crur, armillata. leprda. chetopus. sordidata. sordidata. minax. FROM THE INDIAN REGION. is sometimes provided with two white waxy spots. Second joint ofantenna=lmm, F.1[=6;B.= GET, so s6occacoeacouNouONC 6°. Tibize of palps not four times as long as broad. a’, Cephalothorax spinous on each side. Tubercle brown. A row of strong teeth aHROMING! NS OVEN Gos caacacasyscecnos0ed 6. Cephalothorax not spinous. a'*, Legs more or less hairy. a. Body brown with a darker longitudinal band through the scutum; antennz and palps yellow, and the legs brown. b'*. Colour black. Antennze and. last two joints of the palps yellowish-brown .. 6, Legs either not hairy or with but few small hairs. a's, Tubercle completely smooth above. a6, Abdomen completely black. a, One or two lighter bands between the tubercle and the front. a. be, Tubercle yellow; one or two spines. A single lighter band between the tubercle and the front. Second joint of the antenna=12 mm. F. I[=9; IS, S13; WO, ogo ge cone 2600080 Tubercle black; two spines. Two yellow bands between the tubercle and thefront. Second joint of the antenna=13 mm. 10, Woes(GF s [BAe WONG 6 503 6“. No lighter band between the tubercle and the front,......... 6°, Abdomen not completely black. a, Abdomen black with a yellowish longitudinal band. Tubercle high and in front granular below. Ventral segments bear transverse rows of granules ............-- 6, Scutum bears a yellowish spot on each side before the spine ...... cl’, Abdomen black with yellow sides. GF, Whole body granular. Smee ett 62, Tubercle, seen from the front, almost twice as broad as high. a**, Tubercle and palps completely SNK saeapoagto.d ho 00 1 0l4oK 6°*, Tubercle armed in front, under the eyes, with a row of strong ee the cae a i aiRueer. mM Bileteceheraclers el®. Body granular, with large and low granules, which are placed distant from each other. Colour greyish, with an abdominal longitudinal darker band ; underside white, with darker spots. Tubercle smooth; Palpstcoothedy treme aie t rrr tester 7'®. Brown with black spots.......... 61, Tubercle more or less toothed above. a**, Abdomen more or less black. a, Abdomen completely black. a®, No lighter band between the tubercle and the front ...... b», A lichter band between the tubercle and the front........ 6%, Abdomen not completely black. a*°, Body black with exception of the brown antenne and the terminal joints of the palps. _Scutum on each side of the spine bears six yellow spots ‘Of /a) Wary) MA LUNG Ms teltene ey segnata. quadrivittata. CENeSCENS. laticlavia. PUPESCENS. Stoliczke. mirabilis. umperator, var. [unispinosa, 2. atrata. Fee. flavimaculata. FROM THE INDIAN REGION. 6°, Body black with exception of two yellow spots at the hind- most part of the abdomen. Tubercle with arow of granules. c2, Cephalothorax with a white triangle; scutum with two large white spots. Second joint of antenna 2 mm. long. HF. Ij=13i; B=8 mm. d@°, Body black, with the exception of the usual triangle and a longitudinal yellow spot on each side of the spine ...... e®, Scutum black, with a large waxy white spot in the front and a smaller one behind. One spine. The ventral segments are yellow with black spots. The coxse are black ...........- 6*, Abdomen yellow or brown. a7, Body yellowish. Cephalothorax with a yellow spot infront. The spine, a spot at its base, and marginal spots black. Tubercle rather high and armed with a row of teeth. Second joint of the antenna 13 mm. long. F.1=14; B.=5mm......... 27, Body bears one short spine. Tibiee twice as long as broad. F. [=8; F. 11=14; B.=45 mm. Colour yellow or yellowish- brown, and there is sometimes a more or less marked longitudinal darker band. Cephalothorax bears a yellow triangular spot, encompassing the tubercle. Palps, antennze, and legs yellow- ish-brown or yellow.........- 7, Body armed with two long spines. Tubercle provided with teeth above. Colour reddish- PYOWD ... 2002 - eset Hol olkcs a. Tubercle provided ah small teeth above..........+++20% LINN. JOURN.—ZOOLOGY, VOL. XXVIII. 487 binotata. histrionica. [var. fusca. triangularis, ceylonensis. cervina. triangularis. umperator. emperator, forma typica.. 30 488 MR. C. WITH ON PHALANGIIDE b°*. Tubercle provided with larger [var. dentata. teeth above and under theeyes. wnperator, d*, Scutum reddish-brown. One black [wnispinosa ° SPC Aki icant, aeicyuca tne ey tere ele imperator, var. 1, First pair of femora not more than four times as long as the second joint of the antenne. a’. Tubercle provided with strong teeth around THE'CYES! “eae pees Gir ate sis ee Oe plebera. 6**. 'Tubercle smooth or granulated. a, Body granular above and below °°. Free ventral segments smooth. a. Body much raised towards the single SHUN a ono TAO ss ek onn aon ed aGe Hansenir. 6°’. Body not more raised than usual. a°*, 9. Body completely black. Spine smooth. ¢. Terminal joints of the as RAINE take trastis. palpsiyellOw as ces-b cage tars sasiets ste atrata. b°*, ©. Blackish-brown. Spine granular. 3. Palps completely yellow ...... varrans. 6. Scutum with two spines in a transverse row. UIHeELE SIM OOLD.) saree varchar cea shane Roe eo taee se bispinosa. GAGRELLA DENTATA, 0. Sp. @. Cephalothorax almost triangular. Scutum bears two very long spines: the first directed upwards and forwards, the second almost perpendicular. Body raised towards the first spine, and sloping behind the second. Penultimate segment of the cephalo- thorax narrow, the last is not so broad as the articulations between the cephalothorax and the scutum. Frontal eminence well developed and armed with a short distinct thorn, directed upwards and forwards. A distinct lamina frontalis is very well developed, triangular and spinous. Granulation of the body, with the exception of the well-granuled coxe, so very little marked that it becomes smooth. Tubercle, seen from the side, rather high and convex, as high as long; seen from the front it is much broader than high. In front there is a thorn on each side of the groove, as long as the diameter of the eyes. There are two granules under the eyes, and the diameter of the eyes is much less than the distance between them, and greater than the distance between their lower edge and the base of the tubercle. Basal joint of the. antenna almost smooth. Procursus maxillaris internus long, slender, pointed, and pale ; the upper branch very long, the lower very short; its lower FROM THE INDIAN REGION. 489 edge granular. Palps hairy but without teeth, with the excep- tion of some very small ones on the patelle and tibie. Patelle long and enlarged towards their extremities. Inwardly there is a long slender apophysis, as long as the patelle are broad. Tibiz five times as long as broad. The legs are very long and slender ; the femora spinous. F. [=12 (53), I1=24 (120), IJI=10 (45), IV=6 (70) mm. Body 5-7 mm. Second joint of the antenna 1} mm. long. Colour white with some few dark spots. The procursusfrontales, tubercle, and articulations are yellowish-brown. A longitudinal broad band from the first spine to the end of the abdomen is indistinctly brown. The top of the spines is black, the coxe and genital plate brown, the ventral segments yellow with white spots on the side, the point of coxa 4 with a yeilow spot in front. The antenne and palps are yellow, the legs brown. One full-grown female from “‘ Large Nicobar.” Another specimen from Little Nicobar has no patellar apophysis. The procursus frontales are smaller. The colour is bluish-white, the longitudinal band yellow. The segments of the scutum are marked by a row of depressed dark spots. The last specimen, also from Little Nicobar, is a little different from the other two. Instead of the frontal spine there is a short apophysis, armed with three short pointed teeth. Beneath this apophysis there is a short pointed tooth. The scutum is finely granular. Apophysis of the patella short. The femora of the palps bear an inner row of granules. Tubercle black. This beautiful and odd species was taken by the Danish * Galathea’ Expedition. GAGRELLA NOBILIS, n. sp. © (?). Cephalothorax almost square; the scutum bears one thorn; the area between the ocular tubercle and the spine is almost plane. The upperside is densely and finely granular with the exception of the tubercle, spme, and free dorsal segment. The free ventral segments are smooth; cox partly smooth, partly beset with large distinct granules. The lamina frontalis is undeveloped; the procursus are small, narrow, well separated, and armed with few teeth. The tubercle is very characteristic ; seen from the side it is somewhat higher than long, its upperside iy almost flat; the front side sloping and higher than the back. 35% 4.90 MR. C. WITH ON PHALANGIIDA Seen from the front, the tubercle narrows beneath the eye and is as high as broad. ‘The crest is grooved and provided with five long hairs on each side. Diameter of the eyes both less than the distance between them, and than the space from their lower edge to the base of the tubercle. Basal joint of the antenna once and a half as long as broad, aud smooth. Procursus max. internus two-branched and pale; its lower edge bears dark teeth. The femora of the palps are as long as the tibie+patelle and the tarsi. Beneath and at the base of the patelle they are toothed. Patelle hairy and toothed. Tibi four times as long as broad. Tarsi with few teeth below, and the femora of the legs hairy and with the usual small teeth. F. [=9 (46), W=17, JI1=8 (438), 1V=12 mm. Body4mm. Second joint of antenna 1} mm. long. On both sides of the tubercle the body is white; the hindmost part of the cephalothorax and the abdomen is yellowish and brown, the three colours merging into each other. A dark depressed spot on each side of the tubercle. Tubercle black, and connected with the front by a broadening black band. Five or six light yellow spots are placed in a row on each side of the scutum near the middle. Last two free dorsal segments black, with a longitudinal white spot on each side. Free ventral segments yellowish-white with indistinct black spots. First pair of cox white, the second black, the third and fourth white at the base, and black spotted with yellow at their extremities. The trochanters and the femora of the palps are black; their terminal joints, the antenne, and the mouth-organs are yellow, the legs brownish, and the points of the second and fourth tibie white. The spine is black. One female (?) from Silouri. GAGRELLA CRUX, 0. sp. Q@. Cephalothorax almost triangular. Tubercle placed near the articulation, since the next last segment of the cephalothorax 1s small. The last segment well developed. ‘The five segments of the scutum marked by darker transverse striz. The body granular above, with the exception of some parts of the cephalo- thorax, the spine, and the tubercle; the coxe are almost smooth, and the free ventral segments completely so. The cephalothorax FROM THE INDIAN REGION. 491 presents the usual depressed triangle. On each side along the margin there is a depressed groove, and near the tubercle a black depressed spot. Procursus frontales more or less separated and spinous. Tubercle low, convex, and longer than high, and seen from the side it is on each side beset with two small teeth. In front it is broader than high. There ig the usual groove above. The diameter of the eyes is less than the distance between them, and larger than the space between their lower edge and the base of the tubercle. Basal joints of the antenne granular. Procursus max. int. two-branched and pale, the upper branch short or absent ; the lower edge granular. The palps are slender and hairy, but without teeth, with the exception of the patelle, which have a few at the side. Femora as long as the tarsi and shorter than the sum of patelle and tibie. Patelle have an inner apophysis, which varies very much in the different specimens, even in the two palps. It may be slender or stout, pointed or obtuse, well developed or almost absent. Tubiz five times as long as broad. Femora of the legs spinous, and the second pair of tibiz with false articulations. Hel 7> (a2), Ll — 2s (64), Mil —/65 (05), LV— 92 Goa) Body 3mm. Second joint of the antenna ? mm. long. Cephalothorax yellow, with the exception of some depressed spots and a more or less narrow indistinct band between the tubercle and the front. Tubercle brown, in front yellow. The seements between the tubercle and the black front margin of the scutum are black, the articulations lighter, and this black transverse band is continued along the margins of the scutum, through which there runs a longitudinal broad black band. Space between the central and marginal black bands yellow, but subdivided by four indistinct narrow transverse black bands denoting the segments. The first two free dorsal segments bear a central and two marginal black spots and two that are white. Genital plate and free ventral segments yellow; coxe white with a brown base. Antenne yellow, the second joint with transverse black spots. Palps and legs yellow with the exception of the black trochanters. This species is similar to G. Jepida, Thor., but differs in its colour and shorter legs. Three females from Punkabari. 492 MR. C. WITH ON PHALANGIID” GAGRELLA ARMILLATA, Thor. Thorell, (10) p. 629. G. LEPIDA, Thor. Thorell, (10) p. 626. The tarsus of the palps of the males is smooth. G. cHzTorts, Thor Thorell, (10) p. 631. G. sorpipata, Thor. Thorell, (10) p. 634. The tarsus of the palps of the male is smooth. G. minax, Thor. Thorell, (10) p. 638. The tarsus of the palps of the male is smooth. G. EREBEA, Thor. Thorell, (10) p. 636. 3. The femora of the palps alone bear small teeth. The patelle have no inner apophysis. The tarsi bear the usual inner row of granules, which are small and placed somewhat apart. Colour brown, with indistinct black spots above. Cephalothorax yellowish-brown. The articulations are yellow, the cox@ brown, and the ventral segments yellowish. Both sides of the cephalo- thorax in front and the coxe partly covered with a white waxy material. Antenne and palps yellowish-brown; the legs yellow. As the malé described here is in all other respects similar to G. erebea, Thor., and as it is one of Fea’s original specimens, I conclude that the differences are only such as occur between males and females. G. sprinuxtosa, Thor. Thorell, (10) p. 657. The tarsi of the palps of the males are smooth. G. HIRTA, 0. sp. Q@. Cephalothorax almost square. Body somewhat raised towards the single spine, falling gradually behind. It is not very densely granular above, and the coxe, the free dorsal segments, the spine, and the genital plate are less so. Lamina, procursus FROM THE INDIAN REGION. 493 frontales, tubercle, free ventral segments, and spots on cephalo- thorax smooth. Cephalothorax with the usual depressed tri- angle. The whole body is densely hairy, the “hairs” curved and. yellow. Frontal eminence well marked. Procursus frontales well separated, small, and spinous at their points. The tubercle, seen from the side, is higher than long and convex, the front higher than the back. The top is convex and surmounted by a row of yellow hairs on each side of the groove. Seen from the front it is as broad as high, and narrows towards its base. The diameter of the eyes is as great as the distance between them, and a little less than the distance between their lower edge and the base of the tubercle. Basal joint of antenna smooth, and almost twice as long as broad. Procursus maxillaris internus two-branched, the upper branch small and pale; the lower larger, with a granular lower edge. wine Femora of the palps as long as the patelle and tibie, but shorter than the tarsi; toothed below and above at the base of the patelle. The patelle are enlarged towards their extremities and densely toothed, chiefly on the inner side. Tibie twice and a half as long as broad and granular. Femora of the legs strongly hairy. Ray tall FB. 1=73 (85), I1=18 (70), WI=73 (85), 1V= 103 (60) mm. Body 54 mm. The second joint of the antenna is 13 mm. long. Body light brown, with a longitudinal broad darker band through the abdomen. Cephalothorax has darker spots. Base of the tubercle brown; eyes and parts around them black. A yellow band, narrow in the front and broad behind, runs between the eyes. Coxe and genital plate brown; the free segments are lighter. Antenne yellow. Palps and legs brown; the latter are lighter towards their extremities. One female from Punkabari. GAGRELLA FRAGILIS, 0. sp. . 2,6. Cephalothorax almost square. Abdomen raised towards the single spine. Segments separated from each other by broad articulations in the pregnant females, which also have the lateral parts of the ventral segments well developed. Cephalothorax presents the usual depressed triangle, open in the front. The upperside is granular ; the cox and genital plate less densely granular. The articulations, the spine, the lamina frontalis, and 494, MR. C. WITH ON PHALANGIID® the last two dorsal segments are almost, and the free ventral segments are completely, smooth. Frontal eminence indistinct, and only bears a few large granules. Scutum hairy behind the spine. Lamina frontalis indistinct; procursus small, partly separated and toothed. Tubercle, seen from the side, is as long as high; convex above and behind. The front face is the higher. It is broader than high, as seen from the front, and somewhat narrower at the base. The top is deeply grooved, hairy, but hardly ever granular. The diameter of the eyes is as great as the distance between them, and greater than that between their lower edge and the base of the tubercle. Basal joint of the antenne smooth. Procursus max. internus two-branched. Femora of the palps both shorter than the patella +tibie and than the tarsi, and toothed below. Patellz enlarged towards their extremities, and toothed. Tibie, which are almost smooth, are twice and a half as long as broad. ‘The tarsi are hairy, and in the males armed with the usual row of conical teeth at the inner lower edge. Femora of the legs strongly hairy; the legs very brittle. F. I = 63 (0), 11 = 103 @), NI =6, 1V=82 mm. Body 5-7 mm. The second joint of the antenna is 13 mm. long. Colour blackish-brown, the underside and the articulations lighter. Antenne and last two joints of the palps yellowish- brown ; legs brown, but lighter towards their extremities. A great number of specimens from Darjeeling. Var. b¢spinosa is similar to the described species, but it has only two spines. One specimen from Darjeeling. GAGRELLA Fra, Thor. Thorell, (10) p. 648. The tarsi of the palps (¢) bear the usual row of teeth. G. Nocricotor, Thor. Thorell, (10) p. 651. The tarsi of the palps (¢) bear the usual row of teeth. G. atrata, Sfol. Stoliezka, (5) 1869, p. 213. @. Cephalothorax semilunar. Scutum raised towards the single spine. Free segments, as usual in the pregnant females, OO ————— ee S—- FROM THE INDIAN REGION. A495 weli separated from each other. This fact is mostly evident from above, on account of the granulation. Last segment of the cephalothorax separated both from the scutum and from the rest of the cephalothorax by articulations, as broad as the seg- ment itself. Articulations between the free dorsal segments broader than the segments. Space between the scutum and the free ventral segments very broad, in front as broad as the secutum, which seems to be disposed as a cap on the top of the abdomen. ‘The lateral parts of the free ventral segments well developed, the ventral segments being transversely grooved near their hinder margin. In the young females the shape of the body is as in the males. Cephalothorax presents the usual depressed triangle. Scutum, cephalothorax, first free dorsal segments, the coxe, the genital plate, and the lateral parts of the first free ventral segments distinctly granular. Middle part of the ventral segments, the last two dorsal, the spine, and the tubercle smooth. Procursus frontales partly united, stout and toothed. Tubercle very low, convex, and longer than it is high. The top is smooth or on both sides of the groove crested with a few small teeth. In front under the eyes a row of five teeth. The basal joints of the antenne bear black granules above. Procursus max. internus two-branched, the upper branch is slender, the lower edge toothed. Femora of the palps bear inwardly a row of short conical teeth, and outwardly a row more irregularly placed. In the middle they are smooth. Both tibie and patelle are toothed, and the tibie are twice and a half as long as broad. Tarsi hairy, and sometimes with a few teeth below. Femora shorter than tarsi and longer than patelle + tibie. Femora of the legs granular. F. I= 5 (23), IT = 8 (44), III = 5 (23), IV =7% (83) mm. Body 5-7 mm. Second joint of the antenna 17 mm. long. Body black. Free ventral segments lighter. Space between the eyes in some specimens light brown, and the cephalothorax beset with lighter spots. Articulations, palps, and antenne yellowish-brown ; legs blackish-brown ; the base of the femora lighter. 3. Femora of the tarsi toothed below, with no difference between the outer and inner row. Tibie smooth. Tarsi bear a long row of teeth (30-50), which extends from the base to the 496 MR. C. WITH ON PHALANGIID™® last third part. Legs of different lengths, but generally longer than in the females. F. I = 5-7 (82), IT = 9-103 (50), II = 5-7 (82), IV = 8-92 (40) mm. Body 53 mm. Space between the eyes and between the tubercle and the front yellowish-brown. Antenne yellow; the second joint spotted with transverse black bands. Terminal joints of the palps yellow. The femora of the legs show a distinct light- brown ring. Legs usually black, but in two specimens brown. Articulations between the coxe and the trochanter white. A great number from Calcutta. This species is without doubt identical with G. atrata, Stol. There is only a slight difference in the granulation; while Stoliezka’s species is also granular below. G. atrata is very similar to G. Fee and G. nocticolor, Thor., but especially to G. varians, n. sp. It is almost impossible to distinguish the females from each other. GaGREeLLA Marnpront, S72. H. Simon, (16) 1897, p. 296. G. stanata, Stoliczka. Stoliczka, (5) 1869, p. 214. G. QUADRIVITTATA, Sim. E. Simon, (9) 1887, p. 115. G. mzNESCENS, Thor. Thorell, (10) 1889, p. 643. With regard to the difference between the palps of the males and the females, consult Thorell. The tarsi of the males bear a row of granules, which begin near the base and are continued towards the extremity. The teeth in the proximal half are placed relatively close to each other, while in the distal part they are distant. Femur IIT=21 mm. Thorell has only 11. It is probably a printer’s error. G. LaTICLAVIA, Thor. Thorell, (10) 1889, p. 641. G. RUFESCENS, Thor. Thorell, (10) 1889, p. 645. FROM THE INDIAN REGION. 497 GAGRELLA STOLICZKS, 0. sp. @. Cephalothorax somewhat triangular. Body somewhat raised towards the single spine, with granules above flat, not very densely placed. Spine granular. Free dorsal segments, the coxe, and genital plate bear a few small granules. Free ventral segments, lamina frontalis, and tubercle smooth. Cephalothorax presents the usual depressions. Lamina frontalis well developed ; procursus frontales long, enlarged towards their edges, and toothed. The tubercle, seen from the side, is low and convex. The front is in breadth once and a half its height, and is scarcely narrower towards its base. The top is grooved, hairy, and smooth ; but in the front under the eyes there is a row of well- developed teeth, curved upwards. Diameter of the eyes smaller than the distance between them, but larger than the distance between their lower edge and the base of the tubercle. | Basal joint of the antenne about twice as long as broad and granular above. Procursus max. internus two-branched. Femora of the palps longer than the patelle and tibie, but shorter than the tarsi, toothed below and above at the base of the patella. Tibiz and patelle spinous. Tibie twice and a half as long as broad. Femora of the legs bear small teeth. F. I=8, 11=14 (ce. 80), T1T=74, 1V=114. The second joint of the antenna is 13 mm. long. Body 6 mm. Upperside brown; scutum with transverse black bands to mark the segments. Colour lighter in front and behind. Tubercle yellow between the eyes. Underside lighter. An- tenn and palps yellowish-brown. Trochanters, spine, and base of the femora black. Legs light brown, with darker patelle. Articulations between trochanters and coxe white. One female from Tenasserim. This species is very similar to G. rufescens, Thor. GAGRELLA MIRABILIS, 0. sp. Cephalothoraz almost triangular. Body raised towards the single spine, which presents a very singular appearance. It is short and thick, and its pot is bent downwards along its lower part. Ido not at all think that this shape is natural, but that it has been caused by violence while undergoing ecdysis. The body bears relatively few, large, and low granules. The spine is also granular. Free ventral segments smooth. Lamina 498 MR. C. WITH ON PHALANGIID& frontalis small; the procursus frontales distinct, pale, and toothless. Tubercle rather long and smooth. The front edge is seen to decline; the back is perpendicular. Viewed from the side, it is almost as high as long. Diameter of the eyes almost as large as the distance between them, and a little larger than the dis- tance between their lower edge and the base of the tubercle. Basal joint of the antenna smooth. Procursus max. internus two-branched and smooth, but with the branches almost parallel and bent downwards and forwards. Femora of palps bear long pointed teeth; patelle and tibie bear smaller. Tibize three times as long as broad. Femora of the legs spinous. F. I=10 (50), I[=20 (86), III=10 (42), IV=14 (65). The second joint of the antenna is 14 mm. long. Body 5 mm. Colour greyish, with an indistinct darker longitudinal band. Cephalothorax with a few dark depressed spots. Coxe and ventral segments white with darker spots, especially distinct at the coxe. Extremities of the coxe and marginal teeth black. Antenne yellow, palps and legs yeilowish-brown. One mutilated specimen from Mooleyit (500-600 m.). This animal was mounted as G@. laticlavia, Thor., in the Zoolo- gical Museum of Copenhagen; but as it differs much from the description of G. laticlavia, and as Thorell’s specimens of G. laticlavia are from Thagata, Jiva, I think there has been a mistake, and I accordingly establish a new species. GAGRELLA FLAVIMACULATA, D. sp. 3. Cephalothorax broad and semicircular. Segments of the scutum (which bears a spine) marked by distinct transverse grooves. Along the margins of the first half part of the abdo- men there is a longitudinal groove. Granulation of the upper- side fine and dense ; cox beset with fewer and longer granules. Free ventral segments smooth. Procursus frontales small, pointed, and smooth. Tubercle, seen from the side, almost tee above, and longer than high, beset with a few granules. Seen from the front, it is as high as broad, and narrower below. Diameter of the eyes larger than the distance between their lower edge and the base of the tubercle, but smaller than the distance between them. Basal joint of the antenne smooth. Femora and patelle of the palps toothed, the tibie smooth. Tibie three and a half FROM THE INDIAN REGION. 499 times as long as broad. Tarsi (¢) bear two rows of teeth: the inner proximal row extends to near the middle, and con- sists of twelve small flat teeth, placed near to each other with the exception of the last; the outer distal row begins before the middle, but does not reach the extremity, and its few stout teeth are placed very far apart. Femora of the legs bear small teeth. F, [=143, 11=20(c. 100), 111=133, IV=18. Second joint of the antenna 17 mm. in length. Body 53 mm. Colour black, with the exception of the second joint of the antenna and the terminal joints of the palps, which are brown. On each side of the abdomen are six yellow spots of a waxy material, and the coxe are covered with a similar mass. The tubercle is brown. One mutilated male without locality. G-AGRELLA BINOTATA, Szm. Simon, (9) 1887, p. 115. G. HistRIonica, Thor. Thorell, (10) 1889, p. 652. G. CEYLONENSIS, Karsch. Karsch, (13) 1892, p. 308. G. CERVINA, Szm., Simon, (9) 1887, p. 115; Thorell, (10) 1889, p. 655. G. TRIANGULARIS, Nn. sp. 9, d. Cephalothorax square or triangular. Body higher towards the single spine and sloping gradually behind, its upper surface densely granular. Spine, coxe, genital plate, and some parts of the cephalothorax less granular; the free ventral seg- ments, lamina, and procursus frontales quite smooth. Depressions of cephalothorax little marked. Lamina frontalis well developed, with slender and pointed procursus. Ocular tubercle convex and almost as high as long; when viewed from the side, its front face higher and more sloping. The tubercle is, on each side of the groove, surmounted by larger and smaller teeth, which are continued as a row of granules under the eyes. Seen from the front, it is almost as 500 MR. C. WITH ON PHALANGIIDE high as broad, and narrower towards its base. Diameter of the eyes as large as the distance between them, and much larger than the distance between their lower edge and the base of the tubercle. Basal joint of the antenne smooth. ‘ Fingers” in some specimens irregularly toothed; the labrum (epistoma) in some club-like. The procursus max. internus two-branched and toothed. Femora of the palps strongly spmous below. Teeth arranged in two rows—an inner of flat and stout, an outer of more or less pointed spines, but not in a regular row. Patelle spinous, and with a small inner apophysis. Tibie granular and twice and a half as long as broad. Tarsi of the females hairy and smooth; and in one specimen with indistinct tarsal teeth, placed in two rows. Tarsi of the males, with regard to their dentation, similar to G. flavimaculata. The lower side has two rows of teeth—a proximal row beginning near the base and extending almost to the middle near the inner side; it consists of c. 20 small flat teeth, which are placed near to each other with the exception of the last. The distal row along the outer margin begins near the middle and extends to the extremity ; it con- sists of c. 10 small pointed teeth, placed far apart. Legs long, stiff, and with small teeth. F. [=82 (35), I1=13 (62), 111 = 8 (83), TV = 103 (45) mm. Body 44 mm. Second joint of the autenna 1 mm. long. Colour red-brown, with larger or smaller irregular yellow spots; and sometimes there runs through the abdomen a darker longitudinal band. Cephalothorax bears a yellow triangular spot, turned forwards and encompassing the tubercle. This triangle has two darker spots on each side and a pointed narrow one in front. Ventral segments and coxe lighter. Spine blackish-brown. Antenne and the palps yellow. Legs yel- lowish-brown. Var. FUSCA. Body black, with the usual yellow triangle and a long yellow spot on each side of the spine. Tubercle black, but yellow between the eyes. Last segments of the cephalothorax brown with yellow spots. Coxe brown. Ventral segments, with antenne, palps, and legs, yellowish-brown. The missionary Mr. Loventhal has captured many specimens (only two of the variety) at Vellore, near Madras. FROM THE INDIAN REGION. 501 GAGRELLA IMPERATOR, Ni. Sp. 2. Cephalothorax somewhat semilunar. Scutum bears two spines, placed on the first and on the second segment; the foremost shorter and directed forwards and upwards ; the hindmost more perpendicular. The body is higher towards the spine, and behind it slopes gradually. It is very dilated in the gravid females, and the articulations are well developed. The body and, especially, the top of the scutum are finely granular above. Cephalothorax less densely granular; the coxe and the spine with fewer and larger granules. Ventral segments and articulations smooth. Cephalothorax presents the usual depressions. Frontal eminence well marked; lamina frontalis distinct and the procursus rather long, separated at their outer ends, and spinous. Tubercle low and, seen from the side, as high as long. The front area is higher than the back and not perpendicular. Seen from the front, it is broader than high and narrowed at its base. Five small teeth on each side of the groove. Diameter of the eyes larger than the distance between their lower edge and the base of the tubercle, but less than their distance apart. Basal joint of the antenne almest smooth above. Procursus max. internus two-branched and yellow, the lower edge with a few granules. Femora of the palps shorter than the tarsi and longer than the patelle + tibiz. Femora toothed below. Tibie twice and a half as long as broad. Both tibie and patelle are granular, chiefly on their inner areas. . Femora of the legs, as usual, granular. Tibie 2 have false articulations. | F. I=8 (35), (=17 (ce. 80), INT =84 (85), IV =12 (45). Se- cond joint of the antenna 17 mm. long. Body 83 mm. Colour a little different in the three specimens—one com- pletely yellow, and soft as if in the condition for ecdysis; the two others light brown, but more or Jess spotted with yellow. Spines somewhat darker. Coxe brown. Three specimens ( 2 ) from the Andamans. Var. DENTATA. Granules larger. Tubercle surmounted by large teeth and provided with a row of granules under the eyes. Basal joint of the antenne granular. One female from the Andamans. 502 MR. C. WITH ON PHALANGIIDE Var. UNISPINOSA. 2. Scutum has only one long spine ; in one specimen there is a little spine in front of the long one. Tubercle smooth. Legs long. iB 1=12)(60) S26 i Vi IG: Colour brown with black spots, or with the last segments of the cephalothorax darker. Antenne yellow. Palpi yellowish- brown. Legs brown. 3. Tubercle bears a few teeth above; tarsi bear an inner row of about 20 teeth, placed rather far apart. Cephalothorax black or blackish-brown ; the scutum brown with a black spine. Three free dorsal segments black in the front, yellow-brown bebind. Underside blackish. The Danish ‘ Galathea’ Expedition has taken three specimens from the “ Little” and three from the “ Large ’’ Nicobar. Though the var. wnispinosa is somewhat different from the main type, I prefer at present to place them in the same species. The characters which distinguish them seem to be fluctuating, as, for instance, the length of the legs, the armature of the scutum, and the tubercle. When we have acquired richer material from the different islands of the two groups, it will be possible to estimate better the value of these differences. GAGRELLA PLEBEIA (Thor.). Thorell, (40) 1889, p. 659. (Melanopa plebeia.) i Lateral parts of the ventral segments partly granular. Tarsi of the males smooth. The two specimens, which are mounted as Thorell’s original specimens in the Museum at Copenhagen, are very different with regard to the femora of the legs. The femora in one resemble those of an ordinary Gagrella, but in the other they are very odd. The femora of especially the fourth pair of legs are very thick and gradually thicken from the base to the extremity, the outer and under sides are strongly curved inwards; and it is probably such femora which suggested to Thorell the genus Melanopa. GAGRELLA TRISTIS (Thor.). Thorell (10) 1889, p. 662. (Melanopa tristis.) Palps of the males with the usual tarsal row of teeth. FROM THE INDIAN REGION. 508 GAGRELLA HANSENII, n. sp. 3. Cephalothorax triangular. Body much raised towards the single spine, the point of which is curved backwards. Free ventral segments transversely grooved, whereby the back margin becomes more marked. Lateral parts well developed, and their hindmost (non-granular) parts merge into the corresponding part of the central one. Body beset with large, dense granules. The foremost part of the lateral pieces of the ventral segments bears smaller granules, and the genital plate bears very few and large. Cephalothorax presents a well-marked depressed triangle. Lamina frontalis indistinct ; procursus stout, partly united, and toothed. Tubercle low, seen from the side it is somewhat longer than high. The top is deeply grooved, but smooth. A row of granules under the eye. The diameter of the eyes is as large as the distance between them, and larger than the distance between their lower edge and the base of the tubercle. Basal joint of the antennz smooth above. Proc. max. internus two-branched and low, the upper branch pointed and slender, the lower stout and obtuse. The lower edge granular. Femora and patelle of the palps toothed. Tibi almost smooth, and twice and a half as long as broad. ‘The tarsi bear a long row of teeth, which are stout, obtuse, and pressed downwards. Femora of the legs spinous ; false articulations of the second pair of the tibie very distinct. Po i=4) (22), ti =7 (35), 1 =4 @2), 1V => 63 mm.. Body 5 mm. Second joint of the antenna 1 mm. long. Body black. Antenne, the two terminal joints. of the palps, and the mouth-organs yellowish-brown. Legs black, with the exception of a brown ring near the base of the femora. One male from Todaspoor. GAGRELLA VARIANS, 0. sp. @. Cephalothorax almost semilunar. Body a little raised towards the single short spine, finely and densely granular, with the exception of the top of the spine, the free ventral segments, and the articulations. Procursus frontales united, smooth, and toothed. Tubercle low, convex, and longer than high; seen from the front it is broader than high. The top on each side of the low groove is provided with small teeth; others are placed under LINN, JOURN.—ZOOLOGY, VOL, XXVIII. 36 O04 MR. C. WITH ON PHALANGIIDE the eyes. Diameter of the eyes as large as the distance between them. Basal joint of the antenne granular above. Procursus max. internus two-branched. Femora of the palps toothed below and partly above. Patelle short and toothed. Tibize twice as long as broad, and with small teeth below. Femora of the legs spinous; tibie of the second pair with very distinct false articulations. Second pair of legs very slender. F.1 = 4-5 (c. 22), II = 63-8 (85), III 4-5 (22), IV = 6-7 (28) mm. Body6 mm. Second joint of the antenna 1 mm. long. Colour black, space between the eyes lighter. Antenne and the palps yellowish-brown, spotted with black. Femora and the tarsi of the legs light brown, tibiz and patelle black. ; Two females from Berbhoom District. The described female is very similar to G. atvata. Its common aspect and colour are a little different; the femora are more slender. Var. DENTATA. 2. Upperside and coxe beset with large dense granules. Spine almost granular to its point. Tubercle on each side of the deep groove crested with a row of rather long teeth. Colour blackish-brown, with the cephalothorax and especially the space between the eyes lighter. Ventral segments and the coxe light brown. Antenne and the terminal joints of the palps yellow. Legs more or less brown, with darker patelle and tibie. One female from Berbhoom. o. Body beset with dense and large granules. Tubercle surmounted by a row of strong granules, continued under the eyes. ‘libie of the palps three and a half times as long as broad. ‘Tarsi with the usual row of strong conical teeth. Body 5 mm. Colour brown, with black spots; most distinct are the trans- verse spots of the scutum which mark the segments. Spine black. The single specimen has a longitudinal darker band through the scutum. Space between the eyes yellowish. Ventral segments lighter. Antenne and the palps yellow. Legs brown. Two specimens from Berbhoom. I have great doubt as to the limitation of this species, since the females are in most characters similar to G. atrata, Stol. FROM THE’ INDIAN REGION, 505 On the other hand, it seems unnatural for me to refer it to another species than that represented by the male; and as this is very well characterized, I have found it most correct to grant the species the afore-named limitation. GAGRELLA BISPINOSA, Karsch. Karsch, (13) 1892, p. 309. G. monacanTua (Herbst). Herbst, (1) 1798, p. 19. I think it impossible to identify this species from the description. Scotomenta, Thor., 1889. ScOTOMENIA CETRATA, Thor. Thorell, (10) 1889, p. 665. With regard to the other genera described from India, I refer to the following papers :— Thorell, 10. Simon, 8. C. L. Koch, 2. On full consideration it seems natural for me to refer the genera Syleus, Thor., Systenocentrus, Sim., and Oncobunus, Thor., to the same genus. Besides the species herein described, I have had some young specimens which it has been impossible to define. Two large specimens from Punkabari were very well marked. They had relations both with Gagrella and Liobunum (the marginal pro- jections, for instance, were wanting); but it was impossible to decide whether they were full-grown or not. As their genus was thus dubious, I have not described them. SUPPLEMENT. GAGRELLA SEPIA, Loman. J. C. C. Loman, (14) 1894, p. 8. 2. Spine with only few granules at the base. A triangle in front, including the frontal eminence, is smooth, but beset with a row of three large granules along the frontal margin. Front and back of the high tubercle perpendicular. The top is convex, 506 MR. C. WITH ON PHALANGIIDE provided with a row of rather strong teeth. The transverse row of hairy granules on the ventral segments wanting. ’g. Four frontal granules, not placed in a single row. A small tubercle on the first abdominal segment before the spine. Tarsi of the palps with the usual row of teeth. In all other respects this species is similar to G. sepia, Lom., and as it was taken by Loman at the same locality, it must be regarded as the same species. - GAGRELLA SPLENDENS, 0. sp. 3. Cephalothorax almost semicircular. Body raised towards the single spine, sloping gradually behind. It is granular, with dense and low granules above, the coxe bearing a few larger granules. Ventral segments smooth. Frontal eminence well marked. Lamina frontalis distinct, broad, and merging on each side into the small but stout procursus frontales. Tubercle, seen from the side, rather low, convex, and as high in front as behind. Seen from the side, broad and scarcely narrower towards its base. It is granular on both sides of the groove. Diameter of the eyes larger than the distance between their lower edge and the base of the tubercle, and smaller than the distance between them. Basal joint of antennew smooth. Procursus max. internus two- branched and pale ; branches short, and lower edge beset with a row of granules. Femora of the palps with a few granules below, arranged in an outer and an inner row, and above towards their extremities. Patelle short, thinner at their bases, and beset with a few granules. Tibize longer than:the patelle, twice as long as broad, cylindrical, and very thick—as thick as the femora (compare the male of G. enescens). They are somewhat thinner towards their extremities and base, and a little convex below; underneath they are hairy with long stiff hairs, and above only with short. The tibiw are only a little longer than the patelle. Tarsi hairy and much thinner than the tibie, with a row of comparatively few teeth, of which the 13 proximal are placed near to each other from the base to the middle, and the 4 distal are widely separated. Femora of the legs toothed. Tibi II have false articulations. BF. f= 112 42); l= 172 (0), 11 = 107 41), LY —aaieo: Second joint of the antenna 13 mm. long. Body 5 mm. Colour black above, with a beautiful metallic gloss. Between € FROM THE INDIAN REQION. 507 the tubercle and the front a longitudinal yellow band, broader behind, and including a black pointed spot, starting from the front, the point being turned backwards. Near the margin of the cephalothorax are other yellow spots. Along the margins of the scutum there are three such spots on each side. Coxe yellowish-brown, with darker spots. Free ventral segments yellowish, with darker spots. Legs brown, with the trochanters, patelle, and tibie black. Antenne and palps yellow. Yellow spots larger in another specimen, and the hinder margins of the free dorsal segments are yellow. Mr. Schonau has taken two males at Woosung (China ’). This species is similar to G. enescens, Thor., but the colour is very different and the legs are stronger. BIBLIOGRAPHY. 1. J. F. W. Herssr.— Natursystem der ungeflugelten Insekten. Zweites Heft. Berlin, 1798, p. 19. 2. C. L. Kocu.—Die Arachniden, vol. xvi. (p. 61, fig. 1541). Niirnberg, 1848. 3. C. L. DorescHatt.—Bijdrage tot de Kennis der Arachniden van den Indischen Archipel. Natuurkundig Tijdschrift voor Nederlandsche Indié, xiii. (ser. 3, i11.), 1857. 4. C. L. Donescuatt.—Bijdrage tot de Kennis der Arachniden van Ind. Arch. Acta Soe. Scient. Indo-Neerlandice (Ver- handl. d. Natuurkundig Vereen in Nederl. Indié), vol. v., 1859. 5. F. SrortczKa.—Contributions towards the Knowledge of the Indian Arachnoidea. Journal of the Asiatic Society of Bengal, vol. xxxvill. part ii., Phys. Science, No. 4, 1869, pp- 212-215. 6. T. Tuorene.—Descrizione di alcune specie di Opilioni dell’ Archipelago Malese. Annali del Museo Civico di Storia Naturale di Genova, vol. ix., 1876, pp. 115-123. 7, BE. Stmon.— Etudes Arachnologiques, 5° Mém. ix. Arachnides recueillis aux iles Philippines, etc. Ann. de la Soc. Ent. de France, 2° sér. vil., 1877, p. 96. 8. E. Stwon.—Arachnides recueillis au Cambodja. Actes de la Société Linnéenne de Bordeaux, 1886, 4° série, xl, pp. 137-166. 508 9. 10. lye 12. 13. 14. 15. 16. aie 18. 19: 20. 21. 22. 23. 24. MR. C. WITH ON PHATLANGITD& FE. Staon.— Etudes sur les Arachnides de |’ Asie méridionale, te.: I. Journal of the Asiatic Society of Bengal, lvi. partii., Natural Science, No. 1, p. 115. Calcutta, 1887. T. THorELL.—Arachnidi Artrogastri Birmani. Viaggio di Leonardo Fea: XXI. Ann. del Museo Civico d. Stor. Nat. d. Genova, xxvu. 1889, pp. 521-729. T. Tworetu.—Arachnidi di Nias e di Sumatra. Ann. del Museo Civico d. Stor. Nat. d. Genova, ser. 2°, vol. x., 1890, pp. 5-106. | T. THorEeny.—Opilioni nuovi a poco conosciti. Ann. del Museo Civico d. Stor. Nat. d. Genova, ser. 2?, vol. x., 1891, pp- 669-770. ¥. Karscu.—Arachniden von Ceylon. Berliner entomolo- gische Zeitschrift, xxxvi. pt. ii. 1892, pp. 308-309. J. C. C. Loman. — Opilioniden von Sumatra, Java und Flores. Zoologische Ergebnisse einer Reise in N.-Ost- Ind. (Max. Weber), Band iii. Leiden, 1894, pp. 1-12. T. THoRELL. — Forteckning 6fver Arachnider fran Java. Bibang till K. Svenska Vet.-Akad. Handlingar, Band xx. Afd. iv. No. 4, Stockholm, 1894, pp. 16-25. E. Stmon.—Arachnides recueillis 4 Matheran. Bull. Mus. Paris, 1897, pp. 289-297. R. I. Pococx.—Description of some new Oriental Opiliones. Ann. Nat. Hist. xix. 1897, pp. 283-292. J.C. C. Loman.—Die Opilioniden der Sammlung Plate: X. Zool. Jahrb., Supplem. IV., ii. 1899, pp. 1-4. EK. Sruon.—Contribution 4 la Faune de Sumatra. Arach- nides recueillis ete. 2° Mém. Ann. Soc. Ent. de Belgique, 43, 1899, p. 78. J.C. C. Lomay.—Ueber die geographische Vorbecttune der Opilioniden. Zool. Jahrb., Abth. f. Syst. 13. B., 1900, pp. 72-104. E. Stuton.—On the Arachnida collected during the “ Skeat ” Expedition. Proc. Zool. Soc. Lond. 1901, vol. ii. pp. 80-84. J.C. C. Loman.—Neue aussereuropaische Opilioniden. Zool. Jahrb., Abth. f. Syst. 16. B., 1902, pp. 163-216. Stern Brzrz.—Beretning om Corvetten Galatheas Reise omkring Jorden, 1845-47. Forste Del. Kjobenhayn, 1849, pp- 250-400. KE. Stwon.—Les Arachnides ae France. ‘TT. vii., Paris, 1879. FROM THE INDIAN REGION. 509 Posrscriet.—In a paper published in 1901, E. Simon has described one new genus and in all seven new species of Gagrel- line from the Malay Peninsula, viz.: Gagrella bicarnigera, G. semigranosa, G. patalungensis, G. illusa, G. biseriata, G. atro- rubra, and Verpulus spumatus. They seem all to be well defined from those established by myself. Simon’s paper (21) was unfor- tunately overlooked when I was engaged on the present treatise, and his species have not been taken into account in the synoptic keys. The last-named paper of Loman (22) was published after I had sent in the manuscript; his species have been inserted in the list (pp. 466-68) on the geographical distribution together with those of Simon (21), when I looked over the proof. C. WitH. 21st February, 1903, INDEX. [Synonyms and native names are printed in i¢alics. A star is added to names which appear to be here used for the first time. | Abraxas grossulariata, Linn., 370. Acanthocystis, Carter, 340. aculeata, Hertwig § Lesser, 341. cheetophora, Ledy, 340. erinaceus, Penard, 341. flava, Greeff, 341. pallida, Greeff, 340. paludosa * , West, 340; mentioned, 342, Pertyana, Archer, 341. spinifera, Greeff, 341. turfacea, Carter, 340. viridis, Greeff, 340. Acanthodrilus, Perr., 109, 124. sydneyensis, JMetcher, 124, 129, 131, 1386, 138, 139. Acantholophus, mentioned, 470. Acartia Clausii, Giesbrecht, 306. Accessory glands of Australian Harth- worms (Sweet), 154. Acervulina inherens, Schultze, 198. Achzeus Cranchii, Leach, 292. Actineda, Koch, 374. Actinophrys, Hhrenb., 334. difformis, Ehrenb., 334. Hichhornii, Khrenb., 336. picta, Leidy, 335. Sol, Hhrenb., 334. subalpina * , West,335; mentioned, 342. viridis, Khrenb., 340. Actinospherium, Stein, 556. Hichhornii, Stein, 336. Adelosina polygonia, Schlumberger, mentioned, 19. Agalma, Eschsch., 254. Allorchestes, Dana, 299. aquilina, Costa, 299; mentioned, 307. plumicornis, Heller, 299; men- tioned, 307. Alveolina Boscii, Defrance, 400, 414. Alysidium Lafontii, Awd., mentioned, 2 Amblyplana, Graff, mentioned. 41. Awmmochares, Grube, 254, 255, 259: assimilis, Sars, 259, 260. filiformis, D. Chiaje, 259, 260. fusiformis, D. Chiaje, 259. ottonis, Grube, 259. Ammocharide, On the Structure and Habits of the Polychxta of the 256, Family, by A. T. Watson, 230- 260. Ammoconia sagenelia, Haeckel, men- tioned, 5. Ammoconide, 5. Ammodiseus, Reuss, 29. incertus, @’Orb., 29, fig. 32. Ammothea echinata, Hodge, 506. Ameeba, Hhrenb., 308, 309, 311. bilimbosa, Auerbach, 312. brachiata, Dujardin, 310. lateritia, Fresenius, 353. natans, Perty, 309. princeps, Oarter, 309. princeps, Khrenb., 509. proteus, Leidy, 309. quadrilineata, Carter, 509. radiosa, Hhrenb., 3i0. ramosa, Dujardin, 510. sabulosa, Leidy, 310. verrucosa. Khrenb., 309. villosa, Wallich, 309, 310. Amecebea, 309. Ampelisca diadema, Costa, 300. tenuicornis, Ziljeb., 300. Ampeliscidze, 500. Amphilochus brunneus, Della Valle, 300, SUL. LINN. JOURN.—ZOOLOGY, VOL. XXVIII. 37 512 Amphilochus Marionis, S¢ebbing, men- tioned, 301. melanops, Walker, 300, 301. neapolitanus, Della Valle, 301. Amphipoda of the British Isles, by C. Chilton, 140-161. Amphistegina, d’Orb., 404 ; mentioned, 2, 163, 164, 165. Lessonii, @’ Orb., 204, 209, 382, 387, 394, 407, 414. Amphistomina, 308, 319, 332. Amphithoé aquilina, Costa, 299. Amphithoide, Stebbing, 306. Amphitrema, Archer, 319, 332. Wrightianum, Archer, 332. Amphizonella vestita, Archer, 313. Anarthropora monodon, forma minus- cula, Smitt, 86. Anchialus agilis, Sars, 293. Anomalina, @’ Orb., 195. ammonoides, Jeuss, 406. coronata, Parker & Jones, 382, 406. grosserugosa (Giimbel), 406. (Rotalina) ammonoides, Reuss, 195. Anomura, 292. Anthurideze, 295. Aora gracilis, Spence Bate, 395. Aoride, Stebbing, 305. Apherusa bispinosa, Spence Bate, 303. Aphis, Linn. (Currant), 370. Aphrothoraca, 334. Aporrhais, Da Costa, 487. pes-pelecani, Linn., 436, 442. Apteryx, Shaw, 347. Arcella, Hhrenb., 314, 316, 317, 330. aculeata, Ehrenb., 315. artocrea, Leidy, 315, 342. constricta, Hhrenb., 317. discoides, Hhrenb., 314. hemispherica, Perty, 314. hyalina, Ehrenb., 380. mitrata, Leidy, 515. vulgaris, Lhrenb., 314, 315. , var. gibbosa, Penard, 314. vulgaris, Leidy, 314. Arcellina, Du Pless, 312. vulgaris, Carter, 314. Arenicola, Leach, 232. Argillecia, Sars, 419. acuminata, Miiller, 419. affinis * , Chapm., 419, 433. eburnea, Brady, 420. Argulide, 213. Arthrocentrus atratus, Thor. (Upper Burma), 468. Articulina conico-articulata, 399. funalis, Brady, 399. 300, 195, 209; mentioned, Batsch, INDEX. Articulina funalis, var. inornata, Brady, 381, 399. lineata, Brady, 399. Artotrogus, Boeck, 213. Ascaris, Linn., 245. Asellota, Erichs., 296. Assulina, Lhrenb., 329. seminulum, Hhrenb., 329. Asterocheres. Boeck, 213. Astrorhizide, 4, 182. Athenas nitescens, Leach, 292. Atylide, 304. Atylus Coste, Heller, 304. Australian Earthworms, On _ the Structure of the Spermiducal Glands and Associated parts in, by Miss G. Sweet, 109-139. Autonoé longipes, Liljeb., 805. Bairdia, M‘Coy, 420; mentioned, 397. amygdaloides, Brady, 420. attenuata, Brady, 428. Crosskeiana, Brady, 421. Saba, Reuss, 418. foveolata, Brady, 422, 423. hirsuta, Brady, 422. Milne-Edwardsii, Brady, 422. simplex, Brady, 422. tenera, Brady, 422. ventricosa, Brady, 423. Woodwardiana, Brady, 421. Bairdiidx, 420. Bdelloidina, Carter, 7 ; 6, 8. ageresata, Carter, 7; mentioned, 8, 9, fig. 26, 3888, 393, 400. Beania magellanica, Busk, mentioned, 49, Benthos species of Foraminifera, 414. Bicellaria Alderi, Busk, mentioned, 58. Bifarina limbata, Brady, 401, 409, 417. Biloculina, d@’O7b., 169. elongata, @’Orb., 169. irregularis, @’ Orb., 381, 415. oblonga, d’ Orb., 169, 206, 398. ringens, Lamarck, 170, 206, 391, 398. ——.,, var. denticulata, Brady, 398. subspheerica, d’Orb., 169, 206. Biomyxa vagans, Leidy, 311. Bipaliidee, 41. Bipalium kewense, Moseley, mentioned, 34. Black-Currant Gall-mite, Hriophyes (Phytoptus) ribis, W estw., Life-history of, by Cecil Warburton and Alice L. Embleton, 366-378. Black-Currant plant disease (War- burton & Embleton), 366-378. Bolivina, @’ Ord., 186 ; mentioned, 163. mentioned, INDEX. Bolivina arenosa, Chapm., 401. dilatata, Reuss, 401. limbata, Brady, 187, 208, 382. porrecta, Brady, 401. punctata, d’Ord., 186, 207, 401. robusta, Brady, 401. textularioides, Reuss, 401. tortuosa, Brady, 187, 208, 382. Bolodon, Owen, mentioned, 282. Bomolchus, Nordi., 222. Boruta, Wrzesniowshi, 154, 158. Brachyura, Latr., 292. Brain of Macroscelides and other In- sectivora, Notes on the, by G. Hlliot Smith, 445-448. Branchiomma vesiculosa, 253. Brettia australis, Busk, mentioned, 51. cornigera, Bush, mentioned, 51. frigida*, Waters, 51; mentioned, 98, 101. minima * , Waters, 52; mentioned, 98, 101. Broom, Robert, On the Karly Condition of the Shoulder-girdle in the Poly- protodont Marsupials, Dasyurus and Perameles, 449-454. “Bryozoa from Franz Josef Land, collected by the Jackson- Harmsworth Expedition, 1896-97, by Arthur W. Waters, 43-105. Bacco Dysoni, G. &. Gray, 357. Bugula avicularia, Linn., mentioned, 54. fruticosa, Packard, mentioned, 52. Harmsworthii * , Waters, 54 ; men- tioned, 98. hexacantha, Ortmann, mentioned, doen Murrayana, Johnst.,52; mentioned, 51, 98. , var. fruticosa, Packard, 53, 98. quadridentata, Simitt, mentioned, 51. Bulimina convyoluta, Well., 401. elegans, @’Orb., var. exilis, Brady, 401. elegantissima, @’ Orb., 401. subteres, Brady, 401. variabilis, @’ Orb., mentioned, 28. Buliminingz, 186. Bunotheria, 283. Bythocythere, Sars, 432. arenacea, Brady, 432. armata*, Chapm., 4382, 433. pumilio, Brady, 432. velifera, Brady, 432. Caberea Hllisii, Smitt, mentioned, 58. Calamodon, Cope, 284. Calanus gracilis, Dana, 306. 513 Calearina, @’Orb., 2, 15, 163, 164, 16a 196. Calcar, @’ Orb., var. hispida, Brady ftnote, 15, 197. — Defrancii, @ Orb., 197, 209, 407. hispida, Brady, ftnote 15, 196, 209, 382, 387, 394, 407, 414. , var. pulchella *, Chapm., 15, fir. 26; mentioned, 382, 413. Spengleri, Linn., 382, 335, 4138. Callianassa, Leach, 213. Calliopiidee, 303. Calliopius subterraneus, Chilton, 157. Campecopea corallina ?, [?sso, 296. Cancellaria, Zam., mentioned, 439. Candeina nitida, @’ Ord., 404, 415. Canis, Linn., mentioned, 266. Caprella acanthifera, Leach, 306. Caprellidz, 306. Caprimulgus europeeus, Linn., 553. Capromyidze, 285. Capromyine, 285. Capulidee, 437. Capulus, Montf., mentioned, 436. Carcinus meenas, Penn., mentioned, 292. Cariama, Briss., ftnote 348. Carinatee, 343. Carpenteria, Gray, 2, 12, 195. balaniformis, Gray, 15; mentioned, 388, 390, 392, 3938, 394, 406. , var. proteiformis, Goés, 195. monticularis, Carter, 14; men- tioned, 138, 15, 388, 389, 392, 393, 406. proteiformis, Goés, 195, 209, 210, 406. rhaphidodendron, Mobius, 388, 589, 393, 395, 406, 413, 416. serialis*, Chapm., 390 ; tioned, 391, 593, 413. utricularis, Carter, 12; mentioned, 18, 15, 195, 388, 389, 405, 416. Carterina, Brady, 184. spiculotesta, Carter, 165, 184, 207, 400. Cassidulina, @ Orb., 408. calabra, Sequenza, 402. erassa, @’Orb., 402. levigata, d’Orb., 402. — — (Orthoplecta) clavata, Brady, 402, Caudenia nitida read Candeina nitida, d@ Orb., 415. Cavia, Klein, 262-278. Cobaya, Gmel., 262-290. -—., premolar of, 283. ——, upper molar of, 290. Cayiidz: Tooth-Genesis in, by H. W. Marett Tims, 261-290. Cecidomyid fly, 375. Cellarina scabra, van Beneden, 54. ternata, var, gracilis, Verrill, 55. 37* men- 514 Cellepora, Gmel., 2, 77, 93. armata, Hass., mentioned, 95. avicularis, Hass., mentioned, 99. bilaminata, Hincks, 92. cervicornis, Busk, 94. conica, Bush, mentioned, 95. coronopus, S. Woods, mentioned, 94, 95. cylindritormis, Lusi, mentioned, Oi). Eatoniensis, Buss, mentioned, 96. incrassata, Lamarck, mentioned, 46, 79, 94. inerassata, Smitt, 93-96, 100, fig. 104, 105. pertusa, Smit, mentioned, 91. plicata, Smitt, 92. pumicosa, Bush, 95, 100, fig. 105. ramulosa, Linn., mentioned, 95, 96. ramulosa, Manzoni, 80. scabra, Sinitt, 91. tridens, Kirchenpauer, 80. ventricosa, Lorenz, 96, 100, fig. 104. verruculata, 49. Celleporaria incrassata, Smitt, 93. surcularis, Packard, 94. Celleporella, Gray, 70. Cellularia Peachii, Bush, 58. scabra, Smitt, 54. ternata, var. gracilis, Smitt, 55. Centetes, Z//., mentioned, 445. Centropyxis, Stein, 315, 316. aculeata, Stein, 315, 342. -——, var. ecornis, Leidy, 317. levigata, Penard, 317. Ceratobunus, Thorell, 476; mentioned, 474, 483. annulatus Thor., (Upper Burma), 476; mentioned, 468, 473. bimaculatus, TZhor., 476; tioned, 468. brevipes * , tioned, 468, 472, 474. caleuttensis *, C. With, 477; men- tioned, 468. lugubris, Thor. (Upper Burma), 468, 476. pulcher*, C. With, 476; men- tioned, 468. quadricornis, 467. Ceratophyllum, Hat., 332. Ceriopora globulus, Reuss, 198. Cetacea, dental lamina in, 277. Chalarothoraca, 339. Chapman, F., On some New and Interesting Foraminifera from the Funafuti Atoll, Hllice Islands, 1-27. Smitt, mentioned, men- Thor. (Sumatra), C. With, 478; men- ! INDEX. Chapman, F., On some Foraminifera of Tithonian Age from the Stramberg Limestone of Nesselsdorf, 28-82. Foraminifera from the Lagoon at Funafuti, 161-210. On the Foraminifera collected round the Funafuti Atoll from shallow and moderately deep water, 379-417. On some Ostracoda from Funa- futi, 417-4383. Chara hispida, Linn., 314, 339. Cheilostomella ovoidea, Reuss, 402. Chenopus (Aporrhais) pes-pelecani, Linn., mentioned, ftnote 437. Chilton, C., The Subterranean Amphi- poda of the British Isles, 140-161. Chirogide, 284. incisors of, 284. Chlamydophora, Har/., 336-447. Chrysochloris, Cwv., mentioned, 447, 448. Chrysomelinz, 212. Chytra, Moore, 434-442. — Kirk, Sm., 454, 486, 437, 441, 442, Cirolanidz, 296. Clathrulina, Cienkowski, 341. eleoans, Crenk., 341. Clausidiidee, 213, 219. Clausidium, Kossm., 213, 215, 218, 219, 221,228. testudo, Kossm., 213. Clavulina, @’Orb., 186. angularis, @’ Orb., 382, 387, 401. communis, @’ Orb., 401. cylindrica, Hantken, 401. Parisiensis, @’ Orb., 186, 207. Cliona, Grant, 389. Clymene, Sav., 232. Coccinella bipunctata, Linn., 370. Cochliopodium, Hertwig g Lesser, 312. bilimbosum, Leidy, 312. longispinum *, West, 342; men- tioned, 313. minutum*, JVest, 342, mentioned, 312. pellucidum, ol2 pilosum, Hertwig & Lesser, 313. vestitum, Archer, 313, 314. Ceelogenys, Z//., 285. Copepoda, 306. Copilia, Dana, 221. Corallina, Lin., 397. Corallum of Turbinaria, by 8. Pace, 358-365. Cornuspira, Schultze, 178. involvens, Rewss, 178, 207, 381, 400. Hertwig & Lesser, INDEX. Corvus frugilegus, Linn., 357. Coryexide, 221, 222. Coryezxus obtusus, Dana, 306. Couymoiorsllie, Hincks, mentioned, 52. Cosmobunus, Sim., mentioned, 470. Cotyloplanidie, 41. Crangon sculptus, Bel/, 292. Crangonyx, Spence Bate, 142, 146, 154. compactus, Chalton, 157. mucronatus, Horbes, 157. subterraneus, Spence Bate, 144, 155, 157. Crania, Refz., 389, 391, 592. Cressia dubia, Spence Bate, 301. Cressidze, Stebbing, 301. Cribrilina acanthoceros, MacG., 64. annulata, Habr., 64, 99. Balzaci, Azd., 64. clithridata, Waters, 64. figularis, Johnst., 64. furcata, Hincks, 64. Gattyx, Busk, 64. hippocrepis, Hincks, 64. lahiosa, Busk, 64. latimarginata, Busk, 64. monoceros, Bush, 49, 64. nitido-punctata, Smitt, 64. philomela, Bush, 64. , var. adnata, Bush, 64. punctata, Hassall, 48, 62, 64, 86, 99: 142, radiata, Awd., 64. reniformis, Ortmann, 64. scutulata, Busk, 64. setirostris, MacG., 64. setosa, Waters, 64. speciosa, Hincks, 64. tubulifera, Hinchs, 64 Cristellaria, Lamarck, 31. articulata, Rewss, 403. Bronni, Rémer, 31, fig. 32. ealva, Wisniowski, 31, fig. 52. erepidula (/, § M.), 403. cultrata, Mont., 32, 403. gibba, @’ Orb., 31, 403. mirabilis*, Chapm., 410; men- tioned, 403, 417. obtusata, Reuss, 403. orbicularis, @ Orb., 4035. reniformis, @ Orb., 403. rotulata, Lam.,403 ; mentioned, 28, 32. Schloenbachi, Rewss, 403. tenuis (Born), 4038. variabilis, Reuss, 403. yarians, Bornemann, mentioned, 28. Cryptodifflugia, Penard, 326. Cryptodrilus, etcher, 109, 113. cooraniensis, Spencer, 114, 128, 129. 127, 515 Cryptodrilus illawarree, Fletcher, 118, yy, Wes NBS) Bi), weil, ASB) lleley 139. Cryptura, Z7/., 347. Cuma pulchella, Sars, 254. scorpioides, Mont., 294. Cumacea, 294. Cumella pygmeza, Sars, 294. Cyanops asiatica, Lath., 352. Cyclaspoides cornigera, Sars, 294. Cyclicopora, Hincks, mentioned, 71. Cycloclypeine, 21. Cycloclypeus, Carpenter, 21; mentioned, 6, 414. Carpenteri, Brady, 22; mentioned, 7, 21, 24, 25, fig. 27, 388, 390, 391, 394. —— form A, 407, 412. form B, 407, 412. Guembelianus, Brady, 22; tioned, 21-25. Cyclopidz, 222. Cymbalopora, Hagenow, 189; mentioned, + men- Poeyi, @’Orb., 189, 208, 382, 387, 405. ——, var. squamosa, @’ Orb., 382, 385, 405. , depressed var., Brady, 385. tabelleeformis, Brady, 189, 208, 382, 405. (Tretomphalus) bulloides, @’Ord., 189, 208, 382, 405, 411, 415. ( ) inversa*, Chapm., 411; mentioned, 405, 417. Cyphoderia, Schlumberger, 330. ampulla, Leidy, 330. margaritacea. Schlumberger, 330. , var. major, Penard, 330. Cypridid, 418. Cypridinide, 482. Cythere, Miller, 424. acupunctata, Brady, 424. caudata, Brady, 426. deltoides, Brady, 426. fortificata, Brady, 425. lactea, Brady, 426. obtusalata, Brady, 424. pectunculata *, Chapm., 425, 426, 453. phylloides * , Chapm., 424 433. prava, Baird, 426. stolonifera, Brady, 425. Wyville-LThomsoni, Lrady, 425. Cythereis prava, Baird, 426. Cytherella, Jones, 433. cingulata, G. S. Brady, 433. venusta, G. S. Brady, 459. Cytherellide, 453. Cytheride, 424. Cytheridea margaritea, Brady, 429. 516 Cytheropteron, Sars, 431. assimile, Brady, 431. elatum, Sars, 431. intermedium, Brady, 431. longicaudatum, Brady, 431. scaphoides, Brady, 431. Cytherura, Sars, 480. marcida, Brady, 480. Dactylospherium, Heriwig § Lesser, 310. radiosum, 342. vitreum, Hertwig § Lesser, 310. Dasyprocta, J//., 285. Dasyuride, 445. Dasyurus, Geoffr., mentioned, 277, 445. viverrinus, Shaw, 449, 451, 454. Dasyurus and Perameles, Shoulder- Girdle in, by Robert Broom, 449-454. Desmothoraca, 341. Dexamine dolichonyx, Nebeski, 304. spinosa, Mout., 304. Thea, Boeck, 304. Diamysis bahirensis, Sars, 294. Diastylis rugosa, Sars, 294. Diastyloides biplicata, Sars, 294. Diazeuxia, Jullien, 70.~ Didelphys, Zinn., dental lamina in, 277 ; mentioned, 446. Didymogaster, [Vctcher, 109. Difflugia, Leclerc, 317. acropodia, Hertwig & Lesser, 520. acuminata, Hhrenb., 308. ——,, var. amphora, Penard, 319. Blochmann, 310, 311, 342. amphora, Leidy, 319. ampulla, Ehrenb., 380. areolata, Ehrenb., 327. assulata, Ehrenb., 323. avellana, Penard, 317. bacillariarum, Perty, 319. bacillifera, Penard, yar. inflata, Penard, 317. carinata, Archer, 322. ciliata, Ehrenb., 327. collaris, Khrenb.. 321. compressa, Carter, 317. constricta, Leidy, 317. corona, Wallich, 320. crenulata, Leidy, 320. enchelys, Ehrenb., 329. entochloris, Leidy, 317. fallax, Penard, 317. globulosa, Dujardin, 320, 342. globularis, Wallich, 320. helix, Cohn, 321. lageniformis, Wallich, 318. lanceolata, Penard, 317. lobostoma, Leidy, 320. var. elegans, Penard, 319, | INDEX. Difflugia dwcida, Penard, 317. marsupiformis, Wallich, 317. peltigeracea, Carter, 321. platystoma, Penard, 317. pyriformis, Perty, 317, 318, 342. , var. vas, Leidy, 318. saxicola, Penard, 317. seminulum, Whrenb., 329. sphagni, Leidy, 323. spiralis, Ehrenb., 321. strigosa, Khrenb., 327. symmetrica, Wallich, 321, 323. urceolata, Carter, 318, 319. vas, Leidy, 318. (Catharia) elegans, Leidy, 325. Digaster, Perr., 109, 121. armifera, Mletcher, 121, 128, 129, 130, 131, 138. brunneus, Spencer, 122, 127, 128, IDB), ay, Tls}8). excavata, Wvletcher, 128, 128,' 129, NB BBE si, Nels. gayndahensis, Spencer, 122, 127, 128, 129. minor, Spencer, 122, 127, 128, 129. queenslandica, Hetcher, 124, 128, 129, 133, 188. sylvatica, Fletcher, 123, 128, 129. Digby, Lettice, On the Structure and Affinities of the Tanganyika Gastro- poda Chytra and Limnotrochus, 434442. Diplophrys, Barker, 332. Archerii Barker, 332. Diplotrema, Speacer, 109, 114. fragilis, Spencer, 114, 128, 129, 130, 134, 136, 1388, 139. | Diporocheeta, Beddard, 109, 110. Bakeri, Fletcher, 116, 127, 128, 129, 131, 138. Copelandi, Spencer, 115, 127, 128, 129, 130; 131. Discopora coccinea forma Peachii, Smitt, forma labiata, Sinitt, 90. elongata, Smitt, 97. emucronata, Smitt, mentioned, 89. labiata, Smitt, 90. reticularis, va Beneden, 72. sincera, Smitt, 72. Skenei, Smit, mentioned, 80. Discorbina, Parker & Jones, 11, 164, 190. acuminata*, Chapm., 385; men- tioned, 382, 413, 416. allomorphinoides (Reuss), 405. araucana, d’Orb., 190, 208, 382, 405. Bertheloti (d’Orb.), 405. biconcava, Parker & Jones, 405. concinna, Brady, 191, 208, 382, AOD. INDEX. Discorbina globularis, @’ Orb., 190, 208, 382, 387, 405. opercularis, @’ Orb., 405. orbicularis, Terguem, 191,208, 382, 405. parisiensis, @’Orb., 405. patelliformis, Brady, 4005. pileolus, d’Ord., 12, 191, 208. pulvinata, Brady, 405. rarescens, Brady, 192, 208, 405. rosacea, @’Orh., 382, 405. rugosa, @’ Orb., 190, 191, 208, 382, 405. Saulcii, d’Orb., 190, 208. tabernacularis, Brady, 382,385,413. tuberocapitata *, Chapman, 11 ; mentioned, 12, 192, 208, 405. turbo, @’Orb., 405. Vilardeboana, d’Orb., 191, 208, 382, 405. Ditrema, Blees., 319. Dolichoplana, Mose/., mentioned, 41. Dolichotis, Desm., molar of, 277 ; upper molar of, 285. platycephalica, molar of, 290. - Dromeus, Vieill., 345, 344, 346, 347. novee-hollandiz, Viei//., 355. Ducts (Vasa Deferentia and Spermidu- cal Ducts) of Australian Earthworms, 127 Dynamene rubra, Mont., 296. Ameghino, wpper Earthworms, On the Structure of the Spermiducal Glands and associated parts in Australian, by Miss G. Sweet, 109-139. Eehinopyxis aculeata, Clap. & Lachm., 315. Echiurus unicinctus, Drasche, 211, 215, 227. Ehrenbergina serrata, Ieuss, 402. Embletou, Miss A. L., Goidelia japonica, a new Hntozoic Copepod from Japan, associated with an Infusorian (Tricho- dina), 211-229. Hmbleton, Miss A. L., and Warburton, C., The Life-history of the Black- currant Gall-mite, Hriophyes (Phy- toptus) ribis, Westw., 866-378. Endothyrine, 29. Entalophora, Zam., mentioned, 50. Entozoic Copepod from Japan, 211- 229: Koecardia, Ameghino, 285. Eregasilide, 222. Hrinaceus, Linn., mentioned, 445-448. Eriophyes (Phytoptus) ribis, Westw., Life-history of, by C. Warburton & Alice L. Embleton, 366-378. Briophyide, 367. 517 Eriopis, Brugz,, 146. Eriopsis read Hriopis, Brugz., 146. Eromyidx read Hryomyide, 285. EKryomyide, 283. Erythrops elegans, Sars, 293. Eschara, Lam., 71. cervicornis, Busk, 77 ; mentioned, forma Lepralie, Smitt, 78. elegantula, d’Orb., mentioned, 50, 81, 82. elegantula, Sritt, 81. fascialis, d’ Orb., mentioned, 81. grandipora, Blainville, mentioned, 5 lobata, Lamarck, mentioned, 85. palmata, Sars, 71. pavonina, @’Orb., mentioned, 82, fig. 103. propinqua, Smitt, mentioned, 79. rosacea, Busk, mentioned, 85. saccata, Busk, 81, 82. Skenet, var. tridens, Busk, 80. solida, Vergelius, 71. verrucosa, Smitt, 79. Hscharella, Gray, 71. Jacotint, forma lamellosa, Smitt, 88. Landsborovii, Simtt, mentioned, 90 Legentilii, forma prototypa, Smitt, 66, 73. palmata, Smt¢, 71. porifera, forma typica, Smitt, 75. Escharipora punctata, Smitt, 62. Escharoides rosacea, Bush, mentioned, 85, 86. Sarsii, Simitt, 85; mentioned, 48, 62, 63, 74, 78, 99. verruculata, Bush, 49. Esthonyx, Cope, 283, 284, 283. Eucalanus attenuatus, Dana, 306. Hucheta marina, Prestandrea, 306. Eucrangonyx, Séebbing, 146, 15+. Vejdovskyi, Stebbing, 155, 156. Huglypha, Dujardin, 327. alveolata, Dujardin, 327. brachiata, Leidy, 328. brunnea, Leidy, 329. ciliata, Leidy, 327. ——, var. strigosa, Ehrenb., 327. compressa, Carter, 327. curvata, Perty, 330. globosu, Carter, 328. levis, Perty, 327. margaritacea, Wadllich, 330. minima, Perty, 327. mucronata, Leidy, 328. spinosa, Carter, 328. strigosa, Leidy, 327. tincta, Archer, 529. tuberculata, Dujardin, 327. 518 INDEX EHuglyphina, 327. Eupagurus angulatus, Asso, 292. Euromyidz read Eryomyide, 285. EHurostopus nigripennis, Ramsay, 353. Hurydice achata, Slabécr, 296. pulchra, Leach, 296. Husyllis monilicornis, Malmgr., 106. Flabellifera, 295. Fletcherodrilus, Michaelsen, 109, 115. unicus, Fletcher, 115, 128, 129, 131, 138. Flustra, Linn., 71. abyssicola, Sa7s, mentioned, 59. carbasea, HU/. § Sol., 59, 98. membranacea, IVitsche, mentioned, ftnote 67. membranaceo-truneata, Smztt, 59, 98. securifrons, Pal/., mentioned, 59. separata, Waters, mentioned, ftnote 59. serrulata, Busk, mentioned, 59. solida, Stimpson, 71. spitzbergensis, Bidenkap, men- tioned, 59. Foraminifera, On some New and Inter- esting, from the Funafuti Atoll, by F. Chapman, 1-27. , On some Foraminifera of Titho- nian Age from the Stramberg Lime- stone of Nesselsdorf, by F. Chapman, 28-52. from the Lagoon at Funafuti, by F. Chapman, 161-210. collected round the Funafuti Atoll from shallow and moderately-deep water, by F. Chapman, 379-417. Franz-Josef Land, Bryozoa from, col- lected by the Jackson-Harmsworth Expedition, 1896-97, by A.W. Waters, 48-105 Frondicularia spathulata, Brady, 403, 417. Funafuti, On some New and Interesting Foraminifera from the Funafuti Atoll, Ellice Islands, by F. Chapman, 1-27. Gagrella, Stol., 474, 483. acuaria, Thor. (Sumatra), 467. snescens, Zhor. (Upper Burma), 496 ; mentioned, 468, 473, 486, 506, 507. Albertisii, Thor. (New Guinea), 466. albicoxa, Loman (Java), 467. albitarsis, Sim. (Sumatra), 467. amboinensis, Zhor. (Amboina), 467. armillata, Zhor., 492; mentioned, 468, 484. Gagrella atrata, S/ol.,494; mentioned, 467-469, 483, 485, 486, 496, 504. atrorubra, Sim. (Lower Burma and Malay Peninsula), 468. bicornigera, Sim. (Lower Burma and Malay Peninsula), 468. bidentata, 7o7., mentioned, 468. binotata, Sim., 499; mentioned, 467 ; (Lower Burma and Malay Peninsula), 487. bipeltata, Zor. (Aru), 467. biseriata, Stm. (Lower Burma and Malay Peninsula), 468. bispinosa, Karsch, 505 ; mentioned (Ceylon), 468, 488. celerrima, Loman (Java), 467. cervina, Szm., 499: mentioned (Lower Burma and Malay Pen- insula), 467, 487. ceylonensis, Karsch, 499; men- tioned, 468, 487. cheetopus, Zhor., 492; mentioned, 468, 484. concinna, Thor. (Sumatra), 467. conspersa, Thor. (Java), 467. crux*, C. With, 490; mentioned, 468, 484. dentata*, C. With, 488; men- tioned, 467, 469, 472, 484. docilis, Loman ?, mentioned, 468. Doleschallii, Zhor. (New Guinea), 466. elegans, Sim. (Philippines), 466. ephippiata, Thor. (Sumatra), 467. erebea, Zhor., 492; mentioned, 468, 485. Fee, Thor., 494; mentioned, 467, 485, 486, 496. ——, var. humeralis, C. With, 485. ferruginea, Loman, (China), 468 ; (Japan), 468. flavimaculata*, C. With, 498; mentioned, 468, 472, 473, 486. fragilis * , C. With, 493 ; mentioned, 468, 485. , var. bispinosa * , C. With, 494. Hansenii*, C. With, 503; men- tioned, 468, 4738, 483, 488. Hasseltii, Thor. (Sumatra), 467. hirta* , C. With, 492; mentioned (India), 468, 472, 485. histrionica, Thor., 499 ; mentioned (Upper Burma), 468. 487. illusa, Sim. (Lower Burma and Malay Peninsula), 468. imperator *, C. With, 501; men- tioned (Nicobar), 467, 483. 487. , var. dentata *, C. With, 501; mentioned, 488. INDEX. Gagrella imperator, var. unispinosa * , C. With, 502; mentioned (Anda- mans), 467, 469, 471, 488. insculpta, Pocock (Borneo and Java), 467. laticlavia, Zhor., 496; mentioned (Upper Burma), 468, 486, 498. lepida, Zhor. (Lower Burma and Malay Peninsula), 468, 491. Lomanii, Zhor. (Java), 467. longipalpis, Zor. (Borneo), 467. luzonica, Loman (Philippines), 466. Maindroni, Sim., 496; mentioned (India), 468, 485. minax, Thor, 492; mentioned (Upper Burma), 468, 484. mirabilis*, C. With, 497; men- tioned, 468, 485. monacantha, Herbst, 468, 505. monticola, Thor. (Sumatra), 467. niasensis, Zhor. (Nias), 467. nobilis * , C. With, 489; mentioned, 468, 474, 484. nocticolor, Thor., 494; mentioned (Upper Burma), 468, 485, 496. obscura, Sim. (Philippines), 466. patalungensis, Sim. (Lower Burma -and Malay Peninsula), 468. plebeia, TZhor., 502; mentioned, 473, 488. pullata, Zhor. (Sumatra), 467. quadrivittata, Sim., 496; men- tioned, 468, 486. ramicornis, Thor. (Java), 467. rufescens, Zhor., 496; mentioned, 486, 497. serobiculata, Thor. (Borneo), 467. semigranosa, Sim. (Lower Burma and Malay Peninsula), 468. sepia, Loman, 505; mentioned (Java), 467, 472. signata, Stoliczka, 496; mentioned (India), 468, 486. simplex, Loman, (Java), 467. sordidata, Thor., 492; mentioned (Upper Burma), 468, 471, 484. spinulosa, Zhor., 492; mentioned (Upper Burma), 468, 472, 485. splendens*, C. With, 506: men- tioned (China), 468, 470, 472, 473. Stoliezez *, C. With, 497; men- tioned (Lower Burma and Malay Peninsula), 468, 486. tenuis, Loman (Flores), 467. testacea, Thor. (Java), 467. triangularis *, C. With, 499; men- tioned (India), 468, 472, 473, 487. mentioned, 519 Gagrella triangularis, var. fusca*, C. With, 500; mentioned, 487. tristis, Zhor., 502; mentioned, 488. varians*, C. With, 503; men- tioned (India), 468, 483, 496. —-—, var. dentata*, C. With, 504. variegata, Dol. (Java), 467. vestita, Zhor. (Java), 467. viridis, Dol. (Java), 467. volcanica, Dol. (Java), 467. _ xanthostoma, Zhor. (New Guinea), 466. (Melanopa) Aurivillii, Zor. (Java), 467. ( )plebeia, Zor. (Upper Burma), 468. (——-) tristis, Zzor.(Upper Burma), 468, 472. Galathea, Rathke, sp., 292. Galeopithecus, Padl., mentioned, 445, 448, Gallus bankiva, var. domestica, Lriss., 356. Gammarella brevicaudata, Milne-Hdw., 305. Gammaridz, 298, 304. Gammarus (Fabr.), Dana, 146. fontanus, Moniez, 153. marinus, Leach, 304. pulex, Fabr., 141. puteanus, Hamann, 151, 153, 155. puteanus, Koch, 145, 147. scissimanus, Costa, 305. subterraneus, Leach, 141, 142, 147. Gastropods Chytra and Limnotrochus, On the Structure and Affinities of the Tanganyika, by Lettice Digby, 434-442. Gaudryina attenuata*, Chapm., 409; mentioned, 401, 416. baceata, Schwager, 409. rotunda *, Chapm., 409; tioned, 401, 416. rugosa, @’ Orb., 401. Gemellaria loricata, Zinw., 98; men- tioned, 50. Geodesmus bilineatus, Mecznikojf, men- tioned, 42. Geoplanide, 41. Giardella, 214, 215, 217, 219, 221, 2238. Gitana Sarsii, Boeck, 301. Globigerina, @’ Orbigny, 164, 157. xquilateralis, Brady, 404, 415. bulloides, @Orb., 187, 208, 404, 415. , var. triloba, Reuss, 188, 208, 404, 415. conglobata, Brady, 404, 415. cretacea, Brady, +10. digitata, Brady, 404, 415. men- 520 INDEX. | Globigerina dubia, Hgger, 404, 415. Halimedon recetirostris, Della Valle, Dutertrei, d@’ Orh., 404, 415. 3038. pachyderma, Ehr., 404, 415. rubra, d Orb., 404, 415. sacculifera, Brady, 382, 404, 415. subcretacea * , Chapm., 414; men- tioned, 404, 415, 417. Globigerina-ooze, 420, 425. Globigerine, 410. Globigerinidee, 187. Glyciphagus domesticus, de Geer, men- tioned, 63. Goidelia japonica—a new Entozoic Copepod from Japan, associated with an Infusorian (Trichodina), by Miss A. L. Embleton, 211-229. Goidelia japonica*, Hmbleton, 2293 Goodrich, E. 8., Observations on Syllis ' yivipara, Krohn, 105-108. Goplana, Wrzegniowski, 154, 158. polonica, Wrzesniowski, 154. Gromia, Dwardin, 312, 313, 381. fluviatilis, Dujardin, 331. granulata, Schulze, 382. hyalina, Schlumberger, 331. stagnalis *, West, 331 ; mentioned, 342. terricola, Leidy, 331. Guerna coalita, Norm., 304. Gymnophrys, Cienkowski, 311. cometa, Cienkowshi, 311, 342. Gymnura, Vig. § Horsf., mentioned, 266, 287, 447, 448, Gypsina, Carter, 198; mentioned, 2, 163, 164. globulus, Rewss, 198, 209, 382, 407. 227, inherens, Schultze, ftnote 17, 198, 209. 382, 387, 407. melobesioides, Carter, 201. melobesoides, Carpenter, 396. yesicularis, Parker & Jones, 198, 209, 210, 387, 407. ——-, yar. discus, Goés, 199, 209, 200; mentioned, 209, 210, 407. ——-, var. squamiformis* , Chapin., 200; mentioned, 165, 209, 210, 407. Haddonia, Chapm., 6; mentioned, 8, 164, 183. minor * , Chapin., 884 ; mentioned, 382, 400, 416. Torresiensis, Chapm.,6 ; mentioned, 7, 183, 207, 384, 389, 391-393, 400, 416. Halimeda, mentioned, 1, 6, 165, 182, 183, 184, 195, 389, 391, 397. , var. monticulus * , Chapm., Haplophragmium, Reuss, 28. agglutinans, @ Orb., 28 ; mentioned, 400. ealeareum, Brady, 409. cassis, Parker, 408; mentioned, 400, 416. nanum, Brady, 400. neocomianum, Chapm., 29, fig. 32. ~ tesselatum *, Chapm., 408; men- tioned, 400, 40%, 416. Harpinia crenulata, Boeck, 306. neglecta, Sars, 300. Hauerina, d’Orb., 178; 164. compressa, @’Orb., 381, 399. ornatissima, Karrer, 178, 207, 381, 399. Hauerinine, 178. Hedimys, Ameghino, 285. Heleopora, Lezdy, 323. petricola, Leidy, 323. picta, Leidy, 323. Heliophrys, Greeff, 336. variabilis, Greett. 336. varians, West, 336. Heliozoa and Rhizopods, Freshwater, by G. S. West, 308-342. Helix nemoralis, Linvn., 33. Hemeschara cruenta, Smitt, 73. sincera, Busk, 72. Hemicentetes, Mivart, mentioned, 447, 448. Hermellidz, 256. Hersilia, Philippi, 212, 213, 214, 224. apodiformis, Philippi, 212, 225. Hersiliidze, 212, 213. Hersiliodes, Canu, 214, 215, 216, 217, 219, 221, 223, 224. Heterophrys, Archer, 337. Fockei, Archer, 338. myriapoda, Archer, 337. Pavesii, Garbini, 338. radiata *, West, 337; mentioned, 342. spinifera, Hertwig & Lesser, 337. varians, Schulze, 336. Heterostegina, d’Orb., 18; mentioned, 2, 163, 164, 165, 205. costata, d’ Orb., mentioned, 20. curva, Mobius, 205. depressa, @’ Orb.,18; mentioned, 20, 205, 209, 382, 387, 394, 412, 414, ——, form A, 407, 412. , form B, 407, 412. , var. simplex, d’Orb., 19. helvetica, Kaufmann, 19. simplex, VOrb., 19. mentioned, mentioned, INDEX. 521 Hippolyte gracilis, Hel/er, 292. varians, Leach, 292. viridis, O¢to, 292. Hipponycidee, 437. Hipponyx conieus, Schumacher, men- tioned, 435. Hippothoa, Zam, mentioned, 50. divaricata, Zam., mentioned, 70. expansa, Dawson, 69 ; mentioned, 99. expansa, Hincks, 69. hyalina, Linn., 70; mentioned, 48, 99. rugosa, Stimps., 59. Hornera, Lam., mentioned, 89. lichenoides, Pont., mentioned, 49, 62, 63. Hyale aquilina, Della Valle, 299. pontica, Rathke, 298. Prevostii, W/.-Edw., 299. Hyalodiscus rubicundus, Hertwig & Lesser, 334. Hyalolampe fenestrata, Greeff, 339. Hyalosphenia, Stein, 325, 330. euneata, Stez2, 325, 342. elegans, Leidy, 325. lata, Schulze, 325. Hydra, Linn., 228. Hydrocherus, 285. Hyperammina, Brad, 182. ramosa, Brady, 182, 207. Hyperiidea, 298. Hypsibunus, Zhor., 474; mentioned, 483. diadematus, Thor., mentioned, 474. vigilans*, C. With, 475; men- tioned, 468, 472, 474. Hyssura, Norm. & Stebbing, 295. producta, Norm. § Stebbing, men- | tioned, 296. spinieauda, Walker *, 295; men- tioned, 307. Hystricidae, 286. Hystricomorpha, cheek-teeth of, 283. Hystrix leucura, Sykes, premolar of, 283, 286; upper molar of, 290. Idmonea atlantica, Forbes. mentioned, | 50. Involutina, Terquem, 29. conica, Schlumberger, 30, fig. 32. Jonesi, Terquem 4 Piette, 29. Remesiana * , Chapm., 29, fig. 32. Iphimedia minuta, Sars, 303. Iphimedidze, 303. Iphinoé serrata, Norm., 294. Tsoétes, Linn., 312, 331, 359, 341. Isopoda, Latr., 295. Jeropsis Dollfusi, Norm., 296. Janiride, 296. Kinetoskias arborescens, Kor. § Dan., mentioned, 58. eyathus, W. Thoms., mentioned, 49. Krithe, Brady, Crosskey § Robertson, 427. producta, Brady, 427. tumida, Brady, 427. Lagena acuticosta, Reuss, 402. lagenoides, V2Ul., 402. radiato-marginata, Parker § Jones, 402. spiralis, Brady, 402. striata, @’ Orb., 402. Lagenide, 30, 187. Lagenipora, Hincks, mentioned, 93. Lasius niger, Fabr., 370. Lecquereusia, Schlumberger, 321. Jurassica, Schlumberger, 321. spiralis, Blochmann, 321. Lecythium hyalinum, Hertwig & Lesser, dol. Leieschara coarcta, Sars, 68. crustacea, Levinsen, 64. subgracilis, Levinser, mentioned, 45, Lemna minor, Linn., 331, 333. Lenticulites rotulata, Lam., 32. Lepralia, Johnst., 71, 87, 93. borealis * , Waters, 73; mentioned, 66, 99. cervicornis, Smitt, 78. cruenta, Norm.,73; mentioned, 74, 13599: elegantula, Levinsen, 81. eliminata, Waters, mentioned, 92. foliacea, H//. §: So/., mentioned, 66, 76, 83. hippopus, Smitt, 75; mentioned, 99; 102: incisa, Busk, mentioned, 75. Jeffreysii, Norman, 88. margaritifera, Quoy §° Gatm., men- tioned, 83, 85. megastoma, Bush, mentioned, 75. plana, Dawson, 65. porifera, Smitt, '75, 99, 101. porifera, Hincks, 75. sincera, Smitt, 72, 99. Smitti, Kirchenp., 66. spatulifera, Smitt, 87. trispinosa, Johnst., 88. tubulosa, Norm., 86. vitrea, Lorenz, mentioned, 75. Leptocheirus guttatus, Grube, 805. Leptochlamys*, West, 325, 326. ampullacea * , West, 325, 342. Leptomysis apiops, Sars, 294. var. Hincks, 522 Leptophoxus falcatus, Sa7s, mentioned, 300. Leucothoé, Leach, 302. spinicarpa, Abildgaard, 302. euryonyx*, 4. O. Walker, 302, 307. Leucothoide, 302. Life-history of the Black-currant Gall- mite, Eriophyes (Phytoptis) ribis, Westw., by C. Warburton and Miss A. L. Embleton, 366-3878. Ligia italica, Fabr., 298. Ligiide, 298. Liljeborgia pallida. Spence Bate, 305. Liljeborgidx, Stebbing, 3005. Limacopside, 41. Limnicythere, Brady, 427; mentioned, 418 fijiensis, Brady, 427. LimnotrochusThomsoni, Bowv., 487-442. Lingulina, d@’ Orb., 30. carinata, d’Orb., 403, 417. , var. seminuda, Hanthen, 403. nodosaria, Feuss, 30, fig. 32. ovalis, Schwager, 31, fig. 32. Liobunum, Koch, 470,471; distribution of, 469; mentioned, 505. rotundum, Latr., mentioned, 470. rupestre, Herbst, mentioned, 470. Lithocolla, Schulze, 388. globosa, Schulze, 338. Lithophyllum, Pxi/., 201, 396. onkodes, Heydr., 201. Lithothamnion, PAi/., mentioned, 1, 25, 9G, 184, 201, 389, 892, 396, 397. Philippii, var. funafutiensis, Fos/., 388, 389, 390. Littorina suleata, Pi/é., mentioned, 439. Lituola cassis, Parker, 408. Lituolide, 6, 28, 185. Lituoline, 6, 28, 183. Lobosa, 309. Loxoconcha, Sais, 397, 427. alata, Brady, 427. australis, Brady, 427. honoluliensis, Brady, 428. tumida* , Chapm., 428, 433. Lysianasside, 299. Macrocypris, Brady, 420. decora, Brady, 420. Macroscelides, Smith, and other In- sectivora, Notes on the Brain of, by G. Elliot Smith, 443-448. Macroscelides proboscideus, Shaw, 443— 448. Macrura, 292. Madrepora, Linn., 358, 359. Mera integrimana, Heller, mentioned, 305. scissimana, Costa, 305. | Mediterranean : INDEX. Meera truncatipes, Della Vaile, 305. Malacostracan Fauna of the Mediter- ranean, Contributions to the, by A. O. Walker, 290-506. Maplestoniasimplex, MacG., mentioned, ol. Marginopora vertebralis, blainville, var. plicata, Dana, 181, 584. Marginulina costata (Batsch), 403. glabra, @’Orb., 403. soluta, Reuss, 403. Marsupials Dasyurus and Perameles, On the Early Condition of the Shoulder-Girdle in the Polyprodont, by Robert Broom, 449-454. Marsupials, pre-milk dentition in, 217. Marthana columnaris, Thor. (Borneo), cuspidata, Loman (Java), 467. turrita, Zhor., mentioned, 468. Contributions to the Malacostracan Fauna of the Mediter- ranean, by A. O. Walker, 290-3506. Megalema marshallorum, Swinhve, 351, 3852, dd0- Megaluropus agilis, Norm., 305. Megamphopus cornutus, Norm., 308. Megascolex, Zemp/let., 109, 116. Dendyi, Spencer, 118, 127, 129, 183, 138: dorsalis, Fletcher, 118, 127, 128, 129; 131, 133; Wes Waonlss: 139. Fielderi, Spencer, 117, 128, 129, 138. Frenchi, Spencer, 116, 128, 129, 133, 136, 138. Hoggii, Spencer, 118, 127, 129. Illidgei, Spencer, 121, 128, 129. intermedius, Spencer, 120, 127, 129, 138. minor, Spencer, 119, 128, 129. rubra, Spencer, 119, 127, 128, 129. tasmanica, Spencer, 117, 128, 129, 133, 136, 139. Megascolides, McCoy, 109, 110. attenuatus, Spencer, 112, 127, 128, 129, 130, 154, 136. australis, McCoy, 112, 128, 129, 130. bobartensis, Spencer, 111, 127, 128, 129, 130, 131. insularis, Spencer, 111, 127, 128, ZS), USO, USI, Isis), intermedius, Spencer, 110, 127, 128, 129, 130, 134, 136, 158, 139. roseus, Spencer, 112, 127, 128, 129, 130, 138. ‘ tuberculatus, FVetcher, 129, 131. 128, 113, 128, INDEX. Megascolides victorix, Spencer, 111, 127, E38), 129), sil, tats), Melanopa, Zhor., mentioned, 474, 485. Aurivillii, Zhor. (Java), 467. conspersa, Zhor. (Java), 467. plebeia, Vhor., 502; mentioned (Upper Burma), 468. | tréstis, Thor., 502 ; mentioned, 468, | 78. Melita palmata, Mont., 305. Melphidipella, G. O. Sars, 304. Me!phidippa, Boeck, 504, 307. | n. sp., 4. O. Walker, 304. 2 sp., 307. | Melphidippidee, Stebbing, 304. Membranipora, Blainv., 72. arctica, d’Ord., 60, 98. arctica, Lorenz, 60. armifera, Hinehs, mentioned, 62. | eatenularia, Jam., mentioned. 60. cornigera, Busi, mentioned, 62. crassimarginata, inchs, mentioned, | 49 , Var. incrustans, Busi, men- tioned, 49. | eraticula, Alder, mentioned, 62. | cymbformis, Hincks, mentioned, 61. Flemingii, Busk, mentioned, 62. flustroides, Hznck:s, mentioned, 61. galatea, var. multifida, Bust, men- tioned, 49. hians, Aincks, mentioned, 91. incrustans, Waters, mentioned, 49. lineata, Zinw., mentioned, 49, 60, 98. lineata, f. Sophie, Smitt, 60. macilenta, Jullien, 61, 9Y. membranacea, Jinn., mentioned, 62. monostachys, Bush, 59, 98. perfragilis, Mac Gil/, mentioned, 49. Sophiz, form matura, mentioned, 61. Sophie, Busk, 60. spinifera, Johnst., 61, 98. spitzbergensis, Bidenkap, tioned, 60, 61. trifolium, S. Wood, mentioned, 62. unicornis, FYem., mentioned, 61. Membraniporide, 49. Menipea, Susk, mentioned, 50, 56. arctica, Busk, 55. duplex, Levinsen, 57. gracilis, Levinsen, 5), 56. Jeffreysii, Norm., mentioned, 57. Normani, Nordg., mentioned, 58. Smittii, Kirchenpauer, 57. Smittiz, Norm., 57. ternata, Hincks, 55. men- , forma duplex, Smit, 57. 523 Meniscoessus, Cope, mentioned, 282. Metaphoxus, Bonnier, 299. Fultoni, Z. Scott, 299. pectinatus, 4. O. Walker, 299, 307. pectinatus, Chevreux, 300. typicus, Bounier, 299. Microdeutopus algicola, 305. Microplana, )jdoushy, mentioned, 41. humicola, Vejdousky, 42. Microporella, Hincks, mentioned, 87. impressa, Awd., mentioned, 71. spatulifera, Smtt, LOO, fig. LOF. Miliolide, 3, 168. Miholina, Wil/iamson, 163, 164, 172. agelutinans, @ Orb., 176, 207, 381, 383, 399. alveoliniformis, Brady, 177, 2 399. amygdaloides, Brady, 399. Della Valle, mentioned, 2, bicornis, Walker & Jacob, 176, 178, 207, 381, 399. Bosciana, @ Orb., 177, 207, 210, 381, 383, 399. Boueana, d@'Orb., 177, 207, 381, 413. circularis, Bornemann, i173, 206, 381, 398. , var. sublineata, Brady, 381, 398. cultrata, Brady, 399. Cuvieriana, d’Orb., 3881, 399. Ferussacii, @’ Orb., 175, 176, 207, 381, 383, 386, 399. Funafutiensis * , Chapm., 178, 207, 210; 381, 399. labiosa, d@’ Orb., 173, 206, 381, 383, 399. Linneana, @’Orb., 176, 207, 381, 399. Linneana, &Orb., 176. macilenta, Brady, 399. oblonga, Montagu, 177, 178, 207, 381, 386, 399. Parkeri, Brady, 175, 207, 381, 386, 399. pulchella, @’Ord., 399. reticulata, d@ Orb., 174, 206, 381, 386, 399. scrobiculata, Brady, 178, 207, 399. seminulum, Linn., 176, 381, 386, Bu), subrotunda, Montagu, 173, 206, 381, 399. Terquemiana, Brady, 174. tricarinata, d’Orb., 174, 206, 381,. 399. , var. Bertheliniana, Brady,. 174, 206, 381, 386, 413. , reticulated var., Millett, 174. 524: Miliolina tricarinata, striate var., W/7/- lett, 174. , var. Terquemiana, Brady, 174, 206, 381, 386, 399. trigonula, Lam., 381, 586, 413. undosa, Karrer, 175, 206, 381, 399: valvularis, Fess, 172, 206, 383. —- (Triloculina) reticulata, dOrb., 174. Miliolinine, 169. Miliolites ringens, Lamarck, 170. Millepora compressa, Sowerby, 77. miniacea, Pallas, 16, 200. rubra, Lamarck, 16. Mitopus, Zhor., mentioned, 470. Mitraria, Raf., 240, 254, 255, 256, 259. Mollia granifera, Barrois, mentioned, 71. hyalina, forma hyalina, Smitt, 70. Monalysidium * , Chapm., 3, 179. cylindraceus, Lam., 381, 384, 415. lituus, Gmelin, 400. polita *, Chapm., 4. Sollasi* , Chapm., 3, 400. Monoculodes griseus, Della Valle, 308. Monoporella spinwifera, Hincks, 74. Montipora, Quoy § Gaim., ftnote 362. Morphology of the Palate in Neo- | enathe, by W. P. Pycraft, 343-357. Mougeotia, Rhizopods feeding on cell- contents of, 333, 3842. Mucronella, Hincks, mentioned, coccinea, Abi/d., mentioned, coccinea, Bidenkap, 87. cruenta, Nordgaard, 74. labiata, Levinsen, 90. | sincera, Nordgaard, 72. spinulifera, Hincks, 74. ventricosa, Hincks, 89. yentricosa, var. connectens, idlev, mentioned, 90. Munnide, 297. Muscide, 213. Myodocopa, 482. Myriophyllum, Rhizopods scarce among, 338, 339. Myriozoum, Donati, 65. coarctwm, Bidenkap, 68. coarctum, Sars, 65, 67, 68; men- tioned, 99. crustaceum, Smitt, 64. 89. 87, 88. ? planum, Hincks, 65. pulchrum, Ortmann, 69. subgracile, d’ Orb., 69; mentioned, 45, 65, 67, 68, 99. truncatum, Pal/., 65, 67. _Myrmecobius, Waterh., mentioned, 277, 446, Mysidopsis, Sars, 293. angusta, Sars, mentioned, 293. INDEX. Mysidopsis didelphis, Norm., tioned, 293. gibbosa, Sars, 293. serraticauda *, Waters, 293, 307. men- Nannastacus longirostris, Sars, 294. Nassopsis, Smith, mentioned, 437. nassa, Smith, 442. Nautilus Beccarii, Linn., 196. crispus, Linn., 203. depressulus, Walker & Jacobs, 202, macellus, var. a, Fichtel & Moll, 208. pertusus, Forskal, 179. repandus, Fichtel & Moll, 196. striatopunctatus, Fichtel & Moll, 208. wmbilicatulus, Montagu, 202. ——. (Lituwus) arietinus, Batsch, 179. Navicula divergens, Ralfs, 318. major, Kuetz.. 318. viridis, Kwetz., 318. Nebela, Leidy, 321. ambigua, Cash, 321. barbata, Letdy, 323. bigibbosa, Penard, 323. carinata, Leidy, 322. collaris, Leidy, 321, 322. , var. lageniformis, Penard, o2i. dentistoma, Penard, 322. flabellulum, Leidy, 322. galeata, Penard, 323. hippocrepis, Leidy, 323. militaris, Penard, 821. Nematodemus, Graff; mentioned, 41. Neognathz, Some points in the Morpho- logy of the Palate in, by W. P. Py- craft, 843-357. Neoniphargus, Stebbing, 145, 145. Nereis diversicolor, O. . Miiller, 107. | Nerine, Johnst., 248. Nesselsdorf, On some Foraminifera of Tithonian Age from the Stramberg Limestone of, by F. Chapman, 28-32. | Nicea plumicornis, Heller, 299. Niphargus, Schiddte, 141, 144, 146. aquilex, Schiodte, 141, 142, 147. croaticus, Jurinac, 157. fontanus, Spence Bate, 142, 144, Naess Forelii, Humb., 149. Kochianus, Spence Date, 142, 144, LO oly Moniezi, Utrzesiowshi, 143. puteanus, Koch, 142, 148, 144, 147, 152, 156. : puteanus, Hosius, 152. , var. Forelii, Humbert, 149. var. Vejdoyskyi, Wrzes~- niowski, 149. INDEX. Niphargus stygius, Schiddte, 141. stygius, Westwood, 147. subterraneus, Leach, 144, 147, 152, 153, 161. tatrensis, Vrzesniowshki, 149. Virei, Chevrewxr, 143. Nitzschia, Hassad/, 331. Nodosaria, Zam., mentioned, 3. comata, Batsch, 402. communis, @'Orb., 402. consobrina, @’Orb., 402. filiformis, @’ Orb., 402. intercellularis, Brady, 403. limbata, @’Orb., 402. obliqua, Linn., 403. pauperata, d Orb., 402. plebeia, Reuss, 402. proxima, S¢vestri, 402. punctata, d’Orb., mentioned, 3. Roemeri, Neugeboren, 402. scealaris, Batsch, 402. , var. separans, Brady, 402. subcanaliculata, Newgeboren, 403. vertebralis, Batsch, 403. Nodosariine, 30. Nonionina, d@’ Orb., 202; mentioned, 164. depressula, Walker § Jacob, 202, 209, 407. scapha, #. g W., 407. umbilicatula, Montagu, 202, 163, 209, Nothoprocta perdicarius, Kittl., 355. Notorycetes, Stirling, mentioned, 446. Nototropis guttatus, Costa, 304. spinulicauda, Costa, 304. Nubecularia, Defrance, 168. Bradyi, Millett, 169, 206, 210, 381, 383, 413. Bradleyi read Bradyi, Millett, 383. divaricata, Brady, 168, 206, 210, 381, 586, 398. inflata, Brady, 169. lacunensis*, Chapm., 169, 206, 210, 381, 386, 413. lucifuga, Defrance, 168, 206, 381, 386, 398. tibia, Jones §& Parker, 898. Nubeculariine, 168. Nummulinide, 18, 202. Nummulitine, 18, 204. Nyctibius, Véei2/., 353. Nyctophilus, Leach, mentioned, 447. Obesiella*, Ridew., 463, 465. lyonsiellee *, Ridew., 463-465. Obesiella lyonsielle * , a new Species of Copepod Crustacean, by W. G. Ride- wood, 463-465. Oceanodroma leucorrhoa, Vietll., 356. Octodontide, 285. Oculinidee, 390. QGidiceride, 303. Cidicerus griseus, Della Valle, 303. Oithina, Ald. § Hane., 222. spinifrons, Boeck, 306. Oncea mediterranea, Claus, 506. Oncobunus, 7hor., mentioned, 505. galeatus, Thor. (Upper Burma), 468. Oniscoida, 298. Operculina incertus, d’Orb., 29. anvolvens, Reuss, 178. Ophthalmidium cornu *, Chapm., 399, 408, 416. inconstans, Brady, 408. tumidulum, Brady, 399. Opiliones palpotares, mentioned, 468. Orbiculina, Lam., 412. Orbitoides, Schaff., mentioned, 18. Orbitolites, Lamarck, mentioned, 2, 163, 164, 180. complanata, Lam., 181, 207, 381, 386, 392, 394, 400, 414. ——, var. Jlaciniata, Carpenter, 384. , var. plicata, Dana, 209, 382, 384, 413. duplex, Carpenter, 182, 381, 400 414. laciniatus, Brady, 384. marginalis, Lam., 180, 207, 210 381, 386, 394, 400, 412, 414. tenuissima, Carpenter, 24. Orbitulina versicularis, Parker & Jones, 198. Orbulina universa, d’Orb., 404, 415. Orchestiidx, 298. Orchomene Batei, Sars, 299. humilis, Costa, 299. Orthoplecta clavata, Brady, 402. Oryzoryctes, Grand., mentioned, 447, 448, Osthimosia, Jullien, mentioned, 50, 93. eveza, Jullien, mentioned, 96. Ostracoda from Funafuti, by F. Chap- man, 417-433. Othelosoma, Gray, mentioned, 41. Otocyon, Licht., 282. Owenia, Della Chiaje, 230, 239, 255, 260. filiformis, Della Chiaje, 250, 260. 251, 209, Pace, 8., On the Anatomy of the Progo- branch Genus Pontiothauma, 2, A, Smith, 455-462. On the Corallum of Turbinaria, 398-365. Pachygrapsus marmoratus, Mabr., 292. Palemon Sguilla, #abr., 292. 526 Palate in the Neognathz, Some points in the Morphology of, by W. P. Py- eratt, 343-397. Paleocavia, 285. Palmicellaria S/enei, var. tridens, Hincks, 80. Pamphagus, Bailey, 330. euryus, Leidy, 331, 335, 342. hyalinus, Letdy, 330. Panthalis, Kinb., 248. Pantopoda, 306. Paracrangonyx, Stehbing, 146, 154. Paranthura nigro-punctata, Lucas, Paratylus guttatus, Costa, 304, 507. vedlomensis, Spence Bate, men- tioned, 304. Patellina corrugata, W2/l., 382, 384, 405. Pelagic Foraminifera, 415. Pelomyxa, Grecff, 309. palustris, Greeff, 309, 310. villosa, Leidy, 510. Peltidiidx, 213. Peneroplidine, 3, 178. Peneroplis, Montfort, 3, 164, 179. arietinus, Datsch, 179, 207, 386, 400. lituus, Gmelin, 381. pertusus, Lorskal, 179, 207, 386, 400. pertusus, Forsk., type ce, Brady, 179. pertusus, Korsk., type d, Brady, 180. Sollasi, Chapm., 381. (Monalysidium) cylindraceus, La- marck, 180, 881, 384, 413. ( ) lituus, Ginelin, 400. (——) polita*, Chapm., 4, fig. 26. (——) (?)politus, Chapm., 180, 207, 210. ( ) Sollasi, Chapm.*, 3, 180, 207, 400. Perameles, Geoffr., mentioned, 444-446. obesula, Geoffr., 451, 454. Perameles and Dasyurus, Shoulder- 296. girdle in, by Robert Broom, 449-454. Perioculodes longimanus, Spence Bate, 303. Perissogaster, M/etcher, 109. Petrogale, Gray, 449, 452. Phalangiide, New and Old, from the Indian Region, by C. With, 466- 509. Phanomys, Ameghino, 285. Phascologale, Wagn., mentioned, Pherusa bispinosa, Nebeski, 305. Philonotis fontana, Brid., 322, Philontis fontana, read Blonoris fon- tana, Brid., 322. Photide, 305. Photis longicaudata, Spence Bate, 306. Reinhardi, Della Valle, 306. Phoxocephalidee, 299. Ore al/. | INDEX. Pp as chelatus, Della Valle, 299. pectinatus, 4. O. Walker, 299. simplex, Calman, 299. simplex, Norm., 299, 300. Phoxus simplex, Spence Bate § Westw., 300, 3507. Phrosina semi-lunulata, Risso, 298. Phtisica marina, Slabber, 306. Phylactelia, Hincks, mentioned, 90, 91. collaris, var., Norm., mentioned, 91. grandis, Hincks, 90. labiata, Seitt, 90, 100. Phytoptide, 367. Phytoptis read Phytoptus ribis, Westw., 366. Placorephalus kewensis, Moseley, men- tioned, 54. Placocysta, Leidy, 328. spinosa, Leidy, 328. Placopsilina, @’Ord., 164, 183. cenomana, @’Orb., 183, 207, 400. Plagiaulacide, 284; incisors of, 284; molar of, 282. Plakopus ruber, Schulze, 334. Planarian Worm, a new European Species of Terrestrial (Rhynchodemus Howesi), by R. F. Scharff, 33-42. Planispirina communis, Seguenza, 399. exigua, Brady, 381, 399. Planorbulina, @’ Orb., 163, 192. aceryalis, Brady, 382, 392, 413. , var. fimbriata * ,Chapm., 193, 208, 210, 405. larvata, Parker § Jones, 165, 193, 382, 405. , var. crispata * , Chapm., 193, 208, 210, 405. Mediterranensis, d’Ord., 192, 208, 385, 405. reticulata, Cajzek, 194. retinaculata, Parker & Jones, 382, 885, 413. vulgaris, var. Jones, 193. Planularia Bronnt, Romer, 51. Platybunus minus, Loman (Sumatra), 467 ; mentioned, 469. Platycopa, 433. Platydemus, Graff, mentioned, 41. Plecanium rugosum, Reuss, 185. Pleonexes gammaroides, Spence Bate, 306. Pleurocope dasyura, A. O. Walker * 297, 307. Pleurogonium, mentioned, 297, 298. Pleuromma abdominale, Claus, 306. Pleurophrys? amphitrematoides, Archer, 380. spherica, .Clap et Lachm., 3380. Pleurotomaria, Defr., mentioned, 437. larvata, Parker & INDEX. Pleurotomaria Beyrichu, Hilyendor/, 442, Pleurotomella, Verril/, mentioned, 456, 459. Podargus humeralis, Vig. § Horsf., 354. Podiceps cristatus, Linn., 356. Podocopa, 418. Polycheta of the Family Ammocharide, by A. T. Watson, 230-260. Polymastodon, Kraatz, incisors of, 284. foliatus, Cope, 284. taoénsis, Cope, mentioned. 282. Polymastodontidex, molar tooth of, 282; incisors of, 284. Polymorphina communis, d’Orb., 403. compressa, @’Orb., 403. gutta, d’ Ord., 403. Seguenzana, Brady, 403. Polymorphinine, 187. Polystomella, Lam., 164, 203. crispa, Linn., 208, 209, 382, 407. macella, Fichtel § Moll, 203, 209, 382, 407. striatopunctata, Hichtel g Moll, 203, 204, 209, 382, 413. subuodosa, Miinster, 203, 204, 209. Polystomellinz, 202. Polytrema, fisso, 1, 2, 15, 164, 200. brunnescens, Dana, 390. megentericum, Carter, mentioned, 17. miniaceum, Pallas, 16, 200, 382, 387-393, 407, 416. , var. alba, Carter, 201, 209, 388, 393, 407. , var. album, Carter, 201. , var.involva * , Chapin., 17, 18, 201, 396. planum, Carter, 201, 209, 210, 387- 393, 396, 407, 413, 416. rubra, Lam., 16. utriculare, Carter, 12, 195. Pomatoceros triqueter, Linn., 107. Pompholyx ophrys, Archer, 339. punicea, Archer, 339. Ponthiothauma abyssicola, 1. A, Sinzth, 459 209, mirabile, EF. A. Smith, mentioned, 455, 456-462. Ponthiothauma, EH. A. Smith, On the Anatomy of the Prosobranch Genus, by S. Pace, 455-462. Pontocypris, Brady, 418. attenuata, Brady, 419. faba, Reuss, 418. sicula, Brady, 419. Pontoporeiidz, 299. Porcellidium viride, Phi/ipp?, 306. Porella, Cleve, 50, 76, 81. acutirostris, Smitt, 66, 76, 78, 83, 85, 99. 527 Porella bicornis, Busk, mentioned, 80. cervicornis, JV/.-Hdw., mentioned, 18; 00> Es compressa, Sow., 77, '78, 85, 99. compressa, Hineks, 77. concinna, Busk, 77; 48, 76, 79, 80, 99. , granular var., Hincks, 79. elegans, A/der, mentioned, 80. elegantula, d’Ord., 76; mentioned, 82; fig. 103. elegantula, Tievinsen, 81. glaciata * , Waters, 78, 99. inflata *, Waters, 83; mentioned, 76, 84, 85, 99. levis, Flem., mentioned, 76, 81, 83. , forma Jepralia, mentioned, mentioned, , var. subcompressa, Busk, mentioned, 84. levis, Smitt, 83. lorea, Alder, mentioned, 76, 80. major, Hincks, 83. malleolus, Hincks, 76. margaritifera, Quoy ¢ Gaim., 76. marsupium, MacG.,76; mentioned, 85. minuta, Norm., mentioned, 76. nitidissima, Hineks, 76. obesa*, Waters, 84; mentioned, 85, 100. patens, Smitt, 77. perpusilla, Bus, mentioned, 76, 81, 82. plana, Hincks, 79, 99, fig. 104. proboscidea, Hincks, 79. propinqua, Simitt, 78, 104. rostrata, Hincks, 76. saccata, Bush, 81, 82; mentioned, 50, 76, 78, 85, 99. Skenei, E/Z. § Sol., 79. , forma plana, Hincks, 79. , var. proboscidea, Hincks, 79, 8). ——., var. tridens, Kirchenpauer, &0, fig. 103. struma, Norm., 77. surcularis, Busk, mentioned, 77. —— (Lepralia) margaritifera, Quoy & Gaim., mentioned, 85. Porina, d’ Orb., mentioned, 87. ciliata, forma dura, Smitt, 74. tubulosa, Norm., 86, 100. Porites, Lam., 364. Posidonia, Bronn, mentioned, 291. Caulini, Kon., 292. Prolagus sardus, f. Wagner, mentioned, 275. Psammocora. Dana, 889. Pseudarthromerus, Karsch, mentioned, 472, LINN. JOURN.—ZOOLOGY, VOL. XXVIII. 38 a28 Pseudarthromerus Karsch (Ceylon), 468, 469. Pseudvchirus, Ogilby, 449, 452. Pseudocuma cercaria, v. Beneden, 294. Pseudodifflugia, Schlumber ‘ger, 330. er acilis, Schlumberger, 330. Pseudofinstra palmata, Sars, 71, 99. solida, Bidenkap, 71. Pseudomma sp., 298. Psittacotherium, Cope, 284. Pteroceras, Lam., mentioned, 437 Pullenia obliquiloculata, Parker & Jones, 404, 415. Pulyinulina, Parker & Jones, 196. auricula, Fichtel & Moll, 407. canariensis, @’ Orb., 407, 415. concentrica, Parker & Jones, 407. elegans, @’ Orb., 407. exigua, Brady, 407. Hauerii, @’ Orb., 407. lateralis, Brady, 407. Menardii, @’Orb., 406, 415. oblonga, Williamson, 196, 209. patagonica, d’Orb., 407. proceda, Brady, 407. punctulata, d’Orb., var. Chapm., 407, 411, 417. repanda, Hichtel ¢ ‘Moll, 196, 209, 407. ,var. concamerata, Mont., 407. tumida, Brady, 406, 415. Pustularia rosea, Gray, 16. Pycraft, W. P., Some Points in the LON phology of the Neognathe, 343- 357. spurius, scabra * , Baeoecslis: Wagl., 357. teniata, Peale, 356. Pyripora catenularia, Jameson, tioned, 50. men- Quadrula, F. #. Schulze, 323. irregularis, Archer, 324, 342 monensis, Cash, 324. symmetrica, . £. Schulze, 323, 324. Quinqueloculina agglutinans, dOrb., 176. Bosciana, Q’Orb., 177, 178. Boueana, d Orb., 177. Ferussacii, d’'Orb., IUDs ornatissima, Karrer, 178. reticulosa, d’Orb., 175 undosa, Karrer, 175. Ramulina cervicornis, Chapm., men- tioned, 5. globulifera, Brady, 404. Reophax scorpiurus, Monéfort, 400. Reptoflustrina arctica, dOrb., 60,61. Retepora Beaniana, King. 97. cellulosa, Linn., 96, 100. | INDEX. Retepora cellulosa, Smitt, 96. elongata, Smmitt, 97, 100. elongata, Levinsen, 97. tenella, Ortmann, 97. Wallichiana, Hincks, 97. Reticulosa, 311. Rhabdammina, Sars, mentioned, 5. Rhabdamminine, 4, 182. Rhamphostomella, Lorenz, 50, 91, 93. bilaminata, Hincks, 104. bilaminata, Lorenz, 92, 104. costata, Lorenz, 91, 100, fig. 104. plicata, Smtt, 92, 100, fig. 104. plicata, Lorenz, 92. scabra, Smitt, mentioned, 92. Rhaphidiophrys, Archer, 339. pallida, Schulze, 339, 342. viridis, Archer, 339. Rhapidodendron album, Mobius, 399. Rhea, Moehr., 344, 345, 346. americana, Viedll., 355. Rhizammina, Brady, meutioned, 5. Rhizopoda, Duj., 309. Rhizopods and Heliozoa, Freshwater, by G. S. West, 308-342. Rhynchodemide, 41. Rhynchodemus, Leidy, 35-42. albicollis, Graff, mentioned, 41. bilineatus, Metschner, mentioned, 35, 41. Howesi *, Scharff, 33, 38, 39. nematoides, Loman, mentioned, ochroleucus, Graff, mentioned, 35. pyrenaicus, Graff, mentioned, 41. Scharfi, Graff, mentioned, 35, 39, 40. terrestris, O. F. Miiller, mentioned, 39, 42. Rhynchodemus Howesi: a new Euro- pean Species of Terrestrial Planarian Worm, by R. F. Scharif, 33-42. Rhynchopora, Hincks, mentioned, 93. longirostris, Hincks, mentioned, 49. Ridewood, W. G., On Obesiella lyon- siellze, a new Genus of Copepod Crus- tacean, 463-465. Rissa tridactyla, Linn., 356. Robulina swbnodoas, Minster, 203. Robulus cultratus, Montfort, 32. Rodentia, two dentitions in, 277. Rosalina araucana, d’Orb., 190. bulloides, d’Orb., 189, ftnote 411. globularis, d’Orb., 190 orbicularis, Terq., 191. Poeyi, @ Orb., 189. rugosa, d’Orb., 190. Sauleii, V’Orb., 190. squamosa, VOrb., 385. Vilardeboana, @Orb., 191. INDEX. 529 Rotalia, Lamarck, 196. Beccarii, Linn., 196, 209. spiculotesta, Carter, 184. sguamosa, d’Orb., 385. Rotaliidee, 10, 188. Rotaliine, 11, 189. Rotalina ammonoides, Reuss, 195. oblonga, Williamson, 196. reticulata, Czjzek, 194. Ungeriana, dOrb., 194. Sagenelia, Brady, 4, 5, finote 182. frondescens, Brady, 4; mentioned, 5, 6, fig. 26. Sagenina*, Chapm., 4, 163, 164, 169, 182. frondescens, Brudy, 4; mentioned, 5, 6, 26, 182, 207, 388, 391-393, 400. Sagrina, d’Orb., emend. Parker § Jones, 187; mentioned, 3. columellaris, Brady, 494. divaricata, Brady, 168. raphanus, Parker & Jones, 187, 208, 382, 387, 403. Salmacina Dysteri, Huxley, 107. Sapphirina, Zhomps., 221. _ Sapphirinide, 213. Sarsiella, Norm., 432. sculpta, Brady, 432. Seharff, R. F., Rhynchodemus Howesi : anew Huropean Species of Terrestrial Planarian Worm, 33-42. Schismopora, MaeGillivray, mentioned, 50, 93. Schizopoda, Latr., 293. Schizoporella, Hincks, mentioned, 84, 93. auriculata, Hass., 66. biaperta, Mich., 66; mentioned, 84. candida, S7impson, 66. cincta, Hincks, 65. cruenta, Hincks, 73. crustacea, Smit, 64, 99. crustacea, Lorenz, 64. elmwoodize * , Waters, 66, 99. Harmsworthii * , Waters, 65, 99. hyalina, Zinn., mentioned, 50. limbata, Lorenz. 66. linearia, Hass., 66. pulcherrima, MacGilliv., tioned, 75. sinuosa, Bus, 66. stylifera, Levinsen, 66. unicornis, Johnst., mentioned, 91. Sciuridez, 278. Sciurus Brookei, Yhomas, mentioned, 278. ’ Prevosti, Desm., 278. vulgaris, Linn., 278. men- Scotomenia, Sim., mentioned, 471, 474. cetrata, Thor. (Upper Burma), 468, 472, 505. Serupocellaria, van Beneden, mentioned, 50, 79. : Delilti, Alder, 54, 55. elongata, Smtz, 58,98 ; mentioned, 56. gracilis, Bus/:, mentioned, 56, 57. Peachii, Bush, 58, 98. scabra, van Beneden, 54, 98; ‘men- tioned, 58, 80. scabra, Norm., 54. , forma eongata, Bidenkap, = 58. Smittii, Norm., 57,98 ; mentioned, 56, 58. ternata, var. gracilis, Smmitt, 55, 98; mentioned, 58. Semiflustrellaria arctica, d’Ord., 60, 61- Serpula, Linn., 388, 389, 390, 392, 416. bicornis, Walker & Jacob, 176. Shoulder-girdle, On the Harly Condi- tion of the, in the Polyprotodout Marsupials Dasyurus and Perameles, by Robert Broom, 449-454. Sigmoilina celata, Costa, 381, 399. Siphogenerina (Sagrina) raphanus, Parker & Jones, 187. Siphonostomide, 213. Siriella Clausii, Sars, 293. Smith, G. Elliot, Notes on the Brain of Macroscelides and other Insectivora, 443-448. Smittia, Hincks, 77, 91; mentioned, abyssicola, Norm., mentioned, 90. Jacksonii *, Waters, 87, 100, fig. 105. lamellosa, Smitt, mentioned, 88, fig. 105. Landsborovii, Johnst., var., 90, 100, 104. , var. porifera, Hincks, 75. Peachii, Johnst., 89, 90, 100. , var. octodentata, Hincks, mentioned, 89. plicata, Hincks, 92. poriferd, Lorenz, 70. preestans, Hincks, mentioned, 88. reticulata, MacGilliv., mentioned, (8% 20. reticulo-punctata, tioned, 90. rigida, Lorenz, mentioned, 90. trispinosa, Jo/zst., mentioned, $9, 90. , var. arborea, Levinsen, 88. , var. lamellosa, Smit, 88, 100. yariolosa, Johnst., mentioned, 89. Hincks, men- 530 Smittia ventricosa, Hass., mentioned, 89, 90. , var., 89, 100, fig. 105. Spaniomys, Ameghino, 286. riparius, Ameghino, upper molar of, 290. : Spergo, Dall, 456. glandiniformis, Dadi, mentioned, 456. Spermiducal Glands, &c., in Australian Farth-worms, by Miss G. Sweet, 109- NBS). Spermophilus, Cuv., incisors in, 279. citillus, Linn., mentioned, 278. leptodactylus, Lichtenstein, men- tioned, 278. Spherastrum, Greeff, 338. conglobatum, Greeff, 338. Fockei, Archer, 338. Spheroidina dehiscens, Parker § Jones, 404, 415. Spheroma serratum, Fabr., 296. Spheromide, 296. Sphagnum, Dillen., 309, 315, 821-825, 328, 339. cuspidatum, Hhrh., 338. Sphenoderia, Schlumberger, 328. Jjissirostris, Penard, 329. lenta, Schlumb., 328, 329. ——, yar. fissirostra, 329. Spirillina, Lhrenb., 10, 164, 188. decorata, Brady, 410. , var. unilatera *, 405, 410, 417. inequalis, Brady, 10, 11, 382, 404. limbata, Brady, 11, 404. , var. denticulata, Brady, 11. spinigera * , Chapm., 10, 188, 208, 210, 382, 404. tuberculata, Brady, 10, 11. tuberculato-limbata * , Chapm., 11, 188, 208, 404. vivipara, Hhrenb., 382, 404. Spirillinine, 10, 188. Spirolina, Lam., mentioned, 3. agglutinans, VOrb., 28. —— (Spirolinites) cylindracea, Lam., 180. Spiroloculina, d’ Orb., 164, 170, 175. acutimargo, Brady, 172, 206, 381, 398. Penard, Chapm., antillarum, d’Orb., 171, 206, 381, - 386, 398. asperula, Karrer, 398. canaliculata, d’ Orb., 381, 413. orenata, Karrer, 381, 398. excavata, @’Orb., 170, 171, 206, 398. foveolata, Egger, 171. INDEX. Spiroloculina fragillissima, Brady, 598. grata, Terquem, 171, 206, 381, 386, 398. impressa, Terquem, 170, 171, 398, 408. lamella, Egger, 171. limbata, @ Orb., 398. —~, Brady, var., 408. nitida, @’ Orb., 171, 381, 386, 398. , var. foveolata, Egger, 171, 206, 210, 381, 386, 398. parvula * , Chapm., 398, 408, 416. planulata, Lam., 171. robusta, Brady, 170, 206, 398. tenuiseptata, Brady, 172, 206, 398. tortuosa * , Chapm., 171, 206, 210. Steatornis, Humb., 354. caripensis, Humb., 357. Stichomys, Ameghino, 286. constans, Ameghino, upper molar of, 290. Strombide, 437. Strombus, Zinn., mentioned, 440. Struthio, Zinn., mentioned, 347. Syleus, Zhor., mentioned, 505. niger, C. L. Koch (India), 468. Syllis Armandi, Claparéde, 105. incisa, Fabr., 106. prolifera, Krohn, 105. vivipara, Krohn, 105. Syllis vivipara, Krohn, Observations on, by E. 8. Goodrich, 105-108. Synchelidium haplocheles, Grube, 303. Systenocentrus, Szm., mentioned, 505. quinquedentatus, Sim. (Cambodja), 467. Sweet, Miss G., On the Structure of the Spermiducal Glands and Associated Parts in Australian Farthworms 109-139. Talpa, Linn., mentioned, 445, 447, 448. Tanganyika Gastropods Chytra and Limnotrochus, On the Structure and Affinities of the, by Lettice Digby, 434-449, Tarsonemus, Can. Fanz., 375. Temora dubia, Lubbock, 306. Testacea, 312. Testes and Sperm-sacs of Australian Harthworms (Sweet), 125. Tetrapteryx, Thunb., 356. paradisea, Licht., 350, 357. Textularia, Defrance, 163, 164, 184. agglutinans, d Orb., 401. , var. porrecta, Brady, 401. concava, Karrer, 401. conica, d’Orb., 30, 185, 207, 382, 401. crispata, Brady, 401. INDEX. Textularia folium, Parker & Jones, 184, 207, 400. gramen, @’Orb., 387, 401. rugosa, Reuss, 185, 207, 382, 387, 401 sagittula, Defrance, 401. , var. fistulosa, Brady, 401. siphonifera, Brady, 387, 401. transyersaria, Brady, 401. trochus, d Orb., 401. Textulariide, 9, 30. Textulariine, 9, 30, 18+. Thalestris rufocincta, Norm., 306. Thecameeba guadripartita, Fromontel, 309. Tillodontia, 283. Tillotherium, Marsh, 284; premolars of, 286. Tims, H. W. Marett, Tooth-Genesis in the Caviidee, 261-290. Tinoporine, 15, 197. Tinoporus, Montfort, emend. Carpenter, 197; mentioned, 2, 164, 165. baculatus, Montf., ftnote 165, 197, 209, 382, 387, 394, 407. —, var. florescens*, Chapm., 386; mentioned, 382, 413, 416. vesicularis, Parker & Jones, 199. Tooth-Genesis in the Caviide, by H. W. Marett Tims, 261-290. Tretomphalus bulloides, d’Orb., 189, 382, 405, 415. inversa* , Chapm., 405, 411, 417. Trichameeba hirta, Fromontel, 309. Trichoda chetophora, Schrank, 340. Sol, Miller, 334. Trichodina, n. sp. * , Hmbleton, 211, 227, 228,.229,. Trichosurus vulpecula, Kerr, mentioned, 449, 452, 454. Triloculina circularis, Bornemann, labiosa, d’Orb., 173. Linneiana, d’O7b., 176. reticulata, d'Orb., 174. tricarinata, d’Orb., 174. valvularis, Reuss, 172. Trinema, Dujardin, 329 acinus, Dujardin, 529. complanatum, Penard, 329. enchelys, Leidy, 329. lineare, Penard, 329. Triteta gibbosa, Spence Bate, 304. Trochammina squamata, Parker & Jones, 400. Trochamminine, 29, 184. Trochus, Linn., mentioned, 437. turbinatus, Born., 442. Pruncatulina, d Ord., 164, 193. Akneriana, d’ Ord., 387, 406. culter, Parker & Jones, 406. 531 Truveatulina echinata, Brady, 406. Haidingerii, d’Ord., 406. humilis, Brady, 406. lobatula, W. & J., 382, 405. preecincta, Karrer, 406. pygmexa, Hantken, 194, 209, 406. refulgens, Montfort, 405. reticulata, Czjzek, 194, 209, 382, 406. rosea, d@’ Orb., 406. rostrata, Brady, 194, 209, 382, 406. soluta, Brady, 406. tenuimargo, Brady, 405. Ungeriana, d@’Orb., 194, 209, 406. variabilis, d’Orb., 193, 209, 382, 405. Wuellerstorfi, Schwager, 405. (Rotalina) reticulata, Czjzek, 194. Turbinaria, Oken, 365, 389. auricularis, Bernard, ftnote 362. calicularis, Bernard, ftnote 362. foliosa, Bernard, ftnote 361. gracilis, Bernard, ftnote 363. magna, Bernard, ftnote 364. peltata, Lsper, ftnote 362. radicalis, Bernard, 362. reniformis, Bernard, ftnote 361. Turbinaria, corallum of, by 8S. Pace, 358-365. Typhobia, Moore, mentioned, 437, 440. Tyroglyphus, Latr., 372. Udenodon Baini, Owen, 454. Urothoé, Dana, sp., 299. Utricularia minor, Linn., 309, 315. vulgaris, Linn., 339. Uvigerina angulosa, Wi//., 403. asperula, Czjzek, 403. interrupta, Brady, 403. porrecta, Brady, 403. —— (Sagrina) raphanus, Jones, 187. Parker & Vaginulina, @’Or6., 31. legumen, Linn., 403. linearis, Mont., 403. truncata, Reuss, 31, fig. 32. Valvulina, @’Orb., 9, 30, 186. cuneiformis * , d’Orb., 30, fig. 32. Davidiana*, Chapm., 9; men- tioned, 10, 186, 207, 382, 387, 413. fusca, Wiil., 401. Oviedoiana, d’Orb., mentioned, 10 paleotrochus, var. compressa, Brady, mentioned, 30. pileolus, d’Ord., ftnote 12, 191. triangularis d’Ord., mentioned, 9. 532 Vampyrella, Cienkowshki, 308, 333. lateritia, Leidy, 335, 542. pedata, Klein, 334. Spirogyre, Cienkowski, 333. Vampyrellida, 308, 335. Vanheurckia rhomboides, Brés., 318. , var. saxonica, Rahenh , 319. Vasa deferentia of Australian Earth- worms (Sweet), 128. | Vermiculum oblongwm, Montagu, 17. . subrotundum, Montagu, 173. Verneuilina, @’Orb., 164, 185. Davidiana, read Valvulina David- jana, Chapm., 4138. propinqua, Brady, 409. spinulosa, fewss, 185, 207, 382, 384, 401. Verpulus spumatus, Sim., 468, 509. Vespa, Linn., nest of, 391. Virgulina Schreibersiana, Czjzeh, 401. -subsquamosa, Egger, 401. Voluta, Linn., mentioned, 439. Volvox proteus, Pallas, 309. Vorticella, Miiller, 228. Walker, A. O., Contributions to the Malacostracan Fauna of the Medi- terranean, 290-306. Warburton, C., and Embleton, Miss A. L., The Life-history of the Black- currant Gall-mite, Eriophyes (Phy- toptus) ribis, Westw., 366-378. Waters, A. W., Bryozoa from Franz- Josef Land, collected by the Jackson- Harmsworth Expedition, 1896-1897, 43-105. Watson, A. T., On the Structure and Habits of the Polycheeta of the Family Ammocharidz, 230-260. Webbina, @’ Orb., mentioned, 5. West, G. S., On some Freshwater Rhizopods and Heliozoa, 308-342. INDEX. With, C., New and Old Phalangiid from the Indian Region, 466-509. Xenophoride, 436. Xenophorus, Blackwall, 436. Xestoleberis, Sars, 428. acuminalis *, Chapm., 429, 433. curta, Brady, 430. ? foveolata, Brady, 430. depressa, Sars, 428. gracilis, Brady, 429. granulosa, Brady, 428. margaritea, Brady, 429. nana, Brady, 430. setigera, Brady, 428. tumefacta, Brady, 430. variegata, Brady, 429. mentioned, Zaleptus, Thor., 479; mentioned, 474. festivus, Thor. (Lower Burma and Malay Peninsula), 467, 473, 479. fuscus *, 479. hirsutus * , C. With * (India), 468, 482; mentioned, 474. inermis, Sim. (Philippines), 466. minutus *, C. With (India), 481 ; mentioned, 468, 471, 474. ramosus, Thor. (Sumatra), 467. simplex, Zhor. (Sumatra), 467. subcupreus, Zhor. (Lower Burma and Malay Peninsula), 467, 479. sulphureus, Thor. (Lower Burma and Malay Peninsula), 467, 479. Thorellii*, C. With (India), 468, 480. trichopus, Thor. (Borneo), 467 ; (Java), 467. Zobesula, Geoffr., 451, 454. Zostera oceanica, Al/m., 292. C. With (India), 468, END OF THE TWENTY-HIGHTH VOLUME. PRINTED BY TAYLOR AND FRANCIS, RED LION COURT, FLEBY STBEET. Jury $l. Price 128. THE JOURNAL OF THE LINNEAN SOCIETY. Vou. XXVIII. ZOOLOGY. No. 179. CONTENTS. Page J. On some New and Interesting Foraminifera from the Funafuti Atoll, Ellice Islands. By Freperick Cuapman, A.L.S., F.R.M.S. (Plates 1-4.)........ 1 II. On some Foraminifera of Tithonian Age from the Stramberg Limestone of Nesseldorf. By Freprrick Coarman, A.L.S., F.R.M.S. (Plate 5.) .......... 28 il. Rayycuopsmus Howzst—A new European Species of Terrestrial Planarian Worm. By R. F. Scuarer, Ph.D., B.Sc., Keeper Nat. Hist. Coll., National Musenm, Dublin. (Communicated by Prof. G. B.- Howées, Sec. Linn. Soc.) (Plate 6.) ............ 33 TV. Bryozoa from Franz-Josef Land, collected by the Jackson-Harmsworth Expedition, 1896-1897. By Arruurn WiitramM Waters, F.L.S. (Plates 7-12.) 43 See Notice on last page of Wrapper. LONDON: SOLD AT THE SOCIETY'S APARTMENTS, BURLINGTON HOUSE, PICCADILLY, W., AND BY LONGMANS, GREEN, AND CO., AND WILLIAMS AND NORGATE. 1900. LINNEAN SOCIETY OF LONDON. LIST OF THE OFFICERS AND COUNCIL. Elected 24th May, 1900. PRESIDENT. Prof. Sydney Howard Vines, M.A., D.Sc., F.R.S. VICE-PRESIDENTS. C. B. Clarke, M.A., F.R.S. F. Du Cane Godman, F.R.S. Trank Crisp, LL.B., B.A. Albert C. L. G. Giinther, Ph.D., ¥.R.S. TREASURER. Frank Crisp, LL.B., B.A. SECRETARIES. B. Daydon Jackson, Esq. | Prof. G. B. Howes, LL.D., F.R.S. COUNCIL. C. B. Clarke, M.A., F.R.S. | B. Daydon Jackson, Esq. Frank Crisp, LL.B., B.A. | A. D. Michael, F.Z.S., F.R.M.S. Prof. J. B. Farmer, M.A. | Clement Reid, F.RB.S. F. ID. Godman, F.B.S8. , A. Barton Rendle, M.A., D.Sc. Henry Groves, Esq. | Dukinfield H. Scott, Ph.D., F.R.S. A.C. L. G. Giinther, M.A., M.D., F.R.S.| Rev. Thomas R. R. Stebbing, F.R.S. Prof. G. B. Howes, LL.D., F.R.S. | Prof. Sydney H. Vines, D.Se., F.R.S. A. Smith Woodward, F.G:S. ASSISTANT SECRETARY. James Edmund Harting, F.Z.S. LIBRARIAN. CLERK. A. W. Kappel. A. R. Hammond. LIBRARY OOMMITTEE. This consists of nine Fellows (three of whom retire annually) and of the four officers ex officio, in all thirteen members. The former are elected annually by the Council in June, and serve till the succeeding Anniversary. The Committee meet at 4 p.M., at intervals during the Session. The Members for 1900-1901, in addition to the officers, are :-— i. G. Baker, Esq. Prof. F. W. Oliver, D.Se. Prof. J. B. Farmer, M.A. Howard Saunders, F.Z.S. Albert C. L. G. Ginther, M.D., F.P.S,| Roland Trimen, F.R.S. W. B. Hemsley, F.R.S. F. N. Williams, L.R.C.P. H. W. Monckton, F.G.S. Notr.—The Charter and Bye-Laws of the Society, as amended to the 19th March, 1891, may be had on application. ‘SHAOUD AUNT ‘NOSHOVE NOCGAVC ‘a seonrpny { ‘dda) ANOLNY ‘NOLMONOW “AM HOVYWOH ‘wainspaty, ‘ASTHO MNVUA OLZ PGlLF wee eces sewers eeses so veeeeee seer csoneere ‘0061 ‘T30T AvIT “yet100 punoj puB SLISTONOA Sy YIM pesuLuIvXe Woeq AVY S}UNODR SuULOsEL0} OUT, * (qsonbeg poomyse MA ) spostOG Bane © pana tai - QO GI 16¢ atatulaiaietelststeretersiajerate(atmielet eistelctaleistorareterst ale (puny Awveaquy @ SB qsoubog, 9 cueyy weg, “IIT “"ZIA) yooyg “300 ged > fuedaog Kemprery odplig qatog 6 0 g00l Saraseet eek meen ares eieigiaisivicciatelsieteieicveters ayelalsievoetite ta Piniistwiateisivialete(clalsrateisielelctslsite.vistaiivs/eta 038 peesaeiensg “qued rad ¢ Gc BONERS: BlUSULDeT UBIPUT JvatLx) (©) 2 (ah f6Y0) Fl Meee manage , sain ils Roe Nee R Mth laten < ie 51s ee eg ME a, el Rey Sc Ps e 210038 “que aed § £e Syr0 iM ao pavog. mez jodo.yey OT 9 GLOV SC a ray eee eee sere Cc eo i ard - S]OStLOD, ys F ‘OOBL ‘420 jap uo syuampsoauy 8 & PVGE 8 & PGF ae 668 Tidy 430g wo sxoyueg ye couvyeg eee San pOsEBe Dena Giouelacinsaeeuete sia be : ae BORDON DBONEDUG suorj1soda10g jo nour}seauT G G BL s[osuoy jo eleg Ag ems Pee BOCs ont gee “(e8uysog puv vay, durpnjout) sesuedxg Ay0q Peete Ole = aie trevereescscee KTOTONVIG PUB SUNULIG SNOUB]eDSsTPL Te 5A 0) be emerson t LT LL G9 cicct “e+ HOTMAIL iG. QOS 3 po0ace sireeeeseeeveeeeees SULOMBIISNITT (US ee tereressesadigaay snoamereosl py Gue OlEee oe Bunun OL Gl Fee aS —! suotqvorlqng jo sosuedxgy She Ge “* sonSoeyegQ pues ssutpeed0.q 9 CAL LOG —————_——- R ST ZSl Se ee ee a S[BuANo (? Gee ane Sod6 “DOSHe Kr HobbosennQda6 +s Surpulg 8 Gl GRP ce - suorjousuuay, 9 IL pelR Feoeeeeeee Seal ace eRe eae --syooq —: suotyeoyang Jo sepeg —: Auwgquy 0 0 28 eiefeievaletetetaetotate pereece lo [elelaloralelsin'siein|siele wleibie bimisie’s)=/e(0/uleln see suorisoduiog 0 OL Gal, kev veeeeeeres wA(b isis vip siaw et sinisic ein cc's ences en eeee aos SOLLeleG Gi (hea oe Booonor aeecnerenee avlosccesersce SuOTINgIyUOD [BnuUy ie ees Spa pe secede its sisishccnan aie “2+ se puR s[ZOy OO BLL rei tereceeeeeeestes gagir UOISsTOIpY Gj AO ne eet GO GUDE GU TOU ODRCELE SUG Ria uyierina yap amieeSHauOlsh sp G9 FOG creme: gquatsaauy Uo ysacaqUy 0 16 SGoeaeeooe Mchgatennta Seg oabee ‘ss @oURANSUT pure sox, TE 9 682 voce @EgT KepE ae] Wo sroquug: ye counpeg P Bae eee 2 “sqhra0a re gsy hn ‘fija190 yn fo sjuau whog pup sydrovayy i IMPORTANT NOTICE. © Journal of Zoology. Volume XXVIII. commences with the present Part 179. The Journal (both Zoological and Botanical) is now issued in THREE PARTS PER ANNUM as follows :— Part I., containing papers read from November to the middle of January, on April Ist. Parr II., containing papers read from the middle of January to the end of April, on July Ist. Parr III., containing papers read in May and June, on Wovember Ist. : A GENERAL INDEX to the first twenty Volumes of the Journal (Zoology). may be had on application, either in cloth or in sheets for binding. Price to Fellows, 15s.; to the Public, 20s. A new CATALOGUE of the LIBRARY may be had on applica- _ tion. Price to Fellows, 5s.; to the Public, 10s NovemsBer 1. Price 10s. THE JOURNAL OF THE LINNEAN SOCIETY. Vou. XXVIII. ZOOLOGY. No. 180. CONTENTS. Page I. Observations on Syllis vivipara, Krohn. By Epwiy S; Gooprica, MEAL Pi.S..:(Plate 13.) oO) dagneGriahinddo.co Sug pnb gba SuEscuanOCeEe OME aGede nce ice ee Oe teetwereesenry SLOSCG : “LOGI “v70¢ pd uo spUIUSIvUT Pate oe A CORIDD OOOO OCOD OSU COO OSONOUCOOOtMOO OUST Don ot eal aes Seer | aed O LI IsseF | O- 21 1SScrF | ae ener ied GEG See a aes as TOSI ‘THdy Wig uo saoyung ye course: OG GGG retirements srosnog, jo dug Zl Sal 7 (e8ujs0g pus vay, Surpnyou) sosuedxy Ay jaf gL ee Keuoneyg pur sarjmrg SNOdUL| [eos eel ee 8 II 9G See cercecece WOT Ng LAy81Cy 6 TI 666 “' SUOIBIYSN IT 8 9 [Li reo ach ap hacer nomiees sy aie | Ge ot = : —: suoHvorqng Jo sosuadxy it “"" sanSo][Rj}VO puwB ssulpsooo01g C0) =o ¢ LT Ig Se SSeS SOON OUI Hint © FILS! Sonor EE pa ee ee SU LIET Q SL IQR cue “* suoTjoBSsURAT GM POF cere cereecee cnet ener ces ge | —: suorwonqng Jo sepug Bere eee ewes seer esoeeredeserser sree eraeSOcesesersreseen suorjisod atog —: Aariqry 0 0 ¢F voree sores |g gt eagy serimetntafrreapeceesseesoicininunstei gis eae 8B) pur Byeory 0 -0--9ar Se aes ee ee Onan par saredayy TL OL GGL ctr etttteetersereseeeeeeteseee s}UsUt}saAuT UO 48ed0q Uy OO ever meee e reer ceseerssescececes eoUuBINSUy pue SOXUT, $ 9 Gl toes e steer erceercosce 0061 ‘Keyl 48] uo SloyuBg 4B eouRleg: ‘3s 5 ‘szuauhing Wie A ae “399007 st hope woul ‘fyatoog unauut rt 2019 of Peer meron rc eres eeecesece ssssettenenneeeeionee cassseesesnegydigogay snOoUU]aOStTY SP eee eearn cree eeororsceresseceseesnrsonverses Sane Sey NOTICE. re} Journal of Zoology. Volume SX 2VITI. commenced with Part 179. The Journal (both Zoological and Botanical) is now issued in THREE PARTS PER ANNUM as follows :— Parr I., containing papers read from November to the middle of January, on April Ist. Part TI., containing papers read from the middle of January to the end of April, on July Ist. Part III., containing papers read in May and June, on November Ist. The publication of the Journal of Botany (No. 244) is un- avoidably delayed. A GENERAL INDEX to the first twenty Volumes of the Journal (Zoology) may be had on application, either in cloth or in sheets for binding. Price to Fellows, 15s.; to the Public, 20s. A. new CATALOGUE of the LIBRARY may be had on appli- cation. Price to Fellows, 5s.; to the Public, 10s, Aprit 1. Price 8s. THE JOURNAL OF THE LINNEAN SOCIETY. Vou. XXVIII. ZOOLOGY. No. 184. CONTENTS. Page I. The Life-history of the Black-Currant Gall-mite, Eriophyes (Phytoptis) ribis, Westwood. By Cxcrn Warsurton, M.A., F.Z.8., Zoologist to the Royal Agricultural Society of England, and Anrce L. Emeeton, B.Sc., 1851 Exhibition Science Research , Scholar, Associate of the University of Wales (Cardiff College). (Plates 338 & 34.) ©... ...2........+-0. 366 II. On the Foraminifera collected round the Funafuti Atoll from Shallow and Moderately Deep Water. By Freprerrck Cuapman, A.L.S., E.R.M.S. (Plates BP So) SO!) cae eis ls jn fo epee teh einen) sep Metuchen age 379 TIT. On some Ostracoda from Funafuti. By Freprrick Cuapman, A.L.S., F.R.M.S. (Plate 37.).......... 417 See Notice on last page of Wrapper. LONDON: SOLD AT THE SOCIETY'S APARTMENTS, BURLINGTON HOUSE, PICCADILLY, W., ' AND BY LONGMANS, GREEN, AND CO., AND WILLIAMS AND NORGATEH. 1902. LINNEAN SOCIETY OF LONDON. LIST OF THE OFFICERS AND COUNCIL. Hlected 24th May. 1901. PRESIDENT. Prof. Sydney Howard Vines, M.A., D.Sc., F.B.S. VICE-PRESIDENTS. William Carruthers, F.R.S. F. DuCane Godman, D.C.L., F.R.S8. Frank Crisp, LL.B., B.A. - D. H. Scott, M.A., Ph.D., F.R.S. TREASURER. Frank Crisp, LL.B., B.A. SECRETARIES. B. Daydon Jackson, Esq. | Prof.G.B. Howes, D.Se., LL.D., F.B.S8. COUNCIL. William Carruthers, F.R.S. | Rev. Canon A. M. Norman, M.A., Frank Crisp, LL.B., B.A. | D.C.L., F.B.S. Herbert Druce, F.Z.8. Clement Reid, F.R.S8. F. D. Godman, D.C.L., F.RB.S. Dukinfield H. Scott, M.A., Ph.D., Prof. J. Reynolds Green, M.A., D.Sc., E.R.S. E.R.S. Rey. Thomas R. R. Stebbing, M.A., Henry Groves, Esq. E.R.S. W. Botting Hemsley, F.R.S. | Prof. Sydney H. Vines, M.A., D.Sc., Prof. G. B. Howes, D.Sc., LL.D., F.RB.S. EBS. B. Daydon Jackson, Esq. | A. Smith Woodward, LL.D. ASSISTANT SECRETARY. James Edmund Harting, F.Z.8. LIBRARIAN. CLERK. A. W. Kappel. A. R. Hammond. LIBRARY OOMMITTHE. This consists of nine Fellows (three of whom retire annually) and of the four officers ex officio, in all thirteen members. The former are elected annually by the Council in June, and serve till the succeeding Anniversary. ‘The Committee meet at 4 P.M., as required during the Session. The Members for 1901-1902, in addition to the officers, are :— E. G. Baker, Hsq. Prof. F. W. Oliver, D.Sc. Alfred W. Bennett, M.A., B.Sc. A. B. Rendle, M.A., D.Se. Herbert Goss, F.G.S. Howard Saunders, F.Z.8. Albert C. L. G. Giinther, M.D., F.R.S,) F. N. Williams, L.R.O.P. H. W. Monckton, F.G.S8. : Notr.—The Charter and Bye-Laws of the Society, as amended to the 20th June, 1901, may be had on application. ‘TO61 ‘WOT Avy ‘suopspny fe ‘SHMOH 2‘) ‘SHAOUD AUNTH | ‘qoeqt00 punoy pus ; YaWMTVM ‘O CHYATVY DNIGEHLS YM LI SLOYONOA OY} UJI PAUTUXO WEdq OAVY s}UNOdDB SUIOSALOJ OUT, 8 § GLOOF “wounsna4T, “ASTHO MNVUA Il e FES of, 6 ® 8 e 6S see ee ereene ae ee rane a ee ee ete aN ee i OOP” SCE GEREN (qsonbeg poom4se AA ) sposuog 0 a1 g1e Yo fLZIDO O OGR TT (puny Areaqry & se qsonbog s,wmeqqueg “1yy “2tA) yo09g “yu00 sod F Cuvdwmop Kemprey odputge yoy Zz IL Gz Z DG | GB Fnac cane ccecsesesinesvecvecescenstoscerttresccace Iieis eieip/e's waieleie seerenete OF SARTO Aynu0y Lemyresy UNSMIUEg UBIpUT yor G FI OLIT of, 601 ® e II 6L01 DDC OUICOCTUICICOCOOGOCONOCRICICO UIC COC OOOH OOD UOCOOCHOCOOCIOOUDOOOIOGHONOOFAONNA yooyg “qUL90 aod go SYLOAA Jo pivog uvyppodo.yoy G . 6 IGLe of, 96 ® d OL QBRGF tt tts recce reese eee e cece cc esetersee cess eseereteeesessseane Cove ere enon reese reer eseeneeresres &.: seesbeeesenis BORIC TIES “LOGT “Y70E pd py wo szuausoauy 0 LI 1eeaF O LL 18eex LT 6ce LOST ‘THY WI0g UO sroyUEg yw comer ZL Sol (e8eysog pue vey, Surpnjout) sosuedxg Ay30qg OI EG ee oeeereoseeceosecsee Areuoryeyg pue suruig snooue][eosipy (SAS laps Soeeeeme ee. TOG oe eas ener eer ONG TBTCE mt 19 6 g 9 ; Tee nena eens tee eeasenenstesserseeeetnssnssasetenees GTOGTTC OolR Ue ee § Sats ee Q 9 |S a cea ari A OOD I Fa Gon ey 6 GI OLT ae ee - —: suoHyvor[qng jo sosuodxgy O) eT “"* sondoTeyVQ pues ssulpeooo1g G 0 OhG __ -—- : Sas g LI 18 Pees seo reese eee esesseseseene SVU. Pf © FIL eosee EEC OD AEE oy OPEL OPO AEGL OSH a\ints 9 QL LGR cee scomoesuery, Zo POG eee eee eeeeeeees roger —: suoHyworTgng. Jo sopeg Fp Ree She EE EN EE —: Aaeiquy 0 0 Gh eteetetttss teetteeeeseesteeereeceeeseeenatenes suorpisodurog OL GGL Beer gee Se eae oe amaG Q OL GGL “etter eee suonngiymo) Tenuy eee ee reorterervecsersetiien Sy) PUB BTHO(). 0 0 921 Hiereeereeseesees sgQur HOISSIMpY 9 Ae: SoonsUnOnogeGenoGocD oe pur sardey” TL OL SGE itititesereseees gimammsoauy UO 4801097 afatsto(eCaeie/siniafelvlesie\elsievelsvisisinisfnieie(e¥slnsre/sisirelelove UBINSUT pues soxey, | SSO OOOO OURO SOOO OnooE ‘ABT JS] UO sioyURg 4v oouvye se #5 ‘spuauhog D's gs pease T wd 4 a “IO6T “Y708 1244p 07 ‘OOGT ‘281 Kup mou ‘hgar00y wvawury ayz fo syuauhog pun sdracagz ae Oe a ee ee Manx oO oO) ei a = _ = ee = a Sn Oe ee dk a A NS te NOTICKH. ‘O Journal of Zoology. Volume XXVIII. commenced with Part 179. The Journal (both Zoological and Botanical) is now issued in THREE PARTS PER ANNUM as follows :— Part I., containing papers read from November to the middle of January, on April Ist. Parr [I., containing papers read from the middle of January to the end of April, on July Ist. Part III., containing papers read in May and June, on November Ist. A GENERAL INDEX to the first twenty Volumes of the Journal (Zoology) may be had on application, either in cloth or in sheets for binding. Price to Fellows, 15s.; to the Public, 20s. A CATALOGUE of the LIBRARY may be had on application. © Price to Fellows, 5s.; to the Public, 10s, Jury 1. Price As. THE JOURNAL OF THE LINNEAN SOCIETY. Vou. XXVIII. ZOOLOGY. No. 185. CONTENTS. Page I. On the Structure and Affinities of the Tanganyika Gastropods Chytra and Limnotrochus. By Lerrice Diesy. (Plates 38-40.).....-.-.--eseeee eee eeee 434 II. Notes on the Brain of Macroscelides and other Insectivora. By G. Exutor Suits, M.D., Fellow of St. John’s College, Cambridge; Professor of Anatomy, Egyptian Government School of Medicine, Cairo. (Figs. 1-2.) 2.2 eee cece cece cee eee cence ences III. On the Early Condition of the Shoulder-Girdle in the Polyprotodont Marsupials Dasyurus and Perameles. By Rozerr Broom, M.D., C.M., B.Sc. (Plate 41.) 449 See Notice on last page of Wrapper. LONDON: SOLD AT THE SOCIETY'S APARTMENTS, BURLINGTON HOUSE, PICCADILLY, W., AND BY LONGMANS, GREEN, AND CoO., AND WILLIAMS AND Se ae fs “1902. ' VeW Wi Poti Wie teak oy ¥ LINNEAN SOCIETY OF LONDON. LIST OF THE OFFICERS AND COUNCIL. Hlected 24th May, 1902. PRESIDENT. Prof. Sydney Howard Vines, M.A., D.Sc., F.R.S. VICE-PRESIDENTS. William Carruthers, F.R.S. Herbert Druce, F.Z.S. I’rank Crisp, LL.B., B.A. Rev. Thomas R. R. Stebbing, F.R.S. TREASURER. Frank Crisp, LL.B., B.A. SECRETARIES. Prof.G.B. Howes, D.Se., LL. D., F.R.S. | D. H. Scott, Ph.D., F.R.S. COUNCIL. William Carruthers, F.R.S. G. S. Saunders, Esq. Frank Crisp, LL.B., B.A. Dukinfield H. Scott, M. nO Ph.D., Herbert Druce, F.Z.S8. F.B.S. Prof. J. Reynolds Green, M.A., D.Sc.,| Rev. Thomas R. R. Stebbing, M.A., F.B.S. E.B.S. W. Botting Hemsley, F.R.S. Col. C. Swinhoe, F.Z.8., F.E.S. Prof. G. B. Howes, D.Se., LL.D., F.R.S.| A. G. Tansley, B.A. B. Daydon Jackson, Hsq. Prof. Sydney H. Vines, D.Sc., F.R.S. George Massee, Hsq. Alfred O. Walker, Esq. CENERAL SECRETARY. B, Daydon Jackson, Esq. LIBRARIAN. CLERK, A. W. Kappel. A. R. Hammond. LIBRARY COMMITTEE. This consists of nine Fellows (three of whom retire annually) and of the four officers ex officio, in all thirteen members. The former are elected annually by the Council in June, and serve till the succeeding Anniversary. The Committee meet at 4 P.m., as required during the Session. The Members for 1902-1803, in addition to the officers, are :— Herbert Goss, F.G.S8. Prof, KF. W. Oliver, D.Sc. Albert C. L. G. Giinther, M.D., F.R.&.| A. B. Rendle, M.A., D.Sc. W. Botting Hemsley, F.R.S. Howard Saunders, F.Z.S. G. R. M. Murray, F.R.S. FB. N. Williams, L.R.C.P. P. Chalmers Mitchell, M.A., F.Z.S. “s40p pny poesie NOCGAV ‘& “NOLMONOW'M WOVYOH ‘PNISAALS “WW uy) 'ZOGL ‘IGT Aepy AWISWGH ONIGLOG M ‘XHWIVM‘O GuauaTVY ‘SHUNIA ‘H AWNCAS *‘Poidoo punoy pues pejipny & I I6egoF ‘wainspa4T “TSTHO MNVUA BOC iciry DOOR COU UCC UIC OOGNNE gO go neO nnn icin inininnrriri ir iri (qsanbog poomyse AA sTOSsuO. q SWeypuog “IPL “ZI4) yoo3g “yu0o ted F Luedurog Avmpreyy ae aieoe Tiere gf ser) Ayinuay Avaprey ensurveg uerpuy yeerg 6 GI Gcsz 86 OS § 6H 0 G 09¢ %o%ZIQO O OGF & Il G66 oJ GG ae I GP Gr FI OLIT 0 G0LOe II 6L01 daa OURS CS Roreteire Vistefalatnlalalulelelglela/nretu’al ele!aisvelels/alelelelerelmerarsvereraininistevereieletavereictove yo0jg “quo aed £0 sy10 0 don i Hl eae o/5 o66 ® J, OL (oh Ae} Cle aD aeRO DG see ecoe eos eeeses fo -cereeee see agree ae astrneettcas oad sesee ca tiaatee erage erage "GOBL ‘YI0G Mp wo squaupsonur I Gl 8686s ee am I GI 868¢F 6 8 ssl “rs ZOGT ‘Tady qWIOg Uo sroyURG 48 couse Se G GE Zit 7 (eseys0g pure voy, sutpnqour) sosuedxm Aq}0q é I OL I0L se ererseesrseoseccscs Aqau0yeyg pur SuyuLnig snoouvypeostpy 0202003 == 8 G Lt SPOS e oreo ssossercesesassscesssseon uOoTyNgI4SICy G OL 806 Peer oeesereneseroesersceossesesscs SUOT BL sn It 9 QaGF ttt eeeeeseeeeseesseeeesBamme OG OV ee | eOiapony Snoaueljeasuit —-: suoryvor[qng jo sesuedxqy OF SG ee 1 it soe —— 09 @ “"" sondO[BIVO puw SdUTpeeD0.1g Tn 7 E veeeeeeeee Sorepurgy OB ae Ge a ee ee oe BILINIO ES O 8 QGP reece eseeespoger OL 4 SGLF rere srooRsURLT, —: Aveaqyy er ae Se pee nd TORRES Q GL pel ee eee reece enerseen eerste sneer pera 0 0 06 fe ge ae ee eee, SUOIEOU UU GB LB ete trttettcteeectesstensrecersisreee ge pure a[BOQ 0 8 G66FT eee revecenesezecereesereeeree STOTINGIIUOD [BN Q 8 1G cere e ganar” pue saredesy 0 O PF Se oe ee ee SOO USS TUNE B SE SL reteset gomeansuy pity sexy, 0 &I S8T ireeneceseessessorrereseses SUUUTISOAUT UO 4sato} UT "p as) od ‘spuauhog eg LL 696 Coe rere reso eseoseeven LOGI ‘ARIAL 4ST uo sioyueg qe oourreg (yet eeesiceg “spdvaoany “GOGT ‘7708 147 02 “TOGT ‘281 ogy woul ‘hzar00y unauury ay} fo sjuaumhng pun sydvaooay — CZ06T ‘497s Ley ing Crpneics oe ‘Suyeoyy Aresioatuny oy 4v poyuoserg) Say eee NOTICE. ° Journal of Zoology. Volume XXVIII. commenced with Part 179. The Journal (both Zoological and Botanical) is now issued in THREE PARTS PER ANNUM as follows :— Parr I., containing papers read from November to the middle of January, on April Ist. Part II., containing papers read from the middle of January to the end of April, on July Ist. Parr III., containing papers read in May and June, on November Ist. A GENERAL INDEX to the first twenty Volumes of the Journal (Zoology) may be had on application, either in cloth or in sheets for binding. Price to Fellows, 15s.; to the Public, 20s. A CATALOGUE of the LIBRARY may be had on application. Price to Fellows, 5s.; to the Public, 10s, VUUVOY & May 1. Price 7s. THE JOURNAL OF THE LINNEAN SOCIETY. Vou. XXVIII. ZOOLOGY. No. 186. CONTENTS. Page I. On the Anatomy of the Prosobranch Genus Pontio- thauma, B. A. Smith. By S. Pace. (Communicated by Prof. G. B. Howes, LL.D., D.Sc., F.R.S., Sec...) CE CRY GIES Se RoR Bee became corer nem nn on 455 Il. On Obesiella lyonsielle, a new Genus of Cmebel Crus- tacean. By W. G. Rrpewoop, D.Sc., F.L.S. (With 463 DUMGUEES) Welle. cae le oe ee sete ese ee Jil. New and Old Phalangiide from the Indian Region. By C. Wire. (Communicated by H. J. Hanssy, BXM.LS.) .. 0.2 cece eet eee ee cert e eee ees 466 Tadi, Title-page, and Contents. See Notice on last page of Wrapper. LONDON: SOLD AT THE SOOCIETY’S APARTMENTS, BURLINGTON HOUSE, PICCADILLY, W,, AND BY LONGMANS, GREEN, AND AND onian Institugy ” a 1903. ‘ LINNEAN SOCIETY OF LONDON. LIST OF THE OFFICERS AND COUNCIL. Elected 24th May, 1902. PRESIDENT. Prof. Sydney Howard Vines, M.A., D.Sc., F.R.S. VICE-PRESIDENTS. William Carruthers, F.R.S. Frank Crisp, LL.B., B.A. Herbert Druce, F.Z.8. Rey. Thomas R. R. Stebbing, F.R.S. TREASURER. Frank Crisp, LL.B., B.A. SECRETARIES. Prof.G. B. Howes, D.Sc., LL.D., F.R.S. | D. H. Scott, M.A., Ph.D., FBS. COUNCIL. William Carruthers, F.R.S. Frank Crisp, LL.B., B.A. Herbert Druce, F.Z.8. Prof. J. Reynolds Green, M.A., D.Sc., E.R.S. W. Botting Hemsley, F.R.S. Prof. G. B. “Howes, D.Se., LL.D., F.R.S. B. Daydon Jackson, Esq. George Massee, sq. G. S. Saunders, Esq. Dukinfield H. Scott, M.A., Ph.D., B.R.S. Rev. Thomas R. R. Stebbing, M.A., E.R... Col. C. Swinhoe, F.Z.8., F.E.S. A. G. Tansley, M.A. Prof. Sydney H. Vines, D.S8c., E. B.S. Alfred O. Walker, Esq. CENERAL SECRETARY. B. Daydon Jackson, Hsq. LIBRARIAN. A. W. Kappel. CLERK. A, R. Hammond. LIBRARY COMMITTEE. This consists of nine Fellows (three of whom retire annually) and of the officers ex officio. The former are elected annually by the Council in June, and serve till the succeeding Anniversary. as required during the Session. the officers, are :— Herbert Goss, #.G.S8. Albert C. L. G. Ginther, M.D., F.R.& W. Botting Hemsley, F.R.S. G. R. M. Murray, F.R.S. P. Chalmers Mitchell, M.A., F.Z.S. The Committee meet at 4 p.m., The Members for 1902-1908, in addition to Prof. F. W. Oliver, D.Sc. A. B. Rendle, M.A., D.Se. Howard Saunders, #.Z.8. F.N. Williams L.R.C.P. ‘scomppnyr { Mosmove NOGAV ‘€ ‘NOLMONOW'M HOVYOH ‘ONIATAIS “WW a ‘COL “MGT Lepr AWISWGH ONILLOL '“M ‘SHWIVM‘O GHYAIV ‘SHNTA ‘H XGNCAS "J00A10) punog pure poytpny & 1 WG9oF “wainspony “TSTHO MNVUA 6 Sl GEz o/, 856 7) g ¢ 6G g Sieisiolajeseis sinid}ejcTels\ale“eiel¥re ocaielvtare’ele_s.e\e\eiotelelaieiaftzs'areieialorsisicisiert ieieisietereisteie HalerGen CIM ooe ice GOUCKIOAUACRADOGUAGE Cc (qsonbagr po0o0a4se AA ) s[osuog 0 ¢ o9¢ %HeIOO O OGr oc: “(puny Areaqry @ se ysonbag SMUqUI “APL “Zi4) yooxg “yueo aad 7 Luedutog AvMrey odpltg 44.107 z IL 26 GO DG [Gh ii ctecee reese canvases eees secven tescene seen eeicveris deeeeee treeeeree oF surg AjINUUY Seaprexy VINSUIMed UBIpUyT Fwatey G PI OLTT oF, 60I ® e II 6L0T weeteece SOOO ere ereereene Pee terre newer ee reeneeseeseerecssesseren yooys "qu90 aad fa SYLO AA jo pizog uvyzipodo.uyeazy IL al aoe of, £6 ®) 2 OT QaoeF eer eee Seem ere ert as rene Cnn ear) Cee sae ees see ene tess vevavenesageree ees Sere w rene Oe reer eee eeee Pee eseeres Ore e veer eesenne s[038u09 ps F "GOGL ‘Y70G usd wo SPUIULPSOQUT CT pee Ee) = Teen eapes 6 § 881 “CC GOBT Tdy 190¢ uo sxoyueg qe ooulRlyEy Se ParGScaiies eee (oseysog pur vay, Sutpnyaur) sosuedxa Ayjog Tao 10 sos Kouoyng pue sunuag snoouv|poostpy 0 0 908 = : | 8 5 LF sdugonnoon oad “— cornqmyarg G OL 80G Be a ar SUON RAST] eS padaouetoogodptonuocGsnond -noodsenerDdcoE, a Il 9 ocax SooEEO eh gemteae eats eects Saat - | ee = s}dieooyy snoeuulpeosipy + SuoHvoriqng Jo sesuadxgy a 9-2 “""* sengopezZQ pue sSurpee00.1g T IT &0¢ eauObOOD. Suga dseoUnSEDcoOnSOGoR AA: OSB 16g a es eau IT € 18 Teese SUTpUIE, J GEL ge otter srromaesues PAY ye) ds ae cso cca. oben ot eceor creer OL 4 SclF MM uy DIG: Be —! SUOT}woTGNg Jo sarug pA a ee soe Ct ee AEIQUy 0 0 06 Peete ese eneweees. ee es es 2 eauOnsodntog 0 aT Fé4 Seaneinas seUpoUEoOnoDS Shae NdCACUGAEAATeID OREO senles Uo) /4 eee dees cea Sezer ee eR “* SaOrNgIIqUOD [enooy G2 Ne 3 SB) pue STROH ens Tab ace egy eager rdlee evataslcn ele nGme t Tan . Fe OTD EEE EOG ach OLDE Sap ctpens i MERC Cts 0 0 Wf UOlsstupy ee re, ee c = is DRE IY ND) Goi 686 ee) (081 ‘AVL JB] uO sroqueg qe souRlege PES eee spdaeoagy “GOGT “Y708 22d 02 ‘TORT ‘asT hon morf ‘hjawoy unauury ay) fo spuauhng pwo sidraoaar — (ZO6T ‘CIES Ae qi ‘Surjoayy ATVSLOAIU OY oy} 4B poqmoserg) 3 = NOTICE. te) Journal of Zoology. Volume XXVIII. commenced with Part 179. The Journal (both Zoological and Botanical) is now issued in THREE PARTS PER ANNUM as follows :— Part I., containing papers read from November to the middle of January, on April Ist. Paer II., containing papers read from the middle of January to the end of April, on July Ist. Part III., containing papers read in May and June, on November Ist. A GENERAL INDEX to the first twenty Volumes of the Journal (Zoology) may be had on application, either in cloth or in sheets for binding. Price to Fellows, 15s.; to the Public, 20s. A CATALOGUE of the LIBRARY may be had on > aaaiae Price to Fellows, 5s.; to the Public, 10s, f ie ira mer ee iN AAV Beer ie A , ale ANAS AP a aa ar A Ape ARRAN ACE A VAAAAAA A AA AAA Aaa eons SeneANel ac An BARA MAAAAAAAARA, ‘A AAR { AYA AAA, a AAR RANA. AAAAA Aaaas ’ WAAARAAARAA Ne APEPRRRAA A | A : \- RAE ARAAAAAA AAAAA AA) Resonant tene AA A\\A\ AnAAAY | NIAARAARAAAAA «AAR ARAT AAR ARR ARR RRR AY).T YY. Ann \ A ARAAA ala | A\AA\AAVAY AAWAAAAA A AAAE an \AAARAAAARA Wan RAAAAAAAAAAAARA —AAAAAA AN WAAAAR Aman AAA An nn nanaasnaa anne aan aannnan nanan ae A ane Aa AAA R Reena AAAARARAARRR AAAS A AAA iA BAARARARAAARAAAAAAAA ARRAS a ’ fn fe vy | nr Anaaane AARAAAAAAAAAARAARARAAR Aaa aana Anannnnenn nae Dannanngnnannan AAAANAAD [way AY NS